Mechanistic insights into nickel-catalyzed cycloisomerizations

Article abstract of DOI:10.1021/ol101852w

A nickel-catalyzed cycloisomerization coupling an allylic alcohol with an alkyne has been developed. Mechanistic insights gained through deuterium-labeling crossover studies and stereochemical probes illustrate that oxidation of the allylic alcohol to a metal-free enone is not involved in the cyclization pathway. The intermediacy of a metallacycle directly derived from oxidative cyclization of Ni(0) with the allylic alcohol and alkyne is consistent with the results obtained. The simple mechanistic probes employed could be useful in the study of many classes of cycloisomerization processes.

Full text of DOI:10.1021/ol101852w

ORGANIC
LETTERS
2010
Vol. 12, No. 20
4556-4559
Mechanistic Insights into
Nickel-Catalyzed Cycloisomerizations
John H. Phillips and John Montgomery*
930 North UniVersity AVenue, Department of Chemistry, UniVersity of Michigan,
Ann Arbor, Michigan 48109-1055, United States
jmontg@umich.edu
Received August 7, 2010
ABSTRACT
A nickel-catalyzed cycloisomerization coupling an allylic alcohol with an alkyne has been developed. Mechanistic insights gained through
deuterium-labeling crossover studies and stereochemical probes illustrate that oxidation of the allylic alcohol to a metal-free enone is not
involved in the cyclization pathway. The intermediacy of a metallacycle directly derived from oxidative cyclization of Ni(0) with the allylic
alcohol and alkyne is consistent with the results obtained. The simple mechanistic probes employed could be useful in the study of many
classes of cycloisomerization processes.
Transition metal catalyzed couplings involving the union
of two different π-components have been developed in
Scheme 1
.
C-C Bond Formations Mediated by Redox
many contexts.1 In some cases, a completely atom-
economical rearrangement of starting materials leads to
products.1a,b In other cases, reducing agents are employed,
leading to a net two-electron reduction during the coupling
event.1c-i The substrate combinations that efficiently par-
ticipate in redox-neutral rearrangements often differ structur-
ally from the combinations that participate in reductive
coupling processes. Coupling reactions that normally require
external reducing agents can potentially be made more atom
economical and cost-effective by developing strategies where
the redox partner is an integral component of the starting
material. Recent work from our laboratory has illustrated that
nickel-catalyzed enal-alkyne-alcohol three-component cou-
plings possess such a feature.2 As illustrated (Scheme 1),
during the C-C bond-forming event producing compound
Exchange
1, reduction of the alkyne to an alkene simultaneously with
oxidation of an aldehyde to an ester removes the need for
an external reducing agent commonly utilized in processes
of this type. The extensive body of work from Krische
involving transfer hydrogenation-mediated couplings shares
this characteristic as well.3 As examples in the formation of
2 and 3 illustrate, during the C-C bond forming process,
(1) (a) Trost, B. M.; Frederiksen, M. U.; Rudd, M. T. Angew. Chem.,
Int. Ed. 2005, 44, 6630–6666. (b) Trost, B. M.; Krische, M. J. Synlett 1998,
1–16. (c) Montgomery, J. Acc. Chem. Res. 2000, 33, 467–473. (d)
Montgomery, J. Angew. Chem., Int. Ed. 2004, 43, 3890–3908. (e)
Montgomery, J.; Sormunen, G. J. Top. Curr. Chem. 2007, 279, 1-23. (f)
Moslin, R. M.; Miller-Moslin, K.; Jamison, T. F. Chem. Commun. 2007,
4441–4449. (g) Ikeda, S. Angew. Chem., Int. Ed. 2003, 42, 5120–5122. (h)
Skukcas, E.; Ngai, M.-Y.; Komanduri, V.; Krische, M. J. Acc. Chem. Res.
2007, 40, 1394–1401. (i) Jeganmohan, M.; Cheng, C. H. Chem.sEur. J.
2008, 14, 10876–10886.
(2) (a) Herath, A.; Li, W.; Montgomery, J. J. Am. Chem. Soc. 2008,
130, 469–471. (b) Li, W.; Herath, A.; Montgomery, J. J. Am. Chem. Soc.
2009, 131, 17024–17029.
10.1021/ol101852w  2010 American Chemical Society
Published on Web 09/15/2010

either an alkyne or an allene may be reduced to an isolated
alkene simultaneously with the oxidation of the starting
alcohol to a carbonyl component that is captured in the bond-
forming process. Processes of this type stand to improve the
efficiency of the underlying transformations that otherwise
require external reductants.
of substrate 6 illustrated that ligand choice and base
additive were both essential elements for promoting
efficient cyclizations to produce 7 (Table 1). All cycliza-
The union of allylic alcohols with alkynes is one of the
few substrate combinations that undergo efficient cou-
plings in both reductive and nonreductive manifolds.
Developments from Trost and Dixneuf illustrated that
atom-economical couplings of allylic alcohols with alkynes
employing ruthenium catalysis proceed to afford substi-
tuted ketone products 4,4,5 whereas Micalizio6 and Cha7
independently reported that titanium-mediated couplings of
similar substrates proceed with net reduction and elimination
to provide 1,4-diene products 5 (Scheme 2).8 To better
Table 1. Optimization of Allylic Alcohol-Alkyne Coupling
temp addition
entry ligand
base
none
t-BuOK PhCH3
t-BuOK PhCH3
t-BuOK PhCH3
t-BuOK PhCH3
solvent
PhCH3
(°C)
timea
yield
1
2
3
4
5
6
7
8
9
10
none
none
IPrb
PBu3
PCy3
PCy3
PCy3
PCy3
PCy3
PCy3
90
90
90
90
90
60
90
60
rt
3 h
3 h
3 h
3 h
3 h
3 h
30 min
30 min
30 min
none
4
20
18
7
35
17
58
71
45
17
Scheme 2. Couplings of Allylic Alcohols and Alkynes
none
PhCH3
t-BuOK PhCH3
t-BuOK PhCH3
t-BuOK PhCH3
t-BuOK PhCH3
60
a Addition time refers to the ynal addition time by syringe drive. b 10
mol % ligand was used.
tions were conducted with 10 mol % Ni(COD)2 as the
precatalyst. Cyclizations in the absence of added ligands
were inefficient (entries 1 and 2). The use of the
N-heterocyclic carbene 1,3-bis(2,6-diisopropylphenyl) im-
idazol-2-ylidene (IPr) was inefficient, as was PBu3 (entries
3 and 4). Modestly improved yields were seen with PCy3
in the presence of KO-t-Bu, and yields with PCy3 in the
absence of this base additive were lower (entries 5 and
6). Using the combination of PCy3 and KO-t-Bu with 30
min syringe drive addition of the ynal, comparisons of
reactions conducted at various temperatures illustrated that
reactions conducted at 60 °C provided an optimized yield
of 71% (entries 7-9). Yields without syringe drive
addition of the ynal were markedly lower since competing
pathways such as alkyne trimerization were minimized
with slow addition (entry 10). On the basis of these
experiments, the optimized conditions from entry 8 were
examined with different substrates.
understand the potential for developing new classes of metal-
catalyzed coupling events that proceed by redox-mediated
pathways without the use of an external reductant, we have
investigated couplings of allylic alcohols with alkynes and
employed crossover studies and stereochemical probes to
better define how the redox-neutral pathways proceed. The
insights gained from this study illustrate mechanistic features
that may guide further reaction discovery efforts.
Our mechanistically focused efforts have primarily
examined intramolecular variants of allylic alcohol-alkyne
couplings. Initial explorations of the cycloisomerization
(3) For a review, see: (a) Bower, J. F.; Kim, I. S.; Patman, R. L.; Krische,
M. J. Angew. Chem., Int. Ed. 2009, 48, 34–46. (b) For alkynes: Patman,
R. L.; Chaulagain, M. R.; Williams, V. M.; Krische, M. J. J. Am. Chem.
Soc. 2009, 131, 2066–2067. (c) For allenes: Bower, J.; Skucas, E.; Patman,
R. L.; Krische, M. J. J. Am. Chem. Soc. 2007, 129, 15134–15135
.
(4) (a) Trost, B. M.; Martinez, J. A.; Kulawiec, R. J.; Indolese, A. F.
J. Am. Chem. Soc. 1993, 115, 10402–10403. (b) Trost, B. M.; Indolese,
A. F.; Muller, T. J. J.; Treptow, B. J. Am. Chem. Soc. 1995, 117, 615–623.
(c) Trost, B. M.; Surivet, J. P.; Toste, F. D. J. Am. Chem. Soc. 2001, 123,
2897–2898. (d) De´rien, S.; Jan, D.; Dixneuf, P. H. Tetrahedron 1996, 52,
Using the optimized protocol, cyclizations bearing either
aromatic or silyl functionality on the alkyne were found
to proceed efficiently (Table 2, entries 1-3). However,
terminal alkynes were poor substrates for the cyclization
(entry 4). The allylic alcohol functionality tolerated either
aromatic (entries 1-3) or aliphatic (entry 5) substitution
on the secondary hydroxyl. Primary allylic alcohols,
however, failed to participate in the cycloisomerizations
(entry 6). A six-membered ring cyclization also proceeded
in moderate yield with a phenyl-substituted alkyne (entry
7). In addition to the above examples involving cyclization, a
5511–5524
(5) For a recent synthetic application, see: Nicolaou, K. C.; Li, A.; Ellery,
S. P.; Edmonds, D. J. Angew. Chem., Int. Ed. 2009, 48, 6293–6295
(6) Kolundzic, F.; Micalizio, G. C. J. Am. Chem. Soc. 2007, 129, 15112–
.
.
15113.
(7) Lysendo, I. L.; Kim, K.; Lee, H. G.; Cha, J. K. J. Am. Chem. Soc.
2008, 130, 15997–16002.
(8) For related reductive cyclizations of simple enynes, see: (a) Chen,
M.; Weng, Y.; Guo, M.; Zhang, H.; Lei, A. W. Angew. Chem., Int. Ed.
2008, 47, 2279–2282. (b) Chai, Z.; Wang, H. F.; Zhao, G. Synlett 2009,
1785–1790. (c) Trost, B. M.; Rise, F. J. Am. Chem. Soc. 1987, 109, 3161–
3163. (d) Jang, H. Y.; Krische, M. J. J. Am. Chem. Soc. 2004, 126, 7875–
7880. (e) Jang, H. Y.; Hughes, F. W.; Gong, H. G.; Zhang, J. M.; Brodbelt,
J. S.; Krische, M. J. J. Am. Chem. Soc. 2005, 127, 6174–6175.
Org. Lett., Vol. 12, No. 20, 2010
4557

development of ruthenium(II)-catalyzed processes.4 An enone
intermediate could not be detected during the nickel-
catalyzed processes, and injection of an enone to the reaction
mixture inhibits product formation. Although these observa-
tions illustrate that appreciable buildup of enone was not
occurring, the possibility remained that enones were rapidly
formed and consumed during the reaction pathway.
Table 2. Scope of Allylic Alcohol-Alkyne Couplingsa
entry
R1
Ph
R2
n
yield %a
To more completely examine the potential involvement
of an enone intermediate, a crossover experiment utilizing
substrates 12 and 13 was examined (Scheme 4). The
1
2
3
4
5
6
7
Ph
1
1
1
1
1
1
2
71
75
82
7
55
0
Ph
Ph
Ph
p-anisole
TMS
H
n-butyl
H
Ph
Ph
Scheme 4. Crossover Experiments
Ph
Ph
46
a Reaction conditions: allylic alcohol (1.0 equiv), Ni(COD)2 (0.1 equiv),
PCy3 (0.2 equiv), tBuOK (0.1 equiv), toluene (0.1 M), 60 °C. The allylic
alcohol was added via syringe drive over a 30 min period.
single example of an intermolecular coupling proceeded in
modest yield (eq 1).
metallacycle mechanism involves a unimolecular hydrogen
atom transfer process (via ꢀ-hydride elimination and reduc-
tive elimination), and therefore no crossover would be
anticipated for this pathway. In contrast, any pathway that
involves an enone intermediate may be subject to crossover
in this experiment if dissociation of the enone from the
catalyst during the sequence occurs (i.e., at the stage of
intermediate 11). Additionally, nickel hydride-mediated
pathways such as those involved in hydrovinylation processes
with electrophilic Ni(II) species would also lead to crossover
in experiments of this type.10,11 In the reaction involving 12
and 13, no crossover was observed, which is consistent with
either involvement of metallacycle 10 or with a mechanism
involving enone formation where enone dissociation from
the catalyst does not occur.
With the above preliminary data in hand, we were
interested in evaluating mechanistic features of the process
to elucidate the features that would allow the rational
development of new processes that involve redox interchange
of reactive components during a C-C bond-forming event.
A number of possible mechanistic pathways may be envi-
sioned for the above catalytic transformations. One possibility
involves initial alkoxide-assisted formation of metallacycle
10 from oxidative cyclization of Ni(0) with the allylic
alcohol-alkyne substrate (Scheme 3). A second possibility
Another mechanistic probe in evaluating the potential
involvement of enones is the transfer of chirality from
starting materials to products (eq 2). Using enantiopure
substrate 14, if enone formation occurs, then racemic product
15 would be anticipated unless both the enone and alkyne
remain coordinated to nickel without dissociation of the
enone unit. Alternatively, achiral intermediates are not
involved in the mechanism involving metallacycle 10, and
some level of chirality transfer would be expected. Upon
conducting this experiment, enantiopure substrate 14 was
Scheme 3. Mechanistic Pathways for Coupling
(9) Cuperly, D.; Petrignet, J.; Crevisy, C.; Gree, R. Chem.sEur. J. 2006,
12, 3261–3274.
(10) For leading references to hydrovinylation processes: (a) RajanBabu,
T. V. Synlett 2009, 85, 3–885. (b) RajanBabu, T. V.; Nomura, N.; Jin, J.;
Radetich, B.; Park, H. S.; Nandi, M. Chem.sEur. J. 1999, 5, 1963–1968.
(11) A recent report from Louie illustrates that nickel hydride-mediated
cycloisomerizations may involve ortho-metallation of an NHC ligand. As
our work involves phosphine complexes, nickel hydride generation by this
mechanism is unlikely in the present manuscript. Tekavec, T. N.; Louie, J.
Tetrahedron 2008, 64, 6870–6875.
involves oxidation of the allylic alcohol to an enone, followed
by rapid cyclization by a process such as that depicted for
11 to 9. This latter sequence could most likely be initiated
by an adventitiously generated Ni(II) species.9 These path-
ways are analogous to pathways proposed by Trost in the
4558
Org. Lett., Vol. 12, No. 20, 2010

converted to product 15 with 30% ee,12 and recovered
starting material in an experiment run to partial completion
was recovered in enantiopure form. Therefore, while the
degree of chirality transfer is insufficient for synthetic utility,
the involvement of a mechanism involving free enone or
enone solely bound to nickel through the alkyne may be ruled
out.
occur. In addition to crossover and chirality transfer experi-
ments that support the metallacycle-based pathway, use of
a Ni(0) precatalyst is most consistent with metallacycle-based
pathways in comparison to related Ni(0)-catalyzed transfor-
mations previously reported.1c-g The mechanism of coupling
of allylic alcohols and alkynes may differ according to the
catalyst employed.
In summary, a base-promoted, nickel-catalyzed cycloi-
somerization of allylic alcohol-alkyne substrates has been
developed. On the basis of crossover experiments and
stereochemical probes, a mechanism involving metallacycle
formation seems most likely. Further examination of other
processes that take advantage of the unique mechanistic
features of the nickel-catalyzed system is in progress.
On the basis of these experiments, the formation of
metallacycle 10 as the operative mechanistic pathway seems
most likely for the nickel-catalyzed process. Notably, the
methyl ether or silyl ether of the starting allylic alcohol fails
to undergo cyclization under the reaction conditions devel-
oped, lending support to the notion that alkoxide coordination
as depicted in structure 10 may be important for catalysis to
Acknowledgment. The authors acknowledge receipt of
NSF grant CHE-1012270 in support of this work. Ananda
Herath (Burnham Institute) is thanked for helpful suggestions.
Supporting Information Available: Full experimental
details and copies of NMR spectral data. This material is
available free of charge via the Internet at http://pubs.acs.org.
(12) The absolute configuration of the enantioenriched ketone was not
determined.
OL101852W
Org. Lett., Vol. 12, No. 20, 2010
4559