Asymmetric alkylation of glycine imines using in situ generated phase-transfer catalysts

Article abstract of DOI:10.1016/S0040-4039(02)01982-2

In this paper we report an investigation into the possibility of effecting the asymmetric alkylation of glycine imines using chiral quaternary ammonium salt catalysts that are generated in situ. It is demonstrated that O-alkyl N-alkyldihydrocinchonidinium salts can be assembled from the parent alkaloid under the reaction conditions required for the liquid-liquid phase-transfer alkylation of glycine imines, and that reactions performed in this way give enantioselectivities comparable to those obtained using pre-prepared catalysts. Utilization of this methodolgy in the generation and screening of catalyst libraries is also demonstrated.

Full text of DOI:10.1016/S0040-4039(02)01982-2

TETRAHEDRON
LETTERS
Pergamon
Tetrahedron Letters 43 (2002) 8015–8018
Asymmetric alkylation of glycine imines using in situ generated
phase-transfer catalysts
a,
a
b
c
Barry Lygo, * Benjamin I. Andrews, John Crosby and Justine A. Peterson
a
School of Chemistry, University of Nottingham, Nottingham NG7 2RD, UK
b
AstraZeneca, Process R&D, Silk Road Business Park, Charter Way, Macclesfield, Cheshire SK10 2NA, UK
c
Department of Chemistry, University of Salford, Salford M5 4WT, UK
Received 20 August 2002; accepted 13 September 2002
Abstract—In this paper we report an investigation into the possibility of effecting the asymmetric alkylation of glycine imines
using chiral quaternary ammonium salt catalysts that are generated in situ. It is demonstrated that O-alkyl N-alkyldihydrocin-
chonidinium salts can be assembled from the parent alkaloid under the reaction conditions required for the liquid–liquid
phase-transfer alkylation of glycine imines, and that reactions performed in this way give enantioselectivities comparable to those
obtained using pre-prepared catalysts. Utilization of this methodolgy in the generation and screening of catalyst libraries is also
demonstrated. © 2002 Elsevier Science Ltd. All rights reserved.
The utility of N-anthracenylmethyl substituted cin-
chona alkaloids, e.g. 1 as phase-transfer catalysts for
ternary ammonium salts of this type have wide utility in
asymmetric synthesis, and consequently efficient meth-
ods of generating and/or optimizing the catalyst struc-
ture are of significant interest.
1
the asymmetric alkylation of glycine imines 2 (Scheme
1
) was first reported in 1997. This chemistry can be
2 3
effected under liquid–liquid, solid–liquid, and micellar
phase-transfer conditions, and is also similarly effec-
4
Salts 1 are normally prepared via quaternization of the
tive under homogeneous reaction conditions involving
appropriate cinchona alkaloid, e.g.
4
with 9-
5
phosphazene bases.
chloromethylanthracene, followed (when necessary) by
biphasic alkylation of the free hydroxyl group (Scheme
2,3
It has also been shown that the same catalysts 1 are
effective in a number of other enantioselective CꢀC
2). It has been known for some time that the O-alkyl-
ation step can be effected in situ during the liquid–liq-
6
2a,f,9
bond-forming reactions and also in the highly enan-
uid phase-transfer alkylation of imine 2,
and we
7
tioselective epoxidation of a,b-unsaturated ketones.
have recently been investigating whether it is also possi-
ble to effect the initial N-quaternization in the same
reaction system.
Cinchona alkaloid derivatives bearing a number of
other N-alkyl substituents have also been shown to be
effective in a variety of asymmetric phase-transfer alkyl-
8
ation processes. These results demonstrate that qua-
In order to test this, we initially examined the effect of
exposing imine 2 to catalytic amounts of dihydrocin-
chonidine 4 and anthracenylmethyl halides (1.3 equiv.)
at room temperature, under typical liquid–liquid phase-
transfer conditions (Scheme 3). It was found that when
9
-chloromethylanthracene was employed, none of the
Scheme 1.
*
0
Corresponding author.
Scheme 2.
040-4039/02/$ - see front matter © 2002 Elsevier Science Ltd. All rights reserved.
PII: S0040-4039(02)01982-2

8
016
B. Lygo et al. / Tetrahedron Letters 43 (2002) 8015–8018
desired alkylation product 6 could be isolated. In addi-
tion, no quaternary ammonium salt could be detected
in the reaction mixture, which suggests that this reac-
tion failed because quaternization of the cinchonidine
did not take place. In contrast, when the more reactive
The final part of this study involved examination of
whether three different alkylating agents could be uti-
lized. The first two of these would be involved in
generation of the quaternary ammonium catalyst (N-
then O-alkylation), and the third should then partici-
pate in the reaction with the imine substrate 2.
9-bromomethylanthracene 5 was used, efficient alkyla-
tion of the imine 2 was observed (Scheme 3), and
HPLC analysis of the corresponding N-benzoate
derivative indicated that this process was highly
enantioselective.
We considered that this type of protocol would be most
useful for the automated screening of potential catalyst
structures. Consequently, we opted to investigate its
application in the parallel synthesis and screening of a
small library of catalysts. For this study we chose the
alkylation of glycine imine 2 with 3-bromo-2-methyl-
propene as the assay reaction. This particular alkyla-
tion reaction had not previously been optimized in our
laboratories and so we considered that this would be a
useful test of the methodology. Thus, using a simple
robotic system, a library of 20 catalysts were generated
in parallel, and the enantioselectivities monitored by
HPLC (Fig. 1). Initially we utilized similar reaction
conditions to those outlined in Scheme 4. Unfortu-
nately, although the desired transformations took
place, a number of the quaternary ammonium salts
generated exhibited poor solubility in toluene. This
resulted in poor reproducibility in some of the screens.
Fortunately this problem was overcome by adding
2
g
We also attempted to investigate the use of 9-
iodomethylanthracene in this reaction process, however
the instability of this material made it impractical for
1
0
routine use. This initial result established that it is
possible to generate an effective asymmetric phase-
transfer catalyst in situ, and also provides a highly
effective method 1for accessing anthracenylmethyl
1
glycine derivatives.
Extension of this protocol to allow utilization of two
different alkylating agents was then examined. In this
case we chose imine 7 as the target since it is straight-
forward to assess the enantiomeric excess of this mate-
rial by chiral HPLC. Initial experiments established
that the optimal conditions for in situ generation of
N-anthracenylmethyldihydrocinchonidinium bromide 1
12
dichloromethane to the reaction mixture after the
(
R=H, X=Br) involved treatment of the starting alka-
loid with 1.2–1.5 equiv. 9-bromomethylanthracene 5, at
0–75°C in toluene for 5 h. After this time a second
initial N-quaternization had taken place, and using this
13
modification enantioselectivities could be reproduced
to ±3% over a number of runs.
6
alkylating agent, in this case benzyl bromide, could be
added along with aqueous base and imine 2. This
resulted in rapid formation of the desired phenylalanine
imine 7. None of the corresponding anthracenylmethyl
imine 6 appeared to be formed under these conditions,
and as far as we could detect, the hydroxyl group of the
catalyst had only been alkylated with benzyl bromide.
Independent synthesis and testing of selected examples
of these catalysts confirmed that the screening protocol
generated valid results, and utilizing this approach,
1
H NMR analysis of the reaction mixture indicated that
all the starting 9-bromomethylanthracene 5 had been
consumed, indicating that the slight excess of this
reagent, required to ensure complete quaternization, is
degraded during catalyst formation.
The enantioselectivity (93% e.e.) of this process is com-
parable to that obtained if pre-formed, recrystallized,
N-anthracenylmethyldihydrocinchonidinium bromide 1
(
R=H, X=Br) is employed, suggesting that minor
amounts of less selective catalytic elements that could
have been generated have minimal effect on reaction
selectivity.
Figure 1.
Scheme 3.
Scheme 4.

B. Lygo et al. / Tetrahedron Letters 43 (2002) 8015–8018
8017
Crosby, J.; Peterson, J. A. Tetrahedron Lett. 1999, 40,
671; (f) Lygo, B.; Crosby, J.; Lowdon, T. R.; Wain-
8
wright, P. G. Tetrahedron 2001, 57, 2391; (g) Lygo, B.;
Crosby, J.; Lowdon, T. R.; Peterson, J. A.; Wainwright,
P. G. Tetrahedron 2001, 57, 2403; (h) Lygo, B.; Crosby,
J.; Peterson, J. A. Tetrahedron 2001, 57, 6447; (i) Lygo,
B.; Humphreys, L. D. Tetrahedron Lett. 2001, 43, 6677;
(
j) Ma, D.; Zhang, T.; Wang, G.; Kozikowski, A. P.;
Lewin, N. E.; Blumberg, P. M. Bioorg. Med. Chem. Lett.
001, 11, 99.
. (a) Corey, E. J.; Xu, F.; Noe, M. C. J. Am. Chem. Soc.
997, 119, 12414; (b) Corey, E. J.; Noe, M. C.; Xu, F.
Tetrahedron Lett. 1998, 39, 5347; (c) Horikawa, M.;
Busch-Petersen, J.; Corey, E. J. Tetrahedron Lett. 1999,
40, 3843; (d) Boisnard, S.; Carbonnelle, A.-C.; Zhu, J.
Org. Lett. 2001, 3, 2061; (e) Jotterand, N.; Pearce, D. A.;
Imperiali, B. J. Am. Chem. Soc. 2001, 66, 3224; (f) Chen,
G.; Deng, Y.; Gong, L.; Mi, A.; Cui, X.; Jiang, Y.; Choi,
M. C. K.; Chan, A. S. C. Tetrahedron: Asymmetry 2001,
12, 1567.
2
3
1
Scheme 5.
quaternary ammonium salt 1 (R=Bn, X=Br) was
identified as the most effective catalyst for this transfor-
mation, generating the desired imine 8 in 92% e.e.
14
The results shown in Fig. 1 also demonstrate that both
N- and O-substituents in the catalyst can play a signifi-
cant role in determining the level of enantioselectivity.
This highlights the need to vary both of these sub-
stituents when search for optimal catalysts, and empha-
sizes the utility of the methodology described here.
4. Okino, T.; Takemoto, Y. Org. Lett. 2001, 3, 1515.
5. (a) O’Donnell, M. J.; Delgado, F.; Hostettler, C.;
Schwesinger, R. Tetrahedron Lett. 1998, 39, 8775; (b)
O’Donnell, M. J.; Delgado, F.; Pottorf, R. S. Tetrahedron
1
999, 55, 6347; (c) O’Donnell, M. J.; Delgado, F.;
It is also worth noting that the above reactions were
performed at ambient temperature (25–30°C) simply for
convenience. If required, it is usually possible to
increase the level of enantioselectivity simply by reduc-
ing the reaction temperature. For example, by using the
optimal catalyst 1 (R=Bn, X=Br) identified in Fig. 1,
Dominguez, E.; de Blas, J.; Scott, W. L. Tetrahedron:
Asymmetry 2001, 12, 821.
6. See for example: (a) Corey, E. J.; Bo, Y.; Busch-Petersen,
J. J. Am. Chem. Soc. 1998, 120, 13000; (b) Zhang, F.-Y.;
Corey, E. J. Org. Lett. 2000, 2, 1097; (c) Perrard, T.;
Plaquevent, J.-C.; Desmurs, J.-R.; H e´ brault, D. Org.
Lett. 2000, 2, 2959; (d) Zhang, F.-Y.; Corey, E. J. Org.
Lett. 2001, 3, 639.
15
and reducing the reaction temperature to 0°C, imine 8
can be generated in 98% e.e. (Scheme 5).
7
. (a) Lygo, B.; Wainwright, P. G. Tetrahedron Lett. 1998,
39, 1599; (b) Lygo, B.; Wainwright, P. G. Tetrahedron
1999, 55, 6289; (c) Lygo, B.; To, D. C. M. Tetrahedron
Lett. 2001, 42, 1343.
In conclusion, this study has demonstrated that cin-
chona alkaloid-derived quaternary ammonium salts can
be generated in situ during the liquid–liquid phase-
transfer alkylation of glycine imine 2, and that this
protocol results in enantioselectivities similar to those
obtained with pre-prepared catalysts. We believe that
his chemistry is likely to be particularly useful for the
identification of optimal catalysts structures for a given
transformation, and that this approach should be appli-
cable to a wide variety of phase-transfer reaction
processes.
8. See for example: (a) Hughes, D. L.; Dolling, U.-H.;
Ryan, K. M.; Schoenewaldt, E. F.; Grabowski, E. J. J. J.
Org. Chem. 1987, 52, 4745; (b) Arai, S.; Tsuge, H.;
Shioiri, T. Tetrahedron Lett. 1998, 39, 7563; (c) Jew, S. S.;
Jeong, B. S.; Yoo, M. S.; Huh, H.; Park, H. G. Chem.
Commun. 2001, 1244; (d) Park, H. G.; Jeong, B. S.; Yoo,
M. S.; Park, M. K.; Huh, H.; Jew, S. S. Tetrahedron Lett.
2001, 42, 4645; (e) Thierry, B.; Plaquevent, J. C.; Cahard,
D. Tetrahedron: Asymmetry 2001, 12, 983.
9. O’Donnell, M. J.; Wu, S.; Huffman, J. C. Tetrahedron
Acknowledgements
1994, 50, 4507.
0. Fernandez, M.-J.; Gude, L.; Lorente, A. Tetrahedron
1
Lett. 2001, 42, 891.
We thank the EPSRC for funding and AstraZeneca for
a studentship (to J.A.P.).
11. For applications of anthracenylmethyl glycine deriva-
tives, see: (a) Nestor, J. J.; Ho, T. L.; Simpson, R. A.;
Horner, B. L.; Jones, G. H.; McRae, G. I.; Vickery, B. H.
J. Med. Chem. 1982, 25, 795; (b) Hohsaka, T.; Kajihara,
D.; Ashizuka, Y.; Murakami, H.; Sisido, M. J. Am.
Chem. Soc. 1999, 121, 34.
References
1
. For an authoritative review of glycine imine chemistry,
see: O’Donnell, M. J. Aldrichim. Acta 2001, 34, 1.
. (a) Lygo, B.; Wainwright, P. G. Tetrahedron Lett. 1997,
12. Sufficent dichloromethane was added so as to give a final
toluene:dichloromethane ratio of 7:3. This solvent ratio
has previously been reported to be optimal for alkyla-
tions of this type, see: Esikova, I. A.; Nahreini, T. S.;
O’Donnell, M. J. In ACS Symposium Series 659: Phase-
Transfer Catalysis—Mechanisms and Syntheses; Halpern,
M. E., Ed.; ACS: Washington, DC, 1997; p. 89.
2
3
8, 8595; (b) Dehmlow, E. V.; Wagner, S.; M u¨ ller, A.
Tetrahedron 1999, 55, 6335; (c) Lygo, B.; Crosby, J.;
Peterson, J. A. Tetrahedron Lett. 1999, 40, 1385; (d)
Lygo, B. Tetrahedron Lett. 1999, 40, 1389; (e) Lygo, B.;

8
018
B. Lygo et al. / Tetrahedron Letters 43 (2002) 8015–8018
13. Typical procedure for catalyst screening: Dihydrocin-
R (Chiralcel OD-H, 99:1, hexane:IPA, 232 nm, 0.5 ml/min)
t
chonidine (2.5 mg, 0.0085 mmol) in toluene (390 ml) was
treated with 9-bromomethylanthracene (110 ml of a 0.085
M solution in toluene, 0.0093 mmol) and the mixture
stirred at 75°C for 5 h. After cooling to room temperature,
dichloromethane (430 ml) was added, and the mixture
stirred for a further 1 h. Benzyl bromide (100 ml of a 0.0085
M solution in toluene, 0.0085 mmol) and 9 M aqueous
potassium hydroxide (250 ml, 2.3 mmol) were then added
and the mixture stirred for 1 h. 3-Bromo-2-methylpropene
8.3 min (S-isomer), 10.1 min (R-isomer). If desired, the
product could be isolated by extraction with ethyl acetate
(3×1 ml). The combined organic extracts needed to be
filtered through a short column of finely-ground anhydrous
sodium sulfate, then concentrated under reduced pressure
to give the crude alkylated imine 8. Material isolated in this
1
way was usually sufficiently pure (as judged by H NMR)
for use in subsequent reactions.
14. This value is the average e.e. (four runs) obtained using
catalyst 1 (R=Bn, X=Br) generated in situ. When pre-pre-
pared catalyst was used under similar conditions, imine 8
was obtained in 93% e.e.
(
2
100 ml of a 1.0 M solution in toluene, 0.10 mmol) and imine
(300 ml of a 0.28 M solution in toluene, 0.085 mmol) were
then added and the mixture stirred vigorously (ca. 1500
rpm) at room temperature until reaction was complete
15. The aqueous phase was saturated with KBr in order to
prevent freezing.
(4–18 h). Enantioselectivities were monitored via HPLC,