Application of the asymmetric hydrogenation of enamines to the preparation of a beta-amino acid pharmacophore

Article abstract of DOI:10.1016/j.tetasy.2005.12.016

(3R)-3-[N-(tert-Butoxycarbonyl)amino]-4-(2,4,5-trifluorophenyl)butanoic acid 7a has been synthesized by an asymmetric hydrogenation of enamine ester 3 using chiral ferrocenyl ligands I and II in conjunction with [Rh(COD)Cl] ₂. The direct reduction of 3 provides amino ester 1b in 93% ee, which was isolated as an (S)-camphorsulfonic acid salt to upgrade the enantiomeric excess to >99%. A more concise approach was developed involving the in situ protection of 1b using di-tert-butyldicarbonate. This approach provided the desired N-Boc amino ester 7b directly from the hydrogenation with 97% ee, which was upgraded to >99% ee upon crystallization.

Full text of DOI:10.1016/j.tetasy.2005.12.016

Tetrahedron:
Asymmetry
Tetrahedron: Asymmetry 17 (2006) 205–209
Application of the asymmetric hydrogenation of enamines
to the preparation of a beta-amino acid pharmacophore
Michele Kubryk* and Karl B. Hansen
Department of Process Research, Merck Research Laboratories, PO Box 2000, Rahway, NJ 07065, USA
Received 26 October 2005; revised 30 November 2005; accepted 14 December 2005
Available online 23 January 2006
Abstract—(3R)-3-[N-(tert-Butoxycarbonyl)amino]-4-(2,4,5-trifluorophenyl)butanoic acid 7a has been synthesized by an asymmetric
hydrogenation of enamine ester 3 using chiral ferrocenyl ligands I and II in conjunction with [Rh(COD)Cl]2. The direct reduction of
3 provides amino ester 1b in 93% ee, which was isolated as an (S)-camphorsulfonic acid salt to upgrade the enantiomeric excess to
>99%. A more concise approach was developed involving the in situ protection of 1b using di-tert-butyldicarbonate. This approach
provided the desired N-Boc amino ester 7b directly from the hydrogenation with 97% ee, which was upgraded to >99% ee upon
crystallization.
Ó 2006 Published by Elsevier Ltd.
1. Introduction
ester 24 followed by a Mitsunobu reaction. Although this
approach has been successfully implemented on large
scale, a more direct synthesis of beta-amino acid deriva-
tives was desired. To this end, we recently developed a
method for the asymmetric reduction of enamines,
which provides unprotected beta-amino amides or esters
in high yield and enantioselectivity.5 This paper
describes the application of this methodology to the
enantioselective hydrogenation of enamine 3, conver-
sion of the resultant amine to its N-tert-butylcarbamate
and a crystallization method for the upgrade of the
Beta-amino acids possess biologically interesting prop-
erties and their enantioselective synthesis has been the
subject of considerable interest.1 Derivatives of beta-
homophenylalanine and in particular those of com-
pound 1 (Scheme 1)2 have recently been reported to be
potential new treatments for type II diabetes.3 Methods
have been developed in these laboratories to prepare
compounds such as 1 in enantiomerically pure form by
asymmetric reduction of the corresponding beta-keto
F
CO2R
NH2
F
1a R = H
1b R = Me
F
F
F
F
F
CO2Me
CO2Me
O
2
NH2
3
F
F
Scheme 1.
*
Corresponding author. E-mail: michele_kubryk@merck.com
0957-4166/$ - see front matter Ó 2006 Published by Elsevier Ltd.
doi:10.1016/j.tetasy.2005.12.016

206
M. Kubryk, K. B. Hansen / Tetrahedron: Asymmetry 17 (2006) 205–209
material to near enantiomeric purity. The application of
this methodology to the preparation of 1 results in a
more efficient and concise preparation of this
pharmacophore.
tic acid formed in situ since enamine formation is revers-
ible under acidic conditions. Solutions treated in this
fashion can be subjected to aqueous work-up to remove
inorganic salts, however in the case of 3, the methanol
solvent was removed and the mixture dissolved in ethyl
acetate to precipitate the excess ammonium acetate,
which was removed by filtration. Enamine 3 was then
crystallized by switching the solvent by distillation to
heptane and cooling to 0 °C. Enamine 3 was isolated
in 91% yield with >98 wt % purity and was used directly
in the following step.
2. Results and discussion
2.1. Synthesis of enamine 3
Enamine ester 3 was prepared in two steps starting from
2,4,5-trifluorophenyl acetic acid 4 (Scheme 2). Activa-
tion of carboxylate 4 by the method of Masamune with
carbonyldiimidazole (CDI) and reaction with the mag-
nesium enolate of mono-methyl potassium malonate
provide beta-keto ester 2. The enolization procedure
(triethylamine, magnesium chloride) was crucial to the
success of the reaction on large scale. Charging magne-
sium chloride to a mixture of triethylamine and the mal-
onate salt resulted in a thick slurry that was difficult to
stir and gave inconsistent assay yields of 2. Addition
of the amine base to a mixture of the malonate salt
and magnesium chloride resulted in a thinner slurry,
which was considerably easier to stir. The malonate slur-
ry was heated to 50 °C for 8 h to ensure complete depro-
tonation. Shorter age times resulted in incomplete
conversion of 4.
2.2. Asymmetric hydrogenation of enamine 3
Me
Me
PtBu2
PtBu2
PPh2
P(4-CF3-Ph)2
Fe
Fe
I
II
The direct asymmetric hydrogenation of enamines, such
as 3, is catalyzed by the metal complex generated in situ
from [Rh(COD)Cl]2 and chiral ferrocenyl ligands I or II.
Typically, the hydrogenation of enamine-ester sub-
strates requires the use of ligand II in conjunction with
trifluoroethanol as a solvent. Indeed, reduction with
0.1 mol % Rh-II at 40 °C and 200 psig H2 resulted in
the formation of the beta-amino ester 1b with 93% ee
and 89% yield (Table 1, entry 1).
The rate of addition of the CDI adduct to the enolate
was also found to be important. Rapid addition resulted
in the formation of large amounts of ketone 5, most
likely due to the concentration of the activated carboxyl-
ate building up in solution. Slow addition, over several
hours, gave consistently high assay yields of 2 with mini-
mal formation of 5 (<1 HPLC area%) after acid quench
and aqueous work-up.
Although the asymmetric hydrogenation yielded beta-
amino ester 1b with enantiomeric purity of 92–94%,
material of near enantiomeric purity (99% ee) was
desired for further use. Amino ester 1b was not observed
to crystallize at ambient temperatures and was also
found to decompose slowly. We wished to find a crystal-
line salt, which could allow for the isolation of 1b with
an upgrade in enantiomeric purity. After screening sev-
eral chiral acid salts, it was found that crystallization of
the (S)-camphorsulfonic acid (S-CSA) salt of 1b resulted
in efficient upgrade of its chemical and enantiomeric
purity with an overall yield of 70% starting from enam-
ine 3. The S-CSA salt 6 was converted to the N-Boc car-
boxylic acid 7a in a one-pot procedure by breaking the
salt and saponifying the ester with LiOH in THF/water,
followed by treatment with di-tert-butyldicarbonate
(Boc2O). Following aqueous work-up, the protected
amino acid was isolated by crystallization in 90% yield
with 97 wt % purity (Scheme 3).
F
F
O
F
F
5
F
F
Following aqueous work-up, keto-ester 2 could be iso-
lated by crystallization from isopropanol/water but
was typically converted directly to enamine 3 without
purification. The organic extracts of 2 were switched
into methanol by distillation and an excess of ammo-
nium acetate was charged. The mixture was heated at
reflux for several hours until complete conversion to 3
was observed.6 Upon cooling to room temperature, an
equivalent of NH4OH was charged to neutralize the ace-
F
F
F
CO2Me
CO2Me
i, ii
iii
CO2H
O
NH2
95%
91%
F
F
F
2
4
3
F
F
F
Scheme 2. Reagents and conditions: (i) 1-10-carbonyldiimidazole, CH3CN; (ii) mono-methyl potassium malonate, Et3N, MgCl2, CH3CN, 30 °C; (iii)
NH4OAc, MeOH, reflux.

M. Kubryk, K. B. Hansen / Tetrahedron: Asymmetry 17 (2006) 205–209
Table 1. Asymmetric hydrogenation of enamine ester 3
207
Entry
Ligand
Catalyst loading (mol %)
Solvent
Pressure (psig)
Yield (%)a
ee (%)
1
2
3b
II
I
0.1
1.0
0.6
TFE
MeOH
MeOH
200
90
90
89
80
88
93
93
97
I
a HPLC assay (LCAP).
b Reaction performed with 1.1 equiv of Boc2O at 25 °C.
F
F
F
i, ii
CO2Me
CO2Me
NH2
NH2
70%
S
-CSA
F
F
3
6
F
iii, iv
99%
F
F
v
CO2R
75%
HN
Boc
F
7a R = H
7b R = Me
Scheme 3. Reagents and conditions: (i) 0.1 mol % [Rh(COD)Cl]2, 0.1 mol % II, 200 psig H2, trifluoroethanol, 40 °C; (ii) (1S)-(+)-10 camphorsulfonic
acid, isopropanol; (iii) LiOH–H2O, THF, water; (iv) 1.1 equiv Boc2O, LiOH–H2O; (v) 0.6 mol % [Rh(COD)Cl]2, 0.6 mol % I, 1.1 equiv Boc2O,
90 psig H2, MeOH, 25 °C.
While our initial goal of applying asymmetric hydro-
genation to the synthesis of 1 had been achieved, several
aspects of the process could be improved. First, the
reduction conditions required the use of trifluoroethanol
as a solvent, which is considerably more expensive than
methanol. The catalyst derived from ligand I can be
used with methanol as a solvent, however, higher load-
ings of this catalyst were required to achieve yields com-
parable to the Rh-II/TFE system (Table 1, entry 2).
Additionally, the crystallization of 1b as its S-CSA salt
was necessary to upgrade its enantiomeric purity. A
more efficient process would involve a single isolation
of the desired N-Boc amino acid or ester directly with-
out the use of a chiral acid.
3. Conclusion
Two approaches to the enantioselective synthesis of beta
amino acid derivative 1 via asymmetric hydrogenation
of unprotected enamine 3 have been presented. Direct
hydrogenation of enamine 3 using ligand II in trifluoro-
ethanol affords the desired amino ester in 93% ee. While
this material can be elaborated to a protected and iso-
lable form of 1, a more concise approach is the in situ
protection of 1b as the desired N-Boc carbamate. The
use of Boc2O in the hydrogenation allows for methanol
to be used as a solvent and for the reaction temperature
to be significantly reduced. These conditions result in a
higher enantioselectivity for the reduction and the
N-Boc amino ester 7b can be isolated in high enantiomeric
and chemical purity directly from the hydrogenation
stream.
We recently reported that product inhibition occurs in
the asymmetric reduction of enamines and that it can
be largely eliminated by in situ protection of the product
with (Boc)2O.7 In addition to increasing the efficiency of
catalyst Rh-I, application of these conditions to the
reduction would avoid the use of trifluoroethanol as a
solvent, as well as provide the desired N-Boc amino ester
1b in a single step. Thus, reduction of 3 using 0.6 mol %
catalyst at 90 psig H2 at room temperature resulted in
90% conversion to 1b. The enantioselectivity under these
conditions increased to 97% upon reduction of the reac-
tion temperature. In contrast to amino ester 1b, carba-
mate 7b is a crystalline solid and was isolated directly
from the reaction mixture in 75% yield. The enantio-
meric purity of 7b was gratifyingly increased to >99%
when it was isolated in this fashion.
4. Experimental
4.1. General procedure
Melting points were determined on an open capillary
apparatus and were uncorrected. HPLC assays were car-
ried out using a C-18 reverse phase column eluted with
acetonitrile and 0.1% HClO4. Assay yields were deter-
mined by HPLC using pure compounds as standards.
Enantiomeric excess assays were determined by using a
Chiralpak AD-H column eluted with isopropanol and
n-heptane. All reagents and solvents were obtained from

208
M. Kubryk, K. B. Hansen / Tetrahedron: Asymmetry 17 (2006) 205–209
commercial suppliers and used without further purifica-
tion. All 1H NMR and 13C NMR data were recorded on
a 400 and 100 MHz spectrometer, respectively.
4.1.3. Methyl (3R)-3-amino-4-(2,4,5-trifluorophenyl)but-
anoate 1b. A hydrogenation vessel under a nitrogen
atmosphere was charged with [Rh(COD)Cl]2 (18.5 mg,
0.05 mol %) and Josiphos ligand II (51 mg, 0.1 mol %).
Enamine 3 (18.4 g, 0.075 mol) was dissolved in degassed
trifluoroethanol (150 mL) and charged to the hydro-
genation vessel and the entire mixture was degassed by
vacuum/N2 backfill cycles. The vessel was agitated vig-
orously for 10 min to dissolve the solids and was then
hydrogenated at 200 psig H2 at 40 °C for 24 h. The
hydrogenation stream yielded 16.7 g of 1b (90% assay
yield) with an enantiomeric excess of 93%. The enantio-
meric excess was determined by treating an aliquot of
the crude hydrogenation stream with an excess of 3,5-
dinitrobenzoylchloride. The reaction was aged for
30 min and analyzed via HPLC (chiralpak AD-H
column, 20% iPrOH/heptane, 1 mL/min, isocratic
for 20 min).
4.1.1. Methyl 4-(2,4,5-trifluorophenyl)acetoacetate 2.
A
solution of 2,4,5-trifluorophenyl acetic acid 4 (3.5 kg,
18.4 mol) in CH3CN (21 L) was charged to a 50 L flask
containing a slurry of 1,10-carbonyldiimidazole (3.28 kg,
20.2 mol) and CH3CN (9 L) over 0.5 h. The resulting
solution was aged for 3 h. To a separate 100 L flask,
mono-methyl malonate potassium salt (3.59 kg,
23 mol), Et3N (7.8 L, 55.2 mol) and CH3CN (42 L) were
charged followed by a portionwise addition of magne-
sium chloride (1.9 kg, 20.2 mol). The 100 L reaction
slurry was heated to 50 °C and aged for 8 h. The slurry
was cooled to 30 °C and was charged with the CDI
adduct over 2 h. The combined reaction mixture was aged
for 24 h at 30 °C. The reaction was cooled to 0 °C and
was charged with 35 L of 3 M HCl over 2 h. Methyl
tert-butyl ether (MTBE) (25 L) was charged and the
biphasic mixture was settled and the aqueous layer
discarded. The organic extracts were washed sequentially
with 1 M HCl, 7 wt % NaHCO3 and 24 wt % NaCl.
The MTBE organic extracts were switched into iPrOH
and concentrated to a final volume of 17 L upon which
the product crystallized. The slurry was warmed to
30 °C for complete dissolution and was cooled to
20 °C and seeded with 0.1 wt % of 2.8 Water (36 L)
was charged to the slurry over 6 h. The resulting suspen-
sion was filtered and the collected solids were dried
under vacuum (temperature <30 °C) to provide 2
(3.94 kg, 84%) as an off-white crystalline solid: mp:
4.1.4. Methyl (3R)-3-amino-4-(2,4,5-trifluorophenyl)but-
anoate (1S)-(+)-10 camphorsulfonic acid 6. A methanol
solution of 1b (18.5 g, 75 mmol) was switched into
iPrOH and concentrated to 75 mL. (1S)-(+)-10-Cam-
phorsulfonic acid (17.4 g, 82.5 mmol) was dissolved in
iPrOH (30 mL) and charged to 1b via an addition funnel
followed by a 10 mL iPrOH rinse. The cloudy solution
was seeded with 0.1 wt % of 6 and the title compound
crystallized out. The slurry was heated to 40 °C, aged
for 3 h and cooled to 20 °C. The solids were filtered,
washed with iPrOH and dried in a vacuum oven at
50 °C to provide 6 (25.9 g, 72%) as a white crystalline
solid with >99% enantiomeric excess. Mp: 137–138 °C;
[a] = +15.7 (c 1.02, CHCl3); 1H NMR (CD3OD) d
7.39 (m, 1H), 7.21 (m, 1H), 3.9 (m, 1H), 3.7 (s, 3H),
3.3 (s, 1H), 3.07 (m, 2H), 2.8 (s, 1H), 2.75 (s, 2H), 2.62
(m, 1H), 2.35 (m, 1H), 2.07 (s, 2 H), 1.90 (m, 1H),
1.66 (m, 1H), 1.42 (m, 1H), 1.11 (s, 3H), 0.86 (s, 3H);
13C NMR (DMSO) d 216.3, 170.4, 157.8, 155.4, 150.1,
147.6, 145.1, 120.1, 106.3, 58.4, 52.0, 47.7, 47.4, 47.2,
42.5, 42.4, 36.2, 31.1, 26.73, 24.5, 20.2, 19.8; 19F NMR
(DMSO) d ꢀ118.4, ꢀ136.4, ꢀ143.9; IR (KBr) 2953,
2097, 1748, 1633, 1522, 1443, 1211, 1048, 850,
818 cmꢀ1; Anal. Calcd for C21H28F3NO6S: C, 52.6; H,
5.89, N, 2.92. Found: C, 52.53; H, 5.93; N, 2.92.
1
40–41 °C; H NMR (CDCl3) d 7.08 (m, 1H), 6.94 (m,
1H), 3.85 (s, 2H), 3.77 (s, 3H), 3.55 (s, 2H); 13C NMR
(CDCl3) d 197.9, 167.1, 156.9, 154.5, 150.4, 147.9,
145.5, 119.4, 117.0, 105.5, 52.4, 48.3, 41.9; 19F NMR
(CDCl3) d ꢀ119.0, ꢀ134.7, ꢀ143.0; IR (KBr) 3096,
3054, 2945, 1738, 1635, 1522, 1440, 1343, 1236, 1214,
1157, 1091, 995, 948 cmꢀ1; Anal. Calcd for C11H9F3O3:
C, 53.67; H, 3.68. Found: C, 53.63; H, 3.36.
4.1.2. (Z)-Methyl 3-amino-4-(2,4,5-trifluorophenyl)but-2-
enoate 3. A 1 L flask was charged with 2 (50 g,
0.20 mol), ammonium acetate (78 g, 1.0 mol) and meth-
anol (500 mL).9 The reaction solution was refluxed for
2 h. The mixture was cooled to 25 °C and the solvent
was switched to EtOAc and concentrated to 100 mL.
Salts crystallized out during the solvent switch and were
filtered. The filtrate was concentrated to 100 mL and
heptane (100 mL) was charged. The solution was seeded
with 0.1 wt % of 3 and aged for 0.5 h. An additional
200 mL of heptane was charged to the seed bed and
the resulting suspension was cooled to 0 °C, filtered
and the collected solids were dried under vacuum to pro-
vide 3 (45.5 g, 91%) as a white crystalline solid: mp: 70–
4.1.5. (3R)-3-[N-(tert-Butoxycarbonyl)amino]-4-(2,4,5-
trifluorophenyl)butanoic acid 7a. To a 500 mL flask
was charged 6 (20 g, 41.7 mmol), THF (100 mL), water
(100 mL) and lithium hydroxide monohydrate (5.25 g,
125 mmol). The biphasic mixture was aged for 24 h.
Once the ester hydrolysis was complete, the mixture
was cooled to 0 °C and was charged with di-tert-butyldi-
carbonate (10 g, 45.9 mmol) and lithium hydroxide
monohydrate (1.75 g, 41.7 mmol). The reaction mixture
was warmed to ambient temperature and was stirred for
12 h. The reaction was worked-up by adjusting the pH
to 2.5 using 10 wt % NaHSO4 and extracting 7a twice
with iPrOAc. The organic extracts were combined,
switched into heptane and the final volume was concen-
trated to 200 mL at which point the product crystallized.
The resulting slurry was heated to 50 °C for 2 h and
cooled to 20 °C. The solids were collected via filtration
and washed with neat heptane to provide 7a (14 g,
1
71 °C; H NMR (CDCl3) d 7.08 (m, 1H), 6.94 (m, 1H),
4.55 (s, 1H), 3.62 (s, 3H), 3.4 (s, 2H); 13C NMR (CDCl3)
d 170.2, 159.4, 156.9, 154.5, 150.5, 148.0, 145.5, 119.8,
118.4, 105.5, 84.7, 50.1, 34.5; 19F NMR (CDCl3) d
ꢀ119.5, ꢀ135.0, ꢀ142.6; IR (KBr) 3490, 3340, 1666,
1559, 1507, 1424, 1167, 773 cmꢀ1; Anal. Calcd for
C11H10F3NO2: C, 53.88: H, 4.11; N, 5.71. Found: C,
53.83; H, 3.84; N, 5.53.

M. Kubryk, K. B. Hansen / Tetrahedron: Asymmetry 17 (2006) 205–209
209
95%) as a white crystalline solid; mp: 124–125 °C;
ꢀ119.7, ꢀ136.3, ꢀ143.6; IR (KBr) 3365.8, 3058.7,
2968.7, 1737.6, 1684, 1523.5, 1422, 1253.4, 1160.2,
852.9; Anal. Calcd for C16H20F3NO4: C, 55.33; H,
5.80; N, 4.03. Found: C, 55.52; H, 5.85; N, 3.96.
1
[a] = +32.3 (c 1.0, CHCl3); H NMR (CD3OD) d 7.18
(m, 1H), 7.07 (m, 1H), 4.15 (broad, 1H), 2.93 (dd,
J = 4.7 and 13.8 Hz, 1H), 2.68 (m, 1H), 2.51 (m, 2H),
1.34 (s, 9H), 13C NMR (CD3OD) d 173.1, 157.6,
156.0, 155.1, 149.9, 147.4, 145.1, 122.1, 119.0, 104.6,
78.5, 38.6, 32.9, 27.2, 26.9; 19F NMR (CD3OD) d
ꢀ121.0, ꢀ139.6, ꢀ146.7; IR (KBr) 3362, 2985, 1693,
1521, 1424, 1274, 1233, 1170, 1054, 840 cmꢀ1; Anal.
Calcd for C15H18F3NO4: C, 54.05; H, 5.44; N, 4.20.
Found: C, 54.02; H, 5.42; N, 4.16.
References
1. See reviews: (a) Abdel-Magid, A. F.; Cohen, J. H.;
Maryanoff, C. A. Curr. Med. Chem. 1999, 6, 955; (b) Cole,
D. C. Tetrahedron 1994, 50, 32; (c) Book: Enantioselective
Synthesis of b-Amino Acids; Juaristi, E., Ed.; Wiley-VCH:
New York, 1997; (d) Ma, J.-A. Angew. Chem., Int. Ed.
2003, 42, 4290–4299.
2. Xu, J.; Ok, H. O.; Gonzalez, E. J.; Colwell, L. F.;
Habulihaz, B.; He, H.; Leiting, B.; Lyons, K. A.; Marsilio,
F.; Patel, R. A.; Wu, J. K.; Thornberry, N. A.; Weber, A.
E.; Parmee, E. R. Biorg. Med. Chem. Lett. 2004, 14, 4759–
4762.
3. (a) Weber, A. J. Med. Chem. 2004, 48, 4135–4141; (b)
Drucker, D. J. Exp. Opin. Investig. Drugs 2003, 12, 87–100;
(c) Weideman, P. E.; Trevillyan, J. M. Curr. Opin. Investig.
Drugs 2003, 4, 412–420.
4. Angelaud, R.; Zhong, Y.-L.; Maligres, P.; Lee, J.; Askin, D.
J. Org. Chem. 2005, 70, 1949–1952.
5. Hsiao, Y.; Krska, S.; Rivera, N.; Rosner, T.; Njolito, E.;
Wang, F.; Sun, Y.; Armstrong, J.; Grabowski, E. J.; Tillyer,
R.; Spindler, F.; Malan, C. J. Am. Chem. Soc. 2004, 126,
9918–9919.
6. HPLC analysis of the reaction was performed under basic
conditions with potassium phosphate buffer and ACN as
the diluents since the reaction is reversible under acidic
conditions.
7. Hansen, K. B.; Rosner, T.; Kubryk, M.; Dormer, P.;
Armstrong, J. D. Org. Lett. 2005, 7, 4935–4938.
8. Seed was generated by removing an aliquot of the desired
compound, cooling to 0 °C and charging antisolvent until
the desired compound crystallized out of solution. The
slurry was then charged back to the crystallization solution.
9. Through process starting from 4.
4.1.6.
Methyl
(3R)-[(tert-butoxycarbonyl)amino]-4-
(2,4,5-trifluorophenyl)butanoate 7b. A hydrogenation
vessel under a nitrogen atmosphere was charged with
[Rh(COD)Cl]2 (91 mg, 0.3 mol %) and Josiphos ligand
I (206 mg, 0.6 mol %). Enamine 3 (15 g, 0.61 mol) was
dissolved in degassed methanol (150 mL) and charged
to the hydrogenation vessel. The hydrogenator was sha-
ked vigorously for 10 min to dissolve the solids and was
hydrogenated at 90 psig H2 at 30 °C for 24 h. The
hydrogenation stream yielded 18.5 g of 7b (87% assay
yield) with an enantiomeric excess of 97% (HPLC, chir-
alpak AD-H column, 10% isopropanol/heptane, 1 mL/
min, isocratic for 15 min). The reaction volume was con-
centrated to 130 mL of methanol and warmed to 45 °C
upon which 45 mL of water was charged to the solution.
Aminoester 7b crystallized and the slurry was cooled to
ambient temperature. The solids were collected via fil-
tration and washed with 3:2 methanol/water to provide
7b (16 g, 75% yield) as a fluffy off-white crystalline solid
with a purity of 100 wt % by HPLC and >99% enantio-
meric excess. Mp: 88–88.5 °C; [a] = +15.2 (c 1.0,
1
MeOH); H NMR (CDCl3) d 7.14 (m, 1H), 6.97 (m,
1H), 5.29 (m, 1H), 4.31 (m, 1H), 3.95 (s, 3H), 3.10 (m,
2H), 2.67 (m, 2H), 1.60 (s, 9H); 13C NMR (CDCl3) d
171.1, 154.9, 150.3, 147.5, 145.5, 121.2, 118.9, 105.3,
79.5, 51.7, 47.6, 37.1, 32.9, 28.1; 19F NMR (CDCl3) d