THE OXIDATIOY O F CHROMOTX TO CHROMIC SULFATE1 BURTOS B. IiSAPP
BSD
JAMES H. WALTOS
Department of Chemistry, University o j W i s c o n s i n , d f a d i s o n , W i s c o n s i n Receii,ed December 29. 195’6 I. T H E R E A C T I O S BETWEEX CHIIOMOUS SL LFATE A X D SULFURIC A C I D
TT’hile investigating the autoxidation of chromous sulfate it TT as found b y the authors that in the presence of metallic silver a qolution of this compound imdergoes oxidation TI ith the liberation of hydrogen. Thic paper deal. n ith certain conditions which influence this reaction.
Beagents ann‘ method The chroiiiou. ~ ~ i l f a nt ra- prepared by the electrolytic reduction of rhromic Eulfatc as deccrihed by Alqmanov(1). The ialt u-as precipitated hy alcohol and washed by cthcr, the+ operation. bring carried out in a n atmosphere of liydrogeii Ain e e w a r y precaution in the preparation of t h e salt is the removal of nioi-tiire, a. n-ell as disolved oxygen and pcroxides, from t h e ether u-ed i i l waihing. The chronioiis sulfate p i t a hydrate ohtaified by thi. mrtliod iTta\ 93 per cent pure, the impurity coniistinp of the chromic -alt By .raling the salt in tubes under an atmcspherc of nitrogrn, it TT as po-ihle t o licep it for n eeks a t room tcmpcraturc without any oxidation This c h o w that under these conditions there i i no apprcciablc reaction betncen tlic chroinous ialt and the natcr of hydration. “~Iolecular”~ I I ri T prepared by t h c method dewribed b y Comberg ( 3 ) n a s used as the c a t a l p t . I t was thoroughly mixed, thus providing a product n hirh gave s a m p l c ~of uniform surface. The method of mea-tiring the rate at n-hich the reaction 1)rorrcdcd ha. been d+crihrd by the heiiior author ( 2 ) . The rhroniou* .ulfatc w a i n cighed into gla*s capiiilc- TT h k h TT ere w-pended in the iiccks of ipecial reaction flaiky nhirh coiitairied 25 re. of u-ater and the wbLtaJicC nhoqe cffect was to bc s t u d i d . The Aaqk TT as placed in a thermostated chakiiig de.\ ice and connected t o 11 ater-jacbeted 1)urctL in which the rvolrcd hydrogen TT a‘: incaiiired Thc ctarting of tlic ~halicrdroppcd the capsule containing the chromou- ciilfatc into t h e fla-k, and after thc salt n a c dis1 This investigation 11 a i financed by a grant from the Research Committee of the University of \Visconsin, Dean I:. B. Fred, C h n i ~ ~ ~ i a n .
679
680
BURTOX B. KNAPP AND JAMES H . WALTON
solved the volume of hydrogen evolved was measured a t various time intervals. Since the reaction must be carried out in the absence of oxygen, the air in the flask and the buret was replaced by pure nitrogen. All measurements, unless otherwise stated, were carried out a t 25°C. THE EFFECT O F SHAKIKG
The agitation of the solution prevents supersaturation of the hydrogen, keeps the silver dispersed, and brings the catalyst into contact with fresh portions of the chromous sulfate, thereby making constant the time necessary for a chromous ion to reach a point a t which it will be adsorbed by the silver. In reactions which occur a t a n interface the shaking efficiency is of utmost importance. When stannous chloride solutions are oxidized by shaking with air, for example, a point is reached a t which more vigorous shaking is without effect on the rate of oxidation (5). On the other hand, cuprous chloride solutions when shaken with air show increased speed of oxidation as the shaking of the solution is increased (2).
Data
TABLE 1 OJ a typical r u n
_ _ _ _ _ ~ _ _ _
Time (in min-
I
Utes). . . . . . . . . 10 !20 /SO 140 ‘50 160 170 180 190 1100 Mg. of H,. . . . . 0.1001 0.1881 0.274 0 366( 0.458, 0.5511 0 . 6 5 2 ~0.754’ 0.8611 0.968 K X 103, . . . . . . 10.0 , 1 9 . 4 1 9 . 1 1912 9.2 9 . 2 9 . 3 , 9.4 9.6 9.7
.I
~
~
,
~
Total volume of hydrogen = 43.4 cc. (3.36 mg.) K = mg. of hydrogenltime in minutes.
T o determine the effect of shaking, experiments were carried out in n-hich the speed of the particular shaker used was varied from 590 t o 1162 and then increased to 1780 “shakes per minute.” The corresponding values for the half-life of the chromous sulfate solution agree within experimental error, showing that a maximum shaking efficiency had been reached. The data of a typical run are given in table 1. The solution contained 0.4 g. CrS04.5H20,0.1 g. Ag, and 1.5 g. Cr2(S04)3.5H20 in 25 cc. of 0.9 sulfuric acid. This will be referred to hereafter as the “standard solution.” The sulfuric acid and chromic sulfate were added because the former is a reactant, t h e latter is a product and, as will be shown later, each compound has a definite effect upon the speed of the reaction. By adding a n excess of sulfuric acid and chromic sulfate to the solution, however, the changes in concentration due to the reaction are 10 small that the effect on the velocity of the reaction may be disregarded. For the first part of this reaction the values of K correspond to zero order. The average is 9.4. In a duplicate run the value 9.36 was obtained. In general duplicate experiments did not m r y more than 5 per cent. The zero
681
OXIDATIOX O F CHROMOUS TO CHROMIC SULFATE
order holds only for about the first half of the reaction when the concentration of chromous sulfate is sufficiently great to saturate the surface of the catalyst .
Efect of chromic sulfate This substance acts as a negative catalyst. The data in tahlc 2 were obtained with the standard solution, using 0.5 g. of silver. The constancy of the data in the last line of table 2 shows that the retarding effect of the chromic ions is directly proportional to the increase in chromic-ion concentration until the higher concentration of the salt is reached. This effect can be explained by the adsorption of chromic ions by the silver. To test this point the conductances of dilute solutions of chromic sulfate were measured before and after the addition of 0.5 g. of molecular TABLE 2 Data obtained w i t h the standard solution
.I
I
Cr2(S04)3.5H1O (in grams). . . . . . . . . . 0 0.251 0.75i 1. O O ~ 1.25, 1.50,2.00' 3 00 Half-life (in minutes). . . . . . . . . . . . . . . . . 5 . 4 9 . 5 ~ 1 60 i19.5 24.8 828.2 i38.3 ~ 7 4 . 2 Increase in half-life 16 4 I14 1 14 1 15 5 '15 2 ~ 1 64 ~ 2 29 Crz(S04)a.5 H 2 0 (in grams) I
TA4BLE 3 T h e effect o j increase in acid concentration &So4 (normality). . . . . . . . . . . . . . . . . . . . . . . . . ' 0.904 Half-life (in minutes). . . . . . . . . . . . . . . . . . . . . . 92.4 Ho (acidity function). . . . . . . . . . . . . . . . . . . . . . . . 0.31
~
~
I
1.97 63.3 -0.11,
1 . ,
2.92 47.0 -0.41
~
~
3.94 30.1 -0.70
silver. The increase in resistance after shaking with silver showed that some of the salt had been adsorbed. Since the data are of a qualitative nature only, they are not given. With the higher concentrations used the surface of the silver became so saturated that the silver was practically without effect as a catalyst.
E$ect of acid concentration To find the effect of increase in acid concentration experiments were carried out in which the acid concentration was increased to about 4 N . The results are listed in table 3. The standard solution was used. At higher concentrations the speed of the reaction is approximately in direct proportion to the normality of the acid. With hydrochloric acid t h e same relationship was found to exist, the ratio of the half-life being about 10 to G for equivalent solutions of hydrochloric acid and sulfuric acid.
682
BCRTON B. KNAPP AKD JAMES H. WALTON
When the acidity functions of the sulfuric acid (Ho above) as determined by Haiiiniett and Deyrup (4) are plotted against the half-life, a linear relationship i,. obtained. The effect of the acid, consequently, is to increase the speed of the reaction in direct proportion to the acidity function. In the above experiments the reactions were zero order. With very dilute solutions of sulfuric acid, when the relative concentrations of chromous arid hydrogen ions had the value 2 to 1, second-order reaction constants were obtained.
Temperature e$ect With the standard solution the velocity was measured a t 25", 30°, and 35OC. The valueq of K in cc. per minute for these temperatures were TABLE 4 D a t a obtained using standard solution w i t h varying amounts u j silver Silver (in grams) . . . . . . . . . . . . . . . . . . . Half-life (in minutes). . . . . . . , , . . . . . . Silver (in grams) . . . . _ . . . . . . . . . ' 9 . 2 Reciprocal of half-life .___
.-
TABLE 5 -
Increase in adsorption w i t h increasing atomic weight
_SOLUTION
0.001 N 0.001 N 0.001 N 0.001 N
I
LiI KaI
KI RbI
0.139, 0.124, and 0.109, giving a temperature coefficient of 1.27. This indicates that the effect of temperature change is largely an effect on the rate of diffusion.
Effect of silver The data obtained uaing the standard solution n-ith varying amount5 of silver follow in table 4. When the concentration of the bilver is plotted against the reciprocal of the half-life a straight line is obtained, as evidenced by the constancy of the values of the last line in table 4. Kithin the limit3 of these concentrations the velocity of the reaction is directly proportional to the concentratioii of the silver. Thebe data furnish additional confirmation of the efficiency of the shaking of the solution.
OXIDATIOK O F CHROMOUS TO CHROMIC SULFATE
683
The effect of various salts Using the standard solution with 0.5 g. of silver results for quarter-life periods mere obtained with normal solutions of the following halides : potassium chloride, 14.2 min. ; potassium bromide, 105 min. ; potassium iodide, immeasurably slow. The quarter-life in the absence of these salts was 14.1 min. The quarter-life period was used here because of the slowneSs of the reaction. When the conductances of these solutions (0.001 N ) were measured before and after shaking 100 cc. with 0.5 g. of silver, the following data were obtained: potassium chloride, practically no change; potassium bromide, resistance change from 815 to 842; potassium iodide, resistance change from 810 to 915. These measurements show t h a t the effects of these salts may again be ascribed to adsorption. While no attempt has been made to study the nature of the adsorption, the data in table 5 are of interest as showing that with both anions and cations the adsorption tends to increase with increasing atomic weight. It was of interest to comI;are the effect of other sulfates with that of the chromic sulfate. Normal solutions of sodium sulfate, magnesium sulfate, and aluminum sulfate gave half-life values of 13,9.1, and 17.8 min., respectively, when the standard solutions containing 0.5 g. of silver were used. Without these salts the half-life is 7 min. The slight effect of these salts parallels their lack of adsorption, as evidenced by practically unchanged conductance before and after shaking with silver.
Summary
X solution of chromous sulfate in the absence of air is oxidized very slowly a t ordinary temperatures. Finely divided silver increases the rate very slightly. The addition of hydrogen ions in the form of sulfuric acid also causes a slight increase in the rate of oxidation. With silver and sulfuric acid together the reaction takes place rapidly. The presence of chromic ions retards the rate of oxidation. The speed of reaction is directly proportional to the acidity function of the solution, even when the concentration of the acid is forty times as great as the chromous-ion concentration. I n this reaction a chromous ion gives an electron to a hydrogen ion, and the latter is evolved in the form of molecular hydrogen. It seem9 probable that the chromous ions are adsorbed b y the “molecular silver” and t h a t they react with the hydrogen ions in solution. Increasing the hydrogenion concentration results in more impacts per unit of time. If the assumption is made t h a t the surface of the silver is continually saturated with chromous ions the zero order of reaction would be explained, inasmuch as the sulfuric acid is present in such a high concentration that it undergoes n o appreciable change during the reaction. I t seems probable then that the adsorbed chromous ionr react v i t h the hydrogen ions in the solution rather than with
684
BURTON B. XNAPP AND JAMES H. WALTON
adsorbed hydrogen ions, a n assumption that is supported by the linear relationship b e h e e n acid concentration and reaction velocity. The existence of a complex resulting as the product of an equilibrium reaction between chromous sulfate and sulfuric acid must not be overlooked. Such a compound might be more readily oxidized than the chromous sulfate, just as H2SnCI4is more rapidly oxidized than SnCln. The existence of such a complex is quite in accordance with many of the results of this study. Here, as in many other reactions of this kind, the lack of information concerning the nature of adsorption and the laws which govern the formation and behavior of adsorption compounds makes it impossible to do more than speculate concerning the mechanism of this reaction. 11. T H E AUTOXIDATIOK O F CHROMOUS SULFATE
For the study of the autoxidation of chromous sulfate by oxygen gas the apparatus used was similar to that described in section I, the only change being the use of an atmosphere of pure oxygen instead of nitrogen over the chromous sulfate solutions. When this solution was shaken the oxygen was used up, and the rate of oxidation could be followed by measuring the rate of disappearance of the oxygen, the method being similar to t h a t used in the study of the autoxidation of stannous chloride (2). Inhibitors The reaction between oxygen and chromous sulfate was found to be practically instantaneous. In a n attempt to find any similarity between this reaction and the autoxidation of stannous chloride, a number of compounds which inhibit the latter reaction were investigated from the standpoint of inhibiting the chromous sulfate oxidation. Small amounts of the following compounds when dissolved in the chromous sulfate solution were found to be without effect: tartaric acid, acetanilide, succinic acid, thiourea, chromic chloride, ethanol, benzene, toluene, aniline, trinitrotoluene, picric acid, p-toluidine, p-aminophenol, pyrogallol, pyridine, n-butyl alcohol, iodobenzene, glycine, potassium cyanide, ammonium chloride, and acetaldehyde. RIany of these compounds, especially those containing nitro groups, were extremely potent as inhibitors of the stannous chloride reaction ( 5 ) . It should be noted that these nitro compounds were in general reduced b y the chromous sulfate. The presence of sufficient water to dissolve the chromous sulfate seems to be essential to its rapid oxidation. In a few experiments the solid chromous sulfate was added to such liquids as acetone, formic acid, acetic acid, and ethanol. The salt is practically insoluble in these reagents, and in this state undergoes practically no oxidation.
OXIDATIOX O F CHROMOYS TO CHRONIC SVLFATE
68.5
Coupled o x d a t i o n In thc autoxidation of acid solutions of +taniioiii chloride (5), the p r e w i c e of allyl alcohol cauies the absorption of an excess of oxygen, oning to the induced oxidation of the latter by a peroxide intermediate. The autoxidation of chroniou. wlfate caiiied no coupled oxidation n ith allyl alcohol nor n-ith mjiuni arsenite, qodiiini iulfitc, or ferrous sulfatc I t v a s found, hov c w r , that nhile a nmtral aqueou? rolution of qtaiiiioiichloride a b s o r b oxygen very slowly (0.1 cc. per hour), coupled oxidation of the \taniioiiy chloride occur-, in the presence of chromouq siilfate. The coupled oxidation n as complete within t n o minute.. This oxidation i-, not due to the chromic biilfate formed, for thii compound liaq no effect on the spced of the autoxidation of the itannoiiq chloridP. Using equiniolecular amounts of chroiiioiii sulfate and btannoui chloride, all of thr chromous sulfate waq oxidized aiid about 52 per cent of the staniious chloride. The anioiint of coupled oxidation depciidq upon the relative amomits
~
I
C1.
Polarogram of chromous sulfate arid chromic sulfate
of itannoii. cliloridc and chrcmous 5ulfate. It increaied n ith the amount ulfatc until 95 per cent of t h e ~taniioiischloride was oxidized. T h e lice of oxygeii in.tcad of air increa.cd the amount of coupled oxidation. -411ii1crca.e of the oxygen p e w i r e to 2 atmospherei i n c r e a 4 thc amowit of itannouq rhloridc oxidized hy 20 per cent. The anioiiilt of coupled oxidation thu. appear. to he dependent on the amount of chromous qulfate and di~-olvedoxygcm in +elution. Expoiure to ultra-xiolct rays froni a mercury arc light had no effect 011 the amount of coupled oxidation. T h e v re+ult*are best explained by the peroxide t hcory of autoxidation ; this postulate. the formation of a very reactive intermediate peroxide hich caui.ei the coupled oxidation of certain oxidizable wb+tances. Several polarogranis of chromic and chronioiiq ohtaiiicd 1)y Prajzlcr (ti). The fir4t hrcak ilidicstes the reductio11 of cliroiiiic ion to cliromou. io11 aiid tlic. +coiid the reduc*tionof cliromoui ion t o cliroiiiium. \T7hcn y i i c w y ~ i v esmoiuit. of oxygen v, ere h1o~v-nthrough a eliromou\ .ulfate wlution, a \ reprewited in curves 3 and 4, and 5 and 6, a third brcak in the c'urvc was found. Thi. third break indicate\ the lirewicc~of anothcr c'oinpoiund TI itli a higher oxidationrcductioii potential. It can not be .tatccl tlcfinitcly that thi; compound i y a pcroxidc, but theriiiodyiiaiiuc~ally it 11%. a11oxiclatioii-recluc.tion potential of the riglit order of magnitude. s21111 Vl nl
y
1. Tlic cffwt of a i1wiil)c.r oi organic :inti inorganic cmnpowid~on the oxidation of chromou. iulfatr by oxygen hay been itudied. 2 . Staiiiious c~hlorickwa5 the only yiiliytaiice I\ lie+ oxidation could bc induced by tlic autoxidation of cahroniou\ \illfate 3 . Evidence hay been found for thc formatioil of a pcroxidc intermediate in the autoxidation of cliromou- \ulfato.
