You are on page 1of 503

Developments in Petroleum Science, 1

GEOCHEMISTRY OF OILFIELD WATERS

This Page Intentionally Left Blank

Developments in Petroleum Science, 1

GEOCHEMISTRY OF OILFIELD WATERS


A. GENE COLLINS
Bartlesville Energy Research Center Bureau o f Mines United States Department of the Interior Bartlesville, Oklahoma, U.S.A.

ELSEVIER SCIENTIFIC PUBLISHING COMPANY Amsterdam

Oxford

New York 1975

ELSEVIER SCIENTIFIC PUBLISHING COMPANY 335 Jan van Galenstraat P.O. Box 211,Amsterdam, The Netherlands AMERICAN ELSEVIER PUBLISHING COMPANY, INC. 52 Vanderbilt Avenue New York, New York 10017

Library of Congress Card Number: 73-89149 ISBN 0-444-41183-6 With 132 illustrations and 87 tables Copyright 0 1975 by Elsevier Scientific Publishing Company, Amsterdam All rights reserved. No part of this publication may be reproduced, stored in a retrieva system, or transmitted, in any form or by any means, electronic, mechanical, photo copying, or otherwise without the prior written permission o the publisher, Elseviei f Scientific Publishing Company, Jan van Galenstraat 335,Amsterdam Printed in The Netherlands

PREFACE

The purpose of this book is to provide information relevant to the analytical chemistry and geochemistry of oilfield waters. The book attempts to recognize the importance of subsurface oilfield waters as they are related t o origin, migration, accumulation, and maturation of oil and gas and thus their relationship t o exploration for and production of oil and gas. One chapter presents a simplistic introduction t o the origin of oilfield waters. Because oilfield waters can constitute an environmental pollution hazard, the book describes and comments on methods of their disposal or of recovering valuable constituents from them. The numerous references indicate that the book relies heavily upon the work of others. The reader will vastly expand his knowledge of the subject by consulting these references. The writer appreciates the understanding and thoughtfulness of his Wife, Barbara, and children, Sandy and Mike, during the preparation of part of this book at our home. He acknowledges With appreciation the criticisms, opinions, and suggestions of various portions of the book by O.C. Baptist, W.H. Caraway, P.H. Dickey, G.L. Gates, R.V. Huff, P.H. Jones, and C.C. Linville. M.E. Crocker and Ms. C.A. Pearson, did an invaluable service of proof-reading and index preparation. He extends appreciation t o Ms. D.J. Forbes, Ms. M.G. Goff, and Ms. J. Haimson for typing the manuscript; t o D.W. Anderson, Ms. E.S. Baldwin, J.A. Chidester, G.E. Fletcher, R.M. Horn, and W.A. McClung for preparing the figures; and to authors, book publishers, companies, and technical journals who granted permission t o use various illustrations. Permission t o publish this manuscript was granted by the Director of the United States Bureau of Mines. Bureau of Mines officials who generously helped obtain this permission were: J.S. Ball, R.T. Johansen, and J.W. Watkins. Finally inasmuch as it is the writers belief that this book is not perfect, he takes this opportunity to solicit constructive criticism from its readers.
A. GENE COLLINS Bartlesville Energy Research Center U.S.Bureau of Mines Bartlesville, Oklahoma

This Page Intentionally Left Blank

CONTENTS

Preface

. . . . . . . . . . . . . . . . . . . . . . . . . .

V 1 5

Chapter 1. Introduction References . . . . .

. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

Chapter 2 . Sampling subsurface oilfield waters . . . Drill-stem test . . . . . . . . . . . . . . Sample containing dissolved gases . . . . . . . Sampling at the flow line . . . . . . . . . . . Sampling at the wellhead . . . . . . . . . . . Sample for determining unstable properties or species . Sample for stable-isotope analysis . . . . . . . Sample containers . . . . . . . . . . . . . Tabulation of sample description . . . . . . . References . . . . . . . . . . . . . . . .

7
8

. . .

12 13 14 14 15 16 17 17 19 19 22 25 27 27 29 31 32 35 37 37 40 43 44 45 47 50 51 52 53 54 57 58 59 61 62 63

Chapter 3. Analysis of oilfield waters for some physical properties and inorganic chemical constituents . . . . . . . . . . . . . . . . Quality control . . . . . . . . . . . . . . . . . . . . . . . Preliminary sample treatment . . . . . . . . . . . . . . . . . . Reporting the analytical results . . . . . . . . . . . . . . . . . Synthetic brine . . . . . . . . . . . . . . . . . . . . . . . Determination of pH . . . . . . . . . . . . . . . . . . . . . Determination of Eh . . . . . . . . . . . . . . . . . . . . . Suspended solids . . . . . . . . . . . . . . . . . . . . . . . Resistivity . . . . . . . . . . . . . . . . . . . . . . . . . Specific gravity . . . . . . . . . . . . . . . . . . . . . . . TITRIMETRIC METHODS . . . . . . . . . . . . . . . . . . Acidity. alkalinity. and borate boron . . . . . . . . . . . . . . . Calcium and magnesium . . . . . . . . . . . . . . . . . . . Ammonium nitrogen . . . . . . . . . . . . . . . . . . . . Chloride . . . . . . . . . . . . . . . . . . . . . . . . . Bromide and iodide . . . . . . . . . . . . . . . . . . . . . Oxygen . . . . . . . . . . . . . . . . . . . . . . . . . Carbon dioxide . . . . . . . . . . . . . . . . . . . . . . Sulfide . . . . . . . . . . . . . . . . . . . . . . . . . Sulfur compounds . . . . . . . . . . . . . . . . . . . . . FLAME SPECTROPHOTOMETRIC .METHODS . . . . . . . . . . . Lithium . . . . . . . . . . . . . . . . . . . . . . . . . Sodium . . . . . . . . . . . . . . . . . . . . . . . . . Potassium . . . . . . . . . . . . . . . . . . . . . . . . Rubidium and cesium . . . . . . . . . . . . . . . . . . . . Manganese . . . . . . . . . . . . . . . . . . . . . . . . Strontium . . . . . . . . . . . . . . . . . . . . . . . . Barium . . . . . . . . . . . . . . . . . . . . . . . . .

. . .

. . . . . . . . . . . . . . . . . .

VIII ATOMIC ABSORPTION METHODS . . . . . . . . . . Interferences . . . . . . . . . . . . . . . . . . . Burners and solvents . . . . . . . . . . . . . . . . . Lithium . . . . . . . . . . . . . . . . . . . . . . Sodium . . . . . . . . . . . . . . . . . . . . . Potassium . . . . . . . . . . . . . . . . . . . . Magnesium (1) . . . . . . . . . . . . . . . . . . . Calcium (1) . . . . . . . . . . . . . . . . . . . . . Magnesium (2) . . . . . . . . . . . . . . . . . . . Calcium (2) . . . . . . . . . . . . . . . . . . . . Strontium . . . . . . . . . . . . . . . . . . . . Barium . . . . . . . . . . . . . . . . . . . . . Manganese . . . . . . . . . . . . . . . . . . . . Iron . . . . . . . . . . . . . . . . . . . . . . Copper . . . . . . . . . . . . . . . . . . . . . Zinc . . . . . . . . . . . . . . . . . . . . . . Lead(1) . . . . . . . . . . . . . . . . . . . . . Lead(2) . . . . . . . . . . . . . . . . . . . . . EMISSION SPECTROSCOPY . . . . . . . . . . . . . Barium, boron, iron, manganese, and strontium . . . . . . . Beryllium . . . . . . . . . . . . . . . . . . . . Aluminum . . . . . . . . . . . . . . . . . . . . MASS SPECTROMETRIC METHODS FOR STABLE ISOTOPES . Deuterium . . . . . . . . . . . . . . . . . . . . Oxygen-18 . . . . . . . . . . . . . . . . . . . . COLORIMETRIC METHODS . . . . . . . . . . . . Interferences . . . . . . . . . . . . . . . . . . . Iron . . . . . . . . . . . . . . . . . . . . . . Copper . . . . . . . . . . . . . . . . . . . . . Nickel . . . . . . . . . . . . . . . . . . . . . Lead . . . . . . . . . . . . . . . . . . . . . . Zinc . . . . . . . . . . . . . . . . . . . . . . Cadmium . . . . . . . . . . . . . . . . . . . . Phosphate . . . . . . . . . . . . . . . . . . . . Silica . . . . . . . . . . . . . . . . . . . . . . Nitrate nitrogen . . . . . . . . . . . . . . . . . . Arsenic . . . . . . . . . . . . . . . . . . . . . Fluoride . . . . . . . . . . . . . . . . . . . . . Iodide . . . . . . . . . . . . . . . . . . . . . Selenium . . . . . . . . . . . . . . . . . . . . Barium . . . . . . . . . . . . . . . . . . . . . GRAVIMETRIC METHODS . . . . . . . . . . . . . Sulfate . . . . . . . . . . . . . . . . . . . . . Barium . . . . . . . . . . . . . . . . . . . . . OTHER METHODS . . . . . . . . . . . . . . . . . Sodium . . . . . . . . . . . . . . . . . . . . . Dissolved solids . . . . . . . . . . . . . . . . . . Spent acid . . . . . . . . . . . . . . . . . . . . Acetic acid solutions . . . . . . . . . . . . . . . . . References . . . . . . . . . . . . . . . . . . . . Chapter 4 . Interpretation of chemical analysis of oilfield waters Calculating probable compounds . . . . . . . . . .

CONTENTS

. . . . . . . . . . . . . . .

. . . . . . . . . . . . . . . . . . . . . . . .

. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

. . . .

. . . . . . . . . . . . . . . . . . .

. . . .

. . . . . . . . . . . . . . . . . . .

. . . .

. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

. . . .

. . . .

65 66 66 68 68 70 71 72 74 75 76 77 78 79 80 80 81 : 82 . 83 . 83 89 90 . 91 91 91 . 92 93 94 96 98 99 101 103 105 107 107 108 109 110 11 1 114 . 114 114 115 116 116 117 118 120 121

. . . . . . . 125 . . . . . . . 125

CONTENTS Determining a sought compound Graphic plots . . . . . . . References . . . . . . . .

IX

. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

127 128 132 133 133 136 138 140 141 141 142 143 145 147 149 149 150 151 151 152 152 153 155 156 157 158 158 158 159 160 161 161 162 164 166 174

Chapter 5 . Significance of some inorganic constituents and physical properties of oilfield waters . . . . . . . . . . . . . . . . . . . . . Lithium . . . . . . . . . . . . . . . . . . . . . . . . . . Sodium . . . . . . . . . . . . . . . . . . . . . . . . . . Potassium . . . . . . . . . . . . . . . . . . . . . . . . . Rubidium . . . . . . . . . . . . . . . . . . . . . . . . . Cesium . . . . . . . . . . . . . . . . . . . . . . . . . . Beryllium . . . . . . . . . . . . . . . . . . . . . . . . . Magnesium . . . . . . . . . . . . . . . . . . . . . . . . . Calcium . . . . . . . . . . . . . . . . . . . . . . . . . . Strontium . . . . . . . . . . . . . . . . . . . . . . . . . Barium . . . . . . . . . . . . . . . . . . . . . . . . . . Manganese . . . . . . . . . . . . . . . . . . . . . . . . . Iron . . . . . . . . . . . . . . . . . . . . . . . . . . . Copper . . . . . . . . . . . . . . . . . . . . . . . . . . Zinc . . . . . . . . . . . . . . . . . . . . . . . . . . . Mercury . . . . . . . . . . . . . . . . . . . . . . . . . . Lead . . . . . . . . . . . . . . . . . . . . . . . . . . . Cadmium . . . . . . . . . . . . . . . . . . . . . . . . . Boron . . . . . . . . . . . . . . . . . . . . . . . . . . Aluminum . . . . . . . . . . . . . . . . . . . . . . . . . Silica . . . . . . . . . . . . . . . . . . . . . . . . . . . Ammonium nitrogen . . . . . . . . . . . . . . . . . . . . . Phosphorus . . . . . . . . . . . . . . . . . . . . . . . . . Arsenic . . . . . . . . . . . . . . . . . . . . . . . . . . Oxygen . . . . . . . . . . . . . . . . . . . . . . . . . . Sulfur . . . . . . . . . . . . . . . . . . . . . . . . . . Selenium . . . . . . . . . . . . . . . . . . . . . . . . . Fluorine . . . . . . . . . . . . . . . . . . . . . . . . . . Chlorine . . . . . . . . . . . . . . . . . . . . . . . . . . Bromine . . . . . . . . . . . . . . . . . . . . . . . . . . Iodine . . . . . . . . . . . . . . . . . . . . . . . . . . Significance of some physical properties . . . . . . . . . . . . . . . References . . . . . . . . . . . . . . . . . . . . . . . . . Chapter 6. Organic constituents in saline waters . Nitrogen-free organic compounds . . . . . Hydrocarbons containing nitrogen . . . . . Fatty acids . . . . . . . . . . . . . . Naphthenic and humic acids . . . . . . . Determination of oil in water . . . . . . . Organic acids in oilfield brines . . . . . . References . . . . . . . . . . . . . .

. . . . . . . . Chapter 7. Origin of oilfield waters . . . . . . Definitions of some water terms . . . . . . . Sedimentary rocks . . . . . . . . . . . . Composition of oilfield waters . . . . . . .

. . . . . . . . . . . .

Research studies related t o the originof oilfield brines

. . . . . . . . . . . . .

. . . . . . . . . . . . .

. . . . . . . . . . . . .

. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

. 177 . 178 . 182 183 . 185 . 186 . 188


188

. 193 . 194 195 . 213 . 219

X
Conclusions References .

CONTENTS

. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
. . . . . . . . . . . . . . . . . . . .

245 246

Chapter 8 . Classification of oilfield waters Palmers classification . . . . . . . Sulins classification . . . . . . . . Modification of Sulins system by Bojarski Chebotarevs classification . . . . . . Schoellers system . . . . . . . . Oilfield brine analyses . . . . . . . Application of the classification systems . Discussion . . . . . . . . . . . Conclusions . . . . . . . . . . References . . . . . . . . . . .

. . . . . . . . . . . . 253 . . . . . . . . . . . . 254

. . . . . . . . . . . . . .

. . . . . . . .

. . . . . . . .

. . . . . . . .

. . . . . . . .

. . . . . . . .

. . . . . . . .

. . . . . . . .

. . . . . . . .

. . . . . . . .

. . . . . . . .

. . 257 . . 260 . . 262 . . 267 . . 272 . . 274 . . 289 . . 291


291 293 294 295 298 299 304 304

Chapter 9. Some effects of water upon the generation. migration. accumulation. and alteration of petroleum . . . . . . . . . . . . . . . . . Compaction . . . . . . . . . . . . . . . . . . . . . . . . Generation and migration . . . . . . . . . . . . . . . . . . . . Accumulation . . . . . . . . . . . . . . . . . . . . . . . . Alteration . . . . . . . . . . . . . . . . . . . . . . . . . Conclusions . . . . . . . . . . . . . . . . . . . . . . . . References . . . . . . . . . . . . . . . . . . . . . . . . .

Chapter 10. Geochemical methods of exploration for petroleum and natural gas . . 307 Introduction . . . . . . . . . . . . . . . . . . . . . . . . 307 Hydrogeochemical research and methods . . . . . . . . . . . . . . . 313 Reviews . . . . . . . . . . . . . . . . . . . . . . . . . . 322 Case history of the Delaware sand (Bell Canyon formation). Texas. by Visher (1961) 322 Formation water maps of others areas . . . . . . . . . . . . . . . . 330 Concluding remarks . . . . . . . . . . . . . . . . . . . . . . 335 References . . . . . . . . . . . . . . . . . . . . . . . . . 337 Chapter 11. Geopressured reservoirs . . Geopressure . . . . . . . . . . . Origin of abnormal pressures . . . . . Abnormal pressures in the Gulf Coast area Detection of abnormal pressures . . . . References . . . . . . . . . . . .

. . . . . . . . . . . . . . . . . . . . . . . . 367 . . . . . 368 . . . . . 370 Solubilities of the sulfates of barium and strontium in saline solutions . . . . . 372 Experimental determination of some solubilities of the sulfates of barium and strontium . . . . . . . . . . . . . . . . . . . . . . . . . 372 Resultsand discussion of the experimental investigation . . . . . . . . . . 373 Brine stabilization . . . . . . . . . . . . . . . . . . . . . . 380 Mixing of subsurface waters . . . . . . . . . . . . . . . . . . . 382 References . . . . . . . . . . . . . . . . . . . . . . . . . 386
Chapter 13 Valuable minerals in oilfield waters . . Recovery of iodine and bromine from oilfield brines

. . . . Chapter 12. Compatibility of oilfield waters . . . . . . . Wellbore and formation damage . . . . . . . . . . . Solubility of calciumcompounds invarioussaltsolutions . .

. . . . . .

. . . . . .

. . . . . .

. . . . . .

. . . . . .

. . . . . .

. . . . . . . . . . . . . . .

. . . . . . .

. . . . . . .

343 343 344 346 362 364

. . . . . . . . . . . 389 . . . . . . . . . . . 390

CONTENTS Minerals recovered from saline waters . . . . Fresh-water production . . . . . . . . . . Preliminary economic evaluation . . . . . Disposal brines . . . . . . . . . . . . Worth and value estimates . . . . . . . . . Conclusions . . . . . . . . . . . . . References . . . . . . . . . . . . . .

XI

. . . . . . .

. . . . . . .

. . . . . . .

. . . . . . .

. . . . . . .

. . . . . . .

. . . . . . . 392 . . . . . 401 . . . . . . 402 . . . . . 411 . . . . . 411 . . . . . 417 . . . . . 417

Chapter 1 4.Subsurface disposal . . . . . . . History of brine disposal operations . . . . . . Subsurface injection . . . . . . . . . . . . Present-day technology in subsurface disposal . . Economics and oilfield brine disposal . . . . . Injection well versus disposal well . . . . . . Acceptable geologic areas . . . . . . . . . . Suitable disposal zones . . . . . . . . . . . Evaluation of the disposal zone . . . . . . . Considerations during drilling and well completion . Fluid travel . . . . . . . . . . . . . . . Hazards of underground waste disposal . . . . State regulations and tax incentives . . . . . . Costs of disposal systems . . . . . . . . . . Conclusions . . . . . . . . . . . . . . References . . . . . . . . . . . . . . .

. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

. . . . . . . . . . . . . . . .

. . . . . . . . . . . . . . .

Chapter 1 5 Solubilities of some silicate minerals in saline waters Composition and structure of minerals studied . . . . . . Silicate solubilities a t 25C and 1 atm . . . . . . . . . Experimental equipment . . . . . . . . . . . . . Experimental method . . . . . . . . . . . . . . Fundamental equations . . . . . . . . . . . . . . Experimental dataandempirical equations . . . . . . . Summary and conclusions . . . . . . . . . . . . . References . . . . . . . . . . . . . . . . . .

. . . . . . . 419 . . . . . . . 419 . . . . . . 420 . . . . . . . 421 . . . . . . . 422 . . . . . . . 424 . . . . . . 425 . . . . . . 426 . . . . . . . 427 . . . . . . . 432 . . . . . . 433 . . . . . . . 434 . . . . . . . 434 . . . . . . 437 . . . . . . 438 . . . . . . 438 . . . . . . . 441 . . . . . . . 441 . . . . . . . 443 . . . . . . . 447 . . . . . . . 449 . . . . . . . 449 . . . . . . . 453 . . . . . . . 457 . . . . . . . 459

Chapter 16 Environmental impact of oil- and gas-well drilling. production and associated waste disposal practices . . . . . . . . . . . . Drilling . . . . . . . . . . . . . . . . . . . . . . . . . . Production . . . . . . . . . . . . . . . . . . . . . . . . . Disposal . . . . . . . . . . . . . . . . . . . . . . . . . . References . . . . . . . . . . . . . . . . . . . . . . . . . Reference Index Subject Index .

. 461
461 467 471 474 477 485

. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

This Page Intentionally Left Blank

Chapter 1 .

INTRODUCTION

Petroleum, known to exist long before an oil well was drilled, first found limited use as a medicine, lubricant, and waterproofing agent. The American Indians knew of several oil and gas springs and gave this information t o the early American settlers. Early settlements were commonly located close to salt licks which supplied salt to the population. Often these salt springs were contaminated with petroleum, and many of the early efforts to acquire more salt by digging wells were rewarded by finding unwanted increased amounts of oil and gas associated with the saline waters. In the Appalachians: many saline water springs occurred along the crests of anticlines (Rogers and Rogers, 1843). In 1855 it was found that distillation of petroleum produced a light oil similar to coal oil, which was better than whale oil as an illuminant (Howell, 1934, p.2). This knowledge spurred the search for saline waters which contained oil. Colonel Edward Drake, utilizing the methods of the salt producers, drilled a well on Oil Creek, near Titusville, Pennsylvania, in 1859. He struck oil at a depth of 21 m, and this first oil well produced about 35 barrels of oil per day (Dickey, 1959). The early oil producers did not realize the significance of the oil and saline waters occurring together. In fact, it was not until 1938 that the existence of interstitial water in oil reservoirs was generally recognized (Schilthuis, 1938). Torrey (1966) was convinced as early as 1928 that dispersed interstitial water existed in oil reservoirs, but his belief was rejected by his colleagues because most of the producing oil wells did not produce any water upon completion. Occurrences of mixtures of oil and gas with water were recognized by Griswold and Munn (1907), but they believed that there was a definite separation of the oil and water, and that oil, gas, and water mixtures did not occur in the sand before a well tapped the reservoir. It was not until 1928 that the first commercial laboratory for the analysis of rock cores was established (Torrey, 1966); the first core tested was from the Bradford Third Sand (from the Bradford field, McKean County, Pennsylvania). The percent saturation and percent porosity of this core were plotted versus depth t o construct a graphic representation of the oil and water saturation. The soluble mineral salts that were extracted from the core led Torrey to suspect that water was indigenous t o the oil productive sand. Shortly thereafter a test well was drilled near Custer City, Pennsylvania, which encountered higher than average oil saturation in the lower part of the Bradford Sand. This high oil saturation resulted from the action of an un-

INTRODUCTION

suspected flood, the existence of which was not known when the location for the test well had been selected. The upper part of the sand was not cored. Toward the end of the cutting of the first core with a Baker cable tool core barrel, oil began t o come into the hole so fast that it was not necessary t o add water for the cutting of the second section of the sand. The lower 1 m of the Bradford Sand therefore was cut with oil in a hole free from water. Two samples from this section were preserved in sealed containers for saturation tests, and both of them, when analyzed, had a water content of approximately 20% of pore volume. This well made about 10 barrels of oil per day and no water after being shot with nitroglycerine. Thus, the evidence developed by the core analysis and the productivity test after completion provided a satisfactory indication of the existence of immobile water, indigenous t o the Bradford Sand oil reservoir, which was held in its pore system and which could not be produced by conventional pumping methods (Torrey, 1966). Fettke (1938) was the first t o report the presence of water in an oilproducing sand. However, he thought that it might have been introduced by the drilling process. It was recognized by Munn (1920)that moving underground water might be the primary cause of migration and accumulation of oil and gas. However, this theory had little experimental data t o back it until Mills (1920)conducted several laboratory experiments on the effect of moving water and gas on water-oil-as-sand and water-oil-sand systems. Mills concluded that the up-dip migration of oil and gas under the propulsive force of their buoyancy in water, as well as the migration of oil, either up or down dip, caused by hydraulic currents, are among the primary factors influencing both the accumulation and the recovery of oil and gas. This theory was seriously questioned and completely rejected by many of his contemporaries. Rich (1923)postulated that hydraulic currents, rather than buoyancy, are effective in causing accumulation of oil or its retention. He did not believe that the hydraulic accumulation and flushing of oil required a rapid movement of water, but rather that the oil was an integral constituent of the rock fluids and that it could be carried along with them whether the movement was very slow or relatively rapid. The effect of water displacing oil during production was not recognized in the early days of the petroleum industry in Pennsylvania. Laws were passed, however, to prevent operators from injecting water into the oil reservoir sands through unplugged wells. In spite of these laws, some operators at Bradford surreptitiously opened the well casing opposite shallow groundwater sands in order to start a waterflood in the oil sands. Effects of artificial waterfloods were noted in the Bradford field, McKean County, Pennsylvania, in 1907, and became evident about 5 years later in the nearby oilfields of New York (Torrey, 1950). Volumetric calculations of the oil-reservoir volume which were made for engineering studies of these waterflood opera-

INTRODUCTION

tions proved that interstitial water was generally present in the oil sands. Publications by Garrison (1935) and Schilthuis (1938) give detailed information concerning the distribution of water and oil in porous rocks, and of the origin and occurrence of connate water with information concerning the relationship of water saturation to formation permeability. The word connate was first used by Lane and Gordon (1908) to mean interstitial water that was deposited with the sediments. The processes of rock compaction and mineral diagenesis result in the expulsion of large amounts of water from sediments and movement out of the deposit through the more permeable rocks. It is therefore highly unlikely that the water now in any pore is the same as that which was there when the particles that surround it were deposited. White (1957) redefined connate water as fossil water; it has been out of contact with the atmosphere for an appreciable part of a geologic time period. Connate water is thus distinguished from meteoric water which has entered the rocks in geologically recent times, and from juvenile water which has come from deep in the earths crust and has never been in contact with the atmosphere. Meanwhile petroleum engineers and geologists had learned that waters associated with petroleum could be identified with regard to the reservoir in which they occurred by a knowledge of their chemical characteristics. Commonly the waters from different strata differ considerably in their dissolved chemical constituents, making the identification of a water from a particular strata easy. However, in some areas the concentrations of dissolved constituents in waters from different strata do not differ significantly, and the identification of such waters is difficult or impossible. The amount of water produced with the oil often increases as the amount of oil produced decreases. If this is edge water, nothing can be done about it. If it is bottom water, the well can be plugged back. However, it often is intrusive water from a shallow sand gaining access t o the well from a leaky casing or faulty completion and this can be repaired. Enormous quantities of water are produced with the oil in some fields, and it is necessary to separate the oil from the water. Most of the oil can be removed by settling. Often, however, an oil-in-water emulsion forms which is very difficult t o break. In such cases, the oil is heated and various surfaceactive chemicals are added to induce separation. In the early days, the water was dumped on the ground where it seeped below the land surface. Until about 1930, the oilfield waters were disposed into local drainage, frequently killing fish and even surface vegetation. After 1930, it became common practice t o evaporate the water in earthen pits or to inject it into the producing sand or another deep aquifer. The primary concern in such disposal practice is to remove all oil and basic sediment from the waters before pumping them into injection wells, to prevent clogging of the pore spaces in the formation receiving the waste water. Chemical compatibility of waste water and host aquifer water must also be assured. Waters produced with petroleum are growing in importance. In years past,

INTRODUCTION

these waters were considered waste and had t o be disposed of in some manner. Injection of these waters into the petroleum reservoir rock serves three purposes: (1) it produces additional petroleum (secondary recovery); (2) it utilizes a potential pollutant; and (3) in some areas it controls land subsidence. The volume of water produced with petroleum in the United States is very large. In 1970, daily production was about 3.78 trillion liters of water with about 1.51 trillion liters of oil. In older fields, the production is frequently 95%water and 5%petroleum. . Waterflooding in petroleum production is expanding rapidly, and in 1970 one-third to one-half of the production in the United States came from fields into which water was injected. The volume of injected water has grown each year. In many fields the volume of petroleum produced by secondary recovery by waterflooding is equal to the volume recovered by primary met hods. To inject these waters into reservoir rocks, suspended solids and oil must be removed from the waters to prevent plugging of the porous formations. Water injection systems require separators, filters, and, in some areas, deoxygenating equipment utilizing chemical and physical control methods to minimize corrosion and plugging in the injection system. In waterflooding most petroleum reservoirs, the volume of produced water is not sufficient t o efficiently recover the additional petroleum. There. fore, supplemental water must be added t o the petroleum reservoir. The use of waters from other sources requires that the blending of the produced water with supplemental water must yield a chemically stable mixture so that plugging solids will not be formed. For example, a produced water containing considerable calcium should not be mixed with a water containing considerable carbonate because calcium carbonate may precipitate and prevent injection of the flood water. The design and successful operation of a secondary recovery waterflood requires a thorough knowledge of the composition of the waters used. Chemical analyses of waters produced with oil are useful in oil production problems, such as identifying the source of intrusive water, planning waterflood and salt-water disposal projects, and treating t o prevent corrosion problems in primary and secondary recovery. Electrical well-log interpreta tion requires a knowledge of the dissolved solids concentration and composi tion of the interstitial water. Such information also is useful in correlationof stratigraphic units and of the aquifers within these units, and in studiesof the movement of subsurface waters. It is impossible to understandthe processes that accumulate petroleum or other minerals without insight in to the nature of these waters.

REFERENCES

References
Dickey, P.A., 1959. The first oil well. J. Pet. Technol., 11:14-26. Fettke, C.R., 1938. Bradford oil field, Pennsylvania, and New York. Pa. Geol. Surv., Fourth Ser., Bull., M21:l-454. Garrison, A.D., 1935. Selective wetting of reservoir rock and its relation to oil production. In: Drilling and Production Practice. American Petroleum Institute, New York, N.Y., pp.130-140. Griswold, W.T. and Munn, M.J., 1907. Geology of oil and gas fields in Steubenville, Burgettstown and Claysville Quadrangles, Ohio, West Virginia and Pennsylvania. U.S. Geol. Sum. Bull., No.318, 196 pp. Howell, J.V., 1934. Historical development of the structural theory of accumulation of oil and gas. In: W.E. Wrather and F.H. Lahee (Editors), Problems of Petroleum Geology. American Association of Petroleum Geologists, Tulsa, Okla., pp.1-23. Lane, A.C. and Gordon, W.C., 1908. Mine waters and their field assay. Bull. Geol. SOC. A m . , 19:501-512. Mills, R. van A., 1920. Experimental studies of subsurface relationships in oil and gas fields. Econ. Geol., 15:398-421. Munn, M.J., 1920. The anticlinal and hydraulic theories of oil and gas accumulation. Econ. Geol., 4:509-529. Rich, J.L., 1923. Further notes on the hydraulic theory of oil migration and accumulation. Bull. Am. Assoc. Pet. Geol., 7:213-225. Rogers, W.B. and Rogers H.D., 1843. On the connection of thermal springs in Virginia with anticlinal axes and faults. A m . Geol. Rep., 1313.323-347. Schilthuis, R.J., 1938. Connate water in oil and gas sands. In: Petroleum Development and Technology, AIME, pp.199-214. Torrey, P.D., 1950. A review of secondary recovery of oil in the United States. In: Secondary Recovery of Oil in the United States. American Petroleum Institute, New York, N.Y., pp.3-29. Torrey, P.D., 1966. The discovery of interstitial water. Prod. Monthly, 30:8-12. White, D.E., 1957. Magmatic, connate, and metamorphic water. Bull. Geol. SOC.A m . , 68:1659-1682.

This Page Intentionally Left Blank

Chapter 2.

SAMPLING SUBSURFACE OILFIELD WATERS

Subsurface waters associated with petroleum are subjected to forces that promote mixing and homogeneity, but the assumption cannot be made that they are so well mixed that no attention to sampling technique is required. Localized conditions within an aquifer are commonly such that a given subsurface body of water may not be of uniform composition. The composition of subsurface water commonly changes with depth, and also laterally in the same aquifer. Changes may be brought about by the intrusion of other waters, and by discharge from and recharge to the aquifer. It is thus difficult to obtain a representative sample of a given subsurface body of water because any one sample is a very small part of the total mass, which may vary widely in composition. To develop a comprehensive picture of the composition characteristics of the total mass, it is generally necessary t o obtain and analyze many samples. Also, the samples may change with time as gases come out of solution and supersaturated solutions approach saturation. The sampling sites should be selected, if possible, t o fit into a comprehensive network t o cover an oil-productive geologic basin. Considerations in selecting sampling sites are as follows: (1)Which sites will better fit into an overall plan to evaluate the chemistry of the waters on a broad basis? (2) Which sites will yield the better information for correlation with data obtained from other sites? (3) Which sites are more representative of the total chemistry of brines from a given area? The value of the sample is directly proportional to the facts known about its source; therefore, sites should be selected for which the greater source knowledge is available. For surveillance purposes, samples can be collected from the same site at sufficiently frequent intervals that no important change in quality will occur between sampling times. Change in composition may result from changes in rate of water movement, pumpage rates, or infiltration of other water. Changes that can occur in petroleum-associated water are illustrated in Table 2.1. Well 1 shows the sort of change that commonly occurs. The water from well 2 did not change between 1947 and 1957, within the accuracy of the analytical determination. Water from well 3 changed drastically, suggesting the intrusion of water from a different source.

8
TABLE 2.1

SAMPLING SUBSURFACE OILFIELD WATERS

Changes in the composition of petroleum-associated waters (mg/l) Constituent Well 1 Well 2 Well 3

1947
Sodium and potassium Magnesium Calcium Bicarbonate Sulfate Chloride Total dissolved solids

1957 25,000 1,200 5,500 12 50 51,800 83,562

1947 46,038 2,011 14,200 24 3 102,100 164,376

1957 45,924 2,200 14,400 12 52 102,800 165,388

1956 1,491 30 60 600 200 2,000 4,381

1959 856 2 10 1,800 0 300 2,968

29,062 1,100 5,900 34 14 58,500 94,610


~-

_ .

There is a tendency for some petroleum-associated waters t o become more dilute as the oil reservoir is produced. Such dilution may result from the movement of dilute water from adjacent compacting clay beds into the petroleum reservoir as pressure declines with the continued removal of oil and brine (Wallace, 1969). The composition of petroleum-associated water varies with the position within the geologic structure from which it is obtained. For example, if the water table is tilted, the more dilute water probably will be on the structurally high side. In some cases the salinity will increase upstructure t o a maximum at the point of oil-water contact. Usually this is caused by infiltrating meteoric waters. Few of the samples collected by drill-stem test are truly representative formation-water samples. During drilling, the pressure in the well bore is intentionally maintained higher than that in the formations. Filtrate from the drilling mud seeps into the permeable strata, and this filtrate is the first liquid to enter the test tool. The most truly representative formation-water sample usually is obtained after the oil well has produced for a period of time and all extraneous fluids adjacent t o the wellbore have been flushed out. Samples taken immediately after the well is completed may be contaminated with drilling muds, with drilling fluids, and/or with well completion fluids, such as filtrate from cement, tracing fluids, and acids, which contain many different chemicals.

Drill-stem test
The drill-stem test, if properly made, can provide a reliable formation water sample. Mud filtrate wl be the first fluid to enter the drill-stem test il tool, and it will be found at the top of the fluid column immediately below

DRILL-STEM TEST

tester

Multiple closed i pressure s a m p l e

Flow stream pressure recorde V e r t i c a l and rot

Locked down

Blanked o f f pressure r e c o r d

RUNNING IN HOLE

Fig. 2.1. Multiple closed-in-pressure subsurface sampler. (Courtesy of the Halliburton Company.)

10

SAMPLING SUBSURFACE OILFIELD WATERS


Droin
YOIV4

Oroln

'
Rubber doughnut Sample

Floottnq p l s t o n
h

SAMPLE UNIT,' for l o w p e r m o b i l i t i e s

Mud

Dump chamber v o l v e ( l o c k s open)

"
MECHANICAL U N I T F I M - A

- J

SAMPLE U N I T

SURFACE CONTROL I N 0 ICATIDNS

m E ?
SP Tester positioning depth S A M P L I N G PRESSURE Pod set Tool open I n i t i a l shut-in prss1ure action Sampling pressure

RECORDED TESTER POSITION

3
RECORDED LOG

F i n a l shul-in pressure Hydroltotic head

Fig. 2.2. Formation interval tester. (Courtesy of Schlumberger Well Services Company.)

DRILL-STEM TEST

11

the oil. At some point down the column a representative formation-water sample can be found. The point is variable and will be influenced by rock characteristics, mud pressure, type of mud, and duration of the test. It is best to sample the water after each stand of pipe is removed. Normally, the total dissolved solids content will increase downwards and become constant when pure formation water is obtained, if the concentration continues t o increase all the way t o the bottom, no representative sample can be obtained. A test that flows water will give even higher assurance of an uncontaminated sample. If only one drill-stem test water sample is taken for analysis, it should be taken just above the tool, as this is the last water to enter the tool and is least likely t o show contamination. Fig. 2.1 and 2.2 illustrate two drill-stem test tools with their various components. Fig. 2.1 illfistrates a Halliburton Company tool; Fig. 2.2 illustrates a Schlumberger Well Services Company tool. Other companies supply equally adequate tools, and reference t o specific brands throughout this test is made for identification only and does not imply endorsement by the US. Bureau of Mines. The drill-stem test can provide pressure head and head decline and buildup data useful in permeability calculation (Bredehoeft, 1965) and other information for the determination of additional reservoir conditions, such as the gas/oil ratio and reservoir depletion (McAlister et al., 1965). A stratigraphic interval of interest is isolated in the drilled hole by use of packers attached t o the drill string. Opening the tester valve in the test string allows the formation fluid to enter the drill pipe. Pressures are recorded by gages in the bottom of the test tool. To insure that a representative sample is obtained, the pH, resistivity, and chloride content of samples taken at intervals down the drill pipe can be determined. Usually a transition zone will be found below which apparently uncontaminated formation water will be located. The pH, resistivity, and chloride content will vary above the transition zone, and they will become constant below it. The sample taken for analysis in the laboratory can yield positive evidence of contamination, if present. The two most indicative tests are pH and the color of a filtered sample. If the filtered sample remains tan or brown and the color cannot be removed even with pressure filtration, it probably is contaminated with drilling-mud filtrate. A sample can be placed on a white-spot plate for color evaluation. For positive identification of the presence of mud filtrate, a sample of the drilling mud used in drilling the well can be obtained and allowed t o react with distilled water, the reacted water .is analyzed to determine the mud-contributed ions, and the suspected contaminated sample is analyzed t o determine if it contains these ions. Analyses of water obtained from a drill-stem test of Smackover Limestone water in Rains County, Texas, show how errors can be caused by improper sampling of drill-stem test water. Analyses of top, middle, and bottom samples taken from a 15-m fluid recovery are shown in Table 2.11. These data show an increase in salinity with depth in the drill pipe, indicating that the first water was contaminated by mud filtrate (Noad, 1962). The middle

12 .1 TABLE 2 1

SAMPLING SUBSURFACE OILFIELD WATERS

Drill-stem test recovery of Smackover Limestone water Constituent Concentration (mg/l) top Sodium Calcium Magnesium Bicarbonate Sulfate Chloride Total dissolved solids middle bottom

29,600 8,100 600 500 2,000 59,900 101,000

43,500 13,100 900 500 1,300 91,800 151,000

71,800 22,400 1,400 400 500 154,000 251,000

sample is approximately half mud filtrate and half formation water. The bottom sample is the most representative of Smackover water. No single procedure is universally applicable for obtaining a sample of oilfield water. For example, information may be desired concerning the dissolved gas or hydrocarbons in the water, or the reduced species present such as ferrous or manganous compounds. Sampling procedures applicable to the desired information must be used. Sample containing dissolved gases Knowledge of certain dissolved hydrocarbon gases is used in exploration. Methane is quite soluble in water, but samples must be collected in a sampler that keeps the subsurface pressure on the sample until it is opened in the laboratory. The testing tool is kept open until the head of water in the drill pipe is equalized with the formation pressure or until water flows at the surface. The pressure equalization may require 4 or more hours. However, a surface recording subsurface pressure gage can be lowered into the drill pipe to determine when the pressure has equalized. After equalization of pressure, formation-water samples can be obtained by lowering a subsurface sampler into the drill pipe (Buckley et al.,1958). Zarrella et al. (1967) determined the content of dissolved benzene. For this it is not necessary t o use a subsurface sampler; the samples are caught in buckets on opening the pipe string, and immediately transferred from the buckets t o new narrow-necked glass or metal containers. A preferred method of obtaining a sample for subsequent gas analysis is t o catch the aqueous sample in a metal container of about one-quart capacity. This sample is immediately transferred to another metal sample container. The second container should be filled completely t o the top, then the sides of the can are lightly squeezed t o allow for fluid expansion, and the lid is sealed tightly. A foil-lined (not plastic) lid should be used. If possible, the

SAMPLING AT THE FLOW LINE

13

sample should be analyzed immediately. If this is not possible, cool or freeze the sample. Sampling at the flow l n ie Another method of obtaining a sample for analyses for dissolved gases is to place a sampling device in a flow line. Fig. 2.3 illustrates such a device.

container

Valve P i p e line

I -

Valve

Rubber tube

Fig. 2.3. Flow-line sampler.

The device is connected to the flow line, and water is allowed t o flow into and through the container, which is held above the flow line, until 10 or more volumes of water have flowed through. The lower valve on the sample container is closed and the container removed. If any bubbles are present in the sample, the sample is discarded and a new one is obtained.

14

SAMPLING SUBSURFACE OILFIELD WATERS

Sampling at the wellhead It is common practice in the oil industry to obtain a sample of formation water from a sampling valve at the wellhead. A plastic or rubber tube can be used to transfer the sample from the sample valve into the container. The source and sample container should be flushed t o remove any foreign material before a sample is taken. After flushing the system, the end of the tube is placed in the bottom of the container, and several volumes of fluid are displaced before the tube is slowly removed from the container and the container is sealed. Fig. 2.4 illustrates a method of obtaining a sample at the wellhead. An extension of this method is to place the sample container in a larger container, insert the tube to the bottom of the sample container, allow the brine to overflow both containers, withdraw the tube, and cap the sample under the fluid. At pumping wellheads the brine will surge out in heads and will be mixed with oil. In such situations a larger container equipped with a valve at the bottom can be used as a surge tank or an oil-water separator, or both. To use this device, place the sample tube in the bottom of the large container, open the wellhead valve, rinse the large container with the well fluid, allow the large container t o fill, and withdraw a sample through the valve at the bottom of the large container. This method will serve to obtain samples that are relatively oil-free.
We1 l h e a d

O i l and water

Fig. 2.4. Schematic of method of obtaining a sample at the wellhead.

Sample for determining unstable properties or species The pH, Eh, and various species of elements are unstable and will change with changes in pressure and temperature, and when the formation water is exposed to the atmosphere. The pH of the sample will change because of the oxidation of reduced species, because of release of dissolved gases, and because of hydrolysis reactions such as:
c03-*

+ HCO,-+

H+

H+

H,CO,

SAMPLE FOR STABLE-ISOTOPE ANALYSIS

15

Because the pH of the formation water sample will change, the pH should be determined using a flowing sample. A pH/Eh flow sampling chamber (Collins, 1964) is shown in Fig. 2.5. The Eh determination is difficult and
for corroboration it should be checked using a knowledge of the dissolved

Fe+*/Fe+3ratio of the water.


Ther mocompensator

Fig. 2.5. Flow chamber for use in determining pH and Eh at the wellhead.

Ferrous iron will oxidize to ferric and should be determined immediately after collecting a fresh sample. Some of the other dissolved constituents that should be determined immediately after securing a fresh sample are oxygen, hydrogen sulfide, thiosulfate, and manganous manganese.

Sample for stable-isotope analysis


A sample that is t o be analyzed for stable isotopes should be collected with care. If possible, such a sample should be taken at reservoir temperatures and pressures to minimize any isotope fractionation. However, because this usually is impossible, caution should be exercised t o insure that a representative sample is collected at the prevailing wellhead temperature and pressure. The sample should be collected at the wellhead. If this proves impossible,

16

SAMPLING SUBSURFACE OILFIELD WATERS

it may be feasible t o collect the sample from a nonheated separator or heater; samples are not to be taken of water that has been heated or treated with any chemicals. Glass sample bottles (about 100 ml usually is sufficient) should be used, and the sample should overflow the bottle. The bottle should be closed with a cap equipped with a plastic insert, and the top should be sealed with wax to minimize exchange reactions with air. Sample containers Various factors influence the type of sample container that is selected. Containers that are used include polyethylene, other plastics, hard rubber, metal cans, and borosilicate glass. Glass will absorb various ions such as iron and manganese, and may contribute boron or silica to the aqueous sample. Plastic and hard rubber containers are not suitable if the sample is to be analyzed to determine its organic content. A metal container is used by some laboratories if the sample is to be analyzed for dissolved hydrocarbons such as benzene.
TABLE 2.111 Description needed for each petroleum-associated water sample
__

--

Sample number Field Farm or lease Well No. in the of Section Township Range __ County State Operator Operators address (main office) -___ Date Sample obtained by -_____ Address Representing Sample obtained from (lead line, separatory flow tank, etc.) Elevation of well ___ Completion date of well Name of productive zone from which sample is produced Sand -Shale Lime Other Names of formations Name of productive formation well passes through Bottom of formation Depths: Top of formation Top of producing zone Bottom of producing zone Total depth drilled Present depth Bottom hole pressure and date of pressure Bottom hole temperature b e any chemicals If yes, Date of last workover added to treat well? -what? Well production Initial Present Casing service record: Oil, barrelslday Water, barrels/day _______ Gas, cubic feetlday Method of production (primary or secondary)

__

Remarks: (such as casing leaks, communication, or other pays in same well, lease, or field)

TABULATION OF SAMPLE DESCRIPTION

17

The type of container selected is dependent upon the planned use of the analytical'data. Probably the more satisfactory container, if the sample is to be stored for some time before analysis, is the polyethylene bottle. All polyethylenes are not satisfactory because some contain relatively high amounts of metals contributed by catalysts in their manufacture. The approximate metal content of the plastic can be determined using a qualitative emission spectrographic technique. If the sample is transported during freezing temperatures, the plastic container is less likely to break than glass. The practice of obtaining two samples and acidifying one sample so that the heavy metals will stay in solution works better if the plastic container is used. Some of the heavy metals are adsorbed by glass even if the sample is acidified. Tabulation of sample description The sample is of little value if detailed information concerning it is not available. Information such as that in Table 2.111 should be obtained for each sample of petroleum-associated water, and for certain types of studies, additional information may be needed. References
Bredehoeft, J.D., 1965. The drill-stem test: the petroleum industry's deep-well pumping test. Ground Water, 3:15-23. Buckley, S.E., Hocott, C.R. and Taggart, Jr., M.S., 1958.Distribution of dissolved hydrocarbons in subsurface waters. In: L.C. Weeks (Editor), Habitat of Oil. American Association Petroleum Geologists, Tulsa, Okla., pp.850-882. Collins, A.G., 1964. Eh and pH of oilfield waters. Prod. Monthly, 28:ll-12. McAlister, J.A., Nutter, B.P. and Lebourg, M., 1965. A new system of tools for better control and interpretation of drill-stem tests. J. Pet. Technol., 17 :207-214. Noad, D.F., 1962. Water analysis data, interpretation and applications. J. Can. Pet. TechnoL , 1 :82-89. Wallace, W.E., 1969. Water production from abnormally pressured gas reservoirs in South Louisiana, J. Pet. Technol., 21 :969-982. Zarrella, W.M., Mousseau, R.J., Coggeshall, N.E., Norris, M.S. and Schrayer, G.T., 1967. Analysis and significance of hydrocarbons in subsurface brines. Geochim. Cosmochim. Acta, 31 :1155-1166.

This Page Intentionally Left Blank

Chapter 3. ANALYSIS OF OILFIELD WATERS FOR SOME PHYSICAL PROPERTIES AND INORGANIC CHEMICAL CONSTITUENTS

Water analyses are used by the petroleum industry in studies related to subsurface formation identification, pollution problems, water compatibilities, corrosion, water-quality control, waterflooding, and exploration. Efforts to standardize methods applicable to analyzing oilfield waters have been made by the American Petroleum Institute (1968),and currently similar efforts are being made by the American Society for Testing and Materials. The methods discussed in this chapter include wet chemical procedures for calcium, magnesium, barium, carbon dioxide, sulfide, sulfur compounds, selenium, oxygen, spent acid, fluoride, chloride, bromide, and iodide. Instrumental methods are described for pH, Eh, specific gravity, resistivity, suspended solids, acidity, alkalinity, oxygen isotopes, ammonium nitrogen, phosphate, boron, arsenic, copper, nickel, lead, manganese, zinc, cadmium, and silica. Also described are emission and atomic absorption methods for lithium, sodium, potassium, rubidium, cesium, magnesium, calcium, barium, manganese, zinc, copper, iron, and lead; and emission spectroscopic methods for aluminum, beryllium, boron, iron, manganese, and strontium. ._ The methods used to analyze oilfield waters should be capable of producing precise and accurate results. Methods applicable to analyzing fresh waters may or may not be directly applicable to a petroleum-associated water, but in general such a method will need modification or complete redevelopment because the petroleum-associated water contains a more complex and concentrated array of dissolved salts than the fresh water. Quality control Data provided by the analytical laboratory are used in decision-making, and the data must describe precisely and accurately the characteristics or concentrations of the constituents in the sample. Usually an approximate or incorrect result is less valuable than no result because it leads t o faulty interpretations. The analyst needs t o be aware of his responsibility to provide results that reliably describe the sample. Further, he should know that the procedures that he uses, his professional competence, and his reported values may be used or challenged. To meet any challenge his results must be adequately

20

ANALYSIS OF OILFIELD WATERS

documented. The value of research investigations which use oilfield brine analyses depends upon the validity of the laboratory results. A program to insure the reliability of analytical data is mandatory because of the importance of the laboratory results and the actions that they produce. An established routine control program applied to analytical tests is necessary to assure the precision and accuracy of the final results. The use of spiked samples can measure quality, while the use of analytical-grade reagents is a control measure. Quality control varies with the type of as*sis. For example, the frequent standardization of the titrant used in a titration is an element of quality control, while instrument calibration in an instrumental method is also a quality control function. The specific methodology employed should be carefully documented regardless of the method used; thus the data user or reviewer can apply the associated precision and accuracy when interpreting the laboratory data.

Choosing an analytical method


Widespread use of an analytical method indicates that it probably is reliable and will produce valid results. Use of a little-known procedure forces the data user t o accept the judgment of the analyst. The following criteria are useful in selecting analytical methods: (1)The desired constituent should be measured with sufficient precision and accuracy in the presence of the interferences normally found in oilfield waters. (2) The method must utilize the skills and equipment available in the oilfield water laboratory. (3) The method should be sufficiently tested and used by several laboratories to establish its validity.

Precision
Precision is the reproducibility among replicate observations, and in quality control it is determined on actual water samples containing interfering constituents. Several methods to determine precision are available and the following is representative: (1) Study four separate concentration levels, including a low concentration near the sensitivity level of the method, two intermediate concentrations, and a concentration near the upper limit of application of the method. (2) Make seven replicate determinations at each of the concentrations tested. (3) To allow for changes in conditions, the precision study should use at least 2 hours of normal laboratory operation. (4) To permit the maximum interferences in sequential operation, the samples should be run in the following order: high, low, intermediate, intermediate. Repeat this series seven times to obtain the desired replication.

QTJALITY CONTROL TABLE 3 1 . Precision data on oilfield brine samples for boron Sample
.

21

___
~~

Concentrations of boron found (mg/l) Taylor


-

Eagle Ford

_ _ _ _-

Paluxy

Douglas
-

10.1 1. 01 10.2 10.3 10.1 10.2 10.2


Average Standard deviation

15.2 15.3 15.1 15.2 15.3 15.2 15.1 15.2 01 .

20.1 20.1 20.3 20.2 20.3 20.3 20.1 20.2 01 .

30.3 30.2 30.1 30.1 30.3 30.2 30.1 30.2 0.2

10.2 01 .

(5) The precision statement includes a range of standard deviations over the t A e d range of concentrations. Thus, four standard deviations are obtained over a range of four concentrations, but the statement contains only the extremes of standard deviations and concentrations studied. An example of data generated from such an approach is shown in Table 3.1. Using the data of Table 3.1 the precision statement would read: In a single ldboratory, using oilfield water samples containing concentrations of 10.2 and 30.2 mg B/1, the standard deviation was kO.1.
Accuracy
The degree of difference between observed and actual values is accuracy. The accuracy of a method can be determined as follows: (1) Add known amounts of the constituent t o be determined to actual samples at concentration levels where the precision of the method is adequate. The added amount should double the concentration of the lowconcentration sample and bring the concentration of an intermediate sample to about 75% of the upper limit of application of the method. (2) Make seven replicate determinations at each concentration. (3) Report the accuracy as the percent recovery found in the spiked sample, where the percent at each concentration is the mean of the seven replicate tests. Table 3.11 illustrates the application of this approach, where two of the samples used in the precision study, Table 3.1, were used. An appropriate accuracy statement is: In a single laboratory, using oilfield water samples g containing concentrations of 20.2 and 35.3 m B/1, recoveries were 100.0% and 100.3%,respectively.

22

ANALYSIS OF OILFIELD WATERS

TABLE 3.11 Accuracy data on oilfield brine samples for boron Sample Concentrations of boron found (mg/l) Taylor (added 10 mg/l boron)
20.2 20.2 20.1 20.1 20.3 20.3 20.4

Paluxy (added 15 mg/l boron)


35.5 35.4 35.2 35.2 35.3 35.2 35.1 35.3 35.3

Average Percent recovery

20.2

20-2 ] x 100 = 100.0 [ 10.2 + 10

[ 20.2 + 15

] x 100 = 100.3

The precision and accuracy data are valuable in determining that the analyst and the method are capable of generating valid data. Once this is proven, the data can be used to evaluate systematic performance. This can be done by using spiked samples about 10% of the time to determine that the accuracy is favorable, and evaluating replicate samples to determine that the precision is favorable. Preliminary sample treatment The following determinations should be made in the field immediately after sampling: (1) temperature (in "C),(2) pH, (3) dissolved oxygen, (4) resistivity, ( 5 ) acidity, (6) alkalinity, (7) sulfide, and (8) carbon dioxide. If possible, the oilfield water sample should be filtered immediately after sampling in the field. A preferred method-is to use pressure filtration through a 0.45-micrometer (pm) membrane filter. A liter of filtrate usually is sufficient and the following determinations can be made on aliquots: (1) iodide, (2) bromide, (3) chloride, (4) selenium, ( 5 ) sulfate, (6) nitrogen, (7) phosphate, (8) silica, (9) boron, (10) potassium, (11) sodium, and (12) lithium. If a field-filtered sample cannot be provided, a laboratory-filtered sample may be substituted with slightly less confidence in the reported data.

PRELIMINARY SAMPLE TREATMENT

23

Standard solutions
Examples of standardization procedures are given for some of the methods. The concentrations of standard solutions are indicated as the weight of a given element equivalent to, or contained in, 1 ml of solution. The strength of acids and bases are given in terms of molarities or normalities.

Accuracy of measurements
In the instructions for making the analysis and preparing the solutions, significant figures are utilized to define the accuracy of weights and measures. Required accuracy for measurement of volume in the analysis and preparation of reagents is also shown. Standard solutions are prepared in and measured from volumetric glassware.

Reagent chemicals and solutions


All of the chemicals used in the analytical procedures should conform to the specifications of the Committee on Analytical Reagents of the American Chemical Society. Chemicals not listed by this organization can be tested according to procedures given by Rosin (1955). Primary standard chemicals can be obtained from the National Bureau of Standards or from companies marketing chemicals of the same purity. Water used to dilute samples or t o prepare chemical solutions should first be demineralized by passage through mixed cation-anion exchange resins or by distillation. Its specific conductance a t 25C must not exceed 1.5 pmho/ cm, and it should be stored in polyethylene bottles. Carbon-dioxide-free water may be prepared by boiling and cooling demineralized water immediately before use. Its pH should be between 6.2 and 7.2. Ammonia-free water should be prepared by passing distilled water through a mixed-bed ion-exchange resin.

Sampling
A field-filtered acidified sample also should be taken. It is pressured filtered using a 0.45-pm membrane filter and then the filtrate is immediately acidified to a pH of 3.0 or less with reagent-grade HCl. The acidified sample is used for the following determinations: (1) aluminum, (2) arsenic, (3) barium, (4) cadmium, (5) calcium, (6) copper, (7) iron, (8) lead, (9) magnesium, (10) manganese (11)nickel, (12) strontium, and (13) zinc.

24 TABLE 3.111
Units in which water analyses may be reported milligrams per liter = mg/l part per million = ppm milligrams per liter ppm = specific gravity of the water

ANALYSIS OF OILFIELD WATERS

1 grain per U.S. gallon = 17.12 mg/l 1 grain per Imperial gallon = 14.3 mg/l 1 ppm = 0.012 milligram atom per liter

T o convert compounds expressed as parts per million t o ions expressed,as parts per million (where compound is A, Bm):
ppm ion A = ppm compound A, ppm ion B = ppm compound A, ,(atomic weight A) Bm molecular weight A, B, ,(atomic weight B) Bm molecular weight A, , B

To convert parts per million t o equivalents per million (epm): Example: sample contains 28.3 ppm Ca , what is the concentration of calcium in epm? Solution; atomic weight Ca = 40.08; valence = 2; equivalent weight =

40.08 2
28.3 ppm Ca+ =

= 20.04; then: = 1.41 epm Ca+

28 3 20.04

Titrime tric analysis


(ml x N of standard solution) x milliequivalent weight of determined ion ml of sample used
106

= mg/l of determined ion

Gravimetric analysis
(grams of preninitnta\ .,.F.U,

, atomic weightweigdetermined element . molecular of ,ht of precipitate Y

106

. . . _ .

ml of sample used
%O

= me/l determined .. ._.~

element

= parts per thousand or g/kg = mass in grams of silver required to precipitate the halogens in

Chlorinity (CZ) Salinity (S)

328.5233 g of sea water


= total amount of solid material, in grams contained in 1 kg of

sea water when all of the bromide and iodide have been replaced by the equivalent amount of chloride, when all of the carbonate is converted t o oxide and when all the organic matter is completely oxidized
%o

S = 1.805 x

%o

Cl + 0.03

REPORTING THE ANALYTICAL RESULTS

25

Reporting the analytical results


A study conducted by the American Petroleum Institute (1968) indicated that some laboratories reported the results of oilfield water analysis as parts per million (ppm) or as milligrams per liter (mg/l) without regard to the specific gravity of the sample. For example, a sample with a specific gravity of 1.200 containing 12,000 mg/l of calcium does not contain 12,000 ppm of calcium but contains 12,000/1.200 = 10,000 pprn of calcium. Such an error obviously is more serious in reporting the analytical results for a brine than in reporting the results for a fresh water. The unit ppm means parts per million by weight, while the unit mg/l means milligrams per liter or weight per unit volume; therefore, they are not interchangeable until the volume is changed to a unit weight. Table 3.111 indicates the relation between various units of measurement. Because the American Petroleum Institute now recommends that oilfieldwater analysis be reported in units of mg/l, other associations will no doubt recommend the same uniform practice. Such standardization implements studies concerned with the chemistry and geochemistry of waters.

Sign i f ican t figures


The term significant figure (Ballinger et al., 1972) is used rather loosely to describe some judgment of the number of reportable digits in a result. Often the judgment is not soundly based and meaningful digits are lost or meaningless digits are accepted. Proper use of significant figures gives an indication of the reliability of the analytical method used. The following definitions and rules are suggested for retention of significant figures. A number is an expression of quantity. A figure or digit is any of the characters 0, 1, 2, 3, 4, 5, 6, 7, 8, 9, which, alone or in combination, serves to express a number. A significant figure is a digit that denotes the amount of the quantity in the place in which it stands. Reported values should contain only significant figures. A value is made up of significant figures when it contains all digits known to be true and one last digit in doubt. For example, if a value is reported as 18.8 mg/l, the 18 must be a firm value while the 0.8 is somewhat uncertain and may be 0.7 or 0.9. The number zero m a y or may not be a significant figure: (a) Final zeros after a decimal point are always significant figures. For example, 9.8 g t o the nearest milligram is reported as 9.800 g. (b) Zeros before a decimal point with other preceding digits are significant. With no other preceding digit, a zero before the decimal point is not significant. (c) If there are no digits preceding a decimal point, the zeros after the

26

ANALYSIS OF OILFIELD WATERS

decimal point but preceding other digits are not significant. These zeros only indicate the pbsition of the decimal point. (d) Final zeros in a whole number may or may not be significant. A good measure of the significance of one or more zeros before or after another digit is to determine whether the zeros can be dropped by expressing the number in exponential form. If they can, the zeros are not significant. For example, no zeros can be dropped when expressing a weight of 100.08 g in exponential form; therefore, the zeros are significant. However, a weight of 0.0008 g can be expressed in exponential form as 8 x g, and the zeros are not significant. Significant figures reflect the limits of the particular method of analysis. It must be decided beforehand whether this number of significant digits is sufficient for interpretation purposes. If not, there is little that can be done within the limits of normal laboratory operations t o improve these values. If more significant figures are needed, a further improvement in method or selection of another method will be required t o produce an increase in significant figures. Once the number of significant figures is established for a type of analysis, data resulting from such analyses are reduced according t o set rules for rounding off.

R o unding-of f numbers
Rounding off of numbers is a necessary operation in all analytical areas. It is automatically applied by the limits of measurement of every instrument and all glassware. However, it is often applied in chemical calculations incorrectly by blind rule or prematurely, and in these instances can seriously affect the final results. Rounding off should normally be applied only as follows.

Round ing-of f rules (a) If the figure following those to be retained is less than 5 , the figure is dropped, and the retained figures are kept unchanged. As an example, 11.443 is rounded off to 11.44. (b) If the figure following those to be retained is greater than 5, the figure is dropped, and the last retained figure is raised by 1 As an example, 11.446 . is rounded off to 11.45. (c) When the figure following those t o be retained is 5 , and there are no figures other than zeros beyond the 5, the figure is dropped, and the last place figure retained is increased by 1 if it is an odd number, or it is kept unchanged if an even number. As an example, 11.435 is rounded off t o 11.44,while 11.425 is rounded off to 11.42. Rounding-off single arithmetic operations (a) Addition: when adding a series of numbers, the sum should be rounded off to the same number of decimal places as the addend with the

SYNTHETIC BRINE

27

smallest number of places. However, the operation is completed with all decimal places intact, and rounding off is done afterward. As an example, 11.1+ 11.12 + 11.13 = 33.35, and the sum is rounded off t o 33.4. (b) Subtraction : when subtracting one number from another, rounding off should be completed before the subtraction operation, to avoid invalidation of the whole operation. (c) Multiplication: when two numbers of unequal digits are t o be multiplied, all digits are carried through the operation; then the product is rounded off to the number of significant digits of the less accurate number. (d) Division: when two numbers of unequal digits are t o be divided, the division is carried out on the two numbers using all digits. Then the quotient is rounded off t o the number of digits of the less accurate of the divisor or dividend. (e) Powers and roots: when a number contains n significant digits, its root can be relied on-for n digits, but its power can rarely be relied on for n digits. Synthetic brine Synthetic brine solutions are used in many of the analytical procedures for analyzing oilfield waters (American Petroleum Institute, 1968). Such solutions are a necessity in the development of analytical methods to study the effects of possible interfering ions. Often these synthetic solutions are used as an integral part of the analytical technique (Collins, 1967). Preparation of a fairly stable synthetic brine involves saturating distilled water with carbon dioxide by bubbling carbon dioxide through it, followed by adding the bicarbonate and sulfate compounds to one portion of the C 0 2-saturated water, adding the alkali chlorides to one portion, and adding the alkaline earth chlorides to one portion. The alkali chloride solution is mixed with the bicarbonate-sulfate solution, and t o this mixture the alkaline earth chloride solution is added. Carbon dioxide is bubbled through the synthetic brine to mix it, and the synthetic brine container is sealed immediately after removing the carbon dioxide source. Determination of pH The pH of the water can be determined with a pH meter which utilizes the principle of measuring the electrical potential between an indicator electrode and a reference electrode (Potter, 1956, p.56). pH meters measure the electrical potential between two suitable electrodes immersed in the solution to be tested. The reference electrode assumes a constant potential, and the indicating electrode assumes a potential dependent on the pH of the solution. Electrode potential is the difference in poteptial between the electrode and the solution in which it is immersed. The calomel electrode, which is a widely used reference electrode in water analysis, consists of a mercurycalomel rod immersed in a saturated solution of potassium chloride; this

28

ANALYSIS OF OILFIELD WATERS

electrode has a potential of +0.246 V. Electrical connection with the sample is provided through porous fibers sealed into the immersion end. A hydrogen-ion-selective glass electrode is normally used as an indicating electrode. The glass electrode has several features that recommend it for pH measurements. Among the most important are that it is not affected by oxidizing or reducing substances in the sample and that it can be used to measure the pH of turbid samples and/or colloidal suspensions. The basic design is a silver-silver chloride or mercury-mercurous chloride electrode immersed in a solution of known pH and the whole completely sealed in glass. The mechanism by which the glass membrane responds to hydrogen-ion activity involves absorption of hydrogen ions on both sides of the membrane proportionally to the activity of the hydrogen ions in solution. The cell for measuring the pH of a solution is of the following type: Ago :AgC1

1I

solution of known pH;

glass membrane;

glass electrode

I 1 Hgo :HgC1

solution of unknown pH

The voltage of the glass electrode is a logarithmic function of the difference in hydrogen-ion activity of the solutions on either side of the glass membrane. To measure this voltage an electron-tube voltmeter is used because the resistance of the glass membrane is so great. The pH should be determined at the time of sampling. A device similar to that shown in Fig. 2.5, can be used, or the electrodes can be placed in a container and then a stream of the sample allowed to flow from the oilwater separator (Fig. 2.4.) into the container while the pH is measured. If accurate results are desired, at least two pH buffer solutions should be used to calibrate the pH meter and electrodes before determining the pH. Because
TABLE 3.IV pH buffer solutions (pH values of NBS standards from 0-30C) Temperature 0.5M potassium ("C) tetroxalate Potassium acid tartrate (sat. at 25OC)
0.05M potassium acid phthalate 0.025M potassium dihydrogen phosphate + 0.025M sodium dihydrogen phosphate 6.98 6.92 6.90 6.88 6.86 6.85 0.01M sodium tetraborate 9.46 9.33 9.27 9.22 9.18 9.14

0 10 15 20 25 30

1.67 1.67 1.67 1.68 1.68 1.69

4.01 4.00

4.00
4.00 4.01 4.01

3.56 3.55

DETERMINATION O F Eh TABLE 3.V Performance characteristics of typical pH meters Normal scale


~~~

29

Expanded scale 1 pH (* 100 mV)

Range Smallest scale division Accuracy Reproducibility Temperature compensation Input impedance -

0-14 pH (+ 1,400 mV) 0.1 pH (10 mV) f 0.05 pH (5 5 mV) f 0.02 pH (+ 2 mV)

0.005 pH (0.5 mV) f 0.002 pH (+ 2%of reading) f 0.002 pH (+ 0.2 mV)

>

O-IOO~C (manual or automatic) 1014

> 1013

the pH probably will fall between 5 and 7, the standard pH buffer solutions used could be for pH 5 and pH 7. Standard buffer solutions, covering a range of pH, may be purchased from almost any chemical supply house and are satisfactory for routine use. Table 3.IV gives a list of NBS buffers (easily made in the laboratory) and the resulting pH at several different temperatures. An idea of the effect of temperature on pH may be obtained by observing temperature versus pH of various buffers shown in Table 3.IV. Theoretically, the potential response of the electrode system changes 0.20 mV per pH unit per degree Celsius. Since all pH meters measure potential but read out in pH, a variable compensation is used. A rough rule of thumb is that temperature compensation is about 0.05 pH unit per 5' increase in temperature. Performance data of a conventional and an expanded scale pH meter are shown in Table 3.V. Determination of Eh The Eh, called the oxidation-reduction potential or the redox potential, is a measure of the relative intensity of oxidizing or reducing conditions in a chemical system. It is expressed in volts, and at equilibrium it is related to the proportions of oxidized and reduced species present. Standard equations of chemical thermodynamics express the relationships (Collins, 1964). Eo is the standard potential of a redox system when unit activities of participating substances are present under standard conditions. Eo is related to standard free energy change in a reaction by the equation:

A P

= -nfEo

where n is the number of unit negative charges (electrons) shown in the redox reaction and f is the Faraday constant in units that give a potential in volts (94,484absolute coulombs). Standard free energy values are given in texts such as that of Latimer (1952).

30

ANALYSIS OF OILFIELD WATERS

When the system is not under standard conditions, the redox potential is expressed by the Nernst equation: RT E h = E o + - log oxidized species) nf (reduced species) where R is the gas constant (1.987 calories per degree mole) and T is the temperature in degrees Kelvin. Geochemical literature and biochemical literature such as that of Pourbaix (1949) use increasing positive potential values to represent increasing oxidizing systems, and decreasing potential values to represent reducing systems. The sign of Eh used in this manner is opposite to standard American practice in electrochemistry. Reagents. An Eh standard' which can be used is a solution of M/300 K3Fe(CN)6 and M/300 KqFe(CN), in M/10 KC1 (Zobell, 1946). The Eh of this mixture is 0.430 V at 25'C. Equipment. A pH meter equipped with a thermometer, a glass electrode, a calomel electrode, a platinum electrode and a thermocompensating electrode. Eh flowchamber, a design similar to Fig. 2.5 can be used. Procedure. Buff the platinum electrode lightly with a fine abrasive cloth and wipe it carefully with a dry soft tissue. Install the glass electrode, the calomel electrode, the platinum electrode, the thermocompensator, and the thermometer in the flowchamber. Standardize the instrument using the Eh standard. Connect a line t o the wellhead or waterline and install an oil-water separator if oil and water both are present. Connect the flowchamber to the waterline, allowing the water to flow into the bottom and out the top. Make certain that all air bubbles are excluded at the top. Take at least three readings of the Eh (in mV), and the temperature at 10-minute intervals. These readings should agree; if they do not, continue making readings until three successive readings do agree. Make certain the water is continually flowing, that there are no air bubbles in the flowchamber, and that the solution is being stirred. It may be necessary t o remove and rebuff the platinum electrode. Calculation. Because a thermocompensator is used in determining the pH, a temperature correction need not be made. However, if a thermocompensator is not used, a temperature correction should be made. The Eh value is obtained by algebraically adding the measured voltage E and the voltage of the constant voltage reference electrode, which in this case is the saturated calomel electrode. The potential of the saturated calomel electrode at 25OC is 0.242 V. Therefore, if the millivolt reading of the sample is +300:

SUSPENDED SOLIDS

31

Eh = E + voltage of reference electrode Eh= +300+242=+542 However, if the millivolt reading is -300, then:

E = +0.242 - 7.6x

( t - 24)

Note: the calculations for Eh are correct only if the temperature of the brine is 25OC at the time of measurement. If the temperature is not 25"C, a correction should be made. For example, the potential of the saturated calomel electrode is 0.246 V at 20C and 0.238 V at 30C. The following formula can be used to obtain the correct potential:

= +0.242 - 7 6 x .

( t - 24)

where t is in degrees Celsius. Suspended solids Various inorganic and organic materials are found in petroleum-associated water. Knowledge of the composition of such material is useful in determining the source of the material and what treatments can be used t o remove it or prevent it from recurring. Such material may be particles of oxides of the metals from well casings, pumps, or precipitates caused by oxidation of the formerly reduced species, such as iron or manganese. Other suspended solids may be silt, sand, and clay. An estimation of the amount of material in suspension can be accomplished by using a turbidimeter (Rainwater and Thatcher, 1960).This is done by comparing the intensity of light passing through the solution with the Tyndall effect produced by lateral illumination of the solution with the same source of light.

Turbidimeter
Instruments for the measurement of turbidity employ principles of design related to transmission or reflectance of light. The lack of a primary standard for turbidity, however, has resulted in a complete absence of uniformity among the available instruments. Further, the Jackson candle turbidimeter, which does not depend upon the use of a primary standard, is a primitive instrument, subject to many interferences, and the measurements generally are not reproducible. Recently developed turbidimeters often use for calibration a suspension of formazin permanently embedded in a cylinder of Lucite. These instruments produce reproducible readings up t o 40 Jackson candle units (JCU), and samples containing turbidities in excess of 40 JCU should be diluted to

32

ANALYSIS O F OILFIELD WATERS

values below this level and the results multiplied by the correct dilution factor. To obtain maximum accuracy and precision the following precautions should be observed: (a) Protect the Lucite standard from scratches, nicks, and fingerprints. (b) While calibrating the instrument, use a constant orientation of the Lucite standard. (c) Use a homogeneous sample in the sample cuvette; do not take readings until finely dispersed bubbles have disappeared. (d) Dilute samples containing. excess tubidity to some value below 40 JCU; take reading, and multiply results by correct dilution factor.
Suspended solids analysis

To determine the composition of the suspended solids they can be removed by filtration using a 0.45-pm membrane or less porous filter. The filtered solids can then be subjected to chemical analysis. To determine the exact composition of the solids may require the filtration of a large sample in order to procure enough solid material. The heavy-metal content can be determined by subjecting a portion of the sample t o an emission spectrometric analysis; X-ray diffraction can be used to determine which, if any, clays are present; extraction with organic solvents followed by infrared mass spectrometric, chromatographic, and gas chromatographic analysis will give an indication of organic compounds present; thermogravimetric analysis will provide clues; wet chemical analysis can be used t o determine many of the anions; and X-ray fluorescence can be used to determine some of the anions. Resistivity The resistivity of petroleum-associated waters is used in electric log interpretations (Wyllie, 1963), and for such use the values must be adjusted t o the formation temperature. This can be done by referring t o curves such as those shown in Fig.3.1, which gives resistivity values for sodium chloride solutions. The resistivity of a formation water will not be exactly the same as that of a pure sodium chloride solution of equal dissolved solid (DS) content, but for practical purposes the assumption that the resistivities are approximately equal is satisfactory. It is possible to calculate the resistivity from water-mineral analysis by using methods such as those developed by Dunlap and Hawthorne (1951) or Jones (1944). The calculated values are less accurate and usually lower than the directly measured resistivities. The direct-measurement method is essentially the electrical resistance of a cube of oilfield water. In well-logging practice, the edge of the cube considered is 1 m in length. Therefore, resistivity of an oilfield water is expressed in ohm-meters ( a m ) . Temperature has a profound effect on resistivity; therefore, all resistivities

RESISTIVITY

33

R E S I S T I V I T Y . .hm-n.ttrl

Fig. 3.1. Plots of resistivity of aqueous solutions containing various concentrations of sodium chloride.

should be determined at a known constant temperature. The sample should be freshly filtered and free of oil. Nonionized silica and other materials in suspension in an oilfield water can affect the resistivity determination, but in general such interferences can be ignored. Cell polarization can be troublesome with highly mineralized waters and will vary directly with the current that flows between the electrodes and inversely with the frequency of the current. High input voltage t o the bridge or low cell resistance (highly mineralized waters) increases the likelihood of polarization. Cell resistance can be increased by increasing the cell constant. Reagents. The necessary reagents are standard potassium chloride solutions of l.OOON, O.lOOON, and 0.01OON (use only certified reagent-grade KC1 that has been oven-dried t o constant weight at 110OC);chromic-sulfuric acid cleaning solution; platinizing solution (dissolve 3 g of chloroplatinic acid and 0.02 g of lead acetate in 100 ml of water); and a 10%aqueous sulfuric acid solution. Equipment. The necessary resistivity measurement equipment includes a Wheatstone bridge; resistivity cells, either dip or pipet type, with platinum

34

ANALYSIS OF OILFIELD WATERS

electrodes; water bath, complete with stirrer, thermostat, and thermometer, with 0.loC graduations; source of alternating current, 25- t o 60-cycle a.c. galvanometer, and an appropriate isolating transformer. Selection of the cell constant is limited by the accuracy and sensitivity of the bridge when measuring very high and very low resistivities. Also, current frequency should not be excessively high since a.c. resistance is a complex function of frequency; e.g., at frequencies necessary to avoid polarization, the differences between a.c. resistance and d.c. resistance may be appreciable unless the cell has been carefully designed t o minimize this difference. In essence, the ideal single apparatus for measurement of resistivity throughout a wide range necessarily incorporates compromises between low input voitage, high cell constant, high current frequency, and accuracy and sensitivity of the bridge.

Cell preparation
To prepare the cell, clean it with chromic-sulfuric acid solution and rinse thoroughly with water. Immerse the cell or fill it, depending upon whether a dip or pipet cell is used, in the platinizing solution. Connect the electrodes of the cell to three dry cells (1-1/2 V each) in parallel through a limiting resistance of approximately 1,000 52. Reverse the direction of the current once a minute for 6 minutes or until the shiny platinum surface is covered with a dense black coating. Repeat the electrolytic process using 10% sulfuric acid solution to remove chlorine. Remove the electrodes, rinse with distilled water, and store in distilled water. Note: new cells should be cleaned and platinized before use. They should be cleaned and replatinized whenever the readings become erratic or when the platinum black flakes off.

Cell resistance
To determine the cell resistance using the standard potassium chloride solutions, adjust the temperature of each potassium chloride solution to exactly 25OC and obtain a reading in ohms for each solution with the cell. Calculate the cell constant using the following formula:

C = R K C l x specific conductance of standard KC1 solution


where R K C l = reading obtained in ohms for standard KC1 solution. Note: the specific conductivities of the standard KC1 solutions a t 25C are as follows (Hodgman et al., 1962, p. 2690):

1.OON KCl = 0.11173 mho/cm 0.1ON KC1 = 0.012886 mho/cm 0.01N KCl = 0.0014114 mho/cm

SPECIFIC GRAVITY

35

Method of determination Procedure. To determine the resistivity of the petroleum-associated water, filter the sample to remove oil and transfer the sample t o the cell or cell container and place it in a water bath. Allow sample sufficient time to adjust to bat!i temperature, and measure resistance of sample and record the temperature to nearest 0.1Oc. Calculation. The resistivity calculation is dependent upon the type of cell and bridge used, but in general the following formula will apply:

R,

=-

D2 V

4LXT

where R, = resistivity of water, a m ; V = difference in potential between potential-measuring electrodes, V; I = current flowing through the cell, A; D = inside diameter of potential-measuring electrodes, m; and L = distance between potential-measuring electrodes, m. Because D and L are constant for any one cell and I is held constant for most waters, these values can be combined into a single constant, K , and the following simplified equation used :

R,

= KV

Calculated resistivity
The resistivity of petroleum-associated waters often is calculated using the laboratory analysis (Dunlap and Hawthorne, 1951). The concentrations of the ionic constituents are used in the calculation method. Dunlap and Hawthorne (1951) caution users of their calculation method that the sulfate factor 0.50 may give unreliable results if the water contains appreciable concentrations. of sulfate. If the sulfate concentration exceeds 2,500 mg/l, a factor of 0.40 will give a better calculated resistivity value. Specific gravity Specific gravity is the ratio of the weight of a given volume of material to the weight of an equal volume of some other material used as a standard (Mellon, 1956, p.306), and pure water is the usual standard for liquids and solids. Depending upon the accuracy desired, the specific gravity of a petroleum-associated water can be determined with a pycnometer, specific gravity balance, or hydrometer. Because any oil in or on the sample will interfere with the specific gravity determination, the sample should be filtered.

TABLE 3.W Approximate relation of specific gravity (Sp. gr.) t mg/l of dissolved solids (DS) o

Sp. gr.
1.ooo

DS
0 1,400 2,800 4,200 5,600 7,000 8,300 9,700 11,100 12,400 13,700 15,200 16,600 17,800 19,100 20,500 21,900 23,200 24,500 25,900 27,300 28,500 29,800 31,000 32,400 33,900 35,100 36,400 37,700 39,100 40,400 41,700 43,000 44,300 45,600 46,900 48,300 49,500

Sp.gr.

DS
'50,800 52,000 53,300 54,600 55,900 57,100 58,300 59,600 60,900 62,100 63,400 64,600 65,900 67,100 68,400 69,600 70,900 72,000 73,300 73,600 75,800 77,100 78,200 79,400 80,600 81.800 83,100 84,300 85,600 86,700 87,800 89,100 90,300 91,500 92,700 93,900 95,100 96,200

Sp. gr.

DS
97,400 98,700 99,800 101,000 102,200 103,400 104,600 105,800 106,900 108,100 109,300 110,400 111,600 11 2,800 114,000 115,100 116,200 117,400 118,600 119,600 120,800 122,000 123,100 124,400 125,500 126,700 127,800 128.800 130,000 131,100 132,300 133,400 134,500 135,600 136,800 137,900 139,100 140.1 00

Sp.gr.

DS
141,200 142,300 143,400 144,500 145,600 146,700 147,900 148,900 150.00 0 151,100 152,100 153,200 154,400 155,500 156,600 157,700 158,800 159,900 161,000 162,000 163,100 164,100 165,200 166,200 167,300 168,400 169,400 170,400 171,500 172,500 173,600 174,700 175.7 00 176,800 177,900 178,900 180,000 181,100

Sp. gr.

DS
182,100 183,200 184,200 185,300 186,300 187,400 188,400 189,500 190,500 191,600 192,600 193,600 194,700 195,700 196,700 197,800 198.800 199,800 200,900 201,900 202,900 203,900 204,900 206,000 207,000 208,000 209,000 210,000 211,000 212,000 213,000 214,000 215,000 216,000 217,000 218,000 219.000 22o;ooo

Sp.gr.

DS
221,000 222,000 223,?00 224,000 225,000 226,000 227,000 228,000 229,000 230,000 230,800 231,800 232,800 233,700 234,700 235,700 236,700 237,600 238,600 239,500 240,500 241,500 242,400 243,400 244,300 245,300 246,200 247,700 248,100 249,100 250,000 250,900 251,900 252,800 253,800 254,700

1.001 1.002 1.003 1.004 1.005 1.006 1.007 1.008 1.009 1.010 1.011 1.012 1.013 1.014 1.015 1.016 1.017 1.018 1.019 1.020 1.021 1.022 1.023 1.024 1.025 1.026 1.027 1.028 1.029 1.030 1.031 1.032 1.033 1.034 1.035 1.036 1.037

1.038 1.039 1.040 1.041 1.042 1.043 1.044 1.045 1.046 1.047 1.048 1.049 1.050 1.051 1.052 1.053 1.054 1.055 1.056 1.057 1.058 1.059 1.060 1.061 1.062 1.063 1.064 1.065 1.066 1.067 1.068 1.069 1.070 1.07 1 1.072 1.073 1.074 1.075

1.076 1.077 1.078 1.079 1.080 1.081 1.082 1.083 1.084 1.085 1.086 1.087 1.088 1.089 1.090 1.091 1.092 1.093 1.094 1.095 1.096 1.097 1.098 1.099 1.100 1.101 1.102 1.103 1.104 1.105 1.I06 1.107 1.108 1.109 1.110 1.111 1.112 1.113

1.114 1.115 1.116 1.117 1.118 1.119 1.120 1.121 1.122 1.123 1.124 1.125 1.126 1.127 1.128 1.129 1.130 1.131 1.132 1.133 1.134 1.135 1.136 1.137 1.138 1.139 1.140 1.141 1.142 1.183 1.144 1.145 1.146 1.147 1.148 1.149 1.150 1.151

1.152 1.153 1.154 1.155 1.156 1.157 1.158 1.159 1.160 1.161 1.162 1.163 1.164 1.165 1.166 1.167 1.168 1.169 1.170 1.171 1.172 1.173 1.174 1.175 1.176 1.177 1.178 1.179 1.180 1.181 1.182 1.183 1.184 1.185 1.186 1.187 1.188 1.189

1.190 1.191 1.192 1.193 1.194 1.195 1.196 1.197 1.198 1.199 1.200 1.201 1.202 1.203 1.204 1.205 1.206 1.207 1.208 1.209 1.210 1.211 1.212 1.213 1.214 1.215 1.216 1.217 1.218 1.219 1.220 1.221 1.222 1.223 1.224 1.225

3 m

TITRIMETRIC METHODS

37

Knowledge of the specific gravity of the sample is necessary to convert the analytical data determined for the sample from milligrams per liter to parts per million. In addition, the specific gravity will give an indication of the amount of dissolved solids present in the sample, as indicated in Table 3.VI.
TITRIMETRIC METHODS

Acidity, alkalinity, and borate boron If the pH of the water is less than 4.5, the water possesses what is called mineral-acid acidity. The acidity of a petroleum-associated water may indicate a contaminant because of acid treatment of the well or it could indicate the presence of various dissolved gases and salts. Most petroleumassociated waters contain little or no acidity. If a water contains acidity, it does not contain alkalinity. The acidity of a water is determined by adding a standard base such as 0.02N sodium hydroxide to the water until the pH of the water is 4.5 (Collins et al., 1961) as monitored with a pH meter. To obtain a value close to natural conditions, the acidity should be determined at the sampling point. The alkalinity of a water is determined by adding a standard acid such as 0.05N hydrochloric acid t o the water and recording the volume used to neutralize it to pH 8.1 and pH 4.5. The amounts of hydroxide, carbonate, and/or bicarbonate can then be calculated using the relationships shown in Table 3.VII. Because the alkalinity will change when the sample is exposed to the atmosphere, the alkalinity should be determined as rapidly as possible after sampling.

TABLE 3.VII
Relationships for determining alkalinity after neutralization with a standard acid Volume of standard acid used OH
P=O P = 1/2T P = 1/2T P > 1/2T P=T

co3
0 2P 2P 2(T-P) 0

HC03

<

0 0 0 2P-T T

T T - 2P 0 0 0

P = volume used t o titrate t o pH 8 . 1 ; T = volume used t o titrate t o pH 4.5 plus volume used to titrate to pH 8.1.

38

ANALYSIS OF OILFIELD WATERS

Reagents. The necessary reagents are standard hydrochloric acid, standard sodium hydroxide, pH buffer solutions (preferably for pH 4, 7, and lo), mannitol, and nitrogen gas. Equipment The necessary equipment are a pH meter, 10-ml microburets, boron-free glassware, and boron-free reflux condensers. Standardization of 0.02N sodium hydroxide
The 0.02N sodium hydroxide solution should be standardized t o determine its exact normality. One of the better methods is to standardize it with potassium acid phthalate. Obtain a potassium acid phthalate sample of known purity, such as a National Bureau of Standards standard sample, and dry the salt at 105C. Weigh 0.1 g of salt, dissolve it in 50 ml o distilled f water, and titrate with sodium hydroxide to a pH of 7.0. Reaction:
HOOC ,
NaOH +

,
/

,COONa ChH,
-

-ChH, COOK

+ H,O

KOOC

Normality calculation: weight KHCs H, 0, = 0.20422 x ml NaOH

Standardization of 0.05N hydrochloric acid


If constant boiling-point hydrochloric acid is not used in preparing 0.05N hydrochloric acid, the normality should be checked. One method is t o use a potassium iodate sample of known purity; for example, a National Bureau of Standards standard sample. Dry the salt at 180C for 2 hours, weigh 0.1 g, dissolve it in 50 ml of distilled water, add 2 g of potassium iodide and 2 g of sodium thiosulfate, and titrate with hydrochloric acid t o a pH of 7.0. Reaction:

KI03 + 5KI = 6HC1 I2 + 2Na2S203


Normality calculation : weight KI03 0.03567 x ml HC1

6KC1+ 312 + 3H20 Na2S406+ 2NaI

N=

TITR I ETR IC METHODS M

39

Procedure. Calibrate the pH meter to pH 4 and 7 with appropriate buffer solutions .and recheck the calibration often. Transfer an undiluted 50- or 100-ml sample t o a beaker and determine the pH and record it. If the pH is above 8.1, titrate it to 8.1 with 0.05N hydrochloric acid and record the titer for the carbonate calculation. Continue the titration to pH 4.5 and record the titer for the bicarbonate calculation. If salts of organic acids are in the water sample, special precautions must be taken to separate the bicarbonate titer from that required for the organic salts. This may be done by extracting the acids, usually naphthenic, with a neutral organic solvent such as petroleum ether. If the initial pH is below 4.5,titrate to pH 7 with 0.02N sodium hydroxide and record the titer for the acidity calculation. . Next reduce the pH to 3 5 with 0.05N hydrochloric acid, and reflux the sample 5 minutes. Remove the sample and immediately cool in an ice-water bath. Carefully adjust the pH of the cooled brine to 7 with sodium hydroxide while nitrogen is aspirated gently over the top. Add 10 g of mannitol and titrate the sample back to pH 7 with 0.02N sodium hydroxide. Record this titer for the borate boron calculation. If more than 1 mg of boron is present in the titrated sample, the results may be low. Calculations. If the initial pH was more than 8.1, the titer for carbonate and bicarbonate is determined:
m1 HC1
ml sample

30y000= mg carbonate per liter

To convert carbonate t o bicarbonate, multiply the carbonate value by 2.03. If the initial pH is less than 8.1 but more than 4.5,only bicarbonate is present :
ml HC1 x N x 61,000 = mg bicarbonate per liter ml sample

If the initial pH is below 4.5,the brine is acid:


ml NaOH x N x 50,000 = acidity as mg CaCO, per liter ml sample If the total titration is equal to the titer found to pH 8.1, only hydroxide is present: ml HC1 x N x 50,000 = acidity as mg CaC0, per liter mg sample The borate boron is calculated by using the titer for borate boron: ml NaOH x N x 10,820 = mg borate boron per liter mg sample

40

ANALYSIS OF OILFIELD WATERS

Calcium and magnesium Probably the most common method currently used to determine calcium and magnesium in waters is the complexometric method (American Petroleum Institute, 1968)which utilizes a salt such as disodium ethylenediaminetetraacetic acid (EDTA) or disodium 1,2-cyclohexanediaminetetraacetic acid (CDTA) to chelate calcium or magnesium. At a pH of 10,both calcium and magnesium are chelated, while at a pH of 12, only calcium is chelated because magnesium will precipitate as the hydroxide. Disodium ethylenediaminetetraacetate has the following structure (Welcher, 1957,p.128):
- CHI
\

0 -CO

CHI - C O - 0

Na

N - (CH2) - N
CH,

, Na
H
/

\ 0 - CO - CH, /

- CO - 0

Its molecular weight is 372.254, and it forms 1:l complexes with most cations according t o the following equations: Me+ + H2Y- Me+3 + H2Y- Me+4 + H2Y-

* MeY- * MeY* MeY

+ 2H+ + 2H+ + 2H+

where Me = the cation, H2Y = EDTA, and MeY = the complex. Therefore, 1 gram-ion of EDTA reacts with 1 gram-ion of the metal, regardless of its valence. The resulting complexes have the same composition, differing only in the charge they carry. A metal indicator in an EDTA titration can be represented by the following expression : M-In + EDTA

* M-EDTA + In

where M-In = the metal indicator complex, M-EDTA = the metal-EDTA complex, and In = the indicator. The metal indicator complex must be weaker than the metal-EDTA complex. The color change occurs because the metalindicator complex ionizes, and the free metal is completely complexed by the EDTA, leaving a free indicator.

Sample size
Because many petroleum-associated waters contain high concentrations of dissolved solids including calcium and magnesium, it usually is necessary t o dilute them or to use a micropipet t o obtain a small sample before performing a complexometric titration. The dilution and subsequent aliquot

TITRIMETRIC METHODS TABLE 3.VIII Aliquot size for total hardness, calcium, and magnesium determination Specific gravity
1 .ooo--1.010 1.010-1.025 1.025-1 .O50

41

Dilution none none dilute 25 ml to take 50 ml dilute 25 ml to take 25 ml dilute 25 ml to take 25 ml dilute 25 ml to take 25 ml

Aliquot (ml)
50 25 12.5 6.25 1.25 0.625

100 ml, 100 ml, 500 ml, 1,000 ml,

1.050-1.090 1.090-1.1 20 1.120-1.150

TABLE 3.IX Comparison of errors in direct reading of sample size using a micropipet versus the dilution technique Direct-reading using micropipet sample size error (ml)
~~

Dilution technique ml taken after diluting 10 ml to 100 ml


2 5 10 10 20 50

error (ml)
f 0.01
f 0.025

200 h 500 h 1,000 h 1ml 2 ml 5 ml

f 0.004

f 0.01

* 0.002
f 0.006 f 0.006
f 0.01

f 0.005

* 0.005 f 0.009 * 0.02

size usually can be determined by using data such as that illustrated in Table 3.VII1, which is applicable to most oilfield brines. A more rapid method of obtaining a fraction of a milliliter of a liquid sample is direct measurement using a micropipet. A micropipet in the hands of a competent analyst can also yield a more accurate sample size than the dilution technique illustrated in Table 3.VIII; e.g., two reading errors are omitted because only one meniscus reading is necessary with the direct measurement as compared to three using the dilution. Table 3.IX illustrates a comparison of errors in sample sizes of the direct reading method versus the dilution method. Reagents: CDTA (disodium 1,2-cyclohexanediaminetet,raacetic acid) standard solution, approximately 0.025M:dissolve 10.66 g CDTA in water and dilute

42

ANALYSIS OF OILFIELD WATERS

to 1liter. (To standardize, dissolve 2.4971 g calcium carbonate in the smallest amount of hydrochloric acid possible and dilute to 1 liter with water (0.025M Ca solution). Pipet triplicate 10-ml aliquots into flasks and dilute to approximately 50 ml with water. Begin at paragraph 2 of the procedure, and carry the standard solutions through all the steps of that paragraph. Calculate the molarity using the following equation: M=

10.0 x 0.025 0.25 or V V

where V = volume of CDTA required to titrate 1 0 ml of 0.025M calcium solution. Sodium hydroxide: dissolve 320 g sodium hydroxide pellets in water and dilute to 1liter. Ammonium chlorideammonium hydroxide buffer solution : dissolve 67.5 g ammonium chloride in approximately 200 ml of water. Add 570 ml concentrated ammonium hydroxide and dilute to 1 liter with water. l-hydroxy-2-naphthylaz0)-6Eriochrome Black T indicator (sodium 1-( nitro-2-naphtol-4-sulfonate): dissolve 0.5 g of the indicator and 4.5 g of hydroxylamine hydrochloride in 100 ml of water. 2-hydroxy-l-naphthylazo)-2-naphthol-4Calcon: dissolve 0.4 g sodium 1-( sulfonate in 100 ml of methanol. Triethanolamine: dilute 30 ml triethanolamine to 100 ml with water.

Procedure. Filter the sample to remove undissolved solids and traces of oil from the water. Transfer, by means of Lambda pipet or volumetric transfer pipet, an aliquot of sample containing not more than 10 mg of calcium into an Erlenmeyer flask. Dilute t o approximately 50 ml with water. (No more than 10 ml of standard CDTA are t o be used in a titration for either calcium or magnesium.) Add two to three drops of triethanolamine solution and approximately 4 ml of sodium hydroxide solution. The pH of the solution at this point should be 12.0-12.5. Add six drops of calcon indicator and titrate with standard CDTA solution until the indicator blue endpoint is reached. Record the volume of CDTA titrant used t o titrate calcium. Using the same pipet, pipet another aliquot into another Erlenmeyer flask and dilute t o 50 ml with water. Add two to three drops of triethanolamine solution, 3-5 ml of the ammonium chloride ammonium hydroxide buffer solution, and three to four drops of Eriochrome Black T indicator. Refill the buret with the same standard CDTA titrant and titrate the sample until the color changes from wine to pure blue. This endpoint is sometimes delayed, so proceed cautiously with the titration near the endpoint. Record the volume of CDTA used t o titrate calcium and magnesium, or hardness (Ca + Mg).

TITRIM ETRIC METHODS

43

Calcula t ions :

B x M x 100,100 sample volume


A x M x 40,000 sample volume

= mg total hardness as CaC03 per liter

= mg

Ca per liter

(B-A) x M x 24,300 sample volume

mg Mg per liter

where A = calcium titer (blank); B = hardness titer (blank); and M = molarity of CDTA. The complexometric determination of calcium and magnesium will give a precision of about 2% of the amounts present. The accuracy is dependent upon the interferences present, and the major interferences are strontium and barium, both of which will be complexed along with calcium, thus producing high results. In the absence of strontium and barium the accuracy of the method is about 4% of the amount of calcium and magnesium present. Ammonium nitrogen Organic compounds containing nitrogen decompose in a reducing environment and form ammonia and the ammonium ion. A reducing environment is characteristic of a petroleum genetic environment (Collins et al., 1969). Bogomolov et al. (1970) call it an indicator of petroleum. The ammonium ion is too weak an acid t o be successfully titrated; however, when treated with formaldehyde, hexamethylenetetramine and a strong acid are produced. This strong acid can be titrated with a base using indicators or a potentiometer to determine the endpoint. Reagents. The necessary reagents are hydrochloric acid, 12N; sodium hydroxide standard, OJN, and 0.02N; formaldehyde. and Equipment. The necessary equipment includes an expanded-scale pH meter, a hotplate, microburets, flasks, and an ice-water bath. Procedure. The method should not be used if less than 5 mg/l of ammonium nitrogen is present. Acidify the brine or water when sampling t o a pH of about 1.5 with 12N HC1. The acid will stabilize the sample by changing to the ammonium ion any ammonium hydroxide which could volatilize as ammonia and be lost. Transfer a 100-ml aliquot of the acidified sample to a 250-ml Erlenmeyer flask and boil the solution for 5 minutes on a hotplate. Cool the solution as quickly as possible to about 25C. An ice bath will facilitate .rapid adjustment to this temperature.

44

ANALYSIS OF OILFIELD WATERS

Using an expanded-scale pH meter, adjust the pH of the sample to 7.0 with 0.1N NaOH, add 5 ml of 37 wt.% formaldehyde solution, and heat the mixture t o 4OoC. (Do not exceed this temperature.) Cool immediately t o ambient temperature using an ice water bath, and titrate the sample with 0.02N NaOH to pH 8.6 using an expanded-scale pH meter t o detect the endpoint The weak hydroxide titrant must be protected from atmospheric carbon dioxide, and a reagent blank must be determined because formaldehyde contains formic acid. The reactions are :

6HCHO + 4NH4 C1 (CH2 )6 N4 + 4HC1+ 6H2 0 HC1+ NaOH + NaCl + H20


--f

Calculation:
(ml NaOH x N used for sample) -

(ml NaOH x N used for reagent blank) x 14,007 = mg/l NH4N ml sample

Chloride
A modification of the Mohr method (Furman, 1962) is satisfactory for the determination of chloride in petroleum-associated waters. Common interferences are bromide, iodide, sulfide, and iron. Sulfide can be removed by acidifying the sample with nitric acid and boiling. Iron can be removed by ion exchange or precipitation with sodium hydroxide or sodium peroxide followed by filtration. Because most petroleum-associated waters contain high concentrations of chloride, it usually is necessary to dilute the sample before titrating with silver nitrate, because the voluminous precipitate masks the endpoint. About 50 mg of chloride is maximum for a satisfactory titration. The indicator usually is potassium chromate or sodium chromate, and at the endpoint the chromate ion combines with excess silver to form the slightly soluble red silver chromate: Ag+ + C1+

AgCl
--f

ZAg+ +

Ag2Cr04

The specific gravity of the sample can be used t o estimate the correct aliquot size. Table 3.X indicates aliquot sizes that will contain less than 50 mg of chloride. The micropipet can be used as demonstrated in the calcium-magnesium procedure and Table 3.IX.

TITRIMETRIC METHODS TABLE 3.X Aliquots that contain less than 50 mg of chloride as estimated from the specific gravity Specific gravity 1.000-1.002 1.003-1.004 1.005-1.012 1.01 3-1.019 1.020-1.032 1.033-1.064 1.065-1.087 1.088-1.162 1.163 Dilution none none dilute 10 ml to dilute 10 ml to dilute 10 ml to dilute 25 ml to dilute 20 ml to dilute 10 ml to dilute 10 ml to Aliquot (ml) 100 50 5.0 2.0 1.0 0.5 0.4 0.2 0.1

45

>

100 ml, take 50 ml 100 ml, take 20 ml 100 ml, take 1 0 ml 500 ml, take 10 ml 500 ml, take 10 ml 500 ml, take 10 ml 1,000 ml, take 10 ml
-

Reagents. The necessary reagents include silver nitrate, standard solution, 0.05N; potassium or sodium chromate, neutral 5% aqueous solution; and nitric acid, 0.1N (nitrous free); and sodium bicarbonate. Equipment. The necessary equipment includes a hotplate, a 10-ml microburet, flasks, and pipets.

Procedure. After removal of interferences and selection of correct aliquot size, dilute the sample to 20 ml or more, adjust the pH to 8.3 with sodium bicarbonate or 0.1N nitric acid, add 1 ml of a 5% aqueous potassium chromate solution, and titrate with an 0.05N silver nitrate solution until the red endpoint just persists.
Calculation: ml AgN03 x N x 35,500 = mg,l clml sample The precision and accuracy of the method are about 1% and 2%, respectively, of the amount present. Bromide and iodide Bromide and iodide are present in almost all petroleum-associated waters. In the following procedure, iodide is selectively oxidized t'o iodate with bromine water; excess bromine is reacted with sodium formate. The iodate reacts with added iodide t o produce iodine which is titrated with thiosulfate. Hypochlorite is added to another sample to oxidize both bromide and iodide to bromate and iodate, respectively. Excess hypochlorite is reacted with sodium formate, and the bromate and iodate are reacted with iodide t o liberate iodine for titration with thiosulfate.

46

ANALYSIS OF OILFIELD WATERS

Iron, manganese, and organic matter can interfere but are removed in the procedure. Fluoride is added to mask interference from any remaining traces of iron.

Reagents. The necessary reagents include a 2% ammonium molybdate solution; glacial acetic acid; calcium hydroxide; calcium carbonate; 0.05N hydrochloric acid; 6N hydrochloric acid; potassium iodide; sodium fluoride; starch indicator solution; 0 . O l N sodium thiosulfate (standardize prior t o use); 3.8M sodium formate (prepare fresh daily); saturated bromine water; and methyl red indicator solution. Equipment. The necessary equipment includes a mechanical shaker, 200-ml bottles, a hot-water bath, flasks, pipets, and microburets. Procedure. To remove iron, manganese, and organic matter from the sample, add exactly 100 ml of sample to a stoppered bottle. Add 1 g of calcium hydroxide, and place the mixture in a shaker for 1 hour. Allow the mixture to stand overnight and filter through a dry folded filter, discarding the first 20 ml that comes through. Brines with specific gravities of less than 1.009 may be filtered without standing overnight. Prepare a blank in the same manner. Transfer an aliquot of the filtrate containing 1-2 mg of iodide t o a 250-ml Erlenmeyer flask. Add sufficient water t o make the total volume 75 ml, and three drops of methyl red indicator. Add 0.05N hydrochloric acid until the mixture is just slightly acid, add 10 ml of sodium acetate solution, 1 ml of glacial acetic acid, and 4 ml of bromine water, and allow to stand for 5 minutes. Next add 2 ml of sodium formate solution, blow out any bromine vapor from the neck of the flask, and wash down the sides with water. When the solution is completely colorless, add 0.2 g of sodium fluoride and 0.5 g of potassium iodide. Mix until dissolved and add 15 ml of 6N hydrochloric acid. Titrate with 0.01N sodium thiosulfate using starch indicator. Disregard any return of blue color after the endpoint. Record this titration volume for the iodide calculation. Transfer another aliquot of the filtrate containing 1-2 mg of bromide t o a 250-ml Erlenmeyer flask and add sufficient water to make the total volume 75 ml. Add 10 ml sodium hypochlorite solution and approximately 0.4 g of calcium carbonate (or enough so that approximately 0.1 g will remain after the next step). Adjust the pH of the solution with 3N hydrochloric acid to between 5.5 and 6.0 and heat in a water bath t o 90C for 10 minutes. (A small amount of undissolved calcium carbonate should remain at this point.) Remove the flask and cautiously add 10 ml of sodium formate solution, return the flask to the water bath, and keep the contents hot for 5 minutes more and observe the timing very closely. Rinse down the inside of the flask with a few milliliters of distilled water and allow the solution to cool t o room temperature. (Do not use a cold water bath.) To the ambient solution

TITRlMETRIC METHODS

47

add three drops of ammonium molybdate solution, 0.5 g sodium fluoride (if iron is present), and 0.5 g potassium iodide, mix until dissolved, and acidify with 15 ml of 6N hydrochloric acid. Titrate with 0 . O W sodium thiosulfate using starch indicator. Disregard any return of blue color after the endpoint. Record this titration for the bromide calculation. Calculations. Iodide: ml of Na, S2O 3 for sample - ml of Na, S2O3 for blank corrected ml of Na, S2O 3 : (ml x N) N a 2 S 2 0 3x 21,150- mg/l Iml sample Bromide: ml of Na2S2O 3 for sample - ml of Na2 S2O3 for blank = corrected : ml of Na2S203 (ml x N) Na, S 2 0 3 x 13.320 - mg/l I- x 0.63 = mg/l Brml sample The precision and accuracy of the method are about 3%and 676, respectively, of the amounts of bromide and iodide present. Oxygen The solubility of a gas varies directly with pressure and inversely with temperature and usually is reduced by the presence of dissolved minerals. Most petroleum-associated waters contain little or no dissolved oxygen in situ at depth. Knowledge of the dissolved oxygen content of waters that are to be reinjected for waterflooding or disposal is needed to determine treatment required t o prevent corrosion. Instrumental and wet chemical methods (American Petroleum Institute, 1968) are available for the determination of dissolved oxygen. Instrumental methods usually are modifications of the rotating platinum electrode method (Marsh, 1951), but with them the residual current (when no oxygen is present) is difficult to determine. The modified Winkler method probably is the most accurate wet chemical method available (Watkins, 1954). In the Winkler method for quantitatively determining dissolved oxygen in water, a glass-stoppered bottle is completely filled with the water to be tested. Manganous sulfate (MnS04) and potassium hydroxide (KOH) are added, forming a precipitate of manganous hydroxide (Mn(OH), ) in accordance with the following reaction: MnS04 + 2KOH + Mn (OH), + K2S04 The manganous hydroxide combines with the oxygen dissolved in the water to form a higher oxide of uncertain composition, assumed t o be manganese hydroxide (MnO(OH), ), as follows: 2Mn (OH), + 0,
+

MnO (OH),

48

ANALYSIS OF OILFIELD WATERS

On acidification in the presence of an iodide, the higher oxide of manganese liberates a quantity of iodine stoichiometrically equivalent t o the quantity of dissolved oxygen present in the sample in the following manner:
Z MnO (OH), + 2H2 SO4 +. Mn(S04 ) + 3H20 Mn(S04), + 2KI +. MnS04 + K2S04 + I2

The quantity of iodine liberated is determined by titrating an aliquot portion of the sample with a standard solution of sodium thiosulfate (Na2S203) using starch solution as an indicator, as shown by the equation: 2Na2S, O3 + I,
+ Na, .

S40, + 2NaI

The iodine modification of the Winkler method depends upon the conversion of any hydrogen sulfide t o hydrogen iodide and free sulfur by reducing the iodine added t o the brine. This reaction proceeds as follows: H,S+I2 + 2 H I + S Tests have shown that interfering substances other than hydrogen sulfide that might be present in oilfield brines also are counteracted by the iodine added.

Reagents. It is important to use sterile glassware or polyethylene bottles in preparing and storing reagents for this test to prevent contamination and t o make longer storage of reagents possible without appreciable changes in their normality. Iodine solutions, 0.5N and 0.W. Hydrogen sulfide water: saturate distilled water (which has been boiled and cooled recently t o drive off dissolved oxygen) with hydrogen sulfide gas. Starch solution. Manganous sulfate solution: dissolve 480 g of manganous sulfate 0) 0) (MnS04 *4H2 or 400 g of manganous sulfate (MnS04 *2H2 in distilled water, filter, and dilute to 1 liter. Alkaline iodide solution: dissolve 700 g of potassium hydroxide (KOH) or 500 g of sodium hydroxide (NaOH) and 150 g of potassium iodide (KI), or 135 g of sodium iodide (NaI).in distilled water, and dilute the solution t o 1 liter. If a white carbonate precipitate is formed, separate the precipitate by settling and then siphoning off the supernatant liquid. The solution should give no color with starch indicator when diluted and acidified, which indicates the absence of nitrates, iodates, and ferric salts. Sulfuric acid, concentrated. Sodium thiosulfate solution, 0.1N. Standard sodium thiosulfate solution, 0.025N.

TITRIMETRIC METHODS

49

Equipment. The necessary equipment includes glass-stoppered bottles, pipets, flasks, and microburets.
Sa mp 1ing Care must be taken t o obtain uncontaminated samples of water for determining dissolved gases. Glass-stoppered bottles should be used for sample containers. To determine dissolved oxygen in water, 300-ml bottles with pointed, ground-glass stoppers and overflow lips of the type used for biochemical oxygen-demand tests are particularly suitable. These bottles are so designed that samples may be obtained without contamination by atmospheric oxygen and so the necessary chemical reagents may be introduced during the analysis without excessive overflow from the lip of the bottle. Before a sample is taken, rinse the bottle three times with the water to be sampled and fill through a rubber tube extending to the bottom of the bottle. A quantity of water equal t o at least three times the capacity of the bottle should be allowed to overflow the bottle, and the rubber tube should be withdrawn slowly so that the space in the bottle occupied by the tube is filled simultaneously with water. The glass stopper, when placed in the mouth of the bottle, will displace all excess water. If any bubbles are seen, the sample is immediately analyzed. If the temperature of the water taken for analysis of dissolved gases is above 2OoC, a cooling coil should be used to cool the sample before the water enters the bottle. It is important that the samples contain no included atmospheric oxygen or carbon dioxide, as errors may be introduced into many of the analyses if extraneous oxygen or carbon dioxide is present in the water.

Procedure. All reagents in the following steps 1 through 8 should be added slowly and carefully under the surface of the water near the bottom of the bottle, using pipets, permitting the displaced water t o overflow the top of the bottle. The quantities of reagents added should be recorded for use in the final calculation. After each reagent is added, the stopper should be carefully replaced and the bottle inverted gently several times so as not to introduce air into the bottle while adding and mixing reagents. Collect the sample as described previously. Add excess 0.5N iodine solution to give the sample a yellow color and let stand 5 minutes. Add saturated hydrogen sulfide water until the sample is a very light straw-yellow, and 1ml of starch solution as an indicator. Add dilute hydrogen sulfide water until the blue color just disappears and then add, drop by drop, 0.1N iodine solution until a faint blue color persists. Add 1 ml of manganous sulfate solution, 1 ml of alkaline iodide solution, and 1ml of concentrated sulfuric acid, letting it run down the neck of the bottle. Transfer 200 ml of the solution by pipet from the sample bottle t o a 500-ml Erlenmeyer flask. Titrate the 200-ml sample in the Erlenmeyer flask with 0.025N sodium thiosulfate solution. The starch indicator should be

50

ANALYSIS OF OILFIELD WATERS

added when the yellow color of free iodine has been almost eliminated by the sodium thiosulfate titration, and the titration should be continued until one drop changes the solution from a light blue to colorless. (Subsequent blue recoloration should be disregarded.) If no hydrogen sulfide or other interfering substances are present, the first six steps of the determination may be eliminated, using only the part of the procedure starting with the addition of the alkaline iodide solution. Calculation. The dissolved oxygen content of the water is determined by the following equations:

v = 200 x-(Y-1) u=-200 w


V
where U = dissolved oxygen content, ppm; V = volume of sample titrated, ml; W = volume of 0.025N sodium thiosulfate required, ml; X = volume of sample bottle, ml; Y = total volume of all reagents added, ml; and 1 = the 1 ml of acid added, which does not change the effective oxygen-tested volume of the sample because it is added after all the oxygen has been absorbed. The factor used to take into account the volume of reagents added may involve a slight error, because it is based on the assumption that the reagents contain no dissolved oxygen.

Carbon dioxide
Petroleum-associated waters containing carbon dioxide and bicarbonate or carbonate will contain a weak acid H2C03 or its salt, which buffers the solution. This combination controls the pH of waters in the range of about pH 4.5-8.0. Such buffering is caused by the presence of slightly dissociated acids or bases, and when H+ or OH- ions are added they first convert the undissociated acid or base to its salt or vice versa. Loss of carbon dioxide will disturb the carbon dioxide-bicarbonatecarbonate buffer systems. For example, the pH probably will change and precipitation of calcium carbonate or other compounds may occur. An increase in carbon dioxide will shift t h e . carbon dioxide-carbonatebicarbonate equilibria, allowing more material such as calcite t o go into solution. Bacterial reduction of sulfate can cause the amount of dissolved carbon dioxide and hydrogen sulfide in petroleum-associated waters t o be quite high. Several hundred milligrams per liter of C 0 2 can be present in such waters. Knowledge of the amount of carbon dioxide in solution is useful in carbonate equilibria studies (Garrels and Christ, 1965) and in water compatibility studies (Watkins, 1954).

TITR IMETR IC METHODS

51

Reagents. The necessary reagents are 0.05N sodium carbonate solution and phenolphthalein indicator solution. Procedure. Collect the water sample in the same manner used in taking the sample t o be analyzed for dissolved oxygen. Pipet 100 ml of the water into a flask and add five drops of phenolphthalein indicator. If the sample turns red, no free carbon dioxide is present; if it remains colorless, titrate the sample with the standard sodium carbonate solution to a red endpoint.
Calcula tion : ml Na2CO, x N x 22,000 = mg/l C02 ml sample Sulfide

As mentioned above, the bacterial reduction of sulfate causes some petroleum-associated waters t o contain appreciable concentrations of hydrogen sulfide. Knowledge of the amount of dissolved sulfide present is necessary for corrosion and water compatibility studies (Watkins, 1954). The following method depends upon the reduction of iodine by the hydrogen sulfide in the brine, as shown by the following equation: H2S+I2 + 2 H I + S
Because of the unstable nature of the hydrogen sulfide in solution in waters and brines, the sulfide is not titrated directly. To prevent the loss of hydrogen sulfide t o the air, an excess of iodine solution is added, and the sample is back-titrated with standard sodium thiosulfate solution, in accordance with the following equation: 2Na2 S2 0, + I2

+. Na2S4 O6 + 2NaI

Experiments conducted by the US. Bureau of Mines indicate that residual reducing agents that cannot be removed by aeration or boiling are present in some oilfield brines. Brine from the Arbuckle (siliceous) Limestone formation originally containing 96 mg/l hydrogen sulfide showed such residual reducing agents t o equal 9 mg/l of hydrogen sulfide after air has been bubbled through the brine for 28 hours. This dropped to 4 mg/l after standing another 24 hours. Further tests in which the hydrogen sulfide was driven off by boiling indicated the presence of 5 mg/l residual reducing agents. When the brine was neutralized with hydrochloric acid (using methyl orange indicator) before boiling, residual reducing agents equal t o 2 mg/l hydrogen sulfide remained.

52

ANALYSIS OF OILFIELD WATERS

Reagents. The necessary reagents are potassium iodide, standard sodium thiosulfate, 0.1N and 0.01N; standard iodine solutions, 0.1N and 0.01N; and starch indicator solution. Procedure. Collect the sample in a glass-stoppered bottle (approximately 200-mi capacity) in the manner previously described for dissolved oxygen. Analysis should be made as soon as possible after sampling. Pipet 5 ml of 0.1N or 0.01N standard iodine solution, depending upon the hydrogen sulfide concentration expected, into each of two Erlenmeyer flasks. It may be necessary to use a larger quantity of 0.01N solution if the hydrogen-sulfide content of the sample is high. Add approximately 1 g of potassium iodide crystals t o each flask. (This step usually may be omitted in determinations on brine samples because of the high mineral content of the water.) Add 50 ml of distilled water t o the flask to be used for a blank determination, and pipet 50 ml of the water sample into the other flask. Titrate both the distilled water blank and the water sample with standard sodium thiosulfate solution of the same normality as the iodine solution used, adding 1 ml of starch indicator near the end of the titration. Record the milliliters of thiosulfate used in each titration. Calculation. Subtract the milliliters of thiosulfate solution used for the sample from the milliliters used for the blank and use the difference in the following formula:
(ml x N ) I2 - (ml x N) N a 2 S 2 0 3x 17,000 = mg/l H 2 S ml sample

Sulfur compounds The redox potential of petroleum-associated waters indicates that sulfur compounds other than sulfate and sulfide may exist in solution. When the water is brought t o the land surface, the change in pressure and temperature will affect the redox potential and, if the sample is allowed t o come into contact with the atmosphere, the equilibria of the sample will start t o change immediately. Better methods are needed t o determine the composition of a water in situ. The following method can be used to gain a semiqualitative estimation of the amomts of thiosulfate, sulfite, and sulfide in a water.

Reagents. Zinc carbonate suspension: add zinc acetate to a solution of sodium carbonate, filter and wash the precipitate with several volumes of cold water. Prepare the zinc carbonate suspension by vigorously shaking the precipitate with water. The other reagents are glycerol iodine, 0.01N; sodium thiosulfate, 0.01N; starch indicator solution; glacial acetic acid; and formaldehyde.

FLAME SPECTROPHOTOMETRIC METHODS

53

Determination o f thiosulfate, sulfate, and sulfide Procedure. Collect a water sample as described in the dissolved oxygen procedure. Pipet 100 ml of the sample into a 300-ml flask, and add 20 ml of glycerol, 100 ml of an aqueous suspension of zinc carbonate, and 70 ml of distilled water. Shake the mixture vigorously for 1 minute, filter, and discard the precipitate. Pipet 50 ml of the filtrate into a flask and add 5 ml of formaldehyde, and 3 ml of acetic acid, add starch indicator and titrate to the blue endpoint with 0.01N iodine. Record the amount of iodine used to calculate thiosulfate (A). Pipet another 50-ml aliquot of the filtrate into another flask; add 0.01N iodine until the solution remains yellow. Add starch indicator and titrate to a colorless endpoint with 0.Ol.N sodium thiosulfate. Record the amount of iodine used for thiosulfate plus sulfite (B). Pipet 25 ml of water that was not treated with the zinc carbonate into a flask and add an excess of 0.Ol.N iodine, 3 ml of acetic acid, add starch indicator and titrate t o the colorless endpoint with 0.01N sodium thiosulfate, sulfite, and sulfide (C). Calculations. Milliliters iodine used in A = X ml

X ml

x N x 112,000 = mg/l S2 03-2 ml sample

Milliliters iodine used in A - milliliters iodine used in B = Y ml

Y ml x N x 40,000 = mg/l SO,-2 ml sample


Milliliters iodine used in C - milliliters iodine used in B = 2 ml
2 ml x N x 16,000

ml sample

= mg/l

S-

FLAME SPECTROPHOTOMETRIC METHODS

When a metal salt in solution is sprayed into a flame, the solvent evaporates and the salt decomposes and vaporizes, producing atoms. Some of these atoms can be raised t o an excited state by the thermal energy of the flame, although a major portion of the atoms present in the flame remain at the grourid state. The return of the excited atoms to the ground state results in the emission of radiant energy characteristic of the element atomized. The quantitative measurement of this radiation is the basis of emission flame spectrophotometry, and the essential difference between this form of analysis and classical arc-emission spectrography is the temperature of the source used to excite the atoms. Because the g a s a i r and gas-oxygen flames

54

ANALYSIS OF OILFIELD WATERS

are much cooler than the spark and arc sources used in spectrography, analysis by emission flame spectrophotometry is usually limited t o the more easily excitable elements - lithium, sodium, and potassium. Instrumentation requirements include: (1) A method of introducing the sample into the flame for vaporization. (2) A method of detecting and recording the radiation intensity emitted. (3) A method of selecting the correct wavelength, ordinarily a variable monochromator. A more complete discussion of the theory and instrumentation can be found in books by Burriel-Marti and Ramirez-Munoz (1957) and Dean (1960), as well as in publications of commercial instrument manufacturers. Lithium Lithium usually is calculated as a part of the sodium content in reporting the results of oilfield water analyses rather than being determined and reported separately. One of the more accurate methods t o determine lithium in petroleum-associated waters is the flame spectrophotometric method (Collins, 1962). Reagents. The reagents are lithium, standard solutions, 0.1 mg/ml and 0.01 mg/ml; and n-propanol. Equipment. The necessary equipment includes a flame spectrophotometer, 10-ml'microburets, and volumetric flasks. Preliminary calibration curves. Preliminary calibration curves are useful in determining approximately how much lithium is in the sample and in determining the optimum amount of standard lithium solution t o use in the analysis. Because n-propanol is easier t o work with, it usually is used; however, if additional sensitivity is needed, the acetone-n-amyl alcohol mixture can be used (Collins, 1965). To prepare the preliminary calibration curves, transfer t o 50-ml volumetric flasks aliquots of diluted standard lithium solution containing the following amounts of lithium: 0.01 mg, 0.05 mg, 0.1 mg, 0.15 mg, and 0.2 mg. Add 20 ml of n-propanol to each flask and dilute to volume with distilled water. Aspirate, burn, and record the emission intensity of each of these five standards at 670.8 mp and their background at about 665 mp. Record several peaks for each standard at various sensitivity levels and slit widths. Plot the results on linear graph paper by plotting milligrams of lithium versus intensity. Prepare a curve for each sensitivity level and slit width used, as illustrated in Fig. 3.2. The sensitivity of the instrument will determine the optimum concentrations of lithium and this will require some experimentation. The analyst may find it convenient t o scan all the emission lines of

FLAME SPECTROPHOTOMETRIC METHODS

55

26 -

'I

2 220 0

- 20-x 18L 5

0.01 mm r l i t ~ 1,620 volta to I T 1 F W 6836 tOppri 02 5 p s i C2H2 12.5 mm burner height

10

mg L i / m l 5 0 % n - P R O P A N O L

Fig. 3.2. Preliminary calibration curves for use in selecting optimum standard additions: 5 Instrument: 0.01-mm slit, 1,620 V to ITT FW 6836, 1 0 psi 02, psi C2HZ,and 12.5-mm burner height.

interest; e.g., lithium, sodium, potassium, rubidium, cesium, and perhaps others. This will give information concerning what elements are present.

Procedure. To determine the amount of lithium in the petroleum-associated water, transfer an aliquot of about 10 ml of brine to a 50-ml volumetric flask, add 20 ml of n-propanol, and dilute to volume with distilled water. (The size of the aliquot will vary with the sample. The specific gravity can be used to help decide the aliquot size. For a brine with a specific gravity of 1.1, an aliquot of 10 ml or less probably will be sufficient.) Aspirate the sample into the flame and read or record the emission intensity of the background at 665 mp and lithium line at 670.8 mp. With these readings and the preliminary calibration curves, calculate approximately how much lithium is in the sample. Determine an aliquot size that will contain about 0.05 mg of lithium. Transfer equal aliquots to three 50-ml volumetric flasks. Add no lithium standard to the first flask, 0.05 mg to the second flask, and 0.1 mg to the third flask. Add 20 ml of n-propanol t o each flask and dilute to volume with distilled water. Aspirate and record the background at 665 mp and the emission intensity of each sample at 670.8 mp. Optimum accuracy is attained by this method when the two standard additions are respectively equal to and twice the amount of lithium in the sample. Care should be taken that too much lithium is not present in the final samples, because self-absorption will cause errors.

56
5

ANALYSIS OF OILFIELD WATERS

4
v)

13

2 3 W a c a
.x2
I

u
1

L
I
COI

2s

4 5 6 tENTRATION OF STANDARD ADDITIONS

I
-

I
7

Fig. 3.3. Standard-addition calculation graph. In this ideal case the unknown would contain 2 x the dilution factor ( 2 could be 2 mg or 2 pg or whatever unit the analyst used).

Calculation. A graph can be used in the calculation, as illustrated in Fig. 3.3. Plot the concentrations in milligrams of the standard-addition samples on the horizontal axis of linear graph paper and the emission intensities on the vertical axis. Plot the emission intensity of the sample to which no standard lithium soiution was added at 0 concentration. The plot should produce a straight line as shown in Fig. 3.3. Multiply the chart reading at 0 concentration by 2, place this value on the y-axis, and draw a line parallel to the x-axis until it intersects the line plotted. From this point, draw a line parallel to the y-axis until it intersects the x-axis. The vrlue obtained in milligrams can be converted to milligrams per liter by the following formula:
mg Li x 1,000 = mg/l Li+ ml sample The formula, shown in Table 3.X1, can be used to calculate the amount of lithium in the sample, using the flame spectrophotometric readings in lieu of the graph method. Optimum accuracy is attained with this method using either type of calculation when the two standard additions respectively are equal to and twice the amount of lithium that is present in the sample. The addition of alcohols t o the aqueous phase before aspiration into the flame increases the sensitivity of the flame method, allowing the use of more dilute solutions and consequently less dissolved solids, which reduces burner plugging. The average precision and accuracy of the lithium method are about 2% and 4%, respectively, of the amount present.

FLAME SPECTROPHOTOMETRIC METHODS TABLE 3.XI


C Formula for standard-addition calculation C = (rx - r b ) , r-r, where the following are true*:
.. -

57

Solution

Concentration

Reading

Unknown Mixture
.

c , c ,
=c + ,

r,

*C is a standard addition.

Sodium The flame spectrophotometer offers an excellent instrumental technique for determining sodium in a petroleum-associated water. The flames containing alkali metals give strong resonance lines of these metals plus some additional continyous radiation. The strongest line for sodium results from a transition between the lowest excited level and the ground state. The yellow doublet of sodium at 589.0-589.6 mp results from such a transition. Reagents. The necessary reagents are sodium standard solutions, 1 mglml and 0.01 mg/ml; and n-propanol. Preliminary Calibration curves. Preliminary calibration curves similar to those shown in Fig. 3.2 should be used to determine the approximate amount of sodium in the sample. These curves are prepared in the same manner as the lithium curves, except that standard sodium solutions are used; the emission intensity of the sodium at 589 mp is determined, minus a background at about 582 mp. Procedure. To analyze the petroleum-associated water, transfer an aliquot of water t o a 50-ml volumetric flask, add 20 ml of n-propanol, and dilute to volume with distilled water. (The size of the aliquot will vary with the sample. The specific gravity can be used to help decide the aliquot size. For a water with a specific gravity of 1.1,an aliquot of 1ml or less probably will be sufficient.) Aspirate the sample into the flame and record the emission intensity of the background at 582 mp and sodium line at 589 mp. With these readings, calculate approximately how much sodium is in the sample by using the preliminary calibration curves. Determine the aliquot size that will contain about 0.05 m of sodium. g Transfer equal aliquots t o three 50-ml volumetric flasks. Add no sodium

58

ANALYSIS OF OILFIELD WATERS

standard to the first flask, 0.05 mg to the second flask, and 0.1 mg t o the third flask. Add 20 ml of n-propanol to each flask and dilute to volume with water. Aspirate and record the emission intensity of each sample at 589 mp and its background at 582 mp. Calculation. Use the graph or formula illustrated in the lithium method. The value obtained in milligrams can be converted to milligrams per liter by the following formula: mg Na x 1,000 = mg/l Na' ml sample The precision and accuracy of the method are approximately 3%and 6%, respectively, of the amount of sodium present. Some elements, when present in the solution being analyzed, will cause a change in the emission intensity of the sodium. The use of a standard addition technique largely compensates for these interferences. Potassium Potassium usually is included with sodium without any differentiation in reporting the results of brine analyses, although potassium is known to be present in many oilfield brines. Potassium compounds often are dissolved before sodium compounds; however, they do not remain dissolved as readily because they are readily adsorbed and enriched in clays. In sea water and oilfield brines, only a small part of the originally dissolved potassium remains in solution. The fact that many oilfield brines are low in potassium with respect to sodium, whereas surface waters and young volcanic waters are enriched in potassium with respect to sodium, is an important criterion in identifying the sources of brines. The flame spectrophotometer provides a sensitive method for the determination of potassium. The strongest lines for potassium detection in a flame are the doublet at 766.5 and 769.9 mp. Reagents. The necessary reagents are potassium standard solution, 0.1 mg/ml; and n-propanol. Preliminary calibration curves. Preliminary calibration curves are useful in determining the approximate amount of potassium in the sample, so that the optimum sample size for standard addition can be selected for the analysis. These curves can be prepared in the same manner used in the preparation of the lithium preliminary calibration curves (Fig.3.2) except that standard potassium solutions are used. The emission intensity of the potassium line at 766.5 mp minus the background at about 750 mp can be used in preparing the curves.

FLAME SPECTROPHOTOMETRIC METHODS

59

Procedure. To determine the amount of potassium in the sample, transfer an aliquot of sample to a 50-ml volumetric flask, add 20 ml of n-propanol, and dilute to volume with distilled water. The specific gravity can be used to help decide the aliquot size. For a brine with a specific gravity of 1.1,an aliquot of 5 ml or less probably will be sufficient. Aspirate the sample into the flame and record the emission intensity of the background at 750 mp and potassium line at 766.5 mp. With this reading, use the preliminary calibration curves and calculate approximately how much potassium is in the sample. Determine an aliquot size that will contain about 0.05 mg of potassium. Transfer equal aliquots t o three 50-ml volumetric flasks. Add no potassium standard to the first flask, 0.05 m to the second flask, and 0.1 mg t o the g third flask. Add 20 ml of n-propanol to each flask and dilute t o volume with distilled water. Aspirate and record the emission intensity of each sample at 766.5 mp and the background at 750 mp. Optimum accuracy is attained by this method when the two standard additions are respectively equal to and twice the amount of potassium in the sample. Care should be taken that too much potassium is not present in the final samples, because self-absorption will cause errors.
Calculation. The graph or formula illustrated in the lithium method can be used. The value obtained in milligrams can be converted to milligrams per liter by the following formula: mg K x 1,000 = mg/l K+ ml sample The precision and accuracy of the method are approximately 2% and 4% of the amount present. Several elements can interfere in the flame analysis of potassium. Elements which ionize easily will lower the degree of ionization of potassium, and elements which are difficult to ionize or have high ionization energies will give the opposite effect. By using the Saha equation (Herrmann and Alkemade, 1963), it is possible t o estimate such interferences. Generally, the use of a standard addition compensates for interferences. Rubidium and cesium The flame spectrophotometer provides one of the most sensitive methods available for determining rubidium and cesium. Cesium has a pair of emission lines at 852.1 and 894.4 mp. Both lines are of about equal intensity, but water produces a molecular band system at 900 mp which can interfere at 894.4 mp. Rubidium also has two strong lines in the red region at 780.0 and 794.8 mp. It is necessary t o use a photomultiplier with an S-1 response t o detect cesium and rubidium at the levels found in many waters. Examples of such tubes are ITT type 6836/FW118, RCA types 1P22 and 7102, and DuMont

60

ANALYSIS OF OILFIELD WATERS

type 6911. Such tubes also are useful for lithium and potassium determinations. Several elements can interfere in the determination of cesium and rubidium. However, because a solvent extraction or standard-addition technique is used most interferences are either removed or compensated (Collins, 1965).

Reagents. The necessary reagents are cesium standard solution, 0.01 mg/ml; rubidium standard solution, 0.01 mg/ml; buffer solution, pH 6.6 (adjust the pH of a 1M sodium citrate solution to 6.6 with 0.5M nitric acid); sodium tetraphenylboron, 0.05M (dissolve 0.855 g of sodium tetraphenylboron in distilled water and dilute t o 50 ml - prepare a fresh solution daily); nitroethane; hydrochloric acid, 0.1N; sodium hydroxide, 0.W;synthetic brine solution. Procedure. To determine the amount of rubidium and cesium in the petroleum-associated water, transfer an aliquot of brine containing 0.005 to 0.05 m of cesium and rubidium to a 100-ml beaker and add 25 ml of the g citrate buffer solution. Transfer the solution to a 125-ml Teflon-stoppered separatory funnel and adjust to 100-ml volume. Add 2 ml of 0.05M sodium tetraphenylboron aqueous solution and 1 0 ml of nitroethane, and shake the mixture vigorously for 2 minutes. Allow the phases t o separate for 30 minutes, after which time withdraw the aqueous phase. Centrifuge the nitroethane phase. Determine the cesium and rubidium emission intensities by burning the nitroethane phase in the flame spectrophotometer and automatically scanning the 780.0 mp, 794.8 mp, and 894.4 mp lines. Calibration curves. Prepare calibration curves by using appropriate portions of the standard cesium and rubidium solutions. Add 5 ml of synthetic brine solution t o each standard sample before buffering and extraction. Plot the resultant emission intensities versus milligrams of cesium or rubidium or linear graph paper. Calculation. Determine the milligrams of cesium or rubidium in the sample by referring t o the calibration curves. The milligrams can be converted to mg/l by the following formula:
mgx 1,000 = mg/l Cs+ or Rb' ml sample Fig.3.4 illustrates the relative emission intensities obtained with cesium and rubidium in nitrobenzene, nitroethane, 1-nitropropane, and 2-nitropropane. 15 ml of each of these solvents.are used t o extract 0.1 mg each of cesium and rubidium tetraphenylboron from aqueous solutions. The organic phases then are aspirated directly into the flame, and the peaks scanned automatically. Good resolution is obtained with a 0.01 mm slit width. Amy1 alcohol gives poorer results than nitrobenzene.

FLAME SPECTROPHOTOMETRIC METHODS


NITROBENZENE NITROETHANE I- NITROPROPANE

63

2 - NITROPROPANE

Fig. 3.4. Relative intensities obtained by burning organic solvents containing tetraphenylboron salts of cesium and rubidium.

Standard-addition technique to determine rubidium Some waters contain sufficient rubidium to enable use of the standardaddition technique. To analyze such waters, preliminary calibration curves similar to those used to determine lithium (Fig.3.2) are recommended, to aid in selecting the optimum amount of standard rubidium solution to use. Manganese The amounts of sodium, potassium , calcium, and strontium in most petroleum-associated waters are too high t o permit determination of manganese with the flame spectrophotometer without preliminary separations. These interferences can be obviated by extracting the manganese into a chloroform 8-hydroxyquinoline solution. The chloroform is removed by evaporation, and the manganese hydroxyquinoline is dissolved in n-propanol. This solution is burned in the flame spectrophotometer, and the emission intensity of its resonance triplet at 403.2 mp is recorded (Collins, 1962). Reagents. The necessary reagents are standard manganese solution (dissolve 0.583 g of manganese dioxide in 10 ml of hydrochloric acid and dilute to 1 liter with distilled water, transfer a 100-ml aliquot of this solution t o another

62

ANALYSIS OF OILFIELD WATERS

1-liter flask, add 10 ml of hydrochloric acid, and dilute to volume with distilled water; ( 1 ml of this solution contains 10 pg of manganese); chloroform solution of 8-hydroxyquinoline (dissolve 1.O g of 8-hydroxyquinoline in 100 ml of chloroform); hydrogen peroxide (3% solution); ammonium hydroxide ( 3 N ) ; sodium potassium tartrate (10% solution); ammonium fluoride (5%solution); n-propanol; and chloroform.

Procedure. Transfer an aliquot of brine containing up to 150 pg of manganese to a 100-ml beaker; add 1ml of hydrogen peroxide, 5 ml of ammonium fluoride, and 10 ml of sodium potassium tartrate; and adjust the pH of the mixture to 9.0 with ammonium hydroxide. Transfer the solution to a 125-ml Teflon-stoppered separatory funnel, add 10 ml of 8-hydroxyquinoline chloroform solution, and bring the mixture t o equilibrium by shaking it vigorously for 1minute. Draw the chloroform phase off into a 100-ml beaker and strip the aqueous phase by an additional extraction with chloroform. Evaporate the combined chloroform extracts to dryness over a hotplate, taking care t o prevent the residue from charring. Dissolve the residue in n-propanol and make to 50 ml volume with n-propanol. Aspirate the n-propanol solution directly into the flame and determine the net emission by subtracting the background emission at 400 mp. Calculate the amount of manganese in the sample from a calibration curve prepared by adding known amounts of manganese t o a synthetic brine solution. The calibration curve should be linear for up t o 150 pg of manganese when the emission intensity is plotted versus micrograms of manganese on linear graph paper. Calculation :
pg Mn (from curve) = mg/l Mn +* ml sample The intensity of the emission of manganese in a flame spectrophotometer is enhanced by a factor of 16 by using n-propyl alcohol rather than water as the solvent. With this increased intensity, the sensitivity of the method is about 1 mg/l, although additional sensitivity is attainable by concentrating the brine by evaporation. The precision of the method is about 3%,and the accuracy is about 6% of the amount present.

Strontium
Several flame photometric methods are available for determining strontium in oilfield brines; a standard curie may be unreliable if there are instrument changes, such as a slightly plugged burner, change of resistance in the amplifying circuit, or other variables. Chemical precipitation of strontium as the sulfate does not satisfactorily separate strontium from barium

FLAME SPECTROPHOTOMETRIC METHODS

63

and calcium without several preliminary separations. Precipitations as the carbonate or oxalate have the same disadvantages, and precipitation as the nitrate and subsequent solvent extraction of calcium with butylcellosolve still leaves barium in the precipitate. The use of a standard addition flame photometric method gives reproducible results without the necessity of several separations. Reagents. The necessary reagents are standard strontium solution, 1mg/ml; and n-propanol. Preliminary calibration curves. To determine approximately how much strontium is present in the samples, it is advantageous t o prepare preliminary calibration curves. A procedure similar to that used in the lithium method can be used, except that the strontium emission should be determined at 680 mp with a background reading at 690 mp. The data are plotted in a manner similar to Fig. 3.3. Procedure. To determine the amount of strontium, transfer an aliquot of brine to a 50-ml volumetric flask, add 20 ml of n-propanol, and dilute t o volume with distilled water. Aspirate the sample into the flame and read; record the emission intensity of the background at 690 mp and the strontium line at 680 mp. With these readings and the preliminary calibration curves, calculate approximately how much strontium is in the sample. Determine an aliquot size that will contain about 1.0 mg of strontium. Transfer equal aliquots t o three 50-ml volumetric flasks. Add no strontium standard t o the first flask, 1.0 mg t o the second flask, and 2.0 mg t o the third flask. Add 20 ml of n-propanol to each flask and dilute t o volume with distilled water. Aspirate and record the background at 690 mp and the emission intensity of each sample at 680 mp. Calculation. A graph can be used in the calculation as illustrated in Fig.3.3. The value obtained in milligrams can be converted t o milligrams per liter by the following formula: mg Sr x 1,000 = mg/l Sr+* ml sample The formula, shown in Table 3.X1, can be used to calculate the amount of strontium in the sample using the flame spectrophotometric readings in lieu of the graph method. Barium A flame spectrophotometric method was developed which utilizes the chromate precipation followed by dissolution in nitric acid, mixing with an alcohol, and burning in the flame (Collins, 1962). The flame method is

64

ANALYSIS OF OILFJELD WATERS

subject to few interferences except from calcium, but by using the chromate precipitation, calcium is eliminated and barium is concentrated. Reagents. The necessary reagents are barium standard solution, 1 mg/ml; ammonium chromate solution (dissolve 10 g of ammonium chromate in distilled water and dilute to 100 ml); 10%ammonium acetate aqueous solution; nitric acid ( 4 N ) ;n-propanol; acetic acid; and synthetic brine solution (use carbon dioxide-saturated distilled water and dissolve the following amounts of constituents in 1 liter of water: sodium bicarbonate, 0.4 g; sodium chloride, 61 g; potassium. chloride, 5 g; calcium chloride, 19 g; magnesium chloride, 1 2 g; and strontium chloride, 5 g). Procedure. Transfer an aliquot of the sample containing 0.5-15 mg of barium t o a 100-ml beaker, add 1 ml of the ammonium acetate solution, 10 ml of the ammonium chromate solution, and adjust the pH t o 4.6 using acetic acid. Cover the beaker with a watchglass; heat the solution t o near boiling (90C), remove from the hotplate, and allow to stand for 1 hour. Filter the solution through a 0.45-pm membrane filter using vacuum. Take care to transfer all of the precipitate from the beaker to the filter funnel. Use ammonium chromate solution rather than distilled water to aid in this transfer. Wash the precipitate with 50 ml of ammonium chromate or until strontium and calcium are absent. Wash the precipitate with 50 ml of hot water to remove excess chromate. Add 5 ml of 4N nitric acid t o the filter and swirl the solution on the filter gently to dissolve the precipitate. A clean test tube should be placed below the filter to catch the dissolved precipitate. When all of the precipitate is dissolved, turn on the vacuum and catch the solution in the test tube. Repeat this procedure using an additional 5 ml of 4N nitric acid. Transfer the solution from the test tube t o a 50-ml volumetric flask. Carefully wash the test tube with two 5-ml portions of water. Add 25 ml of n-propanol, dilute to 50 ml volume with water, and mix the solution thoroughly. Burn the sample in the flame spectrophotometer and record the emission intensity at 873 mp and the background at 900 mp. Prepare calibration curves by adding up t o 25 mg of barium to 10 ml portions of the synthetic brine followed by analysis according t o the foregoing procedure, and use in the calculation. Calculation: mg Ba x 1,000 = mg/l Ba+ ml sample

ATOMIC ABSORPTION METHODS ATOMIC ABSORPTION METHODS

65

Atomic absorption is complementary to flame spectroscopy. The spectra emitted are analyzed by absorption of resonance lines by free atoms of a constituent in the vapor phase. The unexcited or ground-state atoms produced in the flame can absorb radiant energy when supplied by a suitable external radiation source at a frequency coinciding with that of the emission frequencies of the element atomized. The measurement of this radiation absorbed forms the basis of absorption flame spectrophotometry - or atomic absorption spectrophotometry, as it is usually called. At temperatures up t o 2,7OO0C, ground-state atoms usually account for more than 90%of the atoms in the vapor phase. Hollow cathode discharge tubes generally are used as a light source. The sensitivity of detection does not depend upon the spectral response of the light receiver, since the absorption coefficient is a measure of the relative intensity of the light which passes through an absorption cell versus that which does not. Additional theory can be found in a book by Robinson (1966). Atomic absorption is useful in water and brine analysis, and there are several publications on the subject. Publications oriented to oilfield and sea water analysis are Fabricand et al. (1966), and Angino and Billings (1967). Table 3.XII illustrates the sensitivities that can be obtained using atomic absorption t o determine some metals in aqueous solutions. The sensitivities listed are obtainable if no interferences are present. Interference usually

TABLE 3.XII Approximate sensitivities for some metals to atomic absorption Metal
--

Wavelength
. -

(A)

Sensitivity (mg/l)
1.o 0.2 0.1 0.04 0.08 0.15 0.2 0.3 0.5 0.02 0.15 0.01 0.15 0.1 0.03 0.04

Fuel and oxidant nitrous oxide-acetylene nitrous oxide-acetylene nitrous oxide-acetylene air-acetylene air-acetylene air-acetylene air-acetylene air-acetylene air-acetylene air-acetylene air-acetylene air-acetylene air-acetylene air-acetylene air-acetylene air-acetylene

Aluminum Barium Beryllium Cadmium Calcium Chromium Copper Iron Lead Magnesium Manganese Mercury Nickel Silver Sodium Zinc

3093 5536 2348 2288 4226 3579 3247 2483 2833 2852 2794 2536 2320 3281 5890 21 38

66

ANALYSIS OF OILFIELD WATERS

results from lack of absorption of atoms bound in molecular combination in the flame and can occur when the flame is not hot enough to dissociate the molecule. It also occurs when a dissociated atom immediately oxidizes to a compound that cannot dissociate further at the temperature of the flame. Interferences

Ionization
When a significant number of the atoms of the element being determined are ionized in the flame, an error in the analysis can result. This ionization is because of excessive flame temperature, which, however, can be changed to control this interference. Another type of interference can be caused by the presence in the sample of other, more easily ionizable elements than the one sought. The resulting increase can be controlled by the addition of a sufficient amount of the interfering element t o both sample and standards t o produce a plateau in the absorbance above which no further increase occurs. Che m ica 1
A chemical interference is caused by the formation, in the flame, of salts of the element sought which are difficult to decompose, thus reducing the amount of the element available for absorption. The formation of such compounds may often be precluded by the addition of another element, such as lanthanum, which forms a less-soluble salt with the interfering anion than does the element desired. The interfering anion is thus removed from the flame, and the interference is eliminated. Phosphate combines with calcium and magnesium and produces an interference; however, the addition of lanthanum largely overcomes this interference. Addition of an excess of a cation having a similar or lower ionization potential usually reduces interference problems.

Matrix
Matrix interference is caused by unequal amounts of dissolved solids in the standards and samples. This can cause error because of differences in aspiration rates through the atomizer. Often this can be controlled by matching the specific gravities of the standards and samples or by adding salts t o the standards. Burners and solvents Various types of burners are used with atomic absorption spectrophotometers. For example, a Boling burner usually is used for aqueous solutions,

ATOMIC ABSORPTION METHODS

67

while a premix burner is used for organic solutions. A nitrous oxide burner head with. a 2-inch slot is used for determining aluminium, barium, and beryllium because overheating is often encountered wit,h a 3-inch slot burner. The use of concentration steps, such as solvent extraction of a chelated compound, enables sensitivities lower than those shown in Table 3.XII to be achieved. For example, aluminium and beryllium can be complexed with 8-quinolinol and extracted with chloroform; cadmium and lead can be complexed with ammonium pyrrolidine dithiocarbamate and extracted with methyl isobutyl ketone. When burning the organic solvents, it usually is necessary t o reduce the fuel air ratio because the burning organic solvent contributes to the fuel supply producing an undesirable luminescent flame and may also lift the flame off the burner. An optimum fuel/air ratio can be found by noting the characteristics of the flame before burning the organic solvent and then reducing the fuel flow, while burning the organic solvent until the flame characteristics are similar t o those noted before the organic solvent was burned. Ramirez-Munoz (1968) provides additional information. Burner height is very important and adjustment often is necessary when changing from one element t o another. Some instruments have a Vernier adjustment for reproducing burner-height settings and some do not. Fig. 3.5 illustrates a device which can be used for reproducing exact burner height (Ballinger et al., 1972).

0-m

from hollow cathode lamp

Fig. 3.5. Device for reproducing burner height for emission and atomic absorption spectrometers.

68

ANALYSIS OF OILFIELD WATERS

Lithium Lithium is determined at the 6707.8 A wavelength with an air-acetylene flame.

Interferences. Ionization interference is suppressed by adding 1,000 pg/ml of potassium. Reagents. The necessary reagents are: (1) Potassium solution: see reagents preparation under Sodium. (2) Standard lithium solution: obtain commercially or dissolve 5.324 g of 0 lithium carbonate, Liz CO, , in a minimum volume of one part Hz to one part of HC1 (1+ 1). Dilute to 1liter with water. 1ml of this solution contains 1,000 pg of lithium. Preliminary calibration. Prepare standard lithium solutions containing 1-5 pg/l of lithium using the standard lithium solution and 50-ml volumetric flasks. Add to each of these and to a blank, 0.5 ml of the potassium stock solution. Aspirate these standards and the blank as recommended in the calcium method and determine the absorbance at a wavelength of 6707.8 A. Procedure. Transfer an aliquot of brine to a 50-ml volumetric flask. The specific gravity of the brine can be used as a guide in estimating the size of an aliquot containing about 0.05 mg of lithium. Add 0.5 ml of the potassium stock solution, dilute t o volume with water, and aspirate. Calculate the approximate sample concentration from the preliminary calibration readings and determine the aliquot size that will contain about 0.05 mg of lithium. Transfer equal aliquots containing about 0.05 mg of lithium t o three 50-ml volumetric flasks. Add no lithium standard t o the first flask, 0.05 mg of the lithium standard t o the second flask, and 0.10 mg t o the third. Add 0.5 ml of the potassium stock solution t o each of the three flasks and dilute to volume. Aspirate and record the absorbance readings for each sample. Calculations. See calculations under Lithium in the flame spectrophotometric section, Fig.3.3, or Table 3.XI:
mgLix 1,000 = mg/l Li+ ml sample

Precision. In a single laboratory using oilfield water samples containing concentrations of 90 and 190 mg Li+/l, the standard deviations were k 3 and +5, respectively. The recoveries were 100.6% and 92.996, respectively.
Sodium Two wavelengths are used: the 5890-5896

A doublet for the 1-mg/l

ATOMIC ABSORPTION METHODS

69

aliquots and the 3302-3303 doublet for the 100-mg/l aliquots. Because of the wide .range of sodium concentrations found in brines, the higher wavelength can be used for the lower gravity brines and the lower wavelength for the higher gravity brines, thus avoiding making two dilutions with some of the heavier brines. It is usually necessary t o make a preliminary determination so that the correct aliquot can be used with the standard additions.

Interferences. Ionization interference is usually overcome by adding potassium. Reagents. The necessary reagents are: (1) Potassium solution: dissolve 190.70 g of potassium chloride, KC1, in water and dilute to 1 liter. 1 ml of this solution contains 100 mg of potassium. (2) Standard sodium solution: obtain commercially or dissolve 25.420 g of sodium chloride in 1 liter of water. 1 ml of this solution contains 10 mg of sodium. Dilute 1 0 ml of this solution t o a liter. 1 ml of this solution contains 100 pg of sodium. Preliminary calibration. Prepare standard solutions containing 1.O-5.0 and 100-500 pg/ml of sodium using the standard sodium solutions and 50-ml volumetric flasks. Add to each of these, and to a blank, 0.5 ml of the potassium stock solution. Aspirate these standards and blank as recommended in for the the calcium method and determine the absorbance at 5890-5896 for the 100-500 pg/ml Na 1.0-5.0 pg/ml Na solutions and at 3302-3303 solutions.

Procedure. Transfer an aliquot of brine t o a 50-ml volumetric flask. The specific gravity of the brine can be used as a guide in estimating the size of an aliquot containing either about 0.05 mg or about 5 mg of sodium. Add 0.5 ml of the potassium stock solution, dilute t o volume, and aspirate. Calculate the approximate sample concentration from the preliminary calibration readings. Determine the aliquot size that will contain either about 0.05 m or 5 m of sodium, depending on the wavelength to be used. g g Transfer equal aliquots t o three 50-ml volumetric flasks. For the 0.05-mg aliquots, add no sodium standard t o the first flask, 0.05 mg of sodium standard to the second flask, and 0.10 m to the third. For the 5-mg g g aliquots, add no sodium standard to the first flask, 5 m t o the second, and 10 mg t o the third. Add 0.5 ml of the potassium stock solution t o each flask and dilute to volume. Aspirate and record the absorbance readings for each sample. Calculations. See calculations under Lithium in the flame spectrophotometric section, Fig. 3.3, or Table 3.XI:
mg Na x 1,000 = mg/l Na+ ml sample

70

ANALYSIS OF OILFIELD WATERS

Precision. In a single laboratory using oilfield water samples containing concentrations of 22,700 and 43,200 mg Na+/l, the standard deviations were +485 and ?1,890, respectively. The recoveries were 100.8% and 100.9%, respectively.
Potassium Potassium is determined at the 7664.9 A wavelength with an air-acetylene flame.

Interferences. Ionization interference is suppressed by adding 1,000 pg/ml of sodium. Reagents. The necessary reagents are: (1) Sodium solution: dissolve 254.20 g of sodium chloride in 1 liter of water. 1 ml of this solution contains 100 mg of sodium. (2) Standard potassium solution: obtain commercially or dissolve 1.907 g of potassium chloride, KCl, in 1liter of water. 1 ml of this solution contains 1,000 pg of potassium. Preliminary calibration. Prepare standard solutions containing 1-5 pg/l of potassium using the standard potassium solution and 50-ml volumetric flasks. Add 0.5 ml of the sodium stock solution t o each of these and to a blank. Aspirate these standards and the blank as recommended in the calcium method and determine the absorbance at a wavelength of 7664.9 A. Procedure. Transfer an aliquot of brine t o a 50-ml volumetric flask. The specific gravity of the brine can be used as a guide in estimating the size of an aliquot containing about 0.05 mg of potassium. Add 0.5 ml of the sodium stock solution, dilute t o volume with water, and aspirate. Calculate the approximate potassium concentration from the preliminary calibration readings and determine the aliquot size that will contain about 0.05 mg of potassium. Transfer equal aliquots containing about 0.05 mg of potassium t o three 50-ml volumetric flasks. Add no potassium standard to the first flask, 0.05 mg of the potassium standard t o the second flask, and 0.10 mg to the third. Add 0.5 ml of the sodium stock solution to each flask and dilute t o volume. Aspirate and record the absorbance readings for each sample. Calculations. See calculations under Lithium in the flame spectrophotometric section, Fig. 3.3, or Table 3.XI:
mg K x 1,000 = mg/l K+ ml sample

ATOMIC ABSORPTION METHODS

71

Precision. I n a single laboratory using oilfield water samples containing concentrations of 456 and 5,680 mg K+/1, the standard deviations were *25 and
+325, respectively. The recoveries were 93.7% and 97.8%, respectively. Magnesium (1) Magnesium is determined at the 2852.1 acetylene flame.

A wavelength with an air-

Interferences. The silicon and aluminum suppression of the magnesium


absorption is generally removed by the addition of lanthanum or by the use of a nitrous oxide-acetylene flame.

Reagents. The reagents are: (1)Lanthanum solution (same as used in the calcium procedure).
(2) Standard magnesium solution: obtain commercially or dissolve 1.OOO g of magnesium ribbon in a minimum of (1 + 1) HC1, and dilute to 1 liter (v/v) HC1. 1ml of this solution contains 1,000 pg of magnesium per with 1% ml and should be made up daily t o use for the standard additions.

Preliminary calibration. Prepare standard solutions containing 0.1-0.5 pg/l


of magnesium using the standard magnesium solution and 50-ml volumetric flasks. Add t o each of these and t o a blank 5 ml of the stock lanthanum solution. Aspirate as suggested in the calcium method and determine the absorbance at 2852.1 A.

Procedure. Transfer an aliquot of brine to a 50-ml volumetric flask. The


specific gravity of the brine often can be used as a guide in estimating the size of an aliquot containing about 0.005 mg of magnesium. Add 5 ml of the lanthanum stock solution, dilute t o volume with water, and aspirate. Calculate the approximate sample concentration from the preliminary calibration readings, and determine the aliquot size that will contain about 0.005 mg magnesium. Fig. 3.6 illustrates a plot of the concentration of magnesium found in some oilfield brines compared to their specific gravity. This figure cannot necessarily be applied to all oilfield brines, however, because some will contain more and some less. The concentrations of magnesium in brines from the same formation at about the same depth often are similar. Transfer equal aliquots containing about 0.005 mg magnesium to three 50-ml volumetric flasks. Add no magnesium standard t o the first flask, 0.005 mg t o the second flask, and 0.010 mg t o the third. Add 5 ml of the lanthanum stock solution t o each of the three flasks and dilute t o volume. Aspirate and record the absorbance readings for each sample.

72

ANALYSIS OF OILFIELD WATERS

4.000

-L

1.00

1.05

1.10

1.15

1.20

I. 5

SPECIFIC G R A V I T Y

Fig. 3.6. Relationship of the concentration of magnesium to specific gravity for some oilfield brines.

Calculations. See calculations under Lithium in the flame spectrophotometric section, Fig. 3.3, or Table 3.XI: mgMgx 1,000 = mg/l Mg+2 rnl sample Precision. In a single laboratory using oilfield water samples containing concentrations of 1,470 and 2,000 mg Mg+2/1, the standard deviations were k36 and +128, respectively. The recoveries were 97.3%and 103.2% respectively. Calcium (1) Calcium is determined at the 4226.7 A wavelength with an air-acetylene flame.

ATOMIC ABSORPTION METHODS

73

Interferences. The chemical suppressions caused by silicon, aluminium, and phosphate are controlled by adding lanthanum. The lanthanum also controls a slight ionization interference. A pH above 7 causes low calcium values, so dilute HC1 is added t o standards and samples. For samples containing large amounts of silica, it often is preferable t o use the nitrous oxide-acetylene flame. The analysis appears to be free from chemical suppressions, but a large amount of alkali salt should be added' t o control ionization interferences. Reagents. The reagents are: (1) Lanthanum solution: wet 58.65 g of L a 2 0 3 with water, add 250 ml concentrated HC1 very slowly until the material is dissolved and dilute t o 1 liter. This provides a 5% lanthanum solution in 25% (v/v) HC1. (2) Standard calcium solution: obtain commercially or prepare by adding 50 ml of water t o 0.2497 g of primary standard calcium carbonate, CaC03. Add dropwise a minimum volume of HC1 to dissolve all of the CaCO, and dilute to 1 liter. 1ml of solution contains 100 pg of calcium. Preliminary calibration. Use the standard calcium solution (1ml-100 pg Ca) and transfer the following amounts t o six 50-ml volumetric flasks. To the first flask add 0.5 ml, t o the second 1.0 ml, t o the third 1.5 ml, to the fourth 2.0 ml, and to the fifth 2.5 ml; and the sixth flask should have 0.0 ml. To each flask add 5 ml of the lanthanum solution and sufficient distilled water to adjust the volume t o 50.0 ml. The first flask now contains 1.0 pg/ml Ca, the fifth contains 5.0 pg/ml Ca, and the sixth is a blank. Aspirate these five standards and the blank into an air-acetylene flame and determine the absorbance at 4226.7 A. If the atomic absorption instrument has curvature correction controls, make the necessary adjustments to obtain a linear relationship between absorbance and the actual concentration of the standards. If the instrument does not have these controls, plot the results on linear graph paper as illustrated in Fig. 3.2 by substituting absorbance for intensity. Procedure. Transfer an aliquot of brine t o a 50-ml volumetric flask. The specific gravity of the brine can be used as a guide in estimating the size of an aliquot containing about 0.05 mg of calcium. Add 5 ml of the lanthanum stock solution, dilute t o volume with water, aspirate the sample into an airacetylene flame, and determine the absorbance of 4226.7 A. Calculate the approximate sample concentration from the preliminary calibration readings, and determine the aliquot size that will contain 0.05 m of calcium. g Transfer equal aliquots containing 0.05 mg Ca+2 t o three 50-ml volumetric flasks. Add no calcium standard t o the first flask, 0.5 mg to the second flask, and 0.10 m t o the third. g

74

ANALYSIS OF OILFIELD WATERS

Add 5 ml of the lanthanum stock solution to each of the three flasks and dilute to volume. Aspirate and record the absorbance readings for each sample.

Calculations. See calculations under Lithium in the flame spectrophotometric section, Fig.3.3, or Table 3.XI: mg Ca x 1,000 = mg/l Ca+ ml sample Precision. In a single laboratory using oilfield water samples containing cong centrations of 17,400 and 32,500 m Cat /1, the standard deviations were k430 and +1,090, respectively. The recoveries were 103.5% and 100.336, respectively.
Magnesium (2) The following method for the determination of magnesium in an oilfield water was supplied through courtesy of the Halliburton Company (1970), and can be used t o determine all concentrations of the magnesium ion in a brine.

Reagents. The necessary reagents are magnesium standard solution, 1 mg/ml; lanthanum solution, 1g/ml; and hydrochloric acid. Magnesium standard working so 1ut ions
Pipet 1.0 ml of the magnesium standard stock solution into a 1liter flask, add 11.0 ml t o a second 1-liter flask, and add 21.0 ml to a third 1-liter flask. To each flask add 50 ml of concentrated hydrochloric acid, 10 ml of the lanthanum stock solution, and dilute each to an overall 1,000 ml volume with water. This yields standards of 1.0, 11.0, and 21.0 mg/l of magnesium in the first, second, and third flasks, respectively.

Procedure. Filter the sample with the micropore filter apparatus t o remove solids and traces of hydrocarbons from the water. Transfer, by means of Lambda pipet or volumetric transfer pipet, an aliquot of sample t o contain not more than 1.0 mg magnesium into a 100-ml volumetric flask. Add 5.0 ml hydrochloric acid, 1.0 ml lanthanum stock solution, and sufficient water to dilute to exactly the 100-ml mark. Mix thoroughly. Aspirate the 5-mg/l standard through the burner, positioning the burner angle as necessary until the recorder indicates a stable reading of about 25% absorption using a wavelength setting of 2852 a. Record the reading and aspirate distilled water through the burner until the recorder returns t o the original baseline. Next, aspirate the sample through the burner until a maximum stable reading is obtained on the recorder. Record the reading and if the sample

ATOMIC ABSORPTION METHODS

75

reading on the recorder is greater than the 5-mg/l standard, aspirate the 9-mg/l standard through the burner until a maximum stable reading is obtained. Record the reading and if the sample reading on the recorder is less than the 5-mg/l standard, aspirate the 1-mg/l standard through the burner until a maximum, stable reading is obtained, and record the reading.

Calculations:

where %A = percent absorption of high standard; %I2 = percent absorption of low standard; %A, = percent absorption of sample; mg/ll = mg Mgt2 /1 of high standard; mg/12 = mg Mg+2/1 of low standard; mg/l, = mg Mg+2/1 of sample; and DF= dilution factor of sample (100/ml sample). Derivation of above equation: %A 1-%A 2 - %A s-%A 2 mg/l1-mg/l2 mg/l,-mg/l2

or :

when mg/l, = 11,mg/12 = 1 ; A mg/l(l-2) = 1 0 when mg/ll = 21, mg/12 = 11;A mg/l( - 2 = 1 0 1 mg/ll -mg/12 = 10, when standards of 21 mg/l and 1 mg/l or 11 mg/l and 1mg/l are used.

Calcium (2)
The same apparatus used in determining magnesium by atomic absorption can be used t o determine calcium.

Reagents. The necessary reagents are calcium standard solution, 1 mg/ml; lanthanum solution, 1g/ml; and hydrochloric acid. Calcium standard working solutions
Pipet 1.0 ml of the calcium standard stock solution into a 1-liter flask, add 11.0 ml t o a second 1-liter flask, and 21.0 ml to a third 1-liter flask. To each flask add 50 ml of concentrated hydrochloric acid, and 10 ml of the lanthanum stock solution, and dilute each to an overall 1,000 ml volume

76

ANALYSIS OF OILFIELD WATERS

with water. This yields standards of 1.0, 11.0, and 21.0 mg/l of calcium in the first, second, and third flasks, respectively.

Procedure. Filter the sample through the micropore filter apparatus to remove
solids and traces of hydrocarbons from the water. Transfer, by means of micropipet or volumetric transfer pipet, an aliquot of sample containing not more than 2.0 mg calcium into a 100-ml volumetric flask. Add 5.0 ml hydrochloric acid, 1.0 ml lanthanum stock solution, and sufficient water t o dilute to exactly the 100-ml mark and mix thoroughly. Aspirate the ll mg/l standard through the burner, positioning the burner angle as necessary until the recorder reaches a maximum stable reading of about 22% absorption using a wavelength setting of 4227 A. Record the reading and aspirate distilled water through the burner until the recorder returns t o the original baseline. Remove and aspirate the sample through the burner until a maximum stable reading is obtained on the recorder. Record the reading and aspirate distilled water through the burner until the recorder returns t o the original baseline. If the sample reading on the recorder is greater than the 1 mg/l standard, 1 aspirate the 21 mg/l standard through the burner until a maximum stable reading is obtained. Record the reading and if the sample reading on the recorder is less than the 1 mg/l standard, aspirate the 1 mg/l standard 1 through the burner until a maximum, stable reading is obtained. Record the reading.

Calculations: (%A,--%A 2 ) 10 + mg/12 x DF = mg/l Ca+2 %A1 -76 2


where %A = percent absorption of high standard; %A2 = percent absorption of low standard; %A, = percent absorption of sample; mg/ll = mg Ca+2/lof high standard; mg/12 = mg Ca+2/l of low standard; mg/l, = mg Ca+?/l of sample; and DF = dilution factor of sample (100/ml sample).

Strontium
Strontium is determined at the 4607 A wavelength with an air-acetylene flame.

Interferences. The chemical suppression caused by silicon, aluminum, and phosphate is controlled by adding lanthanum. The lanthanum also controls ionization interference. The nitrous oxide-acetylene flame can be used t o control chemical interferences, but a large excess of alkali salt should be added t o control ionization.

ATOMIC ABSORPTION METHODS

77

Reagents. The necessary reagents are: (1) Lanthanum solution (same as used in the calcium standard-addition procedure). (2) Standard strontium solution: obtain commercially or dissolve 2.415 g of strontium nitrate, Sr(N03)2,in 1 liter of 1% (v/v) HNO,. 1 ml of the solution contains 1,000 pg of strontium. Preliminary calibration. Prepare standard strontium solutions containing 1-10 pg/ml of strontium using the standard strontium stock solution and 50 ml of volumetric flasks. Add to each of these and t o a blank, 5 ml of the lanthanum stock solution. Aspirate these standards and the blank as suggested in the calcium method and determine the absorbance of strontium at 4607 A. Procedure. Transfer an aliquot of brine to a 50-ml volumetric flask. The specific gravity of the brine can be used as a guide in estimating the size of an aliquot containing about 0.1 mg of strontium (see Fig. 3.6). Add 5 ml of the lanthanum stock solution, dilute t o volume, and aspirate. Calculate the approximate sample concentration from the preliminary calibration readings, and determine the aliquot size that will contain about 0.1 mg strontium. Transfer equal aliquots containing about 0.1 mg of strontium t o three volumetric flasks. Add no strontium standard to the first flask, 0.1 mg to the second, and 0.2 t o the third. Add 5 ml of the lanthanum stock solution to each of the three flasks and dilute to volume. Aspirate and record the absorbance readings for each sample. Calculations. See calculations under Lithium in the flame spectrophotometric section, Fig. 3.3, or Table 3.XI:
mg Sr x 1,000 = mg/l Sr+2 ml sample

Precision. In a single laboratory using oilfield water samples containing concentrations of 840 and 2,250 mg Sr+2/1,the standard deviations were +48 and +110, respectively. The recoveries were 106.8%and 103.1%, respectively.

Barium
Barium is determined at the 5336 nitrous-oxide flame.

A wavelength with an acetylene

Interferences. Ionization interference is suppressed by adding 1,000 pg/ml of sodium.

78

ANALYSIS OF OILFIELD WATERS

Reagents. The necessary reagents are: (1) Sodium solution: see reagents preparation under Potassium. (2) Standard barium solution: obtain commercially or dissolve 1.5161 g of BaClz in 1 liter of water. 1 ml of this solution contains 1,000 pg of barium. Preliminary calibmtion. Prepare standard barium solutions containing 2-1 0 pg/ml of barium using the standard barium solution and 50-ml volumetric flasks. Add t o each of these and to a blank, 0.5 ml of the sodium stock solution. Aspirate these standards and the blank as recommended in the calcium method and determine the absorbance at a wavelength of 5336 8. Procedure. Transfer an aliquot of brine to a 50-ml volumetric flask. The specific gravity of the brine can be used as a guide in estimating the size of an aliquot containing about 0.1 m of barium. Add 0.5 ml of the sodium g stock solution, dilute t o volume with water, and aspirate. Calculate the approximate sample concentration from the preliminary calibration readings and determine the aliquot size that will contain about 0.1 mg of barium. Transfer equal aliquots containing about 0.1 mg of barium to three 50-ml volumetric flasks. Add no barium standard t o the first flask, 0.1 mg of the barium standard t o the second flask, and 0.2 mg to the third. Add 0.5 ml of the sodium stock solution t o each of the three flasks and dilute to volume. Aspirate and record the absorbance readings for each sample. Calculations. See calculations under Lithium in the flame spectrophotometric section, Fig. 3.3 or Table 3.XI:
mg Ba x 1,000 = mg/l Ba+ ml sample

Precision. In a single laboratory using oilfield water samples containing concentrations of 7 and 8 m Ba+/l, the standard deviations were k0.5 and g kO.9, respectively. The recoveries were 108.2% and 97.3% respectively.
Manganese Manganese is determined at the 2794.8 8 wavelength with an air-acetylene flame.

Reagents. The necessary reagent is a standard manganese solution: obtain commercially or dissolve 1.000 g of manganese in a minimum volume of (1 +1)nitric acid. Dilute t o 1 liter with 1% (v/v) HC1.l ml of solution contains 1mg of manganese. Preliminary calibration. Prepare standard manganese solutions containing 1-5 pg/ml using the standard manganese solution and 50-ml volumetric flasks. Aspirate these standards arid a blank as recommended in the calcium method, and determine the absorbance at a wavelength of 2794.8 8.

ATOMIC ABSORPTION METHODS

79

Procedure. Transfer an aliquot containing about 0.05 mg of manganese to a 50-ml volumetric flask. Dilute t o volume and aspirate. Calculate the approximate sample concentration from the preliminary calibration readings and determine the aliquot size that will contain 0.05 mg of manganese. Transfer equal aliquots containing about 0.05 mg of manganese to three 50-ml volumetric flasks. Add no manganese standard t o the first flask, 0.05 mg of the manganese standard to the second flask, and 0.10 mg t o the third. Dilute t o volume. Aspirate and record the absorbance readings for each sample.

Calculations. See calculations under Lithium in the flame spectrophotometric section, Fig. 3.3, or Table 3.XI:
mg Mn x 1,000 = mg/l Mn+ ml sample

Precision. In a single laboratory using oilfield water samples containing concentrations of 20 and 97 mg Mn+2/1,the standard deviations were k 1 and +3, respectively. The recoveries were 102.2% and 105.4% respectively. Iron
Iron is determined at the 2483.2 flame.

A wavelength with an air-acetylene

Interferences. The sensitivity is reduced if nitric acid and nickel are present. This effect can be controlled by using a very lean (hot) flame. Reagents. The necessary reagent is a standard solution: obtain commercially or dissolve 1.000 g of iron wire in 50 ml of (1+ 1)nitric acid and dilute to 1 liter with water. 1 ml of solution contains 1mg of iron. Preliminary calibration. Prepare standard iron solutions containing 1-5 Mg/ml using standard iron solution and 50-ml volumetric flasks. Aspirate these standards and a blank as recommended in the calcium method and determine the absorbance at a wavelength of 2483.2 A. Procedure. Transfer an aliquot containing about 0.05 mg of iron to a 50-ml volumetric flask. Dilute t o volume and aspirate. Calculate the approximate sample concentration from the preliminary calibration readings and determine the aliquot size that will contain about 0.05 mg of iron. Add no iron standard t o the first flask, 0.05 mg of the iron standard to the second flask, and 0.10 mg to the third. Dilute to volume. Aspirate and record the absorbance readings for each sample.

80

ANALYSIS OF OILFIELD WATERS

Calculations. See calculations under Lithium in the flame spectrophotometric section, Fig. 3.3, or Table 3.XI:
mg Fe x 1,000 = mg/l Fe+ ml sample

Precision. In a single laboratory using oilfield water samples containing concentrations of 6.3 and 6.8 mg Fe+2/1, the standard deviations were k0.5 and k0.3, respectively. The recoveries were 115.6% and 97%,respectively.
copper Copper is determined at the 3247.5 8 wavelength with an air-acetylene flame.

Reagents. The necessary reagent is a standard copper solution: obtain commercially or dissolve 1.000 g of copper metal in a minimum volume of (1 + 1) nitric acid. Dilute 1 liter with 1% (v/v) nitric acid. 1 ml of solution contains 1 mg of copper. Preliminary calibration. Prepare standard copper solutions containing 1-5 pg/ml using the standard copper solution and 50-ml volumetric flasks. Aspirate these standards and a blank as recommended in the calcium method and determine the absorbance at a wavelength of 3247.5 8. Procedure. Transfer an aliquot containing about 0.05 mg of copper t o a 50-ml volumetric flask. Dilute t o volume and aspirate. Calculate the approximate sample concentration from the preliminary calibration readings and determine the aliquot size that will contain 0.05 mg of copper. Transfer equal aliquots containing about 0.05 mg of copper to three 50-ml volumetric flasks. Add no copper t o the first flask, 0.05 mg of the copper standard to the second flask, and 0.10 mg t o the third. Dilute t o volume. Aspirate and record the absorbance readings for each sample. Calculations. See calculations under Lithium in the flame spectrophotometric section, Fig. 3.3, or Table 3.XI:
mgCux 1,000 = mg/l CU+ ml sample

Precision. In a single laboratory using an oilfield water sample containing a concentration of 3 mg Cu+ /1, the standard deviation was k0.2. The recovery was 100.5%.

zinc
Zinc 1- determined at the 2138.6 flame.

A wavelength with an air-acetylene

ATOMIC ABSORPTION METHODS

81

Reagents. The necessary reagent is a standard zinc solution: obtain commercially or dissolve 0.500 g of zinc metal in a minimum volume of ( 1 +1) HC1 and dilute to 1 liter with 1%(v/v) HCl. 1 ml of solution contains 500 pg of zinc. Preliminary calibration. Prepare standard zinc solutions containing 0.2-1.0 pg/ml using the standard zinc solution and 50-ml volumetric flasks. Aspirate these standards and a blank as recommended in the calcium method and determine the absorbance at a wavelength of 2138.6

a.

Procedure. Transfer an aliquot containing about 10 pg of zinc to a 50-ml volumetric flask. Dilute t o volume and aspirate. Calculate the approximate sample concentration from the preliminary calibration readings and determine the aliquot size that will contain 10 pg of zinc. Transfer equal aliquots containing about 10 pg of zinc t o three 50-ml volumetric flasks. Add no zinc standard to the first flask, 10 pg of the zinc standard t o the second flask, and 20 pg t o the third. Dilute t o volume. Aspirate and record the absorbance readings for each sample. Calculations. See calculations under Lithium in the flame spectrophotometric section, Fig. 3.3, or Table 3.XI:

Precision. In a single laboratory using oilfield water samples containing cong centrations of 27 and 120 m Zn+2/1, the standard deviations were +1. The recoveries were 103.5%and 102.3%,respectively. Lead (1)
Lead is determined at the 2833.1 A wavelength with an air-acetylene flame.

Reagents. The necessary reagent is a standard lead solution: obtain commercially or dissolve 1.598 g of lead nitrate, Pb(N03)2,in 1 liter of 1%(v/v) HN03. 1 ml of solution contains 1,000 pg of lead. Preliminary calibration. Prepare standard lead solutions containing 2-10 pg/ml using the standard lead solution and 50-ml volumetric flasks. Aspirate these standards and a blank as recommended in the calcium method and determine the absorbance at a wavelength of 2833.1 8. Procedure. Transfer an aliquot containing 100 pg of lead t o a 50-ml volumetric flask. Dilute to volume and aspirate. Calculate the approximate sample concentration from the preliminary calibration readings and determine the aliquot size that will contain 100 pg of lead.

82

ANALYSIS OF OILFIELD WATERS

Transfer equal aliquots containing about 100 pg of lead t o the three 50-ml volumetric flasks. Add no lead standard t o the first flask, 100 pg of lead standard to the second flask, and 200 pg t o the third. Dilute t o volume. Aspirate and record the absorbance readings for each sample.

Calculations. See calculations under Lithium in the flame spectrophotometric section, and Fig. 3.3, or Table 3.XI: mg Pb x 1,000 = mg/l Pb+ ml sample Precision. In a single laboratory using an oilfield water sample containing a concentration of 16 mg Pb+*/l,the standard deviation was k2.6. The recovery was 74.8%.
Lead (2) Lead is determined by chelating with ammonium pyrollidine dithiocarbamate (APDC) and extracting with methyl isobutyl ketone (MIBK) (Brooks et al., 1967). The organic extract is analyzed by means of atomic-absorption spectrophotometry. Interferences have not been observed in the airacetylene flame.

Reagents. The necessary reagents are methyl isobutyl ketone (MIBK); 0.3M hydrochloric acid; ammonium pyrollidine dithiocarbamate (APDC) (dissolve 1.0 g of APDC in 100 ml of distilled water); bromphenol blue indicator solution (dissolve 0.1 g bromphenol blue in 100 ml of 50% ethanol); 2.5M sodium hydroxide; and lead standard solution. The latter can be bought commercially or made from lead nitrate. The presence of 0.5% nitric acid in the lead standards of low concentrations retards the plating of the lead on the sides of the container. Procedure. Pipet the sample into a 200-ml volumetric flask and adjust the volume to approximately 100 ml with distilled water. Add two drops of the bromphenol blue indicator solution. Adjust the pH by adding 2.5M NaOH by drops until a blue color persists. Add 0.3M HC1 until the blue color disappears. Add 2.0 ml of HC1 in excess. The pH should be 2.4. Add 2.5 ml of the APDC solution and mix. Add 10 ml of MIBK and shake vigorously for 1 minute. Allow the layer to separate and add distilled water until the ketone layer is in the neck of the flask. Aspirate the ketone layer for lead content. Prepare a calibration curve by adding known amounts of lead t o a synthetic brine solution. Calculations:
mg Pb (from curve) - mg/l Pb ml sample

EMISSION SPECTROMETRY EMISSION SPECTROSCOPY

a3

The basic requirements for all spectroscopic measurements are a source, a dispersion element, and a detector. The source may be an emitter whose emission is to be measured, or it may be a continuum that emits all wavelengths, within a certain range, so that absorption by material in the light path may be measured. In general, emission spectra are concerned with transitions from upper state to lower state electronic levels in atoms and in simple molecular species. Some flames are hot enough to excite upper electronic levels in neutral atoms (un-ionized) and in molecules. Electric discharges produce more vigorous excitation, and a high-voltage spark tends t o increase the ionization of the emitters. In spectrographic analysis the light source first vaporizes and dissociates the sample and second excites the atoms causing them to radiate characteristic spectra. The intensities of the spectral lines of elements excited in a light source are proportional to the concentration of the elements in the sample, thus providing a basis for quantitative analysis. Excitation is mainly thermal in the sources, flames, arcs, and sparks. Temperature is very important in spectrographic analysis because some elements are not easily excited in a thermal source while others are. The ionization potential of the element determines the ease of exciting its spectra. The alkali elements with ionization potentials of 4-5 V are excited in low energy sources while the rare gases with ionization potentials up to 25 V require high temperatures t o be excited. A Bunsen flame gives a tempera,0'; ,0'; ture of about 1 7 0 C an oxyacetylene flame, about 2 7 0 C an electric arc, 3 7 0 C 6 7 0 C ,0'-,0'; and an electric spark, about 9,700"C. the followIn ing procedures a plasma arc source was used, capable of temperatures up t o

7 7 0 C. ,0'

The plasma arc was adapted to analytical spectrography by Scribner and Margoshes (1961).The temperature of a direct current arc is increased by thermal-pinch effect. The internal standard method is used in the following procedures, and with this method the intensity of a line of the element present in unknown concentration is measured relative to that of an invariant line of a reference element. With this method the intensity ratio must be highly reproducible. Barium, boron, iron, manganese, and strontium The emission characteristics of barium, boron, iron, manganese, strontium, and lanthanum in 10 solvent systems have been studied (Collins, 1967).The greatest emission enhancement was found in a mixture consisting of 30 ml of water plus 20 ml of 35% n-amyl-alcohol and 65% acetone, as .. illustrated by Fig. 3 7 Because n-propanol is easier t o work with, it w a s used in the following procedure; however, if additional sensitivity is needed, the n-amyl-alcohol-acetone mixture can be used.

84

ANALYSIS OF OILFIELD WATERS


35 percent N-Amy1 65 percent acetone

"'i
90

Proplonlc Add, Acetone N-Proponoi

a
I -

2 50
4 30 ~ t h y l ~ l f ~ ~ i d ~

2
0

3 4 5 6 7 8 CARBON, grams per SOml

101I

Fig. 3.7. Relative intensity of lanthanum versus grams of carbon in the solvent aspirated into a plasma arc.

Reagents. The necessary reagents are: Helium. Eastman Kodak D-19 developer. Eastman Kodak rapid fixer and hardener. Stop-bath solution, e.g. 5%acetic acid. Standard spectroscopic stock solutions containing 1 mglml of the following metals (use spectroscopic grade reagents): (1) barium: dissolve 1.4368 g of barium carbonate in a minimum amount of hydrochloric acid and dilute t o 1 liter with distilled water; (2) strontium: dissolve 1.6848 g of strontium carbonate in a minimum amount of hydrochloric acid and dilute to 1 liter with distilled water; (3) boron: dissolve 5.7153 g of boric acid in distilled water and dilute to 1 liter with distilled water; (4) manganese: dissolve 1.5823 g of manganese dioxide in hydrochloric acid and dilute to 1 liter with 6N hydrochloric acid; (5) iron: dissolve 1.00 g of iron wire in aqua regia and dilute t o 1 liter with 6N hydrochloric acid. Internal standard solution: dissolve 2.3455 g of lanthanum oxide in a minium amount of hydrochloric acid and dilute t o 1 liter with distilled water. 1 ml contains 2 m of lanthanum. g Standard solution: prepare a composite standard containing 0.025 mg/ml of manganese, 0.075 mg/ml of iron, and 0.03 mg/ml of strontium by transferring appropriate quantities of the standard spectroscopy stock solutions to a 1-liter volumetric flask. Dilute the resultant mixture t o volume with 6N hydrochloric acid. Synthetic brine solution: prepare a solution containing the following ions, in mg/l: sodium, 32,000; calcium, 4,000; and magnesium, 2,500. Dissolve 73.766 g of sodium carbonate in hydrochloric acid, 9.989 g of calcium carbonate in hydrochloric acid and 2,500 g of magnesium metal in

EMISSION SPECTROMETRY

85

hydrochloric acid. Evaporate these acid solutions to dryness, dissolve the residues in distilled water, combine, and dilute t o 1liter with distilled water. Hydrochloric acid, concentrated. n-Propanol.

Equipment. The necessary equipment includes a spectrograph; a d.c. arc source, 18-A minimum; a plasma arc assembly; a plate-developing machine; a microphotometer; 50-ml volumetric flasks; 10-ml microburets; pipets; and spectrographic plates, Eastman Kodak Type 1-N. Spectrochemical excitation conditions. The conditions which are used to determine barium, boron, iron, manganese, and strontium are as follows: Source, d.c. arc. Current, 18-25 A (keep constant). Voltage, 220 d.c. Pre-exposure time, 5 seconds. Exposure time, 15 seconds. Spectral region, 3200-5200 8, first order and second order. Dispersion, reciprocal linear 8.21 8 / m m first order, 4.02 8 / m m second order or better. Plasma arc assembly. Helium lift gas, 7 liter per minute. Helium tangential gas, 60 liter per minute. Atomizer, Beckman Model 4030 with medium-bore capillary. Arc length above orifice, 7 mm. Full arc length, anode t o cathode, 18 mm. Portion of arc viewed, 2 mm above orifice. Orifice electrodes; lower anode, Ultra Carbon 106 drilled to 3.97 mm in center hole, tapered to 9.53 mm at bottom; center ring, neutral Ultra Carbon 861 drilled to 5.95 mm center hole, tapered to 9.53 mm at bottom. Cathode electrode, vertical position, Ultra Carbon 6.35 mm graphite rod with pointed tip. Slit, 20-p. Filter, 3-step. Spectrographic plate development conditions: 5 minutes in Eastman Kodak D-19 at 2OoC with constant agitation; 30 seconds in stop bath at 3OoC with agitation; 5 minutes in Eastman Kodak rapid fixer and hardener solution at 2OoC with constant agitation; 30 minutes in water rinse with constant fresh supply of water; 30 seconds rinse with distilled water; and 30 minutes in constant air bath to dry. Microphotometer criteria. Slit, 1 2 p wide and 0.5 mm high. Read the background and intensity of the following lines: Ba, 11, 4554.03

86

ANALYSIS OF OILFIELD WATERS

8;B, I, 4995.46 8; 11, 5198.80 8;Mn, 11, 5152.20 8;Sr, 11,4215.52 8; Fe, La, 11, 4086.62 8; La, 11, 4429.90 8. and
The background and intensity of the following lines can be read if some of those above are too intense or if more than one line for a given element is wanted: Ba, 11, 4934.09 8; B, I, 4993.56 8; Fe, 11, 4196.74 8; Mn, 11, 5187.46 8; La, 11, 4077.35 8; La, 11, 4123.23 8; 11, 4077.71 8;and Sr, Sr, I, 4607.33 8.

Calibration. A preliminary curve, gamma curve, and calibration curves are needed unless a direct-reading instrument is used. To make a preliminary curve, record an iron spectrum using d.c. arc current excitation at about 4 A. ' ) Read the percent transmittance (% 2 of several iron lines at 100%unfiltered portion. (Any filter can be used as long as the 5% T is known.) Plot the 100% unfiltered lines versus the 63.10%filtered lines. The % T of these lines should vary from about 10%T t o 90%T to give a good preliminary curve, shown in Fig. 3.8. After the preliminary curve is plotted, the gamma or emulsion calibration curve is made, as shown in Fig.3.9. There are several methods of establishing a gamma curve. The following example is given: 98 on x-axis set t o equal 0.2, and 96 on the y-axis intersects curve at the same point on the curve that 98 does on the = x-axis. The filter factor is now used. In this case, it is 100%/63.10% 1.585.
%?'

98 = 0.2 96 = 0.2

1.585

Relative intensity arbitrarily set at 0.2 0.317

Owwit). rotio of filter i s 1.585

100

0
FILTERED, percent

Fig. 3.8. Preliminary curve for emission spectrometry.

EMISSION SPECTROMETRY

87

20-

- 40-

P
v)

I
R E L A T I V E INTENSITY

Fig. 3.9. Gamma or emulsion calibration curve for emission spectrometry.

A t 96 on the x-axis, find the curve intersection point on the y-axis; in this case, it is 91.

%T
91 = 0.317 x 1.585

Relative intensity 0.502

Repeat above procedure to obtain the following data: 81 = 0.502 x 1.585 0.796 1.262 63 = 0.796 x 1.585 2.000 38 = 1.262 x 1.585 3.17 19.5 = 2.000 x 1.585 9.5 = 3.17 x 1.585 5.024 4.6 = 5.024 x 1.585 7.963 12.621 2 = 7.963 x 1.585 Plot the gamma curve using the above values and plot the values on 3-cycle semilogarithmic paper. Place the 7% T values on the linear portion, usually the x-axis, and place the relative-intensity values on the log portion. The resultant curve should be an inverted S if the linear portion or % T is the x-axis. (Theoretically, only one gamma curve need be plotted for all plates with the same emulsion number.) After the gamma curve is plotted, a calibration curve for each element desired can be plotted, as shown in Fig. 3.10. To do this, spectra are recorded for various concentrations of the element in question. The % T of each of the desired lines is determined, and these % T are referred to the gamma curve to obtain their relative intensities. Ordinarily, internal standards are used t o permit a ratio of the relative intensity of the internal standard line to the relative intensity of the element line to be calculated for each concentration of the element. These ratios are plotted versus the element concentration on 2 x 2-cycle logarithmic paper.

88

ANALYSIS OF OILFIELD WATERS

I '

0.2

0.4

I I 1 I 0.6 0.8 10 . 2.0 INTENSITY RATIO

4.0

t 0

Fig. 3.10. Calibration curve for emission spectrometry.

To obtain data for calibration curves for barium, boron, iron, manganese, and strontium, use size 50-ml volumetric flasks. To one flask add no standard solution; add 1.0 ml t o the second flask; and add 2.5 ml, 5.0 ml, 7.5 ml, and 10.0 ml of standard solution to the third, fourth, fifth, and sixth flasks, respectively. (These aliquots will vary with the sensitivity of your instrument.) Add 2 ml of concentrated hydrochloric acid, 2 ml of internal standard solution, 5 d of synthetic brine solution, 20 ml of n-propanol, and sufficient distilled water t o adjust the final volume t o 50 ml at ambient temperature. For optimum accuracy, prepare duplicate or triplicate samples. Aspirate and burn the samples using the excitation conditions, the development conditions, and the microphotometer conditions described above; plot the curves using the above procedure. The water sample should be adjusted t o a pH of about 1.5 at the time of sampling t o prevent precipitation and adsorption. The sample should be contained in a good quality plastic bottle that has been rinsed first with dilute nitric acid and then with distilled water. Transfer to a 50-ml volumetric flask an aliquot of the sample of sufficient size to provide absolute quantities of the elements which will fall within the calibration curves. The optimum aliquot size will vary from brine to brine; however, equal-size aliquots often can be used for waters with similar specific gravities from the same geologic formation. Add 2 ml of concentrated hydrochloric acid, 2 ml of internal standard solution, 5 ml of synthetic brine solution (or try to approximate the ionic composition of the

EMISSION SPECTROMETRY

89

standard), 20 ml of n-propanol, and sufficient distilled water to adjust the volume to 50 ml at ambient temperature. For optimum accuracy, prepare duplicate or triplicate samples. Aspirate and excite the sample, develop the plate, and read the plate as suggested above. Determine the relative intensity ratios for the following: Ba 4554.03/La 4429.90; B 4995.46/La 4429.90; Mn 5152.20/La 4086.72; Sr 4215.52/La 4086.72; and Fe 5198.80/La 4086.72. Calculations. Refer the calculated ratio to the appropriate calibration curve to determine milligrams of tested ion in the sample. Convert this value to milligrams per liter by use of the following equation: mg from curve x 1,000 = mg/l ml sample The relative intensity ratios for other line pairs can be calculated and used if desired. The precision and accuracy of the method are approximately 2-3% and 4-696, respectively, for strontium and barium; and 5 4 %and 10-1196, respectively, for boron, iron, and manganese. Beryllium Beryllium forms a complex with acetylacetone which can be extracted into chloroform from an aqueous solution. The chloroform extracted is aspirated into a plasma arc, and the beryllium I1 line at 3131.07 A is read. An apparent carbon line at 3036.3 A is used for an internal standard. Reagents. Spectrographic plates, Eastman Kodak Type SA No. 1. Standard beryllium stock solution: dissolve 1.00 g of fused metallic beryllium (spectroscopic grade) in a small amount of 6N hydrochloric acid and dilute t o 1 liter with 1% hydrochloric acid. 1 ml contains 1 mg of beryllium. Standard beryllium solution: prepare a standard by transferring a suitable aliquot of the standard stock solution t o a 1-liter volumetric flask and hydrochloric acid. The standard prepared wl il diluting t o volume with 1% depend upon the resolution and dispersion of the spectrograph. However, for many instruments, a 0.01 pg/ml solution should be adequate. EDTA solution: dissolve 10 g of disodium ethylenediaminetetraacetic acid and 2 g of sodium hydroxide in water and dilute to 100 ml. Synthetic brine solution: dissolve 80 g of sodium chloride, 30 g of calcium chloride, 10 g of magnesium chloride, 5 g of strontium chloride, and 3 g of potassium chloride in distilled water that is saturated with carbon dioxide and dilute t o 1liter. Hydrochloric acid, concentrated. Sodium hydroxide, 0.5N. Chloroform. Acetylacetone.

90

ANALYSIS OF OILFIELD WATERS

Procedure. The spectrochemical excitations used are the same as those shown in the procedure t o determine barium, etc., with the exception that the spectral region is 2300-3300 8, first order and the slit is 10 p. The plate development conditions are the same as those shown in the procedure t o determine barium, etc., and the microphotometer conditions are the same except that the background and the intensity of only the following lines are read: Be, 11, 3131.07 internal standard line, 3036.3 8;or if the 3131.07-8 line is too intense, the Be, 11, 3130.42-a line can be used. To prevent precipitation and adsorption, immediately acidify the clean, oil-free sample to a pH of approximately 1.5 with concentrated hydrochloric acid. Store the sample for transportation t o the laboratory, in a good quality plastic bottle which previously was washed with dilute nitric acid, rinsed with distilled water, and dried. Transfer an aliquot of the sample estimated to contain 0.01-0.05 pg of beryllium t o a 100-ml beaker, adjust the pH to 0.5 with concentrated hydrochloric acid, adjust the volume to about 30-50 ml with distilled water, boil gently for 5 minutes, and then cool. Add 2 ml of the EDTA solution and adjust the pH of the mixture to 7.0 with 0.5N sodium hydroxide. Add 2 ml of acetylacetone, readjust the pH t o 7.0, mix thoroughly, and allow the solution t o stand for 15 minutes. Transfer the sample to a 125-ml Teflonstoppered, separatory funnel and adjust the volume t o 75 ml with distilled water, add 10 ml of chloroform, and shake the mixture vigorously for 2 minutes. After the phases separate, extract the chloroform phase and centrifuge it. Aspirate the centrifuged extract into the plasma arc using the above excitation conditions. For optimum accuracy, prepare duplicate samples. Develop the plates, make background corrections, and determine the relative intensity ratios for the following lines:

a,

Be 3130.42 Be 3131.07 and 3036.2 3036.3 Determine the concentration of beryllium using a calibration curve prepared by using 0.01-0.05 pg of beryllium standard. This concentration in micrograms can be converted to milligrams per liter by this formula: pg Be (from curve) = mg/l Be+2 ml sample Less than 1 ppb of beryllium can be detected with this method, the precision and accuracy of the method are about 2%and 496, respectively, of the amount present.

Aluminum,
Petroleum-associated water containing more than 5 mg/l of aluminum can be analyzed using the same procedure and internal standard that are

MASS SPECTROMETRIC METHODS

91

described above for barium, boron, iron, manganese, and strontium; the can be used. However, if the alumialuminum emission lines at 3082.5 num concentration is less than 5 mg/l, the aluminum should be separated and concentrated from the aqueous phase. This can be done by adjusting the pH of a sample containing up t o 100 pg of aluminum to pH 0.4 with hydrochloric acid, adding 10 ml of a 6% aqueous solution of cupferron, adjusting the pH t o 4.8 with sodium acetate, and extracting the aluminum complex into chloroform. The chloroform phase then is aspirated into the plasma arc using the same conditions and internal standard line that is described above for beryllium.

MASS SPECTROMETRIC METHODS FOR STABLE ISOTOPES

The ratios of the stable isotopes of deuterium and hydrogen and of oxygen-18 and oxygen-16 differ in water taken from various sources. These differences are useful in studying the origin of a water, and of studying paleoenvironments if the water is geologically old. The isotopic ratios are measured on a mass spectrometer and are always compared to the ratios found in a standard material because such a comparison proyides greater precision than direct analysis of absolute ratios. Deuterium Friedman and Woodcock (1957) developed a method whereby deuterium is converted t o hydrogen gas by reacting a 0.01-ml sample with hot uranium metal. A mass spectrometer (Friedman, 1953) is used t o compare the deuterium/hydrogen ratio in the emitted gas to the ratio in a standard gas. Replicates agreeing within k0.176 usually are considered satisfactory. The results usually are expressed as deuterium enrichments (+6 values) or depletion (-6 values) relative t o SMOW (standard mean ocean water, with a D/H ratio of 158 x (Craig, 1961b). The standard deviation is about 0.2%, and a sample with a 6 value of -5 has 5% less deuterium than SMOW. Oxygen-18 Epstein and Mayeda (1953) developed a method t o analyze water samples for l 8 0 . A 10-ml sample of water is equilibrated with carbon dioxide at 25OC and an aliquot of the COz is analyzed using a mass spectrometer for l 8 0 . The isotope ratios in the sample are compared to those in a standard material, using the mass spectrometer, which gives a greater precision than direct analysis of the absolute ratios. The standard generally used in SMOW (standard mean ocean water) which is distributed by the National Bureau of Standards (Craig, 1961a). Delta units express the isotopic data as:

92

ANALYSIS OF OILFIELD WATERS

where R is the isotope ratio such as 180/160 or D/H, and the delta values are expressed in per mil like salinity, and &MOW = O, %.
COLORIMETRIC METHODS

The instrumental measurement of the absorption of radiant energy at a certain wavelength involves spectrophotometry. The essential components of a spectrophotometer include: (1)Radiant energy source such as a tungsten-filament incandescent lamp for the visible region, while hydrogen or deuterium discharge lamps usually are used for the ultraviolet region. (2) A monochromator, which is a device that isolates a narrow band of the radiant energy. (3) Containers, cells, or cuvettes usually made of glass to hold the solution being analyzed. (4) A detector, which is a device (usually a phototube) that measures the radiant energy passed through the solution. In the application of spectrophotometric analysis the two terms transmittance and absorbance are important. Transmittance is:

T =-I 2
I 1

where T = transmittance; II = radiant energy incident upon the first surface of the sample; and I2 = radiant energy leaving the sample. The term absorbance is defined as: 1 A = -1ogIJ = lOg1,T or the negative logarithm of the transmittance. In the preparation of spectrophotometric curves of light-intensity ratio plotted against concentration, it is preferable, for convenience, t o use absorbance as the basis of the plot. Under these conditions a system that conforms t o Beers law gives a straight-line plot, and the commonly used colorimetric systems that do not conform will usually show only a moderate curvature (Willard et al., 1965). Extreme curvature, when the curve is plotted on the basis of absorbance data, is sometimes a sign that the system is not sufficiently stable for analytical purposes. Semicolloidal suspensions of colored substances often give extreme curvatures. When transmittance data are used for plotting, a curve is always obtained unless semilogarithmic coordinates are used. The modern. spectrophotometers have an absorbance calibration as well as the conventional percent transmittance, and it is common practice t o use the absorbance scale. The relations between trans-

COLORIMETRIC METHODS

93

mittance and absorbance plots for potassium permanganate solutions at three wavelengths are illustrated by Mellon (1950, p.95). Several other terms for light absorption are given in the literature and are still found on the printed scales of some photometers. Optical density is often used; it is the same as absorbance. Interferences In spectrophotometric determinations, interferences often result from the presence in the sample of dissolved or suspended foreign material that either absorbs radiant energy or reacts with the color reagent to form a complex that absorbs radiant energy. In either case, the absorbance of the sample is decreased. Where the interference results from the formation of an absorbing complex by ions in solution, dilution of the sample can eliminate the interference if the sensitivity of the color reagent for the element sought is sufficiently greater than for the interfering ions. If this is not the case, other methods must be found t o increase the selectivity of the method. Among such methods are: (1)pH adjustment: if pH is an important factor in complex ion formation, its adjustment can favor the formation of the complex of the element desired instead of the interfering ions. (2) Masking: compounds such as EDTA (ethylenediaminetetraacetic acid) are added t o the sample t o form a stable complex with interfering ions, thus preventing their reaction with the color reagent. (3) Solvent extraction: preferential solubility of some ions in organic solvents permits the removal of interfering ions. Another common source of interference in spectrophotometry is the use of color reagents that absorb at the wavelength at which the complex of the element desired is measured. Such interference usually can be reduced or eliminated by the use of a reagent blank. In some samples a significant source of interference results from the presence of natural color. The natural color in water samples often gives appreciable absorbance and requires either compensation or elimination. In some cases it is possible t o select a spectrophotometric reagent of sufficient sensitivity that the absorbance of the constituent sought will exceed the absorbance of the natural color by a large factor. If this factor is 50 or higher the error caused by the natural color is 2% or less. Knowledge of the relative sensitivity of the constituent to be determined relative to the natural color in the sample is necessary before such a factor can be used. If the relative sensitivity is unknown the natural color of the sample should be compensated for or removed. This can be done by determining the absorbance of the test sample versus the blank specified for the procedure. Determine the absorbance of the naturally colored sample versus distilled water. The difference is the corrected absorbance and is used to determine concentration values.

94

ANALYSIS OF OILFIELD WATERS

Iron The spectrochemical procedure will give values only for total iron and will not differentiate ferrous iron from ferric iron. The following procedure can be used t o determine F-+* and Fe+3 in a freshly sampled water (Collins et al., 1961). Reagents and apparatus. Standard iron solution: dissolve 1.00 g of hydrogen-reduced iron in a minimum of hydrochloric acid and dilute to 1 liter with distilled water. This solution contains 1mg/ml of iron. Transfer 1 0 ml of this solution to a l-liter flask and dilute to volume with distilled water. 1ml of this solution contains 0.01 mg of iron. Hydroquinone solution: dissolve 1 g of hydroquinone in 100 ml of distilled water.

IRON, m i l l i g r a m

Fig. 3.11. Plot of the optical density at 522 m p of the ferrous iron complex with 2,2'bipyridine.

COLORIMETRIC METHODS

95

O-phenanthroline or 2,2-bipyridine (either reagent can be used, how ever, 2,2-bipyridine is subject t o less interferences): dissolve 0.5 g of either reagent in 100 ml of distilled water. The solution can be warmed to 60C t o effect more rapid dissolution. Sulfuric acid, approximately 9N (441.36 g per liter): cautiously pour 270 ml of pure concentrated sulfuric acid into 650 ml of distilled water. Carefully mix the solution, cool, and dilute to 1liter with distilled water. Spectrophotometer capable of measurements at 508 mp or 522 mp, glasselectrode pH meter, 100-ml volumetric flasks, 10-ml microburet, and pipets.

Procedure. Prepare a calibration curve by transferring aliquots of the standard iron solution, containing from 0.02 mg t o 0.20 mg of iron, t o 100-ml volumetric flasks. To separate aliquots, add 5 ml of the sodium citrate solution and determine how much sulfuric acid is necessary to adjust the pH t o 3.5. Add this amount to the aliquots in the volumetric flasks. Add reagents in the following order: 5 ml of hydroquinone solution, 5 ml of 2,2-bipyridine or O-phenanthroline solution, and 5 ml of sodium citrate. The citrate must always be added last. Convert t o volume with distilled water, mix well, and let stand for 1hour. Prepare a reagent blank in the same manner. Determine the absorbance at 522 mp if 2,2-bipyridine is used or 508 mp if O-phenanthroline is used. Plot the absorption versus iron concentration on coordinate graph paper. The resulting curve should be linear, as shown in Fig. 3.11. Obtain a clean sample of brine, free of oil. Determine ferrous iron, by following the above procedure, but omit the addition of hydroquinone. To determine dissolved iron, filter the sample and follow the above procedure. To determine total iron, do not filter the sample. The amount of ferric iron can be calculated from the difference. Calculations: 1,000 x mg iron from curve = mg/l Fe+2 or Fe+ sample volume Concentrating copper, iron, lead, and nickel by ion exchange
To determine accurately, using colorimetric methods, copper, nickel, lead, zinc, and cadmium in oilfield brines, they should be separated from interfering ions. Many oilfield brines contain metals in such minute amounts that they must be concentrated before analyses can be made. Concentration methods investigated were ion exchange, electro-deposition, solvent extraction, and evaporation. An ion-exchange method proved t o be the most practical for concentrating copper, nickel, and lead, because it is less time consuming and requires less expensive equipment than any of the other methods studied.

96

ANALYSIS OF OILFIELD WATERS

Acidifying the samples to pH 3.5 with acetic or hydrochloric acid minimizes precipitation and adsorption. If acetic acid is used, 2 ml of formaldehyde per liter of sample should be added to retard mold growth. These precautions will aid in obtaining representative heavy metal analyses; however, to obtain optimum results, the samples should be analyzed as quickly after sampling as possible. If it is necessary to store the samples, they should be stored in a cool, dark place and should not be moved frequently. Light accelerates photochemical reactions, and high temperatures and moving accelerate chemical reactions. Once the seal of the cap of the sample bottle has been broken, the sample should be analyzed immediately. A chelating ion-exchange resin such as Dowex A-1 can be used to separate copper, iron, nickel, and lead from an aqueous solution. Slurry the resin into a plastic column about 36 cm long and 1.7 cm in diameter. Convert the resin to the sodium form by washing with 2 volumes of distilled water, 1 volume being equal to the amount of resin used, followed by 2 volumes of 1 N sodium hydroxide, and then with 10 volumes of distilled water. Because the resin expands more than 100%when changing from the hydrogen form to the sodium form, the column must be backwashed frequently t o reduce compaction of the resin and to prevent shattering of the column. Pass the brine which has been neutralized to pH 7.0 with sodium hydroxide through the column. 2 liters or more probably will be necessary, depending upon the amount of heavy metals present in the brine. Elute the chelated metals with 2 volumes of 2N hydrochloric acid and water effluents t o a small volume; cool and adjust t o a predetermined volume (for example, 200 ml) with water. Use aliquots of this solution for determining copper, iron, nickel, and lead. The resin must be changed back t o the sodium form as soon as the metals have been eluted, because the resin tends to lose its chelating capacity if left in the water-rinsed hydrogen form for longer than a few hours. If this happens, the resin can be regenerated by heating it at 6OoC in a 30-50% sodium hydroxide solution for 24 hours. Once the metals are separated from the brine and concentrated, they can be analyzed using various methods such as atomic absorption spectrometry, flame spectrometry, emission spectrometry, or colorimetry (Collins et al., 1962).

The compound 2,9-dimethyl-1,lo-phenanthroline, assigned the name neocuproine (Diehl and Smith, 1958, p.23), has the following structure:

COLORIMETRIC METHODS

97

This reagent is used t o determine copper because of its nearly specific reaction with cuprous copper. The combining ratio is 2 moles of neocuproine to 1 mole of copper. The increased selectivity of neocuproine for copper is caused by a steric hindrance effect. The cuprous neocuproine compound is formed over a pH range of 3-10 and is bright orange. The compound can be extracted with n-amyl alcohol, isoamyl alcohol, n-hexyl alcohol, or chloroform. The maximum absorption of the compound in isoamyl alcohol occurs at a wavelength of 454 mp. Hydroxylamine hydrochloride can be used t o reduce the cupric ion t o cuprous. Citrate will hold any iron present in solution when the pH is adjusted to between 5 and 6. The chromate ion can cause low results; however, this effect does not occur when iron is present, which is almost always the case with an oilfield brine. The anions such as sulfide, cyanide, periodate, nitrate, t hiocyanate, and ferricyanide can interfere by reacting with hydroxylamine; however, they are eliminated in the ion exchange separation. Reagents. Neocuproine solution: dissolve 1 g of 2,9-dimethyl-l,1 O-phenanthroline in 1liter of ethyl alcohol. Hydroxylamine hydrochloride solution : dissolve 10 g of hydroxylamine hydrochloride in 100 ml of water. Isoamyl alcohol, analytical reagent grade. Sodium citrate solution: dissolve 300 g of sodium citrate in 1 liter of water, add 2 ml of the hydroxylamine hydrochloride solution, add 1 ml of neocuproine solution, and extract with 10-ml portions of chloroform until a colorless chloroform extract is obtained. Standard copper solution: dissolve 0.100 g of copper in 5 ml of nitric acid and 5 ml of water by heating gently to dissolve the copper. Add 5 ml of perchloric acid and evaporate to fumes of perchloric acid. Cool, dilute with water, transfer t o a l-liter volumetric flask, and dilute t o volume. Pipet a 100-ml aliquot of this solution t o another l-liter volumetric flask. Dilute to volume with water. This solution contains 10 mg/ml of copper. Sodium acetate.

Procedure, Add 5 ml of 10% hydroxylamine hydrochloride solution and 20 ml of 30% sodium citrate solution to a sample of effluent from the ion exchange column containing 4-150 pg of copper, and adjust the pH of the mixture to between 5 and 6 with 1 g or more of sodium acetate. Extract with a 10-ml portion of isoamyl alcohol. Separate the liquids and discard the alcohol layer. Add 10 ml of 0.1%neocuproine solution and 10 ml of isoamyl alcohol, and shake the mixture vigorously for 1 minute. Let the phases separate and transfer the alcohol layer to a 50-ml volumetric flask. Make additional extractions until the alcohol layer remains colorless. Dilute the combined alcohol extracts t o 50 ml with isoamyl alcohol, mix, and measure the absorbance at 454 mp in a l-cm cell with a spectrophotometer.

-.#

30

ANALYSIS OF OILFIELD WATERS

Calculations. Estimate the amount of copper present by using a calibration curve prepared by using about 10-200 pg of copper:
pg Cu (from curve)

ml sample Nickel

mg/l C U + ~

Nickel forms a wine-red or brown compound with dimethylglyoxime (Sandell, 1959, p.555). The structure of the chelate on the basis of available evidence is:
H3C - C-C

//

CH3

H3C - C

/I

/I

CH,

Dimethylglyoxime gives a nearly specific reaction with nickel that has been oxidized t o its higher valences with an oxidizing agent such as bromine. The wine-red compound is somewhat unstable; therefore, the absorbance measurements should be made within 10 minutes after formation of the nickel dimethylglyoximate. Cobalt and copper also give colored compounds with dimethylglyoxime, but they can be removed by washing the chloroform extract of nickel dimethylglyoximate with dilute ammonium hydroxide. Iron interference is removed by extracting the nickel dimethylglyoximate with chloroform from a solution containing citrate. Palladium, platinum, and gold also give colored compounds when nickel dimethylglyoximate is extracted with chloroform; however, they are removed, if present, by the ion-exchange separation.

Reagents. Dimethylglyoxime solution: dissolve 1 g of dimethylglyoxime in 100 ml of ethyl alcohol. Saturated bromine water. Ammonium hydroxide solution, approximately 4N: add 200 ml of concentrated ammonium hydroxide to 800 ml of water. Standard nickel solution: dissolve 0.100 g of nickel in dilute nitric acid by heating gently. Cool, dilute with water, transfer to a 1-liter volumetric flask, and dilute to volume. Pipet a 100-ml aliquot of this stock solution into another 1-liter volumetric flask and dilute to volume. This solution contains 10 pg/ml of nickel. Hydrochloric acid, approximately 6N: cautiously add 500 ml of concentrated hydrochloric acid to 500 ml of water. Chloroform.

COLORIMETRIC METHODS

99

Ammonium citrate solution: dissolve 1 0 g of ammonium citrate in water and dilute to 100 ml. Hydroxylamine hydrochloride solution: dissolve 1 0 g of hydroxylamine hydrochloride in water and dilute to 50 ml.

Procedure. Add 10 ml of ammonium citrate solution, and 5 ml of hydroxylamine hydrochloride solution to a sample of effluent containing up to 100 pg of nickel, and adjust the pH t o 8 with ammonium hydroxide. Transfer the mixture t o a 125-ml separatory funnel, add 10 ml of dimethylglyoxime solution and 10 ml of chloroform, and shake the mixture vigorously for 1 minute. Let the phases separate and extract the chloroform phase into another 125-ml separatory funnel. Make additional extractions of the sample with 10-ml portions of chloroform until a colorless chloroform extract is obtained. Add 10 ml of 4N ammonium hydroxide solution to the combined chloroform extracts in the 125-ml separatory funnel, and shake the mixture vigorously for 1 minute. Let the phases separate and discard the ammonium hydroxide phase. Acidify the chloroform phase with 1 ml of 6N hydrochloric acid, shake the mixture vigorously for 2 minutes, let the phases separate, and discard the chloroform phase. Add 10 ml of chloroform t o the acid phase, shake the mixture vigorously for 1 minute, and discard the chloroform phase. Adjust the pH of the acid phase to 6.9, transfer it t o a 100-ml volumetric flask, add bromine water until a yellow color persists, swirl the mixture, and allow it to stand for 10 minutes. Add 10 ml of 4N ammonium hydroxide and 10 ml of dimethylglyoxime solution. Swirl to mix, cool to room temperature in an ice water bath, and adjust t o 100-ml volume with water. After 5 minutes determine the absorbance at 445 mp using a 1-cm cell and a spectrophotometer. Calculations. Calculate the nickel concentration in the water by using a calibration curve prepared by using about 10-100 pg of nickel: pg Ni (from curve) = mg/l Ni+ ml sample

Lead Dithizone (Sandell, 1959, p. 665) is an excellent reagent for the determination of traces of lead. Lead dithizonate probably has a formula similar to the following:
r
C6HS

100

ANALYSIS OF OILFIELD WATERS

Lead can be extracted from a basic solution with dithizone in chloroform or carbon tetrachloride in the presence of citrate or tartrate, which prevent the precipitation of several metal hydroxides. The optimum pH range for extraction of the lead dithizonate with chloroform is 8.5-11. Cyanide will complex all interfering metals except bismuth, thallium, and stannous tin. Because these metals are separated by ion,exchange, their interference is eliminated. Ferric iron can form a ferricyanide that will oxidize dithizone; however, this reaction can be prevented by adding a reducing agent such as hydroxylamine hydrochloride. Excess of calcium, magnesium, and phosphorus retards the lead dithizonate extraction, but thz ion exchange separation excludes phosphorus as well as much of the calcium and magnesium. The lead dithizonate in chloroform absorbs at 510 mp. The amount of lead in the chloroform phase should not be much greater than 2.5 mg/l for optimum results.

Reagents. Hydroxylamine hydrochloride solution : dissolve 10 g of hydroxylamine hydrochloride in water and dilute to 50 ml. Standard lead solution: dissolve 0.100 g of lead in 10-15 ml of nitric
acid. Dilute to 1liter volume with water. Pipet a 100-ml aliquot of this stock solution into another 1-liter volumetric flask, add 1 0 ml of nitric acid, and dilute t o 1 liter volume with water. This solution contains 10 pg/ml of lead. Ammonia-cyanide-sulfite solution: add 350 ml of concentrated ammonium hydroxide, 30 ml of a 10%potassium cyanide solution, and 1.5 g of sodium sulfite, t o a 1-liter volumetric flask and dilute to volume with water. Dithizone solution: dissolve 0.01 g of dithizone in 200 ml of chloroform. Chloroform.

Procedure. Transfer a sample of the ion exchange effluent containing up t o 80 pg of lead to a 125-ml separatory funnel, and add 5 ml of hydroxylamine hydrochloride solution, 75 ml of ammonia-cyanidesulfite solution, and 10 ml of chloroform. Shake the mixture vigorously for 1minute, let the phases separate, and discard the chloroform phase. Add 1 ml of 0.005% dithizonechloroform solution, shake the mixture vigorously for 1 minute, let the phases separate, and extract the dithizone-chloroform phase into a 25-ml volumetric flask. If the dithizone-chloroform phase is green or some color other than cinnabar red, three possibilities exist: (1)there is no lead present; (2) there is an oxidizing agent present; or (3) an excess of dithizone has been used. In any event, if the dithizone-chloroform phase is not red, acidify it with 15 ml of 1 : l O O nitric acid, shake the mixture for 1 minute t o transfer the lead to the nitric acid phase, and discard the chloroform. Treat the nitric acid phase with hydroxylamine hydrochloride solution, ammonia-cyanidesulfite solution, and make another dithizone-chloroform extraction using 0.5 ml or less of the dithizone-chloroform solution. If the dithizonechloroform phase still does not turn red, take a larger sample of the effluent. However, if the original dithizone-chloroform extraction did turn red, make

COLORIMETRIC METHODS

101

additional extractions until the dithizone-chloroform phase remains green. Dilute the combined red dithizone-chloroform phases in the 25-ml volumetric flask t o volume with chloroform, mix well, and determine the absorbance with a spectrophotometer at 510 mp.
Calculations. Prepare a calibration curve by using aliquots of the standard lead solution containing 10-80 pg of lead:

pg Pb (from curve) = mg/l Pb+2 ml sample zinc Extraction of zinc with dithizone from a weakly ammoniacal solution containing a reducing agent and citrate prevents the precipitation of iron. Extraction of zinc at a pH of 4.75 in the presence of sodium thiosulfate largely eliminates interference from copper, mercury, lead, and cadmium. The zinc dithizonate complex can be broken in 0.02N hydrochloric acid, whereas cupric dithizonate cannot. Lead and cadmium dithizonates will dissociate in 0.02N hydrochloric acid, but only traces of them should be present after the preliminary extractions. More accurate results are obtained by applying a zincon (Platte and Marcy, 1959) method t o the zinc which has been isolated by the dithizone extractions than by making another dithizone extraction of the isolated zinc and using it for absorption measurements. Therefore, the following method is a combination of the dithizone and zincon methods. Traces of any remaining interferences can be complexed. Zinc reacts with dithizone to form a compound similar to:

Zinc reacts with zincon:


OH
1

CsHS

to form a 1:lblue complex that absorbs at a wavelength of 620 mp.

102

ANALYSIS OF OILFIELD WATERS

Reagents. Standard zinc solution: dissolve 1.00 g of zinc metal in hydrochloric acid and dilute to 1 liter with water. Dilute 1 0 ml of the stock solution to 1liter to prepare a standard containing 1 0 pg/ml of zinc. Sodium citrate solution: dissolve 1 0 g of sodium citrate in water and dilute to 100 ml. Hydroxylamine hydrochloride solution : dissolve 1 0 g of hydroxylamine hydrochloride in water and dilute to 100 ml. Buffer solution, pH 4.75: dissolve 130 g of sodium acetate and 57 ml of glacial acetic acid in water and dilute to 1 liter. Dithizone solution: dissolve 0.1 g of dithizone in a liter of alcohol-free carbon tetrachloride. Extract any alcohol from the carbon tetrachloride by shaking it with distilled water. Keep a water blanket on the extracted carbon tetrachloride when storing it. Potassium cyanide solution: dissolve 1.0 g of potassium cyanide in water and dilute to 100 ml. Buffer solution, pH 9.0: dilute 213 ml of l sodium hydroxide to 600 ml N with water. Dissolve 37.3 g of potassium chloride and 31.0 g of boric acid in water, mix with the sodium hydroxide, and dilute to 1liter. Zincon solution: dissolve 0.13 g of zincon in 2 ml of 1N sodium hydroxide and dilute to 100 ml with water. Chloral hydrate solution: dissolve 10 g of chloral hydrate in water and dilute to 100 ml. Hydrochloric acid, 0.02N:add 1.7 ml of concentrated hydrochloric acid to water and dilute to 1liter. Ammonium hydroxide. Sodium ascorbate. Sodium thiosulfate solution: dissolve 25 g of sodium thiosulfate in water and dilute to 100 ml. Procedure. Add 1 0 ml of the hydroxylamine hydrochloride solution to an aliquot of brine containing up to 200 pg of zinc, mix, add 1 0 ml of sodium citrate solution, and adjust the pH to 8.3 with ammonium hydroxide. Transfer the sample to a separatory funnel, add 3 ml of 0.01% dithizone solution, and shake the mixture vigorously for 1 minute. Let the phases separate and note the color of the dithizone phase. If any zinc is present, the dithizone phase will be red or violet, but not green. If the dithizone phase is green, take a larger aliquot of brine. If the dithizone phase is red or violet, extract it into another separatory funnel containing 1ml of sodium thiosulfate solution and 1 0 ml of pH 4.75 buffer solution. Make additional extractions of the brine solution with dithizone solution until the dithizone remains green, which indicates that all the zinc has been extracted. This is important because the final dithizone phase must be green, not violet. Discard the brine solution and wash the combined dithizone extracts by mixing them vigorously for 1minute with the buffer solution. Let the phases separate, extract the dithizone phase into another separatory funnel con-

COLORIMETRIC METHODS

103

taining 10 ml of 0.02N hydrochloric acid, and shake this mixture vigorously for 2 minutes. Let the phases separate, and extract and discard the carbon tetrachloride phase. Wash the acid phase twice with carbon tetrachloride, extract, and discard the carbon tetrachloride. Transfer the acid phase to a 50-ml volumetric flask and make t o volume with water. Pipet 10-ml aliquots from the 50-ml volumetric flask to two 50-ml Erlenmeyer flasks. To both flasks add 0.5 g of sodium ascorbate followed by, in this order and with mixing, 1 ml of potassium cyanide solution, 5 ml of pH 9.0 buffer solution, and 3 ml of zincon solution. To one sample add 3 ml of chloral hydrate solution, and t o the other (which is the reference solution) add 3 ml of water. Within 2-5 minutes after adding the last reagent, measure the absorbance of the sample versus the reference solution at 620 mp in 1-cm cells with a spectrophotometer.

Calculations. Prepare a calibration using aliquots of the standard zinc solution containing 10-80 pg of zinc, and use the curve t o calculate the amount of zinc in the sample:
pg Zn (from curve) = mg/l Zn+* ml sample

Cadmium
Cadmium can be extracted from aqueous solutions as cadmium dithizonate into carbon tetrachloride or chloroform. Cadmium dithizonate is extracted more readily into carbon tetrachloride than is zinc dithizonate, but zinc dithizonate is extracted more readily into chloroform than the cadmium compounds. Therefore, because many oilfield brines contain more zinc than cadmium, the cadmium extraction should be made with carbon tetrachloride to insure the best possible separation from zinc. Although citrate and tartrate do not hinder the cadmium dithizonate extraction, they do impede the extraction of lead and zinc. Cadmium dithizonate can be extracted from an alkaline solution containing cyanide and tartrate; the dithizonates of nickel, copper, silver, and tin are not extracted. Most of the interference from iron can be eliminated by oxidizing it with peroxide and filtering. Cadmium reacts with dithizone to form a compound of the type:

104

ANALYSIS O F OILFIELD WATERS

Cadmium dithizonate in carbon tetrachloride absorbs strongly at a wavelength of 620 mp.

Reagents. Standard cadmium solution: dissolve 0.100 g of cadmium metal in hydrochloric acid and dilute t o 1 liter with water. Pipet a 10-ml aliquot of this stock solution into another l-liter volumetric flask and dilute to volume. This solution contains 1pg/ml of cadmium. Ammonium chloride, 1N: dissolve 53.5 g of ammonium chloride in water and dilute to 1liter. Rochelle salt solution: dissolve 100 g of rochelle salt (KNaC4H406 4H20) in water and dilute t o 1liter. Sodium citrate solution: dissolve 100 g of sodium citrate in water and dilute to 1 liter. Hydrogen peroxide, reagent grade 30%hydrogen peroxide. Tartaric acid solution: dissolve 20 g of tartaric acid in 1 liter of water. Store in a refrigerator and discard if any mold is present. No.1 dithizone reagent: dissolve 0.12 g of dithizone in 1 liter of carbon tetrachloride. Store in a refrigerator in a dark bottle. No.2 dithizone reagent: dilute 5 ml of No.1 reagent t o 100 ml with carbon tetrachloride. Store in the refrigerator. Hydroxylamine hydrochloride solution : dissolve 10 g of hydroxylamine hydrochloride in 50 ml of water. Prepare fresh weekly. Sodium hydroxide (35%)-potassium cyanide (1%) solution: dissolve 175 g of sodium hydroxide and 0.5 g of potassium cyanide in water and dilute to 1 liter. Ammonium hydroxide 5M: dilute 16.0 ml of concentrated ammonium hydroxide (14.8M) to 50 ml. Sodium hydroxide, 5% solution: dissolve 5 g of sodium hydroxide in water and dilute to 100 ml.

Procedure. Filter the brine through Whatman No.4 filter paper (double thickness). Transfer 900 ml or less of the filtered brine to a 2-liter beaker, add 5 ml of 30% hydrogen peroxide, and heat until complete decomposition of the excess hydrogen peroxide is attained. Cool the solution and filter if any precipitate is present. Add 100 ml of ammonium chloride solution, 10 ml of rochelle salt solution, 25 ml of sodium citrate solution, and adjust the pH to between 8 and 8.5 with 5M ammonium hydroxide. Transfer the solution to a liter separatory funnel, add 15 ml of the No.1 dithizone solution, and shake the mixture vigorously for 5 minutes. Let the phases separate and extract the dithizone phase into a 50-ml separatory funnel. Reextract the brine with another 15 ml of No.1 dithizone solution. Separate the dithizone phase into the 50-ml separatory funnel and discard the brine phase. Add 10 ml of tartaric acid solution to the combined dithizone extractions in the 50-ml separatory funnel and shake the mixture vigorously for 2

COLORIMETRIC METHODS

105

minutes. Discard the carbon tetrachloride phase and wash the tartaric acid phase twice with a 3-ml portion of carbon tetrachloride. To the tartaric acid phase add 1 ml of hydroxylamine hydrochloride solution, 5 ml of the 35% sodium hydroxide-1% potassium cyanide solution, and 10 ml of the No. 2 dithizone solution Shake the mixture vigorously for 1minute. Let the phases separate and extract the dithizone phase into another 50-ml separatory funnel. Reextract the aqueous phase with 10 ml of No.2 dithizone solution, and add the dithizone extraction of the previous separation. Wash the aqueous phase with 5 ml of carbon tetrachloride, extract the carbon tetrachloride, and combine it with the two dithizone extractions. Discard the aqueous phase. Add 15 ml of 5% sodium hydroxide solution to the combined dithizone extractions, shake the mixture vigorously for 1 minute, extract the carbon tetrachloride phase, and determine its absorbance at 620 mp in a l-cm cell with a spectrophotometer.

Calculations Run a blank and make appropriate corrections, using a calibration curve prepared by using aliquots of the standard cadmium solution containing 1-7 pg of cadmium:
pg Cd (from curve) = mg/l Cd+2 ml sample Phosphate Only orthophosphate will respond to the test. Polyphosphates must be reverted to orthophosphates by boiling with acid (American Petroleum Institute, 1968).

Interferences. Color development in the test is inhibited when the dissolved solids content of the sample is greater than 8% (a specific gravity greater than 1.06) or when the total iron is greater than 50 mg/l. In such cases the sample taken for analysis must be diluted with distilled water so that these limits are not exceeded. Sulfide interferes by giving high results, and should be destroyed by adding potassium permanganate solution t o the acidified sample. Reagents. Hydrochloric acid, concentrated. Reagent No. 1: dissolve 46 g of ammonium molybdate [ (NH), )a Mo, 02, 4H201 in 700 ml of distilled water. The ammonium molybdate used should consist of white crystals without a bluish-green tinge. Add 2.5 ml of concentrated ammonium hydroxide to the solution and dilute t o 1liter with distilled water. Amino solution: dissolve 10 g (about 1 level tablespoon) of amino powder mixture in 100 ml of distilled water. If solution remains turbid, filter. Store solution in a well-stoppered, brown glass bottle and prepare fresh at least every 2 weeks.

106

ANALYSIS OF OILFIELD WATERS

The amino powder mixture is made up by adding 5 g of sodium sulfite and 1.0 g of l-amino-2-naphthol-4-sulfonic to a dry mortar. Grind the acid materials to a fine powder. Transfer the powder t o a large wide-mouthed bottle containing 66.5 g of sodium bisulfite (meta, powder, Naz Sz 0, ) and 35 g of sodium sulfite. Mix well by shaking. If the mix is not uniform, it should be passed through a 20-mesh screen and again shaken in the large bottle. Store mixture in a well-stoppered, wide-mouthed brown bottle. Standard phosphate solution: dissolve in distilled water 0.1335 g of potassium dihydrogen phosphate (KHz PO4 ) which has been dried in an oven at 105C. Dilute t o 1 liter. 1ml of this solution is equivalent to 0.1 mg sodium metaphosphate (NaP03).

Procedure. Thoroughly shake a freshly drawn sample to disperse the solids and pipet 100-ml aliquot into each of two 250-ml beakers. If the expected concentration of sodium metaphosphate is greater than 10 mg/l, take smaller aliquots diluted t o 100 ml with distilled water. Note: phosphate-free glassware must be used in this determination. The glassware should be soaked in dilute hydrochloric acid, followed by rinsing with distilled water. Add 7 ml of concentrated hydrochloric acid to one of the samples. If it is suspected t o contain sulfide, stir the solution vigorously for a minute to remove as much of the sulfide as possible, then add potassium permanganate solution (8%) dropwise until the solution just turns pink. Boil solution vigorously for 30 minutes while maintaining the volume between 75 and 100 ml by adding distilled water. Cool sample t o a temperature between 70" and 95F and dilute t o 107 ml with distilled water in a graduated cylinder bearing a mark at the 107-ml level. Add 7 ml of concentrated hydrochloric acid t o the unboiled sample and treat with permanganate as above if sulfide is suspected. Filter both boiled and unboiled samples if turbid. Add 5 ml of reagent No.1 to both samples and mix well. Add 5 ml of amino solution t o both and again mix well. Ten minutes after the amino solution addition, measure the color with a spectrophotometer at a wavelength of 690 mp, after adjusting the meter to 100%transmittance with a proper blank. Calibration curve. Prepare a calibration curve by using aliquots of the standard phosphate solution containing up t o 10 mg/l of sodium metaphosphate. Calculations. Refer the spectrophotometer readings t o the calibration curve (expressed as milligrams of NaP03 versus photometer reading) t o obtain the sodium metaphosphate concentration. The results on the heated sample correspond to total phosphate, whereas, those on the unheated sample correspond to orthophosphate, the difference being polyphosphate, usually expressed as sodium metaphosphate (NaP03):

COLORIMETRIC METHODS

107

where A = mg NaPO, (heated sample), and B = mg NaPO, (unheated sample).

Precision. The precision is about k3% of the amount present.


Silica
Silicon is the second most abundant element in the earths crust and is found in most rocks as the oxide Si02 or as a silicate such as Mg,Si2O5 (OH), . The solubilities of silicate minerals in saline waters are a function of temperature, pressure, pH, Eh, dissolved gases, and other ions in solution. A limited amount of research has been done concerning silicate solubilities (Collins, 1969) in saline solutions. Some investigators believe that most silica exists in solution as H4Si04 (White et al., 1956);others that it exists both in colloidal form and as H4Si04 (Krauskopf, 1956).Hydration of silica gives the following reaction: Si02 + 2H20 + Si(OH), or H4 Si04
A method developed by Schrink (1965) was used to study silicate solubilities in saline waters (Collins, 1969)and it gave satisfactory results. It also has been used to analyze some petroleum-associated waters. The method involves adding 1 ml of a 4% ammonium molybdate solution in 0 7 molar .5 sulfuric acid solution to an appropriate aliquot of the water sample; add 15 ml of 4.5N sulfuric acid; extract for 1 minute with ethyl acetate; and determine the absorbance of the ester extract with a spectrophotometer at a wavelength of 335 mp.

Nitrate nitrogen Nitrate is the most highly oxidized form of nitrogen and is the most stable form in an oxidizing environment. Many fertilizers contain nitrate, and waters will leach the nitrate from soil or rock. Most rocks do not contain much nitrate; therefore, it is unlikely that petroleum-associated waters contain appreciable quantities of nitrate. The nitrate in deep waters also may be depleted through anion exchange (George and Hastings, 1951). Chloride is a serious interference in many of the methods used to determine nitrate nitrogen. Oxidizing or reducing agents such as ferric or ferrous iron also interfere. The Brucine method (Fisher et al., 1958)can be applied to a petroleum-associated water. To determine the nitrate concentration, transfer an aliquot of the sample containing up to 15 pg of nitrate into a 50-ml Erlenmeyer flask, add 15 ml of water, 1 ml of Brucine reagent (2%

108

ANALYSIS OF OILFIELD WATERS

aqueous solution of Brucine hydrochloride), acid, place in a dark area and allow to cool to 30C. Determine the absorbance of the sample with a spectrophotometer at a wavelength of 410 mp.
Arsenic

The determination of arsenic in brines has received little attention despite its toxic relationship t o fish and animals. The arsenic content of sea water was first investigated by Gautier (1903),who found inconsistent variations. He attributed the higher amounts found at great depth to volcanic influences, and the higher amounts found at the surface t o evaporation and disturbances caused by marine animals. and Gorgy et al. (1948)also studied arsenic in Rakestraw and Lutz (1933) sea water. They conclude that 50-60% of the arsenic is in the arsenite form, with 8--10% each of arsenate, dissolved organic arsenic, and arsenic suspended in particulate matter. Smales and Pate (1952)used an activation analysis method t o determine submicrogram quantities of arsenic in sea . water. They found an average of 2 6 pg of arsenic per liter, with a range of 1 .6-5.0 pg/l. The water analyzed is believed representative for Atlantic Ocean water. The Gutzeit method can be used to analyze a petroleum-associated water for arsenic (Collins et al., 1961). Arsenic is reduced t o arsine with zinc in acid solution. A yellow t o brown stain is produced when AsH3 passes through paper impregnated with mercuric chloride or mercuric bromide. The color- yellow, A s ( H ~ B ~ ) ~ - brown, and ation is produced by A s H ( H ~ B ~ ) ~ As2Hg3 - black. By comparing unknowns with a series of standard papers prepared with known amounts of arsenic, a quantitative estimation can be made. Papers prepared from mercuric bromide can be preserved for several months in a dark, dry atmosphere. Arsenic silver diethyldithiocarbamate method Arsine gas is liberated from arsenic compounds upon the addition of zinc in an acid medium (Stratton and Whitehead, 1962). The arsine gas is passed through a lead acetate scrubber and into an absorbing tube containing silver diethyldithiocarbamate solution. The arsine and the silver diethyldithiocarbamate solution react forming a red color that can be measured spectrophotometrically. Apparatus. Arsine generator, scrubber, and absorber. Spectrophotometer set at the following operating conditions: wavelength - 535 mp; cells - 10 mm; phototube -blue sensitive; and slit width - 0.02 mm. Reagents. Standard arsenious oxide solution: dissolve 1.320 g of As203 in

COLORIMETRIC METHODS

109

10 ml of 1OM NaOH and dilute to 1 liter with distilled water. 1 ml of this g Dilute this stock standard solution as solution contains 1.00 m of A% 0 3 . required. Hydrochloric acid, concentrated, analytical-grade. Lead acetate solution: dissolve 1 0 g of Pb(C2H302)2 3 H 2 0 in distilled * water and dilute to 100 ml. Potassium iodide solution: dissolve 1 5 g of KI in distilled water and dilute to 100 ml. Store in an amber colored bottle. Silver diethyldithiocarbamate solution: dissolve 1 g of AgS[SN(C2H5 )2 ] in 200 ml of pyridine. Store in an amber colored bottle. Stannous chloride solution: dissolve 40 g of arsenic-free SnC12*2H2 in 0 1:3 HC1 and dilute to 100 ml with the same acid. Zinc, 20 mesh, arsenic-free.

Procedure. Place a 25-ml sample, or suitable aliquot, containing less than 20 pg of arsenic in a Gutzeit generator. Add to the flask successively, 5 ml of concentrated HCl, 2 ml of KI solution, and eight drops of SnC12 solution. Thoroughly mix after each addition. Allow 15 minutes for reduction of the arsenic to the tervalent state. Insert a plug of glass wool that has been impregnated with the lead acetate solution into the scrubber. Assemble the generating apparatus and add 4 mi of the silver diethyldithiocarbamate solution to the absorber. Glass beads should be added to the absorber until the liquid just covers them. Add 3 g of zinc to the generator and reconnect immediately. Allow 30 minutes for complete evolution of the arsine. Warm the generating flask gently to assure complete evolution of the arsine and then pour the solution from the absorber directly into the spectrophotometer cells. Make the determinations within 30 minutes as the color developed is not permanent.

CuZcuZutions. The quantity of arsenic in the sample is determined from a plot of absorbances of the standards:
pg As (from curve) = mg/l As ml sample

Fluoride Because of interferences from large amounts of chloride present in petroleum-associated waters, a standard addition method was developed which is accurate in the presence of large amounts of chloride and sulfate and is more rapid than methods requiring distillation (Collins et al., 1961). Up to 0.01 mg of phosphate in the aliquots taken for analysis can be tolerated. Larger amounts of phosphate than this decolorize the zirconium

110

ANALYSIS OF OILFIELD WATERS

cyanine R complex completely, and distillation is necessary to remove the phosphate.

Reagents. Eriochrome Cyanine R stock solution: dissolve 1.80 g of Eriochrome Cyanine R in 200 ml of distilled water. Zirconyl nitrate stock solution: dissolve 0.40 g of zirconyl nitrate dihydrate in 100 ml of concentrated hydrochloric acid and dilute to 200 ml. Fluoride indicator solution: add 20.0 ml of the Eriochrome Cyanine R solution to about 500 ml of water, stir and add 10.0 ml of the zirconyl nitrate solution, 75 ml of concentrated hydrochloric acid, and 4 g of barium chloride. This mixture is stable for 4 - 6 months. Thiosemicarbazide, powdered solid. Procedure. Measure equal amounts of brine containing less than 0.03 m of g fluoride into each of three 50-ml volumetric flasks. Add lOpg of fluoride to one of the flasks and add 20 pg to another. Add a few milligrams of solid thiosemicarbazide and 25 ml of fluoride indicator solution to each 50-ml volumetric flask. If sulfate is present, it will precipitate as barium sulfate and must be centrifuged out of suspension. Arbitrarily adjust the transmission of the blank (25 ml of fluoride indicator solution made to 50-ml volume with distilled water) at 540 mp to 32% and measure the transmission of the three solutions. Calculations. Using coordinate graph paper, plot the transmission of the standard-addition samples on the y-axis and their concentrations in milligrams of fluoride per liter on the x-axis. Multiply the sample reading at 0 concentration by 2, and from this point on the y-axis, draw a line parallel to the x-axis until it intersects the line plotted. From this point of intersection, draw a line parallel to the y-axis until it intersects the x-axis. This value from the x-axis multiplied by the dilution factor equals the amount of fluoride in milligrams per liter. Fig. 3.3 illustrates this procedure.
Iodide

A rapid, accurate method for the determination of iodide suitable for field work utilizes the principle whereby iodide is oxidized to iodine with nitrous acid and extracted into carbon tetrachloride. Hydrogen sulfide will interfere, but it can be removed by acidifying the sample and boiling (Collins et al., 1961).

Reagents. Bromphenol blue: dissolve 0.1 g of bromphenol blue in 100 ml of distilled water. Carbon tetrachloride. Iodide standard solution: dissolve 1.3081 g of potassium iodide in distilled water and dilute to 1,000ml. 1 ml contains 1 m of iodide. g

COLORIMETRIC METHODS

111

Potassium nitrite solution: dissolve 10 g of potassium nitrite in 100 ml of distilled water. Sulfuric acid, 9N.

Procedure. Pipet a sample containing less than 3 m of iodide into a separag tory funnel, and add three drops of the bromphenol blue solution and a few drops of 9N sulfuric acid until the indicator turns yellow. Add 10 ml of carbon tetrachloride and 1 ml of a 10% aqueous potassium nitrite solution, and vigorously mix the combined phases. Extract the carbon tetrachloride phase into a glass-stoppered cylinder. A violet color in the carbon tetrachloride indicates iodine. Repeat the extractions with 5-ml portions of carbon tetrachloride until all of the iodine is extracted. Dilute the combined extracts to 25 ml with carbon tetrachloride and measure the absorbance using a spectrophotometer at a wavelength of 517 mp. Use a calibration curve prepared with standard iodide solutions t o determine the milligrams of iodide in the sample. Calculation:
mg I (from curve) x 1,000 = mg/l ml sample Selenium Selenium can be reduced t o the elemental form with sulfur dioxids (Collins et al., 1964), hydrazine, hydroxylamine hydrochloride, hypophosphorous acid, ascorbic acid, and stannous chloride. From hydrochloric acid solutions exceeding 8N,selenium is precipitated free of tellurium when the reducing agent is sulfur dioxide. Both selenium and tellurium are precipitated by sulfur dioxide from 3 t o 5N hydrochloric acid solutions. Traces of nitric acid should be removed before sulfur dioxide reduction. When precipitating selenium, it is important that the temperature of the solution be kept below 30C because the volatile selenium monochloride easily can form and be lost. A large excess of reducing agent helps to prevent loss of the monochloride. Selenium can be determined semiquantitatively by comparing the color of the red amorphous form, or it can be adjusted to the quadrivalent form, reacted with 3,3-diaminobenzidine to form the monopiazselenol, and quantitatively determined spectrophotometrically. If sufficient selenium is present, it also can be determined gravimetrically . Selenate (VI) can be reduced to selenite (IV) by heating in concentrated hydrochloric acid. Selenite is the only form that reacts with 3,3-diaminobenzidine; the reaction is :
% N
H2 N

- NH2 + H, SeO,

N=

NH2 + 3Ha0 NH2

NH2

SeN

112

ANALYSIS OF OILFIELD WATERS

Selenium adsorption on glassware can introduce a significant error. Much of this adsorption can be eliminated by treating the glassware with a solution of chlorosilane.

Reagents. Hydrobromic acid, 48%. Selenium, stock solution: dry some selenium dioxide by placing it in a desiccator over phosphorous pentoxide for 24 hours. Dissolve 0.141 g of the dry selenium dioxide in water, add 80 ml of 48% hydrobromic acid, and dilute to 1 liter with water. 1 ml of this solution contains 0.1 mg of selenium. (Note: particles of red selenium may appear in this stock solution after long standing as a result of reduction. When this happens, a new stock solution must be prepared.) Selenium solution: pipet 100 ml of the selenium stock solution into a 1-liter volumetric flask, add 80 ml of 48% hydrobromic acid, and dilute with water. 1ml of this solution contains 0.01 mg, or 10 pg of selenium. Sulfur dioxide selenium free. Hydrochloric acid, concentrated. Sulfuric acid, concentrated. 3,3'-diaminobenzidine hydrochloride: dissolve 0.25 g of 3,3'-diaminobenzidine hydrochloride in 50 ml of water. Prepare a fresh solution each day. Formic acid, 2.5M: dissolve 11.5 g of formic acid in water, and dilute to 100 ml with water. Toluene, spectro-grade. Ammonium hydroxide: dilute 10 ml of concentrated ammonium hydroxide to 100 ml with water. Barium chloride solution: dissolve 5 g of barium chloride in 100 ml of water. EDTA solution, 0.1M: dissolve 37.225 g of disodium ethylenediaminetetraacetate in water and dilute t o 1liter.
Procedure. Pipet an aliquot of brine (50 ml or less) into a 100-ml volumetric flask and dilute to volume with concentrated hydrochloric acid. If desired, the detection limit can be increased by first concentrating the brine by careful evaporation after adjusting the pH t o 2 with hydrochloric acid. Mix the solution and allow it t o stand until most of the sodium chloride precipitates. Carefully withdraw 50 ml of the supernatant clear liquor into a 150-ml beaker and add 10 ml of concentrated hydrochloric acid. Heat the mixture to near boiling for 10 minutes. Place the beaker in an ice-water bath beneath an exhaust hood, let the mixture cool t o the temperature of the ice water, and then bubble sulfur dioxide gas rapidly into the solution for about 8 minutes. If a heavy turbidity develops, filter the solution through a micropore filter. Wash the precipitate with 20 ml of cold water if a 30-ml crucible is used, or with 5 ml if a 1.5-ml crucible is used. Take care that no air is

COLORIMETRIC METHODS

113

pulled through the precipitate until the entire filtration and washing process is complete. Transfer the filter and precipitate back t o the 150-ml beaker and add 5 ml of a 1:l mixture of hydrochloric acid and nitric acid. Heat the mixture to near boiling for a few minutes, taking care not t o let the mixture boil violently or to dryness. Examine the mixture carefully t o make sure that all of the selenium has dissolved. Place the beaker containing the mixture in a vacuum desiccator over anhydrous magnesium perchlorate and sodium hydroxide and let the mixture evaporate t o dryness. Dissolve the residue in 5 ml of concentrated hydrochloric acid and heat the mixture to near boiling for a few minutes. Cool the mixture, add 20 ml of water, and filter it through Whatman No.4 filter paper into a 100-ml volumetric flask. Adjust the volume to 100 ml and pipet an aliquot containing 1-100 pg of selenium (IV) from the 100-ml flask into a 100-ml beaker. Add 5 ml of 0.1M EDTA and 2 ml of 2.5M formic acid and adjust the pH to 1.5 with hydrochloric acid. Adjust the volume to about 50 ml with water, add 2 ml of 3,3'-diaminobenzidine solution, mix, and let stand for 30 minutes. Adjust the pH to 8 and transfer the solution to a 1 2 5 4 Teflon-stoppered separatory funnel containing 10.0 ml of toluene. Shake this mixture vigorously for 2 minutes and let the phases separate. Extract the toluene phase, which now contains the monopiazselenol, into a centrifuge tube. Centrifuge briefly t o clear the toluene of water droplets. If a centrifuge is not available, the organic phase can be filtered through a dry filter paper to which has been added 100 mg of anhydrous sodium sulfate. Determine the absorbance of the toluene phase at 420 mp versus a reagent blank. 1-cm cells are used; however, longer path length cells will increase sensitivity.

Calculation. Prepare a calibration curve by plotting log I,JI, which is the extinction or optical density of the solution versus the concentration, using solutions containing known amounts of selenium and treated as previously described. Estimate the amount of selenium from this curve, and calculate as follows:
g Se = mg/l Se-* ' ml sample

Semiquantitative determination of selenium


Pipet a 20- to 50-ml aliquot of brine into a 100-ml volumetric flask and dilute t o volume with concentrated hydrochloric acid. (To increase the detection limit, the brine can first be concentrated by careful evaporation after acidifying it to pH 2 with hydrochloric acid.) Mix the solution and allow it to stand until most of the sodium chloride precipitates. Withdraw an aliquot of the supernatant clear liquor into a small beaker,

114

ANALYSIS OF OILFIELD WATERS

add additional concentrated hydrochloric acid if necessary, and bubble sulfur dioxide gas into the solution for 3 minutes. If the solution remains clear, there is less than 25 pg of selenium present. Filter the solution through a 5-ml micropore filter. Compare the color in the crucible with a series of color standards comprising 3-20 pg of selenium. These cofar standards are prepared with known amounts of selenium and will give the fobwing colorations (in pg of selenium): 3 - very pale yellow; 6 - very pale orange; 10 - pale orange; 15 - orange; and 20 - red orange. Barium Qualitative test This test can be used to detect barium and strontium in an oilfield brine. It is possible to detect barium and strontium individually by using chromate to precipitate the barium. Transfer an aliquot of brine to a test tube, add a few millimeters of 0.5% aqueous sodium rhodizonate solution, stopper the tube, and shake the mixture vigorously. Barium and/or strontium is present if a bright red, a brownish-red, or a yellow-red precipitate forms. The deeper brown indicates barium, while the lighter yellow may indicate strontium. In any event, if a precipitate forms, barium and/or strontium is present. A series of standards can be prepared to help in determining the approximate amounts present. To differentiate between barium and strontium, a few milliliters of a 10% aqueous solution of ammonium chromate can be added to a sample brine 30-60 minutes before the sodium rhodizonate solution is added. The more soluble strontium chromate will react with the rhodizonate while the less soluble barium chromate will not.

GRAVIMETRIC METHODS

Gravimetric methods involve isolating a compound and determining its weight. Their use can involve considerable time because preliminary separations often are necessary to remove interfering elements; e.g., to determine barium as the sulfate, all strontium should be removed before the final precipitation of the barium sulfate. One constituent present in most oilfield waters that has resisted development of a good instrumental method of analysis is sulfate, and perhaps the most accurate method to determine sulfate in oilfield waters is still the gravimetric method. Sulfate Sulfate is precipitated as barium sulfate from an acid solution. The precipitate is baked, cooled, and weighed.

GRAVIMETRIC METHODS

115

Reagents. Hydrochloric acid, concentrated. Barium chloride solution, 10%aqueous. Procedure. Use an aliquot that will produce no more than 100 mg of precipitated barium sulfate. Dilute the aliquot t o 250 ml with distilled water and add 1 ml of hydrochloric acid. If the sample volume itself is larger than 250 ml, add 1 ml of hydrochloric acid per 250 ml of volume. Heat t o boiling and add an excess of hot, 10% barium chloride solution while stirring. Cover the solution and allow it t o stand for about 4 hours at a temperature of about 85OC. Filter through a very retentive filter paper such as Munktells No.OOH or Whatman No.42, and wash with hot water until the filtrate is chloride free. Place the filter plus precipitate in a tared crucible, char slowly without igniting, and bake at 800C for 1 hour. Place the crucible in a desiccator t o cool and then weigh. Calculation:

Barium
Interest in the knowledge of the barium concentration in most petroleum-associated waters has spurred the development of several types of methods to determine barium. Perhaps the most rapid but least accurate method is the turbidimetric method, which measures the turbidity of the sample caused by precipitated barium sulfate after addition of excess sulfate. The gravimetric method also measures precipitated barium sulfate or more preferably barium chromate. It is a more time-consuming method than the turbidimetric method, but will yield more accurate results. The double precipitation as chromate reduces the interference from calcium and strontium.

Reagents. Ammonium chromate solution: dissolve 10 g of ammonium chromate in water, dilute t o 100 ml, and filter. Ammonium chromate wash solution: dissolve 5 g of ammonium chromate in water and dilute to 1 liter. Adjust the pH of this solution to 4.6 with ammonium acetate or acetic and filter. Ammonium acetate solution: dissolve 30 g of ammonium acetate in water, dilute to 100 ml, and filter. Nitric acid, 4N: cautiously add 30 ml of concentrated nitric acid t o 90 ml of water. Ammonium hydroxide, concentrated. Hydrogen peroxide, 30%solution. Procedure. Because iron will interfere if present, it should be removed by

116

ANALYSIS OF OILFIELD WATERS

adding a few drops of hydrogen peroxide to the sample before heating to near boiling, adding ammonium hydroxide dropwise, and stirring until the odor of ammonia is faint but distinct. Heat to boiling to remove excess peroxide and flocculate any precipitate, and then filter out the iron hydroxide. To an aliquot of iron-free filtered water (the water should be filtered even if iron is not specifically removed) containing less than 500 mg of barium and strontium, add acetic acid until the pH is 4 6 Then add 1 0 ml of .. ammonium chromate and 1 ml of ammonium acetate. Readjust the pH to 4 6 The final volume should be about 200 ml. Boil the mixture for 5 .. minutes, stirring occasionally. Remove the mixture from the heat, cool it fairly rapidly to room temperature, and allow it to stand at room temperature for 1hour. Filter the solution through a fine porosity filter. Wash the precipitate from the beaker into the filter with the ammonium chromate wash solution. Since a second precipitation is made, it is not necessary to police the beaker. Wash the precipitate on the filter with 50 ml or more of ammonium chromate wash solution or until calcium and strontium are absent. Dissolve the precipitate in 3 or 4 ml of 4N nitric acid. Transfer the dissolved precipitate back to the beaker and repeat the precipitation. The same filter can be used, but make sure that it is acid-free. Dry the second precipitate for 1 hour at llOC or until it reaches a constant weight.

Calcuhtions. Weigh the barium chromate and calculate barium as follows: mgBaCr04 x 542
ml sample
= mg/l Ba+

Precision and accuracy. The precision and accuracy of this method with optimum conditions are 1% 2% respectively, of the amount of barium in and
the sample.
OTHER METHODS

The approximate concentration of sodium in an oilfield water can be calculated by using a knowledge of the amounts of other major cations and anions in the sample. Likewise, the dissolved solids concentration in an oilfield water can be calculated. Sodium The practice of determining sodium by calculation does not give an accurate sodium value. For example, this value is calculated after determining all of the major common anions plus two or more cations, usually calcium and

OTHER METHODS

117

magnesium. The excess of equivalents per million of anions over cations is assumed to be sodium plus potassium, and this practice includes all the errors of the analysis plus the undetermined ions in the combined sodium plus potassium value. The ions are converted t o milliequivalents per liter (me/l) by dividing each ion concentration (mg/l) by its milliequivalent weight (mglme) to give the milliequivalents per liter for each ion determined. After adding the milliequivalents per liter for both the anions and the cations, the difference is multiplied by the milliequivalent weight of sodium to give the calculated milligrams per liter of sodium.
Procedure. The calculation method is demonstrated as follows:

chloride (50,000 mg/1)/(35.5 mg/me) = 1,410 me/l sulfate (1,290 mg/1)/(48.0 mg/me) = 27 me/l bicarbonate (204 mg/1)/(61.0 mg/me) = 3 me/l Total anions 1,440 me/l Cations: calcium (5,900 mg/1)/(20.0 mg/me) = magnesium (2,000 mg/1)/(12.1 mg/me) = Total cations Determination of sodium: (1,440 - 459 me/l) x 23.0 mg/me = 22,600 mg/l Dissolved solids The dissolved solids determination can be used t o estimate the accuracy of the resistivity determination. The specific gravity determination, and the evaporation method can be used t o double check the calculated total dissolved solids. Theoretically, if all the dissolved solids are accurately determined, their sum will equal the weight of the residue left after evaporation of the water. The dissolved solids include all the solid material in solution which is ionized, or which is not ionized but does not include suspended material, colloids, or gases. The residue method involves evaporating a filtered sample t o dryness followed by drying the residue in an oven at 180C for 1 hour. The cooled residue is weighed and the total dissolved solids are calculated; e.g., if 100 ml of brine is evaporated and the residue weighs 3.0 g, then the dissolved solids equal 30,000 mg/l. The evaporation method is subject to errors when hygroscopic material such as calcium chloride is in the water, as is usually the case in oilfield waters. The calculation method simply involves adding the sum of all the analyzed constituents as follows: 295 me/l 164 me/l 459 me/l

Anions:

118

ANALYSIS OF OILFIELD WATERS

Constituent Na+

K+
Li+ Ca+ Mg+ c1 Br- HC03
S04-2

Concentration(mg/l) 13,500 400 10 2,000 1,200 23,500 500 1,200 1,200 43,500

Total dissolved solids Spent acid

Hydrochloric acid is the oldest and most common solution used in oil-well acidizing (Halliburton Company, 1970)..Many additives and other acids may be used in conjunction with HC1, for example HF and HAc. Normally, 15% HCl is used; however, other strengths are quite common. These acid solutions are pumped into carbonate formations t o dissolve and remove a part of the formation. After reacting with carbonate rocks or being spent on the formation, the solutions are returned t o the surface by various means. Often, they are mixed with formation water, and an operator may want t o know when the spent acid has been recovered, or if formation water or a mixture of solution and water is being produced. When 15% HC1 is completely spent on CaC03 or M g C 0 3 , the resulting solution will contain 90,000 mg/l Ca or Ca equivalent. The normal formation water contains only about 10,000 mg/l Ca or Ca equivalent. The procedure is based on these differences. Reagents and equipment. The necessary reagents and equipment include: Calcium carbonate, 10-mesh. NH4OH, reagent. Whatman No.31 filter paper. Plastic funnel. 150-ml beakers. Graduated cylinder, 25 ml. 1-ml syringe or pipette, preferably plastic. 0.5% Eriochrome Black T indicator. 2N NaOH solution. CDTA solution (disodium dihydrogen 1,2-cyclohexanediamine-tetraacetate): dissolve 100.0 g CDTA in 900 ml water and dilute to 1 liter. 1 ml equals 9.0 mg Ca.

OTHER METHODS

119

Buffer solution: 67.5 g NH4C1, 570 ml NH40H made to 1 liter with distilled water. Procedure: (1) Determine the pH of the returned water. If pH is below 4, the presence of HC1 is indicated. (2) Pour 10-15 ml of the sample into a beaker containing 10 g of 10-mesh CaC03. Bring t o a boil, remove from the hotplate, allow to settle for about 5 minutes, and filter. (3) Pipet 1.0 ml of the filtrate into a beaker containing 50 ml of HZO. Heat to boiling, add 1 ml NH40H while stirring, remove from heat, let settle for a few minutes, and filter through Whatman No.31 paper. Wash the beaker and filter twice, using 25.0 ml H20 for each wash. (4)Add 0.5 g Eriochrome Black T indicator t o the filtrate and 10 ml of the buffer solution (pH should be 10). Titrate with standard CDTA solution (1ml = 9.0 m Ca) t o a permanent clear blue endpoint. Record the milliliters g of CDTA used. Refer t o a curve to determine the percent spent acid in the sample. (5) To determine a blank, take 1.0 ml of the formation water through the procedure, starting at step 3 and determine 0% spent acid, or the blank correction. Curve construction It is desirable t o construct a curve tpercent spent acid versus milliliter CDTA) for the determination of spent acid. On rectangular graph paper, plot

Example: I m l formation water = 1.3 ml CDTA

10 0

I ml return water = 5.8 ml CDTA p H r e f w n woter = 6.0 I ml CDTA = 9.0 mg Ca

Fig. 3.12. Graph for use in calculating the amount of spent mineral acid in a water sample.

120

ANALYSIS OF OILFIELD WATERS

the blank titration (formation water) as 0% spent acid. Draw a straight line from this point through the intersection of the 100% spent acid and the 10.0-ml CDTA lines as illustrated in Fig. 3.12. This procedure corrects for any Ca+ or Mg+2 present in the dilution water. In cases where it is impossible t o obtain formation water for the 0%spent acid, a reasonable approximation can be made by titrating 100 ml of the water used for washing and dilution. To this volume of CDTA, add 1.3 ml. This value can then be used for the 0%spent acid point on the plot.

Free HCl
When free HC1 is indicated (pH below 4), and it is t o be determined, an additional sample is required. Withdraw 1.0 ml of clear sample. Start with step 3 and follow the procedure. The free HC1 is determined by the difference of the two titrations:
% free HC1= (A -B) x 1.5

where A = ml standard CDTA to titrate CaCO, treated sample, and B = ml standard CDTA t o titrate sample.

Acetic acid solutions


Generally, acetic acid solutions are mixtures of acetic acid and HC1.

Example: 10% HCI t 5% acetic acid I m l = 7.0 ml CDTA I m l formation waler = 1.3 ml CDTA I ml return water = 4 . 2 ml CDTA I ml CDTA = 9.0 mg Ca Return canlains 51% rpent acid

ml, CDTA

Fig. 3.13. Graph for use in calculating the amount of spent mineral and organic acid in a water sample.

REFERENCES

121

Various proportions of each are common. The determination is complicated by the fact that acetic acid will not completely spend itself on calcium and magnesium carbonates. At a pH of 5-6, considerable free acetic acid is still present in the solution and this necessitates a modification of the procedure. In this case, it is necessary to have a representative sample, or t o prepare a sample of the original acid mixture used on the acid job. Take 10-15 ml of the treating acid and 10-15 ml of the returned water through the same procedure as outlined for HC1. Again a plot is constructed, percent spent acid versus milliliters CDTA. Plot the milliliters CDTA used by the formation water as 0%spent acid and the milliliters CDTA used by the injected acid sample as 100%spent acid as illustrated in Fig.3.13. Connect these points by a straight line. From the curve, determine the percent spent acid in the sample of returned water. Other acid mixtures are sometimes used in oil wells. The handling of these are usually too complicated for a rapid field determination. References
American Petroleum Institute, 1968. API Recommended Practice for Analysis o f Oilfield Waters. Subcommittee on Analysis of Oilfield Waters, API RP 45, 2nd ed., 49 pp. Angino, E.E. and Billings, G.K., 1967. Atomic Absorption Spectrometry in Geology. American Elsevier, New York, N.Y., 144 pp. Ballinger, D.G., Booth, R.L., Midgett, M.R., Kroner, R.C., Kopp, J.F., Lichtenberg, J.J., Winter, J.A., Dressman, R.C., Eichelberger, J.W. and Longbottom, J.E., 1972. Handbook f o r Analytical Quality Control in Water and Wastewater Laboratories. National Environmental Research Center, Cincinnati, Ohio, 107 pp. Bogomolov, G.V., Kudelskii, A.V. and Kozlov, M.F., 1970. Ammonium as one of the indications of oil-gas content. Dokl. Akad. Nauk S.S.S.R., 195:938-940 (in Russian). Brooks, R.R., Presley, B.J. and Kaplan, I.R., 1967. APDC-MIBK extraction system for the determination of trace elements in saline waters by atomic absorption spectrophotometry. Talanta, 14:809-816. Burriel-Marti, F. and Ramirez-Munoz, J., 1957. Flame Photometry. American Elsevier, New York, N.Y., 531 pp. Collins, A.G., 1962. Methods of analyzing oilfield waters: flame-spectrophotometric determination of potassium, lithium, strontium, barium, and manganese. US. Bur. Min. Rep. Invest., No. 6047,18 pp. Collins, A.G., 1964. Eh and pH of oilfield waters. Prod. Monthly, 29:ll-12. Collins, A.G., 1965. Methods of analyzing oilfield waters: cesium and rubidium. U.S. Bur. Min. Rep. Invest., No. 6641,18 pp. Collins, A.G., 1967. Emission spectrometric determination of barium, boron, iron, manganese, and strontium in oilfield waters. Appl. Spectrosc., 21 :16-19. Collins, A.G., 1969. Solubilities of some silicate minerals in saline waters. U.S. O f f .Saline Water Res. Dev. Progr. Rep., No. 472, 27 pp. Collins, A.G., Castagno, J.L. and Marcy, V.M., 1969. Potentiometric determination of ammonium in oilfield brines. Environ. Sci. Technol., 3:274-275. Collins, A.G., Waters, C.J. and Pearson, C.A., 1964. Methods of analyzing oilfield waters: selenium and tellurium. U.S.Bur. Min. Rep. Invest., No.6474, 19 pp. Collins, A.G., Pearson, C., Attaway, D.H. and Ebrey, T.G., 1962. Methods of analyzing oilfield waters metallics: copper, nickel, lead, iron, manganese, zinc, and cadmium. US.Bur. Min. Rep. Invest., No. 6087,24 pp.

122

ANALYSIS OF OILFIELD WATERS

Collins, A.G., Pearson, C., Attaway, D.H. and Watkins, J.W., 1961. Methods of analyzing oilfield waters: iodide, bromide, alkalinity, acidity, borate boron, total boron, organic boron, potassium, calcium, magnesium, iron, fluorides, and arsenic. US. Bur. Min. Rep. Invest., No.5819, 39 pp. Craig, H., 1961a. Isotopic variations in meteoric waters. Science, 133:1702-1703. Craig, H., 1961b. Standards for reporting concentrations of deuterium and oxygen-18 in natural waters. Science, 133:1833-1834. Dean, J.A., 1960. Flame Photometry. McGraw-Hill, New York, N.Y., 354 pp. Diehl, H. and Smith, G.F., 1958. The Copper Reagents: Cuproine, Neocuproine, Bathocuproine. G. Frederick Smith Chemical, Columbus, Ohio, 48 pp. Dunlap, H.F. and Hawthorne, R.R., 1951. The calculation of water resistivities from chemical analyses. J. Pet. Technol., 7:17. Epstein, S . and Mayeda, T., 1953. Variation of "0 content of waters from natural sources. Geochim. Cosmochim. Acta. 4:213-224. Fabricand, B.P., Imbimbo, E.S., Brey, M.E. and Watson, J.A., 1966. Atomic absorption analysis of lithium, magnesium, potassium, rubidium, and strontium in ocean waters. J. Geophys. R e s , 71:3917-3921. Fisher, F.L., Ibert, E.R. and Beckman, H.F., 1958, Inorganic nitrate, nitrite, or nitratenitrite. Anal. Chem., 30:1972-1974. Friedman, I., 1953. Deuterium content of natural waters and other substances. Geochim. Cosmochim Acta, 4:213-224. Friedman, I. and Woodcock, A.H. 1957. Determination of deuterium/hydrogen ratios in Hawaiian waters. Tellus, 9:553-556. Furman, N.H., 1962. Standard Methods o f Chemical Analysis. D. Van Nostrand, Princeton, N.J., 6th ed., 332 pp. Garrels, R.M. and Christ, C.L. 1965. Solutions, Minerals, and Equilibria. Harper and Row, New York, N.Y., 450 pp. Gautier, A., 1903. The arsenic content of some biologic materials. Compt. Rend., Acad. Sci. Fr., 137:232. George, W.O. and Hastings, W.W., 1951. Nitrate in the groundwaters of Texas. A m . Geophys. Union Trans., 32:450-456. Gorgy, S., Rakestraw, N.W. and Cox, D.L., 1948. Arsenic in the sea. J. Mar. Res., 7 :2 2-41; Halliburton Company, 1970. Chemical Research and Development. Halliburton Services, Procedures 110.14 and 110.15, unpublished. Herrmann, R. and Alkemade, C.T.J., 1963. Chemical Analysis by Flame Photometry. Interscience, New York, N.Y., 644 pp. Hodgman, C.D., Weast, R.C., Shankland, R.S. and Selby, S.M., 1962. Handbook of Chemistry and Physics. Chemical Rubber, Cleveland, Ohio, 44th ed., 3604 pp. 1. Jones, P.J., 1944. Properties of water found in reservoirs, 1 1 Oil Gas J., 43( 28):205-209. Krauskopf, K.B., 1956. Dissolution and precipitation of silica a t low temperatures. Geochirn Cosmochirn Acta, 1O:l-26. Latimer, W.M., 1952. Oxidation Potentials. Prentice-Hall, New York, N.Y., 2nd ed., 392 PP. Marsh, G.A., 1951. Portable dissolved oxygen meter for use with oilfield brines. Anal. Chern, 23:1427. Mellon, M.G., 1950. Analytical Absorption Spectroscopy. John Wiley and Sons, New York, N.Y., 618 pp. Mellon, M.G.,. 1956. Quantitative Analyses. Thomas F. Crowell, New York, N.Y., 694 pp. Platte, J.A. and Marcy, V.M., 1959. Photometric determination of zinc with zincon: application to water containing heavy metals. Anal. Chem., 31 :1226-1228. Potter, E.C., 1956. Electrochemistry. MacMillan, New York, N.Y., 418 pp.

REFERENCES

123

Pourbaix, M.J., 1949. Thermodynamics of Dilute Aqueous Solutions. Edward Arnold, London, 136 pp. Rainwater, F.H. and Thatcher, L.L., 1960. Methods for collection and analysis of water samples. U S . Geol. Surv. Water Supply Paper, No.1454, p.70. Rakestraw, N.W. and Lutz, F.B. 1933. Determination of arsenic in sea water. Biol. Bull., 65:397. Ramirez-Munoz, J., 1968. Atomic Absorption Spectroscopy and Analysis by Atomic Absorption Flame Photometry. American Elsevier, New York, N.Y., 315 pp. Robinson, J.W., 1966. Atomic Absorption Spectroscopy. Marcel Dekker, New York, N.Y., 204 pp. Rosin, J., 1955. Reagent Chemicals and Standards. D. Van Nostrand, New York, N.Y., 561 pp. Sandell, E.B., 1959. Colorimetric Determination of Traces o f Metals. Interscience, New York, N.Y., 1032 pp. Schrink, D.R., 1965. Determination of silica in sea water using solvent extraction. Anal. Chem., 37:764-765. Scribner, B.F. and M. Margoshes, 1961. Excitation of solutions in a gas-stabilized arc source. Natl. Bur. Standards Rep., No.7342, 8 pp. Smales, A.A. and Pate, B.D., 1952. The determination of sub-microgram quantities of arsenic by radioactivation, 11. The determination of arsenic in sea water. Analyst, 7 7 :188-195. f Stratton, G. and Whitehead, H.C., 1962. Colorimetric determination o arsenic in water with silver diethyldithiocarbamate. J. A m . Water Works Assoc., 54:861-863. Watkins, J.W., 1954. Analytical methods of testing waters to be injected into subsurface oil-productive strata. U.S. Bur. Min. Rep. Invest., No.5031, 29 pp. Welcher, F.J., 1957. The Analytical Uses o f Ethylenediaminetetraacetic Acid. D. Van Nostrand, Princeton, N.J., 356 pp. White, D.E., Brannock, W.W. and Murata, K.J., 1956. Silica in hot-spring waters. Geochim. Cosmochim. Acta, 10:27-59. Willard, H.H., Merritt, Jr., L.L. and Dean, J.A., 1965. Instrumental Methods of Analysis. D. Van Nostrand Co., Princeton, N.J., 4th ed., 250 pp. Wyllie, M.R.J., 1963. The Fundamentals of Well Log Interpretation. Academic Press, New York, N.Y., 3rd ed., 238 pp. Zobell, C.E., 1946. Studies on redox potential of marine sediments. Bull. A m . Assoc. Pet. Geol., 30:477-513.

This Page Intentionally Left Blank

Chapter 4.

INTERPRETATION OF CHEMICAL ANALYSES OF OILFIELD WATERS

Water analyses may be used to identify the source of a water. In the oilfield one of the prime uses of these analyses is to determine the source of extraneous water in an oil well, so that casing can be set and cemented to prevent such water from flooding the oil or gas horizons. In some wells a leak may develop in the casing or cement, and water analyses are used to identify the water-bearing horizon so that the leaking area can be repaired. With the present emphasis on water pollution prevention, it is very important to locate the source of a polluting brine, so that remedial action can be taken. Comparisons of water-analysis data are tedious and time-consuming; therefore, graphical methods are commonly used for positive, rapid identification. A number of systems have been developed, all of which have some merit. Calculating probable compounds The hypothetical combinations of dissolved constituents found in waters are commonly calculated by combining the positive and negative radicals in the following order: calcium magnesium sodium potassium bicarbonate sulfate chloride nitrate

Calcium is combined with bicarbonate, and if more calcium is available than that consumed by bicarbonate, it is combined with sulfate, chloride, and nitrate until exhausted. Conversely, any excess bicarbonate is combined with magnesium, sodium, and potassium until consumed. Other radicals can and should be added for most petroleum reservoir waters. These include lithium, strontium, barium, iron, borate, phosphate, bromide, and iodide. They can be grouped in the appropriate column and then in the calculations each positive and negative radical is totally combined, the next following radical is combined until both the cations and anions are exhausted. If the analysis is correct, the cations and anions will be present in approximately equivalent amounts. To calculate the hypothetical combinations, the reacting values of the positive and negative radicals or ions are calculated as follows: reacting

126
TABLE 4.1 Reaction coefficients Cation Calcium Magnesium Iron Sodium TABLE 4.11 Reacting values (RV) Cation (mg/l) Ca 4,000x Mg 3,000x Fe 100 x Na 9,400 x RV

INTERPRETATION OF CHEMICAL ANALYSES

Anion

0.0499 0.0823 0.0358 0.0435

bicarbonate sulfate chloride

0.0164 0.0208 0.0282

Anion (mg/l) HC03

RV

0.0499 = 0.0823 = 0.0358 = 0.0435 =

199.6 246.8 3.6 408.9 858.9

so4
C1

500 x 0.0164 = 8.2 200 x 0.0208 = 4.2 30,000 x 0.0282 = 846.3 858.7

TABLE 4.111 Reacting value distribution Ca Ca Ca Mg Fe Na as as as as as as calcium bicarbonate calcium sulfate calcium chloride magnesium chloride iron chloride sodium chloride

82 . 4.2 187.2 246.8 36 . 408.9 858.9

values (RV) or equivalents per million (epm) = mg/l of ion x valence of ion/ molecular weight of ion. The term valence of ion/molecular weight of ion is called reaction coefficient and the positive and negative ions have values as shown in Table 41 .. Table 41 indicates how the results of a water analysis are converted t o .1 reacting values. The reacting values are a measure of the cations and anions dissolved in the water. The 4,000 mg/l of calcium with a reacting value of 199.6 can combine with all the bicarbonate, all the sulfate, and 187.2 epm of the ., chloride. Magnesium will combine with 246.8,iron with 3 6 and sodium with 408.9 epm of chloride. Thus the reacting values can be considered to be distributed as shown in Table 4 1 1 .1.

DETERMINING A SOUGHT COMPOUND TABLE 4.IV Combination factors Reaction values given Ca o r C 0 3 Ca or SO4 Ca or C1 Mg or C 0 3 Mg or SO4 Mg or C1 Fe or C03 Fe o r S 0 4 Fe orC1 Na or C03 Na or SO4 Na or C1 Compound sought CaC03 CaS04 CaClz MgCO3 MgS04 MgClz FeC03 FeS04 Fa12 Na~C03 Naz SO4 NaCl Combination factor 50.1 68.1 55.5 42.2 60.1 47.6 57.8 76.0 63.4 53.1 71.0 58.4

127

TABLE 4.V Hypothetical combinations Ca(HC03 ) to CaC03 CaS04 CaC12 MgCh FeClz NaCl *In mg/l. 8.2 x 50.1 4 . 2 68.1 ~ 187.2 x 55.5 246.8 x 47.6 3.6 x 63.4 858.9 x 58.4
= 411 CaC03* = 286CaS04 = 10,390 CaC12 = 11,748 MgClz
= =

228 FeCIz 50,160 NaCl

Determining a sought compound It is necessary to multiply the reacting value by a combination factor to determine a hypothetical compound. This factor is necessary to convert the reported radical into the desired compound. For example, the factor for converting Ca to CaCO, is 2.50 and the reaction coefficient for Ca is 0.0499. Therefore, the combination factor to convert the reacting value for Ca to CaCO, is 2.50 + 0.0499 = 50.1. Table 4.IV illustrates some combination factors. The combination factors given in Table 4.IV can be used to calculate the hypothetical combinations shown in Table 4.V, using the analysis shown in Table 4.111.

128

INTERPRETATION OF CHEMICAL ANALYSES

Graphic plots
Graphic plots of the reacting values can be made to illustrate the relative amount of each radical present. The graphical presentation is an aid t o rapid identification of a water, and classification as t o its type, and there are several methods that have been developed.

Tickell diagram
The Tickell (1921) diagram was developed using a 6-axis system or star diagram. Percentage reaction values of the ions are plotted on the axes. The percentage values are calculated by summing the epms of all the ions, dividing the epm of a given ion by the sum of the total epms, and multiplying by 100.
Na Ca+Mg Na Ca+Mg

2-

ci
(a) CI RV=49.92%
(b)

So4

h 9 2 ma / I i tar

Fig. 4.1. Tickell (a) and modified Tickell (b) diagram for Gulf Coast water, sample No.1.

Na

Ca+Mg

Ca + Mg

(a)

CI ~ v = 4 9 . 2 9 %

(b)

1i07 ma/ lltar

Fig. 4.2. Tickell (a) and modified Tickell (b) diagram for Anadarko Basin water, sample No. 2.

GRAPHIC PLOTS

129

(a)

so4 C I RV=49.92 %

(b)

so4 5,708 m e / l i t e r

Fig. 4.3. Tickell (a) and modified Tickell (b) diagram for Williston Basin water, sample N0.3.

No

Ca+Ma

Co+Mg

c i\
(b)

so4 1.769 me / liter

Fig. 4.4. Tickell (a) and modified Tickell (b) diagram for Gulf Coast and Anadarko Basin waters, mixed 1:l.

Na

Co+Mg

Na

Ca+Mg

CI

so4

(b) C I

7 $.
2870 me / i i t r r

Fig. 4.6. Tickell (a) and modified Tickell (b) diagram for Gulf Coast, Williston, and Anadarko Basin waters, mixed 1 :1 :1.

130

INTERPRETATION OF CHEMICAL ANALYSES

Fig. 4.1.illustrates the Tickell diagram using reaction values in percentage in the diagram on the left, and total reaction values in the diagram on the right. The plots of total reaction values, rather than of percentage reaction values, are often more useful in water identification because the percentage values do not take into account the actual 'ion concentrations. Water differing only in concentrations of dissolved constituents cannot be distinguished. To illustrate differences in patterns for different waters, Fig. 4.1-5 were . prepared using the Tickell method. Fig. 4 1 represents a water from the Gulf Coast Basin, taken from the Wilcox formation of Eocene age. Fig. 4.2 is of a sample from the Mer?.mec formation of Mississipian age in the Anadarko Basin. Fig. 4.3 is of sample from a Devonian age formation in the Williston Basin. Fig. 4.4 represents a 1:l mixture of waters of the Gulf Coast and Anadarko Basins, and Fig. 4.5 is a 1:1:1 mixture of all three waters.

REISTLE SYSTEM

Fig. 4.6. Water-analysis interpretation, Reistle system the samples of Fig. 4.1-3.

- sample numbers correspond to

GRAPHIC PLOTS

131

Reistle diagram
Reistle (1927) devised a method of plotting water analyses using the ion concentrations as shown in Fig. 4.6. The data are plotted on a vertical diagram, with the cations plotted above the central zero line and the anions below. This type of diagram often is useful in making regional correlations or studying lateral variations in the water of a single formation, because several analyses can be plotted on a large sheet of paper.

St iff diagra m
Stiff (1951) plotted the reaction values of the ions on a system of rectangular coordinates as illustrated in Fig. 4.7. The cations are plotted to the left and the anions to the right of a vertical zero line. The end points then are connected by straight lines to form a closed diagram, sometimes called a butterfly diagram. To emphasize a constituent that may be a key t o interpretation, the scales may be varied by changing the denominator of the

Fig. 4.7. Water-analysis interpretation, Stiff method - sample numbers correspond to the samples of Fig. 4.1-3.

132

INTERPRETATION OF CHEMICAL ANALYSES

ion fraction usually in multiples of 10. However, when looking at a group of waters all must be plotted on the same scale. Many investigators believe that this is the best method of comparing oilfield water analyses. The method is simple, and nontechnical personnel can be easily trained t o construct the diagrams.

Other methods
Several other water identification diagrams have been developed, primarily for use with fresh waters, and they will not be discussed here. The Piper (1953) diagram and the Stiff (1951) diagram were adapted to automatic data processing by Morgan et al. (1966),and Morgan and McNellis (1969).The Piper (1953)diagram uses a multiple trilinear plot t o depict the water analysis, and this quaternary diagram shows the chemical composition of the water in terms of cations and anions. Angino and Morgan (1966) applied the automated Stiff and Piper diagrams to some oilfield brines and obtained good results.

References
Angino, E.E. and Morgan, C.O., 1966. Application of pattern analysis t o the classification of oilfield brines. Kans. State Geol. Sum.,Comput. Contrib., No.7, pp.53-56. Morgan, C.O. and McNellis, J.M., 1969. Stiff diagrams of water-quality data programmed for the digital computer. Kuns. State Geol. Sum., Spec. Distrib. Publ., No.43, 27 pp. Morgan, C.O., Dingman, R.J. and McNellis, J.M., 1966. Digital computer methods for water-quality data. Ground Water, 4:35-42. Piper, A.M., 1953. A graphic procedure in the geochemical interpretation of water analyses. US.Geol. Surv. Ground Water Note, No.12, 1 4 pp. Reistle, C.E., 1927. Identification of oilfield waters by chemical analysis. U.S.Bur. Min. Tech. Paper, No.404, 25 pp. Stiff, H.A., 1951. The interpretation of chemical water analysis by means of patterns. J. Pet. Technol., 3:15-17. Tickell, F.G., 1921. A method for graphical interpretation of water analysis. Calif. State Oil Gas Superv., 6:5-11.

Chapter 5.

SIGNIFICANCE OF SOME INORGANIC CONSTITUENTS AND PHYSICAL PROPERTIES OF OILFIELD WATERS

In general, the concentrations of the constituents in various natural solids of reservoir rocks must be considered along with the amounts that are found in associated oilfield waters. Some possible chemical reactions between host rock and reservoir water may deplete or enrich the concentration of the constituents in oilfield waters. Another important factor is the solubility of a constituent. The ionic potential, determined by dividing the ionic radius by the valence, influences the solubility of elements. For example, elements with low ionic potential are more likely t o remain in true ionic solution. Elements commonly found in oilfield waters have the following ionic potentials: sodium, 0.95; calcium, 0.50; magnesium, 0.33; chlorine, 1.81; bromine, 1.95; and iodine, 2.16. Apparently the cation (magnesium) and the anion (chlorine) would be the most likely to remain in true ionic solution; however, several other variables occur during diagenesis which lead to depletion or enrichment of constituents in waters.

Lithium
Lithium is the lightest alkali metal; it has a distinctly smaller radius, 0.60 8,than the other alkalies and is the smallest of all singly charged cations. It is one of the less abundant elements, and its abundance in the earths crust is about 6.5 x wt.% (Fleischer, 1962). Here again, it is an exception because in general, the lighter elements tend to be more abundant than the heavier elements. It is lithophilic in that it tends t o be associated with the silicate phase in rocks (Ahrens, 1965); however, because of its small size, it supposedly cannot replace the abundant alkali metals in mica. It and the other alkali metals exist in a uniform positive one state of oxidation and are inherently ionic. Their chemical behavior depends almost entirely upon electron loss, and their chemistry is simpler than that of any of the other metallic elements (Moeller, 1954). Lithium is potentially toxic to plants (Hem, 1970), yet it is regularly found in plant ashes, which indicates that it normally is present in soil waters (Goldschmidt, 1958). Coal ashes of Neurode, Silesia, contained up to 198 ppm lithium, whereas soils in northeast Scotland contain 30-5,000 ppm. The content of lithium in sediments ranges up to 6 ppm in quartzites and sandstones, up to 15 ppm in calcareous rocks, and up t o 120 ppm in clays and shales.

134
TABLE 5.1

INORGANIC CONSTITUENTS AND PHYSICAL PROPERTIES

Properties of the alkali metals Property Atomic number Nonhydrated radius (A) Hydrated radius (A) Outer electronic configuration Atomic weight Ionization potential (V) Lithium Sodium Potassium Rubidium Cesium

3 0.60 3.82

11

19 1.33 3.31 3' 3p6 4 ' s s 39.102 4.339

37 1.48
-

55

0.95 3.58 2s22p6 3 ' s 22.990 5.138

1.69

5 5p6 6 ' s ' s 132.905 3.893

1s' 2 ' s 6.939


5.390

4s24p6 5 ' s 85.47 4.176

TABLE 5.11 Five relative concentration changes of some dissolved ions during evaporation of sea water and brine* Constituents Concentrations (mg/l) Sea water Lithium Sodium Potassium Rubidium Magnesium Calcium Strontium Boron Chloride Bromide Iodide CaSO4 NaCl MgS04

KCI

MgC12

0.2 2 11,000 98,000 350 3,600 01 . 1 1,300 13,000 400 1,700 60 7 5 40 19,000 178,000 65 600 0.05 2

11 12 140,000 70,000 23,000 37,000 6 8 74,000 80,000 100 10 1 10 300 310 275,000 277,000 4,000 4,300 5 7

27 13,000 26,000 14 130,000 0 0 750 360,000 8,600 8

34 12,000 1,200 10 153,000 0 0 850 425,000 10,000 8

*Approximate mg/l. Columns headed sea water, CaS04, etc., represent stages in sea water evaporation. For example, sea water contains 0.2 mg/l of lithium, after calcium sulfate has precipitated the residual brine contains about 2 mg/l of lithium, after sodium chloride has precipitated the residual brine contains about 11 mg/l of lithium, the residual brine contains about 12 mg/l of lithium after magnesium sulfate precipitates, 27 mg/l of lithium after potassium chloride precipitates, and 34 mg/l of lithium after magnesium chloride precipitates.

LITHIUM

135

The hydrated radius of lithium is 3.82 as shown in Table 5.1 (Moeller, 1954). The ionic potential is 0.60, and the polarization is 1.67. The polarization is quite high and is a measure of its replacing power in an exchange system. Apparently it can replace strontium, calcium, and magnesium since their polarizations are 1.77, 2.02, and 3.08, respectively. Some surface waters of the volcanic sodium chloride type are enriched in lithium (White, 1957). Lithium from Searles Lake brine is recovered as Li2NaP04 (Brasted, 1957). The content of lithium in oilfield waters is usually less than 10 mg/l but in some Smackover formation waters from east Texas, concentrations up t o 500 mg/l are present. When a brine containing lithium goes through an evaporite sequence, lithium is one of the elements whose concentration does not decrease, as illustrated in Table 5.11, in the liquid phase as various minerals precipitate (Collins, 1970). Fig. 5.1 illustrates the enrichment of lithium as compared t o an evaporite sequence in some subsurface brines from Tertiary, Cretaceous, and Jurassic age sediments. Fig. 5.2 illustrates a similar enrichment for some brines taken from Pennsylvanian and Mississippian age sediments (Collins, 1969a). Possibly lithium was liberated and potassium was depleted by exchange reactions with clay minerals, degradation of lithium containing minerals, or simply a leaching of minerals, primarily silicates, which contain lithium. Lithium substitutes in the structure of several common minerals and forms few minerals of its own. If the minerals in which it has substituted should degrade or break down with depth, the lithium might be resolubilized, thus increasing its concentration in the aqueous phase. White et al. (1963) postulated that because the lithium concentration in magmatic waters is related to volcanic

a,

LITHIUM, mgll

Fig. 5.1. Comparison of the lithium concentrations in some Tertiary (T),Cretaceous (C), and Jurassic (J) age formation waters from Louisiana with an evaporating sea water.

136

INORGANIC CONSTITUENTS AND PHYSICAL PROPERTIES

u
1 0

20

LITHIUM, mg/l Fig. 5.2. Comparison of the lithium concentrations in some Mississippian (M) and Pennsylvanian (P) age formation waters from Oklahoma with an evaporating sea water.

emanations, the increase in the lithium content of deeper waters might be related to the same cause.

Sodium
The most abundant member of the alkali-metal group is sodium, ranking number 6 with respect t o all the metallic elements. The radius of the sodium ion is 0.95 A, and its geochemistry is controlled to some extent by calcium because of the similarity of their ionic radii. Its abundance in the earth's . crust is about 2.8 wt.% (Fleischer, 1962). Table 5 1 shows that its outer electronic configuration is 2s' 2p6 3s' , with a first ionization potential of 5.138 V, indicating that its single outer electron is less firmly held than in the lithium atom with a first ionization potential of 5.390 V. The ionization potential is a measure of the chemical reactivity - the lower the potential, the greater the reactivity. Table 5.1 (Moeller, 1954) also illustrates some qf its other properties. According t o Ahrens (1965),sodium is lithophilic, and many distinctly lithophile elements have valence electrons outside a closed shell of eight electrons. The ionic radius decreases as the charge on the cation increases. Sodium does readily participate in solid solution relationships because its radius is small, making replacement of cations with 30% larger radii difficult. The amounts of sodium in argillaceous sediments and marine shales are about 1,000ppm and 1,300ppm, respectively (Goldschmidt, 1958).

SODIUM

137

Sodium in solution tends to stay in solution; it does not readily precipitate with an anion, and it is less easily adsorbed by clay minerals than are cesium, rubidium, potassium, lithium, barium, and magnesium. The major source of sodium in sea water can be attributed t o the weathering of rocks. Some sodium probably was derived through volcanic activity. The ocean and evaporite sediments contain the bulk of the sodium. Igneous rocks contain appreciably more sodium than sedimentary rocks with the exception of evaporites. Sea water contains about 11,000 mg/l of sodium, as illustrated in Table 5.11. The concentration of sodium increases in brine as it evaporates, t o about 140,000 mg/l, when halite precipitates. Most oilfield waters contain more sodium than any other cation, and most oilfield waters are believed to be of marine origin. Fig.5.3 is a log-log plot of the chloride concentration versus sodium of some subsurface brines taken from sediments of Tertiary, Cretaceous, and Jurassic age. The straight line is a plot of chloride versus sodium concentrations for some evaporite waters, and indicates the enrichment of sodium ions until halite (NaC1) precipitates - at a chloride concentration of about 140,000 mg/l (compared t o that of normal sea water, 19,000 mg/l). The plot of the concentrations of sodium versus chloride for these subsurface brines falls very near the normal evaporite curve, indicating that the concentration mechanism may be related to an evaporite process (Collins, 1970). Fig. 5.4 is a similar plot for some subsurface brines taken from sediments of Pennsylvanian and Mississippian age (Collins, 1969a). Several of these samples are somewhat depleted in sodium which indicates that

SODIUM,

g/l

Fig. 5.3. Sodium versus chloride concentrations for some formation waters taken from Tertiary (T), Cretaceous (C), and Jurassic (J) zge sediments and compared to evaporating sea water.

138

INORGANIC CONSTITUENTS AND PHYSICAL PROPERTIES

500F - / 200t0

- Nor ma1 evaoor it e


associated b i n e

I-

/"

SODIUM, mg/l
Fig. 5.4. Sodium versus chloride concentrations for some formation waters taken from Pennsylvanian (P) and Mississippian (M) age formation sediments and compared t o evaporating sea water.

diagenetic processes, such as ion-exchange or ultra-filtration reactions involving clays and/or carbonates, may operate to deplete the sodium concentration in waters in older sediments. Potassium The second most abundant member of the alkali-metal group is potassium; its abundance in the crust of the earth is about 2.55 wt.% (Fleischer, 1962). Like the other alkali metals, it is lithophilic, and with its large ionic radius (see Table 5.1) it participates in forming solid solutions and forms its own minerals, such as feldspar and mica. The potassium feldspars are resistant to leaching by water, which may account for the low potassium concentrations in many natural waters. Clay minerals readily adsorb potassium, and in illite it is incorporated into the crystal structure in such a manner that it cannot be removed by ion-exchange reactions (Lyon and Buckman, 1960). Potassium is less easily hydrated than sodium, and is more easily adsorbed by colloids; therefore, it is retained in sediments and soils in greater abundance than sodium. It is an essential element t o plants and animals. According to Gol&chmidt (1958),potassium in pulverized potassium feldspars is absolutely unavailable t o plants. The concentrations of potassium in carbonates, sandstones, and shales is about 2,700, 10,700,and 26,600 ppm, respectively (Mason, 1966). Potas-

POTASSIUM
.lvv
~~

139

200

- 100 -

1
POTASSIUM,
g/ I

I I I IIll

Fig. 5.5. Potassium versus chloride concentrations for some formation waters taken from Tertiary (T), Cretaceous (C), and Jurassic (J) age sediments and compared to evaporating sea water.

sium concentrates primarily in hydrolysates (clay minerals), such as illite and glauconite, and in evaporites. Table 5.11 illustrates how the concentration of potassium in the aqueous phase increases until sylvite (KC1) precipitates. The concentration of potassium in some subsurface brines usually is depleted with respect to an evaporite-associated sea water. Fig.5.5 illustrates the relation of potassium in some subsurface brines taken from sediments of Terti500 -

, '

-Nmal

evaporite curve

5-

m
POTASSIUM, mg/l Fig. 5.6. Comparison of the potassium concentrations in some Pennsylvanian (P) and Mississippian (M) age formation waters from Oklahoma with an evaporating sea water.

140

INORGANIC CONSTITUENTS AND PHYSICAL PROPERTIES

ary, Cretaceous, and Jurassic ages to an evaporite-associated sea water (Collins, 1970). Fig.5.6 illustrates the same relation for some subsurface brines taken from Pennsylvanian and Mississippian age sediments (Collins, 1969a). The depletion of potassium in subsurface brines might be caused by its uptake by clays. For example, montmorillonite-type clay minerals systematically change to illite with increasing depth of burial, due to thermal diagenesis; and, as a result of this transformation, they lose interlayer (bound) water (Burst, 1969). This change appears t o begin at a temperature above 90C. (This freed interlayer water can be readily expelled, and its movement probably is important in the first migration stage of hydrocarbons.) Laboratory experiments at elevated temperatures and pressures indicate that montmorillonite loses its interlayer water and transforms into illite in the presence of potassium-enriched water (Khitarov and Pugin, 1966). The structural variations of the expandable minerals in clays apparently are also influenced by the potassium content of the associated waters.

Rubidium
Rubidium, like the other alkali metals is lithophilic, and its abundance in the earths crust is about 3.0 x wt.%, which is greater than that of lithium (Fleischer, 1962). It tends to be removed from solution more readily than lithium, primarily because of its ability to replace potassium in mineral structures. Table 5.11 indicates that it precipitates from an evaporite along with sylvite to a greater extent than lithium, and it has a high chemical reactivity. The radius of its ion, 1.48 is only about 10% larger than the potassium ion, so it can be accommodated into the same crystal lattices. Because of this, it forms no minerals of its own. Rubidium and cesium occur sympathetically in nature; that is, both are commonly found in amazonite, vorobyevite, and beryl (Goldschmidt, 1958). Rubidium is a member of series NH4-K-Rb-Cs, and members of this series are more similar in their chemical and physical properties than are the members of any other group, with the exception of the halogens. Rubidium concentrates in the late crystallates, particularly those of granitic derivation, and it has a greater tendency t o be adsorbed by clays than has potassium. It is removed from igneous rocks by water leaching and then adsorbed by hydrolysate sediments and soils. Shales contain about 250 ppm of rubidium; deep-sea red clays, about 400 ppm; and some glauconites, about 500 ppm (Goldschmidt, 1958).Sea water contains about 0.12 mg/l of rubidium; subsurface brines contain up t o 4 mg/l. Higher concentrations of rubidium probably can be found in brines associated with rocks containing potassium minerals, such as microcline feldspars, or lepidolite mica.

a,

CESIUM

141

Cesium Cesium is the heaviest alkali metal and also the rarest, with an abundance of about 7 x wt.% in the earths crust (Fleischer, 1962). It has an ionic radius of 1.69 8,which is distinctly larger than potassium, and it cannot replace potassium in minerals as easily as rubidium; probably because of this, it forms its own minerals. It is leached from igneous and metamorphic rocks by water during weathering, and is adsorbed by hydrolysate sediments and soils more readily than rubidium or potassium. Its low ionization potential indicates that it has the greatest chemical reactivity of the alkali metals. Cesium and rubidium were discovered in 1860 by Robert Bunsen by use of spectral analysis, a method which he and Kirchhoff invented. Cesium concentrates primarily like rubidium, in marine argillaceous sediments. Some shales contain about 15 ppm; deep-sea red clays, 20 ppm; and glauconite, 15 ppm of cesium (Goldschmidt, 1958). Sea water contains 5x mg/l of cesium, and some subsurface brines contain up to 1mg/l. Beryllium Beryllium is a member of the alkaline earth group in the periodic chart of the elements, but few of its properties are similar t o the more abundant members, such as magnesium, calcium, and strontium. Beryllium, like lithium, is a light element with an atomic weight of 9.012 (Table 5.111; see also Moeller, 1954), and like lithium, it is an exception t o the rule that light elements are more abundant than heavy elements. The earths crust contains about 6 x wt.% of beryllium (Fleischer, 1962). In sedimentary rocks, beryllium is restricted primarily to hydrolysates and especially to bauxites enriched in aluminum (Goldschmidt, 1958). Shales contain about 6 ppm, and some coal ashes contain up to 8,000 ppm, although generally only about 4 ppm. The concentration of beryllium in sea

TABLE 5.111 Properties of the alkaline earth metals Property Beryllium Magnesium 12 0.65 2s 2p6 3s 24.31 7.644 Calcium 20 0.99 3s2 3p6 4s2 40.08 6.111 Strontium 38 1.13 4s 4p6 5s2 87.62 5.692 Barium 56 1.35 5s2 5p66s 137.34 5.210

Atomic number 4 Ionic radius (A 1 0.31 Outer electronic configuration 1s 2s Atomic weight 9.012 Ionization 9.320 potential (V)

142

INORGANIC CONSTITUENTS AND PHYSICAL PROPERTIES

water is about 5 x lo-' mg/l, and some subsurface brines contain 0.02-4.2 mg/l. Since beryllium is highly toxic, waters containing it should be handled with caution.

Magnesium
One of the more abundant members of the alkaline earth group of metals, magnesium makes up about 2.1 wt.% (Fleischer, 1962) of the crust of the earth. Magnesium is dissolved during chemical weathering, mainly as the chloride and sulfate. Ferromagnesian minerals in igneous rocks and magnesium carbonate in carbonate rocks are generally considered t o be the principal sources of magnesium in natural waters. Carbon dioxide plays an important role in the dissolution of magnesium from silicate and carbonate minerals. Waters associated with either granite or siliceous sand may contain less than 5 mg/l of magnesium, whereas those associated with either dolomite or limestone may contain over 2,000 mg/l of magnesium. Elements commonly found in oilfield waters have the following ionic potentials: sodium, 0.95; calcium, 0.50; magnesium, 0.33; chlorine, 1.81; bromine, 1.95; and iodine, 2.16. Apparently the cation (magnesium) and the anion (chlorine) would be the most likely to remain in true ionic solution; however, several other variables occur during diagenesis which lead to depletion of magnesium in waters. Depletion of magnesium in some waters probably is a result of the replacement reaction t o form dolomite, CaMg(C0, ) 2 . Whole mountain masses are made of dolomite, which is formed by the regular substitution in the calcite

2oo

t
500

C
J

?$

/
2,000

Normal evaporite curve

'so0

rpoo

5ooO

lop00

2m 0 ,

5Q(

MAGNESIUM, mg I I

F g 5.7. Comparison of the magnesium concentrations in some Tertiary (T), Cretaceous i. (C), and Jurassic (J) age formation waters from Louisiana with an evaporating sea water.

CALCIUM

143

c
Normal evaporite curve

500

M M P

r
20
1 0

1,000

I 0,000 lO0,OoO MAGNESIUM, mg/l

Fig. 5.8. Comparison of the magnesium concentrations of some Pennsylvanian (P) and Mississippian (M) age formation waters from Oklahoma with an evaporating sea water.

crystal lattice of alternate ions of calcium and magnesium. The large differences in the ionic radii of Ca (0.99A) and Mg (0.65A) are the reason for this diadochy. Magnesium ions in aqueous solution have a large attraction for water molecules and probably are surrounded by six water molecules in octahedral arrangement. This may account for the paucity of magnesium in soils, because the small cation becomes large by hydration. Sodium has a similar reaction, but potassium, which does not, is readily adsorbed by soil colloids. Shales, sandstones, and carbonates contain 15,000,7,000,and 47,000 ppm of magnesium, respectively (Mason, 1966). Subsurface brines contain from less than 100 mg/l t o more than 30,000 mg/l; however, many subsurface brines are depleted in magnesium if compared to a sea water evaporite .1. sequence, (Table 5 1 ) Sea water contains about 1,300 mg/l. Fig. 5.7 is a plot of chloride versus magnesium for some subsurface brines taken from Tertiary, Cretaceous, and Jurassic age sediments. The position of the normal evaporite curve indicates that all of these waters were depleted in magnesium with respect to this curve (Collins, 1970). Fig. 5 8 is a plot showing similar . depletion of some subsurface brines taken from some sediments of Pennsylvanian and Mississippian age.

Calcium
The abundance of calcium in the crust of the earth is about 3.55 wt.% (Fleischer, 1962),making it the most abundant of the alkaline earth metals,

144

INORGANIC CONSTITUENTS AND PHYSICAL PROPERTIES

but only in the crust; in the earth as a whole, magnesium is much more abundant. Calcium is dissolved as bicarbonate as a result of chemical weathering of calcium-bearing minerals. Waters associated with limestone, dolomite, gypsum, or gypsiferous shale usually contain an abundance of calcium, but waters associated with granite or silicious sand may contain less than 10 mg/l of calcium. Slight changes in the pH of waters containing calcium bicarbonate will cause calcium carbonate to precipitate, and calcium carbonate is one of the most common deposits found in plugged oilfield lines, equipment, and reservoirs. Precipitation of calcium carbonate in the sea is the prime mode of the origin of limestone. The solubility of calcium carbonate in sea water increases with salinity and increasing partial pressure of carbon dioxide, but it decreases with increasing pH, calcium content, and temperature. The solubility of calcium sulfate decreases with increasing temperature. Shales, sandstones, and carbonate rocks contain about 22,100, 39,100, and 302,300ppm of calcium, respectively (Mason, 1966).Sea water contains 400 mg/l and subsurface brines often contain 2,000-3,000 mg/l, with some . as high as 30,000 mg/l. Fig. 5 9 is a plot of chloride versus calcium concentrations for some subsurface waters taken from Tertiary, Cretaceous, and Jurassic age sediments. The amount of calcium in these waters increases with increasing salinity, and the waters from the older sediments appear to contain more calcium. Fig. 5.10 is a similar plot for some subsurface brines taken from sediments of Pennsylvanian and Mississippian age. These samples all appear to be enriched in calcium relative t o the evaporite curve, and the concentration of calcium appears to increase with increasing salinity.

200

Normal evaporite curve

\ 0

100-

1
500
1 , m

I 1 I I I111
p

5poo

lop00

29ooo

CALCIUM, mg/l

Fig. 5.9. Comparison of the calcium concentrations of some Tertiary (T), Cretaceous (C), and Jurassic (J) age formation waters from Louisiana with an evaporating sea water.

STRONTIUM

145

&\ --

Normal evaporite curve

Ii
201

M M
1

00
CALCIUM, mg/ I Fig. 5.10. Comparison of the calcium concentrations of some Pennsylvanian (P) and MEsissippian (M) age formation waters from Oklahoma with an evaporating sea water.

Strontium Strontium, a minor element compared t o calcium and magnesium comprises about 0.03 wt.% of the earth's crust (Fleischer, 1962). Table 5.111 illustrates some of its properties, and it resembles calcium chemically. Strontium has a tendency to work upward during fractional crystallizaticn because of its relatively large radius (Goldschmidt, 1958).It occurs abundantly with potassium in volcanic rocks, alkali rocks, and pegmatites. Dissolved strontium results from water leaching of rocks, and it has been postulated that the strontium in petroleum-associated waters also may be a byproduct of the organic decay processes which originally formed petroleum. Strontium is only a microconstituent in most terrestrial animals, but several species of marine animals contain considerable quantities of strontium in their skeletons (Odum, 1951). Table 5 1 indicates that strontium may reach a concentration of 60 mg/l .1 during sea-water evaporation, and then most of it precipitates with calcium sulfate. The amount of sulfate in the water influences the amount of strontium that remains in solution. Data by Sillhn and Martell (1964)indicate that if the sulfate activity in a water is 100 mg/l, the strontium activity can be about 28 mg/l. Davis and Collins (1971)studied the solubility of strontium sulfate in strong electrolyte solutions and found that 958 mg/l of strontium is soluble in a synthetic brine solution of ionic strength 3.05,

146

INORGANIC CONSTITUENTS AND PHYSICAL PROPERTIES

containing ions of sodium, calcium, magnesium, potassium, chloride, bromide, and iodide. Calcium chloride concentration apparently has a very pronounced effect upon the solubility of strontium sulfate. Celestite and strontianite occur commonly in sediments. Carbonate sediments contain up t o 1,200 ppm of strontium; dolomites, usually less than

I
C -Cretaceous J -Jurassic C

I
J

c cc 2,000
C cC

T
1 0

20

50

I IIII

10 0

2 a

I IIL

STRONTIUM, mgll

Fig. 5.11. Comparison of the strontium concentrations of some Tertiary (T), Cretaceous (C), and Jurassic (J) age formation waters from Louisiana with an evaporating sea water.
500

200 -

P
I

Ill
50
10 0 ZOO STRONTIUM, mg/l

500

1,000

Fig. 5.12. Comparison of the strontium concentrations of some Pennsylvanian (P) and Mississippian (M) age formation waters from Oklahoma with an evaporating sea water.

BARIUM

147

170 ppm; and secondary gypsum, up t o 1,100 ppm (Goldschmidt, 1958). Sea water contains about 8 mg/l of strontium, but subsurface brines contain up to 3,500 mg/l. Fig. 5.11 is a plot of chloride versus strontium content for some subsurface brines taken from some Tertiary, Cretaceous, and Jurassic age sediments. Most of these samples were enriched in strontium compared to the evaporite-associated water, and it is possible that a mechanism similar to dolomitization could cause the enrichment. In comparison t o calcium, the strontium appears to be increasingly accumulated; for example, only five samples (from Tertiary sediments) fell within the normal evaporite curve. Fig. 5.12 is a similar plot for some subsurface brines showing similar results taken from sediments of Mississippian and Pennsylvanian age.
Barium
Barium, like strontium, is a minor element, comprising 0.04 wt.%, of the earths crust; it is more concentrated in igneous rocks and less concentrated in sedimentary rocks (Fleischer, 1962). It, like the other alkaline earth metals, is predominantly lithophile. Table 5 1 1 illustrates some of the .1 properties of barium; its ionic radius, 1.35 A, permits it t o replace potassium, but usually not calcium and even less commonly magnesium. Barium forms more of its own minerals than does strontium. Barium is readily absorbed by colloids, like potassium, and is therefore retained by soils or precipitated with hydrolysates; it is also concentrated in deep-sea manganese nodules (Hem, 1970). Barium dissolves as bicarbonate, chroride, or sulfate during weathering processes, and migrates in aqueous solutions as these compounds. The solubility of barium sulfate increases when hydrochloric acid or chlorides of the alkali or other alkaline earth metals are present in solution. The properties of barium are similar t o those of strontium. Both precipitate through loss of carbon dioxide from a bicarbonate-bearing solution, or as sulfates by the action of sulfuric acid, sulfides, or sulfates. Strontium, however, is less likely t o be absorbed by clays than barium, because its ionic radius is smaller and its ionic potential is higher. Encrustation deposits taken from plugged pipes of waterflood systems for secondary recovery of oil, where barium is present, usually contain barium, calcium, strontium, iron, and traces of other metals. Barium may cause problems in waterflood systems by reacting with the chromate-type oxygencorrosion inhibitors, forming water-insoluble barium chromate. The amount of barium found in sandstones, shales, and carbonates is about 180, 450, and 90 ppm, respectively (Goldschmidt, 1958).Sea water contains about 0.03 mg/l, and subsurface brines may contain more than 100 mg/l; however, many brines contain less than 10 mg/l. Davis and Collins (1971)found that 59 mg/l of barium sulfate is soluble in a synthetic brine with an innic strength of 3.0487, containing sodium, calcium, magnesium,

TABLE 5.IV Properties of aluminum. copper. iron, lead, manganese, and zinc
property

Aluminum 13 0.50 2s22p63s23p1 26.98 5.984

Copper 27 0.96(+1) 0.691+21 . , 3s23p63d'04s' 63.54 1.723

Iron 26 0.76(+2) 0.64(+3) . . 3s23p63d64s' 55.54 1.165

Lead 82 1.20(+2) 0.84(+4) . . 4d'05s'5p64f'5d106s'6p' 207.19 7.415

Manganese 25 0.80(+2) 0.46c+7 1.. 3s' 3p6 3d54s' 54.938 1.168

Zinc 30 0.74 3s2'3p63d" 4s' 65.37 9.391

Atomic number Ionic radius (A) Outer electronic configuration Atomic weight Ionization potential (V)

MANGANESE

149

potassium, chloride, bromide, and iodide ions. Many analyses performed by wet chemical methods indicate rather high concentrations of barium in some subsurface brines. Some of these high results probably should be attributed to strontium plus barium rather than barium only, because satisfactory separation of the two in wet chemical methods is very difficult to accomplish. Manganese Manganese is a member of the VII B group of elements and is well known for its multiplicity of oxidation states. Essentially it is cationic, and the Mn+4 oxidation state usually is found in sediments. Its (+2) ionic radius is 0.80 8,while the ferric iron radius is 0.76 8 (see Table 5.IV); reasonable amounts of interchange in crystal lattices between these two ions are possible. The abundance of manganese is about 0.1 wt.% of the earths crust (Fleischer, 1962). Manganese is present in many oilfield brines because it is readily dissolved by waters containing carbon dioxide and sulfate. Except for titanium, manganese is the most abundant trace element in igneous rocks. Nearly all mineral groups of petrological importance contain manganese. During weathering, manganese is dissolved mainly as the bicarbonate. Decomposition of the bicarbonate leads to the formation of M d 4 compounds. In a reducing type of environment Mn+ compounds migrate in aqueous solutions. Mn+ compounds are less mobile, and Mn+4 compounds precipitate from aqueous solutions. In general, manganese remains in solution at a low redox potential and precipitates at a high redox potential. According to Goldberg (1963), manganese oxide nodules on the ocean bottom occur in both shallow water and deep water environments. He attributes these deposits to slow oxidation of dissolved manganese in areas where the waters contact an oxide surface. In most subsurface brines, the manganese is in the reduced form (Mn+*)because the redox potential is low and the pH is less than 7.0. Any in subsurface brines probably would be suspended with particulate matter or complexed by organic compounds, rather than in ionic solution. Shales and carbonates contain about 850 ppm and 1,100 ppm, respectively, of manganese (Mason, 1966). Sea water contains about 0.002 mg/l, and many subsurface brines contain 1.0 to 6.0 mg/l of manganese. Iron Iron is a member of the VIII group of elements and is predominantly siderophile. However, because it has an affinity for sulfur, it is also thiophile; and because it commonly enters into silicate minerals, it is lithophile as well. It is an ubiquitous element, with an abundance of about 5 wt.% of the earths crust (Fleischer, 1962).

150

INORGANIC CONSTITUENTS A N D PHYSICAL PROPERTIES

Iron, cobalt, and nickel possess atomic radii that differ only about 2% or less, so that the crystal chemistry of the three are related. The divalent ions of nickel, magnesium, cobalt, and iron have similar ionic radii; consequently, their chemistries in the sequence of isomorphous crystallization of mixtures are similar. The trivalent ions of iron and cobalt are similar in size, but the high oxidation potential of cobalt prevents much replacement (Goldschmidt, 1958). The solubility of iron compounds in ground waters is a function of the type of iron compound involved, the amounts and types of other ions in solution, the pH, and the Eh. According t o Larson and King (1954), 100 ppm of ferrous iron can stay in solution at pH 8 and pH 7; the theoretical maximum is about 10,000 ppm. The effects of many other ions, plus temperature and pressure differentials, such as those common to oilfield waters, have not been thoroughly studied. When a ground water in which ferrous iron is dissolved contacts the atmosphere, the following reaction can occur: 2Fe2++ 4HCO3- + H20 + 1 / 2 0
2

+ 2Fe(OH), + 4C02

Sandstone contains iron oxide, iron carbonate, and iron hydroxide, and shales and carbonate rocks contain oxides, carbonates, and sulfides of iron. Oilfield waters with characteristic low redox potentials dissolve some iron from the surrounding rock. The iron occurs in such waters at two levels of oxidation, ferrous or ferric. Knowledge of the amount and type of iron compounds in oilfield waters is used to estimate the amount of corrosion that is occurring in the production system, and t o determine the type of treatment required if the water is t o be used for waterflooding. This knowledge also enables determination of the Eh of the in situ water, because the Eh can be calculated from the Fe+2 and Fe+ values. Shales, sandstones, and carbonates contain about 47,200, 9,800, and 3,800 ppm, respectively, of iron (Mason, 1966). Sea water contains about 0.01 mg/l, and subsurface brines contain from traces to over 1,000 mg/l of iron. Copper Copper is a member of the VIII group of elements, and it is characteristically thiophile; the largest concentrations of it are found in various sulfur compounds. The earths crust contains about 0.01 wt.% of copper (Fleischer, 1962). Its compounds are dissolved easily during weathering, if the pH of the solution is less than 4.5. Many of the water-soluble copper compounds are salts of organic acids such as acetic, citric, and naphthenic. Much of the copper that is dissolved is precipitated afterward as sulfide. Traces of copper remain in the oceans, but its content is kept low because of the adsorption on, or combination with, marine organisms. Miholic (1947) presented an age

ZINC

151

division for mineral waters based on the presence of heavy metals in waters associated with joints and faults caused by tectonic movements of different geological ages. He placed copper as the predominant heavy metal in the Caledonian Group of the Orogenic Epoch (post-Silurian). Biochemical processes are known to be responsible for enriching a deposit in metals such as uranium, copper, and vanadium; therefore, this classification is restricted to waters of igneous origin. Most shales and carbonates contain about 45 and 4 ppm, respectively, of copper, with sandstones containing less than 1 ppm (Mason, 1966). Sea water contains about 0.003 mg/l, and most subsurface brines analyzed in this laboratory contained from less than 0.5 mg/l up to about 3 mg/l. The solubility of copper generally decreases with decreasing redox potential and increases with increasing redox potential if reduced sulfur is present. Most subsurface oilfield brines have relatively low redox potentials. zinc Zinc is a member of the I1 B group of elements and is predominantly thiophile. Its abundance in the crust of the earth is about 0.013 wt.% (Fleischer, 1962). Its geochemistry results from the similarity of its divalent ionic radius and the radii of Mg+, Ni+?, Co+, Fe+, and Mn+ (Goldschmidt, 1958). Zinc is dissolved readily as sulfate or chloride from acid rocks, such as granite, during weathering. Conversely, zinc is not dissolved easily from limestone with which it is deposited. Most alkaline waters do not extract zinc; however, a solution of NH,, NH,NO,, and NaC10, can extract and hold small quantities of zinc; the more acidic the water, the greater the amount of zinc extracted. Zinc is precipitated as the sulfide, oxide, carbonate, or silicate. Traces of zinc are found in sea water, but eventually zinc is deposited in carbonated sediments or in bottom muds or sapropels as sulfide. Shales, sandstones, and carbonates contain about 95, 16, and 20 ppm, respectively, of zinc (Mason, 1966). Sea water contains about 0.01 mg/l, and subsurface brines contain traces to more than 500 mg/l of zinc. Mercury Mercury is a member of the I1 B group of elements, which also includes zinc and cadmium. It is relatively abundant for a heavy element, but still must be considered scarce, with an abundance of about 4 x lo- wt.% of the crust of the earth (Fleischer, 1962). Most commercial deposits of mercury are of hydrothermal origin and are related to magmatic rocks; the commercial ore is cinnabar, HgS, or the liquid metal itself (Goldschmidt, 1958). Mercury is predominantly thiophile, and its geochemistry is controlled by the fact that it is volatile, with a boiling point of 357C, and can be reduced to the metal by ferrous iron. Therefore, in a magmatic environment

152

INORGANIC CONSTITUENTS AND PHYSICAL PROPERTIES

the temperature and the redox potential control its occurrence. It is transported in hot springs (White et al., 1963). Shales, sandstones, and carbonates contain about 0.4, 0.03, and 0.04 ppm, respectively, of mercury. Sea water contains 3 x lo- mg/l, and subsurface oilfield brines contain 0-0.15 mg/l. The samples containing 0.15 mg/l of mercury were found in relatively dilute brines taken from the Cymric and the Rio Bravo oilfields in California. Free mercury is found in the oils produced from these fields, and the ages of the producing formations range from Eocene t o Pleistocene. The mercury content of natural waters has been used t o locate cinnabar deposits (DallAglio, 1968). The amounts of mercury in waters appear t o increase with increasing bicarbonate concentration. Karasik et al. (1965) found that saline waters containing 200,000 mg/l of chloride contain very small amounts of mercury, which suggests that anionic complexes such as HgC14-* may not be important transporters of mercury. Brackish waters containing up t o 3,000 mg/l dissolved solids, up to 400 mg/l of bicarbonate, and the iodide ion sometimes contain up to 10 ppb of mercury, while stronger brines contain <0.1 ppb of mercury, which suggests that mercury may be transported as Hg14-* in brackish waters.

Lead
Lead is a member of the IV A group of elements; it is ubiquitous in the earth, but its abundance in the crust is only about 0.002 wt.% (Fleischer, 1962). It is extracted from its minerals during weathering and migrates in the form of soluble-stable compounds. It is particularly soluble in acetic and other acids. Because the bicarbonate form is more soluble than the carbonate, lead can be transported as the bicarbonate. Most of the lead is precipitated from waters before they reach the sea. Hemley (1953) studied lead sulfide solubility related to ore deposition from saline waters. He concluded that lead-complex concentrations increase with increasing concentrations of bivalent sulfur and decrease at pH values above 7. The solubility of lead is limited primarily by the solubility restrictions of its sulfide and sulfate in reducing and oxidizing systems. How its solubility is influenced by many other ions, such as those found in a brine, has not been sufficiently studied. Shales, sandstones, and carbonates contain about 20, 7, and 9 ppm of lead, respectively. Sea water contains about 0.003 mg/l, and subsurface brines contain trace amounts to more than 100 mg/l of lead. Cadmium Cadmium is a member of the I1 B group of elements and may be considered one of the rarer elements; its abundance is about 3 x lo- wt.% of the earths crust (Fleischer, 1962). It is strongly thiophile, but its chemistry

BORON

153

differs from that of zinc in that it will precipitate from a strong acid solution, whereas zinc will not. There are few independent cadmium minerals, and its distribution is mainly that of a guest atom or ion in minerals. It frequently is present in lead-zinc deposits and occurs in solid solution in hypogene sulfides. A main carrier of cadmium is sphalerite, and oxidation of sphalerite or other sulfides containing cadmium will release the soluble cadmium sulfate. Shales and carbonates contain about 0.3 and 0.035 ppm of cadmium, respectively, and sandstones contain less than 0.01 ppm (Mason, 1966). Sea water contains about 0.0001 mg/l, and the subsurface oilfield brines may contain from 0 to about 0.001 mg/l of cadmium. Subsurface brines of the sulfate type in contact with lead-zinc deposits probably contain higher concentrations of cadmium. Boron Boron is a member of the I11 A group of elements, and it is an oxyphile and lithophile element. Its abundance in the crust of the earth is about 0.001 wt.% (Fleischer, 1962). It has small atomic and ionic radii. Knowledge of the presence of boron compounds in oilfield waters is important for several reasons. Boron is useful in identifying the sources of brines intrusive t o oil wells, or in fresh-water lakes or streams. In concentrations exceeding 100 mg/l, it affects electric log deflections. Boron is present in oilfield brines as boric acid, inorganic borates, and organic borates. When it is present as undissociated boric acid, it is an important buffer mechanism, being second only to the carbonate system. It may be precipitated as the relatively insoluble calcium and magnesium borates. Kazmina (1951) calculated the borate-chloride coefficient of some Russian oilfield waters. With a plot of the borate-chloride coefficient in logarithmic coordinates as a function of chloride content, he distinguished genetic groups of natural waters found in oil-bearing regions. Mitgarts (1956) studied the significance of boron and other elements in petroleum prospecting. In general, boron, together with bromine and iodine, is always associated with waters accompanying petroleum. Like chlorine, it can be considered an element of marine origin. The solubility of most boron compounds, the hydrolytic cleavage of boron salts, and their ability t o be occluded and coprecipitated with other compounds account for the extensive migration of boron. Soluble-complex boron compounds in brines and connate waters probably are there as a result of the decay of the same plants and animals that were the source of petroleum. Shales, sandstones, and carbonates contain about 100, 35, and 20 ppm, respectively, of boron. Sea water contains about 4.8 mg/l, and subsurface oilfield water contains from trace amounts to more than 100 mg/l. Fig. 5.13 is a plot of chloride versus boron concentrations of some oilfield brines taken from some sediments of Tertiary, Cretaceous, and Jurassic age. The plot

154

INORGANIC CONSTITUENTS AND PHYSICAL PROPERTIES

---I200
T T

Normal evaporite curve,

a
50
0

"

30

"k/~ /

c
J

C "

T c

BORON,mg/I

Fig. 5.13.Comparison of the boron concentrations of some Tertiary (T), Cretaceous (C), and Jurassic (J) age formation waters from Louisiana with an evaporating sea water.

BORON, m g / l

Fig. 5.14. Comparison of the boron concentrations of some waters from Pennsylvanian (P) and Mississippian (M) age sediments with an evaporating sea water.

indicates that the majority of these brines are enriched in boron relative to a normal evaporite-formed brine, and that the samples that were depleted in boron may have contained dissolved halite. Fig. 5.14 is a similar plot for some samples taken from some Pennsylvanian and Mississippian age sediments. Boron is one of the elements whose concentration in the aqueous phase increases as a brine is evaporated, as illustrated in Table 5.11.

ALUMINUM

155

Aluminum Aluminum is the third most abundant element in the earths crust, but its concentration in natural waters usually is less than 1 mg/l. The ionic radius of trivalent aluminum is 0.57 (Goldschmidt, 1958),and it usually behaves as a cation when 6-coordinated with oxygen compounds. However, when 4-coordinated, it usually acts like the central atom of an anion. The 4-coordination usually, but not exclusively, is associated with minerals formed at high temperatures, but the 6-coordination is associated with minerals formed at low temperatures, which includes most sediments in the petroleum environment. The clay minerals illite, kaolinite, and montmorillonite often contain about 13.5, 21, and 11%aluminum, respectively. Quartzites, sandstones, limestones, and shales contain about 0.7, 3.0, 0.6, and 10% aluminum, respectively. During weathering silica will leach out and leave aluminum hydroxide behind (Pirsson and Knopf, 1947), and sedimentation processes leave only about 0.4 mg/l aluminum in sea water. According t o Hem (1970), the cation A P 3 predominates in solutions with a pH of 4.0 or less. Above pH 4.5, polymerization gives rise to an aluminum species with a gibbsite (aluminum hydroxide) structural pattern. Above pH 7.0, the dissolved form is the anion A1 (OH),-. The pH of the water is the main control of the amount of alumium that is likely to be present in natural waters. A water with a pH less than 4.0 may contain 1% more of aluminum; for example, waters associated with acid or mine drainage. Oilfield waters contain trace amounts t o more than 100 mg/l of aluminum.

A 1ha 1inity
Alkalinity is defined as the capacity of a solution to neutralize an acid, usually t o a pH of 4.5. A solution with a neutral pH of 7.0 may have a considerable amount of alkalinity; therefore, alkalinity is a capacity function, in contrast to pH, which is an intensity function. The alkalinity-pH ranges originally coincided with methyl orange and phenolphthalein color end points. The potentiometric titration produces more accurate alkalinity results, and it utilizes an end point where the most abrupt pH change occurs while specific increments of a standard acid are added. Alkalinity usually is caused by the presence of bicarbonate, carbonate, or hydroxyl ions in a water; however, the weak acids such as silicic, phosphoric, and boric can contribute titratable alkalinity species. Carbon dioxide, which is dissolved in circulating waters as bicarbonate or carbonate as a result of the carbon cycle, is the prime source of alkalinity in shallow ground waters. However, in deep subsurface brines, additional carbon dioxide probably is dissolved as a result of diagenesis of inorganic and organic compounds. Most oilfield waters contain no hydroxyl ions, and most of them contain

156

INORGANIC CONSTITUENTS A N D PHYSICAL PROPERTIES

no carbonate ions, but they do contain bicarbonate ions. Some oilfield waters in the Rocky Mountain area are alkaline and contain both primary and secondary alkalinity, where primary alkalinity is that associated with the alkali metals and secondary alkalinity is that associated with the alkaline earth metals. For example, the Green River formation waters that are in or near trona beds may contain more than 20,000 mg/l of carbonate and 5,000 mg/l of bicarbonate. Most oilfield waters from other areas contain from about 100 to 2,000 mg/l of bicarbonate.

Acidity
The basis of acidity is the solvated hydrogen ion H 3 0 + ,which is found in nature. Volcanic emanations produce HF, HCl, and H2SO4, probably formed by reactions between water and constituents associated with the magma. Waters associated with peat may contain organic acids, rain waters may contain carbonic acid, and waters associated with reducing conditions and anaerobic bacteria may contain H2 S. Acidity, as contrasted to alkalinity, is the capacity of a solution to neutralize a base, usually from below pH 4.5 t o pH 7.0. Most oilfield brines normally do not contain acidity. New wells or reworked wells often are acidified or "acidized" with a strong mineral acid or a combination of mineral and organic acids. This treatment causes the produced water to contain a certain amount of acidity until all of the acid is neutralized or diluted. Because of the large quantities of acids used in some treatments, it m a y take 6 months or more for the water produced from a treated well t o return to normal. Organic acids and organic acid salts commonly are found in oilfield waters, and the concentration ranges from trace amounts to more than 3,000 mg/l. Silica Silicon is the second most abundant element in the earth's crust, which contains about 27 wt.% of it (Fleischer, 1962). It always occurs in a combined form. Most of the silicon compounds involve structures with oxygen, and there are about a thousand silicate minerals in the earth's crust; however, those which are predominant are relatively few in number. The solubility of silica in water is a function of temperature, pressure, pH, and other ions in solution. Most silica in natural water probably is in the form of monomolecular silicic acid, H4 Si04 or Si(OH)4. Collins (196913) studied the solubility of a serpentine in solutions of calcium chloride and sodium chloride a t temperatures from 30" to 200C and pressures from 176 t o 1,055 kg/cm2. The solubility calculated as silicon molarity in solution increased with increasing concentrations of sodium chloride, increasing pressure, and increasing temperature up to about 125C. Between about 125" and 2OO0C, the solubility decreased with increasing temperature. The solu-

AMMONIUM NITROGEN

157

bility of silicon in the presence of calcium chloride solutions decreased with increasing Concentration of calcium chloride and with increasing temperature above about 100C. The solubility increased with increasing pressure and with increasing temperature between 3' and about 100C. Subsurface 0 petroleum-associated brines usually contain less than 30 mg/l of dissolved silica; Rittenhouse et al. (1969) report that their silica content ranges from about 1 to 500 ppm as silicon, and that some low salinity waters contained a higher median content of silica than more saline waters in other areas. Ammonium nitrogen Ammonium contains nitrogen in the N-3 oxidation state, a reduced form. Nitrogen can occur in all of its states of oxidation, ranging from -3 t o +5. Oxidation of the reduced forms produces nitrogen gas, N 2 , and other nitrogen species up to nitrate, NO3-. Ammonia, NH3, forms during the anaerobic decay of organic nitrogenous material. The petroleum genetic environment produces ammonia, which transforms to ammonium, NH4, in many petroleum-associated waters because the redox potential is too low to oxidize the ammonia to nitrate. The ammonium ion is too weak t o be successfully titrated; however, Collins et al. (1969), developed a technique using formaldehyde, whereby a produced strong acid can be titrated.
TABLE 5.V Ammonium content of 10 subsurface brine samples Sample State Utah Utah Utah Okla. Okla. Utah Utah Okla. Okla. Okla. Formation Navaho Green River Lower Green River Morrow Rue Uinta Surface, Green River Green River Hunton Oswego Chester

1 2 3 4 5 6 7 8 9 10

30-143 0 852-1,719 71 91 914-1,737 2,713-2,715 0 116 1,958-1,963 717-1,111 2,069 5 68 1-1,5 5 6 2,696-2,77 1 434 1,928-1,951 233 23 2,408-2.437

The NH4N content of several oilfield brines was determined and a wide variation in concentration was found. Table 5.V illustrates the amounts of NH4N found in 10 samples taken from subsurface rocks in Oklahoma and Utah.

158

INORGANIC CONSTITUENTS A N D PHYSICAL PROPERTIES

Phosphorus

Phosphorus occurs in the earths crust almost exclusively as the ion, and a large percentage of it is contained in the apatite group of minerals, which primarily are related to igneous rocks. The crust of the earth contains about 0.12 wt.% of phosphorus. It is a member of the V A group of elements with oxidation states ranging from -3 to +5. In contrast t o nitrogen, phosphoric acid and phosphates are not oxidizing agents. The phosphorus species present in most natural waters probably is the phosphate anion, and it usually is reported as an equivalent amount of the orthophosphate ion (PO4 )-3, the final dissociation product of phosphoric acid, H3PO4. This dissociation occurs in four steps, giving four possible phosphate forms: H3PO4, H2PO4-, HP04-2, and P 0 4 - 3 . In the alkalinity titration, any HP04-2 is converted to H2P04- and appears as bicarbonate. Shales, sandstones, and carbonates contain about 700, 170, and 400 ppm, respectively, of phosphorus. Sea water contains about 0.07 mg/l. A detailed study of the content of phosphorus in subsurface brines has not been made, but of the few that have been analyzed, most have contained less than 1 mg/l. Arsenic Arsenic is a member of the V A group of elements and probably occurs in nature mainly in the form of arsenides and sulfarsenides; it rarely occurs in its elemental form. It is comparatively rare, and the earths crust contains about 0.0005 wt.% of it (Fleischer, 1962). In an acidic environment, the oxidized ion, A s O ~ - ~is mobile, and mineral arsenates tend t o be solu, bilized. The arsenates usually are formed in oxidation zones in contact with atmosphere and free oxygen, and arsenic will precipitate with ferric iron hydroxide. Glauconitic sediments have been found which contain up t o 70 ppm of arsenic (Goldschmidt, 1958). Subsurface oilfield brines may contain arsenic as HAs02- or H2 As04, depending upon the Eh and pH. A low Eh may favor the HAs02- form. Shales, sandstones, and carbonates contain about 13, 1, and 1 ppm, respectively, of arsenic (Mason, 1966). Sea water contains about 0.003 mg/l and subsurface oilfield brines contain from 0 to 10 mg/l. Compounds containing arsenic sometimes are used in corrosion inhibitors; therefore, information concerning well treatments should be obtained before assuming that any arsenic found occurs naturally. Oxygen Oxygen is the most abundant element in the earths crust, which contains about 49 wt.% of it (Fleischer, 1962). I t is capable of existing in many types of combinations, and even though it is highly active, it occurs extensively in

SULFUR

159

the free form. Most combined oxygen is ionic; however, it forms a covalent molecule with hydrogen, namely water. It also forms complex oxy-salts with various metals. The oxygen content of rocks decreases with depth. The solubility of oxygen in water is primarily a function of temperature and pressure, and surface waters at ambient conditions may contain 7.63 mg/l a t 3OoC and 11.33 mg/l at 10C (Hem, 1970). The amounts of dissolved oxygen in subsurface petroleum-associated waters is usually low, and in most in situ conditions it is undetectable because of the low redox potential of the environment. It can cause corrosion problems in the well pipes, but in most cases it is atmospheric oxygen that mixes with the produced brine during production operations that causes oxygen corrosion. Sulfur Sulfur is a member of the VI A group of elements and is widely dispersed in sedimentary and igneous rocks as metallic sulfides. The crust of the earth contains about 0.05 wt.% of sulfur (Fleischer, 1962). Free sulfur often is related to volcanic activity and can be deposited directly as a sublimate. Many commercial deposits, however, are associated with sedimentary gypsum, and probably result from biogenic activity such as that. of anaerobic bacteria. Large deposits of sulfur are found in caprocks of anhydrite overlying some salt domes. Hydrogen sulfide, often found in oilfield waters, is formed by anaerobic bacteria. One such species of bacteria is the Desulphouibrio,which obtains its oxygen from sulfate ions, causing them to be reduced to hydrogen sulfide. complexed with Sulfur in surface water usually occurs in the form ( S 6 ) oxygen as the sulfate anion S04-2. As previously mentioned, the conversion of oxidized sulfur t o a reduced form commonly involves a biogenic process, and such a reduction may not occur unless these bacteria are present. The Eh of subsurface oilfield brines usually is somewhat reducing, and the sulfur species in such environments can include hydrogen sulfide (H2 S), sulfite and thionates (S406-). Detailed studies of the sulfur species in subsurface brines have not been made, and it is likely that other forms of sulfur are present in some brines. The temperature, pressure, Eh, pH, and other constituents in solution all influence the types of dissolved sulfur that occur in oilfield brines. Shales, sandstones, and carbonates contain about 2,400, 240, and 1,200 ppm, respectively, of sulfur (Mason, 1966). Sea water contains 900 mg/l of sulfur as sulfate, and subsurface oilfield brines contain from none up to several thousand milligrams per liter. The amount of sulfate in the brine is influenced by bacterial activity and by how much calcium, strontium, and barium is present. If these three cations are present in relatively high concentrations, the amount of sulfate present will be low. However, some brines containing high concentrations of magnesium and low concentrations of the other alkaline earth metals may contain high concentrations of sulfate.

160

INORGANIC CONSTITUENTS AND PHYSICAL PROPERTIES

Selenium Selenium is a member of the VI A group of elements and occurs in -2,0, +4,and +6 valence states, respectively. It is a scarce element with an abun0 dance of about 9 x 1 " wt.% of the crust of the earth (Fleischer, 1962). Large areas of North America are underlain by seleniferous rocks and soils. These seleniferous rocks are of sedimentary origin and range in age from Late Paleozoic to Holocene. Selenium is the only known element that can be absorbed by plants in sufficient amounts to make them lethal when eaten by animals (Trelease, 1945).Fig. 5.15 illustrates the distribution of seleniferous vegetation. Sandstones, shales, and carbonates contain about 0.6, 0.05,and 0.08 ppm, respectively, of selenium. Sea water contains about 0.004 mg/l of selenium. A few subsurface oilfield brines from areas where selenium is present in soils were analyzed at this laboratory, but no selenium was detected in the brines analyzed. Most brines are present in a petroleum environment under reducing conditions, and in such an environment, selenium likely is reduced to the element and precipitated. However, in areas where outcrop water flows through petroleum-bearing formations, it is possible that selenium in the form of t)e anion Se03-2 may be present.

1
Fig. 5.15. Distribution of seleniferous vegetation in the United States.

FLUORINE

161

Fluorine Fluorine is a member of the VII A group of elements and is the most electronegative of all the elements. Its ionic radius is 1.33 A, which is about the size of OH- and O-*; therefore, it enters a variety of minerals. The earths crust contains about 0.03 wt.% of fluorine (Fleischer, 1962). In solutions, fluorine usually forms the fluoride F- ion; at a low pH, the HFo form might occur. It also can form strong complexes with aluminum, beryllium, and ferric iron. Fluorine occurs in several minerals, but the only common industrial source is fluorspar (CaF2). It occurs as HF or SiF in volcanic emanations, and even as the free element in (stinkfluss) stinking fluorspar of Wolssendorf, Bavaria. The solubility of calcium fluoride (fluorite) in water at 25OC is . about 8 7 ppm of fluoride (Aumeras, 1927);this solubility could be affected by other dissolved constituents. Sodium fluoride is very soluble, and magnesium fluoride is more soluble than calcium fluoride; therefore, a petroleumassociated water that is deficient in calcium and has been in contact with rocks containing fluoride minerals will contain appreciable quantities of fluoride. Shales, sandstones, and carbonates contain about 740, 270 and 330 ppm of fluorine, respectively (Mason, 1966). Sea water contains about 1 3 mg/l, . and natural waters with a dissolved solids concentration of less than 1,000 mg/l usually contain less than 1 mg/l of fluoride. However, concentrations up t o 50 mg/l have been reported (Hem, 1970). Not many subsurface petroleum-associated brines have been analyzed for fluoride, but a few are known t o contain up t o 5 mg/l. Chlorine Chlorine is a member of the VII A group of elements and is the most important member of the group with respect to water. The crust of the earth contains about 0.19 wt.% of chlorine (Fleischer, 1962);some estimates place the fluorine abundance above the chlorine abundance. Volcanic activity produces the gas hydrogen chloride and sometimes chlorine, but much less frequently. The caliche evaporite deposits in Chile contain the perchlorate ion C104-; however, the mechanism by which it formed is not clear. Several minerals contain the chloride ion. The chloride ion does not form low-solubility salts. I t is not easily adsorbed on clays or other mineral surfaces. It is not significant in oxidation and reduction reactions, and it forms no important solute complexes. Chloride is very mobile in the hydrosphere, yet it is relatively scarce in the earths crust. It is the predominant anion in sea water and in most petroleum-associated waters. It is found in all natural waters, and its average concentration in rainwater is about 3 mg/l (Hem, 1970). Chloride salts are

162

INORGANIC CONSTITUENTS AND PHYSICAL PROPERTIES

very soluble; therefore, chloride is usually not removed from solution except during freezing or evaporation processes and in hyperfiltration, as water moves through some types of clay beds (White, 1965). Shales, sandstones, and carbonates contain about 180, 10, and 150 ppm, respectively, of chloride. Sea water contains about 19,000 mg/l of chloride, the principal anion in the sea. The chloride content of the hydrosphere is much greater than can be accounted for by weathering of rocks, and it has been postulated that the primordial atmosphere may have been rich in chlorine compounds. The volcanic emission of chlorine gases appears a more plausible explanation, however. Oilfield brines usually contain relatively high concentrations of chloride; in some brines the concentration may be 200,000 mg/l or more. Chloride usually is the predominant anion in oilfield brines. Table 5.1 illustrates how its concentration can increase in an evaporite-associated brine. Evaporation probably is the only geochemical process which appreciably affects the chloride content of the seas. Bromine Bromine is a member of the VII A group of elements and it behaves somewhat similarly t o chlorine. The crust of the earth contains about 0.0005 wt.% of bromine (Fleischer, 1962). It usually occurs as the ion bromide Br-, and it does not form its own minerals when sea water evaporates (Valyashko, 1956). It forms an isomorphous admixture with chloride in the solid phases. The order of crystallization (see Table 5.1) is halite (NaCl), sylvite (KCl), carnallite (MgC12-KC1*6H2 and/or kainite (MgS04 *KC1=3H2 and at 0), 0), the eutectic point, bischofite (MgCl? -6H2 0). Each of these chlorides entrains bromide in the solid phase. This distribution accounts for the relative enrichment of bromide in the liquid phase because with each crystallization more bromide is left in solution than is entrained in the solid phase. Mun and Bazilevich (1962) reported that, in fresh-water lakes, bromide accumulates in the muds, that its concentration is proportional t o the organic-matter concentration in the sediments, and that it is not influenced by the pelitic fraction. In muds of salt lakes, the higher the bromide concentrations in the brine, the higher it is in the muds. In general, the bromide content in the pore solutions increased with depth, but the bromide content in muds decreased with depth, owing to more complete decomposition of organic bromine compounds. Bromide is two t o three times more concentrated in carnallite than in sylvite and five to ten times more than in halite (Myagkov and Burmistrov, 1964). Apparently, concentration and dilution are responsible for the complex distribution of bromide in rocks of a carnallite zone. The determining factor in the replacement of chloride by bromide is the mineral composition rather than the bromide concentration in the brine.

BROMINE

163

Braitsch and Herrmann (1963) found that the absolute bromide content of rocks can be used to determine primary and secondary paragenesis. Distribution of bromide between solution and crystals of halite, sylvite, carnallite, and bischofite, and the effects of other ions plus temperatures between 25" and 83OC, confirm this. This method was also applied to determine the temperature of primary potash deposits. An investigation of the bromide/ sodium chloride relation in salt deposits revealed that bromide can be used to determine the stratigraphy of evaporite-salt deposits (Baar, 1963). Derivation of theoretical profiles of bromide thickness versus salt thickness indicated that, with constant inflow, evaporation, and reflux, the thickness profiles were all monotomic logarithmic functions. The irregular and high bromide concentrations of some salt deposits were attributed to inflow of bromide-rich bitterns from an adjacent potash basin (Holser, 1966). Shales, sandstones, and carbonates contain about 4,1,and 6 ppm, respectively, of bromide (Mason, 1966). Sea water contains about 65 mg/l of bromide, and subsurface petroleum-associated brines contain from less than 50 to more than 6,000 mg/l of bromide. Fig. 5.16 illustrates the bromide concentration plotted versus the chloride concentration for some subsurface brines taken from Tertiary, Cretaceous, and Jurassic age sediments. This plot indicates that the waters from these Tertiary age sediments are depleted in bromide relative to a normal evaporite brine, whereas those from the Cretaceous and Jurassic age sediments are enriched'in bromide.

BROMIDE, mg / I

Fig. 5.16. Comparison of the bromide concentrations in some formation waters from Tertiary (T), Cretaceous (C), and Jurassic (J) age sediments from Louisiana with an evaporating sea water.

164

INORGANIC CONSTITUENTS A N D PHYSICAL PROPERTIES

Normal evaporite curve

P M PI

300

1,000 BROMIDE, m g l l

3000

1 0

Fig. 5.17. Comparison of the bromide concentrations in some formation waters from Pennsylvanian (P) and Mississippian (M) age sediments from Oklahoma with an evaporating sea water.

Fig. 5.17 is a similar plot for some brines taken from some Pennsylvanian and Mississippian age sediments. The bromide concentrations in these brines do not appear t o be significantly different. Brines containing 1,500 to 8,000 mg/l of bromide, with calcium and magnesium chloride as the major constituents, are formed by evaporation of sea water and associated sedimentation rather than by dissolution of salts. Increase in temperature causes a phase shift in the solid and brine phases, resulting in an increase of bromide in solution. Iodine Iodine is a member of the VII A group of elements, and of the four members discussed in this chapter, it is the least abundant, since it comprises only about 3 x lo-' wt.% of the earth's crust (Fleischer, 1962). I t forms three minerals of its own; namely, iodoargyrite (AgI), iodoembolite [Ag(Cl,Br,I)], and miersite [(Ag,Cu)I]. Marine plants, such as kelp and plankton algae, concentrate iodine. The distribution of iodide in marine and oceanic silts and interstitial waters indicates that near-shore ocean Sediments contain more iodide than deep-sea sediments. Red clays and calcareous sediments contain less iodide than organic-bearing argillaceous sediments. The iodide concentration in the marine and oceanic sediments decreases with depth, but the iodide concen-

IODINE

165

tration in the interstitial waters increases with depth (Shishkina and Pavolva, 1965). The iodide in bottom water layers and in the interstitial water of muds in some Japanese lakes was found t o be selectively captured by flocculated iron, manganese, and aluminum hydroxides which sank to anaerobic layers (Sugawara et al., 1956). Reduction of the hydroxides releases iodide to the bottom waters. However, the release of iodide is incomplete, and the flocculates reach the bottom muds where the Eh is even more negative, resulting in high accumulation of iodide in interstitial water of muds. The primary source of organic matter in marine and oceanic basins is photosynthesis by plankton algae. Algae are directly or indirectly the food resource of all the remaining life in the basins, and the proliferation rate differential and the types of feeding organisms influence the sediment deposition rate as well as the amount of iodide and bromide in the sediment (Bordovskii, 1965). Shales, sandstones, and carbonates contain about 2.2, 1.7, and 1.2 ppm, respectively, of iodide (Mason, 1966). Sea water contains about 0.05 mg/l, and most subsurface petroleum-associated brines contain less than 10 mg/l; however, some have been found to contain up t o 1,400 mg/l. Fig. 5.18 is a plot of the chloride concentrations versus the iodide concentrations for some brines taken from some Pennsylvanian and Mississippian age sediments. Iodide is tremendously enriched in all of these brines compared to the normal evaporite-associated brine. Some mechanisms such as leaching or solubilization of iodine, iodate, or iodide compounds, ion filtration, anion exchange, and desorption had t o occur, t o account for this enrichment of iodide in the aqueous phase. A similar plot for some waters taken from Tertiary, Cretaceous, and Jurassic age sediments gave similar results except that these particular brines were not as heavily enriched in iodide. The iodide concentration of some subsurface waters is dependent on the proximity of argillaceous deposits containing organic matter, rather than on dissolved mineralization. Gas may play an important part in the accumulation of iodide in subsurface waters. Some gas structures are bounded by iodide-rich waters, and the iodide content is depleted at a distance from the gas structure (Ovchinnikov, 1960). Studies of some reservoirs, Holocene to Miocene in age, in lagoonal sedimentary basins of thick sediments with wide areal extent, indicate that a genetic relation exists between iodide in the formation waters and the accompanying natural gas (Marsden and Kawai, 1965). Possibly the high concentrations of iodide are the result of concentration by algae and other marine organisms from ancient sea waters; their remains became part of the sediments, and later the iodide was solubilized. However, because the iodide usually is strongly incorporated in the sediment, such sediments must contain large quantities of iodide, and other mechanisms must operate to solub i k e the iodide in associated waters.

166
1 ,

INORGANIC CONSTITUENTS AND PHYSICAL PROPERTIES

120 ,0
~ w mevaporite i
\
CUM

cn

I#OOO

500 200

w '

2 s I
O

loot
t

/1
/

, ,
/-

'

P P

v s
MM M M F P M MhP

20

50E

IODIDE, mg/l
Fig. 5.18. Comparison of the iodide concentrations of some formation waters from Pennsylvanian (P) and Mississippian (M) age sediments from Oklahoma with an evaporating sea water.

Theoretically, only iodate is thermodynamically stable in sea water (SillCn, 1961). The exact form of iodine in oilfield brines has not been investigated. These forms probably will vary with the salinity, Eh, and other factors. Sugawara and Terada (1957)established that both iodide and iodate are present in comparable amounts in sea water. Biologists found that iodine-concentrating algae ultilize only the iodate form (Shaw, 1962). Significance of some physical properties

Redox potential
The redox potential often is abbreviated as Eh, and may also be referred to as oxidation potential, oxidation-reduction potential, or pE. It is expressed in volts, and at equilibrium it is related to the proportions of oxidized and reduced species present. Standard equations of chemical thermodynamics express the relationships. Eo is the standard potential of a redox system when unit activities of participating substances are present under standard conditions. Eo is related to standard free energy change in a reaction by the equation:

where n is the number of unit negative charges (electrons) shown in the

PHYSICAL PROPERTIES

167

redox reaction and f is the Faraday constant in units that give a potential in volts (94,484 absolute coulombs). Standard free energy values are given in texts such as that of Latimer (1952). When the system is not under standard conditions, the redox potential is expressed by the Nernst equation: Eh = E o species) + RfTlog (oxidizedspecies) n (reduced

where R is the gas content (1.987 cal. degree mole), and T is the temperature in degrees Kelvin. Geochemical literature and biochemical literature, such as that of Pourbaix (1950), present increasing positive potential values to represent increasing oxidizing systems and decreasing potential values to represent reducing systems. The sign of Eh used in this manner is opposite to standard American practice in electrochemistry. Zobell (1946) established basic procedures for measuring the Eh of geologic-related materials. The Zobell solution containing 0.003M potassium ferrocyanide and 0.003M potassium ferricyanide in a 0.1M potassium chloride solution has an Eh of 0.428 V at 25OC. Minor temperature variations can be calculated using the equation: Eh = 0.428+).0022 ( t - 25) where t = temperature of the sample in degrees Celsius. Garrels and Christ (1965) describe procedures for determining Eh equilibria of mineral substances. Particularly useful are the procedures described for constructing diagrams showing fields of stability for various mineral substances as functions of pH and Eh. Fig. 5.19 is an Eh/pH diagram. Such stability field diagrams might be constructed for the substances comprising petroleum and should be of considerable help in understanding the mechanisms of origin, accumulation, and chemical stability of petroleum. Unfortunately, this approach does not yield simple results because most oxidation reactions involving hydrocarbons and other petroleum constituents are not reversible in the usual sense. Furthermore, thermodynamic data are available for only a small fraction of the large number of reactions and products that are possible. Attempts t o obtain useful results from Eh measurements in natural media involve numerous difficulties. In a natural medium, such as petroleumassociated water, there are many variables, none of which is controlled, which individually or collectively may have little or great influence on Eh measurements made on the water. Many chemical substances, such as ferric or ferrous ions, various organic oxidation-reduction systems, sulfides, and sulfates, may be present in the water in large or small amounts. Even controlled systems in the laboratory often produce unaccounted-for variances. In the field, the lack of knowledge of actual participating species may seriously impair proper interpretation of Eh readings. Eh measurements made

168

INORGANIC CONSTITUENTS A N D PHYSICAL PROPERTIES

1,200 -

H2O

000 600

1
Fa++ Acid

<
W

400-

I
I I

/Modern Alkaline

sea water

200

on poorly poised media (media with poor Eh stability), such as some oilfield waters, involve additional uncertainties. Response of electrodes in such solutions is sluggish, and electrodes are easily contaminated with trace amounts of substances which will produce invalid readings. In the natural environment, reactions occur that involve protons and electrons, such as: FeS04 + 2H2 0

* S04-2

+ FeO-OH + 3H+ + e-

Such reactions depend upon both the pH and the Eh of the system, and the equilibrium line of such reactions is Eh = E o - 59 a/n pH mV, where a is the number of protons. Knowledge of the redox potential is useful in studies of how compounds such as uranium (Naumor, 1959), iron, sulfur (Hem, 1960), and other minerals (Cloke, 1966; Pirson, 1968) are transported in aqueous systems. The solubility of some elements and compounds is dependent upon the redox potential and the pH of their environment. The Eh/pH diagram shown in

PHYSICAL PROPERTIES

169

Fig. 5.19 can be used to predict that ferrous ions are the more common form of dissolved iron and that ferric ions will precipitate in an oxidizing environment if the pH is above 1. Similar diagrams can be drawn for other constituents. Some water associated with petroleum is connate water, and Fig. 5.19 indicates that such water has a negative Eh; this has been proven in various field studies (Buckley et al., 1958). The Eh of some petroleum-associated waters in the Anadarko Basin ranged from -270 mV to -300 mV (Collins, 1969a). Knowledge of the Eh is useful in determining how t o treat a water before it is injected into a subsurface formation (Ostroff, 1965). For example, the Eh of the water will be oxidizing if the water is open to the atmosphere, but if it is kept in a closed system in an oil-production operation, the Eh should not change appreciably as it is brought to the surface and then reinjected. In such a situation, the Eh value is useful in determining how much iron will stay in solution and not deposit in the well bore. Organisms that consume oxygen cause a lowering of the redox potential. In buried sediments, it is the aerobic bacteria that attract organic constituents which remove the free oxygen from the interstitial water. Sediments laid down in a shoreline environment will differ in degree of oxidation as compared t o those laid down in a deep-sea environment (Pirson, 1968). For example, the Eh of the shoreline sediments may range from -50 t o 0 mV, but the Eh of deep-sea sediments may range from -150 to -100 mV. The aerobic bacteria die when the free oxygen is totally consumed; the anaerobic bacteria attack the sulfate ion, which is the second most important anion in the sea water. During this attack, the sulfate reduces t o sulfite and then t o sulfide; the Eh drops to -600 mV; H2S is liberated, and CaC03 precipitates as the pH rises above 8.5 (Dapples, 1959).

The term pH means the logarithm (base 1 0 ) of the reciprocal of the hydrogen-ion concentration, and the pH of pure water at 25OC is 7.0, which means that there is lo- mole per liter of H+ in solution. When other constituents are solubilized by water, the pH probably will change because the chemical equilibrium shifts as new ions combine with H+ or OH-. The presence of slightly dissociated acids or bases will tend to buffer the solution, and the addition of H+ or OH- will shift the pH only a small amount until the acids or bases are changed to salts. The pH of oilfield waters usually is controlled by the carbon dioxidebicarbonate system. Because the solubility of carbon dioxide is directly proportional to temperature and pressure, the pH measurement should be made in the field if a close-to-natural-conditions value is desired. The pH of the water is not used for water identification or correlation purposes, but it will indicate possible scale-forming or corrosion tendencies of a water. The

170

INORGANIC CONSTITUENTS A N D PHYSICAL PROPERTIES

pH also may indicate the presence of drilling-mud filtrate or well-treatment chemicals. A detailed study indicates that virtually no environment exists on or near the earths surface where the pH/Eh conditions are incompatible with organic life (Baas Becking et al., 1960). Because COz is the main byproduct of organic oxidation and the building material of plant and much bacterial life, it must be expected t o play a dominant role. It dissolves in HzO, producing the bicarbonate ion and a free hydrogen ion. The concentration of moles per liter (pH 7) at 25OC in pure water, the hydrogen ion is 1 x but when saturated with COz, it rises t o 1 x lo- moles per liter (pH 5). The equilibrium conditions of carbon dioxide, carbonic acid, and the bicarbonate ion are: Hz 0 + COz

* HzCO, * HC03-

+ H+

* 2H+ + C03-

and the pH of each equilibrium in ocean water is pH 5, pH 6.3, and pH 10.3. This reaction moves to the right with increasing temperature in a closed system. In the presence of organic constituents, the equilibria are modified, and the pH range can extend from 2 to 12.

Fig. 5.20. Changes in pH as a result of the addition of carbonate ions to distilled water and water solutions containing sodium and chloride ions.

The pH of concentrated brines usually is less than 7.0, and the pH will rise during laboratory storage, indicating that the pH of the water in the reservoir probably is appreciably lower than many published values. Addition of the carbonate ion to sodium chloride solutions will raise the pH, as illustrated in Fig. 5.20. If calcium were present, calcium carbonate would precipitate. The reason why the pH of most oilfield waters rises during storage in the laboratory is because of the formation of carbonate ions as a result of bicarbonate decomposition.

PHYSICAL PROPERTIES

171

Ionic radii
Table 5.VI contains data concerning the radii of the nonhydrated ions, the hydrated ions, the ionic potential, and the polarization. The size of the ions is of interest concerning the mobilities or the relative transport coefficient of a given ion through a clayshale membrane system or the replacement coefficient in a clay ion-exchange system. The ionic potential is of interest because elements with low ionic potentials are the most likely to remain in true solution. The polarization, which is equivalent to the valency divided by the ion radius, is of interest because the larger the polarization, the lower the replacing power in an exchange system (Collins, 1970). The ionic potentials of the constituents involved in the diagenesis are important (Hem, 1960). Those that stay in true ionic solution to rather high pH levels include Na+, K+, Mg+, Fe+, Mn+, Ca+, Sr+*,and Ba+ ;they are the soluble cations, and their ionic potentials range from 0.3 to 1.3, where the ionic potential is the ratio between the ionic charge and the ionic radius. Constituents that are precipitated by hydrolysis are those with ionic potentials of 3-12 and include such ions as A P 3 , Fe+3, S P 4 , and M r P 4 . Constituents which form soluble complex ions and usually go into true ionic solution include B+3, C 4 , + 5 , P+, S 6 , N and Mn+ ;their ionic potentials are over 12. In general, the hydroxides of the soluble cations possess ionic bonds; therefore, they are soluble. The hydrolysates, or those ions precipitated by hydrolysis from hydroxyl bonds, and the soluble complex ions both have hydrogen bonds.

TABLE 5.VI Radii, valence, ionic potentials, and polarization Constituents Nonhydrated radius (A) Valence Hydrated Ionic radius (A) potential Polarization

Lithium Sodium Potassium Calcium Magnesium Strontium Barium Boron Chloride Bromide Iodide Sulfate

0.60 0.95 1.33 0.99 0.65 1.13 1.35 0.23 1.81 1.95 2.16 2.90

+1 +1 +1 +2 +2 +2 +2 +3 1 -1 -1 -2

3.82 3.58 3.31 4.12 4.28 4.12 4.04

3.32 3.30 3.31 3.79

0.60 0.95 1.33 0.50 0.33 0.57 0.68 0.08 1.81 1.95 2.16 1.45

1.67 1.05 .75 2.02 3.08 1.77 1.48

0.55 0.51 0.46 0.69

172

INORGANIC CONSTITUENTS AND PHYSICAL PROPERTIES

Density Equations that were developed for sea water can be applied to oilfield waters to obtain approximate values for engineering studies. The density values (ao) at OC and atmospheric pressure are related to the chlorinity (CZ) as follows:

-0.069 + 1.4708 C - 0.00157 C12 + 0.0000398 C13 = UO Z


where C = chlorinity (see Table 3 1 1 . Z .1) The density is very dependent upon temperature:

where D = a complex function of ao, and temperature and D values can be obtained from Knudsens Hydrographic Tables (Knudsen, 1901).

Vapor pressure
The relative lowering of the vapor pressure of oilfield water can be calculated with the following equation: Ap/po = 0.538 x

where p o = the vapor pressure of distilled water at the same temperature, and S = the salinity (see Table 3 1 1 (Kellog and Company, 1956, 1966, .1)

1968).
Boiling point
A first approximation of the boiling point elevation can be calculated from:

At = 0.0158S where S = the salinity. Freezing point An empirical equation which can be used t o estimate the freezing points is :

= 4.0086 - 0.064633 ((TO)

- 0.0001055 ( 0 0 ) ~

See Density for an explanation of terms.

PHYSICAL PROPERTIES

173

Viscosity The viscosity will increase with decreasing temperature and with increasing salinity. The viscosities of sodium chloride solutions of the same ionic strength can be used to estimate oilfield water vicosities. Osmotic pressure

A relationship between osmotic pressure ( P o ) and the depression of the freezing point at 0C is (in atmospheres):

The osmotic pressure at other temperatures can be estimated (Kellog and Company, 1956,1966,1968):

Po

(1+ 0.00367t)

Specific heat The values for the specific heat, c p , of oilfield waters can be approximated from the values of an equivalent sodium chloride solution. Thermal conductivity The thermal conductivity coefficient, A, can be calculated from thermal capacities because the ratio of thermal conductivities of two materials is the same as that of the thermal capacities of equal volumes. The values for X at various temperatures are available in a Saline Water Conversion Technical Data Book (Kellog and Company, 1956,1966,1968).

Surface tension
The surface tension of an oilfield water increases with decreasing temperature and with increasing salinity. An empirical formula which can be used t o calculate it is:

75.64 - 0.144t + 0.0399 Cl = surface tension (dynes/cm*)

Z .1; where t = temperature in Celcius, and C = the chlorinity (see Table 3 1 1 Kellog and Company, 1956,1966,1968).

174

INORGANIC CONSTITUENTS AND PHYSICAL PROPERTIES

References
Ahrens, L.H., 1965. Distribution o f the Elements in Our Planet. McGraw-Hill, New York, N.Y., 110 pp. Aumeras, M., 1927. Studies of ionic equilibria, 11. Equilibrium of calcium fluoride in dilute hydrochloric. J. Chem. Phys., 24:548-571. Baar, C.A., 1963. How to use the bromine test to determine the stratigraphical position in rock salt series. Neues Jahrb. Mineral. Geol. Palaontol., Monatsh., 7( 1):145-153. Baas Becking, L.G.M., Kaplan, I.R. and Moore, D., 1960. Limits of the natural environment in terms of pH and oxidation-reduction potentials. J. Geol., 68:243-284. Bordovskii, O.K., 1965. Source of organic matter in marine basins. Mar. Geol., 3( 1/2):5-31. Braitsch, 0. and Herrmann, A.G., 1963. Zur Geochemie des Broms in Salinaren Sedimenten, Teil I. Experimentelle Bestimmung der Br-Verteilung in verschiedenen naturlichen Salzsystemen. Geochim. Cosmochim. Acta, 27 :361-391. Brasted, R.C., 1957. The halogens. In: M.C. Sneed, J.L. Lewis and R.C. Brasted (Editors), Comprehensive Inorganic Chemistry, 3. Van Nostrand, New York, N.Y., 250 pp. Buckley, S.E., Hocott, C.R. and Taggart, Jr., M.S., 1958. Distribution of dissolved hydrocarbons in subsurface waters. In: L.G. Weeks (Editor), Habitat of Oil. American Association of Petroleum Geologists, Tulsa, Okla., pp.850-882. Burst, J.F., 1969. Diagenesis of Gulf Coast clayey sediments and its possible relation to petroleum migration. Bull. A m . Assoc. Pet. Geol., 53:73-93. Cloke, P.L., 1966. The geochemical application of E h - p H diagrams. J. Geolog. Ed., 14:140-1 48. Collins, A.G., 1969a. Chemistry of some Anadarko Basin brines containing high concentrations of iodide. Chem. Geol., 4:169-187. Collins, A.G., 1969b. Solubilities of some silicate minerals in saline waters. U.S. Off. Saline Water Res. Dev. Progr. Rep., No. 472, 27 pp. Collins, A.G., 1970. Geochemistry of some petroleum-associated waters from Louisiana. US.Bur. Min. Rep. Invest., No.7326, 31 pp. Collins, A.G., Castagno, J.L. and Marcy, V.M., 1969. Potentiometric determination of ammonium nitrogen in oilfield brines. Environ. Sci. Technol., 3:274-275. DallAglio, M., 1968. The abundance of mercury in 300 natural water samples from Tuscany and Latium (Central Italy). In: L.H. Ahrens (Editor), Origin and Distribution o f the Elements. Pergamon Press, Oxford, p.1065-1081. Dapples, E.C., 1959. The behavior of silica in diagenesis. In: H.A. Ireland (Editor), Silica in Sediments - SOC.Econ. Paleontol. Mineral., Spec. Publ., No.7, pp.36-55. Davis, J.W. and Collins, A.G., 1971. Solubility of barium and strontium sulfates in strong electrolyte solutions. Environ. Sci. Technol., 5:1039-1043. Fleischer, M., 1962. Recent estimates of the abundances of the elements in the earths crust. U.S. Geol. Surv. Circ., No.285, 9 pp. Garrels, R.M. and Christ, C.L., 1965. Solutions, Minerals, and Equilibria. Harper and Row, New York, N.Y., 450 pp. Goldberg, E.D., 1963. The oceans as a chemical system. In: M.N. Hill (Editor), Composition of Sea Water: Compamtive and Descriptive Oceanography. Interscience, New York, N.Y., 2:3-25. Goldschmidt, V.M., 1958. Geochemistry. Oxford University Press, London, 730 pp. Hem, J.D., 1960. Some chemical relationships among sulfur species and dissolved iron : chemistry of iron in natural water. U.S.Geol. Surv., Water Supply Paper, No, 145942, pp.57-73. Hem, J.D., 1970. Study and interpretation of the chemical characteristics of natural waters. U.S. Geol. Suru., Water Supply Paper, No. 1473, 363 pp. Hemley, J.J., 1953. A study of lead sulfide solubility and its relation to ore deposition. Econ. Geol., 48:113-137.

REFERENCES

175

Holser, W.T., 1966. Bromide geochemistry of salt rocks. In: Jon L. Rau (Editor), 2nd Symposium on Salt. The Northern Ohio Geological Society, Cleveland, Ohio, 1:248-275. Karasik, M.A., Goncharov, Yu. and Vasilevskaya, A.Ye., 1965. Mercury in waters and brines of the Permian salt deposits of Donbas. Geochem. Int., 2:82-87. Kazmina, T.I., 1951. The boron-chloride coefficient in oilfield waters. Dokl. Akad. Nauk S.S.S.R . , 77 :301. Kellog, M.W. and Company, 1956, 1966, 1968. Saline Water Conversion Technical Data Book and Supplements 1 and 2. (Catalog No. 1, 1.7712: Sa3, Supp.1, Supp.2, Supt. of Documents), U.S. Government Printing Office, Washington, D.C. Khitarov, N.I. and Pugin, V.A., 1966. Behavior of montmorillonite under elevated temperatures and pressures. Geochem. Int., 3:621-626. Knudsen, M., 1901. Hydrographic Tables. G.E.C. Gadd, Copenhagen, 6 3 pp. Larson, T.E. and King, R.M., 1954. Corrosion by water at low flow velocity. J. A m . Water Works Assoc., 46 :1-9. Latimer, W.M., 1952. Oxidation Potentials. Prentice-Hall, New York, N.Y., 2nd ed., 352 PP. Lyon, T.L. and Buckman, H.D., 1960. The Nature and Properties o f Soils. Macmillan, C., New York, N.Y., 567 pp. Marsden, S.S. and Kawai, K., 1965. SuiyBsei-Tennengasu - a special type of Japanese natural gas deposit. Bull. Am. Assoc. Petrol. Geol., 49:286-295. Mason, B., 1966. Principles of Geochemistry. John Wiley and Sons, New York, N.Y., 329 PP. Miholic, S., 1947. Ore deposits and geologic age. Econ. Geol., 42:713. Mitgarts, B.B., 1956. The composition of subterraneous waters in its value for petroleum prospecting, according to the data from the Fergana region. Vses. Nauch. Issled., Geol. Inst., Vopr. Neftepoiskovi Gidrogeol., 18:58-93 (in Russian). Moeller, T., 1954. Inorganic Chemistry. John Wiley and Sons, New York, N.Y., 966 pp. Mun, A.I. and Bazilevich, Z.A., 1962. Distribution of bromine in lacustrine bottom muds. Geochemistry, 2:199-205. Myagkov, V.F. and Burmistrov, D.F., 1964. Distribution of bromine in the carnallite beds of the Verkhnekamsk deposit. Geochem. Int., 1:701-703. Naumor, G.B., 1959. Transportation of uranium in hydrothermal solution as a carbonate. Geochemistry, 1:5-20. Odum, H.T., 1951. Notes on the strontium content of sea water, celestite radiolaria, and strontianite snail shells. Science, 114:211. Ostroff, A.G., 1965. Introduction t o Oilfield Technology. Prentice-Hall, Englewood Cliffs, N.J., 412 pp. Ovchinnikov, N.V., 1960. Patterns in the alteration of the chemical composition of subsurface waters of the Azqv-Kuban Trough and the distribution of iodine and bromine therein. Izv. Vyssh. Uchebn., Zaved., Geol. Razvedka, 3:134-138. Pirson, S.J., 1968. Redox log interprets reservoir potential. Oil Gas J., 66:69-75. Pirsson, L.V. and Knopf, A., 1947. Rocks and Rock Minerals. John Wiley and Sons, New York, N.Y. 3rd ed., 179 pp. Pourbaix, M.J.N., 1950. Thermodynamics o f Dilute Aqueous Solutions. Longmans, Green and Co., New York, N.Y., 151 pp. Rittenhouse, G., Fulton, Jr., R.B., Grabowski, R.J. and Bernard, J.L., 1969. Minor elements in oilfield waters. Chem. Geol., 4:189-209. Shaw, T.I., 1962. Physiology and biochemistry of algae. In: R.A. Lewin (Editor), Halogens. Academic Press, New York, N.Y., pp. 247-253. Shishkina, O.V. and Pavolva, G.A., 1965. Iodine distribution in marine and oceanic muds and in their pore fluids. Geochem. Int., 2:559-565. Sillbn, L.G., 1961. Physical chemistry of sea waters. A m . Assoc. Adv. Sci. Publ., No.67, pp. 549-581.

176

INORGANIC CONSTITUENTS AND PHYSICAL PROPERTIES

Sill&, L.G. and Martell, A.E., 1964. Stability constants of metal-ion complexes. Chem. Soc. Lond., Spec. Publ., No.17, 754 pp. Sugawara, K. and Terada, K., 1957. Iodine distribution in the western Pacific Ocean. J. Earth Sci., 5:81-102. Sugawara, K., Koyama, T. and Terada, K., 1956. Co-precipitation of iodide ions by some metallic oxides with special reference to iodide accumulation in bottom water layers and interstitial waters of muds in some Japanese lakes. UNESCO/NSRIC/65,presented at Int. Congr. Theor. Appl. Limnol., Helsinki, Finland, 1956, 1 pp. 1 Trelease, S.F., 1945. Selenium in soils, plants, and animals. Soil Sci., 160:125-131. Valyashko, M.G., 1956. Geochemistry of bromine in processes of halogenesis and use of bromine content as a genetic and prospecting criterion. Geochemistry, 6:33-49. White, D.E., 1957. Magmatic, connate and metamorphic waters. Bull., Geol. Soc. A m . , 68 :1669. White, D.E., 1965. Saline waters of sedimentary rocks. In: A. Young and J.E. Galley (Editors), Fluids in Subsurface Environments - A m . Assoc. Pet. Geol., Mem. 4 , pp.342-366. White, D.E., Hem, J.D. and Waring, G.A., 1963. Data of geochemistry. U.S. Geol. Surv. Prof. Paper, No.440-F, pp. FI-F67. Zobell, C.E., 1946. Studies on the redox potential of marine sediments. Bull. A m . Assoc. Pet. Geol., 30:477-513.

Chapter 6.

ORGANIC CONSTITUENTS IN SALINE WATERS

Water is a peculiar solvent and has been considered t o possess at ambient temperature a quasi-crystalline, open structure which will allow solute molecules to fill the space between the lattice points (Eley, 1939). As a nonpolar molecule dissolves in water at ambient temperature, the structure of the water in its immediate vicinity becomes more crystalline, or a microscopic iceberg surrounds the solute (Frank and Evans, 1945). Water also has been considered t o be an equilibrium mixture of an icelike and a close-packed structure, and with a molecule of gas as a solute, it reacts with the icelike structure filling one of the cavities to form a gas-hydrate and shifting the equilibrium from the close-packed structure to the icelike structure (Namoit, 1961). Another theory is that water is composed of clusters of highly hydrogenbonded molecules which are surrounded by a closely packed structure of monomeric water. These flickering clusters form and dissolve perpetually as a result of local energy fluctuations. Therefore, a water molecule can have a solute molecule as a neighbor along with its four H-bonded water neighbors. Interactions between the solute and water molecules will depress the energy level of the tetrabonded water molecule. However, large numbers of water molecules surround an unbonded molecule, and if it acquires a solute neighbor after the latter replaces a water molecule, the energy level is raised. Changes in the energy levels cause a shift of water molecules between various levels in accordance with the Boltzmann distribution law, giving an increase in the icelikeness and an increase in the clusters of water molecules near the surface of the solute molecule (Nemethy and Scheraga, 1962). When a hydrocarbon molecule transfers from the pure liquid to the solution hydrocarbon-water, interactions are established while hydrocarbonhydrocarbon interactions are broken. The amount, kind, and state (suspended, dissolved, or colloidal) of organic matter in petroleum-associated waters is important in determining the origin and migration of petroleum, and in problems concerning pollution of fresh waters by petroleumassociated waters. Probably the most plausible theory concerning the origin of petroleum is that it originated from organic constituents which are recognized as remnants or degradation products of living organisms of past ages; these organic source materials entered fine-grained aquatic sediments where biochemical and chemical conversions and fractionations occurred (Erdman, 1965). As increased sedimentation took place, the resulting overburden pressure and compaction caused the interstitial water, which con-

178

ORGANIC CONSTITUENTS IN SALINE WATERS

tained minute quantities of hydrocarbon, to be squeezed out of the consolidating sediments, and subsequently the oil was accumulated in sands and left behind in structural traps (Kidwell and Hunt, 1958). Obviously, the waters associated with petroleum play a very important part in the origin, migration, accumulation, and subsequent production of petroleum - the accumulation and production of petroleum being totally dependent upon hydraulic flow in response to geostatic and hydrostatic pressures. Consider briefly that 99% of the oil found and produced typically occurs within the pore spaces of sedimentary rocks (Hedberg, 1964).About 59% of the production comes from sandstone reservoirs, 40% from carbonates, and 1% from other types of rock. Petroleum in igneous and metamorphic rocks occurs primarily in fracture pore spaces and probably has migrated to these rocks from its place of origin. The solubilities of petroleum hydrocarbons in water increase with temperature and decrease as the salinity of the water increases. A temperature drop from 150 t o 25C reduces the solubility of petroleum in water by a factor of 4.5-20.5. Such a mechanism can account for the accumulation of petroleum because as upward moving subsurface waters containing dissolved hydrocarbons decrease in temperature and pressure and meet more saline waters, they will release dissolved hydrocarbons (Price, 1973). Information concerning dissolved organic matter in sea water was published as early as 1892 (Duursma, 1965). Palmitic acid, stearic acid, acrolein, and organic nitrogen were tentatively identified. The dissolved organic matter was found t o be about 2 mg/l in the open sea, increasing to about 15 mg/l in water taken near the coast of Greece, all of which was attributed t o saponification of the fats of dead organic organisms. Phytoplankton organisms comprise most living marine organic matter, 10% of which eventually becomes animal matter. The bulk of the organic particulate matter in the sea results from dead animal matter, but the dissolved organic constituents appear to be derived from dead phytoplankton and detritus rather than excretions from living cells. Decomposition of organic matter results primarily from the activities of heterotrophic bacteria. Organic matter decomposes more rapidly in a nearshore environment, where there is an abundance of such matter and bacteria, than in a deep-sea environment, where both the matter and bacteria are diluted. The dissolved organic matter can be classified into groups as follows: (1)nitrogen-free (for example, carbohydrates); (2)nitrogen-containing (for example, proteins); (3)lipids (for example, esters of fatty acids); and (4) complexes comprised of mixtures of the preceding three groups (for example, humic acids).
Nitrogen-free organic compounds

Many petroleum-associated waters contain methane; however, in Japan, a there is a type of natural-gas deposit called suiy6sei-tenynengasuy, dry gas,

NITROGEN-FREE ORGANIC COMPOUNDS

179

which is found dissolved in subsurface brines (Marsden and Kawai, 1965). The major reservoirs in which it is found are marine or lagoonal sedimentary basins with thick sediments and of wide areal extent. Some of the associated brines contain more than 80 mg/l of iodide, which is the only commercial source of iodine in Japan. Some of these brines also contain dissolved ethane, propane, isobutane, butane, isopentane, and pentane. An interesting note is that large Soviet deposits of natural gas in a solid state totaling about 15 trillion m3 were reported to the U.S.S.R. Committee for Inventions and Discoveries. According to U.S.S.R. investigators, molecules of ground water attract molecules of natural gas and convert them to a hydrate, which resembles silvery-grey ice, where the pressure is 250 atm and the temperature 25C or less. 1m3 of the hydrate contains up t o 200 m3 of natural gas. These solid hydrate deposits are found in permafrost zones at depths to 2,500 m. Because of the high electrical resistance, they are discoverable by geophysical methods. The hydrate can be converted to gas by sinking a well and reducing the pressure and/or pumping a catalyst such as methyl alcohol into it (Anonymous, 1970). The solubility of the hydrocarbons benzene, toluene, o-xylene, rn-xylene, p-xylene, naphthalene, biphenyl, diphenylmethane, and phenanthrene was found t o increase with increasing silver-ion concentration, indicating that a complex formed (Andrews and Keefer, 1949). Evidence slightly soluble 1-1 was obtained that two water-soluble complexes formed with silver and each aromatic hydrocarbon tested. Potassium nitrate causes a reduction in the solubility of aromatics in aqueous solutions (salting-out effect), but silver nitrate increases the solubility of toluene about 73% compared to its solubility in pure water. Apparently this effect with silver ions results from the formation of n-complexes between the benzene ring and the cation. Benzene hydrocarbons exhibit a minimum solubility in water near 18C which corresponds to a zero heat of solution. The actual volume occupied by a hydrocarbon with one benzene ring in water solution apparently influences its degree of solubility, and the larger the molecule the less soluble it is in water (Bohon and Claussen, 1951). However, naphthalene and biphenyl, which are larger in size and are multiring compounds, were 7 to 10 times more soluble, indicating that some property of the benzene ring may influence the solubility. I t was postulated that a positive heat of solution resulted in the heat of cavity formation, while a negative heat of solution resulted from the formation of icelike structures around the dissolved hydrocarbons and/or a n-electron complex of the aromatic nucleus where the n-electrons functioned as a base and the water as an acid. The heat of cavitation would predominate above 18C and would cancel the negative heat reaction at 18C ,and below 18C the negative heat would be larger. A study of the effect that the salts sodium fluoride, sodium chloride, lithium chloride, ammonium chloride, sodium iodide, cesium chloride, tetramethylammonium bromide, etc., have upon the activity coefficient of benzene in aqueous solutions indicates that the salting-out effect varies con-

180

ORGANIC CONSTITUENTS IN SALINE WATERS

siderably among the electrolytes (McDevit and Long, 1952). A limiting law for determining the influence of electrolytes on the activity of nonpolar solutes was developed, which related the magnitude of the salt effect t o the volume changes associated with salt and water mixing. Molecular hydrogen was found in oilfield waters in the Lower Volga region (Zinger, 1962). Up to 43%of the dissolved gas in these waters was hydrogen; other gases dissolved in the waters were methane, ethane, butane, pentane, carbon dioxide, nitrogen, helium, and argon. The pH of these waters was as low as 3.4, and the iron content was as high as 1,100 ppm. The solubility of methane increases with pressure and decreases with increased salt concentration at ambient temperature in NaC1-H2 0 and CaC12-H20 systems (Duffy et al., 1961). From the experimental data, it was estimated that 1 cubic foot of sedimentary rock with 20% porosity buried 300 m deep and saturated with a brine containing 50,000 ppm of NaCl could accommodate 0.3 mole of methane in solution. A gas-liquid partition chromatographic technique was used t o determine the solubilities of C1-C9 paraffin and branched-chain paraffins, four cycloparaffins, and five aromatic hydrocarbons in water (McAuliffe, 1963). Later this study was extended to seventeen paraffins, seventeen olefins, nine acetylenes, seven cycloparaffins, seven cycloolefins, and six aromatic hydrocarbons (McAuliffe, 1966). The data indicated that the solubilities of the hydrocarbons in water increased as unsaturate bonds were added t o the molecule, with ring closure, with addition of unsaturate bonds in the ring, and with addition of bonds which decreased the hydrocarbon molar volume. Branching increased the solubility in water for paraffin, olefin, and acetylene hydrocarbons but not for the cycloparaffin, cycloolefin, and aromatic hydrocarbons. Plots of the log of the solubility in water were a linear function of hydrocarbon molar volume for each homologous series of hydrocarbons. A capillary-cell method was used to measure the diffusion of methane, ethane, propane, and n-butane in water (Witherspoon and Saraf, 1964). At 25OC, the results indicated that the diffusion coefficients times los cm2/sec were 1.88, 1.52, 1.21, and 0.96, respectively, for methane, ethane, propane, and n-butane. The coefficients increased with higher temperatures. Near the critical solution temperature and about 300 atm, the solubility versus pressure curves for some hydrocarbon-water systems show a sharp maximum. However, pressure has a negative effect on solubility beyond this maximum, and a second two-phase region appears. The five binary hydrocarbon-water systems studied were benzene, n-heptane, n-pentane, 2-methyl-pentane, and toluene (Connolly, 1967). The accommodation of CI2-C& n-alkanes in distilled water was determined as a function of hydrocarbon supply, settling time, filtration poresize, and mode of introduction (Peake and Hodgson, 1967). Apparently it is possible to accommodate hydrocarbons in water at levels higher than solubil-

NITROGEN-FREE ORGANIC COMPOUNDS

181

ity levels, and such accommodation systems are stable for several days. range was found at Preferential accommodation of alkanes in the C16-C20 the expense of other 'alkanes with lower and higher carbon numbers. A gas chromatographic method for the determination of petrol in water was developed whereby the petrol was extracted from the water into nitrobenzene and the extract was analyzed using a column polyethylene glycol 1,500 on silanized Chromosorb W (Jeltes-and Veldink, 1967). The methods were sensitive t o 0.1 mg/l, and for concentrations > 0.5 mg/l the precision was about 5% for the major components. Low-molecular-weight hydrocarbons in the C1-C4range were detected in sea water. Generally the concentration tended to decrease with depth (Swinnerton and Linnenbom, 1967). Methane was the most abundant hydrocarbon found, but smaller amounts of ethane, ethylene, propane, propylene, n-butane, isobutane, and some butenes also were detected and measured. Hundreds of drill-stem samples of brine from water-bearing subsurface formations in the Gulf coastal area of the United States were analyzed to determine their amounts and kinds of hydrocarbons (Buckley et al., 1958). The chief constituent of the dissolved gases usually was methane, with measurable amounts of ethane, propane, and butane present. The concentration of the dissolved hydrocarbons generally increased with depth in a given formation and also increased basinward with regional and local variations. In close proximity t o some oilfields, the waters were enriched in dissolved hydrocarbons, and up to 14 standard cubic feet of dissolved gas per barrel of water was observed in some locations. The ratio of toluene t o benzene in 27 crude oils from various sources ranged from 2.0 t o 11.3. Toluene is less soluble than benzene in distilled water, where the ratio is about 0.3 (McAuliffe, 1966). A method of prospecting for petroleum, utilizing information concerning the amount of benzene dissolved in subsurface waters, was patented (Coggeshall and Hanson, 1956). Gas chromatographic methods proved t o be good for determining the amount of benzene and other hydrocarbons in the petroleumassociated waters (Zarrella et al., 1967). Collected information indicates that the concentration of benzene in petroleum-associated water varies with different types of hydrocarbon accumulations, that the benzene concentration decreases with increasing distance from the hydrocarbon accumulation, and that benzene is specific for detecting the occurrence of petroleum hydrocarbon accumulation in a given geologic horizon. A brine sample taken i from a horizon separated by 27 m of shale from an o l pool contained 0.02 ppm of benzene, indicating that low-permeability shale prevents movement of hydrocarbons. Chromatographic techniques were developed for the determination of sugars and phenols in sea waters and in sediments (Degens and Reuter, 1964). Biogeochemical differences were observed between the sugars in the sea and in the sediments.

182

ORGANIC CONSTITUENTS IN SALINE WATERS

Wilson e t al. (1970) found that ethylene, propylene, and carbon monoxide are produced in illuminated sea water to which dissolved phytoplankton was added. Higher saturated gaseous hydrocarbons and methane were not produced. Bonoli and Witherspoon (1968) measured the diffusion coefficients of methane, ethane, propane, n-butane, n-pentane, benzene, toluene, ethylbenzene, cyclopentane, methylcyclopentane, and cyclohexane in pure water at temperatures ranging from 2" to 60C using the capillary-cell method. The effect of sodium chloride was studied, and the largest decrease in diffusion coefficients was found for the paraffin hydrocarbons. They attributed the decrease to the effects of ions in water acting as structure breakers as well as obstacles to diffusion because of obstructions and hydrations. Hydrocarbons containing nitrogen Chromatographic techniques were developed for determining humic acids, amino acids, and indoles in saline waters and in sediments (Degens and Reuter, 1964). Arginine was found in the particulate matter in sea water and decreased in concentration with depth. Relatively abundant concentrations of ornithine, serine, and glycine were found in sea water. The total concentrations of amino acids found in some petroleumassociated waters ranged from 20 to 230 pg/l (Degens et al., 1964). In general, the amino acid content increased with salinity. Adjustment of the salinity of the brines t o that of modern sea water indicated a similarity between the amino acid spectra in the two. High concentrations of serine and the presence of threonine and phenylalanine and glutamic and aspartic acids were found in the petroleum-associated waters. It was postulated that the amino acids occurred in the petroleum waters in a combined state as nonproteinaceous acid complexes and that the solubility of these complexes probably is a function of salinity. This postulate was based on information which indicated that serine is thermally unstable. More recent information indicates that serine, lysine, threonine, glycocol, histine, isoleucine, and leucine are fairly stable up t o 180C (Califet and Louis, 1965). Liquid-exchange chromatography was used to determine the amounts of amino acids in some saline waters (Siege1 and Degens, 1966). The results indicated the bulk of the amino acids dissolved in the sea are tied up in complexes and are not in a free form. A study of the organic solutes in sea water led t o the conclusion that coprecipitation methods are the most versatile for their isolation (Chapman and Rae, 1967). Some of the organics that can be isolated by this method include glucose, glutamic acid, aspartic acid, citric acid, succinic acid, glycollate, glycine, and lysine. The percent of recovery of these solutes by this method varied from 16 t o 90%. The method involved the coprecipitation of these organic solutes with iron or copper. Most of the nitrogen in humic acid is located in the large and intermediate

FATTY ACIDS

183

size molecules, with the smallest molecules being practically nitrogen-free. This is attributed to the fact that the link involving nitrogen is more susceptible t o oxidation than the rest of the molecule. Amino acids can be released from humic acid by acid hydrolysis, alkaline hydrolysis, and reduction with sodium amalgam (Piper and Posner, 1968). A widely used procedure for concentrating and recovering trace organics is the carbon-adsorption method developed by Braus et al. (1951).A modification of this method by Robinson et al. (1967) allowed recovery of organics using three activated carbon filters in series, with the final two receiving acidified water. Krause (1962)investigated the decomposition of organic matter in natural waters and found that immediately after the death of an organism that amino acids and keto acids appeared in the water. After 24 hours of aerobic decomposition, a qualitative and quantitative maximum was reached by both groups, and the amino acids present were alanine, aspartic, glutamic, glycine, leucine, lysine, methionine, phenylalanine, serine, tyrosine, and raline; and the keto acids present were pyruvic, oc-ketoglutaric, oxaloacetic, and glyoxylic. After 10 days, the only acids that remained of the amino group were glutamic, glycine, lysine, and serine; and of the keto group, pyruvic and oc-ketoglutaric. Litchfield and Prescott (1970)analyzed sea water, and pond water, and spent algal media and found aspartic acid in all of the samples. Other amino acids frequently found were serine, glycine, alanine, and arginine. Techniques employed in the analysis were dansylation, extraction, and thin-layer chromatography . Fatty acids Ralston and Hoerr (1942) determined the solubilities of the normal saturated fatty acids from caproic to stearic acid, whose number of carbon atoms ranges from 6 to 18 in water, ethanol, acetone, 2-butanone, benzene, 0 0C and glacial acetic acid from ' to about 6'. In general, the solubilities increased with increasing temperature. Free fatty acids and hydroxyl ions form when soaps hydrolyze. The rate and percentage of hydrolysis is pH dependent, generally the potassium soaps are more hydrolyzed than the corresponding sodium soaps, and free fatty acid never separates as such from pure soap solutions unless reacted with an excess of acid such as carbon dioxide (McBain et al., 1948). Quantitative recovery of organic constituents from saline waters without alteration is difficult. Temperature and pressure changes, bacterial actions, adsorption, and the high inorganic/organic constituents ratio in most petroleum-associated waters are some reasons why quantitative recovery is difficult. Some of these factors apply also to sea waters, and Jeffery and Hood (1958) evaluated five methods which proved effective for isolation of portions of the soluble organic compounds in sea water. They were: (1)

184

ORGANIC CONSTITUENTS IN SALINE WATERS

dialysis or electrodialysis; (2) ion exchange;, (3) solvent extraction; (4) coprecipitation; and (5) carbon adsorption. Their results showed that the total organic material was most efficiently removed by electrodialysis or by coprecipitation with ferric hydroxide. A study of the differential uptake of organic compounds by montmorillonite and kaolinite revealed that montmorillonite adsorbed the compounds more efficiently than kaolinite. The compounds studied were aspartic acid, alanine, glucose, and sucrose (Williams, 1960). Approximately 13%of the aspartic acid was removed from solution by montmorillonite, while kaolinite removed only about 2%. The following saturated, monosaturated, and diunsaturated long-chain fatty acids were found in sea water: saturated Cl0, C12,CI4,C16,CIS,CzO, and CZ2; diunsaturated CIS; and monounsaturated CI6 and c 8 (Emery and 1 Koerner, 1961). Also isolated were CIS,C1,, and C19 acids which might or might not have been originally present. A gas chromatographic method was developed for the determination of trace amounts of the following fatty acids in water: n-valeric, isovaleric, n-butyric, isobutyric, propionic, and acetic (Emery and Koerner, 1961). The gas chromatograph was equipped with a flame ionization detector and a column of Tween 80 on Chromosorb W. The fatty acids lauric, myristic, palmitic, stearic, hyristoleic, palmitoleic, oleic, linoleic, and linolenic were identified in sea water using solvent extraction, esterification, and gas-liquid chromatography (Slowey et al., 1962). Samples of deep-sea water contained less unsaturated acids and shorter-chain acids than surface samples. Saturated straight-chain fatty acids were found in petroleum-associated waters from two reservoirs. The carbon numbers were CI4through C30.The same acids were identified in a shale-core sample from a petroleum reservoir. The even-numbered acids predominated over the odd-numbered acids in the amounts found in every case. The identification methods consisted of extraction by refluxing, esterification, gas chromatography, and mass spectrometry (Cooper, 1962). A gas chromatographic method capable of separating unesterified fatty were identified using acids was developed (Metcalfe, 1963). Acids up to CzO a thermal conductivity detector and a column composed of phosphoric acidtreated ethylene glycol succinate polyester on Chromosorb W. Bordovskii (1965) studied the sources of organic matter in marine basins, the sedimentation of organic matter in water, and the transformation of organic matter in sediments and its early diagenesis. He also pointed out that the organic matter in water is present in true solution, as colloidal organic detritus, and as live organisms in suspension. Bacteria play an important part in altering the composition of the organic material in the aqueous phase as well as in the sediments. Wangersky (1965) found that organic carbon was present in freshly distilled water and that it survived triple distillation and distillation from

NAPHTHENIC A N D HUMIC ACIDS

185

alkaline permanganate. He correlated this with algal growth in city reservoirs and suggested that the organic carbon content of distilled water must be considered by anyone growing organisms in distilled water. He also found that bubbling sea water caused organic compounds to form aggregates on the surface and further theorized that such reactions may be related to the origin of life. Kabot and Ettre (1963) developed gas chromatographic methods capable of determining free fatty acids. They analyzed different mixtures of the normal fatty acids using both packed and Golay columns in conjunction with a flame ionization detector. They concluded that the quantitative analysis of free fatty acids is possible. Naphthenic and humic acids Davis (1968) examined the organic fractions of artesian well waters from a Texas oil-bearing Eocene age aquifer, using infrared and chromatographic methods. He found that the water coproduced with oil contained 1,000 times more naphthenic acids than water located updip from the oil. He also found a phthalic acid ester dissolved in the petroleum-associated water but concluded that it may be common t o ground waters in general. Shaborova et al. (1961) state that the presence in subsurface waters of organic acids in the form of salts of various metals or in a free state indicates a current process of leaching of organic matter from the enclosing rock. The presence of organic acids in subsurface waters is one of the evidences for the existence in the earths crust of chemical processes of decomposition of preserved organic matter. In turn, the organic acids are broken down into simpler compounds by decarboxilization. It is known that decarboxilization of organic acids is accompanied by the formation of hydrocarbons. In nature, this process is a real geochemical factor. Consequently, the organic acids and their salts that are dissolved in subsurface waters can be regarded as one of the sources for the generation of hydrocarbons. Using a steam distillation method, organic acids were found in concentrations from 663 to 2,242 mg/l in subsurface waters taken from a Kazhim stratigraphic well. The average molecular weight of the acids was from 46 to 58, and the waters taken from Devonian age sediments contained higher concentrations of the acids than waters taken from Carboniferous age sediments. Lochte et al. (1949) analyzed waters produced with high-pressure gas wells and identified the following acids: acetic, propionic, isobutyric, n-butyric, isovaleric, n-valeric, n-hexanoic, and other C6isomers. Crude oils were treated with ammonia solution followed by electroprecipitation of the aqueous phase to remove naphthenic acids (Agaev, 1961). Further isolation of the naphthenic acids was accomplished by heating the aqueous phase to decompose the ammonium salts and remove ammonia and water. Oden (1919) recognized fulvic acid, humus acid, and hymatomelanic acid

186

ORGANIC CONSTITUENTS IN SALINE WATERS

in soils. Later, Page and Dutoit (1930) modified the name humus acid to humic acid. Sestini (1898) demonstrated that the humic acids are of complex composition and contain ethereal and anhydride components in addition to alkyl, hydroxyl, and ketonic groups. Burges (1960) suggested that humic acid is a single chemical substance or a group of similar substances, and that primarily it is nonnitrogenous. Steelink et al. (1960) fused soil humic acids and found the following degradation products: catechol, profocatechuic acid, and resorcinol. Steelink and Tollin (1962) determined the presence of two free-radical species in humic acid using an electron paramagnetic-resonance spectrometer. They believed that one could be a semiquinone radical and the other a quinhydrone radical. Fulvic acids, humic acids, and hymatomelanic acids have been found in natural waters (Wilson, 1959; Black and Christman, 1963; Packman, 1964). The brown color, characteristic of many natural waters, is attributed to complex organic compounds which probably are derived from water-soluble peptizable components of soil humus. A method that can be used t o determine the organic acids in petroleumassociated waters was published by the Natural Gasoline Association of America (1953). The water is treated with lime water to convert the organic acids to their calcium. salts, which are titrated with a standard mineral acid. Determination of oil in water The following method was developed by Nalco Chemical Company (1971) and is applicable t o waters and brines where the oily matter is hydrocarbons or hydrocarbon derivatives and all liquid or unctuous substances that have a boiling point above 90C and are extractable from waters or brines at pH 5.0 or lower using benzene, chloroform, or carbon tetrachloride. The sample is extracted with a fluorocarbon solvent which is evaporated off in a specially designed vessel and the residual oil measured volumetrically in a microsyringe.
Pear-shoped l o p , capacity opprox.17 m l , o f f s e t odditionol opening so th0t"popped" somple w i l l be r e t o i n e d

S y r i n g e , 5 0 0 kl, 10-pl divisions

Fig. 6.1. Microsyringe-evaporatingflask.

DETERMINATION OF OIL IN WATER

187

Apparatus. The necessary apparatus consists of: (1)Microsyringe - evaporating flask (see Fig. 6.1): this assembly consists of a single-neck flask of approximately 20 ml volume which tapers opposite and slightly offset to the neck into a microsyringe equipped with a gas-tight Teflon-tipped plunger and calibrated to measure 0-500 pl. (2) 1,000-ml pear-shaped separatory funnel. (3) Hotplate or hot-water bath: capable of being controlled in the range of 45O-55'~ at * ~ O C . (4) 500-ml Berzelius, tall form beaker. Reagents. The necessary reagents are: (1)50%hydrochloric acid solution, reagent-grade. (2) pH paper indicating strip or pH meter. (3) 1,1,2-trichlorotrifluoroethane(Freon TF) reagent-grade, purified, 48OC boiling point. Sampling. Collect a composite or spot sample representative of stream to be measured. Volume to be taken will be dependent on content of oily material and should be in the range of 1-5 liters. Sample should be caught in glass container. Procedure. Extraction: adjust pH of entire sample t o pH 5 or below using hydrochloric acid added in small increments. Thoroughly mix the sample and allow it t o stand 15 minutes. Measure the volume of entire sample and transfer to separatory funnel. Add portion of 1,1,2-trichlorotrifluoroethane extraction fluid (see Note 1) to sample container, thoroughly rinsing any adhering oil material. Add this and balance of fluorocarbon to separatory funnel. Shake thoroughly for 5 minutes and let stand t o separate layers. Draw off fluorocarbon layer into suitable beaker, filtering any entrained solids, if necessary, and warm gently to boiling point (see Note 2). Continue boiling until volume remaining can be contained in measuring flask. Transfer to measuring flask with fluorocarbon rinse of beaker, and immerse the flask and contents into 500-ml beaker partially filled with water and warmed to 65C on hotplate or in hot-water bath. Be sure open neck of flask is clear of upper edge of beaker (can be maintained by extension of syringe piston). Continue until volume is reduced to that of syringe volume. Draw fluid into syringe and increase heat slowly t o remove last traces of solvent, indicated by lack of bubbles forming in the syringe column. Measure amount of oil material in graduated syringe using graduations midway in syringe. Note 1: for single extractions fluorocarbon volume should be one-tenth of the original sample volume. In double extractions for better accuracy and reproducibility use two volumes of fluorocarbon, each onetwentieth of the original sample volume. Note 2: although fluorocarbon is essentially considered nontoxic, the

188

ORGANIC CONSTITUENTS IN SALINE WATERS

evaporation of the solvent should be done in a well-ventilated area or under an exhaust hood with adequate draft to handle high-density vapors.

Calculation :
c

1-11measured = mg/l oily matter mi sampIe

Organic acids in oilfield brines This method measures the bulk of the organic acid salts in oilfield waters.

Reagents and apparatus. Acetone; NaOH, 0.02N; HC1, 0.05N; acetic acid, 10 mg/l; a hotplate; and a standard pH meter. Procedure. Pipet 25 ml of filtered brine into a 250-ml beaker; add 25 ml of acetone and adjust the pH t o precisely 6. Titrate the sample to a pH of 3.5, and record the amount of 0.05N HC1, used in the adjustment from pH 6 to 3.5. Boil the solution for 5 minutes. Cool and titrate back to pH 6 with 0.02N NaOH. Make a blank determination for NaOH and HC1. To calculate, subtract blank from NaOH and HC1 titrations. Calculation:
(ml HC1 x HC1 N ) - (ml NaOH x NaOH N ) x 60,030 = mg/l organic acids ml sample as acetic acids. References
Agaev, A.A., 1961. Separation of ammonium salts of naphthenic acids from crude oil in an electric field. Izv. Vyssh. Uchebn. Zaved., N e f t Gaz, 4:95-98 (in Russian). Andrews, L.J. and Keefer, R.M., 1949. Cation complexes of compounds containing carbon-carbon double bonds, 11. The solubility of cuprous chloride in aqueous maleic acid solutions. J. A m, Chem. Soc., 71 :2379-2380. Anonymous, 1970. Solid natural gas discovered. Ind. Res., 12:70. Black, A.P. and Christman, R.F., 1963. Chemical characteristics of fulvic acids. J. A m . Water Works Assoc., 55:897-912. Bohon, R.L. and Claussen, W.F., 1951. The solubility of aromatic hydrocarbons in water. J. A m . Chem. Soc., 73:1571-1578. Bonoli, L. and Witherspoon, P.A., 1968. Diffusion of paraffin, cycloparaffin, and aromatic hydrocarbons in water and some effects of salt concentration. In: Advances in Organic Geochemistry - Proc. 4th Int. Meet. Org. Geochem., Amsterdam. Pergamon Press, New York, N.Y., 9:16-18. Bordovskii, O.K., 1965. Accumulation and transformation of organic substance in marine sediments. Mar. Geol., 3:3-114. Braus, H.,Middleton, F.M. and Walton, G., 1951. Organic chemical compounds in raw and filtered surface waters. Anal. Chem., 23:1160-1164.

REFERENCES

189

Buckley, S.E., Hocott, C.R. and Taggart, Jr., M.S., 1958. Distribution of dissolved hydrocarbons-in subsurface waters. In: L.G. Weeks (Editor), Habitat o f Oil. American Association of Petroleum Geologists, Tulsa, Okla., pp. 850-882. Burges, A., 1960. The nature and distribution of humic acid. Sci R o c . R . Dublin SOC. Ser. A , 4:53-51. Califet, Y. and Louis, M., 1965. Contribution to the knowledge concerning the stability of amino acids contained in sedimentary rocks. Compt. Rend., Acad. Sci. Fr., 261 :3645-3646. Chapman, G. and Rae, A.C., 1967. Isolation of organic solutes from sea water by coprecipitation. Nature, 214:627-628. Coggeshall, N.D. and Hanson, W.E., 1956. Method of geochemical prospecting. US. Patent, No. 2,767,320. Connolly, J.F., 1967. Solubility of hydrocarbons in water near the critical solution temperatures. J. Chem. Eng. Data, 11:13-16. Cooper, J.E., 1962. Fatty acids in recent and ancient sediments and petroleum reservoir waters. Nature, 193:744-746. Davis, J.B., 1968. Distribution of naphthenic acids in an oil-bearing aquifer. Presented at Annual Geol. SOC. A m . and Assoc. SOC. Meet., Mexico, D.F., November, 1968. Program, p. 7 1. Degens, E.T. and Reuter, J.H., 1964. Analytical techniques in the field of organic geochemistry. In: U. Colombo and G.F. Hobson (Editors), Advances in Organic Geochemistry. Pergamon Press, New York, N.Y., pp.377-402. Degens, E.T., Hunt, J.M., Reuter, J.H. and Reed, W.E., 1964. Data on the distribution of amino acids and oxygen isotopes in petroleum brine waters of various geologic ages. Sedimentology, 3 :199-225. Duffy, J.R., Smith, N.O. and Nagy, B., 1961. Solubility of natural gases in aqueous salt solutions, I. Liquidus surfaces in the System CH4-Hz O-NaC1-CaCl2 at room temperatures and at pressures below 1,000 psia. Geochim. Cosmochim. Acta, 24:23-31. Duursma, E.K., 1965. The dissolved organic constituents of sea water. In: J.P. Riley and G. Skirrow (Editors), Chemical Oceanography. Academic Press, New York, N.Y., 1:433-475. Eley, D.D., 1939. Solubility of gases, 1. Inert gases in water. Trans. Faraday SOC., 35 :1281-1293. Emery, E.M. and Koerner, W.E., 1961. Gas chromatographic determination o trace f amounts of the lower fatty acids in water. Anal. Chem., 33:146-147. Erdman, J.G., 1965. Petroleum - its origin in the earth. In: A. Young and J.E. Galley (Editors), Fluids in Subsurface Environments - A m . Assoc. Pet. Geol., Mem. 4 , pp.20-52. Frank, H.S. and Evans, M.E., 1945. Free volume and entropy in condensed systems. J. Chem. Phys., 13:507-532. Hedberg, H.G., 1964. Geologic Hspects of origin of petroleum. Bull. A m . Assoc. Pet. Geol., 48 (11):1755-1803. Jeffery, L.M. and Hood, D.W., 1958. Organic matter in sea water; an evaluation of various methods for isolation. J. Mar. Res., 17:247-271. Jeltes, R. and Veldink, R., 1967. The gas chromatographic determination of petrol in water. J. Chromatogr., 27:242-245. Kabot, F.J. and Ettre, L.S., 1963. Gas chromatography of free fatty acids, 111. Quantitative aspects. J. Gas Chromatogr., 1:7-10. Kidwell, A.L. and Hunt, J.M., 1958. Migration of oil in sediments. In: L.G. Weeks (Editor), Habitat of Oil. American Association of Petroleum Geologists, Tulsa, Okla., pp.7 90-8 17. Krause, H.R., 1962. Investigation of the decomposition of organic matter in natural waters. F A 0 Fish. Biol. Rep., No.34, 1 4 pp.

190

ORGANIC CONSTITUENTS IN SALINE WATERS

Litchfield, C.D. and Prescott, J.M., 1970. Analysis by dansylation of amino acids dissolved in marine and freshwaters. Limnol. Oceanogr., 15:250-256. Lochte, H.L., Burnam, C.W. and Meyer, H.W.H., 1949. Organic acids produced by high pressure gas wells The Pet. Eng., 21 :C34-C40. Marsden, S.S. and Kawai, K., 1965. "Suiy6sei-Ten'nengasu" - a special type of Japanese natural gas deposit. Bull. A m . Assoc. Pet. Geol., 49:286-295. McAuliffe, C., 1963.Solubility in water of C l - C g hydrocarbons. Nature, 200:1092-1093. McAuliffe, C., 1966. Solubility in water of paraffin, cycloparaffin, olefin, acetylene, cycloolefin, and aromatic hydrocarbons. J. Phys. Chem., 70:1267-1275. McBain, J.W., Laurent, P. and John, L.M., 1948. The hydrolysis of soap solutions, 111. Values of pH and the absence of fatty acid as free liquid or solid. J. A m . Oil Chem. SOC.,25:77-84. McDevit, W.F. and Long, F.A., 1952. The activity coefficient of benzene in aqueous salt solutions. J. A m . C h e m SOC.,74:1773-1777. Metcalfe, L.D., 1963. The gas chromatography of fatty acids and related long chain compounds on phosphoric acid treated columns. J. Gas Chromatogr., 1 :7-11. Nalco Chemical Company, 1971. Analytical Method f o r the Determination o f Oil in Water. Unpublished method (permission obtained through H.C. Noe). Namoit, A.Y., 1961.Solubility of nonpolar gases in water. Zh. Strukt. Khim., 2:408-417. Natural Gasoline Association of America, 1953. Condensate Well Corrosion. NGAA, Tulsa, Okla, 203 pp. Nemethy, G. and Scheraga, A.A., 1962. Structure of water and hydrophobic bonding in proteins, 1. A model for the thermodynamic properties of liquid water. J. Chem. Phys., 36:3382-3400. Oden, S., 1919. Humic acids, studies in their chemistry, physics, and soil science. Kolloidchem. Beih., 11:75-260. Packman, R.F., 1964. Studies of organic color in natural water. h o c . SOC. Water Treat. Exam., 13:316-329. Page, H.J. and Dutoit, M.M.S., 1930. Studies on the carbon and nitrogen cycles in the soil, 111. The formation of natural humic matter. J. Agric. Sci., 20:478-488. Peake, E. and Hodgson, G.W., 1967. Alkanes in aqueous systems, 11. The accommodation 44:696-702. of C12-C36n-alkanes in distilled water. J. A m . Oil Chem. SOC., Piper, T.J. and Posner, A.M., 1968. On the amino acids found in humic acid. Soil Sci., 106 :188-192. Price, L.C., 1973. Solubility of petroleum in water as function of temperature and salinity and its significance in primary migration. Bull. A m . Assoc. Pet. Geol., 57 :Sol. Ralston, A.W. and Hoerr, C.W., 1942.The solubilities of the normal saturated fatty acids. J. Org. Chem., 7:546-555. Robinson, L.R., O'Conner, J.T. and Engelbrecht, R.S., 1967.Organic materials in Illinois groundwaters. J. A m . Water Works Assoc., 59:227-236. Sestini, F., 1898. The nitrogen content associated with humic acids in soil. Landw. Vers. Sta., 51:153-160. Shaborova, N.T., Tunyak, A.P. and Nektarova, M.B., 1961. Study of organic acids in underground waters. Geol. Nefti Gaza, 5:50-51 (in Russian). Siegel, A. and Degens, E.T., 1966. Concentration of dissolved amino acids from saline waters by ligand-exchange chromatography. Science, 151 :1098-1101. Slowey, J.F., Jeffery, L.M. and Hood, D.W., 1962.The fatty-acid content of ocean water. Geochim Cosmochim. Acta, 26:607-616. Steelink, C. and Tollin, G., 1962. Humic acid contains stable free radicals. Chem. Eng. News, 40:53-54. Steelink, C., Berry, J.W., Ho, A. and Nordby, H.E., 1960.Alkaline degradation products of soil humic acid. Sci. Proc. R. Dublin SOC.,Ser. A , 1(4):59-67. Swinnerton, J.W. and Linnenbom, V.J., 1967. Gaseous hydrocarbons in sea water: determination. Science, 156 :1119-1120.

REFERENCES

191

Wangersky, P.J., 1965.The organic chemistry of sea water. A m . Sci., 53~358-374. Williams, P.M., 1960. Organic Acids Found in Pacific Ocean Waters. Ph.D. Dissertation, Univ. of Calif., Los Angeles, Calif., 74 pp. Wilson, A.L., 1959. Determination of fulvic acids in water. J. Appl. Chem., 9:501-510. f Wilson, D.F., Swinnerton, J.W. and Lamontagne, R.A., 1970. Production o carbon monoxide and gaseous hydrocarbons in sea water: relation to dissolved organic carbon. Science, 168:1577 1 579. Witherspoon, P.A. and Saraf, D.N., 1964. Diffusion of methane, ethane, propane, and n-butane in water. Presented at A m . Chem. SOC.,Div. Pet. Chem. Meet., Chicago, Ill., 1964, Preprints, pp.303-311. Zarrella, W.M., Mousseau, R.J., Coggeshall, N.D., Norris, M.S. and Schrayer, G.J., 1967. Analysis and significance of hydrocarbons in subsurface brines. Geochim. Cosmochim. Acta, 31 :1155-1166. Zinger, AS., 1962. Molecular hydrogen in gas dissolved in waters of oil-gas fields, Lower Volga region. Geochemistry, 10 :lo15-1023.

This Page Intentionally Left Blank

Chapter 7 ORIGIN OF OILFIELD WATERS .

The five spheres of the earth are the lithosphere (rocks), the pedosphere (soils, till, and other surficial materials), the hydrosphere (natural waters), the atmosphere (gases), and the biosphere (living organisms). Oilfield waters are a part of the hydrosphere, and petroleum is a product of the biosphere. The total volume of water in the hydrosphere is about 1,338 x 10'' liters, and about 8.4 x 10" liters is ground water (Skinner, 1969). Most of the water, 1,300 x 10" liters, is in the oceans. Less than 50% of the total ground water is in strata below 1 km. The total amount of water in sedimentary rocks and associated with liquid and gaseous hydrocarbons is less than 1.3 x 1 O I 8 liters. All of the petroleum recovered t o date has been taken from oil wells drilled into the upper 8 km of the earth's crust. The average thickness of the

0 km I 7 km 400km

1,000 k m

2.900km

5,000km 6,371 k m

Fig. 7.1. Regions of the interior of the earth.

194

ORIGIN OF OILFIELD WATERS

earths crust is about 1 7 km, ranging from 5 km under the oceans t o about 35 km under the continents (Clark and Ringwood, 1964). Fig. 7.1 illustrates the various regions of the interior of the earth, with the distance from the surface of the crust to the center of the inner core being 6,731 km. In this discussion we are concerned only with the crust to a distance of 0.08% of the depth to the center of the earth. Hydrocarbons are believed t o have originated from organic material in sedimentary material which was produced by weathering and erosion of the earths surface. This eroded material is carried away by water, ice, or wind and redeposited, ultimately forming sedimentary rocks. The major sedimentary, minerals are clays, quartz, calcite, gypsum, anhydrite, dolomite, and haiite. Most of the large bodies of sedimentary rocks were formed in marine environments; smaller sedimentary deposits formed in lakebeds and river floodplains. Definitions of some water terms

Meteoric water. White (1957) defined it as water that was recently involved
in atmosphere circulation and further that the age of meteoric groundwater is slight when compared with the age of the enclosing rocks and is not more than a small part of a geologic period.

Sea water. The composition of sea water may vary somewhat, but in general will have a composition relative t o the following (in mg/l): chloride 19,375, bromide - 67, sulfate - 2,712, potassium - 387, sodium - 10,760, magnesium - 1,294, calcium - 413, and strontium - 8.
Table 7.1 (Anonymous, 1964) gives a more comprehensive picture of the constituents found in sea water. The analyses given in Table 7.1. are in parts per million.

Interstitial water. Interstitial water is the water contained in the small pores
of spaces between the minute grains or units of rock. Interstitial waters are: (1) syngenetic (formed at the same time as the enclosing rocks); or (2) epigenetic (originated by subsequent infiltration into rocks).

Connate water. The term connate implies born, produced, or originated together, connascent. Therefore, connate water probably should be considered t o be an interstitial water of syngenetic origin. White (1957) called connate water of this definition a fossil water, i.e., water that has been out of contact with the atmosphere for at least a large part of a geologic period. As White (1957) pointed out the implication that connate waters are only those born with the enclosing rocks is an undesirable restriction. Diagenetic water. Diagenetic waters are those waters that have changed
chemically and physically, both before, during, and after sediment consolida-

SEDIMENTARY ROCKS TABLE 7.1 Average composition of sea water Element Chlorine Sodium Magnesium Sulfur Calcium Potassium Bromine Carbon Oxygen Strontium Boron Silicon Fluorine Nitrogen Argon Lithium Rubidium Phosphorus Iodine Barium Indium Aluminum Iron Amount (PPm) Element Zinc Molybdenium Selenium Copper Arsenic Tin Lead Uranium Vanadium Manganese Titanium Thorium Cobalt Nickel Gallium Cesium Antimony Cerium Yttrium Neon Krypton Lanthanum Silver Bismuth Cadmium Amount (PPm 1 Element Tungsten Germanium Xenon Chromium Beryllium Scandium Mercury Niobium Thallium Helium Gold Praseodymium Gadolinium Dysprosium Erbium Ytterbium Samarium Holmium E uro pi um Thulium Lutetium Radium Protactinium Radon Amount (PPm)
1 1 1 5x 5 4x 3x 1 1 5x 4x

195

18,980 10,560 1,270 880 400 380 65 28 8 8 4.8 3.0 1.3 0.8 0.6 0.2

0.12 0.07 0.05 0.03 0.02 0.01 0.01

0.01 0.01 0.004 0.003 0.003 0.003 0.003 0.003 0.002 0.002 0.001 0.0007 0.0005 0.0005 0.0005 0.0005 0.0005 0.0004 0.0003 0.0003 0.0003 0.0003 0.0003 0.0002 0.0001

lo4 lo4 lo4 1010-~ 1010-~ 1" 0 lod

2x 2 2 2 2 2 8x 4x 4x 4x 3 x lo-" 2 x lo-'* 9

tion. Some of the reactions that occur in or t o diagenetic waters include bacterial, ion exchange, replacement (dolomitization), infiltration by permeation, and membrane filtration.

Formation water. Formation water as here defined is water that naturally occurs in the rocks and is present in them immediately before drilling (Case, 1955). Juvenile water. Water that is in primary magma or derived from primary magma is juvenile water (White, 1957).
Sedimentary rocks At least a portion of the water found in petroleum reservoirs consisting of sedimentary rocks w a s deposited with the sediments before they were transformed into rock. As the sediment compacted to form rock, the composition

196
TABLE 7 1 .1

ORIGIN OF OILFIELD WATERS

Average composition o igneous and some types of sedimentary rocks (ppm) f Element Igneous rocks Sedimentary rocks resistates Si A 1 Fe Ca Na K Mg Ti P Mn F S C hydrolyzates precipitates Evaporites (halite)

c 1

co
Pb Th

Rb Sr Ba Cr Zn Ni cu Li N Sn

cs
U

Be As B Br I Cd Se

H g
Ra

277,200 367,500 272,800 24,200 90 81,300 25,300 81,900 4,300 20 11 50,000 9,900 47,300 4,000 36,300 39,500 22,300 304,500 930 28,300 3,300 9,700 370 325,000 25,900 11,000 27,000 2,700 800 20,900 7,100 14,800 47,700 460 4,300 4,400 960 0.8 1,180 350 740 175 trace 620 385 1 1,000 600-900 510 250 20 520 2,800 2,600 1,100 770 320 13,800 15,300 113,500 70 trace 314 200 586,000 0 310 273 300 300 < 26 170 425-765 250 170 460 120 2 200 68-200 410-680 2 132 < 20 200-1,000 < 50 1 80 28 24 0 0 70 3 192 20.2 65 17 46 < 26 46 40 40 0 8 0 23 16 20 20 51 0 2 11.5 (? ) 6.1 10.1 1.1 < 0.2 7 12 <4 0 6 (3 5 - 5 4 1.2 1.2 1.3 0.01 3 9-31 310 3 <2 1.62 < 0.2 60 0.07-0.55 < 2 0.3 0.15 0 0.3 0.09 06 . < 0.1 < 0.5 0.08-0.5 01 . 0.3 0.03 13 x . 0.7 x 1.08 x lod 0.42 x lod -

of the interstitial water changed because of reactions with the rock. A simplistic view of sedimentary rocks and their relation to oilfield waters should include consideration of weathering; erosion; transportation mechanisms; sorting of weathered products; depositional environments of clastics, carbonates, evaporites, organic matter, and silica; sediment compactions; sediment diagenesis; and petroleum and natural gas.

SEDIMENTARY ROCKS

197

The volume of the earth is about 1,100 billion km3 and the volume of the oceans is about 1.3 .billion km3; however, the oceans with an area of 360 million km2 cover 70% of the surface of the earth. The average composition of some of the igneous and sedimentary rocks of the earths crust is shown in Table 7.11, which was taken from Clarke and Washington (1924) and Rankama and Sahama (1950). The resistate rocks referred to in Table 7.11 are composed of residues not chemically decomposed in the weathering of the parent rocks. Hydrolyzate rocks are the insoluble products formed by chemical reactions during weathering of parent rocks. Precipitate rocks are those formed by chemical precipitation of minerals from aqueous solution. Evaporite rocks are marine evaporites which were produced when the water in which they were dissolved was evaporated. Sedimentary rocks comprise about 5% of the lithosphere, while the igneous rocks form 95%. The three main types of sedimentary rocks are shale, sandstone, and iimestone, and their relative abundance determined from geochemical data ranges from 70 t o 83% shale, from 8 to 16%sandstone; and from 5 t o 14% limestone (Pettijohn, 1957). Levorsen (1966) noted that oil and gas are found in reservoir rocks consisting primarily of sandstones, limestones, and dolomites.

Weathering
Weathering is a most important factor in producing the source material for the creation of sedimentary rocks. Processes that cause weathering are chemical, physical, and biological (Ross, 1943). The weathering of rock by physical methods includes temperature changes brought about by climate changes. Examples are the breaking of rock by thermal expansion (heat), the breaking of rock by the expansion of freezing water in the pores or cracks, or the mechanical breaking of rock as a glacier moves over it. Breaking the rock causes the surface area to become larger without significantly changing the chemical composition. Biological weathering includes the cracking of rock as a result of plant roots and the action of acids derived from plants, animals, and bacteria. The biotic factor includes bacteria, algae, protista, protozoa, higher animals and plants, during both life and subsequent necrotic decomposition which furnish Ht ions, colloids, complexing agents, and dispersants. Chemical weathering involves the action of water upon the parent rock and upon the products of physical and biological weathering. In chemical weathering the composition of the source rock is changed by solution, hydrolysis, oxidation, and reduction reactions. The Ht ion when concentrated in aqueous solution is a very important energy factor because it will cause rapid chemical reactions with parent rocks. The redox potential influences the rate of removal of elements, such as iron and manganese from the parent rock, and if it is a reducing potential, these elements are more likely to remain in solution after solubilization.

198

ORIGIN OF OILFIELD WATERS

Erosion
Erosion is the opposite of deposition (the processes are reversible), but erosion must occur before deposition can proceed. The products of weathering are eroded and transported to a new location by the action of water and wind. The water serves to transport the majority of these products, and it can transport them by dissolution, suspension, or pushing of larger particles.

Transpor ta t ion mechanisms


Both wind and water can transport the products of weathering, however, this discussion will consider only water. The transport mechanisms considered are chemistry, physics, and hydraulics. Perhaps the primary solvents of weathered products are carbonated water, organic acids, and sulfate solutions. Elements that dissolve readily in carbonated waters are lithium, sodium, potassium, magnesium, calcium, strontium, iron, manganese, phosphorus, and others. The organic acids will dissolve iron and manganese, while sulfate solutions will dissolve copper, iron, and manganese compounds. The chemistry of the water is a prime factor in the dissolution of the rock; if the pH is acidic, the transition group metals are more likely to dissolve, while if it is basic, elements such as silica are more likely to dissolve. Salts of the alkali and alkaline earth metals will dissolve if the pH is either acidic or basic; however, if the pH is above 10, some of the alkaline earths such as magnesium will precipitate. The pH of the water is influenced by the dissolution of carbon dioxide. For example, as carbon dioxide dissolves in water, the pH will change. The pH of pure water in equilibrium with carbon dioxide can be calculated and is 5.65 (Hem, 1970). The pH is calculated using the mass-law equations in which the activity of water is unity in dilute solution, and h , = constant equal to the product of the activities of H+ and OH-. Introduction of another phase such as calcite into the water carbon dioxide system will change the pH. Garrels and Christ (1965) calculated that such a system in equilibrium with the atmosphere will have a pH of 8.4. Additional ions such as those found in ocean water will produce other pH values. For example, if the system is ocean water in equilibrium with carbon dioxide, the pH at each equilibrium step is approximately: H20+C02 H2 co3 HC03+H+

* A

=+

H2C03 HC03-+H+ 2H++C03-2

7PH 5) (PH 6-31 (pH 10.3)

Some chemicals when dissolved in water act as buffers, where a buffer is defined as something that .produces an effect which inhibits a large pH change when an acid or a base is added t o the water. Therefore, as a water

SEDIMENTARY ROCKS

199

becomes more concentrated with certain dissolved solids, the pH can be more stable because of buffer effects. A solution containing an activity of orthosilicic acid greater than its solubility product ( K ) is oversaturated and SiOz will precipitate. If the activity s.p of orthosilicic acid is less than its K s p , SiQz can dissolve until (H4Si0,) = Ksp. The same applies for other constituents. Complex compounds and ion pairs are important chemical transport mechanisms. An example of a complex compound is a metal chelate such as copper chelated with ethylenediamine tetraacetate, with the interacting ligands immediately adjacent to the metal cation. An ion pair is formed if the coordinated water is retained in forming the complex (Stumm and Morgan, 1970). The redox potential (Eh) also influences the transport of metals in solution. Most surface waters in contact with the atmosphere will have an oxidizing potential, Fig. 7.2. However, many subsurface waters have low redox potentials, and metals such as iron and manganese are transported as reduced species.

600 400r
W

I I
I
Acid Fe++

/M,odern

sea water

200

I
I

Alkaline

-400 -600

-eooo

6
PH

1 0

H2, 12

14

Fig. 7.2. Diagram of some EhlpH relationships.

200

ORIGIN OF OILFIELD WATERS

Sorting of weathered products


Products of weathering are transported by water by rolling along the streambed, by suspension of the smaller particles, and by solution of soluble components. As the larger rocks and pebbles roll along the streambed, they are abraded by bumping against each other and by abrasive action with the rocks in the streambed. The size of the clastics, which are detritus transported mechanically t o the point of sedimentation and portions solubilized by water before sedimentation, decreases in the downcurrent direction (Pettijohn, 1957), and this change in grain size is primarily a sorting effect. The effect is noted in both fluvial and marine deposits.
Mobile belt
f

Geosynclinal trough For eland (stable)

Borderland

r
Shelf area

Fig. 7.3. Idealized depositional basin. (After Moore, 1969.)

Knowledge of the sorting of the clastics is used in reconstructing the ancient environment (Visher, 1965).This knowledge can be applied to exploration for petroleum and other valuable minerals. A simplistic deposiil tional basin is shown in Fig. 7.3; the deposited clastics wl be found on the borderland side of the basin and not on the foreland side.

Depositional environments of clastics


Depositional environments of the clastics include eolian, fluvial, regressive marine, transgressive marine, deltaic, bathyal-abyssal, and lacustrine. Eolian deposits are sands that are drifted and arranged by currents of air or wind. Fluvial deposits are those related t o streams, rivers, and ponds. Water in these environments usually contains less than 10,000 mg/l of dissolved solids. Regressive marine deposits are land-derived sediments that are transported seaward and settle in the ocean. The salinity of the water transporting these sediments will vary, it is fresher at its source and becomes more brackish as it nears the sea. The dissolved solids in contemporary sea water are about 35,000 mg/l, while some estuary waters contain about 20,000 mg/l of dissolved solids. Transgressive marine deposits usually are small in volume compared t o fluvial and regressive-marine deposits. Such deposits are formed mainly by

SEDIMENTARY ROCKS

201

erosion and redeposition of sediment deposits. The salinity of the sea may change somewhat during transgression but probably not much. Deltaic deposits result from a combination of environmental factors and are related t o both fluvial and regressive marine processes. During flood times the rivers transport tremendous volumes. of material, both clastic and organic, into delta areas. Deltaic deposition is a very important factor in the formation of petroleum because of the tremendous amount of organic material deposited. Bathyal-abyssal deposits are formed in deep-water areas in the sea, and turbidity currents are responsible for most of the clastic deposition (Emery, 1960). Lacustrine deposits are those that are formed in lakes. If the lake is a fresh-water lake, the dissolved solids may be less than 1,000 mg/l; in a salt-water lake, the dissolved solids may be greater than 35,000 mg/l. Consider a simplistic sedimentation area where the borderland area is the prime source of sediments. The coarse- t o fine-grained clastics which are weathering products of the high-mountains borderland are deposited near their source. The clastics are detritus transported mechanically to the point of sedimentation and are not solubilized by the water before deposition. Primarily they are the sandstones and shales (clays). They will not be found on the for.eland side of a depositional basin. Clay deposition can be detrital or authigenic; illite often is detrital. There are at least two dozen clay minerals, many of which occur in very minute grains and most of which cannot be resolved by high-power petrographic microscopes. The electron microscope, X-ray diffraction, and differential thermal analysis are used to determine the type of clay. The clays are very important in relation t o petroleum and gas because they are the major component in the shales from which petroleum and gas are generated. The clays also possess base exchange properties which will react with constituents in water and petroleum. The detrital clays settle in low-energy waters and they settle more rapidly from a saline water than from a fresh water.

Depositional environments of the carbonates


Limestones and dolomites are the dominant carbonate reservoir rocks, while the sandstones are the dominant clastic reservoir rocks (Ham, 1962). The carbonates were precipitated at the place where the rocks first formed, while clastics were primarily transported grains. Plumley et al. (1962) classified the carbonates according t o an energy index of the water from which they precipitated. They divided them into five types. Type I is deposited in quiet water; it consists of calcite, 15-5096 clay, and < 5% detrital quartz. Type I1 is deposited in intermittently agitated water and consists of calcite, < 25% clay, and < 50%detrital quartz. Type I11 is deposited in slightly agitated water and it consists of calcite with up to 50%detrital quartz. Type IV is deposited in moderately agitated water

202

ORIGIN OF OILFIELD WATERS

and consists of calcite with up to 50% detrital quartz. Types I11 and IV are similar in percentage of minerals; however, they are further differentiated according t o grain size, sorting, roundness, and fossils, as are the other types. Type V is deposited in strongly agitated (high-energy) water and it consists of calcite, < 5% clay, and < 25% detrital quartz. The calcium in the carbonates is released during rock weathering and goes into solution as bicarbonate. The solubility of calcium carbonate in water is also dependent upon the amount of carbon dioxide in solution. If the amount of dissolve carbon dioxide decreases, calcium carbonate is precipitated; therefore, the amount of calcium carbonate precipitation increases in warm water because the amount of carbon dioxide in solution is less than in cold water. Considerable amounts of carbonate precipitation occur in warm environments (Illing, 1954). Aquatic plants also absorb carbon dioxide and cause carbonate precipitation. The deposited carbonates can be pure or mixed with sand, clay, iron, manganese, and organic matter. Modern carbonate deposition occurs as deep-water oozes, and reefs, and on shallow shelves. The deepwater oozes form along the flanks of ocean basins at depths of less than 6,000 m. They do not form at depths greater than 6,000 m because the calcium carbonate solubility increases with the increased pressure. On the flanks of the basins, terrigenous material mixes with the calcium carbonate. Often the mixture is 65--89% percent calcium carbonate with silt making up the remainder (Gevirtz and Friedman, 1966). Reef carbonates develop in open oceans on shallow platforms forming atolls, as isolated patches on the ocean shelves, or along the margins of shelf areas as fringing reefs. Fringing reefs generally occur in tropical regions on the western side of the ocean basin. Reefs form as a result of living organisms which form the framework of the sediment (Ginsburg and Lowenstam, 1958). Present-day shelf carbonates are developing in Florida Bay, on the Bahama Banks, on the Australian shelf, and on shelves off British Honduras, Yucatan, and India. The precipitation often occurs as shallow carbonate mud banks. The sediment in the shallow shelf areas often consists of about 10%skeletal material mixed with oolites, mud aggregates, grapestone, aragonite needles, calcareous algae, etc. The average rate of carbonate accumulation on the Bahama Banks is 50 mg cm-2 yr-I (Broecker and Takahashi, 1966). The salinity of the water ranges from 36,000 mg/l of dissolved solids t o 46,500 mg/l. The more saline waters occur in lagoonal areas. The pH ranges from about 8.0 to 8.2 and is lowest at the end of the day because of C02 extraction from the water by marine plants. In the simplistic depositional basin shown in Fig. 7.3 (Moore, 1969), limestones and reef limestones will be deposited on the foreland side, formed from water soluble constituents that precipitated from the saline solutions.

SEDIMENTARY ROCKS

203

Depositional environments of evaporites Calcium carbonate precipitates from sea water after it begins to evaporate (Usiglio, 1849). Removal of calcium ions from solution increases the Mg/Ca ratio in the residual brine, and the precipitated calcium carbonate reacts with the magnesium enriched brine t o form dolomite (Deffeyes et al., 1964). The common order of evaporite deposition in a basin cut off from the open sea is limestone > dolomite > gypsum > halite > potash. Evaporite deposition can be stopped in a basin by a change in climatic regimen or tectonism. If it were changed and new water were allowed to enter, the already deposited salts probably would be effectively protected by the superadjacent high-density brine, and the lighter waters either meteoric or sea water would float on top, developing euxinic conditions similar to a situation found in the Black Sea. Toxic conditions in euxinic areas tend t o preserve deposited organic matter. The preserved organic matter later can be transformed to hydrocarbons and petroleum. Hypersaline lagoons often are very prolific in the production of organic matter (Phleger and Ewing, 1962). Algal pads and an abundance of organic organisms are characteristic of a high-pH, high-salinity environment (Carpelan, 1957) such as exists before euxinic conditions develop. The deposition of evaporites occurs when water evaporates under restricted environmental conditions. The restrictions usually are caused by tectonism such as an uplift, regression of the strand line leaving a relict sea, or biological building of a reef. Sloss (1953) outlined five environments from which evaporites deposit; they are normal marine, euxinic, penesaline, saline, and supersaline. Primary carbonates and dolomitized carbonates precipitate from the normal marine environment where the water contains about 35,000 mg/l of dissolved solids. Limestones rich in organic matter and black shales often are related t o euxinic environments. In the penesaline environment carbonates and primary dolomite form. The salinity of the water is toxic to normal marine life but not sufficiently saline so that halite precipitates. The dissolved solids in the water range from about 250,000 mg/l t o 350,QOOmg/l. From a saline environment carbonates, sulfates, and halites will precipitate. From a supersaline environment potassium compounds will precipitate, and the amount of dissolved solids in the solution will be about 500,000 mg/l. A simplistic model of an evaporite basin is a closed basin initially filled with sea water and evaporated to dryness. In the isochemical system a layer of calcium carbonate is deposited over the entire basin floor as evaporation proceeds. Next, gypsum is deposited and when the water is reduced to about 10%of the original volume, halite precipitates. Halite will deposit only in the deeper parts of the basin and if the solution goes t o dryness the more soluble salts will deposit in any remaining depressions and on top of the halite.

204

ORIGIN OF OILFIELD WATERS

Deffeyes et al. (1964) proposed that dolomitization of limestone results from the evaporation of sea water and precipitation of gypsum causing the ratio of magnesium to calcium in the water t o increase. The concentrated water flows downward, because it is more dense, into the underlying sediments where it reacts with limestone to form dolomite. Modern evaporite deposits are thin and cover relatively small areas of the earth; however, the ancient environments indicate that these depositions were widespread in the United States (Krumbein, 1951) and in the world (Lotze, 1938). The majority of the major evaporite bodies are of marine origin, and range in age from Cambrian to Tertiary. They form in arid marine climates where water lost by evaporation equals or exceeds that supplied by rainfall, rivers, or the open sea. They also form in deep-water environment (Brongersma-Sanders, 1971). The data in Table 7.111 illustrate how the concentrations of some of the dissolved constituents change as sea water is evaporated to dryness (Collins, 1969a). As sea water is evaporated to dryness and the chloride concentration approaches 178,000 mg/l, calcium sulfate precipitates. Halite precipitates when the chloride concentration approaches 27 5,000 mg/l, magnesium sulfate precipitates during the next stage as the chloride concentration approaches 277,000 mg/l. The concentrations of the ions in solution at these various stages approximate those shown in Table 7.111. The data in Table 7.111 indicate that, as sea water evaporates, the concentrations of lithium, magnesium, boron, chloride, bromide and iodide in the residual liquor increase, and that the concentrations of sodium, potassium, rubidium, calcium, and strontium decrease. In most depositional areas, the brines never reached the concentration necessary for the deposition of potassium and magnesium

TABLE 7.111 Concentration changes during evaporation of sea water and brine Element Lithium Sodium Potassium Rubidium Magnesium Calcium Strontium Boron Chloride Bromide Iodide Total Sea water CaS04.l NaCl

MgS04.l

KCI

MgClz.1

2 0.2 11,000 98,000 350 3,600 1 0.1 13,000 1,300 1,700 400 60 7 40 5 178,000 19,000 600 65 2 0.05 295,000

11 140,000 23,000 6 74,000 100 10 300 275,000 4,000 5

12 70,000 37,000 8 80,000 10 1 310 277,000 4,300 7 469,000

27 13,000 26,000 14 130,000 0 0 750 360,000 8,600 8 538,000

34 12,000 1,200 10 153,000 0 0 850 425,000 10,000 8 602,000

517,000

SEDIMENTARY ROCKS

205

salts; or if they did, these salts were later removed by leaching so that their occurrence is relatively rare. Holser (1963) analyzed some brine inclusions in halite from Permian age evaporites. He found that the Br/Cl and Mg/Cl ratios in many of the brine inclusions are similar to those found in the late stages of halite deposition. He concluded that some of the inclusions were connate bitterns with few diagenetic changes, and that the Br/Mg ratio of sea water has remained relatively constant since Permian time. Some diagenetic changes were evident in a few of the inclusions in which a large ratio of Ca/C1 and a low ratio of SO4/Cl compared t o sea water were found. Sediments commonly associated with evaporites are red beds, quartzose sandstones, subgraywacke sandstones, carbonate rocks, and marine shales (Krumbein, 1951). Normal marine evaporite successions are found in intercratonic basins such as the Michigan and Williston Basins. Euxinic black shales sometimes are associated with evaporites. Low redox potentials have been found in modern evaporite (Morris and Dickey, 1957; Quaide, 1958). Examples of modern depositional evaporites are the Karaboghaz Gulf on the eastern side of the Caspian Sea, the Great Bitter Lake of Suez, the Rann of Cutch in northwest India (Grabau, 1920), and the Persian Gulf sabkhas (Evans et al., 1963; Butler, 1969). Deposition of organic matter The organic matter can be biogenic (produced by living systems) or abiogenic (not produced by living systems). The source of biogenic matter can be both terrestrial and marine; for example, considerable plant and animal debris is collected from the land by streams and rivers and carried to the sea, while in the sea large quantities of plant and animal matter live and die. The organic matter that is deposited with sediment usually decomposes if the conditions are right; however, if the environment is reducing some of it may be preserved. The preserved organic matter is transformed into other organic compounds (Kvenvolden, 1964). During sediment diagenesis the organic matter is transformqd to insoluble organic matter (kerogen) and soluble petroleum hydrocarbon (Hunt and Jamieson, 1958). Chemical, bacterial, and catalytic reactions are involved in these conversions. Temperature and pressure affect the reaction rates. Some of the chemical reactions are as follows:

(1)Oxidation : (2) Reduction: (3) Elimination: (4) Polymerization: (5) Cracking:

C2Hm + 0 2 R'OH + H2 R'COOH (small units)

nC02 + '/znH,O +H20 + R'H + co2 + big molecule


-+

+ R'H

c-c-c-c-c-c~oc-c-c-

c-c

206

ORIGIN OF OILFIELD WATERS

The organic matter produced by photosynthesis in the oceans is estimated t o be sufficient t o produce 11 million metric tons of hydrocarbon precursors annually (Riley, 1944). A very small amount of this organic material preserved in sedimentary rocks each year through geologic time would supply all of the known oil and gas fields plus many undiscovered giant fields. Shales consisting of organic material, siltstones, claystones, and limy mud mixtures are found in the trough of a basin (Fig.7.3.). A simplistic idealized view of the trough area is that organic matter deposited in the trough is preserved in the stagnant low-Eh environment. If rapid subsidence occurs, the organic material later is transformed into hydrocarbons which move up and out of the trough into stratigraphic traps on the foreland side or structural traps on the borderland side of the basin. The time interval between deposition of the organic material and conversion to petroleum is millions of years, during which time the trough or ocean basin is filled with sediment and buried, and the sediment is compacted to rock. Some of the water in which the sediments deposited will remain in the rocks as interstitial water.

Deposition of silica
Most of the silica in sediments is of the detrital variety but some is authigenic. Silica is dissolved by waters with high pH potentials and precipitated from water with a low pH. The precipitated silica often acts as a cement.

Sediment compact ion


Sediments compact or consolidate in response t o an imposed load, and in the natural environment, the load is the weight of overlying sediments (Weller, 1959). Compaction of a sediment results in a reduction of the interstitial volume concurrent with expulsion of interstitial water and deformation of the sediment skeleton (the solid granular framework exclusive of bound interstitial water). The grains and the interstitial water are almost incompressible, and the rate of expulsion of interstitial water is about identical with the rate of compaction (Taylor, 1956). Terzaghi and Peck (1968) studied the expulsion of pore water from unconsolidated clays, and determined that the rate at which clay compaction occurs is dependent upon the clay permeability, its volume compressibility, and the square of the thickness of the bed which is compacted. Shales decrease in porosity when compacted; for example, their porosity can decrease about 80%as they compact during burial. White (1957) noted that very large quantities of water are removed from sediments during their compaction. For example, water-saturated shale decreasing in porosity from 20 t o 10% loses 10" liters of water per km3 of sediment. Such sediment could yield per km3 a water supply of 20 liters per

SEDIMENTARY ROCKS

207

minute for 10,000 years. Compaction water is further considered in a subsequent chapter concerned with the accumulation of petroleum. The resistance t o flow of bound interstitial water is greater than the resistance to deformation of the sediment framework during the first stages of compaction. As pressure is increased and sediment compacts, the sediment framework increases resistance to deformation, which also governs the deformation rate of the bound-water films and the outflow of the interstitial found that bound water has a higher viscosity than water. Rosenqvist (1962) unbound insterstitial water. Permeability decreases with compaction, and this together with the increased water viscosity leads to additional resistance to expulsion of bound interstitial water during subsequent compaction. Porosity decreases during compaction, and at infinite depth it would become infinitely small. Porosity and permeability are two important factors in determining the amount of petroleum and/or water that can be recovered from a given reservoir or aquifer (Pollard and Reichertz, 1952;Caraway and Gates, 1959). The porosity of an aquifer indicates how much fluid the aquifer can hold, and the permeability indicates how fluid can move through the aquifer. If the porosity is high but the permeability is low, the reservoir may contain large amounts of oil, gas, or water, but they cannot be recovered unless special techniques are used to increase the permeability.

Sediment diagenesis
Mineralogical and chemical changes occur in the sediments as they compact. The mineralogic composition of recent marine sediments and ancient marine sedimentary rocks are different, as is the composition of the interstitial water in the recent sediments compared with the waters in ancient stratigraphic units. Chemical reactions occur between the sediments and their interstitial water (Chave, 1960). It has been shown that chemical changes can be measured in recent sediments, e.g., below the sedimentwater interface, changes in pH and Eh result from degradation of sulfate ions by bacteria (Emery and Rittenberg, 1952). Numerous chemical inhomogeneities occur within a single core sample of recent sediments (Degens and Chilingar, 1967). The magnesium concentration in the interstitial w&er decreases slightly with depth, while the concentrations of calcium and potassium increase (Siever et al., 1965). The interstitial waters from continental shelf sediments have higher chloride concentrations and higher ratios of Ca/C1, K/Cl, and Rb/C1 than the overlying sea water; the ratios of Li/Cl and Mg/Cl are about the same as in the overlying sea water except that the Li/C1 ratios are higher in the innershelf samples than in the outershelf samples. The Sr/Cl ratio is higher in the overlying sea water, while the pH and Eh values are lower in the sediments than in the overlying sea water (Friedman et al., 1968). A detailed study indicates that virtually no environment exists on or near the earths surface where the pH/Eh conditions are incompatible with

208

ORIGIN OF OILFIELD WATERS

organic life (Baas Becking et al., 1960). Because COz is the main byproduct of organic oxidation and the building material of plant and much bacterial life, it plays a dominant role. Carbon dioxide dissolves in water, producing the bicarbonate ion and a free hydrogen ion. The concentration of the hydrogen ion is lo- equiv./l (pH 7) at 2OoC in pure water, but when saturated with COz it rises t o lo- (pH 5). This reaction moves to the right with increasing temperature in a closed system. In the presence of organic constituents, the equilibrium is modified, and the pH range can extend from 2 to 12. The ionic potentials of the constituents involved in diagenesis are important (Cloke, 1966). Those that stay in true ionic solution up t o rather high pH levels are Na+, K+, Mg+, Fe+, Mn+2,Ca+, Sr+, Ba+, etc.; they are the soluble cations, and their ionic potentials range from 0 t o 3, where the ionic potential is the ratio between the ionic charge and the ionic radius. Constituents that are precipitated by hydrolysis are those with ionic potentials from 3 t o 12 and include such ions as A P 3 , Fe+3, S P 4 , and Constituents that form soluble complex ions and usually go into true ionic and solution include B+3 , V 4 , N+ P+ , S6, Mn+ ; their ionic potentials are over 12. In general, the hydroxides of the soluble cations possess ionic bonds; therefore, they are soluble, the hydrolysates or those precipitated by hydrolysis form hydroxyl bonds, and the soluble complex ions have hydrogen bonds. Organisms that consume oxygen cause a lowering of the redox potential, and in buried sediments it is the aerobic bacteria that attract the organic constituents and remove the free oxygen from the interstitial water. Sediments laid down in a shoreline environment often differ in degree of oxidation from those laid down in a deep-sea environment (Pirson, 1968). For example, the Eh of the shoreline sediments may range from -50 t o 0 mV while the Eh of deep-sea sediments may range from -150 to -100 mV. The aerobic bacteria die when the free oxygen is totally consumed, and the anaerobic bacteria attack the sulfate ion which is the second most important anion in the sea water. During this attack, the sulfate is reduced to sulfite and then t o sulfide. Also the Eh drops t o -600 mV (Fig.7.2). Sulfide is liberated and CaC03 precipitates as the pH rises above 8.5 (Dapples, 1959). Sulfur has two stable isotopes, * S and S, with a mass differential of 6%. The isotopes are fractionated during the change of S04-2 to Sv2, and SZ is enriched in the more energetic 32S isotope. The average ratio of 3zS/34S in normal sea water sulfate is about 21.76 (Ault, 1959). The sulfate isotopes are useful in interpreting ancient diagenetic stages. Reactions occur during sediment diagenesis and affect the composition of the interstitial water. Calcite is precipitated if the pH is high, or it is dissolved if the pH is low. Dolomitization occurs as follows:

2CaC03 + MgClz

* CaMg(C03)2 + CaClz

SEDIMENTARY ROCKS

209

Other exchange reactions occur, whereby montmorillonite transforms to illite (Burst,1969). Ions are adsorbed by negatively charged clays from the sea water (Krauskopf; 1956). Dissolved salts hydrolyze; e.g., olivine hydrolyzes to serpentine as follows: 2Mg2 SiO, + 3H20 =+3Mg
*

2Si02

2H20 + Mg(OH)2

Hydration and dehydration reactions occur, such as the gypsum-anhydrite relation : CaS0,

+ 2 H 2 0 + CaSO,

2H20

As depth of burial increases, many mineralogical changes take place in the sediments. In the Gulf Coast Tertiary, the clay mineral montmorillonite gradually disappears at depths between 2,500 and 3,000 m (Burst, 1969). It is replaced by mixed layer and illitic clay minerals. This change involves chemical alteration and also the release of water of crystallization. This change appears t o be temperature dependent with the reactions starting at about 100C. Other mineral changes occur in the sandstones such as the deposition of secondary silica overgrowths on the quartz grains causing resultant loss in porosity. Available data differ on the loss of porosity with burial depth (Atwater and Miller, 1965; Philipp et al., 1963). Much of the silica may come from outside the porous sand, such as from shales whose pore water is traversing the sand as a result of compaction. Authigenic clay minerals such as kaolinite also form in the pores. Interstitial water in deeply buried sediments sometimes is at a pressure close t o the weight of the overburden. This pressure may be sufficient t o burst the rock and allow the water t o move out through the fissures which it forms. These fissures are principally vertical and extend upward into zones of lower temperature. The water forming and subsequently filling the fissures usually is a mixture of salty interstitial sedimentary pore water and water from the clay minerals released by their recrystallization. If it is hot, and saturated with silica and other minerals, it could force its way upward and as it cools deposit quartz, feldspar, calcite, and other minerals. Possibly many hydrothermal ore veins were formed by interstitial sedimentary waters rather than by juvenile waters. These hydrothermal veins contain metallic minerals composed of compounds containing copper, zinc, lead, gold, and silver. The process is a geothermal convection cell and it is able to concentrate and segregate useful minerals. The process has many points of resemblance to the concentration and segregation of petroleum - the principal difference being that the geothermal convection cells operate at higher temperatures.

210

ORIGIN OF OILFIELD WATERS

Petroleum and natural gas


The total amount of organic matter dispersed in the sedimentary rocks of the earth has been estimated to be about 2,700 trillion metric tons; of this amount 50 trillion metric tons are dispersed petroleum hydrocarbons, of which 0.5 trillion metric tons exist in petroleum reservoirs (Hunt, 1968). The types of hydrocarbons in a petroleum often indicate its origin; for example, if it contains predominately odd-numbered n-alkanes in the low molecular-weight range, it probably was formed from marine organisms. Petroleums from the Uinta Basin contain a predominance of odd-numbered hydrocarbons in the vax fraction, indicating a nonmarine organic source. Waxes derived from land plants contain a predominance of hydrocarbons with carbon numbers of C27,CZ9,and C31,while hydrocarbons derived from marine plankton may contain more hydrocarbons with carbon numbers of c 7 and c 9 1, 1. The water in the sediments containing the organic matter contains many dissolved organic constituents such as salts of the humic, fatty, and naphthenic acids, sugars, heterocyclics, and aromatic oxygen compounds. Degens et al. (1964) observed that as the salt concentration in petroleumassociated waters increases, the concentration of dissolved amino acids increases. Petroleum is generated in organic-rich shales, but the mechanisms whereby it migrates from the shales and concentrates in porous reservoir rocks are not understood. Petroleum precursors leave the shale with the water as the water is expelled by compaction. As burial proceeds, pressures and temperatures increase. With increasing temperature, chemical changes in the solids are accelerated and the organic matter first generates petroleum, which ultimately is converted to methane and finally graphite. The clay minerals continue their recrystallization, and finally metamorphism to slates, phyllites, and schists occurs: These processes involve a continuing loss of porosity with the release of additional pore water. The solubility of petroleum hydrocarbons in water increases with increasing temperature and pressure. However, at ambient temperature and pressure the solubility in pure water is rather low (McAuliffe, 1969). Waterwet shale has no permeability t o immiscible fluids such as gas or oil, so the petroleum or petroleum precursors probably do not move as droplets. Peake and Hodgson (1966) report accommodations of specific hydrocarbons in water up to about 30 ppm. Cartmill and Dickey (1970) found that a colloidal suspension was able to pass through water-wet sands, but the tiny droplets coalesced at points where the grain size decreased. Neruchev and Kovacheva (1965) offered some evidence that the amount of extractable hydrocarbon decreases in shales for the first few meters away from the reservoir rocks, as if removed by some flushing action. Bruderer (1956) suggested that oil deposits originated from sea water

SEDIMENTARY ROCKS

21 1

containing dissolved hydrocarbons, which subsequently deposited them in sedimentary reservoir ,rocks. Altovskii et al. (1961) believe that petroleum forms in subsurface formation waters and moves to traps as an emulsion or in aqueous solution. Movement of hydrocarbons from the source bed to the reservoir via aqueous solution or as molecular films was postulated by Brod (1960). Baker (1960) suggested that, as the sediments compact, the expressed water may contain solubilizers capable of releasing sediment hydrocarbons into aqueous colloidal solution, and subsequent changes in the aqueous equilibria cause the colloids to become oil droplets in reservoirs. The soluble salts of adenosine triphosphate are capable of solubilizing numerous inorganic and organic compounds in neutral or slightly alkaline medium and of keeping these compounds in solution (Mandl et al., 1952). Hydrocarbon solubilization processes are essential to the migration of petroleum and acquired evidence indicates that nature provides solubilizing agents. Numerous solvents are evaluated by Mandl (1953), many of which are useful agents for solubilizing relatively insoluble inorganic and organic compounds. Organic as well as inorganic compounds will enter the aqueous phase to the limit of their solubilities. Interactions of the solubilized compounds affect the solubility product of other solubilized compounds. For example, the presence of dissolved sodium chloride may inhibit or increase the solubility of a normal alkane. For any given system, equilibria of all components will be attained only at a given temperature and pressure. If the temperature changes, if the pressure changes, or if more inorganic or organic constituents are added or subtracted, the equilibria will change causing solubilization or deposition. Hydrocarbons can exist in the aqueous phase as emulsions or colloids, as suspended particles, or in true solution. Suspended particles will settle from an aqueous phase because of gravity, whereas colloids will remain in suspension. Suspended particles possess colligative effects, but colloidal sols do not (Van Nostrand Press, 1958). A dissolved particle usually is smaller than 5 mp, while colloidal particles usually range in size from 0.25 mp to 6 mp. Phenols and alkalinaphthenates capable of acting as emulsifying agents and causing oil-in-water emulsion t o migrate have been detected in crude oils and in the associated brines (Neumann and Jobelius, 1967). It has been demonstrated that finite amounts of paraffinic, aromatic, napthenic hydrocarbons and other organic derivatives are dispersed in recent marine sediments (Smith, 1954). Surfactant organic acids stabilize hydrocarbon particles of relatively large size and contribute t o solubilization and subsequent migration of hydrocarbons from sediments (Baker, 1960). A shift in the equilibria caused by a pH or Eh change, adsorption of the surfactant, etc., results in deposition of the pseudo-soluble hydrocarbon. Therefore, solubilization and mobilization of hydrocarbons from sediments into the aqueous phase occurs when the equilibria are shifted t o the correct conditions. Subsequent migration of the hydrocarbons in the aqueous phase

212

ORIGIN OF OILFIELD WATERS

occurs if the equilibria in the aqueous phase remain constant while the water moves through the sedimentary rocks. Deposition or accumulation of hydrocarbons occurs if the equilibria in the aqueous phase shifted, causing desolubilization or precipitation of the hydrocarbons. Temperature gradients in sedimentary basins usually are about 1C per 46 m of depth, and the rate of heat flow to the surface is approximately 1.2 x loV6 cal cmV2 sec-I (Birch, 1954). Temperature is believed to be a primary cause in the conversion of organic matter in rocks t o petroleum (Philippi, 1965), it is also believed that lipids are the major precursors of petroleum and that most petroleum is generated by chemical reactions occurring at temperatures above 115C. Nonmarine sources are recognized for many crude oils, in contrast to the once general belief that such sources are unfavorable for the generation of petroleum. Perhaps the best known examples in the United States are the nonmarine sequences in the Eocene of the Uinta Basin in Utah. Other examples of oil and gas with continental source sediments are basins such as the Dzungaria, Tsaidam, Tarim, Turfan, Ordos, Pre-Nan Shan, and Sungliao of China (Meyerhoff, 1970). There is considerable nonmarine Tertiary age strata in the Cook Inlet-Kena Basin in coastal Alaska.
TABLE 7.IV Tertiary system - highest concentration of a constituent found, average concentration. and number of samples analyzed Constituent Concentration (mg/l) highest Lithium Sodium Potassium Rubidium Cesium Calcium Magnesium Strontium Barium Boron Copper Chloride Bromide Iodide Bicarbonate Carbonate Sulfate Organic acid as acetic Ammonium 27 103,000 1,200 0.6 0.4 38,800 5,800 420 240 450 1 201,300 1,300 35 3,600 300 8,400 1,900 2,700 average 4 39,000 2 20 0.24 0.20 2,530 530 130 60 36 0.63 64,600 85 28 560 75 3 20 140 230 169 379 176 11 9 37 6 368 142 140 170 3 380 323 322 3 64 8 139 53 64 Number of samples

COMPOSIDION OF OILFIELD WATERS TABLE 7.V

213

Cretaceous system - highest concentration of a constituent found, average concentration, and number of samples analyzed Constituent Concentration (mg/l) highest Lithium Sodium Potassium Rubidium Cesium Calcium Magnesium Strontium Barium Boron Chloride Bromide Iodide Bicarbonate Sulfate Ammonium
13 88,600 580 0.10 0.10 37,400 8,000 980 67 0 70 187,000 1,760 190 1,660 7,100 35

Number of samples

average
4 31,000 130 0.10 0.10 7,000 900 200 40 20 62,OOb 550 25 260 280 23 26 987 38 1 1 987 987 39 34 38 987 173 172 864 776 2

Composition of oilfield waters To illustrate the variety of oilfield waters found in subsurface petroleumbearing formations, samples were taken, and analyzed, from formations of the following ages: Tertiary, Cretaceous, Jurassic, Permian, Pennsylvanian, Mississippian, Devonian, Silurian, Ordovician, and Cambrian. Insufficient samples from Triassic age strata were available. These data are given in Tables 7.IV-XIII, and all of the samples were taken from sedimentary basins in the United States. Each table gives the highest value found in this study in milligrams per liter for a given constituent, the average values, and the number of samples used t o estimate the average value. The constituents determined for most of the brines include lithium, sodium, potassium, rubidium, cesium, calcium, magnesium, strontium, barium, boron, copper, chloride, bromide, iodide, bicarbonate, carbonate, sulfate, organic acid as acetic (actually organic acid salts but calculated as acetic acid), and ammonium. Comparison of the lithium content in the samples from Tertiary age strata (Table 7.IV) with the lithium content of sea water (Table 7.1), indicates that the average lithium content of 169 oilfield waters is enriched by a factor of 20. At least one sample of oilfield water contained 27 mg/l of lithium or an enrichment factor of 135 compared t o sea water. Table 7.11 indicates that igneous rocks contain up t o 65 ppm of lithium and sedimentary rocks contain up to 46 ppm of lithium.

214
TABLE 7.VI

ORIGIN OF OILFIELD WATERS

Jurassic system - highest concentration of a constituent found, average concentration, and number of samples analyzed Constituent Concentration (mg/l) highest Lithium Sodium Potassium Rubidium Cesium Calcium Magnesium Strontium Barium Boron Chloride Bromide Iodide Bicarbonate Sulfate Organic acid as acetic average
-

Number of samples

400 120,000 900 0.10 0.10 56,300 5,200 2,080 50 50 210,000 6,000 40 2,640 1,480 12

10 57,300 140 0.10 0.10 25,800 2,500 320 10 13 141,000 1,200 16 140 21 0

80 85 9 1 1 85 84 9 7 9 85 80 8 72 78
1

12

TABLE 7.VII Permian system - highest concentration of a constituent found, average concentration, and number of samples analyzed Constituent Concentration (mg/l) Number of samples highest Lithium Sodium Potassium Rubidium Cesium Calcium Magnesium Strontium Boron Copper Chloride Bromide Iodide Bicarbonate Carbonate Sulfate Organic acid as acetic Ammonium average

6 109,000 40 5 2 0.20 22,800 5,800 10 20 0.88 177,000 68 3 281 36 3,400 2 20 24

3 47,000 170 0.80 0.13 8,600 2,000 7 8 0.88 92,700 46 3 77 36 7 30 170 24

3 54 3 3 3 54 53 3 3 1 64 3 1 49

1
41 2 3

COMPOSITION OF OILFIELD WATERS TABLE 7.VIII

21 5

Pennsylvanian system - highest concentration of a constituent found, average concentration, and number of samples analyzed Constituent Concentration (mg/l) highest Lithium Sodium Potassium Rubidium Cesium Calcium Magnesium Strontium Barium Boron Manganese Chloride Bromide Iodide Bicarbonate Carbonate Sulfate Organic acid as acetic Ammonium
35 101,000 710 2.30 8.50 205,000 15,000 4,500 640 70 105 270,000 3,900 1,410 1,200 70 5,400 2,300 3,300

Number of samples

average
7 43,000 170 0.55 0.15 9,100 1,900 600 30 15 60 87,600 490 210 130 40 430 430 300 45 951 57 25 19 9 50 947 70 41 54 2 950 57 52 897 2 7 56 44 51

Table 7.V indicates that oilfield water samples taken from Cretaceous age strata were enriched in lithium with respect to sea-water. The highest lithium concentration found in 26 samples was 13 mg/l. Table 7.VI indicates that oilfield waters taken from Jurassic age strata contain up t o 400 mg/l of lithium, which, compared with sea water (Table 7.1), represents a concentration factor of 2,000. Compared t o the hydrolyzates in sedimentary rocks (Table 7.11), the concentration factor is about 9. The lithium concentration in oilfield waters taken from Permian age strata averaged 3 mg/l (Table 7.VII). Only three samples were available for use in determining this average. Table 7.VIII indicates that the lithium concentration averaged 7 mg/l in 45 samples taken from Pennsylvanian age strata. This represents a concentration factor of 35 compared with sea water (Table 7.1). Table 7.IX indicates that the lithium concentration in oilfield waters taken from Mississippian age strata is enriched by a factor of 45 compared with sea water. Table 7.X indicates a similar enrichment factor of 250 for oilfield waters taken from Devonian age strata. For waters from Silurian age strata (Table 7.XI) the enrichment factor found was 185; for the Ordovician

216

ORIGIN OF OILFIELD WATERS

TABLE 7.IX Mississippian system - highest concentration of a constituent found, average concentration, and number of samples analyzed Constituent Concentration (mg/l) highest Lithium Sodium Potassium Rubidium Cesium Calcium Magnesium Strontium Barium Boron Copper Manganese Chloride Bromide Iodide Bicarbonate Carbonate Sulfate Organic acid as acetic Ammonium
55 115,800 5,000 5 2 37,800 11,200 3,390 20 240 3 36 206,000 1,800 620 1,590 450 3,500 3,070 700

Number of samples

average
9 41,500 430 1 0.40 8,900 1,600 630 5 40 3 12 85,000 410 110 185 450 540 370 210 81 210 80 47 37 209 202 52 44 86 2 5 210 88 89 198 1 191 84 83

age (Table 7.XII), it w a s 100; and for the Cambrian age (Table 7.X111), it was 85. The data in Tables 7.VI-XI11 indicate that waters taken from sediments that formed during the various geologic ages do not all have the same chemical composition and that the waters have evolved considerably in comparison to modern sea water composition (Table 7.1). The manner whereby this evolution occurred is not completely understood; however, recent studies have shed some light on the problem. Note the amount of organic acid as acetic found in waters taken from the sedimentary rocks (Tables 7.VI-XIII). The organic acids are present in the oilfield waters as organic acid salts. These organic compounds possibly are a precursor of petroleum and serve as a transportation mechanism for migration. The exact composition of each organic acid salt has not been determined. Knowledge of the composition of these organic acid salts would aid in geochemical studies of petroleum. Rittenhouse et al. (1969) studied the minor elements in 823 oilfield-water samples taken from subsurface formations in the United States and Canada. The data that they found are shown in Table 7.XIV as 25% quartiles, median concentrations, and 75% quartiles. The dissolved solids are given in grams per

COMPOSITION OF OILFIELD WATERS TABLE 7.X

217

Devonian system - highest concentration of a constituent found, average concentration, and number of samples analyzed Constituent Concentration (mg/l) highest
~~

Number of samples

average

Lithi um Sodium Potassium Rubidium Cesium Calcium Magnesium Strontium Barium Boron Copper Manganese Chloride Bromide Iodide Bicarbonate Carbonate Sulfate Organic acid as acetic Ammonium

170 101,000 11,600 11 1.4 129,000 26,000 2,300 120 90 2 200 259,000 3,500 120 1,000 60 1,700 67 0 560

50 48,000 3,100 4 0.5 18,000 2,900 1,000 40 30 2 175 115,000 1,060 30 155 30 450 130 110

29 85 30 12 11 85 82 8 7 30 1 2 85 32 32 67 2 74 27 32

liter, the data followed by p are given in parts per billion (ppb), and the other data are given in parts per million. They analyzed samples from several basins as illustrated in Table 7.XIV, and the elements analyzed included lithium, magnesium, manganese, nickel, cobalt, chromium, copper, potassium, tin, strontium, titanium, vanadium, and zirconium. Rittenhouse et al. (1969) concluded that elements in oilfield waters commonly are present in the following concentrations:

5% 5% or ppm > 100 ppm


1-100 ppm ppb (most oilfield waters) ppb (some oilfield waters)

Na, C1 Ca, SO4 K, Sr Al, B, Ba, Fe, Li Cr, Cu, Mn, Ni, Sn, Ti, Zr Be, Co, Ga, Ge, Pb, V, W, Zn

They found no relationship between the constituents in the brine and the minerals in the aquifer rocks except for potassium. They postulated that exchange reactions occurred between the clays in the rocks and potassium in the water to control the dissolved potasssium.

218

ORIGIN OF OILFIELD WATERS

TABLE 7.XI Silurian system - highest concentration of a constituent found, average concentration, and number of samples analyzed Constituent Concentration (mg/l) highest Lithium Sodium Potassium Rubidium Cesium Calcium Magnesium Strontium Barium Boron Chloride Bromide Iodide Bicarbonate Sulfate Organic acid as acetic Ammonium TABLE 7.XII Ordovician system - highest concentration of a constituent found, average concentration, and number of samples analyzed Constituent Concentration (mg/l) highest Lithium Sodium Potassium Rubidium Cesium Calcium Magnesium Strontium Barium Boron Manganese Chloride Bromide Iodide Bicarbonate Carbonate Sulfate Organic acid as acetic Ammonium
70 89,100 2,890 6 0.5 39,000 10,900 900 10 80 56 205,600 7 20 70 2,260 60 7,600 3,300 630
-

Number of samples

average
37 49,100 1,900 4 0.4 21,000 4,300 7 30 15 30 122,000 520 17 115 830 90 80 8 14 11 2 2 14 12 2 1 10 14 11 10 11 13 9 10

90 89,000 8,400 8 0.4 41,000 12,000 880 15 90 195,000 1,700 30 27 0 3,500 220 20 0

Number of samples

average 20 31,000 990 2 0.2 6,100 1,300 340 6 20 56 62,000 300 25 270 25 1,070
5 20 140 15 609 15 11 9 609 607 12 10 18 1 609 19

16
598 26 583 14 16

RESEARCH STUDIES TABLE 7.XIII

21 9

_______--__

Cambrian system - highest concentration of a constituent round, average concentration, and number of samples analyzed
~~

__

Constituent

Concentration (mg/l) highest


._____

Number of samples
_

average
_

Lithium Sodium Potassium Rubidium Cesium Calcium Magnesium Strontium Boron Chloride Bromide Iodide Bicarbonate Sulfate Organic acid as acetic Ammonium

40 43,000 2,000 3.3 0.6 14,500 8,800 360 13 95,000 1,170 40 790 2,600 50 120

17 23,400 440 3.3 0.6 4,000 1,300 125 7 46,100 520 18 260 1,170 30

8 23 10 1 1 23 22 7 5 23 5 3 23 22 3 3

60

Compared with sea water the 823 brines were enriched in manganese, lithium, chromium, and strontium, and depleted in tin, nickel, magnesium, and potassium. Generally the silicon content varied inversely with the dissolved solids content. This agrees with a study of the solubilities of silicate minerals where Collins (1969b) found that in general the silicon solubilities decreased with increasing concentrations of dissolved salts at ambient conditions. Research studies related to the origin of oilfield brines Tables 7.IV-XIV indicate that the compositions of oilfield brines are not consistent, and that they are not formed by the simple evaporation or dilution of sea water. Oilfield brines are found in deep formations that sometimes contain fresher water nearer surface outcrop areas, in formations containing evaporites or in close proximity to soluble minerals, and in formations close to surface saline waters. The amounts and ratios of the constituents dissolved in oilfield waters are dependent upon the origin of the water and what has occurred t o the water since entering the subsurface environment. For example, some subsurface waters found in deep sediments were trapped during sedimentation, while other subsurface waters have infiltrated from the surface through outcrops.

N
ES

TABLE 7.XIV Minor elements in 823 oilfield brine samples in United States and Canada*' Number of samples Illinois Basin 22 Louisiana and Texas Gulf Coast 79 East Texas 88 North Texas 24 West Texas and New Mexico 148 Permian only 74 Pennsylvanian only 34 Silurian and Devonian only 15 Ordovician and Cambrian only 21 Anadarko Basin*' 118 Williston Basin, postPaleozoic 25 Williston Basin, Paleozoic 55 Powder River Basin 22 Other Wyoming 28 Colorado 18 California 116 Sea Water Estimated detection limit Lithium
q25 md 10 15

Magnesium
q75 q25 25 4 ND 15 25 25 20 25 25 35 10 50 2 45 ND ND 3,000 15 150 3,000 500 500 500 200 500 900 10 300 10 20 10 35

Manganese
q75 q25

Nickel
q75 750p >5,OOOp >5,OOOp 90 >5,OOOp >5,OOOp >5,OOOp q25

Tin md ND
q75 q25

md
6,000 250 250 5,000 1,000 1,000 1,000 400

md

md

q75 5P 1P 1lP 25p 3P 3P 3P 5P 4P 4P

8,000
550 800 6,000 1,650 2,000 1,500 560 1,000 3,000 2,000 2,000 225 200 300 175

8Op 175p 3.5 8OOp 1 , 8 0 0 ~ 3.3 25 45 2OOp 18Op 500p 30p 1.8 1.7 2.8

ND

ND

ND ND ND
3 2 3 4 10 ND

ND ND ND
15 10 10 10 15 10

< lp < < 1P


15p

3p ND 3p ND 150p ND l p ND 3p ND l p <Ip l p ND l p <1 15p ND 3p ND 3p 3p 3p 3p 35p

< 1P < 1P
3P 12p

300p >5,OOOp >5,OOOp >5,OOOp 450p 1,200~ 2,000~

800
1,550 250 600 40 100 30 90 1,272 10

150p 400p 600p 5.6 9Op 300p 2OOp 300p 60p 9Op 300p 660p 450p 300p 300p 950p lp-lop 1P

< lp < 1P < lp < lp < lp


6p

< < < <

< 1P < 1P < 1P


IP IP 2P

ND
18 ND ND ND ND

ND
35 ND ND ND ND 0.1 2

<

1,000~
750p 2,800~

< 3p < < ND < 3p < ND < 3p <


lop 5.4p

ND ND ND ND ND

< lp < lp < lp < lp

<lop <lop 2.5~ 4.5~ 3P 1P

< lp < lp < lp < lp

_ 1P_

___

*' Medium (md - Rittenhouse et al., 1969)and quartile (9) concentrations in each area; ND = below detection limits; p = concentration in ppb, otherwise ppm.
** No data, less sensitive methods of analysis used.
*3

Includes Oklahoma Platform and Ardmore Basin.

TABLE 7.XIV (continued)*' Number of samples Dissolved solids (gll) md


-

q75
119

Cobalt md
~ ~

Chromium
~

Copper

Potassium

q75
~

q75 3P
<lop

md

q75 75p <25p < 1 450p 10P 10P 5P

q25 180 160


ND ND

md

q75 400 400 300 100 ,0 500 750 400 450 650 500 400 <5,000 400 700 400 70

22 Illinois Basin Louisiana and Texas Gulf 79 Coast 88 East Texas 24 North Texas West Texas and New 148 Mexico 74 Permian only Pennsylvanian only 34 Silurian and Devonian 15 only Ordovician and 21 Cambrian only Anadarko Basin*3 118 Williston Basin, postPaleozoic 25 Williston Basin, Paleozoic 55 22 Powder River Basin Other Wyoming 28 Colorado 18 California 116 Sea water Estimated detection limit

70 30 27 173 61 70
80

98 69 66
222

ND

ND

ND

lop

300 300 <50 300 350 400 300 300 400 250 300
800

131 116 241 173 215 168 72 128 203


88

ND ND <5P ND*~ ND*~ ND*~ ND ND ND ND ND ND ND ND ND ND ND ND ND ND ND ND ND ND ND ND ND

< 2P
2P 35P 4P 4P 4P 4P

<25p < 1P 25P


ND ND ND

<25p < 1 150p 1P 2P 1P 4~ 4~ lop <25p 3P <25p


ND

111 143 115 55 67 137 59 173 5 5 5 18 35

120 160
200

<

42 53 51 9 115 3 4 3 5

<

1P

1 5 ~ 170 1 5 ~ 200 20 25p 2Op 5P 70p 30 <25p 2Op


200

<5P
ND

3P N 4 2 5 ~ < 1P

<5P
ND

<5P
ND

< 2P < 6~ 20 < 6;


IP
~~

<25p

296 11 11 15 30

<5P
ND ND ND

<5P
ND

<5P
ND

1 5 ~ ND <25p
ND

.0 40
200 ND
200

<5P
ND

<5P 2P

ND

0.27~ 1P

5p 15p 0.04~-0.07~

<25p 2p

<25p 5p lp-15p 1P

ND

300 300 300 45 380 50

_ _

. . -

*'

*'

*3

Medium (md - Rittenhouse et al., 1969)and quartile (4) concentrations in each area; ND = below detection 1imits;p = concentration in ppb,otherwise ppm. No data, less sensitive methods of analysis used. Includes Oklahoma Platform and Ardmore Basin.

to to to

Number of samples Illinois Basin Louisiana and Texas Gulf Coast East Texas North Texas West Texas and New Mexico Permian only Pennsylvanian only Silurian and Devonian only Ordovician and Cambrian only Anadarko Basin Williston Basin, postPaleozoic Williston Basin, Paleozoic Powder River Basin Other Wyoming Colorado California Sea water Estimated detection limit
~

Strontium q25 md 140 300


__.

Titanium q75 q25


~

Vanadium
_____. -

Zirconium q25 md <lop <lop q75 <lop <lop

md
<1 op

q75

q25

md

22 79 88 24 148 74 34 15 21 118 25 55 22 28 18 116

400 < l o p

<lop <lop 1P

ND ND ND ND ND ND ND ND ND ND ND ND ND ND ND ND

ND
<lop

85 200 ND 45 75 350 750 ND 150 450 700 < l o p


1 5 200 65 90 180 300 75 90 400 ND 300 ND 450 ND 300 < l o p

<lop

ND
7P <lop <lop <lop <lop <lop <lop

<

2OP
<lop <lop <lop <lop <lop <lop <lop

ND ND ND ND ND ND ND ND ND ND
<lop

ND
<lop

ND
lop

ND ND ND ND ND
<lop

ND
<lop

ND ND ND
2Op

100 250 400 ND 90 300 650 ND 20 1 0 0 200 50

ND

ND

ND ND
<lop

ND
<lop <lop

ND
10 7

ND

95 450 < l o p 25 50 < l o p 45 ND 20 20 60 < l o p 1 0 22 < l o p 13 16

<lop 25P <lop <lop <I op <lop <lop <1 op <lop 7P present 10P
.

ND ND ND

ND
<lop

ND
<lop

ND ND
1OP

ND

*' Medium (md; from Rittenhouse e t al., 1969) and quartile ( 4 ) concentrations in each area; ND = below detection limits; p = concentration
in ppb, otherwise ppm.

** No data, less sensitive methods of analysis used.


*3

Includes Oklahoma Platform and Ardmore Basin.

RESEARCH STUDIES

223

Some waters are mixtures of the infiltration water and trapped ancient sea water.. Also, the rocks containing the waters often contain soluble constituents which dissolve in the waters or contain chemicals which will exchange with chemicals dissolved in the waters causing alterations of the dissolved constituents. The amounts of dissolved constituents found in oilfield waters range from less than 10,000 mg/l t o more than 350,000 mg/l. This salinity distribution is dependent upon several factors including hydraulic gradients, depth of occurrence, distance from outcrops, mobility of the dissolved chemical elements, soluble material in the associated rocks, ion exchange reactions, and clay membrane filtration. Concentration of sea water can occur by surface evaporation, and there are at least three independent processes that can cause major changes in buried, isolated sea water: (1) Dilution with meteoric or fresher waters which have entered outcrops. (2) Reactions with minerals in the sediments and sedimentary rocks (the reactions are often temperature and pressure dependent). (3) Membrane filtration through clays and shales as a result of pressure and osmosis.

Playa deposits
Jones et al. (1969) studied the composition of brines in shallow, finegrained playa deposits in the Great Basin. The concentrations of dissolved solids in the water in these sediments often were as much as five times greater than in the water in the associated lakes. They attributed the concentration processes t o capillary evaporation and entrapment of fossil brines when the salinity of the lake water was greater (lake nearly dry).

Continental Slope drill holes


Manheim and Bischoff (1969) analyzed pore waters from drill holes on the Continental Slope of the northern Gulf of Mexico. A relationship between the salinities of the waters and the proximity of diapiric structures was found. This indicated that salts are leached from salt-bearing sediments to increase the salinities of the pore waters. In some samples the high bromide and potassium concentrations suggested that late-stage evaporitic minerals such as carnallite and polyhalite were leached from salt bodies. They postulated that molecular diffusion is a major mechanism which influences the distribution of salt in the pore waters. Similar conclusions have been made for saline waters in other areas.

Relation to petroleum accumulations


Van Everdingen (1968) suggested that major circulation systems of for-

224

ORIGIN OF OILFIELD WATERS

mation water exist in the Western Canada Sedimentary Basin, that the flow systems affected accumulations of hydrocarbons in the basin, and further that pressure and salinity variations might be explained by membrane properties of the shales. He saw a need for studies of the hydrodynamics of the basin. Hydrodynamics and geochemistry of the Paradox Basin were studied by Hanshaw and Hill (1969). The ground-water movement in the basin is generally southwestward from the high outcrop areas in western Colorado, flowing toward the Colorado River discharge areas. Hydrodynamic conditions exist in lower Paleozoic strata which are favorable to accumulations of petroleum in stratigraphic traps. Paleozoic aquifers in northwestern New Mexico have very high potentiometric surfaces and these aquifers may be the outflow receptors of an osmotic membrane system operating within the San Juan Basin. This regional study was excellent and of value in exploration for petroleum and gas. Parker (1969) studied brines and waters in five aquifers of Cretaceous age in the East Texas Basin. He found that the composition of the waters in the older, more deeply buried aquifers were modified more than waters in younger, less deeply buried aquifers. Most of the modifications were made by exchange reactions, dilution by meteoric waters, and loss of sulfate because of bacterial reduction. Hydrodynamic movement of the waters in the Woodbine formation contributed t o the giant oil accumulation in the East Texas Basin. Much of the stratigraphic trapped oil probably was trapped in part because of this type of flow. Sabkha sediments and transport of valuable ores Bush (1970) discussed the origin of chloride-rich brines from Sabkha sediments and how they are related t o inclusion brines and lead-zinc deposits of the type found in the Mississippi Valley. He noted that Helgeson (1964) and Barnes and Czamanske (1967) have shown that chloride-rich scrlutions can transport lead and zinc as chloride complexes. According t o Bush (1970), Sabkha brines free of sulfur are expelled by sediment compaction, migrate, and become enriched in base metals until they contact a zone of higher temperature and pressure. In this zone, sulfides are present as a result either of inorganic reduction of sulfate, anaerobic reduction, or hydrocarbon reduction of anhydrite. The sulfides cause the base metals to precipitate. Formation brines are the medium in which several metals, in addition to hydrocarbons, migrate prior to deposition in ore deposits. A current theory is that the metals travel primarily as chloride complexes in solutions that are depleted in reduced sulfur species (Dunham, 1970). The metals subsequently are precipitated when a source of reduced sulfur is met. An example of a source of reduced sulfur is an area where anaerobic bacteria are reducing sulfate. This occurs in waters near petroleum-bearing formations, and such waters in carbonate reservoirs often contain considerable amounts of sulfide.

RESEARCH STUDIES

225

Brine classification
A study of the evolution of subsurface brines in Israel by Bentor (1969) led him to classify brines into four groups. The first group consists of brines similar t o sea water except for an increased concentration of calcium and a decreased concentration of magnesium, which he attributed to dolomitization. The second group was similar t o sea water but contained two to three times higher concentrations of dissolved salts, deficient in sulfate and magnesium, and enriched in bromide and iodide. The sulfates were lost by organic reduction, the magnesium was lost by exchange reactions with clays, and bromide and iodide were added by organic sources. The third group was a high-salinity calcium chloride-type brine formed by surface evaporation and later modified in the subsurface by differential ultrafiltration, The fourth group was a highly saline, calcium chloride type with Ca/Na ratios greater than one. This group was divided into two subgroups where the first subgroup is a highly saline and highly differentiated Early Paleozoic brine, while in the second subgroup they are old Paleozoic brines which were submitted t o an additional cycle of surface concentration by evaporation.

Ion association Truesdell and Jones (1969) studied ion association in brines and found that, except for the chloride ion, the major simple ions form ion pairs, while the minor and trace metals in brines form coordination complexes. Selective ion electrodes can be used to determine directly the ionic activities of sodium, potassium, chloride, fluoride, and sulfide in brines. Experimental data were used t o calculate chemical models for ion association and coordination complexes in brines. These models are useful in explaining the chemical behavior of brines. Relation t o lithology Kramer (1969) used factor analysis to study the relationships of the brines to the type of rock from which they were taken. His results indicated that the major ions in most brines are sodium, calcium, and chloride; brines are enriched in calcium and bicarbonate and are deficient in magnesium and sulfate relative t o sea water. The factor groupings did not reflect the lithology of the rocks from which the brines were taken, indicating that such a relationship does not exist or is difficult to detect. The brine analyses used in the study were primarily macro analyses and did not include pH, minor, or trace constituents. A study of this type would benefit significantly if the following conditions were met: (1) use only the best available sampling use methods; (2) use field analysis techniques; (3) positive lithology identification; and (4) use only the best available laboratory methods of analysis.

226

ORIGIN OF OILFIELD WATERS

Kramer (1969) did not have these controls because he used only the published data of various laboratories. Carpenter and Miller (1969) used statistical and thermodynamic methods in an effort t o determine the origin of the dissolved chemical constituents in saline subsurface waters in north-central and northwestern Missouri. Statistical analysis of scatter diagrams indicated that the concentrations of lithium, sodium, potassium, and bromide and the ion activity ratios of K+/H+, Ca+/Mg+*, and Sr+ /Ba+ in the waters are influenced by reactions with constituents in the aquifer rocks. They concluded that the ion ratios are of little value in determining the origin of the waters because the concentrations of the dissolved constituents in the waters had reacted with minerals in the aquifer rocks. This study was excellent because it did show that the concentrations of constituents in the water are controlled to some extent by reactions with the aquifer rocks. Additional work of this kind is needed in the study of deep brines. A study of brines from the Sylvania formation in the Michigan Basin indicated that evaporation and dolomitization were two dominant controls for their dissolved concentrations of calcium, magnesium, sodium, strontium, and bromide (Egleson and Querio, 1969). Mechanisms responsible for concentrations of elements such as potassium, lithium, rubidium, ammonia, boron, and iodide were believed t o be reactions with sedimentary rocks, leaching of organic constituents, and bioconcentration.

Relation to depth and salinity

A study of the chemical composition of some selected Kansas brines indicated that in general the concentrations of calcium, sodium, and chloride increase with increasing salinity, while the sulfate concentrations decrease (Dingman and Angino, 1969). However, the Ca/C1 ratio, concentrations of calcium and salinity, did not generally increase with geologic age or with depth of the aquifer. Dickey (1969) surveyed the analyses of oilfield waters from many areas of the United States and concluded that in general the Ca/Mg ratio increases with increasing salinity while the ratio Na/(Ca + Mg) decreases irregularly with increasing. salinity and depth. This observation is compatible with the findings of several investigators. The dominant anion in subsurface waters usually changes with depth; in near-surface waters, it is sulfate; at depths exceeding 520 m it is bicarbonate; and in deep brines, it is chloride (Chebotarev, 1964). The Ca/Na ratio usually increases with depth and age of the associated rocks, while the Mg/Na ratio decreases.
Iodide
Collins (1969a) studied the chemistry of some oilfield brines from the Anadarko Basin which contain high concentrations of iodide. The concen-

RESEARCH STUDIES

227

trations of bromide in many of these brines are lower than the iodide which is unusual. Localized sedimantary rock deposits enriched in organic iodine are the source of the high iodide concentrations in these brines (Collins et al., 1971). Hot brines Hot brines containing minor and trace amounts of several metallic elements in addition to macro concentrations of some alkalies, alkaline earths, and chloride are found in drill holes in southern California, in the Caspian Sea, and in deeps in the Red Sea. These brines were formed from evaporites dissolved by meteoric water, and the metallic elements were leached from country rocks by the hot brines (Tooms, 1970). Laboratory reactions of 2M and 4M sodium chloride with andesite and shale at 300"-500C have produced solutions containing metallic elements in concentrations similar to the hot brines (Ellis, 1968). Comparison of oilfield brines with evaporated sea water Bromide does not form its own minerals when sea water evaporates. It forms an isomorphous admixture with chloride in the precipitates (Valyashko, 1956; Braitsch and Herrmann, 1963). As sea water evaporates, the carbonates precipitate first, followed by the sulfates. Little or no bromide precipitates, or if it does, it is occluded with these. Halite (NaCl) begins t o precipitate when the chloride concentration is about 275,000 mg/l (Table 7.111) compared with that of normal sea water, 19,000 mg/l. Some bromide is entrained with chloride in the precipitate.
300

2oo

Normal evaporite curve/

.c
C

Fig. 7.4. Use of the bromide ion to differentiate some Tertiary (T), Cretaceous (C), and Jurassic (J) age brines.

228

ORIGIN OF OILFIELD WATERS

TABLE 7 .XV Bioconcentrated bromide and iodide in seaweeds and corals Bromide (ppm) Seaweed Laminaria digitata (dry matter) Laminaria saccharina (dry matter) Desmaresta (ash) Iodide (ppm)

1,380 340 6,800

5 10-8,000 2,000 5,200

Corals* Gorgonia uerrucosa Gorgonellidae Isididae

16,200 19,800 7,400

69,200 22,100 20,300

* After Vinogradov (1953).


However, with each crystallization, more bromide is left in solution than is entrained in the precipitate. Sylvite (KC1) begins to precipitate when the chloride concentration is about 360,000 mg/l (Table 7.111), followed by carnallite (MgC12*KCI *6H20 ) and bischoffite (MgC12 *6H, 0).During evaporation the concentration rate of bromide in solution increases. The change in the slope of the curve in Fig. 7.4 illustrates this approximately. Other concentration mechanisms operate t o account for the high bromide concentrations (6,000 mg/l and up) found in some brines. One of the mechanisms is related to bioconcentrators such as seaweeds and corals. The seaweeds and corals concentrate the bromide, they die, and are buried with the sediments. Later the bromide is leached by the surrounding waters. Table 7.XV illustrates some of the concentrations of bromide and iodide that Vinogradov (1953) found in various seaweeds and corals. Laboratory experiments have demonstrated that bromide is accommodated in the halite crystal lattice and replaces chloride in solid solution (Borchert and Muir, L964). The weight percentage of bromide in solid solution in the halite lattice is related to its weight percentage in the parent brine as :

C = wt.% Br (in solution)


where C = the pa@ition coefficient. In most natural environments C = 0.14 (Braitsch and Herrmann, 1964). In a marine salt sequence the wt.% Br/NaCl rises from about 0.007 wt.% at the bottom to 0.02 wt.% at the beginning of potassium precipitation. However, the bromide concentration with a given natural halite sequence may vary considerably, even though theoretically it should increase continuously

wt.% Br (in halite)

RESEARCH STUDIES

229

from the bottom t o the top of the depositional strata. These variations can be attributed t o inflow of fresh sea water during the deposition or subsequent leaching after deposition. Rittenhouse (1967) developed a method to classify oilfield waters based upon the bromide concentrations. Fig. 7.4 is a log-log plot of chloride versus bromide concentrations for some Louisiana oilfield waters. The T, C, and J on the figure refer t o Tertiary, Cretaceous, and Jurassic, indicating the ages of the rocks from which the waters were taken. The normal evaporite curve was plotted by using data from Table 7.111. The data in the figure indicate that most of the Tertiary waters are deficient in bromide when compared to an evaporite water, whereas the Cretaceous and Jurassic waters are enriched in bromide (Collins, 1967). The Tertiary waters contain dissolved halite, which accounts for their low bromide concentration, while the waters that are enriched in bromide contain bitterns or have leached bromide from sediments that were enriched in bioconcentrated bromide. The bromide content of oilfield brines can be used t o distinguish between brines that originated because of evaporation of sea water and those formed by the dissolution of evaporite minerals. This can be done by using Fig. 7.4. If the bromide concentration falls to the right of the normal evaporite curve, the brine contains evaporated sea water, while if it falls t o the left of the curve, it contains dissolved evaporite minerals. Fig.7.5 illustrates how closely the concentration of sodium of some Louisiana oilfield waters taken from Tertiary, Cretaceous, and Jurassic age rocks follow the sodium concentration of a brine associated with normal evaporation (Collins, 1970).
300
200
0 ,

- 100
W

Normal evaporite curve

50
0
J

20

lo

5,000

10,000

20,000

SODIUM, m g / l

'

&bob I llob!OOO

'

1
500 300

Fig. 7.5. Relationships of the chloride concentrations to sodium concentrations in a normal evaporite brine to oilfield brines taken from formations of Tertiary (T), Cretaceous (C), and Jurassic (J) age in the United States,

23 0

ORIGIN OF OILFIELD WATERS

Ion exchange
Ion exchange reactions on clay minerals are reversible and they follow the law of mass action. The number of exchange sites governs the reaction, and other important factors include temperature, pressure, solution concentrations, and bonding strength of exchangeable ions. Ion exchange between clay minerals and a brine will stop when equilibrium is attained. As the waters move in their subsurface environment, their dissolved ions have a tendency to exchange with those in the rocks. There are two extreme types of adsorption in addition to intermediate types of adsorption. The extreme types are: (1) a physical adsorption or Van der Waals adsorption with weak bonding between the adsorbent and the constituent adsorbed; and (2) a chemical adsorption with strong valence bonds. Cations can be fixed at the surface and in the interior of minerals. These fixed cations can exchange with cations in the water. Under the right physical conditions of the adsorbent, similar exchange can occur with the anions. Some of the constituents in formations that are capable of exchange and adsorption are argillaceous minerals, zeolites, ferric hydroxide, and certain organic compounds. Particle size influences the exchange rates and capacities if the solids are clays such as illite and kaolinite. The rate increases with decreasing particle size. However, if a larger mineral has a lattice, the exchange can easily occur on the plates. The concentration of exchangeable ions in the adsorbent and in the water is important. More exchange usually occurs when the solution is highly concentrated. According t o Grim (1952), the replacing power of some ions in clays is:

(1) In NH, ,kaolinite:


Cs > Rb > K > Ba > Sr > Ca > Mg> H > Na > Li

(2) In NH, ,montmorillonite:


Cs > Rb > K > H > Sr > Ba > Mg> C a > Na > Li

These two clays often are present in sedimentary rocks and the replacing order indicates that lithium and sodium are more likely to be left in solution, while cesium and rubidium are more likely t o be removed from solution. Fig. 7.6 is a plot of the chloride content versus the lithium content of some oilfield waters taken from the Smackover formation. The .lithium enrichment results at least in part from exchange reactions on clays. Lithium has a small radius, a low atomic number, a larger hydrated radius than sodium, and a larger polarization than sodium. Because of these, its replacing power in the lattices of clay minerals is low (Kelley, 1948). Other ions such as barium, strontium, calcium, magnesium, cesium, rubidium, potassium, and sodium will preferentially replace lithium in clay minerals, thus releasing lithium t o solutions. Furthermore, the solubility products of most lithium

RESEARCH STUDIES
1,000 000 C

23 1

600

400 -

Louisiana Mississippi Alabama Arkansas Texas


0 0

40 20d

0 0

1 0

O I , I

300 200

- 100
m

2
0

50 30 20

C- J
T

,I
19
3

T
200

1 I II,]

1 1 1 I I I L
5,000
0 10 1 0

100

500 1,000 2,000 POTASSIUM, m g / I

Fig. 7.7. Relationships of the concentrations of chloride and potassium in a normal evaporite-formed brine to oilfield brines taken from formations of Tertiary (T), Cretaceous (C), and Jurassic (J)age in the United States.

compounds are higher than those of other alkalies and alkaline earths. Therefore lithium tends to stay in solution. Fig. 7.7 compares the potassium concentration of some Louisiana oilfield waters with those of waters subjected to evaporation. All of these waters are depleted in potassium with respect to a brine subjected t o evaporation, indicating that potassium was lost to the associated sediments during

23 2

ORIGIN OF OILFIELD WATERS

diagenesis. It has been shown that a tendency exists for potassium t o be adsorbed and fixed by clay minerals, mica, and potassium feldspar in normal low-temperature processes (White, 1965; Khitarov and Pugin, 1966; Grim, 1952). The data in Tables 7..III-XIV indicate that the concentration of calcium in oilfield waters generally is enriched relative t o sea water. Cation exchange reactions with clays accounts for some of this enrichment: 2Na+ (solution) + Ca (clay) + Ca+* (solution) + 2Na (clay) Collins (1972) found that the ratio Na/(CL + Mg) tends t o decrease as the dissolved solids concentration increases in some oilfield waters from the East Texas Basin. This depletion of sodium with respect to calcium plus magnesium was attributed to diagenesis of the waters and it correlated with an index of base exchange (Schoeller, 1955), indicating that the alkali metals in the waters exchanged with alkaline earth metals on the argillaceous minerals t o decrease the dissolved alkali metals and increase the dissolved alkaline earth metals. Fig. 7.8 is a plot of the calcium concentration in some oilfield waters taken from the Smackover formation. All of these waters are enriched in calcium relative to the evaporated sea water. Krejci-Graf (1963) found that solutions predominantly concentrated in chloride can force an exchange of calcium and bromide from clay minerals for sodium and chloride from the solution. If this type of reaction occurred

Fig. 7.8. Relationships of the concentrations of chloride t o calcium in an evaporiteformed brine to oilfield brines taken from the Smackover formation in five states of the United states.

RESEARCH STUDIES
1,000 000 600 -

233

400 -

A
0 0

- 200 \ 0

Louisiana Mississippi Alabama Arkansas Texas

0 100,

5
V

80
60 40

20
101I<I
I

10 0

200

400

I 1 1 1 1 1 1

' I ' I ' l ' ' ' 1 ' 1,000 2,000 4,000 10,000

40,000

'

BROMIDE, mg/l

Fig. 7.9. Relationships o f the concentrations of chloride t o bromide in an evaporiteformed brine t o oilfield brines taken from the Smackover formation in five states of the United States.

t o the Smackover brines, it explains their enrichment of calcium and bromide. Kozin (1960) wrote about a "reverse" exchange of anions when the cations exchange on clays: C1- (solution) + Br (clay) + Br- (solution) + C1 (clay) Such a reaction also helps to account for the bromide enrichment found in most oilfield waters taken from the Smackover formation (Fig. 7.9). A similar reaction for iodide:

, C - (solution) + I (clay) -I- (solution) + C1 (clay) 1


would help explain the tremendous enrichment of iodide in oilfield brines (Collins, 1969a) with respect t o sea water as demonstrated in Tables 7.1V-XIII. Fig. 7.10 shows that boron usually is enriched relative to the normal evaporite curve in Smackover oilfield brines. Boron, like lithium, has a small radius, a low atomic number, and large polarization. Therefore, its replacing power in the lattices of clay minerals is low. Also, boron does not have a tendency to enter silicate lattices of the common rock-forming minerals. Because of these factors, it usually remains in solution until late-stage crystallization.

234
1,000 -

ORIGIN OF OILFIELD WATERS

000 600 -

400-

A
0

:
LL

- 200 w
100:

Louisiana Mississippi Alabama Arkansas Texas

006040 20 -

BORON, mg/l

Fig. 7.10. Relationships of the concentrations of chloride to boron in an evaporiteformed brine to oilfield brines taken from the Smackover formation in five states of the United States.

Mineral formation
A study of some brines taken from Devonian age reservoir rocks indicated that dolomitization probably is the most important mechanism in determining the calcium, strontium, and magnesium content of these brines (Egleson and Querio, 1969). It also was concluded that the relative amounts of ammonium, iodide, and lithium in these brines were too high to be derived directly from sea water, and the ammonium and iodide probably were enriched in the brines as a result of bioconcentration and subsequent leaching of organic debris. Dolostone deposits owe their origin t o hypersaline brines (Friedman and Sanders, 1967). Some dolomite, including diagenetic and epigenetic forms, originates from subsurface brines. In the geologic columns in several oil-

TABLE 7.XVI Approximate sea-water composition before and after gypsum precipitation (mg/l)

In o
Calcium Magnesium Bromide Sulfate

Before precipitation

After precipitation

390 1,300 65 2,580

0 1,300 65 2,580

RESEARCH STUDIES TABLE 7.XVII Approximate sea-water composition after dolomitization or bacterial reduction (mg/l) Ion Calcium Magnesium Bromide Sulfate After dolomitization After bacterial reduction

23 5

0 883 65 0

0
1,300

65 0

productive basins, mixtures of dolomite and anhydrite occur, which indicates that sulfate may have been removed from the associated waters by dolomitization as well as by bacterial reduction. Table 7.XVI illustrates the approximate amounts of calcium, magnesium, bromide, and sulfate that could exist in a waterbefore and after precipitation of gypsum. Assuming that the residual sulfate (1,644 mg/l) was removed by the dolomitization reaction: MgC12 + 2CaC0, += CaC12 + CaMg(C0, )* - CaC12- + MgS04- += CaS04 + MgCi2 MgSO, + 2CaC03 + CaS04 + CaMg(C0, ) 2 then the Mg/Br ratio would be about 883/65 = 13.6, as illustrated by the data in Table 7.XVII. However, if the residual sulfate was removed by bacterial reduction : C,H,

+ Na2 SO4

Na2C03 + H 2 S + C 0 2 + H 2 0

the Mg/Br ratio would be about 1300/65 = 20. Magnesium will react with CaC03 (calcite) t o form dolomite, thus increasing the concentration of calcium in the brine. However, the total calcium plus magnesium in the brine should remain constant. This can be calculated as (24.31/40.08) x mg/l calcium + mg/l magnesium = total equivalent magnesium or Mg. The ratio Mg/Mg will vary, depending upon the availability of calcite, and the ratio should be indicative of the degree of dolomitization. For example, brines that are in equilibrium with sandstones should have a relatively low Mg/Mg ratio, those in equilibrium with dolomite should have higher ratios, and those in equilibrium with limestone should have the highest ratios. The average ratio for some Smackover brines is 7 (Table 7.XVIII), which indicates that the brines were in equilibrium with limestone and dolomite. Brines from some Tertiary age rocks which were primarily

236

ORIGIN OF OILFIELD WATERS

TABLE 7.XVIII Concentration ratios and excess factor ratios for some constituents in Smackover brines Constituent Average composition (mg/l) sea water Lithium Sodium Potassium Calcium Magnesium Strontium Barium Boron Copper Iron Manganese Chloride Bromide Iodide Sulfate Mg'*
0.2 10,600 380 400 1,300 8 0.03 4.8 0.003 0.01 0.002 19,000 65 0.05 2,690 1,543

Concentration ratio*'

Excess factor*'

Smackover brines
17 4 66,975 2,841 34,534 3,465 1,924 23 134 1.1 41 30 171,686 3,126 25 446 24,362 870 6 8 86 3 241 767 28 359 4,049 14,957 9 48 501 0.2 16 18.1 0.1 0.2 1.8 0.1 5 16 0.6 7.5 84.2 31 1 0.2 1 10.4 0.003 0.3

Number of Smackover samples


71 283 82 284 280 85 73 71 64 90 69 284 74 73 27 1 284

*' Amount in brine/amount in sea water.


*3

** Concentration ratio of a given constituent/concentrationof bromide.


Mg' = (24.31/40.08) x mg/l Ca + mg/l Mg.

sandstone (Table 7.XIX) had an average ratio of 2.8, while brines from some Cretaceous age rocks had an average ratio of 6.0 (Table 7.XX). Bromide does not form its own minerals when sea water evaporates. Some of it is lost from solution because it forms an isomorphous admixture with chloride with the halite precipitate. However, more bromide is left in solution than is entrained in the precipitate. Therefore, relative t o chloride, the bromide concentration in the brine increases exponentially. Bemuse of this, the bromide concentration in the brine is a good indicator of the degree of sea water concentration, assuming that appreciable quantities of biogenic bromide have not been introduced. Table 7.XVIII presents data that were obtained by comparing the average composition of some Smackover brines with that of sea water. The concentration ratio was calculated by taking the mean average for a given constituent in the Smackover brines and dividing it by the amount of the constituent found in normal sea water. The excess factor was determined by dividing the concentration ratio of a constituent by the concentration ratio of bromide. The calculation for Mg' or total equivalent magnesium was previously explained, and the number of Smackover samples indicates how

RESEARCH STUDIES TABLE 7.XIX

237

Concentration ratios of some constituents in some brines taken from Tertiary age rocks Constituent Average composition (mg/l) sea water Lithium Sodium Potassium Calcium Magnesium Strontium Barium Boron Chloride Bromide Iodide Sulfate Mg' Tertiary brines Concentration ratio*' Excess factor *

'

0.2 10,600 380 400 1,300 8 0.03 4.8 19,000 65 0.05 2,690 1,543

3 37,539 226 2,077 686 148 73 20 63,992 79 21 104 1,947

15 3.5 0.6 5.2 0.5 18.6 2,439 4.1 3.4 1.2 426 0.03 1.3

12.5 2.9 0.5 4.3 0.4 15.5 2,033 3.4 2.8


1

355 0.03 1.1

*'

*' Amount in brine/amount in sea water.


*3

Concentration ratio of a given constituent/concentration brymide. of Magnesium e,quivalent of calcium plus magnesium in brine: Mg = (24.31/40.08) x mg/l Ca + mg/l Mg .

TABLE 7.XX Concentration ratios of some constituents in some brines taken from Cretaceous age rocks Constituent Average concentration (mg/l) sea water Lithium Sodium Potassium Calcium Magnesium Strontium Barium Boron Chloride Bromide Iodide Sulfate Mg'*3 Cretaceous brines Concentration ratio*' Excess factor*'

0.2 10,600 380 400 1,300 8 0.03 4.8 19,000 65 0.05 2,690 1,543

4 28,462 193 4,999 606 346 48.3 27.5 54,910 287 37 206 3,643

20 2.7 0.5 12.5 0.5 43.3 1,608 5.7 2.9 4.4 737 0.1 2.4

4.5 0.6 0.1 2.8 0.1 9.8 365 1.3 0.7

1
168 0.02 0.5

*' Amount in brinelamount in sea water.


*'

*'

Concentration ratio of a given constituentlconcentrationof brymide. Magnesium equivalent of calcium plus magnesium in brine: Mg = (24.31/40.08) x mg/l Ca + mg/l Mg

23 8

ORIGIN OF OILFIELD WATERS

many samples were used in the calculation. For example, 71 Smackover brines were analyzed for lithium, while 283 were analyzed for sodium. The concentration ratios (Table 7.XVIII) indicate that all of the determined constituents in the Smackover brines were enriched with respect to sea water except sulfate. However, the excess factor ratios indicate that sodium, potassium, magnesium, chloride, sulfate, and total equivalent magnesium were depleted in the Smackover brines, while lithium, calcium, strontium, barium, copper, iron, manganese, and iodide were enriched. Further, these ratios indicate that the Smackover brines have been altered considerably if it is assumed that they originally were sea water. The concentration ratio 48 for bromide (Table 7.XVIII) is one of the highest that this author has seen. For example, bromide concentration ratios of 1.2 (Table 7.XIX), 4.4(Table 7.XX), and 8.8 and 7.2 were found for brines from Tertiary, Cretaceous, Pennsylvanian, and Mississippian age rocks (Collins, 1967, 1969a, 1970). The concentration ratios and excess factors in Tables 7.XIX and XX indicate several constituents are enriched and several are depleted in these brines also. Almost one-third of the magnesium in sea water and subsequent bitterns can be removed during the dolomitization reaction. The formation of chlorite from montmorillonite requires about 9.2 moles of MgO per mole of chlorite (Eckhardt, 1958):

1.7 A1203 0.9 MgO

8 Si02

- 2 H 2 0 + 9.2 MgO + 6 H 2 0
1.7 A1203 6.4 Si02
*
a
A
0

+ =

10.1 MgO
1,000 -

8 H 2 0 + 1.6 Si02

800 600 400

o a

Louisiana Mississippi Alabama Arkansas Texas

cn

200 100:

* /

W *

g
5
I
0

8060 40 0

Normal curve

evaporite

20 1 III 0
I

1 1 6 1

1 1 8 1

00

Fig. 7.11. Relationships of the concentrations of chloride to magnesium in an evaporiteformed brine to oilfield brines taken from the Smackover formation in five states of the United States.

RESEARCH STUDIES
1,000

23 9

000 600

400

Louisiana Mississippi Alabama Arkansas Texas

\ 0

2 00
000 4 0 ,

: 00
2
0

100

60 40
20
0

00

Fig. 7.12. Relationships of chloride to strontium in an evaporite-formed brine to oilfield brines taken from the Smackover formation in five states of the United States.

Such a reaction could remove large amounts of magnesium from waters. Hiltabrand (1970) has shown that contemporary argillaceous sediments can remove 100 mg/l of magnesium from sea waters. Fig. 7.11 is a plot of the chloride concentrations versus the magnesium concentrations in some oilfield waters taken from the Smackover formation. The figure indicates that the Smackover waters are depleted in the concentration of magnesium with respect t o an evaporite-formed brine. Tables 7.111-XI11 indicate that in general oilfield waters taken from rocks of other formations also are depleted in magnesium. The data also show that generally as the dissolved magnesium decreases the dissolved calcium increases. This is related t o the formation of minerals such as chlorite or dolomite and t o exchange reactions with argillaceous minerals. It is not a result of solubility because most magnesium compounds are more soluble than calcium compounds. Fig. 7.12 is a plot of the concentration of chloride versus the concentration of strontium found in some oilfield brines taken from the Smackover formation. This figure indicates that the strontium concentration is enriched in the Smackover brines relative to sea water. Reactions that account for some of this enrichment are: 2SrC03 + MgCl, +. SrMg(C03), + SrCl, , SrMg(C0, ) + MgCl, +. 2MgC0, + SrC1, The data in Fig. 7.7 and Tables 7.111-XIV indicate that the concentration of potassium in oilfield waters generally is depleted relative to sea water. Montmorillonite-type minerals systematically change t o illite with depth

24 0

ORIGIN OF OILFIELD WATERS

in Gulf Coast shales (Burst, 1969). As a result of this transformation, the montmorillonite-type minerals lose interlayer water. Laboratory experiments at elevated temperatures and pressures indicate that montmorillonite loses its interlayer water and transforms into illite in the presence of potassiumenriched water (Khitarov and Pugin, 1966). The structural variations of the expandable minerals in clays also are apparently influenced by the potassium content of the associated waters. This indicates that oilfield waters tend to become depleted in potassium content where this reaction occurs. Reactions between brines and minerals to form silicates that account for the depletion of dissolved alkali metals are: 3A1, Si2O5(OH), + 2K+ 2KA13Si, O l 0(OH), + 3H20 + 2H+ KA13Si030,0(OH)2+ 6Si02 + 2K+ + 3KA1Si308 + 2H+ A12S i 2 0 5 (OH), + 4sio2 + 2Na+ =+2NaA1Si308 + H2 0 + 2H+ These reactions account not only for the depletion of potassium or sodium, but also for a decrease in pH because of the release of hydrogen ions. The decrease in pH enables the water t o dissolve metallic metals, to convert bicarbonate to carbon dioxide, or to convert bisulfide t o sulfide. The Smackover brines often contain relatively high concentrations of sulfide. Several investigators have attempted t o determine what mechanism is responsible for the increased concentration of calcium and depletion of magnesium relative to sea water in many subsurface brines. Chave (1960) and Von Engelhardt (1960) compared ocean water with subsurface brines containing high concentrations of calcium, and demonstrated that dolomitization cannot account for all of the calcium in the brine solutions. Von Engelhardt (1960) noted that even the formation of chlorite utilizing magnesium with exchange of sodium and calcium does not account for all of the soluble calcium; however, exchange reactions with other clays were not considered. Kramer (1963) assumed that calcium was more abundant in ancient oceans, but White (1965) found this relation t o be untenable and suggested that shale-membrane filtration accounts for increased concentrations of calcium in some brines. Additional data are needed before more definite conclusions can be made. The amounts and ratios of calcium and magnesium vary from one formation water to another as well as within one formation at different geographic areas. Mineral formation, exchange reactions, leaching, and shale-membrane filtration all can alter the composition of the brine. However, in a specific area, one type of reaction may predominate. Mem brane-concen tra ted brines Essentially the postulate that clays and shales act as membranes t o filter dissolved solids from waters results from the fact that synthetic membranes are used to desalinate waters by reverse osmosis. Conceivably, compacted clays and shales may perform as imperfect semipermeable membranes. Solu-

RESEARCH STUDIES

241

tions of salts of different concentrations separated by a semipermeable membrane will cause water from the lower salt-concentration side t o move through the membrahe t o the higher concentration side, producing a greater pressure on the high-concentration side. The pressure differential is the osmotic pressure of the system and can account for abnormal pressures found in some reservoirs. Reverse osmosis occurs when hydraulic pressure in excess of the osmotic pressure is applied t o the high-concentration side, which forces water through the membrane to the low-concentration side. The system is not 100% effective and some dissolved solids move through the membrane (Kimura and Souriragan, 1967). Such a system requires rather high pressure differentials in nature t o produce the highly concentrated brines found in some formations. The osmotic pressure could produce pressure differentials in formations, but the pressure comes t o equilibrium as the two solutions equilibrate. The reverse osmosis system works only as long as the excess hydraulic pressure is applied. In the absence of the excess hydraulic pressure, the system comes t o equilibrium. Larson (1967) reported some desalination results for water with reverse osmosis using cellulose-acetate membranes. With a brackish water containing about 4,300 mg/l of dissolved solids, input pressure of 42 kg/cm2 and temperature of 15.g0C, the ion rejection rates were as high as 99.9%. The rejection order based on the percent rejected was: Ca+ + Mg+

> HC03-2

+ SO4-

> C1->

Na+ > NO3-

Assuming that this mechanism operates in a shale filtration system, the order of ion concentration on the high brine concentration side would be the same. The ion concentrations on the fresher water side would be the reverse or : NO3-

> Na+ > Cl- > SO4-

+ HC03-*

> Ca+ + Mg+

Other investigators have obtained similar results. For example, Loeb and Manjikian (1965) found a rejection order of SO4- > Mg+ > Ca+ > Na+ > HC03- > C1- > NO3-. Michaels et al. (1965) found a rejection order of Ca+2 > Li+ > Na+ > K+ for the pressure independent portion of salt transport in cellulose acetate reverse osmosis desalination membranes. This correlates with the size of the hydrated ion radii because calcium is the largest and potassium the smallest. Further, this indicates that the pore size of the membrane is a controlling factor. The data of Larson (1967) showed that sulfate and carbonate scale formed on the high-pressure side of the membrane and if not removed would cause flow to decrease or stop. The pH on the output or fresh-water side of the membrane decreased.

242

ORIGIN OF OILFIELD WATERS

Russell (1933)considered several processes which could produce subsurface brines more concentrated than sea water. He concluded that evaporation of the water by natural gas generally is not important, water evaporation in coarse-grained rocks generally is not important, gravitational settling of dissolved solids is not greatly important, rocks containing considerable amounts of feldspars and other unstable minerals take up large quantities of hydration water, clays adsorb bases and later expel them into solution causing concentration, and osmosis may occur through semipermeable membranes. DeSitter (1947)noted that oilfield waters are altered as a result of two prominent diagenetic phases. During the first phase magnesium, calcium, sulfate, and carbonate precipitate from the original sea water. During the second phase the concentration of magnesium and calcium ions increases along with the concentration of other dissolved solids. He reasoned that the second phase occurred because of filtration through semipermeable shales. The filtration results because of sediment compaction until a semipermeable membrane develops which allows water molecules t o pass through but retards salt ions. Thus, the more concentrated brines are found where sediment compaction and water flow distance were the greatest. This usually occurs in the deepest portion of a basin. McKelvey et al. (1957) forced aqueous saline solutions through ionexchange resins and found that the effluent solutions contained less dissolved salts than the influent solutions. Effluents from cation-exchange resins were found to contain Na/K ratios similar t o those in the influent; however, the Mg/Ca ratios were at first higher than in the influent but with additional squeezing the ratio decreased t o much lower values. They postulated that similar reactions occur during the compaction of sediments to change the concentrations of constituents dissolved in waters. Pressures of 7 kg/cm2 to 105 kg/cm2 were applied t o force sodium chloride solutions through cation-exchange membranes. The results indicated that the membranes desalted the saline solutions, producing a filtrate containing less salt than the influent. This salt filtering effect was attributed t o the electrical properties of the membrane. Milne et al. (1964)determined the filtering efficiencies of sodium chloride solutions by bentonite membranes. The filtration efficiencies were 94% at 140 kg/cm2 and 88% at 703 kg/cm2 with 0.5N sodium chloride. Increased salinity caused less efficient filtration because filtration efficiencies of 94% for 0.W sodium chloride and 66% for 4 sodium chloride at a pressure of N 352 kg/cm2 were obtained. A similar mechanism could operate in the subsurface to create concentrated brines. Young and Low (1965)performed an experiment using natural rock and demonstrated that osmotic flow of water through shale and siltstone occurs. The osmotic pressures produced were less than theoretical and they were attributed to microcracks in the natural rock which caused them t o be less effective than a perfect membrane.

RESEARCH STUDIES

243

Bredehoeft et al. (1963) developed a mathematical model t o predict the distribution of ions within a formation. They assumed that a hydrostatic head differential opetates between the margin and center of a geologic basin, producing a water movement upward through confining low permeability beds. If these low permeability beds contain clay membranes to restrict the passage of ions, the waters on the upflow, or more permeable, side become more concentrated in dissolved solids. They theorized that this process produced the concentrated brines found in the Illinois Basin, and that their model added weight t o the membrane theory of brine concentration. A major drawback to the model is the tremendous pressures that are necessary to produce a movement of water upward through confining low permeability beds. Graf et al. (1965) found that isotopic fractionation occurred when waters passed through shale micropores in the Illinois, Michigan, Alberta, and Gulf Coast Basins. Their study did not yield sufficient evidence t o estimate the total fraction of water movement in the basins subsequent t o sediment compaction. The 6 " 0 concentrations in brines did not indicate a direct correlation with ancient oceans. A study of the 6D and 6l80 in formation waters indicated that the water was predominantly meteoric, little exchange or fractionation had occurred to alter the deuterium, but extensive exchange between the water and rock had altered the oxygen (Clayton et al., 1966). They postulated that formation waters in the Gulf Coast Basin lost their original connate water because of sediment compaction and flushing, and that the present water is meteoric water which came in through outcrops. This study was good; however, basic studies concerning the fractionation and exchange of isotopes between water, hydrocarbons, and rocks need to be made. Results of such studies should enable more positive interpretations. A simplistic model was derived t o determine the amounts of fresh water and sea water necessary t o create the brine compositions now present in the Illinois and Michigan Basins (Graf et al., 1966). The model assumes: (1) perfect efficiency of shale ultrafilters; (2) complete bacterial reduction of sulfate with replacement in solution of equivalent bicarbonate; (3) complete removal of bicarbonate and equivalent sodium by shale ultrafiltration; and (4) magnesium reaction with calcium carbonate to form dolomite. The dolomitization reaction furnished more soluble calcium than is possible for the Illinois Basin, so another calculation was made assuming complete loss of magnesium t o clay minerals with no return of calcium. The calculations indicated that less fresh water passed through the rocks of the Illinois Basin than those of the Michigan Basin. These data conflicted somewhat with Clayton et al. (1966) in that they argued that the water molecules now in the Illinois Basin originated as fresh water, while the data of Graf et al. (1966) indicated that too few volumes of fresh water passed through the Illinois Basin t o alter the brine significantly. A study of the hydrodynamics of the Illinois Basin indicated that in

244

ORIGIN OF OILFIELD WATERS

recent times, before pumpage, the differences in vertical head in the deep aquifers were insufficient t o cause upward flow through shale, resulting in ultrafiltration (Bond, 1972). In fact the head differentials were barely sufficient to enable upward flow through an open conduit. Berry (1969) outlined the relative factors that influence membrane filtration in geologic environments. The membrane properties of shales are caused by the electrfcal properties of their clays and organic materials. Clays predominantly are cation exchangers with singly charged SiO- and AlOSi-% sites and minor anion exchanges with replaceable OH- ions. Divalent cations are adsorbed in preference to monovalent cations and sodium is hyperfiltratcd with respect t o lithium and strontium with respect to calcium, because of preferential adsorption of ions with ionic potentials most similar t o the ionic potential of the exchange site. The selectivity of hyperfiltration for the halogens is C1 > Br > I > F because of their substitution for O H in the clays. Thus, in waters concentrated by this process the Ca/Na, Na/Li, Sr/Ca, Cl/Br, Br/I, and I/F ratios should increase. These ratio increases have been found in some brine systems, but by no means in all systems. Billings et al. (1969) found five types of formation waters in the Western Canada Sedimentary Basin and postulated the origin of two of the types. One type of water was formed by selective membrane filtration which produced waters containing high concentrations of dissolved solids. A second type was a mixture of membrane-concentrated formation water and bitterns formed after the precipitation of halite but before the precipitation of sylvite. They theorized that the alkalies were filtered selectively by clayshale membranes, producing a concentrated brine, and that the relative concentration pattern is Rb > K > Na > Li. This pattern is the reverse of what occurs by ion exchange but is similar to the surface mobilities of cations along clay surfaces. A detailed study of the Western Canada Sedimentary Basin, including a determination of the rock volume and pore volume (Hitchon, 1968), the effect of topography upon the fluid flow (Hitchon, 1969a), and the effect of geology upon the fluid flow (Hitchon, 1969b), strongly suggested that thermal, electro-osmotic, and chemico-osmotic forces are operating within the basin to affect the fluid energy gradients. Pressure differentials of about 98 kg/cm*along with salinity differences of 200,000 mg/l between formations in close proximity were found which suggest that chemico-osmotic forces are occurring. Hitchon and Friedman (1969) used chemical analyses and stable-isotope analyses for hydrogen and oxygen for surface waters, shallow ground waters, and deep ground waters in a study of the origin of formation waters in the Western Canada Sedimentary Basin. They postulated that surface waters have mixed with diagenetically altered sea water to form the formation waters. Using mass balance data for the deuterium and dissolved solid contents of the formation waters, they calculated not only how much fresh water is

CONCLUSIONS

245

present in the modified sea water but also observed how it redistributed the dissolved solids t o prQduce salinity variations. They concluded that formation waters result from mixing of surface waters with modified marine or nonmarine water in the subsurface rocks, that exchange of oxygen isotopes between the water and rock caused different water types in different basins, and that formation waters that have passed through shale ultrafilters are more depleted in deuterium. A study of the Surat Basin showed that most of its hydrocarbon accumulations are associated with quasi-stagnant waters. The salinities of these quasi-stagnant waters were higher than were the salinities of the waters in the more dynamic recharge areas. The investigators postulated that these high salinity waters were formed by membrane filtration because of crossformational flow and also that the hydrocarbon accumulations in these quasi-stagnant areas resulted from release of hydrocarbons mobilized by a moving water. The hydrocarbons were released because of the higher salinities of the waters in the quasi-stagnant areas (Hitchon and Hays, 1971). A study of waters in sedimentary rocks of Neogene age in the northern Gulf of Mexico Basin was made by Jones (1969). The hydrologic conditions currently found in these sediments are similar to conditions that previously occurred in older sedimentary basins. Osmotic flow has a dominant influence upon the hydrology of normally and abnormally pressured aquifer systems in the northern Gulf Basin. Jones (1969) found that many forces such as gravity, sediment diagenesis, different water salinities, ionic and molecular diffusion, different electrical potentials of sediments, thermal potentials, pressure, and osmotic membrane filtration affect the hydrology in this basin. Fowler (1970) found that salinity variations within the Frio sands in the Chocolate Bayou field, Brazoria County, Texas, are the result of selective concentration of ions by shales acting as membranes. In this field, pressures seem to reflect the flow paths of the waters, and the greatest changes in pressures are found across shaly sections. Analyses of water samples from this field over a 28-year period indicate decreasing salinity with production time caused by dilution of the original brines by waters squeezed from the shales adjacent to the aquifers. Chilingarian and Rieke (1969) reviewed the processes which can alter the chemical composition of formation waters. They concluded that most of the original water was sea water, and that the concentration process in many cases results from compaction and membrane filtration rather than evaporation. Their experimental results indicated that solutions squeezed out of rocks during compaction progressively decrease in dissolved solids concentrations with increasing depth.

Conclusions
The origin of oilfield waters is related to many natural processes. Initially,

246

ORIGIN OF OILFIELD WATERS

meteoric water reacted with weathered rock, soil, and organic matter. The excess waters that did not penetrate the rock or soil caused the rock and soil to erode and channels formed through which the water could move more easily. Forces of gravity caused the water to move from areas of high potential to areas of low potential, and as the waters moved, the concentrations of dissolved solids in them increased. Some of these waters found their way to lakes and the sea. As they entered the lakes or seas their movement slowed, causing some of the suspended particles in them t o deposit. Mixing of the waters with the more saline waters in the sea caused dissolved carbonate and organic compounds to precipitate. Evaporation of the sea and lake waters caused other compounds such as sulfates t o precipitate. The pH of the waters changed slightly because of reactions with the atmosphere, the sediments, and other waters. Each pH change caused precipitation of compounds or dissolution of new compounds. Some of the waters became highly concentrated in dissolved solids in the more shallow marine environments. Evaporites formed in these lagoons, pans, and exposed supratidal sabkhas. Evaporites also formed in deep-water basins when the salinity of the water at the bottom of the basin became sufficiently high. The sediments were buried as additional sediments were deposited on them, and water surrounding the sediment particles also was buried. As the depth of burial increased, the sediments compacted and some of the water was squeezed out. Both the squeezed-out water and the remaining interstitial water reacted with minerals in the sediments t o change the composition of the dissolved solids in the water and the composition of the sediments. Mechanisms that cause the oilfield waters t o differ in composition from water originally deposited with the sediments include ion exchange, infiltrating waters, sediment leaching, mineral formation, sulfate reduction, and ultrafiltration through clay-shale membranes.

References
Al'tovskii, M.E., Kuznetsova, Z.I. and Shvets, V.M.,1961. Origin of Oil and Oil Deposits (English Transl. by Consultants Bureau). Plenum Press, New York, N.Y., 107 pp. Anonymous, 1964. Chemistry of the oceans. Chem. Eng. News, 42:12A. Atwater, G.I. and Miller, E.E., 1965. The effect of decrease in porosity with depth on future development of oil and gas reserves in South Louisiana. Presented at Annual Meet., A m . Assoc. Pet. Geol., New Orleans, La., 1965 -Bull. A m . Assoc. Pet. Geol., 49:334. Ault, W.U., 1959. Isotopic fractionation of sulfur in geochemical processes. In: P.H. Abelson (Editor), Researches in Geochemistry. John Wiley and Sons, New York, N.Y., pp.241-259. Baas Becking, L.G.M., Kaplan, I.R. and Moore, D., 1960. Limits of the natural environn ment i terms of pH and oxidation-reduction potentials. J. Geol., 68 :243-284.

REFERENCES

24 7

Baker, E.G., 1960. A hypothesis concerning the accumulation of sediment hydrocarbons to form crude oil. Geochim. Cosmochim. Acta, 19:309-317. Barnes, H.L. and Czamanske, G.K., 1967. Solubilities and transport of ore minerals. In: H.L. Barnes (Editor), Geochemistry o f Hydrothermal Ore Deposits. Holt, Rinehart, and Winston, New York, N.Y., pp.334-381. Bentor, Y.K., 1969. On the evolution of subsurface brines in Israel. Chem. Geol., 4 :83-1 10. Berry, F.A.F., 1969. Relative factors influencing membrane filtration effects in geologic environments. Chem. Geol., 4:295-301. Billings, G.K., Hitchon, B. and Shaw, D.R., 1969. Geochemistry and origin of formation waters in the Western Canada Sedimentary Basin, 2. Alkali metals. Chem. Geol., 4:211-223. Birch, F., 1954. The present state of geothermal investigations. Geophysics, 19:645+359. Bond, D.C., 1972. Hydrodynamics in deep aquifers of the Illinois Basin. Ill. State Geol. Surv., Circ., No. 470, 7 2 pp. Borchert, H. and Muir, R.O., 1964. Salt Deposits - The Origin, Metamorphism, and Deformation o f Evaporite. D. Van Nostrand, London, 338 pp. Braitsch, 0. and Herrmann, A.G., 1963. Zur Geochemie des Broms in Salinaren Sedimenten, I. Experimentelle Bestimmung der Br-Verteilung in Naturlichen Salzsystemen. Geochim. Cosmochim. Acta, 27:361-391. Braitsch, 0 . and Herrmann, A.G., 1964. Zur Geochemie des Broms in Salinaren Sedimenten, 11. Die Bildungstemperaturen Primarer Sylvin- und Carnallit-Gesteine. Geochim. Cosmochim. Acta, 28 :1081-1 109. Bredehoeft, J.D., Blyth, C.R., White, W.A. and Maxey, G.B., 1963. Possible mechanism for concentration of brines in subsurface formations. Bull. A m . Assoc. Pet. Geol., 47 :257-269. Brod, I.O., 1960. On principal rules in the occurrence of oil and gas accumulations in the world. Int. Geol. Rev., 2:922-1005. Broecker, W.S. and Takahashi, T., 1966. Calcium carbonate precipitation on the Bahama Banks. J. Geophys. Res., 71 :1575-1602. Brongersma-Sanders, M., 1971. Origin of major cyclicity of evaporites and bituminous rocks: an actualistic model. Mar. Geol., 11 :123-144. Bruderer, W., 1956. Les ocbans souterrains fossiles et le phtrole. Assoc. Fr. Technol. Pet. Bull., 120:535-556. Burst, J.F., 1969. Diagenesis of Gulf Coast clayey sediments and its possible relation t o petroleum migration. Bull. A m . Assoc. Pet. Geol., 53:73-93. Bush, P.R., 1970. Chloride-rich brines from sabkha sediments and their role in ore formation. Inst. Min. Metall. Trans., 79:B137-B144. Butler, G.P., 1969. Modern evaporite deposition and geochemistry of coexisting brines, the Sabkha Trucial Coast, Arabian Gulf. J. Sediment. Petrol., 39 :70-89. Caraway, W.H. and Gates, G.L., 1959. Methods for determining water contents of oilbearing formations. U.S. Bur. Min. Rep. Invest., No.5451, 81 pp. Carpelan, L.H., 1957. Hydrobiology of the Alviso salt ponds. Ecology, 38:375-390. Carpenter, A.B. and Miller, J.C., 1969. Geochemistry of saline subsurface water, Saline County (Missouri). Chem. Geol., 4:135-167. Cartmill, J.C. and Dickey, P.A., 1970. Flow of a disperse emulsion of crude oil in water through porous media. Bull. A m . Assoc. Pet. Geol., 54:2438-2447. Case, L.C., 1955. Origin and current usage of the term "connate water". Bull. A m . Assoc. Pet. Geol., 39:1879-1882. Chave, K.E., 1960. Evidence on history of sea water from chemistry of deeper subsurface waters of ancient basins. Bull. A m . Assoc. Pet. Geol., 44:357-370. Chebotarev, I.I., 1964. Metamorphism of natural waters in the crust of weathering. Geochim. Cosmochim. Acta, 8:22-48, 137-170, 198-212.

248

ORIGIN OF OILFIELD WATERS

Chilingarian, G.V. and Rieke, 111, H.H. 1969. Some chemical alterations of subsurface waters during diagenesis. C h e m Geol., 4:235-252. Clark, S.P. and Ringwood, A.E., 1964. Density disturbance and constitution of the mantle. Rev. Geophys., 2:35-88. Clarke, F.W. and Washington, H.S., 1924. The composition of the earths crust. US. Geol. Surv. Prof.Paper, No.127, pp.1-112. Clayton, R.N., Friedman, I., Graf, D.L., Mayeda, T.K., Meents, W.F. and Shimp, N.F., 1966. The origin of saline formation waters, 1. Isotopic composition. J. Geophys. Res., 71 :3869-3881. Cloke, P.L., 1966. The geochemical application of E h - p H diagrams. J. Geol. Educ., 1 4:140-148. Collins, A.G., 1967. Geochemistry of some Tertiary and Cretaceous age oil-bearing formation waters. Environ. Sci. Technol., l :725-730. Collins, A.G. 1969a. Chemistry of some Anadarko Basin brines containing high concentrations of iodide. C h e m Geol., 4:169-187. Collins, A.G., 1969b. Solubilities of some silicate minerals in saline waters. U S . O f f . Saline Water R e s Dev. Progr. R e p . , No.472, 27 pp. Collins, A.G., 1970. Geochemistry of some petroleum-associated waters from Louisiana. U.S. Bur. M i a Rep. Invest., No.7326, 31 pp. Collins, A.G., 1972. Geochemical Classification o f Formation Waters f o r Use in Hydrocarbon Exploration and Production. M.S. Thesis, University of Tulsa, Tulsa, Okla., 63 pp. Collins, A.G., Bennett, J.H. and Manuel, O.K., 1971. Iodine and algae in sedimentary rocks associated with iodine-rich brines. Geol. SOC.A m . Bull., 82:2607-2610. Dapples, E.C., 1959. The behavior of silica in diagenesis. In: Silica in Sediments ( A Symposium) - SOC.E c o n Paleontol. Mineral., Spec. Publ., No.7, pp.36-54. Deffeyes, K.S. Lucia, F.J. and Weyl, P.K., 1964. Dolomitization: observations on the island of Bonaire, Netherlands Antilles. Science, 143 :678+379. Degens, E.T. and Chilingar, G.V., 1967. Diagenesis of subsurface waters. In: G. Larsen and G.V. Chilingar (Editors), Diagenesis in Sediments. Elsevier, Amsterdam, pp.77-502. Degens, E.T., Hunt, J.M., Reuter, J.H. and Reed, W.E., 1964. Data on the distribution of amino acids and oxygen isotopes in petroleum brine waters of various geologic ages. Sedimentology, 3 :199-225. DeSitter, L.Y., 1947. Diagenesis of oilfield brines. Bull. A m . Assoc. Pet. Geol., 31 :2030-2040. Dickey, P.A., 1969. Increasing concentration of subsurface brine with depth. Chem. Geol., 4:361-370. Dingman, R.J. and Angino, E.E., 1969. Chemical composition of selected Kansas brines as a aid to interpreting change in water chemistry with depth. Chem. Geol., 4: 325-339. Dunham, K.C., 1970. Mineralization by deep formation waters - a review. Znst. Metall. Trans., 79:B127-B>36. Eckhardt, F.J., 1958. Uber Chlorite in Sedimenten. Geol. Jahrb., 75:437-474. Egleson, G.D. and Querio, C.W., 1969. Variation in the composition of brine from the Sylvania formation near Midland, Michigan. Environ. Sci. Technol., 3:367-371. Ellis, A.J., 1968. Natural hydrothermal systems and experimental hot waterhock interaction: reactions with NaCl solutions and trace metal extraction. Geochim. Cosmochim. Acta, 32:1356-1363. Emery, K.O., 1960. The Sea O f f Southern California: A Modern Habitat o f Petroleum. John Wiley and Sons, New York, N.Y., 366 pp. Emery, K.O. and Rittenberg, S.C., 1952. Early diagenesis of California Basin sediments in relation t o origin of oil. Bull. A m . Assoc. Pet. Geol., 36:735-806.

REFERENCES

249

Evans, G., Kinsman, D.D.J. and Shearman, D.J., 1963 A reconnaissance survey of the environment of the recent carbonate sedimentation along Trucial Coast, Persian Gulf. In: L.M.J.U. van Straaten (Editor), Developments in Sedimentology, 1 . Deltaic and Shallow Marine Deposits. Elsevier, Amsterdam, pp. 129-1 35. Fowler, Jr., W.A., 1970.Pressures, hydrocarbon accumulation, and salinities - Chocolate Bayou field, Brazoria County, Texas. J. Pet. Technol., 22:411-423. Friedman, G.M. and Sanders, J.E. 1967. Origin and occurrence of dolostones. In: G.V. Chilingar, H.J. Bissel and R.W. Fairbridge (Editors), Carbonate Rocks - Origin, Occurrence and Classification. American Elsevier, New York, N.Y., pp.267-348. Friedman, G.M. Fabricand, B.P., Imbimbo, E.S., Brey, M.E. and Sanders, J.E., 1968. Chemical changes in interstitial waters from continental shelf sediments. J. Sediment. Petrol., 38:1313-1319. Garrels, R.M. and Christ, C.L., 1965. Solutions, Minerals, and Equilibria. Harper and Row, New York, N.Y., 450 pp. Gevirtz, J.L. and Friedman, G.M. 1966. Deep-sea carbonate sediments of the Red Sea and their implications on marine lithification. J. Sediment. Petrol., 36:143-151. Ginsburg, R.N. and Lowenstam, H.A., 1958. Influence of marine bottom communities on the depositional environment of sediments. J. Geol., 66:310-318. Grabau, A.W., 1920. Geology of Nonmetallic Mineral Deposits Other Than Silicate, 1 . Principles of Salt Deposits. McGraw-Hill, New York, N.Y., 435 pp. Graf, D.L., Friedman, I. and Meents, W.F., 1965. The origin of saline formation waters, 11. Isotopic fractionation by shale micropore systems. Ill. State Geol. Surv. Circ., No.393, 32 pp. Graf, D.L., Meents, W.F., Friedman, I. and Shimp, N.F., 1966. The origin of saline formation waters, 1 1 Calcium chloride waters. Ill. State Geol. Surv. Circ. No. 397,60 1. PP. Grim, R.E., 1952. Clay Mineralogy. McGraw-Hill, New York, N.Y., 396 pp. Ham, W.E. 1962. Classification of carbonate rocks. A m . Assoc. Pet. GeoL, Mem. 1 , 279 PP. Hanshaw, B.B. and Hill, G.A., 1969. Geochemistry and hydrodynamics of the Paradox Basin region, Utah, Colorado, and New Mexico. Chem. Geol., 4:263-294. Helgeson, H.C., 1964. Complexing and Hydrothermal Ore Deposition. MacMillan, New York, N.Y., 128 pp. Hem, J.D., 1970. Study and interpretation of the chemical characteristics of natural water. U.S. Geol. Surv. Water Supply Paper, No. 1473,363 pp. Hiltabrand, R.R. 1970. Experimental Diagenesis o f Argillaceous Sediment. Ph.D. Dissertation, Louisiana State University, Baton Rouge, La., 152 pp., unpublished. Hitchon, B., 1968. Rock volume and pore volume data fo; plains region of Western Canada Sedimentary Basin between latitudes 4' and 60 N. Bull. A m . Assoc. Pet. 9 Geol., 52:2318-2323. Hitchon, B. 1969a. Fluid flow in the Western Canada Sedimentary Basin, 1. Effect of topography. Water Resour. Res., 5:186-195. Hitchon, B., 196913. Fluid flow in the Western Canada Sedimentary Basin, 2. Effect of geology. Water Resour. Res., 5:460-461. Hitchon, B. and Friedman, I., 1969. Geochemistry and origin of formation waters in the Western Canada Sedimentary Basin, I. Stable isotopes of hydrogen and oxygen. Geochim. Cosmochim. Acta, 33:1321-1349. Hitchon, B . and Hays, J., 1971. Hydrodynamics and hydrocarbon occurrences, Surat Basin, Queens land, Australia. Water Resour. Res., 7:658-676. Holser, W.T. 1963. Chemistry of brine inclusions in Permian salt from Hutchison, Kansas. In: J.L. Rau (Editor), Symposium o n Salt. Northern Ohio Geological Society, Cleveland, Ohio, pp.86-103. Hunt, J.M., 1968. How gas and oil form and migrate. World Oil, 163:140-150.

250

ORIGIN OF OILFIELD WATERS

Hunt, J.M. and Jamieson, G.W., 1958. Oil and organic matter in source rocks of petroleum. In: L.G. Weeks (Editor), Habitat of Oil. American Association of Petroleum Geologists, Tulsa, Okla., pp. 735-746. Illing, L.V., 1954. Bahaman calcareous sands. Bull. A m . Assoc. Pet. Geol., 38:l-95. Jones, B.F., Vandenburgh, A.S., Truesdell, A.H., and Rettig, S.L., 1969. Interstitial brines in playa sediments. Chem. Geol., 4:253-262. Jones, P.H., 1969. Hydrology of Neogene deposits in the northern Gulf of Mexico Basin. La. Water Resour. Res. Inst. Bull., GT-2, 105 pp. Kelley, W.P., 1948. Cation Exchange in Soils. Reinhold, New York, N.Y., 144 pp. Khitarov, N.I. and Pugin, V.A., 1966. Behavior of montmorillonite under elevated temperatures and pressures. Geochem. Znt., 3:621-626. Kimura, S . and Souriragan, S., 1967. Analysis of data in reverse osmosis with porous cellulose acetate membranes used. A m . Inst. Chem. Eng., 13:497-503. Kozin, A.N., 1960. Geochemistry of bromine and iodine of formation waters in the Kuybyshev area o n the Volga. Pet. Geol., 4:llO-113. Kramer, J.R., 1963. History of the composition of sea water - liquid inclusions compared with a chemical equilibrium model. Geol. SOC.A m . , Spec. Paper, No. 73, 190 pp. Kramer, J.R., 1969. Subsurface brines and mineral equilibria. Chem. Geol., 4:37-50. Krauskopf, K.B., 1956. Factors controlling the concentrations of thirteen rare metals in sea water. Geochim. Cosmochim. Acta, 9:l-32. Krumbein, W.C., 1951. Occurrence and lithologic associations of evaporites in the United States. J. Sediment Petrol., 21:63--81. Krejci-Graf, K., 1963. Uber Rumanische Olfeldwasser. Geol. Mitt. Hydrogeol. Hydrochem., 2:3 51-392. Kvenvolden, K.A., 1964. Hydrocarbons in modern sediments and the origin of petroleum. Min. Mag., Colo. School Min., 54:24-25. Larson, T.J., 1967. Purification of subsurface waters by reverse osmosis. J. Am. Water Works Assoc., 59:1527-1548. Levorsen, A.I., 1966. Geology of Petroleum. W.H. Freeman, San Francisco, Calif., 2nd ed., 724 pp. Loeb, S. and Manjikian, S., 1965. Six-month field test of a reverse osmosis desalination membrane. Ind, Eng. Chem. Process Design Dev., 4: 207-21 2. Lotze, F., 1938. Wichtigen Lager Stattender nicht-erze, III. Steinsalz und Kalisalze Geologie. Borntraeger, Berlin, 936 pp. Mandl, I., 1953. Solubilization of insoluble matter in nature, 11. The part played by salts of organic and inorganic acids occurring in nature. Biochim. Biophys. Acta, 10:540-569. Mandl, I., Grauer, A. and Neuberg, C., 1952. Solubilization of insoluble matter in nature, I. The part played by salts of adenosinetriphosphate. Biochim. Biophys. Acta, 8 :654-663. Manheim, F.T. and Bischoff, J.L., 1969. Geochemistry of pore waters on the Continental Slope of the northern Gulf of Mexico. Chem. Geol., 4:63-82. McAuliffe, C., 1969. Determination of dissolved hydrocarbons in subsurface brines. Chem. Geol., 4:225-234. McKelvey, J.G., Spiegler, K.S. and Wyllie, M.R.J., 1957. Salt filtering by ion-exchange grains and membranes. J. Phys. Chem., 61(2):174-178. Meyerhoff, A.A., 1970. Development in Mainland China, 1949-1968. Bull. A m . Assoc. Pet. Geol., 54:1567-1580. Michaels, A.S., Bixler, H.J. and Hodges, Jr., R.M., 1965. Kinetics of water and salt transport in cellulose acetate reverse osmosis desalination membranes. J. Colloid Sci., 20: 1034-1056. Milne, I.H., McKelvey, J.G. and Trump, R.P., 1964. Semi-permeability of bentonite membranes to brines. Bull. A m . Assoc. Pet. Geol,, 48:103-105.

REFERENCES

251

Moore, C.A., 1969. The occurrence of oil in sedimentary basins. World Oil, 168:69-72. Morris, R.C. and Dickey, P.A., 1957. Modern evaporite deposition in Peru. Bull. A m . Assoc. Pet. Geol., 41:2467-2474. Neruchev, S.G. and Kovacheva, I.S., 1965. The effect of geological conditions on the amount of oil given up by source rocks. Dokl. Akad. Nauk U.S.S.R., 162:913-914. Neumann, H.J. and Jobelius, H., 1967. Detection of emulsifying agents in crude oils and oilfield waters as a contribution to the problem of oil migration. Erdol Kohle Petrochem., 20:622-625. Parker, J. W., 1969. Water history of Cretaceous aquifers, East Texas Basin. Chem. Geol., 4:lll-133. Peake, E. and Hodgson, G.W., 1966. Alkanes in aqueous systems, I. Exploratory investigations on the accommodation of Cz 0-c~ n-alkanes in distilled water and occurrence 3 in natural water systems. J. A m . Oil Chem. SOC.,43:215-222. Pettijohn, F.J., 1957. Sedimentary Rocks. Harper and Brothers, New York, N.Y., 2nd ed., 718 pp. Philipp, W., Drong, H.J., Fuchtbauer, H., Haddenhorst, H.G. and Jankowsky, W.J., 1963. The history of migration in the Gifhorn Trough ( N W Germany), Sixth World Pet. Congr., Frankfurt/Main, June, 1963, Sect. I, Paper, No. 19, pp. 457-481. Philippi, G.T., 1965. On the depth, time and mechanism of petroleum generation. Geochim. Cosmochim. Acta, 29: 1021-1049. Phleger, F.B. and Ewing, G.C., 1962. Sedimentology and oceanography of coastal lagoons in Baja, California, Mexico. Geol. SOC.A m . Bull., 73:145-181. Pirson, S.J., 1968. Redox log interprets reservoir potential. Oil Gas J., 66:6*75. Plumley, W.J., Risley, G.A., Graves, Jr., R.W. and Kaley, M.E., 1962. Energy index for limestone interpretation and classification. In: W.E. Hem (Editor), Classification o f Carbonate Rocks - A m . Assoc. Pet. Geol., Mem.1, pp.85-107. Pollard, T.A. and Reichertz, P.O., 1952. Core-analysis practices - basic methods and new developments. Bull. Am. Assoc. Pet. Geol., 36:230-252. Quaide, W . , 1958. Claymineralsfrom salt concentration ponds. A m . J. Sci., 256:431-437. Rankama, K. and Sahama, T.G., 1950. Geochemistry. Chicago University Press, Chicago, Ill., 991 pp. Riley, G.A., 1944. The carbon metabolism and photosynthetic efficiency of the earth as a whole. J. A m . Sci., 32:134. Rittenhouse, G., 1967. Bromine in oilfield waters and its use in determining possibilities of origin of these waters. Bull. A m . Assoc. Pet. Geol., 51:2430-2440. Rittenhouse, G., Fulton, R.B., Grabowski, R.J. and Bernard, J.L., 1969. Minor elements in oilfield waters. Chem. Geol., 4:189-209. Rosenqvist, I.T., 1962. The influence of physico-chemical factors upon the mechanical properties of clays. Clays Clay Minerals, 9 :12-27. Ross, C.S., 1943. Clays and soils in relation to geologic processes. J. Wash. Acad. Sci., 33:225-235. Russell, W.L., 1933. Subsurface concentration of chloride brines. Bull. A m . Assoc. Pet. Geol., 17:1213-1228. Schoeller, H., 1955. Geochemie des eaux souterraines. Rev. Inst. F r Pet., 10:181-213, '. 219-246, 507-552. Siever, R., Beck, K.C. and Berner, R.A. 1965. Composition of interstitial waters of modern sediments. J. Geol., 73:39-73. Skinner, B.J., 1969. Earth Resources. Prentice-Hall, Englewood Cliffs, N.J., 149 pp. Sloss, L.L., 1953. The significance of evaporites. J. Sediment. Petrol., 23:143-161. Smith, P.V., 1954. Studies on origin of petroleum: occurrence of hydrocarbons in recent sediments. Bull. A m . Assoc. Pet. Geol., 38:377-381. Stumm, W. and Morgan, J.J., 1970. Aquatic Chemistry. Wiley-Interscience, Div. of John Wiley and Sons, New York, N.Y., 583 pp.

252

ORIGIN OF OILFIELD WATERS

Taylor, D.W., 1956. Fundamentals o f Soil Mechanics. John Wiley and Sons, New York, N.Y., 700 pp. Tenaghi, K. and Peck, R.B., 1968. Soil Mechanics in Engineering Pmctice. John Wiley and Sons, New York, N.Y., 84 pp. Topms, J.S., 1970. Review of knowledge of metalliferous brines and related deposits. Inst. Min. Metall. Trans., 79:B116-B126. Truesdell, A.H. and Jones, B.F., 1969. Ion association in natural brines. Chem. Geol., 4:51-62. Usiglio, J., 1849. Analyse de Ieau de la Mediterrank sur les c6tes de France. Ann. Chim. Phys., 3:92-1 07, 27 :172-191. Valyashko, M.G., 1956. The geochemistry of bromine in halogenesis processes and the use of bromine content as a genetic and prospecting criterion. Geochemistry, 1 :33-49. Van Everdingen, R.O., 1968. Studies on formation waters in western Canada; geochemistry and hydrodynamics. Can. J. Earth Sci., 5:523-543. Van Nostrand Press, 1958.Science Encyclopedia. Princeton, N.J., 2nd ed., pp. 371-1528. Vinogradov, A.P., 1953. The elementary chemical composition of marine organisms. Sears. Found. Mar. R e s , Mem., 11:85,91,107,216. Visher, G.S., 1965. Use of vertical profile in environmental reconstruction, Bull. A m . Assoc. Pet. Geol., 49:41-61. Von Engelhardt, W., 1960. On the chemistry of the pore solution of sediments. Uppsala Univ. Geol. Inst. Bull., 40: 189-204 (in German). Weller, J.M., 1959. Compaction of sediments. Bull. A m . Assoc. Pet. Geol., 43:273-310. White, D.E., 1957. Magmatic, connate, and metamorphic waters. Geol. SOC.A m . Bull., 68:1659-1 682 . White, D.E., 1965. Saline waters of sedimentary rocks. In: A. Young and J.E. Galley (Editors), Fluids in Subsurface Environments - A m . Assoc. Pet. Geol., Mem.4, pp.342-366. Young, A. and Low, P.F., 1965. Osmosis in argillaceous rocks. Bull. A m . Assoc. Pet. Geol., 49:1004-1008.

Chapter 8. CLASSIFICATION OF OILFIELD WATERS

Classification of waters provides a basis for grouping closely related waters. Because the grouping is chemical, it is dependent upon the dissolved constituents found in the waters. Most of the classification systems developed t o date have considered only the dissolved major inorganic constituents and have ignored the organic and the minor and trace inorganic constituents. Waters as related to the earth are meteoric, surface, and subsurface. Surface waters can be fresh or saline if the amounts of dissolved constituents in the waters are used to classify them. For example, water from melting snow on a mountain top usually will contain small amounts of dissolved mineral matter and can be classified as fresh water, while water in an ocean will contain about 35,000 mg/l dissolved minerals and is classified as saline. Waters found in rivers connecting the mountain stream t o the ocean may contain varying amounts of dissolved constituents and depending upon the amounts can be classified as fresh or saline. In a similar manner, subsurface waters are classified as fresh or saline. Merely classifying a water as either fresh or saline does not provide a very useful classification. The dissolved constituents that are used in many classification systems depend upon the amounts or ratios of sodium, magnesium, calcium, carbonate, bicarbonate, sulfate, and chloride found in the water. The reason for this is that these are the ions that usually are determined or calculated in a water. (Sodium often is calculated from the difference found in the stoichiometric balance of the determined anions and cations.) The amounts and ratios of these constituents in subsurface waters are dependent upon the origin of the water and what has occurred t o the water since entering the subsurface environment. For example, some subsurface waters found in deep sediments were trapped during sedimentation, while other subsurface waters have been diluted by infiltration of surface waters through outcrops. Some waters have been replaced by infiltration water. Also, rocks containing the waters often contain soluble constituents, which dissolve in the waters or contain chemicals which will exchange with chemicals dissolved in the waters causing alterations of the dissolved constituents. The amounts of dissolved constituents found in subsurface waters can range from a few milligrams per liter t o more than 350,000 mg/L This salinity distribution is dependent upon several factors, including hydraulic gradients, depth of occurrence, distance from outcrops, mobility of the dissolved chemical elements, soluble material in the associated rocks, and the exchange reactions.

254

CLASSIFICATION OF OILFIELD WATERS

Portions of three classification systems (Palmer, 1911; Sulin, 1946; Schoeller, 1955) and Bojarskis (1970) modification of Sulins system were applied to about 4,000 formation waters (U.S. Bureau of Mines, 1965). The waters were analyzed by standard methods (American Petroleum Institute, 1968). The results indicated that the classifications are useful in exploration and production problems. Palmers classification Palmer (1911) observed that the basic characteristics of natural waters are dependent upon their salinity (salts of strong acids) and alkalinity (salts of weak acids). Salts that cause salinity are those that are not hydrolyzed, while alkalinity is caused by free alkaline bases produced by the easily hydrolyzable salts of weak bases. All positive ions (cations) including hydrogen can cause salinity, but of the negative ions (anions), only the strong acids, (e.g., chloride, sulfate, and nitrate) can cause salinity. Because salinity is dependent upon the combined activity of the cations and anions and is limited by the reacting values of the strong acids, its value is determined by multiplying the total value of the strong acids by two. Alkalinity is caused by free alkaline bases as a result of the hydrolytic action of water on dissolved bicarbonates and other weak acid salts. The alkalinity value is calculated by doubling the reacting values of the bases which exceed the reacting values of the strong acids. The ions that commonly are found in waters comprise three groups: (a) alkalies (sodium, potassium, lithium), whose salts are easily soluble in water and do not cause hardness; (b) alkaline earths (magnesium, calcium, strontium, barium), whose salts cause hardness and many of which are sparingly soluble; and (c) hydrogen, whose salts are acids and cause acidity. Geologists know what strong alkalies, alkaline earths, strong acid radicles, weak acid radicles, ions, and reacting values mean generally. To compare several analyses it usually is easier if they are made on a chemical basis. The. proportions of the various ions do not react in proportion t o the various weights given in milligrams per liter but rather in proportion to their capacity for reaction, or reaction value. The reacting value of each ion is determined by multiplying the amount of each radicle by weight (mg/l) by its reaction coefficient, which is the valence of a radicle divided by its atomic weight. The groups of the ions are determined by summing the reacting values of their members, and according to the predominance of reacting values of the groups, five special properties were designated by Palmer. To determine the special properties, the reacting values of a group of cations or anions are doubled so that the full value of a given special property is considered. The terms primary and secondary were used t o qualify the general proper-

PALMERS CLASSIFICATION

25 5

ties of the water; e.g., the principal soluble decomposition products of the oldest rock formations are the alkalies (primary), while more recent rock formations are the principal source of the alkaline earths (secondary), This theory of Palmers that the terms primary and secondary are associated with the age of the rock should not necessarily be considered undisputably true, because primary salinity certainly can be acquired from other soluble material than that derived directly from decomposition products of the oldest rock formations. The five special properties of water are: (1) Primary salinity (alkali salinity); that is, salinity not t o exceed twice the sum of the reacting values of the radicles of the alkalies. (2) Secondary salinity (permanent hardness); that is, the excess (if any) of salinity over primary salinity, not t o exceed twice the sum of the reacting values of the radicles of the alkaline earths group. (3) Tertiary salinity (acidity); that is, the excess (if any) of salinity over primary and secondary salinity. (4) Primary alkalinity (permanent alkalinity); that is, the excess (if any) of twice the sum of the reacting values of the alkalies over salinity. (5) Secondary alkalinity (temporary alkalinity); that is, the excess (if any) of twice the sum of the reacting values of the radicles of the alkaline earths group over secondary salinity. Reacting values in percent are used in this system. The percentage values are determined by summing the milliequivalents of all the ions, dividing the milliequivalents of a given ion by the sum of the total milliequivalents, and multiplying by 100. Waters are classified by numerical values of the relationships of anions to the cations, where a , b , and d represent the percentage values of the alkali cations, alkaline earth cations, and strong acid anions, respectively. Any one of the following five conditions may exist: d may be equal t o or less than a , greater than a and less than a + b y equal to a + b y or greater than a + b . Using these conditions, waters are classified into five classes: Classl: d < a 2d = primary salinity 2(a - d ) = primary alkalinity 2b = secondary alkalinity Class 2: d = a 2u or 2d = primary salinity 2b = secondary alkalinity Class 3 : d > a ; d < (a + b ) 2a = primary salinity 2(d - a ) = secondary salinity 2(a + b - d ) = secondary alkalinity

256

CLASSIFICATION OF OILFIELD WATERS

Class 4: d = (a + b ) 2a = primary salinity 2b = secondary salinity Class 5: d > ( a + b ) 2u = primary salinity 2b = secondary salinity 2(d - a - b ) = tertiary salinity (acidity) These five classes of water are found in nature. Examples of the first three classes are various surface waters, sea water and brines represent class 4, while mine drainage waters and waters of volcanic origin fall in class 5 (Palmer, 1911). Rogers (1917, 1919) studied oilfield waters of the San Joaquin Valley, California, and used the classification system of Palmer (1911). He found that generally the surface waters of the San Joaquin Valley possess secondary salinity rather than primary alkalinity, contain more sulfate than chloride, and contain low amounts of bicarbonate. With increasing depth, the subsurface waters decrease in secondary salinity until primary alkalinity becomes evident. Waters above an oil zone often contained hydrogen sulfide, which was attributed t o reduction of sulfates by hydrocarbons, thus decreasing the amounts of sulfate and increasing the bicarbonate in the water, which Rogers called an altered water. Further he found that, in these altered waters in close proximity to hydrocarbon accumulations, chloride becomes relatively and absolutely important because of the residual chloride from the original (ancient) sea water chlorides as compared to waters above the oil zone which often are freshened because of a more hydrodynamic situation. Altered waters, according t o his definition, can have either primary alkalinity or secondary salinity depending upon their amounts of carbonate and chloride, but normal waters have only secondary salinity. Elliott (1953) used the Palmer system t o determine the chemical characteristics of some Paleozoic age formation waters in Texas. He found that all of the waters in the group that he studied (about 70) contained predominant, primary salinity. Many of these waters contained appreciable concentrations of sulfate; one contained 5,800 mg/l sulfate, and many contained more than 2,000 mg/l. The calcium concentration ranged up to 13,000 mg/l while the bicarbonate concentrations ranged up t o 800 mg/l. Ostroff (1967) used the Palmer classification to classify waters from several basins and t o compare this classification system with two other systems. He found that the Palmer system groups some of the constituents together that are not closely related chemically. Furthermore the system does not consider ionic concentrations or saturation conditions related t o sulfate or bicarbonate.

SULINS CLASSIFICATION

257

Sulins classification
Sulin (1 946), a Russian geochemist, proposed a classification system based upon various combinations of dissolved salts in the waters. The waters are described according to chemical type, subdivided into group, subgroup, and class. He found four basic environments of natural water distribution: (1) Continental (terrestrial) conditions which promote the formation of sulfate waters. Such conditions supply soluble sulfate constituents t o the water and the genetic type of such a water is sulfate-sodium. (2) Continental conditions which promote the formation of sodium bicarbonate waters. The genetic type is bicarbonate-sodium. ( 3 ) Marine conditions and the formation of a chloride-magnesium type of water. (4) Deep subsurface conditions within the earths crust and the formation of a chloride-calcium type of water. The first two types are characteristic of meteoric and/or artesian waters, the third of marine environments and evaporite sequences, and the fourth of deep stagnant conditions.

Types, groups, and subgroups


Water composition is expressed in milligram-equivalents of the separable ions, and the composition is calculated per 100 g of water. The percent of the sum of the equivalents is used t o exclude the degree of water mineralization, and t o compare waters containing different amounts of dissolved solids. The ratio Na/Cl expressed in the percent equivalent form determines the genetic water type. If the value is greater than one, sodium predominates over chloride and the excess sodium can be combined with sulfate or bicarbonate. Therefore waters with a Na/Cl ratio greater than one belong to the bicarbonate-sodium or the sulfate-sodium types. Sulin calculated sodium as the sum of all the alkalies (Li, K, Na etc.) and chloride as the sum of all the halides (Cl, Br, I). The ratio (Na - C1)/S04, if greater than one, indicates that the water is the bicarbonate-sodium type, while if it is less than one it is the sulfatesodium type. Similarly the ratio (Cl- Na)/Na if less than one indicates the chloride-magnesium type, but if greater than one it indicates the chloridecalcium type.

Water classes
Subdivision of the groups of waters were made by Sulin (1946) using the Palmer (1 ) characteristics, because these characteristics express the dis911 solved constituents in the waters in a generalized format. For example, the sum of the alkali chlorides and sulfates corresponds t o primary salinity, and the sum of the alkaline earth chlorides and sulfates corresponds to secondary

258

CLASSIFICATION OF OILFIELD WATERS

salinity and the sodium bicarbonate-calcium stage. N o sodium bicarbonate is present in sulfate-sodium, chloride-magnesium, or chloride-calcium types of water; therefore, these types are classified as follows: (1)Class A2 : secondary alkalinity predominates (alkaline earth carbonates and bicarbonates). (2) Class S2 : secondary salinity predominates (alkaline earth sulfates and chlorides). (3) Class S, : primary salinity predominates (alkali sulfates and chlorides). (4) Class S,: tertiary salinity predominates (iron and aluminum sulfates and chlorides and free strong acids). Bicarbonate-sodium type waters contain sodium bicarbonate and are classified as follows: ( 5 ) Class A2 : secondary alkalinity predominates (alkaline earth carbonates and bicarbonates). (6) Class A , : primary alkalinity predominates (alkali carbonates and bicarbonates). (7) Class S, : primary salinity predominates (alkali chlorides and sulfates). (8) Class A, : tertiary alkalinity predominates (iron and aluminum carbonates and bicarbonates). The water classification is expressed by use of a formula representing decreasing values of the Palmer characteristics. For example, S, S2 A 2 indicates that primary salinity is predominant and is followed by secondary salinity and secondary alkalinity. Therefore, the classes are subdivided into subclasses, and class S1 can include the subclass S, S2 A 2 , S1 A2 S2, S1 S2, and S, . Table 8.1 outlines Sulins method of water characterization. Table 8.11 briefly outlines the relative values of the coefficients which determine the four genetic types of waters. The Palmer characteristics do not account for the interrelations between chloride and sulfate and between calcium and magnesium. Therefore, Sulin calculated the ratio SO4/C1 and Ca/Mg to establish additional subgroups. The complete water characterization included the following: (a) water formula given in Palmer characteristics; (b) coefficients in percent equivalents for S04/C1 and Ca/Mg; (c) sum of the milligram equivalents per 100 g of water (Z r ) t o illustrate the degree of water mineralization; and (d) the genetic coefficients (Na - C1)/ SO4 and (Ca - Na)/Mg t o determine the water type, and Na/C1 t o determine related genetic types of water.

Hydrochemical indicators of hydrocarbons


Sulin (1946) noted that certain properties of subsurface waters were favorable indicators of hydrocarbon accumulations. The bicarbonate-sodium and chloride-calcium types of waters are widely found in oilfields. However, the chloride-calcium type is the more favorable indicator if it has the most characteristic composition plus certain minor or micro constituents. In general, he determined that hydrocarbon accumulations are most commonly

SULINS CLASSIFICATION

25 9

TABLE 8.1
Sulins method of water characterization
_____

Na/Cl

>1
(Na+ - C1-)

Sulfate-sodium type:

so,,-2

<1

Bicarbonate-sodium type:
Bicarbonate group class A1 sodium subgroup class A2 calcium subgroup magnesium subgroup sodium subgroup class S 1 sodium subgroup Sulfate group class S1 sodium subgroup Chloride group class S1 sodium subgroup

(Na+ - C1-)

so4-2

>

Bicarbonate group class A2 calcium subgroup magnesium subgroup Sulfate group class S1 calcium subgroup magnesium subgroup sodium subgroup class Sp calcium subgroup magnesium subgroup Chloride group class S 1 calcium subgroup magnesium subgroup sodium subgroup Na/Cl

<1
(C1-- Na+) < 1 Chloride-calcium type: (Cl- - Na+) > Mg+ Mg+2 Bicarbonate group class A 2 calcium subgroup magnesium subgroup Sulfate group class S1 calcium subgroup magnesium subgroup class s 2 calcium subgroup magnesium subgroup Chloride group class S1 calcium subgroup magnesium subgroup sodium subgroup class s 2 calcium subgroup magnesium subgroup

Chloride-magnesium type:
Bicarbonate group class Aq calcium subgroup magnesium subgroup Sulfate group class s1 calcium subgroup magnesium subgroup class S2 calcium subgroup magnesium subgroup Chloride group class S1 calcium subgroup magnesium subgroup sodium subgroup class S2 calcium subgroup magnesium subgroup

26 0

CLASSIFICATION OF OILFIELD WATERS

TABLE 8.11 Coefficients characterizing the genetic types of waters


~~

Type of water Chloride-calcium Chloride-magnesium Bicarbonate-sodium Sulfate-sodium

Na+/Cl-

(Na - Cl-)/S04-2

(Cl->1 <1

Na+)/Mg+

<1 <1 >1 >1

<O <O
>1 <1

<O <O

related t o water types in this order: chloride-calcium > bicarbonate-sodium chloride-magnesium > sulfate-sodium. Most oilfield waters of the chloride-calcium type belong to the S1 S2 A2 class with a few in the S2 S, A2 class, while most oilfield bicarbonate-sodium waters belong to the S, A, A2 and A , S1 A 2 classes. Other significant indicators were grouped by Sulin; however, none of them can assure the existence of a hydrocarbon deposit, and certainly they cannot provide definite evidence of the size of the accumulation. The groups are as follows: Group I: direct hydrocarbon indicators; for example, naphthenic acid salts and iodide. The naphthenic acids are more soluble in bicarbonate-sodium type waters and are related t o the composition of the hydrocarbon accumulation. Iodide is related to oil because it must have an organic origin. Sulin also noted the dissolved gases in the waters and considered the heavier hydrocarbons such as ethane and butane and the absence of oxygen as direct indicators. Group 11: highly mineralized chloride-calcium or bicarbonate-sodium types of water containing reduced forms of sulfur are important indirect indicators of oil. The sulfate content should be low to indicate interaction with bituminous constituents and/or sulfate-reducing bacteria. Group 111: in this group are constituents which have no genetic relationship to hydrocarbons but appear characteristic of waters that are related t o hydrocarbon accumulations. The constituents are bromide, boron, barium, strontium, radium, and possibly fluoride.

>

f Modification o Sulins system by Bojarski

Bojarski (1970) studied 400 water analyses and differentiated hydrochemical zones within basins in Poland that appear suitable for preservation of hydrocarbon deposits. He distinguished the waters as follows: (1) Waters of the bicarbonate-sodium type. Such waters occur in the upper zone of a sedimentation basin, with intense water exchange (that is, a hydrodynamic situation where the waters are moving at a relatively fast geological rate), which promotes unfavorable conditions for the preservation of petroleum and natural gas deposits. The waters are defined by the ratio

MODIFICATION OF SULINS SYSTEM

26 1

(Na-Cl)/SO, > l . As Sulin (1946) noted, if the ratio Na/Cl in epm is greater than 1, the water contains more sodium than chloride and the excess sodium can react with sulfate or bicarbonate ions. Therefore, such waters belong to the bicarbonate-sodium or sulfate-sodium types. If the ratio (Na - C1)/S04 is greater than 1,it indicates an excess of sodium with respect t o both chloride and sulfate. (2) Waters of the sulfate-sodium type with (Na - Cl)/S04 < 1. This ratio, if less than 1, indicates that all of the sodium will react with chloride or sulfate. (3) Waters of the chloride-magnesium type with (Cl- Na)/Mg < 1. A ratio of this type indicates that all of the chloride will react with sodium and magnesiun. Such a water is characteristic of the transition zone between a hydrodynamic area which is becoming more hydrostatic in the deeper part of the basin, and the amount of dissolved bromide increases directly with the (Cl- Na)/Mg ratio. (4) Waters of the chloride-calcium type with (Cl- Na)/Mg > 1. This ratio indicates an excess of chloride with respect t o sodium and magnesium, and the excess will react with calcium. This type of water occurs in deeper zones which are isolated from the influence of infiltration waters and are hydrostatic or almost hydrostatic. Bojarski observed a large variation in the chemical composition in the chloride-calcium type of water and subdivided this type as follows: (a) The first class, chloride-calcium I with Na/Cl > 0.85 characterizes an active hydrodynamic zone with considerable water movement. It is considered a zone of little prospect for the preservation of hydrocarbon deposits. (b) The second class, chloride-calcium I1 with Na/C1 = 0.85-0.75, characterizes the transition zone between an active hydrodynamic zone and a more stable hydrostatic zone of the sedimentation basin, which is generally considered a poor zone for hydrocarbon preservation. (c) The third class, chloride-calcium I11 with Na/Cl = 0.75-0.65 (0.60), characterizes favorable conditions for the preservation of hydrocarbon deposits.It is designated as a fairly favorable environment for the preservation of hydrocarbons. (d) The fourth class, chloride-calcium IV with Na/C1 = 0 . 6 5 4 . 5 0 , is characterized by complete isolation of the hydrocarbon accumulations as well as by the presence of residual waters. I t is considered a good zone for the preservation of hydrocarbons. (e) The fifth class, chloride-calcium V with Na/C1 < 0.50, is characterized by the presence of ancient residual sea water which has been highly altered since original deposition, both in the concentration of dissolved solids and in the ratios of the dissolved constituents. Bojarski considers a zone of this type to be one of the most likely areas where hydrocarbons are accumulated. Additional characteristics of water associated with hydrocarbon accumulations are as follows: (1) iodide > 1mg/l; (2)bromide > 300 mg/l (increasing

262

CLASSIFICATION OF OILFIELD WATERS

iodide and bromide concenbrations may point to a bitumen accumulation); (3) ratio Cl/Br < 350; and (4) SO4 x lOO/Cl< 1. In addition t o indicating the degree of alteration, bromide and iodide as biophile constituents play a decisive role in the classification Bojarski adopted. This followed because of the increased concentration of biophile elements in the waters accompanying a petroleum deposit. The concentration of iodide in the ground waters depends mainly on the organic substances, whereas theconcentration of bromide up t o a certain limit takes place in an inorganic medium, but an increase in bromide must be evaluated as a positive indication. In many waters accompanying petroleum deposits, large amounts of bromide and smaller amounts of iodide were detected, or vice versa. This probably is related t o the type of bituminous substances which absorb the individual biophile elements in different amounts. Chebotarevs classification Chebotarev (1955),an Australian geochemist, classifies waters on the basis of dissolved bicarbonate, sulfate, and chloride, and he does not consider the acid waters or those that contain free sulfuric or hydrochloric acid. His fundamental assumption is that the anions are independent variables while the cations are dependent. The geochemical types of waters are related t o the products of weathering. Table 8.111 illustrates the cycles and products that are produced by weathering. During the first cycle the igneous rocks are weathered allowing chloride, sulfate, calcium, sodium, silica, and magnesium to go into solution. The second cycle is the weathering of sedimentary rocks with the solution of more of the same products. The third cycle is the weathering of recent drift and yields of the above constituents plus aluminum and iron. Table 8.IV illustrates Chebotarevs (1955) geochemical classification of subsurface waters. The phase of weathering corresponds t o four phases of the solution and redistribution of the chemical constituents in the earths crust and correlates with their relative mobilities. He plotted the relative mobilities of nine chemical constituents using the mobility of chloride as 100%. From this four phases were obtained, namely: (1)chloride and sulfate 100% t o about 58% mobility; (2) calcium, sodium, magnesium and potassium 3% t o about 1.2% mobility; (3) silica about 0.20% mobility; and (4) iron oxide and aluminum oxide, less than 0.05% mobility. The four phases of weathering correspond to the products of weathering shown in Table 8.IV and also to the cycles and products of weathering in Table 8.111. For example the fourth phase in Table 8.IV corresponds with the first cycle in Table 8.111. The genetic types of water shown in the upper portion (A) of Table 8.IV do not correspond directly with the weathering phases since the genetic types overlap the phases. These genetic types are related t o the accumulation products shown in Table 8.111. In the lower portion (B) of Table 8.IV are the

TABLE 8.111 Cycles and types of products of weathering (after Chebotarev, 1955) Cycles of weathering: First cycle (orthoeluvium) from igneous and highly metamorphosed rocks residual products Types of (1) coarse detrital products of weathering (2) calcareous (under vegetation cover) (3)siallitic accumulative products (1) chloride-sulphate (chiefly alluvial) (2) calcareous (chiefly colluvial and proluvial) (3)siallitic Second cycle (paraeluvium) from sedimentary rocks residual products (1) detrital
(2) siallitic

b rn
Third cycle (Neoluvium) from Recent Drift residual products accumulative products accumulative products (1) chloride-sulphate

z rrl
z

(supra-calcareous
(3) allitic (?)

(1) solonez and (1) chloride-sulphate gypsum bearing (2) calcareous (2) leached supra(2) CaC03 chloride-sulphate calcareous (siallitic) (3) unsaturated siallitic*2 (3) unsaturated siallitic (3) alumino- and ferricalcareous chloridesiallitic* siliceous system sulphate

is b =iJ

(4) a l l i t i ~ * ~

(lateritic crust of weathering)

* A large quantity of silica and much of its calcium and sodium compounds are removed; the aluminosilicates pass gradually into residual aluminosilicic acids,
*
*3

i.e., acids of the kaolin type. The action of carbonated atmospheric water is insufficient to replace the absorbed ions by hydrogen. The accumulation of sesquioxides at the expense of the leaching out of the alkalis, alkaline earth, and silica.

TABLE 8.IV Geochemical classification of subsurface waters (after Chebotarev, 1955) ( A ) Relationship of the products of weathering to the genetic types of water Presumable phase of weathering Fourth phase Third and partly second phases Second and partly first phases First phase Products of weathering residual (orthoeluvium and detrital paraeluvium) siallitic drift calcareous accumulation chloride-sulphate accumulation Genetic types of water bicarbonate (alkaline) bicarbonate-chloride (alkaline-saline) chloride-bicarbonate (saline-alkaline) chloride-sulfate (saline) chloride (saline)

(B) Geochemical groups of Waters Major group of water Class Genetic types of water Reacting value in percent

H C 0 3 - + COB- ClBicarbonate

SO4-

Cl-

SO4-

H C 0 , - + CI-

HC03-

SOq-

bicarbonate bicarbonate-chloride chloride-bicarbonate sulphate-chloride sulphate

>40 40-30 30-15 15-5


-

<lo
10-20 20-35

I1
11 1

Na+ + K+ prevail in all types of waters C3+ and Mg less than 2.5 in the water of high saline facies and less than 19.0 in low saline facies

Sulphate

IV

<25

>25 >40

Na+ + K+ prevail in all types of waters Ca+ less than 4.5 in all waters Mg less than 4.0 in all waters
-

Chloride

111

chloride-bicarbonate chloride-sulphate ch 1or ide

30-15 15< 5 5

>20 >20
>40

IV
V

<2 5
<10

Na+ + K+ prevail in all types of waters Ca less than 12.5 in the water of high saline facies Mg+ less than 6.0 in all types of waters

CHEBOTAREVS CLASSIFICATION

26 5

geochemical groups of waters. The three major groups of waters are divided into genetic types, which are determined from the absolute concentrations of the dissolved constituents expressed in reacting values in percent. The bicarbonate group contains three genetic types of water, namely: (1)bicarbonate; (2) bicarbonate-chloride; and (3)chloride-bicarbonate. The amounts of bicarbonate plus carbonate and chloride plus sulfate determines the genetic type. The sulfate group is subdivided into two genetic types: (1) sulfate-chloride; and (2) sulfate. The chloride group is divided into three genetic types: (1)chloride-bicarbonate; (2) chloride-sulfate; and (3) chloride. Chebotarev relates the genetic types to the products of weathering because he believes that although the concentration of dissolved solids in subsurface waters may vary substantially, the types of soluble salts remain largely unchangeable . The water classes are related to the products of weathering, rainfall, and drainage conditions. Class I corresponds t o soluble products from the weathering of orthoeluvium or igneous and highly metamorphosed rocks and their silicate compounds. Class I1 waters are related t o products of weathering from the same silicates and calcareous accumulations. Class I11 waters primarily are related to weathered products from calcareous accumulations. Class IV waters are related to weathering of alluvial, detrital, and gypsum deposits. Class V waters are related to marine deposits plus weathering of the products that derived the Class IV waters. Table 8.IV shows the approximate reacting values in percent for waters n found in some oilfields (Chebotarev, 1955). For example, i such waters in the bicarbonate group, the major cations are sodium plus potassium with the reacting value percentage for calcium and magnesium less than 2.5 if the saline facies are high. (The divisions between his saline facies are shown in Table 8.V.) Table 8.V shows the relationships of hydrodynamic zones t o the geochemistry of the water and the geological environment. The zones are: (1) recharge with active water exchange; (2) pressure with delayed water exchange; and (3) accumulation with stagnant conditions. The equilibrium of the chemical systems (those typical of the major geochemical group of waters) is a criterion called the coefficient of water exchange ( K e ) and is computed as: Na(K)HC03 + (Ca,M&)(HC03)2 Ke = Na(K)Cl + (Ca,Mg)C12 + Naz SO4 + (Ca,Mg)S04 The absolute and relative coefficients of water exchange for the three major groups of waters are as follows:

Ke (absolute)
Bicarbonate waters Sulfate waters Chloride waters

Ke (relative)

15 .5 0.11 0.016

96.9 69 .
1 .o

TABLE 8.V Relationships of hydrodynamic zones to the geochemistry of water and the geological evironment (after Chebotarev, 1955) Hydrodynamic zone rechargedischarge cycle Zone of recharge waterexchange active exchange class Geochemistry of water hydro-chemical facies approximate salinity (ppm) common terms for water fresh Geological environment structures relation to water depth (ft) usually less than 500 examples

I and I1 low saline facies


(some times 111) transitional (typical) facies

180-2,400

different

intensive flush

everywhere

2,400--11,400 brackish

deep portions of delayed flush structures with peculiar geochemical environment hampered flush different deeper portions of structures, folded zones inadequate flush circulation and drainage limited

sometimes

Great Artesian Basin, field and others

5,000-7,000 Rocky Mountain oil-

high saline facies Zone of pressure delayed exchange


111 and
low saline facies

11,400-37.800 saline 400-2,500 2.500-7.400


fresh brackish

IV
trazs. Lional (typical) facks high saline facies

usually less than 1,000 sometimes

everywhere precaucasian Basin, some oilfield areas

3.000-4,000 South Dakota Basin,

7,400--19,300 saline 1.500-20,000


fresh and saline different salt accumulation prevails upon leaching different chiefly in arid regions; deeper portions of many artesian basins: some oilfields

stagnant Zone of accumulation condition

low saline facies

transitional (typical) facies

20.000-90,000 saline and


brines

deeper portions of structures, highly folded zones water exchange manifests on geoloeical scale time

sometimes

8,000-1 3,000

high saline facies

90,000-300,000 brines

many oilfield areas (Louisiana, Alberta, etc.)

SCHOELLERS SYSTEM

26 7

The absolute and relative coefficients were derived from assumed chemical compositions of typical waters of the major geochemical groups. The absolute value can be not lower for the group, but it can be relatively higher. Several thousand analyses of waters associated with oil pools in the world were used to formulate the typical waters. As the water-exchange conditions deteriorate, the type of water changes, and the changes are related to alteration or diagenesis of the waters (metamorphism). The data in Table 8.V give hydrochemical facies, common names of various waters as related to dissolved solids concentrations, geological structures, flush or water circulation, depths, and examples where some of the water types are found. The highly concentrated chloride waters are primarily associated with oil occurrences; however, this is not always true. A prime determinant of the chemical composition of oilfield waters is the hydrodynamic situation and the type of trap. For example, an intensively flushed zone will contain a different type of water from a zone with limited circulation. The type of basin strongly influences the type of water that.is likely t o be found. For example, an open basin probably will contain artesian waters of the bicarbonate group, a partly closed basin may contain artesian or subartesian waters of the bicarbonate or sulfate groups, while a closed basin is more likely t o contain bicarbonate waters on the flanks of the basin with sulfate and chloride waters in the deeper areas. Chebotarev (1955) applied his classification t o 917 subsurface waters in oilfields in the world. The classification indicated that 73.7% of the waters were of the chloride genetic type, 23.0% of the bicarbonate type, and 3.3% of the sulfate type. Most of the sulfate and bicarbonate types were found in the Rocky Mountain areas of the United States and probably were mixtures containing infiltrating meteoric water. Schoellers system

A French geochemist, Schoeller (1955),classified waters on the basis of their dissolved constituents and in the following order of importance: (1) chloride; (2) sulfate; (3) bicarbonate plus carbonate; (4) index of base exchange (IBE); and (5) relationships of anions t o cations. This system separates waters into six primary types based upon their amounts of dissolved chloride and four subgroups based on their concentrations of sulfate. The amounts of bicarbonate and carbonate ions give additional differentiation, and an index of base exchange indicates exchange of ions in the waters with ions in associated clays. Table 8.VI outlines Schoellers classification. As shown in Table 8.VI the chloride concentration separates waters into six types. Waters from several oil-producing regions were classified and the sequence C1> SO4 > HC03 was found to occur in very high chloride waters and in sea waters, especially when they are saturated with CaS04. If the waters are not saturated in CaS04, the sequence Cl- > HC0,- > S04-2 is predominant, and in low-chloride waters the predominant sequence is H C O ~ -> C T > S04-2.

26 8
TABLE 8.VI

CLASSIFICATION O F OILFIELD WATERS

Schoellers scheme for classifying petroleum reservoir waters*

Chloride concentration as C i
Very high if 700 Marine if 420 - 700 High if 140 - 420 Average if 40 - 140 Low if 10 - 40 N o r m a l i f < 10

>

Sulfate concentration as s04-


Very high if 58 High if 24 - 58 Average if 6 24 Normalif< 6 Near saturation when J(S04-)

> -

(Ca+)

> 70
+

Bicarbonate plus carbonate concentmtion as H C 0 3 -

c03-

High if 7 Normal if 2 7 Low if 2 However, he recommends using q ( H C 0 3 - + CO3- )(Ca+ ) rather than HC03- + C03-

> <

Index o f base exchange (IBE)

If Cl-

If Na+

> Na+ then IBE = (Cl- - Na+)/Cl> Cl- then IBE = ( C T - Na+)/(S04-

+ HCO3- + C03-2)

Importance o f anions and cations

c1- > SO^' > co3- co3- > c1-> so4-2 co3- > so4- > C T
Na+ Na+

c - > .co~- so4- 1 >

> Mg > Ca+ > Ca+ > Mg+

All constituents are calculated in epm.

SCHOELLERS SYSTEM

26 9

Also, in very high chloride waters only the sequence Na+ > Ca+ > Mg+2 is found. As the C1- decreases, the sequence Na+ > Mg+ > Ca+ becomes more frequent. In very high chloride waters SO4- > Ca+, but in less concentrated waters the opposite may occur. The sequence HC03-< Ca+ always is found in very high chloride waters. In less concentrated chloride waters, either HCO,- < Ca+ or HC03- > Ca+ may be found, while in low chloride waters HC03- > Ca+ is predominant. x Ca+ to indicate Schoeller used an arbitrary value of 70 for JSO4that a water is saturated with CaS04. (This is not necessarily true because some waters, depending upon their other dissolved constituents, can contain smaller or larger amounts.) He divided waters into four additional types depending upon their amounts of sulfate. Saturation with CaS04 was found to occur only in very high chloride waters. The calcium concentration always is very high - ranging from 150 to 1,100 epm - in high chloride waters which have SO4- > 58 epm and usually is less than 150 epm in high chloride waters where SO4- < 58 epm. All petroleum waters even if saturated in CaS04 have a low S04/C1 ratio which is attributed t o reduction of sulfates and high concentrations of chloride. The ratio never exceeds one except in low or normal chloride waters. The third subgroup contains three additional types depending upon the amounts of bicarbonate and carbonate in the waters. The preferred formula for this calculation is Y(HCO,- + C03- ) (Ca+ ) which is proportional to the gaseous pressure of C 0 2 in equilibrium with CaCO, in the water. As the Cl- increases, the tendency is for Ca+ t o increase and HC03- to decrease; however, because the Ca+ increases, the product of y(HCO,- + C03- ) (Ca+ ) does not vary greatly. As the waters move in their subsurface environment their dissolved ions have a tendency to exchange with those in the rocks. Two extreme types of adsorption can be noted in addition to intermediate types of adsorption. The extreme types are a physical adsorption or the Van der Waals adsorption with weak bonding between the adsorbent and the constituent adsorbed and a chemical adsorption with strong valence bonds. Both of these adsorptions can act simultaneously. Cations can be fixed at the surface and in the interior of the associated minerals. These fixed cations can exchange with the cations in the water. When the exchange occurs, there is an exchange of bases. With the right physical conditions of the adsorbent, similar exchange can occur with the anions. Some of the formation constituents that are capable of exchange and adsorption are argillaceous minerals, zeolites, ferric hydroxide, and certain organic compounds. Particle size influences rates and capacities, if the solids are clays such as illite and kaolinite. The.rate increases with decreasing particle size. However, if a larger mineral has a lattice, the exchange can easily occur on the plates.

27 0

CLASSIFICATION OF OILFIELD WATERS

The concentration of exchangeable ions in the adsorbent and in the water is important. More exchange will occur when the solution is highly concentrated. Schoeller (1955) used the formula:

(a-x)=K

( a E x ) IP

to indicate the relationship that exists between the initial concentration, a , of the cations in milliequivalents in the unreacted water, and x , which equals the final concentration of the cations in milliequivalents in the water after equilibrium or reaction with the rocks. The amounts of cations exchanged by passing from the liquid t o the rock or clay is a - x and the index of base exchange (IBE) = (a - x ) / a . By substitution:

The IBE is used to indicate the ratio between the exchanged ions and the same ions as they originally existed. For example, assume that in the original ' and that when ) , water there were as many equivalents of C1- as (Na+ + K the Na+ and K+ of the water exchanged with the alkaline earths in the rocks alkaline exchange occurred, then: IBE = C1-

- (Na+ + K+)
c1

and this value is positive if the equivalents are Cl- > (Na' + K ' Theoret) . ically all the halides should be included as C1- and all the alkalies as Na+ or (Na+ + K ' ) . However, when the alkaline earth ions in the water exchange for alkali metal ions on the rocks then: Cl- - (Na+ + K + ) IBE = SO4- + HC03- + NO3and this value is negative if the equivalents are Cl- < (Na+ + K+). The lack of equilibrium between the halides and the alkalies is not always a characteristic of base exchange because sea water has a positive value without the occurrence of base exchange. Negative values usually are observed for water coming from altered crystalline rocks. Waters with an IBE equal t o or greater than 0.129 can be true connate petroleum reservoir waters. Waters with a negative IBE are waters of meteoric origin that have infiltrated into marine sediments. Comparison of petroleum-reservoir waters with other types of subsurface waters revealed that the other waters have most of the same characteristics a much higher SO4- concentration and a lower or Kr. Waters that are in contact with organic matter

SCHOELLERS SYSTEM

271

(other than petroleum), such as bitumens, lignites, and coals, resemble petroleum reservoir water, but the frequency of a Kr above normal is greater in petroleum associated waters. Waters related to magmatic reactions commonly possess high concentrations of HC03 . Schoellers (1955) study of petroleum reservoir waters indicated that a positive IBE is more frequent as the C1- increases. A negative IBE is more frequent as the C1- decreases, and a negative value is predominant in low and normal chloride waters associated with petroleum. In fact, this characteristic appears specific for petroleum reservoir waters since in other subsurface waters a positive index occurs as frequently as a negative index. Ancient sea water (connate water) deposited with the sediments usually has an IBE > 0.129 and a Cl/Na > 1.17. Intruding meteoric water in sedimentary marine rock has an IBE < 0.129 and Cl/Na < 1.17. Petroleumreservoir waters with an IBE greater than sea water 0.129 also have the characteristics Cl/Na > 1.17, Cl/Ca < 26.8, Cl/Mg > 5.13, Mg/Ca < 5.24; a very high value for $(HCO,-) (Ca+*) indicating sulfate reduction; low concentrations of HC03-; and frequent high concentrations of NH4 +. Petroleum-reservoir waters containing infiltrating meteoric water mixed with ancient sea water have an IBE less than sea water, 0.129, and the characteristics Cl/Na < 1.17, the ratio Mg/Ca increases and approaches but never equals 5.24, and the ratios Na/Ca and Na/Mg decrease as the dissolved solids increase.

Gases in petroleum-reservoir waters


Schoeller (1955) noted that there should be equilibrium between the free petroleum gases and those dissolved in the water. Considering the solubility of the gases, those in the water should reflect the composition of the petroleum. Components characteristic of petroleum accumulations are ethane, propane, butane, pentane, ethylene, and propylene. Associated components, which may also be present in volcanic waters and in waters in contact with other organic matter such as coal, peat, and lignite as well as in petroleumassociated waters, are favorable components. These are methane, carbon dioxide, organic nitrogen, hydrogen sulfide, helium, radon, and the absence of oxygen. Other components or universal components found in all types of waters are nitrogen and argon. The top waters can contain gases such as Hz S, C 0 2 , and CH4 but because they do not contact the petroleum deposit they are not similar in composition t o the bottom waters. The edge waters are in contact with the petroleum and are characterized by higher amounts of HC03, sulfate reduction, and the presence of H2S , NH4, and small amounts of dissolved hydrocarbons.

27 2

CLASSIFICATION OF OILFIELD WATERS

Oilfield brine analyses About 4,000 oilfield brine analyses were classified. All of these analyses are now in the U.S. Bureau of Mines (1965) open-file report on oilfield brines. The data are on automatic data processing magnetic tape as well as in a file of computer printout sheets listed by State, county, sedimentary basin, formation, etc. These brine data were collected by the Bureau of Mines because the value of oilfield brine analyses in the study of various petroleum-related problems was recognized early in the history of petroleum and natural gas. Before 1928 the Bureau of Mines had indicated in several reports (Ambrose, 1921; Swigart and Schwarzenbek, 1921; Collom, 1922; Mills, 1925; Reistle, 1927) ways in which the analyses could be used. In earlier years, confusion existed because of greatly varying methods of analysis that were used. A paper presenting the methods used by the Bureau of Mines was published by Reistle and Lane (1928). This system of determining the characteristic constituents of oilfield waters and of calculating and reporting results was widely adopted by the petroleum industry. Later the Bureau of Mines cooperated with several interested agencies, and a more detailed report with more modern methods of analyzing oilfield waters was published (American Petroleum Institute, 1968). The Bureau of Mines has an oilfield water analysis laboratory at the Bartlesville Energy Research Center, Bartlesville, Oklahoma. These 4,000 samples were analyzed at this laboratory. Analysis methods The specific gravity of each sample is determined so that a correct aliquot size can be taken for a specific ion analysis. Chloride is determined by titration with silver nitrate, carbonate and bicarbonate are determined by titration with a standard acid, and a pH meter is used to determine the end points. This alkalinity determination should be completed at the time of sampling for accur?te results. However, most of the data for the 4,000 samples were obtained by analysis in the laboratory and were completed within 6 t o more than 48 hours after sampling. Therefore, the alkalinity data cannot be considered absolute but only relative. The calcium was determined by titration of calcium oxalate with permanganate until about 1957, about which time it was determined by complexometric titration such as with disodium ethylenediametetraacetic acid (EDTA) until about 1969; since then, it has been determined by atomic absorption. Magnesium determination has a similar history. I t was precipitated as the pyrophosphate until about 1957, and titrated with EDTA until 1969, from then t o now it has been determined by atomic absorption. Sulfate was determined by precipitation as barium sulfate, and this method still is used. Sodium was determined by calculation from the difference

OILFIELD BRINE ANALYSES

273

between the reacting values of the assumed total anions and cations until about 1960, after which time it was determined by flame photometry or atomic absorption. Several other analyses for dissolved constituents in oilfield brines are now made by the Bureau of Mines. For example, potassium, lithium, rubidium, and cesium are determined by atomic absorption or flame photometry; strontium and barium by atomic absorption; manganese, iron, and boron by atomic absorption or titrimetric methods; and bromide and iodide by titrimetric methods. The precision of the methods is as follows: alkalinity, 2-3% of the amount present; sodium, 2--5% of the amount present; calcium and magnesium, 4--5% of the amount present; sulfate, 1-2% of the amount present; chloride, 1% of the amount present. If sodium is calculated, the precision value reflects the sum of the precision data for the data from which it is calculated plus the undetermined dissolved constituents. The significant figures for the analytical data are all the certain digits and only the first doubtful digit. This number usually is limited to three significant figures except for specific gravity, where four or five are common. It often is recognized that the sampling method is as important as the analytical method. This certainly is true of oilfield brines. Field sampling methods Most of the 4,000 samples were obtained only from wells where reasonable assurance was evident that the formation brine was not contaminated by drilling fluids or by intrusion of water from other formations. Wells were selected on the basis of age, type of completion, and production of fluids. Samples were not taken from some gasfields because of the likelihood of dilution by water condensed from vapor carried up the hole with the gas. Some samples were taken from gas-condensate wells that produced large volumes of brine. In many cases the electrical resistivity measurement was made on the sample at the time the sample was taken, and resulting values were compared with measurements from other samples from the same formation within that field or nearby fields. Obvious discrepancies were eliminated by sampling additional wells. Nearly all samples were withdrawn at production wellheads, and the water was separated from the oil in portable separators. A few samples were taken of brines from formations that did not produce enough water t o permit taking samples at individual wellheads. Such samples were taken from gunbarrels or oil-water separators. Samples were taken in clean, 1-gallon glass jugs that were first rinsed several times with the water sampled and then filled, capped, and labeled. In a few instances samples were obtained by the producer from comparatively isolated small pools or fields and shipped t o the laboratory in 1-gallon polyethylene jugs.

274

CLASSIFICATION OF OILFIELD WATERS

Application of the classification systems Several investigators have applied the classification systems t o determine their usefulness in studies related t o exploration and production of petroleum. Ostroff (1967) concluded that Palmers system is less useful than the systems of Sulin and Schoeller. Further he concluded that the Sulin system is more applicable to petroleum formation waters because many such waters contain more than 2,000 epm of dissolved solids, and the Schoeller system tends to lump these highly concentrated waters together. The index of base exchange (IBE) in Schoellers system, however, appears to have merit for certain interpretations as does theJ Ca+* x SO4-*. Dickey (1966) concluded that the Palmer system does not correlate very well with the geology of oil reservoirs, that Schoellers nomograph is useful, but that the Sulin system appears t o conform better with geology. He also noted that relating water types to geology and flow patterns should be useful in exploration and that distinguishing between a stagnant water and artesian related waters should be highly significant in a new oilfield development area. The Sulin system appears t o be more generally applicable than the other systems in studies of waters from petroleum reservoirs. Because of this and because the Schoeller system appears t o have merit in studying certain types of oilfield waters, it was decided t o apply portions of the Sulin system, the Schoeller system, and Bojarskis modification of the Sulin system t o a study of brines from various sedimentary basins of the United States that are known to be related to petroleum and natural gas.

Calculations
The analyses of most oilfield waters are reported in units of milligrams per liter (mg/l). The conversion of mg/l to equivalents per million (epm) is done using the following formula: mg/l of ion = epm ion atomic weight of ion specific gravity of brine x valence of ion Table 8.VII provides formulas for calculating the epm for many of the common constituents found in oilfield brines. If the constituent is reported in parts per million (ppm), it is not necessary t o divide by the specific gravity of the brine. The sum of the epms (Z epm) shown in Table 8.VII are converted to Zr/100 g of water for the Sulin calculations by moving the decimal to the left in the Z epm and calling this Zr. The term s is used t o indicate the percent of equivalent of a given constituent. The percentage equivalent (s) of each ion is determined by dividing the equivalents per 100 g of water by the total equivalent in 100 g of water. For example, if the r for sodium equals

APPLICATION OF THE CLASSIFICATION SYSTEMS TABLE 8.VII

27 5

Formulas for converting milligrams per liter to equivalents per million for constituents commonly found in oilfield waters Lithium Potassium Sodium Magnesium Calcium Strontium Barium Carbonate Bicarbonate Sulfate Chloride Bromide Iodide mg/l Li+ mg/l K+ mg/l Na+ mg/l Mg+ mg/l Ca+ mg/l Sr+ mg/l Ba+ mg/l co3- mg/l HC03mg/l SO^+ mg/l C1mg/l Brmg/l Ix x x x x x x x

x x
x x x

0.1442/sp. gr.* = O.O256/sp. gr. = O.O435/sp. gr. = O.O823/sp. gr. = O.O499/sp. gr. = O.O228/sp. gr. = O.O146/sp. gr. = O.O333/sp. gr. = O.O164/sp. gr. = 0.0208/sp.gr. = O.O282/sp. gr. = O.O125/sp. gr. = O.O079/sp. gr. = Zepm
=

epm Li+ epm K+ epm Na+ epm Mg+ epm Ca+ epm Sr+ epm Ba+ epm c03- epm HC03eprnS04- epm C1epm Brepm I-

* Specific gravity.

200.6 and the Zr for the total equivalents equals 518.8, then 200.6/518.8 x 100 = 38.7, or 38.7 percentage equivalents for sodium. The Sulin classification considers only the macro constituents; if ions such as potassium, lithium, strontium, barium, bromide, and iodide are determined, they should be added t o their associated macro constituents t o be properly considered in the analysis report. For example, when sodium, potassium, and lithium are determined by atomic absorption the total mg/l of each is reported. Therefore, the epm Na + epm K + epm Li should be added to obtain the correct r value for sodium:
+ + +

rNa= rCa= rC1 =

epmCa epm C1

e p m S r + epmBa 10 10 epm Br 10 epm I 10

The r values then are divided by the Zr and multipIied by 100 to obtain the s value or percentage equivalents. The s values are used to determine the type, class, and other Sulin values of the water as illustrated in Table 8.1, where a = sNa, b = sCa + sMg, and d = sC1+ s S 0 , . The epm values are used to determine the Schoeller characteristics such as the degree of chloridization, the degree of sulfation, IBE, etc., illustrated in Table 8.VI.

epm Na

epm K 10

epm Li 10

27 6

CLASSIFICATION O F OILFIELD WATERS

TABLE 8.VIII Classification of some oilfield waters from 1 0 formations in eight sedimentary basins No. State Formation Basin Depth Concentration (e pm)

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30
31

32 33 34 35 36 37 38 39 40

Kans. Kans. Kans. Kans. Kans. Kans. Kans. Kans. Kans. Kans. Kans. Kans. Kans. Kans. Kans. Kans. Kans. Kans. Kans. Kans. Okla. Okla. Okla. Okla. Okla. Okla. Okla. Okla. Okla. Okla. Ark. Ark. Ark. Texas Texas Texas Texas Texas Texas Texas

Arbuckle Arbuckle Arbuckle Arbuckle Arbuckle Arbuckle Arbuckle Arb u ck le Arbuckle Arbuckle Lansing Lansing Lansing Lansing Lansing Lansing Lansing Lansing Lansing Lansing Wilcox Wilcox Wilcox Wilcox Wilcox Wilcox Wilcox Wilcox Wilcox Wilcox
Nacatoch

Cent .Kans. Cent .Kans. Cent.Kans. Cent.Kans. Cent.Kans. Cant.Kans. Cent.Kans. Cent.Kans. Cent .Kans. Cent .Kans. Cent. Kans. Cent.Kans. Cent.Kans. Cent .Kans. Cent. Kans. Cent. Kans. Cent .Kans. Cent.Kans. Cent.Kans. Cent .Kans. Cherokee Cherokee Cherokee Cherokee Cherokee Cherokee Cherokee Cherokee Cherokee Cherokee
E.Texas

1,050 1,091 1,023 1,102 992 1,152 1,174 949 1,195 1,104 928 1,075 966 999 902 1,009 1,063 1,148 1,172 853 1,228 1,731 904 1,539 1,106 1,020 582 1,432 1,97 2 1,865
360

634.6 581.6 282.0 639.9 430.3 458.7 473.2 356.4 446.4 577.4 1,759.6 570.4 1,899.9 1,728.2 1,903.8 2,514.8 1,854.7 2,087.2 2,414.5 1,898.4 2,366.5 2,188.4 2,161.5 2,365.3 1,997.1 1,990.0 1,587.2 2,459.0 2,701.4 2,635.5
233.4

60.6 58.4 24.8 56.6 38.8 34.4 53.1 57.4 35.4 49.4 266.6 117.4 266.4 232.0 274.5 309.3 225.2 223.6 251.0 210.8 193.3 209.5 163.2 200.8 140.4 161.3 151.2 174.5 153.1 180.6
9.0

133.0 121.8 40.2 13.0 65.4 81.3 119.4 100.4 71.2 112.1 373.9 174.6 442.1 512.4 478.8 532.7 449.8 384.8 519.0 416.7 810.1 529.4 530.3 445.6 457.3 513.7 287.5 511.6 606.0 629.6
22.4

Nacatoch Nacatoch Nacatoch Nacatoch Nacatoch Nacatoch Nacatoch Nacatoch Nacatoch

E.Texas E.Texas E.Texas E.Texas E.Texas E.Texas E.Texas E.Texas E.Texas

465 373 905 70 1 242 650 283 191 181

463.4 300.4 788.2 481.8 274.0 492.7 295.4 451.5 490.6

39.5 25.9 9.4 7.8 8.1 10.4 8.1 17.6 3.6

59.5 36.8 28.7 17.3 14.7 16.5 14.7 44.7 7.8

7.7 6.5 9.5 2.2 8.5 3.4 5.1 12.8 9.0 35.9 1.8 2.4 0.3 1.7 1.1 0.4 0.5 1.4 0.7 0.8 1.1 0.3 0.4 1.1 0.9 0.7 1.8 1.1 0.7 1.1 37 . 1.4 4.3 3.1 3.3 2.5 18.9 2.4 7.1 2.8

49.0 22.7 22.1 46.6 17.5 45.9 36.0 39.9 30.0 45.5 0.0 34.9 0.8 2.5 0.0 1.5 1.1 1.o 2.4 0.7 7.7 9.8 7.3 15.9 7.8 13.3 1.4 6.0 8.4 8.6
0.0

0.0 1.0 0.0 0.0 0.8 0.1 0.9 0.4 0.6

773.8 732.3 313.9 827.0 510.9 523.2 705.0 456.7 512.7 685.8 2,398.9 826.6 2,605.0 2,469.4 2,655.6 3,353.6 2,529.6 2,706.1 3,17 9.9 2,515.5 3,360.6 2,917.2 2,847.4 2,992.0 2,584.3 2,650.8 2,021.6 3,138.8 3,449.1 3,436.6 261.1 559.5 357.7 824.6 503.5 294.2 498.4 313.6 506.2 498,4

* Chloride (epm): (VH)C = 700, (MC) = 420-700, (L)C = 10-40, (N)C = 10. Sulfate (epm): (VH) = =<6.

<

>

> 58, ( H )

(H)C = 140-420, (A)C = 40-140, = 24-58, (A ) = 6-24, : (N )

APPLICATION OF THE CLASSIFICATION SYSTEMS

277

Sulin Zepm type class

Schoeller*

CT

so4-
80.7 52.6 29.8 77.8 33.8 61.1 65.6 62.3 46.2 71.4 0.0 78.0 19.4 36.0 0.0 28.7 22.5 20.4 35.9 17.8 79.1 72.3 62.3 84.3 59.9 82.7 20.5 55.7 71.4 73.6 0.6 0.0 6.0 0.0 0.0 36 . 1.6 36 . 45 . 2.2 20.0 17.4 15.4 8.7 16.8 9.9 14.6 25.2 17.9 52.5 10.7 10.2 3.5 11.6 85 . 4.4 51 . 9.1 6.3 6.6 99 . 3.6 4.7 8.5 71 . 6.4 9.9 87 . 71 . 94 . 6.8 49 . 88 . 65 . 57 . 46 . 18.1 44 . 13.2 40 . 0.18 0.21 0.10 0.23 0.16 0.12 0.19 0.22 01 3 . 0.16 0.27 0.31 0.27 0.30 0.28 0.25 0.27 0.23 0.24 0.25 0.30 0.25 0.24 0.21 0.23 0.25 0.21 0.22 0.22 0.23 0.11 0.17 0.16 0.04 0.04 0.07 0.01 0.06 01 1 . 0.02

1,659.0 Cl-Ca 1,523.6 Cl-Ca 692.6 Cl-Ca 1,585.4 C l - C a 1,071.6 C1-Ca 1,147.2 Cl-Ca 1,492.0 CI-Ca 1,024.0 Cl-Ca 1,105.0 CI-Ca 1,506.4 C1-Ca 4,800.9 CI-Ca 1,726.6 Cl-Ca 5,214.7 Cl-Ca 4,946.5 Cl-Ca 5,314.0 CI-Ca 6,712.5 Cl-Ca 5,061.1 Cl-Ca 5,404.4 Cl-Ca 6,367.7 Cl-Ca 5,043.1 Cl-Ca 6,739.4 Cl-Ca 5,854.9 Cl-Ca 5,710.4 Cl-Ca 6,021.0 Cl-Ca 5,188.0 C l - C a 5,330.0 Cl-Ca 4,051.1 Cl-Ca 6,291.3 C l - C a 6,918.8 Cl-Ca 6,892.3 Cl-Ca 529.8 CI-Ca 1,123.5 C l - C a 726.2 Cl-Ca 1,654.2 C k C a 1.013.8 Cl-Ca 594.5 Cl-Ca 1,037.4 Cl-Mg 635.2 C l - C a 1.027.8 Cl-Ca 1,004.0 C l - C a

278
TABLE 8.VIII (continued)

CLASSIFICATION OF OILFIELD WATERS

No. State

Formation Basin

Depth (m)

Concentration (epm) Na+ Mg+ Ca+ HCOJ- S04-* Cl-

41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65 66 67 68 69 70 71 72 73 74 75 76 77 78 79 80

Ark. Ark. Ark. Ark. Ark. Texas Texas Texas Texas Texas Ark. Ark. Ark. Ark. Texas Texas Texas Texas Texas Texas Texas Texas Texas Texas Texas Texas Texas Texas Texas Texas Ala. Ala. Ala. Miss. Miss. Miss. Miss. Miss. Miss.

Ms. is

Paluxy Paluxy Paluxy Paluxy Paluxy Paluxy Paluxy Paluxy Paluxy Paluxy Rodessa Rodessa Rodessa Rodessa Rodessa Rodessa Rodessa Rodessa Rodessa Rodessa Woodbine Woodbine Woodbine Woodbine Woodbine Woodbine Woodbine Woodbine Woodbine Woodbrine Eutaw Eutaw Eutaw Eutaw Eutaw Eutaw Eutaw Eutaw Eutaw Eutaw

E.Texas E.Texas E.Texas E.Texas E.Texas E.Texas E.Texas E.Texas E.Texas E.Texas E.Texas E.Texas E.Texas E.Texas E.Texas E.Texas E.Texas E.Texas E.Texas E.Texas E.Texas E.Texas E.Texas E.Texas E.Texas E.Texas E.Texas E.Texas E.Texas E.Texas Interior Sal. Interior Sal. Interior Sal. Interior Sal. Interior Sal. Interior Sal. InteriorSal. Interior Sal. Interior Sal. InteriorSal.

1,115 737 1,297 884 1,417 2,340 2,174 1,943 1,512 1,943 1,897 1,241 1,033 711 2,844 2,519 2,722 2,115 2,289 3,062 1,047 1,750 898 841 1,809 1,144 925 1,442 1,596 1,332 1,061 972 1,060 1,444 2,263 1,312 2,443 1,690 1,315 2,469

1,246.8 90.6 586.0 32.2 1,310.3 107.1 934.9 64.8 1,507.2 138.6 1,642.1 22.5
1,515.1 51.1

1,495.4 89.0 205.3 6.4 1,522.8 90.2 1,971.8 157.2 1,943.8 185.1 1,060.3 108.2 538.6 35.9 1,772.1 127.3 1,861.9 217.4 2,068.5 148.5 1,950.3 139.9 2,020.0 141.2 1,877.5 92.7 1,507.0 44.3 1,263.0 33.4 341.8 4.4 1,060.5 32.4 1,271.9 59.6 809.1 12.7 478.8 12.4 1,451.3 26.7 1,447.3 38.3 1,268.0 30.2 1,124.4 45.0 1,102.9 25.6 1,186.7 59.6 1,733.6 82.7 2,009.8 75.9 1,572.7 88.9 1,721.3 256.7 1,925.9 61.6 1,579.6 60.6 2,153.7 87.7

254.1 86.7 278.5 197.6 504.4 378.5 205.9 448.9 10.0 448.1 669.7 563.6 289.8 75.0 878.8 1,084.7 900.9 726.1 741.8 610.0 161.4 56.3 9.3 58.9 154.9 61.1 23.6 213.3 144.1 81.9 164.7 160.5 164.3 287.3 306.8 224.9 458.0 359.9 252.9 518.9

0.4 2.8 2.7 1.5 0.5 00 . 4.7 2.7 14.9 1.8 00 . 0.9 0.0 3.0 1.5 1.1 0.6 1.8 0.8 1.0 42 . 8.2 11.3 00 . 1.0 7.3 7.0 39 . 16 . 8.3 2.7 2.0 29 . 0.9 52 . 09 . 0.0 3.9 23 . 0.0

0.0 33 . . 7 1.1 33 . 9.3 3.7 8.2 6.0 8.3 8.0 11.7 21.4 29 . 2.8 27 . 48 . 8.5 4.2 5.4 3.5 1.8 0.6 0.2 4.9 3.7 01 . 0.0 0.1 40 . 0.0 0.0 00 . 0.0 33.6 0.0 0.0 0.0 0.0 1.0

1,594.4 699.0 1,692.0 1,194.9 2,208.7 2,033.8 1,763.1 2,022.2 202.4 2,050.3 2,789.9 2,679.0 1,436.5 643.4 2,773.2 3,165.1 3,111.6 2,802.4 2,909.8 2,576 9 . 1,705.0 1,342.2 343.5 1,161.4 1,478.2 873.3 507.7 1,685.1 1,629.4 1,367.4 1,330.1 1,288.9 1,411.8 2,108.5 2,358.5 1,890.2 2,433.4 2,352.4 1,893.5 2,765.6

* Chloride (epm): (VH)C 700, (MC) = 420-700, (H)C = 140-420, (A)C = 40-140, (L)C = 10-40, (N)C = 10. Sulfate (epm): (VH) = > 58,(H) = 24-58, (A) = 6-24, (N)
- < 6.

<

=->

APPLICATION OF THE CLASSIFICATION SYSTEMS

279

Sulin Zepm type class

Schoeller*

d c a + ' + so4-'

V(HCO~- coz-2 ) ? ( ~ a +IBE + ~)


38 . 8.8 12.8 76 . 5.4 00 . 16.6 13.6 13.1 11.1 0.5 8.0 0.4 8.8 12.9 11.1 7.3 13.5 7.7 8.4 14.2 15.6 10.6 00 . 5.4 14.8 10.5 14.8 7.4 17.8 10.6 8.8 11.1 6.2 20.3 57 . 00 . 17.8 11.0 0.0 0.22 0.16 0.23 0.22 0.29 0.19 0.14 0.26 0.0 0.26 0.29 0.27 0.26 0.16 0.36 0.41 0.34 0.30 0.31 0.27 0.12 0.06 0.oo 0.09 0.14 0.07 0.06 0.14 0.11 0.07 0.15 0.14 0.16 0.18 0.15 01 7 . 0.29 0.18 0.17 0.22

_-

CT

SO^-'
3.1 17.1 14.1 15.0 41.0 59.5 27.9 60.9 7.7 61.0 73.3 81.3 78.7 14.8 50.1 54.4 66.2 79.0 55.9 57.4 23.9 10.0 2.4 4.0 27.6 15.1 15 . 00 . 37 . 18.1 01 . 0.1 0.1 0.1 101.6 0.0 0.0 00 . 0.0 22.7

3,186.6 CI-Ca 1,410.3 CI-Ca 3,391.5 Cl-Ca 2,395.0 CI-Ca 4,363.0 CI-Ca 4,086.4 Cl-Ca 3,543.9 Cl-Ca 4,066.7 Cl-Ca 445.2 SO4-N 4,121.6 CI-Ca 5,596.8 Cl-Ca 5,384.3 Cl-Ca 2,916.3 CI-Ca 1,299.0 CI-Ca 5,556.0 Cl-Ca 6,333.1 Cl-Ca 6,235.2 Cl-Ca 5,622.2 CI-Ca 5,818.1 CI-Ca 5,163.7 CI-Ca 3,425.6 Cl-Ca 2,705.0 CI-Ca 711.0 CI-Mg 2,313.7 Cl-Ca 2,970.7 CI-Ca 1,767.3 C1-Ca 1,029.9 Cl-Ca 3,379.5 Cl-Ca 3,261.0 Cl-Ca 2,760.0 Cl-Ca 2,667.1 Cl-Ca 2,580.1 Cl-Ca 2,825.4 CI-Ca 4,213.1 Cl-Ca 4,789.9 Cl-Ca 3,777.7 Cl-Ca 4,869.6 Cl-Ca 4,704.8 Cl-Ca 3,789.1 Cl-Ca 5,527.1 Cl-Ca

280

CLASSIFICATION OF OILFIELD WATERS

TABLE 8.VIII (continued) No. State Formation Basin Depth (m) Concentration Na+
2,031.8 1,847.9 1,362.9 2,157.3 1,972.2 1,992.4 2,043.7 2,198.7 1,280.8 1,318.9 1,589.7 2,444.9 2,518.1 2,790.1 2,418.0 2,729.0 4,277.5 2,225.4 1,971.2 681.8 1,718.5 2,035.8 1,206.3 1,874.2 2,138.9 1,057.5 1,379.4 2,119.4 2,132.3 840.1 3,431.2

-___
81 82 83 84

Mg+2
36.8 48.4 15.0 54.1 65.8 38.0 25.5 83.8 36.7 20.9 86.7 275.9 280.4 222.2 279.5 218.6 10.0 149.7 173.5 35.2 44.9 46.5 43.6 40.9 41.9 26.6 41.8 43.7 15.3 26.8 100.6

Ca+2
100.3 94.9 82.5 89.9 95.7 81.9 132.2 90.7 56.3 99.6 1,256.8 1,392.8 1,622.1 1,641.6 1,470.8 1,598.2 34.9 2,282.1 1,668.9 63.6 90.1 124.7 47.2 91.0 96.2 51.0 60.0 100.2 115.2 24.1 993.7 4.5 3.8 4.4 6.8 3.9 3.0 7.2 3.9 9.4 3.9 0.0 2.0 1.8 2.3 0.0 2.4 2.4 0.0 0.0 5.1 1.4 5.6 4.6 0.7 1.4 0.0 2.5 6.1 4.0 7.0 1.5
0.0 0.0 0.0 0.0

c12,163.3 1,985.9 1,455.8 2,114.0 2,122.3 2,108.5 2,194.5 2,369.8 1,364.3 1,434.8 2,940.8 4,105.9 4,413.8 4,648.6 4,163.0 4,540.4 4,312.9 4,654.9 3,810.9 778.6 1,851.2 2,199.9 1,291.2 2,003.8 2,273.4 1,058.4 1,477.6 2,263.4 2,256.4 888.4 4,305.3

Miss.

Ms. is
Miss. Miss, Miss. Miss. Miss. Miss. Miss. Miss. La. Ark. Ark. Ark. Ark. Ark. Ark. La. La. La. La. La. La. La. La.

Wilcox Wilcox Wilcox Wilcox Wilcox Wilcox Wilcox Wilcox Wilcox Wilcox Smackover Smackover Smackover Smackover Smackover Smackover Smackover Smackover Smackover Smackover Wilcox Wilcox Wilcox Wilcox Wilcox Wilcox Wilcox Wilcox Wilcox Wilcox Silurian

Miocene Miocene Miocene Miocene Miocene Miocene Miocene Miocene Miocene Miocene N .Louisiana N.Louisiana N.Louisiana N.Louisiana N.Louisiana N. Louisiana N.Louisiana N.Louisiana N.Louisiana N.Louisiana W. Gulf W. Gulf W. Gulf W. Gulf W. Gulf W. Gulf W. Gulf W. Gulf W. Gulf W. Gulf Williston

1,975 1,748 1,412 1,871 2,330 1,646 2,162 2,268 1,552 1,327 3,109 2,103 2,399 2,509 2,240 2,526 2,615 2,949 3,271 701 1,399 1,814 747 1,561 1,722 666 1,124 2,441 2,158 471 3,633

85 86 87 88 89 90 91 92 93 94 95 96 97 98 99 100 101 102 103 204 105 106 107 108 109 110 111

La.
La. La. La. La. N.D.

7.2 0.0 0.0 0.3 0.0 0.0 8.8 4.6 3.7 1.7 4.1 1.7 0.0 1.4 1.8 2.1 0.2 0.8 0.6 0.8 1.4 0.0 0.6 0.0 8.0 0.0 8.4

* Chloride (epm): (VH)C = 700, (MC) = 420-700, (H)C = 140-420, (A)C = 4040, (L)C = 10-40, (N)C = < 10. Sulfate (epm): (VH) = 58, (H) = 24-58, (A) = 6-24, (N) =<6.

>

>

APPLICATION OF T H E CLASSIFICATION SYSTEMS

281

Cepm

type

class

~ 1 -

SO^-^
~

4,336.9 3,981.1 2,920.8 4,422.4 4,267.3 4,223.9 4,403.4 4,747.3 2,747.7 2,878.2 5,883.0 8,226.2 8,840.1 9,306.8 8,335.6 4,454.9 8,637.9 9,313.8 7,626.5 1,566.7 3,706.6 4,413.6 2,593.7 4,011.7 4,55 3.5 2,193.6 2,962.2 4,533.4 4,531.3 1,786.7 8,840.7

Cl-Ca Cl-Ca Cl-Ca HCO, -Na Cl-Ca Cl-Ca Cl-Ca Cl-Ca Cl-Ca Cl-Ca Cl-Ca Cl-Ca Cl-Ca Cl-Ca Cl-Ca Cl-Ca Cl-Ca Cl-Ca Cl-Ca Cl-Ca Cl-Ca Cl-Ca C1-Ca Cl-Ca Cl-Ca Cl-Mg Cl-Ca Cl-Ca Cl-Ca Cl-Ca Cl-Ca

0.0

Q.0
0.0 0.0

12.7 11.2 11.8 16.1 11.4 9.0 18.9 11.1 17.0 11.5 0.0 17.7 17.9 21.1 0.6 21.3 5.9 0.7 0.6 11.6 5.6 15.8 10.1 3.6 5.9 0.0 7.3 15.2 12.1 9.9 13.0

0.06 0.07 0.06 0.0 0.7 0.06 0.07 0.07 0.06 0.08 0.46 0.40 0.43 0.40 0.42 0.40 0.01 0.52 0.48 0.12 0.07 0.07 0.07 0.06 0.06 0.0 0.07 0.06 0.05 0.05 0.20

26.2 0.0 0.1 5.4 0.0 0.0 103.6 80.0 78.0 53.5 78.0 53.4 0.0 57.5 55.5 11.4 5.1 10.2 5.6 8.8 11.7 0.0 6.1 0.0 29.9 0.0 90.6

282

CLASSIFICATION OF OILFIELD WATERS

Interpretation o f the classification


The majority of the 4,000 oilfield waters that were classified fell in the S ,

S2 A2 class and were of the chloride-calcium type. Table 8.VIII illustrates


the classification data for some of the samples from the Arbuckle, and Lansing Kansas City formations in the Central Kansas Basin; the Wilcox formation in the Cherokee Basin; the Nacatoch, Paluxy, Rodessa, and Woodbine formations in the East Texas Basin; the Eutaw formation in the Interior Salt Basin; the Smackover formation in the North Louisiana Basin; and the Wilcox formation in the western Gulf Basin. Most of the samples of the bicarbonate-sodium and sulfate-sodium types were found at relatively shallow depths. This could indicate that the waters contained infiltrating water. The Cl/Na ratio was determined for each sample as suggested by Bojarski. This ratio was calculated from the epm values, and the ratios ranged from 0.33 to values greater than 1. Values greater than 0.85 are characteristic of hydrodynamic waters, according t o Bojarski (1970),while the more altered waters found in static environments have ratios of less than 0.50. Schoeller's classification uses the chloride concentration t o separate the waters into six types. The majority of the 4,000 oilfield waters analyzed for this study were very high chloride waters (where the chloride epm is equal to or greater than 700). The sulfate concentration of these waters according to his classification was not as consistent. Many of them were normal with sulfate epm less than 6. However, in several waters the sulfate epm was higher than 24. Few waters contained sulfate in excess of 58 epm. The JCa+' x SO4-' exceeded 70 in some waters, indicating that such waters were nearly saturated with calcium sulfate. Schoeller did not consider that the solubility of calcium sulfate increases in the presence of certain other dissolved ions; therefore the value of 70 may not always indicate saturation. However, in primary and secondary recovery operations this value should be considered. The 4 ( H C 0 3 - + C03-' )' (Ca+' ) was used by Schoeller to determine if a water was saturated with calcium carbonate, and such a water should have a value greater than 7. This is not entirely accurate, but the formula does indicate if the water contains an excess of calcium, which decreases the carbonate concentration. Many of the 4,000 waters evaluated had values greater than 7. A distilled water thus saturated would deposit precipitated calcium carbonate, but the activities of other ions dissolved in a brine cause the solubility product t o be different in the brine. The predominant cation sequence for these 4,000 oilfield brines was Na+ > Ca+? > Mg+'. The SO4/Cl ratio ranged from 0.00 to 0.34.The ratio 0.34 was found for a bicarbonate-sodium water sample, which may have contained infiltrating water. None of the chloride-calcium type waters had a S04/Cl ratio greater than 0.17, with many of them 0.00. The IBE indicates that exchange of metal ions dissolved in the water have

APPLICATION OF THE CLASSIFICATION SYSTEMS

283

exchanged with metal ions on the clays. (Schoeller made the arbitrary assumption that the concentrations of sodium and chloride originally present in the water were equal.) If the IBE is a positive number the exchange was alkali metals in the water for alkaline earth metals on the clays, and if the IBE is negative the exchange was alkaline earth metals in the water for alkali metals on the clays. Very few negative numbers were evident when the IBE was determined on the 4,000 oilfield water analyses. This indicates that most of the formation waters associated with hydrocarbons had exchanged dissolved alkali metals for alkaline earth metals on the clays. The few samples that yielded the negative IBE numbers were sulfate-sodium and bicarbonatesodium type waters which, according to Sulin, is indicative of terrestrial derived waters.

Ratios
The ratios Na/(Ca + Mg) and Ca/Mg were calculated from the analytical data for the 4,000 oilfield water samples. Fig.8.1 illustrates a plot of the Na/(Ca + Mg) ratio versus dissolved solids for samples from the East Texas Basin. The ratio tends to decrease with increasing dissolved solids concentration. The depletion of sodium with respect to calcium + magnesium can be attributed to diagenesis of the waters. This correlates with Schoeller's IBE, indicating that the alkali metals in the water exchange with alkaline

I
6

KEY
Tertiary Upper Cretaceous Lower Cretaceous Jurassic

x Pennsylvanian

0
DISSOLVED SOLIDS, g / l

Fig. 8.1. Plot of the Na/(Ca + Mg) ratio versus the concentration of dissolved solids in some formation waters taken from sedimentary rocks in the East Texas Basin.

284

CLASSIFICATION OF OILFIELD WATERS

7-

Fig. 8.2. Plot of the Na/(Ca + Mg) ratio versus the concentration of dissolved solids in some formation waters taken from the Rodessa formation in the East Texas Basin. The line is a least squares fit (y = a + bx + ex2 ) for the scattered data.

earth metals on the clays, decreasing the dissolved alkali metals and increasing the dissolved alkaline earth metals. Fig.8.2 shows the scattered data for the brines taken from the Rodessa formation in the East Texas Basin. The scattered data were submitted t o least squares analysis to determine with more certainty how the ratios varied with depth and with dissolved solids concentrations. The least squares formula used was y = a + bx + c 2or the approximation to a parabola. If the x best fit is a straight line, the solution will regress to it, and this occurred for the data shown in Fig. 8.2. Also, if this occurs, the value for c i n the formula is very small. The curve shown in Fig. 8.2 represents a fit of about 88%of the sum of the mean, a fit of 100%would be ideal. Preliminary plots of the scattered data indicated that a better fit could be obtained with the parabolic least squares approximation than with the least squares approximation for a straight line. A computer was used t o obtain the fit. The least squares curve indicates that the concentration of sodium decreased with respect t o calcium plus magnesium until the dissolved solids were about 210,000 mg/l and then increased. This could be attributed to exchange of sodium in the water for alkaline earths on the minerals until the concentration of dissolved solids reached 210,000 mg/l, at which point the solubility products of the alkaline earths are such that they are unable to stay in solution. This would correlate with Schoellers IBE. However, the IBE does not consider the effect of other ions in solution upon the solubility product of an ion.

APPLICATION OF THE CLASSIFICATION SYSTEMS

285

30 0

Tertiary Upper Cretaceous Lower Cretaceous Jurassic

x Pennsylvanian-

-= 2
0

20-

:Is
100 0 0

.n

n& n
0

35 1

7b

105 I40 I'l5 2;O 245 DISSOLVED SOLIDS, g / l

280

3;5

5 0

Fig. 8.3. Plot of the Ca/Mg ratio versus the concentration of dissolved solids in some n formation waters taken from sedimentary rocks i the East Texas Basin.

Fig. 8.3 illustrates a plot of. the Ca/Mg ratio versus dissolved solids. Here the trends appear t o be that the Ca/Mg ratio increases as the dissolved solids increase. This trend also is related to diagenesis of the waters. The calcium increases as the magnesium decreases, and this may be related to the formation of dolomite and chlorite, or to reactions with argillaceous minerals. . Fig. 8 4 is a plot of Ca/Mg versus dissolved solids. The Ca/Mg ratios for some brines from the Rodessa formation in the East Texas Basin were submitted t o least squares analysis, and the results were used to plot Fig. 8.4. The scattered data in Fig. 8 4 did not yield as good a fit to the formula . y = a + bx + cx2 as did the data in Fig. 8.2, because the fit is about 35%of the mean where 100% is ideal. The line indicates that the ratio increased as the dissolved solids increased. This is the result of magnesium lost from the brine by reactions t o form minerals, ion exchange, or shale membrane filtration. It is not a result of solubility product because most magnesium compounds are more soluble than calcium compounds. Fig. 8.5 is a plot of Na/(Ca + Mg) against depth from which the samples were taken. The trend does not indicate a definite relationship for these samples. Fig. 8.6 is a plot of the least squares analysis data for Na/(Ca + Mg) versus depth for some brines from the Rodessa formation in the East Texas Basin. The plot indicates that the sodium concentration decreases with respect to

286

CLASSIFICATION OF OILFIELD WATERS

calcium plus magnesium until the depth is about 2,350 m and then it increases. The curve in Fig. 8.6 represents a 45% mean fit of the scattered data where an ideal fit is 100%. Fig. 8.7 is a plot of Ca/Mg versus depth, and there appears to be an increase of this ratio with depth which would indicate that magnesium is depleted more with respect t o calcium in brines taken from deeper strata.

7t

21i
I 1
I

20

80

I40 200 DISSOLVED SOLIDS, g/l

260

0 3 1

Fig. 8.4. Plot of the Ca/Mg ratio versus dissolved solids for some formation waters taken from the Rodessa formation in the East Texas Basin. The line is a least squares fit (y = a + bx + cx') for the scattered data.
I "

6C 50 -

o
A

o
x

KEY Cretaceous gulf Cretaceous Comanche Jurassic Pennsylvanian Tertiary

2c 0

..
A

0
0 0

IC

500

1,000

1,500

2,000

2,500

3,000

DEPTH, meters

Fig. 8 5 Plot of the Na/Ca + Mg) ratio versus depth for some formation waters taken .. from sedimentary rocks in the East Texas Basin

APPLICATION OF THE CLASSIFICATION SYSTEMS

287

500

I 1.000

1,500
DEPTH, meters

2,000

2,500

10 0

Fig. 8.6. Plot of the Na/(Ca + Mg) ratio versus depth for some formation waters taken from the Rodessa formation in the East Texas Basin. The line is a least squares fit (y = a + bx + cx2 ) for the scattered data.

3c .
0

o
A

x ' F

I:

2c

KEY Cretaceous gulf Cretaceous Comanche Jurassic Pennsylvanian Tertiary


A

r"
1 0

m
A A

0
A
0

500

I30 f0

I 1,500 2,000 DEPTH, meters

. ..
I

**
3,( 30

2,500

Fig. 8.7. Plot of the Ca/Mg ratio versus depth for some formation waters taken from sedimentary rocks in the East Texas Basin.

This would indicate that magnesium is taken from the brine by reactions to form minerals or absorbed by clays. Fig. 8.8 is a plot of the least squares analysis data for Ca/Mg versus depth for some brines from the Rodessa formation in the East Texas Basin. The plot indicates that the concentration of calcium increases with depth with

288

CLASSIFICATION OF OILFIELD WATERS

1 0

500

I.000

150 .0
DEPTH, meters

2POO

2.500

L 3pOO

Fig. 8.8. Plot of the Ca/Mg ratio versus depth for some formation waters taken from the Rodessa formation in the East Texas Basin. The line is a least squares fit (y = a + bx + cxz ) for the scattered data.

respect t o magnesium. Reactions occurred between the brines and the enclosing rocks to deplete the magnesium or increase the calcium in these brines. The curve in Fig. 8.8 represents a 35% mean fit of the scattered data where an ideal fit is 100%.

Ternary diagram
Ternary or triangular diagrams are useful in studying the distribution of the cations and anions in subsurface waters. The ions which often are used in these diagrams are sodium, calcium, and magnesium or chloride, sulfate, and bicarbonate. To plot these constituents, the equivalent weights of three cation constituents or three anion constituents are determined and summed t o equal 100%.The percentage of the three then are plotted. Dickey (1966) used these types of plots fo study the composition of some deep subsurface waters. Fig. 8.9 is a ternary plot of the normalized values for sodium, calcium, and magnesium concentrations of some brines taken from oil-productive sedimentary basins. This diagram indicates that all of these brines contain more calcium than magnesium. For these particular samples the diagram seems to indicate that the calcium equivalents tend to increase with the age of the rocks from which the brines were taken; this does not, however, apply to all brines.

DISCUSSION

289

Fig. 8.9. Relative concentrations of sodium, calcium, and magnesium in some formation waters taken from sedimentary rocks in oil-productive basins.

Discussion Chebotarev (1955) postulated that most subsurface waters are altered by meteoric water. The three hydrodynamic zones that he considers are: (1) active exchange, where water is influenced by a relatively high degree of hydrodynamic movement with flushing-out of most of the salt thus producing a low salinity water; (2) delayed exchange, where the hydrodynamic flow is less rapid, thus leaving a higher salinity water; and (3) stagnant conditions, where there is little hydrodynamic movement, so that the salinity accumulates and is higher. This system was not applied to the 4,000 waters because it evaluates in a rather complex manner the types and classes of waters from the constituent compositions in reacting values in percent. The method appears useful in certain general types of studies. Palmers system does not consider ionic concentrations or possible conditions of saturation of calcium carbonate or calcium sulfate. Some constituents such as the alkaline earths and chloride and sulfate are lumped togeth-

290

CLASSIFICATION OF OILFIELD WATERS

er. Calcium and magnesium should not be lumped together because even though they are in the same chemical group in the Periodic Table, they often behave quite differently in chemical reactions. For example, the solubilities of many of their salts are considerably different. The same often is true of the chlorides and sulfates. The portion of the Palmer system applied t o the 4,000 brines was used only because it is incorporated in Sulins system. The Palmer system by itself appears to have little value in studying subsurface brines related t o hydrocarbon accumulations. Schoeller separates the waters into six types solely on the basis of the concentration of the chloride ion and into four additional types using the concentration of the sulfate ion. Because of this, his classification of oilfield waters gives considerably more variation in types than the Sulin system, and thus it is more confusing in interpretation. Schoellers formula for determining relative saturations of sulfates and carbonates in the waters has value in production problems. For example, knowledge that the sulfates or carbonates may precipitate from a water under certain conditions is useful so that correct treatment can be applied t o prevent well and/or formation damage. The index of base exchange also is useful in evaluating diagenetic reactions that may have occurred to change the water. Sulins system considers some of the ions in determining the type of water, and the class indicates the predominance of the anion groups. These two characteristics of waters appear useful in studying waters that are likely to be associated with hydrocarbon accumulations. For example, several investigators have determined that the chloride-calcium type in the S1 S2 A2 class of water is the most likely type to be associated with a hydrostatic environment that promotes the accumulation of hydrocarbons. The next most likely type is bicarbonate-sodium in the S1 A l A2 class and the least likely types are chloride-magnesium and sulfate-sodium. Knowledge of the type and class plus what Sulin describes as the significant indicators (direct and indirect) appears useful in hydrocarbon exploration studies. The Sulin system does not consider the degree of saturation of the water with respect to sulfates and carbonates, which is a disadvantage in production problems. Also, the system makes no provision for determining the degree of base exchange which is useful in certain interpretations of diagenesis. Bojarskis modification of the Sulin system appears t o have little value in evaluating a water that is likely to be associated with petroleum. Many of the values that were found for the Na/Cl ratio in the 4,000 samples were greater than 0.85. This value in a chloride-calcium type water according to Bojarski is not likely t o be found in a water associated with petroleum. It is possible that the waters that Bojarski evaluated were consistently very concentrated waters with respect to dissolved solids.

CONCLUSIONS

291

Conclusions Classification of the subsurface waters aids in interpreting the type of water that is more likely t o be associated with a hydrocarbon accumulation. Application of water classification techniques t o 4,000 oilfield brine samples revealed that the majority of the samples were chloride-calcium type of the S1 S2 A2 class, had positive IBE values, had a predominant cation sequence of Na > Ca > Mg, and were very highly concentrated with dissolved chloride. Many of the 4,000 samples had values greater than 7 for C 0 3) (Ca), and some had values greater than 70 for d m , indicating that many of the waters were saturated or nearly saturated with respect to carbonate and sulfate. The Bojarski modification of the Sulin system appears to be of little value. The ratio Na/(Ca + Mg) generally decreased with increasing dissolved solids and depth. The ratio Ca/Mg generally increased with increasing dissolved solids and depth. Studies of the water characteristics and mapping of the more important water properties in conjunction with other geological and geophysical data appear useful in exploration. Maps of certain water characteristics such as the JCa+ x SO,,- and d ( H C 0 3 - + C03- ) (Ca+ ) are useful in solving production problems.

References
Ambrose, A.W., 1921. Underground conditions in oilfields. US.Bur. Min. Bull., 195, 238 PP. American Petroleum Institute, 1968. API Recommended Practice for Analysis o f Oilfield waters. Subcommittee on Analysis of Oilfield Waters, API RP 45, 2nd ed., 49 pp. Bojarski, L., 1970. Die Anwendung der hydrochemischen Klassifikation bei Sucharbeiten auf Erdol. 2.Angew. Geol., 16:123-125. Chebotarev, I.I., 1955. Metamorphism of natural waters in the crust of weathering. Geochim. Cosmochim. Acta, 8:22-48, 137-170,198-212. Collom, R.E., 1922. Prospecting and testing for oil and gas. U.S. Bur. Min. Bull., 201, 170 PP. Dickey, P.A., 1966. Patterns of chemical composition in deep surface waters. Bull. A m . Assoc. Pet. Geol., 50:2472-2478. Elliott, Jr., W.C., 1953. Chemical characteristics of waters from Canyon, Strawn, and Wolfcamp formations in Scurry, Kent, Borden, and Howard Counties, Texas. Pet. Eng., 25 :B7 7-BB9. Mills, R. van A., 1925. Protection of oil and gas field equipment against corrosion. U.S. Bur. Min. Bull., 233, 127 pp. Ostroff, A.G., 1967. Comparison of some formation water classification systems. Bull. A m . Assoc. Pet. Geol., 51:404-416. Palmer, C., 1911. The geochemical interpretation of water analyses. U.S. Geol. Sum. Bull., 749~5-31. Reistle, Jr., C.E., 1927. Identification of oilfield waters by chemical analysis. U.S.Bur. Min. Tech. Paper, No.404, 24 pp. Reistle, Jr., C.E. and Lane, E.C., 1928. A system of analysis of oilfield waters. U.S. Bur. M i a Tech. Paper, No.432, 14 pp.

292

CLASSIFICATION OF OILFIELD WATERS

Rogers, G.S., 1917. Chemical relations of the oilfield waters in San Joaquin Valley, California. U.S. Geol. Sum. Bull., 653:5-116. Rogers, G.S., 1919. Sunset-Midway oil field, California, 11. Geochemical relations of oil, gas, and water. U.S. Geol. Sum. Prof. Paper, No.117, 103 pp. . Schoeller, H ,1955. Geochemie des eaux souterraines. Rev. Znst. Fr. Pet., 10:181-213, 219-246, 507-552. Sulin, V.A., 1946. Waters of Petroleum Formations in the System of Nature Waters. Gostoptekhizdat, Moscow, 96 pp. (in Russian). Swigart, T.E. and Schwanenbek, F.X., 1921. Petroleum engineering in the Hewitt Oilfield, Oklahoma. Cooperative Bulletin published by the US. Bureau of Mines, the State of Oklahoma, and the Ardmore, Oklahoma, Chamber of Commerce, 61 pp. U.S.Bureau of Mines, 1965.Analyses of Oilfield Waters. Open-file Report.

Chupter9.

SOME EFFECTS OF WATER UPON THE GENERATION, MIGRATION, ACCUMULATION, AND ALTERATION OF PETROLEUM

Both water and hydrocarbons are ubiquitous in the crust of the earth and water is in contact with most if not all chemical and physical reactions. The water cycle is comprised of mechanisms whereby it precipitates from the atmosphere t o the crust of the earth and there enters various environments e.g., surface and subsurface, chemical and physical, and biogenic and abiogenic - until it ultimately returns t o the atmosphere. The carbon cycle comprises the reaction of water, minerals, carbon dioxide, and a catalyst (chlorophyll) t o form organic plant material composed of carbohydrates, fats, lignins, proteins, and some hydrocarbons such as alkanes. Production of additional types of organic material results from the consumption of plants by animals. Organic matter from land and marine plants and animals is deposited with inorganic sediments t o form a biomass from which petroleum and natural gas originate. Most of the organic matter that will later produce the hydrocarbons is deposited in an aqueous environment; therefore, it is evident that water profoundly influences not only the hydrocarbon precursors but (as will be demonstrated later) also the generation, migration, accumulation, and alteration of hydrocarbons. The environment controls where and how inorganic and organic matter is deposited. Most of the petroleum source material is marine (autochthonous); however, Corbett (1955) noted that considerable terrestrial (allochthonous) derived organic matter is carried by streams and rivers until the bulk of it often is deposited in deltas. In the marine depositional environment, the sediment characteristics are controlled by hydraulics and by grain size of the detrital material. Most coarse-grained sediment will deposit on beaches, bank tops, and shelves, or where water turbulence is maximum. Finer grained sediments are deposited in shelf areas, while the finest grained are deposited in the deep basin (Emery, 1960). Reducing conditions which aid in preserving organic matter are prevalent in sediments and bottom waters in many marine environments and in general tend t o become more reducing with depth below the water-ediment interface. Therefore, most of the oxidation of organic detritus occurs as it falls t o the sea bottom where the reducing environment will inhibit further oxidation. Rapid deposition will aid in the preservation of more organic matter as well as more of the easily oxidized organic compounds.

294

EFFECTS O F WATER ON PETROLEUM

Compaction Recently deposited sediments have water-filled pore spaces which comprise 8Wo or more of their volume, and deposition of a few more hundred meters of sediments causes this porosity to be reduced to 30% accompanied by the expulsion of water (Hedberg, 1964). Considerable time is necessary for water t o be expelled from low-permeability fine clays (Terzaghi and Peck, 1948). In some cases, pressures approaching geostatic are created because with increasing compaction, the pore water, which cannot easily escape, carries the full load. The pressure will drop to hydrostatic as the water is squeezed out, then the water bears the weight of the overlying water column and the sediments carry the weight of the overlying sediment column minus the buoyancy factor of the water. Compaction occurs in most sediments, but it is greater as muds form shales than when sands form sandstones, and likewise the expulsion of water from muds is greater. Also carbonate compaction will halt before argillaceous mud compaction because carbonate recrystallization causes formation of a rigid, difficult t o compact, structure. The clay, montmorillonite, expands when interlayer water is added and contracts when it is removed. The removal of water from montmorillonite is a function of pressure and temperature (Burst, 1969). Schmidt (1971) found that the temperature is about 93-104OC when clay starts releasing intracrystalline water and that rate of release increases with higher temperatures. Powers (1967) also suggested that it is a function of the amount of potassium which is necessary for the transformation of montmorillonite t o illite. Removal of water from montmorillonite with compaction resulting from deep burial yields large amounts of free water which can migrate. The amount of freed interlayer water depends upon the amount of montmorillonite or other clays in an expanded state in the sediment. The conversion of montmorillonite to chlorite does not require any volume change (Weaver and Beck, 1969), but the conversion of montmorillonite t o illite reduces the clay volume about 5096, so it is apparent that this freed water can be equivalent to 50% of the amount of montmorillonite in the sediment. The dehydration of montmorillonite occurs in three stages and probably a t three different depths of burial since the dehydration is a function of depth. Therefore, the second and third dehydration stages could release water at about 1,500 m and perhaps below 3,000 m thus allowing new supplies of water t o migrate. In the Gulf Coast area, the last water is not removed from montmorillonite until the depth is about 4,500 m. The clays in this area are composed of 50--80% montmorillonite, and the last water out comprises about 20% of the volume of montmorillonite present so it is evident that considerable freed water also is available at depths to 4,500 m. Von Engelhardt and Gaida (1963) noted that compaction of clays appears to proceed more rapidly if the pore solution is enriched in dissolved electrolytes. This may influence the compaction rate as, for example, a nonmarine

GENERATION AND MIGRATION

295

clay may compact more slowly than a marine clay. This could relate t o a nonmarine source evolving a different type of hydrocarbon. Generation and migration Most petroleum and natural gas hydrocarbons are generated from sediment organic matter by low-temperature chemical reactions (Philippi, 1965). Philippi also concluded that lipids probably are the major petroleum precursor and that most petroleum is generated by chemical reactions occurring at temperatures above 115C. A catalytic effect of the surrounding sediment, and in particular the shales, influences the rate and type of hydrocarbon formation. According t o McIver (1963)the maturation of hydrocarbon is influenced by depth, and the heavy hydrocarbons in the liquid crude oil decrease with depth while the lighter hydrocarbons increase with depth. Fig. 9.1 illustrates how an aromatic type oil might generate and mature to graphite with depth, and how a paraffinic type crude oil might generate and maturate t o methane with depth. In reality many crude oils contain both aromatics and paraffins and the ultimate end product with increasing burial depth would be graphite. It has been postulated that oil migrates from the fine-grained source rock to a trap as a discrete oil phase. However, calculations indicate the migration of oil globules in reservoirs required forces several thousand times greater than those produced by hydrodynamic gradients in most reservoirs (Levorsen, 1954). Because water is present in all sediments that contain organic matter which transforms to petroleum and natural gas hydrocarbons, it is reasonable t o assume that water is closely related to petroleum-generation processes. In essence, water controls the depositional environment of the petroleum

S t a r t i n g material

Starting material

1
P a r a f f i n i c crude oil

1
L i g h t e r hydrocarbons plus N, S, and 0 compounds

1
CH 4

Fig.9.1. Schematic of how an aromatic crude oil can maturate to graphite with increasing temperatures associated with increasing depth, and how a paraffinic crude oil can maturate t o dry methane gas. (After McIver, 1963.)

296

EFFECTS O F WATER ON PETROLEUM

precursors, while in the subsurface environment, water appears to control the migration of petroleum hydrocarbons and it is not unlikely that water is a control in the generation of petroleum. The generation of hydrocarbon is temperature dependent (Philippi, 1965). The viscosity of crude oil decreases exponentially with increasing temperature. The solubilities of petroleum hydrocarbons in water increase with temperature and decrease as the salinity of the water increases. A temperature drop from 150 to 25OC reduces the solubility of petroleum in water by a factor of 4.5 t o 20.5 (Price, 1973). Nonhydrocarbon organic acids, esters, and ketones are relatively soluble in water. Therefore it is possible that these compounds may migrate in water solution and be reduced t o hydrocarbons in the reservoir. Baker (1959) demonstrated that artifical solubilizers enhance the ability of saline water to solubilize and transport hydrocarbons. Baker (1962) ultrafiltered dilute petroleum acid soap solutions and found two kinds of colloidal particles present at all finite concentrations. One type he called a small ionic micelle and the other a large neutral micelle. According to him, the presence of these colloidal particles in water increases the water solubility of hydrocarbons by providing hydrocarbonlike regions in which the hydrocarbon solutes preferentially and selectively dissolve. The relatively small, spherical, isotropic micelle appears to enhance hydrocarbon solubility chiefly by incorporating hydrocarbons on its surface. However, hydrocarbons may be accommodated within the interior of the larger, cylindrical, anisotropic micelle, as well as on its surface. He found that paraffinic, naphthenic, and aromatic hydrocarbons possess different solubilities in the different kinds of micelles. It has not been demonstrated that these micelles occur in nature, however. Peake and Hodgson (1966 and 1967) studied the accommodation of C12-C3,n-alkanes in water. They found that it is possible to accommodate various hydrocarbons in water in the range of 0.01-100 ppm. Levels this high are more than adequate t o account for known petroleum reserves. For example, Hunt (1961) estimated a ratio of 27/1 for the volume of dispersed hydrocarbons in sediments t o oil in reservoirs, or that about 3.6% of the dispersed hydrocarbons are trapped in reservoir. Further, it can be shown that only approximately 1.8 ppm of the source hydrocarbons need t o be reservoired to account for the known oil reserves, assuming that the source sediments contain an average of 50 ppm of hydrocarbon. Hodgson et al. (1964) estimated that it would be necessary t o mobilize only about 1 ppm of the hydrocarbons from the source sediment to the reservoirs to account for known reserves. The odd/even carbon ratio is different in recent sediments, from ancient sediments and from most crude oils (Bray and Evans, 1961). Baker (1959) and Peake and Hodgson (1966) concluded that the odd carbon preference does not persist in the mobilizing water because the accommodation of hydrocarbon in water is a function of its properties rather than supply.

GENERATION AND MIGRATION

297

Welte (1965) extracted some bituminous rock samples with oilfield brines which had been cleaned of traces of hydrocarbons. The extracted n-paraffins had an approximately equal carbon number distribution. Cooper and Bray (1963) postulated a mechanism whereby the carbon number distribution may change when fatty acids are dissolved in water. Each fatty acid can lose COz to form an intermediate product which reacts to produce an n-paraffin and a fatty acid. The products have one less carbon atom than the original fatty acids. The oilfield waters that they investigated contained fatty acids with 14-30 carbon atoms with a balanced distribution between even and odd carbon numbers. Kartsev et al. (1959) reported that some oilfield waters contain up t o 5,000 mg/l of sodium naphthenate. Gullikson et al. (1961) found 5,000 t o 8,000 mg/l of organic acid salts in oilfield waters from the Ventura Basin. Collins (1969) reported that some waters from the Anadarko Basin contain up t o 3,000 mg/l of organic acid salts. The presence of 50 mg/l of organic acid salts in waters could transport significant amounts of hydrocarbons out of source rocks. It is apparent that petroleum hydrocarbons are generated from organic matter in sediments, that hydrocarbons are formed primarily at temperatures above 115'C through abiogenic reactions, and that water serves to transport the hydrocarbons from the sediments. It is possible that the water also may transport. petroleum precursors from the sediment. These precursors might be released from the water phase when the environment is right and then be transformed into petroleum hydrocarbons. Fluids in subsurface strata will move from regions of high potential t o a low-energy region. The permeabilities of the strata influence how quickly equilibrium can be attained between high-energy and low-energy fluid potentials. The fluids will move toward the strata offering the least resistance and that have the greatest permeability. Generally the lower energy environment is straight up, but it may also be straight down or lateral. The source sediments may overlie a sand structure, in which case the migration is likely t o be down into the sand structure which may serve as a reservoir. Energy gradients which cause fluids to move are caused by potentials resulting from temperature, pressure, elevation, and osmotic forces which have not been completely described mathematically (Hitchon, 1969). Rapid tectonic uplift can alter the steady-state equilibrium in a basin. Primary transit of hydrocarbon or precursors is this migration from the source sediment and usually is only for a short distance or far enough t o reach a coarse-grained sediment. After they reach the coarse-grained sediment, they may or may not be trapped. Secondary migration occurs after the hydrocarbons or precursors move into the coarser grained sediment, and it may result in long or short-distance migration. The forces suggested above affect the migration. Once the petroleum and gas have separated from the aqueous phase they tend t o move t o higher zones because of gravity segregation.

298

EFFECTS O F WATER ON PETROLEUM

Accumulation Compaction of sediments containing organic matter and water results in the expulsion of water and solubilized organic matter. Conceivably, the organic matter and gases can be solubilized or combined with the water phase as organic acid salts or as other soluble, colloidal, or suspended forms as a result of increased pressure. Meinschein (1959) outlined a method whereby ptroleum hydrocarbons may dissolve in compaction-expelled water, migrate, and later be unloaded to form an accumulation. Cartmill and Dickey (1970) formed a crude oil in water emulsion and found that it freely passed through coarse sand. However, it tended t o coalesce with decreasing grain size. They attributed this to electrochemical and capillary phenomena. Such a process could operate in a subsurface environment and form a hydrocarbon accumulation. Related t o this phenomenon, Kidwell and Hunt (1958) studied the hydrocarbon content produced from sands. They decided, that in the lenticular sands hydrocarbons are being filtered out of the moving stream of water by capillary action. Jones (1968) gives some data illlustrating the relation of pressure differentials to salinity differences in waters attributed to osmotic flow of water through a clay-shale membrane. He found that the salinity differential may exceed 100,000 mg/l between two aquifers separated by a 10-to 50-m layer of clay-shale. Waters carrying dissolved or solubilized hydrocarbons could release them because of salting out when contacting highly saline brines because water will accommodate less hydrocarbon as its dissolved salt concentration increases (Baker, 1967). I t also has been speculated that the solubilized or dissolved hydrocarbons may be unloaded from the aqueous phase because of temperature or presssure changes, filtration, salinity changes, or the attraction of the organic constituents in the water for a hydrocarbon accumulation. According to Baker (1962), the nature of the oil, whether paraffinic, naphthenic, or aromatic, probably depends less on the nature of the hydrocarbons originally available in the sediments for collection than on the proportions of neutral and ionic micelles that were able t o release crude oil droplets for subsequent pooling in the reservoir rock. It appears that the frequency distribution of hydrocarbons in crude oil reflects variations in the kind and size of the micelles in which the sediment hydrocarbons selectively dissolve. The subsequent release of solubilized hydrocarbons from aqueous solution results in the accumulation of oil. The hydrocarbons, after undergoing this solution and release process, would exist in the proportion characteristic of hydrocarbon occurrence in crude oil. Welte (1965), in discussing the micelle theory, noted that Neumann distilled several crude oils and observed that the surface tension of the distillate decreased and the dielectric constants increased with increasing boiling point. This means that the polar surfaceactive molecules which are

ALTERATION

299

necessary for the formation of micelles are enriched in the high molecularweight fraction. The surface-active components were phenols, and it was found that the interfacial tension between oils and a buffered solution was strongly pH dependent, suggesting that pH variations in pore solutions are important for stability of micelles. Hodgson and Hitchon (1966) theorized that surfactants aid in the mobilization of hydrocarbons in water, allowing them to migrate. During migration, the surfactants may decompose, leaving the hydrocarbons in an unstable state in the water phase. If the water is moving, the hydrocarbons will have a tendency not to agglomerate, but in a hydrostatic quasi-stagnant area the hydrocarbons will tend to separate by gravity t o form an accumulation. Hydrocarbons accumulate in areas called traps, and these traps sometimes are structural traps such as an anticline or stratigraphic traps such as permeability barrier. Accumulations of hydrocarbons in oil pools apparently form by the release of hydrocarbons from a water system. The hydrocarbons accumulate subsequent t o their migration, and the volume of water associated with their migration is large; however, tremendous volumes of water are available in the subsurface, and large amounts of water are freed by diagenesis or alteration of clays. Baker (1967) postulated that the tar mats that form near the edges of sedimentary basins can be attributed t o organic solutes moving in a water phase. The water merely evaporates at the sediment-air interface with complete release of the heavy oil solute containing compounds of carbon, hydrogen, nitrogen, oxygen, and sulfur. Because the solubilities of petroleum hydrocarbons increase with temperature and decrease with increasing salinity, a mechanism for petroleum accumulation can be formulated. Upward moving subsurface waters will decrease in temperature, and as they move upward, the pressure will be lower, and they may meet more saline waters. Because of these changing conditions they will release dissolved hydrocarbons. Alteration Hydrocarbons are subject t o alteration as soon as they are formed. The long-chain alkanes may react with other chemicals associated with the water phase, resulting in bond breaking and the formation of a shorter chain alkane. Organic acid salts may form which are relatively soluble in water and they may be further altered if the redox potential or pH of the water changes. The pH may change because of release of carbon dioxide associated with a hydrocarbon reaction. The hydrogen proton in the low pH water will react with dissolved organic acid salts, causing a reversion to the less soluble organic acid, which will disaccommodate from the aqueous phase. A positive change in the redox potential of the water could cause oxidation of dissolved or associated hydrocarbons. A negative change in the redox

300

EFFECTS O F WATER ON PETROLEUM

potential will aid in preserving the hydrocarbon, but if the redox should become extremely reducing, it could cause hydrogenation of the hydrocarbons, this is, however, very unlikely. Reservoired oil becomes lighter with depth in all productive sedimentary basins; this usually is attributed to increased cracking of the larger hydrocarbon molecules into smaller ones, and is associated with increased temperatures (Philippi, 1965). This process requires more hydrogen for the smaller molecules, which may result in condensation and polymerization of large molecules. Ultimately, with deeper burial and higher temperature, the hydrocarbons would be converted t o methane and graphite (Fig.9.1).

Water washing
Water washing is a simple dissolving of light hydrocarbons by water moving past a reservoir. The water washing causes the hydrocarbon accumu-

O i l m i g r a t e d l o here

Fig.9.2. Alteration of an oil accumulation caused by tectonism and moving water.

ALTERATION

301

lation t o become more concentrated in heavy hydrocarbons with a lower API degree gravity because of fractionation. Water washing occurs when fresh water infiltrates through an outcrop and flows through the zone containing a structurally trapped hydrocarbon accumulation. Examples of this type are found in the Rocky Mountain area. Tectonic disturbances can cause the hydraulic potential associated with a hydrocarbon accumulation t o become dynamic. The resulting hydrodynamic situation can cause severe alteration of the hydrocarbon accumulation. Fig. 9.2illustrates how oil may migrate to another reservoir if the strata are tilted, causing oil t o spill from a structural trap. This phenomenon can be observed in the Rocky Mountain area, where the first row of anticlines at the edge of a basin often are barren of hydrocarbon accumulation.

Bacterial
Biological degradation of a hydrocarbon accumulation is more complex than water washing. I t results from the introduction of bacteria into the reservoir or the reactivation of dormant species. Alteration of petroleum by bacteria cannot be directly termed alteration because of water, but water indirectly aids the alteration because water must be present so that the bacteria can live. Aerobic bacteria can exist and thrive at the surface or in the subsurface only if several conditions are met, and and Davis (1967)include these requirements as listed by Beerstecher (1954) the following: (1) Water: water must be present t o serve as a medium in which the bacteria can live. (2) Bacteria: the bacteria themselves must be present. (3) Nutrients: certain inorganic trace nutrients must be present. (4)Temperature: bacteria can exist at temperatures up t o about 90C but thrive at temperatures of less than 60C. (5)Food: a supply of food such as organic matter or petroleum must be present. (6) Toxics: bacterial poisons such as hydrogen sulfide must not be present in excessive concentrations. (7)Oxygen: aerobic bacteria require free oxygen dissolved in the water medium in which they live. Obviously all of these conditions are most easily met at the surface and the heavy oils and tars common t o oil seeps are products of fractionation and aerobic degradation of crude petroleum. Bacterial alteration of petroleum can also occur below the surface of the earth if the prerequisite conditions are met. These conditions, especially the presence of dissolved oxygen, occur only where subsurface waters are in hydrodynamic connection with the surface and are actively receiving surface-recharge waters. The presence of free oxygen is critical although it generally exists in very small concentrations, Kuznetsov (1957)measured the dissolved oxygen content in.

302

EFFECTS OF WATER ON PETROLEUM

waters from wells in the Groznyv oil district and found up t o 10.6 mg/l at a depth of 249 m. In general, dissolved oxygen was found only in the zone of active water circulation (surface recharge) and was lacking in the zone of stagnate water. Dissolved oxygen in smaller concentrations (less than 1 ppm) sustains aerobic bacteria and controls the rate at which they consume food (organic matter and petroleum). Since the oxygen content of subsurface waters usually is low, alteration and consumption of petroleum in reservoirs may require thousands or even millions of years to achieve a significant degree of change. 957) made bacterial counts in produced subsurface waters Kuznetsov (1 and found that the number of bacteria varied directly with the rate of exchange of water (recharge). The amount of dissolved oxygen varied directly with the rate of recharge, and the size of the bacterial population is dependent upon the amount of oxygen available. Aerobic bacteria in outcropping rocks were carried into the subsurface by recharge waters and were sustained by dissolved atmospheric oxygen in such waters. Apparently the rate at which bacteria are able t o move depends only on the amount of available oxygen and food. All digestible organic matter in the carrier bed along the recharge path would have t o be removed before the bacteria could proceed. Biodegradable organic matter in the carrier bed would be depleted as the oxygen front advanced and would permit free oxygen to reach great distances into the subsurface. The ultimate advance of bacterial alteration depends principally on time and temperature. If favorable conditions are present for sufficient time, bacterial alteration can reach great distances if the temperature remains below about 6OoC. Although some forms of bacteria can exist at temperatures up to about 90C, they apparently are in a dormant state above about 6OoC (Davis, 1967). Zobell (1949, 1950, 1952) found that all kinds of hydrocarbons appear to be susceptible t o bacterial action, even though the process is selective. In general, aliphatic hydrocarbons and aliphatic side chains on cyclic compounds are most susceptible t o attack while naphthenes and aromatics are least affected. Bacteria prefer the normal alkanes over the branched alkanes and concentrate on the heavier molecules, rarely touching hydrocarbons as light as C 3 . Bacteria consume alkanes at a hydrocarbon fermentation plant at Grangemouth, England (Anonymous, 1971). The nearly pure mixture of paraffins is prepared from crude oil by molecular sieving. The microorganisms assimilated the material with little waste and produced microbial protein which is used in animal feed. After consumption of the normal alkanes, branched chain alkanes are attacked (Winters and Williams, 1971), as are long chains attached t o aromatic and naphthenic rings (Davis, 1967). Bacteria usually reject ring compounds and many aromatics are toxic t o bacteria. However, with ideal conditions, bacteria can break naphthenic rings and consume them. Because

ALTERATION

303

bacterially altered oils contain less paraffinic hydrocarbons than unaltered oils of the same type, they have higher correlation index values (Smith, 1940) and a lower API gravity. Winters and Williams (1971) found an increase in nitrogen and optical activity disproportionate to n-paraffin loss in altered Powder River Basin ofis in Wyoming and suggested that microbially produced materials are added t o altered oils. Bailey and Krouse (1970) observed an increase in weight percent sulfur and asphaltenes and a decrease in saturates (alkanes) in altered Williston Basin crudes. The relative increase in asphaltenes because of the selective loss of lighter saturates leads t o increased viscosity of highly altered oils. The results of bacteria alteration are evidenced by heavy residual oils common t o oil seeps and breached oil accumulations such as the Athabasca tar sands. A bacterially altered paraffinic crude differs from a similarly altered asphaltic crude. Altered paraffinic crudes contain higher percentages of asphaltene and naphthene compounds than their unaltered counterparts, and usually are liquid and mobile. In contrast, severe alteration of asphaltic crude produces a solid, immobile tar. In the subsurface, the formation of tar seals halts bacterial alteration and may even act as a trapping mechanism for an accumulation. An example of o l residue providing a seal occurs in the i Bolivar Coastal fields of the Maracaibo Basin in Venezuela (Brenneman, 1960). The Maracaibo Basin oils have an asphaltic base, and bacterial alterahion has produced heavy residues which provides a seal (Rubio, 1959). Apparently, secondary migration of the oils occurred in a carrier-bed system exposed at the surface. Oil migrated updip without being trapped until it reached meteoric recharge waters about 5-10 km from the outcrop. Here, aerobic bacteria reduced the n-paraffin content of the oil, causing an asphaltic gel t o precipitate which reduced the porosity and permeability, preventing further migration and allowing oil to accumulate. An example are the vast reserves of Lagunillas field found in a massive sand downdip from an asphaltic seal in an area devoid of structural of stratigraphic txappng mechanisms. Three types of oil exist in the Bell Creek field, which is a large stratigraphic trap on the northeast flank of the Powder River Basin. The field contains common reservoirs, but the oil at the south was not altered and contained a full spectrum of n-paraffins, while in the middle of the field all n-paraffins were absent and the oil at the north was deficient in n-paraffins up to C1,. Selective loss of n-paraffins up t o C1, caused by microbiological activity occurred (Winters and Williams, 1969). The Bell Creek altered oils also had lower API gravity, higher nitrogen content, and higher optical rotation than their unaltered counterparts. The Bell Creek field is associated with fresh meteoric water in its central and northern portions but has no oil-water contact at its southern end, the downdip limit of production in that area being defined by absence of reser-

304

EFFECTS OF WATER ON PETROLEUM

voir sand. Evidence of aerobic organisms was found in waters produced with altered Bell Creek oils; oils in the Hackberry field, Louisiana, were bacterially altered by loss of n-paraffins and singly branched isoparaffins, and in North Africa an oil was depleted in n-paraffins.' Conclusions A suitable mechanism for removing water and organics from source sediments is compaction and squeezing of the fluids into adjacent beds either vertically, laterally, or below. Accommodation of hydrocarbons or hydrocarbon precursors in the aqueous phase appears t o be a probable means of primary migration. Variations in salinity, pressure, temperature, pH, redox potential, capillarity, hydrodynamics, or lithology may cause dissolution of mobilized hydrocarbons and serve t o begin an accumulation. Finally, the moving waters can serve t o alter an accumulation. Alteration of crude oils in the reservoir by water washing and biodegradation is common t o many producing areas and affects a significant portion of the world's reserves. Biodegradation is most prevalent in shallow-basin flank environments where fresh meteoric water reaches into the subsurface and carries with it aerobic bacteria and the oxygen necessary t o sustain them. Bacterial alteration selectively removes n-alkanes first, followed by isoalkanes and aliphatic side chains on cyclic compounds, leaving an o l relatively i enriched in naphthenes, aromatics, asphaltenes, and sulfur compounds. The API gravity is lowered, and the nitrogen content and optical activity are increased. Asphaltic base oils may be reduced t o an immobile tarry residue that can block porosity and form a seal which may trap oil. Alteration of paraffinic based crudes produces the so-called naphthenic oils, which, because of their relatively small asphaltene content, remain mobile even under advanced alteration conditions. References
Anonymous, 1971. Food from oil. Nature, 229:79. Bailey, N.J.L. and Krouse, H.R., 1970. Chemical aspects of crude oil preservation (abstract). Bull. Am. Assoc. Pet. Geol., 54:834-835. Baker, E.G., 1959. Origin and migration of oil. Science, 129 (3353):871-874. Baker, E.G., 1962. Distribution of hydrocarbons in petroleum. Bull. A m . Assoc. Pet. Geol., 46 (1):76-84. Baker, E.G., 1967. A geochemical evaluation of petroleum migration and accumulation. In: B. Nagy and U. Colombo (Editors), Fundamental Aspects of Petroleum Geochemistry. American Elsevier, New York, N.Y., pp.299-329. Beerstecher, Jr., E., 1954. Petroleum Microbiology - A n Introduction to Microbiological Engineering. American Elsevier, New York, N.Y., 375 pp. Bray, E.E. and Evans, E.D., 1961. Distribution of n-paraffins as a clue to recognition of source beds. Geochim. Cosmochim. Acta, 22:2-15. Brenneman, M.C., 1960. Chemical study of crudes of the Maracaibo Basin: Venezuela. Direc. Geol., Bol. Geol., Publ. Espec., No.3, Part 3, pp.1025-1069 (in Spanish).

REFERENCES

305

Burst, J.F., 1969. Diagenesis of Gulf Coast clayey sediments and its possible relation to petroleum migration. Bull. A m . Assoc. Pet. Geol., 53 (1):73-93. Cartmill, J.C. and Dickey, P.A., 1970. Flow of a disperse emulsion of crude oil in water through porous media. Bull. A m . Assoc. Pet. Geol., 54 (12): 2438-2447. Collins, A.G., 1969.Chemistry of some Anadarko Basin brines containing high concentrations of iodide. Chem. Geol., 4:169-187. Cooper, J.E. and Bray, E.E., 1963. A postulated role of fatty acids in petroleum formations (preprints). A m . Chem. SOC.,Div. Pet. Chem., 8 (2):A17-A28. Corbett, C.S., 1955. In situ origin of McMurray oil of northeastern Alberta and its relevance to general problem of origin of oil. Bull. A m . Assoc. Pet. Geol., 39

(8):1601-1649.
Davis, J.B., 1967. Petroleum Microbiology. American Elsevier, New York, N.Y., 604 pp. Emery, K.O., 1960. The Sea Off Southern California: A Modern Habitat of Petroleum. John Wiley and Sons, New York, N.Y., 366 pp. Gullikson, D.M., Caraway, W.H. and Gates, G.L., 1961. Chemical analysis and electrical resistivity of selected California oilfield waters. U.S.Bur. Min. Rep. Invest., No.5736, 21 PP. Hedberg, H.D., 1964. Geologic aspects of origin of petroleum. Bull. A m . Assoc. Pet. G e o i , 48 (11):1755-1803. Hitchon, B., 1969. Fluid flow in the Western Canada Sedimentary Basin, 2. Effect of geology. Water Resour. Res., 5:460-469. Hodgson, G.W. and Hitchon, B., 1966.Research trends in petroleum genesis. In: Petroleum --roc. Eighth Common. Min. Metall. Congr., Aust. N.Z. Paper 34, 59-19. Hodgson, G.W., Hitchon, B. and Taguchi, K., 1964.The water and hydrocarbon cycles in the formation of oil accumulations. In: I. Miyake and T. Koyamer (Editors), Recent Researches in the Fields of Hydrosphere, Atmosphere, and Nuclear Geochemistry. Maruzen, Tokyo, pp.217-242. Hunt, J.M., 1961. Distribution of hydrocarbons in sedimentary rocks. Geochim. Cosmochim. Acta, 22:37-49. Jones, P.H., 1968. Geochemical hydrodynamics - a possible key to the hydrology of certain aquifer systems in the northern part of the Gulf of Mexico Basin. Proc. 23rd Int. Geol. Congr., Prague, 1968, 17:113-125. Kartsev, A.A., Tabasaranskii, Z.A., Subbotta, M.I. and Mogilevskii, G.A., 1959. Geochemical Methods of Prospecting and Exploration for Petroleum and Natural Gas. University of California Press, Berkeley, Calif., 349 pp. Kidwell, A.L. and Hunt, J.M., 1958. Migration of oil in recent sediments in Pedernales, Venezuela. In: L.C. Weeks (Editor), Habitat o f Oil. American Association of Petroleum Geologists, Tulsa, Okla., pp.790-817. Kuznetsov, S.I., 1957. A review of fundamental research on the microflora of petroleum deposits. Mikrobiologiya, 26 (6):651-658. Levorsen, A.I., 1954. Geology of Petroleum. W.H. Freeman, San Francisco, Calif., 703 PP. "McIver, R.D., 1963. Maturation of oil, an important natural process. Geol. SOC. A m . Annual Meet., 1963 Paper, 15 pp. Mehschein, W.G., 1959. Origin of petroleum. Bull. A m . Assoc. Pet. Geol., 43:925-943. Peake, E. and Hodgson, G.W., 1966.Alkanes in aqueous systems, I. Exploratory investigations on the accommodation of Czo-C33 n-alkanes in distilled water and occurrence in natural water systems. J. A m . Oil Chem. SOC.,43 (4):215-222. Peake, E. and Hodgson, G.W., 1967.Alkanes in aqueous systems, 11. The accommodation of C, *-C36 'n-alkanes in distilled water. J. A m . Oil Chem. SOC.,44 (12):696-702. Philippi, G.T., 1965. On the depth, time, and mechanisms of petroleum generation. Geochim. Cosmochim. Acta, 29:1021-1049.

306

EFFECTS OF WATER ON PETROLEUM

Powers, M.C., 1967. Fluid-release mechanisms in compacting marine mudrocks and their importance in oil exploration. Bull. Am. Assoc. Pet. Geol., 51:(7)1240-1254. Price, L.C., 1973. Solubility of petroleum in water as function of temperature and salinity and its significance in primary migration. Bull. Am. Assoc. Pet. Geol., 57:801. Rubio, F.E., 1959. The conditions of the accumulation of petroleum of the Costaueros Fields, in the Bolevar District, Lake Maracaibo. Tercer Congr. Geol. Venezolana, Mem., 3:1009-1023 (in Spanish). Schmidt, G.W., 1971. Interstitial Water Composition and Geochemistry o f Deep Gulf Coast Shales and Sands. M.S. Thesis, University of Tulsa, Tulsa, Okla., 121 pp. Smith, H.M., 1940. Correlation index to aid in interpreting crude-oil analyses. U.S. Bur. Min. Tech. Rep., No.610, 34pp. Terzaghi, K. and Peck, R.B., 1948. Soil Mechanics in Engineering. John Wiley and Sons, New York, N.Y., 566 p9. Von Engelhardt, W. and Gaida, K.H., 1963. Concentration changes of pore solution during the compaction of clay sediments. J. Sediment. Petrol., 33 (4):919-930. Weaver, C.E. and Beck, K.C., 1969. Changes in the clay-water system with depth, temperature, and time. O f f Water Resour. Res. Project N0.A-008-GA WRC-0769, Georgia Inst. Technol., Completion Rep., 95 pp. Welte, D.H., 1965. Relation between petroleum and source rock. Bull. Am. Assoc. Pet. Geol., 49 (12):2246--2268. Winters, J.C. and Williams, J.A., 1969. Microbiological alteration of crude oil in the reservoir (preprints). Am. Chem. SOC.,14(4):E22-E31. Winters, J.C. and Williams, J.A., 1971. Microbiologic alteration of crude oil in muddy sandstone and other reservoirs. Bull. Am. Assoc. Pet. Geol., 55:369. Zobell, C.E., 1949.Action of microorganisms on hydrocarbon. Am. Pet. Inst. Rep. Progr. 1946-1947, pp. 107-132. Zobell, C.E., 1950. Assimilation of hydrocarbons by microorganisms. Adu. Enzymol.,

10:443--468.
Zobell, C.E., 1952. Part played by bacteria in petroleum formation. J. Sediment. Petrol.,

22:42-49.

Chapter 10. GEOCHEMICAL METHODS OF EXPLORATION FOR PETROLEUM AND NATURAL GAS

The current worldwide concern for energy resources results from the fact that man eases his burdens and accomplishes enormous amounts of work by harnessing mineral energy. His material progress is reflected by the amount of energy his machines consume, and the continued upward thrust of such progress is dependent upon the supply of mineral fuels. In the United States this energy has eased the burdens of man t o the extent that every person in the U.S.A. has available the energy equivalent of the output of 260 men. About 33% of the total mineral energy demand is supplied by natural gas and about 42% by crude oil. In 1967 the U S . demand for natural gas was 17.7 trillion cubic feet and the demand for crude oil was 4.5 billion barrels. The demand for both is expected t o double by the year 2000. Improved methods of finding petroleum and gas are needed. Many investigators believe that the application of geochemical methods in conjunction with geological and geophysical methods can markedly improve the discovery ratio. Data presented in this chapter primarily are in the form of an annotated survey of the available literature and many of these data were taken from Petroleum Abstracts with permission of H.O. McLeod (1971). Introduction

Type of methods
Geochemical methods of exploring for petroleum and gas can be termed direct and indirect. The direct methods are as follows: (1) Analysis of soil samples t o determine the amounts of hydrocarbon gases adsorbed by the soil. (2) Analysis of free hydrocarbons in soils. (3) Analysis of soils to determine the amounts of soil wax, paraffin dirt, and other bitumens. (4)Analyses of waters t o determine their amounts of dissolved hydrocarbons. These methods are classified as direct because the determined constituents are present in accumulations of petroleum and gas or derived from them. The indirect methods are as follows: (1)Measure the oxidation-reduction potential of soils, rocks, and waters. (2) Analyze soil samples for various salts such as bromides, chlorides, carbonates, and sulfates.

308

EXPLORATION FOR PETROLEUM AND GAS

(3) Analyze plants t o determine if their growth is affected by petroleumtype bitumens. (4) Analyze soil samples for bacteria such as the types that consume hydrocarbons. The indirect methods are so classified because the determined properties may have originated from something other than petroleum or gas accumulation (Rosaire, 1939; Kartsev et al., 1959). Collected data from both the direct and indirect methods usually are plotted or contoured to form maps to identify anomalies and to locate a target area for drilling. Case histories A geochemical soil survey led t o the discovery of an oilfield in the Texas Gulf Coast (Stormont, 1939). Three hundred geochemical surveys were made in 1939 with a greater than 50%success ratio (Simons, 1940). A gas discovery at Stroud, Oklahoma, was attributed t o geochemistry (Rosaire et al., 1940). Pirson (1942) determined the success ratios for the following exploration methods: random drilling, 5.8%;geology, 8.2%;geophysics, 14.9%; and geochemistry, 57.8%. Geochemistry predicted the discovery in Oklahoma of the West Edmond field (Bronston, 1947). The Hardy field in Texas was found by drilling into a geochemical anomaly where reflection seismographs gave no indication of a stratigraphic trap (Ransone, 1947). The Soviet Union attributes a 70% success ratio to wells drilled into geochemical anomalies (Sokolov et al., 1959). The success of geochemical methods is outstanding when compared to the overall success ratio for wildcats (Anonymous, 1960b). A company using an indirect inorganic geochemical technique achieved a 25.5%success ratio, which is more than double the usual wildcat ratio (Anonymous, 1959). Success ratios of 65% and 75%, respectively, were obtained using geochemical anomalies in the Texas Gulf Coast and North Texas areas (Anonymous, 1960a). The Kohav oilfield in Israel was discovered solely through hydrocarbon geochemistry, and geochemistry successfully predicted that stepout wells in this field and the Heletz field would be dry (Davidson, 1963). Analytical techniques, type of hydrocarbon anomalies, and successful discoveries made by hydrocarbon geochemical studies in Texas were discussed by Horvitz (1969). According t o him, the anomalies become weak and tend to disappear after an oil accumulation is produced. Note: The author has been told by representatives of various companies
that research performed by them has produced evidence indicating that

success ratios claimed for some of these exploration techniques cannot be met in actual practice. However, the author is at a loss t o cite references to this effect because there are no documented published data available.

INTRODUCTION

309

Source rocks
There are three basic elements relevant t o hydrocarbon accumulations: (1) source rock; (2) trap; and (3)reservoir. Before any hydrocarbons can accumulate in commercial quantitites, a source rock must be present. Concepts of hydrocarbon generation from source rocks have been reviewed by Hedberg (1964), Erdman (1965), Philippi (1965), Welte (1965), Landes (1967), Tissot et al. (1971), and Klemme (1972). Crude oil is liquid in its natural state and is composed primarily of hydrocarbons often combined with nitrogen, sulfur, and oxygen. Examples of some hydrocarbons found in petroleum are shown in Fig. 10.1. Natural gas is a gas in its natural state composed primarily of hydrocarbons often mixed with carbon dioxide, hydrogen sulfide, and nitrogen. Examples of hydrocarbons found in natural gas are shown in Fig. 10.2. Organic matter in rocks usually is divided into the part that is soluble in common organic solvents and the part that resists these solvents. The nonsoluble organic matter is called kerogen, and the amount of kerogen and soluble matter can be estimated from their respective carbon contents. The hydrocarbons in the soluble matter are of primary value in evaluating a source rock. The evaluation of total organic carbon in rocks can serve as an
ALKANES

c-c-c-c-c-c-c
Porottin

c-c-c-c-c
C

Bronchrd Porottin

Monocyclic

Bicyclic

Tricyclic

CYCLOPARAFF INS (nophthrnrsl

AROMATICS

Pyrrne

Fig.lO.1. Some of the types of hydrocarbons found i petroleum. n

310
NORMAL PARAFFINS H

EXPLORATION FOR PETROLEUM AND GAS

H-C-H H Methane

H H-C-C-H

I I

H H Ethone

I 1 I H-C-C-C-H I I I
H H

Propane

Fig.lO.2. Some of the hydrocarbons found in natural gas.

indication of a source rock for further estimation of the hydrocarbon potential of a target area. Organic matter in all sedimentary rocks is about 2% of the rock mass, and The Clarke for organic the amount of trapped oil is about 1.25 x carbon is 1.14% in shales and 0.24% in carbonates (Gehman, 1962). Forsman and Hunt (1958) found that the ratio of organic material t o organic carbon in rocks ranges from 1.07 to 1.22. Philippi (1969) found that the ratio between soluble carbon and total carbon CJCt must be greater than 3 for shale before it can be construed to be a source rock. Good source rocks contain more than 130 ppm of petroleumlike hydrocarbons; fair source rocks, 40-130 ppm; and poor source rocks, less than 40 ppm (Hunt and Meinert, 1954). Philippi (1957) extracted several shale samples and made a similar conclusion. His ranges were more than 500 ppm good, 50-500 ppm fair, and less than 50 ppm noncommercial. Analysis of the gases in cores and cuttings can be used to identify source rocks (FeugGre and Gkrard, 1970). Hydrocarbons with carbon numbers below CI4 usually occur in source rocks (Dunton and Hunt, 1962), but the recent sediments do not contain light hydrocarbons (Erdman, 1967), indicating that long periods of time are necessary in the hydrocarbon generation process. Commercial accumulations of oil usually are associated with marine shales or with organic-rich, fine-grained limestones. A shale commonly contains 0.5% or more organic matter and a limestone 0.2% before oil begins to occur. The kind and amount of organic matter in a rock indicate its source potential (Hunt and Meinert, 1954). Less than 5% of the organic matter in sediments eventually becomes petroleum; therefore, about 95% of it remains in the source rocks (Hunt and Jamieson, 1956; Philippi, 1957). Trask and Patnode (1942) studied about 35,000 rock samples and found that the average amount of organic matter in rocks in close proximity t o oilfields was about 1.5%. According t o Hunt (1967), more hydrocarbons appear t o be generated by fine-grained carbonate rocks than by other types of rocks containing the same amounts of organic matter. Philippi (1969) concluded that petroleum is generated by chemical reactions at relatively low temperatures, from source rocks. He further concluded that bacteria alone do not transform organic material into petroleum,

INTRODUCTION

311

and that radioactive bombardment of organic matter contributes little t o the formation of petroleum. His study of the shales and oil from the Los Angeles Basin indicated that the Upper Miocene divisions of the D and E shales form the major source of oil; this was the first documentation of oil source rock identified by comparing the composition of crude oil and shale hydrocarbons. He believes that lipids are the major precursors but that precise data concerning the chemical conversion of lipids t o petroleum components are lacking. Silverman (1964) summarized available data and concluded that petroleum is derived from lipids and that the r3C/12C 6-values indicate that the ranges for the lipid fractions of marine and land plants cover the ranges found in all petroleums. He suggested that polymers are a petroleum hydrocarbon nrecursor. Organic acids were suggested as hydrocarbon sources by Cooper and Bray (1963). Ferguson (1962) derived a quantitative method of determining hydrocarbons in sediments, using benzene (rather than mixed solvents), the benzene is evaporated, and the extract residues are resolved using silica gel chromatography. This method was used in an organic geochemistry study of the Cherokee Group rocks in Kansas and Oklahoma (D.R. Baker, 1962). A quantitative method using electron spin resonance spectrometry was used to evaluate the maturation stage of potential source rocks (Pusey, 1973). The method is effective for all types of rocks and all types of kerogens.

Water and hydrocarbons


Because of the need to locate gas and oil reserves, better hydrogeochemical exploration methods are needed. Water affects the accumulation and migration of hydrocarbons. Water flowing from compacting sedimentary rocks moves the oil precursors from source rocks before they concentrate in a trap area. Peake and Hodgson (1966, 1967) found that water can accommodate up to 150 ppm of C,,-C,, hydrocarbons as a fine colloidal suspension. Lowmolecular-weight hydrocarbons such as benzene, methane, ethane, and propane are somewhat soluble in brines and water; however, the normal paraffins and cycloparaffins with carbon numbers greater than C8 are rather insoluble (McAuliffe, 1969), and the insoluble ones are the main constituents of most crude oils. E.G. Baker (1962) postulated that micelles of oil (colloids) may be transported by water; however, the effects of temperature, p~ssure, and flow through sedimentary rock need further study. Spencer and Koons (1970) theorized that hydrocarbon precursors may be transported in water as compounds of nitrogen, oxygen, or sulfur, which later are reduced t o hydrocarbons. Cartmill and Dickey (1970) stabilized an emulsion of crude oil in water where the globules averaged 1 p in diameter, and the emulsion passed through sand but coalesced as the grain size decreased.

312

EXPLORATION FOR PETROLEUM AND GAS

According to them, the coalescing process may be an electrochemical rather than a capillary phenomenon. Baker (1960) suggested that salts of organic acids form clusters or micelles in water solution and that these colloidal particles increase the solubility of hydrocarbons by providing hydrocarbonlike regions within the water. An. equation was developed which postulates a relation between the abundance of a hydrocarbon in a crude oil and micellar water solubility.

Fig.lO.3. Illustrations of (a) structural and (b) stratigraphic traps.

The precise manner whereby oil leaves a source rock is not known,but it apparently is related t o the flow of water from compacting sediments. A better understanding of the migration process will aid in locating reserves of oil and especially those in stratigraphic traps or in a combination of stratigraphic and structural traps. Fig. 10.3 illustrates these types of traps.

HYDROGEOCHEMICAL RESEARCH AND METHODS

313

Hydrogeochemical research and methods

Organic compounds Saturated hydrocarbons Davis and Yarbrough (1969)invented a geochemical prospecting method which involves the analysis of formation waters for the saturated hydrocarbons: normal decane, isodecane, butyl cyclohexane, and pentyl cyclohexane. The presence of these hydrocarbons in the formation water is indicative of petroleum accumulations. studied the dissolved hydrocarbon gases in waters of Buckley et al. (1958) petroleum-bearing strata and determined the escaping pressure or the pressure at which the gas started t o come out of solution. Most of the gas in solution was methane with lesser quantities of ethane and heavier gases. The results indicated that the gases generally diffuse into the edgewater rather short distances from the margins of oil fields. outlined a general solution for the problem of deterKortsenshtein (1965) mining the oil and gas potential of subsurface strata from data of the gas saturation of subsurface waters under dephased equilibrium conditions. His study indicated that water that is saturated or supersaturated with hydrocarbon gases can positively be used t o predict whether oil and gas are present in traps, while waters that are not saturated with hydrocarbon gases cannot be used to make a positive prediction. It was observed: (1)that .the direction of increasing saturation indicates the direction of the accumulation; (2) that a relatively accurate target location can be obtained by determining where the increasing saturation pressure intersects by using data from two or more wells; (3) that an increase in pressure of saturation stratigraphically upward in a well may indicate an approach t o a gas-water contact; (4)that a constant saturation pressure laterally presents an unsolvable problem, unless one has data on changes with depth; and (5)that interpretation of the size of the postulated accumulation is difficult (Kortsenshtein, 1964). Continuous investigations made in the oilfields of the Dnepr-Donets Basin revealed a number of regularities in the composition and degree of gas saturation of groundwaters. In the presence of oil accumulations, the gas saturation of waters and the composition of the dissolved gases were determined with respect to the oil pool in a vertical and horizontal direction. The regularities of the gas saturation of these waters serve as a basis for oil exploration (Gutsalo and Krivosheya, 1965). London et al. (1961)found three classes of gas-water relationships. The first is represented by formation waters undersaturated with gas in which the hydrocarbons dominate the chemistry of the water. Commercial gas pools are not probable under such conditions. Examples of these exist in the Jurassic and Cretaceous sediments of the West Siberia Lowland. A system in which predominantly hydrocarbon gases and formation water are in equi-

314

EXPLORATION FOR PETROLEUM AND GAS

librium (that is, the gases are at saturation pressure) represents the second class. The third class is represented by a system in which the pressure of the dissolved gas is greater than the saturation value, free gas escapes, and the hydrocarbons are oxidized. Gas pools are likely to be found near the highly saturated water. Bond (1962)patented a geochemical exploration method which complements conventional gravimetric, magnetic, seismic, and electrical geologic survey methods. The technique utilizes the analysis of gas samples, collected from earth or water samples obtained from locations in the proximity of hydrocarbon-containing reservoirs. The gas is analyzed t o determine the /13C1H4. the samples .of oil-gas have not diffused a If isotopic ratio '*C1H4 considerable distance through the earth, the normal ratio ordinary methane/ heavy methane is in the range of 89.5-93.5. If the gas has diffused up from a considerable depth, the ratio will be 0.5-2.5 units above normal, which in turn indicates the presence of petroleum.

Aromatic hydrocarbons Zarrella et al. (1967)found that the amount of benzene in formation waters directly reflects the occurrence of a petroleum accumulation in a formation. They believe that vertical migration of hydrocarbons between aquifers is restricted and that lateral migration is limited. For example, Fig. 10.4 illustrates how the concentration of a hydrocarbon in a brine may vary with the distance of the brine from an oil pool. Zinger and Kravchik (1969)believe that toluene and benzene in water are hydrochemical indexes which indicate the presence of oil and gas. Accumulations of oil and gas are the principal source of benzene and toluene in oilfield . waters. The ratio benzene/toluene in these waters is greater than 1 Benzene in waters is considered one of the most important and direct indicators of oil and gas content (Kortsenshtein, 1968;Kartsev et al., 1969). If the concentration of a measured aromatic hydrocarbon in a sample of formation water is equal to a target value, the point from which the sample was obtained is close to a reservoir of crude oil. Greater differences between these two values represent greater distance to the crude oil accumulation. The target value is determined by contacting the type of crude oil expected in the sampled reservoir with water solutions of salt and measuring the concentration of the aromatic hydrocarbon in the solutions. This determines the variation in concentration of this aromatic hydrocarbon as a function of the salinity of the dissolving water. The salinity of the sample of the water makes it possible to determine the target value; that is, the value of concentration of this aromatic hydrocarbon existing at the point of contact between crude oil and formation water (Schmidt, 1970). daturnted, unsaturated, and aromatic hydrocar'bons McAuliffe (1 969) determined hydrocarbons dissolved in water by gas chromatographic methods using a number of techniques to separate the

HYDROGEOCHEMICAL RESEARCH AND METHODS

315

Fig.10.4. Relationship of the concentration of benzene in brines and proximity to an oil accumulation. (After Zarella, 1969.)

hydrocarbons from water. Separation methods include : (1)direct injection o the water sample; (2) equilibration of aqueous sample with immiscible f solvent; and (3) equilibration of aqueous sample with gas. Some of the solubilities found in water were: n-hexane, 10 ppm; n-heptane, 3 ppm; n-octane, 0 6 ppm; cyclopentane, 156 ppm; benzene, 1,780 ppm; and iso. propylbenzene, 50 ppm (McAuliffe, 1966). McAuliffe (1967)obtained a patent for a geochemical method of prospecting for petroleum, which utilized the concentrations o dissolved hydrocarbons in oilfield brines to f determine the location of a hydrocarbon accumulation.

Fatty acids Cooper and Kvenvolden (1967)analyzed samples of water from the subterranean formations t o determine the ratios of selected fatty acids. These ratios indicate the presence of a petroleum reservoir if the ratio of fatty acids containing an even number of carbon atoms to those containing an odd number of carbon atoms is not more than 1.6.

316

EXPLORATION FOR PETROLEUM AND GAS

Organic acids Shvets and Shilov (1968) used a quantitative method t o determine watersoluble organic substances in waters. Waters near oil accumulation deposits contained up to 3,500mg/l of organic acids. Data on organic carbon, organic acids, and oil hydrocarbons in waters could be used as indicators of oil accumulations. Organic acids dissolved in subsurface waters in the form of salts, or in the free state, indicate the removal of organic substances from rocks. This indicates the existence of geochemical processes in which deeply buried organic substances are being broken by decarboxylation, accompanied by the formation of hydrocarbons. Thus, organic acids and their salts may be considered a source for the formation of hydrocarbon accumulations (Shabarova et al., 1961). Inorganic and organic compounds
Kolodii (1969) found two types of hydrochemical anomalies in the Pliocene deposits of the South Caspian Basin. The first is associated with structures in which the red bed part of the producing formation is at a shallow depth. It manifests itself in sudden increase in dissolved solids in the water up to 300 g/l. The second anomaly is related t o the lower section and decreased dissolved solids 5-40 g/l. These waters are often of bicarbonatesodium type and were formed by mixing oilfield water with condensate water which reacted with the host rocks. This hydrochemical inversion is related to processes accompanying migration of hydrocarbons into the zones of reduced pressure and temperature. The genetic relationship between hydrochemical anomalies and the presence of oil (gas) deposits can be used in oil exploration. Compared to a general background, methane, heavy hydrocarbons, hydrogen sulfide, and carbonic acid concentrations in water increase with the approach t o a gas and oil accumulation, while the content of sulfates decreases. The amount of helium in the zone depends upon the difference between its saturation in the waters and in the accumulation (Savchenko et al., 1965). Sudo (1967) studied the major oil- and gasfields of Japan as related to subsurface water for the exploration of oil and gas deposits. The analytical data suggested that: (1) the SO,-2 concentration is extremely low, perhaps resulting from microbiological activity in the initial sedimentary environment; and (2) dissolved hydrocarbons and naphthenic acid salts in the brine are direct indicators, and I- and HC03 are indirect indicators of oil and gas accumulations. Oilfield waters (as well as waters of gasfields and bituminous formations) are characterized by enrichments of the biophile elements K+, B+3,Br-, and I-. Alkaline waters with higher concentrations of carbonates and sulfates, and with naphthenic acids, are associated with naphthenic oils. Alkaline-

HYDROGEOCHEMICAL RESEARCH A N D METHODS

317

earth chloride waters with low-numbered fatty acids are associated with paraffinic oils ( Krejci-Graf, 1962).

In0 rganic compounds


Serebriako and Tronko (1969) found that the ammonium content of subsurface waters correlates with hydrocarbon deposits. The amount of ammonium found was above 80 mg/l in oil- and gas-bearing areas with lesser amounts in nonproductive areas. Korobov (1965) determined that the main source of lithium, strontium, barium, manganese, copper, chromium, and aluminum in oilfield waters is related to the geochemical and biochemical environments which produced the oil and gas deposit accumulations. The elements potassium, boron, barium, iodide, strontium, barium, and gallium are enriched in oilfield waters (Krejci-Graf, 1962) and are an indication of bituminiferous formations. The first transition series metals - iron, cobalt, manganese, nickel, vanadium, titanium, chromium, and scandium - are soluble in the reduced state but insoluble in the oxidized state and are useful in determining the proximity of petroleum deposits according t o a patent issued to Billings (1969). The method involves analyzing formation waters for one or more of the transition metals to determine a trend in the amounts of the metal in an area. According to London (1964), the main indication of the presence of oil-gas in rocks is the presence of hydrogen sulfide and biogenic nitrogen, and the degree of sulfate reduction. The author recommends using the degree of sulfate reduction as a reliable exploratory criterion. The sulfate concentration decreases in the direction toward an oil accumulation. Sulfates cannot be regarded as reliable criteria if the water-bearing formation contains salt and gypsum. The single most important common ground between hydrodynamics and log interpretations is the determination of formation resistivity (Rw ), which is related to the chemical makeup of the formation water. In waters with a high chloride-ion content, the chloride ion is (for all purposes) the resistivity-determining ion; but in cases where low-chloride salinity waters are present, other ions, particularly calcium and magnesium, can make material changes from the chloride-calculated resistivity. Chemical content and its variations can be used directly as a tool for finding oil. Variations in the chemical makeup of formation waters can be used to delineate separate reservoirs within the same formation, thereby indicating potential stratigraphic traps. The production engineer also benefits from detailed chemical analysis (Schwab, 1965).

Radioactive compounds Filonov (1969) investigated radioactive elements in underground waters of the Devonian deposits in the Pripvatsky Depression. He found a uniformity in distribution of uranium and radium and a correlation between the increase

318

EXPLORATION FOR PETROLEUM AND GAS

in dissolved solids of the waters and the increase in their radium content. He concluded that the radium concentration may be used as an indirect indicator of oil. Gutsalo (1967) studied the geochemical relationship between radium anomalies and oil and gas deposits. In waters of t h e Dnepr-Donets Basin associated with hydrocarbons, the amount of Ra concentration depends on the concentration of dissolved solids and the age of the water-bearing rocks. Positive hydrogeochemical anomalies exist within the areas of hydrocarbon influence. The Ra anomalies are related t o the gas component of waters; Ra concentration is dependent on the partial pressure of hydrocarbons dissolved in water, which usually amounts t o more than 90%of the total gas pressure. The Ra concentration in these waters has a direct linear relationship to the hydrocarbon pressure. The Ra anomalies in the ground waters of the Dnepr-Donets Basin, near the gas-water or oil-water contacts, appear to owe their origin t o the hydrocarbon gases, which are genetically related t o the hydrocarbon deposits. The formation of radioactive anomalies on the earth's surface, above oil and gas deposits, is related t o the migration of hydrocarbons, salts, and ions in water solutions. It is assumed that the radioactive substance is carried toward the surface by ascending waters and hydrocarbons. The migration of the water and hydrocarbons is controlled by the permeability of the rock strata. The infiltration of gaseous hydrocarbons into the surface zone may increase evaporation, which leads to the increased migration of the water from surrounding areas. This water may be capable of dissolving radioelements, redepositing them in these places of greatest evaporation, which leads to the increased migration of water from surrounding areas. Liquid hydrocarbons, because of fractionation, may deposit bituminous material which is able to extract radioactive material from ground waters. There are two schools of thought on the matter of formation of surface radioactive anomalies. They are: (1) the radioelements have a surface origin; and (2) the radioelements have a deep-seated origin (Sikka, 1963). The deep-seated origin is substantiated by MacElvain (1963). In fact, he concluded that the determination of *l0Pband 206 Pb in near-surface samples should be more rewarding in oil exploration than the beta radiation techniques. Furthermore, the 210Pb/206Pb ratio might indicate the age of the underlying hydrocarbon accumulation. The amount of gases in surface soils is a function of the weather and is influenced by temperature, rain, and wind. Therefore, geochemical exploration techniques using soil samples must be designed so that samples are taken from a depth that is not influenced by the weather (MacElvain, 1963). Filonov (1964) studied the radium and uranium contents in different oilfield waters occurring in the platformal and geosynclinal deposits, in particular, the variations of the Ra/U ratio. The results revealed: (1) that oilfield waters contained an increased Ra content and that chloride-calcium types had a higher Ra concentration than the bicarbonate-sodium waters; (2)

HYDROGEOCHEMICAL RESEARCH A N D METHODS

319

that the water enrichment in Ra occurred .because of leaching from the reservoir rocks and because of the radioactive decomposition of uranium; (3) that uranium was either absent or present in insignificant amounts in the bicarbonate-sodium waters, where it occurred in the form of a complex anion [(UOzC03)3]*; (4) that the Ra/U ratio in the waters a t the oilwater contacts was shifted in the direction of increasing R a content; and (5) that the organic matter in oil had a definite effect on the shift of the Ra/U ratio because it removed uranium from the salt components dissolved in water, and supplied additional quantities of Ra because of the radioactive decomposition of uranium in oil. A linear relationship between .the radium concentration and dissolved solids concentration of some water from Paleozoic and Mesozoic age strata was found (Gutsalo, 1964). Waters in contact with hydrocarbon accumulations were more highly concentrated with radium. Vilonov (1962) investigated water-oil contact zones in several of the Soviet producing regions and found that a definite regularity in the distribution of radioactive elements exist. It was found that, in formation waters of oil accumulations, there is no definite relationship between the radium and the uranium content, and that usually uranium and thorium are present only in small quantities. The influence of oil occurrence on the distribution of radioactive elements in the formation waters of the oil accumulations appears most clearly in the zone of the water-oil contact; radium and its isotopes are concentrated in the water of this zone, and there is a relative decrease of uranium content, sometimes to zero. Gutsalo (1969) concluded the following:. (1) enrichment of chloridecalcium brines with helium results from migration through rocks and is directly proportional t o the amount of radium in solution - radium leached from the rocks by the underground water determines the amount of helium lost from the rocks; (2) a positive helium anomaly is formed in underground water around oil and gas accumulations; (3) the helium anomaly coincides with a positive radium anomaly; (4) the absolute value of the helium concentration at any point in the anomaly zone is dependent upon the radium concentration at the point and the time of formation of the accumulation; and ( 5 ) for any deposit of oil or gas, the areal extent of the positive helium anomaly in the formation water is dependent upon the value of the ratio of the partial pressure of the hydrocarbons to the total gas pressure. Physical properties Vdovykin (1963) determined the Eh and pH of formation waters and waters from rivers, lakes, and seas. Fig. 10.5 illustrates the results that he obtained. Unfiltered formation waters from petroleum producing wells had a lower Eh and a higher pH than waters taken from rivers, lakes, and seas. Filtration of the formation waters prior to analysis fof Eh and pH caused higher Eh and pH readings.

320
600

EXPLORATION FOR PETROLEUM AND GAS

SO0

I I
I

400

I I
I

I
> 300

200

0
I I

---- 1------I I

QI
9

100

(3
I
I
7

I
1 0

I
I1

PH

Fig.lO.5. Approximate pH and Eh of waters from unfiltered petroleum producing wells (A); filtered petroleum producing wells ( B ) ;and surface rivers, lakes, and seas (C). (After Vdovykin, 1963.

Fluid mechanics
Roach (1965) describes how to apply fluid mechanics t o petroleum exploration. The definition of fluid mechanics as he used it encompasses complete study of subsurface fluids including physical and chemical characteristics, whether hydrodynamic or hydrostatic, and a complete study of the characteristics of the reservoir rock.

Maps
Chloride-ion concentrations in water produced from rocks of various ages and depths were mapped in Lea County, New Mexico, using machine mapplotting techniques and trend analyses. Anomalously low chloride concentrations (1,000-3,000 mg/l) were found along the western margin of the Central Basin Platform in the San Andres and Capitan Limestone formations of Permian age. These low chloride-ion concentrations may be caused by preferential circulation of ground water through the more porous and permeable rocks (Hiss et al., 1969). Hanshaw and Hill (1969) studied aquifer systems from: (1) Mississippian age rocks; (2) Pinkerton Trail Limestone; (3) Paradox member of the Her-

HYDROGEOCHEMICAL RESEARCH AND METHODS

321

mosa formLtion; (4)Honaker Trail formation; and (5) Permian age rocks. Recharge in the Paradox Basin occurs on the west flank of the San Juan Mountains and along the west side of the Uncompahgre Uplift. A series of potentiometric surface maps were prepared for the five systems studied. With a few exceptions, most wells in formations above the Pennsylvanian age strata contain fresh to moderately saline water. Much of the strata below the Permian age rocks contained waters with dissolved solids concentrations greater than 35,000 mg/l and some areas favorable for hydrocarbon accumulations. Some of the brines in the Paradox formation contained up to 400,000 mg/l of dissolved solids. Cambrian age strata in much of Colorado is favorable for the accumulation of hydrocarbons. Chemical analyses of water from five Cretaceous aquifers were used to compute ion ratios, which were used in conjunction with structural and stratigraphic information t o interpret hydrologic conditions in the East Texas Basin. Ion ratio comparisons made by maps and diagrams show that the aquifers contain water of distinctive character, and that there are interconnections between aquifers, especially near the Mexia-Talco Fault zone and the Sabine Uplift. A hypothesis is offered that water moves along an unconformity from the Sabine Uplift eastward toward the East Texas oilfield where it enters the Woodbine Sandstone. Ion-ratio maps show the effect of time and of rock composition upon the relative kind and amount of dissolved solids in the water because- of reactions with minerals and organic material in the rocks. The hydrodynamic component of the water in the Woodbine formation from east to west helped form and contain the giant East Texas oilfield (Parker, 1969). Karim et al. (1966) studied three exploratory wells drilled on the east plunge of the Cordillera Isabella, Nicaragua, and all three had gas shows. A stratigraphic cross section and a localized map showing the relationship of magnetic highs obtained from an aeromagnetic survey and results of fluoroanalysis and water-gas surveys are included. The prospect of finding petroleum in coastal northeastern Nicaragua appears fair. Water analysis integrated with the existing knowledge of the geologic framework of an area provides supplementary information t o assist the exploration geologist in solving geologic problems on both a local and a regional scale. Isoconcentration maps showing regional variations in the total solids content of the waters within a given stratigraphic unit are important. Inorganic water analyses data are useful in the correlation of porous zones, and benzene analysis is a promising hydrocarbon exploration tool (Noad, 1966). Maps were prepared delineating: (1) surface outcrop areas of Upper Cretaceous rocks; (2) axial lines of Upper Cretaceous anticlinal structures; (3) zones of water types; (4)zones of water groups; and (5) zones of mixed waters. A diagram of chemical composition of the waters was also prepared. Using these maps, Galin and Plyushchenko (1963) selected areas in Dagestan that are favorable for the accumulation of oil and gas.

322

EXPLORATION FOR PETROLEUM AND GAS

Reviews Gerard and Feug&re (1969) concluded that geochemical exploration techniques are useful in offshore areas. Kroepelin (1967)reviewed geochemical prospecting as applied to petroleum and found that several companies had success ratios up to 59% with it in exploration. Johnson (1970) states that success ratios of about 35% can be attributed to the use of an inorganic technique of prospecting for oil and gas. The method is useful in locating stratigraphic trap accumulations and is based on the postulate that heavy-metal salts concentrate in the soil profile as a result of vertical migration of waters above an accumulation of oil. According to Boyle and Garrett (1970),geochemical prospecting will play an ever increasing role in the discovery of hidden ore deposits and accumulations of hydrocarbons. The methods need no longer be sold since it has now been generally recognized that they are the only direct approach to the problems of mineral, oil, and natural gas exploration. The methods are direct and have generally proved most successful when applied in conjunction with geological and geophysical exploration techniques. Geochemical methods applied t o petroleum prospecting have not reached their potential, expecially in the United States. Karaskiewicz (1966)discussed the geomicrobiological and hydrochemical research done by the Polish Petroleum Institute in 1962-63 in the LubelNadbuzan region. Specific methods such as the determination of gas dissolved in water were applied in this area, and 215 water samples were studied from the 9,500-km2 area. Anomalies produced by the bacterial activities which oxidize methane, propane, and butane were determined. Where methane was present, the bacterial microflora activity was lowest; where the biological activity was high, methane was absent. Important aspects of geochemical prospecting are : (1) analytical methods; (2) transport mechanisms of the hydrocarbons; (3) anomalies associated with hydrocarbon accumulations; (4) statistical treatment of the data; and (5) the final result. Significant findings are mapped and interpreted to locate a target area for drilling (Kroepelin, 1967). Case study of the Delaware sand (Bell Canyon formation), Texas, by Visher

(1961)
Preliminary work was carried out on the relation of sapropelic material found in both dark shales and laminated silts, and the oil occurring in reservoir rocks. The study indicated that the frequency distribution of the radicals, - benzene, straight chain, CH3, and CH2 -, was identical in both the black shales and the residual crude present in oil saturated reservoir rocks. The entire sequence of Permian rocks from the Bone Springs Limestone through the Lamar Limestone member of the Bell Canyon formation is composed of dark organic sediments, deposited in a reducing environment.

CASE STUDY OF THE DELAWARE SAND

3 23

The total quantity of the sapropelic material available for the formation of oil was not determined, but 1% would be a conservative estimate, indicating that source material in the Delaware Basin is of sufficient quantity t o produce volumes of oil greater than presently discovered. The oilfields which have been discovered in the basin primarily are confined t o the upper porous and permeable sands of the Bell Canyon formation. The present distribution of oil appears t o be controlled by hydrodynamic conditions. Therefore, the tracing of the times and paths of migration is dependent upon reconstructing the paleohydrodynamics.

Potentiometric surface of the upper Delaware sand


A potentiometric surface map of the Bell Canyon formation was made from a two-dimensional electric analog model of the central and eastern portions of the Delaware Basin in West Texas (Fig. 10.6). The majority of pressures were bottom-hole measurements from existing fields. Only a few shut-in pressures from drill-stem tests were usable because of the short shutin time commonly used in this area, and consequently, few tests reached true formation pressures. A pressure buildup method should have been used. The total dissolved solids of the formation water range from a low of 90,000 mg/l in the Ford field to over 250,000 mg/l at the South Pyote field. All pressure readings were corrected for effects of varying total dissolved solids. The assumption was made that the concentrations of total dissolved solids are stratified with little mixing. Therefore, the weighted mean is between 50,000 and 120,000 mg/l. The potentiometric surface has a hydrodynamic gradient from west to east with a component of northward flow. In the southeastern portion of the mapped area, the hydrodynamic gradient is reversed because of the influence of the eastern flank of the basin. Stratigraphic traps are formed in areas where linear sand fingers show an updip decrease in permeability and porosity. The change in permeability and porosity between the permeability barrier and adjacent reservoir rocks, however, is not great. The Saber field, for example, has an average porosity of 25% and permeability of 70 md, and the barrier rock an average porosity of 12% and permeability of 3 md. Since in some areas this barrier rock would be considered a possible reservoir, something in addition to these changes in porosity and permeability is necessary to prevent the movement of oil into the barriers. Under equilibrium conditions water flowing through a formation will have a greater pressure gradient across a tight zone than a more permeable zone (see Fig.lO.7). Therefore, the differential pressure in the barrier is greater than in the reservoir, and varies directly with the decrease in permeability between the two. When the oil phase reaches the barrier zone, the pressure gradient increases updip, making i t increasingly more difficult for the oil to

324

EXPLORATION FOR PETROLEUM AND GAS

-zI

CASE STUDY OF THE DELAWARE SAND

325

246 k g A q cm

Fig.10.7. Relationships of hydrodynamic gradients t permeability; I = decreased hydroo dynamic gradient because of increased permeability; 2 = increased hydrodynamic gradient because of decreased permeability; 3= average hydrodynamic gradient; 4 = decreased hydrodynamic gradient.

invade the water-filled rock pores. Finally, the entry pressure of the barrier rock is greater than the invading force of the oil and migration ceases. The downdip hydrodynamic flow increases the efficiency of the trap by reducing the buoyancy effect of the oil. The hydrodynamic enforcement of stratigraphic traps can increase the oil column many times over what it would be under hydrostatic conditions and probably accounts for the development of commercial stratigraphic oil accumulations in the Delaware Basin.

Formation waters
The initial approach in the study of stratigraphic problems within the Bell Canyon formation was by the use of formation waters. Over 300 samples of formation water were collected, analyzed, and processed by computer techniques. Data collected in this manner were posted on maps (Fig.10.8-10) and contoured. Several aspects of the waters (relative concentration percentages of SO4,Mg, Ca, and total solids) show systematic variations over the basin. Variation in these parameters is related t o proximity to outcrop and the degree of transmissibility of the formation. The highest sulfate content is near the outcrop belt t o the west; the calcium and total dissolved solids concentrations increase toward more impermeable rocks and areas of low circulation. In the center of the basin, the waters are characterized by very high total dissolved solids, high calcium, and low sulfate content, but in the porous and permeable fingers near the outcrop, salinites are low and sulfate is high. All gradations between these two extremes exist in the Bell Canyon formation. In areas where the sand fingers pinch out very rapidly into lowpermeability sediments, the transition between these two extremes of water composition may take place in a matter of a few well locations. An excellent

3 26

EXPLORATION FOR PETROLEUM AND GAS

example of this is the Saber field (Fig.10.8-10) where a range of waters is evidenced in one homogeneous, continuous sand body. The reason for these rapid changes in formation water compositions may be explained by permeability changes within the Bell Canyon formation. In areas of low permeability there is less circulation, less dilution, and more chance for the maintenance of an equilibrium relation between formation water and sediment. This was substantiated by the distribution of magnesium in the waters. A series of multiple regression analyses was made on the relation of various dissolved ions in the waters to their total dissolved solids. First, all the waters were analyzed as a unit to determine the correlation coefficients and the degree of variability explained by the chosen ions. The second stage was the breakdown of the waters into three arbitrarily defined groups (based principally on salinities) t o see if there were any noticeable changes in either correlation coefficients or degree of explained variance. The only significant change was in the relation of magnesium t o total solids. In those waters containing relatively low concentrations of total dissolved solids, there was a significant positive correlation, but in those with high concentrations of total dissolved solids, there was a significant negative correlation. This indicates that the relative concentration of magnesium decreases in waters of high total dissolved solids. These waters are precisely those that are found in low-permeability, fine-grained, argillaceous rocks in which magnesium would most likely be taken out of waters by diagenetic alteration of clay minerals. The variations found in the formation waters within the Bell Canyon formation can be used as the basis of an exploration technique. Since the composition of formation waters is related t o permeability, and permeability is related t o producibility of reservoir rocks, a workable relation exists between exploration objectives and water compositions. The refining of the maps of the distribution of the composition of the waters aids in defining the distribution of the sand fingers. The updip edges of the permeable sqnd fingers show increased concentrations of total dissolved solids and decreased magnesium which are related to the presence of a barrier (or trap) updip from the reservoir sands.

Formation water maps


Maps of the total dissolved solids content (Fig. 10.8), the chloride content (Fig. 10.9), and the calcium content (Fig. 10.10) of formation waters were prepared. The inference which may be drawn from these maps is the empirical association of oilfield occurrence versus the iso-mg/l contours of the various constituents. This empiricism shows some remarkable alignments and permits formation-water composition maps to be added to the suite of exploration tools. Some of the subtleties of constituent composition versus rock properties do not lend themselves readily t o mapping techniques but are useful for consideration (i.e., magnesium content versus low permeability

CASE STUDY OF THE DELAWARE SAND

327
I

328

EXPLORATION FOR PETROLEUM AND GAS

d
Y

CASE STUDY OF THE DELAWARE SAND

329

33 0

EXPLORATION FOR PETROLEUM AND GAS

and fine-grained rocks, sulphate concentration variation with relationship to outcrop, etc.). It is important to note on each of the constituent maps that the local variations of the iso-mg/l contours are of greatest importance and not the precise value of the contour. For example, on the total dissolved solids map (Fig. 10.8) the concentration in the Ford field is only 50,000 mg/l and ranges in a re-entrant to about 150,000 mg/l, while in the Wheat field, the range is from 150,000 mg/l to nearly 250,000 mg/l. The overall appearance of this map (Fig. 10.8) is a series of fingering expressions. The various oilfields seem to have an occurrence relationship in the transition zone from higher to lower concentration. The Mason, Tunstill, Olds, and Saber fields occur along a transition zone from 250,000 mg/l to 50,000 mg/l. The El Mar and Grice fields are on a transition zone from 250,000 mg/l to 150,000 mg/l. The Two Freds field has a transition zone from 250,000 mg/l to about 150,000 mg/l. The Wheat and also the Ford fields occur in a similar transition zone. This relationship of oil occurrence and differential concentration is of great significance. Even in this limited area it appears that in the block from longitude 1033000 to 1034500 and from latitude 304500 to 320000 there are several places which, from this empirical relationship, have some potentialities. The block immediately south of this, trending southwesterly from the Wheat and also from the Two Freds fields, needs additional study for better delineation. The area east of the Two Freds field lacks adequate control but basically shows the possibility of favorable development. The chloride map (Fig.10.9) has a configuration similar to the total solids map. Again it is not the precise iso-mg/l contour which is of prime concern but the variation in the limited area. This rate of change from higher to lower concentration appears to be a principal key to occurrence. The calcium content map (Fig. 10.10) does not show the prominent fingering, almost pseudodeltaic, effect that the total solids and chloride maps have. Perhaps this is because of the smaller range of values mapped. Some of the high to lower concentration effect is present and in other areas, the Wheat and Two Freds fields, the iso-mg/l closure is developed. This map is less diagnostic than the others; however, considered in conjunction with the other two maps, the coexistence of accumulation and the transition zone, even closure cannot be missed. Formation water maps of other areas Fig. 10.11 is a potentiometric surface map of the Arbuckle formation group in parts of Kansas, Missouri, and Oklahoma drawn by Chenoweth (1964). As ,noted on the figure, the arrows indicate the theoretical direction of the water flow which was inferred from the tilted oil-water interfaces. Detailed pressure data along with reservoir transmissibility data could be used to construct a similar map for determining detailed flow pattern.

FORMATION WATER MAPS

331

Kilometers 55.60

I:l,wom

LEGEND

drrpru indicate ItmoretlcoI dimction of artasion flow. Infarred tmm tilted oil
i"l,,t.aCe. in B"ter,Ellswrth, ord Pownee Counties. Kansas.
w01.r

Arbuckle b n t . " R e W o n amd'or Gmnite Wash beneath hnnaylvonion. Ellia.Rush and Barton Countlea. Kanroa .

Fig.lO.11. Potentiometric surface map of the Arbuckle formation group in South Kansas, North Oklahoma, and southeast Missouri.

The map shown in Fig.10.12 also was constructed by Chenoweth (1964) and it is a chloride map of the Arbuckle formation group in Kansas and Oklahoma. Note the dilute brines near the outcrop areas, which are diluted by meteoric recharge waters entering the outcrop. As a general rule the trapped petroleum in this group of formations is found associated with the more saline brines, and in the transition areas. Fig. 10.13 is a map that illustrates the variation in salinity at the bottom of the lower Wilcox formation in portions of Texas, Louisiana, Arkansas, Mississippi, and Alabama. The most saline brines occur in the deeper basin areas with the dilute brines nearer outcrop areas. Fig. 10.14 is a similar map which illustrates the salinity variations in the top of the lower Wilcox formation in the same area. Fig. 10.15 illustrates the salinity variations at the base

332

EXPLORATION FOR PETROLEUM AND GAS

Fig.10.12. Map of the chloride concentrations (mg/l) in the Arbuckle formation waters in Kansas and Oklahoma

of the upper Wilcox formation, while Fig. 10.16 illustrates the salinity variations at the top of the upper Wilcox formation. Fig. 10.17 is a salinity map of the lower Tuscaloosa and Woodbine formations, again most of the trapped petroleum is found in areas where the more saline waters occur and in transition areas. Salinity maps are useful as a primary tool in petroleum exploration because they provide information concerning sand fingering, diagenetic changes that affect reservoir and source rocks, and stratigraphic traps. Occurrences of petroleum accumulations often correlate with salinity transition zones, i.e., where the salinity ranges from 50,000 mg/l to 100,000 mg/l.

EEL

S d V N IZIBLVM NOILVNIZIOd

334

EXPLORATION FOR PETROLEUM AND GAS

Fig.10.15. Salinity concentrations in waters taken from the base of the Upper Wilcox formation in portions of Texas, Louisiana, Arkansas, and Mississippi.

Less than 5,600

Fig.lO.16. Salinity concentrations in waters taken from the top of the Upper Wilcox n formation i portions of Texas, Louisiana, Arkansas, Mississippi, and Florida.

CONCLUDING REMARKS

335

LEGEND

II

Greater than

im,ooo

mg/l (as NOCI)

5.600 lo 70.000 mp/l

Less than 5.600 mg/I

Fig.lO.17. Salinity concentrations in waters taken from the Woodbine (Dexter) and Lower Tuscaloosa formations in .portions of Texas, Oklahoma, Arkansas, Louisiana, Mississippi, Georgia, South Carolina, North Carolina, and Florida.

Concluding remarks Organic acid salts, petroleum hydrocarbons, and other organic compounds are soluble in water. The ionic composition, the pH, and the Eh of the water influence the solubilities of the organic compounds. The aqueous solubility of petroleum hydrocarbons increases with increasing temperature and pressure, and decreases with increasing water salinity. The aqueous solubility of organic acid salts increases with increasing pH. A mechanism for the migration of petroleum or petroleum precursors, therefore, is water. It is known that petroleum hydrocarbons are generated from organic-rich rocks. The organic material in the petroleum source rocks is transformed by physicochemical reactions into petroleum precursors and/or hydrocarbons which are solubilized by water. The water phase moves the solubilized organics from the source to the reservoir where, because of temperature, pressure, salinity, pH, filtration, or organic salting-out phenomena, the organic phase separates from the water. In the reservoir the petroleum precursors and/or hydkocarbons mature to crude oil and gas, primarily because of temperature and time. Thermal alter-

336

EXPLORATION FOR PETROLEUM AND GAS

ation proceeds both in the fine-grained source rock and in the reservoir at temperatures above 115OC by abiogenic reactions. With increasing temperature the quality of the crude oil improves; however, at higher temperatures the crude oil is destroyed, leaving methane and pyrobitumen. The primary mechanism in the migration of petroleum involves water, therefore, it follows that knowledge of certain characteristics of the water is useful in exploration for oil and gas. The chapters Classification of oilfield waters, and Some effects of water upon the generation migration, accumulation, and alteration of Petroleum discuss some of these characteristics. Fig. 10.18 illustrates some characteristics related to waters that are likely to indicate an oil or gas accumulation, and some characteristics related t o waters that are likely to indicate a dry reservoir.

/2 C

a type water

(a)

Fig.lO.18. Genetic indicators in a water associated with an oil and gas accumulation (a) compared to indicators in a water associated with a dry reservoir (b).

REFERENCES

337

References
Anonymous, 1959. Is geochemistry worth the effort. Pet. Week, 23-1-1959:20-24. Anonymous, 1960a. Geochemistry to get full-fledged test as exploration technique. Pet. Week, 2-12- 1960 :34-35. Anonymous, 1960b. 10.9% of U.S. wildcats produce in 1959. World Oil, 1 5 0 : l l l . Baker, D.R. 1962. Organic geochemistry of Cherokee group in southeastern Kansas and northeastern Oklahoma. Bull. A m . Assoc. Pet. Geol., 46:1621-1642. Baker, E.G., 1960. A hypothesis concerning the accumulation of sediment hydrocarbons t o form crude oil. Geochim. Cosmochim. Acta, 29:309-317. Baker, E.G., 1962. Distribution of hydrocarbons in petroleum. Bull. A m . Assoc. Pet. Geol., 46 :76-84. Billings, G.K., 1969. Geochemical petroleum exploration method, U.S. Patent, No.3,428,431. Bond, D.C., 1962. Geochemical process. U.S. Patent, No.3,033,287. Boyle, R.W. and Garrett, R.G., 1970. Geochemical prospecting - a review of its status and future. Earth-Sci. Rev., 6:51-75. Bronston, A., 1947. Geochemical data forecast major field. World Oil, 127:112-116. Buckley, S.E., Hocutt, C.R. and Taggart, M.S., 1958. Distribution of dissolved hydrocarbons in subsurface waters. In: L.G. Weeks (Editor), Habitat o f Oil. American Association of Petroleum Geologists, Tulsa, Okla., pp.850-882. Cartmill, J.C. and Dickey, P.A., 1970. Flow of a disperse emulsion of crude oil in water through porous media. Bull. A m . Assoc. Pet. Geol., 54~2438-2447. Chenoweth, P.A., 1964. Potentiometric Surface Map and Salinity Map o f the Arbuckle Formation Group in South Kansas, North Oklahoma and Southeast Missouri. Sinclair Oil Company, unpublished. Cooper, J.E., 1962. Fatty acids in recent-and ancient sediments and petroleum reservoir waters. Nature, 193:7 44-7 46. Cooper, J.E. and Bray, E.E., 1963. A postulated role of fatty acids in petroleum formation. Geochim. Cosmochim. Acta, 27:1113-1127. Cooper, J.E. and Kvenvolden, K.A., 1967. Method for prospecting for petroleum. US. Pa ten t , No. 3,30 5,31 7. Davidson, M.J., 1963. Geochemistry can help find oil if properly used. World Oil, 157:94-106. Davis, J.B. and Yarbrough, H.R., 1969. Geochemical exploration. U.S. Patent, No.3,457,044. Dunton, M.L. and Hunt, J.M., 1962. Distribution of low-molecular-weight hydrocarbons in recent and ancient sediments. Bull. A m . Assoc. Pet. Geol., 46:2246-2248. Erdman, J.G., 1965. Petroleum - its origin in the earth. In: A. Young and J.E. Galley (Editors, Fluids in Subsurface Environments. American Association of Petroleum Geologists, Tulsa, Okla., pp.20-52. Erdman, J.G., 1967. Geochemical origins of the low-molecular-weight-hydrocarbon constituents of petroleum and natural gases. In: Proceedings 7th World Petroleum Congress, Mexico, D.F. Elsevier, London, 2:13-24. Ferguson, W.S., 1962. Analytical problems in determining hydrocarbons in sediments. Bull. A m . Assoc. Pet. Geol., 46:1613-1620. Feugere, G. and Gerard, R.E., 1970. Geochemical logging - a new exploration tool. World Oil, 170:37-40. Filonov, V.A., 1964. Influence of organic material in oil on the shift of radioactive equilibrium in the watersof oil deposits. Sou. Geol., 6:144-146 (in Russian). Filonov, V.A., 1969. The problem of use of radioactivity of underground waters as an indirect hydrochemical indicator of oil content. Nauch. Tekh. Sb. Ser. Neftegazov. Geol. Geofiz., 3:32-35 (in Russian).

338

EXPLORATION FOR PETROLEUM AND GAS

Forsman, J.P. and Hunt, J.M., 1958. Insoluble organic matter (kerogen) in sedimentary rocks of marine origin. In: L.G. Weeks (Editor), Habitat of Oil. American Association of Petroleum Geologists, Tulsa, Okla., pp.747-778. Galin, V.L. and Plyushchenko, V.G., 1963. Hydrogeology of the Upper Cretaceous deposits of Dagestan as related to their oil- and gas-bearing properties. Zzu. Vyssh. Uchebn. Zaued. Geol. Razued., 4:120-217 (in Russian). Gehman, Jr., H.M., 1962. Organic matter in limestones. Geochim. Cosmochim. Acta, 26 :885-899. Gerard, R.E. and Feug&re, G., 1969. Results of an experimental offshore geochemical prospection study. In: P.A. Schenck and I. Havenaar (Editors), Advances in Organic Geochemistry. Pergamon Press, New York, N.Y., pp.355-372. Gutsalo, L.K., 1964. Some regularities of radium distribution in underground waters of middle part of Dnepr-Pon Depression. Geokhimiya, 12:1305-1312 (in Russian). Gutsalo, L.K., 1967. Geochemical relation of radium anomalies in groundwaters to oil and gas deposits. Dokl. Akad. Nauk S.S.S.R., 172:1174-1176 (in Russian). Gutsalo, L.K., 1969. The nature and regular features of distribution of helium anomalies in underground waters adjoining oil and gas deposits, Sou. Geol., 8:112-123 (in Russian ). Gutsalo, L.K. and Krivosheya, V.A., 1965. Certain regularities in the gas-saturation of groundwaters associated with oil-gas-bearing structures in the central DDB (DneprDonets Basin) and their significance for oil-gas prospecting. Geol. Nefti Gaza, 3:51-53 (in Russian). Hanshaw, B.B. and Hill, G.A., 1969. Geochemistry and hydrodynamics of the Paradox Basin region, Utah, Colorado, and New Mexico. Chem. Geol. 4:263-294. Hedberg, H.D., 1964. Geologic aspects of the origin of petroleum. Bull. A m . Assoc. Pet. Geol. 48:1755-1803. Hiss, W.L., Peterson, J.B. and Ramsey, T.R., 1969. Saline water in southeastern New Mexico. Chem. Geol., 4:341-360. Horvitz, L., 1969. Hydrocarbon gqochemical prospecting after thirty years. In: W.B. Heroy (Editor), Unconventional Methods in Exploration f o r Petroleum and Natural Gas. Southern Methodist University, Dallas, Texas, pp. 205-21 8. Hunt, J.M., 1967. The origin of petroleum in carbonate rocks. In: G.V. Chilingar, H.M. Bissel and R.W. Fairbridge (Editors), Developments in Sedimentology, 9B. Carbonate Rocks -Physical and Chemical Aspects. Elsevier, Amsterdam, pp.225-251. Hunt, J.M. and Jamieson, G.W., 1956. Oil and organic matter in source rocks of petroleum. Bull. A m . Assoc. Pet., Geol., 40:477-488. Hunt, J.M. and Meinert, R.N., 1954. Petroleum prospecting. U.S. Patent, No.2,854,396. Johnson, A.C., 1970. How to hunt oil and gas using the inorganic surface-geochemical method. Oil Gas J., 68:llO-112. Karaskiewicz, J., 1966. Forecasting of oil- and gasbearing capacity in the light of geomicrobiological and geochemical investigations. Nafta, 22 (12):19-20 (in Polish). Karim, M, Chilingar, G.V. and Hoylman, H.W., 1966.Northeast Nicaragua has gas and oil i, indications. World O l 162:84,86, 91,92,94,96. Kartsev, A.A., Dudova, M. Ya. and Diterikhs, O.D., 1969. Benzene homologs in underground waters and their relation to oil. Geol. Nefti Gaza, 7:41-45 (in Russian). Kartsev, A.A., Tabasaranskii, Z.A., Subbota, M.I. and Moglevskii, G.A., 1959. Geochemical Methods of Prospecting and Explomtion f o r Petroleum and Natural Gas (English transl. edited by P.A. Witherspoon and W.D. Romey) University of California Press, Berkely, Calif., 349 pp. Klemme, H.D., 1972. Geothermal gradients. Oil Gas J . , 69:136,141-144; 70:76-78. Kolodii, V.V., 1969. Origin of some hydrochemical anomalies in petroleum provinces. Nauch. Tekh. Sb. Ser. Neftegazou. Geol. Geofiz., 8:37-40 (in Russian).

REFERENCES

339

Korobov, D.S., 1965. Distribution of trace elements in water and rock of oil deposits in the Saratov-Volgograd region of the Volga and its significance in petroleum exploration. In: Soviet Advances in Nuclear Geophysics (English transl. of Yadernaya Geofizika). Consultants Bureau, New York, N.Y., pp.171-178. Kortsenshtein, V.N., 1964. The estimation of the possible oil-gas presence according to the groundwater analyses and the evaluation of the forecast oil-gas reserves. Dokl. Akad. Nauk S.S.S.R., 158:856-859 (in Russian). Kortsenshtein, V.N., 1965. Estimating the oil and gas potential from data on gas saturation of groundwater under dephased equilibrium conditions. Dokl. Akad. Nauk S.S.S.R., 150:635-638 (in Russian). Kortsenshtein, V.N., 1968. A comparative description of the benzene content in stratal waters of the Mesozoic complexes of South Mangyshlak and of East Caucasia. Dokl. Akad. Nauk S.S.S.R., 180:697-699 (in Russian). Kravchinskii, Z.Ya. 1960. Comparison of chemical characteristics of waters in productive red-colored formations. Geol. NeftiGpza, 12:42-44 (in Russian).. Krejci-Graf, K., 1962. Oilfield watek. Erdol Kohle Petrochem., 15:102-109 (in German). Kroepelin, H., 1967. Geochemical prospecting. In: Latest Developments within the Oil Industry. Proceedings 7 t h World Petroleum Congress. American Elsevier, New York, N.Y., 1B:37-57. Landes, K.K., 1967. Eometamorphism and oil and gas in time and space. Bull. A m . Assoc. Pet. Geol., 51 :828-841. London, E.E., 1964. The degree of groundwater saturation with dissolved hydrocarbons and sulfates as a criterion for the evaluation of oil-gas prospects. Geol. Nefti Gaza, 11:41-47 (in Russian). London, E.E., Zorkin, L.M. and Vasilev; V.G., 1961. Location of gas reserves based on dissolved gas content of formation waters. Geol. Nefti Gaza, 3:35-40 (in Russian). MacElvain, R.G., 1963. What d o near-surface signs really mean in oil finding. Oil Gas J., 61(7):133-136; 61(8):139-146. McAuliffe, C.D., 1966. Solubility in water of paraffin, cycloparaffin, olefin, acetylene, cycloolefin, and aromatic hydrocarbons. J. Phys. Chem., 70:1267-1275. McAuliffe, C.D., 1967. Geochemical method of prospecting for petroleum. U.S. Patent, No.3,345,137. McAuliffe, C., 1969. Determination of dissolved hydrocarbons in subsurface brines. Chem. GeoL, 4:225-233. McLeod, H.O., 19711 Bibliogmphy Related to Research and Geochemical Methods for Petroleum and Gas Exploration. Petroleum Abstracts, 200 pp., unpublished. Noad, D.F., 1966. Water analysis: a key to exploration. Can Petrol., 7:12-14; 7:16-18. Parker, J.W., 1969. Water history of Cretaceous aquifers, East Texas Basin. Chem. Geol., 4:111-133. Peake, E. and Hodgson, G.W., 1966, 1967. Alkanes in aqueous systems, I and 1 . J. Am. 1 Oil Chem. SOC.,I, 43:215-222; 11, 44:696-702. Philippi, G.T., 1957. Identification of oil source beds by chemical means. Proc. 20th Int. Geol. Congr., Mexico City, 1957, Sect. III, pp.25-40. Philippi, G.T., 1965. On depth, time, and mechanism of petroleum generation. Geochim. Cosmochim. Acta, 29:1021-1049. Philippi, G.T., 1969. Essentials of the petroleum formation process are organic source material and a subsurface temperature controlled chemical reaction mechanism. In: P.A. Schenck and I. Havenaar (Editors), Advances in Organic Geochemistry. Pergamon Press, New York, N.Y., pp.25-46. Pirson, S.J., 1942. Theoretical and economic significance of geodynamic prospecting. World Petrol., 13:38-42. Pusey, 111, W.D., 1973. How to evaluate potential gas and oil source rocks. World Oil, 176:7 1-75.

340

EXPLORATION FOR PETROLEUM AND GAS

Ransone, W.R., 1947. Geochemical history of the Hardy oilfield, Jones County, Texas. Geophysics, 12: 384-392. Roach, J.W., 1965. How to apply fluid mechanics to petroleum exploration. World Oil, 160:7 1-7 5. Rosaire, E.E., 1939. The Handbook of Geochemical Prospecting. Rosaire, Houston, Texas, 120 pp. Rosaire, E.E., McBermott, E. and Fash, R.H., 1940. Discussion of geochemical exploration. Bull. Am, Assoc. Pet. GeoL, 24:1434-1463. Savchenko, V.P., Vinogradov, V.L. and Yakoylev, Yu. I., 1965. Front- and back-wall effect and its importance in prospecting. Geol. Nefti Gaza, 7:36-40 (in Russian). Schmidt, G.W., 1970. Geochemical prospecting method. US.Patent, No.3,524,346. Schwab, R., 1965. Logging important aspect of hydrodynamic studies. Oilweek, 16:36-37. Serebriako, 0.1. and Tronko, I.V., 1969. Ammonium content in groundwaters of the northwestern Caspian region as an indication of oil and gas. GeoL Nefti Gaza, 9:57-60 (in Russian). Shabarova, N.T., Tunyak, A.P. and Nektarova, M.B., 1961. Study of organic acids in subsurface waters. Geol. Ne f t i Gaza, 11:50-5 1(in Russian). Shvets, V.M. and Shilov, I.K., 1968. On organic matter in underground waters of the southwestern part of the Azov-Kuban Artesian Basin. GeoL Nefti Gaza, 8:46-49 (in Russian). Sikka, D.B., 1963. Mechanisms explaining the formation of radiometric anomalies. Izv. Akad. Nauk S.S.S.R., Ser. GeoL, 6:73-87 (in Russian). Silverman, S.R., 1964. Investigations of petroleum origin and evaluation mechanisms by carbon-isotope studies. In: H. Craig, S.L. Miller and G.J. Wasserburg (Editors), Isotopic and Cosmic Chemistry. North-Holland, Amsterdam, pp.92-102. Simons, H.F., 1940. Scope of soil analysis increased during year. Oil Gas J., 38:54. Sokolov, V.A., Alexeyev, F.A., Bars, E.A., Geodekyan, A.A., Mogilevskii, G.A., Yurovskii, Yu.M. and Yasenev, B.P., 1959. Investigations into direct oil detection methods. Proc. 5th World Petrol Congr., Sect. I , Paper, No.36, pp.667-687. Spencer, D.W. and Koons, C.B., 1970. Studies on origin of crude oil data. Presented at SPE Symp., 44th Annual Meet., Calgary, A l t a , June 22-24, 1970. Stormont, D.H., 1939. Operation of a Gulf Coast field on soil survey information. OilGas J., 38:28-29. Sudo, Y., 1967. Geochemical study of brine from oil and gas fields in Japan. J. Japan Assoc. Pet. Technol., 32: 286-296 (in Japanese). Tissot, B., Califet-Debyser, Y., Deroo, G. and Oudin, J.L., 1971. Origin and evolution of hydrocarbons in Early Toarcian shales, Paris Basin, France. Bull. Am. Assoc. Pet. GeoL, 55:2177-2193. Trask, P.D. and Patnode, H.W., 1942. Source Beds o f Petroleum. American Association of Petroleum Geologists, Tulsa, Okla., 566 pp. Vdovykin, G.P., 1963. Oxidation-reduction potential of formation waters of the northwest Cis-Caucasus and of some surface waters. Pet. GeoL, 7:286--290. Vilonov, V.A., 1962. A feature of the distribution of radioactive elements in the wateroil contact zone. Geokhim. Nefti Neftyanykh Mestorzhdenii, Akad. Nauk S.S.S.R., Moscow, pp.199-206 (in Russian). Visher, G.S., 1961. Petrologic study of the Delaware sand (Bell Canyon formation), Texas. Rep. Sinclair Oil Company, unpublished. Welte, D.H., 1965. Relation between petroleum and source rock. Bull. Am. Assoc. Pet. Geol., 49:2246-2268. Zarrella, W.M., 1969. Applications of geochemistry to petroleum exploration. In: W.B. Heroy (Editor), Unconventional Methods in Exploration for Petroleum and Natuml Gas. Southern Methodist University, Dallas, Texas, pp.29-41.

REFERENCES

341

Zarrella, W.M., Mousseau, R.J., Coggeshall, N.D., Norris, M.S. and Schrayer, G J , 1967. .. Analysis and significance of hydrocarbons in subsurface brines. Geochim. Cosmochim. A h , 31 :1155-1166. Zinger, A S . and Kravchik, T.E., 1969. On direct hydrochemical indices in the presence of n oil and gas and the role of micellar solubility in the case of hydrocarbon migration i aqueous solutions. DokL Akad. Nauk S.S.S.R., 189:180-184 (in Russian).

This Page Intentionally Left Blank

Chapter 11. GEOPRESSURED RESERVOIRS

The composition of the waters in normally pressured reservoirs often differs from the composition of the waters in geopressured or abnormally high-pressured reservoirs. There are several theories concerning the cause of the geopressured zones; many papers have been written about their occurrence and causes (Burst, 1969; Dickey et al., 1968, 1972; Fowler, 1970; Harkins and Baugher, 1969; Hottmann and Johnson, 1965; Jones, 1969; Powers, 1967;Schmidt, 1973;Wallace, 1969). Knowledge of how to locate geopressured zones is important in drilling operations, because if such a zone is drilled into without adequate preparation, the well may blow out, perhaps causing a fire, loss of the well, loss of the drilling rig, or even loss of life. The usual precaution, if the driller knows of a high-pressure zone, is to increase the weight of the drilling mud; however, the continual use of heavyweight mud is much more expensive than drilling with a lighter weight mud. Drilling rig time is worth about $2,000 per day, and it costs about $44,000 per kilometer to drill a well on land. A drilling barge in the bay can cost from $4,000 to $lO,O.OO day while a drilling ship plus a full crew costs about per $25,000 per day. Considering.the foregoing costs plus the cost for a special crew t o extinguish a fire at an ignited blowing well can be very expensive because the initial fee for the fire extinguishing personnel is about $25,000. Steps, therefore, are taken by the drilling company to assure that an adequate drilling rig is used, that the optimum size borehole is drilled, that the correct weight drilling mud is pumped down, that strong enough casing is inserted into the well, and that blow-out preventers are operative. Geopressure Dickinson (1953)defined abnormally high pressure (geopressure) as any pressure exceeding the hydrostatic head of a column of water (extending from the subsurface tapped stratum to the land surface) containing 80,000 mg/l of dissolved solids. Formations with equal or less pressures are considered normal or subnormal. In the Gulf Coast area the normal pressure gradient is about 0.107 kg m-l, or about equal to 0.21 g cm-3 of drilling mud (Harkins and Baugher, 1969). Normal pressure in the Rocky m-', although excep Mountain region has a gradient of 0.100 kg tions occur in western Montana, the Denver Basin, the Powder River Basin, and the San Juan Basin, mostly in Cretaceous rocks (Finch, 1969). (A gradient of 0.118 kg m-l is normal in the Williston Basin in North

344

GEOPRESSURED RESERVOIRS

Dakota.) Normal pressure is that which is normal for the particular area involved and is related t o the salinity of the reservoir water, rock types, and geologic setting, but in general, it is that pressure exerted by a column of water from the surface t o the observed subsurface formation, which is equal to and will balance the subsurface formation pressure. Abnormally high pressures are those which exceed this normal hydrostatic head. Geostatic ratio Abnormal pressures can be expressed in terms of a geostatic ratio, which is the ratio of the observed fluid pressure in a subsurface formation t o the overburden pressure of the overlying sediments. This load at a given m down t o depths of more than depth is approximately 0.231 kg 6,100 m, because the density of rocks changes slowly with depth (Pennebaker, 1968). Any abnormal pressure will therefore have a geostatic ratio in the Gulf Coast area and between between 0.0327 and 0.0703 kg in the Rocky Mountain area. 0.0304 and 0.0703 kg Compaction model The compaction concept was demonstrated using a model consisting of perforated metal plate separated by metal springs in water and enclosed in a cylindrical tube (Terzaghi and Peck, 1948). The springs were used to simulate communication between deposited particles and with the initial pressure upon the upper plate, the springs d o not move because all of the pressure is supported by the water, assuming that water does not escape from the system. Relating the fluid pressure (FP) t o the total pressure (TP) one can derive an equation X=FP/TP to record various formation pressures (X)o determine the geostatic ratio using the model. t

Origin of abnormal pressures


Abnormally high pressures in a formation can be caused by compaction. Factors which may cause them are, according to Hottmann and Johnson (1965), the ratio of shale t o sand thickness, the mean formation permeability, the elapsed time since deposition, the rate of deposition, and the amount of overburden. These parameters are interrelated in compaction, which is the controlling factor in fluid pressures within subsurface sedimentary environments (Harkins and Baugher, 1969). Dickinson (1953)reports that the fluid pressures within sediments are predominately controlled by two the factors; namely: (1)the compression as a result of compaction; and (2) resistance t o expulsion of water. Compaction begins with sedimentation and deposition of soft muds composed of up to 90% water (Wallace, 1969). In an environment where deposition continues, gradual compaction occurs whereby the muds become clay

ORIGIN OF ABNORMAL PRESSURES

345

minerals and shales. The shales are primarily clay minerals with flat or tabular grain shapes; with additional overburden, the pressure packs the grains closer together, with a resultant expulsion of water from the intervening spaces. In the early stages of compaction, the shale possesses high porosity and permeability, and the expelled water always flows to areas of least resistance and pressure (often porous sand). As the overburden increases, the porosity and permeability of the shale decrease until equilibrium is approached and the pressure in all directions is equal. A t this point, expulsion of additional water is limited. Tectonics, of course, could alter the subsurface environment. Deposition and sedimentation of sand are somewhat different because the sand grains are in contact in the first stage and sand compaction is about complete with deposition. However, reduction of porosity can occur by: (1) solution of the sand grains at contact points; and (2) rearrangement of the grains because of very high pressures. Clay beds separating aquifers are often referred to as semipermeable membranes. Such beds can separate aquifers containing waters of different salinities, causing a hydrostatic head in the direction of the more saline water.

Fig. 11.1. Sand dikes in the Simpson Sand formed by the actiqn of highly pressured subsurface waters forcing the lighter colored sand intrusively into the primary sandstone. The primary sand was formed from white beach sands during Ordovician time.

346

GEOPRESSURED RESERVOIRS

Osmotic pressure can develop, which is dependent upon osmotic efficiency of the clay bed and the differences in salinities of the two aquifers (Young and Low, 1965).According to Jones (1969),stepwise increments of osmotic pressure may develop wiih depth through a series of bedded sands and clays acting as a multistage pump, thus producing the high reservoir pressures in the northern Gulf of Mexico basin. Fertl and Timko (1972)discuss 17 possible causes of abnormally high pressures. They are rate of sedimentation, tectonic activities, potentiometric surface levels, reservoir structures, areal salt deposition, shallow-reservoir repressuring, paleopressures, mud volcanoes, secondary precipitation of cementation constituents, diagenesis of volcanic ash, rehydration of anhydrite, diagenesis of clays, osmosis, permafrost, earthquakes, chemical, thermal chemical, and biochemical effects, and tidal disturbances. Fig. 11.1 illustrates one type of action that results from high pressures, where sand dikes formed by the action of highly pressured subsurface waters forcing the lighter colored sand intrusively into the primary sandstone. The primary sandstone was formed from white beach sands during Ordovician time. Abnormal pressures in the Gulf Coast area In the Gulf Coast area, the abnormal pressure seems t o be related t o rapid deposition of sediments and low regional transmissibility. Fluid pressures are near hydrostatic where there is continuity with normally pressured aquifers and where the sands are sufficiently permeable to dissipate the expelled water from the compacting fine-grained rock. In some of the deep oil and gas wells of the Gulf Coast, the pressure of the interstitial fluids (oil, gas, or water) in kilograms per square centimeter is normally the depth in meters multiplied by 0.107.This is slightly more than the pressure required t o sustain a column of water to the surface. At great depths where the geological section is mostly shale, fluids at abnormally high pressures are found. Sometimes the pressures are very high, approaching 0.2 kg cme2 m-l. Often the increase in fluid pressure is abrupt, taking place in a vertical interval of 30 m or less. In other areas, the increase in pressure is more gradual, extending over 300 m of vertical section. The depth at which the pressure starts t o increase ranges over a wide interval. Abnormal pressures are found at depths as shallow as 1,000m in some offshore fields, and wells in some areas have been drilled deeper than 7,000 m without encountering abnormal pressures. Forty-one formation water samples from gasfields in southwestern Louisiana were obtained and analyzed to determine the relationships of the chemical composition of the waters to normal and abnormally pressured geologic zones (Dickey et al., 1972).The concentration of dissolved solids in the waters from the overpressured zones is generally less than in the normal pressure zones, and this knowledge is significant in electric log interpretation.

ABNORMAL PRESSURES IN THE GULF COAST AREA

347

Fig. 11.2. Slash lines showing the general area in Louisiana where the samples were obtained.

Previous work in the area suggested that the abnormally fresh waters were found in the same part of the section as were the abnormally high pressures (Dickey et al., 1968). The general locations of the wells are shown in Fig. 11.2. They were from the South Lewisburg, Church Point, Branch, South BOSCO, North Duson, Duson, Ridge, and Andrew fields, all in Acadia, and Lafayette Parishes, Louisiana. The water samples were analyzed chemically by using the procedures published by the American Petroleum Institute (1968). The analytical data are summarized in Table 11.1. A subsurface cross section, Fig. 11.3, was constructed in a general northsouth direction showing the stratigraphy and structure across seven oilfields in the area of study (Fajardo, 1968). The initial pressures of the shallower reservoirs are normal. However, below 2,450 m many reservoirs contain fluids with abnormally high pressures. The 4.9-m amplified normal curve was used to recognize the first appearance of abnormal pressures in the shale section. The fluid pressure gradients were estimated following the method described by Hottmann and Johnson (1965). Shale resistivity and fluid pressure gradient versus depth were plotted for 50 wells in different fields of the study area, and of these, 22 are included in the cross section. All of the 41 waters belong t o the chloride-calcium class of Sulin (1946), and none has the composition of meteoric water. The principal cation is sodium, although the concentration of calcium is always high. In some of the more concentrated brines, the calcium concentration is nearly 40,000 mg/l and constitutes over half the reacting value of the sodium. Magnesium is variable in amount, and in two samples it is absent. Chloride is the predominant anion, amounting always t o more than 49.5% of the total reacting values. Sulfate usually is absent and never is present in concentrations greater than 0.5% of the total reacting value. In Fig. 11.3, the top of the section is 2,100 m below sea level. The electric logs indicate the lithology, which is quite sandy down t o a depth of 2,700 m

TABLE 11.I Formation-waters sample locations, constituents found in the waters, shale resistivity (SR), and fluid pressure gradients (FPG) Sample Location o f number of well (S-Twp-R) Depth (m)
Zone

Specific gravity

Concentration (mg/l) HCO3


SO,

SR at
B

FPG spl. depth (kg cm-2

(60/600F) CI

Br

Na

Ca

Mg

Li

Sr

Ba

NH4 organic acid as acetic 246 295 238 650 178


200

6)

1*I 21 3* 41 5*1

6
7 8 9 10
11

73-105-34E 2l-lOS-03E 2*lOSFo3E 20-10S-aE 17-10S43E O8-lOS43E 25-07S43E 25-07603E

4,66+4,664 3,643-3,645 3,625-3,627 3.613-3,615 4,060-4,063 3,047-3.049 3,325-3,327 3.293-3.296

U. Camerina

12 13 14 15 16 17** 18 19 20 21 22* 23 24* 25 26 27 28 29 30 31 32 33* 3483 35*3 36 37 38 39*3 40 41*3

1.035 1.062 1.057 L. Camerina 1.059 U. Camerina 1.085 Bolivina-mex Discorbis 1.045 1.060 U. Tweedel 1.069 L. Tweedel U. Nodosaria 1.083 Daigle 1.051 1.065 U. Nodosaria 1.069 Tweedel 1.062 Tweedel 1.128 Struma Frio 1.061 Nodosaria 1.149 Klumpp D 1.090 Frio 1.058 Frio 1.058 Frio 1.057 1.144 Frio 1.090 U. Texana U.Texana 1.092 1.220 Frio 1.202 Frio l.lS3 Frio 1.139 Nodosaria 1.070 Marg howei 1.144 Homeseeker 1.145 Nodosaria A Horn-eker D-4 1.088 1.055 Klumpp E 1.089 Brookshire Brookshire 1.089 Brookshire 1.082 Brookshire 1.089 Nodosaria 1.140 Homeseeker 2-D 1.120 1.069 Marg tex 1.082 U. Moicene 1.050 Marginulina Bolivina-mex

39,000 55,600 56,600 50,000 72.800 33,300 51,700 58,000 50,900 45,300 45,600 57,900 52,600 116,000 50,900 135,000 79,300 46,600 47,900 45,800 125,000 84.500 80,300 201,000 184,000 111,000 119,000 61,600 100,000 109,000 80,000 44,400 77,500 78,000 72,300 75,400 129,000 120,500 55,500 74,400 49,700

387 541 826 630 448 503 180 363 507 545 574 586 579 322 330 92 334 741 788 694 135 419 363 0 0 112 76 550 66 73 270 244 171 206 234 203 80 240 539 249 482

0 407 234 38 tr. 50 0 tr. 33 0 tr. tr. 0 0 67 223 0 60 72 ND 0 122 tr. 352 tr. ti. 0 0 0

49 62 62 37 43 32 18 26 35 28 29 26 34 38 23 52 48 46 48 44 47 40 45 75 67 42 52 67 42 0 39 77 34 130 36 0 18 0 18 0 33 0 18 0 41 0 43 102 52 0 26 8 8 43

35 61 52 57 81 37 21 41 70 38 35 56 45 128 43 154 169 40 52 47 64 20 62 213 204 94 117


14

18 22 21 21 19 16 15 18 23 18
22

17,800 32,300 34,200 29,500 41,100 19.200 34,800 34,400 27,500

25 20 26 18 24 74 23 21
22

26,000 24,900 33,300 31,700 49,600 29,600


66,800 46,400 24,700 26,600 ND 61,900 45,800 45,800 80,600 68,900 53,600 52,700 35,200 40,600 51,800 47,800 25,800 44,200 44,400 41,200 42,600 77,800 68,800 32,500 42,800 29,600

201 71 110 70 81 82 58 162 174 134 40 79 60

23 22 20 18 19 24 28 5 21 21 34 30 21 19 18 19 24 38 26 18 35

1,070 78 2,210 369 1,380 213 1,380 194 3,850 583 1,390 224 2,730 544 2,020 194 3,050 719 2,310 167 2,950 447 2,660 389 1,570 0 21,600 2.180 1,890 408 15,200 1,270 2,950 447 2,660 1,010 2,180 303 ND ND 15,700 159 7,390 565 3,300 972 38,800 2,140 33,200 5,770 14,300 428 18,400 1,200 3.610 17 18,300 1,090 14,400 700 2,760 35 1,510 447 836 3,530 3,270 972 3,270 564 3,370 894 4.560 0 5,610 564 3,210 136 2,950 855 1,780 141

518 247 200 204 267 85 208 230 162 134 262 in1 192 427 31 5 813 427 172 166 157 830 324 376 782 640 798 771 137 1,150 631 392 166 236 235 294 232 ND 375 176 264 71

10 7 6 6 6 3 4 5 6 4 4 4 5 9 9 9 10
6

ND ND ND ND ND P:D ND ND ND ND ND ND ND ND ND ND

5 4 15 7 9 17 17 12 18 5 17 15 10 5 2 2 3 2 ND 5 5
2

ND ND ND YD ND ND ND ND ND ND ND ND ND ND ND ND 0 50 0 5 0 5 0 8 0 19 ND ND 0 33 ND ND ND ND ND ND ND ND ND ND N D l ND ND ND 0 110 ND ND 140 97 128 109 265 41 171 102 ND ND ND ND 0 7 195 85 0 4

538 301 362 279 210 254 364 218 230 250 202 377 206 96 243 110 142 294 282 295 279 180 222 368 349 195 219 214 258 306 152 295 179 167 160

84 48 96 72 96 120 24 76 24 96 72 96 120 144 48 24 144 120 96 72 48 96 144 48 120 48 24 168 72 96 24 12 96 624 144 48 ND 96 192 432 312

0.77 0.59 0.62 ND 0.38


1.1

ND ND 1.02 0.83 0.97 ND ND 0.90 ND ND 0.60 0.84 ND


1.o

0.185 0.159 0.157 ND 0.195 0.107 ND ND 0.107 0.107 0.107 ND

ND
0.107 ND ND 0.157 0.107 ND 0.107
0.107

0.94 0.35 0.70 0.50 0.35 1.4


1.o

0.38 1 .o ND ND 0.83 ND ND ND ND 0.95 0.92 0.65 ND 0.57

0.191 0.131 0.191 0.203 0.107 0.107 0.193 0.107 ND ND 0.107 ND ND ND ND 0.107
0.107

0.152 ND 0.16X

* Abnormal pressure. but normal chemically. * 3 Samples 33-36 are from the Abheville field or the south and of the area sampled and appear to be in a different chemical family, S-Twp-R = section-township-range; ND = not determined.

* Abnormal pressure.

This Page Intentionally Left Blank

ABNORMAL PRESSURES IN THE GULF COAST AREA

357

? 6 *33a34

KEY X AbnzDressure Normal pressure

4.5

30

il
I

I
50

1
60

40

I 70

I
80

I
90

I
100

1
10 1

I
120
I D

CHLORIDE, g / l

Fig. 11.4. Plot of the depth of the wells versus concentrations of chloride in the formation waters.

--I
2.5

33.

3p

034

/ KEY X Abnormal pressure/ ?Normal presauro

0.5
BICARBONATE, p/l

Fig. 11.5. Plot of the depth of the wells versus concentrations of bicarbonate i the n formation waters.

358

GEOPRESSURED RESERVOIRS

in the north to 3,050 m in the south. Below this depth, the sands become less abundant and less widespread. The first abnormal pressure as calculated from shale resistivity is indicated by an arrow. The location of a producing horizon from which a water sample was taken is shown by the sample number in a circle. When the water sample was taken from a nearby well, not shown on the section, it was projected onto the section and shown as the sample number inside a square in Fig. 11.3. There is a general tendency for the dissolved salt concentration of the water samples t o increase with depth. This is shown in Fig. 11.4, which shows chloride plotted against depth. Since chloride is the predominant

KEY )< Abnormal pressure 0 Normal pressure

25

K
27

\
\
I

24

1 0

CALCIUM, g/l

Fig. 11.6. Plot of the depth of the wells versus concentrations of calcium in the formation waters.

ABNORMAL PRESSURES IN THE GULF COAST AREA

359

anion, it serves as an indication of the degree of concentration. The samples of water from abnormally pressured sands are shown as circles in x'es. All of them except 17, 22, 24, and 25 fall below the average concentration line, that is, they are less concentrated than they should be for their depth of burial. Sample 1 especially is much too weak. Bicarbonate, while occurring in much smaller quantities, shows the reverse relation, decreasing in amount with depth, as shown in Fig. 11.5. The waters from horizons with abnormal pressures have more bicarbonate than they should, considering their depth of burial. Calcium increases with depth, as shown in Fig. 11.6. It would be more correct t o say that there are two types of water. Type 1includes waters with less than 5,000 mg/l calcium, all of which are shallower than 3,800 m; type 2 is water with more than 5,000 mg/l calcium, most of which is deeper than 3,500 m. The only minor constituent that indicated a significant change with depth was potassium, and it appears to increase relative to sodium. The abnormally pressured waters seem deficient in potassium for their depth.

Normal pressure
x

Abnormal pressu&e-Solution

o N o r m a l pressure- A l t e r e d relict bittern

100 O\&o

0%
'0

a0 0

"

"\

\
\ "\
. \ O

O \

60-

\ 0

A
0
0

0
SODIUM, g/l

Fig. 11.7. Comparison of some brines of a bittern type from the Michigan Basin with some brines from some normal and abnormally pressured reservoirs in Louisiana.

36 0
I50

GEOPRESSURED RESERVOIRS

Normal pressure

I25

x Abnormal pressure

I00

2 5 0

75
X

/
- x
X
X*

I n
50

25
Sodium=mg/l

Na

+ 40 mg/l

Ca

0.05

00 1
BROMIDE,

0.15

0.20

!5

g/l

Fig. 11.8. Plot of Na versus Br from some brines from normal and abnormal pressured reservoirs in Louisiana.

Four of the waters from high-pressure sands (17, 22, 24, 25) have normal concentrations of dissolved solids for their depth. The other waters from high-pressure sands (1-5, 13, 28, 39, 41) have lower concentrations than normal. They also have less calcium, more bicarbonate, and a higher Cl/K ratio. About 80% of the material in the Gulf Coast shale is clay. Assuming that the waters have reacted with montmorillonite, there should be a direct relationship of calcium t o sodium. Plotting the calcium and sodium data in Table 11.1 plus some data for some brines from the Michigan Basin (as shown in Fig. 11.7) indicate that a relationship of calcium t o sodium does exist in the Gulf Coast waters and that they probably have reacted with montmorillonite. Fig. 11.7 also indicates that the Gulf Coast waters are not an altered relict bittern as are the Michigan Basin brines. In an ion exchange reaction with montmorillonite, 2 moles of sodium are exchanged for 1 mole of calcium, therefore, if salt is redissolved the bromide content in solution should be proportional to the original redissolved solu-

ABNORMAL PRESSURES IN THE GULF COAST AREA

36 1

tion. However, because of the exchange reaction the sodium in solution should be Na + 46/40 Ca or Na'. Fig. 11.8 is a plot of Na' versus Br for the 41 samples. The data scatter to some extent but this can be expected if biogenic derived bromide is present and the presence of iodide indicates that such is the case. Fig. 11.8 indicates that re-solution of salt is a control in these samples. Fig. 11.9 shows further evidence that the Louisiana brines were formed by re-solution of salt. For example, the dashed line in the left portion of Fig. 11.9 is a plot of Na' versus Br of salt dissolved in distilled water, and the solid line just t o the right is a replot of Na' versus Br for the Louisiana brines. The next dashed line t o the right is Na' versus Br for evaporating sea water, and the curved dashed line is Na' versus Br for relict brines from the Michigan Basin. Notable differences in the waters found in the normally and abnormally pressured rocks are evident (Schmidt, 1973). The dissolved solids in the

-Re-solution solt in pure water Southwestern Louisiano brines

\*

I
I

*\

\*

25
I
I

9vaporoting seo water

t
\

*\

f *\

Sodium I=mg/l sodium

+%mg/l calcium
3

.3:
m>

2 BROMIDE, g/l

Fig. 11.9. Replot of Na' versus Br of the Louisiana brines (Fig. 11.8);plus data for relict Michigan brines, evaporating sea water, and resolution of salt. Resolution of salt is an important control for the Louisiana brines

362

GEOPRESSURED RESERVOIRS

normally pressured sandstones range from 600 to 180,000 mg/l, while in the geopressured sandstones the range is from 16,000 to 26,000 mg/l. The dissolvedsolids in the water in the pores of the shales adjacent to normally pressured sandstones are lower than the dissolved solids in the water in the sandstones, but the dissolved solids concentrations are similar in the waters of the adjacent high-pressure sandstones and shales. The concentration order > HC03- > Cl-, and in normally pressured in shale pore water is sandstone water it is C - > HC03- > S 0 4 - 2 . 1 The temperature gradient in the geopressured zone is about O.8l0C/25 m, while in the normally pressured zone it is about 0.44OC/25 m. This change in temperature gradient is believed t o be related t o the porosity, where a greater porosity causes a decreased thermal conductivity (Schmidt, 1973). The clay mineral composition in the geopressured zone is predominantly a nonexpandable type, while in the normally pressured zone montmorillonite, an expandable type, frequently occurs. This change is believed to be related t o the temperature, and the heat allows the release of water from the clays at temperatures of about 93-104OC. This released water will dilute the pore water and cause the dissolved solids to decrease. The total amount of water released by Gulf Coast shales in geopressured zones is about 13%of the total in the system (Schmidt, 1973). This can be a cause of the lower salinity of the waters found in the geopressured zones. Fowler (1970) studied the Chocolate Bayou field in Texas and evaluated the relationships between geopressure and the migration and accumulation of hydrocarbons. He concluded that faults tend t o act as barriers separating fluid systems in the area; however, cross-formational flow occurs with geopressure causing shale ultrafiltration of the waters. The ultrafiltration produces salinity variations in the waters. Hydrocarbon accumulation in the area is controlled by the hydrodynamic flow. According to Fowler (1970), hydrocarbons are trapped in the upper sands because of slight pressure differentials across fault traps in the West Chocolate Bayou field. However, in deeper strata, abnormal pressures have caused hydrodynamic flow and pressures greater than the displacement pressure in the fault, resulting in no trapped hydrocarbons. In essence then, sands with pressure gradients greater than 0.20 kgcm-' m-' in the Chocolate Bayou field do not contain commercial amounts of hydrocarbons. It also appears that the size of the accumulation may decrease with increasing pressure gradients up t o about 0.16 kg cm-* m-'. The accumulation size may increase with pressure gradients in the range of 0.16-0.19 kg cm-2 m-' and then decreases. Detection of abnormal pressures Estimation of formation pressures from electrical surveys is related to the following assumptions, concerning the origin of abnormal pressures (Foster and Whalen, 1966):

DETECTION OF ABNORMAL PRESSURES

363

(1) Shale porosity is a function of net overburden pressure and normally decreases with an increase in depth. (2) Shales with abnormal pressure will have a higher porosity than normally pressured shales at the same depth, because of the greater amounts of interstitial fluids. (3) Sand bodies (confined by lensing, faulting, etc.) surrounded by shale will have a pressure similar t o those in the shales. Data from acoustic and resistivity logs can be used to establish a shale transit time or shale resistivity versus depth of normal hydrostatically pressured formations. Deviation from the derived curve is used t o determine abnormal pressures (Hottmann and Johnson, 1965). The acoustic log is a function of porosity and lithology; therefore, in any given shale sequence it is primarily a measure of porosity. The acoustic response in normally pressured shales decreases in travel time (velocity increases) with increasing depth. This is the normal compacted trend, and the pressures in the shale are normal, or hydrostatic. Deviation from the normal compaction trend indicates an abnormally pressured zone. Relating the difference in the travel time of the observed formation pressure ( A T , ) t o a normal formation pressure (AT,) t o the formation pressure gradient (calculated from known depths and pressures of wells in the area), a pressure gradient (AT, -AT,) can be determined. The reservoir pressure can be found by multiplying this gradient by the depth. Fertl and Timko (1970) discuss several methods, using the theory of departure from the normal t o detect abnormally pressured zones. Methods they discuss are as follows: (1) Bulk density -this is a measurement of the intensity of back-scattered electrons produced by gamma-ray bombardment; this intensity varies with the bulk density of the rocks surrounding the borehole. (2) Conductivity measurements - measure of an induction log. Electromotive forces set up a current, which is detected by a receiver and recorded. Overpressured shales are noted by greater-than-normal conductivity reading resulting from higher-than-normal water content and porosity. (3) Borehole temperature - geopressured as usually associated with an increase in temperature. (4)Presence of gas in mud - this is not always a good detector, for gas can evolve from formation cuttings, as they come to the surface. One of the best means of obtaining subsurface information, other than drilling, is the use of the reflection seismograph. This geophysical tool is a measure of time between the earths surface and various subsurface reflecting horizons. The differences in interval velocities between these different horizons (formations) can be used to obtain a plot of average interval travel time, which varies exponentially with depth. The degree of departure from a normal plot of this travel time versus depth is related t o abnormally pressured reservoirs in the Gulf Coast area (Pennebaker, 1968). This departure is noted as an increase in the normally

364

GEOPRESSURED RESERVOIRS

decreasing travel time with depth, because of the undercompacted formations. To measure the formation pore pressure, plots of equal pore pressure gradients are compared, by an overlay, to the abnormally pressured interval travel time depth plots. Forgotson (1969),by experience with wells in the Gulf of Mexico, noted that the presence of high background gas and high trip gas, together with lower than normal shale density, does not necessarily indicate the proximity of an abnormally pressured reservoir. He believes that a minimum of 200% increase in the shale penetration rate when drilling is the best available means to predict abnormal pressures. A recent series of papers explains how downhole temperatures and pressures can affect drilling (Fertl and Timko, 1972;Timko and Fertl, 1972). Methods of detecting abnormal pressures, compensating for them, and evaluating the hydrocarbon potential of geopressured strata are discussed.

References
American Petroleum Institute, 1968. API Recommended Practice f o r Analysis of Oilfield Waters. Subcommittee on Analysis of Oilfield Waters, API, RP 45, 2nd ed., 49 pp. Burst, J.F., 1969. Diagenesis of Gulf Coast clayey sediments and its possible relation t o petroleum migration. Bull. A m . Assoc. Pet. GeoL, 53:73-93. Dickey, P.A., Collins, A.G. and Fajardo, I., 1972. Chemical composition of deep formation waters in southwestern Louisiana. Bull. Am. Assoc. Pet. GeoL, 56:1530-1533. Dickey, P.A., Shiram, C.R. and Paine, W.R., 1968. Abnormal pressures in deep wells of southwestern Louisiana. Science, 160:609-615. Dickinson, G., 1953. Geological aspects of abnormal reservoir pressures in Gulf Coast Louisiana, Bull. A m . Assoc. Pet. GeoL, 37:410-432. Fajardo, I., 1968. A Study of the Connate Waters and Clay Mineralogy. M.S. Thesis, University of Tulsa, Tulsa, Okla., 50 pp. . Fertl, W.H. and Timko, D.J., 1970. Overpressured formations, 2 How abnormal pressure-detection techniques are applied. Oil Gas J., 68:62-71. Fertl, W.H. and Timko, D.J., 1972. How downhole temperatures, pressures affect drilling. World Oil, 174(7):67-70; 175(1):47-49; 175(2):36-39, 66;175(4):45-50; 176(2): 47-50. Finch, W.D., 1969. Abnormal pressure in the Antelope field, North Dakota. J. Pet. Technol., 21:821-835. Forgotson, J.M., 1969. Indication of proximity of high pressure fluid reservoir, Louisiana and Texas Gulf Coast. Bull Am. Assoc. Pet. GeoL, 53:171-173. Foster, J.B. and Whalen, H.E., 1966. Estimation of formation pressures from electrical surveys - offshore Louisiana. J. Pet. TechnoL, 18:165-171. Fowler, Jr., A.W., 1970. Pressures, hydrocarbon accumulation, and salinities - Chocolate Bayou field, Brazoria County, Texas. J. Pet. TechnoL, 22:411-423. Harkins, K.S. and Baugher, 111, J.W., 1969.Geological significance of abnormal formation pressures. J. Pet. TechnoL, 21:961-966. Hottmann, C.E. and Johnson, R.K., 1965. Estimation of formation pressures from logderived shale properties J. Pet. TechnoL, 17:717-721. Jones, P.H., 1969. Hydrodynamics of geopressure in the North Gulf of Mexico Basin. J. Pet. TechnoL, 21:803-810. Pennebaker, E.S., 1968. Detection of abnormal pressure formations from seismic field records. Presented at API Southern Dist. Meet., San Antonio, Texas, March 6-8, 1968, API Paper, No. 926-13C.

REFERENCES

365

Perry, D.R., 1969. A Correlation of Reserves and Drive Mechanisms with Reservoir Pressure Gradients on Geopressured Gas Reservoirs in Southwest Louisiana. M.S. Thesis, Southwest Louisiana University, Lafayette, La., 54 pp. Powers, M.C., 1967. Fluid-release mechanisms in compacting marine mudrocks and their importance in oil exploration. Bull. Am. Assoc. Pet. GeoL, 51:1240-1254. Schmidt, G.W., 1973. Interstitial water composition and geochemistry of deep Gulf Coast shales and sandstones. Bull. A m . Assoc. Pet. Geol., 57:321-377. Sulin, V.A., 1946. Waters of Petroleum Formation in the System o f Natural Waters. Gostoptekhizdat, Moscow, 96 pp. Terzaghi, K. and Peck, R.R., 1948. Soil Mechanics in Engineering Practice. John Wiley and Sons, New York, N.Y., 56 pp. Timko, D.J. and Fertl, W.H., 1972. How downhole temperatures, pressures affect drilling. World Oil, 175(5):73-81; 175(6):79-82; 175(7):5*62; 176(1):45-48; 176(4):. 62-65. Wallace, W.E., 1969. Water production from abnormally pressured gas reservoirs in South Louisiana. J. Pet. Technol., 21 :969-982. Young, A. and Low, P.F., 1965. Osmosis in argillaceous rocks. Bull Am. Assoc. Pet. Geol., 49:1004-1008.
\

This Page Intentionally Left Blank

Chapter 12. COMPATIBILITY OF OILFIELD WATERS

Waters used for the secondary recovery of oil by waterflooding usually contain a number of inorganic salts and sometimes organic salts in solution. It is common practice t o test the compatibility of the injection water and water in the formation before starting a waterflood operation. Often this test is performed by mixing the injection water with the formation water in a glass container and observing t o determine if a precipitate forms. The precip itate or scale can be analyzed to determine its composition. Waters are compatible if they can be mixed without producing chemical reactions between the dissolved solids in the waters and precipitating insoluble compounds. The precipitated insoluble compounds are undesirable because they can reduce the permeability of a porous petroleum-productive rock formation, plug input wells in waterflood systems, and cause scale formation in water pumps and lines. Some of the more common ions that frequently occur in oilfield waters and that cause precipitation in incompatible waters are: Ca+2,S P 2 , Ba+2, Fe+?,HC03-, and Common reactions are: CaC1, + Na2S04 CaC1, +MgS04 c a w 0 3 )2 CaC12 + 2NaHC03 SrC1, + NaS04 SrClz +MgSO, BaC12 +NaS04 BaCl2 +MgSO, Fe + H2S Fez03 + 6H2S
+= +=
-b

+= += += += +=
-b

+=

2NaC1+ CaSO, MgClz +CaS04 C 0 2 + H 2 0 + CaCO, 2NaC1+ C 0 2 + H 2 0 + CaCO 2NaC1+ SrSO, MgC12 + S r S 0 4 2NaC1+ BaSO, MgC12 +BaS04 H2 + FeS 6H,O + 2Fe2 S3

33J-

33J-1

335.

A relatively insoluble compound CA where C is the cation and A is the anion will precipitate from an aqueous solution if:

where ac = the cation activity, a A = the anion activity in the solution, and

SCA = the solubility product of the compound CA. When two salts with a
common cation (CAI and CA2) are in equilibrium in a solution, the following will hold:

36 8

COMPATIBILITY OF OILFIELD WATERS

a~~

I ~ A , = &!Al PCA,

If a A l l a A ,
a~~

> SCA /SCA,, CA,

will precipitate, and CA2 will dissolve if

I ~ A , <SCA,ISCA,.

Deposition of scale in both primary and secondary recovery producing wells and formations is a very costly problem in the petroleum industry. The scale not only restricts production but also causes inefficiency and production equipment failure. Scale deposits are caused by mixing incompatible waters and by environmental changes during the production of well fluids. For example, as production begins, the pressure drops in the vicinity of the wellbore, allowing dissolved gases to escape from solution. The loss of C 0 2 can cause calcium carbonate t o precipitate. The decrease in pressure also can cause the vapor pressure of the brine to increase. The temperature of the brine will decrease because heat energy is required to vaporize the water, causing calcium sulfate t o precipitate. Wellbore and formation damage In several case studies Vetter and Phillips (1970) found that calcium sulfate deposits form in both primary and secondary petroleum production operations. The scale forms within the formation and causes production loss and permanent damage. In many cases damage to the formation cannot be corrected even by fracturing. Research has indicated that sodium carbonate can cause the metathesis of anhydrite and gypsum to calcium carbonate. This might work in a formation that is partially plugged. For example, a water containing sodium carbonate could be injected into the formation and allowed t o react with the scale. An acidified water then could be injected into the formation to remove the carbonate and hopefully clean the formation, allowing recovery of more oil. Potential scale deposition should be predicted as soon as the well begins production, and the correct inhibitor should be added immediately rather than following the common practice which is to pull the tubing after a production decline and find scale on the metal surface. Scaling can occur within the formation and never show up on the tubing. Pressure drops are the primary cause of calcium sulfate scaling within a formation, when the formation brine is saturated with calcium sulfate. Important variables related t o scaling are: (1)Temperature of the formation in relation to solubility of the possible scale former in the fluids passing through it. CaS04 becomes less soluble

WELLBORE AND FORMATION DAMAGE

369

with increasing temperature (Blount and Dickson, 1969), while BaS04 becomes more soluble (Templeton, 1960). (2) Subsurface pressures change for any system, with the highest pressure found while the fluid flows through the formation. The greatest pressure change is at the sand face of the producing well (Vetter and Phillips, 1970), which causes this area t o be where solubility changes are the greatest. Deposition of scale at this point is the most damaging to oil production and the most difficult t o discover or to remedy. Very few data are available on pressuresolubility relations of most scale forming compounds, but CaSO, has been shown to decrease in solubility with decrease in pressure at NaCl concentrations t o 10%(Fulford, 1968). (3) Brine concentration, exclusive of precipitating compounds, also influences scale formation. Most electrolytes in ionic form cause an increase in the solubility of compounds which form scales. The solubility normally increases with increasing electrolyte concentration unless some other solubility equilibrium is reached. This can occur, for example, when BaS04 saturation level is reduced because of increasing amounts of Ca+2 ion in the solution (Davis and Collins, 1971). Other properties of brine known to influence the solubility levels of scale formers are gases in solution, hydrogenion concentration, ion pairs, and dissolved organic chelates (Weintritt and Cowan, 1967). Waterflooding of petroleum reservoirs has been successfully carried out for many years. Large quantities of petroleum are produced through secondary recovery by forcing water (usually a brine) into an oil sand which has become unproductive by primary production methods. However, the efficiency of the operation is often low, and the amount of petroleum remaining in the sand after waterflooding can be as high as 50% of the original accumulation (Shaffer, 1967). One of the reasons such a high proportion of the oil remains unrecovered is because the injection pressures become economically too large to continue forcing water through the sand t o displace the oil. The gradual deposition of solid material precipitating from the water closes the permeable channels and slows the flow at the producing well. Dilution of the water injected into a formation often occurs, and additional makeup water is necessary. The slow mixing of connate (interstitial) waters of the formation or the introduction of water from associated aquifers, both underground and on the surface, contributes to the instability of the injection water. The deposition of scale in wellbores, sand faces, and piping has reduced oil production in many fields. Removal of scale is difficult, often impossible, and methods t o avoid its formation need additional development. Scaling results from the precipitation of a solid from a formation water or from injection water in waterflood operations. The most common causes of scale are: (1) temperature and/or pressure changes t o which the formation water is subjected; (2) dilution with makeup water (in secondary recovery opera-

370

COMPATIBILITY O F 0ILF IELD WATERS

tions) or mixing with other formation water containing incompatible ions; (3) evaporation causing increased concentrations of dissolved solids allowing saturation t o be reached; (4) supersaturation caused by formation water flowing through and dissolving slightly soluble solids. When the composition and temperature of a brine, saturated with CaSO,, remain constant, precipitation will occur if the pressure drops. Scaling is not likely with increasing pressure under the same conditions. However, these conclusions must be modified if the brine is flowing through beds conkaining soluble compounds of calcium or sulfate or if another water source is altering the brine concentration. Because of moderate rise in brine temperature as it travels betwsen the injection wellhead and the bottomhole and the rapid rise in pressure, scaling of CaS0, is not likely in the injection well. The formation of BaS0, scale is worthy of special attention. Most barium compounds are relatively insoluble, and large volumes of brine often are necessary t o cause heavy BaS0, scale. The most unique characteristic of this compound is its crystal growth (Weintritt and Cowan, 1967). It will remain in a supersaturated solution for an unpredictable time and then will precipitate slowly and in a crystal form which has not been duplicated in the laboratory. Some observers attribute this phenomenon t o the requirement of a unique solid crystal acting as a seed to promote BaSO, precipitation. Furthermore, the forming crystal adheres to other larger solids suspended in the solution or attached to the associated solid phase. This causes the scale to occur in larger quantities than if it were pure barite. In the Raleigh field, Smith County, Mississippi, a scale consisting of concentric rings of prismatic barite commonly occurs. The barite prisms are about 0.5 mm in length and contain up t o 1% strontium and lead. The pumping equipment in the Pisgah field, Rankin County, Mississippi, often is plagued with a scale composed of metallic lead containing small fragments of steel. The steel is from the pumping mechanism but the lead must be from the formation water because the amount of dissolved lead ranges up t o 100 mg/l. Maintenance of the wells to remove the lead scale occurs as often as every 10 days. Knowledge of the solubilities of BaS0, and SrS0, in solutions containing NaCl, CaCl,, and NaHCO, needs to be increased t o better understand various precipitation reactions that occur when waters containing these salts mix. Information concerning the effects of heat and pressure upon these reactions is lacking. Solubility of calcium compounds in various salt solutions Frear and Johnston (1929) measured the solubility of calcite in water at saturated with carbon dioxide and obtained an activity of 4.8 x 25OC. Ellis (1963) determined that the solubility of calcite was significantly less in the laboratory salt solutions than in hydrothermal solutions with similar ionic strength.

SOLUBILITY OF CALCIUM COMPOUNDS

371

Stiff (1952) developed a graphic method of predicting the tendency of oilfield waters to deposit calcium sulfate. Diagrams can be used to find the maximum solubility of a salt in waters of similar composition. This information is useful in predicting that a given brine has a scale forming tendency. However, better pressure and temperature data in respect t o their effect on scale formation are needed. Akin and Lagerwerff (1965) studied the solubility of calcite in relation t o ionic strength. The soluble salts used were NaC1, NaHCO,, and CaC12; the ionic strength of the solutions ranged up t o about 0.09 and their data agreed well with the Debye-Huckel theory. They also studied the effect of Mg" and S04-2 and found that the solubility of calcite was enhanced by these ions relative to theoretical values. Ostroff and Metler (1966) determined the solubility of calcium sulfate dihydrate in the system NaC1-MgC12-H20 in 5.50 molal NaCl and 0.340 molal MgC12 admixtures at 28", 38', 50, 70, and 90'C. Their results indicate that the solubility increases in the presence of small amounts of MgC12 in NaCl solutions up t o about 2.5 molal NaC1. The MgC12 effect decreases in higher molalities of NaCl until at about 4 mold NaCl a plateau is reached. Shaffer (1967) studied the solubility of gypsum in sea water and sea-water concentrates. He found that the solubility product of gypsum is greater in the highly concentrated brines and also that in these brines it increased with increasing temperature. Glater et al. (1967) developed a method t o measure calcium sulfate scaling thresholds in saline water samples at 100'C. They found a correlation of ionic strength with calcium sulfate solubility, and used a gSaphical method to relate scaling threshold to the concentration of calcium and sulfate ions in saline water. Pytkowicz et al. (1967) measured in situ the pressure coefficient of the aragonitic oolites with pH electrodes, Their results indicate that the pressure coefficient or the apparent solubility must be known to obtain accurate solubility data at high pressures. Fulford (1968) found that the solubility of gypsum or anhydrite increases with pressure because of a small decrease in total volume as the scale dissolves. Subsequently with a pressure drop a supersaturated solution forms and gypsum precipitates. In very concentrated brines this does not occur because the solubility of gypsum in very concentrated brines is less dependent upon pressure. He presented several equations t o calculate anhydrite and gypsum solubilities. Blount and Dickson (1969) determined the solubility of anhydrite in NaCl solutions at 100'-450'C and 1-1,000 bars. They found that anhydrite solubility increased with temperature and NaCl concentrations. Glew and Hames (1970) determined the solubilities cf gypsum, disodium pentacalcium sulfate, and anhydrite in sodium chlor,de solutions. Their results indicated that the solubilities of these comr#ounds decreased in chloride solutions with molalities greater than 3.5.

372

COMPATIBILITY OF OILFIELD WATERS

Vetter and Phillips (1970) included the effects of complicated downhole phase equilibria to develop an improved thermodynamic method to predict deposition of calcium sulfate. According to them the calculated solubility is as accurate as the experimentally determined solubility; however, additional data are needed concerning the solubility of gypsum in brines at high pressures. These data are needed to determine which CaS04 compounds are formed under high pressure in brines. Knowledge of pressure drops either at the wellhead or within the reservoir is important to determine where scale deposition occurs. Solubilities of the sulfates of barium and strontium in saline solutions Neuman (1933) published results of studies of BaS0, solubility in aqueous solutions of potassium, magnesium, and lanthanum as chlorides and nitrates. His data show that BaS04 solubility increases with the increasing complexity of the major solute, and in the order (3, -1) > (2, -1) > (1,-1) of equal molality solutions. Gates and Caraway (1965) analyzed California oil-well scale and found in a BaS0,-type scale significant amounts of strontium along with iron, calcium, magnesium, and some carbonate. Weintritt and Cowan (1967) studied the unique characteristics of BaS04-scale deposition and concluded, the presence of strontium in barium sulfate scales deposited from oilfield waters appears to be common. All of the sulfate deposits analyzed contained strontium sulfate in concentrations ranging from 1.2 to 15.9%, and barium sulfate in concentrations ranging from 63.7 t o 97.5%. Templeton (1960) studied the solubility of BaS04 in solutions at 25OC and at sodium chloride molalities between 0.1 and 5.0. He found that at constant ionic strength the solubility of BaS0, increases with increasing temperature, and observed that calcium sulfate exhibits an inverse reaction with increasing temperature. Experimental determination of some solubilities of the sulfates of barium and strontium

A radioisotope-tagged solution of Na2SO4 was prepared from which aliquots were taken (Davis and Collins, 1971). The radioactive isotope was 35S. One aliquot was used to precipitate BaS04 by addition of an excess of BaCl,; a second portion was used to precipitate SrS04 by addition of appropriate equivalents of SrC12.These suspensions were stirred and allowed t o settle. Following a 2 4 t o 48-hour settling period, the precipitates were washed onto a 0.45-pm pore size filtering medium, and the washings were continued until the sulfate ion in the filtrates could not be further reduced. The tagged precipitate was removed from the filter, dried in an oven at 105OC, and transferred t o storage vials. Standard samples of the sulfates were prepared by chelation in a 0.W solution of EDTA. Various strengths of

RESULTS AND DISCUSSION

373

5-80 mg/l of BaSO, and 50-800 mg/l of SrS04 were made and used as reference counting samples for all of the sulfate determinations. A nonionic detergent (Triton X-100) and toluene emulsion (Patterson and Greene, 1965) were prepared, whereby l-cm3 sulfate samples in brine could be counted with greater than 20%efficiency. The emulsion forms a clear gel and permits a homogenous dispersion of the aqueous phase in the fluor with no salting out. Solutions of various salts, such as those usually found in formation waters, were made up in strengths of 0.005-1.77 molal, and tagged solid barium or strontium sulfate was added. The chlorides of sodium, calcium, magnesium, and potassium were prepared, as were solutions of sodium bicarbonate, sodium borate, and potassium bromide. All of the solutions were stored in plastic bottles. To determine sulfate solubility, a 20-cm3 portion of one of the prepared salt solutions was transferred to a small plastic stoppered vial, and 0.1 g of the solid, tagged sulfate was added. This suspension was shaken in a wrist-action type shaker for 72 hours and then allowed t o settle a minimum of 24 hours without opening the vial. The samples were prepared in duplicate to assure equilibrium, and the operation was repeated when better precision was needed. The temperature of the suspension was raised briefly above the stabilized room temperature (25OC k l 0 C ) with a heat lamp during the shaking period, but no change was permitted during the last 24 hours nor during the settling period. When the sample container was opened, it was quickly filtered through a double Whatman No.42 filter paper, and 1 ml was transferred t o a counting vial which contained 12 ml of the Triton emulsion and 7 ml of deionized water. The sample then was counted in a liquid scintillation counter for 50 minutes. The chelated sulfate standards were counted in the same time period. By this method, the correction for radioactive decay could be omitted and the soluble sulfate values determined from a graph of the chelated standards (in mg/l) versus the counts per minute. Barium was analyzed by emission spectroscopy, but adequate precision at levels of 1 mg/l and less was difficult to achieve in the presence of ionic-strength salts encountered in some solutions. Results and discussion of the experimental investigation The values obtained from solubility measurements are shown in Table 12.1. The amounts of the alkaline sulfates which dissolve in other electrolyte solutions are tabulated alongside the total ionic strength of each solution. Ionic strength is the most useful concept yet developed t o include the combined effects of the activities of several ionic species in a solution. Lewis and Randall (1923) state, in dilute solutions, the activity coefficient of a given strong electrolyte is the same in all solutions of the same ionic strength. It is defined as s = H m 1 2, , where m 1 = the ionic molality, and 2 , = the charge of the ion in solution, the summation being taken over all ions, positive and negative. By definition, the activity of the dissolved species approaches the concentration value (molality) a t infinite dilution.

374

COMPATIBILITY OF OILFIELD WATERS

Since the thermodynamic solubility product Ku = x mso, x y2 and since y equals unity at zero ionic strength, a plot of log mso, versus the ionic strength function would extrapolate to zero concentration where log KuS = log rnso,. Fig. 12.1-3 give plots representing six electrolytes and the values of Ku'h are determined graphically. The value for the sulfate solubilities in pure water was determined experimentally and agreed with values in the literature. In Fig. 12.1, the plot of BaS04 solubility versusds for the six electrolytes is almost identical at low ionic strength, a phenomenon to be expected from the statements above. The extrapolated KuS values of all systems are 1.05 x loe5 (within experimental limits). This fact must be correlated with the nature of the equation defining ionic strength. The square of the ionic valence gives the Mg+' and Ca+' ions four times the numerical weight of the Na+ and K+ ions. Molality values would indicate that the bivalent ions cause increased solubility effects. Because borates are present in many oilfield waters, sodium borate was included t o find differences in sulfate solubility in electrolytes containing a complex ion. As shown in Fig. 12.1 and 3, the solubility deviated from that of monatomic electrolytes, and the relationship described does not hold at higher solubilities of electrolytes containing complex ions. Another ion commonly found in mineral waters is bicarbonate. Many water-bearing zones contain limestone and dolomite which slowly erode in water of low pH. The water carries away carbonates and bicarbonates. In this study, NaHCO, solutions of 0.005-1.0 molal were saturated with tagged
10.0

6.0-

0.00

KEY CoC12 MgC12 No2 84 07 KBr NaCl KCI

__

4.0
0
I

A A
0

0
x

2.0 -

0
v)

m
l. I

1.0

0.8 0.6

0+

SO4 Mo Io I i t ies

k
_I

a 0.4 J
0

0.2

0. I

0.5

I .o

I .5

2.0

2.5

4 l O N l C STRENGTH

Fig. 12.1. Concentration of saturated B a s 0 4 in strong electrolyte solution.

RESULTS AND DISCUSSION


TABLE 12.1 Solubility of Bas04 and SrS04 in electrolyte solution Major solute (molality) Bas04 major solute system Bas04 (mg/l) total ionic strength

375

5r504
major solute system SrS04 (mg/l) total ionic strength

ca Cl2
0.010 0.015 0.020 0.025 0.045 0.050 0.090 0.100 0.136 0.200 0.226 0.300 0.400 0.456 0.500 0.934 1.000 2.000
M m 2

(5.0)*

6.2 7.6
11.5

0.03016 0.06013

214 247 295 260

0.0347 0.0504 0.0815 0.1403 0.2819 0.4197 0.6959 1.3947 1.8438

0.15020 508 0.30027 590 0.60030 757 0.90028 1.20028 1,152 1.50028 1,942 3.00019 6.00004 172 203 233 295 394 0.0188 0.0344 0.0499 0.0805 0.1571 0.2324 0.2866 0.7782 1.5993

15.5 (17.3)* 17.7 16.2 16.3 16.6 (16.3)* (10.8)*


11.3 25 .

0.005 0.010 0.015 0.020 0.025 0.049 0.050 0.074 0.099 0.125 0.196 0.254 0.474 0.525 0.902 1.637

(2.9)*

5.4 69 . 9.8 13.3 18.0 25.9

0.02979 0.05982 0.14927

422 0.29693 530 0.58831 731 1.42244 1,063 (44.5)* 32.0 33.2 2.70655 4.93257

See footnote at end of Table.

376
TABLE 12.1 (continued) Major solute (molality) Bas04 major solute system Bas04 (mg/l)

COMPATIBILITY OF OILFIELD WATERS

5r504
major solute system total ionic strength SrS04 (mg/l) total ionic strength

Na Cl 0.010 0.015 0.020 0.025 0.050 0.086 0.100 0.172 0.200 0.257 0.431 0.500 0.869 1 .ooo 1.771 2.000
KCl 0.010 0.015 0.020 0.025 0.050 0.067 0.100 0.200 0.202 0.338 0.500 0.684 1.000 1.396 2.000

(5.3)* (5.6)*

3.6 42 . 5.4

0.01006 0.02007 0.05009 0.10012

134 149 172 199 265 332

0.0129 0.0182 0.0288 0.0543 0.0914 0.1756 0.2667 0.4423 0.8840 1.7875

(7.3)* (11.3)*

71 . 10.0 14.8

0.20017 420 525 0.50025 699 1.00035 760 2.00047 0.01007 0.02008 167 0.0286 0.0754 0.2109 0.3492 0.7001 1.4139 0.05011 375 0.10015 0.20019 396 502 0.50029 742 1.00037 802 2.00047 144 169 0.0131 0.0185

(22.3)* (35.7)* (3.7)*

20.2 27.2 4.2 49 . 6.3 8.6 11.2 16.8 21.6

(25.8)*

27.2

See footnote at end of Table.

BaS04. However, only trace amounts of barium were found in solution, though the sulfate content increased with the amount of NaHCO, in solution. This apparent anomaly can be reconciled by the ionization of the HC03- ion into CO,-*, which in appreciable concentration would reduce the Ba+2 ion concentration according to the solubility product K,, = M B x~
MC03.

RESULTS AND DISCUSSION TABLE 12.1 (continued) Major solute (molality) Bas04 major solute system Bas04 (mg/l) XBr 0.010 0.015 0.020 0.042 0.050 0.084 0.100 0.126 0.200 0.211 0.426 0.500 0.866 1.000 2.000
Na2 B4 0 1

377

5r504
major solute system total ionic strength SrS04 (mg/l) total ionic strength

(3.6)*

41 . 47 . 6.3 8.2 11.0 16.2

0.01007 0.02008

152 163 215

0.0133 0.0186 0.0467 0.0900 0.1335 0.2207 0.4378 0.8812

0.05011 262 0.10014 320 0.20019 420 509 0.50028 669 1.00037 2.00046 0.03010 320 0.0462 0.1307 0.2100 0.06014 600 0.15022 690 0.30036 0.60060 0.00004 114 0.0025

(23.9)*

21.8 26.7 60 . 78 . 12.9

0.010 0.013 0.020 0.039 0.050 0.065 0.100 0.200


Pure Water

(23.7)* (33.9)*

21.0 34.8 2.5

Parentheses indicate barium ion and sulfate. ion determinations made separately.

The effect of high concentration of CaClz on BaS04 solubility is indicated by the solid curve of Fig. 12.1. At concentrations of 2 molal, the BaS04 solubility has dropped t o values close to that of the compound in pure Hz0. The maximum value is reached between 0.2 and 0 4 molal where . the decline begins. The accompanying broken line of Hg. 12.1, which is a plot of Ba+2 + SO4-' ions determined separately, shows the reduced solu-

378

COMPATIBILITY OF 0IL F IE LD WATERS

-1 IONIC

STRENGTH

Fig. 12.2. Concentrations of saturated SrS04 in strong electrolyte solutions of NaCl, KCl, and KBr.

- K a = 2.4 x
I

I
I I I I I

4 IONIC

STRENGTH

Fig. 12.3. Concentrations of saturated SrSO4 in strong electrolyte solutions of MgClz, CaClz , and Na2 B4 0,.

RESULTS AND DISCUSSION

379

bility of the Ba ion caused by the equilibrium Ca+ + S04-2 CaSO, ( K s p 10-4). The effect of the ions of strong electrolyte solutions on SrS04 solubility is similar t o that observed when BaS04 solubility was studied. The Na+, K+, C1-, and Br- ions have approximately equal effect, and all determined values fall on a common curve (Fig. 12.2). The increase in sulfate solubility is marked in dilute solutions but reaches a maximum a t concentrations with ionic strength near 1. This is the average value calculated for sea water. When bivalent ions Mg+ and Ca+2 are used in the strong electrolyte (Fig. 12.3), the SrS04 solubility remains of the same relation t o the total ionic strength as for the monovalent ions. A study of the system SrS04-NaHC03 -H2 0 was limited by the insolubility of SrCO, . The ionization of the bicarbonate to H+ and COSw2would result in the precipitation of any Sr+ which dissolves and leaves the S04-2 in solution. This relationship is similar t o the BaS04-NaHC03 -H2 0 system and is worthy of special note. That is, when carbonate or bicarbonate waters are diluted or intermixed with waters containing barium or strontium, an unstable solution is formed. Experimental data indicate that maximum sulfate solubility in strong electrolytes begins at an ionic strength of approximately 1.When the principal cation in solution is the Ca+2 ion, sulfate solubility decreases after the ionic strength exceeds unity. Blount (1965), when measuring solubility of CaSO, in the system CaS04-NaClLH2 0, and Lucchesi and Whitney (1962),

< 1.95

TABLE12.11 Sulfate solubilities in synthetic brines


~ ~ ~~~

Concentration (molal) brine 1 Na+ Ca+ Mg+ K+ Br11.2179 0.0250 0.0206 0.0051 1.3019 0.0125 0.0001 60 2.57 x D4 1.3600 813 44.26 x lo4 1.3777 brine 2 1.7399 0.0374 0.0823 0.0051 1.9650 0.0188 0.0000 63 2.70 x D4 2.1038 922 50.19 x lo4 2.1239 brine 3 2.4359 0.0499 0.0411 0.0193 2.6113 0.0250 0.0001 66 2.83 x 3.0278

c1

Barium sulfate solubility Bas04 (mg/l) Bas04 (molality) Ionic strength (s) Strontium sulfate solubility SrS04 (mg/l) SrS04 (molality) Ionic strength (s)

o4

958 52.18 x lo4 3.0487

380

COMPATIBILITY OF OILFIELD WATERS

when measuring SrS04-NaC1-H2 0 solubility equilibria, found similar maximums. By using ionic-strength calculations in place of weight per unit volume, the predictions of mineral water stability become more accurate and dilutions more feasible. Three synthetic brines were made with salts concentrations in the range of many formation waters and containing the major salts found in these waters. Table 12.11 gives these concentrations and results of a BaS04 and a SrS04 solubility determination. The values found when plotted against the ionicstrength function of the brine fall on the same curve as the barium salt in Fig. 12.1 and the strontium salt in Fig. 12.2. No carbonates were added t o these synthetic brines.

Brine stabilization
Efforts to stabilize the brines used in petroleum production have been extensive and successful in many cases, but the complexity of the problem in other cases is reported. Water treating units are considered necessary in waterflooding operations, but none fully satisfy the operators apprehension that there may be plugging within the reservoir. Addition of solubilizing, chelating, and clarifying agents to the brine has helped, but economics limit the quantities used. Tests for compatibility of the fluids as they exist in the formation and in the wellbore give erroneous results because the subsurface environment cannot be fully duplicated a t the surface. To aid brine stabilization programs, several studies of the solubilities of various relatively insoluble compounds have been made as previously discussed. Usually the results of these studies are reported as solubility products of various pure compounds (CaS04, BaS04, CaCO, , etc.) in the presence of other ions, dissolved gases, ion pairs, and various sized crystals of the compound under study. Some efforts have been made using mixed cations and anions in solution with the compound under study, including limited study of sulfates in sea water or synthetic sea water (Shaffer, 1967). Various easily measured parameters such as percent chlorides, total solids, and ionic strength have been plotted against solubility product of the potential scale former. Very little correlation suitable for direct field application has been found. For example, the author has measured BaS04 solubility in CaC12, MgC12, NaC1, and other salt solutions using ionic strength as the common property. However, a synthetic sea water containing these compounds and having comparable ionic strength will dissolve double the weight of BaS04 in milligrams per liter with respect to any solution containing a single salt. Fulford (1968) and Vetter and Phillips (1970) proposed useful formulas and graphs t o use in predicting scaling from calcium sulfate. Fig. 12.443 are included for possible use in predicting potential scale problems from calcium sulfate, strontium sulfate, and barium sulfate. The figures are plots of molal solubility versus ionic strength. The advantage of this plot is that the ionic strength of any given water can be calculated from its chemical composition,

BRINE STABILIZATION

381

IONIC S T R E N G T H

Fig. 12.4. Solubility o f CaSO4 versus ionic strength of aqueous solutions (Ostroff and Metler, 1966).

12-

KEY

1
IONIC S T R E N G T H

Fig. 12.5. Solubility of SrS04 versus ionic strength of aqueous solutions containing CaClz, MgClz, NaCl, KCl, and KBr (Davis and Collins, 1971).

and the solubility of a given compound is a direct function of the ionic strength of the solution. Therefore, a very good approximation of the solubility of a given compound in a given water solution can be made. For example, if a water with an ionic strength of 0.1 contains 0.001 molal of strontium sulfate, it can be assumed that the water is undersaturated with respect to strontium sulfate as illustrated in Fig. 12.5. However, if the water contains 0.003 molal of strontium sulfate it is oversaturated and some treatment should be made if the water is to be reinjected.

382

COMPATIBILITY OF OILFIELD WATERS

0 1 I a I I ,111 0 0 0.02 0.040.0601 .1


I

I IIII

I , 1 , 1 ,

a2

04 060.0 1.0

6 8 3

IONIC STRENGTH

Fig. 12.6. Solubility of B a s 0 4 versus ionic strength of aqueous solutions containing CaClz, MgClz, NaCI, KCl, and KBr (Davis and Collins, 1971).

Similar curves can be made using appropriate solubility data for calcium carbonate and for iron compounds. However, it should be noted that application of this technique only gives an estimation of the maximum solubility of a compound in waters of similar ionic compositions. Better data on pressure and temperature and how they affect the solubilities are needed before adequate prediction equations can be developed. Mixing of subsurface waters Mixing of surface and subsurface waters results in solutions which are either saturated or undersaturated with relatively insoluble compounds such as calcium carbonate, calcium sulfate, strontium sulfate, and barium sulfate. These compounds are considered because they often are found in scales formed because of mixing of formation waters. Hydrodynamic potentials caused by differences in elevation, weight of the overlying fluids and rocks, secondary cementation of rock pores (Levorsen, 1967), temperature differences, osmotic pressures, and chemical and physical reactions cause subsurface waters t o move (Hubbert, 1953).Popov and Goldshteyn (1957) described a large hydrodynamic system of descending fresh water and ascending saline water which could mix to form a fresh-saline water mixture. Henningsen (1962)found that recharge waters into Trinity aquifers were two types of water from strata of different lithology and with basinward movement of the waters they mixed to form a third type of water. Mixing of fresh waters with encroaching sea water Columbus (1965), and Upson (1966). occurs according t o Kohout (1960),

MIXING OF SUBSURFACE WATERS

383

Estimating strontium sulfate saturation in waterflood makeup brines (Biles, 19 72)


The data in Tables 1 . and I1 were used by Biles (1972) estimate the 21 to saturation point of strontium sulfate in waterflood makeup waters. According to him, brines used as makeup water for waterflood operations often are more concentrated in dissolved solids than are the single solute samples shown in Table 1 . . However, considering that the sodium concentrations 21 are 96, 93,and 97 mole %, respectively, in the synthetic brines 1, 2, and 3 shown in Table 1 . 1 it appears reasonable in lieu of experimental data to 21, extend the NaCl data to the higher.concentration range with these data. Fig. 12.7 is a plot of the milligrams per liter of strontium sulfate in solution as a function of the total ionic strength of the solution. The data 21 were taken from Table 1 . and 11. A smooth curve can be plotted for Fig. 12.7 if the strontium sulfate value at 1.7875 total ionic strength (Table 12.1) is ignored. This curve can be extrapolated for use in estimating the amount of strontium sulfate in milligrams per liter that is likely t o be soluble in more concentrated brines. A similar curve could be plotted for the solubility of barium sulfate. Consider a brine that does not contain the stoichiometric combining weight ratio of strontium and sulfate as shown in Table 1 . 1 .To compare 211 the amount of strontium sulfate apparently at equilibrium in this brine with the solubility data in Tables 12.1 and 11, it is necessary to use another approach. The solubility product of a solute A, B is determined by the m molalities of the ions composing the solute and their activity coefficients:

"--"I
4

z 0 1.200I a

I
I1 )

I I 2.0 3.0 TOTAL IONIC STRENGTH

Fig. 12.7. Solubility of strontium sulfate versus ionic strength of the solution (J. Biles, written communication, Cities Service Oil Company, Tulsa, Okla., 1972).

384
TABLE 1 . 1 211

COMPATIBILITY OF OILFIELD WATERS

Composition of a brine that does not contain a stoichiometric combining weight ratio of strontium and sulfate* Ion Na+ mg/l me/l Molalit y

K+
ca+' Mg+' Ba+'

sr+'
Fe"

c1

s04-'

49,000 220 11,500 2,400 25 1,000 101 106,140 170 170,556

2,130 6 574 197 < 1 23 4 2,990 4

2.24 0.00589 0.303 0.104 0.00018 0.012 0.00189 3.15 0.00189

Total

Total ionic strength = 3.54;density at 22OC = 1,120 g/l; grams H20/1= 950.

The activity coefficients are determined primarily by the total ionic strength of the solution, and in a solution saturated with the solute A, B, :

If the total ionic strength is unchanged, Y B and [K,,/(YA~ Y B ~ ) are ~ ] constant. Therefore, the concentration of A in equilibnum with a given concentration of B in a saturated solution of A, B, is defined:

21 Plotting the data in Tables 1 . and I1 for the solubility of SrS04 in sodium chloride and synthetic brine solutions as the product of the molalities of strontium and sulfate versus total ionic strength, as shown in Fig. 12.8,indicates that the brine in Table 1 . 1 is undersaturated in SrS04 by 211 d(280 x l-) - (227 x l-) molal. This method is in error to the extent o' o' that the SrS04 solubility is affected differently by the ions in the brine in 211 21. Table 1 . 1 than by the ions in the brines shown in Table 1 . 1 Nevertheless, this approach is valuable in that a reasonable estimate can be made of the degree of undersaturation of SrS04. Now consider the advisability of mixing the brine shown in Table 1 . 1 211 with another brine which contains less dissolved solids and a comparable percentage of cations as sodium and about 1,850 mg/l of sulfate. To determine the solubility of SrS04 in various mixtures of waters, the product of the weighted average molalities of strontium and sulfate was determined for

MIXING OF SUBSURFACE WATERS

385

240

200 -

KEY
In NaCl solutions A In synthetic brines
0

3
TOTAL IONIC STRENGTH

Fig. 12.8. Solubility of strontium sulfate as a product of the molalities of strontium and sulfate versus the ionic strength of the solution (J. Biles, written communication, Cities Service Oil Company, Tulsa, Okla., 1972). Filled square shows the product of the molalities of (Sr )(SO4 ) in Table 12.111 brine.

6
t J

9
E

a w a .
0 I a 500 a
3

600

-514

g / kiloliter

I-

3 a
2
0

* 200
100

??
m

100
TABLE 1 . BRINE, 2X

percent

Fig. 12.9. Plot o the supersaturation of a mixture of the brine shown in Table 12.111 with f a brine containing 1,850 mg/l sulfate versus the Table 12.111 brine in percent (J. Biles, written communication, Cities Service Oil Company, Tulsa, Okla., 1972).

386

COMPATIBILITY OF OILFIELD WATERS

each mixture and compared with the comparable values in Tables 12.1 and I1 and Fig. 12.8. It was determined that a maximum supersaturation of 514 g/kl occurred when the mixture contained 60% of the brine shown in Table 12.111 as illustrated in Fig. 12.9. Mixtures containing less than 9%and more than 97% of the brine shown in Table 12.111 were undersaturated in &SO4 when mixed with a brine containing 1,850 mg/l of sulfate. The same error mentioned in the above paragraph will be present, but the correction would not greatly affect the value obtained using solubility data from Table 12.11. References
Akin, G.W. and Lagerwerff, J.V., 1965. Calcium carbonate equilibria in aqueous solutions 1 open t o the air, I. The solubility of calcite in relation to ionic strength; 1 . Enhanced solubility of CaC03 in the presence of Mg and SO4-. Geochim. Cosmochim. Acta, 29: 343-360. Blount, C.W., 1965. The Solubility o f Anhydrite in the Systems C a S 0 4 - H z 0 and CaS04-NaCl-H2 0 and Its Geologic Significance. Ph.D. Dissertation, University of California, Riverside, Calif., 179 pp. Blount, C.W. and Dickson, F.W., 1969. The Solubility of anhydrite (CaS04) in NaCl-HO from 100 t o 45OoC and 1 to 1000 bars, Geochim. Cosmochim. Acta, 33:227-245. Columbus, N., 1965. Viscous model study of sea water intrusion in water table aquifers. Water Resour. Res., 1:318-323. Davis, J.W. and Collins, A.G., 1971. Solubility of barium and strontium sulfates in strong electrolyte solutions. Environ. Sci TechnoL , 5:1039-1043. Ellis, A.J., 1963. The solubility of calcite in sodium chloride solutions at high tempera. tures. A m . J S c i , 261:259-267. Frear, G.L. and Johnstonb J., 1929. Solubility of calcium carbonate (calcite) in certain aqueous solutions a t 25 J. A m . Chem. SOC.,51:2082-2093. Fulford, R.S., 1968. Effects of brine concentration and pressure drop o n gypsum scaling in oil wells. J. Pet. Technol., 20:559-564. Gates, G.L. and Caraway, W.H., 1965. Oil well scale formation in waterflood operations using ocean brines, Wilmington, Calif. US.Bur. Min. Rep. Invest., No.6658, 28 pp. Glater, J., Ssutu, L. and McCutchan, J.W., 1967. Laboratory method for predicting calcium sulfate scaling thresholds, Environ. Sci Technol, 1:41-52. Glew, D.N. and Hames, D.A., 1970. Gypsum, disodium pentacalcium sulfate, and anhydrite solubilities in concentrated sodium chloride solutions, Can. J. Chem., 48:3734-3738. Henningsen, E.R., 1962. Water diagenesis in Lower Cretaceous Trinity aquifers of Central Texas. Baylor Univ. Geol. Studies, Bull., 3~38. Hubbert, M.K., 1953. Entrapment of petroleum under hydrodynamic conditions, Bull. A m . Assoc. Pet. Geol., 37:1954-2026. Kohout, F.A., 1960. Cyclic flow of salt water in the Biscayne aquifer of southeastern Florida, J. Geophys. Res., 65:2133-2141. Levorsen, A.I., 1967. Geology o f Petroleum (revised by F.A.F. Berry). W.H. Freeman, San Francisco, Calif., 724 pp. Lewis, G.N. and Randall, H.M., 1923. Thermodynamics. McGraw-Hill, New York, N.Y., 723 pp. Lucchesi, P.J. and Whitney, E.D., 1962. Solubility of strontium sulfate in water and aqueous solution of hydrogen chloride, sodium chloride, sulfuric acid and sodium sulfate by the radiotracer method. J. AppL Chem. (London), 12:277-279. Neuman, E.W., 1933. Solubility relations of barium sulfate in aqueous solutions of strong electrolytes. J. Am. Chem. SOC.,55:879-884.

REFERENCES

387

Ostroff, A.G. and Metler, A.V., 196:. Soltbility of calcium sulfate dihydrate in the system NaCl-MgC12-H20 from 28 to 70 C. J. Chem. Eng. Data, 11:346-350. Patterson, M.S. and Greene, R.C., 1965. Measurement of low energy beta-emitters in aqueous solution by liquid scintillation counting of emulsions. Anal. Chem., 37 :85 4-85 7. Popov, A.I. and Goldshteyn, R.I., 1967. Hydrologic zoning of hydrostatic systems as a mineralizing factor in the stratal cover of Central Asia. Dokl. Akad. Nauk S.S.S.R., Earth Sci Sect., 17:118-120 (transl.). Pytkowicz, R.M., Disteche, A. and Disteche, S., 1967. Calcium carbonate in sea water at in situ pressures. Earth Planet. S c i Lett., 2:430-432. Shaffer, L.H., 1967. Solubility of gypsum in sea water and sea water concentrates at temperatures from ambient t o 6 5 C. J. Chem. Eng. Data, 12:183-188. Stiff, H.A., 1952. A method for predicting the tendency of oilfield waters to deposit calcium sulfate. AIME, Pet. Trans., 195:25-28. Templeton, C.C., 1960. Solubility of barium sulfate in sodium chloride solutions from 25' to 95'C. J. Chem. Eng. Data, 5:514-516. Upson, J.E., 1966. Relationships of fresh and salty groundwater in the Northern Atlantic Coastal Plain of the United States. U.S. Geol. Surv. Prof. Paper, No.550-C, pp. 2 35-2 4 3. Vetter, O.J.G. and Phillips, R.C., 1970. Prediction of deposition of calcium sulfate scale under down-hole conditions. J. Pet. TechnoL, 22:1299-1308. Weintritt, D.J. and Cowan, J.C., 1967. Unique characteristics of barium sulfate scale deposition. J. Pet. TechnoL, 19:1381-1394.

This Page Intentionally Left Blank

Chapter 13. VALUABLE MINERALS IN OILFIELD WATERS

In the early days of the oil industry, oilfield brines were allowed t o flow by natural drainage into streams until it was noted that some of the once good fishing streams contained less fish. Fur-bearing animals had disappeared in these areas and dead trees and barren soils now bordered these same streams that once had luxurious vegetation. A few years prior t o 1935, litigation pertaining t o pollution of fresh water was taking a heavy toll from oil operators. In certair, older oil producing areas, extensive plots of ground still are barren, with no living vegetation. The litigations against oil operators combined with legislation for fresh-water protection to force better disposal techniques. At first, evaporation ponds were employed; however, usually more brine drained into fresh-water aquifers than evaporated. Until recently a widely employed practice for disposal was the dumping of oil brines into salt-water bodies when they existed nearby. This disposal method was practiced along the Gulf of Mexico and in California. Authorities in these areas insisted that oil separation be highly efficient to prevent damage to fish and oyster populations. Recently the State pollution boards have ruled that oilfield brines can no longer be dumped into surface salt-water bodies. In California excess oilfield waters are being injected into porous subsurface formations as rapidly as the injection systems can be constructed. The Plains States are not only situated in a hard water beit, but seldom have they had an overabundance of usable or surface ground waters. For this reason, State legislatures passed laws for the protection of fresh-water supplies, allowing the return of oilfield brines t o subsurface formations and allowing the repressuring or waterflood of oil properties with salt water. Subsurface brine disposal has since become the common practice. Since the laws were passed to allow subsurface disposal, more legislation has both forced such disposal and set up tight controls for it. A survey of cost data on subsurface injection in 1968 showed that subsurface disposal costs ranged from 6.6 to 19.8 cents per m3. These figures were based on operating costs plus 5-year amortization. Costs vary with the amount of treatment necessary before injection, the number of production wells per injection well, and the costs of drilling injection wells or the depth of the injection formation. The depths of disposal wells normally encountered required no injection pressure. The brines flow readily into the receiving formations under the gravity head alone.

390

VALUABLE MINERALS IN OILFIELD WATERS

Most of the 1.23 billion m3 of saline water that is produced yearly with petroleum is an expense t o oil producers even though some of these waters contain salts yielding valuable elements which might be economically recovered (Angino, 1967). Elements found in some brines in economic concentrations are magnesium, calcium, potassium, lithium, boron, bromine, and iodine. Many of them are recovered by chemical companies from sea water, salt lakes, and subsurface saline waters (Collins, 1966; Brennan, 1966). The recovery of minerals from saline waters dates back to the first time that someone precipitated a compound from a salt solution. Precipitation is the most used separation process employed in separating minerals from sea water or subsurface brines. Research continued on the separation methods which show economic promise in mineral separation from saline waters. The Office of Saline Water, U S . Department of the Interior, supports research aimed at mineral recovery processes to be integrated with freshwater plants. The object of this research is to reduce the cost of the produced fresh water by selling the extracted minerals at a profit. Now consider mineral recovery as a means of reducing the cost of oilfield brine disposal. There are additional advantages t o mineral removal other than profits from the sale of the mineral. For instance, magnesium in sea water causes great expense because of scale formation in fresh-water plants. Recovery of iodine and bromine from oilfield brines

Iodine
Iodine consumption in the United States exceeds domestic production. The Dow Chemical Company is the sole domestic producer of iodine. 75% of our domestic consumption is imported from Japan and Chile (Miller, 1965). Chilean nitrate deposits furnish most of the worlds supply of iodine. The United States and Japan obtain iodine from subsurface brines. In Michigan, Dow Chemical liberates iodine from brines by chlorination and blows t h e iodine out with air. Japan recovers iodine from brine by the cuprous iodine, electrolytic, or active carbon methods. In the United States, iodine was discovered in an oilfield brine by C.W. Jones in Louisiana in 1926. The Dow Chemical Company and Jones combined t o produce iodine from a brine well in Louisiana in 1928. At that time, iodine sold at a price between $9 and $11per kg. In 1929 General Salt Company began extracting iodine from oilfield brines in California. General Salt halted operations when Chile cut the iodine price to $3.30 per kg. In 1931 Deepwater Chemical Company began to produce iodine from oilfield brines in California. Deepwater Chemical halted recovery of iodine from brines in the late 1950s. The Dow Chemical Company moved its iodine recovery operation from Louisiana to the California oilfield brines in 1932. The move was made for two reasons. The first reason was that California brines contained 60 ppm

RECOVERY OF IODINE AND BROMINE

391

iodine as compared to Louisianas 35 ppm. Secondly, the Dow Chemical Company was producing the brine in Louisiana from its own brine wells. In California the brine was produced by oil producers, because older wells produced 1 0 m3 of brine for every cubic meter of oil. The Dow Chemical Company used two methods to obtain brine in California. The first was by paying royalties t o oil producers for brines of high quality which were delivered at one pick-up point. The second was from an extensive brine gathering system which Dow built t o collect the brines from independent producing companies. The second method of disposal was done for the producers in lieu of royalties. At one time, Dow operated three iodine recovery plants in California. Only one of the plants utilized a complete iodine recovery process. In 1961 Dow began iodine recovery from Michigan brines at Midland, Michigan. These brines are not oilfield brines and although the Michigan brines contain only 35 ppm, compared to Californias 60 ppm, Dow found the Michigan operation less costly. Oilfield brines of California have two disadvantages. First, the brine source near DOWS operation dwindled, and secondly, production costs in California rose. Several economic advantages were available in the Michigan operation. For example, the iodine recovery process was integrated with processes for the recovery of calcium chloride, magnesium hydroxide, magnesium sulfate, bromine, potassium chloride, and magnesium chloride. The Midland operation boosted iodine recovery by using brines which were heated t o 91C for other extraction processes. The absence of oil in the Michigan brines negated the cost of oil removal. In California, oil removal is necessary t o prevent interference with the oxidation step in the recovery process. The brine feed for the Midland operation is composed of brines produced from various strata in order t o obtain the desired feed for the most economical products.

Bromine
Bromine is another element that is recovered from oilfield brines. One plant that is located in Arkansas recovers bromine from the Smackover formation in the Catesville field. The bromine recovery project was originally included as part of the plans to unitize the field in 1956.
TABLE 13.1 Bromide recovery economics at Catesville Minimum economical production Designed production Designed brine feed Plant cost Plant payout period 900,000 kg/year 1,800,000 kglyear 1,400 m3/day $ 1,000,000 6 years

392

VALUABLE MINERALS IN OILFIELD WATERS

TABLE 13.11
Profitability of Catesville bromide project Return o n investment Profit as sales percent Profit per m3 processed Investment Profit per year 16.7% 14.3% $ 0.346 $ 1,100,000 $ 180,000

Location of a bromine plant at Catesville offered several important advantages such as high bromide content (up to 6,000 mg/l) of the brine, field operation under a single company (unitization), excellent rail and road facilities, low-cost fuel, and regional market outlets. Production of Smackover brine in 1956 was approximately 190 m3 daily from four oil wells. This quantity of brine was not quite economical for a bromine recovery plant. Additional pumping equipment was installed in some of the wells in order t o provide 795 m3 of brine daily for the bromine recovery project. Depleted oil wells later were employed for brine production to raise the plant feed to 1,430 m3 daily. Table 13.1 shows the initial economics associated with the bromine project as reported by Kincaid (1956). The economic data presented in Table 13.1 are based on the bromine prices of 1956. In 1956 the price of bromine was 66 cents per kg. The price fluctuates with supply and demand. The data shown in Table 13.11 were calculated by assuming that all of the bromine is sold at 70 cents per kg, that the total investment is not more than $1.1million, and that the payout time is 6 years (Cox, 1967). Minerals recovered from saline waters

Sodium chloride
Minerals are recovered from practically every type of saline water. By far the largest recovery is that of sodium chloride in solar evaporation processes. From the point of view of oilfield brine disposal, where solar evaporation is possible, the cost of disposal is small. The salts recovered, if any, would probably pay for the construction of evaporation pits.

Lithium
Lithium is produced from brines by Foote Mineral Company at Silver Peak, Nevada, and by American Potash and Chemical Corporation at Trona, California. American Potash and Chemical Corporation recovers a coproduct lithium sodium phosphate from Searles Lake, California, brines. However, the largest lithium production is from lithium ore mined in North Carolina

MINERALS RECOVERED FROM SALINE WATERS

393

by Lithium Corporation of America. Domestic production of lithium has not been reported since the mid-l950s, because individual companies do not want t o disclose confidential data. In 1954 about 36,000 metric tons were produced in the United States. The staff of the U.S. Bureau of Mines (1968) reports that both the lithium industry and the government are hampered by restrictions on publishing statistical data on the production and consumption of lithium metal, alloys, and compounds. These restrictions inhibit the determination of requirements, the evaluation of market potentialities, and the planning of future action.

Potassium
Tallmadge et al. (1964) report that the commercial recovery of potassium from brines only has been attempted on a pilot plant scale. Precipitation appears the most promising either by the addition of a selective agent specific t o potassium, or by fractional crystallization of saturated brines. Potassium compounds occur in many rocks and minerals, but the commercial sources are limited t o soluble salts in bedded salt deposits and brines. The major deposits of potassium salts in the United States are part of the Permian Salt Basin that underlies parts of Colorado, Kansas, Oklahoma, Texas, and New Mexico, and the Paradox Basin of southwestern Colorado and southeastern Utah. However, commercial beds of potassium minerals have been found only in New Mexico. Commercial operations have been limited t o about 1 4 0 km2 east of Carlsbad, New Mexico. These deposits were discovered by oil well drillers. Commercial recoveries on a limited scale are made from the brines of Searles Lake, California, and Bonneville Flats, Utah.

Rubidium
The rubidium-producing industry is very small. During 1958 rubidium production in the United States was only about 100 kg annually, and during that year some new technical-grade rubidium compounds were prepared from alkali carbonate residues of lithium operations. As with many other minerals found in oilfield brines, the production of rubidium is not published because it is withheld as confidential company data. However, with current accumulated stocks and a very small consumption, it is doubtful that the recovery of rubidium from brines would be economical even at $935 per kilogram.

Cesium
Cesium, both as a metal and as an industry, is similar to rubidium. The demand for both is small, and the known uses are few. Both cesium and rubidium are obtained commercially from lepidolite, a lithium mineral.

394

VALUABLE MINERALS IN OILFIELD WATERS

Cesium and rubidium are byproducts of the lithium industry, and both are recovered from the residues of the lithium production process are precipitation from solution. The high concentrations of cesium and rubidium in the residues and the fact that the amount therein greatly exceeds demand virtually preclude their removal from oilfield brines on a competitive basis.

Magnesium
Magnesium comprised one-third of the value which Collins (1966) attributed to the minerals wasted by oilfield brine disposal, and the price used was that of magnesium metal. In the primary.meta1 form, magnesium commands its highest price. When magnesium is sold as contained in other compounds, its value is less than 2 cents per kg as compared to 77 cents per kg for primary magnesium. Magnesium and magnesium compounds are produced from the following four raw material sources: (1) sea water; (2) dolomite; (3) ores other than dolomite; and (4) evaporite deposits and lake and well brines. In 1963, well brines, bitterns, and sea water combined with calcined dolomite or lime accounted for more than half of the domestic production of magnesium compounds used as chemicals, filters or bases in many industrial products including basic refractories. Magnesium and magnesium compounds are produced and recovered by several companies in the United States. The Dow Chemical Company produces magnesium chloride crystals, magnesium chloride fluxes, and magnesium hydroxide from well brines and calcined dolomite at Ludington, Michigan. The Michigan Chemical Company produces precipitated magnesium carbonate, magnesium hydroxide, and magnesium oxide from well brines and calcined dolomite at St. Louis, Missouri. The Dow Chemical Company produces magnesium chloride, caustic-calcined magnesia, and magnesium hydroxide from sea water and oyster shells at Freeport, Texas. Magnesium compounds are recovered from solution by precipitation of magnesium hydroxide. This method is so economical that a large part of the production of magnesium and magnesium compounds in the United States is derived from sea water. The Dow Chemical Company is the major source of primary magnesium and in 1963 it had a capacity at Freeport, Texas, of 50,000 metric tons per year; at Velasco, Texas, the capacity was 34,000 metric tons. In 1963 the U S . production of primary magnesium was 69,000 metric tons. The Dow process for magnesium recovery from sea water first precipitates magnesium hydroxide. The hydroxide source is calcium hydroxide made from oyster shells. After settling and thickening, a slurry of 17%magnesium hydroxide is attained and neutralized with hydrochloric acid to form a 15% solution of magnesium chloride. After evaporation and dehydration, the resultant 48% magnesium chloride solution is mixed with dried magnesium chloride t o form a paste. The paste is dried t o granules which consist of 74% magnesium chloride, and the granular material is the feed to electrolytic cells

MINERALS RECOVERED FROM SALINE WATERS

395

which produce magnesium metal and chlorine. The chlorine is then converted t o hydrochloric acid which is used in the neutralization step. Shreve (1956) lists three economic factors of importance in the precipitation of magnesium hydroxide. They are: (1) the source of the hydroxide; (2)the dewaterirg procedures used for removal of the magnesium hydroxide from the dilute solution; and (3) the purification of precipitates. The source of the hydroxide is the major economic deterrent factor against the increased use of well brines. Sea water provides the magnesium, and the sea also furnishes the oyster shells for calcium hydroxide production. For well brine feed, dolomite often is employed, and a large source of dolomite must be economically available. Tallmadge et al. (1964) report that waste sodium hydroxide has been tested in Japan, but in most areas, calcium salts are the least expensive sources of the hydroxide. Thus the choice of a raw material must be belanced in cost against plant size and market. While Michigan brines contain four t o five times the magnesium concentration of sea water, the reduced size in necessary equipment for processing the brine does not completely overcome the cost of producing and disposing of the brine. This would appear t o make oilfield brines more attractive than other subsurface brines if a hydroxide source is available at an equivalent expense. Tallmadge et al. (1964) report methods for extracting magnesium from brines by methods other than hydroxide precipitation. However, none appear economically attractive when compared to precipitation unless combined with other processes or products. Among those studied are solar evaporation to produce chloride, use of ion-exchange resins with lime and carbon dioxide or waste liquor from the ammonia-soda process, and electrolysis.

Calcium
Calcium production from brines does not appear economical when compared to the source of the worlds calcium consumption. The largest amount of calcium is produced by the mining of mineral deposits (notably gypsum) found extensively throughout the world. Proposals for methods t o recover calcium from brines have been made and are under study, but to compete commercially beyond extremely small, local demands, considerable research is needed.

Mixed salts
Mixed salts are precipitated by evaporation of sea water and brines, producing crude separations. The costs of these separations are low compared t o those of highly purified compounds or metals. There are several drawbacks which prevent greater use of this type of recovery. The product does not command a high price, the plant must be at the brine sburce, there must be solar evaporation conditions, and a local market must exist for the majority

396

VALUABLE MINERALS IN OILFIELD WATERS

of the mixed salts. Uses which have been suggested include heat-treating salt baths in the steel industry, raw materials for refractory or catalyst manufacture, and fertilizer components. Precipitation other than by solar evaporation is accomplished by cooling or adding chemical agents. Simple cooling may be all that is necessary for more concentrated brines, but fractional crystallization is necessary for dilute brines such as sea water. Again local markets dictate whether cooling or freezing processes will yield the correct products for a particular area. Adding chemicals t o precipitate a specific product is the most fruitful of the nonsolar evaporation processes. Most of the processes have been aimed at the production of fertilizer. Potassium and magnesium are the minerals in sea water that are most valuable for use in fertilizers. Salutsky and Dunseth (1962) report that metal ammonium phosphates (MAP) containing magnesium, calcium, iron, manganese, copper, and many other trace metals comprise a high-analysis fertilizer. The production of metal ammonium phosphates (MAP) in the United States was started by W.R. Grace and Company in 1960 on a semicommercial scale. The method which Grace used to produce MAP was not disclosed until 1962 after it was patented. The fertilizers are nonburning, long-lasting sources of nitrogen, phosphorus, and various trace metals. Because of their low solubility, MAPs will not cause salt injury t o seeds or plants. In magnesium ammonium phosphate, practically all of the P z 0 5 is available, and the size of the MAP granules applied to plants determines how long the nutrients will be available. Thus, availability of nutrients can be controlled by granulation and, since growing time varies from crop t o crop, MAPS can be tailored to a specific crop (Anonymous, 1961). Therefore, fewer applications are necessary with MAPS than with fertilizers of higher solubility and high nitrification rates. W.R.Grace and Company developed the MAP process for two purposes. First, it is useful t o remove scale-forming materials from sea water before desalination. Secondly, it would yield the valuable, high-analysis fertilizer, magnesium ammonium phosphate. In 1962, W.R. Grace and Company (Anonymous, 1962) reported that the process was ready for the pilot plant. The process is based on phosphate precipitation. To descale sea water and produce high-analysis fertilizer at the same time, wet-process phosphoric acid and anhydrous ammonia are added continuously t o raw sea water. This precipitates the scale-forming elements - calcium, magnesium, iron, and other metals - as metal ammonium phosphates and other phosphates. The precipitated solids are removed by settling, and the descaled sea water is pumped to the saline water conversion plant. The descaled water holds only 1% the original magnesium and 5% of the original calcium. The slurry of of MAPs is dewatered t o about 35--40% solids by continuous centrifuges and thin it is heated t o 90C. This converts MAP hexahydrate t o monohydrate. The slurry is filtered, washed, mixed with recycle fines, and granulated. Fig.

MINERALS RECOVERED FROM SALINE WATERS

397

S e t t l i n g ond thickeninq

Descoled sea w o t e k

1
Dehydration

1
Wash Filtrotion ond wos hing Gronulotion

Drying

Crushing

1
Undersize Screeninq Oversize

Finished product to storoge

Fig. 13.1. Diagramatic flowsheet for producing descaled sea water and fertilizer.

13.1 shows a process flowsheet for producing descaled sea water and fertilizer. Several questions surround the economics of the process. For a plant descaling 3,800 m3 of sea water per day (output: about 10,000 metric tons per year of fertilizer), the fertilizer would have to command a price higher than that of conventional farm fertilizers. The estimate assumes 1962 market prices for raw materials (phosphoric acid and ammonia) and does not take credit for the increased value of the descaled sea water. The cost is just about the same for Graces present method of producing MAPS. Because of its premium quality, MAP can go t o the market as a specialty product. In the phosphoric acid-ammonia process, 2 moles of ammonia per mole of MAP are lost. to ammonium chloride in neutralizing the phosphoric acid. Using disodium phosphate in place of the acid loses no ammonia, and using monosodium phosphate only loses 1mole of ammonia. If a cheap method were developed for producing the sodium phosphates, ammonia waste would be reduced. The simplest method for producing sodium phosphates involves

398

VALUABLE MINERALS IN OILFIELD WATERS

the neutralization of phosphoric acid with either dilute sodium hydroxide or soda ash. Caustic soda and chlorine can both be produced from sodium chloride brines.

Chlorine
Chlorine is the most abundant element in oilfield brines, and the removal of all chlorides from oilfield brines would virtually desalt the brine. However, present commercial methods for the extraction of chlorine from brines requires evaporation to acquire a saturated brine. Solar evaporation first produces the saturated brine, then electrolytic processes are employed to generate chlorine gas. The largest production of chlorides from brines is from solar evaporation of sea water and salt lake brines. Many oilfield brines are very close t o saturation with sodium chloride at surface temperatures. Subsurface brines are employed as raw materials for chlorine production, but the amount produced is not significant when compared t o surface brine production. Shreve (1956)describes the methods still in use for chlorine production. Caustic soda and chlorine are coproducts in the electrolytic process. Purification of the brine is necessary to produce a purer caustic soda and lessen clogging of the cell diaphragm with consequent increased voltage demand. Calcium, iron, magnesium, and sulfate must be removed. The higher concentrations of magnesium and calcium in oilfield brines cause greater expense in brine purification unless the removed compounds can be sold. Hydrogen, caustic soda, and chlorine are the products of this type of recovery. The hydrogen presents a disposal problem and is frequently made into other compounds such as hydrochloric acid or ammonia or is employed for hydrogenation of organic compounds.

Iodine and bromine


Iodine and bromine are two minerals which have closely related processes for recovery from brines. Both are displaced from ions to elements in solution by chlorination and then stripped from solution by air. When bromine is liberated by chlorination, iodine is oxidized to the iodate ion. After bromine is stripped from solution, the iodate ion can be reduced t o free iodine by treatment with ferrous chloride and then stripped by air as was the bromine. The greater portion of bromine production in the United States is from well brines. Slightly over 50% of the domestic production is from well brines, 35% comes from sea water, and the remainder comes from oil well brines and saline lake brines. Substantial expansion recently completed by two producers of bromine from brines should give the industry sufficient capacity for several years t o supply the expected increase in markets (Miller, 1965). The 2-million-kilogram-per-year plant designed for oilfield brines in Arkansas by Michigan Chemical represented 50% of the domestic production

MINERALS RECOVERED FROM SALINE WATERS

399

at that time. In 1963 the domestic production of bromine was about 12 million kg. The recovery of bromine from brines and sea water is accomplished by displacement of the bromide ion with chlorine. The resulting free bromine dissolved in water is then stripped from the solution with air and recovered. A large number of modifications t o the basic process (developed by the Dow Chemical Company) have been proposed and patented from time to time. Tallmadge et al. (1964) report that studies are being carried out on the effect of pH, temperature, organic impurities, chlorine concentration, and foreign ion concentration on the displacement reaction between the bromide ion and chlorine gas. The use of chlorine water rather than gas for the displacement step has been suggested t o reduce the adverse effects of magnesium and calcium interference where more concentrated brines are used. Activated carbon with adsorbed chlorine also has been proposed as a means of carrying out the displacement step. Iodine production in the United States uses oilfield brines and sub-surface brines exclusively. Dow Chemical is the sole domestic producer of crude iodine. Dow extracts iodine from oilfield brines in California and from deepwell brines in Michigan. Roughly 1.1 million kg of crude iodine were imported from Chile in 1963 and about 0.5 million kg were imported from Japan. Japanese production of iodine is almost exclusively from deepwell brines, while Chilean production is from nitrate deposits containing the minerals lautarite and dietzite. Miller (1965) reports that Chilean reserves are in excess of 1 billion tons as a byproduct of the nitrate minerals industry. It was price cuts by Chile that forced all domestic producers except Dow out of iodine production; however, the recent nationalization of the Chilean mines has changed the picture and pushed the price of iodine to about $5.06 per kg. Multiproduct production Tallmadge et al. (1964) report that it is very probable that the most economic system for removal of minerals from sea water may involve two or more recovery steps in some integrated fashion. Only a few multiproduct processes are operated on a commercial scale. The Dow plant in Midland, Michigan, is such a plant. The first step in studying multiproduct processes is t o determine how much of each product can be sold. The second step is to determine the engineering design and production costs for such a plant. Angino (1967) has pointed out that several processes exist that can be set up t o recover elements from petroleum-associated waters. These existent methods and new methods should be utilized and developed to conserve the natural resources dissolved in brines and to aid in the abatement of soil and fresh water pollution. The recovery methods used may include desalination (Christensen et al., 1967), ion exchange (Klein et a1.,'1968), and ion exchange plus other methods (George et al., 1967; Waters and Salutsky, 1968).

400

VALUABLE MINERALS IN OILFIELD WATERS

Fig. 13.2. How gas, oil, and brine are separated after production from subsurface strata.

Fig. 13.2 illustrates a wellhead through which gas, oil, and water are produced from a subsurface formation. Often they are produced as a mixture and it is necessary t o separate them in a tank such as that illustrated and sometimes referred to as a gunbarrel. In this tank the water or brine will settle to the bottom with the oil interface forming over the brine, and the gas will rise to the top. The gas is drawn off the top, the oil is pumped off the top of the water and stored in an oil tank, and the water is siphoned from the bottom into a skimming tank for further oil-water separation. The water is siphoned from the bottom of the skimming tank into a settling pond where additional oil-water separation occurs. The brine or water could be pumped from the settling pond t o a chemical plant for recovery of valuable elements or to an injection well. Fig. 13.3 illustrates a possible scheme for recovery of some elements plus fresh water from the brine. For example, the raw brine could be concentrated by a desalination process which would also yield fresh water. Sulfate then could be taken from the concentrated brine and used t o produce sulfur or sulfur compounds. Next, iodine could be recovered, then bromine, followed by calcium, sodium chloride, and magnesium as suggested in the figure. The remaining sludge could be dried and disposed of as a solid or it could be recycled for additional recovery of elements.

FRESH-WATER PRODUCTION
RAW B R I N E

401

I I I
I

1
DESALINATION PROCESS CONCENTRATED BRINE

I-*

PRODUCT F R E S H WATER

SULFATE PRECIPITATION

I-*

PRODUCT SULFUR AND SULFUR COMPOUNDS PRODUCT I O D I N E AND I O D I N E COMPOUNDS PRODUCT BROMINE AND BROMINE COMPOUNDS

IODINE RECOVERY

t .
BROHINE RECOVERY

PRODUCT CALCIUM

coMp0uM)s

SODIUM CHLORIDE RECOVERY

PRODUCT SODIUM CHLORIDE OR SODA ASH AND CHLORINE PRODUCT MAGNESIUM AND MAGNESIUM COMPOUNDS

- l -

SLUDGE CONTAINING CALCIUM, STRONTIUM, BARIUM, MAGNESIUM AND OTHER ELEMENT COHPOUNDS. THE SLUDGE CAN BE D I S P O S E D A S A S O L I D OR RECYCLED FOR CHEMICAL RECOVERY

Fig. 13.3. Diagramatic flowsheet for producing fresh water and valuable elements from brines.

Fresh-water production Dwindling fresh water supplies and polluted supplies have increased research on how t o best obtain fresh water from saline water. Several plants throughout the world produce fresh water from sea water. The price of water for municipal purposes is a highly specific thing. The availability of fresh water and costs of obtaining it vary from place to place. Conventional

402

VALUABLE MINERALS IN OILFIELD WATERS

water supplies range in cost from a few dollars per 1,000 m3 to over $260 per 1,000 m3. The average cost of conventional water supplies in the United States was $100 per 1,000 m3 in 1952. This was chosen as the goal for saline water conversion costs. Several authors have estimated ultimate costs of saline water conversion based on thermodynamic considerations. Dodge and Eshaya (1960) have examined the minimum expected costs for saline water conversion. Prior t o their calculations, other authors reported sea water conversion costs to be ultimately less than $79 per 1,000 m3. Dodge and Eshaya expanded earlier work t o look at departure from isothermal operation, finite
product recovery, differential as opposed to single stage operation, and salt

concentration in the feed. They found that $90 dollars per 1,000 m3 is the smallest cost for desalination of sea water. Consider the case for converting brackish water with 5,000 ppm sodium chloride. For converting 50% of the feed t o fresh water, 187 kWhr per 1,000 m3 was the power requirement. For 35,000 pprn sea-water conversion, the power requirement was 1,530 kWhr per 1,000 m3. Both calculations were for 50% recovery of fresh water from feed, where the average power costs used in determining conversion costs are 1.5 cents per kWhr. At this rate, the difference in power costs for sea water over brackish water is $20 dollars per 1,000 m3. Oilfield brines contain up to seven times the concentrations of dissolved salts compared with sea water. Would the power be seven times again as expensive per 1,000 m3? At over $132 for power and $92 for other costs, the cost of obtaining fresh water from oilfield brines probably would be prohibitive when consideration is given to the other sources for feed t o a conversion plant in the same area. An additional factor is that most oilfield brines with their high concentrations are nearly saturated. Removing 50% of the water would in essence leave a precipitated salt. Therefore, since no conversion processes under study deal with saturated brine effluents, it is not technologically feasible to completely desalt oilfield brines at this time. Preliminary economic evaluation The brine refinery concept (Collins, 1966) yields a processing plant the size of a large petroleum refinery. The market prices used were for the recovery and sale of the pure elements. The $3 billion in sales from 0.95 billion m3 of brine is the highest sales income possible that would result from recovering and selling the minerals in the form that gives the highest unit price. Consider what is probably the best case of a brine refinery, a system that would gather 22 million m3 per year. The cost of gathering and disposing of this brine would be approximately 9.4 cents per m3. The question is whether or not minerals could be sold at a profit such that the disposal expense would be negated or a profit made. First, the minerals t o be

PRELIMINARY ECONOMIC EVALUATION

403

sold must be determined. At 7 ppm lithium, 163 metric tons per year could be produced. This is a large fraction of present consumption and probably would depress the sale price. The same holds true for most other elements of such a refinery. The assumptions lead to a brine refinery that would process 22 million m3 per year and sell $35 million of minerals. Assuming a 15% return on investment and a profit of 15% of sales, the plant would require $35 millioninvestment and yield 23.6 cents per m3 of brine processed. The original disposal operation without mineral recovery was such that only $6 million was invested. The brine refinery would turn brine disposal into a profit. But the new investment is six times that for disposal only. It is doubtful that any large oil producer would be interested in a 15%return on investment, and small ones would never gather the cash. Would a chemical company be interested in such an operation since they operate at about a 15% return? Companies that currently remove minerals from brines use brines that are more concentrated in the minerals desired. It is doubtful that a process could combine several less economical operations into a more economical one, and this would probably be true even if the brine was supplied t o a chemical company free of charge. Only in the special case where an oilfield brine contained a concentration very near t o a brine that would be the most economical for separation would the oilfield brine be a best alternate. Therefore, a tax incentive for pollution abatement or some other economic incentive such as price increase of recovered chemicals is necessary.

Other economic factors


Table 13.111 illustrates the approximate amount of valuable chemicals per 1 million kg of brine produced from a given depth should contain before it can be considered of economic value at present market conditions. The values shown in Table 13.111 should allow a profit if conventional or better recovery operations are utilized. The marketed end product will influence the selection of the recovery operation as well as the delivered price. The

TABLE 13.111
Dollar value of dissolved chemicals a brine should contain per million kg of brine produced from a given depth Value of dissolved chemicals
$ 462 $ 968 $ 1,430

Depth of well (m)

760 2,130 3,050

404 TABLE 13.W

VALUABLE MINERALS IN OILFIELD WATERS

Amount of element necessary in 1 million kg of brine to produce a chemical worth $ 550 at the market Element in the brine Sodium Potassium Lithium Magnesium Calcium Strontium Boron Bromide Iodide Sulfur Concentration of element (ppm/106 kg of brine)
50,000 14,000 170 8,000 11,000 4,000 1,400 1,700 250 5,300

Market product sodium chloride potassium chloride lithium chloride magnesium chloride strontium chloride strontium chloride sodium borate bromine iodine sodium sulfate

price information used to make the approximations was taken from the U.S. Bureau of Mines (1968). Factors that must be considered in evaluating a saline water as an economic ore are the cost of bringing it t o the factory, the cost of the recovery process, and the cost of transporting the recovered products to market. Assuming that a brine is produced only for the purpose of recovering its il dissolved chemicals, a prime factor is the cost of pumping the brine. It wl cost less to produce the brine from a shallow well than from a deep well. Therefore, neglecting other factors, a brine must contain a certain amount of recoverable chemicals before it can be considered economically valuable, and the farther it must be pumped, the more chemicals it must contain. Today the possibility of recovering elements from brines that are pumped t o the surface is increasingly important because the brines present a pollution hazard if their disposal is improper. Consider the fact that 1 m3 of brine containing 100,000 ppm of chloride is capable of polluting 400 m3 of fresh water so that they are unfit for human consumption. Table 13.IV illustrates the value that chemicals recovered from brines have at the market; however, because the market fluctuates, these values are approximate. The column on the left indicates the elements that are found in petroleum-associated brines, and the second column indicates the concentration that a given brine must contain before it can be used to produce a given amount of chemical. For example, a brine containing 50,000 ppm of sodium will contain sufficient sodium in 1 million kg of brine to produce sodium chloride worth about $550. The data in Table 13.IV indicate that some petroleum-associated waters contain sufficient sodium to establish them as economic for the production of sodium chloride. This is not necessarily true, because factors such as

PRELIMINARY ECONOMIC EVALUATION

405

market demand, ease of recovery, and proximity to market may be discouraging in certain geographic areas. Such factors must be fully considered before startup of a chemical from brine recovery operation. One important goal that should not be discounted nor overlooked is developing a means of ultimately disposing of these brines so that they are not a pollution hazard. Coupling of this goal with the fact that many of these brines contain economic concentrations of several elements should make such recovery operations more attractive. Additionally, several important chemicals can be produced from these elements instead of those shown in the market product column in Table 13.IV. An example is soda ash, which is a basic chemical in many manufacturing processes. Furthermore, the figures shown in Table 13.111 are applicable only if the brine is produced solely for the recovery of its dissolved chemicals. If the brines are pooled from several petroleum production operations, the cost of pumping the brine becomes less, and the necessary amounts of chemicals dissolved in 454,000 kg of brine become less. At the present time, many petroleum-associated brines are injected into subsurface strata, and it is assumed that they are thus disposed of permanently (Crouch, 1964). However, this method of disposal appears subject to question, because in some instances, fresh waters apparently have been polluted by disposal of brines. Subsurface disposal operations are suspected in certain areas as possibly contributing to increased earthquakes and ground tremors (Evans, 1966; Bardwell, 1966). The storage of brines in earthen pits is known to cause pollution of nearby soils and streams. Such ponds which have been abandoned for 10 years still contribute to soil pollution (Bryson et al., 1966). Sound conservation should favor the recovery of valuable elements from brines, and with proper planning, the recovery processes should aid in the ultimate disposal of unwanted brines. Conservation of this type not only will develop new resources, but will benefit the oil producer and the national economy and will aid in abating pollution of soils, potable waters, and streams.

Work necessary f o r an exact preliminary evaluation


Aries (1954) spells out the marketing research techniques employed in the chemical industry. There are ways to quickly determine where a market for a product is. Usually these places are currently served by some producer or another. If the competition is located far from the market, then an evaluation of a closer area source is readily made. To find product users, the following methods and approaches are utilized: advertising, company analysis, product analysis, industry analysis, use analysis, and other miscellaneous methods. If new markets must be found, the following types of work are utilized: personal interview, questionnaire, trade analysis, company records, and published sources. Before an economic analysis can be made for a given area, probably several man-months of the listed methods would be

406

VALUABLE MINERALS IN OILFIELD WATERS

required. The product of this type of market study would be a list of elements and compounds that could be sold from a given place. The quantities and prices obtgined would then allow an economic calculation of the production costs. With the quantities, prices, and production costs in hand, it is still not a simple matter t o determine what type of plant t o operate. Regardless of whom the investor might be, he will want to know what return on investment he will get, what risk is involved, and what payout period exists for the project. Depending on the investor, he may want t o limit the plant size by the amount of money he can invest. This does not simply scale down the plant. It may rearrange various ratios of certain products produced in order to give the investor the combination of profits, return on investment, risk, and actual size of investment that he desires.

Locations o f valuable brines


Table 13.V lists the approximate geographic locations where subsurface saline waters containing valuable elements are found. The numbers in the left column of the table correspond to the numbered arrow locations on Fig. 13.4. The second column in the Table indicates the age of the geologic strata from which the waters were obtained. These waters are in or near oilproductive sedimentary basins. Concentrations of various elements present in the waters are given in columns 3 through 1 of Table 13.V. These concen1 trations are representative of one or more subsurface waters from each location; however, the concentrations should not be considered typical of all subsurface waters in an area or stratum. For example, some waters near location 1 from Mississippian age strata may contain 1,000 ppm of bromine, while other waters 80 km away, but from the same geologic strata, may contain 3,600 ppm of bromine. The elemental composition of ocean water is consistent; the composition of subsurface saline waters is inconsistent. Fig. 13.5 is a map showing some areas in the United States where brines containing high concentrations of sodium are found. The solid circles on the figure represent areas where brines containing 75,000-80,000 mg/l of sodium can be found, the open circle represents brines containing 80,000-95,000 mg/l, and the triangle represents brines containing more than 95,000 mg/l. Fig. 13.6 is a map showing some areas in the United States where brines containing high concentrations of calcium are found. On this figure the solid circle represents brines containing 20,000-30,000 mg/l of calcium, the open circle 30,000-50,000 mg/l, and the triangle more than 50,000 mg/l. Fig. 13.7 is a map illustrating some of the areas in the United States where high concentrations of magnesium in brines are found. On this figure the solid circle indicates brines containing 5,000-10,000 mg/l of magnesium, the open circle 10,000-30,000 mg/l, and the triangle more than 30,000 mg/l.

PRELIMINARY ECONOMIC EVALUATION

l l
c

TABLE 13.V Geographic location, geologic age of saline water-bearing strata, and concentration of some of the elements found in the brine's Location* Age of subsurface strata 1 2 3 4 5 6 7 8 9 10 1 1 12 13 Concentration (ppm) lithium sodium potassium magnesium
40 28,000 100 1,000 55,000 2,500 66,000 10,000 8,000 ND 51,000 100 68,000 4,000 73,000 600 74,000 700 14,000 8,000 16,000 9,000 58,000 3,000 72,000 2,000 10,000 25,000 9,000 5,000 11,000 600 5,000 6,000 5,000 15,000 11,000 5,000 4,000

calcium
60,000 100 30,000 40,000 20,000 10,000 30,000 30,000 30,000 70,000 14,000 20,000 35,000

strontium boron
3,000 5 ND 2.000 ND 1,000 ND ND 900 1,500 800 1,000 ND 40 ND 90 ND ND 10 ND 60 ND 300 ND ND ND

sulfur
400 20,000 400 100 350 30 60 60 600 400 20 60 4

chloride
179,000

bromide
3,200 ND 1,200 700 ND 600 5,000 200 2,000 2,500 1,500 1,300 1,800

iodide
40 ND 25 20 ND 1,000 10 20 40 40 40 30 20

Mississippian 10 Permian 30 Permian 40 Devonian 100 Cambro-Ordovician ND Pennsylvanian 5 Jurassic 100 Miocene 15 Devonian 25 Devonian 60 * Mississippian Devonian 10 Devonian 90

9,000
166,000 198,000 79,000 98,000 175,000 ia4,ooo 182,000 200,000 90,000 143,000 186,000

See arrows in Fig. 13.4. ND = not determined.

PRELIMINARY ECONOMIC EVALUATION

409

Fig. 13.5. Approximate geographic locations of brines containing high concentrations of sodium.

Fig. 13.6. Approximate geographic locations of brines containing high concentrations of calcium.

410

VALUABLE MINERALS IN OILFIELD WATERS

'

LEGEND
0

5,000-1 , 0 mg/l 000 10,000-30,000 > 30,000

Fig. 13.7. Approximate geographic locations of brines containing high concentrations of magnesium.

Fig. 13.8. Approximate geographic locations of brines containing high concentrations of bromine.

DISPOSAL BRINES

411

Fig. 13.8 is a map illustrating some of the areas in the United States where high concentrations of bromide in brines are located. On this figure the solid circle represents brines containing 1,500-2,000 mg/l of bromide, the open circle 2,000-3,000 mg/l, and the triangle more than 3,000 mg/l. Disposal brines In an attempt to acquire as much information as possible about salt water disposal facilities in the various States, the literature was surveyed and State and Federal agencies and oil companies were contacted. The acquisition of a true compilation of the exact number of disposal facilities, disposal wells, and total number of barrels of brine injected was an impossible task without a large surveillance force; however, the data in Table 13.VI are reasonably representative. To determine the value of the minerals in the brines flowing into these disposal systems, samples were obtained from 40 systems. The samples were analyzed for concentrations of lithium, sodium, potassium, magnesium, calcium, boron, ammonium, sulfate, bicarbonate, chloride, bromide, and iodide. Table 13.VII lists the state, county, subsurface formation, and dissolved solids (DS) from which the brine samples were obtained. The specific gravity of each sample plus the ionic values determined in the laboratory also are given in Table 13.VII. Sample 1is sea water, sample 2 is a brine that contains a high concentration of iodide, and sample 3 is a brine from which bromine currently is extracted.

Worth and value estimates


The estimates of the value of a brine are related to the market and the recovery process. The market is dependent upon demand; however, in the following estimates the demand was not considered. According to Christensen et al. (1967): Brine value = market value of products - operating costs exclusive of brines value - fixed costs - profit The maximum value of a brine can be found by letting the return on investment equal zero, and the brine worth is: Brine worth = brine value + profit = market value of products ating costs - fixed costs

- oper-

The brine value is less than the brine worth by the amount of profit expected. Although the information necessary to obtahi an accurate calcula-

TABLE 13.VI States where oilfield brines are disposed, total number of subsurface salt water disposal wells (SWDW) and largest disposal facilities State Ah. Alaska Ariz. Ark. Calif. Colo. Fla. Ill. Ind. Kans. KY. La. Mich. Miss. Mont. Nebr. Nev. N. M. N. Y. N. D. Ohio Okla. Pa. S. D. Texas Utah W.Va. wyo. Total number of SWDW Largest SWDW
( m3 /day)

SWDW

SWDW

Remarks

2 4700 m3/day

2 1590 m3 /day

14* no data 4 421 216 25* 5 no data 325 3,150 60** 1,304 52 2 370 20 0 3 300

2,385 191 9,221 25,120 2,544 954 159 3,180 11 1


318 1,113 239 135
0

5
0 0 0 0 0 0 0 0 0 0 0 0
0 0

0 14
0

+ bromine plant effluent

0 2** 0
0 0

average SWDW is 21 7 m3 /day average SWDW is 32 m3 /day

2,703
318 318

62 28* 4,900

0 0 0 4 0 0 0
0 0

only 7 producing wells for State

2
most of this is by ponding ponding used number of SWDW permitted from 1950-1971

143 159

2 P

7,173** 3** 95** 19*


0 0
0 0

_.

2 --

* Salt water disposal Systems that may have more than one salt water disposal well per system. * * Approximate.

TABLE 13.VII Analyses of some disposal brines, seawater, and a proven economic brine* Brine State County Formation Sp. gr. Constituents (mg/l) Na
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42

Ca
400 8,300 44,440 2,480 1,450 760 2,630 1,200 4,300 3,500 3,800 3,800 34,500 38,500 36,300 37,300 27,600 1,500 1.850 2,840 2,400 1,970 8,650 10,320 3.300 10;530 840 6,750 630 5,630 6.750 8,860 1,650 3,960 2,335 1,855 70 1,520 10,120 13,270 50

Mg
1,350 260 4,340 700 490 260 690 320 1,300 900 1,030 930 3,950 3,850 4,040 3,895 1,315 500 500 340 770 410 365 345 1,130 690 110 205 970 40 480 550 680 600 230 135 780 5 50 1,640 2,460 10

K
380 180 4,410 260 75 70 190 105 160 200 160 140 1,845 1,945 1,370 2,000 3,500 245 370 20 150 560 400 105 790 860 550 360 250 50 400 320 460 310 300 200 200 15 20 1.000 980 10

Li
0.17 14 370 20 3 3 10 5 5 5 5 5 160 180 170 165 230 5 10 2 10 10 5 2 35 2 5 2 10 2 5 10 10 5 5 5 1
1

B
4.6 18 200 10 10 0 5 3 12 10 16 12 140 150 140 140 160 10 0 0 40 5 5 40 10 0 5 10 20 10 0 12 30 25 10 12 5 0 0 40 0 0

NH4
-

CI
19.000 98,300 202,050 32,850 19,460 10,380 30,500 17,YOO 45,100 42,200 43,100 42,400 178,100 180,800 197,600 182,600 150,000 17,800 32,650 24,300 42,600 37,240 22,400 49,100 135,500 61,300 57,100 31,800 68,700 8,350 29,400 37,000 47,540 62,500 74,600 68,900 28,870 330 11,600 115,500 138,600 3,861

Br
65 1,500 5,725 60 50 20 50 50 150 500 400 600 2,450 2,340 4,800 3,390 3,500 50 30 25 210 40 40 270 210 400 370 180 100 70 40 50 30 70 25 30 15 2 20 326 540 3

I
0.06 1,300 15 10 5 2 2 3 10 10 10 10 5 5 5 10 5 3 0 0 5 2 0 30 30 35 30 35 25 5 10 10 12 20 20 20 20 0 12 150 10 0

SO4
3,468 180 220 2,000 2,350 1,400 2,880 1,100 2,270
0 0 0

HCOi
140 50 95 450 350 60 315 250 260 170 160 60 100 190 200 600 200 1,000 500 600 380 490 590 400 0 300 400 450 300 500 190 160 140 380 140 300 240 130 170 120 85 255

DS
35,308 166,652 335,865 53,290 34,123 18,855 54,072 30,332 76,895 67,439 69,819 68,495 286,420 292,560 345,235 295,955 241,250 32,329 56,428 42,687 69,055 62,137 38,865 89,232 207,045 103.1 57 103,200 55,107 112.185 15,417 49,135 59,742 76,652 101,410 124,115 114,549 45,616 803 21,165 187,096 225,365 7,479

Seawater Okla. Ark. Kans Kans. Kans. Kans. Kans Kans. Ark. Ark. Ark. Ark. Ark. Ark. Ark. Ark. N.M. N.M.

N.M. N.M.
N.M. Texas Texas Texas Texas Texas Texas Texas Texas Ala. Ala. Ala. La. La. La. Calif. Calif. Ariz. Okla. Okla. Miss.

1.025 Kin