Process development of (1S,2S,5R,6S)-spiro[bicyclo[3.1.0]hexane-2′, 5′-dioxo-2,4′-imidazolidine]-6-carboxylic acid, (R)-α- methylbenzenemethanamine salt (LSN344309)

Article abstract of DOI:10.1021/op049829e

Process development and a pilot-plant process for the synthesis of 4 and its resolution to obtain (1S,2S,5R,6S)-spiro[bicyclo[3.1.0]hexane-2′, 5′-dioxo-2,4′-imidazolidine]-6-carboxylic acid, (R)-α- methylbenzenemethanamine salt (5) are described. Starting from the inexpensive raw 2-cyclopenten-1-one and sulfur ylide 1 the racemic bicyclo keto ester 2 was synthesized. Reaction of 2 with potassium cyanide and ammonium carbonate under Buecherer-Berg's reaction conditions affords racemic 3 in 80% yield. Hydrolysis of 3 followed by the resolution with (R)-(+)-α- methylbenzylamine gave 4 in excellent yield and purity under optimized conditions. The improvement of the original discovery process to accommodate safety and environmental requirements for scale-up in manufacturing facilities is also discussed.

Full text of DOI:10.1021/op049829e

Organic Process Research & Development 2006, 10, 28−32
Full Papers
Process Development of (1S,2S,5R,6S)-
Spiro[bicyclo[3.1.0]hexane-2′,5′-dioxo-2,4′-imidazolidine]-6-carboxylic Acid,
(
R)-r-Methylbenzenemethanamine Salt (LSN344309)
,
†
†
†
†
‡
‡
Ossama M. Rasmy,* Radhe K. Vaid, Michael J. Semo, Erik C. Chelius, Roger L. Robey, Charles A. Alt,
‡
†
Gary A. Rhodes, and Jeffery T. Vicenzi
Chemical Product Research and DeVelopment, Eli Lilly and Company, Indianapolis Laboratories, Indiana 46285, and
Tippecanoe Laboratories, Indiana 47905, U.S.A.
Abstract:
hydantoin acid comprising four chemical steps with three
isolated intermediates is shown in Scheme 1.
Process development and a pilot-plant process for the synthesis
of 4 and its resolution to obtain (1S,2S,5R,6S)-spiro[bicyclo-
Evaluation of the processes before scale up revealed some
undesirable features such as: (a) emission of the air
pollutants dimethyl sulfide and ethyl bromoacetate during
the filtration in step 1 and (b) irreproducibility of the yield
and purity of 5 when acetone/water was used as a crystal-
lization medium.
Therefore, our development efforts were focused on
addressing these issues and improving processing.
Herein, we describe the results of these efforts.
[3.1.0]hexane-2′,5′-dioxo-2,4′-imidazolidine]-6-carboxylic acid,
(R)-r-methylbenzenemethanamine salt (5) are described. Start-
ing from the inexpensive raw 2-cyclopenten-1-one and sulfur
ylide 1 the racemic bicyclo keto ester 2 was synthesized.
Reaction of 2 with potassium cyanide and ammonium carbonate
under B u1 cherer-Berg’s reaction conditions affords racemic 3
in 80% yield. Hydrolysis of 3 followed by the resolution with
(R)-(+)-r-methylbenzylamine gave 4 in excellent yield and
purity under optimized conditions. The improvement of the
original discovery process to accommodate safety and environ-
mental requirements for scale-up in manufacturing facilities is
also discussed.
Results and Discussion
The bicyclic ketone 2 is readily synthesized by the
reaction of 2-cyclopentenone with either a diazoester or with
6,7
a sulfur ylide. Although use of a diazoester avoids the odor
and emission problems associated with the use of a sulfur
ylide and particularly the sulfur ylide generated from 1, the
safety issues associated with handling diazoesters preclude
their use in a commercial manufacturing setting. Thus,
process development focused on the use of a sulfur ylide.
In the initial discovery synthesis, the sulfonium salt 1
resulting from reaction of dimethyl sulfide and ethylbro-
moacetate was isolated as a crystalline solid. Both the
isolation of 1 and its use in step two pose significant odor
and volatile emission control problems. One attractive
solution to this problem is the use of a nonvolatile polymeric
sulfide in the formation of the sulfonium salt and corre-
sponding ylide in steps 1 and 2. However, all our attempts
to synthesize 2 using a polymeric sulfide in reaction with
ethylbromoacetate to prepare the desired sulfonium salt and
corresponding ylide were unsuccessful. Thus, efforts focused
on odor and emission minimization by the in situ generation
of 1. Ideally, by combining steps 1 and 2, all processing
involving the use or generation of dimethyl sulfide could be
conducted in a closed vessel. The dimethyl sulfide would
Introduction
LY544344 hydrochloride 6 is a new chemical entity under
investigation by Eli Lilly & Company as a potential treatment
of neurological or psychiatric disorders related to the
1
mammalian central nervous system (CNS). Compound 5 is
a key intermediate in the synthesis of 6. The original route
involved the synthesis of racemic hydantoin acid 4 followed
by resolution using (R)-(+)-(R)-methylbenzyylamine in
acetone/water (Scheme 1).2 The synthesis of the racemic
-5
*
To whom correspondence should be addressed. E-mail: rasmy_osama_m@
lilly.com.
†
Chemical Product Research and Development, Eli Lilly and Company.
Tippecanoe Laboratories.
‡
(
1) Bueno Melendo, A. B.; Coffey, D. S.; Dantzing A. H.; De Dios, A.;
Dominguez-Fernandez, C.; Merin, M.; Hillgren, K. M.; Martin, J. A.; Martin-
Cabrejas, L. M.; Martinez-Grau, M. A.; Massey, S. M.; Moher, E. D.; Monn,
J. A.; Montero Salgado, C.; Pedersen, S. W.; Pedregal-Tercero, C.; Sweetana,
S. A.; Valli, M. J. WO 02/055481, 2002; Chem Abstr. 2002, 137, 94004.
2) Helton, D. R.; Kallman, M. J.; Monn, J. A.; Schoepp, D. D.; Tizzano, J. P.
U.S. Patent 5,661,184, 1997; Chem. Abstr. 1997, 127, 205300.
3) Monn, J. A.; Valli, M. J.; Massey, S. M.; Wright, R. A.; Salhoff, C. R.;
Johnson, B. G.; Howe, T.; Alt, C. A.; Rhodes, G. A.; Robey, R. L.; Griffey,
K. R.; Tizzano, J. P.; Kallman, M. J.; Helton, D. R.; Schoepp, D. D. J.
Med. Chem. 1997, 40, 528-537.
(
(
(5) Schoepp, D. D.; Johnson, B. G.; Wright, R. A.; Salhoff, C. R.; Mayne, N.
G.; Wu, S.; Cockerham, S. L.; Burnett, J. P.; Belegaje, R.; et al.
Neuropharmacology 1997, 36, 1-11.
(6) Payne, G. B. J. Org. Chem. 1967, 32, 3351-3355.
(7) La Porta, E.; Piarulli, U.; Cardullo, F.; Paio, A.; Provera, S.; Seneci, P.;
Gennari, C. Tetrahedron Lett. 2002, 43, 761-766.
(
4) Monn, J. A.; Valli, M. J.; Massey, S. M.; Hansen, M. M.; Kress, T. J.;
Wepsiec, J. P.; Harkness, A. R.; Grutsch, J. L., Jr.; Wright, R. A.; Johnson,
B. G.; Andis, S. L.; Kingsston, A.; Tomlinson, R.; Lewis, R.; Griffey, K.
R.; Tizzano, J. P.; Schoepp, D. D. J. Med. Chem. 1999, 42, 1027-1040.
2
8
•
Vol. 10, No. 1, 2006 / Organic Process Research & Development
10.1021/op049829e CCC: $33.50 © 2006 American Chemical Society
Published on Web 12/01/2005

Scheme 1. Original process for the synthesis of LSN344309
Table 1. Solvent screening for the synthesis of 1 and 2
Table 2. Screening of bases for the formation of 2
a
combined steps
base
yield of 2 (%)
1
and 2
isolated
yield of 1 (%)
in situ yield
of 2 (%)
isolated yield
of 2 (%)
solvent
1,8-diazabicyclo[5.4.0]undec-7-ene
,1,3,3-tetramethylguanidine
1,5-diazabicyclo[4.3.0]non-5-ene
1,4-diazabicyclo[2.2.2]octane
N,N-diisopropylethylamine
triethylamine
77
77
52
0
0
0
1
a
acetone
acetonitrile
dichloromethane
ethyl acetate
tetrahydrofuran
toluene
methyl-tert-butyl ether
heptane
N,N-dimethylformamide
81.7
60.0
56.3
51.1
46.9
31.2
16.4
13.8
9.8
61.2
88.7
85.4
51.6
50.1
78.0
62.5
28.9
85.5
43.5
75.9
35.9
a
a
pyridine
potassium carbonate/TBAB
0
0
a
exo:endo (97:3).
a
exo:endo (97:3).
Table 3. Impacts of sequential addition times of DBU, and
2-cylopentenone on the yield of 3
then be removed by distillation and destructive vapor
incineration during reaction concentration and product isola-
tion at the end of step 2.
DBU
addition time
2-cyclopentenone
addition time (min)
yield of
3 (%)
(min)
A solvent screen was conducted to find a common solvent
for the formation of 1 and its subsequent use in step 2. The
results in Table 1 show that, while good yields of the
sulfonium salt 1 are obtained in acetone, acetonitrile, and
dichloromethane, the best combined step 1/step 2 process
yield was obtained with acetonitrile.
1
42
0
15
15
15
10
60
58.7
55.9
53.2
58.4
53.2
49.6
6
1
1
1
0
5
5
5
150
With the choice of acetonitrile made as the common
solvent for steps 1 and 2, various bases were screened for
their impact on the yield of 2. The data in Table 2 indicate
that 1,8-diazabicyclo[5.4.0]undec-7-ene (DBU) and 1,1,3,3-
tetramethylguanidine gave the best yield of 2 with a ratio of
exo/endo isomers of 97:3.
Thus, DBU was chosen as the base for the formation of
the ylide due to its commercial availability. The crude oil
of 2 was then purified by crystallization from heptane to
obtain a 66% yield of the pure exo isomer.
there was concern that omitting the purification of 2 could
affect the quality of 3, but when this approach was actually
tried, it performed adequately, and 3 was obtained in good
yield without compromising purity (98-99%).
An important critical process parameter was observed
during our process development wherein longer addition
times for either DBU or 2-cyclopentenone lowered the yield
of 3. Further investigation confirmed this observation as
indicated by the results in Table 3. Hence, the addition times
of the DBU and 2-cyclopentanone were held to 15-20 min
during the scale-up.
We believed omitting the isolation of intermediate 2 and
continuing processing to form 3 by solvent exchange at the
end of step 2 would improve process efficiency. However,
Vol. 10, No. 1, 2006 / Organic Process Research & Development
•
29

Scheme 2. Potential formation of 7 from 6 employing
chemistry for the synthesis of 4
Scheme 5. B u1 cherer-Berg’s Reaction for the synthesis of
9 and 10
Scheme 3. NMR study for the conversion of 6 to
2-cycopentenone by DBU
Table 4. Solvent screening study for B u1 cherer-Berg’s
reaction
Scheme 4. Formation of ketals
time for reaction
completion (h)
in situ ratio of
solvent system
MeOH
9 and 10 %
72
58
40
44
72
72
72
72
72
92:8
89:11
88:12
89:11
90:10
87:13
88:12
91:9
EtOH
EtOH/H
2
O (1:1)
O (1:1)
methanol/H O (1:1)
n-propanol/H
-propanol/H
DMF/H
2
Prior to the manufacture of 3, purchased 2-cylopenten-
2
1-one was found to contain 3% 3-cyclopenten-1-one. This
2
2
O (1:1)
O (1:1)
raised a concern for introduction of a new impurity such as
2
7
during B u¨ cherer-Berg’s reaction as shown in Scheme 2.
A literature survey revealed that compounds similar to 6
2
ethylene glycol/H O (1:1)
2
DMSO/H O (1:1)
83:17
undergo rearrangement of the double bond in the presence
8
9-13
of acid or strong base.
Therefore, we expected 6 would
of 1:1 ethanol/water at 60 °C for 5 h resulted in the formation
of 3 (a mixture of the hydantoin esters 9 and 10) in 80%
yield. The diastereoisomeric ratio of 9 to 10 was 78:22.
Recrystallization of this mixture from 2-propanol gave pure
also rearrange to 2-cyclopentenone in the presence of the
base DBU. A study by NMR involving treatment of 6 in
CD CN with a catalytic amount of DBU showed the
3
conversion of 6 to 2-cyclopentenone. A change in the
methylene and olefin proton resonances at δ 2.80 and 6.08
of 6 to the corresponding resonances in 2-cyclopentenone
are shown in Scheme 3. Furthermore, subjecting 6 to the
same reaction conditions for the conversion of 2-cyclopen-
tenone to 3 gave the expected quality and yield for 3.
Therefore, our concern was alleviated by these results.
Another concern was the formation of ketals of 8 during
the solvent exchange from tert-butyl methyl ether to 3A-
alcohol in preparation for step 3 as shown in Scheme 4. These
ketals did not undergo B u¨ cherer-Berg’s reaction to give 3.
A solution for this problem was achieved by the treatment
of slightly acidic solution of 2 in tert-butyl methyl ether with
aqueous sodium bicarbonate to remove traces of acid that
promote the formation of ketals before the solvent exchange.
The formation of hydantoins 9 and 10 were performed
using B u¨ cherer-Berg’s reaction conditions14 as shown in
Scheme 5. Thus, treatment of 2 with 1.1 equiv of potassium
cyanide and 2 equiv of ammonium carbonate in a mixture
9
in 55% yield.
To enhance the selectivity in B u¨ cherer-Berg’s reaction
in favor of 9, a solvent-screening study was conducted to
choose the solvent or solvent mixture that favors this
selectivity. Table 4 shows the results of the screen.
Slight selectivity was observed employing methanol,
methanol/water, and ethylene glycol/water. However a
mixture of ethanol/water was chosen because it shortens
reaction times. Optimal conditions were then obtained by
adjusting the ethanol/water ratio and the temperature as
shown by the results in Table 5.
The original resolution procedure for 5 utilized 1.0 equiv
of (R)-(+)-R-methylbenzylamine in 1.4:1 acetone/water. This
procedure initially worked well at laboratory scale; however,
the initial scale-up did not achieve the desired resolution,
and the product was difficult to filter. NMR and X-ray
crystallography analysis of the product indicated the product
is a mixture of racemic salts of 5, and they exist as a unique
crystal form in a double salt, rather than a mixture of the
two enantiomerically pure forms. It appeared that the racemic
double salt was the most thermodynamically stable crystal
form, and the desired resolution of 5 could be achieved only
under kinetic conditions. To overcome this problem a solvent
screen was conducted to find the right solvent combination
to provide the desired enantiomer salt of 5. The results are
shown in Table 6.
(
(
8) Collins, D. J.; Tomkins, C. W. Aust. J. Chem. 1977, 30, 433-450.
9) Collins, S.; Hong, Y.; Kataoka, M.; Nguyen, T. J. Org. Chem. 1980, 55,
3
395-3398.
10) Johnson, W. S.; Petersen J. W.; Gutsche, C. D. J. Am. Chem. Soc. 1947,
9, 2942-2955.
11) Hamanaka N.; Nakai, H.; Kurono, M. Bull. Chem. Soc. Jpn. 1980, 53, 2327-
329.
(
(
(
(
(
6
2
12) Fuzhenkova, A. V.; Galyautdinov, N.; Shaikhullina, R. F. J. Gen. Chem.
USSR (Engl. Transl.) 1983, 53, 1997-2003.
13) Brown, R. F.; Burge, G. L.; Collins D. J. Aust. J. Chem. 1983, 36, 117-
The data show that tetrahydrofuran/water (2:1) gave the
best enantioselectivity. Intermediate 5 was obtained in 42%
1
34.
14) Bucherer, H. T.; Lieb, V. A. J. Prakt. Chem. 1934, 141, 5-43.
30
•
Vol. 10, No. 1, 2006 / Organic Process Research & Development

Table 5. Impact of the ratio of ethanol/water on the ratio of
facturing scale. Also, a robust resolution process for the
synthesis of 5 was achieved and demonstrated at manufactur-
ing scale with good yield and excellent purity (>99% ee).
These processes addressed manufacturing concerns with
respect to safety, throughput, and quality.
a
9:10
reaction
temperature
(°C)
in situ
ratio of
9 and 10 (%)
reaction
time (h)
solvent system
EtOH/H
EtOH/H
EtOH/H
EtOH/H
EtOH/H
EtOH/H
EtOH/H
2
2
2
2
2
2
2
O (4:1)
O (3:2)
O (2:3)
O (1:4)
O (4:1)
O (4:1)
O (1:1)
25
25
25
25
35
40
60
24
48
48
72
16
6
88:12
86:14
81:19
81:19
87:13
87:13
78:22
Experimental Section
Melting points were obtained using a Thomas-Hoover
capillary melting point apparatus and are uncorrected. NMR
data were obtained at 300 MHZ using a Bruker instrument
with TMS as an internal standard. All reagents were
commercially available and were used without further
purification, unless otherwise stated. 3-Cyclopenten-1-one
was obtained from Wacker Chemicals. 1 and 2 were prepared
in dry reaction vessels in an atmosphere of dry nitrogen.
Reaction completion for 1, 2, 3, and 4 were monitored by
HPLC using a Hitachi L-7100 system equipped with a
Zorbax SB-Phenyl 25-cm column and the following condi-
tions: eluting solution, 80% 50 mM sodium phosphate
monobasic solution adjusted to pH 2.0 with o-phosphoric
acid and 20% acetonitrile; flow rate equal to 1 mL/min;
wavelength equal to 220 nm; and column temperature set at
5
a
This study indicated that a ratio of (4:1) ethanol to water at 40 °C gave the
optimum conditions in favor of 9 (9/10, 87:13) with reaction completion in 6 h.
The optimum condition for crystallization and isolation of 9 was achieved by
addition of water to adjust the ethanol/water ratio to 1:1. Isolated 3 is a mixture
of 9 and 10 in a 98:2 ratio. Hence, the overall yield for the combined three-step
process gave 3 in 55-58% yield.
Table 6. Solvent-screening study for optimization the purity
and yield of 5
solvent system (ratio)
yield of 5 (%)
ee (%)
tetrahydrofuran/water (1:1)
tetrahydrofuran/water (2:1)
tetrahydrofuran/water (3:1)
tetrahydrofuran/water (4:1)
acetonitrile/water (1:1)
36
42.4
45
46
42.0
28.5
99.5
99.9
93.0
97.2
98.2
96.6
4
0 °C. Chiral HPLC to determine chiral purity of 5 was run
under the following conditions: column: Daicel Chiralpak
AD, 25 cm × 4.6 mm column, 10 µm particle size; column
temperature: ambient; flow rate: 0.8 mL/min; detection
wavelength equal to 230 nm; injection volume 10 µL;
isocratic program with mobile phase of 35/65/0.2 (v/v/v) IPA/
hexane/TFA; sample preparation: 20 mg dissolved in 10 mL
of mobile phase; retention times: 5.4 min (1R,2R,5S,6R),
and 9.4 min 1S,2S,5R,6S (5).
2-propanol/water (1:1)
yield and enantomeric enhancement (ee) of 99.5%. The scale-
up of step 5 using this condition was robust for yield and
ee.
Thus, the original synthesis of 5 using four isolated
intermediates was modified to incorporate only one isolated
intermediate as shown in Scheme 6. The changes improved
overall efficiency and increased the yield from 48% to 55%.
Consolidated Synthesis of Racemic (1SR,2SR,5RS,6SR)-
Spiro[bicyclo[3.1.0]hexane-2′,5′-dioxo-2,4′-imidazolidine]-
6
-carboxylic Acid (4). (i) Synthesis of Carboethoxymethyl
Dimethylsulfonium Bromide (1). Acetonitrile (27 mL), ethyl
bromoacetate (24.4 g, 0.146 mol), and dimethyl sulfide (10.4
g, 0.17 mol) were added to a dry vessel at 10-15 °C and
stirred for 30 min. Carboethoxymethyl dimethylsulfonium
bromide 1 (5 mg) was added, and the reaction temperature
was adjusted to 20-25 °C to promote crystallization of 1.
The reaction mixture was stirred for 24 h to complete the
Conclusions
A consolidated process for the synthesis of racemic
(1SR,2SR,5SR,6SR)-bicyclo[3.1.0]hexane-2-spiro-5′-hydan-
toin-6-carboxylic acid (4) has been developed and demon-
strated for the preparation of LY544344‚HCl on a manu-
Scheme 6. Modified process for the synthesis of LSN344309
Vol. 10, No. 1, 2006 / Organic Process Research & Development
•
31

crystallization of 1. This slurry was used directly in the
synthesis of 2.
Synthesis of Racemic (1SR,2SR,5RS,6SR)-Spiro[bicyclo-
[3.1.0]hexane-2′,5′-dioxo-2,4′-imidazolidine]-6-carboxyl-
ic Acid (4). Water (120 mL) and sodium hydroxide solution
A portion of the slurry was filtered and washed with
acetonitrile to isolate 1 for physical data. 1: mp 42-44 °C
5
0% were added to 3 and stirred until hydrolysis completion
3
[Lit. mp 42-44 °C].
was confirmed by HPLC (approximately 4 h). The pH of
the solution was adjusted to a range between 1.5 and 1 with
concentrated hydrochloric acid and was stirred for 30 min
at 20-25 °C to promote the crystallization of 4. The resulting
slurry was cooled gradually to 0-5 °C and stirred for 1 h.
The product 4 was filtered, washed with cold water (57 mL),
and dried at 55-60 °C under vacuum. Overall yield is 57%
(ii) Synthesis of Racemic (1SR,5RS,6SR)-2-Oxobicyclo-
[3.1.0]hexane-6-carboxylic Acid Ethyl Ester (2). The slurry
of 1 (prepared above) was cooled to 0-5 °C, and acetonitrile
52 mL) was added. DBU (22.3 g, 0.146 mol) was then added
(
over 10-15 min, and the temperature was maintained
between 0 and 5 °C. The mixture was stirred until all the
crystals dissolved (10-15 min), and then 2-cyclopenten-1-
one (10 g, 0.122 mol) was added over 10-15 min, maintain-
ing the temperature between 0 and 5 °C. The reaction mixture
was warmed gradually to 20-30 °C over 30 min, and then
stirred for 8 h. Reaction completion was confirmed by HPLC
analysis. Water (50 mL) was added to quench the reaction,
and the pH of the media was adjusted to pH 6-8 with
concentrated sulfuric acid. Acetonitrile was removed by
vacuum distillation at 30-35 °C. The pH of the reaction
media was then adjusted to 1-1.5 with sulfuric acid. The
product 2 was extracted with 150 mL of tert-butyl methyl
ether. The resulting organic solution was washed with 20
mL of 5% aqueous sodium bicarbonate. The organic solution
was concentrated to an oil by vacuum distillation at 35-40
3
from 2-cyclopentenone. 4: mp 278-280 °C [Lit. mp 278-
2
80 °C].
Synthesis of (1S,2S,5R,6S)-Spiro[bicyclo[3.1.0]hexane-
′,5′-dioxo-2,4′-imidazolidine]-6-carboxylic Acid, (R)-r-
2
Methylbenzenemethanamine Salt (5). Compound 4 (10.0
g, 0.472 mol) was dissolved in THF (100 mL) and water
(48 mL) and then heated to 55-60 °C. (R)-(+)-R-Methyl-
benzylamine (4.04 g, 0.333 mol) was added over 20 min,
and then the mixture was stirred for 1 h before it was cooled
gradually to 0-5 °C. The reaction slurry stirred for 2 h before
it was filtered and washed with chilled aqueous THF (40
mL). The product was dried at 55-60 °C under vacuum to
give 40-44% yield of 5: mp 165-168 °C; IR (KBr) 3233,
-
1 1
°
C.
2
937, 1771, 1727, 1619, 1554, 1398, 1384, 1246 cm ; H
NMR (300 mHz, DMSO-d ) δ 1.32 (2H, m), 1.39 (3H, d, J
6.7), 1.68 (5H, m), 1.96 (1H, m), 4.19 (1H, q, J ) 6.6),
.26 (1H, t, J ) 7.1), 7.33 (2H, t, J ) 7.6), 7.43 (2H, d, J
(iii) Synthesis of Racemic (1SR,2SR,5RS,6SR)-Spiro-
6
[bicyclo[3.1.0]hexane-2′,5′-dioxo-2,4′-imidazolidine]-6-car-
)
7
boxylic Acid Ethyl Ester (3). Ethanol (66 mL) was added to
(prepared above) and stirred until it dissolved. Water (16
2
)
7.7), 8.06 (1H, br s); [R]
D
) -30.10° (c ) 1.01, H
O).
2
O),
mL), ammonium carbonate (19.9 g, 0.21 mol), and potassium
cyanide (7.43 g, 0.11 mol) were added to the solution of 2,
and the slurry was heated to 30-35 °C and stirred until
reaction completion was confirmed by HPLC. The reaction
slurry was cooled to 20-25 °C, and water (50 mL) was
added over 30 min to promote crystallization of 3. The
product slurry was stirred for 30 min, cooled gradually to
3
Lit. -27.00° (c ) 1.01, H
2
Acknowledgment
We thank Lilly Analytical Laboratories for providing
analytical support for the project. We also thank Dr. D. Kjell
for useful suggestions, discussions, and review of the
manuscript.
0
-5 °C, and stirred for 1 h to maximize the crystallization
of 3. The crystals of 3 were filtered, washed with cold water
57 mL), and used as is in the following reaction. A portion
(
of 3 was dried at 40-45 °C under vacuum for determination
of the melting point. 3: mp 218-221 °C [Lit. 219-221
Received for review September 14, 2004.
OP049829E
3
°
C].
32
•
Vol. 10, No. 1, 2006 / Organic Process Research & Development