2778
COMMUNICATIONS TO THE EDITOR
different solvated complex in the presence of EDTA and excess H202. Indeed, when EDTA was added to a ferrate solution prepared from the dry solid K2Fe04 in 9$! KOH, the minimum usyally occurring a t 3900r2. was shifted to 400CJA., and the maximum usually occurrigg a t 5000A. was shifted to approximately 5500A. These displacements were somewhat dependent on the concentration of EDTA. In determining the FeO4-- concentration spectrophotometrically, it was assumed that the polar extinction coefficient for Fe04-- a t the 5180 A. maximum was 1070 1. mole-' cm.-', the same as determined by Wood3 and reproduced by us a t 5050A. Bttempts to prepare FeOd-- as described above, in the absence of EDTA, were unsuccessful. When solid Fe(N03)a.9H20 was added to the mixture with no EDTA present, purple flashes of color appeared on the crystals and in the immediate area of the crystals but quickly disappeared. It has been reported by other^,^^^ and observed by us, that addition of H 2 0 2to FeO4-- in strong alkaline media causes rapid reduction of Fe04-to Ft?(OH)3. Likewise, addition of EDTA to an alkaline Fe04-- solution prepared from solid K2Fe04results in an increase in the rate of thermal decomposition of this solution. \Ye cannot explain the apparent stabilizing effect of EDTA on the forination of Fe04-- with Hz02. EDTA generally is considered to be a cation complexing agent, but in this strongly alkaline solution, Fe(V1) as FeOl-and Fe(II1) as Fe02- or Fe03--- are anionic species. Moreover, as observed by Duke and Haas,' addition of EDTA to a Fe(OH)3 precipitate in alkaline medium appears to have no effect on the precipitate. Thanks are due to H. A. Mahlman and J. W. Royle for suggesting this phase of the work, and to IIT.F. I'aughan who assisted in the initial preparation of Fe04-- by this method. This work was supported by the Oak Ridge Institute of h'uclear Studies in cooperation with Michigan State University and the Oak Ridge National Laboratory.
Vol. 83 TABLE I
Compound
(CHaCH2)zC"=O CH3CH~C"OzH CHaCH2ClaO*CH3 CH3CHzC"DzOH (CH3CH2)tC13DOH (CH3CH2)3C'30H CHaCH2Ci3(0H)(CHa)z CH3CHzCl3(C1)(CHa)2 "CH3 I
JH-C-O-H(C.P.S.)
JCla-C-C-E(c.p.S.)
4.7 f0.1 5.5 f . 2 5 . 3 f .1 6.4 f .2 4.0 i .2 4 . 5 * .1 4 . 4 f .2 5 7 f .1
7.5 f 0 . 1 7.4 i .1 7.4 f . 1 7.3 f . 2 6.8 & .3 7.5 f . 2 7.3 f .2 7.5 f .2
*
Ci3Hs-C(OH)@CH~CHd (a) 3 . 2 .2 ( 6 ) 3 . 0 =I= . 2 aCH3
I
C i 8 ~ 8 - ~ ( ~ i ) Q ~ 1)(. ~ 4~ . a2 (p) 3 . 2 . f
.2 .2
tion that the Fermi contact term is the dominant one predict such a correlation, has led to the conclusion that the contact term is indeed the important term in both JCl3-H and J C ~ L C - HWe . wish to point out that such does not seem to be the case in JCI~LC-C-H. Spin-spin coupling constants (JCILC-C-H) for ten compounds are reported in Table I. The enrichment in the samples varied from 30% to 60% excess C13. All spectra were taken with a model V4300D Varian Associates Spectrometer, a t 60 Mc. ; spin-spin coupling constants were measured by the standard side band technique.4 The following observations and conclusions are varies irregularly and there is made : (1) JC~P-C-C-H no obvious correlation between its magnitude and the hybridization of C13 atomic orbitals. (2) No correlation is possible involving the electronegativity of the group attached to C13. (3) The constancy of the corresponding JH-C-C-H excludes the possibility that the variations are due to different populations of rotational isomers; there is no reason to suggest that, in the same compound, CI3 sees the methyl protons in different conformations than the methylene hydrogens do. (4) 0. N. Hinsvark, Master's Thesis, Michigan S t a t e University, East Lansing, Michigan, 1952. Calculations analogous to those reported previously,2 and based on the assumption that the ELECTRONUCLEAR RES. DIV. OAKRIDGET \ T LABORATORY ~ ~ ~ ~ ~ ~ ~ Fermi contact term is the dominant term, lead to the OAK RIDGE,TENNESSEE G. L. KOCHANNY, JR. following results: (1) for sp3 hybridized el3, CHEMISTRY DEPARTMENT Jc~~-c-c-H = JH-C-C-H X 0.3036. (2) For sp2 MICHIGAT STATEUNIVERSITY EASTLANSITG, MICHIGAN A. TIMNICKhybridized el3,JCia-C-C-H = JH-C-C-HX 0.4046. Using the experimental values of JH-C-C-Hgiven in RECEIVED APRIL 27, 1961 Table I, one obtains for sp3hybridized C13JC~LC-C-H = -2.3 C.P.S. and for sp2hybridized C13 JCN-C-H SPIN-SPIN COUPLING CONSTANTS BETWEEN = -3.0 C.P.S. The random variation of the experiNON-BONDED C'3 AND PROTON. III.',' ANOMALIES mental JcLC-C-H values suggests that the Fermi IN J c ~ I - c - c - H contact term is not t h ~only important term. Sir: I t has been shown that a direct correlation Subtler factors, resulting to higher than predicted values for JC~~LC-C-H, must be operative. We hope exists between the magnitude of either JCIP-H to study the problem further by examining rigid or JCLC-Hand the degree of sp hybridization of the C13 atomic orbital^.^^^ This finding, coupled systems where the dihedral angle is fixed, and with the fact that calculations based on the assump- systematic variations of the hybridization of and the groups attached to it can be performed. (1) Acknowledgment is made t o t h e donors of t h e Petroleum ReWe wish to comment briefly on the fact that for search Fund, administered by t h e American Chemical Society. the same compound Jc~z-c-C-H is occasionally ( 2 ) Previous paper, G. J. Karabatsos, J. D. Graham a n d F. Vane, greater than JC1a-C-H. It was shown2that JC~LC-H J Phys. Chem., in press. (3) (a) J. N . Shoolery, J . Chcm. P k y s . , 81, 1427 (1858): (b) N. Muller and D. E Pritchard, ibid., 31. 768, 1471 (196Q).
(4) ,T
T. Arnold and M. E. Packntd, { b i d , 19, 1608 (1951),
June20, 1961
COMMUNICATIONS TO THE EDITOR
2779
for sp3 hybridized C13 has values of about 4.0 rnents of the mass peak heights for Hz, H D and Dr C.P.S. The data presented here show that for spy in the completely decomposed sample. Assuming hybridized C13, JC1a-C-C-H is either equal or con- the hydrogen originated only from SbH3and SbD3, siderably greater than the corresponding Jc1a-c-H our preliminary results gave a ratio of 15 : 1 for the in the same compound. Such anomalies-spinhydride to the deuteride ratio. spin coupling constants do not decrease monoMost of the readily conceivable mechanisms for tonically with the number of bonds separating the the reactions producing stibine would be expected interacting nuclei-have been observed in various to yield SbH3, SbHnD, SbHDz and SbD3 in ratios metal alkyl^.^ The general impression has been which would be predicted statistically from the that JM-C-H is abnormally small. We wish to over-all H to D ratio in the stibine. For our suggest that perhaps, as in the case of C13,JM-C-C-H experiment, a statistical distribution would correis abnormally large, and more intensive studies on spond to the most abundant deuterated product J ' s between interacting nuclei separated by three being SbHZD. Hence, it was surprising to find bonds may explain decreases which are not mono- SbD3 as the only deuterated product. The D in tonic with the number of separating bonds. the stibine did not originate from borohydride since (5) (a) P. T. Narasimhan and M. T. Rogers, J. A m . Chem. Soc., the hydrogens of borohydride ion are known not to 82, 34 (1960); (b) R. E. Dessy. T. J. Flautt, H. H. JaffC and G. F. exchange4 with the hydrogen of an aqueous solvent. Reynolds, J . Chem. P h y s , 30, 1422 (1959); ( c ) E. B. Baker, ibid., One possibly is the existence of two independent 26, 960 (1957). paths, one resulting in SbH3 and the other SbDa. KEDZIECHEMICAL LABORATORY OF CHEMISTRY GERASIMOS DEPARTMENT J. KARABATSOSThe path leading to hydride formation would The second MICHIGAN STATEUNIVERSITY JOHN D. GRAHAM involve reactions only with BH,-. EASTLANSIXG, MICHIGAN FLOIEVANE path leading to the deuteride would involve reacRECEIVED APRIL21, 1961 tions of deuteriodiborane with Sb(II1). I n concentrated sulfuric acid some diborane5 is formed ANOMALOUS HYDROGEN-DEUTERIUM from borohydride ion; and diborane has exchangeDISTRIBUTION IN STIBINE PREPARED FROM able6 hydrogens. Such a two-path mechanism can ANTIMONY(II1) AND BOROHYDRIDE IN HEAVY be used to explain the appearance of SbDa (without WATER other deuterio species) in addition to the expected Sir: SbH3. We wish to report an unexpected deuteriumWork is in progress to elucidate the mechanism hydrogen distribution in the preparation of stibine. of this deuteration and to study the analogous This preparation yielded SbH3 to SbD3 in a ratio of with the corresponding ils, Sn and Ge about 15 : 1 with no partially deuterated stibine reactions compounds. detectable. The synthesis' consisted of slowly We wish to acknowledge assistance of Mr. N. dropping a heavy water solution containing 2.5F Neunke and thank Dr. H. Papazian for measuring NaOD, 0.5F KSbO(C4H406), and 0.4F KBHl into a the infrared spectrum. heavy water solution 2 F in D2S04 while vigorously (4) P. R. Girardot and R. W. Parry, J. A m . Chcm. Soc.. 7S, 2365 bubbling nitrogen through the reaction mixture a t (1951). 11 cm. total pressure. The total exchangeable (5) H.G.Weiss and I. Shapiro, ibid., 81, 6167 (1959). (6) S.Kaye and T. Freund, unpublished results. hydrogen was 97% deuterium. The gaseous stibine RESEARCH LABORATORY was trapped and purified as recommended by COWAIRSCIENTIFIC THOMAS FREUXD 5001 KEARNEY VILLAROAD Jolly. The infrared spectrum of our preparation showed SANDIEGO,CALIFORNIA RECEIVED APRIL22, 1961 peaks in the rock salt region a t 1945, 1882, 1825, 1406, 1362, 1320, 832 and 775 cm.-'. These eight THE LEAVING GROUP AS A FACTOR IN THE peaks, the only peaks in the spectrum of our sample, ALKYLATION OF AMBIDENT ANIONS are in agreement2 with the four fundamental vibra- Sir: tions of SbH3 and the two fundamental vibrations The reaction of nitroparaffin salts with alkyl of SbD3 which were in the range of the prism used. Although the intensity3 measurements could not be halides may occur as carbon-alkylation or as used to determine the relative amounts of SbH3 oxygen-alkylation. The latter is productive of to SbD3 it should be mentioned that the relative nitronic esters (11) which are not isolated; instead absorbancy of v1 for SbH3to SbD3 was about 10: 1. the carbonyl compound and oxime are obtained and R' Xass spectrometric analysis of the stibine showed a broad peak in the 120-130 region, no oxygen, a R--[-NO* tf R-C=N little water with HzO greatest and DzO least, and small amounts of Hz, H D and Dz with an H to D R"CH& R"CH2X ratio of about 15. The sample was decomposed in the gas phase by R' R' I I .oan electrical discharge, using a n ordinary Tesla R-C-NO~ R-C=h(iCH,R , coil leak detector external to the glass system. I The ratio of H to D was estimated from measureCH2-R" (1) W. L. Jolly, J. A m . Chcm. Soc., 88, 335 (1961); s.R. Gunn, I I1 W. L. Jolly and L. G. Green,J . Phys. Chcm., 84, 1334 (1960). it is generally assumed that they arise from the (2) W. H. Haynie and H. H. Nielsen, J . Chcm. Phys., 11, 1839 nitronic ester.I (1953); D. C. Smith, ibid., 19, 384 (1951); American Petroleum
[
'
+
