Influence of electronic and steric factors on 2,3-dihydroimidazo[1,2-a] pyridine-based enantioselective acylation catalysts

Article abstract of DOI:10.1016/j.tet.2005.08.119

The catalytic activity and enantioselectivity of chiral 2,3-dihydroimidazo[1,2-a]pyridine (DHIP) derivatives was examined as a function of the steric environment of the nucleophilic nitrogen and the electronic properties of the pyridine ring. 2-Phenyl-6-trifluoromethyl-DHIP (CF ₃-PIP) was confirmed to be the best catalyst in this series. In addition, three second-generation catalysts derived from the 1,2-dihydroimidazo[1,2-a]quinoline (DHIQ) core were tested and proved to be considerably more active than CF₃-PIP.

Full text of DOI:10.1016/j.tet.2005.08.119

Tetrahedron 62 (2006) 285–294
Influence of electronic and steric factors on
2,3-dihydroimidazo[1,2-a]pyridine-based
enantioselective acylation catalysts
*
Vladimir B. Birman, Ximin Li, Hui Jiang and Eric W. Uffman
Department of Chemistry, Washington University, Campus Box 1134, One Brookings Drive, St. Louis, Missouri 63130, USA
Received 17 May 2005; revised 4 August 2005; accepted 31 August 2005
Available online 10 October 2005
Abstract—The catalytic activity and enantioselectivity of chiral 2,3-dihydroimidazo[1,2-a]pyridine (DHIP) derivatives was examined as a
function of the steric environment of the nucleophilic nitrogen and the electronic properties of the pyridine ring. 2-Phenyl-6-trifluoromethyl-
DHIP (CF3-PIP) was confirmed to be the best catalyst in this series. In addition, three second-generation catalysts derived from the 1,2-
dihydroimidazo[1,2-a]quinoline (DHIQ) core were tested and proved to be considerably more active than CF3-PIP.
q 2005 Elsevier Ltd. All rights reserved.
1. Introduction
Testing these compounds as catalysts in acetylation of
racemic phenyl ethyl carbinol provided the basis for their
initial comparison. We determined that introduction of the
trifluoromethyl at C6 produced not only a more reactive, but
also a more enantioselective catalyst (7c) than either the
stronger (NO2) or the weaker (Br, H) electron-withdrawing
groups (Table 1).
The challenge of developing easily accessible and effective
non-enzymatic asymmetric acylation catalysts has attracted
many research groups around the world over the last
decade.1,2 Two years ago, we envisioned a new class of
enantioselective nucleophilic catalysts based on dihydro-
imidazo[1,2-a]pyridine (DHIP)3 (1, RZXZH).1n Although
no prior reports of the catalytic activity of this heterocyclic
compound were to be found in the literature, we recognized
its practical potential as a core structure for rational catalyst
design. Chiral derivatives of DHIP (1, RsH) could be
obtained from chiral 2-amino alcohols (4) and substituted
2-halopyridines (3) in only two steps (Scheme 1). The
commercial availability of a variety of these starting
materials was expected to greatly facilitate the variation of
the steric environment of the nucleophilic nitrogen and the
tuning of the electronic properties of the system in search of
the optimal catalytic activity and enantioselectivity.
Optimization of the reaction conditions (use of N,N-
diisopropylethylamine as auxiliary base and propionic
anhydride as the stoichiometric acyl donor and lowering
the reaction temperature to 0 8C) allowed us to obtain useful
levels of selectivity (20–85) in kinetic resolution of a range
of acyclic secondary benzylic alcohols with only 2 mol%
catalyst loadings of CF3-PIP. The most important selectiv-
ity/reactivity trends are illustrated with representative
examples 14–18 (Fig. 1). The absolute sense of
Table 1. The first catalytic activity/selectivity test
After establishing the catalytic activity of the parent
compound, we proceeded to investigate the efficacy of its
chiral derivatives in kinetic resolution.4 2-Phenyl-DHIP 7a
(abbreviated H-PIP) and three of its analogs bearing
electron-withdrawing groups at C6 (6-Br-PIP 7b, 6-CF3-
PIP 7c and 6-NO2-PIP 7d) were prepared from (R)-
phenylglycinol (5), which happened to be the most abundant
chiral 2-amino alcohol found in our laboratory (Scheme 2).
a
a
X
% eeE
% eeA
% Conversiona
Selectivitya
H
Br
CF3
NO2
49
74
80
82
13
25
49
13
21
25
38
11
3.3
8.6
14
Keywords: Organocatalysis; Non-enzymatic asymmetric acylation cata-
lysts; Enantioselective acyl transfer; Kinetic resolution.
11
*
Corresponding author. Tel.: C1 314 935 9188; fax: C1 314 935 4481;
e-mail: birman@wustl.edu
a Averaged results of duplicate runs.
0040–4020/$ - see front matter q 2005 Elsevier Ltd. All rights reserved.
doi:10.1016/j.tet.2005.08.119

286
V. B. Birman et al. / Tetrahedron 62 (2006) 285–294
Scheme 1.
Scheme 2.
Figure 1. Kinetic resolution of secondary benzylic alcohols using CF3-PIP.
enantioselectivity observed in the kinetic resolution of
substrates 14–16 is consistent with our proposed transition
state model 13, which is based on p–p5 and/or cation-p6
interactions of the aromatic ring of the substrate with the
pyridinium moiety of the acylated catalyst. This model also
explains why the cyclic benzylic alcohol, 1-indanol 17,
which cannot adopt the required conformation, is acylated
extremely slowly and with no detectable selectivity, while
cyclohexyl methyl carbinol 18, lacking an aromatic ring,
simply fails to react to any appreciable extent.
2. Results and discussion
2.1. The influence of substitution at C2 on the
enantioselectivity
To answer the first of these questions, we prepared the
2-benzyl (21a), 2-isopropyl (21b) and 2-t-butyl (21c)
analogs of CF3-PIP as well as the conformationally
restricted version of the latter, tetracycle 21d, starting
with the commercially available chiral 2-amino alcohols
(R)-phenylalaninol 19a, (R)-valinol 19b, (S)-tert-leucinol
19c and (1R,2S)-1-amino-2-indanol 19d, respectively
(Scheme 3).
The success of this initial study prompted us to seek answers
to the following two questions: (a) was our fortuitous choice
of the phenyl group as the C2-substituent critical for the
enantioselectivity? and (b) could we further improve the
catalytic activity of our system by changing the substitution
of the pyridine ring?
Testing these catalysts under the conditions optimized for
CF3-PIP1n demonstrated how fortunate we were to have
started our investigation with phenylglycinol (Table 2). 21a

V. B. Birman et al. / Tetrahedron 62 (2006) 285–294
287
Scheme 3.
and 21b displayed catalytic activities comparable to CF3-
PIP, but drastically reduced enantioselectivities. The t-butyl
derivative 21c was even less selective and reacted extremely
slowly, which ostensibly reflected the steric encumbrance of
the nucleophilic nitrogen. Finally, the indane derivative 21d
reacted noticeably faster than CF3-PIP and with barely
detectable selectivity, suggesting that the two faces of the
catalyst were no longer effectively discriminated. It is
noteworthy that the low enantioselectivity seen in this case
was opposite to that with the rest of the catalysts having
(R)-configuration at C2 (7c, 21a,b), which also suggests a
change in the transition state. The same sense of chiral
recognition was previously observed in the kinetic
resolution of a-phenethylamine with an indane-based
stoichiometric enantioselective acyl donor structurally
similar to 21d.7
which the plane of the ring is almost perpendicular to the
C2–2 bond) is very different from that in the conformation-
ally restricted analog 21d. Currently, we surmise that the
role of the phenyl group consists not only in blocking the
approach of the nucleophile from one of the faces, but also
in forcing the acyl group to adopt the conformation required
for high selectivity. It will be of interest to examine
modification of the phenyl ring, particularly by introducing
ortho-substituents. However, substituted phenylglycinols
are not commercially available and therefore, the prep-
aration of such catalysts must await further study.
2.2. The influence of substitution of the pyridine ring
Since we had obtained the best results with a substituent of
intermediate electron-withdrawing strength—trifluoro-
methyl—we decided to synthesize and test two additional
catalysts, having substituents with a somewhat lower and a
somewhat higher Hammett spara values.8,9 For practical
reasons, we chose to prepare the 6-diethylcarbamoyl and
6-cyano derivatives CONEt2-PIP (7e) and CN-PIP (7f). We
also wanted to verify our initial assumption that position
ortho- to the nucleophilic nitrogen must remain unsub-
stituted. The commercially available 2-chloro-3-fluoropyri-
dine afforded 8-F-PIP (7g) having the smallest possible non-
hydrogen substituent at position 8. 6,8-Br2-PIP (7h) was
readily available by a simple modification of the procedure
used to prepare Br-PIP (7b)1n,10 (Scheme 4).
The above results indicate that the selectivity critically
depends not only on the nature of the C2 substituent, but
also on its orientation. Clearly, the thermodynamically
favored conformation of the benzene ring in CF3-PIP (in
Table 2.
a
a
In addition, we had recently begun to explore the second
generation of asymmetric acylation catalysts based on the
structure of 1,2-dihydroimidazo[1,2-a]quinoline (DHIQ)
25, which is simply the 5,6-benzannellated analog of the
DHIP core. Chiral derivatives of DHIQ are just as easily
accessible as those of DHIP, via the standard two-step
procedure starting with 2-haloquinolines instead of 2-halo-
pyridines.11 Thus, the first three members of the series—
H-PIQ 27a, 7-Cl-PIQ 27b, and 5-Me-PIQ 27c were
prepared from the commercially available 2-chloroqui-
noline 25a, 2,6-dichloroquinoline 25b and 2-chloro-4-
methylquinoline 25c (Scheme 5).
Catalyst
% eeE
% eeA
% Conversiona
Selectivitya,b
7c
91
56
51
K31
K14
59
34
23
39
38
31
7d
36
4.9
3.9
K2.0
K1.6
21a
21b
21c
21d
c
—
K19
58
a Averaged results of duplicate runs.
b Catalysts 7c, 21a and 21b produced the (R)-ester and the (S)-alcohol (the
absolute configurations shown). Catalysts 21c and 21d gave the opposite
selectivity.
c Too low for reliable HPLC measurements.
d Estimated from the 1H NMR spectra of the crude reaction mixtures and
used to calculate the selectivity factor.4

288
V. B. Birman et al. / Tetrahedron 62 (2006) 285–294
Scheme 4.
Scheme 5.
As mentioned above, the first catalyst performance test was
carried out using acetic anhydride and without any auxiliary
base. Since we already knew that addition of i-Pr2NEt led to
dramatic rate acceleration and propionic anhydride pro-
duced considerably higher selectivities than acetic anhy-
dride, we modified the original set of conditions accordingly
(Table 3). Furthermore, since the catalysts being studied this
time were expected to differ by several orders of magnitude
measured reliably, and the measurements were repeated at
0.01 M catalyst concentrations. The results of these
determinations are shown in the last column of Table 3.
The most practically important conclusion that may be
drawn from these data is that the new generation of catalysts
based on the DHIQ core is considerably more active than
even the best catalyst in the DHIP series. The somewhat
higher enantioselectivity of Cl-PIQ compared with CF3-PIP
observed in this trial is consistent with our recent study.11
1
in their activity, we monitored each reaction by H NMR
spectroscopy and stopped it upon reaching approximately
the same conversion (45–50%), rather than after a fixed
period of time.
The status of CF3-PIP as the most active and selective DHIP
catalyst has remained unchallenged, with the two new
6-substituted contenders, CONEt2-PIP and CN-PIP, falling
far behind. Br-PIP, on the other hand, displayed selectivity
comparable to CF3-PIP and also proved to be the second
best in the DHIP series in terms of catalytic activity. It is
noteworthy that CONEt2-PIP is less active and selective
than Br-PIP, although the spara value assigned to the amide
group (0.36) is intermediate between that of Br (0.23) and
CF3 (0.54).9 The order of the respective smeta values (0.35,
0.39, and 0.43)9 is qualitatively consistent with the
performances of these three catalysts in this trial. It remains
to be seen whether the effect of other substituents on the
In addition, we sought to obtain ‘fair’ comparison of the
catalytic activities. Since the conversions observed during
kinetic resolution are dependent on the enantioselectivities
and therefore, can serve only as an approximate measure of
activity, we also determined the half-life of the fast-reacting
(R)-enantiomer of phenyl ethyl carbinol with each catalyst
under the conditions identical to those described above,
except that 0.5 M concentration of (R)-11 was used instead
of 1.0 M (G)-11. In the case of the most reactive catalysts—
CF3-PIP and especially the three DHIQ derivatives—the
half-lives with 0.05 M catalyst were too short to be

V. B. Birman et al. / Tetrahedron 62 (2006) 285–294
289
Table 3.
1
t ⁄ (R)
#
Catalyst
Time
% eeE
% eeA
% CHPLC
S
2
7a
7b
7c
7d
7e
7f
7g
7h
28a
28b
28c
H-PIP
11 h
3 h
40 min
146 h
4 h
14 h
46 h
120 h
15 min
15 min
27 min
73
83
85
30
77
74
57
56
81
84
84
65
79
77
26
74
72
45
46
89
88
82
47
49
47
46
49
49
48
45
52
51
50
12.5
25
28
2.4
17
14
5.6
5.5
28
33
29
2 ha
6-Br-PIP
6-CF3-PIP
6-NO2-PIP
6-CONEt2-PIP
6-CN-PIP
8-F-PIP
6,8-Br2-PIP
H-PIQ
7-Cl-PIQ
5-Me-PIQ
22 mina
4.5 min,a 45 minb
4 daysa
34 mina
96 mina
11.5 ha
4 daysa
!2 min,a 10.5 minb
!2 min,a 12 minb
z2 min,a 17 minb
a Measured at 0.05 M catalyst concentration.
b Measured at 0.01 M catalyst concentration.
catalysts’ performance can be reliably predicted from
Hammett’s parameters. At this point, however, it may be
noted that both the trifluoromethyl and the bromine—the
two most successful groups—exert purely inductive
electron-withdrawing effect, in contrast to the rest of the
C6-substituents tested.
3. Conclusions
We have examined the influence of several steric and
electronic factors on the activity and enantioselectivity of a
series of chiral catalysts derived from the DHIP core. The
currently available data indicate the following:
(1) A conformationally unrestricted phenyl group at C2
provides by far the highest levels of enantioselectivity;
(2) Substitution at C8 is highly detrimental to both the
catalytic activity and the enantioselectivity of DHIP
catalysts;
(3) Trifluoromethyl is the optimal C6-substituent in the
DHIP series, bromine being the second best;
(4) All three DHIQ-based catalysts examined are consider-
ably more active than CF3-PIP. Cl-PIQ is also more
selective than the latter.
As expected, the presence of a C8-substitutuent (cf. 8-F-PIP
and 6,8-Br2-PIP) proved to be highly detrimental to the
catalytic activity and selectivity, presumably due to its
interference with the acyl carbonyl in the acylated
intermediate. In the case of the small fluoro substituent, it
is not clear whether this effect is primarily due to the steric
or the electronic repulsion with the carbonyl oxygen.
Perhaps it is more surprising that 6,8-Br2-PIP reacted at
all, even producing modest enantioselectivity. Monitoring
the reaction by NMR spectroscopy provided a possible
explanation of this result. Virtually no reaction was
observed during the first 10 h, but later the product
gradually began to appear at an apparently increased rate.
We speculate that in this case the catalytically incompetent
6,8-Br2-PIP was slowly transformed into an active (and
enantioselective) species, which catalyzed the acylation.
The exact nature of this compound and the mechanism of its
formation cannot be ascertained at this point, but we surmise
that it may be a DHIP derivative lacking the bromine at C8.
In conclusion, CF3-PIP is likely to remain the best among
the first-generation DHIP-based catalysts—at least those
that can be conveniently prepared from commercially
available starting materials. On the other hand, the second
generation of enantioselective catalysts, based on the DHIQ
core and represented so far by only three members, already
offers significant improvement over the DHIP catalysts.
There is little doubt that the structure-selectivity and
structure-activity trends originally discovered and studied
in the DHIP series will be applicable to DHIQ derivatives
and thus will prove quite valuable for further optimization
of our catalyst design.
Mention must also be made here of the low chemical
stability of NO2-PIP and, to a lesser extent, CN-PIP under
the reaction conditions. In the acylations catalyzed by these
compounds, decomposition was evident from the red colour
and some precipitation developing in the course of the
reaction. In the case of NO2-PIP, the discoloration was
observed even during the first trial conducted under
relatively neutral conditions (see Table 1). Apparently, the
presence of Hunig’s base in the present experiment further
facilitated the decomposition, which accounts for the very
slow reaction and the low enantioselectivity. This last result
is in sharp contrast to that obtained under the base-free
conditions, wherein NO2-PIP displayed selectivity compar-
able to CF3-PIP.1m
4. Experimental
4.1. General methods
Preparation of catalysts 7a–d and 28b, methods of
compound characterization, determination of ee by chiral
HPLC and calculation of the selectivity factors and
conversion in kinetic resolution have been described.1n,11
(R)-enantiomer of phenyl ethyl carbinol was prepared by
kinetic resolution of the racemic alcohol (on 5.0 g scale)

290
V. B. Birman et al. / Tetrahedron 62 (2006) 285–294
using 2 mol% of (S)-Cl-PIQ and was enantiomerically pure
within the limits of HPLC detection (R99.9% ee). Other
reagents were obtained from commercial sources and used
as received.
4.2.3. (R)-N-(5-trifluoromethylpyridyl-2)-valinol (20b).
Standard Procedure A was followed [237 mg (1.31 mmol)
of 9, 122 mg (1.18 mmol) of R-valinol, 0.185 g i-Pr2NEt;
105G5 8C for 2 days; 10% i-PrOH/hexanes] to afford
188 mg of colorless oil. (64% yield).
4.2. Preparation of the catalysts
1H NMR (300 MHz, CDCl3) d 8.25 (apparent s, 1H), 7.56
(dd, J1Z8.9 Hz, J2Z2.3 Hz, 1H), 6.51 (d, JZ8.9 Hz, 1H),
5.10 (br d; JZ5.8 Hz, 1H), 3.85–3.66 (m, 3H), 3.14 (s, 1H),
1.96 (octet; JZ6.7 Hz, 1H), 1.00 (d, JZ6.7 Hz, 3H), 0.99
(d, JZ6.7 Hz, 3H); 13C NMR (75 MHz, CDCl3) d 160.8,
145.4, 134.4, 124.4 (q, JZ270.2 Hz), 115.3 (q, JZ32.4 Hz),
107.3, 63.7, 59.3, 29.7, 19.2, 18.7; IR (film, cmK1): 3434 (br),
1640; MS: HR-FAB calcd for C11H15F3N2OLi (MCLiC)
m/z: 255.1297, found: 255.1298; [a]D C48 (c 1.77, MeOH).
4.2.1. (R)-N-(5-trifluoromethylpyridyl-2)-phenylalaninol
(20a). Standard procedure A: a 25-mL pressure tube
charged with 2-chloro-5-trifluoromethylpyridine 9 (493 mg,
2.72 mmol), (R)-(C)-phenylalaninol (345 mg, 2.28 mmol),
i-Pr2NEt (334 mg, 2.58 mmol) and a stir bar was flushed
with nitrogen several times, stoppered and heated at 120G
5 8C for 3 days. The tube was allowed to cool to room
temperature, the contents was diluted with a small amount
of CH2Cl2 and chromatographed (8% i-PrOH/hexanes) to
afford 563 mg of white fluffy solid (84% yield).
4.2.4. (R)-2-isopropyl-6-trifluoromethyl-2, 3-dihydro-
imidazo[1,2-a]pyridine (21b). Standard procedure B was
followed [93 mg (0.38 mmol) 20b, 0.08 mL SOCl2, 2 mL of
CHCl3; 20% i-PrOH, 2% NEt3/hexanes] to afford 45 mg of
product (52% yield).
1H NMR (300 MHz, CDCl3): d 8.28 (s, 1H), 7.54 (dd, J1Z
8.8 Hz, J2Z2.2 Hz, 1H), 7.21–7.34 (m, 5H), 6.40 (d, JZ
8.8 Hz, 1H), 5.23 (s, 1H), 4.18–4.23 (m, 1H), 3.83 (dd,
J1Z11.0 Hz, J2Z3.3 Hz, 1H), 3.69 (dd, J1Z11.0 Hz, J2Z
6.0 Hz, 1H), 2.99 (dd, J1Z13.9 Hz, J2Z6.6 Hz, 1H), 2.91
(dd, J1Z13.9 Hz, J2Z7.7 Hz, 1H), 2.71 (s, br, 1H); 13C
NMR (75 MHz, CDCl3): d 160.1, 145.6 (q, JZ4.5 Hz),
137.9, 134.6 (q, JZ3.0 Hz), 129.3, 128.8, 126.8, 124.6 (q,
JZ270 Hz), 115.7 (q, JZ33.2 Hz), 107.8, 64.4, 55.0, 37.5;
IR (film, cmK1): 3409 br, 1617, 1324; MS: HR-FAB calcd
for C15H15F3N2OLi (MCLiC) m/z: 303.1297, found:
303.1302; mp 92.5–94.5 8C; [a]D C37.4 (c 1.07, MeOH).
1H NMR (300 MHz, CDCl3) d 7.31 (apparent s; 1H), 6.87
(dd, J1Z9.7 Hz, J2Z2.0 Hz, 1H), 6.40 (d, JZ9.7 Hz, 1H),
4.05–3.94 (m, 2H), 3.69 (dd, J1Z16.2 Hz, J2Z13.7 Hz,
1H), 1.79 (octet; JZ6.6 Hz, 1H), 1.00 (d, JZ6.6 Hz, 3H),
0.91 (d, JZ6.6 Hz, 3H); 13C NMR (75 MHz, CDCl3) d
155.5, 133.6 (q, JZ3.7 Hz), 13.8, 123.7 (q, JZ268.5 Hz),
115.2, 106.0 (q, JZ34.6 Hz), 70.5, 52.0, 33.4, 18.5, 17.9; IR
(film, cmK1): 1664, 1334; MS: HR-FAB calcd for
C11H14F3N2 (MCHC) m/z: 231.1109, found: 231.1116;
mp 75–76.5 8C; [a]D C63 (c 2.0, MeOH).
4.2.2. (R)-2-benzyl-6-trifluoromethyl-2, 3-dihydro-
imidazo[1,2-a]pyridine (21a). Standard procedure B: a
solution of 20a (782 mg, 2.6 mmol) in 15 mL of CHCl3 was
treated with SOCl2 (0.55 mL, 7.5 mmol) added dropwise at
room temperature, then heated to reflux in an oil bath kept at
65 8C. After 1.5 h, the flask was taken out of the oil bath,
allowed to cool somewhat and treated cautiously with 2–3
drops of MeOH (vigorous gas evolution!), then heated again
for 5 min. The mixture was rotary evaporated and the
evaporation residue was extracted with water. The aqueous
extract was decanted from the gummy residue, brought to
pH 7–8 with aqueous NaHCO3 and extracted with CH2Cl2
(three times). More aqueous NaHCO3/NaOH was added to
the aqueous phase to pH 12 and extraction was continued
until organic extracts were pale-yellow (four times). The
organic phase was dried over Na2SO4 and then rotary
evaporated. The crude mixture was chromatographed (20%
i-PrOH, 2% NEt3/hexanes) to give 545 mg of yellow solid
(74% yield).
4.2.5. (R)-N-(5-trifluoromethylpyridyl-2)-tert-leucinol
(20c). Standard procedure A was followed [403 mg
(2.22 mmol) of 9, 243 mg (2.07 mmol) of (S)-tert-leucinol,
330 mg i-Pr2NEt; 120G5 8C for 3 days; 7% i-PrOH/
hexanes] to afford 277 mg of white fluffy solid (51% yield).
1H NMR (300 MHz, CDCl3) d 8.26 (s, 1H), 7.59 (dd,
J1Z8.8 Hz, J2Z2.5 Hz, 1H), 6.57 (d, JZ8.8 Hz, 1H), 5.16
(d, JZ7.4 Hz, 1H), 3.98 (dd, J1Z11.0 Hz, J2Z3.3 Hz, 1H),
3.84 (ddd, J1Z8.5 Hz, J2Z7.4 Hz, J3Z3.3 Hz, 1H), 3.60
(dd, J1Z11.0 Hz, J2Z8.5 Hz, 1H), 2.58 (s, br, 1H), 1.01 (s,
9H); 13C NMR (75 MHz, CDCl3) d 161.6, 145.6 (q,
JZ4.3 Hz), 134.7 (q, JZ3.2 Hz), 124.6 (q, JZ270 Hz),
115.5 (q, JZ33.2 Hz), 107.2, 63.4, 62.4, 34.5, 27.1; IR (film,
cmK1): 3329 (br), 1618, 1325; MS: HR-FAB calcd for
C12H17F3N2OLi (MCLiC) m/z: 269.1453, found:
269.1445; mp 109.5–110.5 8C; [a]D C15.0 (c 1.10, MeOH).
4.2.6. (S)-2-tert-butyl-6-trifluoromethyl-2, 3-dihydro-
imidazo[1,2-a]pyridine (21c). Standard procedure B was
followed [101 mg (0.385 mmol) of 20c, 0.071 mL SOCl2,
3 mL CHCl3; 5% i-PrOH, 0.5% NEt3/hexanes] to afford
82 mg of yellow crystalline product (86% yield).
1H NMR (300 MHz, CDCl3) d 7.21–7.34 (m, 6H), 6.92 (dd,
J1Z9.8 Hz, J2Z1.9 Hz, 1H), 6.45 (d, JZ9.8 Hz, 1H),
4.47–4.58 (m, 1H), 3.94 (t, JZ11.0 Hz, 1H), 3.73 (dd, J1Z
11.0 Hz, J2Z8.0 Hz, 1H), 3.21 (dd, J1Z13.7 Hz, J2Z
5.0 Hz, 1H), 2.71 (dd, J1Z13.7 Hz, J2Z9.1 Hz, 1H); 13C
NMR (75 MHz, CDCl3) d 155.8, 137.8, 133.8 (q, JZ
4.3 Hz), 132.0, 129.2, 128.5, 126.5, 123.8 (q, JZ268 Hz),
115.3, 106.1 (q, JZ34.6 Hz), 65.8, 53.7, 42.7; IR (film,
cmK1): 1666, 1335, 1283, 1150, 1105; MS: HR-FAB calcd
for C15H13F3N2 (MCHC) m/z: 279.1109, found: 279.1106;
mp 122.5–123.5 8C; [a]D C169 (c 1.10, MeOH).
1H NMR (300 MHz, CDCl3) d 7.30 (s, 1H), 6.84 (dd, J1Z
9.9 Hz, J2Z2.2 Hz, 1H), 6.37 (d, JZ9.9 Hz, 1H), 3.73–3.99
(m, 3H), 0.93 (s, 9H); 13C NMR (75 MHz, CDCl3) d 155.6,
133.8 (q, JZ3.7 Hz), 131.8, 124.0 (q, JZ269 Hz), 115.4,
105.9 (q, JZ34.6 Hz), 74.5, 50.5, 34.3, 25.8; IR (film, cmK1):
1665, 1340, 1327, 1281, 1150, 1106; MS: HR-FAB calcd

V. B. Birman et al. / Tetrahedron 62 (2006) 285–294
291
for C12H16F3N2 (MCHC) m/z: 245.1266, found: 245.1277;
mp 55–56 8C; [a]D K187 (c 1.01, MeOH).
121.0, 106.7, 66.4, 58.3, 25.0, 13.3; IR (film, cmK1): 3330
(br), 1607; MS: HR-FAB calculated for C18H23N3O2Li
(MCLiC) m/z: 320.1950, found: 320.1953; [a]D K71.7 (c
0.99, MeOH).
4.2.7. (1R, 2S)-1-[(5-trifluoromethylpyridyl-2)-amino]-2-
indanol (20d). Standard procedure A was followed [576 mg
(3.17 mmol) of 9, 448 mg (3.00 mmol) of (1R, 2S)-1-amino-
2-indanol 19d, 475 mg of i-Pr2NEt; 105G5 8C for 2 days;
8% i-PrOH/hexanes] 640 mg of grayish solid 9 (73% yield).
4.2.10. (R)-5-diethylcarbamoyl-2-phenyl-2, 3-dihydro-
imidazo[1,2-a]pyridine (7e). A solution of 63 mg
(0.20 mmol) of 6e in anhydrous CH2Cl2 (2 mL) was cooled
to 0 8C under N2 atmosphere and NEt3 (0.085 mL,
0.61 mmol) was added followed by MsCl (0.024 mL,
0.31 mmol). The mixture was stirred at 0 8C for 1 h. The
solvent was removed and residue was extracted with warm
water. The aqueous extract was decanted from the residue,
brought to pH 7–8 with aqueous NaHCO3 and extracted
with CH2Cl2. More aqueous NaHCO3/NaOH was added to
the aqueous phase to pH 12 and extraction was continued
until organic extracts were pale-yellow. The organic phase
was dried over Na2SO4 and then rotary evaporated. The
crude mixture was chromatographed (30% i-PrOH, 3%
NEt3/hexanes) to give 40 mg of yellow solid (68% yield).
1H NMR (300 MHz, CDCl3) d 8.36 (s, 1H), 7.65 (dd, J1Z
8.8 Hz, J2Z2.5 Hz, 1H), 7.22–7.31 (m, 4H), 6.63 (d, JZ
8.8 Hz, 1H), 5.60 (d, JZ7.4 Hz, 1H), 5.47 (dd, J1Z7.4 Hz,
J2Z5.2 Hz, 1H), 4.77 (td, J1Z5.2 Hz, J2Z2.9 Hz, 1H),
3.24 (dd, J1Z16.4 Hz, J2Z5.2 Hz, 1H), 3.04 (dd, J1Z
16.4 Hz, J2Z2.9 Hz, 1H); 13C NMR (75 MHz, CDCl3) d
160.4, 146.0 (q, JZ4.0 Hz), 141.2, 140.3, 134.7 (q, JZ
2.9 Hz), 128.5, 127.3, 125.6, 124.7 (q, JZ271 Hz), 124.6,
116.2 (q, JZ33.2 Hz), 107.5, 73.8, 60.0, 39.8; IR (film,
cmK1): 3331 (br), 1617, 1325; MS: HR-FAB calcd for
C15H13F3N2OLi (MCLiC) m/z: 301.1140, found:
301.1137; mp 130.5–133 8C (partially decomposed before
mp); [a]D K10.1 (c 1.00, MeOH).
1H NMR (300 MHz, CDCl3) d 7.42 (d, JZ1.9 Hz, 1H),
7.27–7.38 (m, 5H), 7.00 (dd, J1Z9.6 Hz, J2Z1.9 Hz, 1H),
6.51 (d, JZ9.6 Hz, 1H), 5.31 (dd, J1Z11.3 Hz, J2Z8.5 Hz,
1H), 4.46 (dd, J1Z11.3 Hz, J2Z11.0 Hz, 1H), 3.93 (dd,
J1Z11.0 Hz, J2Z8.5 Hz, 1H), 3.41 (q, JZ7.1 Hz, 4H), 1.20
(t, JZ7.1 Hz, 6H); 13C NMR (75 MHz, CDCl3) d 167.8,
156.7, 143.7, 136.0, 135.9, 128.7, 127.4, 126.5, 113.2,
113.1, 67.0, 57.2, 41.6, 13.5; IR (film, cmK1):1653, 1609;
MS: HR-FAB calcd for C18H22N3O (MCHC) m/z:
296.1763, found: 296.1757; [a]D C256 (c 1.13, MeOH).
4.2.8. (4bR, 10aS)-8-trifluoromethyl-4b, 10a-dihydro-
11H-indeno[10, 20, 4, 5]imidazo [1,2-a]pyridine (21c).
General procedure B was followed, except that the SOCl2
was added in two portions and the mixture was refluxed
overnight [152 mg (0.517 mmol) of 20c, 2!0.092 mL
SOCl2, 3 mL CHCl3; 5% i-PrOH, 1% NEt3/hexanes] to
afford 43 mg of yellow solid (31% yield).
1H NMR (300 MHz, CDCl3) d 7.52 (d, JZ6.9 Hz, 1H), 7.37
(s, 1H), 7.21–7.31 (m, 3H), 6.83 (dd, J1Z9.9 Hz, J2Z
7.4 Hz, 1H), 6.36 (d, JZ9.9 Hz, 1H), 5.67 (d, JZ9.3 Hz,
1H), 4.92 (dd, J1Z9.3 Hz, J2Z7.4 Hz, 1H), 3.61 (dd, J1Z
17.0 Hz, J2Z7.4 Hz, 1H), 3.25 (d, JZ17.0 Hz, 1H); 13C
NMR (75 MHz, CDCl3) d 155.9, 143.3, 138.7, 132.8 (q, JZ
5.4 Hz), 131.9, 128.5, 128.2, 125.6, 125.1, 123.9 (q, JZ
269 Hz), 116.4, 107.4 (q, JZ34.6 Hz), 76.2, 64.1, 40.3; IR
(film, cmK1): 1661, 1332, 1161, 1145, 1108, 1054; MS: HR-
FAB calcd for C15H12F3N2 (MCHC) m/z: 277.0953, found:
277.0951; mp 121–123.5 8C (partially decomposed before
mp); [a]D C636 (c 0.99, MeOH).
4.2.11. (R)-N-(5-cyanopyridyl-2)-phenylglycinol (6f). A
solution of 277 mg (2.00 mmol) of 2-chloro-5-cyanopyr-
idine 23, 274 mg (2.00 mmol) of (R)-phenylglycinol 5 and
300 mg of i-Pr2NEt in 1 mL of i-PrOH was heated to reflux
in an oil bath kept at 100 8C. After 12 h, the flask was cooled
and the solvent was removed under reduced pressure.
302 mg (63% yield) of colourless oil was produced after
chromatography (EtOAc/hexanesZ1:1/2.5:1).
1H NMR (300 MHz, CDCl3) d 8.22 (d, JZ2.2 Hz, 1H), 7.46
(dd, J1Z8.8 Hz, J1Z2.2 Hz, 1H), 7.26–7.38 (m, 5H), 6.32
(s, 1H), 6.29 (d, JZ8.8 Hz, 1H), 4.88 (m, 1H), 3.97 (dd,
J1Z11.3 Hz, J2Z4.1 Hz, 1H), 3.87 (dd, J1Z11.3 Hz, J2Z
6.6 Hz, 1H), 3.23 (s, br, 1H); 13C NMR (75 MHz, CDCl3) d
159.6, 152.8, 139.8, 138.9, 129.0, 128.0, 126.7, 118.4,
107.5, 97.1, 66.4, 58.1; IR (film, cmK1): 3348 (br), 2218,
1605, 1509; MS: HR-FAB calcd for C14H13N3OLi (MC
LiC) m/z: 246.1219, found: 246.1222; mp 132–133 8C
(partially decomposed before mp); [a]D K131 (c 1.13,
MeOH).
4.2.9. (R)-N-(5-diethylcarbamoyl-pyridyl-2)-phenylgly-
cinol (6e). (a) 6-Chloro-N,N-diethyl-nicotinamide 2212
was prepared from 6-chloronicotinoyl chloride (1.01 g,
5.7 mmol) and diethylamine (3.0 mL, 29 mmol) in 5 mL
CH2Cl2. 1.06 g (88% yield) was obtained after chromatog-
raphy (20% i-PrOH/hexanes).
(b) General procedure
A was followed [719 mg
(3.38 mmol) of 6-chloro-N,N-diethyl-nicotinamide 22,
619 mg (4.52 mmol) of (R)-phenylglycinol 5, 725 mg of
i-Pr2NEt; 150 8C for 3 days; 20% i-PrOH/hexanes to afford
817 mg of the product (77% yield).
4.2.12. (R)-6-cyano-2-phenyl-2, 3-dihydroimidazo[1,2-
a]pyridine (7f). Standard procedure B was followed
[181 mg (0.757 mmol) of 6f, 0.14 mL of SOCl2, 3.8 mL
of CHCl3; 20% i-PrOH, 2% NEt3/hexanes] to afford 126 mg
of yellow solid (75% yield).
1H NMR (300 MHz, CDCl3) d 8.08 (d, JZ1.7 Hz, 1H),
7.20–7.38 (m, 6H), 6.21 (d, JZ8.5 Hz, 1H), 5.01 (s, br, 1H),
4.77 (m, 1H), 3.87 (dd, J1Z11.3 Hz, J2Z3.0 Hz, 1H), 3.75
(dd, J1Z11.3 Hz, J2Z7.4 Hz, 1H), 3.38 (q, JZ6.2 Hz, 4H),
1.15 (t, JZ6.2 Hz, 6H); 13C NMR (75 MHz, CDCl3) d
169.6, 158.8, 146.3, 139.9, 136.4, 128.3, 127.1, 126.5,
1H NMR (300 MHz, CDCl3) d 7.50 (d, JZ1.7 Hz, 1H),
7.26–7.40 (m, 5H), 6.88 (dd, J1Z9.9 Hz, J2Z1.7 Hz, 1H),
6.48 (d, JZ9.9 Hz, 1H), 5.34 (dd, J1Z11.3 Hz, J2Z8.7 Hz,
1H), 4.26 (t, JZ11.3 Hz, 1H), 3.92 (dd, J1Z11.3 Hz, J2Z

292
V. B. Birman et al. / Tetrahedron 62 (2006) 285–294
8.7 Hz, 1H); 13C NMR (75 MHz, CDCl3) d 154.9, 142.9,
141.9, 135.5, 128.9, 127.8, 126.5, 117.5, 115.4, 87.6, 67.7,
57.0; IR (film, cmK1): 2215, 1653; MS: HR-FAB calcd for
C14H12N3, (MCHC) m/z: 222.1031, found: 222.1026; [a]D
C370 (c 1.09, MeOH).
The crude product is pure by 1H NMR. If necessary it may
be chromatographed (30% EtOAc, 1.5% NEt3/hexanes).
1H NMR (300 MHz, CDCl3) d 7.27–7.37 (m, 6H), 7.10 (d,
JZ1.9 Hz, 1H), 5.33 (dd, J1Z11.5 Hz, J2Z8.8 Hz, 1H),
4.51 (dd, J1Z11.5 Hz, J2Z11.0 Hz, 1H), 3.99 (dd, J1Z
11.0 Hz, J2Z8.8 Hz, 1H); 13C NMR (75 MHz, CDCl3) d
195.6, 143.6, 141.4, 133.1, 128.8, 127.6, 126.8, 109.2, 92.9,
67.0, 59.7; IR (film, cmK1): 1635, 1557; MS: HR-FAB
calcd for C13H11Br2N2, (MCHC) m/z: 352.9289, found:
352.9301; mp 147–149 8C (dec); [a]D C183 (c 1.12,
MeOH).
4.2.13. (R)-N-(3-fluoropyridyl-2)-phenylglycinol (6g).
Standard procedure A was followed [530 mg (4.03 mmol)
of 2-chloro-3-fluoropyridine 24, 466 mg (3.40 mmol) of 5,
0.89 mL of i-Pr2NEt; 165G5 8C for 2.5 days, 40% EtOAc/
hexanes] to afford 363 mg of colorless liquid product (46%
yield).
1H NMR (300 MHz, CDCl3) d 7.79 (m, 1H), 7.25–7.40 (m,
5H), 7.15 (ddd, J1Z11.0 Hz, J2Z7.7 Hz, J3Z1.4 Hz, 1H),
6.51–6.57 (m, 1H), 5.22 (s, br, 1H), 5.11–5.14 (m, 1H), 4.64
(br s, 1H), 3.94 (d, JZ6.0 Hz, 2H); 13C NMR (75 MHz,
CDCl3) d 148.6 (d, JZ11.5 Hz), 147.5 (d, JZ252 Hz),
142.4 (d, JZ6.0 Hz), 140.4, 128.9, 127.8, 126.9, 120.7 (d,
JZ15.2 Hz), 112.8, 67.9, 58.3; IR (film, cmK1): 3427 (br),
1622, 1504; MS: HR-FAB calcd for C13H13FN2OLi (MC
LiC) m/z: 239.1172, found: 239.1173; [a]D K121 (c 1.06,
MeOH).
4.2.17. (R)-N-(quinolyl-2)-phenylglycinol (27a). Standard
procedure A was followed (0.144 g, 0.880 mmol) of
2-chloroquinoline 26a, 0.128 g (0.933 mmol) of (R)-
phenylglycinol 5, 0.156 g (1.20 mmol) of i-Pr2NEt; 130G
5 8C for 43 h; 5/10% i-PrOH, 1% NEt3/hexanes to afford
0.183 g of white solid (78% yield).
1H NMR (300 MHz, CDCl3) d 7.86 (d, JZ9.1 Hz, 1H),
7.76–7.23 (m, 9H), 6.69 (d, JZ9.1 Hz, 1H), 6.2 (br s, 1H),
5.5 (br s, 1H), 5.11 (br apparent d; JZ3.0 Hz, 1H), 4.07 (dd,
J1Z11.5 Hz, J2Z8.0 Hz, 1H), 3.99 (dd, J1Z11.5 Hz, J2Z
3.3 Hz, 1H); 13C NMR (75 MHz, CDCl3) d 156.6, 145.8,
140.1, 138.3, 130.2, 129.0, 128.0, 127.5, 126.8, 125.0,
123.2, 122.9, 112.0, 68.6, 60.2; IR (film, cmK1): 3260 (br),
1608, 1620; MS: HR-FAB calcd for C17H16N2OLi (MC
LiC) m/z: 271.1423, found: 271.1425; mp 178.5–180 8C;
[a]D K173 (c 0.25, MeOH).
4.2.14. (R)-8-fluoro-2-phenyl-2, 3-dihydroimidazo[1,2-
a]pyridine (7g). Standard procedure B was followed
[62 mg (0.27 mmol) of 6g, 0.049 mL of SOCl2, 2 mL of
CHCl3; 20% i-PrOH, 2% NEt3/hexanes] to afford 44 mg of
yellow oil after chromatography (77% yield).
1H NMR (300 MHz, CDCl3) d 7.25–7.36 (m, 5H), 6.89 (JZ
6.9 Hz, 1H), 6.66 (dd, J1Z9.9 Hz, J2Z6.9 Hz, 1H), 5.66
(m, 1H), 5.36 (dd, J1Z11.5 Hz, J2Z9.5 Hz, 1H), 4.50 (t,
JZ11.5 Hz, 1H), 3.98 (t, JZ9.5 Hz, 1H); 13C NMR
(75 MHz, CDCl3) d 152.0 (d, JZ26.0 Hz), 149.5 (d, JZ
253 Hz), 143.9, 129.5 (d, JZ4.9 Hz), 128.5, 127.2, 126.5,
116.8 (d, JZ14.9 Hz), 100.7 (d, JZ4.9 Hz), 67.6, 57.7; IR
(film, cmK1): 1659, 1577, 1547; MS: HR-FAB calcd for
C13H12FN2 (MCHC) m/z: 215.0985, found: 215.0983; [a]D
C385 (c 1.09, MeOH).
4.2.18. (R)-2-phenyl-1,2-dihydroimidazo[1,2-a]quinoline
(28a). A slightly modified procedure B was followed. A
solution of 105 mg (0.397 mmol) of 27a, in 2 mL of CHCl3
was treated with 0.23 mL (3.15 mmol) of SOCl2 and heated
to reflux. After 1.5 h, the solution was allowed to cool,
treated with a few drops of MeOH, and then heated again for
10 min. The reaction mixture (without prior rotary
evaporation) was treated with aqueous NaHCO3 until the
aqueous phase reached pH 8, then with 1 M NaOH to pH 14,
and extracted with CH2Cl2. The organic extract was dried
over Na2SO4, concentrated, and chromatographed (10%
IPA, 2% NEt3/hexanes) to afford 847 mg of yellow oil
(87%), which later crystallized.
4.2.15. (R)-N-(3, 5-dibromopyridyl-2)-phenylglycinol
(6h). A literature procedure10 was used. A solution of
112 mg (0.523 mmol) N-pyridyl-phenylglycinol 6a1n in
2.5 mL of CH3CN was treated with 187 mg (1.05 mmol)
of N-bromosuccinimide. The mixture was stirred at room
temperature for 2 h, and then concentrated and chromato-
graphed (15/20% EtOAc/hexanes) to afford 189 mg of
white solid (98% yield).
1H NMR (300 MHz, CD3CN) d 7.38–7.24 (m, 8H), 6.94 (td,
J1Z7.4 Hz, J2Z1.1 Hz, 1H), 6.77 (d, JZ8.0 Hz, 1H), 6.53
(d, JZ9.9 Hz, 1H), 5.28 (dd, J1Z11.4 Hz, J2Z8.8 Hz, 1H),
4.38 (dd, J1Z11.4 Hz, J2Z10.4 Hz, 1H), 3.74 (dd, J1Z
10.4 Hz, J2Z8.8 Hz, 1H); 13C NMR (75 MHz, CD3CN) d
157.2, 146.3, 140.5, 137.8, 131.5, 129.6, 129.2, 128.0,
127.8, 122.0, 120.9, 118.0, 112.8, 68.9, 54.8; IR(film, cmK1):
1638, 1573, 1457; MS: HR-FAB calcd for C17H15N2
(MCHC) m/z: 247.1235, found: 247.1233; mp 130.5–
132 8C; [a]D C675 (c 0.26, MeOH).
1H NMR (300 MHz, CDCl3) d 8.01 (d, JZ2.2 Hz, 1H), 7.76
(d, JZ2.2 Hz, 1H), 7.27–7.41 (m, 5H), 5.74 (d, JZ6.6 Hz,
1H), 5.17–5.22 (m, 1H), 3.91–4.01 (m, 2H), 2.81 (br s, 1H);
13C NMR (75 MHz, CDCl3) d 153.1, 147.1, 141.6, 140.0,
128.9, 127.8, 126.8, 106.5, 106.0, 67.1, 57.9; IR (film,
cmK1): 3411 (br), 1579, 1490; MS: HR-FAB calcd for
C13H12Br2ON2Li, (MCLiC) m/z: 376.9476, found:
376.9480; [a]D K5.0 (c 1.11, MeOH).
4.2.19. (R)-N-(7-chloroquinolyl-2)-phenylglycinol (27b).
Standard procedure A was followed, except for the workup
[1.352 g (6.83 mmol) of 2,6-dichloroquinoline, 0.964 g,
(7.03 mmol) of (R)-phenylglycinol 5, 0.991 g (7.67 mmol)
of i-Pr2NEt; 130G5 8C for 2.5 days]. The tube was allowed
to cool to room temperature and the content was diluted
with CH2Cl2, which led to precipitation of a white solid. The
4.2.16. (R)-6, 8-dibromo-2-phenyl-2, 3-dihydro-
imidazo[1,2-a]pyridine (7h). Standard procedure B was
followed [355 mg (0.954 mmol) of 6h, 0.18 mL SOCl2,
5 mL CHCl3] to afford 295 mg of yellow solid (88% yield).

V. B. Birman et al. / Tetrahedron 62 (2006) 285–294
293
mixture was washed with saturated aqueous NH4Cl to
remove i-Pr2NEt and then with saturated aqueous NaHCO3.
The solution was dried over Na2SO4 and evaporated to
afford 1.770 g of the crystalline product (87%), which was
sufficiently pure for the next step. If necessary, the product
can be chromatographed (6% i-PrOH, 0.8% NEt3/hexanes).
125.9, 123.7, 123.5, 122.5, 112.0, 68.8, 60.2, 18.7; IR (film,
cmK1): 3317 (br), 1617, 1494; MS: HR-FAB calcd for
C18H18N2OLi (MCLiC) m/z: 285.1579, found: 285.1572;
mp 168–170 8C; [a]D K145 (c 0.26, MeOH);
4.2.22. (R)-5-methyl-2-phenyl-1,2-dihydroimidazo[1,2-
a]quinoline (28c). Procedure described for 28a was
followed (106 mg (0.381 mmol) of 27c, 0.07 mL of
SOCl2, 2 mL of CHCl3) to afford 0.0758 g of the product
(77% yield).
1H NMR (300 MHz, CDCl3) d 7.73 (d, JZ9 Hz, 1H), 7.62
(d, JZ8.7 Hz, 1H), 7.62 (d, JZ8.7 Hz, 1H), 7.55 (d, JZ
2.2 Hz, 1H), 7.48 (dd, J1Z9 Hz, J2Z2.2 Hz, 1H), 7.28–
7.45 (m, 5H), 6.67 (d, JZ9 Hz, 1H), 5.85 (s, 1H), 5.16 (dt,
J1Z7.4 Hz, J2Z3.6 Hz, 1H), 4.05 (dd, J1Z11.3 Hz, J2Z
7.4 Hz, 1H), 3.00 (dd, J1Z11.3 Hz, J2Z3.6 Hz, 1H); 13C
NMR (75 MHz, CDCl3) d 156.7, 144.8, 140.0, 137.0, 130.5,
129.0, 128.05, 127.98, 126.88, 126.81, 126.2, 123.9, 113.0,
68.3, 59.8; IR (film, cmK1) 3316 (br) 1618.7, 1498.6,
1398.4; MS: HR-EI calcd for C17H15ClN2O (MC) m/z:
298.0873, found: 298.0904; mp 172–174 8C; [a]D K154 (c
0.26, MeOH).
1H NMR (300 MHz, CD3CN) d 7.52 (dd, J1Z7.7 Hz, J2Z
1.1 Hz, 1H), 7.40–7.23 (m, 6H), 6.98 (td, J1Z7.7 Hz; J2Z
1.1 Hz, 1H), 6.79 (dd, J1Z8.0 Hz, J2Z0.7 Hz, 1H), 6.45 (d,
JZ1.1 Hz, 1H), 5.25 (dd, J1Z11.3 Hz, J2Z8.6 Hz, 1H),
4.39 (dd, J1Z11.3 Hz, J2Z10.4 Hz, 1H), 3.74 (dd, J1Z
10.4 Hz, J2Z8.6 Hz, 1H), 2.34 (d, JZ1.1 Hz, 3H); 13C
NMR (75 MHz, CD3CN) d 157.0, 146.4, 145.1, 140.1,
131.3, 129.5, 128.0, 127.8, 126.0, 122.5, 120.7, 116.6,
113.0, 68.6, 55.0, 19.5; IR (film, cmK1): 1643, 1574, 1459,
1399; MS: HR-FAB calcd for C18H17N2 (MCHC) m/z:
261.1392, found: 261.1404; mp 118.5–120 8C; [a]D C525
(c 0.33, MeOH).
4.2.20. (R)-7-chloro-2-phenyl-1,2-dihydroimidazo[1,2-
a]quinoline (28b). Procedure described for 28a was
followed (365 mg, 1.22 mmol) of 27b, 0.23 mL of SOCl2,
6.4 mL of CHCl3 to afford 311 mg of the product (91%
yield) as yellow oil, which crystallized on standing.
4.3. Kinetic resolutions and measurements
1H NMR (300 MHz, CDCl3) d 7.20–7.40 (m, 8H), 6.76 (d,
JZ9.3 Hz, 1H), 6.65 (d, JZ8.8 Hz, 1H), 5.40 (dd, J1Z
11.4 Hz, J2Z8.4 Hz, 1H), 4.40 (dd, J1Z11.4 Hz, J2Z
10.3 Hz, 1H), 3.88 (dd, J1Z10.3 Hz, J2Z8.4 Hz, 1H); 13C
NMR (75 MHz, CDCl3) d 156.2, 143.9, 137.5, 135.8, 130.2,
128.7, 127.5, 127.4, 126.6, 125.4, 122.2, 118.4, 112.7, 67.7,
54.1; IR (film, cmK1) 1640.3, 1568.9, 1433.9; MS: HR-EI
calcd for C17H13ClN2 (MC) m/z: 280.0767, found:
280.0770; mp 134–136 8C; [a]D C746 (c 0.27, MeOH).
4.3.1. Variation of the C2 substituent (Table 2). The
previously published procedure1n was followed exactly. The
data obtained from individual runs are given in (Table 4).
4.3.2. Variation of substitution on the pyridine ring. The
stock solution of each catalyst was prepared by dissolving
0.05 mmol of the catalyst (7a–h and 28a–c) and 98 mg
(0.75 mmol) of i-Pr2NEt in CDCl3 in a 1.00 mL volumetric
test tube and bringing the volume to the mark.
4.2.21. (R)-N-(4-methylquinolyl-2)-phenylglycinol (27c).
Standard procedure A was followed [212 mg (1.19 mmol)
of 2-chloro-4-methylquinoline, 170 mg (1.24 mmol) of (R)-
phenylglycinol 5, 208 mg, of i-Pr2NEt; 130G5 8C for 46 h;
5/10% i-PrOH, 1% NEt3/hexanes to afford 328 mg of
white solid (99% yield).
4.3.3. Enantioselectivity test (Table 3). A one dram vial
was charged with 68 mL (0.50 mmol) of (G)-phenyl ethyl
carbinol and 0.50 mL of the stock solution of the catalyst.
48 mL (0.375 mmol) of propionic anhydride was added (at
which point the timing was started), the contents was mixed
and transferred into a 5 mm NMR tube. The reaction was
1H NMR (300 MHz, CDCl3) d 7.71 (dd, J1Z12.9 Hz, J2Z
8.0 Hz, 2H), 7.53 (t, JZ8.0 Hz, 1H), 7.42–7.23 (m, 6H),
6.45 (s, 1H), 5.10 (dd, J1Z7.4 Hz, J2Z3.3 Hz, 1H), 4.02
(dd, J1Z11.0 Hz, J2Z7.4 Hz, 1H), 3.97 (dd, J1Z11.0 Hz,
J2Z3.3 Hz, 1H), 2.48 (s, 3H); 13C NMR (75 MHz, CDCl3)
d 156.6, 146.4, 145.9, 140.4, 129.6, 129.0, 127.9, 126.8,
monitored by H NMR at room temperature by comparing
1
integration values of peaks at d 4.5 and d 5.6 ppm and
stopped by pouring the contents into a vial with MeOH upon
reaching conversion of 45–50%. Further workup procedure
and determination of ee’s of the product and the unreacted
alcohol have been previously described.1n
Table 4
Catalyst
Expt (#)
% eeE
% eeA(%)
% CHPLC(%)
S
6c
1
2
1
2
1
2
1
2
1
2
90.4
90.6
55.2
57.1
51.7
51.2
K32.5
K30.2
K14.1
K14.1
60.8
57.2
35.1
33.2
21.7
24.7
40.2
38.7
38.9
36.8
29.6
32.5
7b
36.8
36.0
4.8
5.0
3.9
21a
21b
21c
21d
3.9
a
—
K2.0
K1.9
K1.5
K1.6
a
—
K16.4
K22.3
7b
53.7
61.3
a Too low for a reliable measurement by HPLC.
b Estimated from the 1H NMR spectra of the crude reaction mixtures and used to calculate the selectivity factor.4

294
V. B. Birman et al. / Tetrahedron 62 (2006) 285–294
Dalaigh, C. O.; Hynes, S. J.; Maher, D. J.; Connon, S. J. Org.
Biomol. Chem. 2005, 3, 981. (q) Kano, T.; Sasaki, K.;
Maruoka, K. Org. Lett. 2005, 7, 1347. (r) Yamada, S.; Misono,
T.; Iwai, Y. Tetrahedron Lett. 2005, 46, 2239.
4.3.4. Half-life measurements (Table 3). The procedure
described above was followed, except 34 mL (0.25 mmol) of
(R)-phenyl ethyl carbinol was used instead of 68 mL of the
racemate. The conversion was checked at appropriate
1
intervals and plotted versus time. The t ⁄ value was
determined from the resulting graph.
2. Reviews: (a) Vedejs, E.; Jure, M. Angew. Chem., Int. Ed. 2005,
44, 3974. (b) Dalko, P. I.; Moisan, L. Angew. Chem., Int. Ed.
2004, 43, 5138. (c) Jarvo, E. R.; Miller, S. J. Asymmetric
Acylation. In Comprehensive Asymmetric Catalysis; Jacobsen,
E. N., Pfaltz, A., Yamamoto, H., Eds.; Springer: Berlin,
Heidelberg, 2004; Supplement 1, Chapter 43. (d) France, S.;
Guerin, D. J.; Miller, S. J.; Lectka, T. Chem. Rev. 2003, 103,
2985. (e) Dalko, P. I.; Moisan, L. Angew. Chem., Int. Ed. 2001,
40, 3726. (f) Spivey, A. C.; Maddaford, A.; Redgrave, A. J.
Org. Prep. Proced. Int. 2000, 32, 331. (g) Somfai, P. Angew.
Chem., Int. Ed. Engl. 1997, 36, 2731.
2
Acknowledgements
We gratefully acknowledge financial support of this study
provided by NIGMS (NIH Grant R01 GM072682) and
Washington University. Mass spectrometry was provided
by the Washington University Mass Spectrometry
Resource, an NIH Research Resource (Grant No.
P41RR0954).
3. (a) For a review of the synthesis of DHIP derivatives, see:
Sulojeva, E.; Yure, M.; Gudriniece, E. Chem. Heterocycl.
Compd. 1999, 35, 1121. (b) For a review of the properties of
DHIP derivatives, see: Sulojeva, E.; Yure, M.; Gudriniece, E.
Chem. Heterocycl. Compd. 2000, 36, 885.
References and notes
4. For theoretical aspects of kinetic resolution, see: (a) Kagan,
H. B.; Fiaud, J. C. Top. Stereochem. 1988, 18, 249. (b) Sih,
C. J.; Wu, S.-H. Top. Stereochem. 1988, 18, 249.
1. For leading references to nonenzymatic asymmetric acylation
catalysts, see: (a) Fu, G. C. Acc. Chem. Res. 2004, 37, 542. (b)
Vedejs, E.; Daugulis, O. J. Am. Chem. Soc. 2003, 125, 4166.
(c) Kawabata, T.; Stragies, R.; Fukaya, T.; Nagaoka, Y.;
Schedel, H.; Fuji, K. Tetrahedron Lett. 2003, 44, 1545. (d)
Terakado, D.; Koutaka, H.; Oriyama, T. Tetrahedron:
Asymmetry 2005, 16, 1157. (e) Miller, S. J. Acc. Chem. Res.
2004, 37, 601. (f) Spivey, A. C.; Zhu, F.; Mitchell, M. B.;
Davey, S. G.; Jarvest, R. L. J. Org. Chem. 2003, 68, 7379. (g)
Naraku, J.; Shimomoto, N.; Hanamoto, T.; Inanaga, J.
Enantiomer 2000, 5, 135. (h) Lin, M.-H.; RajanBabu, T. V.
Org. Lett. 2002, 4, 1607. (i) Priem, G.; Pelotier, B.;
Macdonald, S. J. F.; Anson, M. S.; Campbell, I. B. J. Org.
Chem. 2003, 68, 3844. (j) Jeong, K.-S.; Kim, S.-H.; Park, H.-J.;
Chang, K.-J.; Kim, K. S. Chem. Lett. 2002, 1114. (k)
Matsumura, Y.; Maki, T.; Murakami, S.; Onomura, O.
J. Am. Chem. Soc. 2003, 125, 2052. (l) Mizuta, S.;
Sadamori, M.; Fujimoto, T.; Yamamoto, I. Angew. Chem.,
Int. Ed. 2003, 42, 3383. (m) Suzuki, Y.; Yamauchi, K.;
Muramatsu, K.; Sato, M. Chem. Commun. 2004, 2770. (n)
Birman, V. B.; Uffman, E. W.; Jiang, H.; Li, X.; Kilbane, C. J.
J. Am. Chem. Soc. 2004, 126, 12226. (o) Ishihara, K.; Kosugi,
Y.; Akakura, M. J. Am. Chem. Soc. 2004, 126, 12212. (p)
5. Hunter, C. A.; Lawson, K. R.; Perkins, J.; Urch, C. J. J. Chem.,
Soc. Perkin Trans. 2 2001, 651.
6. Ma, J. C.; Dougherty, D. A. Chem. Rev. 1997, 97, 1303.
7. Maezaki, N.; Furusawa, A.; Uchida, S.; Tanaka, T. Tetra-
hedron 2001, 57, 9309.
8. Hammett, L. P. J. Am. Chem. Soc. 1937, 59, 96.
9. Hansch, C.; Leo, A.; Unger, S. H.; Kim, K. H.; Nikaitani, D.;
Lien, E. J. J. Med. Chem. 1973, 16, 1207. The spara and smeta
values given in the paper for the CO(NHMe) group were
deemed sufficiently close to those for CO(NEt2).
10. Canibano, V.; Rodriguez, J. F.; Santos, M.; Sanz-Tejedor,
M. A.; Carreno, M. C.; Gonzalez, G.; Garcia-Ruano, J. L.
Synthesis 2001, 2175.
11. (a) Birman, V. B.; Jiang, H. Org. Lett. 2005, 7, 3445. (b) This
work has been presented in part: Birman, V. B.; Jiang, H.
Abstracts of papers, 229th National Meeting of the American
Chemical Society, San Diego, CA, March 13–17, 2005;
American Chemical Society: Washington, DC, 2005; ORGN
313.
12. Yamanaka, H.; Araki, T.; Sakamoto, T. Chem. Pharm. Bull.
1988, 36, 2244.