Homobenzotetramisole: An effective catalyst for kinetic resolution of aryl-cycloalkanols

Article abstract of DOI:10.1021/ol703119n

Homobenzotetramisole (HBTM), a ring-expanded analogue of the previously reported catalyst BTM, displays higher catalytic activity and a different structure-selectivity profile. It displays good enantioselectivities in kinetic resolution of secondary benzylic alcohols but is particularly effective for 2-aryl-substituted cycloalkanols.

Full text of DOI:10.1021/ol703119n

ORGANIC
LETTERS
2008
Vol. 10, No. 6
1115-1118
Homobenzotetramisole: An Effective
Catalyst for Kinetic Resolution of
Aryl-Cycloalkanols
Vladimir B. Birman* and Ximin Li
Department of Chemistry, Washington UniVersity, Campus Box 1134,
One Brookings DriVe, St. Louis, Missouri 63130
birman@wustl.edu
Received December 28, 2007
ABSTRACT
Homobenzotetramisole (HBTM), a ring-expanded analogue of the previously reported catalyst BTM, displays higher catalytic activity and a
different structure-selectivity profile. It displays good enantioselectivities in kinetic resolution of secondary benzylic alcohols but is particularly
effective for 2-aryl-substituted cycloalkanols.
Over the past several years, we have developed a new class
of enantioselective acyl transfer catalysts 1-4 represented
in Figure 1.1 These compounds have proved to be effective
nones (5-8) (Figure 2). Throughout our structure optimiza-
tion studies, we have kept constant one structural feature:
Figure 2. Previously resolved classes of substrates.
Figure 1. Previously developed catalysts.
the chiral imidazoline moiety (9) (Figure 3). Likewise, all
of the substrates investigated so far have shared the same
general patternsa π-system located R- to a nucleophilic atom
(10)sand displayed the same absolute sense of enantio-
selection in kinetic resolutions. These observations lend
support to our supposition that their chiral recognition occurs
via a common mechanism involving π-π and/or cation-π
interactions, as illustrated by structure 11.
in kinetic resolution (KR)2,3 of secondary benzylic, allylic
and propargylic alcohols, and 4-aryl-substituted oxazolidi-
(1) (a) Birman, V. B.; Uffman, E. W.; Jiang, H.; Li, X.; Kilbane, C. J.
J. Am. Chem. Soc. 2004, 126, 12226. (b) Birman, V. B.; Jiang, H. Org.
Lett. 2005, 7, 3445. (c) Birman, V. B.; Li, X. Org. Lett. 2006, 8, 1351. (d)
Birman, V. B.; Jiang, H.; Li, X.; Guo, L.; Uffman, E. W. J. Am. Chem.
Soc. 2006, 128, 6536. (e) Birman, V. B.; Guo, L. Org. Lett. 2006, 8, 4859.
(2) Kagan, H. B.; Fiaud, J. C. Top. Stereochem. 1988, 18, 249.
(3) For review of nonenzymatic kinetic resolution of alcohols and
alternative catalyst designs, see: Vedejs, E.; Jure, M. Angew. Chem., Int.
Ed. 2005, 44, 3974.
Having achieved good to excellent selectivities in the
aforementioned cases, we became interested in exploring
further structural variations of our catalysts, as well as
10.1021/ol703119n CCC: $40.75
© 2008 American Chemical Society
Published on Web 02/16/2008

Scheme 1. Preparation of 15 and 16
Figure 3. Proposed general mode of chiral recognition.
identifying new types of substrates. In this communication,
we report the synthesis of a new type of enantioselective
acyl transfer catalyst and its use in kinetic resolution of cyclic
alcohols.
We have recently examined a new class of achiral
acylation catalysts, 12-14, containing a tetrahydropyrimi-
dine, rather than imidazoline, moiety4 (Figure 4). The highly
The efficacy of the new catalysts was first examined in
KR of (()-1-phenylpropanol 22, employed as the test
substrate in our previous studies (Table 1). As anticipated,
Table 1. KR of 1-phenylpropanola
entry
catalyst (mol %)
time, h
% convn
s
1
3 (10)
4 (4)
15 (1)
16 (1)
3
31
49
47
48
28
72
30
26
2b
3
2.2
1.7
1.5
4
a Conditions: 0.25 M 23, 0.75 equiv of (EtCO)2O, 0.75 equiv of i-Pr2NEt,
CDCl3, Na2SO4, rt. b Data from previous work.1c
Figure 4. Design of HTM and HBTM.
15 and 16 proved to be substantially more catalytically active
than 3 and 4 (entries 3 and 4 vs 1 and 2). Surprisingly, HTM
15 displayed essentially the same catalytic activity as HBTM
16, despite the fact that in the achiral series, THTP 13 was
several times less active than its benzannulated analogue
DHPB 14.4 The enantioselectivity of both 15 and 16,
however, was only moderate, especially by comparison with
BTM 4. We turned our attention to enantioselective acylation
of other classes of chiral alcohols. Keeping in mind that
π-interactions with aromatic rings were beneficial for chiral
recognition of the previously investigated classes of sub-
strates, we decided to examine kinetic resolution of trans-
phenylcyclohexanol 24, in which the phenyl group is two
carbon atoms away from the hydroxyl (Table 2).8,9
active catalyst 13 developed by our group and 14 discovered
at the same time by Okamoto and Kobayashi5 appeared to
be especially suitable as leads for designing their chiral
analogues. We were especially interested in examining their
2-phenyl-substituted derivatives, 15 and 16 (Scheme 1),
which might be viewed as ring-expanded versions of catalysts
3 and 4. Accordingly, the new structures were dubbed HTM
(for HomoTetraMisole) and HBTM (for HomoBenzoTetra-
Misole), respectively.
Chiral γ-aminoalcohol 17 was prepared in enantiopure
form according to a literature procedure.6 Its condensation
with 18 afforded a moderate yield of aminothiazoline
derivative 19,7 which was cyclized with thionyl chloride to
produce HTM 15. Preparation of HBTM 16 was ac-
complished using the protocol previously developed for BTM
4.1c
Tetramisole 3 produced only modest enantioselectivity and
apparently underwent deactivation, so that low conversion
(8) For enzymatic kinetic resolution of 24 and its utility as a chiral
auxiliary, see: Whitesell, J. K.; Chen, H.-H.; Lawrence, R. M. J. Org. Chem.
1985, 50, 4663.
(9) Efficient nonenzymatic kinetic resolution of 24 has been previously
reported by: (a) Oriyama, T.; Hori, Y.; Imai, K.; Sasaki, R. Tetrahedron
Lett. 1996, 37, 8543 (s ) 200). (b) Copeland, G. T.; Miller, S. J. J. Am.
Chem. Soc. 2001, 123, 6496 (s > 50). (c) Jeong, K.-S.; Kim, S.-H.; Park,
H.-J.; Chang, K.-J.; Kim, K. S. Chem. Lett. 2002, 1114 (s ) 21).
(4) Birman, V. B.; Li, X.; Han, Z. Org. Lett. 2007, 9, 37.
(5) Kobayashi, M.; Okamoto, S. Tetrahedron Lett. 2006, 47, 4347.
(6) Liu, S.; Mu¨ller, J. F. K.; Neuburger, M.; Schaffner, S.; Zehnder, M.
HelV. Chim. Acta. 2000, 83, 1256.
(7) McKay, A. F.; Whittingham, D. J.; Kreling, M.-E. J. Am. Chem. Soc.
1958, 80, 3339.
1116
Org. Lett., Vol. 10, No. 6, 2008

Table 2. KR of (()-trans-2-Phenylcyclohexanol 25a
Table 4. Kinetic Resolutions with HBTM 16
entry catalyst (mol %) anhydride time, h % convn
s
1
2
3
4
5
6
7
3 (20)
4 (8)
2 (8)
15 (2)
16 (2)
16 (2)
16 (2)
(EtCO)2O
(EtCO)2O
(EtCO)2O
(EtCO)2O
(EtCO)2O
(MeCO)2O
(i-PrCO)2O
3
6
12
1.8
1.4
0.5
19
48
45
43
47
59
39
5.5
25
11
14
29
19
26
48
a Conditions: 0.25 M substrate, 0.75 equiv of (EtCO)2O, 0.75 equiv of
i-Pr2NEt, CDCl3, rt.
was obtained even at a 20 mol % catalyst loading (entry 1).
BTM 4 produced a synthetically useful selectivity factor with
this substrate; however, the reaction was quite slow, neces-
sitating a high catalyst loading (entry 2). Cl-PIQ 2, our most
active catalyst for asymmetric acylation of benzylic alcohols,
turned out to be rather slow in the case of substrate 24 and
displayed only moderate levels of enantioselectivity (entry
3). In contrast, acylations of 24 catalyzed by both HTM 15
and HBTM 16 proceeded rapidly at 2 mol % loadings (entries
4 and 5). In addition, HBTM produced the highest selectivity
factor and thus was selected for further optimization. We
tested the suitability of acetic and isobutyric anhydrides as
achiral acyl donors (entries 5 and 6), only to confirm that
propionic anhydride was indeed the best. Variation of the
reaction temperature was examined next (Table 3, entries
Table 3. Optimization of Reaction Conditionsa
entry mol % of 16 solvent temp, °C time, h % convn
s
1
2
3
4
5
6
7
8
2
2
2
4
4
2
2
2
2
2
2
4
CDCl3
CDCl3
CDCl3
CDCl3
CDCl3
MeCN
CH2Cl2
THF
23
0
-20
-40
-55
23
23
23
23
23
1.4
2.3
3
4
4
2.5
1.8
1.8
1
0.7
1.7
3
47
52
53
53
46
48
47
50
54
56
54
46
29
43
53
86
122
15
23
25
30
41
9
PhMe
TA
TA
10
11
12
-10
-40
72
101
a Absolute configuration of the fast-reacting enantiomer is shown
b Conditions: 0.25 M substrate, 0.75 equiv of (EtCO)2O, 0.75 equiv of
i-Pr2NEt, 2 mol % HBTM, Na2SO4, CDCl3, room temp c Conditions: 0.25
M substrate, 0.55 equiv of (EtCO)2O, 0.55 equiv of i-Pr2NEt, 4 mol %
HBTM, Na2SO4, 1:1 tert-amyl alcohol/toluene, -40 °C d Conditions: 0.25
M substrate, 0.75 equiv of (EtCO)2O, 0.75 equiv of i-Pr2NEt, 4 mol %
HBTM, Na2SO4, tert-amyl alcohol, -10 °C e Only 0.1 M of the substrate
was used due to its poor solubility f Not Determined due to the poor
solubility of the substrate g Two molar percent of the catalyst was used.
TA-PhMe
a Conditions: 0.25 M substrate, 0.75 equiv of (EtCO)2O, 0.75 equiv of
i-Pr2NEt.
1-5). We were pleased to find that, in contrast to BTM,
which was rarely effective below 0 °C, HBTM produced
convenient reaction rates at temperatures as low as -55 °C,
which allowed us to obtain a 4-fold increase in the enantio-
selectivity (entries 5 vs 1). Interestingly, the decrease in the
reaction rate was not as significant as we had anticipated.
Similar observations were previously made by Vedejs et al.10
Org. Lett., Vol. 10, No. 6, 2008
1117

Screening various solvents (entries 6-10) revealed that,
unlike all our previous catalysts, which were usually suc-
cessful only in chloroform, HBTM was more tolerant of the
reaction medium. In fact, tert-amyl alcohol (TA) proved to
be clearly superior to chloroform at room temperature.11
Because of its high freezing point (-12 °C), however, it
could not be used at temperatures below -10 °C. Therefore,
a 1:1 mixture of toluene and tert-amyl alcohol, which could
be cooled without freezing to -40 °C (acetonitrile-dry ice
bath), was selected as a compromise.
Kinetic resolution of several structurally different sub-
strates was studied next at room temperature in deuterated
chloroform. On the basis of the enantioselectivities and
reaction rates observed during this initial screening, we
subsequently repeated some of these reactions at lower
temperatures in tert-amyl alcohol or its mixture with toluene
and obtained improved selectivity factors. Cyclohexanols
bearing a trans-aryl or -heteroaryl group at C2 displayed
good to excellent levels of enantioselectivity in HBTM-
catalyzed acylations (Table 4, entries 1-3). The analogous
trans-phenylcyclopentanol was also resolved with high
selectivity (entry 4). Cis-isomer of phenylcyclohexanol 29
reacted rather slowly, but also reached a respectable selectiv-
ity factor of 28 (entry 5). On the other hand, cyclohexanols
containing nonaromatic substituents produced only moderate
selectivity factors (entries 6-9). In particular, menthol 33
reacted slowly and with very low selectivity (entry 9). These
results suggested that the presence of a π-system is necessary
for effective chiral recognition. The rigidity of the substrate
molecule also appears to be important, as suggested by the
modest selectivity factor achieved in KR of acyclic alcohol
34, a conformationally flexible analogue of trans- and cis-
phenylcyclohexanols (entry 10).12 As expected, conducting
KR of benzylic alcohols 22 and 35 at low temperatures
produced high selectivities (entries 11 and 12), although these
results were still below those achieved with BTM in our
earlier work.1c
In conclusion, we have demonstrated that 2-aryl-cyclo-
alkanols can be efficiently resolved using our new catalyst,
HBTM. A seemingly straightforward modification in the
structure of BTMsexpansion of the imidazoline ring by one
carbonshas led not only to an increased catalytic activity,
but also to a significantly altered structure-selectivity profile.
HBTM affords useful levels of selectivity in KR of benzylic
alcohols but is particularly effective in KR of aryl-cyclo-
alkanols, whereas BTM displays the opposite trend. Further
investigation of the new type of enantioselective acylation
catalysts and their applications is underway.
Acknowledgment. We thank NIGMS (NIH R01
GM072682) for financial support of this work. Mass
spectrometry was provided by the Washington University
Mass Spectrometry Resource, an NIH Research Resource
(Grant No. P41RR0954).
Supporting Information Available: Experimental pro-
cedures and 1H and 13C NMR spectra of compounds. These
materials are available free of charge via the Internet at
http://pubs.acs.org.
OL703119N
(10) Vedejs, E.; and Daugulis, O. J. Am. Chem. Soc. 2003, 125, 4166.
(11) For use of tert-amyl alcohol as a solvent for KR, see, e.g.: Ruble,
J. C.; Tweddell, J.; Fu, G. C. J. Org. Chem. 1998, 63, 2794.
(12) Interestingly, BTM produced higher enantioselectivity in this case,
although the reaction was quite slow (s ) 9.9 at 15% conversion after 8 h
at rt; 8 mol % catalyst loading).
1118
Org. Lett., Vol. 10, No. 6, 2008