Serial ligand catalysis: A highly selective allylic C-H oxidation

Article abstract of DOI:10.1021/ja0500198

We are reporting a mild, chemo-, and highly regioselective Pd(II)-catalyzed allylic oxidation of α-olefins to furnish branched allylic esters that proceeds via a novel serial ligand catalysis mechanism in which two different ligands (i.e., vinyl sulfoxide 2 and BQ) interact sequentially with the metal to promote distinct steps of the catalytic cycle (i.e., C-H cleavage and π-allyl functionalization, respectively). Copyright

Full text of DOI:10.1021/ja0500198

Published on Web 04/21/2005
Serial Ligand Catalysis: A Highly Selective Allylic C-H Oxidation
Mark S. Chen, Narayanasamy Prabagaran, Nathan A. Labenz, and M. Christina White*
Department of Chemistry and Chemical Biology, HarVard UniVersity, Cambridge, Massachusetts 02138
Received January 3, 2005; E-mail: white@chemistry.harvard.edu
Table 1
Conventional homogeneous catalysis relies on one transition
metal/ligand combination to promote all steps within a catalytic
cycle. This approach is suboptimal when different steps within the
cycle place different demands on the catalyst. Herein we report
for the first time a serial ligand catalysis mechanism in which two
different ligands interact sequentially with the metal to promote
different product-forming steps of the same catalytic cycle.
We recently reported sulfoxide-promoted, catalytic Pd(OAc)2/
benzoquinone (BQ)/AcOH R-olefin allylic oxidation systems1 that
have the interesting feature of furnishing either predominantly linear
or branched allylic acetates depending on whether DMSO or bis-
sulfoxide ligands are used, respectively.1a While investigating the
bis-sulfoxide-promoted system, we discovered that 1 partially
decomposes2 under the reaction conditions to generate vinyl
sulfoxide 2 (Table 1). We tested commercially available 2 and found
that 10 mol % 2/Pd(OAc)2 effectively promotes the oxidation
reaction to furnish branched products with no decomposition.1b,3
Reducing the equivalents of AcOH significantly improves regio-
selectivities (Table 1, entries 3a,b) by suppressing a background
Pd(II)-mediated isomerization.4 We now report a vinyl sulfoxide-
promoted catalytic system for the mild, chemo- (R- versus internal
olefins), and highly regioselective C-H oxidation of R-olefins to
furnish allylic alkyl and aryl esters that proceeds via a mechanism
in which two different ligands are responsible for promoting
different steps in the catalytic cycle (Table 3).
a Average of 2-3 runs. Yields are corrected for response factor variations.
b Complex A (10 mol % based on Pd), 72 h. c Methyl-1,4-benzoquinone.
d 2,6-Dimethyl-benzoquinone. e Duroquinone. f 1 equiv of Pd(OAc)2/2, 24
h.
Scheme 1. Serial Ligand Catalysis
Mechanistic studies were carried out to establish the fundamental
steps of this catalytic cycle and the role of vinyl sulfoxide 2 and
BQ therein. When stoichiometric mixtures of 1-undecene, Pd(OAc)2,
1
and 2 were heated and monitored by H NMR, dimeric π-allyl-
palladium acetate complex A was observed in ca. 59% yield (eq
1).5a-c When BQ was then added to this reaction mixture, formation
of allylic acetate product was observed with yields and regio-
selectivities similar to those observed for the stoichiometric reaction
run in the presence of BQ (eq 1, Table 1, entry 3h). In the absence
of 2, with and without BQ, formation of complex A was not
observed. These data are consistent with 2, and not BQ, acting as
a ligand to effect Pd-mediated allylic C-H cleavage to likely form
a monomeric π-allylpalladium intermediate that is detected in the
form of dimeric complex A.
concentration disfavoring dimerization.5d Moreover, no olefin
exchange was observed upon exposure of A to 1-benzyloxy-4-
pentene (10 equiv) in the presence of AcOH (40 equiv) and 2 after
48 h at 43 °C in dioxane. This indicates that formation of the
π-allylpalladium moiety is not reversible and that a reaction by a
pathway separate from this intermediate is unlikely. Thus, while
A does not lie within the catalytic cycle, it can re-enter the cycle,
albeit slowly, upon exposure to the reaction conditions presumably
as its monomeric version II or III (Scheme 1).
We examined the effect of individual reaction components on
the functionalization of synthetic A5b,c under conditions that mimic
the reaction of a monomeric Pd-π-allyl intermediate during one
catalytic reaction cycle (ca. 3.3 mM Pd, 20 equiv of BQ, 40 equiv
of AcOH, 1 equiv of 2, Table 2). BQ was uniquely effective at
promoting product formation with similar yields and the same
regioselectivities as those observed under standard catalytic condi-
tions (Table 2, entry 3). Importantly, the addition of both vinyl
sulfoxide 2 and BQ did not significantly impact yields or regio-
selectivities (Table 2, entry 4). These data are consistent with BQ,
and not vinyl sulfoxide 2, acting as a ligand to effect functional-
ization from a monomeric Pd-π-allyl intermediate.6
A combination of A (10 mol % Pd) and 2 (10 mol %) catalyzed
the allylic oxidation reaction with similar yields and the same
regioselectivities as those observed with Pd(OAc)2/2 (10 mol %)
but with lower rates of product formation (Table 1, entries 3b,c).
Significantly, A is not observed by H NMR under the catalytic
reaction conditions, consistent with II being formed only at low
1
9
6970
J. AM. CHEM. SOC. 2005, 127, 6970-6971
10.1021/ja0500198 CCC: $30.25 © 2005 American Chemical Society

C O M M U N I C A T I O N S
Table 2 a
The high functional group compatibility of this mild method is
demonstrated in Table 3. Incorporation of benzylic, acidic, and basic
functionality is possible without a loss in yields or regiocontrol
(Table 3, entries 2-4, 7, and 8). Significantly, with diolefin
substrates, chemoselectivity is observed for the R-olefin (Table 3,
entries 5 and 6). Variation in the steric and electronic properties of
the carboxylic acid component is also well-tolerated (Table 3, entries
8-13).
In conclusion, this report describes a mild, chemo-, and highly
regioselective Pd-catalyzed allylic oxidation reaction that proceeds
via a novel mechanism where two different ligands interact serially
with Pd to promote different steps of the catalytic cycle. Further
studies are targeted toward elucidating the role of 2 in promoting
C-H cleavage, developing asymmetric versions of this reaction,
and exploring the generality of serial ligand catalysis for effecting
other challenging transition metal-mediated transformations.
entry
conditions
GC yield, B t
)
6 h
[B:L]
1
2
3
4
AcOH (40 equiv)
2 (1 equiv)
BQ (20 equiv)
2 (1 equiv)
BQ (20 equiv)
PPh3 (20 equiv)
dppec (10 equiv)
<1%
<1%
58%
62%
-
-
32:1
34:1
5
6
42%
44%
1:1b
1:1d
a Average of 3 runs. b L, 46%. c 1,2-Bis(diphenylphosphino)ethane. d L,
47%.
Table 3 a
Acknowledgment. M.C.W. gratefully acknowledges the Henry
Dreyfus Foundation and Harvard University for financial support.
We are grateful to Dr. G. Dudek and Ms. Q. Liao for HRMS and
to a reviewer for suggesting the experiment in Table 2, entry 6.
Supporting Information Available: Detailed experimental pro-
cedures and full characterization. This material is available free of
charge via the Internet at http://pubs.acs.org.
References
(1) Pd(II): (a) Chen, M. S.; White, M. C. J. Am. Chem. Soc. 2004, 126, 1346.
(b) 2 in solvent quantities gives no reaction. (c) Fraunhoffer, K. J.;
Bachovchin, D. A.; White, M. C. Org. Lett. 2005, 7, 223. (d) Hansson,
S.; Weumann, A.; Rein, T.; Åkermark, B. J. Org. Chem. 1990, 55, 975.
(e) McMurry, J. E.; Kocovsky, P. Tetrahedron Lett. 1984, 25, 4187. (f)
Macsari, I.; Szabo, K. Tetrahedron Lett. 1998, 39, 6345. (g) Yu, J.-Q.;
Corey, E. J. J. Am. Chem. Soc. 2003, 125, 3232. Se(IV): (h) Umbreit,
M. A.; Sharpless, K. B. J. Am. Chem. Soc. 1977, 99, 5526.
(2) Khiar, N.; Araujo, C.; Alcudia, F.; Fernandez, I. J. Org. Chem. 2002, 67,
345.
(3) Pd(II)/DMSO/O2 oxidations/functionalizations: (a) Grennberg, H.; Gogoll,
A.; Ba¨ckvall, J.-E. J. Org. Chem. 1991, 56, 5808. (b) Larock, R. C.;
Hightower, T. R.; Hasvold, L. A.; Peterson, K. P. J. Org. Chem. 1996,
61, 3584. (c) Steinhoff, B. A.; Fix, S. R.; Stahl, S. S. J. Am. Chem. Soc.
2002, 124, 766. (d) Zhou, C.; Larock, R. C. J. Am. Chem. Soc. 2004,
126, 2302.
(4) (a) Isomerization is suppressed in the presence of R-olefin (SI). (b)
Overman, L. E.; Knoll, F. M. Tetrahedron Lett. 1979, 4, 321. (c) Lowering
AcOH equivalent in the Pd(OAc)2/DMSO system leads to significantly
decreased product yields.
a Data based on an average of 3-4 runs. b 2 and Pd(OAc)2 must be mixed
neat. c Ratio based on GC analysis of crude. d Ratio based on 1H NMR
analysis of crude. e 20 mol % 2. f Temperatures below 40 °C (e.g., 38 °C)
result in decreased yields. g 48 h.
(5) (a) The structure of A was confirmed by independent synthesis5b,c and
chloride anion trapping (SI). (b) Trost, B. M.; Metzner, P. J. J. Am. Chem.
Soc. 1980, 102, 3572. (c) Robinson, S. D.; Shaw, B. L. J. Organomet.
Chem. 1965, 3, 367. (d) 2 + Pd(OAc)2 is the resting state of the catalyst
observed by 1H NMR. (e) Lowering BQ concentrations in the catalytic
reaction results in increasing amounts of A (SI).
A catalytic cycle consistent with these data is outlined in Scheme
1. The first step involves electrophilic allylic C-H cleavage of the
R-olefin via a 2/Pd(OAc)2 complex (I) to afford Pd-π-allyl
intermediate II.7 Both the sulfoxide and vinyl moieties of 2 are
necessary for effecting Pd-mediated C-H cleavage; ethyl 3 and
ketone 4 analogues do not promote this reaction (Table 1, entries
4 and 5). In the absence of BQ, monomer II dimerizes to give the
observed complex A. Under the reaction conditions, II reacts
directly with BQ (present in high concentrations)5e to give III that
is activated toward nucleophilic functionalization.8 In further support
of the well-precedented role of BQ as a functionalization-promoting
ligand,8b-f no reaction was observed with other standard Pd(0)/
Pd(II) oxidants and decreasing yields and regioselectivities were
observed with increasing steric hindrance of BQ (Table 1, entries
3d-g).6 Moreover, the branched regioselectivity of these reactions
is consistent with functionalization of an electronically dissymmetric
π-allyl intermediate that may be generated from ligands with
differential trans effects such as BQ and carboxylate.9 In support
of this, when complex A was treated with either PPh3 or dppe,
ligands capable of both breaking up dimeric A and displacing
anionic ligands to generate electronically symmetric Pd-π-allyl
intermediates, high yields of product formation with no regio-
selectivity were observed (Table 2, entries 5 and 6).
(6) Preliminary data suggest that the Pd(OAc)2/DMSO system proceeds via
a different mechanism than 2/Pd(OAc)2, which may account for the
different regioisomeric outcomes observed.1b,4c For example, the Pd(OAc)2/
DMSO system is significantly less sensitive to the steric hindrance of
BQ, suggesting that BQ may not be necessary for functionalization. This
is consistent with previous observations that DMSO promotes formation
of linear allylic acetates from R-olefins in stoichiometric Pd(OAc)2
reactions: Kitching, W.; Rappoport, Z.; Winstein, S.; Young, W. G. J.
Am. Chem. Soc. 1966, 88, 2054. Additionally, structural changes to the
DMSO ligand result in different regioselectivities, but changes to 2 do
not (SI). A complete reversal in regioisomeric outcomes as a result of a
change in mechanism (i.e., SN2′ on σ-allyl Pd vs SN2 on a π-allyl Pd) in
the amination of π-allyl PdCl complexes has been reported: Åkermark,
B.; Åkermark, G.; Hegedus, L.; Zetterberg, K. J. Am. Chem. Soc. 1981,
103, 3037.
(7) No binding of 2 with Pd(OAc)2 is detected by 1H NMR or solution IR
(SI).
(8) (a) Giovannina, R.; Knochel, P. J. Am. Chem. Soc. 1998, 120, 11186. (b)
Temple, J. S.; Riediker, M.; Schwartz, J. J. Am. Chem. Soc. 1982, 104,
1310. (c) Ba¨ckvall, J.-E.; Bystrom, S. E.; Nordberg, R. E. J. Org. Chem.
1984, 49, 4619 (d) Grennberg, H.; Gogoll, A.; Ba¨ckvall, J.-E. J. Org.
Chem. 1991, 56, 5808. (e) Backvall, J. E.; Gogoll, A. Tetrahedron Lett.
1988, 29, 2243. (f) Reference 1e.
(9) (a) von Matt, P.; Pfaltz, A. Angew. Chem., Int. Ed. Engl. 1993, 32, 566.
(b) Janssen, J. P.; Helmchen, G. Tetrahedron Lett. 1997, 38, 8025.
JA0500198
9
J. AM. CHEM. SOC. VOL. 127, NO. 19, 2005 6971