Aminoalkyl-substituted indenylnickel(II) complexes (η3:η0-Ind(CH2)2 NMe2)Ni(PR3)X and [(η3

Article abstract of DOI:10.1021/om0341348

Reaction of LiInd(CH₂)₂NMe₂ with (PR₃)₂NiCl₂ gave the neutral complexes (η³:η⁰-Ind(CH₂)₂ NMe₂)Ni(PCy₃)Cl (3), was obtained

Full text of DOI:10.1021/om0341348

Organometallics 2003, 22, 4759-4769
4759
Am in oa lk yl-Su bstitu ted In d en yln ick el(II) Com p lexes
(η3:η0-In d (CH2)2NMe2)Ni(P R3)X a n d
[(η3:η1-In d (CH2)2NMe2)Ni(P R3)][BP h 4]: In flu en ce of th e
P h osp h in e in Liga n d Exch a n ge a n d P olym er iza tion
Rea ction s
Laurent F. Groux and Davit Zargarian*
De´partement de Chimie, Universite´ de Montre´al, Montre´al (Que´bec), Canada H3C 3J 7
Received August 26, 2003
Reaction of LiInd(CH2)2NMe2 with (PR3)2NiCl2 gave the neutral complexes (η3:η0-Ind(CH2)2-
NMe2)Ni(PR3)Cl (R ) Ph (1) or Me (2)), while the PCy3 analogue, (η3:η0-Ind(CH2)2NMe2)-
Ni(PCy3)Cl (3), was obtained by reacting 1 with PCy3. These Ni-Cl species react with R′Li
or NaBPh4 to form, respectively, the corresponding Ni-R′ derivatives (η3:η0-Ind(CH2)2NMe2)-
Ni(PR3)R′ (R ) Ph, R′ ) Me (4) or CCPh (5); R ) R′ ) Me (6)) or the cationic species [(η3:
η1-Ind(CH2)2NMe2)Ni(PR3)]+ (R ) Ph (7), Me (8), or Cy (9)), in which the NMe2 moiety is
coordinated to the nickel center. These complexes have been fully characterized, including
solid state structure determinations by X-ray crystallography for complexes 2, 4, 5, 6, and
9. Inspection of the structural data showed that replacing the Cl ligand by the more strongly
donating ligands CCPh and Me reinforces the Ni-P and Ni-Ind interactions. On the other
hand, electrochemical measurements showed that the reduction potentials of the Ni-Cl
compounds are intermediate between the Ni-R′ derivatives, which are more resistant to
reduction, and the cationic species, which are the easiest to reduce. The cationic complexes
are single-component catalysts for the polymerization of styrene, giving poly(styrene) with
Mw in the range of 104-105, and the hydrosilylation of styrene and 1-hexene with PhSiH3
and Ph2SiH2. The nature of the phosphine ligand has an important influence on the catalytic
reactivities, the PMe3 analogue 8 being the most active catalyst.
Sch em e 1
In tr od u ction
An important class of hemilabile ligands consists of
functionalized cyclopentadienyl ligands represented by
∧
Cp∧L, with denoting the side chain linking the Cp
ligand to a functional group L such as NRR′,1 OR,2
PRR′,3 SR,3 AsRR′,3 and CdC.4 The main role of the Cp
moiety is to anchor these multidentate ligands to metal
centers, while the reversible coordination of L modulates
the reactivities of the metal center. In principle, sub-
strates can displace the hemilabile group L from the
metal center in order to initiate reactivity; on the other
hand, since L is never far from the metal, it can
re-coordinate readily in the absence of substrate to
prevent the decomposition of the catalyst (Scheme 1).
The potential of this class of compounds in catalysis has
spurred research efforts in this area and resulted in the
preparation of many transition metal complexes bearing
Cp∧L type ligands or their indenyl analogues. Examina-
tion of the reactivities of some of these complexes has
demonstrated the dramatic influence of the hemilabile
ligands in improving catalytic reactivities, stabilizing
otherwise unstable species, changing solubility proper-
ties, introducing chirality, and so on.5
We became interested in this area of research during
our investigations on the catalytic reactivities of inde-
nyl-nickel(II) complexes.6 What inspired us to examine
the influence of a hemilabile moiety in our complexes
was the observation that the highly reactive, in situ
generated cationic species [IndNi(PR3)]+ (Ind ) indenyl
and its substituted derivatives) are rapidly converted
to the inactive compounds [IndNi(PR3)2]+ in the absence
of substrates. The presence of this deactivation pathway
meant that these catalysts had to be generated in situ
and in the presence of a large excess of substrate;
otherwise, the formation of the bis(phosphine) deriva-
tives would inhibit the catalysis. We reasoned that the
incorporation of a hemilabile moiety in the vicinity of
the Ni center might circumvent catalyst deactivation,
thereby improving catalyst lifetimes.
(1) Review on Cp∧NRR′ ligands: J utzi, P. Eur. J . Inorg. Chem. 1998,
663.
(2) Review on Cp∧OR ligands: Siemeling, U. Chem. Rev. 2000, 100,
1495.
(5) For reviews of this topic see: (a) Mu¨ller, C.; Vos, D.; J utzi, P. J .
Organomet. Chem. 2000, 600, 127. (b) J utzi, P.; Redeker, T. Eur. J .
Inorg. Chem. 1998, 663. (c) J utzi, P.; Siemeling, U. J . Organomet.
Chem. 1995, 500, 175. (d) J utzi, P.; Dahlaus, J . Coord. Chem. Rev.
1994, 137, 179.
(3) Review on Cp∧L (L ) PRR′, AsRR′, and SR) ligands: Buten-
schoen, H. Chem. Rev. 2000, 100, 1527.
(4) General review on functionalized Cp ligands: Mu¨ller, C.; Vos,
D.; J utzi, P. J . Organomet. Chem. 2000, 600, 127.
(6) Zargarian, D. Coord. Chem. Rev. 2002, 233-234, 157.
10.1021/om0341348 CCC: $25.00 © 2003 American Chemical Society
Publication on Web 10/09/2003

4760 Organometallics, Vol. 22, No. 23, 2003
Groux and Zargarian
a
Sch em e 2
This assertion was borne out by the results of studies
on the influence of an amino tether on the reactivity of
the complexes.7 Thus, we found that the cationic com-
plexes [(η3:η1-Ind∧NR2)Ni(PPh3)]+ were active, single-
component catalysts (i.e., no activation or in situ
generation needed) for polymerization of styrene and
norbornene. Interestingly, the Mw and solubilities of the
products obtained from these reactions were different
from those of the products obtained from reactions
promoted by the in situ generated [(Ind)Ni(PPh3)]+,
implying that the hemilabile moiety might also influ-
ence the course of the catalysis.7b Isolation and complete
characterization of the new complexes, as well as a
study of their ligand exchange reactions, allowed an
evaluation of the NfNi binding as a function of the
incoming ligand’s nucleophilicity7a and different amine
substituents.7d
As a follow up to our previous studies, we have
prepared analogous cationic compounds with PMe3 and
PCy3 instead of PPh3 in order to examine the influence
of the phosphine ligand on the binding of the tether to
the Ni center and on the reactivities of these complexes.
The present report describes the preparation and char-
acterization of the chloro derivatives (η3:η0-Ind(CH2)2-
NMe2)Ni(PR3)Cl (R ) Me (2) and Cy (3)), the cationic
complexes [(η3:η1-Ind(CH2)2NMe2)Ni(PR3)]+ (R ) Me (8)
and Cy (9)), and (η3:η0-Ind(CH2)2NMe2)Ni(PR3)R′ (R )
Ph, R′ ) Me (4) and CCPh (5); R ) R′ ) Me (6)). The
catalytic reactivities of these compounds in the polym-
erization of styrene and phenylacatylene, oligomeriza-
tion of PhSiH3, and the hydrosilylation of styrene and
1-hexene are also reported herein.
a
a ) LiInd(CH2)2NMe2; b ) LiMe; c ) LiCCPh; d ) PCy3;
e ) NaBPh4; f ) HCl; g ) HBF4.
Previous studies had shown that the complexes (Ind∧-
NR2)PPh3Ni(Cl) display variable degrees of dynamic
NfNi binding: temperature-dependent H and 13C{1H}
1
NMR spectra were obtained, with the room-temperature
1H NMR spectra displaying very broad signals, while
the corresponding 13C{1H} NMR spectra contained few
of the expected resonances. In the case of complexes 2
and 3, however, no signal broadening was observed in
the H NMR spectra and all the anticipated 13C NMR
1
resonances were accounted for. Moreover, the N(CH3)2
groups in these complexes were found to be equivalent,
which indicates that the amine moiety was not coordi-
nating to the Ni center. (Since these complexes have C1
symmetry, NfNi coordination would be expected to
render the Me groups chemically inequivalent.) There-
fore, we conclude that complexes 2 and 3 do not undergo
a dynamic exchange process involving amine chelation,
in contrast to their previously studied PPh3 analogues.
This difference is presumably related to the higher
electron density of the Ni center in 2 and 3, which is in
turn brought about by the more strongly donating
phosphines PMe3 and PCy3. This issue will be addressed
in the next section, along with the solid state structure
of 2.
The chloro complexes 1-3 have been used to prepare
the new Ni-Me and Ni-CC-Ph derivatives (η3:η0-
Ind(CH2)2NMe2)Ni(PR3)R′ (R ) Ph, R′ ) Me (4) and
CCPh (5); R ) R′ ) Me (6)) and the chelated cations
[(η3:η1-Ind(CH2)2NMe2)Ni(PR3)]+ (R ) Me (8) and Cy
(9)). The main motivation for preparing the neutral,
nonchelating derivatives was an earlier observation that
seemed to indicate that even when the amine moiety
in the precatalysts is not chelated, its proximity to the
Ni center seems to have a marked influence over the
course of the catalytic reactions and the nature of their
products. Thus, the Ni-Me derivatives 4 and 6 were
obtained by reacting MeLi with 1 or 2, respectively,
while the Ni-CCPh derivative 5 was prepared in a
similar manner by the metathetic reaction between 1
and LiCCPh (Scheme 2); recrystallization of the crude
products from hexane gave pure compounds. Reacting
these compounds with HCl gives back the Ni-Cl
precursors.
Resu lts a n d Discu ssion
Syn th esis a n d Sp ectr oscop ic Ch a r a cter iza tion .
The previously reported7a preparation of (η3:η0-Ind(CH2)2-
NMe2)(PPh3)Ni-Cl, 1, served as a model for the syn-
thesis of (η3:η0-Ind(CH2)2NMe2)(PMe3)Ni-Cl, 2. Thus,
slow addition of 1 equiv of LiInd(CH2)2NMe2 to 1 equiv
of Ni(PMe3)2Cl2 in THF gave a dark red solution, from
which pure 2 precipitated gradually. On the other hand,
the PCy3 analogue (η3:η0-Ind(CH2)2NMe2)(PCy3)Ni-Cl,
3, was readily obtained by adding 1.5 equiv of PCy3 to
a solution of 1 in Et2O (Scheme 2).
The new complexes have been fully characterized by
NMR spectroscopy (1H, 13C, and 31P), elemental analy-
sis, and in the case of 2 X-ray diffraction studies. The
NMR spectra of complexes 2 and 3 display the charac-
teristic signals observed for the analogous nonfunction-
alized compounds (1-Me-Ind)Ni(PR3)Cl (R ) Ph, Me,
Cy).8 For instance, the 31P{1H} NMR spectra showed
singlets for the phosphine ligands at -11 ppm for 2 (cf.
-10.6 ppm for its 1-Me-Ind analogue) and 37.0 ppm for
3 (cf. to 37.2 ppm for its 1-Me-Ind analogue). The 1H
and 13C{1H} NMR spectra also served to establish the
degree of NfNi interaction in these complexes, as
described below.
(7) (a) Groux, L. F.; Be´langer-Garie´py, F.; Zargarian, D.; Vollmer-
haus, R. Organometallics 2000, 19, 1507. (b) Groux, L. F.; Zargarian,
D. Organometallics 2001, 20, 3811. (c) Groux, L. F.; Zargarian, D.;
Simon, L. C.; Soares, J . B. P. J . Mol. Catal. A: Chem. 2003, 193, 51.
(d) Groux, L. F.; Zargarian, D. Organometallics 2003, 22, 3124.
(8) (a) Huber, T. A.; Bayrakdarian, M.; Dion, S.; Dubuc, I.; Be´langer-
Garie´py, F.; Zargarian, D. Organometallics 1997, 16, 5811. (b) Fon-
taine, F.-G.; Dubois, M.-A.; Zargarian, D. Organometallics 2001, 20,
5156.
The new complexes were characterized by NMR
spectroscopy, and their solid state structures were
determined by X-ray crystallography.9 For instance, the

Aminoalkyl-Substituted Indenylnickel(II) Complexes
Organometallics, Vol. 22, No. 23, 2003 4761
Sch em e 3
interactions, and we have seen that complex 1 can adopt
one or the other structure depending on crystallization
conditions.12
The Ni-Ind interaction is fairly symmetrical (Ni-C1
≈ Ni-C3; Ni-C3a ≈ Ni-C7a) in the Ni-Me complexes
4 and 6, quite unsymmetrical in 2 (Ni-C1 > Ni-C3),
and somewhat unsymmetrical in 5 (Ni-C1 > Ni-C3
by greater than 13 esd) and 9 (Ni-C1 < Ni-C3 by about
15 esd). As described in detail elsewhere,6 these obser-
vations can be attributed to the relative trans influences
of the PR3 and X ligands in the neutral complexes (X )
Cl, Me, CCPh, etc.) or the geometrical constraints
imposed by the chelation in the cationic species. The
Ind hapticity, as measured by the slip parameter ∆(M-
C),13 seems to vary as a function of PR3 basicity, showing
a higher Ind hapticity (i.e., smaller ∆(M-C)) in 1 (0.23
Å)7a versus 2 (0.32 Å), 4 (0.18 Å) versus 6 (0.20 Å), and
7 (0.26 Å) versus 9 (0.32 Å). On the other hand, for
complexes having the same phosphine ligand, the Ind
hapticity increases (i.e., ∆(M-C) decreases) with the
stronger basicity of the X ligand, as follows: 1 < 5 < 4;
2 < 6.
The Ni-alkynyl bond length in 5 (1.852(4) Å) is much
shorter than the Ni-Me distances in 4 (1.9508(17) Å)
and 6 (1.983(4) Å), presumably because of the greater
sp character of the alkynyl carbon. On the other hand,
the Ni-P distances are shortest in the Ni-Me com-
plexes (ca. 2.13 Å) compared to the Ni-Cl and Ni-CCPh
complexes (ca. 2.16 Å) or the cationic complex 9 (ca. 2.24
Å). This observation prompted us to compare the Ni-
PPh3 bond lengths as a function of the ligand X in the
complexes IndNi(PPh3)X, many of which have been
characterized structurally.6 This comparison confirmed
a trend in the Ni-P bond distances, which are shorter
in the Ni-alkyl derivatives (ca. 2.12-2.13 Å), followed
by derivatives of other anionic ligands such as chloro,
alkynyl, phthalimidato, and thienyl (ca. 2.16-2.19 Å),6
and the cations featuring the chelating amino moieties
(ca. 2.20 in 77b and ca. 2.22 Å in [{η3:η1-IndCH2(2-
pyridine)}Ni(PPh3)]+).7d A similar trend is noted for the
Ni-Ind interactions, which are stronger (i.e., the slip
parameter ∆(M-C) is smaller) with the Ni-Me deriva-
tive 4, followed by the Ni-Cl derivative 1 and the
cationic 7.
31P{1H} NMR spectra showed singlets for the phosphine
ligands at 46.9 ppm for 4 (cf. 47.7 ppm for its 1-Me-Ind
analogue),8a 38.8 ppm for 5 (cf. 40.4 for its 1-Me-Ind
analogue),10 and -4.0 ppm for 6 (cf. -3.7 ppm for its
1-Me-Ind analogue).11 In addition, the characteristic
doublet resonances for the Ni-CH3 moieties in 4 and 6
3
were observed at -0.65 ppm (4, J P-H ) 5.6 Hz) and
3
1
-0.73 ppm (6, J P-H ) 6.3 Hz) in the H NMR spectra
2
and -18.3 ppm (4, J P-C ) 24.8 Hz) and -21.7 ppm (6,
2J P-C ) 25.2 Hz) in the 13C{1H} NMR spectra. As before,
the equivalence of the N(CH3)2 groups confirmed the
absence of any Ni-N interactions. On the other hand,
the absorption signal for ν(CC) in the IR spectrum of 5
appeared at 2099 cm-1, very close to the corresponding
signals in the 1-Me-Ind analogue (2090 cm-1 10
) and free
phenylacetylene (2110 cm-1), implying little or no Ni-
CCPh back-bonding in these complexes. The results of
the X-ray diffraction studies will be discussed in the
next section.
The cationic complexes 8 and 9 were prepared in
analogy with their PPh3 analogue, 7,7b by reacting the
chloro precursors with NaBPh4 or the Ni-Me precursor
6 with HBF4 (Scheme 2); these new complexes were
purified by multiple recrystallizations from Et2O/CH2-
Cl2 mixtures. The 31P{1H} NMR spectra of these com-
pounds displayed one new singlet resonance at ca. -20.8
ppm for 8 and ca. 24.9 ppm for 9; the absence of AA′
doublet resonances in these spectra indicated that
formation of the corresponding bis-phosphine complexes
had been circumvented by the chelation of the amine
moiety. On the other hand, the NfNi chelation was
supported by the observed inequivalence of IndCH2CH2N-
(CH3)2 signals and confirmed by the solid structure of
9, which is discussed below.
Solid State Str u ctu r es an d Electr och em ical Stu d-
ies. Suitable crystals for X-ray diffraction studies were
obtained for 2, 4, 5, 6, and 9, as described in the
Experimental Section, and studied at 223 K. The details
of data collection and the structure refinement param-
eters are listed in Table 1, while bond distances and
angles are reported in Table 2. The overall geometry in
all of the complexes studied can be described as dis-
torted square planar, with the Ind moiety occupying two
coordination sites and the largest distortion arising from
the small C1-Ni-C3 angle of ca. 67°.6 In complexes 2,
4, 5, and 6, no Ni-N interaction is detected: the NMe2
moiety is pointed away from the Ni, extending along
the plane of coordination (as in 2, 4, and 6) or almost
perpendicular to it (as in 5). The different orientation
of the amine moiety is likely a result of packing
In the search for a relationship between the above-
noted structural trends and the relative electron rich-
ness of these complexes, we undertook electrochemical
studies, with the following results. Cyclic voltammetry
measurements of complexes 1-9 showed that they
undergo irreversible reductions at potentials ranging
from -1.16 to -2.33 V (vs SCE), as reported in Table
3. Inspection of the electrochemical data shows the
following:
(a) As expected, the cationic species are more easily
reduced than the neutral complexes.
(b) The Ered values for the cationic species 8 (-1.22
V) and 7 (-1.16 V) follow the donor ability of the
phosphine (PMe3 > PPh3), but the reduction potential
(12) Groux, L. F.; Zargarian, D. Acta Crystallogr. 2001, E57, m547.
(13) The slip parameter, ∆(M-C), is determined according to the
relationship (M-C3a + M-C7a)/2 - (M-C1 + M-C3)/2. Thus, a ∆-
(M-C) value of 0 would signal a perfectly η5 coordination of Ind,
whereas increasing degrees of slippage toward η3 coordination would
result in larger values: (a) Baker, R. T.; Tulip, T. H. Organometallics
1986, 5, 839. (b) Westcott, S. A.; Kakkar, A.; Stringer, G.; Taylor, N.
J .; Marder, T. B. J . Organomet. Chem. 1990, 394, 777.
(9) It should be noted, however, that the results of combustion
analysis for complex 6 were not acceptable, leading us to suspect that
this compound is thermally unstable with respect to the reductive
coupling of the Ind∧NMe2 and the methyl ligands.
(10) Wang, R.; Be´langer-Garie´py, F.; Zargarian, D. Organometallics
1999, 18, 5548.
(11) Fontaine, F.-G.; Zargarian, D. Organometallics 2002, 21, 401.

4762 Organometallics, Vol. 22, No. 23, 2003
Groux and Zargarian
Ta ble 1. Cr ysta l Da ta , Da ta Collection , a n d Str u ctu r e Refin em en t P a r a m eter s
2
4
5
6
9
formula
C
16H25NPNiCl
C32H34NPNi‚Hex
dark red
block
C39H36NPNi‚THF
dark red
block
C17H28NPNi
dark red
needle
C55H69NBPNi‚CH2Cl2
dark red
block
cryst color
cryst habit
cryst dimens, mm
symmetry
space group
a, Å
dark red
needle
0.78 × 0.13 × 0.07
monoclinic
P21/n
6.3037(2)
15.6826(5)
18.2702(6)
90
96.106(2)
90
1795.92(10)
4
1.3185
Bruker AXS
SMART 2K
223(2)
0.96 × 0.40 × 0.26
0.46 × 0.24 × 0.13
0.84 × 0.08 × 0.08
0.17 × 0.20 × 0.20
triclinic
triclinic
triclinic
triclinic
P1h
P1h
9.998(3)
13.887(4)
14.716(5)
111.58(3)
99.68(3)
99.55(3)
1814.2(10)
2
1.2349
Nonius CAD-4
P1h
P1h
9.3419(3)
12.8669(5)
13.9322(5)
98.748(3)
103.145(2)
106.493(3)
1521.01(10)
2
6.4485(2)
8.5720(3)
16.7049(5)
84.18(3)
80.64(3)
85.70(3)
904.77(5)
2
10.9625(1)
13.4241(2)
17.6136(2)
83.939(1)
87.023(1)
78.308(1)
2522.84(5)
2
b, Å
c, Å
R, deg
â, deg
γ, deg
volume, Å3
Z
D(calcd), g cm-1
diffractometer
1.2345
1.234
1.224
Bruker AXS
SMART 2K
223(2)
Bruker AXS
SMART 2K
223(2)
Bruker AXS
SMART 2K
223(2)
temp, K
223(2)
λ (Cu KR), Å
1.54178
3.677
ω scan
72.75
-6 e h e 7
-19 e k e 19
-22 e l e 22
2927
multiscan
SADABS
0.398, 0.773
0.0363, 0.0967
1.003
1.54178
1.576
ω scan
72.64
-11 e h e 11
-15 e k e 14
-17 e l e 17
5440
multiscan
SADABS
0.389, 0.662
0.0411, 0.1153
1.086
1.54178
1.489
ω/2θ scan
69.93
-12 e h e 12
-16 e k e 16
-17 e l e 17
4139
integration
ABSORB
0.5977, 0.8501
0.0428, 0.0948
1.014
1.54178
2.290
ω scan
72.62
-7 e h e 7
-10 e k e 10
-20 e l e 20
3405
multiscan
SADABS
0.464, 0.833
0.0751, 0.1961
1.005
1.54178
2.094
ω scan
72.95
-13 e h e 13
-16 e k e 16
-21 e l e 21
7639
multiscan
SADABS
0.620, 0.700
0.0441, 0.1268
1.011
µ, mm-1
scan type
θ
max (deg)
h,k,l range
reflns used (I > 2σ(I))
abs correction
T (min, max)
R[F2>2σ(F2)], wR(F2)
GOF
Ta ble 2. Selected Bon d Dista n ces (Å) a n d An gles (d eg) for 2, 4, 5, 6, a n d 9
2
4
5
6
9
Ni-P
2.1608(6)
2.1868(6)
2.1184(19)
2.0389(19)
2.0344(19)
2.3817(19)
2.4035(18)
1.406(3)
1.419(3)
1.468(3)
1.417(3)
1.467(2)
1.495(3)
0.32
97.78(5)
95.86(2)
99.48(6)
66.92(8)
163.78(6)
166.36(5)
113.42(15)
110.75(16)
12.5(2)
2.1277(5)
1.9508(17)
2.0870(17)
2.0810(17)
2.1012(16)
2.2659(15)
2.2811(16)
1.421(2)
1.409(2)
1.441(2)
1.429(2)
1.456(2)
1.503(2)
0.18
94.39(7)
93.23(6)
107.25(5)
66.68(6)
157.91(8)
168.83(5)
113.53(14)
110.89(14)
8.2(2)
2.1573(11)
1.852(4)
2.092(2)
2.061(3)
2.051(3)
2.283(3)
2.280(3)
1.413(3)
1.395(4)
1.454(4)
1.428(4)
1.445(4)
1.493(3)
0.21
94.02(14)
95.64(12)
103.60(8)
66.88(11)
160.71(14)
169.81(8)
115.4(3)
116.5(2)
9.6(3)
2.1290(12)
1.983(4)
2.100(3)
2.090(4)
2.081(4)
2.285(3)
2.294(3)
1.418(5)
1.420(6)
1.445(6)
1.430(5)
1.467(5)
1.516(5)
0.20
96.90(15)
93.18(11)
103.05(12)
66.66(15)
162.69(16)
169.59(11)
113.0(3)
111.8(3)
9.4(2)
2.2424(5)
2.022(5)
2.0660(18)
2.0466(19)
2.096(2)
2.4172(19)
2.3932(19)
1.419(3)
1.408(3)
1.463(3)
1.413(3)
1.461(3)
1.503(5)
0.32
82.70(17)
111.11(16)
100.45(6)
66.42(8)
148.08(18)
165.24(6)
109.8(4)
107.4(4)
12.6(2)
Ni-Xa
Ni-C1
Ni-C2
Ni-C3
Ni-C3A
Ni-C7A
C1-C2
C2-C3
C3-C3A
C3A-C7A
C7A-C1
C1-C8
∆(M-C)b
C1-Ni-Xa
P-Ni-Xa
P-Ni-C3
C1-Ni-C3
C3-Ni-Xa
P-Ni-C1
N-C9-C8
C1-C8-C9
HAc
FAd
13.2(2)
7.5(2)
9.2(3)
7.3(2)
13.2(2)
a
b
X ) Cl, C12, or N. ∆(M-C) ) {Ni-Cav (for C7a and C3a) - Ni-Cav (for C1 and C3}. c HA ) angle between C1/C2/C3 plane and
d
C1/C3/C3A/C7A plane. FA ) angle between C1/C2/C3 plane and C3A/C4/C5/C6/C7/C7A plane.
of the PCy3 analogue 9 is only slightly more negative
(-1.17 V) than that of the PPh3 analogue; this might
be due to the greater steric volume of PCy3 that results
in a very long Ni-P bond and, presumably, a less
effective electron donation.
(c) The Ni-Me derivatives have more negative reduc-
tion potentials than their Ni-Cl counterparts: -2.33 V
(4) versus -1.41 V (1); -2.12 V (6) versus -1.27 V (2).
The Ered value for the Ni-CCPh derivative 5 (-1.72 V)
is intermediate between those of its chloro and cationic
analogues.
Therefore, the electrochemical measurements are in
fairly good agreement with the solid state data and
signal a correlation between the apparent electron-
richness of the Ni center and its interactions with the
phosphine and Ind ligands.14 On a first approximation,
the above observations indicate that the most electron-
rich Ni centers (i.e., those with the strongly donating
alkyl ligands) form the shortest Ni-P bonds and the
strongest Ni-Ind interactions, whereas the longest
Ni-P bonds and weakest Ni-Ind interactions occur
with the least electron rich centers (i.e., the cationic

Aminoalkyl-Substituted Indenylnickel(II) Complexes
Organometallics, Vol. 22, No. 23, 2003 4763
F igu r e 3. ORTEP plot of 5. Hydrogen and solvent atoms
are omitted for clarity.
F igu r e 1. ORTEP plot of 2. Hydrogen atoms are omitted
for clarity.
F igu r e 4. ORTEP plot of 6. Hydrogen atoms are omitted
for clarity except on the Ni-Me.
F igu r e 2. ORTEP plot of 4. Hydrogen (except on the Ni-
Me) and solvent atoms are omitted for clarity.
Ta ble 3. Ch a r a cter iza tion of Com p lexes 1-9
1H (ppm)
31P{1H}
(ppm)
ERed (V vs
SCE)
a
H2
H3
H4
ERed (V obs)
1b
2b
3b
4b
5c
6b
7c
8d
9d
30.8
-11.0
37.0
46.9
38.8
-4.0
29.1
-20.8
24.9
6.70 3.42 6.11
6.57 3.66 6.59
6.76 4.17
6.39 4.22 6.54
6.45 3.97 6.14
6.24 4.47 6.99
6.78 3.94 5.54
6.79 4.29 6.78
7.15 4.41 6.79
-1.70
-1.56
-1.96
-2.62
-2.01
-2.41
-1.45
-1.51
-1.46
-1.41
-1.27
-1.67
-2.33
-1.72
-2.12
-1.16
-1.22
-1.17
a
b
d
CH3CN. C6D6. c CDCl3. CD2Cl2.
species). This correlation is evident from a graph of ∆-
(M-C) values and Ni-P distances against the Ered
values (Figure 6). We have also noted an empirical
correlation between the two structural parameters
F igu r e 5. ORTEP plot of 9 with the major orientation of
the tether. Hydrogen atoms, counterion, other orientation
of the tether, and disordered CH2Cl2 are omitted for clarity.
(14) Strictly speaking, the reduction potential of an irreversible
reduction (or oxidation) process might vary as a function of many
different parameters such as electrolyte composition and nature of
electrode. It is common practice, however, to consider that in situations
where an analogous series of compounds are being studied under
identical conditions the values of reduction potential can be related,
to a good approximation, to structural or electronic properties of the
compounds. In the present case, since we are dealing with a family of
complexes possessing very similar structural features and composi-
tions, and since the electrochemical studies were performed under
identical conditions, we believe that the reduction potential values
represent fairly accurately the electron richness of the metal centers
in this family of complexes.
(Ni-P distances and ∆(M-C) values) and the reduction
potentials of these complexes, on one hand, and their
31P chemical shifts, on the other (Table 3). Thus, the
Ni-Me derivatives and the cations show the most
downfield and upfield shifts, respectively, whereas the
Ni-Cl and Ni-CCPh deriviatves have intermediate
shifts, as follows (δ values given in ppm): for PMe3
complexes, 6 (-4.0) > 2 (-11.0) > 8 (-20.8); for PCy3

4764 Organometallics, Vol. 22, No. 23, 2003
Groux and Zargarian
Ta ble 4. P olym er iza tion Exp er im en ts
monomer
equiv
T
time
Mw
cat.
equiv (°C) (days) (10-3
)
PDI TONc
0
1
2
3
4
5
6
7
8
9
7a
7a
8
8
9
styrene
styrene
styrene
styrene
styrene
styrene
2000 20
2000 80
2000 20
2000 80
2000 20
2000 80
100 20
100 20
100 20
100 20
100 20
7
2
2
2
2
2
1
1
1
1
1
77.3
38.1
243.4 4.0
11.6 1.8
147.9 4.2
3.2
2.9
364
23
630
57
350
9
7, 8, or 9 PhCCH
7/MAOb PhCCH
8/MAOb PhCCH
no reaction
34.5
57.7
41.5
3.4
3.2 <10
1.8 <10
1.8 <10
1.5 <10
10 9/MAOb PhCCH
11 5/MAOb PhCCH
a
b
See ref 7b. [Al]/[Ni] ) 10. c Based on isolated yield.
ment is generally governed by an equilibrium process
of which the Keq depends on the relative nucleophilicities
of the incoming ligand and the tethered amine. In
addition, the activities of the cations in the polymeri-
zation of styrene were found to correlate with the
lability of the NfNi bond, more labile cases showing
higher catalytic activities. (One exception was the
complex bearing a pyridine moiety wherein NifN π
back-bonding gave a stronger NfNi interaction while
displaying higher catalytic activity.) In the present
study, we have performed similar studies in order to
evaluate the influence of different phosphines on the
lability of the Ni-N bonds and the relative effectiveness
of these complexes in catalysis.
Initial tests showed that even relatively weak ligands
such as styrene and norbornene displace the chelating
NMe2 moiety in 8 and 9. To compare the Ni-N bonding
strength in these and the previously studied complexes,
we reacted 8 and 9 with increasing portions of pyridine
(py) and monitored the ensuing equilibria by NMR
spectroscopy. The new species [(η3:η0-Ind(CH2)2NMe2)-
Ni(PMe3)(py)]+, 10, formed in the equilibrium between
py and complex 8; the Keq was determined to be 41 ( 6
M-1, which is larger than the Keq of 9 ( 1 M-1 found
for the reaction of 7 with py. In the cases of complex 9,
only 1 equiv of py was sufficient for completely convert-
ing this complex into [(η3:η0-Ind(CH2)2NMe2) Ni(PCy3)-
(Py)]+ (11). On the basis of these observations, the Ni-N
binding strength in the cationic compounds follows the
order 7 (PPh3) > 8 (PMe3) > 9 (PCy3). Therefore, both
the greater steric bulk of PCy3 and the greater basicity
of PMe3 increase the lability of the Ni-N bond. As will
be discussed in the next section, the relative catalytic
activities of these complexes are largely reflected in the
observed order of Ni-N bond lability.
F igu r e 6. Correlation between electrodensity on the Ni
and ligand donation.
complexes, 3 (37.0) > 9 (24.9); for PPh3 complexes, 4
(46.9) > 5 (38.8) > 1 (30.8) > 7 (29.1).
The above findings may be explained in terms of the
hard-soft acid-base theory (or the E-C model), as
proposed by Bergman and co-workers15 for the analo-
gous Cp* complexes: when bonded to a soft ligand such
as an alkyl group, the Ni center becomes softer and
binds more effectively with other soft ligands such as
PR3 and Ind; conversely, when bonded to a hard ligand
such as Cl in the neutral compounds or an amine in
the cationic species, the Ni center binds less effectively
with soft ligands, resulting in weaker Ni-P and Ni-
Ind interactions. Whereas this explanation applies fairly
well to all of the complexes studied here, alternative
explanations involving NifP π-back-bonding and ClfNi
π-donation account for the observed structural data in
case of the PPh3 and the chloro complexes only.16
Liga n d Exch a n ge Rea ction s. The hemilabile che-
lation of the tethered amine in the cationic complexes
[(η3:η1-Ind∧NR2)Ni(PR3)]+ is particularly important in
view of using these compounds as precatalysts. In this
sense, the ease with which the N-Ni binding can be
displaced by incoming substrates should have a direct
bearing on the catalytic activities of these complexes.
In our previous studies,7b,d we have measured the degree
of Ni-N lability in some of these compounds using
ligand exchange reactions. These studies showed that
a number of strong ligands (e.g., pyridine, dppe, etc.)
can displace the chelating amine moiety; this displace-
(15) Holland, P. L.; Smith, M. E.; Andersen, R. A.; Bergmann, R.
G. J . Am. Chem. Soc. 1997, 119, 12815.
(16) One of the reviewers of our manuscript has suggested an
alternative explanation for the structural observations, as follows.
First, it is suggested that larger slip-fold distortions (i.e., larger ∆M-C
values) of the Ind ligand should not be taken as an indication of weaker
Ni-Ind bonds, because average Ni-C (Ind) bonds are shorter (and
hence the Ni-Ind interaction is stronger) in the Ni-Cl derivatives (e.g.,
2) vs the Ni-Me derivatives (e.g., 6). We disagree on the last point:
the average Ni-C distance is smaller in 2 (2.064 vs 2.090 Å) only if
we take into account the allyl portion of the Ind ligand; if, however,
distances over all five Ind carbons are considered, the Ni-Me deriva-
tive 6 has shorter Ni-C bond lengths (2.170 vs 2.195 Å in 2). In other
words, the Ni-Ind interaction appears to be stronger in the system
possessing the better donor (Me vs Cl). This reviewer also suggests
that shorter Ni-P distances (and, by extension, stronger Ni-P bonds)
be attributed to more extensive NifP π-back-bonding present in the
more electron-rich complexes such as the Ni-Me derivative 6. Although
we do not rule out this possibility, it seems to us that PMe3 is not
well-known for significant π-back-bonding interactions with metals.
Furthermore, it is not clear why this type of interaction would not take
place in the Ni-Cl derivative wherein the Ind ligand could serve as a
good source of electron density.
Ca ta lytic P olym er iza tion Rea ction s. Previous
studies7b had indicated that complex 7 is unreactive
toward styrene at room temperature, but heating to 80
°C initiated the polymerization reaction; this require-
ment for heating was attributed to the difficulty of
displacing the chelating NMe2 moiety by styrene. Since
8 and 9 feature more labile Ni-N bonds, we set out to
determine if the polymerization of styrene by these
precatalysts would take place more readily. Indeed,
complexes 8 and 9 do polymerize styrene at room
temperature, but high levels of catalytic activity and
large molecular weights are obtained only at higher
temperatures (Table 4, runs 3-6). The level of catalytic
activity in these reactions seems to be affected mainly
by steric effects, the PMe3 derivative 8 giving the

Aminoalkyl-Substituted Indenylnickel(II) Complexes
Organometallics, Vol. 22, No. 23, 2003 4765
highest activity, while the PCy3 derivative 9 and its
PPh3 analogue 7 displayed comparable activities. On the
other hand, the Mw of the polymer appears to correlate
with electronic effects: the precatalyst 8, featuring the
best donor phosphine ligand (and the compound with
the most negative Ered value), gives polymers with the
longest chains, followed by 9 and 7. That longer polymer
chain length and/or higher catalytic activity arise from
systems having the better donating phosphines PCy3 or
PMe3 suggests that the phosphine ligand remains
coordinated to the Ni center during the polymerization
reaction.
We have also examined the reactivities of complexes
7, 8, and 9 with phenylacetylene for the following
reason. Previous studies10,17 have shown that combining
the precursors (1-Me-indenyl)Ni(PR3)X (R ) Ph or Cy,
X ) Cl, Me, CCPh, thienyl) with methylaluminoxane
(MAO) forms catalytically active species that polymerize
phenylacetylene to cis,transoidal-poly(phenylacetylene)
(PPA); this material is of interest for applications
requiring nonlinear optical and magnetic susceptibili-
ties, photoconductivity, and gas permeability.18 Al-
though the interaction of MAO with these precursors
gives rise to cationic species, the role of the latter in
the polymerization reaction has not been established
with certainty. In an earlier attempt to address this
issue, we studied the reaction of phenylacetylene with
(1-i-Pr-indenyl)(PPh3)Ni(OTf) (OTf ) OSO2CF3), a com-
pound that can generate the unsaturated cations by the
in situ displacement of the triflate anion.19 This reaction
gave poor yields of a polymeric material possessing
lower solubility and diminished Mw compared to the
samples obtained from the MAO-cocatalyzed reactions;
these results implied that the cationic species present
in the latter reactions are not be responsible for the
formation of PPA. Access to the cationic complexes 7,
8, and 9 provided a good opportunity to further probe
this question, as described below.
As shown in Table 4, the cationic complexes do not
promote the polymerization of phenylacetylene (run 7).
Given the earlier discussed lability of the Ni-N bond
in complexes 8 and 9, it is reasonable to presume that
excess phenylacetylene should compete effectively with
the chelating NMe2 moiety for coordination to the Ni
center; therefore, what hinders the polymerization in
the absence of MAO is not the monomer’s access to the
Ni center. A more likely reason is that unlike the
polymerization of styrene, which can proceed by an
electrophilic pathway, the polymerization of pheny-
lacetylene proceeds by an insertion mechanism requir-
ing a Ni-R moiety for initiation; thus, the cationic Ni
center must be converted into a neutral derivative
possessing a moiety that can facilitate the initiation
step. To test whether MAO could activate the cationic
complexes, we carried out the polymerization of pheny-
lacetylene in the presence of MAO (ca. 1:10 ratio of Ni:
Al); these reactions gave PPA having properties similar
to those of the samples obtained from previous MAO-
cocatalyzed reactions (Table 4, runs 8-10). The Mw of
the polymers obtained from these reactions are compa-
rable to the values obtained with previously studied
systems not bearing tethered amine moieties on the Ind
ligand, but the activities of the present systems are
inferior.
In an attempt to gain insight on how MAO activates
the cationic complexes, we monitored the NMR spectra
of a mixture consisting of 7 (ca. 0.015 M in CD2Cl2) and
10 equiv each of MAO and phenylacetylene. The 31P-
{1H} NMR spectrum showed the conversion of 7 to the
analogous Ni-Me and Ni-CCPh derivatives, as evi-
denced by the singlet resonances at 45.5 and 39.9 ppm,
respectively; evidently, MAO can bring about the forma-
tion of the requisite Ni-R moieties. We were intrigued
to find, however, that polymerization reactions using
combinations of MAO and independently prepared (as
opposed to in situ formed) samples of the Ni-Me and
Ni-CCPh derivatives 4 and 5, respectively, yielded PPA
samples with much lower Mw values (e.g., see run 11
for the reaction of 5 + MAO). Since the combination of
MAO and the analogous neutral Ni-R derivatives (R
) Me or CCPh) bearing nonfunctionalized Ind ligands
polymerize phenylacetylene,10,18 it is clear that the
amine moiety has a detrimental effect on these reac-
tions. We conclude, therefore, that the cationic species
cannot polymerize phenylacetylene in the absence of
MAO, but the precise role of the latter in these reactions
remains unknown.
Olefin Hyd r osilyla tion Rea ction s. The complexes
Ind(PR3)Ni(X), either used directly (X ) alkyl) or
generated in situ (X ) H or positive charge), are known
to catalyze the dehydrogenative oligomerization of Ph-
SiH311,20 and the hydrosilylation of olefins and ketones.21
We have studied the influence of the tethered amine
moiety on the course of these reactions, beginning with
the reactivity of complexes 4-6 in the oligomerization
of PhSiH3. Thus, addition of neat PhSiH3 (200 equiv)
to solid samples of 4-6 caused an immediate evolution
of gas (presumably H2); 1H NMR spectra of aliquots
taken at various intervals showed the gradual conver-
sion of the monomer to cyclic and linear (PhSiH)n.22 The
viscous oil produced after 3 days was analyzed by NMR
and GPC and found to consist primarily of cyclic and
linear oligomers in ca. 3:1 ratio for 5 and 1:3 ratio for 4
and 6. Qualitatively, the oligomers obtained from reac-
tions promoted by complexes 4-6 and their nonfunc-
tionalized counterparts are quite similar,11 but the
present precatalysts give higher conversions (monomer
conversion was >95% in all cases).
Next, we examined the effectiveness of complexes 7-9
in the hydrosilylation of styrene and 1-hexene; most of
the experiments were carried out at room temperature
and on a NMR scale in CD2Cl2 using 100 equiv each of
olefin and silane (Table 5). Analysis of the reaction
(17) Wang, R.; Groux, L. F.; Zargarian, D. J . Organomet. Chem,
2002, 660, 98.
(18) (a) Chien, J . C. W. Polyacetylene Chemistry, Physics and
Material Science; Academic Press: New York, 1984. (b) Bredas, J . L.;
Street, G. B. Acc. Chem. Res. 1985, 18, 309. (c) Masuda, T.; Higash-
imura, T. Adv. Polym. Sci. 1986, 81, 121. (d) Yoshimura, T.; Masuda,
T.; Higashimura, T.; Ishihara, T. J . Polym. Sci., Polym. Chem. Ed.
1986, 24, 3569.
(20) Fontaine, F.-G.; Kadkhodazadeh, T.; Zargarian, D. J . Chem.
Soc., Chem. Commun. 1998, 1253-1254.
(21) (a) Fontaine, F.-G. Ph.D. Thesis, Universite´ de Montre´al, 2002.
(b) Fontaine, F.-G.; Nguyen, R.-V.; Zargarian, D. Can. J . Chem. 2003,
in press.
(19) Wang, R.; Groux, L. F.; Zargarian, D. Organometallics 2002,
21, 5531.
(22) For details on the analysis of these materials by NMR see ref
11.

4766 Organometallics, Vol. 22, No. 23, 2003
Groux and Zargarian
Ta ble 5. Hyd r osilyla tion of Olefin sa
this case, three other signals emerged over a few hours
(43.9, -2.8, and -8.1 ppm) in addition to the one at 32.7
ppm that had been observed in the previous experiment.
Addition of styrene to this sample caused the immediate
disappearance of the signals at 43.9, -2.8, and -8.1
ppm and gave rise to the same signals as in the previous
experiment (32.7, -5.8, and -8.3 ppm) along with the
cat. olefin (equiv) silane (equiv) product (conversion, %)
1
2
3
4
5
7
8
9
8
8
styrene (100)
styrene (100)
styrene (100)
PhSiH3 (100) Ph(PhSiH2)CHCH3 (70)
PhSiH3 (100) Ph(PhSiH2)CHCH3 (100)
PhSiH3 (100) Ph(PhSiH2)CHCH3 (45)
styrene (1000) PhSiH3 (1000) Ph(PhSiH2)CHCH3 (93)
styrene (100)
Ph2SiH2 (100) Ph(Ph2SiH)CHCH3 (70)
PhCH2CH2SiHPh2 (10)
1
signal for 8. The H NMR spectrum showed the forma-
6
8
1-hexene (100) PhSiH3 (100) 1-(PhSiH2)hexane (20)
tion of Ph(PhSiH2)CHCH3 as soon as the styrene was
added.
products23 obtained from the hydrosilylation of styrene
with PhSiH3 confirmed the exclusive formation of Ph-
(PhSiH2)CHCH3 with a conversion of 45% (9), 70% (7),
and 100% (8) (Table 5, runs 1-3); no evidence was found
for the formation of poly(styrene) or (PhSiH)n. The
reaction of complex 8 was repeated on a larger scale and
found to give a very good catalytic turnover number (run
4). Replacing PhSiH3 by Ph2SiH2 gave the R-addition
isomer Ph(Ph2SiH)CHCH3 as the major product (70%)
plus a small quantity (<10%) of the â-addition isomer
Ph(CH2)2SiHPh2 (run 5). Finally, reaction of 1-hexene
with PhSiH3 in the presence of 7 led to the formation
of 1-(PhSiH2)hexane with only 20% conversion (run 6).
The above results demonstrate that the cationic
complexes 7-9 can act as single-component precatalysts
for the hydrosilylation reaction. It is noteworthy that
the presence of hydrosilanes in the reaction mixture
inhibits the polymerization of styrene, while the pres-
ence of styrene inhibits or minimizes the dehydrogena-
tive oligomerization of PhSiH3. That the most active
precatalyst is the one bearing the most strongly donat-
ing phosphine ligand (PMe3) suggests that the hydrosi-
lylation reaction does not involve phosphine dissocia-
tion; by inference, we believe that the catalysis involves
the dissociation of the chelating amine moiety. Although
the mechanistic details of this reaction are not known
with certainty, a number of observations from previous
studies21 have pointed to the following sequence of
steps: (a) the transfer of a hydride from the hydrosilane
to the cationic Ni center forms a Ni-H intermediate;
(b) insertion of the olefin gives a Ni-alkyl species; (c) a
concerted, σ-bond metathesis reaction between the
hydrosilane and the alkyl intermediate releases the
hydrosilylation product and regenerates the Ni-H
species. Precedent for the last step has been observed
in our studies on the oligomerization of silanes.11 In an
attempt to extract mechanistic clues for the first two
steps, we monitored near-stoichiometric reactions by
NMR, with the following results.
The above results confirm that the first step in the
hydrosilylation reaction takes place between the precur-
sor and PhSiH3 and generates at least four P-containing
intermediate species. We believe that one of these
intermediates is a Ni-H species formed via the transfer
of H- from PhSiH3;24 unfortunately, however, we were
unable to detect the Ni-H signal in the 1H NMR
spectrum. This might be due to the weak intensity of
this signal as a result of coupling to the P nucleus. To
gain some indirect support for the conversion of complex
8 to a Ni-H species, we repeated the hydrosilylation
reactions in CDCl3; given the propensity of many metal
hydrides to react with chloroform, we reasoned that the
course of the hydrosilylation reaction should be affected
in this medium. Indeed, these reaction mixtures turned
dark blue immediately and the 31P{1H} NMR spectrum
revealed two new signals at 40.7 and 38.9 ppm, but no
trace of the hydrosilylation product was detected in the
1H NMR spectrum. This observation is consistent with
the postulated formation of Ni-H, but it does not
validate it.
The postualted Ni-H intermediate would presumably
promote the formation of Si-Si bonds in the absence of
styrene, but insertion of styrene should form a Ni-alkyl
species and drive the reaction toward hydrosilylation.
This Ni-alkyl intermediate should exhibit a 31P{1H}
NMR signal close to the corresponding signal for the
Ni-Me complex 6 (-4.0 ppm); thus, of the 31P{1H} NMR
signals detected at the end of the hydrosilylation reac-
tions, those appearing at ca. -6 and -8 ppm might be
due to the Ni-alkyl species.
Con clu sion
The results of the present study on the chemistry of
the complexes [(η3:η1-IndCH2CH2NMe2)Ni(PR3)]+ dem-
onstrate the influence of phosphine ligands on the
hemilabile nature of the Ni-N bond and the reactivities
of these complexes. Although the lability of the NfNi
binding, as measured by how easily the chelating amine
moiety is displaced by pyridine, increases in the order
PPh3 < PMe3 < PCy3, the enhanced lability is not
translated into superior catalytic activities in all cases.
For instance, we found that the significant steric bulk
of the PCy3 ligand attenuates the impact of the hemi-
labile Ni-N moiety; as a result, the more labile Ni-N
bond in the PCy3 derivative does not confer a significant
reactivity advantage to this complex relative to the PPh3
derivative. On the other hand, the increased lability of
the Ni-N bond is reflected in the superior catalytic
activities of the complex bearing the strongly donating
and relatively nonbulky phosphine PMe3. The results
No reaction took place between complex 8 (ca. 0.02
M in CD2Cl2) and styrene (5 equiv) over 2.5 h at room
temperature, but addition of PhSiH3 (5 equiv) to this
sample resulted in the partial conversion of 8 to a new
species displaying a 31P{1H} NMR signal at 32.7 ppm.
Over time, two additional 31P{1H} NMR signals ap-
peared (-5.8 and -8.3 ppm), but the starting material
(-19.6 ppm) was still the major P-containing species
1
even after 24 h. The H NMR spectrum contained the
characteristic signals of the reaction product Ph(PhSiH2)-
CHCH3, but the putative Ni-H species was not de-
tected. In a similar experiment, PhSiH3 was added to 8
first and the sample was studied by 31P{1H} NMR; in
(23) Analytical data for the hydrosilylation products discussed
matched those given in: (a) Fu, P.-F.; Brard, L.; Li, F. C.; Marks, T. J .
J . Am. Chem. Soc. 1995, 117, 7157. (b) Rubin, M.; Schwier, T.;
Gevorgyan, V. J . Org. Chem. 2002, 67, 1936.
(24) The fate of the resulting silylium cation is not known, but it
might be stabilized by forming an adduct with the NMe2 moiety.

Aminoalkyl-Substituted Indenylnickel(II) Complexes
Organometallics, Vol. 22, No. 23, 2003 4767
and 115.7 (C4/C7), 100.4 (C2), 91.1 (C1), 75.3 (C3), 59.1
(CH2N), 45.7 (NCH3), 24.0 (Ind-CH2), -18.3 (d, J C-P ) 24.8
Hz, Ni-Me). Anal. Calcd for C32H34NPNi: C, 73.59; H, 6.56;
N, 2.28. Found: C, 73.62; H, 6.83; N, 2.67.
of the present study nicely complement those of our
previous studies on the importance of the amine moiety
for catalytic activities; the lessons learned from these
studies should lead to the development of highly active
precatalysts bearing hemilabile amine moieties.
2
Rea ction of 4 w ith HBF 4. HBF4‚OEt2 (7 µL, 0.057 mmol)
was added to a solution of 4 (27.8 mg, 0.053 mmol) in CDCl3
(ca. 0.8 mL). The solution was then transferred to an NMR
tube and shaken to ensure complete mixing, and the NMR
Exp er im en ta l Section
1
spectra were recorded 10 min later. The 31P{1H} and H NMR
Gen er a l Com m en ts. All manipulations were performed
under an inert atmosphere of N2 using standard Schlenk
techniques and a drybox. Dry, oxygen-free solvents were
employed throughout. Preparation of (η3:η0-Ind(CH2)2NMe2)-
Ni(PPh3)Cl (1),7a [(η3-η1-Ind(CH2)2NMe2)Ni(PPh3)][BPh4] (7),7a
and (PMe3)2NiCl28b have been reported previously. LiCCPh has
been prepared by deprotonation of HCCPh with BuLi in
hexanes. All other reagents used in the experiments were
obtained from commercial sources and used as received. The
elemental analyses were performed by the Laboratoire d’Analyse
EÄ le´mentaire (Universite´ de Montre´al). The spectrometers used
for recording the NMR spectra are as follows: Bruker AMXR400
(1H (400 MHz), 13C{1H} (100.56 MHz), and 31P{1H} (161.92
MHz)) and Bruker AV300 (1H (300 MHz) and 31P{1H} (121.49
MHz)).
(η3:η0-In d (CH2)2NMe2)Ni(P Me3)Cl (2). A THF solution of
Ind(CH2)NMe2 (700 mg, 3.93 mmol) and BuLi (1.6 mL of a 2.5
M solution in hexane) was stirred for 3 h and then transferred
(dropwise over 2 h) to a stirred solution of Ni(PMe3)2Cl2 (1.1
g, 3.93 mmol) in 30 mL of THF at 50 °C. The resulting solution
is evaporated and extracted with 90 mL of hot Et2O. The Et2O
solution is reduced to 40 mL and cooled (-20 °C). Dark red
powder (365 mg, 26% yield) precipitated as pure 2. Recrys-
tallization from hot hexane/Et2O gave crystals suitable for
X-ray diffraction analysis. 31P{1H} NMR (C6D6): -11.01 ppm.
spectra showed characteristic signals of 7. Addition of more
HBF4‚OEt2 (14 µL, 0.114 mmol) provoked the decomposition
of the product.
Rea ction of 4 w ith HCl. HCl (17 µL of a 2 M solution in
Et2O, 0.8 equiv) was added to a solution of 4 (22.7 mg, 0.043
mmol) in CDCl3 (ca. 0.8 mL). The solution was then transferred
to an NMR tube and shaken to ensure complete mixing, and
the NMR spectra were recorded 10 min later. The 31P{1H} and
1H NMR spectra showed ca. 70% conversion of 4 into 1.
(η3:η0-In d (CH2)2NMe2)Ni(P P h 3)CCP h (5). A solution of
LiCCPh (67 mg, 0.62 mmol in 15 mL of benzene) was added
dropwise to a solution of 1 (270 mg, 0.50 mmol) in benzene
(15 mL) and stirred for 2 h. The mixture was then concentrated
to ca. 3 mL, hexanes (ca. 25 mL) were added, and the mixture
was cooled to -20 °C to give a dark red solid. Repeated
recrystallization gave pure product (152 mg, 50% yield). 31P-
{1H} NMR (CDCl3): 38.8 ppm. 1H NMR (C6D6): 7.7-7.0 (PPh3,
CCPh, H6 and H7), 6.82 (H5), 6.45 (H2), 6.14 (H4), 3.97 (H3),
3.13 and 2.91 (CH2N), 2.91 (IndCH2), 2.25 (NCH3). 13C{1H}
2
NMR (CDCl3): 134.1 (d, J P-C ) 10.4 Hz, o-C of PPh3), 133.2
1
(d, J P-C ) 45.8 Hz, i-C of PPh3), 132.2, 131.1 (CCPh, o-C),
130.1 (p-C of PPh3), 128.8, 128.5, 128.1 (d, 3J P-C ) 9.7 Hz, m-C
of PPh3), 127.4 (CCPh, m-C), 125.0 and 124.5 (C5/C6/CCPh,
p-C), 123.0, 121.3, 117.9, and 116.7 (C4/C7), 101.8 (C2), 100.1,
99.0, 74.6 (C3), 58.4 (CH2N), 45.7 (NCH3), 25.6 (Ind-CH2). IR
(KBr, cm-1): 3050, 2924, 2099 (CC), 1591, 1477, 1431, 1383,
1095, 814, 745, 692, 530. Anal. Calcd for C39H36NPNi‚H2O: C,
74.78; H, 6.12; N, 2.24. Found: C, 74.66; H, 6.23; N, 2.25.
(η3:η0-In d (CH2)2NMe2)Ni(P Me3)Me (6). MeLi (0.7 mL of
a 2 M solution in Et2O, 1.32 mmol) was added slowly to a
solution of 2 (314 mg, 0.88 mmol) in Et2O (40 mL) and stirred
for 45 min. Desoxygenated water (0.7 mL) was then added,
and the mixture was stirred for 10 min, filtered, dried (MgSO4),
and evaporated to give crude 6 as a sticky solid. Compound 6
is quite unstable in solution, decomposing to form a black
insoluble powder; this prevented us from obtaining analytically
pure samples. However, a small batch of crystals suitable for
X-ray analysis was obtained after multiple recrystallizations
from hexane solutions. 31P{1H} NMR (C6D6): -3.99 ppm. 1H
NMR (C6D6): 7.23 (d, 3J H-H ) 7.8 Hz, H7), 7.07 (m, H5 or H6),
7.01 (m, H5 or H6), 6.99 (H4), 6.24 (H2), 4.47 (H3), 2.75 (m,
1H NMR (C6D6): 7.11 (d, 3J H-H ) 7.7 Hz, H7), 6.99 (t, 3J H-H
)
3
7.3 Hz, H5 or H6), 6.91 (t, J H-H ) 7.3 Hz, H5 or H6), 6.59 (d,
3J H-H ) 7.3 Hz, H4), 6.57 (H2), 3.66 (H3), 2.80 and 2.65 (m,
3
IndCH2), 2.38 and 2.29 (CH2N), 2.17 (NCH3), 0.74 (d, J H-P
)
9.4 Hz, PCH3). 13C{1H} NMR (C6D6): 130.0 (C7A), 126.5 (C3A),
125.8 (C4), 125.6 (C5), 118.4 (C6) 116.1 (C7), 104.4 (C1), 102.8
(C2), 59.8 (C3), 57.3 (CH2N), 45.6 (NCH3), 24.7 (Ind-CH2), 14.7
(d, 2J C-P ) 29.1 Hz, PCH3). Anal. Calcd for C16H25NPNiCl: C,
53.91; H, 7.07; N, 3.93. Found: C, 53.73; H, 7.24; N, 3.77.
(η3:η0-In d (CH2)2NMe2)Ni(P Cy3)Cl (3). Complex 1 (500 mg,
0.92 mmol) and PCy3 (390 mg, 1.38 mmol) were mixed together
in 80 mL of Et2O and stirred for 3 h. The solution was then
concentrated to 40 mL and cooled to -20 °C. After 24 h, a dark
red solid precipitated as pure 3 (480 mg, 93% yield). 31P{1H}
NMR (C6D6): 37.02 ppm. 1H NMR (C6D6): 7.06 (m, H7/H6),
6.90 (m, H5/H6), 6.76 (H2), 4.17 (H3), 2.87 and 2.67 (m,
IndCH2), 2.33 (CH2N), 2.21 (NCH3), 1.95-1.09 (m, PCy3). 13C-
{1H} NMR (C6D6): 130.2 (C7A), 128.9 (C3A), 126.3 and 125.4
(C4/C5), 118.6 (C6/C7), 103.5 (C2), 102.7 (C1), 59.0 (CH2N),
3
IndCH2), 2.61 (CH2N), 2.19 (NCH3), 0.64 (d, J H-P ) 9.0 Hz,
3
PCH3), -0.73 (d, J H-P ) 6.3 Hz, Ni-CH3). 13C{1H} NMR
(C6D6): 126.7 (C7A), 122.0 (C4/C5), 120.9 (C3A), 116.4 and
1
57.4 (C3), 45.7 (NCH3), 35.2 (d, J P-C ) 19.4 Hz, i-C), 30.1 (d,
116.2 (C6/C7), 100.0 (C2), 91.4 (C1), 70.1 (C3), 59.5 (CH2N),
3
2J P-C ) 6.2 Hz, o-C), 27.9 and 27.8 (d, J P-C ) 4.5 Hz, m-C),
2
45.8 (NCH3), 24.9 (Ind-CH2), 16.1 (d, J C-P ) 44.5 Hz, PCH3),
2
26.7 (s, p-C), 24.8 (ind-CH2). Anal. Calcd for C31H49NPNiCl‚
H2O: C, 64.32; H, 8.88; N, 2.42. Found: C, 64.24; H, 9.13; N,
2.07.
-21.7 (d, J C-P ) 25.2 Hz, Ni-Me). Anal. Calcd for C17H28
-
NPNi: C, 60.76; H, 8.40; N, 4.17. Found: C, 55.67; H, 8.44;
N, 3.88.
(η3:η0-In d (CH2)2NMe2)Ni(P P h 3)Me (4). A solution of MeLi
(0.374 mL of a 1.5 M solution in hexane) was added dropwise
to a solution of 1 (203 mg, 0.374 mmol in 60 mL of Et2O) and
stirred for 1 h. The mixture was then filtered and evaporated
to dryness. Recrystallization from hot hexane gave pure
product (80 mg, 41% yield) as dark red crystals suitable for
X-ray diffraction analysis. 31P{1H} NMR (C6D6): 46.9 ppm. 1H
Rea ction of 6 w ith HBF 4. HBF4‚OEt2 (7 µL, 0.057 mmol)
was added to a solution of 6 (27.8 mg, 0.053 mmol) in CDCl3
(ca. 0.8 mL). The solution was then transferred to an NMR
tube and shaken to ensure complete mixing, and the NMR
1
spectra were recorded 10 min later. The 31P{1H} and H NMR
spectra showed characteristic signals of 8. Addition of more
HBF4‚OEt2 (14 µL, 0.114 mmol) provoked the decomposition
of the product.
[(η3:η0-In d (CH2)2NMe2)Ni(P Me3)][BP h 4] (8). A CH2Cl2
mixture of 2 (228 mg, 0.64 mmol) and NaBPh4 (1.095 g, 3.2
mmol) was stirred at room temperature for 4 h and filtered.
The orange filtrate was concentrated to ca. 1 mL, and hexanes
(40 mL) were added to precipitate an orange-red solid, which
3
NMR (C6D6): 7.7-7.0 (PPh3, H5/H6/H7), 6.54 (d, J H-H ) 7.7
Hz, H4), 6.39 (H2), 4.22 (H3), 2.8-2.5 (m, CH2N and IndCH2),
2.19 (NCH3), -0.65 (d, 3J H-P ) 5.6 Hz, Ni-CH3). 13C{1H} NMR
2
(CDCl3): 134.2 (i-C of PPh3), 133.7 (d, J P-C ) 19.0 Hz, o-C of
3
PPh3), 129.7 (p-C of PPh3), 128.5, 127.9 (d, J P-C ) 14.5 Hz,
m-C of PPh3), 122.0 (C5/C6), 119.9 and 119.6 (C3A/C7A), 116.7

4768 Organometallics, Vol. 22, No. 23, 2003
Groux and Zargarian
was filtered, redissolved in ca. 1 mL of CH2Cl2, and precipi-
tated by adding Et2O (40 mL). Filtration and washing with
Et2O gave pure 8 (282 mg, 69% yield). 31P{1H} NMR (CD2-
relative to the PhSiH3 signal at 4.22 ppm (ca. 90% for reactions
of 4 and 5; >95% for the reaction of 6).
Hyd r osilyla tion of Styr en e. P r ep a r a tion of P h CH(Me)-
(SiP h H2). Styrene (140 µL, 1.2 mmol, 100 equiv) and PhSiH3
(150 µL, 1.2 mmol, 100 equiv) were added to a solution of 7
(10.0 mg, 0.0121 mmol) in CD2Cl2 (ca. 0.8 mL). The sample
was left to stand in an ultrasonic bath over 24 h, during which
the original orange color changed to dark red. 31P{1H} NMR
(CD2Cl2): 41.2 (br) and -4.9 (free PPh3) ppm. The new signals
in 1H NMR (CD2Cl2): 4.59 (m, PhSiH2), 2.82 (m, PhCH), 2.19
1
3
Cl2): -20.77 ppm. H NMR (CD2Cl2): 7.42 (d, J H-H ) 7.2 Hz,
H5), 7.34 (o-H, BPh4), 7.24 (m, H6 and H7), 7.04 (m, m-H,
3
BPh4), 6.89 (m, p-H, BPh4), 6.79 (s, H2), 6.78 (d, J H-H ) 8.5
Hz, H4), 4.29 (m, H3), 3.19 and 2.80 (m, IndCH2), 2.32 and
3
2.05 (s, NCH3), 2.19 and 1.94 (m, CH2N), 1.12 (d, J H-P ) 8.8
Hz, PCH3). 13C{1H} NMR (CD2Cl2): 164,0 (4-line multiplet,
J B-C ) 49.2 Hz, i-C, BPh4), 136.0 (m-C, BPh4), 129.2 and 127.5
(C5 and C6), 126.0 and 124.8 (C3A and C7A), 125.7 (o-C, BPh4),
121.8 (p-C, BPh4), 117.9 (C4 and C7), 107.5 C2), 107.1 (d, 2J C-P
3
(d, J H-H ) 7.56 Hz, PhCH(CH3)). Monomer conversion:
(determined by integration of the signals due to PhSiH3 vs
PhCH(Me)PhSiH2) 70%.
3
) 8.3 Hz, C1), 75.2 (d, J C-P ) 4.1 Hz, CH2N), 65.4 (C3), 52.8
2
and 50.3 (NCH3), 24.5 (Ind-CH2), 14.7 (d, J C-P ) 28.4 Hz,
The experiment was repeated in the same manner for the
other precatalysts. For 8: 100% conversion; 31P{1H} NMR (δ,
CD2Cl2) -5.6 (br) and -19.7 (8). For 9: 45% conversion; 31P-
{1H} NMR (δ, CD2Cl2) 58.9 (major) and 30.9 (trace). The
reaction of 8 with 1000 equiv each of styrene and PhSiH3 in
CD2Cl2 led to similar results after 24 h (conversion 93%). No
hydrosilylation was observed in CDCl3.
PCH3). Anal. Calcd for C40H45NPNiB‚H2O: C, 72.98; H, 6.90;
N, 2.13. Found: C, 72.93; H, 7.10; N, 2.14.
[(η3:η0-In d (CH2)2NMe2)Ni(P Cy3)][BP h 4] (9). The mixture
of 3 (400 mg, 0.71 mmol) and NaBPh4 (1.20 g, 3.57 mmol) in
CH2Cl2 was stirred at room temperature for 4 h and then
filtered. The red filtrate was concentrated to 1 mL and Et2O
(40 mL) added to precipitate a red solid, which was filtered
and redissolved in ca. 1 mL of CH2Cl2 and precipitated by
adding Et2O (ca. 40 mL). Filtration and washing with Et2O
gave pure 9 (395 mg, 66% yield). Crystals suitable for X-ray
analysis were obtained from a cold solution of 9 in CH2Cl2/
P r ep a r a tion of P h CH(Me)(SiP h 2H) a n d P h CH2CH2-
(SiP h 2H). Styrene (140 µL, 1.2 mmol, 100 equiv) and Ph2SiH2
(222 µL, 1.2 mmol, 100 equiv) were added to a solution of 8
(7.7 mg, 0.0121 mmol) in CD2Cl2 (ca. 0.75 mL). The sample
was left to stand in an ultrasonic bath over 24 h, during which
the original orange color changed to dark red. 31P{1H} NMR
(δ, CD2Cl2): 28.3, 26.9, -5.5, -7.8, -19.6 (major). The new
signals in 1H NMR (δ, CD2Cl2): 5.27 (t, PhCH2CH2SiHPh2),
5.21 (d, PhCH(Me)(SiHPh2)), 3.14 (m, PhCH(Me)(SiHPh2)),
3.05 (m, PhCH2CH2SiHPh2), 1.82 (m, PhCH2CH2SiHPh2), 1.77
1
Et2O. 31P{1H} NMR (CD2Cl2): 24.88 ppm. H NMR (CD2Cl2):
3
7.39 (d, J H-H ) 7.3 Hz, H5), 7.30 (o-H, BPh4), 7.21-7.10 (m,
H6, H7, and H2), 7.01 (m, m-H, BPh4), 6.87 (m, p-H, BPh4),
3
6.79 (d, J H-H ) 7.5 Hz, H4), 4.41 (m, H3), 3.10 and 2.80 (m,
IndCH2), 2.22 and 2.13 (s, NCH3), 2.1-1.1 (m, CH2N and PCy3).
13C{1H} NMR (CD2Cl2): 164.0 (4-line multiplet, J B-C ) 49.5
Hz, i-C, BPh4), 135.9 (m-C, BPh4), 131.8 and 130.6 (C3A and
C7A), 128.8 and 128.3 (C5 and C6), 125.6 (o-C, BPh4), 121.8
(p-C, BPh4), 119.3 and 118.2 (C4 and C7), 107.8 C2), 106.0 (d,
2J C-P ) 8.3 Hz, C1), 75.5 (s, CH2N), 65.4 (C3), 52.0 (NCH3),
3
(d, J H-H ) 7.56 Hz, PhCH(Me)(SiHPh2)). Total conversion
80%; ratio of PhCH(Me)(SiHPh2)) to PhCH2CH2SiHPh2 was
7:1.
P r ep a r a tion of CH3(CH2)5(SiP h H2). The above protocol
was carried out using 1-hexene (100 µL, 1.2 mmol, 100 equiv),
PhSiH3 (150 µL, 1.2 mmol, 100 equiv), and 8 (7.7 mg, 0.0121
mmol). 31P{1H} NMR (δ, CD2Cl2): 87.3, -4.9. The new signal
in 1H NMR (δ, CD2Cl2): 4.46 (t, SiH). Conversion: 20%.
Cyclic Volta m m etr y. Electrochemical measurements were
performed on an Epsilon electrochemical analyzer using 0.002
M solutions of the Ni(II) complexes in a 0.1 M CH3CN solution
of n-Bu4NPF6. Cyclic voltammograms were obtained in a
standard, one-compartment electrochemical cell using a graph-
ite-disk electrode as working electrode, a platinum wire as the
counter electrode, and an Ag-AgNO3 (0.01 M in CH3CN)
reference electrode. The experiments were performed in the
potential range of -2.8 to 0.8 V (CH3CN) using a scan rate of
100 mV/s. Under these conditions, E1/2 for the Fc+-Fc couple
was 90 mV.25
1
34.9 (d, J P-C ) 18.8 Hz, i-C), 30.4 and 29.8 (s, o-C), 27.6 (m,
m-C), 26.2 (s, p-C), 23.0 (Ind-CH2). Anal. Calcd for C55H69
-
NPNiB‚CH2Cl2: C, 72.36; H, 7.70; N, 1.51. Found: C, 72.05;
H, 7.81; N, 1.55.
P olym er iza tion of Styr en e. 8 (13.5 mg, 0.021 mmol) or 9
(15.5 mg, 0.018 mmol) and styrene (2000 equiv) were stirred
at room temperature (Table 4, runs 3 and 5) or at 80 °C (runs
4 and 6) for 2 days in dichloroethane (6 mL). Removal of the
solvent and unreacted styrene gave a white solid (run 3: 70
mg, 23 turnovers; run 4: 1.39 g, 630 turnovers; run 5: 128
mg, 57 turnovers; run 6: 670 mg, 350 turnovers), which was
1
isolated and analyzed by GPC (THF). H NMR (CDCl3): 7.07
(br), 6.59 (br), 1.87 (br), 1.45 (br). 13C{1H} (CDCl3): 145.4 (ipso-
C), 128.0 (o- and m-C), 125.9 (p-C), 44.1 and 40.8 (alkyl chain).
P olym er iza tion of P h en yla cetylen e. To a solution of the
Ni precatalyst (ca. 0.0364 mmol) in 2 mL of THF was added
phenylactylene (0.40 mL, 3.64 mmol, 100 equiv) and MAO
(0.24 mL of a 10% ww solution in toluene, 10 equiv with
respect to Ni), and the reaction mixture was stirred for 24 h
at room temperature and under nitrogen. (No reaction is
observed in the absence of MAO.) The reaction was quenched
by adding a solution of ethanol/acetic acid; the resulting yellow
precipitate (PPA, 5-10% yield) was filtered, washed with
hexane, dried in vacuo, and analyzed by GPC (THF). 1H NMR
[(η3:η0-In d(CH2)2NMe2)Ni(P Me3)(P y)][BP h 4] (10) in Equ i-
libr iu m w ith 8. Pyridine (0.0371 mmol, 3.0 µL, 1.08 equiv)
was added to a solution of 8 (22.0 mg, 0.0344 mmol) in CD2-
Cl2 (ca. 0.75 mL), allowed to stand for 15 min, and analyzed
1
by H and 31P{1H} NMR spectroscopy. By integrating the 31P-
{1H} NMR signals for 10 (-9.19 ppm) and 8 (-20 ppm) a 1:1
ratio was established. 1H NMR (CD2Cl2): 7.6-6.9 (BPh4, Ind,
and Py), 6.43 (H2), 4.45 (br, H3), 2.55 and 2.28 (IndCH2CH2),
3
2.16 (s, NCH3), 1.05 (d, J H-P ) 8.8 Hz, PCH3).
More pyridine (1.08 equiv, 3.0 µL, 0.0371 mmol) was then
added to the above sample, and the NMR spectra were
recorded 15 min later, showing a 2.3:1 ratio of 10 and 8.
Repeated addition of pyridine allowed the determination of the
equilibrium constant: Keq ) 41 ( 6 M-1. Evaporation of the
solution to dryness gave back the starting complex 8.
[(η3:η0-In d (CH2)2NMe2)Ni(P Cy3)(P y)][BP h 4] (11). Pyri-
dine (0.0309 mmol, 2.5 µL, 1.02 equiv) was added to a solution
of 9 (25.5 mg, 0.0302 mmol) in CD2Cl2 (ca. 0.75 mL), and the
1H and 31P{1H} NMR spectra were recorded 15 min later. The
3
(CDCl3): 6.94 (m, m- and p-H), 6.62 (d, J H-H ) 6.8 Hz, o-H),
5.84 (s, vinylic H).
Deh yd r op olym er iza tion of P h SiH3. Addition of PhSiH3
(2.0 mmol, 0.25 mL) to solid samples of 4, 5, or 6 (0.01 mmol)
led to the evolution of gas (H2), which was most vigorous with
the mixture of 6. Stirring the mixtures for 3 days gave thick
oils consisting of various mixtures of cyclic and linear polysi-
1
lanes, as determined by the H NMR spectra: broad peaks at
5.6-5.1 ppm (cyclic) and 4.8-4.4 ppm (linear). The cyclic to
linear ratio was determined for each case by integration of
these peaks (25:75 for reaction with 4 and 6; 78:22 for reaction
with 5), while the monomer conversion was determined
(25) Pavlishchuk, V. V.; Addison, A. W. Inorg. Chem. Acta 2000,
298, 97.

Aminoalkyl-Substituted Indenylnickel(II) Complexes
Organometallics, Vol. 22, No. 23, 2003 4769
31P{1H} NMR spectrum showed the disappearance of the signal
for 9 (24.88 ppm) and the emergence of a new signal at 33.20
ppm. 1H NMR (CD2Cl2): 8.27, 7.75, 7.65, 7.40, 7.20 (m, Py and
Ind), 7.31 (o-H, BPh4), 7.01 (m, m-H, BPh4), 6.87 (m, p-H,
BPh4), 6.49 (d, H4), 4.68 (m, H3), 2.48 (m, IndCH2), 2.13 (s,
NCH3), 1.8-0.9 (m, CH2N and PCy3). Removing the volatiles
under vacuum and recording the 31P{1H} NMR spectrum of
the redissolved solid showed that 11 remained as the main
product.
Cr ysta l Str u ctu r e Deter m in a tion s. Dark red crystals of
5 were obtained from a cold (-20 °C) THF/hexanes solution.
The crystal data for 5 were collected on a Nonius CAD-4
diffractometer with graphite-monochromated Cu KR radiation
at 223(2) K using the CAD-4 software.26 Refinement of the cell
parameters was done with the CAD-4 software, while the data
reduction used NRC-2 and NRC-2A.27
Dark red crystals of 2, 4, 6, and 9 were obtained from cold
solutions 2 in Et2O/hexanes, 4 or 6 in hexanes, and 9 in CH2-
Cl2/Et2O. The crystal data were collected on a Bruker AXS
SMART 2K diffractometer with graphite-monochromated Cu
KR radiation at 223(2) K using SMART.28 Cell refinement and
data reduction used SAINT.29 All five structures were solved
by direct methods using SHELXS9730 and difmap synthesis
(SHELXL96).31 The refinements were done on F2 by full-matrix
least squares. All non-hydrogen atoms were refined anisotro-
pically, while the hydrogens (isotropic) were constrained to the
parent atom using a riding model. Solving the structure of 5
entailed an interesting problem: the CC triple bond first
obtained was much shorter than expected.32 Careful inspection
of the data revealed that the starting material 1 cocrystallized
with 5 in a 9:91 ratio (also confirmed by NMR spectroscopy).
Taking into account this ratio and solving for a THF molecule
disordered over two positions (occupancy of 0.62 and 0.38)
resulted in a good R factor (ca. 4.3%). The structure of 4 also
contained a disordered solvent molecule of hexane, which was
situated on the inversion point with two orientations (oc-
cupancy of 0.35 and 0.15). The relatively high R factor for the
structure of 6 (ca. 7.5%) is due to poor quality of the crystals
(small, twinned needles). Finally, the crystal structure of 9
presented a disorder on the chelated tether (two positions with
occupancy of 0.72 and 0.28) and also contained a CH2Cl2
molecule disordered over three positions (occupancy of 0.20,
0.38, and 0.42). These disorders were solved and allowed an
R factor of 4.41%. Crystal data and experimental details for
2, 4, 5, 6, and 9 are listed in Table 1, and selected bond
distances and angles are listed in Table 2.
Ack n ow led gm en t. The Natural Sciences and En-
gineering Research Council of Canada, le Fond FCAR
of Quebec, and the University of Montreal are gratefully
acknowledged for financial support. H. Terrien is thanked
for her help with the GPC.
Su p p or tin g In for m a tion Ava ila ble: Complete details on
the X-ray analysis of 2, 4, 5, 6, and 9, including tables of crystal
data, collection and refinement parameters, bond distances
and angles, anisotropic thermal parameters, and hydrogen
atom. This material is available free of charge via the Internet
at http://pubs.acs.org.
(26) CAD-4 Software, Version 5.0; Enraf-Nonius: Delft, The Neth-
erlands, 1989.
(27) Gabe, E. J .; Le Page, Y.; Charlant, J .-P.; Lee, F. L.; White, P.
S. J . Appl. Crystallogr. 1989, 22, 384.
(28) SMART, Release 5.059; Bruker Molecular Analysis Research
Tool, Bruker AXS Inc.: Madison, WI 53719-1173, 1999.
(29) SAINT, Release 6.06; Integration Software for Single Crystal
Data, Bruker AXS Inc.: Madison, WI 53719-1173, 1999.
(30) Sheldrick, G. M. SHELXS, Program for the Solution of Crystal
Structures; University of Goettingen: Germany, 1997.
(31) Sheldrick, G. M. SHELXL, Program for the Refinement of
Crystal Structures; University of Goettingen: Germany, 1996.
OM0341348
(32) Twenty-four structures containing a Ni-CtC group have been
found on the Cambridge Structural Database with
distance of 1.204 Å.
a mean CtC