Dehydrocoupling of organosilanes with a dinuclear nickel hydride catalyst and isolation of a nickel silyl complex

Article abstract of DOI:10.1021/om100887v

The dehydrocoupling of organosilanes is an efficient method for producing polysilanes. Although this is traditionally done with group 4 metallocene catalysts, there are a few examples of nickel catalysts that are effective for this reaction. We report the

Full text of DOI:10.1021/om100887v

Organometallics 2010, 29, 6527–6533 6527
DOI: 10.1021/om100887v
Dehydrocoupling of Organosilanes with a Dinuclear Nickel Hydride
Catalyst and Isolation of a Nickel Silyl Complex
Erin E. Smith, Guodong Du,† Phillip E. Fanwick, and Mahdi M. Abu-Omar*
Brown Laboratory, Department of Chemistry, Purdue University, 560 Oval Drive, West Lafayette,
Indiana 47907, United States . †Current address: Department of Chemistry, University of North Dakota,
Grand Forks, ND 58202.
Received September 14, 2010
The dehydrocoupling of organosilanes is an efficient method for producing polysilanes. Although
this is traditionally done with group 4 metallocene catalysts, there are a few examples of nickel cata-
lysts that are effective for this reaction. We report the dehydrocoupling of phenylsilane and phenyl-
methylsilane with [(dippe)Ni(μ-H)]2 (1) (dippe=1,2-bis(diisopropylphosphino)ethane). As expected
from thermodynamic and steric evidence, the primary silane is more active toward dehydrocoupling
than the secondary silane. This catalyst compares favorably in required reaction conditions, molecular
weight of polysilane product, and selectivity for linear oligomers. Possible mechanisms for the
dehydrocoupling of silane are discussed. The hypothesized intermediate is a hydrido silyl nickel com-
plex. We report the isolation and single-crystal X-ray structure of a stable analogue of the proposed
catalytic intermediate, (dippe)Ni(SiCl3)Cl (2).
1. Introduction
of masked disilenes2 and electroreduction of dichlorosilanes.3
The most efficient alternative is dehydrocoupling of organo-
silanes with transition metal catalysts (eq 1). This method was
first studied by Harrod and co-workers using derivatives of
titanocene and zirconocene.4 Since then, much more work has
been done with group 4 catalyst systems.5 Other metal com-
plexes used for dehydrocoupling include groups 5-12 and the
lanthanide series.6-14
Polysilanes are of interest for their conductivity and optical
properties. Traditional industrial preparation of these
€
materials is done by the Wurtz-Fittig coupling reaction. This
method involves the coupling of diorganodihalosilanes using
molten sodium.1 The functional groups that are tolerated by
this method are limited, and despite high molecular weights,
the percent cyclic oligomers is also high. Other methods that
can be utilized to produce polysilanes include polymerization
*Corresponding author. E-mail: mabuomar@purdue.edu.
(1) (a) Corey, J. Y. The Chemistry of Organic Silicon Compounds;
Patai, S., Rappoport, Z., Eds.; John Wiley and Sons: New York, 1989;
pp 1-56. (b) West, R. The Chemistry of Organic Silicon Compounds; Patai, S.,
Rappoport, Z., Eds.; John Wiley and Sons: New York, 1989; pp 1207-1240.
(2) Sakamoto, K.; Obata, K.; Hirata, H.; Nakajima, M.; Sakurai, H.
J. Am. Chem. Soc. 1989, 111, 7641.
(3) Shono, T.; Kashimura, S.; Ishifune, M.; Nishida, R. J. Chem.
Soc., Chem. Commun. 1990, 1160.
(4) (a) Aitken, C.; Harrod, J. F.; Samuel, E. J. Organomet. Chem.
1985, 279, C11. (b) Aitken, C.; Harrod, J. F.; Samuel, E. J. Am. Chem. Soc.
1986, 108, 4059–4066. (c) Aitken, C.; Harrod, J. F.; Samuel, E. Can. J.
Chem. 1986, 64, 1677–1679. (d) Aitken, C.; Harrod, J. F.; Gill, U. S. Can. J.
Chem. 1987, 65, 1804–1809.
(5) (a) Mu, Y.; Aitken, C.; Cote, B.; Harrod, J. F. Can. J. Chem. 1991,
264, 164. (b) Campbell, W. H.; Hilty, T. K. Organometallics 1989, 8, 2615–
2618. (c) Woo, H. G.; Tilley, T. D. J. Am. Chem. Soc. 1989, 111, 3757–3758.
(d) Woo, H. G.; Tilley, T. D. J. Am. Chem. Soc. 1989, 111, 8043–8044. (e)
Woo, H. G.; Walzer, J. F.; Tilley, T. D. J. Am. Chem. Soc. 1992, 114, 7047–
7055. (f) Tilley, T. D. Acc. Chem. Res. 1993, 26, 22–29. (g) Bourg, S.; Corriu,
J. P.; Enders, M.; Moreau, J. J. E. Organometallics 1995, 14, 564–566.
(h) Peulecke, N.; Thomas, D.; Bauman, W.; Fischer, C.; Rosenthal, U. Tetra-
hedron Lett. 1997, 38, 6655–6656. (i) Hengge, E.; Gspaltl, P.; Pinter, E.
J. Organomet. Chem. 1996, 521, 145–155. (j) Corey, J. Y.; Zhu, X. H.; Bedard,
T. C.; Lange, L. D. Organometallics 1991, 10, 924–930. (k) Corey, J. Y.;
Huhmann, J. L.; Zhu, X. H. Organometallics 1993, 12, 1121–1130. (l) Wang,
Q.; Corey, J. Y. Can. J. Chem. 2000, 78, 1434–1440. (m) Banvetz, J. P.; Stein,
K. M.; Waymouth, R. M. Organometallics 1991, 10, 3430–3432. (n) Dioumaev,
V. K.; Harrod, J. F. J. Organomet. Chem. 1996, 521, 133–143. (o) Choi, N.;
Onozawa, S.; Sakakura, T.; Tanaka, M. Organometallics 1997, 16, 2765–2767.
(p) Obora, Y.; Tanaka, M. J. Organomet. Chem. 2000, 595, 1–11.
(6) (a) Corey, J. Y. Adv. Organomet. Chem. 2004, 51I, 1–52. (b) Gauvin,
F.; Harrod, J. F.; Woo, H. G. Adv. Organomet. Chem. 1998, 42, 363–405.
(c) Tilley, T. D. Comments Inorg. Chem. 1990, 10, 37–51.
(7) Aitken, C.; Barry, J. P.; Gauvin, F.; Harrod, J. F.; Malek, A.;
Rousseau, D. Organometallics 1989, 8, 1732–1736.
(8) Itazaki, M.; Ueda, K.; Nakazawa, H. Angew. Chem., Int. Ed.
2009, 48, 3313–3316.
(9) (a) Ojima, I.; Kogure, T.; Nagai, Y. J. Organomet. Chem. 1973, 55,
C7. (b) Chang, L. S.; Corey, J. Organometallics 1989, 8, 1885–1893.
(c) Rosenberg, L.; Davis, C.; Yao, J. J. Am. Chem. Soc. 2001, 123, 5120–
5121. (d) Rosenberg, L.; Kobus, D. J. Organomet. Chem. 2003, 685, 107–
112. (e) Fryzuk, M. D.; Rosenberg, L.; Rettig, S. J. Organometallics 1991,
10, 2537–2539. (f) Fryzuk, M. D.; Rosenberg, L.; Rettig, S. J. Inorg. Chim.
Acta 1994, 222, 345–364.
(10) Diversi, P..; Marchetti, F.; Ermini, V.; Metteoni, S.
J. Organomet. Chem. 2000, 593-594, 154–160.
(11) (a) Boudjouk, P.; Rajkumar, A. B.; Parker, W. L. J. Chem. Soc.,
Chem. Commun. 1991, 245–246. (b) Fontaine, F. G.; Kadkhodazadeh, T.;
Zargarian, D. J. Chem. Soc., Chem. Commun. 1998, 1253–1254. (c) Groux,
L. F.; Zargarian, D. Organometallics 2001, 20, 3811–3817. (d) Fontaine,
F. G.; Zargarian, D. Organometallics 2002, 21, 401–408. (e) Fontaine, F. G.;
Zargarian, D. J. Am. Chem. Soc. 2004, 126, 8786–8794.
(12) (a) Brown-Wensley, K. Organometallics 1987, 6, 1590–1591.
(b) Tanaka, M.; Kobayashi, T.; Hayashi, T.; Sakakura, T. Appl. Organomet.
Chem. 1988, 2, 91–92. (c) Chauhan, B.; Shimizu, T.; Tanaka, M. Chem. Lett.
1997, 785–786.
r
2010 American Chemical Society
Published on Web 11/10/2010
pubs.acs.org/Organometallics

6528 Organometallics, Vol. 29, No. 23, 2010
Smith et al.
Figure 1. (a) Plot of[PhSiH3] versus time. (b) Log-scale. Conditions: 3.72 mM(1.0 mol%)1 and0.372 Mphenylsilaneintoluene at25 °C.
Some interesting work has also been done on the catalytic
dehydrocoupling of organophosphines15-20 and heterode-
hydrocoupling of organophosphines with other compounds
such as boranes19-21 and the elements of group 14.19,21,22
Oligomeric phosphines have been prepared using catalysts
- 16
such as Cp*2Zr(H)3
and (N3N)Zr(PHR) (N3N = N(CH2CH2NSiMe3)3
,
Cp*Rh(allyl)2,17 (dippe)Rh(allyl),18
3- 17
.
Harrod and co-workers studied the dehydrocoupling of orga-
nophosphines with [(EBTHI)Ti(μ-H)]2 (EBTHI = ethylene-
1,2-bis(η5-4,5,6,7-tetrahydro-1-indenyl), which is a structural
analogue of the catalyst system discussed inthisreport, as they
both feature a bridging hydride.21
Nickel catalysts are of interest due to their low cost when
compared with precious metals. Examples of dehydrocoupling
of organosilanes with nickel are rare and have some limitations.
Nickel indenyl complex precursors require activators such
as AgBF4, MAO, or LiAlH4.11b,c It is believed that activation
of the catalyst precursor gives a cationic nickel species and,
in the case of LiAlH4, aneutralnickelhydridecomplex, whichis
presumed to be the active catalyst.
Mechanisms proposed for the dehydrocoupling of organo-
silanes with titanium and zirconium metallocene complexes
involve an intermediate complex containing a hydride and
silyl group.5c-f,23,24c The oligomerization of organosilanes with
late transition metals such as nickel likely begins with oxida-
tive addition of the silane to give a metal silyl complex. Exam-
ples of nickel silyl complexes have been isolated and char-
Figure 2. Plot of volume of dihydrogen versus time. Conditions:
16 mM (1.0 mol %) 1 and 1.6 M phenylsilane in toluene at 25 °C.
acterized.24 We report in this contribution dehydrocoupling
of PhSiH3 and PhMeSiH2 catalyzed by [(dippe)Ni(μ-H)]2 (1)
(dippe = 1,2-bis(diisopropylphosphino)ethane). The molec-
ular weight distributions (by GPC) of the polysilane products,
the catalytic activity of 1, and selectivity toward production
of linear oligomers in comparison to conventional group 4
metallocene catalysts are also reported. A mechanistic pro-
posal is advanced in which a nickel silyl hydride is the active
dehydrocoupling catalyst. In support of our mechanistic
hypothesis, we report the preparation, isolation, and X-ray
molecular structure of (dippe)Ni(SiCl3)Cl (2).
(13) (a) Sakakura, T.; Lautenschlager, H. J.; Nakajia, M.; Tanaka,
M. Chem. Lett. 1991, 913–916. (b) Forsyth, C. M.; Nolan, S. P.; Marks, T. J.
Organometallics 1991, 10, 2543–2545.
(14) Li, Z.; Iida, K.; Tomisaka, Y.; Yoshimura, A.; Hirao, T.;
Nomoto, A.; Ogawa, A. Organometallics 2003, 26, 1212–1216.
(15) Waterman, R. Organometallics 2007, 26, 2492.
(16) Fermin, M. C.; Stephan, D. W. J. Am. Chem. Soc. 1995, 117,
12645.
(17) Bohm, V. P. W.; Brookhart, M. Angew. Chem., Int. Ed. 2001, 40,
4694.
(18) Han, L.-B.; Tilley, T. D. J. Am. Chem. Soc. 2006, 128, 13698.
(19) (a) Waterman, R. Curr. Org. Chem. 2008, 12, 1322. (b) Waterman,
R. Dalton Trans. 2009, 18.
2. Results and Discussion
2.1. Oligomerization of Phenylsilane with [(dippe)NiH]2
(1). Oligomerization of a primary silane, PhSiH3, was carried
out in toluene under ambient conditions with 1 as the pre-
catalyst at 1.0 mol % loading (eq 1). The consumption of PhSiH3
1
was followed by H NMR. Figure 1 shows a typical kinetic
profile. The log-scale plot (Figure 1b) deviates from linearity,
and the observed rate decreases as the reaction progresses. This
deviation from pseudo-first-order kinetics is attributed to cata-
lyst deactivation during the reaction. Deviation from linearity is
also observed for the second-order plot of 1/[PhSiH3] versus
(20) Roering, A. J.; MacMillan, S. N.; Tanski, J. M.; Waterman, R.
Inorg. Chem. 2007, 46, 6855.
(21) (a) Xin, S.; Woo, H. G.; Harrod, J. F.; Samuel, E.; Lebuis, A.-M.
J. Am. Chem. Soc. 1997, 119, 5307. (b) Xin, S.; Harrod, J. F.; Samuel, E.
J. Am. Chem. Soc. 1994, 116, 11562.
(22) Dorn, H.; Singh, R. A.; Massey, J. A.; Nelson, J. M.; Jaska,
C. A.; Lough., A. J.; Manners, I. J. Am. Chem. Soc. 2000, 122, 6669.
(23) Spaltenstein, E.; Palma, P.; Kreutzer, K.; Willoughby, C. A.;
Davis, W. M.; Buchwald, S. L. J. Am. Chem. Soc. 1994, 116, 10308–
10309.
(24) (a) Iluc, V. M.; Hillhouse, G. L. Tetrahedron 2006, 62, 7577–
7582. (b) Adhikari, D.; Pink, M.; Mindiola, D. J. Organometallics 2009, 28,
2072–2077. (c) Iluc, V. M.; Hillhouse, G. L. J. Am. Chem. Soc. 2010, 132,
11890–11892.

Article
Organometallics, Vol. 29, No. 23, 2010 6529
Table 1. Summary of GPC Data for Dehydrocoupling of PhSiH3
and PhMeSiH2 Catalyzed by 1a
Table 3. Direct Comparison the Dehydrocoupling of PhSiH3 with
Nickel versus Zirconium Catalystsa
b
peak # area (%) Mn Mw PDI
b
silane
% yield GPC
H2
catalystb
solvent t (h)
peak # Mw PDI % area
c
phenylsilane (PhSiH3)
1
2
1
2
3
32
68
66
17
17
1750 3200 1.82
521 549 1.05
951 1160 1.22
475 479 1.01
320 330 1.03
Cp2ZrMe2
Cp2ZrMe2
toluene
neat
2
1
79
98
nd
nd
nd
68
32
nd
31
69
32
68
phenylmethylsilane (PhMeSiH2)
1
2
2390 1.33
580 1.05
Ind2ZrMe2
toluene 96
neat 24
25
78
nd
nd
1
2
1
2
3200 1.14
1530 1.05
3200 1.83
549 1.05
a [PhSiH3] = 2.78 M, [PhMeSiH2] = 2.27 M, and 2.0 mol % of 1 in
toluene at 25 °C. b Determined by GPC with polystyrene standards.
[(dippe)NiH]2 toluene 1.5
81
Table 2. Performance of Selected Group 4 Dehydrocoupling
Catalysts from the Literaturea
a Dehydrocoupling of phenylsilane; all reactions at 21 °C. b Results
from reactions with catalyst loading of 2.0 mol %. c Measured by volume
displacement using a gas evolution apparatus.
b
catalyst
solvent
t
Mw PDI % cyclicc ref
Cp2Zr(NMe2)2 neat
CpCp*Zr[Si(SiMe3)3]Me neat
∼1 min 2810 1.44d
14d
nd
5l
5f
5n
5m
Scheme 1. σ-Bond Metathesis Mechanism for Dehydrocoupling
of Phenylsilane with 1
15 min 3100 1.8
Cp2ZrCl2/2nBuLi
(EBI)ZrCl2/2nBuLi
toluene 10 days 2450 1.18
toluene 7 h 2963 2.28
45d
13.8
a Dehydrocoupling of phenylsilane; all reactions at 20-25 °C. b High
molecular weight peak only. c Determined by 1H NMR. d Calculated
from reported data.
time. The incomplete consumption of PhSiH3 is also consistent
with catalyst death. The reaction was monitored as well by
following dihydrogen evolution (Figure 2). The results are
consistent with those obtained for PhSiH3 consumption, the
observed rate decreases as the reaction progresses, and only
69% conversion is observed based on the volume of H2
produced. Some of the H2 evolved is from the dihydride cata-
lyst, but since the catalyst concentration is low (1-2 mol %),
this volume is negligible. The conditions between the silane
consumption and dihydrogen evolution experiments differ,
hence, the difference in reaction times and conversion
between Figures 1 and 2. Some scatter is observed late in
the reaction due to error in peak integration at low silane
concentrations.
Increasing the catalyst loading to 2.0 mol % resulted in
improved conversion, 81% yield based on dihydrogen gas
evolved. The resulting polysilane was analyzed by gel per-
meation chromatography (GPC) and showed two peaks
(Figures S1 and S2 and Table 1). Peak 2 accounts for the
majority of the product (68%) and corresponds to oligomers
incorporating six monomers. The polydispersity index (PDI)
is quite narrow (1.05). Peak 1 corresponds to longer oligomers/
polymer containing an average of 16 phenylsilane monomers.
However, this peak is the minor product (32%) and exhibits a
broad PDI (1.82).
Zirconium catalysts have been used widely as dehydrocou-
pling catalysts. Table 2 shows representative examples of the
more successful zirconium catalysts. The molecular weight
distributions of the resulting silane polymer fall in a relatively
narrow range, 2450-3100 g mol-1. Both time-scale and percen-
tage of cyclic oligomer vary among different catalysts.
To ascertain reliable comparisons, we prepared two zirco-
nium catalysts (Cp2ZrMe2 and Ind2ZrMe2) and investigated
their dehydrocoupling catalysis of phenylsilane. GPC of the
resulting oligomers/polymers and percent yields for each of the
catalysts are presented in Table 3. For the high molecular weight
polyphenylsilane obtained with 1, Mw was considerably higher
than that observed for Cp2ZrMe2 (3200 versus 2390 g mol-1).
However, the distribution of high and low molecular weight
oligomers is very similar for Ind2ZrMe2 and 1. It is worth noting
that the zirconium catalysts required neat conditions to
afford higher conversions, and in general the time of reaction
for the nickel catalyst is more favorable. Also, forInd2ZrMe2,
the “lower” molecular weight peak consists of much higher
molecular weight oligomers than the other catalyst.
The percentage of cyclic oligomers in the product from
dehydrocoupling of PhSiH3 with catalyst 1 was determined
using 1H NMR. The chemical shift of Si-H groups for cyclic
phenylsilane oligomers lies between 4.8 and 5.3 ppm, while
the chemical shift for linear oligomers lies between 4.4 and
4.8 ppm.5m The resulting oligomer/polymer mixture using
catalyst 1had 19% cyclic oligomers after 17 h. Time was allowed
to reach thermodynamic equilibrium as linear oligomers are
slowly converted to cyclic. Catalyst 1 gives relatively high linear/
cyclic selectivity in comparison with group 4 metallocene cata-
lysts, which have cyclic percentages in the range 3-55%, as
1
determined by H NMR.5l-p In one of the nickel systems
investigated by Zargarian and co-workers, (1-Me-indenyl)Ni-
(PPh3)Me and Me2AlCH2PMe2 activator, mostly cyclic oligo-
mers are generated with Mw = 470 and PDI = 1.03 in less than
15 min with >95% conversion.11e
2.2. Oligomerization of Phenylmethylsilane with 1. The
dehydrocoupling reaction of phenylmethylsilane (PhMeSiH2)
was investigated with 2.0 mol % loading of 1 in toluene under
ambient temperature. Based on the amount of dihydrogen pro-
duced, the reaction proceeded to 76% conversion of PhMeSiH2.
This conversion is high compared to only 16% for Ind2ZrMe2
under the same conditions and after 1 day. Gel permeation
chromatography (GPC) analysis of the product showed three
distinguishable peaks (Figure S2 and Table 1). Light scatter-
ing detection by GPC has been shown to underestimate the

6530 Organometallics, Vol. 29, No. 23, 2010
Scheme 2. Oxidative-Addition Mechanism for Dehydrocoupling of Phenylsilane with 1
Smith et al.
molecular weight of polysilane by 20% when referenced versus
polystyrene standards.25 Therefore, peak 1 accounts for the
majority of the product (66%) and corresponds to an average
oligomer length of ten monomers. Peaks 2 and 3 account for
chains of three, four, or five monomers of phenylmethylsilane.
The high molecular weight peaks obtained using catalyst 1
with phenylmethylsilane average 10 monomers. This is very
high when compared with secondary silane oligomerization
with other late transition metal catalysts.
By comparison to results for phenylsilane, it can be con-
cluded that phenylsilane is more active for dehydrocoupling
than phenylmethylsilane. Zirconium catalysts also follow
this trend. Oligomerization of phenylmethylsilane with the
bis-indenyl zirconium catalyst [(Ind)2Zr(CH3)2] gave 16%
yield compared with 25% for phenylsilane. The same reac-
tion using the bis-cyclopentadienyl zirconium catalyst
[(Cp)2Zr(CH3)2] gave a 25% yield compared with 98% for
phenylsilane.
demanding), which results in a higher activation barrier for
silicon coupling at the metal center as the chain length increases.
2.3. Dehydrocoupling Mechanism. Two possible mecha-
nisms for the catalytic dehydrocoupling of phenylsilane have
been envisaged. The first is σ-bond metathesis (Scheme 1),
which is similar to the mechanism proposed by Tilley and co-
workers for the group 4 metallocene catalyst.5c-f The dimeric
precatalyst dissociates to active monomeric nickel in solution.
This step is induced by the presence of the organosilane as
a substrate (Ia ligand-exchange mechanism). η2-Silane com-
plexes of nickel are known.24a The mononuclear nickel
hydride undergoes σ-bond metathesis with silane to form a
nickel silyl complex. Additional σ-bond metathesis with silane
would regenerate nickel hydride and disilane. Repetition of
this sequence results in growth to oligomers and polymers.
An alternative mechanism (Scheme 2) is formation of a
nickel(0) from H2 extrusion,27 followed by oxidative addition
of phenylsilane to afford a silyl hydride complex. The latter
initiates via a silylene intermediate, followed by propagation.
Termination of the disilyl complex regenerates Ni(0) and
yields cyclic oligomers. Chain termination of the silyl hydride
affords linear polysilane. Recently, Hillhouse and co-workers
reported a reaction of [(dtbpe)NiH]2 with Mes2Si(H)K where
they were able to isolate both a nickel silyl complex and, with
The extent and rate of polymerization decrease moving
from primary to secondary silane. This trend is rationalized
by bond energies and steric hindrance. There is a difference of
5 kJ mol-1 for the silicon-hydrogen bond strength between
phenylsilane (377 kJ mol-1) and phenylmethylsilane (382 kJ
mol-1).26 Secondary silanes are also more crowded (sterically
(25) Devaux, J.; Daoust, D.; De Mahieu, A. F.; Strazielle, C. In
Inorganic and Organometallic Oligomers and Polymers; Laine, R. M.,
Harrod, J. F., Eds.; Kluwer Publishers: Amsterdam, 1991; pp 49-60.
(26) Brook, M. A. Silicon in Organic, Organometallic, and Polymer
Chemistry; John Wiley and Sons: McMaster University, Hamilton, Ontario,
Canada, 2000; p 172.
(27) (a) Vicic, D. A.; Jones, W. D. J. Am. Chem. Soc. 1997, 119,
10855–10856. (b) Garcia, J. J.; Jones, W. D. Organometallics 2000, 19,
5544–5545. (c) Vicic, D. A.; Jones, W. D. J. Am. Chem. Soc. 1999, 121,
4070–4071. (d) Tran, B. L.; Pink, M.; Mindiola, D. J. Organometallics 2009,
28, 2234–2243.

Article
Organometallics, Vol. 29, No. 23, 2010 6531
is an in-house built residual gas analyzer that consists of a single
quadrupole electron impact mass spectrometer.28 The forma-
tion of 2 from 1 with trichlorosilane corroborates the oxidative-
addition mechanism outlined in Scheme 2. It is likely that a
transient nickel-silyl hydride complex is formed in route to
product 2 (eq 2). NMR provided evidence for the intermediacy
of (dippe)Ni(SiCl3)H. The 1H NMR spectrum of the product
mixture from reaction 2 shows two sets of CH peaks for the
dippe ligand and a peak at -4.0 ppm, which is typical for a
Ni-H. Furthermore, two species are observed in the 31P spec-
trum at δ 76.1 (d, JPP = 25.0 Hz) and 89.7 (d, JPP = 25.0 Hz)
and a singlet at 59.9 ppm. The doublets are assigned to com-
pound 2, and the singlet is assigned to the proposed inter-
mediate, (dippe)Ni(SiCl3)H. The absence of splitting can be
explained by dynamic behavior via η2 species as shown in eq 2.
Two species are also observed by29SiNMR: δ 16.5 (dd, JSiP
=
79.4 and 303 Hz) and a triplet at 92.2 (t, JSiP = 20.9 Hz). The
doublet of doublets is assigned to 2; the chemical shift and
coupling constants compare well with similar compounds in
the literature.24a The triplet at 92.2 ppm is assigned to the
proposed intermediate, (dippe)Ni(SiCl3)H.
Figure 3. ORTEP drawing of (dippe)Ni(SiCl3)(Cl) (2). Ellipsoids
are shown at 50% probability. Hydrogens have been omitted for
clarity. Selected bond distances (A) and angles (deg): Ni(1)-P(3) =
˚
2.2171(8), Ni(1)-P(4) = 2.1433(8), Ni(1)-Cl(1) = 2.1851(8),
Ni(1)-Si(2) = 2.2390(8), Si(2)-Cl(21) = 2.1050(10), Si(2)-
Cl(22) = 2.0861(11), Si(2)-Cl(23) = 2.0935(11); Cl(1)-Ni(1)-
P(3)=91.31(3), Cl(1)-Ni(1)-Si(2)=85.08(3), Cl(1)-Ni(1)-P(4) =
172.87(4), Si(2)-Ni(1)-P(4) = 95.74(3), P(3)-Ni(1)-P(4) =
88.81(3), Si(2)-Ni(1)-P(3) = 171.42(3), Cl(21)-Si(2)-Cl(22) =
98.65(4), Cl(22)-Si(2)-Cl(23) = 104.81(5), Cl(23)-Ni(1)-
Cl(21) = 98.96(5).
the addition of Cp2FeþB(ArF)4-, a silylene complex showing
an agostic interaction with the silyl hydrogen.24c
There is another synthetic route for the preparation of 2
that is better understood and easier to reconcile. The reaction
of 1 with chlorobenzene affords (dippe)Ni(Ph)(Cl), 3, in
79% yield and releases dihydrogen (eq 3). This reaction has
previously been studied by Jones and co-workers.27c The
reaction of 3 with trichlorosilane yields 2 and PhSiCl3.
Phenyltrichlorosilane was observed as the major silicon-
containing product by GC-MS with phenyldichlorosilane
as the minor product. This observation supports a nickel-
silyl hydride intermediate (eq 4). The formation of PhSiCl3 is
accompanied by the formation of the intermediate, which
reacts further with SiHCl3 to give 2. The minor products,
(dippe)NiCl2 and PhSiHCl2, are formed by the direct sub-
stitution of a chloride for the phenyl group.
In an attempt to distinguish between the two mechanisms,
we investigated the reaction of triethylsilane-d1 (Et3SiD)
with precatalyst 1. We knew a priori that 1 does not catalyze
disilane (Et3Si-SiEt3) formation from triethylsilane. We
wanted to probe if triethylsilane would induce H2 or HD
evolution and formation of Ni(0). Such a reaction would
support the mechanism in Scheme 2. Neither H2 nor HD was
formed. Instead scrambling of Et3SiD with nickel hydrides
was observed, as evidenced by the appearance of Et3SiH
in the 1H NMR spectra. The source of the proton was
confirmed by the observation of Ni-D at -9.0 ppm by 2H
NMR. This result indicates that triethylsilane interacts with
the nickel center, but no hydrogen/deuterium atom from the
silane is eliminated as H2 or HD gas. The lack of dihydrogen
evolution and Si-D/H scambling can be taken in support of
σ-bond metathesis but, of course, not a proof. The mecha-
nistic evidence is not conclusive because Et3SiH is not coupled
by 1 to give Et3Si-SiEt3. Furthermore, considering ample
precedence for 1 acting as a source of Ni(0),27 the oxidative-
addition mechanism cannot be ruled out.
2.4. Synthesis and Characterization of (dippe)Ni(SiCl3)(Cl)
(2). Attempts to isolate or detect a nickel silyl complex under
catalytic conditions were not successful. Therefore, we
reacted 1 with trichlorosilane (Cl3SiH), a tertiary silane that
does not dimerize in the presence of 1. The reaction afforded
(dippe)Ni(SiCl3)(Cl) (2) (eq 2) as the major product in good
yields (61% isolated). The 31P spectrum is characteristic of
two different phosphorus environments: two doublets with
resonances in benzene-d6 at δ 76.1 ppm (JP-P = 25.0 Hz)
and 89.7 ppm (JP-P = 25.0 Hz). Minor products of this reaction
included (dippe)NiCl2 (31P NMR δ 87.2 ppm in benzene-d6).
Residual gas analysis (RGA) was used to observe dihydrogen
gas, which was evolved during the course of this reaction. RGA
(28) Zdilla, M. J.; Lee, A. Q.; Abu-Omar, M. M. Inorg. Chem. 2009,
48, 2260.

6532 Organometallics, Vol. 29, No. 23, 2010
Smith et al.
The molecular structure of complex 2 was solved by X-ray
crystallography. An ORTEP rendition is shown in Figure 3
alongside selected bond distances and angles, and crystal-
lographic data are tabulated in the Supporting Information.
The structure features square-planar nickel(II) with the
silicon of the silyl group in an approximate tetrahedral
Phenylsilane (666.6 μL, 5.408 mmol) was added by syringe
directly to Cp2Zr(CH3)2 (27.2 mg, 0.108 mmol, 2%). The result-
ing solution was yellow. The evolution of dihydrogen was
monitoredusinga gasevolution apparatus. The yieldofhydrogen
gas was 98%. The resulting product mixture was a reddish solid.
Phenylsilane (534.6 μL, 4.337 mmol) was added by syringe
directly to Ind2Zr(CH3)2 (30.5 mg, 0.0807 mmol, 2%). The
resulting solution was colorless. The evolution of dihydrogen
was monitored using a gas evolution apparatus. The yield of
hydrogen gas was 78%. The resulting product mixture was orange
and extremely viscous.
˚
geometry. The Ni-Si bond of 2.24 A is consistent with that
observed for other nickel silyl complexes.24
3. Conclusions
4.4. Gel Permeation Chromatography. Gel permeation chro-
matography was done using a Waters 2695 Separations Module,
equipped with a vacuum degasser, and a Waters 2410 differential
refractometer. Separation was done with Polymer Laboratories
PLgel 5 μm Mixed-C columns preceded by a PLgel 5 μm guard
column. Analyses were performed with certified grade THF as
eluent with a flow rate of 1.0 mL/min. The column and detector
were heated to 35 °C. Samples were prepared in THF at about
0.5% w/v solids and filtered through 0.45 μm PTFE syringe
filters. An injection volume of 100 μL was used, and data were
collected for 25min. Data collectionandanalyses wereperformed
using ThermoLabsystems Atlas chromatography software and
Polymer Laboratories Cirrus GPC software. Molecular weight
averages were determined relative to a third-order calibration
curve created using polystyrene standards of molecular weight
580-2 300 000. The precision and accuracy of the analysis of
these samples has not been established. In general, the repeat-
ability of conventional GPC can be expected to be (5-10%.
4.5. Reaction of 1 with Triethylsilane-d1 by 1H NMR. A
solution of triethylsilane-d1 (71.3 μL, 0.706 mmol) and diphenyl-
methane (3.7 μL, 0.023 mmol) internal standard was prepared in
500 μL of benzene-d6. A 1H NMR spectrum was collected, and no
signal was observed at 3.8 ppm, the expected chemical shift for
protio-triethylsilane. A solution of 1 (14.3 mg, 0.0222 mmol,
3.1 mol % loading) was prepared in 200 μL of benzene-d6. The
solution of 1 was added to the solution of triethylsilane-d1. After
10 min, 1H and 31P NMR spectra were collected. The growth of a
sextuplet at 3.8 ppm indicated scrambling of triethylsilane-d1
deuterium with a proton. The 31P NMR had only a singlet at
78 ppm, indicating that the catalyst was still in the form of 1.
4.6. Reaction of 1 with Triethylsilane-d1 by 2H NMR. A
solution of 1 (14.3 mg, 0.0222 mmol, 3.1 mol % loading) was
prepared in 700 μL of benzene. Triethylsilane-d1 (71.3 μL, 0.706
mmol) and benzene-d6 (10.5 μL, 1.18 mmol) were added to the
The dinuclear nickel hydride complex 1 produces poly-
silane with molecular weights and linear-to-cyclic ratios that
are comparable to those obtained using group 4 metallocene
catalysts. The dehydrocoupling reaction can proceed by
either σ-bond metathesis or an oxidative-addition mecha-
nism with a nickel chloro hydride intermediate. Compound 2,
(dippe)Ni(SiCl3)(Cl), preparedandisolatedviatwopathways,
represents a stable analogue of the hypothesized intermediate
and provides support for the nickel silyl intermediate in cata-
lytic dehydrocoupling of phenylsilane.
4. Experimental Methods
4.1. General Procedures and Methods. All reactions were
carried out under a nitrogen atmosphere using standard Schlenk
and drybox techniques. Solvents were distilled from sodium/
benzophenone ketyl and degassed. Deuterated NMR solvents
(Cambridge Isotope Laboratories) were stored over molecular
sieves. All glassware was dried in an oven at 120 °C. All reagents
were used as received from commercial sources. 1H and 31P
NMR spectra were collected using a Varian 300 MHz spectro-
2
meter. 29Si and H NMR spectra were collected using Bruker
DRX 500 MHz spectrometers. A gas evolution apparatus pur-
chased from ChemGlass was used for the measurement of the
volume of evolved hydrogen gas. Gas analysis was done with an
in-house built RGA equipped with a Varian model SH 100 vacuum
pump and a Stanford Research Systems RGA 100 mass spectro-
meter with an Alcatel ATH131 Series turbopump.28 Complex 1,
[(dippe)NiH]2, was prepared from the published procedure.27a
Cp2Zr(CH3)2 was purchased from Strem.
4.2. Monitoring the Oligomerization of Phenylsilane. The
kinetic data for the oligomerization of phenylsilane was obtained
by two methods. The first method used (plotted in Figure 1) was
NMR. Phenylsilane has a chemical shift of 4.2 ppm; the decrease
in area of this peak was quantified by integration against a diphe-
nylmethane internal standard (3.7 ppm). Peaks for oligomers
and polymers were observed below 4.8 ppm, and cyclic oligomers
were observed above 4.8 ppm.5m The second method used to
monitor the kinetics was measuring the volume of dihydrogen gas
evolved. This was done using a buret connected to an equalizing
bulbfilledwithmineraloil. The displacementofoilwithhydrogen
gas was used to determine the volume evolved. The data from the
gas evolution method are plotted in Figure 2.
4.3. Preparation of Samples for Gel Permeation Chromatog-
raphy. Compound 1 (23.4 mg, 0.0363 mmol, 2.0%) was dis-
solved in benzene (0.4 mL). Phenylsilane (223.9 μL, 1.82 mmol)
was also dissolved in toluene (0.176 mL) and added by syringe to
the solution. The evolution of dihydrogen was monitored using
a gas evolution apparatus. The yield of hydrogen gas was 81%
after several hours. The remaining solution was red-brown and
viscous.
Compound 1 (17.9 mg, 0.0278 mmol, 2%) was dissolved in
toluene (0.308 mL). This solution was added to phenylmethyl-
silane (192.8 μL, 1.31 mmol). The evolution of dihydrogen was
monitored using a gas evolution apparatus. The yield of hydro-
gen gas was 76%. The remaining solution was red-brown and
viscous.
2
solution. After 10 min, the H NMR spectrum was collected.
The appearance of a peak at -9.0 ppm indicated that the
deuterium from the triethylsilane-d1 was scrambling with the
hydride from 1.
4.7. Synthesis of (dippe)Ni(SiCl3)(Cl) (2). Complex 2 was
prepared using 1 (50.0 mg, 0.780 mmol) dissolved in benzene
(50 mL) and adding trichlorosilane (156.0 μL, 1.544 mmol). The
reaction was left to stir for two hours and then filtered through a
frit of medium porosity. The filtrate was collected, and the ben-
zene was removed under vacuum. The resulting yellow solid was
dried overnight under vacuum. Yield: 61%. Hydrogen gas evol-
ving from this reaction was detected by RGA-MS. NMR for 2 in
benzene-d6: 1H, δ 0.67 (dd, JHH = 12.6, 6.9 Hz, 6H, CH3), 0.76
(dd, JHH = 13.5, 6.9 Hz, 6H, CH3), 0.92 (m, 2H, P(CH2)2P), 1.17
(dd, JHH = 16.4, 7.1 Hz, 6H, CH3), 1.27 (dd, JHH = 18.6, 7.2 Hz,
6H, CH3), 1.40 (m, 2H, P(CH2)2P), 2.03 (m, 2H, CH), 2.67 (m,
2H, CH); 31P{1H}, δ 76.1 (d, JPP = 25.0 Hz), 89.7 (d, JPP = 25.0
Hz); 29Si{1H}, δ 16.5 (dd, JSiP = 79.4 and 303 Hz, SiCl3).
4.8. Synthesis of (dippe)Ni(Ph)Cl (3). Complex 3 was pre-
pared using 1 (50.0 mg, 0.78 mmol) dissolved in THF (30 mL)
and adding chlorobenzene (198 μL, 1.98 mmol, 25 times excess).
The solution was left to stir for 2 h and was reduced to 5 mL by
vacuum. A 30 mL amount of hexane was added, and the mixture
was filtered through a medium frit. An orange solid was isolated
and dried overnight under vacuum. Yield: 79%. NMR spectra

Article
Organometallics, Vol. 29, No. 23, 2010 6533
for 3 in THF-d8: 1H, δ 1.04 (dd, JHH = 15.3, 7.2 Hz, 6H, CH3),
1.20 (dd, JHH = 14.0, 6.9 Hz, 6H, CH3), 1.29 (dd, JHH = 12.5,
7.1 Hz, 6H, CH3), 1.48 (dd, J = 15.5, 7.1 Hz, 6H, CH3), 1.57 (m,
2H, P(CH2)2P), 1.82 (m, 2H, P(CH2)2P), 2.16 (m, 2H, CH), 2.33
˚
a = 8.8414(4) A, b = 13.6132(5) A, c = 18.7005(8) A, V =
2250.79(16) A . For Z = 4 and fw = 490.97 the calculated
˚
˚
3
˚
density is 1.45 g/cm3. The refined mosaicity from DENZO/
SCALEPACK29 was 0.40°, indicating good crystal quality. The
space group was determined to be P212121 (#19) by the program
XPREP.30 The data were collected at a temperature of 150(1) K
to a maximum 2θ of 54.9°.
(m, 2H, CH), 6.64 (t, JHH = 7.5 Hz, 1H, p-H), 6.79 (dt, JHH
=
1.5, 7.8 Hz, 2H, m-H), 7.40 (t, JHH = 6.9 Hz, 2H, o-H). 31P{1H},
δ 72.2 (d, JPP = 20.3 Hz), 75.4 (d, JPP = 20.4 Hz). This
procedure was modified from the synthetic method for 3 pub-
lished by Jones and co-workers.27
Frames were integrated with DENZO-SMN.29 Lorentz and
polarization corrections were applied to the data. The linear
absorption coefficient is 15.3/mm for Mo KR radiation. An
empirical absorption correction using SCALEPACK29 was
applied. The crystallographic data are provided in the Support-
ing Information. The structure was solved by direct methods
using SIR2004.31 The remaining atoms were located in succes-
sive difference Fourier syntheses. Scattering factors were taken
from the International Tables for Crystallography.32 Refine-
ment was performed on a LINUX PC using SHELX-97.30
Crystallographic drawings were done using the programs
ORTEP33 and PLUTON.34
4.9. Preparation of 2 from 3. Complex 2 was prepared using
3 (15.0 mg, 0.03459 mmol) dissolved in benzene (20 mL).
Trichlorosilane (5.2 μL, 0.05146 mmol) was added to the solu-
tion. The solution was left to stir for 2 h and was filtered through
a frit of medium porosity. The filtrate was collected, and the
benzene was removed under vacuum. The resulting yellow solid
was dried overnight under vacuum. Yield: 61%.
5.0. Crystallographic Studies. Orange needle crystals of 2
were obtained by solvent diffusion in toluene with hexane for
the diffusing solvent after several days at -35 °C in a glovebox. A
crystal was mounted on a fiber in a random orientation. Pre-
liminary examination and data collection were performed with
Acknowledgment. Thank you to the U.S. Department
of Energy for financial support (grant no. DE-FG-02-
06ER15194) and to the Dow Corning Corporation for
both GPC analysis and financial support.
˚
Mo KR radiation (λ = 0.71073 A) on a Nonius KappaCCD
equipped with a graphite crystal, incident beam monochromator.
The orthorhombic cell parameters and calculated volume were
(29) Otwinowski, Z.; Minor, W. Methods Enzymol. 1997, 276, 307.
(30) Sheldrick, G. M. Acta Crystallogr. 2008, A64, 112.
(31) Burla, M. C.; Caliandro, R.; Camalli, M.; Carrozzini, B.;
Cascarano, G. L.; De Caro, L.; Giacovazzo, C.; Polidori, G.; Spagna,
R. J. Appl. Crystallogr. 2005, 38, 381.
(32) International Tables for Crystallography; Kluwer Academic Pub-
lishers: Utrecht, The Netherlands, 1992; Vol. C, Tables 4.2.6.8 and 6.1.1.4.
(33) Johnson, C. K. ORTEPII, Report ORNL-5138; Oak Ridge
National Laboratory: Oak Ridge, TN, USA, 1976.
Supporting Information Available: Crystallographic data
(CIF), table of crystallographic data, and GPC for nickel and
zirconium catalysts. This material is available free of charge via
the Internet at http://pubs.acs.org.
(34) Spek, A. L. PLUTON, Molecular Graphics Program; University of
Utrecht: The Netherland, 1991.