Mesomorphism of ionic allylpalladium(II) complexes containing pz R2py as ligands and [BF4]-, [PF 6]- or [CF3SO3]- as counteranions

Article abstract of DOI:10.1039/b601637e

The synthesis and thermal behaviour is reported of some ionic liquid-crystalline complexes formed when a mesogenic, two-chained, pyrazolylpyridine ligand is bound to a η³-allylpalladium(ii) fragment. The phase behaviour of the complexes is dominated by the formation of the SmA phase and, while the stability of the crystal phase appears to be controlled by intermolecular solid-state interactions, the stability of the liquid crystal phases is controlled by the length of the ligand alkoxy chains. Some of the complexes also show a complex polymorphism leading to the observation of double-melting behaviour. The Royal Society of Chemistry 2006.

Full text of DOI:10.1039/b601637e

View Article Online / Journal Homepage / Table of Contents for this issue
PAPER
www.rsc.org/dalton | Dalton Transactions
Mesomorphism of ionic allylpalladium(II) complexes containing pzR2py as
ligands and [BF4]−, [PF6]− or [CF3SO3]− as counteranions†‡
Mari Carmen Torralba,a Jose´ Antonio Campo,a Jose´ Vicente Heras,a Duncan W. Bruceb and Mercedes Cano*a
Received 2nd February 2006, Accepted 15th May 2006
First published as an Advance Article on the web 30th May 2006
DOI: 10.1039/b601637e
The synthesis and thermal behaviour is reported of some ionic liquid-crystalline complexes formed
3
when a mesogenic, two-chained, pyrazolylpyridine ligand is bound to a g -allylpalladium(II) fragment.
The phase behaviour of the complexes is dominated by the formation of the SmA phase and, while the
stability of the crystal phase appears to be controlled by intermolecular solid-state interactions, the
stability of the liquid crystal phases is controlled by the length of the ligand alkoxy chains. Some of the
complexes also show a complex polymorphism leading to the observation of double-melting behaviour.
On these bases, the related mesogenic N,Nꢀ-bidentate 2-[3,5-
1
Introduction
bis(4-alkyloxyphenyl)pyrazol-1-yl]pyridine (pzR2py) ligands were
designed strategically and complexed with PdCl2, giving rise to
the expected [Pd(Cl)2(pzR2py)] compounds which also exhibited
liquid crystal behaviour (Scheme 1).18 Those results suggested us
to consider that the unsymmetrical pzR2py ligands can be used
successfully to produce an elongated molecular shape and, at the
same time, a lower symmetry of the complexes; both conditions
prove to be useful in designing mesogens. By contrast, the related
[Pd(Cl)2(pzRpy)] complexes containing the non-mesomorphic
pzRpy ligands bearing only one substituent at the 3-position of the
pyrazole group, were not liquid-crystalline,19 in agreement with
the shortening of the molecular length in relation to that of the
homologues containing pzR2py ligands (Scheme 1). Surprisingly,
when these N,Nꢀ-bidentate pzRpy ligands were coordinated to the
The study of metallomesogens, that is liquid crystals based on
metal complexes, has developed significantly as a subject in the
last twenty years and a wide range of different metals has been
used in their construction, including d- and f-block elements.1–6
Considering this work as a whole, it is clear that the majority
of systems described might formally be regarded as coordina-
tion complexes, while a smaller number can be classified as
organometallic (defined here as complexes containing M–C bonds
including to CO and isonitriles).7 Many organometallic systems
derived from orthometallated imines and azo compounds bound
to manganese(I), rhenium(I) and (particularly) palladium(II) and
platinum(II) have been described, but there are fewer complexes
where the organometallic moiety is coordinated with a hapticity of
two or more (with the notable exception of ferrocene mesogens).8
3
[Pd(g -C3H5)]+ fragment they gave rise to the new metallome-
2
4
Examples include g -alkene complexes of Pt(II),9 g -butadiene
3
sogenic [Pd(g -C3H5)(pzRpy)]+.19 Unfortunately, these complexes
6
complexes of Fe(0),10 and g -arene complexes of Cr(0).11
exhibited high clearing temperatures, giving rise to partial or full
decomposition. Scheme 1 summarises the above results concerning
the pyrazolylpyridine ligands and their complexes.
Recently, we reported the synthesis of Pd(II) mesogens contain-
3
ing an g -allyl group to which were also coordinated mesogenic
pyrazoles. Previous work in our laboratory dealt with the studies
of the mesomorphic properties of pyrazole ligands containing
C6H4OCnH2n+1 as substituents at the 3- or at the 3- and 5-
positions of the ring (HpzR and HpzR2, respectively) as well as
that of some of their metal complexes.12–17 Most of them behave
as liquid crystal molecular materials, irrespective of whether
mesomorphic (HpzR2) or non-mesomorphic (HpzR) ligands are
used. So, for instance, it was interesting to note how the non-
mesomorphic HpzR ligands induced liquid crystal behaviour on
the complexes t-[Pd(Cl)2(HpzR)2], which exhibited a conventional
rod-like molecular geometry.14
Following those precedents, we are now interested in proving
that coordination of the longer and mesogenic pzR2py ligands
3
to the [Pd(g -C3H5)]+ fragment should produce more adequate
and hopefully more stable mesogenic derivatives. We wished to
attain the possibility of improving the mesomorphic properties
3
of these [Pd(g -C3H5)(pzRpy)]+ systems by modifying the rod-
like molecular shape. So, we thought that by introducing a new
substituent group at the 5-position of the pyrazole ring, the
anisometric ratio between molecular length and width of the
3
resulting [Pd(g -C3H5)(pzR2py)]+ complexes should be increased
in relation to that of those containing pzRpy ligands, so favouring
the mesomorphic properties.
On the other hand, mesomorphism on palladium complexes
containing bidentate ligands has been restricted to that of neutral
cyclopalladated N(sp2)–Pd–C(sp2) species,20–26 and some others
produced by coordination of several N,N-coordinated bipyridine-
type ligands towards PdCl2.27 In addition, ionic mesogenic palla-
dium derivatives have scarcely been reported.28–31
By contrast, linear two-coordinated cationic silver com-
plexes with alkoxystilbazoles have been studied extensively,32
their liquid crystal behaviour depending on the anion and
aDepartamento de Qu´ımica Inorga´nica I, Facultad de Ciencias Qu´ımicas,
Universidad Complutense, E-28040, Madrid, Spain. E-mail: mmcano@
quim.ucm.es; Fax: +34-91-3944352
bDepartment of Chemistry, University of York, Heslington, York, UK YO10
5DD
† Dedicated to Dr J. A. Abad (University of Murcia, Spain) with occasion
of his retirement.
‡ Electronic supplementary information (ESI) available: Spectroscopic
data of all compounds, microphotographs from SmA mesophases of I16
and II18, and seven selected DSC traces. See DOI: 10.1039/b601637e
3918 | Dalton Trans., 2006, 3918–3926
This journal is
The Royal Society of Chemistry 2006
©

View Article Online
Scheme 1 Schematic molecular representation of the pzR2py (a) and pzRpy (b) ligands, and their complexes: [Pd(Cl)2(pzR2py)] (c), [Pd(Cl)2(pzRpy)] (d)
3
and [Pd(g -C3H5)(pzRpy)][X] (e).
the number and length of alkoxy chains, among other
factors.
This type of complexes constitute one of the few examples of
ionic mesogenic palladium derivatives that contain unsymmetrical
N,Nꢀ-bidentate ligands in a rigid five-membered metallocyclic
core.
In the present work we describe the preparation, characteri-
3
sation and thermal study of a family of ionic complexes [Pd(g -
C3H5)(pzR2py)][X] (pzR2py = 2-[3,5-bis(4-alkyloxyphenyl)pyrazol-
1-yl]pyridine, R = C6H4OCnH2n+1, n = 10 (dp), 12 (ddp), 14 (tdp),
−
−
−
16 (hdp), 18 (odp)), X = BF4 (I), PF6 (II) and CF3SO3 (III).
Scheme 2 shows the complexes studied in this work including the
nomenclature used. All complexes of these three classes containing
chains with 14–18 carbon atoms and the compound with 12 carbon
atoms at the chains and BF4− (I12) have been proved to have liquid
crystal properties.
2
Experimental
2.1 Materials and physical measurements
All commercial reagents were used as supplied. The starting Pd-
3
complex [Pd(l-Cl)(g -C3H5)]2 was purchased from Sigma-Aldrich
Scheme 2
This journal is
The Royal Society of Chemistry 2006
Dalton Trans., 2006, 3918–3926 | 3919
©

View Article Online
and used as supplied. The 2-[3,5-bis(4-alkyloxyphenyl)pyrazol-1-
yl]pyridine (pzR2py) ligands (R = C6H4OCnH2n+1, n = 12, 14, 16,
18) were prepared as described previously.18,33
Elemental analyses for carbon, hydrogen, nitrogen and sulfur
were carried out by the Microanalytical Service of Complutense
University. IR spectra were recorded on a FTIR ThermoNicolet
200 or on a FTIR Nicolet Magna-550 spectrophotometers with
samples as KBr pellets in the 4000–400 cm−1 region: vs (very
strong), s (strong), m (medium), w (weak).
1H NMR spectra were performed on a Bruker AC200, on a
Bruker DPX-300 or on a Varian VXR-300 spectrophotometers
(NMR Service of Complutense University) from solutions in
CDCl3. Chemical shifts d are listed relative to TMS using the
signal of the deuterated solvent as reference (7.26 ppm), and
coupling constants J are in hertz. Multiplicities are indicated as s
(singlet), d (doublet), t (triplet), m (multiplet), br (broad signal).
The atomic numbering used in the assignment of the signals is
shown in Scheme 2. The 1H chemical shifts and coupling constants
are accurate to 0.01 ppm and 0.3 Hz, respectively.
Phase studies were carried out by optical microscopy using an
Olympus BX50 microscope equipped with a Linkam THMS 600
heating stage. The temperatures were assigned on the basis of optic
observations with polarised light.
Measurements of the transition temperatures were made using
a Perkin Elmer Pyris 1 differential scanning calorimeter with the
sample (1–4 mg) sealed hermetically in aluminium pans and with
a heating or cooling rate of 5–10 K min−1.
2.3 Preparation of [Pd(g3-C3H5)(pzR2py)][BF4] (R =
C6H4OCnH2n+1; n = 10 (I10), 12 (I12), 14 (I14), 16 (I16), 18 (I18))
3
To a solution of [Pd(l-Cl)(g -C3H5)]2 (100 mg, 0.273 mmol) in
dry acetone (20 mL) was added AgBF4 (106.3 mg, 0.546 mmol)
under a nitrogen atmosphere. The mixture was stirred in the
absence of light at least for 6 h and then filtered over Celite. The
corresponding pzR2py (0.546 mmol) in acetone (40 mL) was added
to the resulting colourless solution and allowed to stir for 24 h at
room temperature under nitrogen. Then the solvent was removed
in vacuo and the solid crystallised from dichloromethane—hexane
leading to the precipitation of a colourless solid, which was filtered
off, washed with hexane and dried in vacuo. Yields and analytical
data are collected in Table 1. Spectroscopic data are given
as ESI.‡
2.4 Preparation of [Pd(g3-C3H5)(pzR2py)][X] (X = PF6−, R =
C6H4OCnH2n+1; n = 12 (II12), 14 (II14), 16 (II16), 18 (II18); X =
CF3SO3−, R = C6H4OCnH2n+1; n = 12 (III12), 14 (III14), 16 (III16),
18 (III18))
Complexes II and III were prepared in a same way to that above
described for the BF4 derivatives I, but starting from AgPF6 or
AgCF3SO3. Yields and analytical data are collected in Table 1.
Spectroscopic data are given as ESI.‡
3
Results and discussion
The X-ray diffractograms at variable temperature were recorded
on a Panalytical X’Pert PRO MPD diffractometer in a h–h
configuration equipped with a Anton Paar HTK1200 heating stage
(X-Ray Diffraction Service of Complutense University).
3.1 Synthetic studies
The syntheses of [Pd(g -C3H5)(pzR2py)][X] (R = C6H4OCnH2n+1
,
3
−
−
−
n = 10, 12, 14, 16, 18; X = BF4 (I10–18), PF6 (II12–18), CF3SO3
(III12–18)) was carried out using a procedure similar to that used
for the related complexes containing pzRpy. The complexes were
2.2 Preparation of 2-[3,5-bis(4-decyloxyphenyl)pyrazol-1-
yl]pyridine (pzdp2py)
3
−
obtained by reaction of [Pd(g -C3H5)(acetone)2][X] (X = BF4
,
PF6−, CF3SO3−), prepared in situ by reaction of the dimer [Pd(l-
The new ligand pzdp2py was prepared in a similar way used for
analogous ligands.18,33 This synthesis is based on the condensation
of the corresponding 1,3-bis(4-alkyloxyphenyl)propane-1,3-dione
and 2-hydrazinopyridine in ethanol using an acidic media to
increase the yield. Mp 68 ◦C. Yield: 36%. Elemental analyses:
found: C 78.5, H 8.9, N 6.5%; calc. for C40H55N3O2: C 78.8, H 9.1,
N 6.9%. Spectroscopic data are given as ESI.‡
3
Cl)(g -C3H5)]2 and AgX in acetone,19,34 with the corresponding
ligand pzR2py (Scheme 3).
3.2 Spectroscopic and structural characterisation
All the compounds were characterised by analytical and spectro-
1
scopic (IR and H NMR) techniques (see Experimental section
Table 1 Yields and analytical data for the new complexes
Elemental analysis: calc (found) (%)
Compound
Yield (%)
C
H
N
S
3
[Pd(g -C3H5)(pzdp2py)][BF4]
I10
62
65
70
67
60
60
63
54
58
70
81
76
79
61.2
(61.1)
(62.6)
(64.0)
(65.0)
(66.2)
(58.6)
(60.2)
(61.4)
(62.8)
(60.0)
(61.3)
(62.7)
(63.7)
7.2
(7.0)
(7.5)
(8.0)
(8.2)
(8.5)
(6.9)
(7.4)
(7.8)
(8.1)
(7.0)
(7.5)
(7.8)
(8.0)
5.0
4.7
4.5
4.2
3.9
4.4
4.1
3.9
3.7
4.4
4.1
3.9
3.7
(5.0)
(4.8)
(4.4)
(4.4)
(3.9)
(4.5)
(4.3)
(4.0)
(3.8)
(4.5)
(4.3)
(4.0)
(3.8)
—
—
—
—
—
—
—
—
—
3.3
3.2
3.0
2.8
—
—
—
—
—
—
—
—
3
[Pd(g -C3H5)(pzddp2py)][BF4]
I12
62.7
64.0
65.2
66.3
58.9
60.4
61.7
62.9
59.9
61.3
62.6
63.7
7.6
8.0
8.4
8.7
7.2
7.6
7.9
8.2
7.1
7.5
7.9
8.2
3
[Pd(g -C3H5)(pztdp2py)][BF4]
I14
3
[Pd(g -C3H5)(pzhdp2py)][BF4]
I16
3
[Pd(g -C3H5)(pzodp2py)][BF4]
I18
3
[Pd(g -C3H5)(pzddp2py)][PF6]
II12
II14
II16
II18
III12
III14
III16
III18
3
[Pd(g -C3H5)(pztdp2py)][PF6]
3
[Pd(g -C3H5)(pzhdp2py)][PF6]
3
[Pd(g -C3H5)(pzodp2py)][PF6]
—
3
[Pd(g -C3H5)(pzddp2py)][CF3SO3]
(3.4)
(3.1)
(2.9)
(2.9)
3
[Pd(g -C3H5)(pztdp2py)][CF3SO3]
3
[Pd(g -C3H5)(pzhdp2py)][CF3SO3]
3
[Pd(g -C3H5)(pzodp2py)][CF3SO3]
3920 | Dalton Trans., 2006, 3918–3926
This journal is
The Royal Society of Chemistry 2006
©

View Article Online
Scheme 3
and ESI‡), and all data agree with their proposed molecular
formulation.
The IR spectra of the complexes in the solid state show the
characteristic bands of the pyrazolylpyridine ligands scarcely
modified upon coordination and confirmed the presence of the
and H4 protons are not modified by the presence of the different
counteranions. It is of interest to mention that the H3 proton is
shifted upfield with respect to the free ligand, whilst an opposite
effect is observed for the H4, H5 and H6 protons.
The non-equivalence of the lateral substituents at the pyrazolyl
ring remains upon coordination of the ligand, the spectra showing
two signals for each type of their protons. One set of the Ho
protons of the C6H4 groups is shifted downfield with respect to
the free ligand, while the other is shifted upfield. However, the
two signals of the Hm protons are slightly displaced at higher
frequencies. The H4 signal of the pyrazolyl group is unchanged
after coordination.
=
counteranions. It is interesting to mention the m(C N) band at ca.
1610 cm−1 from the pyrazolyl group as well as the characteristic
bands for the different counteranions. So, for complexes I, a broad
m(B–F) band at ca. 1087–1052 cm−1 and d(F–B–F) at ca. 518 cm−1
were observed from the BF4− counteranion. Analogously the m(P–
F) and d(F–P–F) bands at ca. 844 and 558 cm−1, respectively,
−
from the PF6 counteranion were observed for complexes II. The
−
spectra for complexes III with CF3SO3 as counteranion display
the md(SO3) and m(CF3) bands overlapped at ca. 1260 cm−1 and the
ms(SO3) absorption at ca. 1029 cm−1. The remaining bands are not
certainly assigned because of the superposition with many others
of the ligand.
3.3 Structural studies
The relationship between chemical structure and mesomorphism
has been investigated extensively in the field of liquid crystals,
but drawing relationships between solid-state structures and
mesomorphism is fraught with danger. Nonetheless, it is often
instructive to look at these solid-state structures as there can be
insights to be gained.
The 1H NMR spectra of the new complexes at room temperature
in CDCl3 solution (see ESI‡) show all the expected ligand signals
slightly modified in each series of compounds. Some resonances
appear to be overlapped or masked by other signals. The main
difference in each series of compounds is the integration of the
(CH2)n signal. In all cases the 1H NMR spectra display four groups
of signals corresponding to the allyl, pyridyl, alkyloxyphenyl and
pyrazolyl groups. For the allyl groups five resonances are observed
corresponding to the meso, syn and anti protons, the last two being
split in two signals due to the unsymmetric environment produced
by the pyrazolylpyridine ligand. In general those syn and anti
signals are very broad and overlapped each other or with the OCH2
groups of the lateral chains showing the dynamic character of the
allyl group. There are many examples in the literature describing
the dynamic behaviour of allylpalladium complexes.34–37 In the
complexes studied herein, those signals were well resolved at room
temperature in three compounds. So, for I14 and I16, four doublets
were observed at 4.58 and 3.39 ppm (Hsyn), and 3.65 and 3.02 ppm
(Hanti), while for II16 these signals appear slightly shifted at higher
field (4.43 and 3.40 ppm for Hsyn, 3.59 and 3.02 ppm for Hanti).
In same other cases these four signals could also be observed at
lower temperatures (for instance, at −50 ◦C for II18), but not well
resolved.
Following these considerations the X-ray structures of
3
complexes [Pd(g -C3H5)(pzbp2py)][X] (pzbp2py
=
2-[3,5-bis(4-
−
−
butoxyphenyl)pyrazol-1-yl]pyridine; X = BF4 (I4), CF3SO3
(III4)) were solved and reported recently by us.38 We consider it
of interest to comment herein on the main relevant results as
well as, for comparative purposes, those from the X-ray structure
of the related complex [Pd(g -C3H5)(pzhppy)][BF4] (pzhppy = 2-
3
[3-(4-hexyloxyphenyl)pyrazol-1-yl]pyridine) (IV), which we also
reported recently.19 Firstly, it should be mentioned that the three
complexes, I4, III4 and IV, exhibited a related molecular shape
which can be defined as rod-like type with respect to the molecular
length-to-width ratio (Fig. 1).
3
On
a
supramolecular level, the structure of [Pd(g -
C3H5)(pzhppy)][BF4] (IV) was consistent with a one-dimensional
assembly produced through weak hydrogen-bonds between the
counteranion BF4 and the cations [Pd(g -C3H5)(pzhppy)]+, so
defining a zigzag chain almost parallel to the b axis.19 In addition,
extended hydrogen bonds hold together the chains in a two-
dimensional network, in which molecules are arranged in bilayers
with the chains interdigitated between them (Fig. 2).19 Therefore
we could deduce that the counteranion and the pzhppy ligand
function as a source of hydrogen-bonds which generated a two-
dimensional supramolecular assembly.
−
3
Concerning the pyridyl group, a doublet is observed for the H6
−
proton at ca. 8.89 ppm for complexes I and III (with BF4 and
−
CF3SO3 as counteranions) being shifted upfield by ca. 0.15 ppm
−
for the PF6 complexes (II). A similar behaviour is also shown
The literature of calamitic ionic liquid crystals28 is dominated
by the observation of SmA phases (although refs. 29 and 32
for the H5 proton, this signal appearing partially masked by the
Ho protons of one C6H4 group. The signals corresponding to H3
This journal is
The Royal Society of Chemistry 2006
Dalton Trans., 2006, 3918–3926 | 3921
©

View Article Online
As indicated earlier, the thermal behaviour of the complexes
3
[Pd(g -C3H5)(pzRpy)][BF4] (R = C6H4OCnH2n+1, n = 6, 10, 12,
14, 16, 18) was investigated,19 and those compounds with
12–18 carbon atoms in the chain exhibited SmA mesophases
at high temperatures. The X-ray diffraction study at variable
3
temperature of [Pd(g -C3H5)(pzhdppy)][BF4] (pzhdppy = 2-[3-(4-
hexadecyloxyphenyl)pyrazol-1-yl]pyridine), as a representative
example, was consistent with a smectic type layered structure
containing the molecules organised in interdigitated bilayers.19
Therefore, there were some parallels between the arrangement in
the mesophase and the structure observed in the crystalline solid
state.
3
In our current work the complexes studied, [Pd(g -
C3H5)(pzR2py)][X], are related to those containing pzRpy as
ligand and from the structural point of view should have a
similar molecular shape. Thus, it was felt that the presence of
two substituents at the 3- and 5-positions of the pyrazole ring
should not modify the molecular skeleton appreciably or the
solid-state arrangement involving the counteranions. However
the potential two-dimensional supramolecular assembly could be
more affected.
3
In fact, the X-ray single crystal structures of [Pd(g -
C3H5)(pzbp2py)][X] (X = BF4 (I4), CF3SO3 (III4)) showed the
presence of intermolecular H · · · O/F hydrogen-bonds between
the cationic metal complexes and the counteranions, which were
responsible for zigzag chains in a one-dimensional assembly.38
In addition and concerning the counteranions, new H · · · O/F
interactions involved interchain interactions that gave rise to layer-
like arrangements, defining a two-dimensional network. In these
layer-like structures, the flexible chains are interdigitated (Fig. 3).38
−
−
Fig. 1 (a) ORTEP plot of I4,38 and (b) ORTEP plot of IV,19 in which a
rod-like shape could be suggested (the structure of III4 is similar to that
of I4)38.
Fig. 2 Two-dimensional network of IV,19 showing the interdigitation
between bilayers.
contain exceptions from metallomesogen chemistry), and the
formation of the smectic layers is undoubtedly promoted by
intermolecular Coulombic attractions. In these complexes, the
solid-state structure shows evidence of intermolecular hydrogen
bonding, too, but consideration of the thermal behaviour of these
complexes (vide infra) suggests that these forces influence the
stability of the crystalline phases rather than the liquid-crystal
phases.
Fig. 3 Two-dimensional network of I4,38 showing the interdigitation of
the chains between layers.
3922 | Dalton Trans., 2006, 3918–3926
This journal is
The Royal Society of Chemistry 2006
©

View Article Online
3
Table 2 Phase properties of pzR2py and [Pd(g -C3H5)(pzR2py)][X] (In, IIn,
These structural features were similar to those of the re-
3
IIIn) determined by optical microscopy and DSC
lated compound [Pd(g -C3H5)(pzhppy)][BF4], mentioned above,
for which the hydrogen-bonding C–H · · · F interactions along
Compound
Transition
T/◦C
DH/kJ mol−1
19
˚
˚
(3.08 A) and between (3.27 A) chains compared well with
3
pzdp2py
pzddp2pya
pztdp2pya
pzhdp2pya
pzodp2pya
I10
Cr → I
68.2
71.0
(44.1)
(48.8)
52.1
84
126
76
148
(108)
78
148
(139)
76
142
145
157
85
137
162
149
53
48.4
47.3
(−4.4)
(−5.7)
59.6
0.5
37.6
˚
the values of 3.06 and 3.16 A, respectively, observed in [Pd(g -
C3H5)(pzbp2py)][BF4].38 Then we deduce that, in cationic al-
lylpalladium(II) complexes containing pzR2py ligands, the 3,5-
disubstitution at the pyrazole ring does not significantly modify
the molecular assembly with respect to the analogous compounds
containing only one substituent at the 3-position on the pyrazole,
being in both cases described as the layer-like type in which the
layers have formed a molecular arrangement defining a bilayer
(Fig. 4). As a consequence, the number of chains interdigitated
between layers is higher, leading to an increased filling of interlayer
space.
Cr → I
(I → SmA)
(I → SmA)
Cr → I
Cr → Crꢀ
Crꢀ → I
I12
I14
I16
Cr → Crꢀ
Crꢀ → Crꢀꢀ → I
(I → SmA)
Cr → Crꢀ
Crꢀ → I
20.7
48.1b
(−1.3)
6.2
57.4
(I → SmA)
Cr → Crꢀ
Crꢀ → Crꢀꢀ
Crꢀ → SmA
SmA → I
Cr → Crꢀ
Crꢀ → SmA
SmA → I
Cr → I
(−1.8)
13.8
56.2b
2.0
12.8
56.2
2.1
I18
II12
II14
28.7
50.5
Cr → SmA
SmA → Crꢀ
Crꢀ → I
82
−15.2
132
(96)
48
19.2
(I → SmA)
Cr → Crꢀ
Crꢀ → Crꢀꢀ
Crꢀꢀ → I
(−1.5)
10.6
9.0
II16
II18
86
138
(114)
63
18.0
(I → SmA)
Cr → SmA
SmA → Crꢀ
Crꢀ → I
(−1.9)
72.2
88
−12.7
19.0
134
(124)
88
123
98
122
(92)
113
123
(107)
119
(115)
47
(I → SmA)
Cr → Crꢀ
Crꢀ → I
(−1.9)
III12
III14
5.7
54.4
8.3
56.1
Cr → Crꢀ
Crꢀ → I
(I → SmA)
Cr → Crꢀ
Crꢀ → I
(−1.5)
III16
III18
9.0
59.0
(I → SmA)
Cr → I
(−1.8)
81.2
(I → SmA)
SmA → Cr
Cr → SmAc
SmA → Ic
(−2.1)
−21.1
27.1
52
117
1.9
a Data from the pyrazolylpyridine ligands are given for comparative
purposes and taken from ref. 18. b The peaks overlapped; combined
enthalpy values are given. c Second heating.
calorimetry (DSC) and, in some cases, X-ray diffraction at variable
3
temperature. The complete mesomorphism of complexes [Pd(g -
C3H5)(pzR2py)][X] (R = C6H4OCnH2n+1, n = 10, 12, 14, 16, 18; X =
−
−
−
BF4 (I), PF6 (II), CF3SO3 (III)) is reported in Table 2.
3
Fig.
4
Schematic representation of the bilayers of [Pd(g -
The complexes I–III displayed only one type of mesophase,
namely SmA mesophase, which was recognised on the basis of
the optical texture, and so a typical oily streak texture was seen
on cooling as well as batonnets that sometimes formed from the
isotropic liquid (Fig. 5 and ESI,‡ Fig. S1). It is also worth noting
that on a number of occasions, the texture appeared very strongly
homeotropic. However, the overall thermal behaviour was not
straightforward and is described now in more detail.
C3H5)(pzRpy)][X] (a) and [Pd(g -C3H5)(pzR2py)][X] (b) (the values of 43.3
(a) and 37.2 A (b) correspond to the d001 spacing from the mesophases for
[Pd(g -C3H5)(pzhdppy)][BF4] (IV) and [Pd(g -C3H5)(pzodp2py)][BF4] (I18),
3
˚
3
3
˚
respectively, and those of 34 (a) and 68 A (b) to the calculated molecular
length of the same compounds).
3.4 Thermal studies
The mesomorphism of the new complexes was studied by po-
larised light optical microscopy (POM), differential scanning
The shortest-chain homologue of complexes I, I10, was not
mesomorphic and simply melted to the isotropic state at 126 ◦C,
This journal is
The Royal Society of Chemistry 2006
Dalton Trans., 2006, 3918–3926 | 3923
©

View Article Online
Complex I14 was a little more straightforward in that following
a crystal–crystal transition at 78 ◦C, the complex melted directly
to isotropic at 148 ◦C and a monotropic SmA phase formed at
139 ◦C, which persisted down to room temperature. Re-heating the
SmA phase led◦to partial crystallisation, the remaining SmA areas
cleared at 139 C, while the crystalline parts melted to isotropic
at 146 ◦C. Complex I16 also showed crystal–crystal transitions (76
and 142 ◦C) before melting into a SmA phase at 145 ◦C. The
◦
liquid-crystal phase persisted to 157 C whereupon it cleared to
give◦an isotropic liquid. Cooling the SmA phase was formed at
155 C and persisted down to room temperature. On re-heating,
partial crystallisation was observed, the solid parts re-melted at
146 ◦C while the SmA phase all cleared at 155 ◦C.
In common with all complexes containing octadecyloxy chains,
I18 was the most straightforward and showed an enantiotropic
◦
SmA phase between 137 and 162 C. Cooling once more led to
the SmA phase at 162 C, which was maintained down to room
Fig. 5 Textures of the smectic A mesophase, observed on cooling, for II18
◦
at 111 ◦C.
temperature, and on re-heating the SmA phase cleared at 162 ◦C.
For complexes II, once more the shortest-chain homologue
was non-mesomorphic and II12 melted directly to the isotropic
phase◦at 149 ◦C. Complex II14 was more interesting and melted
at 53 C to give a SmA phase. However, on further heating, the
SmA phase crystallised at 82 ◦C, and the event is recorded clearly
in the DSC trace (Fig. 7). Further heating to 132 ◦C led this
new crystalline phase to melt to the isotropic liquid. On cooling,
nothing was seen until 96 ◦C when the SmA phase re-formed,
which persisted down below 50 ◦C. Reheating the sample showed
that crystallisation had not occurred and all that was observed
was the clearing transition of the SmA phase at 96 ◦C (ESI,‡ Fig.
S3). The same pattern of behaviour was also seen for II18 where
melting to SmA first occurred at 63 ◦C, with the SmA phase
subsequently crystallising at 88 ◦C. The new crystalline phase
melted to the isotropic liquid at 134 ◦C and cooling revealed a
monotropic SmA phase at 124 ◦C. On cooling II18, crystallisation
of the SmA phase was seen at 45 ◦C, but this was almost certainly
incomplete as on re-heating, melting to SmA was observed at the
slightly lower temperature of 52 ◦C with, more significantly, a
much lower enthalpy of transition (DH = 72.2 kJ mol−1 on the
first heating compared with 28.4 kJ mol−1 on the second heating).
The second heating showed no evidence of cold crystallisation of
following a crystal–crystal transition at 84 ◦C. There was no
evidence for any monotropic mesophase on cooling. Complex I12
◦
also showed a crystal–crystal transition, this time at 76 C, and
then from around 140 ◦C there was a complex series of events. The
DSC trace for this area is reproduced as Fig. 6 and shows what we
believe to be a melting event followed rapidly by recrystallisation,
followed by melting once more—this time to isotropic liquid. This
assignment is based on the fact that the peak for the first melting
event at about 147 ◦C, is not symmetric, and the steeper slope to
high temperature rather suggests a catastrophic event consistent
with recrystallisation. In the absence of an identifiable texture by
microscopy, we were unable to assign the phase formed during this
first melting event. When the isotropic phase was cooled, a readily
identifiable SmA phase did form at 108 ◦C and persisted down to
room temperature. Crystallisation was not observed and so on re-
heating, both optical microscopy and DSC show a simple clearing
◦
event at 108 C (see ESI,‡ Fig. S2). However, heating above this
temperature showed a small, residual peak at about 146 ◦C which
we attribute to the melting of some crystal that did re-form during
a slow crystallisation process that took during the cooling and
re-heating cycle. The third heat–cool cycle reproduces the second
exactly.
◦
the SmA phase, which persisted to clearing at 123 C. Complex
−
II16 showed a pattern observed in the BF4 complexes in that
following crystal–crystal transitions at 48 and 86 ◦C, the complex
melted directly to the isotropic phase at 138 ◦C and showed a
monotropic SmA phase at 114 ◦C, which persisted down to room
temperature. DSC data for II16 and II18 are reproduced as Fig. S4
and S5, respectively (see ESI‡).
Complex III12 was non-mesomorphic simply melting at 123 ◦C,
but the remaining three homologues showed a SmA phase. For
III14 and III16, this was monotropic with melting points observed
at 122 and 123 ◦C, respectively, and clearing points at 92 and
107 ◦C, respectively. Both complexes also showed two crystal
polymorphs, and this behaviour was confirmed by observations
made by DSC. On cooling, the mesophase in both compounds was
maintained down below 50 ◦C with no evidence of crystallisation,
indeed the viscosity suggested that the system had formed a
smectic glass. Complex III18 melted directly to the isotropic
phase at 119 ◦C and the SmA phase appeared monotropically
at 115 ◦C. Crystallisation of the SmA phase was around 47 ◦C and
Fig. 6 DSC trace of I12 in the 120–160 ◦C range (first heating).
3924 | Dalton Trans., 2006, 3918–3926
This journal is
The Royal Society of Chemistry 2006
©

View Article Online
can deduce that the pzR2py ligands have allowed us to obtain
3
more stable mesophases upon complexation to [Pd(g -C3H5)]+
than those with pzRpy.
An interesting feature of three of these complexes is the presence
of more than one crystal polymorph in several samples. By this we
do not refer to the various crystal–crystal transition mentioned in
the text, rather the observation that some complexes (e.g. II18) show
◦
melt◦ing points at two different temperatures (63 C to SmA and
134 C to isotropic for II18). In II14 and II18, cold crystallisation
of the SmA phases is observed when the phase forms from the
lower-melting of the two solid phases, showing clearly that the
higher-melting phase is thermodynamically more stable. Thus, the
thermodynamically less stable phase is obtained from solution
crystallisation. However, for III18, the opposite seems to be true
as the higher-melting solid phase is seen on first heating (Cr–I at
119 ◦C), while the lower-melting modification appears on second
heating with no observed cold crystallisation (Cr–SmA at 52.5 ◦C).
Thus, the liquid crystal range of these complexes is potentially
rather wide, but is suppressed by the high melting point of the
more thermodynamically stable crystal phase.
In order to confirm the smectic nature of the mesophases, X-ray
diffraction experiments at variable temperature were undertaken
for representative examples, namely I18, II18 and III18.
The X-ray diffractograms for I18 agree with the observations by
POM, so that the sample enters to the mesophase between 135
and 140 ◦C and clears at slightly above 165 ◦C. On cooling the
phase reappears at 162 ◦C, but crystallisation is not observed
even at 50 ◦C. The pattern shown in Fig. 8 was recorded at
140 ◦C on heating the sample. Two, sharp, small-angle reflections
Fig. 7 DSC trace of II14 in the 50–100 ◦C range (first heating).
on re-heating the solid, the compound melted directly to a SmA
phase at 52 ◦C which persisted to the clearing point, measured
◦
at 117 C on heating. Once more, these events were matched by
corresponding events observed in the DSC trace. Thus, once more
there are two accessible crystal forms giving rise to the SmA phase
at different temperatures. DSC traces for III14, III16 and III18 are
shown as Fig. S6–S8 (see ESI‡).
In comparing the mesomorphism of the different series of
materials, we first note that the decyloxy, dodecyloxy and octade-
cyloxy ligands were non-mesomorphic, while the tetradecyloxy
and hexadecyloxy homologues showed monotropic SmA phases
with clearing points below 50 ◦C (Table 2). Thus, complexation
leads to materials whose mesophases are much more thermally
stable, clearing in all cases at temperatures above 90 ◦C, and in
some cases above 150 ◦C.
˚
were observed at 37.2 (d001) and 18.5 (d002) A corresponding to
a layered structure. In addition, a broad halo was located at
˚
4.5 A indicating the ligand-like order of the molten chains and
thus a disordered phases. Furthermore, the lattice spacing is
almost constant, changing by less than 5%, within the bounds
of experimental error, with temperature.
Within a given series of complexes, the melting points of
the mesomorphic homologues are relatively invariant with chain
length (e.g. between 137 and 148 ◦C for I, between 132 and 138 ◦C
for II and between 119 and 123 ◦C for III, ignoring the lower-
melting polymorphs discussed below), while the clearing point of
the SmA phases increases in a rather pronounced manner with
chain length, the one exception being II18 whose clearing point is
lower than that of II16. These observations suggest that the melting
transition may well be controlled by some of the intermolecular
interactions and ionic forces discussed above and that these can
exert a great effect than that of chain length alone. However, chain
length does control the stability of the liquid crystal mesophase
and the fairly sharp increase in clearing point with n suggests that
the increasing chain length helps to overcome the steric effect of
3
the g -allyl group, which will act to reduce the overall anisotropy
of the system.
Comparing the clearing points of the three series shows that
those of the smallest anion (BF4−), I, are the highest while those
of the largest anion (CF3SO3−), III, are the lowest, which suggests
that the anion size contributes negatively to the anisotropy of the
materials.
3
Fig. 8 X-Ray diffractogram of [Pd(g -C3H5)(pzodp2py)][BF4] (I18) at
140 ◦C.
Similar patterns were recorded for II18 and III18, with spacing of
˚ ˚
Comparison between the thermal behaviour of these new
d001 of 38.8 and 35.3 A and d002 of 19.6 and 17.7 A, respectively, with
3
complexes with the monosubstituted analogues,19 [Pd(g -
the broad halo at 4.4 A (data taken from the diffractograms at 78
˚
◦
C3H5)(pzRpy)][X] is, unfortunately, not directly possible as the
SmA phases of the latter were at rather high temperature and
decomposition precludes a meaningful comparison. However we
and 110 C on cooling, respectively). However, the solidification
of these compounds was observed at ca. 50 ◦C, and thus, the
diffractogram at lower temperatures corresponds to a solid phase.
This journal is
The Royal Society of Chemistry 2006
Dalton Trans., 2006, 3918–3926 | 3925
©

View Article Online
For I18, the length of the cationic part in the most extended
3 S. R. Collinson and D. W. Bruce, in Transition Metals in Supramolecular
Chemistry, ed. J. P. Sauvage, Wiley, Chichester, 1999, ch. 7, p. 285.
4 B. Donnio, Curr. Opin. Colloid Interface Sci., 2002, 7, 371.
5 K. Binnemans and C. Go¨rler-Walrand, Chem. Rev., 2002, 102, 2303.
6 B. Donnio, D. Guillon, R. Deschenaux and D. W. Bruce, Comprehensive
Coordination Chemistry II, ed. J. A. McCleverty and T. J. Meyer,
Elsevier, Oxford, UK, 2003, vol. 7, ch. 7.9, p. 357.
7 B. Donnio, D. Guillon, R. Deschenaux and D. W. Bruce, Comprehensive
Organometallic Chemistry III, ed. R. H. Crabtree and D. M. P. Mingos,
Elsevier, Oxford, UK, in press.
8 R. Deschenaux and J. W. Goodby, in Ferrocene: Homogeneous Cataly-
sis, Organic Synthesis, Materials Science, ed. A. Togni and T. Hayashi,
VCH, Weinheim, 1995, ch. 9, p. 471.
9 J. P. Rourke, F. P. Fanizzi, D. W. Bruce, D. A. Dunmur and P. M. Maitlis,
J. Chem. Soc., Dalton Trans., 1992, 3009.
10 (a) L. Ziminski and J. Maltheˆte, J. Chem. Soc., Chem. Commun., 1990,
1495; (b) P. Jacq and J. Maltheˆte, Liq. Cryst., 1996, 21, 291.
11 E. Campillos, R. Deschenaux, A. M. Levelut and R. Ziessel, J. Chem.
Soc., Dalton Trans., 1996, 2533.
12 M. C. Torralba, M. Cano, J. A. Campo, J. V. Heras, E. Pinilla and M. R.
Torres, J. Organomet. Chem., 2001, 633, 91.
13 M. C. Torralba, M. Cano, J. A. Campo, J. V. Heras, E. Pinilla and M. R.
Torres, J. Organomet. Chem., 2002, 654, 150.
14 M. C. Torralba, M. Cano, J. A. Campo, J. V. Heras, E. Pinilla and M. R.
Torres, Inorg. Chem. Commun., 2002, 5, 887.
15 M. C. Torralba, M. Cano, S. Go´mez, J. A. Campo, J. V. Heras, J. Perles
and C. Ruiz-Valero, J. Organomet. Chem., 2003, 682, 26.
16 M. C. Torralba, P. Ovejero, M. J. Mayoral, M. Cano, J. A. Campo, J. V.
Heras, E. Pinilla and M. R. Torres, Helv. Chim. Acta, 2004, 87, 250.
17 M. C. Torralba, M. Cano, J. A. Campo, J. V. Heras and E. Pinilla,
Z. Kristallogr.-New Cryst. Struct., 2005, 220, 615.
18 M. J. Mayoral, M. C. Torralba, M. Cano, J. A. Campo and J. V. Heras,
Inorg. Chem. Commun., 2003, 6, 626.
19 M. C. Torralba, M. Cano, J. A. Campo, J. V. Heras, E. Pinilla and M. R.
Torres, J. Organomet. Chem., 2006, 691, 765.
20 M. Ghedini and A. Crispini, Comments Inorg. Chem., 1999, 21, 33.
21 M. Ghedini, D. Pucci, A. Crispini, I. Aiello, F. Barigelletti, A. Gessi
and O. Francescangeli, Appl. Organomet. Chem., 1999, 13, 565.
22 J. Buey, G. A. D´ıez, P. Espinet, S. Garc´ıa-Granda and E. Pe´rez-Carren˜o,
Eur. J. Inorg. Chem., 1998, 1235.
23 M. J. Baena, J. Buey, P. Espinet and C. E. Garc´ıa-Prieto, J. Organomet.
Chem., 2005, 690, 998.
24 L. D´ıez, P. Espinet, J. A. Miguel and M. P. Rodr´ıguez-Medina,
J. Organomet. Chem., 2005, 690, 261.
25 F. Neve, M. Ghedini and A. Crispini, Chem. Commun., 1996, 2463.
26 F. Neve, M. Ghedini, A. Francescangeli and S. Campagna, Liq. Cryst.,
1998, 24, 673.
crystalline conformation, calculated on the basis of the X-ray
3
structure of [Pd(g -C3H5)(pzbp2py)][BF4],38 was estimated to be of
˚
˚
ca. 68 A, and the value of the spacing was 37.2 A. Therefore these
results suggest significant interdigitation of the layers. Similar
features can also be suggested for II18 and III18.
An interesting comparison is that of the spacing (d001) of
related complexes containing mono or disubsituted pyrazole
rings, i.e. pzRpy and pzR2py, and the same counteranion BF4
.
−
So, d001 for [Pd(g -C3H5)(pzhdppy)][BF4] (V) was 43.3 A, and a
3
˚
˚
molecular length, l, of ca. 34 A was estimated in the fully extended
configuration.19 Because the d spacing was larger than l but smaller
than 2l, an interdigitated bilayer structure related to that observed
in the solid was proposed for these complexes (Fig. 4).19 For I18, it
is possible to suggest a similar structural arrangement, so that the
˚
molecular length of ca. 68 A is now close to twice the observed
˚
d-spacing of 37.2 A. Thus, we can confidently propose for I18 that
the molecules are arranged in interdigitated bilayers and we note
that such an arrangement exists also in the solid state (Fig. 4).
Similar deductions can be made for complexes II18 and III18.
Conclusions
We have described the synthesis and thermal behaviour of a series
of ionic palladium(II) allyl complexes bound to a series of difunc-
tionalised pyrazolylpyridine ligands, pzR2py. Consistent with their
ionic nature, the complexes show SmA mesophases and the trends
in transition temperature point to a crystalline phase who stability
is determined largely by intermolecular Coulombic and hydrogen-
bonding interactions. The stability of the mesophase, however, is
governed first by the size of the anion (larger anions give less
stable SmA phases) and by the length of the flexible alkyl chains
(longer chains give more stable phases). An important feature of
these materials is the crystalline polymorphism shown by some
complexes so that some appear to show two melting points due to
the metastability of the crystalline phase obtained from solution
crystallisation.
27 D. Pucci, G. Barberio, A. Crispini, M. Ghedini and O. Francescangeli,
Mol. Cryst. Liq. Cryst., 2003, 395, 155.
28 K. Binnemans, Chem. Rev., 2005, 105, 4148.
29 F. Neve, Adv. Mater., 1996, 8, 277.
Acknowledgements
30 M. Ghedini and D. Pucci, J. Organomet. Chem., 1990, 395, 105.
31 A. El-ghayoury, L. Douce, A. Skoulios and R. Ziessel, Angew. Chem.,
Int. Ed., 1998, 37, 1255.
We are grateful to the Direccio´n General de Investigacio´n/MEC
of Spain (Project No. BQU2003-07343) for financial support. We
also thank to Comunidad de Madrid for financial support (Project
GR/MAT/0511/2004) and the contract to M. C. T. We thank
Dr B. Donnio (Institut de Physique et Chimie de Mate´riaux de
Strasbourg IPCMS, Universite´ de Strasbourg, France) for the X-
ray diffraction experiments at variable temperature for compound
I18 and for his helpful assessment in the interpretation of these
XRD data.
32 D. W. Bruce, Acc. Chem. Res., 2000, 33, 831.
33 J. A. Campo, M. Cano, J. V. Heras, M. C. Lagunas, J. Perles, E. Pinilla
and M. R. Torres, Helv. Chim. Acta, 2002, 85, 1079.
34 F. Go´mez-de la Torre, A. de la Hoz, F. A. Jalo´n, B. R. Manzano, A.
Otero, A. M. Rodr´ıguez and M. C. Rodr´ıguez-Pe´rez, Inorg. Chem.,
1998, 37, 6606.
35 F. Go´mez-de la Torre, A. de la Hoz, F. A. Jalo´n, B. R. Manzano, A. M.
Rodr´ıguez, J. Elguero and M. Mart´ınez-Ripoll, Inorg. Chem., 2000, 39,
1152.
36 J. Elguero, A. Fruchier, A. de la Hoz, F. A. Jalo´n, B. R. Manzano, A.
Otero and F. Go´mez-de la Torre, Chem. Ber., 1996, 129, 589.
37 F. A. Jalo´n, B. R. Manzano and B. Moreno-Lara, Eur. J. Inorg. Chem.,
2005, 100.
38 M. C. Torralba, M. Cano, J. A. Campo, J. V. Heras, E. Pinilla, M. R.
Torres, J. Perles and C. Ruiz-Valero, J. Organomet. Chem., 2006, 691,
2614.
References
1 J. L. Serrano, Metallomesogens: Synthesis, Properties and Applications,
VCH, Weinheim, 1995.
2 B. Donnio and D. W. Bruce, Struct. Bonding (Berlin), 1999, 95, 193.
3926 | Dalton Trans., 2006, 3918–3926
This journal is
The Royal Society of Chemistry 2006
©