ARTICLE IN PRESS

Journal of

Trace Elements
in Medicine and Biology
Journal of Trace Elements in Medicine and Biology 20 (2006) 143153 www.elsevier.de/jtemb

BIOAVAILABILITY

Trace elements in rock salt and their bioavailability estimated from solubility in acid$
Georg Steinhauser, Johannes H. Sterba, Karin Poljanc, Max Bichler, Karl Buchtela
sterreichischen Universita Vienna University of Technology, Atominstitut der O ten, Stadionallee 2, 1020 Vienna, Austria Received 11 January 2006; accepted 8 June 2006

Summary
In this study, 18 partly commercially available samples of rock salt from Austria, Germany, Pakistan, Poland, Switzerland, and Ukraine were investigated with respect to their content of trace elements using instrumental neutron activation analysis. Elements detected were Al, Ba, Br, Ca, Ce, Cl, Co, Cr, Cs, Eu, Fe, Hf, La, Mn, Na, Rb, Sb, Sc, Sm, Sr, Ta, Tb, Th, and Zn, some of them only in individual cases. An estimation of the bioavailability of these trace elements was performed by dissolving an equivalent of the sodium chloride samples in diluted hydrochloric acid (simulating stomach acid), ltering off the insoluble components, and analyzing the evaporated ltrate. It could be shown that in most cases bioactive trace elements like Fe can be found in rock salt in the form of almost insoluble compounds and are therefore not signicantly bioavailable, whereas thorium, for example, was partly bioavailable in two cases. A signicant contribution to the recommended daily intake of metal trace elements by using rock salt for nutrition can be excluded. r 2006 Elsevier GmbH. All rights reserved.
Keywords: Rock salt; Nutrition; Neutron activation analysis; Sodium chloride; Bremsstrahlung spectrum

Introduction
During the last years, the use of natural or organic rock salt for nutrition purposes has become more and more popular in Central Europe. Therefore, rock salt from various more or less exotic regions of the world can be obtained almost everywhere in Europe. In many cases, a high content of trace elements which has a highly positive connotation in everyday language is advertised, reasoning positive effects on human (and animal) health. Only in exceptional cases, a detailed
$

Corresponding author. Tel.: +43 1 58801 14189;

Reverentially dedicated to Professor Helmut Rauch.

fax: +43 1 58801 14199. E-mail address: georg.steinhauser@ati.ac.at (G. Steinhauser). 0946-672X/$ - see front matter r 2006 Elsevier GmbH. All rights reserved. doi:10.1016/j.jtemb.2006.06.001

declaration of the respective trace elements and their concentrations can be found for the respective salt. In 2002, Javor [1] lamented the lack of routinely published trace element analyses of rock salt. Detailed reports on major element impurities like e.g., Ca2+, Mg2+, K+, and SO2 4 in salt can often be found [2,3]. Therefore the aim of this study was to apply instrumental neutron activation analysis (INAA) to the determination of trace elements in rock salt samples from different countries. Although INAA is a wellproven analytical tool for bulk analyses of geological samples [4], it has been used for the investigation of rock salt only in a few cases before [59]. Other methods for the determination of trace elements in rock salt or sodium chloride brine, respectively, like inductively coupled plasma-atomic emission spectroscopy

ARTICLE IN PRESS
144 G. Steinhauser et al. / Journal of Trace Elements in Medicine and Biology 20 (2006) 143153

(ICP-AES), atomic absorption spectroscopy (AAS) or total reexion X-ray uorescence (TXRF) require extensive sample preparation because of interferences with the sodium chloride matrix [10,11]. Since rock salt is one of the most important inorganic raw materials [12], the question of impurities in sodium chloride is essential. Moreover, another aim of this study was to indicate biological effects of the trace elements in salt. To the authors knowledge no study on the bioavailability of the rock salts constituents has been performed yet. Therefore, it had to be shown which trace elements in rock salt have an impact on the recommended daily intake of trace elements with regard to the typical consumption of salt. In some rare cases, esoteric aspects like salt-stored biophotons or salt-stored bioenergetic vibrations are used to praise benecial health effects of rock salt. Since such arguments lack any scientic fundament, they appear to be a questionable marketing strategy at best, taking advantage of the customers nescience. These aspects or other scientically not veriable effects are therefore not topic of this study.

Materials and methods

Sample materials
In this study, 18 partly commercially available samples of rock salt from Austria, Germany, Pakistan, Poland, Switzerland, and Ukraine were investigated. To check the performance of modern brine purication, one puried and evaporated salt from Pakistan was analyzed, too. The samples and their respective origin are presented in Table 1.

ingested particles or compounds to the body, their solubility in stomach acid is a highly signicant indicator [13]. Thus, 3 g of each homogenized rock salt sample (except samples I and N, which consisted of only one crystal of a few grams), were dissolved in diluted hydrochloric acid with an equivalent concentration to stomach acid when stimulated ([H+] approximately 100 mmol l1, see [14]), stirred for 5 min at approximately 37 1C, and ltered twice using Schleicher & SchuellTM red and blue ribbon lters. The ltrate was evaporated in a drying oven; the resulting treated salt was homogenized again by grinding. It was concluded that trace elements found in the original sample but not in the evaporated ltrate can hardly be bioavailable. Because of the formation of volatile hydrogen bromide, bromide concentrations were ignored in the interpretation of the analytical results of treated salt. Unfortunately, the steam of hydrochloric acid led to corrosion of the drying oven. Some of the treated samples were contaminated with corroded metal particles and thus contained more chromium and/or zinc than the untreated rock salt sample. Therefore chromium and zinc concentrations of the treated sample were ignored for analytical discussion with all samples of treated salt. Nevertheless, the concentrations of all the other trace elements of the treated salt are reasonable. A detailed summary of trace elements in nutrition and their biological function can be found in [15] and [16].

Instrumental neutron activation analysis

For INAA, about 50100 mg of each sample were sealed into SuprasilTM quartz glass vials. Blank values for these irradiation vials were measured and found negligible. The samples were irradiated together with international certied reference materials for approximately 40 h in the central irradiation thimble of the TRIGA Mark II-reactor of the Atominstitut der sterreichischen Universita O ten, Vienna, Austria, at a thermal neutron ux of about 1013 cm2 s1. The multielement standards used for quantitative analysis were CANMET reference soil SO1, NIST SRM 1633b coal y ash, light sandy soil BCR no. 142, MC Rhyolite GBW 07113, and NIST SRM 2702 inorganics in marine sediment. Additionally, sodium chloride p.a. was used as a standard material for quantication of sodium and chloride. The quartz glass sample vials were decontaminated and packed into PE vials tting the automatic sample changer of the Atominstitut in Vienna. The automatic sample changer was adjusted to a xed measurement position at a distance of 4 cm beside the detector.

Sample preparation
Neutron activation analysis does in principle not require extensive sample preparation, but as in any bulk analysis, homogenization is a prerequisite. Sample preparation followed the same procedure for every sample. About 50 g (or more, if available) of the rock salt sample was crushed with exclusively polyethylene (PE) tools to avoid any contamination by abraded metal particles. Homogenization by grinding in an agate mortar to grain sizes o3 mm was followed by drying to constant weight at 110 1C.

Trace element bioavailability

For the estimation of the bioavailability of trace elements in the rock salt samples, the following simplied approach was used: For the availability of

Table 1.
Country/location Pakistan Altaussee/Austria Altaussee/Austria Ukraine Pakistan Pakistan Pakistan No No No No No No No Yes Coarse crystals Part of a rock salt drill core No Coarse crystals No Veda, Ottensheim Free sample offered by Salinen Austria AG Free sample offered by Salinen Austria AG Yes Yes Yes Sonnentor GmbH, Spro gnitz Yes Fine salt crystals Light pink Source Purchasable? Sample description (shape) Color

Sample materials and their characteristics

Sample code

Sample name

B C D

Ayurvedisches Zaubersalz Weisses Gold Mittelkorn Bergkern Steinsalz Ukraine Karpatensalz White rock salt Coarse crystals Rock salt lumps Coarse salt crystals with some dark-colored stones Coarse crystals Pink White

Light pink Dark pink Light gray to light blue or green White

Pink rock salt

Free sample offered by HUB-PAK Salt Renery Free sample offered by HUB-PAK Salt Renery Energena, Graz

Energena Himalaya Kristallsalz Granulat Rock salt Riburg Saline Riburg near Mo hlin/ Switzerland Wielicka/Poland Personal geological collection of Max Bichler Personal geological collection of Max Bichler Personal geological collection of Max Bichler Personal geological collection of Max Bichler Personal geological collection of Max Bichler Free sample offered by Vereinigte Schweizerische Rheinsalinen Personal geological collection of Max Bichler

White to gray Colorless and clear Pink to orange Dark gray

Wielicka salt

J K

Hallstatt Fasersalz Springerwerk dunkelgrau

One crystal with an edge length of ca. 1 cm Rock salt lumps Rock salt lumps Rock salt lumps

Springerwerk rot

Dark pink to red

Springerwerk gelb

ARTICLE IN PRESS

Tonwu rfelsalz

Hallstatt/Austria Altaussee/Austria salt mine Springerwerk Altaussee/Austria salt mine Springerwerk Altaussee/Austria salt mine Springerwerk Hallstatt/Austria

Clear almost square-edged crystal lumps One single crystal grown in a clay matrix No Lumps of large salt crystals interbedded with clay

Stetten/Germany

Free sample offered by Wacker-Chemie GmbH

Colorless to pale yellow Clear and colorless with some clay inclusions Gray salt with dark gray clay inclusions

Stetten/Germany

Free sample offered by Wacker-Chemie GmbH Free sample offered by Wacker-Chemie GmbH

No

Lumps of large salt crystals interbedded with clay No Lumps of large salt crystals interbedded with clay, containing parts of a clear salt lense No Yes Large grown crystals Fine crystals

G. Steinhauser et al. / Journal of Trace Elements in Medicine and Biology 20 (2006) 143153

Stetten/Germany

Light gray salt with dark gray clay inclusions Pale yellow to gray

R Pakistan

Stetten/Germany

Clear and colorless White

Wacker-Chemie Tonzwickelsalz (derb), Wetterstrecke/70 West; Mitte Wacker-Chemie Tonsalz, Wetterstrecke/147 West; aus dem Hangenden (oben) Wacker-Chemie Tonzwickelsalz; (grobspa tig gelblich), mit Teilen einer Salzlinse; Wetterstrecke/ 110 Ost; Mitte Wacker-Chemie part of a salt lense from Q Rened salt A

Free sample offered by Wacker-Chemie GmbH Free sample offered by HUB-PAK Salt Renery

145

ARTICLE IN PRESS
146 G. Steinhauser et al. / Journal of Trace Elements in Medicine and Biology 20 (2006) 143153

After a decay time of 7 d, a rst g-spectrum was measured to obtain the activities of the short- and medium-lived activation products 24Na, 82Br, 140La, and 153 Sm. Three weeks later, a second measurement sequence was started to detect the long-lived activation products of Ba, Ce, Co, Cr, Cs, Eu, Fe, Hf, Rb, Sb, Sc, Ta, Tb, Th, and Zn (see Table 2). The measuring times were 1800 s and 20 000 s, respectively. For analysis of the activation products 28Al, 38Cl, 49Ca, and 56Mn, short time activation analysis was performed: About 80 mg of the samples and the standard materials described above were weighed into PE vials and irradiated using a pneumatic transfer system (neutron ux approx. 2 1012 cm2 s1). Irradiation time was 30 s, after a decay time of 4 min the rst measurement (measuring time 300 s) was started for determination of aluminum and calcium. The samples were irradiated and measured sequentially in this case. Approximately 4 h later, a second measurement (duration: 1800 s) was started for the determination of manganese.
Table 2. Elements, activation products, half-lives, and gphoton energies (sorted by measurement sequences) Element Activation product Half-life (T1/2) 2.2414 min 8.718 min 37.24 min 2.5789 h 14.951 h g-Energy (keV) 1778.8 3084.4 2167.5 1811.2 1368.6; 2754.1

Table 3. Reference materials used for quantication of certain element concentrations Element Al Ba Br Ca Ce Cl Co Cr Cs Eu Fe Hf La Mn Na Rb Sb Sc Sm Sr Ta Tb Th Zn SO1 | | | | | | | | | | | LSS | | | | | | | | | | | | | | | | | | | | | GBW | | | | CFA | | | | | | | | | | | | | | | | | | | IMS | | NaCl p.a.

| | | | | | | | | | | | |

| | | | | |

| | | |

| | | | | |

| |

| |

Short time activation analysis 28 Al Al 49 Ca Ca 38 Cl Cl 56 Mn Mn 24 Na Na

| reference materials used for quantitative analysis. SO1 CANMET reference soil SO-1; LSS light sandy soil BCR No. 142; GBW MC rhyolite GBW 07113; CFA NIST SRM 1633b Coal y ash; IMS NIST SRM 2702 Inorganics in Marine Sediment.

Measurement 7 days after long-term neutron irradiation 24 Na 14.951 h 1368.6; 2754.1 Na 82 Br Br 35.282 h 776.5 140 La La 1.6781 d 1596.2 153 Sm Sm 46.284 h 103.2 Measurement 28 days after long-term neutron irradiation 131 Ba 11.50 d 496.3 Ba 141 Ce Ce 32.508 d 145.5 60 Co Co 1925.28 d 1173.2 51 Cr Cr 27.7025 d 320.1 134 Cs Cs 2.0652 a 795.8 152 Eu Eu 13.506 a 1408.1 59 Fe Fe 44.495 d 1099.2 181 Hf Hf 42.39 d 482.2 86 Rb Rb 18.642 d 1076.6 124 Sb Sb 60.11 d 1691.0 46 Sc Sc 83.79 d 1120.5 85 Sr Sr 64.84 d 514.0 182 Ta Ta 114.43 d 1221.3; 1189.1 160 Tb Tb 72.3 d 879.4 233 Th Paa 26.975 d 311.9 65 Zn Zn 243.66 d 1115.5
Data on half-lives are taken from [29]. a 233 Pa is the decay product of 233Th (T1/2 22.3 min), which is the activation product of 232Th, and can be detected more sensitively.

The g-spectrometry was performed with a 151 cm3 HPGe detector (1.8 keV resolution at the 1332 keV 60Co peak; 50.1% relative efciency), connected to a PCbased multi-channel analyzer with preloaded lter and loss-free counting system [17,18]. The energies and standards used for quantitative analysis are given in Table 2 and 3.

Analytical problems
Sodium chloride is a non-trivial matrix for INAA because of some analytical problems. In contrary to typical geological samples with silicate or calcium carbonate matrices, where INAA can be applied without major troubles [19], sodium chloride causes a high photon-background. Since 23Na (natural occurrence 100%) and 37Cl (natural occurrence 24.23%) can easily be activated (total cross-sections for the capture of thermal neutrons are 0.53 b for 23Na(n,g)24Na and 0.24 b for 37Cl(n,g)38Cl, [20]), the g-background of their activation products complicates the detection of radionuclides with similar half-lives.

ARTICLE IN PRESS
G. Steinhauser et al. / Journal of Trace Elements in Medicine and Biology 20 (2006) 143153 147

1.0E+07

1.0E+06

1.0E+05

counts

1.0E+04

1.0E+03

1.0E+02

1.0E+01
Th) 312keV

1.0E+07

1.0E+06

1.0E+05

counts

1.0E+04

1.0E+03

1.0E+02

1.0E+01 0 200 400 600 800 Energy [keV] 1000 1200 1400 1600

Fig. 1. Comparison of the g-spectra of a distinctive salt sample (P) on the top and a typical geological sample (in this case coal y ash SRM) on the bottom, showing the characteristic bremsstrahlung-background with the salt sample originating mainly from 35 Cl(n,a)32P.

Pa (from

K 1461keV (natural)

Ta 1189, 1221keV

Rb 1077keV

Co 1173keV

Co 1333keV

Sc 1120keV

Fe 1099keV

Fe 1291keV

Eu 1408keV

Cs 605keV

Cs 796keV

Sr 514keV

Sc 889keV

Cr 320keV

ARTICLE IN PRESS
148 G. Steinhauser et al. / Journal of Trace Elements in Medicine and Biology 20 (2006) 143153

Short time activation for the determination of shortlived radionuclides is mainly disturbed by high 38Cl activities (T1/2 37.2 min). Therefore the detection limits of e.g., 28Al (T1/2 2.2 min), 27Mg (T1/2 9.5 min), 49Ca (T1/2 8.7 min), and 128I (T1/2 25.0 min) are much higher than in other geological matrices. Although INAA is one of the most sensitive analytical methods for the determination of iodine, iodide concentrations were lower than the detection limit in every single case of this study. For determination of the content of iodine in salt using neutron activation analysis, the authors suggest to use epithermal neutrons by means of a cadmium shielding, because of an advantageous Cd ratio of 2.6 for iodine [21]. However, wet chemical determination of iodide and iodate in salt is performed routinely by the saltprocessing industry. Aim of this study was to concentrate on metal traces in salt, which are determined only in exceptional cases. Values for Mg can hardly be obtained using INAA, because of the nuclear reaction 27Al(n,p)27Mg (sn,p 0.0043 b) and the wide-ranging aluminum contents of the reference materials used. A few hours after activation, the 38Cl activities are getting more and more irrelevant with respect to the analytical problems mentioned, whereas the 24Na activities (T1/2 15.0 h) increase in their signicance. The high 24 Na activity thwarts the determination of the radionuclides 56Mn (T1/2 2.6 h) and 165Dy (T1/2 2.3 h) in most cases. Only in exceptional cases concentrations of manganese could be determined, whereas dysprosium could not be detected. Additionally to the 24Na activities, rock salt usually has a high content of bromide of approximately some 10100 mg kg1, which causes a high g-background as well, because of the activation product 82Br (T1/2 35.3 h). Therefore, the measurement after 1 week (see Table 2) cannot be regarded as completely undisturbed. It was supposed that 3 or 4 weeks after irradiation, the g-background would be low enough to sensitively detect the long-lived activation products (see Table 2). Nevertheless, a high bremsstrahlung-background caused by the b-emitters 32P and 36Cl, produced by 35Cl(n,a)32P (sn,a 0.00008 b) and 35Cl(n,g)36Cl (sn,g 43.7 b), could be observed and lead to a bremsstrahlung spectrum (see Fig. 1 for comparison of such a spectrum with a typical gspectrum of the coal y ash standard used in this study). This bremsstrahlung caused a much higher count rate than one would have expected. However, due to the loss-free counting system, this photon-background can be regarded as hardly troublesome for the determination of the longlived radionuclides. The neutron activation analyzer [22] of the Atominstitut was applied to comparative quantication of elements, which produce short-lived radionuclides (especially 28Al and 49Ca). Results obtained were comparable to those obtained by the slow pneumatic

Table 4. Element Al Ca Sc Cr Mn Zn Fe Co Br Sr Rb Sb Cs Ba La Ce Sm Eu Tb Hf Ta Th

Detection limits of the elements analyzed Detection limit 6 104 2 106 0.2 300 7 104 100 3100 3 310 2350 150 4 4 5600 270 280 40 1 2 11 1 15

All values in ppb (mg/kg).

transfer system (produced by General AtomicTM). Although the neutron activation analyzer in principle can perfectly be used for the determination of 20F, in this case the sodium chloride matrix inhibited any information on the uoride concentrations because of the nuclear reaction 23Na(n,a)20F (sn,a 0.0006 b). A summary of all detection limits (within a 3 s range) in this study can be found in Table 4. Detection limits were calculated by comparison of the sample Wacker Tonsalz (P), containing almost all trace elements detected, and the treated sample of Springerwerk rot (treated salt L) where almost no trace elements could be detected.

Results
INAA results are presented in Table 5. Additionally to the rock salt samples, one sample of puried and evaporated salt was analyzed in this study to show the efciency of modern brine purication [2327]. After each rock salt sample, the results of the treated salt samples are listed. Contents of sodium and chloride in the treated salt were not plotted in order to keep the table concise and because it can be assumed that both sodium and chloride are fully bioavailable. Cobalt, for example, is only bioavailable to an extent of approximately 10%, as shown by the samples from Stetten.

Table 5.
% of RDI 412 412 413 413 398 406 405 416 409 404 212 211 0.01 7 5 0.05 1 0.5 6 0.04 2 410 3 395 410 424 384 401 407 442 439 445 443 1010 1 3360 2 403 409 404 62.8 o1 63.0 o1 62.2 o1 62.6 o1 59.2 61.9 o1 o1 418 437 935 3050 6 2 65.2 o1 460 1.2 7 7 1.06 3.77 4.02 1.29 7 5 2 5 6212 23754 24804 8258 62.9 o1 444 61.0 o1 431 63.3 62.1 o1 o1 447 438 63.9 o1 451 62.5 o1 441 62.6 o1 442 79 410 0.02 1 62.5 o1 441 0.8 410 5 64.1 o1 452 0.01 2 63.4 o1 448 64.1 o1 452 63.3 o1 447 0.01 3 Cl Err. % % of RDI Al Err. % Sc Err. % Ca Err. % % of RDI Cr Err. % % of RDI

Results of the INAA

Sample code

Na

Err. %

37.8

o1

salt A

37.8

o1

salt B

37.9

o1

salt C

37.9

o1

salt D

36.5

o1

salt E

37.2

o1

salt F

37.1

o1

salt G

38.1

o1

salt H

37.5 37.0

o1 o1

salt J

36.2

o1

salt K

37.6

o1

salt L

38.9

o1

salt M

35.2 36.8

o1 o1

ARTICLE IN PRESS

salt O

37.3

o1

salt P

36.9

o1

0.16 0.61 0.05 0.64 0.02 0.19

o1 o1 2 o1 3 1

salt Q

37.5

o1

A Treated B Treated C Treated D Treated E Treated F Treated G Treated H Treated I J Treated K Treated L Treated M Treated N O Treated P Treated Q Treated R Treated S

salt R

37.0

o1

Sample code

Mn

Err. %

% of RDI

Fe

Err. %

% of RDI

Co

Err. %

Zn

Err. %

% of RDI 0.28 0.35 410 410

Br

Err. %

Rb

Err. %

Sr

Err. %

G. Steinhauser et al. / Journal of Trace Elements in Medicine and Biology 20 (2006) 143153

0.020.03 17.1 0.020.04 47.4 36.0 34.1 38.2

2 1 1 o1

salt A

0.02 0.01 0.01

8 410 10

salt B

salt C

15 15 41 23 83 56 18 0.02 0.01 8 410 1

5 4 2 3 2 2 4

salt D

33 9 22 5 54 12 68 11

410 9 410 410 3 410 7 410

1 0 1 0 2 0 3 0

A Treated B Treated C Treated D Treated E Treated F 80 5 3 0.03

salt E 410 0.32 10 0.020.04 25.6 1

262 35

2 3

149

150

Table 5. (continued ) Fe Err. % % of RDI 0 0.01 6.1 2 410 28.6 o1 14 5 Co Err. % Zn Err. % % of RDI Br Err. % Rb Err. % Sr Err. %

Sample code

Mn

Err. %

% of RDI 6 410

salt F

salt G

salt H 0.38 0.33 0.38 97.2 0.63 2.68 3.83 1.00 14.0 73.4 5 0.060.12 7.3 2 0.230.46 12.9 1 1 2 1 23 7 10 3 0.040.08 48.2 0.160.32 8.5 1 10 1.82 8.06 7.26 2.48 1 5 6 2 3 15 6 410 0.020.05 77.0 1 410 0.020.04 42.7 1 0.66 8 10 2.2 0.020.05 49.7 4 1

salt J

0.01 0.03 0.01 0.07 0.02 410

410 5 410 9

salt K

9 144 9 195 10 22 6 410 0

410 4 410 3 410 6

0 6 0 8 0 1

142 80

2 1

salt L

salt M

42 207

410 5

513 2562

salt O

161

410

1948

salt P

1214 1729 143 1749 39 562

1 1 4 1 4 2

49 69 6 70 2 22

513 55

1 2

salt Q

0.08 0.67 0.08 0.75 0.07 0.25 0.02

3 2 3 1 410 3 9

Treated G Treated H Treated I J Treated K Treated L Treated M Treated N O Treated P Treated Q Treated R Treated S

salt R

ARTICLE IN PRESS

Sample code

Sb

Err. %

Cs

Err. %

Ba

Err. %

La

Err. %

Ce

Err. %

Sm

Err. %

Eu

Err. %

Tb

Err. %

Hf

Err. %

Ta

Err. %

Th

Err. %

salt A

0.01

salt B 7 410

salt C

salt D

G. Steinhauser et al. / Journal of Trace Elements in Medicine and Biology 20 (2006) 143153

salt E 8 0.02 410

0.02

410

0.01

salt F

salt G

A Treated B Treated C Treated D Treated E Treated F Treated G Treated H Treated I J 3

salt H 0.004 410

0.03

410

0.06

salt J 0.004 10 0.004 410 0.05 410

0.03 410

0.01 0.04

10 3

salt K

0.02

salt L

salt M 24 1.6 33 0.6 410 0.91 7 0.08 10 0.08 410 0.016 410 1.3 410 2.14 3 0.10 4 0.035 10 4 5 410 0.72 3.43 0.26 0.12 0.19 3 410 5 410 2 0.05 0.18 410 410 0.014 0.048 6 10

0.17 0.05

6 5

salt O

0.09

salt P

0.04

410

0.013 5 0.056 5 0.023 4 0.045 6 0.014 5 0.017 10

0.01 0.05 0.01 0.03 0.01 0.01

410 410 7 410 410 410

0.15 0.56 0.09 0.47 0.07 0.24

3 1 5 4 410 2

salt Q

0.18 1.43 0.23 0.67 0.02 0.45 0.04

2 o1 3 2 6 2 410

Treated K Treated L Treated M Treated N O Treated P Treated Q Treated R Treated S

salt R

G. Steinhauser et al. / Journal of Trace Elements in Medicine and Biology 20 (2006) 143153 151

All values given in mg kg1, except Na, Cl, and Ca in percent. % of RDI shows the percentage of the recommended dietary intake of the essential elements, assuming a consumption of 6 g sodium chloride in the form of the particular rock salt. Where no value is given, the limit of detection was not reached. Errors due to counting statistics.

ARTICLE IN PRESS

ARTICLE IN PRESS
152 G. Steinhauser et al. / Journal of Trace Elements in Medicine and Biology 20 (2006) 143153

Table 6. Adults recommended daily intake of elements analyzed in this study (data taken from [15]) Element Recommended dietary intake for adults 550 mg 850 mg 1g 15 mg 510 mg 25 mg 0.0020.1 mg 0.0030.1 mg Average daily uptake with food 1.76.9 g 2.610.6 g
b b

Na Cl Ca Fe Zn Mn Coa Cr
a b

622 mg 248 mg
b

5200 mg

Co is regarded to be essential only in the form of vitamin B12. No data available due to large variations depending on the type of food.

However, it must be emphasized at this point that cobalt is an essential trace element only in the form of vitamin B12. Iron compounds are even less bioavailable. Data on the daily requirements and the average uptake are taken from [15] and are listed in Table 6. For the estimation whether the use of rock salt for nutrition purposes can signicantly contribute to the daily requirements of trace elements, an average uptake of 6 g sodium chloride was assumed [12,28]. In a separate column, the percentage of the contribution of rock salt to the daily adult requirement of trace elements is shown. This percentage would, of course, imply that only this salt was consumed. Since nearly all food contains salt from its production process, this assumption is unrealistic. Two samples of treated salt contain thorium an aemitter whose needless ingestion should be avoided under all circumstances. Aluminum values can be regarded as good indicators for clay impurities in the salt.

contribute a signicant part of the RDI of chromium and manganese, humans are supplied with these elements sufciently with normal food. Thus, from an analytical point of view, there is no health reason to use unpuried rock salt for nutrition. However, there are a few drawbacks to the use of unpuried salt, as hygroscopic compounds like magnesium or calcium chloride, as well as compounds of toxic heavy metals like chromium or thorium are not removed. This hypothesis is supported by the fact that the sample of rened salt (S) did not contain detectable concentrations of trace elements except bromine and strontium. Rock salt deposits are never built up homogenously, since they were formed by evaporation of natural brine. This led to a sequential precipitation of compounds depending on their solubility in water, starting with calcium carbonate and gypsum and ending with the chlorides of potassium, magnesium, and calcium. However, especially rare-earth element (REE) concentrations can often be used to obtain a chemical ngerprint, which can be used to identify the origin of an unknown sample. In principle, INAA could therefore help identify fake exotic and therefore expensive rock salt. In the case of this study, the sample number from each region was too small to collect signicant data for this purpose. Therefore, more analytical information is needed for a conclusive result in this respect.

Acknowledgments
We are grateful to the donators of the samples, in particular Evi Cech-Haberl, Herbert Stieb (Salinen Austria AG), Ismail Suttar (HUB-PAK Salt Renery, Pakistan), Stefan Trachsel (Vereinigte Schweizerische Rheinsalinen AG, Switzerland), Christian Da hne, and Alfred Ho llerbauer (both Wacker-Chemie GmbH, Germany). Furthermore, the authors wish to express their thanks to Thomas Schwarzgruber for medical consultancy and Georg P. Westphal for providing access to the neutron activation analyzer of the Atominstitut.

Discussion and conclusion

In most samples, the concentrations of trace elements in rock salt were too low to show biological effects. Salt can therefore only signicantly contribute the essential elements sodium, chlorine, and, if added on purpose, uorine and iodine to human nutrition. The contribution of all other traces (e.g., zinc) in salt to the average daily human requirements can be neglected, even if the solubility and bioavailabilty were assumed to be 100% in the case of zinc. Essential trace elements like Fe or Ca do not occur in mostly soluble (and therefore bioavailable) form, but probably in the form of iron oxides and gypsum. Therefore, their contribution to the human recommended dietary intake of trace elements can be neglected. Even though some rock salt samples might

References
[1] Javor BJ. Industrial microbiology of solar salt production. J Ind Microbiol 2002;28:427. [2] Steinhauser G. Sodium chloride. In: Ullmanns encyclopedia of industrial chemistry, electronic release. 7th ed. Weinheim: Wiley-VCH; 2005 [Chapter 5]. [3] Lefond SJ. Handbook of world salt resources. New York: Plenum Press; 1969. [4] Steinnes E. Neutron activation analysis in the geosciences: lost territory, or New Deal? J Radioanal Nucl Chem 2004; 261:7018.

ARTICLE IN PRESS
G. Steinhauser et al. / Journal of Trace Elements in Medicine and Biology 20 (2006) 143153 153

[5] Tobola T. Preliminary results of rare earth elements contents study in rock salts of the Bochnia deposit. Przegl Geol 2002;50:7626 [in Polish]. [6] Yui M, Kikawada Y, Oi T, Honda T, Sun D, Shuai K. A preliminary investigation on the determination of lanthanoids, thorium and uranium in brine and deposit of a salt lake in China. J Radioanal Nucl Chem 1997;231: 836. [7] Yui M, Kikawada Y, Oi T, Honda T, Nozaki T. Abundance of lanthanoids in rock salts determined by neutron activation analysis. J Radioanal Nucl Chem 1998;238:36. [8] Yui M, Kikawada Y, Oi T, Honda T, Nozaki T. Abundance of uranium and thorium in rock salts. Radioisotopes 1998;47:48892. [9] Zhang S. INAA of rock salt samples from a block basin in East China. He Huaxue Yu Fangshe Huaxue 1991;13:1278 [in Chinese]. [10] Slavin W. The determination of trace metals in seawater. Atom Spectrosc 1980;1:6671. [11] Lynch WH. Trace mineral salt analysis with atomic absorption spectrophotometry. In: Symp. Salt Third Meeting Date 1969, vol. 2.Cleveland, OH: N. Ohio Geol. Soc.; , 1970. p. 1409. [12] Steinhauser G. Sodium chloride. In: Ullmanns encyclopedia of industrial chemistry, electronic release. 7th ed. Weinheim: Wiley-VCH; 2005 [Chapter 6]. [13] Elles MP, Armstrong QA, Lee SY. Characterization and solubility measurements of uranium-contaminated soils to support risk assessment. Health Phys 1997;72: 71626. [14] Kunsch K, Kunsch S. Der Mensch in Zahlen, 2nd ed. Heidelberg: Spektrum Akad. Verl.; 2000 [in German]. [15] Belitz H-D, Grosch W, Schieberle P. Lehrbuch der Lebensmittelchemie, 5th ed. Berlin: Springer; 2001 [in German]. [16] World Health Organisation (WHO). Trace elements in human nutrition and health. Geneva: WHO; 1996.

[17] Westphal GP. Real-time correction of counting losses in nuclear pulse spectroscopy. J Radioanal Nucl Chem 1982;70:387410. [18] Westphal GP, Cadek GR, Kero N, Sauter T, Thorwartl PC. Digital implementation of the preloaded lter pulse processor. J Radioanal Nucl Chem 1995;193:818. [19] Peltz C, Schmid P, Bichler M. INAA of Aegean pumices for the classication of archaeological ndings. J Radioanal Nucl Chem 1999;242:36177. [20] Pfennig G, Klewe-Nebenius H, Seelmann-Eggebert W. Karlsruher Nuklidkarte, Forschungszentrum Karlsruhe GmbH, Karlsruhe, 1995. [21] Parry S. Activation spectrometry in chemical analysis. In: Winefordner JD, Kolthoff IM, editors. Chemical analysis, a series of monographs on analytical chemistry and its applications, vol. 119. New York: Wiley; 1991. [22] Westphal GP, Lemmel H, Grass F, De Regge PP, Burns K, Markowicz A. A neutron activation analyzer. J Radioanal Nucl Chem 2005;263:7039. [23] Steinhauser G. Split-precipitation an ecological alternative in the purication of crude sodium chloride brine. Berg Hu ttenmaenn Monatsh 2005;150:2836. [24] Schwaiger H. Methods for the purication of raw brines from ALPINE Salt Mines. Berg Hu ttenmaenn Monatsh 1998;143:12832 [in German]. [25] Klade M, Schwaiger H, Stieb H. In-situ brine purication. Report on the experience of a one year large-scale eld test. Berg Hu ttenmaenn Monatsh 2003;148:506. [26] Flachberger H. The present day technology in salt processing. Part 1. Berg Hu ttenmaenn Monatsh 1998;143: 11523 [in German]. [27] Flachberger H, Krenn K. The present day technology in salt processing. Part 2. Berg Hu ttenmaenn Monatsh 1999;144:23440 [in German]. [28] Verein Deutsche Salzindustrie e.V. (VDS). Salz als Lebensmittel, Bonn, 2002 [in German]. [29] Brookhaven National Laboratory. National Nuclear Data Center, 2005. /http://www.nndc.bnl.gov/index.jspS.