Benzoimidazole-pyrrolidine (BIP), a highly reactive chiral organocatalyst for aldol process

Article abstract of DOI:10.1016/j.tetlet.2004.08.171

A new chiral benzoimidazole-pyrrolidine ligand (BIP) was shown to catalyze an aldol process in the presence of an equimolar amount of a Br?nsted acid, leading to the aldol adduct in high yield and enantioselectivity. Remarkably, the aldol reaction was sti

Full text of DOI:10.1016/j.tetlet.2004.08.171

Tetrahedron
Letters
Tetrahedron Letters 45 (2004) 8035–8038
Benzoimidazole–pyrrolidine (BIP), a highly reactive chiral
organocatalyst for aldol process
Eric Lacoste,a Yannick Landais,a,* Kurt Schenk,b Jean-Baptiste Verlhaca and
Jean-Marc Vincenta,*
a
´
University Bordeaux-1, 351 Cours de la Liberation, 33405 Talence Cedex, France
bInstitute ofCrystallography, University ofLausanne, BPS Dorigny, CH-1015 Lausanne, Switzerland
Received 30 July 2004; revised 30 August 2004; accepted 31 August 2004
Available online 17 September 2004
Abstract—A new chiral benzoimidazole–pyrrolidine ligand (BIP) was shown to catalyze an aldol process in the presence of an equi-
molar amount of a Bro¨nsted acid, leading to the aldol adduct in high yield and enantioselectivity. Remarkably, the aldol reaction
was still effectively catalyzed when starting from equimolar amounts of aldehyde and ketone in THF.
ꢀ 2004 Elsevier Ltd. All rights reserved.
1
N
Organocatalyzed reactions have recently enjoyed a re-
N
H
3
N
H
O
OH
newed interest, and spectacular progress has been made
in new catalytic methods based on the use of metal-free
chiral organic molecules.1 Amongst the most efficient
inducers, proline occupies a prominent place, and was
shown to act as a bifunctional acid–base catalyst in a
large array of organic transformations including aldol
and Michael reactions, leading to products with gener-
ally high levels of diastereo- and enantiocontrol.2 How-
ever, there remain transformations in which the use of
proline gives unsatisfactory results, suggesting the need
for more efficient organocatalysts having a tunable
structure. In this context, we recently started an investi-
gation on a new class of ligand, which may be useful for
both metal and metal-free catalyzed organic processes.3
We thus report on the design and the utilization of a
simple bifunctional benzoimidazole–pyrrolidine 1 (BIP)
(Scheme 1), which can be considered as a nitrogenated
proline analogue,4 providing, under acid catalysis, aldol
adducts in high yields and stereoselectivities. Such a cat-
alyst was found to mediate at 5mol% loading, the aldol
reaction using an equimolar amount of aldehyde and
ketone, thus circumventing the limitation of the proline
catalyzed process where a large excess of the ketone is
required.
2
H
O
1 (BIP)
O
O2N
2
4
O2N
3
Scheme 1. Aldol reaction catalyzed by BIP 1.
Such catalysts are easily available by simply heating L-
proline and a substituted 1,2-diaminobenzene under
acidic conditions.5 Under these conditions, 1 is obtained
in 60% yield without racemization. Interestingly, the
X-ray crystal structure of 1 incorporates a water mole-
cule, which binds two BIP ligands through a four hydro-
gen bond network (Fig. 1). The oxygen atom of water is
ligated to the two N–HÕs of 1 with O–HN(3) and
˚
O–HN(2) bond lengths of 2.71 and 1.87A, respectively.
The two hydrogen atoms of the water molecule are also
ligated to the electron lone pairs of the two nitrogen
atoms of a second BIP molecule.
Preliminary investigations on the aldol reaction starting
from acetone 2 and p-nitrobenzaldehyde 3 in DMSO
using 1 as a catalyst (30mol%) led to the aldol adduct
4 with moderate yield and 44% ee (Table 1, entry 1).
Under similar conditions, proline was shown to provide
4 in 64% yield and 76% ee.6 We found that addition of a
Bro¨nsted acid into the reaction medium led to a signifi-
cant improvement in terms of reaction rate and yield
Keywords: Organocatalyst; Aldol; Proline; Benzoimidazole; Enamine.
*
Corresponding authors. Tel.: +33 5 400022 89; fax: +33 5 400062
86 (Y.L.); tel.: +33 5 400089 42; fax: +33 5 400061 58 (J.-M.V.);
e-mail addresses: y.landais@lcoo.u-bordeaux1.fr; jm.vincent@
lcoo.u-bordeaux1.fr
0040-4039/$ - see front matter ꢀ 2004 Elsevier Ltd. All rights reserved.
doi:10.1016/j.tetlet.2004.08.171

8036
E. Lacoste et al. / Tetrahedron Letters 45 (2004) 8035–8038
proline was found to be inactive under such conditions.
Finally, when the reaction was carried out under the
same conditions but with only 5mol% of 1 aldol 4
was obtained in good yield and enantioselectivity, albeit
at a slower rate (entry 14).
BIP–TFA appears to be much more reactive than pro-
line for this reaction, which requires the use of a large
excess of acetone (ꢀ25equiv) with respect to the alde-
hyde and 20–30mol% of catalyst loading. It is also note-
worthy that aldol experiments have been repeatedly
carried out on 10mmol scale, demonstrating the practi-
cal value of this protocol. As a comparison, we showed
that the combination of 1 and a Lewis acid such as
Zn(OTf)2 may also efficiently catalyze the aldol process,
suggesting that these ligands also have the potential as
bidendate ligands for metal-catalyzed asymmetric proc-
esses (entry 15).
Figure 1. X-ray crystal structure of 1-0.5H2O.
(entries 2 and 3 vs 1).7,8 The reaction conditions were
then systematically varied using 20mol% of catalyst.
The rate of the aldol process was shown to be solvent
dependent, in the order: THF > CH2Cl2 > DMF (en-
tries 4–7). Better yields but lower eeÕs were observed in
CH2Cl2 (entry 7). Varying the temperature led to the
observation that the reaction was still efficient and gen-
erally more enantioselective at À20ꢁC. But the most
remarkable improvement of the process was observed
by varying the nature of the proton source. Using the
stronger CF3CO2H (TFA) instead of AcOH led to both
an enhancement of the yield and the enantioselectivity
(entry 9 vs 8). The use of even stronger triflic acid did
not bring any further change (entry 12). It was observed
that only 2mol% of the ligand/TFA system, in neat ace-
tone, was sufficient to lead to 5 with a high yield and an
optimal 82% ee (entry 11).9
In contrast to proline,6 BIP 1 has two potential nucleo-
philic sites. When adding an acid, protonation occurs on
the more basic site, that is, the pyrrolidine nitrogen cen-
ter (pKa ꢀ 9–10at N3 vs p Ka ꢀ 5–6 at N1). Formation
of the enamine may still occur on the pyrrolidine moiety
as rate enhancements have been observed during acid-
mediated aldol processes using pyrrolidine-based
catalysts.7 However, the occurrence of an alternative
intermediate with the enamine located on the
benzoimidazole ring could not be ruled out. To test this
hypothesis, a reductive amination reaction was carried
out on 1 in the presence of acetone using NaBH3CN
as a reducing agent (Scheme 2).10 Whether TFA was
added in the reaction mixture or not, only the BIP deriva-
tive 5 with the isopropyl group on the pyrrolidine moiety
was obtained. A BIP derivative with an isopropyl on
the benzoimidazole ring or the bis-isopropyl compound
were not detected. Clearly, under these reaction condi-
tions, the benzoimidazole is not nucleophilic enough to
generate the enamine function. Accordingly, when the
aldol process was performed using 5,11 from which the
More importantly, we found that the aldol process was
still efficiently catalyzed when starting from equimolar
amounts of 2 and 3, using THF as a solvent (entry
13). The aldol 4 was thus obtained in 67% yield and
82% ee in THF at À5ꢁC in the presence of 1
(20mol%) and TFA (20mol%) (entry 13). Noteworthy,
Table 1. BIP 1-catalyzed asymmetric aldol process between acetone 2 and aldehyde 3
Entry
1 (mol%)
Acid (mol%)
T (ꢁC)
Solvent
Time (h)
%Yielda
eeb
c
1
30—
20
DM20SO
AcOH (100)
8
0.5
1
40
65
65
65
61
70
92
62
82
80
87
80
67
69
87
44
50
46
54
75
64
36
2
20
DMSOc
c
3
)
)
)
)
)
)
20AcOHD(M3200SO
20AcOH (20
20AcOH (20
20AcOH (20
20AcOH (20
20AcOH (20
20TFA (20
d
4
À20DMSO
À20DMF
À20THF
À20CH
À20Acetone
À20Acetone
À20Acetone
À5
4
c
c
5
24
8
6
c
7
2Cl2
18
78
86
95
24
84
24
48
17
8
4
18
9
)
101)0
11
12
13
14
15
TFA (10
24
Acetone
24
THFe
THFe
Acetone
2
TFA (2)
TFA (5)
82
)
20TfOH (20
20TFA (20
5
À20Acetone
À5
)
82
80
74
À5
20Zn(OTf)
(20)
20
2
a Isolated yield after purification.
b ee Determined by 1H NMR spectroscopy of the corresponding MosherÕs ester.
c Acetone (25equiv).
d DMSO/acetone 1:2.5.
e Acetone (1.1equiv).

E. Lacoste et al. / Tetrahedron Letters 45 (2004) 8035–8038
8037
the process is noteworthy as equimolar amounts of ace-
tone and aldehyde may be employed, thus opening new
perspectives for such aldol reactions. Moreover, the
presence on the ligand of several nucleophilic and basic
sites acting synergistically, as well as various functional-
izable sites, suggest that 1 is a potentially attractive tem-
plate. Further refinement of the ligand structure and
extension of the utility of this new organocatalyst class
to synthetically relevant organic processes are under
active investigation and will be reported in due course.
H
N
H
N
1. CF3CO2H (1 equiv.)
N
N
H
2. acetone
NaBH3CN (2 equiv.)
N
N
H
H
1 (BIP)
5
Scheme 2. Reductive amination on BIP 1.
reaction may only take place with an enamine function
located on the benzoimidazole ring, no reaction occurs.
In all studied cases, aldol adduct 4 was shown to possess
the same (R)-absolute configuration as that obtained
with L-proline.6 The stereochemistry of the process
may be rationalized invoking transition state as in Fig-
ure 2, with the aldehyde binding to one of the protons
on the benzoimidazolium ring.12 Based on estimated
pKa values (enamonium pKa ꢀ 4 vs benzoimidazole
pKa ꢀ 5–6) protonation is likely to occur on the benzo-
imidazole ring. Within this arrangement, the aldehyde
thus approaches the Si-face of the enamine, to provide
the (R)-aldol. Increased acidity of the benzoimidazole
N–H proton would thus allow a better activation of
the aldehyde. This would explain the higher enantiose-
lectivity observed with TFA (pKa ꢀ À1) where complete
protonation should occur in comparison with AcOH
(pKa ꢀ 4.7) in which the protonation is only partial.
Acknowledgements
´
This work was supported by the CNRS, the Region
Aquitaine and the Institut Universitaire de France. E.L.
thanks the MNERT for a Ph.D. fellowship. We grate-
fully acknowledge Dr. D. Bassani for helpful
discussions.
References and notes
1. For recent reviews on organocatalysis, see: (a) Dalko, P.
I.; Moisan, L. Angew. Chem., Int. Ed. 2001, 40, 3727–3748;
(b) Jarvo, E. R.; Miller, S. J. Tetrahedron 2002, 58, 2481–
2495; For recent examples, see: (c) Alexakis, A.; Bernar-
dinelli, G.; Andrey, O. Org. Lett. 2003, 5, 2559–2561; (d)
Jørgensen, K. A.; Melchiorre, P. J. Org. Chem. 2003, 68,
4151–4157; (e) Mocquet, C. M.; Warriner, S. L. Synlett
2004, 356–358; (f) Jørgensen, K. A.; Aburel, P. S.;
Halland, N. Angew. Chem., Int. Ed. 2003, 42, 661–665;
(g) Barbas, C. F., III; Chowdari, N. S.; Suri, J. T. Org.
Lett. 2004, 6, 2507–2510; (h) Northrup, A. B.; Mangion,
I. K., F.; Hettche, F.; MacMillan, D. W. C. Angew.
Chem., Int. Ed. 2004, 43, 2152–2154.
H
N
H
N
H
N
O
X
Ar
2. For recent studies on aldol processes with proline as a
catalyst, see: (a) List, B. Tetrahedron 2002, 58, 5573–5598;
(b) Gro¨ger, H.; Wilken, J. Angew. Chem., Int. Ed. 2001,
40, 529–532; (c) Suzuki, N.; Hayashi, Y.; Shoji, M.;
Tsuboi, W. J. Am. Chem. Soc. 2003, 125, 11208–11209;
(d) Zhong, G. Angew. Chem., Int. Ed. 2003, 42, 4247–4250;
(e) Hayashi, Y.; Tsuboi, W.; Ashimine, I.; Urushima, T.;
Shoji, M.; Sakai, K. Angew. Chem., Int. Ed. 2003, 42,
3677–3680; (f) Duthaler, R. O. Angew. Chem., Int. Ed.
2003, 42, 975–978; (g) List, B.; Vignola, N.; Hoang, L.;
Pidathala, C. Angew. Chem., Int. Ed. 2003, 42, 2785–2788;
(h) MacMillan, D. W. C.; Northrup, A. B. J. Am. Chem.
Soc. 2002, 124, 6798–6799; (i) Barbas, C. F., III; Notz, W.;
Me
H
Figure 2. TS model for BIP 1-mediated aldol reaction.
In an attempt to further extend the methodology, we
submitted cyclopentanone 6 (1.1equiv) to the aldol
process with 3 in the presence of BIP 1 (20mol%). Ald-
ols 7a,b were thus obtained in a satisfying 67% yield and
more importantly with 88% and 86% ee for anti- and
syn-diastereomers, respectively (Scheme 3).13 This com-
pares favorably with the results previously reported for
the related proline catalyzed aldol reaction.6b,7b,14
`
Cordova, A. J. Org. Chem. 2002, 67, 301–303; (j) Anders,
B.; Nagaswamy, K.; Jørgensen, K. A. Chem. Commun.
2002, 620–621; (k) Palomo, C.; Oiarbide, M.; Garcia, J.
O
H
´
´
M. Chem. Eur. J. 2002, 8, 36–44; (l) Cordova, A.; Sunden,
H.; Casas, J. Tetrahedron Lett. 2004, 45, 6117–6119.
3. Vincent, J.-M.; Philouze, C.; Pianet, I.; Verlhac, J.-B.
Chem. Eur. J. 2000, 6, 3595–3599.
O
O
O
OH
Ar
OH
Ar
O2N
(1 equiv.)
1 (20 mol%)
+
4. For the use of a closely related proline tetrazole as an
organocatalyst, see: (a) Cobb, A. J. A.; Longbottom, D.
A.; Shaw, D. M.; Ley, S. V. Chem. Commun. 2004, 1808–
1809; (b) Cobb, A. J. A.; Shaw, D. M.; Ley, S. V. Synlett
2004, 558–560; (c) Torii, H.; Nakadai, M.; Ishihara, K.;
Saito, S.; Yamamoto, H. Angew. Chem., Int. Ed. 2004, 43,
1983–1986; (d) Hartikka, A.; Arvidsson, P. I. Tetrahedron:
Asymmetry 2004, 15, 1831–1834.
6
7a : 7b
d.r. 47 : 53 (67% yield)
TFA (20 mol%)
THF, 24h
88% e.e.
86% e.e.
Scheme 3. BIP 1-mediated aldol reaction of cyclopentanone.
In summary, we described here a highly efficient asym-
metric aldol process catalyzed by a new benzoimidaz-
ole–pyrrolidine ligand 1 (BIP) in the presence of an
equimolar amount of Bro¨nsted acid. The efficiency of
5. 5,6-Dimethyl-2-pyrrolidin-2-yl-1H-benzoimidazole (BIP)
(1). L-Proline (6.8g, 59mmol) was reacted with 4,5-
dimethyl-1,2-phenylenediamine (6.2g, 46mmol) in a 4M

8038
E. Lacoste et al. / Tetrahedron Letters 45 (2004) 8035–8038
aqueous solution of HCl (50mL) under reflux for four
was monitored using TLC (hexanes–EtOAc 7/3). The
crude reaction mixture was then purified by flash column
chromatography on silica gel (hexanes–EtOAc 7/3) to give
the aldol adduct 4. Enantiomeric excesses were determined
using 1H NMR spectroscopy of the corresponding
MosherÕs ester.
days. The solution was then neutralized with a 4M
aqueous solution of NaOH, then filtered, and the precipi-
tate was washed with diethyl ether (10·) to afford 1 as a
beige powder (5.9g, 60%). Mp 94ꢁC (MeOH/H2O); IR
(KBr): m, cmÀ1 3280(NH), 2970, 2871, 2282, 1634, 1444,
1310, 825. 1H NMR (250MHz, DMSO-d6): d 7.30(2H, s),
4.46–4.41 (1H, m), 3.06–2.98 (2H, m), 2.34 (6H, s), 2.22–
2.13 (1H, m), 2.02–1.98 (1H, m), 1.85–1.77 (2H, m). 13C
NMR (75.55MHz, DMSO-d6): d 155.2, 136.9, 131.2,
115.1, 56.3, 46.5, 32.2, 25.5, 20.2. MS (EI) 215 (75, M+),
216 (20), 187 (55), 186 (37), 173 (100), 172 (95), 160 (75),
147 (30), 145 (20). Anal. Calcd for C13H17N3 + 1.5H2O: C,
64.46; H, 8.26; N, 17.35. Found: C, 64.50; H, 7.85; N,
17.25. (a) Cohen, V. I. J. Heterocycl. Chem. 1979, 16, 13–
16; (b) McKee, V.; Zvagulis, M.; Dogdigian, J. V.; Patch,
M. G.; Reed, C. A. J. Am. Chem. Soc. 1984, 106, 4765–
4772.
10. 2-(1-Isopropyl-pyrrolidin-2-yl)-5,6-dimethyl-1H-benzoimi-
dazole (5). Compound 1 (50mg, 0.23mmol) and TFA
(18lL, 0.23mmol) were stirred in acetone for 10min, then
NaBH3CN (30mg, 0.5mmol) was added by portion. After
24h at rt, the resulting solution was treated with water and
Na2CO3. The reaction mixture was extracted with ethyl
acetate and the resulting organic layer dried over MgSO4.
The solvent was evaporated in vacuo to afford 5 as a white
powder (39mg, 65%). Mp 160–178ꢁC; IR (KBr): m, cmÀ1
2966, 1635, 1417, 1309. 1H NMR (300MHz, CDCl3): d
7.25 (2H, s), 4.11–4.08 (1H, m), 3.15–3.05 (1H, m), 2.85–
2.75 (1H, m), 2.60–2.40 (1H, m), 2.28 (6H, s), 2.25–2.15
(1H, m).13C NMR (75.55MHz, DMSO-d6): d 159.2, 131.3,
115.4, 59.7, 53, 50.3, 33.8, 24.7, 22.2, 20.7,19.4. MS
(FAB+) 280([M+Na] +), 258 (100, [M+H]+), 257 (26), 256
(80), 214 (65), 212 (22). HRMS for C16H24N3 ([M+H]+):
258.196543, found: 258.197023.
11. The following conditions were used: 5 (10mol%), TFA
(10mol%) in neat acetone from À20ꢁC to rt over 48h.
12. (a) Mase, N.; Thayumanavan, R.; Tanaka, F.; Barbas, C.
F., III. Org. Lett. 2004, 6, 2527–2530; (b) Mase, N.;
Tanaka, F.; Barbas, C. F., III. Angew. Chem., Int. Ed.
2004, 43, 2420–2423; (c) Brown, S. P.; Brochu, M. P.;
MacMillan, D. W. C.; Sinz, C. J. J. Am. Chem. Soc. 2003,
125, 10808–10809.
6. (a) List, B.; Lerner, R. A.; Barbas, C. F., III. J. Am. Chem.
Soc. 2000, 122, 2395–2396; (b) Sakthivel, K.; Notz, W.;
Bui, T.; Barbas, C. F., III. J. Am. Chem. Soc. 2001, 123,
5260–5267; (c) List, B.; Houk, K. N.; Bahmanyar, S.;
Martin, H. J. J. Am. Chem. Soc. 2003, 125, 2475–2479.
7. For a recent study on the influence of a proton source on
the rate of aldol process, see: (a) Barbas, C. F., III;
Tanaka, F.; Mase, N. Org. Lett. 2003, 5, 4369–4372; (b)
Nakadai, M.; Saito, S.; Yamamoto, H. Tetrahedron 2002,
58, 8167–8177; (c) Brochu, M. P.; Brown, S. P.; MacMil-
lan, D. W. C. J. Am. Chem. Soc. 2004, 126, 4108–4109.
8. (a) Oka, N.; Saigo, K.; Wada, T. J. Am. Chem. Soc. 2003,
125, 8307–8317; (b) Greco, M. N.; Maryanoff, B. E.
Tetrahedron Lett. 1992, 33, 5009–5012.
13. The enantiomeric excess was determined by chiral phase
HPLC analysis (Chiralcel ODꢂ column; hexane–i-PrOH
9:1).
14. Proline-catalyzed aldol reaction of cyclopentanone led to a
1:2 syn/anti isolated ratio with 63% and 69% ee, respec-
tively, in the presence of 20mol% of catalyst.6b
9. Representative experimental aldol reaction: (Table 1,
entry 14): BIP 1 (10.4mg, 45lmol) and TFA (3.8lL,
49lmol) were stirred with acetone (81lL, 1.1mmol) and
THF (1mL) for 5min, then p-nitrobenzaldehyde
(150.6mg, 1mmol) was added. The reaction progress