DISSOLUTION

OF ALKALINE

PRESENCE

EARTH

SULFATES

IN THE

OF M O N T M O R I L L O N I T E DENNIS D. EBERL

U.S. Geological Survey, Box 25046, Mail Stop 420, Federal Center, Denver, CO 80225, U.S.A.

and EDWARD R. LANDA U.S. Geological Survey, Mail Stop 413, National Center, Reston, VA 22092, U.S.A.

(Received March 12, 1984; revised September 11, 1984) Abstract. In a study of the effectof montmorilloniteon the dissolution of BaSO4 (barite), SrSO4 (celestite), and 226RafromU milltailings,it was found that: (1) More of these substances dissolvein an aqueous system that contains montmorillonite than dissolve in a similar system without clay, due to the ion exchange properties of the clay; (2) Na-montmorilloniteis more effectivein aiding dissolution than is Ca-montmorillonite; (3) the amount of Ra that moves from mill railings to an exchanger increases as solution sulfate activity decreases. Leaching experiments suggest that 226Ra from HzSOa-circuit U mill tailings from Edgemont, South Dakota, is not present as pure Ra sulfate or as an impurity in anhydrite or gypsum; it is less soluble, and probably occurs as a trace constituent in barite.

I. Introduction The interaction of cations with naturally occurring ion exchangers such as clay minerals, zeolites, and organic matter has been recognized and investigated for many decades. The potential role of such exchange materials in enhancing the dissolution of relatively insoluble substances in natural systems, by contrast, has received little attention. Several investigators (Honda et al., I953; Osborn, 1953; Rich, 1963) have demonstrated that the dissolution of barite and other sparingly soluble salts is accelerated and promoted in aqueous laboratory systems by the ion exchange ability of synthetic ion exchange resins. Bradfield (1932) found that sparingly soluble sulfates and oxalates dissolve in the presence of Na-exchanged clays, and calculated from his experiments that a hectare of soil containing about 225000 kg of colloidal clay could render 3810 kg of BaSO 4 soluble. G r a h a m (1941) showed that H÷-saturated clay was an effective agent for dissolving anorthite, thereby rendering its Ca more available as a plant nutrient. More recent studies by Bunzl and Sansoni (1972) and Bunzl (1981) have demonstrated the dissolving power of soil organic matter and Ca-saturated bentonite for a variety of sparingly soluble compounds, including salts of heavy metal ions. Thus, it appears that reactions of this type may be significant in the pedologic and geologic environment. The present paper investigates the influence of Ca and Na-saturated montmorillonite on the dissolution of celestite (SrSO4) and barite (BaSO4). It was hoped that if the effect could be understood and predicted quantitatively for these alkaline earth sulfates, then the results could be used to predict the behavior of Ra sulfate in similar systems of Water, Air, and Soil Pollution 25 (1985) 207-214. 0049-6979/85.15. © 1985 by D. Reidel Publishing Company.

208

DENNIS

D. E B E R L A N D E D W A R D

R. L A N D A

geological and environmental interest. Specific attention is focused on 226Ra in U mill tailings derived from a H2SO 4 acid leach process. 2. Materials and Methods

Two montmorillonites were used in these experiments: Na and Ca-exchanged, < 2 gm Wyoming bentonite (SWy - 1 obtained from the Source Clays Repository of the Clay Minerals Society; Moll e t a L , 1975), and Na-saturated, < 2 ~tm Kinney bentonite (Khoury and Eberl, 1981). The cation exchange capacities of these clays are 91 and 125 meq (100 g)-i, respectively. The clays were size-fractionated by centrifugation (Jackson, 1975) and saturated with Ca or Na by washing them three times in one-normal chloride solution. Then the clays were washed three times in deionized water and dialyzed overnight to remove excess salt. Dialysis bags used in the experiments (24 average pore diameter) were washed three times and shaken in deionized water overnight to remove impurities. BaSO 4 (Baker*, anhydrous, reagent grade) and SrSO 4 (purified, anhydrous, from Barium and Chemicals, Inc., Steubenville, Ohio), which proved to be barite and celestite by X-ray diffraction (XRD) analysis, were washed and then dialyzed overnight against deionized water prior to use to remove readily soluble impurities. Then they were dried at 200 ° C. One gram aliquots of the sulfates were placed in polycarbonate centrifuge tubes. To each tube were added 15 mL of deionized water and a sealed dialysis bag containing either 10 mL of a clay suspension (approximately 2% clay by weight) or 10 mL of deionized water. The tubes were shaken at room temperature from 1 to 54 days. On the sampling day, the dialysis bags (if unbroken) were retrieved, washed externally with deionized water, and the interior clay suspension was removed by means of a syringe and hypodermic needle. The tube was centrifuged and the supernate filtered through a 0.2-gin membrane filter. The filtrate was analyzed for sulfate by a turbidimetric procedure (Picketing, 1983) and for Na, Ba, Sr, and Ca by atomic absorption spectrometry. The clay suspensions were dried at 70 ° C, and the dried materials ground by use of a Wig-L-Bug. The ground clay was dissolved using a Li metaborate fusion-HNO3 procedure (Van Loon and Parissis, 1969) and analyzed by atomic absorption spectrometry. Similar experiments were performed using U mill tailings rather than sulfates to study the effect of clay on the release of 226Ra. A 15 g sample of sandy tailings (143pCi226Rag -1) from an inactive H2SO 4 acid-circuit, U mill tailings pile at Edgemont, South Dakota, was contacted twice in succession (1 h shake at room temperature) with 150 mL of deionized water. The sample was centrifuged (5000 rpm), and the supernate filtered (0.2 pm filter). Some fine particles were lost to the filter upon decantation; therefore, multiple leaches were not attempted. The filtrate was analyzed for major cations and anions, Ba, Sr, 226Ra, pH, and specific conductance by the methods of Skougstad et al. (1979) and Thatcher et al. (1977). * The use ofbrand namesin this report is for identificationpurposes onlyand doesnot implyendorsement by the U.S. GeologicalSurvey.

209

D I S S O L U T I O N OF ALKALINE EARTH SULFATES

The effect of adding clay to a similar system was studied by placing a dialysis bag containing 0.51 g of the Na-saturated Wyoming bentonite into a bottle containing 200 mL deionized water and 5 g tailings. Another bottle contained 200 mL deionized water and 5 g tailings, but no added clay. The bottles were gently agitated on a rotary shaker for 5 days at room temperature. Then the dialysis bag was removed, and the clay samples (including a clay sample not exposed to the railings, to determine the native 226Ra content) were analyzed for 226Ra by the Rn-emanation method (Yang, 1980). The kinetics ofBaSO 4 dissolution in a Na-Wyoming bentonite system were followed, using a Lazar conductivity probe with an Altex Selection 5000 ion meter. Ten mL of the clay suspension was pipetted into 50 mL of deionized water and allowed to equilibrate for approximately 2 hr while being stirred from above. Then 1 g of BaSO 4 was added and the conductivity of the stirred slurry was monitored through time.

3. Experimental Results The experiments revealed that no measurable systematic changes occurred in solution compositions with contact times lasting between 1 and 54 days, which suggests that equilibrium had been attained in 1 day or less, Hence, the solution compositions for runs lasting 1 to 54 days are presented as averages in Table I. The reaction of BaSO 4 with Na-montmorillonite was followed kinetically by measuring the change in solution conductivity through time. Equilibrium for this experiment, conducted without a dialysis bag, was attained within an hour. The weight of alkaline earth sulfate dissolved by the experimental system was calculated from the experimental data (Table I) by three methods, based on: (1) Cations in solution (Na + and Ba2+ or Sr2+), (2) anions in solution (SO42-), and (3) Ba2+ or

TABLE I ' Solution compositions in mg L - i a System

Na

Sr

Ba

Ca

SO 4

BaSO 4 SrSO 4 Na-Wyoming BaSO 4 + NaWyoming SrSO 4 q- NaWyoming Na-Kinney BaSO 4 + NaKinney Ca-Wyoming BaSO 4 + CaWyoming

0.5 1.2 27.7

0 59.3 0

2.9 n.d. 0

n.d. b n.d. 0.9

3.4 63.3 11.9

43.5

n.d.

4.2

n.d.

55.7

115.5 48.5

16.6 n.d.

n.d. 0

n.d. n.d.

208.8 7.2

66.3 1.2

n.d. n.d.

0 0

n.d. 9.4

47.0 10.9

3.1

n.d.

0

11.3

12.2

Arithmetic average of up to 10 run compositions taken over the course of up to 54 days. b n.d. = not determined. a

210

DENNIS D. EBERL AND EDWARD R. LANDA TABLE II Calculated values for the dissolution of BaSO 4 and SrSO 4 in equilibrium with montmorillonite

System

BaSO4 SrSO 4 BaSO 4 + NaWyoming SrSO 4 + NaWyoming BaSO4 + NaKinney BaSO 4 + CaWyoming

Calculated dissolution, in mg, based on

Equivalent fraction Ba or Sr in clay (based on anions in solution)

Cations in solution

Anions in solution

Clay

0.1 3.1

0.2 3.0

-

-

2.5

2.6

2.0

0.15

9.2

9.4

9.6

0.68

2.2

2.9

2.0

0.10

0.2

0.1

n.d. a

0.01

a n.d. = not determined.

Sr 2 + in the clay. The methods gave comparable results for the quantity of alkaline earth sulfate dissolved when a clay was present (Table II). A comparison between systems shows that those containing Na-montmorillonite dissolved > 10 times more BaSO 4 and > 3 times more SrSO 4 than those without clay, and that Ca-montmorillonite systems were much less effective than Na-montmorillonite in promoting BaSO 4 dissolution, presumably due to the greater replacing power of Ba 2 + for Na + as compared to Ca 2 +. 226Ra concentrations in the 1:10 U mill tailings:water leachates were 6.8 and 5.6 pCi L - 1, respectively in the first and second extractions; the corresponding sulfate concentrations were 380 and 25 mg L - i, respectively. The ion activity products (IAP) for RaSO 4 for leaches 1 and 2 were 4.2 x 10- 17 and 5.7 × 10- i8, which are smaller than the solubility product (Ks) for RaSO 4 (4.25 × 1 0 - i i ; Vdovenko and Dubasov, 1975), the Ra phase most likely to be present if 226Ra is there as a separate mineral in the tailings. It is probable, however, that the 226Ra is present in trace amounts in the crystal structure of another mineral, such as barite. Such an isomorphous substitution would render Ra much less soluble due to the fact that the mole fraction of RaSO 4 in the mixed solid is very low (Stumm and Morgan, 1970; Lind, 1918 as cited by Seeley, 1977). Studies by Langmuir et al. (1983) showing barite-radium sulfate solid solution control on the concentration of dissolved 226Ra concentrations in U mill tailings solutions support the view of BaSO 4 as the likely host mineral. Also, recent nuclear emulsion microscopy studies of U mill tailings (Stieff, 1984) suggest barite as a major host for 226Ra. The effect of clay on the dissolution of Ra is given in Table III. In the system containing tailings and water, 0.45 ~o of the 2 2 6 R a in the tailings is mobilized, whereas in the system containing tailings, water, and clay, a total of 2 . 3 ~ is mobilized. While we cannot definitely state that this 226Ra is derived from the dissolution of barite or some other sparingly soluble alkaline earth sulfate salt (as opposed for example to desorption

DISSOLUTION OF ALKALINE EARTH SULFATES

211

T A B L E III Leaching of 226Ra from U mill tailings in the presence and absence of montmorillonite clay

System

Tailings + water Tallings + water + clay Clay only

226Ra in solution

226Ra in clay

pCi L- 1a

~/ob

pCi g- 1

%b

12.8 10.9 -

0.45 0.38 -

28.8 2.8

1.94° -

1 pCi of 226Ra = 10- 12g. u Percent of the total Z26Ra present in 5 g tailings. ° Corrected for original clay 226Racontent. from a variety of possible substrates), it is clear that the clay is acting as a sink for Ra and thereby promoting its release from the tailings.

4. Discussion The experiments have shown that in a closed system clay can aid in the dissolution of sparingly soluble sulfates, and the kinetic data suggest that there is little kinetic inhibition. Equations based on equilibrium thermodynamics (Eberl, 1985) give reasonably accurate predictions for the dissolution o f B a S O 4 and SrSO 4 by montmorillonite; these equations also can be applied in an approximate manner to the dissolution of sparingly soluble compounds containing Ra in U mill tailings in the presence of an ion exchanger as solution composition changes through time. Such an equation for predicting the amount of R a S O 4 dissolved at equilibrium in the presence of a Ca-exchanger is: F (R-a) (CEC)g LK s ] mg R a S 0 4 dissolved = L 2 x ~ + ~R~,[S042- ] (FW) (1000),

(1)

where m

(Ra) =

KcKs [Ca 2+ ] [SO42- ] +

KcK s

In Equations (1) (Ra) is the equivalent fraction of 2 2 6 R a in the clay, (CEC) is the cation exchange capacity of the clay in meq (100 g - a), g is the weight of clay in the system, L is the volume of solution, 7R, is the activity coefficient of 226Ra2 +, [8042- ] and [Ca 2+ ] are the activities of these ions, (FW) is the formula weight of RaSO4, Kc is the selectivity coefficient for Ca-Ra exchange, and K s is the solubility product for the dissolution of RaSO 4. Figure 1 graphs this equation for 1000 g of an exchanger with a CEC of 25 meq (100 g - 1) and an average K c of 1.1 for Ca-Ra exchange (Kc based on ionic size o f R a 2 + ; see Benson, 1983). The IAP for RaSO4 for leach 1 is used to approximate K s. As a

212

D E N N I S D. EBERL A N D E D W A R D R. L A N D A

i

I

i

i

3

4

I I I

I1

4

,,,

>C9

> .J

0 CD C~

0 ~"

5

U) co

Ir CO <

m 0

6

._1 ,-I

0

~ 0

T

._1 I

10 1

2

- OO, o

5

1

Fig. 1. Calculated weight of Ra sulfate dissolved by 1 kg of Ca-exchanger (CEC = 25) and 1 L of solution (top curve), and by ! L of solution only (bottom curve) as a function of changing ~ulfate activity (based on Equation (1) and parameters given in text).

D I S S O L U T I O N OF ALKALINE EARTH SULFATES

213

tailings solution becomes more dilute through time either, for example, by leaching action of rain water or by erosive dispersal of tailings into fresh water, Figure 1 shows that Ra will move from the sparingly soluble phase to the exchanger, and thereby will become available for cation exchange. Thus, more 226Ra may be released to interacting waters through time than would be predicted from a model based solely on a consideration of the solubility product. The same considerations would appear to be applicable in the disposal of Ra-bearing BaSO4 sludges in a landfill (Oak Ridge National Laboratories, 1979), and the proposed incorporation of fuel reprocessing wastes in a Ba sulfate matrix prior to geologic disposal (Briggs et al., 1982). Recent studies at Pacific Northwest Laboratory (Uziemblo et al., 1981) have demonstrated the precipitation of jarosite [(K,Na,H)F%(SO4)2(OH)6 ] and alunite [KA13(S 0 4)2(OH)6 ] in clay liner material (largely smectite) contacted with acidic U mill tailings effluent. Jarosite-group minerals, produced in U mill railings during the milling process or as a result of weathering processes, have been shown to contain up to 80000 pCi 226Ra g-1, about 1000 times the 226Ra content of the bulk tailings from which they were separated (Kaiman, 1977); such minerals would thus appear to be potentially important sinks for 226Ra in U mill tailings. This study suggests that over the long term, in such intimate admixtures of Ra-rich jarosites and high exchange capacity clay minerals, 226Ra transfer to the exchange sites of the clay may occur concomitant with dissolution of the jarosite. No attempt was made to remove S 0 4 z - from the system in our experiments. Hence, the law of mass action will mitigate the amount of dissolution. In natural systems, SO42- removal could be achieved by solute transport, by biological uptake and(or) transformation (for example, sulfate-reducing bacteria), or by sorption on earth materials, such as goethite, gibbsite, and volcanic ash (Parfitt, 1978). Whereas these experiments have focused on alkaline-earth sulfates and clay minerals, the process need not be limited to either. Other sparingly soluble salts of geologic or agronomic importance (for example, calcite, apatite) can be involved, as can other exchangers such as organic matter (Tan, 1980), zeolites, and hydrous oxides (Fischer and Puchlet, 1972).

Acknowledgments The assistance of the Tennessee Valley Authority, Chattanooga, and Silver King Mines, Inc., Edgemont; South Dakota, in obtaining the uranium mill tailings sample, and the analytical support of the U.S. Geological Survey National Water Quality Laboratory, Arvada, Colorado, are gratefully acknowledged. The authors thank J. Cleveland and T. Henderson for their reviews of the initial manuscript.

References Benson, L. V.: 1983,Environ. Geol. 4, 23-29. Bradfield, R.: 1932,J. Phys. Chem. 36, 340. Briggs,A., Jones,D. V. C., and Cole,G. B.: 1982,'Hot-PressedBarium Sulphate CeramicWasteForms for Direct Immobilization of Medium Level Magnox Waste', in S. V. Tropp (ed.), The Scientific Basis for

214

DENNIS D. EBERLAND EDWARDR. LANDA

Nuclear Waste Management, Materials Research Society Symposia Proceedings, Vol. 6, pp. 541-548, Elsevier. Bnnzl, K.: 1981, 'The Dissolution of Sparingly Soluble Salts of Metal Ions by Clay Minerals in the Soil', in Internat. Conf. on Heavy Metals in the Environment, Amsterdam, pp. 717-720. Bunzl, K. and Sansoni, B.: 1972, Z. Pflanzenernaehr. Bodenk. 133, 132. Eberl, D. D.: 1985, Clays and Clay Minerals, submitted. Fischer, K. and Puchelt, H.: 1972, 'Barium:', in K. H. Wedepohl (ed.), Handbook of Geochemistry, Vol. II-3, Chapter 56, Springer-Verlag. Graham, E. R.: 1941, Soil Sci. 51, 65. Honda, M., Yoshino, Y., and Wabiko, T.: 1953, Chem. Abstr. 47, 2568. ]J. Chem. Soc. Japan, 73, 348-349 (1952)]. Jackson, M. L.: 1975, Soil Chemical Analysis-Advanced Course, 2d edn., Madison, Wisc. Kaiman, S.: 1977, 'Mineralogical Examination of Old Tailings from the Nordic Lake Mine, Elliot Lake, Ontario', Canada Centre for Mineral and Energy Technology Report ERP/MSL 77-190 (IR). Khoury, H. N. and Eberl, D. D.: 1981, N. Jb. Miner. Abh. 141, 134. Langmuir, D., Riese, A. C., and Melchior, D. C.: 1983, Abstracts with programs, 96th Ann. Meet. Geol. Soc. Amer., 622. Moll, W. F. Jr., Johns, W. D., and Van Olphen, H.: 1975, Internat. Clay Conf. Proc., Mexico, 465. Oak Ridge National Laboratories: 1979, 'Formerly Utilized MED/AEC Sites Remedial Action Program. Radiological Survey of the St. Louis Airport Storage Site, St. Louis, Missouri', Final report. U.S. Dept. of Energy Report DOE/EV-0005/16. Osborn, G. H.: 1953, Analyst 78, 220. Parfitt, R. L.: 1978, 'Anion Adsorption by Soils and Soil Materials', in Advances in Agronomy 30, pp. 1-50. Picketing, R.J.: 1983, 'Analytical Methods: Sulfate Determination', U.S. Geol. Surv. Quality of Water Branch Tech. Memorandum 83.07. Rich, R.: 1963, J. Chem. Educ. 40, 414. Seeley, F. G.: 1977, Hydrometallurgy 2, 249-263. Skougstad, M. W., Fishman, M.J., Friedman, L C., Erdmann, D. E., and Duncan, S. S. (eds.): 1979, 'Methods for Determination of Inorganic Substances in Water and Fluvial Sediments', U.S. GeoL Surv. Techniques of Water-Resources Investigations, Book 5, Chapter A1. Stieff, L. R.: 1984, U.S. Department of Energy report GJ/TMC-15. Stumm, W. and Morgan, J. J.: 1970, Aquatic Chemistry, Wiley, New York, pp. 206-207. Tan, K. H.: 1980, Soil Sci. 129, 5. Thatcher, L. L., Janzer, V.J., and Edwards, K.W.: 1977, 'Methods for Determination of Radioactive Substances in Water and Fluvial Sediments', U.S. GeoL Surv. Techniques of Water-Resources Investigations, Book 5, Chapter A5. Uziemblo, N. H., Erickson, R. L., and Gee, G.W.: 1981, 'Contact of Clay Liner Materials with Acidic Tailings Solutions', in Proc. of the 4th Symp. on Uranium Mill Tailings Management, Colorado State University, p. 597. Van Loon, J. C. and Patissis, C. M.: 1969, Analyst 94, 1057. Vdovenko, V. M. and Dubasov, Y. V.: 1975, Analytical Chemistry of Radium, Wiley, New York, p. 36. Yang, I. C.: 1980, Health Phys. 39, 1059.