The first supported N-pyridinyl proline-derived chiral catalyst for the kinetic resolution of racemic alcohols

Article abstract of DOI:10.1055/s-2003-38354

A family of polymer-supported chiral acylation catalysts derived from N-4′-pyridinyl-α-methyl-proline has been synthesised and screened in the kinetic resolution of racemic alcohol 1. The best supported catalyst 4 behaves in most cases as its solution phase analog 13 in the enantioselective acylation of a range of alcohols and can be recycled without loss of activity or selectivity.

Full text of DOI:10.1055/s-2003-38354

LETTER
679
The First Supported N-Pyridinyl Proline-derived Chiral Catalyst for the
Kinetic Resolution of Racemic Alcohols
N
-Pyridinyl Proli
é
ne-derived
a
Chiral Cata
t
lyst rice Pelotier,a Ghislaine Priem,b Ian B. Campbell,b Simon J. F. Macdonald,b Mike S. Anson*a
a
GlaxoSmithKline, Medicines Research Centre, Chemical Development, Gunnels Wood Road, Stevenage, SG1 2NY, UK
Fax +44(143)8764414; E-mail: mike.s.anson@gsk.com
b
GlaxoSmithKline, Medicines Research Centre, Medicinal Chemistry, Gunnels Wood Road, Stevenage, SG1 2NY, UK
Received 19 December 2002
Abstract: A family of polymer-supported chiral acylation catalysts
derived from N-4′-pyridinyl-α-methyl-proline has been synthesised
and screened in the kinetic resolution of racemic alcohol 1. The best
supported catalyst 4 behaves in most cases as its solution phase
analog 13 in the enantioselective acylation of a range of alcohols
and can be recycled without loss of activity or selectivity.
Key words: asymmetric catalysis, kinetic resolution, alcohols,
supported catalysts, enantioselective acylations
Scheme 1 Kinetic resolution of alcohol 1 with solution phase
catalysts A
Heterogeneous catalysis is an extensively growing pro-
cess due to the advantages it offers in the facile recovery
and recycling of the catalyst. This field has gained a re-
newed interest with the emergence of chiral heteroge-
neous catalysts for a wide range of metal-catalyzed
asymmetric reactions.1 Among the different techniques to
immobilize an enantioselective catalyst, the covalent at-
tachment of the chiral homogeneous catalyst to an organic
or inorganic polymer is an attractive approach. This com-
bines the advantages of solution phase catalysis (high ac-
tivity and selectivity) with those of solid phase catalysis
(recycling and stability of the catalyst). In this paper, we
describe a new chiral supported catalyst for the kinetic
resolution of racemic alcohols.
with a polymer-supported N-4′-pyridinyl proline-derived
catalyst.
A set of 8 solid supported catalysts (3–10, Scheme 2)
were prepared by coupling N-4′-pyridinyl-α-methyl-L-
proline 2 with different aminoalkylresins. The resins were
chosen in order to get supported catalysts with different
loadings (low loading -LL- or high loading -HL-), differ-
ent physical properties (swelling or not) and different
structural environments (no linker or linkers with variable
steric and electronic effects). The synthesis of acid precur-
sor 2 bearing a methyl substituent in the α-position of the
carboxyl moiety was developed to overcome the problems
of racemization encountered during the coupling reaction
in solution phase.11 The overnight heating of the ami-
noresin (1 equiv) with an excess of acid 2 (5 equiv) acti-
vated by HATU in the presence of diisopropylethylamine
in DMF afforded a quantitative coupling as confirmed by
a negative TNBS test and the mass increase of the poly-
mer after reaction.12
Over the last few years, efforts have been targeted to the
development of non-enzymatic nucleophilic catalysts for
asymmetric acyl transfer reactions.2 The discovery of 4-
dialkylaminopyridines as potent acylation catalysts3 led
chemists to design and synthesize new chiral catalysts
based on that structure, a few of which proved to be very
selective for the kinetic resolution of alcohols.4–6 Recent-
ly, we reported a new family of chiral acylating catalysts,
based on N-4′-pyridinyl-α-methyl proline structure A,
easily accessible in two synthetic steps.7 The best catalyst
was found to resolve alcohol 1 with a high selectivity
index8 (s = 13.2, Scheme 1). At this stage, we decided to
take advantage of the straightforward synthetic accessibil-
ity of these catalysts to rapidly develop a polymer-sup-
ported version.9 4-Dialkylaminopyridines anchored on a
solid support are known to exhibit catalytic properties in
acylation and hydrolysis reactions.10 Herein, we wish to
report the first example of enantioselective acylation
Catalysts 3–5 and 8–10 were thus obtained in one step
from commercially available solid supports. Catalysts 6
and 7 were prepared from aminoalkyl resins 1113 and 12,14
derived from Wang and amino-Merrifield resins by at-
tachment of an aromatic or an alkyl spacer unit respective-
ly (Scheme 3).
The chiral polymer-supported catalysts were screened in
the kinetic resolution of alcohol 1 with isobutyric anhy-
dride as the acylating agent.15 Dichloromethane instead of
toluene was chosen as solvent for its better swelling prop-
erties towards the resins. Furthermore, a previous study
had shown that higher selectivities could be achieved in
this solvent.7 As solid phase chemistry often requires
longer reaction times, an excess of isobutyric anhydride
Synlett 2003, No. 5, Print: 10 04 2003.
Art Id.1437-2096,E;2003,0,05,0679,0683,ftx,en;D28602ST.pdf.
© Georg Thieme Verlag Stuttgart · New York
ISSN 0936-5214

680
B. Pelotier et al.
LETTER
Scheme 2 Synthesis of solid supported catalysts 3–10
Scheme 3 Synthesis of amino-precursors 11 and 12
Conditions: (a) (i) DIC, DMAP, Wang resin, DMF, r.t., o/n; (ii) 20% piperidine in DMF, r.t., 1 h; (b) (i) HATU, DIPEA, LL amino-Merrifield
resin, DMF, 65 °C, o/n; (ii) 20% piperidine in DMF, r.t., 1 h.
(1.2 equiv) was added from the beginning and the reac- All supported catalysts proved to be enantioselective
tions were followed by HPLC. In order to compare the (s = 3.0–11.9). After 16 hours reaction (entries 1–8), cat-
performance of the different catalysts, all reactions were alysts 3, 4, 6 and 7 showed both higher activity
run in parallel and stopped at the same time. The results (conversion = 49–53%) and selectivity (s = 10.2–11.9).
are given in Table 1.
Although the reaction is slower with macroporous Argo-
pore-derived catalyst 5 compared to catalysts 3 or 4, the
selectivity is still high (s = 8.9, entry 3). The limited
Synlett 2003, No. 5, 679–683 ISSN 0936-5214 © Thieme Stuttgart · New York

LETTER
N-Pyridinyl Proline-derived Chiral Catalyst
681
Table 1 Kinetic Resolution of Alcohol 1 with Supported Catalysts
3–10a
therefore to test the recycling ability of these polymer-
supported acylation catalysts. For this purpose, the kinetic
resolution of alcohol 1 was performed several times with
the same sample of catalyst 4. Catalyst 4 was chosen for
practical reasons: more straightforward synthesis than
catalysts 6 and 7 and lower loading than catalyst 3. At the
end of each run, the supported catalyst was filtered off,
washed thoroughly with dichloromethane and added to a
fresh solution of reagents. As shown in Table 2, the cata-
lyst remained both very active and selective over the runs
and alcohol (–)-1 could still be recovered in 98% enantio-
meric purity after a fourth use of the catalyst.
Entry
Catalyst
Conversion Ee (%)c
(%)b
Selectivity
1
2
3
4
53
51
31
50
49
25
37
22
67
67
65
75
71
32
71
66
23
24
14
92
93
92
10.9
10.7
8.9
3
5
4
6
11.9
10.2
6.4
5
7
6
8
Table 2 Recycling of Catalyst 4 in the Kinetic Resolution of Alco-
hol 1a
7
9
3.0
Run
Time
(h)
conversion ee
Selectivity
8
10
3
3.3
(%)b
(%)c
9d
10d
11d
7.8
1
2
3
4
24
24
24
72
55
77
82
84
98
10.2
13.0
13.0
10.0
4
8.5
55
6
8.9
55
a Standard conditions: Catalyst 3–10 (5 mol%), alcohol 1 (1 equiv,
0.05 mmol, 13.1 mg), isobutyric anhydride (1.2 equiv, 0.06 mmol, 10
µL) in CH2Cl2 (500 µL), r.t., 16 h. Analysis: Chiralcel OD-H column,
227 nm, 1 mL/min, 5% EtOH/heptane on filtered crude mixtures.
b Conversion (%) = 100 × ee (recovered alcohol)/[ee (recovered alco-
hol) + ee (ester formed)].8
69
a Standard conditions: Catalyst 4 (0.33 mmol/g, 5 mol%, 5 µmol, 14.8
mg), alcohol 1 (1 equiv, 0.1 mmol, 26.3 mg), isobutyric anhydride
(1.5 equiv, 0.15 mmol, 25 µL) in CH2Cl2 (1 mL), r.t. Analysis: Chiral-
cel OD-H column, 227 nm, 1 mL/min, 5% EtOH/heptane.
b Conversion (%) = 100 × ee (recovered alcohol)/[ee (recovered alco-
hol) + ee (ester formed)].8
c Enantiomeric excess of the recovered alcohol.
d Reaction performed on a 0.1 mmol scale of alcohol 1 with 1.5 equiv
of isobutyric anhydride and stopped after 72 h. Results obtained after
purification.
c Enantiomeric excess of the recovered alcohol.
Finally, we conducted the kinetic resolution of various al-
cohols with solid-supported catalyst 415 and compared
this latter with the solution phase analog 13.7 The results
are reported in Table 3. Unless otherwise stated, all reac-
tions were performed in chloroform, the best solvent for
both the swelling of the resin and solubility of the alcohol
substrates. Alcohols 14 and 15 were poorly resolved, as
was the case in solution phase (entries 1 and 2). Catalyst
4 proved to be enantioselective in the kinetic resolution of
all other alcohols investigated. In general, comparable re-
sults between solution phase and solid phase catalysis
were obtained. For aminoalcohols 17 and 18, enantiomer-
ic purities greater than 90% were achieved in all cases for
the recovered alcohols (entries 4 and 5). Quite surprising-
ly however, the kinetic resolution of serine-derived alco-
hol 16 was not as effective in solid phase with a selectivity
decrease from 12.0 in solution phase to 2.9 (entry 3). This
unexpected result is difficult to rationalize: the direct
acylation of this more reactive primary alcohol may occur
more rapidly than the enantio-differentiating H-bonding
interaction between this alcohol and the solid phase cata-
lyst (see ref.7).
mobility and therefore accessibility of the catalytic site
due to the rigid nature of this resin might be the origin of
the decreased activity. Catalyst 8 with a poly(ethylenegly-
col) unit (PEG) between the polymer backbone and the
catalytic site did not perform as well as catalysts 3 and 4
without spacer, possibly because of competing hydrogen-
bonding the PEG unit can give rise to. We previously
demonstrated that hydrogen-bonding is involved in the
mechanism of the kinetic resolution of alcohols such as 1
with our solution phase catalysts.7 Finally, the presence of
bulky rink amide moieties close to the active site as in
catalysts 9 and 10 dramatically decreased the performance
of the catalysts (entries 7 and 8). According to the results
obtained with the best supported catalysts, it appears that
the loading of the polymer (3 vs 4) and the presence or ab-
sence of a spacer unit (3, 4 vs 6, 7) are not major factors
for the performance of the catalyst. After 72 hours, the ki-
netic resolution of alcohol 1 with best catalysts 3, 4 and 6
(entries 9–11) afforded alcohol (–)-1 in excellent enantio-
meric purity (92–93% ee). Most importantly, these sup-
ported catalysts achieved selectivities comparable to their
solution phase analogs.
In conclusion, we have synthesized chiral solid-phase
catalysts easily accessible in two steps from α-methyl-L-
proline and identified the best candidates for enantioselec-
tive acylation reactions. The recovery and recycling of
these catalysts was successfully demonstrated with poly-
One of the main advantages of hetereogeneous catalysis is
that the catalyst can be easily recovered by simple filtra-
tion at the end of the reaction and potentially reused with-
out loss of activity or selectivity. The next step was
Synlett 2003, No. 5, 679–683 ISSN 0936-5214 © Thieme Stuttgart · New York

682
B. Pelotier et al.
LETTER
Table 3 Kinetic Resolution of Different Alcohols with Supported Catalyst 4 vs Solution Phase Catalyst 13a
Entry
1b
Alcohol
Catalyst 4
Catalyst 137
Ee (%) @ conversion (%)
s
Ee (%) @conversion (%)
s
14
11 @ 59
1.3
22 @ 74
1.4
2b
3
15
16
17
18
6 @ 42
1.3
2.9
6.1
8.3
11 @ 65
99 @ 69
96 @ 59
98 @ 74
1.2
61 @ 71
91 @ 71
97 @ 72
12.0
18.8
7.9
4
5
a P=COC6H4-p-NMe2.
b Reactions run in CH2Cl2 with supported catalyst 4 and in toluene with catalyst 13.
mer-supported catalyst 4. Promising results were obtained
in the kinetic resolution of racemic alcohols and constitute
the first example of enantioselective acylation with a sup-
ported N-4′-pyridinyl proline-derived catalyst.
References
(1) Reviews: (a) Shuttleworth, S. J.; Allin, S. M.; Sharma, P. K.
Synthesis 1997, 1217. (b) Fodor, K.; Kolmschot, S. G. A.;
Sheldon, R. A. Enantiomer 1999, 4, 497. (c) Bein, T. Curr.
Opin. Solid State Mater. Sci. 1999, 4, 85. (d) Shuttleworth,
S. J.; Allin, S. M.; Wilson, R. D.; Nasturica, D. Synthesis
2000, 1035.
Acknowledgment
(2) Reviews: (a) Spivey, A. C.; Maddaford, A.; Redgrave, A. J.
Org. Prep. Proc. Int. 2000, 32, 333. (b) Long, Y. O.;
Paquette, L. A. Chemtracts 2000, 13, 1.
(3) Reviews: (a) Höfle, G.; Steglich, W.; Vorbrüggen, H.
Angew. Chem., Int. Ed. Engl. 1978, 17, 569. (b) Scriven, E.
F. V. Chem. Soc. Rev. 1983, 12, 129.
We are grateful to the European Commission for financial support
of this work via Marie Curie Fellowships (B. P. and G. P., HPMI-
CT-1999-00021).
(4) (a) Ruble, J. C.; Tweddell, J.; Fu, G. C. J. Org. Chem. 1998,
63, 2794. (b) Tao, B.; Ruble, J. C.; Hoic, D. A.; Fu, G. C. J.
Am. Chem. Soc. 1999, 121, 5091.
Synlett 2003, No. 5, 679–683 ISSN 0936-5214 © Thieme Stuttgart · New York

LETTER
N-Pyridinyl Proline-derived Chiral Catalyst
683
(5) (a) Kawabata, T.; Nagato, M.; Takasu, K.; Fuji, K. J. Am.
Chem. Soc. 1997, 119, 3169. (b) Kawabata, T.; Yamamoto,
K.; Momose, Y.; Yoshida, H.; Nagaoka, Y.; Fuji, K. Chem.
Commun. 2001, 2700.
(6) Spivey, A. C.; Fekner, T.; Spey, S. E. J. Org. Chem. 2000,
65, 3154.
mmol/g, 1 equiv, 0.75 mmol, 664 mg) followed by 4-
dimethylaminopyridine (0.1 equiv, 0.075 mmol, 10 mg)
were added. After 16 h stirring at r.t., the resin was filtered
and thoroughly washed with DMF (3 ×) and CH2Cl2 (5 ×).
The resin was dried overnight at 50 °C in a vacuum oven
(946 mg). A loading of 0.81 mmol/g was calculated from the
mass increase of the polymer. The dry resin was treated with
a solution of 20% piperidine in DMF (5 mL) for 20 min and
filtered and washed with DMF. The operation was repeated
twice and resin 11 was finally washed with CH2Cl2 (5 ×) and
dried in a vacuum oven at 50 °C.
(7) Priem, G.; Pelotier, B.; Macdonald, S. J. F.; Anson, M. S.;
Campbell, I. B. J. Org. Chem. 2003, in press.
(8) Kagan, H. B.; Fiaud, J. C. Top. Stereochem. 1988, 18, 249.
(9) For the only other example of chiral polymer-supported
catalyst for the kinetic resolution of alcohols, see: Clapham,
B.; Cho, C.-W.; Janda, K. D. J. Org. Chem. 2001, 66, 868.
(10) (a) Goe, G. L.; Marston, C. R.; Scriven, E. F. V.; Sowers, E.
E. Chem. Ind. (London) 1990, 40, 275; and references
therein. (b) Taylor, S. J.; Morken, J. P. Science 1998, 280,
267.
(11) Priem, G.; Anson, M. S.; Macdonald, S. J. F.; Pelotier, B.;
Campbell, I. B. Tetrahedron Lett. 2002, 43, 6001.
(12) General Procedure for the Coupling of Acid 2 with
Aminoresins: To a solution of acid 2 (5 equiv) in DMF (7
mL/mmol) was added HATU (4.9 equiv). The mixture was
warmed to 65 °C and stirred until all the reagents had
dissolved. Diisopropylethylamine (10 equiv) was added,
followed by the aminoresin (1 equiv) and the resulting
suspension was stirred at that temperature overnight. The
resin was then filtered into an Alltech tube, washed
thoroughly with DMF (3×) and CH2Cl2 (5×) and dried in a
vacuum oven at 50 °C. A negative TNBS test indicated all
amine sites had reacted.
(14) Aminoresin 12: A solution of 8-(Fmoc-amino)octanoic acid
(5 equiv, 0.36 mmol, 137 mg) in DMF (2 mL) was cooled to
0 °C and HATU (4.9 equiv, 0.35 mmol, 133 mg) was added.
The solution was stirred for 1 h under nitrogen and amino-
Merrifield resin (0.36 mmol/g, 1 equiv, 0.072 mmol, 200
mg) was added, followed by diisopropylethylamine (10
equiv, 0.72 mmol, 93 mg). The resulting suspension was
stirred at r.t. overnight. The resin was filtered, washed with
DMF (3 ×) and CH2Cl2 (5 ×) and dried in a vacuum oven at
50 °C (231 mg). A loading of 0.31 mmol/g was calculated
from the mass increase of the polymer. The dry resin was
treated with a solution of 20% piperidine in DMF (4 mL) for
20 min and filtered and washed with DMF. The operation
was repeated twice and resin 12 was finally washed with
CH2Cl2 (5×) and dried in a vacuum oven at 50 °C.
(15) Typical Procedure for the Kinetic Resolution of Alcohols
with Supported Catalysts: The supported catalyst (5
mol%) was weighed in an Alltech tube and swollen in
CH2Cl2 for 1 h. The solvent was then filtered and a solution
of alcohol (1 equiv) in CH2Cl2 or CHCl3 (10 mL/mmol) was
added, followed by isobutyric anhydride (1.2 equiv). The
resulting mixture was shaken at r.t.
(13) Aminoresin 11: A solution of 4-(Fmoc-aminomethyl)benz-
oic acid (6 equiv, 4.5 mmol, 1.68 g) in DMF (8 mL) was
cooled to 0 °C and diisopropylcarbodiimide (5.3 equiv, 4.0
mmol, 660 µL) was added. After 45 min stirring at that
temperature, the ice bath was removed and Wang resin (1.13
Synlett 2003, No. 5, 679–683 ISSN 0936-5214 © Thieme Stuttgart · New York