Enantioselective synthesis of (1 R,2 S)-1-amino-2- vinylcyclopropanecarboxylic acid ethyl ester (vinyl-ACCA-OEt) by asymmetric phase-transfer catalyzed cyclopropanation of (E)- N -phenylmethyleneglycine ethyl ester

Article abstract of DOI:10.1021/op100070d

A concise asymmetric synthesis of (1R,2S)-1-amino-2- vinylcyclopropanecarboxylic acid ethyl ester, a key intermediate in the preparation of many hepatitis C virus inhibitors, is described. Stereoselective cyclopropanation of (E)-N-phenylmethyleneglycine ethyl ester was effected by treatment with trans-1,4-dibromo-2-butene in the presence of a catalytic amount of a chiral phase-transfer catalyst. Microscale high-throughput experimentation techniques were successfully used to identify a cinchonidine-derived catalyst that provided (1R,2S)-1-(E)-N-phenylmethyleneamino-2-vinylcyclopropanecarboxylic acid ethyl ester in up to 84% ee. This was translated to a lab scale process to attain 78% yield and 77.4% ee. Chiral purity upgrade and isolation of the ester was accomplished via preparatory supercritical fluid chromatography followed by crystallization of the ester as its tosylate salt. The improved synthesis described herein represents a potentially more economical preparation of this valuable intermediate.

Full text of DOI:10.1021/op100070d

Organic Process Research & Development 2010, 14, 692–700
Enantioselective Synthesis of (1R,2S)-1-Amino-2-vinylcyclopropanecarboxylic Acid
Ethyl Ester (Vinyl-ACCA-OEt) by Asymmetric Phase-Transfer Catalyzed
Cyclopropanation of (E)-N-Phenylmethyleneglycine Ethyl Ester
Kevin M. Belyk,* Bangping Xiang,* Paul G. Bulger, William R. Leonard, Jr., Jaume Balsells, Jingjun Yin, and Cheng-yi Chen
Merck Research Laboratories, Department of Process Research, 126 East Lincoln AVenue, Rahway, New Jersey 07065, U.S.A.
Abstract:
A concise asymmetric synthesis of (1R,2S)-1-amino-2-vinylcyclo-
propanecarboxylic acid ethyl ester, a key intermediate in the
preparation of many hepatitis C virus inhibitors, is described.
Stereoselective cyclopropanation of (E)-N-phenylmethyleneglycine
ethyl ester was effected by treatment with trans-1,4-dibromo-2-
butene in the presence of a catalytic amount of a chiral phase-
transfer catalyst. Microscale high-throughput experimentation
techniques were successfully used to identify a cinchonidine-
derived catalyst that provided (1R,2S)-1-(E)-N-phenylmethylene-
amino-2-vinylcyclopropanecarboxylic acid ethyl ester in up to 84%
ee. This was translated to a lab scale process to attain 78% yield
and 77.4% ee. Chiral purity upgrade and isolation of the ester
was accomplished via preparatory supercritical fluid chromatog-
raphy followed by crystallization of the ester as its tosylate salt.
The improved synthesis described herein represents a potentially
more economical preparation of this valuable intermediate.
Figure 1. Structures of HCV NS3/4A protease inhibitors.
Introduction
Interest in the development of novel treatments for the
hepatitis C virus (HCV) has led to the identification of multiple
antiviral targets for HCV.1 In particular, several potent HCV
NS3/4a protease inhibitors have recently been reported.2 The
macrocyclic BILN 2061 (1, Figure 1) was the first small-
molecule peptidomimetic demonstrated to show antiviral activity
in patients infected with HCV genotype 1.3 One common
structural motif present in these inhibitors is derived from
(1R,2S)-1-amino-2-vinylcyclopropanecarboxylic acid (vinyl-
ACCA) (2). This key synthetic intermediate is typically ap-
pended at the P1 position of HCV inhibitors such as 3.2b The
terminal vinyl-ACCA group can also be fused via ring-closing
metathesis4 (RCM) to form a P1-P3 macrocyclic constraint
as found in 1. We required a cost-effective synthesis of ACCA
derivative 4 (vinyl-ACCA-OEt) to minimize the overall cost
related to the synthesis of a NS3/4a protease inhibitor in
development. Several preparations of various ACCA derivatives
have been reported,5 however, few synthetic routes have been
developed for large-scale implementation.
Scheme 1. Boehringer-Ingelheim enzymatic resolution
approach
The synthesis of vinyl-ACCA-OMe (9) via enzymatic
resolution was recently reported by Boehringer-Ingelheim5a to
be amenable to multikilogram scale (Scheme 1). While this
process was fairly cost-effective, a further reduction in manu-
facturing cost was desired for our purposes. Two other potential
alternative preparations of vinyl-ACCA derivatives were re-
cently described. Fox and co-workers5b reported on the catalytic
asymmetric synthesis of 4 which utilized asymmetric Pd-
catalyzed allylic alkylation of 3,4-epoxy-1-butene with glycine
equivalents derived from the benzophenone imine of either
glycine ethyl ester or 2-aminoacetonitrile. The desired (1R,2S)-
* Authors for correspondence. E-mail: kevin_belyk@merck.com;
bangping_xiang@merck.com.
(1) Gordon, C. P.; Keller, P. A. J. Med. Chem. 2005, 48, 1.
692
•
Vol. 14, No. 3, 2010 / Organic Process Research & Development
10.1021/op100070d  2010 American Chemical Society
Published on Web 04/07/2010

Scheme 2. Phase-transfer catalysis screening conditions
1-amino-2-vinylcyclopropane-1-carboxylic acid ethyl ester (4)
was obtained in 8% overall yield and 88% ee. The Charette
group5c also demonstrated that Cu(I)-catalyzed asymmetric
cyclopropanation of methyl nitroacetate-derived phenyliodo-
nium ylides with 1,3-butadiene provided the corresponding
1-nitro-2-vinyl-cyclopropyl carboxylate in good yield (84%) and
enantioselectivity (90% ee), albeit with moderate diastereose-
lectivity (82:18 dr). While both of these routes provided high
selectivity for each of the key enantio-differentiating steps, the
cost and commercial availability of the raw materials associated
with each of these catalytic steps eliminated them from
consideration. Re-evaluation of the other reported syntheses of
ACCA derivatives5 provided few potential alternatives for us
to consider. A more detailed analysis of potential cost improve-
ments to the existing enzymatic resolution approach seemed
warranted based on the limited number of alternatives available.
The use of inexpensive glycine equivalent 6a in the resolu-
tion approach (Scheme 1), coupled with use of an inexpensive
enzyme for the resolution of racemic-8, set a cost objective
that was difficult to improve upon. Any alternative process
would almost certainly need to include a stereoselective reaction.
One attractive possibility would involve an asymmetric variant
of the dialkylation reaction between a glycine equivalent and
trans-1,4-dibromo-2-butene (7). Indeed, such an approach
employing asymmetric phase-transfer catalysis (PTC) was
using catalyst 10
previously investigated by the Boehringer-Ingelheim team;
however, their initial attempts met with poor results.6 At first
glance, this type of approach appeared limited because a more
costly imine-protected tert-butyl glycinate derivative (e.g., 6b)
would likely be required in order to achieve the most useful
enantioselectivities based on O’Donnell’s pioneering investiga-
tions with cinchona alkaloid-based catalysts7a and Maruoka’s
investigations with N-spiro C2-symmetric chiral quaternary
ammonium bromide catalysts.7b However, recent success in the
asymmetric phase-transfer catalyzed mono- and dialkylation of
the cheaper methyl and ethyl glycinate derivatives offered a
potential area for further investigation.8
We now report the development of a concise asymmetric
synthesis of (1R,2S)-1-amino-2-vinylcyclopropanecarboxylic
acid ethyl ester (4) featuring a diastereo- and enantioselective
dialkylation reaction under PTC conditions as the key step.
(2) (a) McCauley, J. A.; Rudd, M. T.; Nguyen, K. T.; McIntyre, C. J.;
Romano, J. J.; Bush, K. J.; Varga, S. L.; Ross, C. W.; Carroll, S. S.;
DiMuzio, J.; Stahlhut, M. W.; Olsen, D. B.; Lyle, T. A.; Vacca, J. P.;
Liverton, N. J. Angew. Chem., Int. Ed. 2008, 47, 9104. (b) Liverton,
N. J.; Holloway, K. M.; McCauley, J. A.; Rudd, M. T.; Butcher, J. W.;
Carroll, S. S.; DiMuzio, J.; Fandozzi, C.; Gilbert, K. F.; Mao, S.-S.;
McIntyre, C. J.; Nguyen, K. T.; Romano, J. J.; Stahlhut, M.; Wan,
B.-L.; Olsen, D. B.; Vacca, J. P. J. Am. Chem. Soc. 2008, 130, 4607.
(c) Vendeville, S.; Nilsson, M.; Kock, H.; Lin, T.; Antonov, D.;
Classon, B.; Ayesa, S.; Ivanov, V.; Johansson, P.; Kahnberg, P.;
Eneroth, A.; Wikstrom, K.; Vrang, L.; Edlund, M.; Lindstro¨m, S.;
Vreken, W.; McGowan, D.; Tahri, A.; Hua, L.; Lenz, O.; Delouvroy,
F.; Dooren, M.; Kindermans, N.; Surleraux, D.; Wigerinck, P.;
Rosenquist, A.; Samuelsson, B.; Simmen, K.; Raboisson, P. Bioorg.
Med. Chem. Lett. 2008, 18, 1689. (d) Lamarre, D.; Anderson, P. C.;
Bailey, M. D.; Beaulieu, P. L.; Bolger, G.; Bonneau, P.; Bo¨s, M.;
Cameron, D. R.; Cartier, M.; Cordingley, M. G.; Faucher, A.-M.;
Goudreau, N.; Kawai, S. H.; Kukolj, G.; Lagace´, L.; LaPlante, S. R.;
Narjes, H.; Poupart, M.-A.; Rancourt, J.; Sentjens, R. E.; St-George,
R.; Simoneau, B.; Steinmann, G.; Thibeault, D.; Tsantrizos, Y. S.;
Weldon, S. M.; Yong, C.-L.; Llina`s-Brunet, M. Nature 2003, 426,
186. (e) McCauley, J. A.; Rudd, M. T.; McIntyre, C. J.; Nguyen, K. T.;
Romano, J. J.; Butcher, J. W.; Gilbert, K. F.; Bush, K. J.; Holloway,
M. K.; Wan, B.-L.; Carroll, S. S.; DiMuzio, J. M.; Graham, D. J.;
Ludmerer, S. W.; Mao, S.-S.; Stahlhut, M.; Fandozzi, C.; Trainor,
N.; Olsen, D. B.; Vacca, J. P.; Liverton, N. J. J. Med. Chem. 2010,
53, 2443.
Results and Discussion
Initial evaluation of the phase-transfer catalyzed reaction
between 6c and 7 (Scheme 2) in the presence of 20 mol %
N-benzylcinchoninium bromide (10) under standard PTC condi-
tions (toluene/50% aqueous KOH, 0 °C) showed rapid conver-
sion (<1 h) of 6c. Unfortunately, multiple species unrelated to
the cyclopropanation adduct were formed. Exposing authentic
racemic cyclopropanation product (rac-11) to the same condi-
tions also resulted in rapid decomposition. It was assumed that
ester hydrolysis was rapidly occurring under these conditions
so we evaluated a variety of inexpensive aqueous and solid
inorganic bases (NaOH, KOH, K2CO3, and K3PO4) under
phase-transfer conditions to try to identify clean dialkylation
conditions. In general, carbonate and phosphate bases provided
little conversion of 6c to the cyclopropanation product (1S,2R)-
11 even at 20 °C. Powdered KOH, however, provided 60%
conversion of 6c to (1S,2R)-11 while powdered NaOH cleanly
provided complete conversion of 6c at -5 °C (16 h), though
the enantioselectivity was only 30% ee. The cinchonine-derived
catalysts, in general, provided the undesired (1S,2R)-trans-
enantiomer 11.
(3) Hinrichsen, H.; Benhamou, Y.; Wedemeyer, H.; Reiser, M.; Sentjens,
R. E.; Calleja, J. L.; Forns, X.; Erhardt, A.; Cro¨enlein, J.; Chaves, R.;
Yong, C.-L.; Nehmiz, G.; Steinmann, G. G. Gastroenterology 2004,
127, 1347.
(4) (a) Farina, V.; Zeng, X.; Wei, X.; Xu, Y.; Zhang, L.; Haddad, N.;
Yee, N. K.; Senanayake, C. H. Catal. Today 2009, 140, 74. (b) Shu,
C.; Zeng, X.; Hao, M.-H.; Wei, X.; Yee, N. K.; Busacca, C. A.; Han,
Z.; Farina, V.; Senanayake, C. H. Org. Lett. 2008, 10, 1303.
(5) (a) Beaulieu, P. L.; Gillard, J.; Bailey, M. D.; Boucher, C.; Duceppe,
J. S.; Simoneau, B.; Wang, X.-J.; Zhang, L.; Grozinger, K.; Houpis,
I.; Farina, V.; Heimroth, H.; Krueger, T.; Schnaubelt, J. J. Org. Chem.
2005, 70, 5869. (b) Fox, M. E.; Lennon, I. C.; Farina, V. Tetrahedron
Lett. 2007, 48, 945. (c) Moreau, B.; Charette, A. B. J. Am. Chem.
Soc. 2005, 127, 18014. (d) Zhu, S.; Perman, J. A.; Zhang, X. P. Angew.
Chem., Int. Ed. 2008, 47, 8460. (e) Jime´nez, J. M.; Rife´, J.; Ortun˜o,
R. M. Tetrahedron: Asymmetry 1996, 7, 537. (f) Cativiela, C.; D´ıaz-
de-Villegas, M. D.; Jime´nez, A. I. Tetrahedron: Asymmetry 1995, 6,
177. (g) Fliche, C.; Braun, J.; Le Goffic, F. Synth. Commun. 1994,
24, 2873.
While the Maruoka-type phase-transfer catalysts were also
considered for asymmetric cyclopropanation, our limited in-
vestigations with them proved inferior to the cinchona alkaloid-
(6) See ref 5a. No experimental details or results were provided.
(7) (a) O’Donnell, M. J.; Bennett, W. D.; Wu, S. J. Am. Chem. Soc. 1989,
111, 2353. (b) Ooi, T.; Kameda, M.; Maruoka, K. J. Am. Chem. Soc.
2003, 125, 5139.
(8) (a) Lygo, B.; Allbutt, B. Synlett 2004, 2, 326. (b) Ooi, T.; Uematsu,
Y.; Maruoka, K. Tetrahedron Lett. 2004, 45, 1675.
Vol. 14, No. 3, 2010 / Organic Process Research & Development
•
693

Table 1. Screening results summary for asymmetric phase-transfer catalyzed dialkylation of 6c
based catalysts.9 We subsequently demonstrated that the use
of the pseudo-enantiomeric cinchonidine-derived catalysts (e.g.,
12, Table 1) led to the formation of the desired (1R,2S)-trans-
product 11, some in useful enantioselectivities. In order to
thoroughly investigate the effect of the cinchona alkaloid-
derived catalyst structure, over 100 cinchonidine-derived cata-
lysts were prepared from a diverse set of commercially available
benzyl bromide derivatives. This collection of catalysts was
evaluated for asymmetric cyclopropanation10 and also served
as a valuable resource for screening other PTC reactions. These
catalysts were efficiently screened against various glycine
equivalents using microscale high-throughput experimentation
techniques previously described by Merck.11 All phase-transfer
screening experiments were conducted at total reaction volumes
of 80-100 µL using only 0.5 µmol phase-transfer catalyst.12
Screening of the reaction between glycine equivalent 6c and
7 using powdered NaOH in toluene13 at 0 to -10 °C showed
that the enantioselectivity was strongly influenced by both the
steric and electronic nature of the substituents present at the
ortho- and meta-positions of catalyst 12. While N-benzylcin-
chonidinium bromide (12a) provided almost no selectivity (entry
1), the incorporation of a trifluoromethyl group at the meta-
position of the catalyst provided (1R,2S)-11 in 36% ee (compare
entries 2-4). The enantioselectivity was doubled to >70% ee
if another electron-withdrawing group was also incorporated
at the ortho-position such as for catalysts 12e-g. The bis(tri-
fluoromethyl) catalyst 12g provided (1R,2S)- 11 in 84% ee at
76% conversion of 6c (entry 7). The importance of the 2,5-
substitution pattern for high selectivity was reflected in the
substantial decrease in enantioselectivity when the positioning
of the trifluoromethyl groups was changed (entries 8 and 9).
This dramatic change suggested there was a significant steric
element involved in how the catalyst and substrate enolate
interacted during the enantiodifferentiating step of the final
alkylation.
All possible fluorine-substituted variations of N-(fluoro-
substituted) benzylcinchonidinium bromide phase-transfer cata-
lysts represented by 12 (18 in all) were also evaluated under
the same conditions described in Table 1.14 The N-(2′,3′,4′,5′-
tetrafluoro)benzylcinchonidinium bromide catalyst (12j) pro-
vided the highest selectivity (71% ee) amongst this series (Table
1, entry 10). As another indication of the sensitivity of the
enantioselectivity of this reaction to catalyst structure, the
pentafluoro analogue 12k, however, provided only 8% ee (entry
11). Capping the hydroxyl group of the catalyst resulted in a
(9) For a review see: Maruoka, K. Org. Process Res. DeV. 2008, 12, 679.
Maruoka’s catalyst, (11bS)-(+)-4,4-dibutyl-4,5-dihydro-2,6-bis(3,4,5-
trifluorophenyl)-3H-dinaphth[2,1-c:1′,2′-e]azepinium bromide, pro-
vided (1R,2S)-11, albeit in poor selectivity (17% ee) at 9% conversion
of 6c using the conditions described in Table 1.
(10) Only those phase-transfer catalysts most relevant to the discussion of
the reported results were included here in the interest of brevity.
694
•
Vol. 14, No. 3, 2010 / Organic Process Research & Development

Table 2. Optimization experiments for asymmetric
enantioselectivity for (1R,2S)-11 and conversion of 6c (compare
entries 1, 4, and 7 with entries 2, 5, and 8, respectively).
Addition of 400 mol % water accelerated the rate of the
reactions;15 however, enantioselectivities were still noticeably
lower than seen in screening experiments (entries 3, 6, and 9).
Catalyst 12g, however, still provided the desired trans-product
(1R,2S)-11 in good selectivity (79% ee). Replacement of NaOH
in the phase-transfer reaction with bases such as KOH,
CsOH ·H2O, or RbOH ·H2O failed to increase the observed
enantioselectivity for the dialkylation (entries 10-12).16
Further optimization of the phase-transfer conditions dem-
onstrated that the loading of catalyst 12g could be lowered to
1 mol % if a slightly increased amount of water was added
(entry 14).17 The use of increased amounts of water at lower
catalyst loadings led to more hydrolysis-related byproduct in
subsequent lab-scale experiments (data not shown). Ultimately,
it was found that the correct balance of catalyst loading and
water provided a good reaction rate and yield for the dialky-
lation. The final optimized conditions provided (1R,2S)-11 in
up to 78% assay yield (77.4% ee) if the amount of NaOH
charged was decreased to 3.0 equiv and the amount of water
was decreased to 250 mol % (entry 15). Use of pin-milled
NaOH with a mean particle-size of ∼30 µm resulted in 99%
conversion of 6c within 24 h at 0 °C. Much longer reaction
times (>45 h) and lower yields (<70%) occurred when com-
mercial powdered NaOH was utilized instead.18 The lower yield
was presumably due to the competing hydrolysis pathway. The
aza-Cope derived byproduct 14a, arising from the undesired
cis-cyclopropanation product,19 was also observed in all phase-
transfer catalyzed reactions. A level of ∼6% of 14a occurred
under the final optimized conditions described above.20 None
of the cis-cyclopropanation product was ever observed by 1H
NMR, so diastereoselectivity for the phase-transfer catalyzed
process was estimated to be 94:6 dr if 14a may be assumed to
be an indicator for the cis-cyclopropanation product.21
cyclopropanation of 6c with 7a
catalyst
entry (mol %)
water
% ee (1R,2S)-11
base
(mol %) (% conversion of 6c)b
1c
2
12f (20) NaOH
-
-
79 (77)
53 (13)
72 (99)
82 (80)
63 (35)
60 (98)
84 (76)
76 (43)
79 (99)
79 (55)
61 (96)
52 (98)
79 (99)
80 (98)
77.4 (99)f,g
12f (5)
12f (5)
NaOH
NaOH
3
400
-
4c
5
12l (20) NaOH
12l (5)
12l (5)
NaOH
NaOH
-
6
400
-
7c
8
12g (20) NaOH
12g (5) NaOH
12g (5) NaOH
-
9
400
-
10
11
12
13
14
12g (5) 85% KOH
12g (5) CsOH ·H2O
12g (5) RbOH ·H2O
12g (3) NaOH
-
-
500
500
250
12g (1) NaOH
15d 12g (3) NaOHe
a Reactions were run on 100 mg 6c with 1.3 equiv of 7, 25 mL toluene/g 6c,
600 mol % base, 40-48 h at 0 °C unless otherwise stated. b %ee and conversion
of 6c were monitored by SFC at 254 nm. c Reaction was run at 190 mL toluene/g
6c using Table 1 conditions. d Reaction conducted on 11-g scale of 6c. e 3.0 equiv
of pin-milled NaOH used. f The yield of 11 was 78% by 1H NMR analysis. g The
ratio of trans-product 11 to aza-Cope product 14a was 15:1 by 1H NMR analysis
of the crude reaction mixture.
dramatic decrease in selectivity (11% ee) and conversion (20%)
as observed for catalyst 13 (compare entries 12 and 13). This
result suggests that the free hydroxyl group of the catalyst plays
a critical role in facial recognition of the substrate enolate.
Identification of the top performing catalysts 12f, 12g, and
12l from screening-scale experiments provided a good starting
point for development of a scalable, asymmetric, cyclopropa-
nation process. Phase-transfer reactions performed with these
catalysts on a larger scale (Table 2) under more concentrated
conditions (25 mL of toluene/g 6c) and at a more reasonable
catalyst loading (5 mol % 12) resulted in a drop in both
A series of other glycine equivalents were also evaluated22
under the final optimized asymmetric phase-transfer conditions
(Table 3) to determine if modification of either the nitrogen
protecting group or the ester group could improve the observed
(15) For a review on the role of water in phase-transfer catalyzed processes
see: Albanese, D.; Landini, D.; Maia, A.; Penso, M. Ind. Eng. Chem.
Res. 2001, 40, 2396.
(16) Use of other bases (LiOH, Ca(OH)2, and Mg(OH)2 as nanoparticles)
resulted in no conversion of 6c.
(17) A portion of catalyst 12g (∼40%) could be recovered unchanged from
the PTC reaction and recycled with no observed drop in selectivity.
The catalyst could be recovered by filtration of the PTC reaction and
washing the recovered solids with MTBE.
(11) For examples of applying this technique to reaction screening and
parallel synthesis, see: (a) Pei, T.; Tellers, D. M.; Streckfuss, E. C.;
Chen, C.; Davies, I. W. Tetrahedron 2009, 65, 3285. (b) Dreher, S. D.;
Dormer, P. G.; Sandrock, D. L.; Molander, G. A. J. Am. Chem. Soc.
2008, 130, 9257. (c) Tellers, D. M.; McWilliams, J. C.; Humphrey,
G.; Journet, M.; DiMichele, L.; Hinksmon, J.; McKeown, A. E.;
Rosner, T.; Sun, Y.; Tillyer, R. D. J. Am. Chem. Soc. 2006, 128, 17063.
(d) Tellers, D. M.; Bio, M.; Song, Z. J.; McWilliams, J. C.; Sun, Y.
Tetrahedron: Asymmetry 2006, 17, 550. (e) Shultz, C. S.; Dreher, S. D.;
Ikemoto, N.; Williams, J. M.; Grabowski, E. J. J.; Krska, S. W.; Sun,
Y.; Dormer, P. G.; DiMichele, L. Org. Lett. 2005, 7, 3405. (f) Maligres,
P. E.; Krska, S. W.; Humphrey, G. R. Org. Lett. 2004, 6, 3147.
(12) See the Experimental Section for details.
(18) The powdered NaOH purchased from Sigma-Aldrich had a mean
particle-size of ∼70 µm.
(19) The cis-cyclopropanation product is presumed to spontaneously
undergo aza-Cope rearrangement and then a [1,3]-hydride shift to
provide 14. See ref 5a for a discussion regarding the formation of
14.
(20) The aza-Cope by-product 14 and the intermediate mono-alkylated
product of 6c and 7 were not resolved by our chiral SFC methods.
1
These were readily evaluated by H NMR analysis of crude reaction
mixtures.
(13) Other solvents (e.g. o-xylene, m-xylene, p-xylene, MTBE, CH2Cl2)
were also evaluated, but none provided significantly superior perfor-
mance to that of toluene.
(14) For a complete table of experimental results for the N-(fluoro-
substituted)benzylcinchonidinium catalysts see Table 1 in the Sup-
porting Information. For a detailed study on the electronic effects of
N-(fluoro-substituted)benzylcinchonidinium catalysts see: Jew, S.; Yoo,
M.-S.; Jeong, B.-S.; Park, Y.; Park, P. Org. Lett. 2002, 4, 4245.
(21) Another major by-product observed to form but not fully characterized
was a dimeric species containing two glycine units and one butene
unit. This species was not resolved from 14a under the HPLC
conditions utilized for monitoring the lab-scale reaction. The combined
amount of 14a and this species was 14 area % under the HPLC
conditions described in the Experimental Section.
(22) For a review of phase-transfer catalysis with achiral Schiff base esters
see: O’Donnell, M. J. Acc. Chem. Res. 2004, 37, 506.
Vol. 14, No. 3, 2010 / Organic Process Research & Development
•
695

Table 3. Results for asymmetric phase-transfer catalyzed alkylation of 6a-6l and 15
dialkylation enantioselectivity. The methyl glycinate 6a provided
similar selectivity for the desired trans-product (entry 2) while
the benzyl glycinate 6d was observed to readily undergo ester
hydrolysis (entry 3). The tert-butyl glycinate 6b, surprisingly,
gave substantially lower selectivity (58% ee, entry 4). Evaluation
of several aldimine-protected ethyl glycinate derivatives also
failed to provide results that improved upon the best enanti-
oselectivity obtained with 6c (Table 3, entry 1).23 Aldimines
6e-6h all provided 11 in 78% ee under the optimized
conditions. Evaluation of aldimines 6i-6k under the original
asymmetric phase-transfer screening conditions (reported in
Table 1) with our chiral phase-transfer catalyst library failed to
provide any catalyst superior to 12g.24 Only monoalkylation of
benzophenone imine derivative 6l occurred even after warming
to 20 °C. Asymmetric cyclopropanation of the aldimine-
protected R-aminoacetonitrile 15 provided the cyclopropanation
product with 35% ee.25 Overall, glycine equivalent 6c provided
the best balance of both raw material cost and enantioselectivity
for the asymmetric cyclopropanation. Both chiral purity upgrade
and isolation of (1R,2S)-vinyl-ACCA-OEt (4) were all that
remained for demonstration of the new process.
Chiral purity upgrade and isolation of the cyclopropanation
product (1R,2S)-11 from the phase-transfer reaction was readily
accomplished by preparatory supercritical fluid chromatography
(SFC) on Chiralcel OJ-H chiral phase (Scheme 3). Chiral SFC
separation provided (1R,2S)-11 in 55% yield from 6c with a
chiral purity of >99% ee. Deprotection of (1R,2S)-11 was
accomplished by treatment of the purified material with
TsOH ·H2O in 2-propanol. The tosylate salt 4 was isolated by
crystallization from the reaction mixture in 91% yield from
(1R,2S)-11 with a chiral purity of >99% ee and an overall
isolated yield of 50% from 6c.
(23) Glycinate derivatives 6a, 6b, and 6d-6l were prepared only for
1
screening purposes and characterized by H NMR.
(24) The trans-cyclopropanation products 11 and aza-Cope derived by-
products 14 prepared from glycinate derivatives 6a-6l were identified
only by 1H NMR characterization of the crude reaction mixtures. These
intermediates were not isolated due to poor stability during purification.
Enantioselectivities for 11 were determined by SFC analysis of the
crude reaction mixture.
(25) Absolute configuration of the product was not determined.
696
•
Vol. 14, No. 3, 2010 / Organic Process Research & Development

Scheme 3. Process for the preparation of
(1R,2S)-vinyl-ACCA-OEt (4)
glovebox unless otherwise noted. The stirring mechanism used
for all microscale experimentation was a heated/cooled tumble
stirring26 module obtained from Symyx Technologies. Analyti-
cal and Preparative SFC was conducted on a Berger Instru-
ments, Inc. Analytical, and Multigram II instrument, respec-
tively. Melting points were obtained on a Bu¨chi B-545 Melting
Point instrument and are uncorrected. Optical rotations were
measured on a Perkin-Elmer 241 polarimeter, and data are
reported as [R]20D (concentration in g/100 mL, solvent). 1H and
13C NMR spectra were recorded on Bruker DPX-400 and DPX-
500 NMR spectrometers. Chemical shifts are reported in ppm
relative to the residual deuterated solvent. Reactions were
monitored by TLC or reverse-phase HPLC on a Hewlett-
Packard 1100 series instrument. High-resolution mass spectra
were recorded on a Micromass QTOF Ultima API US mass
spectrometer by electrospray ionization.
Conclusion
A concise asymmetric cyclopropanation of (E)-N-phenyl-
methyleneglycine ethyl ester (6c) with trans-1,4-dibromo-2-
butene (7) to provide (1R,2S)-1-(E)-N-phenylmethyleneamino-
2-vinylcyclopropanecarboxylic acid ethyl ester (11) under
solid-liquid phase-transfer conditions has been developed.
Application of high-throughput experimentation techniques was
found to be a very useful tool for the rapid screening of a large
number of phase-transfer catalysts. A series of cinchonidinium-
derived phase-transfer catalysts provided the desired cyclopro-
panation product (1R,2S)-11 in up to 84% ee when using
powdered NaOH as a base in toluene at 0 °C. Critical to the
successful completion of the reaction under more concentrated
conditions was the inclusion of water to accelerate the reaction
rate. Various glycine equivalents were evaluated, of which 6c
provided the best balance of both raw material cost and
enantioselectivity for the key asymmetric cyclopropanation step.
Preparatory SFC chromatography provided a suitable means
of upgrading the chiral purity of (1R,2S)-11 to >99%. The
(1R,2S)-vinyl-ACCA-OEt (4) was isolated as the tosylate salt
by acid treatment of (1R,2S)-11 to cleave the aldimine. With
further optimization of the process described herein, we would
expect this asymmetric phase-transfer catalyzed approach to
provide a cost-efficient synthesis of 4 superior to those previ-
ously reported. We anticipate that classical resolution or an
efficient chromatographic separation for the chiral purity
upgrade of 4 would provide a viable manufacturing route to
this high-value pharmaceutical intermediate. Work on further
improving this process is currently ongoing.
General Procedure for the Preparation of Phase-
Transfer Catalysts. A 100 mL three-necked round-bottomed
flask equipped with a magnetic stir bar, condenser, thermo-
couple, and nitrogen inlet was charged with cinchonidine (1
equiv), benzyl bromide (1.2 equiv), and toluene or 2-propanol
(17 mL/g of cinchonidine) at RT. The stirred slurry was heated
to 105 °C (toluene) or 85 °C (2-propanol) and maintained at
that temperature until complete consumption of the cinchonidine
starting material (generally requires <3 h, monitored by HPLC).
The slurry was then cooled to RT and filtered, washing with
fresh solvent (3×). The resulting crude product cake was dried
at 35 °C under vacuum with a nitrogen sweep for 24 h. The
catalyst product was ground in a mortar and pestle, and then
used without further purification in the PTC step.
HPLC conditions: Zorbax Eclipse Plus C18 column, 4.6 mm
× 50 mm, 1.8 µm pore size; column temperature 25 °C; flow
rate 1.5 mL/min; linear gradient 10:90 MeCN/0.1% aq H3PO4
to 95:5 in 5 min, then hold at 95:5 for 1 min, then to 10:90 in
0.1 min, then hold at 10:90 for 1.9 min; UV detection at 210
nm.
Commercial sources of cinchonidine typically contain a small
amount of the dihydrocinchonidine analogue, which is not
rejected during the catalyst preparation step. The levels of the
corresponding dihydro-PTC analogue present in the catalyst
batches reported here averaged 6-8%.
Experimental Section
Material and Methods. Pin-milling of NaOH was carried
out using a Hosokawa Mikro LPM. LiOH, CsOH ·H2O, and
RbOH·H2O were purchased from Aldrich and ground in a
mortar under a nitrogen atmosphere inside a glovebox prior to
use. Phase-transfer catalysts 12a, 12l, and 13 are commercially
available. All other cinchonidine-derived phase-transfer catalysts
were prepared using previously reported procedures. All cin-
chonidine-derived catalysts discussed here have been previously
prepared except for catalysts 12b, 12e-12h, and 12p. All
N-arylmethyleneglycine ester derivatives used for screening
purposes were prepared using a previously reported procedure.5a
All high-throughput experimentation and optimization experi-
ments were carried out under a nitrogen atmosphere inside a
N-(2-Trifluoromethyl)benzylcinchonidinium Bromide
(12b). Prepared according to the general procedure from
cinchonidine (2.87 g, 9.75 mmol) and (2-trifluoromethyl)benzyl
bromide (2.8 g, 11.71 mmol) in toluene. Isolated the title
compound (4.57 g, 88%) as an off-white powder and as a
toluene solvate. Mp 158-159 °C dec; HPLC retention time
3.69 min; [R]20D -151.0° (c 1.12, MeOH); 1H NMR (400 MHz,
(26) Unlike the typical rotary magnetic stirrer, the tumble stirrer causes a
magnetic stir bar to tumble vertically end-over-end inside the microvial
instead of spinning the stir bar about the horizontal plane.
Vol. 14, No. 3, 2010 / Organic Process Research & Development
•
697

dmso-d6) δ 8.99 (d, J ) 4.5 Hz, 1H), 8.23 (d, J ) 7.5 Hz, 1H),
8.10 (dd, J ) 8.5, 1.5 Hz, 1H), 7.98 (d, J ) 8.0 Hz, 1H), 7.92
(dd, J ) 7.5, 7.5 Hz, 1H), 7.86-7.82 (m, 3H), 7.73 (dd, J )
7.5, 1.5 Hz, 1H), 6.85 (d, J ) 3.5 Hz, 1H), 6.57 (s, 1H), 5.67
(ddd, J ) 17.5, 10.5, 6.5 Hz, 1H), 5.52 (d, J ) 13.5 Hz, 1H),
5.42 (d, J ) 13.5 Hz, 1H), 5.24 (d, J ) 17.5 Hz, 1H), 4.92 (d,
J ) 10.5 Hz, 1H), 4.38 (br s, 1H), 4.13 (dd, J ) 8.5, 8.5 Hz,
1H), 4.03 (d, J ) 12.0 Hz), 3.45 (dd, J ) 12.0, 12.0 Hz, 1H),
3.15 (ddd, J ) 11.0, 11.0, 3.5 Hz, 1H), 2.71 (s, 1H), 2.13-2.04
(m, 2H), 1.99 (br s, 1H), 1.81 (br s, 1H), 1.23-1.17 (m, 1H);
13C NMR (100 MHz, dmso-d6) δ 150.1, 147.6, 145.1, 137.9,
137.1, 132.7, 131.2, 130.2 (q, JC-F ) 29.0 Hz), 129.8, 129.4,
127.8 (q, JC-F ) 5.5 Hz), 127.1, 125.6, 124.2, 124.1 (q, JC-F
) 255.0 Hz), 123.7, 122.7, 119.9, 116.4, 67.9, 64.7, 59.7, 58.8,
51.2, 37.2, 25.4, 24.3, 21.2; HRMS calcd for C27H28F3N2O+
[M+] 453.2148, found 453.2149.
150.1, 147.6, 145.1, 140.9, 138.0, 133.1 (q, JC-F ) 3.5 Hz),
131.8, 129.8, 129.4, 128.9 (q, JC-F ) 3.5 Hz), 128.2 (q, JC-F
) 32.5 Hz), 127.6, 127.2, 124.2, 123.6, 123.5 (q, JC-F ) 272.5
Hz), 119.9, 116.4, 67.7, 64.5, 59.1, 58.9, 51.2, 37.1, 25.4, 24.4,
21.1; HRMS calcd for C27H2735ClF3N2O+ [M+] 487.1759, found
487.1758.
N-[(2,5-Bis)trifluoromethyl]benzylcinchonidinium bro-
mide (12g). Prepared according to the general procedure from
cinchonidine (2.155 g, 7.32 mmol) and [(2,5-bis)trifluorometh-
yl]benzyl bromide (2.7 g, 8.79 mmol) in toluene. The title
compound was isolated (3.897 g, 89%) as a white powder. Mp
215 °C dec; HPLC retention time 4.15 min; [R]20D -152.8° (c
1
1.03, MeOH); H NMR (400 MHz, dmso-d6) δ 8.99 (d, J )
4.5 Hz, 1H), 8.65 (s, 1H), 8.41 (d, J ) 8.5 Hz, 1H), 8.26-8.22
(m, 2H), 8.10 (d, J ) 8.5 Hz, 1H), 7.86-7.81 (m, 2H), 7.74
(dd, J ) 8.5, 1.0 Hz, 1H), 6.89 (d, J ) 3.5 Hz, 1H), 6.57 (s,
1H), 5.66 (ddd, J ) 17.5, 10.5, 6.0 Hz, 1H), 5.55 (abq, J )
15.0 Hz, 2H), 5.26 (d, J ) 17.5 Hz, 1H), 4.93 (d, J ) 10.5 Hz,
1H), 4.40 (br s, 1H), 4.16-4.10 (m, 2H), 3.52 (dd, J ) 11.0,
11.0 Hz, 1H), 3.23-3.18 (m, 1H), 2.68 (s, 1H), 2.14-2.01 (m,
3H), 1.84-1.72 (br m, 1H), 1.23-1.14 (br m, 1H); 13C NMR
(100 MHz, dmso-d6) δ 150.2, 147.6, 145.0, 137.9, 134.1 (q,
N-(2-Fluoro-5-trifluoromethyl)benzylcinchonidinium Bro-
mide (12e). Prepared according to the general procedure from
cinchonidine (2.5 g, 8.49 mmol) and (2-fluoro-5-trifluoro-
methyl)benzyl bromide (2.62 g, 10.19 mmol) in toluene. Isolated
the title compound (4.51 g, 96%) as a white powder. Mp
234-235 °C dec; HPLC retention time 3.86 min; [R]20
D
1
-141.9° (c 1.12, MeOH); H NMR (500 MHz, dmso-d6) δ
JC-F ) 30.0 Hz), 133.7 (q, JC-F ) 3.0 Hz), 132.6 (q, JC-F
33.5 Hz), 129.8, 129.6 (q, JC-F ) 5.5 Hz), 129.5, 128.2 (q,
JC-F ) 3.5 Hz), 127.4, 127.2, 124.2, 123.6, 123.4 (q, JC-F
)
8.98 (d, J ) 4.5 Hz, 1H), 8.49 (d, J ) 5.0 Hz, 1H), 8.37 (d, J
) 8.5 Hz, 1H), 8.10 (d, J ) 8.0 Hz, 1H), 8.09-8.07 (m, 1H),
7.84 (dd, J ) 8.0, 8.0 Hz, 1H), 7.81 (d, J ) 4.5 Hz, 1H), 7.77-7.71
(m, 2H), 6.82 (d, J ) 4.0 Hz, 1H), 6.57 (s, 1H), 5.68 (ddd, J )
17.5, 10.5, 6.5 Hz, 1H), 5.37 (d, J ) 13.0 Hz, 1H), 5.23 (d,
J ) 13.0 Hz, 1H), 5.21 (d, J ) 17.5 Hz, 1H), 4.95 (d, J )
10.5 Hz, 1H), 4.36 (br s, 1H), 4.05 (dd, J ) 8.5, 8.5 Hz,
1H), 3.94 (d, J ) 12.0 Hz, 1H), 3.47 (dd, J ) 11.5, 11.5
Hz, 1H), 3.26 (dd, J ) 11.0, 4.0 Hz, 1H), 2.68 (s, 1H),
2.15-2.06 (m, 2H), 2.00 (s, 1H), 1.83-1.79 (m, 1H),
1.29-1.25 (m, 1H); 13C NMR (100 MHz, dmso-d6) δ 163.9
(d, JC-F ) 255.0 Hz), 150.1, 147.6, 145.1, 137.9, 133.3-133.2
(m, 1C), 130.6-130.4 (m, 1C), 129.8, 129.4, 127.2, 125.8
)
275.0 Hz), 123.2 (q, JC-F )273.5 Hz), 119.9, 116.5, 68.0, 64.8,
59.2, 58.2, 51.2, 37.3, 25.4, 24.4, 21.3; HRMS calcd for
C28H27F6N2O+ [M+] 521.2022, found 521.2027.
N-[(2,4-Bis)trifluoromethyl]benzylcinchonidinium Bro-
mide (12h). A 100 mL three-necked round-bottomed flask
equipped with a magnetic stir bar, condenser, thermocouple,
and nitrogen inlet was charged with cinchonidine (1.914 g, 6.5
mmol), [(2,4-bis)trifluoromethyl]benzyl bromide (2.42 g, 7.82
mmol, 1.2 equiv), and 2-propanol (32.5 mL) at RT. The stirred
slurry was heated to 85 °C and maintained at that temperature
for 19 h. The resulting dark-red solution was cooled to RT and
aged for a further 7 h. The solution was then concentrated to a
small volume (∼5 mL), and then 1:1 toluene/MTBE (100 mL)
was added with stirring. The resulting slurry was aged at RT
for 3 h then filtered, washing with MTBE (3×). The product
cake was dried at 35 °C under vacuum with a nitrogen sweep
for 24 h, to give the title compound (3.569 g, 73%) as a tan
(dq, JC-F ) 33.0, 3.5 Hz), 124.2, 123.7, 123.6 (q, JC-F
)
272.5 Hz), 120.0, 117.8 (d, JC-F ) 24.0 Hz), 116.9 (d, JC-F
) 15.5 Hz), 116.4, 67.7, 64.2, 59.1, 55.6, 51.0, 37.1, 25.6,
24.3, 21.0; HRMS calcd for C27H27F4N2O+ [M+] 471.2054,
found 471.2061.
N-(2-Chloro-5-trifluoromethyl)benzylcinchonidinium Bro-
mide (12f). Prepared according to the general procedure from
cinchonidine (2.5 g, 8.49 mmol) and (2-chloro-5-trifluoro-
methyl)benzyl bromide (2.79 g, 10.19 mmol) in toluene. Isolated
the title compound (4.73 g, 98%) as a white powder. Mp
powder and as a toluene solvate. Mp 153-155 °C dec; HPLC
1
retention time 4.22 min; [R]20 -113.5° (c 1.08, MeOH); H
D
NMR (500 MHz, MeOH-d4) δ 8.97 (d, J ) 4.5 Hz, 1H), 8.54
(d, J ) 8.0 Hz, 1H), 8.52-8.47 (m, 1H), 8.24 (s, 1H), 8.18 (d,
J ) 8.0 Hz, 1H), 8.12-8.10 (m, 1H), 8.01 (d, J ) 4.5 Hz,
1H), 7.86-7.81 (m, 2H), 6.68 (s, 1H), 5.74-5.67 (m, 3H), 5.29
(d, J ) 17.5 Hz, 1H), 4.97 (d, J ) 10.5 Hz, 1H), 4.64-4.59
(m, 1H), 4.26 (dd, J ) 9.0, 9.0 Hz, 1H), 4.18 (d, J ) 12.0 Hz,
1H), 3.73-3.62 (m, 1H), 3.31-3.26 (m, 2H), 2.80 (br s, 1H),
2.26-2.08 (m, 3H), 1.92-1.86 (m, 1H), 1.33-1.29 (m, 1H);
13C NMR (125 MHz, MeOH-d4) δ 151.0, 148.5, 147.8, 139.7,
138.6, 134.2 (q, JC-F ) 33.5 Hz), 133.9 (q, JC-F ) 30.5 Hz),
131.5, 131.4, 130.8-130.6 (m, 1C), 130.1, 129.3, 126.3-126.2
(m, 1C), 126.1, 125.0 (q, JC-F ) 274.5 Hz), 124.7, 124.6 (q,
JC-F ) 272.5 Hz), 121.4, 117.8, 70.2, 66.9, 62.6, 60.1, 53.8,
233.5-234 °C dec; HPLC retention time 3.96 min; [R]20
D
1
-176.9° (c 1.05, MeOH); H NMR (500 MHz, dmso-d6) δ
8.99 (d, J ) 4.5 Hz, 1H), 8.51 (s, 1H), 8.37 (d, J ) 8.5 Hz,
1H), 8.11 (d, J ) 8.5 Hz, 1H), 8.02 (dd, J ) 8.5, 1.5 Hz, 1H),
7.97 (d, J ) 8.5 Hz, 1H), 7.86-7.81 (m, 2H), 7.75 (dd, J )
8.5, 8.5 Hz, 1H), 6.88 (d, J ) 3.5 Hz, 1H), 6.59 (s, 1H), 5.67
(ddd, J ) 17.0, 10.5, 6.5 Hz, 1H), 5.47 (d, J ) 13.0 Hz, 1H),
5.30 (d, J ) 13.0 Hz, 1H), 5.23 (d, J ) 17.0 Hz, 1H), 4.94 (d,
J ) 10.5 Hz, 1H), 4.48 (br s, 1H), 4.06-4.03 (m, 2H), 3.49
(dd, J ) 11.5, 11.5 Hz, 1H), 3.24 (ddd, J ) 11.0, 11.0, 3.5 Hz,
1H), 2.68 (s, 1H), 2.14-2.05 (m, 2H), 2.01 (s, 1H), 1.79 (br s,
1H), 1.24-1.19 (m, 1H); 13C NMR (100 MHz, dmso-d6) δ
698
•
Vol. 14, No. 3, 2010 / Organic Process Research & Development

39.3, 27.5, 26.1, 23.0; HRMS calcd for C28H27F6N2O+ [M+]
521.2022, found 521.2028.
N-(2,3,5-Trifluoro)benzylcinchonidinium Bromide (12p).
Prepared according to the general procedure from cinchonidine
(1.943 g, 6.6 mmol) and (2,3,5-trifluoro)benzyl bromide (1.8
g, 8.0 mmol) in 2-propanol. Isolated the title compound (2.813
Analytical SFC conditions for reaction monitoring: Chiracel
OJ-H column; 250 mm × 4.6 mm; column temperature 35 °C;
flow rate 2.0 mL/min; pressure 200 bar; isocratic 4% 2-propanol
(containing 25 mM isobutylamine) in CO2; UV detection at
254 nm. Retention times: 3.53 min (desired enantiomer (1R,2S)-
11), 4.17 min (undesired enantiomer (1S,2R)-11), 4.90 min
(starting imine 6c), 5.84 min (aza-Cope 14a + monoalkylated
product of 6c and 7).
g, 64%) as a peach powder. Mp 182-183 °C dec; HPLC
1
retention time 3.49 min; [R]20 -137.0° (c 1.08, MeOH); H
D
NMR (500 MHz, dmso-d6) δ 8.99 (d, J ) 4.5 Hz, 1H), 8.33
(d, J ) 8.5 Hz, 1H), 8.10 (d, J ) 8.5 Hz, 1H), 7.90-7.83 (m,
2H), 7.80 (d, J ) 4.5 Hz, 1H), 7.77-7.73 (m, 2H), 6.80 (d, J
) 3.5 Hz, 1H), 6.52 (s, 1H), 5.68 (ddd, J ) 17.5, 10.5, 6.5 Hz,
1H), 5.28 (d, J ) 12.5 Hz, 1H), 5.20 (d, J ) 17.5 Hz, 1H),
5.13 (d, J ) 12.5 Hz, 1H), 4.96 (d, J ) 10.5 Hz, 1H), 4.29 (br
s, 1H), 4.02 (dd, J ) 8.5, 8.5 Hz, 1H), 3.84 (d, J ) 11.5 Hz,
1H), 3.52 (dd, J ) 12.0, 12.0 Hz, 1H), 3.32-3.28 (m, 1H),
2.66 (s, 1H), 2.17-2.07 (m, 2H), 2.01 (s, 1H), 1.80 (s, 1H),
1.27 (dd, J ) 11.0, 11.0 Hz, 1H); 13C NMR (100 MHz, dmso-
d6) δ 156.9 (ddd, JC-F ) 244.0, 11.0, 2.5 Hz), 150.1, 150.0
General Procedure: Optimization Studies for the Di-
alkylation of Imine 6c Using Hydroxide Bases, Water as
Additive, with Phase-Transfer Catalyst 12 (Table 3). The
following procedure using NaOH is representative. A chilled
(0 °C) solution of imine 6c (100 mg, 0.523 mmol) in 0.65 mLof
toluene was added to a chilled (0 °C) suspension of 12g (9.4
mg, 15.6 µmol, 0.03 equiv) in 1.2 mL of toluene in an 8 mL
glass vial containing a Teflon-coated stir bar (1/2 in. × 1/8 in.).
The resulting mixture was agitated at 700 rpm on a tumble
stirrer for 0.5 h. A chilled (0 °C) solution of trans-1,4-dibromo-
2-butene (137 mg, 0.64 mmol, 1.22 equiv) in 0.65 mL of toluene
was then added followed by addition of powdered NaOH (125
mg, 3.12 mmol, 6.0 equiv) and water (47 µL, 2.6 mmol). The
resulting mixture was agitated at 700 rpm on a tumble stirrer
at +1 to -3 °C for 24-48 h. Reaction analysis was carried
out as described above.
(ddd, 1JC-F ) 249.0, 14.5, 14.5 Hz), 147.5, 146.9 (ddd, JC-F
)
247.0, 13.5, 4.0 Hz), 145.1, 137.9, 129.8, 129.4, 127.2, 124.2,
123.6, 120.0, 118.5 (dd, JC-F ) 13.0, 10.5 Hz), 117.1 (dd, JC-F
) 25.0, 2.5 Hz), 116.4, 108.6 (dd, JC-F ) 28.0, 21.5 Hz), 67.8,
64.3, 59.4, 55.5, 51.1, 37.1, 25.5, 24.4, 20.9; HRMS calcd for
C26H26F3N2O+ [M+] 439.1992, found 439.1992.
Preparative Scale Synthesis of (1R,2S)-1-(E)-N-Phenyl-
methylene-amino-2-vinylcyclopropanecarboxylic Acid Eth-
yl Ester (1R,2S)-11. Toluene (280 mL) cooled to -1 °C
(internal temperature) was treated with imine 6c (11.28 g; 59
mmol, 1.0 equiv), the flask was rinsed down with 20 mL of
toluene, and N-[(2,5-bis)trifluoromethyl]benzyl cinchonidinium
bromide (1.064 g, 1.77 mmol, 0.03 equiv) was charged. This
slurry was aged for 5 min, then 1,4-dibromo-2-butene (16.74
g, 77 mmol, 1.3 equiv) was charged as a solid, and the slurry
was further aged for 10 min. Sodium hydroxide (pin-milled,
7.30 g, 177 mol, 3.0 equiv) was added in portions at <0 °C.
The resulting suspension was aged for another 10 min at 0 °C,
then water (2.66 mL, 147 mmol, 2.5 equiv) was added over 8
min at <0 °C. The reaction was agitated at 500 rpm at 0 °C for
22 h to reach >98% conversion (monitored by HPLC). The
reaction mixture was filtered through solka-floc, rinsing with
ice-cold MTBE (100 mL) (if it was desired to recycle the
catalyst, the mixture was instead filtered through filter paper,
rinsing with ice-cold MTBE). The filtrate was then concentrated
by rotary evaporation under vacuum, maintaining the water bath
at <30 °C, to give crude (1R,2S)-11 (22.38 g, 50 wt % as
determined by 1H NMR analysis against 1,3,5-trimethoxyben-
zene internal standard, 78% assay yield, 77.4% ee). The crude
product mixture was purified by preparative SFC to give the
title compound (8.49 g, 93 wt %, >99% ee, 55% overall yield
from 6c) as a colorless oil. [R]20D -102.3° (c 10, CHCl3); 1H
NMR (400 MHz, CDCl3) δ 8.37 (s, 1H), 7.77-7.75 (m, 2H),
7.44-4.40 (m, 3H), 5.79 (ddd, J ) 17.0, 10.5, 8.5 Hz, 1H),
5.30-5.26 (m, 1H), 5.13 (dd, J ) 10.5, 1.5 Hz, 1H), 4.26 (q,
J ) 7.0 Hz, 2H), 2.31-2.26 (m, 1H), 2.01 (dd, J ) 8.0, 5.5
Hz, 1H), 1.71 (dd, J ) 9.5, 8.0 Hz, 1H), 1.31 (t, J ) 7.0 Hz,
3H); 13C NMR (100 MHz, CDCl3) δ 169.7, 160.3, 136.1, 134.1,
130.9, 128.6, 128.2, 117.3, 61.2, 54.1, 36.0, 21.7, 14.3; HRMS:
product fragmented under the ionization conditions with cleav-
High-Throughput Experimentation Studies for the Di-
alkylation of Imine 6 Using Sodium Hydroxide and Phase-
Transfer Catalysis (Table 1). General Protocol. The catalysts
were arrayed in microvials, and the phase-transfer reactions were
run in parallel using a 96-well plate format to maximize
screening efficiency. In a typical experiment, the glycine
equivalent (2.5 µmol) and excess alkylating reagent were
dispensed as solutions in reaction solvent to an 800 µL microvial
predosed with the phase-transfer catalyst (0.5 µmol). Excess
base (6 equiv) was then added, and the resulting mixture was
agitated at 450 rpm27 to effect the PTC reaction. Screening
experiments were efficiently conducted at total reaction volumes
of 80-100 µL and substrate concentrations of about 0.03 M.
RepresentatiVe Procedure. A chilled (0 °C) solution of 6c
(0.48 mg, 2.5 µmol, 1.0 equiv) in 40 µL toluene was added to
the phase-transfer catalyst (0.54 µmol, 0.2 equiv) contained in
a 0.65 mm diameter × 30 mm height glass vial containing a
1.98 mm diameter × 4.80 mm length super-tumble stir dowel.
The resulting mixture was agitated at 400 rpm on a tumble
stirrer for 45 min. A chilled (0 °C) solution of trans-1,4-
dibromo-2-butene (7) (0.68 mg, 3.2 µmol, 1.3 equiv) in 20 µL
toluene was then added to the reaction vial followed by 25 µL
of a chilled (0 °C) suspension of powdered NaOH (0.6 mg, 15
µmol, 6 equiv) in toluene. The reaction vial was sealed with a
Teflon-lined cover and the resulting mixture agitated at 450 rpm
on a tumble stirrer at -10 to 0 °C for 40-48 h. The reaction
was diluted with 600 µL of chilled (0 °C) toluene, and the solids
were allowed to settle prior to chiral analysis by supercritical
fluid chromatography (SFC).
(27) Stir rates of >400 rpm were necessary to obtain reproducible results.
For systematic studies investigating stir rate effects on phase-transfer
catalyzed reactions see: Dehmlow, E. V.; Dehmlow, S. S. Phase
Transfer Catalysis; VCH: Weinheim, 1993.
Vol. 14, No. 3, 2010 / Organic Process Research & Development
•
699

age of the aldimine CdN bond to generate the free amine -
calcd for C8H14NO2 [MH]+ 156.1019, found 156.1027.
Analytical SFC conditions for reaction monitoring: Chiralcel
OJ-H column, 250 mm × 4.6 mm; column temperature 35 °C;
flow rate 2.0 mL/min; pressure 200 bar; isocratic 4% 1:1
2-propanol (containing 25 mM isobutylamine)/n-heptane in
CO2; UV detection at 254 nm. Retention times: 3.89 min
(desired enantiomer (1R,2S)-11), 4.79 min (undesired enanti-
omer (1S,2R)-11).
Analytical HPLC conditions for reaction monitoring: Xterra
RP18 column, 4.6 mm × 250 mm; column temperature 25 °C;
flow rate 1.5 mL/min; linear gradient 40:60 MeCN/10 mM aq
sodium tetraborate dicarbonate + 5% acetonitrile (pH 9) to
85:15 in 8 min, then hold at 85:15 for 3 min; UV detection at
254 nm. Retention times: 5.45 min (imine 6c), 7.84 min
(toluene), 8.74 min (dialkylation product 11), 10.0 min (aza-
Cope product 14a and dimeric products).
Preparative SFC conditions: Chiralcel OJ-H column, 25 mm
× 3 cm; column temperature 35 °C; flow rate 70.0 mL/min;
pressure 100 bar; isocratic 5% 2-propanol in CO2; UV detection
at 290 nm. Retention times: 3.7-4.4 min (desired enantiomer
(1R,2S)-11), 4.6-5.4 min (undesired enantiomer (1S,2R)-11).
(1R,2S)-1-Amino-2-vinylcyclopropanecarboxylic Acid Eth-
yl Ester Tosylate Salt (4). A solution of dialkylation product
(1R,2S)-11 (2.0 g, 93 wt %, 7.64 mmol, 1.0 equiv) in 2-propanol
(6 mL) was treated with TsOH ·H2O (1.89 g, 9.94 mmol, 1.3
equiv) in one portion. The mixture was stirred for 10 min, and
then the solvent was switched to isopropyl acetate by rotary
evaporation, and the mixture was concentrated to a target
volume of 4 mL. n-Heptane (8 mL) was then added over 5
min, and the resulting slurry was aged 1 h. The slurry was
filtered, and the cake was washed with 2:1 n-heptane/isopropyl
acetate (1 × 4 mL) then n-heptane (1 × 4 mL), then dried under
vacuum for 4 h to give the title compound (2.28 g, 91%) as a
white powder. Mp 136-138 °C, [R]20D +30.7° (c 1.03, MeOH);
1H NMR (400 MHz, CDCl3) δ 8.62 (s, 3H), 7.74 (d, J ) 8.0
Hz, 2H), 7.17 (d, J ) 8.0 Hz, 2H), 5.60 (ddd, J ) 17.0, 10.5,
8.5 Hz, 1H), 5.19 (d, J ) 17.0 Hz, 1H), 5.09 (d, J ) 10.5 Hz,
1H), 4.15-4.03 (m, 2H), 2.58-2.53 (m, 1H), 2.36 (s, 3H), 1.93
(dd, J ) 10.0, 6.5 Hz, 1H), 1.53 (dd, J ) 8.0, 6.5 Hz, 1H),
1.18 (t, J ) 7.0 Hz, 3H); 13C NMR (100 MHz, CDCl3) δ 167.2,
141.3, 140.5, 131.7, 128.9, 125.9, 119.2, 62.3, 40.1, 30.4, 21.3,
19.4, 13.9; HRMS calcd for C8H14NO2 [MH]+ 156.1019, found.
156.1020.
Acknowledgment
We thank Zainab Pirzada, Wes Shafer, Tanja Brkovic, and
Margret Figus for assistance in developing all chiral assays.
We thank Lisa DiMichele for assistance in NMR characteriza-
tion. We also thank Albert Wieczorek for all pin-milling
operations and Laura Artino for all particle size measurements.
Supporting Information Available
Complete results table for the N-(fluoro-substituted)benzyl-
1
cinchonidinium catalyst screening, H and 13C NMR spectra
for catalysts 12b, 12e-12h, and 12p, dialkylation product
(1R,2S)-11, and tosylate salt 4, chiral SFC spectra for dialky-
lation product (1R,2S)-11, and particle size distribution com-
parison between powdered and pin-milled NaOH. This material
is available free of charge via the Internet at http://pubs.acs.org.
Received for review March 12, 2010.
OP100070D
700
•
Vol. 14, No. 3, 2010 / Organic Process Research & Development