Contemporaneous dual catalysis: Chemoselective cross-coupling of catalytic vanadium-allenoate and π-allylpalladium intermediates

Article abstract of DOI:10.1021/ja204817y

This paper presents a detailed investigation of a dual catalytic system that combines a vanadium-catalyzed Meyer-Schuster rearrangement and a palladium-catalyzed allylic alkylation. The implementation of this novel reaction relies on matching the formation rates of vanadium-allenoate and π-allylpalladium intermediates with their bimolecular coupling rate in order to minimize the undesired protonation or O-alkylation of the catalytically generated intermediates. Chemoselectivity in this dual catalytic process was successfully achieved by adjusting ligand structure and catalyst loading ratios of the vanadium and palladium catalysts. A great range of coupling partners for both the propargyl alcohol and allyl carbonate components are readily accommodated in this new transformation, which in turn provides a novel avenue to a variety of α-allylated α,β-unsaturated ketones, esters, and amides in moderate to excellent isolated yields.

Full text of DOI:10.1021/ja204817y

ARTICLE
pubs.acs.org/JACS
Contemporaneous Dual Catalysis: Chemoselective Cross-Coupling of
Catalytic VanadiumꢀAllenoate and π-Allylpalladium Intermediates
Barry M. Trost,* Xinjun Luan, and Yan Miller
Department of Chemistry, Stanford University, Stanford, California 94305-5080, United States
S Supporting Information
b
ABSTRACT:
This paper presents a detailed investigation of a dual catalytic system that combines a vanadium-catalyzed MeyerꢀSchuster
rearrangement and a palladium-catalyzed allylic alkylation. The implementation of this novel reaction relies on matching the
formation rates of vanadiumꢀallenoate and π-allylpalladium intermediates with their bimolecular coupling rate in order to
minimize the undesired protonation or O-alkylation of the catalytically generated intermediates. Chemoselectivity in this dual
catalytic process was successfully achieved by adjusting ligand structure and catalyst loading ratios of the vanadium and palladium
catalysts. A great range of coupling partners for both the propargyl alcohol and allyl carbonate components are readily
accommodated in this new transformation, which in turn provides a novel avenue to a variety of R-allylated R,β-unsaturated
ketones, esters, and amides in moderate to excellent isolated yields.
’ INTRODUCTION
economic processes.4b In the course of our recent studies, we
became interested in coupling two catalytically generated
intermediates in order to access a certain reactivity pattern
and structural motif. Reactions catalyzed by two catalysts
cooperatively are of great interest universally because they
can exhibit reactivity and selectivity not observed in a single-
catalyst system.6 Although some studies have been conducted
on the design and utilization of bifunctional catalysis systems in
organic synthesis,7 the number of dual-catalyzed reactions that
have been developed is still dramatically lower than that of
single catalyst systems. The paucity of dual catalytic processes
arises in part from issues regarding the compatibility of the two
catalysts with each other, the additional complexity of operating
two separate catalytic cycles concurrently, and the low con-
centrations of the two reactive intermediates that must undergo
a bimolecular addition to each other.
Inspired by the finite nature of chemical feedstocks and the
growing global demand for fine chemicals, the development of
new synthetic processes that proceed with high efficiency, low
energy consumption, and minimal waste has become an im-
portant area of research.1 Our research group is particularly
interested in improving synthetic efficiency following two
specific principles: (1) the development of reactions where
the maximum number of atoms of reactants appear in the
products (atom economy)2 and (2) the use of reactions that
occur only at the desired functional groups in a molecule
(chemoselectivity).3 The goal is to prepare complex molecules
in the smallest possible number of steps, while high efficiency
and chemoselectivity are maintain in each step. To date,
chemists have unimpeachably witnessed much success employ-
ing the principles of atom economy and chemoselectivity in a
great number of practical applications.4
Contemporaneous dual catalysis may exemplify efficient
synthesis by effectively employing the principles of atom econ-
omy and chemoselectivity. A general description of the concept
The hallmark of atom economical processes are simple
addition reactions where two or more versatile building blocks
are brought together by a catalyst without the need for addi-
tional stoichiometric reagents.5 Our laboratory has a rich
history of pursuing and developing these types of atom
Received: May 27, 2011
Published: June 29, 2011
r
2011 American Chemical Society
12824
dx.doi.org/10.1021/ja204817y J. Am. Chem. Soc. 2011, 133, 12824–12833
|

Journal of the American Chemical Society
ARTICLE
Scheme 1. General Description of Contemporaneous Dual
Catalysis
Scheme 3. Formation of Ester and Amide Products via
Contemporaneous Dual Catalysis
reactions. Moreover, the formation rates of I* and II* must be
relatively similar (kI ≈ kII) so that overgeneration of one
intermediate does not facilitate side reactions. Thus, the
execution of contemporaneous dual catalysis relies on careful
design of a compatible catalyst system, the appropriate choice
of the substrates, and a high affinity of the two reacting
intermediates for each other.
Recently, we reported in a preliminary communication an
effective example of contemporaneous dual catalysis that merges
a vanadium-catalyzed MeyerꢀSchuster rearrangement reaction
and a palladium-catalyzed allylic alkylation reaction to generate
R-allylated R,β-unsaturated ketones (Scheme 2).8 In this reaction, a
vanadiumꢀallenoate, which cannot be obtained directly by simple
deprotonation of a R,β-unsaturated carbonyl compound, is gen-
erated in situ by rearrangement of a propargylic alcohol.9 Instead
of undergoing simple protonation to form the enone product, the
allenoate is sufficiently reactive to intercept the π-allylpalladium
intermediate, which is also generated catalytically in the reaction.
The reaction proceeds efficiently despite the potential side reac-
tions for each intermediate (protonation and O-allylation).
Scheme 2. Palladium and Vanadium Contemporaneous
Dual Catalysis
In the development of this dual catalytic process, certain
substituents (i.e., aromatic groups) at the propargylic position
of the propargylic alcohol were necessary for the reaction to
occur. For example, when Y was an alkyl substituent, the reaction
did not proceed at all under various conditions (see Scheme 2).
We believe this result is due to the inability of the vanadium
catalyst to promote the required MeyerꢀSchuster rearrange-
ment with these nonactivated substrates. With this challenge in
mind, we decided to further explore the substrate scope of the
dual catalytic reaction because of the potential synthetic applica-
tions and because of the possibility of getting a deeper insight
into the possible mechanism. We hypothesized that placement of
an electron-donating heteroatom (oxygen or nitrogen) at the
terminus of the alkyne would sufficiently activate the propargyl
alcohol for MeyerꢀSchuster rearrangement to occur,10ꢀ12 al-
lowing us to employ alkyl groups at the propargylic position
(Scheme 3). Such a variation should enable the formation of a
highly reactive ketene acetal or ketene aminal intermediate and
further promote the synthesis of corresponding R-allylated R,β-
unsaturated ester and amide. However, there are no previous
reports on the interception of these MeyerꢀSchuster rearrange-
ment intermediates with an eletrophile to form new carbonꢀ
carbon bonds. In this report, we will describe the full investigation
of our dual catalytic reaction involving two metal-catalyzed pro-
cesses which, to our knowledge, represents the first example of
contemporaneous dual metal catalysis. Furthermore, we will also
propose a plausible mechanism for this transformation.
of dual catalysis is depicted in Scheme 1. Double activation of
both substrates (I and II) by two independent catalysts
(catalystI and catalystII) would selectively generate the corre-
sponding reactive intermediates (I* and II*) catalytically.
Subsequent bimolecular coupling of I* and II* would then
produce the desired product I*ꢀII* (pathway A). It is note-
worthy that each intermediate is surrounded by a large excess
of reactive starting materials and therefore can be easily
intercepted via pathways B (or B0) and C (or C0) to generate
undesired byproducts. In such a system, intermediates I* and
II* are present in low concentrations and will not undergo
efficient bimolecular coupling unless the rate of formation of
product (kA) is significantly faster than the rates of formation
of the undesired byproducts (kB, kB , kC, and kC ). Conse-
quently, the rapid formation of even a single undesired
byproduct would prevent formation of the dual-catalyzed
product. To conquer this challenge, it is critical to enhance
the affinity between the two transient intermediates (I* and
II*) in order to increase the reaction rate of pathway A to such
an extent that it could largely outcompete the potential side
0
0
’ RESULTS AND DISCUSSION
Experimental Plan. The development of catalytic methods
employing propargyl alcohols as latent allenoate precursors
represents a facile practical entry to enolates that cannot be
12825
dx.doi.org/10.1021/ja204817y |J. Am. Chem. Soc. 2011, 133, 12824–12833

Journal of the American Chemical Society
ARTICLE
Table 1. Optimization of the Ligand Structure
entry
Ar
L2
conv (%)a
3:4:5b
E:Zof 3b
entry
Ar
L2
conv (%)a
3:4:5b
E:Z of 3b
1
2
3
4
Ph
Ph
Ph
Ph
DPPM
DPPE
DPPP
DPPB
100
91
100:9:1
6:91:1
1:13:11
2:1:89
5:1
4:1
4:1
4:1
5
6
7
8
Ph
Ph
Ph
DPPF
85
48
77
82
13:2:1
15:72:1
6:85:1
19:1:6
5:1
2:1
10:1
4:1
Trost-LS
BINAP
DPPM
53
96
4-Cl-Ph
a Determined by 1H NMR spectroscopy by observing consumption of 1. b Determined by 1H NMR spectroscopy.
Table 2. Optimization of Solvent
Table 3. Studies of the [V]/[Pd] Catalyst Ratios
entry [V] (mol %) [Pd] (mol %)a conv (%)b
3:4:5c
E:Z of 3c
entry
solvent
conv (%)a
3:4:5b
E:Z of 3b
1
2
3
4
5
DCE
100
100
98
100:9:1
32:5:1
38:4:1
40:2:1
17:1:8
5:1
2:1
2:1
2:1
7:1
1
2
5.0
1.0
5.0
1.5
1.5
5.0
5.0
1.0
1.0
1.0
100
93
47:3:1
17:1:12
16:30:1
65:5:1
3:1
3:1
6:1
6:1
7:1
THF
DME
3
6
dioxane
14
4
5d
62
toluene
60
100
5:1:0
1
a Determined by 1H NMR spectroscopy by observing consumption of 1.
a Pd2(dba)3 CHCl3/DPPM = 1:2.4. b Determined by H NMR spec-
3
b Determined by 1H NMR spectroscopy.
troscopy with the consumption of 1. c Determined by 1H NMR
spectroscopy. d Reaction was performed with 1.2 equiv of 1 at 80 °C
for 16 h to give 98% isolated yield of 3.
generated by simple deprotonation. Capturing this enolate with
eletrophilic partners such as aldehydes and imines has proven
possible.13 Encouraged by the success of these aldol-like pro-
cesses, we wondered what other carbonꢀcarbon bond forming
reactions could be accessed by trapping this nucleophilic all-
enoate intermediate. An alternative possible eletrophile is a
π-allylpalladium cationic species, which is obtained from an
alkene containing an allylic leaving group and a zerovalent palladi-
um catalyst. Nucleophiles such as alcohols, amines, enolates, and
malonates have been successfully employed in the palladium-
catalyzed allylic alkylation reactions to generate new allylic CꢀO,
CꢀN, or CꢀC bonds.14 The challenge, however, is to favor the
intermolecular reaction between these two catalytically gener-
ated intermediates, while disfavoring the two potential compet-
ing side reactions: (1) simple protonation9 of the vanadium
allenoate and (2) nucleophilic attack15 on the π-allylpalladium
unit by the starting propargyl alcohol. Herein, we describe our
findings related to the development of a novel vanadium- and
palladium-catalyzed coupling of propargyl alcohols and allylic
carbonates.
complex. The results indicated that the ligands played an
important role in the successful execution of the desired con-
temporaneous dual catalysis reaction. In the preliminary exam-
ination (Table 1), triphenylsilanol and DPPM (diphenylphos-
phinomethane) were found to be the most effective ligands
for the vanadium and palladium catalysts, respectively. In the
presence of 5 mol % OdV(OSiPh3)3 and 2.5 mol % Pd2-
(dba)3 CHCl3 with 6.0 mol % DPPM, the anticipated product
3
3 was isolated in 98% yield with good stereoselectivity (E:Z =
5:1) (entry 1). In contrast, all other phosphine ligands screened
in the reaction resulted in either low conversions or undesired
byproducts 416 and 5.17 For example, when phosphine ligands
were employed that retard the rate of π-allyl formation or the
rate of alkylation of the electrophile, enone 4 became the pre-
dominant product of the reaction (entries 2, 6, and 7). On the other
hand, when DPPP and DPPB were employed, the O-alkylation
process became much faster than the desired dual catalytic
transformation, affording the desired product 3 in low yield (entries
3 and 4). Furthermore, the more electron poor vanadium catalyst
did not perform as well as the original vanadium catalyst. In this
case, the unwanted allylated alcohol 5 was generated in larger
amounts (entry 8).
Standard Reaction Optimization. To test the feasibility of this
new proposal, we began by choosing propargyl alcohol 1 and allyl
carbonate 2 as the standard coupling partners, along with a catalyst
combination of OdV(OSiAr3)3 and a palladium(0)ꢀphosphine
12826
dx.doi.org/10.1021/ja204817y |J. Am. Chem. Soc. 2011, 133, 12824–12833

Journal of the American Chemical Society
ARTICLE
Table 4. Survey of the Allyl Carbonate Scope
Table 5. Survey of the Propargyl Alcohol Scope for Making
r,β-Unsaturated Ketones
a Double bond geometry was assigned by analogy with compound 3, and
E/Z ratio was determined by 1H NMR spectroscopy. b 1.5 mol %
Pd2(dba)3 CHCl3, 3.6 mol % DPPM, and 4.5 mol % OdV(OSiPh3)3
3
were used. c 1.5 equivalent of 1 was used and the reaction run for 48 h.
Having determined the optimal catalyst structures for the
reaction, we next studied the effects of solvent on the efficiency
and selectivity of the reaction (Table 2). Although several
solvents were suitable for this new process, 1,2-dichloroethane
(DCE) turned out to be the best solvent, giving the product in
higher yield and stereoselectivity with regard to double bond
geometry.
We next determined the effect of catalyst ratios and catalyst
loadings on the overall efficiency of the process. The results
summarized in Table 3 revealed that the performance of the dual
catalytic system was affected dramatically by the vanadium to
palladium ratio. For example, when an excess of the palladium
catalyst with respect to the vanadium catalyst was employed, the
O-allylated byproduct 5 was formed in greater amounts along
a Double bond geometry was assigned by analogy with compound 3, and
1
E/Z ratio was determined by H NMR spectroscopy. b Run for 48 h.
c 1.5 mol % Pd2(dba)3 CHCl3, 3.6 mol % DPPM, and 4.5 mol % OdV-
3
(OSiPh3)3 were used. d tert-Butyl cinnamyl carbonate was used instead
of 2. e 1.5 equiv of propargyl alcohol was used and the reaction run at
100 °C for 48 h.
12827
dx.doi.org/10.1021/ja204817y |J. Am. Chem. Soc. 2011, 133, 12824–12833

Journal of the American Chemical Society
ARTICLE
Scheme 4. A Comparison of the Two Standard Reaction Conditions
with the formation of product 3 (entry 2). Similarly, when the
vanadium catalyst was present in excess relative to palladium
catalyst, MeyerꢀSchuster rearrangement product 4 is generated
as the major product (entry 3). Further optimization revealed
that the reaction could proceed smoothly at fairly low catalyst
loadings (1.5% vanadium and 1% palladium) to yield 3 with
excellent isolated yield (98%) and double bond stereoselectivity
(E:Z = 7:1) (entry 5). The stereochemistry of compound 3 was
confirmed by 1H NMR nuclear Overhauser enhancement
(NOE) study. When irradiation was performed on the β-proton
in the major isomer of R,β-unsaturated ketone 3, no NOE was
observed for the allylic protons, so the major product was
assigned as E-configuration; when irradiation was performed
on the same β-proton in the minor isomer, +2.0% NOE was
observed for the allylic protons, so the minor product was then
assigned as the Z-configuration.
Reaction Scope of Allyl Carbonates. With the optimized
reaction conditions in hand, the reaction scope was examined
by varying the substitution patterns on both coupling part-
ners. First, various allylic carbonate substrates were tested in
the reaction using propargyl alcohol 1 as a model substrate
(Table 4). These studies confirmed that a broad range of
substitution patterns on the allylic carbonates were tolerated,
wherein the products were obtained in moderate to excellent
yields (57ꢀ98% yield) with high stereoselectivity (E:Z up to
12:1). Aryl (entries 2 and 3), heteroaryl (entries 5 and 6), and
alkyl (entry 8) substituents could be present at the terminal
position of the olefin. In addition, substitution on the internal
carbon of the olefin was tolerated (entry 7), as well as
substitution on the carbon atom adjacent to the oxygen
(entries 4 and 8).
Reaction Scope of Propargyl Alcohols (Preparation of r-
Allylated r,β-Unsaturated Ketones). Our next goal was to
investigate the scope of the reaction with regards to the pro-
pargylic alcohol (Table 5). Overall, most of the desired R-
allylated R,β-unsaturated ketones were prepared successfully,
affording good yields (up to 98%) with up to >19:1 (E:Z)
stereoselectivity. Remarkably, a wide range of substituents, with
varied steric and electronic properties, on the acetylene terminus
were allowed. Meanwhile, substrates bearing various conjugated
functional groups, including electron-rich and electron-poor
arenes (entries 2ꢀ4), heterocycles (entries 1 and 7), and olefins
(entry 4), on the propargylic position participated quite well in
the reaction. A tertiary propargylic alcohol was also employed,
and the corresponding product was isolated in good yield, albeit
with no stereocontrol of the double bond geometry (entry 8).
The reaction efficiency decreased significantly, however, when
the conjugated stabilizing groups were absent from the pro-
pargylic position. For instance, when both substituents on the
tertiary propargylic alcohol were alkyl groups, the reaction only
proceeded at higher temperature (100 °C) to give a moderate
yield of the product (entry 9). When just a single aliphatic group
was present at the alcohol center, only a trace amount of the
desired product was obtained under a variety of reactions
conditions (entry 10).
Reaction Scope of Propargyl Alcohols (Preparation of r-
Allylated r,β-Unsaturated Esters). One limitation of our
current methodology described above is the low reactivity of
propargylic alcohols lacking an aryl or vinyl group at the
propargylic center. As a solution to this problem, we envisioned
that introduction of a heteroatom such as oxygen atom at the
alkyne terminus of propargyl alcohol could activate the alcohol
toward MeyerꢀSchuster rearrangement. In addition, this simple
variation would allow us to obtain R-allylated R,β-unsaturated
esters as products. In order to test this hypothesis, we synthesized
propargyl alcohol 6, with the challenging double alkyl substitu-
tion pattern on the propargylic position, from commercially
available ethoxy acetylene and cyclohexanone via a one-step
addition reaction.11 Under the standard reaction conditions, the
desired R-allylated R,β-unsaturated ester 7 was obtained only in
trace amounts due to the rapid formation of enone 8 via a
MeyerꢀShuster rearrangement of alcohol 6 (Scheme 4a). As a
result, a preliminary optimization of the catalysts’ ratio and
loadings were performed. To our delight, the optimal condition
was identified quickly; in the presence of 0.5 mol % vanadium
and 2.5 mol % palladium, product 7 was prepared in excellent
isolated yield (86%) (Scheme 4b). Due to the rapid rate of
MeyerꢀShuster rearrangement of alcohol 6, it was necessary to
decrease the loading of the vanadium catalyst and increase the
palladium catalyst loading in order to obtain the desired coupling
product in good yield. Most importantly, the vanadium catalyst
must be added after the solvent was charged into the reaction
vessel, otherwise the high concentration neat propargyl alcohol 6
would be converted into side product 8 in a few minutes once it is
exposed to the vanadium catalyst. Significantly, these observa-
tions demonstrated that the interception of this vanadium ketene
acetal intermediate with electrophilic π-allylpalladium is far more
challenging than the previous cases. To the best of our knowl-
edge, this is the first example of a ketene acetal, generated from
the MeyerꢀSchuster rearrangement of the corresponding pro-
pargyl acohol, being trapped by an electrophile to form a new
carbonꢀcarbon bond.11,18
12828
dx.doi.org/10.1021/ja204817y |J. Am. Chem. Soc. 2011, 133, 12824–12833

Journal of the American Chemical Society
ARTICLE
Table 6. Survey of the Propargyl Alcohol Scope for Making
r,β-Unsaturated Esters
Table 7. Survey of the Allyl Carbonate Scope for Making r,β-
Unsaturated Esters
Next, we explored the scope of the ester-forming reaction with
respect to the allyl carbonates (Table 7). Gratifyingly, the allylic
carbonate coupling component could be substituted at the
terminus (entry 1) or the internal carbon (entry 4) of the olefin.
In addition, alkyl, aryl, and heteroaryl substituents are tolerated
on the terminus of the olefin (entries 1ꢀ3).
To exemplify the utility of the protocol, cholestane-derived
substrate 9 and estrone-derived substrate 11 were tested
(Scheme 5). We were pleased to find that these complex
substrates readily reacted when a combination of 0.5 mol %
vanadium and 5.0 mol % palladium catalysts was utilized at
80 °C for 16 h. From this result, we believe that the current
methodology could be an additional tool for the structural
modification of steroids and related natural products. In
addition, the geometry of the carbonꢀcarbon double bond
in product 12 was tentatively assigned as Z-configuration by
using a combination of two-dimensional NMR experiments
COSY, NOSY, HSQC, and HMBC.
Reaction Scope of Propargyl Alcohols (Preparation of r-
Allylated r,β-Unsaturated Amides). The generality of this
reactivity pattern was further explored using a nitrogen atom in
place of the oxygen atom on the terminus of the acetylene in the
propargyl alcohols under the same catalytic conditions de-
scribed above. Ynamide 13 was prepared by copper-catalyzed
coupling of N-benzylsulfonamide with 1-bromo-2-triisopropyl-
silylethyne, followed by desilylation, deprotonation with LDA,
and addition to 3-phenylpropanal.20,21 Illustrative examples of
combining the substrate 13 with three different allyl carbonates
1
a 2.0 equiv of propargyl alcohol was used. b Determined by H NMR
spectroscopy.
Encouraged by these positive findings, we next tested a variety
of propargylic alcohols bearing the terminal ether functionality
(Table 6). The dual catalytic transformation is competent across
a wide range of propargyl alcohols, including those substituted
with alkyl, aryl, and vinyl groups at the propargylic position. For
example, the five-, six-, and seven-membered ring systems could
be incorporated (entries 1ꢀ4). Furthermore, a noncyclic sub-
strate with disubstitution at the propargylic center (entry 5) and
substrates bearing aryl and vinyl groups reacted in high yield
(entries 6 and 7). Finally, the previously unreactive substrate
bearing a single alkyl substituent reacted successfully, and the
corresponding product was isolated in extremely high yield
(91%) (entry 8). The product in entry 6 is a known compound,
and its stereochemistry was assigned according to literature
data.19 The double bond geometries of products in entries 7
and 8 were then given by analogy with this compound.
12829
dx.doi.org/10.1021/ja204817y |J. Am. Chem. Soc. 2011, 133, 12824–12833

Journal of the American Chemical Society
ARTICLE
Scheme 5. Structural Modification of Steroid Derivatives
Table 8. Survey of the Propargyl Alcohol Scope for Making
r,β-Unsaturated Amides
Scheme 6. A Successful Example of Employing a Primary
Propargylic Alcohol
primary propargylic alcohol 14 was examined. Compound 14
was obtained by copper-catalyzed coupling of N-methylsulfo-
namide with ((3-bromoprop-2-yn-1-yl)oxy)(tert-butyl)diphenyl-
silane followed by desilylation.20,22 Under the standard reaction
condition, substrate 14 underwent efficient MeyerꢀSchuster
rearrangement and coupling with the π-allylpalladium inter-
mediate to generate the β,β-unsubstituted R,β-unsaturated
amide 15 in good yield (Scheme 6).
As demonstrated in the above sections, a large family of
R-allylated R,β-unsaturated ketones, esters, and amides were
successfully prepared by using a novel contemporaneous dual
catalytic system that employs both vanadium and palladium
catalysts. The unsaturated carbonyl compounds are versatile
building blocks for the synthesis of biologically active natural
products, pharmaceuticals, agrochemicals, fragrances, and other
useful fine chemicals. Our method offers a new one-step dis-
connection that uses propargyl alcohols and allyl carbonates that
could be accessed from widespread industrial raw materials like
alkynes, aldehydes, ketones, and allylic alcohols. Moreover, our
approach also exemplifies a high level of atom economy and is
more environmentally benign, generating tert-butanol and car-
bon dioxide as the only byproducts.
The value of the dual catalytic process is further amplified by
the control experiments using allyl halides as eletrophiles
(Scheme 7). When we attempted to trap the vanadiumꢀalleno-
ate derived from various propargyl alcohols with allyl bromide or
allyl iodide, none of anticipated R-allylated unsaturated carbonyl
compounds were observed. This observation demonstrates the
need for both catalysts in the reaction, and we believe that the
π-allylpalladium intermediate, being positively charged, is exclu-
sively able to trap the vanadium allenoate.
a Determined by 1H NMR spectroscopy.
are shown in Table 8. The catalytic results proved to be
promising, and all of the desired R-allylated R,β-unsaturated
amides were prepared in high yields. The stereochemistry of the
product in entry 1 was revealed by NOE study. When irradia-
tion was performed on the β-proton in the major isomer of this
R,β-unsaturated amide compound, no NOE was observed for
the allylic protons, so the major product was assigned the
E-configuration; when irradiation was performed on the same
β-proton in the minor isomer, +1.8% NOE was observed for
the allylic protons, so the minor product was then assigned the
Z-configuration. The double bond geometries of products in
entries 2 and 3 were then given by analogy with this compound.
To the best of our knowledge, this is the first example of a
ketene aminal generated from the MeyerꢀSchuster rearrange-
ment of the corresponding properly alcohol being trapped by
an electrophile to form a new carbonꢀcarbon bond.12 More-
over, the reaction scope was significantly expanded when a
12830
dx.doi.org/10.1021/ja204817y |J. Am. Chem. Soc. 2011, 133, 12824–12833

Journal of the American Chemical Society
ARTICLE
Scheme 7. Attempts To Trap VanadiumꢀAllenoate by Allyl
Scheme 8. Formation of Competing Side Products
Halides
Scheme 9. Survey of Byproducts 4 and 5 as Potential
Intermediates
O-allylated alcohol (viii) (cycle B0). These concerns were
supported by experimental results from the reactions in which a
single vanadium or palladium catalyst was employed (Scheme 8).
Under the standard reaction conditions, and in the presence of
only the vanadium catalyst, alcohol 1 underwent MeyerꢀSchuster
rearrangement to produce enone 4 exclusively (Scheme 8a).
In contrast, when only the palladium catalyst was employed,
O-allylated alcohol 5 was isolated in 85% yield (Scheme 8b).
Significantly, neither of the two reactions resulted in even a trace
amount of desired product 3.
We next considered the possibility that the dual catalytic
system might be proceeding through a consecutive two-step
process. In order to preclude this possibility, several control
experiments were conducted (Scheme 9). We found that enone 4
resulting from vanadium-catalyzed rearrangement of the pro-
pargylic alcohol was not converted to the desired R-allylated R,β-
unsaturated ketone 3 by the palladium catalyst (Scheme 9a). In
addition, the vanadium catalyst was not able to facilitate the
rearrangement of allylated alcohol 5 to product 3. Therefore, we
concluded that the contemporaneous dual catalysis indeed
involves the coupling of the in situ generated vanadiumꢀallenoate
and π-allylpalladium intermediates.
Once we had established that contemporaneous dual catalysis
was indeed responsible for product formation, we set out to
establish some general guiding principles for maintaining effi-
ciency in the reaction across substrate classes. It is critical to
harmonize the palladium and vanadium catalysts in order to favor
coupling of the two reactive intermediates. The OdV(OSiPh3)3
catalyst was sufficiently reactive to promote the MeyerꢀSchuster
rearrangement of propargyl alcohols containing certain func-
tional groups, including those substituted with an aromatic group
at the propargylic position or with a heteroatom at the terminus
of the triple bond. Importantly, the latter propargyl alcohols
could undergo rearrangement more readily and their corre-
sponding reactions were achieved with extremely low vanadium
catalyst loading (0.5 mol % V). On the other hand, the differences
Figure 1. Proposed mechanism for the contemporaneous dual catalysis.
Mechanistic Discussion. The proposed reaction mechanism
is outlined in Figure 1. The combination of vanadium catalysis
cycle (cycle A) and palladium catalysis cycle (cycle B), in a dual
catalysis manifold, leads to the formation of R-allylated R,β-
unsaturated carbonyl products vi via coupling of the reactive
intermediates from each cycle. For cycle A, the vanadium catalyst
enters by transesterification with the propargylic alcohol (ROH)
to generate vanadium ester iii. Ester iii is then rearranged via a
1,3-transposition of the oxygen atom to give vanadiumꢀallenoate
(iv). For cycle B, initial binding of palladium to the olefin of the
substrate, followed by ionization of the allylic carbonate, leads
to π-allylpalladium intermediate v. These two in situ generated
intermediates then intercept each other to yield the desired
product (vi) and regenerate the active vanadium and palladium
catalysts.
The challenge of contemporaneous dual catalysis is that the
coupling of the two reactive intermediates must occur in
preference to two competing side reactions. On one hand,
vanadium allenoate (iv) could be quenched via protonation by
propargyl alcohol (i) to generate enone (vii) (cycle A0). On the
other hand, π-allylpalladium intermediate (v) also has the
potential to be attacked by the propargylic alcohol to generate
12831
dx.doi.org/10.1021/ja204817y |J. Am. Chem. Soc. 2011, 133, 12824–12833

Journal of the American Chemical Society
ARTICLE
Scheme 10. Comparative Trials of the Variations on the
Vanadium and Palladium Loadings and Ratios
success is the need to control the relative catalyst loading for each
process. In addition, in spite of the need for a bimolecular process
between two catalytic intermediates, low catalyst loadings are still
possible. To the best of our knowledge, this is the first case of
contemporaneous dual catalyst being effected by two separate
metal catalysts. The ability of allenoates to function as suitable
nucleophiles toward π-allylpalladium complexes is also note-
worthy. Furthermore, the results reported herein represent the
first cases of using ester and amide enolates as nucleophiles in
Pd-catalyzed allylic alkylation. It also represents the first exam-
ples of such enolates being intercepted in a MeyerꢀSchuster type
process.
’ EXPERIMENTAL SECTION
Representative Procedure for the Preparation of r-Ally-
lated r,β-Unsaturated Ketones. To a dry V-shaped vial fitted with
a stirring bar were added allyl tert-butyl carbonate (31.6 mg, 0.20 mmol),
1-phenylhept-2-yn-1-ol (45.2 mg, 0.24 mmol), and OdV(OSiPh3)3
(2.7 mg, 0.003 mmol), and the vial was flushed with argon. In another
vial, Pd2(dba)3 CHCl3 (1.0 mg, 0.001 mmol) and DPPM (0.9 mg,
3
in reactivity of the allyl carbonates are consistent with the
difficulties of the formation of the corresponding π-allylpalla-
dium species. If an allyl carbonate is difficult to activate, higher
palladium catalyst loadings were generally required. For example,
the sterically more demanding substrate 1,3-dimethylallyl car-
bonate needed 3 mol % palladium (Table 4, entry 8), in contrast
to the parent standard substrate allyl carbonate that required only
1 mol % palladium (Table 4, entry 1). On the basis of the
appropriate adjustments of catalysts loading and ratio, a large
number of interesting R-allylated R,β-unsaturated ketones, es-
ters, and amides were prepared. The importance of these
parameters was further emphasized by a series of comparative
trials (Scheme 10). When catalyst loading ([V]/[Pd]) was
increased from 4.5/3.0 to 5.0/5.0 mol % while maintaining other
condition constants for substrates 1 and 16, the yield of desired
product was increased to 97% from 88% without erosion of
stereoselectivity (Scheme 10a). Alternatively, when the palla-
dium catalyst loading for substrate 18 was decreased from 2.5 to
1.0 mol %, the yield of desired product 19 was strikingly
decreased (88% versus 27%) (Scheme 10b). Also, when the
vanadium catalyst loading was increased from 0.5 mol % to
1.0 mol % for substrate 13, the yield of desired product 20 was
dramatically reduced to 10% (versus 85%) (Scheme 10c). These
results establish the need to balance the catalyst loadings and
ratios, depending on the inherent reactivity of the allylic carbon-
ate or propargylic alcohol starting materials.
0.0024 mmol) were dissolved in dry DCE (0.2 mL) and stirred for
10 min. The freshly made palladium catalyst solution was transferred
into the V-shaped vial through a cannula. The vial was then sealed with a
Teflon cap under an argon stream and heated at 80 °C for 16 h. The
crude reaction mixture was then subjected to a silica gel column using
dichloromethane/petroleum ether (1:2) as an eluent to afford 4-benzy-
lidenenon-1-en-5-one (44.8 mg, 98% yield, E:Z = 7:1). The sample for
analytical data was obtained by using preparative TLC. E-isomer:
Colorless oil. 1H NMR (400 MHz, CDCl3): δ 7.63 (s, 1H),
7.44ꢀ7.32 (m, 5H), 6.01ꢀ5.92 (m, 1H), 5.09ꢀ5.00 (m, 2H), 3.28 (t,
J = 4.0 Hz, 2H), 2.80 (t, J = 7.6 Hz, 2H), 1.73ꢀ1.63 (m, 2H), 1.43ꢀ1.34
(m, 2H), 0.95 (t, J = 7.2 Hz, 3H). 13C NMR (CDCl3, 100 MHz): δ
202.05, 139.80, 139.09, 135.95, 135.58, 129.33, 128.77, 128.53, 115.59,
37.61, 30.83, 27.06, 22.55, 14.02. IR: 2956, 1667, 1447, 1170, 915, 753,
698. HRMS (ESI): m/z calculated for C16H21O [M + 1]+ 229.1592,
observed 229.1587. Z-isomer: Colorless oil. 1H NMR (400 MHz,
CDCl3): δ 7.34ꢀ7.16 (m, 3H), 7.17 (d, J = 8.0 Hz, 2H), 6.64 (s,
1H), 5.91ꢀ5.80 (m, 1H), 5.17ꢀ5.12 (m, 2H), 3.10 (dd, J = 6.8, 1.2 Hz,
2H), 2.24 (t, J = 7.2 Hz, 2H), 1.47ꢀ1.43 (m, 2H), 1.16ꢀ1.11 (m, 2H),
0.76 (t, J = 7.2 Hz, 3H). 13C NMR (CDCl3, 100 MHz): δ 210.20, 142.94,
136.39, 134.70, 130.48, 128.50, 128.42, 127.98, 117.69, 42.88, 39.52,
26.06, 22.12, 13.82. IR: 2957, 1689, 1377, 920, 754, 699.
Representative Procedure for the Preparation of r-Ally-
lated r,β-Unsaturated Esters. To a dry V-shaped vial fitted with a
stirring barwere added allyl tert-butyl carbonate (31.6 mg, 0.20 mmol)
and 1-(ethoxyethynyl)cyclohexanol (40.4 mg, 0.24 mmol), and the vial
was flushed with argon. In another vial, Pd2(dba)3 CHCl3 (2.6 mg,
3
0.0025 mmol) and DPPM (2.3 mg, 0.006 mmol) were dissolved in dry
DCE (0.2 mL) and stirred for 10 min. The freshly made palladium
catalyst solution was transferred into the V-shaped vial through a
cannula. Catalyst OdV(OSiPh3)3 (0.9 mg, 0.001 mmol) was added,
and the vial was then sealed with a Teflon cap under an argon stream
and heated at 80 °C for 16 h. The crude reaction mixture was then
subjected to a silica gel column using dichloromethane/petroleum
ether (1:2) as an eluent to afford ethyl 2-cyclohexylidenepent-4-enoate
(36.0 mg, 86% yield) as a colorless oil. 1H NMR (400 MHz, CDCl3): δ
5.80ꢀ5.76 (m, 1H), 5.06ꢀ4.97 (m, 2H), 4.18 (q, J = 7.2 Hz, 2H), 3.05
(d, J = 6.0 Hz, 2H), 2.45 (t, J = 4.8 Hz, 2H), 2.21 (t, J = 4.8 Hz, 2H),
1.62ꢀ1.52 (m, 6H), 1.28 (t, J = 7.2 Hz, 3H). 13C NMR (CDCl3, 100
MHz): δ 170.10, 149.33, 135.99, 122.59, 115.16, 60.23, 33.60, 32.78,
31.38, 28.34, 28.14, 26.55, 14.36. IR: 2977, 2926, 2853, 1712, 1637,
1446, 1281, 1198, 1137, 1097, 1028, 992, 911. HRMS (ESI): m/z
calculated for C13H21O2 [M + 1]+ 209.1542, observed 209.1536.
’ CONCLUSION
In conclusion, we have disclosed a new mode of catalysis,
which we call contemporaneous dual catalysis. This catalytic
method features two reactive chemical intermediates, which are
generated by two independent catalysts from their respective
substrates. These reactive intermediates then preferentially un-
dergo bimolecular coupling in the presence of stochiometric
reagents that are able to quench both active moieties. The current
example was carried out via a combination of a vanadium-
catalyzed 1,3-transposition of propargylic alcohols and a palla-
dium-catalyzed alkylation of allylic carbonates. Two competing
side reactions, including trapping the vanadiumꢀallenoate and
π-allylpalladium intermediates with a large excess of propargyl
alcohol, were effectively overcome. An important aspect for
12832
dx.doi.org/10.1021/ja204817y |J. Am. Chem. Soc. 2011, 133, 12824–12833

Journal of the American Chemical Society
ARTICLE
Representative Procedure for the Preparation of r-Ally-
lated r,β-Unsaturated Amides. To a dry V-shaped vial fitted with a
stirring bar were added allyl tert-butyl carbonate (31.6 mg, 0.20 mmol)
and N-benzyl-N-(3-hydroxy-5-phenylpent-1-yn-1-yl)-4-methylbenze-
nesulfonamide (167.8 mg, 0.40 mmol), and the vial was flushed with
(5) For books on catalytic methods, see:(a) Hartwig, J. F. Organo-
transition Metal Chemistry: From Bonding to Catalysis; University Science
Books: Mill Valley, CA, 2010. (b) Organocatalysis; Reetz, M. T., List, B.,
Jaroch, S., Weinmann, H., Eds.; Springer: Berlin, 2008.
(6) For reviews on bifunctional catalysis systems, see:(a) Lee, J. M.;
Na, Y.; Han, H.; Chang, S. Chem. Soc. Rev. 2004, 33, 302. (b) Shibasaki,
M.; Kanai, M.; Matsunaga, S.; Kumagai, N. Acc. Chem. Res. 2009,
42, 1117. (c) Shao, Z.; Zhang, H. Chem. Soc. Rev. 2009, 38, 2745.
(d) Zhou, J. Chem. Asian J. 2010, 5, 422.
(7) For selected examples on dual catalysis, see:(a) Sawamura, M.;
Sudoh, M.; Ito, Y. J. Am. Chem. Soc. 1996, 118, 3309. (b) Trost, B. M.;
McEachern, E. J.; Toste, F. D. J. Am. Chem. Soc. 1998, 120, 12702.
(c) Nakoji, M.; Kanayama, T.; Okino, T.; Takemoto, Y. Org. Lett. 2001,
3, 3329. (d) Kamijo, S.; Yamamoto, Y. Angew. Chem., Int. Ed. 2002,
41, 3230. (e) Sammis, G. M.; Danjo, H.; Jacobsen, E. N. J. Am. Chem. Soc.
2004, 126, 9928. (f) Ibrahem, I.; Cꢀordova, A. Angew. Chem., Int. Ed.
2006, 45, 1952. (g) Komanduri, V.; Krische, M. J. J. Am. Chem. Soc. 2006,
128, 16448. (h) Mukherjee, S.; List, B. J. Am. Chem. Soc. 2007,
129, 11336. (i) Nicewicz, D. A.; MacMillan, D. W. C. Science 2008,
322, 77. (j) Hu, W.; Xu, X.; Zhou, J.; Liu, W.; Huang, H.; Hu, J.; Yang, L.;
Gong, L. J. Am. Chem. Soc. 2008, 130, 7782. (k) Shi, Y.; Roth, K. E.;
Ramgren, S. D.; Blum, S. A. J. Am. Chem. Soc. 2009, 131, 18022. (l) Ikeda,
M.; Miyake, Y.; Nishibayashi, Y. Angew. Chem., Int. Ed. 2010, 49, 7289.
(8) Trost, B. M.; Luan, X. J. Am. Chem. Soc. 2011, 133, 1706.
(9) Pauling, H.; Andrews, D. A.; Hindley, N. C. Helv. Chem. Acta
1976, 59, 1233.
argon. In another vial, Pd2(dba)3 CHCl3 (2.6 mg, 0.0025 mmol) and
3
DPPM (2.3 mg, 0.006 mmol) were dissolved in dry DCE (0.2 mL) and
stirred for 10 min. The freshly made palladium catalyst solution was
transferred into the V-shaped vial through cannula. Catalyst OdV-
(OSiPh3)3 (0.9 mg, 0.001 mmol) was added, and the vial was then sealed
with a Teflon cap under an argon stream and heated at 80 °C for 16 h.
The crude reaction mixture was then subjected to a silica gel column
using dichloromethane/petroleum ether (2:1) as an eluent to afford
2-allyl-N-benzyl-5-phenyl-N-tosylpent-2-enamide (78.0 mg, 85% yield,
E:Z = 2:1) as a colorless oil. Due to the existence of two isomers, two sets
of NMR signals were observed. 1H NMR (400 MHz, CDCl3): δ 7.63 (d,
J = 8.3 Hz, 2H), 7.54 (d, J = 8.3 Hz, 2H), 7.22ꢀ7.09 (m, 20H), 7.03 (dd,
J = 7.8, 0.8 Hz, 2H), 6.94ꢀ6.93 (m, 2H), 5.95 (t, J = 7.2 Hz, 1H),
5.55ꢀ5.47 (m, 2H), 5.33ꢀ5.30 (m, 1H), 4.93ꢀ4.78 (m, 6H), 4.69 (s,
2H), 2.83ꢀ2.81 (m, 2H), 2.55 (t, J = 7.7 Hz, 4H), 2.44 (t, J = 7.8 Hz,
2H), 2.37ꢀ2.33 (m, 8H), 1.87ꢀ1.86 (m, 2H). 13C NMR (CDCl3, 100
MHz): δ 172.58, 171.24, 144.49, 140.95, 139.50, 136.31, 135.22, 134.84,
134.28, 133.99, 131.13, 129.49, 129.36, 128.80, 128.67, 128.61, 128.58,
128.55, 128.49, 128.45, 128.43, 128.40, 128.36, 128.14, 127.78, 127.73,
126.25, 117.88, 116.53, 50.88, 50.14, 37.2, 35.07, 34.55, 32.51, 31.02,
30.07, 21.73, 21.71. IR: 3064, 3029, 2918, 2850, 1685, 1357, 1166, 1088,
699, 664. HRMS (ESI): m/z calculated for C28H30NO3S [M + 1]+
460.1946, observed 460.1942.
(10) For a review on the MeyerꢀSchuster rearrangement, see:Engel,
D.; Dudley, G. B. Org. Biomol. Chem. 2009, 7, 4149.
(11) (a) Engel, D.; Dudley, G. B. Org. Lett. 2006, 8, 4027. (b) Lopez,
S. S.; Engel, D. A.; Dudley, G. B. Synlett. 2007, 949. (c) Engel, D. A.;
Lopez, S. S.; Dudley, G. B. Tetrahedron 2008, 64, 6988.
(12) Egi, M.; Yamaguchi, Y.; Fujiwara, N.; Akai, S. Org. Lett. 2008,
10, 1867.
’ ASSOCIATED CONTENT
(13) (a) Trost, B. M.; Oi, S. J. Am. Chem. Soc. 2001, 123, 1230.
(b) Trost, B. M.; Chung, C. K. J. Am. Chem. Soc. 2006, 128, 10358.
(14) (a) Consiglio, G.; Waymouth, R. M. Chem. Rev. 1989, 89, 257.
(b) Trost, B. M.; Van Vranken, D. L. Chem. Rev. 1996, 96, 395. (c) Trost,
B. M. Acc. Chem. Res. 2002, 35, 695.
S
Supporting Information. Experimental procedures and
b
analytical data of new products. This material is available free of
charge via the Internet at http://pubs.acs.org.
’ AUTHOR INFORMATION
(15) Haight, A. R.; Stoner, E. J.; Peterson, M. J.; Grover, V. K. J. Org.
Chem. 2003, 68, 8092.
(16) Iwasawa, N.; Maeyama, K.; Saitou, M. J. Am. Chem. Soc. 1997,
119, 1486.
Corresponding Author
bmtrost@stanford.edu
ꢀ
(17) Sanz, R.; Martínez, A.; Alvarez-Gutiꢀerrez, J. M.; Rodríguez, F.
Eur. J. Org. Chem. 2006, 1383.
(18) (a) Sugawara, Y.; Yamada, W.; Yoshida, S.; Ikeno, T.; Yamada,
T. J. Am. Chem. Soc. 2007, 129, 12902. (b) Rieder, C. J.; Winberg, K. J.;
West, F. G. J. Am. Chem. Soc. 2009, 131, 7504. (c) Rieder, C. J.; Winberg,
K. J.; West, F. G. J. Org. Chem. 2011, 76, 50.
(19) (a) Mali, R. S.; Babu, K. N. Helv. Chim. Acta 2002, 85, 3525.
(b) Ferguson, M. L.; Senecal, T. D.; Groendyke, T. M.; Mapp, A. K.
J. Am. Chem. Soc. 2006, 128, 4576.
’ ACKNOWLEDGMENT
We thank the NSF (CHE 0948222) for financial support of
this project. X.L. is grateful for a Swiss National Science
Foundation postdoctoral fellowship. Dr. David Michaelis is
acknowledged for proofreading the manuscript.
(20) Tracey, M. R.; Zhang, Y.; Frederick, M. O.; Mulder, J. A.;
Hsung, R. P. Org. Lett. 2004, 6, 2209.
’ REFERENCES
(21) (a) Couty, S.; Meyer, C.; Cossy, J. Angew. Chem., Int. Ed. 2006,
45, 6726. (b) Rodríguez, D.; Martínez-Esperꢀon, M. F.; Castedo, L.; Saꢀa,
C. Synlett. 2007, 1963.
(22) (a) Barbazanges, M.; Meyer, C.; Cossy, J. Org. Lett. 2007,
9, 3245. (b) Barbazanges, M.; Meyer, C.; Cossy, J. Tetrahedron. 2008,
49, 2902.
(1) (a) Trost, B. M. Angew. Chem., Int. Ed. Engl. 1995, 34, 259. (b)
Anastas, P. T.; Warner, J. C. Green Chemistry Theory and Practice; Oxford
University Press: New York, 1998. (c) Li, C.; Trost, B. M. Proc. Natl.
Acad. Sci. U.S.A. 2008, 105, 13197.
(2) Trost, B. M. Science 1991, 254, 1471.
(3) (a) Trost, B. M. Science 1983, 219, 245. (b) Shenvi, R. A.;
O’Malley, D. P.; Baran, P. S. Acc. Chem. Res. 2009, 42, 530.
(4) For reviews on atom-economic reactions, see:(a) Trost, B. M.;
Toste, F. D.; Pinkerton, A. B. Chem. Rev. 2001, 101, 2067. (b) Trost,
B. M. Acc. Chem. Res. 2002, 35, 695. (c) Trost, B. M.; Frederiksen, M. U.;
Rudd, M. T. Angew. Chem., Int. Ed. 2005, 44, 6630. (d) Li, C. Acc. Chem.
Res. 2010, 43, 581. (e) Basavaiah, D.; Reddy, B. S.; Badsara, S. S. Chem.
Rev. 2010, 110, 5447. (f) Aubert, C.; Fensterbank, L.; Garcia, P.;
Malacria, M.; Simonneau, A. Chem. Rev. 2011, 111, 1954. (g) Corma,
A.; Leyva-Prez, A.; Sabater, M. J. Chem. Rev. 2011, 111, 1657.
12833
dx.doi.org/10.1021/ja204817y |J. Am. Chem. Soc. 2011, 133, 12824–12833