Effects of methyl substituents on the activity and enantioselectivity of homobenzotetramisole-based catalysts in the kinetic resolution of alcohols

Article abstract of DOI:10.1002/adsc.200900383

Substitution of the tetrahydropyrimidine ring in the enantioselective acyl transfer catalyst homobenzotetramisole (HBTM) 6 with methyl groups exerts a dramatic influence on its performance in the kinetic resolution of secondary alcohols. The syn-3-methyl analogue of HBTM (9a) has proved to be superior to the parent compound in terms of catalytic activity, enantioselectivity, and synthetic accessibility.

Full text of DOI:10.1002/adsc.200900383

UPDATES
DOI: 10.1002/adsc.200900383
Effects of Methyl Substituents on the Activity and
Enantioselectivity of Homobenzotetramisole-Based Catalysts
in the Kinetic Resolution of Alcohols
Yuhua Zhanga and Vladimir B. Birmana,*
a
Department of Chemistry, Washington University, Campus Box 1134, One Brookings Drive, St. Louis, Missouri 63130,
USA
Fax : (+1)-314-935-4481; phone: (+1)-314-935-9188; e-mail: birman@wustl.edu
Received: June 2, 2009; Revised: July 27, 2009; Published online: October 1, 2009
Supporting information for this article is available on the WWW under
http://dx.doi.org/10.1002/adsc.200900383.
Abstract: Substitution of the tetrahydropyrimidine
ring in the enantioselective acyl transfer catalyst ho-
mobenzotetramisole (HBTM) 6 with methyl groups
exerts a dramatic influence on its performance in
the kinetic resolution of secondary alcohols. The
syn-3-methyl analogue of HBTM (9a) has proved
to be superior to the parent compound in terms of
catalytic activity, enantioselectivity, and synthetic
accessibility.
utility in the kinetic resolution (KR)[2] of several
classes of secondary alcohols[3] and 2-oxazolidinones[4]
bearing a p-system a- to the nucleophilic atom. Sever-
al applications of catalysts 3 and 4 have been recently
reported by other groups.[5]
More recently, we investigated their tetrahydropyri-
midine analogues, homotetramisole (HTM) 5 and ho-
mobenzotetramisole (HBTM) 6.[6] HBTM proved to
be effective in the KR of both a- and b-aryl-substitut-
ed alcohols[6a] and of a-substituted carboxylic acids.[6b]
It was also considerably more active and more toler-
ant of reaction conditions than the previously devel-
oped imidazoline-based catalysts. Furthermore, com-
pared to the latter, the structure of HBTM provides
more opportunities for introducing additional sub-
stituents in the non-aromatic ring, which could alter
Keywords: acylation; catalyst design; kinetic resolu-
tion; organocatalysis
Enantioselective acyl transfer catalysis has been suc- its conformation and/or provide additional interac-
cessfully applied to a variety of synthetically useful tions with the substrate molecules, thus potentially en-
asymmetric processes.[1] Therefore, there is a contin- hancing its catalytic activity and enantioselectivity.
ued need to develop new catalysts that will be highly With this in mind, we decided to synthesize methyl-
active, enantioselective, general, and, last but not substituted HBTM analogues 9a, 9b, 10a and 10b and
least, easily obtainable. Our efforts in this area initial- evaluate their performance in acylation reactions.
ly resulted in a series of imidazoline-based catalysts Thanks to recent advances in the stereo- and enantio-
1–4 illustrated in Figure 1. We demonstrated their selective Mannich reactions, the four requisite 1,3-
amino alcohols are easily available in enantiopure
form (Scheme 1), thus obviating the need for the clas-
sical resolution employed in the original synthesis of
6.[6a]
Chiral 1,3-amino alcohol 7a was prepared via the
syn-selective proline-catalyzed asymmetric Mannich
reaction according to the protocol described by Haya-
shi et al.[7] followed by removal of the p-methoxy-
phenyl group by periodate oxidation[8] (Scheme 1). Its
diastereomer 7b was synthesized from N-propionyl-
(S,S)-pseudoephedrine via the highly anti-selective
Mannich reaction disclosed by Badia et al.[9] Prepara-
tion of the diastereomeric amino alcohols 8a and 8b
relied on the asymmetric Mannich reaction procedure
Figure 1. Amidine-based catalysts.
Adv. Synth. Catal. 2009, 351, 2525 – 2529
ꢀ 2009 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
2525

Yuhua Zhang and Vladimir B. Birman
UPDATES
Scheme 1. Preparation of HBTM analogues. Conditions: (a) l-proline, DMF, À208C; then NaBH4;[7] (b) HIO4, H2SO4,
MeCN-H2O;[8] (c) 2-chlorobenzothiazole, (i-Pr)2NEt; (d) MsCl, NEt3, CH2Cl2; (e) LDA; ZnCl2;[9] (f) H2SO4, MeOH;[9] (g)
LiCl, NaBH4, THF, EtOH, 08C (96%); (h) l-proline, acetone;[10] (i) NaBH4, MeOH, 08C (61% anti-, 38% syn-); (j)
CF3CO2H, CH2Cl2 (quant.).
reported by List et al.[10] Reaction of amino alcohols nol 11 (Table 2) and (Æ)-trans-2-phenylcyclohexanol
7a, 7b, 8a and 8b with 2-chlorobenzothiazole followed 12 (Table 3). Again, 9a emerged as the most active of
by cyclization delivered 9a, 9b, 10a, and 10b, respec- the five catalysts in both reactions. It was somewhat
tively. With sufficient quantities of the four methyl- inferior in terms of selectivity to HBTM in the first
substituted HBTM analogues in hand, we undertook test, but surpassed it in the second. Catalyst 10a,
their systematic comparison with the parent catalyst albeit less active than HBTM, produced the highest
(6). The first test reaction – acylation of methanol selectivity factors in the acylation of both alcohols.
with acetic anhydride – revealed the significant influ- On the other hand, catalysts 9b and 10b displayed
ence of the substitution pattern on the catalytic activi- drastically reduced activities in both tests, requiring
ty (Table 1). Catalyst 9a demonstrated the best perfor- 10-fold higher catalyst loadings to achieve acceptable
mance, whereas 9b and 10a were somewhat less active reaction rates.
than HBTM. In sharp contrast, 10b proved to be far
less active than the rest.
Catalyst 9b produced dismal selectivities in the KR
of both 11 and 12, whereas 10b was only effective in
Even more remarkable differences were observed the case of the former substrate.
in the KR of secondary alcohols (Æ)-1-phenylpropa-
Taking into account its synthetic accessibility, cata-
lyst 9a (dubbed HBTM-2) was deemed to be the most
Table 1. Methanol test.
Table 2. KR of (Æ)-1-phenylpropanol 11.[a]
[a]
Entry
Catalyst
t1/2
Entry Catalyst (mol%) Time [h] Conversion [%]
S
1
2
3
4
5
HBTM 6
6 min
1
2
3
4
5
6 (1)
9a (1)
9b (10)
10a (1)
10b (10)
1.5
0.90
4.0
4.1
24
48
50
49
49
48
26
22
4.4
40
29
syn-3-Me-HBTM 9a
anti-3-Me-HBTM 9b
syn-4-Me-HBTM 10a
anti-4-Me-HBTM 10b
4.5 min
13 min
12 min
7.5 h
[a]
[a]
Time required to achieve 50% conversion was deter-
mined by monitoring the reactions by H NMR.
Conditions: 0.25M 11, 0.75 equiv. (EtCO)2O, 0.75 equiv.
(i-Pr)2NEt, CDCl3, room temperature.
1
2526
asc.wiley-vch.de
ꢀ 2009 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Adv. Synth. Catal. 2009, 351, 2525 – 2529

Effects of Methyl Substituents on the Activity and Enantioselectivity
UPDATES
Table 3. KR of (Æ)-trans-2-phenylcyclohexanol 12.[a]
comparison. Several cyclohexanol derivatives with a
p-system b to the hydroxy were resolved with equal
or higher enantioselectivities than those previously
achieved with HBTM (entries 1–5). In the case of the
structurally similar, but fully saturated substrate men-
thol, HBTM-2 was found to be completely ineffective
(entry 6). Useful levels of selectivity were also ob-
tained with representative benzylic (11), propargylic
(20) and cinnamyl (21) alcohols, although they fell
short of those previously obtained with benzotetrami-
Entry Catalyst (mol%) Time [h] Conversion [%]
S
1
2
3
4
5
6 (1)
9a (1)
9b (10)
10a (1)
10b (10)
3.0
1.0
13
7.0
40
49
49
54
47
19
31
38
1.2
44
sole and
(BTM)
4
7-chloro-2-phenyl-2,3-
dihydroimidazoHCATUNGTERN[NUNG 1,2-a]quinoline (Cl-PIQ) 2 (entries 7–
2.9
9). However, HBTM-2 proved to be clearly superior
in the KR of benzylic amino alcohol 22, which under-
goes rapid background acylation and thus cannot be
resolved efficiently with the less reactive BTM.[11] It is
noteworthy that the acylation of 21 catalyzed by 9a
[a]
Conditions: 0.25M 12, 0.75 equiv. (EtCO)2O, 0.75 equiv.
(i-Pr)2NEt, CDCl3, room temperature.
practical and thus was selected for further study. In- was quite rapid, whereas the structurally similar BTM
vestigation of solvent and temperature effects on the was inhibited by this substrate.[3c]
KR of 12 with HBTM-2 (Table 4) revealed qualita-
In conclusion, we have developed HBTM-2, a
tively similar trends to those displayed by HBTM,[6a] highly active and versatile enantioselective acylation
except that the former performed equally well in pure catalyst obtainable from inexpensive starting materi-
toluene and in the toluene-tert-amyl alcohol mixture als in four steps. The asymmetric route used to pre-
at À408C, which led to a more convenient experimen- pare it can be easily adapted for the synthesis of vari-
tal protocol (entries 9 and 13). It is also noteworthy ously substituted HBTM analogues. We have also
that HBTM-2 typically produced higher selectivity demonstrated that the substitution pattern of HBTM
factors in shorter times than HBTM at half the cata- analogues exerts a profound influence on their cata-
lyst loadings of the latter (e.g., entries 9 vs. 14, 13 vs. lytic activity and enantioselectivity. Theoretical stud-
15.)
ies linking these effects with the conformational and
The substrate scope of 9a was evaluated using 1% steric changes induced by the methyl substituents are
catalyst loadings in toluene at À408C (Table 5). The ongoing and will be reported in due course.
best results previously obtained with amidine-based
catalysts 2, 4 and 6 are listed in the right column for
Experimental Section
Table 4. Optimization of reaction conditions for 9a[a]
A: (2S,3S)-3-[(2-Benzothiazolyl)amino]-2-methyl-3-
phenylpropanol-1 (23a)
Entry Catalyst Solvent,
(mol%) t [8C]
Time Conversion
S
G
[h]
[%]
Amino alcohol 7a was prepared in enantio- and diastereo-
merically pure form (>99% ee and de by HPLC) via a com-
bination of two literature procedures[7,8] and repeated recrys-
tallization of the N-PMP derivative from ethyl acetate. A
25-mL pressure tube charged with 2-chlorobenzothiazole
(0.378 g, 2.2 mmol), 7a (0.348 g, 2.0 mmol), (i-Pr)2NEt
(0.70 mL, 4.0 mmol) and a stir bar was flushed with N2 sev-
eral times, stoppered and heated at 135Æ58C for 24 h. After
cooling the tube to 30–408C, the viscous reaction mixture
was treated with 2 mL of CH2Cl2 and left at room tempera-
ture to dissolve. The diluted mixture was applied directly to
a chromatographic column and eluted with EtOAc/hexanes/ACHTUNG-
1
2
3
4
5
6
7
8
9a (1)
9a (1)
9a (2)
9a (2)
9a (1)
9a (1)
9a (1)
9a (1)
9a (2)
9a (2)
9a (1)
9a (2)
9a (2)
6 (4)
CDCl3, 23
CDCl3, 0
1.4
1.5
49
51
54
53
51
48
51
54
52
25
54
47
50
45
46
38
49
66
75
20
30
33
44
131
127
53
87
132
67
CDCl3, À20 1.7
CDCl3, À40 2.0
MeCN, 23
CH2Cl2, 23
THF, 23
1.5
1.1
1.0
0.67
2.3
PhMe, 23
PhMe, À40
PhMe, À63
TA, 23
9
10
11
12
13
14
15[b]
4.0
0.35
0.42
2.3
2.5
3.0
TERNN(UG 1:2) to afford the title compound as a white solid; yield:
TA, À8
0.459 g (77%). 1H NMR (300 MHz, CDCl3): d=7.52–7.22
(m, 8H), 7.07–7.02 (m, 1H), 5.58 (br, 1H), 5.08–4.98 (m,
1H), 3.70–3.52 (m, 2H), 2.58–2.40 (m, 1H), 0.82 (d, J=
6.9 Hz, 3H); 13C NMR (75 MHz, CDCl3): d=168.3, 151.6,
139.3, 130.3, 128.7, 127.7, 126.2, 121.8, 121.1, 118.8, 64.5,
TAT, À40
PhMe, À40
TAT, À40
6 (4)
101
[a]
Conditions: 0.25M 12, 0.75 equiv. (EtCO)2O, 0.75 equiv.
(i-Pr)2NEt.
62.1, 40.6, 12.5; IR (film): n=3270, 1599, 1538, 1446 cmÀ1
;
Data from Ref.[6]; TA=tert-amyl alcohol; TAT=tert-
amyl alcohol/toluene 1:1 mixture.
HR-ESI-MS: m/z=299.1212, calcd. for C17H19N2S+ (M+
H+): 299.1213; [a]2D0: À116.08 (c 0.1, MeOH); mp 149–1508C.
[b]
Adv. Synth. Catal. 2009, 351, 2525 – 2529
ꢀ 2009 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
asc.wiley-vch.de
2527

Yuhua Zhang and Vladimir B. Birman
UPDATES
Table 5. Substrate scope of HBTM-2 9a.
Entry
1
(Æ)-Substrate[a]
Result with HBTM-2 9a[b]
S=112 (52%/11 h)
Best previous result
HBTM,[6] S=107 (51%/10 h)
2
3
S=66 (51%/11 h)
S=16 (40%/11 h)
HBTM,[6] S=44 (44%/10 h)
HBTM,[6] S=10 (55%/12 h)
4
5
6
S=10 (52%/11 h)
S=5.7 (43%/11 h)
S=1.2 (55%/11 h)
HBTM,[6] S=10 (26%/10 h)
HBTM,[6] S=5.6 (28%/10 h)
HBTM,[6] S=3.2 (33%/10 h)
7
8
9
S=54 (49%/11 h)
S=21 (53%/11 h)
S=23 (54%/11 h)
S=19 (48%/11 h)
BTM,[3c] S=145 (45%/36 h)
BTM,[3d] S=31 (59%/11 h)
Cl-PIQ,[3b] S=27 (44%/8 h)
BTM,[11] S=8.2 (44%/5.5 h)
10
[a]
Absolute configuration of the fast-reacting enantiomer is shown.
Conditions: 0.25M substrate, 0.55 equiv. (EtCO)2O, 0.55 equiv. (i-Pr)2NEt, 2 mol% HBTM-2, PhMe, À408C.
[b]
B: (2S,3R)-3,4-Dihydro-3-methyl-2-phenyl-2H-
pyrimido-[2,1-b]benzothiazole (9a)
C: Kinetic Resolution using HBTM-2 (9a); General
Procedure
A solution of 23a (0.459 g, 1.54 mmol) in anhydrous CH2Cl2
(40 mL) was cooled to 08C under N2 and treated with NEt3
(0.66 mL, 4.7 mmol) followed by MsCl (0.24 mL, 3.1 mmol).
The mixture was stirred at 08C for 1 h and then warmed to
room temperature. Methanol (0.2 mL) was added to quench
the excess MsCl. NEt3 (1.5 mL) was added and the mixture
was refluxed for 5 h. After cooling, the mixture was washed
with a small amount of water, dried over MgSO4 and rotary
evaporated. The crude product was purified by chromatog-
raphy (5% isopropyl alcohol, 1% NEt3 in hexanes) to give
9a as light yellow solid; yield: 0.311 g (72%). 1H NMR
(300 MHz, CDCl3): d=7.33–7.15 (m, 7H), 7.01–6.95 (m,
1H), 6.73 (d, J=7.8 Hz, 1H), 4.71 (d, J=4.2 Hz, 1H), 3.79
(dd, J=11.1, 5.1 Hz, 1H), 3.33 (dd, J=11.1, 7.5 Hz, 1H),
2.44–2.32 (m, 1H), 0.78 (d, J=6.9 Hz, 3H); 13C NMR
(75 MHz, CDCl3): d=157.8, 141.3, 141.0, 128.3, 128.0, 127.1,
126.1, 123.0, 122.1, 122.0, 107.7, 63.1, 46.6, 29.5, 14.1; IR
(film): n=1619, 1580, 1477, 1462 cmÀ1; HR-ESI-MS: m/z=
281.1105, calcd. for C17H17N2S+ (M+H+): 281.1107; [a]D20:
+212.08 (c 0.1, MeOH); mp 88–898C.
A stock solution was prepared by dissolving catalyst 9a
(3 mg, 0.01 mmol) and (i-Pr)2NEt (95 mL, 0.55 mmol) in
PhMe in a 2.0-mL volumetric test tube and bringing the
volume to the mark. Each KR experiment was performed in
duplicate. A one-dram vial was charged with 0.25 mmol of a
substrate, 100 mg of Na2SO4, 0.50 mL of PhMe and 0.50 mL
of the stock catalyst solution and stirred at À408C for
15 min followed by addition of propionic anhydride (18 mL,
0.14 mmol). After 11 h, the reaction mixture was quenched
by addition of methanol, worked up and analyzed as de-
scribed previously.[3,6a] The results are presented in Table 5
above.
Acknowledgements
This study was supported in part by NIGMS (NIH R01
GM072682). Mass spectrometry was provided by the Wash-
2528
asc.wiley-vch.de
ꢀ 2009 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Adv. Synth. Catal. 2009, 351, 2525 – 2529

Effects of Methyl Substituents on the Activity and Enantioselectivity
UPDATES
ington University Mass Spectrometry Resource, an NIH Re-
search Resource (Grant No. P41RR0954).
Birman, X. Li, Org. Lett. 2006, 8, 1351; d) V. B.
Birman, L. Guo, Org. Lett. 2006, 8, 4859.
[4] V. B. Birman, H. Jiang, X. Li, L. Guo, E. W. Uffman, J.
Am. Chem. Soc. 2006, 128, 6536.
[5] See, e.g.: a) F. R. Dietz, H. Grçger, Synlett 2008, 663;
b) I. Shiina, K. Nakata, Y. S. Onda, Eur. J. Org. Chem.
2008, 35, 5887; c) V. C. Purohit, A. S. Matla, D. Romo,
J. Am. Chem. Soc. 2008, 130, 10478; d) Q. Xu, H.
Zhou, X. Geng, P. Chen, Tetrahedron 2009, 65, 2232.
[6] a) V. B. Birman, X. Li, Org. Lett. 2008, 10, 1115; b) X.
Yang, V. B. Birman, Adv. Synth. Catal. 2009, 351, 2301.
[7] Y. Hayashi, W. Tsubo, I. Ashimine, T. Urushima, M.
Shoji, K. Sakai, Angew. Chem. 2003, 115, 3805; Angew.
Chem. Int. Ed. 2003, 42, 3677.
[8] J. M. M. Verkade, L. J. C. van Hemert, P. J. L. M.
Quaedflieg, P. L. Alsters, F. L. van Delft, F. P. J. T.
Rutjes, Tetrahedron Lett. 2006, 47, 8109.
[9] J. L. Vicario, D. Badꢁa, L. Carrillo, Org. Lett. 2001, 3,
773.
[10] a) J. W. Yang, M. Stadler, B. List, Angew. Chem. 2007,
119, 615; Angew. Chem. Int. Ed. 2007, 46, 609; b) This
variant of the asymmetric Mannich reaction is also ap-
plicable to the synthesis of HBTM 6 and HBTM-2 9a.
See also: J. W. Yang, C. Chandler, M. Stadler, D.
Kampen, B. List, Nature 2008, 452, 453.
References
[1] For a recent review, see: S. E. Denmark, G. L. Beutner
Angew. Chem. 2008, 120, 1584; Angew. Chem. Int. Ed.
2008, 47, 1560.
[2] For principles of kinetic resolution, see: a) H. B.
Kagan, J. C. Fiaud, Top. Stereochem. 1988, 18, 249. For
a review of non-enzymatic kinetic resolution, see: b) E.
Vedejs, M. Jure, Angew. Chem. 2005, 117, 4040; Angew.
Chem. Int. Ed. 2005, 44, 3974 and Ref.[1] For leading
references to selected contributions by other groups,
see: c) E. Vedejs, O. Daugulis, J. Am. Chem. Soc. 2003,
125, 4166; d) G. C. Fu, Acc. Chem. Res. 2004, 37, 542;
e) S. J. Miller, Acc. Chem. Res. 2004, 37, 601; f) T.
Kano, K. Sasaki, K. Maruoka, Org. Lett. 2005, 7, 1347;
g) D. Terakado, H, Koutaka, T. Oriyama, Tetrahedron:
Asymmetry 2005, 16, 1157; h) Y. Kosugi, M. Akakura,
K. Ishihara, Tetrahedron 2007, 63, 6191.
[3] a) V. B. Birman, E. W. Uffman, H. Jiang, X. Li, C. J.
Kilbane, J. Am. Chem. Soc. 2004, 126, 12226; b) V. B.
Birman, H. Jiang, Org. Lett. 2005, 7, 3445; c) V. B.
[11] V. B. Birman, H. Jiang, X. Li, Org. Lett. 2007, 9, 3237.
Adv. Synth. Catal. 2009, 351, 2525 – 2529
ꢀ 2009 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
asc.wiley-vch.de
2529