Conformational change of N-benzylideneanilines in crystals

Article abstract of DOI:10.1107/S0108768104016623

The crystal structures of N-(4-nitrobenzylidene)aniline (1), N-(4-chlorobenzylidene)-4-methylaniline (2) and N-(4-methylbenzylidene)-4- methylaniIine (3) were determined by X-ray diffraction analyses at various temperatures. A dynamic disorder was observe

Full text of DOI:10.1107/S0108768104016623

research papers
Acta Crystallographica Section B
Structural
Science
Conformational change of -benzylideneanilines in
crystals
ISSN 0108-7681
Received 7 April 2004
Jun Harada, Mayuko Harakawa
and Keiichiro Ogawa*
The crystal structures of N-(4-nitrobenzylidene)aniline (1), N-
(4-chlorobenzylidene)-4-methylaniline (2) and N-(4-methyl-
Accepted 7 July 2004
benzylidene)-4-methylaniline (3) were determined by X-ray
diffraction analyses at various temperatures. A dynamic
disorder was observed in the crystal structures of all
compounds. The dynamic disorder is accounted for in terms
of a conformational change involving a pedal motion in the
crystals.
Department of Chemistry, Graduate School of
Arts and Sciences, The University of Tokyo,
Komaba, Meguro-ku, Tokyo 153-8902, Japan
Correspondence e-mail:
ogawa@ramie.c.u-tokyo.ac.jp
1. Introduction
Solid-state conformational changes have been studied in a
wide range of ®elds in chemistry (Gavezzotti & Simonetta,
È
1982, 1987; Harris & Aliev, 1995; Burgi, 2002). Among the
several methods that have been applied to the subject, X-ray
diffraction analysis is one of the most powerful. When
conformational changes take place in crystals, each molecule
adopts one of several possible, alternative conformations. X-
ray diffraction studies give access to the populations of each
conformer by analysis of diffraction data with a disorder
model. The occurrence of conformational changes in crystals
can be detected on the basis of temperature-dependent
populations of the conformers. Thus, variable-temperature X-
ray diffraction measurement and analysis of the disorder
provide not only static molecular structures, but also
substantial information on the dynamic aspects of molecules
in the crystals.
We have studied the conformational changes of (E)-stil-
benes and azobenzenes in crystals using disorder analysis
(Harada et al., 1997; Harada & Ogawa, 2001, 2004). In the
disordered crystals of these compounds the molecules adopt
two orientations related by an approximate twofold rotation
about the longest axis of the molecules (Fig. 1).
Variable-temperature X-ray diffraction analyses showed
that the populations depend on the temperature, which
behavior is indicative of a conformational interconversion in
the crystal. The conformational change was inferred to take
place through a pedal motion: a pair of benzene rings moving
like the pedals of a bicycle (Fig. 2). Solid-state NMR spec-
troscopy (Ueda et al., 1988; McGeorge et al., 1998), measure-
ments of heat capacity (Saito et al., 1995, 2000, 2004) and
molecular mechanics calculations (Galli et al., 1999) have
substantiated the occurrence of the pedal motion. The same
motion was reported to be a key process of the photo-
chromism of salicylideneanilines and the photodimerization of
trans-cinnamides in the crystalline state (Harada et al., 1999;
Ito et al., 2000; Ohba et al., 2001).
In this study we investigated the solid-state conformational
change of some derivatives of N-benzylideneaniline, which is
isoelectronic with (E)-stilbene and azobenzene. Although the
# 2004 International Union of Crystallography
Printed in Great Britain ± all rights reserved
Acta Cryst. (2004). B60, 589±597
doi:10.1107/S0108768104016623 589

research papers
crystal structures of some N-benzylideneanilines are known to
show a disorder similar to that observed for (E)-stilbenes and
azobenzenes (Bernstein et al., 1976; Bar & Bernstein, 1983),
the dynamic properties of the disorder have yet to be studied.
In the preceding paper we described the results of the vari-
able-temperature X-ray diffraction analyses of some N-
benzylideneanilines and showed that a torsional vibration
about the CÐPh and NÐPh bonds, which is similar to the
pedal motion, takes place in the crystals (Harada et al., 2004).
In this paper we report on the variable-temperature X-ray
diffraction analyses of some additional disordered N-benzyl-
ideneanilines: N-(4-nitrobenzylidene)aniline (1), N-(4-
chlorobenzylidene)-4-methylaniline (2) and N-(4-methylben-
zylidene)-4-methylaniline (3). Disorder analyses of the crystal
structures revealed that disorder in the crystals of these N-
benzylideneanilines is dynamic and that conformational
changes through pedal motion take place. Detailed analysis of
the temperature dependence of the crystal structure of (3)
showed that conformational changes may be frozen in, so that
nonequilibrium states are found at low temperature.
Cryosystems) open-¯ow gas cryostat (Cosier & Glazer, 1986),
except for the measurements at room temperature. The
temperature in the nozzle was held constant within Æ 0.2 K
during the measurement. The crystals were cooled or warmed
at a rate of 1 K minꢁ1, unless otherwise mentioned. The crystal
and experimental data are summarized in Table 1.
Three sets of diffraction data (at room temperature, 200 K
and 90 K) were collected for (1) and (2). The temperature
dependence of the crystal structure of (3) was investigated in
more detail. A total of 10 sets of diffraction data were
collected at various temperatures using the same crystal, seven
of which are discussed in this paper. Temperature changes and
diffraction measurements were carried out according to the
following procedures. Diffraction measurements were carried
out at 300, 250, 200, 150 and 90 K in this order. Before each
measurement the crystal was warmed to room temperature
Figure 1
Disorder for stilbenes and azobenzenes.
2. Experimental
All compounds were obtained by the dehydration condensa-
tion between corresponding benzaldehydes and anilines.
Crystals of (1) and (2) were obtained by recrystallization from
methanol. Crystals of (3) were obtained from ethanol. Melting
points were determined on a micro-hot-stage apparatus and
are uncorrected.
All diffraction measurements were carried out using a
Bruker SMART 1000 CCD area-detector system with
Ê
graphite-monochromated Mo Kꢀ radiation (ꢁ = 0.71073 A).
Approximately 2500 frames of data were collected per data set
with a scan width of 0.3ꢀ in ! and 5 or 10 s exposure times. A
semi-empirical absorption correction was applied to the data
using the SADABS program (Sheldrick, 2002). The tempera-
ture of the crystals was controlled using a Cryostream (Oxford
Figure 2
Pedal motion of stilbenes and azobenzenes.
ꢂ
590 Jun Harada et al.
Conformational changes
Acta Cryst. (2004). B60, 589±597

research papers
Table 1
Experimental details.
(1) at RT
(1) at 200 K
(1) at 90 K
(2) at RT
Crystal data
Chemical formula
Mr
Cell setting, space group
C13H10N2O2
226.23
C13H10N2O2
226.23
C13H10N2O2
226.23
C14H12ClN
229.70
Monoclinic, P21/n
14.6363 (11), 10.8175 (8),
14.7228 (11)
101.943 (1)
2280.6 (3)
8
1.318
Mo Kꢀ
7219
Monoclinic, P21/n
14.4053 (11), 10.7367 (8),
14.7627 (12)
101.617 (2)
2236.5 (3)
8
1.344
Mo Kꢀ
10 646
Monoclinic, P21/n
14.2105 (12), 10.6596 (9),
14.8045 (12)
101.228 (2)
2199.6 (3)
8
1.366
Mo Kꢀ
13 353
Monoclinic, P21/a
5.9663 (5), 7.3989 (7),
13.7221 (12)
99.120 (2)
598.09 (9)
2
1.275
Mo Kꢀ
3013
Ê
a, b, c (A)
ꢂ (ꢀ)
3
Ê
V (A )
Z
Dx (Mg mꢁ3
)
Radiation type
No. of re¯ections for cell
parameters
ꢃ range (ꢀ)
2.2±28.5
0.09
2.4±29.5
0.09
2.2±30.0
0.10
2.8±30.0
0.29
ꢄ (mmꢁ1
)
Temperature (K)
Crystal form, colour
Crystal size (mm)
300
Block, yellow
0.46 Â 0.40 Â 0.28
200
Block, yellow
0.46 Â 0.40 Â 0.28
90
Block, yellow
0.46 Â 0.40 Â 0.28
300
Block, colourless
0.48 Â 0.34 Â 0.18
Data collection
Diffractometer
Bruker SMART 1000CCD
diffractometer
! scan
Multi-scan (based on
symmetry-related
measurements)
0.959
Bruker SMART 1000CCD
diffractometer
! scan
Multi-scan (based on
symmetry-related
measurements)
0.959
Bruker SMART 1000CCD
diffractometer
! scan
Multi-scan (based on
symmetry-related
measurements)
0.958
Bruker SMART 1000CCD
diffractometer
! scan
Multi-scan (based on
symmetry-related
measurements)
0.874
Data collection method
Absorption correction
Tmin
Tmax
0.975
34 291, 6663, 3616
0.974
33 619, 6529, 4886
0.974
33 043, 6400, 5521
0.950
8950, 1760, 1271
No. of measured, indepen-
dent and observed
re¯ections
Criterion for observed
re¯ections
Rint
ꢃmax (ꢀ)
I > 2ꢅ(I)
I > 2ꢅ(I)
I > 2ꢅ(I)
I > 2ꢅ(I)
0.041
30.1
0.024
30.0
0.023
30.0
0.019
30.0
Range of h, k, l
ꢁ20 ) h ) 20
ꢁ15 ) k ) 15
ꢁ20 ) l ) 20
ꢁ20 ) h ) 20
ꢁ15 ) k ) 15
ꢁ20 ) l ) 20
ꢁ20 ) h ) 19
ꢁ15 ) k ) 15
ꢁ20 ) l ) 20
ꢁ8 ) h ) 8
ꢁ10 ) k ) 10
ꢁ19 ) l ) 19
Re®nement
Re®nement on
R[F2 > 2ꢅ(F2)], wR(F2), S
No. of re¯ections
No. of parameters
H-atom treatment
F2
0.065, 0.174, 1.03
6663
399
Mixture of independent and
constrained re®nement
w = 1/[ꢅ2(Fo2) + (0.0436P)2
+ 0.878P], where P = (Fo2
+ 2Fc2)/3
F2
0.047, 0.131, 1.02
6529
387
F2
0.039, 0.112, 1.02
6400
387
F2
0.060, 0.195, 1.08
1760
72
Re®ned independently
Re®ned independently
Constrained to parent site
Weighting scheme
w = 1/[ꢅ2(Fo2) + (0.0542P)2
w = 1/[ꢅ2(Fo2) + (0.0629P)2
w = 1/[ꢅ2(Fo2) + (0.0841P)2
+ 0.1347P], where P = (Fo2
+ 2Fc2)/3
+ 0.6974P], where P = (Fo2
+ 0.6739P], where P = (Fo2
+ 2Fc2)/3
0.001
+ 2Fc2)/3
0.001
(Á/ꢅ)max
<0.0001
0.18, ꢁ0.29
<0.0001
0.24, ꢁ0.36
Ê ꢁ3
Áꢆmax, Áꢆmin (e A
Extinction method
Extinction coef®cient
)
0.27, ꢁ0.25
0.42, ꢁ0.19
None
±
None
±
None
±
SHELXL
0.101 (16)
(2) at 200 K
(2) at 90 K
(3) at 300 K
(3) at 250 K
Crystal data
Chemical formula
Mr
Cell setting, space group
C14H12ClN
229.70
C14H12ClN
229.70
C15H15N
209.28
C15H15N
209.28
Monoclinic, P21/a
5.9174 (5), 7.2879 (6),
13.70490 (11)
99.233 (2)
583.37 (7)
2
1.308
Mo Kꢀ
5419
Monoclinic, P21/a
5.8827 (6), 7.1953 (7),
13.6919 (14)
99.385 (2)
571.79 (10)
2
1.334
Mo Kꢀ
5419
Monoclinic, P21/c
9.8766 (7), 4.8839 (4),
12.0187 (9)
90.499 (1)
579.72 (8)
2
1.199
Mo Kꢀ
2827
Monoclinic, P21/c
9.8649 (6), 4.8663 (3),
11.9484 (8)
90.518 (1)
573.57 (6)
2
1.212
Mo Kꢀ
3145
Ê
a, b, c (A)
ꢂ (ꢀ)
3
Ê
V (A )
Z
Dx (Mg mꢁ3
)
Radiation type
No. of re¯ections for cell
parameters
ꢃ range (ꢀ)
2.8±30.0
2.8±30.0
3.4±29.9
3.4±30.0
ꢂ
Acta Cryst. (2004). B60, 589±597
Jun Harada et al.
Conformational changes 591

research papers
Table 1 (continued)
(2) at 200 K
(2) at 90 K
(3) at 300 K
(3) at 250 K
ꢄ (mmꢁ1
Temperature (K)
Crystal form, colour
Crystal size (mm)
)
0.30
200
0.30
90
0.07
300
0.07
250
Block, colourless
0.48 Â 0.34 Â 0.18
Block, colourless
0.48 Â 0.34 Â 0.18
Plate, colourless
0.40 Â 0.30 Â 0.08
Plate, colourless
0.40 Â 0.30 Â 0.08
Data collection
Diffractometer
Bruker SMART 1000CCD
diffractometer
! scan
Multi-scan (based on
symmetry-related
measurements)
0.871
Bruker SMART 1000CCD
diffractometer
! scan
Multi-scan (based on
symmetry-related
measurements)
0.868
Bruker SMART 1000CCD
diffractometer
! scan
Multi-scan (based on
symmetry-related
measurements)
0.973
Bruker SMART 1000CCD
diffractometer
! scan
Multi-scan (based on
symmetry-related
measurements)
0.972
Data collection method
Absorption correction
Tmin
Tmax
0.949
8648, 1707, 1454
0.948
8393, 1668, 1536
0.995
8496, 1689, 1330
0.994
8404, 1677, 1406
No. of measured, indepen-
dent and observed
re¯ections
Criterion for observed
re¯ections
Rint
ꢃmax (ꢀ)
I > 2ꢅ(I)
I > 2ꢅ(I)
I > 2ꢅ(I)
I > 2ꢅ(I)
0.016
30.0
0.017
30.0
0.022
30.0
0.019
30.0
Range of h, k, l
ꢁ8 ) h ) 8
ꢁ10 ) k ) 10
ꢁ19 ) l ) 19
ꢁ8 ) h ) 8
ꢁ10 ) k ) 10
ꢁ19 ) l ) 19
ꢁ13 ) h ) 13
ꢁ6 ) k ) 6
ꢁ16 ) l ) 16
ꢁ13 ) h ) 13
ꢁ6 ) k ) 6
ꢁ16 ) l ) 16
Re®nement
Re®nement on
F2
F2
F2
F2
R[F2 > 2ꢅ(F2)], wR(F2), S
No. of re¯ections
No. of parameters
H-atom treatment
Weighting scheme
0.053, 0.148, 1.12
1707
72
Constrained to parent site
w = 1/[ꢅ2(Fo2) + (0.0574P)2
+ 0.1885P], where P = (Fo2
+ 2Fc2)/3
0.059, 0.141, 0.60
1668
72
Constrained to parent site
w = 1/[ꢅ2(Fo2) + (0.0866P)2
+ 1.6982P], where P = (Fo2
+ 2Fc2)/3
0.052, 0.149, 1.05
1689
89
Constrained to parent site
w = 1/[ꢅ2(Fo2) + (0.061P)2
+ 0.122P], where P = (Fo2
+ 2Fc2)/3
0.048, 0.135, 1.08
1677
89
Constrained to parent site
w = 1/[ꢅ2(Fo2) + (0.0552P)2
+ 0.1239P], where P = (Fo2
+ 2Fc2)/3
(Á/ꢅ)max
<0.0001
0.29, ꢁ0.37
<0.0001
0.68, ꢁ0.65
<0.0001
0.23, ꢁ0.20
<0.0001
0.22, ꢁ0.20
Ê ꢁ3
Áꢆmax, Áꢆmin (e A
Extinction method
Extinction coef®cient
)
SHELXL
0.083 (10)
SHELXL
0.106 (10)
SHELXL
0.206 (18)
SHELXL
0.126 (13)
(3) at 200 K
(3) at 150 K
(3) at 90 K
Crystal data
Chemical formula
Mr
Cell setting, space group
C15H15N
209.28
C15H15N
209.28
C15H15N
209.28
Monoclinic, P21/c
9.8565 (6), 4.8494 (3),
11.8748 (7)
90.561 (1)
567.57 (6)
2
1.225
Mo Kꢀ
3629
Monoclinic, P21/c
9.8477 (6), 4.8327 (3),
11.7991 (7)
90.631 (1)
561.50 (6)
2
1.238
Mo Kꢀ
4040
Monoclinic, P21/c
9.8184 (6), 4.8159 (3),
11.7536 (7)
90.689 (1)
555.72 (6)
2
1.251
Mo Kꢀ
4495
Ê
a, b, c (A)
ꢂ (ꢀ)
3
Ê
V (A )
Z
Dx (Mg mꢁ3
)
Radiation type
No. of re¯ections for cell
parameters
ꢃ range (ꢀ)
3.4±30.0
0.07
3.5±29.9
0.07
3.5±29.9
0.07
ꢄ (mmꢁ1
)
Temperature (K)
Crystal form, colour
Crystal size (mm)
200
Plate, colourless
0.40 Â 0.30 Â 0.08
150
Plate, colourless
0.40 Â 0.30 Â 0.08
90
Plate, colourless
0.40 Â 0.30 Â 0.08
Data collection
Diffractometer
Bruker SMART 1000CCD
diffractometer
! scan
Multi-scan (based on
symmetry-related
measurements)
0.972
Bruker SMART 1000CCD
diffractometer
! scan
Multi-scan (based on
symmetry-related
measurements)
0.972
Bruker SMART 1000CCD
diffractometer
! scan
Multi-scan (based on
symmetry-related
measurements)
0.972
Data collection method
Absorption correction
Tmin
Tmax
0.994
0.994
0.994
ꢂ
592 Jun Harada et al.
Conformational changes
Acta Cryst. (2004). B60, 589±597

research papers
Table 1 (continued)
(3) at 200 K
(3) at 150 K
(3) at 90 K
No. of measured, indepen-
dent and observed
re¯ections
8329, 1661, 1428
8213, 1641, 1453
8100, 1624, 1473
Criterion for observed
re¯ections
Rint
ꢃmax (ꢀ)
I > 2ꢅ(I)
I > 2ꢅ(I)
I > 2ꢅ(I)
0.017
30.0
0.015
30.0
0.013
30.0
Range of h, k, l
ꢁ13 ) h ) 13
ꢁ6 ) k ) 6
ꢁ16 ) l ) 16
ꢁ13 ) h ) 13
ꢁ6 ) k ) 6
ꢁ16 ) l ) 16
ꢁ13 ) h ) 13
ꢁ6 ) k ) 6
ꢁ16 ) l ) 16
Re®nement
Re®nement on
F2
F2
F2
R[F2 > 2ꢅ(F2)], wR(F2), S
No. of re¯ections
No. of parameters
H-atom treatment
Weighting scheme
0.045, 0.134, 1.08
1661
89
Constrained to parent site
w = 1/[ꢅ2(Fo2) + (0.0651P)2
+ 0.1155P], where P = (Fo2
+ 2Fc2)/3
0.043, 0.126, 1.06
1641
89
Constrained to parent site
w = 1/[ꢅ2(Fo2) + (0.0652P)2
+ 0.132P], where P = (Fo2
+ 2Fc2)/3
0.041, 0.121, 1.07
1624
89
Constrained to parent site
w = 1/[ꢅ2(Fo2) + (0.0687P)2
+ 0.1224P], where P = (Fo2
+ 2Fc2)/3
(Á/ꢅ)max
<0.0001
0.27, ꢁ0.18
<0.0001
0.31, ꢁ0.18
<0.0001
0.36, ꢁ0.18
Ê ꢁ3
Áꢆmax, Áꢆmin (e A
Extinction method
Extinction coef®cient
)
SHELXL
0.074 (12)
SHELXL
0.050 (10)
SHELXL
0.035 (10)
² Computer programs used: SMART and SAINT (Bruker, 1998), SHELXS97 (Sheldrick, 1990), SHELXL97 (Sheldrick, 1997), SHELXTL (Bruker, 2000).
and then cooled to the required temperature. The data sets are
referred to as 300, 250, 200, 150 and 90 K, hereafter. After the
®ve sets of data collection, another data set was collected at
300 K. The agreement of these data (300 K2) with 300 K and,
therefore, the reversibility of the crystal structural change with
temperature was con®rmed. Then the crystal was ¯ash-cooled
to 90 K. The ¯ash cooling was carried out according to the
procedure described in the literature (Harada & Ogawa,
2004). Two consecutive measurements were carried out at this
temperature (90 K2 and 90 K3). No difference between 90 K2
and 90 K3 was observed. The crystal was warmed to room
temperature for annealing, and then cooled slowly to 90 K at
the rate 1 K minꢁ1. An X-ray diffraction analysis was carried
out at 90 K (90 K4). Afterwards the crystal was warmed to
room temperature and the diffraction analysis was carried out
at 300 K (300 K3). The results from 300 K3 showed no
difference from those of 300 K and 300 K2, which con®rmed
the total reversibility of the crystal structural change with
temperature. Results obtained from 300 K2, 90 K2, 90 K3,
90K4 and 300 K3 are given in the supplementary material.1
For nondisordered molecules all the H atoms were located
from difference-Fourier electron density maps and re®ned
isotropically, and all C, N and O atoms were re®ned aniso-
tropically. For disordered molecules the re®nement proce-
dures are given in Appendix A.
3. Results and discussion
3.1. Crystal structure of (1)
There are two crystallographically independent molecules
(referred to as molecules A and B hereafter) in the asym-
metric unit (monoclinic, P21/n, with Z = 8). Perspective views
of the molecules are shown in Fig. 3. For molecule A no
disorder was observed at any temperature.
For molecule B, a disorder similar to that observed in
crystals of (E)-stilbenes or azobenzenes was detected at room
temperature (Fig. 3). The populations of the two orientations
re®ned to 0.900 (3):0.100 (3). The population of the minor
orientation decreased as the temperature was lowered. At
200 K the difference Fourier synthesis gave a few weak resi-
dual peaks corresponding to the atoms of the minor orienta-
tion in the vicinity of the C N bond. Re®nement with the
disorder model was not successful, however, because of the
small population of the minor orientation. At 90 K no residual
peaks corresponding to atoms of the minor orientation could
be detected. The temperature dependence of the populations
of the two orientations clearly shows that the disorder is
dynamic and that a conformational change takes place in the
crystal of (1). The conformational change can be explained to
take place through the pedal motion resembling that observed
in the crystals of (E)-stilbenes and azobenzenes.
3.2. Crystal structure of (2)
1 Supplementary data for this paper are available from the IUCr electronic
archives (Reference: BK5008). Services for accessing these data are described
at the back of the journal.
The crystal structure observed at room temperature is
identical with that reported in the literature (monoclinic unit
ꢂ
Acta Cryst. (2004). B60, 589±597
Jun Harada et al.
Conformational changes 593

research papers
tions A and B are related by a
pseudo-twofold rotation around
the longest axis of the molecule.
The relationship between the
two orientations is similar to
that in the disordered crystals of
(E)-stilbenes and azobenzenes.
Orientations
A and C are
related by inversion. The rela-
tionship between orientations B
and D, and that between C and
D, are identical with that
between A and C, and A and B,
respectively. Disorder arising
from the coexistence of two or
more molecular orientations at
the same site is frequently
observed in the crystal struc-
tures of N-benzylideneanilines
(Bernstein & Schmidt, 1972;
Bernstein & Izak, 1975, 1976;
Bar & Bernstein, 1982, 1987).
The presence of two molecules
(Z = 2) in the unit cell of the
P21/a crystal of (2) requires
Figure 3
Perspective views of N-(4-nitrobenzylidene)aniline (1) with the atom-numbering scheme. The ellipsoids are
drawn at the 50% probability level. H atoms are shown as spheres of a ®xed arbitrary size. The major
molecule at room temperature is drawn with ®lled bonds and shaded ellipsoids. The minor molecule is drawn
with open bonds and spheres.
crystallographic
disorder
cell, P21/a, with Z = 2; Bar & Bernstein, 1983; Haller et al.,
1995). A perspective view of the molecule is shown in Fig. 4.
The crystal structure of (2) was reported to exhibit a four-
fold disorder, as illustrated in Fig. 5. Each molecule in the
crystal adopts one of four alternative orientations. Orienta-
around an inversion center. Consequently, the population of
orientation A is equal to that of C. The same holds for the
relationship between B and D. However, the populations of
orientations A and B are not necessarily equal.
Re®nement with the disorder model gave a ratio of
0.361 (2):0.139 (2) at room temperature, in good agreement
with that reported [0.351 (3):0.149 (3)] (Haller et al., 1995).
The ratio was determined to be 0.389 (2):0.111 (2) and
0.408 (1):0.092 (1) at 200 and 90 K, respectively. The
temperature dependence of the populations clearly shows that
the disorder is dynamic and that a conformational change
takes place through the pedal motion.
3.3. Crystal structure of (3)
The crystal structure observed at 300 K is identical with that
reported in the literature (monoclinic unit cell, P21/c, with Z =
2; Bernstein et al., 1976). A perspective view of the molecule is
shown in Fig. 6.
As reported, the crystal structure of (3) shows fourfold
disorder identical to that of crystal (2), except that molecule
(3) is homo-disubstituted by methyl groups at its para posi-
tions (Fig. 5). As in the crystal of (2), the molecule resides on
an inversion center. The population of orientation A equals
that of C and is not necessarily equal to that of B.
Measurements of the diffraction data of the crystal (3) were
carried out according to the procedure described in x2. As in
the crystals of (1) and (2), lowering of the temperature
increases the population of the major orientation (A) and
decreases that of the minor orientation (B); see Table 2. The
Figure 4
Perspective view of the N-(4-chlorobenzylidene)-4-methylanilines (2)
with the atom-numbering scheme. The ellipsoids are drawn at the 50%
probability level. H atoms are shown as spheres of a ®xed arbitrary size.
The major molecule is drawn with ®lled bonds and shaded ellipsoids. The
minor molecule is drawn with open bonds and spheres.
ꢂ
594 Jun Harada et al.
Conformational changes
Acta Cryst. (2004). B60, 589±597

research papers
Table 2
Populations of the two orientations in the crystals of N-(4-methylbenzyl-
idene)-4-methylaniline (3).
The nonlinearity of the van't Hoff plot and the cooling-rate
dependence of the populations can be explained in terms of a
thermodynamic nonequilibrium state generated by a freezing
in of the conformational change at low temperature. A similar
phenomenon has recently been reported for crystals of (E)-
stilbene and azobenzene (Harada & Ogawa, 2004). The
freezing in of the conformational change during cooling
increases the population of the minor orientation compared
with the equilibrium state and, accordingly, the van't Hoff plot
becomes nonlinear. During the ¯ash cooling, the two orien-
Temperature (K)
300
250
200
150
90
90 (90 K2²)
90 (90 K4²)
0.292 (2):0.208 (2)
0.309 (2):0.191 (2)
0.333 (2):0.167 (2)
0.363 (1):0.137 (1)
0.369 (1):0.131 (1)
0.335 (1):0.165 (1)
0.369 (1):0.131 (1)
² The de®nition of the data is given in x2.
temperature dependence of the populations proves that a
conformational change also takes place in the crystal of (3).
A detailed investigation of the temperature dependence of
the populations revealed that the conformational change is
frozen in at low temperature. Avan't Hoff plot gives an almost
straight line in the high-temperature range (300±200 K). At
the low-temperature range (200±90 K; Fig. 7), however, the
points do not fall on the same straight line.
The populations at 90 K showed a signi®cant dependence
on the cooling rate. In the ¯ash-cooled crystal (data 90 K2; see
x2), the population of the minor orientation (B) increased
compared with that in the slowly cooled crystal (90 K and
90 K4).
Figure 6
Perspective view of the N-(4-methylbenzylidene)-4-methylanilines (3)
with the atom-numbering scheme. The ellipsoids are drawn at the 50%
probability level. H atoms are shown as spheres of a ®xed arbitrary size.
The major molecule is drawn with ®lled bonds and shaded ellipsoids. The
minor molecule is drawn with open bonds and spheres.
Figure 5
Illustrations of the fourfold disorder in the crystals of N-(4-chlorobenzyli-
dene)-4-methylanilines (2) and N-(4-methylbenzylidene)-4-methylanil-
ines (3): (a) orientation A, which is the reference orientation; (b)
orientation B, which is twofold rotation-related with orientation A; (c)
orientation C, which is inversion-related with orientation A; (d)
orientation D, which is inversion-related with orientation B.
Figure 7
van't Hoff plot (ln K versus 1/T) for the two orientations in the crystal of
N-(4-methylbenzylidene)-4-methylanilines (3). The data of 300, 250, 200,
150 and 90 K are denoted by ®lled circles and the data of 90 K2 (¯ash-
cooled) by an open circle. K is the ratio of the populations of the major
orientation (orientation A) to that of the minor one (orientation B).
ꢂ
Acta Cryst. (2004). B60, 589±597
Jun Harada et al.
Conformational changes 595

research papers
tations in the crystal start to deviate from equilibration at a
higher temperature than during the slow cooling, because
faster conformational interconversion is necessary for equili-
bration. Therefore, the deviation from equilibrium and thus
the population of the minor orientation is larger in the ¯ash-
cooled crystal compared with the crystal which is cooled
slowly. The lack of a difference between the two serial data
(90 K2 and 90 K3) observed after ¯ash cooling shows that no
time evolution of the crystal structure takes place at 90 K
during ꢃ 10 h of data collection time and that the conforma-
tional change stops completely at this temperature.
(Cl1) and methyl C (C8) atoms were placed at distances of
Ê
1.74 and 1.53 A from the bonded atom C4 (Fig. 4). They were
also ®xed on the external bisectors of the angles /
AÐCl1A and N1CÐCl1C) were
re®ned anisotropically, those of the minor orientations
(N1BÐCl1B and N1DÐCl1D) were re®ned isotropically. The
chlorine and methyl C atoms of the major orientations were
re®ned with the same anisotropic temperature factors. All N
and C atoms of the minor orientations (N1BÐC7B and
N1DÐC7D) were re®ned with a common isotropic tempera-
ture factor. H atoms were re®ned according to the riding
model. Populations were determined from the same re®ne-
ment as above, except that all atoms were re®ned isotropically
using a common temperature factor for all the N and C atoms
and another isotropic temperature factor for the Cl atoms. The
populations were held constant during the subsequent aniso-
tropic re®nement.
4. Conclusions
In this study we carried out variable-temperature X-ray
diffraction analyses of some N-benzylideneaniline derivatives:
N-(4-nitrobenzylidene)aniline (1), N-(4-chlorobenzylidene)-
4-methylaniline (2) and N-(4-methylbenzylidene)-4-methyl-
aniline (3). In the crystal structures of all the compounds a
dynamic disorder was observed. The dynamic disorder was
accounted for in terms of a conformational change through a
pedal motion in the crystals. The results show that the
conformational change through the pedal motion is a wide-
spread type of molecular motion in crystals. It was also shown
that the conformational change in the crystal of (3) freezes in
at low temperature.
A3. -(4-Methylbenzylidene)-4-methylaniline (3)
The overlapping C and N atoms of orientations A and C
(Fig. 5) were constrained to have the same coordinates,
displacement parameters and occupation factors. The same
constraints were applied to orientations B and D. The benzene
rings of the minor orientations were constrained to be regular
Ê
hexagons with bond lengths of 1.39 A. Non-H atoms of the
major orientations (N1AÐ C8A and N1CÐC8C) were re®ned
anisotropically (Fig. 6) and those of the minor orientations
(N1BÐC8B and N1D± C8D) were re®ned with a single
common isotropic temperature factor. H atoms were re®ned
according to the riding model. Populations were determined
from the same re®nement as above, except that all non-H
atoms were re®ned isotropically using a common temperature
factor. The populations were held constant during the subse-
quent anisotropic re®nement.
APPENDIX
Refinement of the disordered structures
A1. -(4-Nitrobenzylidene)aniline (1) at room temperature
Only ®ve atoms (N1, C1, C7, C8 and H7) were re®ned at
separate positions in the disorder model (Fig. 3). Four atoms
of the major orientation (N1B, C1B, C7B, C8B) were re®ned
anisotropically, and those of the minor orientation (N1C, C1C,
C7C, C8C) were re®ned isotropically. The other non-H atoms
were re®ned anisotropically. The H atoms of the C N bond
(H7B and H7C) were re®ned according to the riding model.
The other H atoms were re®ned isotropically without any
constraint. The length of the CÐPh bond (C1ÐC7) of the two
This work was supported by the Grant-in-Aid for Scienti®c
Research from the Ministry of Education, Culture, Sports,
Science and Technology, Japan, and by Mitsubishi Chemical
Corporation Fund.
Ê
orientations was restrained to be 1.47 A with an e.s.d. of
Ê
0.01 A. The length of the N±Ph bond (N1ÐC8) of the two
Ê
orientations was restrained to be 1.41 A with an e.s.d. of
References
Ê
0.01 A. The length of the C N bond of the two orientations
Ê
was restrained to be equal with an e.s.d. of 0.01 A. Populations
were determined from the same re®nement as above, except
that the eight disordered atoms (N1B, N1C, C1B, C1C, C7B,
C7C, C8B and C8C) were re®ned with a common isotropic
temperature factor. The populations were held constant
during the subsequent anisotropic re®nement.
Bar, I. & Bernstein, J. (1982). Acta Cryst. B38, 121±125.
Bar, I. & Bernstein, J. (1983). Acta Cryst. B39, 266±272.
Bar, I. & Bernstein, J. (1987). Tetrahedron, 43, 1299±1305.
Bernstein, J., Bar, I. & Christensen, A. (1976). Acta Cryst. B32, 1609±
1611.
Bernstein, J. & Izak, I. (1975). J. Cryst. Mol. Struct. 5, 257±266.
Bernstein, J. & Izak, I. (1976). J. Chem. Soc. Perkin Trans. 2, pp. 429±
434.
Bernstein, J. & Schmidt, G. M. J. (1972). J. Chem. Soc. Perkin Trans.
2, pp. 951±955.
Bruker (1998). SMART and SAINT. Bruker AXS Inc., Madison,
Wisconsin, USA.
A2. -(4-Chlorobenzylidene)-4-methylaniline (2)
The overlapping C and N atoms of the orientations A and C
(Fig. 5) were constrained to have the same coordinates,
displacement parameters and occupation factors. The same
constraints were applied to orientations B and D. The chlorine
Bruker (2000). SHELXTL, Version 6.10. Bruker AXS Inc., Madison,
Wisconsin, USA.
È
Burgi, H. B. (2002). Faraday Discuss. 122, 41±63.
Cosier, J. & Glazer, A. M. (1986). J. Appl. Cryst. 19, 105±107.
ꢂ
596 Jun Harada et al.
Conformational changes
Acta Cryst. (2004). B60, 589±597

research papers
Galli, S., Mercandelli, P. & Sironi, A. (1999). J. Am. Chem. Soc. 121,
3767±3772.
Gavezzotti, A. & Simonetta, M. (1982). Chem. Rev. 82, 1±13.
Gavezzotti, A. & Simonetta, M. (1987). Organic Solid State
Chemistry. Studies in Organic Chemistry 32, edited by G. R.
Desiraju, ch. 11. Amsterdam: Elsevier.
Haller, K. J., Rae, A. D., Heerdegen, A. P., Hockless, D. C. R. &
Welberry, T. R. (1995). Acta Cryst. B51, 187±197.
Harada, J., Harakawa, M. & Ogawa, K. (2004). Acta Cryst. B60, 578±
588.
Harada, J. & Ogawa, K. (2001). J. Am. Chem. Soc. 123, 10884±
10888.
McGeorge, G., Harris, R. K., Batsanov, A. S., Churakov, A. V.,
Chippendale, A. M., Bullock, J. F. & Gan, Z. (1998). J. Phys. Chem.
A, 102, 3505±3513.
Ohba, S., Hosomi, H. & Ito, Y. (2001). J. Am. Chem. Soc. 123, 6349±
6352.
Saito, K., Okada, M., Akutsu, H., Sato, A. & Sorai, M. (2004). J. Phys.
Chem. B, 108, 1314±1320.
Saito, K., Okada, M., Akutsu, H. & Sorai, M. (2000). Chem. Phys.
Lett. 318, 75±78.
Saito, K., Yamamura, Y., Kikuchi, K. & Ikemoto, I. (1995). J. Phys.
Chem. Solids, 56, 849±857.
Sheldrick, G. M. (1990). Acta Cryst. A46, 467±473.
È
Sheldrick, G. M. (1997). SHELX97. University of Gottingen,
Germany.
Sheldrick, G. M. (2002). SADABS, Version 2.03 and 2.05. University
Harada, J. & Ogawa, K. (2004). J. Am. Chem. Soc. 126, 3539±3544.
Harada, J., Ogawa, K. & Tomoda, S. (1997). Acta Cryst. B53, 662±672.
Harada, J., Uekusa, H. & Ohashi, Y. (1999). J. Am. Chem. Soc. 121,
5809±5810.
Harris, K. D. M. & Aliev, A. E. (1995). Chem. Br. pp. 132±136.
Ito, Y., Hosomi, H. & Ohba, S. (2000). Tetrahedron, 56, 6833±6844.
È
of Gottingen, Germany.
Ueda, Y., Nakamura, N. & Chihara, H. (1988). J. Phys. Soc. Jpn, 57,
4063±4064.
ꢂ
Acta Cryst. (2004). B60, 589±597
Jun Harada et al.
Conformational changes 597