Temporary Generation of a Cyclopropyl Oxocarbenium Ion Enables Highly Diastereoselective Donor-Acceptor Cyclopropane Cycloaddition

Article abstract of DOI:10.1002/anie.201601340

A novel formal [3+2] cycloaddition of cyclopropylacetals and aldehydes was developed, and the resulting trisubstituted tetrahydrofurans display three new chiral centers formed with highly diastereoselectivity. This method is stereocomplementary to most previously reported cycloadditions of malonate diesters, relies on the transient generation of cyclopropyl oxocarbenium ions, proceeds under mild conditions, and is based on the concept of temporary activation of an otherwise inert protecting group. Hide-and-seek with a dipole: A novel formal [3+2] cycloaddition of cyclopropylacetals and aldehydes was developed. This reaction affords trisubstituted tetrahydrofurans displaying three newly formed chiral centers with high diastereoselectivity. The reaction relies on the transient generation of cyclopropyl oxocarbenium ions under mild conditions and is based on the concept of temporary activation of an otherwise inert protecting group.

Full text of DOI:10.1002/anie.201601340

Angewandte
Communications
Chemie
International Edition: DOI: 10.1002/anie.201601340
German Edition: DOI: 10.1002/ange.201601340
Cycloaddition
Temporary Generation of a Cyclopropyl Oxocarbenium Ion Enables
Highly Diastereoselective Donor–Acceptor Cyclopropane
Cycloaddition
Juliette Sabbatani and Nuno Maulide*
Abstract: A novel formal [3+2] cycloaddition of cyclopro-
pylacetals and aldehydes was developed, and the resulting
trisubstituted tetrahydrofurans display three new chiral centers
formed with highly diastereoselectivity. This method is stereo-
complementary to most previously reported cycloadditions of
malonate diesters, relies on the transient generation of cyclo-
propyl oxocarbenium ions, proceeds under mild conditions,
and is based on the concept of temporary activation of an
otherwise inert protecting group.
effect[6] and can therefore undergo a broad range of reactions
[5]
=
for transforming the C C double bond.
We envisaged that cyclopropyl oxocarbenium ions, which
are potentially accessible from a-cyclopropylacetals, might be
prone to a similar electronic activation (Scheme 1b). In
particular, we were intrigued by the possibility that appro-
priately substituted cyclopropyl acetals might behave as
entirely new D–A cyclopropanes and thereby enable formal
[3+2] cycloaddition reactions. Notably, such a cycloaddition
would generate three stereogenic centers at once, a rare
occurrence in the D–A cyclopropane literature.[7] As daunting
as this prospect may appear, we were also aware that, if the
acetal moiety was retained intact in the product, a significant
increase in the synthetic versatility of those cycloadducts
would result.
Herein, we report the first [3+2] cycloaddition of cyclo-
propylacetals with aldehydes under mild conditions to
stereoselectively afford trisubstituted tetrahydrofuran prod-
ucts that carry an easily functionalizable masked aldehyde.
A preliminary dipolarophile screening showed that cin-
namaldehyde (2) efficiently undergoes cycloaddition with the
phenyl-substituted cyclopropyl acetal 1 (Table 1, entry 5). We
thus continued our investigations with 1 and 2 as model
P
olysubstituted tetrahydrofurans are prevalent cores in
various biologically active substances.[1] Over the past few
years, perhaps the best developed method for the formation
of these moieties has been the [3+2] cycloaddition of donor–
acceptor (D–A) cyclopropanes with carbonyl derivatives.[2]
Most notably, the groups of Johnson and Waser have
intensively investigated this area and have contributed
significantly to its development towards a highly efficient
and stereoselective method.[3] The quintessential acceptor
group in this chemistry is a malonate diester moiety, which is
activated by Lewis acid coordination and decisively weakens
À
the central cyclopropyl C C bond. Nevertheless, the presence
of this malonate diester moiety in the final cycloadducts
renders the subsequent functionalization of the newly formed
products cumbersome.[4]
Table 1: Selected examples for optimization of the cycloaddition.
As shown in Scheme 1a, a,b-unsaturated acetals are
known to generate vinyl oxocarbenium ions in the presence
of an acid catalyst.[5] These intermediates display a highly
polarized double bond as a result of the so-called vinylogy
Entry 2 [equiv] Catalyst Solvent T [8C] d.r. [a/b/c][a] Yield [%][a]
[b]
1
2
3
4
5
6
7
8
1
1
1
1
1
1
1
1
1
1
1
2
2
2
TfOH
In(OTf)3 CH2Cl2
Dy(OTf)3 CH2Cl2
SnCl4
Sn(OTf)2 CH2Cl2
Sc(OTf)3 CH2Cl2
SnCl2
AuCl3
InCl3
CH2Cl2
À78
23[c]
23[c]
23[c]
23[c]
23[c]
23[c]
23[c]
40[c]
–
–
–
–
–
–
–
[d]
[d]
CH2Cl2
traces
65
34
50
35
66
49
56
82
47:42:11
40:40:20
48:30:22
51:46:3
71:29:0
CH2Cl2
CH2Cl2
CH2Cl2
9
10
11
12
13
14
TMSOTf CH2Cl2
À78 67:33:0
Scheme 1. Cyclopropyl oxocarbenium ions as D–A cyclopopanes.
TMSOTf CH3NO2 À25 87:13:0
TMSOTf CH3NO2 À25 87:13:0
TBSOTf
CH3NO2 08C
83:17:0
99
97
TBSOTf CH3NO2 À25 90:10:0
[*] J. Sabbatani, Prof. Dr. N. Maulide
[a] Determined by 1H NMR spectroscopic analysis of the non-purified
reaction mixture with 1,3,5-trimethoxybenzene as the internal standard.
[b] Degradation. [c] All of the screening experiments were started at
À788C, and if no conversion was observed, the temperature was slowly
raised until some reactivity was observed. [d] No conversion was
observed.
Fakultät für Chemie, Institut für Organische Chemie
Universität Wien, Währinger Straße 38, 1090 Wien (Austria)
E-mail: nuno.maulide@univie.ac.at
Homepage: http://maulide.univie.ac.at
Supporting information for this article can be found under:
http://dx.doi.org/10.1002/anie.201601340.
6780
ꢀ 2016 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Angew. Chem. Int. Ed. 2016, 55, 6780 –6783

Angewandte
Communications
Chemie
substrates. Weak Brønsted acids were not able to promote any
conversion, whereas strong acids such as trifluoromethane-
sulfonic acid led to the rapid degradation of 1 (Table 1,
entry 1) even at low temperature. Hence, we focused our
attention on Lewis acid catalysts. Although some of the
classically used Lewis acids in cycloaddition chemistry
displayed no catalytic activity in this reaction (Table 1,
entries 2–4), several others proved to be efficient (Table 1,
entries 5–8), including tin-, scandium-, and gold-based spe-
cies. However, in these early experiments, a distinct lack of
diastereoselectivity was observed, with synthetically unap-
pealing mixtures of three out of the four possible diastereo-
isomers (3a, 3b, 3c) being routinely observed in varying
amounts.
Pleasingly, indium(III) trichloride and trimethylsilyl tri-
fluoromethanesulfonate remarkably improved the diastereo-
selectivity in favor of the formation of 3a (Table 1, entry 9).
After optimizing the solvent, the temperature, and the
stoichiometry of the reaction we found that the best results
were reproducibly obtained when using 10 mol% of tert-
butyldimethylsilyl trifluoromethanesulfonate and 2 equiva-
lents of cinnamaldehyde in nitromethane as the solvent, at
À258C (Table 1, entry 14). Under these conditions, 3a was
obtained in an excellent 97% yield as measured by NMR
(90% yield of isolated product) as a 9:1 mixture of
diastereomers.
From the outset, the observed dominance of trans,trans-
tetrahydrofuran 3a as a product was striking; the vast
majority of the reported examples of [3+2]-cycloadditions
between D–A cyclopropanes and aldehydes describe stereo-
selective access to cis-2,5-tetrahydrofuran adducts.[2] This
renders our method, which is able to stereoselectively deliver
a 2,5-trans-configured tetrahydrofuran, stereocomplementary
to those prior approaches. Interestingly, the starting cyclo-
propylacetal features a trans relationship between the arene
and acetal, which is reversed to cis in the cycloaddition
product.
Scheme 2. Scope of the cycloaddition: variation of the aldehyde
partner. Reaction conditions: 1a (0.1 mmol), 2 (0.2 mmol), TBSOTf
(0.01 mmol), nitromethane (1 mL), À258C, 2 h. Yields refer to isolated
major diastereoisomers; d.r. values were measured by 1H-NMR
spectroscopic analysis of the non-purified mixture. [a] TMSOTf was
used as a Lewis acid.
Importantly, control experiments showed that the diaste-
reomeric ratio of product 3 under these conditions remains
constant at lower conversions, and is identical to the final
value. This suggests that the observed stereoselectivity is not
the result of subsequent thermodynamic equilibration.[8]
To explore the generality of the reaction, various unsatu-
rated aldehydes were subjected to [3+2] cycloaddition with
1 (Scheme 2). A wide range of substrates was studied, and
smooth cycloadditions occurred within a short reaction time
(2 h).[9] The reaction displayed satisfactory functional-group
tolerance, and both internal and terminal alkenes (8 and 12)
were tolerated. Additionally, aliphatic and aromatic halides
(13 and 15), esters (14), and protected alcohols (10) proved to
be compatible with the mild conditions employed in this
transformation. Notably, high diastereoselectivity was
obtained in favor of the products shown. As depicted, all
products contain the masked aldehyde ultimately responsible
for the observed reactivity.
Scheme 3. Scope of the cycloaddition: variation of the cyclopropane
partner. Reaction conditions: 1 (0.1 mmol), 2 (0.2 mmol), TBSOTf
(0.01 mmol), nitromethane (1 mL), À258C, 2 h. Yields refer to isolated
major diastereoisomers; d.r. values were measured by 1H-NMR
spectroscopic analysis of the non-purified mixture. [a] The reaction was
performed at À158C.
At this juncture, we were eager to examine diverse
cyclopropyl acetal derivatives and study their applicability to
this transformation, and the results are shown in Scheme 3.
Different aromatic residues, including naphthyl, p-tolyl, and
thienyl residues, were tolerated in the reaction, smoothly
delivering the tetrahydrofuran cycloadducts. Although the
yields are mostly moderate (a consequence of the somewhat
Angew. Chem. Int. Ed. 2016, 55, 6780 –6783
ꢀ 2016 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.angewandte.org
6781

Angewandte
Communications
Chemie
lower stability of cyclopropane acetals 1 under these con-
ditions), the increase in structural and stereochemical com-
plexity enabled by this highly stereoselective cycloaddition
reaction is noteworthy.
In spite of the considerable synthetic potential provided
by using a,b-unsaturated aldehydes as dipolarophiles (see
below), given the novelty of this system, we were interested in
studying the behavior of aromatic aldehydes in our cyclo-
addition reaction. In the event, these were considerably more
reactive than their enal analogues. Indeed, electron-rich
aromatic aldehydes afforded satisfying yields and diastereo-
selectivity under even milder conditions (À508C).
The reaction enabled the use of a thiophene carbaldehyde
(25, Scheme 4), and even a free phenol (28) was tolerated
under these conditions. However, the use of electron-neutral
or electron-poor substrates afforded lower stereoselectivity,
Scheme 5. Possible mechanisms for the cycloaddition of cyclopropyla-
cetals and aldehydes.
pathways, as well as to clarify whether the substitution step is
SN1- or SN2-like (via an intimate ion pair, as originally
proposed by Johnson).[2j]
These reactions deliver densely functionalized tetrahy-
drofuran products that can be readily modified through
standard synthetic procedures. As shown in Scheme 6, acetal
removal to give the aldehyde is easily achieved without
noticeable epimerization and in good yield. Once revealed,
the aldehyde moiety can readily engage in olefination,
reduction, or reductive amination reactions. Additionally,
the olefinic moiety of product 4 can itself be used as
a surrogate for an aldehyde, generating the differentiated
(through protection) dialdehyde 34 by oxidative cleavage.
It should be noted that this mode of activation of the
Scheme 4. Scope of the cycloaddition with aryl-substituted aldehydes.
Reaction conditions: 1a (0.1 mmol), 2 (0.2 mmol), TBSOTf
(0.01 mmol), nitroethane (1 mL), À508C, 2 h. Yields refer to isolated
major diastereoisomers; d.r. values were measured by 1H-NMR
spectroscopic analysis of the non-purified mixture. [a] The reaction was
performed in nitromethane at À258C. [b] The reaction was performed
at À408C. [c] Yield refers to the isomer mixture. [d] The temperature
was raised to 508C.
À
cyclopropane C C bond is not limited to cycloaddition
although we observed faster conversion of the cyclopropane
acetal (40, Scheme 4), and aliphatic aldehydes were not
reactive enough, even at higher temperatures (41, Scheme 4).
These observations, along with the qualitatively higher
reactivity of aromatic aldehydes versus enals, allow two
possible interpretations concerning the actual mechanism of
this cycloaddition (Scheme 5). Assuming that cyclopropyla-
cetal 1 undergoes Lewis acid triggered ring opening to give
stabilized carbenium ion C, attack by the aldehyde carbonyl
would lead to the intermediate D (Scheme 5, cycle I), from
which cyclisation to product E could take place. While this
pathway is attractive, given our success with electron-rich
aromatic aldehydes, the observed trans,trans-diastereoselec-
tivity is not easily rationalized on this basis. Alternatively,
Mukaiyama-type aldol reaction (C’!D’) followed by ether–
carbenium ion cyclisation (Scheme 5, cycle II) might also
account for the formation of E. Additional mechanistic
studies will be required to distinguish between these two
Scheme 6. Functionalization of tetrahydrofuran 4.
6782 www.angewandte.org
ꢀ 2016 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Angew. Chem. Int. Ed. 2016, 55, 6780 –6783

Angewandte
Communications
Chemie
Y. Sugita, K. Kawai, Heterocycles 2001, 55, 135 – 144; h) K.
reactions, and the nucleophilic addition of a simple alcohol
can also be achieved at room temperature under tin(III)
tin(II) catalysis. This gives the ether 35 in high yield
(Scheme 7).
Oshima, Z. Han, S. Uehira, T. Tsuritani, H. Shinokubo,
Tetrahedron 2001, 57, 987 – 995; i) J. Johnson, P. Pohlhaus, J.
Am. Chem. Soc. 2005, 127, 16014 – 16015; j) P. D. Pohlhaus, S. D.
Sanders, A. T. Parsons, W. Li, J. S. Johnson, J. Am. Chem. Soc.
2008, 130, 8642 – 8650; k) J. Johnson, A. Parsons, J. Am. Chem.
Soc. 2009, 131, 3122 – 3123; l) S. Xing, W. Pan, C. Liu, J. Ren, Z.
Wang, Angew. Chem. Int. Ed. 2010, 49, 3215 – 3218; Angew.
Chem. 2010, 122, 3283 – 3286; m) J. Johnson, A. Smith, M. Slade,
Org. Lett. 2011, 13, 1996 – 1999; n) G. Yang, Y. Shen, K. Li, Y.
Sun, Y. Hua, J. Org. Chem. 2011, 76, 229 – 233; o) A. Dobbs, J.
Dunn, Tetrahedron Lett. 2012, 53, 2392 – 2395; p) J. Waser, F.
Benfatti, F. Nanteuil, Org. Lett. 2012, 14, 386 – 389; q) J. Waser,
F. Benfatti, F. Nanteuil, Chem. Eur. J. 2012, 18, 4844 – 4849; r) G.
Yang, Y. Sun, Y. Shen, Z. Chai, S. Zhou, J. Chu, J. Chai, J. Org.
Chem. 2013, 78, 5393 – 5400; s) J. Waser, F. Nanteuil, E. Serrano,
D. Perrotta, J. Am. Chem. Soc. 2014, 136, 6239 – 6242.
Scheme 7. Ring opening of cyclopropane 1a with butanol.
In summary, we present herein the first examples of the
generation and exploitation of cyclopropyl oxocarbenium
ions as reagents in synthesis. Suitably substituted derivatives
thereof can serve as donor–acceptor cyclopropane building
blocks that engage aldehydes in highly diastereoselective
formal [3+2] cycloaddition. This chemistry enables the
de novo generation of three stereogenic centers with high
diastereoselectivity and proceeds in a stereocomplementary
fashion to the known chemistry of cyclopropane diesters.
Notably, the use of an acetal as the acceptor moiety leads to
the interesting feature of having a protecting group serve as
a platform for the temporary generation of an electrophile, an
interesting concept for which there are only scattered
examples in the literature to date[10] and which is bound to
find growing utility in organic synthesis.
[3] Reviews: a) M. Kerr, C. Carson, Chem. Soc. Rev. 2009, 38, 3051 –
3060; b) J. Waser, Top. Heterocycl. Chem. 2013, 32, 225 – 270;
c) S. France, M. Cavitt, L. Phun, Chem. Soc. Rev. 2014, 43, 804 –
818; d) T. Schneider, J. Kaschel, D. Werz, Angew. Chem. Int. Ed.
2014, 53, 5504 – 5523; Angew. Chem. 2014, 126, 5608 – 5628.
[4] Elegant examples: a) M. Kerr, I. Young, J. Am. Chem. Soc. 2007,
129, 1465 – 1469; b) M. Kerr, M. Johansen, Org. Lett. 2008, 10,
3497 – 3500.
[5] Selected examples: a) M. Harmata, P. Rashatasakhon, Tetrahe-
dron 2003, 59, 2371 – 2395; b) E. Nakamura, M. Nakamura, H.
Isobe, Chem. Rev. 2003, 103, 1295 – 1326; c) R. Hsung, C. Ko, J.
Feltenberger, S. Ghosh, Org. Lett. 2008, 10, 1971 – 1974; d) R.
Hsung, J. Feltenberger, C. Ko, J. Deng, S. Ghosh, Heterocycles
2012, 84, 843 – 878; e) R. Hsung, J. Deng, C. Ko, Org. Lett. 2012,
14, 5562 – 5565; f) P. Nagorny, A. Borovika, Tetrahedron 2013,
69, 5719 – 5725.
[6] a) R. Fuson, Chem. Rev. 1935, 16, 1 – 27; b) R. Fuson, R. Christ,
J. Am. Chem. Soc. 1937, 59, 893 – 897.
Acknowledgments
[7] a) J. Johnson, M. Campbell, A. Parsons, P. Pohlhaus, S. Sanders,
J. Org. Chem. 2010, 75, 6317 – 6325 and Ref. [3d].
[8] Taking each one of the isolated product isomers and resubmit-
ting them to the reaction conditions led to no isomerization even
after several hours. This rules out epimerization or interconver-
sion of isomers under the reaction conditions.
[9] 2 h was the chosen standard time to reach full conversion. When
reactions were completed in a shorter time, prolonged stirring
with the catalyst at À258C caused no erosion in yield or d.r.. See
the Supporting Information for details.
Support of this research by the University of Vienna and the
Deutsche Forschungsgemeinschaft (Grant MA 4861/3-1) is
generously acknowledged. Dr. H.-P. Kählig (U. of Vienna) is
thanked for extensive assistance with NMR analysis and
assignment. R. Oost (U. Vienna) is gratefully acknowledged
for initial experiments.
Keywords: cycloaddition · cyclopropane · donor–
acceptor systems · oxocarbenium ions · tetrahydrofuran
[10] An early example: a) S. S. Vershinin, R. M. Makayeva, L. N.
Zorina, V. V. Zorin, D. L. Rahmankulov, Russ. J. Gen. Chem.
1996, 66, 1147 – 1149. A masked donor– acceptor cyclopropane
carrying an olefin: b) V. Ganesh, T. Kundu, S. Chandrasekaran,
J. Org. Chem. 2013, 78, 380 – 399. Catalysis through temporary
intramolecularity: c) N. Guimond, M. J. MacDonald, V.
Lemieux, A. M. Beauchemin, J. Am. Chem. Soc. 2012, 134,
16571 – 16577.
How to cite: Angew. Chem. Int. Ed. 2016, 55, 6780–6783
Angew. Chem. 2016, 128, 6892–6895
[1] J. Wolfe, M. Ha, Tetrahedron 2007, 63, 261 – 290.
[2] a) H. Reissig, Tetrahedron Lett. 1981, 22, 2981 – 2984; b) A.
Daniewski, T. Kowalczyk-Przewloka, J. Org. Chem. 1985, 50,
2976 – 2980; c) H. Reissig, H. Holzinger, G. Glomsda, Tetrahe-
dron 1989, 45, 3139 – 3150; d) H. Reissig, C. Brueckner, H.
Holzinger, J. Org. Chem. 1988, 53, 2450 – 2456; e) E. Nakamura,
S. Yamago, J. Org. Chem. 1990, 55, 5553 – 5555; f) I. Yokoe, Y.
Sugita, K. Kawai, Heterocycles 2000, 53, 657 – 664; g) I. Yokoe,
Received: February 5, 2016
Revised: March 7, 2016
Published online: April 21, 2016
Angew. Chem. Int. Ed. 2016, 55, 6780 –6783
ꢀ 2016 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.angewandte.org
6783