Development of a general Pd(II)-catalyzed intermolecular hydroalkoxylation reaction of vinylphenols by using a sacrificial alcohol as the hydride source

Article abstract of DOI:10.1021/ol062222t

A general hydroalkoxylation of vinylphenols has been successfully developed wherein primary, secondary, and tertiary alcohols can be used as nucleophiles. The key conceptual breakthrough is the use of sec-phenethyl alcohol at relatively low concentrations as the sacrificial alcohol to undergo oxidation and provide the proposed Pd-H intermediate.

Full text of DOI:10.1021/ol062222t

ORGANIC
LETTERS
2006
Vol. 8, No. 24
5557-5560
Development of a General
Pd(II)-Catalyzed Intermolecular
Hydroalkoxylation Reaction of
Vinylphenols by Using a Sacrificial
Alcohol as the Hydride Source
Yang Zhang and Matthew S. Sigman*
Department of Chemistry, UniVersity of Utah, 315 South 1400 East,
Salt Lake City, Utah 84112
sigman@chem.utah.edu
Received September 7, 2006
ABSTRACT
A general hydroalkoxylation of vinylphenols has been successfully developed wherein primary, secondary, and tertiary alcohols can be used
as nucleophiles. The key conceptual breakthrough is the use of sec-phenethyl alcohol at relatively low concentrations as the sacrificial
alcohol to undergo oxidation and provide the proposed Pd−H intermediate.
Transition-metal-catalyzed direct addition of N-H or O-H
bonds across olefins represents an attractive method for
C-N1 or C-O2 bond construction. Although significant
advancements have been reported in the area of metal-
catalyzed hydroamination processes,3 transition-metal-
catalyzed hydroalkoxylation reactions are much less common
and remain a challenging transformation presumably due to
the weaker Lewis basicity and the decreased nucleophilicity
of oxygen nucleophiles, especially employing simple alco-
hols.4,5 Recently, as part of our program aimed at the
development of Pd(II)-catalyzed olefin functionalization
reactions, we discovered a hydroalkoxylation reaction of
o-vinylphenols (eq 1).6 Initial mechanistic studies suggest a
unique process in which a Pd(II)-catalyzed alcohol oxidation
of the solvent was proposed to generate a Pd(II)-hydride B,
followed by olefin insertion and formation of an o-quinone
methide intermediate D (Scheme 1). The addition of a second
equivalent of the alcohol solvent to the o-quinone methide
intermediate is proposed to lead to product formation.
(1) For recent reviews, see: (a) Hultzsch, K. C. AdV. Synth. Catal. 2005,
347, 367. (b) Beller, M.; Seayad, J.; Tillack, A.; Jiao, H. Angew. Chem.,
Int. Ed. 2004, 43, 3368. (c) Hong, S.; Marks, T. J. Acc. Chem. Res. 2004,
37, 673.
(2) For recent reviews, see: (a) Messerle, B. A.; Vuong, K. Q. Pure
Appl. Chem. 2006, 78, 385. (b) Wolfe, J. P.; Thomas, J. S. Curr. Org. Chem.
2005, 9, 625.
(3) For recent examples, see: (a) Johns, A. M.; Utsunomiya, M.;
Incarvito, C. D.; Hartwig, J. F. J. Am. Chem. Soc. 2006, 128, 1828. (b)
Patil, N. T.; Lutete, L. M.; Wu, H.; Pahadi, N. K.; Gridnev, I. D.; Yamamoto,
Y. J. Org. Chem. 2006, 71, 4270. (c) Hii, K. K. Pure Appl. Chem. 2006,
78, 341. (d) Abbiati, G.; Beccalli, E.; Broggini, G.; Martinelli, M.; Paladino,
G. Synlett 2006, 73.
(4) For intramolecular hydroalkoxylation reactions, see: (a) Oe, Y.;
Ohta, T.; Ito, Y. Synlett 2005, 179. (b) Taylor, J. G.; Whittall, N.; Hii, K.
K. Chem. Commun. 2005, 5103. (c) Grant, V. H.; Liu, B. Tetrahedron Lett.
2005, 46, 1237. (d) Coulombel, L.; Favier, I.; Dunach, E. Chem. Commun.
2005, 2286. (e) Yang, C. G.; Reich, N. W.; Shi, Z.; He, C. Org. Lett. 2005,
7, 4553. For intermolecular hydroalkoxylation reactions, see: (f) Qian, H.;
Han, X.; Widenhoefer, R. A. J. Am. Chem. Soc. 2004, 126, 9536. (g)
Matsukawa, Y.; Mizukado, J.; Quan, H.; Tamura, M.; Sekiya, A. Angew.
Chem., Int. Ed. 2005, 44, 1128. (h) Yang, C. G.; He, C. J. Am. Chem. Soc.
2005, 127, 6966.
10.1021/ol062222t CCC: $33.50
© 2006 American Chemical Society
Published on Web 10/26/2006

Scheme 1. Mechanistically Inspired Approach to an Improved
Pd(II)-Catalyzed Hydroalkoxylation of o-Vinylphenols
Table 1. Evaluation of the Solvent and Sacrificial Alcohol in
the Hydroalkoxylation of 2-Propenyl Phenol with MeOH
entry cosolvent
ROH
MeOH/ROH
1i/1ii/1iiib
1
2
3
4
5
6
7
8c
9
-
-
-
-
(CH3)2CHOH
(CH3)2CHOH
(CH3)2CHOH
(CH3)2CHOH
(CH3)2CHOH
(CH3)2CHOH
PhCh(OH)CH3 10 equiv:2 equiv
PhCh(OH)CH3 10 equiv:2 equiv
PhCh(OH)CH3 4 equiv:2 equiv
1:7 (M/M)
1:1 (M/M)
5:1 (M/M)
100:0 (M/M)
10 equiv:2 equiv
10 equiv:2 equiv
1:1:0
3:1:0
15:1:0
0:0:>99
12:1:0
14:1:0
>99:0:0
>99:0:0
82:18:0
CH2Cl2
toluene
CH2Cl2
CH2Cl2
CH2Cl2
Two issues limit the application of the current method:
(1) only alcohols readily oxidized by Pd(II) are able to gen-
erate the proposed Pd-H intermediate,7 therefore alcohols
such as methanol or tertiary alcohols fail to yield the hydro-
alkoxylation product, and (2) the reaction utilizes a nucleo-
philic alcohol as solvent, which clearly prohibits the use of
precious alcohols or alcohols in the solid state. To address
these issues, we sought to develop a Pd(II)-catalyzed hydro-
alkoxylation reaction that exploits the use of a sacrificial
alcohol to undergo Pd(II)-catalyzed alcohol oxidation pro-
moting the formation of the requisite Pd-H intermediate B
(Scheme 1). It was envisioned that using a hindered readily
oxidized sacrificial alcohol in combination with a nucleo-
philic alcohol in higher concentrations would facilitate the
hydroalkoxylation of a diverse set of alcohol substrates.
Herein, we report a general hydroalkoxylation reaction of
o-vinylphenols using sec-phenethyl alcohol as the sacrificial
alcohol and offer new insights into the mechanistic features
of the reaction.
Considering that secondary alcohols are more hindered
nucleophiles and are also generally excellent substrates for
Pd[(-)-sparteine]Cl2-catalyzed alcohol oxidations,7a we rea-
soned they would be a logical choice for a sacrificial alcohol.
To test this hypothesis, 1a was subjected to Pd(II)-catalyzed
hydroalkoxylation reaction conditions using mixtures of
2-propanol and methanol as the solvent (entries 1-3, Table
1). It was found that using a 5:1 methanol/2-propanol mixture
resulted in high selectivity (94%) for the hydroalkoxylation
product 1i derived from the addition of methanol (entry 3).
It should be noted that without added 2-propanol only
dialkoxylation8 of the olefin occurred, alluding to the
necessity of a readily oxidizable alcohol to obtain the
hydroalkoxylation product (entry 4).7e,f Although these
experiments provide a proof of concept, the use of the
nucleophile as the solvent still limits the scope of the reaction.
Therefore, mixtures of methanol and several secondary
alcohols in various solvents were examined (entries 5-9).9
It was found that when CH2Cl2 or toluene was utilized as
the cosolvent significantly lower levels of both the sacrificial
alcohol and the alcohol nucleophile could be used (entries 5
and 6). Moreover, sec-phenethyl alcohol is an excellent
sacrificial alcohol at low concentrations where only product
1i is observed (entry 7). Under these reaction conditions, a
balloon of air can be used in lieu of O2 (entry 8).10
a 5 mol % of Pd[(-)-sparteine]Cl2, 10 mol % of CuCl2, 3 Å MS, O2, rt.
b Ratios were determined by GC yields using 5-nonanone as the internal
standard. c A balloon of air was used.
Using sec-phenethyl alcohol as the sacrificial alcohol with
CH2Cl2 as the cosolvent, we explored the scope of this
reaction with various vinylphenols and methanol as the
nucleophilic alcohol (Table 2, entries 1-7). Both electron-
rich and electron-poor substrates lead to good to excellent
isolated yields of methanol hydroalkoxylation products.
Interestingly, low enantiomeric excesses were measured for
some of the hydroalkoxylation products implying that
asymmetric catalysis may be possible (up to 9% ee for 3e).9
This is in contrast to our previous report.6
To further evaluate the scope, diverse alcohols were sub-
jected to the reaction conditions using vinylphenol 1b and 2
equiv of sec-phenethyl alcohol as the sacrificial alcohol
(Table 2, entries 8-19). Secondary alcohols were found to
be excellent substrates (entries 8-12), and using enantio-
merically enriched secondary alcohols, we observed modest
diastereomeric ratios (entries 11-13: 4d, dr 1.4:1; 4e, dr
1.9:1; and 4f, dr 2.0:1).9 It is interesting to consider that even
large alcohols such as norborneol and menthol are success-
fully added to 1b in the presence of sec-phenethyl alcohol
(entries 10 and 13). This result is attributed to the lower
concentration of sec-phenethyl alcohol relative to the nu-
cleophile.
(5) Hartwig and co-workers have recently reported that metal-triflate-
catalyzed hydroalkoxylation reactions may proceed by simple Brønsted acid
catalysis. See: Rosenfeld, D. C.; Shekhar, S.; Takemiya, A.; Utsunomiya,
M.; Hartwig, J. F. Org. Lett. 2006, 8, 4179-4282.
(6) Gligorich, K. M.; Schultz, M. J.; Sigman, M. S. J. Am. Chem. Soc.
2006, 128, 2794.
(7) For reviews of Pd(II)-catalyzed aerobic alcohol oxidation, see: (a)
Sigman, M. S.; Jensen, D. R. Acc. Chem. Res. 2006, 39, 221. (b) Stahl, S.
S. Angew. Chem., Int. Ed. 2004, 43, 3400. (c) Stoltz, B. M. Chem. Lett.
2004, 33, 362. (d) Muzart, J. Tetrahedron 2003, 59, 5789. For difficulties
in the oxidation of methanol with Pd(II) catalysts, see: (e) Lloyd, W. G. J.
Org. Chem. 1967, 32, 2816. (f) Nishimura, T.; Kakiuchi, N.; Onoue, T.;
Ohe, K.; Uemura, S. J. Chem. Soc., Perkin Trans. 1 2000, 1915.
(8) Schultz, M. J.; Sigman, M. S. J. Am. Chem. Soc. 2006, 128, 1460.
(9) See Supporting Information for details.
(10) The use of Pd[(-)-sparteine]Cl2 as the catalyst results in a
simultaneous oxidative kinetic resolution of sec-phenethyl alcohol with krel
values ∼10. See ref 7 for details.
5558
Org. Lett., Vol. 8, No. 24, 2006

Table 2. Substrate Scope
a Average isolated yield of two reactions. b Toluene was used as solvent, and the reaction was carried out at 45 °C.
Excitingly, tert-butanol was an effective nucleophile for
the transformation (entry 14), which, to the best of our
knowledge, represents the first example of a metal-catalyzed
addition of a tertiary alcohol across an olefin. This result
suggests that the size of the nucleophilic alcohol will have
minimal impact on the future scope of this transformation.
In contrast to our previous report, benzyl alcohol can now
be utilized as a substrate leading to a high yield of the
corresponding hydroalkoxylation product (entry 15). Alco-
holic nucleophiles containing additional functional groups
that may chelate or inhibit catalysis were also competent for
the hydroalkoxylation reaction under slightly modified reac-
tion conditions (entries 17-19).
After exploring the scope of the hydroalkoxylation reac-
tion, we wanted to confirm our mechanistic rationale in
designing this significantly improved method. Specifically,
is the sacrificial alcohol acting as a hydride source in the
hydroalkoxylation process? In the absence of the sacrificial
alcohol and using only MeOH under the standard conditions,
a <5% conversion of the starting material and no hy-
droalkoxylation product are observed which is consistent with
the requirement of an alcohol that readily undergoes oxida-
tion (eq 2). Additionally, methanol is not the source of the
hydride incorporated into the product considering that the
use of CD3OD does not result in any deuterium incorporation
into the olefin skeleton of 3h (eq 3). This also rules out the
action of Brønsted acid catalysis in the hydroalkoxylation
process. As previously observed using CD3CD2OD as the
solvent,6 the use of PhCD(OH)CH3 as the sacrificial alcohol
results in two isotopomers 3i and 3j in a 2.5:1 ratio,
consistent with the proton incorporated into the product being
derived from the oxidation of sec-phenethyl alcohol (eq 4).11
As an additional experiment, the number of millimoles of
sacrificial alcohol oxidized per millimole of product formed
(11) The mixture of isotopomers is proposed to result from reversible
hydride insertion (see ref 6). A total of 91% incorporation of a single
deuterium atom is observed and consistent with reversible hydride insertion
that does not include dissociation of the olefin.
Org. Lett., Vol. 8, No. 24, 2006
5559

was subjected to the reaction conditions using 1b as the
substrate in the absence of a nucleophilic alcohol (eq 5). The
endo-cycloaddition product was isolated in 19% yield as a
single diastereomer, and the yield of the cycloaddition
product could be increased to 30% when 10% NEt3 was
added to the reaction. These results provide strong evidence
for the intermediacy of an o-quinone methide under the
hydroalkoxylation reaction conditions. Of particular interest,
the use of styrenyl precursors to generate o-quinone methides
is quite rare and typically requires forcing thermal conditions.
Additionally, o-quinone methides derived from vinylphenols
generally undergo dimerization reactions.13 Therefore, the
mild conditions under which we are able to generate and
intercept o-quinone methide intermediates from vinylphenols
may have future synthetic utility.
In conclusion, a general hydroalkoxylation of vinylphenols
has been successfully developed wherein primary, secondary,
and tertiary alcohols can be used as nucleophiles. The key
conceptual breakthrough is the use of sec-phenethyl alcohol
at relatively low concentrations as the sacrificial alcohol to
undergo oxidation and provide the proposed Pd(II)-H
intermediate. Isotopic labeling studies confirm that the
hydrogen atom incorporated into the product is derived from
oxidation of the sacrificial alcohol, and for every alcohol
oxidation, approximately 1 equiv of olefin is converted to
product. Evidence for a well-defined o-quinone methide
intermediate was garnered through trapping the intermediate
via a cycloaddition reaction with an enol ether. Future work
will focus on exploiting the coupling of Pd(II)-catalyzed
alcohol oxidations with new olefin functionalization pro-
cesses and generating o-quinone methide intermediates under
mild conditions.
was measured. To achieve this, aliquots of the hydroalkoxy-
lation reaction of o-vinylphenol with methanol were sampled
over a period of time and analyzed by GC. The GC yield of
acetophenone and the GC yield of hydroalkoxylation product
3b were measured for each aliquot. By correlating the
number of millimoles of each component at a given time, a
linear relationship was observed (Figure 1). The slope (0.93)
Figure 1. Correlation of the number of millimoles of the sacrificial
alcohol oxidized per millimole of the product formed.
Acknowledgment. This work was supported by the
National Institutes of Health (NIGMS RO1 GM3540). M.S.S.
thanks the Dreyfus Foundation (Teacher-Scholar) and Pfizer
for their support. We are grateful to Johnson Matthey for
the gift of various Pd salts. We thank Keith Gligorich at the
University of Utah for initiating this work and insightful
discussions.
of the line indicates that for every molecule of sec-phenethyl
alcohol oxidized nearly 1 equiv of olefin is converted to
product. This result highlights why low levels of the
sacrificial alcohol can be successfully used in this reaction
and bodes well for the use of a sacrificial alcohol as a hydride
source in related olefin functionalization reactions.
The final question we wanted to address is whether the
hydroalkoxylation product does indeed arise from a well-
defined quinone methide intermediate.12 To probe this possi-
bility, 2 equiv of ethyl vinyl ether, a commonly used dieno-
phile in cycloaddition reactions with o-quinone methides,13
Supporting Information Available: Experimental pro-
cedures and full spectroscopic data for all new compounds.
This material is available free of charge via the Internet at
http://pubs.acs.org.
OL062222T
(13) (a) Nakatani, K.; Higashida, N.; Saito, I. Tetrahedron Lett. 1997,
38, 5005. (b) Arduini, A.; Bosi, A.; Pochini, A.; Ungaro, R. Tetrahedron
1985, 41, 3095.
(12) Van De Water, R. W.; Pettus, T. R. Tetrahedron 2002, 58, 5367.
5560
Org. Lett., Vol. 8, No. 24, 2006