Ligand symmetry significantly affects spin crossover behaviour in isomeric [Fe(pybox)2]2+complexes

Article abstract of DOI:10.1039/d0dt03978k

The understanding of the correlation between the spin-state behaviour and the structural features in transition-metal complexes is of pronounced importance to the design of spin crossover compounds with high performance. However, the study of the influence of ligand symmetry on the spin crossover properties is still limited due to the shortage of suitable structural systems. Herein we report the magneto-structural correlations of three mononuclear Fe(ii) isomers with respect to their ligand symmetry. In this work, two phenyl-substitutedmesoandoptically purepybox ligands were employed to constructmeso(1),optically pure(2), andracemic(3) ligand types of [Fe(pybox)₂]²⁺complexes. Their magnetic susceptibilities were measuredviatemperature-dependent paramagnetic¹H NMR spectroscopy. We fitted the midpoint temperatures of the transition (T_1/2) of 260 K for1(ClO₄), 247 K for2(ClO₄), and 281 K for3(ClO₄). The influence of structural symmetry on spin crossover was rationalized through density functional theory calculations. The optimized structures of [Fe(pybox)₂]²⁺complex cations show that the geometric distortion of the central FeN₆coordination sphere is mainly caused by the steric congestions between adjacent phenyl substituents. In these compounds, there is a distinct correlation that more steric congestions produce larger coordination distortion and favor the electron configuration in the high-spin state, which reflects in the increase ofT_1/2. Additionally, the influence of the counter anion and lattice solvent on themesoseries compounds was inspected. It is revealed that multiple factors dominate the spin-state behaviour in the solid state. This work provides deep insight into the effect of ligand symmetry on the spin transition behaviour in spin crossover compounds. It demonstrates that molecular symmetry should be considered in the design of spin crossover compounds.

Full text of DOI:10.1039/d0dt03978k

Dalton
Transactions
View Article Online
View Journal | View Issue
PAPER
Ligand symmetry significantly affects spin
crossover behaviour in isomeric [Fe(pybox)2]
complexes†
2
+
Cite this: Dalton Trans., 2021, 50,
3369
a
b
c
a
Run-Guo Wang,
Yin-Shan Meng,
Fang-Fang Gao, Wan-Qing Gao,
a
c
b
a,b
Chun-Hua Liu, Anyang Li, * Tao Liu * and Yuan-Yuan Zhu
*
The understanding of the correlation between the spin-state behaviour and the structural features in tran-
sition-metal complexes is of pronounced importance to the design of spin crossover compounds with
high performance. However, the study of the influence of ligand symmetry on the spin crossover pro-
perties is still limited due to the shortage of suitable structural systems. Herein we report the magneto-
structural correlations of three mononuclear Fe(II) isomers with respect to their ligand symmetry. In this
work, two phenyl-substituted meso and optically pure pybox ligands were employed to construct meso
2
+
(
1), optically pure (2), and racemic (3) ligand types of [Fe(pybox)
2
]
complexes. Their magnetic suscepti-
1
bilities were measured via temperature-dependent paramagnetic H NMR spectroscopy. We fitted
the midpoint temperatures of the transition (T1/2) of 260 K for 1(ClO
(ClO ). The influence of structural symmetry on spin crossover was rationalized through density func-
tional theory calculations. The optimized structures of [Fe(pybox)
metric distortion of the central FeN coordination sphere is mainly caused by the steric congestions
4 4
), 247 K for 2(ClO ), and 281 K for
3
4
2
+
2
]
complex cations show that the geo-
6
between adjacent phenyl substituents. In these compounds, there is a distinct correlation that more steric
congestions produce larger coordination distortion and favor the electron configuration in the high-spin
state, which reflects in the increase of T1/2. Additionally, the influence of the counter anion and lattice
solvent on the meso series compounds was inspected. It is revealed that multiple factors dominate the
spin-state behaviour in the solid state. This work provides deep insight into the effect of ligand symmetry
on the spin transition behaviour in spin crossover compounds. It demonstrates that molecular symmetry
should be considered in the design of spin crossover compounds.
Received 20th November 2020,
Accepted 27th January 2021
DOI: 10.1039/d0dt03978k
rsc.li/dalton
4
–7
complexes bearing d
transition metals to which organic
Introduction
3
ligands with suitable ligand field strength are coordinated. In
The development of spin crossover (SCO) transition metal these complexes, the electronic configuration of the central
complexes has been widely recognized as promising metal involves the reversible rearrangement between high-spin
approach for constructing bistable magnetic materials at the (HS) and low-spin (LS) states under physical stimuli such as
a
1
,2
single-molecule level. The SCO phenomenon exists in some temperature, pressure, light, solvent, magnetic field, etc. The
consequence of the change in the count of unpaired electrons
can switch the magnetic moment directly and influence other
a
School of Chemistry and Chemical Engineering, Anhui Key Laboratory of Advanced
Catalytic Materials and Reaction Engineering, Hefei University of Technology, Hefei
physical properties including the dielectric constant, electrical
resistance, and crystallographic phase. The rearrangement of
the electron configuration in the d orbital is triggered by the
relative energy magnitude of ligand field splitting between the
2
b
30009, China. E-mail: yyzhu@hfut.edu.cn
State Key Laboratory of Fine Chemicals, Dalian University of Technology, 2
Linggong Road, Dalian 116024, China. E-mail: liutao@dlut.edu.cn
c
t
2g and e
trons (P). When Δ > P, pairing of electrons is favoured. When Δ
P, electrons tend to separately locate in maximum d orbitals.
g
molecular orbitals (Δ) and pairing energy of elec-
Key Laboratory of Synthetic and Natural Functional Molecule of Ministry of
Education, College of Chemistry and Materials Science, Northwest University, Xi’an,
7
10127, P. R. China. E-mail: liay@nwu.edu.cn
<
†
Electronic supplementary information (ESI) available: Synthesis, characteriz-
Due to the splitting energy, Δ is sensitive with respect to the
ligand field strength, so an appropriate coordination environ-
ment plays a vital role in the occurrence of SCO in a measur-
able temperature range. The fine-tuning of the ligand field
ation, and NMR spectra of complexes, X-ray single crystal structures, PXRD
pattern, and structural analysis of complexes and supplementary magnetic data.
CCDC 2045167–2045172. For ESI and crystallographic data in CIF or other elec-
tronic format see DOI: 10.1039/d0dt03978k
This journal is © The Royal Society of Chemistry 2021
Dalton Trans., 2021, 50, 3369–3378 | 3369

View Article Online
Paper
Dalton Transactions
may affect SCO behaviour to a large extent. Extensive research netic properties in solution and in the solid state. Compared
on SCO complexes shows that the ligand’s substituent, counter with their enantiopure and racemic analogues, the diverse
ion, co-crystallized solvent, and polymorphism in the solid SCO behaviour caused by the variation in chirality in the pybox
state are operative factors to influence their spin-state pro- ligand is observed. Density Functional Theory (DFT) calcu-
4
perties. Very recently, chirality in enantiopure ligands that lations are carried out to optimize the geometries of three
2
+
could influence SCO was investigated in some mononuclear forms of [Fe(pybox)
complexes and coordination polymers.
2
]
complex cations. The calculated results
5
–8
For example, the are rationalized with the experimental magnetic and structural
enantiopure and racemic isomers are crystallized in chiral and data. Furthermore, the magneto-structural correlation is
achiral space groups, respectively. The discriminations in the discussed.
molecular structure and packing probably change their ligand
field in which the tunable SCO properties can be attained.
Tridentate pincer-type ter-imine molecules constitute a
Results and discussion
Synthesis of the ligand
large family of organic ligands in constructing SCO-active com-
9
pounds. We first demonstrated that a new type of tridentate
nitrogen ligand, pyridine-2,6-bis (oxazoline) (pybox), is In this work, the meso type pybox ligand (L
efficient in constructing an SCO-active mononuclear thesized for the first time. The detailed synthetic procedure is
meso
) was syn-
1
0
complex. Subsequently, the tuning of SCO properties via the shown in Scheme 1. First, amidation of dimethyl pyridine-2,6-
ligand structure, counter anion, and co-crystallized solvent was dicarboxylate with (R)- and (S)-phenylglycinol stepwise gener-
1
1–13
studied by us and other groups.
pybox ligands contain a chiral centre, [Fe(pybox)2] type com- line rings bearing opposite chiral centres through intra-
Considering that some ated 3 with a moderate yield. Then the formation of two oxazo-
2
+
meso
plexes are an ideal platform to evaluate the possible corre- molecular cyclization afforded L
lation between chirality and SCO behaviour. More recently, the ture and purity of L
Halcrow group and we independently discovered the subtle mass spectroscopy, and elemental analysis. L
in a high yield. The struc-
meso
1
13
were confirmed by H and C NMR,
meso
and its opti-
R/S
structural diversity of homogeneous and heterogeneous [Fe cally pure analogue (L ) possess totally different molecular
2
+
R/S
(
pybox)
2
]
complexes, in which two same enantiopure ligands symmetries. In the optically pure form L , there exists one C
2
R/S
and a pair of racemic ligands are assembled into the com- rotation axis. The overall symmetry of L
plexes, respectively, resulting in significantly different spin- group C
is of the point
2
; so this molecule shows optical activity. Nevertheless,
belongs to the C point group. A symmetry mirror (σ)
1
4,15
meso
state behaviours.
In comparison with heterogeneous com-
L
s
meso
pounds, their homogeneous analogues have lower symmetry divides the molecule equally, making L
both in the molecular structure and the crystallized space cule (see Fig. 1(a)).
group. The reduced molecular symmetry produces structural
an achiral mole-
Synthesis and characterization of the complexes
distortion of the coordination sphere of the metal ion and ulti-
1
6
meso
mately affects the spin-state behaviour. To further explore The synthetic procedure of L
series complexes bearing
the influence of ligand symmetry in isomers, we thought to different counter anions is as follows. Treatment of 1 equiv. of
meso
investigate and compare the spin-state behaviour of [Fe iron(II) salt with 2 equiv. of L
in methanol afforded the fol-
2
+
(
pybox)
2
]
complexes when the ligand’s chirality is in the opti- lowing complexes as dark red-coloured precipitates: [Fe
meso
meso
cally pure, racemic and meso forms. This research will deepen (L
) ][ClO ]
(1(ClO )), [Fe(L
) ][BF ]
4
(1(BF )), [Fe
2
4 2
4
2
4 2
the understanding of SCO behaviour in transition metal com- (Lmeso)2][PF ] (1(PF )), and [Fe(L
meso
) ][BPh ] (1(BPh )). Then
6 2
6
2
4 2
4
plexes when the ligands contain a chiral element. In this con- the crystals of the four complexes suitable for single X-ray diffr-
tribution, the phenyl-substituent meso type pybox ligand action analysis were obtained through recrystallization. The
meso
meso
(
L
) is synthesized for the first time. Using L
as the detailed operation was carried out via slow gas evaporation of
ligand, the Fe(II) mononuclear complexes bearing four diethyl ether into the diluted acetonitrile solution of com-
4 4 6 4
different counter anions (ClO , PF , and BPh ) are plexes. All crystals show dark red colour. Among them, 1(ClO4)
−, BF
−
−
−
obtained. We investigated their temperature-dependent mag- is a solvent-free crystal while the other three contain a lattice
meso
Scheme 1 The synthetic route of ligand L
.
3370 | Dalton Trans., 2021, 50, 3369–3378
This journal is © The Royal Society of Chemistry 2021

View Article Online
Dalton Transactions
Paper
intramolecular steric congestions between phenyl substituents in
1
(ClO ) and 2(ClO ), causing the deviation of coplanarity
4 4
between two oxazoline rings and the central pyridine ring (see
Fig. 1(b)). In contrast, four phenyl substituents in 3(ClO ) are sep-
4
arately located at different positions. Each pair of phenyl substi-
tuents is ‘perfectly’ parallel to the pyridine ring from the other
ligand. The sums of the steric phenyl pair are one in 1(ClO
4
), two
in 2(ClO ) and null in 3(ClO ). The steric congestion has a great
4
4
impact on the molecular structure that not only affects the π–π
stackings between phenyl and pyridine rings but also causes the
distortion of the oxazoline ring (see Fig. 1(b)).
The symmetry discrimination of the three complexes in solu-
1
1
tion was investigated via H NMR spectroscopy. The H NMR
spectra of three Fe(II) complexes and isostructural Zn(II) com-
plexes (perchlorate as the counter anion) are shown in Fig. S14†
meso
S
Fig. 1 The structures of ligands (a) L
and L . (b) The structures of
) (middle) and 3(ClO ) (right).
1
meso
S/R
complex cations in 1(ClO
4
) (left), 2(ClO
4
4
and Fig. 2. Although the H NMR spectra of L
and L are
similar (see Fig. S13†), the proton signals in three Fe(II) series
complexes and their diamagnetic Zn(II) analogues exhibit signifi-
solvent. The sums of acetonitrile per complex are included in cant differences both in the chemical shift and in sum of peaks
1
their names. For comparison of the structural and magnetic (see Fig. S14† and Fig. 2). H NMR spectra reveal that the meso
discrimination imposed by ligand symmetry, the homochiral ligand type 1(ClO )-Zn shows the lowest symmetry. The doubling
4
S
complex [Fe(L )
2
][ClO
4
]
2
4
(2(ClO )) and the heterochiral proton peaks in the phenyl ring range is observed compared
complex [Fe(L )(L )][ClO ] (3(ClO )) were synthesized. Their with its optically pure and racemic forms due to two phenyl sub-
R
S
4
2
4
structures and purity were confirmed through single crystal stituents in one ligand being inequivalent in the chemical
X-ray diffraction analysis (see Fig. S19–S30†) and paramagnetic environment (see Fig. 2). A similar phenomenon occurs in their
1
H NMR spectroscopy (see Fig. S6–S9†). The phase purity of pyridine and oxazoline proton ranges. It is evident that the
bulk samples was determined through powder XRD analysis differently arranged types of phenyl rings can considerably
(see Fig. S37–S42†). In addition, the content of the lattice change the electronic state and thus affect their ligand field.
solvent was further determined by elemental analysis and TGA
measurement (see Fig. S31–S36†).
Solution-state magnetic studies
Single crystal X-ray diffraction analyses revealed that four The geometric coordination and SCO behaviour of complexes
complexes 1(ClO4), 1(BF )·MeCN, 1(PF )·MeCN, and
in the solid state are greatly affected by intermolecular coop-
O adopt the achiral space group C2/c, Pbca, erativity, lattice solvent, and crystal packing. Therefore, it is
/n, respectively (see Fig. S19–S30†). Their octahe- inaccurate to inspect the unique influence of the ligand struc-
1
4
6
4 2
(BPh )·MeCN·Et
C2/c, and P2
1
dra of Fe(II) centres vary in geometric distortion due to the ture on their SCO properties through magnetic measurement
influence of the anion and lattice solvent in some cases (see in the solid state. Nevertheless, the complex molecules in solu-
Fig. S3†). A large anion and lattice solvent tend to cause great tion are well isolated by the solvent; so such effects can be
distortion and thus favour a HS state. Two control complexes 2
4 4
(ClO )·MeOH and 3(ClO )·MeOH, which were grown from
liquid–liquid diffusion in methanol, crystallized in the chiral
space group P2 2 2 and the achiral space group P2 /n,
1
1
1
1
respectively.
For a comparison of the X-ray structures in the solid state,
their structural diversity caused by ligand symmetry was investi-
gated in three perchlorate series complexes. The crystal structures
of complex cations in 1(ClO
Fig. 1(b). First, they possess different structural symmetries. The
approximate molecular point groups of complex cations are C
for 1(ClO ), C for 2(ClO ), and S for 3(ClO ). Among them, only
(ClO ) displays optical activity. In 3(ClO ), a pair of racemic
4 4 4
), 2(ClO ) and 3(ClO ) are shown in
1
4
2
4
4
4
2
4
4
ligands are coordinated to the central Fe(II) cation; so it is actually
a meso type complex. Second, the positions of phenyl substitu-
ents are variable with respect to the chirality of ligands in the
complex. In these phenyl substituent complexes, the intra-
molecular π–π stackings between phenyl and pyridine rings are
1
Fig. 2 The merged H NMR spectra of 1(ClO
)-Zn in CD CN at 298 K.
4 4
)-Zn, 2(ClO )-Zn, and 3
14,15,17
thought to stabilize the complex structure.
There exist (ClO
4
3
This journal is © The Royal Society of Chemistry 2021
Dalton Trans., 2021, 50, 3369–3378 | 3371

View Article Online
Paper
Dalton Transactions
excluded to a great extent. It is widely accepted that the para-
1
magnetic H NMR spectroscopy method is reliable to measure
the temperature-dependent magnetic susceptibilities in
1
8–20
solution.
Hence, the temperature-dependent magnetic
susceptibilities of three complexes, 1(ClO ), 2(ClO ), and 3
4
4
1
(
4 3
ClO ), were measured in CD CN via paramagnetic H NMR
spectroscopy. As shown in Fig. S15a, S16a and S17a,† their
NMR proton signals all exhibit a considerable chemical shift
towards downfield in accordance with the increase in tempera-
ture, causing a gradual rise of the HS state fraction. The χ
M
T
values at specific temperatures were calculated using the
2
1
Evans method (see eqn (1)):
ꢀ
ꢁ
3
Δf
dia
χMT ¼
þ χM ꢀ T
ð1Þ
4
mf
where m is the concentration of the paramagnetic solution in g
−
1
mL , f is the spectrometer frequency in Hz, Δf is the shift of
M 4 4
Fig. 3 Temperature dependence of χ T for 1(ClO ), 2(ClO ), and 3
the CD
pure CD
3
CN peak in the paramagnetic solution compared to (ClO ) in CD CN. The plots represent the experimental data which were
4
3
dia
M
3
CN in Hz (see Fig. S15b, S16b and S17b†), and χ
=
collected between 243 and 333 K with an interval of 10 K. The lines
correspond to the best fit found for each compound using the regular
solution model.
.5 M × 10− cm mol (M is the molecular weight). The cal-
6
3
−1
22
0
M
culated data were plotted in the χ T vs. T form, which indi-
cated that the three complexes all underwent gradual SCO
within the measured temperature range, from 243 to 333 K.
Table 1 Solution-state thermodynamic parameters in SCO for the
three complexes
These detached χ
M
T plots were modeled as a gradual and com-
plete SCO curve using the following regular solution model
(
eqn (2)). This fitting can derive some thermodynamic para-
a
ΔH (kJ mol−1
ΔS (J mol−1 K)
b
No
T
1/2 (K)
)
LS
2
3,24
meters in the SCO process:
1
2
(ClO
(ClO
4
)
)
260
247
281
22.8
22.4
23.3
88
76
83
0.9812
0.9931
0.9944
χ T
4
M
ðmaxÞ
χMT ¼
ꢀ
ꢁ
ð2Þ 3(ClO4)
ΔH ΔS
ꢁ
a
b
RT
R
T1/2 = ΔH/ΔS. Least squares.
1
þ e
where χ
M M
T(max) is the maximum χ T value for the complete HS
state. As per our previous experience with this family of com-
3
−1
plexes, the saturated χ T value of 3.9 cm K mol was used in
M
eqn (2) (Fig. 3).
For the data of the three complexes, good fitting with the
least squares value larger than 0.98 was obtained and the
thermodynamic parameters, ΔH, ΔS, and T1/2, were fitted (see
Table 1). Our fitting results for 2(ClO
ent with the results reported by the Halcrow group. The SCO
midpoint temperatures (T1/2) of the three complexes in CD CN
follow the order of 3(ClO ) > 1(ClO ) > 2(ClO ), suggesting that
racemic ligands in the complex impose a stronger ligand field
than the other two forms. The lowest T1/2 of 2(ClO ) reveals
that two homochiral ligands provide the weakest ligand field,
which is probably caused by the geometric distortion of the
coordination sphere due to two steric congestions of the
phenyl substituents in the structure.
4 4
) and 3(ClO ) are consist-
1
4
Fig. 4 The optimized quintet structures of (a) 1, (b) 2, and (c) 3.
3
4
4
4
Some key structural parameters derived from the optimized
structures which can quantitatively reflect the geometric dis-
4
tortion of the FeN
Table 2. For the three structural parameters, bigger ∑ and
smaller θ and ϕ imply larger geometric distortion of the FeN
6
coordination sphere are presented in
6
coordination sphere. In the three series of structures, the
degrees of distortion all follow the order of quintet > singlet.
The Fe–Nav lengths in different spin states are in line with the
Theoretical studies
To rationalize the varied SCO behaviour and structural stability well-concluded experimental observation for Fe(II) SCO
1
4,15
for the three types of complexes, three complex cations (1–3) were complexes.
optimized at the M06/6-31G(d) level
Generally, a length decrease of ∼0.2 Å occurs
25,26
in two spin multiplicities, from the quintet to singlet state.
high spin (S = 2) and low spin (S = 0). The optimized structures in
the singlet and quintet states are shown in Fig. 4 and S47.†
In addition, continuous shape measurements (CShMs) were
used to identify the deviation of the coordination sphere to the
3372 | Dalton Trans., 2021, 50, 3369–3378
This journal is © The Royal Society of Chemistry 2021

View Article Online
Dalton Transactions
Paper
Table 2 Summary of some key structural parameters of the optimized structures 1–3 at different spin states
Complex
Spin state
∑ (°)
θ (°)
ϕ (°)
Fe–Nav (Å)
S(O
h
)
h
S(T )
1
1
2
2
3
3
Singlet
Quintet
Singlet
Quintet
Singlet
Quintet
91.38
145.50
88.18
156.75
91.55
89.17
84.06
88.22
82.71
89.98
88.59
178.52
168.04
179.61
178.88
179.92
179.99
1.962
2.161
1.962
2.164
1.967
2.155
2.16
5.78
2.14
5.55
2.28
4.97
9.50
6.39
9.53
6.30
9.26
6.45
142.53
3
1
4
∑
: the sum of the deviations from 90° of the cis angles; θ: the dihedral angle between the planes of the pybox ligands; ϕ: the trans-N[pyri-
4
dine]–Fe–N[pyridine] angle; Fe–Nav: the average value of the six Fe–N bond lengths in one FeN
6 h
coordination sphere; S(O ): the result of the
CShM calculation denotes the deviation value of ideal O symmetry; S(T ): the result of the CShM calculation denotes the deviation value from
h
h
2
7
h
ideal T after removing pyridine N.
2
7
ideal octahedron (O ) and tetrahedron (T ). When the spin the solid state. On comparison of relative energies for the
h
d
state is changed from the quintet to singlet state, significant three complexes, 3 possesses the lowest energies in both spin
decreases of the values of S(O ) and S(T ) are observed, states. Conversely, the two states for 2 are the highest. The
h
d
suggesting that the symmetry increases from the high-spin to energy calculations disclose that the relative structural stability
low-spin state. Overall, the structural parameters of six calcu- is in the order of 3 > 1 > 2, consistent with the structural inves-
lated structures quantitatively demonstrate the structural diver- tigation for intramolecular steric congestion. The discrimi-
1
sity in different ligand symmetries and spin states. The trend nation of stability between 2 and 3 was also confirmed by H
in the calculated structures is concomitant with the single NMR experiments. When equimolar amounts of (R)-2(ClO4)
crystal structures in specific spin states, also implying the and (S)-2(ClO
reliability of the calculated results. was performed and the sole crystals of 3(ClO ) were grown
Based on the optimized structures of the three complexes after vapor diffusion of ether into the solution (see
4
) were dissolved in acetonitrile, racemization
4
3
0
in specific spin states (S = 0 and 2), their relative energies are Fig. S18†). The energy gap between 2 and 3 in the quintet
2
8
−1
presented in Fig. 5. Similar to other SCO systems, the high- state is just 5.58 kJ mol , which is small enough for the occur-
spin state of complex cations 1–3 with S = 2 is the ground rence of racemization at ambient temperature. Additionally,
state. The energy gaps between high spin (S = 2) and low spin the spin density maps of the three complex cations at quintet
S = 0) for 1–3 are 14.90, 13.14, and 14.27 kJ mol− , respect- states were calculated, showing that the spin density is par-
1
(
ively. Neese and co-worker have pointed out that the energy tially delocalized at the coordinated N atoms and pyridine ring
−
1
difference of 0 and 25.12 kJ mol is proposed to be indicative (see Fig. S48†).
2
9
of SCO behaviour. The energy gaps of the three complexes
are within this range in which the spin transitions are Solid-state magnetic studies
observed via magnetic measurements both in solution and in
Subsequently, the temperature-dependent magnetic suscepti-
bilities for four complexes constructed from the meso ligand, 1
ClO ), 1(BF )·MeCN, 1(PF )·MeCN, and 1(BPh )·MeCN·Et O,
(
4
4
6
4
2
in the solid state were measured using SQUID. The effect of
the counter anion and/or co-crystallized solvent in these com-
3
2–36
plexes was disclosed.
Because the rear three crystals were
grown in the solvated form, they were protected with Parafilm
to minimize solvent loss in the measurement of the first cycle.
After completion of the first cycle, the samples were main-
tained at 400 K for 2 hours to completely remove the lattice
solvent. Then the later cycles revealed the magnetic properties
of desolvated samples. The temperature-dependent magnetic
susceptibilities for the four complexes under successive cycles
are shown in Fig. 6 and S43,† indicative of the variable SCO
behaviour of meso type complexes as per the counter anion
and co-crystallized solvent.
3
−1
4 M
For 1(ClO ), its χ T value at 300 K is only 0.64 cm K mol
and gradually decreases to zero at low temperature, indicating
that most of the iron(II) ions are located at the low-spin state in
the range of 2–300 K. Upon warming, the χ T value gradually
M
3
−1
increases to 2.05 cm K mol at 400 K, but just reaching the
Fig. 5 Energy diagram for different spin multiplicities for the complex
cations of 1, 2, and 3.
high-spin population of ∼53% (see Fig. 6(a), assuming that the
This journal is © The Royal Society of Chemistry 2021
Dalton Trans., 2021, 50, 3369–3378 | 3373

View Article Online
Paper
Dalton Transactions
at the high-spin state regardless of the removal of the lattice
solvent (see Fig. S46†). These results reveal that the structural
symmetry, counter anion and lattice solvent jointly influence
3
7
the spin-state behaviour of the complex in the solid state.
Magneto-structural correlation studies
Finally, the structural factors that can directly affect the spin-
state behaviour are discussed. First, to reveal the unique effect
by structural symmetry, we focus on the optimized calculated
complex cations to compare their structural diversity caused
by ligand symmetry. From the six optimized structures, it is
found that the oxazoline rings exhibit considerable geometric
distortion. To describe the degree of distortion quantitatively,
a structural parameter μrms is introduced. This parameter is
defined as the root mean square of the distances (μ, Å) of
M 4 4
Fig. 6 Temperature dependence of χ T for 1(ClO ) (a), 1(BF
)·MeCN (b), atoms in the oxazoline ring to the mean coordinated plane
1
(PF
6
)·MeCN (c), and 1(BPh
4 2
)·MeCN·Et O
(d) upon two repeated constructed by the central Fe(II) ion and three N atoms (shown
scannings.
in light green, see Fig. 7(a)). The μrms values of each structure
are calculated from the projection distances of twenty atoms in
four oxazoline rings to the coordination plane and are shown
3
saturated χ
mol ). 1(BF
M
T value at the complete HS state is 3.9 cm
K
in Fig. 7(b) and Table S7.† 2 displays the largest deviation com-
pared with 1 and 3 in two spin states. Conversely, the relative
μrms values of 3 are the smallest. Additionally, 2 shows the
largest variation in the μrms value between singlet and quintet
states. According to the analyses of the data, two conclusions
can be obtained: (1) the oxazoline rings in 2 show the largest
−
1
4
)·MeCN also displays an incomplete SCO in the
temperature range of 2–400 K. At 400 K, the χ T value only
reaches 0.70 cm K mol , far away from its complete HS state
see Fig. 6(b)). The SCO behaviour in 1(PF )·MeCN can be
switched via lattice acetonitrile. The solvated sample is a com-
plete LS compound in the temperature range of 2–300 K (see
M
3
−1
(
6
3
8,39
geometric distortion in two states due to steric congestion
the χ
M
T vs. T curves in the first cooling/heating cycle). Upon
heating from 300 K, its χ T value undergoes an abrupt
and (2) the spin transition causes oxazoline rings in 2 to
encounter the greatest structural change. Generally, large geo-
metric distortion in coordination will weaken the ligand field
M
−1
3
increase to 3.06 cm K mol at 400 K as a result of the
removal of the lattice solvent. Afterwards, the desolvated
sample displays a two-step SCO behaviour (see Fig. 6(c)). The
first transition at a higher temperature range shows no
thermal hysteresis and its T1/2 value is ∼360 K. In the second
SCO, a hysteresis loop of ΔT = 20 K appears (T1/2↓ = 260 K and
4
0,41
and thus favor the high-spin state.
The great structural
change in high-spin and low-spin states makes the occurrence
4
of SCO more difficult. The structural feature can well explain
why 2 possesses the lowest T1/2. A similar phenomenon is also
observed in another structural parameter θ (see Fig. S49†). The
4
2
T
1/2↑ = 280 K). This loop is stable and repeatable at the temp-
angle θ is the dihedral angle between two ligand planes.
−1
erature-scan rate of 3 K min (see Fig. S44†). The loss of the
lattice solvent releases the crowd, thus reducing the geometric
distortion of the Fe(II) complex cation, making the electron
Owing to the high symmetry of 3, a minor change in θ takes
place during the SCO process. The correlation between single
point energy and T1/2 is plotted (see Fig. S50†), indicating that
the structure of the lower energy level favours the low-spin
state.
4 2
configuration favor the low-spin state. 1(BPh )·MeCN·Et O fea-
tures an incomplete SCO. It locates in a complete high-spin
state above 300 K. Upon cooling, the spin transition takes
The structural features of the available single crystal struc-
3
−1
2
+
M
place and ends at ∼72 K with a χ T value of 1.34 cm K mol ,
tures of phenyl-substituted [Fe(pybox)2] are summarized and
indicative of the residual HS population of ∼34%. In the
second cycle, desolvation results in the spin transition region
moving to a high temperature of ∼40 K (see Fig. 6(d)).
Moreover, the solvent effects in optically pure and racemic
ligand type complexes were tested (see Fig. S45 and S46†). Two
complexes with methanol as the lattice solvent were grown (see
Fig. S27–S30†). Before and after the removal of lattice metha-
4
nol, 2(ClO )·MeOH displays a two-step SCO. The transitions of
solvated and desolvated forms are both different compared
with previously reported 2(ClO )·MeCN (see Fig. S45†). It
4
suggests that the annealing phases from two solvated com-
plexes are different. Unlike 3(ClO )·MeCN which is SCO-active
4
Fig. 7 (a) Definition and equation of the μrms value for the deviation of
atoms in the oxazoline ring from coordination planarity; (b) the μrms
both in solvated and desolvated forms, 3(ClO )·MeOH remains values of three calculated structures in specific spin multiplicities.
4
3374 | Dalton Trans., 2021, 50, 3369–3378
This journal is © The Royal Society of Chemistry 2021

View Article Online
Dalton Transactions
Paper
4
3
4
compared. Except 1(ClO ) which is in a desolvated form, all Synthesis of 2
others contain lattice solvents. The co-crystallized solvents in
the unit cell affect the packing of complex molecules. In most
cases, they produced great geometric distortion in the solid
state. After the loss of the solvent, SCO emerges or moves to a
higher temperature. It is evident that the removal of the
solvent results in the release of structural distortion of the
complex cation. The lattice solvent, together with the steric
congestion of adjacent phenyl rings and the counter anion,
contributes to the twisting of the phenyl substituent.
Normally, larger phenyl twisting can be observed in the opti-
cally pure and solvated compounds (see Tables S2–S5†).
A mixture of 1 (3.71 g, 19.03 mmol), (S)-2-amino-2-pheny-
lethan-1-ol (2.35 g, 17.12 mmol) and MeOH (30 mL) was added
to a sealed tube; then the mixture was stirred at 110 °C for
4
8 hours. After removing the solvent under vacuum, the crude
product was purified by column chromatography (SiO , pet-
roleum ether/ethyl acetate, 1 : 2) to give compound 2 as a light
2
1
yellow solid (3.07 g, 59%). H NMR (600 MHz, CDCl
3
): δ 9.12
= 0.9 Hz, 1H), 8.20
dd, J = 7.8 Hz, J = 1.0 Hz, 1H), 7.94 (t, J = 7.8 Hz, 1H), 7.42
(
(
(
1 2
d, J = 7.3 Hz, 1H), 8.24 (dd, J = 7.8 Hz, J
1
2
d, J = 7.4 Hz, 2H), 7.36 (t, J = 7.6 Hz, 2H), 7.29 (t, J = 7.3 Hz,
1H), 5.30 (dd, J = 12.2 Hz, J = 7.2 Hz, 1H), 4.08–4.05 (m, 2H),
1 2
4.04 (s, 3H), 3.40 (t, J = 6.2 Hz, 1H).
Synthesis of 3
Conclusions
A mixture of 2 (3.05 g, 10.2 mmol), (R)-2-amino-2-phenylethan-
In summary, we have investigated and compared the discrimi- 1-ol (2.05 g, 14.9 mmol) and MeOH (20 mL) was added to a
nation of structural symmetry and SCO behaviour in optically sealed tube; then the mixture was stirred at 110 °C for
2
+
pure, racemic, and meso ligand type [Fe(pybox)2] isomers 48 hours. After removing the solvent under vacuum, the crude
bearing phenyl substituents. The diverse orientations of the product was purified by column chromatography (SiO , ethyl
2
1
chiral centre in the ligand result in totally different molecular acetate) to give compound 3 as a white solid (3.44 g, 83%). H
symmetries of the complexes. In this series of complexes, NMR (400 MHz, DMSO-d ): δ 9.31 (d, J = 9.0 Hz, 1H), 8.26–8.13
6
chiral discrimination has a great impact on their SCO behav- (m, 2H), 7.47 (d, J = 7.2 Hz, 2H), 7.35 (t, J = 7.4 Hz, 2H), 7.28 (t,
iour both in solution and in the solid state. The ‘mismatched’ J = 7.2 Hz, 1H), 5.20 (dd, J = 14.5 Hz, J = 6.7 Hz, 1H), 5.10 (t, J
1
2
1
3
chiral arrangement produces steric congestion of phenyl sub- = 5.3 Hz, 1H), 3.83 (pd, J = 11.3 Hz, J = 7.1 Hz, 2H). C NMR
1
2
stituents thereby distorting the FeN coordination octahedron. (151 MHz, DMSO): δ 163.3, 149.2, 140.8, 139.5, 128.3, 127.1,
6
DFT calculations give the energy profile of three type com- 127.0, 124.7, 64.5, 55.6. HRMS (MALDI-TOF) calcd for
+
plexes in different spin multiplicities. The racemic type C H N O [M + H] 406.17, found 406.08. Anal. calcd for
2
3
23 3 4
complex displays the highest symmetry and possesses rela- C H N O : C, 68.13; H, 5.72; N, 10.36. Found: C, 68.01; H,
2
3
23 3 4
tively low energies in two multiplicities. According to the ana- 5.62; N, 10.56.
lyses of the optimized calculation structures, less geometric
meso
distortion in 3 is found. The high symmetry of coordination Synthesis of L
geometry imposes a higher ligand field and thus favours the
low-spin state. It is consistent with the experimental investi-
A solution of 3 (1.13 g, 2.8 mmol), TsCl (1.32 g, 6.8 mmol),
DMAP (34 mg, 6.8 mmol) and Et N (2 mL) in CH Cl was
3
2
2
gation in acetonitrile that reports that 3(ClO ) has the highest
4
stirred at 0 °C for 3 hours. Subsequently, the mixture was
heated to 110 °C for 48 hours. After cooling to ambient temp-
erature, the solution was diluted with CH Cl (10 mL) and
T
1/2 in this series. In the meso series complexes, SCO pro-
perties are also strongly affected by the counter cation and
lattice solvent. This work systematically discusses the influence
of ligand symmetry on the spin state and reveals that subtle
structural discrimination in isomers can fine-tune the SCO
behaviour.
2
2
washed with 1 N NaOH aqueous solution (10 mL), the organic
phase was washed with DI water (30 mL × 3) and brine (30 mL)
and subsequently dried over anhydrous sodium sulfate. After
the solvent was removed, the solid was washed with ethanol
meso
(
10 mL) and then filtered to give L
as a white crystalline
1
powder (560 mg, 53%). H NMR (600 MHz, CDCl ): δ 8.34 (d, J
3
=
7.8 Hz, 2H), 7.92 (t, J = 7.8 Hz, 1H), 7.41–7.27 (m, 10H), 5.46
Experimental
Synthesis of 1
(
t, J = 9.5 Hz, 2H), 4.97–4.90 (m, 2H), 4.43 (t, J = 8.6 Hz, 2H).
1
3
C NMR (151 MHz, CDCl ): δ 165.1, 148.3, 143.3, 139.1, 130.5,
3
1
29.5, 128.4, 128.0, 79.0, 78.8, 78.6, 71.9. HRMS (MALDI-TOF)
A solution of 2,6-pyridinedicarboxylic acid (8.35 g, 0.5 mol)
and concentrated H SO (1 mL) in methanol (50 mL) was
2 4
heated under reflux for 3 hours. After cooling to ambient
temperature, the solvent was removed under vacuum. The
crude product was dissolved in ethyl acetate and washed with
water (50 mL × 3) and brine (50 mL) successively and dried
+
calcd for C23
H
19
N
3
O
2
[M + H] 370.15, found 370.03. Anal.
calcd for C H N O : C, 74.78; H, 5.18; N, 11.37. Found: C,
23 19 3 2
74.98; H, 5.23; N, 11.30.
Synthesis of 1(ClO
4
)
over anhydrous sodium sulfate. After the solvent was removed A solution of Fe(ClO
4 2 2
) ·6H O (36 mg, 0.1 mmol) in methanol
under reduced pressure, the product was obtained as a white (5 mL) was added to a stirred methanol solution (5 mL) of
meso
powder without the need for further purification (8.87 g, 91%).
L
(74 mg, 0.2 mmol). The dark violet solution was stirred
This journal is © The Royal Society of Chemistry 2021
Dalton Trans., 2021, 50, 3369–3378 | 3375

View Article Online
Paper
Dalton Transactions
for 30 minutes and concentrated under vacuum. Dark violet Synthesis of 3(ClO
cubic crystals (73 mg, 74%) were obtained by slow diffusion of
4
)·MeOH
4
The procedure is identical to that of 2(ClO )·MeOH except that
1
ether vapor into an acetonitrile solution of the complex.
NMR (600 MHz, CD CN, 298 K): δ 55.30 (br, 2H), 53.54 (br,
H), 25.55 (s, 2H), 12.32 (s, 2H), 10.43 (s, 2 H), 9.21 (s, 2H),
.91 (s, 2H), 7.62 (s, 4H), 6.59 (s, 4H), 6.23 (s, 6H), 4.11 (s, 2H),
.23 (s, 4H), 2.44 (s, 4H). Anal. calcd for C46 FeN 12: C,
5.61; H, 3.86; N, 8.46. Found: C, 55.21; H, 3.62; N, 8.79.
H
S
S
R
L was replaced by equal amounts of L and L . Dark red block
3
1
crystals were obtained with a yield of 57%. H NMR (600 MHz,
CD CN, 298 K): δ 40.83 (br, 4H), 21.35 (s, 2H), 7.48 (s, 4H),
3
.22 (s, 4H), 6.86 (s, 8H), 6.18 (s, 4H), 4.52 (br, 4H), 3.43 (s,
H). Anal. calcd for C H Cl FeN O : C, 55.04; H, 4.13; N,
.19. Found: C, 55.35; H, 3.82; N, 8.51.
2
7
3
5
7
8
8
H
38Cl
2
6
O
4
7
42
2
6 13
Synthesis of 1(BF
4
)·MeCN
Synthesis of analogous Zn complexes
meso
The procedure is identical to that of 1(ClO ) except that Fe
4
Three Zn(II) complexes, [Zn(L
(
) ][ClO ] (1(ClO )-Zn), [Zn
2 4 2 4
R
R
S
(
ClO
cubic crystals (62 mg, yield: 64%) were obtained. H NMR
600 MHz, CD CN, 298 K): δ 55.39 (br, 2H), 53.57 (br, 2H),
4 2 2 4 2 2
) ·6H O was replaced by Fe(BF ) ·6H O. Dark purple
L ) ][ClO ] (2(ClO )-Zn), and [Zn(L L )][ClO ] (3(ClO )-Zn),
2 4 2 4 4 2 4
were synthesized using a similar procedure to that of their
archetypic Fe(II) complexes except that Zn(ClO ) was used as
1
(
3
4
2
2
2
5.74 (s, 2H), 12.25 (s, 2H), 10.29 (s, 2H), 9.03 (s, 2H), 7.94 (s,
H), 7.65 (s, 4H), 6.58 (s, 2H), 6.22 (s, 6H), 5.98 (br, 2H), 3.97
the metal source; the products were afforded as white powders
1
after drying under vacuum. 1(ClO
4
)-Zn: H NMR (600 MHz,
(
C
s, 2H), 3.19 (s, 4H), 2.35 (s, 4H). Anal. calcd for
FeN : C, 57.12; H, 4.09; N, 9.71. Found: C, 56.91;
H, 3.87; N, 9.39.
CD CN, 298 K): δ 8.42 (t, J = 7.9 Hz, 2H), 8.32 (d, J = 7.9 Hz,
3
48
H
41
B
2
F
8
7 4
O
2
H), 7.61 (d, J = 7.9 Hz, 2H), 7.18 (t, J = 7.2 Hz, 2H), 7.15–7.04
m, 10H), 7.01 (t, J = 7.7 Hz, 4H), 6.65 (d, J = 7.3 Hz, 4H),
.49–5.35 (m, 4H), 5.21 (t, J = 7.4 Hz, 2H), 4.91–4.77 (m, 2H),
4.54–4.42 (m, 4H). Anal. calcd for C H Cl ZnN O : C, 55.08;
(
5
Synthesis of 1(PF )·5MeCN
6
4
6
38
2
6 12
The procedure is identical to that of 1(ClO ) except that Fe H, 3.82; N, 8.38. Found: C, 54.81; H, 4.01; N, 8.02. 2(ClO
4
)-Zn:
CN, 298 K): δ 8.47 (t, J = 7.9 Hz, 2H),
8.04 (d, J = 7.9 Hz, 4H), 7.21 (t, J = 7.4 Hz, 4H), 7.08 (t, J = 7.7
NMR (600 MHz, CD CN): δ 55.81 (br, 2H), 54.00 (br, 2H), 25.88 Hz, 8H), 6.76 (d, J = 7.3 Hz, 8H), 5.26–5.20 (m, 4H), 5.17 (t, J =
4
1
(
ClO ) ·6H O was replaced by FeCl ·4H O and KPF . Dark
H NMR (600 MHz, CD
3
4
2
2
2
2
6
1
purple cubic crystals (86 mg, yield: 79%) were obtained.
H
3
(
s, 2H), 12.37(s, 2H), 10.31 (s, 2H), 9.07 (s, 2H), 7.94 (s, 2H), 10.6 Hz, 2H), 4.74 (dd, J
1
= 10.7 Hz, J
H38Cl ZnN O12: C, 55.08; H, 3.82; N, 8.38. Found:
2 6
2
= 8.8 Hz, 2H). Anal.
7
2
.65 (s, 4H), 6.58 (s, 2H), 6.22 (s, 6H), 5.97 (br, 2H), 3.95 (s, calcd for C46
H), 3.13 (s, 4H), 2.32 (s, 4H). Anal. calcd for C, 55.01; H, 3.99; N, 8.09. 3(ClO
CN, 298 K): δ 8.46 (t, J = 7.9 Hz, 2H), 7.99 (d, J = 7.9 Hz,
4H), 7.28 (t, J = 7.4 Hz, 4H), 7.15 (t, J = 7.7 Hz, 8H), 6.94 (d, J =
.2 Hz, 8H), 4.99 (dd, J = 10.3 Hz, J = 9.5 Hz, 2H), 4.79 (t, J =
1
4
)-Zn: H NMR (600 MHz,
C H F FeN O P : C, 52.15; H, 4.14; N, 11.95. Found: C, CD
5
2
47 12
9 4 2
3
5
1.48; H, 3.66; N, 9.98.
7
1
2
Synthesis of 1(BPh )·MeCN·Et O
9.2 Hz, 4H), 4.73–4.66 (m, 4H). Anal. calcd for
4
2
C H Cl ZnN O : C, 55.08; H, 3.82; N, 8.38. Found: C, 54.93;
H, 4.15; N, 7.98.
4
6
38
2
6 12
The procedure is identical to that of 1(ClO
4
) except that Fe
(
ClO ) ·6H O was replaced by FeCl ·4H O and NaBPh . Dark
4
2
2
2
2
4
1
purple cubic crystals (110 mg, yield: 76%) were obtained. H
NMR (600 MHz, CD CN, 298 K): δ 56.01 (br, 2H), 54.19 (br,
H), 25.89 (s, 2H), 12.44 (s, 2H), 10.29 (s, 2H), 9.05 (s, 2H),
.94 (s, 2H), 7.63 (s, 4H), 7.34 (br, 16H), 7.03 (t, J = 7.0 Hz, 16
Computational details
3
2+
The structures of three [Fe(pybox)2] complex cations were fully
optimized without restrictions using the M06 functional with a
2
7
6-31G(d) basis set in the gas phase. Tables S9–S14† list the
H), 6.88 (t, J = 7.0 Hz, 8 H), 6.58 (s, 2H), 6.22 (s, 2H), 5.98 (br,
H), 3.97 (s, 2H), 3.07 (s, 4H), 2.30 (s, 4H). Anal. calcd for
FeN : C, 77.57; H, 5.92; N, 6.33. Found: C, 77.98;
H, 5.56; N, 6.87.
Cartesian coordinates for complexes 1–3 in different spin states.
All calculations were carried out using the Gaussian 09 software.
Caution! Although not encountered in our experiments,
perchlorate salts in the presence of organic ligands are poten-
tially explosive. Only a small amount of the materials should
be prepared and handled with care.
2
C
100
H
91
B
2
7 5
O
Synthesis of 2(ClO )·MeOH
4
S
A solution of L (74 mg, 0.2 mmol) in methanol and CH
2
Cl
2
(v/v, 10 : 1, 6 mL) was placed in the bottom of a test tube,
methanol (10 mL) was gently layered on the top of the solu-
Conflicts of interest
4 2 2
tion, and then a solution of Fe(ClO ) ·6H O (36 mg, 0.1 mmol)
in methanol (6 mL) was carefully added as the top layer. After
There are no conflicts to declare.
a few weeks, dark red block crystals were collected (46 mg,
1
4
3
4
8%). H NMR (600 MHz, CD
3
CN, 298 K): δ 56.42 (br, 4H),
0.03 (br, 4H), 24.96 (s, 2H), 17.52 (s, 4H), 5.03 (s, 4H), 4.66 (s,
H), 3.34 (s, 8H), −3.03 (s, 8H). Anal. calcd for
Acknowledgements
C
47
H
42Cl
2
FeN
6
O
13: C, 55.04; H, 4.13; N, 8.19. Found: C, 55.41; This work is partially supported by the National Natural
H, 3.92; N, 8.56.
Science Foundation of China (Grant No. 21771049, 21421005,
3376 | Dalton Trans., 2021, 50, 3369–3378
This journal is © The Royal Society of Chemistry 2021

View Article Online
Dalton Transactions
Paper
9
1422302, 21429201, 21871039 and 21801037), the State Key
and Applications, John Wiley & Sons Ltd, Chichester, 2013,
pp. 281–301.
Laboratory of Fine Chemicals at Dalian University of
Technology (KF 1607) and the Fundamental Research Funds 21 D. F. Evans, J. Chem. Soc., 1959, 2003–2005.
for the Central Universities of China (Grant No 22 C. Piguet, J. Chem. Educ., 1997, 74, 815.
PA2019GDPK0055 and PA2020GDJQ0028).
23 S. Rodriguez-Jimenez and S. Brooker, Inorg. Chem., 2017,
6, 13697–13708.
5
24 S. Rodriguez-Jimenez, A. S. Barltrop, N. G. White,
H. L. C. Feltham and S. Brooker, Inorg. Chem., 2018, 57,
Notes and references
6266–6282.
1
2
3
P. Gütlich, A. B. Gaspar and Y. Garcia, Beilstein J. Org. 25 M. J. Frisch, G. W. Trucks, H. B. Schlegel, G. E. Scuseria,
Chem., 2013, 9, 342–391.
M. A. Robb, J. R. Cheeseman, G. Scalmani, V. Barone,
B. Mennucci, G. A. Petersson, H. Nakatsuji, M. Caricato,
X. Li, H. P. Hratchian, A. F. Izmaylov, J. Bloino, G. Zheng,
J. L. Sonnenberg, M. Hada, M. Ehara, K. Toyota, R. Fukuda,
J. Hasegawa, M. Ishida, T. Nakajima, Y. Honda, O. Kitao,
H. Nakai, T. Vreven, J. A. Montgomery Jr., J. E. Peralta,
F. Ogliaro, M. Bearpark, J. J. Heyd, E. Brothers,
K. N. Kudin, V. N. Staroverov, R. Kobayashi, J. Normand,
K. Raghavachari, A. Rendell, J. C. Burant, S. S. Iyengar,
J. Tomasi, M. Cossi, N. Rega, J. M. Millam, M. Klene,
J. E. Knox, J. B. Cross, V. Bakken, C. Adamo, J. Jaramillo,
R. Gomperts, R. E. Stratmann, O. Yazyev, A. J. Austin,
R. Cammi, C. Pomelli, J. W. Ochterski, R. L. Martin,
K. Morokuma, V. G. Zakrzewski, G. A. Voth, P. Salvador,
J. J. Dannenberg, S. Dapprich, A. D. Daniels, O. Farkas,
J. B. Foresman, J. V. Ortiz, J. Cioslowski and D. J. Fox,
GAUSSIAN 09, Revision E.01, Gaussian, Inc., Wallingford,
CT, 2009.
M. A. Halcrow, Spin-Crossover Materials: Properties and
Applications, John Wiley & Sons, Chichester, 2013.
P. Gutlich and H. A. Goodwin, in Spin Crossover in
Transition Metal Compounds I, ed. P. Gutlich and
H. A. Goodwin, 2004, vol. 233, pp. 1–47.
4
5
M. A. Halcrow, Chem. Soc. Rev., 2011, 40, 4119–4142.
X.-P. Sun, Z. Tang, Z.-S. Yao and J. Tao, Chem. Commun.,
2
020, 56, 133–136.
6
7
8
9
L. F. Qin, C. Y. Pang, W. K. Han, F. L. Zhang, L. Tian,
Z. G. Gu, X. Ren and Z. Li, Dalton Trans., 2016, 45, 7340–
7
348.
Y. Sekimoto, M. R. Karim, N. Saigo, R. Ohtani,
M. Nakamura and S. Hayami, Eur. J. Inorg. Chem., 2017,
1
049–1053.
J. Ru, F. Yu, P.-P. Shi, C.-Q. Jiao, C.-H. Li, R.-G. Xiong,
T. Liu, M. Kurmoo and J.-L. Zuo, Eur. J. Inorg. Chem., 2017,
3
144–3149.
H. A. Goodwin, in Spin Crossover in Transition Metal 26 Y. Zhao and D. G. Truhlar, Theor. Chem. Acc., 2008, 120,
Compounds I, ed. P. Gutlich and H. A. Goodwin, 2004, vol.
33, pp. 59–90.
0 Y.-Y. Zhu, C.-W. Liu, J. Yin, Z.-S. Meng, Q. Yang, J. Wang,
215–241.
2
27 M. Llunell, D. Casanova, J. Cirera, P. Alemany and
S. Alvarez, SHAPE, version 2.0, Universitat de Barcelona,
Barcelona, Spain, 2010.
1
1
1
T. Liu and S. Gao, Dalton Trans., 2015, 44, 20906–20912.
1 Y.-Y. Zhu, H.-Q. Li, Z.-Y. Ding, X.-J. Lu, L. Zhao, Y.-S. Meng, 28 B. Dey, A. Mondal and S. Konar, Chem. – Asian J., 2020, 15,
T. Liu and S. Gao, Inorg. Chem. Front., 2016, 3, 1624–1636. 1709–1721.
2 Y. Pan, Y.-S. Meng, Q. Liu, W.-Q. Gao, C.-H. Liu, T. Liu and 29 S. Ye and F. Neese, Inorg. Chem., 2010, 49, 772–774.
Y.-Y. Zhu, Inorg. Chem., 2020, 59, 7398–7407.
3 A. Kimura and T. Ishida, Inorganics, 2017, 5, 52.
30 T.-T. Ma, X.-P. Sun, Z.-S. Yao and J. Tao, Inorg. Chem. Front.,
2020, 7, 1196–1204.
1
1
4 K. E. Burrows, S. E. McGrath, R. Kulmaczewski, 31 P. Guionneau, M. Marchivie, G. Bravic, J. F. Létard and
O. Cespedes, S. A. Barrett and M. A. Halcrow, Chem. – Eur.
J., 2017, 23, 9067–9075.
5 W.-Q. Gao, Y.-S. Meng, C.-H. Liu, Y. Pan, T. Liu and
Y.-Y. Zhu, Dalton Trans., 2019, 48, 6323–6327.
D. Chasseau, Top. Curr. Chem., 2004, 234, 785–786.
32 F. Kobayashi, Y. Komatsumaru, R. Akiyoshi, M. Nakamura,
Y. Zhang, L. F. Lindoy and S. Hayami, Inorg. Chem., 2020,
59, 16843–16852.
1
1
6 Z. Y. Li, H. Ohtsu, T. Kojima, J. W. Dai, T. Yoshida, 33 X. H. Zhao, S. L. Zhang, D. Shao and X. Y. Wang, Inorg.
B. K. Breedlove, W. X. Zhang, H. Iguchi, O. Sato, Chem., 2015, 54, 7857–7867.
M. Kawano and M. Yamashita, Angew. Chem., Int. Ed., 2016, 34 D. Shao, L. Shi, F. X. Shen, X. Q. Wei, O. Sato and
5, 5184–5189. X. Y. Wang, Inorg. Chem., 2019, 58, 11589–11598.
7 K. E. Burrows, R. Kulmaczewski, O. Cespedes, S. A. Barrett 35 C. Rajadurai, Z. Qu, O. Fuhr, B. Gopalan, R. Kruk,
5
1
1
and M. A. Halcrow, Polyhedron, 2018, 149, 134–141.
M. Ghafari and M. Ruben, Dalton Trans., 2007, 3531–
8 L. J. Kershaw Cook, R. Kulmaczewski, R. Mohammed,
3537.
S. Dudley, S. A. Barrett, M. A. Little, R. J. Deeth and 36 M. Nihei, L. Han and H. Oshio, J. Am. Chem. Soc., 2007,
M. A. Halcrow, Angew. Chem., Int. Ed., 2016, 55, 4327–4331. 129, 5312–5313.
9 S. Rodriguez-Jimenez, M. Yang, I. Stewart, A. L. Garden and 37 X. P. Sun, R. J. Wei, Z. S. Yao and J. Tao, Cryst. Growth Des.,
S. Brooker, J. Am. Chem. Soc., 2017, 139, 18392–18396. 2018, 18, 6853–6862.
0 M. P. Shores, C. M. Klug and S. R. Fiedler, in Spin-state 38 I. Prat, A. Company, T. Corona, T. Parella, X. Ribas and
switching in solution. Spin-Crossover Materials: Properties M. Costas, Inorg. Chem., 2013, 52, 9229–9244.
1
2
This journal is © The Royal Society of Chemistry 2021
Dalton Trans., 2021, 50, 3369–3378 | 3377

View Article Online
Paper
Dalton Transactions
3
9 R. Kulmaczewski, F. Bamiduro, N. Shahid, O. Cespedes and 41 S.-G. Wu, M. N. Hoque, J.-Y. Zheng, G.-Z. Huang, N. V. Ha
M. A. Halcrow, Chem.
092.
–
Eur. J., 2021, 27, 2082–
Anh, L. Ungur, W.-X. Zhang, Z.-P. Ni and M.-L. Tong, CCS
Chem., 2020, 453–459.
2
4
0 S. Q. Su, S. Q. Wu, M. L. Baker, P. Bencok, N. Azuma, 42 M. A. Halcrow, I. Capel Berdiell, C. M. Pask and
Y. Miyazaki, M. Nakano, S. Kang, Y. Shiota, K. Yoshizawa, R. Kulmaczewski, Inorg. Chem., 2019, 58, 9811–9821.
S. Kanegawa and O. Sato, J. Am. Chem. Soc., 2020, 142, 43 P. Drabina, P. Funk, A. Růžička, J. Hanusek and M. Sedlák,
1434–11441. Transition Met. Chem., 2010, 35, 363–371.
1
3378 | Dalton Trans., 2021, 50, 3369–3378
This journal is © The Royal Society of Chemistry 2021