Lattice-solvent controlled spin transitions in iron(ii) complexes

Article abstract of DOI:10.1039/b703700g

A series of spin transition (ST) iron(ii) compounds of the type [Fe ^IIL₂](X)₂·{S}₂ (where L is 4′-(4?-cyanophenyl)-1,2′:6′1″- bispyrazolylpyridine, X = ClO₄^- or BF₄ ^-,

Full text of DOI:10.1039/b703700g

View Article Online / Journal Homepage / Table of Contents for this issue
PAPER
www.rsc.org/dalton | Dalton Transactions
Lattice-solvent controlled spin transitions in iron(II) complexes†
Chandrasekar Rajadurai, Zhirong Qu, Olaf Fuhr, Balaji Gopalan, Robert Kruk,
Mohammed Ghafari and Mario Ruben*
Received 12th March 2007, Accepted 7th June 2007
First published as an Advance Article on the web 9th July 2007
DOI: 10.1039/b703700g
A series of spin transition (ST) iron(II) compounds of the type [FeIIL2](X)2·{S}2 (where L is
−
4ꢀ-(4ꢀꢀꢀ-cyanophenyl)-1,2ꢀ:6ꢀ1ꢀꢀ-bispyrazolylpyridine, X = ClO4 or BF4−, and S is acetonitrile) was
synthesized and magnetically investigated. The effects of the removal of the lattice-solvent molecules
and of their different positions relative to the iron(II) cations on the ST process were investigated.
Crystallization yields orange block (A·{S}2) crystals of the composition [FeII(L)2](ClO4)2·{S}2, and two
polymorphic compounds of the stoichiometry [FeII(L)2](BF4)2·{S}2 as red coffin (B·{S}2) and orange
˚
block (C·{S}2) crystals. The Fe–N bond distances of A·{S}2 (from 1.921(9) to 1.992(3) A; at 150 K),
˚
˚
B·{S}2 (from 1.943(2) to 2.017(2) A; at 180 K) and C·{S}2 (from 1.883(3) to 1.962(3) A; at 180 K)
indicate low spin (LS) states of the respective iron(II) ions. Notably, the observed small difference in the
Fe–N distances at 180 K for the two polymorphs B·{S}2and C·{S}2 are due to different positions of the
acetonitrile molecules in the crystal lattices and illustrate the sensitivity of the spin transition properties
on lattice-solvent effects. Variable-temperature single crystal X-ray studies display single-crystal
thermochroism (red (LS) ↔ orange (HS)) for A·{S}2 and B·{S}2 and ca. 3.6% decrease in the unit cell
3
3
˚
˚
volume of A·{S}2 from 4403 A at 300 K to 4278 A at 150 K. The temperature dependent magnetic
susceptibilities of A·{S}2 and B·{S}2 demonstrate systematic increase of the spin transition
temperatures (T1/2) and continuous decreases of the hysteresis loop width (DT1/2) upon slow
lattice-solvent exclusion.
spin transition and removal of solvates from the ST compound
increases the spin transition temperature but mostly at the expense
of the hysteresis loop width.4
Introduction
Spin transition (ST) compounds displaying high transition tem-
peratures (T1/2), wide thermal hysteresis (DT1/2) and thermochro-
ism are increasingly the focus of scientific interest due to their
potential applicability in molecular devices.1–3 It is apparent
that the character of a thermal ST compound is strongly influ-
enced by factors such as ligand-field strength, lattice-solvents,4
counterions1,5 and polymorphism.6 These factors determine the
propensity of iron(II) complexes towards high-spin (HS) ↔ low-
spin (LS) state switching, their abruptness, multistep nature, T1/2
The goal of this work is to demonstrate that, by manipulating the
solvent molecules in crystal lattices of iron(II) complexes, the ST
behaviour can be rationally fine-tuned. Herein, we investigate the
change in the ST parameters (T1/2 and DT1/2) of crystalline iron(II)
complexes (i) by removing the lattice-solvent molecules from the
bulk crystals in a step-wise manner and (ii) by varying the positions
of the solvent molecules within the crystal lattice. Towards this
goal, we report on the single crystal X-ray investigation of
the complexes [FeII(L)2](ClO4)2·2CH3CN (A·{S}2) and of two
polymorphs of [FeII(L)2](BF4)2·2CH3CN (orange (B·{S}2) and red
(C·{S}2 morphologies), where L is (4ꢀ-(4ꢀꢀꢀ-cyanophenyl)-1,2ꢀ:6ꢀ1ꢀꢀ-
bispyrazolylpyridine). Simultaneously, we study the change of
the ST properties of A·{S}2 and B·{S}2 upon in situ slow
removal of the lattice-acetonitrile molecules during the SQUID
magnetic measurements by four continuous heating and cooling
cycles. Room temperature Fourier Transform Infrared (FTIR)
analysis confirms solvate removal thereby supporting the magnetic
investigations. Powder X-ray diffraction (XRD) investigations
further verify the transformation of the crystal lattices due to the
solvate removal.
and DT1/2
.
In ST compounds the presence of an intermolecular solid state
cooperative effect is very important in order to achieve high DT1/2
values. In this context, the lattice-solvents play a critical role
in spreading the elastic interactions2e caused by the Fe–N bond
˚
distance changes of ca. 0.2 A. Importantly, the presence or absence
of lattice-solvent molecules in a ST compound determines the
degree of intermolecular cooperativity in the solid state. Hence
the understanding of the function of lattice-solvent molecules in
a ST compound is essential in order to tune the ST parameters
such as T1/2 and DT1/2. Despite this fact, only a few systematic
studies are dedicated to the investigation of the influence of the
lattice-solvents on the critical T1/2 and DT1/2 parameters of ST–
iron(II) compounds. Sometimes, solvate molecules trap or allow
Experimental
Institute of Nanotechnology, Research Centre, Karlsruhe, PB-3640,
D-76021, Germany. E-mail: Mario.Ruben@int.fzk.de; Fax: +49 724-782-
6781; Tel: +49 724-782-6434
General
4-Cyanophenylboronic acid, Pd(PPh3)4, Fe(BF4)2hydrate, and
Fe(ClO4)2hydrate were obtained from Aldrich and used as
† CCDC reference numbers 634174–634176. For crystallographic data in
CIF or other electronic format see DOI: 10.1039/b703700g
This journal is
The Royal Society of Chemistry 2007
Dalton Trans., 2007, 3531–3537 | 3531
©

View Article Online
XꢀPert diffractometer in Bragg–Brentano geometry with Cu-Ka
˚
(1.54184 A) radiation.
Synthesis of ligand and iron(II) complexes
4-Iodo-2,6-di-pyrazol-1-ylpyridine was synthesized following re-
ported procedures.7
4ꢀ-(4ꢀꢀꢀ-cyanophenyl)-1,2ꢀ:6ꢀ1ꢀꢀ-bispyrazolylpyridine (L). 4-iodo-
2,6-di-pyrazol-1-ylpyridine (0.69 g, 2 mmol) and 4-cyanophenyl-
boronic acid (0.31 g, 2 mmol) was suspended in N2 bubbled
solvents of MeOH–toluene (1 : 1, 100 mL) and 2 M Na2CO3
(8 mL). The mixture was heated to 70 ◦C for 2 d under a
nitrogen atmosphere. The crude reaction mixture was poured
into ice water (500 mL) and extracted with CH2Cl2 solvent. The
separated organic layer was dried over MgSO4 and the solvent
was removed by evaporation. The solid residue was purified by
column chromatography on silica with CH2Cl2 as the eluent
to obtain compound L as a white powder. Yield 0.35 g, 56%.
1H NMR (300 MHz, CDCl3): d = 8.61 (d, 2H), 8.10 (s, 2H),
7.91 (d, 2H), 7.82 (d, 2H), 7.79 (d, 4H), 6.54 (d, 2H) ppm. 13C
◦
NMR (75 MHz, CDCl3, 25 C): 152.48, 152.27, 143.08, 142.30,
133.27, 128.38, 127.70, 118.69, 113.83, 108.71, 107.74 ppm. FT–
IR (KBr): m/cm−1 = 2229 (ms, CN band). MALDI-TOF m/z
(relative intensity of isotopic distribution in %): Observed: 312.69
(100%), 313.70 (22%), 314.71 (2%). Simulated: 312.11 (100%),
313.11 (22%), 314.11(2%).
Scheme 1 Representation of the components used in the synthesis of spin
transition complexes (A·{S}2, B·{S}2 and C·{S}2) and the investigation of
the solvate role in the magnetic properties.
[FeII(L)2](ClO4)2·2CH3CN (A·{S}2). In a 100 mL Schlenk
tube, a solution of L (103 mg, 0.33 mmol) and Fe(ClO4)2·6H2O
received. Toluene, dichloromethane, diisopropyl ether, acetonitrile
and methanol solvents were used without any further purification.
1H and 13C NMR spectroscopic data were recorded on a Bruker
DPX 300 spectrometer with solvent proton as the internal stan-
dard. Infrared spectra were recorded using KBr pressed pellets on
a Perkin Elmer Spectrum GX FT–IR spectrometer. MALDI-TOF
MS data were acquired with a Voyager-DE PRO Bio spectrometry
work station. Elemental analyses were carried out on a Vario
MICRO cube.
◦
(40 mg, 0.15 mmol) in acetonitrile (15 mL) was heated at 80 C
for 6 h. The reaction mixture was cooled to room temperature and
80 mL of diisopropyl ether was added under a N2 flow at RT to
yield an orange–yellow precipitate of the complex A·{S}2. Orange
coloured crystals were grown from diffusing the diisopropyl ether
into an acetonitrile solution of the complex under N2 at RT. Yield
ca. 65 mg. Elementalanalysis calcdfor, C40H30Cl2FeN14O8, C 49.95,
H 3.12, N 20.36%. Found: C 48.86, H 2.92, N 19.96%.
Caution. Although we have experienced no difficulties in han-
dling this compound, metal-organic perchlorates are potentially
explosive and should be handled with care in small quantities.
Magnetic susceptibility
Temperature and solvate dependent static susceptibilities of
complexes A·{S}2 and B·{S}2 were recorded for four continuous
cooling and heating cycles (2 K min−1) using a MPMS-5S (Quan-
tum Design) SQUID magnetometer over a temperature range of
4.5↔380 K in a homogeneous 0.1 Tesla external magnetic field.
Gelatin capsules were used as sample containers for measurements
taken in the temperature range of 4.5↔380 K. The very small
diamagnetic contribution of the gelatin capsule had a negligible
contribution to the overall magnetization, which was dominated
by the sample. The diamagnetic corrections of the molar magnetic
susceptibilities were applied using well-known Pascal’s constants.
[FeII(L)2](BF4)2·2CH3CN (B·{S}2) and (C·{S}2). A similar
procedure as for A was applied using Fe(BF4)2·6H2O (50 mg,
0.15 mmol). Crystallization yielded two distinctive orange block
(B·{S}2) and red coffin (C·{S}2) shaped crystalline polymorphs.
(Yield ca. 50 mg of B·{S}2). Elemental analysis calcd for,
C40H30B2FeN14F8 (B·{S}2), C 51.31, H 3.21, N 20.95%. Found:
C 50.17, H 3.10, N 20.60%.
Result and discussion
Synthesis of complexes
Single crystal and powder X-ray diffraction
Ligand L (4ꢀ-(4ꢀꢀꢀ-cyanophenyl)-1,2ꢀ:6ꢀ1ꢀꢀ-bispyrazolylpyridine) was
obtained via Suzuki cross-coupling reaction of 4-iodo-2,6-di-
pyrazol-1-ylpyridine7 with 4-cyanophenylboronic acid as a white
powder in 56% yield. All complexes were prepared by the
reaction of 2 equivalents of L with one equivalent of the
respective iron(II) salt in acetonitrile at 80 ◦C. Single crystals
of A·{S}2, B·{S}2 and C·{S}2 were obtained quantitatively by
Single X-ray data collection was performed on a STOE IPDS
II diffractometer with graphite monochromated Mo-Ka radiation
˚
(0.71073 A). The structure was solved by direct methods (SHELX-
97). Refinement was performed with anisotropic temperature
factors for all non-hydrogen atoms. The powder diffraction data
of polycrystalline samples were collected by using a Phillips
3532 | Dalton Trans., 2007, 3531–3537
This journal is
The Royal Society of Chemistry 2007
©

View Article Online
diffusing diisopropyl ether into an acetonitrile solution of the
complexes under a N2 atmosphere at room temperature. After
one week, crystallization of complex [FeII(L)2](ClO4)2·2CH3CN
yielded orange block crystals of solvated complex A·{S}2 (Fig. 1).
Crystallization of complex [FeII(L)2](BF4)2·2CH3CN resulted in
two solvated polymorphic forms, with a major amount (ca. 50
mg) of orange block B·{S}2 and a minor amount of red coffin
C·{S}2 morphologies (Fig. 2). These crystals are found to be stable
at room temperature without any decomposition. For magnetic
analysis the two polymorphs (B·{S}2 and C·{S}2) were carefully
separated under a microscope (vide infra). Simple inspection of
the colours of the two polymorphs gives the first indication that
their spin states are different. The colour of the single crystals
also changed from orange (HS state) to dark red (LS state) upon
cooling from 300 to 150 K for both A·{S}2 and B·{S}2 indicating
the ST behaviour of these complexes (see Fig. 3)
X-Ray structure of complex A·{S}2†
Table 1 summarizes the experimental crystallographic data. Se-
lected bond lengths and bond angles are given in Table 2. The
molecular structures of the complexes are displayed in Fig. 1
and 2. Single crystal X-ray structure analysis reveals monoclinic
P2/c symmetry of complex A·{S}2 (Fig. 1). Noticeably, complex
A·{S}2 contains two non-bonded CH3CN solvates per one iron(II)
complex cation. The asymmetric units in A·{S}2 consist of the
[Fe(L)2] cation, two corresponding ClO4 anions (one of them
disordered), and two CH3CN molecules. The complex A·{S}2
Fig. 1 Left figure: ORTEP view of complex dication in the crystal structures of [FeII(L)2](ClO4)2·2CH3CN (A·{S}2) (30% probability ellipsoids) unit.
Hydrogen atoms, ClO4 counteranions and lattice-acetonitrile molecules are omitted for clarity. Right figure: crystal packing diagram of complex A·{S}2.
H atoms have been removed for clarity. [FeII(L)2] dications, counteranions and lattice-acetonitrile molecules are shown in capped sticks, ball and stick
and spacefill representations, respectively.
Fig. 2 Left figures: ORTEP view of polymorphs [FeII(L)2](BF4)2·2CH3CN (B·{S}2) and [FeII(L)2](BF4)2·2CH3CN (C·{S}2) (30% probability
ellipsoids) units. Hydrogen atoms, BF4 counteranions and lattice-acetonitrile molecules are omitted for clarity. Right figures: crystal packing diagram of
B·{S}2 and C·{S}2. [FeII(L)2] dications, counteranions and lattice-acetonitrile molecules are shown in capped sticks model, ball and stick and spacefill
representations, respectively.
This journal is
The Royal Society of Chemistry 2007
Dalton Trans., 2007, 3531–3537 | 3533
©

View Article Online
Table 1 Crystallographic parameters of A·{S}2, B·{S}2 and C·{S}2
[FeII(L)2](ClO4)2·2CH3CN
(A·{S}2)
Two polymorphs of [FeII(L)2](BF4)2·2CH3CN
(B·{S}2)
(C·{S}2)
Formula
C40H30Cl2O8FeN14
961.53
Orange
150(2)
0.71073
Monoclinic
P2/c
10.808(2)
13.925(3)
28.685(6)
90
97.77
C40H30B2F8FeN14
936.25
Orange
180(2)
0.71073
Monoclinic
P2/c
10.691(2)
13.868(3)
28.711(6)
90
97.64
C40H30B2F8FeN14
936.25
Red
Formula weight
Crystal colour
Temperature/K
180(2)
˚
Wavelength/A
0.71073
Orthorhombic
Pna21
19.503(4)
19.129(4)
11.409(2)
90
Crystal system
Space group
˚
a/A
˚
b/A
˚
c/A
a/◦
b/◦
c /◦
90
90
90
90
3
˚
V/A
4277.5(15)
4, 1.493
0.548
1968
0.48 × 0.27 × 0.15
1.49 to 25.71
R1 = 0.0375, wR2 = 0.0892
R1 = 0.0508, wR2 = 0.0940
0.00140(19)
4219.2(15)
4, 1.474
0.443
1904
0.6 × 0.4 × 0.22
1.43 to 25.71
R1 = 0.0564, wR2 = 0.1528
R1 = 0.0679, wR2 = 0.1607
0.0178(13)
4256.3(15)
4, 1.461
0.439
Z, qcalcd/mg m−3
l (Mo-Ka)/mm−1
F(000)
1904
Crystal size/mm
0.45 × 0.27 × 0.23
h range for the data collection/◦
Final R indices [I > 2r(I)]
R indices (all data)
Extinction coefficient
GoF on F2
1.49 to 25.66
R1 = 0.0514, wR2 = 0.1429
R1 = 0.0545, wR2 = 0.1460
0.0219(16)
1.026
1.013
1.042
Fig. 3 The vM T vs. T plot of A·{S}2 and B·{S}2 measured in the temperature range of 4.5↔380 K at four continuous cycles (↓ cooling mode and
↑heating mode) with an applied magnetic field of 0.1 T. Bottom photographs show the temperature dependence of the single crystal colour for the
acetonitrile solvate of A·{S}2 and B·{S}2 (orange and dark red colours correspond to the HS and LS states, respectively).4b
˚
and C·{S}2 consist of the [Fe(L)2] cation, two corresponding
shows Fe–N bond lengths ranging from 1.921(9) to 1.992(3) A
BF4 anions, and two CH3CN molecules. The positions of the
acetonitrile molecules within the crystal lattices of B·{S}2 and
C·{S}2 are different. In the case of B·{S}2, two acetonitrile
molecules are distributed above and below the [Fe(L)2] cation
core, while in C·{S}2 they form a continuous 2D layer structure
along the crystallographic ac-plane (Fig. 2). At 180 K, the two
polymorphs show Fe–N bond lengths ranging from 1.943(2) to
at 150 K indicating the LS state of the iron(II) ion. Variable
temperature single crystal X-ray studies show a dramatic increase
(ca. 3.6%) in the unit cell volume with values of 4278, 4305, 4374
3
˚
and 4403 A at 150, 200, 250 and 300 K temperatures, respectively.
X-Ray structures of polymorphic complexes B·{S}2 and C·{S}2†
˚
˚
The single-crystal X-ray analysis of the two polymorphs B·{S}2
and C·{S}2 reveals monoclinic P2/c and orthorhombic Pna21
symmetry, respectively (Fig. 2). The asymmetric units in B·{S}2
2.017(2) A for B·{S}2 and from 1.883(3) to 1.962(3) A for C·{S}2
which indicate a situation close to the LS state but with more or
less an admixture of a residual HS fraction. The slight differences
3534 | Dalton Trans., 2007, 3531–3537
This journal is
The Royal Society of Chemistry 2007
©

View Article Online
Table 2 Selected bond distances and angles of A·{S}2, B·{S}2 and C·{S}2
agrees with the LS state (S = 0) of an iron(II) ion. In addition,
the first cycle also displays a two step transition centred at ca.
150 K with different T1/2 values in both cooling (T1/2↓ = 224 K)
and heating (T1/2↑ = 218 K) mode measurements and a resulting
ca. 6 K hysteresis loop. It has to be mentioned that the observed
discrepancy between the magnetic moment (high spin) and the
single crystal X-ray structure (low spin) at the same temperature
(180 K) points towards a possible cohabitation of two different
species in the bulk of the crystals due to partial solvent removal,
which is in agreement with the observation of a two-step transition
during the first heating cycle. Unfortunately, due to the poor yield,
it was not possible to measure the temperature dependent magnetic
susceptibilities of polymorph C·{S}2. Nevertheless, the average
A·{S}2
B·{S}2
C·{S}2
Temperature/K
Bond distances/A
150
180
180
˚
Fe(1)–N(9)
Fe(1)–N(3)
Fe(1)–N(1)
Fe(1)–N(5)
Fe(1)–N(7)
Fe(1)–N(11)
1.924(0)
1.921(9)
1.971(9)
1.992(3)
1.984(6)
1.988(4)
1.943(2)
1.945(2)
1.985(2)
1.996(2)
2.011(3)
2.017(2)
1.883(3)
1.891(3)
1.953(3)
1.957(3)
1.960(3)
1.962(3)
Bond angles/◦
˚
N(9)–Fe(1)–N(3)
N(9)–Fe(1)–N(1)
N(3)–Fe(1)–N(1)
N(9)–Fe(1)–N(5)
N(3)–Fe(1)–N(5)
N(1)–Fe(1)–N(5)
N(9)–Fe(1)–N(7)
N(3)–Fe(1)–N(7)
N(1)–Fe(1)–N(7)
N(5)–Fe(1)–N(7)
N(9)–Fe(1)–N(11)
N(3)–Fe(1)–N(11)
N(1)–Fe(1)–N(11)
N(5)–Fe(1)–N(11)
N(7)–Fe(1)–N(11)
174.75(4)
96.07(8)
79.40(1)
105.39(6)
79.21(1)
158.46(4)
79.09(0)
103.64(0)
92.63(3)
89.87(3)
79.03(0)
98.42(6)
92.82(3)
92.84(3)
157.88(6)
173.47(9)
96.39(9)
78.26(9)
106.71(9)
78.81(9)
156.82(9)
78.30(9)
105.51(9)
92.71(9)
90.15(10)
78.34(9)
98.08(9)
93.32(9)
93.20(10)
156.37(9)
179.23(12)
100.37(12)
80.00(11)
99.33(11)
80.29(12)
160.27(11)
80.01(11)
99.32(12)
91.53(13)
90.86(13)
80.08(11)
100.60(12)
92.00(12)
92.40(12)
160.08(11)
Fe–N bond distance (1.982 A) obtained from the single-crystal
X-ray analysis shows that the compound is close to the LS state at
180 K.
In order to elucidate further the role of the lattice-solvates
on the ST properties, the cooling and heating cycle SQUID
measurements of the samples A·{S}2 and B·{S}2 were continued
without interruption (see Table 3). Interestingly, during the 2nd
cycle, both complexes displayed increasing T1/2 (173 K for A·{S}2
and 221 K for B·{S} ) and decreasing DT1/2 values (4 K for
A·{S}2 and 2 K for B·{2S}2). Additional 3rd and 4th measurement
cycles showed further increasing T1/2 values from 181 K to 185 K
and from 231 K to 236 K, respectively for A·{S}2 and B·{S}2.
Moreover, a reduced loop width of only 3 K was observed for
A·{S}2 after the 4th measurement cycle, while the hysteresis loop
had completely vanished for B·{S}2. In addition, a careful look
at the evolution of the ST curves of both A·{S}2 and B·{S}2 at
different cycles reveals striking disparity in their abruptness. In
A·{S}2, a two-step transition is observed during the first cycle,
which becomes more pronounced in the subsequent cycles (from
2nd to 4th).4f The observed two-step transition boundary was
shifted from 173 K to 185 K at the end of the 4th cycle with an
emerging abruptness during the 2nd step. While, the ST curve of
B·{S}2 became smoother upon continuous measurement cycles
and an abrupt transition was observed only during the 4th cycle.
Both complexes A·{S}2 and B·{S}2 display no change in the
magnetic behaviour beyond the 4th cycle. This is interpreted in
terms of complete removal of solvates from the crystal lattice.
in the Fe–N bond distances in B·{S}2 and A·{S}2 at 180 K reveal
the influence of different spatial distributions of the same number
of solvent molecules within the crystal lattice on the ST properties.
Temperature and solvate dependent magnetic properties of A·{S}2
and B·{S}2
The temperature and solvate dependent magnetic susceptibilities
of polycrystalline compounds A·{S}2 and B·{S}2 were measured
at four continuous heating (↑) and cooling (↓) cycles in the
temperature range of 380↔4.5 K (Fig. 3).
During the first cycle, the product of the molar magnetic
susceptibility and temperature (vMT) of compound A·{S}2 is ca.
3.36 emu K mol−1 at 380 K, which is close to the value expected
for a HS state (S = 2) of an iron(II) ion. Upon cooling, the vMT
decreases sharply (with two steps) to a value of ca. 0.09 emu K
mol−1 at 125 K, indicating the LS state (S = 0) of an iron(II) ion.
The first cycle also reveals a broad ca. 34 K hysteresis loop with
different T1/2 values in both cooling (T1/2↓ = 196 K) and heating
(T1/2↑ = 162 K) mode measurements.
Powder XRD data
The phase structures of the isocrystallographic complexes A·{S}2
and B·{S}2 before and after SQUID magnetic measurement cycles
were examined by powder X-ray diffraction (XRD) studies at
293 K. The XRD pattern of A·{S}2 and B·{S}2 are given in Fig. 4.
Noticeably, the single crystal XRD studies performed on a fresh
crystal of A·{S}2 show that the crystal system (monoclinic, P2/c)
remains unchanged during the cooling mode measurements from
For polymorph B·{S}2, the first cooling and heating cycle shows
a vMT value of ca. 3.63 emu K mol−1 at 380 K indicating the HS
state (S = 2) of an iron(II) ion. At 130 K the vMT value sharply
decreases to a minimum value of ca. 0.03 emu K mol−1, which
Table 3 Spin transition temperatures (T1/2) and hysteresis loop widths (DT1/2) of complexes A·{S}2 and B·{S}2 measured during four continuous cooling
(↓) and heating (↑) cycles
[FeII(L)2](ClO4)2·2CH3CN A·{S}2
[FeII(L)2](BF4)2·2CH3CN B·{S}2
Cycles(↓↑)
1/2/K
1
2
3
4
1
2
3
4
T
179
34
173
4
181
4
185
3
166
6
221
2
231
0
236
0
DT1/2/K
This journal is
The Royal Society of Chemistry 2007
Dalton Trans., 2007, 3531–3537 | 3535
©

View Article Online
Fig. 4 Room temperature powder XRD data of A·{S}2 and B·{S}2 before
Fig. 5 FTIR spectra (2500–1800 cm−1 region) of KBr pelleted complexes
(A·{S}2 and B·{S}2) before 1st and after the 4th cooling and heating
cycles. The spectra of ligand L is shown for comparison. * show the –C≡N
stretching frequencies from the lattice-acetonitrile.8
and after the four cooling and heating cycles.
293 K to 150 K ruling out the possibility of any phase transition in
the single crystal XRD measurements conditions. Both powdered
samples of A·{S}2 and B·{S}2 show slightly different peak patterns
apparently as a result of slow solvent loss during the room
temperature powder XRD data collection conditions (see Fig. 4).
The comparison of the room temperature powder XRD data
of A·{S}2 and B·{S}2 with the corresponding calculated low
spin powder XRD data (180 K) reveals completely different
patterns. These results support the fact that the spin transition
behaviours observed during the 1st heating cycle (under a SQUID
environment) in both A·{S}2 and B·{S}2 are accompanied by
phase transitions. This is also in accordance with the observation
of two-step spin transitions for A·{S}2 and B·{S}2 during the
magnetic measurements. To explore further alteration of the
crystal structure due to solvate desorption, the powder XRD
patterns before the 1st measurement cycle and after the 4th
measurement cycle were collected and compared. As expected
complexes A·{S}2 and B·{S}2 show absolutely different patterns
before and after the SQUID measurements, which ascertain that
the final structural change is caused by step-wise desorption
of lattice-acetonitrile molecules from the crystals. In addition,
the initially isocrystallographic compounds A·{S}2 and B·{S}2
develop into two different diffraction patterns after the 4th
magnetic cycle due to solvate removal. This observation is clearly
in agreement with the observation of different magnetic behaviours
observed during the 4th magnetic cycle for both compounds.
the lower and higher frequency bands further confirms the solvate
band assignment.8 Noticeably, both complexes show absolute
disappearance of the two characteristic acetonitrile mC≡N stretching
bands after the completion of the fourth magnetic measurement
cycle (see * in Fig. 5). This result unequivocally confirms the
slow removal of lattice-acetonitrile molecules from the crystalline
samples during the continuous magnetic measurement cycles and
their complete removal at the end of fourth cycle.
The observed increase in T1/2 and decrease in DT1/2 in strong
dependence of the number of heating/cooling cycles can be
attributed to the diminution of solvate assisted cooperativity
between the ST centres due to the slow solvate removal.4 In-
depth Mo¨ssbauer investigations are in progress at different heating
and cooling and cycles in the temperature range of 5–380 K
to investigate the cooperativity associated with the slow solvent
removal and also the relaxation dynamics of the complexes.
Conclusions
In conclusion, the results presented here illustrate the effect of
the removal of lattice-solvent molecules and of their relative
positions within the crystal-lattice on the magnetic spin states
of the iron(II) ion, in particular on the ST parameters (T1/2 and
DT1/2). Despite the complexity of the considered three molecular
systems A·{S}2, B·{S}2, and C·{S}2 exhibiting a high degree of
freedom in the positions of the anions and solvent molecules so
leading to different polymorphs, this study proves that the control
of the lattice-solvent molecules can be used rationally to achieve
technologically attractive spin transition temperatures.
FTIR spectra of complexes
In order to further confirm the entire solvate removal from the
polycrystalline samples, room temperature FTIR measurements
for A·{S}2 and B·{S}2 were performed before and after magnetic
measurements (Fig. 5). The cyano functional group of the ligand L
shows a low frequency mC≡N stretching band centred at 2229 cm−1.
Before the SQUID measurement, both complexes A·{S}2 and
B·{S}2 show characteristic mC≡N stretching frequencies bands
centred at 2223 and 2234 cm−1for the cyano groups of the coor-
dinated ligands. Additionally, complexes A·{S}2 and B·{S}2 show
distinctive low (2251 and 2254 cm−1) and high frequency (2291
and 2292 cm−1) bands for the lattice-acetonitrile molecules. The
frequency separation of ca. 40 cm−1and the typical intensities of
Acknowledgements
The authors thank the Deutsche Forschungsgemeinschaft (DFG)
for financial support within the frame of the project “Kondo
Molecules”. We thank Prof. Dr Horst Hahn for providing
instrument facilities and the referees for constructive comments.
3536 | Dalton Trans., 2007, 3531–3537
This journal is
The Royal Society of Chemistry 2007
©

View Article Online
2003, 9, 4422–4429; (d) E. Breuning, M. Ruben, J.-M. Lehn, F. Renz,
Y. Garcia, V. Ksenofontov, P. Gu¨tlich, E. Wegelius and K. Rissanen,
Angew. Chem., Int. Ed., 2000, 39, 2504–2507.
References
1 For general reviews: (a) Spin Crossover in Transition Metal Compounds I,
Topics in Current Chemistry, ed. P. Gu¨tlich and H. A. Goodwin, Springer,
Berlin, 2004, vol. 233; (b) H. Spiering, Spin Crossover in Transition Metal
Compounds III, Topics in Current Chemistry, Springer-Verlag, Berlin,
2004, vol. 235, p. 171 and references therein; (c) P. Gu¨tlich, Struct.
Bonding, 1981, 83–195; (d) P. Gu¨tlich and A. Hauser, Coord. Chem.
Rev., 1990, 97, 1–22; (e) E. Ko¨nig, Prog. Inorg. Chem., 1987, 35, 527–
623; (f) J. A. Real, A. B. Gaspar, V. Niel and M. C. Mun˜oz, Coord. Chem.
Rev., 2003, 236, 121; (g) A. B. Gaspar, V. Ksenofontov, M. Seredyuk and
P. Gu¨tlich, Coord. Chem. Rev., 2005, 249, 2661; (h) P. Gu¨tlich, Y. Garcia
and T. Woike, Coord. Chem. Rev., 2001, 839, 219–221; (i) P. Gu¨tlich,
A. Hauser and H. Spiering, Angew. Chem., Int. Ed. Engl., 1994, 33,
2024–2054; (j) M. A. Halcrow, Coord. Chem. Rev., 2005, 249, 2880.
2 (a) O. Kahn and C. Jay Martinez, Science, 1998, 279, 44; (b) J. A. Real,
E. Andres, M. C. Mun˜oz, M. Julve, T. Granier, A. Bousseksou and F.
Varret, Science, 1995, 268, 265–267; (c) O. Kahn, J. Kro¨ber and C. Jay,
Adv. Mater., 1992, 4, 718; (d) A. Galet, A. B. Gaspar, M. C. Mun˜oz, G. V.
Bukin, G. Levchenko and J. A. Real, Adv. Mater., 2005, 17, 2949–2953;
(e) I. Sanner, E. Meissner, H. Ko¨ppen and H. Spiering, Chem. Phys.,
1984, 86, 227.
4 (a) A. Galet, M. C. Mun˜oz and J. A. Real, Chem. Commun., 2006,
4321–4323; (b) V. Niel, A. L. Thompson, M. C. Mun˜oz, A. Galet, A. E.
Goeta and J. A. Real, Angew. Chem., Int. Ed., 2003, 42, 3760–3763;
(c) Y. Garcia, P. J. van Koningsbruggen, R. Lapouyade, L. Fourne´s,
L. Rabardel, O. Kahn, V. Ksenofontov, G. Levchenko and P. Gu¨tlich,
Chem. Mater., 1998, 10, 2426–2433; (d) M. Hostettler, K. W. To¨rnroos,
D. Chernyshov, B. Vangdal and N. B. Bu¨rgi, Angew. Chem., Int. Ed.,
2004, 43, 4589–4594; (e) S. M. Neville, B. Moubaraki, K. S. Murray and
C. J. Kepert, Angew. Chem., Int. Ed., 2007, 46, 1–5; (f) G. J. Halder, C. J.
Kepert, B. Moubaraki, K. S. Murray and J. D. Cashion, Science, 2002,
298, 1762.
5 M. Yamada, H. Hagiwara, H. Torigoe, N. Matsumoto, M. Kojima, F.
Dahan, J.-P. Tuchagues, N. Re and S. Iijima, Chem.–Eur. J., 2006, 12,
4536–4549.
6 (a) A. Galet, A. B. Gasper, M. C. Mun˜oz, J. A. Real, G. Levchenko and
J. A. Real, Inorg. Chem., 2006, 45, 9670–9679; (b) A. L. Thompson, A. E.
Goeta, J. A. Real, A. Galet and M. C. Mun˜oz, Chem. Commun., 2004,
1390; (c) D. L. Reger, J. R. Gardinier, M. D. Smith, A. M. Shahin, G. J.
Long, L. Rebbouh and F. Grandjean, Inorg. Chem., 2005, 44, 1852–1866.
7 (a) C. Rajadurai, F. Schramm, S. Brink, O. Fuhr, R. Kruk, M. Ghafari
and M. Ruben, Inorg. Chem., 2006, 45, 10019–10021; see also the ESI;
(b) C. Rajadurai, O. Fuhr, R. Kruk, M. Ghafari, H. Hahn and M. Ruben,
Chem. Commun., 2007, 2636.
3 (a) M. Ruben, F. J. Rojo, J. Romero-Salguero and J.-M. Lehn, Angew.
Chem., Int. Ed., 2004, 43, 3644–3662; (b) M. Ruben, U. Ziener, J.-M.
Lehn, V. Ksenofontov, P. Gu¨tlich and G. B. M. Vaughan, Chem.–
Eur. J., 2005, 11, 94–100; (c) M. Ruben, E. Breuning, J.-M. Lehn, V.
Ksenofontov, F. Renz, P. Gu¨tlich and G. B. M. Vaughan, Chem.–Eur. J.,
8 R. A. Nyquist, Appl. Spectrosc., 1990, 44, 1405–1407.
This journal is
The Royal Society of Chemistry 2007
Dalton Trans., 2007, 3531–3537 | 3537
©