Oxidant-controlled stereoselectivity in the Pd-catalyzed allylic oxidation of cis-vinylsilanes

Article abstract of DOI:10.1021/ja2089102

The allylic oxidation of cis-vinylsilanes is reported. The reaction requires a low catalyst loading of Pd(OAc)₂ without the need for an external ligand. Interestingly, trans-vinylsilanes are unreactive, whereas allylic oxidations of cis-vinylsi

Full text of DOI:10.1021/ja2089102

COMMUNICATION
pubs.acs.org/JACS
Oxidant-Controlled Stereoselectivity in the Pd-Catalyzed Allylic
Oxidation of cis-Vinylsilanes
Christopher T. Check, William H. Henderson, Brenda C. Wray, Matthew J. Vanden Eynden,
and James P. Stambuli*
Department of Chemistry, The Ohio State University, 100 West 18th Avenue, Columbus, Ohio 43210, United States
S Supporting Information
b
reports that describe metal-catalyzed allylic oxidation reactions of
vinylsilanes.
ABSTRACT: The allylic oxidation of cis-vinylsilanes is
reported. The reaction requires a low catalyst loading of
Pd(OAc)2 without the need for an external ligand. Inter-
estingly, trans-vinylsilanes are unreactive, whereas allylic
oxidations of cis-vinylsilanes proceed in good yields giving
a single diastereo- and regioisomer of the branched allylic
acetate trans-vinylsilane when benzoquinone is employed.
The use of PhI(OAc)2 as oxidant in place of benzoquinone
provides the branched, cis-vinylsilane as the major product.
Additionally, the first intramolecular allylic CÀH etherifica-
tions of cis-vinylsilanes to give oxygen heterocycles are also
described.
Initial experiments began with the exposure of trans-triethyl-
(heptenyl)silane (2) to the reaction conditions shown in eq 1.
The desired allylic oxidation product was not observed by GC or
1H NMR spectroscopy when the reaction was conducted at 23 or
70 °C. Only starting material (2) was observed in the 1H NMR
spectrum. This unreactivity was not too surprising given that
trans-2-hexene was also unreactive under our previous condi-
tions. To decrease steric hindrance about the alkene even further,
cis-triethyl(heptenyl)silane (4) was prepared.
he renewed interest toward the discovery of catalysts for the
T
selective allylic CÀH oxidation of terminal olefins demon-
strates the utility of the products in these reactions. Recent literature
reports describe the selective allylic oxidation of terminal olefins
to linear1À4 or branched allylic acetates,5À7 and the use of this
chemistry in the synthesis of complex molecules.8À10 Arguably,
linear allylic acetates are less synthetically useful than branched
allylic acetates because the former lack a stereogenic center. The
use of disubstituted olefins in these reactions would produce
stereogenic centers.
Submission ofcis-vinylsilane 4to 10 mol % Pd(OAc)2,10mol% 1,
and BQ in AcOH at 23 °C did not produce any oxidized pro-
duct (eq 2). Increasing the temperature of the reaction to 70 °C
gave 40% yield of an allylic oxidation product. Prior to spectro-
scopic analysis, the allylic oxidation product was expected to
be the allylic acetate 3 or 5. Analysis of the purified reac-
tion mixture showed the product to be the branched allylic
acetate containing a trans-vinylsilane (5). The product was
1
assigned based on 1- and 2-D H NMR spectroscopic analysis
There are numerous reports of allylic oxidations of disubsti-
tuted olefins, such as cyclohexene.11À14 However, acyclic disub-
stituted olefins are less commonly reported.14 For example, cis/
trans-2-hexene was unreactive under our reported conditions
that convert terminal olefins to linear allylic acetates (Pd(OAc)2,
thioether (1, PhS(CH2)2OC6H4-p-CH3), and benzoquinone
(BQ) in acetic acid).1 We hypothesized that the sterics about
the substituted alkene may inhibit binding to the metal catalyst,
which is likely required for the allylic oxidation to occur.2 According
to this hypothesis, increasing the sp2Àsp3 carbonÀcarbon bond
length (CÀCH3, ∼1.5 Å) of the olefin should decrease the steric
hindrance about the metal and increase binding to the metal
catalyst to allow a reaction to occur. To investigate this hypoth-
esis, vinylsilanes were employed because the CÀSi bond length
is 0.3 Å longer than the CÀC bond. We were also aware of the
α-silyl allyl acetate product that is formed from the allylic oxidation
may undergo a [3,3] rearrangement15,16 with an additional catalyst
in the same pot or in a second step to produce the branched allylic
acetate product. Such a transformation would be useful since
vinylsilanes are versatile intermediates in synthetic chemistry.17
Moreover, vinylsilanes are easily prepared by hydrosilylation of
the corresponding alkyne. Currently, there are no documented
(Supporting Information) by comparison of the alkene coupling
constants with previously reported trans-vinylsilanes.18,19 Inter-
estingly, this product was the only observed regioisomer by GC
and 1H NMR spectroscopic analysis of the crude reaction. The
1
only other vinylic resonances in the H NMR spectrum of this
reaction were assigned to a small amount of cis-vinylsilane 4.
After some optimization, the use of 2 mol % Pd(OAc)2 at 90 °C
without external ligand (1) in the presence of BQ provided the
optimal catalytic conditions.
Additional reaction screening showed that the optimal silicon
group was triethylsilyl (Table 1, entry 1). Replacing the triethylsilyl
group with the smaller trimethylsilyl group lowered the reaction
yield from 66% to 41% (entry 2). Increasing the steric bulk to the
Received: May 10, 2011
Published: October 19, 2011
r
2011 American Chemical Society
18503
dx.doi.org/10.1021/ja2089102 J. Am. Chem. Soc. 2011, 133, 18503–18505
|

Journal of the American Chemical Society
COMMUNICATION
Table 1. Effect of Silyl Group Substitution on Allylic
Table 3. Intramolecular Allylic Etherifications of
Oxidationsa
cis-Vinylsilanesa
a Isolated yields are an average of two (1 mmol) reactions.
Table 2. Intermolecular Allylic Oxidations of cis-
Vinylsilanesa
a Isolated yields are average of at least two (1 mmol) reactions.
However, the use of PhI(OAc)2 as oxidant increased the yields of
the oxidation products (entries 7À13). For example, using BQ as
oxidant with phthalimide substrate (18) gave 48% yield of
oxidation product 29 (entry 5), while the same substrate with
PhI(OAc)2 gave 63% yield of the desired allylic oxidation product
(entry 10). The increased yields are likely an effect of the faster
reaction times (e1.5 h) with this oxidant. All products isolated
using PhI(OAc)2 retained the cis-vinylsilane as the major product.
The metal-catalyzed intramolecular cyclization of alcohols
containing terminal olefins has been known for quite some time.
One drawback to these methods is that the formed products
typically contain a methyl group at the 2-position, limiting the
diversity of the products in these reactions (eq 3).20 Moreover,
the typical need to employ substrates containing geminal dis-
ubstitution to invoke the ThorpeÀIngold effect also limits re-
action scope. We discovered that exposure of vinylsilane 38 to a
catalytic amount of Pd(dba)2 in a 10:1 (v/v) solution of acetone
and acetic acid provided 66% yield of the desired α-vinyl tetrahy-
drofuran product 45 (Table 3, entry 1). The acetylated product was
also recovered along with a minor amount of the intermolecular
acetate product. This reaction seemed general as α-substituted
alcohols reacted smoothly to produce the desired vinyl tetrahydro-
furan products in 50À67% yield (entries 2À6). Moreover, allylic
etherification to form the tetrahydropyran 51 occurred in 55% yield
(entry 7). The method is complementary to Wacker-type oxidative
cyclizations, which typically employ phenol nucleophiles.21
a Isolated yields are average of at least two (1 mmol) reactions.
b Reactions were conducted with 5 mol % of Pd(OAc)2. c Isolated as a
mixture of cis (major) and trans (minor) mixture of diastereomers.
triphenylsilyl group lessened the yield of the reaction even
further to 16% (entry 3). Other silicon groups such as tert-butyl
dimethylsilyl and benzyl dimethylsilyl provided the desired
product in 44 and 57% yield, respectively (entries 4À5).
The proposed reaction mechanism for the allylic oxidation of
terminal olefins is thought to proceed via π-allylpalladium
intermediates. In the presence of BQ as oxidant, coordination
of the vinylsilane (I) followed by CÀH activation can produce
the anti-πÀallylpalladium intermediate II (Scheme 1). Complex
II can either undergo reductive elimination to produce the cis-
vinylsilane product or can undergo πÀσÀπ isomerization22 to
produce the syn-πÀallylpalladium complex (IV) via (III). In the
presence of PhI(OAc)2, complex I likely reacts with PhI(OAc)2
to produce the Pd(IV) species (V).14a Insertion of palladium into
The substrate scope of the reaction was examined under the
optimized conditions (Table 2). Most all of the products within
the table are regio- and diastereomerically pure when BQ was
used as oxidant (entries 1À2, 4À6). Moreover, the majority of
the reactions proceeded using only 2 mol % of Pd(OAc)2. An
increase in catalyst loading did not greatly improve the yield since
this also likely increased unproductive consumption of the vinylsilane.
18504
dx.doi.org/10.1021/ja2089102 |J. Am. Chem. Soc. 2011, 133, 18503–18505

Journal of the American Chemical Society
COMMUNICATION
Scheme 1. Oxidant-Controlled Diastereoselectivity
’ ASSOCIATED CONTENT
S
Supporting Information. Experimental procedures and
b
spectral characterization of all new compounds. This material is
available free of charge via the Internet at http://pubs.acs.org.
’ AUTHOR INFORMATION
Corresponding Author
stambuli@chemistry.ohio-state.edu
’ ACKNOWLEDGMENT
the allylic CÀH bond gives complex VI, which can undergo
CÀO bond forming reductive elimination to give the cis-vinylsi-
lane product. We cannot discount the potential role of Pd(III)
intermediate complexes in reactions with PhI(OAc)2 as oxidant,
as such intermediates would also increase the rate of CÀO bond
forming reductive elimination.23
The authors thank The Ohio State University for the generous
support of this research and the Ohio BioProduct Innovation
Consortium for supporting the mass spectrometry facility.
’ REFERENCES
(1) Henderson, W. H.; Check, C. T.; Proust, N.; Stambuli, J. P. Org.
Lett. 2010, 12, 824.
(2) (a) Campbell, A. N.; White, P. B.; Guzei, I. A.; Stahl, S. S. J. Am.
Chem. Soc. 2010, 132, 15116. (b) Lin, B.-L.; Labinger, J. A.; Bercaw, J. E.
Can. J. Chem. 2009, 87, 264.
(3) Mitsudome, T.; Umetani, T.; Nosaka, N.; Mori, K.; Mizugaki, T.;
Ebitani, K.; Kaneda, K. Angew. Chem., Int. Ed. 2006, 45, 481.
(4) Thiery, E.; Aouf, C.; Belloy, J.; Harakat, D.; Le Bras, J.; Muzart, J.
J. Org. Chem. 2010, 75, 1771.
(5) Chen, M. S.; Prabagaran, N.; Labenz, N. A.; White, M. C. J. Am.
Chem. Soc. 2005, 127, 6970.
(6) Chen, M. S.; White, M. C. J. Am. Chem. Soc. 2004, 126, 1346.
(7) Covell, D. J.; White, M. C. Angew. Chem., Int. Ed. 2008, 47, 6448.
(8) Fraunhoffer, K. J.; Bachovchin, D. A.; White, M. C. Org. Lett.
2005, 7, 223.
(9) Fraunhoffer, K. J.; Prabagaran, N.; Sirois, L. E.; White, M. C.
J. Am. Chem. Soc. 2006, 128, 9032.
(10) Stang, E. M.; White, M. C. Nature Chem. 2009, 1, 547.
(11) B€ackvall, J. E.; Hopkins, R. B.; Grennberg, H.; Mader, M. M.;
Awasthi, A. K. J. Am. Chem. Soc. 1990, 112, 5160.
(12) Hansson, S.; Heumann, A.; Rein, T.; Åkermark, B. J. Org. Chem.
1990, 55, 975.
(13) McMurry, J. E.; Kocovsky, P. Tetrahedron Lett. 1984, 25, 4187.
(14) (a) Pilarski, L. T.; Selander, N.; Bose, D.; Szaboꢀ, K. J. Org. Lett.
2009, 11, 5518. (b) Pilarski, L. T.; Janson, P. G.; Szabꢀo, K. J. J. Org. Chem.
2011, 76, 1503.
(15) Sparks, M. A.; Panek, J. S. J. Org. Chem. 1991, 56, 3431.
(16) Kim, A. I.; Kimmel, K. L.; Romero, A.; Smitrovich, J. H.;
Woerpel, K. A. J. Org. Chem. 2007, 72, 6595.
(17) Luh, T.-Y.; Liu, S.-T. The Chemistry of Functional Groups;
Rappoport, Z., Ed. John Wiley & Sons, Inc.: New York, 1998.
(18) Miller, R. B.; Mcgarvey, G. J. Org. Chem. 1978, 43, 4424.
(19) Takeuchi, R.; Tanouchi, N. J. Chem. Soc., Perkin Trans. 1 1994,
2909.
(20) (a) Dzudza, A.; Marks, T. J. Org. Lett. 2009, 11, 1523. (b) Seo,
S.; Yu, X.; Marks, T. J. J. Am. Chem. Soc. 2009, 131, 263. (c) Qian, H.;
Han, X.; Widenhoefer, R. A. J. Am. Chem. Soc. 2004, 126, 9536.
(21) McDonald, R. I.; Liu, G.; Stahl, S. S. Chem. Rev. 2011, 111, 2981.
(22) (a) Faller, J. W.; Thomsen, M. E.; Mattina, M. J. J. Am. Chem.
Soc. 1971, 93, 2642. (b) Trost, B. M.; Machacek, M. R.; Aponick, A.
Acc. Chem. Res. 2006, 39, 747.
From this proposed mechanism, some assumptions can be
made. First, the selective formation of the trans-allylic acetate
product when using BQ can be explained by slow CÀO bond
forming reductive elimination from complex (II), which allows
the system time to funnel to the typically more thermodynami-
cally stable syn-πÀallylpalladium complex (IV), which forms the
trans-vinylsilane after reductive elimination. The faster reductive
elimination from the Pd(IV) complex (VI) is likely the reason
the cis-product is the major product using this oxidant.14a,24
However, the rate of reductive elimination is not fast enough to
completely outcompete anti to syn isomerization, hence, the
appearance of trans product. Once the cis- and trans-allylic acetates
are formed, these do not interconvert under the conditions with
BQ or PhI(OAc)2 as oxidant since submission of a 5:1 mixture of
cis/trans vinylsilanes under either reaction condition did not alter the
ratio of the diastereomers. These results suggest that the catalyst
does not insert into the products to re-form complexes II, IV, or VI.
Upon replacing AcOH from the optimized reaction condi-
tions with a 4:1 (v/v) solution of AcOH/H2O in the presence of
Pd(OAc)2 and PhI(OAc)2 as the oxidant, cis-vinylsilane 4 was
converted to the cis-vinylsilane allylic acetate in 45% yield. The
1
corresponding trans product was not observed by H NMR
spectroscopy of the crude reaction mixture. As a control experi-
ment, the corresponding trans-allylic acetate product (5) was
subjected to the above reaction conditions containing water and
did not decompose. This result suggests that both products are
not being formed concurrently. The starting material is comple-
tely consumed in the reaction. The amount of water appears to
increase catalyst or reaction intermediate decomposition rate,
thus, decreasing the reaction yield. The selectivity observed by
the addition of water may be caused by an increase in the rate of
CÀO reductive elimination or by somehow shutting down the
syn- to anti-πÀallylpalladium isomerization pathway.25
In summary, the allylic oxidation of cis-vinylsilanes to produce
branched allylic acetate cis- or trans-vinylsilane compounds oc-
curs in the presence of catalytic Pd(OAc)2. This reaction with
BQ as oxidant solely provides the branched, trans-silyl allylic
acetate product, while the use of PhI(OAc)2 as oxidant provides
the cis-silyl allylic acetate as the major product. The first intra-
molecular allylic CÀH etherification of cis-vinylsilanes produced
five- and six-membered oxygen heterocycle products that re-
tained the vinylsilane functionality. The reaction pathway is
being examined in order to suppress the deleterious side reaction
and better understand the observed selectivities.
(23) Powers, D. C.; Ritter, T. Nat. Chem. 2009, 1, 302.
(24) Yin, G; Wu, Y.; Liu, G. J. Am. Chem. Soc. 2010, 132, 11978.
(25) For another report describing an interesting, unexplained effect
of water in πÀallylpalladium chemistry, see:Fernandes, R. A.; Stimac, A.;
Yamamoto, Y. J. Am. Chem. Soc. 2003, 125, 14133.
18505
dx.doi.org/10.1021/ja2089102 |J. Am. Chem. Soc. 2011, 133, 18503–18505