Resolution of racemic 1-benzyl-5-oxo-3-pyrrolidinecarboxylic acid with enantiopure (S)-phenylalanine N-benzylamide via diastereomeric salt formation

Article abstract of DOI:10.1016/j.tetasy.2007.11.030

The resolution of racemic 1-benzyl-5-oxo-3-pyrrolidinecarboxylic acid 1, a potent chiral synthon with high pharmacological activity, was investigated with a variety of basic chiral resolving agents via diastereomeric salt formation. The findings in the optimization of resolution conditions aiming at an industrial-scale production revealed that (S)-phenylalanine benzylamide (S)-10 and 2-propanol containing ca. 4 mol % of water to (RS)-1 were the best conditions for obtaining enantiopure less-soluble diastereomeric salt, (S)-1/(S)-10/0.5H₂O (81%, 98% de, E 79%). X-ray crystallographic analysis of the salt clearly revealed that water molecules play a key role in crystallizing the enantiopure salt crystals, while stabilizing the crystal structure via three types of hydrogen-bond network associated with water molecules in addition to usual acid-base hydrogen bond.

Full text of DOI:10.1016/j.tetasy.2007.11.030

Available online at www.sciencedirect.com
Tetrahedron: Asymmetry 18 (2007) 2869–2875
Resolution of racemic 1-benzyl-5-oxo-3-pyrrolidinecarboxylic acid
with enantiopure (S)-phenylalanine N-benzylamide via
diastereomeric salt formation
Masao Morimoto,a,* Atsushi Yamakawa,a Hiroshi Katagirib and Kenichi Sakaia,*
aSpecialty Chemicals Research Laboratory, Toray Fine Chemicals Co., Ltd, Minato-ku, Nagoya, Aichi 455-8502, Japan
bGraduate School of Science and Engineering, Yamagata University, 4-3-16 Jonan, Yonezawa, Yamagata 992-8510, Japan
Received 3 September 2007; accepted 19 November 2007
Abstract—The resolution of racemic 1-benzyl-5-oxo-3-pyrrolidinecarboxylic acid 1, a potent chiral synthon with high pharmacological
activity, was investigated with a variety of basic chiral resolving agents via diastereomeric salt formation. The findings in the optimization
of resolution conditions aiming at an industrial-scale production revealed that (S)-phenylalanine benzylamide (S)-10 and 2-propanol
containing ca. 4 mol % of water to (RS)-1 were the best conditions for obtaining enantiopure less-soluble diastereomeric salt, (S)-
1/(S)-10/0.5H2O (81%, 98% de, E 79%). X-ray crystallographic analysis of the salt clearly revealed that water molecules play a key role
in crystallizing the enantiopure salt crystals, while stabilizing the crystal structure via three types of hydrogen-bond network associated
with water molecules in addition to usual acid–base hydrogen bond.
Ó 2007 Elsevier Ltd. All rights reserved.
1. Introduction
be a useful chiral building block in synthetic organic
chemistry. Although it is known that (RS)-1 can be
produced effectively, only a few methods of producing
enantiopure 1 have been reported, such as enzymatic
stereoselective hydrolysis of its methyl ester,5 resolution
of its amide by HPLC8 and diastereomeric resolution of
(RS)-1 with cinconine.9 Unfortunately, these methods are
considered difficult to apply to an industrial-scale produc-
tion from economical and scale-up technical points of
view.
It is well known that a number of pyrrolidinone com-
pounds are useful key intermediates for various types of
drugs, such as antidepressants [Rolipram, (R)-2]1, anti-
hypertensive [Cromakalim, (S)-3],2 or muscarinic anta-
gonist [Oxotremorine, 4].3 Various researches have been
reported which show how to synthesize such func-
tionalized pyrrolidinone compounds4 and to clarify
their SARs (structure–activity relationships).5 Among
them, especially, enantiopure 1-benzyl-5-oxo-3-pyrrolidine-
carboxylic acid (S)-1 is the most likely one to
exhibit interesting pharmacological activity because the
oxo-analog, enantiopure paraconic acid (S)-5,6 acts as
an inducer of cytodifferentiation in many streptomycetes
(Fig. 1).
To establish an industrial production process for
preparing enantiopure 1, we have investigated the
resolution of (RS)-1 with chiral basic resolving agents.
As a result, we found that (S)-phenylalanine benzylamide
(S)-10 is the most preferable resolving agent for the
resolution of (RS)-1, and the water molecule contained
in the solvent plays a crucial role in crystallizing the
enantiopure less-soluble salt. We herein report a new
efficient preparation method of enantiopure 1 by resolu-
tion via diastereomeric salt formation suited for the
industrial-scale production, and discuss the molecular
mechanism of the resolution reaction on the basis of the
X-ray crystallographic analysis of less-soluble diastereo-
meric salt.
(RS)-1 is commonly prepared via the Michael reaction of
itaconic acid with benzylamine.5,7 The carboxyl group
on the pyrrolidinone ring of 1 can be easily converted
into the corresponding ester, amide, or hydroxyl group.
For this reason, therefore, compound 1 is considered to
*
Corresponding authors. E-mail: masao_morimoto@tfc.toray.co.jp
0957-4166/$ - see front matter Ó 2007 Elsevier Ltd. All rights reserved.
doi:10.1016/j.tetasy.2007.11.030

2870
M. Morimoto et al. / Tetrahedron: Asymmetry 18 (2007) 2869–2875
CO2H
CO2H
CO2H
O
O
O
N
O
N
CH2Ph
CH2Ph
(
S)-5
rac-1
(
S)-1
OMe
O
O
N
O
N
NC
OH
N
HN
O
O
4
(
S)-3
(R)-2
Figure 1.
2. Results and discussion
2.1. Screening of resolving agents
of the salt crystals such as shape, rigidity, and thickness.
And it is true that appearances vary remarkably, depending
on the resolving agent used. On the basis of this empirical
knowledge, the salt crystal with (S)-10 was the best in terms
of separability of the salt from mother liquor. Thus, the
resolution condition of (RS)-1 with (S)-10 in 2-propanol
was optimized in detail.
Racemic 1 was prepared according to the method proposed
by Felluga et al.5 and was used for resolution experiments
as a starting material. First, 15 commercially available ba-
sic resolving agents classified in seven (7) types of basic
structures, such as enantiopure (a) alanine derivatives, (b)
phenylalanine derivatives, (c) pyrrolidine derivatives, (d)
aminocyclohexanol derivatives, (e) 1-phenylethylamine
derivatives, (f) aromatic amines, and (g) aliphatic diamines.
The structures of the resolving agents used in the experi-
ments are illustrated in Figure 2. Screening of resolving
agents was performed using approximately 2 mmol-scale
of (RS)-1, an equivalent amount of resolving agent, and
1 mL of pure or mixed solvent (water, methanol, or 2-pro-
panol). As a result, 2-propanol was chosen as a universal
solvent for the resolution experiments, judging from the
solubility and appearance of the obtained salt, whereas
no crystal was precipitated due to the excessive solubility
of the intended salt in the solvents when methanol or water
was used as a solvent. The screening results using 2-propa-
nol as a solvent are summarized in Table 1.
2.2. Optimization of the solvent for the resolution of (RS)-1
with (S)-10
Resolution of (RS)-1 with (S)-10 in 2-propanol was opti-
mized to obtain enantiopure 1. The test results are summa-
rized in Table 2. As shown in the Table 2, the enantiomeric
purity of the salt obtained from pure 2-propanol was as
low as 29% de [(S)-enantiomer rich, entry 1]. We tried to
change the solvent property, by adding small amounts of
water into 2-propanol.10 As a result, it was surprisingly
found that the enantiomeric purity drastically increased
while the resolution efficiency (E)11 remained nearly
unchanged in a range of water contents between 0.6 and
3.7 mol %, and large rigid and clear salt crystals containing
(S)-1 with higher enantiomeric purity (98% de) were
obtained when 3.7 mol % of water was added to (RS)-1
(entry 6). This result clearly indicates that water molecule
leads the chiral recognition as observed in our previous
studies.10,12 Resolved (S)-1 in the diastereomeric salt
obtained could be easily isolated in the usual manner,
where cleavage with alkaline water and extraction with
solvent were carried out, followed by acidification and
filtration to give enantiopure acid (S)-1 without any deteri-
oration of the enantiomeric purity of the diastereomeric
salt. To determine the role of the water molecule, the crys-
tal structure of the less-soluble diastereomeric salt was
determined by X-ray crystallographic analysis.
As shown in Table 1, fine salt crystals were precipitated
with (S)-alanine benzylamide (S)-6 (entry 1), (S)-phenylal-
anine benzamide (S)-9 (entry 4), and (S)-phenylalanine
benzylamide (S)-10 (entry 5), while no crystal was precipi-
tated or scaled solids were obtained with the other resolv-
ing agents. These test results indicate that the ‘aromatic’
group of the effective resolving agents 6, 9, and 10 plays
an extremely important role for realizing salt crystalliza-
tion. Furthermore, these phenomena obviously suggest
that the critical parameter of salt crystallization would
not be the main backbone with a stereogenic center but
the moiety of the aromatic amide such as a benzamide or
benzylamide (entries 1, 4, and 5). On the other hand,
solid–liquid separability of the salt crystals from mother
liquor is also an extremely important factor for indus-
trial-scale production and is affected by the appearances
2.3. Role of water in resolution: crystal structure of the less-
soluble diastereomeric salt
X-ray crystallographic analysis of the less-soluble salt
revealed that molar ratios of substrate, resolving agent,

M. Morimoto et al. / Tetrahedron: Asymmetry 18 (2007) 2869–2875
2871
NH2
NH2
NH2
H
N
NH2
NHCH3
O
(
O
O
(
S
)-7
(
S)-8
S)-6
NH2
NH2
H
N
H
N
O
O
(
S)-10
(
S)-9
OH
NH2
OH
O
N
N
N
H
N
H
(S)-14
(
S)-13
(
S)-12
(S)-11
NH2
NH2
NH2
H3C
N
Cl
H
(R)-18
(R)-17
(R)-16
(1R,2R)-15
NH2
H2N
NH2
(
S)-20
(R)-19
Figure 2. Resolving agents.
and water were 1:1:0.5 [(S)-1/(S)-10/0.5H2O]. This result
was also supported by the elemental analysis of the salt.
The crystal data and structure of the less-soluble salt are
shown in Table 3 and Figure 3, respectively.
action between two types of phenyl rings in one (S)-10 mol-
ecule, and CH–CH intermolecular interaction between two
phenyl rings of adjacent (S)-10 molecules, CH–p intermo-
lecular interaction between phenyl ring of (S)-1 molecule
and the phenyl ring of (S)-10 molecules, CH–CH inter-
molecular interaction between benzylic methylene of (S)-1
molecule and phenyl ring of (S)-10 molecules. Further-
more, it is thought that the van der Waals interactions were
closely related to the stability of the crystal formation.
Figure 3, as seen from the a-axis, demonstrates that the
less-soluble salt crystallizes in space group P1, crystal sys-
tem ‘triclinic’ with Z = 1. The unit lattice consists of four
molecules of (S)-10 (resolving agent), four molecules of
(S)-1, and two molecules of water. All the phenyl rings of
(S)-1 and (S)-10 located on the [001] and [002] planes form
hydrophobic space which stabilizes the outer surface along
+ and ꢀ sides of the c axis. Although we did not measure
the interaction lengths between all of the atoms in detail,
there is no doubt that the crystal consists of not only
hydrogen bonds between some functional groups, but also
a variety of interactions, such as p–p intramolecular inter-
The X-ray crystal structure analysis reveals that the car-
boxyl group of (S)-1 strongly bonds to the amino group
of the resolving agent based on a common acid–base inter-
action, NH–O hydrogen bond. In addition, the crystal con-
tains three types of hydrogen bonds associated with water
molecules, (a) O(water)–HN(10), (b) OH(water)–O@C(10),
and (c) OH(water)–O@C(1). It is clear that these

2872
M. Morimoto et al. / Tetrahedron: Asymmetry 18 (2007) 2869–2875
Table 1. Screening results of resolving agents in 2-propanol
Entry
Main structure
Resolving agent for (RS)-1
Solubility
Crystallization
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
Alanine
Phenylalanine
(S)-Alanine N-benzylamide 6
(S)-Phenylalanine amide 7
A
B
a
NA
c
a
a
c
c
b
c
c
(S)-Phenylalanine N-methylamide 8
(S)-Phenylalanine N-benzamide 9
(S)-Phenylalanine N-benzylamide 10
(S)-3-Hydroxypyrrolidine 11
(S)-1-Benzyl-3-hydroxypyrrolidine 12
(S)-1-Benzyl-3-aminopyrrolidine 13
(R)-3-Benzyloxypyrrolidine 14
(1R,2R)-2-Cyclopropylaminocyclohexanol 15
(R)-1-Phenylethylamine 16
A
A
A
A
A
A
A
A
A
A
A
A
A
Pyrrolidine
Aminocyclohexanol
1-Phenylethylamine
b
b
b
c
(R)-1-(4-Chlorophenyl)ethylamine 17
(R)-1-(4-Methylphenyl)ethylamine 18
(R)-1-Phenyl-3-butylamine 19
Aromatic amine
Aliphatic diamine
(S)-1,2-Diaminopropane 20
c
Solubility = A: dissolved at less than 65 °C, B: not dissolved. Crystallization = a: crystallized, b: scaled, c: not crystallized, NA: not applicable.
Table 2. Resolution of (RS)-1 with (S)-10 in 2-propanol with or without
Table 3. Crystal data and structure refinement for (S)-1/(S)-10/0.5H2O
water
Empirical formula
4(C16H18N2O)ꢂ4(C12H13NO3)ꢂ2H2O
1930.26
Entry
Water/1
(mol %)
Yield
(%)
Diastereomeric
excess (% de)
Resolution
efficiencya (E, %)
Formula weight
Temperature
93 K
˚
1
2
3
4
5
6
7
0.0
0.6
1.0
1.4
1.8
3.7
5.9
96
107
95
95
81
80
6
29
69
80
84
95
98
86
28
74
76
80
77
78
5
Wavelength (A)
1.54187
Triclinic
P 1
a = 11.486 (8) A, a = 108.65 (5)°
b = 14.015 (10) A, b = 90.72 (5)°
Crystal system
Space group
Unit cell dimensions
˚
˚
˚
c = 18.328 (13) A, c = 114.06 (4)°
2517(3) A
3
˚
Volume
Z
1
a Resolution efficiency (E, %) = yield (%) ꢁ diastereomeric excess (%
de) ꢁ 2/100.
Density (calculated) (g/cm3) 1.273
Absorption coefficient
(mmꢀ1
0.703
)
F(000)
1028
interactions contribute critically to the formation of three-
dimensional networks in the hydrophilic layer and stabilize
the crystal structure. Accordingly, it was concluded that
the crystal structure of the less-soluble salt is stabilized
by the supraspace formed between the [001] and [002]
planes, composed of the four types of hydrogen bonds,
the acid–base interaction based on NH–O hydrogen bond
between the carboxyl group of (S)-1 and the amino group
of (S)-10, and three hydrogen bonds via water molecules as
mentioned above.
Crystal size (mm3)
Theta range for data
collection (°)
0.10 ꢁ 0.05 ꢁ 0.05
3.68–67.50
Index ranges
ꢀ13 6 h 6 13, ꢀ16 6 k 6 16,
ꢀ21 6 l 6 21
Reflections collected
Independent reflections
Completeness to
theta = 67.50° (%)
Absorption correction
15,128
15,128 [R(int) = 0.0000]
93.4
Empirical
Max. and min. transmission 0.9657 and 0.9330
Refinement method
Data/restraints/parameters
Goodness-of-fit on F2
Final R indices
Full-matrix least-squares on F2
15,128/9/1280
3. Conclusion
0.888
R1 = 0.1189, wR2 = 0.2616
[I > 2sigma(I)]
The resolution of racemic 1-benzyl-2-oxo-3-pyrrolidine
carboxylic acid (RS)-1 via diastereomeric salt formation
was investigated aiming at an industrial production. The
findings in the optimization of the resolution parameters
for obtaining (S)-1 were that (S)-phenylalanine benzyl-
amide (S)-10 and 2-propanol containing water (ca.
4 mol %) are the most suitable combination of resolving
agent and solvent, respectively. The X-ray crystallographic
analysis of the less-soluble salt revealed that water mole-
cules play a key role in forming stable rigid salt crystals
with high enantiomeric purity. The three types of hydrogen
bonds associated with water molecules and the usual acid–
base hydrogen bond between molecules 1 and 10 form
three-dimensional hydrophilic supraspace.
R indices (all data)
Largest diff. peak and hole
R1 = 0.2200, wR2 = 0.3291
0.003 and ꢀ0.003
ꢀ3
˚
(e A
)
4. Experimental
4.1. General
Racemic 1-benzyl-2-oxo-3-pyrrolidine carboxylic acid
(RS)-1 (CP > 99% by HPLC) was prepared according to
the conventional method5 using the Michael reaction of ita-
conic acid with benzylamine. All the basic resolving agents

M. Morimoto et al. / Tetrahedron: Asymmetry 18 (2007) 2869–2875
2873
Figure 3. Crystal structure of (S)-1/(S)-10/0.5H2O viewed along the a-axis. Red balls and dotted lines indicate water molecules and hydrogen bonds,
respectively.
used were prepared according to the methods disclosed in
the patents13 and had high enantiomeric purities with over
99% ee or 99% de.
2.68–2.82 (m, 3H); Anal. Calcd for C28H31N3O4ꢂ0.5H2O:
C, 69.69; H, 6.68; N, 8.71. Found: C, 69.68; H, 6.66; N,
8.70.
X-ray crystallographic analysis was performed on a Rigaku
The less-soluble salt obtained was recrystallized from
2-propanol containing 3.7 mol % of water to the amine
component of the salt. The filtrated crystal was treated with
toluene and aqueous sodium hydroxide. The resulting
aqueous layer was separated and extracted with tetra-
hydrofuran under acidic conditions with sulfuric acid.
The upper layer (oil) was separated and concentrated to
give the crystalline (S)-1 with 99% ee.
R-axis Rapid diffractometer with Cu Ka radiation
˚
(k = 1.54187 A) at 93 K. Optical rotations were measured
with a JASCO DIP-140 polarimeter. Melting points were
obtained on a YAMATO apparatus MODEL MP-21. IR
spectrum was measured on a PERKIN ELMER SYSTEM
2000 spectrometer in KBr pellet. 1H NMR spectrum was re-
corded on a JEOL JNM-AL400 spectrometer (400 MHz for
a proton), using CDCl3 as solvent and tetramethylsilane as
the internal reference. Elemental analysis was run on Perkin
Elmer CHNS/O 2400II Analyzer. Enantiomeric purity (dia-
stereomeric excess, de %) of the salt was directly determined
25
(S)-1: ½aꢃD ¼ þ15:5 (c 0.5, EtOH); 99% ee; mp 100 °C (lit.5a
99–100 °C); IR (KBr) cmꢀ1:3409, 2923, 2882, 2800, 2495,
1907, 1709, 1632, 1497, 1465, 1455, 1438, 1426, 1359,
1332, 1263, 1202, 1177, 1110, 1070, 1040, 1005, 975, 931,
917, 888, 852, 820, 805, 759, 703, 677, 612; 1H NMR
(CDCl3, 400 MHz): 7.60–7.20 (m, 5H), 4.53 (d,
J = 14.4 Hz, 1H), 4.40 (d, J = 14.4 Hz, 1H), 3.45–3.60
(m, 2H), 3.15–3.30 (m, 1H), 2.79 (dddd, J = 8.0 Hz,
8.0 Hz, 8.0 Hz, 8.0 Hz, 2H).
by HPLC using
a
CHIRALPAK AD-RH (ID
4.6 mm ꢁ 150 mm, DAICEL). Analytical conditions for
the HPLC were as follows; mobile phase; aqueous phospho-
ric acid (pH 2)/acetonitrile = 84/16 (v/v), detector: UV
(220 nm), column temperature: 40 °C, elution rate: 0.5
(mL/min.). Retention times; (R)-enantiomer 15.2 min, (S)-
enantiomer 16.8 min.
4.3. X-ray crystallographic structure analysis (Fig. 4 and
Table 3)
4.2. Resolution of (RS)-1 with (S)-10
A typical resolution procedure is as follows. To a 100 mL-
flask were added 5.02 g of (RS)-1 (22.9 mmol), 5.80 g of
(S)-10 (22.8 mmol), 32.7 g of 2-propanol, and 1.5 g of
water (3.7 mol % to (RS)-1) followed by heating to obtain
a homogeneous solution at 65 °C. After cooling, the pre-
cipitated diastereomeric salt was filtrated and washed with
2-propanol (80.6% on the basis of (S)-enantiomer con-
tained in the fed (RS)-1, 98% de, E = 79%). Analytical data
Single crystals of (S)-1/(S)-10/0.5H2O [4(C16H18N2O)ꢂ
4(C12H13NO3)ꢂ2(H2O), MW = 1930.26] suitable for X-ray
analysis were grown by slowly cooling a hot 2-propanol/
water (0.89 g/0.07 g) solution of (S)-1/(S)-10/0.5H2O
(100 mg) to ambient temperature, and a single colorless
crystal with dimensions 0.10 ꢁ 0.05 ꢁ 0.05 mm was selected
for intensity measurements. The unit cell was triclinic with
the space group P1 with Z = 1. Crystal data and structure
refinement for (S)-1/(S)-10/0.5H2O are summarized in
Table 3. Crystallographic data for (S)-1/(S)-10/0.5H2O
have been deposited with Cambridge Crystallographic Data
Centre as supplementary publication number CCDC
666198. The ORTEP drawing is shown in Figure 4.
for the recrystallized salt is as follows: (S)-1/(S)-10/
25
0.5H2O: ½aꢃD ¼ þ20:9 (c 10.0, MeOH); >99% de; mp
102–105 °C; IR (KBr) cmꢀ1: 3512, 3424, 3323, 3248,
3088, 3064, 3030, 2940, 2158, 1683, 1669, 1560, 1496,
1457, 1395, 1271, 1225, 1199, 1078, 951, 918, 853, 732,
697, 601; 1H NMR (CDCl3, 400 MHz): 7.60–7.20 (m,
15H), 4.71–4.37 (m, 4H), 3.73 (dd, J = 4.4 Hz, 9.2 Hz,
1H), 3.51–3.43 (m, 2H), 3.27 (dd, J = 14.0 Hz, 4.4 Hz,
1H), 3.19 (dddd, J = 8.0 Hz, 8.0 Hz, 8.0 Hz, 8.0 Hz, 1H),
The absolute configuration was determined from the
known configuration of (S)-10. All calculations were
performed using the CrystalStructure crystallographic

2874
M. Morimoto et al. / Tetrahedron: Asymmetry 18 (2007) 2869–2875
Figure 4. ORTEP drawing of (S)-1/(S)-10/0.5H2O.
software package.14 The structure was solved by direct
methods using the program SHELXS-9715 and refined by
full-matrix least squares methods on F2 using SHELXL-
97.16 All non-hydrogen atoms were refined anisotropically.
References
1. Meyers, A. I.; Snyder, L. J. Org. Chem. 1993, 58, 36.
2. Bergmann, R.; Gericke, R. J. Med. Chem. 1990, 33, 492.
3. Nilsson, B. M.; Ringdahl, B.; Hacksell, U. J. Med. Chem.
1990, 33, 580.
4. (a) Rigo, B.; Fasseur, D.; Cherepy, N.; Couturier, D.
Tetrahedron Lett. 1989, 30, 7057; (b) Ma, D.; Jiang, J.
Tetrahedron: Asymmetry 1998, 9, 575; (c) Arvanitis, E.;
Motevalli, M.; Wyatt, P. Tetrahedron Lett. 1996, 37, 4277; (d)
Moody, C. M.; Young, D. W. Tetrahedron Lett. 1994, 35,
7277; (e) Thottathil, J. K.; Moniot, J. M.; Mueller, R. H.;
Wong, M. K. Y.; Kissick, T. P. J. Org. Chem. 1986, 51,
3140.
5. (a) Felluga, F.; Pitacco, G.; Prodan, M.; Pricl, S.; Visintin,
M.; Valentin, E. Tetrahedron: Asymmetry 2001, 12, 3241; (b)
Felluga, F.; Fermeglia, M.; Ferrone, M.; Pitacco, G.; Pricl,
S.; Valentin, E. Tetrahedron: Asymmetry 2002, 13, 475.
6. Crawforth, J. M.; Rawlings, B. J. Tetrahedron Lett. 1995, 36,
6345.
The very small crystal size and weak diffraction led to a low
2hmax for the data used in the refinement. The positions of
all the hydrogen atoms were calculated geometrically and
refined as a riding model. The hydrogen atoms of the water
molecules and the amino groups which could not be
located from difference Fourier maps. Since the number
of possible hydrogen moieties that could be involved was
quite large, no attempt was made to place the hydrogen
atoms at calculated positions.
Crystals of (S)-1/(S)-10/0.5H2O indicated twinning. Com-
bined data with complete or no overlap were used for
refinement, and the twinning prevented the merging of
equivalent reflections before refinement. The partially over-
lapped reflections were rejected, resulting in a low data
completeness value of 93.4%. The TwinRotMat procedure
in PLATON17 was used to generate a HKLF5 file containing
15,128 reflections.
7. Watanabe, S., Japan Patent Kokai, 112868, 1992; Chem.
Abstr. 1992, 117, 90156. The reaction of itaconic acid with
benzylamine was carried out at 130 °C for 2 h.
8. Irving, M.; Krueger, C. A.; Wade, J. V.; Hodges, J. C.;
Leopold, K.; Collins, N.; Chan, C.; Shaqair, S.; Shornikov,
A.; Yan, B. J. Comb. Chem. 2004, 6, 478.
9. Fles, D.; Ghyczy, T. Croat. Chem. Acta 1964, 26, 27.
10. Sakai, K.; Sakurai, R.; Yuzawa, A.; Hirayama, N. Tetrahe-
dron: Asymmetry 2003, 14, 3713.
Acknowledgment
11. Sakurai, R.; Yuzawa, A.; Sakai, K.; Hirayama, N. Cryst.
Growth Des. 2006, 6, 1606.
12. Saigo, K.; Sakai, K. In Chirality Drug Res.; WILEY-VCH,
GmbH & KGaA, 2006; Vol. 33, pp 127–154.
The authors greatly thank Professor Eiji Yashima of
Nagoya University, (Nagoya, Japan) for his valuable
suggestions.

M. Morimoto et al. / Tetrahedron: Asymmetry 18 (2007) 2869–2875
2875
13. (a) Alanine or phenylalanine derivatives: Nakai, S.; Sato,
H.; Fujino, T., Japan Patent Kokai, 59517, 1996; Chem.
Abstr. 1996, 125, 11463; (b) 3-Hydroxypyrrolidine: Ito, N.;
Sato, H., Japan Patent Kokai, 279326, 1993; Chem. Abstr.
1994, 120, 217268; (c) 1-Benzyl-3-hydroxypyrrolidine: Ito,
N.; Sato, H., Japan Patent Kokai, 13659, 1992; Chem.
Abstr.1996, 125, 11463; (d) 1-Benzyl-3-aminopyrrolidine:
Nakai, S.; Sato, H., Japan Patent Kokai, 124595, 1997;
Chem. Abstr. 1997, 127, 65681; (e) 3-Benzyloxypyrrolidine:
Morimoto, M.; Yamakawa, A., WO, 52578, 2007; (f)
Cyclopropylaminocyclohexanol: Takezaki, H.; Sato, H.,
Japan Patent Kokai, 139528, 2001; Chem. Abstr. 2001,
134, 366619; (g) 1-Phenylethylamine derivatives; Hasegawa,
S.; Sato, H., Japan Patent Kokai, 306942, 1990; Chem.
Abstr. 1991, 114, 184966; (h) 1,2-Diaminopropane; Nakai,
S.; Sato, H., Japan Patent Kokai, 18057, 1992; Chem. Abstr.
1992, 116, 255170.
14. (a) Crystal Structure, version 3.6.0: Crystal Structure Anal-
ysis Package, Rigaku and Rigaku/MSC, 9009 New Trails
Drive, The Woodlands, TX 77381, USA. (2000–2004); (b)
CRYSTALS, Issue 10: Watkin, D. J., Prout, C. K., Carru-
thers, J. R. and Betteridge, P. W., Chemical Crystallography
Laboratory, Oxford, UK. (1996).
15. Sheldrick, G. M. SHELXS-97: Program for the Solution of
Crystal Structures; University of Go¨ttingen: Go¨ttingen,
Germany, 1997.
16. Sheldrick, G. M. SHELXL-97: Program for the Refinement of
Crystal Structures; University of Go¨ttingen: Go¨ttingen,
Germany, 1997.
17. Spek, A. L. . J. Appl. Cryst. 2003, 36, 7.