Tetraphenylborate as a novel bridging ligand in a zwitterionic nickel(I) dimer

Article abstract of DOI:10.1002/anie.200502905

An uncommon interaction characterizes the Ni^I-Ni^I zwitterionic complex presented here. The μ-Ph moiety of the tetra-phenylborate ligand coordinates synfacially to the two Ni centers (see picture) and provides two pairs of electrons through an η²(C2,C3) :η²(C5,C6) interaction that is strengthened as a result of the electrostatic attractions between the anion and the formally cationic Ni ₂ fragment. (Chemical Equation Presented).

Full text of DOI:10.1002/anie.200502905

Angewandte
Chemie
Cationic Ni2 Fragment
DOI: 10.1002/anie.200502905
Tetraphenylborate as a Novel Bridging Ligand in a
Zwitterionic Nickel(i) Dimer**
Yaofeng Chen, Christine Sui-Seng, and Davit Zargarian*
Heterolytic abstraction of anionic ligands such as halides or
alkyl substituents can convert coordinatively saturated or-
ganometallic complexes into highly electrophilic, usually
transient, cationic species that can serve as efficient initiators
in various catalytic reactions. This strategy has been used in
the development of many catalytic processes based on (quasi)
square-planar d8 metal precursors, including the recently
reported systems for olefin polymerization reactions.[1]
In this context, we have shown that abstraction of Clꢀ
from the complexes [(R-ind)Ni(PPh3)Cl] (R-ind = indenyl
and its substituted derivatives)[2] leads to the generation of a
transient species (presumed to be [(R-ind)Ni(PPh3)]+) that
reacts with excess substrate to promote the oligomerization or
polymerization of various olefins,[3] PhC CH,[4] and PhSiH3,[5]
ꢁ
or catalyzes the hydrosilylation of olefins and ketones[6]
[*] Dr. Y. Chen, C. Sui-Seng, Prof. D. Zargarian
DØpartement de Chimie
UniversitØde MontrØal
MontrØal, PQ H3C3J7 (Canada)
Fax: (+1)514-343-2468
E-mail: zargarian.davit@umontreal.ca
[**] The authors gratefully acknowledge NSERC of Canada for financial
support of this study and Profs. F. Schaper and D. Gusev for many
useful discussions.
Supporting information for this article is available on the WWW
under http://www.angewandte.org or from the author.
Angew. Chem. Int. Ed. 2005, 44, 7721 –7725
ꢀ 2005 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
7721

Communications
(Scheme 1). We have also shown that abstraction of Clꢀ in the
repeated recrystallizations allowed us to identify it as 3,
which is a dinuclear zwitterion composed of two (PPh3)NiI
cations bridged by the anions m-[PPh2]ꢀ and syn,m-[(h2:h2-
Ph)BPh3]ꢀ (Figure 1). The main features of this crystal
structure are consistent with the solution spectra: the A2X
signals observed in the 31P NMR spectrum of 3 can be
assigned to a m-PPh2 and two PPh3 moieties,[9] while the
upfield signals in the 1H and 13C NMR spectra can be assigned
to the (h2:h2-Ph)BPh3 moiety.
presence of suitable ligands L gives [(R-ind)Ni(PPh3)L]+ (L =
PR’3; R’’CN, tBuNC).[3a,4c,5d] If, on the other hand, Clꢀ is
Scheme 1. Reaction pathways following abstraction of Clꢀ from the
complexes [(R-ind)Ni(PPh3)Cl].
abstracted in the absence of sufficiently strong coordinating
ligands or substrates, we obtain [(R-ind)Ni(PPh3)2]+ in low
yields.[3a] Varying amounts of the latter species also form
during the above-noted catalytic reactions, thus resulting in
lower catalytic activities.
In the search for a strategy to circumvent this built-in
deactivation pathway, we set out to determine whether
placing sterically bulky substituents on the ind ligands might
prevent the formation of [(R-ind)Ni(PPh3)2]+ during catalysis.
An initial study showed that the catalytic activities of [(R-
ind)Ni(PPh3)Cl] for the hydrosilylation of styrene do increase
with more bulky ind ligands (1-Me-ind < 1-SiMe3-ind < 1,3-
(SiMe3)2-ind).[7] To answer this question more directly, we
have studied the abstraction of Clꢀ from [(1-SiMe3-3-R-
ind)Ni(PPh3)Cl] (R = H, 1; SiMe3, 2) and report our findings
herein.
Figure 1. ORTEP view of 3. Thermal ellipsoids are shown at the 30%
probability level. Selected distances [Š] and angles [8]: Ni1-Ni2
2.4471(11), Ni1-P1 2.1397(16), Ni2-P1 2.1395(13), Ni1-P2 2.2055(15),
Ni2-P3 2.1841(16), Ni1-C1 2.670(4), Ni1-C2 2.120(4), Ni1-C3 2.094(4),
Ni1-C4 2.617(4), Ni2-C1 2.666(5), Ni2-C4 2.522(5), Ni2-C5 2.075(5),
Ni2-C6 2.125(5), C1-C2 1.421(6), C2-C3 1.415(6), C3-C4 1.396(6), C4-
C5 1.392(6), C5-C6 1.415(6), C6-C1 1.414(6); P1-Ni1-P2 116.27(6), P1-
Ni2-P3 110.98(6), P2-Ni1-Ni2 168.82(5), P3-Ni2-Ni1 165.66(5), Ni2-P1-
Ni1 69.76(5), P1-Ni1-Ni2 55.12(4), P1-Ni2-Ni1 55.12(4).
Treatment of 1 with NaBPh4 gave an initial product whose
NMR spectra are consistent with the formation of [(1-SiMe3-
ind)Ni(PPh3)2][BPh4],[8] although the ready decomposition of
this new species prevented its isolation and full character-
ization. It appears, therefore, that the presence of one SiMe3
substituent on ind destabilizes the bis(phosphane) complex
but does not block its formation. In contrast, treatment of 2
with NaBPh4 gave a product that was stable and isolable, but
whose spectral features were not those of a bis(phosphane)
cation. For instance, the 31P{1H} NMR spectrum of this
product displays an A2X system with a doublet at around
d = 22 ppm and a triplet at around d = 126 ppm (JP, P = 35 Hz),
The most notable structural feature of 3 is the novel
bonding mode of the borate moiety: although there are many
precedents for complexes featuring [(h2–6-Ph)nBPh4ꢀn]ꢀ,[10,11]
the formation of a complex in which one of the Ph groups
bridges two metals is without precedent.[12] Interestingly, the
synfacially coordinated m-PhBPh3 moiety interacts with the
two NiL2 moieties through two nonadjacent double bonds
ꢀ
ꢀ
from opposite edges of the m-arene ring, i.e., C2 C3 and C5
C6. We propose that this unusual bonding mode[13] minimizes
the steric interactions between the Ph groups on the PPh3
ligands and the borate moiety in 3.[14]
A more complete picture of the bonding in the Ni2-
while the H and 13C NMR spectra contain many signals in
[m,h2,h2-PhBPh3] moiety emerges from a careful inspection of
1
regions typical of phenyl groups (1H: d = 6.5–7.3 ppm; 13C:
d = 122–138 ppm) in addition to a few peaks farther upfield
(1H: d = 5.5–5.9 ppm; 13C: d = 86–102 ppm). Significantly, the
signals expected for the (SiMe3)2-ind moiety were absent from
the latter spectra. In a search for clues regarding the fate of
this moiety, we analyzed the reaction mixture by GC-MS and
1H NMR spectroscopy, which enabled the detection of 3-
phenyl-1,1-bis(trimethylsilyl)indene.
the individual Ni C distances, which show that the m-Ph ring
straddles the Ni–Ni axis and is somewhat tilted, giving the
ꢀ
ꢀ
ꢀ
ꢀ
following unsymmetrical Ni C interactions: Ni1 C2 > Ni1
C3; Ni2 C6 > Ni2 C5; Ni(1/2) C1 > Ni(1/2) C4. Interest-
ingly, these interactions result in only minor and more or less
uniform lengthening of the C C bond lengths within the m-Ph
ring (1.39–1.42 Š), although the dihedral angle of about 158
between the planes (C1,C2,C3,C4) and (C1,C6,C5,C4) signals
a relatively significant butterfly distortion of the ring.[15,16]
It is instructive to compare the unusual Ni–arene inter-
actions found in 3 to those in complex 4[17], which is a rare
ꢀ
ꢀ
ꢀ
ꢀ
ꢀ
These spectral analyses did not, however, allow an
unambiguous characterization of the main product, but X-
ray diffraction analysis of single crystals obtained from
7722
www.angewandte.org
ꢀ 2005 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Angew. Chem. Int. Ed. 2005, 44, 7721 –7725

Angewandte
Chemie
reasonable, as is the rearrangement of the initially formed
1-phenyl-1,3-bis(trimethysilyl)indene to the observed 3-
phenyl-1,3-bis(trimethysilyl)indene isomer by a silyl migra-
tion (step e).[23] Finally, comproportionation of NiII and Ni0
species to give NiI dimers (step d) has been reported,[24]
although the origin of the putative “Ni0(PPh3)2” fragment is
not known at this point.
Ni2(m-arene) species.[18] In contrast to 3, the Ni2(m-C6H6)
interactions in 4 are antifacial and involve two adjacent
We conclude that the greater steric bulk in [{(SiMe3)n-
ind}Ni(PPh3)Cl] indeed hinders the generation of the cationic
bis(phosphane) side-products that form with less bulky
analogues, and this might contribute to the observed increase
in the catalytic reactivity of these complexes.[7] On the other
hand, the presence of SiMe3 substituents on ind also opens a
new decomposition pathway that, in the case of the 1,3-
(SiMe3)2-ind derivative 2, results in the formation of the NiI
dimer 3. The observation of a [(m,h2,h2-Ph)BPh3]ꢀ ligand in 3
expands the range of coordination modes that should be
taken into account when contemplating the use of BPh4ꢀ and
its various analogues as weakly coordinating counterions.
Preliminary experiments have shown that 3 does not react
with PhSiH3 or styrene, thus implying that this compound is
not involved in the hydrosilylation catalysis promoted by its
precursor. In view of the interesting catalytic reactivities
exhibited by some PdI–PdI dimers,[25] however, future studies
are planned to probe the reactivities of 3 towards other
substrates.
ꢀ
double bonds, i.e., a 1,3-diene moiety. Furthermore, the Ni C
distances are significantly shorter in 4 (ca. 2.00–2.02 vs. 2.10 Š
in 3), which is presumably due to greater backbonding
possible in the Ni0 complex. It should be noted, however, that
the interactions of the Ni centers in 3 with (m-PhBPh3) are
reinforced by the electrostatic attractions within this zwitter-
ionic compound. Accordingly, the arene moiety in 3 cannot be
displaced by aromatic solvents, even when heated for
extended periods.
The structural parameters discussed above are consistent
with a simple bonding picture involving the contribution of
2
2
ꢀ
ꢀ
two pairs of electrons from [(m,h ,h -Ph)BPh3] ; when the Ni
Ni bond is taken into consideration,[19] each Ni center ends up
with a total of 16 valence electrons. This is in agreement with
the conclusions of a theoretical analysis[20] showing that the
analogous fragment [{(PH3)Pd}2(m-Br)]+ has only two low-
lying vacant orbitals of appropriate symmetry to interact with
the two HOMO orbitals of a (m-C6H6) ligand.
The unexpected formation of 3 from 2 raises the question
of how this transformation might take place. Although
definitive conclusions regarding the reaction mechanism
must await further experimental evidence, the sequence of
steps shown in Scheme 2 serves to rationalize the conversion
Experimental Section
A slurry of [{1,3-(SiMe3)2-ind}Ni(PPh3)Cl] (191 mg, 310 mmol) and
Na[BPh4] (113 mg, 330 mmol) was stirred in Et2O (20 mL) at room
temperature for two days. The reaction mixture
was then filtered and the solid residues extracted
with CH2Cl2 (total volume ca. 15 mL). (Analysis
of the filtrate is described below.) Concentration
of the extracts to about 2 mL and layering with
hexane (ca. 5 mL) resulted in the formation of
dark-brown crystals overnight, which were iso-
lated, washed quickly with acetone, and dried
under vacuum to give 3 (30 mg, 25% yield).
1H NMR (400 MHz, CDCl3, 258C): d = 7.33–6.50
(m, 55H; Ph-H), 5.92 (brs, 2H; benzene-H2,6/3,5),
5,79 (brs, 1H; benzene-H4), 5.52 ppm (brs, 2H;
benzene-H2,6/3,5); 13C{1H} NMR (100.56 MHz,
CDCl3, 258C): d = 138.5, 136.1, 135.6, 134.8,
133.3, 132.1, 129.5, 127.1, 125.8, 122.3, 102.6,
97.1, 86.2 ppm; 31P{1H} NMR (161.92 MHz,
CDCl3, 258C): d = 126.5 (t, 2JP, P = 35 Hz),
22.3 ppm (d, 2JP, P = 35 Hz); elemental analysis
calcd (%) for C72H60BP3Ni2: C 75.44, H 5.28;
found: C 74.70, H 5.25.
Crystal data for 3: monoclinic (P21/n); a =
18.6631(6), b = 13.5095(5), c = 25.0944(9) Š; b =
Scheme 2. Suggested pathway for the conversion of 2 into 3.
110.350(2)8; V= 5932.1(4) Š3; Z = 4; 1calcd
=
1.283 gcmꢀ1; Bruker AXS diffractometer; m =
1.869 mmꢀ1, l = 1.54178 Š (CuKa); w scan, qmax
=
60.138; temperature: 223(2) K; h,k,l range: ꢀ20 ꢂ h ꢂ 20; ꢀ14 ꢂ k ꢂ
15; ꢀ28 ꢂ l ꢂ 27; 5072 reflections used (I > 2s(I)); absorption correc-
tion multiscan SADABS; R(F2>(2sF2)), wR(F2): 0.0591, 0.1336;
GOF: 0.929. CCDC-279748 contains the supplementary crystallo-
graphic data for this paper. These data can be obtained free of charge
from the Cambridge Crystallographic Data Centre via www.ccdc.ca-
m.ac.uk/data_request/cif.
ꢀ
of 2 into 3. The proposed addition of a P Ph bond (step b) is
well known;[21] indeed, a particularly relevant example of such
ꢀ
P C bond activation has been observed during the formation
of [Pd2(tBu2PH)2(m-tBu2P)(m-C6H5O)], which is a closely
related, isoelectronic analogue of 3.[22] The reductive elimi-
nation from a NiIV intermediate (step c) is also quite
Angew. Chem. Int. Ed. 2005, 44, 7721 –7725
ꢀ 2005 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.angewandte.org
7723

Communications
Analytical data for the filtrate from the reaction mixture: MS: m/z:
doerfer, Organometallics 1998, 17, 2619; e) F. Torres, E. Sola, M.
Martin, C. Ochs, G. Picazo, J. A. Lopez, F. J. Lahoz, L. A. Oro,
Organometallics 2001, 20, 2716; h2-coordination: f) M. Pasquali,
C. Floriani, A. Gaetani-Manfredotti, Inorg. Chem. 1980, 19,
1191; g) A. D. Horton, J. H. G. Frijns, Angew. Chem. 1991, 103,
1181; Angew. Chem. Int. Ed. Engl. 1991, 30, 1152; h) W. J. Evans,
C. A. Seibel, J. W. Ziller, J. Am. Chem. Soc. 1998, 120, 6745; h1-,
h2-, and h4-coordination: i) G. B. Deacon, C. M. Forsyth, Chem.
Commun. 2002, 2522.
336 [M]+, 321 [MꢀCH3]+, 263 [MꢀSiMe3]+, 248 [MꢀSiMe3ꢀCH3]+,
233 [MꢀSiMe3ꢀ2CH3]+, 218 [MꢀSiMe3ꢀ3CH3]+, 73 [SiMe3]+;
1H NMR (400 MHz, CDCl3, 258C): d = 7.5–6.9 (aromatic protons),
6.74 (s, H2 of the indenyl ring), ꢀ0.03 ppm (s, SiMe3).
Received: August 16, 2005
Published online: November 8, 2005
[12] Multiple searches of the CCSD (up to August 6, 2005) did not
Keywords: bridging ligands · chloride abstraction · nickel ·
P ligands · zwitterions
ꢀ
.
lead to any complex in which one of the Ph rings of BPh4 is
coordinated to two or more metal centers. These searches also
indicated that complex 3 is the first instance of a Ni complex
bearing a coordinated [PhBPh3]ꢀ.
[1] a) S. D. Ittel, L. K. Johnson, M. Brookhart, Chem. Rev. 2000, 100,
1169; b) V. C. Gibson, S. K. Spitzmesser, Chem. Rev. 2003, 103,
283.
[2] D. Zargarian, Coord. Chem. Rev. 2002, 233–234, 157.
[3] a) R. Vollmerhaus, F. B. GariØpy, D. Zargarian, Organometallics
1997, 16, 4762; b) M.-A. Dubois, R. Wang, D. Zargarian, J. Tian,
R. Vollmerhaus, Z. Li, S. Collins, Organometallics 2001, 20, 663;
c) M.-A. Dubois, MSc Dissertation, UniversitØ de MontrØal,
2000.
[4] a) R. Wang, F. B. GariØpy, D. Zargarian, Organometallics 1999,
18, 5548; b) R. Wang, L. F. Groux, D. Zargarian, J. Organomet.
Chem. 2002, 660, 98; c) R. Wang, L. F. Groux, D. Zargarian,
Organometallics 2002, 21, 5531; d) E. Rivera, R. Wang, X. X.
Zhu, D. Zargarian, R. Giasson, J. Mol. Catal. A 2003, 204–205,
325.
[5] a) F.-G. Fontaine, T. Kadkhodazadeh, D. Zargarian, Chem.
Commun. 1998, 1253 – 1254; b) F.-G. Fontaine, D. Zargarian,
Organometallics 2002, 21, 401; c) F.-G. Fontaine, D. Zargarian, J.
Am. Chem. Soc. 2004, 126, 8786; d) F.-G. Fontaine, PhD
Dissertation, UniversitØ de MontrØal, 2002.
[13] The following complexes featuring m(h2:h2)-arene ligands coor-
=
dinated through nonadjacent C C bonds have been reported:
a) [{(tBu3SiO)3Ta}2(m,h2:h2-C6H6)]
(D. R. Neithamer, L.
Pꢁrkꢁnyi, J. F. Mitchell, P. T. Wolczanski, J. Am. Chem. Soc.
1988, 110, 4421); b) {AgN(SO2F)2(C6H6)}n (M. J. Begley, D. B.
Sowerby, R. D. Verma, A. Vig, J. Organomet. Chem. 1994, 481,
243);
c) [{Fe(CO)3}2{m,h2:h2-1,4-(CH2NPh)2-3,6-h1,h1-
{Fe(CO)3}2-C6H2}] (W. Imhof, A. Gꢂbel, J. Organomet. Chem.
2000, 610, 102); d) [Pd2(GaBr4)2{m(h2:h2-arene)}2] (M. Gorlov, A.
Fischer, L. Kloo, J. Organomet. Chem. 2004, 689, 489).
[14] Structural simulations have shown that rotating the borate anion
in such a way as to orient a 1,3-diene moiety of the m-PhBPh3
ligand towards the Ni atoms results in significant steric
interactions between the Ph groups of the PPh3 ligand and the
borate anion. The authors are grateful to Prof. F. Schaper for
carrying out these simulations.
[15] For comparison, values of about 208 and 438 have been reported
for the corresponding fold angles in the complexes [Cp2V2(m-
H)2(m-Ph)] (K. Jonas, V. Wiskamp, Y.-H. Tsay, C. Krꢃger, J. Am.
Chem. Soc. 1983, 105, 5480) and [Cp2Fe2(m-C6R6)] (K. Jonas, G.
Koepe, L. Schieferstein, R. Mynott, C. Krꢃger, Y.-H. Tsay,
Angew. Chem. 1983, 95, 637; Angew. Chem. Int. Ed. Engl. 1983,
22, 620).
[16] For a comprehensive review on the various bonding modes
observed in m-arene complexes, see: H. Wadepohl, Angew.
Chem. 1992, 104, 253; Angew. Chem. Int. Ed. Engl. 1992, 31, 247.
[17] I. Bach, K.-R. Pꢂrschke, R. Goddard, C. Kopiske, C. Krꢃger, A.
Rufinska, K. Seevogel, Organometallics 1996, 15, 4959.
[18] To our knowledge, the only other m-arene complexes of Ni for
which structural data are available are triple-decker complexes
that involve, respectively, h6- and h3:h3-arene ligands: a) J. L.
Priego, L. H. Doerrer, L. H. Rees, M. L. H. Green, Chem.
Commun. 2000, 779; b) J. J. Schneider, D. Spickermann, D.
Blꢄser, R. Boese, P. Rademacher, T. Labahn, J. Magull, C.
Janiak, N. Seidel, K. Jacob, Eur. J. Inorg. Chem. 2001, 1371; for
the structure of a Ni dimer bridged by a novel benzdiyne ligand,
(m-C6H2), see: M. A. Bennett, J. S. Drage, K. D. Griffiths, N. K.
Roberts, G. B. Robertson, W. A. Wickramasinghe, Angew.
Chem. 1988, 100, 1002; Angew. Chem. Int. Ed. Engl. 1988, 27,
941.
[6] F.-G. Fontaine, R.-V. Nguyen, D. Zargarian, Can. J. Chem. 2003,
81, 1299.
[7] Y. Chen, C. Sui-Seng, S. Boucher, D. Zargarian, Organometallics
2005, 24, 149.
[8] For example, the 31P{1H} NMR spectrum shows two very broad
peaks at around d = 30 and 33 ppm (width at half-height
approximately 400 Hz), which are assigned to the two inequiva-
lent PPh3 groups that are in
a dissociation/re-association
equilibrium. The observation in the 1H NMR spectrum of a
significantly downfield-shifted signal for H3 (from d = 3.5 ppm
in 1 to about d = 5 ppm in the product) is also consistent with the
greater ind!Ni donation in such cationic species (ref. [2]). This
spectrum also shows the anticipated signals for H2 and the
aromatic protons at about d = 6.5–7.5 ppm, as well as an upfield
singlet for the Si-CH3 protons at around d = ꢀ0.12 ppm.
[9] For comparison, the 31P NMR signals of the m-PPh2 moieties in
[(CNR)2Ni2(m-PPh2)2(m-Ph2PNHPPh2)2] (E. Simꢀn-Manso, C. P.
Kubiak, Inorg. Chem. Commun. 2003, 6, 1096) and [Fe2(CO)6(m-
PPh2)2] (E. P. Kyba, R. E. Davis, C. N. Clubb, S.-T. Liu, H. O. A.
Palacios, J. S. McKennis, Organometallics 1986, 5, 869) are found
at around d = 190 and 143 ppm, respectively. On the other hand,
the general range for the 31P NMR signal of a PPh3 coordinated
to Ni is d = 25–50 ppm.
ꢀ
[19] The Ni Ni bond length of 2.4471(11) Š in 3 is within the
I
I
ꢀ
expected range of 2.30–2.50 Š found in related Ni Ni dimers:
a) B. L. Barnett, C. Krꢃger, Y. H. Tsay, Chem. Ber. 1977, 110,
3900; b) R. A. Jones, A. L. Stuart, J. L. Atwood, W. E. Hunter,
R. D. Rogers, Organometallics 1982, 1, 1721; c) R. Hanko,
Angew. Chem. 1985, 97, 707; Angew. Chem. Int. Ed. Engl. 1985,
24, 704; d) M. J. Tenorio, M. C. Puerta, P. Valerga, J. Chem. Soc.
Dalton Trans. 1996, 1305; e) M. D. Fryzuk, G. K. B. Clentsmith,
D. B. Leznoff, S. J. Rettig, S. J. Geiib, Inorg. Chim. Acta 1997,
265, 169 – 177; f) C. E. Kriley, C. J. Woolley, M. K. Krepps, E. M.
Popa, P. E. Fanwick, I. P. Rothwell, Inorg. Chim. Acta 2000, 300–
302, 200.
[10] The main coordination modes of BPh4ꢀ and its derivatives have
been reviewed: a) S. H. Strauss, Chem. Rev. 1993, 93, 927; b) M.
Bochmann, Angew. Chem. 1992, 104, 1206; Angew. Chem. Int.
Ed. Engl. 1992, 31, 1181.
[11] For examples of the variouꢀs coordination modes encountered in
complexes bearing BPh4 fragments, see: h6-coordination:
a) D. C. Ware, M. M. Olmstead, R. Wang, T. Henry, Inorg.
Chem. 1996, 35, 2576; b) R. F. Winter, F. M. Hornung, Inorg.
Chem. 1997, 36, 6197; c) M. Manger, J. Wolf, M. Laubender, M.
Teichert, D. Stalke, H. Werner, Chem. Eur. J. 1997, 3, 1442; d) H.
Werner, M. Manger, U. Schmidt, M. Laubender, B. Webern-
[20] L. Zhu, N. M. Kostic, Organometallics 1988, 7, 665.
7724
www.angewandte.org
ꢀ 2005 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Angew. Chem. Int. Ed. 2005, 44, 7721 –7725

Angewandte
Chemie
[21] a) D. R. Fahey, J. E. Mahan, J. Am. Chem. Soc. 1976, 98, 4499;
b) P. E. Garrou, Chem. Rev. 1985, 85, 171; c) R. A. Dubois, P. E.
Garrou, Organometallics 1989, 8, 466; d) H. Omori, H. Suzuki,
Y. Take, Y. Moro-oka, Organometallics 1989, 8, 2270.
[22] M. Sommovigo, M. Pasquali, P. Leoni, D. Braga, P. Sabatino,
Chem. Ber. 1991, 124, 97.
[23] For a discussion of this type of rearrangements in indenes, see:
a) M. Stradiotto, P. Hazendonk, A. D. Bain, M. A. Brook, M. J.
McGlinchey, Organometallics 2000, 19, 590; b) M. Stradiotto,
M. J. McGlinchey, Coord. Chem. Rev. 2001, 219–221, 311.
[24] For an example of a NiI–NiI dimer formed from the reaction of
Ni0 and NiII precursors, see ref. [19c]. For the analogous
formation of PdI–PdI dimers, see: T. Murahashi, H. Kurosawa,
Coord. Chem. Rev. 2002, 231, 207.
[25] a) J. P. Stambouli, R. Kuwano, J. F. Hartwig, Angew. Chem. 2002,
114, 4940; Angew. Chem. Int. Ed. 2002, 41, 4746; b) M. Prashad,
X. Y. Mak, Y. Liu, O. Repic, J. Org. Chem. 2003, 68, 1163; c) U.
Christmann, R. Vilar, A. J. P. White, D. J. Williams, Chem.
Commun. 2004, 1294.
Angew. Chem. Int. Ed. 2005, 44, 7721 –7725
ꢀ 2005 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.angewandte.org
7725