Well-defined NHC-rhodium hydroxide complexes as alkene hydrosilylation and dehydrogenative silylation catalysts

Article abstract of DOI:10.1039/c2dt31339a

Alkene hydrosilylation and dehydrogenative silylation reactions, mediated by [Rh(cod)(NHC)(OH)] complexes (cod = 1,5-cyclooctadiene; NHC = N-heterocyclic carbene) are described. The study details a comparison of the catalytic activity and steric characteristics of four rhodium complexes bearing different NHC ligands. The novel [Rh(cod)(Ii-PrMe)(OH)] complex (Ii-PrMe = 1,3-diisopropyl-4,5-dimethylimidazol-2-ylidine) was designed to improve the reactivity of Rh(i)-hydroxides and proved to be a successful promoter of hydrosilylation and dehydrogenative silylation, displaying good stereo- and regiocontrol. The Royal Society of Chemistry 2013.

Full text of DOI:10.1039/c2dt31339a

Dalton
Transactions
View Article Online
View Journal | View Issue
PAPER
Well-defined NHC-rhodium hydroxide complexes as
alkene hydrosilylation and dehydrogenative silylation
catalysts†
Cite this: Dalton Trans., 2013, 42, 270
Byron J. Truscott, Alexandra M. Z. Slawin and Steven P. Nolan*
Alkene hydrosilylation and dehydrogenative silylation reactions, mediated by [Rh(cod)(NHC)(OH)] com-
plexes (cod = 1,5-cyclooctadiene; NHC = N-heterocyclic carbene) are described. The study details a com-
parison of the catalytic activity and steric characteristics of four rhodium complexes bearing different
NHC ligands. The novel [Rh(cod)(Ii-PrMe)(OH)] complex (Ii-PrMe = 1,3-diisopropyl-4,5-dimethylimidazol-
2-ylidine) was designed to improve the reactivity of Rh(I)-hydroxides and proved to be a successful
promoter of hydrosilylation and dehydrogenative silylation, displaying good stereo- and regiocontrol.
Received 28th September 2012,
Accepted 15th October 2012
DOI: 10.1039/c2dt31339a
www.rsc.org/dalton
successfully employed as hydrosilylation promoters.4 Of par-
ticular interest to us are the carbene complexes, specifically
Introduction
Silicon–carbon bond creation by hydrosilylation of CvC those of the type [Rh(cod)(NHC)X] (X = Cl, Br; NHC = N-hetero-
bonds is a key process in the preparation of organosilicons.1 cyclic carbene) which have shown promising results in the
Hydrosilylation has been catalysed by transition metals since selective hydrosilylation of simple and functionalized alke-
the first use of Speier’s catalyst in the late 1950’s.2 This was fol- nes.1b,c,4,11 The use of ionic liquids and in some cases super-
lowed by an improved system in the form of Karstedt’s Pt(0) critical fluids has boosted activity of Rh–(NHC) and phosphine
vinyl-siloxane system, which was to become the benchmark of functionalized systems,1b,8b,11a,12 enabling hydrosilylation at
hydrosilylation catalysts.1d,3 In spite of its usefulness, Pt-cata- loadings as low as 0.005 mol%, with TON’s of 20 000.12
lyzed hydrosilylation is plagued by competing side-reactions
We have recently shown that Rh(I)-hydroxides stabilized by
such as oligomerization, polymerization, hydrogenation, redis- a NHC moiety are potent catalysts for the conjugate addition
tribution, dehydrogenation and dehydrogenative silylation.1d,4 of arylboronic acids to activated alkenes, where [Rh(cod)(ICy)-
Regio- and stereoselectivity is a major concern, most notably (OH)] (3b) in particular, boasted TONs and TOFs of 100 000
when hydrosilylating alkynes. Hence, there is great interest in and 6600 h−1, respectively.13
the development of highly selective silylation catalysts.1d
Encouraged by these precedents, we have investigated and
Apart from advances made with Pt-catalysts,1d,5 metal com- herein report the use of Rh–(NHC) complexes in alkene hydro-
plexes of iridium, palladium, nickel, ruthenium1d,4,6 and more silylation. In particular, a series of (NHC)-rhodium hydroxides
recently, non-noble metals7 have been employed as efficient were used in this study, demonstrating their utility in silylation
hydrosilylation catalysts. Rhodium complexes have been inves- chemistry.
tigated extensively in the last half century, particularly those of
the general formula [Rh(PR3)3X] (e.g. X = Cl, R = Ph for Wilkin-
son’s catalyst)8 or [Rh(CO)(PR3)2X].6b Variations of these cata-
Results and discussion
lysts include the highly active dimethylglyoxime-containing
Rh(I–III) complexes, active at catalyst loadings as low as
0.004 mol%.9 Rh-dimers containing π-acceptor ligands, e.g.
[Rh2X2Y4] (X = Cl, R, OSiOMe3, and Y = olefin, CO, cod, P(OR)3,
Cp or Cp*)4,6b,10 and [Rh(cod)Cl]2/L (L = PR3) have been
In the conjugate addition study,13 steric variations between
three NHC ligands (IPr (2a), ICy (2b) and IDD (2c), Scheme 1)
were found to be responsible for observed differences in cataly-
tic activity. Complexes 3a–c were synthesized in a one-pot
process from [Rh(cod)Cl]2
(Scheme 1).13
In addition to this series, we designed a novel complex
bearing the small 1,3-diisopropyl-4,5-dimethylimidazol-
2-ylidine (Ii-PrMe) ligand. We hoped to challenge our hypoth-
esis that catalytic reactivity of Rh(I)-hydroxides is inversely
1
as previously reported
EastCHEM School of Chemistry, University of St Andrews, St Andrews, KY16 9ST, UK.
E-mail: snolan@st-andrews.ac.uk
†Electronic supplementary information (ESI) available: Optimization tables and
product characterization. CCDC 867621 (3c) and 867622 (3d). For ESI and crystal-
lographic data in CIF or other electronic format see DOI: 10.1039/c2dt31339a
270 | Dalton Trans., 2013, 42, 270–276
This journal is © The Royal Society of Chemistry 2013

View Article Online
Dalton Transactions
Paper
Scheme 1 Preparation of [Rh(cod)(NHC)(OH)] complexes 3a–d.
Fig. 1 Catalyst activity over 24 h of [Rh(cod)(NHC)(OH)] complexes 3a–d in the
reaction between HSiEt3 and 1-hexene.
Table 1 Reaction of HSiEt3 and 1-hexene in the presence of [Rh(cod)(NHC)-
(OH)] 3a–d at 60 °Ca
Table 2 Reaction of HSiEt3 and 1-hexene using 3a–d at rta
Entry
Catalyst
NHC
Conversionb (%)
6ab (%)
7a (%)
Entry
Catalyst
NHC
Conversionb (%)
6ab (%)
7ab (%)
1
2
3
4
3a
3b
3c
3d
IPr
ICy
IDD
Ii-PrMe
85
94
93
93
86
83
84
86
14
17
16
14
1
2
3
4
3a
3b
3c
3d
IPr
ICy
IDD
Ii-PrMe
42
82
89
>99
72
75
71
93
28
25
29
7
a Reaction conditions: HSiEt3 (4a) (0.3 mmol), 1-hexene (5a)
(0.6 mmol), 3a–d (0.02 mol%), toluene (1 mL), 60 °C, 24 h. b Product
yields calculated from GC-MS (conversion relative to concentration of
HSiEt3) and confirmed by 1H NMR.
a Reaction conditions: HSiEt3 4a (0.3 mmol), 1-hexene 5a (0.6 mmol),
3a–d (1.0 mol%), toluene (1 mL), rt, 24 h. b Product yields calculated
from GC-MS (conversion relative to concentration of HSiEt3) and
confirmed by 1H NMR.
proportional to the steric bulk of the NHC ligand, and provide
a new, highly active complex. Thus, [Rh(cod)(Ii-PrMe)(OH)] 3d
was prepared from 1 and the free carbene 2d, according to the
established procedure (Scheme 1) and fully characterized.†
In order to test the series of Rh-complexes 3a–d in alkene
hydrosilylation, the reaction between triethylsilane (4a) and
1-hexene (5a) (eqn (1)) was selected as the model reaction.
Reaction conditions were optimized (see ESI†); 4a was reacted
with two equivalents of 5a in toluene at 60 °C in the presence
of [Rh(cod)(NHC)(OH)] (0.02 mol%) (3a–d). Reaction progress
was monitored by gas chromatography (GC), following the con-
version of 4a to products. Gas chromatography-mass spec-
trometry (GC-MS) analysis of the reaction mixture, followed by
1H NMR analysis indicated that two products were present,
triethyl(hexyl)silane (6a) and (E)-triethyl(hexenyl)silane (7a)
(eqn (1)). It was clear at this point that competitive dehydro-
genative silylation was leading to the formation of the vinyl-
silane 7a. Selectivity was to become a key issue in these
reactions since dehydrogenative silylation is often promoted
by metals belonging to the iron- or cobalt-triad complexes.4
At 60 °C, all four catalysts 3a–d showed high activity, with
conversions of 85% for 3a and 93–94% for 3b–d, with similar
selectivity for the alkylsilane 6a (Table 1).
In order to obtain a clear comparison between the four
complexes 3a–d the reaction was repeated under milder con-
ditions. Reaction profiles for each entry at room temperature
showed considerable differences in performance (Fig. 1).
Paying particular attention to the first 8 h, activity increased in
the following order: 3a < 3c < 3b < 3d. Complex 3d displayed
the highest catalytic activity, boasting full conversion after 8 h.
Noteworthy is that while Rh-(IDD) 3c was slow to initiate, over
time the reaction progress surpassed that of the cyclohexyl-
complex 3b. The selectivity towards alkylsilane 6a decreased
for all entries at lower temperature (71–75%) except that of
Rh–(Ii-PrMe) 3d which showed only 7% vinylsilane production
(Table 2, entry 4).
In order to rationalize the variance in catalyst activity
between complexes 3a–d, the steric and electronic character-
istics of the NHC ligands concerned (2a–d) were investigated.
To compare the steric demand of each ligand buried volumes
(%Vbur) were calculated for each metal centre (Table 3). The %
Vbur refers to the percentage of a sphere (with a radius r)
that is occupied by atoms of a ligand of interest.14 Single
crystal X-ray structures of complexes 3a–d were obtained
from crystals grown in hexane (3c and 3d shown in Fig. 2
and 3, previously reported for 3a and 3b13). The %Vbur was cal-
culated for each ligand (2a–d) using the SambVca software15
and increases in the following order: 2b < 2d < 2a < 2c
(Table 3).
This journal is © The Royal Society of Chemistry 2013
Dalton Trans., 2013, 42, 270–276 | 271

View Article Online
Paper
Dalton Transactions
increases in the order 2a < 2b–2d < 2c. A comparison of the
bond lengths in the solid state was measured from single crys-
tals and showed that the Rh–NHC bond of Rh–(ICy) 3b
(Table 3, entry 2) is considerably shorter than the other
entries. More importantly, the Rh–OH bond of the Rh–(Ii-
PrMe) complex 3d (Table 3, entry 4) is substantially longer
than that found for the other complexes. From the results
shown in Table 3, it is clear that the four ligands (2a–d)
possess very similar electronic properties. In this vein, Nolan
and co-workers have shown, for a series of cycloalkane-substi-
tuted NHCs, that the size of cycloalkyl N-substituents have very
little effect on the electronics of the ligand and that differences
in performance are more likely to be a result of steric varian-
ces.16a The %Vbur of ICy is similar to that of Ii-PrMe (Table 3,
entries 2 and 4). However, in the reaction between 4a and 5a at
room temperature, 3b showed considerably lower activity than
3d. The difference in Rh–OH bond lengths between 3b and 3d
(Table 3, entries 2 and 4) might indicate a more labile hydroxyl
moiety on the Rh–(Ii-PrMe) complex and hence a better
leaving group. However, it must be noted that these measure-
ments are taken from solid-state crystal structures and do not
necessarily describe the behaviour in solution.
Table 3 Steric and electronic characteristics of NHC ligands on complexes
3a–d with selected bond lengths
a
Entry
NHC
%Vbur
TEP
Rh–NHC (Å)
Rh–OH (Å)
1
2
3
4
IPr 2a
ICy 2b
IDD 2c
Ii-PrMe 2d
35.1
27.1
47.9
28.0
2051.5c
2049.6b
2049.0b
2049.6d
2.046(6)
2.022(3)
2.054(8)
2.044(3)
2.030(4)
2.036(2)
2.024(6)
2.074(3)
a Calculated using SambVca software (sphere radius = 3.5, distance
from centre of sphere
= 2.00, mesh spacing = 0.5, hydrogens
omitted);15 TEP: calculated from νco([Ir(CO)2Cl(NHC)] in DCM.
b Calculations by Nolan et al. 2010.16a c Calculations by Nolan et al.
2008.16b d DFT calculations from [Ni(CO)3(NHC)].16c
Another explanation can be offered by comparing the reac-
tion profiles shown in Fig. 1. Catalysts 3b and 3d show com-
parable activity initially, but over time, the reaction with 3b
slows down and finally plateaus after just six hours. This may
be as a result of decomposition of the complex, leading to cata-
lyst death, whereas 3d appears to be more robust.
Fig. 2 ORTEP representation of [Rh(cod)(IDD)(OH)] (3c), showing 50% thermal
ellipsoid probability. H atoms have been omitted for clarity. Selected bond
lengths (Å) and bond angles (°): Rh(1)–O(1): 2.024 (6), Rh(1)–C(1): 2.054 (8),
C(1)–Rh(1)–O(1): 89.4 (3).†
Rh-(IDD) 3c was slow to initiate hydrosilylation at room
temperature, but demonstrated high activity after 24 h. This
would indicate that high stability and inherent flexibility of the
ligand may be important in enabling a variable steric environ-
ment. Nolan and co-workers have shown that the cyclododecyl
rings are able to flex in solution, providing fluctuating sterics
and enabling lower energy barriers between different confor-
mations.16a It is interesting that the steric data gathered on 3c
in the solid state (Table 3, entry 3) indicates poor potential as
a catalyst. The IDD ligand 2c has a very high %Vbur (47.9%)
and hence, should have provided the least active catalyst.
However, 3c showed great potential as a hydrosilylation cata-
lyst, promoting high yield, particularly at elevated temperature
(93% conversion for reaction between 4a and 5a (Table 1, entry
3); and 94% for 4a with styrene (5b) (see ESI†) at 0.02 mol%,
60 °C). Based on these results and those reported by Nolan,16a
it is evident that the high flexibility of the cyclododecyl rings
means that structural data collected in the solid state does not
necessarily reflect behaviour in solution or at elevated
temperatures.
Fig. 3 ORTEP representation of [Rh(cod)(Ii-PrMe)(OH)] (3d), showing 50%
thermal ellipsoid probability. H atoms have been omitted for clarity. Selected
bond lengths (Å) and bond angles (°): Rh(1)–O(1): 2.074 (3), Rh(1)–C(1):
2.044 (3), C(1)–Rh(1)–O(1): 90.53 (13).†
To provide a comparison of the electronic character of each
ligand, Tolman electronic parameters (TEP) from the literature
have been included in Table 3.16 The TEP method of electronic
quantification employs the fact that electron density from the
ligand is not only passed to the metal but also to the π* orbital
of a CO ligand bound to that metal. Infrared data of Ni, Ir or
[Rh(CO)3L] systems are employed, where a lower stretching fre-
quency of the CO band indicates stronger σ-donation from the
ligand under investigation.14 From Table 3, σ-donation
The hydrosilation of alkenes in the presence of rhodium
has been described by a modified Chalk–Harrod mechanism
(Scheme 2).6b,17 A tertiary silane (ii) oxidatively adds to the
metal (i) to give a silyl-M(III)-hydride (iii). Alkene insertion into
the metal–silyl bond (iv), followed by C–H reductive elimi-
nation (v) then releases the organosilicon product (vi). In the
case of a Rh(I)–OH pre-catalyst, we were unable to isolate any
intermediates and mechanistic investigations are ongoing.
272 | Dalton Trans., 2013, 42, 270–276
This journal is © The Royal Society of Chemistry 2013

View Article Online
Dalton Transactions
Paper
Table 4 Reaction of silanes with terminal alkenes towards selective
hydrosilylationa
3d
(mol%)
T
(°C)
Convb
(%)
6b
(%)
7b,c
(%)
Entry
Product
1
2
3
4
5
6
7
8
1.00
0.02
0.50
0.02
1.00
0.05
1.00
0.02
0.50
0.02
0.50
0.02
1.00
0.02
0.50
0.02
1.00
0.02
1.00
0.02
1.00
0.02
1.00
0.02
1.00
0.05
0.50
0.02
rt
60
rt
60
rt
60
rt
60
rtf
60f
rtf
60
rt
60
rt
60
rt
60
rt
60
rt
60
rt
60
rt
>99
93
95
94
65
91
92
76
>99
>99
>99
99
>99
99
>99
94
97
67
>99
95
>99
38
96
93
86
50
67
32
21
61
76
>99
>99
>99
>99
92
7
14
50
33
68
79
29d
16e
—
—
—
—
8
Scheme 2 Modified Chalk–Harrod mechanism.6b
Having identified 3d as the most efficient catalyst, we inves-
tigated a scope (eqn (2), Table 4) for the hydrosilylation of
terminal alkenes to demonstrate the versatility of the method
and investigate the effect of varying the alkene and the silane
on catalyst activity and selectivity. In each case, reactions were
performed under optimised conditions at room temperature
and at 60 °C with catalyst loadings varying between 0.02 and
1.00 mol% depending on the reaction temperature and the
reactivity of the substrates under investigation. Reactions with
triethylsilane (4a) showed high conversions, particularly with a
loading of 1.0 mol% at room temperature. Reaction with linear
alkenes (Table 4, entries 1 and 7) gave better selectivity
towards the desired alkyl silanes under mild conditions.
Triethoxysilane (4b) was a very efficient hydrosilylation sub-
strate, with completely selective production of alkylsilane in
each case, with the exception of chlorostyrene (Table 4, entries
9–16). Methyldiphenylsilane (4c) also showed high conversions
at rt with 1.00 mol% loading, delivering the hydrosilylated pro-
ducts with pleasing selectivity, particularly when coupled to
the linear alkenes (Table 4, entries 17 and 23). Triphenylsilane
proved to be a relatively unreactive substrate, giving low yield
when coupled with 4a and predominantly the dehydrogenated
product when coupled with styrene (Table 4, entries 25–28).
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
93
7
>99
>99
>99
99
88
60
73
57
>99
>99
97
>99
35
—
—
—
<1
12
40
27
43
—
—
3
64
30
32
>99
>99
60
rt
60
—
65
65
35
a Reaction conditions: HSiR3 4 (0.3 mmol), alkene 5 (0.6 mmol), 3d,
toluene (1 mL), 24 h (unless stated otherwise). b Product yields
calculated from GC-MS (conversion relative to concentration of HSiR3)
and confirmed by 1H NMR. c Stereochemistry determined as (E) for
each entry by 1H NMR unless stated otherwise. Reaction also produces
2-(triethylsilyl)-1-octene. d 10%. e 7%. f Reaction time = 18 h.
(NHC)X] have shown results similar to 3d at catalyst loadings
of 0.02 mol%, with the aid of ionic liquids.1b Other highly
active systems include the family of Rh-complexes bearing
phosphine functionalized imidazolium ligands reported by Lai
and co-workers. These catalysts boast full conversions at cata-
lyst loadings as low as 0.005 mol%, with the assistance of ionic
liquids.11a,12 Karstedt’s Pt-catalyst and the dicarbonyl Rh(I)
compound tested by Ganicz are highly active at loadings as low
as 0.004 mol%, albeit under harsher conditions.9
ð2Þ
Intrigued by the competition between hydrosilylation and
dehydrogenative silylation, we investigated the selective prep-
aration of vinylsilanes from alkenes using 3d, under similar
conditions. The ability to prepare vinylsilanes directly from
alkenes is a valuable process, particularly if stereocontrol is
achieved, since hydrosilylation of alkynes is often
accompanied by side reactions and can lead to isomers of the
olefinic product.18 Complexes of Fe, Ni, Co, Pd and Rh have
provided active dehydrogenative silylation catalysts, particu-
larly where alkenes with strongly electronegative substituents
are used, such as styrenes.17d The product determining step in
the reaction mechanism is a competitive β-H transfer from
either the σ-alkyl or σ-silylalkyl fragments of intermediate (viii)
(Scheme 3). In the case of dehydrogenative silylation, β-H
Clearly, there is strong competition between hydrosilylation
and dehydrogenative silylation in the reactions described
(Table 4).
The selectivity is fairly substrate specific. The alkylsilanes
are more susceptible to dehydrogenative silylation than the
silyl ether (4b) and the conjugation of the styrene systems (5b
and 5c) make them good substrates for dehydrogenative silyl-
ation. Selectivity towards dehydrogenative silylation ranges
from almost non-existent when HSi(OEt)3 is used (Table 4,
entries 9–16) to being the dominant mechanism when styrenes
are coupled to triethyl- or triphenylsilane (Table 4, entries 5, 6
and 27, 28).
The reactivity shown by 3d is comparable to some of the
most active catalysts available. Rh-systems of the type [Rh(cod)-
This journal is © The Royal Society of Chemistry 2013
Dalton Trans., 2013, 42, 270–276 | 273

View Article Online
Paper
Dalton Transactions
dehydrogenative silylation towards vinylsilane as the major
product (Table 6, entries 1–4, 10, 11, 13 and 14), with five of
the reactions showing a complete shift in selectivity (Table 6,
entries 1, 2, 4, 10 and 11). HSi(OEt)3 showed only mild suscep-
tibility to dehydrogenative silylation in the presence of a ten-
fold excess of 5c (Table 6, entry 7) and no dehydrogenation for
the other entries (Table 6, entries 5, 6 and 8).
Scheme 3 Competitive β-H transfer leading to either alkyl- or vinylsilanes.17d
Table 5 Reaction of HSiEt3 with varied concentrations of styrenea
1H NMR analyses of products indicated the exclusive for-
mation of the (E)-vinylsilane products (undetermined for 7l,
Table 6, entry 15). The results presented are comparable to
those obtained with complexes of Pd and Re, both which have
been used at 1.0 mol% loadings, although with better regio-
selectivity in some cases.17d,19 In addition, phosphine-amido
and -amine Rh-chelates have shown high activity and reason-
able selectivity at catalyst loadings as low as 0.1 mol%.20
Entry
5bb
Conv.c (%)
6bc (%)
7bc (%)
1
2
3
4
1
4
7
10
93
60
19
11
5
40
82
89
95
>99
>99
>99
a Reaction conditions: 4a (0.3 mmol), 5b. b (molar equivalence relative
to 4a), 3b 1.0 mol%, toluene (1 mL), 60 °C, 2.5 h. c Conversion and
product ratio calculated by GC-MS and confirmed by 1H NMR.
Conclusions
Table 6 Reaction of silanes with terminal alkenes towards selective dehydro- In summary, we have investigated the catalytic potential of a
genative silylationa
series of [Rh(cod)(NHC)(OH)] complexes, including the novel
Ii-PrMe-complex 3d. The four complexes proved to be efficient
Entry
Product
Conv.b (%)
6b (%)
7
b,c (%)
catalysts for the hydrosilylation of terminal alkenes. An investi-
gation into the steric nature of the NHC ligands used (2a–d)
ensued. The structural data proved useful in identifying reac-
tivity trends and the catalytic potential of these complexes.
Rh–(Ii-PrMe) 3d was further tested in hydrosilylation reactions
involving silanes and terminal alkenes, proving to be a versa-
tile catalyst in this regard, achieving full conversions at catalyst
loadings as low as 0.02 mol% at 60 °C. Regioselectivity proved
to be a key issue, with dehydrogenative silylation competing
with hydrosilylation to give vinylsilane products. We were able
to exploit this competitive side reaction, preparing vinyl-
silanes, by dehydrogenative silylation in the presence of 3d
(0.05 mol%). The dehydrogenative silylation showed good
regio- and stereocontrol at low catalyst loading. This study has
extended the scope of activity and provided insight into reactiv-
ity trends displayed by Rh(I)-hydroxides. These results will be
paramount to our understanding of Rh(I)-hydroxides and
enable further pursuit of highly active organometallic species
for homogeneous catalysis.
1
2
99
43
5
57
95
92
60d
—
—
35
—
15
62
>99
>99
92
3
8
4
35
>99
>99
65
>99
85
38
5
>99
>99
>99
>99
>99
96
6
7
8
9
10
11
12
13
14
93
38
83
12
8
62
17
88
92e
>99
99
>99
a Reaction conditions: HSiR3 4 (0.3 mmol), alkene 5 (3.0 mmol), 3d
(0.05 mol%), toluene (1 mL), 60 °C, 24 h. b Product yields calculated
from GC-MS (conversion relative to concentration of HSiR3) and
confirmed by 1H NMR analysis. c Stereochemistry determined by 1H
NMR. d Reaction also produces 8a (5%). e Stereochemistry unclear.
Experimental
General considerations
transfer from the silylalkyl fragment will lead to a dihydride
complex, which will protonate a second alkene fragment (ii),
giving the alkane (xi). Hence, dehydrogenative silylation is
favoured by a high alkene : silane ratio.17d
During reaction optimization it was evident that increasing
the alkene concentration did in fact drive the dehydrogenative
process. In particular, when the styrene concentration was
increased in the reaction with 4a, the selectivity could be
increased from 40% (Table 5, entry 1) to 95% (Table 5, entry 4).
Under similar conditions as before, silanes 4a–d were
reacted with a ten-fold excess of the alkene 5a–d at 60 °C
(Table 6). In eight of the fourteen entries, we observed selective
Synthesis of Rh-complexes was performed inside an MBraun
Glovebox under inert conditions. All reagents were supplied by
Aldrich and used without further purification and solvents
were distilled and dried as required. NMR data was obtained
using either a Bruker 400 MHz or 300 MHz spectrometer in
the specified deuterated solvent. All chemical shifts are given
in ppm and coupling constants in Hz. Spectra were referenced
to residual protonated solvent signals (C6D6: 1H δ 7.16 ppm,
13C δ 128.0 ppm; CDCl3: 1H δ 7.26 ppm). Elemental analyses
were performed at the London Metropolitan University. Gas
274 | Dalton Trans., 2013, 42, 270–276
This journal is © The Royal Society of Chemistry 2013

View Article Online
Dalton Transactions
Paper
chromatography (GC) analyses were performed on an Agilent hexane as eluent. Stereochemistry of vinylsilanes was deter-
1
1
7890A apparatus equipped with a flame ionization detector mined by H NMR. H NMR and MS were compared to litera-
and a (5%-phenyl)-methylpolysiloxane column (30 m, 320 μm, ture values (reported in ESI.†).
film: 0.25 μm). Flow rate 1 mL min−1 constant flow, inlet temp-
erature 260 °C, column temperature 50 °C, 20 °C min−1
increase to 70 °C (held for 1 min), 50 °C min−1 increase to
300 °C (held for 1 min), total time 7.6 min. GC-MS measure-
Acknowledgements
ments were performed on a Trace GC Ultra gas chromatograph The ERC (FUNCAT to SPN) and Sasol (BJT) are gratefully
equipped with a DSQ II mass selective detector and a RXI®- acknowledged for support. Umicore AG is thanked for their
1ms column (30 m, 0.25 mm, film: 0.25 μm). [Rh(cod)(NHC) generous gift of materials.
(OH)] complexes 3a–d were prepared in the manner previously
reported.13
[Rh(cod)(Ii-PrMe)(OH)] (3d). The reported procedure was fol-
Notes and references
lowed using [Rh(cod)Cl]2 1, free Ii-PrMe 2d and CsOH in THF
1
to give 3d (80%) as a yellow solid; H NMR (300 MHz, C6D6): δ
6.46 (quint, J = 7.1, 2H), 5.32–5.18 (m, 2H), 3.16–3.05 (m, 2H),
2.65–2.36 (m, 4H), 2.14–1.99 (m, 2H), 1.96–1.80 (m, 2H), 1.60
1 (a) E. de Wolf, E. A. Speets, B.-J. Deelman and G. van
Koten, Organometallics, 2001, 20, 3686–3690; (b) J. Li,
J. Peng, Y. Bai, G. Lai and X. Li, J. Organomet. Chem., 2011,
696, 2116–2121; (c) N. Imlinger, K. Wurst and
M. R. Buchmeiser, Monatsh. Chem., 2004, 136, 47–57;
(d) D. Troegel and J. Stohrer, Coord. Chem. Rev., 2011, 255,
1440–1459.
2 (a) J. L. Speier, J. A. Webster and G. H. Barnes, J. Am. Chem.
Soc., 1957, 79, 974–979; (b) J. L. Speier and D. E. Hook,
United States Pat., US2823218, 1958.
(s, 6H), 1.32 (d, J = 7.0, 6H), 1.29 (d, J = 7.0, 6H); 13C NMR
1
(75 MHz, C6D6): δ 184.9 (d, 1JRhC = 56.3), 124.6, 95.9 (d, JRhC
=
8.3), 71.7 (d, 1JRhC = 14.3), 53.8, 34.0, 29.2, 22.4, 22.3, 9.8; Anal.
Calcd for C19H33N2ORh (MW 408.38): C, 55.88, H, 8.14, N,
6.86. Found: C, 55.76; H, 8.24; N, 6.88.
X-ray diffraction studies
Single crystal diffraction data were collected on a Rigaku
Saturn (3c) and Mercury (3d) CCD diffractometer, 007 rotating
anode generator (graphite monochromated Mo Kα radiation,
λ = 0.71073 Å) with the aid of the CrystalClear program. An
empirical (multi-scan) absorption correction was applied at
the data reduction stage and the structure was solved by direct
methods with SHELXS-97.
CRYSTAL DATA FOR [Rh(cod)(IDD)(OH)] (3c). Colourless platelets
were obtained by slow evaporation of a concentrated solution
of hexane under an inert atmosphere. C35H61N2ORh, M =
628.77, crystal dimensions: 0.20 × 0.10 × 0.01 mm, orthorhom-
bic, a = 29.075 (7), b = 15.582 (3), c = 14.620 (3) Å, V = 6623(3)
Å, T = 93 K, space group Pbca, Z = 8, 63 525 measured reflec-
tions, 6075 independent reflections, Rint = 0.0722, final R1
values [I > 2σ(I)] = 0.1109, final wR2 = 0.2333.
3 B. D. Karstedt, France Pat., FR1548775, 1968.
4 B. Marciniec, in Advances in Silicon Science, ed. B. Marciniec,
Springer Science+Business Media, Australia, 1st edn, 2009,
vol. 1, pp. 3–51.
5 (a) I. E. Markó, S. Stérin, O. Buisine, G. Mignani,
P. Branlard, B. Tinant and J. P. Declercq, Science, 2002, 298,
204–206; (b) G. Berthon-Gelloz, O. Buisine, J.-F. Brière,
G. Michaud, S. Stérin, G. Mignani, B. Tinant, J.-P. Declercq,
D. Chapon and I. E. Markó, J. Organomet. Chem., 2005, 690,
6156–6168; (c) I. E. Markó, S. Stérin, O. Buisine,
G. Berthon, G. Michaud, B. Tinant and J.-P. Declercq, Adv.
Synth. Catal., 2004, 346, 1429–1434; (d) J. J. Hu, F. Li and
T. S. A. Hor, Organometallics, 2009, 28, 1212–1220;
(e) J. J. Dunsford, K. J. Cavell and B. Kariuki, J. Organomet.
Chem., 2011, 696, 188–194; (f) M. A. Taige, S. Ahrens and
T. Strassner, J. Organomet. Chem., 2011, 696, 2918–2927;
(g) J. W. Sprengers, M. J. Mars, M. A. Duin, K. J. Cavell and
C. J. Elsevier, J. Organomet. Chem., 2003, 679, 149–152.
6 (a) T. Hayashi, in Comprehensive Asymmetric Catalysis, ed.
E. N. Jacobson, A. Pfaltz and H. Yamamoto, Springer-
Verlag, New York, 4th edn, 1999, vol. 1, pp. 319–333;
(b) B. Marciniec, Silicon Chem., 2002, 1, 155–175.
CRYSTAL DATA FOR [Rh(cod)(Ii-PrMe)(OH)] (3d). Yellow prisms
were obtained by slow evaporation of a concentrated solution
of hexane under an inert atmosphere. C19H33N2ORh, M =
408.38, crystal dimensions: 0.10 × 0.10 × 0.02 mm, monoclinic,
a = 14.999(5), b = 9.986(3), c = 13.137(4) Å, β = 106.884 (8) , V =
1882.8 (10) Å, T = 93 K, space group P2(1)/c, Z = 4, 11
414 measured reflections, 3417 independent reflections, Rint
0.0722, final R1 values [I > 2σ(I)] = 0.0422, final wR2 = 0.1164.
=
7 A. M. Tondreau, C. C. H. Atienza, K. J. Weller, S. A. Nye,
K. M. Lewis, J. G. P. Delis and P. J. Chirik, Science, 2012,
335, 567–570.
General method for catalytic hydrosilylation
[Rh(cod)(NHC)(OH)] (4a–d) was dissolved in toluene (1 mL)
and removed from the glove-box. HSiR3 4a–d and alkene 5a–d
were added and the reaction mixture was stirred at the stated
temperature for 24 h. Conversion to products was determined
8 (a) I. Ojima, M. Nihonyanagi, T. Kogure, M. Kumagai,
S. Horiuchi, K. Nakatsugawa and Y. Nagai, J. Organomet.
Chem., 1975, 94, 449–461; (b) J. Li, J. Peng, G. Zhang, Y. Bai,
G. Lai and X. Li, New J. Chem., 2010, 34, 1330.
by GC and GC-MS (conversion relative to concentration of
HSiR3) and confirmed by H NMR after separation from start- 10 B. Marciniec, P. Krzyżanowski, E. Walczuk-Gusciora and
9 T. Ganicz, Appl. Catal., A, 2004, 259, 49–55.
1
ing materials by flash chromatography on silica gel with
W. Duczmal, J. Mol. Catal. A: Chem., 1999, 144, 263–271.
This journal is © The Royal Society of Chemistry 2013
Dalton Trans., 2013, 42, 270–276 | 275

View Article Online
Paper
Dalton Transactions
11 (a) J. Li, J. Peng, D. Wang, Y. Bai, J. Jiang and G. Lai,
J. Organomet. Chem., 2011, 696, 263–268; (b) K. H. Park,
Organometallics, 2008, 27, 202–210; (c) D. G. Gusev, Organo-
metallics, 2009, 28, 6458–6461.
S. Y. Kim, S. U. Son and Y. K. Chung, Eur. J. Org. Chem., 2003, 17 (a) B. Marciniec, W. Duczmal, W. Urbaniak and
2003, 4341–4345; (c) J. E. Hill and T. A. Nile, J. Organomet.
Chem., 1977, 137, 293–300; (d) S. Díez-González, N. Marion
andS. P. Nolan, Chem. Rev., 2009, 109, 3612–3676.
12 J. Li, J. Peng, Y. Bai, G. Zhang, G. Lai and X. Li, J. Organo-
met. Chem., 2010, 695, 431–436.
13 B. J. Truscott, G. C. Fortman, A. M. Z. Slawin and
S. P. Nolan, Org. Biomol. Chem., 2011, 9, 7038–7041.
14 T. Droge and F. Glorius, Angew. Chem., Int. Ed., 2010, 49,
6940–6952.
E. Śliwińska, J. Organomet. Chem., 1990, 385, 319–327;
(b) S. B. Duckett and R. N. perutz, Organometallics, 1992,
11, 90–98; (c) S. Sakaki, M. Sumimoto, M. Fukuhara,
M. Sugimoto, H. Fujimoto and S. Matsuzaki, Organometal-
lics, 2002, 21, 3788–3802; (d) B. Marciniec, Coord. Chem.
Rev., 2005, 249, 2374–2390; (e) P. B. Glaser and
T. D. Tilley, J. Am. Chem. Soc., 2003, 125, 13640–13641;
(f) S. Park and M. Brookhart, Organometallics, 2010, 29,
6057–6064.
15 A. Poater, B. Cosenza, A. Correa, S. Giudice, F. Ragone, 18 (a) R. Takeuchi and N. Tanouchi, Chem. Commun., 1993,
V. Scarano and L. Cavallo, Eur. J. Inorg. Chem., 2009, 1759–
1766.
1319–1320; (b) M. L. Christ, S. Sabo-Etienne and
B. Chaudret, Organometallics, 1995, 14, 1082–1084.
16 (a) G. C. Fortman, A. M. Slawin and S. P. Nolan, Dalton 19 Y. Jiang, O. Blacque, T. Fox, C. M. Frech and H. Berke,
Trans., 2010, 39, 3923–3930; (b) R. A. Kelly III, H. Clavier, Chem.–Eur. J., 2009, 15, 2121–2128.
S. Giudice, N. M. Scott, E. D. Stevens, J. Bordner, 20 L. J. Hounjet, R. McDonald, M. J. Ferguson and M. Cowie,
I. Samardjiev, C. D. Hoff, L. Cavallo and S. P. Nolan,
Inorg. Chem., 2011, 50, 5361–5378.
276 | Dalton Trans., 2013, 42, 270–276
This journal is © The Royal Society of Chemistry 2013