Cyclic enolates of Ni and Pd: Equilibrium between C- and O-bound tautomers and reactivity studies

Article abstract of DOI:10.1002/chem.200500622

2-Acylaryl complexes of Ni and Pd containing chelating diphosphines react with KtBuO to give metallacyclic enolate complexes. While coordination through the carbon atom is preferred in the case of Pd, the nickel O-enolate compounds are formed as the corresponding O-tautomers. Slow equilibration between O- and C-enolate tautomers is observed for the nickel complex with an unsubstituted enolate function (M-O-C=CH₂). Theoretical DFT calculations suggest that the barrier for the tautomer exchange has its origin in the rigidity of the metallacycle. Whilst the C-enolate tautomer is unreactive towards aldehydes, the corresponding O-enolate adds to MeCHO and PhCHO, giving rise to products that retain the enolate functionality. The carbonylation of these products cleanly leads to the formation of enol lactones in a highly selective manner.

Full text of DOI:10.1002/chem.200500622

FULL PAPER
DOI: 10.1002/chem.200500622
Cyclic Enolates of Ni and Pd: Equilibrium between C- and O-Bound
Tautomers and Reactivity Studies
Juan Cµmpora,*[a] Celia M. Maya,[a] Pilar Palma,[a] Ernesto Carmona,*[a]
Enrique GutiØrrez,[b] Caridad Ruiz,[b] Claudia Graiff,[c] and Antonio Tiripicchio[c]
Abstract: 2-Acylaryl complexes of Ni
and Pd containing chelating diphos-
phines react with KtBuO to give metal-
lacyclic enolate complexes. While coor-
dination through the carbon atom is
preferred in the case of Pd, the nickel
O-enolate compounds are formed as
the corresponding O-tautomers. Slow
equilibration between O- and C-eno-
late tautomers is observed for the
nickel complex with an unsubstituted
allacycle. Whilst the C-enolate tauto-
mer is unreactive towards aldehydes,
the corresponding O-enolate adds to
MeCHO and PhCHO, giving rise to
products that retain the enolate func-
tionality. The carbonylation of these
products cleanly leads to the formation
of enol lactones in a highly selective
manner.
ꢀ ꢀ
enolate function (M O C=CH2). The-
oretical DFT calculations suggest that
the barrier for the tautomer exchange
has its origin in the rigidity of the met-
Keywords:
aldol reactions
·
lactones · metallacycles · nickel ·
O ligands
Introduction
direct consequence of the workcarried out in the last deca-
des. Thus, transition-metal enolates have become indispensa-
ble in a number of valuable synthetic methodologies,[2] in-
cluding some that are effected catalytically.[1e,f,h–k,3] Optimiza-
tion of these methodologies demands a deeper understand-
ing of their mechanistic details.
The chemistry of late-transition-metal enolates is a subject
of considerable synthetic interest due to the high levels of
selectivity that such compounds introduce in many organic
reactions.[1] The detailed study of these complexes is essen-
tial for the development of new synthetic applications.
Indeed, many of those currently known have emerged as a
Earlier studies have disclosed that the coordination mode
of the enolate ligand exerts a strong influence in its reactivi-
ty.[2a,4] For instance, in contrast to O-enolates, coordination
through the carbon atom often causes low enolate-like reac-
tivity and a chemical behaviour more akin to that of metal
alkyls, for example, toward migratory insertion.[2e,5] Al-
though some aldol-type additions of C-bound enolates are
known,[2f,6] the active participation of the O-bound tautomer
in these transformations cannot be ruled out, as the two
forms may have similar energies. Both O- and C-binding
have been ascertained for late transition metals,[2b–d] but the
latter is more common for the softer Lewis acids derived
from the heavier elements of these groups.[7,5c] Accordingly,
while for nickel O- and C-enolates are common,[8,2f] in gen-
eral palladium tends to favour C-enolate coordination.[7,5c]
In some systems the tautomeric C- and O-enolate forms
have been found to exist.[7,2d]
[a] Dr. J. Cµmpora, Dr. C. M. Maya, Dr. P. Palma, Prof. E. Carmona
Departamento de Química Inorgµnica
Instituto de Investigaciones Químicas
Universidad de Sevilla
Consejo Superior de Investigaciones Científicas
Centro de Investigaciones Científicas “Isla de la Cartuja”
Avda. AmØrico Vespucio, s/n, Isla de la Cartuja, 41092 Sevilla (Spain)
Fax : (+34)954-460-565
E-mail: campora@iiq.csic.es
guzman@us.es
[b] Prof. E. GutiØrrez, Dr. C. Ruiz
Instituto de Ciencia de Materiales
Consejo Superior de Investigaciones Científicas
Campus de Cantoblanco, 28049 Madrid (Spain)
[c] Dr. C. Graiff, Prof. A. Tiripicchio
To gain more insight into the different chemical behaviour
of the two isomeric structures, we have prepared Ni and Pd
enolate complexes for which interconversion between the O
and C isomers is hindered due to their metallacyclic nature.
Following Vicenteꢀs preparation of cyclic palladium enolates
Dipartimento di Chimica Generale ed Inorganica
Chimica Analitica, Chimica Fisica, Università di Parma
Parco Area delle Scienze 17 A, 43100 Parma (Italy)
Supporting information for this article is available on the WWW
under http://www.chemeurj.org/ or from the author.
Chem. Eur. J. 2005, 11, 6889 – 6904
ꢁ 2005 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
6889

such
as
[PK d{CH2C(O)-2-CL H-3,4,5-(OMe)3}(tmed)][7e]
be attributed to the organic carbonyl unit of 2, whereas the
palladium-bound enolate carbon resonates at d=52.1 ppm,
as a doublet of doublets, with trans and cis JCP couplings of
6
(tmed=N,N,N’,N’-tetramethylethylenediamine) and our
2
own extensive workon metallacycles of nickel and palladi-
um of composition [MK (CH2CMe2-2-CL H4)(L)2][9] for differ-
71 and 9 Hz, respectively. In the aliphatic region of the
6
3
ent mono- and bidentate neutral L ligands, we have attempt-
ed the preparation of related enolate structures using o-ace-
tylphenyl derivatives as starting materials. The results of
these studies are reported herein. Part of this workhas ap-
peared in preliminary form.[10]
1H NMR spectrum a triplet due to accidentally equal JHP
couplings between the Pd-CH2 protons and the two inequi-
valent 31P nuclei appears at d=2.26 ppm (7.6 Hz). These
and other spectroscopic data collected in the Experimental
Section are similar to those reported for the related pallada-
cycle [PK d(CH2CMe2-2-CL H4)(PMe3)2].[9a,b]
6
Results and Discussion
Synthesis of nickel enolates: Extension of this chemistry to
the analogous nickel compounds is not straightforward. As
represented in Scheme 2 the synthesis of chloro- or bro-
Palladium compounds: Since experimentally the nickel
system involves more difficulties than the palladium com-
plexes, the latter are described first.
Treatment of the chloro-bridged palladium allyl dimer
[{Pd(h3-C3H5)(m-Cl)}2] with NaC5H5, followed by the addi-
tion of PMe3 and CH2=CHCO2Me is known to generate the
reactive Pd0 species [Pd(CH2=CHCO2Me)(PMe3)2].[11] This
olefin complex is a suitable source of Pd0 and its use circum-
vents the experimental complications due to the byproduct
contamination commonly associated with the well-known
Pd0/PPh3, or other related compounds. As represented in
Scheme 1, its reaction with 2’-bromoacetophenone at 508C,
Scheme 2.
moaryl compounds related to 1 can be readily accomplished
by using [Ni(cod)2] and the appropriate haloacetophenone
substrate. The corresponding complexes, 3 and 4, are isolat-
ed as orange crystalline solids. At variance with IR data for
1, n(C=O) data for these compounds (ca. 1630 cmꢀ1) is ap-
proximately 70 cmꢀ1 lower than in the corresponding organ-
ic precursors, suggesting the existence of a weakaxial inter-
action between the acetyl group and the Ni atom, as previ-
ously observed in the crystal structure of a related Ni com-
pound.[12] Also in contrast with the Pd complex 1, treatment
of 3 and 4 with KtBuO does not afford the expected cyclic
enolate with a structure analogous to that of 2. Instead the
dinuclear, sparingly soluble, compound 5 is obtained and
isolated in the form of a yellow fine powder. The proposed
structure for 5 finds support in microanalytical and spectro-
scopic data (see Experimental Section). The presence of a
coordinated hydroxide ligand is evinced by an IR band at
3300 cmꢀ1 and by a 1H NMR resonance with a chemical
shift (d=ꢀ5.46 ppm) similar to that reported for other tran-
sition-metal hydroxide compounds.[13] The two inequivalent
PMe3 ligands appear as singlets (d=ꢀ10.7 and ꢀ5.0 ppm) in
the 31P{1H} NMR spectrum, whereas the analysis of the aryl
region of the 1H NMR spectrum reveals the presence of
eight multiplets that correspond to two nonequivalent Ni–
(ortho-substituted)phenyl units, as deduced also from 2D
Scheme 1.
allows the isolation of the bromoaryl complex 1 in the form
of a colourless crystalline solid. The observation of a 31P{1H}
singlet at d=ꢀ18.4 ppm and of a virtually coupled triplet in
1
both the H (d=0.92; Jap =3.6 Hz) and the 13C{1H} (d=13.9;
Jap =15 Hz) NMR spectra is indicative of a trans-trimethyl-
phosphine arrangement. Moreover the similarity between
the n(CO) values in the IR spectra of 1 (1665 cmꢀ1) and the
starting organic reagent, namely, o-bromoacetophenone (ca.
1700 cmꢀ1) suggests little or no apical interaction at all be-
tween the metal and the acetyl oxygen atom. This observa-
tion is in accord with the strong tendency of PdII to remain
four-coordinate.
COSY and H-13C one-bond heterocorrelation experiments.
1
The organic ligand that bridges the two nickel centres of 5
results from the aldolic condensation of the acetylphenyl
ligand of two molecules of either 3 or 4. Even though the
formation of 5 requires action of adventitious water, the de-
liberate addition of H2O does not improve the yield of 5.
The use of other bases like Na[N(SiMe3)2] or TlEtO gives
rise to complex reaction mixtures. To gain additional infor-
Reaction of the bromoaryl compound 1 with KtBuO gives
the C-enolate 2 as air-stable white crystals. The coordination
of the carbon atom of the enolate terminus within the met-
1
allacyclic structure is supported by IR, H and 13C{1H} NMR
spectroscopic data. Thus, an IR absorption at 1630 cmꢀ1 can
6890
www.chemeurj.org
ꢁ 2005 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Chem. Eur. J. 2005, 11, 6889 – 6904

Metallacyclic Enolates
FULL PAPER
mation on this complex transformation 31P{1H} NMR moni-
toring of the conversion of 3 to 5 was undertaken at ꢀ808C.
Deprotonation appears to be fast and complete under these
conditions. Moreover, a new species that gives broad reso-
nances at d=8.6 and 40.5 ppm is observed. Quenching the
reaction by addition of HCl gives backthe starting com-
pound 3. It thus appears that a highly reactive nickel eno-
late, possibly analogous to 2, is initially formed. Neverthe-
less its lability prevents its isolation as a pure complex. In
an attempt to increase the stability of this species the nature
of the ancillary ligand has been systematically modified and
the effects of this variation monitored.
enolate 2, spectroscopic data for the nickel derivative 8a are
consistent with coordination of the oxygen atom of the eno-
late functionality. For example, the two enolic carbon atoms
give 13C resonances at d=176.1 and 74.4 ppm, the latter
1
being characterized by a JCH coupling of 154 Hz, indicative
of sp2 hybridization. Additionally, an IR absorption at
1590 cmꢀ1 may be assigned to n(C=C) of the O-enolate
ligand, whereas the higher energy band that could be ex-
pected for n(C=O) (at 1630 cmꢀ1 in 2) is conspicuously
absent. As reported in a preliminary form, definite structur-
al comfirmation of the proposed O-enolate coordination has
ꢀ
been provided by X-ray studies, that reveal Ni O and C=C
Treatment of [Ni(cod)2] with 2’-bromoacetophenone, in
the presence of 2,2’-bipyridyl furnishes a complex reaction
mixture from which no pure compound can be isolated. The
same holds for the analogous reaction with iPr2PCH2CH2-
PiPr2 (dippe), although useful compounds of this diphos-
phine may be isolated by a subtle modification of this reac-
tion procedure (see below). At variance with these observa-
tions the use of PPh3 and Ph2PCH2CH2PPh2 (dppe) permits
the isolation of the corresponding NiII aryl compounds (Ex-
perimental Section). Nonetheless their reactions with
KtBuO, or related basic reagents, lead to very unstable and
complex reaction mixtures from which no clean products
could be separated.
The diphosphine iPr2PCH2CH2PiPr2 (dippe), has proved
to be the most suitable auxiliary ligand for the synthesis of
aryl–nickel organometallic complexes that would serve as
precursors for enolate compounds. While as mentioned
above its addition to the [Ni(cod)2]/2’-bromoacetophenone
reaction mixture proves fruitless, the use of 2’-chloroaceto-
phenone (Scheme 3a) permits the isolation of the desired
chloro–o-acetylphenyl–nickel compound 6a. Long reaction
times are needed (3 days, 458C), but 6a can be isolated by
this procedure in 70–80% yield. Alternatively, addition of
dippe to 3 gives a mixture of 6a and the cationic complex 7
(Scheme 3b) when Et2O is employed as the reaction solvent,
whereas the use of toluene produces compound 6a as the
main reaction product.
bond lengths within the metal enolate linkage of 1.857(10)
and 1.31(2) Š, respectively.[10] The related enolates 8b and
8c, featuring Me substitution at one or the two olefinic sites
of 8a can be prepared by the same methodology (Scheme
4a). The chloroaryl precursors needed for these syntheses
are not commercially available, but are obtained by the re-
action of 2-chlorobenzonitrile with the appropriate Grignard
reagent, followed by acid hydrolysis of the resulting
imine.[14] Their oxidative addition to [Ni(cod)2] in the pres-
ence of dippe (458C, 24 h) yields the expected nickel ortho-
substituted aryls 6b and 6c, as orange crystalline solids, with
spectroscopic properties similar to those of 6a. Deprotona-
tion of 6b and 6c to the enolates 8b and 8c, respectively,
occurs upon treatment with KtBuO. Interestingly, whereas
the conversion of 6c to 8c requires stirring for 12 h at room
temperature, the formation of 8b is more facile and takes
place spontaneously during the synthesis of 6b, even in the
absence of base, although only to a limited extent. Signifi-
cant decomposition takes place if the heating is prolonged.
As represented in Scheme 4b, compound 6b can also be ob-
tained if a more reactive h2-butene complex is used as a pre-
cursor. This may be generated in situ and then oxidatively
reacted with 2’-chloropropiophenone to give 6b, accompa-
nied with minor amounts of enolate 8b. The latter becomes
the main reaction product when the transformation is per-
formed in the presence of NEt3. These observations indicate
that the ease of deprotonation increase in the order 6c<
6a<6b. While the formation of 8b benefits of the stabiliz-
ing effect of the alkylation of the double bond, the unfav-
ourable steric interaction of the mutually cis aromatic and
methyl groups renders the de-
Treatment of a solution of 6a in THF with one equivalent
of KtBuO allows the preparation of the nickel enolate 8a in
good yields (Scheme 3a). At variance with the related Pd
protonation of 6c difficult.
Compounds 8b and 8c ex-
hibit spectroscopic properties
similar to those of 8a. An IR
absorption centred around
1600 cmꢀ1 may be attributed to
the stretching of the C=C
bond, while in the 1H NMR
spectrum the olefinic proton of
8b resonates at d=5.27 ppm as
a quartet, due to coupling with
the adjacent methyl protons
(3JHH =6.7 Hz). For 8c the ole-
Scheme 3.
finic methyl substituents give
Chem. Eur. J. 2005, 11, 6889 – 6904
ꢁ 2005 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chemeurj.org
6891

E. Carmona, J. Cµmpora et al.
ꢀ
O-enolate ligand. The Ni O
bond length of 1.853(3) Š is
within the 1.85–2.10 Š range
ꢀ
characteristic of Ni O distan-
ces in square-planar alkoxide,
aryloxide and hydroxide nickel
derivatives with terminal O-
bound ligands.[15] The olefinic
ꢀ
C7 C8 bond length is normal
(1.368(6) Š), being only slight-
ly longer than in ethylene
(1.34 Š) and, as expected,
ꢀ
much shorter than the Csp2 Csp3
bond between the olefinic
carbon C8 and the methyl sub-
ꢀ
stituent (C8 C9=1.496(7) Š).
These features compare well
with those of other nickel com-
pounds containing O-bound
Scheme 4.
rise to singlets at d=2.40 and 2.42 ppm. NOESY experi-
ments reveal that the former resonance corresponds to the
methyl group trans to the oxygen atom. In compound 8b it
is the vinylic H atom that occupies the position in trans with
respect to the oxygen atom. Since the quality of the X-ray
data obtained for 8a is not very high and therefore a precise
analysis of its bonding parameters is not justified, we have
also characterized the analogous enolate 8b by X-ray single-
crystal studies. Figure 1 shows an ORTEP view of the struc-
ture, which is characterized by a slightly distorted square-
enolates[8] or related functionalities.[16]
By means of a similar methodology, summarized in
Scheme 5, a related nickel O-enolate 12 has been prepared.
The starting nickel alkenyl can be obtained from trans-[Ni-
ꢀ
planar coordination of the metal. The Ni P1 distance of
ꢀ
2.163(1) Š is shorter than the Ni P2 (2.206(1) Š) reflecting
Scheme 5.
the higher trans influence of the aryl as compared with the
(CH3)Cl(PMe3)2], by successive insertion of CO and PhCꢁ
CH, as reported previously.[12] Whereas its direct reaction
with KtBuO gives rise to a complex mixture of products,
prior PMe3 substitution by dippe, followed by deprotonation
permits isolation of the desired compound 12. The spectro-
scopic features of this compound are similar to those of
8a,b. Hence, it is likely that all these compounds show simi-
lar chemical reactivity, although that of 12 has not been ex-
plored.
Equilibria between C- and O-bound enolates—mechanistic
studies on their formation: The preparation of the nickel
enolates by the reaction of the appropriate halo aryl precur-
sors with KtBuO leads selectively to the O-bound tautomers
8a–8c. Nevertheless at temperatures around 508C (Scheme
3a) solutions of compound 8a in different solvents equili-
brate slowly with the C-enolate 9. At
room temperature the conversion of
8a into 9 is very slow, thereby indicat-
ing a kinetic control of the selectivity
in the formation of the O-enolate 8a
by the route depicted in Scheme 3a.
A reasonable explanation for this se-
lectivity is that the deprotonation of
Figure 1. View of the molecular structure of complex 8b together with
the atomic numbering system. Selected bond lengths [Š] and angles [8]:
ꢀ
ꢀ
ꢀ
ꢀ
Ni O1 1.853(3), Ni C1 1.913(4), Ni P1 2.1627(12), Ni P2 2.2064(12),
ꢀ
ꢀ
ꢀ
ꢀ
ꢀ
O1 C7 1.342(5), C7 C8 1.368(6), C6 C7 1.464(6), C1 C6 1.398(6), C8
C9 1.496(7); O1-Ni-C1 86.17(15), C1-Ni-P1 99.77(12), O1-Ni-P2
86.10(10), O1-C7-C8 119.7(4), Ni-O1-C7 115.4(2), O1-C7-C6 113.0(3),
C6-C7-C8 127.3(4), C7-C8-C9 122.6(5), P1-Ni-P2 88.50(5).
6892
www.chemeurj.org
ꢁ 2005 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Chem. Eur. J. 2005, 11, 6889 – 6904

Metallacyclic Enolates
FULL PAPER
the acetyl group by the base is
helped by previous coordina-
tion of the oxygen atom to the
Ni center, as indicated.[17]
The isomer ratio 8a/9 varies
very little in the solvents used
(from ca. 3:1 in THF to ca.
1.6:1 in benzene, toluene or
cyclohexane, at 508C). Mea-
surement of Keq for 8a/9 in tol-
uene at several temperatures
in the range 42–1128C pro-
vides the following thermody-
namic
values:
DG298
DH=
=
ꢀ0.31(1) kcalmolꢀ1
;
ꢀ0.28(1) kcalmolꢀ1 and DS=
+0.12(3) calmolꢀ1 Kꢀ1, indicat-
ing that the reaction is essen-
tially thermoneutral. The equi-
Scheme 7.
libration follows first-order kinetics[18] over the temperature
range from 52 to 928C and it is characterized by activation
parameters DH° =18.5(3) kcalmolꢀ1, DS° =ꢀ22(1) calm-
olꢀ1 Kꢀ1, and DG°298 =25.3(3) kcalmolꢀ1. A concerted mecha-
nism, with an ordered h3-oxoallyl transition state may be
suggested. Indeed, h3-oxoallyl structures have been pro-
posed as intermediates for the interconversion of C- and O-
enolates.[2d] It is worth mentioning in this regard that the de-
protonation of 6a by Li[N(CHMe2)2] yields a 3:1 mixture of
8a/9. Moreover, the addition of LiCl to solutions of 8a in
THF at room temperature causes an almost instant isomeri-
zation of 8a to a 3:1 mixture of 8a/9. Accordingly, LiCl is
proposed to catalyze the tautomerization as depicted in
Scheme 6, that is, via an oxoallyl intermediate structure.
the Pd–enolate carbon bond of 13. Chloride abstraction
occurs upon reaction of 6a and 14 with NaBPh4, yielding
the desired cationic derivatives 15 and 16. Metal coordina-
tion of the carbonyl oxygen atom in the two complexes is
hinted by the low n(CO) values of 1575 and 1580 cmꢀ1,
found for 15 and 16, respectively. Note that in the nickel–
aryl complexes 3, 4 and 6a, ortho substituted by an acetyl
group, the corresponding n(CO) IR bands appear around
1630 cmꢀ1, while the frequencies are close to 1700 cmꢀ1 in
the organic precursors. An X-ray structural determination
carried out with the nickel compound 15 (Figure 2) demon-
ꢀ
strates the coordination of the acetyl unit. The Ni O1 bond
length of 1.935(2) Š is, as expected, longer than in the
ꢀ
nickel enolate 8b. Also in 15 the Ni P2 distance of
2.1425(7) Š is shorter than the
ꢀ
Ni P1 (2.2202(6) Š), reflect-
ing the higher trans influence
of the aryl as compared with
the O-carbonyl of the acetyl
ligand. The bond length of the
ꢀ
carbonyl O1 C2 1.259(3) is
quite normal. Compound 15
reacted with KtBuO to yield
Scheme 6.
the O-bound nickel enolate
8a, but the Pd complex gave
Additional experiments on the formation of the nickel
and palladium enolates, 8a and 2, respectively, were per-
formed with the aim of improving our mechanistic knowl-
edge. Firstly, to checkwhether the deprotonation of the
acetyl group coordinated through the oxygen atom to nickel
(and palladium) is feasible, the cationic compounds 15 and
16 (Scheme 7b) were prepared by treatment of the chloroar-
yl 6a and 14 with NaBPh4. The chloroaryl palladium com-
plex 14 cannot be obtained directly by oxidative addition of
2’-chloroacetophenone to the zerovalent palladium(0) pre-
cursor. Nevertheless, as represented in Scheme 7a, com-
pound 14 is easily obtained from the enolate 2 by means of
PMe3 displacement by dippe, followed by HCl addition to
the C-bound palladium analogue, 13, as the only observable
product. This result illustrates once more the strong prefer-
ence of Pd for the C-bound configuration.
Secondly, if the formation of the O-enolate 8a is the
result of an intermolecular attackby tBuOꢀ onto the acetyl
group, (vide supra) then an intramolecular version of this
process, whereby an alkoxide or aryloxide base has been in-
corporated to the metal coordination sphere, could be feasi-
ble, although alcohol elimination should give rise to the C-
enolate isomer 9. With this aim in mind the trans-nickel–
aryl–aryloxide complexes 17a and 17b were synthesized
(Scheme 8). Both react rapidly with dippe at room tempera-
ture and while partial decomposition occurs, one of the
Chem. Eur. J. 2005, 11, 6889 – 6904
ꢁ 2005 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chemeurj.org
6893

E. Carmona, J. Cµmpora et al.
tion was not attempted in these cases. Since the iPr groups
allow for a number of conformers, the calculation of the
structures of these compounds was initiated by a molecular
mechanics conformational analysis, by using the MerckMo-
lecular Force Field. The positions of the atoms in the metal-
lacyclic unit were fixed, and the structures of the five more
stable conformers were optimized with the DFT without im-
posed restrictions. The results of this study are summarized
in Table 1.
Table 1. Results of the theoretical study.
M
R
O-Enolate TS C-Enolate DE(O!C) Barrier(O!C)
[kcalmolꢀ1
]
[kcalmolꢀ1
]
Figure 2. View of the structure of the cationic portion of compound 15 to-
gether with the atomic numbering system. Selected bond lengths [Š] and
Ni
Pd
H
H
O-Ni
O-Pd
TS C-Ni
C-Pd
TS’ C-Ni’
TS’’ C-Ni’’
+2.07
+7.23
+2.11
+0.03
25.2
–
25.4
25.4
ꢀ
ꢀ
ꢀ
ꢀ
angles [8]: Ni O1 1.935(2), Ni C4 1.939(2), Ni P1 2.2202(6), Ni P2
–
ꢀ
ꢀ
ꢀ
ꢀ
2.1425(7), O1 C2 1.259(3), C1 C2 1.482(3), C2 C3 1.463(3), C3 C4
1.420(3); O1-Ni-C4 85.25(9), C4-Ni-P2 96.96(8), O1-Ni-P1 89.87(6), O1-
C2-C1 118.9(2), Ni-O1-C2 114.0(2), O1-C2-C3 117.1(2), C1-C2-C3
124.02(2), P1-Ni-P2 88.97(3).
Ni Me O-Ni’
Ni iPr O-Ni’’
Figure 3 shows the structures calculated for the model
complexes O/C-Ni, O/C-Pd, and TS. Some key distances
and angles are shown in the figure. The geometric parame-
ters of substituted model compounds O/C-Ni’ and O/C-Ni’’
show no significant differences to those of the dpe models
and have been omitted for the sake of simplicity. The availa-
bility of the crystal structures of compounds 8a and 8b
allow for some interesting comparisons. In general, the
structural features of these compounds are well reproduced
in the model complexes. As in the experimental structure,
the metallacyclic unit is essentially planar. The calculated
major species of both reactions is the expected C-enolate 9,
while the O-enolate 8a was not observed.
DFT study of the interconversion of C- and O-enolate tau-
tomers: To provide theoretical support to the mechanism
proposed for the exchange between the tautomeric enolate
forms, and to gain a better understanding of the properties
of these complexes, DFT calculations were performed on
model complexes of the simplified diphosphine ligand
H2PCH2CH2PH2 (from now on dpe) by using the BP86
functional and the numerical basis set DN*, as implemented
in the Spartan Pro package. For Ni complexes, this study has
been further expanded to the methyl and isopropyl P-substi-
tuted complexes, the latter corresponding to the full repre-
sentation of the real compounds. In the case of the simple
model containing the dpe ligand, the calculation of the sta-
tionary points was verified with a frequency calculation. The
structures of the models containing substituted diphosphines
were built from those of the dpe complexes and optimized
with the same DFT method, although a frequency calcula-
ꢀ
Ni O bond length (1.859 Š) matches well the experimental
values (8a: 1.857(10); 8b: 1.853(3) Š). Other important pa-
rameters are also correctly reproduced. For example, the
ꢀ ꢀ
ꢀ
Ni O C (8a: 118.0(8)8; 8b: 115.4(2)8) and O C=C (8a:
119(1)8; 8b: 119.7(4)8) angles are close to the calculated
values, 115.9 and 121.28, respectively. The exocyclic double
bond is slightly bent out of the plane of the metallacycle, in
order to avoid the nearby aromatic hydrogen atom. This
small distortion is also found in the real complexes. The Pd
O-enolate displays very similar features. In contrast with the
Scheme 8.
6894
www.chemeurj.org
ꢁ 2005 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Chem. Eur. J. 2005, 11, 6889 – 6904

Metallacyclic Enolates
FULL PAPER
alkyl groups replace the hydrogen atoms. Since the energy
calculation does not take into account the solvation effects,
it can be concluded that the calculation reasonably reprodu-
ces the thermoneutral character of the O-Ni/C-Ni equilibri-
um.
As discussed previously, the experimental data suggest
that for the Ni derivatives the equilibration of the tautomers
could involve an intermediate h3-enolate species. According-
ly, we have searched for such an intermediate. Starting with
the more simple dpe ligand, two stationary points were lo-
cated in the energy surface, each of them characterized by
one imaginary frequency. They correspond to transition
states in which the Ni atom interacts simultaneously with
the C and O atoms of the enolate fragment, and differ only
in the conformation of the Ni–dpe chelate. These transition
states were used as a starting point to build the structures
corresponding to the methyl and isopropyl derivatives, and
in the latter case the structures of the ten (=2”5) more
stable conformers were fully optimized. Since the conforma-
tional change of the diphosphine ligand is thought to be
facile, we have selected the lower energy transition states
for the calculation of the energy barrier. The structure of
the transition state corresponding to the dpe ligand is shown
in Figure 3. As can be seen, the O-C-CH2 unit is approxi-
mately perpendicular to the coordination plane (93.38), al-
lowing a p-interaction with the Ni atom. The lengths of the
ꢀ
ꢀ
Ni O and Ni C bonds suggest that the interaction is not
symmetrical, and can be considered closer to h2-C=O than a
3
-
–
h -pseudoallylic interaction. The relatively long C- - O and
short C- - C bonds further support this view; hence, an early
-
–
transition state in the O!C reaction coordinate is indicat-
ed.[19] The energy difference between TS and O-Ni (ca.
25.2 kcalmol, 24.5 kcalmol with ZPE correction) is in good
accord with the experimental DG° for 8a!9 (25.3 kcal
molꢀ1), although it is somewhat higher than the activation
enthalpy (DH° =18.5 kcalmol). The introduction of methyl
or iPr substituents in the diphosphine does not alter signifi-
cantly the relative energy of the corresponding transition
states (TS’ and TS’’, respectively). Therefore, the origin of
the relatively high barrier to the tautomeric exchange does
not reside in the steric hindrance caused by the dippe
ligand, but in the distortion of the metallacyclic ring during
Figure 3. Schematic representation of the calculated geometries of simpli-
fied O- and C-cyclic enolates showing some selected geometric parame-
ters
for
Ni
and
Pd
(in
parenthesis):
a) [M(o-C6H6C(=
CH2)O)(H2PCH2CH2PH2)], M=Ni (O-Ni) and Pd (O-Pd); b) [M(o-
C6H6C(=O)CH2)(H2PCH2CH2PH2)], M=Ni (C-Ni) and Pd (C-Pd) (top
view); c) transition state (TS).
ꢀ
flat structure exhibited by the O-M models, the metallacycle
ring of the C-enolates shows a significant puckering. A simi-
lar conformation is found in the Pd C-enolate complex [Pd-
(CH2C(=O)-2-C6H-3,4,5-(OMe)3(tmed)], but the different
co-ligands present in this compound prevents a more precise
comparison with their structural features.
The relative energies of the model complexes are shown
in Table 1. As expected, the C-tautomeric form is strongly
favored for Pd. For Ni, the C-enolate is favored by only
about 2 kcalmolꢀ1. This difference further diminishes to
1.7 kcalmolꢀ1 if the zero-point energy (ZPE) correction is
taken into account. The introduction of carbon substituents
on the P atoms of the phosphine does not change signifi-
cantly the energy balance, although there seems to be a ten-
dency to favor the O-enolate form when electron-donor
the process. As can be seen in Figure 3, the aryl C(O) bond
has to bend noticeably (the angle formed by this bond and
the aromatic ring is close to 908), in order to maintain the
M enolate interaction. This restriction is absent in open-
ꢀ
chain enolates, where the interconversion is facile. It is also
worth to recall that the presence of bulky substituents on
the diphosphine ligand seems to have a small influence on
the equilibrium constant or the exchange rate between the
two tautomers.
Reactions of enolates and Brønsted acids: As represented in
Scheme 9 the reaction of 8a with HCl reverts enolate for-
mation and regenerates the starting chloroaryl compound
6a. The analogous reaction with [H(OEt2)2][BAr’4] (Ar’=
3,5-C6H3(CF3)2) also occurs with protonation of the enolate
Chem. Eur. J. 2005, 11, 6889 – 6904
ꢁ 2005 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chemeurj.org
6895

E. Carmona, J. Cµmpora et al.
Reactions with aldehydes: Despite the fact that enolizable
carbonylic compounds like MeC(O)H may undergo acid–
base reactions with metal enolates, giving rise to equilibrium
mixtures of different enolates, we considered this mode of
ꢀ
reactivity unlikely for 8a, as the Ni Caryl bond of the puta-
tive protonation product would keep the carbonylic func-
tionality in the vicinity of the metal coordination environ-
ment. In accord with expectations, compound 8a and
MeC(O)H react at room temperature with formation of the
condensation product 19a, resulting from nucleophilic
attackof the enolate carbon atom onto the carbonylic alde-
hyde carbon atom (Scheme 11). Most remarkably, 19a is an
Scheme 9.
functionality and yields the
cationic aryl compound 15 as
ꢀ
the corresponding BAr’4 salt,
15’. The palladium enolates 2
and 13 also react with HCl
with selective cleavage of the
Scheme 11.
ꢀ
Pd CH2 bond (see Scheme 7a
for the reaction of the dippe
derivative 13).
enolate related to 8a, formally derived from the addition of
ꢀ
one of the enolate C H bonds to the carbonyl group. Fur-
Reactions with carbon monoxide: Upon bubbling CO
through solution of 8a, reductive carbonylation to
[Ni(CO)2(dippe)] occurs (Scheme 10) with concomitant for-
ther reaction of 19a with an excess of MeC(O)H does not
take place, possibly due to steric hindrance. In agreement
with this assumption, the methyl-substituted enolates 8b
and 8c do not react with acetaldehyde. Although it is widely
a
ꢀ
accepted that aldehyde coordination precedes the C C
bond-forming step in aldol transformation induced by transi-
tion metals,[2a,4c,5a–5f] this mechanistic hypothesis appears
doubtful in the present case due to the rigid nature of the
enolate fragment. Instead, aldehyde attackby the enolate
carbon and a prototropic shift (rather than Ni2+ shift) seems
more likely.
The NMR spectra of 19a are very similar to those of the
parent enolate 8a. The =CH part of the new enolate ligand
Scheme 10.
gives 1H and 13C signals at d=4.73 and 97.3 ppm (1JCH
=
153 Hz), respectively, while the hydroxyl group is responsi-
ble for an IR absorption around 3200 cmꢀ1 and a proton res-
onance at d=5.77 ppm. The low thermal stability of this
product has precluded characterization by microanalysis.
However, in the presence of CO it is quantitatively convert-
ed into the lactone 20a (Scheme 11), that has been isolated
in a pure form.
Similarly to acetaldehyde, the non-enolizable aldehyde
PhCHO reacts with 8a giving rise to a new b-hydroxyeno-
late complex 19b, which is structurally similar to 19a. In
contrast with the latter, 19b is stable enough to allow its iso-
lation in analytically pure form. It can also be carbonylated
in situ, to afford the corresponding lactone in good isolated
yield (Scheme 12).
mation of lactone 18a. The related enolates 8b and 8c react
similarly producing the analogous lactones 18b and 18c.
The organic compounds can be separated from the Ni0 or-
ganometallic product by standard chromatographic tech-
niques. They have been characterized by high-resolution
mass spectrometry and by IR and NMR spectroscopy (see
Experimental Section). It is worth pointing out that the se-
lectivity of this reaction provides additional support for the
structure of the parent enolates. Thus, the carbonylation of
8b (R1 =Me, R2 =H) yields a single isomer, 18b, for which
NOESY experiments reveal the same Z configuration of the
C=C bond present in 8b. This selectivity contrasts with pre-
vious results on the generation of related lactones, obtained
as mixtures of their Z and E isomers,[20] and has been used
as structural probe for the nature of the products resulting
from the reactions of 8a with MeC(O)H and PhC(O)H, to
be discussed below.
Experimental conditions prove to be important for the
successful preparation of 19b. Thus, relatively concentrated
reaction mixtures (ca. 0.16m in each reactant) lead to full
conversion within approximately 4 h, but the reaction be-
6896
www.chemeurj.org
ꢁ 2005 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Chem. Eur. J. 2005, 11, 6889 – 6904

Metallacyclic Enolates
FULL PAPER
One final comment which is
worthy of note concerns the re-
activity of the enolate tauto-
mers 8a and 9 (Scheme 3a) to-
wards aldehydes. Only the O-
bound structure 8a exhibits the
enolate reactivity presented in
Scheme 11 and 12. Thus when
an approximate 2:1 mixture of
8a/9 is treated at room temper-
ature with either MeC(O)H or
PhC(O)H, 31P{1H} NMR moni-
toring reveals consumption of
8a, while 9 remains unaltered
(208C, 24 h). Hence under
these conditions only the O-
enolate is sufficiently nucleo-
philic to add to aldehydes.
Scheme 12.
comes too slow when more diluted solutions are used. In
practice, 19b is best obtained using a large excess of
PhCHO (1:10). Under these conditions, the product is clean-
ly formed within 5 min, but on standing for longer periods
of time, it slowly converts into a second product (21) the
transformation being complete after about 48 h. The NMR
spectra of 21 are strikingly similar to those of its precursor
(19b), both showing almost the same number of signals lo-
cated at very close positions. The main differences are found
1
in the H spectrum of the former, which lacks the hydroxyl
resonance and displays an apparently anomalous CH signal
with a relative intensity of 0.5H. These features suggest that
21 has the binuclear structure shown in Scheme 12. This pro-
posal was confirmed by the EI-MS spectrum of the corre-
sponding carbonylation product 22, which shows the expect-
ed molecular ion at m/z 380.
The formation of 21 from two molecules of 19b requires
the release of two molecules of water and one of PhCHO,
therefore an excess of the latter is not needed. Indeed, it
has been observed that pure samples of 19b dissolved in
THF evolve into 21, albeit much more slowly (ca. 10 days)
than observed in the reaction of 8a with an excess of benzal-
dehyde. The apparent catalytic effect of PhCHO is probably
due to traces of benzoic acid, which are difficult to remove.
In fact, the formation of 21 was retarded when the benzalde-
hyde was distilled immediately prior to use. However, the
presence of small amounts of benzoic acid can be consid-
ered unavoidable, since this could be formed from the well-
known autoxidation reaction of benzaldehyde (Cannizzaro
reaction), catalyzed by the basic Ni enolate complexes.
Scheme 13 shows a likely mechanism for the acid-cata-
lyzed formation of compound 21. Protonation of 19b, fol-
lowed by loss of water produces a very reactive enone, acti-
vated by coordination to the metal center. This intermediate
readily undergoes the addition of a second molecule of 19b.
The species formed is sterically crowded and readily elimi-
nates PhCHO (in the form of PhC(OH)H+), to give the
final product. Our observation that related nickel enolates
react with a,b-unsaturated ketones gives some support to
this mechanistic proposal.[8a]
Scheme 13.
Conclusion
Cyclic palladium enolates stabilized by either PMe3 or
iPr2PCH2CH2PiPr2 (dippe) auxiliary ligands can be prepared
following a conventional organometallic synthetic methodol-
ogy. Only the C-enolate isomers are observed. Isolation of
analogous nickel enolates has only been possible when the
chelating diphosphine dippe is employed. The failure to
obtain nickel enolates containing monodentate phosphine li-
gands could be due to the formation of binuclear oxygen-
bridged species by means of phosphine dissociation. This
would facilitate an intramolecular aldol condensation, ulti-
mately leading to compound 5.
While C-coordination is favored for the Pd enolate com-
plex, O-coordination is preferred by the nickel complexes.
Nonetheless for compound 8a, which features primary alkyl
coordination in the C-bound form (compound 9, see Sche-
me 3a), equilibrium mixtures of the two tautomers may be
generated. Still, the O-enolate predominates. DFT calcula-
tions suggest that the tautomeric exchange is hindered by
Chem. Eur. J. 2005, 11, 6889 – 6904
ꢁ 2005 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chemeurj.org
6897

E. Carmona, J. Cµmpora et al.
at ꢀ808C and a solution of PMe3 in toluene (1m, 2 mL, 2 mmol) was
added. The mixture was warmed to room temperature, stirred for 15 min,
and then cooled to ꢀ808C. 2’-Chloroacetophenone (0.13 mL, 1 mmol)
was then added, and the mixture was stirred at RT for 24 h and then
evaporated under reduced pressure. The residue was extracted with Et2O
(40 mL) and filtered. The solution was concentrated and cooled to
ꢀ208C furnishing compound 3 as orange crystals in 85% yield. 1H NMR
the incorporation of the enolate functionality into a rigid
metallacyclic fragment. The parent O-enolate complex 8a
reacts with enolizable and non-enolizable aldehydes, giving
products that retain the enolate functionality. Interestingly,
the C-enolate tautomer 9 does not react with aldehydes, il-
lustrating the relationship between coordination mode and
reactivity in this class of compounds.
*
(C6D6, 208C): d=0.76 (t, JHP =3.5 Hz, 18H; P(CH3)3), 2.38 (s, 3H; CH3),
6.71 (pseudo-t, J=7.5 Hz, 1H; CHar), 6.88 (pseudo-t, J=7.5 Hz, 1H;
CHar), 7.41 (d, 3JHH =7.6 Hz, 1H; CHar), 7.76 ppm (d, 3JHH =7.4 Hz, 1H;
CHar); 31P{1H} NMR (C6D6, 208C): d=ꢀ16.2 ppm (s); 13C{1H} NMR
(C6D6, 208C): d=12.7 (t, 1JCP =13 Hz; CH3), 26.3 (s, CH3), 120.8 (s,
CHar), 128.7 (s, CHar), 130.4 (s, CHar), 136.9 (t, 5JCP =4 Hz; CHar), 142.8
Experimental Section
(s, Car C=O), 171.8 (t, 2JCP =35 Hz; Car Ni), 199.1 ppm (s, C=O); IR
(Nujol): n˜ =1635 cmꢀl (C=O); elemental analysis calcd (%) for
C14H25ClP2ONi: C 46.0, H 6.9; found: C 45.6, H 7.3.
ꢀ
ꢀ
Microanalyses were performed by the Analytical Service of the Instituto
de Investigaciones Químicas. The spectroscopic instruments used were
Bruker Model Vector 22 for IR spectra, and Bruker DPX-300, DRX-400
and DRX-500 for NMR spectroscopy. The 13C resonance of the solvent
was used as an internal standard, but chemical shifts are reported with re-
spect to SiMe4. The 13C{lH} NMR assignments were helped in most cases
with the use of gate-decoupling techniques. 31P{lH} NMR shifts are refer-
enced to external 85% H3PO4. All preparations and other operations
were carried out under oxygen-free nitrogen by conventional Schlenck
techniques. Solvents were dried and degassed before use. The petroleum
ether used had a b.p. of 40–608C. The compound [Ni(h3-C3H5)Cl(PMe3)2]
was prepared by oxidative addition of allyl chloride to [Ni(cod)2] in the
presence of PMe3. The Grignard reagents Mg(R)Cl (R=Et, iPr) and the
sodium salts NaArO (Ar=2,4-dimethylphenoxide and 2,6-dimethylphen-
oxide) were prepared following the conventional methodologies. The li-
gands PMe3 and iPr2PCH2CH2iPr2 (dippe), the acid [H(OEt2)][BAr’4]
(BAr’4 =B[3,5-C6H3-(CF3)2]4) and its salt NaBAr’4,[21] and the complexes
The bromide derivative 4 was similarly prepared, although in this case
the oxidative addition of the 2’-bromoacetophenone to [Ni(cod)(PMe3)2]
*
requires only 3 h. Yield: 80%; 1H NMR (C6D6, 208C): d=0.81 (t, JHP
=
3.8 Hz, 18H; P(CH3)3), 2.35 (s, 3H; CH3), 6.70 (t, 3JHH =7.4 Hz, 1H;
3
3
CHar), 6.88 (t, JHH =7.2 Hz, 1H; CHar), 7.40 (d, JHH =7.7 Hz, 1H; CHar),
7.74 ppm (d,; 3JHH =7.5 Hz, 1H; CHar); 31P{1H} NMR (C6D6, 208C): d=
ꢀ16.4 ppm (s); 13C{1H} NMR (C6D6, 208C): d=13.6 (t, 1JCP =14 Hz; P-
(CH3)3), 26.4 (s, CH3), 121.0 (s, CHar), 129.0 (s, CHar), 130.7 (s, CHar),
136.5 (t, 5JCP =4 Hz; CHar), 142.8 (s, Car C=O), 173.7 (t, 2JCP =35 Hz;
ꢀ
ꢀl
ꢀ
Car Ni), 199.3 (s, C=O); IR (Nujol): n˜ =1635 cm (C=O); elemental
analysis calcd (%) for C14H25BrP2ONi: C 41.0, H 6.2, found: C 41.4, H
5.8.
Synthesis of compound 5: KtBuO (112 mg, 1 mmol) was added to a solu-
tion of 4 (410 mg, 1 mmol) in THF (50 mL) at ꢀ808C. After stirring the
mixture at room temperature for 4 h, the solvent was removed in
vacuum. The resulting yellow solid was extracted with CH2Cl2 (20 mL)
and the suspension was centrifuged to separate the KCl and partially
concentrated. After addition of some petroleum ether and cooling at
ꢀ208C, compound 5 was obtained as a yellow powder. Yield: 55%;
1H NMR (C6D6, 208C): d=ꢀ5.46 (s, 1H; OH), 0.53 (s, 3H; CH3), 0.93
(d, 2JHP =10.0 Hz, 9H; P(CH3)3), 1.27 (d, 2JHP =9.3 Hz, 9H; P(CH3)3),
2
[11]
[Ni(cod)2],[22]
[Pd(h -CH2=CH CO2Me)(PMe3)2]
and
[Pd(h3-
ꢀ
[23]
C3H5)Cl]2 were prepared according to literature methods. The nickel
compounds [Ni(CH2Ph)Cl(PMe3)2][24] and [Ni(C(Ph)=CH CO CH3)Cl-
ꢀ
ꢀ
(PMe3)2][12] were previously prepared in our laboratories.
Synthesis of [Pd(C6H4-o-C(O)CH3)Br(PMe3)2] (1): 2’-Bromoacetophe-
2
ꢀ
none (0.27 mL, 2 mmol) was added to a solution of [Pd(h -CH2=CH
CO2Me)(PMe3)2] (340 mg, 1 mmol) in Et2O (40 mL). The mixture was
stirred and heated to 508C for 14 h. The solution was evaporated under
vacuum and the residue extracted with CH2Cl2 (30 mL). After partial
2
2
3.00 (d, JHH =18.1 Hz, 1H; CH2), 4.85 (d, JHH =18.1 Hz, 1H; CH2), 6.37
(dm, 3JHH =7.5 Hz, 1H; CHar), 6.42 (d, 3JHH =7.6 Hz, 1H; CHar), 6.62
(tm, 3JHH =7.4 Hz, 1H; CHar), 6.79 (tm, 3JHH =7.4 Hz, 1H; CHar), 6.87
(tm, 3JHH =7.2 Hz, 1H; CHar), 6.92 (tm, 3JHH =7.3 Hz, 1H; CHar), 7.07
(dm, 3JHH =7.4 Hz, 1H; CHar), 7.67 ppm (d, 3JHH =7.6 Hz, 1H; CHar);
31P{1H} NMR (C6D6, 208C): d=ꢀ10.7 (s, PMe3), ꢀ5.9 ppm (s, PMe3);
13C{1H} NMR (C6D6, 208C): d=13.4 (d, 1JCP =30 Hz; P(CH3)3), 14.3 (d,
1JCP =28 Hz; P(CH3)3), 26.6 (s, CH3), 63.9 (s, CH2), 82.6 (s, C-O), 121.3 (s,
CHar), 123.0 (s, CHar), 123.5 (s, CHar), 124.4 (s, CHar), 125.7 (s, CHar),
126.7 (s, CHar), 137.3 (s, CHar), 138.4 (d, JCP =9 Hz; CHar), 141.7 (d,
evaporation of the solvent and cooling at ꢀ208C, complex 1 was isolated
*
as white crystals in 65% yield. 1H NMR (C6D6, 208C): d=0.92 (t, JHP
=
3.6 Hz, 18H; P(CH3)3), 2.41 (s, 3H; CH3), 6.84 (t, 3JHH =7.4 Hz, 1H;
3
3
CHar), 6.97 (t, JHH =7.3 Hz, 1H; CHar), 7.50 (d, JHH =7.7 Hz, 1H; CHar),
7.63 ppm (d, 3JHH =7.5 Hz, 1H; CHar); 31P{1H} NMR (C6D6, 208C): d=
ꢀ18.4 ppm (s); 13C{1H} NMR (C6D6, 208C): d=13.9 (t, 1JCP =15 Hz; P-
(CH3)3), 28.1 (s, CH3), 122.2 (s, CHar), 129.9 (s, CHar), 130.3 (s, CHar),
136.8 (m, CHar), 143.5 (s, Car C=O), 162.0 (t, 2JCP =6 Hz; Car Pd),
200.0 ppm (s, C=O); IR (Nujol): n˜ =1665 cmꢀ1 (C=O); elemental analysis
calcd (%) for C14H25BrP2OPd: C 36.7, H 5.5; found: C 36.4, H 5.6.
ꢀ
ꢀ
2JCP =33 Hz; Car Ni), 148.2 (d, 2JCP =45 Hz, Car Ni), 149.5 (s, Car C O),
ꢀ
ꢀ
ꢀ ꢀ
ꢀ
168.3 (s, Car C=O), 206.9 ppm (s, C=O); IR (Nujol): n˜ =3300 (OH),
1655 cmꢀl (C=O); elemental analysis calcd (%) for C22H32P2O3Ni: C 50.4,
Synthesis of [Pd(CH2C(O)-o-C6H4)(PMe3)2] (2): KtBuO (123 mg,
1.1 mmol) was added to a cold solution (ꢀ808C) of 1 (457 mg, 1 mmol)
in THF (40 mL). The mixture was stirred at room temperature for 1 h
and then taken to dryness. The residue was extracted with CH2Cl2
(25 mL) and the resulting solution was centrifuged. After concentration,
addition of some Et2O and cooling to ꢀ208C, enolate 2 was obtained as
H 6.2; found: C 50.6, H 6.0.
Synthesis of [Ni(C6H4-o-C(O)CH3)Br(dppe)]: The ligand dppe (398 mg,
1 mmol) was added to a suspension of [Ni(cod)2] (275 mg, 1 mmol) in tol-
uene (50 mL) at ꢀ788C. The mixture was allowed to warm to room tem-
perature and stirred for 15 min, cooled to ꢀ808C and 2’-bromoacetophe-
none (0.14 mL, 1 mmol) was added. The resulting solution was then stir-
red at RT for 1.5 h, during which time the initial yellow colour due to
[Ni(cod)(PMe3)2] turned red-brown. The solvent was evaporated under
vacuum and the residue extracted with of toluene (80 mL). After partial
concentration of the solution and cooling to ꢀ308C, the product was iso-
lated as an orange solid. Yield: 85%; 1H NMR (C6D6, 208C): d=2.10 (s,
3H; CH3), 2.14–2.34 (m, 4H; CHH), 6.73 (t, 3JHH =7.4 Hz, 1H; CHar),
6.93 (t, 3JHH =7.0 Hz, 1H; CHar), 7.14–7.64 (m, 20H; CHar), 7.76 (d,
3JHH =6.4 Hz, 1H; CHar), 8.02 ppm (m, 1H; CHar); 31P{1H} NMR (C6D6,
1
2
colourless crystals. Yield: 75%; H NMR (C6D6, 208C): d=0.75 (d, JHP
=
=
7.3 Hz 9H; P(CH3)3), 0.88 (d, 2JHP =7.3 Hz, 9H; CH3), 2.26 (t, JHH
2
7.6 Hz, 2H; CH2), 7.20 (m, 2H; CHar), 7.71 (m, 1H; CHar), 8.20 ppm (m,
1H; CHar); 31P{1H} NMR (C6D6, 208C): d=ꢀ23.1 (A, AX spin system),
ꢀ23.9 ppm (X, AX spin system;
J
AX =30 Hz); 13C{1H} NMR (C6D6,
208C): d=16.1 (d, JCP =22 Hz; P(CH3)3), 17.5 (d, JCP =21 Hz; P(CH3)3),
1
1
52.1 (dd, 2JCP =71, 9 Hz; CH2), 124.1 (s, CHar), 124.6 (d, JCP =4 Hz;
ꢀ
CHar), 126.7 (d, JCP =9 Hz; CHar), 136.4 (m, CHar), 149.0 (s, Car C=O),
164.1 (dd, 2JCP =125, 9 Hz; Car Pd), 202.9 ppm (t, 3JCP =8 Hz; C=O). IR
ꢀ
(Nujol): n˜ =1630 cmꢀl (C=O); elemental analysis calcd (%) for
208C): d=41.8 (A, AX spin system), 54.3 ppm (X, AX spin system; JAX
=
2
34 Hz); 13C{1H} NMR (C6D6, 208C): d=25.2 (dd, 1JCP =26 Hz, JCP
=
C14H24P2OPd: C 44.6, H 6.4; found: C 44.5, H 6.2.
Synthesis of [Ni(C6H4-o-C(O)CH3)X(PMe3)2] (X=Cl, 3; X=Br, 4): A
suspension of [Ni(cod)2] (275 mg, 1 mmol) in toluene (50 mL) was cooled
14 Hz; CH2), 26.2 (s, CH3), 30.6 (dd, 1JCP =27 Hz, 2JCP =23 Hz; CH2),
121.9 (s, CHar), 128.4–134.2 (m, CHar), 129.9 (d, JCP =5 Hz; CHar), 130.7
6898
www.chemeurj.org
ꢁ 2005 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Chem. Eur. J. 2005, 11, 6889 – 6904

Metallacyclic Enolates
FULL PAPER
(s, CHar), 139.2 (s, CHar), 143.8 (s, Car C=O), 174.7 (dd, 2JCP =85, 36 Hz;
ꢀ
[Ni(C6H4-o-C(O)CH(CH3)2)Cl(dippe)] (6c): 1H NMR (C6D6, 208C): d=
0.99 (dd, 3JHP =13.6 Hz, 3JHH =7.0 Hz, 12H; CH3), 1.28 (d, 3JHH =6.8 Hz,
6H; CH3), 1.34 (dd, 3JHP =12.5 Hz, 3JHH =7.1 Hz, 6H; CH3), 1.44 (dd,
ꢀ
Car Ni), 205.9 ppm (d, 4JCP =6 Hz, C=O); IR (Nujol): n˜ =1640 cmꢀl (C=
O); elemental analysis calcd (%) for C44H37BrP2ONi: C 67.6, H 4.8;
found: C 67.9, H 5.0.
3
3JHP =15.3 Hz, JHH =7.2 Hz, 6H; CH3), 1.49 (m, 2H; CH2), 1.78 (m, 2H;
CH2), 2.15 (m, 2H; CH), 2.43 (m, 2H; CH), 3.66 (sept, 3JHH =6.4 Hz,
Synthesis of [Ni(C6H4-o-C(O)CH3)Br(PPh3)2]: This complex was pre-
pared by a analogous route to that employed for the related dppe com-
plex, although it was extracted with CH2Cl2 and crystallized from a
CH2Cl2/Et2O mixture. Yield: 85%. 1H NMR (C6D6, 208C): d=1.94 (s,
3H; CH3), 6.41 (t, 3JHH =7.4 Hz, 1H; CHar), 6.52 (t, 3JHH =7.5 Hz, 1H;
CHar), 6.66 (d, 3JHH =6.9 Hz, 1H; CHar), 7.73 (m, 1H; CHar), 6.78–
7.70 ppm (m, 30H; CHar); 31P{1H} NMR (C6D6, 208C): d=25.9 ppm (s);
13C{1H} NMR (C6D6, 208C): d=25.9 (s, CH3), 121.5 (s, CHar), 125.6–132.4
3
3
1H; CH), 6.92 (tm, JHH =7.7 Hz, 1H; CHar), 7.08 (tm, JHH =7.3 Hz, 1H;
3
3
CHar), 7.58 (t, JHH ꢂJHP =6.3 Hz, 1H; CHar), 7.70 ppm (dm, JHH
7.8 Hz, 1H; CHar); 31P{1H} NMR (C6D6, 208C): d=65.6 (A, AX spin
system), 73.0 ppm (X, AX spin system,
AX =33 Hz); 13C{1H} NMR
=
J
(C6D6, 208C): d=17.3 (s, CH3), 17.3 (t, JCP =16 Hz; CH2), 17.9 (s, CH3),
19.3 (s, CH3), 19.8 (s, CH3), 23.6 (partially hidden dd, CH2), 23.9 (d,
1JCP =23 Hz; CH), 24.6 (d, 1JCP =22 Hz; CH), 34.7 (s, CH), 121.8 (s,
CHar), 129.7 (m, CHar), 130.0 (d, JCP =6 Hz; CHar), 138.6 (s, CHar), 144.6
ꢀ
(m, CHar), 129.1 (s, CHar), 132.7 (s, CHar), 136.9 (s, CHar), 143.5 (s, Car
C=O), 168.6 (m, Car Ni), 200.6 ppm (m, C=O);. IR (Nujol): n˜ =1640 cm
(C=O); elemental analysis calcd (%) for C34H31BrP2ONi: C 62.2, H 4.8;
ꢀl
(s, Car C=O), 175.1 (dd, 2JCP =85, 36 Hz; Car Ni), 209.8 ppm (m, C=O).
IR (Nujol): n˜ =1625 cmꢀl (C=O); elemental analysis calcd (%) for
C24H43ClP2ONi: C 57.2, H 8.6; found: C 57.1, H 8.7.
ꢀ
ꢀ
ꢀ
found: C 62.6, H 5.2.
Synthesis of [Ni(C6H4-o-C(O)CHR1R2)Cl(dippe)] (R1 =R2 =H, 6a; R1 =
H, R2 =Me, 6b; R1 =R2 =Me, 6c): The compounds 6b and 6c were pre-
pared from 2-chloropropiophenone and 2-chloroisobutirophenone. These
precursors were previously synthesized by treating 2-chlorobenzonitrile
with Mg(Br)Et or Mg(Br)iPr followed by acid hydrolysis with H2SO4. 2-
Chloropropiophenone: 79% yield; 1H NMR (C6D6, 208C): d=1.18 (t,
3JHH =7.3 Hz, 3H; CH3), 2.93 (q, 2H; CH2), 7.29, 7.35, 7.39, 7.42 ppm (m,
1H; CHarom). 2-Chloroisobutirophenone: 43% yield; 1H NMR (C6D6,
208C): d=1.03 (d, 3JHH =7.3 Hz, 3H; CH3), 3.03 (sept, 1H; CH); 6.98
(m, 1H; CHarom), 6.98 (m, 2H; CHarom), 7.14 ppm (m, 1H; CHarom).
ꢀ
Synthesis for the complexes [Ni{o-C6H4 C=C(HR)O}(dippe)] (R=H,
8a; R=Me, 8b): KOtBu (123 mg, 1.1 mmol) was added to a cooled
(ꢀ788C) solution of [Ni(C6H4-o-C(O)CH3)(Cl)(dippe)], 6a, (475 mg,
1 mmol) in anhydrous THF (80 mL) under N2. After stirring at room
temperature for 1 h, the solvent was removed in vacuo. The resulting
yellow solid was extracted with toluene (100 mL) and this solution was
centrifuged to remove the KCl. Recrystallization from THF provided
277 mg (65%) of the enolate 8a as yellow crystals. 1H NMR (C6D6,
208C): d=0.77 (dd, 3JHP =11.8 Hz, 3JHH =7.0 Hz, 6H; CH3), 0.94 (dd,
3JHP =13.0 Hz, JHH =7.1 Hz, 6H; CH3), 0.98 (m, 2H; CH2), 1.10 (m, 2H;
3
CH2), 1.13 (dd, 3JHP =17.4 Hz, 3JHH =7.3 Hz, 6H; CH3), 1.34 (dd, JHP
=
3
A suspension of [Ni(cod)2] (275 mg, lmmol) in toluene (40 mL) was
cooled at ꢀ808C, and treated with dippe (0.31 mL, 1 mmol). The mixture
was warmed to RT, stirred for 15 min and then 2’-chloroacetophenone
(0.13 mL, 1 mmol) was added. The mixture was stirred at 458C for 3 d,
after which time the solvent was removed in vacuo. The resulting orange
solid was extracted with CH2Cl2 (30 mL) and this solution was filtered.
15.4 Hz, 3JHH =7.2 Hz, 6H; CH3), 1.99 (m, 2H; CH2), 2.07 (m, 2H; CH),
4.62 (s, 1H; CHH), 4.79 (d, 2JHH =1.7 Hz, 1H; CHH), 7.14 (m, 2H;
3
CHar), 7.73 (pseudo-t, JHH =6.6 Hz, 1H; CHar), 7.74 ppm (m, 1H; CHar);
31P{1H} NMR (C6D6, 208C): d=72.3 (A, AX spin system), 78.1 ppm (X,
AX spin system, JAX =25 Hz); 13C{1H} NMR (C6D6, 208C): d=16.0 (dd,
The volume was reduced and cooled to ꢀ308C to provide complex 6a as
1JCP =20 Hz, 2JCP =10 Hz; CH2), 18.1 (d, 2JCP =5 Hz; CH3), 18.1 (s, CH3),
3
orange crystals. Yield: 75%; 1H NMR (C6D6, 208C): d=0.87 (dd, JHP
=
19.2 (d, 2JCP =4 Hz; CH3), 21.1 (d, 2JCP =6 Hz; CH3), 22.4 (dd, JCP
=
=
=
1
13.7 Hz, 3JHH =6.9 Hz, 12H; CH3), 1.21 (dd, 3JHP =12.4 Hz, 3JHH =7.1 Hz,
26 Hz, 2JCP =23 Hz; CH2), 24.0 (d, 1JCP =17 Hz; CH), 25.2 (d, JCP
1
3
3
6H; CH3), 1.33 (dd, JHP =15.2 Hz, JHH =7.1 Hz, 6H; CH3), 1.46 (m, 2H;
CH2), 1.64 (m, 2H; CH2), 2.01 (m, 2H; CH), 2.29 (m, 2H; CH), 2.53 (s,
3H; CH3), 6.80 (tm, 3JHH =7.4 Hz, 1H; CHar), 6.97 (tm, 3JHH =7.3 Hz,
21 Hz; CH), 75.9 (s, CH2), 122.9 (s, CHar), 123.4 (s, CHar), 124.9 (d, JCP
7 Hz; CHar), 137.4 (m, CHar), 156.2 (s, Car C O), 158.6 (dd, 2JCP =27,
ꢀ ꢀ
87 Hz; Car Ni), 176.1 (d, 3JCP =14 Hz; C O); IR (Nujol): n˜ =1590 cm
(C=C); elemental analysis calcd (%) for C22H38P2ONi: C 60.2, H 8.7;
found: C 60.0, H 8.3.
ꢀl
ꢀ
ꢀ
3
3
3
1H; CHar), 7.49 (t, JHH ꢂ JHP =6.3 Hz, 1H; CHar), 7.56 ppm (dm, JHH
7.7 Hz, 1H; CHar); 31P{1H} NMR (C6D6, 208C): d=65.9 (A, AX spin
system), 71.9 ppm (X, AX spin system;
AX =32 Hz); 13C{1H} NMR
=
J
1
2
Complex 8b can be prepared according to the same methodology (yield:
65%) or, alternatively, by the following procedure: a solution of LiMe
was added (0.63 mL of 1.6m solution in hexane, 1 mmol) to a solution of
complex 10 (199 mg, 0.5 mmol) in THF (40 mL) at ꢀ788C. The mixture
was stirred at RT during 20 min, after which time SiMe3Cl (0.63 mL) was
added to neutralize the LiMe excess. 2’-Chlorophenylethylcetone
(0.07 mL, 0.5 mmol) and NEt3 (0.70 mL) were then added at ꢀ808C and
the resulting mixture was stirred at room temperature for 12 h. The sol-
vent was removed in vacuo and the residue extracted with toluene
(100 mL) and the solution centrifuged and taken to dryness. The residue
was recrystallized from THF. Yield: 65%; 1H NMR (C6D6, 208C): d=
(C6D6, 208C): d=17.2 (s, CH3), 17.5 (dd, JCP =20 Hz, JCP =12 Hz; CH2),
1
18.1 (s, CH3), 19.4 (s, CH3), 23.4 (t, JCP =23 Hz; CH2), 24.0 (d, JCP
=
17 Hz; CH), 24.5 (d, 1JCP =22 Hz; CH), 27.2 (s, CH3), 121.9 (s, CHar),
ꢀ
130.0 (d, JCP =6 Hz; CHar), 130.7 (s, CHar), 138.3 (s, CHar), 146.1 (s, C C=
O), 173.7 (dd, 2JCP =82, 37 Hz; Car Ni), 203.2 ppm (d, 4JCP =4 Hz; C=O);
ꢀ
IR (Nujol): n˜ =1630 cmꢀl (C=O); elemental analysis calcd (%) for
C22H39ClP2ONi: C 55.6, H 8.3; found: C 55.4, H 8.3.
Compounds 6b and 6c were prepared according to the same procedure,
although in both cases the oxidative addition only requires 24 h. Both
products are crystallized from THF in 80% (6b) and 65% (6c) yields.
[Ni(C6H4-o-C(O)CH2CH3)Cl(dippe)] (6b): 1H NMR (C6D6, 208C): d=
0.98 (dd, 3JHP =13.7 Hz, 3JHH =7.0 Hz, 12H; CH3), 1.29 (t,; 3JHH =7.3 Hz,
3H CH3), 1.35 (dd, 3JHP =12.5 Hz, 3JHH =7.1 Hz, 6H; CH3), 1.46 (dd,
3
3
0.75 (dd, JHP =11.8 Hz, JHH =7.0 Hz, 6H; CH3), 0.80 (m, 2H; CH2), 0.94
(dd, JHP =12.9 Hz, JHH =7.1 Hz, 6H; CH3), 1.05 (m, 2H; CH2), 1.14 (dd,
3JHP =17.1 Hz, 3JHH =7.3 Hz, 6H; CH3), 1.38 (dd, 3JHP =15.4 Hz, JHH
3
3
3
3
=
3JHP =15.3 Hz, JHH =7.2 Hz, 6H; CH3), 1.50 (m, 2H; CH2), 1.78 (m, 2H;
3
CH2), 2.15 (m, 2H; CH), 2.43 (m, 2H; CH), 3.09 (quart, 3JHH =7.3 Hz,
2H; CH2), 6.91 (tm, 3JHH =7.3 Hz, 1H; CHar), 7.08 (tm, 3JHH =7.3 Hz,
7.2 Hz, 6H; CH3), 2.00 (m, 2H; CH), 2.10 (m, 2H; CH), 2.36 (d, JHH
=
6.7 Hz, 3H; CH3), 5.27 (quart, 3JHH =6.7 Hz, 1H; =CH), 7.17 (m, 2H;
CHar), 7.37 (m, 1H; CHar), 7.73 ppm (m, 1H; CHar); 31P{1H} NMR (C6D6,
3
3
1H; CHar), 7.58 (t, JHH ꢂJHP =6.4 Hz, 1H; CHar), 7.68 ppm (dm, JHH
=
7.7 Hz, 1H; CHar); 31P{1H} NMR (C6D6, 208C): d=72.8 (A, AX spin
208C): d=65.4 (A, AX spin system), 72.5 ppm (X, AX spin system, JAX
32 Hz); 13C{1H} NMR (C6D6, 208C): d=11.4 (s, CH3), 16.2 (dd, JCP
20Hz, 2JCP =11 Hz; CH2), 18.1 (d, 2JCP =4 Hz; CH3), 18.2 (s, CH3), 19.2
(d, 2JCP =4 Hz; CH3), 21.1 (d, 2JCP =6 Hz; CH3), 22.4 (t, 1JCP =24.2 Hz;
CH2), 24.0 (d, 1JCP =17 Hz; CH), 25.2 (d, 1JCP =21 Hz; CH), 86.1 (s, =
CH), 121.3 (s, CHar), 123.3 (s, CHar), 123.9 (d, JCP =7 Hz; CHar), 137.5
=
1
J
AX =21 Hz); 13C{1H} NMR
=
system), 77.4 ppm (X, AX spin system;
(C6D6, 208C): d=8.7 (s, CH3), 17.2 (s, CH3), 17.4 (dd, 1JCP =20 Hz, JCP
=
2
12 Hz; CH2), 17.9 (s, CH3), 19.3 (s, CH3), 19.4 (s, CH3), 23.6 (t, JCP
=
23 Hz; CH2), 24.0 (d, 1JCP =18 Hz; CH), 24.6 (d, 1JCP =25 Hz; CH), 31.8
(s, CH2), 121.9 (s, CHar), 129.5 (m, CHar), 130.0 (d, JCP =6 Hz; CHar),
ꢀ
ꢀ
(m, CHar), 156.5 (s, Car C O), 158.0 (dd, 2JCP =27, 88 Hz; Car Ni),
169.7 ppm (d, 3JCP =13 Hz, C O); IR (Nujol): n˜ =1610 cm (C=C); ele-
mental analysis calcd (%) for C23H40P2ONi: C 61.0, H 8.9; found: C 60.8,
H 8.9.
138.4 (s, CHar), 145.7 (s, Car C=O), 173.5 (dd, 2JCP =85, 37 Hz; C Ni),
206.2 (d, 4JCP =4 Hz; C=O); IR (Nujol): n˜ =1645 cmꢀl (C=O); elemental
analysis calcd (%) for C23H41CIP2ONi: C 56.4, H 8.4; found: C 56.4, H
8.4.
ꢀ ꢀ
ꢀ
ꢀl
ꢀ
Chem. Eur. J. 2005, 11, 6889 – 6904
ꢁ 2005 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chemeurj.org
6899

E. Carmona, J. Cµmpora et al.
ꢀ
Synthesis of [Ni{o-C6H4 C=C(CH3)2O}(dippe)] (8c): KtBuO (123 mg,
4.34 (t, 2JHH =1.5 Hz, 1H; CHH), 4.58 (d, 2JHH =1.1 Hz, 1H; CHH), 6.77
(dd, 4JHP =8.3, 6.1 Hz, 1H; =CH), 7.02 (t, 3JHH =7.3 Hz, 1H; CHar), 7.13
(t, 3JHH =7.3 Hz, 2H; CHar), 7.33 ppm (d, 3JHH =7.1 Hz, 2H; CHar);
31P{1H} NMR (C6D6, 208C): d=73.5 (A, AX spin system), 76.1 ppm (X,
AX spin system, JAX =29 Hz); 13C{1H} NMR (C6D6, 208C): d=15.4 (dd,
1JCP =20 Hz, 2JCP =11 Hz; CH2), 17.6 (d, 2JCP =6 Hz; CH3), 18.3 (s, CH3),
1.1 mmol) was added to a solution of complex 6c (504 mg, 1 mmol) in
THF (80 mL) at ꢀ788C. After stirring the mixture at RT for 24 h, the sol-
vent was removed under vacuum. The yellow residue was extracted with
toluene (100 mL) and the resulting solution was centrifuged to separate
the KCl. The solution was evaporated and the residue recrystallized from
THF providing the enolate 8c as yellow crystals in 65% yield. 1H NMR
(C6D6, 208C): d=0.76 (dd, 3JHP =11.8 Hz, 3JHH =7.0 Hz, 6H; CH3), 0.80
(m, 2H; CH2), 0.92 (dd, 3JHP =12.7 Hz, 3JHH =7.1 Hz, 6H; CH3), 1.09 (m,
2H; CH2), 1.14 (dd, 3JHP =17.0 Hz, 3JHH =7.3 Hz, 6H; CH3), 1.39 (dd,
3JHP =15.6 Hz, 3JHH =7.2 Hz, 6H; CH3), 1.98 (m, 2H; CH), 2.11 (m, 2H;
CH), 2.40 (s, 3H; CH3), 2.42 (s, 3H; CH3), 7.14 (m, 1H; CHar), 7.23 (t,
3JHH =7.5 Hz, 1H; CHar), 7.43 (t, 3JHH =7.0 Hz, 1H; CHar), 7.93 ppm (d,
3JHH =7.9 Hz, 1H; CHar); 31P{1H} NMR (C6D6, 208C): d=72.4 (A, AX
spin system), 72.7 ppm (X, AX spin system, JAX =19 Hz); 13C{1H} NMR
(CD2Cl2, 208C): d=16.7 (dd, 1JCP =20 Hz, 2JCP =10 Hz; CH2), 18.8 (d,
2JCP =5 Hz, 4C; CH3), 19.8 (d, 2JCP =5 Hz; CH3), 20.7 (s, CH3), 21.1 (s,
CH3), 21.8 (d, 2JCP =6 Hz; CH3), 23.0 (dd, 1JCP =26 Hz, 2JCP =22 Hz, 1C;
CH2), 24.5 (d, 1JCP =17 Hz; CH), 25.9 (d, 1JCP =21 Hz; CH), 97.7 (s, =C),
2
1
19.4 (d, JCP =4 Hz; CH3), 20.1 (t, JCP =23 Hz; CH2), 23.9 (d, JCP =22 Hz;
CH), 24.2 (d, 1JCP =17 Hz; CH), 80.4 (s, CH2), 122.8 (s, Car), 123.7 (s,
3
CHar), 126.4 (d, JCP =2 Hz; CHar), 127.1 (s, CHar), 146.6 (d, JCP =4 Hz; =
CH), 166.4 (dd, 2JCP =83, 26 Hz; =C Ni), 178.7 ppm (dd, 3JCP =16, 2 Hz;
ꢀ
ꢀl
ꢀ
C O); IR (Nujol): n˜ =1600 cm (C=C); elemental analysis calcd (%) for
C26H40P2ONi: C 62.0, H 8.7; found: C 61.3, H 8.6.
Synthesis of [Pd(CH2C(O)-o-C6H4)(dippe)] (13): The ligand dippe
(0.31 mL, 1 mmol) was added to a cooled (ꢀ808C) solution of complex 2
(376 mg, 1 mmol) in THF (30 mL), and the mixture warmed slowly to
room temperature. The solvent was then evaporated and the residue ex-
tracted with toluene (30 mL) and filtered. After concentration and the
addition of some Et2O complex 13 was obtained as colourless crystals.
Yield: 90%; 1H NMR (C6D6, 208C): d=0.69 (m, 12H; CH3), 0.91 (m,
12H; CH3), 1.08 (m, 2H; CH2), 1.18 (m, 2H; CH2), 1.77 (m, 2H; CH),
1.92 (m, 2H; CH), 3.44 (dd, 3JHP =7.3, 6.2 Hz, 2H; CH2), 7.26 (m, 2H;
CHar), 7.88 (m, 1H; CHar), 8.32 ppm (m, 1H; CHar); 31P{1H} NMR (C6D6,
122.6 (s, CHar), 123.1 (d, JCP =6 Hz; CHar), 124.0 (s, CHar), 138.2 (m,
2
ꢀ ꢀ
ꢀ
CHar), 155.0 (s, Car C O), 160.5 (dd, JCP =87, 25 Hz; Car Ni), 162.9 ppm
(d, 3JCP =14 Hz; C O); IR (Nujol): n˜ =1595 cm (C=C); elemental anal-
ꢀl
ꢀ
ysis calcd (%) for C24H42P2ONi: C 61.7, H 9.1; found: C 62.0, H 8.9.
208C): d=68.4 (A, AX spin system), 72.9 ppm (X, AX spin system, JAX
=
=
=
2
Synthesis of [Ni(h3-C3H5)Cl(dippe)] (10): A solution of the complex [Ni-
(h3-C3H5)Cl(PMe3)2] (278 mg, 1 mmol) in Et2O (40 mL) was cooled to
ꢀ808C and treated with dippe (0.31 mL, 1 mmol). After the solution was
warmed to room temperature, it was pumped to dryness and the remain-
ing solid was dissolved in Et2O (50 mL) and filtered. Its concentration
and cooling to ꢀ208C produced orange crystals in quantitative yield.
15 Hz); 13C{1H} NMR (C6D6, 208C): d=17.9 (s, CH3), 19.1 (d, JCP
2
6 Hz; CH3), 19.8 (d, 2JCP =8 Hz; CH3), 20.0 (dd, 1JCP =39 Hz, JCP
18 Hz; CH2), 24.8 (d, 1JCP =15 Hz; CH), 25.0 (d, 1JCP =12 Hz; CH), 47.0
(dd, 2JCP =66,8 Hz; CH2), 124.1 (s, CHar), 124.9 (d, JCP =4 Hz; CHar),
ꢀ
126.5 (d, JCP =8 Hz; CHar), 138.6 (dd, JCP =10, 5 Hz; CHar), 149.2 (s, Car
C=O), 163.3 (dd, 2JCP =21, 8 Hz; Car Pd), 201.7 ppm (t, 3JCP =7 Hz; C=
ꢀ
lH NMR (CD2Cl2, 208C): d=1.14 (brs, 14H; CH3, CH2), 1.26 (brs, 12H;
O); IR (Nujol): n˜ =1625 cmꢀl (C=O); elemental analysis calcd (%) for
2
CH3, CH2), 2.05 (brs, 2H; CH), 2.37 (brs, 1H; CH), 2.70 (d, JHH
=
C22H38P2OPd: C 54.3, H 7.9; found: C 54.1, H 7.9.
8.2 Hz, 2H; CHH), 4.37 (brs, 2H; CHH), 5.05 ppm (q, 3JHH =10.8 Hz,
1H; CH); 31P{1H} NMR (C6D6, 208C): d=81.6 ppm (s); 13C{1H} NMR
(CD2Cl2, 208C): d=18.9 (s, CH3), 19.6 (m, CH3), 21.9 (m, CH2), 26.5 (d,
Synthesis of [Pd(C6H4-o-C(O)CH3)Cl(dippe)] (14): A solution of com-
plex 13 (487 mg, 1 mmol) in THF (30 mL), cooled to ꢀ808C, was treated
with HCl (10 mL of a 0.1m solution in THF, 1 mmol). The mixture was
stirred at RT for 15 min, and the solvent evaporated. The resulting white
solid was extracted with CH2Cl2 and the solution centrifuged. Concentra-
tion of this solution and cooling to ꢀ208C furnished white crystals in
quantitative yield. 1H NMR (CD2Cl2, 208C): d=1.03 (dd, 3JHP =14.8 Hz,
3JHH =7.0 Hz, 12H; CH3), 1.20 (dd, 3JHP =13.4 Hz, 3JHH =7.1 Hz, 6H;
CH3), 1.34 (dd, 3JHP =16.1 Hz, 3JHH =7.2 Hz, 6H; CH3), 1.59 (m, 2H;
CH2), 1.87 (m, 2H; CH2), 2.32 (m, 2H; CH), 2.53 (m, 2H; CH), 2.53 (s,
3H; CH3), 6.94 (tm, 3JHH =7.4 Hz, 1H; CHar), 7.07 (tm, 3JHH =7.3 Hz,
1JCP =25 Hz; CH), 63.4 (d, JCP =13 Hz; CH2), 113.5 ppm (s, CH).
2
ꢀ
ꢀ
Synthesis of [Ni(C(Ph)=CH CO Me)Cl(dippe)] (11): A solution of the
ꢀ
ꢀ
complex [Ni(C(Ph)=CH CO Me)Cl(PMe3)2], (392 mg, 1 mmol) in Et2O
(50 mL), cooled to ꢀ808C, was treated with dippe (0.31 mL, 1 mmol).
After the solution was warmed to room temperature, it was pumped to
dryness and the remaining solid was dissolved in Et2O (50 mL). The sol-
vent was again evaporated to remove the PMe3. The extraction with tolu-
ene (60 mL), filtration, partial concentration and addition of some Et2O,
furnished the desired complex as an orange solid. Yield: 70%; 1H NMR
(C6D6, 208C): d=0.41 (dd, 3JHP =11.3 Hz, 3JHH =7.0 Hz, 3H; CH3), 0.82
(m, 6H; CH3), 0.86 (m, 2H; CH2), 0.95 (dd, 3JHP =16.6 Hz, 3JHH =7.2 Hz,
3
3
1H; CHar), 7.47 (t, JHH ꢂJHP =7.6 Hz, 1H; CHar), 7.63 ppm (dd, JHH
=
7.6 Hz, JHP =1.6 Hz, 1H; CHar); 31P{1H} NMR (CD2Cl2, 208C): d=70.3
(A, AX spin system), 73.5 ppm (X, AX spin system, JAX =15 Hz); 13C{1H}
3
3
1
3H; CH3), 1.09 (dd, JHP =15.8 Hz, JHH =7.0 Hz, 3H; CH3), 1.11 (m, 2H;
CH2), 1.49 (m, 8H; 2CH3, 2CH), 1.67 (dd, 3JHP =14.2 Hz, 3JHH =7.1 Hz,
3H; CH3), 2.00 (m, 1H; CH), 2.15 (s, 3H; CH3), 2.78 (m, 1H; CH), 6.52
NMR (CD2Cl2, 208C): d=16.7 (s, CH3), 18.0 (s, CH3), 18.3 (dd, JCP
=
20 Hz, 2JCP =10 Hz; CH2), 19.1 (d, 2JCP =4 Hz; CH3), 23.6 (dd, JCP
=
1
26 Hz, 2JCP =22 Hz; CH2), 24.2 (d, 1JCP =18 Hz; CH), 28.9 (s, CH3), 122.4
(s, CHar), 129.7 (d, JCP =8 Hz; CHar), 130.2(d, JCP=7 Hz; CHar), 137.0 (d,
(dd, 4JHP =14.8, 4.3 Hz, 1H; =CH), 7.11 (t, JHH =7.3 Hz, 1H; CHar), 7.20
3
(t, 3JHH =7.6 Hz, 2H; CHar), 8.17 (d, 3JHH =8.0 Hz, 2H; CHar); 31P{1H}
NMR (C6D6, 208C): d=66.4 (A, AX spin system), 73.2 ppm (X, AX spin
J
CP =3 Hz; CHar), 146.3 (s, Car C=O), 167.4 (d, 2JCP =135 Hz; Car Pd),
ꢀ
ꢀ
202.2 ppm (d, JCP =3 Hz; C=O); IR (Nujol): n˜ =1675 cmꢀl (C=O); no an-
4
system, JAX =33 Hz); 13C{1H} NMR (C6D6, 208C): d=16.0 (d, JCP =7 Hz;
alytical data available.
2
CH3), 16.8 (d, 2JCP =6 Hz; CH3), 17.1 (dd, 1JCP =19 Hz, 2JCP =12 Hz;
Synthesis of [M{C6H4-o-C(O)CH3}(dippe)]+[BAr4]ꢀ (M=Ni, Ar=Ph, 15;
Ar=3,5-C6H3(CF3)2, 15’; M=Pd, Ar=Ph, 16): These compounds were
prepared following the same procedure. Compound 15: NaBPh4 (171 mg,
0.5 mmol) was added to a solution of complex 6a (238 mg, 0.5 mmol) in
THF (50 mL) cooled to ꢀ808C. The mixture was stirred at room temper-
ature for 1 h after which time the solvent was evaporated to dryness. The
residue was dissolved in CH2Cl2 and centrifuged. After removal of the
solvent in vacuo, the product was crystallized from a mixture of acetone/
Et2O as yellow crystals in 90% yield. 1H NMR (CD2Cl2, 208C): d=1.38
(m, 12H; CH3), 1.49 (m, 12H; CH3), 1.62 (m, 2H; CH2), 1.90 (m, 2H;
CH2), 2.40 (m, 2H; CH), 2.56 (m, 2H; CH), 2.65 (s, 3H; CH3), 7.30 (tm,
3JHH =7.5 Hz, 1H; CHar), 7.38 (tm, 3JHH =6.4 Hz, 1H; CHar), 7.51 (t,
CH2), 17.6 (s, CH3), 18.0 (d, JCP =3 Hz; CH3), 18.5 (d, 2JCP =3 Hz; CH3),
2
18.6 (d, 2JCP =3 Hz; CH3), 20.1 (d, 2JCP =4 Hz; CH3), 20.2 (d, 2JCP =4 Hz;
1
1
CH3), 22.2 (d, JCP =22 Hz; CH), 22.5 (t, JCP =24 Hz; CH2), 23.1 (d, JCP
=
23 Hz; CH), 25.6 (d, 1JCP =17 Hz; CH), 27.3 (d, 1JCP =23 Hz; CH), 29.7
(s, CH3), 126.6 (s, CHar), 127.5 (s, CHar), 134.1 (s, =CH), 149.7 (s, Car),
199.4 (d, 4JCP =5 Hz; C=O), 209.5 (dd, 2JCP =84, 31 Hz; =C Ni); IR
ꢀ
(Nujol): n˜ =1635 (C=O), 1505 cmꢀl (C=C); elemental analysis calcd (%)
for C24H41ClP2ONi: C 57.5, H 8.2; found: C 57.3, H 8.1.
ꢀ
Synthesis of [Ni{CH(Ph)=CH C(=CH2)O}(dippe)] (12): Following the
same methodology than for 8c, the enolate 12 was prepared from com-
plex 11, although in this case, only one hour was required for the depro-
tonation. It was isolated as brown crystals after recrystallization from
3
3JHH ꢂ JHP =7.5 Hz, 1H; CHar), 7.66 ppm (dm, 3JHH =7.7 Hz, 1H; CHar);
l
Et2O. Yield: 35%; H NMR (C6D6, 208C): d=0.57 (m, 6H; CH3), 0.68
31P{1H} NMR (CD2Cl2, 208C): d=75.4 (A, AX spin system), 84.4 ppm
(X, AX spin system, JAX =30 Hz); 13C{1H} NMR (CD2Cl2, 208C): d=16.6
(dd, 1JCP =26 Hz, 2JCP =7 Hz; CH2), 18.8 (d, 2JCP =5 Hz; CH3), 19.0 (s,
(m, 2H; CH2), 0.96 (dd, 3JHP =13.0 Hz, 3JHH =7.1 Hz, 6H; CH3), 0.96 (m,
2H; CH), 0.97 (m, 2H; CH2), 1.33 (m, 12H; CH3), 2.07 (m, 2H; CH),
6900
www.chemeurj.org
ꢁ 2005 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Chem. Eur. J. 2005, 11, 6889 – 6904

Metallacyclic Enolates
FULL PAPER
CH3), 19.8 (d, 2JCP =3 Hz; CH3), 22.1 (d, 2JCP =4 Hz; CH3), 22.3 (dd,
128.1 (s, CHar), 130.2 (s, CHar), 130.6 (s, CHar), 137.7 (s, CHar), 143.5 (s,
1
1JCP =29 Hz, 2JCP =18 Hz; CH2), 24.5 (s, CH3), 24.8 (d, JCP =19 Hz; CH),
Car CH3), 165.2 (s, Car O), 170.9 (m, Car Ni), 199.2 ppm (s, C=O); IR
(Nujol): n˜ =1655 cmꢀl (C=O); elemental analysis calcd (%) for
C22H34P2O2Ni: C 58.6, H 7.6; found: C 58.2, H 7.5.
ꢀ
ꢀ
ꢀ
26.7 (d, 1JCP =24 Hz; CH), 126.6 (s, CHar), 131.9 (s, CHar), 136.7 (d, JCP
=
ꢀ
6 Hz; CHar), 138.5 (d, JCP =3 Hz, CHar), 148.4 (s, Car C=O), 165.9 (dd,
2JCP =77, 28 Hz; Car Ni), 222.1 ppm (t, 3JCP =10 Hz; C=O); IR (Nujol):
Synthesis of O=C[C6H4-o-C(=CR1R2)O] (R1 =R2 =H, 18a; R1 =H, R2 =
Me, 18b; R1 =R2 =Me, 18c): CO was bubbled for 5 min at room temper-
ature through a solution of complex 8a (100 mg, 0.1 mmol) in THF
(15 mL). After this time the initial orange solution turned pale yellow.
The solvent was evaporated under vacuum and the oily residue was ex-
tracted with petroleum ether (30 mL) and filtered. Compound 18a was
separated from the complex [Ni(dippe)(CO)2] by spinning band chroma-
tography, with petroleum ether as eluent. Yield: 60%; 1H NMR (C6D6,
ꢀ
n˜ =1575 cmꢀ1 (C=O); elemental analysis calcd (%) for C46H59BP2ONi: C
72.8, H 7.8; found: C 72.4, H 7.9.
The same procedure was employed for the synthesis of 15’ and 16, but
using NaBAr’4, for the former. Compound 15’ was crystallized from Et2O
and 16 from CH2Cl2/Et2O, both in approximately 80–85% yield.
[Ni(C6H4-o-C(O)CH3)(dippe)]+[BAr’4]ꢀ (15’): 1H NMR (CD2Cl2, 208C):
3
d=1.23 (dd, 3JHP =13.3 Hz, 3JHH =7.0 Hz, 6H; CH3), 1.23 (dd, JHP
=
2
2
14.2 Hz, 3JHH =7.1 Hz, 6H; CH3), 1.32 (dd, 3JHP =16.4 Hz, 3JHH =7.3 Hz,
208C): d=4.61 (d, JHH =2.8 Hz, 1H; =CHH), 4.82 (d, JHH =2.8 Hz, 1H;
=CHH), 6.80 (t, 3JHH =6.7 Hz, 1H; CHar), 6.91 (m, 2H; CHar), 7.48 ppm
(d, 3JHH =8.2 Hz, 1H; CHar); 13C{1H} NMR (C6D6, 208C): d=90.0 (s,
3
3
6H; CH3), 1.37 (dd, JHP =18.7 Hz, JHH =7.3 Hz, 6H; CH3), 1.67 (m, 2H;
CH2), 1.94 (m, 2H; CH2), 2.28 (m, 2H; CH), 2.46 (m, 2H; CH), 2.55 (s,
3H; CH3), 7.10 (tm, 3JHH =7.4 Hz, 1H; CHar), 7.26 (tm, 3JHH =6.4 Hz,
ꢀ
CH2), 120.1 (s, CHar), 124.8 (s, CHar), 125.2 (s, Car C=O), 129.9 (s, CHar),
ꢀ ꢀ
ꢀ
3
3
3
133.6 (s, CHar), 138.7 (s, Car C O), 151.9 (s, C O), 165.9 ppm (s, C=O);
IR (Nujol): n˜ =1770 (C=O), 1660 cmꢀl (C=C); HREIMS: m/z calcd for
C9H6O2: 146.0368; found: 146.0369.
1H; CHar), 7.34 (t, JHH ꢂ JHP =7.5 Hz, 1H; CHar), 7.53 ppm (dm, JHH
=
7.6 Hz, 1H; CHar); 31P{1H} NMR (CD2Cl2, 208C): d=75.1 (A, AX spin
system), 84.4 ppm (X, AX spin system,
J
AX =30 Hz); 13C{1H} NMR
(CD2Cl2, 208C): d=16.2 (dd, 1JCP =26 Hz, 2JCP =7 Hz; CH2), 18.3 (d,
2JCP =5 Hz; CH3), 18.5 (s, CH3), 19.5 (d, 2JCP =3 Hz; CH3), 21.8 (dd,
1JCP =29 Hz, 2JCP =18 Hz; CH2), 21.9 (d, 2JCP =4 Hz; CH3), 23.9 (s, CH3),
24.6 (d, 1JCP =19 Hz; CH), 26.6 (d, 1JCP =24 Hz; CH), 126.5 (s, CHar),
Compounds 18b and 18c were prepared in the same manner. The former
was purified by spinning band chromatography using a 4:1 mixture of pe-
troleum ether/Et2O as eluent, while the latter was crystallized from pe-
troleum ether.
Compound 18b: 1H NMR (C6D6, 208C): d=1.67 (d, 4JHH =7.2 Hz, 3H;
ꢀ
131.6 (s, CHar), 136.6 (d, JCP =6 Hz; CHar), 138.4 (s, CHar), 148.4 (s, Car
C=O), 165.7 (dd, 2JCP =77, 28 Hz; Car Ni), 222.5 (d, 3JCP =9 Hz; C=O);
CH3), 4.99 (quart, 4JHH =7.2 Hz, 1H; =CH), 6.83 (t, 3JHH =7.4 Hz, 1H;
ꢀ
IR (Nujol): n˜ =1580 cmꢀl (C=O); elemental analysis calcd (%) for
3
CHar), 6.98 (m, 2H; CHar), 7.56 ppm (d, JHH =7.7 Hz, 1H; CHar); 13C{1H}
C54H51BF24P2ONi: C 49.8, H 3.9; found: C 49.3, H 4.4.
NMR (C6D6, 208C): d=10.8 (s, CH3), 102.9 (s, =CH), 119.1 ( s, CHar),
[Pd(C6H4-o-C(O)CH3)(dippe)]+[BPh4]ꢀ (16): 1H NMR (CD2Cl2, 208C):
d=1.24 (m, 24H; CH3), 1.63 (m, 2H; CH2), 1.96 (m, 2H; CH2), 2.31 (m,
2H; CH), 2.47 (m, 2H; CH), 2.65 (s, 3H; CH3), 7.22 (t, 3JHH =8.1 Hz,
1H; CHar), 7.39 (m, 2H; CHar), 7.71 ppm (m, 1H; CHar); 31P{1H} NMR
ꢀ
124.6 (s, Car C=O), 124.8 (s, CHar), 128.8 (s, CHar), 133.4 (s, CHar), 139.3
ꢀ ꢀ
ꢀ
(s, Car C O), 146.4 (s, C O), 166.0 ppm (s, C=O); IR (Nujol): n˜ =1770
ꢀl
ꢀ
(O C=O) 1685 cm (C=C); HR EIMS: m/z calcd for C10H8O2: 160.0524;
found: 160.0524.
Compound 18c: 1H NMR (C6D6, 208C): d=1.54 (s, 3H; CH3), 1.77 (s,
3H; CH3), 6.83 (t, 3JHH =7.5 Hz, 1H; CHar), 7.02 (t, 3JHH =7.6 Hz, 1H;
CHar), 7.15 (d, 3JHH =7.5 Hz, 1H; CHar), 7.69 ppm (d, 3JHH =7.7 Hz, 1H;
CHar); 13C{1H} NMR (C6D6, 208C): d=17.9 (s, CH3), 19.6 (s, CH3), 117.7
(s, =C-(CH3)2), 122.2 (s, CHar), 125.2 (s, CHar), 127.8 (s, CHar), 128.2 (s,
(CD2Cl2, 208C): d=77.6 (A, AX spin system), 87.7 ppm (X, AX spin
1
system, JAX =20 Hz); 13C{1H} NMR (CD2Cl2, 208C): d=16.6 (dd, JCP
=
2
2
24 Hz, JCP =6 Hz; CH2), 18.1 (d, JCP =3 Hz; CH3), 18.5 (s, CH3), 19.0 (d,
2JCP =4 Hz; CH3), 20.9 (d, JCP =3 Hz; CH3), 23.8 (dd, JCP =31 Hz, JCP
=
2
1
2
19 Hz; CH2), 24.9 (d, 1JCP =19 Hz; CH), 25.4 (d, JCP =3 Hz; CH3), 26.1
(d, 1JCP =27 Hz; CH), 125.9 (s, CHar), 132.9 (s, CHar), 133.0 (s, CHar),
ꢀ
ꢀ ꢀ
ꢀ
Car C=O), 133.2 (s, CHar), 138.5 (s, Car C O), 141.4 (s, C O), 165.9 ppm
137.4 (s, CHar), 147.8 (s, Car C=O), 173.4 (d, 2JCP =118 Hz; Car Pd),
ꢀl
ꢀ
ꢀ
ꢀ
(s, C=O); IR (Nujol): n˜ =1770 (O C=O), 1695 cm (C=C); HR EIMS:
3
221.9 ppm (dd, JCP =8, 3 Hz; C=O); IR (Nujol): n˜ =1580 cmꢀl (C=O); el-
m/z calcd for C11H10O2: 174.0681; found: 174.0680.
emental analysis calcd (%) for C46H59BP2OPd: C 68.5, H 7.4; found: C
68.4, H 7.4.
ꢀ
Synthesis of [Ni{o-C6H4 C(=CHCHOHMe)O}(dippe)] (19a): Acetalde-
hyde in CH2Cl2 (2 mL of a 0.17m solution) was added to complex 8a
(150 mg, 0.34 mmol), which was placed in a Schlencktube, at ꢀ788C.
The mixture was stirred at room temperature for 30 min, and the solvent
was evaporated under reduced pressure, to give a red solid. This solid
was washed twice with petroleum ether (2”10 mL) and was judged pure
by 1H, 13C{1H} and 31P{1H} NMR spectroscopy. No satisfactory analytical
data could be obtained for this complex, which decomposed upon at-
tempted recrystallization, even under an inert atmosphere. 1H NMR
(C6D6, 208C): d=1.18 (m, 12H; CH3), 1.18 (s, 3H; CH3), 1.33 (m, 12H;
CH3), 1.45 (m, 2H; CH2), 1.75 (m, 2H; CH2), 2.21 (m, 2H; CH), 2.37 (m,
2H; CH), 4.70 (m, 1H; =CH), 4.73 (m, 1H; CH-OH), 5.77 (brs, 1H;
OH), 6.76 (t, 3JHH =7.1 Hz, 1H; CHar), 6.81 (t, 3JHH =7.3 Hz, 1H; CHar),
Synthesis of [Ni(C6H4-o-C(O)CH3)(OAr)(PMe3)2] (17a and 17b):
Sodium 2,6-dimethylphenoxide (2.1 mL of a 0.48m solution in THF,
1 mmol) was added to a solution of complex 3 (365 mg, 1 mmol) in THF
(50 mL) cooled to ꢀ808C. The mixture was stirred for one hour at room
temperature, and the solvent was pumped off. The residue was extracted
with Et2O (30 mL) and the solution centrifuged. Reduction of the
volume and cooling to ꢀ208C produced yellow-green crystals of complex
1
17a in essentially quantitative yield. H NMR (C6D6, 208C): d=0.52 (brs,
18H; CH3), 2.38 (s, 3H; CH3), 2.72 (s, 3H; CH3), 3.36 (s, 3H; CH3), 6.82
3
(t, JHH =7.3 Hz, 1H; CHar), 6.69 (m, 2H; CHar), 7.24 (m, 2H; CHar), 7.35
(d, 3JHH =7.8 Hz, 1H; CHar), 7.82 ppm (d, 3JHH =7.5 Hz, 1H; CHar);
31P{1H} NMR (C6D6, 208C): d=ꢀ20.4 ppm (s); 13C{1H} NMR (C6D6,
208C): d=11.8 (brs, CH3), 18.6 (s, CH3), 19.1 (s, CH3), 26.4 (s, CH3),
112.4 (s, CHar), 120.7 (s, CHar), 122.5 (s, CHar), 127.9 (s, CHar), 128.6 (s,
7.09 (d, 3JHH =7.4 Hz, 1H; CHar), 7.13 ppm (t, JHH ~3JHP ~6.6 Hz, 1H;
3
CHar); 31P{1H} NMR (C6D6, 208C): d=73.0 (A, AX spin system),
78.7 ppm (X, AX spin system, JAX =24 Hz); 13C{1H} NMR (C6D6, 208C):
d=16.7 (dd, 1JCP =22 Hz, 2JCP =10 Hz;, CH2), 18.6 (s, CH3), 18.7 (d,
ꢀ
ꢀ
CHar), 130.2 (s, CHar), 137.6 (s, CHar), 143.5 (s, Car CH3), 166.3 (s, Car
O), 167.9 (t, 2JCP =45 Hz; Car Ni), 198.8 ppm (s, C=O); IR (Nujol): n˜ =
2JCP =5 Hz; CH3), 18.8 (d, 2JCP =5 Hz; CH3), 19.0 (s, CH3), 19.6 (d, JCP
=
2
ꢀ
1645 cmꢀl (C=O); elemental analysis calcd (%) for C22H34P2O2Ni: C 58.6,
2
2
4 Hz; CH3), 19.9 (d, JCP =5 Hz; CH3), 21.9 (d, JCP =6 Hz; CH3), 22.0 (d,
2JCP =6 Hz; CH3), 22.4 (dd, 1JCP =27 Hz, 2JCP =21 Hz; CH2), 24.2 (d,
H 7.6; found: C 58.8, H 7.4.
1JCP =17 Hz; CH), 24.5 (s, CH3), 24.7 (d, JCP =17 Hz; CH), 26.0 (d, JCP
=
1
1
The related complex 17b was obtained also in quantitative yield follow-
23 Hz; CH), 26.1 (d, 1JCP =22 Hz; CH), 66.4 (s, CH-OH), 97.4 (s, =CH),
ing the same preparation, but using 2,4-dimethylphenoxide. 1H NMR
121.1 (s, CHar), 123.6 (s, CHar), 125.3 (d, JCP =6 Hz, CHar), 138.1 (s, CHar),
(C6D6, 208C): d=0.59 (brs, 18H; CH3), 2.38 (s, 3H; CH3), 2.41 (s, 3H;
154.3 (s, Car C O), 158.1 (dd, 2JCP =28, 85 Hz; Car Ni), 169.8 ppm (d,
3
CH3), 2.53 (s, 3H; CH3), 6.70 (t, 3JHH =7.4 Hz, 1H; CHar), 6.87 (t, JHH
=
ꢀ ꢀ
ꢀ
3JCP =14 Hz; C O); IR (Nujol): n˜ =3200 (OH), 1605 cm (C=C).
ꢀl
ꢀ
7.5 Hz, 1H; CHar), 7.07 (brs, 1H; CHar), 7.26 (brs, 2H; CHar), 7.37 (d,
3JHH =7.7 Hz, 1H; CHar), 7.85 ppm (brs, 1H; CHar); 31P{1H} NMR (C6D6,
Synthesis of [Ni{o-C6H4 C(=CHCHOHPh)O}(dippe)] (19b): Complex
ꢀ
1
208C): d=ꢀ18.2 ppm (s); 13C{1H} NMR (C6D6, 208C): d=11.9 (t, JCP
=
8a (160 mg, 0.36 mmol) was dissolved in THF (2 mL). PhCHO (0.37 mL,
3.6 mmol) was added at room temperature. After 10 min the solvent was
removed under reduced pressure. An oily residue was obtained that was
12 Hz CH3), 18.1 (s, CH3), 20.7 (s, CH3), 26.2 (s, CH3), 118.7 (s, CHar),
ꢀ
ꢀ
119.8 (s, Car C=O), 120.6 (s, CHar), 127.1 (s, Car CH3), 127.6 (s, CHar),
Chem. Eur. J. 2005, 11, 6889 – 6904
ꢁ 2005 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chemeurj.org
6901

E. Carmona, J. Cµmpora et al.
washed twice with petroleum ether (20 mL), and then recrystallized from
toluene/Et2O to afford 19b (112 mg, 55% yield) as yellow crystals.
1H NMR (C6D6, 208C): d=0.76 (m, 14H; 4CH3 and 1CH2), 1.01 (m, 2H;
CH2), 1.10 (m, 6H; CH3), 1.34 (m, 6H; CH3), 2.03 (m, 4H; CH), 5.48
(dd, 5JHP =4.9 Hz, 4JHH =1.2 Hz, 1H; =CH), 6.91 (d, 4JHH =3.4 Hz, 1H;
(s, =CH), 121.9 (s, CHar), 122.8 (s, CHar), 123.3 (s, CHar), 123.5 (d, JCP =
7 Hz; CHar), 127.1 (s, CHar), 129.2 (s, CHar), 137.6 (s, CHar), 152.5 (s, Car),
3
157.4 (s, Car), 159.0 (dd, 2JCP =88, 26 Hz; Car Ni), 167.6 ppm (d, JCP
=
ꢀ
ꢀl
ꢀ
15 Hz; C O); IR (Nujol): n˜ =1595 cm (C=C); elemental analysis calcd
(%) for C51H80P4O2Ni2: C 63.4, H 8.3; found: C 62.9, H 8.5.
3
OH), 7.10 (m, 4H; CHar), 7.28 (m, 3H; CHar), 7.48 (d, JHH =7.3 Hz, 1H;
ꢀ
Synthesis of [O=C-o-C6H4C(=CH )O)]2CHPh (22): This compound was
CHar), 7.93 ppm(d, 3JHH=7.9 Hz, 1H; CHar); 31P{1H} NMR (C6D6, 208C):
prepared by the procedure given for 18a, starting from complex 21 and
using a 3:1 mixture of petroleum ether/Et2O as eluent. Yield: 45%.
1H NMR (C6D6, 208C): d=5.63 (t, 3JHH =9.7 Hz, 1H; Ph-CH), 6.01 (d,
d=73.3 (A, AX spin system), 78.1 ppm (X, AX spin system, JAX
=
2
22 Hz); 13C{1H} NMR (C6D6, 208C): d=15.1 (dd, 1JCP =21 Hz, JCP
=
2
2
3
3
3JHH =9.7 Hz, 2H; =CH), 7.29 (t, JHH =7.3 Hz, 1H; CHar), 7.39 (t, JHH
=
10 Hz; CH2), 17.9 (s, CH3), 18.0 (d, JCP =4 Hz; CH3), 18.1 (d, JCP =5 Hz;
CH3), 18.3 (s, CH3), 19.4 (d, JCP =5 Hz; CH3), 19.5 (d, 2JCP =5 Hz; CH3),
2
3
7.6 Hz, 2H; CHar), 7.48 (d, JHH =7.7 Hz, 2H; CHar), 7.58 (m, 2H; CHar),
21.2 (s, CH3), 21.3 (s, CH3), 21.9 (dd, 1JCP =26 Hz, 2JCP =22 Hz; CH2),
3
7.74 (m, 4H; CHar), 7.91 ppm (d, JHH =7.6 Hz, 2H; CHar); 13C{1H} NMR
23.8 (d, 1JCP =16 Hz; CH), 24.2 (d, 1JCP =16 Hz; CH), 25.2 (d, JCP
=
1
ꢀ
(C6D6, 208C): d=39.3 (s, Ph CH), 108.5 (s, =CH), 120.4 (s, CHar), 125.0
22 Hz; CH), 25.4 (d, 1JCP =22 Hz; CH), 73.2 (s, CH OH), 97.3 (s, =CH),
122.0 (s, CHar), 123.8 (s, CHar), 124.9 (d, JCP =7 Hz; CHar), 125.9 (s,
CHar), 126.8 (s, CarH), 127.9 (s, CarH), 137.4 (m, CarH), 148.5 (s, Car),
ꢀ
ꢀ
(s, Car C=O), 125.6 (s, CHar), 127.4 (s, CHar), 127.9 (s, CHar), 129.3 (s,
ꢀ ꢀ
CHar), 130.4 (s, CHar), 134.8 (s, CHar), 139.7 (s, Car), 142.1 (s, Car C O),
ꢀ
ꢀ
146.2 (s, =C O), 166.7 ppm (s, C=O); IR (Nujol): n˜ =1775 (O C=O),
1695 cmꢀ1 (C=C); HREIMS: m/z calcd for C25H16O4: 380.1049; found:
380.1045.
155.1 (s, Car), 157.9 (dd, 2JCP =28, 85 Hz; Car Ni), 169.8 ppm (d, JCP
=
3
ꢀ
ꢀl
ꢀ
14 Hz; C O); IR (Nujol): n˜ =3250 (OH), 1605 cm (C=C); elemental
analysis calcd (%) for C29H44P2O2Ni: C 63.9, H 8.1; found: C 64.2, H 7.8.
Computational details: All calculations were performed using the pack-
age Spartan Pro.[25] Initial guess of the molecular geometry was obtained
with the semiempirical PM3 method, and the resulting structures were
fully optimized without restrictions with DFT methods, using the BP86
functional and the numerical basis set DN*, which includes d-type polari-
zation functions for all non-hydrogen atoms. The gradient correction was
included in a perturbative manner only after convergence on a local po-
tential was achieved (pBP method). The geometry of the minima and
saddle-points of the potential surface were checked with a frequency cal-
culation, only for the more simple models with H substituents on the
phosphine ligand. For the calculation of the geometry of “real” molecules
containing isopropyl substituents, a guess model of the molecular geome-
try was built starting from result of a previous calculation on the simpli-
fied molecules, and those were subjected to a molecular mechanics con-
formational analysis (MMFF, MerckMolecular Force Field), while main-
taining fixed the positions of the core atoms. The geometries of the five
most stable conformers were subjected to full optimization without re-
straints by using the DFT method, and the lower computed energies
were used in the thermochemical and activation barrier calculations.
Synthesis of O=C-o-C6H4C(=CHCHOHR)O (R=Me, 20a; R=Ph, 20b):
CO was bubbled through a solution of 19a (100 mg, 0.18 mmol) in THF
(15 mL). After 5 min the initial light orange solution turned pale yellow.
The solvent was evaporated under vacuum and the resulting oil was ex-
tracted with petroleum ether (30 mL) and filtered. Compound 20a was
separated from the complex [Ni(dippe)(CO)2] by spinning band chroma-
tography with Et2O as eluent. Yield: 76 mg, 40%; 1H NMR (C6D6,
3
208C): d=1.26 (d, JHH =6.4 Hz, 3H; CH3), 3.89 (brs, 1H; OH), 4.94 (m,
1H; CH-OH), 5.35 (d, 3JHH =8.3 Hz, 1H; =CH), 6.78 (t, 3JHH =7.3 Hz,
1H; CHar), 6.90 (m, 2H; CHar), 7.53 ppm (d, 3JHH =7.5 Hz, 1H; CHar);
13
1
ꢀ
C{ H} NMR (C6D6, 208C): d=23.7 (s, CH3), 63.04 (s, CH OH), 112.6
(s, =CH), 120.2 (s, CHar), 125.1 (s, Car-C=O), 125.4 (s, CHar), 129.9 (s,
ꢀ ꢀ
ꢀ
CHar), 134.0 (s, CHar), 139.7 (s, Car C O), 144.8 (s, C O), 166.2 ppm (s,
ꢀl
ꢀ
C=O); IR (Nujol): n˜ =3250 (OH), 1785 (O C=O), 1690 cm (C=C); HR
EIMS: m/z calcd for C10H8O2: 190.0630; found: 190.0631.
Compound 20b was prepared in a similar way. A 3:2 mixture of Et2O/pe-
troleum ether was used as eluent. (45% yield). 1H NMR (C6D6, 208C):
3
d=4.42 (s, 1H; OH), 5.60 (d, 3JHH =9.0 Hz, 1H; =CH), 6.02 (d, JHH
=
9.0 Hz, 1H; CHOH), 6.75 (m, 1H; CHar), 6.76 (d, 3JHH =6.9 Hz, 1H;
CHar), 6.85 (t, JHH =7.5 Hz, 1H; CHar), 7.06 (t, JHH =7.1 Hz, 1H; CHar),
7.18 (t, 3JHH =7.4 Hz, 2H; CHar), 7.50 (d, 3JHH =7.8 Hz, 1H; CHar),
7.55 ppm (d, 3JHH =7.5 Hz, 2H; CHar); 13C{1H} NMR (C6D6, 208C): d=
X-ray structure determination of 8band 15 : A summary of the funda-
mental crystal and refinement data are given in Table 1 in the Supporting
Information. CCDC-273534 (8b) and CCDC 273533 (15) contain the sup-
plementary crystallographic data for this paper. These data can be ob-
tained free of charge from The Cambridge Crystallographic Data Centre
via www.ccdc.cam.ac.uk/data_request/cif.
3
3
ꢀ
ꢀ
68.1 (s, CH OH), 110.9 (s, =CH), 119.9 (s, CHar), 124.7 (s, Car C=O),
124.9 (s, CHar), 126.0 (s, CHar), 128.0 (s, CHar), 128.4 (s, CHar), 129.4 (s,
ꢀ ꢀ
ꢀ
CHar), 133.5 (s, CHar), 139.1 (s, Car C O), 143.5 (s, Car), 144.8 (s, C O),
Compound 8b: The intensity data of compounds 8b were collected at
173 K on a ENRAF Nonius CAD4 single-crystal diffractometer equip-
ped with a fine-focus sealed tube graphite monochromated radiation
source (radiation type CuKa, l=1.54180 Š), using a q/2q scan method. To
determine the cell parameters 24 reflections between 24 and 358 were
used; one standard reflection every 100 counts was measured to monitor
crystal decay.
ꢀ
165.8 ppm (s, C=O); IR (Nujol): n˜ =3330 (OH), 1785 (O C=O),
1685 cmꢀl (C=C); HR EIMS: m/z calcd for C16H12O3: 252.0786; found:
252.0780.
Synthesis of binuclear complex 21
Method a: Enolate 8a (160 mg, 0.36 mmol) was disso1ved in THF
(2 mL). Freshly distilled C6H5C(O)H (0.37 mL, 3.6 mmol) was added to
this solution at ꢀ788C. The mixture was stirred for 48 h at room temper-
ature, after which time the solvent was removed under reduced pressure.
The residue was washed twice with petroleum ether (20 mL) and then re-
crystallized from Et2O to afford 21 as yellow crystals.
The structure was solved by Fourier methods and refined by full-matrix
least-squares procedures (based on F2o)[26] with anisotropic thermal pa-
rameters in the last cycles of refinement for all the non-hydrogen atoms.
The data were corrected empirically[27] by absorption effect. The hydro-
gen atoms were introduced into the geometrically calculated positions
and refined riding on the corresponding parent atoms. In the final cycles
of refinement a weighting scheme, w=1/[s2Fo2 +(0.0891P)2 +4.4001P], in
which P=(F2o +2F2c)/3, was used.
Method b: A solution of complex 19b (100 mg, 0.18 mmol) in THF
(2 mL) was stirred for 48 h to effect dimerization to 21 in quantitative
yield. 1H NMR (C6D6, 208C): d=ꢀ0.81 (m, 16H; 4CH3, 2CH2), 0.88 (m,
4H; CH2), 0.96 (m, 12H; CH3), 1.18 (m, 12H; CH3), 1.38 (m, 12H; CH3),
1.94 (m, 2H; CH), 2.04 (m, 6H; CH), 5.56 (d, 3JHH =6.8 Hz, 1H; CH),
5.65 (t, 3JHH =7.2 Hz, 2H; =CH), 7.05 (t, 3JHH =7.4 Hz, 1H; CHar), 7.11
(m, 4H; CHar), 7.31 (t, 3JHH =7.4 Hz, 2H; CHar), 7.37 (s, 2H; CHar), 7.65
(s, 2H; CHar), 7.92 ppm (d, 3JHH =7.5 Hz, 2H; CHar); 31P{1H} NMR
Compound 15: A crystal with well-defined faces was coated with a poly-
fluoroether oil coated and cooled at 173 K. It was mounted on a Bruck-
er-Siemens Smart CCD diffractometer equipped with a normal focus,
2.4 kW sealed tube X-ray source (MoKa radiation, l=0.71073 Š) operat-
ing at 50 kV and 20 mA. Data were collected by a combination of two
frame sets covering more than a hemisphere of the reciprocal space. The
cell parameters were determined and refined by a least-squares fit of all
reflection collected. Each frame exposure time was of 20 s covering 0.38
in w. The crystal to detector distance was 5.02 cm. Coverage of the
unique set was over 94% complete to at least 308 in q. The first 50
(C6D6, 208C): d=73.7 (A, AX spin system), 77.2 ppm (X, AX spin
system, JAX =23 Hz); 13C{1H} NMR (C6D6, 208C): d=15.7 (dd, JCP
=
1
2
19 Hz, JCP =10 Hz; CH2), 18.1 (s, CH3), 18.2 (s, CH3), 18.7 (s, CH3), 19.6
(d, 2JCP =4 Hz; CH3), 20.2 (d, 2JCP =5 Hz; CH3), 21.3 (s, CH3), 22.8 (t,
J
CP =24 Hz; CH2), 23.5 (d, 1JCP =17 Hz; CH), 24.2 (d, 1JCP =17 Hz; CH),
25.3 (d, 1JCP =21 Hz; CH), 25.4 (d, 1JCP =21 Hz; CH), 39.8 (s, CH), 100.3
6902
www.chemeurj.org
ꢁ 2005 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Chem. Eur. J. 2005, 11, 6889 – 6904

Metallacyclic Enolates
FULL PAPER
frames were recollected at the end of the data collection to monitor crys-
tal decay. A multiscan absorption correction (SADABS[26]) was applied.
The structure was solved by Multan and Fourier methods using
SHELXS.[26] Full-matrix least-square refinement was carried out using
SHELXTL[26] minimizing w(F2oꢀF2c)2. Weighted R factors (Rw) and all
goodness-of-fit (S) values are based on F2, conventional R factors (R) are
based on F.
nometallics 2004, 23, 3398; c) A. C. AlbØniz, N. M. Catalina, P. Espi-
net, R. Redón, Organometallics 1999, 18, 5571; d) J. Vicente, J. A.
Abad, R. Bergs, P. G. Jones, D. J. Bautista, Chem. Soc. Dalton Trans.
1995, 3093; e) J. Vicente, A. Arcas, J. M. Femµndez-Hemµndez, D.
Bautista, Organometallics 2001, 20, 2767; f) P. Veya, C. Floriani, A.
Chiesi-Vil1a, C. Rizzoli, Organometallics 1993, 12, 4899; g) W. Hen-
derson, J. Fawcett, R. D. W. Kemitt, P. McKenna, R. R. Russell,
Polyhedron 1997, 16, 2475; h) L. R. Falvello, R. Garde, E. M. Mi-
queleiz, M. Tomas, J. P. Urriolabeitia, Inorg. Chim. Acta 1997, 297;
i) R. A. Wanat, D. B. Col1um, Organometallics 1986, 5, 120; j) M.
Green, J. A. K. Howard, P. Mitrprapachon, M. Pfeffer, J. L. Spencer,
P. Woodward, J. Chem. Soc. Dalton Trans. 1979, 306; k) D. R.
Russel1, P. A. Tcker, J. Chem. Soc. Dalton Trans. 1975, 2222;
l) M. A. Bennett, G. B. Robertson, P. O. Whimp, T. Yoshida, J. Am.
Chem. Soc. 1973, 95, 3028.
Acknowledgements
Financial support from the DGI (Project PPQ2003–000975) and Junta de
Andalucía is gratefully acknowledged. C.M.M. thanks the Dirección Gen-
eral de EnseÇanza Superior for a PFPI studentship.
[8] a) J. Cµmpora, C. M. Maya, P. Palma, E. Carmona, C. Graiff, A. Tir-
ipicchio, Chem. Commun. 2003, 1742; b) K. K. D. Amarasinghe,
S. K. Chowdhury, M. J. Heeg, J. Montgomery, Organometallics 2001,
20, 370.
[1] a) I. Paterson in Comprehensive Organic Synthesis, Vol. 2. (Eds.:
B. M. Trost, I. Fleming, C. H. Heathcock), Pergamon, Oxford,
1991, pp. 301–319; b) C. Palomo, M. Oiarbide, J. M. García, Chem.
Soc. Rev. 2004, 33, 65; c) J. Montgomery, Acc. Chem. Res. 2000, 33,
467; d) B. C. Hamann, J. F. Hartwig, J. Am. Chem. Soc. 1997, 119,
12382; e) M. Palucki, S. L Buchwald, J. Am. Chem. Soc. 1997, 119,
11108; f) T. Satoh, Y. Kawamura, M. Miura, M. Nomura, Angew.
Chem. 1997, 109, 1820; Angew. Chem. Int. Ed. Engl. 1997, 36, 1740;
g) J. Ahman, J. P. Wolfe, M. V. Troutman, M. Palucki, S. L. Buch-
wald, J. Am. Chem. Soc. 1998, 120, 1918; h) A. Fuji, E. Hagiwara,
M. Sodeoka, J. Am. Chem. Soc. 1999, 121, 5450; i) O. Fujimura, J.
Am. Chem. Soc. 1998, 120, 10032; j) J. Krüger, E. M. Carreira, J.
Am. Chem. Soc. 1998, 120, 837; k) B. L. Pagenkopf, J. Krüger, A.
Stojanovic, E. M. Carreira, Angew. Chem. 1998, 110, 3312; Angew.
Chem. Int. Ed. 1998, 37, 3124.
[9] a) J. Cµmpora, J. A. López, P. Palma, P. Valerga, E. Spil1ner, E. Car-
mona, Angew. Chem. 1999, 111, 199; Angew. Chem. Int. Ed. 1999,
38, 147; b) J. Cµmpora, J. A. López, P. Palma, D. del Rio, E. Carmo-
na, P. Valerga, C. Graiff, Inorg. Chem. 2001, 40, 4116; c) J. Cµmpora,
M. M. Conejo, K. Mereiter, P. Palma, C. PØrez, M. L. Reyes, C.
Ruiz, J. Organomet. Chem. 2003, 683, 220; d) J. Cµmpora, E. GutiØr-
rez, A. Monge, P. Palma, M. L. Poveda, C. Ruiz, E. Carmona, Orga-
nometallics 1994, 13, 1728; e) J. Cµmpora, M. Llebaria, J. M.
Moretó, M. L. Poveda, E. Carmona, Organometallics 1993, 12, 4032;
f) E. Carmona, GutiØrrez-Puebla, J. M. Marín, A. Monge, M. Pane-
que, M. L. Poveda, C. Ruiz, J. Am. Chem. Soc. 1989, 111, 2883.
[10] J. Cµmpora, C. M. Maya, P. Palma, E. Carmona, E. GutiØrrez-
Puebla, J. Am. Chem. Soc. 2003, 125, 1482.
[2] a) G. A. Slough, R. G. Bergman, C. H. Heathcock, J. Am. Chem.
Soc. 1989, 111, 938; b) J. F. Hartwig, R. A. Andersen, R. G. Berg-
man, J. Am. Chem. Soc. 1990, 112, 5670; c) L. F. Hartwig, R. G.
Bergman, R. A. Andersen, Organometallics 1991, 10, 3344; d) J. F.
Hartwig, R. G. Bergman, R. A. Andersen, Organometallics 1991, 10,
3326; e) J. G. Stack, J. J. Doney, R. G. Bergman, C. H. Heathcock,
Organometallics 1990, 9, 453; f) R. R. Burkhardt, R. G. Bergman,
C. H. Heathcock, Organometallics 1990, 9, 30.
[3] a) G. M. Mahandru, A. R. L. Skauge, S. K. Chowdhury, K. K. D.
Amarasinghe, M. J. Heeg, J. Montgomery, J. Am. Chem. Soc. 2003,
125, 13481; b) S. Ikeda, H. Miyahita, M. Taniguchi, H. Kondo, M.
Okani, Y. Sato, K. Odasuma, J. Am. Chem. Soc. 2002, 124, 12060;
c) M. Sodeoka, R. Tokunoh, F. Miyakazi, E. Hagiwara, M. Shibasa-
ki, Synlett 1997, 463; d) M. Sodeoka, K. Ohrai, M. Shibasaki, J. Org.
Chem. 1995, 60, 2648.
[4] a) D. A. Culkin, J. F. Hartwig, J. Am. Chem. Soc. 2001, 123, 5816;
b) G. A. Slough, R. G. Bergman, C. H. Heathcock, J. Am. Chem.
Soc. 1989, 111, 938; c) G. A. Slough, R. Hayashi, J. R. Ashbaugh,
S. L. Shamblin, A. M. Aukamp, Organometallics 1994, 13, 890.
[5] a) P. Veya, P. G. Cozzi, C. Floriani, F. P. Rotzinger, A. Chiesi-Villa,
C. Rizzoli, Organometallics 1995, 14, 4101; b) P. Veya, C. Floriani,
A. Chiesi-Villa, C. Rizzoli, Organometallics 1994, 13, 208; c) P.
Veya, C. Floriani, A. Chiesi-Villa, C. Rizzoli, Organometallics 1993,
12, 4892; d) P. G. Cozzi, P. Veya, C. Floriani, F. P. Rotzinger, A.
Chiesi-Vil1a, C. Rizzoli, Organometallics 1995, 14, 4092; e) D. A.
Evans, L. R. McGee, Tetrahedron Lett. 1980, 21, 3975; f) Y. Yama-
moto, K. Maruyama, Tetrahedron Lett. 1980, 21, 4607; g) J. Manri-
quez, P. J. Fagan, T. J. Marks, J. Am. Chem. Soc. 1978, 100, 7112;
h) D. C. Sonnenberger, E. A. Mintz, T. J. Marks, J. Am. Chem. Soc.
1984, 106, 3484; i) T. Mukaiyama, K. Banno, K. J. Narasaka, J. Am.
Chem. Soc. 1974, 96, 7503; j) J. R. Stil1e, R. H. Grubbs, J. Am.
Chem. Soc. 1983, 105, 1664; k) M. D. Curtis, S. Thanedar, W. M.
Butler, Organometallics 1984, 3, 1855.
[11] J. Cµmpora, C. Graiff , P. Palma, E. Carmona, A. Tiripicchio, Inorg.
Chim. Acta 1998, 269, 191.
[12] E. Carmona, E. GutiØrrez-Puebla, A. Monge, J. M. Marín, M. Pan-
eque, M. L. Poveda, Organometallics 1989, 8, 967.
[13] a) E. Carmona, J. M. Marín, P. Palma, M. Paneque, M. L. Poveda,
Organometallics 1985, 4, 2053; b) E. Carmona, J. M. Marín, M. Pan-
eque, M. L. Poveda, Organometallics 1987, 6, 1757; c) R. A. Michel-
in, M. Napoli, J. Organomet. Chem. 1979, 175, 239; d) V. V. Grushin,
H. Alper, Organometallics, 1993, 12, 1980; e) J. C. M: Ritter, R. G.
Bergman, J. Am. Chem. Soc. 1977, 99, 119, 2580; f) J. C. M. Ritter,
R. G. Bergman, J. Am. Chem. Soc. 1998, 120, 6826.
[14] B. L. Jensen, Tetrahedron 1978, 34, 1627.
[15] a) J. Cµmpora, P. Palma, D. del Río, M. M. Conejo, E. lvarez, Or-
ganometallics 2004, 23, 5653; b) J. Cµmpora, P. Palma, D. del Río, E.
lvarez, Organometallics 2004, 23, 1652; c) P. T. Matsunaga, G. L.
Hillhouse, J. Am. Chem. Soc. 1993, 115, 2075; d) K. Koo, G. L. Hill-
house, A. L. Rheingold, Organometallics 1995, 14, 456; e) J.
Cµmpora, J. A. López, C. Maya, P. Palma, E. Carmona, P. Valerga,
J. Organomet. Chem. 2002, 643, 331; f) A. L. Seligson, R. L. Cowan,
W. C. Trogler, Inorg. Chem. 1991, 30, 3371; g) H. F. Klein, A. Dal, T.
Jung, S. Braun, C. Rçhr, U. Flçrke, H. Haupt, Eur. J. Inorg. Chem.
1998, 621; h) H. Klein, A. Dal, T. Jung, U. Flçrke, H. Haupt, Eur. J.
Inorg. Chem. 1998, 2027.
[16] a) H. F. Klein, X. Li, H. Sun, A. Brand, M. Lemke, J. Flçrke, H. J.
Haupt, Inorg. Chim. Acta 2000, 298, 70; b) J. D. Martin, K. A.
Abboud, K. H. Dahmen, Inorg. Chem. 1998, 37, 5811; c) H. Qui-
chen, X. Minzhi, Q. Yanlong, X. Weihua, S. Meicheng, T. Youqi, J.
Organomet. Chem. 1985, 287, 419.
[17] As observed for the PMe3 derivatives 3 and 4, the frequency of the
n(C=O) band of the acetyl group of 6a–c is about 70 cmꢀ1 lower
than in the corresponding organic precursors, suggesting the exis-
tence of a Ni···O interaction in these molecules. However, if this in-
teraction exists, it must be weak, as 1H and 13C NMR spectra of the
these compounds indicate free rotation of the aryl group relative to
[6] a) E. R. Burkhardt, J. J. Doney, R. G. Bergman, C. H. Heathcock, J.
Am. Chem. Soc. 1987, 109, 2022.
[7] a) P. Dotta, R. G. A. Kumar, P. S. Pregosin, A. Albinati, S. Rizzato,
Organometallics 2003, 22, 5345; b) D. A. Culkin, J. F. Hartwig, Orga-
ꢀ
the Ni C bond.
Chem. Eur. J. 2005, 11, 6889 – 6904
ꢁ 2005 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chemeurj.org
6903

E. Carmona, J. Cµmpora et al.
[18] The form of the equilibrium relaxation rate law can be reduced to
simple first-order kinetics. See: K. A. Connors, Chemical Kinetics:
The Study of Reaction Rates in Solution, VCH, New York, 1990.
[19] A similar transition state has been characterized for a C!O enolate
transformation: H. P. Hratchian, S. K. Chowdury, V. M. GutiØrrez-
García, K. K. D. Amarasinghe, M. J. Heeg, H. B. Schlegel, J. Mont-
gomery, Organometallics, 2004, 23, 4636.
[23] Y. Tsatsumo, T. Yoshida, S. Otsuka, Inorg. Synth. 1990, 28, 343.
[24] E. Carmona, M. Paneque, M. L. Poveda, Polyhedron 1989, 8, 285.
[25] Mac Spartan Pro, Vers. 1.0.4. Wavefunction, Inc., 18401 Von
Karman, Suite 370, Irvine, California 92612, 2000.
[26] G. M. Sheldrick, University of Gçttingen, Gçttingen, Germany,
1997.
[27] a) N. Walker, D. Stuart, Acta Crystallogr. Sect. A 1983, 39, 158; b) F.
Ugozzoli, Comput. Chem. 1987, 11, 109.
[20] Y. Uozumi, E. Mori, M. Mori, M. Shibasaki, J. Organomet. Chem.
1990, 399, 93.
[21] M. Brookhart, B. Grant, A. F. Volpe Organometallics 1992, 11, 3920.
[22] M. M. Colquhoun, J. Holton, D. J. Thompson, M. V. Twiggs, New
Pathways for Organic Synthesis. Practical Applications of Transition
Metals, Plenum Press, New York 1984, p. 389.
Received: June 1, 2005
Published online: September 21, 2005
6904
www.chemeurj.org
ꢁ 2005 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Chem. Eur. J. 2005, 11, 6889 – 6904