Amphiphilic iron(ii) spin crossover coordination polymers: Crystal structures and phase transition properties

Article abstract of DOI:10.1039/c8tc05580g

Iron(ii) coordination polymers with an N₂O₂ coordinating Schiff base-like equatorial ligand bearing different alkyl chain lengths (C16, C18, C20, and C22) and 1,2-bis(4-pyridyl)ethyne, 1,2-bis(4-pyridyl)ethene or 1,2-bis(4-pyridyl)et

Full text of DOI:10.1039/c8tc05580g

Journal of
Materials Chemistry C
View Article Online
View Journal
PAPER
Amphiphilic iron(II) spin crossover coordination
polymers: crystal structures and phase transition
properties†
Cite this:DOI: 10.1039/c8tc05580g
Johannes Weihermuller,a Stephan Schlamp,a Wolfgang Milius,b Florian Puchtler,b
¨
b
d
Josef Breu,
Philipp Ramming,c Sven Huttner,c Seema Agarwal,
¨
Christoph Gobel,a Markus Hund,e Georg Papastavrou
and Birgit Weber
*
e
a
¨
Iron(II) coordination polymers with an N2O2 coordinating Schiff base-like equatorial ligand bearing different
alkyl chain lengths (C16, C18, C20, and C22) and 1,2-bis(4-pyridyl)ethyne, 1,2-bis(4-pyridyl)ethene or
1,2-bis(4-pyridyl)ethane as bridging ligand are synthesized. All complexes display a rather similar abrupt
spin transition above room temperature, which is investigated using magnetic measurements and
Mossbauer spectroscopy. Variation of the bridging ligand and the alkyl chain lengths allows fine tuning of
¨
the transition temperature in the range between 338 K and 357 K. Single crystal X-ray structure analysis of
two coordination polymers and one of the starting complexes reveals the formation of a lipid layer-like
arrangement of the amphiphilic complexes in all cases. Further characterization by thermal gravimetric
analysis, differential scanning calorimetry, X-ray powder diffraction, and polarized optical microscopy show
in all cases solid–solid phase transitions. Those transitions determine the spin crossover behavior and
depend on the crystal packing that is controlled by the alkyl chains in the outer periphery of the ligand.
Thus, with the presented system the spin crossover properties are controlled by small alterations of the
ligand structures. With respect to technological applications, spin coating is shown to be suitable for the
processing of the complexes as thin films and furthermore thin platelets of the complexes can be generated
by delamination techniques.
Received 6th November 2018,
Accepted 9th December 2018
DOI: 10.1039/c8tc05580g
rsc.li/materials-c
through chemical stimuli, the spin state of the metal center
can be switched between a high spin (HS) and a low spin (LS)
1. Introduction
Bistability is a property frequently observed for hexa-coordinated state, a phenomenon known as spin crossover (SCO). Due to the
complexes of 3d transition metals with d4–d7 electron pronounced property changes upon SCO, these switchable
configuration.1–3 Through external perturbations like changes molecular materials have a high potential for a variety of
in temperature, pressure, by electromagnetic irradiation, or different applications.4–8 One example would be bio-sensors
for nano-thermometry9–11 or the detection of biologically relevant
parameters such as pH. In order to achieve such applications, the
a
¨
Department of Chemistry, Inorganic Chemistry VI, Universitat Bayreuth,
synthesis of nanostructured and/or composite materials is investi-
gated very actively.12–17 Alternatively, attempts are made to combine
the SCO with additional properties like softness (metallomesogens,
amphiphilic molecules)18–21 leading to multifunctionality, new
patterning possibilities, and by this enlarging the range of potential
SCO applications. The structural changes upon spin state change
could trigger a LC phase transition or, alternatively, the phase
transition could trigger the spin transition.18,22,23 With regard to
this, Seredyuk et al. demonstrated that the SCO can be influenced
by crystal–liquid crystal phase transitions (PTs) of metal complexes
functionalized with long alkyl chain substituents18,24,25 Hayami and
co-workers observed interesting phenomena like a reverse ST due
to PTs for amphiphilic cobalt complexes.22,26 Further approaches
by the group of Real yielded scan rate dependent cooperative spin
¨
Universitatsstrasse 30, NW I, 95440 Bayreuth, Germany.
E-mail: weber@uni-bayreuth.de
b
c
¨
Department of Chemistry, Inorganic Chemistry I, Universitat Bayreuth,
¨
Universitatsstrasse 30, NW I, 95440 Bayreuth, Germany
¨
Department of Chemistry, Macromolecular Chemistry I, Universitat Bayreuth,
¨
Universitatsstrasse 30, NW I, 95440 Bayreuth, Germany
d
e
¨
Department of Chemistry, Macromolecular Chemistry II, Universitat Bayreuth,
¨
Universitatsstrasse 30, NW I, 95440 Bayreuth, Germany
¨
Department of Chemistry, Physical Chemistry II, Universitat Bayreuth,
¨
Universitatsstrasse 30, NW I, 95440 Bayreuth, Germany
† Electronic supplementary information (ESI) available: Experimental section,
¨
powder X-ray diffraction data and single crystal X-ray diffraction data, Mossbauer
spectra of 7–10 and magnetic measurements of 2, TGA and DSC measurements,
POM micrographs of all complexes, AFM images of thin films of 7. CCDC
1835195, 1835196 and 1044903. For ESI and crystallographic data in CIF or other
electronic format see DOI: 10.1039/c8tc05580g
This journal is ©The Royal Society of Chemistry 2018
J. Mater. Chem. C

View Article Online
Paper
Journal of Materials Chemistry C
transitions for complexes with short alkyl chains27 and a more able to fine-tune the SCO parameters through control of the
gradual SCO and crystal–liquid crystal phase transition for crystal packing – an essential point for future applications.
complexes with longer alkyl chains.28 Further optimization Additionally, the investigation of amphiphilic SCO systems is one
of the system showed that a phase transition in meltable step further towards synthesis of self-assembled (mono-)layers
complexes can be used to control the spin transition.29 Inves- of SCO molecules. In contrast to the coordination polymers
tigations of Albrecht et al. on amphiphilic iron(III) complexes with amphiphilic ligands reported so far, the system presented
in solution30 and in the solid31 revealed that an increase of the here compromised neutral polymer chains. The influence of
length of the alkyl chains did lead to higher cooperativity of this differences on the SCO properties and film formation will
the spin crossover due to an improved self-assembly. Further- be discussed.
more, the combination of LC and SCO properties offers the
opportunity of an orientation of the complexes via the electrical
field. In the case of amphiphilic systems new strategies to
process bulk SCO compounds into thin films also motivated
2. Results and discussion
2.1 Synthesis
such studies. Those could be obtained by techniques such as
the Langmuir–Blodgett, spin coating, or drop casting.21,30,32–34
In Scheme 1, the general structure of the ligands and complexes
discussed in this work and the used abbreviations are given.
Such self-assembled monolayers of SCO molecules as thin films
can be studied by scanning tunneling microscopy (STM).35 Self-
The ligands H2L(y+1) were synthesized following procedures
described in literature for similar systems.44 The reaction with
assembling spin crossover complexes based on alkylated ligands
iron(II) acetate48 in methanol yielded the corresponding parent
reported so far are either based on mononuclear complexes
complexes [FeL(y+1)(MeOH)2] with two methanol molecules as
(e.g. Langmuir–Blodgett film formation of the [Fe(L2)(NCS)2] or
[Fe(L)3]2+ 2NCSꢀ system with L = 2,20-bipyridine substituted in
axial ligand. For [FeL(20)(MeOH)2], single crystals of high
position 4 and 40 with long alkyl chains),36,37 triazole-based coordi-
enough quality were obtained to determine the crystal structure
that is discussed in the following. In the next step the axial
nation polymers with the triazole carrying alkyl trails (Langmuir–
ligands (MeOH) were substituted by 1,2-bis(4-pyridyl)ethyne49,50
Blodgett film formation, solid state properties),38–40 or complexes
with amphiphilic counter ions (thin film formation).41
(bpey), 1,2-bis(4-pyridyl)ethene (bpee), or 1,2-bis(4-pyridyl)ethane
(bpea), respectively to yield complexes 1–6 (ESI,† Scheme S1).
Single crystals of 4ꢁtol and 6ꢁtol were obtained by slow diffusion
Please note that the SCO may be modified by the functiona-
lization of ligands with long alkyl chains and by the processing,
thus a prediction of the results so far is highly difficult.37
setups whose structures are also discussed in the following. All
iron(II) complexes are very air sensitive in solution and in part
However, for a purposeful synthesis of SCO-based materials, it
also in the solid state. For comparison purpose, the corres-
is indispensable to be able to predict the impact of changes in
ponding oxidized m-O-complexes 7–10 were synthesized as well.
the ligand structure on the SCO parameters.
For iron(II) complexes with amphiphilic Schiff base-like
ligands used in our group, self-assembly behavior was observed
2.2. X-ray structure analysis
but no LC properties.42,43 The complexes crystallize in lipid Platelet-like crystals of [FeL(20)(MeOH)2] and 4ꢁtol and spicular
layer like structures. In agreement with the results from crystals of 6ꢁtol suitable for X-ray structure analysis were
Albrecht et al. longer alkyl chains support the formation of obtained either directly from the synthesis or by slow diffusion
lipid layers and by this lead to improved spin crossover setups. The crystal data were collected at 133 K ([FeL(20)(MeOH)2]
properties.44 In one case an over 20 K wide thermal hysteresis and 4ꢁtol) and 200 K (6ꢁtol) and are summarized in the ESI,†
loop is observed.45 So far we were not able to investigate the PT Table S1. For 4ꢁtol a non-mathematical twin was obtained.
properties of those complexes as they decompose at higher Therefore, the refinement of the crystal structure is incomplete
temperatures (above 350 K), a typical behavior for mono- and and it will be discussed as a structural motif only. Thus, only the
dinuclear complexes of this type. For the synthesis of more general packing of the molecules is considered but no exact
stable coordination polymers, relatively large bridging ligands bond lengths or angles are given. Please note that it is very
as 1,2-bis(4-pyridyl)ethane (bpea), 1,2-bis(4-pyridyl)ethene difficult to obtain large enough single crystals of such amphi-
(bpee), or 1,2-bis(4-pyridyl)ethyne (bpey) are necessary.46,47 For philic complexes and that the molecular weight per iron center is
the realization of those systems the self-assembly parameter with more than 1000 g molꢀ1 very high. Thus, the R values are in
(sap) needs to be considered, where interplay of the coordination all cases larger than the ones usually obtained for smaller
number, the size of the axial ligands attached at the iron(II) molecules. All complexes crystallized in the triclinic space group
center, and the alkyl chain length is summarized.42 Through P1. Selected bond lengths and angles within the first coordina-
%
application of this parameter, we are now able to predict the tion sphere of the iron center are listed in Table 1. An ORTEP
successful synthesis of the coordination polymers with Schiff drawing of the asymmetric unit is given in Fig. 1.
base-like ligand with 16–22 carbon atoms in the alkyl chain that
The iron center of [FeL(20)(MeOH)2] has an N2O4 coordina-
crystallize in a lipid-layer like arrangement. Please note that tion sphere build by the equatorial N2O2-coordinating Schiff
despite of the number of examples it is still difficult for a given base-like ligand and two axially coordinating methanol, as
system to predict the packing of the molecules in the crystal and shown at the top of Fig. 1. The average bond lengths are
by this the spin crossover properties. Here we show that we are 2.10 Å (Fe–Neq), 2.00 Å (Fe–Oeq), and 2.19 Å (Fe–Oax) and the
J. Mater. Chem. C
This journal is ©The Royal Society of Chemistry 2018

View Article Online
Journal of Materials Chemistry C
Paper
Scheme 1 General structure of the coordination polymers and the m-O-complexes discussed in this work and their abbreviations. (y + 1) denotes to the
number of carbon atoms of the alkyl chain, bpea, bpee, and bpey denote to the bridging ligands 1,2-bis(4-pyridyl)ethane, 1,2-bis(4-pyridyl)ethene and
1,2-bis(4-pyridyl)ethyne.
Table 1 Selected bond lengths [Å] and angles [1] of [FeL(20)(MeOH)2]*, 4ꢁtol, and 6ꢁtol within the first coordination sphere and their dimensions [Å] and sap42
a
Compound
Fe–Neq
Fe–Oeq
Fe–Oax/Nax
Oeq–Fe–Oeq
109.38(14)
Lax–Fe–Lax
166.88(11)
-Lax
a
b
H
B
L
sap
0.7
[FeL(20)(MeOH)2]
2.097(4)
2.100(4)
1.9
1.9
1.885(4)
1.908(3)
1.990(4)
2.019(3)
2.0
1.9
1.917(3)
1.953(3)
2.199(4)
2.184(3)
2.0
2.0
2.012(3)
2.003(4)
—
162
176
8
15
33
4ꢁtol
6ꢁtol
86
176
5
112
140
131
158
14
13
15
14
29
29
1.0
0.9
90.05(10)
175.91(16)
88
a
Torsion angle between the axial pyridine rings.
O–Fe–O angle is 1091. Those values are in the region typical for The details (distances and angles) of the hydrogen bonds are
octahedral HS iron(II) complexes of this ligand type.46,51–53 The given in Table 2.
Lax–Fe–Lax angle of 1661 deviates slightly of the expected 1801
ORTEP drawings of the asymmetric unit of 4ꢁtol and 6ꢁtol are
for a perfect octahedral coordination sphere. An analysis of the given in the center and at the bottom of Fig. 1. In both cases the
packing of the molecules in the crystal, given at the top of iron center has an N4O2 coordination sphere build of the
Fig. 2, reveals that they are ordered in a lipid layer-like equatorial N2O2 coordinating ligand and the bridging N coor-
arrangement. The alkyl chains build parallel layers with an dinating ligand. The average bond lengths within the first
approximate layer to layer distance of about 4.2 Å. This distance coordination sphere are 1.89 Å/1.90 Å (Fe–Neq), 1.96 Å/1.94 Å
is typical for stabilizing van der Waals interactions (London (Fe–Oeq), and 1.99 Å/2.01 Å (Fe–Nax), respectively. The O–Fe–O
Dispersion forces) between the alkyl chains and a similar angles are 881 in average. Those values are in the region typical
behavior is observed for other amphiphilic complexes, not only for LS iron(II) complexes of this ligand type.50,54 The Lax–Fe–Lax
for this general ligand type, but also for others that are less angle of 1761 (4ꢁtol and 6ꢁtol) deviates only slightly of the
related.42 The iron containing head groups are oriented to each expected 1801 for a perfect octahedral coordination sphere.
other between the layers of the alkyl chains. Two intermolecular The torsion angle between the axial pyridine rings is 51 for 4ꢁtol
hydrogen bonds are observed within the layer of the head and 881 for 6ꢁtol. The remaining electron density of 6ꢁtol
groups. The hydrogen bond O(9)–H(9)ꢁ ꢁ ꢁO(3) connects the indicates included solvent molecules. However, due to a strong
complex molecule in the same lipid-like layer to form infinite disorder, they could not be further refined and therefore
chains along [1 0 0] and O(10)–H(10A)ꢁ ꢁ ꢁO(2) connects the head SQUEEZE from PLATON55 was used. A total number of 97
groups of two opposite layers through the formation of dimers. electrons were removed from the refinement with a void
This journal is ©The Royal Society of Chemistry 2018
J. Mater. Chem. C

View Article Online
Paper
Journal of Materials Chemistry C
Fig. 1 ORTEP drawing of the asymmetric unit of the crystal structure of
[FeL(20)(MeOH)2] (top), 4ꢁtol (center) and 6ꢁtol (bottom). Hydrogen atoms
are omitted for clarity. Thermal ellipsoids shown at the 50% probability
level.
Fig. 2 Molecular packing of [FeL(20)(MeOH)2] along [0.3 2.3 0] (top),
4ꢁtol along [1 0 0] (center) and 6ꢁtol along [1 0 0] (bottom) illustrating
selected intermolecular distances discussed in the manuscript. Hydrogen
atoms are omitted for clarity.
volume of 946 Å3. Consequently, intermolecular contacts
between the polar head groups cannot be discussed for both
complexes.
The packing of the molecules in the crystal is for both
complexes very similar to that of [FeL(20)(MeOH)2]. The dis-
tances between the alkyl chains correspond to a maximum of
stabilizing van der Waals interactions (approx. 4.2 Å), in good
agreement with the particular high ordering of the alkyl chains
without major bending. As shown in Fig. 2 (center and bottom),
the molecules are ordered in the crystal in a lipid-layer like
arrangement. One significant difference between the three
structures are the angles between the plain of the chelate cycle
and the alkyl chains. Here we can define two angles: the
bending of the alkyl chains relative to the plane of the equatorial
ligand (angle a) and the shift sideward in the plane of the chelate
cycle (angle b). A schematic presentation of the angles is shown
in Fig. 3. For [FeL(20)(MeOH)2] the bending a is 1621 and the
shifting b is 1761. This bending and shifting is more pronounced
for the two complexes 4ꢁtol and 6ꢁtol with the longer chains. The
values for 4ꢁtol are a = 1121 and b = 1311 and for 6ꢁtol a = 1401
and b = 1581, respectively. This deviation of an ideal linear
Table
[FeL(20)(MeOH)2]*
2
Distances [Å] and angles [1] of the hydrogen bonds of
Bond
D–H
Hꢁ ꢁ ꢁA
Aꢁ ꢁ ꢁD
D–Hꢁ ꢁ ꢁA
O(9)–H(9)ꢁ ꢁ ꢁO(3)a
0.84
0.84
1.87
1.95
2.672(7)
2.767(4)
161
165
O(10)–H(10A)ꢁ ꢁ ꢁO(2)b
a
b
1 + x, y, z. 1 ꢀ x, 2 ꢀ y, 2 ꢀ z.
hexa-coordinated fashion. The orientation compensates the
sterical demand of the axial ligand and, therefore, provides a
lipid layer-like ordering.
As discussed in a previous work,42 there is a relation between the
size of the head group (height H plus broadness B) and the length L
of the molecule which is called sap (self-assembly parameter):
ðH þ BÞ
sap ¼
L
arrangement along the equatorial ligand (both angles 1801) is A lipid layer-like arrangement can be expected for a sap E 1.
responsible for the possibility of the complexes to crystallize in a The calculated values for [FeL(20)(MeOH)2], 4ꢁtol and 6ꢁtol are
J. Mater. Chem. C
This journal is ©The Royal Society of Chemistry 2018

View Article Online
Journal of Materials Chemistry C
Paper
between the layers of the alkyl chain in the region of 4.3 Å.42,56
Especially for the coordination polymers 1–4 with bpey as
bridging ligand strong similarities are observed thus a very
similar packing of the molecules in the crystal is likely. For the
more flexible ligands bpee and bpea, the obtained samples are
less crystalline with slightly broader reflexes. However, the
overall pattern is still very similar to that observed for 1–4
and to that of the calculated patterns from the single crystals.
For comparison purpose, the XRPD spectra of the corresponding
the m-O-complexes 7–10 were recorded as well and are given in
the ESI,† Fig. S1.
Fig. 3 Schematic representation of the angles for the bending a and
shifting b between the plane of the chelate cycle and the alkyl chains.
In contrast to the XRPD spectra of 1–6, for the m-O-complexes
7–10 a set of reflexes is observed in the region of 3.51–6.51 2y,
whereas in the region 6.51–81 2y no reflexes are observed. In the
201–251 2y, on the other side, some reflexes are observed as in
the case of the coordination polymers. For all samples strong
reflexes are observed below 3.51 2y. In the case of the coordina-
tion polymers, the signal appears around 2.51 2y. By using
Bragg’s law (nl = 2d sin y; n = 1, l = 1.54184 Å, d = interplanar
distance, y = scattering angle) the related distances can be
calculated and correlated with distances observed in the crystal
packing of 4ꢁtol and 6ꢁtol (Table 3 and Fig. 2). They fit very well to
the Fe–Fe distance in the lipid layer-like structure. Furthermore,
the trend for 1, 3, 4, 5, and 6 shows that the distance depends
on the alkyl chain length which determines the thickness of the
layer. Only 2 deviates slightly with two signals appearing, one a
bit lower and one a bit higher than expected. It is possible that
during the crystallization process two slightly different phases
shown in Table 1. 4ꢁtol and 6ꢁtol have a sap of around 1 which fits
very well to the obtained lipid layer-like structure in the crystal
packing. However, [FeL(20)(MeOH)2] also shows this kind of
arrangement despite of having a sap of 0.7. In literature, so far
all examined complexes have values around 1 or higher, so it is
possible that values below 1 can also lead to lipid layer-like
structures. Another reason might be, that for [FeL(20)(MeOH)2]
the structure is additionally stabilized by the intermolecular
hydrogen bond network. A similar behavior is observed for other
methanol complexes of this ligand type with shorter alkyl chains.44
In order to analyze, if the fine crystalline samples of 1–6
assume similar structures, the calculated X-ray powder diffrac-
tion pattern of 4ꢁtol and 6ꢁtol are compared with the measured
XRPD patterns of 1–6. The results are given in Fig. 4. Indeed, in
the region of 61–81 2y and 201–251 2y strong similarities in the
diffraction patterns are observed. This can be used as first
indication that in all cases coordination polymers were formed
that assemble in a lipid layer like arrangement of the amphi-
philic molecules with an approximate distance between the
iron centers within the polymer chain of 13–14 Å and a distance
¨
were formed. However, results from Mossbauer spectroscopy
and magnetic measurements indicate that only one independent
iron species is present.
2.3. Magnetic properties
Magnetic measurements were done for all coordination polymers
(1–6), the results are displayed in Fig. 5 as plot of the HS fraction
gHS vs. T. The wMT vs. T plot is shown in Fig. S2 (ESI†). At room
temperature the complexes 1–5 are clearly diamagnetic with a wMT
product in the range between 0.02 and 0.27 cm3 K molꢀ1. 6
undergoes already a partial spin transition at this temperature
and has a wMT value of 0.89 cm3 K molꢀ1. The spin state is
¨
confirmed by room temperature Mossbauer spectroscopy, the
details are summarized in Table 4. For the complexes 1–5
the average values determined at room temperature are
Table 3 XRPD data and the calculated interplanar distances of 1–6
Compound
2y [1]
d [Å]
1
2
3
2.750
2.525/2.765
2.540
32
35/32
35
4
2.465
36
4ꢁtol
2.503a
2.510
35a
35
5
6
6ꢁtol
a
2.360
37
2.316a
38a
Fig. 4 XRPD spectra of 1–6 in the range of 21–301 2y at room tempera-
ture and the calculated XRPD data of the single crystal of 4ꢁtol and 6ꢁtol.
The vertical lines were included as guide for the eye.
Calculated values from the single crystal X-ray structure using
Mercury.57
This journal is ©The Royal Society of Chemistry 2018
J. Mater. Chem. C

View Article Online
Paper
Journal of Materials Chemistry C
HS fraction of 0.17. Considering the differences of the
¨
Lamb–Mossbauer parameters for iron(II) in the HS and the LS
state, this is in good agreement with the results from the
magnetic measurement which show a gHS value of 0.26 at room
temperature.
Upon heating to 400 K the complexes 1–4 with different alkyl
chain lengths and bpey as bridging ligand show a very similar
abrupt, irreversible SCO from LS to HS with T1/2 of 354 K for 1,
347 K for 2, 340 K for 3 and 344/351 K for 4 (top of Fig. 5). The
room temperature wMT product after annealing is with an
average value of 2.79 cm3 K molꢀ1 lower than expected for an
iron(II) in the HS state (details see Table 4). This is most likely
due to the phase transition observed by temperature dependent
polarized optical microscopy and powder XRD. The now
observed gradual SCO already starts around room temperature
and by this leads to reduced wMT values. In order to confirm
that a spin transition took place and to check if it is complete or
¨
not, Mossbauer spectra were recorded of the annealed (heating
to 380 K for a few minutes) complexes at room temperature.
Those measurements confirm that after the first heating the
samples are in the HS state at room temperature (see Fig. 6 and
Table 5). After heating to 380 K (above SCO temperature, but
below the melting point, see below), for the complexes 3–6 a
single HS site is observed. However, in the case of 1 and 2 a
second doublet is observed whose parameters cannot be related
to iron(II) in the HS or the LS state. Comparison with the room
Fig. 5 Magnetic measurements of 1–4 (top) and 4–6 (bottom) with gHS
plotted against T. Temperature sequence: first heating (a), cooling (b),
second heating (c).
¨
temperature Mossbauer parameters of the oxidized complexes
d = 0.342 mm sꢀ1 and DEQ = 1.193 mm sꢀ1. Those are 7–10 (ESI,† Fig. S3 and Table 4) confirms that this doublet
characteristic for a low spin Fe(II) in an octahedral N4O2 belongs to a Fe(III) HS species. This indicates that the compound
coordination sphere.58–60 For complex 6, on the other hand, oxidized either during the heating process or during the Mossbauer
¨
two quadrupole split doublets are observed (Fig. 6 left, Table 4), measurement itself (around 5 days at room temperature). During
of which one is characteristic for iron(II) in the low spin state, the magnetic measurements in the SQUID magnetometer, no
while the other one is characteristic for iron(II) in the HS state. indications for the formation of oxidized species during the heating
Analysis of the relative area of the two doublets reveals a progress are observed. Here, subsequent cooling of the complexes
Table 4 Magnetic properties of 1–10. The magnetic susceptibility data of 1–6 is summarized on the left side. T1/2 is used for the first heating and T1/2* is
used for the following cooling/heating cycles after annealing. On the right side the room temperature 57Fe Mossbauer data are summarized for 1–10.
¨
The SCO coordination polymers 1–6 were measured before and after annealing
¨
Mossbauer studies
Magnetic measurements
Compound wMT (rt) [cm3 K molꢀ1
]
gHS (rt) gHS (50 K) T1/2 and T1/2* [K] Species
d [mm sꢀ1
]
DEQ [mm sꢀ1
]
G/2 [mm sꢀ1
]
Area [%]
1
0.15
2.58
0.05
0.86
—
0.04
354
225
Fe(II) LS
Fe(II) HS 0.96(8)
Fe(III) HS 0.3(3)
Fe(II) LS
Fe(II) HS 0.936(18)
Fe(III) HS 0.35(9)
Fe(II) LS
Fe(II) HS 0.924(4)
Fe(II) LS 0.345(4)
Fe(II) HS 0.951(12)
Fe(II) LS 0.355(3)
Fe(II) HS 0.879(7)
Fe(II) LS 0.389(17)
Fe(II) HS 0.90(7)
Fe(II) HS 0.94(3)
Fe(III) HS 0.334(16)
Fe(III) HS 0.335(13)
Fe(III) HS 0.330(16)
Fe(III) HS 0.31(4)
0.338(5)
1.173(10)
2.11(16)
0.8(6)
0.163(7)
0.238(6)
0.3(2)
0.160(6)
0.18(3)
0.22(12)
0.142(2)
0.158(6)
0.141(6)
0.178(19)
0.155(4)
0.178(11)
0.231(14)
0.21(6)
100
69(14)
31(14)
100
77(8)
22(8)
100
100
100
100
100
1annealed
2
0.02
2.83
0.01
0.94
—
0.28
347
238
0.331(4)
1.175(8)
2.21(4)
0.81(17)
1.203(3)
2.181(8)
1.206(8)
2.14(3)
1.207(5)
2.220(14)
1.10(3)
2.24(17)
2.23(6)
0.77(3)
0.74(2)
0.75(3)
0.73(6)
2annealed
3
0.21
2.79
0.07
2.94
0.27
2.82
0.89
0.07
0.92
0.02
0.98
0.09
0.94
0.26
—
0.08
—
0.18
—
0.79
—
340
216
344/351
199
340/369
—
0.343(2)
3annealed
4
4annealed
5
5annealed
6
100
338
83(4)
17(4)
100
100
100
100
100
6annealed
7
8
9
2.85
—
—
—
—
0.95
—
—
—
—
0.56
—
—
—
—
199
—
—
—
—
0.19(5)
0.17(2)
0.200(18)
0.18(2)
0.23(5)
10
J. Mater. Chem. C
This journal is ©The Royal Society of Chemistry 2018

View Article Online
Journal of Materials Chemistry C
Paper
For 4–6 the alkyl chain length of C22 was kept constant, but
the axial ligand was varied from bpey, to bpee and bpea. At the
bottom of Fig. 5 the results from the temperature dependent
measurement are given that show that the influence of the
bridging ligand on the SCO behavior is stronger in comparison
to a change of the alkyl chain length. This is due to stronger
interactions of the iron center with the bridging ligand than
with the alkyl chains at the outer periphery. From bpey (4) to
bpee (5) and bpea (6) the SCO of the first heating becomes more
and more gradual. This can be explained with the decreasing
rigidity of the bridging ligand. A more rigid system tends to
have a more abrupt SCO. T1/2 of the abrupt part of the SCO is
344/351 K for 4, 340/369 K for 5, and 338 K for 6. Upon cooling
to 50 K, the second SCO is incomplete for bpey (4) and bpea (6)
with a gHS fraction of 0.18 and 0.56, respectively, and almost
disappears for bpee (5) with a gHS fraction of 0.79. In order to
analyze the spin crossover and potential phase transition
behavior in more detail, the magnetic measurements were
repeated for 2 using slightly varied conditions. The sample
was heated up to 354 K (only slightly above T1/2 and signifi-
cantly below any further phase transitions, see TGA/DSC) with
three repeating heating and cooling cycles (Fig. S4, ESI†).
Independent of this, the same initial abrupt SCO is observed
followed by a gradual SCO during the subsequent cooling and
heating cycles. Thus, the SCO behavior is not influenced by any
subsequent phase transitions/melting of the sample above
354 K. When 2 is measured for three heating and cooling cycles
it can be seen that at 50 K gHS increases after each cycle from
0.30 to 0.56 to 0.76 and the gradual SCO is slowly disappearing.
2.4. TGA and DSC
In both single crystal X-ray structures of 4ꢁtol and 6ꢁtol solvent
molecules were observed in the crystal packing. From the
literature it is known that SCO phenomena can be triggered
or influenced by a solvent loss.61,62 Consequently, TGA mea-
surements of the fine crystalline samples 1–6 were performed
to analyze if solvent is included in the crystal packing that
could influence the SCO behavior and explain the differences
between the first and all subsequent heating modes. Please
note that the complexes were dried in vacuum for several hours
prior to all characterization to remove as much solvent as
possible to reduce such effects. The results of the TGA mea-
surements are displayed in the ESI,† Fig. S5. In the case of 2, 3,
5 and 6, the TGA shows a small step around the SCO tempera-
ture which can be associated with some solvent loss. On the
other side, for the complexes 1 and 4 no indications for the
presence of additional solvent in the sample is observed (ESI,†
Table S2). The temperature range and percentage of the weight
loss corresponds best to the inclusion of 0.5 to 1 methanol
molecules per repeating unit of the coordination polymer
(probably from the starting material). However, according to
elemental analysis, the inclusion of toluene is more likely.
Fig. 6 Mossbauer spectra of 1–6 measured at room temperature (left:
¨
before annealing, right: after annealing at 380 K for about 10 min). The blue
doublet corresponds to iron(II) in the low spin state and the read doublet
corresponds to iron(II) in the high spin state. The orange doublet is
characteristic for the corresponding m-O-iron(III) species (high spin state).
The corresponding parameters are summarized in Table 4.
to 50 K reveals a gradual, reversible SCO from HS to LS for all As the magnetic properties of all SCO coordination polymers
complexes with T1/2* values of 225 K for 1, 238 K for 2, 216 K for 3, is similar independent of the presence or absence of included
and 199 K for 4. This SCO is incomplete and at 50 K gHS varies solvent molecules, an influence of the solvent molecules can be
between 0.04 for 1, 0.28 for 2, 0.08 for 3, and 0.18 for 4, respectively. ruled out. An irreversible phase transition accompanying the
This journal is ©The Royal Society of Chemistry 2018
J. Mater. Chem. C

View Article Online
Paper
Journal of Materials Chemistry C
Table 5 Spin-crossover temperature T1/2 obtained from magnetic mea-
surements and thermodynamic parameters calculated from DSC for 1–6
Compound SCO T1/2 [K] DSC Tmax [K] DH [kJ molꢀ1] DS [J Kꢀ1 molꢀ1
]
1
2
3
4
5
6
354
347
340
344/351
340/369
338
354
351
351
353
336/362
340
30.15
45.18
69.57
78.83
73.24
67.85
84.26
125.19
213.49
237.57
203.33
197.32
Fig. 8 Influence of the chain length and of the flexibility of the bridging
ligand on the enthalpy DH and entropy DS.
spin transition could be another reason for the different SCO
properties for the first heating and all subsequent cycles. In
order to analyse this, DSC measurements were performed with
two heating and cooling cycles. The results are summarized in
the ESI,† Fig. S6 and in Table 5. While there are no differences
between the first and the second cooling cycle, in the heating
mode pronounced differences between the first and the second
cycle are observed. For the first heating, each coordination
polymer undergoes an endothermic process around the T1/2 of
the SCO, as illustrated in Fig. 7. The calculated enthalpy (DH)
and entropy (DS) changes exceed by far the expected values for
a phase transition related to those alkyl chains. Furthermore,
the values of DH and DS increase by decreasing flexibility of
the bridging ligand (4–6) (Fig. 8), however, here the changes
are not as pronounced. The evaporation of additional solvent
molecules also has an influence on DH and DS. However, in
our case the influence is too small to be observable as only the
number of C atoms appears to be relevant for the correlation
shown in Fig. 8. The two cooling cycles and the second heating
cycle confirm the assumption that additional phase transi-
tions take place next to the spin transition. The related
structural changes were investigated further using tempera-
ture dependent powder XRD.
Temperature dependent XRPD measurements were carried
out to analyze the structural changes associated to the com-
bined spin and phase transition. Those will be reflected in
changes in the XRPD patterns and can be related to character-
istic Fe–Fe distances in the crystal packing (e.g. between the
layers or interchain). Please note that the XRPD measurements
mentioned in the previous paragraph were measured in a
capillary, whereas the temperature dependent XRPD measure-
ments were measured on a flat plate. Thus, small differences in
the 2y values between those two methods can occur. In Fig. 9,
the temperature dependent changes in the 2.0–3.5 2y region is
displayed for the complexes 2, 4 and 5 as typical examples. The
corresponding powder diffraction patterns of the other three
complexes together with the diffraction patterns in the whole
2y range are given in the ESI,† Fig. S7 and S8. As already
discussed in the X-ray structure analysis section and illustrated
in Table 3, the 2.0–3.5 2y region is characteristic for the
distance between the layers of the lipid-like structure. Upon
heating of 4, three different phases are observed. Starting at
an iron(II) SCO (DH = 10 kJ molꢀ1 and DS = 40 J Kꢀ1 molꢀ1
,
Table 5).1,2,63 This points towards a phase transition taking
place during the spin transition. Either the spin crossover
triggers the phase transition or vice versa. The absolute values
of DH and DS strongly depend on the chain length of the
equatorial ligand and increase with the extension of the alkyl
chain (1–4). Thus, the spin transition is most likely triggered by
Fig. 7 DSC measurements (red, dashed line) and the first derivative of the
magnetic measurements (black, solid line) of 1–6 in the temperature range
between 320 and 380 K illustrating a good agreement between both
methods. The slight difference between the temperatures is most likely
due to the different measurement velocities of the two methods.
Fig. 9 Temperature dependent XRPD spectra of 2, 4 and 5 displayed in
the 2.01–3.51 2y range. The temperatures were selected based on the DSC
measurements and the phase transitions observed therein.
J. Mater. Chem. C
This journal is ©The Royal Society of Chemistry 2018

View Article Online
Journal of Materials Chemistry C
Paper
room temperature upon heating to 338 K a first phase with a
peak maximum at 2.51 2y is observed. At 343 K a second peak
appears at 2.71 2y, that increases in intensity at 348 K while the
intensity of the first peak decreases. At 358 K the first peak
disappeared completely and a third peak starts to appear at 3.01
2y. At 363 K this is the only peak. The increase in the 2y values
upon increasing temperature corresponds to a decrease of the
distances between the layers, thus a rearrangement of the alkyl
chains of the complexes takes place. The other complexes show
similar phase transitions with either two (2 and 5) or three
(1, 3, 4, and 6) different phases. No systematic trend can be
observed if the peak maximum shifts to higher (4 and 5) or
lower (2 and 6) 2y values. In the case of 3 the intermediate
phase at 363 K is shifted to higher 2y values and upon further
heating it goes back to almost the original value, whereas for 1
an opposite trend is observed with the highest temperature
peak in the middle between the two others. Thus the statement
that can be derived from those results is that the phase
transition is accompanied by an irreversible rearrangement
of the alkyl chains leading to changes in the distances between
the polymer chains. Those changes are always in the 2.4–2.9
2y region corresponding to distance changes in the range
between 36.8 to 30.5 Å. The numbers indicate that this
involves a significant structural re-organization that leads to
a loss of cooperativity (and crystallinity) and a gradual SCO is
observed afterwards. Those significant rearrangements also
lead to a shift of peaks in the other 2y regions, as illustrated in
the ESI,† Fig. S8. In order to analyze the irreversible nature of
this rearrangement in more detail, in the case of 1 two
subsequent heating and cooling cycles were investigated. In
the ESI,† Fig. S9, the room temperature XRD pattern before
annealing (1st cycle), after heating to 413 K (2nd cycle) and
after heating to 373 K (3rd cycle) are given. It can be seen that
after the first heating the XRPD pattern did change, however
no further changes are observed for subsequent heating
cycles.
Fig. 10 POM micrographs of 4. All images were taken under crossed
polarized light. Left: With retardation plate, right: without retardation plate.
(A and B) Crystalline powder from synthesis; (C and D) after combined SCO
and phase transition; (E and F) melted sample; (G and H) formation of
birefringent domains after cooling down.
2.5. Polarized optical microscopy
In order to analyze the phase transitions observed by DSC and properties after the first heating and for all subsequent heat-
XRPD in more detail, polarized optical microscopy (POM) ings. For the samples 1–3, 5 and 6 a similar behavior is
pictures of all coordination polymers (1–6) were taken at observed that is illustrated in the ESI,† Fig. S10–S14. In all
different temperatures in the heating and cooling mode. The cases the phase transition associated to the spin transition is
micrographs were recorded with and without a retardation reflected in changes of the POM micrographs. A chain length
plate (first order). In Fig. 10, the different phases of 4 as dependent difference is observed for the final structures. For
function of temperature are shown as typical example. In the complexes 1, 2, and 3 with C16–C20 alkyl chains, spherulites
Fig. 10A and B the crystalline phase at room temperature is are observed after cooling down from the melt, while for the
shown. Upon heating above the SCO temperature (Fig. 10C and D), complexes 4–6 with 22 carbon atoms in the alkyl chain,
the solid–solid phase transition detected by DSC measurements birefringent domains are observed after cooling. The corres-
is reflected in changes in the POM micrograph. Further heating ponding POM micrographs are summarized in Fig. 11. The
resulted in a melting of the sample around 384 K (Fig. 10E and changes of the crystalline phase after heating to 400 K were
F), in line with the outcomes of the DSC measurements. When additionally investigated using scanning electron microscopy
the sample is cooled down after the initial melting, the for- for complex 4. The results are shown in the ESI,† Fig. S15. After
mation of ordered, birefringent domains is observed (Fig. 10G crystallization from solution, the sample consists of plate-like
and H). Thus, from the melt the complex crystallizes in a crystals with a thickness of about 65 nm. After heating and
different phase compared to the crystallization from solution. crystallization from the melt, again plate-like crystals are
This is in good agreement with the differences in the magnetic observed with a similar average thickness (82 nm).
This journal is ©The Royal Society of Chemistry 2018
J. Mater. Chem. C

View Article Online
Paper
Journal of Materials Chemistry C
measure the film thickness a small scratch was carved in the
film with a needle. Due to this some of the removed material
was deposited next to the edge and will be disregarded. Height
profiles at different locations of the film were extracted and the
film thickness and the RMS (root mean square) roughness was
determined. With a concentration of 0.2 mg mLꢀ1 (Fig. S16(A1),
ESI†) no film formation can be observed. However, a network is
formed with bridges of about 2 nm height. By increasing the
concentration to 1.0 mg mLꢀ1 a film with cavities is formed.
The film thickness is 4 nm (Fig. S16(B2), ESI†) and the film
RMS roughness is 1.4 nm (Fig. S16(B1), ESI†). A concentration
of 5 mg mLꢀ1 increases the film thickness up to 22 nm
(Fig. S16(C2), ESI†) while the number of cavities decreases
(Fig. S16(C1), ESI†). The cavities still reach down to the surface
of the silicon wafer. Thus, the RMS roughness increases to
4.8 nm. By further increasing the concentration to 10 mg mLꢀ1
the film thickness increases up to 30 nm (Fig. S16(D2), ESI†). At
this concentration the cavities don’t reach the silicon wafer
anymore and the RMS roughness is decreased to 1.3 nm. The
film becomes more and more homogenous (Fig. S16(D1), ESI†).
For a concentration of 10 mg mLꢀ1 the influence of the spin
speed was investigated, too. An increase from 2000 rpm to
5000 rpm still resulted in the formation of a thin film with a
thickness of 15 nm (Fig. S16(E2), ESI†) and an RMS roughness
of 1.2 nm (Fig. S16(E1), ESI†).
Motivated by those results in the following the iron(II) spin
crossover complex 4 was characterized with regard to film
formation. TEM samples of 4 were prepared to analyze if the
same behavior is observed as for the iron(III)-m-O-complex and
to investigate the structure of the film in more detail. Further-
more, the complex was dissolved in toluene or suspended in
iso-octane to investigate the impact of the solvent on the film
formation. The results are illustrated in Fig. 12. An incomplete
film formation with gaps between the patterns was observed for
the sample of 4 from toluene. The results are similar to those
obtained for the AFM measurements done with 7 with low
concentrations (Fig. S16(A1 and B1), ESI†). It appears that a
similar film formation behavior is observed for both, the
coordination polymer and the dimeric iron(III) complex if
toluene is used as solvent. As 4 was insoluble in iso-octane
the suspension was vortexed and ultrasonicated for some
minutes. In the corresponding TEM pictures thin platelets
and agglomerates of thin platelets are observed. In comparison
Fig. 11 POM micrographs with cross shaped spherulites of 1–3 and the
birefringent domains of 4–6 after the first heating.
2.6. Processing as thin films
One of the advantages of amphiphilic complexes is that they
offer an easy approach towards thin film formation. This is of
importance for potential applications and the construction of
functional devices. Consequently, the suitability of the com-
plexes described in this work for film formation was tested
using the spin coating approach. Due to the air sensitivity of the
iron(II) complexes 1–6, the corresponding m-O-complex 7 was
used for first preliminary investigations on the general suit-
ability of these complexes for spin coating experiments. 7 was
dissolved in toluene and spin coated with different concentra-
tions and spin coating speeds on silicon wafers (Table 6). Spin
coating on glass slides and ODTS functionalized silicon wafers
was also tested, but did not result in a homogeneous film and is
therefore not further discussed. The morphology of the films
was then analyzed by atomic force microscopy (AFM) in tapping
mode (Fig. S16, ESI†). Images with a resolution of 20 ꢂ 20 mm
and 1 ꢂ 1 mm or 10 ꢂ 10 mm were recorded. To be able to
Table 6 Spin coating parameters and film properties of 7
Concentration
Spin speed
[rpm]
Average film
thickness [nm]
RMS roughness
[nm] (image)
[mg mLꢀ1
]
0.2
1.0
5.0
10.0
10.0
2000
2000
2000
2000
5000
—
4
22
30
15
1.5 (A1)
1.4 (B1)
4.8 (C1)
1.3 (D1)
1.2 (E1)
Fig. 12 TEM images of 4 prepared in toluene (A) and iso-octane (B).
J. Mater. Chem. C
This journal is ©The Royal Society of Chemistry 2018

View Article Online
Journal of Materials Chemistry C
Paper
ligand design it is possible to predict the packing of the
molecules in the crystal. The Fe–Fe distance between the layers
depends on the length of the alkyl chains. Temperature depen-
dent XRPD supports the assumption that the spin transition is
coupled to a phase transition which is triggered by a rearrange-
ment of the alkyl chains. This rearrangement is also observable in
POM micrographs and in addition to that ordered, birefringent
domains appear in the solid phase after the melting process. The
film formation behavior was tested for complex 7 using spin
coating. By increasing the concentration from 0.2 mg mLꢀ1 to
10 mg mLꢀ1 the properties of the film significantly improved and
with 10 mg mLꢀ1 a homogenous film with a thickness of 30 nm
and a RMS roughness of 1.3 nm was formed. As expected, an
increase of the spin speed resulted in a reduction of the film
thickness. Preliminary TEM and AFM measurements were done
on films and delaminated crystalline layers of the coordination
polymer 4. Further studies on the film vs. platelet formation and
the corresponding magnetic properties are in progress.
Fig. 13 AFM images of 4 prepared in iso-octane with the corresponding
height profiles.
to the SEM measurements (Fig. S15, ESI†) done for the same
complex before and after annealing at 400 K, the platelets
appear to be much thinner and separated layers can be identi-
fied as seen in Fig. 12B. It is possible that the ultrasonication
procedure in the unpolar solvent triggered a delamination of
the layer-structure observed in the single crystal XRD. For the
dried TEM sample the formation of agglomerates of the thin
plates, as seen on the left side in Fig. 12B, was observed, too. To
determine the thickness of the platelets AFM measurements of
4 dispersed in iso-octane were conducted. The results are
illustrated in Fig. 13 and further images are given in the ESI,†
Fig. S17. The images show again small agglomerates where the
platelets are not perfectly stacked above each other but are
piled up in a random way. This can be seen in the height
profiles as some slopes are increasing constantly while others
increase step-wise. The thickness of the platelets is roughly
between 75 nm (Fig. S17(C), ESI†) and 260 nm (Fig. S17(A),
ESI†). However, for the height of 260 nm it was difficult to
distinguish between multistacking and thicker platelets. Please
note that for the un-treated sample the SEM images (Fig. S15,
ESI†) do not reveal such very thin platelets and the crystallites
appear to be thicker.
Conflicts of interest
There are no conflicts to declare.
Acknowledgements
Financial support of the University of Bayreuth and the German
Science formation (WE 3546/5-1 and SFB 840) is gratefully
acknowledged. L. Peng, P. Pineda, J. Kronawitt and K. Ku¨spert
(MC, University of Bayreuth) are thanked for help with the
calorimetric measurement and the polarized optical microscopy.
We thank P. Mayer (University of Munich) for the collection of
single crystal XRD data of 6ꢁtol.
3. Conclusion
References
In this manuscript the synthesis of six new iron(II) coordination
polymers (1–6) with amphiphilic ligands is reported. All of
them show an abrupt SCO above room temperature. After
initial heating, 1–5 show a gradual and incomplete SCO at
lower temperatures, while 6 remains in the HS state. The spin
state before and after the first heating was confirmed by room
1 Spin-Crossover Materials, ed. M. A. Halcrow, John Wiley &
Sons Ltd, Chichester, 2013.
2 Spin Crossover in Transition Metal Compounds I–III, ed.
P. Gu¨tlich and H. A. Goodwin, Springer Berlin/Heidelberg,
2004, 233–235.
3 P. Gu¨tlich, A. B. Gaspar and Y. Garcia, Spin state switching
in iron coordination compounds, Beilstein J. Org. Chem.,
2013, 9, 342–391.
4 J. Linares, E. Codjovi and Y. Garcia, Pressure and Tempera-
ture Spin Crossover Sensors with Optical Detection, Sensors,
2012, 12, 4479–4492.
5 O. Kahn, C. Jay, J. Krober, R. Claude and F. Groliere, Spin
transition chemical compounds, and devices comprising read-,
memory-, and erase-units, active medium which contains at
least one of those compounds, EP0666561, 1995.
6 O. Kahn, Spin-Transition Polymers: From Molecular Materials
Toward Memory Devices, Science, 1998, 279, 44–48.
¨
temperature Mossbauer spectroscopy. DSC measurements of
the six complexes reveal DH and DS values around the SCO
temperature which are too high to be only associated with a
spin transition. In addition, DH and DS increases with increas-
ing alkyl chain length. This indicates the occurrence of a
second process, namely a phase transition that is coupled to
the spin transition and depends on the alkyl chain length and
by this on the crystal packing. Single crystal X-ray structure
analysis of three complexes ([FeL(20)(MeOH)2], 4ꢁtol and 6ꢁtol)
reveals a lipid layer-like packing of the complexes in the crystal.
The very similar PXRD patterns of all six coordination polymers
indicates a similar packing in all cases that is dominated by the
van der Waals interactions between the alkyl chains. Thus, by
´
7 J.-F. Letard, P. Guionneau and L. Goux-Capes, in Spin Cross-
over in Transition Metal Compounds I–III, ed. P. Gu¨tlich and
This journal is ©The Royal Society of Chemistry 2018
J. Mater. Chem. C

View Article Online
Paper
Journal of Materials Chemistry C
H. A. Goodwin, Springer Berlin/Heidelberg, 2004, 22 S. Hayami, in Spin-Crossover Materials, ed. M. A. Halcrow,
pp. 221–249. John Wiley & Sons Ltd, Chichester, 2013, pp. 321–345.
8 A. Galet, A. B. Gaspar, M. C. Munoz, G. V. Bukin, 23 H. S. Scott, B. Moubaraki, N. Paradis, G. Chastanet, J.-F.
˜
G. Levchenko and J. A. Real, Tunable Bistability in a
Three-Dimensional Spin-Crossover Sensory- and Memory-
Functional Material, Adv. Mater., 2005, 17, 2949–2953.
9 C. D. S. Brites, P. P. Lima, N. J. O. Silva, A. Millan,
Letard, S. R. Batten and K. S. Murray, 2,20-Dipyridylamino-
based ligands with substituted alkyl chain groups and their
mononuclear-M(II) spin crossover complexes, J. Mater.
Chem. C, 2015, 3, 7845–7857.
V. S. Amaral, F. Palacio and L. D. Carlos, Thermometry at 24 M. Seredyuk, A. B. Gaspar, V. Ksenofontov, Y. Galyametdinov,
the nanoscale, Nanoscale, 2012, 4, 4799–4829.
10 O. Kraieva, C. M. Quintero, I. Suleimanov, E. M. Hernandez,
D. Lagrange, L. Salmon, W. Nicolazzi, G. Molnar, C. Bergaud
J. Kusz and P. Gu¨tlich, Does the Solid–Liquid Crystal Phase
Transition Provoke the Spin-State Change in Spin-Crossover
Metallomesogens?, J. Am. Chem. Soc., 2008, 130, 1431–1439.
and A. Bousseksou, High Spatial Resolution Imaging of 25 M. Seredyuk, A. B. Gaspar, V. Ksenofontov, Y. Galyametdinov,
Transient Thermal Events Using Materials with Thermal
Memory, Small, 2016, 12, 6325–6331.
J. Kusz and P. Gu¨tlich, Iron(II) Metallomesogens Exhibiting
Coupled Spin State and Liquid Crystal Phase Transitions
near Room Temperature, Adv. Funct. Mater., 2008, 18,
2089–2101.
´
11 L. Salmon, G. Molnar, D. Zitouni, C. Quintero, C. Bergaud,
J.-C. Micheau and A. Bousseksou, A novel approach for
fluorescent thermometry and thermal imaging purposes 26 S. Hayami, Y. Shigeyoshi, M. Akita, K. Inoue, K. Kato,
using spin crossover nanoparticles, J. Mater. Chem., 2010,
20, 5499.
K. Osaka, M. Takata, R. Kawajiri, T. Mitani and Y. Maeda,
Reverse Spin Transition Triggered by a Structural Phase
Transition, Angew. Chem., Int. Ed., 2005, 44, 4899–4903.
´
12 S. Cobo, G. Molnar, J. A. Real and A. Bousseksou, Multilayer
Sequential Assembly of Thin Films That Display Room- 27 K. Senthil Kumar and M. Ruben, Emerging trends in spin
Temperature Spin Crossover with Hysteresis, Angew. Chem.,
Int. Ed., 2006, 45, 5786–5789.
crossover (SCO) based functional materials and devices,
Coord. Chem. Rev., 2017, 346, 176–205.
´
´
13 E. Coronado, J. R. Galan-Mascaros, M. Monrabal-Capilla, 28 M. Seredyuk, M. C. Munoz, V. Ksenofontov, P. Gutlich,
´
´
´˜
J. Garcıa-Martınez and P. Pardo-Ibanez, Bistable Spin-
Crossover Nanoparticles Showing Magnetic Thermal Hysteresis
near Room Temperature, Adv. Mater., 2007, 19, 1359–1361.
14 J. Larionova, L. Salmon, Y. Guari, A. Tokarev, K. Molvinger,
Y. Galyametdinov and J. A. Real, Spin Crossover Star-Shaped
Metallomesogens of Iron(II), Inorg. Chem., 2014, 53, 8442–8454.
29 T. Romero-Morcillo, M. Seredyuk, M. C. Munoz and J. A.
Real, Meltable Spin Transition Molecular Materials with Tun-
able Tc and Hysteresis Loop Width, Angew. Chem., Int. Ed.,
2015, 54, 14777–14781.
30 P. N. Martinho, Y. Ortin, B. Gildea, C. Gandolfi, G. McKerr,
B. O’Hagan, M. Albrecht and G. G. Morgan, Inducing
hysteretic spin crossover in solution, Dalton Trans., 2012,
41, 7461–7463.
´
G. Molnar and A. Bousseksou, Towards the Ultimate Size
Limit of the Memory Effect in Spin-Crossover Solids, Angew.
Chem., Int. Ed., 2008, 47, 8236–8240.
15 M. Cavallini, I. Bergenti, S. Milita, G. Ruani, I. Salitros, Z.-R.
Qu, R. Chandrasekar and M. Ruben, Micro- and Nano-
patterning of Spin-Transition Compounds into Logical
Structures, Angew. Chem., Int. Ed., 2008, 47, 8596–8600.
16 A. D. Naik, L. Stappers, J. Snauwaert, J. Fransaer and
Y. Garcia, A Biomembrane Stencil for Crystal Growth and
31 P. N. Martinho, C. J. Harding, H. Mu¨ller-Bunz, M. Albrecht
and G. G. Morgan, Inducing Spin Crossover in Amphiphilic
Iron(III) Complexes, Eur. J. Inorg. Chem., 2010, 675–679.
¨
Soft Lithography of a Thermochromic Molecular Sensor, 32 Y. Bodenthin, U. Pietsch, H. Mohwald and D. G. Kurth,
Small, 2010, 6, 2842–2846.
17 Y. Raza, F. Volatron, S. Moldovan, O. Ersen, V. Huc,
C. Martini, F. Brisset, A. Gloter, O. Stephan, A. Bousseksou,
Inducing Spin Crossover in Metallo-supramolecular Poly-
electrolytes through an Amphiphilic Phase Transition,
J. Am. Chem. Soc., 2005, 127, 3110–3114.
L. Catala and T. Mallah, Matrix-dependent cooperativity 33 J. A. Kitchen, N. G. White, C. Gandolfi, M. Albrecht, G. N. L.
in spin crossover Fe(pyrazine)Pt(CN)4 nanoparticles, Chem.
Commun., 2011, 47, 11501–11503.
18 A. B. Gaspar, M. Seredyuk and P. Gu¨tlich, Spin crossover in
metallomesogens, Coord. Chem. Rev., 2009, 253, 2399–2413.
19 A. B. Gaspar and M. Seredyuk, Spin crossover in soft matter,
Coord. Chem. Rev., 2014, 268, 41–58.
20 A. B. Gaspar and B. Weber, in Molecular Magnetic Materials,
ed. B. Sieklucka and D. Pinkowicz, Wiley-VCH Verlag GmbH
& Co. KGaA, Weinheim, Germany, 2017, pp. 231–252.
21 Y. Bodenthin, G. Schwarz, Z. Tomkowicz, M. Lommel,
Jameson, J. L. Tallon and S. Brooker, Room-temperature
spin crossover and Langmuir–Blodgett film formation of an
iron(II) triazole complex featuring a long alkyl chain sub-
stituent: the tail that wags the dog, Chem. Commun., 2010,
46, 6464.
34 P. Coronel, A. Barraud, R. Claude, O. Kahn, A. Ruaudel-
Teixier and J. Zarembowitch, Spin transition in a Langmuir–
Blodgett film, J. Chem. Soc., Chem. Commun., 1989, 193–194.
35 K. Bairagi, O. Iasco, A. Bellec, A. Kartsev, D. Li, J. Lagoute,
C. Chacon, Y. Girard, S. Rousset, F. Miserque, Y. J. Dappe,
A. Smogunov, C. Barreteau, M.-L. Boillot, T. Mallah and
V. Repain, Molecular-scale dynamics of light-induced spin
cross-over in a two-dimensional layer, Nat. Commun., 2016,
7, 12212.
¨
T. Geue, W. Haase, H. Mohwald, U. Pietsch and D. G.
Kurth, Spin-crossover phenomena in extended multi-
component metallo-supramolecular assemblies, Coord.
Chem. Rev., 2009, 253, 2414–2422.
J. Mater. Chem. C
This journal is ©The Royal Society of Chemistry 2018

View Article Online
Journal of Materials Chemistry C
Paper
´
´
¨
36 H. Soyer, E. Dupart, C. J. Gomez-Garcıa, C. Mingotaud and 51 B. Weber and E.-G. Jager, Structure and Magnetic Properties
2ꢀ
`
P. Delhaes, First Magnetic Observation of a Spin Crossover in
of Iron(II/III) Complexes with N2O2 Coordinating Schiff
a Langmuir–Blodgett Film, Adv. Mater., 1999, 11, 382–384.
Base Like Ligands, Eur. J. Inorg. Chem., 2009, 465–477.
37 H. Soyer, C. Mingotaud, M.-L. Boillot and P. Delhaes, Spin 52 B. Weber, Spin crossover complexes with N4O2 coordination
Crossover of a Langmuir–Blodgett Film Based on an Amphi-
sphere – The influence of covalent linkers on cooperative
philic Iron(II) Complex, Langmuir, 1998, 14, 5890–5895.
interactions, Coord. Chem. Rev., 2009, 253, 2432–2449.
¨
38 T. Fujigaya, D.-L. Jiang and T. Aida, Switching of Spin States 53 C. Lochenie, K. Schotz, F. Panzer, H. Kurz, B. Maier,
¨
Triggered by a Phase Transition: Spin-Crossover Properties
of Self-Assembled Iron(II) Complexes with Alkyl-Tethered
Triazole Ligands, J. Am. Chem. Soc., 2003, 125, 14690–14691.
39 O. Roubeau, J. M. Alcazar Gomez, E. Balskus, J. J. A. Kolnaar,
F. Puchtler, S. Agarwal, A. Kohler and B. Weber, Spin-
Crossover Iron(II) Coordination Polymer with Fluorescent
Properties: Correlation between Emission Properties and
Spin State, J. Am. Chem. Soc., 2018, 140, 700–709.
J. G. Haasnoot and J. Reedijk, Spin-transition behaviour in 54 C. Lochenie, J. Heinz, W. Milius and B. Weber, Iron(ii) spin
chains of FeII bridged by 4-substituted 1,2,4-triazoles carry-
ing alkyl tails, New J. Chem., 2001, 25, 144–150.
crossover complexes with diaminonaphthalene-based Schiff
base-like ligands: mononuclear complexes, Dalton Trans.,
2015, 44, 18065–18077.
´
40 O. Roubeau, B. Agricole, R. Clerac and S. Ravaine, Triazole-
Based Magnetic Langmuir–Blodgett Films, J. Phys. Chem. B, 55 A. L. Spek, PLATON – A Multipurpose Crystallographic Tool,
2004, 108, 15110–15116. Utrecht University, Utrecht, The Netherlands, 2008.
41 D. G. Kurth, P. Lehmann and M. Schutte, A route to 56 C. Lochenie, W. Bauer, A. P. Railliet, S. Schlamp, Y. Garcia
hierarchical materials based on complexes of metallosupra-
molecular polyelectrolytes and amphiphiles, PNAS, 2000,
97, 5704–5707.
and B. Weber, Large Thermal Hysteresis for Iron(II) Spin
Crossover Complexes with N-(Pyrid-4-yl)isonicotinamide,
Inorg. Chem., 2014, 53, 11563–11572.
42 S. Schlamp, P. Thoma and B. Weber, Influence of the Alkyl 57 C. F. Macrae, P. R. Edgington, P. McCabe, E. Pidcock,
Chain Length on the Self-Assembly of Amphiphilic Iron
Complexes: An Analysis of X-ray Structures, Chem. – Eur. J.,
2014, 20, 6462–6473.
G. P. Shilds, R. Taylor, M. Towler and J. van de Streek,
Mercury, J. Appl. Crystallogr., 2006, 39, 453–457.
58 W. Bauer, T. Pfaffeneder, K. Achterhold and B. Weber,
Complete Two-Step Spin-Transition in a 1D Chain Iron(II)
Complex with a 110-K Wide Intermediate Plateau, Eur.
J. Inorg. Chem., 2011, 3183–3192.
43 S. Schlamp, K. Dankhoff and B. Weber, Amphiphilic iron(ii)
complexes with short alkyl chains – crystal packing and spin
transition properties, New J. Chem., 2014, 38, 1965–1972.
´
44 S. Schlamp, P. Thoma and B. Weber, New Octahedral, Head- 59 B. Weber, E. S. Kaps, C. Desplanches, J.-F. Letard,
Tail Iron(II) Complexes with Spin Crossover Properties, Eur.
J. Inorg. Chem., 2012, 2759–2768.
45 S. Schlamp, B. Weber, A. D. Naik and Y. Garcia, Cooperative
spin transition in a lipid layer like system, Chem. Commun.,
2011, 47, 7152–7154.
K. Achterhold and F. G. Parak, Synthesis and Characterisa-
tion of Two New Iron(II) Spin-Crossover Complexes with
N4O2 Coordination Spheres – Optimizing Preconditions
for Cooperative Interactions, Eur. J. Inorg. Chem., 2008,
4891–4898.
¨
46 W. Bauer, W. Scherer, S. Altmannshofer and B. Weber, Two- 60 B. Weber, Mossbauer Spectra of FeII Spin crossover Complexes
Step versus One-Step Spin Transitions in Iron(II) 1D Chain
Compounds, Eur. J. Inorg. Chem., 2011, 2803–2818.
with N4O2 coordination sphere, Moessbauer Eff. Ref. Data J.,
2012, 35, 238–254.
47 W. Bauer, S. Schlamp and B. Weber, A ladder type iron(ii) 61 R. Nowak, W. Bauer, T. Ossiander and B. Weber, Slow Self-
coordination polymer with cooperative spin transition,
Chem. Commun., 2012, 48, 10222.
48 B. Weber, R. Betz, W. Bauer and S. Schlamp, Crystal Structure
Assembly Favors Hysteresis above Room Temperature for an
Iron(II) 1D-Chain Spin-Crossover Complex, Eur. J. Inorg.
Chem., 2013, 975–983.
´
of Iron(II) Acetate, Z. Anorg. Allg. Chem., 2011, 637, 102–107. 62 B. Weber, E. S. Kaps, C. Desplanches and J.-F. Letard,
49 M. Tanner and A. Ludi, A facile synthesis of 4,40-dipyridyl-
acetylene, Chimia, 1980, 34, 23–24.
Quenching the Hysteresis in Single Crystals of a 1D Chain
Iron(II) Spin Crossover Complex, Eur. J. Inorg. Chem., 2008,
2963–2966.
50 K. Dankhoff, C. Lochenie, F. Puchtler and B. Weber, Solvent
Influence on the Magnetic Properties of Iron(II) Spin-Crossover 63 P. Gu¨tlich, A. Hauser and H. Spiering, Thermal and Optical
Coordination Compounds with 4,40-Dipyridylethyne as Linker,
Eur. J. Inorg. Chem., 2016, 2136–2143.
Switching of Iron(II) Complexes, Angew. Chem., Int. Ed.,
1994, 33, 2024–2054.
This journal is ©The Royal Society of Chemistry 2018
J. Mater. Chem. C