Can. J. Chem. Downloaded from www.nrcresearchpress.com by MICHIGAN STATE UNIV on 01/16/17 For personal use only.

ON THE SOLUBILITY OF ANHYDROUS CALCIUM SULPHATE AND OF GYPSUM IN CONCENTRATED SOLUTIONS OF SODIUM CHLORIDE AT 25" C, 30" C, 40" C, AND 50" C1

ABSTRACT The solubilities of anhydrous CaSO, and CaS0,.2H?O in concentrated aqueous solutions of NaCl a t a series of temperatures-25" C, 30" C, 40" C, and 50" C, have been determined. I t was found that the transition temperature for the reaction which in pure water has a value of 42" C, is shifted progressively to lower temperatures with increasing NaCl concentration. It was also found that the variation of the transition temperature with S a c 1 concentration could be adequately represented by the equatiou AG = A H - T A S f 2 R T In P/Po, where AG, AH, and AS are the Gibbs free etiergy, enthalpy, and entropy of the reaction and P Oand Y are the vapor pressures of pure water and the solution respectively a t the absolute temperature T . INTRODUCTION

I t has long been known that the gypsum-anhydrite transition point is depressed by addition of strong electrolytes. However, it appears that no systematic study of this aspect of this system has ever been made, and only isolated experiments are reported in the literature (1). Since many naturally occurring gypsum deposits which are used as raw material for the commercial production of plaster of paris and building plaster are contaminated with various amounts of sodium chloride, it was decided to carry out a systematic study of the behavior of the system gypsum-anhydrite in the presence of varying amounts of sodium chloride. The results of such a study should afford a clearer understanding of the factors that govern the deposition of anhydrite or gypsum, as the case may be, from a slurry of hemihydrate (CaSO4.3H2O) and water, contaminated with NaCI. EXPERIMENTAL

The salts used in this research, i.e. CaS04 anhyd., CaS04.2I-120, and NaCl, were all Fisher certified reagents of high purity. No recrystallization of the salts was attempted since it was felt that nothing would be gained by it in view of the high initial purity of the materials. An ignition analysis of CaS04.2H20 showed a loss in weight equal to the theoretically expected value of 20.8%. The anhydrous CaS04 was heated a t 900" C for 1 hour and subsequently kept in a desiccator a t a temperature of 60" C. Saturated solutions of gypsum and anhydrite were obtained as follows. Excess solid was added to an aqueous solution of sodium chloride of known con~position.The latter was contained in an all-hard-pyrex-glass solubility cell (Fig. I ) . The cell and contents were then immersed in a water bath the temperature of which was colltrolled by an electronic relay and was constant to within 0.05" C. The temperature of the water bath was read by means of a Beckmann thermometer. The latter was calibrated against a standard platinum resistance thermometer. The solutions were agitated for a t least 48 hours by means of a pyrex-glass stirrer. This time interval was found to be quite 'Manuscript receiaed Ape1 17, 1961. Contribudion from the Department of Chemistry, Unksrskty of Ma~aitoba, Winnipeg, Menitoba. Can. J. Chem. Vol. 39 (1061)

1746

Can. J. Chem. Downloaded from www.nrcresearchpress.com by MICHIGAN STATE UNIV on 01/16/17 For personal use only.

BOCK: SOLUBILITY O F ANHYDROUS CaSOa AND CaS01.2Hz0

FIG.I .

Solubility cell.

satisfactory in the case of the more stable solid pliase. Thus periodic checlts in which tlie solution \\?as stirred for 48 hours, 72 hours, and '36 hours showed no difference in the solubility, i.e. the solubility was identical \\Tithin tlie limits of experimental error of +%. 111the case of the less stable solid phase some difficulties were encountered. I t was found that the solubility decreased \vitli longer periods of stirring. I t was ~lecessaryto make several solubility cleter~ninations for different durations of stirring. Tlie maximum solubility value was talcen as the "true" solubility of the phase. After it was cleemecl that the solution had been stirred lor a sufficient length of time, the excess solicl \\,as separated fro111 the liquid phase by quicltly emptying the contents of the solubility cell into a large sintered-glass filter and applying suction to the suction flaslt by means of a water aspirator. T h e act of filtration never lasted longer than 1 minute, thus malting any possible loss of water due to evaporation from the saturatecl solution negligible. The analysis of the saturated solution was carried out as follows. Two 50-1111salllples of the solutioll were put into two previously weighed Erlenmeyer flaslts. The latter were immediately stoppered with rubber stoppers. Tlie flaslts and contents were weighed on an analytical balance. Next, the calciul~lion concentration was determined by titration with sta~idardE.D.T.A. using "Calred" as indicator. T h e sodiunl chloride content was similarly determined on two different samples by titration with standard AgNOa solution following Mohr's method. Blanks were run in each case.

Can. J. Chem. Downloaded from www.nrcresearchpress.com by MICHIGAN STATE UNIV on 01/16/17 For personal use only.

13 01

W

0, Cir+WLiirP

rn&A60A0 m

o

t

l

OCn+WL3rr

.o .o .c o p c p Cnm444a13 mWwmaMa +morw+rn~

n

t.s&~~&rnb l *-IcCOao

o .c .c c . o . c . c . am4COCO4t3

W a 4 W r C n - I r+orarCn

occ~,cocc . c . o . o . o . o . .o

am-1-1-101C DO-1440N r0xLiimrOtii !+0+-1or0 ~COamWOlii 4 C O r ~ w + q

3

COCCOO

C C 0 0 0 0 0

Can. J. Chem. Downloaded from www.nrcresearchpress.com by MICHIGAN STATE UNIV on 01/16/17 For personal use only.

BOCK. SOLUBILI'I Y O F AXE-IYDKOLS CaSOi

~i

I

,1 I

i

I

I

I

.-Kin CaSOi.2H?O

I

I

0

20

Temperature

OC

FIG.3. Solubility in pure water: 8 CaSOd anhyd., 0 CaS04.2H?0. DISCUSSION OF RESULTS

Fro111 the phase rule i t is known t h a t the number of degrees of freedom F, i.e. the number of undetermined variables of a system consisting of C components which are in mutual equilibrium and exist in P phases, is given b y the expression

I

An aqueous solution of gypsum or anhydrite, containing, in addition, dissolved soclium chloride, constitutes a system of three components: &SO4, NaCl, and H 2 0 . If the temperature and pressure are fixed a t some arbitrary value, the vapor phase ma>- be disregarded. Hence if all sodium chloride is in solution there are only two phases present: the liquid phase-solution, and the solid phase-CaS04 anhycl. or CaS04.21-120. Fro111 eq. [I] it is apparent t h a t F, the rlumber of degrees of freedom, is unity. However, a s soon a s a second solid phase appears, the system will become invariant, i.e. F = 0. I t is clear t h a t when F = 0 the solution must be saturated with respect to both solid phases, C a S 0 4 anhyd. and CaS04.21-I?O. I11 other words, the two solids will be in equilibrium with one another. By determining the solubility of both salts separately and constructing a plot of solubility versus sodium chloride content of solution, it is possible from the intersection of the two corresponcling isothernls to obtain the transition-point te~nperatui-e of the two salts. T h a t is, the point of intersection of the two curves represents t h a t state of affairs in which both salts have the same solubility and this, a s has been show11 above, is just the condition for equilibl-ium. Hence the temperature of the isotherm is the transition temperature for t h e system in the presence of a certain definite quantity of sodium chloride, the actual quantity of which is defined b y the intersection of the two isotherms. By plotting the temperature of the isotherlns versus the sodium chloride

C:\N:\DI..\N JOUKN.4L O F CHEMISTRY. VOL. 39.

Can. J. Chem. Downloaded from www.nrcresearchpress.com by MICHIGAN STATE UNIV on 01/16/17 For personal use only.

1750

.lW

1961

-

C 0

I

I

2

NaCl

.

3

4

5

m o l e s / l O O O q water

6

7

N o C l moles / IOOOq wale!

Fig. 1. Isotherms: ( a ) 25" C , ( b ) 30" C , ( c ) -LO0 C , (d) 50" C ; 9 CnSO., a ~ t h y d . ,0 C ~ S O I . " I H O .

coiltent of the solutions corresponding to the i~ltersectioilpoints a graph is obtained which represents the variation of the transition-point temperature with sodium chloride content of solution. By means of this graph (Fig. 2) oile can predict the nature of the precipitate that will result from a n evaporation of a solution of CaS04 containing a ltnown quantity of sodium chloride. Thus, for all solutions of a given sodium chloride content, CaSOl

BOCK: SOLUBlLlTY O F ANHYDROUS CaSO4 A N D CaS04.2HzO

1751

Can. J. Chem. Downloaded from www.nrcresearchpress.com by MICHIGAN STATE UNIV on 01/16/17 For personal use only.

anhyd. will be deposited provided the temperature is above the curve in Fig. 2 or, alternatively, gypsum will be the solid thrown out if the temperature is below the curve. T h e variation of the transition-point tenlperature with sodium chloride content ma17 be also derived from theoretical considerations. T h e problem is to derive an expression for the variation of the equilibrium constant with temperature for the reaction En~ployingthe known relationships

where AGO, A H o , and ASn are the standard Gibbs free energy, enthalpy, and entropy and K the equilibrium constant for the reaction, it is possible to arrive a t a value of K for any temperature T provided the other quantities involved are known. This procedure was employed by h4acdonald (2) in the calculation of the transition point for the above system. A few words should be said concerning the water term on the right-hand side of eq. [ 2 ] .This term refers to pure water. But in presence of NaCl the water is no longer pure; its activity is said to be depressed by the presence of NaC1. T h e extent of this depression may be calculated approximately from vapor pressure values of aqueous sodium chloride solution. These are extensively recorded in the International Critical Tables (3). Inserting the nun~ericalvalues for A H , A S , and R the gas constant (4), and recognizing that a t equilibrium, AG = 0, one obtains

Table I1 shows a nuniber of calculated transition-point tenlperatures along with those TABLE I 1 Variation of transition point with temperature -

Moles of NaCl 1000 g water

in

5.30

Temperature, "C Experimental

3.28 2.30

20.0" 25.0 30.0 35 0

0.00

43.0

4.30

CalcuIated 19.0 25.2 31 . O

32.5 40.0

experimentally determined. T h e agreement is quite satisfactory if one taltes into acco~iiit the approximations involved in arriving a t eq. [5]. ACKNOWLEDGMENT T h e author is indebted to Western Gypsum Products Co. Ltd. of Winnipeg, A~Ianitoba, for a generous grant. REFERENCES 1. E. POSXJAK. Am. J. Sci. . 4 , 3 5 , 2 4 i (1938). W. M. MADGIN a n d L). .A. S\V.AILS. J. Appl. Chem. (Iao1~tlon), 6, 482 (195G). 2. J. F. MACDONALD. AIII.J. Sci. 251, 884 (1953). 3. INTERNATIONAL CRIT~CAL TABLES.WlcGraw-Hill Book Co., New Yorli.-1933. 4. I