Spin Crossover vs. High-Spin Iron(II) Complexes in N4S2 Coordination Sphere Containing Picolyl-Thioether Ligands and NCE (E=S, Se and BH3) Co-Ligands

Article abstract of DOI:10.1002/ejic.202100355

To study the effect of the chelate ring size on the magnetic properties of thioether-based iron(II) metal complexes, two ligands have been envisaged, synthesised and characterised. The two ligands correspond to the bidentate benzylpicolylthioether (PySBn) and tetradentate 2,3-bis(((2-pyridylmethyl)thio)methyl)quinoxaline (QuinoxS). Five iron(II) complexes have been synthesised, containing either two bidentate ligands or one tetradentate ligand, and two N-bond NCE co-ligands (E=S, Se or BH₃): trans-[Fe^II(PySBn)₂(NCE)₂] (1 a–b) and cis-[Fe^II(QuinoxS)(NCE)₂] (2 a–c), a for E=S, b for E=Se and c for E=BH₃. The iron(II) complexes have been characterised by standard techniques, X-ray crystallography (except for complex 1 a) and VT-magnetic measurements in the solid state. X-ray crystallography showed that all the complexes are isolated in the high spin (HS) state, based on the relatively long Fe?L bond lengths, Fe?N>2.0 ? and Fe?S≈2.5–2.6 ?. VT-magnetic measurements demonstrated that complexes 1 a and 2 a-c are stabilised in the HS-state, showing orbital contribution to g and zero field splitting. However, complexes 1 b shows a relatively abrupt, hysteretic, and incomplete at the low-end spin conversion, with T_1/2↓=92, T_1/2↑=98 K and ΔT_1/2=6 K at 5 K min^?1, moreover, the hysteresis loop is scan rate dependent increasing up to 11 K at 10 K min^?1. An analysis of structural and electronic parameters has been performed to rationalise the differing magnetic properties of the metal complexes, such as metallacycle size, bond lengths and angles, and cis- vs. trans-coordination mode. A comparison with the literature-reported spin crossover iron(II) complexes in N₄S₂ coordination sphere containing NCE co-ligands has been conducted as well, finding that, as previously reported, the Fe?N?C(E) bond angle is diagnostic for determining the spin lability of the metal complexes, and in addition we have found that the N?C(E) bond length is too useful.

Full text of DOI:10.1002/ejic.202100355

European Journal of Inorganic Chemistry
Accepted Article
Title: Spin crossover vs. high-spin iron(II) complexes in N4S2
coordination sphere containing picolyl-thioether ligands and NCE
(E = S, Se and BH3) co-ligands
Authors: Diego Plaza-Lozano, Agustín Conde-Gallardo, and Juan
Olguin
This manuscript has been accepted after peer review and appears as an
Accepted Article online prior to editing, proofing, and formal publication
of the final Version of Record (VoR). This work is currently citable by
using the Digital Object Identifier (DOI) given below. The VoR will be
published online in Early View as soon as possible and may be different
to this Accepted Article as a result of editing. Readers should obtain
the VoR from the journal website shown below when it is published
to ensure accuracy of information. The authors are responsible for the
content of this Accepted Article.
To be cited as: Eur. J. Inorg. Chem. 10.1002/ejic.202100355
Link to VoR: https://doi.org/10.1002/ejic.202100355
01/2020

10.1002/ejic.202100355
European Journal of Inorganic Chemistry
FULL PAPER
Spin crossover vs. high-spin iron(II) complexes in N4S2
coordination sphere containing picolyl-thioether ligands and NCE
(E = S, Se and BH3) co-ligands
Diego Plaza-Lozano[a], Agustín Conde-Gallardo[b] and Juan Olguín*[a]
[a]
D. Plaza-Lozano and Dr. J. Olguín
Departamento de Química
Centro de Investigación y de Estudios Avanzados del IPN (Cinvestav)
Avenida IPN 2508, Col. San Pedro Zacatenco, Ciudad de México 07360, México
E-mail: jolguin@cinvestav.mx.
[b]
Dr. A. Conde-Gallardo
Departamento de Física
Centro de Investigación y de Estudios Avanzados del IPN (Cinvestav)
Avenida IPN 2508, Col. San Pedro Zacatenco, Ciudad de México 07360, México
Supporting information for this article is given via a link at the end of the document.
Abstract: To study the effect of the chelate ring size on the
magnetic properties of thioether-based iron(II) metal complexes, two
ligands have been envisaged, synthesised and characterised. The
two ligands correspond to the bidentate benzylpicolylthioether
possessing d4 to d7 electronic configurations, mostly in
octahedral geometries.[1–3] Due to the differences in the
physicochemical properties of the HS and LS-states, this type of
materials can be used as molecular switches and sensors,[4–6]
and when the spin transition is accompanied by hysteresis these
materials could be used as nanodevices for information
storage.[5,7–12]
The most widely studied SCO system is iron(II) complexes
where the metal centre is surrounded by six nitrogen donors,
however there are fewer examples of other metal centres and
geometries.[13–16] In addition to the N6 coordination sphere,
recently, it has been demonstrated that N4S2 systems can
produce iron(II) SCO-active complexes, with only a handful of
ligands that can provide the right ligand field for SCO to
occur,[17–23] Figure 1. All the S-based ligands contain the same
thioether functional group, in combination with amine, pyridine,
triazole or pyrazole nitrogen donors, and some of them contain
NCE (E = S, Se or BH3) co-ligands. An advantage of
synthesizing S-containing iron(II) complexes is the protection
toward oxidation of the metal centre, as it has been shown that
some S-based ligands are redox non-innocent,[24–27] preventing
the metal centre to oxidize to iron(III), thus ensuring the
preservation of the SCO-properties.
(PySBn)
and
tetradentate
2,3-bis(((2-
pyridylmethyl)thio)methyl)quinoxaline (QuinoxS). Five iron(II)
complexes have been synthesised, containing either two bidentate
ligands or one tetradentate ligand, and two N-bond NCE co-ligands
(E = S, Se or BH3): trans-[FeII(PySBn)2(NCE)2] (1a-b) and cis-
[FeII(QuinoxS)(NCE)2] (2a-c), a for E= S, b for E = Se and c for E =
BH3. The iron(II) complexes have been characterised by standard
techniques, X-ray crystallography (except for complex 1a) and VT-
magnetic measurements in the solid state. X-ray crystallography
showed that all the complexes are isolated in the high spin (HS)
state, based on the relatively long Fe-L bond lengths, Fe-N > 2.0 Å
and Fe-S  2.5-2.6 Å. VT-magnetic measurements demonstrated
that complexes 1a and 2a-c are stabilised in the HS-state, showing
orbital contribution to g and zero field splitting. However, complexes
1b shows a relatively abrupt, hysteretic, and incomplete at the low-
end spin conversion, with T1/2 = 92, T1/2 = 98 K and T1/2 = 6 K at
5 K min-1, moreover, the hysteresis loop is scan rate dependent
increasing up to 11 K at 10 K min-1. An analysis of structural and
electronic parameters has been performed to rationalise the differing
magnetic properties of the metal complexes, such as metallacycle
size, bond lengths and angles, and cis- vs. trans-coordination mode.
A comparison with the literature-reported spin crossover iron(II)
complexes in N4S2 coordination sphere containing NCE co-ligands
has been conducted as well, finding that, as previously reported, the
Fe-N-C(E) bond angle is diagnostic for determining the spin lability
of the metal complexes, and in addition we have found that the N—
C(E) bond length is too useful.
Introduction
The spin crossover (SCO) effect, that is the reversible
interconversion between the low- and high-spin state (LS and
HS, respectively) by means of an external stimulus (e.g.,
temperature, light, magnetic field, solvation/desolvation), can be
observed in some first row transition metal complexes
Figure 1. All seven ligands feature thioether donors capable to provide SCO-
active FeII complexes.
1
This article is protected by copyright. All rights reserved.

10.1002/ejic.202100355
European Journal of Inorganic Chemistry
FULL PAPER
To expand the number and type of ligands that can produce
SCO-active iron(II) complexes we have synthesised two ligands
containing pyridyl and thioether moieties, Figure 2, inspired by
the bpte[22] and bptPh[23] ligands (Figure 1). The first ligand is
the bidentate picolylbenzylthioether (PySBn), to evaluate if such
a simple ligand can produce SCO active complexes as it
contains both S and N donors, and upon coordination to iron(II),
five-membered chelate rings are exclusively formed (Figure 2),
as well as in the literature reported complexes of bpte and
bptPh ligands. The second ligand is the tetradentate 2,3-bis(((2-
but in the case of QuinoxS-based complexes they are found in
a cis-arrangement, whereas in the PySBn-complexes are found
in a trans-manner. The effect of chelate ring size on the
magnetic properties of the complexes synthesised in this work
and the SCO-active complexes reported in the literature has
been evaluated (Figure 2), and an analysis of the structural
factors affecting the magnetic properties, such as Fe−L bond
lengths and angles is presented as well.
pyridylmethyl)thio)methyl)quinoxaline
(QuinoxS),
that
Results and Discussion
structurally resembles to bptPh, however, the phenyl ring has
been replaced by a quinoxaline heterocycle, and an extra
methylene group has been incorporated in between the aromatic
and sulphur groups, thus the resulting metal complex will contain
a seven-membered chelate ring, Figure 2, instead of a five-
membered ring as in the bptPh-based complex. In addition, the
fused heterocycle quinoxaline in QuinoxS could promote -
interactions, and H-bonds due to the spare sp2 nitrogen atoms,
such supramolecular interactions are desired to be present in
iron(II) complexes to promote cooperativity in the system,
increasing the odds of obtaining SCO systems.
Synthesis of the ligands
The PySBn ligand was synthesised according to the published
method with slight modifications,[29] Figure 3, by reacting an
equimolar mixture of benzyl chloride, sodium hydroxide and
picolylthiol[30] in refluxing ethanol. The ligand PySBn was
obtained in a 91 % yield as a red oil, which crystallised at 0-5 °C,
and it was used without further purification. The ligand QuinoxS
was synthesised according to the procedure reported in the
literature,[31] Figure 3, as a dark beige solid in 86 % yield. All the
ligands were characterised by 1H- and 13C{1H}-NMR, IR and
mass spectrometry, confirming the proposed structures, see
supporting information.
Figure 3. Synthetic procedure for the obtention of PySBn and QuinoxS ligands.
Synthesis of the metal complexes
The family of complexes 1 were obtained by reaction of one
equivalent of in situ synthesised [FeII(Solv.)4(NCE)2] (E = S or
Se), with two equivalents of PySBn, in an adequate solvent
under Ar, Figure 4. It is important to mention that the synthesis
of the E = BH3 analogue for PySBn was attempted, however it
was not possible to obtain a clean sample, and due to COVID-
19 pandemic, we have not been able to access research
facilities in Mexico since the begging of the health emergency.
Whereas, the family of complexes 2 was obtained by treatment
of the corresponding trans-[FeII(Py)4(NCE)2][32] (E = S, Se or
BH3) complex with one equivalent of QuinoxS ligand, in an
adequate solvent under Ar, Figure 4.
To obtain crystals suitable for X-ray crystallography, in most
cases, the complexes were obtained by slow diffusion of a
solution of the precursor metal complex layered over a solution
of the ligand in an adequate solvent (see experimental section),
after slow diffusion, crystals suitable for X-ray diffraction were
obtained in moderate or good yield, Table 1. However, in the
case of 1b the filtered reaction solution was subjected to diethyl
ether vapor diffusion in air at room temperature, and after 24-48
h crystals suitable for X-ray crystallography were obtained.
Figure 2. Top: structure of the ligands used in this work. Bottom: structure of
the complexes synthesised in this work and the literature reported SCO-active
complex, highlighting the chelate ring size.
A series of five metal complexes of the type [FeII(L)n(NCE)2], n =
1 and E = S, Se or BH3, for QuinoxS, whereas n = 2 and E = S
or Se, for PySBn, have been synthesised and characterised by
standard methods and magnetometry. In addition, single crystal
X-ray crystallography has been used for the characterisation of
most of the complexes. The use of NCE anions is very common
to produce SCO-complexes, since the crystal field can be
increased in the order S < Se < BH3.[28] Thus, two types of metal
complexes in N4S2 octahedral geometry have been successfully
obtained and magnetically characterised. One class of
complexes contain the metal centre surrounded by two bidentate
ligands, where each ligand strand encompasses a pyridyl unit as
N-donor, and a thioether group as S-donor. The second type
contains one tetradentate ligand strand, comprising
picolylthioether pendant arms. For both types of complexes two
NCE co-ligands complete the octahedral coordination sphere,
2
This article is protected by copyright. All rights reserved.

10.1002/ejic.202100355
European Journal of Inorganic Chemistry
FULL PAPER
room temperature data set was not further pursued. Relevant
bond lengths and angles, and single crystal data and structure
refinement parameters for the iron(II) complexes are shown in
Tables 2 and S1-S5 (CCDC: 2005424, 2005425, 2005427,
2005430, 2005431, 2005432 and 2005433). Complex 1b
crystallised in the C2/c space group, Figure 5 and S16. Complex
2a crystallised in the P21/c space group, Figure 6, while
complexes 2b and 2c are isostructural and isomorphs,
crystallizing in the
space group, Figures 6 and S17. Except
for complexes 2b and 2c, all of them are solvent free. The
asymmetric unit of complex 1b comprises half complex molecule.
The asymmetric unit of complex 2a comprises one complex
molecule at both acquisition temperatures (298(2) and 177(2) K),
whilst 2b shows two molecules independent by symmetry and
half acetonitrile molecule solvent at both acquisition
temperatures 298(2) and 177(2) K. However, at both
temperatures the crystal structure of 2b showed disordered
solvent molecules that could not be appropriately modelled, thus
the SQUEEZE[37] routine in PLATON was performed, removing
their contribution to the overall intensity data. An improvement
was observed in the final refinement parameters and the
electron density removed was in accordance with acetonitrile
molecules (see supporting information), thus the formula from X-
ray at both temperatures is 2b1¼MeCN. Complex 2c also
shows two molecules independent by symmetry and one and a
quarter acetonitrile molecule at 298(2) K, but two complex
molecules independent by symmetry and half acetonitrile
molecule at 175(2) K, suggesting a loss of solvent, however,
there was disordered solvent molecules that could not be
appropriately modelled at 175(2) K, thus the SQUEEZE[37]
routine in PLATON was performed, removing their contribution
to the overall intensity data. An improvement was observed in
the final refinement parameters and the electron density
removed was in accordance with acetonitrile molecules (see
supporting information), thus the formula for this complex is
2c1¼MeCN at both temperatures.
Figure 4. Synthetic methodologies for the obtention of complex families 1 and
2.
Table 1. Crystallisation method, yield, and relevant IR parameters for complex
families 1 and 2.
 NCE[c]
Crystallisation
Yield (%)
27 (80)[a]
20 (15)[a]
 B-H (cm-1)
Complex
1a
(cm-1)
method
-
2059
-
-
Et2O vapor
diffusion
1b
2059
2a
2b
2c
Slow diffusion
Slow diffusion
Slow diffusion
41
57
45
2070, 2053
2061, 2040
2179
-
-
2349, 2321
[a] Yields in parenthesis were obtained from the reaction using precursor
complex trans-[FeII(Py)4(NCE)2] E = S or Se, see Experimental Section.
The metal complexes were characterised by standard
techniques, single crystal X-ray crystallography, except for
complex 1a, and magnetic measurements. In Table 1 a
summary of diagnostic IR vibrations is shown, demonstrating
that in all cases the NCE co-ligands are coordinated through the
N-atom.[33–36] In the case of the symmetric trans complexes (1a-
b), there is only one  N≡C(E) (E = S, Se) band in the IR spectra,
expected for this type of isomers, whereas for the cis complexes
2a and 2b there are two bands, but for complex 2c only one
band is observed.
X-ray crystallography
Single crystal X-ray diffraction data set was acquired at room
and low temperature for all complexes on the same crystal,
except for complex 1b which was collected only at 199 K, due to
a technical difficulty on the diffractometer cooling system, and
due to the stabilisation of the HS-state at such temperature the
Figure 5. Perspective view of complex trans-[FeII(PySBn)2(NCSe)2] (1b)
acquired at 199(2) K. Ellipsoids are drawn at the 50 % probability level.
Hydrogen atoms are omitted for the sake of clarity.
3
This article is protected by copyright. All rights reserved.

10.1002/ejic.202100355
European Journal of Inorganic Chemistry
FULL PAPER
Figure 6. Perspective view of the family complex cis-[FeII(QuinoxS)(NCE)2] E = S, Se and BH3; complexes (2a), (2b) and (2c) acquired at 177(2) K, 177(2) K and
175(2) K; left, centre and right respectively. Ellipsoids are drawn at the 50 % probability level. Hydrogen atoms and acetonitrile solvent molecules are omitted for
the sake of clarity.
metallocycles upon coordination. On the other hand, the
complexes containing the tetradentate ligand QuinoxS, 2a-c,
also show a distorted N4S2 octahedral geometry for the iron(II)
centre, but the NCE ligands are found in a cis coordination mode,
Table 2. Average bond distances (Å) and angles (°) and ΣOh parameter for the
complex families 1 and 2.
instead of trans as 1b. Due to the multidentate nature of the
Complex
Temperature (K)
Space group
1b
2a
2b
2c
ligand, it is possible to identify a seven-membered and two five-
membered metallocycles. There is no difference in the structural
parameters found for complexes 2a-c acquired at room and low
temperature, suggesting that these complexes are not SCO-
active.
199(2)
177(2)
177(2)
175(2)
C2/c
P21/c
P-1
P-1
Av. Fe-NPy (Å)
Av. Fe-NNCE (Å)
Av. Fe-S (Å)
2.182(4)
2.105(4)
2.6004(13)
180.00
2.177(4)
2.094(4)
2.5964(13)
158.70(10)
2.094(8)
2.091(10)
2.6044(32)
156.59(23)
2.127(6)
2.107(7)
2.5899(23)
158.87(17)
Based on the data shown in Table 2, it is possible to observe
that the structural data of the iron(II) complexes agree with the
parameters found for related iron(II) complexes, cis-
[FeII(bpte)2(NCE)2],
bpte
=
S,S′-bis(2-pyridylmethyl)-1,2-
thioethane, E = S, Se and BH3,[20,22] Table S7. The shortest bond
type for all complexes comprises the Fe-NNCE bond, as expected
due to the anionic nature of the NCE co-ligands. The Fe-N bond
length values agree for the high spin state stabilisation of the
iron centre in all complexes (> 2.0 Å).[40] Additionally, as
expected, the Fe-S bonds are the longest bond type for all
complexes.
trans-NPy-Fe-NPy (°)
trans-NNCE-Fe-NNCE
180.00
-
-
-
-
(°)
Av. trans-S-M-NNCE
172.505(13)
173.20(29)
172.00(20)
(°)
The cis and trans bond angles for the thioether-complexes, 1b
and 2a-c, are dispersed and away from the ideal values, 90 and
180 ° respectively, however, for complex 1b the trans angle
values are 180 °, which could be due to the all-trans geometry
found for this complex and the fact that only a half complex
molecule is found in the asymmetric unit, thus generating the
trans-S-Fe-S (°)
Av. cis-N-Fe-N (°)
Av. cis-S-Fe-S (°)
Av. cis-N-Fe-S (°)
Av. Fe-N-C(E) (°)
180.00
90.00(21)
-
-
-
-
96.56(26)
99.28(3)
83.35(20)
161.1
96.96(54)
101.37(7)
82.81(43)
163.4
95.99(38)
101.81(6)
83.62(30)
166.5
90.00(23)
155.30
81.02(44)
[a]
other half by symmetry. The cis angles are found in a 78.1
—
101.9 and 77.8 102.7 ° range, for 1b and 2a-c, respectively.
—
The trans angles for complexes 2a-c range from 154.7 to
176.7 °. The more distorted trans angles correspond to the NPy-
[a]
ΣOh
82.00(33)
89.32(69)
80.06(67)
Fe-Npy angle, 154.7
169.0
176.7 °. The octahedral distortion parameter[38,39] (Oh
values are found within the 80.06 91.58 ° range (see Tables 2
—159.9 °, whilst the NNCE-Fe-S range from
Av. = average, Py = pyridine,
defined as the sum of the absolute values of the difference of each of the
twelve cis angles from 90°. ΣOh
geometry.[38,39]
The octahedral distortion parameter, ΣOh, is
—
)
—
= 0 indicates a perfect octahedral
and S5) which is in perfect agreement with the high-spin state[40]
assignment based on the Fe-L bond lengths.
In the crystal structures of 1b it is possible to observe that the
iron(II) centre is bonded to two PySBn bidentate ligands, in an
all trans fashion, that is all the identical donors are found trans to
each other, where the NCSe co-ligands are found in apical
positions, resulting in a distorted N4S2 octahedral geometry.
Moreover, the bidentate ligand strands produce five-membered
Finally, an analysis of the supramolecular interactions present in
all the crystal structures reported in this work can be found in the
supporting information, see page S19, that correspond to -
interactions CH, and CHE bonds.
4
This article is protected by copyright. All rights reserved.

10.1002/ejic.202100355
European Journal of Inorganic Chemistry
FULL PAPER
Magnetic properties
value ranging from 4.9 to 5.5 B, although for the majority of
complexes the value is higher than the expected S.O. = 4.9 B, it
is still in good agreement with the high spin state and suggests a
considerable orbital contribution to g,[22,27,41] in addition the
values are in agreement with the literature.[20,42–44] The room
temperature eff value for complex 1a containing the PySBn
ligand (Figure 7) indicates that it is stabilized in the HS-state, the
eff value is maintained relatively constant down to 25 K, below
this temperature zero field splitting (ZFS) is evident. Finally, for
the complexes containing QuinoxS ligand (2a-c, Figure 8), eff
ranges from 4.9-5.6 B, observing a decrease below c.a. 25 K
due to ZFS.
All complexes were characterised by VT-magnetic studies in the
solid state under a 0.1 T external magnetic field, in the 300 to 5
K temperature range at a 5 K min-1 scan rate. The magnetic
properties were acquired using either crystalline or powder
samples. The crystalline samples, same sample used for SC-
XRD, correspond to complexes 2a-c, whereas the powder
sample correspond to complex 1a-b, either because of a small
quantity was obtained from crystallization or no crystals could be
grown at all. In Figures 7 and 8 the effective magnetic moment
(eff) vs. temperature (K) profile is shown, and in Table 3 a
summary of the eff values at room temperature and 50 K is
presented.
Table 3. Effective magnetic moment for all metal complexes at 300 and 50 K,
and spin state assignment.
Complex

eff at 300 K
(B)

eff at 50 K
(B)
Comments
High spin
1a
1b
5.5
5.2
5.4 / 5.4 
2.3  / 2.4 
Relatively abrupt, incomplete
and hysteretic spin crossover
2a
2b
2c
5.6
4.9
5.1
5.1
4.7
4.6
High spin
High spin
High spin
Complex 1b is the only one showing SCO behaviour, Figure 7.
Complex 1b is stabilised in the pure HS-state at 300 K with 5.4
B, this value is relatively constant down to 150 K. A relatively
abrupt decrease of the magnetic moment is observed below 140
K, reaching a plateau of 2.3-2.4 B between 50
—20 K, that is a
Figure 7. VT-magnetic behaviour for complexes 1a and 1b in the 300-5 K
temperature range at 0.1 T external magnetic field and a 5 K min-1 scan rate.
The cooling (blue ) and warming (red ) behaviour for 1b is shown.
c.a. 20 % of the iron centres are locked in the HS-state,
considering the value of 5.4 B at 300 K for the pure HS-state
and 0 B for the LS-state,[45] indicating an incomplete transition in
the lower temperature range. Below 20 K a sharp decrease is
observed due to ZFS of the remaining HS-state. A cooling
transition temperature (T1/2) of 92 K can be obtained from the
first derivative of the spin transition curve. Upon warming, a
similar behaviour is observed, however, the warming transition
temperature (T1/2) is 98 K, that is a hysteresis loop width
(T1/2) of 6 K is observed. To ensure that the hysteresis loop is
real, the magnetic profile was acquired at 1, 5 and 10 K min-1
rate, Table 4 and Figures S23 and S24. At the slowest scan
rates, 1 and 5 K min-1, the hysteresis loop is not affected, T1/2
=
5 and 6 K, respectively, but at 10 K min-1 the width increases to
11 K, in good agreement with the literature showing that at the
fastest scan rates the hysteresis loop should be the widest,[46–48]
therefore confirming the hysteretic profile.
Table 4. Hysteresis loop scan rate dependence for complex 1b.
Scan rate (K min-1)
T
1/2 (K)[a]+
T
1/2 (K )[a]
98
T1/2 (K)
Figure 8. VT-magnetic behaviour for complexes 2a-c in the 300-5
K
temperature range at 0.1 T external magnetic field and a 5 K min-1 scan rate.
1
5
93
92
91
5
6
98
All metal complexes are stabilised in the HS-state along the
entire experimental temperature range, except for complex 1b
which shows spin crossover behaviour (see below). At room
temperature the complexes stabilized in the HS-state show a eff
10
102
11
[a] Obtained from the first derivative of the spin transition curve.
5
This article is protected by copyright. All rights reserved.

10.1002/ejic.202100355
European Journal of Inorganic Chemistry
FULL PAPER
Discussion
complexes show Fe-N-C(E) angles closer to 180 °. It is
important to note that there is a small number of SCO-active
complexes possessing N4S2 coordination sphere and containing
NCE co-ligands,[20,22] however, an analysis of the Fe-N-C(E) (E =
S, Se or BH3) bond angle range obtained from X-ray
crystallography for the complexes reported in this work and
those reported in the literature for the SCO-active complexes in
the HS-state varies as follows (Figure 10): for the SCO-active
complexes the experimental ranges observed are 165.9 ° for S
(only one complex has been reported), whereas for Se and
The differing magnetic behaviour of the metal complexes can be
explained based on the electronic and structural differences
among them. Two main differences among these complexes can
be observed, the metallacycle size and the positioning of the co-
ligands, cis vs. trans. The coordination mode of the NCE co-
ligands seems not to have a direct influence on the magnetic
properties of the complexes. In the previously reported SCO-
active iron(II) complexes in N4S2 octahedral geometry, it was
found that they contain such co-ligands in a cis-fashion,[20,22]
whilst the SCO-active complexes reported in this work are found
in a trans-mode. The complexes containing the QuinoxS ligand
(2a-c) comprise two five-membered and one seven-membered
metallacycles, in addition an electron withdrawing quinoxaline
moiety is present. These structural properties result in a weak
field ligand, and HS-only iron(II) complexes. Nonetheless, one of
the two complexes based on PySBn ligand is SCO-active, 1b.
The presence of five-membered metallacycles and a good -
acid thioether ligand, that can stabilise the LS-state due to d—
d back bonding,[49–51] permit the modulation of the crystal field
by co-ligands and conferring a stronger field (NCE, E = Se),
producing in this way a SCO-active compound. Complex 1b
shows a relatively abrupt hysteretic SCO.
The sample used for the magnetic measurements for the SCO-
active complex was in powder form, obtained by precipitation
from the reaction mixture, although it was possible to obtain
crystals for complex 1b, the amount was only enough for SC-
XRD. Therefore, the hysteretic magnetic profile for 1b is
surprising, because, as it is known, the systems possessing high
cooperativity within the crystal structure are more likely to show
hysteresis, as in most of the crystalline samples; while in
powders the supramolecular interactions responsible for the
cooperativity may be debilitated.[47,52,53] Although, the sample
used for the magnetic measurements is different from the
crystals used for SC-XRD, it is interestingly to note that the
supramolecular interactions present in the crystal structure at
199(2) K are actually medium to weak slipped - stackings and
non-classical C-HSe bonds, Figure S18. The former interaction
occurs between a pyridine ring of a complex molecule and the
pyridine ring of a neighbouring molecule, where the aromatic
rings are almost coplanar with an angle between the plane
formed by both pyridine rings of 3.6 °, whereas the distance
between both centroids is 3.64 Å. The second type of interaction
occurs between the para carbon of the benzyl group of a
complex molecule and the Se-CN of a neighbouring molecule, a
CSe distance of 3.92 Å is found. The latter molecule shows the
same interaction with a neighbouring molecule, but interacting
through the benzyl group, producing a 1D supramolecular
polymer that is intertwined with neighbouring chains by the
above mentioned pyridyl-pyridyl - interactions forming 2D
sheets, thus is highly likely that the powder and crystalline
samples possess the same interactions that promotes
cooperativity and thus hysteresis in the magnetic profile.
BH3 the ranges are 155.3
complex), respectively. In the case of the HS-only complexes
the angle range at room temperature is 152.0 170.2, 150.0
168.6 and 161.4 172.7 ° for S, Se and BH3, respectively. Due
—176.3 and 161.9 ° (only one
—
—
—
to the small number of examples is difficult to find a trend,
particularly for the S- and BH3-based complexes, which only one
complex is SCO-active of each. Nonetheless, for the Se-based
complexes the angle range for the SCO-active complexes is
closer to linearity than the HS-only complexes that show smaller
angles.
In addition, we have analysed other structural features closely
related to the Fe-N-C(E) angle such as the (N)C—E and N—
C(E) bond lengths and the N-C-E angle of the co-ligands, Figure
10 and Tables S8 and S9. The ranges for the (N)C—E and N—
C(E) bond lengths for the SCO-active complexes, in the HS
state, corresponding to the S, Se and BH3 anions are 1.618
(only one complex), 1.765
—1.803 and 1.583 Å (only one
complex), and 1.158 (only one complex), 1.108
—
1.165 and
1.141 (only one complex) Å, respectively. The range for the
(N)C—E bond lengths in the case of the HS-only complexes, at
room temperature, show a range of 1.622
1.564 1.577 Å, whilst the N—C(E) bond length varies 1.097
1.129, 1.093 1.166 and 1.121 1.134 Å for S, Se and BH3,
—1.648, 1.713—1.799,
—
—
—
—
respectively. Whereas the N-C-E angle of the co-ligands for the
SCO-active complexes, in the HS-state, range is as follow:
179.6 (only one complex), 177.2
complex), and the range for HS-only complexes, at room
temperature, are 178.2 178.6, 176.9 179.4 and 176.7—
—179.3 and 177.5 ° (only one
—
—
179.0 ° for S, Se and BH3, respectively. In the case of the
(N)C—E bond length, there is no clear difference between the
SCO-active and the HS-only complexes. In the case of the N—
C(E) bond length value ranges, at least for the S and BH3-based
complexes, indicate that in the SCO-active complexes the N—C
bond elongates in comparison to the HS-only complexes,
suggesting a much more electron density donation towards the
metal centre, debilitating the N—C bond in SCO-active
complexes. The N-C-E angle trend is like the one found for Fe-
N-C(E), that is the angle in the SCO-active complexes is closer
to linearity than the HS-only complexes, except for the NCBH3
anion that have smaller angle values. Thus, from these results it
seems that the more diagnostic parameters to evaluate spin
lability are the Fe-N-C(E) angle and N—C(E) bond length values,
Figure 9. A plot of these two parameters shows that the N—C(E)
bond length value is a better marker for determining the
probability of observing the SCO effect, as the complexes with
longer values are SCO-active. From Figure 9 it is possible to
observe that 10 out of 14 complexes that have a bond length
larger than 1.14 Å are SCO-active, that is a 71 %. It is important
to mention that there are just a few of SCO-active complexes,
compared to HS-only, which could hamper an in depth analysis
and to stablish an accurate bond length and angle range for
Brooker and co-workers[36] have shown that the Fe-N-C(E) (E =
S, Se or BH3) bond angle is diagnostic for determining the spin
lability of the metal complexes for a family of structurally
characterised iron(II) complexes containing 3-pyridyl-4-
substituted-1,2,4-triazoly ligands (Rdpt) and NCE co-ligands in
N6 coordination sphere. The SCO-active complexes showed an
angle range of 162
locked complexes is 142
—
178 °, whilst the range found for the HS-
159 °. It is clear that the SCO-active
—
6
This article is protected by copyright. All rights reserved.

10.1002/ejic.202100355
European Journal of Inorganic Chemistry
FULL PAPER
determining spin lability of an iron(II) complex based on
structural parameters for this type of S-containing complexes.
produce SCO-active complexes in combination with a right co-
ligand. The NCSe-based complex 1b shows a relatively abrupt
and hysteretic SCO profile, whereas the magnetic profile for the
NCS complex 1a is in agreement for a complex stabilised in the
HS-state. In contrast, the tetra-dentate ligand QuinoxS
produced HS-only complexes, being the main differences with
the PySBn-based complexes the combination of five- and
seven-membered rings present in the complexes of the former
ligand, and the electron withdrawing quinoxaline heterocycle.
Another important difference is the disposition of the NCE co-
ligands, for the QuinoxS complexes the co-ligands are in a cis-
manner but in the PySBn complexes the disposition is trans,
however both types of complexes can be SCO-active, for
example the family cis-[FeII(bpte)(NCE)2] and complex trans-1b.
An analysis of the structural parameters of the complexes
reported in this work and literature-based SCO-active complexes
with N4S2 coordination spheres containing NCE co-ligands
demonstrated, as previously reported, that the Fe-N-C(E) bond
angle is a suitable parameter for determining the potential SCO-
character of a complex, as in the SCO-active complexes such
angle is usually almost linear, close to 180 °. In addition, we
have demonstrated that the N—C(E) bond length is as well a
suitable parameter when considering the feasibility of SCO, as in
SCO-active complexes such bond is longer than for HS-only
complexes, N—C(E) bond > 1.14 Å.
Figure 9. Fe-N-C(E) angle vs. N—C(E) bond length values plot. The blue
dotted box highlights the area where all the SCO-active complexes can be
found, and the green solid box shows the highest density of SCO-active
complexes showing higher Fe-N-C(E) angles and N—C(E) bond length values
than the HS-only complexes. The SCO-active complexes are shown in purple
symbols, and the HS-only complexes are shown in red symbols.
Finally, we have shown that readily available and simple
thioether based complexes can produce SCO-active complexes
with hysteresis, the latter property is actually not easy to achieve,
and in our case, it is observed in powder materials, which is
important due to the difficulty of obtaining crystalline materials.
Therefore, our work has expanded the ligand repertoire of S-
based ligands that produce SCO-active materials, that can be
easily modified to tune the magnetic properties by introducing
different N-heterocycles on one arm and electron-
donating/withdrawing substituents on the other arm of the
thioether group, which can positively impact the magnetic
properties.
Conclusion
The ligands containing thioether functional groups can produce
SCO-active complexes, as previously reported by other authors.
However, we have shown that the chelate ring size and the
selection of NCE co-ligands are important for tuning the crystal
field adequately but are not the only factors. The packing
interactions in the resulting metal complexes is quite important
as the presence/absence of solvent molecules can promote or
deactivate SCO, polymorphism, or other crystallographic
phenomena. in addition, the nature of the substituents on the
ligand scaffold, is important as well, see below. In the case of
PySBn ligand, it forms five-membered metallacycles, and can
7
This article is protected by copyright. All rights reserved.

10.1002/ejic.202100355
European Journal of Inorganic Chemistry
FULL PAPER
Figure 10. Bar plots for the Fe-N-C(E) and N-C-E bond angle range (top) and for the (N)C—E and N—C(E) bond length range (bottom) for SCO-active and HS-
locked complexes of the type
[FeII(N2S2)(NCE)2].
collected at 1 and 10 K min-1 rates. Data were corrected for the
diamagnetism of the sample.
Experimental Section
General
X-ray crystallography
All reactions were carried out under argon atmosphere using standard
Schlenk techniques. Solvents were purified and dried by standard
procedures (tetrahydrofuran over Na/benzophenone, methanol and
ethanol over Mg/I2, acetone and acetonitrile over P2O5) and were distilled
under argon prior to use. All other chemicals were used as purchased
from Sigma-Aldrich, Merck, and J. T. Baker. Column chromatography
was carried out utilising Sigma-Aldrich Silica Gel (40-63 μm particle size).
TLC was performed with Macherey-Nagel Alugram®Xtra Sil G/UV254
plates employing UV light for visualization. All compressed gases were
obtained from Infra. Argon (> 99.9 %) and nitrogen (> 99.5 %) were used
as supplied without purification. FT-IR spectra were recorded utilising a
FT-IR Perkin Elmer Spectrum GX as KBr pellets (4000-400 cm-1) and
utilizing a Varian 640-IR Spectrometer with Attenuated Total Reflectance
(ATR) (4000-500 cm-1). The 1H (300, 400 and 500 MHz), 13C{1H} (75, 101
and 126 MHz) spectra were recorded in either a Bruker Advance DPX
300 MHz, Jeol Eclipse 400 MHz or Jeol Eclipse ECA 500 MHz at room
temperature, 1H and 13C{1H} spectra are referenced internally using the
residual proton and carbon solvent resonances relative to
tetramethylsilane. High-resolution mass spectra (HRMS) were obtained
by HPLC 1100 coupled/MSD-TOF with APCI as ionization source in an
Agilent Technologies HR-MSTOF G1969A. Mass spectra were obtained
on an Agilent HPLC-SQ-MS, HPLC 1200 series and MS-ESI-SQ model
6120. Elemental analyses were performed on a Thermo-Finningan Flash
1112 elemental analyser in the Chemistry Department at CINVESTAV.
X-ray diffraction measurements were collected on a Bruker APEX-II CCD
diffractometer, using graphite-monochromated Mo k radiation ( =
0.71073 Å). The structures were solved by direct methods, using
SHELXS-2014 included in WinGX v.2014.1 and refined by a full-matrix
least-squares method based on F2.[54] Absorption corrections were
performed by Multi-Scan. All non-hydrogen atoms were refined with
anisotropic thermal displacement coefficients unless specified otherwise.
Due to badly disordered solvent molecules in complexes 2b and 2c the
SQUEEZE[37] routine in PLATON was performed removing its
contribution to the overall intensity data. An improvement was observed
in the final refinement parameters. For more details of the structure
refinement of complexes 2b and 2c see supporting information. The
crystallographic information file (CIF) for compounds 1b and 2a-c
(CCDC: 2005424, 2005425, 2005427, 2005430, 2005431, 2005432 and
2005433) contains the supplementary crystallographic data for this paper.
Synthesis
2-Pycolythiol (B1). The synthesis of B1 was based
on the literature with slight modifications.[30] In a 100
mL round-bottom flask under argon atmosphere 2-
(chloromethyl)pyridine hydrochloride (3.000 g,
18.288 mmol, 1 eq.), thiourea (1.520 g, 20.117
Magnetic measurements
mmol, 1.1 eq.) and 50 mL of distilled and degassed water were added.
The resulting orange solution was heated to 80-85 °C for 1 h. Once the
amber solution reached room temperature it was cooled to 0 °C, followed
by addition of sodium hydroxide (2.340 g, 58.530 mmol, 3.2 eq.). The
resulting mixture was stirred for 30 minutes at 0°C followed by further
stirring at room temperature for 12 h, before the yellowish solution was
extracted with 200 mL of diethyl ether. The basic aqueous phase was
cooled to 0 °C and neutralised by careful dropwise addition of a
hydrochloric acid solution (37 %, w/w) until obtaining a pH value of 7.
The magnetic susceptibility measurements were carried out on
a
Quantum Design Physical Property Measurement System (PPMS)
Dynacool-9T equipped with a vibrating sample mount (VSM). The
measurements were performed in the temperature range of 300-5 K and
at 5 K min-1 rate with an applied field of 0.1 T (1000 Oe). Microcrystalline
or powder samples of complexes 1a-b and 2a-c were employed in the
measurements. In addition, the magnetic data for complex 1b was
8
This article is protected by copyright. All rights reserved.

10.1002/ejic.202100355
European Journal of Inorganic Chemistry
FULL PAPER
Then, the neutralised aqueous phase was extracted with 200 mL of
diethyl ether, and the combined organic extracts were dehydrated with
sodium sulphate, filtered and the solvent was removed in vacuo, to afford
B1 (1.442 g, 11.522 mmol, 63 %) as a yellow oil, which was used in next
synthetic step without further purification. The spectroscopic data were in
good agreement with the literature.[55] 1H-NMR (300 MHz, CDCl3 @7.26)
δ 8.54 (d, J = 4.3 Hz, 1H, Ha), 7.65 (td, J = 7.8, 1.7 Hz, 1H, Hc), 7.33 (d, J
= 7.8 Hz, 1H, Hd), 7.16 (t, J = 7.0 Hz, 1H, Hb), 3.85 (d, J = 7.9 Hz, 2H, He),
2.03 (t, J = 7.7 Hz, 1H, Hf). 13C{1H}-NMR (126 MHz, CDCl3 @ 77.16) δ
(ppm): δ 160.25 (5), 149.54 (1), 137.02 (3), 122.41 (2), 122.08 (4), 30.95
(6).
Complex synthesis
Synthetic methodology for trans-[FeII(Py)4(NCE)2] E = S, Se, BH3.
The synthesis of the complexes is based on reference.[32] A Schlenk
flask under argon atmosphere was charged with two equivalents of the
corresponding cyano-derivative salt (NH4NCS, KNCSe or NaNCBH3),
which was dissolved in 30 mL of distilled and degassed water, followed
by addition of six equivalents of pyridine. Under vigorous stirring, one
equivalent of iron(II) chloride tetrahydrate salt was added, observing the
instantaneous precipitation of yellow solids for NCS and NCBH3 or olive
green solid for NCSe derivatives. The corresponding resulting mixture
was stirred at room temperature for 1 h. The resulting precipitate solid
was filtered via cannula and washed with distilled and degassed water (3
x 50 mL) to remove pyridine excess. Finally, the solid was dried in vacuo
affording the corresponding complex in 85 % (NCS, yellow), 70 %,
(NCSe, olive green) and 95 % (NCBH3, pale-yellow) yield. NCS: IR-ATR
ν (cm-1): 2059 (s), 1597 (s), 1485 (m), 1440 (s), 1213 (w), 1147 (w), 1068
(m), 1038 (m), 1005 (m), 806 (w), 764 (m), 754 (m), 711 (s), 699 (s), 652
(w), 622 (m). NCSe IR-ATR ν (cm-1): 2060 (s), 1597 (s), 1484 (m), 1440
(s), 1356 (w), 1213 (m), 1147 (w), 1068 (m), 1038 (m), 1005 (m), 764 (m),
754 (m), 710 (s), 700 (s), 652 (w), 623 (m), 585 (m), 567 (m). NCBH3 IR-
ATR ν (cm-1): 2337 (m), 2179 (m), 1597 (m), 1486 (w), 1439 (m), 1217
(m), 1154 (w), 1120 (m), 1068 (w), 1039 (m), 1008 (m), 753 (m), 697 (s),
627 (m), 593 (w), 569 (s).
Benzylpicolylthioether (PySBn). The
synthesis of PySBn was based on the
literature with slight modifications.[29]
A
Schlenk flask under argon atmosphere
was charged with B1 (1.442 g, 11.252
mmol, 1 eq.), benzyl chloride (1.29 mL, 11.252 mmol, 1 eq.), sodium
hydroxide (450 mg, 11.252 mmol, 1 eq.) and 100 mL of anhydrous and
degassed ethanol. The resulting pale-yellow solution was heated to reflux
for 12 h. After that time, the resulting amber mixture was cooled to room
temperature, and filtered. The solution was then extracted with 200 mL of
dichloromethane, and the combined organic phases were dehydrated
with sodium sulphate, filtered, and the solvent was removed in vacuo to
afford PySBn as a reddish oil (2.205 g, 10.239 mmol, 91 %), which
crystallized at 0-5 °C as prismatic red crystals. 1H-NMR (500 MHz, CDCl3
@7.26) δ 8.55 (dd, J = 4.8, 0.8 Hz, 1H, Ha), 7.62 (td, J = 7.7, 1.7 Hz, 1H,
Hc), 7.39 – 7.28 (m, 3H, Hd, Hg y Hh), 7.26 – 7.21 (m, 1H, Hi), 7.15 (dd, J
= 7.4, 4.9 Hz, 1H, Hb), 3.75 (s, 2H, He), 3.69 (s, 2H, Hf). 13C{1H}-NMR
(126 MHz, CDCl3 @ 77.16) δ (ppm): 158.67 (5), 149.41 (1), 138.12 (8),
136.70 (3), 129.12 (9), 128.55 (10), 127.07 (11), 123.20 (4), 121.94 (2),
37.57 (6), 35.96 (7). IR-ATR ν (cm-1): 3055 (w), 2967 (w), 1588 (m), 1563
(m), 1493 (m), 1467 (m), 1452 (m), 1431 (s), 1306 (w), 1423 (w), 1216
(w), 1146 (w), 1088 (m) 1069 (m), 1043 (m), 1206 (m), 994 (m), 926 (w),
903 (w), 791 (m), 770 (m), 745 (s), 705 (s), 669 (m), 629 (w), 579 (m),
562 (m). SQ-ESI(+)MS (m/z) MeCN: Calculated for [C13H14NS]+ 216.1,
found 216.1.
Synthetic methodology for complexes [FeII(L)n(NCE)2]
All the metal complexes were synthesised following two possible
methodologies.
Method A: A Schlenk flask under argon atmosphere was charged with
iron(II) chloride tetrahydrate (1 eq.), A[NCE] salt (2 eq., A = NH4 for E =
S, A = K for E = Se and A = Na for E = BH3) and an adequate anhydrous
and degassed solvent. The resulting solution was stirred for 1 h at room
temperature and filtered via canula to remove inorganic salts. The filtrate
solution was added via cannula to a solution of the corresponding ligand
(1 or 2 eq.) in an adequate solvent. The mixture was stirred for 2 h at
room temperature observing the precipitation of a solid. Finally, the solid
was filtered via canula and dried under reduced pressure to afford the
corresponding product as powders.
2,3-bis(((2-picolyl)thio)methyl)
quinoxaline (QuinoxS). The synthesis of
QuinoxS was based on the literature with
slight modifications.[31] A Schlenk flask
under argon atmosphere was charged
Method B: A Schlenk flask under argon atmosphere was charged with
precursor complex [FeII(Py)4(NCE)2] (1 eq.) and 5 mL of anhydrous and
degassed solvent. The resulting suspension was layered with fresh and
degassed solvent. Finally, a solution of the corresponding ligand (1 or 2
eq.) in an adequate solvent was carefully deposited over the previous
layers. The three layers were mixed by liquid-liquid slow diffusion over a
few days, obtaining crystals.
with
2,3-bis(bromomethyl)quinoxaline
(1.487 g, 4.705 mmol, 1 eq.), B1 (1.178 g, 9.411 mmol, 2 eq.), sodium
hydroxide (376 mg, 9.411 mmol, 2 eq.) and 100 mL of anhydrous and
degassed ethanol. The resulting brown solution was heated to reflux for 3
h. After cooling to room temperature, the solution was filtered off and
extracted utilising 100 mL of distilled and degassed water and 200 mL of
degassed dichloromethane. The combined organic extracts were
dehydrated with sodium sulphate, and the solvent was removed in vacuo
to provide a brownish solid, which was further purified by column
chromatography on silica gel with dichloromethane/acetone gradient
(100/0 to 75/25) as the eluent. The solvent was removed under reduced
pressure to afford QuinoxS (1.638 g, 4.048 mmol, 86 %) as an off-white
solid. 1H-NMR (400 MHz, CDCl3 @7.26) 8.49 (ddd, J = 4.9, 1.8, 0.9 Hz,
2H, Ha), 8.00 (dd, J = 6.3, 3.5 Hz, 2H, Hg), 7.70 (dd, J = 6.4, 3.4 Hz, 2H,
Hh), 7.56 (td, J = 7.7, 1.8 Hz, 2H, Hc), 7.35 (d, J = 7.8 Hz, 2H, Hd), 7.08
(ddd, J = 7.5, 4.9, 1.1 Hz, 2H, Hb), 4.16 (s, 2H, Hf), 3.84 (s, 2H, He).
13C{1H}-NMR (101 MHz, CDCl3, @7.26) 158.00 (5), 152.47 (8), 149.72
(1), 140.86 (9), 136.62 (3), 129.68 (11), 128.89 (10), 123.37 (4), 122.05
(2), 37.98 (6), 35.45 (7). IR-ATR ν (cm-1): 3055 (w), 2971 (w), 2928 (w),
1584 (m), 1567 (m), 1479 (m), 1435 (m), 1415 (w), 1361 (w), 1325 (m),
1267 (w), 1180 (w), 1126 (w), 1051 (w), 995 (m), 937 (w), 898 (w), 830
(w), 805 (w), 788 (w), 732 (s), 746 (s), 709 (w), 683 (m), 614 (w), 581 (w),
557 (s). SQ-ESI(+)MS (m/z) MeCN: Calculated for [C22H21N4S2]+ = 405.1,
found 405.1.
Complex trans-[FeII(PySBn)2(NCS)2 (1a). Method A: iron(II) chloride
tetrahydrate (92 mg, 0.465 mmol, 1 eq.), ammonium isothiocyanate (106
mg, 1.395 mmol, 3 eq.) and 10 mL of anhydrous and degassed ethanol
were stirred at room temperature for 1 h then filtered to remove
ammonium chloride and ammonium isothiocyanate excess. The purple
solution was dried in vacuo to isolate bis-isothiocyanate iron(II) salt as a
yellow solid, which was dissolved in 10 mL of anhydrous and degassed
acetone, resulting in a purple solution. A pale-red solution of PySBn (200
mg, 0.930 mmol, 2 eq.) in 10 mL of anhydrous and degassed acetone
was added via cannula. The resulting amber solution was stirred for 3 h
at room temperature observing the precipitation of an off-white solid. The
solid was filtered off via canula to afford complex 3a (76 mg, 0.127 mmol,
27 %) as a white solid, which was dried under reduced pressure.
Modified method B: To
a suspension of precursor complex
[FeII(Py)4(NCS)2] (220 mg, 0.450 mmol, 1 eq.) in 10 mL of anhydrous and
degassed ethanol was added a solution of PySBn (194 mg, 0.900 mmol,
2 eq.) in 5 mL of anhydrous and degassed ethanol via cannula resulting
in a purple solution. After a few minutes, a white solid had precipitated,
which was filtered off and dried in vacuo resulting in a white powder
9
This article is protected by copyright. All rights reserved.

10.1002/ejic.202100355
European Journal of Inorganic Chemistry
FULL PAPER
(218.3 mg, 0.360 mmol, 80 %). IR-ATR ν (cm-1): 2059 (s), 1598 (m),
1570 (w), 1478 (m), 1438 (m), 1309 (w), 1150 (w), 1098 (w), 1056 (w),
1017 (w), 964 (w), 910 (w), 841 (w), 774 (m), 752 (m), 711 (s), 699 (s),
674 (m), 641 (m), 624 (m), 593 (m), 567 (m). SQ-ESI(+)MS (m/z) DMSO:
Calculated for [FeII(C13H13NS)2(Cl)]+ 521.1, found 521.1. Elemental
analysis (%): Calculated for [FeII(C13H13NS)2(NCS)2]·H2O C: 54.19, H:
4.55, N: 9.03; found: C: 54.30, H: 4.26, N: 10.51.
Acknowledgements
This work was financially supported by CINVESTAV and SEP-
CONACyT (CB 2016/286346). Diego Plaza-Lozano thanks
CONACyT for a PhD scholarship (Beca 591879).
Keywords: Spin crossover • Magnetic properties • iron(II)
compounds • S-based ligands • Hysteresis
Complex trans-[FeII(PySBn)2(NCSe)2] (1b). Method A: similar to
complex 3a, PySBn (200 mg, 0.930 mmol, 2 eq.), iron(II) chloride
tetrahydrate (92 mg, 0.465 mmol, 1 eq.) and potassium isoselenocyanate
(202 mg, 1.395 mmol, 3 eq.). Complex 3b was isolated as a white solid
(66 mg, 0.094 mmol, 20 %). Modified method B: similar to complex 3a.
Complex [FeII(Py)4(NCSe)2] (125 mg, 0.216 mmol, 1 eq.) and PySBn (93
mg, 0.432 mmol, 2 eq.). Complex 3b was isolated as pale grey solid (22
mg, 0.032 mmol, 15 %). Single crystals suitable for X-ray crystallography
were obtained by diethyl ether vapor diffusion into the filtrate solution. IR-
ATR ν (cm-1): 3028 (w), 2059 (s), 1598 (m), 1568 (w), 1494 (m), 1476 (m),
1437 (m), 1416 (w), 1399 (w), 1308 (m), 1265 (w), 1246 (w), 1025 (w),
1149 (m), 1098 (m), 1055 (m), 1018 (m), 965 (w), 910 (m), 841 (m), 773
(s), 751 (s), 711 (s), 699 (s), 671(m), 640 (m), 595 (m), 576 (m).
[1]
P. Gütlich, H. A. Goodwin, in Spin Crossover Transit. Met. Compd. I
(Eds.: P. Gütlich, H.A. Goodwin), Springer Berlin Heidelberg, Berlin,
Heidelberg, 2004, pp. 1–47.
[2]
[3]
J. A. Real, A. B. Gaspar, M. C. Muñoz, Dalton Trans. 2005, 2062–2079.
M. A. Halcrow, Ed., Spin‐Crossover Materials: Properties and
Applications, John Wiley & Sons, Ltd, 2013.
[4]
[5]
O. S. Zhao-Yang Li, Z.-S. Yao, S. Kang, S. Kanegawa, in
Spin‐Crossover Mater. (Ed.: M.A. Halcrow), John Wiley & Sons, Ltd,
2013.
J.-F. Létard, P. Guionneau, L. Goux-Capes, in Spin Crossover Transit.
Met. Compd. III, Springer Berlin Heidelberg, Berlin, Heidelberg, 2004,
pp. 221–249.
Complex cis-[FeII(QuinoxS)(NCS)2] (2a). Method B: [FeII(Py)4(NCS)2]
(110 mg, 0.225 mmol, 1 eq.) in 10 mL of anhydrous and degassed
acetonitrile. Over this solution 10 mL of anhydrous and degassed
acetonitrile were carefully added, forming two liquid layers. A third liquid
layer of a yellow solution of QuinoxS (100 mg, 0.225 mmol, 1 eq.) in 10
mL of anhydrous and degassed chloroform was carefully deposited over
the previous layers. The three layers were mixed by liquid-liquid slow
diffusion over 4 days, obtaining a pale red solution, and amber crystals
with prismatic morphology of complex 4a which were suitable for single
crystal X-ray diffraction (53 mg, 0.092 mmol, 41 %). IR-ATR ν (cm-1):
2070 (m), 2059 (m), 1600 (m), 1482 (m), 1439 (m), 1390 (d), 1359 (d),
1315 (d), 1265 (d), 1242 (d), 1201 (d), 1162 (d), 1125 (w), 1103 (w),
1052 (w), 1017 (m), 879 (w), 851 (w), 822 (w), 767 (s), 696 (w), 640 (m),
611 (m), 563 (s). SQ-ESI(+)MS (m/z) DMF: Calculated for
[FeII(C22H20N4S2)(NCS)]+ 518.0, found 518.1. Elemental analysis (%):
Calculated for [FeII(C22H20N4S2)(NCS)2] C: 50.00, H: 3.50, N: 14.58;
found: C: 50.21, H: 3.30, N: 14.74.
[6]
[7]
S. Brooker, J. A. Kitchen, Dalton Trans. 2009, 7331–7340.
A. Bousseksou, G. Molnár, L. Salmon, W. Nicolazzi, Chem. Soc. Rev.
2011, 40, 3313–3335.
[8]
[9]
H. Li, H. Peng, Curr. Opin. Colloid Interface Sci. 2018, 35, 9–16.
G. Molnár, S. Rat, L. Salmon, W. Nicolazzi, A. Bousseksou, Adv. Mater.
2018, 30, 1703862.
[10] G. Molnár, L. Salmon, W. Nicolazzi, F. Terki, A. Bousseksou, J. Mater.
Chem. C 2014, 2, 1360–1366.
[11] R. Pei, E. Matamoros, M. Liu, D. Stefanovic, M. N. Stojanovic, Nat.
Nanotechnol. 2010, 5, 773.
[12] R. N. Muller, L. Vander Elst, S. Laurent, J. Am. Chem. Soc. 2003, 125,
8405–8407.
[13] J. Olguín, Coord. Chem. Rev. 2020, 407, 213148.
[14] D. J. Harding, P. Harding, W. Phonsri, Coord. Chem. Rev. 2016, 313,
38–61.
[15] H. A. Goodwin, in Spin Crossover Transit. Met. Compd. II (Eds.: P.
Gütlich, H.A. Goodwin), Springer Berlin Heidelberg, Berlin, Heidelberg,
2004, pp. 23–47.
Complex cis-[FeII(QuinoxS)(NCSe)2] (2b). Method B: Similar to
complex 4a. [FeII(Py)4(NCSe)2] (131 mg, 0.225 mmol, 1 eq.) in 10 mL of
acetonitrile. Layered with a solution of QuinoxS (100 mg, 0.225 mmol, 1
eq.) in 10 mL of CHCl3. Complex 4b was obtained as prismatic yellow
crystals (59 mg, 0.101 mmol, 57 %). IR-ATR ν (cm-1): 2061 (w), 1650 (w),
1595 (w), 1476 (w), 1441 (w), 1342 (w), 855 (w), 758 (m), 700 (w), 646
(w), 609 (m), 584 (m), 550 (s). SQ-ESI(+)MS (m/z) DMSO: Calculated for
[FeII(C22H20N4S2)(Cl)]+ 495.0, found 495.0.
[16] S. Hayami, Y. Komatsu, T. Shimizu, H. Kamihata, Y. H. Lee, Coord.
Chem. Rev. 2011, 255, 1981–1990.
[17] V. A. Grillo, L. R. Gahan, G. R. Hanson, R. Stranger, T. W. Hambley, K.
S. Murray, B. Moubaraki, J. D. Cashion, J. Chem. Soc. Dalton Trans.
1998, 2341–2348.
[18] C. Reus, K. Ruth, S. Tüllmann, M. Bolte, H.-W. Lerner, B. Weber, M. C.
Holthausen, M. Wagner, Eur. J. Inorg. Chem. 2011, 1709–1718.
[19] R. W. Hogue, H. L. C. Feltham, R. G. Miller, S. Brooker, Inorg. Chem.
2016, 55, 4152–4165.
Complex cis-[FeII(QuinoxS)(NCBH3)2] (2). Method B: Similar to
complex 4a. [FeII(Py)4(NCBH3)2] (102 mg, 0.225 mmol, 1 eq.) in 10 mL of
acetonitrile. Layered with a solution of QuinoxS (59 mg, 0.101 mmol, 2
eq.) in 10 mL of CHCl3. Complex 4c was obtained as prismatic yellow
crystals (59 mg, 0.101 mmol, 45 %). IR-ATR ν (cm-1): 2349 (m), 2179 (m),
1601 (m), 1587 (w), 1483 (m), 1438 (m), 1392 (w), 1360 (w), 1268 (w),
1114 (m), 1057 (w), 1019 (w), 851 (w), 759 (s), 694 (w), 642 (w), 592 (s),
574 (s), 564 (s). SQ-ESI(+)MS (m/z) DMF: Calculated for
[FeII(C22H20N4S2)(NCBH3]+ 500.1, found 500.1. Elemental analysis (%):
Calculated for [FeII(C22H20N4S2)2(NCBH3)2]·(CH3CN) C: 53.74, H: 5.03,
N: 16.87; found: C: 53.22, H: 5.14, N: 17.41.
[20] A. Arroyave, A. Lennartson, A. Dragulescu-Andrasi, K. S. Pedersen, S.
Piligkos, S. A. Stoian, S. M. Greer, C. Pak, O. Hietsoi, H. Phan, S. Hill,
C. J. McKenzie, M. Shatruk, Inorg. Chem. 2016, 55, 5904–5913.
[21] S. Yergeshbayeva, J. J. Hrudka, J. Lengyel, R. Erkasov, S. A. Stoian, A.
Dragulescu-Andrasi, M. Shatruk, Inorg. Chem. 2017, 56, 11096–11103.
[22] A. Lennartson, A. D. Bond, S. Piligkos, C. J. McKenzie, Angew. Chem.
Int. Ed. 2012, 51, 11049–11052.
[23] J. England, R. Gondhia, L. Bigorra-Lopez, A. R. Petersen, A. J. P.
White, G. J. P. Britovsek, Dalton Trans. 2009, 5319–5334.
[24] C. G. Young, J. Inorg. Biochem. 2007, 101, 1562–1585.
[25] B. S. Stadelman, M. M. Kimani, C. A. Bayse, C. D. McMillen, J. L.
Brumaghim, Dalton Trans. 2016, 45, 4697–4711.
[26] M. M. Kimani, H. C. Wang, J. L. Brumaghim, Dalton Trans. 2012, 41,
5248–5259.
[27] D. Plaza-Lozano, D. Morales-Martínez, F. J. González, J. Olguín, Eur. J.
Inorg. Chem. 2020, 1562–1573.
[28] H. Toftlund, J. J. McGarvey, in Spin Crossover Transit. Met. Compd. I
(Eds.: P. Gütlich, H.A. Goodwin), Springer Berlin Heidelberg, Berlin,
Heidelberg, 2004, pp. 151–166.
10
This article is protected by copyright. All rights reserved.

10.1002/ejic.202100355
European Journal of Inorganic Chemistry
FULL PAPER
[29] O. Prakash, P. Singh, G. Mukherjee, A. K. Singh, Organometallics 2012,
31, 3379–3388.
[30] Y.-J. Kim, J. Han, Bull. Korean Chem. Soc. 2012, 33, 48–54.
[31] F. Saleem, G. K. Rao, P. Singh, A. K. Singh, Organometallics 2013, 32,
387–395.
[32] R. A. A. George B. Kauffman Fred L. Harlan, R. Scott Stephens,
Ronald O. Ragsdale, in Inorg. Synth. (Ed.: R.W. Parry), 1970, pp. 251–
256.
[33] J. L. Burmeister, L. E. Williams, Inorg. Chem. 1966, 5, 1113–1117.
[34] K. Nakamoto, in Infrared Raman Spectra Inorg. Coord. Compd. Part B
(Ed.: Wiley-Interscience), 2008, pp. 1–273.
[35] A. H. Norbury, in Adv. Inorg. Chem. Radiochem. (Eds.: H.J. Emeléus,
A.G. Sharpe), Academic Press, 1975, pp. 231–386.
[36] S. Rodríguez-Jiménez, S. Brooker, Inorg. Chem. 2017, 56, 13697–
13708.
[37] A. Spek, Acta Crystallogr. Sect. C 2015, 71, 9–18.
[38] F. Anthony Deeney, C. J. Harding, G. G. Morgan, V. McKee, J. Nelson,
S. J. Teat, W. Clegg, J. Chem. Soc. Dalton Trans. 1998, 1837–1844.
[39] M. G. B. Drew, C. J. Harding, V. McKee, G. G. Morgan, J. Nelson, J.
Chem. Soc. Chem. Commun. 1995, 1035–1038.
[40] M. A. Halcrow, in Spin‐Crossover Mater. (Ed.: M.A. Halcrow), 2013.
[41] G. Novitchi, S. Jiang, S. Shova, F. Rida, I. Hlavička, M. Orlita, W.
Wernsdorfer, R. Hamze, C. Martins, N. Suaud, N. Guihéry, A.-L. Barra,
C. Train, Inorg. Chem. 2017, 56, 14809–14822.
[42] G. Carver, P. L. W. Tregenna-Piggott, A.-L. Barra, A. Neels, J. A. Stride,
Inorg. Chem. 2003, 42, 5771–5777.
[43] A. Ozarowski, S. A. Zvyagin, W. M. Reiff, J. Telser, L.-C. Brunel, J.
Krzystek, J. Am. Chem. Soc. 2004, 126, 6574–6575.
[44] T. Sugaya, T. Fujihara, T. Naka, T. Furubayashi, A. Matsushita, H.
Isago, A. Nagasawa, Chem. – Eur. J. 2018, 24, 17955–17963.
[45] The calculation of the remaining HS-state percentage at low
temperature for complex 1b was performed using the MT parameter,
instead of the eff value, due to the linear relationship of MT of the spin
state mixture: MT(mixture)= HS(MT)HS+ LS(MT)LS,  = mole fraction.
[46] R. Kulmaczewski, J. Olguín, J. A. Kitchen, H. L. C. Feltham, G. N. L.
Jameson, J. L. Tallon, S. Brooker, J. Am. Chem. Soc. 2014, 136, 878–
881.
[47] S. Brooker, Chem. Soc. Rev. 2015, 44, 2880–2892.
[48] A. J. Fitzpatrick, E. Trzop, H. Muller-Bunz, M. M. Dirtu, Y. Garcia, E.
Collet, G. G. Morgan, Chem Commun 2015, 51, 17540–17543.
[49] H. Beraldo, L. Tosi, Inorganica Chim. Acta 1983, 75, 249–257.
[50] D. Sellmann, J. Utz, N. Blum, F. W. Heinemann, Coord. Chem. Rev.
1999, 190–192, 607–627.
[51] M. A. Halcrow, Chem. Lett. 2014, 43, 1178–1188.
[52] S. Vela, H. Paulsen, Inorg. Chem. 2018, 57, 9478–9488.
[53] P. Gütlich, Y. Garcia, H. Spiering, in Magn. Mol. Mater. IV (Eds.: J.S.
Miller, M. Drillon), John Wiley & Sons, Ltd, 2003, pp. 271–344.
[54] G. Sheldrick, Acta Crystallogr. Sect. A 2008, A64, 112–122.
[55] P. Remuzon, D. Bouzard, P. DiCesare, M. Essiz, J.-P. Jacquet, A.
Nicolau, A. Martel, M. Menard, C. Bachand, Tetrahedron 1995, 51,
9657–9670.
11
This article is protected by copyright. All rights reserved.

10.1002/ejic.202100355
European Journal of Inorganic Chemistry
FULL PAPER
Entry for the Table of Contents
The effect of the chelate ring and NCE co-ligand on the magnetic behaviour in complexes of the type [FeII(N2S2)(NCE)2] based on
picolylthioether ligands is shown. A structural analysis has been performed, confirming that the Fe-N-C(E) angle is a good marker
that could permit predicting SCO-behaviour, and in addition, we have shown that the N-C(E) bond length is useful too as a predicting
marker.
Institute and/or researcher Twitter usernames: @Leolguin (Dr. J. Olguín)
12
This article is protected by copyright. All rights reserved.