MEYERL. FREEDMAN
2072 [CONTRIBUTIOS FROM THE
Vol.
so
REFRACTORY METALS LABORATORY, GEXERAL ELECrRIC CO.]
Polymerization of Anions : The Hydrolysis of Sodium Tungstate and of Sodium Chromate BY MEYERL. FREEDMAN RECEIVED OCTOBER 5 , 1957
A coordination theory for the formation of isopolytungstates was developed from a study of the modc of polymerizatirtn of WO4-- and of Cr04--. The pH of Na2WO4 and of NazCrO4 solutions was measured as a function of concentration, ionic strength and time. The relation between polyion charge and polymerization number was determined from the slope of linear pH-log C plots. Structures of polynuclear species, consistent with hydrolysis equilibria, were deduced from various considerations based upon the principle that chromium retains tetracovalency while tungsten reversibly expands to octahedral coordination. The primary aggregation process of wo4-- produces linear polymers of doubly linked octahedra ill which one of each pair of shared oxygen atoms also binds a proton. Secondary aggregation processes involve coordination of this “01” group to produce tetracovalent oxygen. Direct condensation of polyacids takes place in more acid solutions.
Introduction The hydrolysis studies of Na2W04and of KaaCrOd Tungsten is one of the best known iso- and hetero- reported here were made to determine the mode polyacid forming elements, while polymerization of polymerization of the anion during the initial of chromate yields only simple pyro ions.2 The stages of aggregation. A dilution method was principles governing the formation of these complex used so that the normal tungstate or chromate ion, anions are not known. Classical theories, such as respectively, would be the predominant species a t those of Miolati and Rosenheim, have been made all times. Under conditions where the classical obsolete by the many crysthl structure determina- hydrolysis constant ratio may be applied, that is in tions of polyanions which have been r e p ~ r t e d . ~dilute solution or a t constant ionic strength, the Information is required concerning the aggregation slope of the dilution curve fixes the relation between processes in solution which lead to the formation of polyion charge and polymerization number.8 the known solid phase structures. Previous work nA-~ H * O+-+ A?: POI$in this area has been limited to determinations of the average composition of polyions in solution without regard to the mode of polymerization or to z = %n - p is required for electroneutrality while structural considerations. LE- = -1The sulfate, chromate, molybdate and tungstate dlogC 1 + p ions are all believed to be tetrahedral complexes, both in the solid state and in ~ o l u t i o n . ~On follows from the condition of ecpilibriuni. The acidification the chromate ion is converted to composition and structure or‘ the polyion car1 then dichromate which is believed to retain tetrahedral be deduced for each value of n by assuming a value coordination.2 Acidified tungstate solutions yield for the coordination number. Dilution curves were obtained on both a one hour paratungstate, metatungstate and tungstic acid and a one month basis so that initial as well as solid phases in which octahedral coordination prevai1s.j Dissolved polynuclear species in acidified final equilibria could be observed. Molar solutions tungstate solutions have been reported as various of NaaCr04 and of Na~WO4 were diluted with hexatungstate ions as well as a tritungstate ion, all 3 M NaN03 and with M Na2S04 to maintain an of unknown structure.6 Octahedral coordination ionic strength of 3.00. Water was used as the is assumed here also by analogy with the known diluent to study the hydrolysis in dilute solution. water-derived solid polytungstates. The poly- NaN03 rather than KaCIOa was selected as the tungstates obtained from fused melts are exceptions inert supporting electrolyte in order to retain the in that the coordination number of tungsten de- desirable saturated KCI salt bridge. The Na2SOA pends on the ratio of tungsten to oxygen. Thus, was used to test for interaction between hydrolyzed the ditungstate obtained by fusion has a structure and unhydrolyzed species. The Sod-- is inert to containing both octahedral and tetrahedrally co- hydrolysis in the pH range under consideration and ordinated tungsten.’ In water solution it is the might be expected to imitate the action of both pH rather than the tungsten to oxygen ratio which Cr04-- and LX’04-- in complex ion formation. An electrostatic W to 0 bond has been postulated determines the coordination number of tungsten. for the polytungstatesP However, the fact that (1) Presented before the Division of Physical and Inorganic Chemisthe known structures violate three out of four of try, 132nd National Meeting of the American Chemical Society, S e w Pauling’s rules for the linking of polyhedra in ionic P o r k , N. Y., Sept. 13, 1957. crystals is evidence for covalency. The slow rate (2) A. F. Wells, “Structural Inorganic Chemistry,” Oxford University Press, New York, N. Y . , 1947. of oxygen exchange observed by Spitsyn:O in his
+
(3) “Symposium on t h e Structure a n d Properties of Heteropoly Anions,” 130th Meeting of American Chemical Society, Atlantic City, N. J , , Sept. 1956. ( 4 ) I. Lindquist. N ~ J WAcln R ~ ~ i t iSeo r SCi b’psaliensis, 15, Sci. I\’. No 1 (1960). ( 5 ) J . C. Bailar, Jr., Editor, “The Chemistry of t h e Coiirdination Compounds,” Reinhold Publ. Cory., New York, N. Y . , l95li ( G ) D J. Bettinger and S. Y. Tyree, Jr., THE J O I I X N A I , , 79, 3855 (19.57). ( 7 ) I. Lindciuiat .Ir!a C i i c r i i .Siiziiil.. 4 , lotiti (l!lXl).
+
(8) Kuan P a n and Tong Aling Hseu, Bicll. C i ! e n i . SOC.J a p n l t , 26, 126 (1933). T h e y obtained different results in applying a similar CrOd a n d ITa2W0,. However, their r:ite of dilution appears to ha\.e been slow. Also, they used alcohol i l l rrcrystallizing t h e salts. I t has been shown b l - J . Bye, A n n , Ciiiiii , 20, 403 (1945), t h a t alcohol can produce containinalion b y isolioly s,tlts (9) I,. Pauling, “h‘aturc of the Chemical Bond,” Cornel1 Uiiiversity P r e s Itlidca, K. Y . , 19-40. (10) \’. I . Spithyn. C . . I 60, l ( l . 1!lI il’,J,-~Ii),
May 3 , 1958
HYDROLYSIS OF SODIUM TUNGSTATE AND SODIUM CHROMATE
isotope studies further supports the covalent bond concept. It should thus be possible to derive the known polyanion structures by the rules of coordination chemistry, provided that the mode of polymerization is known. Experimental Reagents.-Five lots of Folin grade NapWO4.2HzOwere obtained from three manufacturers. Molar solutions prepared from these salts and purified water gave p H values ranging from 9.4 to 10.3. Values reported in the literature range from 9.158 to 10.5.’] After repeated recrystallizations a pH of 9.05 was obtained. The recrystallized salt analyzed 10.94 f 0.01% H20 (fusion) and 70.27 f O . O l ~ oWOS (precipitation and weighing WOa).12 Calculated values are 10.92 and 70.28’%, respectively. Neither powder X-ray diffraction patterns nor chemical analyses were clearly affected by recrystallization. Potentiometric titration of the molar solutions with 0.1 N HC1 showed the presence of from 0.01 to 0.037, of excess alkali, calcd. as NaOH, in the salts. Extrapolation of the common, straight line portion of the neutralization curves indicated a theoretical p H value of 8.85 for the molar solution. Since this degree of purity could not be attained, dilutions were made with the series of salts to determine the effect of the excess alkali. Other reagents were of analytical reagent grade. Distilled water was purified by condensing the middle portion of distillate from a Pyrex fractionating column in a quartz condenser. The purified water and all solutions were stored in polyethylene bottles under nitrogen pressure. Apparatus.-A dilution cell was constructed from an 8 oz. polyethylene bottle by sealing on side arms capped with rubber sleeves through which electrodes were inserted. Diluents were added from a Pyrex buret which fitted into the cell. Air was excluded by passing a stream of nitrogen which had been bubbled through NaOH and H2S04 solutions. The Beckman 1190-60 glass and saturated calomel electrodes were used with the Model G potentiometer for p H measurement. The outer barrel of the calomel electrode was replaced by a polyethylene tube through which a small Pyrex fiber was sealed. Temperature was maintained a t 25 f 0.2” by means of an external infrared heater controlled by a mercury thermoregulator. Magnetic stirring was used. The Beckman standard buffer solutions were used for calibration. Procedure.-In the “forward” dilutions 50 ml. of test solution was pipetted into the dilution cell and increments of diluent added from the buret. p H readings were taken after each addition when constancy was observed for a 1minute period. Periodically, when the total volume reached 150 ml., 100 ml. of the solution was removed by pipet and stored in a polyethylene bottle a t 25”. The p H of the stored solutions were measured a t intervals of a week or more until constancy was observed. Salt solution diluents were adjusted to a p H of 6.8-7.0 with 0.05 N NaOH. In the “reverse” dilution 100 ml. of diluent was placed in the cell and titrated with 25 ml. of test solution.
Results and Discussion (A) Na2Cr04 Hydrolysis.-The dilutions required approximately one hour. Final p H values were measured after one week and did not change appreciably after a second week. Similar dilution curves were obtained with freshly prepared NanCrOl solutions and with solutions which had aged for 10 days before dilution. The slope of the dilution curve defines the reaction which takes place as the equilibrium is shifted by dilution. Equilibrium concentrations of HCr04- or Cr207-- present in the original solution do not interfere, since the degree of hydrolysis of the predominant Cr04-- increases as dilution progresses. (11) K. Saddington and R. Cahn, J. Chem. SOL.,3526 (1950). (12) “Scott’s Standard Methods of Chemical Analysis,” 5th Ed.. D. Van Nostrand Co., New York. h’. Y.,1939.
2073
Dilution curves are shown in Fig. 1, while the indicated reactions are summarized in Table I. The dilution curve with water is considered only a t low concentrations where the activity coefficient ratio can be assumed to be constant. Similarly, the dilution curve with Na2S04 is considered only a t low Na2Cr04 concentrations where the Sod-concentration is approximately constant. 9.5
85
PH.
75t,,,
I,, ,
,, , , ,
I,,, ,\\J
,, , , ,
,, ,
YOLS
Fig. 1.-Dilution
0001
001
010
10
Na2Cr04
PER
LITER,
curves for KazCrOa. Final values were taken after 1 week.
The initial hydrolysis reaction a t low ionic strength appears to be the binding of two protons by a hydrated CrOd-- to produce a labile neutral molecule. This initial product rearranges slowly a t low ionic strength to chromic acid, which is largely ionized to HCr04-. The rearrangement may involve the transfer of protons from bound water of hydration to the central Cr04--. This rearrangement is more rapid a t high ionic strength, since HCr04- is the initial product observed in the dilution with NaN03 solution. The near linearity of the NaN03 curve shows that Nos- and Cr04-interact only to a minor extent. However, sod-reacts to produce a sulfato-chromate complex ion as an initial product. Both HCr04- and Crz07-are unstable a t high ionic strength. The final dilution curves with both NaN03 and Na2S04 solutions coincide and show Cr207-- to be the final hydrolysis product in both cases. Since the final equilibrium is a t a lower p H thaii that producing HCrOd-, the latter ion does not appear to be an intermediate in Cr207-- formation. Direct condensation of HCr04- to CrLC)7-- has been postulated by Xeuss and Rieman.13 They assumed a mechanism Cr04--
Ki
+-+
Kz
HCrOl-
