STRUCTURE OF AND METAL-METAL BONDING IN Rh(CO)2

and perchlorate of the order of 2 X 10~4 molar ... for Max Wien1 andall subsequent experimenters invariably ... of the metal carbonyls and related com...
2 downloads 0 Views 265KB Size
1761

April 5,1961

COMMUNICATIONS T O T H E E D I T O R Sir :

NEGATIVE WIEN EFFECTS

We wish to report that solutions of uranyl nitrate and perchlorate of the order of 2 X low4molar concentration exhibit a decrease in conductance upon the application of high electrical fields. This is in contrast with the usual behavior of electrolytes, for Max Wien' and all subsequent experimenters invariably have observed an increase of conductance of electrolytes under such circumstances. The decrease in conductance, expressed as the high field conductance quotient, AX/X(O), a t 65' amounts to 1.3% a t a field of 200 kilovolts/cm., and is many times greater than any possible experimental errors. The figure 1.3% is corrected for the Wien effect of the reference electrolyte, and should be considered a fractional change of conductance on the part of the uranyl nitrate or perchlorate alone. The phenomenon has not been observed with any other electrolytes thus far tested, including uranyl sulfate and fluoride. The phenomenon is dependent upon temperature. Both uranyl nitrate and perchlorate exhibit small positive Wien effects a t 5 O , small negative Wien effects a t 25" with multiple crossover of the zero conductance quotient axis as a function of increasing field, and a t 65" a sizable negative Wien effect or decrease of conductance a t all fields. I t is also dependent upon PH. The uranyl salt solutions are themselves hydrolyzed, but if acid or base is purposely added, or if a strong electrolyte such as potassium nitrate is added, the magnitude and trend of the effect as a function of field is decidedly altered. The effect of adding acid is the most significant, decreasing the magnitude of the negative Wien effect. The decrease in conductance with application of field may be due either to a decrease in the number of conducting ions or to an actual change in the method of conduction. These results and others to be reported later dealing with a high field conductance study of uranyl ion solutions2 lead 11sto conclude that the reaction U0zc+

+ Ha0

UOsOH+

+ H+

the occurrence of which is favored by higher temperatures as part of the hydrolysis scheme of uranyl is responsible for the phenomenon in these solutions. Under the influence of the applied field, hydrogen ion, with its abnormally high mobility, may be expected to overtake and collide with the slower-moving uranyl ions, reversing the above hydrolysis reaction and decreasing the number of conducting species. If such a mechanism is responsible, then it should be possible to demonstrate the phenomenon of negative Wien effects in other electrolytes, particularly with aquo complexes of transition metal ions. If a change in mechanism of conduction is involved, (1) M. Wien and J. Malsch, Ann. PhysiA, [4] 88, 305 (1927). (2) J. P. Spinnler, Dissertation, Yale University, 1961. (3) J. A. Hearne and A. G. White, J . C h s n . SOL, 3168

(1957).

it is not easy to predict in what kinds of chemical systems the decrease of conductance upon application of a field should be sought. STERLING CHEMISTRY FREDERICK E. BAILEY,JR. LABORATORY JOSEPHF. SPINNLER YALEUNIVERSITY NEWHAVEN,CONNECTICUT ANDREW JR. PATTERSON, RECEIVED FEBRUARY 8, 1961

STRUCTURE OF AND METAL-METAL BONDING IN Rh(c 0)&1

Sir: A detailed X-ray investigation of Rh(CO)&l has produced results of unusual interest, which not only clarify previous work but also appear to provide some insight into the nature of bonding of the metal carbonyls and related compounds. The compound, generously made available to us by H. Sternberg and I. Wender of the Bureau of Mines, was first prepared by Hieber and Lagally.' Their freezing point depression data established a dimeric species, Rh2(C0)4C12,in solution. Hieber and Heusinger2proposed a dimeric planar chlorinebridge structure which has been accepted by From infrared studies both in solution and Nujol mull, Yang and Garland6and later Hinds6 suggested a second possible model with only RhRh bonds linking the dimers. Further infrared, magnetic susceptibility, and dipole moment studies by Wilt? showed the compound to be diamagnetic with dipole moment p = 1.64 f 0.03 D in benzene. Wilt7 postulated a third structure involving the intersection of two planar Rh(C0)zCl groups along the C1-Cl line with a dihedral angle not equal to 180'. Precession and Weissenberg pictures revealed the crystals to be tetragonal with lattice constants a = 14.23 A., c = 9.32 k. The calculated density for 16 Rh(C0)ZCl species per unit cell is 2.74 g./cc. in good agreement with an experimental density of 2.72 g./cc. determined by the flotation method. The probable space group, I a d , was determined from systematic absences and the structure ultimately found. This space group has 16-fold general positions which results in one Rh(C0)2C1 species in the asymmetric unit. A complete three-dimensional Patterson analysis of intensity data taken with hIoKa radiation located the rhodium positioiis; the other atoms were found from subsequent three-dimensional Fourier maps. Isotropic thermal refinement by (1) W. Hieber and H. Lagally, 2. anorg. u nilgem. Chem., 261, 96 (1943). (2) W. Hieber and H. Heusinger, Angeu. Chem., 68, 678 (1956). (3) L. Vallarino, J . Chem. Soc., 2287 (1957). (4) H. J. Emeleus and J. S.Anderson, "Modern Aspects of Inorganic Chemistry," D. Van Nostrand Co., Inc , N e w York, N. Y., 1960, p. 309. ( 5 ) A. C. Yang and C. W. Garland, J . P h y s . Chem., 61, 1510 (1957). (6) L. de C. Hinds, S.B. Thesis, M.I.T., May. 1958. (7) J. R. Wilt, S. B. Thesis, M.I.T., May, 1960.

1762

COI\IMUNICATIONS TO THE

EDITOR

VOl. 53

nieric compounds of the type [Fe(CO)yX]z (X = S, Se, SCzHB,SeC2HB)possess chalcogen bridges and non-planar configurations. Our results support this proposal, and we furthermore believe that these compounds possess bent metal-metal bonds. We wish to acknowledge the financial assistance obtained from the Petroleum Research Fund (No. 471A), and the use of the facilities of the Kunierical Analysis Laboratory a t the University of Wisconsin. (22) National Science Foundation Fellow. '\~56'-7' ',..:I

'\"/

Fig. 1.-Bond

lengths and angles in Kh(C0)ZCl.

LAWRENCE F. DAHL DEPARTMENT OF CHEMISTRY OF WISCONSIN CALVINMARTELL UNIVERSITY ~UADISON 6, WISCONSIN DALEL. W A M P L E R ~ ~ RECEIVED FEBRUARY 20, 1961

least squares8 resulted in a filial discrepancy THE FREE ENERGY OF SILICA factor of R1 = C1.77~ for 230 observed reflections. The principal features of the structure are given Sir: in Fig. 1. Two essentially planar Rh(CO)&l Experimental data on certain metallurgical regroups intersect a t an angle of 124". The re- actions a t high temperatures point to an error sulting dimers apparently are linked by direct in the presently accepted value for the free energy Rh-Rh bonds to form infinite chains. A Rh- of SiOz. Four distinct groups of data lead to Rh distance of 3.31 A. is consistent with X-ray the same conclusion. data for other metal-metal bonds9-l1 which in1.-Kay and Taylor1 have repeated the earlier volve the coupling of unpaired electrons. In measurements of Baird and Taylor2 on the reaction order to explain the compound's diamagnetism SiOp(c) + jrC(graphite) +SiC(6) + 2CO (1) in the solid state we propose a second so-called bent metal-metal bond as indicated in Fig. 1 which -1ccording to their measurements the equilibrium would result from the overlap of d2sp3hybridized pressure of CO reaches 1 atm. at '1800" K. Using u-type orbitals a t an angle of approximately 56". the accepted values3 for the free energy of Si02 and Evidently in solution the weak metal-metal bonds CO, the equilibriuin data lead to a free energy of break to give dimeric molecules. The interatomic S i c which is less negative by 5 kcal. than that distances and angles are in accord with the pro- found by Humphrey, Todd, Coughlin and King.4 posed octahedral environment about each Rh atom. The data of the latter investigators have received Semi-quantitative molecular orbital argumentsI2 experimental confirmation in the work of Chipfor bent metal-metal bonding involving primarily man, Fulton, Gokcen and Caskey5 and more rep orbitals have been presented for C O ~ ( C O ) ~cently that of Kirkwood and Chipman6 using Cz(C6H5)e.l 3 3 l 4 measurements of the solubility of S i c in molten Our work on llh(C0)2Cl suggests that bent iron and lead, respectively. The agreement sugmetal-,metal bonding involving octahedral hybrid- gests that the equilibrium of reaction (1) be used ization occurs in Coz(CO)816 and Co2(CO)9CzH2.16In to calculate the free energy of Si02 rather than the fact octahedral co6rdination rather than square py- reverse calculation which has been reported by ramidal coordination1' readily can be rationalized S ~ x i l t e n s . ~ from the structural results for Fe2(COH)2(CO)6Using the tabulated data of Humphrey, et u Z . , ~ C Z M ~and~ Fe3(C0)8[JCGHS)?C~J2.20 ~ ~ , ~ ~ for thc free energy of 8-Sic and the accurately On the basis of infrared and dipole moment nieas- known datu for CO, the free energy of formation urements Hieber and Beck?' proposed that di- of SiOL a t 1SOO" K . is -138.4 kcal. Coughlin3 gives - 1333.35 for cristobalite and essentially the (8) W , R . Busing and H. Levy, ".2 Crystallographic Least Squares same figure for the other crystalline forms. The Refinement Program for the I.B.11. 704," Oak Ridge Sational Laboratory, 1959. equilibrium data thus lead to a result which is 5 (9) L. F. Dah1 and D. L. W a ~ n p l c r ,J . 2 1 ; ; ~ Chcfia. . Soc., 81, 3160 kcal. iiiore negative than the "third-law" value ( I 959). IL-lYheii a iriolten Fe -Si alloy is equilibrated (10) L. I?. Uahl, E , Isliishi u i d IC. E.iiundlc, J . Cheua. P i i y s . , 26, with ai1 atiriosphere of He arid HzO in a silica cru1750 (1057). (11) F. C. Wilson and I). P. Shoemaker, ihid., 2 7 , SO9 (1957). cible, equilibrium (2) may be established (12) D. A. Brown, J . C h e m P h y s . , 33, 1037 (1960). (13) H . W. Sternberg, H . Greenfield, R . A. Friedel, J. Wotiz, R. Markby and I. Wender, J A m . Chem. Soc., 7 6 , 1457 (1954). (14) W. G . Sly, i b i d . , 81, 18 (1959). (15) G . G. Sumner, H . P. Klug and L. E. Alexander, abstracts of papers, National Meeting of the Smerican Crystallographic Assnciation, Washington, D. C., January, 1080. (16) 0. S. Mills and G. Robinson, Pvoc. Chera. SOL.,1513(1959). (17) 0. S. Nills and G. Robinson, i b i d , 421 (1960) (18) A. A. Hock and 0. S. Mills, ibid., 233 (1958). (19) 0. S. hlills, private communication. (20) R . P. Dodge and V. Schomaker, private communication. (21) W. Hieber and W. Beck, Z . aizoyg. 21. allgem. C h e m . , 306, 265 (1960).

SiOz(C)

+ 2H2 = Si(in F e ) + 2H20

(2)

(1) D. A . R. K a y and J . Taylor, T r a n s . F a r a d a y S o c . , 5 6 , 1372 (1960). (2) J. D. Baird and J. Taylor, ibid.,54, 526 (1958). (3) J. P. Coughlin, E. S. Bureau of Mines Bulletin 542 (1954). (4) G. L. Humphrey, S . S. Todd, J. P. Coughlin and E. G. King, Bur. Mines Rept. In". 4888 (1952). (5) J. Chipman, J. C. Fulton. N.Cokcen aud C. R . Caskey, Acta .Met., 2 , 430 (1954). (6) D. H. Kirkwood and J. Chipman, paper submitted t o J . P h y s . Cheiii. 17) J Smiltens, J . P h y A . Chem, 64, 3G8 (1900).