Asymmetrically substituted distyrylbenzenes and their polar crystal structures

Article abstract of DOI:10.1039/c0nj00732c

The synthesis of twelve asymmetric donor-acceptor distyrylbenzene derivatives with either one nitrile group or one, two or three nitro groups as electron acceptors, and one, two or three methoxy groups as electron donors is reported. Peak potentials obtained from cyclic voltammetry were combined with experimental UV/Vis data and molecular dipole moments obtained from quantum chemical calculations, yielding insight into the influence of the positions of the substituents on the electronic structure and charge distribution of this as yet unexplored class of organic semiconductors. The supramolecular structures of five of these compounds have been studied using single-crystal X-ray diffraction to monitor the influence of the positions of donor and acceptor groups on the organisation of the molecules in the solid state, and three crystal structures have been identified in which the molecular dipoles do not organize themselves in a centrosymmetric lattice. Analysis of the dipoles in the unit cell yields further insight into the possible non-linear optical properties of these three polar structures.

Full text of DOI:10.1039/c0nj00732c

View Article Online / Journal Homepage / Table of Contents for this issue
Dynamic Article Links
NJC
Cite this: New J. Chem., 2011, 35, 649–662
www.rsc.org/njc
PAPER
Asymmetrically substituted distyrylbenzenes and their polar crystal
structuresw
Alain Collas,a Roeland De Borger,a Tatyana Amanova,a
a
Christophe M. L. Vande Velde,z Jan K. Baeke,a Roger Dommisse,b
Christian Van Alsenoya and Frank Blockhuys*a
Received (in Victoria, Australia) 24th September 2010, Accepted 17th December 2010
DOI: 10.1039/c0nj00732c
The synthesis of twelve asymmetric donor–acceptor distyrylbenzene derivatives with either one
nitrile group or one, two or three nitro groups as electron acceptors, and one, two or three
methoxy groups as electron donors is reported. Peak potentials obtained from cyclic voltammetry
were combined with experimental UV/Vis data and molecular dipole moments obtained from
quantum chemical calculations, yielding insight into the influence of the positions of the
substituents on the electronic structure and charge distribution of this as yet unexplored class of
organic semiconductors. The supramolecular structures of five of these compounds have been
studied using single-crystal X-ray diffraction to monitor the influence of the positions of donor
and acceptor groups on the organisation of the molecules in the solid state, and three crystal
structures have been identified in which the molecular dipoles do not organize themselves in a
centrosymmetric lattice. Analysis of the dipoles in the unit cell yields further insight into the
possible non-linear optical properties of these three polar structures.
the properties of these oligomeric organic semiconductor
1. Introduction
materials for their own sake.
Organic semiconductors are materials that enjoy considerable
The structure and electronic properties of one particular
interest these days mainly because of the convenient and easy
class of organic semiconductors, i.e., 1,4-distyrylbenzene
way in which their molecular structures and hence their
(DSB) and its derivatives, which are oligomers of poly-
(p-phenylene vinylene), have already been extensively studied,3–19
properties can be tuned to a specific application. This is in
stark contrast to the present-day CMOS (Complementary
indicating the usefulness of the 1,4-distyrylbenzene structure
Metal Oxide Semiconductor) technology where such tuning
as a template for materials with interesting electro-optical
is virtually impossible. Switching from polymeric to oligomeric
properties. So far, though, these investigations have focused
organic materials is a worthwhile undertaking for the numerous
almost exclusively on symmetrical derivatives, even though a
opto-electronic applications for which these new compounds
number of possible applications of these materials, such as
are used, as has been discussed in depth by Mullen and
¨
Wegner1 and Segura and Martın.2 The most important reason
´
non-linear optics (NLO) and non-volatile organic memories,
require materials with large dipole moments which can be
found only in asymmetrical systems. The most convenient way
of achieving this is by introducing electron-donating on one
and electron-accepting substituents on the other side of the
distyrylbenzene scaffold, thus creating conjugated asymmetrical
push–pull systems which exhibit the desired large molecular
dipoles. Since the synthetic pathways towards distyrylbenzenes
(traditionally the Wittig reaction) allow for a wide variety of
substituents to be introduced at various positions on the
backbone, these oligomers have the potential to generate a
large number of suitable materials for these applications.
[Note that a number of asymmetric DSBs have been known
for some time (see below), but these have remained un-
characterised until now, apart from a single compound bearing
fluorine and tert-butyl substituents (CSD refcode20 REFKUI)11
of which only the crystal structure was reported.]
for making this transition is the fact that it is far easier to
produce, purify and structurally characterise oligomers than
polymers, as they are monodisperse materials. For these
reasons oligomers have been extensively used as model
compounds for polymers but it is far more rewarding to study
a Department of Chemistry, University of Antwerp,
Universiteitsplein 1, B-2610 Antwerp, Belgium.
E-mail: frank.blockhuys@ua.ac.be; Fax: +32 3 820 23 10
b Department of Chemistry, University of Antwerp,
Groenenborgerlaan 171, B-2020 Antwerpen, Belgium
w Electronic supplementary information (ESI) available. CCDC
reference numbers 760975 (5b-1), 760976 (5b-2), 760977 (5c) and
760979 (5i). For ESI and crystallographic data in CIF or other
electronic format see DOI: 10.1039/c0nj00732c
z Current address: Karel de Grote University College, Department of
Applied Engineering, Salesianenlaan 30, B-2660 Antwerp, Belgium.
c
This journal is The Royal Society of Chemistry and the Centre National de la Recherche Scientifique 2011
New J. Chem., 2011, 35, 649–662 649

View Article Online
Fig. 1 Synthetic pathway to and structural formulas and numbering scheme of oligomers 5a-1. Note that 5l was prepared via a different route; see
text for details.
Yet, the fact that applications of organic semiconductors
generally involve the solid state implies that, apart from the
molecular properties, also the properties of the supramolecular
structure are of great importance. Indeed, if, for example, the
crystal structure is centrosymmetric, the molecular dipoles of
asymmetric oligomers are cancelled out by the centre of symmetry
in the crystal, thus nullifying the carefully tuned molecular
properties generated by the substitution pattern. Since it is still
impossible to predict, let alone to control, the space group in
which a compound will crystallize, an empirical approach in
which the preparation of new materials is followed by the analysis
of the supramolecular structure, remains the only option when
assessing the applicability of a compound.
(1),21 E-1-(4-methylphenyl)-2-(4-nitrophenyl)ethene (2a),22
E-1-(4-methylphenyl)-2-(3-nitrophenyl)ethene (2b),23 E-1-
(4-methylphenyl)-2-(4-cyanophenyl)ethene (2d),18 E-1-(4-bromo-
methylphenyl)-2-(4-nitrophenyl)ethene (3a),10 E-1-(4-bromo-
methylphenyl)-2-(2,4-dinitrophenyl)ethene (3c),24 E-1-(4-bromo-
methylphenyl)-2-(4-cyanophenyl)ethene (3d),18 E-triphenyl-
[4-(2-(4-nitrophenyl)ethenyl)benzyl]phosphonium bromide
(4a),10,24
E-triphenyl[4-(2-(4-cyanophenyl)ethenyl)benzyl]phos-
phonium bromide (4d),24 E,E-1-[2-(4-nitrophenyl)ethenyl]-4-
[2-(2,4,6-trimethoxyphenyl)ethenyl]benzene (5a),10,24 E,E-1-
[2-(4-nitrophenyl)ethenyl]-4-[2-(2,4-dimethoxyphenyl)ethenyl]-
benzene (5e),24 E,E-1-[2-(4-nitrophenyl)ethenyl]-4-[2-(4-methoxy-
phenyl)ethenyl]benzene (5f)24 and E,E-1-[2-(2,4-dinitrophenyl)-
ethenyl]-4-[2-(2,4,6-trimethoxyphenyl)ethenyl]benzene (5h)24
were prepared as previously reported. All benzaldehydes used
in the Wittig reactions are commercially available, with the
exception of 2,4,6-trinitrobenzaldehyde (7) which was prepared25
from 2,4,6-trinitrotoluene (6).26 Consequently, 5l was obtained
via an adjusted method allowing 7 to be used in the final step:
one equivalent of 2,4,6-trimethoxybenzaldehyde reacted
with 1,4-bis(triphenylphosphoniummethyl)benzene dibromide,
In this paper the synthesis and characterisation of twelve
new asymmetrical ‘‘donor–acceptor’’ (DA) distyrylbenzenes,
containing (one) nitrile or (one, two or three) nitro groups as
electron-accepting and (one, two or three) methoxy groups as
electron-donating substituents, are reported; the structures of
these twelve oligomers (5a–l) are presented in Fig. 1. Various
combinations of these substituents were used in order to
investigate the influence of the number and position of the
functionalities on the electronic and (supra)molecular structures
of the compounds. The new compounds’ electronic properties
were examined using UV/Vis spectroscopy, cyclic voltammetry
(CV) and quantum chemical calculations using Density
Functional Theory (DFT), and we present the first systematic
study of the solid-state structures of asymmetric DSBs,
determined using single-crystal X-ray diffraction (XRD).
yielding
E-triphenyl[4-(2-(2,4,6-trimethoxyphenyl)ethenyl)-
benzyl]phosphonium bromide (8), which was condensed in a
Wittig reaction with 2,4,6-trinitrobenzaldehyde (7), yielding 5l.
2.2 Molecular properties
2.2.1 Calculated structures. Geometries of the isolated
molecules were optimized in their anti conformation, in which
the two ethenyl spacers are oriented in opposite directions with
respect to the central phenyl ring. The resulting structures
present no surprises as all but three of the twelve compounds
are planar (Cs symmetry). In compound 5b, the methoxy
group in the 4-position points out of the plane of the peripheral
benzene ring. Compounds 5h and 5l have nitro and dinitro
substitution in the 2- and 2,6-positions of the acceptor ring,
2. Results and discussion
2.1 Syntheses
The synthetic pathway for the preparation of 5a–k and
the numbering schemes of the compounds are shown in
Fig. 1. Triphenyl(4-methylbenzyl)phosphonium chloride
c
650 New J. Chem., 2011, 35, 649–662
This journal is The Royal Society of Chemistry and the Centre National de la Recherche Scientifique 2011

View Article Online
Table 1 Calculated dipole moments m (in D) and ionization poten-
tials IP (in eV), experimental peak potentials Epa (in mV vs. Fc/Fc+),
experimental UV/Vis absorption maxima lmax (in nm) and logarithms
of the extinction coefficient e (in l molꢀ1 cmꢀ1) of compounds 5a–5l.
The values in parentheses in the second column represent the
calculated dipole moments (in D) of the molecules in their solid-state
geometries. See text for details
voltammetry; the insolubility of 5l prevented any measurements
for this compound. None of the compounds showed a reduction
peak and, as a consequence, the formal oxidation potentials
could not be determined. Therefore, the anodic peak potentials
(Epa) are listed in Table 1. Measurements at different scan
rates between 0.03 V sꢀ1 and 1 V sꢀ1 showed no significant
shift of the anodic peak potentials with respect to the oxidation
potential of ferrocene, which leads to the conclusion that the
absence of the reverse peak is due to a follow-up chemical
reaction rather than due to the irreversibility of the system.
This has also been observed for a series of self-doping distyryl-
benzene derivatives (see ref. 19 and references therein).
Therefore, the anodic peak potentials can be used as a measure
for the oxidation potential.
m
IP
Epa
lmax
log e
5a
5b
5c
5d
5e
5f
5g
5h
5i
12.20
4.96
5.35
5.47
5.17
5.11
5.30
4.88
5.11
4.90
5.08
5.24
5.19
456
702
938
715
581
742
479
527
463
582
742
—
415
197
390
393
408
400
379
449
390
383
376
502
4.70
4.61
4.74
4.21
4.55
4.50
4.63
4.46
4.69
4.15
4.74
4.09
6.32 (6.19)a
5.79 (9.91)
10.54
10.87 (8.71)
9.01
10.36
13.61
11.33
9.71
8.25
12.41
When assessing the influence of the electron accepting
substituents, a first observation that can be derived from the
data in Table 1 is the fact that a cyano group and a nitro group
have the same influence on the electronic structure: the peak
potentials of the nitro-substituted 5a, 5e and 5f are quite close
to those of their cyano-substituted counterparts 5i, 5j and 5k,
respectively. This effect is also apparent from the calculated
ionization potentials (IP), as similar trends are found. The
addition of a nitro group (5a - 5h) increases the peak
potential, which can be easily linked to the stabilizing effect
of the electron accepting groups on the highest occupied
molecular orbital (HOMO), making the oxidation process
more difficult. The addition of a third nitro group to the
acceptor ring (5h - 5l) can only be inferred from the calculations
which do predict a small further increase in the oxidation
potential. Moving the single nitro group from the 4-position in
5a to the 3-position in 5g has a relatively small effect.
5j
5k
5l
a
For 5b-1.
respectively, and due to the steric hindrance caused by these
nitro groups, the molecules deviate from planarity: the
C2–C1–C7–C8 torsion angles are 22.11 and 32.61 for 5h and
5l, respectively. These nitro groups are themselves twisted out
of the plane of the acceptor ring by 23.51 for the ortho-nitro
group of 5h, and 26.21 and 56.71 for those in 5l, whereas the
nitro groups in the para position remain in the plane of
the ring.
2.2.2 Calculated dipole moments. Considering that the
semiconductor applications of these materials mentioned in
the Introduction require compounds with large dipole
moments, the calculated dipole moments were analyzed in
order to investigate the influence of the nature and position of
the substituents on their values, which are given in Table 1.
The dipole moments of the two related series 5a/5e/5f (4-nitro
substitution) and 5i/5j/5k (4-cyano substitution) indicate that
the dipole moment increases when adding an extra methoxy
group by approximately 1.5 D, as expected. The values are
also consistently higher for the 4-nitro compounds than for the
4-cyano compounds, indicating that the nitro group is a better
electron acceptor. Obviously, 2-, 2,4- and 2,4,6-substitutions
(both for acceptor and donor groups) lead to a better charge
separation and larger dipole moments than 3- or 3,5-substitution
due to the favorable mesomeric conjugation, as can be seen
when comparing the values of 5c and 5e, 5a and 5g, and 5a and
5b. The values for 2,6- and 2,4-dimethoxy substitutions (5d
and 5e, respectively) are comparable. Although adding one
extra nitro group follows the expected trend (5f - 5h), the
2,4,6-trinitro-substituted oligomer (5l) shows a decrease in the
dipole moment of 1.20 D not withstanding the additional
electron acceptor. The reason for this is most likely the
deviation from planarity of the nitro-substituted ring with
respect to the rest of the conjugated system and the out-of-
plane positions of both ortho-nitro groups themselves, leading
to a decrease in conjugation.
When considering the influence of the number and positions
of the electron donating methoxy groups, it is clear that the
addition of a methoxy group considerably decreases the peak
potential, as seen in the series 5f - 5e - 5a and 5k - 5j - 5i.
This can be rationalized when taking into account the
destabilization of the HOMO by the increasing number of
electron donating groups, leading to an easier oxidation and
hence a lower peak potential. As expected, compounds with
methoxy substituents in the 3- and 5-positions fare a lot worse
(5b vs. 5a, 5c vs. 5d and 5e): the poorer mesomeric and
directing properties of these electron donating groups in the
meta-position(s) are clear. Indeed, when comparing 5c and 5f,
the peak potential of the latter is much lower even though the
former has one extra methoxy group.
As can be seen from the values listed in Table 1, the
calculated ionization potentials follow the same trends as the
peak potentials. When the Epa values are plotted against these
adiabatic ionization potentials, a linear correlation is obtained
with R2 = 0.88. This is not a very good correlation per se
and the lower R2 value mainly reflects the neglect of solvent
effects. In any case, the correlation is better than the one
found earlier for a series of symmetrical distyrylbenzene
derivatives.5
These substituent effects can also be observed in the UV/Vis
spectra, as can be seen from Table 1. Removing methoxy
groups from the backbone decreases lmax for both the 4-nitro
(5a/5e/5f) and the 4-cyano (5i/5j/5k) series of compounds.
Adding nitro groups (5a - 5h - 5l) increases lmax. Moving
2.2.3 Electronic properties. The electrochemical character-
istics of compounds 5a–5k were studied using cyclic
c
This journal is The Royal Society of Chemistry and the Centre National de la Recherche Scientifique 2011
New J. Chem., 2011, 35, 649–662 651

View Article Online
the single nitro group from the 4-position in 5a to the 3-position in
5g decreases lmax
the syn conformation, with the two ethenyl spacers oriented in
the same direction with respect to the central phenyl ring, and
is slightly bent along its DA-axis (Fig. 2). The ethenylic link at
the donor end of the molecule is disordered, consisting of a
1801 pedal-like twist of the double bond (Fig. 3), resulting in
about 36% of anti conformer. The three benzene rings and
two ethenyl spacers in 5b-1 are all virtually co-planar, with
dihedral angles between the various rings and spacers of less
than about 71 (Table 4).
The packing scheme given in Fig. 2 indicates that the
molecules are placed head-to-tail and form ribbons through
one CHꢁ ꢁ ꢁO contact between one of the methoxy groups and
the nitro group, i.e., C30–H30Bꢁ ꢁ ꢁO1. These ribbons are
stacked in an anti-parallel fashion by grouping the methoxy-
substituted moieties together, connecting successive ribbons
by five CHꢁ ꢁ ꢁO and two CHꢁ ꢁ ꢁp contacts. In one direction
these are C40–H40Aꢁ ꢁ ꢁO24, C40–H40Aꢁ ꢁ ꢁO25 and
C50–H50Aꢁ ꢁ ꢁCg(B) [not given in Fig. 2; Cg(B) is the centroid
of ring B]. In a second direction these are C15–H15ꢁ ꢁ ꢁO24,
C16–H16ꢁ ꢁ ꢁO1 (not given in Fig. 2), C17–H17ꢁ ꢁ ꢁO23 and
C30–H30Cꢁ ꢁ ꢁCg(C). Thus, alternating layers of molecules,
oriented in an anti-parallel fashion, are formed, creating the
typical herringbone structure also found in 5a.10
.
2.3 Supramolecular structures
Attempts were made to grow single crystals suitable for XRD
of all oligomers using various crystallization techniques.
However, only for compounds 5a, 5b, 5c, 5e and 5i crystals
of sufficient size and quality were obtained. The crystal
structure of 5a was reported earlier.10 Details of the data
collection and structural refinement of 5b (two polymorphs,
5b-1 and 5b-2), 5c and 5i are collected in Table 2. The solid-
state structure of 5e turned out to be dynamically disordered
and this disorder was the subject of a separate study;27 the
latter paper contains the details of the data collection
and structural refinement of 5e, but a description of the
291 K structure is given below. For all compounds the
carbon–hydrogen distances were normalized to 1.083 A
after the refinement and the resulting geometrical parameters
were used in the following discussion of the different inter-
and intramolecular short contacts involving these hydrogen
atoms. The details of these contacts are summarized in
Table 3.
The second, yellow polymorph (5b-2) crystallizes in the
ꢀ
centrosymmetric space group P1, with two molecules in the
2.3.1 E,E-1-[2-(4-Nitrophenyl)ethenyl]-4-[2-(3,4,5-trimethoxy-
phenyl)ethenyl]benzene (5b). Compound 5b is crystallized as a
mixture of two polymorphs. The first, red polymorph (5b-1)
crystallizes in the non-centrosymmetric, polar space group P21
(Table 2). This implies that the molecules do not pair up in a
head-to-tail fashion to form centrosymmetric pairs, and that
the crystal has a macroscopic polarity. The molecule adopts
asymmetric unit (Table 2). It transforms into the red polymorph
(5b-1) upon heating. The two molecules have similar geometries:
both are found in the anti conformation with a quasi-planar
nitro-stilbene moiety, but with a considerably more distorted
trimethoxy-stilbene fragment (Table 4).
Table 2 Details of the data collection and structural refinement for 5b (two polymorphs, 5b-1 and 5b-2), 5c and 5i
5b-1
5b-2
5c
5i
Formula
Formula weight
Crystal colour
Crystal size/mm
Crystal form
Crystal system
Space group
No. of refl. for cell parameters
C
417.44
Red
25H23NO5
C25H23NO5
417.44
Yellow
C24H21NO4
387.42
Red
C26H23NO3
397.45
Yellow
0.4 ꢂ 0.3 ꢂ 0.1
0.3 ꢂ 0.3 ꢂ 0.1
0.3 ꢂ 0.3 ꢂ 0.3
0.4 ꢂ 0.3 ꢂ 0.1
Plate
Prism
Triclinic
Prism
Prism
Triclinic
Monoclinic
P21
25
Monoclinic
P21
25
ꢀ
P1
25
ꢀ
P1
25
y-range (1) for cell measurement
a/A
b/A
c/A
a (1)
b (1)
7.48–17.49
7.856(2)
7.767(2)
17.596(6)
90.00
95.93(2)
90.00
1067.9(5)
2
7.49–17.67
10.221(2)
15.239(3)
15.762 (4)
117.42(2)
92.26(2)
93.52(2)
2168.8(8)
4
5.79–15.90
7.219(2)
7.402(2)
18.163(2)
90.00
95.29(3)
90.00
966.4(5)
2
5.55–14.65
10.203(3)
12.411(2)
18.442(2)
74.350(10)
82.49(3)
70.76(3)
2120.9(7)
4
g (1)
V/A3
Z
T
302(2)
1.298
o/2y scans
2.33
25.33
4025
2105
278
4
63.9(6)
1.047
0.1356
0.0449
0.1057
300(2)
1.278
o/2y scans
1.46
25.32
8388
7906
565
0
—
293(2)
1.331
o/2y scans
1.13
25.33
3989
1919
262
1
—
293(2)
1.245
o/2y scans
1.15
25.32
8212
7741
548
7
85.1(6)
1.004
0.2033
0.0661
0.1643
D/Mg mꢀ3
Data collection method
ymin (1)
ymax (1)
Reflections collected
Reflections used in refinement
Parameters used
Restraints
S.O.F. major conformer (%)
GoOF
Rw
Ru
0.975
1.133
0.1867
0.0610
0.1854
0.1752
0.0594
0.0875
Rall
c
This journal is The Royal Society of Chemistry and the Centre National de la Recherche Scientifique 2011
652 New J. Chem., 2011, 35, 649–662

View Article Online
Table 3 Details (distances d in A, angles y in degrees and symmetry
codes) of the short inter- and intramolecular contacts given in the text;
the lack of an esd on the parameters is due to the normalisation of the
CH distances
Contact
d
y
Symmetry code
5a
C17–H17ꢁ ꢁ ꢁO22
C18–H18ꢁ ꢁ ꢁO26
C2–H2ꢁ ꢁ ꢁCg(C)
C6–H6ꢁ ꢁ ꢁCg(B)
C3–H3ꢁ ꢁ ꢁO24
2.18 117 (Intramolecular)
2.28 102 (Intramolecular)
2.84 138 ꢀx, ꢀ1/2 + y, 3/2 ꢀ z
2.59 166 x, 3/2 ꢀ y, 1/2 + z
2.55 153 ꢀx, ꢀ1/2 + y, 3/2 ꢀ z
2.45 135 ꢀx, 1 ꢀ y, 2 ꢀ z
2.63 119 1 + x, y, ꢀ1 + z
2.20 157 1 + x, 3/2 ꢀ y, ꢀ1/2 + z
2.70 141 x, 3/2 ꢀ y, ꢀ1/2 + z
2.37 148 ꢀ1 + x, y, 1 + z
2.59 138 ꢀ1 – x, 1/2 + y, 1 ꢀ z
2.53 156 ꢀ1 ꢀ x, 1/2 + y, 1 ꢀ z
2.87 159 ꢀ1 + x, y, z
Fig. 3 ORTEP drawing of 5b-1, showing the disorder of the ethenylic
link at the donor end of the molecule. Ellipsoids are drawn at the 30%
probability level.
C22–H22Aꢁ ꢁ ꢁO2
C24–H24Aꢁ ꢁ ꢁO2
C25–H25ꢁ ꢁ ꢁO1
C22–H22Cꢁ ꢁ ꢁO26
5b-1 C30–H30Bꢁ ꢁ ꢁO1
C40–H40Aꢁ ꢁ ꢁO24
C40–H40Aꢁ ꢁ ꢁO25
C50–H50Aꢁ ꢁ ꢁCg(B)
C15–H15ꢁ ꢁ ꢁO24
The packing scheme given in Fig. 4 indicates that the two
molecules in the asymmetric unit are arranged in a parallel
fashion and connected by one CHꢁ ꢁ ꢁO contact
[C50–H50Cꢁ ꢁ ꢁO924]. Furthermore, molecule 1 is arranged in
anti-parallel pairs connected by two CHꢁ ꢁ ꢁO interactions
[C6–H6ꢁ ꢁ ꢁO24 and C7–H7ꢁ ꢁ ꢁO23 (not given in Fig. 4)]. These
interactions lead to the ‘building block’ of four molecules
displayed in Fig. 4. These building blocks are stacked through
five CHꢁ ꢁ ꢁO interactions (not given in Fig. 4) involving the
2.57 160 ꢀx, 1/2 + y, 1 ꢀ z
2.69 145 2 ꢀ x, 1/2 + y, ꢀz
2.59 129 ꢀx, 1/2 + y, 1 ꢀ z
2.77 136 ꢀx, 1/2 + y, 1 ꢀ z
2.34 163 x, y, z
C16–H16ꢁ ꢁ ꢁO1
C17–H17ꢁ ꢁ ꢁO23
C30–H30Cꢁ ꢁ ꢁCg(C)
5b-2 C50–H50Cꢁ ꢁ ꢁO924
C6–H6ꢁ ꢁ ꢁO24
2.31 173 1 ꢀ x, 1 ꢀ y, ꢀz
2.70 127 1 ꢀ x, 1 ꢀ y, ꢀz
2.56 171 1 ꢀ x, 2 ꢀ y, 1 ꢀ z
2.56 138 x, ꢀ1 + y, z
C7–H7ꢁ ꢁ ꢁO23
C940–H41Bꢁ ꢁ ꢁO23
C912–H912ꢁ ꢁ ꢁO24
C912–H912ꢁ ꢁ ꢁO25
C5–H5ꢁ ꢁ ꢁO923
methoxy groups of molecule
1
[C940–H41Bꢁ ꢁ ꢁO23,
2.68 125 x, ꢀ1 + y, z
C912–H912ꢁ ꢁ ꢁO24 and C912–H912ꢁ ꢁ ꢁO25] and molecule 2
[C5–H5ꢁ ꢁ ꢁO923 and C40–H40Bꢁ ꢁ ꢁO923] as acceptors.
Furthermore, five CHꢁ ꢁ ꢁO contacts (not given in Fig. 4) can
be observed involving the nitro groups of molecule 1
[C930–H31Cꢁ ꢁ ꢁO1 and C926–H926ꢁ ꢁ ꢁO2] and molecule 2
[C22–H22ꢁ ꢁ ꢁO91, C30–H30Cꢁ ꢁ ꢁO91 and C26–H26ꢁ ꢁ ꢁO92].
Clearly, the presence of two molecules in the asymmetric unit
leads to a considerably larger number of short contacts. Six
additional weaker interactions (five N–Oꢁ ꢁ ꢁp and one pꢁ ꢁ ꢁp)
have been compiled in Table S1 of the ESIw (note that when
two molecules are present in the asymmetric unit, rings A, B
and C of molecule 1 are renamed to D, E and F, respectively,
in molecule 2).
2.49 143 x, ꢀ1 + y, ꢀ1 + z
2.36 164 1 ꢀ x, 2 ꢀ y, 1 ꢀ z
2.71 119 1 ꢀ x, ꢀy, ꢀz
C40–H40Bꢁ ꢁ ꢁO923
C930–H31Cꢁ ꢁ ꢁO1
C926–H926ꢁ ꢁ ꢁO2
C22–H22ꢁ ꢁ ꢁO91
2.67 148 ꢀx, ꢀy, ꢀz
2.70 168 1 ꢀ x, ꢀy, ꢀz
C30–H30Cꢁ ꢁ ꢁO91
C26–H26ꢁ ꢁ ꢁO92
2.56 112 1 ꢀ x, ꢀy, ꢀz
2.57 138 ꢀx, ꢀy, ꢀz
5c
5e
C30–H30Bꢁ ꢁ ꢁO1
C30–H30Cꢁ ꢁ ꢁO23
C30–H30Aꢁ ꢁ ꢁCg(C)
C18–H18ꢁ ꢁ ꢁO26
C26–H26Bꢁ ꢁ ꢁO1
C24–H24Cꢁ ꢁ ꢁO1
C13–H13ꢁ ꢁ ꢁO24
C16–H16ꢁ ꢁ ꢁO24
C15–H15ꢁ ꢁ ꢁO2
2.69 115 2 + x, y, 1 + z
2.54 152 2 ꢀ x, ꢀ1/2 + y, 1 ꢀ z
2.94 120 2 ꢀ x, 1/2 + y, 1 ꢀ z
2.49
91 (Intramolecular)
2.49 162 ꢀ1 + x, ꢀy, 1/2 + z
2.40 173 ꢀ1 + x, 2 + y, z
2.70 166 x, 2 ꢀ y, ꢀ1/2 + z
2.53 156 1 + x, ꢀ1 + y, z
2.40 129 x, ꢀy, 1/2 + z
2.71 138 x, 1 + y, z
2.12 120 (Intramolecular)
2.25 102 (Intramolecular)
2.09 120 (Intramolecular)
2.25 102 (Intramolecular)
2.60 136 x, y, z
C24–H24Bꢁ ꢁ ꢁO26
C17–H17ꢁ ꢁ ꢁO26
C18–H18ꢁ ꢁ ꢁO22
C917–H917ꢁ ꢁ ꢁO922
C918–H918ꢁ ꢁ ꢁO926
C960–H61Bꢁ ꢁ ꢁO22
C23–H23ꢁ ꢁ ꢁN1
5i
2.3.2 E,E-1-[2-(4-Nitrophenyl)ethenyl]-4-[2-(3,5-dimethoxy-
phenyl)ethenyl]benzene (5c). Compound 5c crystallizes in the
non-centrosymmetric, polar space group P21 (Table 2). This
compound too displays the anti conformation and the
molecules are nearly planar (Table 4). The packing
of 5c (Fig. 5) shows a large number of similarities with the
supramolecular structure of 5b-1: molecules are organized in
2.41 162 ꢀ1 + x, y, 1 + z
C95–H95ꢁ ꢁ ꢁO26
2.61 132 1 ꢀ x, 1 ꢀ y, ꢀz
C960–H61Cꢁ ꢁ ꢁCg(D) 2.72 133 1 ꢀ x, 1 ꢀ y, ꢀz
Fig. 2 Packing scheme of 5b-1, showing the various contacts and the bent form of the molecule. Only the major conformer is shown. See Table 3
for the relevant parameters and symmetry codes.
c
This journal is The Royal Society of Chemistry and the Centre National de la Recherche Scientifique 2011
New J. Chem., 2011, 35, 649–662 653

View Article Online
Table 4 Selected torsion angles (in degrees) for 5b (two polymorphs, 5b-1 and 5b-2), 5c, 5e and 5i. For disordered systems, only the values of the
major conformer are given. The numbering of Fig. 1 is used
5b-2
5i
Molecule 1
Molecule 2
Molecule 1
Molecule 2
5b-1
5c
5e
C2–C1–C7–C8
C1–C7–C8–C11
C7–C8–C11–C12
C15–C14–C17–C18
C14–C17–C18–C21
C17–C18–C21–C22
ꢀ5.6(8)
ꢀ176.5(5)
4.8(8)
178.2(6)
173.6(6)
ꢀ3.7(11)
ꢀ4.4(6)
ꢀ2.7(7)
ꢀ176.2(4)
3.6(6)
ꢀ8.6(8)
ꢀ177.9(5)
8.5(8)
0.5(8)
14.2(7)
176.3(4)
3.9(7)
ꢀ179.6(3)
ꢀ172.0(3)
169.5(3)
170.0(4)
179.1(4)
178.1(4)
ꢀ0.1(6)
ꢀ174.4(3)
6.5(6)
178.0(3)
ꢀ2.7(6)
ꢀ176.8(5)
ꢀ5.9(8)
ꢀ32.2(9)
177.6(6)
ꢀ33.4(5)
ꢀ20.4(6)
ꢀ175.7(3)
ꢀ19.0(6)
13.3(6)
ꢀ174.8(4)
2.5(8)
ꢀ177.1(5)
11.7(6)
0.5(8)
Fig. 4 View of the ‘building block’ of 5b-2, showing the relevant CHꢁ ꢁ ꢁO contacts. See Table 3 for the relevant parameters and symmetry codes.
head-to-tail ribbons through a CHꢁꢁꢁO contact [C30–H30BꢁꢁꢁO1]
and the ribbons are stacked in anti-parallel layers forming a
herringbone structure. The methoxy-substituted moieties are
connected by one CHꢁꢁꢁO contact [C30–H30CꢁꢁꢁO23] and one
CHꢁꢁꢁp contact [C30–H30AꢁꢁꢁCg(C), not given in Fig. 5]. Five
additional weaker CHꢁꢁꢁp interactions involving hydrogen atoms
on the rings have been compiled in Table S1 (ESIw).
the type of disorder is the same as what was observed for 5b-1.
The disorder was fully refined and additional measurements at
lower and higher temperatures were performed to study the
nature of the disorder, which proved to be dynamic; the full
details of this study have been published elsewhere.27
Considering that 5a,10 5b, 5c and 5i (see below) are planar to
within about 201, 5e is somewhat of an outlier. The reason for
this can be found in the intermolecular interactions of the
methoxy-substituted ring C (Fig. 6 and 7). An intramolecular
CHꢁ ꢁ ꢁO contact can be found between the methoxy group
positioned ortho to the ethenylic link and the hydrogen atoms
of that ethenylic link [C18–H18ꢁ ꢁ ꢁO26], as expected9 (Table 3).
The methoxy group in the 2-position of ring C engages in a
CHꢁ ꢁ ꢁO contact with the nitro group of a neighbouring
molecule [C26–H26Bꢁ ꢁ ꢁO1] as is shown in Fig. 6. The methoxy
group in the 4-position is also involved in such a contact
[C24–H24Cꢁ ꢁ ꢁO1] which is presented in Fig. 7. This same
methoxy group simultaneously acts as a hydrogen bond
acceptor for two phenyl hydrogen atoms of two different
2.3.3 E,E-1-[2-(4-Nitrophenyl)ethenyl]-4-[2-(2,4-dimethoxy-
phenyl)ethenyl]benzene (5e). Compound 5e crystallizes in the
non-centrosymmetric, polar space group Pc, again implying
macroscopic polarity due to the fact that the molecules do not
pair up in a head-to-tail fashion to form centrosymmetric
pairs. The molecules display the anti conformation. The main
feature that is immediately obvious from the packing schemes
given in Fig. 6 and 7 is the large deviation of the methoxy-
substituted ring C from co-planarity with the rest of the
molecule (Table 4). The ethenylic link between rings A and
ring B—at the acceptor end of the molecule—is disordered and
Fig. 5 Packing scheme of 5c, showing the relevant intermolecular contacts. See Table 3 for the relevant parameters and symmetry codes.
c
654 New J. Chem., 2011, 35, 649–662
This journal is The Royal Society of Chemistry and the Centre National de la Recherche Scientifique 2011

View Article Online
Fig. 6 CHꢁ ꢁ ꢁO contacts involving the methoxy groups in 5e. Only the major conformer is shown. See Table 3 for the relevant parameters and
symmetry codes.
Fig. 7 Ribbons of head-to-tail arranged molecules of 5e connected by CHꢁ ꢁ ꢁO interactions. The ribbons cross each other at an angle of
approximately 501. Only the major conformer is shown. See Table 3 for the relevant parameters and symmetry codes.
central rings B, in the contacts C13–H13ꢁ ꢁ ꢁO24 (the corres-
ponding contact involving the minor conformer has been given
in Table S1, ESIw) and C16–H16ꢁ ꢁ ꢁO24 (Fig. 6). The overall
packing of 5e consists of ribbons of head-to-tail arranged
molecules which cross each other at an angle of approximately
501 (Fig. 7). In between these ribbons one CHꢁꢁꢁO intermolecular
contact between the nitro group and a hydrogen atom of the
central ring B can be observed: C15–H15ꢁ ꢁ ꢁO2 (Fig. 7, the
corresponding contact involving the minor conformer has
been given in Table S1, ESIw). A final interaction (not given
in Fig. 6 and 7) comprises C24–H24Bꢁ ꢁ ꢁO26.
[C960–H61Bꢁ ꢁ ꢁO22], as shown in Fig. 8. Furthermore,
molecule 1 forms an infinite linear chain through the structure,
in which the successive molecules are linked by CHꢁ ꢁ ꢁN
contacts [C23–H23ꢁ ꢁ ꢁN1] (not given in Fig. 8), whereas
molecule 2 forms anti-parallel couples due to two additional
CHꢁ ꢁ ꢁO [C95–H95ꢁ ꢁ ꢁO26] and CHꢁ ꢁ ꢁp [C960–H61Cꢁ ꢁ ꢁCg(D)]
contacts (not given in Fig. 8). Four additional weaker inter-
actions (three CHꢁ ꢁ ꢁp involving hydrogen atoms on the rings
and one pꢁ ꢁ ꢁp) have been compiled in Table S1 (ESIw).
2.3.5 E,E-1-[2-(4-Nitrophenyl)ethenyl]-4-[2-(2,4,6-trimethoxy-
phenyl)ethenyl]benzene (5a). Before comparing the solid-state
structures of the five available compounds and six available
structures, we will provide a brief description of the crystal
structure of 5a, adapted from ref. 10. Again, the ethenylic
spacers are oriented syn, as can be seen in Fig. 9. The
molecular structure shows a slight deviation from planarity,
with angles of 8.28(11)1 and 11.03(9)1 between rings A and B
and between rings B and C, respectively. The overall
torsion between rings A and C is 17.92(10)1. There are
two intramolecular CHꢁ ꢁ ꢁO contacts: C17–H17ꢁ ꢁ ꢁO22 and
C18–H18ꢁ ꢁ ꢁO26 (Table 3).
2.3.4 E,E-1-[2-(4-Cyanophenyl)ethenyl]-4-[2-(2,4,6-trimethoxy-
phenyl)ethenyl]benzene (5i). Compound 5i crystallizes in the
ꢀ
centrosymmetric space group P1, with two molecules in the
asymmetric unit (Table 2). Both molecules display the syn
conformation around the central ring (Table 4), and are
slightly bent along their DA axis (Fig. 8). Molecule 1 displays
disorder in the ethenylic link at the acceptor end of the
molecule, of the same type as in 5b-1 and 5e, leading to about
15% of anti conformer. The cyano-substituted rings of both
molecules are rotated out of the plane of the central ring by no
more than 151. The two independent molecules of 5i also
display four intramolecular CHꢁꢁꢁO contacts: C17–H17ꢁꢁꢁO26,
C18–H18ꢁꢁꢁO22, C917–H917ꢁꢁꢁO922 and C918–H918ꢁꢁꢁO926
(Table 3).
The molecule crystallizes in the centrosymmetric space
group P21/c, yielding a structure in which the dipoles are lined
up between co-facially stacked pairs of molecules related
through translation/inversion symmetry. These pairs of
molecules form extended double ribbons along the [ꢀ1 0 1]
direction. The symmetry equivalent molecules through the
glide planes form their own ribbons in the same [ꢀ1 0 1]
The four molecules in the unit cell are arranged with their
methoxy-substituted moieties grouped together, connected
by CHꢁ ꢁ ꢁO contacts between molecule 1 and molecule 2
c
This journal is The Royal Society of Chemistry and the Centre National de la Recherche Scientifique 2011
New J. Chem., 2011, 35, 649–662 655

View Article Online
Fig. 8 View along the b axis of the four molecules in the unit cell of 5i, showing the relevant CHꢁ ꢁ ꢁO contacts. Only the major conformer of
molecule 1 is shown. See Table 3 for the relevant parameters and symmetry codes.
Fig. 9 View of the packing of 5a, showing the head-to-tail CHꢁ ꢁ ꢁO contacts and the CHꢁ ꢁ ꢁp contacts responsible for the herringbone structure.
See Table 3 for the relevant parameters and symmetry codes.
direction. The mutually perpendicular ribbons interact
through typical herringbone CHꢁ ꢁ ꢁp contacts originating from
the nitro-substituted ring A to rings B and C (Fig. 9):
C2–H2ꢁ ꢁ ꢁCg(C) and C6–H6ꢁ ꢁ ꢁCg(B). Between these same
molecules, H3 is involved in the contact C3–H3ꢁ ꢁ ꢁO24 (not
given in Fig. 9). Three other short contacts are weak hydrogen
bonds involving the oxygen atoms of the nitro groups:
C22–H22Aꢁ ꢁ ꢁO2 (not given in Fig. 9), C24–H24Aꢁ ꢁ ꢁO2 and
C25–H25ꢁ ꢁ ꢁO1. The methoxy group in the 2-position also
contacts the one in the 6-position of a neighbouring molecule
(C22–H22Cꢁ ꢁ ꢁO26, not given in Fig. 9). A single N–Oꢁ ꢁ ꢁp is
given in Table S1 (ESIw).
infinite strings of likewise oriented molecules. These strings are
then stacked in an anti-parallel fashion with respect to each
other, with the donor and acceptor moieties placed on top of
each other, a packing arrangement which is also present in
REFKUI.11 This logically results in centrosymmetric crystal
structures, which are indeed found for 5a,10 5b-2 and 5i.
Further stabilization of the supramolecular structures is then
obtained from intermolecular contacts (Table 3) involving
primarily the methoxy groups which create a network of
CHꢁ ꢁ ꢁO contacts. Other types of short contacts such as
CHꢁ ꢁ ꢁp, CHꢁ ꢁ ꢁN, NOꢁ ꢁ ꢁp and pꢁ ꢁ ꢁp are observed but more
limited in number. However, three of the crystal structures are
not centrosymmetric: 5b-1 and 5c crystallize in the non-
centrosymmetric, polar space group P21 and 5e in the non-
centrosymmetric, polar space group Pc. Apparently, in these
cases, the dominating effects of the molecular dipoles are
overcome and non-centrosymmetric space groups are obtained.
Strings of dipoles organized in a head-to-tail fashion are still
seen, but the combined effects of the intermolecular CHꢁ ꢁ ꢁO
and CHꢁ ꢁ ꢁp interactions ultimately outweigh those of the
dipoles in the other two directions.
2.3.6 The supramolecular structures of asymmetric
distyrylbenzenes. Three of the six compounds and/or
polymorphs under investigation crystallize in the syn
conformation, i.e., 5a,10 5b-1 and 5i, and this is quite rare
for E,E-distyrylbenzenes: a CSD search (version 5.30, September
2009 update)20 reveals that only five compounds (CSD
refcodes20 GEJTUL,28 GEJVAT,28 PEBRAP,3 QERWUF29
and WEFBOZ)14 show this particular conformation; these are
all symmetrically substituted DSBs.
Complementing these intermolecular CHꢁ ꢁ ꢁO contacts,
intramolecular CHꢁꢁꢁO contacts are found when ortho-methoxy
substitution with respect to the ethenylic link is present,9 as
in the structures of 5a,10 5e and 5i. These contacts are
It is clear that the packing of these asymmetric compounds
is determined in the first instance by the large molecular
dipoles. These are organized in a head-to-tail fashion, creating
c
656 New J. Chem., 2011, 35, 649–662
This journal is The Royal Society of Chemistry and the Centre National de la Recherche Scientifique 2011

View Article Online
undoubtedly the reason why disorder of the ethenylic spacer is
not observed at the donor side of the molecules: when
ortho-methoxy substitution is present the energy barrier for
rotation of the double bond is greatly increased.7 Note that
5b-1 shows disorder of the ethenylic spacer at the donor side of
the molecule but lacks ortho-methoxy substitution, and that
5e and 5i show disorder of the ethenylic spacer at the
acceptor side.
A striking feature is the similarity of the packings of 5b-1
and 5c. Adding a methoxy group to the 3,5-disubstituted
moiety when going from 5c to 5b-1 apparently does not force
the structure to pack in a drastically different way, although
the methoxy group in the 4-position is involved in two CHꢁ ꢁ ꢁO
contacts. Furthermore, given the fact that the structures are
very similar, it seems odd that the major conformer of 5b-1 is
syn as opposed to 5c which has an anti conformation. This can
be rationalized when considering that 5b-1 in the syn
conformation generates an extra CHꢁ ꢁ ꢁO contact which is
absent in its anti conformation.
Keeping the possible applications of these compounds in
mind and considering that they all require solid states with a
macroscopic dipole moment, centrosymmetric structures are
a priori useless, however excellent the molecular properties of a
particular compound may be. Therefore, it is clear that when
multiple polymorphs with different space groups exist for a
certain material, as for 5b, one should carefully assess which
polymorph is formed when constructing devices based on
these compounds.
Fig. 10 Molecular dipoles of the individual molecules in the unit cell
(green) and net dipole of the unit cell (brown) for the polar crystal
structures of (a) 5b-1, (b) 5c and (c) 5e.
2.3.7 Dipole moments of the unit cell. Even though a polar
space group is the first prerequisite for the application of the
materials in the devices mentioned in the Introduction, it is
ultimately the orientation of the molecular dipoles in the unit
cell with respect to the polar axis of the crystal that determines
the relevant properties of a candidate material. Zyss and
Oudar theoretically derived the values of the optimal angles
for maximum bulk quadratic (second-order) NLO behaviour
between the dipolar axes of the molecules in the unit cell and
the polar axis of the crystal, and found that for second-
harmonic generation (SHG) these angles are 54.741 for the
P21 space group (to which 5b-1 and 5c belong) and 35.261 for
Pc (to which 5e belongs).30 The graphical representations of
the molecular dipoles as well as the unit cell dipoles, constituting
the polar axes, of the crystal structures of 5b-1, 5c and 5e are
given in Fig. 10. Consistent with their space group symmetries,
the unit cell dipoles are along the b axis for 5b-1 and 5c in P21
and in the ac plane (the glide plane of the space group given in
blue in Fig. 10c), making angles of about 311 and 591 with the
ꢀa and c axes, respectively, for 5e in Pc.
anti-parallel fashion leading to a unit cell dipole of a mere
0.05 D—due to this limited size, it cannot actually be seen in
Fig. 10b. Consequently, both molecular dipoles are virtually
perpendicular to the unit cell dipole. This means that, even
though 5c crystallises in the non-centrosymmetric, polar space
group P21 and must therefore have a non-zero macroscopic
dipole, the latter’s extremely limited size may preclude any
useful properties, as the system must behave almost as a
centrosymmetric one.
To put the properties of these new compounds into
perspective, the size and orientation of the molecular and unit
cell dipoles of 2-methyl-4-nitroaniline (MNA, CSD refcode
BAJCIY02), a well-known SHG material,31 were analyzed in
the same way. Each of the four molecules in the unit cell32 has
a calculated molecular dipole of 7.93 D, leading to a sizeable
unit cell dipole of 30.42 D. Consistent with its Cc space group,
the unit cell dipole is found in the ac plane, making angles of
about 191 and 1091 with the ꢀa and c axes, respectively. Each
of the four molecular dipoles makes an angle of about 161 with
the unit cell dipole. Comparison of the latter value with the
theoretically predicted one for that space group (35.261)
suggests a more efficient contribution of the molecular hyper-
polarizability to the macroscopic one for MNA than for 5e,
for which the difference with the theoretical value is clearly
larger. For a proper comparison of the NLO properties of
these materials, though, the molecular hyperpolarizabilities of
For 5b-1 (Fig. 10a), the net unit cell dipole amounts to
2.27 D and makes an angle of 791 with the two molecular
dipoles. Comparison of the latter value with the theoretical
value of 54.741 mentioned above suggests that 5b-1 will display
bulk SHG to some extent. For 5e (Fig. 10c), the sizeable unit
cell dipole of 7.01 D makes an angle of about 661 with the two
molecular dipoles. Comparison of this value with the angle of
35.261 given above leads to a similar conclusion as for 5b-1.
For 5c, however, the two large molecular dipoles of the
molecules in the unit cell are oriented in an almost completely
c
This journal is The Royal Society of Chemistry and the Centre National de la Recherche Scientifique 2011
New J. Chem., 2011, 35, 649–662 657

View Article Online
the new compounds will have to be measured; these
experiments are underway.
(150 ml). The mixture was refluxed under a nitrogen
atmosphere for 3 h. After cooling to room temperature, water
(150 ml) was added to the reaction mixture and the precipitate
was filtered off. The product was collected and redissolved in
hot acetone (250 ml). This solution was poured into water
(150 ml), after which the compound was collected by filtration.
In order to obtain the pure E isomer, the compound was
refluxed in p-xylene with a catalytic amount of iodine for 4 h.
The yield was 33%. Mp 182 1C. d1H 2.40 (s, 3H, CH3), 7.22
(d, 3J = 8.24, 2H, H12 and H16), 7.27 (d, 3J = 16.17, 1H, H7),
7.47 (d, 3J = 8.24, 2H, H13 and H15), 7.57 (d, 3J = 16.17, 1H,
H8), 7.97 (d, 3J = 8.70, 1H, H6), 8.40 (dd, 3J = 8.70,
3. Conclusions
Twelve asymmetrical donor–acceptor distyrylbenzenes,
containing one or more nitro or cyano groups as electron
acceptors and one or more methoxy groups as electron donors,
have been synthesized. The electronic properties were examined
using CV, UV/Vis and quantum chemical calculations, and the
effects of the substituents on these properties were described;
although the reduction peak was absent from the voltammo-
grams for all compounds, further experiments showed that the
systems can be regarded as reversible and the anodic peak
potentials were used as a measure for the oxidation potential.
The solid-state structures of five of these new compounds were
analyzed and described in function of the substitution pattern
of the DSB scaffold: in general, the oligomers organise
themselves in a head-to-tail fashion and generate CHꢁ ꢁ ꢁO
networks involving mainly the different methoxy groups.
However, the compounds crystallize in both centrosymmetric
and non-centrosymmetric, polar structures: in three of the six
studied crystal structures the combined effects of the relatively
weak intermolecular CHꢁ ꢁ ꢁO and CHꢁ ꢁ ꢁp interactions
outweigh the otherwise dominating effects of the molecular
dipoles and lead to non-centrosymmetric, polar structures.
Analysis of the relative orientation of molecular and unit cell
dipoles in the polar crystal structures leads to first insights into
the NLO properties of these materials. Further experimental
work to determine these properties is underway.
4
4J = 2.29, 1H, H5), 8.79 (d, J = 2.29, 1H, H3). d13C 21.4
(CH3), 120.2 (C3), 120.7 (C7), 127.0 (C6), 127.6 (C13 and
C15), 128.7 (C5), 129.8 (C12 and C16), 132.9 (C8), 138.2
(C14), 139.0 (C11), 140.3 (C1), 146.1 (C2), 147.4 (C4).
4.1.1.3 E-1-(4-Methylphenyl)-2-(4-cyanophenyl)ethene (2d).
2d was prepared as is outlined in ref. 18. d1H 2.37 (s, 3H,
3
CH3), 7.04 (d, J = 16.33, 1H, H7), 7.19 (m, 3H, H8, H12 and
H16), 7.42 (d, 3J = 8.24, 2H, H13 and H15), 7.56 (d, 3J = 8.69,
2H, H2 and H6), 7.62 (d, 3J = 8.54, 2H, H3 and H5). d13C 21.3
(CH3), 110.4 (C4), 119.0 (CN), 125.8 (C7), 126.7 (C3 and C5),
126.9 (C13 and C15), 129.6 (C12 and C16), 132.41 (C8), 132.44
(C2 and C6), 133.6 (C14), 138.8 (C11), 142.1 (C1).
4.1.2 Bromomethylstilbenes (3a–d)
4.1.2.1 E-1-(4-Bromomethylphenyl)-2-(4-nitrophenyl)ethene
(3a). 3a was prepared as is outlined in ref. 10. d1H 4.52 (s, 2H,
3
3
CH2), 7.15 (d, J = 16.33, 1H, H7), 7.25 (d, J = 16.33, 1H,
H8), 7.42 (d, 3J = 8.24, 2H, H13 and H15), 7.52 (d, 3J = 8.08,
2H, H12 and H16), 7.63 (d, 3J = 8.54, 2H, H2 and H6), 8.23 (d,
3J = 8.85, 2H, H3 and H5). d13C 33.0 (CH2), 124.2 (C12 and
C16), 127.9 (C3 and C5), 127.0 (C7), 127.4 (C13 and C15),
129.6 (C2 and C6), 132.5 (C8), 136.4 (C14), 138.4 (C11), 143.57
(C1), 147.0 (C4).
4. Experimental
4.1 Syntheses
All reagents and solvents were purchased from ACROS and
1
used as received. H and 13C NMR spectra were recorded in
4.1.2.2 E-1-(4-Bromomethylphenyl)-2-(3-nitrophenyl)ethene
(3b). N-Bromosuccinimide (NBS) (7.1 g, 0.040 mol) and a
catalytic amount of 1,10-azobis(cyclohexanecarbonitrile) were
added to a heated solution of 2b (7.6 g, 0.032 mol) in CCl4
(150 ml). The mixture was refluxed overnight. The hot solution
was filtered to remove the succinimide. Then, the solvent was
allowed to cool and partially evaporated. The precipitated
product was collected by filtration. The yield was 47%. Mp
CDCl3, unless stated otherwise, at 25 1C using a Bruker
Avance II spectrometer operating at 400 MHz and 100 MHz,
respectively; chemical shifts d are given in ppm relative to
tetramethylsilane (TMS) and coupling constants J are given in
Hz. UV/Vis absorption spectra were recorded for solutions
(about 20 mM) in CH2Cl2 on a Varian Cary 5 spectro-
photometer.
3
4.1.1 Methylstilbenes (2b–d)
107 1C. d1H 4.51 (s, 2H, CH2), 7.13 (d, J = 16.33, 1H, H7),
4.1.1.1 E-1-(4-Methylphenyl)-2-(3-nitrophenyl)ethene (2b). 2b
was prepared as is outlined in ref. 23. d1H 2.40 (s, 3H, CH3), 7.08
(d, J = 16.33, 1H, H7), 7.21 (m, 3H, H8, H12 and H16), 7.44
7.22 (d, 3J = 16.33, 1H, H8), 7.40–7.60 (m, 5H, H5, H12, H13,
H15, H16), 7.79 (dd, 3J = 7.76, 4J = 1.12, 1H, H6), 8.10 (ddd,
3
4
3J = 7.88, 4J = 2.20, 4J = 1.12, 1H, H4), 8.36 (d, J = 2.20,
3
(d, J = 8.20, 2H, H13 and H15), 7.51 (m, 1H, H5), 7.78 (dd,
1H, H2). d13C 33.14 (CH2), 120.97 (C4), 122.22 (C2), 126.88
(C7), 127.22 (C12 and C16), 129.91 (C13 and C15), 129.68
(C14), 130.99 (C6), 132.31 (C8), 136.51 (C11), 138.04 (C1),
138.96 (C11), 148.81 (C3).
3J = 7.75, 4J = 1.01, 1H, H6), 8.08 (ddd, 3J = 8.17, 4J = 2.26,
4J = 1.01, 1H, H4), 8.35 (dd, 4J = 2.26, 4J = 1.01, 1H, H2). d13C
21.3 (CH3), 120.8 (C4), 121.8 (C2), 125.1 (C7), 126.8 (C13 and
C15), 129.5 (C5), 129.6 (C12 and C16), 131.7 (C14), 132.1 (C6),
133.5 (C8), 138.6 (C1), 139.4 (C11), 148.8 (C3).
4.1.2.3 E-1-(4-Bromomethylphenyl)-2-(2,4-dinitrophenyl)-
ethene (3c). 3c was prepared as is outlined in ref. 24. d1H 4.52
(s, 2H, CH2), 7.26 (d, 3J = 16.17, 1H, H7), 7.45 (d, 3J = 8.24,
2H, H13 and H15), 7.55 (d, 3J = 8.24, 2H, H12 and H16), 7.66
(d, 3J = 16.18, 2H, H8), 7.98 (d, 3J = 8.70, 1H, H6), 8.43 (dd,
4.1.1.2 E-1-(4-Methylphenyl)-2-(2,4-dinitrophenyl)ethene
(2c). A solution of sodium (2.3 g, 0.1 mol) in dry ethanol
(80 ml) was added dropwise to a stirred mixture of 1 (40.3 g,
0.1 mol) and 2,4-dinitrobenzaldehyde (0.1 mol) in dry ethanol
4
4
3J = 8.70, J = 2.29, 1H, H5), 8.83 (d, J = 2.29, 1H, H3).
c
658 New J. Chem., 2011, 35, 649–662
This journal is The Royal Society of Chemistry and the Centre National de la Recherche Scientifique 2011

View Article Online
3
(m, 4H, H12, H13, H15, H16), 7.62 (d, J = 8.86, 2H, H2 and
Due to the limited solubility of 3c no 13C NMR spectrum of
3
sufficient quality could be obtained.
H6), 8.21 (d, J = 8.86, 2H, H3 and H5). d13C 55.4 (C30 and
C50), 100.3 (C24), 104.8 (C22 and C26), 124.2 (C3 and C5), 126.2
(C7), 126.8 (C2 and C6), 127.1 (C13 and C15), 127.4 (C12 and
C16), 128.5 (C17), 129.5 (C8), 132.9 (C18), 135.6 (C14), 137.8
(C21), 139.2 (C11), 143.9 (C1), 146.8 (C4), 161.1 (C23 and C25).
4.1.2.4 E-1-(4-Bromomethylphenyl)-2-(4-cyanophenyl)ethene
(3d). 3d was prepared as is outlined in ref. 18. d1H 4.51 (s, 2H,
3
3
CH2), 7.09 (d, J = 16.33, 1H, H7), 7.19 (d, J = 16.33, 1H,
H8), 7.41 (d, 3J = 8.24, 2H, H13 and H15), 7.50 (d, 3J = 8.24,
3
2H, H12 and H16), 7.58 (d, J = 8.39, 2H, H3 and H5), 7.63
4.1.4.3 E,E-1-[2-(4-Nitrophenyl)ethenyl]-4-[2-(2,6-dimethoxy-
phenyl)ethenyl]benzene (5d). Yield 59%. Mp 151 1C. d1H
(MeOD) 3.89 (s, 6H, H20 and H60), 6.58 (d, J = 8.32, 2H,
3
(d, J = 8.55, 2H, H2 and H6). d13C 33.1 (CH2), 110.9 (C4),
3
118.9 (CN), 126.9 (C2 and C6), 127.3 (C12 and C16), 127.5 (C7),
129.6 (C13 and C15), 131.6 (C8), 132.5 (C3 and C5), 136.6
(C14), 138.1 (C11), 141.6 (C1).
3
H23 and H25), 7.08 (d, J = 16.33, 1H, H7), 7.14–7.23 (m, 2H,
H8 and H24), 7.47–7.58 (m, 8H, H2, H6, H12, H13, H15, H16,
H17 and H18), 8.17 (d, 3J = 8.31, 2H, H3 and H5). d13C (MeOD)
55.83 (C20 and C60), 104.07 (C23 and C25), 114.68 (C21), 120.82
(C17), 124.11 (C3 and C5), 125.47 (C7), 126.72 (C2 and C5),
126.86 (C13 and C15), 127.29 (C12 and C16), 128.44 (C8), 131.55
(C24), 133.20 (C18), 134.76 (C14), 140.15 (C11), 144.09 (C1),
146.59 (C4), 158.78 (C22 and C26).
4.1.3 Phosphonium salts (4b–c)
4.1.3.1 E-Triphenyl[4-(2-(3-nitrophenyl)ethenyl)benzyl]-
phosphonium bromide (4b). Triphenylphosphine (6.6 g, 0.025 mol)
was added to a solution of 3b (7.3 g, 0.023 mol) in acetonitrile
(150 ml). The solution was refluxed for 4 h. After cooling to room
temperature, the product was collected by filtration and washed
with diethyl ether. The yield was 76%. Mp > 250 1C (decomp.).
4.1.4.4 E,E-1-[2-(4-Nitrophenyl)ethenyl]-4-[2-(2,4-dimethoxy-
phenyl)ethenyl]benzene (5e). 5e was prepared as is outlined in
ref. 24. Yield 44%. Mp 172 1C. d1H 3.84 (s, 3H, H40), 3.89 (s, 3H,
H20), 6.48 (d, 4J = 2.40, 1H, H23), 6.53 (dd, 3J = 8.44,
4.1.3.2 E-Triphenyl[4-(2-(2,4-dinitrophenyl)ethenyl)benzyl]-
phosphonium bromide (4c). Triphenylphosphine (6.6 g, 0.025 mol)
was added to a solution of 3c (8.3 g, 0.023 mol) in acetonitrile
(150 ml). The solution was refluxed for 4 h. After cooling to room
temperature, the product was collected by filtration and washed
with diethyl ether. The yield was 39%. Mp 255 1C (decomp.).
3
4J = 2.40, 1H, H25), 7.02 (d, J = 16.43, 1H, H7), 7.13 (d,
3J = 16.43, 1H, H8), 7.26 (d, 3J = 8.44, 1H, H26), 7.26 (d,
3
3J = 16.31, 1H, H17), 7.45 (d, J = 16.47, 1H, H18), 7.51 (d,
3J = 9.02, 2H, H13 and H15), 7.54 (d, 3J = 9.06, 2H, H12 and
H16), 7.63 (d, 3J = 8.88, 2H, H2 and H6), 8.22 (d, 3J = 8.88, 2H,
H3 and H5). d13C 55.45 (C40), 55.57 (C20), 98.57 (C23), 105.19
(C25), 119.36 (C21), 124.17 (C3 and C5), 125.58 (C7), 126.24
(C26), 126.76 (C2 and C6), 127.30 (C17), 127.37 (C13 and C15),
127.38 (C12 and C16), 130.26 (C8), 133.14 (C18), 137.77 (C14),
139.13 (C11), 144.08 (C1), 146.66 (C4), 158.23 (C22),
160.84 (C24).
4.1.4 Oligomers 5b–k. In the general procedure, sodium
(0.4 g, 16.5 mmol) in dry ethanol (40 ml) was added dropwise
to a stirred mixture of the appropriate phosphonium salt 4a–d
(16.5 mmol) and the appropriate methoxybenzaldehyde
(16.5 mmol) in dry ethanol (100 ml). The mixture was refluxed
under a nitrogen atmosphere for 6 h. After cooling to room
temperature, water (125 ml) was added. The mixture was
filtered and the product washed with diethyl ether. To obtain
the pure E,E isomer, the product was refluxed in p-xylene with
a catalytic amount of iodine for 4 h. After cooling, the product
was collected by filtration. NMR spectroscopic data of 5a can
be found in ref. 10.
4.1.4.5 E,E-1-[2-(4-Nitrophenyl)ethenyl]-4-[2-(4-methoxy-
phenyl)ethenyl]benzene (5f). 5f was prepared as is outlined in
ref. 24. Yield 60%. Mp 251 1C. d1H 3.84 (s, 3H, H40), 6.91 (d,
3
3J = 8.80, 2H, H23 and H25), 6.98 (d, J = 16.31, 1H, H7),
3
3
7.12 (d, J = 16.31, 1H, H8), 7.14 (d, J = 16.27, 1H, H17),
7.26 (d, 3J = 16.27, 1H, H18), 7.47 (d, 3J = 8.56, 2H, H22 and
4.1.4.1 E,E-1-[2-(4-Nitrophenyl)ethenyl]-4-[2-(3,4,5-tri-
methoxyphenyl)ethenyl]benzene (5b). Yield 14%. Mp 180 1C
[for the red polymorph (5b-1); the yellow polymorph (5b-2)
transforms into the red upon heating]. d1H 3.88 (s, 3H, H40),
3.93 (s, 6H, H30 and H50), 6.76 (s, 2H, C22 and C26), 7.02 (d,
3J = 16.22, 1H, H18), 7.10 (d, 3J = 16.22, 1H, H17), 7.16 (d,
3
3
H26), 7.51 (d, J = 8.93, 2H, H12 and H16), 7.54 (d, J =
8.93, 2H, H13 and H15), 7.63 (d, 3J = 8.64, 2H, H2 and H6),
8.22 (d, 3J = 8.83, 2H, H3 and H5). d13C 55.37 (C40), 114.27
(C23 and C25), 124.20 (C3 and C5), 125.84 (C7), 125.91 (C17),
126.74 (C12 and C16), 126.80 (C2 and C6), 127.44 (C22 and
C26), 127.88 (C13 and C15), 129.98 (C21), 129.99 (C8), 133.02
(C18), 135.08 (C11), 138.42 (C14), 144.00 (C1), 146.76 (C4),
159.61 (C24).
3
3J = 16.31, 1H, H7), 7.27 (d, J = 16.31, 1H, H8), 7.51–7.58
(m, 4H, H12, H13, H15 and H16), 7.64 (d, 3J = 8.59, 2H, H2
and H6), 8.22 (d, 3J = 8.59, 2H, H3 and H5). d13C 56.21 (C30
and C50), 60.99 (C40), 103.87 (C22 and C26), 124.19 (C3 and
C5), 126.10 (C7), 126.83 (C2 and C6), 126.92 (C13 and C15),
127.47 (C12 and C16), 129.43 (C8), 132.86 (C18), 135.47
(C14), 137.9 (C21), 138.40 (C24), 143.90 (C1), 146.79 (C4),
153.52 (C23 and C25).
4.1.4.6 E,E-1-[2-(3-Nitrophenyl)ethenyl]-4-[2-(2,4,6-tri-
methoxyphenyl)ethenyl]benzene (5g). Yield 42%. Mp 176 1C.
d1H 3.85 (s, 3H, H40), 3.90 (s, 6H, H20 and H60), 6.18 (s, 2H,
3
3
H23 and H25), 7.11 (d, J = 16.30, 1H, H7), 7.22 (d, J =
16.30, 1H, H8), 7.45–7.60 (m, 7H, H3, H12, H13, H15, H16,
3
H17 and H18), 7.79 (dd, J = 7.70, J = 1.0, 1H, H6), 8.07
4
4.1.4.2 E,E-1-[2-(4-Nitrophenyl)ethenyl]-4-[2-(3,5-dimethoxy-
phenyl)ethenyl]benzene (5c). Yield 85%. Mp 172 1C. d1H 3.84
(s, 6H, H30 and H50), 6.42 (t, 4J = 2.25, 1H, H24), 6.68 (d, 4J =
2.25, 2H, H22 and H26), 7.08 (m, 2H, H17 and H18), 7.15
(d, 3J = 16.31, 1H, H7), 7.25 (d, 3J = 16.30, 1H, H8), 7.48–7.58
3
4
4
4
(ddd, J = 8.20, J = 2.20, J = 1.0, 1H, H4), 8.36 (d, J =
2.20, 1H, H2). d13C 55.3 (C40), 55.8 (C20 and C60), 90.9 (C23
and C25), 108.2 (C21), 120.6 (C4), 120.8 (C17), 121.8 (C2),
125.1 (C7), 126.6 (C13 and C15), 127.1 (C12 and C16), 129.2
c
This journal is The Royal Society of Chemistry and the Centre National de la Recherche Scientifique 2011
New J. Chem., 2011, 35, 649–662 659

View Article Online
and 129.5 (C5 and C8), 131.8 and 132.1 (C6 and C18), 134.4
(C14), 139.5 (C1), 140.3 (C11), 148.8 (C3), 159.7 (C23 and
C25), 160.5 (C24).
overnight. Subsequently, water (300 ml) was added, after
which the main by-product (i.e., the symmetric distyrylbenzene)
was filtered off. The solvent was evaporated and the residue
was redissolved in DMF. The remaining precipitate (i.e., the
bisphosphonium starting product) was filtered off and the
solution was placed in the refrigerator overnight. The precipitate
was again filtered off and the solvent was evaporated. Toluene
(200 ml) was added to the residue and the resulting precipitate
was the desired compound, which was washed with ether and
dried in air. The yield was 35%. Mp > 250 1C.
4.1.4.7 E,E-1-[2-(2,4-Dinitrophenyl)ethenyl]-4-[2-(2,4,6-
trimethoxyphenyl)ethenyl]benzene (5h). 5h was prepared as is
outlined in ref. 24. Yield 33%. Mp 210 1C. d1H 3.85 (s, 3H,
H40), 3.90 (s, 6H, H20 and H60), 6.18 (s, 2H, H23 and H25),
3
7.29 (d, J = 16.07, 1H, H8), 7.51 (m, 6H, H12, H13, H15,
3
H16, H17 and H18), 7.62 (d, J = 16.07, 1H, H7), 7.98 (d,
3
3J = 8.80, 1H, H6), 8.40 (dd, J = 8.80, 4J = 2.38, 1H, H5),
8.80 (d, 4J = 2.34, 1H, H3). d13C 55.37 (C40), 55.83 (C20 and
C60), 90.89 (C23 and C25), 107.98 (C21), 119.82 (C3), 120.82
(C17), 126.7 (C7), 126.92 (C5), 127.98 (C13 and C15), 128.79
(C6), 128.92 (C12 and C16), 133.60 (C8), 134.69 (C18), 138.3
(C14), 139.05 (C11), 141.79 (C1), 145.92 (C4), 147.29 (C2),
159.80 (C22 and C26), 160.75 (C24).
4.1.5.2 E,E-1-[2-(2,4,6-Trinitrophenyl)ethenyl]-4-[2-(2,4,6-
trimethoxyphenyl)ethenyl]benzene (5l). Sodium (0.08 g,
3.3 mmol) in dry ethanol (30 ml) was added dropwise to a
stirred mixture of 7 (0.75 g, 3.3 mmol) and 8 (2.1 g, 3.3 mol) in
dry ethanol (60 ml). The mixture was refluxed under nitrogen
atmosphere for 6 h. After cooling to room temperature, the
mixture was filtered and the dark red product washed with
diethyl ether. To obtain the pure E isomer, the product was
refluxed in p-xylene with a catalytic amount of iodine for 4 h.
After cooling, the product was collected by filtration. Yield
25%. Mp 200 1C. d1H (MeOD) 3.81 (s, 3H, H40), 3.85 (s, 6H,
H20 and H60), 6.22 (s, 2H, H23 and H25), 7.50–7.90 (m, 8H,
H7, H8, H12, H13, H15, H16, H17, H18), 8.54 (s, 2H, H5 and
H3). Due to the limited solubility of 5l no 13C NMR spectrum
of sufficient quality could be obtained.
4.1.4.8 E,E-1-[2-(4-Cyanophenyl)ethenyl]-4-[2-(2,4,6-tri-
methoxyphenyl)ethenyl]benzene (5i). Yield 40%. Mp 196 1C.
d1H 3.85 (s, 3H, H40), 3.90 (s, 6H, H20 and H60), 6.18 (s, 2H,
3
3
H23 and H25), 7.06 (d, J = 16.33, 1H, H7), 7.20 (d, J =
16.33, 1H, H8), 7.46 (m, 2H, H17 and H18), 7.48 (d, 3J = 8.54,
2H, H13 and H15), 7.52 (d, 3J = 8.55, 2H, H12 and H16), 7.56
(d, 3J = 8.39, 2H, H2 and H6), 7.62 (d, 3J = 8.39, 2H, H3 and
H5). d13C 55.4 (C40), 55.8 (C20 and C60), 91.0 (C23 and C25),
108.2 (C21), 110.3 (C4), 119.1 (CN), 120.7 (C17), 125.7 (C7),
126.6 (C12 and C16), 126.7 (C2 and C6), 127.1 (C13 and C15),
129.2 (C18), 132.4 (C3 and C5), 134.4 (C14), 140.4 (C11),
142.2 (C1), 159.7 (C22 and C26), 160.5 (C24).
4.2 X-Ray diffraction
For all compounds, attempts were made to grow suitable
crystals via diffusion, slow evaporation and slow cooling, using
various solvents. Both polymorphs of 5b and 5i were obtained
from a CH2Cl2 solution and 5c crystallized by slow evaporation of
a THF solution. The collection of diffraction data of sufficient
quality was performed on an Enraf-Nonius Mach 3 diffracto-
meter with single-point detector using Mo-Ka radiation
(l = 0.71073 A) for compounds 5b (two polymorphs, 5b-1
and 5b-2), 5c and 5i, although the crystals of 5b-2 and 5i were
relatively weakly diffracting and produced data of somewhat
lower quality. Details on the crystallization and data collection
of 5a10 and 5e27 can be found elsewhere. For the non-centro-
symmetric structures 5b-1 and 5c, the Friedel pairs were
collected independently, but were merged in the final refinement;
due to the fact that there are no anomalous scatterers (atoms
with Z > ZSi) an absolute structure could not be determined
and it was chosen arbitrarily.
4.1.4.9 E,E-1-[2-(4-Cyanophenyl)ethenyl]-4-[2-(2,4-di-
methoxyphenyl)ethenyl]benzene (5j). Yield 38%. Mp 162 1C.
d1H 3.84 (s, 3H, H40), 3.88 (s, 3H, H20), 6.48 (s, 1H, H23),
3
3
6.51 (d, J = 8.24, 1H, H25), 7.01 (d, J = 16.33, 1H, H7),
3
3
7.19 (d, J = 16.32, 1H, H8), 7.38 (d, J = 16.32, 1H, H17),
7.41 (d, 3J = 16.32, 1H, H18), 7.49 (d, 3J = 8.49, 2H, H13 and
3
H15), 7.52 (d, J = 8.64, 2H, H12 and H16), 7.53 (d, J =
3
3
8.59, 1H, H26), 7.56 (d, J = 8.35, 2H, H2 and H6). Due to
the limited solubility of 5j no 13C NMR spectrum of sufficient
quality could be obtained.
4.1.4.10 E,E-1-[2-(4-Cyanophenyl)ethenyl]-4-[2-(4-methoxy-
phenyl)ethenyl]benzene (5k). Yield 39%. Mp 245 1C. d1H
3.84 (s, 3H, H40), 6.55 (d, 3J = 8.60, 2H, H23 and H25), 6.97
(d, 3J = 16.19, 1H, H7), 7.08 (d, 3J = 16.19, 1H, H8), 7.11 (d,
3J = 16.31, 1H, H17), 7.21 (d, 3J = 16.31, 1H, H18), 7.49 (d,
3J = 8.60, 2H, H22 and H26), 7.51 (m, 4H, H12, H13, H15
The structures of 5b-1, 5e and 5i showed disorder consisting
of a 1801 twist of one of the ethenylic links. The model used in
the variable-temperature study of the disorder in 5e is reported
elsewhere:27 in it the disorder has been described taking into
account the complete stilbene fragment, i.e., the two rings and
the spacer. For compound 5b-1, however, such a complete
model could not be used, in part due to the limited size of the
crystal. Incorporation of the complete peripheral ring (with or
without all of the methoxy groups) into the model led to
convergence problems (even with 500 LS-cycles) due to the
large number of restraints, or meaningless geometrical and
anisotropic displacement parameters, as did the inclusion of
only parts of the rings. Ultimately, assuming a dynamic pedal
motion (see ref. 27 and references therein), the ethenylic spacer
3
3
and H16), 7.58 (d, J = 8.24, 2H, H2 and H6), 7.63 (d, J =
8.24, 2H, H3 and H5). Due to the limited solubility of 5k no
13C NMR spectrum of sufficient quality could be obtained.
4.1.5 Oligomer 5l
4.1.5.1 E-Triphenyl[4-(2-(2,4,6-trimethoxyphenyl)ethenyl)-
benzyl]phosphonium bromide (8). A solution of sodium
ethanolate in dry ethanol, prepared by reacting sodium
(0.6 g, 0.025 mol) with dry ethanol (50 ml), was added
dropwise to a mixture of 1,4-bis(triphenylphosphonium-
methyl)-benzene dibromide (39.4 g, 0.050 mol) and of 2,4,6-
trimethoxybenzaldehyde (6.0 g, 0.025 mol) in dry ethanol
(200 ml) under a nitrogen atmosphere. The solution was stirred
c
660 New J. Chem., 2011, 35, 649–662
This journal is The Royal Society of Chemistry and the Centre National de la Recherche Scientifique 2011

View Article Online
was treated as a rigid group. The C17–C18 and C17B–C18B
distances were restrained to be similar (SADI) and the
C14–C17 and C21–C18 distances were restrained to 1.45 A
(DFIX). The atoms of the pairs C12/C13, C15/C16, C17/C17B
and C18/C18B were constrained to possess identical ADPs
(EADP). The atoms of the pairs C25/C26 and C22/C23 were
restrained to possess similar ADPs (SIMU). For compound 5i,
a similar model was used. The C7–C8 and C7B–C8B, C21–C8
and C21–C8B, and C4–C7 and C4–C7B distances were
restrained to be similar (SADI). The atoms of the pairs
C7/C7B and C8/C8B were constrained to posses identical
ADPs (EADP). The generation of ‘‘anti-bumping’’ restraints
also proved to be necessary (BUMP).
low to allow any measurement. Extra dry acetonitrile
(o50 ppm water) was purchased from ACROS and stored
on 3 A sieves. Standard voltammograms were recorded at a
scan rate of 0.1 V sꢀ1, while measurements to ascertain the
reversibility of the systems were performed at scan rates of
0.03, 0.05, 0.1, 0.25 and 1 V sꢀ1; the ferrocene/ferrocenium
(Fc/Fc+) couple was used as a reference.43 The working
electrode was a BASi stationary voltammetry electrode
comprising a platinum electrode disc with a diameter of
3 mm. A platinum plate was used as the auxiliary electrode
and a silver wire as the pseudo-reference electrode.
Acknowledgements
Hydrogen atoms were placed in calculated positions and
refined as riding with C–H distances of 0.93 A. For the
analysis of the supramolecular structures, the C–H bond
lengths were normalized to the value derived from neutron
diffraction (1.083 A).33 The CAD-4 EXPRESS34 software was
used for data collection and cell refinement and XCAD-435 for
data reduction. The structures were solved by direct methods
using SHELXS-9736 and refined using SHELXL-97.36 WinGX
(version 1.80)37 and PLATON38 were used to prepare the
material for publication and figures were drawn using
ORTEP339 for Windows and MERCURY (version 2.2).40
The experimental details including the results of the refinements
are given in Table 2. The numbering scheme can be found in
Fig. 1. CCDC 760975 (5b-1), 760976 (5b-2), 760977 (5c) and
760979 (5i) contain the supplementary crystallographic data
for this paper.
A. C. and R. D. B. wish to thank the Institute for the
Promotion of Innovation through Science and Technology
in Flanders (IWT-Vlaanderen) for a research grant. C. V. V.
wishes to thank the Fund for Scientific Research (FWO
Vlaanderen) for a grant as a research assistant. Financial
support by FWO-Vlaanderen under grant G.0129.05 and by
the University of Antwerp under Grant GOA-2404 is gratefully
acknowledged. The authors gratefully acknowledge the
University of Antwerp for access to the university’s computer
cluster CalcUA.
References
1 K. Mullen and G. Wegner, Electronic Materials—The Oligomer
¨
Approach, Wiley-VCH, Weinheim, 1998.
2 J. L. Segura and N. Martin, J. Mater. Chem., 2000, 10, 2403–2435.
3 M. Hakansson, S. Jagner, M. Sundahl and O. Wennerstrom, Acta
Chem. Scand., 1992, 46, 1160–1165.
4.3 Quantum chemical calculations
4 H. Irngartinger, J. Lichtenthaler and R. Herpich, Struct. Chem.,
1994, 5, 283–289.
Quantum chemical calculations were performed using the
Gaussian 0341 suite of programs at the DFT/B3LYP/6-31G*
level of theory; the basis set was used as it is implemented in
the program. Optimized geometries were calculated in C1 (for
5b, 5h and 5l) or Cs (for 5a, 5c, 5d, 5e, 5f, 5g, 5i, 5j and 5k)
symmetry. Frequency calculations were performed to verify
that the resulting structures are energy minima. Ionization
Potentials (IP) were calculated using Koopmans’ theorem.42
Molecular dipoles of the two molecules in the unit cells of
5b-1, 5c and 5e were calculated in their solid-state geometries
(for 5b-1 and 5e, only the major conformer was considered) at
the same level of theory; these values are also given in Table 1.
The resulting vectors were then transformed from the Cartesian to
the unit cell coordinate system.
5 J. K. Baeke, J. Nowaczyk, M. Moens, L. J. Chen, P. Dieltiens,
H. J. Geise, C. Van Alsenoy and F. Blockhuys, J. Electroanal.
Chem., 2007, 599, 1–11.
6 C. M. L. Vande Velde, J. K. Baeke, H. J. Geise and F. Blockhuys,
Acta Crystallogr., Sect. C: Cryst. Struct. Commun., 2005, 61,
o284–o287.
7 C. M. L. Vande Velde, L. J. Chen, J. K. Baeke, M. Moens,
P. Dieltiens, H. J. Geise, M. Zeller, A. D. Hunter and
F. Blockhuys, Cryst. Growth Des., 2004, 4, 823–830.
8 C. M. L. Vande Velde, H. J. Geise and F. Blockhuys, Acta
Crystallogr., Sect. C: Cryst. Struct. Commun., 2005, 61, o21–o24.
9 C. M. L. Vande Velde, H. J. Geise and F. Blockhuys, Cryst.
Growth Des., 2006, 6, 241–246.
10 C. M. L. Vande Velde, R. De Borger and F. Blockhuys, Acta
Crystallogr., Sect. E: Struct. Rep. Online, 2005, 61, o4289–o4291.
11 G. P. Bartholomew, X. H. Bu and G. C. Bazan, Chem. Mater.,
2000, 12, 1422–1430.
12 G. P. Bartholomew, X. H. Bu and G. C. Bazan, Chem. Mater.,
2000, 12, 2311–2318.
13 G. W. Coates, A. R. Dunn, L. M. Henling, J. W. Ziller,
E. B. Lobkovsky and R. H. Grubbs, J. Am. Chem. Soc., 1998,
120, 3641–3649.
14 B. Nohra, S. Graule, C. Lescop and R. Reau, J. Am. Chem. Soc.,
2006, 128, 3520–3521.
15 M. L. Renak, G. P. Bartholomew, S. J. Wang, P. J. Ricatto,
R. J. Lachicotte and G. C. Bazan, J. Am. Chem. Soc., 1999, 121,
7787–7799.
4.4 Electrochemical measurements
Electrochemical measurements were performed using an
AUTOLAB potentiostat/galvanostat connected to a PC equipped
with the ‘‘General Purpose Electrochemical System’’ (GPES)
software. Cyclic voltammetric (CV) measurements were
carried out in dry acetonitrile at 298 K under an argon
atmosphere. Tetrabutylammonium tetrafluoroborate obtained
from ALDRICH was used as the supporting electrolyte.
Concentrations of 0.4 mM and 80 mM for the oligomer and
electrolyte, respectively, were used, except for compound 5f,
for which a solution of only 0.2 mM of oligomer was used, due
to its limited solubility; the solubility of 5l proved to be too
16 J. C. Sancho-Garcia, J.-L. Bredas, D. Beljonne, J. Cornil,
´
R. Martinez-Alvarez, M. Hanack, L. Poulsen, J. Gierschner,
H.-G. Mack, H.-J. Egelhaaf and G. Oelkrug, J. Phys. Chem. B,
2005, 109, 4872–4880.
17 M. Zeller, A. D. Hunter, R. Hoefnagels, H. J. Geise and
F. Blockhuys, Acta Crystallogr., Sect. E: Struct. Rep. Online,
2005, 61, o4183–o4184.
c
This journal is The Royal Society of Chemistry and the Centre National de la Recherche Scientifique 2011
New J. Chem., 2011, 35, 649–662 661

View Article Online
18 R. Erckel and H. Fruhbeis, Z. Naturforsch., B: Chem. Sci., 1982,
37, 1472–1480.
35 K. Harms and S. Wocadlo, XCAD4, University of Marburg,
Germany, 1996.
19 J. K. Baeke, R. De Borger, F. Lemiere, C. Van Alsenoy and
F. Blockhuys, J. Phys. Org. Chem., 2009, 22, 925–932.
20 F. H. Allen, Acta Crystallogr., Sect. B: Struct. Sci., 2002, 58,
380–388.
36 G. M. Sheldrick, Acta Crystallogr., Sect. A: Found. Crystallogr.,
2008, 64, 112–122.
37 L. J. Farrugia, J. Appl. Crystallogr., 1999, 32, 837–838.
38 A. L. Spek, J. Appl. Crystallogr., 2003, 36, 7–13.
21 D. N. Nicolaides, K. E. Litinas, G. K. Papageorgiou and
J. Stephanidoustephanatou, J. Heterocycl. Chem., 1991, 28,
139–143.
22 R. De Borger, C. M. L. Vande Velde and F. Blockhuys, Acta
Crystallogr., Sect. E: Struct. Rep. Online, 2005, 61, o819–o821.
23 P. L’Ecuyer, F. Turcotte, J. Giguere, C. A. Olivier and P. Roberge,
Can. J. Res., Sect. B, 1948, 26, 70–80.
24 G. Manecke and S. Luettke, Chem. Ber., 1970, 103, 700–707.
25 S. Secareanu, Chem. Ber., 1931, 64, 834–841.
26 M. Kroger and G. Fels, J. Labelled Compd. Radiopharm., 2000, 43,
217–227.
27 C. M. L. Vande Velde, A. Collas, R. De Borger and F. Blockhuys,
Chem.–Eur. J., DOI: 10.1002/chem.201002472, in press.
28 Y. H. Liu, T. L. Zhang, J. G. Zhang, L. Yang, J. Y. Guo, W. Yu,
R. F. Wu and K. B. Yu, Can. J. Chem., 2006, 84, 867–873.
29 W. J. Feast, P. W. Lovenich, H. Puschmann and C. Taliani, Chem.
Commun., 2001, 505–506.
39 L. J. Farrugia, J. Appl. Crystallogr., 1997, 30, 565.
40 C. F. Macrae, I. J. Bruno, J. A. Chisholm, P. R. Edgington,
P. McCabe, E. Pidcock, L. Rodriguez-Monge, R. Taylor, J. van de
Streek and P. A. Wood, J. Appl. Crystallogr., 2008, 41, 466–470.
41 M. J. Frisch, G. W. Trucks, H. B. Schlegel, G. E. Scuseria,
M. A. Robb, J. R. Cheeseman, J. A. J. Montgomery, T. Vreven,
K. N. Kudin, J. C. Burant, J. M. Millam, S. S. Iyengar, J. Tomasi,
V. Barone, B. Mennucci, M. Cossi, G. Scalmani, N. Rega,
G. A. Petersson, H. Nakatsuji, M. Hada, M. Ehara, K. Toyota,
R. Fukuda, J. Hasegawa, M. Ishida, T. Nakajima, Y. Honda,
O. Kitao, H. Nakai, M. Klene, X. Li, J. E. Knox, H. P. Hratchian,
J. B. Cross, C. Adamo, J. Jaramillo, R. Gomperts,
R. E. Stratmann, O. Yazyev, A. J. Austin, R. Cammi,
C. Pomelli, J. W. Ochterski, P. Y. Ayala, K. Morokuma,
G. A. Voth, P. Salvador, J. J. Dannenberg, V. G. Zakrzewski,
S. Dapprich, A. D. Daniels, M. C. Strain, O. Farkas,
D. K. Malick, A. D. Rabuck, K. Raghavachari, J. B. Foresman,
J. V. Orti, Q. Cui, A. G. Baboul, S. Clifford, J. Cioslowski,
B. B. Stefanov, A. L. G. Liu, P. Piskorz, I. Komaromi,
R. L. Martin, D. J. Fox, T. Keith, M. A. Al-Laham, C. Y. Peng,
A. Nanayakkara, M. Challacombe, P. M. W. Gil, B. Johnson,
W. Chen, M. W. Wong, C. Gonzalez and J. A. Pople, Gaussian 03,
Gaussian Inc., Pittsburg, PA, 2003.
30 J. Zyss and J. L. Oudar, Phys. Rev. A: At., Mol., Opt. Phys., 1982,
26, 2028–2048.
31 B. F. Levine, C. G. Bethea, C. D. Thurmond, R. T. Lynch and
J. L. Bernstein, J. Appl. Phys., 1979, 50, 2523–2527.
32 G. F. Lipscomb, A. F. Garito and R. S. Narang, J. Chem. Phys.,
1981, 75, 1509–1515.
33 F. H. Allen, O. Kennard, D. G. Watson, L. Brammer, A. G. Orpen
and R. Taylor, J. Chem. Soc., Perkin Trans. 2, 1987, S1–S19.
34 Enraf-Nonius, CAD-4 EXPRESS, 1994, Delft, The Netherlands.
42 T. Koopmans, Physica, 1934, 1, 104.
43 G. Gritzner and J. Kuta, Pure Appl. Chem., 1984, 56,
461–466.
c
662 New J. Chem., 2011, 35, 649–662
This journal is The Royal Society of Chemistry and the Centre National de la Recherche Scientifique 2011