COMMUNICATIONS TO THE EDITOR
Dec. 20, 1961
VARIATION
5029
TABLEI resulted in a 60% yield of silacyclopentadiene nZ5D 1.4265; 232 mp; TOTAL POLARIZATION (P2-)AND DIPOLE (11); b.p. 60-62'; log emax 3.8. MOMENT (J) WlTH TEMPERATURE'
OF
Anal. Calcd. for C1HOSi: C, 58.4; H, 7.35. Found: C, 58.27; H, 7.61. Compound I1 formed an adduct with hexachlorocyclopentadiene uThich melted a t 35-36'. Anal. Calcd. for CgH&Si: C, 30.45; H, 1.70; C1, 60.00. Found: C, 30.22; H, 1.84; 468.2 52.8 1.3Qb C1, 60.31. 490.3 53.6 1.44b Conversion of I1 to the anion I was accomplished a The experimental error is about 1 cc. in P2-,or 0.02 D. Atomic polarization is neglected throughout. * These with potassium sand in tetrahydrofuran, a solvent measurements were made by Rogers and Cannon (ref. 3). in which the product is soluble and colored. The The infrared spectra of the solid and molten dione structure of I is indicated by the n.m.r. spectral were determined from 2-15 p . No bands appeared data: (1) 1,l-dichlorosilacyclopentadienewith a from the liquid which were not shown by the solid, doublet C-H peak (relative areas 1 : l ) a t 2.487; although the relative areas of certain bands changed (2) silacyclopentadiene (11) with doublet C-H peaks a t 2.967 and Si-H peak a t 4.127 (relative slightly upon melting. The Raman spectrum of the solid compound con- areas 1: 1 : 1); (3) anion I in tetrahydrofuran with C-H peaks a t 2.70 and 2.907. and an Si-H peak sisted of seven bands between 726 and 1500 cm.-'. There were 24 infrared bands (Nujol mull) in this a t 3.957. (relative areas 1:l:O.S). There was a region, and there was a coincidence for each noteworthy absence of extraneous peaks. Treatment of I with excess bromobenzene (after Raman band. Such a high degree of coincidence removal of any remaining potassium metal by is inconsistent with I. The conclusion drawn is that 1,4-cyclohexane- amalgamation2) yielded a mixture of l-phenyldione exists predominantly in the flexible form near silacyclopentadiene and 1,l-di p h enylsilacycloroom temperature under all conditions studied in pentadiene. Identification of these two products the present work (solid, liquid and solution in was achieved unequivocally by hydrogenating inert solvents), and also in the gas phase a t higher them catalytically to 1-phenylsilacyclopentane temperatures. This is the first reported example and 1,1-diphenylsilacyclopentane, respectively. of a molecule which exists preferentially in the The latter two compounds were identical in every boat form, not because of steric constraints im- respect (e.g., infrared and retention time by vapor posed by s u b s t i t ~ e n t sbut , ~ because of the nature of phase chromatography) to authentic samples prepared by standard routes. the ring itself. The reaction of I1 with potassium proceeded Acknowledgments.-The authors are indebted to Professor R. C. Taylor, University of Michigan, smoothly in non-polar solvents such as benzene or for determining the Raman spectrum, and for octane, in which I was essentially insoluble and helpful conversation regarding it, and also wish much less reactive toward bromobenzene. It is striking that under identical conditions to thank Professor M. T. Rogers, Michigan State University, for a copy of his manuscript prior to divinylsilane (111), which is the open-chain analog of 11, is essentially unaffected b y potassium sand publication. in benzene or tetrahydrofuran. This would indi(9) R. D. Stolow and C. B. Boyce, J . A m . Chem. SOC.,8.3, 3722 cate that substantial resonance stabilization of the (1961); and references therein. anion is a prerequisite for metalation on the DEPARTMENT OF CHEMISTRY silicon h TL. ALLINGER ~ ~ ~atom.~ ~ WAYNESTATEUNIVERSITY It is of interest that the absorption spectrum 2 , MICHIGAS DETROIT LESLIEA. FREIBERG of I and that of potassium cyclopentadienide in RECEIVED SEPTEMBER 29, 1961 tetrahydrofuran are very similar, implying similar THE FIRST EXAMPLE OF AN AROMATIC SILICON orbital energies. Indeed an LCAO-MO calculation, including silicon d-orbital participation, sugSYSTEM gests considerable resonance stabilization. Sir : The foregoing work clearly implies that the silicon We wish to report the synthesis and characteriatom might be incorporated successfully into an zation of the anion of silacyclopentadiene ( I ) . This constitutes the first example of an aromatic aromatic six-membered ring (e.g., silabenzene) and/ or a seven-membered system (e.g., silatropyliuni system containing the silicon atom. cation). Experiments designed to test these pos/CHZ HzC\ sibilities are presently under way in our laboratory. The authors are grateful to the National Science CH Foundation for a research grant which made this work possible. / / / \ T,
(D) 1.26 1.28 1.32 1.31
Pam
OK.
291.16 298.16 311.14 323.76
P
60.8 61.4 61.7 59.7
H
H
H
H 111
DEPARTMEXT OF CHEMISTRY ROBERT A. BENKESER PURDUE UNIVERSITY RICHARD F. GROSSMAN INDIANA GARTHM. STANTON Reduction of 1,1-dichlorosilacyclopentadienel WESTLAFAYETTE, with lithium aluminum hydride in dimethylcarbitol, RECEIVED NOVEMBER 15, 1961
I
I1
(1) R . A. Benkeser, R. F. Grossman and G. M. Stanton, J . A m . Chsm. SOC.,8.3, 3716 (1961).
(2) H Gilman and co-workers, ibid., 74, 561 (1952).
