Palladium-Catalyzed α,β-Dehydrogenation of Esters and Nitriles

Article abstract of DOI:10.1021/jacs.5b02243

A highly practical and general palladium-catalyzed methodology for the α,β-dehydrogenation of esters and nitriles is reported. Generation of a zinc enolate or (cyanoalkyl)zinc species followed by the addition of an allyl oxidant and a palladium catalyst results in synthetically useful yields of α,β-unsaturated esters, lactones, and nitriles. Preliminary mechanistic investigations are consistent with reversible β-hydride elimination and turnover-limiting, propene-forming reductive elimination.

Full text of DOI:10.1021/jacs.5b02243

Communication
pubs.acs.org/JACS
Palladium-Catalyzed α,β-Dehydrogenation of Esters and Nitriles
Yifeng Chen, Justin P. Romaire, and Timothy R. Newhouse*
Department of Chemistry, Yale University, 275 Prospect Street, New Haven, Connecticut 06520-8107, United States
S
* Supporting Information
Scheme 1. Overview of α,β-Dehydrogenation
ABSTRACT: A highly practical and general palladium-
catalyzed methodology for the α,β-dehydrogenation of
esters and nitriles is reported. Generation of a zinc enolate
or (cyanoalkyl)zinc species followed by the addition of an
allyl oxidant and a palladium catalyst results in syntheti-
cally useful yields of α,β-unsaturated esters, lactones, and
nitriles. Preliminary mechanistic investigations are con-
sistent with reversible β-hydride elimination and turnover-
limiting, propene-forming reductive elimination.
lectronically polarized alkenes are useful functional groups
in organic synthesis because of the abundance of methods
E
by which they can be regioselectively manipulated. Conse-
quently, synthetic access to enones and enals from the
corresponding ketones and aldehydes has been studied since
the 1930s (Scheme 1).1−9 Early approaches to carbonyl α,β-
dehydrogenation utilized nonselective oxidants, such as 2,3-
dichloro-5,6-dicyano-1,4-benzoquinone (DDQ) and SeO2, that
suffer from limited functional group compatibility.2 The range of
substrates that can be oxidized has been broadened through the
introduction of o-iodoxybenzoic acid (IBX)3 as a dehydrogen-
ation reagent; Stahl and co-workers have recently reported a
versatile palladium-catalyzed protocol that leverages an α-C−H
cleavage mechanism to oxidize aldehydes and ketones to the
unsaturated derivatives.4 Recourse to two-step methodologies
has been most widely employed, including the Saegusa oxidation,
which necessitates the synthesis of an unstable enoxysilane as the
first step.5 The ensuing oxidation mediated by palladium has
been reported with catalytic loadings, yet stoichiometric
quantities of precious metal are frequently needed to avoid the
formation of the proto-desilylation byproduct, which is difficult
to chromatographically separate from the unsaturated product.6
Despite these advances in dehydrogenation adjacent to the
relatively acidic aldehydes and ketones, practical methods for the
direct dehydrogenation of less acidic substrates, including
aliphatic esters and nitriles, remains an unmet challenge.7,8 For
less acidic nitrile and ester substrates, approaches involving α-
phenylselenide synthesis followed by a second oxidation step are
most commonly employed; however, this approach requires
multiple operations in addition to stoichiometric quantities of
the expensive and toxic phenylselenyl halide.9 This Communi-
cation details the implementation of a strategy to directly effect
α,β-dehydrogenation of esters and nitriles that likely operates via
transmetalation of an enolate from Zn to Pd and subsequent β-
hydride elimination (Scheme 1).
resulting palladium intermediate 3 could undergo β-hydride
elimination12 to form the desired product 2 and an allylpalladium
hydride species (6), which would undergo reductive elimination
to generate propene and a Pd(0) intermediate (7).13 The
catalytic cycle would be complete after oxidative addition with an
allyl electrophile, such as allyl acetate. Initial attempts highlighted
the difficulty of suppressing the formation of the α-allyl
byproduct (8), which may have proceeded via the pathway
depicted in Scheme 1 or via direct nucleophilic attack of the
enolate at one of the terminal carbon atoms of the (π-
allyl)palladium complex.14,15
We determined the most favorable parameters for the α,β-
dehydrogenation of esters through a series of experiments
summarized in Table 1. Formation of the lithium ester enolate of
It was envisioned that a palladium enolate or cyanoalkyl
species (3)10 could be accessed via deprotonation with a suitable
base11 and transmetalation with a palladium(II) catalyst (5). The
Received: March 2, 2015
© XXXX American Chemical Society
A
DOI: 10.1021/jacs.5b02243
J. Am. Chem. Soc. XXXX, XXX, XXX−XXX

Journal of the American Chemical Society
Communication
a
Table 1. Dehydrogenation Optimization
Scheme 2. Scope of Ester α,β-Dehydrogenation
a
a
a
entry
base
LiTMP
additive
10a (%)
11 (%)
12 (%)
1
2
3
4
5
−
20 (62)
90 (95)
69 (94)
89 (92)
75 (84)
83 (84)
98 (>98)
20
nd
nd
nd
nd
nd
nd
17
nd
nd
nd
nd
nd
nd
LiTMP
ZnCl2
ZnCl2
ZnCl2
−
LiNCy2
LDA
TMPZnCl·LiCl
LiTMP
b
6
c
ZnCl2
7
LiTMP
ZnCl2
a
1
Determined by H NMR analysis using 1,3,5-trimethoxybenzene as
an internal standard. The conversions of 9a are shown in parentheses.
b
nd = not detected. Allyl acetate was used instead of allyl pivalate as
c
the oxidant. ZnCl2 was added at −40 °C. The isolated yield was 89%.
9a using lithium 2,2,6,6-tetramethylpiperidine (LiTMP) fol-
lowed by treatement with (π-allyl)palladium chloride dimer and
allyl pivalate as the oxidant delivered the desired dehydrogen-
ation product 10a in 20% yield (entry 1). Although the product
was obtained under these conditions, incomplete conversion
prevented chromatographic isolation from the starting material,
which is of similar polarity. Furthermore, the undesired π-allyl
substitution product, α-allyl ester 11, was formed in a similar
yield as the desired enoate. The lithium enolate also reacted
directly with allyl pivalate, which afforded the Claisen
condensation side product 12 in comparable yield. Reasoning
that the condensation byproducts could be eliminated through
the utilization of a less reactive metal enolate, ZnCl2 was added at
low temperature before the addition of allyl pivalate. Under these
conditions, the condensation byproduct 12 was not observed by
a
Allyl acetate was used as the oxidant instead of allyl pivalate.
substituted cyclohexane ester (9g) was suitable, and the
difluorosubstituted alkene side product was not observed
under the standard reaction conditions. The ethereal ketal and
tetrahydropyranyl functionality did not interfere, as observed by
the formation of the α,β-unsaturated products 10h and 10i in 79
and 80% yield, respectively. The relatively basic N-benzylpiper-
idine functionality (10j) did not inhibit the dehydrogenation
reaction. Likewise, the N-tosylpiperidine sulfonamide 10k and
the N-Boc-protected prolinol product 10l could be obtained in
moderate to good yields. Dehydrogenation of the N-methyl-
pseudophedrine-derived ester 9m proceeded smoothly.
Although there are a few examples of one-step dehydrogenations
of more acidic lactones, general synthetic methods have not
emerged;18 β-substituted lactones remain a particularly challeng-
ing substrate class.19 Under the optimized conditions, sclareolide
and menthone lactone were oxidized in good yields (10n and
10o).
Complete conversion was also observed for a number of
nitriles using a similar oxidation system; utilization of readily
available allyl acetate provided slightly better results than allyl
pivalate (Scheme 3).20 Normal-, medium-, and large-sized cyclic
alkanes were suitable substrates (13a−e), and the oxidation of
strained bicyclic structures (13f and 13g) also proceeded
smoothly using the optimized procedure. Even the more
challenging primary and secondary acyclic substrates can be
dehydrogenated (13h−k) without the observation of con-
densation byproducts. Heterocyclic motifs, including a Boc-
protected piperidine (14l), furan (14m), and indole (14n), were
tolerated. A number of other functionalities remained intact
under the basic reaction conditions, including a fused carbocycle
(14o), protected phenol (14p), silyl ether (14q), and
methoxymethyl ether (14r). Our observations to date indicate
that a wide variety of functionalities can be tolerated under the
moderately basic reaction conditions, in part because this
1
GC−MS or H NMR spectroscopy after aqueous workup.
Additionally, the α-allylation pathway was also suppressed by the
inclusion of ZnCl2,16 and the dehydrogenated product was
formed in 90% yield (entry 2). Attempts at optimization of the
amide species resulted in less efficient oxidation (entries 3 and 4),
which could be due to the possibility that the amide base alters
the coordination sphere of the catalytically active Pd center.
Alternatively, the speciation of the zinc enolate may be perturbed
through the use of different bases. Utilization of the Knochel
reagent TMPZnCl·LiCl resulted in a depreciation of the yield
(entry 5). For the case of the cinnamyl substrate 9a, allyl acetate
was a competent but slightly less efficient oxidant (entry 6).
Finally, when the enolate formation and transmetalation were
conducted at −40 °C instead of at 0 °C, complete conversion was
1
1
observed by H NMR analysis, and the H NMR yield was
greater than 98%. The isolated yield for the formation of tert-
butyl cinnamyl ester (10a) was 89%.17
With optimization of the reaction conditions complete, the
scope of this process was explored (Scheme 2). Aryl chloride
substrate 9b also underwent efficient dehydrogenation. The
dehydrogenation of the α-terpineol-derived ester 9c, bearing a
trisubstituted cyclic alkene, proceeded in 78% yield. A terminal
alkene (9d) was also tolerated, resulting in a 76% yield of the
desired unsaturated compound. The formation of cyclic alkenes
was feasible in the cases of the methyl- and tert-butylcyclohexane
carboxylates 10e and 10f. The marginally acidic trifluoromethyl-
B
DOI: 10.1021/jacs.5b02243
J. Am. Chem. Soc. XXXX, XXX, XXX−XXX

Journal of the American Chemical Society
Communication
Scheme 3. Scope of Nitrile α,β-Dehydrogenation
Figure 1. Kinetic isotope effect study.
This report details a mode of reactivity for (π-allyl)palladium
catalytic intermediates that has facilitated the development of a
methodology for the α,β-dehydrogenation of esters, lactones,
and nitriles that features low catalyst loadings, an inexpensive
oxidant, and ready purification of the products. Significant
challenges remain in carbonyl α,β-dehydrogenation, including
complementary methodologies to directly introduce unsatura-
tion into other electron-withdrawing-group-containing sub-
strates and selective dehydrogenation of substrates with multiple
groups prone to oxidation. Studies along these lines and further
exploration of catalysis using (π-allyl)palladium intermediates
are underway.
dehydrogenation can proceed at ambient temperature. On larger
scales a reduced loading of Pd could be used; 14a could be
produced in comparable yield on a 10 mmol scale with 0.5 mol %
palladium catalyst.21
To gain insight into which step is turnover-limiting in order to
improve the scope and efficiency of this transformation, we
conducted parallel and intramolecular kinetic isotope effect
(KIE) studies (Figure 1). If a turnover-limiting step were to
involve either β-hydride elimination or reductive elimination, a
primary KIE would be anticipated for the parallel oxidation of 9a
and the corresponding β,β-dideuterio substrate 9a-d2. Thus, in
separate reaction vessels, 9a and 9a-d2 were subjected to identical
dehydrogenation conditions (Figure 1A); it was observed that 9a
was oxidized to 10a at a greater rate than the corresponding
dideuterio substrate. The KIE value determined from the average
of three runs via the method of initial rates was 2.3 0.1 (Figure
1A).22 The observation of a primary parallel KIE rules out
isotope-insensitive steps,23 such as transmetalation and product
dissociation, as turnover-limiting steps and instead requires that
C−H bond cleavage or formation occurs in the turnover-limiting
step or steps (i.e., β-hydride elimination24 or reductive
elimination).25 Given the results of the parallel KIE experiments,
if a smaller value were observed for the intramolecular KIE
experiment, then the product-determining step, namely, the β-
hydride elimination, could not be the turnover-limiting step. The
intramolecular competition experiment with 9a-d1 (Figure 1B)
ASSOCIATED CONTENT
* Supporting Information
■
S
Experimental procedures and spectroscopic data for all new
compounds, including 1H and 13C NMR spectra. The Supporting
Information is available free of charge on the ACS Publications
website at DOI: 10.1021/jacs.5b02243.
AUTHOR INFORMATION
Corresponding Author
*timothy.newhouse@yale.edu
■
Notes
The authors declare no competing financial interest.
ACKNOWLEDGMENTS
■
exhibited a KIE value of 1.0
0.1. The lack of a primary
Financial support for this work was provided by Yale University
and the NSF (GRFP to J.P.R.). The authors thank Prof. Robert
Crabtree, Prof. Jonathan Ellman, Prof. Seth Herzon, Prof. Nilay
Hazari, Prof. Patrick Holland, Prof. James Mayer, and Prof. Scott
Miller for helpful discussions and the use of equipment and
chemicals. We gratefully acknowledge Ms. Aneta Turlik for a
intramolecular KIE suggests that a reversible β-hydride
elimination is followed by a turnover-limiting reductive
elimination.26 However, further theoretical and experimental
mechanistic investigations are ongoing and will be reported in
due course.
C
DOI: 10.1021/jacs.5b02243
J. Am. Chem. Soc. XXXX, XXX, XXX−XXX

Journal of the American Chemical Society
Communication
(16) Zinc ester enolates have previously been shown to undergo α-
allylation when treated with allyl acetate and Pd(PPh3)4. For a seminal
example, see: Boldrini, G. P.; Mengoli, M.; Tagliavini, E.; Trombini, C.;
Umani-Ronchi, A. Tetrahedron Lett. 1986, 27, 4223.
protocol check and Mr. Alexander Thompson for early
experiments.
1
REFERENCES
(17) The difference in the H NMR and isolated yields may be the
■
result of the formation of many minor byproducts, loss of material
during purification, or a systematic error in the measurement of the 1H
NMR yield.
(18) For seminal examples of lactone dehydrogenation under forcing
conditions, see: (a) (PhSeO)2O: Barton, D. H.; Lester, D. J.; Ley, S. V. J.
Chem. Soc., Chem. Commun. 1978, 130. (b) DDQ: Cross, A. D. (Syntex
Corp.). Neth. Pat. 6.503,543 (1965).
(19) For an isolated example of leveraging aromaticity as a driving force
to form coumarins, see: Jagdale, A. R.; Sudalai, A. Tetrahedron Lett. 2008,
49, 3790.
(1) For reviews of α,β-dehydrogenation, see: (a) Walker, D.; Hiebert, J.
D. Chem. Rev. 1967, 67, 153. (b) Reich, H. J.; Wollowitz, S. Org. React.
1993, 44, 1. (c) Muzart, J. Eur. J. Org. Chem. 2010, 3779. (d) Stahl, S. S.;
Diao, T. Comp. Org. Synth. 2014, 7, 178.
(2) For seminal examples of α,β-dehydrogenation, see: (a) Astin, S.;
Newman, A. C. C.; Riley, H. L. J. Chem. Soc., Res. 1933, 391. (b) Agnello,
E. J.; Laubach, G. D. J. Am. Chem. Soc. 1957, 79, 1257.
(3) (a) Nicolaou, K. C.; Zhong, Y. L.; Baran, P. S. J. Am. Chem. Soc.
2000, 122, 7596. (b) Nicolaou, K. C.; Montagnon, T.; Baran, P. S.;
Zhong, Y. L. J. Am. Chem. Soc. 2002, 124, 2245. (c) Nicolaou, K. C.;
Montagnon, T.; Baran, P. S. Angew. Chem., Int. Ed. 2002, 41, 993.
(d) Nicolaou, K. C.; Gray, D. L. F.; Montagnon, T.; Harrison, S. T.
Angew. Chem., Int. Ed. 2002, 41, 996.
(4) (a) Diao, T.; Stahl, S. S. J. Am. Chem. Soc. 2011, 133, 14566.
(b) Diao, T.; Wadzinski, T. J.; Stahl, S. S. Chem. Sci. 2012, 3, 887.
(c) Diao, T.; Pun, D.; Stahl, S. S. J. Am. Chem. Soc. 2013, 135, 8205.
(d) Izawa, Y.; Zheng, C.; Stahl, S. S. Angew. Chem., Int. Ed. 2013, 52,
3672.
(5) (a) Theissen, R. J. J. Org. Chem. 1971, 36, 752. (b) Ito, Y.; Hirao, T.;
Saegusa, T. J. Org. Chem. 1978, 43, 1011. (c) Shimizu, I.; Minami, I.;
Tsuji, J. Tetrahedron Lett. 1983, 24, 1797. (d) Larock, R. C.; Hightower,
T. R.; Kraus, G. A.; Hahn, P.; Zheng, D. Tetrahedron Lett. 1995, 36,
2423.
(6) Similarly, Mukaiyama’s reagent proceeds via a mechanism that
results in low conversion. Difficulty in chromatographic separation of
the starting material from the product resulted in only NMR yields being
reported. See: Matsuo, J.-i.; Aizawa, Y. Tetrahedron Lett. 2005, 46, 407.
(7) For examples of α,β-dehydrogenation of the more acidic α-phenyl
and β-keto nitriles, see: (a) Levine, R.; Sheppard, C. S. J. Org. Chem.
1974, 39, 3556. (b) DiBiase, S. A.; Wolak, R. P., Jr.; Dishong, D. M.;
Gokel, G. W. J. Org. Chem. 1980, 45, 3630. (c) Noda, Y.; Akiba, Y.;
Kashima, M. Synth. Commun. 1996, 26, 4633.
(20) See the Supporting Information for selected reaction conditions
for the α,β-dehydrogenation of nitriles.
(21) With the reported reaction conditions some functionality was not
tolerated, including lactones that lack substitution at the β-position,
protected amino acids, and α-aryl-substituted nitriles. Additionally,
substrates were generally selected such that regioisomers could not be
formed during the β-hydride elimination.
(22) In order to obtain initial rate data, the catalyst loading was
decreased to 0.125 mol %. See the Supporting Information for details.
́
(23) (a) Gomez-Gallego, M.; Sierra, M. A. Chem. Rev. 2011, 111, 4857.
(b) Simmons, E. M.; Hartwig, J. F. Angew. Chem., Int. Ed. 2012, 51, 3066.
(24) Alexanian, E. J.; Hartwig, J. F. J. Am. Chem. Soc. 2008, 130, 15627.
(25) (a) Abis, L.; Sen, A.; Halpern, J. J. Am. Chem. Soc. 1978, 100, 2915.
(b) McCarthy, T. J.; Nuzzo, R. G.; Whitesides, G. M. J. Am. Chem. Soc.
1981, 103, 3396. (c) Michelin, R. A.; Faglia, S.; Uguagliati, P. Inorg.
Chem. 1983, 22, 1831.
(26) Contributions from tunneling or a secondary KIE to the measured
KIE values cannot be determined at this juncture. For a related
discussion, see: Datta, A.; Hrovat, D. A.; Borden, W. T. J. Am. Chem. Soc.
2008, 130, 2726.
(8) For an example of α,β-dehydrogenation of activated esters using
DDQ, see: Clarke, P. D.; Fitton, A. O.; Suschitzky, H.; Wallace, T. W.
Tetrahedron Lett. 1986, 27, 91.
(9) (a) Reich, H. J.; Reich, I. L.; Renga, J. M. J. Am. Chem. Soc. 1973, 95,
5813. (b) Sharpless, K. B.; Lauer, R. F.; Teranishi, A. Y. J. Am. Chem. Soc.
1973, 95, 6137. (c) Trost, B. M.; Salzmann, T. N.; Hiroi, K. J. Am. Chem.
Soc. 1976, 98, 4887.
(10) (a) Palladium cyanoalkyl complexes may exist as C- or N-bound
monomers or higher-order aggregates. See: Culkin, D. A.; Hartwig, J. F.
J. Am. Chem. Soc. 2002, 124, 9330 and references therein. (b) Palladium
ester enolates usually exist as C-bound monomers or higher-order
aggregates. See: Tian, G.; Boyle, P. D.; Novak, B. M. Organometallics
2002, 21, 1462.
(11) In the case of metalated nitriles, solution and solid-state structures
of C- and N-bound alkali-metal aggregate structures have been reported.
See: Collum, D. B.; McNeil, A. J.; Ramirez, A. Angew. Chem., Int. Ed.
2007, 46, 3002 and references therein.
(12) Syn and anti β-hydride elimination are conceivable. See:
(a) Takacs, J. M.; Lawson, E. C.; Clement, F. J. Am. Chem. Soc. 1997,
119, 5956. (b) Lloyd-Jones, G. C.; Slatford, P. A. J. Am. Chem. Soc. 2004,
126, 2690.
(13) Product dissociation may occur before or after the reductive
elimination step. See: Porth, S.; Bats, J. W.; Trauner, D.; Giester, G.;
Mulzer, J. Angew. Chem., Int. Ed. 1999, 38, 2015.
(14) For the stereochemical outcome of nucleophilic attack of anionic
carbonucleophiles, which is generally interpreted as nucleophilic attack
at carbon, see: (a) Trost, B. M.; Weber, L. J. Am. Chem. Soc. 1975, 97,
1611. (b) Trost, B. M.; Van Vranken, D. L. Chem. Rev. 1996, 96, 395.
(15) The role of dimeric or polymeric enolate species or Pd
intermediates cannot be excluded. See: Hruszkewycz, D. P.; Balcells,
D.; Guard, L. M.; Hazari, N.; Tilset, M. J. Am. Chem. Soc. 2014, 136,
7300.
D
DOI: 10.1021/jacs.5b02243
J. Am. Chem. Soc. XXXX, XXX, XXX−XXX