tanium ester solutions is consistent with the existence of two equilibrium constants
The weakness of the TPT-ethylamine interaction was unexpected in view of the relatively high stability of the reported TiC14-aiiiinc comKl = . Y ~ I P ~ Tl ~ ~ ~ , ~ ~ ~ ' H~ z ~ ~ l ' ~ - ~ ~ ~ ~ s l l ~ ~ ~ ~ ~ ~ s pounds. The comparative stabilities of the TP'Y K? = I ~ T P T.E t NII? /zTTl'T. E t?i til YE t N Ji2-I-E t h11% and Tic14 complexes possibly reflect the relative The solid curves on Fig. 2 represent t h a t varia- abilities of the isopropoxide and the chloride groups to the Ti(IV) atoms to which tion of ~ E t h E 1 2 i ~ E , N Hwith 1 X ~EtNH2 t ~ ~ ? ~to~supply ~ s ~ electrons ~ concentration to be expected from (1) and (2) for they are bonded. Alkoxide groups, being less the case where K1 = 3.7 arid K s = 2.5. Behavior electronegative than chlorides, should behave as observed a t 35' can again be accounted for by (1) more effective electron donors to the vacant orbitals and (2), with the best fit of the data being given of the Ti(1V) atoms and accordingly Ti(IV) alkby ATI43.4,Ks-2.3. Less satisfactory fits of the oxides might be anticipated to act a? weaker Lewis data are obtained by assuming that the coniplexing acids than Tic&. Although isopropoxy groups are more bulky of EtNHs is accomplished solely by ( I ) or (2). than are chlorides, the formation of TPT-ethylThe magnitudes of the constants k', and k ' z indicate that the coiirdinatiori compounds TPT. amine complexes is n o t precluded by steric effects. EtNHz and TPT.2EtNH2, if indeed they exist in This can be demonstrated by construction of a solution, have only a low stability. Application of Fisher-Hirschfelder scale model of octaheclral the van't Hoff equation to these K's a t 23 TPT,2EtNH2. EtNH? groups are ol)serr-etl t o and 35' indicates that the cnthalpy change for fit into such a structurc witliout strain. (1) and (2) is < -2 kcal. TVII IIIVG I o u , Ilrr 4i?r
[CONTRIBUTION FROM THE CIIEMICAI. AND ?rfEI'ALLURGICAL D I V I S I O X , SYLVAN1.4 ELECTRIC PRODTCTS, ISC
1
Preparation, Properties and Structure of Cadmium Peroxide BY
c. W. W. HOFFMAN, R . c. ROPP .\ND
Itidun l!l?L' ( 2 ) H . lpiippl 2.a v o y g . a l l g f m . flieni., 291, 12 (19.7)
moil method is the reaction of a metal salt, usually in basic solution, with hydrogen peroxide. I n addition to these preparations, thermal decomposition of the superoxides may yield the peroxides and a recent paper3 describes the reactions of various nietal salts with alkali superoxides in liquid aiiinionia to obtain certain peroxides. Cadmiuni peroxide of doubtful purity has been reportedly prepared by all of these methods except by the decomposition of a defiiiitely established superoxide of cadmium. lIanchot4 reported that the low-temperature oxidation of cadmium produced some CdOz. Teletof,; and later Perkins," utilized hydrogen peroxide to prepare inaterials which probably were fairly pure CdO?. Perkins, in addition, also tried to prepare this peroxide from cadmium dimethyl in ethereal solution with hydrogen peroxide. Perkins' work is the most recent and authoritative discussion of this method. He gives, in addition, references to enrlier inconclusive CdO? preparations. However, Perkins, largely because he depended upon chemical evidence alone, was unable to establish definitely the presence o i CdOp. Recently, Schechter and Kleinbergs clainied a reaction product rich in the corresponding peroxide upon treating cadmium salts in liquid ammonia with alkali superoxides. ) 1). I,. Scliechter :ind J , RI?ini)erK. 'I'III?
w
~ r a n r ~ l o itj ,P i . , 42, (i!im), T e I F t o f , J . R J L YP Y h y ~ Chenr, . .S,>c- , 4 3 , 121 0!111 ) 'I1 I< I ' r r k i i i S . .I ( ' I i t i ~ ! . 5 . 5 ) was necessary for the precipitation of the peroxide. However, the use of strong bases such as KaOH is not recommended because of the possible formation of the cadmate ion. Fa202 may be substituted for HpOp, but CdC12 solutions cannot be used since they lead to the formation of basic cadmium chlorides. Per cent. cadmium was determined by electrodeposition and/or gravimetrically as the sulfate. Anril. Calcd. for CdO?: Cd, 77.8. Found: Cd, 75.1 f 0.6, based on 13 different samples having an average water content of 2-35;, Oxygen was determined in two different ways, neither of which was entirely satisfactory. I n the first method, CdO? was deconiposed in a closed system and the resulting oxygen collected over water. After decomposition, the powder t h a t remained was identified as CdO by X-ray diffraction. Materials partially converted from CdOy to CdO were obtained and their X-ray patterns examined. Only the lines due t o unconverted CdOt or CdO could be found. Unfortunately, CdOp decomposes violently and therefore the conversion t o CdO is difficult t o control. Anal. Calcd. for CdOp: On, 22.16. Found: 0 2 , 21.0 f 0.3. T h e second method involved the determination of "active" oxygen by titration with potassium permanganate after dissolution in sulfuric acid. Anal. Calcd. for CdOY: On, 22.16. Found: 01, 18.4 f 0.2, based on 5 different samples.
per minute were also obtained giving 20 values in good agreement with the powder photographs as well as more accurate relative intensity values. The experimental arrangement used in the latter method eliminated the need for any absorption correction. The observed intensities were measured on the diffractometer tracings by graphical integration. The resulting values were corrected for changes in slit width giving the arbitrary 10values listed in Table I . High-purity tungsten was used as a n internal standard for the 28 measurements. Values calculated for the cell dimension of tungsten agreed closely with published values.8 The samples did not deteriorate during exposures, which normally varied from 3 to 5 hr. Several long exposures of 12 to 20 hr . also were taken to detect and measure the weaker reflections. T h e lines a t d = 1.77 and 1.47 were taken from these films and given an arbitrary intensity of five since they could not be measured accurately from the diffractometer traces.
Structure Determination Initial Debye-Scherrer patterns were readily indexed as cubic (Table I) with the systematic absence of hkl reflections with h k, k I, I h= 2n 1 indicating face-centered symmetry. However, longer exposures showed five additional weak reflections consistent with simple symmetry having h2 k 2 l2 = 5 , 6, 9, 13 and 14. Theisopetric unit cell has an uo value of 5.313 f 0.003 A. obtained as an average of the 16 most intense reflections listed in Table I. The systematic absence of hkO reflections with h = 2n I, especially the absence of the low angle reflections from the 100, 110 and 310 planes indicated that the space group was Th6-Pa3. Following the notation of the International table^,^ it was apparent that the pyrites-type arrangement of dumbbell-shaped anion pairs around the special positions of (4b) with metal atoms in the
+
+
+
+
+ +
+
(7) R. L. Stone, Ohio S t a t e Univ. Engineering Series, Bulletin N o , 140, 1981. ( 8 ) H. E . Swanson and E. T a t g e , Natl. Bur. S t a n d a r d s (U. S.) Circ. No. 539, Vol. 1, 1953, p. 28. (9) "International Tables for X - R a y Crystallography," Kynoch Press, Birmingham, 1952.
: (ti) I ? . Nikitina, S b o r n i k S l o f e i Obshchei Ciiim. A k a d . Nauk .S.S.CR. 1, 41(195.3),
(7) "Gmelin's
Handbuch der anorganishen Chemie, Wolfram," V e r l s g Chemic. 1Veinheim. 1933, p. 130,Vol. 5 4 . ' 8 ) r). Rettinfier a n d S Y, Tyree, 'THIS J O ~ J R N A I . ,79, :33;13 (1057) t!)) 13. Rurhholz %. c i u o r g . CIiem., 244, 149 (1941)).
phenomenon, similar to silica,'O or by cation cxchange, similar to the polyuranates." This investigation was undertaken to clarify the state of knowledge of the tungstic acids. Fifty precipitations were made with HCl or HNO3 a t concentrations of 0.2 to 9.0 ,V and a t temperatures of 25, 50 and 100'. The precipitates were characterized by chemical analysis, X-ray diffraction and potentiometric titration at high ionic strength. The ranges of homogeneity of the unique phases were estimated and their reactions were studied. A series of molybdic acids were prepared for comparison. The chemistry of the tungstic acids then was related to the chciiiistry of W(V1) compounds in aqueous solution and compared with that of Cr(VT), Mo(V1) and U(V1) compounds. Structural Considerations.-The compounds previously written as H2W04.Hz0 and H21V04 are tungstic oxide hydrates rather than acids because they are isomorphous with the yellow - \ 1 0 0 3 . ? & 0 and 11003.Hz0compounds which have been shown to be hydrates.'* Also, the known structure of lIoOa,3H2013,14 defines that of \V03.2Hz0. 110) G . Sears, Anal. Chem., 28, 1981 (1956). 111) J . Maly a n d V. Vesely, J. Inorg. .Tirrl
Cheiii.,
(1958). (1'2) 3 hIaricic and J . S m i t h , J C h c m . .Sot- , SSI; (l!lS'i. (1:O I I.inrlQvist, .lifn Ciiciir S r n , i d , 4 , ti311 f l ! ) X ) I 1 I! J l.indc)vi\t, ! h i d , 10, IXOL' l!I5l7~j.
7 , 1 I'J
