Thermal and Light-Induced Spin Transitions of FeII Complexes with 4- and 5-(Phenylazo)-2,2′-bipyridine Ligands: Intra- vs. Intermolecular Effects

Article abstract of DOI:10.1002/ejic.201501252

Five new spin crossover complexes with 4- and 5-(phenylazo)-2,2-bipyridine (4-/5-PAbipy) ligands were synthesized and investigated with respect to their spin crossover (SCO) behavior. The results are compared to the thermal and light-induced spin transition properties of the parent SCO complexes [Fe(bpz)₂(bipy)] (1) and [Fe(bipy)₂(NCS)₂] (2). [Fe(bpz)₂(4-PAbipy)] (1a) undergoes a stepwise spin transition whereas [Fe(bpz)₂(5-PAbipy)] (1b) exhibits a one-step transition with a 6 K-wide hysteresis. For [Fe(bpz)₂(^tBu5-PAbipy)] (1c) the spin transition to the low-spin state is incomplete. Qualitatively similar changes of the SCO behavior are observed for the complexes [Fe(4-PAbipy)₂(NCS)₂] (2a) and [Fe(5-PAbipy)₂(NCS)₂] (2b). In comparison to the parent system 2, a strengthening of intermolecular interactions leads to a stabilization of the low-spin state. Evidence for the LIESST behavior could be obtained for all new compounds by means of magnetic susceptibility measurements as well as UV/Vis and resonance Raman spectroscopy.

Full text of DOI:10.1002/ejic.201501252

DOI: 10.1002/ejic.201501252
Full Paper
CLUSTER
ISSUE
Spin Crossover
Thermal and Light-Induced Spin Transitions of FeII Complexes
with 4- and 5-(Phenylazo)-2,2′-bipyridine Ligands: Intra- vs.
Intermolecular Effects
Sven Olaf Schmidt,[a] Holger Naggert,[a] Axel Buchholz,[b] Hannah Brandenburg,[a]
Alexander Bannwarth,[a] Winfried Plass,[b] and Felix Tuczek*[a]
Dedicated to Professor Dr. Philipp Gütlich on the occasion of his 80th birthday
Abstract: Five new spin crossover complexes with 4- and 5- transition to the low-spin state is incomplete. Qualitatively simi-
(phenylazo)-2,2-bipyridine (4-/5-PAbipy) ligands were synthe-
sized and investigated with respect to their spin crossover (SCO)
behavior. The results are compared to the thermal and light-
induced spin transition properties of the parent SCO complexes
[Fe(bpz)2(bipy)] (1) and [Fe(bipy)2(NCS)2] (2). [Fe(bpz)2(4-PA-
bipy)] (1a) undergoes a stepwise spin transition whereas
[Fe(bpz)2(5-PAbipy)] (1b) exhibits a one-step transition with a
6 K-wide hysteresis. For [Fe(bpz)2(tBu5-PAbipy)] (1c) the spin
lar changes of the SCO behavior are observed for the complexes
[Fe(4-PAbipy)2(NCS)2] (2a) and [Fe(5-PAbipy)2(NCS)2] (2b). In
comparison to the parent system 2, a strengthening of inter-
molecular interactions leads to a stabilization of the low-spin
state. Evidence for the LIESST behavior could be obtained for
all new compounds by means of magnetic susceptibility meas-
urements as well as UV/Vis and resonance Raman spectroscopy.
Introduction
hand, have been employed to realize photoinduced spin-state
switching at room temperature and in solution.[13,18,19]
Eight decades after the discovery of the spin crossover (SCO)
phenomenon[1] the exploration of new SCO complexes is still
an active and fascinating field of research. Most of these sys-
tems are based on iron(II),[2] which can exist in a diamagnetic
low-spin (S = 0) and a paramagnetic high-spin (S = 2) state.
The possibility to switch between these two states makes SCO
complexes interesting as building blocks for molecular electron-
ics, spintronics[3–8] or memory devices.[9] For these purposes
light might be the most effective stimulus, and in fact different
mechanisms of light-induced spin change have been estab-
lished over the past decades.[10–15] With regard to possible ap-
plications, the spin change has to occur rapidly, reversibly, and
preferably at room temperature. Due to the relatively low bar-
rier between the 5T2 excited and the 1A1 ground state, the
LIESST effect (light-induced excited spin state trapping) is usu-
ally observed at low temperatures.[10,16,17] The LD-LISC (ligand-
driven light-induced spin change) and LD-CISSS (light-driven
coordination-induced spin state switching) effects, on the other
Possible photoactive substituents which might be employed
in LD-LISC systems to switch the ligand-field strength of coordi-
nated ligands are phenylazo groups. Earlier we analyzed iron(III)
salten complexes [salten = azaheptamethylene-1,7-bis(salicyl-
ideneiminate)][20–24] with one photoswitchable phenylazopyr-
idine (PAP) ligand.[25] However, the PAP ligands were found to
be labile towards photodissociation. So we decided to synthe-
size more photostable 2,2′-bipyridine ligands with phenylazo
substituents (= PAbipy ligands) and explore the LD-LISC proper-
ties of derived FeII complexes. Specifically, we attached phen-
ylazo moieties in 4- and 5-position to the 2,2′-bipyridine unit,
leading to the ligands 4-(phenylazo)-2,2′-bipyridine (4-PAbipy)
and 5-(phenylazo)-2,2′-bipyridine (5-PAbipy). The impact of an
additional bulky substituent in these systems was explored by
introduction of a tert-butyl substituent in the 5-PAbipy ligand,
generating the tBu5-PAbipy ligand.
In order to determine how the new PAbipy ligands influence
the electronic and magnetic properties of FeII SCO complexes,
we used them in place of the unsubstituted bipy counterparts
as ligands for the classic SCO systems [Fe(bpz)2(bipy)] {bpz =
dihydrobis(pyrazolyl)borate}[14,19,26–32] (1) and [Fe(bipy)2(NCS)2]
(2).[33–35] The resulting new complexes [Fe(bpz)2(4-PAbipy)] (1a),
[Fe(bpz)2(5-PAbipy)] (1b), [Fe(bpz)2(tBu5-PAbipy)] (1c) as well as
[Fe(4-PAbipy)2(NCS)2] (2a) and [Fe(5-PAbipy)2(NCS)2] (2b) were
intensively studied in solution as potential LD-LISC systems.[36]
In the course of these investigations, however, it became appar-
ent that these systems also exhibit unusual solid-state SCO
[a] Institut für Anorganische Chemie, Christian-Albrechts-Universität Kiel,
Max-Eyth-Str. 2, 24118 Kiel, Germany
E-mail: ftuczek@ac.uni-kiel.de
https://www.ac.uni-kiel.de/de/tuczek
[b] Institut für Anorganische und Analytische Chemie,
Friedrich-Schiller-Universität Jena,
Humboldtstr. 8, 07743 Jena, Germany
Supporting information for this article is available on the WWW under
http://dx.doi.org/10.1002/ejic.201501252.
Eur. J. Inorg. Chem. 2016, 2175–2186
2175
© 2016 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

Full Paper
properties which are of interest in their own right. Herein these bromopyridine which reacts with nitrosobenzene in a Mills-re-
results are presented, analyzed and compared to corresponding action to 2-bromo-(4/5)-(phenylazo)pyridine. In the next step
data obtained with the parent complexes [Fe(bpz)2(bipy)] (1) this compound is used in a Stille-coupling with 2-trimethyl-
and [Fe(bipy)2(NCS)2] (2), respectively.
stannylpyridine, which was prepared by reaction of 2-bromo-
pyridine and trimethyltin chloride. This synthetic route is differ-
ent to the literature-known procedure for the preparation of
4/5-(phenylazo)-2′,2′-bipyridines (Scheme 1).[50–52] tert-Butyl
functionalized 5-(phenylazo)bipyridine is prepared in a slightly
different way: 4-(tert-butyl)pyridine is stannylated by tri-n-butyl-
tin chloride in hexane; afterwards Stille coupling with the 2-
bromo-5-(phenylazo)pyridine leads to the target tBu5-PAbipy li-
gand.
The influence of a functionalization of 2,2-bipyridine on the
SCO properties of SCO complexes with bipy coligands has been
studied before. Complex 1 exhibits a T1/2 of 160 K[26] whereas
the spin transition temperature of complex 2 has been deter-
mined to 213 K.[33–35] Introduction of a styryl group in 4-posi-
tion and a methyl-group in 4′-position was found to lead to
T1/2 values of 264 K and 380 K for the complex [Fe-
(t-mstbpy)2(NCS)2], respectively.[37] Besides changing the SCO
temperature, functionalization of bipy with large side groups
will also modify long and short-range intermolecular interac-
tions between the complex molecules which strongly influence
the SCO process.[38–41] These effects were also studied for a
series of [Fe(bpz)2(L)][31,42] and [Fe(L)2(NCS)2][41,43–45] complexes
and will be analyzed for our new complexes as well.
Based on the phenylazo-functionalized bipyridine ligands,
four new spin crossover complexes were synthesized:
[Fe(bpz)2(4-PAbipy)] (1a), [Fe(bpz)2 (5-PAbipy)] (1b), [Fe(bpz)2
(
tBu5-PAbipy)] (1c), [Fe(4-PAbipy)2(NCS)2] (2a) and [Fe(5-PA-
bipy)2(NCS)2] (2b) (Scheme 2). The synthesis of complexes 1a
and 1b started from Fe(ClO4)2 hexahydrate in methanolic solu-
tion. First, a solution of potassium dihydro(bispyrazolyl)borate
was added and precipitated KClO4 was removed by filtration.[26]
After addition of phenylazobipyridine, dissolved in methanol, a
dark blue-greenish powder in the case of complex 1a and light
green solids for complexes 1b and 1c were obtained.
Results and Discussion
Syntheses
The ligands have been prepared by a 3-step synthesis involving
For the synthesis of complexes 2a and 2b iron(II) trifluoro-
literature-known procedures,[46–49] starting from 4/5-amino-2- methanesulfonate was used as precursor and the synthesis was
Scheme 1. Ligand syntheses of 4-(phenylazo)-2,2′-bipyridine (4-PAbipy) and 5-(phenylazo)-2,2′-bipyridine (5-PAbipy).
Scheme 2. New iron spin-crossover complexes with phenylazo-functionalized 2,2′-bipyridine ligands (PAbipy): [Fe(bpz)2(4-PAbipy)] (1a), [Fe(bpz)2(5-PAbipy)]
(1b), [Fe(bpz)2(5-tBuPAbipy)] (1c), [Fe(4-PAbipy)2(NCS)2] (2a), [Fe(5-PAbipy)2(NCS)2] (2b).
Eur. J. Inorg. Chem. 2016, 2175–2186
www.eurjic.org
2176
© 2016 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

Full Paper
performed in acetonitrile. Solid potassium thiocyanate was
added and a solution of the phenylazobipyridine in CH3CN was
added to this suspension. After heating to 80 °C for 1 h the
desired complexes precipitated and were isolated by filtration,
yielding a blue solid in the case of compound 2a and a green
solid for complex 2b. Efforts to isolate [Fe(tBu5-PAbipy)2(NCS)2]
(2c) prepared in an analogous fashion failed.
1. [Fe(bpz)2(L)], L = 4-PAbipy, 5-PAbipy and tBu5-PAbipy
a) Thermal and Light-Induced Spin Crossover
The SCO properties of the new complexes [Fe(bpz)2(L)], L = 4-
PAbipy (1a), 5-PAbipy (1b) and tBu5-PAbipy (1c) were first inves-
tigated with a PPMS and a SQUID magnetometer equipped
with an irradiation unit, respectively. In case of 1a the transition
from low-spin to high-spin occurs in a very gradual fashion in
a temperature interval ranging from ca. 20 to 190 K. Between
100 and 150 K a steeper SCO transition is observed, and be-
tween 190 and 195 K a second, abrupt transition occurs (Fig-
ure 1). Both, the steeper and the abrupt change of μeff occur in
a temperature interval of 70 K around the T1/2 value of the
parent system 1 (grey points in Figure 1). Irradiation of 1a at
20 K with λexc = 510 nm causes a spin transition of about 20 %
Figure 2. μeff vs. T for [Fe(bpz)2(5-PAbipy)] (1b). Blue: cooling, grey circles:
irradiation at 20 K with 510 nm, red: heating after irradiation. Grey diamonds:
μeff vs. T plot of [Fe(bpz)2(bipy)] (1).
leading to complex 1c. Unexpectedly, this substitution was
found to largely suppress the thermal spin transition (Figure 3).
Upon warming up a sample of 1c which before was slowly
cooled down to 5 K (blue curve in Figure 3) the magnetic mo-
ment first increases from ≈ 3 to ≈ 3.5 μB (magenta curve in
Figure 3). In the temperature range between 20 and 80 K there
is only a slight increase of the magnetic moment detectable. At
5
(red). The population of the metastable T2 state persists up to
1
ca. 50 K where it relaxes to the A1 ground state. At ca. 70 K
the warm-up curve of the irradiated sample (red) and the
cooling curve (blue) of 1a are identical.
Figure 1. μeff vs. T for [Fe(bpz)2(4-PAbipy)] (1a). Blue: cooling, grey circles:
irradiation at 20 K with 510 nm, red: heating after irradiation. Grey diamonds:
μeff vs. T plot of [Fe(bpz)2(bipy)] (1).
In case of 1b the spin transition of 1 does not change to a
stepwise characteristics as observed for 1a, but instead globally
shifts to lower temperature (95 K; Figure 2). It is steeper than
in 1 and exhibits a 6 K-wide hysteresis. These features indicate
an increase of cooperative intermolecular interactions with re-
spect to the parent system 1.[2] Irradiation of 1b under the same
conditions as applied for 1a leads to a ca. 30 % population of
the metastable high-spin state which persists up to 60 K (red).
In order to weaken the intermolecular interactions which
cause the hysteresis observed for 1b (see below), a tert-butyl
group was introduced in the 5-PAbipy ligand in 4′-position,
Figure 3. μeff vs. T for [Fe(bpz)2(5-tBuPAbipy)] (1c), ◊: [Fe(bpz)2(bipy)], red:
warming; blue: cooling; magenta: warming without irradiation, cyan: cooling
of TIESST experiment, orange: warming after rapid cooling. Bottom: detailed
region < 100 K. Inset: detailed view on the photoexcitation of 1c < 10 K.
Eur. J. Inorg. Chem. 2016, 2175–2186
www.eurjic.org
2177
© 2016 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

Full Paper
85 K a small, gradual spin transition occurs, and between 100 to the same intermolecular iron–iron distance but to a stronger
and 250 K again a very gradual increase of μeff is observed. “interdimer coupling” compared to the parent system 1, caused
Around 80 K, the cooling curve (blue) exhibits a slightly more by π–π stacking. We assume thate this increase of cooperativity
gradual spin transition than the heating curve (magenta; Fig- leads to the observed 6 K-wide hysteresis.
ure 3).
Irradiation of the slowly cooled-down sample at 5 K with
exc = 365 nm increases μeff from 2.87 to 3.20 μB, corresponding
to a partial population of the metastable high-spin state (Fig-
ure 3, bottom, red arrow). Warming up the irradiated sample
leads to a similar increase of μeff as observed for the unirradi-
ated sample, but shifted to higher values by an offset of ca.
0.5 μB (red curve in Figure 3). The population of the metastable
high-spin state persists up to ca. 60 K where the system relaxes
to the thermal state of the unirradiated sample.
In case of complex 1a the one-step spin transition of 1 is
modified to a stepwise characteristics, centered around the
T1/2 value of the parent complex 1.[26] So, the ligand-field
strength does not appear to change greatly whereas the
characteristics of the spin transition is influenced by additional
intermolecular interactions. The phenylazo substituent in 4-po-
sition enhances the π-acceptor capability of the bipy ligand,
which should act to increase the ligand field strength. On the
other hand, the electron-withdrawing effect of the PA group
lowers the σ-donor strength (see above); obviously, in the case
of the 4-PAbipy ligand both effects cancel each other.
λ
It is also possible to trap the excited spin state in complex
1c by rapid cooling of the sample down to 2 K. Subsequent
warming to room temperature (orange curve in Figure 3) leads
to nearly the same characteristics as found for warming the
sample which was irradiated at low temperature (red curve).
Obviously a fraction of SCO molecules has been frozen in the
high-spin state due to the rapid cooling process (TIESST, tem-
perature-induced excited spin-state-trapping).[53–58] Around
80 K there is a dip in the μeff vs. T curve which is caused by
relaxation of the metastable high-spin state, and at 82 K a ther-
mal spin transition occurs which leads to saturation of μeff at
ca. 120 K. Slow recooling of the sample from room temperature
to 5 K exhibits a characteristics which above 100 K is congruent
with the warming curve, but at temperatures < 80 K does not
show the TIESST effect (Figure 3, cyan curve). Nevertheless, the
magnetic moments are globally higher in this curve than in the
sample which has not been subjected to initial rapid cooling.
Usually, a two-step spin transition results from an “antiferro-
magnetic” coupling between the low-spin and the high-spin
species.[2] With regard to 1a, π–π interactions between two bi-
pyridine units and/or two 4-phenylazo substituents of neigh-
bouring complex molecules could lead to a situation where an
alternating high-spin/low-spin arrangement of SCO complexes
is favoured. The low-temperature part of the two-step transition
appears to be associated with such a case. The gradual charac-
teristics could be due to short-range, intermolecular interac-
tions that are comparatively “soft”. The much steeper high-tem-
perature part of the two-step transition, on the other hand,
can be attributed to a transition to the high-spin state that is
associated with long-range, cooperative interactions. The ther-
mal spin transition in 1c, finally, is largely suppressed as this
system is “locked” in the high-spin state due to intermolecular
interactions that hinder the relative motion of the complexes
in the crystal lattice.
Incomplete or even fully suppressed SCO transitions have
been observed for other [Fe(bpz)2(L)] complexes.[31,42] These
findings have been explained by π–π stacking interactions.[31,42]
This way, the bidentate coligands L become interdigitated,
which also strongly influences the crystal packing. In such an
arrangement, the tert-butyl group in the PAbipy backbone fur-
ther hinders the contraction of all complex molecules during
transformation to the low-spin state by blocking intermolecular
motions.
The LIESST properties of 1a, 1b and 1c are also influenced
by the different ligand field strengths of the bipy ligands. Ac-
cording to the “inverse energy-gap” law,[16,17] the tunneling rate
constant kHL of the high-spin→low-spin relaxation increases
1
with an increasing energy gap ΔEHL between the A1 and the
[16]
5T2 state which in turn is related to T1/2
.
This explains why
the light-induced metastable high-spin states in 1b and 1c are
stable over a larger temperature range compared to complex
1a: The minima of the (∂ μeff/∂ T) curves (corresponding to
In summary, major, characteristic changes have been de-
tected between the thermal SCO properties of the new com-
plexes 1a–1c and the parent system [Fe(bpz)2(bipy)] (1).[26] The
most simple effect is observed for 1b containing the 5-phen-
ylazo ligand where the entire spin transition curve is shifted by
about 65 K to lower temperature with respect to 1. This indi-
[59]
TLIESST
)
are located at 67 K for 1a, 70 K for 1b and 80 K for
1c; see Supporting Information, Figures S1–S3) whereas TLIESST
of 1 under the same experimental conditions is determined to
52 K.[36]
b) 57Fe-Mössbauer Spectroscopy
cates a weakening of the ligand-field strength of 5-PAbipy com- In order to obtain further information on the spin transitions
pared to unsubstituted 2,2′-bipyridine. The phenylazo unit is of 1a–1c temperature-dependent Mössbauer spectroscopy was
an electron-withdrawing group, which decreases the electron employed. In complex 1a (Figure 4) one doublet with an isomer
density in the bipyridine backbone and thus leads to a decrease shift δIS of 0.46 mm/s and a small quadrupole splitting ΔEQ of
of the σ-donor strength. The phenylazo unit is located at 5- 0.68 mm/s is observed at 80 K; these parameters are character-
position, so the conjugated π-system is not enlarged to en- istic for an iron(II) species in the low-spin state.[60] At about
hance the π-acceptor properties, but the attachment of the 100 K a new doublet with a large quadrupole splitting emerges,
phenylazo unit in 5-position increases the π–π stacking capabil- which can be assigned to the high-spin-state. Starting from this
ities compared to complex 1. Assuming the common stacking temperature, the high-spin doublet further increases and the
motif for this type of complexes[31,42] the phenylazo group in low-spin doublet further decreases in intensity. As observed in
complex 1b is located orthogonally to the Fe–Fe axis. This leads the susceptibility measurements the “high-temperature” SCO of
Eur. J. Inorg. Chem. 2016, 2175–2186
www.eurjic.org
2178
© 2016 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

Full Paper
the two-step curve occurs abruptly at ca. 190 K (Figure 1). This
is confirmed by the Mössbauer spectra at 190 and 200 K:
whereas the 190 K spectrum clearly shows the contribution of
the low-spin species, the 200 K mainly consists of the high-spin
species. In the latter spectrum, however, there is a small inten-
sity present between the two components of the high-spin
doublet which cannot be fitted by a simple (low-spin) doublet.
The residual intensity rather exhibits an asymmetric line shape
with a quadrupole splitting between that of the high-spin and
the low-spin species. This spectrum may be associated with a
fast high-spin/low spin relaxation process,[61–65] which was also
observed in the homoleptic complex [Fe{HB(pz)3}2] at high tem-
peratures.[65] The 300 K Mössbauer spectrum contains a small
contribution of the fast relaxing species as well. The fact that
the relative intensity of this species is even larger than at 200 K
is ascribed to differences in the temperature dependence of the
Debye–Waller-factor.[60]
Figure 5. Mössbauer spectra of complex 1b. Blue: low-spin, red: high-spin.
Between 100 and 80 K, no appreciable change in the relative
intensities of high spin and low spin species is observed. These
findings correspond to the SQUID data shown in Figure 3 (blue/
magenta curves); i.e., between room temperature and 100 K a
partial spin transition occurs, which, upon further cooling, leads
to a large, temperature-independent high-spin fraction.
Figure 4. Mössbauer spectra of complex 1a. Blue: low-spin, red: high-spin.
The Mössbauer spectra of complex 1b show the thermal be-
havior expected for a generic SCO compound (Figure 5).
At 80 K one doublet with an isomer shift of 0.53 mm/s and
a small quadrupole splitting of 0.63 mm/s characteristic for an
iron(II) species in the low-spin state is observed.[60] Heating the
sample to 90 K leads to the appearance of a high-spin signal
with an isomer shift of 1.15 mm/s and a quadrupole splitting
of 2.74 mm/s. Upon further heating by five degrees the spin
transition is almost completed; i.e., the intensity of the high-
spin is much higher than that of the low-spin doublet. The
spectrum at 300 K exclusively contains the high-spin signal with
an isomer shift of 1.01 mm/s and a quadrupole splitting of
2.02 mm/s.
Figure 6. Mössbauer spectra of complex 1c. Blue: low-spin, red: high-spin.
c) UV/Vis Spectroscopy
In the irradiation experiments conducted in the SQUID magne-
Similar to 1b, the Mössbauer spectrum of 1c contains only tometer the light-induced conversion of 1a–1c to the high-spin
one doublet at 300 K with parameters characteristic of the high- state at low temperatures was found to be incomplete (cf Fig-
spin state (Figure 6). Cooling down to 150 K, the doublet of ures 1 and 2, red traces). In order to check whether a larger
the low-spin state appears which upon further lowering of the fraction of the sample can be switched to the metastable high-
temperature increases in intensity. At 80 K the low-spin doublet spin state, LIESST experiments were performed with optically
exhibits an isomer shift δIS of 0.49 mm/s and a quadrupole thinner samples in a UV/Vis spectrometer. To this end com-
splitting ΔEQ of 0.64 mm/s and the high-spin doublet an isomer plexes 1a and 1b were pressed with KBr to transparent pellets
shift of 1.11 mm/s and a quadrupole splitting of 2.70 mm/s. or deposited as thin films by vacuum evaporation on quartz
Eur. J. Inorg. Chem. 2016, 2175–2186
www.eurjic.org
2179
© 2016 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

Full Paper
discs. Corresponding temperature-dependent spectra are
In summary, both 1a and 1b can be switched to the metasta-
shown in Figure 7. Absorption bands at ca. 290 nm, which ap- ble high-spin state with comparable efficiencies as the parent
pear in the parent complex 1 as well,[27] result from π–π* transi- complex 1. Moreover, the MLCT transitions of 1a and 1b are
tions of the aromatic ligands (bipy and bpz). In analogy to FeIII shifted into the NIR by 130 and 150 nm, respectively, with re-
complexes coordinated by phenylazopyridine ligands,[25] the spect to 1. These changes are due to the electron-withdrawing
π→π* transition of the phenylazo unit is located at 320 nm effect of the phenylazo groups. For 1a, the bathochromic shift
where it also appears in the spectrum of the vacuum-deposited is less pronounced than for 1b as it is partly compensated by an
film of 1a (Figure 7, top left). The band at 780 nm, which is increased backbonding capability of the 4-PAbipy ligand with
most intense in the 4 K-spectrum of this complex (blue), results respect to its unsubstituted counterpart (see above).
from a Fe→4-PAbipy MLCT transition. Interestingly, this band is
shifted to lower energy by about 130 nm compared to the par-
ent complex 1.[27] At room temperature the intensity of this
band is significantly lowered (red curve). A similar intensity de-
crease with increasing temperature is also observed for the
other bands, with exception of the π→π* transition of the
phenylazo unit. Almost the same intensity decrease (84 %) is
observed after irradiation of the sample at 4 K (green curve in
Figure 7, top left) with λexc = 519 nm. This indicates a compara-
ble efficiency of spin-state switching as compared to the parent
system (85 % intensity reduction). Thermal relaxation of the
metastable high-spin-state to the low-spin state the tempera-
ture range from 30 to 70 K is indicated by grey spectra. Similar
observations are made with 1a pressed in a KBr disc (Figure 7,
bottom left). Here, 780 nm band exhibits vibronic structure with
maxima at 707 and 792 nm.
Finally, the intensity ratio between the Fe→PAbipy-MLCT
transition and the combined π→π*-transitions is much larger
for the 4- as compared to the 5-PAbipy system. This difference
is due to the extension of the π-system in the former ligand,
increasing the magnitude of the electric-dipole transition ma-
trix element.[66] In contrast, the phenylazo group in 5-position
only acts to make the bipy ligand less electron-rich. This analy-
sis thus conforms the conclusions drawn from the thermal spin
crossover behavior (cf Section a).
2. [Fe(PAbipy)2(NCS)2] Complexes
a) Magnetic Susceptibility Measurements
In contrast to the parent spin crossover system
[Fe(bipy)2(NCS)2] (2) where a very steep spin transition occurs
at 213 K,[33–35] the spin transitions of both [Fe(PAbipy)2(NCS)2]
complexes 2a and 2b are more gradual and occur near room
temperature (Figure 8). However, there are distinct differences
between 2a and 2b. The spin transition of 2a starts at 200 K
and is not complete at 400 K (Figure 8, filled triangles). At this
temperature the slope of the spin transition curve is small, al-
though with a μeff of 3.01 B.M. only ca. 55 % of the sample have
converted into the high-spin-state yet. In complex 2b the spin
transition proceeds in a similar fashion, but with a μeff value of
5.12 B.M. is almost completed at 400 K.[66]
Figure 7. UV/Vis spectra of vacuum deposited films (top) and KBr pellets (bot-
tom) of complex 1a (left) and 1b (right). Red: room temperature spectra;
green: Irradiation at 4 K for 2 min s; blue: 4 K spectra. Inset: detailed view of
the most intensive MLCT absorption band.
The absorption spectra of 1b show a fairly broad π→π* tran-
sition of the phenylazo unit at 350 nm; at ca. 300 nm the π→π*
transitions of the other aromatic ligands are visible as shoulders
(Figure 7, right). With respect to 1a, the Fe→4-PAbipy MLCT
transition is further shifted into the NIR and now is located at
ca. 800 nm. Again, it is most intense in the 4 K-spectrum and
becomes much weaker at room temperature. Irradiation at 4 K
with light of λexc = 519 nm leads to an almost comparable
(80 %) intensity reduction. As for 1a, grey spectra indicate the
Figure 8. μeff vs. T for [Fe(4-PAbipy)2(NCS)2] (2a, filled triangles) and [Fe(5-
PAbipy)2(NCS)2] (2b, empty triangles). Grey: magnetic measurement of
[Fe(bipy)2(NCS)2], adopted from ref.[35]
The gradual spin transitions of 2a and 2b, as compared to
thermal relaxation of the high spin→low spin state which was the very steep spin transition of the parent system, can be as-
monitored in temperature range from 30 to 70 K. cribed to a decrease of cooperative (long-range) interactions
Eur. J. Inorg. Chem. 2016, 2175–2186
www.eurjic.org
2180
© 2016 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

Full Paper
due to the sterically demanding phenylazo groups, which in- interactions. This effect would act to stabilize the low-spin state
crease the intermolecular distances. Similar conclusions have in 2a and 2b in comparison to 2, because the volume increase
been drawn by Real et al. who found that for [Fe(L)2(NCS)2] during the spin transition would become more unfavorable.
complexes a larger flexibility in the ligand backbone, through,
e.g., aliphatic instead of aromatic units, leads to a decrease of
cooperativity.[41] Furthermore, a study on [Fe(PM-L)2(NCS)2]
complexes (PM-L = 2′-pyridylmethylene-4-aniline) by Kahn et
al. demonstrated that the attachment of a phenylazo unit in-
creases the flexibility of the ligand as compared to other sub-
stituents with an aromatic character and thus leads to a de-
crease of cooperativity.[67,68]
In analogy to complexes 1a and 1b, LIESST experiments were
performed in a SQUID magnetometer equipped with an irradia-
tion unit. In contrast to the former systems, however, the light-
induced high-spin state was not stable in the timescale of the
experiment for 2a and 2b after switching off the light source.
These results indicate a shorter lifetime of the metastable high-
spin state than in 1a–1c, in agreement with the inverse energy-
gap law.[16,17]
While the 4- and 5-phenylazo substituents present in both
2a and 2b render the spin transition more gradual than in the
parent system, they influence the SCO behavior in a different
way; i.e., at 400 K the spin transition in 2b is complete and in
2a incomplete. If the incomplete spin transition of 2a is consid-
ered as the low-temperature part of a two-step spin transition,
introduction of phenylazo substituents into the bipy ligand
would have similar effects on the SCO behavior in the
[Fe{(bpz)2}2(bipy)] and [Fe(bipy)2(NCS)2] systems; i.e., the 5-PA-
bipy ligand changes the steepness of the spin transition and/
b) UV/Vis and Resonance-Raman Spectroscopy
In order to obtain further insight into the temperature and
light-induced SCO behavior of 2a and 2b UV/Vis spectroscopy
was performed on KBr pellets between 4 and 297 K (Figure 9).
In the UV/Vis spectrum of complex 2a at room temperature two
major absorption features are observed (Figure 9, left, red). The
more intense absorption band at 333 nm can be ascribed to
the π→π* transition of the 4-PAbipy ligand.
The Fe→4-PAbipy-MLCT transition appears around 700 nm
or the spin transition temperature T1/2 whereas the 4-PAbipy which, compared to the parent complex 2, corresponds to a
ligand modifies a one-step to a two-step transition. In contrast bathochromic shift of about ca. 100 nm.[34] The small intensity
to 1a where the two-step transition is centered around the increase of the MLCT band upon cooling to 4 K indicates that
T1/2 value of the parent system 1, the SCO curve of 2a is shifted 2a mostly exists in the low-spin state at 300 K (Figure 9, left,
to higher temperatures with respect to 2.
blue), in agreement with the susceptibility data. This may also
be the reason why no d→d transition absorption band is de-
tected around 890 nm like in [Fe(bipy)2(NCS)2] at room temper-
ature.[34] After irradiation of the sample at 4 K with 519 nm the
MLCT absorption band decreases by about 10 %, indicating the
LIESST effect (Figure 9, left, green). Fast relaxation might be the
reason for the small decrease of intensity, which is in accord-
ance with the SQUID-measurements.
What is the reason for the high-temperature shift of T1/2 of
2a and 2b as compared to 2? Intramolecular electronic effects
cannot explain this result as the investigation of 1a, 1b and 1c
showed that the 4- and 5-PAbipy ligands exert a weaker ligand
field than unsubstitued 2,2′-bipyridine. It has been shown that
crystal packing of [Fe(α-diimine)2(NCS)2] complexes is deter-
mined by two kinds of intermolecular interactions: π–π interac-
tions of the diimine ligands and “donor–acceptor” interactions
The Fe→5-PAbipy MLCT transition of complex 2b is also
between the sulfur atom of the isothiocyanate groups and two shifted to lower energies compared to the parent spin crossover
carbon atoms of a diimine ligand.[41] In complexes 2a and 2b complex.[34] The intensity of the corresponding absorption
the electron density in the bipy coligands is reduced through band is smaller than for the Fe→4-PAbipy transition in 2a (Fig-
introduction of phenylazo substituents, which in turn could en- ure 9, right, red), in analogy to complexes 1a and 1b. Moreover,
tail a bigger charge donation from the thiocyanate groups and the high-spin fraction of 2b is 50 % at 300 K. Therefore, only a
thus increase the strength of the intermolecular donor-acceptor relatively minor intensity change occurs upon cooling down
Figure 9. UV/Vis spectra of complex 2a (left) and 2b (right) in KBr. Red: room temperature spectrum, green: Irradiation with 519 nm at 4 K for 2 min, Blue:
4 K spectrum. Inset: detailed view on the MLCT transition.
Eur. J. Inorg. Chem. 2016, 2175–2186
www.eurjic.org
2181
© 2016 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

Full Paper
from 300 to 2 K (Figure 9, right, blue). Irradiation of the sample
In case of complex 2a we used 514 nm as excitation wave-
with 519 nm at 4 K leads to a small intensity decrease of the length λexc. In the temperature range from 40 K to 300 K ther-
MLCT absorption band (Figure 9, right, green). This indicates a mal spin crossover as well as LIESST-behavior could be ob-
partial population of the metastable high-spin state.
served (Figure 10, left). At 40 K (dark) the dominant stretching
vibration is located at ca. 2070 cm–1, which corresponds to the
light-induced metastable high-spin state. Obviously, not all mol-
ecules are excited to the high-spin state, because the low-spin
vibrational mode is present at 2100 cm–1, although with low
intensity.[34,69–73] Above 40 K the high-spin vibration band dis-
appears and the low-spin band becomes dominant, due to re-
laxation which becomes faster at higher temperatures (Fig-
ure 10, right). Between 110 and 250 K the sample exists in the
low-spin state. At 260 K the intensity of the low-spin vibrational
mode decreases and the high-spin stretch at 2070 cm–1 appears
again. At 300 K both spin states are present with comparable
intensities.
Thus, UV/Vis spectroscopy at cryogenic temperatures on
complexes 2a and 2b in principle indicates LIESST behavior for
both complexes, although for both systems the effect is small.
So, we decided to study the spin-switching properties of these
[Fe(PAbipy)2(NCS)2] complexes by temperature-dependent reso-
nance Raman spectroscopy, anticipating that permanent irradi-
ation of the sample by the exciting laser might enable monitor-
ing the metastable high-spin state also in the presence of fast
relaxation, as already exemplified for thin films of a series of
[Fe(bpz)2(L)] complexes.[14,42] In particular the C=N stretching
frequencies of the isothiocyanate coligands should serve as a
sensitive probe to monitor the spin state of the iron(II) metal
center.[34,69–73]
Resonance-Raman data recorded for 2b in the temperature
For the resonance Raman experiments KBr pellets of com- range from 42 to 284 K and with an excitation wavelength of
plexes 2a and 2b were prepared, and overview spectra were 488 nm show the thermal spin crossover and LIESST effect as
recorded at room temperature to determine the most suitable well (Figure 11, left). At 42 K a very intense high-spin C=N
excitation wavelength. Afterwards, the samples were cooled stretch is observed at 2066 cm–1. In the temperature range
down to 30 K. Resonance-Raman spectra were collected at low from 42–70 K the metastable high-spin state prevails; a small
temperature and during warm-up of the sample in a frequency population of the low-spin state is indicated by a vibration at
range from 2425 to 1725 cm–1 (Figures 10 and 11).
2119 cm–1. From 117 K to 155 K the intensity of the low-spin
Figure 10. Resonance-Raman spectra of complex 2a (left, λexc = 514 nm) and intensity of the C=N stretching vibration of the low-spin (black) and the high-
spin state (red) as a function of temperature (right). Grey circles represent the susceptibility data out of Figure 8 in the temperature from 5–300 K (right). The
solid lines serve as a guide to the eye.
Figure 11. Resonance-Raman spectra of complex 2b (left, λexc = 488 nm) and intensity of the C=N stretching vibration of the low-spin (black) and the high-
spin state (red; right). Grey circles represent the susceptibility data of 2b in the temperature from 5–300 K (right). The solid lines serve as a guide to the eye.
Eur. J. Inorg. Chem. 2016, 2175–2186
www.eurjic.org
2182
© 2016 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

Full Paper
Bruker Avance 400 and a Bruker Avance 200 at room temperature
and are referenced to the solvent or TMS.
vibration increases such that it becomes more intense than the
high-spin vibration. Nevertheless, there is a small high-spin frac-
tion present. Till 194 K the intensity ratio of both vibrational
modes remains constant. From 205 K the intensity of the high-
spin stretch at 2066 cm–1 increases continuously (Figure 11,
right). At room temperature, the high-spin intensity is higher
than the low-spin intensity, but the spin transition is not com-
plete yet, in agreement with the magnetic susceptibility data.
Magnetic susceptibility data were obtained from powdered samples
using a Quantum-Design MPMS-5 SQUID magnetometer equipped
with a 5 Tesla magnet in the range from 400 to 2 K. For the LIESST
effect studies the fiber optic sample holder (FOSH, Quantum De-
sign) with a 200 W Hg(Xe) arc lamp, a filter wheel (filter 510 nm,
fwhm: 80 nm), shutter and a multicore fibre (all LOT Oriel) was used.
The temperature was scanned with 2 K/min at an applied field of
5000 Oe. As the exact amount of sample in the FOSH sample cell
is difficult to measure and to estimate the correction for sample
holder and diamagnetism of the sample, the sample mass is calcu-
lated from scaling the data to the experimental data from the nor-
mal measurements for the temperature range 300 to 280 K. The
magnetic measurements without irradiation were performed with
an automated low-temperature physical property measurement sys-
tem (PPMS, Quantum Design). Diamagnetic corrections were ap-
plied with the use of the tabulated Pascal's constants. Mössbauer
spectra were recorded on a WISSEL Mössbauer setup equipped with
an Oxford cryostat. The UV/Vis experiments were performed on a
Cary5000 spectrometer in transmission geometry. For temperature
dependence a CryoVac Kryostat with helium cooling was used. For
illumination experiments 3× LED Luxeon Typ LXML-PM01-0080
(519 nm) was used from Sahlmann Photochemical Solutions. Before
the samples were cooled down to 4 K a room temperature spec-
trum was recorded. After measuring of the 4 K spectrum the sample
was irradiated with 519 nm for 2 min leading to the metastable
situation. After the irradiation and recordation of the thermal relaxa-
tion spectra the samples were warmed up to room temperature
and another 300 K UV/Vis-spectra was recorded to assure the intact-
ness of the samples. For Raman spectroscopic measurements a Dilor
XY-Raman spectrometer (Horiba) was used with an Ar+/Kr+ mixed
gas laser (SpectraPhysics GmbH) operating at 514 nm and 488 nm.
Conclusions
Attachment of phenylazo units in 4- and 5-position of 2,2′-bi-
pyridine leads to characteristic changes in the SCO properties
of classic FeII complexes with one or two bipy coligands: (i)
global shifts of the SCO curve to higher or to lower tempera-
tures with or without hysteresis, (ii) appearance of two-step
transitions which are absent in the parent systems, and (iii) al-
most complete suppression of thermal SCO. These findings can
be attributed to a combination of intra- and intermolecular in-
teractions. An important intramolecular factor is the ligand field
strength which in 4-PAbipy is almost equal to and in 5-PAbipy
is reduced with respect to the unsubstituted bipy ligand. Fur-
thermore, the different PAbipy ligands influence the cooperativ-
ity in the bulk material through short- and long-range interac-
tions. Whereas these interactions generally render spin transi-
tions in solids steeper than in solution, they can also act to
suppress spin transitions in crystalline lattices by blocking rela-
tive motions of complex molecules. Such an effect has been
exemplified for 1c, where also temperature-induced excited
spin state trapping (TIESST) could be observed. A further possi-
ble influence of intermolecular interactions could be evidenced
for 2a and 2b where the shift of T1/2 towards higher tempera-
ture with respect to the parent system 2 is attributed to an
increased strength of intermolecular donor-acceptor interac-
tions.
2-Bromo-(4-phenylazo)pyridine: 4-Amino-5-bromopyridine (2.0 g,
11.6 mmol) was dissolved in 75 mL of pyridine, 33.3 mL (92.8 mmol)
25 % TMAH solution was added, and heated to 80 °C. Nitroso-
benzene 1.24 g (11.6 mmol) was solved in 40 mL of pyridine, added
dropwise and heated at 80 °C for 4 h. After cooling the solution
was extracted 3 times with toluene. The combined organic phases
were dried with MgSO4 and the solvent removed in vacuo. The
residue was chromatographed (column, SiO2), by eluting with cyclo-
hexane/EtOAc: 1/3/1 % TEA, yield 2.18 g (6.39 mmol, 61 %).
C11H8BrN3 (262.11): calcd. C 50.4, H 3.1, N 16.0; found C 50.4, H 3.1,
For all investigated systems, evidence for the LIESST effect
was observed at cryogenic temperatures. In agreement with the
inverse energy-gap law, the lifetimes of the metastable high-
spin states were found to be much larger in the bpz complexes
1a–1c than in their [Fe(PAbipy)2(NCS)2] counterparts 2a and 2b.
Nevertheless, a dynamic population of the metastable high-spin
state by continuous laser excitation could also be detected for
the latter systems in resonance Raman measurements.
1
3
5
N 15.5. H NMR (200 MHz, CDCl3, TMS): δ = 8.55 (dd, J = 5.2, J =
4
5
0.6 Hz, 1 H), 8.02–7.96 (m, 2 H), 7.89 (dd, J = 1.7, J = 0.6 Hz, 1 H),
7.70 (dd, 1 H, 3J = 5.2, 4J = 1.7 Hz, 1 H), 7.55 (m, 3 H) ppm.
13C NMR (101 MHz, CDCl3, TMS): δ = 158.9 (Cq), 152.3 (Cq), 151.5 (Ct),
143.3 (Cq), 133.1 (Ct), 129.5 (Ct), 123.8 (Ct), 120.5 (Ct), 116.6 (Ct) ppm.
2-Bromo-(5-phenylazo)pyridine: To
(8.67 mmol) 5-amino-2-bromopyridine and 947 mg (8.85 mmol)
a solution of 1.50 mg
Experimental Section
General: The Stille-type coupling reactions and the complex syn-
nitrosobenzene in 60 mL of pyridine 354 mg (8.85 mmol) sodium
theses were performed under an inert gas atmosphere using hydroxide in 2 mL of water was added under continuous stirring.
Schlenk techniques. The solvents were dried and freshly distilled
under inert gas. The syntheses of the stannylated pyridines and
After 24 h 60 mL of water was added and the solution was extracted
2 times with 150 mL of toluene. The combined organic phases were
the further ligand syntheses were prepared analogous to literature- dried with MgSO4 and the solvent removed in vacuo. The residue
known procedures.[46–49,74] The syntheses of the [Fe(bpz)2(L)] com-
plexes were prepared using literature procedure as well.[26] All
chemicals were reagent grade from Aldrich or abcr. Elemental analy-
ses were performed using a Euro Vector CHNS-O-element analyzer
(Euro EA 3000). Samples were burned in sealed tin containers by a
stream of oxygen. 1H- and 13C-NMR spectra were recorded with
was chromatographed (column, SiO2), by eluting with cyclohexane/
EtOAc: 1/3/2 % TEA, yield 1.68 g (6.42 mmol, 74 %). C11H8BrN3
1
(262.11): calcd. C 50.4, H 3.1, N 16.0; found C 50.7, H 3.2, N 15.6. H
5
NMR (400 MHz, CDCl3): δ = 8.96 (dd, 4J = 2.6, J = 0.6 Hz, 1 H), 8.01
3
4
3
(dd, J = 8.5, J = 2.6 Hz, 1 H), 7.97–7.90 (m, 2 H), 7.62 (dd, J = 8.5,
5J = 0.6 Hz, 1 H, 1 H), 7.57–7.51 (m, 3 H) ppm. 13C NMR (75 MHz,
Eur. J. Inorg. Chem. 2016, 2175–2186
www.eurjic.org
2183
© 2016 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

Full Paper
CDCl3): δ = 152.4 (Cq), 148.2 (Ct), 147.2 (Cq), 144.1 (Cq), 132.2 (Ct), (75 MHz, [D8]toluene): δ = 160.7 (Cq), 158.9 (Cq), 155.8 (Cq), 153.2
129.4 (Ct), 129.1 (Ct), 128.8 (Ct), 123.3 (Ct) ppm.
(Cq), 149.7 (Ct), 148.1 (Cq), 147.7 (Ct), 131.6 (Ct), 129.2 (Ct), 127.4 (Ct),
123.5 (Ct), 121.9 (Ct), 121.4 (Ct), 118.9 (Ct), 34.8 (Cq), 30.5 (Cs) ppm.
(4-Phenylazo)-2,2′-bipyridine (4-PAbipy): Through a mixture of
1.80 g (6.90 mmol) 2-bromo-(4-phenylazo)pyridine and 1.1 mL
(6.20 mmol) (2-trimethylstannyl)pyridine in 30 mL of m-xylene was
degassed by an N2-stream for 1 h. After the addition of 80 mg
(1 mol-%) [Pd(PPh3)4] the solution was heated to 140 °C for 15 h.
After cooling down to room temperature the mixture was added to
[Fe(bpz)2(4-PAbipy)] (1a): To a solution of 150 mg (0.41 mmol)
iron(II) perchlorate hexahydrate in 5 mL of methanol 153 mg
(0.82 mmol) potassium dihydro(bispyrazolyl)borate, dissolved in
5 mL of methanol, was added. After 15 min the precipitated KClO4
was removed by filtration and washed with 5 mL of methanol. A
solution of 107 mg (0.41 mmol) 4-(phenylazo)-2,2′-bipyridine in
10 mL of methanol was added dropwise. A greenish precipitate was
yielded after filtration, washing with methanol and dried in vacuo,
yield 141 mg (230 μmol, 56 %). C28H28B2FeN12 (610.08): calcd. C
55.1, H 4.6, N 27.6; found C 54.7, H 4.9, N 27.2.
40 mL of 2
N sodium hydroxide solution and the phases were sepa-
rated. The aqueous phase was extracted 2 times with 150 mL of
toluene and the organic phases were dried with MgSO4 and the
solvent was removed in vacuo. The residue was chromatographed
(column, SiO2), by eluting with cyclohexane/EtOAc: 1/3/2 % TEA,
yield 1.43 g (5.50 mmol, 80 %). C16H12N4 (260.30): calcd. C 73.8,
H 4.7, N 21.5; found C 73.4, H 4.6, N 21.0. 1H NMR (400 MHz, [D8]tol-
[Fe(bpz)2(5-PAbipy)] (1b): To a solution of 726 mg (2 mmol) iron(II)
perchlorate hexahydrate in 10 mL of methanol 744 mg (4.00 mmol)
potassium dihydro(bispyrazolyl)borate, dissolved in 10 mL of meth-
anol, was added. After 15 min the precipitated KClO4 was removed
by filtration and washed with 5 mL of methanol. A solution of
521 mg (2.00 mmol) 5-(Phenylazo)-2,2′-bipyridine in 10 mL of meth-
anol was added dropwise. A greenish precipitate was yielded after
filtration, washing with methanol and dried in vacuo, yield 673 mg
(1.10 mmol, 55 %). C28H28B2FeN12 (610.08): calcd. C 55.1, H 4.6, N
27.6; found C 55.0, H 4.9, N 27.3.
4
5
3
uene): δ = 9.31 (dd, J = 1.9, J = 0.7 Hz, 1 H), 8.65 (ddd, J = 8.0,
4J = 1.2, J = 0.9 Hz, 1 H), 8.61 (dd, J = 5.1, J = 0.7 Hz, 1 H), 8.51
5
3
5
(ddd, 3J = 4.7, 4J = 1.8, 5J = 0.9 Hz, 1 H), 7.88 (m, 2 H), 7.46 (dd, 3J =
4
3
3
4
5.1, J = 1.9 Hz, 1 H), 7.24 (ddd, J = 8.0, J = 7.5, J = 1.8 Hz, 1 H),
7.12 (m, 3 H), 6.73 (ddd, 3J = 7.5, 3J = 4.7, 4J = 1.2 Hz, 1 H) ppm.
13C NMR (75 MHz, CDCl3): δ = 158.6 (Cq), 158.3 (Cq), 155.8 (Cq), 152.5
(Cq), 150.8 (Ct), 149.5 (Ct), 137.2 (Ct), 132.5 (Ct), 129.4 (Ct), 124.2 (Ct),
123.6 (Ct), 121.4 (Ct), 116.3 (Ct), 114.1 (Ct) ppm.
[Fe(bpz)2(tBu5-PAbipy)] (1c): To a solution of 150 mg (413 μmol)
iron(II) perchlorate hexahydrate in 5 mL of methanol 154 mg
(828 μmol) potassium dihydro(bispyrazolyl)borate, dissolved in 5 mL
of methanol, was added. After 15 min the precipitated KClO4 was
removed by filtration and washed with 5 mL of methanol. A solu-
tion of 133 mg (420 μmol) 4′-tBu-5-(phenylazo)-2,2′-bipyridine in
20 mL of methanol was added dropwise. A greenish precipitate was
yielded after filtration, washing with methanol and dried in vacuo,
yield 92.0 mg (138 μmol, 33 %). C32H36B2FeN12 (666.18): calcd. C
57.7, H 5.5, N 25.2; found C 58.2, H 5.6, N 25.2.
(5-Phenylazo)-2,2′-bipyridine (5-PAbipy): Through a mixture of
642 mg (2.65 mmol) 2-bromo-5-(phenylazo)pyridine and 0.45 mL
(2.40 mmol) (2-trimethylstannyl)pyridine in 30 mL of m-xylene was
degassed by an N2 stream for 1 h. After the addition of 31 mg
(1 mol-%) [Pd(PPh3)4] the solution was heated to 120 °C for 20 h.
After cooling down to room temperature the mixture was added to
40 mL of 2
N sodium hydroxide solution and the phases were sepa-
rated. The aqueous phase was extracted 2 times with 70 mL of
toluene and the organic phases were dried with MgSO4 and the
solvent was removed in vacuo. The residue was chromatographed
(column, SiO2), by eluting with cyclohexane/EtOAc: 1/3/2 % TEA,
yield 468 mg (1.79 mmol, 68 %). C16H12N4 (260.30): calcd. C 73.8,
H 4.7, N 21.5; found C 74.1, H 4.7, N 21.7. 1H NMR (400 MHz, [D8]tol-
uene): δ = 9.36 (dd, 4J = 2.4, 5J = 0.7 Hz, 1 H), 8.74 (dd, 3J = 8.6,
[Fe(4-PAbipy)2(NCS)2] (2a): In a 100 mL Schlenk flask 100 mg
(0.28 mmol) iron(II) trifluoromethanesulfonate and 55.0 mg
(0.57 mmol) potassium thiocyanate were mixed in 5 mL of CH3CN.
Afterwards 221 mg (0.85 mmol) 5-(phenylazo)-2,2′-bipyridine, dis-
solved in 20 mL of warm CH3CN, were added and the mixture was
heated to 80 °C for 3 h. After coolind down a blue solid was filtered
off and dried in vacuo, yield 170 mg (0.25 mmol, 88 %).
3
4
5
5J = 0.7 Hz, 1 H), 8.65 (ddd, J = 8.0, J = 1.2, J = 0.9 Hz, 1 H), 8.49
(ddd, 3J = 4.7, 4J = 1.8, 5J = 0.9 Hz, 1 H), 8.02 (dd, 3J = 8.5, 4J =
2.4 Hz, 1 H), 7.90 (m, 2 H, 8-H), 7.22 (ddd, 3J = 8.0, 3J = 7.5, 4J =
3
3
4
1.8 Hz, 1 H), 7.15 (m, 3 H), 6.72 (ddd, J = 7.5, J = 4.7, J = 1.2 Hz,
1 H) ppm. 13C NMR (126 MHz, CDCl3): δ = 158.0 (Cq), 155.5 (Cq),
152.8 (Cq), 149.5 (Ct), 147.9 (Cq), 147.5 (Ct), 137.2 (Ct), 131.9 (Ct), 129.4
(Ct), 127.6 (Ct), 124.3 (Ct), 123.1 (Ct), 121.8 (Ct), 121.7 (Ct) ppm.
C34H24FeN10S2 (692.60): calcd. C 59.0, H 3.5, N 20.2, S 9.3; found C
58.7, H 3.3, N 19.8, S 9.5.
[Fe(5-PAbipy)2(NCS)2] (2b): In a 100 mL Schlenk flask 100 mg
(0.28 mmol) iron(II) trifluoromethanesulfonate and 55.0 mg
(0.57 mmol) potassium thiocyanate were mixed in 5 mL of CH3CN.
Afterwards 221 mg (0.85 mmol) 5-(phenylazo)-2,2′-bipyridine, dis-
solved in 20 mL of warm CH3CN, were added and the mixture was
heated to 80 °C for 3 h. After cooling down a green solid was fil-
tered off and dried in vacuo, yield 158 mg (0.22 mmol, 81 %).
C34H24FeN10S2 (692.60): calcd. C 59.0, H 3.5, N 20.2, S 9.3; found C
58.6, H 3.4, N 20.4, S 9.5.
4′-tBu-(5-Phenylazo)-2,2′-bipyridine (tBu5-PAbipy): A solution of
800 mg (3.05 mmol) 2-bromo-5-(phenylazo)pyridine and 176 mg
(153 μmol) [Pd(PPh3)4] in 20 mL of toluene was stirred for 20 min.
Afterwards 1.29 g (3.05 mmol) 2-(tributylstannyl)pyridine was
added and the solution was heated for 12 h at 135 °C. After the
reaction was cooled to room temperature the suspension was fil-
tered through celite. 20 mL of satd. NaHCO3 solution was added
and the org. layer was separated. Then the aqueous phase were
extracted with dichloromethane. After combining of the org. phases
the solution was dried with MgSO4 and the solvent was removed
in vacuo. The residue was chromatographed (column, SiO2), by elut-
ing with cyclohexane/EtOAc: 10/1/2 % TEA, yield 614 mg
(1.94 mmol, 64 %). C20H20N4 (316.40): calcd. C 75.9, H 6.4, N 17.7;
found C 76.2, H 6.6, N 17.6. 1H NMR (400 MHz, [D8]toluene): δ =
Acknowledgments
F. T. thanks Deutsche Forschungsgemeinschaft (DFG) for fund-
ing this research (SFB 677). The authors thank Dr. Luca Carella
and Prof. E. Rentschler (Johannes Gutenberg University Mainz)
for the SQUID-measurements on the complex [Fe(bpz)2(bipy)].
The authors thank Maren Rasmussen and Henning Lühmann
4
5
4
5
9.41 (dd, J = 2.4, J = 0.7 Hz, 1 H), 8.91 (dd, J = 2.0, J = 0.7 Hz, 1
H), 8.85 (dd, 3J = 8.6, 5J = 0.7 Hz, 1 H), 8.54 (dd, 3J = 5.2, 5J = 0.7 Hz,
1 H), 8.06 (dd, 3J = 8.6, J = 2.4 Hz, 1 H), 7.90 (m, 2 H), 7.15 (m, 3
4
H), 6.91 (dd, 3J = 5.2, 4J = 2.0 Hz), 1.14 (s, 9 H) ppm. 13C NMR for the TIESST experiments. S. O. S. also thanks Tammy Jacob-
Eur. J. Inorg. Chem. 2016, 2175–2186
www.eurjic.org
2184
© 2016 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

Full Paper
[31] R. Kulmaczewski, H. J. Shepherd, O. Cespedes, M. A. Halcrow, Inorg. Chem.
2014, 53, 9809–9817.
[32] M. Bernien, H. Naggert, L. M. Arruda, L. Kipgen, F. Nickel, J. Miguel, C. F.
Hermanns, A. Krüger, D. Krüger, E. Schierle, E. Weschke, F. Tuczek, W.
Kuch, ACS Nano 2015, 9, 8960–8966.
[33] W. A. Baker Jr., H. M. Bobonich, Inorg. Chem. 1964, 3, 1184–1188.
[34] E. König, K. Madeja, K. J. Watson, J. Am. Chem. Soc. 1968, 90, 1146–1153.
[35] E. W. Müller, H. Spiering, P. Gütlich, Chem. Phys. Lett. 1982, 93, 567–571.
[36] S. O. Schmidt, unpublished results.
sen-Bialas for synthetic work and Stephanie Pehlke for record-
ing resonance Raman measurements.
Keywords: Iron · Spin crossover · Magnetic properties ·
Moessbauer spectroscopy · Resonance Raman spectroscopy ·
LIESST
[37] M.-L. Boillot, S. Chantraine, J. Zarembowitch, J.-Y. Lallemand, J. Prunet,
New J. Chem. 1999, 23, 179–183.
[38] C. P. Köhler, R. Jakobi, E. Meissner, L. Wiehl, H. Spiering, P. Gütlich, J. Phys.
Chem. Solids 1990, 51, 239–247.
[39] A. Hauser, J. Jeftic, H. Romstedt, R. Hinek, H. Spiering, Coord. Chem. Rev.
1999, 190–192, 471–491.
[40] H. Spiering, T. Kohlhaas, H. Romstedt, A. Hauser, C. Bruns-Yilmaz, J. Kusz,
P. Gütlich, Coord. Chem. Rev. 1999, 190–192, 629–647.
[41] J. A. Real, A. B. Gaspar, V. Niel, M. C. Munoz, Coord. Chem. Rev. 2003, 236,
121–141.
[42] H. Naggert, J. Rudnik, L. Kipgen, M. Bernien, F. Nickel, L. M. Arruda, W.
Kuch, C. Näther, F. Tuczek, J. Mater. Chem. C 2015, 3, 7870–7877.
[43] P. Guionneau, M. Marchivie, G. Bravic, J.-F. Letard, D. Chasseau, Top. Curr.
Chem. 2004, 234, 97–128.
[1] L. Cambi, L. Szegö, Ber. Dtsch. Chem. Ges. 1931, 64, 2591.
[2] P. Gütlich, A. Hauser, H. Spiering, Angew. Chem. Int. Ed. Engl. 1994, 33,
202024, 2024–2054–2141; Angew. Chem. 1994, 106, 2109; P. Gütlich, H. A.
Goodwin, Spin Crossover in Transition Metal Compounds I–III, Top. Curr.
Chem., Springer, Berlin, 2004, vol. 233–235; P. Gütlich, H. A. Goodwin,
Top. Curr. Chem. 2004, 233–235.
[3] O. Kahn, J. Kröber, C. Jay, Adv. Mater. 1992, 4, 718–728.
[4] S. Sanvito, Chem. Soc. Rev. 2011, 40, 3336–3355.
[5] J. Letard, P. Guinneau, L. Goux-Capes, Top. Curr. Chem. 2004, 235, 221–
249.
[6] A. Bousseksou, G. Molnar, L. Salmon, W. Nicolazzi, Chem. Soc. Rev. 2011,
40, 3313–3335.
[7] M. C. Munoz, J. A. Real, Coord. Chem. Rev. 2011, 255, 2068–2093.
[8] M. A. Halcrow, Spin-Crossover Materials – Properties and Applications,
Wiley-VCH, Weinheim, Germany, 1st ed., 2013.
[44] J.-F. Letard, J. Mater. Chem. 2006, 16, 2550–2559.
[45] A. Ozarowski, B. R. McGarvey, A. B. Sarkar, J. E. Drake, Inorg. Chem. 1988,
27, 628–635; B. Gallois, J. A. Real, C. Hauw, J. Zarembowitch, Inorg. Chem.
1990, 29, 1152–1158; M. Konno, M. Mikami-Kido, Bull. Chem. Soc. Jpn.
1991, 64, 339–345; J. A. Real, B. Gallois, T. Granier, F. Suez-Panama, J.
Zarembowitch, Inorg. Chem. 1992, 31, 4972–4979; J. A. Real, M. C. Munoz,
E. Andres, T. Granier, B. Gallois, Inorg. Chem. 1994, 33, 3587–3594; J.-F.
Letard, S. Montant, P. Guionneau, P. Martin, A. Le Calves, E. Freysz, D.
Chasseau, R. Lapouyade, O. Kahn, Chem. Commun. 1997, 745–746; J.-F.
Letard, P. Guionneau, E. Codjovi, O. Lavastre, G. Bravic, D. Chasseau, O.
Kahn, J. Am. Chem. Soc. 1997, 119, 10861–10862; Z. J. Zhong, J. Q. Tao,
Z. Yu, C. Y. Dun, Y. L. Liu, X. Z. You, J. Chem. Soc., Dalton Trans. 1998,
327–328; N. Moliner, M. C. Munoz, S. Letard, J.-F. Letard, X. Solans, R.
Burriel, M. Castro, O. Kahn, J. A. Real, Inorg. Chim. Acta 1999, 291, 279–
288; J. Klingele, D. Kaase, M. Schmucker, Y. Lan, G. Chastanet, J.-F. Letard,
Inorg. Chem. 2013, 52, 6000–6010.
[46] C. Brotschi, G. Mathis, C. J. Leumann, Chem. Eur. J. 2005, 11, 1911–1923.
[47] S. H. Wadman, Y. M. van Leeuwen, R. W. A. Havenith, G. P. M. van Klink,
G. van Koten, Organometallics 2010, 29, 5635–5645.
[48] B. W. Michel, L. D. Steffens, M. S. Sigma, J. Am. Chem. Soc. 2011, 133,
8317–8325.
[49] E. V. Brown, G. R. Granneman, J. Am. Chem. Soc. 1975, 97, 621–627.
[50] Y. Uchida, R. Kajita, Y. Kawasaki, S. Oae, Phosphorus Sulfur Silicon Relat.
Elem. 1994, 93–94, 405–409.
[51] O. S. Wenger, L. M. Henling, M. W. Day, J. R. Winkler, H. B. Gray, Polyhedron
2004, 23, 2955–2958.
[9] O. Kahn, C. J. Martinez, Science 1998, 279, 44–48.
[10] S. Decurtins, P. Gütlich, C. P. Köhler, H. Spiering, A. Hauser, Chem. Phys.
Lett. 1984, 105, 1–4.
[11] M.-L. Boillot, J. Zarembowitch, A. Sour, Top. Curr. Chem. 2004, 234, 261–
276.
[12] S. Thies, H. Sell, C. Schütt, C. Bornholdt, C. Näther, F. Tuczek, R. Herges,
J. Am. Chem. Soc. 2011, 133, 16243–16250; S. Thies, C. Bornholdt, F.
Köhler, F. Soennichsen, C. Näther, F. Tuczek, R. Herges, Chem. Eur. J. 2010,
16, 10074–10083.
[13] S. Venkataramani, U. Jana, M. Dommaschk, F. D. Sönnichsen, F. Tuczek,
R. Herges, Science 2011, 331, 445–448.
[14] E. Ludwig, H. Naggert, M. Kalläne, S. Rohlf, S. Kröger, A. Bannwarth, A.
Quer, K. Rossnagel, L. Kipp, F. Tuczek, Angew. Chem. Int. Ed. 2014, 53, 11,
3019–3023; Angew. Chem. 2014, 126, 3063–3067.
[15] A. Witt, F. W. Heinemann, M. M. Khusniyarov, Chem. Sci. 2015, 6, 4599–
4609.
[16] A. Hauser, Coord. Chem. Rev. 1991, 111, 275–290.
[17] A. Hauser, C. Enachescu, M. Lawson Daku, A. Vargas, N. Amstutz, Coord.
Chem. Rev. 2006, 250, 1642–1652.
[18] Y. Hasegawa, S. Kume, H. Nishihara, Dalton Trans. 2009, 280–284.
[19] M. Milek, F. W. Heinemann, M. M. Khusniyarov, Inorg. Chem. 2013, 52,
11585–11592.
[20] N. Matsumoto, S. Ohta, C. Yoshimura, A. Ohyoshi, S. Kohata, H. Okawa,
Y. Maeda, J. Chem. Soc., Dalton Trans. 1985, 2575–2584.
[21] A. Sour, M.-L. Boillot, E. Riviere, P. Lesot, Eur. J. Inorg. Chem. 1999, 2117–
2119.
[22] S. Hirose, S. Hayami, Y. Maeda, Bull. Chem. Soc. Jpn. 2000, 73, 2059–2066.
[23] C. Krüger, P. Augustin, I. Nemec, Z. Travnicek, H. Oshio, R. Boca, F. Renz,
Eur. J. Inorg. Chem. 2013, 902–915.
[52] A. Amar, P. Savel, H. Akdas-Kilig, C. Katan, H. Meghezzi, A. Boucekkine, J.-
P. Malval, J.-L. Fillaut, Chem. Eur. J. 2015, 21, 8262–8270.
[53] G. Ritter, E. König, W. Irler, H. A. Goodwin, Inorg. Chem. 1978, 17, 224–
228.
[54] H. A. Goodwin, K. H. Sugiyarto, Chem. Phys. Lett. 1987, 139, 470–474.
[55] T. Buchen, P. Gütlich, H. A. Goodwin, Inorg. Chem. 1994, 33, 4573–4576.
[56] T. Buchen, P. Gütlich, K. H. Sugiyarto, H. A. Goodwin, Chem. Eur. J. 1996,
2, 1134–1138.
[57] N. Moliner, A. B. Gaspar, M. Carmen Munoz, V. Niel, J. Cano, J. A. Real,
Inorg. Chem. 2001, 40, 3986–3991.
[24] C. Krüger, P. Augustin, L. Dlhan, J. Pavlik, J. Moncol, I. Nemec, R. Boca, F.
Renz, Polyhedron 2015, 87, 194–201.
[25] A. Bannwarth, S. O. Schmidt, G. Peters, F. D. Sönnichsen, W. Thimm, R.
Herges, F. Tuczek, Eur. J. Inorg. Chem. 2012, 2776–2783.
[26] J. A. Real, M. C. Munoz, J. Faus, X. Solans, Inorg. Chem. 1997, 36, 3008–
3013.
[58] K. D. Murnaghan, C. Carbonera, L. Toupet, M. Griffin, M. M. Dirtu, C. Des-
planches, Y. Garcia, E. Collet, J.-F. Letard, G. G. Morgan, Chem. Eur. J. 2014,
20, 5613–5618.
[59] J.-F. Letard, P. Guionneau, L. Rabardel, J. A. K. Howard, A. E. Goeda, D.
Chasseau, O. Kahn, Inorg. Chem. 1998, 37, 4432–4441.
[27] H. Naggert, A. Bannwarth, S. Chemnitz, T. von Hofe, E. Quandt, F. Tuczek,
Dalton Trans. 2011, 40, 6364–6366.
[28] T. G. Gopakumar, F. Matino, H. Naggert, A. Bannwarth, F. Tuczek, R.
Berndt, Angew. Chem. Int. Ed. 2012, 51, 25, 6262–6266; Angew. Chem.
2012, 124, 6367–6371.
[60] P. Gütlich, R. Link, A. Trautwein, Mössbauer Spectroscopy and Transition
Metal Chemistry, Springer, Berlin, Heidelberg, New York, 1978.
[61] P. Adler, H. Spiering, P. Gütlich, Inorg. Chem. 1987, 26, 3840–3845.
[62] P. Adler, H. Spiering, P. Gütlich, Hyperfine Interact. 1988, 42, 1033–1038.
[63] P. Adler, A. Hauser, A. Vef, H. Spiering, P. Gütlich, Hyperfine Interact. 1989,
47, 343–356.
[29] T. G. Gopakumar, M. Bernien, H. Naggert, F. Matino, C. F. Hermanns, A.
Bannwarth, S. Mühlenberend, A. Krüger, D. Krüger, F. Nickel, W. Walter, R.
Berndt, W. Kuch, F. Tuczek, Chem. Eur. J. 2013, 19, 15702–15709.
[30] B. Rösner, M. Milek, A. Witt, B. Gobaut, P. Torelli, R. H. Fink, M. M. Khusni-
yarov, Angew. Chem. Int. Ed. 2015, 54, 44, 12976–12980; Angew. Chem.
2015, 127, 13168–13172.
Eur. J. Inorg. Chem. 2016, 2175–2186
www.eurjic.org
2185
© 2016 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

Full Paper
[64] P. Adler, P. Poganiuch, H. Spiering, Hyperfine Interact. 1989, 52, 47–63.
[65] F. Grandjean, G. J. Long, B. B. Hutchinson, L. Ohlhausen, P. Neill, J. D.
Holcomb, Inorg. Chem. 1989, 28, 4406–4414.
[70] R. H. Herber, Inorg. Chem. 1987, 26, 173–178.
[71] E. König, K. Madeja, Spectrochim. Acta Part A 1967, 23, 45–54.
[72] M. Sorai, S. Seki, J. Phys. Chem. Solids 1974, 35, 555–570.
[73] A. Bousseksou, J. J. McGarvey, F. Varret, J. A. Real, J.-P. Tuchagues, A. C.
Dennis, M. L. Boillot, Chem. Phys. Lett. 2000, 318, 409–416.
[66] L. H. Gade, Koordinationschemie, Wiley-VCH, Weinheim, Germany, 1st ed.,
1998.
[67] V. Ksenofontov, G. Levchenko, H. Spiering, P. Gütlich, J.-F. Letard, Y. Bou-
hedja, O. Kahn, Chem. Phys. Lett. 1998, 294, 545–553.
[74] S. Trofimenko, J. Am. Chem. Soc. 1967, 89, 3170–3177.
[68] P. Guionneau, J.-F. Letard, D. S. Yufit, D. Chasseau, G. Bravic, A. E. Goeta,
J. A. K. Howard, O. Kahn, J. Mater. Chem. 1999, 9, 985–994.
[69] W. A. Baker Jr., G. J. Long, Chem. Commun. 1965, 368–369.
Received: October 30, 2015
Published Online: December 23, 2015
Eur. J. Inorg. Chem. 2016, 2175–2186
www.eurjic.org
2186
© 2016 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim