Catalytic Enantioselective Allenoate-Alkene [2 + 2] Cycloadditions

Article abstract of DOI:10.1021/jacs.5b00563

Catalytic enantioselective [2 + 2] cycloadditions between allenoates and alkenes is disclosed. The method functions well for a variety of alkenes, and the products are generated with excellent levels of enantioselectivity. One of the most significant aspe

Full text of DOI:10.1021/jacs.5b00563

Communication
pubs.acs.org/JACS
Catalytic Enantioselective Allenoate−Alkene [2 + 2] Cycloadditions
#
#
Michael L. Conner, Yao Xu, and M. Kevin Brown*
Department of Chemistry, Indiana University, 800 E. Kirkwood Avenue, Bloomington, Indiana 47405, United States
*
S Supporting Information
Scheme 1. [2 + 2] Cycloadditions
ABSTRACT: Catalytic enantioselective [2 + 2] cyclo-
additions between allenoates and alkenes is disclosed. The
method functions well for a variety of alkenes, and the
products are generated with excellent levels of enantiose-
lectivity. One of the most significant aspects of the present
method is that unactivated alkenes are suitable substrates
for this method, which is distinctly different from nearly all
other catalytic enantioselective [2 + 2] cycloaddition
methods.
ynthesis of cyclobutanes through [2 + 2] cycloadditions of
S
alkenes has emerged as an indispensible method for
1
chemical synthesis. This is due to the presence of the
cyclobutane motif in a variety of important molecules and the
products as reactions with ketenes yet may not suffer from
product inhibition. We were encouraged by three seminal
reports from Snider and Hoffman who in the early to mid-
1980s demonstrated that Al(III)-Lewis acids were capable of
promoting [2 + 2] cycloaddition between allenoates and9
unactivated alkenes (e.g., 1-hexene, cyclopentene, etc.).
Furthermore, in several cases, catalytic turnover of the Lewis
acid was possible. More recently, Loh and co-workers reported
a Lewis acid-catalyzed intramolecular cycloaddition of allenic
ease with which cyclobutanes can be converted readily to a
variety of structures. Consequently, catalytic enantioselective
1
[
2 + 2] cycloadditions to prepare cyclobutanes have emerged as
an attractive target for methods development. The majority of
methods for catalytic enantioselective [2 + 2] cycloadditions
involve the formal cycloaddition between highly polarized
2
,3
electron-rich and electron-poor alkenes. Catalytic enantiose-
lective photochemical [2 + 2] cycloadditions have also recently
4
10
been reported. Several Au-catalyzed cycloadditions of electron-
ketones. These four papers represent the only reported
5
rich allenes and styrenes have been disclosed. Finally,
studies in this area over a ∼30-year period.
transition metal-catalyzed cycloaddition of alkynes with bicyclic
We initiated our studies by examining the cycloaddition of
benzylallenoate 2 (stable and easily prepared in one step) and
allyltrimethylsilane (1) in the presence of a variety of chiral
6
alkenes are known. Despite these efforts, synthesis of
cyclobutanes by catalytic enantioselective [2 + 2] cyclo-
additions remains extremely limited. In particular, methods
that employ weakly activated alkenes are exceptionally rare and
11
Al(III)-Lewis acid catalysts (4−7). While several classes of
chiral Al(III)-Lewis acid catalysts were capable of promoting
the cycloaddition, the enantioselectivities of the processes were
low (4−5, 7). Further evaluation of other highly Lewis-acidic
catalysts ultimately led to the discovery that use of
oxazaborolidine 9 provided the product in 96:4 enantiomeric
4a−i,l
are only known for intramolecular photochemical reactions
and cycloadditions of trisubstituted alkenes.
3a
Herein, we disclose a new class of catalytic enantioselective
[2 + 2] cycloadditions. The process combines readily available
12
ratio (er) but only 20% yield (Scheme 2). Optimization of the
catalyst structure revealed that incorporation of a more
electron-deficient B−Ar unit led to significantly increased
yields at only a small cost in enantioselectivity (Scheme 2,
compare catalysts 9 with 10−13). Fine-tuning of this
substituent ultimately led to the identification of catalyst 13,
which provided cyclobutane 3 in 92% NMR yield and 92:8 er.
Interestingly, related known oxazaborolidine catalyst 8 failed to
allenoates and a variety of alkenes to generate synthetically
versatile cyclobutanes. A key feature of this method is that
weakly or nonpolarized alkenes can be utilized.
Our lab has initiated a program directed toward developing
stereo- and enantioselective [2 + 2] cycloadditions with allenes
and heteroallenes. Along these lines, we have previously
developed a method for Lewis acid-promoted [2 + 2]
7
,8
cycloadditions between ketenes and alkenes (Scheme 1).
13
promote the cycloaddition.
While this method greatly expands the scope of this venerable
reaction, a notable drawback is the requirement for a
stoichiometric amount of Lewis acid due to severe product
inhibition. Thus, development of catalytic, and hence catalytic
enantioselective, variants of this process has been impeded.
To provide a solution to this problem, reactions with
allenoates were evaluated as these would generate similar
With an optimized set of conditions/catalyst in hand, we
explored the scope of this method. As illustrated in Scheme 3 a
variety of alkenes undergo cycloaddition with good yields and
Received: January 17, 2015
Published: March 10, 2015
©
2015 American Chemical Society
3482
DOI: 10.1021/jacs.5b00563
J. Am. Chem. Soc. 2015, 137, 3482−3485

Journal of the American Chemical Society
Communication
a
Scheme 2. Initial Evaluation of Chiral Catalysts
catalytic enantioselective methods (vida supra). (2) In general,
higher yields are observed with more strained alkenes. For
example, the more strained cyclooctene undergoes cyclo-
addition with higher yield than a reaction with less strained
cyclohexene (compare yields of products 18 and 20). (3)
Cycloaddition with 1-hexene (product 16) represents a rare
example of a catalytic enantioselective reaction with a terminal
15
unactivated alkene. In this example, higher catalyst loading
and excess alkene are necessary due to the poor nucleophilicity
of 1-hexene. (4) With less reactive alkenes, the more
electrophilic trifluoroethylallenoate 14 was employed to obtain
higher yields (products 16−19). (5) The catalyst was easily
prepared in one step from commercially available starting
materials. (6) Current limitations of the method are the use of
trisubstituted alkenes and α- or γ-substituted allenes, as low
selectivity was observed (not shown).
Internal nonactivated alkenes are also suitable substrates for
this process (Scheme 4). Reaction of cis-4-octene (22) under
Scheme 4. Reactions with Acyclic Internal Alkenes
a
b
See the Supporting Information for experimental details. Deter-
1
c
mined by H NMR analysis with an internal standard. Determined by
HPLC analysis with a chiral column.
a
Scheme 3. Substrate Scope with Cyclic Alkenes
standard conditions provided 23 in 42% yield, 14:1 E:Z, 93:7
er, and >20:1 dr. Likewise reaction of trans-4-octene furnished
2
5 in 41% yield, 4:1 E:Z, 92:8 er, and >20:1 diastereomeric
ratio (dr). For both isomers of 4-octene the geometry of
starting material is completely conserved in the product. These
results suggest that the cycloaddition event is concerted or
involves an ephemeral intermediate. Furthermore, the
processes are likely highly asynchronous as it has been noted
7,8
for related ketene−alkene [2 + 2] cycloadditions.
regioselectivity of the cycloaddition can also be controlled
Scheme 4b). Thus, nonsymmetric alkene 26 undergoes
The
(
reaction with allenoate 14 to provide the regioisomer in
which the i-Pr-substituent is distal to the unsaturated carbonyl
(4:1 regioisomeric ratio (rr)). The major regioisomer was
generated in 7:1 E:Z and 95:5 er.
a
With 5 equiv alkene. See the Supporting Information for
experimental details. Yields reported are the average of two
experiments. Enantiomeric ratios determined by HPLC analysis with
a chiral column. With 50 mol % catalyst and 10 equiv alkene in the
A rationale that accounts for the observed enantio- and
diastereoselectivity is illustrated in Scheme 5 (X-ray of
derivative 29 established absolute stereochemistry). On the
basis of the above discussion, the reaction is likely concerted
and highly asynchronous or involves a short-lived dipolar
b
absence of CH Cl .
2
2
1
4
enantioselectivity. Several points regarding this process are
noteworthy: (1) Unactivated alkenes can be used in this
process (e.g., cycloheptene, product 19; 1-hexene, product 16).
This stands in sharp contrast to the vast majority of reported
9
,16
intermediate.
In either event, coordination of the allene to
the Lewis acidic boron atom likely occurs according to the
12
model illustrated in Scheme 5 (28). The orientation of the
3
483
DOI: 10.1021/jacs.5b00563
J. Am. Chem. Soc. 2015, 137, 3482−3485

Journal of the American Chemical Society
Communication
Scheme 5. Proposed Model for Selectivity
Scheme 7. Kinetic Resolution of Racemic Alkenes
allene with respect to the catalyst may be fixed by a putative C−
H···O hydrogen bonding interaction as proposed in related
1
2,17
Diels−Alder reactions.
Approach of the alkene can only
occur from the front face of allene as the back face is blocked by
the Ph-substituent on the convex face of the catalyst.
Furthermore, the alkene substituents (R and R ) are pointed
away from the steric bulk of the catalyst.
1
2
allene. We expect that the rate of reaction in which the i-Pr-
group is positioned proximal to the allene was reduced due to
unfavorable steric interactions (model 35), thus allowing
preferential reaction of the other enantiomer of the allylsilane
through pretransition state 36.
Cycloadditions with symmetric and nonsymmetric alkenes
that have the potential to establish multiple stereocenters have
been investigated (Schemes 6 and 7). Thus, cycloaddition of
Scheme 6. Desymmetrization by [2 + 2] Cycloaddition
The utility of this method was demonstrated toward the
1
9,20
formal synthesis of amino acid 39 (Scheme 8).
This
Scheme 8. Synthesis of Intermediate 38
7d
diethyl cyclopentene derivative 30 and allene 14 promoted by
catalyst 13 led to formation of cyclobutane 31 with the
generation of four new stereocenters in excellent selectivity
(
Scheme 6). A model that rationalizes the observed selectivity
molecule was recently identified to be a potent ligand (19 nM)
for the α δ subunit of a voltage gated calcium channel and is
related to anticonvulsants Lyrica and Neurontin. The reported
is shown in Scheme 6 (32). It is likely that the alkene
approached the allenoate-Lewis acid complex with the large Et-
substituents oriented away from the complex to reduce adverse
interactions.
2
route to 39 involves nine steps and a resolution to separate
19
enantiomers from a ketene−alkene [2 + 2] cycloaddition. As
illustrated in Scheme 8, intermediate 38 could be easily
prepared by one-pot catalytic enantioselective [2 + 2]
cycloaddition followed by Michael addition and trans-
Cycloaddition with chiral, racemic allylsilane 33 has also
18
been investigated (Scheme 7). Treatment of allenic ester 2
with an excess of racemic chiral allylsilane 33 in the presence of
catalyst 13 led to the formation of the cyclobutane 34 in 54%
yield, 91:9 er, and 10:1 dr. In this example, one enantiomer of
allylsilane 33 underwent preferential reaction. This reaction
likely proceeded through the putative pretransition states 35
and 36 illustrated in Scheme 7. A key component of these
models is that the C−Si bond is properly aligned to stabilize the
developing positive charge. This alignment positions the i-Pr-
group either proximal (model 35) or distal (model 36) to the
21
esterification in 52% yield > 20:1 dr and 86:14 er. The
ethyl ester of 38 has been converted to 39 through a simple
22
two-step sequence.
In summary, a new and highly enantioselective method for
catalytic [2 + 2] cycloaddition is disclosed. This method is
notable because unactivated alkenes can be utilized. Future
directions are aimed at extending the scope to include related
classes of electron deficient allenes.
3
484
DOI: 10.1021/jacs.5b00563
J. Am. Chem. Soc. 2015, 137, 3482−3485

Journal of the American Chemical Society
Communication
2
014, 53, 7661. (k) Du, J.; Skubi, K. L.; Schultz, D. M.; Yoon, T. P.
ASSOCIATED CONTENT
Supporting Information
Experimental procedures, analytical data for all compounds, and
crystallographic data (CIF). This material is available free of
charge via the Internet at http://pubs.acs.org.
■
Science 2014, 344, 392. (l) Brimioulle, R.; Bach, T. Angew. Chem., Int.
Ed. 2014, 53, 12921.
*
S
(
5) (a) Luzung, M. R.; Mauleon
007, 129, 12402. (b) Teller, H.; Flu
Angew. Chem., Int. Ed. 2010, 49, 1949. (c) Gonzal
́
, P.; Toste, F. D. J. Am. Chem. Soc.
gge, S.; Goddard, R.; Furstner, A.
ez, A. Z.; Benitez, D.;
2
̈
̈
́
Tkatchouk, E.; Goddard, W. A., III; Toste, F. D. J. Am. Chem. Soc.
2011, 133, 5500. (d) Teller, H.; Corbet, M.; Mantilli, L.; Gopakumar,
AUTHOR INFORMATION
Corresponding Author
brownmkb@indiana.edu
Author Contributions
■
G.; Goddard, R.; Thiel, W.; Fu
5331. (e) Suarez-Pantiga, S.; Hernan
Gonzalez, J. M. Angew. Chem., Int. Ed. 2012, 51, 11552. (f) Mauleon
P. ChemCatChem. 2013, 5, 2149.
̈
rstner, A. J. Am. Chem. Soc. 2012, 134,
1
́
́
dez-Díaz, C.; Rubio, E.;
́
́
,
#
M.L.C. and Y.X. contributed equally to this work.
(6) (a) Shibata, T.; Takami, K.; Kawachi, A. Org. Lett. 2006, 8, 1343.
(
b) Fan, B.-M.; Li, X.-J.; Peng, F.-Z.; Zhang, H.-B.; Chan, A. S. C.;
Notes
Shao, Z.-H. Org. Lett. 2010, 12, 304. (c) Hu, J.; Yang, Q.; Yu, L.; Xu, J.;
Liu, S.; Huang, C.; Wang, L.; Zhou, Y.; Fan, B. Org. Biomol. Chem.
The authors declare no competing financial interest.
2
(
013, 11, 2294.
7) (a) Rasik, C. M.; Brown, M. K. J. Am. Chem. Soc. 2013, 135, 1673.
ACKNOWLEDGMENTS
We thank Indiana University for financial support. Drs. M. Pink
and C.-H. Chen (X-ray), Drs. J. A. Karty and A. M. Hanson
mass spectrosc.), and Drs. F. Gao and E. Twum (NMR) are
■
(b) Rasik, C.; Brown, M. K. Synlett 2014, 25, 760. (c) Rasik, C. M.;
Hong, Y. J.; Tantillo, D. J.; Brown, M. K. Org. Lett. 2014, 16, 5168.
(d) Rasik, C.; Brown, M. K. Angew. Chem., Int. Ed. 2014, 53, 14522.
(
(
8) Wang, Y.; Wei, D.; Li, Z.; Zhu, Y.; Tang, M. J. Phys. Chem. A
014, 118, 4288.
9) (a) Snider, B. B.; Spindell, D. K. J. Org. Chem. 1980, 45, 5017.
b) Hoffmann, H.; Ismail, Z. M.; Weber, A. Tetrahedron Lett. 1981, 22,
acknowledged for their support.
2
(
(
REFERENCES
■
(
1) (a) Lee-Ruff, E.; Mladenova, G. Chem. Rev. 2003, 103, 1449.
b) Namyslo, J. C.; Kaufmann, D. E. Chem. Rev. 2003, 103, 1485.
c) Hoffmann, N. Chem. Rev. 2008, 108, 1052. (d) Bach, T.; Hehn, J.
P. Angew. Chem., Int. Ed. 2011, 50, 1000. (e) Yoon, T. P. ACS Catal.
013, 3, 895.
2) For Lewis acid-catalyzed formal [2 + 2] cycloadditions, see:
a) Hayashi, Y.; Narasaka, K. Chem. Lett. 1989, 793. (b) Ichikawa, Y.-
I.; Narita, A.; Shiozawa, A.; Hayashi, Y.; Narasaka, K. Chem. Commun.
989, 1919. (c) Hayashi, Y.; Narasaka, K. Chem. Lett. 1990, 1295.
d) Hayashi, Y.; Niihata, S.; Narasaka, K. Chem. Lett. 1990, 2091.
e) Narasaka, K.; Kusama, H.; Hayashi, Y. Bull. Chem. Soc. Jpn. 1991,
4, 1471. (f) Hayashi, Y.; Otaka, K.; Saito, N.; Narasaka, K. Bull. Chem.
Soc. Jpn. 1991, 64, 2122. (g) Narasaka, K.; Hayashi, K.; Hayashi, Y.
Tetrahedron 1994, 50, 4529. (h) Narasaka, K.; Hayashi, Y.; Shimadzu,
H.; Niihata, S. J. Am. Chem. Soc. 1992, 114, 8869. (i) Engler, T. A.;
Letavic, M. A.; Iyengar, R.; LaTessa, K. O.; Reddy, J. P. J. Org. Chem.
1
953. (c) Snider, B. B.; Ron, E. J. Org. Chem. 1986, 51, 3643.
(
(
(
(
10) Zhao, J.-F.; Loh, T.-P. Angew. Chem., Int. Ed. 2009, 48, 7232.
11) (a) Corey, E. J.; Letavic, M. A. J. Am. Chem. Soc. 1995, 117,
9
616. (b) Ooi, T.; Ohmatsu, K.; Uraguchi, D.; Maruoka, K.
Tetrahedron Lett. 2004, 45, 4481.
12) (a) Ryu, D. H.; Corey, E. J. J. Am. Chem. Soc. 2003, 125, 6388.
b) Corey, E. J. Angew. Chem., Int. Ed. 2009, 48, 2100. (c) Mukherjee,
S.; Scopton, A. P.; Corey, E. J. Org. Lett. 2010, 12, 1836.
13) Payette, J. N.; Yamamoto, H. Angew. Chem., Int. Ed. 2009, 48,
060.
14) The minor Z-isomer is generated in lower enantioselectivity
∼70:30 er) than the E-isomer (see Scheme 3).
15) For selected notable examples, see: (a) Kolb, H. C.;
VanNieuwenhze, M. S.; Sharpless, K. B. Chem. Rev. 1994, 94, 2483.
b) Negishi, E.-I. Dalton Trans. 2005, 827. (c) Coombs, J. R.;
Haeffner, F.; Kliman, L. T.; Morken, J. P. J. Am. Chem. Soc. 2013, 135,
1222.
16) [ 2 + 2 ], [( 2 + 2 )+ 2 ] or a quasi-pericyclic process with
2
(
(
(
(
1
(
(
6
(
8
(
(
(
(
1
999, 64, 2391. (j) Ito, H.; Hasegawa, M.; Takenaka, Y.; Kobayashi,
1
(
T.; Iguchi, K. J. Am. Chem. Soc. 2004, 126, 4520. (k) Takenaka, Y.; Ito,
H.; Hasegawa, M.; Iguchi, K. Tetrahedron 2006, 62, 3380. (l) Canales,
E.; Corey, E. J. J. Am. Chem. Soc. 2007, 129, 12686. (m) Takenaka, Y.;
Ito, H.; Iguchi, K. Tetrahedron 2007, 63, 510. (n) Ishihara, K.; Fushimi,
M. J. Am. Chem. Soc. 2008, 130, 7532. (o) Schotes, C.; Mezzetti, A.
Angew. Chem., Int. Ed. 2011, 50, 3072.
π
s
π
a
π
s
π
s
π s
carbenoid character have all been proposed for related ketene−alkene
2 + 2] cycloadditions. For a lead reference, see: Wang, X.; Houk, K.
N. J. Am. Chem. Soc. 1990, 112, 1754.
17) (a) Corey, E. J.; Lee, T. W. Chem. Commun. 2001, 1321.
b) Paddon-Row, M. N.; Anderson, C. D.; Houk, K. N. J. Org. Chem.
009, 74, 861.
[
(
(
2
(
(
(
3) For amine catalyzed [2 + 2] cycloadditions reactions, see:
(
(
(
a) Ishihara, K.; Nakano, K. J. Am. Chem. Soc. 2007, 129, 8930.
b) Sakakura, A.; Ishihara, K. Bull. Chem. Soc. Jpn. 2010, 83, 313.
c) Talavera, G.; Reyes, E.; Vicario, J. L.; Carrillo, L. Angew. Chem., Int.
Ed. 2012, 51, 4104. (d) Albrecht, Ł.; Dickmeiss, G.; Acosta, F. C.;
Rodríguez-Escrich, C.; Davis, R. L.; Jørgensen, K. A. J. Am. Chem. Soc.
012, 134, 2543. (e) Duan, G.-J.; Ling, J.-B.; Wang, W.-P.; Luo, Y.-C.;
Xu, P.-F. Chem. Commun. 2013, 49, 4625. (f) Qi, L.-W.; Yang, Y.; Gui,
Y.-Y.; Zhang, Y.; Chen, F.; Tian, F.; Peng, L.; Wang, L.-X. Org. Lett.
18) Vyas, D. J.; Oestreich, M. Angew. Chem., Int. Ed. 2010, 49, 8513.
19) Blakemore, D. C.; Bryans, J. S.; Carnell, P.; Carr, C. L.;
Chessum, N. E. A.; Field, M. J.; Kinsella, N.; Osborne, S. A.; Warren,
A. N.; Williams, S. C. Bioorg. Med. Chem. Lett. 2010, 20, 461.
(
corresponding cyclobutanone by oxidative cleavage in high yield and
er; see the Supporting Information for details.
(
of cycloaddition (17) is due to simultaneous conversion of the E- and
Z-isomers of 17 to 38 (see ref 14).
(
20) The cycloaddition product 20 can also be converted to the
2
21) The lower enantiopurity of 36 compared to the direct product
2
(
4
2
014, 16, 6436.
4) (a) Muller, C.; Bauer, A.; Bach, T. Angew. Chem., Int. Ed. 2009,
8, 6640. (b) Guo, H.; Herdtweck, E.; Bach, T. Angew. Chem., Int. Ed.
010, 49, 7782. (c) Muller, C.; Bauer, A.; Maturi, M. M.; Cuquerella,
̈
22) The benzyl ester was used only because a chromophore was
̈
needed for HPLC analysis with a chiral column.
M. C.; Miranda, M. A.; Bach, T. J. Am. Chem. Soc. 2011, 133, 16689.
d) Austin, K. A. B.; Herdtweck, E.; Bach, T. Angew. Chem., Int. Ed.
(
2
2
011, 50, 8416. (e) Brimioulle, R.; Guo, H.; Bach, T. Chem.Eur. J.
012, 18, 7552. (f) Maturi, M. M.; Wenninger, M.; Alonso, R.; Bauer,
A.; Pothig, A.; Riedle, E.; Bach, T. Chem.Eur. J. 2013, 19, 7461.
̈
(
g) Brimioulle, R.; Bach, T. Science 2013, 342, 840. (h) Alonso, R.;
Bach, T. Angew. Chem., Int. Ed. 2014, 53, 4368. (i) Vallavoju, N.;
Selvakumar, S.; Jockusch, S.; Sibi, M. P.; Sivaguru, J. Angew. Chem., Int.
Ed. 2014, 53, 5604. (j) Maturi, M. M.; Bach, T. Angew. Chem., Int. Ed.
3
485
DOI: 10.1021/jacs.5b00563
J. Am. Chem. Soc. 2015, 137, 3482−3485