Synthesis and reactivity of six-membered oxa-nickelacycles: a ring-opening reaction of cyclopropyl ketones

Article abstract of DOI:10.1002/chem.200900929

Cyclopropanecarboxaldehyde (la), cyclopropyl methyl ketone (lb), and cyclopropyl phenyl ketone (Ic) were reacted with [Ni(cod)₂] (cod = l,5-cyclooctadiene) and PBu₃ at 100°C to give η²-enonenickel complexes (2a-c). In the

Full text of DOI:10.1002/chem.200900929

FULL PAPER
DOI: 10.1002/chem.200900929
Synthesis and Reactivity of Six-Membered Oxa-Nickelacycles:
A Ring-Opening Reaction of Cyclopropyl Ketones
Takashi Tamaki, Midue Nagata, Masato Ohashi, and Sensuke Ogoshi*[a]
Abstract:
Cyclopropanecarboxalde-
2-ylidene (IPr), both 1a and 1c rapidly
underwent oxidative addition to
nickel(0) to give the corresponding six-
membered oxa-nickelacycles (6ai, 6ci).
On the other hand, 1b reacted with
nickel(0) to give the corresponding m-
h2:h1-enonenickel complex (3bi). The
molecular structures of 6ai and 6ci
were confirmed by X-ray crystallogra-
phy. The molecular structure of 6ai
shows a dimeric h1-nickelenolate struc-
ture. However, the molecular structure
of 6ci shows a monomeric h1-nickel-
enolate structure, and the nickel(II) 14-
electron center is regarded as having
“an unusual T-shaped planar” coordi-
nation geometry. The insertion of
enones into monomeric h1-nickeleno-
late complexes 6c and 6ci occurred at
room temperature to generate h3-oxa-
allylnickel complexes (8, 9), whereas
insertion into dimeric h1-nickelenolate
complex 6ai did not take place. The
diastereoselectivity of the insertion of
an enone into 6c having PCy3 as a
ligand differs from that into 6ci having
IPr as a ligand. In addition, the stereo-
chemistry of h3-oxa-allylnickel com-
plexes having IPr as a ligand is re-
tained during reductive elimination to
yield the corresponding [3+2] cycload-
dition product, which is consistent with
the diastereoselectivity observed in Ni0/
IPr-catalyzed [3+2] cycloaddition reac-
tions of cyclopropyl ketones with
enones. In contrast, reductive elimina-
tion from the h3-oxa-allylnickel having
PCy3 as a ligand proceeds with inver-
sion of stereochemistry. This is proba-
bly due to rapid isomerization between
syn and anti isomers prior to reductive
elimination.
hyde (1a), cyclopropyl methyl ketone
(1b), and cyclopropyl phenyl ketone
(1c) were reacted with [NiACTHNUTRGENUG(N cod)2]
(cod=1,5-cyclooctadiene) and PBu3 at
1008C to give h2-enonenickel com-
plexes (2a–c). In the presence of PCy3
(Cy=cyclohexyl), 1a and 1b reacted
with [NiACHTUNGTRENNUNG(cod)2] to give the correspond-
ing m-h2:h1-enonenickel complexes (3a,
3b). However, the reaction of 1c under
the same reaction conditions gave a
mixture of 3c and cyclopentane deriva-
tives (4c, 4c’), that is, a [3+2] cycload-
dition product of 1c with (E)-1-phenyl-
but-2-en-1-one, an isomer of 1c. In the
presence of a catalytic amount of [Ni-
ACHTUNGTRENNUNG(cod)2] and PCy3, [3+2] homo-cycload-
dition proceeded to give a mixture of
4c (76%) and 4c’ (17%). At room
temperature, a possible intermediate,
6c, was observed and isolated by repre-
cipitation at À208C. In the presence of
1,3-bis(2,6-diisopropylphenyl)imidazol-
Keywords: cycloaddition
·
cyclo-
propanes
nickel
·
enolates
·
enones
·
Introduction
with a nickel(0) complex to give five-membered hetero-
nickelacycles (Scheme 1A). Earlier studies of hetero-nick-
elacycles, reported by Hoberg
Many nickel-catalyzed transformations by means of
a
hetero-nickelacycle have been reported, in which two differ-
ent key reaction steps to generate a hetero-nickelacycle
from nickel(0) species have been proposed. One proposed
reaction is the oxidative cyclization of two p components
et al., focused on oxidative cyc-
lization, including carbon diox-
ide or isocyanate as a p compo-
nent.[1] We recently reported
oxidative cyclization reactions
of carbon–carbon unsaturated
[a] T. Tamaki, M. Nagata, Dr. M. Ohashi, Prof. S. Ogoshi
Department of Applied Chemistry
Faculty of Engineering, Osaka University
Suita, Osaka 565-0871 (Japan)
compounds
ketone,
and
aldehyde,
with
or
imine
nickel(0).[2] These reactions
have been proposed as an im-
portant key step in multicom-
Fax : (+81)6-6879-7394
E-mail: ogoshi@chem.eng.osaka-u.ac.jp
Scheme 1. Formation
hetero-nickelacycles
of
from
Supporting information for this article is available on the WWW
under http://dx.doi.org/10.1002/chem.200900929.
ponent coupling reactions[3] as nickel(0).
Chem. Eur. J. 2009, 15, 10083 – 10091
ꢀ 2009 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
10083

well as [2+2+2] cycloaddition reactions.[4,5] Formation of
cyclic nickelenolates by a similar oxidative cyclization was
also observed by Montgomery and Schlegel et al.[6]
The other type of reaction to generate a hetero-nickelacy-
cle is the oxidative addition of cyclopropyl ketones and cy-
clopropyl imines to give six-membered hetero-nickelacycles
(Scheme 1B), which has also been proposed as an important
key step in nickel-catalyzed [3+2] cycloaddition reactions to
give cyclopentane derivatives. As a stoichiometric reaction,
ring-opening reactions by oxidative addition of cyclopro-
panes to a transition metal have been reported.[7] However,
catalytic applications have been limited due to the poor co-
ordination ability of cyclopropanes.[8] Therefore, cyclopropyl
compounds with an unsaturated bond that can coordinate to
a low valent transition metal, such as methylenecyclopro-
panes,[9] vinylcyclopropanes,[10] cyclopropyl ketones,[11] cyclo-
propyl imines,[12] and allenylcyclopropanes[13] have been em-
ployed as substrates in transition-metal-catalyzed transfor-
mations that include ring-opening reactions of the cyclo-
propyl ring as a key reaction step.
Scheme 2. Reaction of cyclopropyl ketone with nickel(0) and PBu3.
but-2-en-1-one, respectively.[16] Thus, if dissociation of the
coordinated enones from nickel(0) can occur, this isomeriza-
tion reaction might proceed catalytically. However, in the
presence of 10 mol% of [NiACHTNUGRTNEUNG(cod)2] and 20 mol% of PBu3,
no catalytic isomerization reaction of 1c occurred at 1008C
for 48 h. This finding might be due to strong coordination of
the enone to nickel(0) in 2c.
In the presence of PCy3 (Cy=cyclohexyl), the ring-open-
ing reaction of 1a and 1b occurred to give m-h2:h1-enone-
nickel dimer complexes (3a, 3b) quantitatively (Scheme 3).
The nickel-catalyzed [3+2] cycloaddition reaction was si-
multaneously reported for the first time by our group and
by Montgomery and Liu.[14] Our prior work focused on the
catalytic homo-dimerization of cyclopropyl ketones using
phosphine ligands and observation of key intermediates in
the pathway. The complementary studies of Montgomery
and co-workers demonstrated this catalytic reaction using
N-heterocyclic carbene ligands as well as the crossed reac-
tions of cyclopropyl ketones or cyclopropyl imines with
enones. We have recently focused on providing insights into
the mechanism and origin of diastereoselectivity of these re-
actions, and therefore, we have also addressed the formation
of six-membered hetero-nickelacycles by oxidative addition
of cyclopropanes attached to a carbon–oxygen double bond.
In this paper, we report on the synthesis and structure of
six-membered oxa-nickelacycles generated by oxidative ad-
dition of cyclopropyl ketones to nickel(0) in the presence of
PR3 and 1,3-bis(2,6-diisopropylphenyl)imidazol-2-ylidene
(IPr).[15] These oxa-nickelacycles reacted with enones to give
new nickelacycles with h3-enolate structures. We also discuss
a possible reaction mechanism of [3+2] cycloaddition reac-
tion of cyclopropyl ketones with enones.
Scheme 3. Reaction of cyclopropyl ketone with nickel(0) and PCy3.
The molecular structure of 3b was confirmed by X-ray crys-
tallography.[14a] Under the same reaction conditions, the re-
action of 1c gave a mixture of the corresponding m-h2:h1-
enonenickel complex (3c) and unexpected cyclopentane
products (4c, 4c’). In contrast to the reaction in the pres-
ence of PBu3, 3a, 3b, and 3c were generated even in the
presence of two equivalents of PCy3. The cycloaddition reac-
tion proceeded catalytically to give a mixture of 4c and 4c’
in 93% yield (Scheme 4A). At the end of the reaction, the
formation of 3c (76% based on [NiACHTNUTRGNEUNG(cod)2]) was observed.
Although cyclopropyl methyl ketone did not undergo the
homo-cycloaddition reaction under the reaction conditions
Results and Discussion
Reaction of cyclopropyl ketones with nickel(0): The reaction
of cyclopropanecarboxaldehyde (1a), cyclopropyl methyl
ketone (1b), or cyclopropyl phenyl ketone (1c) with [Ni-
ACHTUNGTRENNUNG(cod)2] (cod=1,5-cyclooctadiene) and PBu3 at 1008C in
[D8]toluene gave an h2-enonenickel complex (2a–c) quanti-
tatively (Scheme 2). This result indicates that the ring-open-
ing reaction of the cyclopropyl ring attached to the carbonyl
group occurred at the proximal position toward the carbonyl
group. The treatment of 2a, 2b, and 2c with carbon monox-
ide (5 atm) led to dissociation of the coordinated enones,
(E)-crotonaldehyde, (E)-3-penten-2-one, and (E)-1-phenyl-
Scheme 4. Catalytic reaction of cyclopropyl ketone.
10084
www.chemeurj.org
ꢀ 2009 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Chem. Eur. J. 2009, 15, 10083 – 10091

Six-Membered Oxa-Nickelacycles
FULL PAPER
in Scheme 4A, the cross-cycloaddition with cyclopropyl
phenyl ketone competed with the homo-cycloaddition of cy-
clopropyl phenyl ketone to give a mixture of 4c, 4c’, 4d,
and 4d’ (Scheme 4B). These results suggest that there might
be an intermediate in the reaction that can supply cyclo-
propyl ketone as a three-carbon unit for this cycloaddition,
that is, six-membered oxa-nickelacycle (Scheme 1B with X=
O). Therefore, a stoichiometric reaction at a lower tempera-
ture was carried out to observe the formation of a key reac-
tion intermediate.
Scheme 5. Isolation of 6c.
pitation at À208C gave 6c as a pale orange solid in 25% iso-
lated yield. Elemental analysis is consistent with the expect-
ed composition. The 13C NMR spectroscopic resonance of
the methylene carbon attached to nickel is coupled with
phosphorus. The 1H and 13C chemical shifts of the nickeleno-
late moiety (-NiOC(Ph)=CH-) are in the range of those for
reported nickelenolates.[6] The treatment of 6c with carbon
monoxide (5 atm) led to the formation of the expected lac-
tone (7c) quantitatively,[16,17] which is consistent with the
structure of 6c depicted in Scheme 5. Its monomeric struc-
ture is inferred from the molecular structure of 6ci, an ana-
logue of 6c (vide infra). The decomposition of 6c at room
temperature for 36 h gave a mixture of 3c (68%) and a
trace amount of 4c.
The reaction of 1c (2 equiv) with [NiACTHNUTRGENUN(G cod)2] (1 equiv) and
PCy3 (1 equiv) in C6D6 at 408C was followed by 1H and
31P NMR spectroscopy. The rapid formation of an h2-ketone-
nickel complex (5c, 32% based on Ni) was observed in
5 min. Then, 5c decreased gradually and a new complex
1
(6c) with a resonance at d=5.21 ppm in the H NMR spec-
trum was generated (40%). After 48 h, 6c disappeared and
3c (60%) and a mixture of 4c and 4c’ (40% as a mixture)
were generated with cyclopropyl phenyl ketone (1 equiv)
and 40% of [Ni
ACHTUNGTRENNUNG
The reaction of 1a with [NiAHCTNUGTRENNNUG
generated the expected new complex (6a), with resonances
at d=4.77 and 6.51 ppm attributed to two vinyl hydrogen
atoms, as an inseparable mixture with 3a at room tempera-
In contrast to the reactions in the presence of PCy3 or
PBu3, IPr promoted the ring-opening reaction much faster,
even at room temperature (Scheme 6). In the presence of
ture. The reaction of 1b with [NiACTHUNRGTNEGN(U cod)2] and PCy3 in C6D6 at
room temperature generated the corresponding h2-ketone-
nickel complex (5b) in poor yield, and no oxidative addition
proceeded for at least 48 h. However, at 408C, only enone
dimer complex 3b was observed as a reaction product. Only
6c seemed to be an isolable key intermediate.
The reaction of 1c with [NiACHTNUGRTNEUNG(cod)2] and PCy3 in toluene
gave 5c in 88% isolated yield (Scheme 5). In THF at room
temperature for 5 h, 5c underwent oxidative addition to
nickel(0) to generate 6c in 60% yield. THF and COD were
removed completely under reduced pressure. The residue
was dissolved in a minimum amount of toluene, and repreci-
Scheme 6. Ni0/IPr-promoted ring-opening reaction.
IPr, both 1a and 1c underwent oxidative addition to give
the expected six-membered oxa-nickelacycles (6ai, 6ci). The
treatment of 6ai and 6ci with carbon monoxide gave the
corresponding lactones (7a, 7c), respectively.[16] Both 6ai
and 6ci were isolated; however, 6ci gradually isomerized
into the corresponding h2-enonenickel dimer complex (3ci)
at room temperature, as observed for 6c. The reaction of 1b
under the same reaction conditions gave an h2-enonenickel
dimer complex (3bi) quantitatively, and no corresponding
six-membered oxa-nickelacycle (6bi) was observed, which
suggests that 6bi is a transient intermediate. On the other
hand, 6ai did not undergo isomerization and was stable
enough to obtain a single crystal, suitable for X-ray crystal-
lography, by recrystallization at room temperature.
Figure 1. Monitoring the formation of 6c at 408C.
Chem. Eur. J. 2009, 15, 10083 – 10091
ꢀ 2009 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chemeurj.org
10085

S. Ogoshi et al.
The molecular structure of 6ai shows a dimeric h1-nickele-
nolate structure (Figure 2). The sum of the bond angles
around nickel along the C1, O1, O2, and C5 is 359.968.
Figure 2. Molecular structure of 6ai. Hydrogen atoms and iPr groups are
omitted for clarity.
Thus, Ni, C1, O1, O2, and C5 are on the same plane and 6ai
is a typical square-planar nickel(II) 16-electron complex.
À
À
Bond lengths C3 C4 (1.328(2) ꢁ) and C4 O1 (1.336(2) ꢁ)
are in the range of a normal carbon–carbon double-bond
length and a carbon–oxygen single-bond length, respectively.
A single crystal of 6ci was obtained by recrystallization at
À208C. The molecular structure of 6ci is an unexpected
monomeric h1-nickelenolate structure (Figure 3A). The sum
of the bond angles around nickel along the C1, O1, and C5
is 359.878. Thus, Ni, C1, O1, and C5 are on the same plane
and 6ci is “an unusual T-shaped square-planar” nickel(II)
Figure 3. A) Molecular structure of 6ci. Hydrogen atoms and iPr groups
are omitted for clarity. B) Space-filling model of 6ci represented from
the front perspective of six-membered oxa-nickelacycle (the color of
atoms: Ni green, O red, C gray).
À
14-electron complex. Bond lengths C3 C4 (1.334(3) ꢁ) and
À
C4 O1 (1.334(2) ꢁ) are also in the range of a normal
carbon–carbon double-bond length and a carbon–oxygen
single-bond length, respectively. A space-filling model of 6ci
clearly indicates that such geometry is mostly due to the
bulkiness caused by the phenyl ring on the C4 atom togeth-
er with the bulky IPr ligand, which prevents other donor
molecules from approaching the fourth coordination site of
1
the square planar (Figure 3B). The H and 13C NMR spectra
Scheme 7. Insertion of enones into an oxa-nickelacycle.
of both 6ai and 6ci in C6D6 are consistent with an h1-nickel-
enolate structure, which suggests that both complexes might
have an h1-nickelenolate structure in solution. As mentioned
above, 6ai did not isomerize into the corresponding h2-eno-
nenickel dimer complex (3ai) at room temperature, al-
though 6ci slowly underwent isomerization into 3ci. This
result might be explained by the dimeric structure of 6ai.
The dimeric structure of 6ai might be too stable to disaggre-
gate into a monomer, which is required to generate a vacant
site to undergo b-hydrogen elimination as the initial step in
isomerization to 3ai.
complex (anti-9c). The stereochemistry of both anti-8c and
anti-9c is regarded as depicted in Scheme 7 on the basis of
NOE correlations among hydrogen atoms attached to C3,
C5, and C6.
The reaction of 6ci with (E)-3-penten-2-one and (E)-1-
phenylbut-2-en-1-one gave syn-8ci and syn-9ci, respectively,
in which the diastereomeric relationship between the benzo-
yl group and the methyl group is different from that of anti-
9c (Scheme 7). Both the molecular structures of syn-8ci and
syn-9ci were determined by X-ray crystallography (Figures 4
and 5). The NOESY measurements of syn-9ci in C6D6
showed that the hydrogen atom attached to C6 has an NOE
contact with that of C4, whereas NOE correlation between
Reaction of oxa-nickelacycles with enones: At room temper-
ature, 6c reacted with (E)-3-penten-2-one to give the ex-
pected nickelacycle complex (anti-8c) (Scheme 7). This reac-
tion proceeded somewhat slowly due to its poor solubility.
Employing (E)-1-phenylbut-2-en-1-one instead of (E)-3-
penten-2-one afforded the corresponding h3-oxa-allylnickel
À
À
H C4 and H C3 is not detected at all, thereby indicating
that the stereochemistry observed in the solid state is main-
tained even in solution. On the other hand, 6ai did not react
10086
www.chemeurj.org
ꢀ 2009 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Chem. Eur. J. 2009, 15, 10083 – 10091

Six-Membered Oxa-Nickelacycles
FULL PAPER
Figure 4. Molecular structure of syn-8ci. Hydrogen atoms except for
those attached to C3, C4, and C6 atoms and iPr groups are omitted for
clarity.
Scheme 8. Prospective diastereomeric relationship between h3-oxa-allyl-
nickel complexes and reductive elimination products.
The deuterium-labeled cyclopropyl phenyl ketone
([D1]1c) was prepared according to the reported proce-
dure[18] and 90% of deuterium was incorporated regioselec-
tively at the a position of the carbonyl group as revealed by
2
1H and H NMR spectroscopy. When preparation of nickele-
Figure 5. Molecular structure of syn-9ci. Hydrogen atoms except for
those attached to C3, C4, and C6 atoms and iPr groups are omitted for
clarity.
nolates 6c and 6ci was applied by using [D1]1c instead of
1c, the reaction products [D1]6c and [D1]6ci incorporated
deuterium at the g position regioselectively (Scheme 9).
with (E)-3-penten-2-one and (E)-1-phenylbut-2-en-1-one at
room temperature. This result indicates that the nickeleno-
late complex was stabilized by its dimeric structure.
As above, the diastereoselectivity of the h3-oxa-allylnickel
complex differs from ligand to ligand. In addition, h3-oxa-al-
lylnickel complexes anti-9c (with PCy3) and syn-9ci (with
IPr) had been assumed as a key intermediate in homo-dime-
rization reactions of 1c by using Ni0/PCy3 (by our group)
and Ni0/IPr (by Montgomeryꢂs group) catalytic systems.[14a,b]
We wondered, nevertheless, why the same major product 4c
was obtained from the two catalytic reactions. Given that re-
ductive elimination of cyclopentane derivative from h3-oxa-
allylnickel complexes progressed with the retention of ste-
reochemistry, the prospective stereochemistry observed in
the reaction products can be accounted for by the processes
A–D shown in Scheme 8; paths A–D are interchanged by
equilibria that involve insertion/reinsertion of enones (iii)
and flipping of the h3-oxa-allyl moiety (ii and iv). In addi-
tion, the introduction of a deuterium atom into the nickele-
nolate complex enabled us to distinguish between two prod-
ucts along path A (from anti’) and along path C (from syn).
Scheme 9. Synthesis of [D1]6c and [D1]6ci.
Treatment of deuterio-nickelenolates with (E)-1-phenylbut-
2-en-1-one resulted in formation of the corresponding g-
deuterio-h3-oxa-allylnickel complexes. During the successive
process for preparation of labeled complexes, no degrada-
tion of deuterium ratio was observed (see the Supporting In-
formation).
When a solution of the isolated syn-[D1]9ci in benzene
was heated at 1008C for 2 h in the presence of two equiva-
lents of (E)-1-phenylbut-2-en-1-one, syn-[D1]9ci was con-
sumed completely to result in the formation of the reductive
elimination products (3-[D1]4c, [D1]4c’) in quantitative yield
Chem. Eur. J. 2009, 15, 10083 – 10091
ꢀ 2009 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chemeurj.org
10087

S. Ogoshi et al.
with excellent syn diastereoselectivity (96:4; Scheme 10).[19]
This result clearly indicates that the diastereomeric relation-
ship in syn-9ci is retained during the reductive elimination
The occurrence of carbon–carbon bond cleavage and the
dissociation of the enone was confirmed by addition of an
excess amount of trans-chalcone to a solution of anti-8c and
syn-8ci (Scheme 12). The replacement of (E)-3-pentene-2-
one with trans-chalcone to give anti-10c and syn-10ci, re-
spectively, was observed.[20]
Scheme 12. Exchange of enone.
Scheme 10. Reductive elimination of cyclopentanes from syn-[D1]9ci.
Reaction path for nickel-catalyzed [3+2] cycloaddition: The
nickel-catalyzed [3+2] cycloaddition of cyclopropyl ketones
with enones might proceed as follows (Scheme 13). The oxi-
step and hence the relationship is reflected in that of the cy-
clopentane derivative 4c. Furthermore, the observed diaste-
reoselectivity in the reductive elimination of 4c from syn-
9ci is strongly consistent with that of Ni0/IPr-catalyzed
homo-dimerization of cyclopropyl phenyl ketone.
On the other hand, reductive elimination from the PCy3-
ligated analogue anti-[D1]9c was found to progress with an
inversion of the stereochemistry (Scheme 11). The deuterio-
Scheme 13. Plausible reaction mechanism for [3+2] cycloaddition.
Scheme 11. Reductive elimination of cyclopentanes from anti-[D1]9c.
dative addition of cyclopropyl ketone to Ni0 generates an
oxa-nickelacycle (A). The insertion of an enone into A
leads to the formation of an h3-oxa-allylnickel intermediate
(C). Reductive elimination from anti-C and syn-C gives the
corresponding cyclopentane derivatives and regenerates the
nickel(0) species. Lability of coordinated enone in a key in-
termediate (B) is caused not only rapid isomerization be-
tween anti-C and syn-C but insertion/reinsertion equilibrium
of enones (paths b, c, and d). This reversible insertion/rein-
sertion process is also supported by Scheme 12. In homo-cy-
cloaddition, the isomerization of cyclopropyl ketone to the
corresponding enone (along paths f, g, and h) might occur to
supply an isomeric enone to the nickel-catalyzed [3+2] cy-
cloaddition.
h3-oxa-allylnickel complex anti-[D1]9c, which was prepared
in situ by treating [D1]6c with (E)-1-phenylbut-2-en-1-one,
underwent gradual reductive elimination at room tempera-
ture to give the same product (3-[D1]4c) as observed in
Scheme 10. The formation of [D1]4c’, the anticipated prod-
uct from anti-[D1]9c with retention of stereochemistry, was
not observed at all. This labeling experiment revealed that
rapid isomerization between syn and anti isomers (equilibri-
um (iii) in Scheme 8) occurs prior to reductive elimination,
and that the reductive elimination product is derived not
from the kinetically controlled anti isomer (along path B in
Scheme 8) but from its syn isomer (along path C in
Scheme 8).
10088
www.chemeurj.org
ꢀ 2009 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Chem. Eur. J. 2009, 15, 10083 – 10091

Six-Membered Oxa-Nickelacycles
FULL PAPER
There are notable differences in the diastereoselectivity of
the insertion of an enone into oxa-nickelacycles with differ-
ent ligands (one with PCy3 and the other with IPr). The stoi-
chiometric reaction of the oxa-nickelacycle with IPr as a
ligand with an enone gave a single isomer of the h3-oxa-al-
lylnickel species with syn stereochemistry. However, the h3-
oxa-allylnickel complex with anti stereochemistry was selec-
tively obtained under the same reaction conditions by using
PCy3 as a ligand (Scheme 7). Nevertheless, labeling experi-
ments clearly revealed that reductive elimination from
either h3-oxa-allylnickel complex, anti-[D1]9c and syn-
[D1]9ci, gives the same product (3-[D1]4c) (Scheme 10 and
11). These observations could be rationalized based on the
results mentioned above. The diastereomeric relationship in
syn-9ci is retained during the reductive elimination step. In
contrast, thermal isomerization of anti-9c with PCy3 would
take place prior to reductive elimination and, as a result, its
syn diastereoisomer predominantly undergoes reductive
elimination to give the same product as observed in that de-
rived from syn-9ci. Reversible isomerization between anti-C
and syn-C can also account for the formation of a minor
product that is derived from the h3-oxa-allylnickel inter-
mediate (anti-C). It should be mentioned that mechanisms
involving direct isomerization between major and minor
products can be excluded since neither 3ci nor 3c, observed
in situ after the homo-dimerization reactions of 1c, cata-
lyzed the isomerization between major and minor products
in the catalytic reactions (see the Supporting Information).
prior to reductive elimination, the reductive elimination
from the h3-oxa-allylnickel with PCy3 as a ligand took place
with a simultaneous inversion of the stereochemistry. These
observations provide deep insights into the nickel-catalyzed
[3+2] cycloaddition reactions of cyclopropyl ketones.
Experimental Section
General: All manipulations were conducted under a nitrogen atmosphere
using standard Schlenk or dry box techniques. 1H, 31P, and 13C NMR spec-
tra were recorded using JEOL GSX-270S, JEOL AL-400, Bruker DPX
400, and Varian UNITY INOVA600 spectrometers at room temperature
unless otherwise stated. The chemical shifts in 1H NMR spectra were re-
corded relative to Me4Si or residual protiated solvent (C6D5H (d=
7.16 ppm) or C7D7H (d=7.02 or 7.13 ppm)). The chemical shifts in the
13C NMR spectra were recorded relative to Me4Si. The chemical shifts in
the 31P NMR spectra were recorded using 85% H3PO4 as external stan-
dard. Assignment of the resonances in 1H and 13C NMR spectra was
based on 1H–1H COSY, HMQC, and HMBC experiments. Elemental
analyses were performed at the Instrumental Analysis Center, Faculty of
Engineering, Osaka University. For some compounds, accurate elemental
analysis were precluded by extreme air or thermal sensitivity and/or sys-
tematic problems with elemental analysis of organometallic compounds.
X-ray crystallographic data were collected using
RAPID Imaging Plate diffractometer.
a Rigaku RAXIS-
Isolation of 6ai: Cyclopropanecarboxaldehyde (37.1 mg, 0.53 mmol) was
added to solution of [Ni(cod)2] (111.8 mg, 0.40 mmol) and IPr
a
ACHTUNGTRENNUNG
(155.1 mg, 0.40 mmol) in toluene (3 mL). The reaction mixture was
stirred at room temperature for 4 h. The solution changed from red to
yellow. The solution was concentrated in vacuo. The residue was dis-
solved in toluene, and the solution was concentrated in vacuo again. The
residue was washed with hexane to give 6ai (203.2 mg, a yellow solid,
98%). A single crystal for X-ray diffraction analysis was prepared by re-
crystallization from toluene/hexane at À208C. 1H NMR (400 MHz,
C6D6): d=7.38 (t, J=7.2 Hz, 4H; IPr), 7.34 (d, J=7.2 Hz, 4H; IPr), 7.24
(d, J=7.2 Hz, 4H; IPr), 6.43 (s, 4H; N(CH)2N), 5.26 (d, J=6.0 Hz, 2H;
-NiOCHCH-), 4.05 (td, J=4.4 Hz, 6.0 Hz, 2H; -NiOCHCH-), 3.55 (t, J=
6.4 Hz, 4H; IPr), 2.83 (t, J=6.4 Hz, 4H; IPr), 1.89 (d, J=6.4 Hz, 12H;
IPr), 1.36 (d, J=6.4 Hz, 12H; IPr), 1.09 (d, J=6.4 Hz, 12H; IPr), 0.99 (d,
J=6.4 Hz, 12H; IPr), 0.47 (d, J=4.4 Hz, 4H; -NiCH2CH2-), À0.09 ppm
(t, J=6.0 Hz, 4H; -NiCH2CH2-); 13C NMR (100 MHz, C6D6): d=188.2
(s; NCN), 147.5 (s; -NiOCHCH-), 146.3 (s; IPr), 146.0 (s; IPr), 137.1 (s;
IPr), 129.5 (s; IPr), 124.7 (s; IPr), 124.1 (s; IPr), 123.5 (s; N(CH)2N), 99.2
(s; -NiOCHCH-), 28.7 (s; IPr), 28.6 (s; IPr), 26.2 (s; IPr), 24.5 (s; IPr),
23.1 (s; IPr), 21.2 (s; -NiCH2CH2-), 2.0 ppm (s; -NiCH2CH2-); elemental
analysis calcd (%) for C62H84N4Ni2O2: C 71.97, H 8.18, N 5.41; found: C
72.14, H 8.25, N 5.28. X-ray data for 6ai: Mr =517.39 (monomer); yellow;
monoclinic; C2/c (no. 15); a=21.712(15), b=11.938(8), c=22.976(16) ꢁ;
Conclusion
We demonstrated for the first time the formation of six-
membered oxa-nickelacycles by oxidative addition of cyclo-
propyl ketones to nickel(0) complexes. IPr is a better ligand
than PCy3 for the generation of six-membered hetero-nick-
elacycles. Molecular structures of oxa-nickelacycle with IPr
were determined by X-ray crystallography, in which dimeric
and monomeric h1-nickelenolate structures exist. In the mo-
lecular structure for monomeric complexes, the nickel(II)
14-electron center is regarded as having “an unusual T-
shaped planar” coordination geometry. The insertion of
enones into a monomeric h1-nickelenolate complex took
place at room temperature to generate h3-oxa-allylnickel
complexes, whereas the insertion into a dimeric h1-nickele-
nolate complex did not occur. We also confirmed that the
insertion of enones into oxa-nickelacycles is reversible. Dif-
fering diastereoselectivity was observed when enones were
inserted into oxa-nickelacycles with different ligands (one
with PCy3 and the other with IPr). Labeled experiments re-
vealed that the stereochemistry of IPr-ligated h3-oxa-allyl-
nickel complexes was retained during reductive elimination,
thereby giving corresponding [3+2] cycloaddition products.
This finding is strongly consistent with the diastereoselectiv-
ity observed in Ni0/IPr-catalyzed [3+2] cycloaddition reac-
tions of cyclopropyl ketones with enones. In contrast, be-
cause of rapid isomerization between syn and anti isomers
b=108.666(6)8; V=5642.1(68) ꢁ3; Z=8; 1calcd =1.218 gcmÀ3
À150.08C; R (Rw)=0.0379 (0.0955).
;
T=
Isolation of 6ci: Cyclopropyl phenyl ketone (150.0 mg, 1.03 mmol) was
added to solution of [Ni(cod)2] (277.1 mg, 1.01 mmol) and IPr
a
ACHTUNGTRENNUNG
(393.1 mg, 1.01 mmol) in cold toluene (7 mL, À208C). The reaction mix-
ture was stirred for 2 min, and the solution changed from red to purple.
The solution was concentrated in vacuo. The residue was dissolved in
cold toluene (À208C), and the solution was concentrated in vacuo again.
The residue was washed with hexane to give 6ci (587.4 mg, a purple
solid, 92%). A single crystal for X-ray diffraction analysis was prepared
by recrystallization from toluene/hexane at À208C. 1H NMR (270 MHz,
C6D6): d=7.84 (d, J=7.6 Hz, 2H; Ph), 7.03–7.29 (m, 9H; IPr, Ph), 6.29
(s, 2H; N(CH)2N), 5.14 (t, J=4.2 Hz, 1H; -NiOC(Ph)CH-), 2.79 (quintet,
J=6.8 Hz, 4H; IPr), 1.60 (d, J=6.8 Hz, 12H; IPr), 1.53 (2H;
-NiCH2CH2-, obscured by IPr peak), 1.06 (d, J=6.8 Hz, 12H; IPr),
0.61 ppm (td, J=4.2 Hz, 6.0 Hz, 2H; -NiCH2CH2-); 13C NMR (100 MHz,
[D8]toluene, À808C): d=180.2 (s; NCN), 156.6 (s; -NiOC(Ph)CH-), 145.1
(s; IPr), 141.0 (s; Ph), 137.3 (s; IPr), 130.1 (s; IPr), 127.1 (s; Ph), 125.0 (s;
Chem. Eur. J. 2009, 15, 10083 – 10091
ꢀ 2009 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chemeurj.org
10089

S. Ogoshi et al.
Ph), 124.7 (s; Ph), 124.0 (s; IPr), 123.0 (s; N(CH)2N), 93.2 (s;
-NiOC(Ph)CH-), 28.6 (s; IPr), 25.6 (s; -NiCH2CH2-), 24.1 (s; IPr),
8.5 ppm (s; -NiCH2CH2-); elemental analysis calcd (%) for C37H46N2NiO:
C 74.88, H 7.81, N 4.72; found: C 74.58, H 7.81, N 4.65. X-ray data for
CCDC 726581 (6ai), 726582 (6ci·0.5
A
ACHTUNGTRENNUNG
supplementary 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.
6ci·0.5
ACHTUNGTRENNUNG(C6H14): Mr =636.57; purple; monoclinic; C2/c (no. 15); a=
22.586(5),
b=14.833(3),
c=23.912(5) ꢁ;
b=116.078(3)8;
V=
7195.5(24) ꢁ3; Z=8; 1calcd =1.175 gcmÀ3; T=À150.08C; R (Rw)=0.0519
(0.1109).
Generation of [Ni{m-h1,h3-CH2CH2CH
ACHUTGTNERN(NGU COPh)CHACHUTNGTREN(NUNG CH3)CH=C(Ph)O}-
Acknowledgements
ACHTUNGTRENNUNG
added to a solution of 6c (0.01 mmol, 4.9 mg) in C6D6 (0.5 mL) at room
temperature. The reaction was followed by 1H and 31P NMR spectra.
After 7 h, the solution changed from orange to deep orange, and the
orange precipitation disappeared. Compound anti-9c was generated
quantitatively. 1H NMR (270 MHz, C6D6): d=7.03–7.99 (m, 10H; Ph),
This work was supported by Grants-in-Aid for Scientific Research (nos.
21245028 and 21750102) and a Grant-in-Aid for Scientific Research on
Priority Areas (no. 19028038, Chemistry of Concerto Catalysis) from the
Ministry of Education, Culture, Sports, Science, and Technology, Japan.
T.T. expresses his special thanks for the Global COE Program “Global
Education and Research Center for Bio-Environmental Chemistry” of
Osaka University.
5.51 (dd, J=9.9, 1.3 Hz, 1H; -CH=
N
ACHTUNGTRENNUNG
35H; including 2H of -NiCH2CH2- at d=1.00 and 1.57 ppm), 0.42–0.54
(m, 1H; -NiCH2CH2-), 0.22–0.33 ppm (m, 1H; -NiCH2CH2-); 31P NMR
(109 MHz, C6D6): d=34.03 ppm (s).
[1] a) S. Derien, E. Dunnach, J. Perichon, J. Am. Chem. Soc. 1991, 113,
8447–8454; b) H. Hoberg, D. Schaefer, G. Burkhart, C. Krueger,
M. J. Romaao, J. Organomet. Chem. 1984, 266, 203–224; c) D.
Walther, G. Braeunlich, R. Kempe, J. Sieler, G. Braeunlich, J. Orga-
nomet. Chem. 1992, 436, 109–120; d) H. Hoberg, B. W. Oster, J. Or-
ganomet. Chem. 1982, 234, C35–C38; e) H. Hoberg, K. Suemmer-
mann, A. Milchereit, Angew. Chem. 1985, 97, 321–321; f) E. Her-
nandez, H. Hoberg, J. Organomet. Chem. 1986, 315, 245–253.
[2] a) S. Ogoshi, M. Oka, H. Kurosawa, J. Am. Chem. Soc. 2004, 126,
11802–11803; b) S. Ogoshi, M. Ueta, T. Arai, H. Kurosawa, J. Am.
Chem. Soc. 2005, 127, 12810–12811; c) S. Ogoshi, K. Tonomori, M.
Oka, H. Kurosawa, J. Am. Chem. Soc. 2006, 128, 7077–7086; d) S.
Ogoshi, H. Kamada, H. Kurosawa, Tetrahedron 2006, 62, 7583–
7588; e) S. Ogoshi, H. Ikeda, H. Kurosawa, Angew. Chem. 2007,
119, 5018–5020; Angew. Chem. Int. Ed. 2007, 46, 4930–4932; f) S.
Ogoshi, T. Arai, H. Kurosawa, Chem. Commun. 2008, 1347–1349.
[3] a) E. Oblinger, J. Montgomery, J. Am. Chem. Soc. 1997, 119, 9065–
9066; b) M. Kimura, A. Ezoe, K. Shibata, Y. Tamaru, J. Am. Chem.
Soc. 1998, 120, 4033–4034; c) M. Takimoto, Y. Hiraga, Y. Sato, M.
Mori, Tetrahedron Lett. 1998, 39, 4543–4546; d) M. Kimura, H. Fuji-
matsu, A. Ezoe, K. Shibata, M. Shimizu, S. Matsumoto, Y. Tamaru,
Angew. Chem. 1999, 111, 410–413; Angew. Chem. Int. Ed. 1999, 38,
397–400; e) G. M. Mahandru, G. Liu, J. Montgomery, J. Am. Chem.
Soc. 2004, 126, 3698–3699; f) K. M. Miller, W.-S. Huang, T. F. Jami-
son, J. Am. Chem. Soc. 2003, 125, 3442–3443; g) J. Montgomery,
Angew. Chem. 2004, 116, 3980–3998; Angew. Chem. Int. Ed. 2004,
43, 3890–3908, and references therein.
Isolation of [Ni{m-h1,h3-CH2CH2CH
(syn-9ci): Cyclopropyl phenyl ketone (65.0 mg, 0.44 mmol) was added to
solution of [Ni(cod)2] (111.1 mg, 0.40 mmol) and IPr (162.3 mg,
ACHTUNGRTEN(UNNG COPh)CHACHTUNRTGEN(NGNU CH3)CH=C(Ph)O}ACHTUNGTREN(NUGN IPr)]
a
ACHTUNGTRENNUNG
0.42 mmol) in toluene (6 mL). The reaction mixture was stirred at ambi-
ent temperature for 5 min. Then (E)-1-phenyl-2-buten-1-one (60.6 mg,
0.41 mmol) was added at room temperature. The solution changed from
purple to dark brown. The reaction mixture was stirred for 20 min, fol-
lowed by concentration in vacuo. The residue was dissolved in toluene,
and the solution was concentrated in vacuo again. The residue was
washed with hexane to give syn-9ci (252.1 mg, a yellow solid) in 85%
yield. 1H NMR (400 MHz, C6D6): d=7.38–7.51 (d, J=7.6 Hz, 6H; Ph,
IPr), 6.95–7.19 (m, 10H; Ph, IPr), 6.60 (s, 2H; N(CH)2N), 4.73 (d, J=
9.6 Hz, 1H; -CH=C(Ph)ONi-), 3.46 (t, J=6,4 Hz, 2H; IPr), 2.93 (m, 1H;
-CH
N
ACHTUNGTRENNUNG
-CH
N
ACHTUNGTRENNUNG
3H; -CH
N
ACHTUNGTRENNUNG
7.2 Hz, 6H; IPr), 1.07 (d, J=7.2 Hz, 6H; IPr), 1.05 (d, J=7.2 Hz, 6H;
IPr), 0.60 (t, J=3.8 Hz, 12.0 Hz, 1H; -NiCH2CH2-), 0.21 (m, 1H;
-NiCH2CH2-), À0.16 ppm (td, J=12.0 Hz 1H; -NiCH2CH2-); 13C NMR
(100 MHz, C6D6): d=201.1 (s; -CHACHTUNTRGENN(UG COPh)CHCAHTUNGTREN(NUGN CH3)-), 192.2 (s; NCN),
155.1 (s; Ph), 146.8 (s; IPr), 146.7 (s; IPr), 141.0 (s; Ph), 138.5 (s; Ph),
136.9 (s; IPr), 131.4 (s; Ph), 130.0 (s; IPr), 128.7 (s; Ph), 128.5 (s; Ph),
128.1 (s; Ph), 127.9 (s; Ph), 124.4 (s; IPr), 124.2 (s; Ph), 75.0 (s; -CH=
C(Ph)ONi-), 54.2 (s; -CH
(CH3)-), 29.4 (s; -NiCH2CH2-), 29.0 (s; IPr), 28.9 (s; IPr), 26.4 (s; IPr),
25.5 (s; IPr), 23.2 (s; IPr), 22.7 (s; IPr), 20.1 (s; -CH(COPh)CH(CH3)-),
ACHTUNGRTEN(UNGN COPh)CHACHUTNRTGENG(UNN CH3)-), 33.5 (s; -CHACHTUNGTREN(NNGU COPh)CH-
ACHTUNGTRENNUNG
A
ACHTUNGTRENNUNG
[4] a) Y. Inoue, Y. Itoh, H. Kazama, H. Hashimoto, Bull. Chem. Soc.
Jpn. 1980, 53, 3329–3333; b) T. Tsuda, S. Morikawa, R. Sumiya, T.
Saegusa, J. Org. Chem. 1988, 53, 3140–3145; c) T. Tsuda, S. Morika-
wa, N. Hasegawa, T. Saegusa, J. Org. Chem. 1990, 55, 2978–2981;
d) J. Louie, J. E. Gibby, M. V. Farnworth, T. N. Tekavec, J. Am.
Chem. Soc. 2002, 124, 15188–15189; e) H. Hoberg, B. W. Oster, Syn-
thesis 1982, 324–325; f) R. A. Earl, K. P. C. Vollhardt, J. Org. Chem.
1984, 49, 4786–4800; g) H. A. Duong, M. J. Cross, J. Louie, J. Am.
Chem. Soc. 2004, 126, 11438–11439.
[5] a) T. Tsuda, T. Kiyoi, T. Miyane, T. Saegusa, J. Am. Chem. Soc.
1988, 110, 8570–8572; b) T. N. Tekevac, J. Louie, Org. Lett. 2005, 7,
4037–4039; c) M. Murakami, S. Ashida, T. Matsuda, J. Am. Chem.
Soc. 2006, 128, 2166–2167.
[6] a) K. K. D. Amarasinghe, S. K. Chowdhury, M. J. Heeg, J. Montgom-
ery, Organometallics 2001, 20, 370–371; b) H. P. Hratchian, S. K.
Chowdhury, V. M. Gutiꢃrrez-Garcꢄa, K. K. D. Amarasinghe, M. J.
Heeg, H. B. Schlegel, J. Montgomery, Organometallics 2004, 23,
4636–4646; c) J. Cꢅmpora, C. M. Maya, P. Palma, E. Carmona, C.
Graiff, A. Tiripicchio, Chem. Commun. 2003, 1742–1743.
[7] a) M. Murakami, Y. Ito, Top Organomet Chem 1999, 3, 97–129, and
references therein; b) M. Rubin, M. Rubina, V. Gevorgyan, Chem.
Rev. 2007, 107, 3117–3179, and references therein; c) R. A. Periana,
0.98 ppm (s; -NiCH2CH2-); elemental analysis calcd (%) for
C47H56N2NiO2: C 76.02, H 7.63, N 3.79; found: C 76.32, H 7.67, N 3.83.
X-ray data for syn-9ci: Mr =739.67; yellow; monoclinic; P21/a (no. 14);
a=14.9301(5), b=15.8322(6), c=18.3235(7) ꢁ; b=111.6129(11)8; V=
4026.7(3) ꢁ3; Z=4; 1calcd =1.220 gcmÀ3; T=À150.08C; R (Rw)=0.0956
(0.3224).
Reaction of syn-[D1]9ci with (E)-1-phenyl-2-buten-1-one: (E)-1-Phenyl-
2-buten-1-one (4.2 mg, 0.028 mmol) was added to a solution of syn-
[D1]9ci (10.6 mg, 0.014 mmol) in C6H6 (0.5 mL). The reaction mixture
was heated at 1008C for 2 h. The solution changed from yellow to brown.
All of the starting material were converted to generate a mixture of 3-
[D1]4c and [D1]4c’. NMR spectroscopic analysis of 3-[D1]4c revealed a
selective incorporation of deuterium at the 3-position.
Reaction of anti-[D1]9c with (E)-1-phenyl-2-buten-1-one: (E)-1-Phenyl-
2-buten-1-one (15.7 mg, 0.11 mmol) was added to a solution of anti-
[D1]9c (0.01 mmol, generated in situ from the procedure mentioned
above) in C6H6 (0.5 mL). The reaction mixture was monitored at room
temperature for 7 d. After 7 d, the solution changed from yellow to
brown. The complex anti-[D1]9c was completely consumed to generate 3-
[D1]4c quantitatively. NMR spectroscopic analysis of 3-[D1]4c revealed a
selective incorporation of deuterium at the 3-position.
10090
www.chemeurj.org
ꢀ 2009 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Chem. Eur. J. 2009, 15, 10083 – 10091

Six-Membered Oxa-Nickelacycles
FULL PAPER
R. G. Bergman, J. Am. Chem. Soc. 1984, 106, 7272–7273; d) D. D.
Wick, T. O. Northcutt, R. J. Lachicotte, W. D. Jones, Organometallics
1996, 15, 4484–4492.
[13] Allenylcyclopropane: M. Murakami, K. Itami, M. Ubukata, I. Tsuji,
Y. Ito, J. Org. Chem. 1998, 63, 4–5.
[14] a) S. Ogoshi, M. Nagata, H. Kurosawa, J. Am. Chem. Soc. 2006, 128,
5350–5351; b) L. Liu, J. Montgomery, J. Am. Chem. Soc. 2006, 128,
5348–5349; c) L. Liu, J. Montgomery, Org. Lett. 2007, 9, 3885–3887;
d) G. C. Lloyd-Jones, Angew. Chem. 2006, 118, 6942–6944; Angew.
Chem. Int. Ed. 2006, 45, 6788–6790.
[8] a) S. C. Bart, P. J. Chirik, J. Am. Chem. Soc. 2003, 125, 886–887;
b) T. H. Tulip, J. A. Ibers, J. Am. Chem. Soc. 1979, 101, 4201–4211;
c) J. Tsuji, M. Morikawa, J. Kiji, Tetrahedron Lett. 1965, 6, 817–819.
[9] Methylenecyclopropane: a) R. Noyori, T. Odagi, H. Takaya, J. Am.
Chem. Soc. 1970, 92, 5780–5781; b) S. A. Baquji, W. B. Motherwell,
M. Shipman, Tetrahedron Lett. 1989, 30, 7107–7110; c) M. Lautens,
Y. Ren, P. H. M. Delanghe, J. Am. Chem. Soc. 1994, 116, 8821–
8822; d) M. Gulꢄas, R. Garcꢄa, A. Delgado, L. Castedo, J. L. Mascar-
eÇas, J. Am. Chem. Soc. 2006, 128, 384–385; e) S. Saito, M. Masuda,
S. Komagawa, J. Am. Chem. Soc. 2004, 126, 10540–10541; f) P.
Binger, M. J. Doyle, R. Benn, Chem. Ber. 1983, 116, 1–10.
[10] Vinyl cyclopropane: a) M. Suginome, T. Matsuda, T. Yoshimoto, Y.
Ito, Organometallics 2002, 21, 1537–1539; b) G. Zuo, J. Louie, J.
Am. Chem. Soc. 2005, 127, 5798–5799; c) P. A. Wender, H. Takaha-
shi, B. Witulski, J. Am. Chem. Soc. 1995, 117, 4720–4721; d) Z. Yu,
P. A. Wender, N. K. Houk, J. Am. Chem. Soc. 2004, 126, 9154–9155.
[15] A portion of this work has been communicated previously
(ref. [14a]).
[16] The treatment of nickel compounds with carbon monoxide can yield
[Ni(CO)4] (extremely toxic) due to the addition of insufficient
amounts of PR3 or IPr, careless handling, or an accident. The reac-
tion mixture must be handled in a well-ventilated fume hood.
[17] a) M. Jimꢃnez-Tenorio, M. C. Puerta, P. Valerga, F. J. Moreno-
Dorado, F. M. Guerra, G. M. Massanet, Chem. Commun. 2001,
2324–2325; b) M. Salisovꢅ, S. Toma, E. Solcaniova, J. Organomet.
Chem. 1987, 327, 77–84.
[18] G. R. Boggs, J. T. Gerig, J. Org. Chem. 1968, 33, 1484–1486.
[19] When this reaction was conducted in the absence of free enone, deg-
radation of diastereoselectivity and scrambling of the deuterium
atom were observed.
[11] Cyclopropyl ketone:
A six-membered oxa-nickelacycle seems a
likely intermediate, although authors did not mention it. T. Ichiyana-
gi, S. Kuniyama, M. Shimizu, T. Fujisawa, Chem. Lett. 1997, 1149–
1150.
[20] The complex syn-10ai was alternatively synthesized by the reaction
of 6ci with trans-chalcone, and its molecular structure was con-
firmed by X-ray diffraction study (see also the Supporting Informa-
tion).
[12] Cyclopropyl imine: a) P. A. Wender, T. M. Pedersen, M. J. C. Scanio,
J. Am. Chem. Soc. 2002, 124, 15154–15155; b) A. Kamitani, N. Cha-
tani, T. Morimoto, S. Murai, J. Org. Chem. 2000, 65, 9230–9233.
Received: April 8, 2009
Published online: August 28, 2009
Chem. Eur. J. 2009, 15, 10083 – 10091
ꢀ 2009 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chemeurj.org
10091