Effects of Solvent and Residual Water on Enhancing the Reactivity of Six-Membered Silyloxyallyl Cations toward Nucleophilic Addition

Article abstract of DOI:10.1021/acs.orglett.6b02194

A new strategy for the generation of six-membered unsymmetrical silyloxyallyl cations using catalytic mild Br?nsted acid is reported. These reactive intermediates were found to readily undergo direct nucleophilic addition with a broad range of nucleophile

Full text of DOI:10.1021/acs.orglett.6b02194

Letter
pubs.acs.org/OrgLett
Effects of Solvent and Residual Water on Enhancing the Reactivity of
Six-Membered Silyloxyallyl Cations toward Nucleophilic Addition
†
†
Joshua A. Malone, Alexander H. Cleveland, Frank R. Fronczek, and Rendy Kartika*
Department of Chemistry, 232 Choppin Hall, Louisiana State University, Baton Rouge, Louisiana 70803, United States
*
S Supporting Information
ABSTRACT: A new strategy for the generation of six-
membered unsymmetrical silyloxyallyl cations using catalytic
mild Brønsted acid is reported. These reactive intermediates
were found to readily undergo direct nucleophilic addition with
a broad range of nucleophiles to produce various α,α′-
disubstituted silyl enol ether structural motifs. The findings
also highlight the significance of the solvent effect and residual
water in enhancing the reaction rate.
he development of synthetic methods that incorporate
indolyl silyl enol ether 5 as a single regioisomer in excellent yield
(Scheme 1). This putative unsymmetrical silyloxyallyl cation
3
T
indole at the α-position of cyclohexanone has recently
gained considerable interest. This emerging area of study has
largely been inspired by synthetic endeavors from the Baran
laboratory, which demonstrated the strategic utility of this type of
molecular scaffold toward concise and elegant total syntheses of
indole alkaloid natural products such as (−)-fischerindole I,
Scheme 1. Limitation of the Previous Methodology
(
+)-hapalinole Q, and (+)-ambiguine H (Figure 1), among
intermediate was efficiently produced from α′-hydroxy starting
material 3 upon ionization with catalytic pyridinium triflate in
toluene at room temperature. Surprisingly, exposure of the
analogous six-membered starting material 6 to these very mild
conditions failed to generate the target product 8. In fact, this
substrate largely remained unreacted. We attributed this
profound lack of reactivity to the destabilization of silyloxyallyl
cation 7 due to the emerging allylic strain that appeared to be
more pronounced in this cationic species compared with that of
Figure 1. Direct addition of indole to oxyallyl cations.
1
others. Over the past several years, a particularly interesting
strategy to construct α-indolyl cyclohexanone has been
exemplified through the generation of zwitterionic oxyallyl
cation 2 from the corresponding cyclohexanone precursor 1,
4
the five-membered variant 4.
Given this limitation, we sought to develop a viable solution to
further broaden the scope of our chemistry. We hypothesized
that the barrier to formation of silyloxyallyl cation 7 could be
which can then be directly captured by indole at the α-carbon
2
(
Figure 1). While generally robust, the scope of this chemistry
5
overcome by applying the Hughes−Ingold rules, which outline
nonetheless suffers from a lack of regiocontrol when α-branched
2f,g
the role of polar solvents in accelerating the rate of reactions
involving the formation of charged intermediates. In our previous
studies, we deliberately employed a nonpolar solvent, such as
unsymmetrical oxyallyl cations are used.
To address this fundamental issue, we recently demonstrated
that simply by “protection” of the forming oxyallyl cation with a
silyl group at the oxygen center (viz. 4), the ensuing addition of
indole could be controlled exclusively at the less-substituted
electrophilic carbon (i.e., the α′-position), thereby producing α′-
Received: July 25, 2016
©
XXXX American Chemical Society
A
DOI: 10.1021/acs.orglett.6b02194
Org. Lett. XXXX, XXX, XXX−XXX

Organic Letters
Letter
a
toluene, to maximize the rate of proton transfer between the
Brønsted acid catalyst and the α′-hydroxyl group while
suppressing protodesilylation. As we embarked on this
investigation, we were naturally cognizant that the use of pola6r
solvents could potentially enhance the rate of protodesilylation,
thus altering the delicate kinetic balance between these two
competitive processes.
Scheme 2. Scope of Substituted Indoles
Table 1 depicts our reaction optimization. As a starting point,
we activated α′-hydroxy silyl enol ether 6 with 0.1 equiv of
a
Table 1. Reaction Optimization
a
Isolated yields after flash column chromatography are shown. All
products were isolated with >20:1 regioselectivity.
a
indole, afforded products 9a and 9b in 71% and 80% yield,
respectively. The use of halogen-containing indoles showed that
this methodology readily produced iodinated and brominated
silyl enol ethers 9c and 9d, which could be further utilized in
downstream reactions to introduce more densely functionalized
materials. Our reaction also appeared to tolerate electron-
deficient methyl indole-5-carboxylate and 5-nitroindole, which
furnished α′-indolyl adducts 9e and 9f, albeit in lower yields at
lower reaction rates. Other substitution patterns, such as
sterically hindered 2-phenylindole as well as protected N-
methylindole and N-benzylindole successfully produced silyl
enol ethers 9g−i in good yields. Compounds 9d and 9g existed as
crystalline materials, thus enabling us to unambiguously confirm
Isolated yields after flash column chromatography are shown.
Starting material 6 was not fully consumed. 0.1 equiv of water was
b
c
added.
pyridinium triflate and 1.1 equiv of indole in a slightly polar
solvent, dichloromethane (entry 1). To prevent reversibility in
the ionization step, 4 Å molecular sieves were introduced to
remove water byproduct from the reaction mixtures.
7
3
d
Interestingly, this reaction produced the intended α′-indolyl
silyl enol ether 8 in a trace amount despite an extended reaction
8
time. Removal of the molecular sieves resulted in an unexpected
improvement in the rate of reaction (entry 2). Nonetheless,
when similar conditions were applied in toluene, the starting
material never fully disappeared even when the reaction was
allowed to run for 2 weeks (entry 3). Astonishingly, ionization of
9
their structures by X-ray crystallography.
Our investigation then continued with the screening of
substituent effects at the α-carbon (Table 2). To enable rapid
diversification at this position, we designed an alternative starting
material in tertiary α-hydroxy silyl enol ether 10 with the
hypothesis that ionization of this compound under the typical
reaction conditions should produce unsymmetrical silyloxyallyl
6
in acetonitrile led to its full consumption in just 26 h, leading to
the isolation of the coupling product 8 in 71% yield as a single
regioisomer (entry 4).
After the solvent screening, the catalyst loading was then
investigated through the systematic use of 0.2, 0.5, and 1.0 equiv
of pyridinium triflate (Table 1, entries 5−7). While the reaction
rate correspondingly accelerated with increasing concentration
of the Brønsted acid, the isolated yield of silyl enol ether product
7
cations 11. Similarly, the ensuing nucleophilic addition by
indole should then occur at the less-substituted electrophilic α′-
carbon and generate the intended coupling product 12. As a
proof of concept, we began with α-methyl silyl enol ether 10a.
Activation of this substrate with catalytic pyridinium triflate and
indole under conditions identical to those for ionization of 6
cleanly afforded α-methyl α′-indole adduct 12a (8) with a
comparable yield and reaction time (entry 1). A further
investigation using n-octyl 10b, allyl 10c, isobutyl 10d, and
cyclohexyl 10e revealed the tolerance of various aliphatic groups
at the α-carbon, as the respective α′-indolyl products 12b−e
were readily isolated in high yields (entries 2−5). Interestingly,
we observed a substantial enhancement in the reaction rate
associated with substrates 10d and 10e. Although the exact origin
of this behavior is currently unclear, we speculated that steric
congestion at the α-carbon might have affected the extent of
ionization of these starting materials, which was presumed to be
reversible and rate-determining.
8
only improved to around 80%. Lastly, our attention turned to
optimizing the amount of indole (entries 8 and 9). We observed
that the use of either 1.0 or 2.0 molar equiv of indole did not
particularly affect the yield or reaction rate. Our attempt to
perform this reaction in a completely anhydrous environment
was again found to be unfruitful (entry 10). In contrast, the
addition of catalytic water into the mixture appeared to accelerate
the reaction but lowered the isolated yield of 8 (entry 11).
Nonetheless, we ultimately established the ionization conditions
listed in entry 5 as optimal for the α′-functionalization of silyl
enol ether 6 with indole.
With these reaction conditions, we then explored the scope of
various substituted indoles (Scheme 2). Unsurprisingly,
electron-rich indoles, such as 5-methoxyindole and 5-methyl-
B
DOI: 10.1021/acs.orglett.6b02194
Org. Lett. XXXX, XXX, XXX−XXX

Organic Letters
Letter
a
Table 2. Scope of Substrates
in our methodology was supported by the experiments depicted
in entries 8 and 9. Treatment of starting material 10h with
catalytic pyridinium triflate and indole led to an extremely
sluggish reaction, implying that the weakly deactivating p-
chlorophenyl substituent readily impeded the formation of the
cationic species. Similarly, the unsubstituted substrate 10i failed
to react, thus indicating that (1) the alternative S 2′ mechanism
N
is unlikely to be responsible for the formation of observed
substitution products, especially since both starting materials 6
and 11a yielded the same product, and (2) substitution effects at
the α-position are crucial to help stabilize the silyloxyallyl cation
intermediate.
We also investigated the suitability of other carbon and
heteroatom nucleophiles beyond substituted indoles using
tertiary α-hydroxy silyl enol ether 10a as the substrate (Scheme
3
). While benzothiophene and benzofuran failed to capture the
a
Scheme 3. Scope of Nucleophiles
a
Isolated yields after flash column chromatography are shown. All
b
products were isolated with >20:1 regioselectivity. 2.0 equiv of 1-
phenyl-1-trimethylsiloxyethylene was employed. 3.0 equiv of methyl
c
trimethylsilyl dimethylketene acetal was employed. The reaction was
carried out at −10 °C and then warmed to rt. See the Supporting
d
Information. 2.0 equiv of alcohols was employed.
putative silyloxyallyl cation 7, the use of pyrrole readily furnished
coupling product 13a in a modest yield. To underscore the
mildness of our reaction conditions, we demonstrated the
compatibility of acid-labile π nucleophiles, such as 1-phenyl-1-
trimethylsiloxyethylene and methyl trimethylsilyl dimethylke-
tene acetal. Such nucleophiles produced the corresponding 1,4-
dicarbonyl-derived products 13b and 13c in good yields. Our
ability to selectively differentiate the two carbonyl groups in these
compounds while creating a new carbon−carbon connectivity in
one single synthetic operation clearly highlights the novelty of
our chemistry. With regard to heteroatom-centered nucleophiles,
applications of both primary and secondary alcohols in p-
methoxybenzyl alcohol and cyclohexanol proceeded to afford
their corresponding α′-alkoxy silyl enol ethers 13d and 13e in
64% and 50% yield, respectively. We also tested thiophenol and
aniline, which yielded products 13f and 13g in 90% and 32%
yield, respectively.
a
Isolated yields after flash column chromatography are shown. All
b
products were isolated with >20:1 regioselectivity. A prolonged
reaction time resulted in the decomposition of the product. The
starting material 10h was not fully consumed.
c
As illustrated in Table 2, entries 6−8, the effect of aromatic
substituents was then examined, starting with α-tolyl substrate
1
0f. While exposure of this compound to the typical indolylation
conditions produced the corresponding α,α′-diaryl silyl enol
ether 12f in 82% yield, the reaction was extremely sluggish. We
believed that this surprisingly longer reaction time was most
likely attributable to the unfavorable 1,3-allylic strain inherent in
the emerging silyloxyallyl cation 11 due to the presumed
planarization between the tolyl group and the oxyallyl cation
moiety, enabling delocalization of π electrons. This stereo-
electronic trade-off appeared to be readily controlled by the
substituent on the aromatic ring. For instance, the introduction
of an electron-rich p-methoxyphenyl substituent in 10g cleanly
afforded silyl enol ether 12g in merely 22 h. The proposed
participation of putative silyloxyallyl cation 11 as an intermediate
In the course of reaction optimization, we observed the
significance of residual water in enhancing the reaction rate. In
fact, kinetic measurements on three sets of activation conditions
(Table 1, entries 5, 10, and 11) unambiguously established this
10
reactivity pattern (Scheme 4). To further probe the role of
water in our methodology, we treated secondary and tertiary α′-
hydroxy silyl enol ethers 6 and 10a individually with 0.2 equiv of
C
DOI: 10.1021/acs.orglett.6b02194
Org. Lett. XXXX, XXX, XXX−XXX

Organic Letters
Letter
Scheme 4. Probing the Role of Water
Notes
The authors declare no competing financial interest.
ACKNOWLEDGMENTS
■
This material is based upon work supported by the National
Science Foundation under CHE-1464788. Generous financial
support from Louisiana State University is gratefully acknowl-
edged. A.H.C. thanks the Louisiana Board of Regents for the
Graduate Fellowship (LEQSF(2015-20)-GF-02). This article is
dedicated to Mrs. Carolyn Forrester on the occasion of her
retirement from the Department of Chemistry at California State
Polytechnic University, Pomona.
REFERENCES
a
■
The reaction progress was monitored by HPLC analyses of the crude
(
1) (a) Richter, J. M.; Ishihara, Y.; Masuda, T.; Whitefield, B. W.;
Llamas, T.; Pohjakallio, A.; Baran, P. S. J. Am. Chem. Soc. 2008, 130,
7938. (b) Baran, P. S.; Maimone, T. J.; Richter, J. M. Nature 2007, 446,
04. (c) Baran, P. S.; Richter, J. M.; Lin, D. W. Angew. Chem., Int. Ed.
005, 44, 609. (d) Baran, P. S.; Richter, J. M. J. Am. Chem. Soc. 2005, 127,
5394. (e) Baran, P. S.; Richter, J. M. J. Am. Chem. Soc. 2004, 126, 7450.
reaction mixtures using 0.1 equiv of naphthalene as an internal
standard. These compounds were assumed to elicit identical LC
responses at 254 nm.
1
4
2
1
(
2) (a) Fort, A. W. J. Am. Chem. Soc. 1962, 84, 2620. (b) Freter, K.
pyridinium triflate in acetonitrile. Upon equilibration for 24 h,
each sample was found to produce a mixture of isomers 6 and
Liebigs Ann. Chem. 1978, 1978, 1357. (c) Fohlisch, B.; Joachimi, R.
̈
Chem. Ber. 1987, 120, 1951. (d) Leitich, J.; Heise, I. Eur. J. Org. Chem.
2001, 2001, 2707. (e) Harmata, M.; Huang, C.; Rooshenas, P.;
Schreiner, P. R. Angew. Chem., Int. Ed. 2008, 47, 8696. (f) Tang, Q.;
Chen, X.; Tiwari, B.; Chi, Y. R. Org. Lett. 2012, 14, 1922. (g) Vander
Wal, M. N.; Dilger, A. K.; MacMillan, D. W. C. Chem. Sci. 2013, 4, 3075.
1
0a, favoring the tertiary alcohol 10a in a roughly 3:1 ratio as
determined by NMR analyses of the crude materials. From these
results, in conjunction with the solvation effect provided by the
subtle polarity in acetonitrile, we proposed that silyloxyallyl
cation 7 is perhaps further stabilized by residual water via a non-
covalent interaction at the tertiary electrophilic center, en route
to nucleophilic addition with indole at the α′-position. This
hypothesis was supported by the fact that the introduction of
catalytic water (entry 11) appeared to induce substantial rate
acceleration during the initial progression of the reaction where
the amount of water generated from the ionization of starting
material 6 was essentially negligible. Whether this underlying
intermolecular force involves solvent separation or more
intimate pairing is unclear at this juncture, and further studies
are currently ongoing.
(
h) Luo, J.; Zhou, H.; Hu, J. W.; Wang, R.; Tang, Q. RSC Adv. 2014, 4,
1
7370. (i) Luo, J.; Jiang, Q.; Chen, H.; Tang, Q. RSC Adv. 2015, 5,
6
7901. (j) Liu, C.; Oblak, E. Z.; Vander Wal, M. N.; Dilger, A. K.;
Almstead, D. K.; MacMillan, D. W. C. J. Am. Chem. Soc. 2016, 138, 2134.
3) (a) Stepherson, J. R.; Fronczek, F. R.; Kartika, R. Chem. Commun.
016, 52, 2300. (b) Stepherson, J. R.; Ayala, C. E.; Dange, N. S.; Kartika,
(
2
R. Synlett 2016, 27, 320. (c) Dange, N. S.; Stepherson, J. R.; Ayala, C. E.;
Fronczek, F. R.; Kartika, R. Chem. Sci. 2015, 6, 6312. (d) Ayala, C. E.;
Dange, N. S.; Fronczek, F. R.; Kartika, R. Angew. Chem., Int. Ed. 2015, 54,
4641. (e) Ayala, C. E.; Dange, N. S.; Stepherson, J. R.; Henry, J. L.;
Fronczek, F. R.; Kartika, R. Org. Lett. 2016, 18, 1084.
(4) While six-membered oxyallyl cations can be generated under base-
promoted activation conditions (see ref 2f−i), the use of milder catalytic
protocols was shown to be ineffective (see ref 2j)
In conclusion, we have successfully developed a new strategy
that addresses the lack of reactivity of six-membered unsym-
metrical silyloxyallyl cations toward direct nucleophilic addition.
Given the mildness of the activation conditions, our method-
ology was readily tolerated by a broad scope of substrates and
nucleophiles. While our initial mechanistic investigation revealed
the significance of acetonitrile and residual water in enhancing
the rate of reaction, we are now pursuing in-depth studies to
further probe the mechanism, including the origin of the
regioselectivity. Our results will be reported in due course.
(
(
5) Hughes, E. D.; Ingold, C. K. J. Chem. Soc. 1935, 244.
6) (a) Schubert, M.; Metz, P. Angew. Chem., Int. Ed. 2011, 50, 2954.
(
b) Fischer, D.; Nguyen, T. X.; Trzoss, L.; Dakanali, M.; Theodorakis, E.
A. Tetrahedron Lett. 2011, 52, 4920. (c) O’Connor, P. D.; Del Signore,
G.; McLachlan, M. M. W.; Willis, A. C.; Mander, L. N. Aust. J. Chem.
2010, 63, 1477. (d) Spivey, A. C.; Martin, L. J.; Grainger, D. M.; Ortner,
J.; White, A. J. P. Org. Lett. 2006, 8, 3891.
(
7) Compounds 6, 10, and 13 were readily prepared in just a few steps
from commercially available materials. See the Supporting Information.
8) Interestingly, ionization of substrate 6 with 0.1 equiv of CSA in
(
dichloromethane at −78 °C in the presence of indole and 4 Å molecular
sieves, following a protocol described in ref 3a, failed to produce the
target α′-indolyl silyl enol ether 8.
ASSOCIATED CONTENT
Supporting Information
■
*
S
The Supporting Information is available free of charge on the
ACS Publications website at DOI: 10.1021/acs.orglett.6b02194.
(9) CCDC 1493044 and 1493045 contain the supplementary
crystallographic data for compounds 9d and 9g, respectively. These
data can be obtained from The Cambridge Crystallographic Data
Centre.
Experimental procedures and spectral data (PDF)
Crystallographic data for 9d (CIF)
Crystallographic data for 9g (CIF)
(10) The lower product yield in Table 1, entry 11 was most likely
caused by competitive water-promoted decomposition of starting
material 6 under the reaction conditions.
AUTHOR INFORMATION
Corresponding Author
■
*rkartika@lsu.edu
Author Contributions
†
J.A.M. and A.H.C. contributed equally.
D
DOI: 10.1021/acs.orglett.6b02194
Org. Lett. XXXX, XXX, XXX−XXX