Hydrogen-bonded complexes between 4-alkoxystilbazoles and fluorophenols: Solid-state structures and liquid crystallinity

Article abstract of DOI:10.1002/chem.201201906

48 new hydrogen-bonded complexes have been prepared by combining 16 fluorophenols of general formula C₆F_nH_5-nOH with three different alkoxystilbazoles (butyloxy-, octyloxy- and dodecyloxy-). Single-crystal X-ray structures were obtained for 10 of the 16 complexes of octyloxystilbazole from which it was found that most of the structures could be collected into one of two groups according to both the motif shown by the complex and by the solid-state packing. Because all but one crystallised in the Pβar 1$ space group, meaningful comparisons could be drawn and it was observed that six structures were extremely close in nature so that significant molecular overlap was found. On this basis, doubt is cast on the significance of some of the weaker intermolecular contacts found in the solid state. 40 of the new complexes showed liquid-crystal properties and it was found that although complexes of butyloxystilbazole were all nematic, almost all of those with dodecyloxystilbazole showed a smectic A (SmA) phase. Complexes of octyloxystilbazole showed a mixture of both. Structure/property correlations showed that clearing points were independent of the pK_a of the phenol. The most stable mesophases were found when the fluorophenol contained a fluorine at the 2-position, which was interpreted in terms of the formation of an intramolecular H···F hydrogen bond to give a six-membered ring linking the two components into a stable, coplanar conformation. The least stable mesophases were found when no such ring formation was possible and the phenol was relatively free to move.

Full text of DOI:10.1002/chem.201201906

FULL PAPER
DOI: 10.1002/chem.201201906
Hydrogen-Bonded Complexes between 4-Alkoxystilbazoles and
Fluorophenols: Solid-State Structures and Liquid Crystallinity
Joanna P.-W. Wong, Adrian C. Whitwood, and Duncan W. Bruce*[a]
Abstract: 48 new hydrogen-bonded
structures were extremely close in
nature so that significant molecular
overlap was found. On this basis, doubt
is cast on the significance of some of
the weaker intermolecular contacts
found in the solid state. 40 of the new
complexes showed liquid-crystal prop-
erties and it was found that although
complexes of butyloxystilbazole were
all nematic, almost all of those with do-
decyloxystilbazole showed a smectic A
(SmA) phase. Complexes of octyloxys-
complexes have been prepared by com-
bining 16 fluorophenols of general for-
mula C6FnH5ꢀnOH with three different
alkoxystilbazoles (butyloxy-, octyloxy-
and dodecyloxy-). Single-crystal X-ray
structures were obtained for 10 of the
16 complexes of octyloxystilbazole
from which it was found that most of
the structures could be collected into
one of two groups according to both
the motif shown by the complex and
by the solid-state packing. Because all
tilbazole showed a mixture of both.
Structure/property correlations showed
that clearing points were independent
of the pKa of the phenol. The most
stable mesophases were found when
the fluorophenol contained a fluorine
at the 2-position, which was interpreted
in terms of the formation of an intra-
molecular H···F hydrogen bond to give
a six-membered ring linking the two
components into
a stable, coplanar
conformation. The least stable meso-
phases were found when no such ring
formation was possible and the phenol
was relatively free to move.
Keywords: fluorophenols · hydro-
gen bonding · liquid crystals · stilba-
zoles · X-ray diffraction
¯
but one crystallised in the P1 space
group, meaningful comparisons could
be drawn and it was observed that six
Introduction
Halogen bonding may be regarded as formally analogous
to the better-known hydrogen bonding, and it was through
this analogy that our interest in the subject grew. Thus, Kato
and Frꢀchet and their co-workers[4] did a great deal to rein-
vigorate work into hydrogen bonding in liquid crystals using
a conceptually simple and elegant approach (Figure 1). With
Liquid crystals are the original supramolecular materials be-
cause their mesophases are stabilised by non-covalent, inter-
molecular interactions, namely anisotropic dispersion forces
that arise from the anisometric nature of the materials
giving rise to the phases. However, in addition, specific in-
termolecular interactions may be employed to modify
liquid-crystal behaviour or even to induce it through the in-
teraction of non-mesomorphic components.[1] One such in-
teraction that has been described in recent years is halogen
bonding,[2] which is an attractive, electrostatic interaction be-
tween a halogen atom (normally iodine) that is electron
poor and a Lewis base (most often nitrogen). Thus, through
the supramolecular chemistry pioneered in Milan,[3] halogen
bonding has become a more widely used interaction in the
formation of what can be quite complex supramolecular spe-
cies.
Figure 1. Examples of hydrogen-bonded mesogens reported by Kato,
Frꢀchet and co-workers.
that inspiration, previous work from this group reported on
liquid-crystalline complexes formed from stilbazoles and
various phenolic[5] and other[6] species (Figure 2), while
others have built elegant structures of higher symmetry
using complementary bonding approaches.[7]
However, more recently aware of halogen bonding, we
published the first example of a liquid-crystalline material
formed using a halogen bond, namely a halogen-bonded
complex between an alkoxystilbazole and iodopentafluoro-
benzene.[8] Other examples followed quickly both from our-
selves[9] and others.[10] In wishing to extend this work, we
[a] J. P.-W. Wong, Dr. A. C. Whitwood, Prof.Dr. D. W. Bruce
Department of Chemistry
University of York
Heslington
YORK YO10 5DD (UK)
Tel : (+44)1904-324085
E-mail: duncan.bruce@york.ac.uk
Supporting information for this article is available on the WWW
under http://dx.doi.org/10.1002/chem.201201906.
Chem. Eur. J. 2012, 18, 16073 – 16089
ꢁ 2012 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
16073

crystallinity of stilbazole complexes of fluorophenols from
which one might learn about the way both hydrogen bond
strength and level and pattern of fluorine substitution affect-
ed mesomorphism.
In planning the study, a decision was made to use stilba-
zoles with three different chain lengths, namely C4, C8 and
C12, with the rationale that complexes of the shortest homo-
logue ought to be dominated by the formation of the nemat-
ic phase, whereas complexes of the longest homologues
would be dominated by smectic phases (probably SmA),
with the C8 derivatives showing a mixture of both. Insofar
as the phenols are concerned, there are 19 isomers of
C6FnH5ꢀnOH with n¼0 (see Figure S1 in the Supporting In-
formation) of which 16 were chosen for study leading to 48
new complexes that were prepared and studied for their
liquid-crystal properties. In considering an accompanying
structural study, a decision was made to look at complexes
of octyloxystilbazole and to this end, single-crystal structures
were obtained for 10 of the possible 16 complexes, along
with 1 complex of a butyloxystilbazole.
Figure 2. Hydrogen-bonded phenols and other systems prepared previ-
ously in this group.
had hoped to prepared and study a range of complexes
based on alkoxystilbazoles and various isomers of iodofluor-
obenzenes, C6FnH5ꢀnI, but many of these proved resistant to
crystallisation and so we undertook a related, structure-only
study of co-crystals formed between C6FnH5ꢀnI and 4-(N,N-
dimethylamino)pyridine (DMAP).[11] One of the things that
emerged from that study was a very good correlation be-
Results
The hydrogen-bonded complexes to be studied were pre-
pared by separately dissolving the appropriate stilbazole and
fluorophenol in the same solvent and then combining the
two solutions with stirring at room temperature for
1.5 hours. Pentane was used as the solvent for complexation
for the 4-butoxy-4’-stilbazole and 4-octyloxy-4’-stilbazole ho-
mologues, whereas hexane was preferred for the 4-dodecy-
loxy-4’-stilbazole complexes. The 1:1 composition of the
crystalline complexes so obtained was established by ele-
mental analysis and the data are given in Table S1 (in the
Supporting Information). Complexes of pentafluorophenol
were a pale yellow colour and, as the degree of fluorination
decreased, the yellow colour became ever paler so that com-
plexes of the three monofluorophenols were colourless. The
colour has its origin in the degree of proton transfer to the
stilbazole nitrogen. Thus, previously we showed that al-
though pure stilbazoles had lmax =320 nm, the fully proto-
nated stilbazolium cation absorbs at 390 nm,[5a] the signifi-
cant shift being in large part due to the fact that alkoxystil-
bazoles are good chromophores for quadratic non-linear op-
tical processes.[13] That the absorption maximum is sensitive
tween the length of the halogen bond, d
number of fluorine atoms on the ring. This correlation im-
proved further when d(N···I) was plotted against the calcu-
lated pKa values for the related phenol, C6FnH5ꢀnOH
(Figure 3),[12] for which the pKa is, of course, a measure of
the stability of C6FnH5ꢀnOꢀ, which is in turn a proxy for the
development of positive electrostatic potential on the iodine
atom of C6FnH5ꢀnI.
ACHTUNGTREN(NGNU N···I), and the
AHCTUNGTRENNUNG
Encouraged by these studies, it then became attractive to
undertake a systematic study of the structure and liquid-
ꢀ
to the position of the hydrogen in the N···H O hydrogen
bond is something we used previously to demonstrate
proton transfer in the liquid-crystal phase behaviour of al-
koxystilbazoles with 2,4-dinitrophenol.[5b]
To study the solid-state behaviour of these complexes,
those from octyloxystilbazole were chosen and structures
were determined for 10 of the 16 complexes prepared. The
single crystals were obtained from pentane or hexane at am-
bient temperature or 48C, and selected crystallographic data
are collected in Table 1. A majority of the complexes crys-
tallised as bunches of needles growing from a single nuclea-
tion point, whereas some grew as plates and others blocks.
Figure 3. Plot of calculated pKa of C6F5ꢀnHnOH versus dACTHNUGRTNEUNG(N···I) for a
series of co-crystals of DMAP with C6F5ꢀnHnI. Adapted from ref. [10] ꢁ
American Chemical Society 2009.
16074
www.chemeurj.org
ꢁ 2012 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Chem. Eur. J. 2012, 18, 16073 – 16089

Hydrogen-Bonded Complexes between 4-Alkoxystilbazoles and Fluorophenols
FULL PAPER
C7b, were constrained to be
equal. All subsequent treatment
and analysis of 2-8 uses a modi-
fied cif file that contains the
stilbazole in its major site occu-
pancy.
Thus, complexes 2-8, 2-7 and
2-6b were found to be nearly
planar (Figure 5) and the angles
that the planes of the pyridine
rings make with the fluorophe-
nol rings were 7.828, 5.848 and
ꢀ
1.858, respectively. The N···H
O bonds (Figure 6) were found
at an angle of 170.968, 165.118
and 158.018, respectively, and
Scheme 1. The three different stilbazoles used in combination with different fluorinated phenols; all Rx sub-
stituents are hydrogen where not indicated as fluorine.
The complexes formed are numbered according to the rep-
resentations shown in Scheme 1.
All short contacts were defined as being less than the sum
of the van der Waals radii minus 0.1 ꢂ ([ꢀvdWꢀ0.1] ꢂ). It
is instructive to note that all but one complex (2-4c) crystal-
¯
lised in the space group P1 and so comparisons between
them are meaningful. The discussion of the complexes is
grouped so that those with similar arrangements in the solid
state are treated together. In addition, crystals were ob-
tained of complex 1-6e for which the structure is reported
although not discussed in any detail.
Figure 5. Top and side view of the molecular structure of complex 2-7.
Description of the crystal structures
Complex 1-6e: The complex crystallised without disorder.
The two stilbazole rings were close to co-planar (angle be-
tween planes=5.068) and the fluorophenol ring was bound
to the pyridyl nitrogen through a hydrogen bond of length
1.708 ꢂ, which was reasonably close to linear at 171.318.
The fluorophenyl ring made an angle of 68.618 to the pyridyl
ring of the stilbazole (Figure 4). The packing of this complex
is closest to that of 2-5a, 2-5d, 2-5 f, 2-4a and 2-4b described
below.
Figure 6. The interaction between the phenolic hydrogen and the pyri-
dine in a) 2-8 and b) 2-6b.
the corresponding hydrogen bond lengths, dN···H were 1.553,
1.507 and 1.729 ꢂ. For 2-8 and 2-7, the planar structure led
to (or perhaps resulted from) a second interaction between
an ortho hydrogen on the pyridine ring and a fluorine ortho
to the hydroxyl function on the phenol; the distances were
2.234 ꢂ (2-8) and 2.360 ꢂ (2-7). In 2-6b, the same distance
was 3.081 ꢂ, the increased distance resulting from the bend-
ing back of the phenol, which was reflected in the smaller
Figure 4. Molecular structure of complex 1-6e.
Complexes 2-8, 2-7 and 2-6b: Complex 2-8 is disordered
and so both of the aromatic rings and the double bond of
the stilbazole exhibited disorder and were modelled in two
positions, the relative occupancy of which refined to
0.77:0.23. For the minor component the aromatic rings were
constrained to be regular hexagons with a bond length of
1.39 ꢂ. The atomic displacement parameters (ADP) of cor-
responding pairs of disordered atoms for example, C7a and
ꢀ
N···H O bond angle.
Two molecules of the complex then dimerise with the
phenols aligned back-to-back. In complex 2-8, the back-to-
Chem. Eur. J. 2012, 18, 16073 – 16089
ꢁ 2012 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chemeurj.org
16075

D. W. Bruce et al.
Table 1. Selected crystallographic properties of the hydrogen-bonded complexes.
2-8
2-7
2-6b
2-6e
2-5a
2-5d
CCDC[a]
F-substitution
fluorophenol pKa
884513
2,3,4,5,6
5.50
884522
2,3,5,6-
5.91
884517
2,3,6-
6.49
884515
3,4,5-
8.60
884516
2,3-
8.65
884518
2,6-
7.07
d
(N···H) [ꢂ]
1.553
1.507
1.729
158.01
P1
1.685
177.08
P1
1.667
173.20
P1
1.580
167.99
P1
qN···H-O [8]
space group
170.96
165.11
¯
¯
¯
¯
¯
¯
P1
P1
crystal system
unit cell dimensions [ꢂ]
triclinic
triclinic
triclinic
triclinic
triclinic
triclinic
a=5.6909(10)
b=12.661(2)
c=16.824(3)
a=99.906(4)
b=95.535(4)
g=92.599(4)
1186.2(4)
a=5.6936(15)
b=12.539(3)
c=16.714(4)
a=99.854(5)
b=95.866(5)
g=92.065(5)
1167.7(5)
a=6.1720(3)
b=11.3711(5)
c=16.6214(7)
a=89.5710(10)
b=89.5410(10)
g=81.6710(10)
1154.17(9)
a=10.4066(6)
b=14.3652(8)
c=15.9121(8)
a=87.7860(10)
b=85.4180(10)
g=87.0740(10)
2366.6(2)
a=7.7812(9)
b=9.7838(11)
c=15.2278(17)
a=87.505(2)
b=89.314(2)
g=88.837(2)
1157.9(2)
a=7.7808(10)
b=9.6495(11)
c=15.3919(19)
a=92.913(3)
b=90.091(3)
g=90.415(3)
1154.1(2)
unit cell angles [8]
volume [ꢂ3]
Z
2
2
2
4
2
2
final R indices [I>2sI]
R1 =0.0370,
wR2 =0.0906
R1 =0.0622,
wR2 =0.1034
0.205 and ꢀ0.177
R1 =0.0482,
wR2 =0.1047
R1 =0.1066,
wR2 =0.1249
0.209 and ꢀ0.257
R1 =0.0422,
wR2 =0.1194
R1 =0.0473,
wR2 =0.1256
0.484 and ꢀ0.255
R1 =0.0434,
wR2 =0.1110
R1 =0.0648,
wR2 =0.1245
0.315 and ꢀ0.202
R1 =0.0448,
wR2 =0.1154
R1 =0.0546,
wR2 =0.1235
0.314 and ꢀ0.239
R1 =0.0439,
wR2 =0.0953
R1 =0.0632,
wR2 =0.1067
0.181 and ꢀ0.187
R indices (all data)
largest diff. peak and
hole [eꢂꢀ3
]
2–5 f
2–4a
2–4b
2–4c
1–6e
884521
3,4,5-(C4 stilb)
8.60
CCDC[a]
F-substitution
fluorophenol pKa
884512
3,5-
8.66
884519
2-
8.73
884520
3-
9.29
884514
4-
9.89
d
(N···H)/ꢂ
1.707
176.82
P1
triclinic
a=7.7371(4)
1.714
175.86
P1
triclinic
a=7.8382(11)
1.716
175.53
P1
triclinic
a=7.6690(9)
1.828
169.59
P1
orthorhombic
a=14.4911(7)
1.708
qN···H-O [8]
171.31
¯
¯
¯
¯
¯
space group
crystal system
unit cell dimensions
[ꢂ]
P1
triclinic
a=9.4570(11)
b=9.7508(5)
c=15.436(8)
a=92.6130(10)
b=90.0900(10)
g=91.7080(10)
1162.83(10)
2
b=9.5985(13)
c=15.231(2)
a=88.188(3)
b=89.391(3)
g=87.520(3)
1144.2(3)
b=9.8547(11)
c=15.2857(17)
a=86.725(2)
b=89.923(2)
g=88.706(2)
1153.0(2)
b=6.1494(3)
c=51.527(3)
a=90
b=90
g=90
b=9.6167(12)
c=12.9396(16)
a=76.006(2)
b=77.024(2)
g=60.548(2)
986.5(2)
unit cell angles [8]
volume [ꢂ3]
Z
4591.7(4)
8
2
2
2
final R indices
[I>2sI]
R indices (all data)
R1 =0.0442,
wR2 =0.1141
R1 =0.0537,
wR2 =0.1218
0.283 and ꢀ0.215
R1 =0.0422,
wR2 =0.1077
R1 =0.0590,
wR2 =0.1178
0.331 and ꢀ0.233
R1 =0.0435,
wR2 =0.1192
R1 =0.0614,
wR2 =0.1338
0.305 and ꢀ0.221
R1 =0.0497,
wR2 =0.1145
R1 =0.0615,
wR2 =0.1223
0.305 and ꢀ0.209
R1 =0.0489,
wR2 =0.1317
R1 =0.0574,
wR2 =0.1415
0.589 and ꢀ0.258
largest diff. peak and
hole [eꢂꢀ3
]
[a] CCDC numbers given contain the supplementary crystallographic data for this paper. These data can be obtained free of charge from The Cambridge
Crystallographic Data Centre via www.ccdc.cam.ac.uk/data_request/cif.
back dimers are offset and are connected by two F···F con-
tacts at 2.719 ꢂ (Figure 7a), with the distance between the
planes of the two parallel pentafluorophenol rings being
0.22 ꢂ. The fluorophenols then propagate perpendicular to
the dimer through F···F contacts of 2.792 ꢂ between fluorine
atoms para to the phenol oxygen with the two parallel rings
separated by 2.707 ꢂ (Figure 7b). The offset nature of the
stacking F···F contacts may suggest that they arise through
crystal packing effects and that perhaps they are not strong-
ly structure directing.
oxygen and a hydrogen meta to the pyridine nitrogen (Fig-
ure 8a). This contact is present in 2-7 and 2-6b, where it is
slightly shorter (see below). Formally, there is also another
F···F (2.78 ꢂ) and even an H···H (2.27 ꢂ) contact showing
up, but it is contended that these are probably not structure
directing. In 2-7 (Figure 8b), the dimers are connected by
two contacts between F3 and H4 (2.396 ꢂ), whereas there
are no such hydrogen–fluorine contacts in 2-6b (Figure 8c).
As in 2-8, the two fluorophenol rings in 2-7 and 2-6b are co-
parallel, their planes being separated by 0.547 and 0.272 ꢂ,
respectively.
The stilbazole/pentafluorophenol unit is then dimeric in
the solid state and these dimers propagate laterally into
sheets, perhaps through the presence of intermolecular H···F
contacts of 2.46 ꢂ between a fluorine ortho to the phenolic
These dimers then drive the structure packing by arrang-
ing side-by-side in a slipped fashion, so that two fluorophe-
nol rings lie next to the two aromatic rings in the stilbazole
16076
www.chemeurj.org
ꢁ 2012 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Chem. Eur. J. 2012, 18, 16073 – 16089

Hydrogen-Bonded Complexes between 4-Alkoxystilbazoles and Fluorophenols
FULL PAPER
ious short contacts. For example, in 2-7 there are intermolec-
ular H···F contacts at 2.408 and 2.454 ꢂ (Figure 8b), where-
as in 2-6b intermolecular H···F contacts are found at 2.354
and 2.611 ꢂ (Figure 8c).
The stacking of the fluorophenol units in 2-8 is reflected
in the packing orthogonal to the sheets shown in Figure 8a,
and so a kind of segregated structure propagates as shown
in Figure 9a. The situation for 2-7 appears very similar when
viewed from the side (Figure 9b), but the difference is that
there are no short, intermolecular contacts in the stacking
direction and in fact the offset of the rings is different in 2-8
and 2-7 when viewed from above (Figure S2 in the Support-
ing Information). In 2-6b, there is no apparent segregation
in the stacking direction (Figure 9c).
Figure 7. Side-to-side (a) and vertical (b) arrangement of the pentafluoro-
phenol units in 2-8, the latter being tilted in view and showing F···F con-
tacts.
¯
Complex 2-6e: Although still in the P1 space group, the
most significant feature of 2-6e is that the fluorophenol ring
is no longer co-planar with the stilbazole. The structure is
most easily described at the local level by considering the
existence of a dimeric unit that consists of two face-to-face
of a neighbouring dimer. This packing forms a flat, one-di-
mensional ribbon-like layer, which may be supported by var-
ꢀ
fluorophenols arranged so that the two O H vectors are for-
Figure 8. Sheet-like arrangement in the crystal packing of a) complex 2-8, b) complex 2-7 and c) complex 2-6b.
Chem. Eur. J. 2012, 18, 16073 – 16089
ꢁ 2012 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chemeurj.org
16077

D. W. Bruce et al.
Figure 9. Stacking of complexes a) 2-8, b) 2-7 and c) 2-6b.
Figure 10. The molecular structure of complex 2-6e showing the general dimeric arrangement and the orthogonality of the stilbazole and phenol.
mally anti to one another through an inversion centre and
are then hydrogen bonded to a stilbazole (Figure 10). Two
different types of these dimers make up the crystal structure
of 2-6e, termed dimer A and dimer B. Dimer A forms
through the stacking of two fluorophenol rings, which inter-
act through a p–p interaction with an interplane separation
of 3.248 ꢂ.
a greater distance, 3.413 ꢂ, which suggests a weaker p–p in-
teraction. Such a conclusion is supported by viewing the
dimer from above (Figure 11b) where it can be seen that
the degree of face-to-face overlap is very much less. In
ꢀ
dimer A, the N···H hydrogen bond is 1.685 ꢂ with an N···H
O angle of 177.098, whereas in dimer B, the hydrogen bond
is longer at 1.729 ꢂ, although the angle is very similar at
176.988. In dimer A, the angle between the pyridine ring
and the fluorophenol ring planes is 58.378, whereas in
dimer B it is 58.238.
Viewed from above, the fluorophenols are stacked with a
ꢀ
displacement of a C C bond length (Figure 11a). Dimer B
differs in that the two fluorophenol planes are separated by
16078
www.chemeurj.org
ꢁ 2012 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Chem. Eur. J. 2012, 18, 16073 – 16089

Hydrogen-Bonded Complexes between 4-Alkoxystilbazoles and Fluorophenols
FULL PAPER
The dimer then lines up side-
by-side in a stepped fashion to
form a flat “segment” as shown
in Figure 12, with all stilbazole
rings almost co-planar or flat.
Two dimer pairs are linked by
two short contacts between O1
and H7, 2.531 ꢂ in length
(97.0% of the sum of the vdW
radii). Two “segments” then fit
together through an interdigita-
tion of the stilbazole alkoxy
chains (Figure 12), while still
keeping all stilbazole aromatic
rings in plane. This extends the
layer into an infinite sheet of
dimer A or dimer B in the xy
plane.
whereas the angle between the aromatic planes of the stilba-
zole moiety is 16.228; both such angles are ranked the larg-
est among all the crystal structures obtained (Figure 13a).
The hydrogen-bond length is 1.667 ꢂ and the angle at hy-
drogen is 173.28. The fluorophenol ring forms a back-to-
back dimer with another fluorophenol ring through two
F···F contacts of 2.833 ꢂ and the distance between the
planes of these two rings in 1.851 ꢂ (Figure 13b).
The structure propagates in the ac plane as follows. The
pyridine ring of each stilbazole forms a p–p interaction with
another stilbazole in an antiparallel fashion with a ring
plane separation of 3.262 ꢂ (Figure 14a). As was the case
with the two fluorophenol rings in complex 2-6e, the rings
Figure 11. View of the two dif-
ferent dimeric arrangements of
phenolic groups in the struc-
ture
of
complex
2-6e
ꢀ
are slipped by one C C bond length. This stilbazole then
a) dimer A and b) dimer B
(see text).
hydrogen bonds into one end of a back-to-back difluorophe-
nol dimer, at the other end of which is a second stilbazole
above which is a further stilbazole and so the motif contin-
ues (Figure 14b).
In complex 2-5d, the angle between the planes of the pyr-
idine and fluorophenol rings is 64.818, whereas the two stil-
bazole aromatic rings made an angle of 13.298. Within the
crystal, there are no short intermolecular contacts that could
reasonably be defined as structure directing (recall that the
cut-off used is 0.1 ꢂ less than the sum of the van der Waals
radii). The dimer motif (Figure 15) of the fluorophenol sees
the closest approach of a hydrogen and a fluorine from dif-
ferent rings (2.627 ꢂ), but neither the distance nor the ge-
ometry of the relationship suggests that this is very signifi-
cant. The planes of the two fluorophenol rings are separated
by 1.448 ꢂ. The structure propagates in the ac plane in the
same way as 2-5a.
Complex 2-5 f has its fluorophenol ring twisted out of the
pyridine plane by 62.768 whereas the aromatic rings in the
stilbazole subtend an angle of 15.468 between their respec-
tive planes. This structure is also lacking in significant inter-
molecular contacts that might be regarded as structure di-
recting, except for a H···F contact at 2.513 ꢂ between a
meta hydrogen of a pyridine ring and the fluorine of a
neighbouring fluorophenol unit. These propagate to give, in
effect, half of the dimer-type organisation seen here, analo-
Complexes 2-5a, 2-5d, 2-5 f, 2-4a and 2-4b: Of these com-
plexes, 2-5a and 2-4b were disordered. In the former, the di-
fluorophenyl group was disordered by rotation about the
C1–C4 axis, which was modelled in two positions with a re-
fined relative occupancy of 84:16. In the latter, the position
of the fluorine was disordered over the two possible meta
positions, with the relative occupancy being refined to 3:1.
In each case, a modified cif file was used to create structures
with the disorder removed.
¯
These complexes, which crystallised in the P1 space
group, shared comparable unit cell dimensions (Table 1),
dimer structures and crystal packing, although unit cell
angles were found as a group of three (4a, 4b, 5a) and as a
group of two (5d, 5 f). The fluorophenol rings were twisted
between 62.768 and 64.818 out of the plane of the pyridine
ring. All of the complexes adopt the same basic structure,
which is described in detail for 2-5a. The only variation of
any significance is the way in which the fluorophenol rings
pair up in a back-to-back fashion.
Complex 2–5a crystallised with the fluorophenol ring
twisted out of the plane of the pyridine ring by 65.728,
Figure 12. View of the propagation of the complexes in the ac plane. Stacks of dimer A are on the left-hand side, whereas those of dimer B are on the
right-hand side. The view eclipses the fluorophenol rings on the right.
Chem. Eur. J. 2012, 18, 16073 – 16089
ꢁ 2012 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chemeurj.org
16079

D. W. Bruce et al.
Figure 13. a) Molecular structure of complex 2-5a and b) a side view showing the back-to-back disposition of the phenolic rings.
Figure 14. a) Top view of the superposition of the pyridine rings of two antiparallel stilbazole moieties and b) propagation of complex 2-5a in the ac
plane.
gous to that in 2-5a and 2-5d. The back-to-back dimer is not
really supported in any way through the fluorophenol rings
because the closest intermolecular separation is an H···F
separation of 2.619 ꢂ (Figure 16).
F···H short contacts (2.549 ꢂ) interacting in a stepped struc-
ture identical to those described before.
Finally here, in complex 2-4b (Figure S3 in the Supporting
Information) the organisation is effectively the same as that
found for 2-5 f, with an unsupported dimer (H···F distance
ꢀ
In complex 2-4a, an N···H O angle was found that was
very nearly linear at 175.868, and the two stilbazole aromatic
rings were almost co-planar making an angle of 14.698 be-
tween the two planes. The fluorophenol ring was, however,
not co-planar with the stilbazole, being twisted 64.408 out of
the plane of the pyridine ring. The primary building block in
the crystal structure was the antiparallel, back-to-back
dimer (Figure 17) supported by two short intermolecular
of 2.614 ꢂ) and a propagating F···H
2.533 ꢂ.
ACHTUNGTREN(NGNU meta) contact of
In considering the fact that the five structures above were
very closely related, further comparison showed that in four
cases of a possible ten, the stilbazole unit from different
structures could effectively be overlaid precisely, namely 2-
4a with 2-5 f, and all combinations of 2-4b, 2-5a and 2-5d.[14]
16080
www.chemeurj.org
ꢁ 2012 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Chem. Eur. J. 2012, 18, 16073 – 16089

Hydrogen-Bonded Complexes between 4-Alkoxystilbazoles and Fluorophenols
FULL PAPER
Figure 15. Two orthogonal views of the molecular structure of complex 2-5d.
Figure 16. Molecular structure of complex 2-5 f.
Figure 17. Molecular structure of complex 2-4a.
In none of these cases did the fluorophenols overlap precise-
ly and some small deviation was seen in the overlap of the
terminal alkyl chains. An example is shown as Figure 18,
whereas the remaining pairwise combinations are found in
Figure S4 (in the Supporting Information).
Figure 18. Overlay of complexes 2-4a and 2-5 f.
Chem. Eur. J. 2012, 18, 16073 – 16089
ꢁ 2012 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chemeurj.org
16081

D. W. Bruce et al.
That the structures of these complexes are so similar, that
they are in the same space group with very similar unit cell
dimensions and to some extent angles, and that some may
be almost precisely overlaid is a point of note. Indeed, com-
bined with the fact that some structures appear to have
short intermolecular contacts, whereas others have few or
none, perhaps the question may be posed as to whether
such interactions have any great structure-directing effect,
or whether they are largely a consequence of the efficient
crystallisation of the complexes. These observations do,
however, suggest that interpretation of intermolecular con-
tacts requires both care and criticality.
The crystal packing of the 2-4c complex sees a polar, zig-
zag arrangement that propagates in the bc plane (Figure 19),
whereas in the a direction, the next such plane has an oppo-
site net polarisation, consistent with the symmetry of the
space group.
Other structural correlations: The other property that it is
instructive to correlate is the calculated pKa of the phenols
with the length of the N···H hydrogen bond and also with
ꢀ
the N···H O angle. Correlation with the latter is not con-
vincing (Figure S6a in the Supporting Information), which is
perhaps not surprising given that there is relatively little an-
gular preference in hydrogen bonding, but there is a correla-
tion with hydrogen bond length (Figure 20). In this case,
data for 2-4c are removed (because it crystallises in a differ-
ent space group) as are data for 2–6c (the unedited plot is
found as Figure S6b in the Supporting Information) to give
a straight line of correlation 0.973. Given the range of fac-
tors affecting packing in the solid state, it is remarkable that
such a good correlation is found. It is also noted that in
none of the cases is the stilbazole protonated, which is con-
sistent with the pale yellow colour observed for complex 2-8
that contains the most acidic phenol (see above).
Complex 2-4c: Complex 2-4c was unique among those stud-
ied because it crystallised in the Pbca space group and had
comparatively large unit cell dimensions. The two compo-
nents of the 2-4c complex are bound through a hydrogen
ꢀ
bond (1.828 ꢂ) with a N···H O angle of 169.598. The aro-
matic rings in the stilbazole are nearly co-planar making an
angle of 7.508, whereas the fluorophenol ring is twisted out
of the plane of the pyridine ring by 23.268 (Figure S5 in the
Supporting Information). There are no short contacts that
could be described as structure directing.
Figure 19. Ziz-zag packing of complex 2-4c in the bc plane.
16082
www.chemeurj.org
ꢁ 2012 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Chem. Eur. J. 2012, 18, 16073 – 16089

Hydrogen-Bonded Complexes between 4-Alkoxystilbazoles and Fluorophenols
FULL PAPER
Table 2. Thermal data for the butyloxystilbazole complexes.
Compound Transition[a] T [8C][b]
DH
[kJmolꢀ1
7.6
DS
[c]
G
]
[Jmolꢀ1 Kꢀ1
]
1-8
Cr–Cr1
Cr1–N
N–Iso
Cr–Cr1
Cr1–N
N–Iso
Cr–Iso
61.4
73
22
–
24
28
0.4
109
–
99
–
80
–
88.5
91.0
26.3
–
1-7
55.0–58.0
82.0
87.2
89.1
ACHTUNGTRENNUNG
7.9
10.1
0.1
39.2
–
35.8
–
30.6
–
1-6a
1-6b
1-6c
1-6d
G
R
ACHTUNGTRENNUNG
E
ACHTUNGTRENNUNG
Cr–Cr1
Cr1–Iso
–
91.5
0.3
29.5
0.9
81
–
A
E
ACHTUNGTRENNUNG
1-6e
1-5a
Cr–Cr1
Cr1–Iso
66.0
76.5
ACHTUNGTRNE(NUNG 73.5)
48.0
74.5
80.9
–
87.0–93.0
–
2
90
–
31
59
1
31.2
–
10.1
20.7
0.5
0.3
28.0
0.3
0.5
24.0
0.7
30.6
4.2
35.0
–
A
Cr–Cr1
Cr1–N
N–Iso
Cr–Cr1
Cr1–Iso
Cr–Cr1
Cr1–Cr2
Cr2–N
N–Iso
Figure 20. Plot of pKa versus N···H hydrogen bond length; the correlation
coefficient is 0.973.
1-5b
1-5c
1
77
0.8
1
66
2
85
12
101
–
–
Liquid-crystal properties: The liquid-crystal properties of
the complexes were determined by polarised optical micro-
scopy and differential scanning calorimetry (DSC), and the
data are collected in Table 2 to Table 4, whereas examples
of DSC traces are given in the Supporting Information. In
general, the complexes melted quite cleanly, which is indica-
tive of well-formed, single-component materials that are
stable in the solid state and are stable up to the melting
point of the complex. In some cases, however, the melting
range was broad and here it was not always possible to as-
certain the precise behaviour giving rise to the event. For
example, it can be the case that at a given temperature, the
two separate components of the complex represent a ther-
modynamically more stable state than the complex itself
and so decomplexation can occur to give the components,
which may or may not then recombine at some higher tem-
perature reforming the complex.[6a] In cases where the origin
of the broad melting is clear then it is described, but where
the exact origin is not apparent then the wide melting range
is simply reported as observed. However, where there is
broad melting behaviour and liquid-crystal properties are
observed, then it was found that the transition into the
liquid-crystal phase was always sharp, indicative of a melt
containing a 1:1 complex.
91.9
95.9
89.8
63.5–72.5
1-5d
1-5e
Cr–Iso
Cr–Cr1
Cr1–Iso
N
G
ACHTUNGTRENNUNG
1-5f
1-4a
82.0–84.5 (broad) 40.1
(79.7) (0.6)
113
–
2
G
E
ACHTUNGTRENNUNG
Cr–Cr1
Cr1–Cr2
Cr2–Iso
Cr–Cr1
Cr1–Iso
Cr–Iso
–
–
0.6
2.1
6
101 (very broad) 20.2
54
0.5
12
101
1-4b
1-4c
–
0.1
4.2
37.7
62.3
101.0
[a] Observations and temperatures are from optical microscopy. [b] As
far as possible, monotropic transition temperatures quoted are the true
thermodynamic temperatures obtained by reheating to isotropic from the
supercooled nematic phase and are indicated as heating transitions, for
example, (N–Iso). [c] Occasionally enthalpies for clearing transitions
could not be determined by DSC, but in most cases this was for mono-
tropic transitions.
11 of the 16 complexes of 4-butoxy-4’-stilbazole were
nematic and 4 showed enantiotropic mesomorphism. Under
the microscope, the nematic phase was observed to be very
homeotropic, but displacing the cover slip on the sample
produced a four-brush schlieren texture as shown in Fig-
ure 21a. The occurrence of the nematic phase observed in
this series is as expected for simple polar mesogens with
short terminal chains.
Figure 21. a) Schlieren nematic texture of complex 1-8 on cooling; b) the
focal conic fan texture of the SmA phase c) the mosaic texture of the
crystal smectic E phase of 2-4c.
Figure 22 is a graphical representation of the thermal data
observed by microscopy arranged in order of decreasing
Chem. Eur. J. 2012, 18, 16073 – 16089
ꢁ 2012 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chemeurj.org
16083

D. W. Bruce et al.
Table 3. Thermal data for the 4-octyloxy-4’-stilbazole complexes.
Compound
Transition
T [8C]
DH
[kJmolꢀ1
DS
G
]
[Jmolꢀ1 Kꢀ1
]
2-8
2-7
Cr–N
N–Iso
Cr–Cr1
Cr1–Cr2
Cr2–N
87.5
91.5
–
40.2
0.3
5.2
3.7
40.8
–
18.1
19.8
0.6
1.2
35.3
0.5
112
1
15
11
112
–
54
59
2
–
90.4
90.6
–
65.1
86.9
92.7
65.3
86.7
92.3
–
90.6
95.6
99.3
60.1
88.5
89.7
75.5
81.1
73.5
78.9
86.4
94.7
76.9
92.0
96.0
99.3
61.5
91.0
61.2
76.7
93.0
N–Iso
2-6a
Cr–Cr1
Cr1–SmA
SmA–N
N–Iso
Cr–SmA
SmA–N
N–Iso
Cr–Cr1
Cr–SmA
SmA–N
N–Iso
Cr - SmA
SmA - N
N–Iso
3
105
1
2-6b
2-6c
0.7
2
12.9
26.6
0.8
1.2
36.3
1.8
1.7
45.1
0.7
41.4
0.2
0.8
14.2
30.9
10.5
1.3
1.0
38.9
5.0
41.6
0.7
52.7
38
73
2
3
109
5
5
129
2
119
0.7
2.28
39
89
29
4
Figure 22. Graphical representation of the thermal behaviour of butylox-
ystilbazole complexes of the fluorophenols.
2-6d
2-6e
2-5a
Cr–N
N–Iso
clearing point for the complexes that were mesomorphic
(melting point is omitted for those showing monotropic
phases) with those that are non-mesomorphic placed at the
right-hand side. Thus, only complexes 1-5c, 1-8,
1-7 and 1-5a showed enantiotropic phases, which is in reality
a commentary on the stability of the crystal phase so that it
is observed that these complexes have the least stable crys-
talline phases. The trifluorophenol complexes were all mon-
otropic, and of these the 2,4,5-substituted fluorophenol com-
plex, 1-6c, had the highest clearing temperature at 97.58C.
The difluorophenol complexes exhibited a mixture of enan-
tiotropic, monotropic and non-mesomorphic behaviour with
complex 1-5c clearing at the highest temperature, exhibiting
an enantiotropic nematic phase between 91.9 and 95.98C.
The monosubstituted fluorophenol complexes were all non-
liquid crystalline. Trends in liquid-crystal phase stability will
be discussed below, but one immediate observation is that
for all the complexes that were liquid crystalline, highest
clearing points were found for those with a 2-fluoro substitu-
ent.
Thermal data for the complexes of octyloxystilbazole are
collected in Table 3 and the data are plotted as a function of
decreasing clearing point in Figure 23. Clearing points for
complexes of 2 are, in general, slightly higher than those of
stilbazole 1, but there are two important differences, namely
that crystal phases are destabilised by the increase in chain
length as witnessed by the fact that all but one complex
showed enantiotropic mesomorphism and that, as might
have been anticipated, SmA phases (Figure 21b) are now
seen in addition to the nematic. The order of decreasing
clearing point does not follow the same progression as in
Figure 22 for complexes of 1, which will be discussed later.
The only other significant difference is that for complex
2-4c, a monotropic crystal smectic E phase is observed. This
phase, characterised by its mosaic texture with feathered
ends (Figure 21c), is also shown by the pure ligand,[15] but
Cr–SmA
SmA–N
N–Iso
Cr - Iso
Cr–Cr1
Cr1–SmA
SmA–N
N–Iso
Cr–SmA
SmA–Iso
Cr–N
N–Iso
Cr–Iso
2-5b
2-5c
3
2-5d
2-5e
2-5f
2-4a
2-4b
115
14
124
2
144
(2)
72
37
0.4
6
(N–Iso)
(81.1)
ACHTUNGTNER(NUNG 0.7)
65.5
95.2
–
–
76.5
24.3
13.8
0.1
1.9
44.7
–
Cr–Iso
Cr–Cr1
Cr1–Cr2
Cr2–Iso
128
–
N
ACHTUNGTRENNUNG
G
G
–
–
43.2
–
–
119
E
2-4c
89.5
[a] Observations and temperatures are from optical microscopy. [b] As
far as possible, monotropic transition temperatures quoted are the true
thermodynamic temperatures obtained by reheating to isotropic from the
supercooled nematic phase and are indicated as heating transitions, for
example, (N–Iso). [c] Occasionally enthalpies for clearing transitions
could not be determined by DSC, but in most cases this was for mono-
tropic transitions. [*] Not observed by microscopy. Described in text.
the temperatures at which it is seen and also the behaviour
observed mean that it is certainly part of the mesomorphic
behaviour of the complex.
As in the previous series, the complexes with pentafluoro-
phenol and 2,3,5,6-tetrafluorophenol showed an enantio-
tropic nematic phase, this time between 87.5–91.5 and 90.4–
90.68C, respectively. On the other hand, the trifluorinated
derivatives were all enantiotropic, as the longer alkyl chain
length diminished crystal phase stability to lower the melt-
16084
www.chemeurj.org
ꢁ 2012 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Chem. Eur. J. 2012, 18, 16073 – 16089

Hydrogen-Bonded Complexes between 4-Alkoxystilbazoles and Fluorophenols
FULL PAPER
Table 4. Thermal data for the 4-dodecyloxy-4’-stilbazole complexes.
Compound
Transition
T [8C]
DH
[kJmolꢀ1
DS
G
]
[Jmolꢀ1 Kꢀ1
]
3-8
Cr–Iso
A
97.8 (broad)
50.9
–
138
–
ACHTUNGTRENNUNG
A
E
–
–
3-7
95.5
ACHTUNGTRNE(NUNG 92.9)
48.1
–
131
–
N
3-6a
3-6b
3-6c
Cr–SmA
SmA–Iso
Cr–SmA
SmA–Iso
Cr–Cr1
78.7
98.7
79.0
98.5
–
53.9
5.0
55.5
5.1
1.2
0.4
31.3
6.2
0.5
42.7
5.2
63.9
–
153
13
158
14
4
Cr1–Cr2
–
1
[
]
*
87.0–89.5
103.5
–
83.1
96.5
86.0
87
16
1
120
14
179
–
SmA–Iso
Cr–Cr1
Cr1–SmA
SmA–Iso
Cr–Iso
3-6d
3-6e
3-5a
3-5b
G
ACHTUNGTRENNUNG
G
A
–
–
Cr–Cr1
–
0.2
58.8
5.5
25.2
27.9
9.7
7.3
9.9
49.5
6.0
55.4
7.0
55.1
3.3
75.0
–
0.6
164
15
74
80
27
20
27
137
16
157
19
162
10
202
–
Cr1–SmA
SmA–Iso
Cr–Cr1
Cr1–Cr2
Cr2–SmA
SmA–Iso
85.8
92.2
–
–
92.0
98.8
ACHTUNGTRNE(NUNG 93.2)
90.8
103.0
79.8
98.5
Figure 23. Graphical representation of the thermal behaviour of octylox-
ystilbazole complexes of the fluorophenols.
ing point. The difluorophenol derivatives were still a mix-
ture of enantiotropic, monotropic and non-mesomorphic
complexes whereas the mono-fluorinated derivatives re-
mained non-mesomorphic with the exception of 2-4b, which
exhibited monotropic E, SmA and N phases.
A
3-5c
3-5d
3-5e
3-5f
Cr–SmA
SmA–Iso
Cr–SmA
SmA–Iso
Cr–SmA
SmA–Iso
Cr–Iso
Thermal data for complexes of dodecyloxystilbazole are
collected in Table 4 and are plotted in Figure 24, once more
in order of descending clearing point. Again as might be
predicted on increasing the terminal chain length further,
SmA phases were promoted strongly so that now only three
complexes show a nematic phase. Indeed, the strong stabili-
65.0–67.4
77.9
100.0
C
ACHTUNGTRENNUNG
G
G
–
–
3-4a
Cr–E
81.0
94.5
99.3
88.0
48.7
9.3
9.0
70.1
–
137
25
24
194
–
E–SmA
SmA–Iso
Cr–Iso
3-4b
3-4c
T
94.3
64.5
–
176
–
A
[a] Observations and temperatures are from optical microscopy. [b] As
far as possible, monotropic transition temperatures quoted are the true
thermodynamic temperatures obtained by reheating to isotropic from the
supercooled nematic phase and are indicated as heating transitions, for
example, (N–Iso). [c] Occasionally enthalpies for clearing transitions
could not be determined by DSC, but in most cases this was for mono-
tropic transitions. [*] Not observed by microscopy. Described in text.
sation of the SmA phase combined with some destabilisa-
tion of the crystal phase means that none of the complexes
derived from stilbazole 3 was monotropic. Once more, a
crystal smectic E phase was observed, this time for com-
plexes 3-5b and 3-4a and its monotropic nature in the
former does not allow it easily to be displayed on the graph.
In both complexes, the fact that the E phase melts into the
SmA phase above the temperature at which the crystal E
phase of the pure ligand forms an isotropic liquid, once
more demonstrates that the phase is a property of the indi-
vidual complexes.
Figure 24. Graphical representation of the thermal behaviour of dodecy-
loxystilbazole complexes of the fluorophenols.
Chem. Eur. J. 2012, 18, 16073 – 16089
ꢁ 2012 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chemeurj.org
16085

D. W. Bruce et al.
The behaviour of complex 3-6c warrants further com-
ments, so that in the temperature range 87.0 to 89.58C, it
was observed to melt in two stages before forming a SmA
phase. Optical microscopy supported the view that at 878C,
the hydrogen bond of the solid complex broke to give the
liquid phenol plus the solid stilbazole and then, as the tem-
perature was raised, the stilbazole itself melted. With both
components in the fluid phase, the hydrogen bond then re-
formed and the complex shows a SmA phase. This is ex-
plained as up to 878C, the crystalline phase of the hydro-
gen-bonded complex was the stable state. Above this tem-
perature, it was preferable thermodynamically for the two
components to exist as the solid form of the stilbazole and
the free phenol until the stilbazole melted, at which point
the two components hydrogen bonded once more, revealing
the mesophase.
C4 stilbazole: 1-6c>1-5c>1-8>1-6a>1-7>1-6b>1-6d>
1-5a
C8 stilbazole: 2-6c>2-5c>2-6a>2-6b>2-8>2-5d>
2-7>2-6d>2-5a
C12 stilbazole: 3-6c>3-5c>3-4a>3-5b>3-6a>3-5d>
3-6b>3-6d>3-8>3-7>3-5a
Thus, although in series 2 there is some change in the
order with respect to 2-8 and 2-7, inspection of the graph
shows that all complexes from 2-6a to 2-6d have very simi-
lar clearing points and so the change is not regarded as sig-
nificant. Similar arguments hold for 3-4a to 3-6b, although
in series 3 the clearing point of complexes 3-8 and 3-7 are
appreciably lower.
However, what is perhaps more revealing is to plot the
data as a function of decreasing clearing point but also ac-
cording to the number of fluorine atoms attached to the flu-
orophenol. This is shown in Figure 26 for complexes of
What is interesting here is that Figure 24 shows a very dif-
ferent order of complexes along the x axis, which will form
part of the discussion below.
Discussion
Generally, the hydrogen-bonded complexes prepared were
observed to clear at increasing temperatures with increasing
alkoxy chain length. The 4-butoxy-4’-stilbazole complexes
were mostly nematogenic, the 4-octyloxy-4’-stilbazole com-
plexes exhibited both the nematic and smectic A phase,
whereas the majority of the 4-dodecyloxy-4’-stilbazole com-
plexes displayed only the smectic A phase. This evolution in
thermal behaviour as a function of alkoxy chain length for
the complexes of 5c is shown in Figure 25.
Figure 26. Clearing point plotted in descending order according to the
number of fluorine atoms on the fluorophenol for complexes of series 3.
series 3 and as Figures S7 and S8 (in the Supporting Infor-
mation) for complexes of series 1 and 2, respectively. What
is remarkable is that in each series, the order is absolutely
identical (obviously only complexes with mesophases will fit
such an analysis).
Figure 25. Evolution of the thermal behaviour.
However, more significant is to try to understand how the
clearing point varies with the nature of the fluorophenol.
Thus, Figures 22 to 24 show clearing points in descending
order and, close examination of the data reveal a very good
degree of conformity between the three and changes in
order are in most cases minimal. For example, clearing
points decrease as follows (complexes that appear in one
line but not the one above are in italics):
So how can this behaviour be understood? Figure 27
shows the different complexes organised by number of fluo-
rine atoms and clearing point and allows a more insightful
structural comparison. What is clear is that for phenols 5
and 6, the highest clearing points are always found where
there is a fluorine atom in the 2-position and, as indicated in
the figure, this is interpreted as arising from the formation
of an additional, intramolecular hydrogen bond between
16086
www.chemeurj.org
ꢁ 2012 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Chem. Eur. J. 2012, 18, 16073 – 16089

Hydrogen-Bonded Complexes between 4-Alkoxystilbazoles and Fluorophenols
FULL PAPER
Figure 27. Hydrogen-bonded complexes arranged in order of decreasing clearing point and by number of fluorine atoms, with phenols 7 and 8 omitted.
this 2-fluorine and the ortho-hydrogen of the pyridine ring
in the stilbazole. That such a motif is a reasonable proposi-
tion is supported by its observation in the crystal structures
of complexes 2-8 and 2-7 in which the intramolecular F···H
distance is 2.234 and 2.36 ꢂ, respectively. The presence of
such an interaction would give a more rigid structure to the
mesogen when compared to the much more flexible ar-
rangement that would be expected through a single point of
contact, which is consistent with the higher clearing points.
Beyond this, the mesophases are clearly most stabilised
when the phenol contains a 3-fluorophenol or 4-fluorophe-
nol motif, both of which provide a terminal dipole that will
enhance the dielectric anisotropy.1
clearing points, whereas the second group consists of com-
plexes in which the phenol does not have a 2-fluoro sub-
stituent and where clearing points are lower for the reasons
explained earlier. For clarity, these two regions are separat-
ed in the figures by a dotted line. However, within or be-
tween these groups there is no dependence of the clearing
point on pKa.2 Thus, in common with our earlier finding and
assertions,[5a,b] we see no evidence whatsoever that clearing
of these complexes is driven by anything other than the nat-
ural clearing behaviour of the hydrogen-bonded complex.
Conclusion
Finally, it is appropriate to determine if there is any corre-
lation between the pKa of the phenol and the clearing point,
as from time to time in the literature it has been speculated
that the clearing of a liquid crystal phase has been driven by
the rupture of the hydrogen bond. Figure 28 shows the cal-
culated pKa values for the different fluorophenols plotted
against the clearing point for complexes of series 2 and 3;
the analogous plot for series 1 complexes is found as Fig-
ure S8 (in the Supporting Information). Analysis of the data
reveals that they can be considered in two groups. The first
of these consists of phenols with a 2-fluoro substituent
which, as noted above, tend to give complexes with higher
This exhaustive and systematic study of the structure and
mesomorphism of fluorophenol complexes of alkoxystilba-
zoles has provided a good deal of information about the be-
haviour of these new, hydrogen-bonded species. By virtue of
the fact that all but one of the complexes of series 2 that
¯
were crystallised did so in the P1 space group, meaningful
comparisons could be made. Central to the results of this
study was the proposition that in many cases, what appeared
to be real, intermolecular interactions were unlikely to be of
any real significance in determining crystal packing and,
consequently, care is suggested in interpreting such interac-
tions in future studies.
1
Although it is, in general, much less reliable to seek trends in melting
points, we did look at them for complexes 2-5a, 2-5d, 2-5 f, 2-4a and 2-
4b (all of which crystallise in P1) and found no correlation with the
2
¯
The data for series 1 (Figure S8 in the Supporting Information) show a
planarity of the complex induced by the presence of a fluorine in the 2-
position.
less clear separation between the two regions, but the form of the plot
is otherwise identical.
Chem. Eur. J. 2012, 18, 16073 – 16089
ꢁ 2012 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chemeurj.org
16087

D. W. Bruce et al.
rafluorophenol (7), a very similar dependence of clearing
point on phenol was observed in all series and, more strik-
ingly, exactly the same dependence of clearing point on fluo-
rine substitution pattern for a given number of fluorine
atoms was found for each of the three stilbazole chain
lengths studied. In particular, it was noted that mesophases
were most stable where the phenol had a 2-fluorine substitu-
ent, which was interpreted in terms of the added stability
provided by the formation of an intramolecular six-mem-
bered ring involving the phenol and the stilbazole, possibly
due to this pattern of substitution.
Finally, it was noted that there was no dependence of the
clearing point on the pKa of the phenol and, by extension,
the strength of the N···H hydrogen bond. This offers strong
support to previous assertions from this group[5a,b] that clear-
ing in hydrogen-bonded mesogens is of the hydrogen-
bonded complex and is not driven by hydrogen-bond rup-
ture.
Experimental Section
All the fluorophenols were used as received from Fluorochem and alkox-
ystilbazoles were prepared as described previously.[15] 1H NMR spectra
were recorded at 270 MHz using a JEOL ECX270 instrument, whereas
optical microscopy employed an Olympus BX50 Optical Microscope
equipped with a Link-Am HFS91 hot stage, TMS92 controller and LNP2
cooling unit. DSC data were collected using a Mettler Toledo DSC 822,
equipped with a TSO801R0 sample robot and calibrated using indium;
samples were run at heating and cooling rates of 108C minꢀ1. Elemental
analysis was carried out on an Exeter Analytical Inc CE 440 Elemental
Analyser and a Sartorius SE2 analytical balance at the University of
York.
Diffraction data were collected at 110 K on a Bruker Smart Apex diffrac-
tometer with MoKa radiation (l=0.71073 ꢂ) using
a SMART CCD
camera. Diffractometer control, data collection and initial unit cell deter-
mination was performed using “SMART”.[16] The crystal was cooled with
an Oxford Crysostream. Frame integration and unit-cell refinement was
carried out with “SAINT+”.[17]Absorption corrections were applied by
SADABS.[18] Structures were solved by “direct methods” using SHELXS-
97 (Sheldrick, 1997)[19] and refined by full-matrix least squares using
SHELXL-97 (Sheldrick, 1997).[20] All non-hydrogen atoms were refined
ꢀ
anisotropically. O H hydrogen atoms were located by difference map
after all other atoms had been located and refined. All other hydrogen
atoms were placed using a “riding model” and included in the refinement
at calculated positions. Three of the structures exhibited disorder, which
was modelled as follows:
Figure 28. Plot of clearing point as a function of pKa for complexes of
(left) octyloxystilbazole (2) and (right) dodecyloxystilbazole (3).
2-8: Both of the aromatic rings and the double bond of the stilbene ex-
hibited disorder and were modelled in two positions, the relative occu-
pancy of which refined to 0.77:0.23. For the minor component, the aro-
matic rings were constrained to be regular hexagons with a bond length
of 1.39 ꢂ. The anisotropic displacement parameters of corresponding
pairs of disordered atoms for example, C7a and C7b were constrained to
be equal.
In terms of liquid-crystal properties, in general terms the
overall pattern of mesomorphism in relation to stilbazole
chain length was as might have been expected for such
simple, dipolar mesogens. Thus, nematic phases were fav-
oured at short chain lengths, whereas the SmA phase
became stabilised when the terminal alkoxy chain was much
longer. More interesting was the dependence of mesophase
stability (N or SmA) on the fluorophenol. With small excep-
tions for complexes of pentafluorophenol (8) and 2,3,5,6-tet-
2-5a: The difluorophenyl group was disordered by rotation about the
Cipso–Cpara axis and was modelled in two positions with a refined relative
occupancy of 84:16.
2-4b: The position of the fluorine was disordered over the two possible
meta positions. The relative occupancy of the positions was refined to
3:1.
16088
www.chemeurj.org
ꢁ 2012 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Chem. Eur. J. 2012, 18, 16073 – 16089

Hydrogen-Bonded Complexes between 4-Alkoxystilbazoles and Fluorophenols
FULL PAPER
[8] H. L. Nguyen, P. N. Horton, M. B. Hursthouse, A. C. Legon, D. W.
Bruce, J. Am. Chem. Soc. 2004, 126, 16.
Acknowledgements
[9] a) D. W. Bruce, P. Metrangolo, F. Meyer, T. Pilati, C. Prꢆsang, G.
Resnati, G. Terraneo, S. G. Wainwright, A. C. Whitwood, Chem.
Eur. J. 2010, 16, 9511; b) C. Prꢆsang, H. L. Nguyen, P. N. Horton,
A. C. Whitwood, D. W. Bruce, Chem. Commun. 2008, 6164; c) C.
Prꢆsang, A. C. Whitwood, D. W. Bruce, Chem. Commun. 2008, 2137;
d) D. W. Bruce, P. Metrangolo, F. Meyer, C. Prꢆsang, G. Resnati,
A. C. Whitwood, New J. Chem. 2008, 32, 477; e) P. Metrangolo, C.
Prꢆsang, G. Resnati, R. Liantonio, A. C. Whitwood, D. W. Bruce,
Chem. Commun. 2006, 3290.
[10] a) J. Xu, X. Liu, J. Kok-Peng Ng, T. Lin, C. He, J. Mater. Chem.
2006, 16, 3540; b) J. Xu, X. Liu, T. Lin, J. Huang, C. He, Macromole-
cules 2005, 38, 3554.
[11] C. Prꢆsang, A. C. Whitwood, D. W. Bruce, Cryst. Growth Des. 2009,
9, 5319.
[12] J. Han, F. M. Tao, J. Phys. Chem. A 2006, 110, 257.
[13] a) M. J. G. Lesley, A. Woodward, N. J. Taylor, T. B. Marder, I. Caze-
nobe, I. Ledoux, J. Zyss, A. Thornton, D. W. Bruce, A. K. Kakkar,
Chem. Mater. 1998, 10, 1355; b) D. W. Bruce, R. G. Denning, M. G.
Grayson, R. Le Lagadec, K. K. Lai, B. T. Pickup, A. Thornton, Adv.
Mater. Opt. Electron. 1994, 4, 293; c) D. W. Bruce, A. Thornton,
Mol. Cryst., Liq. Cryst. 1993, 231, 253.
[14] Overlaying was achieved using the application QtMG available free
from: http://www.ccp4.ac.uk/MG.
[15] D. W. Bruce, D. A. Dunmur, E. Lalinde, P. M. Maitlis, P. Styring,
Liq. Cryst. 1988, 3, 385.
J.P.-W.W. thanks the University of York, Department of Chemistryꢃs
Wild Fund for financial assistance.
[1] D. W. Bruce in: Supramolecular Chemistry: From Molecules to
Nanomaterials (Eds.: J. W. Steed and P. A. Gale), Wiley, Chichester,
2012, pp. 3493–3514.
[2] For a recent review, see: A. C. Legon, Phys. Chem. Chem. Phys.
2010, 12, 7736.
[3] a) P. Metrangolo, H. Neukirch, T. Pilati, G. Resnati, Acc. Chem. Res.
2005, 38, 386; b) P. Metrangolo, G. Resnati, Science 2008, 321, 918.
[4] a) T. Kato, J. M. J. Frꢀchet, J. Am. Chem. Soc. 1989, 111, 8533; b) T.
Kato, J. M. J. Frꢀchet, Macromolecules 1989, 22, 3818; c) U. Kumar,
T. Kato, J. M. J. Frꢀchet, J. Am. Chem. Soc. 1992, 114, 6630; d) T.
Kato, H. Kihara, T. Uryu, A. Fuijishima, J. M. J. Frꢀchet, Macromo-
lecules 1992, 25, 6836; e) T. Kato, H. Kihara, U. Kumar, T. Uryu,
J. M. J. Frꢀchet, Angew. Chem. 1994, 106, 1728; Angew. Chem. Int.
Ed. Eng. 1994, 33, 1644; f) T. Kato, P. G. Wilson, A. Fuijishima,
J. M. J. Frꢀchet, Chem. Lett. 1990, 2003; g) T. Kato, J. M. J. Frꢀchet,
P. G. Wilson, T. Saito, T. Uryu, A. Fuijishima, C. Sin, F. Kaneuch,
Chem. Mater. 1993, 5, 1094; h) T. Kato, A. Fuijishima, J. M. J. Frꢀ-
chet, Chem. Lett. 1990, 912; i) M. Fukumasa, T. Kato, T. Uryu,
J. M. J. Frꢀchet, Chem. Lett. 1993, 65; j) T. Kato, H. Kihara, T. Uryu,
S. Ujiie, K. Iimura, J. M. J. Frꢀchet, U. Kumar, Ferroelectrics 1994,
148, 1303.
[16] “SMART”: control software Bruker SMART Apex X-ray Diffrac-
tometer. v5.625, Bruker-AXS GMBH, Karlsruhe, Germany.
[17] “SAINT+”: integration software for Bruker SMART detectors.
v6.45, Bruker-AXS GMBH, Karlsruhe, Germany.
[18] “SADABS”: program for absorption correction. v2.10. G. M. Shel-
drick, Bruker AXS Inc., Madison, Wisconsin, USA, 2007.
[19] SHELXS-97: program for structure solution. G. M. Sheldrick, Uni-
versity of Gçttingen, Gçttingen, Germany, 1997.
[5] a) D. J. Price, K. Willis, T. Richardson, G. Ungar, D. W. Bruce, J.
Mater. Chem. 1997, 7, 883; b) D. J. Price, T. Richardson, D. W.
Bruce, J. Chem. Soc. Chem. Commun. 1995, 1911; c) D. W. Bruce,
D. J. Price, Adv. Mater. Opt. Electron. 1994, 4, 273.
[6] a) D. J. Price, H. Adams, D. W. Bruce, Mol. Cryst., Liq. Cryst. 1996,
289, 127; b) K. Willis, J. E. Luckhurst, D. J. Price, J. M. J. Frꢀchet, T.
Kato, G. Ungar, D. W. Bruce, Liq. Cryst. 1996, 21, 585; c) B. Friot,
D. Boyd, K. Willis, B. Donnio, G. Ungar, D. W. Bruce, Liq. Cryst.
2000, 27, 605.
[7] For example, see: a) M. Suꢄrez, J. M. Lehn, S. C. Zimmerman, A.
Skoulios, B. Heinrich, J. Am. Chem. Soc. 1998, 120, 9526; b) D.
Goldmann, D. Janietz, C. Schmidt, J. H. Wendorff, J. Mater. Chem.
2004, 14, 1521; c) S. Coco, C. Cordovilla, C. Domꢅnguez, B. Donnio,
P. Espinet, D. Guillon, Chem. Mater. 2009, 21, 3282.
[20] SHELXL-97: program for the Refinement of Crystal Structures.
G. M. Sheldrick, University of Gçttingen, Gçttingen, Germany,
1997.
Received: August 2, 2012
Published online: October 19, 2012
Chem. Eur. J. 2012, 18, 16073 – 16089
ꢁ 2012 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chemeurj.org
16089