01 141 162final

The Florida Institute of Phosphate Research was created in 1978 by the Florida Legislature (Chapter
378.101, Florida Statutes) and empowered to conduct research supportive to the responsible
development of the state’s phosphate resources. The Institute has targeted areas of research
responsibility. These are: reclamation alternatives in mining and processing, including wetlands
reclamation, phosphogypsum storage areas and phosphatic clay containment areas; methods for more
efficient, economical and environmentally balanced phosphate recovery and processing; disposal and
utilization of phosphatic clay; and environmental effects involving the health and welfare of the people,
including those effects related to radiation and water consumption.
FIPR is located in Polk County, in the heart of the central Florida phosphate district. The Institute
seeksto serve as an information center on phosphate-related topics and welcomes information requests
made in person, or by mail, email, or telephone.
Research Staff
Executive Director
Paul R. Clifford
Research Directors
G. Michael Lloyd, Jr. -Chemical Processing

J. Patrick Zhang -Mining & Beneficiation
Steven G. Richardson -Reclamation
Gordon D. Nifong -Environmental Services
Florida Institute of Phosphate Research

1855 West Main Street
Bartow, Florida 33830
(863) 534-7160
Fax: (863) 534-7165
http://www.fipr.state.fl.us
COST-EFFECTIVE REAGENTS AS DEFOAMERS
AND CRYSTAL MODIFIERS TO ENHANCE THE FILTRATION OF
PHOSPHOGYPSUM
FINAL REPORT
Hassan El-Shall, Brij M. Moudgil and

El-Sayed A. Abdel-Aal
Principal Investigators
Department of Materials Science and Engineering

Engineering Research Center for Particle Science and Technology (ERC)
UNIVERSITY OF FLORIDA
Gainesville, Florida 32611
Prepared for
FLORIDA INSTITUTE OF PHOSPHATE RESEARCH

1855 West Main Street
Bartow, Florida 33830 USA
Contract Manager: G. Michael Lloyd, Jr.

FIPR Project Number: 96-01-141
March 1999
PERSPECTIVE
The manufacture of phosphoric acid by the wet process is often characterized as a

method devoted to producing the optimum size and shape phosphogypsum crystals. The
phosphogypsum crystal characteristics determine the speed of filtration that defines the
production rate for the plant and the phosphogypsum cake-washing efficiency that is one
of the more important components contributing to acceptable yields.
Controlling and/or modifying the phosphogypsum crystals has been a long-time

goal of all phosphoric acid producers. Over the years there have been a number of studies
designed to define the role of various naturally occurring metals in the phosphogypsum
crystal formation mechanisms. Organic crystal modifiers have also been investigated and
while they have never been demonstrated to be cost-effective for the dihydrate process,
they have found wide application in the hemihydrate process for phosphoric acid
manufacture. Where crystal modifiers have been used in the dihydrate process they have
often demonstrated variable performance over time that may have been a function of the
changes in minor element content of the phosphate rock.
There is little doubt that a cost-effective crystal modifier would find widespread
application in the phosphoric acid industry and that a crystal modifier that would be
consistently effective for the more highly variable rock that will be mined in the future in
Florida would be a decided asset for the Florida phosphate industry.
iii
ABSTRACT
Phosphoric acid is an important intermediate product for production of fertilizers.

It is mainly produced by the wet process, in which phosphate concentrate is leached with
sulfuric and weak phosphoric acids to produce phosphoric acid. Calcium sulfate
crystallization occurs as leaching is taking place. According to the process adopted,
calcium sulfate dihydrate (gypsum) (CaSO4.2H2O) or calcium sulfate hemihydrate
(CaSO4.0.5H2O) is crystallized. Filtration of gypsum to recover phosphoric acid is the
bottleneck in this process. Therefore, the major goal of this project is to enhance filtration
of phosphogypsum using cost-effective reagents. Clarification of the mechanisms
controlling the effect of such additives is another important objective in this study. In
order to achieve these goals, several tasks have been conducted including: (1) bench-
scale testing of effect of two different surfactants (Crysmod and HiFlo-S5*) on crystal
modification and filtration of phosphogypsum produced from South Florida high
magnesium phosphate concentrate, at low, medium, and high sulfate levels, (2) role of
techniques and point of addition of surfactants, (3) cost-benefit analyses, and (4) basic
studies to elucidate the mechanisms controlling the effect of surfactants under different
conditions.
The results suggest the following:
• Optimum filtration results could be obtained by addition of surfactant as a mixture

with water during nucleation.
• Both surfactants can improve filtration rate at all sulfate levels.
• At any sulfate content, both P2O5 recovery and reaction efficiency are higher with the
addition of surfactants.
• Both surfactants decrease the induction time at different super-saturations. Also, larger
pure calcium sulfate dihydrate crystals with higher mean crystal diameter are produced
in the presence of surfactants.
• A gain of about $1.0 / ton P2O5, could be realized if Crysmod is used. However, the
gain will be about $4.0 / ton P2O5 if HiFlo-S5 is used. Such gain is in addition to the
increase in filtration rates due to use of these surfactants.
____________________
* Patent pending.
v
ACKNOWLEDGMENTS
The Florida Institute of Phosphate Research (FIPR) is acknowledged for

sponsoring this study (FIPR Grant #96-01-141). The Engineering Research Center for
Particle Science & Technology (ERC) at the University of Florida, the National Science
Foundation grant #EEC-94-02988 and the Industrial Partners of the ERC are
acknowledged for their partial financial support. Also, the principal investigators wish to
thank the IMC-Agrico Company for providing the phosphate concentrate sample used in
this study. Efforts conducted by the following research team are gratefully appreciated:
Robert Gaertner (graduate student) and Arlene Maya, Daniel M. Kuncicky, Amy Bowden,
Kathryn Gacek, Caroline Dugopolski (undergraduate students).
vi
TABLE OF CONTENTS
Page
PERSPECTIVE .....................................................................................................iii
ABSTRACT............................................................................................................v
ACKNOWLEDGMENTS ......................................................................................vi
EXECUTIVE SUMMARY .....................................................................................1
INTRODUCTION ..................................................................................................7
METHODOLOGY..................................................................................................9
Materials Characterization ..................................................................................9

Apparatus.........................................................................................................10
Procedure .........................................................................................................10
Estimation of Filtration Rate ......................................................................10

Calculation of Reaction Efficiency,
P2O5 Recovery and Washing Efficiency..................................................11
Determination of Crystal Size Distribution of Produced Phosphogypsum....12
Preparation of Phosphogypsum Slurry........................................................13
Surfactant Addition Techniques .................................................................14
RESULTS AND DISCUSSION............................................................................17
Study of Surfactant Addition Techniques.........................................................17
Addition of Crysmod Surfactant Before Nucleation....................................17

Addition of Crysmod Surfactant During Nucleation ...................................20
Addition of Crysmod Surfactant After Nucleation ......................................23
Testing of Surfactants at Different Sulfate Contents.........................................25
Testing at Low Sulfate Content (1.5-1.8%)................................................25

Testing at Medium Sulfate Content (2.2-2.4%)..........................................28
Washing With Simulated Pond Water....................................................30

Effect of Retention time on Crystal Size Distribution ...........................30
vii
TABLE OF CONTENTS (CONTINUED)
Testing at High Sulfate Content (3.2-3.5%) ...............................................35

Correlation of Crystal Size Distribution of Phosphogypsum
and the Filtration Rate.............................................................................37
Gypsum Morphology.................................................................................39
BASIC STUDIES: PRIMARY NUCLEATION OF

CALCIUM SULFATE DIHYDRATE.............................................................45
Materials .........................................................................................................45
Procedure........................................................................................................45
Turbidity Measurements and Estimation of Induction Time (T)....................... 47
Calculation of Crystal Growth Efficiency (E) .................................................. 47
Calculation of Super-saturation (S)................................................................. 48
Effect of Surfactants on Crystal Growth (Turbidity) at
Different Super-saturation .........................................................................48
Effect of Super-saturation on Induction Time and Crystal
Growth Efficiency With and Without Surfactants.......................................51
Effect of Surfactants’ Concentration on Induction Time
and Crystal Growth Efficiency ...................................................................51
Correlation between Super-saturation and Induction Time ...............................52
Effect of Surfactants on Crystal Size Distribution of Gypsum...........................53
Effect of Surfactants on Gypsum Crystal Morphology .....................................55
COST-BENEFIT ANALYSES..............................................................................57
CONCLUSIONS AND RECOMMENDATIONS .................................................59
REFERENCES .....................................................................................................61
viii
LIST OF FIGURES
Figure Page
1. Simple dihydrate process flow-sheet........................................................ 7
2. Volume % of fine (-10 microns) phosphogypsum with and without

surfactant (before nucleation)............................................................... 19
3. Effect of surfactant/recycle acid mixture addition before

nucleation on phosphogypsum crystal size distribution .......................... 19
4. Volume % of fine (-10 microns) phosphogypsum crystals with

and without surfactant (during nucleation)............................................. 22
5. Effect of surfactant addition during nucleation

on phosphogypsum crystal size distribution ........................................... 22
6. Volume % (–10 microns) of phosphogypsum crystals with

and without surfactant (surfactant in water added after nucleation)........ 24
7. Effect of surfactant addition after nucleation

on phosphogypsum size distribution ...................................................... 25
8. Comparative size distribution of phosphogypsum crystals

(at low sulfate content and after 3 hr)................................................... 27

(at medium sulfate content and after 3 hr) ............................................ 29
10. Volume % (-10 µm) of phosphogypsum crystals

with and without surfactants (at medium sulfate content) ...................... 32
11. Volume % (-38 µm) of phosphogypsum crystals

with and without surfactants (at medium sulfate content) ...................... 33

(without surfactant and at medium sulfate content)................................ 33

(with Crysmod surfactant and at medium sulfate content)...................... 34

(with Hiflo-S5 surfactant and at medium sulfate content)....................... 34
ix
LIST OF FIGURES (CONTINUED)
Figure Page

(at high sulfate content and after 3 hr) ................................................... 37
16. Example of turbidity data plotting and induction time

estimation ............................................................................................. 47
17. Effect of time on turbidity of calcium sulfate dihydrate

(super-saturation 1.018, 100 ppm surfactant) ........................................ 49



21. Relation between super-saturation and induction

time of calcium sulfate dihydrate ........................................................... 52
22. Linear relationship between super-saturation and natural log of

induction time for calcium sulfate dihydrate........................................... 53
23. Differential volume at a super-saturation of 1.502 after

5 minutes (surfactant concentration = 100 ppm) ................................... 54
x
xi
LIST OF TABLES
Table Page
1. Effect of Adding Crysmod Surfactant/Recycle Acid Mixture Before

Nucleation on Phosphogypsum Filtration at Low Sulfate Levels
(1.5-1.8%) .............................................................................................. 1
2. Effect of Crysmod Surfactant Addition during Nucleation on

Phosphogypsum Filtration at Low Sulfate Levels (1.5-1.8%) .................. 2
3. Effect of Crysmod Surfactant Addition after Nucleation on

Phosphogypsum Filtration at Low Sulfate Levels (1.5-1.8%) .................. 3
4. Filtration and Reaction Data at Low Sulfate Content (1.5-1.8%) ............. 3
5. Filtration and Reaction Data at Medium Sulfate Content (2.2-2.4%) ....... 4
6. Filtration and Reaction Data at High Sulfate Content (3.2-3.5%)............. 5
7. Chemical Analysis of High Dolomitic Phosphate Concentrate Sample...... 9
8. Sieve Analysis of High Dolomitic Phosphate Concentrate Sample ........... 9
9. Effect of Adding Surfactant / Recycle Acid Mixture Before

Nucleation on Phosphogypsum Filtration at Low Sulfate Levels
(1.5-1.8%) ............................................................................................ 18
10. Effect of Adding Surfactant / Recycle Acid Mixture Before

Nucleation on Reaction Efficiency, P2O5 Recovery and Washing
Efficiency.............................................................................................. 18
11. Effect of Surfactant Addition during Nucleation on Phosphogypsum

Filtration at Low Sulfate Levels (1.5-1.8%)........................................... 21
12. Effect of Surfactant Addition during Nucleation on Reaction

Efficiency, P2O5 Recovery and Washing Efficiency................................ 21
13. Effect of Surfactant Addition After Nucleation on Phosphogypsum

Filtration at Low Sulfate Levels (1.5-1.8%)........................................... 23
14. Effect of Surfactant Addition After Nucleation on Reaction

Efficiency, P2O5 Recovery and Washing Efficiency................................ 24
xii
LIST OF TABLES (CONTINUED)
Table Page
15. Filtration and Reaction Data at Low Sulfate Content (1.5-1.8%)........... 26
16. Comparative Size Analyses of Gypsum Crystals (at Low Sulfate

Content and After 3 Hr)........................................................................ 26
17. Statistics of Phosphogypsum Crystal Size Distribution (at Low

Sulfate and After 3 Hr) ......................................................................... 27
18. Filtration and Reaction Data at Medium Sulfate Content

(2.2-2.4%) ............................................................................................ 28
19. Comparative Size Analyses of Gypsum Crystals (at Medium

Sulfate Content and After 3 Hr) ........................................................... 29
20. Filtration and Reaction Data at Medium Sulfate Content (2.2-2.4%) ..... 30
21. Comparative Size Analyses of Gypsum Crystals (Without

Surfactant and at Medium Sulfate Content)........................................... 30
22. Comparative Size Analyses of Gypsum Crystals (With

Crysmod and at Medium Sulfate Content)............................................. 31
23. Comparative Size Analyses of Gypsum Crystals (With Hiflo-S5

and at Medium Sulfate Content)............................................................ 31
24. Filtration and Reaction Data at High Sulfate Content (3.2-3.5%)........... 35
25. Comparative Size Distribution of Gypsum Crystals (at High

Sulfate Content and After 3 Hr) ........................................................... 36
26. Statistics of Phosphogypsum Crystal Size Distribution (at High

Sulfate and After 3 Hr) ......................................................................... 36
27. Relationship between % Phosphogypsum Fines and Filtration

Rate at Low Sulfate Content ................................................................. 38
28. Relationship between % Phosphogypsum Fines and Filtration

Rate at Different Sulfate Contents......................................................... 39
xiii
LIST OF TABLES (CONTINUED)
Table Page
29. Amounts of Calcium Hydrogen Phosphate Monobasic and

Sulfuric Acid for Different Super-saturation .......................................... 46
30. Experimental Conditions for Primary Nucleation Studies....................... 46
31. Calculation of Super-saturation ............................................................. 48
32. Effect of Super-saturation on Induction Time (T) and Growth

Efficiency (E) of Gypsum Crystals (Using 100 ppm Surfactant)............. 51
33. Effect of Surfactant Concentration on Induction Time (T) and

Growth Efficiency (E) of Gypsum Crystals (Using 1.502
Super-saturation) .................................................................................. 52
34. Statistics of Size Distribution of Gypsum Crystals ................................. 54
xiii
EXECUTIVE SUMMARY
The major goal of this project is to enhance filtration of phosphogypsum using

cost-effective reagents. Clarification of the mechanisms controlling the effect of such
additives is another important objective in this study. In order to achieve these goals the
following tasks have been conducted:
• Bench scale testing of effect of two different surfactants (Crysmod and HiFlo-S5*)
on crystal modification and filtration of phosphogypsum produced from South
Florida high magnesium phosphate concentrate, at low-, medium- and high- sulfate
levels.
• Role of techniques and point of addition of surfactants; specific techniques involve
adding surfactants:
• As a mixture with recycle acid or water, and
• Before, during, or after nucleation
• Cost-benefit analyses
• Basic studies to elucidate the mechanisms controlling effect of surfactants under
different conditions.
The results are summarized in the following paragraphs.
ADDITION OF SURFACTANT BEFORE NUCLEATION
Tests are carried out in which the Crysmod surfactant (1.5 kg/ton P2O5) is mixed
with recycle acid before nucleation for duration times of 30 and 5 minutes. The filtration
data in Table 1 show that, in general, mixing of the surfactant with recycle acid and its
addition before nucleation result in low filtration rates as compared to the baseline without
surfactant. It may be attributed to an inhibition of regular crystal growth as indicated by
the increase in the amount of fine crystals obtained in presence of surfactant as added
before nucleation.
Table 1. Effect of Adding Crysmod Surfactant/Recycle Acid Mixture Before
Nucleation on Phosphogypsum Filtration at Low Sulfate Levels (1.5-1.8%).
Surfactant Addition Technique
Item Baseline To RA† before To RA† before
(without Nucleation and Nucleation and
Surfactant) Mixed for 30 Min. Mixed for 5 Min.
Filtration rate, 4.24 1.39 3.51
Ton P2O5 /m2.day
Moisture % 28.4 33.3 30.2
Change in filtration rate, % - - 67 - 17
†RA: Recycle acid
_____________________
* Patent pending.
1
ADDITION OF SURFACTANT DURING NUCLEATION
Tests are conducted by mixing the Crysmod surfactant (1.5 kg/ton P2O5) with
either recycle acid or water and adding the mixture during nucleation. The results are
shown in Table 2 below.
Table 2. Effect of Crysmod Surfactant Addition During Nucleation on

Phosphogypsum Filtration at Low Sulfate Levels (1.5-1.8%).

Item Baseline With RA* and Added With Water and Added
(without During the Nucleation During the Nucleation for
Surfactant) for 30 Min. 30 Min.
ton P2O5 /m2.day
Moisture % 28.4 31.6 27.5
Change in filtration - - 39 + 31
rate, %
* RA : Recycle acid
The filtration data show that mixing surfactant with recycle acid during nucleation
leads to a decrease in filtration rate. On the other hand, adding surfactant / water results in
an increase in filtration rate, which corresponds to about 31% improvement over that
obtained in the absence of surfactant. Such improvement is attributed to change in
crystallization as indicated by a decrease in the amount of fine crystals produced in the
presence of surfactant/ water mixture.
ADDITION OF SURFACTANT AFTER NUCLEATION
A test was carried out by mixing the Crysmod surfactant (1.5 kg/ton P2O5) with
water, then adding the mixture continuously for 30 minutes. Feeding the surfactant
mixture started after addition of the phosphate concentrate and sulfuric acid. The filtration
data in Table 3 suggest that filtration rate is slightly improved (8%) due to addition of
surfactant after nucleation.
Thus, it may be concluded that optimum filtration results could be obtained by

addition of surfactant as a mixture with water after nucleation.
2
Table 3. Effect of Crysmod Surfactant Addition After Nucleation on

Item Baseline (without With Water and Added for
Surfactant) 30 Min. After the
Nucleation by 30 Min.
Filtration rate, 4.24 4.56
Ton P2O5 /m2.day
Moisture % 28.4 27.7
Change in filtration rate, % - +8
* RA : Recycle acid
EFFECT OF SURFACTANTS AT LOW SULFATE CONTENT (1.5-1.8%)
Tests have been carried out with and without surfactants addition at low sulfate
contents. The amount of applied Crysmod and Hiflo-S5 surfactants correspond to 1.5 and
0.75 kg/ton P2O5 produced, respectively. The filtration and reaction data are given in
Table 4. The filtration rate is 4.24 ton P2O5 /m2.day without surfactant and improved to
5.54 and 6.88 ton P2O5 /m2.day with the addition of Crysmod and Hiflo-S5 surfactants,
respectively. The percentages of filtration rate improvement correspond to about 31 and
62%, respectively. It is also important to note that the reaction efficiency is increased with
the addition of both surfactants. This may be attributed to decrease of both lattice (co-
crystallized) P2O5 and unreacted P2O5 losses. The lattice (co-crystallized) P2O5, generally,
represent about 60% of the total P2O5 losses. The increase in reaction efficiency is
significant as it approaches 3.9% and 4.1% and the increase in P2O5 recovery is 1.6% and
3.1% with Crysmod and Hiflo-S5 surfactants, respectively.
Table 4. Filtration and Reaction Data at Low Sulfate Content (1.5-1.8%).
Without With Crysmod With Hiflo-S5

Item Additive Surfactant Surfactant
Ton P2O5 /m2.day
Moisture % 28.4 27.5 27.8
Increase in filtration rate, % - 31 62
Reaction efficiency, % 93.2 97.1 97.3
P2O5 recovery, % 92.5 94.1 95.6
Washing efficiency, % 99.3 97.0 98.3
3
EFFECT OF SURFACTANTS AT MEDIUM SULFATE CONTENT (2.2-2.4%)
To cover the range of sulfate contents applied in industry, tests at medium sulfate
content (2.2-2.4%) are carried out with and without addition of surfactants. At medium
sulfate content, the filtration data are given in Table 5. The filtration rate has increased by
25% and 43% with the addition of Crysmod and Hiflo-S5 surfactants respectively. This is
attributed to modification and growth of crystals upon surfactant addition as confirmed by
the gypsum crystals’ size distribution.
It is most interesting that, at medium sulfate content, both P2O5 recovery and
reaction efficiency are higher with the addition of surfactants. The increase in reaction
efficiency is significant as it approaches 2.8% with Crysmod surfactant and 1.9% with
Hiflo-S5 surfactant as well as the increase in P2O5 recovery is 1.4% with Crysmod
surfactant and 1.8% with Hiflo-S5 surfactant.
Table 5. Filtration and Reaction Data at Medium Sulfate Content (2.2-2.4%).
Item Without With Crysmod With Hiflo-S5

Surfactant Surfactant Surfactant
Filtration rate, tonP2O5/m2.day 5.13 6.42 7.34
Moisture % 28.3 27.8 28.6
Increase in filtration rate, % - +25 + 43
P2O5 recovery, % 95.0 96.4 96.8
EFFECT OF SURFACTANTS AT HIGH SULFATE CONTENT (3.2-3.5%)
The excess of sulfuric acid in the reaction slurry is the most effective factor
governing crystallization quality of phosphogypsum. Its effect is not only on the crystal
shape and size, but also on co-crystallized P2O5 losses and the unreacted P2O5 losses.
Therefore, it was decided to test the effect of adding surfactants at high levels of sulfate.
The results are given in Table 6. It is interesting to note that, in this case also, the filtration
rate has improved in the presence of surfactants.
The reaction efficiencies and P2O5 recoveries in the presence and absence of
surfactants are given in the same table. It is most interesting that, even at high sulfate
content, both P2O5 recovery and reaction efficiency are higher due to the addition of
surfactants.
4
Table 6. Filtration and Reaction Data at High Sulfate Content (3.2-3.5%).

Additive Surfactant Surfactant
Ton P2O5 /m2.day
Moisture % 27.9 28.4 28.7
P2O5 recovery, % 96.4 96.9 97.2
Basic studies were performed to understand the effect of surfactants on the

nucleation of pure calcium sulfate dihydrate (gypsum). The results show that both
surfactants decrease the induction time at different super-saturations. Also, larger pure
calcium sulfate dihydrate crystals with higher mean crystal diameter are produced in the
presence of surfactants.
Most importantly, the cost-benefit analyses suggest a gain of about $1.0/ton P2O5,
if Crysmod is used. However, the gain will be about $4.0/ton P2O5 if HiFlo-S5 is used.
Such gain is in addition to the increase in filtration rates due to use of these surfactants.
5
INTRODUCTION
Phosphoric acid is an important intermediate product for production of fertilizers.

It is mainly produced by the wet-process, in which phosphate concentrate is leached with
sulfuric and weak phosphoric acids to produce phosphoric acid. Calcium sulfate
crystallization occurs as leaching is taking place. According to the process adopted,
calcium sulfate dihydrate (gypsum) (CaSO4.2H2O) or calcium sulfate hemihydrate
(CaSO4.0.5H2O) is crystallized. A simple dihydrate process flowsheet is given in Fig. 1.
The primary reaction for the dihydrate process is given below (Becker 1989):
Ca10F2(PO4)6 + 14H3PO4 à 10Ca(H2PO4)2 + 2HF

10Ca(H2PO4)2 + 10H2SO4 + 20H2O à 20H3PO4 + 10CaSO4·2H2O
______________________________________________________________
Ca10F2(PO4)6 +10H2SO4 + 20H2O à 6H3PO4 + 10CaSO4·2H2O+ 2HF
After leaching, the slurry is filtered and counter-current washed to separate the
acid from the phosphogypsum cake. Filtration operation represents a bottleneck in the
wet-process phosphoric acid industry. Using the same filter area, the production capacity
with lower operating costs can be achieved if the filtration rate is increased. It is known
that the filtration rate depends on the characteristics of the filter cake such as crystal size,
size distribution and morphology of the crystals. In other words, large, spherical and
narrow size distribution crystals give better filtration rate.
Recycle Acid 20% P2O5
Phosphate Product
Sulfuric WI
Concentrate Acid
Acid
28% P2O5
REACTION FILTRATION WASHING I
Wash Water W II
Gypsum Cake WASHING III WASHING II
W II 5%P2O5 W I 10%P2O5
Figure 1. Simple Dihydrate Process Flowsheet.
W: Wash liquor
7
High magnesium phosphate concentrate from South Florida was tested for
phosphoric acid production with and without surfactants at different sulfate contents. The
additives are used as crystal modifiers for phosphogypsum. In this regard, there are a
limited number of published reports about industrial crystallization of phosphogypsum
using surfactants. The available data in literature is about nucleation and crystal growth of
gypsum from pure chemicals (Ibragimova 1986; Klima 1987; Witkamp 1990; Hatakka
1997). Also, the influence of different inhibitors on the growth rate of calcium sulfate
dihydrate crystals has been studied (Ibragimova 1986; Klima 1987; Witkamp 1990;
Hatakka 1997). In addition, many research papers are published explaining the effects of
additives and impurities on crystallization of calcium sulfate (Davey 1982; Botsaris 1982;
Budz 1986; Van Rosmalen 1987; Hasson 1990; Martynowicz 1996). There are, however,
numerous reports about improving the filtration rate using polymers (Cody 1991; Liu
1973a; Liu 1973b; Liu 1975; MacCartney 1958; Sarig 1975; Smith 1970; Smith 1971;
Taylor 1989; Shoulian 1996; Moudgil 1995). At the same time, studies dealing with effect
of surfactants on filtration of phosphogypsum are limited (Rocha 1995; Schroeder 1975;
Kopyleva 1987). Even in these studies, crystallization of calcium sulfate dihydrate from
pure chemicals is reported but no data was available about surfactant addition technique.
South Florida phosphate concentrate has a relatively high impurities content (MgO
+ Fe2O3 + Al2O3 + Na2O + K2O = 4.82 %). Generally, some of these impurities lead to
decrease in the filtration rate during the processing of this particular phosphate using the
dihydrate process. Also, excessive foaming is experienced during acidulation due to high
carbonate content. To study this concentrate, dihydrate process conditions were simulated
(Abdel-Aal, 1984, 1989) and two different surfactants were tested. The size distributions
of the gypsum crystals were studied to determine the effect of the surfactants on crystal
growth at different sulfate levels.
The main aim of this study is to enhance the filtration rate of phosphogypsum
using cost-effective reagents. In addition, clarification of the mechanisms controlling the
effect of such additives is another important objective in this study. Achieving these goals
will allow the following advantages:
• High filtration rate (high plant productivity)

• High P2O5 recovery from the phosphate concentrate (high plant efficiency)
• High P2O5 concentration in the filter acid, leading to energy savings during the
concentration
• Low total P2O5 losses in the phosphogypsum by-product.
8
METHODOLOGY
MATERIALS CHARACTERIZATION
IMC-Agrico Company provided the phosphate concentrate sample (high

magnesium). Chemical and sieve analyses of this sample are given in Tables 7 and 8. As
seen in Table 7, the phosphate concentrate contains high MgO content (1.58%). High
MgO contents increase the acid viscosity, consequently decreasing the filtration rate. The
accepted limit in commercial phosphate concentrates is less than 1.0%. The sieve analysis
of the sample is given in Table 8. It contains 51.8% -100 mesh and 29.7% -200 mesh
particle size, which are suitable for the dihydrate process. Most of the sample weight
(80.5%) falls in the 44-295 Fm size range.
Pure (95.5%) sulfuric acid of 1.835 g/ml specific gravity is used for digestion. The
recycle (return) acid is accumulated while carrying out the tests. It is adjusted to the
required P2O5 content before recycling. For the first test (start-up), the required amount of
recycle acid (about 18% P2O5) is prepared by double-stage acidulation, the first with water
and sulfuric acid while the second with about 10% P2O5 phosphoric acid produced from
the previous run and sulfuric acid. The surfactants used are Crysmod (a mixture of C6-C22
Sorbitan esters) supplied by Polimeros Sinteticos, Mexico City, Mexico and Hiflo-S5
(patent pending) surfactants.
Table 7. Chemical Analysis of High Dolomitic Phosphate Concentrate Sample.
Constituent % Constituent %
P2O5 27.22 Al2O3 0.95
CaO 42.79 Na2O 0.73
MgO 1.58 K2O 0.12
Fe2O3 1.56 Insolubles 8.05
Table 8. Sieve Analysis of High Dolomitic Phosphate Concentrate Sample.
Particle Size Cumulative wt. % Passing

Mesh* Inch Mm
-10 0.0650 1.651 100
-48 0.0116 0.295 87.5
-65 0.0082 0.208 68.1
-100 0.0058 0.147 51.8
-150 0.0041 0.104 40.1
-200 0.0029 0.074 29.7
-270 0.0021 0.053 18.6
-325 0.0017 0.044 7.0
* Tylor standard
9
APPARATUS
The reaction is carried out in a cylindrical 1L reactor of 10 cm diameter. It is fitted

with a Teflon-coated stirrer and placed in a water bath adjusted to 80oC. The impeller tip
speed is adjusted at 1.44 m/s (550 rpm). The phosphate concentrate is added continuously
using a vibrating rock feeder. The sulfuric acid is added continuously using a peristaltic
pump with Viton tubing. The surfactant/water suspension is continuously added using a
small graduated separating funnel. Filtration is performed using Buchner type filter of 4.6
inch diameter. Polypropylene filter cloth of 80 mesh aperture size is used. A vacuum pump
is used for filtration. Coulter Laser Diffraction Analyzer model LS230 is used for
determination of size distribution of the produced phosphogypsum crystals. JEOL
Scanning Electronic Microscope model JSM-6400 was used for investigation of crystal
morphology.
PROCEDURE
1. Add the required amount of the return (recycle) acid R.A. to the reactor.
2. Wait until the temperature of R.A. goes up to 80 oC.
3. Add the phosphate concentrate, sulfuric acid, and the surfactant/water
suspension, continuously, for 30 minutes.
4. Continue the reaction for another 2.5 hours.
5. After the required retention time, the slurry is weighed and poured into the
filter.
6. Apply vacuum and record the time from the moment of vacuum application to
just before the surface appears dry to avoid the effects of channeling and
cracking (wall effect).
7. Wash with the required amount of wash I (~10 % P2O5) and repeat step 6.
8. Repeat step 6 with wash II (~5% P2O5), and wash water or simulated pond
water. Then allow the cake to filter for 5 seconds to dry (taking off as much as
possible of the residual liquor from the cake).
9. Weigh the wet cake and determine its moisture by drying at 60 oC for 24 hr.
10. After 0.5, 1.0, 2.0 and 3.0 hours from the start-up of the test, take 3 ml slurry
and disperse in 100 ml methanol. Then sieve using 106 Fm screen and
determine the size distribution using Coulter laser diffraction analyzer.
Estimation of Filtration Rate
In industry the filtration rate is expressed in tons P2O5 produced per square meter
per day. So the same expression was used. The filtration rate was calculated applying the
following equation:
F.R. = SW * SC * F/ T Ton P2O5 /m2.day
10
where:
F.R: Filtration rate, ton P2O5 /m2.day
SW: Slurry weight, g
SC: Solid content, %
F: Filtration factor
T: Total time of filtration, washing and drying, sec
The filtration factor was calculated according to the following equation:
F = 94 * 56 * 27.22 * 104 * 86400 * 94 * q500/381 = 1.688

100 * 172 * 42.79 * 106 * 107.3 * 100
where:
94% : is the % CaSO4.2H2O in solids (gypsum purity)
56 and172 : are the molecular weights of CaO and CaSO4.2H2O
27.22 and 42.79 : are the % of P2O5 and CaO in phosphate concentrate
104 : is the conversion factor from cm2 to m2
86400 : is the conversion factor from sec to day
94% : is the P2O5 recovery
q500/381 : is the pressure difference correction factor

106 : is the conversion factor from gm to ton
107.3 : is the filter area in cm2
After the required retention time, the produced phosphogypsum slurry was filtered
and washed under the following conditions:
Type of filter = Buchner-type funnel

Diameter of filter = 4.6 inch (11.684 cm)
No. of cloth layers = 1 (single)
Filter area = 107.3 cm2
Type of filter cloth = Polypropylene
Pressure Difference = 15 inch. Hg (381 mm. Hg)
Temperature of Slurry = 70-80 o C
Number of washing =3
Wash liquors temperature = 50-60 o C
Calculation of Reaction Efficiency, P2O5 Recovery and Washing Efficiency
Reaction efficiency (digestion or process efficiency, % of extraction or conversion)

is defined as the % of P2O5 removed from the phosphate concentrate into solution,
because some of the P2O5 is lost with an incompletely washed gypsum. The reaction
11
efficiency is calculated from the following equation that obtained form IMC-Agrico
Company (Moudgil, 1995):
Reaction efficiency = 100 - 94*(A - B)*C/(D*E)
where:
A: % Total P2O5 in gypsum cake,

B: % Water-soluble P2O5 in gypsum cake,
C: %CaO in phosphate concentrate used to make the acid,
D: % P2O5 in phosphate concentrate used to make the acid,
E: %CaO in gypsum cake.
Also, P2O5 recovery (overall or plant efficiency, or P2O5 yield) is defined as the %
of P2O5 passing from the phosphate concentrate into the produced phosphoric acid. It is
calculated as follows (Moudgil 1995):
P2O5 recovery = 100 - (94*A*C)/(D*E)
The main objective of the washing operation is to extract, with wash liquor, as
much as possible of the phosphoric acid held by the capillary forces in the interstices of the
gypsum cake. Wash liquor displaces the impregnating phosphoric acid. The washing
(filtration) efficiency is defined as the % of water-soluble P2O5 passing from the slurry into
the produced phosphoric acid. It is calculated applying the following relation (Moudgil
1995):
Washing efficiency = 100 - (94*B*C)/(D*E)
Determination of Crystal Size Distribution of Produced Phosphogypsum
Different tests are carried out to find out the best methodology for sampling
gypsum crystals for Laser diffraction size analyzer including:
• The first test is conducted using about 1.0 g dry gypsum cake. Drying is
o
performed at 60 C for 24 hours. The sample is, then dispersed in 100 ml
methanol.
• The second test is carried out using about 1.0 g air dried gypsum cake and
dispersed in 100 ml methanol.
• In the third test, a 3.0 ml slurry sample is withdrawn at different time intervals
and dispersed in 100 ml methanol.
12
. The fourth test is performed by taking 3.0 ml slurry sample at different time
intervals, and dispersed in 100 ml methanol, then screened using 106 m sieve
before conducting size analyses by Laser Diffraction Technique.
In this report, size distribution analyses have been done as outlined in the fourth test
above.
Preparation of Phosphogypsum Slurry
Phosphogypsum slurry is prepared at 80oC reaction temperature for 3 hours

retention time. Recycle acid (~18% P2O5) is one of the three main starting materials fed to
the reactor with concentrated sulfuric acid (95.5%) and phosphate concentrate.
Preparation of the recycle acid from the same concentrate is very important to simulate the
continuous industrial unit. The level of the impurities in the filter acid depends entirely on
its level in recycle acid. Utilization of pure phosphoric acid or recycle acid produced from
another phosphate concentrate would produce erratic results for the filtration rates and
level of impurities in both filter acid and phosphogypsum. The technique of recycle acid
preparation comprises the following two stages; namely preparation of about 10% P2O5
phosphoric acid and preparation of about 18% P2O5 phosphoric acid (recycle acid).
Amounts of reacted materials are calculated using a material balance computer

program (Abdel-Aal 1984 and 1989). Sulfuric acid and water are used in the first stage.
After the required retention time, the slurry is filtered and the cake is washed one time
with wash water. The same test is repeated under the same conditions except that the
already prepared first filtrate was used instead of water and the gypsum cake is washed
twice, one with the previously produced wash liquor (about 5% P2O5) and the second with
wash water or simulated pond water. It should be noted that the amount of wash liquor
used is also calculated using the computer program which depends on the water balance
all over the process. Sufficient amounts of recycle acid, wash I and wash II were
accumulated from several tests. During the phosphogypsum slurry filtration, recycle acid,
wash I and wash II are produced and recycled in the next run.
For preparation of phosphogypsum slurry the following data were used for the
calculation of feeding amounts of reactants:
Operating parameters:
P2O5 in phosphate concentrate = 27.22 %

CaO in phosphate concentrate = 42.79 %
SO3 in phosphate concentrate = 1.44 %
Moisture in phosphate concentrate = 1.30 %
H2SO4 concentration = 95.5 %
13
Retention time = 180 min
P2O5 in recycle acid = 18.32 %
Free sulfate in filter acid = 2.0 %
P2O5 in filter acid = 27.5 %
Assumptions:
Overall reaction efficiency = 98 %

Gypsum in solids (purity ) = 94 %
Solids in slurry = 35%
P2O5 recovery = 94%
Moisture in gypsum = 25%
Slurry density = 1.50 g/ml
Volatilization rate = 0.2 kg/kg rock
Feeding amounts and operation factors:
Weight of phosphate concentrate = 240 g

Weight of sulfuric acid, 95.5% = 192.5 g
Volume of sulfuric acid, 95.5 % = 104.9 ml
Weight of recycle acid = 559.9 g
Volume of recycle acid = 474.5 ml
Washing factor (water /slurry ratio) = 0.308
Filtration factor = 1.688
Surfactant Addition Techniques
From the industrial point of view, it is very important to add the surfactant without
negative effects on process performance. Thus, different methodologies for surfactants’
addition were studied. For example, effect of surfactant addition to the reactor after its
mixing with recycle acid was tested. Also, mixing of surfactant with water instead of
recycle acid was studied. It should be mentioned that, the slurry solid content and amount
and P2O5 concentration of recycle acid have been kept similar to that of the baseline
(without surfactant). In these tests, the dosage of Crysmod surfactant is 1.5 kg/ton P2O5.
The following is a detailed description of these techniques:
Addition of Surfactant Before Nucleation:
• Mix surfactant with recycle acid for 30 minutes in the reaction vessel before
addition of reactants (phosphate concentrate and sulfuric acid). Then, add reactants
continuously for 30 minutes.
14
. Mix surfactant with recycle acid for 5 minutes before nucleation. Then add the
reactants continuously for 30 minutes.
Addition of Surfactant During Nucleation:
• Mix surfactant with a volume of recycle acid equivalent to 10% of the total
volume of recycle acid. The suspension is then added continuously during the
nucleation for 30 minutes.
• Mix surfactant with water (volume of water is 10% of the total volume of recycle
acid). The suspension is then added continuously during the nucleation for 30 minutes.
Addition of Surfactant After Nucleation:
• Mix surfactant with a volume of water (equivalent to 10% of the total volume of
recycle acid). The suspension is then added continuously starting 30 minutes after
nucleation for 30 minutes addition time. In these experiments, it is assumed that, the
nucleation begins with the start-up of the reaction.
15
RESULTS AND DISCUSSION
South Florida phosphate concentrate (high magnesium) was tested for phosphoric
acid production with and without Crysmod and Hiflo-S5 surfactants. Surfactant addition
techniques are studied in details (before nucleation, during nucleation and after nucleation)
using Crysmod surfactant. During industrial crystallization of phosphogypsum, the
presence of free (excess) sulfuric acid in the reaction medium is the most effective factor
governing crystal quality. Its effect is not only on the crystal shape and size, but also on
co-crystallized P2O5 losses and the unreacted P2O5 losses (Becker 1989). Free sulfate
concentration affects the super-saturation, which is the driving force of both nucleation
and crystal growth. At low sulfate content, the super-saturation is low and the rate of
both nucleation and crystal growth are low and vice versa. The range of free sulfate
industrially applied is 2.0-3.0% (Becker 1989). However, sulfate as low as 1.5% and as
high as 3.5% can be encountered. Consequently, South Florida phosphate concentrate is
tested at different sulfate contents (1.5%-3.5%).
The results of filtration rates, reaction efficiencies, P2O5 recoveries, washing

efficiencies and size distribution of the phosphogypsum crystals are obtained. SEM
photomicrographs of the phosphogypsum crystals are given. Basic studies are also given.
STUDY OF SURFACTANT ADDITION TECHNIQUES
Addition of Crysmod Surfactant Before Nucleation
Tests were carried out in which the Crysmod surfactant (1.5 kg/ton P2O5) is mixed
with recycle acid before nucleation for duration times of 30 and 5 minutes. The data are
given in Tables 9 and 10 and in Figures 2 and 3. The results suggest, in general, mixing of
the surfactant with recycle acid and its addition before nucleation result in low filtration
rates as compared to the baseline without surfactant. It may be attributed to an inhibition
of regular crystal growth as indicated by the increase in the amount of fine crystals
obtained in presence of surfactant as added before nucleation.
Figures 2 and 3 show that % fine crystals of produced gypsum (<10 Fm) after 3
hours are increased to about 83% and 69% for tests of high mixing time (30 min) and low
mixing time (5 min), respectively as compared to lower value of 64% for the baseline. The
reasons for formation of very fine crystals are not yet understood. It may be attributed to
retardation or inhibition of regular crystal growth by surfactant/recycle acid mixture. On
the other hand, addition of surfactant has resulted in an increase in reaction efficiencies,
P2O5 recoveries and the washing efficiencies as compared to the baseline, where surfactant
is not added (see Table 10).
17
Table 9. Effect of Adding Surfactant/Recycle Acid Mixture Before Nucleation on
Item Time, sec

Surfactant Addition Baseline To RA* Before To RA* Before
Technique (without Nucleation and Nucleation and
Surfactant) Mixed for 30 Min. Mixed for 5 Min.
Filter acid 20 69 29
Recycle acid 42 105 47
Wash I 28 94 31
Wash II 20 77 27
Drying 5 5 5
Total time (in sec) 115 350 139
Ton P2O5 /m2.day
Moisture % 28.4 33.3 30.2
Change in filtration rate, % - - 67 - 17
* RA: Recycle acid
Table 10. Effect of Adding Surfactant/Recycle Acid Mixture Before Nucleation on

Reaction Efficiency, P2O5 Recovery and Washing Efficiency.
Surfactant Addition Technique Reaction P2O5 Recovery, Washing

Efficiency, % % Efficiency, %
Without surfactant (Baseline) 93.2 92.5 99.3
To RA* before nucleation and 94.8 91.7 96.7
mixed for 30 min
To RA* before nucleation and 95.2 93.1 97.8
mixed for 5 min
* RA: Recycle acid
18
95
Volume % of Phosphogypsum Crystals, Without Surfactant

90
30 min Mixing With Surfactant
85 5 min Mixing With Surfactant

-10 microns
80
75
70
65
60
After 1 hr. After 2 hr. After 3 hr.
Figure 2. Volume % of Fine (-10 Micron) Phosphogypsum Crystals With and

Without Surfactant (Surfactant Was Mixed With Recycle Acid
Before Nucleation).
Figure 3. Effect of Addition of Surfactant/Recycle Acid Mixture Before

Nucleation on Phosphogypsum Crystal Size Distribution.
19
Addition of Crysmod Surfactant During Nucleation
Tests were conducted by mixing the Crysmod surfactant (1.5 kg/ton P2O5) with
either recycle acid or water and adding the mixture during nucleation. The
surfactant/liquid suspension is added continuously for 30 minutes with the reactants
(during nucleation). The amount of either recycle acid or water is equivalent to 10% of
total volume of recycle acid. The results are given in Tables 11 and 12 and Figs. 4 and 5.
The filtration data in Table 11 show that mixing surfactant with recycle acid during
nucleation gives low filtration rate. On the other hand, adding surfactant/water mixture
during nucleation results in a higher filtration rate (5.54 ton P2O5/m2.day). This increase in
filtration rate corresponds to about 31% improvement.
Data in Figs. 4 and 5 show that mixing surfactant with recycle acid results in an
increase in the amount of fine crystals. On the other hand, less fines are produced due to
adding surfactant/water mixture.
The reasons of formation of very fine crystals upon the addition of surfactant
/recycle acid mixture during nucleation are not yet understood. However, it may be
attributed to reaction of the surfactant with the recycle acid leading to formation of some
compounds that inhibit crystal growth.
The reaction efficiencies, P2O5 recoveries and the washing efficiencies of the
phosphate concentrate with surfactant additions during nucleation are given in Table 12,
together with the data obtained without surfactant addition (baseline). It is clear that the
reaction efficiency is increased with the addition of surfactant. This may be due to
decrease of lattice (co-crystallized) P2O5 losses, which represent the majority of the total
P2O5 losses.
20
Table 11. Effect of Surfactant Addition During Nucleation on
Item Time, sec

Surfactant Baseline With RA* and Added With Water and Added
Addition (without During the Nucleation During the Nucleation for
Technique Surfactant) for 30 Min. 30 Min.
Filter acid 20 38 21
Recycle acid 42 72 27
Wash I 28 47 19
Wash II 20 28 16
Drying 5 5 5
Total time (in sec) 115 190 88
Ton P2O5 /m2.day
Moisture % 28.4 31.6 27.5
Change in filtration - - 39 + 31
rate, %
* RA : Recycle acid
Table 12. Effect of Surfactant Addition During Nucleation on Reaction Efficiency,

P2O5 Recovery and Washing Efficiency.
Surfactant Addition Technique Reaction P2O5 Recovery, Washing

Efficiency, % % Efficiency, %
Baseline 93.2 92.5 99.3
With RA and added during the 95.7 92.8 97.0
nucleation for 30 min
With water and added during 97.1 94.1 97.0
the nucleation for 30 min
* RA: Recycle acid
21
Without Surfactant
Volume % of Phosphogypsum Crystals, -10 microns

100
With Surfactant in R.A.
90 With Surfactant in Water
80
70
60
50
40
30
20
Figure 4. Volume % of Fine (-10 Micron) Phosphogypsum Crystals with and

without Surfactant (During Nucleation).
Figure 5. Effect of Surfactant Addition, During Nucleation,

on Phosphogypsum Crystal Size Distribution.
22
Addition of Crysmod Surfactant After Nucleation
A test was carried out by mixing the surfactant (1.5 kg/ton P2O5) with water then
adding the mixture continuously for 30 minutes. Feeding the surfactant mixture started
after addition of phosphate concentrate and sulfuric acid. The amount of surfactant
corresponds to 1.5 kg/ton P2O5 and the amount of water is equivalent to 10% of total
volume of recycle acid. The results are given in Tables 13 and 14 and Figures 6 and 7.
The filtration data in Table 13 show only 8% improvement in the filtration rate.
Results in Figures 6 and 7 suggest that the amount of fine gypsum crystals (<10
Fm) is decreased to about 37% upon addition of surfactant with water after nucleation.
Note that -10 micron fraction in the baseline without surfactant is 64%.
The reaction efficiencies, P2O5 recoveries and the washing efficiencies obtained
with surfactant addition after nucleation are given in Table 14 together with those
achieved without surfactant (baseline). In this case also, the reaction efficiency has
increased with the addition of surfactant. This may be attributed to decrease of lattice (co-
precipitated) P2O5 losses as mentioned above.
Table 13. Effect of Surfactant Addition, After Nucleation, on Phosphogypsum

Filtration at Low Sulfate Levels (1.5-1.8%).
Item Time, sec

Surfactant Addition Baseline (without With Water and Added for
Technique Surfactant) 30 Min. after the Nucleation
by 30 Min.
Filter acid 20 21
Recycle acid (RA*) 42 40
Wash I 28 24
Wash II 20 17
Drying 5 5
Total time (in sec) 115 107
Filtration rate, 4.24 4.56
Ton P2O5 /m2.day
Moisture % 28.4 27.7
Change in filtration rate, % - +8
* RA : Recycle acid
23
Table 14. Effect of Surfactant Addition After Nucleation on Reaction Efficiency,
P2O5 Recovery and Washing Efficiency.
Surfactant Addition Technique Reaction P2O5 Washing

Efficiency, % Recovery, % Efficiency, %
Baseline 93.2 92.5 99.3
With water and added for 30 min 96.8 93.8 97.8
after addition of phosphate
concentrate and sulfuric acid
100
Volume % of Phosphogypsum Crystals,
Without Surfactant
90
80 With Surfactant in Water

-10 microns
70
60
50
40
30
20
Figure 6. Volume % (–10 Microns) of Phosphogypsum Crystals

With and Without Surfactant (Surfactant in Water
Added After Nucleation).
24
Figure 7. Effect of Surfactant Addition, After Nucleation, on
Phosphogypsum Size Distribution.
TESTING OF SURFACTANTS AT DIFFERENT SULFATE CONTENTS
Testing at Low Sulfate Content (1.5-1.8%)
Tests have been carried out, with and without surfactants addition at low sulfate
contents. The amount of applied Crysmod and Hiflo-S5 surfactants correspond to 1.5 and
0.75 kg/ton P2O5 produced, respectively. The results are given in Tables 15-17 and Figure
8.
The filtration and reaction data are given in Table 15. The filtration rate is 4.24 ton
P2O5 /m2.day without surfactant and improved to 5.54 and 6.88 ton P2O5/m2.day with the
addition of Crysmod and Hiflo-S5 surfactants, respectively. The percentages of filtration
rate improvement correspond to about 31 and 62%, respectively. This is attributed to
modification and growth of produced crystals upon surfactant addition. The size
distributions of gypsum crystals (given in Tables 16 and 17 and Fig. 8) show that the
volume percentages of fine crystals (<10Fm) is decreased to about 31% and 9% upon
addition of Crysmod and Hiflo-S5 surfactants, respectively, while it is as high as 64%
without surfactant addition. Also, the mean of crystal size increases with addition of
surfactants.
The data for reaction efficiencies, P2O5 recoveries and washing efficiencies with
and without additives are given in Table 15. It is important to note that the reaction
25
efficiency is increased with the addition of both surfactants. As mentioned before, this may
be attributed to decrease of both lattice (co-crystallized) P2O5 and unreacted P2O5 losses.
The lattice (co-crystallized) P2O5 losses, generally, represent about 60% of the total P2O5
losses. In this regard, it has been reported that, presence of a surfactant has resulted in
reduction of P2O5 in gypsum and this is attributed to the surfactant effect in reducing the
capture of phosphate ions from solution by gypsum crystals. Regarding the decrease of
unreacted phosphate, it was found that addition of surfactant increases the phosphate
concentrate solubility and consequently increases reaction efficiency. In another study, it
has been found that the presence of surfactant increases the solubility of calcium
phosphate in thermal and wet-process phosphoric acid of 10-40% P2O5 concentration and
at temperature range of 40-75 oC.
The increase in reaction efficiency is significant as it approaches 3.9% and 4.1%

and the increase in P2O5 recovery is 1.6% and 3.1% with Crysmod and Hiflo-S5
surfactants, respectively.
Table 15. Filtration and Reaction Data at Low Sulfate Content (1.5-1.8%).

Ton P2O5 /m2.day
Moisture % 28.4 27.5 27.8
P2O5 recovery, % 92.5 94.1 95.6
Table 16. Comparative Size Analyses of Gypsum Crystals (At Low Sulfate Content
and After 3hr).
Size, Cumulative Volume % Passing

µm Mesh* Without With Crysmod With Hiflo-S5
125 120 100.00 100.00 99.83
106 140 100.00 99.91 97.84
90 170 100.00 99.05 90.95
75 200 99.99 95.68 75.32
63 230 99.82 89.70 57.93
53 270 99.15 82.11 44.09
45 325 97.93 74.91 34.76
38 400 96.21 68.12 27.67
10 - 64.28 30.75 8.95
* ASTM standards
26
Table 17. Statistics of Phosphogypsum Crystal Size Distribution
(At Low Sulfate and After 3 Hr).
Item Phosphogypsum Crystal Mean,

Size Distribution, µm µm
*dx d10 d25 D50 d75 d90

Without surfactant 1.0 2.3 6.3 14.3 25.5 10.3
With Crysmod
surfactant 1.9 7.2 22.2 45.1 63.5 28.1
With Hiflo-S5
surfactant 11.5 35.1 57.5 74.8 88.9 54.5
*dx= diameter of crystals passing x% by volume.
Figure 8. Comparative Size Distribution of Phosphogypsum Crystals

(At Low Sulfate Content and After 3 Hr).
27
Testing at Medium Sulfate Content (2.2-2.4%)
To cover the range of sulfate contents applied in industry, tests at medium sulfate
content (2.2-2.4%) were carried out with and without addition of surfactants. At medium
sulfate content, the obtained filtration data are given in Table 18. The filtration rate is
increased by 25% and 43% due to the addition of Crysmod and Hiflo-S5 surfactants,
respectively. This is attributed to modification and growth of produced crystals upon
surfactant addition as confirmed by the size distribution of gypsum crystals given in Table
19 and Figure 9. It is clear that the volume percentage of coarse phosphogypsum crystals
(>75Fm) is increased with surfactants addition. On the other hand, the volume percentage
of fine phosphogypsum crystals (<38Fm) is decreased with surfactants’ addition.
The reaction efficiencies and P2O5 recoveries in presence and absence of

surfactants at medium sulfate content are given in Table 18. It is most interesting that, at
medium sulfate content, both P2O5 recovery and reaction efficiency are higher with the
addition of surfactants. The increase in reaction efficiency is significant as it approaches
2.8% with Crysmod surfactant and 1.9% with Hiflo-S5 surfactant as well as the increase
in P2O5 recovery is 1.4% with Crysmod surfactant and 1.8% with Hiflo-S5 surfactant.

Surfactant Surfactant Surfactant
Filtration rate, tonP2O5/m2.day 5.13 6.42 7.34
Moisture % 28.3 27.8 28.6
Increase in filtration rate, % - +25 + 43
P2O5 recovery, % 95.0 96.4 96.8
28
Table 19. Comparative Size Analyses of Gypsum Crystals (at Medium Sulfate
Content and After 3 Hr).

µm mesh* Without With Crysmod With Hiflo-S5
125 120 100.00 99.85 99.58
106 140 99.88 99.18 97.65
90 170 98.95 98.06 92.94
75 200 95.63 95.65 83.60
63 230 89.86 90.55 72.01
53 270 82.47 80.54 60.53
45 325 74.98 66.31 51.67
38 400 67.09 49.95 44.66
10 - 22.06 10.38 13.94
* ASTM standards

(At Medium Sulfate Content and After 3 Hr).
29
Washing with Simulated Pond Water. Two tests were carried out to study
effect of using simulated pond water on the filtration rate without surfactant and with
Hiflo-S5 surfactant. The simulated pond water has specific gravity of 1.02 g/ml and
contains 1.5% P2O5 and saturated with phosphogypsum. The results are reported in Table
20 and compared with the results using wash water. Filtration rate was slightly decreased
upon washing with simulated pond water with and without surfactant. It is interesting to
notice a slight decrease in filtration rate whenever pond water is used for washing instead
of tap water.
Item Without Surfactant With Hiflo-S5

Surfactant
Washing liquor Water Pond water Water Pond water
Filtration rate, ton P2O5/m2.day 5.13 4.95 7.34 6.94
Moisture % 28.3 28.5 28.6 28.7
Change in filtration rate, % - -3.5 +43 + 40
Effect of Retention Time on Crystal Size Distribution. Slurry samples were

taken from the reactor for crystal size distribution at different time intervals (after 0.5, 1, 2
and 3 hours) with and without surfactant addition. The results are reported in Tables 21-
23 and Figures 10-14. The results show that, the volume percentages of fine crystals (-38
µm) are highly decreased with increasing the retention time for all the samples. In
addition, the amount of fine crystals is less in presence of surfactants as compared to the
tests without surfactant at all tested retention times.
Table 21. Comparative Size Analyses of Gypsum Crystals (Without Surfactant and
at Medium Sulfate Content).

µm mesh* After 0.5 hr After 1 hr After 2 hr After 3 hr
125 120 99.95 99.89 99.94 100-.00
106 140 99.75 99.76 99.40 99.88
90 170 99.31 99.45 97.68 98.95
75 200 98.25 98.28 93.75 95.63
63 230 96.31 95.64 88.39 89.86
53 270 92.99 91.25 82.33 82.47
45 325 88.28 85.72 76.41 74.98
38 400 81.78 78.86 69.97 67.09
10 - 27.14 27.72 24.24 22.06
* ASTM standards
30
Table 22. Comparative Size Analyses of Gypsum Crystals (With Crysmod and at
Medium Sulfate Content).

µm mesh* After 0.5 Hr After 1 Hr After 2 Hr After 3 Hr
125 120 99.85 99.97 99.89 99.85
106 140 99.53 99.67 99.42 99.18
90 170 99.12 99.18 98.84 98.06
75 200 98.61 98.28 97.53 95.65
63 230 97.53 96.02 94.36 90.55
53 270 94.16 90.38 87.20 80.54
45 325 86.96 79.87 75.30 66.31
38 400 75.33 64.63 59.64 49.95
10 - 17.23 12.89 12.27 10.38
* ASTM standards
Table 23. Comparative Size Analyses of Gypsum Crystals (With Hiflo-S5 and at
Medium Sulfate Content).

µm Mesh* After 0.5 Hr After 1 Hr After 2 Hr After 3 Hr
125 120 99.93 99.92 99.83 99.58
106 140 99.85 99.83 99.75 97.65
90 170 99.48 99.39 99.28 92.94
75 200 97.96 97.51 97.16 83.60
63 230 94.83 93.25 92.32 72.01
53 270 89.97 86.12 84.27 60.53
45 325 83.91 77.11 74.35 51.67
38 400 76.20 66.43 63.05 44.66
10 - 16.82 12.32 11.62 13.94
* ASTM standards
31
35
Without Surfactant
With Crysmod Surfactant

30
Volume % of Phosphoypsum Crystals , -10 microns
WithHiflo-S5 Surfactant
25
20
15
10
0
After 0.5 hr After 1 hr After 2 hr After 3 hr
Figure 10. Volume % (-10 µ m) of Phosphogypsum Crystals

With and Without Surfactants.
32
100
Without Surfactant
90 With Crysmod Surfactant

Volume % of Phosphogypsum Crystals, -38 microns
With Hiflo-S5 Surfactant

80
70
60
50
40
30
20
10
0
After 0.5 hr After 1 hr After 2 hr After 3 hr
Figure 11. Volume % (-38 µ m) of Phosphogypsum Crystals

With and Without Surfactants.

(Without Surfactant at Medium Sulfate Content).
33
(With Crysmod Surfactant at Medium Sulfate Content).

(With Hiflo-S5 Surfactant at Medium Sulfate Content).
34
Testing at High Sulfate Content (3.2-3.5%)
The excess of sulfuric acid in the reaction slurry is the most effective factor
governing crystallization quality of phosphogypsum. Its effect is not only on the crystal
shape and size, but also on co-crystallized P2O5 losses and the unreacted P2O5 losses.
Sulfate concentration affects super-saturation of the medium. At high sulfate content, the
super-saturation is high and the rate of both nucleation and crystal growth are high. The
amounts of applied Crysmod and Hiflo-S5 surfactants correspond to 1.5 and 0.75 kg/ton
P2O5 produced, respectively.
The results obtained are given in Tables 24-26 and Figure 15. It is interesting to
note that the filtration rate (6.77 ton P2O5 /m2.day without additive) has improved to 8.33
and 8.98 ton P2O5/m2day with the addition of Crysmod and Hiflo-S5 surfactants,
respectively. The increase in filtration rate is related to the growth of crystals upon
surfactants addition and formation of high volume percent of coarse crystals (>75 µm).
The size distribution of phosphogypsum crystals given in Tables 25 and 26 and Figure 15
shows that, the volume percentages of coarse crystals (>75 µm) are increased to about
15% and 24% upon addition of Crysmod and Hiflo-S5 surfactants, respectively while it is
as low as 10% without surfactant addition. Also, the mean of crystal size increases with
addition of surfactants.
The reaction efficiencies and P2O5 recoveries in presence and absence of

surfactants are given in Table 24. It is most interesting that even at high sulfate content,
both P2O5 recovery and reaction efficiency are higher with the addition of surfactants. The
increase in reaction efficiency is significant as it approaches 3.9% and 4.1% and the
increase in P2O5 recovery is 1.6% and 3.1% with Crysmod and Hiflo-S5 surfactants,
respectively.
Table 24. Filtration and Reaction Data at High Sulfate Content (3.2-3.5%).

Ton P2O5 /m2.day
Moisture % 27.9 28.4 28.7
P2O5 recovery, % 96.4 96.9 97.2
35
Table 25. Comparative Size Distribution of Gypsum Crystals (At High Sulfate
Content and After 3 Hr).

µm Mesh* Without With Crysmod With Hiflo-S5
125 120 99.91 99.70 98.20
106 140 99.01 97.75 93.79
90 170 96.15 92.97 86.34
75 200 89.76 84.77 75.64
63 230 81.11 76.14 65.20
53 270 71.31 68.25 56.13
45 325 62.10 61.57 49.16
38 400 53.26 55.06 43.31
10 - 17.96 20.07 19.09
* ASTM standards
Table 26. Statistics of Phosphogypsum Crystal Size Distribution (At High Sulfate
and After 3 Hr).
Item Phosphogypsum Crystal Mean,

Size Distribution, µm µm
*dx d10 d25 d50 d75 D90

Without surfactant 3.9 15.8 35.5 56.6 75.4 38.1
With Crysmod
surfactant 3.7 13.2 33.1 61.6 83.7 39.3
With Hiflo-S5
surfactant 3.8 15.5 46.0 74.3 96.9 48.2
36
(at High Sulfate Content and After 3 Hr).
Correlation of Crystal Size Distribution of Phosphogypsum and the Filtration Rate
The relationship between amount of phosphogypsum fines and the filtration rates
are given in Tables 27 and 28. These results show that, at each sulfate level, higher
filtration rates are obtained as the fine crystals (-10 Fm) decrease, and the mean and the
median of crystal size distribution increase.
37
Table 27. Relationship Between % Phosphogypsum Fines and Filtration Rate At
Low Sulfate Content.
Surfactant Phosphogypsum Mean Median, Specific Filtration

Addition Fines (-10 Fm), Diameter, µm Surface Rate, Ton
Technique % µm Area, P2O5/m2.
cm2/g Day
To RA before 83.3 5.31 3.41 29795 1.39
nucleation and
mixed for 30 min
With RA and 81.4 5.98 3.56 13544 2.57
added for 30 min
during the
nucleation
To RA before 68.9 9.33 5.41 8241 3.51
nucleation and
mixed for 5 min
Baseline 64.3 10.32 6.28 10453 4.24
(without
surfactant)
With water and 37.3 16.05 13.51 8464 4.56
added for 30 min
after addition of
phosphate
concentrate and
sulfuric acid
With water and 30.8 28.09 22.18 6599 5.54
added for 30 min
during the
nucleation
RA: Recycle acid
38
Table 28. Relationship Between % Phosphogypsum Fines and Filtration Rate at
Different Sulfate Contents.
Item Phosphogypsum Mean Median Specific Filtration Rate,

Fines (-10 Fm), Diameter, , µm Sur. Area, Ton P2O5/m2.
% µm cm2/g Day
Low Sulfate
Content:
Without surfactant 64.3 10.3 6.3 10453 4.24
With Crysmod 30.8 28.1 22.2 6599 5.54
With Hiflo-S5 9.0 54.5 57.5 1916 6.88
Medium Sulfate
Content:
With Crysmod 10.4 38.4 38.1 3995 6.42
With Hiflo-S5 13.9 44.4 43.5 6565 7.34
High Sulfate
Content:
With Crysmod 20.1 39.3 33.1 7389 8.33
With Hiflo-S5 19.1 48.2 46.0 6133 8.98
Gypsum Morphology
One of the most important factors affecting filtration rate is gypsum morphology
(size and shape of the crystals). For best filtration and washing, crystals of uniform sizes
are most desirable. At low sulfate content, shape of the crystals is changed from tabular
crystals to clusters by addition of the surfactants. On the other hand, at high sulfate
content large clusters are formed upon using the surfactants as shown in SEM
photomicrographs, Plates 1-8. These clusters are composed of many small crystals while
small clusters are formed without addition of surfactants. This may be due to adsorption of
surfactants on crystal surfaces rendering it hydrophobic. Thus, clusters may form due to
hydrophobic interactions between crystals. Similar findings have been reported in literature
(Sirianni and others 1969).
39
Plate 1. SEM Photomicrograph of Phosphogypsum without Surfactant
(Magnification 500).
Plate 2. SEM Photomicrograph of Phosphogypsum without Surfactant

40
Plate 3. SEM Photomicrograph of Phosphogypsum with Crysmod Surfactant

41
Plate 6. SEM Photomicrograph of Phosphogypsum with Hiflo-S5

Surfactant (Magnification 200).
42

43
BASIC STUDIES: PRIMARY NUCLEATION OF
CALCIUM SULFATE DIHYDRATE
Several investigators have studied primary nucleation of calcium sulfate dihydrate

using pure chemicals. However, no data have been reported under simulated industry
conditions. Therefore, these data cannot be extended to explain crystallization of
phosphogypsum. Thus, in our studies, pure chemicals, including phosphoric acid, are
reacted under industrially simulated conditions of temperature, P2O5 and free H2SO4 acid
concentrations. The primary nucleation of calcium sulfate dihydrate with and without
surfactant is followed by turbidity measurement. The primary nucleation occurs when
there are no solids in the medium and it continues until the solids content becomes as high
as 10% by volume as described by Becker (1989). In addition, the induction time (the time
elapsed between achievement of super-saturation by addition of reactants and appearance
of nuclei, turbidity) is measured.
The reaction that takes place can be written as::
CaH4(PO4)2•H2O + H2SO4 + H2O à 2H3PO4 + CaSO4•2H2O ↓
MATERIALS
Pure chemicals including phosphoric and sulfuric acids, and calcium hydrogen
phosphate monobasic (from Fisher Scientific) are used to prepare the following solutions:
Solution # 1: 27.5 % P2O5 and 2.5 % H2SO4

Solution # 2: 20% P2O5
Solution # 3: 32.5 % H2SO4
PROCEDURE
1. Add 500 ml of solution # 1 in 800 ml beaker and heat to 80° C using a constant
temperature water bath.
2. Then add simultaneously the following solutions (see Tables 29-30): Calcium
Hydrogen Phosphate Monobasic (which is prepared by dissolving the required
amount in 100 ml of solution #2 and the required amount of sulfuric acid, solution
#3, and 50 ml of deionized water or water/surfactant solution.
3. Keep reaction at 80° C with constant agitation.
4. Measure turbidity of the resulting suspension at different time intervals during the
course of the reaction using a HACH 2100A Turbidimeter.
5. Take samples (5 ml) from the pregnant liquor for particle size distribution analysis
after 1 minute, 5 minutes, 15 minutes and at the end of the experiment (when the
turbidity value exceeds 1000 NTU).
45
6. For size distribution measurements, disperse the samples in about 50 ml of
methanol and measure the size distribution using a Coulter Laser Size Analyzer.
Table 29. Amounts of Calcium Hydrogen Phosphate Monobasic and Sulfuric Acid
for Different Super-saturation.
Calcium Hydrogen 32.5% H2SO4 Acid, ml

Super-saturation Phosphate Monobasic, g
1.018 10 10
1.222 12 12
1.502 15 15
1.979 20 20
Table 30. Experimental Conditions for Primary Nucleation Studies.
Experiment Super- Surfactant Surfactant Amount of

No. saturation Used Concentration, ppm Surfactant, g
1 1.018 Without 0 0
2 1.018 Crysmod 100 0.8093
3 1.0186 Hiflo-S5 100 0.8093
4 1.222 Without 0 0
5 1.222 Crysmod 100 0.8137
6 1.222 Hiflo-S5 100 0.8137
7 1.502 Without 0 0
8 1.502 Crysmod 100 0.8203
9 1.502 Hiflo-S5 100 0.8203
10 1.979 Without 0 0
11 1.979 Crysmod 100 0.8314
12 1.979 Hiflo-S5 100 0.8314
13 1.502 Crysmod 50 0.4102
14 1.502 Hiflo-S5 50 0.4102
15 1.502 Crysmod 25 0.2051
16 1.502 Hiflo-S5 25 0.2051
17 1.502 Crysmod 12 0.0984
18 1.502 Hiflo-S5 12 0.0984
19 1.502 Crysmod 1.2 0.00984
20 1.502 Hiflo-S5 1.2 0.00984
46
TURBIDITY MEASUREMENTS AND ESTIMATION OF INDUCTION TIME (T)
As mentioned above, turbidity of the solution is measured at different time

intervals. A graph of time vs. turbidity is plotted in Figure 16. The time corresponding to
the point of intersection of the two asymptotic lines represent the induction time (see
example in Figure 16). In this particular figure, T = 130 min.
Induction Time of Calcium Sulfate Dihydrate with Crysmod Surfactant

Supersaturation 1.018
1000
900
800
700
Turbidity (NTU)
600
500
400
300
200
100
0
0 10 20 30 40 50 60 70 80 90 100 110 120 130 140 150
Time (min)
Figure (3). From the Figure Time = 130 min
Figure 16. Example of Turbidity Data Plotting and Induction Time Estimation.
CALCULATION OF CRYSTAL GROWTH EFFICIENCY (E)
Induction times are calculated as described above for the data collected in presence
and absence of surfactants. From the estimated values, crystal growth efficiency (E) is
calculated by the following equation as reported by Georgiev and others (1995):
E = (T0 – T1)
T0
where:
T0 = Induction time in absence of surfactant
T1 = Induction time in presence of surfactant
47
CALCULATION OF SUPER-SATURATION (S)
Super-saturation (S) is calculated (Tavare 1995) and listed in Table 31 as follows:
M.W. CaO
Mass CaO = Mass CaH4(PO4)2•H2O ×
M.W. CaH4(PO4)2•H2O
Mass CaO
% CaO = × 100
Total Mass
% CaO x M.W. CaSO4•2H2O
c = Calcium sulfate dihydrate concentration =
M.W. CaO
c* = Calcium sulfate dihydrate (solute) solubility at the given temperature = 0.83 (Becker,
1989)
c
S = Super-saturation =
c*
Table 31. Calculation of Super-saturation.
Calcium Phosphate % CaO C c* S

Monobasic, g
10 0.275 0.845 0.83 1.018
12 0.330 1.014 0.83 1.222
15 0.406 1.247 0.83 1.502
20 0.535 1.643 0.83 1.979
EFFECT OF SURFACTANTS ON CRYSTAL GROWTH (TURBIDITY) AT

DIFFERENT SUPER-SATURATION
Effect of surfactants on crystal growth (turbidity) at different super-saturation, is

given in Figs. 17-20. The surfactant concentration used is 100 ppm. It is clear that, both
Crysmod and Hiflo-S5 surfactants are crystal growth promoters, because in their presence
the solution turbidity reaches the highest value of 1000 NTU faster than the control
solution (without any surfactant).
48
1200
Without Surfactant
1000
Crysmod Surfactant
800
Turbidity (NTU)
Hiflo-S5
600
400
200
0
0 10 20 30 40 50 60 70 80 90 100 110 120 130 140 150 160
Time (min)
Figure 17. Effect of Time on Turbidity of Calcium Sulfate Dihydrate

(Super-saturation 1.018, 100 ppm surfactant)
1200
Without Surfactant
1000
Crysmod Surfactant
800 Hiflo-S5 Surfactant

Turbidity (NTU)
600
400
200
0
0 10 20 30 40 50 60 70 80 90 100
Time (min)

49
1200
Without Surfactant
1000
Crysmod Surfactant
Turbidity (NTU)
800
600 Hiflo-S5 Surfactant
400
200
0
0 5 10 15 20 25 30
Time (min)

1200
Without Surfactant
1000
Crysmod Surfactant
Hiflo-S5 Surfactant
800
Turbidity (NTU)
600
400
200
0
0 0.5 1 1.5 2 2.5 3 3.5 4 4.5 5 5.5 6 6.5
Time (min)

50
EFFECT OF SUPER-SATURATION ON INDUCTION TIME AND CRYSTAL
GROWTH EFFICIENCY WITH AND WITHOUT SURFACTANTS
Induction time and crystal growth efficiency are determined at different super-
saturation with and without surfactants and given in Table 32. The surfactant
concentration used is 100 ppm. These results confirm that Crysmod consistently lowers
the induction time to a greater degree than Hiflo-S5 when 100 ppm of surfactant is used at
different super-saturation. The induction time (determined from Figures 17-20) of the
solution without surfactant is always higher than that with Crysmod or Hiflo-S5
surfactant. In all three cases, as the super-saturation is increased, the induction time is
decreased.
Table 32. Effect of Super-saturation on Induction Time (T) and Growth Efficiency
(E) of Gypsum Crystals (Using 100 ppm Surfactant).
Super-saturation
Surfactant 1.018 1.222 1.502 1.979
T* E T E T E T E
None 140 - 80 - 24 - 2.9 -
Crysmod 130 0.071 76 0.050 17 0.292 1.70 0.414
Hiflo-S5 137 0.021 77 0.038 17 0.292 2.75 0.052
* in minutes
EFFECT OF SURFACTANTS’ CONCENTRATION ON INDUCTION TIME AND

CRYSTAL GROWTH EFFICIENCY
Effect of surfactants’ concentration on induction time and growth efficiency was

studied in the range of 1.2 to 100 ppm at constant super-saturation of 1.502. The results
are given in Table 33. It is clear that when the concentration of Hiflo-S5 is decreased from
50 ppm to 1.2 ppm, the induction time is decreased and the growth efficiency becomes
higher. This may suggest that Hiflo-S5, could be more efficient since less surfactant
concentration is required in the reacting system. The only exception to this trend is noticed
when 100 ppm of HiFlo-S5 surfactant is used. Crysmod does not show, however, a similar
trend with changing concentrations, possibly because Crysmod is actually composed of
several different surfactants. However, more work is needed to confirm these data since
the differences in induction time are small.
51
Table 33. Effect of Surfactant Concentration on Induction Time (T) and Growth
Efficiency (E) of Gypsum Crystals (Using 1.502 Super-saturation).
Surfactant Concentration, ppm

Surfactant 100 50 25 12 1.2
T* E T E T E T E T E
None 24 - 24 - 24 - 24 - 24 -
Crysmod 17 0.292 22 0.083 17 0.292 21 0.125 18 0.250
Hiflo-S5 17 0.292 23 0.042 22 0.083 21 0.125 17 0.292
* in minutes
CORRELATION BETWEEN SUPER-SATURATION AND INDUCTION TIME
The induction time without and with 100 ppm of surfactant is plotted as a function
of super-saturation, Figure 21. When the natural log of the induction time is plotted
against super-saturation in Figure 22, a linear relationship is observed. This confirmed that
the plot of Figure 8 is indeed exponential. Note that the induction time is inversely
proportional to the super-saturation. Also, Figure 22 shows that, at high supersaturation
the efficiency of the surfactants is higher. It may be inferred from these data that in the
presence of surfactants, retention times may be reduced leading to high throughputs.
However, this needs to be confirmed on plant scale testing.
140
W ithout Surfactant
120 Crysmod Surfactant
Hiflo-S5 Surfactant
Induction Time (min)
100
W ithout Surfactant
Crysmod Surfactant
80
Hiflo-S5 Surfactant
60
40
20
0
1 1.2 1.4 1.6 1.8 2
Supersaturation
Figure 21. Relation Between Super-saturation and Induction Time of

Calcium Sulfate Dihydrate.
52
6
Without Surfactant
Crysmod Surfactant
5
Hiflo-S5
Linear (Without Surfactant)

4
Linear (Crysmod Surfactant)
Linear (Hiflo-S5)
ln tind
2 Without surfactant: y= -4.1137x + 9.2753
With Crysmod: y = -4.6399x + 9.7748
1 With Hiflo-S5: y= -4.1849x + 9.2626
0
1 1.1 1.2 1.3 1.4 1.5 1.6 1.7 1.8 1.9 2
Super-saturation
Figure 22. Linear Relationship Between Super-saturation and Natural

Log of Induction Time for Calcium Sulfate Dihydrate.
EFFECT OF SURFACTANTS ON CRYSTAL SIZE DISTRIBUTION OF

GYPSUM
Gypsum crystal size distribution analysis at a super-saturation of 1.502 after 5

minutes is determined in the presence and absence of 100 ppm surfactants (Figure 23).
The results show that both surfactants increase the gypsum crystal sizes. According to
Table 34, the d50 of the formed crystals is 0.556 µm without surfactant while with
Crysmod and Hiflo-S5 surfactants the d50 values are 1.213 and 1.276 µm, respectively.
53
Figure 23. Differential Volume of Gypsum Crystals Obtained after 5 Minutes at a
Super-saturation of 1.502 (Surfactant Concentration = 100 ppm).
Table 34. Statistics of Size Distribution of Gypsum Crystals.
Item Gypsum Crystals Size Distribution, µm Mean Diameter,

µm
Dx d10 d25 d50 d75 d90
Without 0.157 0.256 0.556 1.806 2.872 1.144
Surfactant
With Crysmod 0.142 0.267 1.213 2.718 3.910 1.637
Surfactant
With Hiflo-S5 0.121 0.223 1.276 2.771 3.959 1.649
Surfactant
54
EFFECT OF SURFACTANTS ON GYPSUM CRYSTAL MORPHOLOGY
Gypsum crystal morphology was investigated without and with 100 ppm
surfactant. The photomicrographs are taken using an Olympus BX60F5 microscope. The
calcium sulfate dihydrate crystals (gypsum) in these photomicrographs are precipitated at
a super-saturation of 1.502. These photomicrographs confirm that Crysmod and Hiflo-S5
surfactants have significantly altered the morphology of the gypsum crystals. The crystals
shown in Plate 9, formed in the absence of surfactant, are needle-like shape with high
aspect ratio. Needle-like crystals give low filtration rate and maintain high amounts of
liquid. On the other hand, Plates 10 and 11 show crystals that are much larger in size and
tabular in shape with low aspect ratio. These crystals are expected to allow high filtration
rates. It is important to mention that the results of the fundamental studies confirm the
data obtained in the digestion work using phosphate concentrate.
Plate 9. Photomicrograph of Gypsum Crystals Formed Without Surfactant.
55
Plate 10. Photomicrograph of Gypsum Crystals Formed with Crysmod
Surfactant (100 ppm).
Plate 11. Photomicrograph of Gypsum Crystals Formed with Hiflo-S5

Surfactant (100 ppm).
56
COST-BENEFIT ANALYSES
For the cost-benefit analyses, the following criteria are used:
• Minimum increase in P2O5 recovery = 1.4%

• Crysmod surfactant consumption = 1.5Kg/ ton P2O5
• HiFlo -5S Surfactant consumption = 0.75Kg// ton P2O5
• Price of ton of P2O5 = $395.00
• Cost of 1.0 Kg. Of Crysmod = $3.00
• Cost of 1.0 Kg of Hi Flo-5S = $2.00
Thus, the following analyses is conducted:
1. In Case of Crysmod:
Cost* = $4.50/ton P2O5
Benefits:
- Gain in recovery
(1.4% @ $395.00/ton P2O5) = $5.53/ ton P2O5
- Increase in filtration rate = 25-30%
Net gain = $ 1.03/ ton P2O5

In addition to 25-30% increase in filtration rate
2. In Case of HiFlo-5S:
Cost = $1.5/ton P2O5
Benefits:
- Gain in recovery
(1.4% @ $395.00/ton P2O5) = $5.53/ ton P2O5
- Increase in filtration rate = 30-40%
Net gain = $ 4.03/ ton P2O5

In addition to 30-40% increase in filtration rate
_________________
*
Cost includes cost of surfactant at the delivery price as quoted by manufacturer. Feeding
and handling costs are neglected since only a feeding tank and dosing pump are needed for
this application.
57
CONCLUSIONS AND RECOMMENDATIONS
The major goal of this project is to enhance filtration of phosphogypsum using

cost-effective reagents. Clarification of the mechanisms controlling the effect of such
additives is another important objective in this study. In order to achieve these goals the
following tasks have been conducted:
1. Bench scale testing of effect of two different surfactants (Crysmod and HiFlo-S5)
on crystal modification and filtration of phosphogypsum produced from South
Florida phosphate concentrate containing high magnesium, at different sulfate
levels (low, medium and high);
2. Role of techniques and point of addition of surfactants; specific techniques involve

adding surfactants :
A) As a mixture with recycle acid or water, and
B) Before, during, or after nucleation
3. Cost-benefit analyses
4. Basic studies to elucidate the mechanisms controlling effect of surfactants under

different conditions
The data obtained in the above studies lead to the following conclusions:
• Surfactant addition technique is very important. Mixing surfactant with water as

well as its addition during reaction (nucleation and crystals growth) gives the
highest filtration rate.
• Effect of Crysmod and Hiflo-S5 surfactants on phosphogypsum filtration rate has

been investigated at different sulfate levels.
• The results indicate that up to 60% increase in filtration rate could be obtained
depending on type of surfactant and the sulfate content.
• The observed improvements are attributed to changes in crystal size
distribution as well as crystal shape. Surfactants result in crystals of narrower
size distribution with higher crystal diameter means than in the absence of
surfactant. In addition, morphology studies suggest that surfactants help in
forming larger clusters of fine crystals.
• Most importantly, surfactants are found to increase reaction efficiency and P2O5
recovery. The increase in reaction efficiency ranges between 1.2% to 4.1% and the
increase in P2O5 recovery ranges from 0.5% to 3.1% depending on surfactant type
and sulfate level.
59
• The fundamental study with pure chemicals indicates that:
• Surfactants shorten the induction time, which leads to increasing nucleation

and crystal growth.
• Crystal size analyses indicate that the two surfactants increase the mean
diameter of formed gypsum crystals.
• Gypsum crystals are needle-like without surfactant and of tabular shape with
lower aspect ratio with surfactants.
• Most importantly, the cost-benefit analyses suggest a gain of $1.03 / ton P2O5, if
Crysmod is used. However, the gain is up to $4.03/ ton P2O5 if HiFlo-S5 is used.
Such gain is in addition to the increase in filtration rates due to use of these
surfactants.
60
REFERENCES
Abdel-Aal EA. 1984. Comparative study on phosphoric acid production from Egyptian
phosphate ore concentrates by the wet process. M. Sc. thesis, Dept. of Chemistry,
Faculty of Science, Cairo University.
Abdel-Aal EA. 1989. Industrial simulation for wet process phosphoric acid production.
Ph.D. dissertation, Dept. of Chemistry, Faculty of Science, Cairo University.
Becker P. 1989. Phosphates and phosphoric acid: raw materials, technology and
economics of the wet process. New York: Marcel Decker Inc.
Botsaris G. 1982. Effects of impurities in crystallization process. Industrial

Crystallization 81: 109-116.
Budz J, Jones AG, Mullin JW. 1986. Effects of selected impurities on the continuous
precipitation of calcium sulphate (gypsum). J. Chem. Tech. Biotechnology 36:
153-161.
Cody AM, Cody RD. 1991. Crystal habit modifications of gypsum from epitaxial-like
adsorption of stereospecific growth inhibitors. Journal of Crystal Growth 113:
508-519.
Davey R. 1982. The role of additives in precipitation process. Industrial Crystallization

81:123-135.
Dyatlova NM, Evseev AM, Nikolaeva LS, Samakeav RKh. 1984. Simulating the
inhibition of calcium sulfate crystallization in the presence of EDTP. Zhurnal
Prikladnoi Khimii 57(10):2237-2240.
Georgiev GS, Djarova MD, Petkova JI, Georgiev GI, Koseva NS. 1995. Effect of poly
(acrylic acid-co-N-vinylpyrrolidone) on calcium oxalate crystallization. Polymer
Journal 27(4): 441-444.
Hasson D, Addai-Mensah J, Metcalfe J. 1990. Filterability of gypsum crystallized in

phosphoric acid solution in the presence of ionic impurities. Ind. Chem. Eng. 29:
867-875.
Hatakka H, Oinas P, Reunanen J, Palosaar S. 1997. Induction time and reaction kinetics
in batch precipitation of calcium sulfate. Acta Polytechnica Scandinavica,
Chemical Technology Series (244): 102-104.
He S, Oddo J, Tomson M. 1994. The inhibition of gypsum and barite nucleation in NaCl
brines at temperatures from 25 to 900. Applied Geochemistry 9:561-567.
61
Ibragimova AN, Ibragimov RN, Mezhidov VKh. 1986. Determination of the nucleation
parameters of calcium sulfate under frictional conditions. Zhurnal Prikladnoi
Khimii 59(5): 965-968.
Klima WF, Nancollas GH. 1987. The growth of gypsum. AIChE Symposium Series
83(253): 23-30.
Kopyleva BB, Reutovich LN. 1987. Effects of surfactants on the filtration of a

phosphogypsum suspension. Zhurnal Prikladnoi Khimii 60(5): 1033-1038.
Kopyleva BB. 1993. Solubility of phosphates in presence of surface active agents.

Proceedings of Conference on Beneficiation of Phosphate: Theory and Practice,
Society of Mining, Metallurgy and Exploration Inc., Littleton, Colorado.
Liu ST, Nancollas, GH. 1973a. Linear crystallization and induction-period studies of the
growth of calcium sulfate dihydrate crystals. Talanta 20: 211-216.
Liu ST, Nancollas GH. 1973b. The crystal growth of calcium sulfate dihydrate in the
presence of additives. Journal of Colloid and Interface Science 44(3): 422-429.
Liu ST, and Nancollas GH. 1975. A kinetic and morphological study of the seeded
growth of calcium sulfate dihydrate in the presence of additives. Journal of
Colloid and Interface Science 52(3): 593-601.
MacCartney ER, Alexander AE. 1958. The effect of additives upon the process of
crystallization: I. Crystallization of calcium sulfate. Journal of Colloid Science 13:
383-396.
Martynowicz ETMJ, Van Rosmalen GM, Witkamp GJ. 1996. The effect of aluminum
fluoride on the formation of calcium sulfate hydrates. Hydrometallurgy 41:171-
186.
Moudgil Brij M. 1995. Enhanced filtration of phosphogypsum. Final report submitted

to Florida Institute of Phosphate Research, Bartow, FL.
Rocha SDF, Ciminalli VST. 1995. Effect of surfactants on calcium sulfate crystallization
in phosphoric acid solution. Minerals and Metallurgical Processing 12(3): 166-
171.
Sarig S, Kahana F, and Leshem R. 1975. Selection of threshold agents for calcium
sulfate scale control on the basis of chemical structure. Desalination 17: 215-229.
Schroeder J, Skudlarska W, Szczepanik A, Sikorska E, Zielinski S. 1976. The influence

of surface-active agents on gypsum crystallization in phosphoric acid solutions.
Ind. Cryst., 6th Proc. Symp., 1975. p 263-268.
62
Shoulian Z. 1996. Modification of crystal size distribution for enhanced filtration of
phosphogypsum using poly(ethylene oxide). Ph.D. dissertation, Dept. of
Materials Science and Engineering, University of Florida, Gainesville, FL.
Sirianni AF, Paillard G, Puddington IE. 1969. Canadian Journal of Chemical

Engineering 47: 210.
Smith BR, Alexander AE. 1970. The effect of additives on the process of crystallization
II. Further studies on calcium sulfate. Journal of Colloid and Interface Science
34(1): 81-90.
Smith BR, Sweett F. 1971. The crystallization of calcium sulfate dihydrate. Journal of
Colloid and Interface Science 37(3): 612-618.
Tavare, NS. 1995. Industrial crystallization: process simulation analysis and design.
The Plenum Chemical Engineering Series. New York: Plenum Press,
p 14.
Taylor JF. 1989. The effect of polymeric additives on the filtration characteristics of
gypsum. M. Sc. thesis, Dept. of Materials Science and Engineering, University of
Florida, Gainesville, FL.
Torocheshnikov NS and others. 1986. Continuous recrystallization of calcium sulfate

dihydrate to the hemihydrate in the phosphoric acid wet process. Journal of
Applied Chemistry of the USSR 59(61): 1109-1113.
Van Rosmalen GM, Daudey PJ, Marchee WGJ. 1982. Quantitative description of the
influence of the inhibitor concentration on the growth rate of calcium sulfate.
Industrial Crystallization 81: 147-153.
Van Rosmalen GM, Weijnen MPC. 1984. The role of additives and impurities in the
crystallization of gypsum. Industrial Crystallization 84: 61-66.
Van Rosmalen GM, Witkamp GJ, de Vreugd CH. 1987. Additive and impurity effects in
the crystallization process. Industrial Crystallization 87: 21-25.
Witkamp GJ, Van der Eerden JP, Van Rosmalen GM. 1990. Growth of gypsum.
Journal of Crystal Growth 102: 281-289.
63

01 141 162final

Uploaded by

Document Information

Original Title

Copyright

Available Formats

Share this document

Share or Embed Document

Sharing Options

Did you find this document useful?

Is this content inappropriate?

Copyright:

Available Formats

01 141 162final

Uploaded by

Copyright:

Available Formats

The Florida Institute of Phosphate Research was created in 1978 by the Florida Legislature (Chapter

G. Michael Lloyd, Jr. -Chemical Processing

Florida Institute of Phosphate Research

Hassan El-Shall, Brij M. Moudgil and

Department of Materials Science and Engineering

FLORIDA INSTITUTE OF PHOSPHATE RESEARCH

Contract Manager: G. Michael Lloyd, Jr.

The manufacture of phosphoric acid by the wet process is often characterized as a

Controlling and/or modifying the phosphogypsum crystals has been a long-time

Phosphoric acid is an important intermediate product for production of fertilizers.

The results suggest the following:

• Optimum filtration results could be obtained by addition of surfactant as a mixture

The Florida Institute of Phosphate Research (FIPR) is acknowledged for

EXECUTIVE SUMMARY .....................................................................................1

Materials Characterization ..................................................................................9

Estimation of Filtration Rate ......................................................................10

RESULTS AND DISCUSSION............................................................................17

Study of Surfactant Addition Techniques.........................................................17

Addition of Crysmod Surfactant Before Nucleation....................................17

Testing of Surfactants at Different Sulfate Contents.........................................25

Testing at Low Sulfate Content (1.5-1.8%)................................................25

Washing With Simulated Pond Water....................................................30

Testing at High Sulfate Content (3.2-3.5%) ...............................................35

BASIC STUDIES: PRIMARY NUCLEATION OF

CONCLUSIONS AND RECOMMENDATIONS .................................................59

1. Simple dihydrate process flow-sheet........................................................ 7

2. Volume % of fine (-10 microns) phosphogypsum with and without

3. Effect of surfactant/recycle acid mixture addition before

4. Volume % of fine (-10 microns) phosphogypsum crystals with

5. Effect of surfactant addition during nucleation

6. Volume % (–10 microns) of phosphogypsum crystals with

7. Effect of surfactant addition after nucleation

8. Comparative size distribution of phosphogypsum crystals

9. Comparative size distribution of phosphogypsum crystals

10. Volume % (-10 µm) of phosphogypsum crystals

11. Volume % (-38 µm) of phosphogypsum crystals

12. Comparative size distribution of phosphogypsum crystals

13. Comparative size distribution of phosphogypsum crystals

14. Comparative size distribution of phosphogypsum crystals

15. Comparative size distribution of phosphogypsum crystals

16. Example of turbidity data plotting and induction time

17. Effect of time on turbidity of calcium sulfate dihydrate

18. Effect of time on turbidity of calcium sulfate dihydrate

19. Effect of time on turbidity of calcium sulfate dihydrate

20. Effect of time on turbidity of calcium sulfate dihydrate

21. Relation between super-saturation and induction

22. Linear relationship between super-saturation and natural log of

23. Differential volume at a super-saturation of 1.502 after

1. Effect of Adding Crysmod Surfactant/Recycle Acid Mixture Before

2. Effect of Crysmod Surfactant Addition during Nucleation on

3. Effect of Crysmod Surfactant Addition after Nucleation on

4. Filtration and Reaction Data at Low Sulfate Content (1.5-1.8%) ............. 3

5. Filtration and Reaction Data at Medium Sulfate Content (2.2-2.4%) ....... 4

6. Filtration and Reaction Data at High Sulfate Content (3.2-3.5%)............. 5

7. Chemical Analysis of High Dolomitic Phosphate Concentrate Sample...... 9

8. Sieve Analysis of High Dolomitic Phosphate Concentrate Sample ........... 9

9. Effect of Adding Surfactant / Recycle Acid Mixture Before

10. Effect of Adding Surfactant / Recycle Acid Mixture Before

11. Effect of Surfactant Addition during Nucleation on Phosphogypsum

12. Effect of Surfactant Addition during Nucleation on Reaction

13. Effect of Surfactant Addition After Nucleation on Phosphogypsum

14. Effect of Surfactant Addition After Nucleation on Reaction

15. Filtration and Reaction Data at Low Sulfate Content (1.5-1.8%)........... 26

Reaction efficiency = 100 - 94(A - B)C/(D*E)

P2O5 recovery = 100 - (94AC)/(D*E)

Washing efficiency = 100 - (94BC)/(D*E)