Silicon-carbon unsaturated compounds. 57. Photolysis of meso- and racemic-1,2-diethyl-1,2-dimethyldiphenyldisilane, direct evidence for a concerted 1,3-silyl shift to ortho-carbon in the phenyl ring

Article abstract of DOI:10.1021/ja960458u

meso- and racemic-1,2-diethyl-1,2-dimethyldiphenyldisilane (2a) and (2b) were synthesized respectively by hydrogenation of meso- and racemic-1,2-diethynyl-1,2-dimethyldiphenyldisilane in the presence of a IrCl(CO)(PPh₃)₂ catalyst. Ir

Full text of DOI:10.1021/ja960458u

J. Am. Chem. Soc. 1996, 118, 6853-6859
6853
Silicon-Carbon Unsaturated Compounds. 57. Photolysis of
meso- and racemic-1,2-Diethyl-1,2-dimethyldiphenyldisilane,
Direct Evidence for a Concerted 1,3-Silyl Shift to ortho-Carbon
in the Phenyl Ring
Joji Ohshita,*,† Hiroyuki Niwa,† Mitsuo Ishikawa,*,† Tokio Yamabe,*,‡
Takao Yoshii,‡ and Kouichi Nakamura‡
Contribution from the Department of Applied Chemistry, Faculty of Engineering,
Hiroshima UniVersity, Higashi-Hiroshima 724, Japan, and Department of Molecular Engineering,
Faculty of Engineering, Kyoto UniVersity, Kyoto 606-01, Japan
ReceiVed February 12, 1996X
Abstract: meso- and racemic-1,2-diethyl-1,2-dimethyldiphenyldisilane (2a) and (2b) were synthesized respectively
by hydrogenation of meso- and racemic-1,2-diethynyl-1,2-dimethyldiphenyldisilane in the presence of a IrCl(CO)-
(PPh3)2 catalyst. Irradiation of 2a with a low-pressure mercury lamp in the presence of isobutene in hexane proceeded
with high diastereospecificity to give (R,S)- and (S,R)-2-(isobutylethylmethylsilyl)-1-(ethylmethylphenylsilyl)benzene
in 77% yield. Similar irradiation of 2b with isobutene produced (R,R)- and (S,S)-isomer. The photolysis of 2a and
2b in the presence of 1,1-diphenylethylene also proceeded diastereospecifically to give the respective adducts. With
2,3-dimethylbutadiene, 2a and 2b produced the corresponding adducts, whose spectrometric analysis showed the
presence of two diastereomers. Results on theoretical studies which were carried out using PhSiH2SiH3 as a model
have also been reported.
Introduction
molytic scission of the silicon-silicon bond of aryldisilanes,
followed by addition of the silyl radical to the ipso position of
the phenylsilyl group, and then a 1,2-silyl shift to give the
rearranged silenes was reported by Sakurai et al.5 However,
In 1975, we found that when phenyl- and p-tolylpentameth-
yldisilane are photolyzed at 254 nm in benzene, photochemical
isomerization takes place to give a silene arising from a 1,3-
trimethylsilyl shift to an ortho carbon atom in the aryl ring.1a
In the photolysis of pentamethyl-p-tolyldisilane, two other
reactions are observed to occur slightly.1b,c One involves
homolytic scission of a silicon-silicon bond, followed by
disproportionation of the resulting silyl radicals leading to
1-methyl-1-p-tolylsilene, while the other comprises extrusion
of dimethylsilylene from the starting disilane. Introduction of
sterically bulky substituents and also radical stabilizing sub-
stituents such as phenyl groups on the silicon atoms2 and
substitution to the ortho position in the aryl ring3 seem to
facilitate the formation of silyl radicals.
The formation of the rearranged silenes is remarkably general
for the photolysis of the benzenoid aryldisilanes that have less
bulky substituents, such as a methyl or ethyl group on the silicon
atoms.1c The silenes thus formed react with various trapping
agents to give addition products.1-4 Although chemical be-
havior of the rearranged silenes has been extensively investi-
gated, the mechanism for the formation of the silenes has not
been fully elucidated. In 1980, a mechanism involving ho-
we concluded that a concerted pathway for the formation of
the rearranged silenes is more favorable than the radical
mechanism on the basis of the following results. Firstly, the
photolysis of 2,2,2-trimethyl(phenyl)bis(trimethylsiloxy)disilane
that proceeds with homolytic scission of the silicon-silicon
bond to give silyl radicals affords no rearranged silene.6
Secondly, picosecond flash photolysis experiments on penta-
methylphenyldisilane indicated that the rearranged silene is
formed with a buildup time of 30 ps after excitation, which is
equal to the decay time of the fluorescence of the disilane.7
These results obviously support the concerted mechanism
derived from a singlet excited state.
Recently, Leigh and Sluggett have investigated the nanosec-
ond flash photolysis of 1,1,1-trimethyltriphenyldisilane and
found that the corresponding rearranged silene was produced
within a time scale shorter than the duration of the excitation
pulse (ca. 10 ns).2b They also demonstrated that the rearranged
silene from the photolysis of 1,1,1-trimethyltriphenyldisilane is
† Hiroshima University.
‡ Kyoto University.
X Abstract published in AdVance ACS Abstracts, July 1, 1996.
(1) (a) Ishikawa, M.; Fuchikami, T.; Sugaya, T.; Kumada, M. J. Am.
Chem. Soc. 1975, 97, 5923. (b) Ishikawa, M.; Fuchikami, T.; Kumada, M.
J. Organomet. Chem. 1976, 118, 155. (c) Ishikawa, M.; Kumada, M. AdV.
Orgenomet. Chem. 1981, 19, 51.
(2) (a) Sluggett, G. W.; Leigh, W. J. Organometallics 1992, 11, 3736.
(b) Sluggett, G. W.; Leigh, W. J. J. Am. Chem. Soc. 1992, 114, 1195. (c)
Leigh, W. J.; Sluggett, G. W. Organometallics 1994, 13, 269. (d) Sluggett,
G. W.; Leigh, W. J. Organometallics 1994, 13, 1005.
(3) Braddock-Wilking, J.; Chiang, M. Y.; Gaspar, P. P. Organometallics
1993, 12, 197.
(5) Sakurai, H.; Nakadaira, Y.; Kira, M.; Sugiyama, H.; Yoshida, K.;
Takiguchi, T. J. Organomet. Chem. 1980, 184, C36.
(6) Nate, K.; Ishikawa, M.; Imamura, N.; Murakami, Y. J. Polym. Sci.
Part A: Polym. Chem. 1986, 24, 1551.
(4) Ishikawa, M.; Kikuchi, M.; Kunai, A.; Takeuchi, T.; Tsukihara, T.;
Kido, M. Organometallics 1993, 12, 3474 and references therein.
(7) Shizuka, H.; Okazaki, K.; Tanaka, M.; Ishikawa, M.; Sumitani, M.;
Yoshihira, K. Chem. Phys. Lett. 1985, 113, 89.
S0002-7863(96)00458-1 CCC: $12.00 © 1996 American Chemical Society

6854 J. Am. Chem. Soc., Vol. 118, No. 29, 1996
Ohshita et al.
Scheme 1
Scheme 2
Photochemical Reactions. First, we carried out the pho-
tolysis of a mixture of 2a and 2b in a ratio of 49:51. Thus,
irradiation of the mixture in the presence of a 20-fold excess of
isobutene in hexane at -78 °C with a low-pressure mercury
lamp for 5 h gave 2-(isobutylethylmethylsilyl)-1-(ethylmeth-
ylphenylsilyl)benzene (3a and 3b) in 85% yield, in addition to
an 11% yield of ethylmethylphenylsilane and 3% of the
unchanged starting compound 2 (Scheme 2). Ethylmethyl-
phenylsilane is probably produced by homolysis of a silicon-
silicon bond, followed by hydrogen abstraction by the resulting
silyl radicals in the solvent cage or concerted dehydrosilation
from a singlet excited state of 2 as proposed by Leigh et al.2d
Although GPC and GLC analysis of the adduct show a
homogeneous peak, its 1H NMR spectrum reveals the presence
of two diastereomeric isomers in a ratio of 49:51. Interestingly,
the photolysis of meso-isomer 2a (>99% de) in the presence
of isobutene under the same conditions produced a single isomer
3a in 77% yield, together with a 12% yield of ethylmethyl-
phenylsilane and 5% of the starting compound 2a.
derived largely from an excited singlet state of the disilane and
free silyl radicals do not play a crucial role in the formation of
the silene.2d,8 Thus, the presence of a triplet quencher in the
photolysis of 1,1,1-trimethyltriphenyldisilane does not lead to
the decrease in the yields of the products originating from the
rearranged silene but suppresses the silyl radical formation.
Furthermore, the photolysis of the aryldisilane affords the
products derived from the rearranged silene in almost the same
yields in the presence of the radical trapping reagent as in the
absence of the radical trapping agent.2c These results also
suggest the concerted mechanism for the formation of the
rearranged silenes.
In order to get more direct evidence for the concerted
mechanism, we investigated the stereochemistry of the pho-
tolysis of aryldisilanes. In this paper, we report the diaste-
reospecific formation of the rearranged silenes in the photolysis
of meso- and racemic-1,2-diethyl-1,2-dimethyldiphenyldisilane.
We also report the results of MO calculations for PhSiH2SiH3
and the rearranged silene as model compounds.
The structure of 3a was confirmed by spectrometric analysis
and combustion elemental analysis. The mass spectrum shows
a parent ion at m/z 354, corresponding to the molecular weight
1
of the isobutene adduct. The H NMR spectrum of 3a shows
two singlets at 0.10 and 0.60 ppm ascribed to two nonequivalent
methylsilyl groups, along with those of ethylsilyl, isobutyl, and
aromatic ring protons. Its 13C NMR spectrum reveals two
signals at -1.8 and -1.7 and seven signals at 7.6, 7.7, 8.38,
8.43, 24.8, 26.1, and 26.3 (two carbons), attributed to methylsilyl
carbons and ethyl and isobutyl carbons, as well as ten signals
due to phenyl and phenylene ring carbons. As expected, the
29Si NMR spectrum shows two signals at -4.48 and -0.15 ppm,
due to two nonequivalent silicon atoms. These spectra reveal
no signals due to the other diastereomer (3b), indicating the
diasteremeric purity for 3a is more than 99% de.
Results and Discussion
Synthesis. A mixture of meso- and racemic-1,2-diethynyl-
1,2-dimethyldiphenyldisilane was obtained by the reaction of
1,2-dichloro-1,2-dimethyldiphenyldisilane with ethynylmagne-
sium bromide in 94% yield, as shown in Scheme 1. NMR
spectrometric analysis of the resulting product reveals that the
product consists of a mixture of diastereomeric isomers in an
approximate ratio of 49:51 (see Experimental Section). Frac-
tional recrystallization of the mixture of meso- and racemic-
isomer from ethanol afforded one isomer as crystals whose
diastereomeric purity was determined to be more than 99%
Similar irradiation of racemic-isomer 2b (80% de) with
isobutene in hexane at -78 °C gave adduct 3b with 78% de
and ethylmethylphenylsilane in 73% and 10% yields, along with
diastereomeric excess (de) by H, 13C, and 29Si NMR spectro-
1
1
3% of the unchanged starting compound 2b. For 3b, the H
metric analysis. The structure of this isomer was characterized
by X-ray crystallographic analysis as meso-1,2-diethynyl-1,2-
dimethyldiphenyldisilane (1a). Removal of the crystals of 1a
from the mixture gave racemic-isomer 1b with 60% de.
Treatment of the mixture with preparative GLC afforded
1b with 78-90% de (diastereomeric ratios of 1a/1b )
11/89-5/95).
Compounds 1a (>99% de) and 1b (78-90% de) thus
obtained were converted into the respective 1,2-diethyl-1,2-
dimethyldiphenyldisilanes (2a and 2b) by hydrogenation in the
presence of Vaska’s catalyst. Compounds 2a and 2b were
obtained as viscous liquid, and therefore, X-ray crystallographic
analysis of these compounds could not be performed. However,
NMR spectrometric analysis of 2a and 2b indicates that the
hydrogenation proceeded with high diastereospecificity to give
2a and 2b having the same diastereomeric purity as that of the
respective starting compounds 1a and 1b (see Experimental
Section). Therefore, we identified 2a as the meso-isomer and
2b as the racemic-isomer.
NMR spectrum shows two singlets at 0.08 and 0.60 ppm
ascribed to two nonequivalent methylsilyl groups, along with
those of ethylsilyl, iso-butyl, and aromatic ring protons. The
13C NMR spectrum reveals methylsilyl carbon at -1.9 and -1.8
ppm and ethyl and isobutyl carbons at 7.6, 7.7, 8.38, 8.40, 24.8,
26.0, 26.2, and 26.3 ppm, along with phenyl and phenylene ring
carbons. Its 29Si NMR spectrum displays two signals at -4.48
and -0.18 ppm. In these spectra, signals due to 3a with low
intensities are also observed, with some overlapping with those
of 3b. The diastereomeric purity for 3b was calculated by the
integration ratio of the signals due to methylsilyl protons at 0.08
ppm for 3b and 0.10 ppm for 3a.
For the formation of 3a and 3b, if the concerted mechanism
is operative, a suprafacial 1,3-silyl shift in (R,R)-1,2-diethyl-
1,2-dimethyldiphenyldisilane should produce silenes A [(E)-
(S,S)] and B [(Z)-(R,S)], according to paths a and b shown in
Scheme 3. The silenes A and B thus formed undergo the ene
reaction with isobutene to give an adduct. In this addition
reaction, isobutene approaches the silicon-carbon double bond
of the silenes from the opposite side of the ethylmethylphenyl-
(8) Leigh, W. J.; Sluggett, G. W. J. Am. Chem. Soc. 1993, 115, 7531.

Silicon-Carbon Unsaturated Compounds
J. Am. Chem. Soc., Vol. 118, No. 29, 1996 6855
Scheme 3
Scheme 5
was determined by an X-ray diffraction study. Recrystallization
of 4b from ethanol afforded colorless crystals, whose X-ray
crystallographic analysis reveals that 4b consists of (R,R)- and
(S,S)-isomers, in agreement with the mechanism proposed in
Scheme 3.
Next, we carried out the photolysis of 2a and 2b in the
presence of 2,3-dimethylbutadiene to obtain further evidence
for a concerted 1,3-silyl shift in the photochemical formation
of the rearranged silenes. Previously, we have reported that
the silenes generated photochemically from aryldisilanes react
with 2,3-dimethylbutadiene to give o-silyl-substituted (2,3-
dimethyl-3-butenylsilyl)arene as the sole product.1b The reaction
of the silenes arising from 2a and 2b with 2,3-dimethylbutadiene
should afford an adduct, 2-[ethylmethyl(2,3-dimethyl-3-butenyl)-
silyl]-1-(ethylmethylphenylsilyl)benzene which has three chiral
centers in the molecule (Scheme 5). If the photochemical
formation of the silenes and the reaction of the resulting silenes
with 2,3-dimethylbutadiene proceed with nonstereospecificity,
four diastereomers should be detected by NMR spectrometric
analysis. Thus, GPC and GLC analysis of the products obtained
from the photolysis of 2a (99% de) in the presence of
2,3-dimethylbutadiene showed a single peak, and the yield of
the adduct was calculated to be 73%. 1H, 13C, and 29Si NMR
spectra of the adduct indicate the presence of two diastereomeric
isomers, 5a and 5a′, in a ratio of 53:47, but not four diastere-
omers. Similar photolysis of 2b (90% de) in the presence of
2,3-dimethylbutadiene afforded a diasrereomeric mixture of
adducts 5a, 5a′, 5b, and 5b′ in a ratio of 4:2:57:37 in 77% yield.
Again, the results obtained can be understood in terms of a
concerted 1,3-silyl shift of the aryldisilane to the ortho carbon
in the aryl ring.
X-ray Crystallographic Analysis of 1a and 4b. The
structures of 1a and 4b were verified by X-ray diffraction
studies. Cell dimensions, data collections and refinement
parameters, and selected bond lengths and angles for 1a and
4b are collected in Tables 1, 2, and 3. All unique diffraction
maxima with 0 < 2θ < 126.0° were recorded on a Rigaku AFC-
6C automated four-circle diffractometer using graphite-mono-
chromated Cu KR radiation (λ ) 1.5418 Å). A total of 1351
observed refractions for 1a and 3930 for 4b (I > 3σ(I)) were
used in the least-squares refinement. The structures were solved
by the Monte-Carlo direct method10 using the MULTAN 78
program system11 and refined by the full-matrix least-squares
program.12 Atomic scattering factors were taken from the
International Tables for X-ray Crystallography.13 Anisotropic
temperature factors were used for refinement of non-hydrogen
atoms. Finally, R and Rw of 0.0743 and 0.0919 for 1b and
0.0508 and 0.0525 for 4b were obtained. ORTEP drawings of
1a and 4b are shown in Figures 1 and 2, along with atomic
Scheme 4
silyl group on the cyclohexadienyl ring to avoid a significant
steric interaction with this group.9 Consequently, (R,R)-2-
(isobutylethylmethylsilyl)-1-(ethylmethylphenylsilyl)benzene
would be produced. On the basis of similar consideration, the
(S,S)-isomer of 2b should afford (E)-(R,R)-silene and (Z)-(S,R)-
silene which react with isobutene to give (S,S)-2-(isobutyleth-
ylmethylsilyl)-1-(ethylmethylphenylsilyl)benzene. Similarly,
(R,S)-isomer 2a produces four types of silene by the concerted
1,3-silyl shift. The 1,3-shift of the silyl group with the
R-configuration to the ortho phenyl carbon should produce (E)-
(R,S)-silene and (Z)-(S,S)-silene, while migration of the silyl
group with the S-configuration is expected to afford (E)-(S,R)-
silene and (Z)-(R,R)-silene. The ene reaction of those four
silenes with isobutene would produce (S,R)- and (R,S)-adducts.
The diastereospecific reactions observed in the present pho-
tolysis are wholly consistent with the concerted mechanism, but
not the radical mechanism.
We also carried out the photolysis of 2a and 2b in the
presence of 1,1-diphenylethylene and 2,3-dimethylbutadiene.
Thus, irradiation of 2a (>99% de) in the presence of a 2.9-fold
excess of 1,1-diphenylethylene in 70 mL of hexane at -78 °C
for 18 h afforded 2-[ethylmethyl(2,2-diphenylethyl)silyl]-1-
(ethylmethylphenylsilyl)benzene (4a) as a pure diastereomer,
in 81% yield (Scheme 4). A 9% yield of ethylmethylphenyl-
silane was also produced in this reaction. No other isomer of
4a was detected in the reaction mixture by spectrometric
analysis. Similar photolysis of 2b (80% de) in the presence of
1,1-diphenylethylene gave adduct 4b, whose diastereomeric
excess was determined to be 78% by the NMR spectrometric
analysis in 85% yield, in addition to a 9% yield of ethylmeth-
ylphenylsilane. These results again indicate that the photo-
chemical reaction of 2a and 2b leading to the rearranged silenes
proceeds with high diastereospecificity. The structure of 4b
(10) Furusaki, A. Acta Crystallogr., Sect. A 1979, 35, 220.
(11) Main, P.; Hull, S. E.; Lessinger, L.; Germain, G.; Declarcq, J.-P.;
Woolfson, M. MULTAN78, A System of Computer Programs for the
Automatic Solution of Crystal Structures from X-Ray Diffraction Data,
University of York, England, and Louvain, Belgium, 1978.
(12) Katayama, C.; Sakabe, N.; Sakabe, K. Acta Crystallogr., Sect. A
1972, 28, S207.
(9) Ishikawa, M.; Fuchikami, T.; Kumada, M. J. Organomet. Chem. 1978,
162, 223. Takaki, K.; Sakamoto, H.; Nishimura, Y.; Sugihara, Y.; Ishikawa,
M. Organometallics 1991, 10, 888.
(13) International Tables for X-ray Crystallography; Kynoch Press:
Birmingham, 1974; Vol. 4.

6856 J. Am. Chem. Soc., Vol. 118, No. 29, 1996
Ohshita et al.
Table 1. Crystal Data, Experimental Conditions, and Summary of
Structural Refinement for Compounds 1a and 4b
Table 3. Bond Distances (Å) and Angles (deg) for Compound 4b
with Their Esd’s in Parantheses
1a
C18H18Si2
290.51
monoclinic
P21/n
4b
C32H38Si2
478.82
triclinic
P1h
Si1-C3
1.868(3) Si1-C4
1.878(4) Si1-C6
1.879(2)
Si1-C20 1.902(5) Si2-C25 1.900(4) Si2-C26 1.874(4)
chem formula
fw
cryst syst
space group
cell dimens
a (Å)
b (Å)
c (Å)
R (deg)
â (deg)
γ (deg)
V (Å3)
Si2-C27 1.875(8) Si2-C29 1.879(2) C4-C5
1.52(1)
1.524(7)
1.408(4)
C6-C7
C8-C9
1.529(3) C7-C8
1.371(4) C8-C13
1.530(3) C7-C14
1.393(4) C9-C10
C10-C11 1.357(6) C11-C12 1.349(6) C12-C13 1.378(4)
C14-C15 1.38(1)
C16-C17 1.362(7) C17-C18 1.35(1)
C20-C21 1.410(5) C20-C25 1.428(8) C21-C22 1.374(8)
C22-C23 1.36(1) C23-C24 1.387(6) C24-C25 1.396(6)
C14-C19 1.377(4) C15-C16 1.39(1)
12.164(2)
10.349(1)
6.885(1)
90.00
83.49(1)
90.00
861.1(2)
2
1.121
16.773(5)
17.260(5)
9.605(2)
95.54(3)
92.77(3)
147.92(1)
1424.6(9)
2
C18-C19 1.39(1)
C27-C28 1.535(5) C29-C30 1.390(4) C29-C34 1.390(5)
C30-C31 1.379(3) C31-C32 1.375(6) C32-C33 1.357(6)
C33-C34 1.373(3)
Z
D
C3-Si1-C4
106.7(3)
112.0(2)
109.1(2)
115.4(2)
109.8(2)
109.4(1)
113.1(5)
115.1(2)
109.3(3)
117.8(3)
119.7(3)
120.0(3)
121.1(3)
121.1(5)
120.6(4)
120.4(8)
120.5(6)
128.7(2)
123.2(6)
118.8(5)
129.4(3)
117.8(4)
122.5(2)
116.1(2)
120.0(3)
120.5(4)
C3-Si1-C6
C4-Si1-C6
108.0(2)
109.2(2)
111.7(2)
110.1(2)
105.8(3)
105.8(2)
118.1(2)
111.9(4)
124.1(3)
118.1(3)
120.7(3)
120.4(4)
120.6(3)
118.2(5)
119.8(9)
120.5(5)
114.4(5)
116.9(4)
119.9(4)
123.3(7)
112.8(4)
117.6(5)
121.3(2)
121.9(4)
119.4(3)
122.1(3)
calcd (g/cm3)
1.117
C3-Si1-C20
C4-Si1-C20
C25-Si2-C26
C25-Si2-C29
C26-Si2-C29
Si1-C4-C5
cryst dimens (mm3)
crystal color
0.60 × 0.58 × 0.33 0.53 × 0.30 × 0.09
C6-Si1-C20
C25-Si2-C27
C26-Si2-C27
C27-Si2-C29
Si1-C6-C7
colorless
1.64
colorless
1.14
Rigaku AFC-6C
293
µ (cm-1
)
diffractmeter
T (K)
Rigaku AFC-6C
293
1.5418
graphite crystal
ω-2θ
4
λ (Cu KR) (Å)
monochrometer
scan type
scan speed (deg/min)
scan width
diffraction geometry
range of h, k, l
h
1.5418
C6-C7-C8
C6-C7-C14
C7-C8-C9
graphite crystal
ω-2θ
4
0 < 2θ < 126
symmetrical A
C8-C7-C14
C7-C8-C13
C8-C9-C10
C10-C11-C12
C8-C13-C12
C7-C14-C19
C14-C15-C16
C16-C17-C18
C14-C19-C18
Si1-C20-C25
C20-C21-C22
C22-C23-C24
Si2-C25-C20
C20-C25-C24
Si2-C29-C30
C30-C29-C34
C30-C31-C32
C32-C33-C34
C9-C8-C13
C9-C10-C11
C11-C12-C13
C7-C14-C15
C15-C14-C19
C15-C16-C17
C17-C18-C19
Si1-C20-C21
C21-C20-C25
C21-C22-C23
C23-C24-C25
Si2-C25-C24
Si2-C27-C28
Si2-C29-C34
C29-C30-C31
C31-C32-C33
C29-C34-C33
0 < 2θ < 126
symmetrical A
-14 < h < 14
0 e k e 12
0 e l e 8
1609
-19 < h < 19
-20 e k e 20
0 e l e 11
5066
4599
3930
k
l
no. of measd rflns
no. of unique rflns obsd
1563
no. of obsd rflns (I > 3σ(I)) 1351
R
0.0743
0.0919
0.0508
0.0525
Rwa
a Weighting scheme is (σ(Fo)2 + 0.0004|Fo| )-1
.
2
Table 2. Bond Distances (Å) and Angles (deg) for Compound 1a
with Their Esd’s in Parantheses
Si1-Si1′ 2.344(2) Si1-C2
Si1-C5 1.869(3) C3-C4
1.864(5) Si1-C3
1.175(5) C5-C6 1.387(5)
1.381(6) C7-C8 1.345(8)
1.841(3)
C5-C20 1.389(6) C6-C7
C8-C9
1.353(7) C9-C20 1.380(7)
Si1′-Si1-C2
Si1′-Si1-C5
C2-Si1-C5
Si1-C3-C4
Si1-C5-C20
C6-C7-C8
C8-C9-C20
110.7(2)
110.9(1)
109.6(2)
176.2(4)
121.5(3)
120.3(4)
120.4(5)
Si1′-Si1-C3
C2-Si1-C3
C3-Si1-C5
Si1-C5-C6
C5-C6-C7
C7-C8-C9
C5-C20-C9
107.3(1)
110.4(2)
108.0(2)
121.6(3)
121.2(4)
120.2(5)
120.9(4)
numbering scheme. As shown in Figure 1, 1a has a Ci center
in the middle of the silicon-silicon bond. Selected bond
distances and angles for compounds 1a and 4b shown in Tables
2 and 3 agree well with the accepted values. There are no
intermolecular contacts less than the van der Waals distances.14
Figure 1. ORTEP drawing of 1a showing the atom-numbering scheme.
Theoretical Studies. In order to learn theoretically the
stereochemistry of the suprafacial 1,3-silyl shift in 1,2-diethyl-
1,2-dimethyldiphenyldisilanes 2a and 2b, we carried out mo-
lecular orbital (MO) calculations for phenyldisilane (PhSiH2-
SiH3, C) and silene (D) arising from a 1,3-silyl shift to one of
the ortho-carbons of a phenyl group, as models of disilanes 2a
and 2b and silenes A and B, respectively. The ab initio MO
calculations were carried out with the Gaussian 92 program16
at the RHF/6-31G** level. The geometries of C and D were
optimized, as shown in Figures 3 and 4. The structure C has
Cs symmetry, and the silyl group is perpendicular to the π-plane
of phenyl ring. In structure D, the Si(2)-C(3) bond length of
1.725 Å corresponds to the double bond.
(16) Frisch, M. J.; Trucks, G. W.; Head-Gordon, M.; Gill, W. P. M.;
Wong, M. W.; Foresman, J. B.; Johnson, B. G.; Schlegel, H. B.; Robb, M.
A.; Replogle, E. S.; Gomperts, R.; Andres, J. L.; Reghavachari, K.; Binkley,
J. S.; Gonzalez, C.; Martin, R. L.; Fox, D. J.; Defrees, D. J.; Baker, J.;
Stewart, J. J. P.; Pople, J. A.; Gaussian 92; Gaussian Inc.: Pittsburgh, PA,
1992.
(14) All computations were performed on the HITAC M-680/180E
system at the Information Processing Center of Hiroshima University using
the CRYSTAN program system.15
(15) Katayama, C.; Honda, M. CRYSTAN, The Computer Center of
Nagoya University Library Program.

Silicon-Carbon Unsaturated Compounds
J. Am. Chem. Soc., Vol. 118, No. 29, 1996 6857
Figure 2. ORTEP drawing of 4b showing the atom-numbering scheme.
Figure 5. Frontier orbitals of phenyldisilane (C): (a) HOMO and (b)
LUMO. The broad arrow in part b indicates the 1,3-silyl shift with
retention of configuration at the migrating silicon.
weakened to form the new C(4)-Si(1) bond because the lobe
of Si(1) toward the π-plane and the lobe of C(4) toward the
perpendicular Si(1) are in the same phase. Hence, the photo-
chemical 1,3-silyl shift in C is expected to occur with a
preference for retention of configuration at the migrating silicon
by the rules for conservation of orbital symmetry.17 This
expectation is coincident with the experimental results. On the
other hand, the C(3)-Si(2) bond is strengthened to make the
bonding π conjugation, leading to the product D.
Figure 3. Calculated RHF/6-31G** geometries of phenyldisilane (C).
All bond lengths are in angstroms and all angles are in degrees.
Conclusions
meso- and racemic-1,2-diethyl-1,2-dimethyldiphenyldisilane
(2a and 2b) were obtained by hydrogenation of meso- and
racemic-1,2-diethynyl-1,2-dimethyldiphenyldisilane (1a and 1b),
respectively, which could be separated from the mixture by
fractional recrystallization. The photolysis of 2a and 2b in the
presence of isobutene, 1,1-diphenylethylene, and 2,3-dimethyl-
1,3-butadiene as silene trapping reagents proceeded with high
stereospecificity to give the respective adducts arising from ene
reaction of the rearranged silenes with olefins.
For the photochemical 1,3-silyl shift in aryldisilanes leading
to rearranged silenes, two types of mechanistic interpretation
have been reported so far. One is a concerted pathway recently
proposed by Leigh and Sluggett,2,8 while the other is a stepwise
mechanism via radical intermediates suggested by Sakurai et
al.5 The results described here are wholly consistent with the
concerted mechanism, but not the radical process. This is the
first direct evidence for a concerted 1,3-silyl shift in aryldisi-
lanes.
Figure 4. Calculated RHF/6-31G** geometries of silene arising from
a 1,3-silyl shift (D). All bond lengths are in angstroms and all angles
are in degrees.
Experimental Section
The reaction mechanism of a 1,3-silyl shift is understood by
using the frontier orbital theory. The highest occupied molecular
orbital (HOMO) and the lowest unoccupied molecular orbital
(LUMO) of C are displayed in Figure 5. Generally, the LUMO
plays an important role in the photochemical intramolecular
reaction.17 With respect to the LUMO of phenyldisilane C, the
Si(1)-Si(2) bond is antibonding and the C(3)-Si(2) bond is
bonding so that in the excited state the Si(1)-Si(2) bond is
General. All reactions were carried out under an atomosphere of
purified argon. Yields of the products from the photolysis of 2a and
2b were determined by GLC based on the starting compounds
photolyzed. 1H, 13C, and 29Si NMR spectra were recorded on a Brucker
(17) (a) Woodward, R. B.; Hoffmann, R. Angew. Chem., Int. Ed. Engl.
1969, 8, 781. (b) Woodward, R. B.; Hoffmann, R. The ConserVation of
Orbital Symmetry; Verlag Chemie: Weinheim, 1970.

6858 J. Am. Chem. Soc., Vol. 118, No. 29, 1996
Ohshita et al.
AMX 400 spectrometer. Mass spectra were measured on a Shimadzu
QP-1000 spectrometer.
Materials. Hexane, THF, and benzene were dried over sodium-
potassium alloy and distilled just before use.
7.55-7.57 (m, 2H); 13C NMR (δ in CDCl3) -6.18, 5.25, 7.90, 127.81,
129.16, 134.31, 136.50; 29Si NMR (δ in CDCl3) -11.40. Anal. Calcd
for C9H14Si: C, 71.92; H, 9.39. Found: C, 71.63; H, 9.42.
Photolysis of 2b in the Presence of Isobutene. A mixture of 0.263
g (0.882 mmol) of 2b (80% de), 2.27 g (40.47 mmol) of isobutene,
15.2 mg (0.0892 mmol) of dodecane as an internal standard, and 70
mL of hexane was irradiated internally with a low-presure mercury
lamp at -78 °C for 3.5 h. At this stage, 97% of the starting compound
2b was photolyzed. The resulting mixture was analyzed by GLC as
being 3b (73% yield) and ethylmethylphenylsilane (10% yield). GLC
retention time and the mass spectrum for ethylmethylphenylsilane were
identical with those of the authentic sample. Compound 3b (78% de)
was isolated by MPLC on a silica gel column eluting with hexane. For
Preparation of 1,2-Diethynyl-1,2-dimethyldiphenyldisilane (1a
and 1b). A 1:1 mixture of 1a and 1b was obtained by the reaction of
1,2-dichloro-1,2-dimethyldiphenyldisilane with ethynylmagnesium bro-
mide as reported in the literature.18 The mixture was recrystallized
from ethanol to give 1a (>99% de) as colorless crystals. Removal of
the crystals gave a 2:8 mixture of 1a and 1b. The mixture was treated
with preparative GLC on a silicone SE-30 column to give 1b with
80-90% de. For 1a: mp 88-89 °C; MS m/z 290 (M+); 1H NMR (δ
in CDCl3) 0.53 (s, 6H), 2.66 (s, 2H), 7.33-7.66 (m, 10H); 13C NMR
(δ in CDCl3) -4.3, 85.5, 97.9, 128.0, 129.5, 133.4, 134.4; 29Si NMR
1
3b: MS m/z 354 (M+); H NMR (δ in CDCl3) 0.08 (s, 3H, MeSi),
1
(δ in CDCl3) -38.68. For 1b: MS m/z 290 (M+); H NMR (δ in
0.60 (s, 3H, MeSi), 0.62 -0.69 (m, 4H, CH2CH and CH3CH2Si), 0.75
(d, 3H, Me2CH, J ) 6.6 Hz), 0.78 (d, 3H, Me2CH, J ) 6.4 Hz), 0.79
(t, 3H, CH2CH2, J ) 6.8 Hz), 0.97 (t, 3H, CH3CH2Si, J ) 7.3 Hz),
1.61 (nonet, CH2CHMe2, J ) 6.6 Hz), 7.26-7.35 (m, 5H, Ph), 7.39-
7.41 (m, 2H, phenylene), 7.66-7.70 (m, 2H, phenylene); 13C NMR (δ
in CDCl3) -1.9, -1.8, 7.6, 7.7, 8.38, 8.40, 24.8, 26.0, 26.2, 26.3, 127.3,
127.6, 127.8, 128.7, 134.5, 136.1, 136.8, 139.4, 142.5, 145.9; 29Si NMR
(δ in CDCl3) -4.48, -0.18.
CDCl3) 0.58 (s, 6H), 2.68 (s, 2H), 7.25-7.52 (m, 10H); 13C NMR (δ
in CDCl3) -4.4, 85.5, 97.7, 127.9, 129.5, 133.2, 134.4; 29Si NMR (δ
in CDCl3) -38.64.
Preparation of meso-1,2-Diethyl-1,2-dimethyldiphenyldisilane
(2a). In a 50-mL autocleave was placed 459.1 mg (1.580 mmol) of
1a, 13.9 mg (0.018 mmol) of IrCl(CO)(PPh3)2, and 30 mL of benzene.
Into the autocleave 150 atm of hydrogen gas was indroduced and the
autocleave was heated at 100 °C for 12 h. After evaporation of the
solvent, the resulting mixture was chromatographed on a silica gel
column eluting with hexane to give 308.7 mg (65% yield) of 2a: MS
Photolysis of 2a in the Presence of 1,1-Diphenylethylene.
A
mixture of 0.268 g (0.898 mmol) of 2a (>99% de), 0.470 g (2.609
mmol) of 1,1-diphenylethylene, 15.6 mg (0.0916 mmol) of dodecane
as an internal standard, and 70 mL of hexane was irradiated internally
with a low-presure mercury lamp at -78 °C for 18 h. At this stage,
90% of the starting compound 2a was photolyzed. The resulting
mixture was analyzed by GLC as being compound 4a (81% yield) and
ethylmethylphenylsilane (9% yield). GLC retention time and the mass
spectrum for ethylmethylphenylsilane were identical with those of the
authentic sample. Compound 4a (>99% de) was isolated by preparative
1
m/z 298 (M+); H NMR (δ in CDCl3) 0.33 (s, 6H), 0.83-0.96 (m,
10H), 7.27-7.32 (m, 6H), 7.35-7.39 (m, 4H); 13C NMR (δ in CDCl3)
-6.4, 5.4, 7.9, 127.7, 128.3, 134.2, 138.0; 29Si NMR (δ in CDCl3)
-18.63. Anal. Calcd for C18H26Si2: C, 72.41; H, 8.78. Found: C,
72.32; H, 8.76 (as a 49:51 mixture with 2b).
Compound 2b was prepared by hygrogenation of 1b in a similar
1
maner to that for 2a in 68% yield: For 2b: MS m/z 298 (M+); H
1
TLC on silica gel. For 4a: MS m/z 449 (M+ - Et); H NMR (δ in
NMR (δ in CDCl3) 0.33 (s, 6H), 0.83-0.96 (m, 10H), 7.27-7.32 (m,
6H), 7.35-7.39 (m, 4H); 13C NMR (δ in CDCl3) -6.3, 5.5, 7.9, 127.7,
128.3, 134.2, 138.0; 29Si NMR (δ in CDCl3) -18.63.
CDCl3) -0.17 (s, 3H), 0.38 (dq, 1H, J ) 14.8, 7.2 Hz), 0.50 (dq, 1H,
J ) 14.9, 7.1 Hz), 0.55 (s, 3H), 0.64 (t, 2H, J ) 7.3 Hz), 0.94 (t, 3H,
J ) 7.6 Hz), 1.05 (br q, 2H, J ) 7.7 Hz), 1.37 (dd, 1H, J ) 14.6, 7.7
Hz), 1.45 (dd, 1H, J ) 14.7, 7.8 Hz), 3.88 (t, 1H, J ) 7.6 Hz), 7.08-
7.19 (m, 10H), 7.21-7.36 (m, 5H), 7.36-7.39 (m, 2H), 7.51 (dd, 1H,
J ) 5.6, 1.7 Hz), 7.71 (dd, 1H, J ) 5.6, 1.7 Hz); 13C NMR (δ in CDCl3)
-2.14, -1.81, 7.42, 7.60, 7.73, 8.43, 22.77, 47.24, 125.84, 127.24,
127.55, 127.67, 127.85, 127.94, 128.18, 128.52, 128.82, 129.08, 134.54,
136.42, 136.80, 139.14, 142.57, 144.87, 146.90, 146.97; 29Si NMR (δ
in CDCl3) -4.52, 0.61. Anal. Calcd for C32H38Si2: C, 80.27; H, 8.00.
Found: C, 80.09; H, 8.04 (as a 1:1 mixture with 4b).
Photolysis of 2a in the Presence of Isobutene. In a 75-mL reaction
vessel fitted internally with a low-pressure mercury lamp bearing a
Vycor filter was placed 0.263 g (0.881 mmol) of 2a, 2.270 g (40.46
mmol) of isobutene, 18.4 mg (0.108 mmol) of dodecane as an internal
standard, and 70 mL of hexane. The vessel was cooled at -78 °C and
the mixture was irradiated for 1 h. At this stage, 95% of the starting
compound 2a was photolyzed. The resulting mixture was analyzed
by GLC as being compound 3a (77% yield) and ethylmethylphenyl-
silane (12% yield). GLC retention time and the mass spectrum for
ethylmethylphenylsilane were identical with those of the authentic
sample. Compound 3a (>99% de) was isolated by MPLC on a silica
gel column eluting with hexane. For 3a: MS m/z 354 (M+); 1H NMR
(δ in CDCl3) 0.10 (s, 3H, MeSi), 0.60 (s, 3H, MeSi), 0.61-0.70 (m,
4H, CH3CH2Si), 0.76 (t, 3H, CH3CH2Si, J ) 6.6 Hz), 0.77 (d, 3H,
Me2CH, J ) 6.6 Hz), 0.78 (d, 2H, Me2CH, J ) 6.9 Hz), 0.97 (t, 3H,
CH3CH2Si, J ) 6.9 Hz), 1.10 (br q, 2H, CH3CH2Si, J ) 6.6 Hz), 1.61
(nonet, CH2CHMe2, J ) 6.6 Hz), 7.28-7.36 (m, 5H, Ph), 7.39-7.42
(m, 2H, phenylene), 7.65-7.71 (m, 2H, phenylene); 13C NMR (δ in
CDCl3) -1.8, -1.7, 7.6, 7.7, 8.38, 8.43, 24.8, 26.1, 26.3 (2C), 127.3,
127.6, 127.8, 128.7, 134.5, 136.2, 136.8, 139.4, 142.5, 145.9; 29Si NMR
(δ in CDCl3) -4.48, -0.15. Anal. Calcd for C22H34Si2: C, 74.50; H,
9.66. Found: C, 74.45; H, 9.77 (as a 49:51 mixture with 3b).
Preparation of Ethylmethylphenylsilane. To a solution of 13.4
mmol of ethylmagnesium bromide in 15 mL of THF was added
dropwise 1.56 g (9.96 mmol) of chloromethylphenylsilane at room
temperature. The resulting mixture was stirred for 15 h at room
temperature and hydrolyzed with water. The organic layer was
separated and the aqueous layer was extracted with ether. The organic
layer and the extracts were combined and washed with water. After
evaporation of the solvent, the residue was distilled under atmospheric
pressure and a fraction boiling in a range of 153-154 °C was collected
(0.800 g, 53% yield): MS m/z 150 (M+); IR νSi-H 2114 cm-1; 1H NMR
(δ in CDCl3) 0.35 (d, 3H, J ) 3.8 Hz), 0.80-0.88 (m, 2H), 1.03 (t,
3H, J ) 8.2 Hz), 4.34 (sept, 1H, J ) 3.5 Hz), 7.36-7.41 (m, 3H),
Photolysis of 2b in the Presence of 1,1-Diphenylethylene.
A
mixture of 0.210 g (0.706 mmol) of 2b (80% de), 0.684 g (3.79 mmol)
of 1,1-diphenylethylene, 14.3 mg (0.084 mmol) of dodecane as an
internal standard, and 70 mL of hexane was irradiated internally with
a low-presure mercury lamp at -78 °C for 30 h. At this stage, 93%
of the starting compound 2b was photolyzed. The resulting mixture
was analyzed by GLC as being compound 4b (85% yield) and
ethylmethylphenylsilane (9% yield). GLC retention time and the mass
spectrum for ethylmethylphenylsilane were identical with those of the
authentic sample. Compound 4b (78% de) was isolated by prepatative
TLC on silica gel. Diastereomeric pure 4b was obtained by recrys-
tallization from ethanol as colorless crystals. For 4b: mp 89 °C; MS
1
m/z 449 (M+ - Et); H NMR (δ in CDCl3) -0.19 (s, 3H), 0.37 (dq,
1H, J ) 14.8, 7.6 Hz), 0.52 (dq, 1H, J ) 14.8, 7.6 Hz), 0.55 (s, 3H),
0.66 (t, 2H, J ) 7.7 Hz), 0.94 (t, 3H, J ) 7.7 Hz), 1.06 (br q, 2H, J )
6.0 Hz), 1.42 (d, 2H, J ) 7.8 Hz), 3.90 (t, 1H, J ) 7.4 Hz), 7.07-7.19
(m, 10H), 7.21-7.34 (m, 5H), 7.36-7.39 (m, 2H), 7.51 (dd, 1H, J )
7.1, 1.3 Hz), 7.71 (dd, 1H, J ) 6.2, 1.3 Hz); 13C NMR (δ in CDCl3)
-2.29, -1.85, 7.45, 7.64, 7.69, 8.39, 22.80, 47.22, 125.80, 125.84,
127.47, 127.51, 127.54, 127.65, 127.83, 128.13, 128.19, 128.79, 134.50,
136.37, 136.77, 139.17, 142.51, 144.86, 146.88, 146.94; 29Si NMR (δ
in CDCl3) -4.54, 0.56.
Photolysis of 2a in the Presence of 2,3-Dimethylbutadiene. A
mixture of 0.152 g (0.574 mmol) of 2a (>99% de), 0.382 g (4.65 mmol)
of 2,3-dimethylbutadiene, 14.9 mg (0.086 mmol) of dodecane as an
internal standard, and 70 mL of hexane was irradiated internally with
a low-presure mercury lamp at -78 °C for 4 h. At this stage, 91% of
the starting compound 2a was photolyzed. The resulting mixture was
(18) Ohshita, J.; Matsuguchi, A.; Furumori, K.; Hong, R.-F.; Ishikawa,
M.; Yamanaka, T.; Koike, T.; Shioya, J. Macromolecules 1992, 25, 2134.

Silicon-Carbon Unsaturated Compounds
J. Am. Chem. Soc., Vol. 118, No. 29, 1996 6859
analyzed by GLC as being 5a and 5a′ (73% combined yield) and
ethylmethylphenylsilane (14% yield). GLC retention time and the mass
spectrum for ethylmethylphenylsilane were identical with those of the
authentic sample. After evaporation of the solvent and excess 1,3-
dimethylbutadiene, the resulting mixture was treated with gel perme-
ation chromatography eluting with chloroform to give a 53:47 mixture
of 5a and 5a′. Products 5a and 5a′ cannot be isolated and analyzed as
the mixture: MS m/z 380 (M+); 1H NMR (δ in CDCl3) 0.10 (s, 1.59H),
0.12 (s, 1.41H), 0.60 (s, 3H, MeSi), 0.61-0.71 (m, 2H), 0.72 (t, 3H,
J ) 9.5 Hz), 0.84 (d, 2H, J ) 6.7 Hz), 0.85 (d, 3H, J ) 6.8 Hz), 0.97
(t, 3H, J ) 7.5 Hz), 1.09 (q, 2H, J ) 7.3 Hz), 1.55 (s, 1.59H), 1.56 (s,
1.41H), 2.21 (sextet, 1H, J ) 6.9 Hz), 4.52 (br s, 1H), 4.55 (s, br s,
0.47H), 4.57 (s, br s, 0.53H), 7.28-7.37 (m, 5H), 7.40 (br d, 2H, J )
6.7 Hz), 7.60-7.70 (m, 2H); 13C NMR (δ in CDCl3) for 5a -1.89,
-1.79, 7.56, 7.73 (overlapping with signals of 5a′), 7.99, 8.45, 18.59,
22.30, 22.81, 37.32, 108.11, 127.41, 127.61, 127.81, 128.75, 134.51,
136.29, 136.80, 136.82, 139.29, 142.54, 145.48, 145.65, 152.44, 152.52
(overlapping with those of 5a′); 13C NMR (δ in CDCl3) for 5a′ -2.12,
-1.81, 8.03, 8.40, 18.54, 22.24, 22.74, 37.37, 108.16, 142.49, other
signals are overlapped by those of 5a; 29Si NMR (δ in CDCl3) -4.45,
0.44 (overlapping). Anal. Calcd for C24H36Si2: C, 75.72; H, 9.53.
Found: C, 75.69; H, 9.52 (as a 30:23:25:21 mixture of 5a, 5a′, 5b,
and 5b′).
Photolysis of 2b in the Presence of 2,3-Dimethylbutadiene. A
mixture of 0.325 g (1.09 mmol) of 2b (90% de), 0.218 g (2.66 mmol)
of 2,3-dimethylbutadiene, 26.4 mg (0.155 mmol) of dodecane as an
internal standard, and 70 mL of hexane was irradiated internally with
a low-presure mercury lamp at -78 °C for 3 h. At this stage, 93% of
the starting compound 2b was photolyzed. GLC analysis of the mixture
showed the formation of a mixture of isomers 5a, 5a′, 5b, and 5b′
(77% combined yield) and ethylmethylphenylsilane (14% yield). GLC
retention time and the mass spectrum for ethylmethylphenylsilane were
identical with those of the authentic sample. After evaporation of the
solvent and excess 1,3-dimethylbutadiene, the resulting mixture was
treated with gel permeation chromatography eluting with chloroform
to give a mixture of 5a, 5a′, 5b, and 5b′ in a ratio of 4:2:57:37.
Products 5a, 5a′, 5b, and 5b′ cannot be isolated and analyzed as the
1
mixture. For 5b and 5b′: MS m/z 380 (M+); H NMR (δ in CDCl3)
0.07 (s, 1.19H), 0.09 (s, 1.81H), 0.60 (s, 3H), 0.67 (dq, 1H, J ) 6.8,
3.5 Hz), 0.70 (dq, 1H, J ) 6.4, 3.5 Hz), 0.77 (t, 3H, J ) 6.4 Hz), 0.82
(d, 3H, J ) 6.9 Hz), 0.87 (d, 2H, J ) 6.8 Hz), 0.97 (t, 3H, J ) 7.4
Hz), 1.09 (q, 2H, J ) 7.4 Hz), 1.55 (s, 3H), 2.21 (sextet, 1H, J ) 6.9
Hz), 4.52 (br s, 1H), 4.55 (br s, 0.61H), 4.57 (br s, 0.39H), 7.26-7.37
(m, 5H), 7.41 (br d, 2H, J ) 6.3 Hz), 7.66-7.70 (m, 2H); 13C NMR
(δ in CDCl3) for 5b -2.17, -1.83, 7.62, 7.74 (overlapping with a signal
of 5b′), 8.13, 8.44, 18.50, 22.22, 22.80 (overlapping with a signal of
5b′), 37.45, 108.22, 127.41, 127.62, 127.82, 128.77, 134.52, 136.25,
136.82, 136.85 (overlapping with those of 5b′), 139.33, 142.46, 145.67,
152.39; 13C NMR (δ in CDCl3) for 5b′ -1.93, -1.74, 8.03, 8.37, 18.65,
22.40, 37.33, 108.12, 139.39, 142.53, 145.52, 152.51, other signals are
overlapped by those of 5b; 29Si NMR (δ in CDCl3) -4.45, 0.40 (5b),
-4.42, 0.36 (5b′).
Acknowledgment. We thank Shin-Etsu Chemical Co. Ltd.,
Nitto Electric Industrial Co. Ltd., Dow Corning Asia Ltd.,
Toshiba Silicone Co. Ltd., Kaneka corporation, and Sumitomo
Electric Industry for financial support. We also thank Dr.
Yoshio Aso for assistance in the single-crystal X-ray analysis
for compounds 1a and 4b.
Supporting Information Available: Tables of atomic
coordinates and equivalent isotropic thermal parameters and
anisotropic thermal parameters for 1a and 4b (5 pages). See
any current masthead page for ordering and Internet access
instructions.
JA960458U