Catalytic hydrosilylation of carbonyls via Re(CO)5Cl photolysis

Article abstract of DOI:10.1021/om2012634

The hydrosilylation reaction between silanes and various carbonyl substrates such as aldehyde, ketone, ester, and carbonate has been catalyzed by Re(CO)₅Cl UV photolysis. Kinetic studies have shown that the reaction is favored for the least sterically hindered silanes with aldehydes followed by aliphatic ketones. The IR spectrum of the rhenium carbonyl dimer HRe ₂(CO)₉(SiR₃) has been recorded in the reaction mixture. This complex is believed to be the resting state of the active catalyst Re(CO)₄SiR₃, which could be released upon photactivation. A catalytic mechanism involving this species has been proposed and shown to be thermodynamically feasible using computational studies. In addition, the relative hydrosilylation rates among the various carbonyl substrates can be explained using the same mechanism.

Full text of DOI:10.1021/om2012634

Article
pubs.acs.org/Organometallics
Catalytic Hydrosilylation of Carbonyls via Re(CO)5Cl Photolysis
Chun Keong Toh, Yin Ngai Sum, Wai Kit Fong, Siau Gek Ang, and Wai Yip Fan*
Department of Chemistry, National University of Singapore, 3 Science Drive 3, Singapore 117543.
ABSTRACT: The hydrosilylation reaction between silanes and various carbonyl
substrates such as aldehyde, ketone, ester, and carbonate has been catalyzed by
Re(CO)5Cl UV photolysis. Kinetic studies have shown that the reaction is favored
for the least sterically hindered silanes with aldehydes followed by aliphatic ketones.
The IR spectrum of the rhenium carbonyl dimer HRe2(CO)9(SiR3) has been recorded in the reaction mixture. This complex is
believed to be the resting state of the active catalyst Re(CO)4SiR3, which could be released upon photactivation. A catalytic
mechanism involving this species has been proposed and shown to be thermodynamically feasible using computational studies. In
addition, the relative hydrosilylation rates among the various carbonyl substrates can be explained using the same mechanism.
dissociates a CO ligand readily upon photolysis (quantum
yield of 0.06 to 0.44 from 366 nm to 313 nm). In light of efforts
to utilize energy from light rather than from fossil fuel sources,
much work would have to be carried out to further explore
light-driven catalytic processes. As it is also desirable for the
catalysts to be air and moisture stable and easy to handle, we
have photoactivated Re(CO)5Cl and other rhenium carbonyls
in order to carry out hydrosilylation reactions on a variety of
carbonyl substrates at room temperature. The objectives of the
work are to determine the relative efficiency of the catalysts
toward different carbonyls, identify possible key catalytic
intermediates, and explore the mechanism of hydrosilylation
using a combination of IR and NMR spectroscopy as well as
computational studies.
INTRODUCTION
■
The reaction of silanes with a wide range of carbonyl
compounds remains important in the generation of useful
protecting groups for organic synthesis. For example, hydro-
silylation across the CO bond of aldehydes and ketones
produces protected alcohols in a single step.1 In the presence of
a catalyst, these reactions could be carried out under mild
conditions, in contrast to using harsh reducing agents such as
lithium aluminum hydride. Hydrosilylation catalysis has been
reported for various transition-metal complexes containing
iron,2,3 tungsten,4 molybdenum,5 manganese,6,7 rhenium,1,8
rhodium,9 silver,10 and most recently ruthenium.11 Recent work
was also done on the mechanistic studies of hydrosilylation
using oxo-rhenium complexes.12,13
While Berke et al.8 have reported the use of different
rhenium(I) catalysts for the hydrosilylation of aldehydes and
ketones, one of the more interesting processes is the
CpFe(CO)2Me-catalyzed reaction between silanes and dime-
thylformamide (DMF) reported by Pannell et al.14 As shown in
Scheme 1, hydrosilylation across the CO bond of DMF first
afforded a silyl ether which further reacted with another
molecule of silane to give siloxane and trimethylamine as final
products. Cutler et al.15 have used the manganese complex
(CO)5MnC(O)CH3 to reduce the much less reactive ester to
ether via a silyl acetal intermediate. Silyl esters are another
important class of silicon-containing compounds, as they find
wide use in the production of polymeric materials.16 Mizuno et
al.16 reported that a range of ruthenium catalysts enables the
conversion of carboxylic acids to silyl esters to take place.
Furthermore, Yamamoto’s group showed that a boron catalyst,
B(C6F5)3, reduces carboxylic acids further to the corresponding
alkanes as depicted in Scheme 1.17 However, most of the
reported carbonyl hydrosilylation catalysts require careful
handling, as they are relatively unstable under ambient
conditions.
RESULTS AND DISCUSSION
■
Hydrosilylation of Aldehydes and Ketones. Preliminary
experiments have been carried out to assess the efficiency of
Re(CO)5Cl UV photolysis toward carbonyl hydrosilylation
using Et3SiH and acetone as substrates (Table 1, entry 1). The
quantum yields for Re(CO)5Cl dissociation at 313 and 366 nm
have been determined to be 0.44 and 0.06, respectively.19 The
light source used in our work is broad band, ranging from 300
to 800 nm. Since Re(CO)5Cl absorbs in the 300−350 nm
region, the quantum yield would also be expected to be similar
to the range quoted. 1H NMR analysis of the mixture after 4 h
of photolysis showed isopropoxytriethylsilane, Et3SiOCH-
(CH3)2 (1a), as the sole product (Figure 1), with a turnover
number (TON) determined to be 70−80 with respect to
Re(CO)5Cl (1% mol loading).
The reaction appears to proceed most efficiently if an excess
of silane is used to achieve full conversion of acetone to the
product. A silane:carbonyl ratio of 3:1 has been used for further
comparison with other carbonyl substrates. However, the silyl
ether 1a does not undergo further reaction with excess silane.
Such is also the case in the Re(CO)5Cl-catalyzed reactions of
Et3SiH with other carbonyl substrates (Table 1, entries 2−6).
Although the commercially available rhenium(I) complexes
Re(CO)5X (X = Br, Cl, Re(CO)5) have been reported to
activate Si−H bonds, their activity toward carbonyl hydro-
silylation has not been explored fully.18 It was previously
reported19 that the air- and moisture-stable Re(CO)5Cl
Received: December 20, 2011
Published: May 16, 2012
© 2012 American Chemical Society
3880
dx.doi.org/10.1021/om2012634 | Organometallics 2012, 31, 3880−3887

Organometallics
Article
Scheme 1. Catalytic Hydrosilylation of Carbonyl Compounds
Table 1. Reactions of Et3SiH with Various Aldehydes and
Ketones
Figure 1. 1H NMR peaks of 1a. Peak assignment: HA, 3.99 ppm
(septet); HB, 1.15 ppm (doublet).
Re2(CO)10 has been found to be an effective catalyst as well
(Table 1, entries 8 and 9) with a TON ratio similar to that of
Re(CO)5Cl. However, hydrosilylation proceeded only slug-
gishly when HRe(CO)5 was used (Table 1, entries 10 and 11).
In comparison to previously reported catalysts7 such as
[Re(CH3CN)3Br2(NO)] (TON = 99 for aldehydes/ketones),
a similar efficiency has been observed for Re(CO)5Cl.
However, hydrosilylation with substrates such as 1,3-diketones,
anthraquinones, and acetylacetone did not occur even after
extended photolysis of up to 12 h.
When the substrate was changed from acetone to
benzaldehyde, an increase in the amount of product formed
over the same period was observed (Table 1, entries 1 and 2).
This can be attributed to the increased reactivity of aldehydes in
comparison to that of ketones. When different silanes such as
Ph2SiH2 and Ph3SiH were used, a decrease in the
corresponding silyl ether yield was observed (Table 2). The
relatively lower rates toward hydrosilylation of acetone appear
to bear a strong relation to the steric hindrance of the silane. In
addition, when the carbonyl substrate was changed from
aliphatic to a hindered aromatic carbonyl, i.e. benzophenone,
acetophenone, and benzaldehyde, a decrease in the TOF was
observed.
a
Product was obtained after 4 hrs of photolysis in vacuo, with 1% mol
b
loading of catalyst. Ratio of silane to carbonyl substrate is 3:1. TOF =
TON/time was calculated based on 4hrs of photolysis in vacuo.
Hydrosilylation of Esters and Carbonates. In the
presence of Re(CO)5Cl, esters and carbonates have been
reduced with silanes, although at a lower rate in comparison to
that for the aldehydes. Under our experimental conditions, the
reaction of ethyl acetate and Et3SiH gave ethoxytriethylsilane as
the main product with a trace amount of diethyl ether.
Acetaldehyde has been detected as an intermediate whereby its
signal intensity decreases upon further photolysis while that of
ethoxytriethylsilane continues to increase (Figure 2). Similar
results have been obtained for methyl phenylacetate and methyl
formate.
Further control experiments have been carried out. When the
mixture was heated from 30 to 120 °C in the absence of light,
the yield of the resultant silyl ether was less than 10%. No
hydrosilylation product was observed when the mixture was
photolyzed in air or upon removal of the rhenium catalyst. In
the presence of PPh3 (about 3 equiv of Re(CO)5Cl), the
Re(CO)5Cl photocatalysis was severely inhibited with a
product yield of less than 5%. An FTIR analysis of the reaction
mixture after 4 h of reaction revealed the presence of
Re(CO)4(PPh3)Cl (2018, 2002, and 1946 cm−1).20
3881
dx.doi.org/10.1021/om2012634 | Organometallics 2012, 31, 3880−3887

Organometallics
Article
c
Table 2. Reactions of Acetone and Benzaldehyde with
Various Silanes (Silane:Carbonyl = 3:1)
Table 3. Ratio of Hydrosilylation Products Formed
a
b
Ratio with respect to acetone. Yield calculated on the basis of the
c
integration of the 1H NMR spectra using toluene as standard. Relative
rates were determined by adding 3:1:1 ratio of Et3SiH with various
carbonyl substrates and acetone.
a
TON calculated on the basis of 1% mol loading of Re(CO)5Cl. The
yield of the product was determined on the basis of the integration of
1
the H NMR spectra using toluene as standard.
preferentially reacted with the O−H group to produce the silyl
ester.
The reaction of Et3SiH with diethyl carbonate afforded a
variety of products, with Et3SiOEt (1) and EtOCH2OSiEt3 (2)
being the major ones. A trace amount of ethyl formate,
EtOCHO (3), has been detected. When the reaction mixture
was left overnight, H NMR peaks belonging to (EtO)2CH2
(4) were detected along with a decrease in 1 and 2.
Comparison of Hydrosilylation Rates for Different
Carbonyls. By comparing the ratio of the products within the
same photolysis period, the following general relationship has
been determined for the relative rate using Re(CO)5Cl and
Et3SiH: aliphatic aldehydes > aromatic aldehydes > aliphatic
ketones > aromatic ketones ≈ esters (Table 3). On comparison
within each functional group, the alkyl analogues are more
reactive than their aryl counterparts. Using pyruvic acid which
has both CO and O−H groups, it was found that Et3SiH
Mechanism. The identification of the organic intermediates
and products suggests that the first step of hydrosilylation
proceeds via addition of R3SiH across the carbonyl group,
except when an O−H group is present. The formation of both
H2 and a strong Si−O bond provides much of the driving force
behind the reaction between the alcohol and silane. However,
our focus is on the addition reaction; hence, the complex
reaction pathways for the diethyl carbonate case are highlighted
as an example (Scheme 2).
The products of the reaction between Et31SiH and
diethylcarbonate were observed and identified by H NMR
(Experimental Section) and were attributed to triethylethox-
ysilane (1), ethyl formate (2), ethoxymethoxytriethylsilane (3),
diethoxymethane (4) and hexaethylsiloxane (5). The formation
of these products led us to postulate the following mechanism.
1
1
Figure 2. H NMR intermediate studies on the amount of intermediates and products produced during the reaction of ethyl acetate and Et3SiH.
3882
dx.doi.org/10.1021/om2012634 | Organometallics 2012, 31, 3880−3887

Organometallics
Article
Scheme 2. Proposed Reaction Scheme of the Stepwise
Reaction
Scheme 3. Formation of A from Re(CO)5Cl Photolysis in
Et3SiH
An unstable silylacetal intermediate is first formed, which
undergoes “condensation” to afford 1 and 2. With excess silane,
ethyl formate (2) can be further reduced to 3. This explains the
relatively small amount of formate left in the mixture. In
addition, the experimental data also suggested that compound 1
undergoes further reaction with 2 to generate the ether 4 and
siloxane 5.
As one of the main objectives of this work is to elucidate the
mechanism of hydrosilylation, FTIR spectroscopy has been
utilized for the detection of any metal carbonyl intermediates
present in the catalytic mixture. Upon photolysis of Et3SiH with
Re(CO)5Cl, a dimeric rhenium carbonyl species with a bridging
hydride (henceforth known as complex A) has been identified
in the mixture (Figure 3). The identification of the complex was
based on the IR spectral resemblance to rhenium complexes21
of formula HRe2(CO)9(SiR3). Its bridging hydride signal at
−9.03 ppm has also been recorded in the NMR spectrum. The
diphenylsilyl analogue, complex B, has also been prepared upon
Re(CO)5Cl photolysis in the presence of diphenylsilane. Other
than A, another hydride signal at −5.77 ppm found in the
reaction mixture has been attributed to HRe(CO)5.
The formation of complex A can be rationalized as follows
(Scheme 3). Upon UV photolysis of Re(CO)5Cl, a CO ligand
dissociates to form the 16-electron Re(CO)4Cl intermediate
followed by Et3SiH coordination. Upon reductive elimination,
either HCl or Et3SiCl is formed together with the
corresponding Et3SiRe(CO)4 or HRe(CO)4 species. A free
CO molecule coordinates back to HRe(CO)4 to give
HRe(CO)5. The bridging of its hydride ligand to Et3SiRe(CO)4
would then result in the formation of A.
When A was isolated and tested for aldehyde hydrosilylation,
the silyl ether was generated about 2−3 times faster in
comparison to Re(CO)5Cl. UV irradiation is still essential for
the catalysis to occur. The IR signals of A persisted even after
the end of catalysis, with a recovery of about 70−80%.
Interestingly, complex A can also be generated upon
Re2(CO)10 photolysis in triethylsilane, which would explain
the similarity in the reaction rate to Re(CO)5Cl. From these
observations, there are reasons to believe that complex A acts as
a resting state in carbonyl hydrosilylation. One of the crucial
steps in the catalysis involves the photocleavage of A to form
HRe(CO)5 and the 16-electron rhenium carbonyl Et3SiRe-
(CO)4. As it was shown earlier that catalysis with HRe(CO)5 is
sluggish, we believe that the rhenium silyl species is most likely
the active catalytic species instead.
A mechanism for the hydrosilylation of carbonyl compounds
is proposed to account for the experimental observations
(Scheme 4). Upon photolysis, A dissociates to afford
Et3SiRe(CO)4. Then the carbonyl substrate undergoes
coordination onto the vacant site and facilitates the silyl ligand
shift onto the oxygen atom. This process results in the
formation of an alkyl ligand bound to the Re center (steps 1
1
Figure 3. IR (left) and H hydride NMR (right) spectra of A obtained from photolysis of Re(CO)5Cl with Et3SiH. IR (cm−1): 2102, 2094, 2085,
2030, 2026, 2020, 2007, 2000, 1978; lit.13 for HRe2(CO)9SiCl3 2150, 2095, 2085, 2047, 2019, 2012, 1999, 1978. H NMR (ppm): −9.03.
1
3883
dx.doi.org/10.1021/om2012634 | Organometallics 2012, 31, 3880−3887

Organometallics
Article
Scheme 4. Proposed Mechanism of the Hydrosilylation
Reaction
Scheme 5. Silane Exchange from B to A during the Reaction
(CH2O) and silane (SiH4) have been used as the substrates as
shown in Scheme 6. Solvent effects have not been computed as
well.
a
Scheme 6. Calculated Catalytic Cycle for SiH4 and CH2O
and 2). Another silane undergoes coordination via a η2-silyl
complex4,7 or a σ-silyl (σH) complex (step 3). The H atom
migrates from the silane to the alkyl group, thus regenerating
the catalyst and releasing the silyl ether product (step 4). When
either the carbonyl or silane has been depleted, the R3SiRe-
(CO)4 coordinates back to HRe(CO)5 and becomes part of the
resting state (A).
While the bond formation of steps 1 and 3 would proceed
thermally, we are uncertain whether the catalytic cycle requires
photons to initiate steps 2 and 4. A crude estimate of the
quantum yield for the production of iPrO(SiEt3) was carried
out by calibrating to the quantum yield (0.06 at 366 nm and
0.44 at 313 nm) for CO dissociation from Re(CO)5Cl. Since
our broad-band irradiation covers both wavelengths, we have
assumed an average quantum yield value of 0.25 for the
experiments. By conducting the measurements during the first
a
Values quoted are for free energies in kJ/mol.
1 h of catalysis, a quantum yield of 7.5
4.8 has been
determined using this method (see the Experimental Section).
The value here suggests that the cycle proceeds without light.
However, due to the large measurement error, further
investigation into this catalytic cycle is required.
The activity of A in the catalytic cycle can be tested by using
its Ph2SiH2 analogue, B, to catalyze the hydrosilylation of
Table 4 shows the relevant thermodynamic parameters and
νCO frequencies for the intermediates (restricted to singlet
states only) and transition states found for the cycle, while the
calculated structures of the relevant intermediates and
transition states are found in Figure 4. The calculations showed
that steps I and III in Scheme 6 are exoergic, due to
coordination of substrates to the relevant rhenium complexes.
In contrast, step II is endoergic, as activation is required for the
silyl transfer to the carbonyl ligand in order to form
intermediate II. The transition state TSI for this step has
been found and optimized with an activation energy of 74.3 kJ/
mol relative to intermediate I or 9.6 kJ/mol relative to CH2O
and Re(CO)4SiH3. The reaction coordinate correctly depicts
the migration of the silyl to the O atom of formaldehyde at an
imaginary frequency of 281i cm−1. Step IV turns out to be
exoergic with a transition state TSII at only 26.0 kJ/mol relative
to intermediate 3. The reaction coordinate for this transition
state shows an almost linear migration of an H atom from silane
to the alkyl group at an imaginary frequency of 798i cm−1. The
formation of the strong C−H bond is the driving force behind
the reaction and accounts for the relatively lower activation
barrier. By comparison of the energies, step II represents the
1
acetone and Et3SiH. Upon completion of catalysis, H NMR
analysis showed the presence of iPrO(SiEt3) as the expected
main product with a small amount of iPrO(SiHPh2). More
importantly, the IR spectrum has changed from that of B to A.
These observations show that silyl exchange has occurred
between B and the free Et3SiH during catalysis and lends
support to A participating in the catalysis (Scheme 5).
Calculations. Computational studies have been carried out
to show the thermodynamic feasibility of the catalytic cycle in
Scheme 4 and to provide a general explanation for the relative
rates of hydrosilylation among different carbonyl substrates.
The energetics and structures of the complexes together with
any transition states have been calculated using the mp2/
lanl2dz level of theory and basis set in Gaussian 03.22
Vibrational frequencies were calculated for the optimized
structures. In order to save computational time, formaldehyde
3884
dx.doi.org/10.1021/om2012634 | Organometallics 2012, 31, 3880−3887

Organometallics
Article
Table 4. Calculated Enthalpies (hartree), Free Energies
(hartree), Carbonyl Vibrational and Imaginary Frequencies
(cm−1), and Activation Barriers to TS1 (kJ/mol, Relative to
Re(CO)4SiH3 and the Respective Carbonyl) for the
Rhenium Intermediates and Transition States of the
a
Catalytic Cycle
complex
enthalpy H
free energy G
νCO or i
Ea
Structures for CH2O Substrate
H3SiRe(CO)4
intermediate I
intermediate II
−535.751 262
−649.819 307
−649.791 523
−535.806 047 1767, 1807, 1809,
1908
−649.880 743 1775, 1784, 1818,
1908
−649.852 45
1779, 1792, 1820,
1916
intermediate III −655.891 354
−655.960 371 1767, 1784, 1808,
1921
TSI
−649.783 078
−655.882 012
−649.840 337 1789, 1799, 1817,
1900, 281i
9.6
TSII
−655.950 457 1797, 1820, 1832,
1921, 798i
Transition States for Different Substrates
TSI for
−763.804 788
−727.961 207
−879.689 834
−763.868 593 1781, 1793, 1821,
67.7
24.0
2.0
HCOOCH3
TSI for
(CH3)2CO
TSI for
PhCHO
1907, 291i
−728.024 76
1777, 1793, 1805,
1900, 279i
−879.758 904 1780, 1792, 1809,
1902, 265i
a
The silane used is SiH4, and the method and basis set used is mp2/
lanl2dz.
rate-determining step in the catalytic cycle. The activation
barrier of 74.3 kJ/mol for step II together with the time needed
to generate sufficient Re(CO)4SiR3 could perhaps account for
the period of several hours required for the completion of
catalysis at 300 K.
The relative hydrosilylation rates can be explained by
comparing the activation barriers of the rate-determining step
for the different substrates. We have focused on optimizing the
respective transition states TS 1 of four representative
carbonyls: formaldehyde, benzaldehyde, acetone, and methyl
formate. The value of the activation barrier is calculated relative
to the starting pair in the cycle: i.e., Re(CO)4SiH3 and the
corresponding carbonyl. The various transition states together
with the essential molecular parameters have been depicted in
Figure 5. On comparison of the values of Ea in Table 4, it can
be seen that the barriers for the aldehydes are lower than that
for the ketone, which itself is much lower than that for the
ester. In the ester case, electronic effects due to the OR group
destabilize the transition state much more than the aldehyde or
ketone counterpart. Interestingly, the calculated activation
energy for the benzaldehyde system is the lowest, despite
experimental kinetic studies suggesting a faster reaction for
aliphatic aldehydes. Although the difference is slight (∼7 kJ/
mol), the inclusion of solvent effects may be able to reverse this
trend.
Figure 4. Calculated structures of intermediates and transition states
for Scheme 6. Essential bond lengths (Å) are shown.
IR spectroscopy and is believed to be the resting state for the
catalysis. Upon its photolysis, the active catalyst Re(CO)4SiH3
is released and participates in the catalytic cycle. Computational
studies have shown that the proposed cycle can be carried out
readily at room temperature and are able to explain the relative
hydrosilylation rates among the various carbonyl substrates.
CONCLUSION
■
Re(CO)5Cl has been found to catalyze the hydrosilylation
reaction between various silanes and carbonyl substrates such as
aldehyde, ketone, ester, and carbonate effectively with a
turnover frequency of 20−25 h−1 for aldehydes. The rates of
hydrosilylation are highest for the least sterically hinderd silanes
with aldehydes, followed by aliphatic ketone. Aromatic ketone,
ester, and carbonate react most slowly with the silanes. The
rhenium dimer complex HRe2(CO)9 has been detected using
EXPERIMENTAL SECTION
■
All chemicals were used as purchased. All reactions were carried out
under vacuum conditions unless otherwise stated. Proton nuclear
magnetic resonance (1H NMR) spectra were obtained on a Bruker
Avance 500 (AV500) spectrometer at room temperature. The
3885
dx.doi.org/10.1021/om2012634 | Organometallics 2012, 31, 3880−3887

Organometallics
Article
in a quartz tube. The mixture was photolyzed in vacuo, using a
Legrand broad-band lamp (200−800 nm, 11 W) until all νCO bands of
Re(CO)5Cl disappeared (2−4 h). The yellow solution was
concentrated and passed through a column using a 1:1 mixture of
hexane and chloroform. The resultant yellow band was then collected
and the solvent removed using vacuum and recrystallized in hexane
solution to afford a yellow powder (the yield with respect to
Re(CO)5Cl is 65%). The products were identified on the basis of a
15
1
comparison of their IR and H NMR spectra to reported values.
HRe2(CO)9(SiEt3): νCO (cyclohexane, cm−1) 2030 (m), 2021 (s),
2016 (sh), 2008 (m), 2000 (m), 1982 (sh), 1978 (s); 1H NMR (ppm)
−9.03; UV (nm) 200−400 (broad with λmax 218 nm); ESI mass
spectrum (m/e) 742.
HRe2(CO)9(SiPh2H): νCO (cyclohexane, cm−1) 2029 (m), 2020
1
(s), 2014 (sh), 2008 (m), 2000 (m), 1981 (sh), 1976 (s); H NMR
(ppm) −9.71.
Synthesis of HRe2(CO)9(SiEt3) from Re2(CO)10. Re2(CO)10
(0.011 mmol) and Et3SiH (0.022 mmol) were dissolved in
cyclohexane in a quartz tube. The mixture was photolyzed in vacuo,
using a Legrand broad-band lamp (200−800 nm, 11 W) until all νCO
bands of Re(CO)5Cl disappeared (2−4 h). The yellow solution was
concentrated and passed through a column using a 1:1 mixture of
hexane and chloroform. The resultant yellow band was then collected
and the solvent removed using vacuum and recrystallized in hexane
solution to afford a yellow powder (the yield with respect to
Re(CO)5Cl is 68%).
Quantum Yield Estimation for Re(CO)5Cl Catalysis of Et3SiH
and Acetone. Re(CO)5Cl (0.011 mmol) was added to cyclohexane
(3 mL) in a quartz tube. The mixture was subjected to freeze−pump−
thaw three times before it was broad band irradiated for 1 h. The
dissociated amount of Re(CO)5Cl (=[Re(CO)5Cl]t=0 − [Re-
(CO)5Cl]t=1 h) was noted as x. The concentrations of the complexes
were derived from Beer’s law via the IR absorbance. Re(CO)5Cl
(0.011 mmol), acetone (1.1 mmol), and Et3SiH (1.1 mmol) were
added to cyclohexane (3 mL) in a quartz tube. The mixture was
subjected to freeze−pump−thaw three times before it was broad band
irradiated for 1 h. The concentration of the iPrO(SiEt3) product (from
1H NMR analysis) was noted as y. This series of reactions was
Figure 5. Optimized transition states for the rate-determining steps of
different substrates. Bond lengths are given in Å.
chemical shifts were recorded relative to tetramethylsilane for spectra
taken in CDCl3.
Infrared (IR) spectra were obtained on a Shimadzu IR Prestige-21
Fourier-transform infrared spectrometer (1000−4000 cm−1, 1 cm−1
resolution, 4 scans coadded for spectra averaging) using a 0.05 mm
path length CaF2 cell. Mass spectra of the organic products were
recorded with a Finnigan Mat 95XL-T spectrometer.
Catalysis using Re(CO)5Cl. Re(CO)5Cl (0.011 mmol) was mixed
with Et3SiH (3.3 mmol) and the carbonyl substrate (1.1 mmol) in a
quartz tube. The mixture was photolyzed in vacuo, using a Legrand
borad-band lamp (200−800 nm, 11 W) for 4 h. Products were
repeated four times.
The estimate of the quantum yield for the catalysis is (y/x) × 0.25.
The value of 0.25 represents an “average” quantum yield of
Re(CO)5Cl dissociation from 300 to 400 nm.19 The x value is
proportional to this quantum yield of 0.25. Under the described
experimental conditions above, y = 0.12 0.05 mmol and x = 0.004
0.002 mmol, and hence the quantum yield for catalysis is 7.5 4.8.
Silyl Exchange between HPh2SiRe(CO)4 and Et3SiH. Et3SiH
(0.22 mmol) and acetone (0.11 mmol) were added to a cyclohexane
solution containing HRe2(CO)9(SiR3) (0.011 mmol) in a quartz tube.
The mixture was photolyzed in vacuo, and an IR spectrum of the
reaction was taken at every 30 min interval.
1
identified using H NMR. For one control experiment, PPh3 (0.03
mmol) was added to the reaction mixture as described above.
Catalytic Reduction of Diethyl Carbonate. Re(CO)5Cl (0.011
mmol) was mixed with Et3SiH (3.3 mmol) and diethyl carbonate (1.1
mmol) in a quartz tube. The mixture was photolyzed in vacuo, using a
1
Legrand broad-band lamp (200−800 nm, 11 W) for 4 h. H NMR
(CDCl3, ppm): triethylethoxysilane (1), 0.60 (m), 0.98 (m), 1.20 (m),
3.65 (m); ethyl formate (2), 1.31 (t), 4.24 (q), 8.04 (s);
diethoxymethane (4), 0.60 (m), 0.98 (m), 1.20 (m), 3.65 (m), 4.86
(m).
AUTHOR INFORMATION
Corresponding Author
*Correspondence: chmfanwy@nus.edu.sg.
Synthesis of HRe(CO)5. HRe(CO)5 was synthesized via a
modification of reported methods.23−25 A 0.011 mmol amount of
Re(CO)5Cl was dissolved in methanol, and the mixture was stirred for
1 h, after which the mixture was cooled using liquid nitrogen. Excess
zinc and acetic acid were then added to the cooled mixture before
evacuation of the air inside the round-bottom flask was done. The
reaction mixture was then stirred for 24 h at room temperature. The
HRe(CO)5 formed was extracted using 3 × 3 mL of hexane and
recrystallized using CHCl3. νCO (cyclohexane, cm−1): 2015 (vs), 2005
■
Notes
The authors declare no competing financial interest.
ACKNOWLEDGMENTS
■
The project was supported by a research grant provided by
NUS 143-000-430-112. C.K.T. acknowledges the support of a
NUS scholarship.
1
(m), 1983 (w). H NMR: −5.77 ppm.
Comparison of Rate of Hydrosilylation Reaction between
Different Substrates. Re(CO)5Cl (0.011 mmol) was mixed with
Et3SiH (3.3 mmol) and two different carbonyl substrates (1.1 mmol
each) in a quartz tube. The mixture was photolyzed in vacuo, using a
Legrand broad-band lamp (200−800 nm, 11 W) for 4 h. Products
were identified using 1H NMR, and yields were calculated with toluene
(0.011 mmol) as standard.
REFERENCES
■
(1) Abu-Omar, M. M.; Du, G. Organometallics 2006, 25, 4920−4923.
(2) Morris, R. H. Chem. Soc. Rev. 2009, 38, 2282−2291.
(3) Shaikh, N. S.; Junge, K.; Beller, M. Org. Lett. 2007, 9, 5429−5432.
Synthesis of HRe2(CO)9(SiR3) from Re(CO)5Cl. Re(CO)5Cl
(0.011 mmol) and R3SiH (0.022 mmol) were dissolved in cyclohexane
(4) Fernandes, A. C.; Fernandes, R.; Romao, C. C.; Royo, B. Chem.
̃
Commun. 2005, 213−214.
3886
dx.doi.org/10.1021/om2012634 | Organometallics 2012, 31, 3880−3887

Organometallics
Article
́
(5) Gądek, A.; Szymanska-Buzar, T. Polyhedron 2006, 25, 1441−
1448.
(6) Son, S. U.; Paik, S.; Chung, Y. K. J. Mol. Catal. A: Chem. 2000,
151, 87−90.
(7) Cavanaugh, M. D.; Gregg, B. T.; Cutler, A. R. Organometallics
1996, 15, 2764−2769.
(8) Dong, H.; Berke, H. Adv. Synth. Catal. 2009, 351, 1783−1788.
(9) Ojima, I.; Nihonyanagi, M.; Kogure, T.; Kumagai, M.; Horiuchi,
S.; Nakatsugawa, K. J. Organomet. Chem. 1975, 94, 449−461.
(10) Wile, B. M.; Stradiotto, M. Chem. Commun. 2006, 4104−4106.
(11) Do, Y.; Han, J.; Rhee, Y. H.; Park, J. Adv. Synth. Catal. 2011,
353, 3363−3366.
(12) Chung, L. W.; Lee, H. G.; Lin, Z.; Wu, Y. J. Org. Chem. 2006, 71,
6000−6009.
(13) Du, G.; Fanwick, P. E.; Abu-Omar, M. M. J. Am. Chem. Soc.
2007, 129, 5180−5187.
(14) Sharma, H. K.; Pannell, K. H. Angew. Chem., Int. Ed. 2009, 48,
7052−7054.
(15) Mao, Z.; Gregg, B. T.; Cutler, A. R. J. Am. Chem. Soc. 1995, 117,
10139−10140.
(16) Ojima, Y.; Yamaguchi, K.; Mizuno, N. Adv. Synth. Catal. 2009,
351, 1405−1411.
(17) Gevorgyan, V.; Rubin, M.; Liu, J.; Yamamoto, Y. J. Org. Chem.
2001, 66, 1672−1675.
(18) Hua, R.; Jiang, J. Curr. Org. Synth. 2007, 4, 151−174.
(19) Wrighton, M. S.; Morse, D. L.; Gray, H. B.; Ottessen, D. K. J.
Am. Chem. Soc. 1976, 98, 1111−1119.
(20) Zingales, F.; Sartorelli, U.; Canziani, F.; Raveglia, M. Inorg.
Chem. 1967, 6, 154−157.
(21) Hoyano, J. K.; Graham, W. A. G. Inorg. Chem. 1972, 11, 1265−
1269.
(22) Frisch, M. J.; et al. Gaussian 03, Revisions B and B.05; Gaussian
Inc., Wallingford, CT, 2004.
(23) Braterman, P. S.; Harrill, R. W.; Kaesz, H. D. J. Am. Chem. Soc.
1967, 89, 2851−2855.
(24) Byers, B. H.; Brown, T. L. J. Am. Chem. Soc. 1977, 99, 2527−
2532.
(25) Li, N.; Xie, Y.; King, R. B.; Schaefer, H. F., III J. Phys. Chem. A
2010, 114, 11670−11680.
3887
dx.doi.org/10.1021/om2012634 | Organometallics 2012, 31, 3880−3887