6508
S. A. GREENBERG [CONTRIBUTION FROM THE LABORATORY FOR
INORGANIC AND
Vol. 80
PHYSICAL CHEMISTRY, UNIVRRSITY OF
LEIDE?;]
The Nature of the Silicate Species in Sodium Silicate Solutions1 BY S. A. GREEN BERG^ RECEIVED JULY 14, 1988
An analysis of the literature data and e.m.f. and conductance measurements of the present study was made to determine the degree of polymerization of the silicate ions. I t was concluded that in relatively dilute solutions and in solutions above pH 10.5, the silicate ion is monomeric. The PK1 and PK2 values for silicic acid Si(0H)d a t 2Oo0are9.85 and 11.8, respectively. The equivalent conductance of the silicate ion SiO(OH),- was estimated to be 35 a t 25 .
Although the nature of sodium silicate solutions is fairly well understood in its broad outlines, confusion exists as to the degree of polymerization of the silicate ions3 The literature data and e.m.f. and conductance measurements of this study were analyzed according to current theories to determine this information. The hypotheses were tested by evaluating the ionization constants and the equivalent conof silicic acid ductance of the silicate SiO(0H)3 - ions. ‘ The equilibria associated with monomeric silicic acid and those active in sodium silicate solutions have not been treated adequately. For example, the negative logarithms of the dissociation constants of silicic acid have been reported4-10 between 9 and 10 for pKl and between 12 and 14.3 for PKz. The molecular weights of silicic acid in sodium silicate solutions have been determined by light scattering, l1 cryoscopic, diffusion and transference measurements12 but no theory has been offered to explain the results. In the present study the dissociation constants of silicic acid were calculated from e.m.f. measurements found in the literature13 and from conductivity measurements on sodium silicate solutions with NazO:SiOz mole ratios of 1:0.5, 1:1, 1:1.5, 1 : 2 and 1:3. The electrical conductivity results and data taken from the literature were used to evaluate the equivalent conductance XO of the SiO(0H) 3 -.ions. Experimental Equipment.-For the equilibrium studies a water-bath held constant to iz0.01” was employed. Conductivity measurements were made with a Philips (Eindhoven) Bridge S o . GM4249 and both Philips and Industrial Instruments dip cells’4 with constants of approximately 1.0 and 0.5 cm.-’. Materials.-The sodium silicate solutions were made from spectroscopically-pure, special bulky silica’4 which contains 85.B70 SiOz and water. Solutions made with analytical reagent grade sodium hydroxide were found to be of sufficient purity after the carbonate was removed. Water of (1) Presented before t h e Division of Physical and Inorganic Chemistry a t t h e 130th meeting of t h e American Chemical Society. Atlantic City, New Jersey, lY56. (2) Portland Cement Association, Skokie, Illinois. ( 3 ) For review see R . K. Iler, “Colloidal Chemistry of Silica and t h e Silicates,” Cornell University Press, Ithaca, N. Y., 1955. (4) P. S. Roller and G. Erwin, TRISJOURNAL, 62, 481 (1940). ( 5 ) W, D. Treadwell and W. Wieland, Helv. Chim. Acta, 13, 843 (l103.zo Although Bacon and W i W 3 considered the p H data of Harmanlo and Bogue* unreliable, eq. 8 and 9 were applied to these data to obtain pK1 and pK1 values a t 25 and 30". The (OH-) values were in these cases given by the authors. The 25" values for pK1 and PKz were found by extrapolation to be 9.7 and 11.9, respectively. From the data of Bogue, pK1 and pKn were estimated to be 9.1 and 11.9 a t 30". A heat of ionization a t infinite dilution for silicic acid of 3.3 kcal./mole was estimated by com(20) T.B. Smith, "Analytical Processes," 2nd Ed., Edward Arnold and Co., London, 1940. (21) H. S. Harned and B. B. Owen, "The Physical Chemistry of Electrolytic Solutions," Reinhold Publ. Corp., New York, N. Y., 1943. (22) S.Glasstone, "Introduction t o Electrochemistry." D. Van Kostrand Co., New York, N. Y., 1940. (23) D. I. Hitchcock and A. C. Taylor, THISJOURNAL, 69, 1812
(1937).
6309
SILICATE SPECIESIN SILICATE SOLUTIONS I O .8
10.6
10.4
IO .2
10.0
9.8
,
1
0.01
0.02
0.03
I
0.04
Ionic Strength,&. Fig. 1.-The plot of pK1* as a function of ionic strength p : 0,NagO:SiO,, 1:1.508: 0 , Na*O:SiOz, 1:1.996.
paring the pK1 values a t 20, 2524 and 35OZ4by means of the van't Hoff equation. From the heat of solution of amorphous silica (2.65 kcal./mole) 26 and the heat of ionization, i t was possible to estimate the heat of neutralization as -7.6 kcal./mole, assuming the heat of ionization of water is -13.6 kcal./mole a t 25'. Conductivity Measurements.-Figure 2 summarizes the conductivity data of the present study on sodium silicate solutions of Na2O:SiOz mole ratios 1:0, 1:0.5, 1:1, 1:1.5, 1 : 2 and 1:3. The results agree fairly well with those of HarmanlB except for the sodium hydroxide and NazO: 0.5 Si02 solutions. For these solutions the results of Harman show considerable scatter. The concentration constant Kl* can be evaluated from the equivalent conductance data [ A N a O H and ANaHrSiO, (Fig. 2)] on solutions with the same sodium ion concentrations. From eq. 5 and 6 an equation can be derived relating the equivalent conductance of the solutions with those of the ions XOH-, XNa+ and hHlS104- and the hydrolysis constant K Hwhich is equal to Kw/K1*
Here use is made of the assumption that the degree of hydrolysis of H3Si04- is small in the PH range 11.2 to 12 and that in this pH range the concentration of hydrogen and H2Si04- ions may be ignored. It is possible to neglect the hydrolysis because the constant is of the order of magnitude of only By plotting known values of the difference A N ~ O H- ANaHfiiO, as a function of [SiO2]-'/* a difference XOH- -XHlSi04in the equivalent conductances of the ions of 145 was obtained as an intercept and from the slope of the curve a $K1* of 9.9 was calculated. This value of 9.9 is quite consistent with the pK1 of 9.77 a t 25" derived from solubility and p H measurements.24 (24) S. A. Greenberg and E. W. Price, J . Phyr. Chcm., 61, 1539 (1957). (26) S. A . Greenberg, ibid., 61, 196 (1957).
S. A. GREENBISKG
6510
Vol. YO
hydrolysis x takeii i r w l several source^^^^^^^' a 1 ~ 1 XO values for silicate ions calculated on the basis of 220 the various x values. In the last column the An values for H3Si04- ions reported by Harinan arc: 210 shown. Some difficulty was encountered in estiniating x , therefore both 20 and 25' data were used 200 since x does not change markedly with temperature. _. 180 The A,, value of 33 for the I :3 mole ratio was coii0 sidered most reliable because x could be estimated 160 accurately for this ratio. The difference between 3 31. and 35 in the 1: 3 and 1: 2 mole ratio solution is 140 u within experimental error, or as is more probably u 9 120 the case (Table II), there are a certain nu~iiber of low molecular weight silicate aggregates present rj .$ 100 it1 the 1 : 3 solution. From the value 35 for A 0 it should follow that XOII-- X F I ~ S ~ O ~a -t infinite 80 dilution is about l(i0. Coiisitlcring the large pessible error in the extrapolation of the data using thc 60 rclatioriship shown i i i cq. IO, which leads t o a tlifference of about 143, we can conclude that tlic 40 agreement is reasonable. Transference Numbers.-A re-examination of the data of Harman2* leads to quite interesting conclusions. Table I1 summarizes the results. It was assuirietl that in these solutions only Na', 0.2 0.1 0.6 0.8 OHand HaSiOa- ions need be considered. By Sq. root of X a + concentration (Na+)''2. using the equation of Harman to calculate mio2, Fig. 2.-Relationships of the equivalent conductances but assuming each Si02 is singly charged as HSwith (Kat.)'/, of solutiolis of sodium hydroxide and sodium SiO4- instead of doubly charged H2Si04= ions as silicates of various mole ratios: (1) KaOH; (2) 1:0.5; ( 3 ) Harman did, the values of n ~ i owere , evaluated. 1 : l ; (4) 1:1.5; (5) 1:2; (6) 1:3. Where the sums of tzxa-l-,Y I O M - and nsion are greater It is possible to calculate the equivalent con- than one, i t is assuined that there is inore than one ductance of HaSi04- ions a t infinite dilution by Si02 per ion atid in the last column of the table the values of A' (no. of Si02 per ion) arc listed. means of the equation Froin a knowledge of iisiol in experiments 5 atid X N n ~ O . f i i 0 ~ XNat (1 - X)AHISIOI XAOII(111 6 , i t was possible by means of the equivalent where x is the fraction of H3Si04- ions hydrolyzed. conductance data to luate Xo for H3Si04- as The procedure consists in choosing linear portions 34 and 33, which is i i i good agreeinent with the of the curves in Fig. 2 at low concentrations where quantities calculatctl frmi conductaiiccx rcsults only Na+, OH- and H3Si04- ions are present (Table I). (@Hanguly29 support this theory. Cryoscopic measTHE EQUIVALENT COSDL~CTASCE OF HaSi04- Ioris AT 25' urements on sodium silicate s o l u t i ~ n sindicate ~~~~ XO, t" H? ..", decreasing effects with increase in Si02 :NasO NanO: I11 h,HsSiO-, S y s c Na?O: %Si$!% I1 Bacon ohs-1 cm.5 Harratio. Unfortunately, the results are often conI11 m a n > & I I1 Harand 25 , I Si01 fusing and disagreements between author^^,^^ ref. ref. ref. (dil. \V:ill;ls mole HaggQ man" ohms-l 13 s o h ) 9 27 25' 20 ratio cm.1 20' are found which perhaps show the experimental 1:0.5 200 0 . 5 9 0 . 3 5 0.80 93 125 . . . . difficulties involved in these experiments. Al1:l 165 .38 .25 .50 65 87 32 60 though light-scattering measurements are rc1:1.5 127 .19 .I2 .27 48 60 33 35 I'ortedll to indicate tllat the molecular weights 92 ,040 ,010 .OR 35 35 32 43 1:2 o f the silicate polymioiis in Na2O :12Si02 solutions 1:3 83 ,012 ,011 013 3 1 31 30 41 with mole ratios 1xt.tliceii I : 0 . 5 and 13.75 increasc f r o m 60 to 4.00, tl1c authors ciiiphasize tllat thew The table lists the various Na& :Si02 mole ratios, cluantities are I i o t absolutt.. I t should I)t. I i o t c s t l the A0 or intercept values, average degrees of
-
cs
+
_v n_ _ _ _ _
+
-
(26) R . A . Robinson and R. H. Stokes, "Electrolytic Solutions," Rutterworth Scientific Publications, London, 1955. (2;) R. U'. H a r m a n , J . P h y s . Chent., SO, 1100 (1926).
( 2 8 ) IS). (2Y) .'F 13, Ganguly. i b i d . , 31, 107 ( 1 9 2 7 ) . (30) I!I (I!l?fit
SYNTILESIS AND PROPERTIES OF ZIRCONIUM DISULFIDE
Dec. 20, 19%
6511
TABLE I1 NUMBERS
TRASSFERENCE
'7Ijxpt. no.28
4 5 G 7 8 9 10
Mole ratio
1:2 1:2 1:2 1:3 1.3 1 :4 I:1
(Na+)
PH
1. 0 0.5 0.1 1. 0 0.5 1.0 0.1
12.12 11.96 11.46 11.29 11.27 10.86 10.75
ma+ 0.42 .34 .45 .40
.45 .53
.44
that in the dilute solutions of sodium silicate used in light-scattering measurements, i t is probable that all the silicate species are in solution as monomeric silicic acid. 2 4 * The silica in solutions appears to be colloidal under two conditions. In alkaline solutions above pH 10.6 low molecular weight silicate species are brought into solution by p e p t i ~ a t i o n . ~Also ~ in concentrated solutions of silicic acid the monosilicic acid behaves as if it were aggregated into [Si(OH)4], particles where n is probably a small number.16 (31) G. E. Alexander, 68, 453 (1954).
1
- ;zya+ 0.58 .66 .55 .60 .55 .47 .56
nsio2
Hydroiysis,25
R
(no. of SiOz,'ion)
1.38 1.88 2.85 0.192 .36 ,071 .57
1 1 1 1.1 1.4 2.2 2 2
7105-
0.45 .33 .47 .68 .78 1.17 1.23
0.13 .33 .08
.. ..
.. .. Acknowledgments.-To Professor J. J. Hermans for suggesting this approach to the problem and for his many suggestions during the course of the study. To Dr. J. Lorimer for helpful discussions. To Miss Joyce van den Berge and Mr. U. Verstrijden for assistance in performing the experiments. Thanks are also due the Foundation for Fundamental Research on Matter (FOM) supported by the Netherlands Organization for Pure Research (ZWO) under whose auspices this research was performed.
W.H. Heston and R. K. Iler, J . Phys. Chew.,
CHICAGO, ILLINOIS
[CONTRIBUTION FROM
THE
TITANIUM ALLOYMFG. DIV. O F THE NATIONAL LEADC O . ]
The Synthesis and Properties of Zirconium Disulfide1 BY ABRAHAM CLEARFIELD RECEIVED MAY7, 1958 Historic methods for preparing zirconium disulfide were re-examined and new and refined methods for obtaining a pure product have been developed. The reaction of carbon disulfide with zirconia yields a mixture of zirconium disulfide and zirconium sulfoxide, ZrOS, in the temperature range 850-1200°. Above 1200" only zirconium disulfide is obtained. This has been correlated with the presence of different forms of zirconia in the two temperature ranges. The air oxidation of the disulfide was examined by means of differential thermograms and high temperature X-ray diffraction patterns.
Published information on the sulfides of zirconium is meager and much of what is reported is conflicting. Therefore, a study of the zirconiumsulfur system was undertaken. The present paper describes the synthesis and some properties of zirconium disulfide. Zirconium disulfide has been synthesized by a variety of methods. Fr&my2claimed to have prepared i t by the action of carbon disulfide on zirconia a t red heat. No analysis was given and i t is questionable that his product was oxygen-free. A number of workers obtained zirconium disulfide by combination of the elements.3-5 Of these only Biltz3 used reasonably pure metal. Biltz and his co-workers also treated zirconium tetrachloride with hydrogen sulfide in a heated tube. They derived a pure product only after reheating the initially obtained material in a hydrogen sulfide atmosphere. PaykulP had been unsuccessful with this reaction (1) Portions of this paper have been presented a t t h e 131st National Meeting of t h e American Chemical Society, Miami, Florida, April 9 , 1957. ( 2 ) E. FrPrny, A n n . ckim. fiizys., 131 38, 326 (1832). (3) W. Biltz, E. F. Strotzer a n d K. Meisel, 2. anorg. allpem. Chcnc.. 2 4 2 , 249 (1939). (4) J. J. Berzelius, A n n . P h y s . , 4, 125 (1823). ( 5 ) 0 . Hauser, Z . unorg. chem., 6 3 , 74 (19071. (1;) S . R. Paykull, Bull. SOc. chim., [ 2 ] 20, 65 (1873).
because he did not rigidly exclude oxygen. VanArkel and DeBoer7 passed zirconium tetrachloride and sulfur vapors over a hot wire and obtained pure zirconium disulfide but in poor yield. Finally, Hagg and Schonberg* reported its preparation by the action of hydrogen sulfide on zirconium metal in the temperature range 550-900 '. Experimental Materials.-Pure zirconium dioxide was prepared by addition of glycolic acid (3 moles) to a solution of zirconyl chloride (1 mole) and calcining the precipitated triglycoA typical sample conlatozirconic acid a t 900-1000 tained 0.001% Al, 0.0027, Ba, 0.02% Ca, 0.002% Cu, 0.001% Fe, 2% Hf,0.01% Mg, 0.002% Na, 0.01% Si, 0.0027, Ti. Zirconium tetrachloride was resublimed from a fused salt (KC1-ZrCla) melt as described by Horrigan.9 The zirconium metal powder was Titanium Alloy P grade ( -200 mesh) and contained 1-27, oxygen. Zirconim Hydride was obtained from Metal Hydrides, Inc. Traces of oxygen were removed from argon (originally 99.97, pure) by bubbling it through alkaline pyrogallol and drymg by passage through tubes containing PlOs. Gaseous impurities in tank hydrogen sulfide (99.8% pure) were minimized by inverting the tank and drawing off liquid H2S. Reagent grade carbon disulfide was used without further purification.
.
(7) A. E. VanArkel and J. H . DeEoer, Z. a m v g . Chern., 148, 345 (1925). (8) G. Higg and N. Schonberg, Arkiu. Kenti., 26, 371 (1951). (9) R. V. Horrigan, J . Ai'elnls, 7 , A I M E Trnizs., 2 0 3 , 1118 (1955).
