This is the peer reviewed version of the following article: Green Chem., 2018,20, 4537-4546, which has been published in final form
https://doi.org/10.1039/C8GC02020E, see http://www.rsc.org/Publishing/Journals/OpenScience/index.asp

Evaluating polymer-supported isothiourea catalysis in industrially-preferable
solvents for the acylative kinetic resolution of secondary and tertiary heterocyclic
alcohols in batch and flow†
a

a

a

a

a

Nitul Ranjan Guha,‡ Rifahath M. Neyyappadath,‡ Mark D. Greenhalgh, Ross Chisholm, Samuel M. Smith, Megan L.
a
a
b
bc
a
a
McEvoy, Claire M. Young, Carles Rodríguez-Escrich, Miquel A. Pericàs, Georg Hähner and Andrew D. Smith*
Polymer-supported Lewis base catalysts, based on the homogeneous isothioureas HyperBTM and BTM, have been synthesised and applied for the acylative
kinetic resolution of secondary and tertiary heterocyclic alcohols. In batch, the use of industrially-preferable solvents was investigated, with dimethyl
carbonate proving to be most generally-applicable. Significantly, the HyperBTM-derived immobilised catalysts were readily recycled, with no loss in either
activity or selectivity. In addition to the kinetic resolution of secondary benzylic, propargylic, allylic and cycloalkanol derivatives, a range of 22 tertiary
heterocyclic alcohols, based on privileged 3-hydroxyoxindole and 3-hydroxypyrrolidinone substructures, were resolved with up to excellent selectivity (s = 7–
190). Finally, the immobilised isothiourea catalysts were applied in a packed bed reactor to demonstrate the first example of the kinetic resolution of tertiary
heterocyclic alcohols in a continuous flow process. High selectivities were obtained for the resolution of 3-hydroxyoxindole derivatives in ethyl acetate (s up
to 70); and for 3-hydroxypyrrolidinones derivatives in toluene (s up to 42).

Introduction
Enantioselective organocatalysis has become a diverse and
vibrant research area over recent decades, with many highly
1
efficient and selective methods having been developed. In the
context of sustainability and potential industrial applicability,
one common criticism of the area is the use of relatively high
loadings of the organocatalyst, which is typically neither
reisolated nor recycled following a given reaction. One
potential general solution is the immobilisation of the
2
organocatalyst onto a heterogeneous polymer support. In
principle, recovery of the catalyst can be achieved following
the reaction by a simple filtration; alternatively the polymersupported catalyst can be incorporated into a continuous flow
3
set-up, improving the ease of catalyst recycling even further.
To compensate the additional time and energy costs
associated with catalyst synthesis, the immobilised catalyst
must be obtained via a straightforward and high yielding
route, and be sufficiently stable to allow recycling multiple
times without any loss in activity or selectivity.
In this area, we recently reported the synthesis of a Merrifield
resin-derived polymer-supported chiral isothiourea based
upon a homogeneous Lewis basic catalyst, HyperBTM, and
applied this immobilised catalyst for the kinetic resolution (KR)
4–8
of secondary alcohols (Scheme 1).
A range of alcohols,
including benzylic, allylic and propargylic alcohols, as well as
cycloalkanol derivatives and a 1,2-diol, were resolved with up
to excellent selectivity factors [where selectivity factor (s) is
defined as the rate constant for the fast reacting enantiomer
divided by the rate constant for the slow reacting enantiomer
9
(s = kfast/kslow)]. Significantly, this immobilised catalyst could
be recycled at least 15 times with no appreciable loss in either
activity or selectivity. The broad scope and excellent catalyst
recyclability in comparison to previously reported KRs using

immobilised catalysts indicated substantial potential of this
4,5
catalyst for future applications.

Scheme 1 Previous work on the KR of secondary alcohols using a Merrifield resinsupported isothiourea catalyst.

Although demonstrating proof-of-concept in this area, one
potential obstacle for industrial application of this method was
the reliance upon using chloroform as the reaction solvent. As
the solvent typically constitutes the majority of the mass in
any synthetic transformation, the nature of the solvent has a
substantial impact on the sustainability of the process. These
considerations are of particular importance in the chemical
industry. Over the last 15 years, considerable effort has been
made to identify the suitability of different solvents, taking
into account environmental impact, health and safety, cost
and life-cycle management considerations. This has resulted in
a range of reports from pharmaceutical companies that have
broadly categorised solvents into a number of groupings, that
for simplicity and the purpose of this manuscript we have
10
classified as “best to avoid”, “usable” and “preferred”.
In order to enhance the potential industrial-applicability of our
polymer-supported isothiourea catalyst, we recognised that
solvent compatibility with industrially-preferable solvents was
of paramount importance. A variety of differentially-

functionalised polymer supports are commercially-available
due to their application in solid phase peptide synthesis
11
(Scheme 2a), where modulation of the polymer functionality
is known to have a substantial effect on the stability and
swelling properties of the polymer in different solvents. The
synthesis of isothiourea catalysts immobilised on different
polymer supports was therefore envisioned to allow
assessment of solvent compatibility and sustainability. In
addition, the change in chemical environment around the
Lewis basic isothiourea in each immobilised catalyst could be
expected to lead to changes in both activity and selectivity.
Herein, we report the synthesis of a range of polymersupported isothiourea catalysts and demonstrate their
application and recyclability for the KR of secondary and
tertiary alcohols in both batch and continuous flow, with a
particular focus on the compatibility of industrially-preferable
solvents (Scheme 2b).

Scheme 3 Preparation of solid-supported HyperBTM and BTM catalysts.

KR of secondary alcohols: solvent compatibility

Scheme 2 a) Commercially-available polymer supports used in this study; b) KR of
secondary and tertiary alcohols using a range of supported isothiourea catalysts in
batch and flow using industrially-preferable solvents.

Results and discussion
Synthesis of polymer-supported isothiourea catalysts
The synthesis of immobilised HyperBTM derivatives 2–4 was
achieved through a Cu-catalysed azide-alkyne cycloaddition
between alkyne-substituted HyperBTM derivative 1 and azidefunctionalised Merrifield, TentaGel and Wang resins (Scheme
12
3). Elemental analysis of each immobilised catalyst allowed
calculation of polymer functionalisation based on nitrogen
13,14
content,
with these values used to determine catalyst
15
loading in all subsequent KRs. A Merrifield resin-supported
derivative of the isothiourea BTM 6 was also synthesised by a
similar procedure starting from alkyne-substituted BTM
derivative 5. In all cases the isothiourea-functionalised
polystyrene resins were obtained in high yield.

Investigations began by comparing the activity and selectivity of
HyperBTM-derived immobilised catalysts 2–4 for the KR of (±)-1(naphthalen-2-yl)ethan-1-ol 7 using isobutyric anhydride in a
14,16
range of solvents (Table 1).
All KRs were conducted at room
temperature to aid practically and to enable future application. The
KRs were first performed in chloroform to provide a direct
4
comparison with the previously-reported methodology. All
immobilised isothioureas catalysed the KR of alcohol 7 with
excellent selectivity (s = 45–80, entries 1–3). The use of
10,17
industrially-preferable solvents was next investigated.
Excellent selectivities were obtained with all catalysts in
toluene (s = 60–70), with Merrifield and TentaGel-derived
catalysts 2 and 3 providing optimal conversion (entries 4–5),
whilst lower conversion was obtained using Wang derivative 4
(entry 6). A more significant solvent effect was observed when
using tert-amyl alcohol, with high conversion and selectivity
only obtained with TentaGel-derived catalyst 3 (entries 7–9).
Table 1 KR of (±)-7 using supported catalysts 2–4 and 6: Effect of solvent

Entry
1c
2
3
4c
5
6
7
8
9
10
11
12
13
14
15
16

Catalyst
2
3
4
2
3
4
2
3
4
2
3
4
2
3
4
6

Solvent
CHCl3
CHCl3
CHCl3
PhMe
PhMe
PhMe
t-AmOH
t-AmOH
t-AmOH
i-PrOAc
i-PrOAc
i-PrOAc
(MeO)2CO
(MeO)2CO
(MeO)2CO
(MeO)2CO

c (%)
50
40
32
52
52
32
17
49
3
52
47
44
50
46
45
27

7 era
95:5
81:19
72:28
98:2
97:3
72:28
59:41
89:11
51:49
97:3
89:11
81:19
94:6
89:11
88:12
33:67

8 erb
96:4
96:4
97:3
95:5
94:6
98:2
93:7
94:6
87:13
93:7
94:6
90:10
94:6
96:4
96:4
4:96

s
80
45
50
70
60
70
16
43
7
49
39
16
45
60
49
38

A
V
O
I
D

U
S
A
B
L
E

P
R
E
F
E
R
R
E
D

First, the KR of 1,2-azidoalcohol 9 was performed with all four
catalysts (Table 2). Each of the HyperBTM derivatives 2–4 gave
similar conversions and selectivities (c = 46–52%; s = 22–29);
however BTM derivative 6 provided only low conversion and
selectivity (c = 16%; s = 3). Each catalyst was then recovered
and applied for the KR of propargylic alcohol 10. All of the
HyperBTM-derived catalysts 2–4 provided optimal conversion,
however more variation in selectivity was observed, with
Wang-derived catalyst 4 providing the highest selectivity (s =
15). In contrast, BTM derivative 6 was completely inactive.
Next, the KR of trans-phenylcyclopentanol 11 with each
HyperBTM-derived catalyst 2–4 provided similar conversions
and selectivities (c = 48–53%; s = 27–29), whilst BTM-derived
Table 2 Recycling supported catalysts 2–4 and 6 for the sequential KR of five secondary
alcohols (±)-9–13

‘Avoid’, ‘Usable’ and ‘Preferred’ indicate the industrial applicability of the
solvent.10 Conditions: (±)-7 (0.2 mmol), 2 or 6 (5 mol%), or 3 or 4 (3 mol%), (iPrCO)2O (0.55 equiv.), i-Pr2NEt (0.6 equiv.), solvent (0.2 M), r.t., 16 h. Conversion
(c) and er determined by chiral HPLC analysis. Selectivity factors (s) calculated
using er of 7 and reaction conversion (see ref. 9a), and rounded according to
estimated associated errors (see ref. 9b). a R:S. b S:R. c Data taken from ref. 4.

This solvent effect is consistent with the expected swelling of
each polymer support in tert-amyl alcohol, with the
polyethylene glycol linker present in TentaGel providing
enhanced swelling in polar protic solvents. The use of acetate
solvents was also tolerated, with good conversions obtained in
isopropyl acetate using each catalyst (entries 10–12). The
selectivity obtained using each catalyst was more variable
however; with Merrifield-derived catalyst 2 providing an s
value of 49 (entry 10), but Wang-derived catalyst 4 providing
an s value of only 16 (entry 12). More consistent data were
obtained when using dimethyl carbonate, with 45–50%
conversion and s = 45–60 obtained for all three catalysts
(entries 13–15). The use of immobilised BTM catalyst 6 also
proved reasonably effective, with good selectivity (s = 38) and
moderate conversion obtained (27%). Although less effective
than immobilised HyperBTM derivatives 2–4, the activity of
this BTM derivative is promising when compared to a
previously-reported polymer-supported variant of BTM, which
was shown to be ineffective for KRs (s = 3 for the KR of 18a
phenylethanol in CDCl3). Based on these results, the use of
dimethyl carbonate was further investigated for the KR of a
range of secondary alcohols using each immobilised catalyst.
Catalyst recycling
The recyclability and robustness of each polymer-supported
isothiourea catalyst was investigated through the sequential
KR of five different secondary alcohols in dimethyl carbonate
(Table 2). In each case, the catalyst was recovered by filtration
and washed sequentially with CH2Cl2/MeOH (1:1), MeOH, THF
and CH2Cl2 before drying under high vacuum for 2 h. The
secondary alcohols were selected from the scope previously
4
reported in chloroform, with these five alcohols chosen to
probe catalyst versatility for the KR of benzylic, allylic,
16
propargylic and cycloalkanols. Differences in catalyst activity
and selectivity for the KR of each substrate, in addition to
changes in activity following catalyst recycling, were assessed.

Catalyst

2

3

4

6

Cycle
1
2
3
4
5
1
2
3
4
5
1
2
3
4
5
1
2
3

Substrate
9
10
11
12
13
9
10
11
12
13
9
10
11
12
13
9
10
11

c (%)
52
55
53
48
52
46
52
48
47
43
50
51
49
47
44
16
0
0

Alc. era
6:94
89:11
95:5c
85:15
> 99:1
13:87
81:19
90:10c
85:15
87:13
10:90
88:12
91:9c
85:15
88:12
54:46
-

Est. erb
10:90
81:19
90:10d
87:13
96:4
9:91
79:21
92:8d
90:10
99:1
8:92
87:13
92:8d
89:11
> 99:1
70:30
-

s
25
10
27
14
140
22
7
29
18
170
29
15
29
16
190
3
-

Conditions: 9–13 (0.2–0.4mmol), 2 or 6 (5 mol%), or 3 or 4 (3 mol%), (i-PrCO)2O
(0.55 equiv.), i-Pr2NEt (0.6 equiv.), (MeO)2CO (0.2 M), r.t., 24 h. Conversion (c) and
er determined by chiral HPLC analysis. Selectivity factors (s) calculated using
alcohol er and reaction conversion (see ref. 9a), and rounded according to
estimated associated errors (see ref. 9b). a R:S. b S:R. c 1R,2S:1S,2R. d 1S,2R:1R,2S.

catalyst 6 again proved to be completely inactive. Based on
these results, the use of BTM-derived catalyst 6 in these
recycling experiments was discontinued. The KR of the next
7l
substrate, aryl-alkenyl-substituted carbinol 12, was achieved
using all HyperBTM derivatives 2–4, with similar conversions
and selectivities obtained once more (c = 47–48%; s = 14–18).
Finally, the KR of benzylic alcohol 13 using each HyperBTM-

derived catalyst 2–4 was achieved with good conversions and
exceptional selectivities (s = 140–190). It should be noted that
the s values obtained in these experiments using dimethyl
carbonate are lower than those previously-reported for the
4
same substrates in chloroform.
Nonetheless, these
experiments demonstrate the feasibility of using an
industrially-prefererable solvent to obtain synthetically-useful
levels of selectivity. It is noteworthy that all of the immobilised
HyperBTM derivatives 2–4 are readily recycled and are
considerably more robust than BTM analogue 6.
Decomposition of the homogeneous isothiourea catalyst BTM
has been reported in the presence of anhydrides and
7i
alcohols, with the operation of a similar deactivation pathway
for the immobilised analogue consistent with the results of
these recycling experiments.

form, further optimisation using Merrifield resin-supported
HyperBTM 2 in dimethyl carbonate was targeted. Although
similar conversions and selectivites were obtained using both
ethyl acetate and dimethyl carbonate, the use of dimethyl
carbonate was considered preferrable due to its reduced
flammability, increased stability and lower environmental
10b,d,e
impact.
The reaction conversion in dimethyl carbonate
was improved to 53% by increasing the equivalents of
anhydride, to provide the optimised conditions for this KR
using Merrifield resin-supported HyperBTM 2 (entry 9).
Consistent with our work using the homogeneous analogues of
these isothiourea catalysts, the use of Merrifield resinsupported BTM derivative 6 led to very low conversion and
7m
selectivity (entry 10).
Table 3 Reaction optimisation for the KR of 3-allyl-3-hydroxyoxindole derivative (±)-14

KR of tertiary alcohols: reaction optimisation and scope
To expand the scope and impact of this method, the polymersupported catalysts 2–4 and 6 were investigated for the KR of
tertiary heterocyclic alcohols in industrially-preferable solvents
in both batch and continuous flow. The acylative KR of tertiary
alcohols is a significant challenge due to the sterically-hindered
nature of the substrates, and the requirement for the catalyst
to differentiate between three substituents at the stereogenic
carbinol centre. As such, only three methods have been
7m,18
previously reported,
including a recent example from
7m
ourselves using the homogeneous version of HyperBTM. The
KR of a range of tertiary heterocyclic alcohols was reported,
however chloroform was used as the optimal solvent and the
ability to recover or recycle the catalyst was not
demonstrated, thus limiting industrial applicability.
Optimisation studies began with the KR of 3-allyl-3hydroxyoxindole derivative 14 using Merrifield resin-supported
16,19
HyperBTM 2 and isobutyric anhydride in batch (Table 3).
Initially, the use of chloroform, which proved most effective in
7m
the homogeneous system, was investigated. After 24 h at
room temperature only 22% conversion was observed,
however encouragingly high selectivity was obtained (s = 60)
(Table 3, entry 1). To provide context, the KR of alcohol 14 in
chloroform at room temperature using homogeneous
HyperBTM (1 mol%) provided a conversion of 53% and an s
7m
value of 90. This indicates the polymer-supported variant of
the catalyst displays reduced activity, but still provides
similarly high enantiodiscrimination. The use of more
10,17
industrially-preferable solvents was investigated next.
Using either toluene or ethyl acetate resulted in significantly
improved conversion, albeit with lower selectivity (entries 2–
3). The KR of 14 in ethyl acetate was also investigated using
TentaGel and Wang resin-supported HyperBTM 3 and 4, with
similar conversions and selectivities obtained in each case
(entries 4–5). Finally, the KR of 14 was examined in dimethyl
carbonate with all three immobilised HyperBTM catalysts 2–4
(entries 6–8). TentaGel-supported catalyst 3 gave the highest
selectivity (s = 29; entry 7), however significantly better
conversion was obtained using Merrifield resin-supported
HyperBTM 2 (entry 6). As conversions of > 50% are desirable in
KRs to allow isolation of substrate in highly enantioenriched

Entry
1
2
3
4
5
6
7
8
9c
10

Catalyst
2
2
2
3
4
2
3
4
2
6

Solvent
CHCl3
PhMe
EtOAc
EtOAc
EtOAc
(MeO)2CO
(MeO)2CO
(MeO)2CO
(MeO)2CO
(MeO)2CO

c (%)
22
46
47
42
47
48
38
36
53
1

14 era
64:36
85:15
87:13
88:12
86:14
88:12
77:23
74:26
93:7
50:50

15 erb
98:2
91:9
91:9
78:22
90:10
90:10
94:6
92:8
88:12
66:34

s
60
21
22
13
20
21
29
19
20
2

U
S
A
B
L
E
P
R
E
F
E
R
R
E
D

‘Usable’ and ‘Preferred’ indicate the industrial applicability of the solvent.10
Conditions: 14 (0.1 mmol), 2, 3, 4 or 6 (5 mol%), (i-PrCO)2O (0.55 equiv.), i-Pr2NEt
(0.6 equiv.), solvent (0.2 M), r.t., 24 h. Conversion (c) and er determined by chiral
HPLC analysis. Selectivity factors (s) calculated using alcohol er and reaction
conversion (see ref. 9a), and rounded according to estimated associated errors
(see ref. 9b). a S:R. b R:S. c 0.7 equiv. of (i-PrCO)2O used.

Encouraged by the promising results using Merrifield resinsupported HyperBTM 2 in dimethyl carbonate at room
16
temperature, the scope of the KR was investigated (Table 4).
The introduction of a sterically-hindered i-Pr group or an
electron- withdrawing CF3 group at C(3) were both tolerated,
with 17 and 18 resolved with good selectivity (s = 13–21). The
incorporation of an aromatic group at C(3) was also successful,
with phenyl-substituted alcohols 19–21 resolved with optimal
conversion and excellent selectivity (s = 60–70) allowing
recovery of the alcohol in highly enantioenriched form in each
case. Aromatic substituents at C(3) bearing both electrondonating and -withdrawing groups were tolerated, as was the
introduction of heteroaromatic substituents, with alcohols 22–
25 all resolved with good selectivity (s = 13–36). The method
was equally applicable for the KR of 3-alkenyl-substituted
alcohol 26 (s = 24). It is noteworthy that these examples
demonstrate the immobilised isothiourea catalyst can
2
differentiate between three sp -hybridised substituents at the
20
tertiary carbinol centre (two -systems and a carbonyl).
Finally, substitution of the benzenoid ring was generally

Table 4 Substrate Scope for the KR of 3-substituted-3-hydroxyoxindole derivatives

c (%)a

Alcohol
er;b yield

Ester
er;c yield

sa

1

56
[56]

> 99:1; 43%

90:10; 51%

60
[60]

2

tolerated, with 4-chloro- and 5-methyl-substituted oxindole
derivatives 27 and 28 resolved with good selectivity (s = 16–
27); however 6- and 7-chloro-substituted derivatives 29 and 30

54
[54]

> 99:1; 46%

92:8; 49%

80
[70]

54
[56]

> 99:1; 43%

92:8; 51%

70
[60]

52
[53]

99:1; 47%

95:5; 50%

90
[70]

53
[56]

99:1; 45%

94:6; 48%

70
[60]

60
[59]

>99:1;c 39%

84:16;b 57%

44
[26]

54
[56]

> 99:1; 45%

93:7; 50%

70
[80]

55
[53]

99:1; 44%

89:11; 52%

37
[27]

> 99:1; 45%

92:8; 53%

89:11; 48%

92:8; 48%

Cycle

3

Substrate

19

4

5

19

6

7

19

8

9

10

Conditions: alcohol (0.1 mmol), 2 (5 mol%), (i-PrCO)2O (0.7 equiv.), i-Pr2NEt (0.6
equiv.), (MeO)2CO (0.2 M), 24–48 h. Conversion (c) and er determined by chiral
HPLC analysis. Selectivity factors (s) calculated using alcohol er and reaction
conversion (see ref. 9a), and rounded according to estimated associated errors
(see ref. 9b). Alcohol er given as S:R, ester er given R:S. a R:S. b S:R.
Table 5 Recycling of polymer-supported HyperBTM 2

19

54
[56]

49
[44]

80
[60]

26
[24]

Conditions: alcohol (0.25–0.4 mmol), 2 (5 mol%), (i-PrCO)2O (0.7 equiv.), i-Pr2NEt
(0.6 equiv.), (MeO)2CO (0.2 M), 18–48 h. Conversion (c) and er determined by
chiral HPLC analysis. Selectivity factors (s) calculated using alcohol er and reaction
conversion (see ref. 9a), and rounded according to estimated associated errors
(see ref. 9b). a Conversion (c) and s data for resolutions using fresh catalyst shown
in italics in brackets. b S:R. c R:S.

provided relatively low selectivity (s = 7–9). Overall, substrates
bearing a variety of N-substituents were successfully applied,
including N-tert-butoxycarbonyl (Boc), N-benzyl, N-allyl and Nmethyl-substituted 3-hydroxyoxindole derivatives. Although
slightly lower than those previously reported using
7m
homogeneous HyperBTM in chloroform, synthetically-useful
levels of selectivity were obtained in the majority of cases.

Catalyst recycling
The robustness and reusability of Merrifield resin-supported
HyperBTM catalyst 2 was further probed through a recycling
experiment in which the catalyst was recovered and reused
ten times (Table 5). In this experiment, six different tertiary
alcohol substrates were applied, with the KR of N-benzyl-3phenyl-substituted hydroxyoxindole 19 performed on
alternate cycles. This approach was adopted to: (i) investigate
if the same sample of catalyst could be used to resolve
different substrates without cross-contamination, and (ii)
provide insight into whether there is any loss in activity or
selectivity of the catalyst upon recycling. All six substrates
were resolved with very comparable conversion and selectivity
to when fresh catalyst was used (this data is provided italicized
and in square brackets in table 5 for clarity). The KR of 19 was
also remarkably consistent, with conversions of 53–56% and s
values of 60–80 obtained for the five cycles. These
experiments indicate excellent stability of Merrifield resinsupported HyperBTM catalyst 2 over the course of this
recycling protocol. Significantly, no cross-contamination
between KRs was observed in any case, demonstrating
excellent potential for further applications.

Table 6 Continuous flow KR of 3-substituted-3-hydroxyoxindole derivatives in
chloroform using the same packed bed reactor

Application in continuous flow
Based on the excellent recyclability of Merrifield resinsupported HyperBTM catalyst 2 for the KR of tertiary alcohols,
application in a continuous flow set-up was investigated. To
the best of our knowledge the KR of tertiary alcohols in
continuous flow has not been reported previously. We
therefore initially investigated the feasibility of the process
under the conditions used previously for the KR of secondary
4
alcohols in continuous flow using chloroform as solvent. The
immobilised catalyst 2 (600 mg, 0.53 mmol) was swollen in a
medium pressure borosilicate glass column in chloroform to
create a packed bed reactor, and solutions of racemic tertiary
alcohol (0.1 M) and a mixture of anhydride (0.08 M) and base
(0.08 M) were passed through using a syringe pump. Based on
14
optimisation studies using hydroxyoxindole derviative 14, a
reaction temperature of 0 °C and an elution rate of 0.1 mL
−1
min , to provide a residence time of 30 min, was chosen and
applied in each case. The packed bed reactor was used for the
sequential KR of six different 3-substituted-3-hydroxyoxindole
derivatives 14, 19, 25, 30, 20 and 28 using isobutyric anhydride
(Table 6, entries 1–6). Good conversions and exceptional
selectivities were obtained in each case (s = 50–190). Each KR
was carried out on a 0.5 mmol scale, with the packed bed
reactor flushed with a solution of 10% methanol in chloroform,
followed by pure chloroform between KRs to regenerate the
catalyst and avoid any cross-contamination.

Conversion and er determined by chiral HPLC analysis. Selectivity factor (s)
calculated using alcohol er and reaction conversion (see ref. 9a), and rounded
according to estimated associated errors (see ref. 9b). Alcohol er given as S:R,
ester er given R:S. a R:S. b S:R.

Next, the use of industrially-preferable solvents for the KR of
3-substituted-3-hydroxyoxindole derivatives was investigated.
Optimisation studies using Merrifield-supported HyperBTM 2
and hydroxyoxindole derviative 14 found that low selectivities
were obtained using dimethyl carbonate (s = 9), and that
14
toluene was unsuitable due to the low solubility of 14. The
use of ethyl acetate was relatively successful however, with a
moderate s and good conversion obtained at 0 °C, albeit when
using an increased concentration of anhydride (0.2 M). The
swelling of Merrifield-supported HyperBTM 2 was inefficient in
ethyl acetate and therefore the use of TentaGel-supported
HyperBTM 3 was also investigated, however only low
14
conversion and selectivity were obtained (c = 10%; s = 9). The
scope of the KR of 3-substituted-3-hydroxyoxindole derivatives
using Merrifield-supported HyperBTM 2 in ethyl acetate was
therefore investigated in more detail (Table 7). The same
packed bed reactor that had been used in the previous studies

was used for the sequential KR of six different 3-substituted-3hydroxyoxindole derivatives 14, 19, 25, 28, 22 and 30. The
diminished swelling of the polymer support in ethyl acetate
resulted in a shorter residence time of 15 min. The resolution
of 3-allyl-substituted derivative 14 proceeded with high
conversion and good selectivity (s = 14; entry 1) to provide the
recovered alcohol with high enantioenrichment (98:2 er). The
resolution of 3-aryl-substituted 3-hydroxyindolin-2-one
derivatives 19, 25, 28 and 22 proceeded with optimal
conversions and high selectivities in each case (s = 29–70;
entries 2–5). In contrast, the KR of 3-methyl-substituted
analogue 30 only provided an s value of 8, albeit still allowing
for the isolation of recovered alcohol (S)-30 with excellent
enantiopurity (> 99:1 er) at 76% conversion (entry 6).
Table 7 Continuous flow KR of 3-substituted-3-hydroxyoxindole derivatives in ethyl
acetate using the same packed bed reactor

The KR of a second class of tertiary alcohol, 3-substituted-37m,21
hydroxypyrrolidin-2-ones,
was next investigated in
continuous flow using the same packed bed reactor. In line
with the previously-reported KR using homogeneous
7m
HyperBTM, the use of acetic anhydride as acylating agent
was required for high conversions. Good solubility of the
substrates in toluene allowed effective resolution, despite the
diminished swelling of the polymer support resulting in a
residence time of only 20 min. The sequential KR of six 3substituted-3-hydroxypyrrolidin-2-one derivatives 31–36 was
performed, with very good conversions and selectivities
16
obtained in each case (c = 40–53%; s = 18–42) (Table 8). The
robustness of the same packed bed reactor was further
exemplified by the 7.25 mmol-scale KR of N-benzyl-3-hydroxy3-phenylpyrrolidin-2-one 33 in a continuous flow process over
a 24 h period (Scheme 4). Significantly, all continuous flow
experiments described in this manuscript, including
Table 8 Continuous flow KR of pyrrolidin-2-one derivatives in toluene using the same
packed bed reactor

Conversion and er determined by chiral HPLC analysis. Selectivity factor (s)
calculated using alcohol er and reaction conversion (see ref. 9a), and rounded
according to estimated associated errors (see ref. 9b). Alcohol er given as S:R,
ester er given R:S. a R:S. b S:R. c 0.25 mmol scale. d 30 (0.07 M), (i-PrCO)2O, i-PrNEt2
(0.14 M).

Conversion and er determined by chiral HPLC analysis. Selectivity factor (s)
calculated using alcohol er and reaction conversion (see ref. 9a), and rounded
according to estimated associated errors (see ref. 9b). Alcohol er given as R:S,
ester er given S:R. a S:R. b R:S.

optimisation studies, were performed with the same sample of
Merrifield resin-supported HyperBTM catalyst 2, resulting in a
total operation time of over 100 h.

Acknowledgements
We thank the Royal Society and the Science and Engineering
Board of India (SERB) for the award of a Royal Society-SERB
Newton International Fellowship (N.R.G.). We also thank the
EPSRC Centre for Doctoral Training in Critical Resource
Catalysis (CRITICAT, grant code EP/L016419/1, R.M.N.P.) for
funding. Financial support from the EPSRC (EP/K000411/1) is
gratefully acknowledged (R.C.). We thank the European
Research Council under the European Union's Seventh
Framework Programme (FP7/2007-2013) ERC grant agreement
no. 279850 (A.D.S). A.D.S. thanks the Royal Society for a
Wolfson Research Merit Award. We also thank the EPSRC UK
National Mass Spectrometry Facility at Swansea University.
C.R.-E. and M.A.P. acknowledge the financial support from
CERCA Programme/Generalitat de Catalunya, MINECO
(CTQ2015-69136-R, AEI/MINECO/FEDER, UE and Severo Ochoa
Excellence Accreditation 2014–2018, SEV-2013-0319) and DEC
Generalitat de Catalunya (Grant 2014SGR827).

Notes and references
Scheme 4 Continuous flow KR of N-benzyl-3-hydroxy-3-phenylpyrrolidin-2-one (±)-33.

Conclusions
The synthesis of polymer-supported Lewis base catalysts,
based on the homogeneous isothioureas HyperBTM and BTM,
was achieved, and these catalysts were applied for the
acylative kinetic resolution of a range of secondary and tertiary
heterocyclic alcohols. In batch, the use of industriallypreferable solvents was investigated, with dimethyl carbonate
proving the most generally applicable. Notably, the HyperBTMderived immobilised catalysts were readily recycled, with no
loss in either activity or selectivity. In contrast, the polymersupported BTM derivative showed diminished reactivity, with
7i
the catalyst being deactivated following a single experiment.
In addition to the kinetic resolution of secondary benzylic,
propargylic, allylic and cycloalkanol derivatives, a range of 22
tertiary heterocyclic alcohols, based on privileged 3hydroxyoxindole and 3-hydroxypyrrolidinone substructures,
were resolved with up to excellent selectivity (s = 7–190). The
immobilised HyperBTM catalysts were applied in a packed bed
reactor to demonstrate the first example of the kinetic
resolution of tertiary alcohols in a continuous flow process.
Although using chloroform as solvent gave the highest
selectivities in flow, high selectivities were also obtained for
the
resolution
of
3-hydroxyoxindole
and
3hydroxypyrrolidinone derivatives in the industrially-preferable
22
solvents of ethyl acetate and toluene.

Conflicts of interest
There are no conflicts to declare.

1
2

3

4
5

6
7

P. I. Dalko, Comprehensive Enantioselective Organocatalysis:
Catalysts, Reactions, and Applications, Wiley-VCH,
Weinheim, 2013.
(a) M. Benaglia, A. Puglisi and F. Cozzi, Chem. Rev., 2003,
103, 3401–3429; (b) A. F. Trindade, P. M. P. Gois and C. A. M.
Alfonso, Chem. Rev., 2009, 109, 418–514; (c) J. Lu and P. H.
Toy, Chem. Rev., 2009, 109, 815–838; (d) T. E. Kristensen and
T. Hansen, Eur. J. Org. Chem., 2010, 3179–3204; (e) B. Altava,
M. I. Burguete, E. Garcia-Verdugo and S. V. Luis, Chem. Soc.
Rev., 2018, 47, 2722–2771.
(a) A. Puglisi, M. Benaglia and V. Chiroli, Green Chem., 2013,
15, 1790–1813; (b) C. Rodríguez-Escrich and M. A. Pericàs,
Eur. J. Org. Chem., 2015, 2015, 1173–1188; (c) I. Atodiresei,
C. Vila and M. Rueping, ACS Catal., 2015, 5, 1972–1985.
R. M. Neyyappadath, R. Chisholm, M. D. Greenhalgh, C.
Rodríguez-Escrich, M. A. Pericàs, G. Hähner and A. D. Smith,
ACS Catal., 2018, 8, 1067–1075.
For other examples of solid-supported catalysts used for the
acylative kinetic resolution of secondary alcohols see: (a) B.
Clapham, C.-W. Cho and K. D. Janda, J. Org. Chem., 2001, 66,
868–873; (b) G. T. Copeland and S. J. Miller, J. Am. Chem.
Soc., 2001, 123, 6496–6502; (c) B. Pelotier, G. Priem, I. B.
Campbell, S. J. F. Macdonald and M. S. Anson, Synlett, 2003,
5, 679–683; (d) K. Ishihara, Y. Kosugi, M. Akakura, J. Am.
Chem. Soc., 2004, 126, 12212–12213; (e) A. Gissibl, M. G.
Finn and O. Reiser, Org. Lett., 2005, 7, 2325–2328; (f) A.
Gissibl, C. Padié, M. Hager, F. Jaroschik, R. Rasappan, E.
Cuevas-Yañez, C,-O. Turrin, A.-M. Caminade, J.-P. Majoral
and O. Reiser, Org. Lett., 2007, 9, 2895–2898; (g) A. Schätz,
M. Hager and O. Reiser, Adv. Funct. Mater., 2009, 19, 2109–
2115; (h) O. Gleeson, R. Tekoriute, Y. K. Gun’ko and S. J.
Connon, Chem. Eur. J., 2009, 15, 5669–5673; (i) A. Schätz, R.
N. Grass, Q. Kainz, W. J. Stark and O. Reiser, Chem. Mater.,
2010, 22, 305–310.
For a review on isothiourea catalysis see: J. Merad, J.-M.
Pons, O. Chuzel and C. Bressy, Eur. J. Org. Chem., 2016, 2016,
5589–5610.
For examples of isothiourea-catalysed acylative KR of
alcohols see: (a) V. B. Birman and X. Li, Org. Lett., 2006, 8,
1351–1354; (b) V. B. Birman and L. Guo, Org. Lett., 2006, 8,
4859–4861; (c) V. B. Birman and X. Li, Org. Lett., 2008, 10,
1115–1118; (d) Y. Zhang and V. B. Birman, Adv. Synth. Catal.,

8

9

10

11

12

13
14
15

16

17
18

2009, 351, 2525–2529; (e) Q. Xu, H. Zhou, X. Geng and P.
Chen, Tetrahedron, 2009, 65, 2232–2238; (f) I. Shiina, K.
Nakata, K. Ono, M. Sugimoto and A. Sekiguchi, Chem. Eur. J.,
2010, 16, 167–172; (g) D. Belmessieri, C. Joannesse, P. A.
Woods, C. MacGregor, C. Jones, C. D. Campbell, C. P.
Johnston, N. Duguet, C. Concellón, R. A. Bragg and A. D.
Smith, Org. Biomol. Chem., 2011, 9, 559–570; (h) I. Shiina, K.
Ono and K. Nakata, Chem. Lett., 2011, 40, 147–149; (i) X. Li,
H. Jiang, E. W. Uffman, L. Guo, Y. Zhang, X. Yang and V. B.
Birman, J. Org. Chem., 2012, 77, 1722–1737; (j) K. Nakata, K.
Gotoh, K. Ono, K. Futami and I. Shiina, Org. Lett., 2013, 15,
1170–1173; (k) I. Shiina, K; Ono and T. Nakahara, Chem.
Commun., 2013, 49, 10700–10702; (l) S. F. Musolino, O. S.
Ojo, N. J. Westwood, J. E. Taylor and A. D. Smith, Chem. Eur.
J., 2016, 22, 18916–18922; (m) M. D. Greenhalgh, S. M.
Smith, D. M. Walden, J. E. Taylor, Z. Brice, E. R. T. Robinson,
C. Fallan, D. B. Cordes, A. M. Z. Slawin, H. C. Richardson, M.
A. Grove, P. H.-Y. Cheong and A. D. Smith, Angew. Chem. Int.
Ed., 2018, 57, 3200–3206.
For previous examples of polymer-supported isothiourea
catalysts see: (a) J. Izquierdo and M. A. Pericàs, ACS Catal.,
2016, 6, 348–356; (b) S. Wang, J. Izquierdo, C. RodríguezEscrich, and M. A. Pericàs, ACS Catal., 2017, 7, 2780–2785;
(c) S. Wang, C. Rodríguez-Escrich and M. A. Pericàs, Angew.
Chem. Int. Ed., 2017, 56, 15068–15072.
(a) H. B. Kagan and J. C. Fiaud, in Topics in Stereochemistry,
ed. E. L. Eliel and S. H. Wilen, John Wiley & Sons, 1988, vol.
18, pp. 249–330; (b) M. D. Greenhalgh, J. E. Taylor and A. D.
Smith, Tetrahedron, DOI:10.1016/j.tet.2018.05.069
(a) K. Alfonsi, J. Colberg, P. J. Dunn, T. Fevid, S. Jennings, T. A.
Johnson, H. P. Kleine, C. Knight, M. A. Nagy, D. A. Perry and
M. Stefaniak, Green Chem., 2008, 10, 31–36; (b) R. K.
Henderson, C. Jiménez-González, D. J. C. Constable, S. R.
Alston, G. G. A. Inglis, G. Fisher, J. Sherwood, S. P. Binks and
A. D. Curzona, Green Chem., 2011, 13, 854–862; (c) D. Prat,
O. Pardigon, H.-W. Flemmin, S. Letestu, V. Ducandas, P.
Isnard, E. Guntrum, T. Senec, S. Ruisseau, P. Cruciani and P.
Hosek, Org. Process Res. Dev., 2013, 17, 1517–1525; (d) H. E.
Eastman, C. Jamieson and A. J. B. Watson, Aldrichimica Acta,
2015, 48, 51–55; (e) D. Prat, A. Wells, J. Hayler, H. Sneddon,
C. R. McElroy, S. Abou-Shehada and P. J. Dunn, Green Chem.
2016, 18, 288–296.
(a) B. Merrifield, Science, 1983, 232, 341–347; (b) E. Bayer,
Angew. Chem. Int. Ed. Engl., 1991, 30, 113–129; (c) I. W.
James, Tetrahedron, 1999, 55, 4855–4946; (d) F. Guillier, D.
Orain and M. Bradley, Chem. Rev., 2000, 100, 2091–2157.
For a review see: (a) A. E. Fernandez, A. M. Jonas and O.
Riant, Tetrahedron, 2014, 70, 1709–1731. For seminal
reports see: (b) V. V. Rostovtsev, L. G. Green, V. V. Fokin, K.
B. Sharpless, Angew. Chem. Int. Ed., 2002, 41, 2596−2599; (c)
C. W. Tornøe, C. Christensen and M. Meldal, J. Org. Chem.,
2002, 67, 3057−3064.
A. Bastero, D. Font and M. A. Pericàs, J. Org. Chem., 2007,
72, 2460–2468.
See the Supporting Information for full details.
Each polymer-supported catalyst was only prepared at a
single functionalisation, and therefore any effect of polymer
functionalisation on catalytic performance cannot be
assessed.
The absolute configurations of the ester product and
recovered alcohol were assigned by comparison of specific
rotation values with the literature, see Supporting
Information.
For the full range of solvents investigated see the Supporting
Information.
(a) E. R. Jarvo, C. A. Evans, G. T. Copeland and S. J. Miller, J.
Org. Chem., 2001, 66, 5522–5527; (b) M. C. Angione and S. J.
Miller, Tetrahedron, 2006, 62, 5254–5261; (c) S. Lu, S. B. Poh,

19
20
21

22

W.-Y. Siau and Y. Zhao, Angew. Chem. Int. Ed., 2013, 52,
1731–1734.
For a review on the bioactivity of 3-substituted-3hydroxyoxindole derivatives see: S. Peddidhotla, Curr. Bioact.
Compd., 2009, 5, 20–38.
For a computational model of alcohol acylation transition
state structures that rationalise this enantiodiscriminiation,
see ref 7m.
For examples documenting the bioactivity of 3hydroxypyrrolidinone derivatives, see: (a) F. Damiani, D.
Hamprecht, A. A. Jaxa-Chamiec, F. Micheli, A. Pasquarello
and G. Tedesco, Int. Pat., WO 03/089409 A1, 2003; (b) K.-K.
Ho, W. Quan and J. Zhou, Int. Pat., WO 2017/005069 A1,
2017; (c) J. L. Castro Pineiro, K. Dinnell, J. M. Elliott, G. J.
Hollingworth, D. E. Shaw and J. C. Swain, Int. Pat., WO
02/102372 A1, 2002; (d) P.-H. Liang, L.-W. Hsin and C.-Y.
Cheng, Bioorganic Med. Chem., 2002, 10, 3267–3276.
The research data underpinning this publication can be
found at DOI: http://dx.doi.org/10.17630/131807b4-8473434e-a77d-ee950ac25464.