Synthesis of dinuclear complexes of rhenium(I) as potential metallomesogens

Article abstract of DOI:10.1039/a910273f

Seven anisotropic diimines and their corresponding dinuclear complexes of rhenium(i) have been synthesized. The crystal and molecular structures of one of the dimines and two of the complexes were determined. In common with related inline ligands which have been synthesized previously, the new diimines were mesomorphic showing smectic C and nematic phases and, in some cases, smectic I and crystal J phases. However, none of the orthometallated, dinuclear complexes showed any liquid crystal phases. The Royal Society of Chemistry 2000.

Full text of DOI:10.1039/a910273f

View Article Online / Journal Homepage / Table of Contents for this issue
Synthesis of dinuclear complexes of rhenium(I) as potential
metallomesogens
Marie-Andrée Guillevic,a Mark E. Light,b Simon J. Coles,b Thomas Gelbrich,b
Michael B. Hursthouseb and Duncan W. Bruce*a
a School of Chemistry, University of Exeter, Stocker Road, Exeter, UK EX4 4QD.
E-mail: d.bruce@exeter.ac.uk
b EPSRC Crystallographic Service, Department of Chemistry, University of Southampton,
Highfield Road, Southampton, UK SO17 1BJ
Received 21st December 1999, Accepted 17th March 2000
Published on the Web 10th April 2000
Seven anisotropic diimines and their corresponding dinuclear complexes of rhenium() have been synthesized. The
crystal and molecular structures of one of the dimines and two of the complexes were determined. In common with
related imine ligands which have been synthesized previously, the new diimines were mesomorphic showing smectic
C and nematic phases and, in some cases, smectic I and crystal J phases. However, none of the orthometallated,
dinuclear complexes showed any liquid crystal phases.
In the synthesis of materials of any kind1 a major motivation is
to be able to exert control over the properties in a predictable
way, ideally via the initial synthesis. In the general field of liquid
crystals this is quite a well developed art and so it is possible
to influence physical properties such as birefringence and
dielectric anisotropy.
The advent of widespread interest in metallomesogens in
the mid-1980s provided an extra challenge in this regard as the
co-ordination of suitable ligands to the metal centre provided
an extra perturbation which needed to be understood. Metal
centres introduce extra parameters into a liquid-crystalline
system and it is necessary to worry about the effects of, for
example, co-ordination geometry, electronic configuration and
reactivity.2
Fig. 1 Structure of some manganese and rhenium imine complexes
studied previously.
synthesis of some ligands with two, isolated imine functions
which would allow the co-ordination of two Re(CO)4 groups.
The results of this study are now reported.
As part of our studies in this area we have looked at the
problem of incorporating six-co-ordinate metal centres into
rod-like (calamitic) liquid crystals. Briefly stated, the potential
problem is that liquid crystal mesophase formation requires
molecules with a rather high structural anisotropy which in
turn leads to anisotropic dispersion forces which stabilise the
mesophase. Tethering rod-like ligands to a planar or linear
metal centre (d8–d10) is consistent with this requirement, but
an octahedral centre can reduce the anisotropy and suppress
mesophase formation.3
We succeeded in the realisation of calamitic liquid crystals
containing six-co-ordinate metal centres initially using ortho-
metallated imine complexes of MnI 4,5 (Fig. 1) although we now
study their rhenium() congeners for choice due to their higher
thermal stability.6,7 We then reported examples of other calam-
itic systems using diazabutadienes8 and 2,2Ј-bipyridines as
ligands to ReI.9 One of the interesting features which a com-
parison of these series of materials revealed was that in the case
of the imine complexes the liquid crystal transition temper-
atures of the ligand were reduced on complexation to the metal
fragment (Re(CO)4), while for the main part they were main-
tained or increased for complexation of the metal fragment
(ReBr(CO)3) to the diazabutadiene and bipyridine systems. This
being the case, would it then also be true that if we were to
introduce more than one Re(CO)4 fragment into the imine
systems the transition temperatures would reduce even further?
In order to attempt to answer this question we undertook the
Results and discussion
Synthesis of the ligands and complexes
The synthetic routes are outlined in Scheme 1 and we note that
related, unsubstituted ‘ligands’ have been reported previously
by Miyajima and Matsunaga.10 Alkoxybenzaldehydes (1–3)
were obtained via a Williamson ether reaction between the
relevant 1-bromoalkane and 4-hydroxybenzaldehyde, while
alkoxyanilines (4–6) were obtained by a reaction between a
1-bromoalkane and 4-acetamidophenol to give 4-alkoxyphenyl-
acetamide, followed by deprotection to give the amine. The
imines 7–12 were then obtained in yields of 70–80% by reaction
of the benzaldehyde with a relevant amine in a Schiff base
condensation and two target, symmetric diimines (15 and 16)
were then produced in yields of around 80% by esterification
of the carboxylic acids 7, 8 with the phenols 10, 11 using
DCC/DMAP
[dicyclohexylcarbodiimide/4(N,N-dimethyl-
amino)pyridine]. For reasons which we cannot fathom, diimine
17 could not be obtained in this way as the esterification failed
and so it was obtained in 55% yield via direct reaction with the
diamine 13 which would represent a preferred, more convergent
future synthesis for symmetric diimines of this type. Finally,
three isomeric diimines, 18–20, were obtained in 80–100%
yields by reaction of the alkoxyanilines 4–6 with the dialdehyde
14. In addition, we synthesized one five-ring diimine, 21. This
was obtained by esterification of octyloxybenzoic acid with
4-hydroxybenzaldehyde to give 4-(4Ј-octyloxybenzoyloxy)benz-
DOI: 10.1039/a910273f
J. Chem. Soc., Dalton Trans., 2000, 1437–1445
This journal is © The Royal Society of Chemistry 2000
1437

View Article Online
Scheme 1 Synthesis of the ligands and complexes. (i) EtOH/Hϩ, (ii) CH2Cl2/DCC/DMAP, (iii) [ReMe(CO)5]/toluene/reflux.
aldehyde, two equivalents of which were then condensed with
1,4-phenylenediamine to give the target diimine.
The related metal complexes (22–28) were finally obtained by
reaction of the diimines with [ReMe(CO)5] in toluene at reflux
for 20 h.
All of the ligands and their complexes were characterised by
1H and 13C NMR spectroscopy and, for the complexes, infrared
spectroscopy which showed νCO at 2090, 1988 and 1922 cmϪ1 11
It is worthy of note that in the course of this work we tried a
variety of methods to obtain the related monometallated
.
1438
J. Chem. Soc., Dalton Trans., 2000, 1437–1445

View Article Online
Table 1 Crystallographic data for compounds 15, 22 and 26
15
22
26
Formula
M
Crystal system
Space group
C35H36N2O4
548.66
C43H34N2O12Re2
1143.12
C51H50N2O12Re2
1255.33
Triclinic
P1
Triclinic
Triclinic
¯
¯
P1
P1
a/Å
b/Å
c/Å
α/Њ
7.2025(6)
10.1729(13)
20.231(3)
83.619(5)
87.892(8)
83.518(8)
1463.3(3)
150(2)
8.5056(2)
11.9274(3)
21.6445(8)
82.0034(19)
82.2317(18)
71.9589(18)
2057.66(10)
150(2)
9.1822(2)
13.527(3)
21.470(4)
74.14(3)
86.93(3)
74.32(3)
2469.1(9)
150(2)
β/Њ
γ/Њ
U/Å3
T/K
Z
2
2
2
µ/mmϪ1
0.081
5.943
4.961
No. reflections measured
No. unique reflections
R1 [I > 2σ(I)]
wR2[I > 2σ(I)]
2.0691
5576
0.0782
0.2201
28995
7333
0.0367
0.0898
46405
25880
0.0372
0.0895
Fig. 2 Molecular structure of ligand 15.
derivatives of ligands 15–20 by using half-stoichiometric
amounts of [ReMe(CO)5] and by varying reaction conditions.
However, while we obtained spectroscopic evidence that the
complex formed, it proved impossible to isolate a pure sample.
joined by the ester group which approaches the stereo-
chemically favoured 90Њ. The angles between the 1st and 2nd and
3rd and 4th rings across the imine double bond are rather similar
to each other.
The molecules within the crystal stack parallel along the
a direction and antiparallel along the b direction. The terminal
chain contains a gauche bond at one end while all those at the
other are trans, which may be due to packing effects optimising
the possibility for the interdigitation of alternate chains.
Crystal and molecular structure of compounds 15, 22 and 26
In order to try to obtain as much information as possible on
these new materials, single crystal structure determinations
were carried out on ligand 15 and complexes 22 and 26 for
which we were able to grow suitable crystals.
Complex 22. Crystals suitable for diffraction studies were
grown from dichloromethane/hexane. The molecules crystal-
¯
Ligand 15. Crystals suitable for diffraction studies were
lised in a triclinic lattice in space group P1 and after refinement
grown from dichloromethane/hexane. The molecules crystal-
lised in a triclinic lattice in space group P1 and after refinement
an R factor of 3.7% was achieved. The chain at one end of the
molecule was modelled as disordered over two conformations.
The molecular structure is shown in Fig. 3.
¯
an R factor of just under 8% was achieved. Crystallographic
parameters are collected in Table 1, selected bond lengths,
angles and least squares mean plane torsion angles in Table 2
and the molecular structure is shown in Fig. 2.
The angles between the least squares planes of the four rings
are not restricted by any chemical constraints and are thus a
feature of the packing arrangement. C5 > C10–C12 > C17
39.81(4), C12 > C17–C19 > C24 76.79(10) and C19 > C24–
C26 > C31 38.68(13)Њ. The largest angle is between the rings
The co-ordination of the metal tetracarbonyl group has
increased the angle between pairs of adjacent rings. The angles
between the least square planes of the four rings are C5 > C10–
C16 > C21 47.93(21), C16 > C21–C23 > C28 84.24(17) and
C23 > C28–C34 > C39 42.55(21)Њ. This can simply be attrib-
uted to the steric demands of the metal carbonyl functions (as
shown in the space-filling model in Fig. 4) which twist the ester-
bridged rings further apart, each nestled between two carbonyl
J. Chem. Soc., Dalton Trans., 2000, 1437–1445
1439

View Article Online
Table 2 Selected bond lengths (Å) and angles (Њ) and least squares mean plane torsion angles (Њ) for compounds 15, 22 and 26
Complex 26
Complex 22
Molecule 1
Molecule 2
Re(1)–C(14)
Re(1)–C(11)
Re(1)–C(13)
Re(1)–C(12)
Re(2)–C(30)
Re(2)–C(29)
Re(2)–C(31)
Re(2)–C(32)
1.924(7)
1.963(7)
1.980(6)
2.000(6)
1.921(7)
1.973(7)
1.980(7)
1.981(7)
Re(1)–C(17)
Re(1)–C(15)
Re(1)–C(16)
Re(1)–C(18)
Re(2)–C(35)
Re(2)–C(36)
Re(2)–C(37)
Re(2)–C(34)
1.92(3)
1.98(2)
1.98(2)
1.99(3)
1.93(2)
1.95(2)
1.97(3)
2.00(3)
Re(3)–C(69)
Re(3)–C(70)
Re(3)–C(68)
Re(3)–C(67)
Re(4)–C(87)
Re(4)–C(89)
Re(4)–C(88)
Re(4)–C(86)
1.92(2)
1.96(3)
1.98(3)
2.02(3)
1.90(2)
1.97(2)
1.97(2)
1.99(2)
C(14)–Re(1)–C(11)
C(14)–Re(1)–C(13)
C(11)–Re(1)–C(13)
C(14)–Re(1)–C(12)
C(11)–Re(1)–C(12)
C(30)–Re(2)–C(29)
C(30)–Re(2)–C(31)
C(29)–Re(2)–C(31)
C(30)–Re(2)–C(32)
C(29)–Re(2)–C(32)
91.8(3)
89.1(3)
93.2(2)
88.6(3)
93.2(2)
92.3(3)
87.8(3)
93.8(3)
89.9(3)
93.1(3)
C(17)–Re(1)–C(15)
C(17)–Re(1)–C(16)
C(15)–Re(1)–C(16)
C(17)–Re(1)–C(18)
C(16)–Re(1)–C(18)
C(35)–Re(2)–C(36)
C(35)–Re(2)–C(37)
C(36)–Re(2)–C(37)
C(35)–Re(2)–C(34)
C(36)–Re(2)–C(34)
89.2(11)
92.6(10)
93.1(9)
86.5(11)
90.5(9)
93.6(9)
90.3(9)
90.7(10)
85.3(9)
92.3(10)
C(69)–Re(3)–C(70)
C(69)–Re(3)–C(68)
C(70)–Re(3)–C(68)
C(69)–Re(3)–C(67)
C(68)–Re(3)–C(67)
C(87)–Re(4)–C(89)
C(87)–Re(4)–C(88)
C(89)–Re(4)–C(88)
C(87)–Re(4)–C(86)
C(89)–Re(4)–C(86)
85.9(9)
95.0(10)
91.5(11)
87.7(9)
90.2(10)
93.2(9)
89.2(10)
93.4(9)
88.6(10)
90.1(8)
O(2)–C(11)–Re(1)
O(3)–C(12)–Re(1)
O(4)–C(13)–Re(1)
O(5)–C(14)–Re(1)
O(8)–C(29)–Re(2)
O(9)–C(30)–Re(2)
O(10)–C(31)–Re(2)
O(11)–C(32)–Re(2)
176.6(5)
178.5(5)
175.0(6)
177.8(7)
177.4(5)
176.7(6)
176.8(6)
175.2(7)
O(2)–C(15)–Re(1)
O(3)–C(16)–Re(1)
O(4)–C(17)–Re(1)
O(5)–C(18)–Re(1)
O(8)–C(34)–Re(2)
O(9)–C(35)–Re(2)
O(10)–C(36)–Re(2)
O(11)–C(37)–Re(2)
177.9(16)
176.8(19)
176(2)
178(2)
174.9(18)
175(2)
O(14)–C(67)–Re(3)
O(15)–C(68)–Re(3)
O(16)–C(69)–Re(3)
O(17)–C(70)–Re(3)
O(20)–C(86)–Re(4)
O(21)–C(87)–Re(4)
O(22)–C(88)–Re(4)
O(23)–C(89)–Re(4)
176.5(18)
178(2)
174(2)
174.0(17)
175.2(17)
178.8(19)
176(2)
179.2(17)
175.8(16)
176.9(16)
(C5 > C10)–(C16 > C21)
C16 > C21)–(C23 > C28)
C23 > C28)–(C34 > C39)
47.93(21)
84.24(17)
42.55(21)
(C9 > C14)–(C20 > C25)
(C20 > C25)–(C27 > C32)
(C27 > C32)–(C38 > C66)
35.21(3)
26.25(2)
51.11(6)
(C90 > C95)–(C79 > C84)
(C79 > C84)–(C72 > C77)
(C72 > C77)–(C61 > C66)
35.73(4)
24.47(3)
48.85(5)
Ligand 15
(C5 > C10)–(C12 > C17)
39.81(4)
(C12 > C17)–(C19 > C24)
76.79(10)
(C19 > C24)–(C26 > C31)
38.68(13)
Fig. 3 Molecular structure of complex 22.
groups on opposing sides of the metal–metal axis. In com-
parison with the rhenium tetracarbonyl complex of benzyl-
ideneaniline,12 the angle between the two rings is smaller due to
the packing arrangement adopted. The molecules pack in
essentially the same way as the parent ligand (15), namely paral-
lel along a and antiparallel along b. While the co-ordination of
both metal centres on the same side of the molecule would seem
energetically unfavourable, the molecules pack back-to-back
such that the metal tetracarbonyl groups interlock, thus form-
ing a stable packing arrangment (Fig. 5).
1440
J. Chem. Soc., Dalton Trans., 2000, 1437–1445

View Article Online
The Re–C bond lengths are Re–C–O angles are comparable
with those usually observed. The C–Re–C angles show slight
deviation from the ideal 90Њ as might be expected due to steric
hindrance of the phenyl ring.
Complex 26. Crystals suitable for diffraction studies were
grown from dichloromethane/hexane. The molecules crystal-
lised in a triclinic lattice in the non-centrosymmetric space
group P1 and after refinement an R factor of 3.7% was
achieved. The molecular structure is shown in Fig. 6.
Although the two molecules in the asymmetric unit appear to
be related by an inversion centre, this symmetry is broken by
two atoms. Thus, O7 and O18, the ester oxygens, deviate signifi-
cantly from the centrosymmetric model. A refinement was tried
Fig. 4 Space-filling model of complex 22 to show the steric influence
of the Re(CO)4 group.
¯
in P1 but the R values were higher than in the non-centro-
symmetric space group, and the ester oxygens had large, elon-
gated ellipsoids. However this difference is very small and can
be ignored in comparison of the packing to that in complex 22.
As with complex 22 the presence of the two Re atoms on the
same side of the molecule is due to the interleaving way the
molecules pack within the unit cell. However, the increased
chain length would appear to have led to greater packing
effects. Thus, the torsion angle between the least square planes
of the ester-bridged rings which is 84.24(17)Њ in 22 has closed up
to 26.25(2) (molecule 1), 24.47(3)Њ (molecule 2), causing more
strain in the central part of the molecule, and resulting in the
ester group adopting two slightly different conformations
which is the origin of the two independent molecules in the
asymmetric unit. The torsion angles between the least square
planes of the 1st and 2nd and 3rd and 4th rings in the two
independent molecules are 35.21(3), 35.73(4) and 51.11(6),
48.85(5)Њ, respectively. In contrast to 15 and 22, these are very
different and this fact arises from the longer chain length having
an increased effect on the mode of packing. All the compounds
show differences in the conformation of the end chains and this
is exaggerated in the longer chain molecule, 26.
The Re–C bond lengths and the Re–C–O angles show values
within the expected range. However the increased steric strain
(two central rings are close to parallel) causes higher distortion
Fig. 5 Packing diagram for complex 22 viewed down the b axis
(hydrogen atoms omitted for clarity).
Fig. 6 Molecular structure of complex 26.
J. Chem. Soc., Dalton Trans., 2000, 1437–1445
1441

View Article Online
Table 3 Thermal data for the new ligands and complexes
∆H/kJ
∆S/J KϪ1
Compound
Transition
T/ЊC
molϪ1
molϪ1
Diimines
15
Crys–J
J–N
N–I (decomp.)
Crys–J
J–Sc
Sc–N
N–I(decomp.)
Crys–SI
(J–SI)
S1–SC
133
155
360
119
129
235
325
111
(110)
124
257
280
196
269
394
115
150
314
334
138
(137)
284
294
88
12.5
32.4
—
28.5
5.6
2.3
0.7
72.3
(22.2)
3.5
5.2
1.8
48.6
—
0.4
10.4
42.8
0.1
30.8
75
—
72
14
4
16
17
1
185
(61)
9
10
3
104
—
0
26
100
0
SC–N
Fig. 7 The nematic phase of compound 16 on cooling at 298 ЊC.
N–I(decomp.)
Crys–SC
Sc–N
N–I(decomp.)
Crys–CrysЈ
CrysЈ–SC
SC–N
N–I(decomp.)
Crys–Sc
(SI–SC)
18
19
2.5
4
20
21
54.7
(4.2)
1.5
2.5
14.2
8.2
8.6
26.9
0.7
1.8
1.4
132
(10)
3
SC–N
N–I(decomp.)
Crys–CrysЈ
CrysЈ–CrysЉ
CrysЉ–J
4
39
21
20
63
2
117
144
151
167
172
290
422
J–SI
SI–SC
a
4
SC–N
N–I(decomp.)
3
—
—
Fig. 8 The transition between the nematic and smectic C phase of
compound 16 at 236 ЊC on cooling showing the classic ’fingerprint’
texture.
Dirhenium complexes
22
23
24
25
26
27
Crys–I(decomp.)
Crys–I(decomp.)
Crys–I
Crys–I(decomp.)
Crys–I
Crys–CrysЈ
CrysЈ–I
220
216
196
238
176
128
148
220
45.2
63.4
72.9
62.4
69.5
6.4
86
128
155
118
154
16
45.9
109
28
CrysЈ–I
a See the text for a discussion of this event.
᎐
᎐
᎐
of both Re–C–O angles and O᎐C–Re–C᎐O angles which range
᎐
from 85.3(9) to 95.0(10)Њ.
Mesomorphism of the ligands
The mesomorphic properties of the diimines were obtained by
polarising optical microscopy to obtain optical textures of the
phases and differential scanning calorimetry (DSC) to evaluate
the enthalpy change at phase transitions. Thermal data for the
ligands are collected in Table 3.
All of the ligands showed a nematic phase (N) which was
identified by its characteristic schlieren texture (Fig. 7) and
highly fluid nature. Below the nematic phase each diimine, with
the exception of 15, also showed a smectic C phase which was
identified on the basis of its characteristic texture (Fig. 8) and
by the appearance of the familiar fingerprint texture at the
smectic C-to-nematic transition (Fig. 9).13 Comparison of the
transition temperatures shows that in 15–17, as the terminal
alkoxy chain is lengthened, the nematic phase is gradually
destabilised (TNI for 15 is 360 ЊC with decomposition, while for
17 it is 280 ЊC also with decomposition) while the smectic C
phase is stabilised (not seen in 15, but clearing at 257 ЊC in 17).
The same trend in nematic phase stability is also seen in
diimines 18–20 although while the smectic C stability increases
from 18 to 19, it decreases at 20. Further, ligands 15–17 also
possess a crystal J phase (Fig. 10) the stability of which drops
Fig. 9 Bar chart to show the mesomorphism of the ligands 15–20.
from 155 ЊC in 15 to 110 ЊC in 17 where it is monotropic. The
more anisotropic ligand, 21, melted at 144 ЊC and showed a
crystal smectic J phase, above which were smectic I and C
phases and, eventually, a nematic phase with a range of over
130 ЊC; the ligand cleared with decomposition at 422 ЊC. The
smectic I phase was identified both on account of the inability
to focus well on the texture and the presence of a schlieren
texture, both factors distinguishing it from the smectic F phase.
Identification of the crystal J phase by microscopy was not
unequivocal, but a J phase would be expected to follow a
smectic I phase (with a crystal G phase following a smectic F).
In addition, there is a direct analogy with related ligands where
the identity of the crystal J phase was established by X-ray
1442
J. Chem. Soc., Dalton Trans., 2000, 1437–1445

View Article Online
Table 4 Microanalytical data (%) for the new ligands and complexes
Calculated (Found)
Compound
Yield
C
H
N
Diimines
15
16
17
18
19
20
21
73
79
55
83
99
78
54
76.6 (76.4)
78.2 (78.4)
79.2 (79.3)
76.6 (76.5)
78.2 (78.2)
79.2 (79.1)
76.9 (77.0)
6.6 (6.5)
7.9 (8.0)
8.9 (8.9)
6.6 (6.5)
7.9 (7.9)
8.9 (8.9)
7.2 (7.2)
5.1 (5.1)
4.2 (4.2)
3.6 (3.6)
5.1 (4.9)
4.2 (4.0)
3.6 (3.5)
3.5 (3.5)
Dirhenium complexes
22
23
24
25
26
27
28
38
42
17
65
64
74
51
45.1 (44.6)
48.8 (48.9)
51.8 (51.6)
45.1 (44.9)
48.8 (48.4)
51.8 (51.4)
50.6 (50.8)
3.0 (2.9)
4.0 (3.9)
4.9 (4.8)
3.0 (2.9)
4.0 (3.9)
4.9 (4.7)
3.9 (3.9)
2.4 (2.3)
2.2 (2.2)
2.0 (2.0)
2.4 (2.3)
2.2 (2.2)
2.0 (2.0)
2.0 (2.0)
Fig. 10 Mosaic texture of the crystal J phase of compound 16
obtained at 118 ЊC on cooling.
methods.14 However, one unresolved aspect was the fact that the
DSC showed a transition at 172 ЊC (∆H = 1.8 kJ molϪ1) which
did not correspond to a significant change in texture. At the
present time we can offer no explanation for this, but in due
course we plan to have the compound examined by X-ray
methods in order to pin this behaviour down.
there is no evidence whatsoever for such an argument at this
time.
The broad pattern of mesomorphism in these ligands is
rather similar to that of related monoimines which we have
synthesized previously when studying related monometallic
complexes.14
Experimental
Elemental analysis was determined by the University of Exeter
Microanalysis Service. NMR spectra were recorded on either
a Bruker ACF-300 or DRX-400 spectrometer, where the chem-
ical shifts are reported relative to the internal standard of the
deuteriated solvent used. Analysis by DSC was carried out on a
Perkin-Elmer DSC7 instrument using heating and cooling rates
of either 5 or 10 K minϪ1. Analysis by hot stage microscopy was
carried out using an Olympus BH40 microscope equipped with
a Link-Am HFS91 hot stage, TMS92 controller and LNP2
cooling unit. Solvents were purified by standard methods before
use. 4-Heptylbenzoic acid, 4-hexyloxyaniline and 4-heptyl-
aniline were obtained from Aldrich. Yields and analytical data
for new compounds are collected in Table 4.
Cell dimensions and intensity data for compounds 15, 22 and
26 were recorded at 150 K, using an Enraf-Nonius KappaCCD
area detector diffractometer mounted at the window of a rotat-
ing anode operating at 50 kV, 50 mA with a molybdenum anode
(λ(Mo-Kα) = 0.71073 Å). The crystal-to-detector distance was
30 mm and φ and ω scans (2Њ increments, 10 s exposure time)
were carried out to fill the Ewald sphere. Data collection and
processing were carried out using the programs COLLECT,15
DENZO16 and MAXUS17 and an empirical absorption correc-
tion was applied using SORTAV.18,19 All structures were solved
via direct methods20 and refined by full matrix least squares20
on F 2. Non-hydrogen atoms were refined anisotropically and
hydrogen atoms were placed in geometrical positions and
refined using a riding model.
Mesomorphism of the complexes
The mesomorphism of the complexes is simply dealt with as it
was non-existent! In each case the complexes melted directly to
the isotropic phase (Table 3), occasionally with decomposition
and in one case (27) passing through a second crystal modifi-
cation. Having found this to be the case for dinuclear complexes
of the four-ring imines, 15–20, we wished to push the study one
step further by synthesizing a dinuclear complex of a five-ring
diimine (21) whose mesomorphism is reported above. In this
case too the dinuclear complex proved to be non-mesomorphic,
simply melting at 220 ЊC. Thus, it is absolutely clear that the
introduction of the second Re(CO)4 unit destroys meso-
morphism. It is, however, interesting to speculate a little on why
this might be.
In the present study the ligands show predominantly smectic
C and nematic phases extending up to, and above, 300 ЊC. In
our previous studies similar ligands displayed a similar meso-
morphism in a similar temperature range. In these previous
studies, complexation of the ligands led to materials which
showed only a nematic phase which formed from the crystal
phase at somewhere between 130–150 ЊC and persisted until
near 200 ЊC. In the present case two factors are apparent. First,
in complexes 22–24, the chain length has relatively little effect
on the melting point which in any case is rather high at about
200 ЊC or above. The same situation is found for the dinuclear
complex of the five-ring imine (28). In these cases, it might be
postulated that the crystal phase is stabilised well above the
temperature at which any mesophase might exist. It is impos-
sible to show this for 22, 23 and 28 which decompose on melt-
ing, but careful studies of 24 show that there is no evidence of a
supercooled (monotropic) nematic phase below the crystal
phase. Curiously, however, for 25–27 there is a much stronger
dependence of the melting point on chain length, decreasing by
90 ЊC from butyloxy to dodecyloxy terminal chains. Once
again, we find no evidence of any monotropic phases only this
time, with melting points as low as 148 ЊC for 27, we might
reasonably have expected to see some evidence of a monotropic
phase unless our strategy has been so successful that the transi-
tion temperatures are very much reduced. It is, of course,
tempting to speculate that this might be the case but of course
CCDC reference number 186/1902.
See http://www.rsc.org/suppdata/dt/a9/a910273f/ for crystal-
lographic files in .cif format.
Preparations
4-Octyloxybenzaldehyde, 2. 4-Hydroxybenzaldehyde (2 g,
16.4 mmol) was dissolved in butanone (100 cm3) and potassium
carbonate (9 g, 65.6 mmol) and potassium iodide (0.2 g) added.
1-Bromooctane (4.8 g, 19.7 mmol) was added and the mixture
heated at reflux under nitrogen for 48 h. On cooling, water (500
cm3) was added and the organic layer extracted with CH2Cl2
(2 × 150 cm3). The combined organic extracts were washed with
5% aqueous NaOH (2 × 100 cm3), dried over anhydrous Na2-
SO4, filtered and evaporated to a give a yellow oil. The crude
product was purified by column chromatography (SiO2, eluent
J. Chem. Soc., Dalton Trans., 2000, 1437–1445
1443

View Article Online
CH2Cl2) and the solvent evaporated to give the pure yellow oil.
Yield: 3.45 g (90%). 1H NMR (400 MHz, CDCl3): δ 0.90 (d, 3,
CH3), 1.45 (m, 10, 5 CH2), 1.75 (m, 2, CH2CH2O), 4.0 (t, 2,
3JHH = 9.1, CH2O), 7.0 (AAЈXXЈ, 2, J = 9), 7.85 (AAЈXXЈ, 2,
J = 9 Hz) and 9.9 (s, 1, CHO).
acid (1.23 g, 8.2 mmol) in CH2Cl2 (60 cm3) was added DCC
(2.03 g, 9.84 mmol) and DMAP (24 mg). The mixture was
stirred under nitrogen for 18 h. The precipitate was filtered off
and the solvent evaporated to give a yellow compound. The
yellow crude product was purified by column chromatography
(SiO2, eluent CH2Cl2) and the solvent evaporated to give a pure
pale yellow powder. Yield 1.36 g (65%). Found: C, 70.8; H, 3.9.
Calc. for C15H8O4, C, 70.8; H, 4.0%. 1H NMR (300 MHz,
CDCl3): δ 7.45 (AAЈXXЈ, 2, J = 7.9), 8.05 (AAЈXXЈ, 2, J = 7.9),
8.10 (AAЈXXЈ, 2, J = 7.9), 8.45 (AAЈXXЈ, 2, J = 7.9 Hz), 10.0
(s, 1, CHO) and 10.2 (s, 1, CHO).
4-Butyloxybenzaldehyde 1 and 4-dodecyloxybenzaldehyde 3
were similarly prepared in yields of 83 and 88%, respectively.
4-Octyloxyaniline, 5. A solution of 1-bromooctane (7.03 g,
36.37 mmol) was added to a suspension of 4-acetamidophenol
(5 g, 33.07 mmol) and potassium carbonate (9.2 g, 66.14 mmol)
in butanone (80 cm3). The mixture was heated at reflux for 18 h.
After cooling, the solution was filtered through a Celite plug
and the solvent evaporated. The residue was washed with Et2O.
After evaporation of the solvent a colourless powder was
obtained. The filtrate was washed with a solution of NaOH
(2 × 100 cm3, 2 mol dmϪ3), then NaCl (2 × 100 cm3, 2 mol
dmϪ3), dried over Na2SO4 and the solvent again removed. The
product was crystallised from hot ethanol (absolute) and a
white solid obtained. The solid was dissolved in ethanol (100
cm3, 95%) and HCl (53.6 cm3, 35%) added. The solution was
heated at reflux for 3 h. After filtration the solid was added to
water (50 cm3) and a solution of NaOH (2 mol dmϪ3) added
until pH 12. The solid was filtered off, dried and stored under
Diimine 16. To a suspension of 4-[4-octyloxybenzylidene-
amino]benzoic acid 8 (1.08 g, 3.07 mmol) and 4-[4-octyloxy-
benzylideneamino]phenol 11 (1.0 g, 3.07 mmol) in CH2Cl2 (100
cm3) were added DCC (0.76 g, 3.68 mmol) and DMAP (9 mg).
The mixture was stirred under nitrogen for 18 h. The precipitate
was filtered off and the solvent evaporated to give a yellow
powder. The solid was crystallised from CH2Cl2/hexane. Yield
1.60 g (79%). 1H NMR (400 MHz, CH2Cl2): δ 0.90 (t, 6, CH3),
1.33 (m, 16, 8 CH2), 1.48 (m, 2, 8 CH2), 1.82 (m, 4, CH2CH2O),
4.03 (m, 4, 2 CH2O), 6.98 (2 AAЈXXЈ, 4), 7.26 (3 AAЈXXЈ, 6),
7.85 (2 AAЈXXЈ, 4), 8.23 (AAЈXXЈ, 2), 8.38 (s, 1, CH᎐N) and
᎐
8.40 (s, 1, CH᎐N). Diimine 15 was similarly prepared.
᎐
1
N2 at sub-ambient temperatures. Yield 4.95 g (56%). H NMR
(300 MHz, CDCl3): δ 0.90 (t, 3, CH3), 1.30 (m, 10, 5 CH2), 1.80
(m, 2, CH2CH2O), 2.90 (br s, 2, NH2), 4.40 (t, 2, CH2O, J = 6.8),
6.55 (AAЈXXЈ, 2, J = 8.8) and 6.70 (AAЈXXЈ, 2, J = 8.8 Hz).
4-Dodecyloxyaniline 6 was similarly prepared in a yield of
60%, while 4-butyloxyaniline 4 was purchased from Aldrich.
Diimine 17. To a suspension of 4-dodecylbenzaldehyde 3
(0.73 g, 2.49 mmol) and 4-aminophenyl4-aminobenzoate 13
(0.19 g, 0.83 mmol) in ethanol (40 cm3, absolute) were added a
few drops of acetic acid (glacial). The mixture was stirred over-
night at room temperature. The resulting solid was filtered off,
crystallised from CH2Cl2/hexane and dried to yield a pure off-
1
4-[4-Octyloxybenzylideneamino]benzoic acid, 8. To a sus-
pension of 4-octyloxybenzaldehyde 2 (0.5 g, 2.1 mmol) and
4-aminobenzoic acid (0.3 g, 2.1 mmol) in ethanol (40 cm3,
absolute) were added a few drops of acetic acid (glacial). The
mixture was heated at reflux for 15 min and then cooled to
room temperature. The solid was filtered off crystallised from
hot ethanol (absolute) and dried to yield a pure, off-white solid.
Yield 0.6 g (81%). Found: C, 74.8 ; H, 7.7; N, 4.2. Calc. for
white solid. Yield 0.35 g (55%). H NMR (300 MHz, CDCl3):
δ 0.90 (t, 6, CH3), 1.30 (m, 32, 8 CH2), 1.50 (m, 4, 8 CH2), 1.90
(m, 4, CH2CH2O), 4.03 (m, 4, CH2O), 6.98 (AAЈXXЈ, 2,
J = 8.4), 6.99 (AAЈXXЈ, 2, J = 8.4), 7.25 (3 overlapping
AAЈXXЈ, 6), 7.90 (2 overlapping AAЈXXЈ, 4, each J = 8.4), 8.20
(AAЈXXЈ, 2, J = 8.4 Hz), 8.38 (s, 1, CH᎐N) and 8.40 (s, 1,
᎐
CH᎐N).
᎐
1
C22H27NO3: C, 74.8; H, 7.7; N, 4.0%. H NMR (400 MHz,
Diimine 19. To a suspension of 4-octyloxyaniline 5 (0.57, 2.59
CDCl3): δ 0.90 (t, 3, CH3), 1.34 (m, 10, 5 CH2), 1.81 (m, 2,
CH2CH2O), 4.03 (t, 2, JHH = 6.4, CH2O), 7.0 (AAЈXXЈ, 2,
J = 8.8), 7.21 (AAЈXXЈ, 2, J = 8.8), 7.83 (AAЈXXЈ, 2, J = 8.8),
mmol) and 4-formylphenyl 4-formylbenzoate 14 (0.3 g, 1.18
mmol) in ethanol (40 cm3, absolute) were added a few drops of
acetic acid (glacial). The mixture was stirred overnight at room
temperature. The resulting solid was filtered off, washed with
hot ethanol (absolute) and dried to yield a pure off-white solid.
3
8.12 (AAЈXXЈ, 2, J = 8.8 Hz), 8.12 (s, 1, CH᎐N) and 9.88 (s, 1,
᎐
CO2H).
1
4-[4-Butyloxybenzylideneamino]benzoic acid 7, 4-[4-dodecyl-
oxybenzylideneamino]benzoic acid 9, 4-[4-butyloxybenzyl-
ideneamino]phenol 10, 4-[4-octyloxybenzylideneamino]phenol
11 and 4-[4-dodecyloxybenzylideneamino]phenol 12 were
similarly prepared in yields 70, 78, 83, 83 and 82% respectively.
Yield 0.77 g (99%). H NMR (400 MHz, CH2Cl2): δ 0.91 (t, 6,
CH3), 1.32 (m, 16, 8 CH2), 1.49 (m, 2, 8 CH2), 1.81 (m, 4,
CH2CH2O), 4.00 (m, 4, CH2O), 6.95 (AAЈXXЈ, 2, J = 8.8), 6.96
(AAЈXXЈ, 2, J = 8.8), 7.26 (AAЈXXЈ, 2, J = 8.8), 7.32 (AAЈXXЈ,
2, J = 8.8), 7.38 (AAЈXXЈ, 2, J = 8.8), 8.01 (AAЈXXЈ, 2,
J = 8.8), 8.08 (AAЈXXЈ, 2, J = 8.8), 8.31 (AAЈXXЈ, 2, J = 8.4
4-Aminophenyl 4-aminobenzoate, 13. To a suspension of
4-nitrophenol (0.83 g, 3.5.98 mmol) was added to a suspension
of 4-nitrobenzoic acid (1 g, 5.98 mmol), DCC (1 g, 7.17 mmol)
and DMAP (17 mg) in CH2Cl2 (60 cm3). The mixture was
stirred under nitrogen for 18 h. The precipitate was filtered off
and the solvent evaporated to give a yellow compound. The
yellow crude product was purified by column chromatography
(SiO2, eluent CH2Cl2) and the solvent evaporated to give a pure
pale yellow powder. The powder was dissolved in THF (30 cm3)
and palladium on carbon (10 mg, 10%) was added. The mixture
was kept under H2 pressure (20 psi) overnight. The solution was
filtered on a Celite plug and the solvent removed. The solid was
crystallised from ethanol/hexane. Yield 0.91 g (67%). 1H NMR
(300 MHz, CDCl3): δ 4.1 (s, 4, NH2), 6.67 (2 overlapping
AAЈXXЈ, 4, each J = 7.9), 7.0 (AAЈXXЈ, 2, J = 7.9) and 8.0
(AAЈXXЈ, 2, J = 7.9 Hz).
Hz), 8.54 (s, 1, CH᎐N) and 8.63 (s, 1, CH᎐N).
Diimines 18, 20 and 21 were similarly prepared.
᎐ ᎐
Complex 23. To a suspension of diimine 16 (55 mg, 0.076
mmol) in toluene (20 cm3) was added pentacarbonylmethyl-
rhenium() (57 mg, 0.167 mmol). The reaction mixture was
stirred under reflux for 20 hours, under a nitrogen atmosphere
and then cooled to room temperature and the solvent removed.
The yellow solid was dissolved in CH2Cl2 and purified by
column chromatography on alumina with CH2Cl2 as eluent.
The solution was concentrated and hexane added to give the
1
pure yellow solid in 42% yield (40 mg). H NMR (400 MHz,
CD2Cl2): δ 0.92 (2 t, 6, 2 CH3), 1.35 (m, 16, CH2), 1.50 (m, 4,
CH2), 1.84 (m, 4, 2 CH2CH2O), 4.11 (2 t, 4, 2 CH2O), 6.72 (2dd,
2, 3JHH = 8.0, 4JHH = 4.0, 2 Hb), 7.36 (2 AAЈXXЈ, 4, J = 8.4), 7.39
(AAЈXXЈ, 2, J = 8.4), 7.44 (AAЈXXЈ, 2, J = 8.8), 7.54 (d, 1,
4
3
4JHH = 4, Ha), 7.57 (d, 1, JHH = 4, Ha), 7.69 (d, 1, JHH = 8.4,
3
4-Formylphenyl 4-formylbenzoate, 14. To a suspension of
Hc), 7.73 (d, 1, JHH = 8.4, Hc), 8.32 (AAЈXXЈ, 2, J = 8.8 Hz),
4-hydroxybenzaldehyde (1 g, 8.2 mmol) and 4-formylbenzoic
8.50 (s, 1, CH᎐N) and 8.5 (s, 1, CH᎐N).
᎐ ᎐
1444
J. Chem. Soc., Dalton Trans., 2000, 1437–1445

View Article Online
8 S. Morrone, D. Guillon and D. W. Bruce, Inorg. Chem., 1996, 35,
All other dirhenium complexes (23, 24–28) were prepared in
the same way.
7041.
9 K. E. Rowe and D. W. Bruce, J. Chem. Soc., Dalton Trans., 1996,
3913.
10 N. Miyajima and Y. Matsunaga, Mol. Cryst., Liq. Cryst., 1995, 260,
Acknowledgements
499.
We would like to thank H. C. Stark for a generous gift of
rhenium, the EU for a Marie Curie Fellowship (to M. A. G.)
and Professor John Goodby of the University of Hull for help
with the microscopy of ligand 21.
11 W. Beck and K. Raab, Inorg. Synth., 1990, 28, 15.
12 V. N. Kalinin, V. I. Zdanovich, P. V. Petrovskii, A. S. Batsanov and
Yu. T. Struchkov. Metalloorg. Khim., 1990, 3, 662.
13 G. W. Gray and J. W. Goodby, Smectic Liquid Crystals: Textures and
Structures, Leonard Hill, Glasgow, 1984.
14 X.-H. Liu, B. Heinrich, I. Manners, D. Guillon and D. W. Bruce,
J. Mater. Chem., 2000, 10, 637.
15 COLLECT data collection software, Nonius B.V., Delft, 1998.
16 Z. Otwinowski and W. Minor, Methods Enzymol., 1997, 276A,
307.
References
1 D. W. Bruce and D. O’Hare, Inorganic Materials, 1996, Wiley,
Chichester, 2nd edn., 1996.
2 B. Donnio and D. W. Bruce, Struct. Bonding, (Berlin), 1999, 95, 193.
17 S. Mackay, C. J. Gilmore, C. Edwards, M. Tremayne, N. Stuart and
K. Shankland, MAXUS, a computer program for the solution and
refinement of crystal structures from diffraction data, University of
Glasgow, Nonius BV, Delft, and MacScience Co. Ltd., Yokohama,
1998.
18 R. H. Blessing, Acta Crystallogr., Sect. A, 1995, A51, 33.
19 R. H. Blessing, J. Appl. Crystallogr., 1997, 30, 421.
20 SHELX 97: Programs for Crystal Structure Analysis (Release 97–2),
G. M. Sheldrick, University of Göttingen, 1998.
3 D. W. Bruce, Adv. Mater., 1994, 6, 699.
4 D. W. Bruce and X.-H. Liu, J. Chem. Soc., Chem. Commun., 1994,
729.
5 X.-H. Liu, M. N. Abser and D. W. Bruce, J. Organomet. Chem.,
1998, 551, 271.
6 D. W. Bruce and X.-H. Liu, Liq. Cryst., 1995, 18, 165.
7 M.-A. Guillevic, M. J. Danks, S. K. Harries, S. R. Collinson, A. D.
Pidwell and D. W. Bruce, Polyhedron, 2000, 19, 249.
J. Chem. Soc., Dalton Trans., 2000, 1437–1445
1445