Does the solid-liquid crystal phase transition provoke the spin-state change in spin-crossover metallomesogens?

Article abstract of DOI:10.1021/ja077265z

Three types of interplay/synergy between spin-crossover (SCO) and liquid crystalline (LC) phase transitions can be predicted: (i) systems with coupled phase transitions, where the structural changes associated to the Cr?LC phase transition drives the spin-state transition, (ii) systems where both transifions coexist in the same temperature region but are not coupled, and (iii) systems with uncoupled phase transitions. Here we present a new family of Fe(II) metallomesogens based on the ligand tris[3-aza-4-((5-C _n)(6-R)(2-pyridyl))but-3-enyl]amine, with C_n = hexyloxy, dodecyloxy, hexadecyloxy, octadecyloxy, eicosyloxy, R = hydrogen or methyl (C_n-trenH or C_n-trenMe), which affords examples of systems of types i, ii, and iii. Self-assembly of the ligands C_n-trenH and C_n-trenMe with Fe(A)₂·xH₂O salts have afforded a family of complexes with general formula [Fe(C_n-trenR)](A) ₂· sH₂O (s > 0), with A = ClO₄ ^-, F^-, Cl^-, Br^- and I^-. Single-crystal X-ray diffraction measurements have been performed on two derivatives of this family, named as [Fe(C₆-trenH)](ClO ₄)₂ (C₆-1) and [Fe(C₆-trenMe)] (ClO₄)₂ (C₆-2), at 150 K for C₆-1 and at 90 and 298 K for C₆-2. At 150 K, C₆-1 displays the triclinic space group P1, whereas at 90 and at 298 K C₆-2 adopts the monoclinic P2₁/c space group. In both compounds the iron atoms adopt a pseudo-octahedral symmetry and are surrounded by six nitrogen atoms belonging to imino groups and pyridines of the ligands C_n-trenH and C _n-trenMe. The average Fe(II)-N bonds (1.963(2) A) at 150 K denote that C₆-1 is in the low-spin (LS) state. For C₆-2 the average Fe(II)-N bonds (2.007(1) A) at 90 K are characteristic of the LS state, while at 298 K they are typical for the high-spin (HS) state (2.234(3) A). Compound C₆-1 and [Fe-(C₁₈-trenH)](ClO ₄)₂ (C₁₈-1) adopts the LS state in the temperature region between 10 and 400 K, while compound C₆-2 and [Fe(C_n-trenMe)](ClO₄)₂ (n = 12 (C ₁₂-2), 18 (C₁₈-2)) exhibit spin crossover behavior at T_1/2 centered around 140 K. The thermal spin transition is accompanied by a pronounced change of color from dark red (LS) to orange (HS). The light-induced excited spin state trapping (LIESST) effect has been investigated in compounds C₆-2, C₁₂-2 and C ₁₈-2. The T_1/2^LIESST is 56 K (C₆-2), 48 K (C₁₆-2), and 56 K (C₁₈-2). On the basis of differential scanning calorimetry, optical polarizing microscopy, and X-ray diffraction findings for C₁₈-1, C₁₂-2, and C ₁₈-2 at high temperature a smectic mesophase S_x has been identified with layered structures similar to C₆-1 and C ₆-2. The compounds [Fe(C_n-trenH)](Cl)₂· sH₂O (n = 16 (C₁₆-3, s = 3.5, C₁₆-4, s = 0.5, C₁₆-5, s = 0), 18 (C₁₈-3, s = 3.5, C₁₈-4, s = 0.5, C₁₈-5, s = 0), 20 (C₂₀-3, s = 3.5, C₂₀-4, s = 0.5, C₂₀-5, s = 0)) and [Fe(C₁₈-tren)](F) ₂·sH₂O (C₁₈-6, s = 3.5, C ₁₈-7, s= 0) show a very particular spin-state change, while [Fe(C₁₈-tren)](Br)₂·3H₂O (C ₁₈-8) together with [Fe(C₁₈-tren)](I)₂ (C ₁₈-9) are in the LS state (10-400 K) and present mesomorphic behavior like that observed for the complexes C₁₈-1, C₁₂-2, and C₁₈-2. In compounds C_n-3 50% of the Fe(II) ions undergo spin-state change at T_1/2 = 375 K induced by releasing water, and in partially dehydrated compounds (s = 0.5) the Cr→S_A phase transition occurs at 287 K (C₁₆-4), 301 K (C₁₈-4) and 330 K (C₂₀-4). For the fully dehydrated materials C_n-5 50% of the Fe(II) ions are in the HS state and show paramagnetic behavior between 10 and 400 K. In the partially dehydrated C_n-4 the spin transition is induced by the change of the aggregate state of matter (solid?liquid crystal). For compound C₁₈-6 the full dehydration to C₁₈-7 provokes the spin-state change of nearly 50% of the Fe(II) ions. The compounds C_n-3 and C₁₈-6 are dark purple in the LS state and become light purple-brown when 50% of the Fe(II) atoms are in the HS state.

Full text of DOI:10.1021/ja077265z

Published on Web 01/09/2008
Does the Solid-Liquid Crystal Phase Transition Provoke the
Spin-State Change in Spin-Crossover Metallomesogens?
†
,‡
†
§
|
M. Seredyuk, A. B. Gaspar,* V. Ksenofontov, Y. Galyametdinov, J. Kusz, and
,
†
P. G u¨ tlich*
Institut f u¨ r Anorganische und Analystiche Chemie, Johannes-Gutenberg-UniVersit a¨ t,
Staudinger-Weg 9, D-55099 Mainz, Germany, Institut de Ci e` ncia Molecular/Departament de
Qu ´ı mica Inorg a` nica, UniVersitat de Val e` ncia, Edifici de Instituts de Paterna, Apartat de Correus
2
2085, E-46071 Val e` ncia, Spain, Kazan Physical Technical Institute, Russian Academy of
Science, Sibirsky Tract 10/7, 420029, Kazan, Russia, and Institute of Physics, UniVersity of
Silesia, 4 Uniwersytecka Street, 40007 Katowice, Poland
Received September 19, 2007
Abstract: Three types of interplay/synergy between spin-crossover (SCO) and liquid crystalline (LC) phase
transitions can be predicted: (i) systems with coupled phase transitions, where the structural changes associated
to the CrTLC phase transition drives the spin-state transition, (ii) systems where both transitions coexist in the
same temperature region but are not coupled, and (iii) systems with uncoupled phase transitions. Here we present
a new family of Fe(II) metallomesogens based on the ligand tris[3-aza-4-((5-Cn)(6-R)(2-pyridyl))but-3-enyl]amine,
with Cn ) hexyloxy, dodecyloxy, hexadecyloxy, octadecyloxy, eicosyloxy, R ) hydrogen or methyl (Cn-trenH or
Cn-trenMe), which affords examples of systems of types i, ii, and iii. Self-assembly of the ligands Cn-trenH and
Cn-trenMe with Fe(A)2‚xH2O salts have afforded a family of complexes with general formula [Fe(Cn-trenR)](A)2‚
-
-
-
-
-
sH2O (s g 0), with A ) ClO4 , F , Cl , Br and I . Single-crystal X-ray diffraction measurements have been
performed on two derivatives of this family, named as [Fe(C6-trenH)](ClO4)2 (C6-1) and [Fe(C6-trenMe)](ClO4)2
(C6-2), at 150 K for C6-1 and at 90 and 298 K for C6-2. At 150 K, C6-1 displays the triclinic space group P 1h ,
whereas at 90 and at 298 K C6-2 adopts the monoclinic P21/c space group. In both compounds the iron atoms
adopt a pseudo-octahedral symmetry and are surrounded by six nitrogen atoms belonging to imino groups and
pyridines of the ligands Cn-trenH and Cn-trenMe. The average Fe(II)-N bonds (1.963(2) Å) at 150 K denote that
C6-1 is in the low-spin (LS) state. For C6-2 the average Fe(II)-N bonds (2.007(1) Å) at 90 K are characteristic of
the LS state, while at 298 K they are typical for the high-spin (HS) state (2.234(3) Å). Compound C6-1 and [Fe-
(C18-trenH)](ClO4)2 (C18-1) adopts the LS state in the temperature region between 10 and 400 K, while compound
C6-2 and [Fe(Cn-trenMe)](ClO4)2 (n ) 12 (C12-2), 18 (C18-2)) exhibit spin crossover behavior at T1/2 centered
around 140 K. The thermal spin transition is accompanied by a pronounced change of color from dark red (LS)
to orange (HS). The light-induced excited spin state trapping (LIESST) effect has been investigated in compounds
LIESST
C6-2, C12-2 and C18-2. The T1/2
is 56 K (C6-2), 48 K (C16-2), and 56 K (C18-2). On the basis of differential
scanning calorimetry, optical polarizing microscopy, and X-ray diffraction findings for C18-1, C12-2, and C18-2 at
high temperature a smectic mesophase SX has been identified with layered structures similar to C6-1 and C6-2.
The compounds [Fe(Cn-trenH)](Cl)2‚sH2O (n ) 16 (C16-3, s ) 3.5, C16-4, s ) 0.5, C16-5, s ) 0), 18 (C18-3, s )
3.5, C18-4, s ) 0.5, C18-5, s ) 0), 20 (C20-3, s ) 3.5, C20-4, s ) 0.5, C20-5, s ) 0)) and [Fe(C18-tren)](F)2‚sH2O
(C18-6, s ) 3.5, C18-7, s ) 0) show a very particular spin-state change, while [Fe(C18-tren)](Br)2‚3H2O (C18-8)
together with [Fe(C18-tren)](I)2 (C18-9) are in the LS state (10-400 K) and present mesomorphic behavior like
that observed for the complexes C18-1, C12-2, and C18-2. In compounds Cn-3 50% of the Fe(II) ions undergo
spin-state change at T1/2 ) 375 K induced by releasing water, and in partially dehydrated compounds (s ) 0.5)
the CrfSA phase transition occurs at 287 K (C16-4), 301 K (C18-4) and 330 K (C20-4). For the fully dehydrated
materials Cn-5 50% of the Fe(II) ions are in the HS state and show paramagnetic behavior between 10 and 400
K. In the partially dehydrated Cn-4 the spin transition is induced by the change of the aggregate state of matter
(
5
solidTliquid crystal). For compound C18-6 the full dehydration to C18-7 provokes the spin-state change of nearly
0% of the Fe(II) ions. The compounds Cn-3 and C18-6 are dark purple in the LS state and become light purple-
brown when 50% of the Fe(II) atoms are in the HS state.
Introduction
The interest in the tailoring of multifunctional materials
combining spin-crossover (SCO) and liquid crystalline (LC)
a recently developed type of mesogenic materials. The so-called
SCO phenomenon is observed in transition-metal ions with 3dn
(n ) 4-7) electronic configurations. The vast majority of SCO
1
2
behavior arises from the potential properties derived from the
(
1) (a) G u¨ tlich, P., Goodwin, H. A., Eds. Spin Crossover in Transition Metal
Compounds. Topics in Current Chemistry; Springer: Berlin, 2004; Vol.
233, 234, 235. (b) Gaspar, A. B.; Ksenofontov, V.; Seredyuk, M.; G u¨ tlich,
P. Coord. Chem. ReV. 2005, 249, 2661-2676. (c) Real, J. A.; Gaspar, A.
B.; Mu n˜ oz, M. C. Dalton Trans. 2005, 2062-2079. (d) Real, J. A.; Gaspar,
A. B.; Niel, V.; Mu n˜ oz, M. C.; Coord. Chem. ReV. 2003, 236, 121-141.
3
presence of metal ions within anisotropic phases. These kinds
4
of materials are new examples of functional metallomesogens,
†
Johannes-Gutenberg-Universit a¨ t.
Universitat de Val e` ncia.
Russian Academy of Science.
University of Silesia.
‡
(
e) G u¨ tlich, P.; Hauser, A.; Spiering, H. Angew. Chem., Int. Ed. Engl. 1994,
§
33, 2024-2054. (f) G u¨ tlich, P.; Garcia, Y. Top. Curr. Chem. 2004, 234,
49-62.
|
10.1021/ja077265z CCC: $40.75 © 2008 American Chemical Society
J. AM. CHEM. SOC. 2008, 130, 1431-1439
9
1431

A R T I C L E S
Seredyuk et al.
compounds reported so far concerns Fe(II) and Fe(III) com-
plexes, to a lesser extent Co(II), and only in a few cases Mn-
(
temperature intervals.11 Hayami and co-workers12 have also
reported mononuclear iron(II) complexes in which the spin
transition and the smectic mesophases appear at different
temperature intervals. Recently, we have investigated the
possibility to synchronize both transitions, spin transition and
solid-liquid crystal transition, in Fe(II) complexes.13 The first
step in the tailoring of these multifunctional materials has been
the choice of SCO systems showing abrupt spin transition near
or above room temperature. This is the range of temperatures
at which usually the LC phase transition is observed. In this
respect, the polymeric SCO compounds based on the triazole-
based ligands are among the best candidates, showing abrupt
spin transition above or near to room temperature accompanied
III) and Cr(II) complexes.1f SCO materials display labile
electronic configurations switchable between the high-spin (HS)
and low-spin (LS) states leading to distinctive changes in
magnetism, color and structure, which may be driven by
1
variation of temperature and/or pressure and by light irradiation
5
(light-induced excited spin state trapping (LIESST) effect).
Their magnetic, optical and structural properties may be altered
drastically in a narrow range of temperature and/or pressure for
cooperative spin transitions. Cooperativity may be accompanied
by hysteresis (memory effect) when the cohesive forces,
communicating between the SCO centers in the solid state,
propagate the structural changes cooperatively throughout the
whole lattice. A crossover between the HS and LS configura-
tions occurs when the change of the Gibbs free energy ∆GHL
1
by a pronounced change of color. The second step in this
approach has concerned the incorporation of the LC moiety into
the SCO system, which implies the attachment of the aromatic
core with the alkyl chains to the triazole ligand. In the
synthesized Fe(II) metallomesogens, [Fe(Cn-trz)3](4-MeC6H4-
)
GHS - GLS is in the range of thermal energy.1
A single material combining spin crossover and liquid
13
crystalline behavior may lead to a number of advantages in
practical applications, for example, processing spin crossover
materials in the form of thin films, enhancement of spin
transition signals, switching and sensing in different temperature
regimes, or achievement of photo- and thermochromism in
metal-containing liquid crystals.1b-c,4a The change of color is
certainly a phenomenon which is of interest in the field of liquid
crystals. The interest lies in the necessity for color in a number
of applications in liquid crystals such as passive blocking filters,
laser addressed devices, polarizers based on dichroic effects,
SO3)2‚xH2O, SCO occurs in the room-temperature range where
the materials show a discotic columnar mesophase Dhd. The
change of the spin state is accompanied by a pronounced change
of color from purple (LS state) to white (HS state). We have
provided experimental evidence that [Fe(Cn-trz)3](4-MeC6H4-
SO3)2 metallomesogens present discotic columnar mesophase
in the temperature interval of 200-475 K where the thermal
spin transition occurs. Realization of the crystalline state of these
materials is difficult, in fact, no melting point has been
1
3
observed. Therefore, we concluded, in contrast to previous
4
14
or the utility of thermochromism. While color change has rarely
work, that the spin-transition behavior in these metallome-
6
been observed for nonchiral organic liquid crystals, several
sogens is not driven by a phase transition from the crystalline
state to the liquid crystal state. On the other hand, spin-state
transition induced by mechanical strength on going to the
amphiphilic mesophase in metallo-supramolecular polyelectro-
lytes organized in LB films has been proposed, however, not
papers have reported thermochromism in nonchiral metal-
containing liquid crystals.7 More successful has been the
8
development of cholesteric liquid crystals for the production
of colored electro-optical films together with the liquid crystal
9
15
OLEDs. Worth mentioning is also the achievement of photo-
yet experimentally demonstrated. Presently, the number of
chromic organic liquid crystals.10
reported SCO metallomesogens is rather scarce, and the spin-
transition phenomena in these materials was found to be
governed by the classical factors which are well-known to affect
the SCO and its characteristics (Tc, hysteresis), namely, subtle
structural and electronic modifications tuned by the crystal
packing, which determines the ligand field strength and the SCO
behavior. These modifications depend essentially on the nature
of the ligands, the non-coordinating anions, the solvent mol-
ecules, and the crystal packing. Complete control of these
variables is a rather difficult task to accomplish.
A first step aiming to achieve a material gathering SCO and
LC properties has led to an iron(III) metallomesogen where both
properties are not synchronous but they appear in different
(
2) (a) Laschat, S.; Baro, A.; Steinke, N.; Giesselmann, F.; H a¨ gele, C.; Scalia,
G.; Judele, R.; Kapatsina, E.; Sauer, S.; Schreivogel, A.; Tosoni, M. Angew.
Chem., Int. Ed. 2007, 46, 4832-4887 and references therein. (b) Donnio,
B.; Guillon, D. AdV. Polym. Sci. 2006, 201, 45-155. (c) Camerel, F.;
Donnio, B.; Bourgogne, C.; Schmutz, M.; Guillon, D.; Davidson, P.; Ziessel,
R. Chem. Eur. J. 2006, 16, 4261-4274. (d) Reddy, R. A.; Tschierske, C.
J. Mater. Chem. 2006, 10, 907-961. (e) Binnemans, K. Chem. ReV. 2005,
1
1, 4148-4204. (f) Paleos, C. M.; Tsiourvas, D. Angew. Chem., Int. Ed.
995, 16, 1696-1711.
1
The question is actually posed as whether the change of
aggregate of the matter (solid-liquid crystal) drives the spin-
(
3) Rowan, S. J. Angew. Chem., Int. Ed. 2005, 44, 4830-4832.
(
4) (a) Serrano, J. L. Metallomesogens; VCH: Weinheim, Germany, 1996 and
references therein. (b) Hudson, S. A.; Maitlis, P. M. Chem. ReV. 1993, 93,
8
61. (c) Donnio, B.; Bruce, D. W. Struct. Bond. 1999, 95, 193-247. (d)
Lemieux, R. P. Acc. Chem. Res. 2001, 34, 845-853. (e) Gimenez, R.;
(9) Aldred, M. P.; Contoret, A. E. A.; Farrar, S. R.; Stephen, M. K.; Mathieson,
D.; O’ Neill, M.; Tsoi, W. C.; Vlachos, P. AdV. Mater. 2005, 17, 1368-
1372.
Lydon, D. P.; Serrano, J. L. Curr. Opin. Solid State Mater. Sci. 2002, 6,
5
8
2
27-535. (f) Serrano, J. L.; Sierra, T. Coord. Chem. ReV. 2003, 242, 73-
5. (g) Donnio, B. Curr. Opin. Colloid, Interface Science State Mater. Sci.
002, 7, 371-394. (h) Bruce, D. W. Acc. Chem. Res. 2000, 33, 831-840.
(10) (a) Frogoli, M.; Mehl, G. H. Chem. Phys. Chem. 2003, 1, 101-103. (b)
Irie, M. Chem. ReV. 2000, 100, 1685.
(
i) Piguet, C.; B u¨ nzli, J. C. G.; Donnio, B.; Guillon, D. Chem. Commun.
(11) Galyametdinov, Y.; Ksenofontov, V.; Prosvirin, A.; Ovchinnikov, I.;
Ivanova, G.; G u¨ tlich, P.; Haase, W. Angew. Chem., Int. Ed. 2001, 40, 4269-
4271.
2
006, 36, 3755-3768.
(
5) (a) Decurtins, S.; G u¨ tlich, P.; K o¨ hler, C. P.; Spiering, H.; Hauser, A. Chem.
Phys. Lett. 1984, 105, 1-4. (b) L e` tard, J.-F. J. Mater. Chem. 2006, 16,
(12) (a) Hayami, S.; Danjobara, K.; Inoue, K.; Ogawa, Y.; Matsumoto, N.;
Maeda, Y. AdV. Mater. 2004, 16, 869-872. (b) Hayami, S.; Motokawa,
N.; Shuto, A.; Masuhara, N.; Someya, T.; Ogawa, Y.; Inoue, K.; Maeda,
Y. Inorg. Chem. 2007, 46, 1789-1794.
2
550-2559.
(
6) Shvartsman, F. P.; Krongauz, V. A. Nature (London) 1984, 309, 608-
6
11.
(
7) (a) Bruce, D. W.; Dunmur, P. A.; Esteruelas, M. A.; Hunt, S. E.; Maitlis,
(13) Seredyuk, M.; Gaspar, A. B.; Ksenofontov, V.; Reiman, S.; Galyametdinov,
Y.; Haase, W.; Rentschler, E.; G u¨ tlich, P. Chem. Mater. 2006, 18, 2513-
2519.
P. M.; Marsden, J. R.; Sola, E.; Stacey, J. M. J. Mater. Chem. 1991, 1,
2
51-255. (b) Ohta, K.; Hasabe, H.; Moriya, M.; Fujimoto, T.; Yamamoto,
I. J. Mater. Chem. 1991, 1, 831-834. (c) Gregg, B. A.; Fox, M. A.; Bard,
(14) Fujigaya, T.; Jiang, D. L.; Aida, T. J. Am. Chem. Soc. 2003, 125, 14690-
A. J. J. Phys. Chem. 1989, 93, 4227-4234.
8) De Filpo, G.; Nicoletta, F. P.; Chidichimo, G. AdV. Mater. 2005, 17, 1150-
14691.
(
(15) Bodenthin, Y.; Pietsch, U.; M o¨ hwald, H.; Kurth, D. G. J. Am. Chem. Soc.
2005, 127, 3110-3114.
1
152.
1432 J. AM. CHEM. SOC.
9
VOL. 130, NO. 4, 2008

Spin-State Changes in Spin-Crossover Metallomesogens
A R T I C L E S
Chart 1
distorted by the gauche conformation of some of the methylene
groups (Figure 1b) and tilted toward the ac plane but do not
intertwine with those of adjacent layers. If compared with the
16
unalkylated complex {Fe[tren(Py)3]}(ClO4)2, the introduction
of alkyl chains into the complex molecule affects the crystal
packing causing in C6-1 the segregation into a layered structure.
In {Fe[tren(Py)3]}(ClO4)2 the hydrogen-bonding network has a
three-dimensional character, while in C6-1 all CH‚‚‚O(Cl)
contacts (2.428(2)-2.719(2) Å) belong to the two-dimensional
hydrogen-bonding network formed within the ionic layers. For
C6-2 (Figure 2a and b) the average Fe(II)-N bonds (2.007(1)
Å) and the distortion parameter Σ (85.54(7)°) at 90 K are
characteristic of the LS state while at 298 K they are typical
for the HS state (2.234(3) Å, 116.59(11)°). Modifications of
the coordination sphere of C6-2 due to spin transition are directly
reflected in the metal-to-ligand bond length, which is illustrated
by the distinguishable divergence of the overlapped HS and LS
structures (Figure 2c). In the HS form the distances Fe-
N(ligand) are longer, consequently, the volume of the [FeN6]
chromophore increases. C6-2 displays similar crystal packing
like observed in C6-1 despite both compounds crystallizing in
different space groups (SFigure 6). The crystal packing in the
LS and HS structures for C6-2 is almost identical, with CH‚‚‚
O(Cl) contacts being 2.308(5)-2.781(5) Å (HS) and 2.301(5)-
a
Arrows show relation of the compounds upon dehydration. For details,
see explanation in text.
state change or only affects the cooperative process of the SCO
phenomena. Three types of interplay/synergy between SCO and
LC phase transitions can be predicted: (i) systems with coupled
phase transitions, where the structural changes associated to the
CrTLC phase transition drives the spin state transition, (ii)
systems where both transitions coexist in the same temperature
region but are not coupled, and (iii) systems with uncoupled
phase transitions. Here we shall present a new family of Fe(II)
metallomesogens based on the ligand tris[3-aza-4-((5-Cn)(6-R)-
2
.718(5) Å (LS).
(
2-pyridyl))but-3-enyl]amine, with Cn ) hexyloxy, dodecyloxy,
Magnetic and Mesomorphic Behavior of Compounds Cn-1
hexadecyloxy, octadecyloxy, eicosyloxy, R ) hydrogen or
methyl (C6,16,18,20-trenH or C6,12,18-trenMe), which affords
examples of systems of types i, ii, and iii. We shall present an
analysis of the factors that govern the SCO process in these
materials and discuss the synergy between SCO and LC phase
transitions observed in materials of type i.
(n ) 6, 18) and Cn-2 (n ) 6, 12, 18). Compound C6-1 and
[Fe(C18-trenH)](ClO4)2(C18-1) adopt the LS state in the tem-
perature region between 10 and 400 K, where the temperature
dependence of the magnetic susceptibility has a constant value
3
-1
of 0.10 cm K mol (SFigure 2). The M o¨ ssbauer spectrum of
C6-1 recorded at 80 K points out two nonequivalent Fe(II) atoms,
both in the LS state (SFigure 3, STable 1). The M o¨ ssbauer
spectrum of C18-1 recorded at 80 K, however, could be
satisfactorily fitted with only one LS doublet. Introduction of
the methyl group in the Cn-trenR ligand has resulted in a
noticeable modification of the crystal-field strength at the iron-
Results and Discussion
Self-assembly of the ligands Cn-trenH and Cn-trenMe with
Fe(A)2‚sH2O salts have afforded a family of complexes with
general formula [Fe(Cn-trenR)](A)2‚sH2O (s g 0), with A )
ClO4 , F , Cl , Br , and I (Chart 1). Single-crystal X-ray
diffraction measurements have been performed on two deriva-
tives of this family of metallomesogens, named as [Fe(C6-
trenH)](ClO4)2 (C6-1) and [Fe(C6-trenMe)](ClO4)2 (C6-2), at 150
K for C6-1 and at 90 and 298 K for C6-2.
Crystal structure of C6-1 and C6-2. Figure 1a and Figure
a,b illustrate the molecular structures of both compounds
together with the corresponding atom numbering scheme. At
50 K, C6-1 displays the triclinic space group P 1h , whereas at
0 and at 298 K C6-2 adopts the monoclinic P21/c space group.
A selection of crystallographic data and bond distances are given
in the Supporting Information STable 1 and STable 2. The iron
atoms adopt a pseudo-octahedral symmetry and are surrounded
by six nitrogen atoms belonging to imino groups and pyridines
of the trifurcated ligands Cn-trenH and Cn-trenMe. In C6-1 each
cell contains two independent complex molecules of opposite
chirality (Figure 1a). The average Fe(II)-N bonds (1.963(2)
Å) and the distortion parameter Σ (67.56(8)°) at 150 K denote
that C6-1 is in the LS state. The amphiphilic nature of the
alkylated molecules results in the self-assembly into bilayered
composite with one layer being made up of polar head groups
together with perchlorate anions. The nonpolar chains from
oppositely directed molecules meet together leading to a
hydrocarbon layer. Almost fully stretched alkyl chains are only
-
-
-
-
-
(II) center. Compound C6-2 and [Fe(Cn-trenMe)](ClO4)2 n )
1
2 (C12-2), 18 (C18-2) exhibit spin crossover behavior at T1/2
centered around 140 K, the temperature at which the number
of molecules in the HS and LS states is equal. Figure 3 depicts
the øMT versus T plots for C6-2, C12-2, and C18-2 (øM stands
for the molar magnetic susceptibility and T the temperature).
The spin transition is complete and relatively continuous for
C6-2, C12-2, and C18-2 with T1/2 ) 146, 120, and 133 K,
respectively. M o¨ ssbauer spectra recorded below and above T1/2
for C6-2, C12-2, and C18-2 are consistent with the magnetic and
structural data reported (SFigure 3). The thermal spin transition
is accompanied by a pronounced change of color from dark
red (LS) to orange (HS). The LIESST effect5 has been
investigated in compounds C6-2, C12-2, and C18-2. The LS
ground state was photoconverted at 4 K to a metastable HS
state after irradiation of samples with light of λ ) 514 nm (25
2
1
9
-
2
21
mW cm ) (Figure 3a). The highest percentage of molecules
that was photoconverted to the HS state inferred from øMT
values at 42, 25, and 42 K for C6-2, C12-2, and C18-2 is 80%,
2
2
5
4%, 50%, respectively. The critical temperature of relaxation
LIESST 5a
of the HS metastable state known as T
is 56 K (C6-2),
(
16) Brewer, G.; Luckett, C.; May, L.; Beatty, A. M.; Scheidt, W. R. Inorg.
Chim. Acta 2004, 357, 2390-2396.
J. AM. CHEM. SOC.
9
VOL. 130, NO. 4, 2008 1433

A R T I C L E S
Seredyuk et al.
Figure 1. (a) Projection of the two independent cations of [Fe(C6-trenH)]2 with atom numbering scheme (150 K). (b) Projection of the molecular
+
packing of C6-1 along the c-axis. Displacement ellipsoids are shown at 50% probability level. Hydrogen atoms and perchlorate anions are omitted for
clarity.
4
8 K (C12-2), and 56 K (C18-2). XRD profiles of the complexes
C18-1, C12-2, and C18-2 have been recorded at 300 and 410 K
SFigure 5). The X-ray diffraction patterns of the mesophase
higher reflections and broadening of the alkyl reflection is
observed. From the maxima of the reflections average interlayer
distances d are obtained in both solid and liquid crystalline states
(Table 1). On the basis of these findings for C18-1, C12-2, C18-2
at high temperature a smectic mesophase SX has been identified
with layered structures similar to C6-1 and C6-2. DSC experi-
ments performed at a rate of 10 K/min, in the warming and
cooling modes, are shown for C18-1, C12-2, C18-2 in SFigure 4.
Table 1 gathers the critical temperatures of melting, isotropi-
(
of these complexes gave a single sharp low angle (10) reflection
and weak high-order reflection(s) all corresponding to the layer
spacing. An additional diffuse reflection observed at larger
angles refers to the average lateral spacing between alkyl chains.
Upon Cr f SA phase transition a shift of the low and high angle
reflections to lower values with change of the intensity for the
1434 J. AM. CHEM. SOC.
9
VOL. 130, NO. 4, 2008

Spin-State Changes in Spin-Crossover Metallomesogens
A R T I C L E S
Figure 2. (a) Projection of the cations [Fe(C6-trenMe)]2+ at 90 K and (b) at 298 K with atom numbering scheme. Hydrogen atoms and perchlorate anions
are omitted for clarity. Displacement ellipsoids are shown at 50% probability level. (c) Projection along Fe‚‚‚N7 axis of the minimized overlay of the cation
2
+
[
Fe(C6-trenMe)] in the LS state at 90 K (blue) and in the HS state at 298 K (red).
the one observed in C6-2. Packed long alkyl chains to some
extent may prevent contraction of the cavity within the
trifurcated ligand with an encapsulated iron(II) ion. The effect
is rather weak causing the shift to lower temperature by ∼15 K
for C18-2.
Magnetic and Mesomorphic Behavior of Compounds Cn-3
to Cn-5 (n ) 16, 18, 20), C18-6 to C18-9. The compounds [Fe-
(
Cn-trenH)](Cl)2‚3.5H2O, n ) 16 (C16-3), 18 (C18-3), 20 (C20-
) and [Fe(C18-tren)](F)2‚3.5H2O (C18-6) show a very particular
spin-state change, while [Fe(C18-tren)](Br)2‚3H2O (C18-8) and
Fe(C18-tren)](I)2 (C18-9) are in the LS state (10-400 K) and
3
[
present mesomorphic behavior like that observed for the
complexes C18-1, C12-2, and C18-2 (Table 1, SFigure 2-5).
Figure 4a) illustrates the øMT versus T plots for C16-3, a partial
hydrate (s ) 0.5) C16-4 and for the completely dehydrated
complex C16-5 (s ) 0) together with the corresponding
M o¨ ssbauer spectra recorded at 80 K. Compound C16-3 is in the
Figure 3. Magnetic properties in the form of øMT versus T for Cn-2
2
1
(
n ) 6, 12, 18) and LIESST experiments with irradiation at 4 K (λ ) 514
-
2
nm, 25 mW cm ) and subsequently moving up in temperature (open
symbols).
3
LS state at 300 K as confirmed by the øMT value of 0.02 cm
-
1
K mol as well as by the M o¨ ssbauer spectrum. It is worth
noting that the M o¨ ssbauer spectrum of C16-3 at 80 K shows
two strongly overlapping doublets corresponding to two non-
equivalent Fe(II) positions in the LS state (STable 3, SFigure
3). Upon heating øMT increases abruptly, attaining the value of
sation and decomposition deduced from differential scanning
calorimetry (DSC), optical polarizing microscopy (OPM), and
thermogravimetric analysis (TGA).
The lengthening of alkyl chains in Cn-2 provokes a decrease
of T1/2, however, not in the linear way. In fact, compounds Cn-2
are obtained at ambient temperature in HS form, and they
adopt packing of molecular parts favoring bulkier HS polar
heads. This could play a role in the shift of SCO in higher
homologues C12-2 and C18-2 to lower temperature compared to
3
-1
1.75 cm K mol at 400 K, where around 50% of the Fe(II)
ions have converted to the HS state. TGA experiments have
shown that dehydration takes place in the same temperature
region where the spin-state change occurs (SFigure 1). Ad-
J. AM. CHEM. SOC.
9
VOL. 130, NO. 4, 2008 1435

A R T I C L E S
Seredyuk et al.
Table 1. Mass Spectroscopy Data, Thermal Transition and Mesomorphism, Interlayer Distances, Enthalpy, and Thermal Stability
Determined by MS, OPM, DSC, and X-ray Diffraction for Compounds Cn-1 to Cn-9
thermal
dCr
[Å]c
dSx, 410K
[Å]
dCr
[Å]e
∆
HCr
fSx
transitions
[K]f
a
[kJ mol-1]g
compound
MS , m/z
+
++
16.6 d
29.1
C6-1
C18-1
C6-2
868 [M+ClO4] , 385 [M]
+
++ b
1373 [M+ClO4] , 637 [M]
31.05
29.1
Cr 370 SX 473 d
49.6
+
17.4 (90 K), 17.9 (290 K) d
910 [M+ClO4]
+
C12-2
C18-2
C16-3
C18-3
C20-3
C18-6
C18-8
C18-9
1163 [M+ClO4]
24.0
29.6
34.7
36.0
37.7
37.8
35.8
32.5
24.8
31.3
39.1
42.3
45.0
38.6
39.1
31.3
24.0
29.5
38.7
40.1
44.1
37.8
39.7
31.3
Cr 370 SX 470 d
Cr 373 SX 470 d
39.0
54.9
28.1
38.5
63.8
53.5
62.4
65.7
+
1415 [M+ClO4]
+
1225 [M+Cl]
Cr 287 SA 456 i(d)
Cr 301 SA 463 i(d)
Cr 330 SA 449 i(d)
Cr 320 SX 439 i(d)
Cr 325 SX 451 i(d)
Cr 393 SX 456 i(d)
+
++
++
++
1309 [M+Cl] , 637 [M]
+
1393 [M+Cl] , 679 [M]
+
b
1293 [M+F] , 637 [M]
+
b
++
1353 [M+Br] , 637 [M]
+
++
1401 [M+I] , 637 [M]
a
ESI or FD. b Trace. c Interlayer distance before first heating. d From the monocrystal X-ray data. e Interlayer distance after first heating. f Measured by
g
CrfSx
DSC on the second heating scan. Isotropisation (decomposition) points are determined by OPM; decomposition confirmed by TGA. DSC ∆H
the second heating scan.
data for
Figure 4. (a) øMT versus T for C16-3, C16-4 and C16-5: The symbol × denotes the temperature (80 K) at which the M o¨ ssbauer spectra (shown on the right)
were recorded (HS doublet, red; LS doublet, blue). (b) Variation of the interlayer distance d with temperature (4) in C16-3 and C16-4 deduced from XRD
profiles after following the sequence 310-410 to 230-410 K together with dehydration experiments in C16-3 monitored by temperature dependence of the
2
3
magnetic susceptibility (O). (c) øMT versus T for C16-4, C18-4, and C20-4 arrows indicate the temperatures at which heating and cooling curves diverge. The
black areas indicate the hysteresis loops.
ditionally, DSC profiles, OPM analysis, and XRD patterns of
C16-3 (SFigures 4, 5) show that the melting occurs at 340 K
emerges: does the solidTliquid crystal phase transition during
the first heating drive the spin transition in C16-3 or is it driven
by the loss of water?
(first heating scan). At this point the following question
1436 J. AM. CHEM. SOC.
9
VOL. 130, NO. 4, 2008

Spin-State Changes in Spin-Crossover Metallomesogens
A R T I C L E S
To answer this question we have performed dehydration
with longer alkyl chains. For compound C18-6 no partial hydrate
can be obtained like in the case of Cn-3, but the full dehydration
provokes the spin-state change of near 50% of the Fe(II) ions
(SFigures 1-5). The compounds Cn-3 and C18-6 are dark purple
in the LS state and become light purple-brown when 50% of
the Fe(II) atoms are in the HS state.
2
3
experiments in C16-3 monitoring the temperature dependence
of the magnetic susceptibility (Figure 4b) and have recorded
temperature-dependent XRD patterns following the sequence
3
10-410 to 230-410 K (SFigure 5). The sample was heated
and subsequently cooled during several cycles between 280 (LS
state) and 350 K. The øMT value after three cycles of
meltingTsolidification (287 K second heating scan) does not
show any noticeable variation. In addition, the temperature
variation of d derived from XRD patterns (Figure 4b) present
an abrupt increase as a consequence of the melting between
Conclusion
In view of the experimental results the answer to the question
posed in the title is yes, for some type of spin crossover
metallomesogens.
3
40 and 350 K, but the increase in the øMT value begins at 360
Compounds C18-1, C18-8, and C18-9 do not present SCO
behavior in the region of temperatures investigated (10-400
K) but exhibit mesomorphic behavior. Compounds C12-2 and
C18-2 can be classified as type iii in view of the different phase
transitions occurring in these materials taking place in very
different temperature regions. Therefore, any interplay/synergy
between different physical phenomena, spin transition, and
change of aggregate state of matter (solidTliquid crystal), can
be anticipated. On the other hand, the compounds Cn-3 and C18-6
K. Consequently, the hypothesis that the phase transition to the
mesophase drives the spin-state change can be ruled out, at least
in the present compound (C16-3). On the contrary, it is clearly
seen in the dehydration experiments that the magnetic suscep-
tibility increases upon losing water, which matches with reported
TGA data, and is completely accomplished at 400 K. For
compound C16-3 50% of the Fe(II) ions undergo spin-state
change at T1/2 ) 375 K induced by releasing water. Synergy
between desolvation and spin-state change as observed in C16-3
has been reported in several cases.1
(type ii) are new examples of materials in which the two phase
,13
transitions coexist in the same temperature region but are not
coupled. These very interesting materials present change of
aggregate state of matter and spin-state change upon dehydration
in a narrow temperature interval. The compounds Cn-4 represent
the first examples of systems in which the solidTliquid crystal
transition drives the spin-state transition.
For the fully dehydrated material C16-5 the M o¨ ssbauer
spectrum recorded at 80 K gives the following relative popula-
tions for the two inequivalent iron sites (1,2) in different spin
states: For Fe(1) they are 25% (HS), 25%(LS); for Fe(2) they
are 25% (HS), 25%(LS) (STable 3, SFigure 3). The øMT versus
T plot for C16-5 agrees well with the M o¨ ssbauer spectrum and
is characteristic of Fe(II) ions in the HS state (Figure 4a). The
magnetic properties of a partially dehydrated C16-4 has also been
investigated, which discloses interesting synergy between spin
transition and change of the aggregate state of matter
As mentioned in the introduction, complete control of the
variables that affect the spin-crossover behavior in solid state
is a rather difficult task to accomplish, since it depends
essentially on the nature of the ligands, the non-coordinating
1
anions, the solvent molecules, and the crystal packing. This is
(
solidTliquid crystal). Figure 4a displays the øMT versus T plot
particularly accentuated for monomeric complexes as the
communication between the SCO centers is achieved exclusively
through intermolecular interactions, such as hydrogen bonding
or π-π interactions between aromatic ligands, anions, and
solvent molecules. In polymeric compounds, partial or total
substitution of these intermolecular interactions by covalent
linkage of the metal centers leads to a more predictable control
of the spin-transition behavior.
for C16-4. The partial dehydration of C16-3 was monitored by
recording øMT versus T during two cycles of warming-cooling
following the sequence 300-400 to 10-400 K and proved by
TGA analysis that C16-4 was left (SFigure 1). For C16-4, the
øMT versus T between 400 and 296 K presents a variation
accompanied by a narrow hysteresis. Below 296 K øMT remains
almost constant until 10 K, where it sharply decreases as a
consequence of the zero-field splitting of the Fe(II) ions
remaining in the HS state (18%, STable 3). The thermally
induced spin transition in C16-4 starts right after the onset of
the first-order phase transition (CrTSA). Since the spin transition
is blocked below the temperature at which the compound
solidifies, one can conclude that the melting process drives the
spin transition in this compound. The fact that the spin transition
inside the SA mesophase is accompanied by hysteresis is not
surprising since in the present state of the material the spin
transition is not expected to follow a Boltzmann distribution of
spin states as observed in the liquid state.1 Indeed, spin
transition accompanied by hysteresis has been observed earlier
in the compounds [Fe(Cn-trz)3](4-MeC6H4SO3)2 in the region
of existence of the Dhd mesophase.13 One remarkable fact is
that the compounds C18-3 and C20-3 behave similarly as C16-3.
The results of physical studies of these materials are presented
in SFigures 1-5. For a comparative analysis Figure 4c shows
the øMT versus T plots for Cn-4 (n ) 16, 18, 20). It is worth
mentioning that the spin-transition temperature is determined
by the melting temperature which is higher for the derivatives
-
The general observation in [Fe(C -trenH)](A) ‚sH O A ) F ,
n
2
2
-
-
-
Cl , Br and I metallomesogens is that the spin state depends
mainly on the degree of hydration of the compound and not on
the state of the matter, solid or liquid. In fact, only for the
particular cases of derivatives C -4 do the structural changes
n
associated to the CrTLC phase transition drive the spin-state
transition. Seemingly, the effect of the different electronegativity
of halogenated counterions is less important in determining the
crystal field strength at the iron(II) center. In contrast, no solvates
were isolated in the case of perchlorate derivatives [Fe(Cn-
trenH)](ClO4)2,where the LS state is the preferred spin state.
Introduction of the alkyl group in the ligand Cn-trenR decreases
the ligand field strength at the Fe(II) center leading to a spin
crossover behavior in [Fe(Cn-trenMe)](ClO4)2.
a
In some instances correlations between anion size and
1
,17
transition temperature have been proposed
but the general
validity has not been established. The effects of the anion and
solvation are, however, not always consistent from one system
to another and are not readily predictable. Replacement of the
anion or solvent molecule is expected to modify the lattice
J. AM. CHEM. SOC.
9
VOL. 130, NO. 4, 2008 1437

A R T I C L E S
Seredyuk et al.
phonon distribution resulting from different crystal packing
geometry or strength of the intermolecular forces. In addition,
changes in the chemical composition of the lattice could impose
different degrees of “chemical pressure” (also known as “image
pressure”) on the spin transition centers and thereby influence
the transition temperature. Hydrogen bonding can be a major
influence on both the transition temperature (in part at least by
affecting the ligand field strength) and the nature of the
transition, providing the structural links for communication
between the SCO centers. Thus the extent to which an anion or
solvate molecule can hydrogen bond with the SCO center will
likely influence the nature of the transition.1
temperature control of better than ( 0.5 K. IR spectra were recorded
at 298 K using a Brooker Tensor 27 spectrometer in the range of 400-
-
1
4
000 cm . Elemental analyses were done on a Vario EL Mikro
1
Elementaranalysator. H NMR spectroscopic measurements were done
on an Advance DRX Bruker 400 MHz Spectrometer. TGA measure-
ments were performed on a Mettler Toledo TGA/SDTA 851, in the
3
1
00-680 K temperature range in nitrogen atmosphere with a rate of
0 K/min. Mass spectra were recorded on Finnegan MAT 95 (Field
Desorption, FD) and Micromass Q-TOF2 (Electro Spray Ionization,
ESI).
X-ray Structure Determination. The data were collected using an
Oxford Diffraction KM4 κ diffractometer with a Sapphire3 CCD
detector and graphite-monochromated Mo KR radiation (0.71073 Å).
The crystals were mounted on a quartz glass capillary and cooled by
a cold, dry nitrogen gas stream (Oxford Cryosystems equipment). The
temperature stability was ( 0.1 K. Accurate cell parameters were
It has been proposed that hydration will generally result in a
stabilization of the LS state, through hydrogen bonding of the
1
water molecule with the ligand . This indeed seems to be the
1
8
determined and refined using the program CrysAlis CCD. For the
case for most hydrates, but in a cationic SCO system where
the ligand is hydrogen bonded to the associated anion only and
this in turn is bonded to the water molecule, the effect can be
the reverse, that is, loss of water can also result in stabilization
of the LS state. Whatever the rationale for the effects, it is clear
that variation in the nature of the anion or the solvation is a
very readily accessible, if not entirely predictable, means of
potentially modulating the transition temperature or the nature
19
integration of the collected data the program CrysAlis RED was used.
The structure was solved using the direct method with SHELXS-97
20
software, and the solution was refined with SHELXL-97. All non-
hydrogen atoms were refined with anisotropic temperature factors.
CCDC 658238-658240 contains the supplementary crystallographic
data for compounds C -1 and C -2. This data can be obtained free of
6 6
charge from The Cambridge Crystallographic Data Centre via www.c-
cdc.cam.ac.uk/data_request/cif.
1
Synthesis of Materials. Starting reagents and solvents were obtained
commercially from Aldrich or Acros and used as received. The synthesis
of the transition.
In this study we have shown that synchronization of spin state
and liquid crystal transitions is achieved by the search for a
parent SCO system suitable, after attaching the liquid crystal
moiety, to possess LS state or SCO properties at the temperature
where the solidTliquid crystal is foreseen (275-400 K). We
have demonstrated that the required criteria could be satisfied
n n
of the precursors for the ligands C -trenH and C -trenMe is described
in the Supporting Information.
n
General Synthetic Procedure for Compounds C -N (n ) 6, 12,
1
6, 18, 20; N ) 1, 2, 3, 6, 8, 9). General Procedure for Alkoxy-
Substituted Systems. A batch of an alkylated picolinaldehyde and a
stoichometric amount of tris(2-aminoethyl)amine (tren) were dissolved
in ethanol, the calculated amount of iron(II) salt was added which was
immediately accompanied by intensive coloration. The mixture was
stirred for 15 min and then transferred into a refrigerator and left at 4
16
by modifying the original SCO system {Fe[tren(Py)3]}(ClO4)2.
Experimental Section
°
C over night. The crystalline precipitate was filtered off, washed with
ethanol, and dried in vacuo.
-1. tren (0.05 g, 0.34 mmol), excess of crude 5-(hexyloxy)-
picolinaldehyde, Fe(ClO ‚xH O (0.087 g, 0.34 mmol). FT-IR (KBr;
cm ): 2928, 2851 ν(C-H), 1616 ν(CdN), 1090, 622 ν(Cl-O). Anal.
Calcd for C42 FeN 11: C, 52.07; H, 6.55; N, 10.12. Found: C,
1.98; H, 6.53; N, 10.09.
18-1. tren (0.05 g, 0.34 mmol), 5-(octadecyloxy)picolinaldehyde
0.38 g, 1.02 mmol), Fe(ClO ‚xH O (0.087 g, 0.34 mmol). FT-IR
KBr; cm ): 2918, 2850 ν(C-H), 1616 ν(CdN), 1093, 622 ν(Cl-
FeN 11: C, 63.57; H, 9.23; N, 6.65.
Found: C, 63.45; H, 8.82; N, 6.46.
-2. tren (0.05 g, 0.34 mmol), excess of crude 6-methyl-5-
hexyloxy)picolinaldehyde, Fe(ClO ‚xH O (0.087 g, 0.34 mmol). FT-
Physical Measurements. Variable-temperature magnetic susceptibil-
ity measurements of samples of C -1 to C -9 (20-30 mg) were recorded
n
n
C
6
with a Quantum Design MPMS2SQUID susceptometer equipped with
a 7 T magnet, operating at 1 T and at temperatures from 1.8 to 400 K.
4
)
2
2
-1
The susceptometer was calibrated with (NH
4
)
2
Mn(SO
4
2 2
) ‚12H O. Ex-
H63Cl
2
7
O
perimental susceptibilities were corrected for diamagnetism of the
constituent atoms by the use of Pascal’s constants. M o¨ ssbauer spectra
5
C
5
7
were recorded in transmission geometry with a Co/Rh source kept at
room temperature and a conventional spectrometer operating in the
constant-acceleration mode. The samples were sealed in Plexiglass
sample holders and mounted in a nitrogen-bath cryostat. The Recoil
(
(
4
)
2
2
-
1
O). Anal. Calcd for C78
H135Cl
2
7
O
1
.03a M o¨ ssbauer Analysis Software (Dr. E. Lagarec; http://
C
6
www.isapps.ca/recoil/) was used to fit the experimental spectra. DSC
measurements were performed on a Mettler model DSC 822e calibrated
with metallic indium and zinc. DSC profiles were recorded at the rate
of 10 K/min and analyzed with Netzsch Proteus software (NETZSCH-
Geraetebau GMBH, http://www.e-thermal.com/proteus.htm). An overall
accuracy of 0.2 K in the temperature and 2% in the heat flow has been
estimated. X-ray powder diffraction measurements were carried out
with a Seifert TT3300 diffractometer (monochromatic Cu KR radiation).
The temperatures and textures of phase transitions were determined
with a polarization microscope, equipped with a hot stage and with
(
4
)
2
2
(
18) CrysAlis CCD, version 1.171.29; Oxford Diffraction Ltd: Poland, 2006;
Wrocław.
(19) CrysAlis RED, version 1.171.29; Oxford Diffraction Ltd: Poland, 2006;
Wrocław.
(
20) Sheldrick, G. M. SHELXS97/SHELXL97; University of G o¨ ttingen: G o¨ t-
tingen, Germany, 1997.
(
21) The sample was irradiated at 4 K (LS state) with light of λ ) 514 nm (25
mW cm- ) until saturation of the magnetic moment was reached. Then
2
irradiation was switched off and ø
mode at a rate of 2 K/min up to 300 K. At temperatures greater than 70 K,
T increases following the spin-transition curve observed for each
compound.
(22) The sharp rise of ø
M
T versus T was recorded in the warming
ø
M
(
17) (a) Garcia, Y.; van Koningsbruggen, P. J.; Lapouyade, R.; Rabardel, L.;
Kahn, O.; Wieczorek, M.; Bronisz, R.; Ciunik, Z.; Rudolf, M. F. C. R.
Acad. Sci., Ser. II C 1998, 1, 523-532. (b) Lavrenova, L. G.; Ikorski, V.
N.; Varnek, V. A.; Oglezhneva, I. M.; Larionov, S. V. Koord. Khim. 1986,
M
T with increasing temperature below the maximum
conversion is due to the well-known zero-field splitting effect.
(23) The hydrated samples were placed in open containers in the SQUID sample
holder and their magnetism measured. The first part of the experiment
comprised several cycles of heating-cooling from 280 until 350 K (empty
squares). Then, the samples were left for 2 h at the reported temperatures
(indicated by arrows) and the magnetic moment was measured (open
circles). Dehydration taking place under these conditions is confirmed by
the thermogravimetric analysis (SFigure 1).
1
2, 207-215. (c) Varnek, V. A.; Lavrenova, L. G. J. Struct. Chem. 1995,
3
6, 104-111. (d) Lemercier, G.; Verelst, M.; Bousseksou, A.; Varret, F.;
Tuchagues, J.-P. In Magnetism: a supramolecular function; Kahn, O., Ed.;
NATO Advanced Study Institute Series; Kluwer Academic Publishers:
Dordrecht, The Netherlands, 1996; Vol. C484, p 335.
1438 J. AM. CHEM. SOC.
9
VOL. 130, NO. 4, 2008

Spin-State Changes in Spin-Crossover Metallomesogens
A R T I C L E S
IR (KBr; cm-1): 2929, 2855 ν(C-H), 1651 ν(CdN), 1080, 622 ν(Cl-
C
18-9. To MeOH/CHCl
(4:1) medium, tren (0.05 g, 0.34 mmol),
3
O). Anal. Calcd for C45
Found: C, 53.72; H, 6.53; N, 9.82.
12-2. tren (0.05 g, 0.34 mmol), excess of crude 6-methyl-5-
dodecyloxy)picolinaldehyde, Fe(ClO ‚xH O (0.087 g, 0.34 mmol).
FT-IR (KBr; cm ): 2914, 2851 ν(C-H), 1651 ν(CdN), 1088, 622
ν(Cl-O). Anal. Calcd for C45 FeN 11: C, 60.73; H, 8.57; N,
.51. Found C, 59.79; H, 8.37; N, 7.74.
18-2. tren (0.05 g, 0.34 mmol), 6-methyl-5-(octadecyloxy)picoli-
naldehyde (0.40 g, 1.02 mmol), Fe(ClO ‚xH O (0.087 g, 0.34 mmol).
FT-IR (KBr; cm ) 2919, 2850 ν(CH), 1650 ν(CdN), 1088, 622 ν-
Cl-O). Anal. Calcd for C78 FeN 11: C, 63.57; H, 9.23; N,
.65. Found: C, 63.73; H, 9.27; N, 6.66.
16-3. tren (0.05 g, 0.34 mmol), 5-(hexadecyloxy)picolinaldehyde
0.35 g, 1.02 mmol), FeCl ‚4H O (0.07 g, 0.34 mmol). FT-IR (KBr;
cm ): 3400br ν(O-H), 2917, 2850 ν(C-H), 1616 ν(CdN). Anal.
Calcd for C72 FeN 6.5: C, 65.29; H, 9.89; N, 7.40. Found: C,
5.63; H, 9.81; N, 7.31.
18-3. tren (0.05 g, 0.34 mmol), 5-(octadecyloxy)picolinaldehyde
0.38 g, 1.02 mmol), FeCl ‚4H O (0.07 g, 0.34 mmol). FT-IR (KBr;
cm ): 3400br ν(O-H), 2918, 2850 ν(C-H), 1616 ν(CdN). Anal.
Calcd for C78 FeN 6.5: C, 66.50; H, 10.16; N, 6.96. Found:
C, 66.74; H, 10.47; N, 6.69.
H69Cl
2
FeN
7
O11: C, 53.47; H, 6.88; N, 9.70.
5-(octadecyloxy)picolinaldehyde (0.38 g, 1.02 mmol), Fe(ClO
(0.087 g, 0.34 mmol), KI (0.11 g, 0.68 mmol) were added. After the
mixing of components the white precipitate of KClO was filtered off,
the solution was evaporated, and the remainder was recrystallized from
hot EtOH. FT-IR (KBr; cm-1): 2917, 2850 ν(C-H), 1613 ν(CdN).
4 2 2
) ‚xH O
C
4
(
4
)
2
2
-
1
H69Cl
2
7
O
Anal. Calcd for C78
61.83; H, 8.84; N, 6.26.
135 2 7 3
H I N O : C, 61.29; H, 8.90; N, 6.41. Found: C,
7
C
Acknowledgment. We acknowledge the financial help from
the Deutsche Forschungsgemeinschaft (Priority Program 1137
Molecular Magnetism”), the Fonds der Chemischen Industrie.
)
4 2
2
-
1
“
(
H135Cl
2
7
O
A.B.G. thanks the Spanish MEC for a Ramon y Cajal research
contract, for the project CTQ2004-03456-BQU, and the Alex-
ander von Humboldt Foundation for work-visiting fellowships.
Y.G. acknowledges the financial support from the DLR/BMBF
project RUS 05/003 and RFBR Grant No. 03-03-32571. We
also thank Ms. E. Muth and Ms. P. R a¨ der of the Max-Planck-
Institute for Polymer Research, Mainz, for their help in
performing TGA and DSC measurements and Dr. Gutmann and
Mr. M. Bach for help in performing temperature-dependent
XRD measurements.
6
C
(
2
2
-
1
H130Cl
2
7
O
6
C
(
2
2
-
1
H142Cl
2
7
O
Supporting Information Available: Synthesis of the precur-
sors for the ligands Cn-trenH and Cn-trenMe, scheme of synthetic
procedure for compounds Cn-N (n ) 6, 12, 16, 18, 20; N ) 1,
, 3, 6, 8, 9). TGA analysis of materials Cn-3, C18-6, C18-8, and
C18-9. Magnetic properties of compounds Cn-1, Cn-3 to Cn-5
n ) 18, 20), C18-6, C18-8, and C18-9 in the form of øMT versus
T. M o¨ ssbauer spectra of compounds Cn-1, Cn-2, Cn-3, to Cn-5
n ) 18 20), C18-6, C18-7, C18-8, and C18-9, at different
C
20-3. tren (0.05 g, 0.34 mmol), 5-(icosyloxy)picolinaldehyde (0.41
-
1
g, 1.02 mmol), FeCl
2 2
‚4H O (0.07 g, 0.34 mmol). FT-IR (KBr; cm ):
3
400br ν(O-H), 2917, 2850 ν(C-H), 1616 ν(CdN). Anal. Calcd for
FeN 6.5: C, 67.58; H, 10.40; N, 6.57. Found: C, 67.83;
H, 10.12; N, 6.47.
18-6. tren (0.05 g, 0.34 mmol), 5-(octadecyloxy)picolinaldehyde
0.38 g, 1.02 mmol), FeBr (0.073 g, 0.34 mmol), TlF in EtOH/H
0.15 g, 0.68 mmol). After mixing of components white precipitate of
2
C
84
H154Cl
2
7
O
(
C
(
(
2
2
O
(
TlBr was filtered off, the solution was evaporated, and the remainder
temperatures. M o¨ ssbauer parameters for Cn-1, Cn-2, Cn-3, to
Cn-5 (n ) 16, 18, 20), C18-6, C18-7, C18-8, and C18-9. DSC
measurements for C -1, C -2, C -2, C -3 to C -5 (n ) 16,
-
1
was recrystallized from hot EtOH. FT-IR (KBr; cm ): 3400br ν(O-
H), 2917, 2850 ν(C-H), 1613 ν(CdN). Anal. Calcd for C81
FeN 6.5: C, 68.09; H, 10.40; N, 7.13. Found: C, 68.32; H, 10.42; N,
.54.
18-8. tren (0.05 g, 0.34 mmol), 5-(octadecyloxy)picolinaldehyde
148 2
H F -
1
8
12
18
n
n
7
O
1
8, 20), C18-6, C18-7, C18-8, and C18-9 in the heating and cooling
mode performed at the rate of 10 K/min. XRD patterns for C18-
, C12-2, C18-2, Cn-3 to Cn-5 (n ) 16, 18, 20), C18-6, C18-7,
6
C
1
-
1
(
3
C
9
0.38 g, 1.02 mmol), FeBr
400br ν(O-H), 2917, 2850 ν(C-H), 1616 ν(CdN). Anal. Calcd for
FeN : C, 62.55; H, 9.56; N, 6.55. Found: C, 62.15; H,
.33; N, 6.43.
2
(0.073 g, 0.34 mmol). FT-IR (KBr; cm ):
C18-8, and C18-9 recorded at different temperatures. Projection
of the molecular packing of C6-2 along c-axis at 90 K.
78
H
141Br
2
7 6
O
JA077265Z
J. AM. CHEM. SOC.
9
VOL. 130, NO. 4, 2008 1439