Solid-state ligand-driven light-induced spin change at ambient temperatures in bis(dipyrazolylstyrylpyridine)iron(II) complexes

Article abstract of DOI:10.1021/ic300030b

We previously reported that an Fe^II complex ligated by two (Z)-2,6-di(1H-pyrazol-1-yl)-4-styrylpyridine ligands (Z-H) presented a solid state ligand-driven light-induced spin change (LD-LISC) upon one-way Z-to-E photoisomerization, although modulation of the magnetism was trivial at ambient temperatures (Chem. Commun.2011, 47, 6846). Here, we report the synthesis of new derivatives of Z-H, Z-CN and Z-NO₂, in which electron-withdrawing cyano and nitro substituents are introduced at the 4-position of the styryl group to attain a more profound photomagnetism at ambient temperatures. Z-CN and Z-NO₂ undergo quantitative one-way Z-to-E photochromism upon excitation of the charge transfer band both in acetonitrile and in the solid state, similar to the behavior observed for Z-H. In solution, these substituents stabilized the low-spin (LS) states of Z-CN and Z-NO₂, and the behavior was quantitatively analyzed according to the Evans equation. The photomagnetic properties in the solid state, on the other hand, cannot be explained in terms of the substituent effect alone. Z-CN displayed photomagnetic properties almost identical to those of Z-H. Z-CN preferred the high-spin (HS) state at all temperatures tested, whereas photoirradiated Z-CN yielded a lower π_MT at ambient temperatures. The behavior of Z-NO₂ was counterintuitive, and the material displayed surprising photomagnetic properties in the solid state. Z-NO₂ occupied the LS state at low temperatures and underwent thermal spin crossover (SCO) with a T_1/2 of about 270 K. The photoirradiated Z-NO₂ displayed a higher value of π_MT and the modulation of π_MT exceeded that of Z-H or Z-CN. Z-NO₂?acetone, in which acetone molecules were incorporated into the crystal lattice, further stabilized the LS state (T _1/2 > 300 K), thereby promoting large modulations of the π_MT values (87% at 273 K and 64% at 300 K) upon Z-to-E photoisomerization. Single crystal X-ray structure analysis revealed that structural factors played a vital role in the photomagnetic properties in the solid state. Z-H and Z-CN favored intermolecular π-π stacking among the ligand molecules. The coordination sphere around the Fe^II nucleus was distorted, which stabilized the HS state. In contrast, Z-NO₂? acetone was liberated from such intermolecular π-π stacking and coordination distortion, resulting in the stabilization of the LS state.

Full text of DOI:10.1021/ic300030b

Article
pubs.acs.org/IC
Solid-State Ligand-Driven Light-Induced Spin Change at Ambient
Temperatures in Bis(dipyrazolylstyrylpyridine)iron(II) Complexes
Kazuhiro Takahashi,† Yuta Hasegawa,† Ryota Sakamoto,† Michihiro Nishikawa,† Shoko Kume,†
Eiji Nishibori,‡ and Hiroshi Nishihara*,†
†Department of Chemistry, Graduate School of Science, The University of Tokyo, 7-3-1, Hongo, Bunkyo-ku, Tokyo, 113-0033 Japan
‡Department of Applied Physics, Nagoya University, Nagoya 464-8603, Japan
S
* Supporting Information
ABSTRACT: We previously reported that an FeII complex ligated by two (Z)-2,6-
di(1H-pyrazol-1-yl)-4-styrylpyridine ligands (Z-H) presented a solid state ligand-
driven light-induced spin change (LD-LISC) upon one-way Z-to-E photoisomeriza-
tion, although modulation of the magnetism was trivial at ambient temperatures
(Chem. Commun. 2011, 47, 6846). Here, we report the synthesis of new derivatives of
Z-H, Z-CN and Z-NO2, in which electron-withdrawing cyano and nitro substituents
are introduced at the 4-position of the styryl group to attain a more profound
photomagnetism at ambient temperatures. Z-CN and Z-NO2 undergo quantitative
one-way Z-to-E photochromism upon excitation of the charge transfer band both in
acetonitrile and in the solid state, similar to the behavior observed for Z-H. In
solution, these substituents stabilized the low-spin (LS) states of Z-CN and Z-NO2,
and the behavior was quantitatively analyzed according to the Evans equation. The
photomagnetic properties in the solid state, on the other hand, cannot be explained in
terms of the substituent effect alone. Z-CN displayed photomagnetic properties
almost identical to those of Z-H. Z-CN preferred the high-spin (HS) state at all temperatures tested, whereas photoirradiated Z-
CN yielded a lower χMT at ambient temperatures. The behavior of Z-NO2 was counterintuitive, and the material displayed
surprising photomagnetic properties in the solid state. Z-NO2 occupied the LS state at low temperatures and underwent thermal
spin crossover (SCO) with a T1/2 of about 270 K. The photoirradiated Z-NO2 displayed a higher value of χMT and the
modulation of χMT exceeded that of Z-H or Z-CN. Z-NO2·acetone, in which acetone molecules were incorporated into the
crystal lattice, further stabilized the LS state (T1/2 > 300 K), thereby promoting large modulations of the χMT values (87% at 273
K and 64% at 300 K) upon Z-to-E photoisomerization. Single crystal X-ray structure analysis revealed that structural factors
played a vital role in the photomagnetic properties in the solid state. Z-H and Z-CN favored intermolecular π−π stacking among
the ligand molecules. The coordination sphere around the FeII nucleus was distorted, which stabilized the HS state. In contrast,
Z-NO2·acetone was liberated from such intermolecular π−π stacking and coordination distortion, resulting in the stabilization of
the LS state.
transfer (MLCT) transitions, followed by trapping of the
metastable HS (S = 2) state.
INTRODUCTION
■
Spin crossover (SCO)1−4 is a phenomenon in which the spin
state of a metal center switches between the low-spin (LS) and
high-spin (HS) states under an external stimulus, such as
temperature, pressure, or light. Octahedral first row transition
metal complexes with electronic configurations of d4−d7 exhibit
SCO. Light-induced excited spin state trapping (LIESST),1−4
in which light-induced electronic transitions involve a change
from the stable LS state to the metastable HS state below the
thermal spin crossover temperature, has been extensively
investigated. Light-controllable systems present tremendous
advantages over systems controlled via other stimuli, including
practical reversibility, controllability over the power supply, and
selectivity with respect to the irradiation wavelength. The
LIESST effects were first reported by Decurtins and co-workers
in 1984 in the context of [FeII(ptz)6](BF4)2 (ptz = 1-
propyltetrazole).5 LIESST is based on the photoexcitation of
the stable LS (S = 0) state via d−d and metal-to-ligand charge
The thermal stability of the metastable HS state, which is
often represented as a convenient observable parameter
T(LIESST),4,6−8 is central to the practical application of
LIESST materials. LIESST phenomena have been primarily
observed only under cryogenic conditions, and much effort has
been devoted to increasing T(LIESST), for example, by using
multidentate ligands4,6−8 or intermolecular cooperative effects,
such as π−π interactions9−12 and hydrogen bonding.13−16
Despite significant efforts, the highest value of T(LIESST) yet
achieved, 132 K,17−19 is substantially below ambient temper-
atures.
A new strategy for achieving photomagnetic switching, that
is, ligand-driven light-induced spin change (LD-LISC), was first
Received: January 5, 2012
Published: April 11, 2012
© 2012 American Chemical Society
5188
dx.doi.org/10.1021/ic300030b | Inorg. Chem. 2012, 51, 5188−5198

Inorganic Chemistry
Article
proposed by Zarembowitch et al.20 The prerequisite for LD-
LISC is that ligand molecules contain organic photochromics,
the photoisomerization of which affects the spin equilibrium
between the LS and HS states. The advantage of LD-LISC over
LIESST is that the former is potentially available even at
ambient temperatures: The lifetimes of metastable states at
ambient temperatures are far longer in the former than in the
latter.
In the solid state, intermolecular interactions play a vital role. Z-
CN, which features intermolecular π−π stacking just as Z-H,
exhibits a little decrease in χMT in the solid state upon Z-to-E
photoisomerization at ambient temperatures. In contrast, Z-
NO2·acetone, which does not feature intermolecular inter-
actions in the crystalline state, presents a counterintuitive LS→
HS transition with large variations in the χMT values, even at
ambient temperatures (acetone: a crystal solvent).
The LD-LISC systems reported thus far have been
demonstrated only in solution or polymer media21−23 because
the photoisomerization of organic photochromics was consid-
ered to require a certain degree of free volume. Recently, solid-
state LD-LISC has been pursued with mixed results. Boillot and
co-workers synthesized an FeII complex containing a pyridine-
incorporated dithienylethene ligand, only to find no photo-
magnetic effects upon photoisomerization of the dithienyle-
thene moiety.24 Boillot and co-workers also investigated
FeII(stpy)4(NCSe)2 (stpy = 4-styrylpyridine).25 In this system,
the authors identified a one-way Z-to-E photochromism in the
solid state. Photoisomerization resulted in LD-LISC, although it
did not affect χMT beyond 250 K; both the E- and Z-isomers
were present solely in the HS state (S = 2).
RESULTS AND DISCUSSION
■
Syntheses. The synthetic details are described in the
Experimental Section. The essence of these procedures is
briefly introduced here. The stilbene-conjugated ligands were
synthesized by Wittig reactions between 2,6-di(1H-pyrazol-1-
yl)isonicotinaldehyde and the corresponding phosphonium
bromides such that the ligands were obtained as mixtures of the
E- and Z-isomers. These isomers were successfully isolated by
alumina column chromatography, and the E-Z configuration
was determined based on the J-coupling value between the two
1
protons on the ethylene moiety in H NMR spectra. These
isomers of the ligands were used to prepare, separately, the
corresponding complexes with the E-and Z-configurations. The
complexes were identified using ESI-TOF mass and FT-IR
spectroscopies, and elemental analysis. The complexes with E-
and Z-configuration were characterized by single-crystal X-ray
diffraction (XRD) analysis. A detailed discussion of the crystal
structures of E- and Z-forms is described in a later section.
1. Photoisomerization Behavior. UV−vis Spectra of the Z-
Isomers. Figure 1 shows the UV−vis spectra of Z-H, Z-CN, and
We independently reported another stilbenoid FeII complex,
Z/E-H, containing two 2,6-di(1H-pyrazol-1-yl)-4-styrylpyri-
dine) ligands (Scheme 1).26 The Z-form underwent a visible
Scheme 1. Photoisomerization from the Z-Form to the E-
Form upon Visible Light (λ > 420 nm) Irradiation in FeII
a
Complexes Studied
a
Counter anions are omitted.
light one-way Z-to-E photochromism to the E-form in the solid
state. Tridentate ligand 2,6-di(1H-pyrazol-1-yl)-4-pyridine
induced a ligand field suitable for thermal SCO at ambient
temperatures.27 Z-H occupied the HS state below 300 K,
whereas E-H was in equilibrium between the HS and LS states
over this temperature range. The χMT values varied from 3.6
cm3 K mol−1 to 3.1 cm3 K mol−1 upon Z-to-E photo-
isomerization at 300 K. Z/E-H, therefore, is the first system to
display solid state LD-LISC at ambient temperatures, although
the photomagnetic changes were not large.
This work has primarily served to improve the amplitude of
photomagnetic modulation in an LD-LISC system. To this end,
the influences of substituents on the stilbenoid ligand were
employed to tune the ligand field properties. Electron-
withdrawing substituents, nitro and cyano groups, were
introduced to reduce the energy of the π* orbital in the 2,6-
di(1H-pyrazol-1-yl)-4-pyridine ligand, which promoted π-back-
donation, a stronger ligand field, and more stabilization of the
LS state.28,29 We also expected that the substituents on the
stilbenoid ligand might tune the intermolecular interactions in
the solid state to significantly influence the SCO behavior.30 In
solution, the electronic substituent effects solely improve the
stability of the LS state, as predicted by the Hammett equation.
Figure 1. UV−vis spectra of Z-H, Z-CN, and Z-NO2 in acetonitrile at
295 K. The inset shows an expanded view of the MLCT absorption
region.
Z-NO2 in acetonitrile. The strong absorption band at 320 nm
was assigned to a mixture of π−π* and n−π* transitions of the
2,6-di(1H-pyrazol-1-yl)-4-styrylpyridine ligands,31 whereas the
weak and broad absorption band at 450 nm was attributed to
metal-to-ligand charge transfer (MLCT) band. This MLCT
band was slightly red-shifted in Z-CN and Z-NO2 compared to
Z-H, suggesting that the introduction of electron-withdrawing
nitro and cyano groups lowered the ligand π* orbital energy
level. Such effects were expected to increase the ligand field
splitting and stabilize the LS state.
Photoisomerization in Solution. In the previous communi-
cation, we reported that Z-H exhibited complete one-way Z-to-
E photoisomerization in acetone at room temperature upon
irradiation with 436 nm light, which corresponded to excitation
of the MLCT band.26 Z-H displayed a similar photoresponse in
5189
dx.doi.org/10.1021/ic300030b | Inorg. Chem. 2012, 51, 5188−5198

Inorganic Chemistry
Article
Figure 2. UV−vis spectral changes of (a) Z-H, (b) Z-CN, and (c) Z-NO2 in acetonitrile upon irradiation with 436 nm light. The spectra of the
corresponding E-form complexes are indicated by black circles, and the concentrations of the complexes were 1.0 × 10−4 M.
acetonitrile at room temperature (Figure 2a). Upon irradiation
with 436 nm light, the absorption intensity of the MLCT band
increased and its absorption maximum shifted toward the red.
The resultant spectrum matched that of the synthesized E-H.
An acetonitrile solution of E-H did not reveal spectral changes
upon excitation of the MLCT band. These results indicate that
one-way Z-to-E photoisomerization proceeded quantitatively in
acetonitrile. Similarly, photoirradiation of Z-CN and Z-NO2
with 436 nm light resulted in one-way Z-to-E photoisomeriza-
tion (Figure 2b, 2c). The molar extinction coefficient of the
MLCT band of the LS FeII complexes far exceeded that of HS
complexes.32 Therefore, the spectral changes that accompanied
the Z-to-E photoisomerization in our complexes suggested that
the fraction of the LS species increased in the E-forms. Note
that this system underwent spectral changes that were unlike
those observed for FeII(stpy)4(NCSe)2. In that system, the
intensity of the MLCT band was reduced upon Z-to-E
photoisomerization, and the authors determined that both
isomers occupied the HS state at room temperature.25 Our
system may be better suited for LD-LISC at ambient
temperatures in the solid state.
spectrum of the photoirradiated Z-NO2 also agreed with that of
E-NO2, and the characteristic Z-NO2 peaks disappeared
(Supporting Information, Figure S1). In Z-CN, a peak derived
from the C−H out-of-plane bending of the alkenyl moieties33,34
at 974 cm−1 was shifted to 971 cm−1, as in other stilbene-
attached derivatives,35,36 including Z-H,26 upon Z-to-E photo-
isomerization. These results suggest that the one-way Z-to-E
isomerization by visible light irradiation was active in the solid
medium, and introduction of cyano and nitro substituents did
not alter the photoresponse.
Photoisomerization in the Crystalline State. Bulk micro-
crystalline samples of Z-H, Z-CN, Z-NO2, and Z-NO2·acetone
were irradiated with visible light (λ > 420 nm) for a few days
until they presented significant color changes (Figure 4). For
Photoisomerization in a KBr Pelletized Sample. The one-
way Z-to-E photoisomerization was investigated using FT-IR
spectroscopy (KBr pellet). KBr pelletized samples of each
complex were irradiated with visible light (λ > 420 nm) to
excite the MLCT absorption band, as in solution. Upon
irradiation, the color changed from yellow to dark orange or red
in all complexes. The photoirradiated samples were the same
color as the synthesized E-forms (Figure 3). The FT-IR
Figure 4. Color changes of the microcrystalline Z-H, Z-CN, Z-NO2,
and Z-NO2·acetone upon irradiation with visible light (λ > 420 nm)
for 115 h in Z-H, 94 h in Z-CN, and 72 h in Z-NO2 and Z-
NO2·acetone.
example, Z-CN turned from yellow to dark orange. The E/Z
ratios in the photostationary state (PSS) were estimated based
on the FT-IR spectra (KBr pelletized state) and the UV−vis
spectra of the microcrystals in the PSS, dissolved in acetonitrile.
Both spectroscopic studies revealed a conversion ratio from the
Z-form to the E-form of no less than 88% (Figure 5). Further
irradiation did not influence the E/Z ratios, presumably because
of the inner filter effects of the crystalline solid.37 Powder XRD
analysis indicated that the photoproducts, which were
dominated by the E-forms, as shown above, lost their
crystallinity (Supporting Information, Figure S2).
2. LD-LISC Behavior. Thermal SCO in Acetone Solutions. The
magnetization of the Z-form in solution was estimated using
the Evans method in acetone-d6.38 The measurements were
performed over the temperature range 220−290 K. All
complexes showed thermal SCO over this temperature range,
and the temperature dependence of the resultant χMT values
was fit using the van’t Hoff equation to calculate the
thermodynamic parameters of thermal SCO (see Experimental
Figure 3. Color changes of the KBr pelletized samples of Z-H, Z-CN,
and Z-NO2 upon irradiation with visible light (λ > 420 nm) for 5 h in
Z-H, 1 h in Z-CN, and 6 h in Z-NO2.
5190
dx.doi.org/10.1021/ic300030b | Inorg. Chem. 2012, 51, 5188−5198

Inorganic Chemistry
Article
Figure 5. FT-IR spectra of photoproducts of (a) Z-H, (c) Z-CN, and (e) Z-NO2 in the microcrystalline solid by irradiation with visible light (broken
lines), together with those of pure Z-(solid lines) and E-isomers (dotted lines). UV−vis spectra of photoproducts of (b) Z-H, (d) Z-CN, and (f) Z-
NO2 in acetonitrile (broken lines), together with those of pure Z-isomers (solid lines). The concentration of each complex was 2.0 × 10−5 M. The
simulated UV−vis spectra (represented as circles) indicated by the circles corresponded to the mixture of (b) Z-H and E-H in a ratio of 12:88, (d)
Z-CN and E-CN in a ratio of 8:92, and (f) Z-NO2 and E-NO2 in a ratio of 12:88.
Section for the details). Resultant van’t Hoff plots for three
complexes are shown in Figure 6 with the regression lines,
Figure 7. χMT vs T plots of Z-H (filled circle), Z-CN (open circle),
and Z-NO2 (triangle) in acetone-d6, determined using the Evans
method. van’t Hoff models using thermodynamic parameters in Table
1 are indicated by the solid lines (a black line for Z-H, a green line for
Z-CN, and a red line for Z-NO2).
Figure 6. van’t Hoff plots for Z-H, Z-CN, and Z-NO2 in acetone-d6.
The regression lines are indicated by the solid lines.
of the electron-withdrawing substituents. This conclusion was
quantitatively supported by the T1/2-Hammett constant σp plots
which afford the thermodynamic parameters listed in Table 1
(Figure 8).39 We note that the χMT values for the E-forms were
and the crossover temperatures T1/2 of 245 K, 259 K, 261 K for
not acquired using the Evans method because of the low
Z-H, Z-CN, and Z-NO2, respectively. The van’t Hoff model
solubility of these forms in solvents.
reproduced well the experimental results (Figure 7). The
LD-LISC Behavior in the Solid State. Variable-temperature
significant increase in T1/2 in Z-CN and Z-NO2 compared to Z-
magnetic susceptibility measurements for Z-H, Z-CN, Z-NO2,
H suggested stabilization of the LS state by the electronic effect
and Z-NO·acetone in the crystalline or solid state were carried
out using SQUID magnetometers, and the resultant χMT−T
Table 1. SCO Temperatures and Thermodynamic
Parameters for Z-H, Z-CN, and Z-NO2, Determined by the
Evans Method
plots are shown in Figure 9. The χMT values at ambient
temperatures, 273 and 300 K, were collected in Table 2
(represented as χMT(i)).
Z-CN exhibited a χMT of 3.3 cm3 K mol−1 at 300 K, which
was a typical value for HS FeII species.2 χMT value did not
change significantly upon cooling to 30 K. The abrupt decrease
of χMT that emerged below 30 K is attributed to zero-field
splitting4 and/or intermolecular antiferromagnetic interaction.
T1/2/K
ΔH/kJ mol−1
ΔS/J mol−1 K−1
Z-H
245
259
261
19.4
20.1
21.1
79
78
81
Z-CN
Z-NO2
5191
dx.doi.org/10.1021/ic300030b | Inorg. Chem. 2012, 51, 5188−5198

Inorganic Chemistry
Article
kind of a phase transition occurred. DSC measurement
supported the occurrence of a phase transition at this
temperature range (Supporting Information, Figure S3). We
could not clarify the detail of this phase transition at this stage,
because we were not able to obtain the single crystal of
unsolvated Z-NO2. Z-NO2·acetone exhibited a χMT of 1.1 cm3
K mol−1 at 300 K and displayed thermal SCO around room
temperatures to reach a χMT of 0.10 cm3 K mol−1 at 170 K,
taking the LS state below this temperature. The transition
temperature T1/2 of Z-NO2·acetone (>300 K) was higher than
that of Z-NO2 (ca. 270 K): Therefore, the LS state was further
stabilized in Z-NO2·acetone.
Magnetic susceptibility measurements were performed on
the photoirradiated microcrystalline samples of Z-H, Z-CN, Z-
NO2, and Z-NO2·acetone. χMT−T plots for the photoproducts
are shown in Figure 10. All photoirradiated Z-form samples
displayed incomplete thermal spin crossover. In Z-CN, a
decrease in magnetization was observed upon Z-to-E photo-
isomerization over the entire temperature range: The photo-
irradiated sample exhibited a χMT value of 2.8 cm3 K mol−1 at
300 K, and showed gradual thermal SCO, reaching a χMT value
of 1.8 cm3 K mol−1 at 100 K. These variations in the magnetic
properties caused by visible light irradiation were similar to
those of Z-H. The ratio of the χMT values varied no more than
20% under ambient temperatures (Table 2). In contrast, the
χMT values of Z-NO2 increased after visible light irradiation.
The irradiated sample yielded a thermal SCO and a constant
χMT value of 2.0 cm3 K mol−1 below 100 K. Variations in the
ratio of χMT upon Z-to-E photoisomerization were as large as
those observed for Z-H and Z-CN at 300 K, although the
modulation was more prominent at 273 K (42%). Microcrystals
of Z-NO2·acetone presented a more profound photomagnetic
effect. As with Z-NO2, Z-to-E photoisomerization in Z-
NO2·acetone increased the values of χMT. The photoirradiated
sample then underwent thermal SCO with a constant χMT of
2.0 cm3 K mol−1 below 100 K. The large variations in the ratio
of χMT values were noteworthy: the variations reached 64% at
300 K and 87% at 273 K (Table 2). Thus, Z-NO2·acetone
displayed the largest variation in χMT among the four samples
at ambient temperatures.
Figure 8. T1/2 vs σ Hammet constant plots for the FeII SCO
complexes.
Figure 9. Temperature dependence of χMT for Z-H, Z-CN, Z-NO2,
and Z-NO2·acetone in the microcrystalline state.
Thus, Z-CN occupied the HS state for the overall temperature
range, just as Z-H with 3.6 cm3 K mol−1 from 300 to 30 K.
Therefore, the introduction of cyano groups did not influence
the magnetic property remarkably in the microcrystalline state
of the Z-form. On the other hand, Z-NO2 showed a smaller
χMT of 2.8 cm3 K mol−1 at 300 K and exhibited gradual thermal
spin crossover, reaching a χMT of 0.20 cm3 K mol−1 at 150 K.
This result indicates that the introduction of nitro groups
stabilized the LS state. In solution, the LS stabilization was
observed in both Z-CN and Z-NO2, whereas only Z-NO2
displayed the stabilization of the LS state in the solid state. The
Hammett constant σp is 0.66 for a cyano group and 0.78 for a
nitro group,39 taking similar values with one another. This fact
indicates that the substituent effect alone cannot sufficiently
account for the contrastive magnetic behavior of Z-CN and Z-
NO2 in the solid state. In Z-NO2, a tiny thermal hysteresis
around 180 K was found in the χMT vs T plot, suggesting that a
Single Crystal X-ray Crystallography. The crystal structures
and crystallographic data for Z-H at 90 K, Z-CN and Z-
NO2·acetone at 113 K are shown in Figures 11a−c and Table
3. Fe−N bond distances and bite angles around the
coordination sphere are good indicators of the spin state of
FeII complexes.2,40,41 Dipyrazolylpyridine ligand complexes
yield an average Fe−N bond distance of 1.93−1.97 Å in LS
complexes and 2.14−2.20 Å in HS complexes.26,27,42−45 Also,
the Σ parameter (deg), which is the summation of the
Table 2. χMT of the FeII Complexes before and after Photoirradiation at 300 and 273 K
300 K
273 K
a
b
c
a
b
c
χMT(i)
χMT(f)
ΔχMT
variation
d
χMT(i)
χMT(f)
ΔχMT
variation
d
(cm3 K mol−1
)
(cm3 K mol−1
)
(cm3 K mol−1
)
ratio (%)
(cm3 K mol−1
)
(cm3 K mol−1
)
(cm3 K mol−1
)
ratio (%)
Z-H
3.6
3.3
2.8
1.1
3.1
2.8
3.2
3.1
0.5
0.5
0.4
2.0
14
15
12
64
3.6
3.3
1.8
0.4
3.0
2.7
3.1
3.0
0.6
0.6
1.3
2.6
17
18
42
87
Z-CN
Z-NO2
Z-NO2
·acetone
a
b
c
d
χMT value before photoirradiation. χMT value at PSS. ΔχMT = |χMT(f) − χMT(i)|. ΔχMT/χMT(i) for Z-H and Z-CN, ΔχMT/χMT(f) for Z-NO2
and Z-NO2·acetone.
5192
dx.doi.org/10.1021/ic300030b | Inorg. Chem. 2012, 51, 5188−5198

Inorganic Chemistry
Article
Figure 10. Temperature dependence of χMT for (a) Z-H, (b) Z-CN, (c) Z-NO2, and (d) Z-NO2·acetone before (solid lines) and after (broken
lines) visible light (λ > 420 nm) irradiation for 115 h in Z-H, 94 h in Z-CN, and 72 h in Z-NO2 and Z-NO2·acetone.
Figure 11. ORTEP diagrams showing 50% probability for (a) Z-H, (b) Z-CN, (c) Z-NO2·acetone, (d) E-CN·MeNO2, and (e) E-NO2·H2O.
Diffraction data were collected at 113 K except for Z-H (90 K). Counter anions and hydrogen atoms are omitted for clarity.
deviations from 90° of the twelve bite angles around the central
FeII ion,46,47 can range from 85° to 96° in LS complexes and
from 145° to 197° in HS complexes. In this study, the average
Fe−N bond distance and Σ parameter were 2.170 Å and 172.4°
for Z-H at 90 K, 2.171 Å and 183.1° for Z-CN, and 1.946 Å and
88.7° for Z-NO2·acetone at 113 K, respectively. These results
suggested that Z-H and Z-CN assumed a HS state and Z-
NO2·acetone assumed a LS state at those temperatures. These
results were consistent with the magnetic susceptibility
measurements (Figure 9). The E-forms were also subjected
to single-crystal XRD analysis. The crystal structure of E-
H·2PC had been reported in the previous communication,26
whereas those of E-CN·MeNO2 and E-NO2·H2O were newly
disclosed. (Figures 11d and 11e and Table 3).
The crystal packing structure notably revealed π−π
interactions between the pyrazole rings of Z-H and between
the 4-cyanophenyl groups of Z-CN (Figures 12a and 12b). The
π−π interactions were expected to distort the coordination
geometry away from ideal octahedral coordination at the FeII
center (i.e., a large Σ parameter), which favored the HS state.
Distortions in octahedral coordination spheres at FeII centers
usually favor the HS state.2,47 On the other hand, such π−π
interactions were absent from the Z-NO2·acetone structure
(Figure 12c), which appeared to assume an ideal octahedral
coordination sphere that stabilized the LS state. Single crystals
of solvent-free Z-NO2 have not yet been obtained, and the
contribution of the acetone molecules to the Z-NO2·acetone
structure remains unknown. Because T1/2 was higher in Z-
NO2·acetone than in the solvent-free Z-NO2, the solvent
5193
dx.doi.org/10.1021/ic300030b | Inorg. Chem. 2012, 51, 5188−5198

Inorganic Chemistry
Article
Table 3. Crystal Data for Z-H, Z-CN, Z-NO2·acetone, E-CN·MeNO2, and E-NO2·H2O
Z-H
Z-CN
Z-NO2·acetone
empirical formula
Fw
C38H30N12B2F8Fe
856.17
C20H14N6BF4Fe0.5
453.11
C41H34N12O5B2F8Fe
1004.27
crystal dimension
crystal system
lattice parameters
0.31 × 0.25 × 0.02 mm
monoclinic
a = 39.672(18) Å
b = 8.106(3) Å
c = 36.949(18) Å
β = 140.303(7)°
V = 7589(6) Å3
C2/c
0.16 × 0.10 × 0.03 mm
monoclinic
a = 14.439(3) Å
b = 8.2491(17) Å
c = 32.940(7) Å
β = 99.954(3)°
V = 3864.6(14) Å3
C2/c
0.25 × 0.11 × 0.06 mm
monoclinic
a = 7.9954(13) Å
b = 36.339(6) Å
c = 15.344(3) Å
β = 98.374(2)°
V = 4410.5(13) Å3
P21/n
space group
Z value
λ
8
8
4
0.7107 Å
0.7107 Å
0.7107 Å
T
90 K
113 K
113 K
μ(Mo Kα)
D(calc)
R1
0.481 mm−1
1.498 g/cm3
0.0927
0.479 mm−1
1.558 g/cm3
0.0596
0.436 mm−1
1.512 g/cm3
0.0815
WR2
0.2659
0.1383
0.2158
GOF
1.098
1.164
1.096
E-CN·MeNO2
E-NO2·H2O
empirical formula
C41H31N13O2B2F8Fe
967.26
C38H30N12O5B2F8Fe
Fw
962.19
crystal dimension
crystal system
lattice parameters
0.17 × 0.16 × 0.06 mm
triclinic
0.10 × 0.10 × 0.07 mm
triclinic
a = 11.9220(17) Å
b = 12.8890(19) Å
c = 14.539(2) Å
α = 87.253(6)°
β = 80.397(5)°
γ = 70.184(4)°
V = 2072.3(5) Å3
a = 10.832(6) Å
b = 13.326(7) Å
c = 16.010(9) Å
α = 107.925(6)°
β = 101.946(5)°
γ = 104.959(6)°
V = 2018.6(19) Å3
space group
Z value
λ
P1
P1
̅
̅
2
2
0.7107 Å
113 K
0.7107 Å
113 K
T
μ(Mo Kα)
D(calc)
R1
0.456 mm−1
1.550 g/cm3
0.0654
0.472 mm−1
1.583 g/cm3
0.0608
WR2
0.1839
0.1646
GOF
1.048
1.094
molecules may further stabilize the LS state, for example, by
filling up the voids in the crystal lattice to support the Z-
NO2·acetone crystal structure without the need for π−π
interactions. Intermolecular interactions and guest solvent
inclusion are expected to dominate the magnetic properties
of complexes rather than the electronic substituent effects. We
note that crystal packing and crystal solvents significantly affect
the magnetic properties of [FeII(2,6-di(1H-pyrazol-1-yl)-4-
pyridine)2] complexes.44,45,47
Because all photoproducts were amorphous, quantitative
analysis of the bond lengths and intermolecular interactions of
the photogenerated E-forms is impossible. Hereafter, limited
discussion is described. The χMT−T plots of the photo-
irradiated Z-NO2 and Z-NO2·acetone resemble one another
(Figure 10). This experimental fact indicates that the two
photogenerated materials are in the same status: In fact, the
FT-IR spectrum of photoirradiated Z-NO2·acetone does not
contain any peaks for acetone molecules (Supporting
Information, Figure S4). Therefore, the acetone molecules as
the crystal solvent are released in the course of Z-to-E
photoisomerization. Greater magnetization in Z-NO2 and Z-
NO2·acetone upon Z-to-E photoisomerization would result
from the emergence of new intermolecular interactions in the
amorphous solid to stabilize the HS state. In other words,
changes in intermolecular interactions, often expressed as
chemical pressure and anisotropy,1,2 would accompany the Z-
to-E photoisomerization, to influence the spin state. This
contribution would work more dominantly than the
intensification of the ligand field induced by the intramolecular
Z-to-E photoisomerization, which stabilizes the LS state. The
steep decline of the χMT values in Z-H and Z-CN may be due
to the cancellation of strong intermolecular π−π interactions,
although strengthening of the ligand field upon Z-to-E
photoisomerization could not be excluded.
Variable-temperature magnetic susceptibility measurements
were also carried out for the synthesized E-CN and E-NO2.
Microcrystalline samples of the two complexes displayed
gradual and incomplete thermal spin crossover to reach a
χMT value of 1.0 cm3 K mol−1 below 90 K (Supporting
Information, Figure S5). The magnetic properties of the
5194
dx.doi.org/10.1021/ic300030b | Inorg. Chem. 2012, 51, 5188−5198

Inorganic Chemistry
Article
Figure 12. Crystal packings of (a) Z-H along c axis, (b) Z-CN along a axis, and (c) Z-NO2·acetone along a axis. Counter anions (a) and hydrogen
atoms (c) are omitted for clarity.
ester were obtained from TCI Ltd. 4-Nitrobenzyl bromide and
iron(II) tetrafluoroborate hexahydrate were received from Aldrich Ltd.
n-Buthyllithium in hexane and manganese dioxide were obtained from
Kanto Ltd. Silica gel 60 (0.040−0.064 mm) for a flash silica column
chromatography was purchased from MERCH Ltd. Dichloromethane
used as a solvent for reactions was distilled from calcium hydride
under a nitrogen atmosphere. Tetrahydrofuran (THF) used as a
solvent for reactions was distilled from sodium and benzophenone
under a nitrogen atmosphere. Spectroscopy grade acetonitrile was
obtained from Kanto Chemical Ltd. Potassium bromide (Wako Ltd.)
for FT-IR spectroscopy was stored in a drybox and used as it was.
Common solvents and anhydrous solvents used for syntheses and
measurements were obtained from Kanto Chemical Ltd.
Synthesis of 4-Cyanobenzyltris(2-(methoxymethoxy)-
phenyl)phosphonium Bromide. Tris(2-(methoxymethoxy)-
phenyl)phosphine (1.00 g, 2.26 mmol), 4-cyanobenzyl bromide (576
mg, 2.94 mmol), and sodium hydrogen carbonate (134 mg, 1.60
mmol) were dissolved in 10 mL of acetonitrile and heated to 80 °C for
10 min. After evaporation of the solvent, the residue was extracted
with dichloromethane (50 mL) and filtered. The filtrate was
concentrated, and ethyl acetate (30 mL) was added to give rise to
pale white precipitate, which was filtered and washed by ethyl acetate.
The titled compound was obtained as a white powder. Yield: 1.34 g
(2.10 mmol, 93%). 1H NMR (acetone-d6, 295 K): δ 7.88 (dd, J = 15.1,
7.9 Hz, 3H, Ph), 7.83−7.79 (m, 3H, Ph), 7.58 (d, J = 8.0 Hz, 2H, Bz)
7.50 (dd, J = 8.3, 2.0 Hz, 2H, Bz), 7.39−7.32 (m, 6H, Ph), 5.30 (d, J =
16.1 Hz, 2H, CH2), 5.21 (s, 6H, CH2), 3.10 (s, 9H, CH3). Analytical
data. Found: C, 60.00; H, 5.38; N, 2.13%. Calcd for C32H33O6NBrP:
C, 60.20; H, 5.21; N, 2.19%.
synthesized E-form complexes were not completely consistent
with those of the photogenerated E-forms, which might stem
from differences in intermolecular interactions, presence or
absence of crystal solvents, and crystal lattice effects.48
CONCLUSIONS
■
We synthesized analogues of an FeII complex, [FeII(2,6-di(1H-
pyrazol-1-yl)-4-styrylpyridine)2](BF4)2, which displayed LD-
LISC behavior at ambient temperatures. Cyano and nitro
groups were introduced at the tips of the stilbene moieties to
enhance modulation of the photomagnetism. These substitu-
ents did not affect the one-way Z-to-E photoisomerization in
solution or microcrystalline solid states. In the solution phase,
cyano and nitro groups stabilized the LS state in proportion to
the groups’ electron-withdrawing strength. In contrast, the
substituents affected the photomagnetic properties in the
microcrystalline and solid states in such a way that was not
fully attributable to the electronic effects. The single crystal X-
ray structures of the Z-forms were associated with the
magnetism of the structures. Z-H and Z-CN, in the HS state
at any temperature, included intermolecular π−π interactions
that distorted the octahedral coordination environment at the
FeII center, thereby stabilizing the HS state. On the other hand,
Z-NO2·acetone preferred the LS state. In this complex, π−π
interactions were absent, and distortion of the FeII coordination
sphere was negligible. The LS state of this structure was thereby
stabilized. Z-NO2 partially destabilized the LS state compared
to Z-NO2·acetone. This result indicated that the solvent
molecules also stabilized the LS state. Z-to-E photoisomeriza-
tion suppressed magnetization in the Z-H and Z-CN structures
and increased χMT in Z-NO2 and Z-NO2·acetone.
NO2·acetone displayed the most profound LD-LISC effects
at ambient temperatures.
Synthesis of (Z)-2,6-di(1H-pyrazol-1-yl)-4-(4-cyanostyryl)-
pyridine. Under a nitrogen atmosphere, 4-cyanobenzyltris(2-
(methoxymethoxy)phenyl)phosphonium bromide (670 mg, 1.05
mmol) was suspended in 50 mL of dry THF. This suspension was
then stirred, and n-buthyllithium (1.65 M in hexane, 0.70 mL, 1.1
mmol) was added dropwise at 0 °C. The reaction mixture was warmed
to room temperature, and was turned from white suspension to orange
suspension after 1 h, and then, 2,6-di(1H-pyrazol-1-yl)-
isonicotinaldehyde (250 mg, 1.05 mmol) in 30 mL of dry THF was
added dropwise at −78 °C. The reaction mixture was kept stirred at
−78 °C for 2 h. After warming to room temperature, THF was
removed under vacuum. The residue was suspended in 20 mL of
chloroform, and filtered through Celite to remove lithium bromide.
The filtrate, which contained Z- and E-isomers, was chromatographed
on alumina with chloroform−hexane (1:1). The titled compound was
EXPERIMENTAL SECTION
■
Materials. Solvents and reagents were used as received unless
otherwise noted. (2,6-di(1H-pyrazol-1-yl)pyridin-4-yl)-
isonicotinaldehyde49,50 and tris(2-(methoxymethoxy)phenyl)-
phosphine51,52 were synthesized according to the literature procedures.
The syntheses of Z-H and E-H were reported in our previous paper.26
4-Cyanobenzyl bromide and (4-cyanobenzyl)phosphonic acid diethyl
1
obtained as a white solid. Yield: 175 mg (0.52 mmol, 50%). H NMR
5195
dx.doi.org/10.1021/ic300030b | Inorg. Chem. 2012, 51, 5188−5198

Inorganic Chemistry
Article
(CDCl3, 295 K): δ 8.54 (dd, J = 2.6, 0.5 Hz, 2H, Pz), 7.71 (d, J = 1.0
Hz, 2H, Pz), 7.66 (s, 2H, Py), 7.54 (dd, J = 8.4, 1.7 Hz, 2H, Ph), 7.37
(d, J = 8.2 Hz, 2H, Ph), 6.85 (d, J = 12.3 Hz, 1H, CHCH), 6.77 (d, J
= 12.3 Hz, 1H, CHCH), 6.49 (dd, J = 2.6, 1.7 Hz, 2H, Pz). ESI-
TOF mass (positive, acetonitrile): m/z = 339.10 [M + H]+, calcd for
C20H15N6, 339.13. Analytical data. Found: C, 70.91; H, 4.41; N,
24.55%. Calcd for C20H14N6: C, 70.99; H, 4.17; N, 24.84%.
Ph), 5.34 (d, J = 16.1 Hz, 2H, CH2), 5.20 (s, 6H, CH2), 3.10 (s, 9H,
CH3). Analytical data. Found: C, 55.79; H, 5.05; N, 2.13%. Calcd for
C31H33O8NBrP·0.5H2O: C, 55.78; H, 5.13; N, 2.10%.
Synthesis of (X)-2,6-di(1H-pyrazol-1-yl)-4-(4-nitrostyryl)-
pyridine (X = Z, E). Under a nitrogen atmosphere, 4-nitrobenzyltris-
(2-(methoxymethoxy)phenyl)phosphonium bromide·0.5H2O (1.38 g,
2.06 mmol) was suspended in 90 mL of dry THF. To the former
stirred solution, n-buthyllithium (1.6 M in hexane, 1.3 mL, 2.1 mmol)
was added dropwise at 0 °C. The reaction mixture was warmed to
room temperature, and turned from a white suspension to a dark-red
suspension after 1 h, and then, 2,6-di(1H-pyrazol-1-yl)-
isonicotinaldehyde (500 mg, 2.09 mmol) in 90 mL of dry THF was
added dropwise at −78 °C. The reaction mixture was kept stirred at
−78 °C for 3 h, and at room temperature for 12 h. Afterwards, THF
was removed under vacuum. The residue was suspended in 40 mL of
chloroform, and filtered through Celite to remove lithium bromide.
The filtrate, which contained both Z- and E-isomers, was chromato-
graphed on alumina with chloroform−hexane (3:1). Z- and E- isomers
were collected separatedly. The pure products were obtained by
recrystallization from dichloromethane/hexane.
Synthesis of Z-CN. Under a nitrogen atmosphere, (Z)-2,6-di(1H-
pyrazol-1-yl)-4-(4-cyanostyryl)pyridine (80 mg, 0.24 mmol) was
dissolved in 10 mL of dry acetone. Fe(BF4)2·6H2O (39 mg, 0.12
mmol) dissolved in 2 mL of dry acetone was added to the former
stirred solution at room temperature. The reaction mixture
immediately turned from colorless solution to red solution. After
stirring for 30 min, most of the solution was evaporated under a stream
of nitrogen at room temperature. The residue was recrystallized from
acetone/diethyl ether. The precipitate was filtered and washed by
diethyl ether. After being dried under vacuum at 100 °C, Z-CN was
obtained as orange powder. Yield: 96 mg (0.11 mmol, 92%). ESI-TOF
mass (positive, acetonitrile): m/z = 366.18 [M]2+, calcd for
C40H28N12Fe, 366.10. Analytical data. Found: C, 52.85; H, 3.34; N,
18.36%. Calcd for C40H28N12B2F8Fe: C, 53.02; H, 3.11; N, 18.55%.
Synthesis of (E)-2,6-di(1H-pyrazol-1-yl)-4-(4-cyanostyryl)-
pyridine. Under a nitrogen atmosphere, (4-cyanobenzyl)phosphonic
acid diethyl ester (267 mg, 1.05 mmol) was suspended in 20 mL of dry
THF. To the former stirred solution, n-buthyllithium (1.63 M in
hexane, 0.70 mL, 1.1 mmol) was added dropwise at 0 °C. The reaction
mixture was warmed to room temperature, and turned from a white
suspension to a yellow suspension after 1 h. Then, 2,6-di(1H-pyrazol-
1-yl)isonicotinaldehyde (252 mg, 1.05 mmol) in 20 mL of dry THF
was added dropwise at room temperature, and the reaction mixture
turned to a green solution. After stirring for 2 h, THF was removed
under vacuum. The residue was suspended in 30 mL of chloroform,
and filtered through Celite to remove lithium bromide. The filtrate,
which contained both Z- and E-isomers, was chromatographed on
alumina with chloroform−hexane (1:1). The title compound was
collected, and recrystallized from dichloromethane/hexane, to afford
the title compound as a white solid. Yield: 247 mg (0.73 mmol, 69%).
1H NMR (acetone-d6, 295 K): δ 8.86 (dd, J = 2.6, 0.6 Hz, 2H, Pz),
8.10 (s, 2H, Py), 8.03 (d, J = 8.3 Hz, 2H, Ph), 7.88 (d, J = 17.6 Hz, 1H,
CHCH), 7.87 (d, J = 8.3 Hz, 2H, Ph), 7.84 (dd, J = 1.6, 0.5 Hz, 2H,
Pz), 7.70 (d, J = 16.4 Hz, 1H, CHCH), 6.59 (dd, J = 2.6, 1.6 Hz,
2H, Pz). MALDI-TOF mass: m/z = 339.12 [M + H]+, calcd for
C20H15N6, 339.13. Analytical data. Found: C, 69.19; H, 4.39; N,
24.11%. Calcd for C20H14N6·0.5H2O: C, 69.15; H, 4.35; N, 24.19%.
Synthesis of E-CN. Under a nitrogen atmosphere, (E)-2,6-di(1H-
pyrazol-1-yl)-4-(4-nitrostyryl)pyridine (100 mg, 0.29 mmol) was
dissolved in 15 mL of dry acetone. To this stirred solution was
added Fe(BF4)2·6H2O (48 mg, 0.14 mmol) dissolved in 2 mL of dry
acetone at room temperature. The reaction mixture immediately
turned from a colorless solution to a red suspension. After stirring for
30 min, the precipitate was filtered and washed by 2 mL of ice-cold
acetone and diethyl ether. After recrystallization from nitromethane/
dietyl ether and under vacuum at 100 °C, E-CN was obtained as
orange solid. Yield: 118 mg (0.13 mmol, 92%). ESI-TOF mass
(positive, acetonitrile): m/z = 366.00 [M]2+, calcd for C40H28N12Fe,
366.10. Analytical data. Found: C, 50.55; H, 3.39; N, 17.48%. Calcd
for C40H28N12B2F8Fe·2.5H2O: C, 50.51; H, 3.50; N, 17.67%.
(Z)-2,6-Di(1H-pyrazol-1-yl)-4-(4-nitrostyryl)pyridine. Obtained as
1
white solid. Yield: 181 mg (0.51 mmol, 25%). H NMR (CDCl3, 295
K): δ 8.54 (dd, J = 2.7, 0.7 Hz, 2H, Pz), 8.11 (d, J = 8.8 Hz, 2H, Ph),
7.70 (dd, J = 1.4, 0.5 Hz, 2H, Pz), 7.68 (s, 2H, Py), 7.43 (d, J = 8.5 Hz,
2H, Ph), 6.90 (d, J = 12.4 Hz, 1H, CHCH), 6.82 (d, J = 12.4 Hz,
1H, CHCH), 6.48 (dd, J = 2.4, 1.7 Hz, 2H, Pz). ESI-TOF mass
(positive, acetonitrile): m/z = 381.1047 [M + Na]+, calcd for
C19H14O2N6Na, 381.1076. Analytical data. Found: C, 63.50; H, 4.11;
N, 23.31%. Calcd for C19H14O2N6: C, 63.68; H, 3.94; N, 23.45%.
(E)-2,6-Di(1H-pyrazol-1-yl)-4-(4-nitrostyryl)pyridine. Obtained as
yellow microcrystals. Yield: 210 mg (0.59 mmol, 29%). 1H NMR
(CD2Cl2, 295 K): δ 8.60 (dd, J = 2.7, 0.7 Hz, 2H, Pz), 8.28 (d, J = 8.8
Hz, 2H, Ph), 8.01 (s, 2H, Py), 7.81 (d, J = 1.0 Hz, 2H, Pz), 7.71 (d, J =
8.8 Hz, 2H, Ph), 7.56 (d, J = 17.0 Hz, 1H, CHCH), 7.29 (d, J = 16.4
Hz, 1H, CHCH), 6.53 (dd, J = 2.7, 1.7 Hz, 2H, Pz). ESI-TOF mass
(positive, acetonitrile): m/z = 381.1073 [M + Na]+, calcd for
C19H14O2N6Na, 381.1076. Analytical data. Found: C, 63.66; H, 4.09;
N, 23.36%. Calcd for C19H14O2N6: C, 63.68; H, 3.94; N, 23.45%.
Synthesis of Z-NO2. Under a nitrogen atmosphere, (Z)-2,6-di(1H-
pyrazol-1-yl)-4-(4-nitrostyryl)pyridine (52 mg, 0.15 mmol) was
dissolved in 10 mL of dry acetone. Fe(BF4)2·6H2O (24 mg, 0.071
mmol) dissolved in 2 mL of dry acetone was added to this solution at
room temperature. The reaction mixture immediately turned from
colorless to red. After stirring for 30 min, most of the solvent was
evaporated under a stream of nitrogen at room temperature. The
residue was recrystallized from acetone/diethyl ether. The precipitate
was filtered and washed by diethyl ether. After drying under vacuum at
100 °C, Z-NO2 was obtained as orange microcrystals. Yield: 55 mg
(0.058 mmol, 82%). ESI-TOF mass (positive, acetonitrile): m/z =
386.10 [M]2+, calcd for C38H28O4N12Fe, 386.09. Analytical data.
Found: C, 48.06; H, 3.19; N, 17.64%. Calcd for C38H28O4N12B2F8Fe:
C, 48.24; H, 2.98; N, 17.76%.
Synthesis of E-NO2. Under a nitrogen atmosphere, (E)-2,6-di(1H-
pyrazol-1-yl)-4-(4-nitrostyryl)pyridine (100 mg, 0.28 mmol) was
dissolved in 12 mL of dry acetone. Fe(BF4)2·6H2O (46 mg, 0.14
mmol) dissolved in 5 mL of dry acetone was added to the former
stirred solution at room temperature. The reaction mixture
immediately turned from colorless solution to dark-red suspension.
After stirring for 30 min, about 2/3 of the volume was slowly
evaporated under a stream of nitrogen at room temperature, forming a
red precipitate. The precipitate was filtered and washed with 2 mL of
acetone. After recrystallization from nitromethane/dietyl ether and
under vacuum at 100 °C, E-NO2 was obtained as red microcrystals.
Yield: 112 mg (0.12 mmol, 86%). ESI-TOF mass (positive,
acetonitrile): m/z = 386.11 [M]2+, calcd for C38H28O4N12Fe, 386.09.
Analytical data. Found: C, 46.09; H, 3.32; N, 16.92%. Calcd for
C38H28O4N12B2F8Fe·2.5H2O: C, 46.05; H, 3.36; N, 16.96%.
Synthesis of 4-Nitrobenzyltris(2-(methoxymethoxy)phenyl)-
phosphonium Bromide. Tris(2-(methoxymethoxy)phenyl)-
phosphine (5.97 g, 13.5 mmol), 4-nitrobenzyl bromide (3.81 g, 17.6
mmol), and sodium hydrogen carbonate (0.81 g, 9.6 mmol) were
dissolved in 60 mL of acetonitrile and heated to 80 °C for 10 min.
After evaporation of the solvent, the residue was extracted with
dichloromethane (100 mL), and insoluble materials were filtered off.
The filtrate was concentrated, and ethyl acetate (60 mL) was added to
form a pale yellow precipitate, which was collected and washed by
ethyl acetate. The title compound was obtained as pale yellow powder.
Yield: 8.33 g (12.7 mmol, 94%). 1H NMR (acetone-d6, 295 K): δ 8.02
(d, J = 8.1 Hz, 2H, Bz), 7.86 (dd, J = 15.1, 7.8 Hz, 3H, Ph), 7.82−7.78
(m, 3H, Ph), 7.56 (dd, J = 8.5, 2.0 Hz, 2H, Bz), 7.38−7.31 (m, 6H,
Photoirradiation Experiments. For solution samples and KBr-
pelletized samples, the light source for solution samples was supplied
by a high-pressure mercury lamp (USH-500D, Ushio), and the bright
5196
dx.doi.org/10.1021/ic300030b | Inorg. Chem. 2012, 51, 5188−5198

Inorganic Chemistry
Article
lines were separated by a monochromator (Jasco CT-10T) to extract
436 nm light with an intensity of 2.5 mW/cm2. For the solid samples,
light from a high-pressure mercury lamp was allowed to get through a
UV cut filter to reduce λ < 420 nm light. The intensity of the light of λ
> 420 nm was about 200 mW/cm2. Photoirradiation was conducted at
295 K.
Single-Crystal X-ray Structure Analysis. Diffraction data were
collected with a Rigaku AFC10 diffractometer with a Rigaku Saturn
70CCD system equipped with a rotating-anode X-ray generator. Single
crystals suitable for this analysis were mounted on a loop fiber with
liquid paraffin. Graphite-monochromated Mo Kα radiation (λ =
0.71073 Å) was utilized. The diffraction data were integrated and
compensated with numerical correction with the Crystal Clear 1.3.6
program. The structure was solved with SIR-9254 and the whole
structure was refined using SHELXL-9755 by the full-matrix least-
squares techniques on F2. All calculations were performed using
WinGX-1.70.0156 software package. All non-hydrogen atoms were
refined anisotropically. Hydrogen atoms of FeII complexes were
located in the idealized positions and were refined using a riding model
with fixed thermal parameters.
Preparation of Microcrystalline Samples. Solvent-free micro-
crystals of Z-H and Z-CN were obtained by recrystallization from
acetone/diethyl ether. On the other hand, recrystallization in various
solvents did not provide us solvent-free microcrystals of Z-NO2;
instead, those including solvent molecules Z-NO2·acetone were
obtained by recrystallization from acetone/diethyl ether. Elemental
analysis and thermogravimetric analysis (TGA) supported that the
microcrystals of Z-NO2·acetone contained one acetone molecule per
one Z-NO2 (Supporting Information, Figure S6). Z-H, Z-CN, and Z-
NO2·acetone were investigated by means of single-crystal XRD
(Figure 11 and Table 3). Solvent-free Z-NO2 was eventually obtained
by vacuuming Z-NO2·acetone at 373 K. Microcrystals of E-CN and E-
NO2 were fabricated by recrystallization from nitromethane/dietyl
ether, the compositions of which proved to be E-CN·MeNO2 and E-
NO2·H2O by means of single-crystal XRD (Figure 11 and Table 3).
Magnetic Susceptibilities in Solution. Magnetic susceptibilities
were measured by a paramagnetic shift of a reference TMS signal in
1H NMR spectra (the Evans method)38 using acetone-d6 containing 1
vol % TMS as a solvent with a Bruker 500 MHz spectrometer
equipped with a variable-temperature unit. A 5 mmϕ clear fused quartz
tube (528-PP-7QZ, Wilmad Labglass, Buena, NJ) and a 2 mmϕ stem
coaxial insert (528-PP-7QZ, Wilmad Labglass, Buena, NJ) were
utilized as a cell. A concentration of FeII complexes was fabricated to
be 1.00 × 10−3 M, and 1.0 equivalent of corresponding ligand was
added to prevent ligand dissociations. The molar magnetic
susceptibilities were corrected with the diamagnetism of the complexes
and ligands, which were calculated from Pascal’s constants.
Diamagnetic susceptibilities were estimated to be −441 × 10−6 cm3
mol−1 for Z-H, −464 × 10−6 cm3 mol−1 for Z-CN, and −466 × 10−6
cm3 mol−1 for Z-NO2 with one equivalent of their corresponding
ligand. Changes in density of the solutions were corrected according to
the literature.53
1
Other Measurements and Apparatus. H NMR spectra were
recorded with a JEOL AL-400 (400 MHz) or a Bruker DRX-500 (500
MHz) spectrometer. ESI-TOF-MS spectra were recorded with a
Micromass LCT, a time-of-flight mass spectrometer. MALDI-TOF-MS
spectra were recorded with a Shimadzu AXIMA-CER. UV−vis spectra
in solutions were recorded with a JASCO V-570 spectrometer at 295
K. Quartz cells (optical path: 1 cm) were used for the measurement.
FT-IR spectra were recorded with a JASCO FT/IT-620 V
spectrometer. All spectra were measured as KBr pellets. A blank
spectrum was measured by the use of clean KBr pellets as a reference
sample before the measurement of each compound.
ASSOCIATED CONTENT
■
S
* Supporting Information
FT-IR spectra of Z-CN and Z-NO2 in KBr pellets before and
after irradiation with visible light (Figure S1), PXRD pattern for
Z-NO2 before and after visible light irradiation (Figure S2),
DSC of Z-NO2 (Figure S3), FT-IR spectrum of Z-
NO2·acetone after visible light irradiation (Figure S4),
temperature dependence of χMT for E-CN and E-NO2 (Figure
S5), TGA chart of Z-NO2·acetone (Figure S6). This material is
available free of charge via the Internet at http://pubs.acs.org.
The fractions of the HS and LS species, γHS and γLS, at each
temperature were calculated from the experimental data under an
assumption of χMT being 3.75 cm3 K mol−1 and 0 cm3 K mol−1 for the
HS and LS states, respectively. The equilibrium constant was defined
as eq 1
AUTHOR INFORMATION
■
Corresponding Author
*E-mail: nisihara@chem.s.u-tokyo.ac.jp.
K = γHS/γLS
Notes
(1)
The authors declare no competing financial interest.
The calculated ln K was ploted versus 1/T, so as to find a linear
relationship in each complex (Figure 6). Provided van’t Hoff eq 2 is
valid, thermodynamic parameters can be extracted from the slope and
intercept of the ln K vs 1/T plot:
ACKNOWLEDGMENTS
■
The authors acknowledge Grants-in-Aid from MEXT of Japan
(Nos. 20245013 and 21108002, area 2107, Coordination
Programming) and the Global COE Program for Chemistry
Innovation for financial support.
(2)
ln K = −ΔH/RT + ΔS/R
where ΔH and ΔS are activation enthalpy and entropy, respectively.
Then T1/2 that gives K = 1 can be calculated as follows:
T
= ΔH/ΔS
(3)
1/2
REFERENCES
■
(1) Gutlich, P.; Hauser, A.; Spiering, H. Angew. Chem., Int. Ed. Engl.
̈
SQUID Measurements. Magnetic susceptibilities in the solid state
were measured with an applied magnetic field of 10000 Oe, in a
temperature range of 5−300 K, and at a rate of 1 K min−1 (sweep
mode) with a Quantum Design MPMS SQUID magnetometer.
Aluminum foils (Nippaku, Inc.) were used as sample containers for the
measurements. The magnetic contribution of aluminum foils was
always subtracted. Typical weights of aluminum foils and samples were
12 mg and 1.5 mg, respectively, and the typical heights of the samples
were 4 mm. The molar magnetic susceptibilities were corrected with
the diamagnetism of the complexes, which were calculated from
Pascal’s constants. Diamagnetic susceptibilities were estimated to be
−324 × 10−6 cm3 mol−1 for Z-H, −340 × 10−6 cm3 mol−1 for Z-CN,
−341 × 10−6 cm3 mol−1 for Z-NO2, and −375 × 10−6 cm3 mol−1 for
Z-NO2·acetone.
1994, 33, 2024−2054.
(2) In Spin crossover in Transition Metal Compounds I-III; Gutlich, P.,
̈
Goodwin, H. A., Eds.; Topics in Current Chemistry; Springer-Verlag:
Berlin, Germany, 2004; Vols. 233−235.
(3) Real, J. A.; Gaspara, A. B.; Munoz, M. C. Dalton Trans. 2005,
̃
2062−2079.
(4) Let
́
ard, J.-F. J. Mater. Chem. 2006, 16, 2550−2559.
(5) Decurtins, S.; Gutlich, P.; Kohler, C. P.; Spiering, H.; Hauser, A.
̈
̈
Chem. Phys. Lett. 1984, 105, 1−4.
́ ́
(6) Letard, J.-F.; Capes, L.; Chastanet, G.; Moliner, N.; Letard, S.;
Real, J. A.; Kahn, O. Chem. Phys. Lett. 1999, 313, 115−120.
́
(7) Marcen, S.; Lecren, L.; Capes, L.; Goodwin, H. A.; Letard, J.-F.
Chem. Phys. Lett. 2002, 358, 87−95.
5197
dx.doi.org/10.1021/ic300030b | Inorg. Chem. 2012, 51, 5188−5198

Inorganic Chemistry
Article
(8) Let
́
ard, J.-F.; Guionneau, P.; Nguyen, O.; Costa, J. S.; Marcen
́
, S.;
(40) Wu, D.-Y.; Sato, O.; Einaga, Y.; Duan, C.-Y. Angew. Chem., Int.
Ed. 2009, 48, 1475−1478.
Chastanet, G.; Marchivie, M.; Goux-Capes, L. Chem.Eur. J. 2005, 11,
(41) Kitchen, J. A.; Jameson, G. N. L.; Tallon, J. L.; Brooker, S. Chem.
Commun. 2010, 46, 3200−3202.
4582−4589.
(9) Niel, V.; Gaspar, A. B.; Munoz, M. C.; Abarca, B.; Ballesteros, R.;
̃
(42) Holland, J. M.; McAllister, J. A.; Kilner, C. A.; Thornton-Pett,
M.; Bridgeman, A. J.; Halcrow, M. A. J. Chem. Soc., Dalton Trans. 2002,
548−554.
Real, J. A.; Zhang, L. Inorg. Chem. 2003, 42, 4782−4788.
(10) Hayami, S.; Kawajiri, R.; Juhasz, G.; Kawahara, T.; Hashiguchi,
K.; Sato, O.; Inoue, K.; Maeda, Y. Bull. Chem. Soc. Jpn. 2003, 76,
1207−1213.
(11) Neville, S. M.; Leita, B. A.; Halder, G. J.; Kepert, C. J.;
Moubaraki, B.; Let
(43) Carbonera, C.; Costa, J. S.; Money., V. A.; Elhaïk, J.; Howard, J.
́
A. K.; Halcrow, M. A.; Letard, J.-F. Dalton Trans. 2006, 3058−3066.
(44) Nihei, M.; Han, L.; Oshio, H. J. Am. Chem. Soc. 2007, 129,
5312−5313.
́
ard, J.-F.; Murray, K. S. Chem.Eur. J. 2008, 14,
10123−10133.
(45) Rajadurai., C.; Qu, Z.; Fuhr, O.; Gopalan, B.; Kruk, R.; Ghafari,
M.; Ruben, M. Dalton Trans. 2007, 3531−3537.
(12) Zhang, L.; Xu, G.-C.; Xu, H.-B.; Mereacre, V.; Wang, Z.-M.;
Powell, A. K.; Gao, S. Dalton Trans. 2010, 39, 4856−4868.
́
(46) Guionneau, P.; Marchivie, M.; Bravic, G.; Letard, J.-F.;
́
(13) Yamada, M.; Fukumoto, E.; Ooidemizu, M.; Brefuel, N.;
Chasseau, D. J. Mater. Chem. 2002, 12, 2546−2551.
(47) Halcrow, M. A. Chem. Soc. Rev. 2011, 40, 4119−4142.
(48) Hostettler, M.; Tornroos, K. W.; Chernyshov, D.; Vangdal, B.;
Matsumoto, N.; Iijima, S.; Kojima, M.; Re, N.; Dahan, F.; Tuchagues,
J.-P. Inorg. Chem. 2005, 44, 6967−6974.
̈
(14) Sunatsuki, Y.; Ohta, H.; Kojima, M.; Ikuta, Y.; Goto, Y.;
Matsumoto, N.; Iijima, S.; Akashi, H.; Kaizaki, S.; Dahan, F.;
Tuchagues, J.-P. Inorg. Chem. 2004, 43, 4154−4171.
Burgi, H.-B. Angew. Chem., Int. Ed. 2004, 43, 4589−4594.
̈
(49) Elhaïk, J.; Pask, C. M.; Kilner, C. A.; Halcrow, M. A. Tetrahedron
2007, 63, 291−298.
(15) Yamada, M.; Ooidemizu, M.; Ikuta, Y.; Osa, S.; Matsumoto, N.;
Iijima, S.; Kojima, M.; Dahan, F.; Tuchagues, J.-P. Inorg. Chem. 2003,
42, 8406−8416.
(16) Ikuta, Y.; Ooidemizu, M.; Yamahata, Y.; Yamada, M.; Osa, S.;
Matsumoto, N.; Iijima, S.; Sunatsuki, Y.; Kojima, M.; Dahan, F.;
Tuchagues, J.-P. Inorg. Chem. 2003, 42, 7001−7017.
(17) Hayami, S.; Gu, Z.-Z.; Einaga, Y.; Kobayashi, Y.; Ishikawa, Y.;
Yamada, Y.; Fujishima, A.; Sato, O. Inorg. Chem. 2001, 40, 3240−3242.
(18) Bonhommeau, S.; Guillon, T.; Daku, L. M. L.; Demont, P.;
(50) Chandrasekar, R.; Frank, F.; Susan, B.; Olaf, F.; Mohamed, G.;
Robert, K.; Mario, R. Inorg. Chem. 2006, 45, 10019−10021.
(51) Jaganathan, S.; Tsukamoto, M.; Schlosser, M. Synthesis 1990,
109−111.
(52) Tsukamoto, M.; Schlosser, M. Synlett 1990, 605−608.
(53) Terent’eva, A. A.; Krumgal’z, B. S.; Gerzhberg, Y. I. Zh. Prkl.
Khim. 1973, 56, 1143−1144.
(54) Altomare, A.; Cascarano, G.; Guagliardi, A.; Burla, M. C.;
Polidori, G.; Camalli, M. J. Appl. Crystallogr. 1994, 27, 435−436.
(55) Sheldrick, G. M. SHELX-97, Programs for Crystal Structure
́
Costa, J. S.; Letard, J.-F.; Molnar, G.; Bousseksou, A. Angew. Chem., Int.
Ed. 2006, 45, 1625−1629.
Analysis (Release 97-2); University of Gottingen: Gottingen, Germany,
̈
̈
(19) Guionneau, P.; Gac, F. L.; Kaiba, A.; Costa, J. S.; Chasseau, D.;
1997.
́
Letard, J.-F. Chem. Commun. 2007, 3723−3725.
(56) Farrugia, L. J. J. Appl. Crystallogr. 1999, 32, 837−838 (an
integrated system of Windows programs for the solution, refinement,
and analysis of single-crystal X-ray diffraction data).
(20) Roux, C.; Zarembowitch, J.; Gallois, B.; Granier, T.; Claude, R.
Inorg. Chem. 1994, 33, 2273−2279.
(21) Hasegawa, Y.; Kume, S.; Nishihara, H. Dalton Trans. 2009,
280−284.
(22) Boillot, M.-L.; Chantraine, S.; Zarembowitch, J.; Lallemand, J.-
Y.; Prunet, J. New. J. Chem. 1999, 179−183.
(23) Boillot, M.-L.; Pillet, S.; Rivier
Inorg. Chem. 2009, 48, 4729−4736.
̀
(24) Senechal-David, K.; Zaman, N.; Walko, M.; Halza, E.; Riviere,
̀
e, E.; Claiser, N.; Lecomte, C.
́
́
E.; Guillot, R.; Feringa, B. L.; Boillot, M.-L. Dalton Trans. 2008, 1932−
1936.
(25) Tissot, A.; Boillot, M.-L.; Pillet, S.; Codjovi, E.; Boukheddaden,
K.; Daku, L. M. L. J. Phys. Chem. C 2010, 114, 21715−21722.
(26) Hasegawa, Y.; Takahashi, K.; Kume, S.; Nishihara, H. Chem.
Commun. 2011, 47, 6846−6848.
(27) Holland, J. M.; McAllister, J. A.; Lu, Z.; Kilner, C. A.; Thornton-
Pett, M.; Halcrow, M. A. Chem. Commun. 2001, 577−578.
(28) Tweedle, M. F.; Wilson, L. J. J. Am. Chem. Soc. 1976, 98, 4824−
4834.
(29) Nakano, K.; Suemura, N.; Yoneda, K.; Kawata, S.; Kaizaki, S.
Dalton Trans. 2005, 740−743.
(30) Dupouy, G.; Marchivie, M.; Triki, S.; Sala-P., J.; Salaun, J.-Y.;
̈
́
Gomez-G., C. J.; Guionneau, P. Inorg. Chem. 2008, 47, 8921−8931.
(31) Marconi, G.; Bartocci, G.; Mazzucato, U.; Spalletti, A.; Abbate,
F.; Angeloni, L.; Castellucci, E. Chem. Phys. 1995, 196, 383−393.
(32) Toftlund, H. Coord. Chem. Rev. 1989, 94, 67−108.
(33) Bree, A.; Zwarich, R. J. Mol. Struct. 1981, 75, 213−224.
(34) Hamaguchi, H.; Iwata, K. Bull. Chem. Soc. Jpn. 2002, 75, 883−
897.
(35) Johnson, C. S.; Chang, R. J. Chem. Phys. 1965, 43, 3183−3192.
(36) Happ, J. W.; Ferguson, A.; Whitten, D. G. J. Org. Chem. 1972,
37, 1485−1491.
(37) Kobatake, S.; Hasegawa, H.; Miyamura, K. Cryst. Growth Des.
2011, 11, 1223−1229.
(38) Evans, D. F. J. Chem. Soc. 1959, 2003−2005.
(39) Hansch, C.; Leo, A.; Taft, R. W. Chem. Rev. 1991, 91, 165−195.
5198
dx.doi.org/10.1021/ic300030b | Inorg. Chem. 2012, 51, 5188−5198