Enantioselective Oxetane Ring Opening with Chloride: Unusual Use of Wet Molecular Sieves for the Controlled Release of HCl

Article abstract of DOI:10.1002/anie.201601844

An unprecedented enantioselective oxetane opening with chloride provides access to a range of highly functionalized three-carbon building blocks. The excellent enantiocontrol is enabled not only by a new catalyst, but also by the unusual use of wet molecular sieves for the controlled release of HCl. The enantioselective ring opening of oxetanes with chloride provides access to a range of highly functionalized three-carbon building blocks. The use of a new catalyst in combination with wet molecular sieves for the controlled release of HCl leads to high enantioselectivities.

Full text of DOI:10.1002/anie.201601844

Angewandte
Communications
Chemie
International Edition: DOI: 10.1002/anie.201601844
German Edition: DOI: 10.1002/ange.201601844
Organocatalysis
Enantioselective Oxetane Ring Opening with Chloride: Unusual Use
of Wet Molecular Sieves for the Controlled Release of HCl
Wen Yang, Zhaobin Wang, and Jianwei Sun*
Abstract: An unprecedented enantioselective oxetane opening
with chloride provides access to a range of highly functional-
ized three-carbon building blocks. The excellent enantiocon-
trol is enabled not only by a new catalyst, but also by the
unusual use of wet molecular sieves for the controlled release
of HCl.
serve as a multi-purpose handle for the attachment of various
nucleophiles, thereby indirectly achieving oxetane desym-
metrization with different nucleophiles (Scheme 1). How-
ever, aside from the weak nucleophilicity of the chloride
anion and the ability of the alcohol product to act as
a competing nucleophile as obvious obstacles, the potential
use of HCl or its precursors would result in strong background
reactions, particularly with chiral acid catalysis. In this
context, we herein describe our successful realization of this
process with excellent stereocontrol under mild conditions.
In view of the weak chloride nucleophilicity relative to the
resulting alcohol product, we surmised that a latent chloride
nucleophile (e.g., a chlorosilane) would not only provide
a chloride ion upon activation, but also concomitantly protect
the resulting hydroxy group in the same catalytic cycle to
inhibit product competition. Indeed, to our delight, a combi-
nation of chlorotriethoxysilane (2a) and a chiral phosphoric
acid could effect the desired transformation [Eq. (1)].[6] In the
N
ew strategies for the efficient synthesis of chiral building
blocks are in great demand in organic synthesis.[1] The
enantioselective desymmetrization of prochiral compounds
has been shown to be one of the most powerful strategies.[2]
Specifically, the enantioselective opening of prochiral oxe-
tanes represents an attractive method for the rapid synthesis
of highly functionalized three-carbon chiral building blocks
(Scheme 1). However, these reactions remain a challenge,
particularly with intermolecular nucleophiles.[3–5]
Scheme 1. Catalytic asymmetric intermolecular ring opening of prochi-
ral oxetanes.
presence of the representative STRIP catalyst (A1, shown in
Table 2), chloride opening of 1a in toluene as the solvent
proceeded smoothly at room temperature. After TBAF work-
up, the desired product 3a was obtained. However, the initial
results showed very low reproducibility. Different runs of the
same reaction with different batches of material gave quite
different conversions and enantioselectivities (20–100%
conversion, 45–58% ee), which were beyond experimental
error. We reasoned that the condition of the chlorosilane and
substrate could influence the result through the presence of
adventitious moisture and/or HCl, which can promote the
background reaction. However, the use of freshly distilled 2a
did not solve the reproducibility issue.
The sensitivity of the results to seemingly random factors
prompted us to probe the effect of water (Table 1). In the
presence of 5 Š molecular sieves (M.S.), almost no conversion
was observed (entry 1). However, the addition of one
equivalent of water resulted in full conversion, but low
enantioselectivity (entry 2). Alternatively, the addition of
both molecular sieves and water gave moderate conversion,
but higher enantioselectivity (entry 3). Increasing the amount
of water to two equivalents improved the reaction to full
conversion and higher enantioselectivity (entry 4). A further
The challenges associated with intermolecular oxetane
desymmetrization include remote stereocontrol and the low
ring-opening propensity of oxetanes. The latter requires
a suitable reaction partner that is sufficiently nucleophilic
for ring opening but does not deactivate the catalyst.
Furthermore, the alcohol product can also be a competing
nucleophile. As a result, such reactions are currently limited
to particular sulfur nucleophiles and have limited applica-
tions.[4c]
Nevertheless, we hypothesized that the desymmetrization
of oxetanes with a chloride nucleophile would be highly useful
as the chloromethyl moiety in the ring-opening product could
[*] W. Yang, Z. Wang, Prof. J. Sun
Department of Chemistry
The Hong Kong University of Science and Technology
Clear Water Bay, Kowloon, Hong Kong SAR (China)
E-mail: sunjw@ust.hk
Supporting information for this article can be found under:
http://dx.doi.org/10.1002/anie.201601844.
6954
ꢀ 2016 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Angew. Chem. Int. Ed. 2016, 55, 6954 –6958

Angewandte
Communications
Chemie
Table 1: Additive screening.
Table 2: Optimization of the reaction conditions.
Entry Additive
Conv. [%][a] ee [%][b]
1
2
3
4
5
6
5 Š M.S. (10 mg)
H2O (1.0 equiv)
5 Š M.S. (10 mg) and H2O (1.0 equiv)
5 Š M.S. (10 mg) and H2O (2.0 equiv)
5 Š M.S. (10 mg) and H2O (3.0 equiv)
wet 5 Š M.S. (12 mg)[c]
<10
100
52
100
100
100
–
46
58
67
63
67
Entry
Catalyst (mol%)
Solvent
Conv. [%][a]
ee [%][b]
1
2
3
4
(S)-A1 (5)
(S)-A2 (5)
(S)-A2 (5)
(S)-A2 (5)
(S)-A2 (5)
(S)-A2 (2)
(S)-A2 (1)
(S)-A2 (2)
toluene
toluene
DCM
100
100
100
76
100
100
100
100
74
91
82
44
93
92
90
93
[a] Determined by 1H NMR analysis using CH2Br2 as the internal
standard. [b] Determined by HPLC analysis on a chiral stationary phase.
[c] An aliquot of premixed wet M.S. (12 mg) containing approximately
2.0 equiv of water was used.
Et2O
5
benzene
benzene
benzene
benzene
6[c]
7[c]
8[c,d]
[a] Determined by 1H NMR analysis using CH2Br2 as the internal
standard. [b] Determined by HPLC analysis on a chiral stationary phase.
[c] Run for 24 h. [d] 0.05m concentration. Cy =cyclohexyl.
increase in the amount of water could not improve the
enantioselectivity. These results indicate that water is essen-
tial to the reaction and more importantly that its amount can
dramatically influence the outcome of the reaction. It is worth
noting that the simultaneous use of water and molecular
sieves has little precedence.[7] However, this combination
proved to be beneficial and made the results reproducible. We
hypothesized that this combination could help the slow
generation of HCl for the reaction. Furthermore, to simplify
the experimental method, we premixed 5 Š M.S. and water on
large scale (12 g), and an aliquot containing about two
equivalents of water was taken for the reaction (entry 6),
which resulted in essentially the same results as those
achieved with the sequential addition of dry M.S. and water
(entry 4).
with respect to the reaction concentration led the optimal set
of parameters (entry 8).
A range of 3-substituted oxetanes smoothly participated
in the ring opening with excellent efficiency and good to high
enantioselectivity (Table 3). Electron-donating and -with-
drawing groups did not affect the reaction. Substrates with
3-alkyl and 3-alkenyl substituents also reacted efficiently.
Heteroatom-substituted oxetanes (e.g., with O-, S-, or N-
containing substituents) were also excellent substrates and
rapidly furnished densely functionalized three-carbon chiral
building blocks (Table 4).[8] Under the mild reaction con-
ditions, a wide variety of functional groups, such as ethers,
esters, silyl-protected alcohols, tosylates, alkenes, alkynes, and
nitriles, are tolerated. Finally, fully substituted stereocenters
could also be generated [Eq. (3)]. Notably, in these two cases,
MgSO4 was used instead of the molecular sieves (see below).
With the above reproducible results as a benchmark, we
next examined different chlorosilanes [Eq. (2)]. As shown
below, (MeO)3SiCl gave the best reactivity and enantioselec-
tivity (76% ee). Bulkier chlorotrialkoxysilanes and chloro-
trialkylsilanes were less reactive (see the Supporting Infor-
mation for details).
Table 3: Substrate scope with respect to 3-carbon-substituted oxetanes.
Entry
R
t [h]
x
3
Yield [%][a]
ee [%]
Next, we further evaluated different catalysts and reaction
parameters (Table 2). The use of various known chiral
phosphoric acids did not improve the enantioselectivity (see
the Supporting Information for details). However, after
considerable efforts, the new catalyst A2, which bears 2,4,6-
tricyclohexylphenyl groups at the 3- and 3’-positions of the
SPINOL backbone, was shown to be superior, providing both
full conversion and excellent enantioselectivity (91% ee,
entry 2). The solvent has a dramatic influence on the
enantioselectivity, with benzene providing the best results
among all evaluated solvents. Decreasing the catalyst loading
to 1–2 mol% does not affect the outcome of the reaction,
albeit at the expense of reaction time. Further investigations
1
2
3
4
5
6
7
8
Ph
24
18
24
24
24
24
24
24
2
2
5
2
2
2
2
2
3a
3b
3c
3d
3e
3 f
91
91
85
90
90
83
92
85
94
94
90
94
96
95
94
96
4-MeC6H4
2-MeOC6H4
3-BrC6H4
3-CF3C6H4
4-CF3OC6H4
4-CNC6H4
2-naphthyl
3g
3h
9
15
5
3i
96
96
10
11
16
24
5
5
3j
93
81
89
71
Bn
3k
[a] Yield of isolated product.
Angew. Chem. Int. Ed. 2016, 55, 6954 –6958
ꢀ 2016 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.angewandte.org
6955

Angewandte
Communications
Chemie
Table 4: Substrate scope with respect to 3-heteroatom-substituted
oxetanes.
Scheme 2. Further functionalization of the desymmetrization products.
dba=dibenzylideneacetone, Ts=para-toluenesulfonyl.
Bz=benzoyl, TBDPS=tert-butyldiphenylsilyl.
To probe the general superiority of our new catalyst A2
over the known versatile catalyst A1 (STRIP), we compared
their performance with several substrates. As shown below,
under otherwise identical reaction conditions, the enantiose-
lectivities achieved with catalyst A2 are all higher by about
20%, although their structural differences appear to be trivial
[Eq. (4)]. Our new catalyst A2 should thus find more
applications.
Scheme 3. Indirect desymmetrization reactions with other nucleo-
philes. PMP=para-methoxyphenyl.
this stage (Scheme 3). We also synthesized triazole 17, an
important intermediate towards the antifungal agents
ZD0870 and Sch45450.[9] The enantioenriched chloride 7c
could be readily synthesized according to our desymmetriza-
tion method. Subsequently, a simple substitution delivered 17.
Its optical purity could be improved to > 99% by a single
recrystallization (Scheme 4).
We carried out a series of control experiments to probe
the mechanism (Table 5). In the absence of catalyst A2, the
reaction still proceeded to yield the desired product, which,
however, was formed as a racemate, representing a back-
ground reaction (entry 1). However, with dry molecular
sieves, almost no conversion was observed, either with or
without catalyst A2. These results further confirm that water
is essential to this reaction and suggest that the in situ
generated HCl is the nucleophile. The direct use of an HCl
The desymmetrization products can be transformed into
other useful chiral molecules. For example, the heterocycles
8–11 could be efficiently synthesized without erosion of
enantiopurity (Scheme 2). Furthermore, this chloride opening
could serve as a bridge to achieve indirect oxetane desym-
metrization with other nucleophiles owing to the versatility of
the alkyl chloride moiety. Simple substitution with different
nucleophiles (e.g., NaCN, NaN3, AcSK, and PMPSH) easily
produced the highly enantioenriched derivatives 12–15, which
cannot be synthesized by direct oxetane desymmetrization at
6956 www.angewandte.org
ꢀ 2016 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Angew. Chem. Int. Ed. 2016, 55, 6954 –6958

Angewandte
Communications
Chemie
Scheme 4. Synthesis of a key intermediate for triazole-containing
antifungal agents.
Table 5: Mechanistic control experiments.
Figure 1. 31P NMR analysis of the possible catalyst states.
2,6-di-tert-butylpyridine was added to the standard reaction as
it is known to change the outcome of proton-involving
pathways. In fact, very low enantioselectivity was observed
[Eq. (6)]. These results are consistent with the hypothesis that
the free acid A2 is the actual catalyst under the standard
reaction conditions.[12]
Entry Variation from the
“standard conditions”
Conv. [%] ee [%]
1
2
3
no (S)-A2
100
<10
<10
0
–
–
dry 5 Š M.S. instead of wet 5 Š M.S.
dry 5 Š M.S. instead of wet 5 Š M.S.,
no (S)-A2
4
5
6
HCl in dioxane (4m, 2.0 equiv),
no (MeO)3SiCl, no 5 Š M.S.
HCl in dioxane (4m, 1.0 equiv),
no (MeO)3SiCl, no 5 Š M.S.
sequential addition of dry MgSO4 and H2O
(2.0 equiv) instead of wet 5 Š M.S.
100
100
100
57
78
92
solution in dioxane resulted in full conversion, but moderate
enantioselectivity (entry 4). Higher enantioselectivity was
observed with a smaller amount of HCl (entry 5), suggesting
that a higher HCl concentration enhances the background
reaction. Furthermore, aside from molecular sieves, MgSO4
worked equally well as an additive in combination with water
(entry 6), suggesting that the role of the molecular sieves is to
serve as a water carrier for its even dispersion in the reaction
mixture for slow HCl generation. The controlled release of
HCl is essential to keep the HCl concentration low and thus
inhibit the background reaction, thereby contributing to the
high enantioselectivity.[10]
In summary, we have developed the first asymmetric
oxetane opening with chloride, providing expedient access to
highly functionalized three-carbon chiral building blocks with
excellent efficiency and stereocontrol. Several notable chal-
lenges, such as the low chloride nucleophilicity, remote
stereocontrol, and possible background reactions, were over-
come by the employment of a suitable chloride source, a new
catalyst, and the unprecedented use of wet molecular sieves
for controlled HCl release. Mechanistic studies suggest that
the free chiral acid is likely to be the actual catalyst. The
unusual reaction conditions should enable the development
of further asymmetric reactions that involve the use of HCl.
To determine the actual catalyst of this transformation
(free A2 or the silyl phosphate A2-Si),[11] we monitored the
reaction by 31P NMR spectroscopy. Indeed, mixing
(MeO)3SiCl and the chiral phosphoric acid A2 did result in
the partial formation of the corresponding silyl phosphate
(A2-Si), whose resonance at À21 ppm was confirmed by its
separate synthesis [Eq. (5)]. However, in the standard
reaction mixture at partial conversion, almost only the free
acid A2 was observed (Figure 1c) as the catalyst resting state,
which is thus highly likely to be the actual catalyst. Finally,
Acknowledgements
Financial support was provided by HKUST and the Hong
Kong RGC (GRF604513, 16304115, ECS605812, and
M-HKUST607/12). We thank Dr. Herman Sung for assistance
with structure determination.
Keywords: asymmetric catalysis · desymmetrization ·
organocatalysis · oxetanes · strained molecules
How to cite: Angew. Chem. Int. Ed. 2016, 55, 6954–6958
Angew. Chem. 2016, 128, 7068–7072
[1] E. M. Carreira, H. Yamamoto, Comprehensive Chirality, Elsev-
ier, Amsterdam, 2012.
[2] a) E. García-Urdiales, I. Alfonso, V. Gotor, Chem. Rev. 2005,
105, 313 – 354; b) S. Rendler, M. Oestreich, Angew. Chem. Int.
Angew. Chem. Int. Ed. 2016, 55, 6954 –6958
ꢀ 2016 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.angewandte.org
6957

Angewandte
Communications
Chemie
Ed. 2008, 47, 248 – 250; Angew. Chem. 2008, 120, 254 – 257;
c) M. D. Díaz de Villegas, J. A. Gµlvez, P. Etayo, R. Badorrey, P.
López-Ram-de-Víu, Chem. Soc. Rev. 2011, 40, 5564 – 5578.
[3] a) J. A. Burkhard, G. Wuitschik, M. Rogers-Evans, K. Müller,
E. M. Carreira, Angew. Chem. Int. Ed. 2010, 49, 9052 – 9067;
Angew. Chem. 2010, 122, 9236 – 9251; b) E. M. Carreira, T. C.
Fessard, Chem. Rev. 2014, 114, 8257 – 8322; c) Z. Wang, Z. Chen,
J. Sun, Org. Biomol. Chem. 2014, 12, 6028 – 6032.
[4] For intermolecular desymmetrization reactions, see: a) M.
Mizuno, M. Kanai, A. Iida, K. Tomioka, Tetrahedron: Asymme-
try 1996, 7, 2483 – 2484; b) M. Mizuno, M. Kanai, A. Iida, K.
Tomioka, Tetrahedron 1997, 53, 10699 – 10708; c) Z. Wang, Z.
Chen, J. Sun, Angew. Chem. Int. Ed. 2013, 52, 6685 – 6688;
Angew. Chem. 2013, 125, 6817 – 6820.
3056; i) D. Parmar, E. Sugiono, S. Raja, M. Rueping, Chem. Rev.
2014, 114, 9047 – 9153; j) C. Zhu, K. Saito, M. Yamanaka, T.
Akiyama, Acc. Chem. Res. 2015, 48, 388 – 398; k) T. Akiyama, K.
Mori, Chem. Rev. 2015, 115, 9277 – 9306; l) T. James, M.
van Gemmeren, B. List, Chem. Rev. 2015, 115, 9388 – 9409.
[7] For an example, see: S. Abbaraju, M. Bhanushali, C.-G. Zhao,
Tetrahedron 2011, 67, 7479 – 7484.
[8] For additional examples with moderate enantioselectivity or
conversion, see Scheme S1 in the Supporting Information.
[9] a) T. Peng, J. N. Galgiani, Antimicrob. Agents Chemother. 1993,
37, 2126 – 2131; b) K. Murakami, H. Mochizuki, F. Boyle, J. M.
Wardleworth, Eur. Pat. Appl. 472392, 1992; for another example
of its synthesis, see: c) M. Suzuki, N. Kato, M. Kanai, M.
Shibasaki, Org. Lett. 2005, 7, 2527 – 2530.
[5] For other asymmetric reactions of oxetanes, including intra-
molecular desymmetrization reactions, see: a) H. Nozaki, S.
Moriuti, H. Takaya, R. Noyori, Tetrahedron Lett. 1966, 5, 5239 –
5244; b) K. Ito, T. Katsuki, Chem. Lett. 1994, 1857 – 1860;
c) M. M.-C. Lo, G. C. Fu, Tetrahedron 2001, 57, 2621 – 2634;
d) R. N. Loy, E. N. Jacobsen, J. Am. Chem. Soc. 2009, 131, 2786 –
2787; e) B. Guo, G. Schwarzwalder, J. T. Njardarson, Angew.
Chem. Int. Ed. 2012, 51, 5675 – 5678; Angew. Chem. 2012, 124,
5773 – 5776; f) Q. Yin, S.-L. You, Org. Lett. 2014, 16, 1810 – 1813;
g) Z. Chen, B. Wang, Z. Wang, G. Zhu, J. Sun, Angew. Chem. Int.
Ed. 2013, 52, 2027 – 2031; Angew. Chem. 2013, 125, 2081 – 2085;
h) Z. Chen, Z. Wang, J. Sun, Chem. Eur. J. 2013, 19, 8426 – 8430;
i) W. Yang, J. Sun, Angew. Chem. Int. Ed. 2016, 55, 1868 – 1871;
Angew. Chem. 2016, 128, 1900 – 1903.
[6] For selected reviews on chiral phosphoric acid and Brønsted acid
catalysis, see: a) T. Akiyama, Chem. Rev. 2007, 107, 5744 – 5758;
b) M. Terada, Chem. Commun. 2008, 4097 – 4112; c) G. Adair, S.
Mukherjee, B. List, Aldrichimica Acta 2008, 41, 31 – 39; d) S.-L.
You, Q. Cai, M. Zeng, Chem. Soc. Rev. 2009, 38, 2190 – 2201;
e) D. Kampen, C. M. Reisinger, B. List, Top. Curr. Chem. 2010,
291, 395 – 456; f) M. Terada, Synthesis 2010, 1929 – 1982; g) J. Yu,
F. Shi, L.-Z. Gong, Acc. Chem. Res. 2011, 44, 1156 – 1171;
h) C. H. Cheon, H. Yamamoto, Chem. Commun. 2011, 47, 3043 –
[10] Notably, it was found that catalyst A2 accelerates the back-
ground reaction, which explains the observed enantioselectivity
(see the Supporting Information for kinetic studies). A similar
phenomenon has been previously observed; see: T. Mita, E. N.
Jacobsen, Synlett 2009, 1680 – 1684. Free HCl has rarely been
used in asymmetric catalysis.
[11] Silyl phosphates or the related sulfonimides have been proposed
to be the actual catalysts generated in situ; see: a) E. B. Row-
land, G. B. Rowland, E. Rivera-Otero, J. C. Antilla, J. Am.
Chem. Soc. 2007, 129, 12084 – 12085; b) P. García-García, F. Lay,
P. García-García, C. Rabalakos, B. List, Angew. Chem. Int. Ed.
2009, 48, 4363 – 4366; Angew. Chem. 2009, 121, 4427 – 4430;
c) C. H. Cheon, H. Yamamoto, Org. Lett. 2010, 12, 2476 – 2479;
d) Q. Wang, M. Leutzsch, M. van Gemmeren, B. List, J. Am.
Chem. Soc. 2013, 135, 15334 – 15337.
[12] A2-Si also catalyzed the reaction, but with diminished enantio-
selectivity (86% ee for 3a). However, it is important to note that
on the basis of the 31P NMR analysis, A2-Si can be hydrolyzed
in situ to yield the free acid A2.
Received: February 22, 2016
Published online: April 27, 2016
6958 www.angewandte.org
ꢀ 2016 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Angew. Chem. Int. Ed. 2016, 55, 6954 –6958