Trimeric liquid crystals assembled using both hydrogen and halogen bonding

Article abstract of DOI:10.1039/b811716k

New liquid crystals are reported that are formed from non-mesomorphic components assembled using both halogen and hydrogen bonding. The Royal Society of Chemistry.

Full text of DOI:10.1039/b811716k

View Article Online / Journal Homepage / Table of Contents for this issue
COMMUNICATION
www.rsc.org/chemcomm | ChemComm
Trimeric liquid crystals assembled using both hydrogen and halogen
bondingw
a
a
b
a
¨
Carsten Prasang, H. Loc Nguyen,z Peter N. Horton, Adrian C. Whitwood
and Duncan W. Bruce*a
Received (in Cambridge, UK) 9th July 2008, Accepted 15th September 2008
First published as an Advance Article on the web 20th October 2008
DOI: 10.1039/b811716k
New liquid crystals are reported that are formed from non-
mesomorphic components assembled using both halogen and
hydrogen bonding.
The use of non-covalent interactions to realise new liquid-
crystalline species has become an area of increasing interest in
recent times.1 Thus, use of quadrupolar2 and charge-transfer3
Fig. 1 Structure of the first halogen-bonded liquid crystals.
interactions has been shown to lead to new mesomorphic
materials, while hydrogen-bonded liquid crystals have been
the subject of a resurgence in interest following the elegant
studies by Kato and Frechet and co-workers in the 1990s.4
We have worked extensively with alkoxystilbazoles as
components of various types of molecular materials5 and, as
part of these studies, have investigated hydrogen-bonded sys-
tems, primarily employing phenols as the hydrogen donor.6
More recently, we became aware of the work pioneered mainly
by the Milan group that uses halogen bonding7,8 to construct a
number of supramolecular motifs. Taking note of the analogy
between hydrogen bonding and halogen bonding, we reported
the first examples of liquid crystals formed employing the latter
interaction.9 Thus, we reported that a complex could be formed
We therefore prepared some complexes of 4-alkoxystil-
bazoles with 4-iodotetrafluorophenol15 (Fig. 2) and were
curious to see whether the mesomorphism would reflect the
angular geometry suggested by the combination of linear
halogen and hydrogen bonds (Fig. 2). Thus, complexes with
a 2 : 1 (stilbazole : iodophenol) stoichiometry (2-4 to 2-10)
were prepared by crystallisation from thf solutions and, in the
case of 2-6, detailed structural information in the solid state
was obtained by solving the single crystal X-ray structure.w
Complex 2-12 was prepared as described below for 4-n.
Two views of the structure are shown in Fig. 3 and they reveal
both that the three components of the complex are close to being
co-planar and that, despite what might have been predicted, the
overall shape of the complex is rather rod-like. This is due to the
non-linear nature of both the halogen and hydrogen bonds, so
that the ipso-C–Npyrꢀ ꢀ ꢀI angle is 164.881, while the ipso-
C–Npyr–Ophenol angle is 161.071, the combination of which
means that the two ipso-C–Npyr vectors are within 4.061 of
being parallel. The halogen bond in the complex is 2.872 A long
and it is often the case that longer bonds tend to lead to greater
between
a 4-alkoxystilbazole and iodopentafluorobenzene
(Fig. 1) and that while neither component was liquid crystalline,
the resulting complexes showed nematic and smectic A meso-
phases. Subsequently, with Metrangolo, Resnati and others we
reported on the formation of trimeric systems from stilbazoles in
combination with a,o-diiodoperfluoroalkanes10 and with
1,4-diiodotetrafluorobenzene.11 Other trimeric and polymeric
examples have also been reported.12 More recently, we have
reported on spontaneous symmetry breaking in the mesophases
of alkoxystilbazole complexes of 1,3-diiodotetrafluorobenzene.13
The levels of control that can be exerted in assembling
molecular units into useful materials potentially increase if
more than one interaction is used and so we were intrigued to
see what might result from using both hydrogen- and halogen-
bonding to assemble new liquid-crystalline species. Indeed,
Aakeroy et al. recently reported the creation of infinite, one-
dimensional chains using isonicotinamide in conjunction with
iodine and with 1,4-diiodotetrafluorobenzene.14
Fig. 2 The 2 : 1 complex of stilbazole and 4-iodotetrafluorophenol
(2-n); the 1 : 1 complex between 4-dodecyloxystilbazole and pentafluoro-
phenol (3) and between stilbazole and 4-iodotetrafluorophenol (4-n).
a Department of Chemistry, University of York, Heslington, York, UK
YO10 5DD. E-mail: db519@york.ac.uk; Tel: +44 1904 434085
b EPSRC Crystallographic Service, Department of Chemistry,
University of Southamton, Southampton, UK SO17 1BJ
w Electronic supplementary information (ESI) available: Crystal data
for 2-6, 3 and 4-1; representation of disorder and helical arrangement
in 4-1. CCDC 694375–694377. For ESI and crystallographic data in
CIF or other electronic format see DOI: 10.1039/b811716k
z Present address: High Force Research Limited, Bowburn North
Industrial Estate, Bowburn, Durham, UK DH6 5PF.
Fig. 3 Two views of the molecular structure of complex 2-6.
ꢁc
This journal is The Royal Society of Chemistry 2008
6164 | Chem. Commun., 2008, 6164–6166

View Article Online
deviations from linear halogen bonds. However, at 164.881, the
deviation is greater than that typically observed for such an
Nꢀ ꢀ ꢀI separation, suggesting that the more efficient crystal
packing offered by the more anisotropic motif is likely respon-
sible for this greater bending of the angle. However, space-filling
models indicate that interaction between the nitrogen non-
bonded electron pair and the s-hole16 on the iodine is still
readily possible. Interestingly, this results in the two I–C–Cring
angles being 123.11 and 120.21, although it is noted that a similar
effect is seen in the structure of 1-8.9 The hydrogen bond is also
distorted from an ideal alignment, contributing to the linearity.
The hydrogen was positioned in riding mode and was placed at
0.91 A from oxygen and 1.69 A from nitrogen with an O–Hꢀ ꢀ ꢀN
angle of 1691. The angle at oxygen (C–O–N) is 1301, deviating
appreciably from the ideal value of close to 1051.
that the handedness in this structure is random and that other
crystals will have the enantiomeric arrangement.
Interestingly, we note that both here and previously,13 solvent-
free, 1 : 1 complexes were obtained only when the terminal chain
was methoxy. This suggests that there is a substantial driving force
for the second iodine to find something with which to interact
(in this case forming a halogen bond with the oxygen of the
methoxy group) and that the stepped, polymeric arrangement
found in each case is sterically incompatible with longer chains.
It is interesting that in this example, the stilbazole coordinated
preferentially via a hydrogen bond, which is intuitively what might
have been expected due to the generally lower strength and higher
lability of halogen bonds in such simple, neutral systems.
However, there is a report from Corradi et al.17 where, in
competitive experiments, the halogen-bonded polymer between
1,2-bis(4-pyridyl)ethane and 1,4-diiodotetrafluorobenzene was
found to crystallise in preference to the analogous polymer
where 1,4-hydroquinone acted as the Lewis acid.
As 4-iodotetrafluorophenol presents two possible sites with
which the stilbazole nitrogen can interact and because we have
already given examples of the lability of halogen bonds at elevated
temperatures,11,13 we then elected to prepare some examples of a
1 : 1 complex. Grown from thf solution, single crystals of 4-4
showed the stilbazole hydrogen bonded to the phenol with a
molecule of thf halogen bonded to the iodine; this structure will
be reported separately. However, grown from diethyl ether, a
solvent-free, 1 : 1 complex of 4-1 was obtained in which the
stilbazole is hydrogen bonded to the phenolic hydrogen (see
Fig. 4). Thus, the stilbazole nitrogen is hydrogen bonded to the
phenolic oxygen and there is an Nꢀ ꢀ ꢀO separation of 2.596 A. The
For the purposes of comparison, the single crystal structure
of the complex between dodecyloxystilbazole and pentafluoro-
phenol was obtained; the molecular structure is shown in
Fig. 5. The structure shows some broad similarities to that
of 2-6 in respect of the hydrogen bond, so that the ipso-
C–Npyr–Ophenol angle in 3 is 166.67(12)1, somewhat straighter
than that found in 2-6 yet still some 131 from linear; the
N–O–C angle is 1281. The hydrogen (again located by differ-
ence map is once more attached to oxygen (dO–H = 0.95 A and
dHꢀ ꢀ ꢀN = 1.69 A) and dNꢀ ꢀ ꢀO = 2.612(3) A.
hydrogen is attached to oxygen (dO–H = 0.80 A and dHꢀ ꢀ ꢀN
=
1.82 A) and the hydrogen bond is not linear so that the
Cipso–N–Ophenol angle is 159.81, with an O–Hꢀ ꢀ ꢀN angle of
1651 (hydrogen located by difference map). The stilbazole itself
is almost planar but makes an angle of nearly 1201 to the plane of
the phenolic ring. Then, in common with the structure of a 1 : 1
complex of methoxystilbazole with 1,3-diiodotetrafluoroben-
zene,13 the iodine forms a halogen bond to the ether oxygen of
the methoxy group, leading to a polymeric arrangement, this time
via both hydrogen and halogen bonding. The Iꢀ ꢀ ꢀO distance is
3.05 A and the C–Iꢀ ꢀ ꢀO angle is 175.161; these data compare with
3.08 A and 171.061, respectively, as reported previously.13 The
stilbazole is disordered by virtue of occupying two positions,
related by a 1801 rotation about the N–O axis. The two sites for
each of the ethylenic carbons were resolved readily and are
included in the model. Separate positions for the carbons of the
aromatic rings could not, however, be resolved, and the disorder
manifested itself by an elongation of the ADP ellipsoids for these
atoms, which increased with proximity to the double bond. The
relative proportions of each stilbazole was refined in the ratio
70 : 30. Representations of the disordered structure are found in
the ESI.w Curiously, the complex crystallised in the chiral space
group P212121, the origin of the chirality being a right-handed,
helical precession of the polymer as shown in the ESI.w We assume
In order to study the thermal properties of the bulk materials,
the complexes were prepared by crystallisation from thf to give
crystalline solids, which were characterised by elemental analysis.
Thus, the thermal properties of the materials were investigated by
polarised optical microscopy and the transition temperatures are
collected in Table 1. The shortest-chain homologue investigated,
2-4, showed a high melting point (ca. 117 1C) below which a
monotropic nematic phase was identified from its characteristic
schlieren texture with two- and four-brush disclinations. The
remainder of the complexes studied showed enantiotropic meso-
phases. Thus, while the clearing points did not vary very much
with chain length, the melting point dropped sharply for 2-6 by
almost 30 1C to 88.4 1C (compared to 2-4), while the clearing point
of the nematic phase was identical to that of 2-4. As the chain
length increased through C8 to C10 and C12, the stability of the
nematic phases dropped slightly reaching a minimum at 2-10
(100 1C), recovering slightly at 2-12 (102 1C), for which homologue
a SmA phase is observed for two degrees above the melting point.
Consistent with their structural similarity, the stability of the
nematic phases of these 2 : 1 complexes is comparable with that of
the 2 : 1 complexes of alkoxystilbazoles with 1,4-diiodotetrafluoro-
benzene11 where, for example, the C8 and C10 homologues clear
(TNI) at 113.5 and 109 1C, respectively. The slightly lower phase
stability in the present case is most likely due to the complexes
adopting a more angular shape in the fluid phase consistent with
Fig. 4 The crystal structure of 4-1, showing the two halogen bonds.
Fig. 5 The molecular structure of 3.
ꢁc
This journal is The Royal Society of Chemistry 2008
Chem. Commun., 2008, 6164–6166 | 6165

View Article Online
Table 1 Thermal behaviour of the new complexes
comparison, it was felt that the data were acceptable and fit for
purpose. Thus, the higher homologues, namely 4-10 and 4-12,
showed enantiotropic SmA phases. Complex 4-10 melted at 85.2
and cleared at 98.1 1C, while 4-10 melted at 87.1 and cleared at
102.1 1C.
Complex
Transition
T/1C
DH/kJ molꢂ1
2-4
Cr–I
116.5
(107.4)
88.4
107.4
85.6
101.7
89.8
100.0
97.8
60.4
(2.8)
57.6
2.1
88.8
2.5
(N–I)a
Cr–N
N–I
2-6
It is interesting to compare these temperatures with those of
complex 3 and also complexes 1-n.9 Thus, 3 also shows a SmA
phase but in this case, the crystal phase is more stable and so it
appears monotropically with TSmA–I at 94 1C. Similarly, 1-12 also
showed a SmA phase, this time enantiotropic melting at 81 and
clearing at 84 1C. Thus, the transition temperatures are lower
where stilbazole is bound to C6F5I when compared to C6F5OH,
and lower where the stilbazole is bound to C6F5OH rather than
p-IC6F4OH. On one hand, this would suggest that the more
electrostatic nature of the hydrogen bond contributes favourably
to phase stabilisation, and on the other it suggests that p-C6F4–I is
a more effective terminal group than p-C6F4–F for phase stabilisa-
tion, presumably due to the great polarisability anisotropy.
We thank EPSRC for support (CP and HLN).
2-8
Cr–N
N–I
Cr–N
N–I
Cr–SmA
SmA–N
N–I
2-10
2-12b
62.0
1.9
—
100.2
102.0
96.0
86.9c
5.4
3d
Cr–I
49
(SmA–I)
Cr–I
Cr–I
(94.0)
105.8
103.6
107.0
(92.4)
85.2
98.1
87.1
102.1
(2.0)
29.9
34.0
56.5
(1.8)
28.1
1.5
4-4b
4-6b
4-8
Cr–I
(N–I)
Cr–SmA
SmA–I
Cr–SmA
SmA–I
4-10b
4-12b
39.9
3.9
a
b
Monotropic transitions in parentheses. Prepared by evaporation of
c
1 : 1 solution in thf. Combined enthalpy for Cr–SmA–N; not
d
resolved by DSC. Data from ref. 5.
Notes and references
1 D. W. Bruce, Struct. Bonding, 2008, 126, 161.
2 C. Dai, P. Nguyen, T. B. Marder, A. J. Scott, W. Clegg and
C. Viney, Chem. Commun., 1999, 2493; N. Boden, R. J. Bushby,
Z. Lu and O. R. Lozman, Liq. Cryst., 2001, 5, 657; N. Boden,
R. J. Bushby, Z. Lu and O. R. Lozman, J. Mater. Chem., 2001, 11,
1612; N. Boden, R. J. Bushby, G. Cooke, O. R. Lozman and Z. Lu,
J. Am. Chem. Soc., 2001, 123, 7915.
3 K. Praefcke and J. D. Holbrey, J. Inclusion Phenom. Mol. Recog-
nit. Chem., 1996, 24, 19; T. Hegmann, J. Kain, S. Diele, G. Pelzl
and C. Tschierske, Angew. Chem., Int. Ed., 2001, 40, 887.
4 For reviews see e.g. T. Kato, in Handbook of Liquid Crystals, ed.
D. Demus, G. W. Gray, J. Goodby, H.-W. Spiess and V. Vill,
Wiley-VCH, Weinheim, 1998; D. Tsiourvas and C. Paleos, Angew.
Chem., Int. Ed. Engl., 1995, 34, 1696.
slightly more linear halogen and hydrogen bonds, able to form
when the constraints of crystal packing are lifted.
In previous studies of trimeric, halogen-bonded liquid crystals
containing two stilbazole molecules,11,13 we had postulated the
dissociation of one of these at the clearing point, evidenced by the
rather large enthalpy changes observed (typically 8 to 9 kJ molꢂ1).
The clearing enthalpies for 2-n are much smaller at between 2 and
3 kJ molꢂ1, which we take as evidence that clearing is not
accompanied by dissociation. Certainly studies carried out on
purely hydrogen-bonded liquid crystals showed conclusively that
clearing was neither driven nor accompanied by rupture of the
hydrogen bond18 and so in the present materials where there is
only one halogen bond, the clearing transition appears to be that
of the intact complex.
5 D. W. Bruce, Adv. Inorg. Chem., 2001, 52, 151.
6 See e.g. (a) K. Willis, D. J. Price, H. Adams, G. Ungar and
D. W. Bruce, J. Mater. Chem., 1995, 5, 2195; (b) D. J. Price,
K. Willis, T. Richardson, G. Ungar and D. W. Bruce, J. Mater.
Chem., 1997, 7, 883; (c) D. W. Bruce and D. J. Price, Adv. Mater.
Opt. Electron., 1994, 4, 273.
1 : 1 Complexes (4-n) between the stilbazoles and 4-iodotetra-
fluorophenol were also obtained for comparison. Preparation of
these complexes was not so straightforward and only for 4-8 was a
crystalline complex obtained from thf. We believe that this
complex had thf halogen bonded to the free iodine, but this thf
is readily removed in vacuo and does not influence the meso-
morphism. Indeed, 4-8 was the first homologue to show any
mesomorphism and, following melting to the isotropic phase at
107.0 1C, a monotropic nematic phase was found at 92.4 1C.
Because of the difficulty associated with preparing crystalline
derivatives, other homologues were obtained by dissolving the
components together in thf, evaporating the solvent slowly and
then finally pumping to ensure complete removal. While we have
found this method to be very good where hydrogen-bonded
materials are concerned, in general it has proved less satisfactory
where halogen-bonded materials are concerned. Nonetheless, we
re-prepared 4-8 by this method and found Cr–I at 105.8 1C and
(I–N) at 90.2 1C. The lower temperatures and slightly broader
transitions are indicative of a material that is not perfectly pure,
which in this case means that it is not single phase. Nonetheless,
the temperatures are rather close and so for the purposes of
7 O. Hassel and C. Rømming, Q. Rev. Chem. Soc., 1962, 16, 1;
A. C. Legon, Angew. Chem., Int. Ed., 1999, 38, 2687; A. C. Legon,
Chem.–Eur. J., 1998, 4, 1980.
8 P. Metrangolo, H. Neukirch, T. Pilati and G. Resnati, Acc. Chem.
Res., 2005, 38, 386; Halogen Bonding: Fundamentals and Applica-
tions, Struct. Bonding, ed. P. Metrangolo and G. Resnati, 2008, 126.
9 H. L. Nguyen, P. N. Horton, M. B. Hursthouse, A. C. Legon and
D. W. Bruce, J. Am. Chem. Soc., 2004, 126, 16.
10 P. Metrangolo, C. Prasang, G. Resnati, R. Liantonio,
A. C. Whitwood and D. W. Bruce, Chem. Commun., 2006, 3290.
11 D. W. Bruce, F. Chaux, P. Metrangolo, F. Meyer, C. Prasang,
G. Resnati and A. C. Whitwood, New J. Chem., 2008, 32, 477.
12 J. Xu, X. Xueming, T. Lin, J. Huang and C. He, Macromolecules,
2005, 38, 3554; J. Xu, X. Liu, J. K.-P. Ng, T. Lin and C. He,
J. Mater. Chem., 2006, 16, 3540.
13 C. Prasang, A. C. Whitwood and D. W. Bruce, Chem. Commun.,
2008, 2137.
14 C. B. Aakeroy, J. Desper, B. A. Helfrich, P. Metrangolo, T. Pilati,
G. Resnati and A. Stevenazzi, Chem. Commun., 2007, 4236.
15 J. Wen, H. Yu and Q. Chen, J. Mater. Chem., 1994, 4, 1715.
16 T. Clark, M. Hennemann, J. S. Murray and P. Politzer, J. Mol.
Model., 2007, 13, 291.
17 E. Corradi, S. V. Meille, M. T. Messina, P. Metrangolo and
G. Resnati, Angew. Chem., Int. Ed., 2000, 39, 1782.
18 D. J. Price, T. Richardson and D. W. Bruce, J. Chem. Soc., Chem.
Commun., 1995, 1911.
ꢁc
This journal is The Royal Society of Chemistry 2008
6166 | Chem. Commun., 2008, 6164–6166