Structural phase transition driven by spin-lattice interaction in a quasi-one-dimensional spin system of [1-(4a?2-iodobenzyl)pyridinium][Ni(mnt) 2]

Article abstract of DOI:10.1021/jp051995g

Crystal structures and magnetic properties were determined for two novel compounds, [1-(4a?2-iodobenzyl)-pyridinium][M(mnt)₂] (mnt ^2- = maleonitriledithiolate; M = Ni (1) or Cu (2)). At room temperature, single crystals of 1 and 2 were isostructural, featuring the formation of segregated columnar structures with regular stacks of cations and anions. For crystal 1, a magnetic transition was observed at a??120 K; furthermore, its magnetic behavior was consistent with that of a regular Heisenberg antiferromagnetic (AFM) chain of S = 1/2 in the high-temperature phase (HT phase) and that of a spin-gap system in the low-temperature phase (LT phase). Such a phenomenon is similar to the spin-Peierls transition. However, the crystal structure of 1 in the LT phase at 100 K revealed that its structural transition is associated with the magnetic transition. Because crystal 2 (S = 0) did not exhibit a structural transition, the structural transition of 1 is driven by spin-lattice interaction. ? 2005 American Chemical Society.

Full text of DOI:10.1021/jp051995g

16610
J. Phys. Chem. B 2005, 109, 16610-16615
Structural Phase Transition Driven by Spin-Lattice Interaction in a
Quasi-One-Dimensional Spin System of [1-(4′-Iodobenzyl)pyridinium][Ni(mnt)2]
X. M. Ren,*,† T. Akutagawa,† S. Nishihara,‡ T. Nakamura,*,† W. Fujita,§ and K. Awaga‡
Research Institute for Electronic Science, Hokkaido UniVersity, Sapporo 060-0812, Japan, and CREST, Japan
Science and Technology Corporation (JST), Kawaguchi 332-0012, Japan, Department of Chemistry, Graduate
School of Science, Nagoya UniVersity, Chikusa-ku, Nagoya 464-8602, Japan, and Research Center of Materials
Science, Nagoya UniVersity, Chikusa-ku, Nagoya, 464-8602, Japan
ReceiVed: April 18, 2005; In Final Form: July 7, 2005
Crystal structures and magnetic properties were determined for two novel compounds, [1-(4′-iodobenzyl)-
pyridinium][M(mnt)2] (mnt2- ) maleonitriledithiolate; M ) Ni (1) or Cu (2)). At room temperature, single
crystals of 1 and 2 were isostructural, featuring the formation of segregated columnar structures with regular
stacks of cations and anions. For crystal 1, a magnetic transition was observed at ∼120 K; furthermore, its
magnetic behavior was consistent with that of a regular Heisenberg antiferromagnetic (AFM) chain of S )
1/2 in the high-temperature phase (HT phase) and that of a spin-gap system in the low-temperature phase (LT
phase). Such a phenomenon is similar to the spin-Peierls transition. However, the crystal structure of 1 in the
LT phase at 100 K revealed that its structural transition is associated with the magnetic transition. Because
crystal 2 (S ) 0) did not exhibit a structural transition, the structural transition of 1 is driven by spin-lattice
interaction.
Introduction
π-π stacking, S‚ ‚ ‚S, or S‚ ‚ ‚M interactions. The magnetic
exchange characteristic is highly sensitive to the intermolecular
contacts and the overlap pattern between neighboring anions.
For example, in the case of adjacent anions of the [M(mnt)2]-
overlap in a slipped M-over-S configuration, the spin-polariza-
tion effect between the large positive spin densities on the M
ions and the small negative spin densities on the S atoms of
the adjacent anions of [M(mnt)2]- may lead to a ferromagnetic
(FM) interaction; in contrast, direct contacts between S‚ ‚ ‚S or
M‚ ‚ ‚M may result in an antiferromagnetic (AFM) interaction6
that an external perturbation could trigger a spin transition.
Accordingly, a key issue in our research is the control of the
[M(mnt)2]- anions during the columnar arrangement. Because
the stacking pattern of the [M(mnt)2]- anions strongly depends
on the molecular topology of the countercation, we proposed
that the formation of a quasi-one-dimensional magnet of
[M(mnt)2]- was attainable by modifying the molecular structure
of the countercation. Derivatives of benzylpyridinium ([RBz-
Py]+) can serve as flexible cations that can be adjusted via
modifying the nature of the groups on the aromatic rings.7,8
Consequently, we have recently prepared a series of [RBzPy]-
[M(mnt)2] ion-pair compounds (the substituent groups of the
cation and the metal ions of the anion are listed in Table 1)
that form a structure with segregated stacks of cation and anion
and show a spin-Peierls-like transition.9 Unfortunately, some
interesting features of this series of compounds remain poorly
understood; for example, although all compounds are isostruc-
tural at room temperature (the anionic stack is regular in the
high-temperature (HT) phase), magnetostructural transition was
observed for some compounds, whereas only magnetic transi-
tion, for others. To investigate this puzzling phenomenon, we
have carried out the design, synthesis, and structural charac-
terization of a series of compounds that do not possess a spin.
Crystalline materials that possess switch property and memory
transduction have a great potential to be used in molecular spin-
electronic or “spintronic” devices, such as next-generation
sensors, molecular switches, and memory devices.1-3 One of
the most spectacular examples of such materials is the spin-
crossover compound, in which a molecular species containing
an octahedrally coordinated transition metal ion with a 3dn (n
) 4-7) electronic configuration exhibits a crossover between
a low-spin state and a high-spin state under external perturba-
tions (change of temperature, application of pressure, light
irradiation, or pulsed magnetic field).3,4 Another molecular
system that has demonstrated a spin-switch property is that of
a radical compound with a quasi-one-dimensional (quasi-1D)
stacking structure. In such a compound, the magnetic interac-
tions within a stack are governed by the intermolecular π orbital
overlap as well as the spin-polarization effect; thus, the property
of the magnetic exchange is sensitive to the intermolecular
contacts; upon application of an external perturbation (that
causes a change in the intermolecular contacts), the spin
transition presumably occurs.2,5,6
Our studies have focused on the design, preparation, and
investigation of a novel spin-transition molecular system that
is based on the molecular architecture of [M(mnt)2]- (mnt2-
)
maleonitriledithiolate; M ) Ni, Pd, or Pt). The formation of
columnar stacks is favorable because such molecules possess
(1) a flat molecular and extended electronic structure and (2)
intermolecular interactions between [M(mnt)2]- anions, such as
* To whom correspondence should be addressed. Phone: +81 11-706-
2849. Fax: +81 11-706-4972. E-mail: xmren@es.hokudai.ac.jp (X.M.R.);
tnaka@imd.es.hokudai.ac.jp (T.N.).
† Hokkaido University and Japan Science and Technology Corporation
(JST).
Herein, we report on two ion-pair compounds [IPyBz]-
‡ Graduate School of Science, Nagoya University.
§ Research Center of Materials Science, Nagoya University.
[M(mnt)2] (IPyBz+ ) 1-(4′-iodobenzyl)pyridinium; M ) Ni or
10.1021/jp051995g CCC: $30.25 © 2005 American Chemical Society
Published on Web 08/11/2005

Phase Transition Driven by Spin-Lattice Interaction
J. Phys. Chem. B, Vol. 109, No. 35, 2005 16611
TABLE 1: Reported Compounds in the Series of
[IPyBz][Cu(mnt)2] (2). The procedure as described above
for 1, but with [1-(4′-iodobenzyl)pyridinium]2[Cu(mnt)2] as the
starting material, was followed to afford 2 (yield: 442 mg, 69%).
Single crystals suitable for X-ray structure analysis were grown
by slowly cooling an acetonitrile solution of 2. Elemental
analysis: Calcd for C20H11N5S4ICu: C, 37.5; H, 1.73; N, 10.9%.
Found: C, 37.3; H, 1.75; N, 10.6%.
[R1R2R3BzPy][M(mnt)2] that Exhibit a Magnetic Transition
Magnetic Susceptibility and Heat Capacity Measurements.
Magnetic susceptibility measurements for polycrystalline samples
over the range 2-350 K and for a crystal with dimensions of
3.0 × 1.0 × 0.2 mm3 and a mass of 0.40 mg over the range
2-300 K were carried out using a Quantum Design MPMS-
XL superconducting quantum interference device (SQUID)
magnetometer under 1.0 T. Heat capacity measurements were
performed following the relaxation method using a Quantum
Design Physical Property Measurement System (PPMS) over
the range 2-150 K on both heating and cooling processes. A
crystal with dimensions of 3.0 × 1.0 × 0.2 mm3 and a mass of
0.40 mg was attached to the sample platform with a small
amount of grease.
R1
R2
R3
M
reference
I
H
H
H
H
H
H
NH2
NH2
NH2
H
H
H
H
H
H
F
H
H
H
H
Ni
Ni
Ni
Ni
Pt
Ni
Ni
Ni
Ni
Ni
present paper
NO2, Cl, Br
8a
8b
8c
8d
8e
8f
8g
8h
8i
F
CN
NO2, Cl, Br
Br
Br
CN
Cl
vinyl
X-ray Structural Analyses. Crystallographic data were
collected using a Rigaku Raxis-Rapid diffractometer with Mo
KR (λ ) 0.710 73 Å) radiation from a graphite monochromator.
Structure refinements were performed using the full-matrix least-
squares method on F2. Calculations were performed using
SHELXL-97 software packages.11 Parameters were refined using
anisotropic temperature factors, with the exception of those for
the hydrogen atom. The hydrogen atoms were introduced at
calculated positions. Details of the crystal parameters, data
collection, and refinement for 1 and 2 at 293 and 100 K are
summarized in Table 2. The bond lengths for the [M(mnt)2]-
(M ) Ni or Cu) anions, along with their estimated standard
deviations, are listed in Supporting Information Table S1. Some
important intermolecular distances are listed in Table 3. The
determination of the crystal structure of 1 at low temperatures
(below ∼140 K) was further complicated by the inaccuracy of
the data due to diffuse scattering that is driven by magnetoelastic
interactions. To minimize the effect of diffuse scattering, the
diffraction intensities of 1 at 100 K were obtained using a
reduced exposure time of 40 s. Although the crystal structure
was consequently solved and refined, parameters such as wR2
and R1 remained somewhat high. Nonetheless, the main
difference in the structures of the HT and low-temperature (LT)
phases was undoubtedly in accordance with the crystal data
analyses.
Cu). The crystals of these two compounds are isostructural with
each other at room temperature, in which the segregated stacks
of cations and anions are regular. The [Ni(mnt)2]- compound
formed a one-dimensional regular AFM chain system with S )
1/2, which exhibited an abrupt drop in magnetic susceptibility
(a spin-Peierls-like transition) below ∼120 K, whereas the
[Cu(mnt)2]- compound (without spin) did not show a structural
transition.
Experimental Section
Preparation of Compounds. Na2mnt, [IPyBz]Br,7 and
[IPyBz]2[M(mnt)2] (M ) Ni or Cu) were prepared according
to published procedures.10
[IPyBz][Ni(mnt)2] (1). To a solution of [IPyBz]2[Ni(mnt)2]
(932 mg, 1.0 mmol) in acetonitrile (5 mL) was added I2 (150
mg, 0.6 mmol) and methanol (30 mL). The mixture was stirred
for 10 min and then placed in a refrigerator at 4 °C for 24 h.
The resulting brown microcrystalline was collected by filtration,
washed with methanol, and dried under vacuum (yield: 520
mg, 82%). Single crystals suitable for X-ray structure analysis
were obtained by diffusing diethyl ether into an acetonitrile
solution of 1. Elemental analysis: Calcd for C20H11N5S4INi:
C, 37.8; H, 1.75; N, 11.0%. Found: C, 37.9; H, 1.91; N, 10.9%.
TABLE 2: Crystallographic Data for 1 and 2a
1 (293 K)
1 (100 K)
2 (293 K)
2 (100 K)
molecular formula
CCDC no.
Mr
space group
a/Å
b/Å
C20H11IN5NiS4
CCDC-260866
635.18
C20H11IN5NiS4
CCDC-260867
635.18
C20H11IN5CuS4
CCDC-260868
640.03
C20H11IN5CuS4
CCDC-260869
640.03
P21/a
7.473(2)
26.373(9)
11.883(4)
90
102.34(1)
90
2287(1)
4
P21/a
P1h
P21/a
7.6048(15)
26.742(5)
12.057(2)
90
102.727(3)°
90
2391.8(8)
4
2.469
7.387(3)
26.18(1)
11.886(8)
88.57(2)
77.28(2)
87.29(2)
2239(2)
4
7.625(1)
26.524(6)
11.984(3)
90
102.614(9)
90
2365.3(8)
4
2.600
c/Å
R/deg
â/deg
γ/deg
V/Å3
Z
µ/mm-1
λ/Å
2.637
0.710 73
1.884
2.688
0.710 73
1.858
0.710 73
1.764
0.0471
0.710 73
1.797
0.039
F/g cm-3
R1
0.248
0.028
wR2
0.1144
0.559
0.128
0.074
2
2
2
a R1 ) ∑(||F0| - |Fc||)/∑|F0|, wR2 ) ∑w(|F0| - |Fc| )2/∑w(|F0| )2]1/2
.

16612 J. Phys. Chem. B, Vol. 109, No. 35, 2005
Ren et al.
Figure 1. Plots of øm and ømT vs T of a polycrystalline sample of 1.
(upper) Open circles represent experimental data; the solid lines
represent the best fit data. (bottom) Plot of ømT reproduced by
subtracting the contributions of diamagnetism and possible van Vleck
paramagnetism (see text).
TABLE 3: Intermolecular and Interplane Distances (Å) in
Both the HT and LT Phases
1 (293 K)
1 (100 K)
2 (293 K)
2 (100 K)
d1
d2
h1
h2
c1
c2
g1
g2
3.981
3.981
3.602
3.635
4.598
4.598
3.699
3.885
3.719
3.889
3.512
3.518
4.393
4.535
3.574
3.793
4.004
4.004
3.578
3.582
4.579
4.579
3.714
3.898
3.913
3.913
3.499
3.450
4.489
4.489
3.636
3.826
Details about Molecular Orbital Calculations. Extended
Hu¨ckel molecular orbital calculations12a were carried out using
the CAESAR2 and SAMOA program packages.12b Double-ú
Slater-type orbitals13 were used for the C, N, and S atoms as
well as for the Ni 3d orbitals.
Results and Discussion
Magnetic Susceptibilities. The temperature dependence of
the magnetic susceptibility (øm) for polycrystalline 1 is displayed
in Figure 1. The value of ømT per mole of 1 at 350 K is 0.322
emu‚K‚mol-1, which is slightly lower than the expected value
for an independent spin of S ) 1/2. The initial smooth decrease
of the ømT values, with cooling temperatures, is indicative of
antiferromagnetic exchange interactions between the nearest
neighboring spins. The abrupt drop of the ømT value at ∼120
K implies the occurrence of a magnetic transition. The
subsequent drop to zero at around 90 K indicates that 1 is nearly
diamagnetic below 90 K. When the temperature was increased
from 2 to 350 K, a very small hysteresis loop (less than 1 K)
was detected, as illustrated in the inset of Figure 1. Temperature
dependences of the magnetic susceptibility of a single crystal
at 1.0 T parallel to the a- and c-axes (cf. Supporting Informa-
tion Figure S1) nearly coincided with each other but parallel
to the b-axis exhibited small differences from the other two
cases.
Figure 2. (a) Molecular structure of 1 and the segregated stacks of
cations and anions along the a-axis; (b) top view; (c) side view.
of various [Ni(mnt)2]- compounds.9 Both the anions and the
cations constructed segregated stacks that were parallel to the
a-axis (Figure 2b and c). Within an anionic stack, the face-to-
face anions slid along the b-axis, to eclipse each other. The
distances between the [Ni(mnt)2]- molecular planes, which are
defined by the Ni ion and four coordinated S atoms, are h1 )
3.602 Å and h2 ) 3.635 Å (Figure 3a), while the adjacent Ni‚
‚ ‚Ni distances are identical (d1 ) d2 ) 3.981 Å), even if the
atoms lie on a general point position (cf. note 14). Within a
cationic stack, the neighboring cations overlap in a boat
conformation, as illustrated in Figure 3b. The central-to-central
distances of the adjacent benzene rings are c1 ) c2 ) 4.598 Å,
and the corresponding interplanar spaces are g1 ) 3.699 and g2
) 3.885 Å.
Crystal Structures. Because 1 and 2 are isostructural at 293
K, only the detailed structure of 1 is presented. The asymmetric
unit within a cell that is comprised of a coupled [Ni(mnt)2]-
anion and IPyBz+ cation of 1 is shown in Figure 2a. Most of
the bond lengths and angles are in good agreement with those
In crystal 2, the [Cu(mnt)2]- anions and the IPyBz+ cations
are stacked in columns that are similar to those in 1. The bond

Phase Transition Driven by Spin-Lattice Interaction
J. Phys. Chem. B, Vol. 109, No. 35, 2005 16613
Figure 3. Separations (Å) between neighboring ions within (a) an anion
stack and (b) a cationic stack.
Figure 5. Temperature dependence of the molar heat capacity of 1.
The open stars and circles represent data in heating and cooling
processes, respectively. The inset shows the excess heat capacity around
the phase transition temperature.
discontinuous change; below 120 K, however, a sharp drop is
observed, thus indicating the existence of a structural transition
in 1.
Except for the shorter intermolecular distances at the lower
temperature, the molecular and stacking structures for crystal 2
at 100 and 293 K were nearly identical. In contrast to 1,
however, uniform compression upon cooling was observed;
specifically, the Cu‚ ‚ ‚Cu distance between adjacent anions in
a stack and the central-to-central distance between adjacent
benzene rings remained unchanged (Supporting Information
Table S1), thus implying that a structural transition does not
occur for 2.
Heat Capacity. The results of the heat capacity measurements
for 1 are shown in Figure 5. A sharp thermal abnormality with
λ shape was observed at ∼120 K. Although the Cp(T)-T plots
in the heating and cooling procedures are superimposable, a
small but significant supercooling/heating phenomenon (less
than 1 K), which coincided with the small hysteresis loop in
the susceptibility measurements, was observed. Furthermore,
the peak temperature at 117 K in the heating procedure coincides
with the temperature at which the abrupt drop of magnetic
susceptibility occurs. The transition in 1, therefore, is not a spin-
Peierls transition.
In the first-order approximation, a normal heat capacity would
appear as a smooth curve.16 Upon subtraction of the contribution
of such a normal heat capacity, the excess heat capacity around
the transition is displayed in the inset of Figure 5. The entropy
of this transition (which includes the contribution of both
structural and spin) is estimated as 4.34 J‚K-1‚mol-1. Because
this value is lower than the theoretical maximum of spin entropy
for a mole of S ) 1/2 ion (R ln 2 ≈ 5.76 J‚K-1‚mol-1),17 it can
be suggested that a substantial short-range order persists over
the temperature of the transition of 1. The short-range order is
an indication of a lowered dimensionality of the magnetic spin
system and is consistent with the chain structure of the
[Ni(mnt)2]- anions in crystal 1.18
Structurally, crystal 1 should exist as a nearly perfect 1D spin
system because the anionic stacks are separated by the diamag-
netic cationic stacks. Magnetic interactions between the anionic
stacks, therefore, should be significantly less than that within
an anionic stack. Qualitative theoretical analyses of the magnetic
exchange interactions between the nearest spins in both the HT
Figure 4. Temperature dependence of the relative cell parameters of
1, which is defined as P(T)/P(293 K), where P(T) represents the lattice
parameters at T K.
lengths and angles for the [Cu(mnt)2]- anion are in good
agreement with those for [TBA][Cu(mnt)2]15 (summarized in
Supporting Information Table S1). To highlight the comparisons
to 1, important intermolecular distances are listed in Table 3
and illustrated in Figure 3; these include c1 and c2 (center-to-
center distances of neighboring benzene rings), d1 and d2
(distances of intermetal ions in an anionic stack), g1 and g2
(interplanar distances between neighboring benzene rings), and
h1 and h2 (interplanar distances between the mean plane defined
by the M ion and coordinated S atoms).
For 1, the space group decreases from P2(1)/a in the HT
phase to P1h in the LT phase at 100 K.15 Moreover, the
asymmetric unit switches from a IPyBz+ and [Ni(mnt)2]- pair
in the HT phase into two ion pairs in the LT phase. Although
the molecular structures are nearly identical, distinct differences
were observed between the stacking structures in the LT and
HT phases. At the phase transition, nonuniform compression
and slippage of both the cation and the anion stacks cause (1)
alternating distances of the adjacent Ni‚ ‚ ‚Ni in an anionic stack
(d1 ) 3.719 Å and d2 ) 3.889 Å) and (2) distortion within a
cationic stack as reflected in the central-to-central separations
between adjacent benzene rings (g1 ) 4.393 Å and g2 ) 4.535
Å). The temperature dependences of the relative lattice param-
eters, which are defined as Pcell(T)/Pcell(293 K), where Pcell(T)
and Pcell(293 K) represent the cell parameters at T and 293 K,
respectively, are depicted in Figure 4. Upon cooling to 120 K,
all parameters decrease smoothly without any significant

16614 J. Phys. Chem. B, Vol. 109, No. 35, 2005
Ren et al.
1
21
Heisenberg model of linear chain with S ) /2 using
2
Ng2µB
A + BX-1 + CX-2
1 + DX-1 + EX-2 + FX-3
øm )
+ ø0
(2)
kBT
where X ) kBT/|J|, in which J is the magnetic exchange constant
of the neighboring spins in a magnetic chain, the coefficients
A-F in the power series are A ) 0.25, B ) 0.149 95, C )
0.300 94, D ) 1.9862, E ) 0.688 54, and F ) 6.0626, and ø0
represents the sum of the diamagnetic and the Van Vleck
paramagnetic components. The best fit to eq 2 resulted in |J|/
kB ) 46.0 K and ø0 ) -2.6 × 10-4 emu‚mol-1 with a fixed
value of g ) 2.017.22 Because the magnetic susceptibility
exhibited a spin-gap feature in the LT phase, best fit calculations
were carried out using23
Figure 6. Schematic (a) top and (b) side views of the spin dimers in
crystal 1. J1 and J2 represent magnetic exchange interaction within a
stack, whereas J3, J4, and J5 represent those between stacks.
R
T
C
T
øm ) exp(-∆/T) + + ø0
(3)
TABLE 4: Values of (∆e)2 a and Ni‚ ‚ ‚Nib Distances of
Various Spin-Exchange Paths at 293 and 100 K for 1,
Where the Spin Dimers Were Identified by Specifying Their
Ni‚ ‚ ‚Ni Distances
where the fitting parameters are R ) 17.2 emu‚K-1‚mol-1, ∆
) 606.2 K, C ) 1.0 × 10-3 emu‚K‚mol-1, and ø0 ) -1.4 ×
10-4 emu‚mol-1
.
The observed phenomena in 1 are very similar to that of a
spin-Peierls transition. However, evidence obtained from the
crystal structure analyses in the HT and LT phases as well as
the temperature dependence of the lattice parameters clearly
demonstrate that the transition is that of a magnetostructural
type. In contrast, crystal 2, which is isostructural with 1 and is
diamagnetic, does not exhibit a structural transition over the
same temperature region. These observations imply that spin-
lattice interactions play an important role in the structural
transition of 1 and that the structure transition of 1 is driven by
spin-lattice interactions.
293 K
100 K
spin dimer
distance
(∆e)2
distance
(∆e)2
(a) Intrastack
8226
J1
J2
3.981
3.981
3.719
3.889
450644
21228
8226
(b) Interstack
J3
J4
J5
11.876
15.038
14.617
0
0
0
11.654
14.808
14.433
2025
2025
2025
a In units of (meV)2. b In units of Å.
and LT phases were carried out using the extended Hu¨ckel
molecular orbital calculation.12 With respect to a spin system
with AFM exchange interaction, the magnetic exchange energy
(JAF) between two spin sites (spin dimer) is described as
Conclusion and Remarks
Investigations were carried out to study a peculiar magnetic
transition for [IPyBz][Ni(mnt)2] (1), which is typically a regular
Heisenberg AFM linear chain in the HT phase but opens a spin
gap in the LT phase. Such behavior is similar to that of a spin-
Peierls transition. A sharp thermal anomaly with a λ shape,
which corresponded with this transition in Cp measurements,
indicated the transition is a first-order one. A supercooling/
heating phenomenon consisting of a small hysteresis loop (less
than 1 K) was observed in magnetic susceptibility measure-
ments. Crystal structural analyses, above and below the transi-
tion, as well as the temperature dependence of the cell
parameters revealed that the transition is that of a magneto-
structural type. In contrast, a structural transition was not
observed for 2, which is diamagnetic and isostructural with 1
at room temperature. These results indicate that the structural
transition in 1 is driven by spin-lattice interactions.
(∆e)2
Ueff
JAF
-
(1)
where the spin-orbital interaction energy (∆e) is defined as
the energy split between the two magnetic orbitals of a
2-
[Ni(mnt)2]2 spin dimer and Ueff is the effective on-site
repulsive energy (nearly constant for a defined magnetic solid).
Therefore, JAF can be estimated with reasonable accuracy using
the corresponding (∆e)2.19,20 Spin interactions between adjacent
spins in both the HT and LT phases are schematically illustrated
in Figure 6. The results of the calculations are presented in Table
4. The values of (∆e)2 show that the magnetic interactions of
J1 and J2 within a stack are identical, whereas those of J3, J4,
and J5 between the stacks are zero in the HT phase. In the latter
three cases, the corresponding highest occupied molecular orbital
(HOMO) exists as a 2-fold degeneracy. Therefore, magnetic
interactions do not exist between two spin sites, and the energy
split between the two magnetic orbitals is zero. These results
demonstrate that this spin system is a perfectly regular one-
dimensional AFM chain in the HT phase. Within a stack in the
LT phase, however, the regular magnetic chain in the HT phase
is distorted to yield unequal values of J1 and J2. Moreover, finite
magnetic interaction between stacks was observed. The magnetic
behavior of 1 in the HT phase is approximately described as
that of a uniform Heisenberg linear chain system. Magnetic
susceptibility in the HT phase was simulated by a regular
Acknowledgment. This work was partly supported by a
Grant-in-Aid for Science Research from the Ministry of Educa-
tion, Culture, Sports, Science and Technology of Japan. The
authors thank Prof. Nomura and Dr. Ichimura for the use of
the SQUID magnetometer. X.M.R., a JSPS fellow (ID no.
P03271), thanks the Japan Society for the Promotion of Science
for financial support.
Supporting Information Available: Tables of the bonding
distances for the anionic moieties of 1 and 2, plots of øm(T)-T
of a single-crystal sample of 1 and a polycrystalline sample of
2, and X-ray crystallographic files (CIF) of 1 and 2 in both the
HT and LT phases. This material is available free of charge via
the Internet at http://pubs.acs.org.

Phase Transition Driven by Spin-Lattice Interaction
J. Phys. Chem. B, Vol. 109, No. 35, 2005 16615
Z. F.; Meng, Q. J. Inorg. Chem. Commun. 2004, 7, 1034. (g) Ni, C. L.; Li,
Y. Z.; Dang, D. B.; Ni, Z. P.; Tian, Z. F.; Yuan, Z. R.; Meng, Q. J. Inorg.
Chem. Commun. 2005, 8, 105. (h) Ni, C. N.; Dang, D. B.; Li, Y. Z.; Gao,
S.; Ni, Z. P.; Tian, Z. F.; Meng, Q. J. J. Solid State Chem. 2005, 178, 100.
(i) Dang, D. B.; Ni, C. L.; Bai, Y.; Tian, Z. F.; Ni, Z. P.; Wen, L. L.;
Meng, Q. J.; Gao, S. Chem. Lett. 2005, 34, 680.
(10) Davison, A.; Holm, H. R. Inorg. Synth. 1967, 10, 8.
(11) Sheldrick, G. M. SHELXL-97, program for the refinement of crystal
structure; University of Go¨ttingen: Go¨ttingen, Germany, 1997.
(12) (a) Hoffmann, R. J. Chem. Phys. 1963, 39, 1397. (b) Dai, D.; Ren,
J.; Liang, W.; Whangbo, M.-H. CAESAR2 and SAMOA (http://chvam-
w.chem.ncsu.edu).
References and Notes
(1) (a) Migliori, J. M.; Reiff, W. M.; Arif, A. M.; Miller, J. S. Inorg.
Chem. 2004, 43, 6875. (b) Brusso, J. L.; Clements, O. P.; Haddon, R. C.;
Itkis, M. E.; Leitch, A. A.; Oakley, R. T.; Reed, R. W.; Richardson, J. F.
J. Am. Chem. Soc. 2004, 126, 8256. (c) Letard, J. F.; Guionneau, P.; Goux-
Capes, L. Top. Curr. Chem. 2004, 235, 221. (d) Galet, A.; Niel, V.; Mun˜oz,
M. C.; Real, J. A. J. Am. Chem. Soc. 2003, 125, 14224. (e) Itkis, M. E.;
Chi, X.; Cordes, A. W.; Haddon, R. C. Science 2002, 296, 1443.
(2) Fujita, W.; Awaga, K. Science 1999, 286, 261.
(3) Kahn, O.; Martinez, C. J. Science 1998, 279, 44.
(4) (a) Reger, D. L.; Gardinier, J. R.; Gemmill, W. R.; Smith, M. D.;
Shahin, A. M.; Long, G. J.; Rebbouh, L.; Grandjean, F. J. Am. Chem. Soc.
2005, 127, 2303. (b) Dorbes, S.; Valade, L.; Real, J. A.; Faulmann, C. Chem.
Commun. 2005, 69. (c) Fujigaya, T.; Jiang, D. L.; Aida, T. J. Am. Chem.
Soc. 2003, 125, 14690. (d) Ko¨nig, E. Struct. Bonding (Berlin) 1991, 76,
51. (e) Gu¨tlich, P. Struct. Bonding (Berlin) 1981, 44, 83.
(5) (a) Bray, J. W.; Hart, H. R.; Interrante, L. V., Jr.; Jacobs, L. S.;
Kasper, J. S.; Watkins, G. D.; Wee, S. H. Phys. ReV. Lett. 1975, 35, 744.
(b) Jacobs, I. S.; Bray, J. W.; Hart, H. R.; Interrante, L. V.; Kasper, J. S.;
Watkins, G. D.; Prober, D. E.; Bonner, J. C. Phys. ReV. B 1976, 14, 3036.
(c) Lovett, B. W.; Blundell, S. J.; Pratt, F. L.; Jestadt, T.; Hayes, W.; Tagaki,
S.; Kurmoo, M. Phys. ReV. B 2000, 61, 12241. (d) Huizinga, S.;
Kommandeur, J.; Sawatzky, G. A.; Thole, B. T.; Kopinga, K.; de Jonge,
W. J. M.; Roos, J. Phys. ReV. B 1979, 19, 4723. (e) Obertelli, S. D.; Friend,
R. H.; Talham, D. R.; Kurmoo, M.; Day, P. J. Phys.: Condens. Matter
1989, 1, 5671. (f) Liu, Q.; Ravy, S.; Pouget, J. P.; Coulon, C.; Bourbonnais,
C. Synth. Met. 1993, 55-57, 1840. (g) Mukai, K.; Wada, N.; Jamali, J. B.;
Achiwa, N.; Narumi, Y.; Kindo, K.; Kobayashi, T.; Amaya, K. Chem. Phys.
Lett. 1996, 257, 538.
(6) (a) Coomber, A. T.; Beljonne, D.; Friend, R. H.; Bre´das, J. L.;
Charlton, A.; Robertson, N.; Underhill, A. E.; Kurmoo, M.; Day, P. Nature
1996, 380, 144. (b) Nishijo, J.; Ogura, E.; Yamaura, J.; Miyazaki, A.; Enoki,
T.; Takano, T.; Kuwatani, Y.; Iyoda, M. Solid State Commun. 2000, 116,
661. (c) Xie, J. L.; Ren, X. M.; Song, Y.; Zou, Y.; Meng, Q. J. J. Chem.
Soc., Dalton Trans. 2002, 2868.
(7) Bulgarevich, S. B.; Bren, D. V.; Movshovic, D. Y.; Finocchiaro,
P.; Failla, S. J. Mol. Struct. 1994, 317, 147.
(8) Montaudo, G.; Caccamese, S.; Finocchiaro, P. J. Am. Chem. Soc.
1971, 93, 4202.
(13) Clementi, E.; Roetti, C. Atomic Data Nuclear Data Tables 1974,
14, 177.
(14) Notes: In fact, the Ni(1) atom lies on a general point position, and
the two neighboring Ni(1)i and Ni(1)ii atoms in a stack are generated by
the symmetric transforms, where i ) -0.5 + x, 0.5 - y, z and ii ) 0.5 +
x, 0.5 - y, z. Defining the fractional coordinates of Ni(1) as (x, y, z), the
coordinates of the two neighbors, (0.5 + x, 0.5 - y, z) of Ni(1)ii and (-0.5
+ x, 0.5 - y, z) of Ni(1)i, are deduced according to the crystal data.
Therefore, the Ni(1)‚ ‚ ‚Ni(1)ii and Ni(1)‚ ‚ ‚Ni(1)i distances are equal to
the moduli of the corresponding vectors of (0.5, 0.5 - 2y, 0) and (-0.5,
0.5 - 2y, 0), respectively, which coincide on the XY-plane of the
crystallographic coordinate system. The two-dimensional coordinate system
(X-O-Y) is orthogonal in the monoclinic system; accordingly, the lengths
of the vectors should be given by |Ni(1)‚ ‚ ‚Ni(1)i|| ) |(0.5, 0.5 - 2y, 0)|
) {0.52 + (0.5 - 2y)2 + 02}1/2 and |Ni(1)‚ ‚ ‚Ni(1)i| ) |(-0.5, 0.5 - 2y,
0)| ) {(-0.5)2 + (0.5 - 2y)2 + 02}1/2 ) {0.52 + (0.5 - 2y)2 + 02}1/2, so
that |Ni(1)‚ ‚ ‚Ni(1)i| ) |Ni(1)‚ ‚ ‚Ni(1)ii|. In other words, the interatomic
distances of Ni(1)‚ ‚ ‚Ni(1)i and Ni(1)‚ ‚ ‚Ni(1)ii are identical.
(15) Forrester, J. D.; Zalkin, A.; Templeton, D. H. Inorg. Chem. 1964,
3, 1507.
(16) Saito, K.; Massalska-Arodz, M.; Ikeuchi, S.; Maekawa, M.;
Sciesinski, J.; Sciesinska, E.; Mayer, J.; Wasiutynski, T.; Sorai, M. J. Phys.
Chem. B 2004, 108, 5785.
(17) Fujita, W.; Awaga, K.; Nakazawa, Y.; Saito, K.; Sorai, M. Chem.
Phys. Lett. 2002, 352, 348.
(18) Bhatia, S. N.; O’Connor, C. J.; Carlin, R. L.; Algra, H. A.; De
Jongh, L. J. Chem. Phys. Lett. 1977, 50, 353.
(19) Whangbo, M.-H.; Koo, H.-J.; Dai, D. J. Solid State Chem. 2003,
(9) (a) Ren, X. M.; Meng, Q. J.; Song, Y.; Lu, C. S.; Hu, C. J.; Chen,
X. Y. Inorg. Chem. 2002, 41, 5686. (b) Xie, J. L.; Ren, X. M.; Song, Y.;
Zheng, W. W.; Liu, W. L.; He, C.; Meng, Q. J. Chem. Commun. 2002,
2346. (c) Xie, J. L.; Ren, X. M.; He, C.; Gao, Z. M.; Song, Y.; Meng, Q.
J.; Kremer, R. K. Chem. Phys. Lett. 2003, 369, 41. (d) Ren, X. M.; Okudera,
H.; Kremer, R. K.; Song, Y.; He, C.; Meng Q. J.; Wu, P. H. Inorg. Chem.
2004, 43, 2569. (e) Ni, C. L.; Dang, D. B.; Song, Y.; Gao, S.; Li, Y. Z.;
Ni, Z. P.; Tian, Z. F.; Wen, L. L.; Meng, Q. J. Chem. Phys. Lett. 2004,
396, 353. (f) Ni, C. L.; Dang, D. B.; Li, Y. Z.; Yuan, Z. R.; Ni, Z. P.; Tian,
176, 417 and references therein.
(20) Whangbo, M.-H.; Dai, D.; Koo, H.-J. Dalton. Trans. 2004, 3019.
(21) (a) Bonner, J. C.; Fisher, M. E. Phys. Rev. A 1964, 135 (3A), A640.
(b) Hatfild, W. E. J. Appl. Phys. 1981, 52, 1985.
(22) Ren, X. M.; Kremer, R. K.; Meng, Q. J. J. Magn. Magn. Mater.
2004, 272-276, 924.
(23) Isett, L. C.; Rosso, D. M.; Bottger, G. L. Phys. ReV. B 1980, 22,
4739.