Chiral Macrocyclic Organocatalysts for Kinetic Resolution of Disubstituted Epoxides with Carbon Dioxide

Article abstract of DOI:10.1021/acs.orglett.7b01838

Among chiral macrocycles 1 synthesized, 1m with the 3,5-bis(trifluoromethyl)phenylethynyl group was the best organocatalyst for the enantioselective synthesis of cyclic carbonates from disubstituted or monosubstituted epoxides and CO₂. The X-ray crystal structure of 1m revealed a well-defined chiral cavity with multiple hydrogen-bonding sites that is suitable for the enantioselective activation of epoxides. A catalytic cycle proposed was supported by DFT calculations.

Full text of DOI:10.1021/acs.orglett.7b01838

Letter
pubs.acs.org/OrgLett
Chiral Macrocyclic Organocatalysts for Kinetic Resolution of
Disubstituted Epoxides with Carbon Dioxide
Tadashi Ema,*,† Maki Yokoyama,† Sagiri Watanabe,† Sota Sasaki,† Hiromi Ota,‡ and Kazuto Takaishi†
‡
†Division of Applied Chemistry, Graduate School of Natural Science and Technology, and Advanced Science Research Center,
Okayama University, Tsushima, Okayama 700-8530, Japan
S
* Supporting Information
ABSTRACT: Among chiral macrocycles 1 synthesized, 1m with the 3,5-
bis(trifluoromethyl)phenylethynyl group was the best organocatalyst for
the enantioselective synthesis of cyclic carbonates from disubstituted or
monosubstituted epoxides and CO2. The X-ray crystal structure of 1m
revealed a well-defined chiral cavity with multiple hydrogen-bonding sites
that is suitable for the enantioselective activation of epoxides. A catalytic
cycle proposed was supported by DFT calculations.
arbon dioxide (CO2) is a sustainable feedstock that can be
used for various chemical transformations.1 Among them,
developed.7 Despite the structural diversity of macrocycles,8
macrocyclic organocatalysts are in the minority partly owing to
laborious synthesis.9 There is, however, much room for the
development of macrocyclic organocatalysts because catalytic
groups can be well arranged on a macrocyclic scaffold. We have
developed a chiral macrocycle called Chirabite-AR (1a) (Figure
1) and its congeners.10 They can be used for chiral
discrimination in NMR, enantiomer resolution in HPLC, gas-
phase recognition in ESI-FT-ICR MS, and fluorescence-
detected anion sensing. Because the lower amide NH groups
of 1a act as a hydrogen-bond donor to bind a variety of guests
C
the formation of cyclic carbonates from epoxides and CO2 is a
useful reaction with high atom efficiency, where a catalyst plays
a pivotal role.2,3 Quite a number of achiral catalysts have been
developed,2 including bifunctional metalloporphyrins showing
very high catalytic activity.3 In contrast, there are only a limited
number of examples of the kinetic resolution of epoxides with
CO2 (Scheme 1),4 most of which are catalyzed by metal salen
Scheme 1. Kinetic Resolution of Epoxides with CO2
complexes. The enantiomers of epoxides and cyclic carbonates
are useful chiral building blocks.5 Although optically active
natural products such as amino acids,6a,b cellulose,6c and β-
cyclodextrin6d can work as an organocatalyst for the production
of cyclic carbonates from epoxides and CO2, none of them is
reported to show enantioselectivity. Only recently, an
enantioselective organocatalytic reaction of epoxides with
CO2 has been reported (s value of up to 1.5).4j It should be
noted that only monosubstituted epoxides have been resolved by
chiral catalysts including metal complexes and organocatalysts,4
while the kinetic resolution of disubstituted epoxides with CO2
is not known at all. The enantioselective reaction of
disubstituted epoxides is a challenging subject because the
reactivity of epoxides remarkably decreases with an increase in
the number of the substituent.3d
Figure 1. Structures of macrocycles 1 and 2.
Received: June 16, 2017
Asymmetric organocatalysis is a rapidly growing research
area, where many excellent organocatalysts have been
© XXXX American Chemical Society
A
DOI: 10.1021/acs.orglett.7b01838
Org. Lett. XXXX, XXX, XXX−XXX

Organic Letters
Letter
including epoxides,10a,b we envisioned that they might work as a
Brønsted acid catalyst to accelerate an organic reaction. Here,
we synthesized and used a series of 1 as a chiral organocatalyst
for the kinetic resolution of epoxides with CO2 (Scheme 1). As
a result, they exhibited catalytic activity and enantioselectivity at
atmospheric CO2 pressure (balloon) under solvent-free
conditions. Even disubstituted epoxides could be transformed
enantioselectively. To the best of our knowledge, this is the first
example of the kinetic resolution of disubstituted epoxides with
CO2.
The 3,3′-positions of the binaphthyl moiety of 1 are the “hot
spots” to adjust the size and shape of the chiral cavity (Figure
1). The Suzuki−Miyaura and Sonogashira reactions successfully
introduced the aryl and arylethynyl substituents, respectively,
into the hot spots of diiodide 2 in good to high yields. A variety
of 1 could be prepared rapidly from a single precursor 2. The
kinetic resolution of trans-stilbene oxide (3a) with catalyst 1 (3
mol %) and co-catalyst (3 mol %) was performed at 75 °C for
72 h at atmospheric CO2 pressure (balloon) under solvent-free
conditions. After separation of 3a and 4a by silica gel column
chromatography, the enantiomeric purities (% ee) were
determined by chiral HPLC, and the degree of enantioselec-
tivity was evaluated by the s value.11 The results are
summarized in Table 1.
and a nucleophile, respectively. When either tetrabutylammo-
nium bromide or tetrabutylammonium chloride was used as a
co-catalyst, the reaction mixture was solidified, leading to a very
slow reaction (not shown). Macrocycles 1 were screened using
the best co-catalyst, TBAI. Catalyst 1b with the phenyl group
resulted in a much lower conversion (6%) with comparable
enantioselectivity (s = 1.5) (entry 2), while 1c with the
phenylethynyl group yielded a comparable conversion (17%)
with improved enantioselectivity (s = 2.7) (entry 3). In view of
these results, we modified the arylethynyl group. Table 1
indicates that the electronic effect is more important for
catalytic activity and enantioselectivity than the steric effect. For
example, 1d−g with the electron-donating group gave modest s
values of 2.4−3.4 (entries 4−7), whereas the electron-
withdrawing group had a better influence on catalytic activity
and enantioselectivity (s = 4.7−6.5) (entries 8−11). A clearer
difference can be seen for the catalysts with two substituents in
the arylethynyl group (entries 12−14). Compound 1m with the
3,5-bis(trifluoromethyl)phenylethynyl group achieved a high
conversion (41%) with the highest enantioselectivity (s = 10)
(entry 13), whereas 1l and 1n with two electron-donating
groups at the 3,5-positions of the phenylethynyl group
exhibited lower conversions (14−29%) with poor enantiose-
lectivity (s = 2.7−3.8) (entries 12, 14). Comparison of 1o and
1q reveals that the substituents at the 2,6-positions of the
phenylethynyl group hinder the reaction (entries 15, 17).
Indeed, 1p with the mesitylethynyl group showed poor
performance (entry 16).
When both catalyst 1a and co-catalyst, tetrabutylammonium
iodide (TBAI), were used together, the reaction proceeded
enantioselectively (s = 1.7) (entry 1). In contrast, no reaction
proceeded in the presence of either 1a alone or TBAI alone
(not shown). It is likely that the lower amide NH groups of 1a
and the I− anion of TBAI act cooperatively as a Brønsted acid
The kinetic resolution of disubstituted epoxides 3b−j was
conducted at 50 °C for 120 h with the best catalyst 1m (3 mol
%) and TBAI (3 mol %) (Table 2). The epoxides with the
electron-donating or electron-withdrawing group were resolved
with good s values of up to 13, and bulky epoxide 3j with the 1-
naphthyl group was also converted into the corresponding
cyclic carbonate 4j with a comparable s value. Table 2 indicates
a
Table 1. Screening of Catalysts
a
Table 2. Kinetic Resolution of Internal Epoxides 3
c
d
% yield (% ee)
b
e
entry
1
c
(%)
20
(R,R)-4a
(S,S)-3a
s value
1
1a
1b
1c
1d
1e
1f
18 (24)
10 (20)
14 (42)
19 (42)
11 (53)
14 (38)
21 (47)
32 (57)
30 (65)
35 (60)
33 (57)
14 (43)
41 (72)
17 (51)
35 (66)
12 (19)
21 (52)
61 (6.1)
77 (1.3)
69 (8.7)
75 (10)
66 (4.8)
65 (6.2)
59 (15)
57 (27)
54 (34)
39 (34)
51 (40)
77 (6.9)
51 (50)
64 (21)
55 (40)
70 (3.7)
64 (16)
1.7
1.5
2.7
2.7
3.4
2.4
3.2
4.7
6.5
5.5
5.3
2.7
10
2
6
17
19
8
3
4
5
6
14
24
32
34
36
41
14
41
29
38
16
24
c
d
% yield (% ee)
7
1g
1h
1i
b
e
entry
3
c
(%)
21
(R,R)-4
(S,S)-3
s value
8
9
1
2
3
4
5
3b
3c
3d
3e
3f
20 (83)
34 (77)
31 (75)
30 (74)
24 (82)
48 (71)
31 (74)
26 (78)
20 (77)
78 (22)
61 (45)
67 (35)
61 (39)
72 (28)
46 (73)
62 (38)
68 (29)
79 (21)
13
12
9.8
9.8
13
13
9.7
11
9.4
10
11
12
13
14
15
16
17
1j
37
32
35
25
51
34
27
21
1k
1l
1m
1n
1o
1p
1q
f
3.8
7.2
1.5
3.7
6
3g
3h
3i
7
8
9
3j
a
a
Conditions: epoxide 3a (1.0 mmol), cat. 1 (3 mol %), TBAI (3 mol
Conditions: epoxide 3 (1.0 mmol), cat. 1m (3 mol %), TBAI (3 mol
b
b
%), CO2 (1 atm, balloon), 75 °C, 72 h. Conversion calculated from c
= ee(3a)/(ee(3a) + ee(4a)). Isolated yield. Determined by HPLC
analysis. Calculated from s = ln[1 − c(1 + ee(4a))]/ln[1 − c(1 −
ee(4a))].
%), CO2 (1 atm, balloon), 50 °C, 120 h. Conversion calculated from
c d
c
d
c = ee(3)/(ee(3) + ee(4)). Isolated yield. Determined by HPLC
e
e
analysis. Calculated from s = ln[1 − c(1 + ee(4))]/ln[1 − c(1 −
ee(4))]. At 60 °C, 78 h.
f
B
DOI: 10.1021/acs.orglett.7b01838
Org. Lett. XXXX, XXX, XXX−XXX

Organic Letters
Letter
that 1m is tolerant of the electronegativity, bulkiness, and
position of the substituent in 3. The substrate scope of 1m for
monosubstituted epoxides 5 was also investigated under the
same conditions (Scheme 2). Because of 2 orders of magnitude
higher reactivity of 5 relative to 3, the reactions were stopped in
only 1 or 0.5 h in most cases. The s values for 5 were 2.5−4.3.
Scheme 3. Proposed Catalytic Cycle
Scheme 2. Kinetic Resolution of Terminal Epoxides 5
intermediate 1 (I1). The subsequent CO2 addition to the
alkoxide intermediate generates a linear carbonate anion (I2),
which is then cyclized by the intramolecular SN2 reaction to
give cyclic carbonate (P). This catalytic cycle was supported by
DFT calculations on a model substrate, ethylene oxide
(Supporting Information).12 The transition-state structure for
the initial ring opening, which is the rate-determining step, is
shown in Figure 3. The breaking C−O bond (1.98 Å) and the
The X-ray crystal structure indicates that 1m has a well-
defined chiral cavity (Figure 2). The cavity is not occluded by
Figure 3. Transition-state structure for the initial ring opening step in
the 1m/TBAI-catalyzed reaction of ethylene oxide with CO2. DFT
calculations were performed at the B3LYP/6-31G(d) level for the H,
C, N, O, and F atoms and at the B3LYP/LanL2DZ level for the I
atom. (a) Space-filling model, where the tetrabutylammonium cation is
shown in brown to clarify the supramolecular contacts. (b) Close-up
view with bond lengths and atomic charges. The I atom and ethylene
oxide are shown in ball-and-stick representation.
Figure 2. X-ray crystal structure of 1m where the 3,3′-substituents are
highlighted by a ball-and-stick representation. A MeOH molecule in
the binding cavity is shown.
forming C−I bond (2.66 Å) are almost the same lengths as
those (2.01 and 2.64 Å, respectively) calculated for a
bifunctional Mg porphyrin catalyst with a quaternary
ammonium iodide.3b The leaving O atom of the epoxide is
hydrogen bonded with the two amide H atoms of 1m, and both
of the O···H distances are short (1.778 and 1.778 Å),13 which
suggests strong hydrogen bonding. This is due to the negative
charge (natural atomic charge of −0.75) on the leaving O atom
that is induced by the nucleophilic attack of I−. The I− anion is
electrostatically stabilized by the H atom adjacent to the
trifluoromethyl group of 1m as well as the tetrabutylammonium
cation with the positive charges on the H atoms, but not on the
central N atom.3b The tetrabutylammonium cation also makes
contacts with the F atom of the trifluoromethyl group and the
O atoms of the nitro group of 1m via hydrogen bonds, forming
a supramolecular assembly. The relatively small energy barrier
(16.7 kcal/mol) is consistent with the high catalytic activity of
1m. All the subsequent anionic intermediates and transition
the 3,5-bis(trifluoromethyl)phenylethynyl group, and the lower
amide NH groups of 1m are hydrogen bonded with the O atom
of a MeOH molecule with distances of 2.208 and 2.253 Å. H
1
NMR titrations exhibited a large downfield shift (Δδ = 0.83
ppm) of the amide NH signal of 1m upon addition of styrene
oxide (R)-5h or (S)-5h, which strongly suggests the activation
of the epoxide by hydrogen bonding. The binding constants
(Ka) of 1m for (R)-5h and (S)-5h in CDCl3 at 21 °C were
determined to be 11.3 and 4.8 M−1, respectively. We suppose
that this chiral recognition leads to the differential ring-opening
transition state in the kinetic resolution. NMR titrations also
suggested the formation of a supramolecular assembly of 1m
and TBAI (Ka = 3.5 M−1 in CDCl3 at 21 °C).
Scheme 3 shows a plausible catalytic cycle. The epoxide is
hydrogen-bonded with the amide NH groups of the catalyst in
the reactant complex (R), and the activated epoxide is ring-
opened by the nucleophilic attack of the I− anion to give
C
DOI: 10.1021/acs.orglett.7b01838
Org. Lett. XXXX, XXX, XXX−XXX

Organic Letters
Letter
B. Chem. Sci. 2012, 3, 2094−2102. (h) Roy, T.; Kureshy, R. I.; Khan,
N. H.; Abdi, S. H. R.; Bajaj, H. C. Catal. Sci. Technol. 2013, 3, 2661−
2667. (i) North, M.; Quek, S. C. Z.; Pridmore, N. E.; Whitwood, A. C.;
Wu, X. ACS Catal. 2015, 5, 3398−3402. (j) Liu, S.; Suematsu, N.;
Maruoka, K.; Shirakawa, S. Green Chem. 2016, 18, 4611−4615.
(5) For other types of kinetic resolution of epoxides, see:
(a) Jacobsen, E. N. Acc. Chem. Res. 2000, 33, 421−431. (b) Pellissier,
H. Adv. Synth. Catal. 2011, 353, 1613−1666.
states are stabilized by double hydrogen bonding in the cavity
(Supporting Information).
In summary, the enantioselective reaction of epoxides with
CO2 was achieved by the synergistic action of chiral
macrocyclic organocatalyst 1m and TBAI. 1m has a chiral
cavity with multiple hydrogen-bonding sites that is suitable for
the enantioselective activation of epoxides. To the best of our
knowledge, this is the first example of the enantioselective
synthesis of cyclic carbonates from disubstituted epoxides and
CO2. Further work is underway to explore the scope and
limitations of the chiral macrocyclic organocatalysts.
(6) (a) Roshan, K. R.; Kathalikkattil, A. C.; Tharun, J.; Kim, D. W.;
Won, Y. S.; Park, D. W. Dalton Trans. 2014, 43, 2023−2031. (b) Yang,
Z.; Sun, J.; Cheng, W.; Wang, J.; Li, Q.; Zhang, S. Catal. Commun.
2014, 44, 6−9. (c) Sun, J.; Cheng, W.; Yang, Z.; Wang, J.; Xu, T.; Xin,
J.; Zhang, S. Green Chem. 2014, 16, 3071−3078. (d) Song, J.; Zhang,
Z.; Han, B.; Hu, S.; Li, W.; Xie, Y. Green Chem. 2008, 10, 1337−1341.
ASSOCIATED CONTENT
* Supporting Information
■
(7) (a) Berkessel, A.; Groger, H. Asymmetric Organocatalysis; Wiley-
̈
S
VCH: Weinheim, 2005. (b) Science of Synthesis: Asymmetric
Organocatalysis; List, B., Maruoka, K., Eds.; Thieme: Stuttgart, 2012.
(8) Choi, K.; Hamilton, A. D. Coord. Chem. Rev. 2003, 240, 101−110.
(9) (a) Diederich, F. Cyclophanes; RSC: Manchester, 1991. (b) Zhao,
H.; Shen, J.; Guo, J.; Ye, R.; Zeng, H. Chem. Commun. 2013, 49,
2323−2325. (c) Blanco, V.; Leigh, D. A.; Marcos, V.; Morales-Serna, J.
A.; Nussbaumer, A. L. J. Am. Chem. Soc. 2014, 136, 4905−4908.
(d) Park, Y.; Harper, K. C.; Kuhl, N.; Kwan, E. E.; Liu, R. Y.; Jacobsen,
E. N. Science 2017, 355, 162−166.
The Supporting Information is available free of charge on the
ACS Publications website at DOI: 10.1021/acs.or-
glett.7b01838.
Synthesis, NMR spectra, determination of binding
constants, and computational details (PDF)
Crystallographic data for 1m (CIF)
(10) (a) Ema, T.; Tanida, D.; Sakai, T. Org. Lett. 2006, 8, 3773−
3775. (b) Ema, T.; Tanida, D.; Sakai, T. J. Am. Chem. Soc. 2007, 129,
10591−10596. (c) Ema, T.; Tanida, D.; Sugita, K.; Sakai, T.;
Miyazawa, K.; Ohnishi, A. Org. Lett. 2008, 10, 2365−2368.
(d) Fraschetti, C.; Filippi, A.; Crestoni, M. E.; Ema, T.; Speranza,
M. J. Am. Soc. Mass Spectrom. 2013, 24, 573−578. (e) Ema, T.; Okuda,
K.; Watanabe, S.; Yamasaki, T.; Minami, T.; Esipenko, N. A.;
Anzenbacher, P., Jr. Org. Lett. 2014, 16, 1302−1305. (f) Akdeniz, A.;
Minami, T.; Watanabe, S.; Yokoyama, M.; Ema, T.; Anzenbacher, P.,
Jr. Chem. Sci. 2016, 7, 2016−2022.
(11) Martin, V. S.; Woodard, S. S.; Katsuki, T.; Yamada, Y.; Ikeda,
M.; Sharpless, K. B. J. Am. Chem. Soc. 1981, 103, 6237−6240.
(12) Frisch, M. J.; et al. Gaussian 09, revision E.01; Gaussian, Inc.:
Wallingford, CT, 2013. For the full citation, see the Supporting
Information.
AUTHOR INFORMATION
■
Corresponding Author
*E-mail: ema@cc.okayama-u.ac.jp.
ORCID
Tadashi Ema: 0000-0002-2160-6840
Kazuto Takaishi: 0000-0003-4979-7375
Notes
The authors declare no competing financial interest.
ACKNOWLEDGMENTS
■
This work was partly supported by JSPS KAKENHI Grant No.
JP16H01030 in Precisely Designed Catalysts with Customized
Scaffolding.
(13) The corresponding O···H distances in reactant complex R
(Scheme 3) are 1.959 and 2.005 Å.
REFERENCES
■
(1) Recent reviews: (a) Tsuji, Y.; Fujihara, T. Chem. Commun. 2012,
48, 9956−9964. (b) Kielland, N.; Whiteoak, C. J.; Kleij, A. W. Adv.
Synth. Catal. 2013, 355, 2115−2138. (c) Maeda, C.; Miyazaki, Y.; Ema,
T. Catal. Sci. Technol. 2014, 4, 1482−1497. (d) Liu, Q.; Wu, L.;
Jackstell, R.; Beller, M. Nat. Commun. 2015, 6, 5933.
(2) (a) Cokoja, M.; Wilhelm, M. E.; Anthofer, M. H.; Herrmann, W.
A.; Kuhn, F. E. ChemSusChem 2015, 8, 2436−2454. (b) Martín, C.;
̈
Fiorani, G.; Kleij, A. W. ACS Catal. 2015, 5, 1353−1370. (c) Lang, X.-
D.; He, L.-N. Chem. Rec. 2016, 16, 1337−1352. (d) Hasegawa, J.;
Miyazaki, R.; Maeda, C.; Ema, T. Chem. Rec. 2016, 16, 2260−2267.
(3) (a) Ema, T.; Miyazaki, Y.; Koyama, S.; Yano, Y.; Sakai, T. Chem.
Commun. 2012, 48, 4489−4491. (b) Ema, T.; Miyazaki, Y.;
Shimonishi, J.; Maeda, C.; Hasegawa, J. J. Am. Chem. Soc. 2014, 136,
15270−15279. (c) Maeda, C.; Taniguchi, T.; Ogawa, K.; Ema, T.
Angew. Chem., Int. Ed. 2015, 54, 134−138. (d) Maeda, C.; Shimonishi,
J.; Miyazaki, R.; Hasegawa, J.; Ema, T. Chem. - Eur. J. 2016, 22, 6556−
6563. (e) Maeda, C.; Sasaki, S.; Ema, T. ChemCatChem 2017, 9, 946−
949.
(4) (a) Lu, X.-B.; Liang, B.; Zhang, Y.-J.; Tian, Y.-Z.; Wang, Y.-M.;
Bai, C.-X.; Wang, H.; Zhang, R. J. Am. Chem. Soc. 2004, 126, 3732−
3733. (b) Paddock, R. L.; Nguyen, S. T. Chem. Commun. 2004, 1622−
1623. (c) Berkessel, A.; Brandenburg, M. Org. Lett. 2006, 8, 4401−
4404. (d) Yamada, W.; Kitaichi, Y.; Tanaka, H.; Kojima, T.; Sato, M.;
Ikeno, T.; Yamada, T. Bull. Chem. Soc. Jpn. 2007, 80, 1391−1401.
(e) Yan, P.; Jing, H. Adv. Synth. Catal. 2009, 351, 1325−1332.
(f) Chang, T.; Jin, L.; Jing, H. ChemCatChem 2009, 1, 379−383.
(g) Ren, W.-M.; Wu, G.-P.; Lin, F.; Jiang, J.-Y.; Liu, C.; Luo, Y.; Lu, X.-
D
DOI: 10.1021/acs.orglett.7b01838
Org. Lett. XXXX, XXX, XXX−XXX