Self-addition of metallacydic nickel enolate complexes stabilized by monodentate phosphine ligands

Article abstract of DOI:10.1021/om060183j

In contrast with their stable counterparts bearing the chelating diphosphines, the nickelacyclic O-enolate complexes stabilized by monodentate ligands Ni(OC(CHR)-o-C₆H₄)L₂ (R = H, Me, L = PMe₃; R = H, L = PMe₂Ph) undergo an unusual self-addition process leading to dinuclear alkoxo-hydroxo derivatives. The α-substituted enolate complexes (R = Me) are less prone to undergo this transformation, and the corresponding nickelacycles with L = PMe₃ or PMe₂Ph can be isolated in pure form.

Full text of DOI:10.1021/om060183j

3124
Organometallics 2006, 25, 3124-3129
Self-Addition of Metallacyclic Nickel Enolate Complexes Stabilized
by Monodentate Phosphine Ligands†
Juan Ca´mpora,* Inmaculada Matas, Celia M. Maya, Pilar Palma, and Eleuterio AÄ lvarez
Instituto de InVestigaciones Qu´ımicas, Consejo Superior de InVestigaciones Cient´ıficas-UniVersidad de
SeVilla, AVenida Ame´rico Vespucio 49, 41092 SeVilla, Spain
ReceiVed February 24, 2006
In contrast with their stable counterparts bearing the chelating diphosphines, the nickelacyclic O-enolate
complexes stabilized by monodentate ligands Ni(OC(dCHR)-o-C6H4)L2 (R ) H, Me, L ) PMe3; R )
H, L ) PMe2Ph ) undergo an unusual self-addition process leading to dinuclear alkoxo-hydroxo derivatives.
The R-substituted enolate complexes (R ) Me) are less prone to undergo this transformation, and the
corresponding nickelacycles with L ) PMe3 or PMe2Ph can be isolated in pure form.
hydes and other organic electrophilic reagents.7 We have
recently shown that the incorporation of a nickel enolate
functionality into a rigid metallacyclic framework hinders the
otherwise facile C-/O-isomerization and allows the examination
of the reactivity of the isomers separatedly.8,9 In addition, the
reactivity of nickelacyclic enolates can be exploited in a number
of synthetically useful transformations leading to the formation
of different types of cyclic and noncyclic organic products in a
highly selective manner.9,10 Unfortunately, the stabilization of
these metallacyclic enolates usually requires the use of chelating
ligands,7b,c,9b and this may have the undesired effect of blocking
some interesting reactions. For example, attempts to produce a
cyclic nickel enolate stabilized by PMe3 led to a dinuclear
product as a result of a formal aldol reaction9b (eq 1). We were
interested in gaining some insight into the mechanism of this
uncommon transformation, and to this end, we have studied
some further examples. In the course of these studies we have
isolated and structurally characterized the different species
involved in this process.
Introduction
Because of their relevance to many metal-mediated organic
transformations,1 late-transition metal enolate complexes have
received continued attention in recent years. The enolate
functionality displays a very rich coordination chemistry, and
several types of complexes displaying η3-coordinated2 or
bridging enolate ligands3 have been reported, although σ
coordination, either through its enol oxygen (O-enolate) or
carbon (C-enolate) atoms, predominate by far.4-7 The coordina-
tion mode influences the reactivity of the enolate functionality,
and while C-enolates often exhibit alkyl-like reactivity (e.g.,
C-C reductive couplings5 or migratory insertion6), O-enolates
display a characteristic nucleophilic reactivity, adding to alde-
† Dedicated to Dr. J. A. Abad (University of Murcia, Spain) on the
occasion of his retirement.
(1) (a) Paterson, I. In ComprehensiVe Organic Synthesis; Trost, B. M.,
Fleming, I., Heathcock, C. H., Eds.; Pergamon: Oxford, 1991; Vol. 2, pp
301-319. (b) Montgomery, J. Angew. Chem., Int. Ed. 2004, 43, 3890. (c)
Ikeda, S.; Miyashita, H.; Taniguchi, M.; Kondo, H.; Okano, M.; Sato, Y.;
Odashima, K. J. Am. Chem. Soc. 2002, 124, 12060. (d) Taggi, A. E.; Hafez,
A. M.; Lectka, T. Acc. Chem. Res. 2003, 36, 1. (e) Hamashima, Y.;
Sasamoto, N.; Hotta, D.; Somei, H.; Umebayashi, N.; Sodeoka, M. Angew.
Chem., Int. Ed. 2005, 44, 1525. (f) Hamashima, Y.; Hotta, D.; Umebayashi,
N.; Tsuchiya, Y. Suzuki, T. Sodeoka, M. AdV. Synth., Catal. 2005, 347,
1576.
(2) (a) Burkhardt, R. R.; Doney, J. J.; Bergman, R. G.; Heathcock, C.
H. J. Am. Chem. Soc. 1987, 109, 2022. (b) Slough, G. A.; Hayashi, R.;
Ashbaugh, J. R.; Shamblin, S. L.; Aukamp, A. M. Organometallics 1994,
13, 890. (c) Zuideveld, M. A.; Kamer, P. C. J.; van Leeuwen, P. W. N. M.;
Kleusener, P. A. A.; Stil, H. A.; Roobeek, C. F. J. Am. Chem. Soc. 1998,
120, 7977. (d) Sun, C.; Tu, A.; Slough, G. A. J. Organomet. Chem. 1999,
582, 235. (e) Murahashi, T.; Kurosawa, H. J. Organomet. Chem. 1999,
574, 142.
(3) (a) Albe´niz, A. C.; Catalina, N. M.; Espinet, P.; Redo´n, R.
Organometallics 1994, 13, 3587. (b) Fuji, A.; Hagiwara, E.; Sodeoka, M.
J. Am. Chem. Soc. 1999, 121, 5450. (c) Ruiz, J.; Rodr´ıguez, V.; Cutillas,
N.; Pardo, M.; Pe´rez, J.; Lo´pez, G.; Chalonner, P.; Hitchcock, P. B.
Organometallics 2001, 20, 1973. (d) Vicente, J.; Arcas, A.; Ferna´ndez-
Herna´ndez, J. M.; Bautista, D. Organometallics 2001, 20, 2767.
(4) (a) Burkhardt, R. R.; Bergman, R. G.; Heathcock, C. H. Organome-
tallics 1990, 9, 30. (b) Hartwig, J. F.; Bergman, R. G.; Andersen, R. A.
Organometallics 1991, 10, 3326. (c) Ketz, B. E.; Ottenwaelder, X. G.
Waymouth, R. M. Chem. Commun. 2004, 5693.
Results and Discussion
As reported previously,8,9b the reaction of 2-acetylphenyl
complex 1a with a stoichiometric amount of KtBuO in THF,
followed by workup involving simply solvent evaporation and
recrystallization from tetrahydrofuran, leads to the dinuclear
complex 3a, isolated as a powdery yellow solid, as shown in
Scheme 1. Monitoring the reaction by 31P{1H} NMR spectros-
copy reveals that the addition of the base immediately gives
rise to a single species characterized by a broad signal located
at ca. -20 ppm. On cooling to -80 °C, this signal splits into
two resonances, at δ -8.6 and -40.5 ppm, whereas addition
(5) Culkin, D. A.; Hartwig, J. F. Organometallics 2004, 23, 3398.
(6) (a) Veya, P.; Floriani, C.; Chiesi-Villa, A.; Rizzoli, C. Organome-
tallics 1993, 12, 4899. (b) Vicente, J. Arcas, A.; Ferna´ndez-Herna´ndez, J.
M. Bautista, D.; Jones, P. G. Organometallics 2005, 24, 2516.
(7) (a) Slough, G. A.; Bergman, R. G.; Heathcock, C. H. J. Am. Chem.
Soc. 1989, 111, 938. (b) Amarasinghe, K. K. D.; Chowdhury, S. K.; Heeg,
M. J.; Montgomery, J. Organometallics 2001, 20, 370. (c) Ketz, B. E.;
Cole, A. P.; Waymouth, R. M. Organometallics 2004, 23, 2835.
(8) Ca´mpora, J.; Maya, C. M.; Palma, P.; Carmona, E.; Gutierrez-Puebla,
E.; Ruiz, C. J. Am. Chem. Soc. 2003, 125, 1482.
(9) (a) Campora, J.; Maya, C. M.; Palma, P.; Carmona, E.; Graiff, C.;
Tiripicchio, A. Chem. Commun. 2003, 1742. (b) Ca´mpora, J.; Maya, C.
M.; Palma, P.; Carmona, E.; Gutie´rrez, E.; Ruiz, C.; Graiff, C.; Tiripicchio,
A. Chem. Eur. J. 2005, 11, 6889.
(10) Lozanov, M.; Montgomery, J. J. Am. Chem. Soc. 2002, 124, 2106.
10.1021/om060183j CCC: $33.50 © 2006 American Chemical Society
Publication on Web 05/17/2006

Metallacyclic Nickel Enolate Complexes
Organometallics, Vol. 25, No. 13, 2006 3125
Scheme 1
of HCl to the solution regenerates the starting complex, 1a, as
expected for the enolate complex 2a.9b Attempts to isolate this
intermediate invariably result in its transformation into 3a.
Despite the anhydrous conditions employed, the formation of
the latter compound can be envisaged as a result of a combina-
tion of hydrolysis and aldol addition reactions (Scheme 1). To
hinder the addition step, we decided to explore the reactivity
of the R-methylated precursor 1b, which should give rise to a
less reactive enolate complex.
Like 1a, the 2-propionylphenyl derivative 1b can be readily
prepared from Ni(cod)2, PMe3, and the corresponding aryl
chloride. The structure of this compound has been established
by X-ray diffraction (see below).
ortho hydrogen in the neighborhood of the central methyne
(d(H-H): 2.87 Å), while the other does not (shortest H-H
distance, 3.44 Å). The calculation indicates that the energy
difference between the two isomers is negligible (less than 0.5
kcal/mol), suggesting a kinetic origin for the observed stereo-
selectivity.
As found for 1a, NMR monitoring of the reaction of 1b with
KtBuO shows the initial formation of a single species, 2b,
characterized by a broad 31P resonance. The position of this
signal can vary within a few ppm from one experiment to other,
but is similar to that of 2a (ca. -15 ppm).
Rather surprisingly, the yields obtained for 3a or 3b do not
improve upon deliberate addition of water to the reaction
The reaction of 1b with KOtBu follows the same path
observed for 1a and affords the corresponding dinuclear
compound 3b, after the usual (nonhydrolytic) workup. The
spectroscopic features of 3b match closely those of 3a. For
instance, the OH and CO functionalities give rise to character-
mixture. In a separate experiment, it was observed that the 31
P
spectrum of 2b is not altered by the addition of water.
Accordingly, water does not trigger the transformation of 2 into
3. Indeed, 2b is fairly stable in THF solution at room
temperature, the appearance of significant amounts of 3b being
detected only upon attempted isolation of the product. Thus,
removal of the reaction solvent under vacuum followed by
extraction of the residue with freshly distilled THF results in
an almost complete transformation of 2b to 3b. Suspecting that
this change is induced by the loss of PMe3, an extra amount of
this ligand was added to the extraction solvent (diethyl ether,
in this case). Under these conditions, a significant amount of
2b remains present and was readily separated from the less
soluble complex 3b. Cooling the ethereal extract led to the
istic absorptions in the IR spectrum at 3400 and 1640 cm-1
,
respectively. The 31P{1H} NMR spectrum shows two pairs of
singlets (δ -5.9, -10.9, and -6.1, -9.7 ppm, respectively)
with intensity ratio 4:1, suggesting that this compound is formed
as a diastereomeric pair. This conclusion is further supported
by the proton spectrum, where two complete sets of signals with
the same relative intensity ratio are observed. The signals
corresponding to the CH-CH3 units were located with the help
of the COSY spectrum. In the 2-D NOESY 1H NMR spectrum,
the major isomer displays a NOE cross-peak linking the
resonances of the central methyne group and one of the doublet
(“ortho”) aromatic resonances. On this basis and with the help
of a DFT (B3LYP/6-31G*) molecular model (Figure 1), the
identity of the stereoisomers can be assigned. As can be seen,
only one of the two structures (R,R or S,S) displays an aromatic
1
isolation of a sample of analytically pure 2b. The H and 13C
NMR spectra of 2b are analogous to those of analogous complex
Ni(OC(dCHMe)-o-C6H4)(dippe) (dippe ) 1,2-bis(diisopropyl-
phosphino)ethane),9b confirming our previous assumption that
this compound has a cyclic O-enolate structure. Thus, the vinyl
enolate proton gives rise to a quartet resonance at δ 5.21, while
the R- and â-olefinic 13C resonances appear at 169.0 and 86.5
ppm, respectively. However, in contrast with the chelating
phosphine derivative, 2b undergoes rapid exchange of the two
chemically inequivalent PMe3 ligands, which give rise to only
one signal in the 1H, 13C, and 31P NMR spectra. The fluxional
process removes the 13C-31P couplings from the metallacycle
signals (clearly observed for the dippe complex), suggesting a
dissociative phosphine exchange mechanism.
To avoid phosphine loss by evaporation, we have replaced
PMe3 by the less volatile PMe2Ph ligand. Treatment of the
corresponding precursors, 1c and 1d, with KtBuO leads to
different results (Scheme 2). The reaction of the 2-acetyl
complex 1c follows the same course observed for 1a and 1b,
Figure 1. Molecular models of the 3b diastereomeric pair (PMe3
ligands have been replaced by PH3): (A) major isomer (R,R or
S,S), showing the C(Et)H‚‚‚H(ortho) approach; (B) minor isomer
(R,S or S,R).

3126 Organometallics, Vol. 25, No. 13, 2006
Ca´mpora et al.
Scheme 2
and the initially formed cyclic enolate, 2c (identified by its
characteristic 31P{1H} spectrum, a broad resonance centered at
-13 ppm) is readily transformed into the corresponding dimeric
hydroxide 3c upon the usual workup. However, complex 1d
affords the stable cyclic enolate 2d, which does not evolve to
the corresponding dinuclear species. The stability of 2d is
probably due to the combination of the decreased reactivity of
the R-substituted enolate functionality with the lower volatility
of the PMe2Ph ligands. As observed for 2b, the NMR spectra
of 2d evidence a rapid exchange of the phosphine ligands at
room temperature. Compounds 3c and 2d have been isolated
as crystalline solids, and their X-ray diffraction structures are
shown in Figures 3 and 4 (vide infra).
to trigger this reaction indicates that hydrolysis of the enolate
complexes 2 is not involved in the process, or at least does not
constitute its rate-limiting step. Even so, the aldol-like structure
of the dinuclear complexes 3 suggests the intermediacy of a
keto intermediate at some stage. However, although the hy-
drolysis of 2 would originate a carbonyl functionality amenable
to nucleophilic attack, we have observed that 1b is indefinitely
stable in the presence of 2b. It is also conceivable that a keto
group could be generated by the previous isomerization of half
of the O-enolate molecules to the C-form. This is also an
unlikely possibility, as we have shown before that O-/C-enolate
isomerization is a very slow process at room temperature in
these rigid metallacyclic complexes.8,9b Moreover, the C-enolate
form would be strongly disfavored in the case of the R-meth-
ylated complex 2b.5
Several features of the self-addition reaction experienced by
complexes 2 are mechanistically valuable. The inability of water
The dissociable phosphine ligands play an important role in
this reaction. The overall stoichiometry of the addition process
involves the net loss of one phosphine unit per Ni atom. Thus,
it is not surprising that the transformation of the relatively stable
enolate 2b is favored when the volatile PMe3 ligand is removed
by evaporation and hindered by addition of extra amounts of
this ligand. The low volatility of PMe2Ph prevents the trans-
formation of 2d, and, in general, the use of the chelating
diphosphine dippe allows the isolation of both substituted and
nonsubstituted (Ni-O-C(R)dCH2) cyclic enolates.8,9b Hence,
it appears likely that phosphine dissociation may occur at the
initial stages of the self-addition reaction. The highly fluxional
behavior of compounds 2 is also consistent with facile phosphine
dissociation. Nickel hydroxides and alkoxides stabilized by
monodentate phosphine ligands display a strong tendency to
form oxygen-bound dimers of type D11 (Scheme 3), and
Figure 2. ORTEP view of 1b. Selected bond lengths (Å) and
angles (deg): Ni1-C1, 1.885(3); Ni1-P1, 2.1907(10); Ni1-P2,
2.1775(10); Ni1-Cl1, 2.2375(11); Ni1-O1, 2.472(2); C7-O1,
1.222(4); C1-Ni1-O1, 77.60(11); Ni1-C1-C2, 118.5(2); P1-
Ni1-P2, 169.15(4).
Figure 4. ORTEP perspective of compound 3c. Selected bond
distances (Å) and angles (deg): Ni1-O1, 1.8945(10); Ni1-O3,
1.8986(11); Ni2-O1, 1.9060(10); N1-P1, 2.1436(4); Ni2-P2,
2.1221(4); Ni1-C4, 1.8790(14); Ni2-C16, 1.8839(14); C9-O2,
1.2227(18); C1-O1, 1.4665(16); Ni1‚‚‚Ni2, 2.8312(3); Ni1-O1-
Ni2, 96.31(4); Ni-O3-Ni2, 96.08(5); Ni1-O1-Ni2-O3, 21.15-
(5).
Figure 3. ORTEP perspective of compound 2d. Selected bond
distances (Å) and angles (deg): Ni1-C1, 1.932(2); Ni1-O1,
1.8595(15); Ni-P1, 2.1391(6); Ni1-P2, 2.2400(7); O1-C(7),
1.347(3); C7-C8, 1.337(3); C1-Ni1-O1, 85.76(8); P1-Ni1-P2,
89.90(3).

Metallacyclic Nickel Enolate Complexes
Organometallics, Vol. 25, No. 13, 2006 3127
Scheme 3
although such compounds have not been observed in our system,
association of complexes 2 in solution through their enolate
oxygen atoms appears likely. The associative process brings
two enolate fragments into proximity and favors the addition
process. Obviously, the structure of the doubly bridged species
D is too rigid to allow any interaction, but a singly bridged
intermediate, C, may be flexible enough to allow the addition
reaction. Self-addition of enolates is a highly unusual process,
and we are unaware of any literature precedent. Therefore, we
consider likely that the C-C bond formation step may actually
be preceded by the tautomerization of one of the enolate units
within the dimeric species C to the C-coordination mode. The
interaction of the enolate oxygen with a second metal center
may facilitate the opening and the subsequent reorganization
of the metallacycle, as indicated in Scheme 3 (inset). In our
previous work on the dippe-based system, we observed that
lithium salts efficiently induce the C-/O-enolate tautomerization,
probably due to the same kind of effect. In the resulting
intermediate E, the nucleophilic and electrophilic centers are
too far apart to interact, but simple ligand reorganization to F
would enable the aldol addition to proceed, the initial addition
products being rapidly trapped by traces of water present in the
reaction medium.
functionality, by increasing its Bro¨nsted acidity of the R-H atoms
and favoring the selective formation of the O-bound enolate
tautomer.
The molecules of 2d display a slightly distorted square-planar
geometry (Figure 3). The Ni-O distance, 1.8595(15) Å, is
somewhat shorter than in the complexes Ni(R)(OAr)(dippe)
(1.907 Å),11a despite the similar coordination environment of
the enolate and aryloxide moieties. However, it is comparable
to those of terminal nickel hydroxides and methoxides13 and
identical, within the experimental error, to that found in the
dippe-containing analogue of 2d.9b Noteworthy, the enolate Cd
C bond length (1.337(3) Å) is shorter than in the dippe complex
(1.368(6) Å) and approaches that of a typical CdC bond. Other
metrical parameters associated with the metallacyclic unit are
very similar in the PMe2Ph and the dippe compounds. As
expected, the P-Ni-P (98.90(3)°) is somewhat wider in 2d
than in the diphosphine derivative (88.49(6)°).
The crystal structure of 3c confirmed our structural proposal
and shows essentially the same features anticipated by the DFT
models discussed above (Figure 4). This unusual molecule
displays a bridging tridentate organic ligand, which originates
five- and seven-membered metallacyclic moieties. The largest
one adopts a boat conformation, with the keto group pointing
away from the nickel center. The Ni-OH and Ni-OR bonds
display similar lengths (ca. 1.9 Å), comparable to those found
in dinuclear hydroxo14 and alkoxo15 nickel complexes. The
central Ni2O2 core is not planar, but slightly puckered (dihedral
angle: 21.13(5)°), and the Ni-Ni distance (2.8312(3) Å) can
be considered normal, very likely not involving any metal-
metal bonding interaction.
X-ray Crystal Structures of Complexes 1b, 2d, and 3c.
The crystal structures of the title compounds are shown in
Figures 2, 3, and 4. As can be seen in Figure 2, compound 1b
displays a pentacoordinated structure, with the carbonyl oxygen
atom occupying the apical position of a distorted square-planar
coordination polyhedron. The coordination of the carbonyl group
causes a decrease of ca. 60 cm-1 of the IR frequency of the
ν(CdO) band, relative to the organic precursor. The Ni-O bond
length (2.472(2) Å) is somewhat shorter than in the previously
reported complex Ni(C(Ph)dC(H)COCH2SiMe3)(Cl)(PMe3)2,12
which displays a similar interaction, characterized by a Ni-O
distance of 2.535 Å. The configuration of the propionyl fragment
is highly reminiscent of that observed in the enolate 2d (vide
infra). Notwithstanding its weakness, the nickel-carbonyl
interaction may play a role in the deprotonation of the acyl
Conclusions
2-Acylaryl derivatives 1a-d are cleanly deprotonated by
potassium tert-butoxide, leading to the metallacyclic enolate
complexes 2a-d. Unlike the analogous complexes containing
(13) (a) Ca´mpora, J.; Del Rio, D.; Palma, P.; AÄ lvarez, E. Organometallics
2004, 23, 1652. (b) Ca´mpora, J.; Palma, P.; Del Rio, D.; Conejo, M. M.;
AÄ lvarez, E. Organometallics 2004, 23, 5653. (c) Ca´mpora, J.; Matas, I.;
Palma, P. Graiff, C.; Tiripicchio, A. Organometallics 2005, 24, 2827.
(14) Carmona, E.; Marin, J. M.; Palma, P.; Paneque, M.; Poveda, M. L.
Inorg. Chem. 1989, 28, 3983.
(15) (a) Ogoshi, S.; Oka, M.; Kurosawa, H. J. Am. Chem. Soc. 2004,
126, 11802. (b) Bertrand, J. A.; Kirkwood, C. E. Inorg. Chim. Acta 1970,
4, 192. (c) Lu, Z.; White, C.; Rheingold, A. L.; Crabtree, R. H. Angew.
Chem., Int. Ed. 1993, 32, 92.
(11) (a) Ca´mpora, J.; Lo´pez, J. A.; Maya, C.; Palma, P.; Carmona, E.;
Valerga P. J. Organomet. Chem. 2002, 643-644, 331. (b) Klabunde, U.;
Ittel, S. D. J. Mol. Catal. 1987, 41, 123. (c) Kalamarides, H. A.; Iyer, S.;
Lipian, J.; Rhodes, L. F.; Day, C. Organometallics 2000, 19, 3983.
(12) Carmona, E.; Gutie´rrez-Puebla, E.; Monge, A. Mar´ın, J. M.;
Paneque, M.; Poveda, M. L. Organometallics 1989, 8, 967.

3128 Organometallics, Vol. 25, No. 13, 2006
Ca´mpora et al.
(s, CarH), 135.8 (t, JCP ) 18 Hz, Car), 137.8 (s, CarH), 143.2 (s,
Car), 168.2 (t, JCP ) 36 Hz, CarNi), 200.9 (s, CO). 31P{1H} NMR
(CD2Cl2, 20 °C, 121 MHz): δ 9.8.
the chelating diphosphine dippe, complexes 2 experience an
unusual self-addition reaction leading to the dinuclear alkoxo-
hydroxide complexes 3. This transformation can be envisaged
as a combination of hydrolysis and aldol reaction. The latter
process is not initiated by the presence of water, but it is
facilitated by the release of the monodentate phosphine ligand,
especially in the case of the volatile PMe3. It is likely that the
mechanism of this reaction is facilitated by the formation of a
dinuclear intermediate bridged through an O-enolate functional-
ity.
Synthesis of Ni(C6H4-o-C(O)CH2CH3)(Cl)(PMe2Ph)2 (1d). To
a cooled solution (-78 °C) of 1.1 g (4 mmol) of Ni(cod)2 in 50
mL of toluene was added 1.14 mL (8 mmol) of dimethylphen-
ylphosphine. The mixture was allowed to reach room temperature,
and 0.54 mL (4 mmol) of 2-chloropropiophenone was added. The
reaction mixture was heated at 45 °C for 4 h. The solvent was
removed under vacuum and the residue extracted with 10 mL of
THF. Dark red crystals of the pure product were isolated after
crystallization at -20 °C. Yield: 58%. Anal. Calcd for C25H31-
ClNiOP2: C, 59.62; H, 6.20. Found: C, 59.74; H, 6.21. IR (Nujol
mull): ν(CdO), 1644 cm-1. 1H NMR (CD2Cl2, 20 °C, 400 MHz):
δ 0.89 (bs, 6H, PMeMePh), 1.19 (t, 3H, 3JHH ) 7.2 Hz, CH2CH3),
Experimental Section
All preparations were carried out under oxygen-free nitrogen by
conventional Schlenk techniques. Solvents were rigorously dried
under nitrogen and degassed before use. Microanalyses were
performed by the Microanalytical Service of the Instituto de
Investigaciones Qu´ımicas (Sevilla, Spain). Infrared spectra were
recorded on a Bruker Vector 22 spectrometer, and NMR spectra
on Bruker DRX 300 and 400 MHz spectrometers. The 1H and 13C-
{1H} resonances of the solvent were used as the internal standard,
but the chemical shifts are reported with respect to TMS. 31P
resonances are referenced to external 85% H3PO4. The compound
2-chloropropiophenone was prepared as described previously.9b
Molecular modeling was carried out with the Spartan 04 package.16
Synthesis of Ni(C6H4-o-C(O)CH2CH3)(Cl)(PMe3)2 (1b). A 1
M solution of PMe3 in toluene (6 mL, 6 mmol) was added to a
suspension of Ni(cod)2 (825 mg, 3 mmol) in toluene (50 mL) at
-78 °C. The mixture was allowed to warm to room temperature,
and 2-chloropropiophenone (0.41 mL, 3 mmol) was added. The
resulting solution was stirred at 45 °C for 4 h, and the initial yellow
color turned dark red. The solvent was evaporated under vacuum
and the residue extracted with Et2O (20 mL). After partial
concentration of the solution and cooling to -20 °C, the product
was isolated as a red-brown solid. Yield: 75%. IR (Nujol mull):
ν(CdO), 1639 cm-1. Anal. Calcd for C15H27ClNiOP2: C, 47.48;
3
1.39 (s, 6H, PMeMePh), 2.79 (q, 2H, JHH ) 7.2 Hz, CH2CH3),
3
3
6.74 (t, 1H, JHH ) 7.4 Hz, CarH), 6.82 (t, 1H, JHH ) 7.1 Hz,
CarH), 7.20 (d, 1H, 3JHH ) 7.4 Hz, CarH), 7.26-7.35 (m, 6H, CarH),
7.44 (bs, 4H, CarH). 13C{1H} NMR (CD2Cl2, 20 °C, 75 MHz): δ
8.8 (s, CH2CH3), 11.5 (t, JCP ) 14 Hz, PMeMePh), 12.9 (t, JCP
)
13 Hz, PMeMePh), 31.4 (s, CH2CH3), 121.6 (s, CarH), 128.2 (s,
CarH), 129.2 (s, CarH), 129.4 (s, CarH), 130.1 (s, CarH), 135.9 (t,
2
JCP ) 19 Hz, Car), 138.0 (s, CarH), 142.8 (s, Car), 167.9 (t, JCP
)
36 Hz, CarNi), 203.5 (s, CO). 31P{1H} NMR (CD2Cl2, 20 °C, 121
MHz): δ -9.9 (s).
Synthesis of 2b. Potassium tert-butoxide (57 mg, 0.5 mmol) was
dissolved in 3 mL of THF and added to a cooled solution (-78
°C) of 190 mg of complex 1b (0.5 mmol) in 10 mL of THF. After
reaching room temperature, the solvent was removed under reduced
pressure, and 8 mL of a solution of PMe3 (0.1 mmol) in Et2O was
added to afford an orange solution. The solution was filtered from
a yellow precipitate of compound 3b and cooled to -20 °C to give
2b as orange crystals. Yield: 15%. Anal. Calcd for C15H26NiOP2:
C, 52.52; H, 7.64. Found: C, 52.99; H, 7.89. IR (Nujol mull): ν-
(CdC), 1614 cm-1. 1H NMR (C6D6, 20 °C, 400 MHz): δ 0.87 (d,
18H, 2JHP ) 7.7 Hz, PMe3), 2.29 (d, 3H, 3JHH ) 6.7 Hz, dCHCH3),
3
1
5.21 (q, 1H, JHH ) 6.7 Hz, dCHCH3), 7.05 (m, 1H, CarH), 7.10
H, 7.17. Found: C, 48.18; H, 7.14. H NMR (C6D6, 20 °C, 300
3
(m, 2H, CarH), 7.58 (d, 1H, JHH ) 7.3 Hz, CarH). 13C{1H} NMR
MHz): δ 0.78 (t, 18H, *JHP ) 3.8 Hz, PMe3), 1.24 (t, 3H, 3JHH
)
7.3 Hz, CH2CH3), 2.76 (q, 2H, 3JHH ) 7.3 Hz, CH2CH3), 6.73 (m,
1H, CarH), 6.89 (dt, 1H, 3JHH ) 7.3 Hz, 4JHH ) 1.5 Hz, CarH), 7.44
(C6D6, 20 °C, 100 MHz): δ 11.8 (s, CHCH3), 16.3 (d, JCP ) 22
Hz, PMe3), 86.5 (s, CHCH3), 121.6 (s, CarH), 123.9 (s, CarH), 124.2
(s, CarH), 136.7 (s, CarH), 155.8 (bs, Car-Ni), 157.4 (s, Car-CO),
169.0 (s, CO). 31P{1H} NMR (C6D6, 20 °C, 162 MHz): δ -13.7
(bs, PMe3).
4
(dd, 1H, 3JHH ) 7.8 Hz, JHH ) 1.3 Hz, CarH), 7.80 (dd, 1H, 3JHH
4
)7.5 Hz, JHH ) 1.0 Hz, CarH). 13C{1H} NMR (C6D6, 20 °C, 75
1
MHz): δ 8.9 (s, CH2CH3), 12.7 (t, JCP ) 13 Hz, PMe3), 31.5 (s,
Synthesis of 2d. A solution of 0.27 g (2.3 mmol) of KtBuO in
5 mL of THF was added to a cooled solution (-78 °C) of 1.05 g
of 1d (2.1 mmol) dissolved in 10 mL of THF. After reaching room
temperature, solvent was removed in vacuo, and 20 mL of toluene
was added. After centrifugation, the solution was taken to dryness
and the residue extracted with 10 mL of THF. Concentration of
this solution and cooling at -20 °C afforded the product as red
crystals. Yield: 42%. Anal. Calcd for C25H30NiOP2: C, 64.28; H,
6.47. Found: C, 63.90; H, 6.44. IR (Nujol mull): ν(CdC), 1573
cm-1. 1H NMR (CD2Cl2, 20 °C, 300 MHz): δ 1.49 (d, 12 H, 2JHP
) 8.0 Hz, PMe2Ph), 1.81 (d, 3H, 3JHH ) 6.8 Hz, CHCH3), 4.72 (q,
CH2CH3), 120.7 (s, CarH), 128.7 (s, CarH), 129.5 (s, CarH), 137.0
2
(t, JCP ) 4 Hz, CarH), 142.4 (s, Car-CO), 171.4 (t, JCP ) 35 Hz,
Car-Ni), 201.9 (s, CO). 31P{1H} NMR (C6D6, 20 °C, 121 MHz): δ
-16.2 (PMe3).
Synthesis of Ni(C6H4-o-C(O)CH3)(Cl)(PMe2Ph)2 (1c). Di-
methylphenylphosphine (1.42 mL, 10 mmol) was added to a cooled
solution (-78 °C) of 1.37 g (5 mmol) of Ni(cod)2 in 50 mL of
toluene. The mixture was allowed to reach room temperature, and
0.65 mL (5 mmol) of 2-chloroacetophenone was added. The reaction
mixture was heated at 45 °C for 4 h. The solvent was removed
under vacuum and the residue extracted with 10 mL of THF. Dark
red crystals of the pure product were obtained after adding some
diethyl ether and cooling at -20 °C. Yield: 50%. Anal. Calcd for
C24H29ClNiOP2: C, 58.88; H, 5.97. Found: C, 58.40; H, 6.07. IR
(Nujol mull): ν(CdO) 1642 cm-1. 1H NMR (CD2Cl2, 20 °C, 300
MHz): δ 0.87 (bs, 6H, PMe2Ph), 1.40 (bs, 6H, PMe2Ph), 2.43 (s,
3H, CH3), 6.75 (td, 1H, 3JHH ) 7.4 Hz, 4JHH ) 1.1 Hz, CarH), 6.82
3
1H, JHH ) 6.8 Hz, CHCH3), 6.55-6.62 (m, 2H, CarH), 6.79 (m,
2H, CarH), 7.10 (m, 1H, CarH), 7.34-7.43 (m, 5H, CarH), 7.76 (t,
3
4H, JHP ) 6.8 Hz, CarH). 13C{1H} NMR (CD2Cl2, 20 °C, 75
1
MHz): δ 10.9 (s, CHCH3), 15.3 (d, JCP ) 24 Hz, PMe2Ph), 87.0
(s, CHCH3), 120.6 (s, CarH), 123.4 (s, CarH), 124.2 (s, CarH), 128.9
2
2
(d, JCP ) 9 Hz, CarH), 130.2 (s, CarH), 131.6 (d, JCP ) 11 Hz,
CarH), 137.7 (s, CarH), 154.4 (s, Car), 155.9 (s, Car), 167.9 (s,OC).
31P{1H} NMR (CD2Cl2, 20 °C, 121 MHz): δ 0.4 (bs).
3
4
3
(td, 1H, JHH ) 7.3 Hz, JHH ) 1.7 Hz, CarH), 7.18 (dd, 1H, JHH
4
) 7.4 Hz, JHH )1.4 Hz, CarH), 7.27-7.34 (m, 5H, CarH), 7.37-
7.46 (m, 5H, CarH). 13C{1H} NMR (CD2Cl2, 20 °C, 75 MHz): δ
11.4 (bs, PMeMePh), 13.0 (bs, PMeMePh), 26.3 (s, CH3), 121.6
(s, CarH), 128.2 (s, CarH), 129.2 (s, CarH), 129.5 (s, CarH), 131.0
Synthesis of the Dinuclear Complex 3b. Potassium tert-
butoxide (115 mg, 1 mmol) was dissolved in 5 mL of THF and
added to a cooled solution (-78 °C) of 370 mg of complex 1b (1
mmol) in 20 mL of THF. After reaching room temperature, the
solvent was removed under vacuum, 10 mL of THF was added,
(16) Spartan 04; Wavefunction, Inc.: Irvine, CA, 2004.

Metallacyclic Nickel Enolate Complexes
Organometallics, Vol. 25, No. 13, 2006 3129
2
and the solution was concentrated, affording 3b as a powdery yellow
solid. This solid retains variable amounts of solvent, preventing
good analytical data from being collected. Yield: 50%. IR (Nujol
techniques (SHELXTL-6.12)20 minimizing w[Fo - Fc2]2. The
methyl groups of a phosphine ligand in the structure of 1b were
found disordered. Therefore a model was included, and the
occupancy factors for the methyl groups on that phosphine were
fixed to 0.5. All non-hydrogen atoms were refined with anisotropic
thermal parameters. Hydrogen atoms were included in calculated
positions (except for those in compound 2d, which were included
in the refinement from observed positions and fully isotropically
refined) and were calculated and allowed to ride on the attached
carbon atoms with the isotropic temperature factors (Uiso values)
fixed at 1.2 times (1.5 times for methyl groups) the Ueq values of
the corresponding carbon atoms.
1
mull): ν(OH), 3397 cm-1; ν(CdO), 1639 cm-1. H NMR (CD2-
Cl2, 20 °C, 300 MHz) 8:2 mixture of isomers; major isomer: δ
-5.65 (s, 1H, OH), 0.27 (t, 3H, 3JHH ) 7.2 Hz, CH2CH3), 0.69 (m,
2
1H, CHHCH3), 0.90 (d, 9H, JHP ) 9.9 Hz, PMe3), 0.95 (m,
obscured by resonance at 0.90 ppm, 1H, CHHCH3), 1.24 (d, 9H,
2JHP ) 9.2 Hz, PMe3), 2.71 (q, 1H, 3JHH ) 7.5 Hz, CHCH3), 3.21
(d, 3H, 3JHH ) 7.5 Hz, CHCH3), 6.27 (d, 1H, 3JHH ) 7.4 Hz, CarH),
3
3
6.35 (d, 1H, JHH ) 7.5 Hz, CarH), 6.63 (td, 1H, JHH ) 7.0 Hz,
4JHH ) 1.1 Hz, CarH), 6.74-6.90 (m, 3H, CarH), 7.07 (dd, 1H, 3JHH
) 7.2 Hz, 4JHH ) 1.6 Hz, CarH), 7.62 (d, 1H, 3JHH ) 7.3 Hz, CarH);
Crystal data for 1b: C15H27ClNiOP2, Mr ) 379.47, orange
needle (0.48 × 0.16 × 0.14 mm3) from diethyl ether, orthorhombic,
space group Pca21 (no. 29), a ) 18.8920(6) Å, b ) 9.2032(3) Å,
c ) 10.5612(3) Å, V ) 1836.24(10) Å3, Z ) 4, Fcalcd ) 1.373 g
3
minor isomer: δ -5.58 (s, 1H, OH), 0.41 (t, 3H, JHH ) 7.2 Hz,
CH2CH3), 0.92 (d, 9 H, 2JHP ) 10.2 Hz, PMe3), 1.24 (d, 9 H, 2JHP
) 10.2 Hz, PMe3), 1.08 (d, 3H, 3JHH ) 7.7 Hz, CHCH3), 3.62 (q,
1H, JHH ) 7.7 Hz, CHCH3). 31P{1H} NMR (CD2Cl2, 20 °C, 121
cm-3, λ(Mo KR1) ) 0.71073 Å, F(000) ) 800, µ ) 1.370 mm-1
,
3
MHz) major isomer: δ -5.9, -10.9; minor isomer: δ -6.1, -9.7.
Synthesis of the Dinuclear Complex 3c. Potassium tert-butoxide
(168 mg, 1.4 mmol) was added to a solution of 1c (632 mg, 1.3
mmol) in THF (50 mL) at -78 °C. After stirring the mixture at
room temperature for 4 h, the solvent was removed under reduced
pressure. The resulting yellow solid was extracted with THF (10
mL), the suspension centrifuged to separate the KCl, and the
solution partially concentrated. After addition of some diethyl ether
and cooling at -20 °C, compound 3c was obtained as yellow
crystals. Yield: 40%. Anal. Calcd for C32H36Ni2O3P2: C, 59.32;
H, 5.60. Found: C, 59.20; H, 5.72. IR (Nujol mull): ν(OH) 3639
T ) 100(2) K; 21 331 reflections were collected in the range 5.78°
< 2θ < 61.06°, index ranges -26 e h e 25, -13 e k e 9, -14
e l e 15, of which 5477 were independent [R(int) ) 0.0365] and
5103 with I > 2σ(I ) were observed; final R1 [I > 2σ(I )] ) 0.0455,
2
wR2 (all data) ) 0.1084, w ) [σ2(Fo ) + (0.0348P)2 + 2.74P]-1
2
where P ) (Max(Fo , 0) + 2Fc2)/3; no. of data/restraints/parameters
5477/64/195; goodness of fit on (F2) ) 1.106. In the final difference
map, the highest peaks (to ca. 0.97 e Å-3) were close to disordered
methyl groups of one phosphine ligand.
Crystal data for 2d: C25H30NiOP2, Mr ) 467.14, red block (0.36
× 0.26 × 0.24 mm3) from tetrahydrofuran, monoclinic, space group
P21/n (no. 14), a ) 10.8446(9) Å, b ) 16.8315(14) Å, c ) 12.7181-
(11) Å, â ) 102.542(2)°, V ) 2266.0(3) Å3, Z ) 4, Fcalcd ) 1.369
1
cm-1; ν(CdO) 1652 cm-1. H NMR (CD2Cl2, 20 °C, 500 MHz):
δ -5.76 (s, 1H, OH), 0.63 (s, 3H, CH3), 0.94 (bs, 3H, PMeMePh),
1.00 (d, 3H, 2JHP ) 9.8 Hz, PMe2Ph), 1.19 (d, 3H, 2JHP ) 9.9 Hz,
g cm-3, λ(Mo KR1) ) 0.71073 Å, F(000) ) 984, µ ) 1.011 mm-1
,
2
PMeMePh), 1.32 (d, 3H, JHP ) 7.6 Hz, PMeMePh), 3.10 (d, 1H,
T ) 100(2) K; 14 369 reflections were collected in the range 4.08°
< 2θ < 56.92°, index ranges -14 e h e 12, -22 e k e 20, -16
e l e 16, of which 5223 were independent [R(int) ) 0.0311] and
4099 with I > 2σ(I ) were observed; final R1 [I > 2σ(I )] ) 0.0390,
2JHH ) 18.2 Hz, CHH), 5.11 (d, 1H, 2JHH ) 18.2 Hz, CHH), 5.95
(d, 1H, 3JHH ) 7.2 Hz, CarH), 6.37 (dd, 1H, 3JHH ) 7.6 Hz, 4JHH
)
1.0 Hz, CarH), 6.42 (dd, 1H, 3JHH ) 7.6 Hz, 4JHH ) 1.2 Hz, CarH),
3
2
6.71 (t, 1H, JHH ) 7.4 Hz, CarH), 6.93 (m, 2H, CarH), 7.19 (dd,
wR2 (all data) ) 0.0884, w ) [σ2(Fo ) + (0.0454P)2 + 0.0000P]-1
1H, JHH ) 6.9 Hz, JHH ) 2.2 Hz, CarH), 7.44 (bs, 6H, CarH),
where P ) (Fo2 + 2Fc2)/3; no. of data/restraints/parameters 5223/
0/382; goodness of fit on (F2) ) 0.972. In the final difference map,
the highest peaks (to ca. 0.84 e Å-3) were close to the nickel atom.
Crystal data for 3c: C32H36Ni2O3P2P2, Mr ) 647.97, prism
yellow (0.20 × 0.20 × 0.11 mm3) from diethyl ether, triclinic, space
group P1h (no. 2), a ) 11.6560(6) Å, b ) 12.2387(6) Å, c )
12.8339(7) Å, R ) 66.2720(10)°, â ) 63.7110(10)°, γ ) 69.2970-
(10)°, V ) 1468.02(13) Å3, Z ) 2, Fcalcd ) 1.466 g cm-3, λ(Mo
KR1) ) 0.71073 Å, F(000) ) 676, µ ) 1.423 mm-1, T ) 100(2)
K; 31 912 reflections were collected in the range 5.98° < 2θ <
61.04°, index ranges -16 e h e 16, -17 e k e 17, -18 e l e
18, of which 8891 were independent [R(int) ) 0.0261] and 7513
with I > 2σ(I ) were observed; final R1 [I > 2σ(I )] ) 0.0281, wR2
3
4
3
7.68 (d, 1H, JHH ) 6.7 Hz, CarH), 7.85 (m, 2H, CarH), 7.99 (bs,
2H, CarH). 13C{1H} NMR (CD2Cl2, 20 °C, 75 MHz): δ 10.58 (d,
1JCP ) 29 Hz, PMeMePh), 10.68 (d, JCP ) 29 Hz, PMeMePh),
1
1
13.7 (bs, PMeMePh), 14.3 (d, JCP ) 6 Hz, PMeMePh), 26.4 (s,
CH3), 64.2 (s, CH2), 82.9 (s, C-O), 121.3 (s, CarH),123.2 (s, CarH),
123.5 (s, CarH), 124.2 (s, CarH), 126.1 (s, CarH), 127.0 (s, CarH),
128.8 (s, CarH), 129.2 (d, 1JCP ) 9 Hz, Car), 130.4 (s, CHar), 130.7
(s, CHar), 132.0 (bs, CHar), 133.9 (d, 1JCP ) 39 Hz, Car), 137.9 (s,
CHar), 138.3 (bs, CHar), 141.2 (bs, Car-Ni), 147.2 (d, 2JCP ) 44 Hz,
Car-Ni), 149.3 (s, Car-C-O), 168.3 (s, Car-CdO), 206.5 (s, CdO).
31P{1H} NMR (CD2Cl2, 20 °C, 202 MHz): δ 9.5 (s), -0.3 (s).
X-ray Crystallography. Crystals coated with dry perfluoropoly-
ether were mounted on a glass fiber and fixed in a cold nitrogen
stream (T ) 100(2) K). Intensity data were collected on a Bruker-
Nonius ×8Apex-II CCD diffractometer (1b and 3c) or a Bruker
SMART Apex CCD diffractometer (2d), both equipped with a Mo
KR1 radiation (λ ) 0.71073 Å) source and graphite monochromator.
The data were reduced (SAINT)17 and corrected for Lorentz
polarization and absorption effects by a multiscan method (SAD-
ABS).18 The structure was solved by direct methods (SIR-2002)19
and refined against all F2 data by full-matrix least-squares
2
(all data) ) 0.0715, w ) [σ2(Fo ) + (0.0353P)2 + 0.5261P]-1 where
2
P ) (Fo + 2Fc2)/3; no. of data/restraints/parameters 8891/1/363;
goodness of fit on (F2) ) 1.057. In the final difference map, the
highest peaks (to ca. 0.65 e Å-3) were close to nickel atoms.
Acknowledgment. Financial support from the DGI (Project
PPQ2003-000975) and Junta de Andaluc´ıa is gratefully ac-
knowledged. I.M. thanks a research fellowship from the Consejo
Superior de Investigaciones Cient´ıficas (I3P program). We thank
Prof. Tim Gallagher for helpful discussions.
(17) SAINT 6.02; Bruker-AXS, Inc.: Madison, WI 53711-5373, 1997-
1999.
(18) Sheldrick, G. SADABS; Bruker AXS, Inc.: Madison, WI, 1999.
(19) Program SIR2002: Burla, M. C.; Camalli, M.; Carrozzini, B.;
Cascarano, G. L.; Giacovazzo, C.; Polidori, G.; Spagna R. J. Appl.
Crystallogr. 2003, 36, 1103.
Supporting Information Available: X-ray crystallographic file
in CIF format is available free of charge via the Internet at
http://pubs.acs.org.
(20) SHELXTL 6.14; Bruker AXS, Inc.: Madison, WI, 2000-2003.
OM060183J