Reactivity of rhodium-triflate complexes with diphenylsilane: Evidence for silylene intermediacy in stoichiometric and catalytic reactions

Article abstract of DOI:10.1002/chem.200400568

Addition of Ph₂SiH₂ to [Rh-(iPr₃P) ₂(OTf)] (1) yielded the thermally unstable Rh^III adduct [Rh(iPr₃P)₂-(OTf)(H)(SiPh₂H)] (2), which decomposed to [Rh(iPr₃P)

Full text of DOI:10.1002/chem.200400568

FULL PAPER
Reactivity of Rhodium–Triflate Complexes with Diphenylsilane: Evidence for
Silylene Intermediacy in Stoichiometric and Catalytic Reactions
Roman Goikhman and David Milstein*[a]
Abstract: Addition of Ph2SiH2 to [Rh-
(iPr3P)2(OTf)] (1) yielded the thermal-
ly unstable RhIII adduct [Rh(iPr3P)2-
(OTf)(H)(SiPh2H)] (2), which decom-
posed to [Rh(iPr3P)2(H)2(OTf)] (3),
liberating (unobserved) silylene. The si-
lylene was trapped by 1, resulting in
the RhI–silyl complex [Rh(iPr3P)2-
(SiPh2OTf)]. Complex 3 was converted
to 2 by addition of diphenylsilane, pro-
viding a basis for a possible catalytic
cycle. The last reaction did not involve
a RhI intermediate, as shown by a la-
beling study. Complex 1 catalyzed the
dehydrogenative coupling of Ph2SiH2
action of Ph2SiH2 with styrene in the
presence of 1 depends on the complex/
substrate ratio; under stoichiometric
conditions olefin hydrogenation pre-
vailed over hydrosilylation, whereas
with excess of substrates hydrosilyla-
tion prevailed. Catalytic hydrosilylation
resulted in double addition giving
Ph2Si(CH2CH2Ph)2. Mechanistic as-
pects of the reported processes are dis-
À
to Ph2HSi SiHPh2. A mechanism in-
volving a silylene intermediate in this
catalytic cycle is proposed. The mecha-
nism is supported by complete lack of
catalysis in the case of the tertiary si-
lanes Ph2MeSiH and PhMe2SiH, and
by a study of individual steps of the
catalytic cycle. The outcome of the re-
cussed, and
a new hydrosilylation
Keywords: dehydrogenative cou-
pling · diphenylsilane · elimination ·
hydrosilylation · silylene
mechanism based on silylene interme-
diacy is proposed.
Introduction
chemistry of unsaturated rhodium complexes, including in-
termolecular silylene transfer, and propose silylene-based
mechanisms for dehydrogenative coupling of silanes and for
the industrially important hydrosilylation of alkenes.
Transition metals catalyze various synthetically important
reactions involving silanes, such as hydrosilylation of olefins
and other unsaturated substrates, dehydrogenative coupling
of silanes, and various polymerization and polycondensation
reactions.[1] The mechanisms of these processes have been
extensively studied, but the operating mechanism in each
case is still under discussion. For example, several mecha-
nisms describing dehydrocoupling of silanes[1f–j] and hydrosi-
lylation of olefins[1a,b,m] have been proposed. They include an
oxidative addition/reductive elimination sequence, s-bond
metathesis, and transition-metal–silylene intermediate com-
plexes. The reported evidence of intermediate silylene com-
plexes in catalytic reactions is mostly indirect. However, a
number of stable metallo–silylene complexes have been syn-
thesized and investigated, as summarized in several re-
views.[1b,d,2] We report here new findings in the silylene
Results and Discussion
Reaction of the RhI complex [Rh(iPr3P)2(OTf)] (1)[3] with
Ph2SiH2 in toluene at À308C quantitatively yielded the oxi-
dative addition RhIII product [Rh(iPr3P)2(OTf)(H)(SiPh2H)]
(2).[4] Two-dimensional 29Si NMR spectroscopy at À308C
showed a clear doublet of triplets at d=19 ppm. This com-
pound was unstable above 08C and decomposed to yield the
dihydride [Rh(iPr3P)2(H)2(OTf)] (3)[3a] in addition to several
minor complexes and a small amount of tetraphenyldisilane
(see detailed discussion below).
When two equivalents of 1 were treated with Ph2SiH2 at
À308C, and the reaction mixture allowed to warm up to
08C and left at this temperature overnight, formation of
complex 3 and a new RhI compound (4) were detected by
NMR spectroscopy. Following the reaction by 31P NMR
spectroscopy, we observed the slow simultaneous disappear-
ance of signals assigned to 1 and 2, and generation of signals
assigned to 3 and 4 in the exact proportions: 1+2=3+4.
[a] Dr. R. Goikhman, Prof. D. Milstein
Department of Organic Chemistry
The Weizmann Institute of Science, Rehovot 76100 (Israel)
Fax : (+972)8-934-4142
E-mail: david.milstein@weizmann.ac.il
Chem. Eur. J. 2005, 11, 2983 – 2988
DOI: 10.1002/chem.200400568
ꢀ 2005 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
2983

The new compound 4 was identified as [Rh(iPr3P)2-
(SiPh2OTf)] by 1H, 31P, and two-dimensional 29Si,1H NMR
measurements. The spectral data of 2 and 4 clearly reflect
A partial silylene character of a SiPh2OTf ligand was ob-
served with a Ru complex based on NMR data, although
the triflate group was found to be covalently bound to the
silicon in nonpolar solvents.[9] Some resonance contribution
of a Rh=Si double bond might make a tricoordinated struc-
ture of 4 more stable. The broad 31P NMR signal probably
indicates a dynamic process. Due to the bulk of the ligands,
a dimeric structure seems to be unlikely. It is possible that
the triflate ligand is involved in (reversible) intramolecular
coordination to the unsaturated rhodium center. Rhodium(i)
14e complexes, stabilized by weak interactions, have been
isolated.[10] A tri-coordinate RhI complex was characterized
spectroscopically at low temperature,[11a] and such a com-
plex, lacking stabilizing interactions, was also isolated very
recently.[11b] Unfortunately, complex 4 is not stable and
cannot be isolated.
1
the difference in Rh oxidation state. The J(Rh,P) coupling
1
constants are 112 and 150 Hz, respectively. Similar J(Rh,P)
coupling constants of 110–117 Hz were reported for [Rh-
1
(iPr3P)2(SiAr3)(H)(Cl)] complexes,[4] whereas values for J-
(Rh,P) higher than 130 Hz usually indicate RhI structures.[5]
A doublet of triplets is observed in the 29Si NMR spec-
2
trum of 2 at d=19 ppm (1J(Rh,Si)=62 Hz, J(P,Si)=14 Hz)
and in that of 4 at d=71.3 ppm (1J(Rh,Si)=106 Hz, 2J-
(P,Si)=19 Hz). 29Si NMR spectroscopy of a RhIII–SiHPh2
complex was reported to give rise to a signal at d=15 ppm
(1J(Rh,Si)=43 Hz).[6] The considerable downfield shift (d=
52 ppm) of the 29Si NMR signal of the SiPh2OTf ligand rela-
tive to that of SiPh2H has been reported for organic com-
pounds: 29Si NMR signals for HSiPh2OTf and H2SiPhOTf[7a]
appear at 27 and 52 ppm downfield from those of H2SiPh2
and H3SiPh,[7b] respectively.
Apparently, the decomposition of 2 involves a-hydrogen
À
elimination from a Rh SiPh2H moiety, followed by elimina-
tion of the silylene SiPh2 from the Rh center. a-Hydrogen
À
Complex 4 was also generated independently from reac-
tion of [Rh(iPr3P)2(h3-CH2Ph)][8] with HSiPh2OTf[7a] (see
Scheme 1). This last reaction resulted in several other prod-
transfer from silicon to a metal center in M SiR2H systems
has been reported with other metal complexes.[2d,12] When
two equivalents of 1 react with one equivalent of the silane,
excess of the RhI starting complex can serve as a silylene
trapping agent. An alternative mechanism involving dihy-
drogen elimination from 2 followed by H2 trapping with 1 is
unlikely, based on the following experiment. When a so-
lution of 2 in mesitylene was decomposed under vacuum
and, in parallel, under atmospheric pressure, no differences
were detected in the ratio of the products after completion
of the reactions. The reactions resulted in the formation of
the expected dihydride complex 3 as a major product rather
than the RhI–silyl 4. The experiment was carried out with
two different concentrations, and no difference whatsoever
À
À
ucts due to side reactions, such as both Si H and Si OTf
À
addition followed by Si C coupling, and compound 4 was
formed in about 15% yield (by NMR spectroscopy). Prepa-
ration of a RhI–silyl complex from a diphosphine–Rh–
benzyl precursor and hydridosilane was reported previously.[5d]
Although the NMR data and the independent generation
of 4 clearly indicate that 4 is a RhI–silyl complex, and that
there is no Rh-Si-Rh bridge according to the 29Si NMR spec-
trum, the exact structure of compound 4 is still ambiguous.
in the 31P and H NMR patterns of the final products was
1
observed. This tends to rule out H2 elimination from 2, since
the liberated H2 would be expected (at least partially) to be
removed under vacuum, decreasing the amount of the dihy-
dride 3 formed.
Other experiments aimed at trapping the silylene product
were carried out with tBuMe2SiOH, since OH bonds are
known to be good trapping agents for silylenes.[13a–d] The si-
lanol was chosen due to its bulk which can minimize side re-
actions, and also due to the easy characterization and high
stability of the desired siloxane product.
Decomposition of complex 2 in the presence of two
equivalents of the silanol resulted in formation of the dihy-
À
dride 3 and the expected unsymmetrical siloxane tBuMe2Si
À
O SiPh2H (over 90% by NMR spectroscopy based on 2).
Following the slow decomposition of 2 in the presence of
the silanol at À308C by 31P NMR spectroscopy revealed
first-order kinetics in complex 2 (kobs =1.3ꢁ10À4 sÀ1). The si-
loxane product was also synthesized independently by reac-
À À
À
tion of tBuMe2Si O H with Cl SiPh2H in the presence of
pyridine. These transformations are summarized in
Scheme 1.[14]
Remarkably, the RhIII–dihydride 3 reacted with Ph2SiH2
at 08C to yield compound 2, which in turn decomposed
Scheme 1. Generation and trapping of diphenylsilylene in the RhI–
Ph2SiH2 system.
2984
ꢀ 2005 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chemeurj.org
Chem. Eur. J. 2005, 11, 2983 – 2988

Rhodium–Triflate Complexes
FULL PAPER
back to 3, liberating the unobserved silylene intermediate.
This reaction sequence forms a basis for catalytic generation
of silylenes under very mild conditions.[15] Again, no RhI in-
termediate was involved in the reaction of 3 with the silane,
as was confirmed by the following labeling experiment. The
deuterio analogue of 3 was prepared from 1 and D2, and
treated with diphenylsilane. At the beginning of the reaction
only [Rh(iPr3P)2(OTf)(D)(SiPh2H)] was detected by NMR
À
spectroscopy, and D H scrambling took place only at later
stages of the reaction. Thus the silane seems to react with
the RhIII–dihydride complex rather than with a RhI inter-
mediate formed by H2 reductive elimination.
Scheme 3. Curtis–Epstein mechanism for dehydrogenative coupling of
Ph2SiH2 catalyzed by a RhI complex.
When a 50-fold excess of the diphenylsilane was added to
complex 1 at room temperature (or at 08C), catalytic dehy-
drogenative coupling to give (Ph2SiH)2 took place (92%
yield based on starting Ph2SiH2, at 08C, 24 h). Since decom-
position of 2 very likely proceeds via a silylene intermediate,
ditions. To ascertain that the methyl substituent does not
play a crucial role in reactivity, we performed the reaction
with the secondary silane PhMeSiH2, resulting in the expect-
ed catalytic formation of disilane (PhMeSiH)2 under the
same conditions that were used in the case of Ph2SiH2.
The difference in the reactivity of Ph2SiH2 (and PhMe-
SiH2) in comparison with that of the tertiary silanes is strik-
ing: no catalysis at all in case of tertiary silanes was detect-
ed, whereas the secondary silane was rapidly converted cata-
lytically to the disilane. Steric reasons can be ruled out by
comparison of cone angles of the silanes. The available cone
angles of phosphines[18] are reported to be identical to those
of analogous silyl ligands.[19] Thus, the cone angles of the
SiHPh2 and SiMe2Ph groups are 1288 and 1228, respectively,
that is, HSiMe2Ph is sterically less demanding than Ph2SiH2.
In addition, since HSiMe2Ph and H2SiMePh are very similar
electronically, major influence of electronic factors on reac-
tivity can be excluded as well.
the mechanism of catalysis may involve silylene insertion
[16]
À
into the Si H bond of a free silane (Scheme 2). A similar
Thus, since the tertiary silanes we studied are very similar
sterically and electronically to diphenylsilane, we have to at-
tribute the strikingly different reactivity to the number of
À
Si H bonds in the silane.
These results demonstrate that a SiH2 group is essential,
and strongly support the suggested mechanism involving ac-
tivation of both hydrogen atoms at one silicon center, appa-
rently generating silylene (free or bound) intermediates.
Our mechanism is supported not only by the silylene trap-
ping in a single step by using two different trapping agents,
Scheme 2. Proposed mechanism for catalytic dehydrogenative coupling of
Ph2SiH2.
but also by extensive previous reports that demonstrated a-
[12]
À
elimination from M SiR2H fragments and insertion of si-
À
mechanism including silane addition to the M=SiR2 bond of
lylenes into Si H bonds.[13,16] Moreover, a very similar mech-
À À
an intermediate complex resulting in formation of H M
anism was recently proposed and investigated for catalytic
dehydrocoupling of stannanes.[20] Our proposed mechanism
includes a silylene elimination step from a hydrido silyl com-
plex, rather than H2 elimination, which was suggested by
Ojima. The mechanism we propose can lead to formation of
À
SiR2 SiR2H, followed by reductive elimination of the disi-
lane was proposed by Ojima et al. 30 years ago,[17a] but there
was no direct evidence for the silylene intermediate, and
this mechanism could not explain formation of higher poly-
silanes. The more accepted Curtis and Epstein mecha-
À
higher silanes as a result of silylene insertion into a Si H
nism[17b] includes silane double addition followed by Si Si
bond of the primary disilane product, although it has not
been observed in our case. Nevertheless, at this stage
Ojimaꢂs mechanism cannot be ruled out completely for our
system.
À
elimination from the metal center (Scheme 3). To clarify the
mechanism we checked the reaction of tertiary silanes that
are similar both sterically and electronically to Ph2SiH2,
namely Ph2MeSiH and PhMe2SiH. In both cases no catalytic
reaction was observed with complex 1 under the same con-
Coming back to the decomposition of 2 above 08C, we
can now identify and explain the resulting mixture. Oxida-
Chem. Eur. J. 2005, 11, 2983 – 2988
www.chemeurj.org
ꢀ 2005 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
2985

D. Milstein and R. Goikhman
tive addition of silanes to RhI complexes is reversible,[21] and
hence small amounts of complex 1 and free diphenylsilane
are present in the reaction mixture. Conversion of 2 to 3 is
accompanied by the generation of diphenylsilylene, which is
trapped by small amounts of 1 and Ph2SiH2, giving complex
4 and HSiPh2SiPh2H, respectively (detected by multinuclear
NMR spectroscopy).
Another potential silylene trapping agent used was sty-
rene. Interestingly, when a 1:1 mixture of styrene and diphe-
nylsilane was added to a stoichiometric amount of 1, forma-
tion of compound 2 was initially observed. Compound 2 de-
composed in the presence of styrene, and the final products
of this reaction sequence were complexes 1 and 4, along
with a small amount of 3. The main organic products were
ethylbenzene and bis(2-phenylethyl)diphenylsilane, formed
in a ratio of 5:1.
The proposed mechanism is depicted in Scheme 4. The
first step (oxidative addition of the silane) was discussed
above. The next expected steps (reversible coordination of
the olefin and formation of intermediate A) are well-de-
fined processes in a catalytic hydrosilylation.[1c] Intermediate
A is similar to complex 2, and it most probably decomposes
in a similar manner through a-hydrogen elimination (com-
pare with Scheme 2) due to the presence of a-SiH and a
labile triflate ligand. However, the resulting silylene inter-
mediate in this case can be intramolecularly trapped by the
alkyl ligand to form irreversibly intermediate B, as was sug-
gested by Tilley et al.[22] Alternatively, intermediate B can
À
be formed from A also by Si C reductive elimination fol-
[1c]
À
lowed by Si H oxidative addition, as shown in Scheme 4.
Intermediate B can easily react with styrene by means of
the traditional hydrosilylation mechanism, resulting in the
final products depicted in Scheme 4.
Apparently, under stoichiometric conditions, coordination
of styrene to the intermediate complex 2 is much slower
than the a-hydrogen elimination process leading to complex
3. Following this step, styrene dehydrogenates the dihydride
complex 3, resulting in 1 and ethylbenzene (Scheme 4 top).
When ten equivalents of each silane and styrene were added
to one equivalent of complex 1, a catalytic reaction took
place, resulting in the formation of ethylbenzene and bis(2-
phenylethyl)diphenylsilane as the main products in a ratio
of 2:5 (as opposed to 5:1 under stoichiometric conditions).
This result is likely to be due to reversible coordination of
styrene to intermediate 2, with excess styrene directing the
process to hydrosilylation.
Interestingly, formation of the product of addition of two
styrene molecules to one molecule of silane is preferred to
the mono-addition product, which was not observed in the
reactions. A possible explanation for this unusual finding is
that intermediate A (Scheme 4) undergoes fast rearrange-
ment to intermediate B, which adds a second olefin mole-
À
cule at a much faster rate than Si H reductive elimination,
À
followed by irreversible Si C elimination. A similar process
was reported with diphenylsilane and ethylene with another
Rh system, in which the mono-addition product was not de-
tected;[23] however, it was detected when ethylene was re-
placed with the less reactive 1-butene. The proposed inter-
mediate step included a bridging silylene complex.[23] Anoth-
er mechanism involving a silylene intermediate complex in
the hydrosilylation reaction as a result of a-hydrogen elimi-
À À
nation from a M Si H moiety has been proposed recent-
ly[24] and studied computationally.[25]
Conclusion
A rare example of a-hydrogen elimination in a Rh–silyl
complex was detected, leading to a (unobserved) silylene in-
termediate, which was trapped with a cationic RhI com-
plex[26] or with an organic silanol. Based on our results, a
mechanism for dehydrogenative coupling of silanes involv-
ing a silylene intermediate was proposed. An interesting
competition between hydrosilylation and hydrogenation in a
reaction of an olefin with a silane was reported and dis-
cussed. A new mechanism for catalytic hydrosilylation re-
sulting in olefin double addition to a silane substrate was
proposed.
Experimental Section
Scheme 4. Proposed mechanism for catalytic hydrosilylation and hydro-
genation of styrene with Ph2SiH2.
General: All the manipulations of air- and moisture-sensitive compounds
were carried out by using a nitrogen-filled Vacuum Atmospheres glove
2986
ꢀ 2005 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chemeurj.org
Chem. Eur. J. 2005, 11, 2983 – 2988

Rhodium–Triflate Complexes
FULL PAPER
box. Solvents were purified by standard procedures, degassed, and stored
over molecular sieves in the glove box. All the reagents were of reagent
grade. Commercially available silanes were used without further purifica-
tion. Rhodium complexes were synthesized according to or similarly to
literature procedures, as referenced in the paper. NMR spectra were ob-
tained with a Bruker AMX 400 spectrometer at ambient probe tempera-
ture in [D8]toluene unless otherwise specified.
(16 mg, 0.12 mmol) in [D8]toluene (0.3 mL) of at À308C. Complex 2 was
formed quantitatively (by NMR spectroscopy). The reaction mixture was
followed by NMR spectroscopy overnight, resulting in slow decomposi-
tion of complex 2 to the dihydride 3. The major organic product was tBu-
1
SiMe2-O-SiHPh2 (90% based on the initial Ph2SiH2 amount). H NMR of
tBuSiMe2-O-SiHPh2: d=7.63 (m, 4H; Ph), 7.16 (m, 6H; Ph), 5.80 (s, 1H;
1
Si-H, J(Si,H)=214 Hz), 0.93 (s, 9H; tBu), 0.08 ppm (s, 6H; Me).
Independent synthesis of tBuSiMe2-O-SiHPh2:
Me2OH (13 mg, 0.1 mmol) and pyridine (8 mg, 0.1 mmol) in C6H6
(0.5 mL) was added dropwise to solution of Ph2SiHCl (22 mg,
A solution of tBuSi-
Reactions of [Rh(iPr3P)2OTf] (1) with Ph2SiH2
Ratio of reagents 1/Ph2SiH2 1:1: A solution of 1 (57 mg, 0.1 mmol) in
[D8]toluene (0.6 mL) was treated with Ph2SiH2 (18 mg, 0.1 mmol) at
À308C. The color of the solution changed immediately from deep violet
to brown. Complex 2 was formed quantitatively (by NMR spectroscopy).
It slowly decomposed at 08C to the dihydride complex 3 in approximate-
ly 80% yield and small amounts of 4 and tetraphenyldisilane. NMR data
of 2: 31P{1H}: d=43.6 ppm (d, 1J(Rh,P)=112 Hz, 2P); 1H NMR: d=7.87
(m, 4H; Si-Ph), 7.20 (m, 4H; Si-Ph), 7.10 (m, 2H; Si-Ph), 5.44 (ddt, 3J-
(P,H)=12.3 Hz, 2J(Rh,H)=2.3 Hz, 3J(H,H)=2.0 Hz, 1H; Si-H), 2.22 (m,
6H; P-CH(CH3)2), 1.04 (m, 36H; P-CH(CH3)2), À21.85 ppm (ddt, 2J-
(Rh,H)=29.6 Hz, 3J(P,H)=13.7 Hz, 3J(H,H)=2.0 Hz, 1H; Rh-H); 29Si:
a
0.1 mmol) in C6H6 (0.5 mL) at room temperature. The reaction mixture
was stirred for half hour, the pyridinium salt was filtered off, and the so-
lution was evaporated to give the desired siloxane in 80% yield (25 mg).
1
The H NMR data were identical to that reported above.
Reaction of [Rh(iPr3P)2(H)2OTf] (3) with Ph2SiH2
Method 1: A solution of complex 3 (17 mg, 0.03 mmol) in [D8]toluene
(0.6 mL) was treated with Ph2SiH2 (5.5 mg, 0.03 mmol) at 08C, producing
complex 2 as a major product within minutes. Compound 2 decomposed
overnight to the starting complex 3 and a mixture of unidentified silicon-
containing products.
1
2
d=19 ppm (dt, J(Rh,Si)=62 Hz, J(P,Si)=14 Hz).
Ratio of reagents 1/Ph2SiH2 2:1: A solution of 1 (57 mg, 0.1 mmol) in
[D8]toluene (0.6 mL) was treated with Ph2SiH2 (9 mg, 0.05 mmol) at
À308C. The color of the solution changed from deep violet to dark
brown and a mixture of complexes 1 and 2 in approximately 1:1 ratio was
observed by NMR spectroscopy. The reaction mixture was warmed up to
08C and was followed by NMR spectroscopy overnight. Slow decomposi-
tion of complex 2 was accompanied by an equivalent consumption of
complex 1 and simultaneous appearance of complexes 3 and a new RhI
complex in 1:1 ratio. The latter compound was assigned as [Rh(iPr3P)2-
(SiPh2OTf)] (4). NMR of 4: 31P{1H}: d=45.2 ppm (d, 1J(Rh,P)=150 Hz,
2P); 1H NMR: d=8.40 (m, 4H; Si-Ph), 7.23 (m, 4H; Si-Ph), 7.13 (m,
2H; Si-Ph), 2.08 (m, 6H; P-CH(CH3)2), 1.02 ppm (m, 36H; P-CH-
Method 2: A solution of [Rh(iPr3P)2(D)2OTf] (prepared analogously to
3[3] from 1 using deuterium bubbling) (17 mg, 0.03 mmol) in [D8]toluene
(0.6 mL) was treated with Ph2SiH2 (5.5 mg, 0.03 mmol) at À108C. Only
[Rh(iPr3P)2(D)(SiPh2H)OTf] was detected by 1H and 2H NMR spectros-
copy at early stages of the reaction. After 15 min a scrambling between
À
À
Rh D and Si H was observed.
Reactions of 1 with a mixture of Ph2SiH2 and styrene
Ratio of reagents 1/Ph2SiH2/styrene 1:1:1:
A
solution of
1
(17 mg,
0.03 mmol) in [D8]toluene (0.5 mL) was treated with
a
mixture of
Ph2SiH2 (5.5 mg, 0.03 mmol) and styrene (3 mg, 0.03 mmol) at À308C.
The color of the solution immediately changed from deep violet to
brown. Complex 2 was formed quantitatively (by NMR spectroscopy).
The reaction mixture was warmed up to 08C and followed by NMR spec-
troscopy overnight. Slow decomposition of complex 2 was observed, ac-
companied by initial appearance of the dihydride 3, which transformed
to complex 1 towards the end of the reaction. The major organic prod-
ucts were ethylbenzene (60% NMR yield, based on the initial amount of
styrene) and bis(phenethyl)diphenylsilane (12% NMR yield, based on
the initial amount of diphenylsilane), along with some unidentified sili-
con-containing products. 1H NMR data of bis(phenethyl)diphenylsilane:
d=7.0–7.5 (overl. m, 20H; Ph), 2.58 (m, 4H; Ph-CH2), 1.33 ppm (t, 4H;
Si-CH2). The NMR spectrum was identical to that of the commercially
purchased compound (Aldrich).
1
2
(CH3)2); 29Si: d=71.3 ppm (dt, J(Rh,Si)=106 Hz, J(P,Si)=19 Hz).
Catalytic experiment with a ratio of reagents 1/Ph2SiH2 1:50: Ph2SiH2
(92 mg, 0.5 mmol) was added to a solution of 1 (6 mg, 0.01 mmol) in
[D8]toluene (0.6 mL), and the reaction was stirred overnight at 08C.
Quantitative conversion of the silane was detected, and (Ph2SiH)2 was
formed in 92% yield (based on starting Ph2SiH2), along with a small
amount of Ph3SiH and traces of other silicon species.
Ratio of reagents 1/Ph2SiH2 1:1 under vacuum: A solution of 1 (80 mg,
0.14 mmol) in mesitylene (2 mL) was treated with Ph2SiH2 (26 mg,
0.14 mmol) at À308C. The color of the solution changed immediately
from deep violet to brown, indicating the formation of adduct 2. The re-
action mixture was separated into two equal portions at À308C. One of
the portions was kept under vacuum for 4 h at room temperature, where-
as another portion was allowed to stand for the same period of time at
normal conditions. 31P NMR spectroscopy of the final mixture was identi-
cal in the two experiments.
Ratio of reagents 1/Ph2SiH2/styrene 1:10:10: A solution of 1 (17 mg,
0.03 mmol) in [D8]toluene (0.5 mL) was treated with
a mixture of
Ph2SiH2 (55 mg, 0.3 mmol) and styrene (30 mg, 0.3 mmol) at À308C. The
color of the solution immediately changed from deep violet to brown,
and complex 2 was formed quantitatively (by NMR spectroscopy). The
reaction mixture was warmed up to 08C and followed by NMR spectros-
copy overnight. Decomposition of complex 2 to 1 was observed, and the
major organic products were ethylbenzene (15% NMR yield) and bi-
s(phenethyl)diphenylsilane (37% NMR yield), along with some unidenti-
fied silicon-containing products.
When the experiment was carried out with 5 mL of mesitylene (instead
of 2 mL) almost no difference in the final product ratio was observed.
Complex 3 was the major product, along with some starting adduct 2 and
a small amount of 4.
Reaction of [Rh(iPr3P)2OTf] (1) with PhMeSiH2: PhMeSiH2 (12 mg,
0.1 mmol) was added to a solution of 1 (6 mg, 0.01 mmol) in [D8]toluene
(0.6 mL), and the reaction was stirred overnignt. Over 90% conversion
of the silane was detected, and (PhMeSiH)2 was formed in 92% yield
(based on starting PhMeSiH2), along with a small amount of other silicon
species.
Acknowledgements
Reactions of [Rh(iPr3P)2OTf] (1) with excess of Ph2MeSiH and PhMe2-
SiH: The silane (0.5 mmol; 68 mg of PhMe2SiH or 99 mg of Ph2MeSiH)
was added to a solution of 1 (6 mg, 0.01 mmol) in [D8]toluene (0.6 mL),
and the reaction was stirred overnignt at 08C. No catalytic reaction was
detected by 1H NMR spectroscopy. Stoichiometric products were not an-
alyzed.
We thank the Israel Science Foundation, Jerusalem, the Minerva Founda-
tion, Munich, and the Helen and Martin Kimmel Center for Molecular
Design for financial support. D.M. is the Israel Matz Professor of Organic
Chemistry.
Reaction of [Rh(iPr3P)2OTf] (1) with a mixture of Ph2SiH2 and tBuSi-
Me2OH: A solution of 1 (34 mg, 0.06 mmol) in [D8]toluene (0.5 mL) was
treated with a mixture of Ph2SiH2 (11 mg, 0.06 mmol) and tBuSiMe2OH
[1] a) M. A. Brook, Silicon in Organic, Organometallic, and Polymer
Chemistry, Wiley-Interscience, New York, 2000; b) T. D. Tilley in
The Chemistry of Organic Silicon Compounds, Part 1 (Eds.: S. Patai,
Chem. Eur. J. 2005, 11, 2983 – 2988
www.chemeurj.org
ꢀ 2005 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
2987

D. Milstein and R. Goikhman
Z. Rappoport), Willey, N.Y., 1989, 1415; c) I. Ojima, Z. Li, J. Zhu,
in The Chemistry of Organic Silicon Compounds, Vol. 3 (Eds.: Z.
Rappoport, Y. Apeloig), Willey, N.Y., 1998, 1687; d) M. S. Eisen, in
The Chemistry of Organic Silicon Compounds, Vol. 3 (Eds.: Z. Rap-
poport, Y. Apeloig), Willey, N.Y., 1998, 2037; e) H. Yamashita, M.
Tanaka, Bull. Chem. Soc. Jpn. 1995, 68, 403; f) C. E. Zybill, C. Liu,
Synlett 1995, 7, 687; g) F. Gauvin, J. F. Harrod, H. G. Woo, Adv. Or-
ganomet. Chem. 1998, 42, 363; h) T. D. Tilley, Comments Inorg.
Chem. 1990, 10, 37; i) T. D. Tilley, Acc. Chem. Res. 1993, 26, 22;
j) J. Y. Corey, Adv. Silicon Chem. 1991, 1, 327; k) J. A. Reichl, D. H.
Berry, Adv. Organomet. Chem. 1998, 43, 197; l) H. K. Sharma, K. H.
Pannell, Chem. Rev. 1995, 95, 1351; m) B. Marciniec, J. Gulinski, J.
Organomet. Chem. 1993, 446, 15, and references therein; n) J. Y.
Corey, J. Braddock-Wilking, Chem. Rev. 1999, 99, 175.
[12] a) B. V. Mork, T. D. Tilley, J. Am. Chem. Soc. 2004, 126, 4375;
b) B. V. Mork, T. D. Tilley, J. Am. Chem. Soc. 2001, 123, 9702;
c) S. R. Klei, T. D. Tilley, R. G. Bergman, J. Am. Chem. Soc. 2000,
122, 1816; d) J. C. Peters, J. D. Feldman, T. D. Tilley, J. Am. Chem.
Soc. 1999, 121, 9871; e) G. P. Mitchell, T. D. Tilley, Angew. Chem.
1998, 110, 2602; Angew. Chem. Int. Ed. Engl. 1998, 37, 2524;
f) R. J. P. Corriu, B. P. S. Chaunan, G. F. Lanneau, Organometallics
1995, 14, 1646.
[13] a) A. G. Brook, M. A. Brook, Adv. Organomet. Chem. 1996, 39, 71;
b) R. Okazaki, R. West, Adv. Organomet. Chem. 232; c) P. P.
Gaspar, R. West, in The Chemistry of Organic Silicon Compounds,
Vol. 3 (Eds.: Z. Rappoport, Y. Apeloig), Willey, N.Y., 1998, 2463;
d) K. P. Steele, W. P. Weber, Inorg. Chem. 1981, 20, 1302, and refer-
ences therein; e) D. Seyferth, D. C. Annarelly, D. P. Duncan, Orga-
nometallics 1982, 1, 1288.
[2] a) H. Ogino, Chem. Rec. 2002, 2, 291; b) S. K. Grumbine, D. A.
Straus, T. D. Tilley, A. L. Rheingold, Polyhedron 1995, 14, 127;
c) P. D. Lickiss, Chem. Soc. Rev. 1992, 21, 271; d) for latest study of
silylene complexes see: B. V. Mork, T. D. Tilley, J. Am. Chem. Soc.
2004, 126, 4375, and references therein.
[3] a) H. Werner, M. Bosch, M. E. Schneider, C. Hahn, F. Kukla, M.
Manger, B. Windmuller, B. Weberndorfer, M. Laubender, J. Chem.
Soc. Dalton Trans. 1998, 21, 3549; b) R. Goikhman, D. Milstein,
Angew. Chem. 2001, 113, 1153; Angew. Chem. Int. Ed. 2001, 40,
1119.
[14] a) An analogous labeling experiment using tBuMe2SiOD (prepared
by deprotonation of tBuMe2SiOH with tBuLi followed by addition
À
of CD3COOD) resulted in H D scrambling. Both tBuMe2SiO-
SiPh2-D and tBuMe2SiO-SiPh2-H were observed in a ratio of 2:3,
À
probably due to reversible Si H oxidative addition. b) Kinetic meas-
urement of silanol disappearance was problematic due to difficulties
in precise integration of 1H NMR signals, which partially overlap
with peaks of the starting compound and product. c) Decomposition
of 2 revealed first-order kinetics at different concentrations and
temperatures.
[4] Similar square-pyramidal [Rh(PiPr3)2(SiAr3)(H)(Cl)] complexes
were structurally characterized, see: K. Osakada, T. A. Koizumi, T.
Yamamoto, Organometallics 1997, 16, 2063.
[15] Catalytic silylene generation was also detected in photochemical de-
composition of oligosilanes: K. H. Pannel, M. C. Brun, H. Sharma,
K. Jones, S. Sharma, Organometallics 1994, 13, 1075.
[5] RhI–silyl complexes exhibit 1J(Rh,P)=130–160 Hz. See, for exam-
ple: a) R. Goikhman, M. Aizenberg, Y. Ben-David, L. J. W. Shimon,
D. Milstein, Organometallics 2002, 21, 5060; b) G. P. Mitchell, D. A.
Straus, T. D. Tilley, A. L. Rheingold, Organometallics 1998, 17, 2912;
c) M. Aizenberg, J. Ott, C. J. Elsevier, D. Milstein, J. Organomet.
Chem. 1998, 551, 81; d) P. Hoffmann, C. Meier, W. Hiller, M.
Heckel, J. Riede, M. U. Schmidt, J. Organomet. Chem. 1995, 490, 51;
e) D. E. Hendriksen, A. A. Oswald, G. B. Ansell, S. Leta, R. V.
Kastrup, Organometallics 1989, 8, 1153.
À
[16] A free silylene insertion into Si H bonds was proposed in catalytic
polymerization of oligosilanes: E. Hengge, M. Weinberger, J. Orga-
nomet. Chem. 1993, 443, 167.
[17] a) I. Ojima, S. I. Inaba, T. Kogure, I. Nagai, J. Organomet. Chem.
1973, 55, C7; b) M. D. Curtis, P. S. Epstein, Adv. Organomet. Chem.
1981, 19, 213.
[18] C. A. Tolman, Chem. Rev. 1977, 77, 313.
[19] D. M. Hester, J. Sun, A. W. Harper, G. K. Yang, J. Am. Chem. Soc.
1992, 114, 5234.
[20] N. R. Neale, T. D. Tilley J. Am. Chem. Soc. 2002, 124, 3802.
[21] M. Aizenberg, R. Goikhman, D. Milstein, Organometallics 1996, 15,
1075.
[22] J. D. Feldman, J. C. Peters, T. D. Tilley, Organometallics 2002, 21,
4065.
[23] M. D. Fryzuk, L. Rosenberg, S. J. Rettig, Organometallics 1996, 15,
2871.
[24] a) P. B. Glaser, T. D. Tilley, J. Am. Chem. Soc. 2003, 125, 13640;
b) M. Itoh, K. Iwata, M. Kobayashi, J. Organomet. Chem. 1999, 574,
241.
[25] C. Beddie, M. B. Hall, J. Am. Chem. Soc. 2004, 126, 13564.
[26] Silylene insertion into a transition-metal–hydride bond is known,
see, for example: D. H. Berry, J. H. Mitstifer, J. Am. Chem. Soc.
1987, 109, 3777.
[6] K. Osakada, S. Sarai, T. A. Koizumi, T. Yamamoto, Organometallics
1997, 16, 3973.
[7] a) W. Uhlig, A. Tzschach, J. Organomet. Chem. 1989, 378, C1;
b) B. J. Helmer, R. West, Organometallics 1982, 1, 877.
[8] H. Werner, M. Schafer, O. Nurnberg, J. Wolf, Chem. Ber. 1994, 127,
27.
[9] D. A. Straus, C. Zhang, G. E. Quimbita, S. D. Grumbine, R. H.
Heyn, T. D. Tilley, A. L. Rheingold, S. J. Geib, J. Am. Chem. Soc.
1990, 112, 2673.
[10] a) Y. W. Yared, S. L. Miles, R. Bau, C. A. Reed, J. Am. Chem. Soc.
1977, 99, 7076; b) H. L. M. Van Gaal, F. L. A. Van den Bekerom, J.
Organomet. Chem. 1977, 134, 237; c) H. Urtel, C. Meier, F. Eisen-
trager, F. Rominger, J. P. Joschek, P. Hofmann, Angew. Chem. 2001,
113, 803; Angew. Chem. Int. Ed. 2001, 40, 781.
[11] a) M. Gandelman, A. Vigalok, L. Konstantinovski, D. Milstein, J.
Am. Chem. Soc. 2000, 122, 9848; b) R. Dorta, E. D. Stevens, S. P.
Nolan, J. Am. Chem. Soc. 2004, 126, 5054.
Received: June 7, 2004
Revised: December 25, 2004
Published online: March 11, 2004
2988
ꢀ 2005 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chemeurj.org
Chem. Eur. J. 2005, 11, 2983 – 2988