Published on Web 02/13/2002
A New Two-Step Four-Component Synthesis of Highly Functionalized
Cyclohexenols by Sequential Nickel-Catalyzed Couplings
Mario Lozanov and John Montgomery*
Department of Chemistry, Wayne State UniVersity, Detroit, Michigan 48202-3489
Received November 21, 2001; Revised Manuscript Received January 21, 2002
The development of new multicomponent coupling processes has
attracted intense interest in recent years.1 Such processes allow the
efficient construction of complex molecules from simple precursors
in a minimum number of steps, and many are ideally suited for the
generation of structurally diverse libraries of small molecules. The
development of tandem reactions that utilize the same catalyst for
more than one step has recently been recognized as an efficient
strategy for effecting multicomponent coupling processes.2 Our
group has developed several classes of nickel-catalyzed reactions
that involve either enones or aldehydes as one of the reactive
components.3 Therefore, we envisioned that enals could be par-
ticularly attractive as substrates in multicomponent reactions since
initial nickel-catalyzed functionalization of the â-carbon of an enal
produces an aldehyde functionality, which could itself then be
further functionalized in a second nickel-catalyzed reaction. Imple-
mentation of this strategy has allowed the development of a new
two-step, four-component coupling of enals, alkynes, acetylenic tin
reagents, and either organozincs or organoboranes to produce highly
functionalized cyclohexenols (eq 1).
tuted (entries 1-2, 5) or substituted at the R-(entry 3) or â-(entries
4, 6-7) positions. The alkyne may be unsubstituted (entries 2-3,
5-6) or substituted with simple (entries 1,7) or functionalized (entry
4) groups. Finally, the acetylenic tin reagent may be substituted
with aromatic (entries 1-3, 5-6) or aliphatic (entries 4, 7) groups.
The combination of these variations suggests that quite a broad
range of substituted enynals should be available by this coupling
process.
A critical feature of the three-component coupling process is that
the aldehyde and alkyne functional groups are introduced in a cis
orientation, which is required for the second nickel-catalyzed
coupling event. Accordingly, treatment of product 1 from the initial
nickel-catalyzed coupling with catalytic Ni(COD)2 and an orga-
nozinc afforded good yields of cyclohexenol 2, which is thus
derived from the coupling of the aldehyde, alkyne, and organozinc
components. Only simple organozincs (Me, entries 3, 5, 6 and Et,
entry 2) were examined, although we anticipate that more complex
organozincs may be incorporated. Prior investigations from our
laboratory demonstrated that ligand structure could be varied to
promote either ethyl group incorporation or hydrogen atom
incorporation during ynal cyclizations with diethylzinc.3 However,
the presence of the internal double bond in substrate 1 completely
suppresses the hydrogen incorporation pathway, irrespective of
ligand structure.
Since hydrogen atom substitution would be advantageous in some
applications, we screened other reducing agents and found that the
use of triethylborane leads to clean hydrogen atom incorporation
if tributylphosphine is employed as a ligand. Triethylborane has
been successfully used as a reducing agent by Tamaru and Kimura5
and Jamison6 in related nickel-catalyzed couplings. The completely
selective incorporation of an ethyl group from diethylzinc and a
hydrogen atom from triethylborane under otherwise identical
conditions was fortunate, albeit surprising. Cyclization of substrate
1 under these modified conditions with triethylborane and tribu-
tylphosphine thus led cleanly to hydrogen atom introduction in each
case examined (entries 1, 4, 7).7 The seemingly similar reactions
that employ either triethylborane or diethylzinc may indeed proceed
by different mechanisms.8
A partially intramolecular variant of the process was next
examined utilizing substrate 3 (eq 2). Treatment of enal 3 with
acetylenic tin 4 in the presence of trimethylsilyl chloride and
catalytic Ni(acac)2/DIBAL cleanly afforded product 5. Cyclization
of 5 with either triethylborane or triethylsilane as the reducing agent
in the presence of catalytic Ni(COD)2/PPh3 cleanly afforded
bicyclononenol 6. Whereas Et3B was the best reducing agent in
cyclizations of 1, Et3SiH afforded better yield and diastereoselec-
tivity in cyclizations of cyclic template 5.
A central challenging requirement for the development of the
desired sequential process is that an acetylenic functionality must
be consumed in an initial coupling process with an enal at the same
time that a second acetylenic functionality is introduced. That
second acetylenic functionality must then undergo cyclization with
the saturated aldehyde that is generated during the initial enal/alkyne
coupling. Important exploratory investigations from Ikeda suggested
that the nickel-catalyzed coupling of enones, alkynes, and acetylenic
tin reagents possessed the critical features outlined above.4 We have,
therefore, explored the scope of enal, alkyne, acetylenic tin
couplings and found that the process is indeed exceptionally
attractive for the generation of ynals 1 that are poised for further
nickel-catalyzed cyclizations. The optimum conditions involved
generation of the active catalyst by reduction of Ni(acac)2 with
DIBAL, followed by the addition of the acetylenic tin, alkyne, enal,
and trimethylsilyl chloride. As demonstrated in Table 1, each of
the three components can be varied to allow access to a broad range
of functionalized enynals. For instance, the enal may be unsubsti-
9
2106 VOL. 124, NO. 10, 2002 J. AM. CHEM. SOC.
10.1021/ja0175845 CCC: $22.00 © 2002 American Chemical Society