Switching the Spin-Crossover Phenomenon by Ligand Design on Imidazole-Diazineiron(II) Complexes

Article abstract of DOI:10.1021/acs.inorgchem.8b02278

The iron(II) complexes of two structural isomers of 2-(1H-imidazol-2-yl)diazine reveal how ligand design can be a successful strategy to control the electronic and magnetic properties of complexes by fine-tuning their ligand field. The two isomers only differ in the position of a single diazinic nitrogen atom, having either a pyrazine (Z) or a pyrimidine (M) moiety. However, [Fe(M)₃](ClO₄)₂ is a spin-crossover complex with a spin transition at 241 K, whereas [Fe(Z)₃](ClO₄)₂ has a stable magnetic behavior between 2 and 300 K. This is corroborated by temperature-dependent M?ssbauer spectra showing the presence of a quintet and a singlet state in equilibrium. The temperature-dependent single-crystal X-ray diffraction results relate the spin-crossover observed in [Fe(M)₃](ClO₄)₂ to changes in the bond distances and angles of the coordination sphere of iron(II), hinting at a stronger σ donation of ligand Z in comparison to ligand M. The UV/vis spectra of both complexes are solved by means of the multiconfigurational wave-function-based method CASPT2 and confirm their different spin multiplicities at room temperature, as observed in the M?ssbauer spectra. Calculations show larger stabilization of the singlet state in [Fe(Z)₃]²⁺ than in [Fe(M)₃]²⁺, stemming from the slightly stronger ligand field of the former (506 cm^-1 in the singlet). This relatively weak effect is indeed capable of changing the spin multiplicity of the complexes and causes the appearance of the spin transition in the M complex.

Full text of DOI:10.1021/acs.inorgchem.8b02278

Article
Cite This: Inorg. Chem. XXXX, XXX, XXX−XXX
pubs.acs.org/IC
Switching the Spin-Crossover Phenomenon by Ligand Design on
Imidazole−Diazineiron(II) Complexes
Naheed Bibi, Eduardo Guimarae
̃
s Ratier de Arruda, Alex Domingo, Aline Alves Oliveira,⊥
†
,‡
†,‡
§
†
§
†
∥
Carolina Galuppo, Quan Manh Phung, Naíma Mohammed Orra, Fanny Beron,
́
⊥
§
,†
Andrea Paesano, Jr., Kristine Pierloot, and Andre
́
Luiz Barboza Formiga*
†
Institute of Chemistry University of CampinasUNICAMP, P.O. Box 6154, Campinas, Sao
̃
Paulo 13083-970, Brazil
§
Department of Chemistry, KU Leuven, Celestijnenlaan 200F, Leuven B-3001, Belgium
⊥
∥
Universidade Estadual de Maringa, Maringa, Brazil
́ ́
Institute of Physics Gleb Wataghin, University of CampinasUNICAMP, Rua Ser
́
gio Buarque de Holanda 777, Campinas, Sao
̃
Paulo 13083-859, Brazil
S Supporting Information
*
ABSTRACT: The iron(II) complexes of two structural
isomers of 2-(1H-imidazol-2-yl)diazine reveal how ligand
design can be a successful strategy to control the electronic
and magnetic properties of complexes by fine-tuning their
ligand field. The two isomers only differ in the position of a
single diazinic nitrogen atom, having either a pyrazine (Z) or a
pyrimidine (M) moiety. However, [Fe(M) ](ClO ) is a spin-
3
4 2
crossover complex with a spin transition at 241 K, whereas
Fe(Z) ](ClO ) has a stable magnetic behavior between 2
[
3
4 2
and 300 K. This is corroborated by temperature-dependent
Mossbauer spectra showing the presence of a quintet and a singlet state in equilibrium. The temperature-dependent single-
crystal X-ray diffraction results relate the spin-crossover observed in [Fe(M) ](ClO ) to changes in the bond distances and
̈
3
4 2
angles of the coordination sphere of iron(II), hinting at a stronger σ donation of ligand Z in comparison to ligand M. The UV/
vis spectra of both complexes are solved by means of the multiconfigurational wave-function-based method CASPT2 and
confirm their different spin multiplicities at room temperature, as observed in the Mo
̈
ssbauer spectra. Calculations show larger
stabilization of the singlet state in [Fe(Z) ] than in [Fe(M) ] , stemming from the slightly stronger ligand field of the former
2+
2+
3
3
−
1
(
506 cm in the singlet). This relatively weak effect is indeed capable of changing the spin multiplicity of the complexes and
causes the appearance of the spin transition in the M complex.
INTRODUCTION
different metal−ligand length and the entropy change related
1−12
■
to the electronic and vibrational degrees of freedom.
Hence, it is possible to trigger the spin transition by external
Memory storage on digital electronics requires materials
capable of switching between two stable and distinguishable
states that represent the 1 and 0 bits of information. Ideally,
such materials should have two long-lived states with different
magnetic or electronic properties under the same external
conditions and allow a fast transition between them triggered
by some external stimuli. Enticing compounds to be used in
these technological applications can be found in the family of
spin-crossover (SCO) materials. The SCO phenomenon
occurs on systems with two spin states of different total spin
number, where the spin transition is accompanied by some
structural change. It is typically found in octahedral transition-
metal complexes of the fourth period with a 3d to 3d metal
center in a ligand field (LF) close to the limit between the low-
spin (LS) and high-spin (HS) regimes. The LS−HS transition
in these complexes induces an expansion of the coordination
sphere and a corresponding change of the LF. The equilibrium
between both spin states of the system is governed by two
competing processes, the enthalpy change related to the
13,14
stimuli affecting the molecular structure (pressure)
vibrational entropy (temperature)
or the
or through an elec-
tronic excitation induced by light radiation in the so-called
15−17
18−24
light-induced excited spin trapping.
The complex interplay of geometrical and electronic factors
behind the SCO phenomenon calls for thorough systematic
studies to unravel the structure−function relationship govern-
25,26
ing the SCO.
Available data for many different SCO
complexes and their derivatives reveal that too large distortions
between the LS and HS states overstabilize one of them, thus
hindering the spin transition. Nonetheless, a sufficient
distortion is needed during the spin transition to switch the
LF between the LS and HS regimes. Hence, many
pseudooctahedral iron(II) complexes with nitrogen-donor
4
7
Received: August 10, 2018
©
XXXX American Chemical Society
A
DOI: 10.1021/acs.inorgchem.8b02278
Inorg. Chem. XXXX, XXX, XXX−XXX

Inorganic Chemistry
Article
chelating ligands are very successful SCO compounds because
they show relatively large structural changes (∼10% on the
Fe−N bond) while still allowing the forward and backward
spin transition to occur. At the molecular level, not only do the
distortions affect the coordination sphere, but they may also
extend to the ligands. Therefore, the bulkiness, flexibility, and
denticity of the ligands can have an impact on thermal SCO
and its transition temperature (T1/2, temperature at 50:50 of
LS and HS), which is a measure of the zero-point-corrected
major goal of designing functional materials with tailored SCO
properties.
We have synthesized and characterized two structural
isomers of 2-(1H-imidazol-2-yl)diazine (Scheme 1) and their
49,50
a
Scheme 1. Structures of 1 and 2 Used in This Work
0
0
energy difference between the HS and LS states (ΔE = E
HL
HS
0
27,28
−
E ).
Relative stabilization of the more voluminous HS
LS
state should lead to lower T1/2, while stabilization of the
compact LS state should increase T1/2. However, even at the
molecular level, control over T1 is not straightforward due to
/2
0
HL
the sensitiveness of ΔE to structural factors.
The effect of stereoisomerism on the magnetic properties of
complexes has been extensively studied. The SCO properties
can be very different from one isomer to another because of
a
The structures correspond to mer isomers because the fac ones were
29−32
not observed experimentally.
the different LFs experienced by the metal ion.
Structural
isomerism also has a very important influence on the magnetic
33−35
respective iron(II) complexes. This family of ligands combines
an electron-deficient diazine with an electron-rich imidazole,
conferring them high versatility. The two isomers differ in the
diazine moiety, having either a pyrazine (Z) or a pyrimidine
properties, and several examples show that both linkage
and substitutional isomers have quite different magnetic
3
6−41
properties.
Isomerism can change the magnetic behavior
of materials even when the structural difference is centered on
51
4
2
(
M), and were partially explored as ligands in the literature.
counterions. The resulting geometric changes are generally
accompanied by changes in both the LF and intermolecular
interactions responsible for the packing in the solid. Associated
with it, cooperative effects are sometimes present in one
isomer and absent in another and, as a consequence, it is
difficult to disentangle the relative importance of both factors
in defining the magnetic properties.
Diazines are electron-deficient heterocycles well-known for
their strong π-acceptor character related to their optic
52,53
properties and noninnocent behavior.
Conversely, imida-
zole-based ligands are prominent not only for their electron-
54,55
rich character, being stronger σ donors than azines,
but
also for the possibility of altering the electronic structure and
reactivity of the compound through deprotonation, hydrogen-
The effect of changing the positions of the heteroatoms in
the ligands has been studied before for the case of tridentated
ligands analogous to terpyridine. Goodwin and co-workers
have reported a comparison between the two isomers of
homoleptic iron(II) complexes obtained with 2,6-di(thiazol-4-
56−59
bonding modulation, or derivatization of the NH group.
Thus, the combination of such properties in this class of
ligands has the potential to provide a tunable LF through
chemical modifications. Moreover, the related SCO iron(II)
complexes can have the required stability and solubility to be
used in the development of novel materials, such as the 2-(1H-
imidazol-2-yl)pyridineiron(II) complex used by Coronado et
al. in a hybrid multifunctional material combining super-
4
3
44
yl)pyridine and 2,6-di(thiazol-2-yl)pyridine; however, both
are LS at room temperature. The substitution of thiazolyl by
pyrazolyl groups has recently been reviewed, and these ligands
have been extensively used to study homoleptic iron(II)
60
conductivity and SCO magnetism.
2+
complexes with the formula [Fe(bpp) ] , in which bpp can be
2
The homoleptic complexes [Fe(Z) ](ClO ) (1) and
3
4 2
either 2,6-di(pyrazol-1-yl)pyridine or 2,6-di(1H-pyrazol-3-yl)-
[
Fe(M) ](ClO ) (2) show strikingly different optical and
4
5
3 4 2
pyridine. Again, both isomers have similar magnetic
properties showing SCO but very sensitive to differences in
packing. Further modification of the ligand by a change from
pyridine to pyrazine as the central ring does not show a
magnetic properties. The preeminent one is the appearance of
a spin transition at 241 K for 2, whereas 1 shows an almost flat
magnetic susceptibility in the same range of temperatures. We
explain the origin behind such differential behavior through a
detailed characterization of the compounds and an in-depth
computational study using multiconfigurational wave-function-
based methods. We have used temperature-dependent experi-
ments with a combination of single-crystal crystallography,
46
significant difference in the SCO properties. More recent
examples report bidentate isomeric ligands used to obtain
iron(II) showing SCO; however, the same combined effects
47,48
(
LF/packing) cannot be easily isolated.
The aim of the present work is to achieve better control of
magnetic susceptibility, and Mo
the spin states of both complexes. Mo
̈
ssbauer spectroscopy to unveil
ssbauer spectroscopy is a
the SCO phenomenon by tuning the LF on the metal center
through structural isomerism of the ligands without serious
perturbation of the packing in the solid state. We will show
that small changes in the ligand structure involving the position
of a single heteroatom can fully switch the SCO behavior of
the complex. Such a limited structural alteration leads to
structure−function relationships lying in the molecular
domain, acting on the LF, and not involving cooperative
effects. Thus, the knowledge obtained from these experiments
is less system-dependent and more likely to be transferred to a
wider range of compounds, which is helpful to attaining the
̈
powerful tool in probing the oxidation and spin states of iron
in coordination compounds. In particular, because of the
isomer shift δ and the quadrupole splitting ΔE , which
Q
presents significantly different values between HS and LS
61
iron(II), we were able to unambiguously assign the spin state
of each complex in different temperatures. We also use highly
elaborate correlated wave-function methods (CASPT2/CC =
second-order perturbation theory based on a multiconfigura-
tional reference wave function, combined with CCSD(T) for a
62
description of the 3s3p correlation on the metal ) to describe
B
DOI: 10.1021/acs.inorgchem.8b02278
Inorg. Chem. XXXX, XXX, XXX−XXX

Inorganic Chemistry
Article
the spin-state energetics of these two complexes and to settle
the electronic structure of their ground state. CASPT2 is also
used to evaluate the transitions to the first manifold of metal-
to-ligand charge-transfer (MLCT) states to resolve their UV/
vis absorption spectra.
[Fe(Z)
](ClO ) (1). Yield: 60% Anal. Calcd for [Fe(C H N ) ]-
3 4 2 7 6 4 3
(
ClO ) (H O): C, 35.46; H, 2.83; N, 23.63. Found: C, 35.88; H, 2.8;
4
2
2
2
+
N, 23.43. ESI-MS (CH CN): 247.0 ([M] ). Molar conductivity in
3
2
−1
H O: 230 S cm mol . FTIR: 637, 760, 1122, 1461, 1633, 3488
2
−
1
cm . Single crystals were obtained by slow diffusion of an CH CN
3
solution of the complex into toluene at room temperature.
[
Fe(M) ](ClO ) (2). Yield: 64% Anal. Calcd for [Fe(C H N ) ]-
3
4 2
7
6
4 3
EXPERIMENTAL SECTION
(ClO
4
)
2
(3H O): C, 33.75; H, 3.23; N, 22.49. Found: C, 34.09; H,
2
■
2+
3
.2; N, 22.08. ESI-MS (CH OH): 279.0 ([M + 2CH OH] ). Molar
conductivity in H O: 155.3 S cm mol . FTIR: 626, 708, 1090, 1411,
470, 1586, 3101, 3403 cm . Single crystals were obtained by slow
3
3
Syntheses of Ligands. All reagents were obtained from
commercial sources and used without purification. Ligands Z and
M were synthesized following an adaptation of the procedure
2
−1
2
−1
1
evaporation of a solution of the complex in a 1:1 mixture of EtOH
and MeOH at room temperature.
63−65
described by Voss et al.
-(1H-Imidazol-2-yl)pyrazine (Z). In a 50-mL round-bottom flask,
2
Physical Measurements. Electronic spectra in the 190−1100 nm
range were acquired by using a 1.00 cm quartz cuvette in a HP8453
diode-array UV/vis absorption spectrophotometer equipped with a
Peltier HP89090A.
a mixture of 0.471 mL (5 mmol) of 2-pyrazinecarbonitrile, 5 mL of
methanol (MeOH), and 0.47 mL (1 mmol) of a 30% solution of
sodium methoxide was stirred at room temperature for 20 min. A total
of 0.727 mL (5 mmol) of 2,2-diethoxyethanamine was added,
followed by the dropwise addition of 0.6 mL of acetic acid. The
system was kept under reflux for 1 h at 50 °C. A total of 10 mL of
MeOH and 2.5 mL of 6 mol L− HCl were added after the mixture
was cooled to room temperature, after which the system was kept
under reflux for 5 h. The solution was evaporated, and the residue was
IR spectra were obtained with an Agilent Cary 630 Fourier
transform infrared (FTIR) spectrometer, using the attenuated total
reflectance method, with a diamond cell. The spectra were recorded
1
−1
in the 4000−400 cm range, with 64 scans and a resolution of 4
−1
cm .
Electrospray ionization mass spectrometry (ESI-MS) measure-
ments were carried out using a Waters Quattro Micro API
spectrometer. Samples were evaluated in the positive mode either
dissolved in 15 mL of water (H O) and extracted with 15 mL of
2
diethyl ether (Et O), which resulted in a brown solution. After the pH
2
−1
was adjusted to 8−9 with 2 mol L NaOH, the product precipitated
as a brown solid, which was filtered and washed with Et O. Yield:
in a 1:1 MeOH/H O solution with the addition of 0.10% (v/v)
2
2
formic acid or in CH CN.
3
4
5
1%. Anal. Calcd for C H N : C, 57.53; H, 4.14; N, 38.33. Found: C,
7.63; H, 4.16; N, 38.14. ESI-MS (MeOH): 147.1 ([M + H] ). H
7 6 4
Elemental analyses for carbon, hydrogen, and nitrogen were
performed using a PerkinElmer 2400 CHN analyzer.
+
1
NMR (500 MHz, DMSO-d ): δ 13.058 (s, 1H), 9.23 (d, 1H, J = 1.5
6
Conductivity measurements of the complex solutions were
obtained using a LAB1000 conductometer and a standard platinum
Hz), 8.649 (dd, 1H, J = 2.5 and 1.5 Hz), 8.595 (d, 1H, J = 2.5 Hz),
1
3
7
1
0
.25 (s, 1H), 7.17 (s, 1H). C NMR (125.7 MHz, DMSO-d ): δ
−3
−1
6
electrode cell with K = 0.1. Data were obtained with 10 mol L
44.90, 144.24, 143.92, 143.70, 141.80, 130.66, 120.27. pKaH = 4.51 ±
aqueous solutions of the complexes.
Potentiometric titrations were used to calculate the pKaH values of
.05.
2-(1H-Imidazol-2-yl)pyrimidine (M). A mixture of 2.62 g (25
the ligands using a 827 pH Lab Metrohm pH meter and a standard
mmol) of 2-pyrimidinecarbonitrile, 10 mL of MeOH, and 0.47 mL of
a 30% solution of sodium methoxide in MeOH was kept under
vigorous stirring at room temperature for 1.5 h in a 100 mL flask.
Then, 1.63 mL (15 mmol) of 2,2-dimethoxyethylamine was added to
the mixture, followed by the dropwise addition of 2.75 mL of acetic
acid. The system was kept under reflux for 30 min. After cooling to
−2
−1
electrode using 1.12 × 10 mol L HCl as the titrant. The initial pH
values for the ligand solutions were 6.82 and 7.78 for Z and M,
respectively.
1H and 13C NMR spectra were acquired in deuterated dimethyl
sulfoxide (DMSO-d ) solutions using Bruker Avance III 500 MHz
6
(
11.7 T) and 600 MHz (14.09 T) spectrometers.
Temperature-dependent direct-current magnetic susceptibilities
−
1
room temperature, 15 mL of MeOH and 12.5 mL of 6 mol L HCl
were added to the solution, which was heated again and kept under
were obtained from dried powder samples (ca. 50 mg) mounted in
a sealed plastic sample holder on a Quantum Design MPMS XL
superconducting quantum interference device (SQUID) magneto-
meter in the temperature range of 2−300 K under an applied field of
0.1 T. The magnetization data for the samples were corrected by
subtraction of the background due to the sample holder and also for
reflux for 5 h. After the reaction time, the solvent was evaporated and
1
2
7.5 mL of a warm 1 g mL− K CO solution was carefully added to
2
3
the residue, yielding a yellowish suspension, which was filtered.
Extraction of the precipitate with 3 × 50 mL of MeOH yielded a
yellowish solution, which was concentrated until precipitation. The
67
resulting pale-yellow solid was filtered, washed with Et O, and dried
diamagnetic contributions estimated by Pascal constants.
Mossbauer spectra were recorded in transmission geometry, using a
Co/Rh source, with a 50 mCi nominal starting activity. A closed-
2
under vacuum. Yield: 65%. Anal. Calcd for C H N : C, 57.53; H,
̈
7
6
4
5
7
4
.14; N, 38.33. Found: C, 57.76; H, 4.22; N, 38.24. ESI-MS
+
1
(
MeOH): 147.1 ([M + H] ). H NMR (600 MHz, DMSO-d ): δ
6
cycle helium cryostat was used for low-temperature measurements.
The spectrometer was calibrated with a metallic iron foil as the
absorber at room temperature, and the spectra were fitted by taking
the absorption lines as Lorentzian functions.
1
(
2.97 (s, 1H), 8.865 (d, 2H, J = 4.8 Hz), 7.43 (t, 1H, J = 4.8 Hz), 7.22
s, 2H). 13C NMR (100.6 MHz, DMSO-d ): δ 158.11, 157.41,
6
1
44.85, 125.98, 120.41. pKaH = 5.07 ± 0.07.
Syntheses of the Iron(II) Complexes. Complexes were obtained
Crystal Structure Determination. Data collection for single
crystals was performed with a Bruker Apex II CCD diffractometer
using either Mo Kα (λ = 0.71703 Å) or Cu Kα (λ = 1.54178 Å)
radiation. An Oxford Cryosystems controller was used in all cases.
Unit cell dimensions and orientation matrices were determined by
least-squares refinement of the reflections obtained by ϕ−ω scans.
The data were indexed and scaled with the ApexII suite [APEX2,
v2014.1-1 (Bruker AXS Inc.), and SAINT, V8.34A (Bruker AXS Inc.,
2013)]. Bruker SAINT and SADABS were used for integration and
66
using the procedure given by Stupka et al. A solution of Fe(ClO ) ·
4
2
6
H O (41.7 mg, 0.115 mmol) in ethanol (EtOH; 2.5 mL) was added
2
dropwise to a solution of 50 mg (0.345 mmol) of the corresponding
ligand in EtOH (2.5 mL). The color of the mixture turned
2
+
immediately to violet in the case of [Fe(Z) ] and reddish orange
3
2+
for [Fe(M) ] . Thereafter, the mixture was stirred for 4 h and
3
concentrated using a rotary evaporator. The resulting concentrated
solution was left in a refrigerator overnight, which resulted in a
colored powder of the desired complexes. In both cases, the powder
68
scaling of the data, respectively. Using Olex2, the structure was
69
was filtered from the corresponding mixture, washed with Et O, and
solved with the ShelXT structure solution program using direct
2
70
dried under reduced pressure. Both complexes are soluble in H O,
methods and refined with ShelXL by a full-matrix least-squares
2
2
EtOH, MeOH, acetonitrile (CH CN), dimethyl sulfoxide (DMSO),
technique on F . All non-hydrogen atoms were refined anisotropically,
3
acetone, and pyridine. (Caution! Perchlorate compounds should be
handled with care due to the risk of explosion.)
and hydrogen atoms were added to the structure in idealized positions
and further refined according to the riding model.
C
DOI: 10.1021/acs.inorgchem.8b02278
Inorg. Chem. XXXX, XXX, XXX−XXX

Inorganic Chemistry
Article
6
2
Powder X-ray diffraction (XRD) data were collected with a
Shimadzu XRD-7000 diffractometer with Bragg−Brentano geometry
and a graphite monochromator. The Cu Kα (1.5406 Å) X-ray
radiation source was operated at 40 kV and 30 mA. The collected 2θ
range was from 5.0 to 50.0° under 2° min− (2θ). The apertures of
the divergence, scattering, and receiving slits were at 1.0, 1.0, and 0.3
mm, respectively.
our previous paper. In this set of calculations, only the valence
electrons were correlated at the CASPT2 level because it was shown
before that CASPT2 generally fails in the description of the metal
9
1
(3s,3p) correlation contribution to the spin-state energetics.
1
Therefore, the latter contribution was instead obtained with
CCSD(T), making use of much smaller ligand basis sets of the type
MINAO (minimal basis sets constructed from the cc-pVTZ basis
sets) combined with aug-cc-pwCVTZ-DK on iron. This approach,
denoted as CASPT2/CC in our recent study, was shown to give
highly accurate spin-state energetics for a broad series of iron
Computational Details. We optimized the geometry of the mer-
2+
[
Fe(M)3]2+ and mer-[Fe(Z)3] complexes in a vacuum using density
7
1−73
functional theory (DFT) with the PBE0 functional
and Def2-
74
62
TZVP basis set. The choice of the PBE0 functional is based on
previous experience showing that it produces accurate molecular
complexes.
In all CASPT2 and CCSD(T) calculations, scalar relativistic effects
were included using the standard second-order Douglas−Kroll−Hess
7
5−77
structures of transition-metal complexes.
Because the mer
92
isomers have no spatial symmetry (C ) because of the orientation
1
Hamiltonian. The CASPT2 calculations use the default IPEA shift
93
of the three bidentate ligands (M or Z), all calculations were
performed without symmetry constraints. The character of the
stationary points on the potential energy surface of the optimized
structures was verified by frequency analysis. DFT/PBE0 calculations
of 0.25 hartree and an imaginary shift of 0.1 hartree to avoid
94
possible weak intruder states. The CASPT2 calculations were
carried out with the Molcas 8 package. Here, the two-electron
integrals were decomposed by means of the Cholesky technique with
a threshold of 10 hartree. For the CCSD(T) calculations, we used
the Molpro V.2012 package. All CCSD(T) calculations are
partially) spin-restricted, on the basis of restricted open-shell
Hartree−Fock orbitals.
Because characterization of the complexes is done in MeOH, we
included solvent effects in the calculations of the electronic spectra
through a polar continuum model (PCM) with the dielectric
constant of MeOH (32.63) and a surface element average area of 0.4
Å . For the excited states, only the fast component of the reaction field
was calculated, while the slow component was taken from the ground
95
78
were carried out with the TURBOMOLE v6.4 package.
We evaluated the spin-state energetics and metal-centered
−6
96
97
2
+
2+
transitions of [Fe(M)3] and [Fe(Z)3] by means of the
(
complete-active-space second-order perturbation theory
79,80
(
CASPT2),
using a complete active space CAS[10,12] formed
by 10 electrons in 12 molecular orbitals (MOs) that we built
81,82
following established procedures.
The 12 active MOs are five
98
Fe(3d)-like orbitals, plus five Fe(4d)-like orbitals to account for the
8
3
so-called double-shell effect and two σ-type orbitals on the
coordinating nitrogen atoms to account for the σ donation. π-back-
donation to π* orbitals of the ligands is negligible in the ground and
LF excited states of both complexes.
2
99
state. Each charge-transfer excitation was fully localized on a single
ligand and, hence, each MLCT transition clearly identified based on
its induced polarization. Therefore, we treat all transitions to the same
ligand molecule with a specific PCM that has its fast response
component equilibrated to one of those states. For instance, the nine
For calculation of the CT spectra, the CAS[10,12] active space was
extended to CAS[10,15] by adding the lowest π* orbital of each
ligand. This CAS[10,15] space allows us to estimate the vertical
transitions in the Franck−Condon region to the first manifold of the
MLCT states. These excitations consist of spin-allowed electron
transfer from the doubly occupied Fe(3d)-like MOs to one π* MO
localized on a M or Z ligand. Hence, the size of the manifold depends
on the total spin of the electronic state of departure, which varies from
1
2+
MLCT transitions of [Fe(Z)3] form three groups of three charge
transfers to each Z ligand. Because all transitions to the same ligand
induce the same qualitative polarizations, we use a common PCM for
each of those three groups.
5
only one possible electron transfer per ligand for HS to three MLCTs
1
per ligand for LS. It is imperative to use a set of MOs fully adapted to
RESULTS AND DISCUSSION
■
the charge-transfer states to properly describe rearrangement of the
charges between the metal center and ligands induced by the
The complexes were synthesized using an adapted procedure
from Stupka et al. with iron(II) perchlorate.
84,85
100
MLCTs.
Thus, we use two different sets of MOs to evaluate each
Reaction
MLCT excitation. The ground state is described by a set of orbitals
resulting from a state-specific CAS[10,12]SCF calculation, whereas
the MOs for the MLCT states are obtained from a CAS[10,15]SCF
state average including exclusively the MLCT excited states of the
manifold in the orbital optimization procedure. Hence, the manifold
of nine singlet MLCTs (three transitions per ligand) is described with
a unique set of MOs obtained from a state average of those nine
kinetics were quick, and the color changed immediately after
the addition of the metal salt to the ligand solution. The Z
complex synthesis resulted in a violet powder, whereas the M
complex was obtained as a reddish-orange powder. Hence,
their electronic spectra are expected to be very distinct from
one another, even though they are isomers (vide infra).
1
5
Another important feature of the complexes is their H NMR
MLCT excited states, while for HS, we use a state average of three
quintet MLCTs (one transition per ligand). Additionally, because the
two states involved in the MLCT transitions are described with
different MO sets, we performed a state-interaction step to ensure the
orthogonality and estimate their oscillator strength.
CASPT2 calculations on the vertical LF and MLCT spectrum were
performed with a basis set of the ANO-RCC type contracted to
spectra showing broad bands preventing peak assignment,
typical of paramagnetic samples.
Magnetic Susceptibilities and Mossbauer Spectra.
̈
86
Figure 1 shows the temperature dependence of the χ T
M
product for both complexes in the 2−300 K range. Both
samples are paramagnetic, with χ T larger than zero in all
M
[
7s6p5d3f2g1h] for iron, to [4s3p2d1f] for nitrogen and carbon, and
87,88
ranges but with very different temperature dependencies. At
to [3s1p] for hydrogen.
In these calculations, all valence electrons
3
00 K, 2 has a χ T value approximately 12 times larger than
and Fe(3s,3p) semicore electrons are correlated at the CASPT2 level.
On the other hand, because we have shown recently that an accurate
description of spin-state energetics requires considerably larger basis
M
that of 1, which decreases until 100 K, where χ T becomes
M
similar to that of the Z complex. The behavior of 2 indicates
that a HS species predominates at room temperature with an
incomplete conversion to a LS species upon cooling.
Inspection of the curve below 20 K hints at a full depopulation
of the HS state as the temperature is decreased. Although this
decrease is more pronounced for 2, the same trend is observed
for 1, indicating that the ground state for both complexes is a
singlet state at 0 K and that the paramagnetic nature of the
62
sets, calculation of the HS−LS splitting was also performed using
correlation-consistent basis sets. For the ligands, we used a
combination of aug-cc-pVTZ-DK on the heavy atoms and cc-
pVTZ-DK on hydrogen, while different basis sets were used for iron,
of the type aug-cc-pwCVnZ-DK, with n = 3−5. Extrapolation to the
complete basis set (CBS) limit on the metal was accomplished by
making use of two-point extrapolation equations (separate for
CASSCF and CASPT2) given in refs 89 and 90 and described in
D
DOI: 10.1021/acs.inorgchem.8b02278
Inorg. Chem. XXXX, XXX, XXX−XXX

Inorganic Chemistry
Article
Figure 1. Temperature dependence of the χ T product for 2 and 1 in
M
the 2−300 K range. The theoretical fitting of the curve exhibiting spin
1
5
transition corresponds to an equilibrium between the LS and HS
iron(II) states with T1/2 = 241 K. The ideal O curve was simulated
h
5
using van Vleck’s equation for a T ground state with λ = −100
2
g
−
1
cm .
samples is a consequence of the thermal population of low-
lying excited states.
Although a plateau is not observed at higher temperatures,
the χ T value for 2 tends to the value expected for an ideal O
T2g state. Taking this into account and considering that the
M
h
5
χ T value for the ground state is zero, the molar fraction of the
M
5
HS species (γ ) can be calculated as a function of the
HS
5
temperature using the theoretical values for the ideal T state.
2
g
The theoretical fit is a sigmoidal curve and reproduces very
well the experimental χ T data with a root-mean-square
M
3
−1
deviation of 0.30 cm K mol (Figures 1 and S6). The
sigmoidal behavior of the spin conversion is an indication that
Figure 2. Mo
̈
ssbauer spectra recorded at 300 and 22 K of (a) 2 fitted
1
5
using two doublets and (b) 1 fitted using one doublet.
a LS− HS equilibrium is in operation with T = 241 K (the
1/2
derivative of the experimental curve gives a broad maximum
ranging from 210 to 250 K).
The nature of the HS and LS species was confirmed by
̈
Mossbauer spectroscopy at two different temperatures. The
temperature, in agreement with the χ T versus T curve
M
(
Figure 1).
spectra shown in Figure 2a,b were fitted using one or two
doublets, and the hyperfine parameters are listed in Table 1.
The recorded spectra confirm the paramagnetic nature of 2,
even at low temperatures. It is clear from Figure 2a that
complex 2 presents two doublets, confirming the coexistence
of two different spin states at both temperatures. Doublet A
Complex 1 (Figure 2b) has only one iron(II) site, the
1
isomer shift value of which may be attributed to a LS state.
The spin configuration does not change with decreasing
temperature to 22 K, as can be verified by the small variation of
the Δ and δ hyperfine parameters. Again, the δ increase is
attributed to a second-order Doppler effect. The absence of the
−
1
always exhibits a large splitting (δ ≥ 1.00 mm s ) consistent
5
HS doublet in this case indicates that the observed increase in
5
with a HS, whereas doublet B presents an isomer shift typical
χT with the temperature for this sample (Figure 1) may be a
consequence of temperature-independent paramagnetism
1
of a LS. At room temperature, the HS area A /A was found
HS
tot
to be 64.5%, decreasing to 9.8% upon cooling to 22 K, in very
(
TIP). Indeed, the experimental data for this sample are very
good agreement with the values obtained from the fitting of the
−4
3
−1
well reproduced if a TIP contribution of 7.7 × 10 cm mol
is added to the Curie component (Figure S7).
5
magnetic curve in which the HS molar fraction was calculated
102
as 72.2% and 6.8%, respectively. The low isomer-shift increase
with decreasing temperature is caused by a second-order
Provided that a spin conversion is observed (as for 2) and
assuming that the system can be described by a regular-
101
Doppler effect, for both doublets. A pronounced change in
103
solution model, thermodynamic parameters can be obtained
the quadrupole splitting was found to accompany the spin
0
HL
using eq 1, in which ΔH denotes the difference in enthalpy
5
conversion. This large quadrupole splitting found for HS
between the LS and HS states and Γ the interaction energy
−
1
(
ΔE > 2 mm s ) indicates that a strong electric-field gradient
Q
concerning magnetic cooperativity between metal centers.
5
permeates the iron atom, suggesting a distorted HS octahedral
complex. It is also worth noting that the subspectral areas
change significantly from room temperature down to 22 K,
clearly indicating a HS−LS crossover under decreasing
0
HL
0
i1 − γ
y
ΔH + Γ(1 − 2γHS)
ΔS
j
z
j
HS z
HL
lnj
z =
−
j
j
z
z
γ
RT
R
HS
{
(1)
k
E
DOI: 10.1021/acs.inorgchem.8b02278
Inorg. Chem. XXXX, XXX, XXX−XXX

Inorganic Chemistry
Article
Table 1. Hyperfine Parameters for 2 and 1 at 22 and 300 K
HS (mm s−1)
LS (mm s−1
)
T (K)
AHS/Atot (%)
δ
ΔEQ
Γ
δ
ΔEQ
Γ
[
[
Fe(M) ]2+
300
64.5
9.8
1.00
1.30
2.04
2.37
0.51
0.84
0.31
0.47
0.32
0.40
0.34
0.43
0.53
0.55
0.41
0.45
0.27
0.39
3
2
2
Fe(Z) ]2+
300
3
2
2
Table 2. Crystallographic Data for Complexes
[
Fe(Z) ](ClO )
[Fe(M) ](ClO )
[Fe(M) ](ClO )
3 4
3
4
2
3
4
2
2
chemical formula
mol wt (g mol−1)
temperature (K)
wavelength (Å)
cryst syst
FeC H N O Cl (H O)
698.04
150.0
Mo Kα, 0.71073
monoclinic
FeC H N O Cl
693.22
150.0
Cu Kα, 1.54178
monoclinic
FeC H N O Cl
21 18 12 8 2
2
1
18 12
8
2
2
0.268
21 18 12
8
2
693.22
296.0
Cu Kα, 1.54178
monoclinic
space group
a (Å)
b (Å)
c (Å)
β (deg)
V (Å3)
P2 /c
P2 /c
P2 /c
1
1
1
15.9269(18)
9.8374(11)
17.1276(19)
101.289(2)
2631.6(5)
4
17.4833(6)
12.8960(5)
12.0829(4)
90.716(2)
2724.05(17)
4
17.710(2)
13.1791(16)
12.1886(12)
90.569(6)
2844.7(5)
4
Z
ρcalcd (g cm−3)
F(000)
1.762
1419
1.690
1408
1.619
1408
μ (mm−1)
θ range (deg)
reflns coll/indep
Rint
0.85
6.89
6.59
2.4−27.8
37738/6279
0.027
30.1−98.1
38366/4976
0.061
8.4−134.3
33446/5084
0.052
data/restraints/param
6279/281/578
0.051, 0.115
0.41, −0.54
1.19
4967/688/503
0.078, 0.218
0.54, −0.64
1.05
5084/509/627
0.078, 0.241
0.58, −0.45
1.03
2
2
2
R [F > 2σ(F )], wR (F )
1
2
diff peak, hole (e Å−3)
2
GOF on F (S)
CCDC
1842589
1842593
1842592
Using the fitted molar fraction curve (Figure S6), a linear fit
transition in this sample would be hypothetically observed at
temperatures not lower than 600 K.
2
of eq 1 with R = 0.9996 in the 200−300 K range results in
0
−1
−1
0
ΔH = 853 ± 3 cm (10.21 ± 0.03 kJ mol ), ΔS = 3.35
Crystal Structures. Single crystals were obtained for both
HL
HL
−
1
−1
−1
−1
±
0.01 cm
K
(40.0 ± 0.1 J K mol ), and Γ = 0 for the
complexes that crystallized in the monoclinic P2 /c space
1
group (Table 2), and data analysis confirms that only the
meridional isomer is present in the samples. The molecular
packing in both samples is very similar, revealing that there is
no direct hydrogen bonding between cationic units, as can be
seen from a comparison of Figures 3 and 4. However,
perchlorate anions form hydrogen bonds with cations through
the NH hydrogen atoms from the imidazole units, giving rise
to a network in the crystal. Similar results were observed in the
crystal structure of the ruthenium complex with ligand M,
spin conversion in 2. The absence of an interaction term for
the process (Γ = 0) means that it corresponds to a true spin-
state equilibrium. Moreover, the enthalpy change is close to
−
1
the usual values (800−1600 cm ) for iron(II) SCO systems.
−
1
−1
The entropy change is comparable to the typical 4−6 cm
K
104
range observed for other iron(II) systems.
Because the
magnetic contribution to the entropy change associated only
with spin can be estimated as ΔS = R[ln(2S + 1)HS − ln(2S
spin
−
1
−1 61
+
1) ] ≈ 1.1 cm K , the low value found for 2 may
105
LS
recently reported. The absence of direct hydrogen bonding
indicate that the LS−HS conversion is in this case not
accompanied by important changes in the crystal packing of
the system. This interpretation is corroborated by the
crystallographic data discussed in the next section. All of
between the complexes may be related to the broad spin
transition observed for the sample because the cooperative
effects due to intermolecular interactions would be less
4
pronounced in this case. Although the unit cell dimensions
these observations can lead us to the conclusion that, for 2, the
are not readily comparable, the cell volumes are very close to
one another, yielding similar values for the calculated densities.
As a consequence of the method used to grow single crystals,
the compositions of powdered samples and crystals differ in
0
ΔH value obtained from eq 1 is a good estimate for the
HL
5
1
energy difference between the HS excited state and the LS
ground state (ΔE ) of the system.
0
HL
Unfortunately, the lack of spin conversion for 1 in the
temperature range used in the experiments prevents calculation
of the thermodynamic data for this complex. However, if we
consider that γHS for this sample at 300 K is close to the values
observed for 2 below 20 K, one can speculate that the spin
the amount of H O content. Because polymorphism and
2
solvatomorphism can play a very important role on the solid-
106,107
state SCO properties,
we have performed powder XRD
experiments for the samples used to obtain magnetic
susceptibilities and Mo
̈
ssbauer spectra. A comparison of the
F
DOI: 10.1021/acs.inorgchem.8b02278
Inorg. Chem. XXXX, XXX, XXX−XXX

Inorganic Chemistry
Article
Figure 3. Crystal structure of 2 at two different temperatures measured from the same single crystal. Unit cell packing along the b axis at (a) 150 K
and (b) 296 K. ORTEP diagram at 50% probability with atom numbering in which hydrogen atoms and perchlorates were omitted for clarity at (c)
1
50 K and (d) 296 K. (e) Overlay of the cations obtained at 150 and 296 K.
Figure 4. Crystal structure of 1 at 150 K. (a) ORTEP diagram at 50% probability with atom numbering in which hydrogen atoms were omitted for
clarity and (b) unit cell packing along the b axis.
experimental and simulated diffractograms using single-crystal
structures obtained at room temperature (Figure S8) reveals
that powdered samples and crystals are structurally very
similar. Although differences between the SCO properties of
the powders and crystals cannot be ruled out, the observed
similarity is strong evidence that the assessments obtained
from the present crystal structures can be extended to the
powdered samples used in magnetic measurements.
As expected, the complexes have distorted octahedral
coordination spheres because of the bidentate nature of the
ligands. The meridional configuration is the only one observed
requiring that two imidazole units are trans to each other while
the third is trans to a pyrimidine unit. In both cases, this results
in different bond distances for similar units.
For sample 2, data collection was performed at 150 and 296
K using the same single crystal. As expected, heating of the
sample yields an expansion of the cell (4.4% in volume), but
G
DOI: 10.1021/acs.inorgchem.8b02278
Inorg. Chem. XXXX, XXX, XXX−XXX

Inorganic Chemistry
Article
2
+
2+
Table 3. Characteristic Geometrical Parameters (Mean Values) of the [Fe(M) ] and [Fe(Z) ] Complexes Obtained from
3
3
1
5
Single-Crystal XRD and at the DFT/PBE0 Level in the Singlet ( LS) and Quintet ( HS) States
Fe(M) ]2
+
[Fe(Z)3]2+
[
3
XRD
DFT
XRD
DFT
1
50 K
296 K
1LS
5HS
150 K
1LS
5HS
d(Fe−Nimidazole) (Å)
d(Fe−Ndiazine) (Å)
bite angle (deg)
1.98
2.00
81.2
93.2
2.12
2.15
78.2
94.6
1.99
2.03
81.2
93.1
2.16
2.28
76.3
95.0
1.96
1.96
81.5
92.8
1.98
2.02
80.9
93.2
2.15
2.27
75.6
95.2
interligand angle (deg)
apart from that, no other important differences are observed
when the two unit cells are compared (Figure 3a,b). The
atomic positions remain almost the same, but the thermal
ellipsoids increase as expected for higher temperatures (Figure
stronger electron-donor character of imidazole. The same
trend was observed in copper(II) complexes of the same
64
ligands. Moreover, the mean values of the Fe−N bond
lengths are comparable with those reported for other related
2
+
2+
3
c,d). Considering that the two selected temperatures
LS iron(II) complexes like [Fe(bpz)3] and [Fe(bpym) ]
3
1
09,110
encompass the observed spin transition for 2, the subtle
changes in the atomic positions are in very good agreement
with the low value for the entropy change obtained from the
magnetic data. It is important to note that the increase in the
temperature does not change the overall hydrogen-bonding
pattern in the unit cell. Although the intermolecular network
formed by the hydrogen bonding between perchlorate anions
and cations may play an important role in defining the overall
temperature-dependent magnetic behavior, these results
suggest that changes in the packing of the sample as a function
of the temperature are not the dominant driving force for the
contrasting magnetic behavior when the two isomers are
compared. Another important fact is that the shortest Fe−Fe
distances are too long to be considered as a mechanism for
magnetic interactions. At 150 K, the distance is 7.247 Å, and it
increases to 7.347 Å at room temperature. This is also in
agreement with the zero value predicted for Γ using eq 1, as
discussed in the previous section. A direct Fe−Fe interaction
mechanism can also be ruled out for sample 1 because the
metal−metal distance is even longer for this sample (8.564 Å)
probably induced by the higher β angle in comparison with 2.
The most important difference between structure 2 at low
and high temperatures is the change in the iron(II)
coordination sphere. With increasing temperature, the bond
distances and angles change, and this can fully account for the
magnetic behavior of this sample. Similar distortions were
observed for other SCO complexes with N-heterocycle
(with 2,2′-bipyrazine and 2,2′-bipyrimidine).
crystallographic data at 150 K and the Mo
Thus, the
ssbauer spectrum of
̈
this sample at low temperatures unambiguously correspond to
a LS state for this complex. The data at 296 K indicate
elongation of all Fe−N bond distances, especially for one of
1
the ligands, which is consistent with a transition from LS to
5
HS iron(II). The more distorted structure at 296 K might be
5
caused by a Jahn−Teller effect in the HS state (corresponding
to a degenerate T in a pure octahedron). The [Fe(Z) ]
5
2+
2
g
3
complex has a coordination sphere very similar to that of
2+
[
Fe(M) ] at 150 K, indicative of the LS nature of this sample,
3
in agreement with the magnetic and Mossbauer results. We
̈
2+
also find that [Fe(Z)3] has slightly shorter Fe−N bond
distances, conforming with the stronger LF exerted by this
ligand.
Spin-State Energetics. To obtain a reliable computational
0
HL
estimate of the adiabatic energy difference ΔE for both
complexes, CASPT2 calculations were performed on structures
for the different spin states obtained from DFT (PBE0
functional). As can be seen from Table 3, the bond distances
and angles of the structures in silico (Tables S6−S8) properly
reproduce the features observed in the XRD results. Zero-point
vibrational energies were also taken from these DFT
calculations. However, as is well-known, obtaining accurate
spin-state energetics from DFT is problematic because the
results are strongly dependent on the applied functional.
Therefore, we rather revert to traditional wave-function
methods for calculation of the electronic energy differences
between the different spin states. Results obtained with
CASPT2 and CASPT2/CC with different basis set combina-
tions (as described in the Computations Details section) are
collected in Table S9. Our “best” results, obtained with
CASPT2/CC in combination with CBS(Q:5) (i.e., CBS
2
7
chelating ligands. Parts c and d of Figure 3 show the FeN
6
2
+
distorted octahedral arrangement of the cationic [Fe(M)3]
species at 150 and 296 K. An overlay of the two structures is
also depicted in Figure 3e. This figure reveals that upon
heating the Fe−N bond distances increase accompanied by a
distortion in the bond angles, with one of the ligands showing
more pronounced distortions.
extrapolation based on the results obtained with iron aug-cc-
If we analyze the geometry of the FeN octahedron on the
6
0
pwCV5Z-DK versus aug-cc-pwCVQZ-DK), are ΔE = 500
basis of the parameters suggested by Guionneau and co-
HL
1
08
−1
2+
−1
2+
cm for [Fe(M) ] and 2012 cm for [Fe(Z) ] . As noted
workers, we obtain the values d
= 2.135 Å and Σ = 82°
3
3
Fe−L
62
5
2+
in our previous paper, the CBS(Q:5) extrapolation might be
slightly biased (by a few hundred wavenumbers) toward the LS
state compared to “full” CBS (that is, including also
extrapolation of the ligand basis sets). However, this error is
compensated for by the inherent bias (by about the same
for the HS [Fe(M) ] . SCO leads to structural modifications,
3
yielding Δr = 0.145 Å and ΔΣ = 19° (24%). Such variations
are in very good agreement with other SCO iron(II)
complexes for which temperature-dependent crystallographic
4
,108
studies were performed.
6
2
Geometric changes can also be clearly seen by data in Table
, showing select mean values for the bond distances and
amount) toward HS states of valence-only CASPT2.
3
Therefore, we believe that the CBS(Q:5) CASPT2/CC results
should give a highly accurate estimate of this property in the
angles. At 150 K, the different Fe−N bond distances are very
0
0
−1
2+
similar, but, remarkably, Fe−N is a little bit longer than
gas phase. ΔE ≃ ΔH = 500 cm for [Fe(M) ] closely
diazine
HL
HL
3
0
−1
Fe−Nimidazole, an asymmetry that reflects the stronger bond and
corresponds to the crystal ΔH = 853 cm , confirming our
HL
H
DOI: 10.1021/acs.inorgchem.8b02278
Inorg. Chem. XXXX, XXX, XXX−XXX

Inorganic Chemistry
Article
observation that the LS−HS conversion is not accompanied by
show distinctive features, which might go beyond the structural
differences between them and reflect the uneven presence of
̈
different spin states observed in the magnetic and Mossbauer
experiments at room temperature. The spectra of both
complexes are dominated by intense absorptions in the 300
nm region, originating from intraligand (IL) transitions. The
change of ligand from Z to M blue-shifts the main peak of the
IL band from 320 to 280 nm.
important changes in the crystal packing of the system.
0
HL
Assuming that the difference in the gas phase ΔH may be
extrapolated to the solid state, our theoretical estimate of 1 for
0
−1
ΔH is 2365 cm .
HL
UV/Vis Spectroscopy. Ligands Z and M are constitutional
isomers (C H N ) only differing in the relative position of one
7
6
4
diazinic nitrogen atom. Because both structures are fully
aromatic and because of the π system of the diazine cycle, we
expect a rich spectroscopic behavior like in other coordination
Furthermore, complex 1 has a less intense and broad band in
the 400−600 nm region of the spectrum, which is completely
missing for 2. On the other hand, the spectrum of the M
complex has a very weak absorption as a shoulder of the IL
band at 440 nm. The assignment of these low-energy bands is
done by computational means, taking into account the
1
11−118
compounds.
Figure 5 shows the UV/vis absorption spectra of 2 and 1
recorded in MeOH at room temperature. The two isomers
1
5
presence of both LS and HS species and evaluating the
MLCT transitions from the occupied Fe(3d) orbitals to the Z/
M(π*), which we suspect to be behind those weaker bands in
the spectra.
We have explored the vertical spin-allowed MLCT
transitions by means of the CASPT2 method. On the basis
of the magnetic and Mossbauer data, we describe the first
̈
2
+
5
quintet MLCT manifold of [Fe(M) ] because the HS
3
species should be the dominant species under the conditions of
the UV/vis spectrum, whereas we calculate the MLCTs from
1
5
2+
1
both the LS and HS states of [Fe(Z) ] because the LS
3
5
state should be dominant but the presence of HS cannot be
discarded. Figure 6 compares the transition energies and
oscillator strengths of the MLCTs calculated for both
complexes with their absorption spectra in the 300−700 nm
5
region. The manifold of MLCTs is formed by three states,
Figure 5. UV/vis absorption spectra of 2 (red) and 1 (black) in
MeOH.
where one electron is transferred to each of the three ligands,
3
2
1
i.e., a [t2g e π* ] electronic configuration. The difference in
g
5
2+
5
1
2+
Figure 6. MLCT transitions from the HS state of [Fe(M) ] (red) and the HS and LS states of [Fe(Z) ] (gray and black) compared to the
3
3
2
+
experimental data (Figure 5). The absorbance of the spectra is adjusted based on the maximum of the [Fe(Z)3] band at 510 nm and its most
intense transition at 537 nm.
I
DOI: 10.1021/acs.inorgchem.8b02278
Inorg. Chem. XXXX, XXX, XXX−XXX

Inorganic Chemistry
Article
energy between these states is 300 cm− or less for both
1
measured their pKaH in H O. Experimental results reveal only
2
complexes, a consequence of the small distortion from the
one protonation equilibrium in the 0−11 pH range, and values
ideal O coordination environment.
are in agreement with the lower pK expected for a 1H-
h
aH
5
The MLCTs are only weakly affected by the PCM(MeOH)
imidazol-2-yl moiety bonded to an electron-deficient group,
−
1
solvent model, being red-shifted by ∼1000 cm for [Fe-
such as a diazine, in comparison to imidazole itself with pKaH
=
=
2
+
2+
122
(
M) ] and twice as much for [Fe(Z) ] . Such a weak
6.99.
The experimental value for free ligand Z is pKaH
3
3
sensitivity to the polar solvent was observed for other iron(II)
complexes with similar chelating ligands as well and justifies
4.51, a lower value than that for isomer M with pKaH = 5.21.
The larger pKaH for M would imply a stronger σ donation
toward coordination with iron(II), resulting in a stronger field
for ligand M, in high contrast with the experimental and
theoretical results previously described. These results confirm
that σ donation from imidazole is not the major electronic
effect defining the LF strength. Unfortunately, diazine
protonation was not observed, but a comparison of these
ligands using NMR spectroscopy shows a strong communica-
tion between the two moieties of the ligands, where the
electron density of the imidazole is delocalized toward the
diazine (discussion in the Supporting Information). CASSCF
calculations revealed that π-back-donation is negligible in the
ground and LF excited states for both complexes, suggesting
that their LF strength is defined by a subtle balance between σ
donation from both units.
119
the use of a relatively simple solvent model without explicit
1
20
solvent molecules.
The manifold of MLCT states of [Fe(Z) ] is broader,
1
2+
3
with nine possible states with a [t2g5e π* ] electronic
configuration. These transitions appear around 5500 cm
0
1
g
−
1
5
lower in energy than the MLCTs and cover a range of
−
1
energies of 3552 cm (Tables S10−S13). Because both the
1
5
LS and HS states of the Z complex and their respective
MLCT states have similar dipole moments, the inclusion of
1
PCM(MeOH) induces a red shift to the MLCTs similar in
5
magnitude to the MLCTs of the same complex. The transition
1
energies and oscillator strengths of the MLCTs closely agree
2+
with the band at 510 nm of the [Fe(Z) ] spectrum. The
3
5
MLCTs are higher in energy and less intense. Nevertheless, a
In conclusion, the slightly stronger LF of Z with respect to
very weak shoulder can indeed be observed in the experimental
1
5
M is sufficient to stabilize the LS state and change the spin of
spectrum at around 400 nm, close to the calculated MLCT
the complex, causing their strikingly different room temper-
ature macroscopic properties. This change in the σ-donor
character between the complexes of M and Z provides a
potential mechanism for tuning the electron density on the
metal center.
band positions. This shoulder should therefore be assigned to
the presence of a small but significant amount of HS complexes
at room temperature, conforming with the paramagnetic
1
behavior observed in its H NMR spectrum in solution and
also with the magnetic measurements in the solid state. The
current approach cannot describe the large broadening of the
CONCLUSIONS
1
■
MLCT band because it depends on dynamical conformational
121
We have shown that the different magnetic and spectroscopic
properties of two iron(II) complexes obtained from two
constitutional isomers of the 2-(1H-imidazol-2-yl)diazine
ligand originate from the subtle differences in their electronic
structures, in the absence of strong cooperativity. A
combination of an electron-rich σ-donor imidazole moiety
with an electron-poorer diazine moiety can be used to fine-
tune the SCO in 2 with a spin transition at 241 K, whereas 1
shows a stable magnetic behavior in the 0−300 K range of
temperatures.
distortions of the complex.
5
Only two MLCT intense transitions at 364 and 366 nm
contribute to the absorption spectrum of [Fe(M)3]2+ in very
good agreement with the observed shoulder at 360 nm in the
experimental spectrum (Figure 6).
LF. Given the distortion of [Fe(M)3]2 and [Fe(Z)3]2+ from
an ideal octahedral coordination, we can qualitatively evaluate
their LF parameters from the LF expression for an octahedral
+
6
d complex. We have performed CASPT2 calculations on spin-
allowed and spin-forbidden metal-centered transitions from the
The interpretation of the magnetic susceptibility data is
straightforward and was corroborated by temperature-depend-
1
3
5
1
LS and HS states of both complexes to cover the A
→
1
g
,1
1
3,1
5
5
T , A
→
T , and T → E transitions in O
2g 2g g h
1
g
1g
̈
ent Mossbauer spectra that reveal an equilibrium between the
quintet and singlet species in the SCO system. Thermody-
symmetry and averaged the energies of electronic states
belonging to the same electronic O term (Table S14). For the
0
h
namic treatment of the magnetic data gives ΔH = 10.21 ±
−
1
0
−1
−1
LS complexes, the Racah parameters B and C may be obtained
0
.03 kJ mol and ΔS = 40.0 ± 0.1 J K mol for the spin
1
3
from the energy differences E( T ) − E( T ) = 2C and
1g
1g
transition between these states.
1
1
E( T ) − E( T ) = 16B, respectively. These repulsion
2
g
1g
The temperature-dependent single-crystal study of 2
confirmed that the changes in the unit cell packing as a
consequence of the temperature are negligible, explaining the
parameters are very similar for both complexes, i.e., B ≃ 500
−
1
−1
cm and C ≃ 4000 cm . The LF strength of the LS
1
1
0
complexes can be obtained from E( T ) − E( A ) = 10 Dq −
1g
1g
low value for the calculated ΔS and pointing to a change in
−
1
2+
C. We obtain 10 Dq = 19075 cm for [Fe(Z) ] and 10 Dq =
the coordination sphere around the iron(II) center. It revealed
an expansion of the pseudooctahedral coordination sphere on
the LS−HS transition in agreement with other similar SCO
complexes, which was corroborated by DFT geometry
optimizations, showing that the metal−ligand bonds are
enlarged by ∼0.1 Å as a function of the temperature and
spin change.
3
−
1
2+
1
[
8569 cm for [Fe(M) ] . The larger 10 Dq value of
Fe(Z) ] , by 506 cm , confirms the larger stabilization of
3
2
+
−1
3
the singlet state by the Z ligand.
6
For a HS d complex, the value of 10 Dq is directly related
5
5
5
to the T → E transition energy [10 Dq = E( E ) −
2
g
g
g
5
−1
2+
E( T )] and diminishes to 10243 cm for [Fe(Z) ] and
2
g
3
−
1
2+
9
936 cm for [Fe(M) ] . As expected, the stronger field of Z
CASPT2 calculations of the spin-state energetics in the gas
phase, making use of correlation-consistent basis sets
3
−
1
is maintained, but by only 307 cm in the HS state.
In order to gain insight into the role of σ donation of
imidazole and diazine units in defining ligands’ strengths, we
2
+
extrapolated to CBS, confirm the SCO behavior of [Fe(M) ]
3
2
+
0
but not [Fe(Z) ] , with ΔH values of 5.98 and 24.05 kJ
3
HL
J
DOI: 10.1021/acs.inorgchem.8b02278
Inorg. Chem. XXXX, XXX, XXX−XXX

Inorganic Chemistry
Article
−
1
mol , respectively. Using CASPT2 calculations, we were also
able to resolve the UV/vis spectra of the complexes in MeOH
and to verify the picture drawn from the spin-state energetics.
Complex 1 has a broad and relatively intense absorption band
402627/2012-1, and 465452/2014-0), INCT-INOMAT, Sao
Paulo Research Foundation (Grants 2013/22127-3 and 2014/
̃
50906-9), and Flemish Science Foundation and Coordenaca
de Aperfeicoamento de Pessoal de Nıvel Superior - Brasil
̧ o
̃
̧
́
1
at 510 nm formed by spin-allowed transitions from the LS
state to the lowest MLCT manifold that is almost missing in
the spectrum of 2 (it appears as a very weak band in
concentrated solutions). Conversely, the latter has a weak
absorption at 360 nm, appearing as a shoulder of the intense IL
band, that rises from spin-allowed transitions from the HS to
the lowest MLCT manifold. Thus, the UV/vis spectra are
composed of both LS and HS species in different proportions
in 1 and 2 at room temperature. The LF parameters obtained
from CASPT2 confirm the stronger LF of Z with respect to M,
with a 10 Dq value that is larger by 300−500 cm . Hence, the
change of the position of the diazinic nitrogen atom does not
produce large structural distortions on the complexes but
induces a weak change of the LF that explains the appearance
of SCO in only one of their complexes (2).
(CAPES) - Finance Code 001−FWO for funding of the
research. The computational resources and services used in this
work were provided by the Centro Nacional de Processamento
1
de Alto Desempenho em Sao Paulo (CENAPAD-SP) and the
̃
VSC (Flemish Supercomputer Center), funded by the
Hercules Foundation and the Flemish Government, Depart-
ment EWI.
5
5
1
5
REFERENCES
■
(
1) Kahn, O.; Martinez, C. J. Spin-Transition Polymers: From
Molecular Materials Toward Memory Devices. Science 1998, 279, 44−
−
1
48.
(2) Gutlich, P., Goodwin, H., Eds. Spin Crossover in Transition Metal
̈
Compounds; Topics in Current Chemistry; Springer: Berlin, 2004;
Vol. 233−235.
(
3) Halcrow, M. A. Spin-Crossover Materials: Properties and
Applications; John Wiley & Sons, 2013.
4) Craig, G. A.; Roubeau, O.; Aromí, G. Spin State Switching in
,6-Bis(Pyrazol-3-Yl)Pyridine (3-Bpp) Based Fe(II) Complexes.
ASSOCIATED CONTENT
Supporting Information
The Supporting Information is available free of charge on the
ACS Publications website at DOI: 10.1021/acs.inorg-
chem.8b02278.
■
(
2
*
S
Coord. Chem. Rev. 2014, 269, 13−31.
(5) Mebs, S.; Braun, B.; Kositzki, R.; Limberg, C.; Haumann, M.
II
Abrupt versus Gradual Spin-Crossover in Fe (phen)
(NCS)
and
2
2
III
1H and 13C NMR spectra and the corresponding
assignments of M and Z in DMSO-d , experimental
and simulated powder XRD data for both complexes,
Fe (dedtc)
copy and Quantum-Chemical Calculations. Inorg. Chem. 2015, 54,
1606−11624.
6) Harding, D. J.; Harding, P.; Phonsri, W. Spin Crossover in
3
Compared by X-ray Absorption and Emission Spectros-
6
1
(
2
+
2+
atomic coordinates of [Fe(M)3] and [Fe(Z)3] in the
singlet, triplet, and quintet lowest states optimized in a
vacuum with DFT/PBE0, estimation of the core
Iron(III) Complexes. Coord. Chem. Rev. 2016, 313, 38−61.
(7) Feltham, H. L. C.; Barltrop, A. S.; Brooker, S. Spin Crossover in
Iron(II) Complexes of 3,4,5-Tri-Substituted-1,2,4-Triazole (Rdpt),
3,5-Di-Substituted-1,2,4-Triazolate (Dpt-), and Related Ligands.
Coord. Chem. Rev. 2017, 344, 26−53.
correlation effect for [Fe(M)3]2 and [Fe(Z)3] at the
CASPT2 level, energy, dipole moment, and oscillator
strength of the spin-allowed transitions to the low-lying
+
2+
(8) Senthil Kumar, K.; Ruben, M. Emerging Trends in Spin
2
+
Crossover (SCO) Based Functional Materials and Devices. Coord.
MLCT manifold from the quintet state of [Fe(M) ]
and the singlet and quintet states of [Fe(Z)3] in a
vacuum and MeOH (PDF)
3
2
+
Chem. Rev. 2017, 346, 176−205.
(9) Kipgen, L.; Bernien, M.; Ossinger, S.; Nickel, F.; Britton, A. J.;
Arruda, L. M.; Naggert, H.; Luo, C.; Lotze, C.; Ryll, H.; Radu, F.;
Schierle, E.; Weschke, E.; Tuczek, F.; Kuch, W. Evolution of
Cooperativity in the Spin Transition of an Iron(II) Complex on a
Graphite Surface. Nat. Commun. 2018, 9, 2984.
Accession Codes
CCDC 1842589 and 1842592−1842593 contain the supple-
mentary crystallographic data for this paper. These data can be
obtained free of charge via www.ccdc.cam.ac.uk/data_request/
cif, or by emailing data_request@ccdc.cam.ac.uk, or by
contacting The Cambridge Crystallographic Data Centre, 12
Union Road, Cambridge CB2 1EZ, UK; fax: +44 1223 336033.
(10) Takahashi, K.; Okai, M.; Mochida, T.; Sakurai, T.; Ohta, H.;
Yamamoto, T.; Einaga, Y.; Shiota, Y.; Yoshizawa, K.; Konaka, H.;
Sasaki, A. Contribution of Coulomb Interactions to a Two-Step
Crystal Structure Phase Transformation Coupled with a Significant
Change in Spin Crossover Behavior for a Series of Charged FeII
Complexes from 2,6-Bis(2-Methylthiazol-4-Yl)Pyridine. Inorg. Chem.
2018, 57, 1277−1287.
AUTHOR INFORMATION
Corresponding Author
E-mail: formiga@unicamp.br.
ORCID
■
*
(
11) Scott, H. S.; Staniland, R. W.; Kruger, P. E. Spin Crossover in
Homoleptic Fe(II) Imidazolylimine Complexes. Coord. Chem. Rev.
018, 362, 24−43.
12) Tsukiashi, A.; Nakaya, M.; Kobayashi, F.; Ohtani, R.;
2
(
Alex Domingo: 0000-0002-9162-8076
Quan Manh Phung: 0000-0001-8205-5328
Kristine Pierloot: 0000-0002-3196-4940
Nakamura, M.; Harrowfield, J. M.; Kim, Y.; Hayami, S. Intermolecular
Interaction Tuning of Spin-Crossover Iron(III) Complexes with
Aromatic Counteranions. Inorg. Chem. 2018, 57, 2834−2842.
(
13) Takano, M.; Nasu, S.; Abe, T.; Yamamoto, K.; Endo, S.;
Andre Luiz Barboza Formiga: 0000-0001-6773-8083
́
Takeda, Y.; Goodenough, J. B. Pressure-induced high-spin to low-spin
Author Contributions
transition in CaFeO . Phys. Rev. Lett. 1991, 67, 3267−3270.
3
‡
Contributed equally to this work.
(14) Gutlich, P.; Ksenofontov, V.; Gaspar, A. B. Pressure effect
̈
studies on spin crossover systems. Coord. Chem. Rev. 2005, 249,
811−1829.
̈
15) Cambi, L.; Szego, L. Uber die magnetische Susceptibilitat der
̈ ̈
Notes
1
(
The authors declare no competing financial interest.
komplexen Verbindungen. Ber. Dtsch. Chem. Ges. B 1931, 64, 2591−
ACKNOWLEDGMENTS
The authors thank Conselho Nacional de Desenvolvimento
Cientıfi
■
2598.
(16) Baker, W. A.; Bobonich, H. M. Magnetic Properties of Some
High-Spin Complexes of Iron(II). Inorg. Chem. 1964, 3, 1184−1188.
́
co e Tecnolog
́
ico−CNPq (Grants 131848/2013-5,
K
DOI: 10.1021/acs.inorgchem.8b02278
Inorg. Chem. XXXX, XXX, XXX−XXX

Inorganic Chemistry
Article
(
17) Koenig, E.; Madeja, K. 5T2−1A1 Equilibriums in some
iron(II)-bis(1,10-phenanthroline) complexes. Inorg. Chem. 1967, 6,
8−55.
18) McGravey, J. J.; Lawthers, I. Photochemically-induced
Prussian Blue Materials Manifested in Molecular Complexes:
Observation of Cyanide Linkage Isomerism and Spin-Crossover
Behavior in Pentanuclear Cyanide Clusters. J. Am. Chem. Soc. 2007,
129, 6104−6116.
4
(
1
5
11
perturbation of the A ⇌ T equilibrium in Fe complexes by
pulsed laser irradiation in the metal-to-ligand charge-transfer
absorption band. J. Chem. Soc., Chem. Commun. 1982, 906−907.
(36) Han, W.-K.; Li, Z.-H.; Zhu, W.; Li, T.; Li, Z.; Ren, X.; Gu, Z.-G.
Molecular Isomerism Induced Fe(II) Spin State Difference Based on
the Tautomerization of the 4(5)-Methylimidazole Group. Dalton
Trans. 2017, 46, 4218−4224.
(
19) Decurtins, S.; Gutlich, P.; Kohler, C. P.; Spiering, H.; Hauser,
̈
̈
A. Light-induced excited spin state trapping in a transition-metal
complex: The hexa-1-propyltetrazole-iron (II) tetrafluoroborate spin-
crossover system. Chem. Phys. Lett. 1984, 105, 1−4.
(37) Burrows, K. E.; McGrath, S. E.; Kulmaczewski, R.; Cespedes,
O.; Barrett, S. A.; Halcrow, M. A. Spin States of Homochiral and
2
+
Heterochiral Isomers of [Fe(PyBox)2] Derivatives. Chem. - Eur. J.
2017, 23, 9067−9075.
(
20) Hauser, A. Intersystem crossing in the [Fe(ptz) ](BF ) spin
6 4 2
crossover system (ptz = 1-propyltetrazole). J. Chem. Phys. 1991, 94,
741−2748.
21) Letard, J.-F. Photomagnetism of iron(II) spin crossover
complexesthe T(LIESST) approach. J. Mater. Chem. 2006, 16,
550−2559.
22) Halcrow, M. A. Trapping and manipulating excited spin states
of transition metal compounds. Chem. Soc. Rev. 2008, 37, 278−289.
23) Sousa, C.; de Graaf, C.; Rudavskyi, A.; Broer, R.; Tatchen, J.;
(38) Arcis-Castíllo, Z.; Zheng, S.; Siegler, M. A.; Roubeau, O.;
Bedoui, S.; Bonnet, S. Tuning the Transition Temperature and
Cooperativity of Bapbpy-Based Mononuclear Spin-Crossover Com-
pounds: Interplay between Molecular and Crystal Engineering. Chem.
- Eur. J. 2011, 17, 14826−14836.
2
(
́
2
(
(39) Gaspar, A. B.; Ksenofontov, V.; Seredyuk, M.; Gutlich, P.
̈
Multifunctionality in Spin Crossover Materials. Coord. Chem. Rev.
2005, 249, 2661−2676.
(40) Childs, B. J.; Craig, D. C.; Scudder, M. L.; Goodwin, H. A.
Structural and Electronic Studies of the Coordination of 6-(Thiazol-2-
Yl)-2,2′-Bipyridine and Related Systems to Fe(II), Co(II) and Ni(II).
Inorg. Chim. Acta 1998, 274, 32−41.
(41) Roux, C.; Zarembowitch, J.; Gallois, B.; Granier, T.; Claude, R.
Toward Ligand-Driven Light-Induced Spin Changing. Influence of
the Configuration of 4 Styrylpyridine (Stpy) on the Magnetic
Properties of FeII(Stpy)4(NCS)2 Complexes. Crystal Structures of
the Spin-Crossover Species Fe(Trans-Stpy)4(NCS)2 and of the High-
Spin Species Fe(Cis-Stpy)4(NCS)2. Inorg. Chem. 1994, 33, 2273−
2279.
(
Etinski, M.; Marian, C. M. Ultrafast Deactivation Mechanism of the
Excited Singlet in the Light-Induced Spin Crossover of [Fe(2,2′-
2+
bipyridine) ] . Chem. - Eur. J. 2013, 19, 17541−17551.
3
(
24) Aubo
̈
ck, G.; Chergui, M. Sub-50-fs photoinduced spin
2
+
crossover in [Fe(bpy) ] . Nat. Chem. 2015, 7, 629−633.
(
3
25) Halcrow, M. A. Structure:function relationships in molecular
spin-crossover complexes. Chem. Soc. Rev. 2011, 40, 4119−4142.
26) Cirera, J.; Ruiz, E. Theoretical Modeling of the Ligand-Tuning
(
Effect over the Transition Temperature in Four-Coordinated FeII
Molecules. Inorg. Chem. 2016, 55, 1657−1663. (PMID: 26824687).
(
27) Hauser, A.; Enachescu, C.; Daku, M. L.; Vargas, A.; Amstutz, N.
Low-temperature lifetimes of metastable high-spin states in spin-
crossover and in low-spin iron(II) compounds: The rule and
exceptions to the rule. Coord. Chem. Rev. 2006, 250, 1642−1652.
(42) Sugahara, A.; Enomoto, M.; Kojima, N. Isomerization Effect of
Counter Anion on the Spin Crossover Transition in [Fe(4-
NH Trz) ](CH C H SO ) ·nH O. J. Phys.: Conf. Ser. 2010, 217,
2
3
3
6
4
3
2
2
(
28) Boilleau, C.; Suaud, N.; Guiher
́
y, N. Ab initio study of the
012128.
influence of structural parameters on the potential energy surfaces of
spin-crossover Fe(II) model compounds. J. Chem. Phys. 2012, 137,
(43) Baker, A.; Goodwin, H. Iron(II) and Nickel(II) Complexes of
2,6-Di(Thiazol-4-Yl)Pyridine and Related Ligands: Magnetic, Spectral
and Structural Studies. Aust. J. Chem. 1986, 39, 209.
(44) Baker, A.; Singh, P.; Vignevich, V. Iron(II) and Nickel(II)
Complexes of 2,6-Di(Thiazol-2-Yl)Pyridine and Related Ligands.
Aust. J. Chem. 1991, 44, 1041.
(45) Kershaw Cook, L. J.; Mohammed, R.; Sherborne, G.; Roberts,
T. D.; Alvarez, S.; Halcrow, M. A. Spin State Behavior of Iron(II)/
Dipyrazolylpyridine Complexes. New Insights from Crystallographic
and Solution Measurements. Coord. Chem. Rev. 2015, 289−290, 2−
12.
2
(
24304.
29) Lathion, T.; Guen
́
ee
́
, L.; Besnard, C.; Bousseksou, A.; Piguet, C.
Deciphering the Influence of Meridional vs Facial Isomers in Spin
Crossover Complexes. Chem. - Eur. J. 2018, DOI: 10.1002/
chem.201804161.
(
30) Holzhacker, C.; Calhorda, M. J.; Gil, A.; Carvalho, M. D.;
Ferreira, L. P.; Stoger, B.; Mereiter, K.; Weil, M.; Muller, D.;
̈
̈
Weinberger, P.; Pittenauer, E.; Allmaier, G.; Kirchner, K. Six-
Coordinate High-Spin Iron(II) Complexes with Bidentate PN
Ligands Based on 2-Aminopyridine − New Fe(II) Spin Crossover
Systems. Dalton Trans. 2014, 43, 11152−11164.
(46) Pritchard, R.; Kilner, C. A.; Halcrow, M. A. Iron(II) Complexes
with a Terpyridine Embrace Packing Motif Show Remarkably
Consistent Cooperative Spin-Transitions. Chem. Commun. 2007,
577−579.
(
31) Furusho, D.; Nishi, K.; Hashibe, T.; Fujinami, T.; Hagiwara, H.;
Matsumoto, N.; Halcrow, M. A.; Iijima, S.; Sunatsuki, Y.; Kojima, M.
Assembly Structures and Spin Crossover Properties of Facial and
Meridional Isomers of Tris[Benzyl(2-Methylimidazol-4-
Ylmethylidene)Amine]Iron(II) Chloride Hexafluorophosphate.
Chem. Lett. 2011, 40, 72−74.
(47) Bushuev, M. B.; Krivopalov, V. P.; Nikolaenkova, E. B.;
Vinogradova, K. A.; Gatilov, Y. V. Hysteretic Spin Crossover in
Isomeric Iron(II) Complexes. Dalton Trans. 2018, 47, 9585−9591.
(48) Rodríguez-Jimenez, S.; Barltrop, A. S.; White, N. G.; Feltham,
́
(
32) Li, L.; Clarkson, G. J.; Evans, D. J.; Lees, M. R.; Turner, S. S.;
H. L. C.; Brooker, S. Solvent Polarity Predictably Tunes Spin
Crossover T1/2 in Isomeric Iron(II) Pyrimidine Triazoles. Inorg.
Chem. 2018, 57, 6266−6282.
Scott, P. Isomeric Fe(II) MOFs: From a Diamond-Framework Spin-
Crossover Material to a 2D Hard Magnet. Chem. Commun. 2011, 47,
1
(
2646.
33) Mekuimemba, C. D.; Conan, F.; Mota, A. J.; Palacios, M. A.;
(49) Real, J. A.; Gaspar, A. B.; Munoz, M. C. Thermal, pressure and
̃
light switchable spin-crossover materials. Dalton Trans. 2005, 2062−
Colacio, E.; Triki, S. On the Magnetic Coupling and Spin Crossover
2079.
II
2
Behavior in Complexes Containing the Head-to-Tail [Fe (μ-SCN) ]
(50) Gamez, P.; Costa, J. S.; Quesada, M.; Aromí, G. Iron Spin-
Crossover compounds: from fundamental studies to practical
applications. Dalton Trans. 2009, 7845−7853.
(51) Orellana, G.; Alvarez-Ibarra, C.; Quiroga, M. L. New binuclear
heterocyclic ligands sharing an α-diimine moiety and their ruthenium-
(II) tris-chelates. Bull. Soc. Chim. Belg. 1988, 97, 731−742.
(52) Das, A.; Scherer, T.; Maji, S.; Mondal, T. K.; Mobin, S. M.;
2
Bridging Unit: A Magnetostructural Experimental and Theoretical
Study. Inorg. Chem. 2018, 57, 2184−2192.
(
34) Sheu, C.-F.; Shih, C.-H.; Sugimoto, K.; Cheng, B.-M.; Takata,
M.; Wang, Y. A Long-Lived Photo-Induced Metastable State of
Linkage Isomerization Accompanied with a Spin Transition. Chem.
Commun. 2012, 48, 5715.
(
35) Shatruk, M.; Dragulescu-Andrasi, A.; Chambers, K. E.; Stoian,
Urbanos, F. A.; Jimenez-Aparicio, R.; Kaim, W.; Lahiri, G. K.
́
Reductive Approach to Mixed Valency (n = 1-) in the Pyrazine
S. A.; Bominaar, E. L.; Achim, C.; Dunbar, K. R. Properties of
L
DOI: 10.1021/acs.inorgchem.8b02278
Inorg. Chem. XXXX, XXX, XXX−XXX

Inorganic Chemistry
Article
2
−
n
2−
Ligand-Bridged [(acac) Ru(μ-L )Ru(acac) ] (L = 2,5-Pyrazine-
(69) Sheldrick, G. M. Crystal Structure Refinement with SHELXL.
Acta Crystallogr., Sect. C: Struct. Chem. 2015, 71, 3−8.
(70) Sheldrick, G. M. A Short History of SHELX. Acta Crystallogr.,
Sect. A: Found. Crystallogr. 2008, 64, 112−122.
(71) Perdew, J. P.; Wang, Y. Accurate and simple analytic
representation of the electron-gas correlation energy. Phys. Rev. B:
Condens. Matter Mater. Phys. 1992, 45, 13244−13249.
(72) Perdew, J. P.; Burke, K.; Ernzerhof, M. Generalized Gradient
Approximation Made Simple. Phys. Rev. Lett. 1996, 77, 3865−3868.
(73) Perdew, J. P.; Ernzerhof, M.; Burke, K. Rationale for mixing
exact exchange with density functional approximations. J. Chem. Phys.
1996, 105, 9982−9985.
2
2
dicarboxylate) through Experiment and Theory. Inorg. Chem. 2011,
0, 7040−7049.
53) Das, A.; Das, D.; Kundu, T.; Lahiri, G. K. Electronic structures
of ruthenium complexes encircling non-innocent ligand assembly. J.
Chem. Sci. 2012, 124, 1181−1189.
54) Santiago, M. O.; Filho, C. L. D.; Moreira, I. d. S.; Carlos, R. M.;
Queiroz, S. L.; Batista, A. A. Photochemical isomerization of trans- to
cis-[RuCl (dppb)(4,4′-X2−2,2′-bipy)] (X = −H, − NO , − Me, −
COOH, − SMe, − O=SMe, − Cl, − OMe) complexes. Polyhedron
5
(
(
2
2
2
(
003, 22, 3205−3211.
55) Bezencon, J.; Wittwer, M. B.; Cutting, B.; Smiesko, M.;
Wagner, B.; Kansy, M.; Ernst, B. pKa determination by 1H NMR
spectroscopy − An old methodology revisited. J. Pharm. Biomed. Anal.
̧
̌
(74) Weigend, F.; Ahlrichs, R. Balanced basis sets of split valence,
triple zeta valence and quadruple zeta valence quality for H to Rn:
Design and assessment of accuracy. Phys. Chem. Chem. Phys. 2005, 7,
3297−3305.
2
(
014, 93, 147−155.
56) Delower Hossain, M.; Ueno, R.; Haga, M.-a. Synthesis and
5
tuning of chemical properties by protonation/deprotonation of novel
dinuclear ruthenium complexes containing 2,6,2′,6′-tetra(4,5-dime-
thylbenzimidazol-2-yl)-4, 4′-bipyridine. Inorg. Chem. Commun. 2000,
(75) Pierloot, K.; Vancoillie, S. Relative energy of the high-( T )
2g
1
2+
2+
and low-( A ) spin states of [Fe(H O) ] , [Fe(NH ) ] , and
1
g
2
6
3 6
2
+
[Fe(bpy) ] : CASPT2 versus density functional theory. J. Chem.
3
3
(
, 35−38.
Phys. 2006, 125, 124303.
57) Díez, J.; Gimeno, J.; Lledos
́
, A.; Suar
́
ez, F. J.; Vicent, C.
(76) Pierloot, K.; Vancoillie, S. Relative Energy of the High-(T2g5)
and Low-(A1g1) Spin States of the Ferrous Complexes [Fe(L)-
(NHS4)]: CASPT2 versus Density Functional Theory. J. Chem. Phys.
2008, 128, 034104.
Imidazole Based Ruthenium(IV) Complexes as Highly Efficient
Bifunctional Catalysts for the Redox Isomerization of Allylic Alcohols
in Aqueous Medium: Water as Cooperating Ligand. ACS Catal. 2012,
2
(
, 2087−2099.
(77) Phung, Q. M.; Vancoillie, S.; Pierloot, K. A Multiconfigura-
tional Perturbation Theory and Density Functional Theory Study on
the Heterolytic Dissociation Enthalpy of First-Row Metallocenes. J.
Chem. Theory Comput. 2012, 8, 883−892.
58) Aakeroy, C. B.; Wijethunga, T. K.; Haj, M. A.; Desper, J.;
Moore, C. The structural landscape of heteroaryl-2-imidazoles:
competing halogen- and hydrogen-bond interactions. CrystEngComm
2
(
Stabilizing volatile liquid chemicals using co-crystallization. Chem.
Commun. 2015, 51, 2425−2428.
(
014, 16, 7218−7225.
59) Aakeroy, C. B.; Wijethunga, T. K.; Benton, J.; Desper, J.
(78) Ahlrichs, R.; Bar, M.; Haser, M.; Horn, H.; Kolmel, C.
̈ ̈ ̈
Electronic structure calculations on workstation computers: The
program system turbomole. Chem. Phys. Lett. 1989, 162, 165−169.
(79) Roos, B. O.; Taylor, P. R.; Siegbahn, P. E. M. A complete active
space SCF method (CASSCF) using a density matrix formulated
super-CI approach. Chem. Phys. 1980, 48, 157−173.
(80) Andersson, K.; Malmqvist, P.-Å.; Roos, B. O. Second-order
perturbation theory with a complete active space self-consistent field
reference function. J. Chem. Phys. 1992, 96, 1218−1226.
(81) Pierloot, K. Nondynamic Correlation effects in Transition
Metal Coordination Compounds. Computational Organometallic
Chemistry; Marcel Dekker, Inc.: New York, 2001; pp 123−158.
(82) Pierloot, K. The CASPT2 method in inorganic electronic
spectroscopy: from ionic transition metal to covalent actinide
complexes*. Mol. Phys. 2003, 101, 2083−2094.
60) Coronado, E.; Gimenez-Marques, M.; Martí-Gastaldo, C.;
́ ́
Espallargas, G. M.; Navarro-Moratalla, E.; Waerenborgh, J. C. Hybrid
Magnetic Superconductors Formed by TaS2 Layers and Spin
Crossover Complexes. Inorg. Chem. 2013, 52, 8451−8460.
(
61) Gutlich, P.; Gaspar, A. B.; Garcia, Y. Spin State Switching in
̈
Iron Coordination Compounds. Beilstein J. Org. Chem. 2013, 9, 342−
91.
62) Phung, Q. M.; Feldt, M.; Harvey, J. N.; Pierloot, K. Toward
3
(
Highly Accurate Spin State Energetics in First-Row Transition Metal
Complexes: A Combined CASPT2/CC Approach. J. Chem. Theory
Comput. 2018, 14, 2446−2455.
(
63) Voss, M. E.; Beer, C. M.; Mitchell, S. A.; Blomgren, P. A.;
(83) Andersson, K.; Roos, B. O. Excitation energies in the nickel
atom studied with the complete active space SCF method and
second-order perturbation theory. Chem. Phys. Lett. 1992, 191, 507−
514.
Zhichkin, P. E. A simple and convenient one-pot method for the
preparation of heteroaryl-2-imidazoles from nitriles. Tetrahedron
2
(
008, 64, 645−651.
64) Nakahata, D. H.; Ribeiro, M. A.; Corbi, P. P.; Machado, D.;
(84) Meyer, B.; Domingo, A.; Krah, T.; Robert, V. Charge Transfer
Processes: the Role of Optimized Molecular Orbitals. Dalton Trans.
2014, 43, 11209−11215.
Lancellotti, M.; Lustri, W. R.; da Costa Ferreira, A. M.; Formiga, A. L.
B. Synthesis, Characterization and Preliminary Antimicrobial Assays
of Copper(II) Complexes with 2-(Imidazole-2-Yl)Heteroaryl Ligands.
Inorg. Chim. Acta 2017, 458, 224−232.
(85) Domingo, A.; Angeli, C.; de Graaf, C.; Robert, V. Electronic
reorganization triggered by electron transfer: The intervalence charge
3
+
2+
(
65) Ratier de Arruda, E. G.; de Farias, M. A.; Venturinelli Jannuzzi,
transfer of a Fe /Fe bimetallic complex. J. Comput. Chem. 2015, 36,
861−869.
S. A.; de Almeida Gonsales, S.; Timm, R. A.; Sharma, S.; Zoppellaro,
G.; Kubota, L. T.; Knobel, M.; Barboza Formiga, A. L. Synthesis,
Structural and Magnetic Characterization of a Copper(II) Complex of
(86) Malmqvist, P.-Å.; Roos, B. O. The CASSCF state interaction
method. Chem. Phys. Lett. 1989, 155, 189−194.
2
,6-Di(1H-Imidazol-2-Yl)Pyridine and Its Application in Copper-
Mediated Polymerization Catalysis. Inorg. Chim. Acta 2017, 466,
56−463.
66) Stupka, G.; Gremaud, L.; Bernardinelli, G.; Williams, A. F.
(87) Roos, B. O.; Lindh, R.; Malmqvist, P.-Å.; Veryazov, V.;
Widmark, P.-O. Main Group Atoms and Dimers Studied with a New
Relativistic ANO Basis Set. J. Phys. Chem. A 2004, 108, 2851−2858.
(88) Roos, B. O.; Lindh, R.; Malmqvist, P.-Å.; Veryazov, V.;
Widmark, P.-O. New Relativistic ANO Basis Sets for Transition Metal
Atoms. J. Phys. Chem. A 2005, 109, 6575−6579.
4
(
Redox State Switching of Transition Metals by Deprotonation of the
Tridentate Ligand 2,6-Bis(Imidazol-2-Yl)Pyridine. Dalton Trans.
2
(
004, 407−412.
67) Bain, G. A.; Berry, J. F. Diamagnetic Corrections and Pascal’s
Constants. J. Chem. Educ. 2008, 85, 532.
68) Dolomanov, O. V.; Bourhis, L. J.; Gildea, R. J.; Howard, J. a. K.;
(89) Halkier, A.; Helgaker, T.; Jørgensen, P.; Klopper, W.; Olsen, J.
Basis-Set Convergence of the Energy in Molecular Hartree−Fock
Calculations. Chem. Phys. Lett. 1999, 302, 437−446.
(90) Helgaker, T.; Klopper, W.; Koch, H.; Noga, J. Basis-Set
Convergence of Correlated Calculations on Water. J. Chem. Phys.
1997, 106, 9639−9646.
(
Puschmann, H. OLEX2: A Complete Structure Solution, Refinement
and Analysis Program. J. Appl. Crystallogr. 2009, 42, 339−341.
M
DOI: 10.1021/acs.inorgchem.8b02278
Inorg. Chem. XXXX, XXX, XXX−XXX

Inorganic Chemistry
Article
(
91) Pierloot, K.; Phung, Q. M.; Domingo, A. Spin State Energetics
Complex [Fe(Bpz)3](ClO4)2·H2O (Bpz = 2,2′-Bipyrazine). Inorg.
Chim. Acta 2004, 357, 2609−2614.
in First-Row Transition Metal Complexes: Contribution of (3s3p)
Correlation and Its Description by Second-Order Perturbation
Theory. J. Chem. Theory Comput. 2017, 13, 537−553.
(111) Glockle, M.; Katz, N. E.; Ketterle, M.; Kaim, W. Qualitatively
̈
and quantitatively different solvatochromism of the MLCT and
MMCT absorption bands of centrosymmetric acceptor-bridged
diiron(II,II) and diiron(III,II) cyanide complexes. Inorg. Chim. Acta
2002, 336, 55−60.
(112) Winnischofer, H.; Formiga, A. L. B.; Nakamura, M.; Toma, H.
E.; Araki, K.; Nogueira, A. F. Conduction and Photoelectrochemical
Properties of Monomeric and Electropolymerized Tetraruthenated
Porphyrin Films. Photochem. Photobiol. Sci. 2005, 4, 359.
(113) Eberlin, M. N.; Tomazela, D. M.; Araki, K.; Alexiou, A. D. P.;
Formiga, A. L. B.; Toma, H. E.; Nikolaou, S. Electrospray Ionization
(
92) Reiher, M.; Wolf, A. Exact decoupling of the Dirac
Hamiltonian. II. The generalized Douglas−Kroll−Hess transforma-
tion up to arbitrary order. J. Chem. Phys. 2004, 121, 10945−10956.
(
93) Ghigo, G.; Roos, B. O.; Malmqvist, P.-Å. A modified definition
of the zeroth-order Hamiltonian in multiconfigurational perturbation
theory (CASPT2). Chem. Phys. Lett. 2004, 396, 142−149.
(
94) Forsberg, N.; Malmqvist, P.-Å. Multiconfiguration perturbation
theory with imaginary level shift. Chem. Phys. Lett. 1997, 274, 196−
04.
95) Aquilante, F.; et al. Molcas 8: New capabilities for
2
(
Tandem Mass Spectrometry of Polymetallic μ-Oxo- and Carboxylate-
+
multiconfigurational quantum chemical calculations across the
periodic table. J. Comput. Chem. 2016, 37, 506−541.
(
Bridged [Ru
3
O(CH
3
COO)
(Py)
6
2
(L)] Complexes: Intrinsic Ligand
(L) Affinities with Direct Access to Steric Effects. Organometallics
2006, 25, 3245−3250.
96) Aquilante, F.; Malmqvist, P.-Å.; Pedersen, T. B.; Ghosh, A.;
Roos, B. O. Cholesky Decomposition-Based Multiconfiguration
Second-Order Perturbation Theory (CD-CASPT2): Application to
the Spin-State Energetics of CoIII(diiminato)(NPh). J. Chem. Theory
Comput. 2008, 4, 694−702.
(114) Nunes, G. S.; Alexiou, A. D. P.; Araki, K.; Formiga, A. L. B.;
Rocha, R. C.; Toma, H. E. Proton-Coupled Redox Chemistry,
Oxidative Reactivity, and Electronic Characterization of Aqua-,
Hydroxo-, and Oxo-Triruthenium Clusters. Eur. J. Inorg. Chem.
2
(
006, 2006, 1487−1495.
(
97) Werner, H.-J.; Knowles, P. J.; Knizia, G.; Manby, F. R.; Schutz,
̈
115) Formiga, A. L. B.; Nogueira, A. F.; Araki, K.; Toma, H. E.
M. Molpro: A General-Purpose Quantum Chemistry Program
Contrasting Photoelectrochemical Behaviour of Two Isomeric
Supramolecular Dyes Based on Meso-Tetra(Pyridyl)Porphyrin
Incorporating Four (μ3-Oxo)- Triruthenium(III) Clusters. New J.
Chem. 2008, 32, 1167.
Package. WIREs Comput. Mol. Sci. 2012, 2, 242−253.
(
98) Cossi, M.; Rega, N.; Scalmani, G.; Barone, V. Polarizable
dielectric model of solvation with inclusion of charge penetration
effects. J. Chem. Phys. 2001, 114, 5691−5701.
(116) Nikolaou, S.; Formiga, A. L. B.; Toma, H. E. Probing the
(
99) Cossi, M.; Barone, V. Solvent effect on vertical electronic
electronic delocalization in a cyclic pyrazine ruthenium cluster
transitions by the polarizable continuum model. J. Chem. Phys. 2000,
12, 2427−2435.
100) Stupka, G.; Gremaud, L.; Williams, A. F. Control of Redox
hexamer. Inorg. Chem. Commun. 2010, 13, 1032−1035.
1
(
(117) Zanvettor, N. T.; Nakahata, D. H.; de Paiva, R. E.; Ribeiro, M.
A.; Cuin, A.; Corbi, P. P.; Formiga, A. L. Copper(II), Palladium(II)
and Platinum(II) Complexes with 2,2-Thiophen-Yl-Imidazole: Syn-
thesis, Spectroscopic Characterization, X-Ray Crystallographic Studies
and Interactions with Calf-Thymus DNA. Inorg. Chim. Acta 2016,
Potential by Deprotonation of Coordinated 1H-Imidazole in
Complexes of 2-(1H-Imidazol-2-yl)pyridine. Helv. Chim. Acta 2005,
8
(
8, 487−495.
101) Gutlich, P. Fifty Years of Mossbauer Spectroscopy in Solid
̈
̈
4
(
43, 304−315.
State Research − Remarkable Achievements, Future Perspectives. Z.
118) Possato, B.; Deflon, V. M.; Naal, Z.; Formiga, A. L. B.;
Anorg. Allg. Chem. 2012, 638, 15−43.
Nikolaou, S. An Extended π-System and Enhanced Electronic
(
102) Shvachko, Y. N.; Starichenko, D. V.; Korolyov, A. V.;
n
Delocalization on Symmetric [Ru O(CH COO) (L) ] Complexes
Yagubskii, E. B.; Kotov, A. I.; Buravov, L. I.; Lyssenko, K. A.; Zverev,
V. N.; Simonov, S. V.; Zorina, L. V.; Shakirova, O. G.; Lavrenova, L.
G. The Conducting Spin-Crossover Compound Combining Fe(II)
Cation Complex with TCNQ in a Fractional Reduction State. Inorg.
Chem. 2016, 55, 9121−9130.
3
3
6
3
Combined with Azanaphthalene Ligands. Dalton Trans. 2017, 46,
926−7938.
119) Saureu, S.; de Graaf, C. On the role of solvent effects on the
electronic transitions in Fe(II) and Ru(II) complexes. Chem. Phys.
014, 428, 59−66.
120) Formiga, A. L. B.; Vancoillie, S.; Pierloot, K. Electronic
7
(
2
(
(
103) Slichter, C. P.; Drickamer, H. G. Pressure-Induced Electronic
Changes in Compounds of Iron. J. Chem. Phys. 1972, 56, 2142−2160.
104) Konig, E. Nature and Dynamics of the Spin-State
Spectra of N-Heterocyclic Pentacyanoferrate(II) Complexes in
Different Solvents, Studied by Multiconfigurational Perturbation
Theory. Inorg. Chem. 2013, 52, 10653−10663.
(
̈
Interconversion in Metal Complexes. Complex Chemistry; Structure
and Bonding; Springer: Berlin, 1991; pp 51−152.
(121) Domingo, A.; Sousa, C.; de Graaf, C. The effect of thermal
(
105) Bibi, N.; Guerra, R. B.; Huamaní, L. E. S. C.; Formiga, A. L. B.
motion on the electron localization in metal-to-ligand charge transfer
Crystal Structure, Electrochemical and Spectroscopic Investigation of
2
+
3
1
excitations in [Fe(bpy)
122) Kozak, A.; Czaja, M.; Chmurzyns
base) equilibria in systems modelling interactions occurring in
biomolecules. J. Chem. Thermodyn. 2006, 38, 599−605.
3
] . Dalton Trans. 2014, 43, 17838−17846.
Mer -Tris[2-(1 H -Imidazol-2-Yl-κ N )Pyrimidine-κ N ]Ruthenium-
II) Bis(Hexafluoridophosphate) Trihydrate. Acta Crystallogr., Sect. E:
Crystallogr. Commun. 2018, 74, 874−877.
106) Tao, J.; Wei, R.-J.; Huang, R.-B.; Zheng, L.-S. Polymorphism
in Spin-Crossover Systems. Chem. Soc. Rev. 2012, 41, 703−737.
107) Bartual-Murgui, C.; Codina, C.; Roubeau, O.; Aromí, G. A
Sequential Method to Prepare Polymorphs and Solvatomorphs of
Fe(1,3-Bpp) ](ClO ) · nH O (n = 0, 1, 2) with Varying Spin-
(
́
ki, L. Investigations of (acid
(
+
(
(
[
2
4
2
2
Crossover Behaviour. Chem. - Eur. J. 2016, 22, 12767−12776.
108) Guionneau, P.; Marchivie, M.; Bravic, G.; Letard, J.-F.;
Chasseau, D. Structural Aspects of Spin Crossover. Example of the
(
́
II
[
Fe Ln(NCS) ] Complexes. Spin Crossover in Transition Metal
2
Compounds II; Springer: Berlin, 2004; Vol. 234; pp 97−128.
109) De Munno, G.; Julve, M.; Real, J. A. Synthesis and Crystal
Structure of the Low-Spin Iron(II) Complex [Fe(Bpym)3](ClO4)2·
(
1
1
(
/4H2O (Bpym = 2,2′-Bipyrimidine). Inorg. Chim. Acta 1997, 255,
85−188.
110) Marilena Toma, L.; Eller, C.; Paul Rillema, D.; Ruiz-Perez, C.;
́
Julve, M. Synthesis and Crystal Structure of the Low-Spin Iron(II)
N
DOI: 10.1021/acs.inorgchem.8b02278
Inorg. Chem. XXXX, XXX, XXX−XXX