Indenylnickel imidato complexes: Synthesis, characterization, and reactivities

Article abstract of DOI:10.1021/om980653c

The complex (1-Me-indenyl)Ni(PPh₃)(phthalimidate) has been prepared and fully characterized by IR and ¹H, ¹³C{¹H}, and ³¹P{¹H} NMR spectroscopy and single-crystal X-ray diffraction. The ana

Full text of DOI:10.1021/om980653c

30
Organometallics 1999, 18, 30-35
In d en yln ick el Im id a to Com p lexes: Syn th esis,
Ch a r a cter iza tion , a n d Rea ctivities
Isabelle Dubuc, Marc-Andre´ Dubois, Francine Be´langer-Garie´py, and
Davit Zargarian*
De´partement de chimie, Universite´ de Montre´al, Montre´al, Que´bec, Canada H3C 3J 7
Received J uly 30, 1998
The complex (1-Me-indenyl)Ni(PPh3)(phthalimidate) has been prepared and fully char-
1
acterized by IR and H, 13C{1H}, and 31P{1H} NMR spectroscopy and single-crystal X-ray
diffraction. The analogous 4,5-dichlorophthalimidato, maleimidato, and succinimidato
complexes have also been prepared and characterized spectroscopically. These complexes
are thermally stable, in contrast to the analogous Ni-amido species, which could not be
prepared. Reactivity studies have shown that the Ni-imidato bond is fairly unreactive in
insertion and nucleophilic reactions, and VT 1H NMR studies suggest that the rotation around
the Ni-N bond in the maleimidato complex is hindered by ca. 11 kcal/mol. The solid-state
structure of the phthalimidato derivative showed that the P, N, C1, and C3 atoms are
arranged around the Ni atom in a distorted-square-planar coordination environment, while
the C2 atom is within bonding distance above the main square plane. The phthalimidato
ligand is η1(N)-coordinated to Ni; the Ni-N bond is relatively short (1.895(4) Å), and the
orientation of the imidato ligand is such that the planes bearing the atoms P, Ni, N and C9,
N, C16 are rotated by ca. 76° with respect to each other. The hapticity of the indenyl ligand
is characterized by an allyl-ene distortion (η5Tη3) and a partial localization of bonding inside
the indenyl ring (η3Tη1,η2). 1H and 13C{1H} NMR spectra indicate that the nonsymmetrical
coordination of the indenyl ligand, which can be attributed to the different trans influences
of the phthalimidato and PPh3 ligands, is maintained in the solution. The character of the
Ni-imidato bond has been discussed in terms of electrostatic-covalent and π-bonding
interactions.
In tr od u ction
results indicated that the equilibrium stabilities of (Ind)-
Ni(PPh3)(halides) follows the order I > Br > Cl, while
the Ni-F derivative could not be prepared.7 Since the
more efficient π-donors (i.e., F and Cl) appeared to form
the more labile Ni-X bonds, we inferred that these
compounds might involve destabilizing halidefNi π-in-
teractions. To investigate further the presence of π-in-
teractions in the Ni-X bonds of this family of com-
pounds, we set out to prepare the analogous Ni-NR2
derivatives and study the stability/lability of the Ni-N
bond as a function of the N substituents. In general,
the NR2 ligands are more suitable for probing the degree
of XfM π-donation because the presence of substituents
on these ligands serves as a means for both modulating
and measuring the extent of π-interaction (e.g., by
determining the degree of planarity around N and the
ease of rotation around the M-N bond). The present
paper reports the preparation of Ni-imidato derivatives
and discusses the apparent lability of the analogous Ni-
amido complexes.
Recent reports by Caulton and others1 have shown
that the stabilities and reactivities of electronically
unsaturated LnMX complexes (X ) halides, OR, SR,
NR2) may be influenced significantly by XfM π-dona-
tions. For example, the 16-electron complexes Cp*Ru-
(PR3)X (X ) halide, OR′, NHR′) are “operationally
unsaturated” species which seem to derive further
stabilization from XfM π-interactions.2 In the case of
d8 square-planar complexes, however, such π-interac-
tions are considered destabilizing3 and the M-X bond
should be stronger when the XfM π-interactions are
minimized.
We became interested in the role played by XfM
π-interactions during our studies on the structures and
reactivities of the indenylnickel complexes [(1-R′-Ind)-
Ni(PPh3)(X)]n+ (Ind ) indenyl; R′ ) H and Me, X ) Cl
and Me, with n ) 0;4 X ) PPh3, PMe3, and MeCN with
n ) 15), including their catalytic activities in the
dehydropolymerization of PhSiH3.6 Some of our initial
(4) (a) Huber, T. A.; Be´langer-Garie´py, F.; Zargarian, D. Organo-
metallics 1995, 14, 4997. (b) Huber, T. A.; Bayrakdarian, M.; Dion, S.;
Dubuc, I.; Be´langer-Garie´py, F.; Zargarian, D. Organometallics 1997,
16, 5811. (c) Bayrakdarian, M.; Davis, M. J .; Reber, C.; Zargarian, D.
Can. J . Chem. 1996, 74, 2115.
(1) (a) Caulton, K. G. New J . Chem. 1994, 18, 25 and references
therein. (b) Tilset, M.; Hamon, J .-R.; Hamon, R. J . Chem. Soc., Chem.
Commun. 1998, 765.
(2) J ohnson, T. J .; Folting, K.; Streib, W. E.; Martin, J . D.; Huffman,
J . C.; J ackson, S. A.; Eisentein, O.; Caulton, K. G. Inorg. Chem. 1995,
34, 488.
(3) This is because the metal-based orbitals with appropriate
symmetry for interaction with the donor atom lone pair(s) are occupied.
(5) Vollmerhaus, R.; Be´langer-Garie´py, F.; Zargarian, D. Organo-
metallics 1997, 16, 4762.
(6) Fontaine, F.-G.; Kadkhodazadeh, T.; Zargarian, D. J . Chem. Soc.,
Chem. Commun. 1998, 1253.
(7) Dubois, M.-A.; Zargarian, D., unpublished results.
10.1021/om980653c CCC: $18.00 © 1999 American Chemical Society
Publication on Web 12/08/1998

Indenylnickel Imidato Complexes
Exp er im en ta l Section
Organometallics, Vol. 18, No. 1, 1999 31
4
7.2, H5), 5.74 (d, J ) 7.0, H4), 4.22 (br, H3), 1.40 (d, J P-H
)
4.9, Me). 31P{1H} NMR (CDCl3): δ 35.0 (s). 31P{1H} NMR (CD3-
CN): δ 36.8. 13C{1H} NMR (CDCl3): δ 177.8 (s, C9 and C16),
Gen er a l Com m en ts. All manipulations and experiments
were performed under an inert atmosphere of nitrogen using
standard Schlenk techniques and/or in an argon-filled glove-
box. Dry, oxygen-free solvents were employed throughout. The
elemental analyses were performed by Laboratoire d’analyze
e´le´mentaire (Universite´ de Montre´al). Unless otherwise stated,
136.6 and 134.9 (s, C10, C11, C14, and C15), 133.6 (d, 2J P-C
)
11.1 Hz, o-C of PPh3), 130.9 (d, 1J P-C ) 43.7, i-C of PPh3), 129.9
(d, 4J P-C ) 2.1, p-C of PPh3), 128.2 (d, 3J P-C ) 9.7, m-C of PPh3),
126.6 and 126,7 (s, C6 and C7), 121.9 (s, C11 and C14), 118.2
and 118.1 (s, C4 and C5), 104.6 (s, C2), 102.6 (d, 2J P-C ) 11.8,
C1), 63.0 (s, C3), 12.2 (s, C8); the signals for C3a and C7a were
not located.
1
the H (400 MHz), 13C{1H} (100.56 and 75.44 MHz), and 31P-
{1H} (161.92 MHz) NMR spectra were recorded at ambient
temperature; the IR spectra were recorded as KBr pellets. The
Ni-Cl precursors IndNi(PPh3)Cl (1) and (1-Me-Ind)Ni(PPh3)-
Cl (2) were prepared as described elsewhere.4a,b
(1-Me-In d )Ni(P P h 3)(m a leim id a te) (5). The procedure
outlined above for 4 was used to give a first crop of 5 (182 mg,
54%), which was shown to be essentially pure by 1H NMR,
with small amounts of PPh3 and Ph3PdO also present.
Repeated recrystallizations of a small portion gave an analyti-
cally pure sample. IR (KBr, cm-1): 3056 (m), 1717 (sh), 1646
(s), 1598 (s), 1436 (m), 1347 (m), 1182 (m), 1120 (m), 695 (s).
1H NMR (C6D6): δ 7.60, 6.99 (m, aromatic protons of PPh3 and
(1-Me-In d )Ni(P P h 3)(p h t h a lim id a t e) (3). Met h od A. A
mixture of 2 (130 mg, 0.268 mmol), phthalimide (53 mg, 0.363
mmol), and NEt3 (150 µL) in CH2Cl2 (20 mL) was stirred for 2
h at 45 °C. The resultant mixture was evaporated and the
residue stirred in Et2O and deoxygenated H2O; separation of
the organic layer followed by drying over MgSO4 (under N2),
filtration, concentration, and layering with hexanes precipi-
tated the product, which was filtered and dried (40 mg, 25%).
3
3
indenyl), 6.83 (d, J H-H ) 3 Hz, H5 or H6), 6.71 (d, J H-H
)
2.9, H2), 6.00 (s, H10 and H11), 5.68 (d, 3J H-H ) 6.6, H4), 3.79
(s, H3), 1.48 (d, J H-H ) 3.9, CH3). H NMR (CD3CN): δ 7.36
(m, aromatic protons of PPh3 and indenyl), 7.07 (t, J H-H
3
1
3
)
Meth od B. A stirred THF (20 mL) suspension of potassium
phthalimide (433 mg, 2.34 mmol) was added dropwise to the
stirred dark red solution of 2 (269 mg, 0.554 mmol) in THF
(25 mL). The resultant dark red mixture was stirred for 2 h
at 40 °C and filtered, and the filtrate was evaporated to
dryness. The residual solid from the evaporation was stirred
for 30 min in 25 mL of Et2O and the solution filtered; the
filtrate was layered with 50 mL of hexanes and cooled to -10
°C to yield dark red crystals. Concentration of the mother
liquor to half the volume and cooling gave second and third
crops for a total yield of 168 mg (51%). IR (KBr, cm-1): 3060
(w), 1654 (s), 1631 (sh), 1605 (sh), 1475 (m), 1480 (m), 1372
3
3
7.4 Hz, H6), 6.80 (t, J H-H ) 7.4, H5), 6.87 (d, J H-H ) 2.6,
3
H2), 6.96 (d, J H-H ) 7.0, H7), 6.09 (s, H10 and H11), 5.67 (d,
3J H-H ) 7.1, H4), 4.08 (br, H3), 1.39 (d, J P-H ) 5.4, Me). 31P-
4
{1H} NMR (C6D6): δ 35.5 (s). 31P{1H} NMR (CD3CN): δ 36.4
(s). 13C{1H} NMR (CDCl3): δ 183.4 (C9 and C12), 136.8 (C10
2
and C11), 133.8 (d, J P-C ) 11.1 Hz, o-C of PPh3), 131.1 (d,
1J P-C ) 43.7, i-C of PPh3), 129.9 (d, J P-C ) 2, p-C of PPh3),
4
3
128.3 (d, J P-C ) 9.7, m-C of PPh3), 128.1 and 126.3 (C3a and
C7a), 126.5, 126.4, 118.2, 117.9 (H4, H5, H6, H7), 104.4 (s,
2
C2), 102.7 (d, J P-C ) 11.8, C1), 62.5 (s, C3), 12.2 (C8). Anal.
Calcd: C, 70.45; H, 4.81;, N, 2.57. Found: C, 70.01; H, 4.85;,
N, 2.63.
1
(m), 1124 (m), 725 (m). H NMR (C6D6): δ 7.62, 7.42, 6.99 (m,
aromatic protons of PPh3 and indenyl), 6.85 and 6.83 (d, 3J H-H
(1-Me-In d )Ni(P P h 3)(su ccin im id a t e) (6). To the stirred
solution of 2 (350 mg, 0.720 mmol) in Et2O (40 mL) and THF
(5 mL) was added a solution of deoxygenated H2O (20 mL)
containing KOH (77 mg, 1.37 mmol) and succinimide (143 mg,
1,44 mmol), and this mixture was stirred for 2 h at room
temperature. Separation of the organic phase followed by
drying over MgSO4 (under N2), filtration, and evaporation to
dryness gave the solid product, which was crystallized from
hexane/CH2Cl2 to yield a red solid (120 mg, 30% crude). This
was shown to be essentially pure 6 by 1H NMR, but small
amounts of PPh3 and Ph3PdO were also present and could
not be eliminated completely after repeated recrystallizations.
IR (KBr, cm-1): 3047 (br, w), 1637 (s), 1601 (s), 1472 (m), 1437
(s), 1350 (s), 1277 (m), 1232 (s), 1095 (m), 758 (s). 1H NMR
(CDCl3, 223 K): δ 7.46, 7.36 (m, PPh3), 7.11 (t, J ) 7 Hz, H6),
7.06 (b, H7), 6.83 (br, H2), 6.81 (br, H5), 5.63 (br, H4), 4.11 (b,
H3), 1.93 (br, C(O)CH2CH2C(O)), 1.48 (d, J ) 5, Me). 1H NMR
(CD3CN): δ 7.43-7.40 (m, PPh3), 7.05 (t, J ) 7.1 Hz, H6), 6.99
(br, H7), 6.83 (br, H5), 6.79 (br, H5), 5.68 (br, H4), 4.15 (br,
3
3
) 3 Hz, H11-14), 6.76 (d, J H-H ) 2.9, H2), 5.70 (d, J H-H
)
6.6, H4), 3.88 (s, H3), 1.50 (d, 3J H-H ) 3.9, CH3). 1H NMR (CD3-
CN): δ 7.42 and 7.30 (m, aromatic protons of PPh3), 7.20 (b,
H11-14), 7.09 (t, 3J H-H ) 7 Hz, H6), 6.97 (d, 3J H-H ) 6.9, H7),
3
3
6.92 (d, J H-H ) 2.6, H2), 6.83 (t, J H-H ) 7.0, H5), 5.72 (d,
3J H-H ) 6.9, H4), 4.18 (s, H3), 1.41 (d, J P-H ) 4.8, CH3). 31P-
4
{1H} NMR (C6D6): 35.5 (s). 31P{1H} NMR (CD3CN): δ 36.51
(s). 31P{1H} NMR (CDCl3): δ 34.9 (s). 13C{1H} NMR (75.345
MHz, CDCl3): δ 180.0 (s, C9 and C16), 137.4 (s, C10 and C15),
133.6 (d, 2J P-C ) 11.7 Hz, o-C of PPh3), 130.9 (d, 1J P-C ) 11.7,
i-C of PPh3), 130.3 (s, C12 and C13), 129.6 (d, 4J P-C ) 2.3, p-C
of PPh3), 128.0 (d, 3J P-C ) 10.3, m-C of PPh3), 126.4 and 126.3
(s, C4 and C7), 126.2 (s, C3a or C7a), 119.6 (s, C11 and C14),
2
118.2 and 117.8 (s, C5 and C6), 104.6 (s, C2), 102.3 (d, J P-C
) 12.4, C1), 62.6 (s, C3), 12.2 (s, C8); the second signal for
C3a/C7a was not located. Anal. Calcd: C, 72.51; H, 4.73; N,
2.35. Found: C, 72.21; H, 4.80; N, 2.49.
(1-Me-In d )Ni(P P h 3)(4,5-d ich lor op h th a lim id a te) (4). To
the stirred solution of 2 (300 mg, 0.620 mmol) in Et2O (30 mL)
and THF (10 mL) was added a solution of deoxygenated H2O
(10 mL) containing KOH (383 mg, 6.82 mmol) and 4,5-
dichlorophthalimide (720 mg, 3.33 mmol), and this mixture
was stirred for 6 h at room temperature. Separation of the
organic phase followed by drying over MgSO4 (under N2),
filtration, concentration, and layering with hexanes precipi-
tated the product, which was filtered and dried (206 mg, 50%
4
H3), 1.90-1.92 (br, C(O)CH2CH2C(O)), 1.45 (d, J P-H ) 4.9,
Me). 31P{1H} NMR (CDCl3): δ 35.38 (s). 31P{1H} NMR (CD3-
CN): δ 36.74. 13C{1H} NMR (CDCl3): δ 189.14 (s, C9 and C12),
2
1
133.83 (d, J P-C ) 11.79 Hz, o-C of PPh3), 131.4 (d, J P-C
)
)
3
39.5, i-C of PPh3), 129.92 (s, p-C of PPh3), 128.2 (d, J P-C
9.71, m-C of PPh3), 128.36 and 126.70 (C3a and C7a), 126.53
and 126,41 (s, C6 and C7), 118.19 and 117.93 (s, H4 and H5),
104.82 (s, C2), 101.37 (d, 2J P-C ) 11.8, C1), 62.73 (s, C3), 31.54
(s, CH2CH2), 12.11 (s, C8).
1
crude). This was shown to be essentially pure 4 by H NMR,
but small amounts of PPh3 and Ph3PdO were also present and
could not be eliminated completely even after three recrystal-
lizations. IR (KBr, cm-1): 3052 (w), 1729 (m), 1655 (s), 1601
Rea ctivities of th e Im id a to Com p lexes. Sample solu-
tions of the imidato complexes 3-6 were combined with 5-10
equiv of various substrates in NMR tubes (C6D6), and the
reactions were monitored by 31P NMR spectroscopy. No reac-
tion took place with phenylacetylene, dimethyl acetylenedi-
carboxylate, 1-hexyne, styrene, 1-hexene, PhNCO, and NaF.
With LiCl, LiBr, LiI, PhSH, HCl, PhCH2Br, PhBr, and MeI
the starting material was converted to the corresponding Ni-X
derivative, as evident from the disappearance or diminution
1
(m), 1430 (m), 1338 (s), 1161 (m), 1094 (m), 748 (s). H NMR
(CDCl3, 223 K): δ 7.52, 7.26 (m, PPh3), 7.57 (b, H7), 7.17 (t, J
) 8 Hz, H6), 7.14 (s, Ha), 7.01 (d, J ) 8, H5), 6.90 (b, H2), 6.85
(t, J ) 8, H5), 5.40 (d, J ) 8, H4), 4.14 (b, H3), 1.45 (d, J ) 5,
1
Me). H NMR (CD3CN): δ 7.30 (m, PPh3), 7.10 (t, J ) 7.6 Hz,
H6), 6.99 (d, J ) 7.6, H7), 6.90 (d, J ) 2.6, H2), 6.86 (t, J )

32 Organometallics, Vol. 18, No. 1, 1999
Dubuc et al.
with R2NH (PhNH2, pKa ) ca. 30;10 Ph2NH, pKa ) ca.
25;10 phthalimide, pKa ) ca. 8.311). Metathetic reactions
between Ni-halide and -amide salts are also known
to lead to Ni-amide species,12,13 but reduction of the
starting halides to paramagnetic Ni(I) species is a
possible side reaction.14 The compound (Ind)Ni(PPh3)-
Cl (1) reacted with LiNHAr (Ar ) Ph, 1,3,5-Me3C6H2),
LiNR2 (R ) i-Pr, SiMe3, Ph), and sodium imidazole in
THF or Et2O to give deep green-blue solutions whose
1H NMR spectra showed the presence of varying quan-
tities of the corresponding HNR2 species, 1,1′-biindene
(Ind-Ind), OdPPh3, and (Ind)2Ni15 and broad signals
characteristic of the known Ni(I) compound (PPh3)3-
NiCl.16 The 31P{1H} NMR spectra of these solutions
contained only weak signals attributable to free PPh3
and Ph3PdO (Scheme 1).
We had previously observed the formation of the
byproducts Ind-Ind and (PPh3)3NiCl in a side reaction
during the preparation of 1.4a Although the mechanism
of this reaction is not established,17 we have found that
its occurrence can be minimized by using the 1-Me-Ind
ligand. Accordingly, reacting (1-Me-Ind)Ni(PPh3)Cl (2)
with LiNHPh did prevent the formation of 1,1′-biindene
and (PPh3)3NiCl, but instead of the expected amido
species we obtained a compound which was identified,
F igu r e 1. ORTEP plot of complex 3a with atom-number-
ing scheme. Selected bond lengths (Å) and angles (deg):
Ni-P ) 2.181(2), Ni-N ) 1.895(4), Ni-C1 ) 2.118(5), Ni-
C2 ) 2.051(5), Ni-C3 ) 2.025(5), Ni-C3a ) 2.337(5), Ni-
C7a ) 2.341(5), C1-C2 ) 1.388(6), C2-C3 ) 1.396(6), C3-
C3a ) 1.444(6), C3a-C7a ) 1.407(6), C7a-C1 ) 1.459(6),
C1-C8 ) 1.488, N-C9 ) 1.378(5), C9-O1 ) 1.224(5), C9-
C10 ) 1.506(6), C10-C15 ) 1.380(6), C15-C16 ) 1.495-
(6), C16-O2 ) 1.219(5), C16-N ) 1.383(5); N-Ni-P )
96.75(12), N-Ni-C3 ) 162.25(18), N-Ni-C2 ) 128.3(2),
N-Ni-C1 ) 96.96, P-Ni-C1 ) 165.51(15), P-Ni-C2 )
126.96(17), P-Ni-C3 ) 100.58(16), C16-N-C9 ) 108.7-
(4), N-C9-C10 ) 108.5(4), N-C16-C15 ) 109.4(4),
N-C9-O1 ) 125.8(5), N-C16-O2 ) 124.5(5).
1
on the basis of its characteristic H and 31P{1H} NMR
spectra, to be the recently reported5 cationic complex
[(1-Me-Ind)Ni(PPh3)2]+ (Scheme 1).
We inferred from the above results that the Ni-N
bond in the putative amido species is very labile and
set out to determine whether strongly electron with-
drawing N substituents would reduce the lability of the
Ni-N bonds in these compounds. Thus, 2 was reacted
with the alkali-metal salts of -N(COR)R′, -N(COR)2,
and -N(SO2CF3)2 with the following results. Whereas
-NPh(COMe) did not react with 2, the sulfonimidate
anion -N(SO2CF3)2 replaced Cl- to form [(1-Me-Ind)Ni-
(PPh3)2]+[N(SO2CF3)2]- instead of the expected Ni-N
species, presumably because of the highly noncoordi-
nating nature of the N atom in this anion. On the other
hand, we obtained the expected imidato derivatives 3
and 4 with the phthalimidate and 4,5-dichlorophthal-
imidate anions, respectively (Scheme 1). The maleimi-
of the signal corresponding to the imidato complex and the
emergence of the signal for Ni-X (δ, ppm (C6D6): 31.1 for
X ) Cl, 33.2 for X ) Br, 37.4 for X ) I, and 34.1 for X ) SPh).
X-r a y Diffr a ction Stu d ies of 3. Dark red crystals of 3 were
grown from Et2O/hexane at -20 °C. Crystallographic data for
NiPO2NC36H28 (M ) 596.27): monoclinic, P21/c, a ) 11.853(7)
Å, b ) 10.938(7) Å, c ) 22.840(9) Å, R ) 90°, â ) 97.27(4)°,
γ ) 90°, V ) 2937(3) Å3, Fcalcd ) 1.348 g/cm3, Z ) 4, µ(Cu KR)
) 1.726 mm-1, λ(Cu KR) ) 1.540 56 Å, 2θmax ) 140.0°, T )
293(2) K. A total of 20 613 reflections were measured using
the ω/2θ scan mode (5590 independent, 2120 observed); the
structure was solved by direct methods using SHELXS96 and
difference map synthesis using SHELXL96. Refinement on F2
by full-matrix least squares gave R1 ) 0.0508 and wR2 (all
data) ) 0.1119. The ORTEP diagram is shown in Figure 1
along with pertinent bond distances and angles. Complete
crystallographic data are included in the Supporting Infor-
mation.
(10) Bordwell, F. G.; Zhang, X.-M.; Cheng, J .-P. J . Org. Chem. 1993,
58, 6410.
(11) Streitweiser, A., J r.; Heathcock, C. H. Introduction to Organic
Chemistry, 2nd ed.; Macmillan: New York, 1981; p 744.
(12) (a) VanderLende, D. D.; Abboud, K. A.; Boncella, J . M. Inorg.
Chem. 1995, 34, 5319. (b) Penney, J .; VanderLende, D. D.; Boncella,
J . M.; Abboud, K. A. Acta Crystallogr. 1995, C51, 2269.
(13) Amides of Pd and Pt have also been prepared via metathetic
routes: (a) Bryndza, H. E.; Fultz, W. C.; Tam, W. Organometallics
1985, 4, 939. (b) Bryndza, H. E.; Fong, L. K.; Paciello, R. A.; Tam, W.;
Bercaw, J . E. J . Am. Chem. Soc. 1987, 109, 1444. (c) Cowan, R. L.;
Trogler, W. C. J . Am. Chem. Soc. 1989, 111, 4750. (d) Seligson, A. L.;
Cowan, R. L.; Trogler, W. C. Inorg. Chem. 1991, 30, 3371. (e)
Villanueva, L. A.; Abboud, K.; Boncella, J . M. Organometallics 1991,
10, 2969. (f) Boncella, J . M.; Villanueva, L. A. J . Organomet. Chem.
1994, 465, 297. (g) Driver, M. S.; Hartwig, J . F. J . Am. Chem. Soc.
1996, 118, 7217 and references therein.
Resu lts a n d Discu ssion
Our initial attempts at the synthesis of the target
amido complexes (1-R′-Ind)Ni(PPh3)(NR2) involved a
number of different approaches which are known to lead
to amido derivatives in other systems. For instance, Ni-
amido compounds have been prepared by reacting the
corresponding Ni-alkyl derivatives with azides (by
elimination of N2)8 or with amines bearing relatively
acidic hydrogens (by elimination of alkane).9 We found,
however, that the Ni-Me derivatives (1-R′-Ind)Ni-
(PPh3)Me did not react with azides (PhN3, Me3SiN3) or
(14) Bradley, D. C.; Hursthouse, M. B.; Smallwood, R. J .; Welch, A.
J . J . Chem. Soc., Chem. Commun. 1972, 872.
(15) Fritz, H. B.; Ko¨hler, F. H.; Schwarzhans, K. E. J . Organomet.
Chem. 1969, 19, 449.
(16) Mealli, C.; Dapporto, P.; Sriyunyongwat, V.; Albright, T. A. Acta
Crystallogr. 1983, C39, 995.
(17) We have proposed a Ni-promoted oxidative coupling of IndLi,4a
but other mechanisms (e.g., electron transfer) cannot be ruled out. The
formation of a number of analogous Cp-Cp compounds is also known,
including Schrock’s observation of Cp*-Cp* during the synthesis of
Cd-NR2 compounds: Cummins, C. C.; Schrock, R. R., Davis, W. M.
Organometallics 1991, 10, 3781.
(8) Matsunaga, P. T.; Hess, C. R.; Hillhouse, G. L. J . Am. Chem.
Soc. 1994, 116, 3665.
(9) Kohara, T.; Yamamoto, T.; Yamamoto, A. J . Organomet. Chem.
1978, 154, C37.

Indenylnickel Imidato Complexes
Organometallics, Vol. 18, No. 1, 1999 33
Sch em e 1
dato and succinimidato derivatives 5 and 6 can also be
prepared by this approach, but these reactions do not
proceed to completion even in the presence of excess
imidate. Indeed, 5 may be converted to the Ni-Cl
precursor 2 by stirring it with excess LiCl in Et2O (vide
infra). In an effort to drive the equilibrium further
toward the Ni-imidato species, we carried out the
synthesis of 5 and 6 in a biphasic medium (Et2O/H2O
+ KOH) to remove Cl- from the organic layer; this
approach led to a higher conversion of 2, but the
reactions were somewhat sluggish and gave other
byproducts (e.g., Ph3PdO). It seems, therefore, that
the ease of formation of Ni-NR2 derivatives from
2 decreases in the order 4,5-dichlorophthalimidate ≈
similar ν(CO) values for the d8 imidato complexes (PR3)2-
Pd(Ph)(imidato) (phthalimidato, 1640-1620 cm-1; suc-
cinimidato, 1615-1605 cm-1) and trans-(PPh3)2M(CO)-
(imidato) (M ) Rh, Ir; phthalimidato, 1640-1630 cm-1
;
succinimidato, 1620-1610 cm-1).
A single-crystal X-ray diffraction study was carried
out on 3 in order to obtain more information on the
Ni-N interactions and also to determine the influence
of the Ni-N bond on the hapticity of the 1-Me-Ind
ligand. Some of the structural features of 3 (Figure 1)
are very similar to those found in the Ni-Cl analogues
1 and 2.4a,b Thus, the atoms P, N, C1, C2, and C3 are
within expected bonding distance from the nickel center,
while the atoms C3a and C7a are considerably farther.
The geometry around Ni may be described alternatively
as distorted square planar (with C1dC2 occupying a
single coordination site) or a highly distorted square
pyramidal environment. An interesting difference be-
tween 3 and the Ni-Cl analogues 1 and 2 is the
conformation adopted by PPh3: in 1 and 2 one of the
phenyl rings points toward the Ind ligand and two point
in the opposite direction; in 3, on the other hand, two
phenyl rings point toward the Ind ligand while the third
points toward the phthalimidato ligand (see Figure 1).
This conformation presumably allows π-stacking inter-
actions between a PPh3 phenyl ring and the benzo ring
of the phthalimidato group, which are ca. 3.9 Å apart.
The phthalimidato ligand is bonded in an η1(N)
fashion to the Ni center. To our knowledge, this is the
first report of a structurally characterized Ni-imidato
compound, but imidato complexes of other metals have
been shown to adopt η1(N)- and µ,η2(N,O)-bonding
modes.20 The Ni-N bond length of 1.895(4) Å in 3 is
shorter than the Ni-N distances found in trans-(PMe3)2-
Ni(1,3,5-Me3C6H2)(NPh(R)) (Ni-N ) 1.974(5) Å for R
) C(O)CHPh2, 1.978(6) Å for R ) C(O)NHBut, and
1.932(3) Å for R ) H)12 but comparable to that in Cp*Ni-
(PEt3)(NHAr) (Ar ) tolyl; Ni-N ) 1.903(5) Å).21 The
orientation of the phthalimidato ligand is such that the
planes bearing the atoms P, Ni, N and C9, N, C16 are
phthalimidate > maleimidate ≈ succinimidate
.
N(COR)R′.
Ch ar acter ization of (1-Me-In d)Ni(P P h3)(im idato).
The 1H and 31P{1H} NMR spectra of the imidato
derivatives 3-6 are similar to those of the Ni-Cl
precursor 2. For instance, the H2 and H3 resonances
for these complexes appear at ca. 6.7-6.9 and 3.8-4.1
ppm, respectively, as compared to 6.3 and 3.4 ppm in
2; the 31P{1H} NMR spectra of these complexes con-
tained a single resonance at ca. 35-36 ppm as compared
to 31.1 ppm for 2.4
The IR spectra of the imidato complexes showed ν-
(CO) bands at ca. 1654 (3), 1655 (4), 1646 (5), and 1637
(6) cm-1 corresponding to the imido carbonyl groups. It
is noteworthy that these values are intermediate be-
tween the corresponding ν(CO) values in primarily ionic
imide derivatives such as potassium phthalimidate (ca.
1600 cm-1) and the more covalent derivatives such as
phthalimide-CH3 (ca. 1705 cm-1) and phthalimide-Ph
(ca. 1710-1720 cm-1).18 Colquhoun19 has reported
(18) The Sadtler Standard Spectra, Infrared Prism Standard Spec-
tra, Sadtler Research Laboratories Inc., 1976: spectra # 10984 (N-K+),
1705 (N-Me), 658 (N-Ph), 47993P, 3495 (N-(p-NO2-Ph)), 35177 (N-
OC(O)But).
(19) Adams, H.; Bailey, N. A.; Briggs, T. N.; McCleverty, J . A.;
Colquhoun, H. M.; Williams, D. J . J . Chem. Soc., Dalton Trans. 1986,
813.

34 Organometallics, Vol. 18, No. 1, 1999
Dubuc et al.
rotated by 75.94(0.21)° with respect to each other. The
planes formed by the atoms P, Ni, and N, on one hand,
and C1, C2, C3, C3a, and C7a, on the other, are nearly
perpendicular to each other (ca. 88°); this feature is also
present in the structures of 1-Me-CpNi(PPh3)I,22 CpNi-
(PPh3)Ph,23 Cp*Ni(acac),24 and Cp*Ni(PEt3)X (X ) Br,
CH2Ph, Me, OMe, N(H)Ar, SAr).21
formations via insertion and displacement reactions, as
exemplified by recent reports by Boncella (insertion of
isocyanates and alkynes into Ni-N bonds),12a Togni (Ir-
catalyzed hydroamination of olefins),28a and Hartwig13g
and Buchwald28b,c (cross coupling of aryl halides and
amines via Pd-N intermediates). On the other hand,
the reactivities of late-metal imidato complexes are less
well-known. To our knowledge, the only report on Ni-
imidato compounds is Yamamoto’s study of the com-
plexes (bipy)Ni(phthalimidate)(R) (R ) Me, Et), which
are thermally stable to above 200 °C but react with R′Br
(R′ ) PhCH2 and Ph; 80-120 °C) to give N-R′-phthal-
imides in high yields.9 Analogous Pd(II), Rh(I), and Ir-
(I) imidates do not promote this reaction.19
We studied the reactivities of complexes 3-6 in
insertion and nucleophilic addition/exchange reactions
in order to examine the lability of the Ni-N bonds; the
results of these are summarized in Scheme 1. Com-
plexes 3-6 are completely inert toward the insertion of
olefins (e.g., styrene, 1-hexene, etc.), alkynes (e.g., MeO2-
CCtCCO2Me, PhCtCH, etc.), and PhNCO. They react
with MeI to give the Ni-I derivative and N-Me-imidate
(40% conversion in 1 h at room temperature in C6D6),
but reactions with PhCH2Br and PhBr are more slug-
gish (<5% conversion to the Ni-Br derivative after 2 h
at 45 °C). Reacting 3-6 with LiCl leads to the formation
of the Ni-Cl species 2; an analogous exchange also
takes place with Br- and I- but not with F-. As expected
from pKa considerations, protonation of the Ni-N bond
in the imidato complexes 3-6 proceeds with relatively
strong acids HX (e.g., PhSH and HCl) to give the
corresponding Ni-X derivatives, but no reaction takes
place with less acidic substrates such as alcohols, water,
acetone, and acetonitrile. The Ni-N bond is also inert
toward PhSiH3. In contrast, the Ni-N bond in the
analogous complexes Cp*Ni(PEt3)(NHAr) reacts with
ROH and R3Si-H,31 whereas the Pt-N bond in L2Pt-
(Me)NR2 reacts with acetone and acetonitrile.13a
The tendency of Ni(II) to form 16-electron complexes
results in a substantial degree of allyl-ene distortion in
the Ind ligand; this “slippage” is reflected in the different
C-C bond lengths inside the five-membered-ring moiety
(C1-C7a and C3-C3a > C1-C2 and C2-C3) and the
Ni-C bond distances (Ni-C(3a,7a) > Ni-C(1-3)).25
A
second type of distortion in the coordination of the Ind
ligand in complex 3 is evident from the unsymmetric
nature of the Ni-C bonds, i.e., Ni-C1 > Ni-C3, which
results in a partial localization of bonding in the allyl
moiety of the Ind ligand (i.e., η3Tη1(C3), η2(C2dC1)).26
Thus, C3 resembles an sp3-hybridized carbon atom,
while C1 is closer to an sp2 hybridization, as reflected
in the closeness of the 13C{1H} NMR chemical shifts of
C1 (102.3 ppm) and C3 (62.6 ppm) in 3 to the corre-
sponding sp2 (ca. 139 ppm) and sp3 (ca. 45 ppm) carbon
atoms in free 1-methyl-substituted indenes.27 These
structural and spectroscopic features are also present
in complexes 1 and 2 and support our proposal4a,b that
(a) the Ind hapticity in complexes of the type IndNi-
(PPh3)X is strongly affected by the relative trans influ-
ences of the ancillary ligands PPh3 and X (Me ≈ PPh3
> Cl ≈ phthalimidato) and (b) the solid-state hapticities
of the Ind ligands in these compounds are maintained
in solution.
Reactivities of (1-Me-In d)Ni(P P h3)(im idato). Late-
transition-metal amido compounds have been impli-
cated in a number of important catalytic reactions,28
industrial processes,29 and biological phenomena.30 An
important reactivity is the promotion of N-C bond
(20) (a) Shaver, A.; Hartgerink, J .; Lal, R. D.; Bird, P.; Ansari, N.
Organometallics 1983, 2, 938 (PtII-η1(N)-imidato). (b) Price, S. J . B.;
DiMartino, M. J .; Hill, D. T.; Kuroda, R.; Mazid, M. A.; Sadler, P. J .
Inorg. Chem. 1985, 24, 3425 (AgI-and AuI-η1(N)-imidato). (c) Refer-
ence 19 (IrI-η1(N)-imidato and PdII-µ,η2(N,O)-imidato). (d) Treichel,
P. M.; Nakagaki, P. C.; Haller, K. J . J . Organomet. Chem. 1987, 327,
327 (MnI-η1(N)-imidato). (e) B.-Strzyzewska, M.; Tosik, A.; Wodka,
D.; Zakrzewski, J . Polyhedron 1994, 13, 1689 (FeII-η1(N)-imidato). (f)
Sahajpal, A.; Robinson, S. D.; Hursthouse, M. B.; Mazid, M. A. J . Chem.
Soc., Dalton Trans. 1993, 393 (RuII-µ,η2(N,O)-imidato).
(21) Holland, P. L.; Smith, M. E.; Andersen, R. A.; Bergman, R. G.
J . Am. Chem. Soc. 1997, 119, 12815.
(22) Ballester, L.; Perez, S.; Gutierrez, A.; Perpinan, M. F.; Guti-
errez-Puebla, E.; Monge, A.; Ruiz, C. J . Organomet. Chem. 1991, 414,
411.
(23) Churchill, M. R.; O’Brien, T. A. J . Chem. Soc. (A) 1970, 161.
(24) Smith, M. E.; Andersen, R. A. J . Am. Chem. Soc. 1996, 118,
11119.
Na tu r e of Ni-N In ter a ction s. The reactivities of
late-transition-metal amido compounds are linked to the
general lability of the M-N bond, a characteristic which
has been attributed to (a) the presumed mismatch
between the “soft” late metals and the “hard” amide
ligands (when R ) H, alkyl, aryl, etc.), and (b) the
destabilizing π-donation from the amide ligands into the
filled orbitals of the low-valent (electron-rich) late
metals.32 On the other hand, electrostatic-covalent (E-
C) bonding models describe M-X bonds primarily in
terms of ionic and covalent (σ) interactions, attributing
less importance to π-effects.33 Thus, M-N bonds may,
in principle, involve both E-C and π-donation interac-
tions, but the relative importance of these interactions
is open to debate and may vary in each specific case. In
this context, Bergman et al. have reported a detailed
study on the thermodynamics of the reactions intercon-
verting Cp*Ni(PEt3)X + HX′ and Cp*Ni(PEt3)X′ + HX
(X, X′ ) halides, NHR, OR, SR, etc.).31 According to
these authors, the Ni-X bond in these compounds has
a substantial electrostatic component and the relative
(25) (a) This type of “slip-fold” distortion may be measured in terms
of such parameters as the slip value, defined as ∆(Μ-C) ) 1/2(M-C3a
+ M-C7a) - 1/2(M-C1 + M-C3), and the hinge and fold angles (HA
and FA), representing the bending of the Ind ligand at C1/C3 and C3a/
C7a, respectively.25b,c The values of these parameters for 3 are very
close to those found for complexes 1 and 2:4a,b ∆Νi-C ) 0.27 Å, HA )
11.1°, and FA ) 12.3°. (b) Baker, R. T.; Tulip, T. H. Organometallics
1986, 5, 839. (c) Westcott, S. A.; Kakkar, A.; Stringer, G.; Taylor, N.
J .; Marder, T. B. J . Organomet. Chem. 1990, 394, 777.
(26) Note that [Ni-C1] - [Ni-C3] > 18σ.
(27) Edlund, U. Org. Magnet. Reson. 1978, 11, 516.
(28) (a) Dorta, R.; Egli, P.; Zu¨rcher, F.; Togni, A. J . Am. Chem. Soc.
1997, 119, 10857. (b) Wolfe, J . P.; Buchwald, S. L. J . Am. Chem. Soc.
1997, 119, 6054. (c) Widenhoefer, R. A.; Buchwald, S. L. Organome-
tallics 1996, 15, 2755 and references therein. (d) Casalnuovo, A. L.;
Calabrese, J . C.; Milstein, D. J . Am. Chem. Soc. 1988, 110, 6738.
(29) Roundhill, D. M. Chem. Rev. 1992, 92, 1.
(31) Holland, P. L.; Andersen, R. A.; Bergman, R. G.; Huang, J .;
Nolan, S. J . Am. Chem. Soc. 1997, 119, 12800.
(32) (a) Fryzuk, M. D.; Montgomery, C. D. Coord. Chem. Rev. 1989,
95, 1. (b) Bryndza, H. E.; Tam, W. Chem. Rev. 1988, 88, 1163.
(33) Drago, R. S. Applications of Electrostatic-Covalent Models in
Chemistry; Surfside: Gainesville, FL, 1994.
(30) Holm, R. H.; Kennepohl, P.; Solomon, E. I. Chem. Rev. 1996,
96, 2239.

Indenylnickel Imidato Complexes
Organometallics, Vol. 18, No. 1, 1999 35
stabilities of the various species are dependent on the
stabilization of the partial negative charge on X. This
reasoning, as opposed to π-bonding arguments, has been
used to account for the observations of higher relative
stabilities for Ni-OR over Ni-NR2 species (oxygen
being the more electronegative atom) and the direct
relationship between the binding of the amide ligands
and the electron-withdrawing nature of their N sub-
stituents.
coordination, presumably to avoid these interactions.
Similar spatial orientations have been adopted by
imidato ligands in other square-planar, d8 complexes.19,20a
Another possibility is a NifN back-bonding interaction,
which may be energetically feasible because the LUMO
of the imidato complexes is primarily the low-lying
imidato π* orbital. Such stabilizing π-back-bonding
would serve to strengthen the Ni-N bond and render
it relatively inert, consistent with our observations. A
pertinent example which illustrates this possibility is
presented by the d6 complexes CpFe(CO)L(phthalimi-
dato), in which the FefN back-bonding is evident from
the increase in the ν(CO,imide) values on going from a
more electron rich center (1640 cm-1 when L ) PPh3)
to a poorer center (1660 cm-1 when L ) CO).36
Con clu sion . The Ni-imidato compounds 3-6 are
much more stable than their Ni-amido counterparts
presumably because of the more strongly electron
withdrawing character of the N substituents in the
imidato complexes. The delocalization of the N lone pair
in the imidato ring and the rotation of the imidato
ligand away from the main plane of coordination serve
to reduce the overlap of the N lone pair with the HOMO
of the complex, thereby minimizing destabilizing NfNi
pπ-dπ interactions. As a result, the Ni-N bonds are
rather robust and inert in nucleophilic and insertion
reactions.
Since we were unable to prepare thermally stable
amido derivatives of our compounds, we could not
measure equilibrium constants for the interconversion
of these derivatives with various Ni-X analogues. We
have, therefore, relied on a combination of structural,
spectroscopic, and reactivity data to glean some infor-
mation on the character of the Ni-imidato bonds in the
1
present complexes. Thus, a VT H NMR study carried
out on complex 5 showed that the vinylic protons of the
maleimidato ligand are inequivalent at lower temper-
atures (coalescence temperature 233 K, ∆ν ) 84 Hz, ∆Gq
) 11.1 kcal/mol);34 this implies a hindered rotation of
the maleimidato ligand about the Ni-N axis which
renders the halves of this otherwise symmetric ligand
nonequivalent.
This hindrance might be caused by the steric interac-
tions between the imidate O lone pairs and the Me
group on the 1-Me-Ind ligand; similarly hindered rota-
tions about the M-N bonds in the complexes Cp*Ni-
(PEt3)(HN(p-C6H4Me)) (∆Gq ) ca. 11 kcal/mol)31 and
Cp*2Hf(H)(NHMe) (∆Gq ) ca. 10 kcal/mol)35 have been
attributed to steric factors. On the other hand, electronic
π-interactions may also be contributing to the observed
hindered rotation. For instance, freely rotating imidato
ligands would experience destabilizing pπ-dπ NfNi
interactions when they approach coplanarity with the
main plane of coordination. A reflection of this phenom-
enon is also observed in the solid-state structure of 3,
in which the phthalimidate moiety was found to be
rotated by ca. 76° with respect to the main plane of
Ack n ow led gm en t. We are grateful to the NSERC
(Canada), the FCAR (Que´bec), and the Universite´ de
Montre´al for financial support and to Prof. R. G.
Bergman and Dr. P. L. Holland for sending us their
then-unpublished manuscripts (refs 21 and 31) and for
useful discussions.
Su p p or tin g In for m a tion Ava ila ble: Complete details on
the X-ray analyses of 3, including tables of bond distances and
angles, anisotropic thermal parameters, and hydrogen atom
coordinates (27 pages). This material is contained in many
libraries on microfiche, immediately follows this article in the
microfilm version of this journal, can be ordered from the ACS,
and can be downloaded from the Internet; see any current
masthead page for ordering information and Internet access
instructions.
(34) The barrier was calculated using the Gutowsky-Holm ap-
proximation, which is based on the Eyring equation for a two-site
exchange process involving a molecualr rotation: Abraham, R. J .;
Loftus, P. Proton and Carbon-13 NMR Spectroscopy; Wiley: New York,
1985; pp 165-168.
OM980653C
(35) Hillhouse, G. L.; Bulls, A. R.; Santarsiero, B. D.; Bercaw, J . E.
Organometallics 1988, 7, 1309.
(36) Zakrzewski, J . J . Organomet. Chem. 1989, 359, 215.