Phenylsilane dehydrocoupling and addition to styrene catalyzed by (R-indenyl)Ni(phosphine)(methyl) complexes

Article abstract of DOI:10.1139/V08-132

This report describes the synthesis and characterization of the Ni-Me complexes (R-indenyl)Ni(PR′³)Me (R = 1-i-Pr, 1-SiMe ₃, and 1,3-(SiMe₃)₂; R′= Me, Ph) and outlines their catalytic reactivities in the dehydrogenative oligomerization of PhSiH₃ and its addition to styrene in the absence of initiators/activators. Observation of higher hydrosilylation activities for PPh₃-based compounds featuring bulky substituents on the indenyl ligand confirms earlier suggestions that phosphine dissociation is an important component of the catalytic cycle for this reaction. In contrast, oligomerization of PhSiH₃ is more facile with PMe₃-based precursors and independent of the steric bulk of the indenyl ligand, implying that this reaction does not involve phosphine dissociation. These conclusions are consistent with the variable-temperature ¹H NMR spectra of {1,3,-(SiMe₃)₂-indenyl}Ni(PR′₃)Me and various structural parameters observed in the solid-state structures of {1,3,-(SiMe₃)₂-indenyl}Ni(PPh₃)Me, {1,3,-(SiMe₃)₂-indenyl}Ni(PMe₃)Cl, and {1-SiMe₃-indenyl}Ni(PMe₃)Me.

Full text of DOI:10.1139/V08-132

280
Phenylsilane dehydrocoupling and addition to
styrene catalyzed by (R-indenyl)Ni(phosphine)(methyl)
complexes1
Yaofeng Chen and Davit Zargarian
Abstract: This report describes the synthesis and characterization of the Ni–Me complexes (R-indenyl)Ni(PR′3)Me (R =
1-i-Pr, 1-SiMe3, and 1,3-(SiMe3)2; R′= Me, Ph) and outlines their catalytic reactivities in the dehydrogenative
oligomerization of PhSiH3 and its addition to styrene in the absence of initiators/activators. Observation of higher
hydrosilylation activities for PPh3-based compounds featuring bulky substituents on the indenyl ligand confirms earlier
suggestions that phosphine dissociation is an important component of the catalytic cycle for this reaction. In contrast,
oligomerization of PhSiH3 is more facile with PMe3-based precursors and independent of the steric bulk of the indenyl
ligand, implying that this reaction does not involve phosphine dissociation. These conclusions are consistent with the
1
variable-temperature H NMR spectra of {1,3,-(SiMe3)2-indenyl}Ni(PR′3)Me and various structural parameters observed
in the solid-state structures of {1,3,-(SiMe3)2-indenyl}Ni(PPh3)Me, {1,3,-(SiMe3)2-indenyl}Ni(PMe3)Cl, and {1-SiMe3-
indenyl}Ni(PMe3)Me.
Key words: nickel-indenyl complexes, hydrosilylation, hydrosilane oligomerization.
Résumé : Ce rapport présente la synthèse et la caractérisation des complexes (R-indényle)Ni(PR′3)Me (R = 1-i-Pr, 1-
SiMe3 et 1,3-(SiMe3)2; R′= Me et Ph) ainsi que la réactivité catalytique de ces composés dans l’oligomérisation déshy-
drogénative du PhSiH3 et dans son addition au styrène en absence d’un initiateur ou d’un activateur. L’observation des
plus grandes activités d’hydrosilylation chez les composés à base de PPh3 et portant des substituants volumineux sur le
ligand indényle indique que le cycle catalytique implique la dissociation de la phosphine. D’ailleurs, l’oligomérisation
du PhSiH3 est plus facile avec les composés à base de la PMe3 et indépendante de la taille des substituents de l’indényle,
ce qui semble indiquer que cette réaction n’implique pas la dissociation de la phosphine. Ces conclusions sont concor-
1
dantes avec d’autres observations tirées des spectres RMN H à températures variables des composés {1,3,-(SiMe3)2-
indényle}Ni(PR′3)Me et des études des structures des composés {1,3,-(SiMe3)2-indényle}Ni(PPh3)Me, {1,3,-(SiMe3)2-
indényle}Ni(PMe3)Cl et {1-SiMe3-indényle}Ni(PMe3)Me.
Mots-clés : composés nickel-indényles, hydrosilylation, oligomérisation des hydrosilanes.
[Traduit par la Rédaction] Chen and Zargarian 287
Introduction
aluminoxanes. The in situ activation of these precursors can
also be done using cationic initiators, such as AgBF4, NaBPh4,
or AlCl3, all of which generate the electronically unsaturated
and highly electrophilic cations [IndNi(PR3)]+ (5); these in-
termediates promote the oligomerization of some olefins (6)
as well as the hydrosilylation of olefins and ketones (7).
Cationic species can also be generated in the absence of ini-
tiators by using the precursors (i-Pr-Ind)Ni(PPh3)(OSO2CF3)
Recent reports have shown that the complexes IndNi(L)X
(Ind= indenyl and its substituted derivatives; L= neutral lig-
ands such as phosphines or N-heterocyclic carbenes; X= an-
ionic ligands such as halides, alkyls, triflate, and so forth)
(1) promote the polymerization of ethylene (2), alkynes (3),
or PhSiH3 (4) in the presence of activators such as methyl-
5
1
(8) or [(η ,η -Ind^NR2)Ni(PR3)]+ (^ = various side chains
tethering the amine moiety to the Ind ligand) (9), or by heat-
ing the complexes [IndNi(PR3)2]+ bearing bulky phosphine
ligands (10).
Received 1 May 2008. Accepted 7 July 2008. Published on
the NRC Research Press Web site at canjchem.nrc.ca on
24 October 2008.
This paper is dedicated to Professor Richard Puddephatt for
his great contributions in Organometallic Chemistry.
A number of recent studies have shown that of the reactions
promoted by IndNi(PR3)X, only those involving hydro-
silanes can proceed in the absence of activators/initiators.
Thus, the Ni–Me complexes (1-Me-Ind)Ni(PR3)Me (R = Me,
Ph, Cy) act as single-component precursors for the dehydro-
genative oligomerization of PhSiH3 (11), whereas the Ni–Cl
complexes (R-Ind)Ni(PPh3)Cl promote the addition of
PhSiH3 to styrene in the absence of any activator/initiator
(12). The available mechanistic information indicates that
these reactions are sensitive to the steric bulk and (or)
Y. Chen2 and D. Zargarian.3 Département de Chimie,
Université de Montréal, Montréal, QC H3C 3J7, Canada.
1This article is part of a Special Issue dedicated to Professor
R. Puddephatt.
2Present address: Shanghai Institute of Organic Chemistry,
Chinese Academy of Science, 354 Fenglin Road, Shanghai
200032, P.R. China.
3Corresponding author (e-mail: zargarian.davit@umontreal.ca).
Can. J. Chem. 87: 280–287 (2009)
doi:10.1139/V08-132
© 2008 NRC Canada

Chen and Zargarian
281
Scheme 1.
sired product (14). Removal of the PMe3 released during the
synthesis by periodic pumping of the reaction mixture helps
minimize this decomposition pathway, but does not elimi-
nate it altogether. In the case of 3, the presence of small
quantities of Ni(PMe3)4 in the reaction mixture complicated
the purification, since both species appear to be highly solu-
ble in most solvents, including hexane. However, solution
NMR spectra and X-ray analysis of crystals obtained by re-
moval of the oily impurities helped establish the identity of
this compound unambiguously (vide infra).
The difficult purification of 3 encouraged us to seek an al-
ternative pathway to the desired Ni–Me compounds. We
found that the direct reaction of Li[R-Ind] with
(PMe3)2NiCl(Me) gave reasonable yields of (R-
Ind)Ni(PMe3)Me (Scheme 1; R = 1-SiMe3 (4, 87%), 1,3-
(SiMe3)2-Ind (5, 64%), i-Pr (6, 77%)). The purification of 5
was also challenging, but 4 and 6 were obtained in good pu-
rity. The characterization of compounds 1–6 by solution
NMR spectra and, in the case of 2–4, by X-ray diffraction
studies was straightforward, as described below.
NMR characterization
All of the complexes showed a single resonance in their
31P{1H} NMR spectra at chemical shifts that are characteris-
tic of each type of complex. Thus, 1 and 2 gave singlets at
~47 and 44 ppm, respectively (cf. 48 ppm for (1-Me-
Ind)Ni(PPh3)Me) (15), 3 gave a singlet at ~ –18 ppm
(cf. –11 ppm for (1-Me-Ind)Ni(PMe3)Cl) (14), and 4–6 gave
singlets at ~ –9 (5) and –4 ppm (4 and 6) (cf. –3.7 ppm for
(1-Me-Ind)Ni(PMe3)Me) (11).
nucleophilicity of the phosphine ligand, the rates following
the order PMe3 > PPh3 > PCy3 for silane oligomerization
(11) and PPh3 > PMe3 for the hydrosilylation (7). The latter
reaction is also favoured by compounds featuring bulkier
indenyl ligands (1,3-(SiMe3)2-Ind > 1-SiMe3-Ind > 1-Me-
Ind) (13). These and a number of other observations led us
to propose that the hydrosilylation requires the dissociation
of the phosphine ligand (12).
As a continuation of our studies on the structure–activity
relationships in Ni–indenyl complexes, we have prepared a
new series of IndNi(PR3)Me complexes and studied their
structures and reactivities in the oligomerization of PhSiH3
and its addition to styrene as a function of Ind substituents
and the phosphine ligands. This report presents the synthe-
sis, characterization, and reactivity studies for the complexes
1
The H NMR spectra showed a number of characteristic
signals for the Ni–Me moiety (~ –0.7 ppm), the SiMe3 sub-
stituents (~0.3–0.5 ppm), the PMe3 ligand (~0.6–0.8 ppm),
and H2 of the Ind ligands (~6.1–6.7 ppm). It is interesting to
note that methylation of the Ni–Cl bond shifts δ H2 upfield
in {1,3-(SiMe3)2-Ind}Ni(PMe3)X (from 6.72 to 6.13 ppm),
{1,3-(SiMe3)2-Ind}Ni(PPh3)X (from ~7 to 6.33 ppm), and
(1-SiMe3-Ind)Ni(PPh3)X (from 6.65 to 6.42 ppm). This up-
field shift likely arises from the greater delocalization in-
duced in the five-membered ring moiety of the Ind ligands in
Ni–alkyl vs. Ni–halide derivatives (1, 15). Finally, the phenyl
rings of the PPh3 ligand exert some shielding influence on
H3 of the Ind ligand that is evident from a comparison of the
δH3 in 1 (4.23 ppm) vs. 4 (4.62 ppm) and 6 (4.44 ppm).
(1-SiMe3-Ind)Ni(PPh3)Me
(1),
{1,3-(SiMe3)2-Ind}Ni-
(PPh3)Me (2), {1,3-(SiMe3)2-Ind}Ni(PMe3)Cl (3), (1-SiMe3-
Ind)Ni(PMe3)Me (4), {1,3-(SiMe3)2-Ind}Ni(PMe3)Me (5),
and {1-i-Pr-Ind}Ni(PMe3)Me (6), including the X-ray dif-
fraction studies for 2–4.
1
We studied the variable-temperature H NMR spectra of 2
and 5 (Fig. 1) and compared them to those of {1,3-(SiMe3)2-
Ind}Ni(PPh3)Cl to probe the issue of phosphine lability in
the Ni–Me derivatives. It is worthwhile to recall that the
low-temperature spectra of {1,3-(SiMe3)2-Ind}Ni(PPh3)Cl
showed a gradual sharpening of the 31P{1H} signal as well
Results and discussion
Syntheses
Reaction of (R-Ind)Ni(PPh3)Cl with MeMgCl gave the
corresponding Ni–Me complexes in moderate yields
(Scheme 1; R = 1-SiMe3 (1, 73%), 1,3-(SiMe3)2-Ind (2,
66%)), but this approach did not work well for the prepara-
tion of the analogous PMe3 complexes because of the diffi-
cult purification of the analogous Ni–Cl precursors. For
instance, reacting equimolar quantities of Li[1,3-(SiMe3)2-
Ind] and Ni(PMe3)2Cl2 gave the desired {1,3-(SiMe3)2-
Ind}Ni(PMe3)Cl (3) in ~50% crude yield. As noted previ-
ously, however, the main challenge in the synthesis of Ni–
indenyl compounds bearing PMe3 is avoiding a side reaction
that generates Ni(PMe3)4 and reducing the yield of the de-
1
as the splitting of the H singlet due to the Si(CH3)3 protons
(Fig. 1a). The relatively small ∆G‡ value of ~10 kcal/mol
(1 cal = 4.184 J) calculated for this exchange process, as
well as a number of parameters from the solid-state structure
of {1,3-(SiMe3)2-Ind}Ni(PPh3)Cl, convinced us that the dy-
namic exchange observed for this compound was caused by
a rapid dissociation/re-association of the PPh3 ligand as op-
posed to the hindered rotation of the 1,3-(SiMe3)2-Ind ligand
(12). The lability of the PPh3 ligand was ascribed to the sig-
nificant steric interaction with the SiMe3 substituents.
© 2008 NRC Canada

282
Can. J. Chem. Vol. 87, 2009
1
Fig. 1. The variable-temperature H NMR spectra (toluene-d8,
400 MHz) for {1,3-(SiMe3)2Ind}Ni(PPh3)Cl (a), {1,3-(SiMe3)2Ind}-
Ni(PPh3)Me (2, b), and {1,3-(SiMe3)2Ind}Ni(PMe3)Me (5, c).
down to –70 °C. These observations can be understood
when we consider that methylation of the Ni–Cl bond is an-
ticipated to reinforce the Ni–P bond strength and increase
the delocalization of bonding in the five-membered-ring
moiety of the Ind ligands; these changes would, in turn, lead
to less facile phosphine dissociation and a less hindered Ind
rotation. Since the NMR spectra indicate that the exchange
process responsible for the equivalence of the SiMe3 sub-
stituents in 2 and 5 is faster than the corresponding process
in the Ni–Cl complex {1,3-(SiMe3)2-Ind}Ni(PPh3)Cl, we
conclude that these observations are more consistent with
the rapid rotation of the 1,3-(SiMe3)2-Ind ligand as opposed
to a dissociation/re-association of PR3. The lower lability of their
phosphine ligands should influence the hydrosilylation activ-
ities of the Ni–Me compounds (vide infra).
Solid-state structures
Single crystals obtained for 2–4 allowed us to examine
their solid-state structures, which were found to be free of
disorder and refined to a good degree of precision, as re-
flected in the R values of 3.77% (2), 4.16% (3), and 3.62%
(4). The ORTEP diagrams for these compounds are shown in
Fig. 2, and a selection of structural parameters is listed in
Table 1; the full list of structural refinement parameters and
the crystal data are provided as Supplementary data.4
In all three compounds, the coordination geometry around
the Ni atom can be described as distorted square-planar, with
the largest distortion arising from the small C1–Ni–C3 an-
gles of ~67°. As commonly observed in this family of com-
plexes (1), the Ind–Ni interaction in 2–4 is primarily through
the allylic carbons, while the Ni–C bond distances for the
benzo carbons are longer. This “slippage” away from an
5
ideal η hapticity is attributed to the tendency of Ni(II) to fa-
vour forming 16-electron complexes.
Another common type of slippage in this family of com-
plexes is the so-called “sideways slippage”, which is re-
flected in the unsymmetrical Ni–C bonds involving the
allylic carbons (Ni–C1 > Ni–C3) and caused by the unequal
trans influences of the PR3 and X ligands (X= halides,
phthalimidato, thiolato, triflato, and so forth). Little or no
“sideways slippage” is evident in the PMe3 compounds 3
and 4 (∆Ni–C(1,3) is within 6 esd values), which is consis-
tent with the similar trans influences of PR3 and Me ligands
(1). On the other hand, complex 2 exhibits fairly different
Ni–C bond lengths (∆Ni–C(1,3) = 50 × esd). We suspect
that this anomalous elongation of the Ni–C3 bond in 2 is due
primarily to the steric repulsion between PPh3 and its neigh-
bouring SiMe3 group. This repulsion is also manifested in
the greater C3–Ni–P angle in 2 (111°) compared with 3
(106°) and 4 (104°). Similar observations have been noted
for the analogous complexes {(SiMe3)n-Ind}Ni(PPh3)Cl (n =
1 or 2) (12).
In contrast to the above discussed temperature-
dependence of the {1,3-(SiMe3)2-Ind}Ni(PPh3)Cl spectra,
low-temperature spectra of the Ni–Me derivatives 2 and 5
showed no change in the linewidths of the 31P signals or the
multiplicities of the Si(CH3)3 singlets (Figs. 1b and 1c)
Finally, the shorter Ni–P bond distance in 4 relative to 2 is
consistent with both the greater nucleophilicity of PMe3 and
the larger Tolman cone angle for PPh3, whereas the shorter
4 Supplementary data for this article are available on the journal Web site (canjchem.nrc.ca) or may be purchased from the Depository of Un-
published Data, Document Delivery, CISTI, National Research Council Canada, Ottawa, ON K1A 0R6, Canada. DUD 3832. For more in-
formation on obtaining material, refer to cisti-icist.nrc-cnrc.gc.ca/cms/unpub_e.shtml. CCDC 686331, 686332, and 686333 contain the
crystallographic data for this manuscript. These data can be obtained, free of charge, via www.ccdc.cam.ac.uk/conts/retrieving.html (Or
from the Cambridge Crystallographic Data Centre, 12 Union Road, Cambridge CB2 1EZ, UK; fax +44 1223 336033; or de-
posit@ccdc.cam.ac.uk).
© 2008 NRC Canada

Chen and Zargarian
283
Fig. 2. ORTEP diagrams for complexes 2 (a), 3 (b), and 4 (c).
Thermal ellipsoids are shown at the 30% probability level. Hy-
drogen atoms are omitted for clarity.
Table 1. Structural parameters for complexes 2–4.
2 (X= C8)
3 (X= Cl)
4 (X= C8)
2.1013(14)
2.0910(14)
2.1763(14)
2.3015(14)
2.2700(15)
1.9591(16)
2.1343(4)
1.8634(15)
1.8684(16)
1.428(2)
1.423(2)
1.466(2)
1.457(2)
1.434(2)
93.82(6)
88.07(5)
173.24(5)
160.13(6)
111.60(4)
66.98(6)
2.108(3)
2.056(2)
2.097(3)
2.308(3)
2.313(3)
2.1915(8)
2.1671(8)
1.873(3)
1.883(3)
1.412(4)
1.418(4)
1.470(5)
1.469(4)
1.422(5)
96.64(8)
90.85(3)
165.21(10)
163.36(8)
105.71(8)
67.63(11)
0.21
2.0999(15)
2.0812(16)
2.1098(16)
2.3075(16)
2.2887(15)
1.9456(19)
2.1129(5)
1.8662(16)
—
1.431(2)
1.404(2)
1.463(2)
1.440(3)
1.428(2
98.66(8)
90.05(7)
170.21(5)
165.32(8)
104.47(5)
66.70(6)
0.19
Ni—C(1)
Ni—C(2)
Ni—C(3)
Ni—C(3a)
Ni—C(7a)
Ni—X
Ni—P
Si(1)—C(1)
Si(2)—C(3)
C(1)—C(2)
C(2)—C(3)
C(1)—C(7a)
C(3)—C(3a)
C(3a)—C(7a)
X–Ni–C(1)
X–Ni–P
C(1)–Ni–P
X–Ni–C(3)
P–Ni–C(3)
C(1)–Ni–C(3)
∆Ni—Ca (Å)
0.15
Catalytic reactivities
The hydrosilylation activities of the Ni–Me complexes
were examined for comparison with those of the correspond-
ing Ni–Cl precursors. In addition, we hoped to determine the
influence of factors such as nature of the phosphine ligand
and the Ind substituents on these reactivities. The addition of
PhSiH3 to styrene was selected as a model system to allow
comparison to our previous studies. The catalysis is initiated
by the direct reaction of PhSiH3 with the Ni–Me moiety, ob-
viating the need for an activator/initiator. The main product
of this reaction, PhCH(Me)(SiPhH2), arises from regio-
selective (>95%) α-addition of the silyl moiety (eq. [1]); as
will be discussed later, appreciable quantities of silane oligo-
mers, (PhSiH)n, are also produced in some cases (vide infra).
The catalytic runs were carried out at room temperature with
Ni/styrene/PhSiH3 ratios of 1:50:75 or 1:100:125; the excess
of PhSiH3 was meant as a compensation for its possible loss
in the competitive silane oligomerization side reaction, thus
ensuring the availability of sufficient quantities of silane for
the desired hydrosilylation reaction. Analysis of the products
and quantification of the yields were done according to pre-
viously published procedures (12).
[1]
The results of the catalytic hydrosilylation reactions are
shown in Table 2. The main point arising from the data is
that the Ni–Me derivatives studied display more or less the
same level of hydrosilylation activity as their analogous
IndNi(PPh3)Cl complexes when the latter are used in the ab-
sence of any initiators (18). For instance, the addition of
PhSiH3 to styrene using the complexes {1,3-(SiMe3)2-
Ind}Ni(PPh3)X under the same conditions proceeded with an
82% yield with X = Cl (12) vs. 79% with X = Me (Run 4).
The similar catalytic activities of the Ni–Me and Ni–Cl pre-
Ni–P bond in 4 relative to 3 reflects the tendency of the
Ni–alkyl compounds to form stronger Ni–P interactions com-
pared to their Ni–Cl counterparts (17). To the extent that
solid state bond distances can correlate with the kinetic
labilities of phosphines, the shorter Ni–P distance in 4 sig-
nals lower hydrosilylation activities for this complex (vide
infra).
© 2008 NRC Canada

284
Can. J. Chem. Vol. 87, 2009
Table 2. Hydrosilylation of Styrene with PhSiH3 catalyzed by IndNi(PR3)Me.
Run
Ind
PR3
Ni:Styrene:PhSiH3
Time (h)
Conversion (%)
a
1
2
3
4
5
1-Me-Ind
PPh3
PPh3
PPh3
PPh3
1:50:75
5
5
5
5
3
37
26
74
79
18
1-SiMe3-Ind
1,3-(SiMe3)2-Ind
1,3-(SiMe3)2-Ind
1-Me-Ind
1:50:75
1:50:75
1:100:125
1:100:125
b
PMe3
PMe3
PMe3
PMe3
5
3
22
29
6
7
8
1-i-Pr-Ind
1:100:125
1:100:125
1:100:125
5
3
31
11
1-SiMe3-Ind
1,3-(SiMe3)2-Ind
5
3
14
35
5
36
Note: Unless otherwise specified, the catalytic runs were conducted at room temperature in 394 mL of ben-
zene using 5 µmol of the Ni precursor.
aThe runs using PPh3-based precursors gave traces only of cyclic and linear (PhSiH)n oligomers.
bThe runs using PMe3-based precursors gave variable quantities of cyclic and linear (PhSiH)n oligomers.
catalysts can be rationalized by noting that both of these pre-
cursors generate the same catalytically active intermediates
beyond the initiation step.
Inspection of the data listed in Table 2 indicates that the
PPh3 derivatives are more active than their corresponding
PMe3 counterparts (Runs 1–3 vs. 5, 7, and 8), particularly
for the precursors featuring the bulky 1,3-(SiMe)3-Ind
ligand, the most effective pre-catalyst being complex 2
Table 3. Dehydrocoupling of PhSiH3 catalyzed by IndNi(PMe3)Me.
(PhSiH)n
Ind
Mw
Mn
Mw/Mn
% lin
1-Me-Ind
1-i-Pr-Ind
1- SiMe3-Ind
1197
1206
1362
1439a,b,c
1283a,c
1354b,c
1318c
1254
1172
1180
1283
1331
1223
1272
1252
1216
1.02
1.02
1.06
1.08
1.05
1.06
1.05
1.03
79
87
82
66
82
67
87
88
(Runs
3
and 4). These observations confirm that
hydrosilylation activity correlates with steric bulk of the Ind
ligand (1,3-SiMe3-Ind > 1-i-Pr-Ind > 1-Me-Ind) and phos-
phine lability (PPh3 > PMe3; Ni–Me < Ni–Cl). Interestingly,
the Ni:substrate ratio does not appear to have a significant
impact on the reactivities (Run 3 vs. 4).
1,3-(SiMe3)2-Ind
Note: Unless otherwise indicated, the catalytic runs were conducted by
stirring the Ni precursor (18 µmol) and PhSiH3 (200 equiv.) in toluene
(0.25 mL) at –35 °C for 7 day.
As mentioned above, the PMe3-based precursors exhibit
lower hydrosilylation activities because of the less labile Ni–
PMe3 (12). On the other hand, the PMe3-based complexes
are efficient promoters of the competitive dehydrocoupling
of PhSiH3 (eq. [2]), whereas their PPh3 counterparts are
nearly inactive for this reaction; this implies that the
dehydrocoupling does not require phosphine dissociation.
Upon a closer examination of the silane dehydrocoupling re-
action with the IndNi(PMe3)Me precursors, we found that
the outcome of the catalysis was virtually independent of the
specific Ind ligand used (Table 3). This implies that all of
the IndNi(PMe3)Me precursors generate the same intermedi-
ate (e.g., Ni0(PMe3)n) that acts as the active catalyst for the
Si–Si bond formation reaction. It is worth noting that previ-
ous studies in our group have shown that most Ni0(PR3)n re-
act with PhSiH3 to produce mixtures of cyclic and linear
oligomers (19), but the involvement of in situ generated Ni0
species in the present studies has not been confirmed.
aThis experiment was run in THF.
bThis experiment was run at 25 °C.
cThis experiment was run for 4 days.
derivatives are better catalyst precursors compared to their
PMe3 counterparts, and that bulky indenyl ligands facilitate
the catalysis, presumably by favouring the phosphine dissoci-
ation step. Contrary to our expectation that stronger Ni–P
interactions in the Ni–Me derivatives (supported by both solu-
tion and solid-state data) should mean a less facile phosphine
dissociation step and hence lower hydrosilylation activities in
these complexes, reactivity data showed that the nature of the
ligand X (Cl or Me) has no significant impact on the cataly-
sis. Studies are in progress to elucidate the intimate mecha-
nisms of hydrosilylation and silane oligomerization reactions
as catalyzed by indenyl–nickel complexes.
Experimental section
[2]
General
All manipulations were performed under nitrogen atmo-
sphere using standard Schlenk techniques and a glovebox.
Dry, oxygen-free solvents were employed throughout. The
Ni precursors Ni(PR3)2Cl2 were prepared from NiCl2·6H2O
and PPh3 or NiCl2 and PMe3, respectively. The syntheses of
1-SiMe3-IndH and 1,3-(SiMe3)2-IndH and their Ni deriva-
Conclusion
This study supports our earlier observation that phosphine
dissociation is an important step in the addition of PhSiH3 to
styrene catalyzed by IndNi(phosphine)X. It follows that PPh3
© 2008 NRC Canada

Chen and Zargarian
(1-SiMe3-Ind)Ni(PPh3)Cl
285
tives
and
{1,3-(SiMe3)2-
28 Hz; Ni–CH3). 31P{H1} NMR (C6D6): 44.0 (s). Anal.
calcd. for C34H41Ni1P1Si2: C, 68.57; H, 6.94. Found: C,
67.94; H, 7.13.
Ind}Ni(PPh3)Cl have been described previously;12 1-i-Pr-
IndH and Li[i-Pr-Ind] were prepared as described previously
(3b, 6a). Ni(PMe3)2MeCl (21) was prepared by reacting
Ni(PMe3)2Cl2 with 1 equiv. of MeMgCl in THF at –30 °C,
{1,3-(SiMe3)2-Ind}NiPMe3Cl (3)
1
and characterized by H NMR and 31P NMR spectroscopy.
A THF solution (100 mL) of Li[1,3-(SiMe3)2-Ind]
(0.611 g, 2.29 mmol) was added dropwise to a stirring solu-
tion of Ni(PMe3)2Cl2 (0.646 g, 2.29 mmol, in 30 mL of
THF) at room temperature. The reaction vessel was placed
under vacuum for 30 s at regular intervals (following the ad-
dition of each 15 mL aliquot of the Li[1,3-(SiMe3)2-Ind] so-
lution) to remove the free PMe3 released from the reaction
(inefficient removal of PMe3 results in the decomposition of
the product and the formation of side-products such as
Ni(PMe3)4). The reaction mixture was stirred for 10 min af-
ter the addition was complete, then the solvent was removed
under vacuum, and the resulting residue was extracted with
hexane (~15 mL). The extracts were concentrated to about
3 mL and allowed to stand at room temperature, which gave
red crystals that were covered with a viscous oil (0.457 g,
51% crude yield). Repeated attempts to purify this product
were not successful due to the high solubility of the desired
product and the main impurity (Ni(PMe3)4).
All other reagents used in the experiments were obtained
from commercial sources and used as received. A Bruker
ARX 400 spectrometer was used for recording 1H
(400 MHz), 13C{H1}(100.56 MHz), and 31P{H1}(161.92 MHz).
Unless otherwise noted, the NMR spectra were recorded at
25 °C. The elemental analyses were performed by the
Laboratoire d’Analyse Élémentaire (Université de Montréal).
The molecular masses of (PhSiH)n oligomers were measured
by a Waters 1525 GPC system with a refractive-index detec-
tor, using THF as eluent and polystyrene standards.
(1-SiMe3-Ind)Ni(PPh3)Me (1)
MeMgCl (0.43 mL of a 3 mol/L solution in THF,
1.30 mmol) was added dropwise to a stirring solution of (1-
SiMe3-Ind)Ni(PPh3)Cl (0.64 g, 1.18, in 40 mL of Et2O) at
room temperature. The reaction mixture was stirred for
30 min after the addition was completed, the solution was
concentrated to about 5 mL under vacuum, and the resulting
residue was extracted into hexane (40 mL). The extracts
were concentrated to about 8 mL and cooled to –30 °C to
give a brown solid (0.45 g, 73% yield).
1H NMR (C6D6): 7.02–6.91 (m; H4, H5, H6 and H7),
2
6.72 (s; H2), 0.83 (d, JP–H = 9.0 Hz; P–CH3), 0.33 (s;
Si(CH3)3). 13C{H1} NMR (C6D6): 134.4, 125.8 and 120.1
(C3A, C7A, C5, C6, C4 and C7), 116.0 (C2), 86.0 (C1 and
1
1H NMR (C6D6): 7.38 (d, 3JH–H = 8.0 Hz; H4 or H7), 7.35
C3), 15.3 (d, JP–C = 28 Hz; P–CH3), 0.4 (Si(CH3)3).
31P{H1} NMR (C6D6): –18.1 (s).
3
(m; aromatic protons of PPh3), 7.28 and 7.05 (t, JH–H
=
7.6 Hz; H5 and H6), 7.00 (m; aromatic protons of PPh3),
3
6.81 (d, JH–H = 8.0 Hz; H4 or H7), 6.42 (s; H2), 4.23 (s;
(1-SiMe3-Ind)Ni(PMe3)Me (4)
3
H3), 0.53 (s; Si(CH3)3), –0.52 (d, JP–H = 6.0 Hz; Ni–CH3).
A THF solution (30 mL) of Li[1-(SiMe3)Ind] (0.232 g,
1.19 mmol) was added dropwise to a stirring THF (30 mL)
solution of Ni(PMe3)2MeCl (0.297 g, 1.13 mmol) at RT. The
reaction vessel was placed under vacuum for 30 s at regular
intervals (following the addition of each ~7 mL aliquot of
the Li[1-(SiMe3)Ind] solution) to remove the free PMe3 re-
leased from the reaction (inefficient removal of PMe3 results
in the decomposition of the product and the formation of
side-products such as Ni(PMe3)4). The reaction mixture was
stirred for 10 min after the addition was complete, the sol-
vent was removed under vacuum, and the residue was ex-
tracted with 20 mL hexane. The solution was concentrated to
about 5 mL and cooled to –78 °C to yield a dark red solid
(0.332 g, 87% yield).
13C{H1} NMR (CDCl3): 134.2 (d, JP–C = 43 Hz; Cipso),
1
2
133.9 (d, JP–C = 11 Hz; Cortho), 128.9 (Cpara), 128.1 (d,
3JP–C = 10 Hz; Cmeta), 125.0, 124.0, 122.9, 121.5, 118.9, and
2
117.7 (C3A, C4, C5, C6, C7and C7A), 106.3 (d, JC–P
=
=
2
2
6 Hz; C2), 82.0 (d, JC–P = 9 Hz; C3), 82.0 (d, JC–P
12 Hz; C1), 0.2 (–SiMe3), –24.4 (d, 2JP–C = 27 Hz; Ni–CH3).
31P{H1} NMR (C6D6): 47.1 (s). Anal. calcd. for
C31H33Ni1P1Si1: C, 71.14; H, 6.36. Found: C, 70.29; H, 6.62.
{1,3-(SiMe3)2-Ind}Ni(PPh3)Me (2)
MeMgCl (0.35 mL of a 3 mol/L solution in THF,
1.05 mmol) was added dropwise to a stirring solution of
{1,3-(SiMe3)2-Ind}Ni(PPh3)Cl (0.59 g, 0.95 mmol, in 30 mL
of Et2O) at room temperature. The reaction mixture was
stirred for 30 min after the addition was complete, the sol-
vent was removed under vacuum, and the residue was ex-
tracted with hexane (~60 mL). The extracts were
concentrated to ~5 mL and allowed to stand at room temper-
ature, which gave a brown solid (0.41 g, 66% yield). Single
crystals of this compound were grown from cold, concen-
trated hexane solutions and subjected to X-ray diffraction
studies.
3
1H NMR (C6D6): 7.51 (d, JH–H= 7.6 Hz; H4 or H7),
3
7.14–7.12 (m; aromatic protons of Ind), 7.04 (t, JH–H
=
7.0 Hz; H5 or H6), 6.27 (bs; H2), 4.62 (b s; H3), 0.60 (d,
2JP–H = 9.1 Hz; P–CH3), 0.50 (s; Si(CH3)3), –0.64 (d, 3JP–H
=
6.4 Hz; Ni–CH3). 13C{H1} NMR (C6D6): 126.6, 125.4,
123.1, 122.0, 119.6 and 117.4 (C3A, C7A, C5, C6, C4 and
C7), 106.6 (C2), 81.7 (d, 2JC–P = 11 Hz; C1), 76.9 (C3), 16.5
1
2
(d, JP–C = 28 Hz; P–CH3), 0.6 (SiMe3), –28.4 (d, JP–C
=
28 Hz; Ni–CH3). 31P{H1} NMR (C6D6): –3.5 (s). Anal.
calcd. for C16H27Ni1P1Si1: C, 57.00; H, 8.07. Found: C,
57.37; H, 8.06.
1H NMR (C6D6): 7.49–7.41 (m; PPh3), 7.06–6.98 (m; aro-
matic protons of Ind and PPh3), 6.33 (s; H2), 0.28 (s;
Si(CH3)3), –0.70 (d, JP–H = 7.0 Hz; Ni–CH3). 13C{H1}
3
2
1
{1,3-(SiMe3)2-Ind}Ni(PMe3)Me (5)
(CDCl3): 134.4 (d, JP–C = 11 Hz; Cor3tho), 134.3 (d, JP–C
=
43 Hz; Cipso), 129.8 (Cpara), 128.0 (d, JP–C = 10 Hz; Cmeta),
The procedure used for preparation of complex 4 was
used here with Li[1,3-(SiMe3)2-Ind] (0.351 g, 1.32 mmol)
and Ni(PMe3)2MeCl (0.328 g, 1.25 mmol). The reaction
125.8, 121.9 and 120.7 (C3A, C7A, C5, C6, C4 and C7),
112.1 (C2), 87.9 (C1 and C3), 0.5 (SiMe3), –24.4 (d, JP–C
2
=
© 2008 NRC Canada

286
Can. J. Chem. Vol. 87, 2009
mixture was stirred for 10 min after the addition was com-
plete, and then the solvent was removed under vacuum.
Recrystallization attempts were unsuccessful because prod-
ucts are highly soluble in non polar solvents such as toluene
and hexane. The residue was sublimated at 85–95 °C at
3 Torr (1 Torr = 101 325/760 Pa), and the initially obtained
oily solid ((Ni(PMe3)4)) was cleaned and the sublimation re-
started to collect the target product as a dark brown solid
(0.301 g, 64% yield).
Acknowledgement
The authors are grateful to the Natural Sciences and Engi-
neering Research Council of Canada (NSERC) for financial
support of this work.
References
1H NMR (C6D6): 7.45–7.43 and 6.99–6.96 (m; aromatic
1. For a review on the structures, characterization, and reactivi-
ties of group 10 metal indenyl complexes, see: D. Zargarian.
Coord. Chem. Rev. 157, 233 (2002).
2. (a) M.-A. Dubois, R. Wang, D. Zargarian, J. Tian, R.
Vollmerhaus, Z. Li, and S. Collins. Organometallics, 20, 663
(2001); (b) L.F. Groux, D. Zargarian, L.C. Simon, and J.B.P.
Soares. J. Mol. Catal. A: Chem. 193, 51 (2003).
3. (a) R. Wang, F. Belanger-Gariepy, and D. Zargarian.
Organometallics, 18, 5548 (1999); (b) R. Wang, L.F. Groux,
and D. Zargarian. J. Organomet. Chem. 660, 98 (2002); (c) E.
Rivera, R. Wang, X.X. Zhu, D. Zargarian, and R. Giasson. J.
Mol. Catal. A, 325, 204 (2003).
2
protons of Ind), 6.13 (s; H2), 0.64 (d, JP–H = 9.2 Hz;
3
P–CH3), 0.42 (s; Si(CH3)3), –0.71 (d, JP–H = 7.2 Hz; Ni–
CH3). 13C{H1} NMR (C6D6): 126.4, 122.2 and 121.4 (C3A,
C7A, C5, C6, C4 and C7), 111.2 (C2), 87.1 (C1 and C3),
1
2
16.3 (d, JP–C = 29; P–CH3), 1.2 (SiMe3), –29.6 (d, JP–C
=
31 Hz; Ni–CH3). 31P{H1} NMR (C6D6): –8.6 (s). Anal.
calcd. for C19H35Ni1P1Si2: C, 55.75; H, 8.62. Found: C,
53.83; H, 8.85.
(1-i-Pr-Ind)Ni(PMe3)Me (6)
The procedure used for preparation of complex 4 was
used here with Li[1-i-Pr-Ind] (0.214 g, 1.30 mmol) and
Ni(PMe3)2MeCl (0.310 g, 1.18 mmol) to give the target
complex as a dark red solid (0.277 g, 77% yield).
1H NMR (C6D6): 7.21–7.00 (m; aromatic protons of Ind),
4. F.G. Fontaine, T. Kadkhodazadeh, and D. Zargarian. Chem.
Commun. 1253 (1998).
5. R. Vollmerhaus, F. Belanger-Gariepy, and D. Zargarian.
Organometallics, 16, 4762 (1997).
6. (a) M.-A. Dubois. M.Sc. Thesis. Universite de Montreal. 2000;
(b) H.-M. Sun, W.-F. Li, X. Han, Q. Shen, and Y. Zhang. J.
Organomet. Chem. 688, 132 (2003); (c) W.-F. Li, H.-M. Sun,
Q. Shen, Y. Zhang, and K.-B. Yu. Polyhedron, 23, 1473 (2004).
7. F.-G. Fontaine, R.-V. Nguyen, and D. Zargarian. Can. J. Chem.
81, 1299 (2003).
3
6.15 (b s, H2), 4.44 (b s; H3), 2.87 (septet, JH–H = 6.8 Hz;
3
CHMe2), 1.39 and 1.35 (d, JH–H = 6.8 Hz, CH(CH3)2), 0.65
2
3
(d, JP–H = 9.0 Hz; P–CH3), –0.77 (d, JP–H = 6.4 Hz; Ni–
CH3). 13C{H1} NMR (C6D6): 122.4, 122.1, 121.5, 120.2,
116.8 and 116.8 (C3A, C7A, C5, C6, C4 and C7), 100.8 (d,
2JC–P = 12 Hz; C1), 97.3 (C2), 70.2 (C3), 25.3 (iPr–CH),
8. R. Wang, L.F. Groux, and D. Zargarian. Organometallics, 21,
5531 (2002).
1
23.8 (iPr–CH3), 21.1 (iPr–CH3), 16.5 (d, JP–C = 28 Hz;
9. (a) L.F. Groux, F. Belanger-Gariepy, D. Zargarian, and R.
Vollmerhaus. Organometallics, 19, 1507 (2000); (b) L.F.
Groux and D. Zargarian. Organometallics, 20, 3811 (2001);
(c) L.F. Groux and D. Zargarian. Organometallics, 22, 3124
(2003); (d) L.F. Groux and D. Zargarian. Organometallics, 22,
4759 (2003).
P–CH3), –22.2 (d, JP–C = 25 Hz; Ni–CH3). 31P{H1} NMR
2
(C6D6): –4.0 (s). Anal. calcd. for C16H25Ni1P1: C, 62.59; H,
8.21. Found: C, 61.93; H, 8.20.
General procedure for catalytic runs
The catalytic runs were carried out by stirring the sub-
strates and the Ni precursors in benzene (hydrosilylations) or
toluene (dehydrogenative oligomerizations) at the specified
temperatures. The characterization of the products and yield
determinations were done on the basis of NMR and GC–
MS, as described in our previous report (12).
10. M. Jimenez-Tenorio, M.C. Puerta, I. Salcedo, P. Valerga, S.I.
Costa, L.C. Silva, and P.T. Gomes. Organometallics, 23, 3139
(2004).
11. F.G. Fontaine and D. Zargarian. Organometallics, 21, 401 (2002).
12. Y. Chen, C. Sui-Seng, S. Boucher, and D. Zargarian. Organo-
metallics, 24, 149 (2005).
13. For instance, the addition of PhSiH3 to styrene catalyzed by
IndNi(PPh3)Cl in the absence of any activator/initiator gives
yields of 7% (1-Me-Ind), 56% (1-SiMe3-Ind), and 82% (1,3,-
(SiMe3)2-Ind) (see ref. 12). It is important to emphasize, how-
ever, that for reactions run in the presence of initiators/activa-
tors, the steric bulk has little or no favourable influence on the
reactivities unless the Ind ligand is substituted at both 1- and
3-positions. For instance, the addition of PhSiH3 to styrene
catalyzed by the combination of (1-R-Ind)Ni(PPh3)Cl and
NaBPh4 is no more effective with R = i-Pr (61%) than Me
(69%) (see ref. 7). This can be understood by recalling that the
complexes (1-R-Ind)Ni(PPh3)X always adopt a configuration
that places the PPh3 ligand away from the more bulky Ind
substituent to avoid steric congestion (see ref. 1); hence, plac-
ing a substituent at the 1-position has no advantage for in-
creasing the lability of the phosphine ligand or the catalytic
activities.
X-ray crystallographic studies
Crystals suitable for X-ray diffraction studies were ob-
tained by recrystallization of the purified solids (for 2 and 4)
or crude material (for 3) from hexane solutions. The crystal-
lographic data were collected on a Bruker AXS SMART 2K
diffractometer with graphite-monochromated Cu Kα radia-
tion (λ = 1.54178 Å) at 220(2) K, using SMART (20). Cell
refinement and data reduction were done using SAINT (21).
A SADABS (22) absorption correction was applied. The
structures were solved by direct methods using SHELXS97
(23) and difmap synthesis using SHELXL97 (24); the refine-
ments were done on F2 by full-matrix least squares. The
PPh3 and Cl groups in 2 are disordered. Details of data col-
lection and refinement are listed in Table 1, while a selection
of structural parameters is listed in Table 2. Complete crys-
tallographic data for these structures are included in the Sup-
plementary data.4
14. F.-G. Fontaine, M.-A. Dubois, and D. Zargarian. Organometallics,
20, 5145 (2001).
© 2008 NRC Canada

Chen and Zargarian
287
21. H.F. Klein and H.H. Karsch. Chem. Ber. 105, 2628 (1972).
22. SMART [computer program]. Release 5.059. Bruker molecular
analysis research tool. Bruker AXS Inc. Madison, WI. 1999.
23. SAINT [computer program]. Release 6.06. Integration soft-
ware for single crystal data. Bruker AXS Inc. Madison, WI.
1999.
24. G.M. Sheldrick. SADABS [computer program]. Bruker area
detector absorption corrections. Bruker AXS Inc. Madison,
WI. 1997.
25. G.M. Sheldrick. SHELXS [computer program]. Program for
the solution of crystal structures. University of Goettingen,
Germany. 1997.
15. T.A. Huber, M. Bayrakdarian, S. Dion, I. Dubuc, F.G.-
Belanger, and D. Zargarian. Organometallics, 16, 5811 (1997).
16. F.-G. Fontaine and D. Zargarian. Organometallics, 21, 401
(2002).
17. Ni–PR3 interactions in IndNi(PR3)X compounds can be under-
stood in the following terms. Relatively soft X ligands like
alkyls result in more effective interactions and shorter Ni–P
bonds, whereas harder X ligands such as halides give less ef-
fective and longer Ni–P bonds. An alternative explanation for
the less effective Ni–X bonds is in terms of cis-labilizing inter-
actions between the occupied dz2 orbital on Ni and the lone
pair of electrons on ligands X (halide, OR, NR2, and so forth).
18. It should be added that the Ni–Cl precursors form much more
efficient catalysts when used in conjunction with cationic initi-
ators such as NaBPh4 (12).
26. G.M. Sheldrick. SHELXL [computer program]. Program for
the refinement of crystal structures. University of Goettingen,
Germany. 1997.
19. M. Morissette. M.Sc. Dissertation. Universite de Montreal. 2001.
20. R. Wang, L.F. Groux, and D. Zargarian. J. Organomet. Chem.
660, 98 (2002).
© 2008 NRC Canada