Palladium-catalyzed elimination/isomerization of enol triflates into 1,3-dienes

Article abstract of DOI:10.1002/anie.201101820

Dying to be Dienes: Substituted 1,3-dienes were synthesized by the title reaction (see scheme; Tf=trifluoromethanesulfonyl). Preliminary studies support a mechanistically distinct pathway that involves an initial β-hydride elimination from a cationic vinyl palladium(II) intermediate, a subsequent regiospecific hydropalladation of the corresponding allene intermediate, and a final β-hydride elimination. Copyright

Full text of DOI:10.1002/anie.201101820

Communications
DOI: 10.1002/anie.201101820
Palladium Catalysis
Palladium-Catalyzed Elimination/Isomerization of Enol Triflates into
1,3-Dienes**
Ian T. Crouch, Timothy Dreier, and Doug E. Frantz*
The prevalence of 1,3-dienes as substrates in some of the most
important transformations in organic synthesis (e.g. Diels–
Alder reactions) engenders this class of compounds as highly
sought after raw materials for carbon–carbon bond construc-
tion. In addition, nature has incorporated 1,3-dienes into a
multitude of biologically active natural products, thus further
solidifying the broad interest of this functional group beyond
simple methodology development. Nonetheless, despite their
pervasiveness and significant efforts from the synthetic
community, the stereoselective synthesis of functionalized,
highly substituted 1,3-dienes remains a formidable chal-
lenge.[1–4]
Our venture into the synthesis of 1,3-dienes began with a
program to identify nontraditional reaction manifolds of
stereodefined enol triflates because of their ability to act as
pluripotent substrates with various catalysts and nucleophiles.
In addition, our ability to readily access enol triflates with
precise control of the acyclic stereochemistry provides the
opportunity to exploit the unique reactivity differences
between stereoisomeric starting materials.[5] Herein, we
report a conceptually new and mechanistically distinct
catalytic reaction of enol triflates that is mediated by a
commercially available palladium(0) catalyst en route to
substituted 1,3-dienes (Scheme 1).
decision was driven by two prevailing ideas. First, we were
committed to identify a commercially available palladium
catalyst for utmost simplicity and practicality in the imple-
mentation of this method. Second, we needed a catalyst that
could rapidly undergo facile reductive elimination of triflic
acid from a [LnPdH(OTf)] intermediate to regenerate the
active palladium(0) catalyst, thus minimizing the lifetime of
the palladium-hydride intermediates. The elegant studies of
Fu and co-workers showed that the reductive elimination of
HX from a [LnPdHX] species can be a kinetically and
thermodynamically favored process when L = P(tBu)3, and
therefore simplified our decision to use this catalyst.[6] In
addition to catalyst choice, from the outset it was not clear
whether we would encounter differential activity between the
Z and E enol triflates given the fact that their internal
chelation environments are significantly different (on the
assumption that a stereospecific oxidative addition of palla-
Table 1: Survey of the reaction conditions.[a]
Entry Enol
Triflate[b]
Solvent Base
Additive
Yield [%][c]
1
2
3
4
5
6
7
8
(E)-1
DMSO Hꢀnig’s
–
–
–
–
–
–
–
–
–
–
–
9
(E)-1
(E)-1
(E)-1
(E)-1
(E)-1
(E)-1
(E)-1
(E)-1
(E)-1
(E)-1
(E)-1
DCE
CH3CN Hꢀnig’s
THF Hꢀnig’s
EtOAc Hꢀnig’s
DMF Hꢀnig’s
Hꢀnig’s
51
66
65
54
47
82
24
14
65
43
80[e]
0
Scheme 1. Catalytic elimination/isomerization of enol triflates into 1,3-
dienes. Tf=trifluoromethanesulfonyl.
toluene Hꢀnig’s
toluene Et3N
9
toluene DIPA
toluene PMP
toluene Cy2NMe
10
11
12
Our initial experiments for the tandem elimination/
isomerization of enol triflates were guided by a rational
decision to use [Pd(PtBu3)2] as our catalyst of choice. This
toluene Na2CO3 H2O[d]
13[f] (E)-1
toluene Na2CO3 P(tBu3)[g]
toluene Na2CO3 H2O[h]
14
15
(Z)-1
(Z)-1
0
toluene Hꢀnig’s TMSOTf[i]
toluene Hꢀnig’s P(tBu3)[g] + TMSOTf[i]
82[h]
0
16[f] (Z)-1
[*] I. T. Crouch, T. Dreier, Prof. D. E. Frantz
Department of Chemistry, The University of Texas at San Antonio
San Antonio, TX 78249 (USA)
[a] Reactions performed at 0.2m using 1.0 equiv of enol triflate and
2.0 equiv of base. [b] Isomeric purity is >99% as determined by LC/MS.
[c] HPLC assay yields for the 2E,4E stereoisomer using an analytical
standard of the product. [d] Used 2 equiv. [e] Isolated in 80% yield as a
10:1 ratio of (2E,4E)/(2E:4Z) stereoisomers. [f] No catalyst. [g] Used
20 mol%. [h] Isolated in 80% yield as a 10:1 ratio of (2E,4E)/(2E:4Z)
stereoisomers. [i] Used 1 equiv. Bn=benzyl, Cy=cyclohexyl, DCE=1,2-
dichloroethane, DIPA=diisopropylamine, DMF=N,N’-dimethylform-
amide, DMSO=dimethyl sulfoxide, Hꢀnig’s=diisopropylethylamine,
PMP=1,2,2,6,6-pentamethylpiperidine, THF=tetrahydrofuran, TMS=
trimethylsilyl.
Fax: (+1)210-458-7428
E-mail: doug.frantz@utsa.edu
Homepage: http://www.utsa.edu/chem/frantz.html
[**] D.E.F. would like to thank the Welch Foundation (AX-1735) and The
University of Texas at San Antonio for generous support of this
work.
Supporting information for this article is available on the WWW
under http://dx.doi.org/10.1002/anie.201101820.
6128
ꢀ 2011 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Angew. Chem. Int. Ed. 2011, 50, 6128 –6132

dium(0) takes place in the first step). Thus, we judiciously
screened both Z and E stereoisomers of enol triflate 1 during
the development phase of our investigations. The results of
our optimization screen are shown in Table 1.
Two separate sets of reaction conditions have been
identified based on the starting olefin geometry of the enol
triflate. For the E-enol triflate 1, various solvent and Brønsted
base combinations provide the corresponding diene 2 in good
yield in the presence of 2.5 mol% of [Pd(PtBu3)2].[7] How-
ever, the optimal reaction conditions we chose for the E enol
triflates were toluene, Na2CO3 (2 equiv), and H2O (2 equiv) at
508C with 2.5 mol% catalyst (Table 1, entry 12). Interestingly,
(Z)-1 completely failed to provide any observable product by
LC/MS regardless of the base or solvent employed. None-
theless, we discovered that the addition of a strong Lewis acid
such as trimethylsilyl triflate (TMSOTf) along with Hꢀnigꢁs
base restored the catalytic activity, thus providing 2 in
comparable yield to its stereoisomer (Table 1, entry 15).
Notably, without a source of palladium(0) the reactions of
both stereoisomers did not provide any product (2), even in
the presence of a phosphine ligand and/or a Lewis acid
(Table 1, entries 13 and 16).
Scheme 3. Mechanistic studies.
With these optimized reaction conditions in hand for both
the E and Z enol triflates, we explored the initial scope and
limitations of this new method for the synthesis of the
corresponding 1,3-dienes. The culmination of our efforts thus
far is highlighted in Table 2. Gratifyingly, a broad range of
acyclic enol triflates were viable substrates and in each case
provided the expected 2E,4E dienoates as the major stereo-
isomeric products. In addition, the successful synthesis of
(2E,4Z)-1,3-cyclooctadiene (8; Table 2, entry 7) from the
corresponding cyclic enol triflate provides, as far as we are
aware, the only known synthesis of this cyclic diene.
Incorporation of a heteroatom-containing substituent on the
enol triflate is highlighted in entries 3 and 11–13 in Table 2,
and includes the synthesis of the dienylboronate ester 14 as a
highly versatile synthetic building block.[8] The ability to
convert either stereoisomer of the starting enol triflate into
Scheme 4. Additional studies.
the corresponding diene in comparable yields and stereose-
lectivity (Table 2, entries 4 and 5) should allow the utmost
flexibility during strategic synthetic planning
when only one stereoisomer of the enol triflate
is synthetically accessible, or a mixture of stereo-
isomers is unavoidable.
Preliminary mechanistic investigations into
these reactions have shed some light on a unifying
reaction pathway that rationalizes the stark
reactivity differences between the E and Z enol
triflates. We propose that the reactions proceed
through the catalytic pathway as outlined in
Scheme 2. Following oxidative addition into the
enol triflate, a b-hydride elimination from a
cationic vinyl palladium(II) complex occurs to
provide the corresponding allenoate.[9] A subse-
quent hydropalladation followed by a second b-
hydride elimination isomerizes the allene to give
the corresponding diene.[10] Catalyst regeneration
occurs by a deprotonation of the resultant
cationic palladium(II) hydride and completes
the catalytic cycle.
Scheme 2. Proposed reaction pathway. L.A.=Lewis Acid.
Angew. Chem. Int. Ed. 2011, 50, 6128 –6132
ꢀ 2011 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.angewandte.org
6129

Communications
Table 2: Preliminary substrate scope.[a]
We believe that the significant
reactivity differences between the E
and Z enol triflates is due to the
inherent intramolecular chelation
environments, which dictate their
relative reactivity after the oxida-
tive addition. In the case of E enol
triflates, a more facile b-hydride
elimination from a cationic vinyl
palladium(II) complex can be
explained by the readily accessible
open coordination site in these
complexes ((E)-15). In contrast,
we believe with Z enol triflates the
carbonyl oxygen occupies the open
coordination site in the cationic
square-planar palladium(II) com-
plex (Z)-15 and this interaction
attenuates the agostic interaction
with the b-hydrogen atoms.
Entry Enol
Triflate[b]
Method,[c]
Cat. [mol%] [h] [8C]
t
T
Product[d]
Yield
[%][e]
1
2
3
A, 2.5
A, 2.5
A, 2.5
12 50
12 50
12 25
3
4
5
87
84
59[f]
4
5
A, 2.5
B, 2.5
12 50
12 50
6
6
54
72
However, strong Lewis acids,
such as TMSOTf, can effectively
compete with the palladium center
for coordination to the carbonyl
oxygen atom, thus freeing a coordi-
nation site for a subsequent b-
hydride elimination to occur.[11]
Nonetheless, despite these signifi-
cant kinetic differences between E
and Z enol triflates, current evi-
dence supports the hypothesis that
the products are under thermody-
namic control.[12]
6
7
B, 10
B, 10
12 80
12 80
7
8
72
92
8
A, 2.5
B, 10
B, 10
B, 2.5
12 50
12 80
12 80
12 50
9
82[g]
Several experiments provide
data consistent with our mechanis-
tic proposal (Scheme 3). First, the
reaction of enol triflate (E)-16 per-
9
10 64
11 90[h]
12 74
formed with
a
stoichiometric
amount of catalyst and in the
absence of exogenous base pro-
vided the diene product 9 in rea-
sonable yield. This is consistent with
the pathway that occurs through an
initial oxidative addition/b-hydride
elimination sequence rather than a
Brønsted base mediated deprotona-
10
11
tion/elimination
pathway.[13]
12
13
B, 10
12 80
12 50
13 81
Second, control reactions in the
absence of a palladium catalyst
with enol triflates (Z)-1 or (E)-1
provide little or no conversion into
alkynoate 17 (or the corresponding
allenoate) after 48 h. This result
supports the critical involvement
of palladium in the first step and is
evidence against alkynoates (or
allenoates) as kinetically viable
B, 2.5
14 41[f]
[a] Reactions performed at 0.2m in toluene under an inert atmosphere of N2. [b] Isomeric purity >99%
as determined by LC/MS. [c] Method A: Na2CO3 (2 equiv) as base and H2O (2 equiv); Method B:
Hꢀnig’s base (2 equiv) and TMSOTf (1 equiv). [d] Diene stereochemistry (both major and minor
isomers) confirmed by nOe analysis. [e] Yields of the isolated products are reported as an average of two
separate experiments (<10% difference between both runs). [f] Isolated as a 5:1 mixture of 4E/4Z
stereoisomers. [g] Isolated as a 7:1 mixture of 4E/4Z stereoisomers. [h] Isolated as an 8:1 mixture of 2E/ entry points into the catalytic cycle
2Z stereoisomers.
by a base-mediated elimination of
6130
www.angewandte.org
ꢀ 2011 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Angew. Chem. Int. Ed. 2011, 50, 6128 –6132

Received: March 14, 2011
Published online: May 13, 2011
triflic acid. Furthermore, to provide support of the interme-
diacy of a p-allyl complex, we subjected enol triflates 18–20 to
the reaction conditions (Scheme 4). The success of these
reactions, which provided isomeric dienes 21–23, is consistent
with the formation of a p-allyl palladium complex capable of
b-hydride elimination from the a-carbon atom. More impor-
tantly, the difficulty of accessing substituted 1,3-butadienes
such as 21–23 by using traditional methods highlights the
unique capacity of this novel reaction of enol triflates.
Keywords: dienes · enol triflates · homogeneous catalysis ·
.
palladium · synthetic methods
[1] Isomerization of alkynoates into dienoates: a) B. M. Trost, T.
Schmidt, J. Am. Chem. Soc. 1988, 110, 2301; b) B. M. Trost, U.
Kazmaier, J. Am. Chem. Soc. 1992, 114, 7933; c) D. Ma, Y. Yu, X.
Lu, J. Org. Chem. 1989, 54, 1105.
Finally, ongoing experiments to explore the extended
scope of this reaction with respect to other enol sulfonate
substrates are shown in Scheme 5. For example, unactivated
[2] Recent cross-coupling routes to dienes: a) H. Yu, W. Jin, C. Sun,
J. Chen, W. Du, S. He, Z. Yu, Angew. Chem. 2010, 122, 5928;
Angew. Chem. Int. Ed. 2010, 49, 5792; b) Y. H. Xu, J. Lu, T. P.
Loh, J. Am. Chem. Soc. 2009, 131, 1372; c) J. P. Ebran, A. L.
Hansen, T. M. Gøgsig, T. Skrydstrup, J. Am. Chem. Soc. 2007,
129, 6931; d) K. S. Yoo, C. H. Yoon, K. W. Jung, J. Am. Chem.
Soc. 2006, 128, 16384; e) G. A. Molander, Y. Yokoyama, J. Org.
Chem. 2006, 71, 2493; f) T. Kurahashi, H. Shinokubo, A. Osuka,
Angew. Chem. 2006, 118, 6484; Angew. Chem. Int. Ed. 2006, 45,
6336; g) C. Fu, S. Ma, Org. Lett. 2005, 7, 1707; h) A. L. Hansen,
J. P. Ebran, M. Ahlquist, P. O. Norrby, T. Skrydstrup, Angew.
Chem. 2006, 118, 3427; Angew. Chem. Int. Ed. 2006, 45, 3349;
i) Y. Hatamoto, S. Sakaguchi, Y. Ishii, Org. Lett. 2004, 6, 4623;
j) J. Takagi, K. Takahashi, T. Ishiyama, N. Miyaura, J. Am. Chem.
Soc. 2002, 124, 8001.
[3] Recent phosphine-mediated routes to dienes: a) D. J. Dong, H.-
H. Li, S.-K. Tian, J. Am. Chem. Soc. 2010, 132, 5018; b) J. T.
Markiewicz, D. J. Schauer, J. Lꢂfstedt, S. J. Corden, O. Wiest, P.
Helquist, J. Org. Chem. 2010, 75, 2061; c) S. Xu, W. Zou, G. Wu,
H. Song, Z. He, Org. Lett. 2010, 12, 3556.
[4] Metathesis approaches to dienes and dienoates: a) W. Zhu, M.
Jimꢃnez, W.-H. Jung, D. P. Camarco, R. Balachandran, A. Vogt,
B. W. Day, D. P. Curran, J. Am. Chem. Soc. 2010, 132, 9175; b) G.
Moura-Letts, D. P. Curran, Org. Lett. 2007, 9, 5; c) R. P. Murelli,
M. L. Snapper, Org. Lett. 2007, 9, 1749; d) T. W. Funk, J. Efskind,
R. H. Grubbs, Org. Lett. 2005, 7, 187.
[5] D. Babinski, O. Soltani, D. E. Frantz, Org. Lett. 2008, 10, 2901.
Despite the prevailing notion that triflates are unstable and
difficult to work with, our experiences with enol triflates derived
from acetoacetates have been quite the contrary.
[6] I. D. Hills, G. C. Fu, J. Am. Chem. Soc. 2004, 126, 13178.
[Pd(tBu3P)2] can be easily handled in air and weighed out on the
benchtop, see: G. C. Fu, Acc. Chem. Res. 2008, 41, 1555.
[7] A quick survey of other palladium catalysts including [Pd-
(PPh3)4], [Pd(PCy3)2], [Pd2(dba)3], and [Pd(PPh3)2Cl2] did not
provide any of the diene product in comparable yield.
[8] N.-W. Tseng, M. Lautens, J. Org. Chem. 2009, 74, 2521.
[9] We have yet to observe the intermediacy of the allene during the
course of these reactions. Nonetheless, b-hydride eliminations
from vinyl palladium(II) complexes to form allene intermediates
have been proposed in other reactions although at much higher
temperatures. L. M. Lutete, I. Kadota, Y. Yamamoto, J. Am.
Chem. Soc. 2004, 126, 1622.
[10] For examples of isomerizations of allenes into dienes , see a) R.
Hayashi, R. P. Hsung, J. B. Feltenberger, A. G. Lohse, Org. Lett.
2009, 11, 2125; b) M. Al-Masum, Y. Yamamoto, J. Am. Chem.
Soc. 1998, 120, 3809.
[11] An alternative explanation for the role of TMSOTf is the
possibility that it promotes an unprecedented Z to E isomer-
ization of the vinyl palladium(II) complex. In either case,
however, the role of the carbonyl oxygen atom in attenuating
reactivity remains the same.
[12] In an effort to study the b-hydride elimination and hydro-
palladation steps more closely reactions with deuterium-labelled
enol triflates were performed, but were inconclusive because of
deuterium scrambling in the diene p system as a result of
Scheme 5. Preliminary studies with other substrates.
vinyl triflate 24 readily participated in the palladium-cata-
lyzed elimination/isomerization reaction to provide the
tetrasubstituted 1,3-diene 25 in excellent yield upon isolation.
Furthermore, although acetoacetate-derived enol triflates are
both readily available, and easily handled, less costly enol
sulfonate derivatives may be more amenable to large-scale
syntheses. In this regard, enol tosylates provide a viable
alternative.[14] Thus, in preliminary experiments, we subjected
enol tosylates (E)-26 and (Z)-26 to the palladium-catalyzed
elimination/isomerization. While the reaction conditions are
harsher and the yields somewhat lower than their enol triflate
counterparts, the viability of enol tosylates to provide the
corresponding 1,3-dienes shows significant promise.
In conclusion, we have discovered a new catalytic
elimination/isomerization pathway of stereodefined enol
triflates capable of providing the corresponding functional-
ized, highly substituted 1,3-dienes in synthetically useful
yields. Preliminary mechanistic studies support a distinct
catalytic pathway that rationalizes the stark reactivity differ-
ences between E and Z enol triflates through stereoisomeric
cationic vinyl palladium(II) complexes. In addition to provid-
ing a practical approach to 1,3-dienes, we believe that this new
mechanism also opens the opportunity to selectively intercept
allenes and p-allyl complexes through tunable cationic vinyl
palladium(II) species. A full report on the further develop-
ment of this approach to additional substrates and a detailed
mechanistic study is forthcoming.
Angew. Chem. Int. Ed. 2011, 50, 6128 –6132
ꢀ 2011 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.angewandte.org
6131

Communications
À
equilibrating Pd H(D) intermediates. See the Supporting Infor-
mation for more details.
the absence of base, however, is evidence against this alternative
pathway.
[14] For the recent use of enol tosylates in cross-coupling reactions,
see Ref [2h] and a) H. Nakatsuji, K. Ueno, T. Misaki, Y. Tanabe,
Org. Lett. 2008, 10, 2131; b) J. M. Baxter, D. Steinhuebel, M.
Palucki, I. W. Davies, Org. Lett. 2005, 7, 215; c) D. Steinhuebel,
J. M. Baxter, M. Palucki, I. W. Davies, J. Org. Chem. 2005, 70,
10124.
[13] The base-mediated elimination of triflic acid from enol triflates
to provide the corresponding alkynoate (or allenoate) could
serve as a potential pathway for these reactions that bypasses the
initial oxidative addition and subsequent b-hydride elimination
from a vinyl palladium(II) complex. The success of reactions in
6132
www.angewandte.org
ꢀ 2011 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Angew. Chem. Int. Ed. 2011, 50, 6128 –6132