This document is the Accepted Manuscript version of a Published Work that appeared in
final form in Journal of the American Chemical Society, copyright © American Chemical
Society after peer review and technical editing by the publisher. To access the final edited
and published work see https://pubs.acs.org/doi/pdf/10.1021/jacs.8b08032

Article

Conducting Anilate-Based Mixed-Valence Fe(II)Fe(III) Coordination
Polymer: Small-Polaron Hopping Model for Oxalate-Type Fe(II)Fe(III)
2D Networks
Suchithra Ashoka Sahadevan, Alexandre Abherve,́ Noemi Monni, Cristina Saenz de Pipaon
José Ramoń Galan-Mascaros, Joaõ
C. Waerenborgh, Bruno J. C. Vieira, Pascale Auban-Senzier,
Sebastien Pillet, El-Eulmi Bendeif, Pere Alemany, Enric Canadell, Maria Laura Mercuri,
and Narcis Avarvari
Laboratoire MOLTECH-Anjou UMR 6200, UFR Sciences, CNRS, Université d’Angers, Bat. K, 2 Bd. Lavoisier, 49045 Angers,
France
Dipartimento di Scienze Chimiche e Geologiche, Università degli Studi di Cagliari, I-09042 Monserrato (Cagliari), Italy
Institute of Chemical Research of Catalonia, The Barcelona Institute of Science and Technology (BIST), Avenida Països Catalans
16, 43007 Tarragona, Spain
ICREA, Passeig Lluís Companys 23, 08010 Barcelona, Spain
Centro de Ciencias e Tecnologias Nucleares, Instituto Superior Tecnico, Universidade de Lisboa, 2695-066 Bobadela LRS,
Portugal
Laboratoire de Physique des Solides, UMR 8502, Bat. 510, CNRS-Université Paris-Sud, 91405 Orsay, France
Université de Lorraine, CNRS, CRM2, F-54000 Nancy, France
Departament de Ciencia de Materials i Química Física and Institut de Química Teorica i Computacional (IQTCUB), Universitat
̀
de Barcelona, Martí i Franques 1, 08028 Barcelona, Spain
Institut de Ciencia de Materials de Barcelona (CSIC), Campus de la UAB, E-08193 Bellaterra, Spain

* Supporting Information
ABSTRACT: The mixed-valence FeIIFeIII 2D coordination polymer formulated as [TAG][FeIIFeIII(ClCNAn)3]·(solvate) 1 (TAG =
tris(amino)-guanidinium, ClCNAn2− = chlorocyanoanilate dianionic ligand) crystallized in the polar trigonal space group P3. In
the solid-state structure, determined both at 150 and at 10 K,
anionic 2D honeycomb layers [FeIIFeIII(ClCNAn)3]− establish in
the ab plane, with an intralayer metal−metal distance of 7.860 Å,
alternating with cationic layers of TAG. The similar Fe−O distances
suggest electron delocalization and an average oxidation state of
+2.5 for each Fe center. The cation imposes its C3 symmetry to the
structure and engages in intermolecular N−H···Cl hydrogen
bonding with the ligand. Magnetic susceptibility characterization
indicates magnetic ordering below 4 K and the presence of a
hysteresis loop at 2 K with a coercive field of 60 Oe. Mössbauer measurements are in agreement with the existence of Fe(+2.5)
ions at RT and statistic charge localization at 10 K. The compound shows semiconducting behavior with the in-plane
conductivity of 2 × 10−3 S/cm, 3 orders of magnitude higher than the perpendicular one. A small-polaron hopping model has
been applied to a series of oxalate-type FeIIFeIII 2D coordination polymers, providing a clear explanation on the much higher
conductivity of the anilate-based systems than the oxalate ones.

■

INTRODUCTION

Molecular materials combining conducting (π-type, delocalized) and magnetic (d-type, localized) electrons have attracted
major interest in molecular science since they can exhibit the
coexistence of two distinct physical properties, furnished by the
two independent networks, or novel and improved properties
when they interact.1 In this context, heterobimetallic oxalatebridged compounds have been thoroughly used as a magnetic

lattice of multifunctional magnetic materials.2 They are formed
by anionic networks [MIIMIII(ox)3]− (ox = oxalate) with
magnetic ions linked through bis-bidentate bridging oxalate
ligands. The second property is provided by the chargecompensating cation, thus combining the long-range magnetic

Article

ordering of the oxalate network with paramagnetism, 3
photochromism,4 electrical conductivity,5 proton conductivity,6 ferroelectricity,7 chirality,8 or single-molecule magnet
behavior.9 In recent years, coordination polymers based on the
3,5-disubstituted-2,6-dihydroxy-1,4-benzoquinone (H2dhbq,
see Scheme 1), also called anilate ligand in its dianionic
Scheme 1. 2,6-Dihydroxy-1,4-benzoquinone (H2dhbq) and
Potassium Chlorocyananilate (KHClCNAn)

form (dhbq2− or X2An2−), have been thoroughly developed.10
They usually present 2D-layered11 or 3D-extended networks12
with larger cavities than those obtained in oxalate-based
coordination polymers due to the larger size of the bridging
ligand. More recently, Miyasaka et al. have been able to
increase the magnetic ordering temperature in the previously
reported ferrimagnet (NBu4)[MnIICrIII(Cl2An)3] (Cl2An2− =
chloranilate) from 10 to 40 K.13 Taking advantage of the
porosity of the material, they inserted Li+ ions into the pores of
the 2D network in order to generate a radical Cl2An•3− (S = 1/
2) and produce a new exchange interaction between the radical
ligand and the metal centers, thus giving further proof of the
potential of the anilate ligand to enhance the magnetic
coupling in the extended network. In this context, coexistence
of electrical conductivity and magnetic ordering in FeIII anilatebased coordination polymers has been reported simultaneously
by two research groups in 2015.14 In both cases, the
conducting properties were attributed to the presence of
radical anilate bridging ligand species. The role played by the
FeII/FeIII and L2−/L3− mixed valency was further discussed by
Robson et al. with a more recent result based on an
interpenetrated 3D network of formula (NBu4)[Fe2(F2An)3]
(F2An2− = fluoranilate).12d Another advantage of the anilate
ligand is that it can be substituted on the 3 and 6 positions of
the aromatic ring by a large variety of substituents.15 Recently,
the asymmetric chlorocyananilate (ClCNAn2−, see Scheme
1)16 ligand has been combined to the redox-active molecule
bis(ethylenedithio)-tetrathiafulvalene (BEDT-TTF) to prepare
an organic semiconductor with formula [HClCNAn]2[BEDTTTF]17 and to lanthanide ions in a series of coordination
polymers of general formula [Ln 2(ClCNAn) 3 (DMF) 6]·
(CH2Cl2)x (Ln = Yb, x = 0; Ln = Nd or Er, x = 1) which
present NIR emission properties as bulk and nanosheets.18 In
this work, we investigated this nonsymmetric anilate ligand to
prepare the mixed-valence coordination polymer of formula
[TAG][FeIIFeIII(ClCNAn)3]·(solvate) (1), where TAG is the
C3-symmetric tris(amino)-guanidinium cation, never used so
far in such coordination polymers. Thorough structural
characterization and study of the magnetic and conducting
properties of this crystalline material are described. Since
mixed-valence FeIIFeIII oxalate-based coordination polymers
were previously reported to present a very poor conductivity,

we also investigated the origin and mechanism of the transport
properties in mixed-valence FeIIFeIII networks based on
oxalate-related bridging ligands, i.e., oxalate, squarate, dhbq2−,
and ClCNAn2−.

■

EXPERIMENTAL SECTION
19

Tris(amino)-guanidinium chloride (TAGCl) and potassium chlorocyananilate (KHClCNAn)16 were prepared according to the
reported procedures. All other chemicals were commercially available
and used as received without further purification.
Synthesis of [TAG][FeII FeIII (ClCNAn) 3]·(solvate) (1). An
aqueous solution (8 mL) of KHClCNAn (64 mg, 0.27 mmol) was
first placed in the bottom of a test tube, then it was carefully layered a
solution of TAGCl (42 mg, 0.3 mmol) in a mixture of water (2 mL)
and THF (2 mL) in the middle, and then a solution of Fe(ClO4)2·
xH2O (51 mg, 0.2 mmol) in acetone (3 mL) was added on the top.
After 1 week black hexagonal crystals suitable for XRD measurement
appeared at the interface. Anal. Calcd for C22H67N9O41Cl3Fe2: C,
19.84; H, 5.07; N, 9.47. Found: C, 19.32; H, 5.23; N, 9.22. FT-IR (ν
max/cm−1): 2221 (νC≡N), 1631, 1492 (νC−O + νC−C), 869 (δC−Cl +
νC−O).
Structural Characterization. Data collection was performed at
150 K on an Agilent Supernova diffractometer with Cu Kα (λ =
1.54184 Å). A single crystal of 1 was mounted on a glass fiber loop
using a viscous hydrocarbon oil to coat the crystal and then
transferred directly to the cold nitrogen stream for data collection.
The structure was solved by direct methods with the SIR97 program
and refined against all F2 values with the SHELXL-97 program using
the WinGX graphical user interface. All non-hydrogen atoms were
refined anisotropically except for the C and N atoms from the cationic
entity CN6H9+, and hydrogen atoms were placed in calculated
positions and refined isotropically with a riding model. The program
SQUEEZE from PLATON was used to calculate the potential
solvent-accessible void volume and the nature of the disordered
solvent molecules. It has indicated a total void space of 885 Å3 and
293 electrons/cell. This corresponds to 29 molecules of water that
have been inserted in the formula of the compound. A summary of
crystallographic data and refinement results are listed in Table 1.
Data collection was performed at 10 K on a SuperNova Microfocus
diffractometer equipped with a two-dimensional ATLAS detector
using Mo Kα radiation (λ = 0.71073 Å) and a Helijet He open-flow
cryosystem. The unit-cell determination and data reduction were
performed using the CrysAlisPRO program suite (Rigaku Oxford
Diffraction, 2017) on the full data set. The data have been indexed
using the trigonal setting with cell parameters a = b = 13.5493(16) Å
and c = 9.484(3) Å. In addition, the diffraction pattern showed the
presence of very weak (1/3 1/3 1/3) superstructure reflections, which
were not taken into account in the structure determination. The
corresponding crystal structure was refined on F2 by weighted full
matrix least-squares methods using the SHELXL program (Sheldrick,
2008). All non-hydrogen atoms were refined anisotropically except for
the C and N atoms from the cationic entity CN6H9+, and hydrogen
atoms were placed in calculated positions and refined isotropically
with a riding model. The program SQUEEZE from PLATON was
used to calculate the potential solvent-accessible void volume and the
nature of the disordered solvent molecules.
CCDC-1858728 (150 K) and 1858527 (10 K) contain the
supplementary crystallographic data for this paper.
Raman Measurements. The Raman spectrum of 1 was carried
out at room temperature on single crystals by using a micro Raman
spectrometer (Horiba Labram 300) equipped with a He−Ne laser (λ
= 632.81 nm) in the 80−2000 cm −1 range with a 20 LWD objective
(<0.25 mW/μm 2 on the crystal). A 180° reflective geometry was
adopted. The samples were mounted on a glass microscope slide, and
the scattering peaks were calibrated against a Si standard (ν = 520.7
cm−1). A typical spectrum was collected with a 300 s time constant at
<1 cm−1 resolution and averaged over three scans. The KHClCNAn
FT-Raman spectrum was recorded at room temperature on a capillary
tube by using a Bruker RFS/100 FT-Raman spectrometer equipped

Article

Table 1. Crystallographic Data for Compound 1 at 150 and
10 K
150 K
empirical formula
fw
cryst color
cryst size (mm3)
wavelength (Å)
cryst syst, Z
space group
a (Å)
b (Å)
c (Å)
α (deg)
β (deg)
γ (deg)
V (Å3)
ρcalcd (g·cm−3)
μ(Cu Kα) (mm−1)
θ range (deg)
data collected
data unique
data obsd
R(int)
no. of parameters/restraints
a
R1(F), I > 2σ(I)
2 b
wR2(F ), all data
S(F2),c all data

10 K

C22H67N9O41Cl3Fe2
1331.90
black
0.20 × 0.20 × 0.05
1.54184
trigonal, 1
P3
13.616(2)
13.616(2)
9.430(4)
90
90
120
1514.1(8)
1.461
5.994

C22H67N9O41Cl3Fe2
1331.90
black
0.22 × 0.19 × 0.06
0.71073
trigonal, 1
P3
13.5493(16)
13.5493(16)
9.484(3)
90
90
120
1507.8(6)
1.467
0.653

3.75−73.74
3465
2610
1360
0.0459
140/4
0.0395
0.1258
0.891

2.15−26.36
8947
3976
1612
0.0604
125/2
0.0715
0.2391
1.095

R1(F) = Σ||F 0| − |F C||/Σ|F0|. bwR2(F2) = [Σw(F 02 − F C2)2/
ΣwF0 4]1/2. cS(F2) = [Σw( F02 − FC2 )2 /(n + r − p)] 1/2.
a

with a Nd:YAG laser (λ = 1064 nm) in a backscattering geometry. No
sample decomposition was observed during the experiments. The
choice of the FT-Raman spectrometer was required because of the
fluorescence of the ligand under He−Ne laser irradiation.
Magnetic Measurements. Magnetic measurements were carried
out on polycrystalline samples with a Quantum Design MPMS-XL-7T
SQUID magnetometer (Quantum Design, Inc., San Diego, CA,
USA). Magnetic measurements (dc) were carried out under an
applied field of 1000 Oe. Zero-field-cooled/field-cooled/remanent
magnetization (ZFC/FC/RM) were collected under an applied field
of 25 Oe. Alternating current susceptibility measurements were
carried out with an alternating magnetic field of 3.95 Oe in the 1−
1500 Hz frequency range.
Mössbauer Spectroscopy. Mössbauer spectra were collected in
transmission mode using a conventional constant-acceleration
spectrometer and a 25 mCi 57Co source in a Rh matrix. The velocity
scale was calibrated using α-Fe foil. Isomer shifts, IS, are given relative
to this standard at room temperature. The low-temperature spectrum
was collected in a bath cryostat with the sample in He exchange gas.
The absorber was obtained by gently packing the powdered sample
into a perspex holder. The spectra were fitted to Lorentzian lines
using a nonlinear least-squares method.
Single-Crystal Transport Measurements. Electrical transport
measurements were performed on hexagonal-shaped single crystals.
Gold wires (17 μm diameter) were glued with silver paste either on
two edges or on both faces of the crystals (for conductivity
measurements parallel and perpendicular to the 2D planes,
respectively). Two-probe dc measurements were performed applying
a constant voltage in the range 0−5 V and measuring the current
using a Keithley 6487 Picoammeter/Voltage Source. Low temperature
was provided by a homemade cryostat equipped with a 4 K pulse
tube.
Theoretical Calculations. Density functional theory (DFT)based calculations were carried out adopting the hybrid TPSSh
functional,20 which has been shown to give good high-spin−low-spin

relative energies for spin-crossover complexes involving iron21 and the
standard double-ζ + polarization basis set 6-31G(d).22 Geometries
were optimized forcing a D3 symmetry and a high-spin configuration
using the Gaussian 09 code.23

■

RESULTS AND DISCUSSION

Synthesis. The synthesis of the mixed-valence compound 1
differs from that of the heterobimetallic oxalate and anilatebased coordination polymers. In such cases, the tetrabutylammonium salt of (tris-oxalato)metal(III) or (tris-anilato)metal(III) complex was first prepared and isolated. The
precursor thus obtained was then reacted with the second
metal salt by diffusion techniques to grow the bimetallic
extended network. Here we have slowly diffused a solution of
FeIIClO4·xH2O and a solution of tris(amino)-guanidinium
chloride (TAGCl) into a solution of KHClCNAn. Due to the
partial oxidation of the FeII ions under aerobic conditions,
black hexagonal crystals of the FeIIFeIII compound 1 were
obtained after 1 week.
Crystal Structure. Compound 1 crystallizes in the trigonal
polar space group P3. The structure is formed by anionic 2D
layers of formula [FeIIFeIII(ClCNAn)3]− in the ab plane,
alternating with cationic layers of TAG. The anionic layer
presents the well-known honeycomb structure, which is similar
to other extended oxalate24 and anilate-based 2D networks.11a−e It consists of a hexagonal layer with FeII and FeIII ions
linked through the anionic bis-bidentate ClCNAn2− ligands
(Figures 1 and S1 and S2). As usual for this type of 2D
networks, the two crystallographically independent metal
centers present alternated chirality (Δ configuration for Fe1
and Λ configuration for Fe2 in the crystal used to solve the
structure). The intralayer metal−metal distance is 7.860 Å.
The average Fe−O distances are very similar between both
metal centers (2.037(12) Å for Fe1 and 2.047(13) Å for Fe2,
see Table 2), which may indicate an electron delocalization
and an average oxidation state of +2.5 for each Fe center (vide
infra). The cationic layer is formed by one crystallographically
independent TAG cation and water molecules. The TAG
cation has an occupancy of 1/3, which is one-half of the Fe
atoms, and is located only on one-half of the vertices of the
hexagons. The structure of the cation is planar, with the C8−
N2 and N2−N3 distances (1.394(15) and 1.558(17) Å
respectively, see Table S1) in agreement with distances
reported in the literature for other TAG-based compounds.25
Anionic and cationic layers present intermolecular H-bonding
interactions between the terminal amino groups of the TAG
cation and the chloro substituents of the anilate ligands (Figure
1). The distance between two anionic layers corresponds to
the value of the c parameter (9.430(4) Å). The cationic and
anionic 2D layers are eclipsed, leading to hexagonal channels
along the c axis which are filled only by solvent molecules.
When compared to the previously reported anilate-based
layered coordination polymers,11a−h here the use of the smaller
cation results in a drastic increase of the void space inside the
hexagonal channels (885 Å3). This represents 58% of the total
volume, thus increasing the porosity of the 2D material (Figure
S3). As a consequence, the compound shows a fast release of
the solvent molecules after filtration, and the nature of these
solvent molecules could not be attributed without ambiguity.
In order to reach a good reliability factor and since 293
electrons per hexagonal cavity were determined by the
SQUEEZE program, 29 molecules of water have been
integrated in the empirical formula (see Table 1). However,

Article

Figure 1. Structure of 1 in the ac plane (top left) and in the ab plane (top right). H-bonding intermolecular interactions (dashed lines) between the
cationic and the anionic layers (bottom). Color code: C, black; H, cyan; O, red; N, blue; Cl, yellow; Fe, green.

Table 2. Selected Fe−O Bond Distances (Å) in Compound
1
Fe−O distances
Fe1−O1
Fe1−O2
Fe2−O3
Fe2−O4

150 K (Å)
2.000(6)
2.086(6)
2.073(7)
2.018(6)

10 K (Å)
1.937(7)
2.103(7)
2.117(7)
1.974(7)

TGA analysis could not confirm the precise nature of the
solvent molecules (Figure S4). Since a mixture of three
solvents has been used during the synthesis (water, THF, and
acetone), the formula of the compound should be defined as
[TAG][FeIIFeIII(ClCNAn)3]·(solvate).
By comparison with the 150 K crystal structure, the overall
honeycomb 2D structural architecture in the trigonal P3 space
group is retained at 10 K. Accordingly, the asymmetric unit still
contains two symmetrically independent iron sites (Fe1 and
Fe2). The unit cell parameters and unit cell volumes are
slightly lower at 10 K, owing to usual thermal contraction
effects. The intralayer metal−metal distance (7.8243(7) Å)
and average Fe−O distances (2.020(14) Å for Fe1 and
2.043(14) Å for Fe2) are not significantly different from the
150 K values. Accordingly, the 10 K crystal structure does not
evidence a specific ordering of the FeII and FeIII ions on the
symmetrically independent Fe1 and Fe2 sites. We can
therefore consider that in this description the FeII and FeIII
ions are spatially distributed and disordered over the two sites.
As mentioned in the Experimental Section, very weak
superstructure reflections were detected on the diffraction
pattern, which indicates that the exact structural ordering may

be more complex than this description in the P3 space group.
However, the quality of the X-ray diffraction data, and
especially the weakness of the superstructure reflections, does
not allow going further. The current description leads to two
FeII and two FeIII different local environments in the crystal,
which is consistent with the results from Mössbauer spectroscopy (vide infra).
Raman Spectroscopy. Raman spectra are valuable probes
to investigate the oxidation state of coordinated benzoquinone
derivatives.26 Therefore, in order to confirm the oxidation state
of the bridging ligand a Raman study at room temperature was
performed and a comparison between the Raman spectra of 1
and the free KHClCNAn ligand is reported in Figure 2.
The strong and broad band centered at ca. 1574 cm −1 can be
assigned to a ν(C C) + ν(C O) combination band, and the
significant observed downshift from 1627 cm−1 for the free
ligand can be attributed to a weakened double-bond character
of these terminal groups because of the coordination with the
metal ion; the weak band centered at ca. 1675 cm−1 present in
the free ligand spectrum can be assigned to the ν(C O)
vibration mode for the uncoordinated C O groups of the free
ligand and in fact is not observed in 1. The two bands observed
in the 1400−1250 cm−1 region are assigned to the ν(C−C) +
ν(C−O) combination band and ν(C−C) vibration, respectively, and according to Harris et al.14c confirm the assignment
of ligand oxidation state as dianionic ClCNAn2−, supporting
structural findings. The observed band at 592 cm−1 can be
assigned to a ν(Fe−O) + ν(C−C) combination stretching
mode, as already found in previously reported dianionic
anilate-based honeycomb-like networks.14a No bands can be
unambiguously assigned to FeII−O and FeIII−O vibrational

Article

Figure 2. Comparison of Raman spectra of 1 and the KHClCNAn
ligand, performed at room temperature with a He−Ne laser (λ =
632,81 nm) and a Nd:YAG (λ = 1064 nm), respectively.

modes, thus supporting extensive electron delocalization
between the Fe centers in 1 at room temperature, as clearly
shown by Mössbauer spectra (vide infra).
Magnetic Properties. The magnetic properties were
measured on a polycrystalline sample of 1. The product of
the molar magnetic susceptibility times the temperature (χmT)
presents a value of 9.7 emu·K·mol −1 at 300 K, which
corresponds to the expected spin-only value (7.38 emu·K·
mol−1) with a g = 2.2 (Figure 3). When the temperature is

Figure 3. Temperature dependence of the product of the molar
magnetic susceptibility times the temperature (χmT) of 1 under an
applied field of 1000 Oe.

lowered, χmT slightly decreases, suggesting weak antiferromagnetic interactions between paramagnetic centers through
the anilate bridges. Below 50 K, χmT increases until reaching a
value of ∼27 emu·K·mol−1 at 7 K, followed by a sharp decrease
at lower temperatures. This suggests a magnetic ordering,
which was confirmed by zero-field-cooled/field-cooled (ZFC/
FC) and remnant magnetization measurements under a very
low magnetic field (Figure S5). The ZFC and FC plots diverge
below 3.5 K, indicating the appearance of an irreversibility or
memory effect. The remnant magnetization becomes nonnegligible at the same temperature, confirming the existence of
spontaneous magnetization below this temperature.

The magnetic ordering was also confirmed by susceptibility
measurements under alternating magnetic field (ac susceptibility, Figure 4a). This shows the appearance of an out-ofphase signal and an ordering temperature Tc slightly dependent
on the frequency, which can be observed for both superparamagnets and spin glasses.27 The fitting of this frequencydependent behavior to a simple Arrhenius model (Figure 4b)
yields parameters with no physical meaning, including a τ0 =
10−12, very different to what is found in superparamagnets (or
single-molecule magnets, with values of τ0 between 10−8 and
10−10 s).28 This τ0 value falls within the range reported for
magnetic spin-glass systems (10−12−10−14 s)29 and is then
consistent with a glassy magnet behavior as observed in many
2D magnetic materials.30 Moreover, the Mydosh parameter φ,
calculated from the ac data,31 has a value of 0.08, in good
agreement with the expected values for a noncanonical spin
glass.14a,32
Isothermal magnetization measurements at low temperatures show a fast increase of the magnetization at low fields
that becomes more gradual at higher fields (Figure 5). The
sharp increase at low fields (H < 1000 Oe) also supports the
appearance of spontaneous magnetization due to strong
interactions between metal centers. The magnetization
saturates at higher fields reaching ∼5 μB, far from the expected
9 μB for parallel alignment of spin carriers. This confirms the
ferrimagnetic nature of the spontaneous magnetization that
stabilizes a ground state with an intermediate spin, characteristic of a glassy ferrimagnet as suggested by the initial decrease
in the χmT at high temperatures. An additional proof of the
magnetic ordering is the presence of a hysteresis loop at 2 K
with a coercive field of 60 Oe.
Mössbauer spectroscopy has been used to confirm the
oxidation state of the Fe metal centers (Figure 6). At 10 K
three broad absorption peaks are observed. They may be
interpreted by two quadrupole doublets. However, due to the
large width of the absorption peaks a significantly better fit is
obtained with four quadrupole doublets. The estimated isomer
shift, IS, and quadrupole splitting, QS, are consistent with the
presence of high-spin Fe3+ and high-spin Fe2+ in octahedral
coordination by anionic oxygen atoms.33 The estimated
relative areas indicate that approximately one-half of the Fe
cations are in the +3 state and the other half in the +2 state.
The two doublets observed for each oxidation state are
consistent with the occupation of Fe(1) and Fe(2) crystallographic sites by both Fe2+ and Fe3+. The room-temperature
spectrum shows only one asymmetric doublet with IS and QS
consistent with an average oxidation state of +2.5.34 The
temperature dependence of the Mössbauer spectra of the
anilate coordination polymer is similar to the behavior
observed for other mixed-valence iron compounds, namely,
molecular complexes.35 The intermediate isomer shift at room
temperature corresponds to a charge-delocalized state on the
Mössbauer spectroscopy time scale of ∼10−7 s, i.e., a chargetransfer frequency ≥ 108 s−1. As the temperatures decreases,
the frequency of charge delocalization gradually decreases, and
at 10 K the Fe2+ and Fe3+ states are localized when compared
to the Mössbauer effect time window (i.e., the lifetime of the
Fe2+ and Fe3+ states becomes longer than 10−7 s).
Transport Properties. Since a mixed-valence state and
electron delocalization have been evidenced in 1, we could
expect this coordination polymer to present transport properties; therefore, electrical conductivity measurements have been
carried out on single crystals. The hexagonal shape of the

Article

Figure 4. (a) Temperature dependence of the in-phase (χ′) and out-of-phase (χ′′) ac susceptibility of 1, and (b) Arrhenius plot for the frequency
dependence of the position of the peak in χ′′ vs 1/T.

Figure 5. (a) Isothermal magnetization of 1 at different temperatures and (b) magnetic hysteresis loop at 2K.

crystals allows determining the direction of the {001} plane
and then to measure the conductivity parallel (σ∥) and
perpendicular (σ⊥) to the 2D layers, corresponding to the ab
plane. The temperature dependence of the resistivity indicates
that 1 is a semiconductor (Figure 7). The parallel roomtemperature conductivity value σ∥ is about 2 × 10−3 S/cm,
almost 3 orders of magnitude higher than the perpendicular
room-temperature conductivity σ⊥ (7 × 10−6 S/cm).
Relationship between the Electrical Conductivity and
the Nature of the Bridging Ligand. The fairly good
conductivity of compound 1 as well as of the few recently
reported anilato-based FeIIFeIII coordination polymers11g,14b is
in sharp contrast with the low conductivity found for the
oxalate-based ones.36 To gain insight into the origin of the
good conductivity in our anilate-based coordination polymer
and to point out the crucial role of the bridging ligand, we have
undertaken a theoretical study on the electron transfer in 2D
FeIIFeIII networks based on bis(bidentate) oxalate-type ligands
in which we consider the conductivity dominated by thermally
activated small-polaron hopping. We are interested in a simple,

pragmatic approach, highlighting the role of the bridging
ligand, and in this comparative work, we decided to focus on
four ligands: oxalate, squarate (C4O42−, dianion of 3,4dihydroxycyclobut-3-ene-1,2-dione), dhbq2−, and ClCNAn2−,
with different electron delocalization capabilities.
The basic reasoning behind the small-polaron hopping
approach to the electron transfer process in either discrete or
extended mixed-valence systems37,38 is qualitatively illustrated
in Figure 8. Consider a system formed by two separated highspin FeIIL6 (t2g 4eg2) and high-spin FeIII L6 (t2g 3eg2) complexes in
close proximity. Although the electron transfer between them
is between levels of the Fe t2g orbital set, which are formally
nonbonding, such transfer leads, in general, to an increase/
decrease of the M−L distances in the FeII/FeIIIL6 units due to
the expansion/contraction of the electron cloud. As a
consequence of the fact that electrons move much faster
than nuclei, the much faster electron transfer occurs in such a
way that the geometry cannot change during the process and
the system cannot exchange thermal energy with the
surroundings. In other words, before the electron transfer

Article

Figure 6. (a) Mössbauer spectra of 1 taken at 295 and 10 K, and (b) estimated parameters from the spectra. Lines over the experimental points are
the calculated functions. On the spectrum taken at 10 K this function is the sum of four quadrupole doublets shown slightly shifted for clarity. IS,
isomer shift relative to metallic α-Fe at 295 K; QS, quadrupole splitting; I, relative areas. Estimated errors are <0.02 mm/s for IS and QS and <2%
for I.

Figure 7. (a) Electrical resistivity ρ∥ plotted as log ρ∥ versus the inverse temperature, measured with a 4 V voltage applied in the ab plane. Red line
is the fit to the data with the law ρ = ρ0 exp(Ea/T) giving the activation energy Ea. (b) Room-temperature resistivity (ρ) and conductivity (σ) values
measured along (∥) and perpendicular (⊥) to the 2D layers.

can actually take place a FeIII species with the FeII geometry
(and viceversa) must be created. This rearrangement is
however energetically unfavorable, so that the electron transfer
will only occur when as a result of some vibrational process the
two Fe centers reach equal coordination geometries (see
Figure 8a). Thus, to understand the differences in thermally
activated conductivity one must focus on the evaluation of the
energetic cost of such “equalization” of the two sites.
A simple yet usually very insightful analysis relies on the
assumption that the structural distortion around each center
may be described by a simple harmonic oscillator Ei = 1/2 ki(di
− d0)2, where ki is the force constant, di the Fe−L distance, and
d0 its equilibrium value. Within this approximation the energy
of the whole system is Etot = kIII(dA − dIII)2 + kII(dB − dII)2,
where for the sake of simplicity the 1/2 factors have been
included in the force constants. Here let us assume that
complex A is in the FeIII state and complex B in the FeII state,

but there is a totally equivalent expression interchanging A and
B. If we plot the two energy surfaces Etot(dA,dB) we see that
they cross along the line dA = dB (Figure 8b). The initial and
final configurations (respectively, top and bottom configurations in Figure 8a) correspond to minima in the lower
surface and the structures for which dA = dB to the crossing
seam. The situation is more conveniently analyzed by using a
contour plot of the bottom surface (Figure 8c), obtained by
joining the two bottom halves of the two intersecting surfaces.
The two minima correspond to the configurations FeIII A II B
···Fe
(zone noted as F in Figure 8c) and FeII A III B (noted as I).
···Fe
The point on the dA = dB line for which the total energy is
minimal (M) is found by taking the derivative of the energy
function with dA = dB = dM and equating it to zero. It is found
that dM = (kIIdII + kIIIdIII)/(kII + kIII) and the corresponding
energy EM = (kIIkIII/(kII + kIII))(dII − dIII)2. In other words, the
minimal energy required to make the coordination environ-

Article

Figure 8. (a) Schematic representation of the three consecutive steps
during the electron transfer with two neighboring pseudooctahedral
Fe(chelate)3 centers shown along the C3 axis. (b) 3D plot of the
energy surface Etot as a function of dA and dB. (c) Contour plot of the
bottom surface from the Etot plot displayed in b.

ments on both Fe centers equal, that is, the height of the
barrier for thermally activated electron transfer, depends
basically on the difference squared between the radii of the
FeIII and FeII coordination environments in their equilibrium
geometries.
The model can be more simply depicted by using a cut
through the two surfaces in which we plot the energy along the

dotted path joining the two minima through the M point,
which is taken as the origin (Figure 9). This gives two
parabolas corresponding to the energy of the whole system
with either an FeIIFeIII or an FeIIIFeII configuration and
crossing at the M point. Note that when the two complexes are
not totally isolated from their surroundings, the energy
necessary to distort the complexes has an additional
contribution from the environment. This is particularly
important in the case of 2D lattices where the distortion of
one site contributes to the distortion of all its neighboring sites.
This contribution is usually included considering a generic
parameter λ, which is the vertical ionization energy from one
minimum to the other curve or in other words the energy
necessary to transfer the electron from A to B without
considering a previous structural equalization of both centers.
The barrier for thermal electron transfer in the absence of
interaction (i.e., the energy difference between the minima and
the crossing point of the two curves) is then simply λ/4. In the
real case there is always some degree of electronic interaction
between the two sites and a gap, 2VAB, is opened at point M,
and the barrier for thermal electron transfer is consequently
lowered to λ/4 − VAB. Unfortunately, there is no simple way to
evaluate VAB without having recourse to long and costly
computations, but since we are here interested only in looking
for trends when the bridging ligand is changed, it seems safe to
consider that λ/4 will be the leading term in the energy barrier,
so that we can make our comparisons neglecting the effects of
the nature of the ligands on VAB.
It is now easy to relate the main parameters (mainly
structural) of this simple model to the transport measurements.
The diagram in Figure 9 is completely general and applicable
to any electron transfer process. In an extended system, when
the dimensions of the zone where the necessary atomic
rearrangement controlling the electron transfer occurs are of
the order of the coordination sphere of a single site as in the
present case, one talks about a system with small polarons.38
The polaron energy, WP, is the energy gained when the system
relaxes after addition of one electron. In this small-polaron
scenario the conductivity is dominated by thermally activated
electron hopping with the mobility given by the equation μ =

Figure 9. (a) Energy plot for a FeIIFeIII to FeIIIFeII electron transfer process. (b) Calculated parameters for the four studied bridging ligands.
Distances are expressed in Angstroms, energies in kcal.mol−1, and k and F in kcal.mol −1·Å−2.

Article

μ0e−WH/k, where W H is the electron hopping barrier, i.e., the
energy cost to reach the geometry under which the electron
transfer is possible. As far as it is assumed that the variation of
energy is a quadratic function of the structural parameters, the
energy cost per site to reach the “equalization” geometry is 1/4
WP, and taking into account that there are two sites involved in
the transfer, it follows that W H = 1/2 WP. Since according to
the Franck−Condon principle the energy to optically excite
one electron from one to the other site (i.e., λ in Figure 9) is
twice the polaron energy it follows that W H = 1/2 W P = λ/4.
Therefore, W H = EM = F(dII − dIII)2, with F = kIIkIII/(kII + kIII)
and dM = (kIIdII + kIIIdIII)/(kII + kIII). In that way it is possible
to correlate the transport (W H) and structural (kII, kIII, dII, and
dIII) features for a series of compounds.
How are these parameters tuned by the nature of the
bridging ligand? The values of kII, kIII, dII, and dIII can be
evaluated from density functional calculations (DFT). We
carried out structural optimizations of FeIIL3 and FeIIIL3
complexes with the four ligands oxalate, squarate, dhbq2−,
and ClCNAn2− using the TPSSh functional, which is known to
give good high-spin−low-spin relative energies for spincrossover complexes involving iron.21 The geometries were
optimized forcing a D3 symmetry and a high-spin configuration. In that way we obtained the dM−L parameter which is
the distance between the Fe atom and the midpoint to the
closest C−C bonds (between the C atoms bonded to the
coordinating O atoms) which corresponds to dII or dIII of the
above discussion. Once the optimal dM−L parameter for each
FeII and each FeIII complex was found, the values of kII or kIII
were obtained by reoptimizing the structure for fixed dM−L ± δ
values and fitting the energy to a second-order polynomial.
The results for the four ligands are reported in Figure 9b.
According to the data in Figure 9b W H = 0.214, 0.057, 0.044,
and 0.037 eV for oxalate, squarate, dhbq2−, and ClCNAn2−,
respectively. Despite the simplicity of the approach, these
numbers are very reasonable when compared with W H
estimations based on experimental data for FeIIFeIII solids
like magnetite (∼0.15 eV)38 and partially substituted ferrites
(∼0.11 eV)39 which are in between those calculated for oxalate
and the other three ligands. Thus, we believe that the simple
approach captures the essence of the electron transfer process,
although the activation energies are smaller than the
experimental ones. This is not unexpected since the Fe sites
are immersed in a quite rigid 2D network, which must lead to a
noticeable increase in the parameter λ and hence in the barrier
estimated here by λ/4. According to the values in Figure 9b
electrons should be much less mobile in oxalate networks than
in anilate ones, which is in good agreement with experimental
observations.36,11f,14b What is the reason for this finding? The F
values for the oxalate ligands are in between those for both the
dhbq2− and the ClCNAn2− ones, yet W H is five times larger. It
is clear that the factor determining the large difference is
(Δd)2, i.e., the square of the difference in the d values for Fe in
the two oxidation states. Because of the quadratic dependence,
this term has an overwhelming influence on the final value of
WH. Comparing the dII and dIII distances for the oxalate,
dhbq2−, and ClCNAn2− ligands in Figure 9b it is clear that
whereas the dIII values are fairly similar, dII for the oxalate
ligand is markedly larger than those for the dhbq2− and
ClCNAn2− ligands. Thus, the determining factor for the lower
conductivity of the oxalate systems lies in the difficulty to cope
with the extra electron of the high-spin FeII situation. The
occurrence of extensive delocalization through the benzene

ring of dhbq2− and ClCNAn2− acts as a buffer for the
electronic rearrangement needed by the presence of the extra
electron. Note that when the ligands are more similar, the F
factor depending on the force constants may become the key
factor. According to Figure 9b this is, for instance, the case
when comparing the dhbq2− and ClCNAn2− ligands. The
squarate ligand is also associated with a small W H value, not
much different from those of the dhbq2− and ClCNAn2−
ligands. However, the (Δd) 2 term is intermediate between
those of the oxalate and dhbq 2− and ClCNAn2− pair of ligands.
Clearly, in that case the F term plays an important role. Both
the geometrical constraints imposed by the four-member ring
as well as the delocalization in the central part of the ligand
influence the electron transfer tendency.
We thus conclude that the present approach provides a
simple, yet insightful, model to rationalize the transport results
for these FeIIFeIII mixed-valence coordination polymers and
that the polymers based on the dhbq2− and ClCNAn2− ligands
rank among the most effective ones in promoting the electron
delocalization through a small-polaron hopping mechanism.

■

CONCLUSIONS

The mixed-valence FeIIFeIII 2D coordination polymer [TAG][FeIIFeIII(ClCNAn)3] based on the asymmetric chlorocyanoanilate ligand and containing, for the first time in such
2D networks, the tris(amino)-guanidinium (TAG) cation has
been synthesized and crystallized in the polar noncentrosymmetric space group P3 thanks to the C3 symmetry of the cation
and its ability to engage in intermolecular hydrogen bonding
with the chlorine atoms of the ligand. Magnetic susceptibility
measurements in combination with Mössbauer spectroscopy
are indicative of a spin-glass behavior with magnetic ordering
below 4 K and the presence of intermediate Fe(+2.5)
oxidation state at RT and charge localization FeIIFeIII at 10
K, with a statistic occupational crystallographic site according
to the 10 K Mössbauer spectra, also confirmed by the X-ray
structure at 10 K. Single-crystal electron transport measurements in the 2D plane and perpendicular on it show
semiconducting behavior of the material with a rather high
RT value of 2 × 10−3 S/cm for the in-plane conductivity, much
higher than the one reported in the oxalate-based 2D FeIIFeIII
coordination polymers. In order to shed light on this difference
and on the electron transport mechanism on these fast
developing multifunctional families of 2D coordination
polymers, the small-polaron hopping approach to the electron
transfer process has been applied to a series of mixed-valence
FeIIFeIII oxalate-related coordination polymers containing as
bridging ligands oxalate2−, squarate2−, dhbq2−, and ClCNAn2−.
The results are clearly indicative of a much lower electron
hopping barrier in the anilate complexes than in oxalates.
These results open the way toward the use of the TAG
cation in such multifunctional binuclear/bimetallic transition
metal or lanthanide coordination polymers, possibly endowed
with multiferroic properties thanks to the crystallization in
polar space groups. Moreover, the mechanism of electron
transport in the mixed-valence FeIIFeIII coordination polymers
with bridging oxalate-type ligands has been disclosed, pointing
out the higher “elasticity” of the anilate network compared to
the oxalate one.

■

Article

ASSOCIATED CONTENT

* Supporting Information
The Supporting Information is available free of charge on the
ACS Publications website at DOI: 10.1021/jacs.8b08032.
Full experimental section including description of the
synthesis and characterization of all new materials and all
of the techniques employed in the research reported
here (PDF)
(CIF)
(CIF)

■

AUTHOR INFORMATION

Corresponding Authors

*p.alemany@ub.edu
*canadell@icmab.es
*mercuri@unica.it
*narcis.avarvari@univ-angers.fr
ORCID

́
José Ramon Galan-Mascaros: 0000-0001-7983-9762
Enric Canadell: 0000-0002-4663-5226
Maria Laura Mercuri: 0000-0002-4816-427X
Narcis Avarvari: 0000-0001-9970-4494
Author Contributions
§

S.A.S. and A.A. contributed equally to this work.

Notes

The authors declare no competing financial interest.

■

ACKNOWLEDGMENTS

■

REFERENCES

The work in France was supported by the CNRS, the
University of Angers, the Erasmus program (mobility grant to
N.M.), the RFI Regional project LUMOMAT (grant to A.A.,
project ASCO MMM), and the PIA project “Lorraine
Université d'Excellence” (reference ANR-15-IDEX-04-LUE).
This work was supported in Italy by the Fondazione di
Sardegna-Convenzione triennale tra la Fondazione di Sardegna
e gli Atenei Sardi, Regione Sardegna-L.R. 7/2007 annualità
2016-DGR 28/21 del 17.05.2015 “Innovative Molecular
Functional Materials for Environmental and Biomedical
Applications” and INSTM. Work in Spain was supported by
́
the Spanish Ministerio de Economia y Competitividad (Grants
FIS2012-37549-C05-05, FIS2015-64886-C5-4-P, CTQ201564579-C3-3-P, and CTQ2015-71287-R) and Generalitat de
Catalunya (2014SGR301, 2017SGR797, and XRQTC and the
CERCA program). E.C. acknowledges support of the Spanish
MINECO through the Severo Ochoa Centers of Excellence
Program under Grant SEV-2015-0496. J.C.W. and B.J.C.V.
acknowledge Fundacao para a Ciencia e a Tecnologia (FCT,
̧̃
Portugal) through the project UID/Multi/04349/2013.

(1) (a) Coronado, E.; Day, P. Chem. Rev. 2004, 104, 5419−5448.
(b) Enoki, T.; Miyazaki, A. Chem. Rev. 2004, 104, 5449−5478.
(c) Kobayashi, H.; Cui, H. B.; Kobayashi, A. Chem. Rev. 2004, 104,
5265−5288. (d) Kurmoo, M.; Graham, A. W.; Day, P.; Coles, S. J.;
Hursthouse, M. B.; Caulfield, J. L.; Singleton, J.; Pratt, F. L.; Hayes,
W.; Ducasse, L.; Guionneau, P. J. Am. Chem. Soc. 1995, 117, 12209−
12217. (e) Martin, L.; Turner, S. S.; Day, P.; Mabbs, F. E.; McInnes,
E. J. L. Chem. Commun. 1997, 15, 1367−1368. (f) Rashid, S.; Turner,
S. S.; Day, P.; Howard, J. A. K.; Guionneau, P.; McInnes, E. J. L.;
Mabbs, F. E.; Clark, R. J. H.; Firth, S.; Biggs, T. J. Mater. Chem. 2001,
11, 2095−2101. (g) Uji, S.; Shinagawa, H.; Terashima, T.; Yakabe, T.;
Terai, Y.; Tokumoto, M.; Kobayashi, A.; Tanaka, H.; Kobayashi, H.

Nature 2001, 410, 908−910. (h) Fujiwara, H.; Fujiwara, E.;
Nakazawa, Y.; Narymbetov, B. Zh.; Kato, K.; Kobayashi, H.;
Kobayashi, A.; Tokumoto, M.; Cassoux, P. J. Am. Chem. Soc. 2001,
123, 306−314. (i) Coronado, E.; Falvello, L. R.; Galan-Mascaros, J.
R.; Gimenez-Saiz, C.; Gomez-García, C. J.; Lauhkin, V. N.; PerezBenítez, A.; Rovira, C.; Veciana, J. Adv. Mater. 1997, 9, 984−987.
(j) Madalan, A. M.; Canadell, E.; Auban-Senzier, P.; Branzea, D.;
Avarvari, N.; Andruh, M. New J. Chem. 2008, 32, 333−339. (k) Pop,
F.; Allain, M.; Auban-Senzier, P.; Martínez-Lillo, J.; Lloret, F.; Julve,
M.; Canadell, E.; Avarvari, N. Eur. J. Inorg. Chem. 2014, 2014, 3855−
3862.
(2) Clemente-Leon, M.; Coronado, E.; Martí-Gastaldo, C.; Romero,
F. M. Chem. Soc. Rev. 2011, 40, 473−497.
(3) (a) Clemente-Leon, M.; Coronado, E.; Galan-Mascaros, J. R.;
Gomez-García, C. J. Chem. Commun. 1997, 1727−1728. (b) Coronado, E.; Galan-Mascaros, J. R.; Gomez-García, C. J.; Martínez-Agudo, J.
M. Adv. Mater. 1999, 11, 558−561. (c) Coronado, E.; GalanMascaros, J. R.; Gomez-García, C. J.; Ensling, J.; Gütlich, P. Chem. Eur. J. 2000, 6, 552−563.
(4) (a) Benard, S.; Yu, P.; Audiere, J. P.; Riviere, E.; Clement, R.;
Guilhem, J.; Tchertanov, L.; Nakatani, K. J. Am. Chem. Soc. 2000, 122,
9444−9454. (b) Aldoshin, S. M.; Sanina, N. A.; Minkin, V. I.;
Voloshin, N. A.; Ikorskii, V. N.; Ovcharenko, V. I.; Smirnov, V. A.;
Nagaeva, N. K. J. Mol. Struct. 2007, 826, 69−74.
(5) (a) Coronado, E.; Galan-Mascaros, J. R.; Gomez-García, C. J.;
Laukhin, V. Nature 2000, 408, 447−449. (b) Alberola, A.; Coronado,
E.; Galan-Mascaros, J. R.; Gimenez-Saiz, C.; Gomez-García, C. J. J.
Am. Chem. Soc. 2003, 125, 10774−10775. (c) Galan-Mascaros, J. R.;
Coronado, E.; Goddard, P. A.; Singleton, J.; Coldea, A. I.; Wallis, J.
D.; Coles, S. J.; Alberola, A. J. Am. Chem. Soc. 2010, 132, 9271−9273.
(d) Coronado, E.; Galan-Mascaros, J. R.; Gomez-García, C. J.;
Martínez-Ferrero, E.; Van Smaalen, S. Inorg. Chem. 2004, 43, 4808−
4810. (e) Zhang, B.; Zhang, Y.; Zhu, D. Chem. Commun. 2012, 48,
197−199.
(6) (a) Okawa, H.; Shigematsu, A.; Sadakiyo, M.; Miyagawa, T.;
Yoneda, K.; Ohba, M.; Kitagawa, H. J. Am. Chem. Soc. 2009, 131,
13516−13522. (b) Pardo, E.; Train, C.; Gontard, G.; Boubekeur, K.;
Fabelo, O.; Liu, H.; Dkhil, B.; Lloret, F.; Nakagawa, K.; Tokoro, H.;
Ohkoshi, S.-I.; Verdaguer, M. J. Am. Chem. Soc. 2011, 133, 15328−
15331. (c) Sadakiyo, M.; Okawa, H.; Shigematsu, A.; Ohba, M.;
Yamada, T.; Kitagawa, H. J. Am. Chem. Soc. 2012, 134, 5472−5475.
(d) Okawa, H.; Sadakiyo, M.; Yamada, T.; Maesato, M.; Ohba, M.;
Kitagawa, H. J. Am. Chem. Soc. 2013, 135, 2256−2262.
(7) (a) Endo, T.; Akutagawa, T.; Noro, S. I.; Nakamura, T. Dalton
Trans 2011, 40, 1491−1496. (b) Pardo, E.; Train, C.; Liu, H.;
Chamoreau, L.-M.; Dkhil, B.; Boubekeur, K.; Lloret, F.; Nakatani, K.;
Tokoro, H.; Ohkoshi, S.-i.; Verdaguer, M. Angew. Chem., Int. Ed.
2012, 51, 8356−8360.
(8) (a) Andres, R.; Gruselle, M.; Malezieux, B.; Verdaguer, M.;
Vaissermann, J. Inorg. Chem. 1999, 38, 4637−4646. (b) Andres, R.;
Brissard, M.; Gruselle, M.; Train, C.; Vaissermann, J.; Malezieux, B.;
Jamet, J. P.; Verdaguer, M. Inorg. Chem. 2001, 40, 4633−4640.
(c) Clemente-Leon, M.; Coronado, E.; Dias, J. C.; Soriano-Portillo,
A.; Willett, R. D. Inorg. Chem. 2008, 47, 6458−6463. (d) Train, C.;
Gheorghe, R.; Krstic, V.; Chamoreau, L. M.; Ovanesyan, N. S.;
Rikken, G. L. J. A.; Gruselle, M.; Verdaguer, M. Nat. Mater. 2008, 7,
729−734. (e) Train, C.; Nuida, T.; Gheorghe, R.; Gruselle, M.;
Ohkoshi, S. J. Am. Chem. Soc. 2009, 131, 16838−16843. (f) Gruselle,
M.; Li, Y.; Ovanesyan, N.; Makhaev, V.; Shilov, G.; Mushenok, F.;
Train, C.; Aldoshin, S. Chirality 2013, 25, 444−448.
(9) Clemente-Leon, M.; Coronado, E.; Gomez-García, C. J.; LopezJorda,́ M.; Camon, A.; Repolles, A.; Luis, F. Chem. - Eur. J. 2014, 20,
1669−1676.
(10) (a) Kitagawa, S.; Kawata, S. Coord. Chem. Rev. 2002, 224, 11−
34. (b) Mercuri, M. L.; Congiu, F.; Concas, G.; Sahadevan, S. A.
Magnetochemistry 2017, 3, 17.
(11) (a) Atzori, M.; Benmansour, S.; Minguez Espallargas, G.;
Clemente-Leon, M.; Abherve,́ A.; Gomez-Claramunt, P.; Coronado,
E.; Artizzu, F.; Sessini, E.; Deplano, P.; Serpe, A.; Mercuri, M. L.;

Article

Gomez García, C. J. Inorg. Chem. 2013, 52, 10031−10040. (b) Atzori,
M.; Pop, F.; Auban-Senzier, P.; Gomez García, C. J.; Canadell, E.;
Artizzu, F.; Serpe, A.; Deplano, P.; Avarvari, N.; Mercuri, M. L. Inorg.
Chem. 2014, 53, 7028−7039. (c) Abherve,́ A.; Clemente-Leon, M.;
Coronado, E.; Gomez-García, C. J.; Verneret, M. Inorg. Chem. 2014,
53, 12014−12026. (d) Abherve,́ A.; Mañas-Valero, S.; ClementeLeon, M.; Coronado, E. Chem. Sci. 2015, 6, 4665−4673. (e) Atzori,
M.; Pop, F.; Auban-Senzier, P.; Clerac, R.; Canadell, E.; Mercuri, M.
L.; Avarvari, N. Inorg. Chem. 2015, 54, 3643−3653. (f) Benmansour,
S.; Gomez García, C. J. Polymers 2016, 8, 89. (g) Benmansour, S.;
Abherve,́ A.; Gomez-Claramunt, P.; Valles-García, C.; Gomez-García,
C. J. ACS Appl. Mater. Interfaces 2017, 9, 26210−26218. (h) PalaciosCorella, M.; Fernandez-Espejo, A.; Bazaga-García, M.; Losilla, E. R.;
Cabeza, A.; Clemente-Leon, M.; Coronado, E. Inorg. Chem. 2017, 56,
13865−13877. (i) Kingsbury, C. J.; Abrahams, B. F.; D’Alessandro, D.
M.; Hudson, T. A.; Murase, R.; Robson, R.; White, K. F. Cryst. Growth
Des. 2017, 17, 1465−1470. (j) DeGayner, J. A.; Jeon, I.-R.; Sun, L.;
Dinca, M.; Harris, T. D. J. Am. Chem. Soc. 2017, 139, 4175−4184.
(k) Benmansour, S.; Perez-Herraez, I.; Lopez-Martínez, G.; Gomez
García, C. J. Polyhedron 2017, 135, 17−25. (l) Benmansour, S.;
Hernandez-Paredes, A.; Gomez-García, C. J. J. Coord. Chem. 2018, 71,
845.
(12) (a) Imaz, I.; Mouchaham, G.; Roques, N.; Brandes, S.; Sutter,
J.-P. Inorg. Chem. 2013, 52, 11237−11243. (b) Benmansour, S.;
Valles-García, C.; Gomez-Claramunt, P.; Minguez Espallargas, G.;
Gomez García, C. J. Inorg. Chem. 2015, 54, 5410−5418. (c) Sumida,
K.; Hu, M.; Furukawa, S.; Kitagawa, S. Inorg. Chem. 2016, 55, 3700−
3705. (d) Murase, R.; Abrahams, B. F.; D’Alessandro, D. M.; Davies,
C. G.; Hudson, T. A.; Jameson, G. N. L.; Moubaraki, B.; Murray, K.
S.; Robson, R.; Sutton, A. L. Inorg. Chem. 2017, 56, 9025−9035.
(13) Taniguchi, K.; Chen, J.; Sekine, Y.; Miyasaka, H. Chem. Mater.
2017, 29, 10053−10059.
(14) (a) Jeon, I.-R.; Negru, B.; Van Duyne, R. P.; Harris, T. D. J. Am.
Chem. Soc. 2015, 137, 15699−15702. (b) Darago, L. E.; Aubrey, M.
L.; Yu, C. J.; Gonzalez, M. I.; Long, J. R. J. Am. Chem. Soc. 2015, 137,
15703−15711. (c) DeGayner, J. A.; Jeon, I.-R.; Sun, L.; Dinca,̆ M.;
Harris, T. D. J. Am. Chem. Soc. 2017, 139, 4175−4184.
(15) (a) Rehwoldt, R. E.; Chasen, B. L.; Li, J. B. Anal. Chem. 1966,
38, 1018−1019. (b) Atzori, M.; Pop, F.; Cauchy, T.; Mercuri, M. L.;
Avarvari, N. Org. Biomol. Chem. 2014, 12, 8752−8763.
(16) Atzori, M.; Artizzu, F.; Marchio,̀ L.; Loche, D.; Caneschi, A.;
Serpe, A.; Deplano, P.; Avarvari, N.; Mercuri, M. L. Dalton Trans
2015, 44, 15786−15802.
(17) Ashoka Sahadevan, S.; Monni, N.; Abherve,́ A.; Auban-Senzier,
P.; Canadell, E.; Mercuri, M. L.; Avarvari, N. Inorg. Chem. 2017, 56,
12564−12571.
(18) Ashoka Sahadevan, S.; Monni, N.; Abherve,́ A.; Marongiu, D.;
Sarritzu, V.; Sestu, N.; Saba, M.; Mura, A.; Bongiovanni, G.; Cannas,
C.; Quochi, F.; Avarvari, N.; Mercuri, M. L. Chem. Mater. 2018,
DOI: 10.1021/acs.chemmater.8b03399.
(19) Weiss, S.; Krommer, H. Chem. Abstr. 1986, 104, 206730.
(20) Tao, J. M.; Perdew, J. P.; Staroverov, V. N.; Scuseria, G. E. Phys.
Rev. Lett. 2003, 91, 146401.
(21) Jensen, K. P. Inorg. Chem. 2008, 47, 10357−10365.
(22) Hariharan, P. C.; Pople, J. A. Theor. Chim. Acta 1973, 28, 213−
222.
(23) Frisch, M. J.; Trucks, G. W.; Schlegel, H. B.; Scuseria, G. E.;
Robb, M. A.; Cheeseman, J. R.; Scalmani, G.; Barone, V.; Mennucci,
B.; Petersson, G. A.; Nakatsuji, H.; Caricato, M.; Li, X.; Hratchian, H.
P.; Izmaylov, A. F.; Bloino, J.; Zheng, G.; Sonnenberg, J. L.; Hada, M.;
Ehara, M.; Toyota, K.; Fukuda, R.; Hasegawa, J.; Ishida, M.;
Nakajima, T.; Honda, Y.; Kitao, O.; Nakai, H.; Vreven, T.;
Montgomery, J. A., Jr., Peralta, J. E.; Ogliaro, F.; Bearpark, M.;
Heyd, J. J.; Brothers, E.; Kudin, K. N.; Staroverov, V. N.; Kobayashi,
R.; Normand, J.; Raghavachari, K.; Rendell, A.; Burant, J. C.; Iyengar,
S. S.; Tomasi, J.; Cossi, M.; Rega, N.; Millam, J. M.; Klene, M.; Knox,
J. E.; Cross, J. B.; Bakken, V.; Adamo, C.; Jaramillo, J.; Gomperts, R.;
Stratmann, R. E.; Yazyev, O.; Austin, A. J.; Cammi, R.; Pomelli, C.;
Ochterski, J. W.; Martin, R. L.; Morokuma, K.; Zakrzewski, V. G.;

Voth, G. A.; Salvador, P.; Dannenberg, J. J.; Dapprich, S.; Daniels, A.
D.; Farkas, Ö .; Foresman, J. B.; Ortiz, J. V.; Cioslowski, J.; Fox, D. J.
Gaussian 09, Revision B1; Gaussian, Inc.: Wallingford CT, 2009.
(24) (a) Tamaki, H.; Zhong, Z. J.; Matsumoto, N.; Kida, S.;
Koikawa, M.; Achiwa, N.; Hashimoto, Y.; Okawa, H. J. Am. Chem. Soc.
1992, 114, 6974−6979. (b) Clemente-Leon, M.; Coronado, E.;
Lopez-Jorda,̀ M.; Mínguez Espallargas, G.; Soriano-Portillo, A.;
Waerenborgh, J. C. Chem. - Eur. J. 2010, 16, 2207−2219.
(25) Okaya, Y.; Pepinsky, R. Acta Crystallogr. 1957, 10, 681−684.
(26) Hartl, F.; Stufkens, D. J.; Vlcek, A. Inorg. Chem. 1992, 31,
1687−1695.
(27) Tholence, J. L. Solid State Commun. 1980, 35, 113−117.
(28) (a) Novak, M. A. J. Magn. Magn. Mater. 2004, 272−276, e707−
e713. (b) Novak, M. A.; Folly, W. S. D.; Sinnecker, J. P.; Soriano, S. J.
Magn. Magn. Mater. 2005, 294, 133−140.
(29) (a) Gunnarsson, K.; Svedlindh, P.; Nordblad, P.; Lundgren, L.;
Aruga, H.; Ito, A. Phys. Rev. Lett. 1988, 61, 754. (b) Laiho, R.;
Lahderanta, E.; Salminen, J.; Lisunov, K. G.; Zakhvalinskii, V. S. Phys.
Rev. B: Condens. Matter Mater. Phys. 2001, 63, 094405. (c) Mattsson,
J.; Jonsson, T.; Nordblad, P.; Aruga Katori, H.; Ito, A. Phys. Rev. Lett.
1995, 74, 4305.
(30) (a) Sellers, S. P.; Korte, B. J.; Fitzgerald, J. P.; Reiff, W. M.; Yee,
G. T. J. Am. Chem. Soc. 1998, 120, 4662−4670. (b) Kaul, B. B.;
Durfee, W. S.; Yee, G. T. J. Am. Chem. Soc. 1999, 121, 6862−6866.
(c) Buschmann, W. E.; Ensling, J.; Gütlich, P.; Miller, J. S. Chem. Eur. J. 1999, 5 (10), 3019−3028. (d) Clerac, R.; O’Kane, S.; Cowen,
J.; Ouyang, X.; Heintz, R.; Zhao, H.; Bazile, M. J.; Dunbar, K. R.
Chem. Mater. 2003, 15, 1840−1850. (e) Krishnamohan Sharma, C.
V.; Chusuei, C. C.; Clerac, R.; Möller, T.; Dunbar, K. R.; Clearfield, A.
Inorg. Chem. 2003, 42, 8300−8308. (f) Zhang, X.-M.; Li, P.; Gao, W.;
Liu, J.-P.; Gao, E.-Q. Dalton Trans 2015, 44, 511−514. (g) Ghosh, S.;
Roy, S.; Liu, C.-M.; Mohanta, S. Dalton Trans 2018, 47, 836−844.
(31) Mydosh, J. A. Spin Glasses: an Experimental Introduction; Taylor
and Francis: London, 1993.
(32) Choi, K. Y.; Wang, Z.; Ozarowski, A.; van Tol, J.; Zhou, H. D.;
Wiebe, C. R.; Skourski, Y.; Dalal, N. S. J. Phys.: Condens. Matter 2012,
24, 246001−6.
(33) Greenwood, N. N.; Gibb, T. C. Mössbauer Spectroscopy;
Chapman and Hall, Ltd. Publishers: London, 1971.
(34) Shilov, G. V.; Nikitina, Z. K.; Ovanesyan, N. S.; Aldoshin, S.
M.; Makhaev, V. D. Russ. Chem. Bull. 2011, 60, 1209−1219.
(35) For example: (a) Lupu, D.; Barb, D.; Filoti, G.; Morariu, M.;
Tarina, D. J. Inorg. Nucl. Chem. 1972, 34, 2803−2810. (b) Maeda, Y.
J. Nucl. Radiochem. Sci. 2006, 7, R13−R18.
(36) Bhattacharjee, A.; Bhakat, D.; Roy, M.; Kusz, J. Phys. B 2010,
405, 1546−1550.
(37) For a very pedagogic presentation of the basic methodology,
see: Launay, J.-P.; Verdaguer, M. Electrons in Molecules: From Basic
Principles to Molecular Electronics; Oxford University Press: Oxford,
2014; Chapter 3.
(38) Austin, I. G.; Mott, N. F. Adv. Phys. 1969, 18, 41−102.
(39) Tsuda, N.; Nasu, K.; Yanase, A.; Siratori, K. Electronic
conduction in Oxides; Springer Series in Solid-State Sciences; Springer
Verlag: Berlin Heidelberg, 1991; pp 163−168.