New catalytic cycle for couplings of aldehydes with organochromium reagents

Article abstract of DOI:10.1021/ol047661b

(Chemical Equation Presented) A new catalytic cycle has been developed to effect all three subgroups of Cr-mediated couplings, i.e., (1) Ni/Cr-mediated alkenylation, alkynylation, and arylation, (2) Co/Cr-mediated 2-haloallylation, alkylation, and proparg

Full text of DOI:10.1021/ol047661b

ORGANIC
LETTERS
2004
Vol. 6, No. 26
5031-5033
New Catalytic Cycle for Couplings of
Aldehydes with Organochromium
Reagents
Kosuke Namba and Yoshito Kishi*
Department of Chemistry and Chemical Biology, HarVard UniVersity,
12 Oxford Street, Cambridge, Massachusetts 02138
kishi@chemistry.harVard.edu
Received November 12, 2004
ABSTRACT
A new catalytic cycle has been developed to effect all three subgroups of Cr-mediated couplings, i.e., (1) Ni/Cr-mediated alkenylation, alkynylation,
and arylation, (2) Co/Cr-mediated 2-haloallylation, alkylation, and propargylation, and (3) Cr-mediated allylation. In the presence of chiral
sulfonamide ligands, good asymmetric inductions can be achieved for some of the Ni/Cr-mediated alkenylation, Co/Cr-mediated 2-haloallylation
and propargylation, and Cr-mediated allylation.
As demonstrated in a number of examples, the Ni/Cr-
mediated coupling reaction has shown its unique potential
Scheme 1
most when applied to a polyfunctional molecule.1 Thus, this
reaction has successfully been used at a late-stage in a
multiple-step synthesis where scalability and practicability
are not necessarily the priority. For application of this process
to a practical synthesis, however, it is highly desirable to
develop a catalytic process for the Ni/Cr-mediated reac-
tions. In 1996, Fu¨rstner and Shi reported seminal work on
a catalytic process of the Ni/Cr-mediated coupling reac-
tion, in which TMS-Cl and Mn(0) are used as a dissociating
agent of chromium-alkoxides (IfII in panel A, Scheme 1)
and a reducing agent of chromium, respectively.2 Electro-
chemically driven Cr(II)-mediated couplings were also
reported.3
(1) For reviews on Cr-mediated reactions, see: (a) Fu¨rstner, A. Chem.
ReV. 1999, 99, 991-1046. (b) Wessjohann, L. A.; Scheid, G. Synthesis
1999, 1-36. (c) Nozaki, H.; Takai, K. Proc. Japan Acad. 2000, 76, 123-
131. (d) Saccomano, N. A. In ComprehensiVe Organic Synthesis; Trost, B.
M., Fleming, I., Eds.; Pergamon: Oxford, 1991; Vol. 1, p 173.
The Fu¨rstner protocol works well also with chiral ligands,
permitting carbon-carbon bond formation in a catalytic
asymmetric manner.4,5 Nevertheless, we wished to make
10.1021/ol047661b CCC: $27.50
© 2004 American Chemical Society
Published on Web 12/02/2004

some improvements on the catalytic process for Cr-mediated
coupling reactions. In particular, it has been often observed
that, with low catalyst-loading, asymmetric catalytic cou-
plings smoothly progress only to a certain degree but not to
completion. Not surprisingly, it was found that this was due
to the formation of TMS-enol ethers of aldehydes. Thus, we
were interested in developing a new catalytic cycle for the
Cr-mediated couplings without use of TMS-Cl or an equiv-
alent reagent. Apparently, formation of a strong O-Cr(III)
bond is a driving force for addition of an organochromium
nucleophile to an aldehyde. Therefore, to establish a catalytic
cycle for the Cr-based reagent, a special means is required
to cleave the resultant strong O-Cr(III) bond. We were
curious about the possibility of transferring the alkoxyl group
to a second metal. Provided that the second metal has a
higher oxophilicity but a lower halophilicity than chromium,
this proposed step should be thermodynamically feasible
(IfIII in panel B, Scheme 1). Extrapolating the oxophilicity
and halophilicity estimated through combustion of a solid
metal propellant,6 we speculated that Be, Al, Zr, and Mg
might meet this condition. Consistent with this speculation,
we found that zirconocene indeed smoothly exchanges its
chloride ligand(s) with an alkoxy group(s),7,8 suggesting that
zirconocene might fulfill our needs both thermodynamically
and kinetically.
almost complete conversion, but with a lower efficiency
(57% yield). Overall, this chemical transformation is de-
scribed by eq 1 and, upon aqueous workup, unprotected
secondary alcohols are obtained.
It was equally exciting to find that the catalytic Ni/Cr-
mediated coupling proceeded in the presence of the chiral
sulfonamide 8a (Figure 1), thereby demonstrating the pos-
Figure 1. Chiral ligands for Ni/Cr-mediated couplings.
With these considerations, we tested a catalytic Ni/Cr-
mediated coupling of dihydrocinnamaldehyde (1, 1.0 equiv)
with 2-iodo-1-hexene (2a, 2.0 equiv)9 using CrCl2 (20 mol
%), NiCl2(dppp) (2 mol %), Cp2ZrCl2 (1.0 equiv), Mn (2.0
equiv), and LiCl (2.0 equiv) at room-temperature overnight
and were delighted to find that the coupling reaction did
proceed smoothly to furnish the expected allylic alcohol 3a
in 71% yield. It is noteworthy that, even with 0.5 equiv of
Cp2ZrCl2 against aldehyde 1, this catalytic process led to
sibility of extending this catalytic cycle to a catalytic asym-
metric process (vide infra). Encouraged with these results,
we then optimized the coupling conditions (Scheme 2).
Several comments are in order. First, with small modifica-
tions, this catalytic process is effective for all three subgroups
of Cr-mediated couplings,4,10,11 i.e., (1) Ni/Cr-mediated alken-
ylation, (2) Co/Cr- and Fe/Cr-mediated 2-haloallylation and
alkylation, and (3) Cr-mediated allylation. Second, the cata-
lyst loading can be lowered to 5 mol % without significant
losses in chemical yields. Importantly, even under these con-
ditions, aldehydes were completely consumed. Third, among
the tested zirconium salts,12 Cp2ZrCl2 was found to be best
by far. Fourth, among the tested reducing agents,13 manga-
nese metal (powder) was found to give the best result. Last,
we would suggest the catalytic cycle depicted in Scheme 3.
Sulfonamide ligands have been shown to effect an asym-
metric induction for the Cr-mediated couplings under both
stoichiometric and catalytic conditions.4 A similar trend is
observed for the new catalytic process. Good asymmetric
inductions were observed for the Ni/Cr-mediated alkenylation
(86% ee for 1 + 2a with 8b14), Co/Cr-mediated 2-haloal-
lylation (90% ee for 1 + 4a with 8c10) and propargylation
(82% ee for 1 + 4c with 8c10), and Cr-mediated allylation
(>90% ee for 1 + 6b with 8c10). However, a ligand
optimization is still needed for the Ni/Cr-mediated alkeny-
(2) (a) Fu¨rstner, A.; Shi, N. J. Am. Chem. Soc. 1996, 118, 2533-2534.
(b) Fu¨rstner, A.; Shi, N. J. Am. Chem. Soc. 1996, 118, 12349-12357.
(3) (a) Grigg, R.; Putnikovic, B.; Urcch, C. J. Tetrahedron Lett. 1997,
38, 6307-6308. (b) Kuroboshi, M.; Tanaka, M.; Kishimoto, S.; Goto, K.;
Tanaka, H.; Torii, S. Tetrahedron Lett. 1999, 40, 2785-2788. (c) Kuroboshi,
M.; Tanaka, M.; Kishimoto, S.; Tanaka, H.; Torii, S. Synlett. 1999, 69-
70. (d) Durandetti, M.; Nedelec, J.-Y.; Perichon, J. Org. Lett. 2001, 3, 2073-
2076.
(4) (a) Wan, Z.-K.; Choi, H.-W.; Kang, F.-A.; Nakajima, K.; Demeke,
D.; Kishi, Y. Org. Lett. 2002, 4, 4431-4434. (b) Choi, H.-W.; Nakajima,
K.; Demeke, D.; Kang, F.-A.; Jun, H.-S.; Wan, Z.-K.; Kishi, Y. Org. Lett.
2002, 4, 4435-4438. (c) Choi, H.; Demeke, D.; Kang, F.-A.; Kishi, Y.;
Nakajima, K.; Nowak, P.; Wan, Z.-K.; Xie, C. Pure Appl. Chem. 2003, 75,
1-17.
(5) (a) Bandini, M.; Cozzi, P. G.; Melchiorre, P.; Umani-Ronchi, A.
Angew Chem., Int. Ed. 1999, 38, 3357-3359. Bandini, M.; Cozzi, P. G.;
Umani-Ronchi, A. Chem. Commun. 2002, 919-927 and references therein.
Jacobsen’s Cr complex was used for these studies. (b) Lombardo, M.;
Licciulli, S.; Morganti, S.; Trombini, C. Chem. Commun. 2003, 1762-
1763. Jacobsen’s Cr complex was used for these studies. (c) Berkessel, A.;
Menche, D.; Sklorz, C. A.; Schroder, M.; Paterson, I. Angew Chem., Int.
Ed. 2003, 42, 1032-1035. Paterson, I.; Bergmann, H.; Menche, D.;
Berkessel, A. Org. Lett. 2004, 6, 1293-1295. (d) Inoue, M.; Suzuki, T.;
Nakada, M. J. Am. Chem. Soc. 2003, 125, 1140-1141. Inoue, M.; Nakada,
M. Org. Lett. 2004, 2977-2980.
(10) Kurosu, M.; Lin, M.-H.; Kishi, Y. J. Am. Chem. Soc. 2004, 126,
12248-12249.
(11) CrCl3 and CrBr3 are as effective as CrCl2 and CrBr2. In consideration
of the commercial availability and cost, anhydrous CrCl2 and CrBr3 have
been used for this study.
(12) These included Zr(OPr-i)4, Zr(acac)4, and ZrCl4.
(13) Among metals tested, Zn was effective, but the reaction with Zn
was not as clean as with Mn.
(6) Kuehl, D. K. U.S. Patent 3133841 19640519, 1964.
(7) (a) Gray, D. R.; Brubaker, C. H., Jr. Inorg. Chem. 1971, 10, 2143-
2146. (b) Femec, D. A.; Silver, M. E.; Fay, R. C. Inorg. Chem. 1989, 28,
2789-2796 and references therein.
(8) On treatment with 2-PrOH (1.0 equiv) and TEA (1.0 equiv) in
CD3CN, a clean and complete ligand exchange was observed between the
Cl of Cp2ZrCl2 and 2-PrOH (NMR).
(9) Corresponding triflate was also a good substrate.
(14) Shi, B.; Cui, S.; Kishi, Y. Unpublished work.
5032
Org. Lett., Vol. 6, No. 26, 2004

or carbazolic groups, cf., the chiral ligands studied by Cozzi
and Umani-Ronchi,5a Lombardo and Trombini,5b Berkessel
and Paterson,5c and Nakada.5d These observations are con-
sistent with the working hypothesis given in the introduction.
The usefulness of the new catalytic system is easily seen
from two examples selected from the ongoing halichondrin
program in this laboratory (Scheme 4).4,10 The first example
Scheme 2. Catalytic Cr-Mediated Coupling Reactionsa
Scheme 4
a i and ii: isolated yields with 10 and 5 mol % catalyst, respec-
tively. b1 (1.0 equiv), 2a-f (2.0 equiv), sulfonamide (11 or 6 mol
%), CrCl2 (10 or 5 mol %), proton sponge or (i-Pr)2(Et)N (11 or 6
mol %), NiCl2(dppp) (2 or 1 mol %), Cp2ZrCl2 (1.0 equiv), Mn
(2.0 equiv), LiCl (2.0 equiv), MeCN (c ) 0.2 M), rt. c1 (1.0 equiv),
4a-e (2.5 equiv), sulfonamide ligand (11 mol %), CrBr3 (10 mol
%), (i-Pr)2(Et)N or (Et)3N (11 mol %), CoPc (0.2 mol %), Cp2ZrCl2
(1.0 equiv), Mn (3.0 equiv), 2,6-lutidine (11 mol %), THF (c )
demonstrates that the efficiency of catalytic asymmetric
2-haloallylation is significantly improved with the new
catalytic system, cf., 4a,b + 12 f 13a,b, respectively. The
second example illustrates that the new catalytic asymmetric
process is applicable even for polyfunctionalized aldehydes
such as 14. Under the TMS-Cl condition, the coupling of
14 with 2e did not proceed to completion, due to the TMS-
enol ether formation. It is worthwhile noting that, upon
aqueous workup, a small amount (10∼20%) of 14 was
recovered but as a nonrecyclable 1:1 mixture of the C31
diastereomers. Under the new system, the coupling proceeded
smoothly to completion to furnish a 15:1 mixture of the C30
diastereomers in 93% yield. A direct crystallization of the
crude product from ethyl acetate-hexanes gave the optically
pure desired product 15 (mp 137 °C) in 75% yield.
In conclusion, a new catalytic reaction has been developed
to achieve Cr-mediated couplings of aldehydes with alkyl,
alkenyl, alkynyl, 2-haloallyl and allyl, propargyl, and aryl
halides. Good asymmetric inductions can be achieved at least
for some of the Ni/Cr-mediated alkenylation, Co/Cr-mediated
2-haloallylation and propargylation, and Cr-mediated ally-
lation.
d
0.2 M), 0 °C. 1 (1.0 equiv), 6a,b (2.0 equiv), CrCl3-3THF (10
mol %), sulfonamide ligand (11 mol %), (i-Pr)2(Et)N or (Et)3N
(11 mol %), Mn (3.0 equiv), 2,6-lutidine (11 mol %), THF (c )
0.2 M), 0 °C.
Scheme 3
Acknowledgment. We are grateful to the National
Institutes of Health (CA 22215) and Eisai Research Institute
for financial support.
Supporting Information Available: Experimental de-
tails. This material is available free of charge via the Internet
at http://pubs.acs.org.
lation with vinyl iodides represented by 2b and 2c as well
as the Co/Cr-mediated alkylation. Related to this, it is
worthwhile noting that the new catalytic cycle was not
interrupted by some of the chiral ligands known in this area,
including 9, 10, and 11,4,10,15 (Figure 1) but was, completely
or partially, interrupted by those ligands bearing phenolic
OL047661B
(15) Chen, C.; Tagami, K.; Kishi, Y. J. Org. Chem. 1995, 60, 5386-
5387.
Org. Lett., Vol. 6, No. 26, 2004
5033