Influence of the lanthanide contraction on the transition temperatures of rare-earth containing metallomesogens with Schiff base ligands

Article abstract of DOI:10.1016/S0009-2614(98)01384-0

We have synthesised liquid-crystalline rare-earth complexes with Schiff base ligands. The stoichiometry of these complexes is Ln(LH)₃(NO₃)₃, where Ln is a trivalent rare-earth ion (Y³⁺, La³⁺-Lu³⁺

Full text of DOI:10.1016/S0009-2614(98)01384-0

5 February 1999
Ž
.
Chemical Physics Letters 300 1999 509–514
Influence of the lanthanide contraction on the transition
temperatures of rare-earth containing metallomesogens with
Schiff base ligands
a
b
c
K. Binnemans a,), R. Van Deun , D.W. Bruce , Yu.G. Galyametdinov
a
K.U. LeuÕen, Department of Chemistry, Coordination Chemistry DiÕision, Celestijnenlaan 200F, B-3001 HeÕerlee, Belgium
b
UniÕersity of Exeter, School of Chemistry, Stocker Road, Exeter EX4 4QD, UK
c
Physical–Technical Institute, Russian Academy of Science, Sibirsky Tract 10r7, 420029 Kazan, Russian Federation
Received 5 June 1998; in final form 2 December 1998
Abstract
We have synthesised liquid–crystalline rare-earth complexes with Schiff base ligands. The stoichiometry of these
3
complexes is Ln LH NO3 3, where Ln is a trivalent rare-earth ion Y q, La3q–Lu3q, except Ce3q and Pm3q , LH is the
Ž
. Ž
.
Ž
.
3
protonated form of the Schiff base 4-alkoxy-N-octadecyl-2-hydroxybenzaldimine. The mesophases were studied by optical
Ž
.
microscopy and by differential scanning calorimetry DSC . For the complexes with the 4-octyloxy-N-octadecyl-2-hydroxy-
benzaldimine ligand, a gradual increase in the melting point and a gradual decrease in the clearing point is found when going
q
from La3 to Lu3q. In this way, the mesophase range narrows over the lanthanide series. This is due to the lanthanide
contraction. The textures observed by optical microscopy reveal the presence of a smectic A mesophase. q 1999 Elsevier
Science B.V. All rights reserved.
.
1. Introduction
4 . In this way, the flat molecular structure of the
ligands is retained and even extended, so that the
structural anisotropy necessary for the formation of
the mesophase is not destroyed. Square planar coor-
dination is however only observed for a limited
number of transition metal ions, and it is desirable to
incorporate other transition metal ions into liquid
crystals in order to take full advantage of the unique
physical properties of these elements.
The investigation of liquid–crystalline metal com-
Ž
.
plexes metallomesogens is a fast growing branch of
w
x
coordination chemistry 1–7 . Metallomesogens are
not restricted to mesomorphic ligands, because the
metal ion can induce mesomorphism in non-meso-
morphic ligands. In the past, the majority of the
Ž .
newly synthesised metallomesogens contained Cu II ,
Ž .
Ž .
Ž .
Ni II , Pt II and Pd II as the central metal ion,
since the coordination of the central metal ion in the
Only a few examples of lanthanide-containing
metallomesogens have been described in the litera-
ture. The first mesomorphic lanthanide compounds
Ž
complexes is square planar coordination numbers
Ž
.
were the substituted bis phthalocyaninato lutetium
w x
Ž
.
III complexes, described by Piechocki et al. 8 ;
these compounds exhibit columnar mesophases.
)
Corresponding author. Tel.: q32-16-32-74-46; fax: q32-16-
32-79-92; e-mail: koen.binnemans@chem.kuleuven.ac.be
Later, Galyametdinov and coworkers described
0009-2614r99r$ - see front matter q 1999 Elsevier Science B.V. All rights reserved.
Ž
.
PII: S0009-2614 98 01384-0

(
)
510
K. Binnemans et al.rChemical Physics Letters 300 1999 509–514
Ž
.
calamitic lanthanide complexes with Schiff bases
element 39 yttrium . In general, the properties of the
3d-element scandium significantly differ from the
properties of the other rare-earth elements. There-
fore, we do not consider scandium in this systematic
study.
w
x
w x
9,10 and b-enaminoketones 11 as the ligands,
showing a smectic A mesophase. We showed that
the transition temperatures of the Schiff base com-
plexes are greatly influenced by the choice of the
w
x
counter-ion 12 . Wang et al. reported complexes
with hemicyanine ligands, showing an unspecified
w
x
mesophase with a mosaic texture 13 .
2. Experimental details
Ž
.
The elements of the lanthanide series La–Lu are
chosen as the central metal ion for several reasons:
All chemicals were used as received, without
further purification. Hydrated rare earth nitrates were
purchased from Aldrich. 1H NMR spectra were
Ž .
Ž
1 Because of the high coordination number coor-
.
dination numbers8 or 9 , the coordination polyhe-
dra in lanthanide complexes are different from those
observed for transition metal complexes. This opens
new possibilities for investigation of the structure-
Ž
recorded on a Bruker AMX400 spectrometer 400
.
Ž
.
MHz . Elemental analyses CHN were performed
on a Perkin-Elmer 2400 Elemental Analyser. The
mesophases were investigated by optical microscopy
Ž .
mesomorphism relation, 2 the lanthanide ion can
Ž
.
and by differential scanning calorimetry DSC . Mi-
croscopy was performed on an Olympus BX60 po-
larising microscope, equipped with
a Linkam
TMS600 hot stage and a Linkam TMS93 tempera-
ture controller. DSC-thermograms were recorded on
a Mettler-Toledo DSC-821e module. The tempera-
Ž
ture and heat flow were calibrated by indium Tm s
156.6"0.38C; D Hs28.45"0.6 J gy1 . The heat-
.
ing and cooling rates were 108C miny1, except when
the melting and clearing peaks showed an overlap. In
those cases, the rate was reduced to 18C miny1. The
samples were sealed in a 40 ml aluminium pan with
a pierced lid. Dry nitrogen was used as the oven
atmosphere.
3. Synthesis of the ligand and complexes
Ž
.
and enthalpies of rare-earth containing liquid crys-
tals. The Schiff base we worked with is 4-octyloxy-
4-Octyloxy-2-hydroxybenzaldehyde was prepared
Ž
.
by refluxing for 3 h 2,4-dihydroxybenzaldehyde
Ž
.
Ž
.
Ž
N-octadecyl-2-hydroxybenzaldimine Fig. 1 . To
avoid confusion: the lanthanides are the elements
with an atomic number between 57 lanthanum and
71 lutetium . By rare earths we understand the
lanthanides together with element 21 scandium and
50 mmol, 6.91 g with 1-bromooctane 50 mmol,
9.68 g in DMF, and with KHCO3 50 mmol, 5.01 g
.
Ž
.
Ž
.
as the base. The raw aldehyde was transformed to
the Schiff base without further purification by reflux-
Ž
.
Ž
.
Ž
.
ing in absolute ethanol 3 h with octadecylamine
Ž
.
Fig. 1. The Schiff base ligand 4-octyloxy-N-octadecyl-2-hydroxybenzaldimine LH .

(
)
K. Binnemans et al.rChemical Physics Letters 300 1999 509–514
511
Ž
.
50 mmol, 13.48 g , using a few drops of glacial
4. Mesomorphic properties: results and discussion
acetic acid as the catalyst. The Schiff base 4-oc-
tyloxy-N-octadecyl-2-hydroxybenzaldimine was pu-
rified by recrystallising three times from absolute
The ligand 4-octyloxy-N-octadecyl-2-hydroxyben-
zaldimine itself does not show mesomorphism, melt-
ing directly to an isotropic liquid at 508C. However,
all the rare-earth complexes exhibit a mesophase.
The mesophase was identified as a smectic A phase
by optical microscopy, using a hot stage and po-
larised light. The sample was placed between two
untreated cover glasses. By cooling from the isotropic
Ž
.
ethanol. The overall yield was 46% 11.60 g . The
ligand was characterised by NMR and IR spec-
troscopy. The rare earth complexes were prepared by
dissolving the Schiff base in absolute ethanol and
adding a solution of the hydrated rare-earth nitrate in
absolute ethanol. Although a complex is formed in a
1:3 metal:ligand ratio, we used an excess of the
hydrated rare-earth nitrate, in order to take the uncer-
tainty of the water content of these reagents into
account. For the calculations, it was assumed that the
melt batonnets form, which coalesce to form a fo-
ˆ
cal–conic fan structure. In the same sample extinct
regions were also observed, indicating homeotropic
alignment of the molecules. This supports the identi-
fication as the smectic A phase. The mesophases
show a high viscosity, which is evident if one tries to
shear the cover glasses at a temperature close above
the melting point. At higher temperatures the viscos-
ity gradually decreases, and a strong increase in
Ž
.
compounds were hexahydrates Y, La–Sm or pen-
Ž
.
tahydrates Eu–Lu . The dissolution of the Schiff
base can be assisted by heating the ethanolic solu-
tion, but reproducible results for the complexes could
only be obtained if the solution is cooled beneath
508C before the solution of the rare-earth salt is
added dropwise. About 15 min after mixing, the
solution turns cloudy and the complex starts to pre-
cipitate. Stirring was continued for at least 3 h. After
filtration, the complex was washed with ice-cold
ethanol and dried in vacuo. The yield of the com-
plexes varied between 70 and 80%. The complexes
were characterised by CHN elemental analysis. No
reproducible results could be obtained for the cerium
complex, probably because of a redox reaction due
fluidity is observed at the clearing point. When
y1
Ž
.
cooling the mesophase 58C min , no crystallisa-
tion is observed in the microscope and the texture of
the mesophase is frozen into the solid state. A glassy
mesophase is formed. The glass formation is noticed
from the fact that it is no longer possible to shear the
cover glasses between which the sample is sand-
wiched. Moreover, when cooling further, conchoidal
fractures are observed in the texture. These con-
choidal fractures also indicate the presence of a
glassy state. The glass state in the mesophase is
closer to the glassy state in polymers than the real
inorganic glasses. In contradistinction to inorganic
glasses which are optically isotropic, the vitreous
mesophase is an anisotropic glass. It should be noted
that by a glass, we mean a structural glass, and not a
spin glass. The DSC traces show crystallisation at a
cooling rate of 108C miny1, although a strong super-
cooling is observed. This difference in behaviour
between the microscopic and DSC observations do
not contradict each other, because we are studying a
thin film under the microscope, whereas a bulk
sample is used for the DSC measurements. The
glassy film is transparent at room temperature. When
the aluminium crucible is opened after the DSC
measurement, the sample is translucent, indicating its
polycrystalline state. No crystallisation of the thin
films was observed, not even after a day. Galyamet-
Ž
.
to Ce III,IV . Promethium was excluded from the
study, because this element is radioactive.
w
x
Although Galyametdinov et al. 10 proposed that
the empirical formula of this type of complexes is
w
Ž
.
xŽ
.
Ln LH L NO3 2 , we now have strong evidence
2
Ž
w
. Ž
.
xŽ
that the empirical formula is Ln LH NO3 3. It is
Ž 3
.
.
difficult to differentiate between Ln LH L NO3
P3H2O and Ln LH NO3 by CHN microanalysis
2
2
Ž
. Ž
.
3
3
1
only. However, the H NMR spectrum of the lan-
thanum complex shows that all the ligands are iden-
tical and that the OH-group of the ligand remains
protonated. Recently, a single crystal X-ray structure
Ž
of an analogous non-mesomorphic derivative LHs
Ž
.
.
w x
CH3OC6 H3 OH CHsNC4 H9 was obtained 14 .
The crystal structure shows that three coordinating
nitrate anions are present and that all three nitrates
are bound in a bidentate, chelating fashion. The
Schiff base ligands are bound through the phenol
oxygen.
w
x
dinov and coworkers 10 describe supercooling of

(
)
512
K. Binnemans et al.rChemical Physics Letters 300 1999 509–514
similar complexes and they observed that the texture
is stable between y208C and 608C. Metallomeso-
gens, and especially ionic complexes, show a strong
tendency to form glassy mesophases. Among the
first to draw attention to this behaviour were Bruce
determined by DSC, whereas the clearing points
have been determined by optical microscopy. The
onset of the clearing peak could not be determined
for the complexes of the heavier rare-earth com-
plexes, because of the overlap of the melting and
clearing peaks. In all cases, it was possible to deter-
mine the clearing peak maximum in the heating run,
although at the end of the lanthanide series the
heating rate had to be reduced to 18C miny1 in order
to resolve the peaks. At a heating rate of 108C
miny1, clearing was only observed as a shoulder of
the melting peak. Microscopy shows that all the
complexes form a mesophase after melting, so that
all our compounds are enantiotropic liquid crystals.
The complexes in the series from lanthanum to
europium show a crystal–crystal transition, which is
absent in the second heating run. No crystal–crystal
transition was observed for the complexes of the
heavier lanthanides. Fig. 3 shows that the melting
temperatures smoothly increase when going from the
lanthanum to the lutetium complex. At the same
time, a decrease is observed for the clearing point.
Therefore, an overall trend is a narrowing of the
mesophase range over the lanthanide series. Since
the clearing point decreases, the mesophase stability
also decreases. The change in the transition tempera-
tures is more pronounced in the beginning of the
lanthanide series than at the end. Whereas an in-
crease in the melting point of 528C is observed when
going from the lanthanum complex to the lutetium
complex, the decrease of the clearing point is only
268C. These trends in mesomorphic behaviour can be
correlated to the ionic radius of the central rare-earth
ion. The radius of the lanthanide ion steadily de-
w
x
Ž .
et al. 15 , who observed glass formation for silver I
complexes. The potential of vitreous metallomeso-
w
x
gens is discussed in a review by Neve 16 .
The mesomorphic properties of the rare-earth
complexes are summarised in Table 1. The DSC
traces for the praseodymium and the erbium complex
are shown in Fig. 2. The reported temperatures were
Table 1
Mesomorphic properties of the rare-earth complexes Ln-
Ž
. Ž
.
LH NO3 3, where Ln represents the rare-earth ion and LH the
3
protonated Schiff base ligand 4-octyloxy-N-octadecyl-2-hydroxy-
benzaldimine
y1
Ln
Y
Transitiona
Temperatureb 8C
D H b kJ mol
Ž
.
Ž
.
Crys–SA
SA –I
Crys–SA
SA –I
Crys–SA
SA –I
Crys–SA
SA –I
Crys–SA
SA –I
Crys–SA
SA –I
Crys–SA
SA –I
Crys–SA
SA –I
Crys–SA
SA –I
Crys–SA
SA –I
Crys–SA
SA –I
Crys–SA
SA –I
Crys–SA
SA –I
Crys–SA
SA –I
132
141
83
165
90
22.7
11.0
2.5
11.8
8.7
La
Pr
163
95
13.2
12.4
11.8
12.4
12.4
15.5
13.3
16.9
12.7
17.6
10.6
21.4
11.8
22.2
10.8
Nd
Sm
Eu
Gd
Tb
Dy
Ho
Er
159
105
154
108
151
121
150
128
148
131
146
134
144
135
142
139
142
138
141
135
139
Ž
creases over the lanthanide series lanthanide con-
.
traction . The fact that a correlation exists between
c
the ionic radius and the mesomorphic properties of
the complexes is also illustrated by comparison of
the transition temperatures of the holmium complex
–
–
–
–
–
–
–
–
c
c
Tm
Yb
Lu
c
Ž
.
Crys™SA: 1348C, SA ™I: 1448C and the yttrium
c
Ž
.
complex Crys™SA: 1328C, SA ™I: 1428C . The
ionic radii of Ho3q and Y3q are nearly identical.
The ion gets smaller over the lanthanide series and
the charge remains, so that the electrostatic attraction
between the lanthanide ion and the coordinating
groups become stronger and therefore the melting
temperature increases. The rapid increase of the
melting point when going from the europium to the
c
c
c
aCrys: crystalline phase; SA: smectic A phase; I: isotropic liquid.
b The error on the temperature measurements is "0.58C and on
the heat flow 0.2 kJ moly1
.
c No transition enthalpies could be determined for the Er–Lu
complexes, because of overlapping peaks.

(
)
K. Binnemans et al.rChemical Physics Letters 300 1999 509–514
513
Ž
.
Ž
. Ž
.
Fig. 2. DSC curves of the praseodymium complex lower trace and erbium complex upper trace 2nd heating run . Endothermic peaks are
Ž
.
pointing upwards ^ endo .
gadolinium complex is most likely due to a differ-
ence in crystal structure between the complexes of
the light and those of the heavy lanthanides. It is
more difficult to explain the reduction of the clearing
point over the lanthanide series. Probably, the reduc-
tion in ion size leads to steric problems in the
mesophase and thus more structural distortions oc-
cur, leading to a less anisotropic complex and reduc-
tion of the clearing point.
not be determined separately for the complexes at
Ž
.
the end of the lanthanide series Er–Lu . The melt-
ing enthalpy exhibits the same trend as the melting
temperature: D Hm increases over the lanthanide se-
ries from 2.2 kJ moly1 for the lanthanum complex to
22.5 kJ moly1 for the holmium complex Table 1 .
However, no distinct trend is observed for the clear-
ing enthalpies. The values are rather constant and
Ž
.
scattered between 10.6 and 13.3 kJ moly1
.
Because of the overlap between melting and clear-
ing peaks, the melting and clearing enthalpy could
5. Conclusions
We have synthesised rare-earth complexes with
the Schiff base 4-octyloxy-N-octadecyl-2-hydroxy-
benzaldimine as the ligand, for all the rare-earth
ions, except scandium, cerium and promethium. The
Ž
. Ž .
complexes have the stoichiometry Ln LH NO3 ,
3
3
where Ln represents the trivalent rare-earth ion. All
the complexes form an enantiotropic mesophase. The
mesophase could be identified as a smectic A phase
Ž
.
SA by optical thermomicroscopy. The melting tem-
peratures smoothly increase when going from the
lanthanum to the lutetium complex, whereas a de-
crease is observed for the clearing point. Therefore,
the mesophase range narrows over the lanthanide
series. This trend can be correlated to the lanthanide
Fig. 3. Change of the mesomorphic properties of the complexes
Ž
. Ž
.
Ln LH NO3 over the lanthanide series.
3
3

(
)
514
K. Binnemans et al.rChemical Physics Letters 300 1999 509–514
contraction. The melting enthalpy follows the same
trend as the melting temperature, whereas the clear-
ing enthalpy is rather constant over the lanthanide
series. A thin film of the complexes supercools to a
glassy mesophase. Further work is in progress to
check whether the trends observed for the 4-oc-
tyloxy-N-octadecyl-2-hydroxybenzaldimine ligand
and the nitrate ion, are also present in ligands with
other chain lengths and for complexes with other
counter-ions.
References
w x
1
A.M. Giroud-Godquin, P.M. Maitlis, Angew. Chem. Int. Ed.
Engl. 30 1991 375.
Ž
.
w x
2
P. Espinet, M.A. Esteruelas, L.A. Oro, J.L. Serrano, E. Sola,
Coord. Chem. Rev. 117 1992 215.
Ž
.
w x
Ž .
3
w x
4
S.A. Hudson, P.M. Maitlis, Chem. Rev. 93 1993 861.
D.W. Bruce, J. Chem. Soc., Dalton Trans. 1993 2983.
A.P. Polishchuk, T.V. Timofeeva, Russian Chem. Rev. 62
Ž
.
w x
5
Ž
.
1993 291.
w x
Ž .
6
w x
7
J.L. Serrano Ed. , Metallomesogens, VCH, Weinheim, 1996.
Ž
.
D.W. Bruce, in: D.W. Bruce, D. O’Hare Eds. , Inorganic
Materials, Chap. 8, 2nd edn., Wiley, Chichester, 1996, p.
429.
w x
8
C. Piechocki, J. Simon, J.J. Andre, D. Guillon, P. Petit, A.
´
Acknowledgements
Ž
.
Skoulios, P. Weber, Chem. Phys. Lett. 122 1985 124.
Yu.G. Galyametdinov, G.I. Ivanova, I.V. Ovchinnikov, Bull.
w x
9
KB is Postdoctoral Fellow of the F.W.O.–Flanders
Ž
.
Acad. Sci. USSR, Div. Chem. Sci. 40 1991 1109.
Ž
.
Belgium . He thanks the F.W.O. for a ‘‘Krediet aan
Navorsers’’ and Prof. C. Gorller-Walrand for provid-
w
x
10 Yu.G. Galyametdinov, M.A. Athanassopoulou, K. Griesar,
O.A. Kharitonova, E.A. Soto-Bustamante, L. Tinchurina,
¨
Ž
.
I.V. Ovchinnikov, W. Haasse, Chem. Mater. 8 1996 922.
Ž
.
ing laboratory facilities GOA 98r03 and F.W.O.
research grant G.0124.95 . RVD specialisatiebursaal
w
w
w
w
w
w
x
11 Yu.G. Galyametdinov, O.A. Kharitonova, O.N. Kadkin, I.V.
Ž
Ž
.
Ovchinnikov, Russian Chem. Bull. 43 1994 1595.
.
IWT is indebted to the Flemish Institute for the
x
12 K. Binnemans, Yu.G. Galyametdinov, S.R. Collinson, D.W.
Encouragement of Scientific and Technological Re-
Ž
.
Bruce, J. Mater. Chem. 8 1998 1551.
Ž
.
x
search in the Industry IWT for financial support.
KB, RVD and DWB wish to thank the British Coun-
cil and the F.W.O.–Flanders for travel grants
13 K.Z. Wang, C.H. Huang, G.X. Xu, Q.F. Zhou, Solid State
Ž
.
Comm. 95 1995 223.
x
14 Yu.G. Galyametdinov, I. Litvinov, A. Polischuk, W. Haase,
A. Gubaydullin, to be published.
Ž
.
Academic Research Collaboration Programme .
DWB and YGG are receiving support from the
x
15 D.W. Bruce, B. Donnio, D. Guillon, B. Heinrich, M. Ibn-
Ž
.
Elhaj, Liq. Cryst. 19 1995 537.
Ž
.
x
Ž
.
INTAS-project contract INTAS 96-1198 .
16 F. Neve, Adv. Mater. 8 1996 277.