Crystal structures of (2-substituted-5-N-tosyl)bicyclo[3.3.0]-5-azacyclooct-2-enone: a pseudo achiral crystal from enantiopure compound and a counter-example of Wallach's rule

Article abstract of DOI:10.1016/j.tetasy.2009.07.013

An unusual crystallization affording an apparently pseudoachiral crystal from single enantiomers 1a and 1b was observed. In cases of enantiopure molecules such as 1a and 1b with structural flexibility, the tendency toward a centrosymmetric arrangement was

Full text of DOI:10.1016/j.tetasy.2009.07.013

Tetrahedron: Asymmetry 20 (2009) 1736–1741
Contents lists available at ScienceDirect
Tetrahedron: Asymmetry
journal homepage: www.elsevier.com/locate/tetasy
Crystal structures of (2-substituted-5-N-tosyl)bicyclo[3.3.0]-
5-azacyclooct-2-enone: a pseudo achiral crystal from enantiopure
compound and a counter-example of Wallach’s rule
Kyung Seok Jeong a, Dong Eun Kim a, Eunsung Lee b, Young Ho Jhon b, Hogyu Han a, Jaheon Kim b,
Nakcheol Jeong a,
*
a Department of Chemistry, Korea University, Seoul 136-701, Republic of Korea
b Department of Chemistry, Soongsil University, Seoul 156-743, Republic of Korea
a r t i c l e i n f o
a b s t r a c t
Article history:
An unusual crystallization affording an apparently pseudoachiral crystal from single enantiomers 1a and
1b was observed. In cases of enantiopure molecules such as 1a and 1b with structural flexibility, the ten-
dency toward a centrosymmetric arrangement was so overwhelming that apparently achiral crystals
were formed even by homochiral compounds via the formation of a pseudo-centrosymmetric asymmet-
ric unit accompanying structural adjustment. In addition, these crystals of 1a constitute as a counter-
example of Wallach’s rule.
Received 9 May 2009
Accepted 2 July 2009
Available online 17 August 2009
Ó 2009 Elsevier Ltd. All rights reserved.
1. Introduction
dency toward centrosymmetric arrangements is intensified partic-
ularly in the case of chiral molecules.3b,c In fact, most crystals
Molecular shape is a primary factor in determining the crystal
packing patterns, when there are no strong intermolecular forces,
such as ionic interactions and hydrogen bonds.1,2 In addition, chi-
rality is another important factor determining the final crystal
structure during the crystallization of chiral molecules. Crystalliza-
tion of chiral molecules can be classified according to the outcome
of crystal chirality and the input of chirality of the molecules sub-
jected to crystallization. Of the four possible manners of crystalli-
zation, all but one have been observed experimentally. Achiral
crystals from racemic mixtures and chiral crystals from enantio-
pure compounds are conceived as natural.3 In addition, as observed
by Pasteur, spontaneous resolutions of racemic mixtures providing
two kinds of crystals with opposite handedness were also reported,
although the cases of this are still rare.4 However, a well-defined
achiral crystal from enantiopure molecules is not allowed by an
intrinsic symmetry requirement. Herein, we report interesting
examples of isomorphic crystals from homochiral molecules with
those from racemic mixtures. These belong to the less explored
class of crystallization of chiral molecules: the formation of appar-
ently achiral crystals from single enantiomers.
formed from racemic mixtures (>90% of reported examples) are
found to have asymmetric units that adopt centrosymmetric
arrangements via enantiomeric pairing. On the other hand, homo-
chiral molecules usually afford crystals belonging to chiral space
groups and, since their symmetry requirement in crystallization
is different from that of racemic mixtures, the crystallization of
homochiral molecules is best understood case by case.5
During our study on crystallization and self-assembly, we ob-
served that crystals of compounds 1 constituted as distinctive
examples of pseudo achiral crystals from homochiral molecules.
This seemed to be motivated by a tendency to mimic a character-
istic feature of most racemates, a centrosymmetric arrangement,6
even with a substantial structural change.
2. Results and discussion
Compounds 1 were obtained either as an enantiomerically pure
form or as a racemic mixture by a Pauson–Khand reaction we have
developed (Scheme 1).7
Normally, organic materials provide thermodynamically stable
crystals by taking on centrosymmetric arrangements.2 This ten-
dency is one of the fundamental working principles in crystalliza-
tion as manifested by the estimated centro-/non-centrosymmetric
arrangement ratio of 4:1 found in all crystal structures.3 This ten-
R
R
chiral catalyst
Ts
N
Ts
N
O
*
H
2a. R = Ph
2b. R = H
1a. R = Ph
1b. R = H
* Corresponding author.
E-mail address: njeong@korea.ac.kr (N. Jeong).
Scheme 1. Synthesis of 1a and 1b by an asymmetric Pauson–Khand reaction.
0957-4166/$ - see front matter Ó 2009 Elsevier Ltd. All rights reserved.
doi:10.1016/j.tetasy.2009.07.013

K. S. Jeong et al. / Tetrahedron: Asymmetry 20 (2009) 1736–1741
1737
Figure 1. Crystal structures of crystal (RS)-1a and crystal (R)-1a. The unit cell origin has been moved to (0.25, À0.25, À0.25) in (b) for easy comparison with (a).
The respective crystals of 1a, (R)-1a and (RS)-1a, were obtained
from single enantiomers and a racemic mixture in ethyl acetate
with a diffusion of n-hexane. Their crystal structures were deter-
mined by a single-crystal X-ray diffraction study and their crystal
data are summarized in Table 1 and their structures are shown
in Figure 1.
ever, since the configuration at the N stereogenic center is
invertible at ambient temperature, the two diastereomers bearing
*
the same configuration at C are often indistinguishable, especially
in solution. Thus we use notations of (R)-1a and (S)-1a to highlight
*
the invariable absolute configuration at C in compound 1a. Com-
pound (R)-1a intrinsically denotes pair of diastereomers,
a
*
*
The crystal structure for (R)-1a indicates an interesting feature
that the asymmetric unit contains two independent molecules.
Two conformers of a single enantiomer were used to achieve the
pseudo-centrosymmetric arrangement. Thus, in order to under-
stand the situation more clearly, it is worthwhile looking at all
the possible stereoisomers of 1 and their structural relationship.
For a detailed description, it is necessary to define the notations.
In a strict sense, compound 1a has four possible stereoisomers be-
[(R(C ),R(N)]-1a (assigned as d1-(R)-1a) and [R(C ),S(N)]-1a (d2-
(R)-1a). Likewise, (S)-1a indicates pair of diastereomers,
and [S(C*),R(N)]-1a (d4-(S)-1a). Thus,
a
*
[(S(C ),S(N)]-1a (d3-(S)-1a)
d1-(R)-1a and d3-(S)-1a, and d2-(R)-1a and d4-(S)-1a are enantio-
meric pairs, respectively. Meanwhile, (RS)-1a is used to indicate a
mixture of (R)-1a and (S)-1a, that is, all four stereoisomers. In a
solution, since there is no significant barrier between two diaste-
reomers having an identical configuration around C*, only one of
them is expected to be present predominantly in solution, if there
is a significant energy difference between two diastereomers.8
The crystal (RS)-1a when formed from a racemic mixture
belongs to an achiral space group P21/n (Table 1), where two
enantiomers, d1-(R)-1a and d3-(S)-1a, are paired to form a centro-
symmetric asymmetric unit (pairing a in Scheme 2).
*
cause there exist two stereogenic centers at C and N (Fig. 2). How-
R
R
O
O
On the other hand, crystal (R)-1a, whose component has an
enantiomeric purity of greater than 99% ee, belongs to a non-cen-
trosymmetric space group P21 (Table 1). Interestingly, two enan-
tiomer-like diastereomers, d1-(R)-1a and d2-(R)-1a (Fig. 3b),
which are in rapid equilibrium at ambient temperature, are paired
to form a pseudo-centrosymmetric asymmetric unit for the imita-
tion of a centrosymmetric asymmetric unit in crystal (RS)-1a
(Fig. 3a). As a result, crystal (R)-1a looked like an achiral crystal,
rather than a chiral crystal of a pure enantiomer.
This unusual pairing of d1-(R)-1a and d2-(R)-1a (pairing b in
Scheme 2) might occur because the shape of d2-(R)-1a was very
similar to that of d3-(S)-1a. The overlay of d2-(R)-1a taken from
crystal (R)-1a on d3-(S)-1a obtained either directly from crystal
(RS)-1a (Fig. 3a and c) or by inverting d1-(R)-1a found in crystal
(R)-1a (Fig. 3b and d) shows their near perfect superimposition.
The only difference between d2-(R)-1a and d3-(S)-1a is the thick-
ness in the middle section of the relatively flat bicyclic skeleton
and the direction of the phenyl ring.
S
O
N
O
O
O
N S
*
*
O
H
H
(
R
(C*),
R
(N))-1a or 1b
(
S
(C*),
S
(N))-1a or 1b
)-1a or 1b]
[
d1-(
R
)-1a or 1b]
[
d3-(
S
R
R
O
S
O
O
N S
N
O
*
*
O
H
H
(
S
(C*),
R
(N))-1a or 1b
)-1a or 1b]
(
R
(C*),
S
(N))-1a or 1b
)-1a or 1b]
[
d2-(
R
[
d4-(
S
( )-1a or 1b
R
(S)-1a or 1b
(
RS)-1a or 1b
a. R = Ph
b. R = H
Thus, under given circumstances, d2-(R)-1a disguises itself as
d3-(S)-1a so that it can be paired with d1-(R)-1a to form a pseu-
do-centrosymmetric asymmetric unit resulting in an apparently
Figure 2. All possible stereoisomers of 1a and 1b.

1738
K. S. Jeong et al. / Tetrahedron: Asymmetry 20 (2009) 1736–1741
pairing a
achiral crystal from enantiopure (R)-1a. Consequently, the overall
crystal structures of (R)-1a and (RS)-1a look very much alike. This
pseudo achiral crystal mimicking a racemic compound constitutes
as an interesting example of the aforementioned case; achiral crys-
tal formation by single enantiomers.9
O
S
O
O
S
O
O
N
O
N
The thermodynamic stability and hence the packing efficiency
of crystal (RS)-1a and crystal (R)-1a were determined by DSC. Crys-
tal (RS)-1a exhibits a substantially higher melting point and en-
thalpy of fusion than crystal (R)-1a: crystal (RS)-1a, 167.8 °C and
36.1 kJ/mol; crystal (R)-1a, 125.3 °C and 25.9 kJ/mol. Therefore,
the racemate is significantly more stable than the pseudo achiral
crystal. However, it is worth pointing out that the density of crystal
(R)-1a (1.359 g/cm3) is noticeably higher than that of crystal (RS)-
1a (1.340 g/cm3) (Table 1). Despite their close resemblance in an
asymmetric unit, a pair of molecules in crystal (R)-1a packs differ-
ently from that in crystal (RS)-1a, which results in crystal (R)-1a
being denser than crystal (RS)-1a.
This is contradictory to the expectation from Wallach’s rule,
which was documented as racemic crystals tended to be denser
than their chiral counterparts.10 Thus, our observation demon-
strated again that density was no longer a good measure of ther-
modynamic stability (the close packing principle),11a,b and
reinforced Yu’s suggestion to use other means, such as eutectic-
melting and/or solubility data, for a fair comparison of the relative
stability of crystals.11c
H
H
pairing b
d1-(R)-1a
d3-(S)-1a
look similar
mirror
O
O
d2-(R)-1a
O
S
N
H
H
S
O
N
O
O
d1-(R)-1a
O
inversion at N
O
O
O
O
O
N
S
N
S
H
H
d4-(S)-1a
d2-(R)-1a
Next, the crystallization of 1b, a congener of 1a, in which the
phenyl ring at the 2-position of bicyclic octenone of 1a is replaced
by a hydrogen atom, was attempted. Although the results are not
much different from those of crystals of 1a (Table 1), a couple of
points are worth mentioning.
mirror
Scheme 2. A schematic representation of the structural relationship between the
diastereomers of 1a in solution.
Firstly, crystal (RS)-1b consists of only enantiomeric stereoiso-
mers, d2-(R)-1b and d4-(S)-1b, which are alternately arranged in
the crystal with a centrosymmetric arrangement. Conversely to
crystal (RS)-1a, crystal (RS)-1b contains d2-(R)-1b and d4-(S)-1b in-
stead of d1 and d3, respectively (Fig. 4a). Secondly, crystal (R)-1b
comprises two diastereomers of (R)-1b, d1-(R)-1b and d2-(R)-1b
(Fig. 4b) as expected by an analogy with 1a. However, the ratio
of diastereomers d1 to d2 in crystal (R)-1b is 38:62 instead of the
expected number of 50:50 (Fig. 4c).
It appears that contrary to 1a, there is no distinct difference be-
tween d1-(R)-1b and d2-(R)-1b in shape and, furthermore, all four
stereoisomers of 1b are not structurally very different from each
other (Scheme 3).
Thus, any combination of pairing does not exhibit a noticeable
difference in stability. As a result, the pairing of d1-(R)-1b and
d2-(R)-1b (pairing b in Scheme 3) could not be easily distinguished
from the self-pairing of two d2-(R)-1b molecules during crystalli-
zation of (R)-1b.
Thus, although there is still a strong tendency to adopt a pseu-
do-centrosymmetric arrangement by the matched pairing of the
two different diastereomers [d1-(R)-1b and d2-(R)-1b] (pairing a
in Scheme 3), a significant proportion of the mismatched pair by
two identical diastereomers d2-(R)-1b is found in the crystal. The
inclusion of the mismatched pair in the crystal is unlikely to make
a significant difference in the overall crystal stability of (R)-1b. In
Figure 3. ORTEP drawings of molecular structures with 30% thermal ellipsoids
depicted for d1-(R)-1a and d3-(S)-1a in crystal (RS)-1a (a) and for d1-(R)-1a and d2-
(R)-1a in crystal (R)-1a (b). (c) Superposition of d2-(R)-1a (solid line) in crystal (R)-
1a onto d3-(S)-1a (open line) in crystal (RS)-1a. (d) Superposition of d2-(R)-1a (solid
line) in crystal (R)-1a onto d3-(S)-1a (open line) which was obtained by inverting
d1-(R)-1a in crystal (R)-1a.
Table 1
Summary of crystal data and refinement results
Crystal
Formula
Space group
a (Å)
b (Å)
c (Å)
a
(deg)
b (deg)
c
(deg)
Z
Dcacld
R1
wR2
GOF
(RS)-1a
(R)-1a
(RS)-1b
(R)-1b
C20H19NO3S
P21/n
P21
P21/n
P21
7.277 (1)
9.272 (1)
6.017 (1)
6.066 (1)
22.243 (1)
14.863 (1)
10.775 (1)
10.767 (1)
11.286 (1)
13.181 (1)
20.613 (1)
20.659 (1)
90
90
90
90
106.42 (1)
108.01 (1)
90.96 (1)
91.53 (1)
90
90
90
90
4
4
4
4
1.340
1.359
1.379
1.366
0.0431
0.0450
0.0498
0.0486
0.1264
0.1042
0.1382
0.0856
1.078
0.853
1.054
0.825
C
20H19NO3S
C14H15NO3S
C14H15NO3S
The bold is used to highlight the violation of Wallach’s Rule.

K. S. Jeong et al. / Tetrahedron: Asymmetry 20 (2009) 1736–1741
1739
will need more relevant data, and thus, the efforts will be put off
until the accumulation of data reaches a certain level.
3. Conclusion
Herein, we have reported examples of apparently achiral crystal
formation by single enantiomers. From the X-ray crystallographic
studies of compound 1a, we noticed that the heterochiral mole-
cules strongly favored the formation of a racemic compound, and
even the homochiral molecules provided a pseudo achiral crystal
of a pure enantiomer through the adoption of a pseudo-centrosym-
metric arrangement accompanying a structural adjustment. This
observation could be explained by the strong propensity of any
molecule to take on centrosymmetric and pseudo-centrosymmet-
ric arrangements in crystals, if possible. In addition, we found that,
especially for 1a, densities of crystals do not necessarily reflect the
thermodynamic stabilities, thus the century-old Wallach’s rule has
to be re-stated.
4. Experimental
Figure 4. ORTEP drawings of crystal and molecular structures of 1b. The unit cell
packing diagrams of crystal (RS)-1b (a) and crystal (R)-1b (b) are drawn along the a-
axis. In crystal (R)-1b, there are two molecules in the asymmetric unit, which
contains a pair of diastereomers, viz. either d1-(R)-1b/d2-(R)-1b (76%) or d2-(R)-1b/
d2-(R)-1b (24%), resulting in an overall d1/d2 ratio of 38:62. One of the two
occupied sites in the asymmetric unit is disordered due to the overlay of the two
different diastereomers, d1-(R)-1b and d2-(R)-1b, at the same site as shown in (c):
d4-(S)-1b (solid line) in crystal (RS)-1b has been super-positioned onto d2-(R)-1b
(hair line) and d1-(R)-1b (dashed line) occupying 24% and 76% of the disordered
sites in crystal (R)-1b, respectively. Notice the mirror symmetry relation between
d4-(S)-1b and d2-(R)-1b.
4.1. Preparation of racemic-1a, (RS)-1a7a
A mixture of Co2(CO)8 (252 mg, 0.737 mmol) and the enyne 2a
(200 mg, 0.615 mmol) in CH2Cl2 (6 mL) was stirred at room temper-
ature for 1 h. After confirming the formation of the enyne-dicobalt
hexacarbonyl complex by TLC analysis, a solution of trimethylamine
N-oxide dihydrate (342 mg, 3.08 mmol) in CH2Cl2 (4 mL) was added.
After stirring at room temperature for 1 h, the reaction mixture was
passed through a small pad of silica gel 60 (230À400 mesh, Merck,
Darmstadt) with 1:2 EtOAc/n-hexane eluent to remove the purple
precipitates. The filtrate was concentrated in vacuo and then
purified by flash column chromatography (EtOAc/n-hexane = 1:5)
on silica gel 60 to afford (RS)-1a (180 mg, 0.510 mmol, 83%) as a
white solid.
mirror
pairing b
4.2. Preparation of (R)-1a in 65% ee7b–d
O
S
O
O
S
O
O
O
N
N
O
N
N
H
H
H
A mixture of (R)-(+)-BINAP (34.4 mg, 0.055 mmol, 9 mol %) and
[Rh(CO)2Cl]2 (7.2 mg, 0.018 mmol, 3 mol %) in THF (5 mL) was stir-
red at room temperature for 10 min under an argon atmosphere.
To the mixture was then added a solution of AgOTf (18.9 mg,
0.074 mmol, 12 mol %) in THF (5 mL). The formation of white
precipitates of silver chloride was observed. To the resulting mix-
ture was added a solution of enyne 2a (200 mg, 0.615 mmol) in
THF (5 mL), and then the argon atmosphere was replaced by a
CO balloon. After refluxing at 80 °C for 4 h, the reaction mixture
was cooled to room temperature and the carbon monoxide was
released into the hood. The resulting mixture was passed through
a small pad of silica gel with 1:3 EtOAc/n-hexane eluent and the
filtrate was concentrated in vacuo. The crude residue was purified
by flash column chromatography (EtOAc/n-hexane = 1:5) on silica
gel to afford (R)-1a (158 mg, 72%, 65% ee). The ee was determined
by HPLC analysis on a chiral column (Chiralpak AD-H, 0.46 Â
25 cm, Daicel) under isocratic conditions (2-propanol/n-hexane =
1:3, 1.2 mL/min flow rate, 254 nm). Retention time: (R)-1a,
d1-(R)-1b
d3-(S)-1b
O
O
O
O
O
S
S
H
d2-(R)-1b
d4-(S)-1b
pairing a
Scheme 3. A schematic representation of the structural relationship between the
diastereomers of 1b in solution.
this case, the ratio of diastereomers in the crystal presumably re-
flects that in solution at the crystallization temperature.
The DSC study shows that crystal (RS)-1b exhibits a higher
melting point and enthalpy of fusion than crystal (R)-1b: crystal
(RS)-1b, 154.0 °C and 32.9 kJ/mol; crystal (R)-1b, 139.8 °C and
24.5 kJ/mol. This indicates that the racemate is still more stable
than the pseudo achiral crystal. However, the stability difference
between the racemic compound and the pseudo achiral crystal of
1b is substantially smaller than that of 1a. This specific example
satisfied Wallach’s rule (Table 1).
sR = 11.16 min; (S)-1a, sR = 13.01 min.
4.3. Preparation of (R)-1a in >91% ee7f
A mixture of (R)-tol-BINAP (312 mg, 0.46 mmol, 30 mol %) and
[Ir(COD)Cl]2 (113 mg, 0.23 mmol, 15 mol %) in toluene (10 mL)
was stirred at 40 °C for 30 min under an atmospheric pressure of
CO. To the resulting yellow mixture was then added a solution of
enyne 2a (500 mg, 1.54 mmol) in toluene (10 mL). After refluxing
Attempts to determine the rationale behind these phenomena
and formulate a method to predict the crystal structure from the
constituting chiral molecules are very tempting.12 However, this

1740
K. S. Jeong et al. / Tetrahedron: Asymmetry 20 (2009) 1736–1741
at 130 °C for 24 h, the reaction mixture was cooled to room tem-
perature and the carbon monoxide was released into the hood.
The resulting mixture was passed through a small pad of silica
gel with 1:3 EtOAc/n-hexane eluent and the filtrate was concen-
trated in vacuo. The crude residue was purified by flash column
chromatography (EtOAc/n-hexane = 1:5) on silica gel to afford
(R)-1a (420 mg, 78%, >91% ee).
4.8. Crystallographic data of crystal (RS)-1a
Colorless block crystal, 0.30 Â 0.22 Â 0.20 mm3, C20H19NO3S,
M = 353.42, monoclinic, space group P21/n (No. 14), T = 296(2) K,
a = 7.277(1) Å, b = 22.243(1) Å, c = 11.286(1) Å, b = 106.42(1)°, V =
1752.3(1) Å3, Z = 4, dcalc = 1.340 g/cm3,
l(Mo Ka
) = 0.203 mmÀ1. A
total of 17,866 reflections were collected in the range
1.83° 6 h 6 28.36° of which 4379 were independent and 3196 were
observed (I >2r(I)). CCDC reference number 711,864.
4.4. Preparation of racemic-1b, (RS)-1b7a
4.9. Crystallographic data of crystal (R)-1a
The title compound (RS)-1b (440 mg, 1.59 mmol, 78%) was ob-
tained from enyne 2b (500 mg, 2.01 mmol) according to the proce-
dure used for the preparation of (RS)-1a from 2a.
Colorless block crystal, 0.34 Â 0.25 Â 0.16 mm3, C20 H19NO3S,
M = 353.42, monoclinic, space group P21 (No. 4), T = 296(2) K,
a = 9.272(1) Å, b = 14.863(1) Å, c = 13.181(1) Å, b = 108.01(1)°, V =
4.5. Preparation of (R)-1b in 85% ee7g
1727.5(1) Å3, Z = 4, dcalc = 1.359 g/cm3,
l(Mo Ka
) = 0.206 mmÀ1. A
total of 32,882 reflections were collected in the range 1.62° 6
A mixture of (R)-(+)-BINAP (250 mg, 0.40 mmol, 20 mol %) and
Co2(CO)8 (138 mg, 0.40 mmol, 20 mol %) in toluene (10 mL) was
stirred at 40 °C for 2 h under an atmospheric pressure of CO. To
the mixture was then added a solution of enyne 2b (500 mg,
2.01 mmol) in toluene (5 mL). After heating at 80 °C with stirring
for 16 h, the reaction mixture was cooled to room temperature
and the carbon monoxide was released into the hood. The resulting
mixture was passed through a small pad of silica gel with 1:3
EtOAc/n-hexane eluent and the filtrate was concentrated in vacuo.
The crude residue was purified by flash column chromatography
(EtOAc/n-hexane = 1:5) on silica gel to afford (R)-1b (200 mg,
36%, 85% ee). The ee was determined by HPLC analysis on a chiral
column (Chiralpak AS, 0.46 Â 25 cm, Daicel) under isocratic condi-
tions (2-propanol/n-hexane = 1:2, 2.0 mL/min flow rate, 254 nm).
h 6 28.28° of which 8452 were independent and 3659 were ob-
served (I >2r(I)). CCDC reference number 711,863.
4.10. Crystallographic data of crystal (RS)-1b
Colorless block crystal, 0.34 Â 0.28 Â 0.18 mm3, C14H15NO3S,
M = 277.33, monoclinic, space group P21/n (No. 14), T = 296(2) K,
a = 6.017(1) Å, b = 10.775(1) Å, c = 20.613(1) Å, b = 90.96(1)°, V =
1336.3(1) Å3, Z = 4, dcalc = 1.379 g/cm3,
l(Mo Ka
) = 0.245 mmÀ1. A
total of 11,401 reflections were collected in the range 2.13° 6
h 6 28.30° of which 3280 were independent and 2171 were ob-
served (I >2r(I)). CCDC reference number 711,866.
4.11. Crystallographic data for crystal (R)-1b
Retention time: (R)-1b,
sR = 21.05 min; (S)-1b, sR = 30.82 min.
Colorless rectangular crystal, 0.28 Â 0.20 Â 0.16 mm3, C14H15NO3S,
M = 277.33, monoclinic, space group P21 (No. 4), T = 296(2) K,
a = 6.066(1) Å, b = 10.767(1) Å, c = 20.659(1) Å, b = 91.53(1)°, V =
4.6. Crystal (RS)-1a and crystal (RS)-1b
1348.8(1) Å3, Z = 4, dcalc = 1.366 g/cm3, ) = 0.243 mmÀ1. A to-
l(Mo Ka
A vial containing (RS)-1a (15 mg, 0.042 mmol) in EtOAc (1 mL)
or (RS)-1b (50 mg, 0.18 mmol) in acetone (1 mL) was placed in a
bottle charged with n-hexane (7 mL). Slow diffusion of n-hexane
into a solution of (RS)-1a in EtOAc or (RS)-1b in acetone provided
a single crystal suitable for the single-crystal X-ray diffraction
experiment. Elemental analysis (wt %): crystal (RS)-1a. Calcd for
C20H19NO3S: C, 67.97; H, 5.42; N, 3.96; S, 9.07. Found: C, 67.59;
H, 5.32; N, 3.93; S, 9.17. Crystal (RS)-1b. Calcd for C14H15NO3S: C,
60.63; H, 5.45; N, 5.05; S, 11.56. Found: C, 60.30; H, 5.43; N,
4.96; S, 11.60.
tal of 12,777 reflections were collected in the range 1.97° 6 h 6 28.32°
of which 5590 were independent and 1989 were observed (I >2r(I)).
CCDC reference number 711,865.
All the intensity data were collected on a Bruker SMART CCD
diffractometer with a normal focus and graphite monochromated
Mo-target X-ray tube (k = 0.71,073 Å), and integrated with the
SAINT+ software package13 with a narrow frame algorithm. Absorp-
tion corrections were applied using SADABS.
14 All the structures were
solved by direct methods and subsequent difference Fourier syn-
theses and refined with the SHELXTL software package.15 All stages
of weighted full-matrix least-squares refinement were conducted
2
4.7. Crystal (R)-1a and crystal (R)-1b
using Fo data. Non-hydrogen atoms were refined anisotropically,
and hydrogen atoms were generated with the ideal geometry.
Since none of the synthetic methods allowed the preparation of
(R)-1a or (R)-1b in greater than 99% ee, it was necessary to perfect
their enantiomeric purity prior to their crystallization. A vial con-
taining (R)-1a (20 mg, 91% ee) in EtOAc (1 mL) was placed in a bot-
tle charged with n-hexane (10 mL). As the first drops of crystals
were formed, the supernatant was taken and its ee was measured.
When the ee of the supernatant exceeded 99%, the crystals were
filtered off and the filtrate was concentrated in vacuo. The white
precipitate thus obtained was then recrystallized by the diffusion
of n-hexane (5 mL) into its solution in EtOAc (1 mL) to give a single
crystal of (R)-1a suitable for the single-crystal X-ray diffraction
experiment. Crystal (R)-1b was obtained according to the proce-
dure used for that of (R)-1a. Elemental analysis (wt %): crystal
(R)-1a. Calcd for C20H19NO3S: C, 67.97; H, 5.42; N, 3.96; S, 9.07.
Found: C, 68.25; H, 5.53; N, 4.30; S, 8.74. Crystal (R)-1b. Calcd for
C14H15NO3S: C, 60.63; H, 5.45; N, 5.05; S, 11.56. Found: C, 61.06;
H, 5.69; N, 4.99; S, 11.96.
Acknowledgments
This work was supported by the Korea Science and Engineering
Foundation (KOSEF) grant (No. R01-2007-000-10534-0) funded by
Korean Government and
University.
a special research grant by Korea
References
1. (a) Horeau, A. In Chemical Methods: Stereochemistry, Fundamentals, and Methods;
Kagan, H. B., Ed.; Georg Thieme: Stuttgart, 1977; pp 51–76; (b) Gavezzotti, A..
In Structure Correlation; Buergi, H. B., Dunitz, J. D., Eds.; VCH: Weinheim, 1994;
Vol. 2, pp 509–542; (c) Zheng, X.; Donkor, I. O.; Miller, D. D.; Ross, C. R., II.
Chirality 2000, 12, 2–5; (d) Dunitz, J. D.; Gavezzotti, A. Angew. Chem., Int. Ed.
2005, 44, 1766–1787.
2. (a) Barlow, W.; Pope, W. J. J. Chem. Soc. 1906, 89, 1675–1744; (b)
Kitaigorodskii, A. I. Organic Chemical Crystallography; Consultant Bureau:
New York, 1961.

K. S. Jeong et al. / Tetrahedron: Asymmetry 20 (2009) 1736–1741
1741
3. (a) Allen, F. H. Acta Crystallogr., Sect. B 2002, 58, 380–388; (b) Jacques, J.; Collet,
A.; Wilen, S. H. Enantiomers, Racemic compounds, and Resolutions; Wiley: New
York, 1981; (c) Collet, A.; Brienne, M.-J.; Jacques, J. Chem. Rev. 1980, 80, 215–
230.
10. (a) Wallach, O. Liebigs Ann. Chem. 1895, 286, 90–143; (b) Brock, C. P.;
Schweizer, W. B.; Dunitz, J. D. J. Am. Chem. Soc. 1991, 113, 9811–9820; A
previously reported counter-example of Wallach’s rule (c) Patrick, B. O.; Brock,
C. P. Acta Crystallogr., Sect. B 2006, 62, 488–497.
4. Pasteur, L. Ann. Chim. Phys. 1848, 24, 442–460.
11. (a) Kitaigorodsky, A. I. Molecular Crystals and Molecules; Academic Press: New
York, 1973; (b) Ebisuzaki, Y.; Askari, L. H.; Bryan, A. M.; Nicol, M. F. J. Chem.
Phys. 1987, 87, 6659–6664; (c) Huang, J.; Yu, L. J. Am. Chem. Soc. 2006, 128,
1873–1878.
5. (a) Gavezzotti, A. CrystEngComm 2002, 4, 343–347. and leading references cited
therein; (b) Gavezzotti, A. J. Am. Chem. Soc. 1991, 113, 4622–4629; (c) Dalhuo,
B.; Gorbitz, C. H. Acta Crystallogr., Sect. B 2000, 56, 715–719; (d) Hendi, M.;
Hooter, P.; Lynch, V.; Davis, R. E.; Wheeler, K. A. Cryst. Growth Des. 2004, 4, 95–
101.
6. There are studies dealing with crystallization through the pseudoinversion
center irrespective of the chirality (a) Desiraju, G. R.; Calabrese, J. C.; Harlow, R.
L. Acta Crystallogr., Sect. B 1991, 47, 77–86; (b) Marsh, R. E. Acta Crystallogr.
1999, B55, 931–936.
7. (a) Jeong, N.; Chung, Y. K.; Lee, B. Y.; Lee, S. H.; Yoo, S.-E. Synlett 1991, 204–206;
(b) Jeong, N.; Sung, B. K.; Choi, Y. K. J. Am. Chem. Soc. 2000, 122, 6771–6772; (c)
Jeong, N.; Sung, B. K.; Kim, J. S.; Park, S. B.; Seo, S. D.; Shin, J. Y.; In, K. Y.; Choi, Y.
K. Pure Appl. Chem. 2002, 74, 85–91; (d) Jeong, N.; Kim, D. H.; Choi, J. H. Chem.
Commun. 2004, 1134–1135; (e) Jeong, N. In Modern Rhodium-Catalyzed Organic
Reactions; Evans, P. A., Ed.; Wiley-VCH: Weinheim, 2005; pp 215–240. Chapter
11; (f) Shibata, T.; Takagi, K. J. Am. Chem. Soc. 2000, 122, 9852–9853; (g) Gibson,
S. E.; Lewis, S. E.; Loch, J. A.; Steed, J. W.; Tozer, M. J. Organometallics 2003, 22,
5382–5384.
12. Since there is probably a group of these types of compounds in the CSD, we
have tried to search similar examples that could be classified as such. More
than a thousand crystal structures with Z = 4 and P21 (a minimum condition for
possible pseudo achiral crystals for the monoclinic crystal system) were
identified from CSD data since 1989. Since scrutinizing all the results would
have been too time-consuming, the search results were further narrowed
down with the constraint that the target molecule contained only C, H, N, O,
and S atoms. The search provided 173 hits. If two molecules in an asymmetric
unit were related to each other via pseudo-inversion or pseudo-mirror planes,
they would have the possibility of forming a pseudo achiral crystal. However,
we were not able to identify many compounds in this class so far. Although we
have some candidates for further elucidation, either the coordinates of the
crystal structures or the data required for the comparison, such as the crystal
structures of the racemic compounds, are lacking. Nevertheless, more cases
would be expected to be found when previous and prospective crystal
structures are scrutinized based on our suggestion. (a) Valente, E. J.; Trager,
W. F.; Lingafelter, E. C. Acta Crystallogr., Sect. B 1976, 32, 277–279; (b) Baert, F.;
Fouret, R.; Oonk, H. A. J.; Kroon, J. Acta Crystallogr., Sect. B 1978, 34, 222–226; (c)
Kroon, J.; van Gurp, P. R. E.; Oonk, H. A. J.; Baert, F.; Fouret, R. Acta Crystallogr.,
Sect. B 1976, 32, 2561–2564; (d) Oonk, H. A. J.; Kroon, J. Acta Crystallogr., Sect. B
1976, 32, 500–504.
*
calculated to be more stable than (R(C*),S(N))-
8. For example, (R(C ),R(N))-1a is
1a. However, the NMR spectra and HPLC analysis of (R)-1a with a chiral column
showed peaks corresponding to only one compound. Therefore, (R)-1a can be
considered as a single chiral molecule as well as a pair of diastereomers. By
analogy, (RS)-1a can be considered as a pair of enantiomers or a racemate of
two chiral molecules as well as all of the four stereoisomers. We also use the
additional notation d#-(R or S)-1a for each of the four stereoisomers as shown
in the bracket, where d# means a specific diastereomer, and (R or S) indicates
the configuration around C*. Note that d1-(R)-1a and d2-(R)-1a are the mirror
images of d3-(S)-1a and d4-(S)-1a, respectively. Similar notations can be
applied for compound 1b.
13. SAINT+ ver. 6.04. SAX Area-Detector Integration Program. Bruker AXS. Madison,
WI, 1997–2001.
14. Sheldrick, G. M. SADABS Version 2.03,
A Program for Empirical Absorption
Correction; Universität Göttingen, 1997–2001.
15. Bruker AXS SHELXTL version 6.10. Structure Determination Package; Bruker AXS:
Madison, WI, 2000.
9. (a) Rosini, G.; Borzatta, V.; Boschi, F.; Candido, G.; Marotta, E.; Righi, P. Chem.
Commun. 2007, 2717–2719; (b) Gavezzotti, A.; Flack, H. Crystal Packing. In
The Third Series of Teaching Pamphlet IUCr Commission on Crystallographic
Teaching; Taylor, C. A., Ed.; International Union of Crystallography, 2007.