Synergetic effect of host-guest chemistry and spin crossover in 3D hofmann-like metal-organic frameworks [Fe(bpac)M(CN)4] (M=Pt, Pd, Ni)

Article abstract of DOI:10.1002/chem.201102357

The synthesis and characterization of a series of three-dimensional (3D) Hofmann-like clathrate porous metal-organic framework (MOF) materials [Fe(bpac)M(CN)₄] (M=Pt, Pd, and Ni; bpac=bis(4-pyridyl)acetylene) that exhibit spin-crossover behavior is reported. The rigid bpac ligand is longer than the previously used azopyridine and pyrazine and has been selected with the aim to improve both the spin-crossover properties and the porosity of the corresponding porous coordination polymers (PCPs). The 3D network is composed of successive {Fe[M(CN)₄]}_n planar layers bridged by the bis-monodentate bpac ligand linked in the apical positions of the iron center. The large void between the layers, which represents 41.7% of the unit cell, can accommodate solvent molecules or free bpac ligand. Different synthetic strategies were used to obtain a range of spin-crossover behaviors with hysteresis loops around room temperature; the samples were characterized by magnetic susceptibility, calorimetric, Moessbauer, and Raman measurements. The complete physical study reveals a clear relationship between the quantity of included bpac molecules and the completeness of the spin transition, thereby underlining the key role of the π-π stacking interactions operating between the host and guest bpac molecules within the network. Although the inclusion of the bpac molecules tends to increase the amount of active iron centers, no variation of the transition temperature was measured. We have also investigated the ability of the network to accommodate the inclusion of molecules other than water and bpac and studied the synergy between the host-guest interaction and the spin-crossover behavior. In fact, the clathration of various aromatic molecules revealed specific modifications of the transition temperature. Finally, the transition temperature and the completeness of the transition are related to the nature of the metal associated with the iron center (Ni, Pt, or Pd) and also to the nature and the amount of guest molecules in the lattice. Copyright

Full text of DOI:10.1002/chem.201102357

FULL PAPER
DOI: 10.1002/chem.201102357
Synergetic Effect of Host–Guest Chemistry and Spin Crossover in
D Hofmann-like Metal–Organic Frameworks
Fe(bpac)M(CN) ] (M=Pt, Pd, Ni)**
3
[
ACHTUNGTRENNUNG
4
[
a, b]
[a]
[a]
[b, c]
Carlos Bartual-Murgui,
Lionel Salmon, Amal Akou, Norma A. Ortega-Villar,
[d] [a] [b]
[
a]
Helena J. Shepherd, M. Carmen MuÇoz, Gꢀbor Molnꢀr, Josꢁ Antonio Real,* and
[
a]
Azzedine Bousseksou*
Abstract: The synthesis and characteri-
zation of a series of three-dimensional
molecules or free bpac ligand. Differ-
ent synthetic strategies were used to
obtain a range of spin-crossover behav-
iors with hysteresis loops around room
temperature; the samples were charac-
terized by magnetic susceptibility, calo-
rimetric, Mçssbauer, and Raman meas-
urements. The complete physical study
reveals a clear relationship between
the quantity of included bpac mole-
cules and the completeness of the spin
transition, thereby underlining the key
role of the p–p stacking interactions
operating between the host and guest
bpac molecules within the network. Al-
though the inclusion of the bpac mole-
cules tends to increase the amount of
active iron centers, no variation of the
transition temperature was measured.
We have also investigated the ability of
the network to accommodate the inclu-
sion of molecules other than water and
bpac and studied the synergy between
the host–guest interaction and the spin-
crossover behavior. In fact, the clathra-
tion of various aromatic molecules re-
vealed specific modifications of the
transition temperature. Finally, the
transition temperature and the com-
pleteness of the transition are related
to the nature of the metal associated
with the iron center (Ni, Pt, or Pd) and
also to the nature and the amount of
guest molecules in the lattice.
(
3D) Hofmann-like clathrate porous
metal–organic framework (MOF) ma-
terials [Fe(bpac)M(CN) ] (M=Pt, Pd,
ACHTUNGTRENNUNG
4
and Ni; bpac=bis(4-pyridyl)acetylene)
that exhibit spin-crossover behavior is
reported. The rigid bpac ligand is
longer than the previously used azopyr-
idine and pyrazine and has been select-
ed with the aim to improve both the
spin-crossover properties and the po-
rosity of the corresponding porous co-
ordination polymers (PCPs). The 3D
network is composed of successive
{
Fe[M(CN) ]} planar layers bridged by
4 n
the bis-monodentate bpac ligand linked
in the apical positions of the iron
center. The large void between the
layers, which represents 41.7% of the
unit cell, can accommodate solvent
Keywords: adsorption · host–guest
systems
·
metal–organic frame-
works · microporous materials ·
spin crossover
Introduction
1) they show complete cooperative thermal bistability at
room temperature with large hysteresis loops up to approxi-
[3]
Increasing attention is being paid to three-dimensional
porous coordination polymers (PCPs), also called metal–or-
mately 50 K; 2) the spin state can be switched by irradiat-
ing the sample with a pulsed laser within the hysteresis
[1]
[4]
ganic frameworks (MOFs), in particular for materials ex-
loop; 3) a reversible switching between the high-spin and
[2]
II
hibiting spin-crossover (SCO) properties. Among them,
Hofmann-like SCO-PCPs are unique in several respects:
low-spin states at the Fe sites takes place concomitantly
[5]
with the uptake of guest molecules; 4) these materials can
[
a] Dr. C. Bartual-Murgui, Dr. L. Salmon, A. Akou, Dr. H. J. Shepherd,
Dr. G. Molnꢀr, Dr. A. Bousseksou
Laboratoire de Chimie de Coordination
CNRS UPR-8241 and Universitꢁ de Toulouse
UPS, INPT, 31077 Toulouse (France)
E-mail: azzedine.bousseksou@lcc-toulouse.fr
[c] Dr. N. A. Ortega-Villar
Departamento de Quꢂmica Inorgꢀnica
Universidad Autꢃnoma de Mꢁxico (UNAM)
Mꢁxico D. F., 04510 (Mꢁxico)
[
d] Prof. Dr. M. C. MuÇoz
Departamento de Fꢂsica Aplicada
Universitat Politꢄcnica de Valꢄncia
Camino de Vera s/n, 46022, Valencia (Spain)
[
b] Dr. C. Bartual-Murgui, Dr. N. A. Ortega-Villar, Prof. Dr. J. A. Real
Instituto de Ciencia Molecular (ICMol)
Universidad de Valencia
[
**] bpac=bis(4-pyridyl)acetylene.
C/Catedrꢀtico Josꢁ Beltrꢀn Martꢂnez 2
Supporting information for this article is available on the WWW
under http://dx.doi.org/10.1002/chem.201102357. It contains additional
crystallographic data and crystal structure views, thermogravimetric
analyses, and Raman spectra.
4
6980 Paterna (Valencia) (Spain)
E-mail: jose.a.real@uv.es
Chem. Eur. J. 2012, 18, 507 – 516
ꢅ 2012 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
507

[
6]
[7]
be prepared as nanocrystals, nanoparticles, and continu-
[8]
ous or nano-patterned thin films, while still retaining the
SCO properties. The versatility of the properties of such
Hofmann-like SCO-PCPs has been demonstrated in the
series {Fe
,4’-azopyridine), which displays enhanced porosity but re-
markably lower cooperativity than the pyrazine (pz) series
ACHTUNGTRENNUNG( azpy)[M(CN) ]}·nH O (M=Ni, Pd, Pt; azpy=
4 2
[9]
4
[2a]
{
Fe(pz)[M(CN) ]}·nH O; this is likely a result of the flexi-
4 2
ble nature of the azpy ligand. Belonging to this family of
compounds, the pillared ligand of the title compounds has
been chosen to affect both the SCO properties and the ca-
pacity to sense invited molecules. Indeed, the bis(4-pyrid-
ACHTUNGTRENNUNGy l)acetylene (bpac) ligand is rigid and longer than azpy and
4
Figure 1. Perspective of a fragment of {Fe ACTHUNGTRNEGNU( bpac)[Pt(CN) ]} coordination
polymer. The inclusion of bpac has been omitted for clarity (color code:
Fe, yellow; Pt, lilac; N, blue; C, gray).
action is observed. In the present paper, we show a com-
plete study of this new series of SCO Hofmann-like coordi-
nation polymers {Fe ACHTUNGTERNNNU(G bpac)[M(CN) ]} (M=Pt, Pd, and Ni).
4
We present different synthetic methods, the study of the
composition, as well as the structural and physical character-
ization of the compounds. It is shown how small variations
of the synthetic procedure may give rise to materials that
differ considerably in composition and physical behavior.
Moreover, with the aim of studying the synergetic effect of
the host–guest chemistry and the spin-crossover phenomen-
on, different strategies have been developed for the inclu-
sion of various guest molecules in these three-dimensional
networks.
Results and Discussion
pz and so it should improve both the cooperativity and the
porosity of the corresponding PCP materials. In recent re-
ports, we have described the structural and magnetic proper-
Synthesis and characterization: Although several methods
for the synthesis of the bpac ligand are found in the litera-
[
11]
[3]
ture, we have developed a new method inspired by the
palladium-catalyzed coupling reaction of haloheteroaromat-
[3a]
ties of the novel {Fe
A
H
U
G
R
N
U
G
and {Fe-
4
[10]
[12]
A
C
H
T
U
N
G
T
R
E
N
N
U
N
G
(bpac) [Ag(CN) ] } compounds. Similarly to the previous-
ic ligands reported by N. Inoue et al. The advantages of
2
2 2
ly reported 3D porous SCO coordination Hofmann poly-
this three-step method (see Scheme S1 in the Supporting In-
formation) are the possibility of working with water as a sol-
vent and the excellent yields of the coupling reaction. The
bpac ligand was prepared in 38% total yield for the three
steps.
[4b]
[9]
mers {Fe(L)[Pt(CN) ]} (L=pz, azpy ), the crystal of the
4
Pt/bpac derivative belongs to the tetragonal P4/mmm space
group. The structure is comprised of an almost regular octa-
II
hedral [Fe N ] site and a square-planar site defined by the
6
2ꢀ
anion [Pt(CN)4] (Figure 1). The equatorial positions of the
octahedron are occupied by four nitrogen atoms of four
Thereafter, two different synthetic approaches have been
used to obtain complexes with different composition. Ac-
cording to the first synthetic method (method A), treatment
of a solution of iron(II) tetrafluoroborate salt containing the
bpac ligand with a solution containing the K [M(CN) ] (M=
2ꢀ
equivalent [Pt(CN)4] groups, which act as bridges linking
four iron(II) atoms that generate {Fe[Pt(CN) ]} layers. The
4
1
layers lie on top of each other and are pillared by the ligand
bpac, which occupies the axial positions of the iron(II), thus
generating the porous 3D framework. The accessible
2
4
Ni, Pd, or Pt) salt at a controlled addition speed of 1 mL per
minute afforded the corresponding {Fe-
ACHTUNGTREN(NGNU bpac)[M(CN) ]}·H O·x(bpac) (M=Pt, x=0.71 (1); M=Pd,
3
volume without invited molecules (293.6 ꢆ ) corresponds to
4
2
4
1.7% of the unit cell at 120 K, which is more than twice
x=0.78 (2); M=Ni, x=0.40 (3)) compounds, in which x was
determined by elemental analyses. The deviation of the Pt/
Fe ratio up to 0.97 and the concomitant detection of traces
of boron atoms by elemental analyses are in agreement with
[4]
the accessible volume of the homologous pz derivative.
This volume is partially occupied by molecules of bpac lo-
calized in the channels with their aromatic rings in front of
those of the pillared bpac ligands, so that p–p stacking inter-
2ꢀ
the presence of [Pt(CN)4] vacancies and the related pres-
508
www.chemeurj.org
ꢅ 2012 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Chem. Eur. J. 2012, 18, 507 – 516

Metal–Organic Frameworks
FULL PAPER
ꢀ
ence of BF4 anions according to charge neutrality. This hy-
pothesis also corroborates the low nitrogen percentage
found for most of the samples. The quantity of included
water molecules was measured by thermogravimetric analy-
ses (see Figure S1 in the Supporting Information) and ap-
proximately one molecule per iron(II) ion was found for all
compounds. Moreover, above 500 K, a significant loss of
mass can be attributed to a fraction of included bpac mole-
cules. One should note that the syntheses have been repro-
duced many times and the obtained products appeared to
have slightly different stoichiometry. Nevertheless, general
tendencies can be established as a function of the synthetic
method used (see magnetic measurements, discussed
below). As confirmed by the large quantity of included bpac
molecules (x) in samples 1–3, syntheses following method A
were carried out with a continuous excess amount of bpac
during the addition of the metallocyanate salt. With the aim
of limiting this inclusion, an alternative method B was used,
which consisted of adding a solution containing both
iron(II) tetrafluoroborate salt and bpac ligand to a solution
of K [Pt(CN) ] salt. In this case, as expected, a quite differ-
Figure 3. X-ray powder diffraction patterns of 1–4 measured at 298 K and
calculated from single-crystal data at 120 K.
a(b) and c axes, respectively. However, no thermal expan-
sion was measured between 148 and 223 K in the low-spin
state and very small thermal expansion was measured be-
tween 340 and 400 K in the high-spin state. This unusual
phenomenon was also reported in the pyrazine derivative
[4b]
[Fe(pz)Pt(CN) ],
and has been explained in other com-
2
4
4
ent composition {Fe ACHTUNGTRNEUN(G bpac)[Pt(CN) ]}·H O·0.09(bpac) (4)
4 2
pounds containing a cyanide-bridged framework as stem-
[13]
with a significantly lower quantity of included bpac mole-
cules was obtained. Although the quantity of included bpac
molecules differs depending on whether method A or B was
used, no significant effect was observed on the quantity of
ming from the great flexibility of the cyanide bridge. Fig-
ure S2 in the Supporting Information shows the thermal var-
iation of the volume of the unit cell for compound 1, which
3
increases significantly during the spin transition (69 ꢆ ,
2ꢀ
vacancies of [Pt(CN) ] . In the following studies of the
9.7%). This value is large compared to the generally ob-
4
physical properties and experiments of guest inclusion, com-
pounds 1’ and 1’’ (4’ and 4’’) correspond to samples formed
in the same manner as sample 1 (4), but with slightly differ-
ent compositions. As an example, Figure 2 shows the color
change of sample 1 from red to yellow when passing from
the low-spin (LS) to the high-spin (HS) state.
served 1–6% variation in SCO complexes with FeN coordi-
6
[14]
nation polyhedra,
although similar in magnitude to the
pyrazine and azopyridine derivatives (13 and 9.3%, respec-
tively). The largest increase in the case of the pyrazine de-
rivative can be explained by the rather small unit-cell
volume of the complex. This increase of the unit-cell
volume is essentially ascribed to the increase of FeꢀN bond
lengths (0.2 ꢆ on average) upon spin transition since ther-
mal expansion is relatively insignificant in the investigated
temperature range. Figure 3 also shows powder X-ray dif-
fraction patterns of 4 (method B) at 298 K. Although all the
patterns are clearly related, most notably with the character-
istic peak at 2q=6.508 indexed as the (001) reflection, there
are also differences in the peak widths between the patterns
that may be attributed to greater disorder within the sample
with an increasing proportion of defects.
Figure 2. Photographs of powder samples of 1 in both spin states.
X-ray powder diffraction: Powder X-ray diffraction patterns
of samples 1–3 at 298 K are shown in Figure 3. Their similar-
ities demonstrate the close structural relationship of the Pt,
Pd, and Ni derivatives, also confirmed by Raman spectros-
copy. Subtle differences for the nickel derivative may be ex-
plained by the more important distortion of the network in
agreement with the lower ionic radius than the platinate and
palladate derivatives. The X-ray powder diffraction profile
of 1 was satisfactorily fitted in the temperature range of 150
Differential scanning calorimetry (DSC) analysis: The calo-
rimetric data (Figure 4) were recorded in the heating and
ꢀ1
cooling modes at 5 Kmin to evaluate the enthalpy (DH)
and entropy (DS) variations associated with the spin transi-
tions of 1–3 in the hydrated and dehydrated forms (which
were obtained by heating at 400 K). For 3, the DSC curve
shows a singularity at T =275 K both for the cooling and
c
heating modes. It was not possible to observe any singularity
for the dehydrated form of sample 3 due to the gradual
nature of the spin crossover (see magnetic measurements
below). The platinum and palladium derivative samples in
the hydrated form show several maxima in the heating and
cooling modes (T ›=312 K, T ›=323 K and T fl=306 K,
to 400 K using the tetragonal P4/mmm model (a=b=
3
7
.42(1), c=14.06(1) ꢆ; V=774.5(7) ꢆ at 380 K) in agree-
ment with the result of the single-crystal X-ray measure-
[3a]
ment.
During the HS-to-LS transition, the unit-cell pa-
rameters show a decrease of 3.3 and 2.6% measured for
c1
c2
c1
Chem. Eur. J. 2012, 18, 507 – 516
ꢅ 2012 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chemeurj.org
509

A. Bousseksou, J. A. Real et al.
dication of subtle structural modifications taking place con-
comitantly with the spin crossover and related to the pres-
ence of water molecules. The estimated DH and DS varia-
tions measured in the hydrated forms are 11 and 35; 13 and
ꢀ
1
ꢀ1
ꢀ1
4
1; and 4 kJmol and 16 JK mol for 1, 2, and 3, respec-
tively. These values are common for iron(II) spin-crossover
compounds; the small values observed for the nickel deriva-
tive are in agreement with a more incomplete spin transi-
tion.
Raman spectroscopy: Raman spectra of the dehydrated
ꢀ
1
compounds (1d–3d), recorded between 200 and 2250 cm ,
display similar features as shown in Figure S3 in the Sup-
porting Information (see also Table S1 in the Supporting In-
formation). In general, the Raman spectra of the bpac, pz,
ꢀ
1
and azpy derivatives are quite similar below 600 cm for a
given metal M (M=Ni, Pd, or Pt) because these modes are
associated mainly with metal–ligand vibration. In particular,
ꢀ
1
Raman modes of 1d (2d) around 495 (474) cm and 405
ꢀ1
(
395) cm in the LS state are assigned to metal–ligand vi-
brations with the main involvement coming from dMCN bend-
ing and nFeꢀNC stretching modes, respectively. In the HS state
ꢀ
1
these modes shift to 342 (322) and 218 (214) cm , which is
in good agreement with the observations made earlier on
[
3b]
the respective pyrazine homologues.
1
Between 600 and
ꢀ
1
700 cm , vibrational modes are attributed to the bpac ring
stretching, bending, or twisting, through taking into account
the spectrum of the pure bpac ligand and the Raman study
[9]
of the azopyridine and pyrazine derivatives. In this region,
and especially for Raman modes of the internal bpac ligand
ꢀ
1
between 950 and 1700 cm , clear variations of intensities
are observed when going from the LS to the HS state. In
ꢀ
1
particular, the HS mode at 1014 cm (assigned to ring
breathing), exhibits in the LS state a frequency shift to
ꢀ
1
1
1
028 cm . In Figure 5, the Raman spectrum of compounds
ꢀ
1
d (method A) and 4d (method B) in the 950–1040 cm
region at 313 K are compared (with that of the pure ligand
ꢀ
1
bpac). The vibrational mode at 992 cm in compound 1d is
consistent with the vibrational mode of the guest bpac
ꢀ
1
ligand, whereas the mode at 1014 cm is attributed to the
coordinated bpac ligand. Conversely, two modes at 1008 and
4 2
Figure 4. Calorimetric analysis for {Fe ACTHUNGTRENN(NUG bpac)[M(CN) ]}·nH O (M=Pt,
Pd, or Ni) powders upon heating and cooling. Pt: 1 (n=1), 1d (n=0);
Pd: 2 (n=1), 2d (n=0); Ni: 3 (n=1).
T fl=314 K for 1; T ›=313 K, T ›=317 K, T ›=323 K
c2
c1
c2
c3
and T fl=311 K, T fl=317 K for 2). For the dehydrated
c1
c2
forms (1d, 2d) only one maximum in both heating and cool-
ing modes was observed (T ›=305 K, and T fl=266 K for
c
c
1
d; T ›=308 K and T fl=261 K for 2d). In comparison
c
c
with the magnetic measurements, the slight deviation of the
transition temperature could be associated with the differ-
ence of the heating and cooling rates. The multistep features
observed in the calorimetric measurements might be an in-
Figure 5. Selected region of the Raman spectrum of 1d and 4d at room
temperature (with the spectrum of the bpac ligand for comparison).
510
www.chemeurj.org
ꢅ 2012 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Chem. Eur. J. 2012, 18, 507 – 516

Metal–Organic Frameworks
FULL PAPER
ꢀ
1
1
014 cm can be ascribed to the breathing of the coordinat-
first cooling the sample from 340 K until 5 K and then heat-
ing from 5 K to 400 K at 2 Kmin . At 340 K, for each
ꢀ1
ed bpac ligand in 4d, which is in agreement with the occur-
rence of a dissymmetric ligand coordination previously sug-
gested. Moreover, the weak intensity of the mode at
sample, the c T value is typical of iron(II) HS compounds
M
3
ꢀ1
(3.20–3.50 cm Kmol ), whereas at 50 K it is typical for an
iron(II) ion in the LS state with a possible small residual
fraction of iron(II) centers in the HS state: 0.50, 0.45, and
ꢀ
1
9
92 cm confirms the small fraction of included bpac mole-
cules in 4d. The temperature dependence of the normalized
Raman intensity ratio I1014 cmꢀ1/(I1014 cmꢀ1 +I1028 cmꢀ1) reported in
Figure S4 in the Supporting Information for 1d and 4d and
showing T fl=242 K and T ›=308 K and T fl=235 and
3
ꢀ1
1 cm Kmol for 1, 2 and 3, respectively. The transition
temperature is T =303 K for 1, whereas a hysteresis is ob-
c
served for 2 (20 K) and 3 (13 K): T fl=308 (2) and 238 K
c
c
c
c
T ›=260 K, respectively, corroborates the magnetic and
calorimetric data.
(3) and T ›=328 K (2), and 251 K (3) for the cooling and
1/2
c
heating modes, respectively. According to the thermogravi-
metric data, a water molecule is lost at approximately 390 K
in each compound. To investigate the role of this molecule
in the SCO, the samples were kept at 400 K in the supercon-
ducting quantum interference device (SQUID) magnetome-
ter for 60 min to dehydrate them in situ. Then, a subsequent
cooling–heating cycle was performed to determine the mag-
netic behavior of the dehydrated samples 1d–3d (Figure 6).
The dehydrated forms 1d and 2d show spin crossover with a
large hysteresis of 68 and 57 K, respectively. The transition
temperatures are T fl=246 (1d) and 260 K (2d), and T ›=
Magnetic measurements: c T versus T plots (cM is the
M
molar magnetic susceptibility) for 1–3 are displayed in
Figure 6. The magnetic measurements were performed by
c
c
3
14 (1d) and 317 K (2d), whereas the barycenters of the
hysteresis loops (T =(T fl+T ›)/2), experience a relatively
c
c
c
small decrease (T =280 (1d) and 289 K (2d)), still remain-
c
ing near room temperature. This contrasts with the case of
the {Fe(pz)[M(CN) ]}·H O and {Fe CAHTUNGTRNEUN(G azpy)[M(CN) ]}·H O
4 2
4
2
(
M=Pd and Pt) series. Upon dehydration, the transition
temperatures of the former increase up to 100 K and the
[2a]
hysteresis widens by up to 25 K, whereas in the latter the
transition temperatures decrease by 100 K and the hysteresis
[9]
widens by 10 K. These data confirm the occurrence of
stronger cooperativity in the bpac derivatives. All these re-
sults clearly demonstrate the strong effect of water mole-
cules on the properties of this family of compounds and
more generally the effect of solvent molecules on the spin-
crossover phenomenon. Moreover, the higher T values ob-
c
tained for the dehydrated form of the pyrazine and bpac ho-
mologues than with the azpy derivative could be explained
by a more significant internal pressure of the three-dimen-
sional network induced by the more important rigidity of
the ligand, which also accounts for the stabilization of the
LS state. Subsequently to the dehydration process, the evo-
lution of the behavior of the nickel derivative is different
from those of the platinum and palladium derivatives;
indeed, sample 3d shows a gradual and incomplete spin
crossover centered at 220 K. The general significant differ-
ences obtained for the nickel complex (already reported for
the azopyridine derivatives) seem to be related to structural
modifications that originate in the smaller ionic radius in
2
+
the case of Ni . Figure 7 shows the comparison of the ther-
mal dependence of the c T product of hydrated (1’’, 1’, 4’,
M
and 4) and dehydrated (1’’d, 1’d, 4’d, and 4d) forms of the
platinum derivatives synthesized following the two different
methods with increasing quantity of included bpac mole-
cules (x). Clear evolution of the SCO behavior can be ob-
served between the four complexes. Although the tempera-
Figure 6. Magnetic properties, in the form of c
(bpac)[M(CN) ]}·nH O (M=Pt, Pd, or Ni) powders upon heating and
cooling. Pt: 1 (n=1), 1d (n=0); Pd: 2 (n=1), 2d (n=0); Ni: 3 (n=1),
d (n=0).
M
T versus T curves, of {Fe-
A
C
H
T
U
N
G
T
R
E
N
N
U
N
G
4
2
3
tures of the transition are similar (T =303, 302, 298, 297 K,
c
Chem. Eur. J. 2012, 18, 507 – 516
ꢅ 2012 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chemeurj.org
511

A. Bousseksou, J. A. Real et al.
T fl=262 and T ›=282; for 1’’,
c
c
1
’, 4’, and 4, respectively. The
cooperativity of the network is
thus improved by both the ab-
sence of inserted water mole-
cules and the presence of guest
bpac molecules.
Mçssbauer spectroscopy: Rep-
5
7
resentative
spectra of 1 and 4 (317 and
0 K) recorded in the cooling
mode are shown in Figure 8
Fe Mçssbauer
8
Figure 7. Magnetic properties, in the form of c
ing and cooling synthesized following method A (1’ and 1’’ (n=1), 1’d and 1’’d (n=0)) and method B (4 and 4’ and values of the Mçssbauer
n=1), 4d and 4’d (n=0)) with increasing amount of included bpac molecules.
M 4
T versus T curves, of {Fe ACHTUNRTGENNUNG( bpac)[Pt(CN) ]} powders upon heat-
(
parameters obtained by least-
for 1’’, 1’, 4’, and 4, respective-
ly), increasingly gradual and in-
complete spin transitions occur
when passing from 1’’ to 4.
These observations reveal a less
cooperative behavior and the
presence of an important resid-
ual HS iron(II) fraction as dem-
onstrated by the high c T prod-
M
3
uct at 150 K of 1.75 cm K for 4.
This effect is clearly correlated
with the proportion of bpac
molecules included within the
pores. The important finding
here is that the increase of the
quantity of included bpac mole-
cules in the network corre-
sponds to the increase of the
completeness of the spin transi-
tion. In other words, based on
the features of the spin transi-
tion of compounds with differ-
ent stoichiometry, the complete
and cooperative spin transition
5
7
Figure 8. Fe Mçssbauer spectra of 1, 4, 4(pz) and 4 ACHTUNGRNENUG( pz-py) in the high- and low-spin states.
seems to imply the presence of
approximately one included bpac molecule per iron atom,
whereas a lower quantity leads to incomplete transitions.
When the quantity of included molecules tends towards
zero, the compound undergoes only a half spin transition.
Even though structural determination of compound 4 was
not possible, there is strong evidence from Raman and
Mçssbauer measurements (see below) that the strong rigidi-
ty of the network leads to a compressed structure that can
only be accommodated by an asymmetrical coordination of
the bridging bpac ligand, which leads to the occurrence of
two distinct iron centers, with only one being spin-crossover
active. For each compound, the dehydration process reveals
hysteresis loops of 63, 42, 28, and 20 K while slightly de-
squares fitting of the spectra are gathered in Table 1. At
317 K, the Mçssbauer spectrum of 1 consists of a unique
quadrupole-split doublet, with an isomer shift (d) of
ꢀ
ꢀ
1
1.039(2) mms
and a quadrupole spitting (DEQ) of
1
0.874(2) mms , in agreement with an iron(II) species in the
HS state. The hyperfine parameters obtained from the least-
squares fitting procedure of the spectra recorded at 80 K
ꢀ1
can be fitted with a doublet with d=0.4558(1) mms and
ꢀ1
DE =0.2391(8) mms in agreement with iron(II) in the LS
Q
state. The two spectra demonstrate the absence of residual
LS and HS fractions at high and low temperature, respec-
tively. Different features supporting the results of the mag-
netic measurements were obtained for sample 4 (method B).
At 317 K, the Mçssbauer spectrum is composed of two dif-
ferent doublets that can be attributed to iron(II) in the HS
creasing the transition temperature with T fl=262 and
c
T ›=325; T fl=274 and T ›=316; T fl=272 and T ›=300;
c
c
c
c
c
512
www.chemeurj.org
ꢅ 2012 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Chem. Eur. J. 2012, 18, 507 – 516

Metal–Organic Frameworks
FULL PAPER
5
7
[a]
Table 1. Representative least-squares-fitted Fe Mçssbauer data for 1, 4, 4(pz), and 4
Sample T HS (site A) HS (site B)
K]
ACHTUNGTRENUNNG( pz-py).
HS (site C)
LS
[
d
DE
Q
G/2
%
d
DE
Q
G/2
%
d
DE
Q
G/2
%
d
DE
0.239(8) 0.215(5) 100
0.192(1) 57(1)
Q
G/2
%
1
4
4
317 1.039(1) 0.874(2) 0.147(4) 100
8
0
0.45(1)
317 0.96(3) 0.85(4) 0.33(3)
59(1) 1.06(3) 2.75(6) 0.32(3) 41(1)
1.18(1) 3.72(3) 0.27(3) 43(2)
8
0
0.4378(6) 0.18*
317 1.027(9) 0.90(2) 0.25*
80
67(3) 1.1(2) 2.97(4) 0.2*
2.3(2) 0.94(4) 2.0(1) 0.39(1) 30.7(9)
(
pz)
1.29(6) 3.9(1) 0.13(1) 3.9(3) 1.00(4) 2.35(8) 0.35(7) 31.3(6) 0.452(7) 0.20(2) 0.18(1)
64.8(3)
80(3)
4
317 0.99(3) 0.84(6) 0.29(5)
81(11)
0.89(9) 2.1(2) 0.23* 19(8)
0.96(3) 2.52(7) 0.25* 20(2) 0.453(7) 0.20(3) 0.23(2)
A( pz-py) 80
C
H
T
U
N
G
T
R
E
N
N
U
N
G
ꢀ
1
ꢀ1
[
[
a] Isomer shifts (d in mms ) refer to metallic iron at room temperature; DE
Q
=quadrupole splitting [mms ]; G/2=half width of the lines at half height
ꢀ
1
mms ]. Statistical standard deviations are given in parentheses; values with an asterisk were fixed during the fitting procedure.
ꢀ
1
state. The low 0.86(4) mms value of DE indicates that the
Q
local symmetry is lower than cubic and the ground orbital
state is rather a doublet, which corresponds to an axial elon-
gation of the octahedral environment around the iron
ꢀ
1
center. Conversely, the large 2.75(6) mms DEQ value indi-
cates a local symmetry lower than cubic, but with a rather
well-isolated ground orbital singlet (in a well-isolated singlet
ꢀ
1
case, DEQ is in the 3–4 mms range), which corresponds to
an axial compression of the octahedral structure. When de-
creasing the temperature, at 80 K, the doublet with the
smaller quadrupole splitting (similar to sample 1), was trans-
formed into the LS state, whereas the second doublet was
not affected by the spin-state change. These results are in
agreement with those of the magnetic studies and the X-ray
crystal structure determination. In the case of 1 (equivalent
composition to the single crystals of 5), the large amount of
included bpac molecules in the network involved in p–p
stacking interactions with the coordinated bpac molecules
decreases the electron density of the bridging ligand and
thus explains the elongation of almost all the axial FeꢀN
Figure 9. Representative thermogravimetric analysis revealing dehydra-
tion and rehydration kinetics for 1.
ꢀ
1
sample has been heated at 10 Kmin from room tempera-
ture until 400 K, which confirmed the loss of approximately
one molecule of water per iron atom. Then, the system was
cooled under a stream of dry air (containing only traces of
water), which led to the partial recovery of water molecules
(gray area in the diagram). Following the small plateau, the
sample was exposed to atmospheric air (white area) showing
additional adsorption of water. A second cycle 293!400!
293 K has been fully realized under atmospheric air and
readsorption of the majority of water molecules occurs in
approximately one hour; complete recovery of the initial
composition takes 24 h.
II
bonds, which represent the SCO-active Fe centers. By con-
trast, for sample 4, only a small amount of guest bpac mole-
cules are included within the network; a proportion of the
iron centers in which the coordinated bpac is not involved
in p–p stacking interactions conserves an axial compression
consistent with the inactive iron(II) centers revealed by
magnetic and Mçssbauer measurements. Moreover, the im-
ꢀ
1
portant DEQ variation (2.75(6)–3.72(3) mms ) of the non-
active HS doublet between 317 and 80 K is characteristic of
a strongly distorted equatorial environment associated with
a small energy gap between the dxy orbital and the thermally
accessible dxz or dyz orbitals.
To extend this work, we have investigated the susceptibili-
ty of the framework to accept inclusions other than water or
bpac molecules and we have studied the synergy between
host–guest chemistry and spin-crossover behavior. Two strat-
egies have been developed, either by preparing crystals by
slow liquid–liquid diffusion in the presence of aromatic mol-
Inclusion of guest molecules: All these results make evident
the high sensitivity of the spin transition in this family of
complexes to the quantity and the type of guest molecules,
which has been specifically demonstrated by the effect of
water and bpac molecules. Similar chemoresponsive effects
have been observed in the corresponding nanoporous
[3a]
ecules that compete with the bpac molecules or by putting
powder samples synthesized following method B in contact
with pyrazine or pyridine vapors. Following the latter strat-
egy, further vapor adsorption experiments were performed
metal–organic framework {Fe(pz)[M(CN) ]} (M=Pt, Ni,
4
[
5]
Pd). Concerning the inclusion of water molecules, comple-
mentary thermogravimetric analysis has been performed to
prove the reversible adsorption of water in 1 (Figure 9). The
with powder samples {Fe ACHTUNGETRNNUNG( bpac)[Pt(CN) ]} prepared by
4
method B (samples 4 and 4’’). Sample 4 was treated over-
night by pyrazine vapor and the resulting 4(pz) sample was
Chem. Eur. J. 2012, 18, 507 – 516
ꢅ 2012 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chemeurj.org
513

A. Bousseksou, J. A. Real et al.
M
Figure 10. Temperature dependence of c T for 4’’, 4’’(py), and 4’’(pz) and the dehydrated forms 4’’d, 4’’d(py), and 4’’d(pz).
studied by Mçssbauer spectroscopy. Figure 8 shows the
Mçssbauer spectrum of 4(pz) acquired at 317 and 80 K and
values of the Mçssbauer parameters obtained by least-
squares fitting of the spectra are gathered in Table 1 and
compared with those of samples 1 and 4. At 317 K the spec-
trum can be satisfactorily fitted by considering three quadru-
pole-split doublets with isomer shift and quadrupole-split-
ting values attributable to iron(II) in the high-spin state. In
comparison with sample 4, the pyrazine vapor that enters
the network modifies the environment of the majority of the
iron(II) centers, so that the intensity of the doublet with the
large DEQ (site B, doublet in black) decreases drastically.
Conversely, the proportion of the doublet with the small
DEQ (site A) increases slightly (from 59 to 67%) and a new
tigation was performed following the exposure of sample
4(pz) to pyridine vapor, thereby leading to sample 4(pz-py).
The Mçssbauer spectrum of 4(pz-py) at 317 K (Figure 8) re-
ACHTUNGTRENNUNG
AHCTUNGTRENNUNG
veals, in comparison with sample 4(pz), the disappearance
of site B and the decrease of site C with a concomitant in-
crease of the proportion of active site A. In this case also, at
80 K, the whole fraction of site A (about 80%) was convert-
ed into iron(II) LS centers while the fraction of site B re-
mained in the HS state. Thermogravimetric analysis coupled
with mass spectroscopy (TGA-MS) measurement at m/z 78
makes evident that 0.75 pyridine molecules per iron atom
were present in compound 4’’(py) and the same measure-
ments at m/z 80 reveals that 0.45 pyrazine molecules per
iron atom was present in compound 4’’(pz) (Figure S5 in the
Supporting Information). Figure 10 shows the magnetic be-
havior of compound 4’’ before and after the adsorption of
pyridine (4’’(py)) or pyrazine (4’’(pz)) vapors in their hydrat-
ed and dehydrated forms. In agreement with the magnetic
result of sample 4, sample 4’’ exhibits a cooperative and in-
complete spin transition centered at 318 K (with a hysteresis
of 14 K). Significant change can be observed in sample
4’’(pz): the magnetic curve reveals a spin transition in two
steps (T =282 K and T =310 K) and the c T value of
ꢀ1
doublet with intermediate DE =2.0(1) mms (site C) ap-
Q
pears. At low temperature, a LS iron(II) doublet appears:
ꢀ
1
its parameters are d=0.452(7) mms
0
and DE =
Q
ꢀ
1
.21(2) mms and similarly to sample 4, only the HS dou-
blet with the smallest quadrupole splitting is affected by the
spin transition. In fact, in 4(pz), only a small amount of pyr-
azine entered the structure, as demonstrated by the ther-
mogravimetric analysis coupled with mass spectroscopy of
the analogous compound 4’’(pz) (see below). As a conse-
quence only a small proportion of the inactive sites B were
transformed into active sites A (8%). This probably occurs
because of p–p stacking interactions between the aromatic
rings of pyrazine molecules and those of the two bpac
ligand moieties around the considered iron atom. Moreover,
even if a small proportion (about 3%) is not affected by the
pyrazine inclusion, the remaining population is transformed
to site C (about 31%), which corresponds to inactive iron
atoms with certainly only one bpac ligand involved in p–p
stacking interactions. It is interesting to note the intermedi-
ate value of DEQ for site C versus those of sites A and B,
c1
c2
M
3
ꢀ1
1.25 cm Kmol at 80 K indicates a lower residual HS frac-
tion. The same effect but with amplification of the phenom-
enon has been measured for 4’’(py). In this case, the second
step is more complete with a transition temperature Tc2
shifted to 233 K. These results clearly demonstrate that the
origin of the two-step transition in these compounds is relat-
ed to the presence of two different iron centers with the
value of Tc2 depending on the vicinity of the iron centers
with either bpac or pyridine guest molecules. The more
complete nature of the spin transition for 4’’(py) versus
4’’(pz) is related to the larger fraction of included pyridine
molecule than that of pyrazine molecule. As already men-
tioned, the dehydration process tends to increase the size of
the hysteresis loop and decreases the transition temperature
without modifying the completeness of the transitions.
ꢀ
1
and its moderate thermal variation (2.0(1)–2.35(8) mms )
compared to site B. This corresponds to a rather well-isolat-
ed ground orbital singlet (weak mean axial compression)
with a weakly distorted equatorial environment associated
with a large energy gap between the dxy orbital and the ther-
mally accessible dxz or dyz orbitals. Further Mçssbauer inves-
514
www.chemeurj.org
ꢅ 2012 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Chem. Eur. J. 2012, 18, 507 – 516

Metal–Organic Frameworks
FULL PAPER
Conclusion
for {Fe
A
H
U
G
R
N
N
4
]}·H
2
O·0.40(bpac) (3): C 51.11, N 19.48, H 2.71;
found: C 51.10, N 18.10, H 2.26.
Synthesis of [Fe(bpac)Pt(CN) ]·H
thesis was performed under an argon atmosphere. A solution of Fe-
(BF ·6H O (0.222 mmol) and bpac (0.222 mmol) in methanol/water
1:1, 20 mL) was added dropwise to an aqueous solution of K [Pt(CN)
(0.222 mmol) leading to the formation of yellow-orange precipitates. The
precipitates were filtered off and washed with methanol and H O and
subsequently dried under vacuum. Yields: 75 and 66% for 4 and 4’, re-
spectively. Elemental analysis calcd (%) for {Fe-
(bpac)[Pt(CN) ]}·H O·0.09(bpac) (4): C 36.02, N 15.20, H 1.88, Pt 34.27,
Fe 9.81; found: C 36.12, N 14.52, H 1.54, Pt 33.38, Fe 9.62, Pt/Fe=0.993;
calcd (%) for {Fe(bpac)[Pt(CN) ]}·H O·0.34(bpac) (4’): C 39.25, N 15.23,
H 2.07, Pt 31.75, Fe 9.09; found: C 39.27, N 14.59, H 1.77, Pt 34.06, Fe
.94, Pt/Fe=0.980.
A
H
U
G
R
N
U
G
4
2
O·x ACHTUNGTENRNUG( bpac) (4)—method B: The syn-
We have investigated a series of 3D Hofmann-like clathrate
metal–organic frameworks exhibiting spin crossover at room
temperature with significant hysteresis loops (up to 68 K).
The main finding of this paper is that the physical and spec-
A
H
U
G
E
N
N
4
)
2
2
(
2
4
]
troscopic
ACHTUNGTRENNUNG( bpac)[M(CN) ]}·H O·x(bpac) (M=Pt, Pd, and Ni) allowed
characterization
of
{Fe-
2
4
2
us to observe how structural characteristics, Raman and
Mçssbauer spectral features, and the spin transition vary
with the sample stoichiometry. In samples with a low pro-
portion of included bpac molecules, the occurrence of differ-
ent iron centers leads to incomplete and less-cooperative
spin transitions. Conversely, when a large proportion of
bpac molecules are included, the p–p stacking interaction
between the bridging and included bpac molecules tends to
homogenize the network and complete and cooperative spin
transitions are observed. Another new feature reported here
A
H
U
G
R
N
U
G
4
2
A
H
U
G
R
N
U
G
4
2
9
Inclusion experiments: Vapor-adsorption experiments were carried out
on freshly prepared powder samples 4 and 4’’ (obtained using method B).
The powder was placed into a small, open vessel. This vessel was placed
in a larger sealed vessel containing pyrazine (ꢁ3 mg in the case of 4 and
4’’) or pyridine (approximately 3 mL in the case of sample 4’’). This setup
was gently heated and maintained at 358C for one night, causing the va-
porization and sublimation of pyridine and pyrazine, respectively. After
the adsorption experiments, samples 4(pz), 4 ACHUTNGRENNUG( pz-py), 4’’(pz), and 4’’(py)
were recovered and their properties were analyzed by Mçssbauer or
magnetic measurements.
5
7
is that the room temperature Fe Mçssbauer spectra give a
direct measure of the amount of spin-transition-active mate-
rial present in the sample. The part of the material that is
spin-transition active shows a doublet with a small quadru-
Microanalyses: Analysis for C, H, and N were performed after combus-
tion at 8508C using IR detection and gravimetry by means of a Perkin–
Elmer 2400 series II device. Analysis for Fe, Pt, and B were performed
by means of a Thermo Scientific iCAP 6000 Series ICP emission spec-
trometer.
ꢀ
1
II
pole splitting (ꢁ0.9 mms ) for HS Fe . Vapor-adsorption
experiments on powdered samples of these 3D compounds
revealed the susceptibility of the framework to accept guests
other than water or bpac molecules. The physical studies of
the modified networks showed the synergy between host–
guest chemistry and spin-crossover behavior. Indeed, the
clathration of various aromatic molecules revealed specific
modification of the transition temperature. These prelimina-
ry results confirm the high sensitivity of these highly porous
compounds towards gas and vapors, already demonstrated
Thermal analysis: Differential thermal analysis and thermogravimetric
(
DTA-TG) data were acquired simultaneously using a Perkin–Elmer Di-
amond thermal analyzer. Coupled mass spectrometry analyses of ele-
ments were realized with a Pfeiffer Vacuum Omnistar quadrupole mass
spectrometer.
X-ray powder diffraction: The powder X-ray diffraction patterns were
collected on a XPert Pro (Theta-Theta mode) Panalytical diffractometer
with l(CuKa1, CuKa2)=1.54059, 1.54439 ꢆ. Variable-temperature meas-
urements were performed on an Antoon Paar TTK 450 Chamber, from
room temperature to 400 K. The extraction of peak positions for indexing
was performed with the fitting program, available in the PC software
package Highscore+ supplied by Panalytical. Pattern indexing was car-
ried out by means of the program DICVOL implemented in the High-
score+ package. The cell refinement was also performed using High-
score+.
[5]
for the pyrazine derivative and open new opportunities for
the elaboration of gas-sensor technologies.
Experimental Section
Magnetic studies: The magnetic properties were measured at various
cooling and heating rates under a field of 1 T using a Quantum Design
MPMS superconducting quantum interference device magnetometer. The
experimental data were corrected for the diamagnetic contribution.
Materials:
and Fe(BF
sources and used as received. The synthesis of the bis(4-pyridyl)acetylene
bpac) ligand was carried out according to Ref. [3a].
Synthesis of [Fe(bpac)M(CN) ]·H O·x(bpac) (M=Pt (1), Pd (2), and Ni
3))—method A: The syntheses were performed under an argon atmos-
phere. A methanolic solution of bpac (0.222 mmol, 20 mL) was added to
K
2
[Pd(CN)
4 2 2 4 2 2 4 2
]·nH O, K [Pt(CN) ]·3H O, K [Ni(CN) ]·nH O,
A
C
H
T
U
N
G
T
R
E
N
N
U
N
G
4
)
2
·6H O salts and solvents were purchased from commercial
2
(
DSC analysis: DSC analysis was carried out on a Netsch DSC 204 instru-
A
C
H
T
U
N
G
T
R
E
N
N
U
N
G
4
2
A
T
N
R
N
N
3
ꢀ1
ment under helium purging gas (20 cm min ) at a heating/cooling rate
(
ꢀ1
of 10 Kmin . Temperature and enthalpy were calibrated using the melt-
ing transition of standard materials (Hg, In, Sn). The uncertainty in the
transition enthalpy (DHHL) and entropy (DSHL) is estimated to be ap-
proximately 10% due to the subtraction of the unknown baseline.
a
solution of Fe
A
C
H
T
U
N
G
T
R
E
N
N
U
N
G
(BF
4 2 2
) ·6H O in methanol/water (1:1) (0.222 mmol,
2
0 mL). K [M(CN)
2
4
] (0.222 mmol) dissolved in water (20 mL) was added
dropwise to the resulting yellow solution leading instantaneously to the
formation of yellow-orange precipitates. The precipitates were filtered
Raman spectroscopy: Raman spectra were collected between 300 and
8
0 K using a LabRAM-HR (Jobin–Yvon) Raman micro-spectrometer
off and washed with methanol and H
vacuum. Yields: 51, 59, 47, 50, and 64% for 1, 1’, 1’’, 2 and 3, respectively.
Elemental analyses calcd (%) for {Fe(bpac)[Pt(CN) ]}·H O·0.71(bpac)
1): C 43.22, N 15.26, H 2.30, Pt 28.66, Fe 8.20; found: C 43.22, N 14.71,
2
O and subsequently dried under
and a Linkam THMS-600 cryostage. The 632.8 nm line of a He–Ne laser
was used as the excitation source and a spectral resolution of approxi-
A
H
U
G
R
N
N
4
2
ꢀ
1
mately 3 cm was obtained.
(
5
7
H
2.03, Pt 27.95, Fe 8.11, Pt/Fe=0.986; calcd (%) for {Fe-
]}·H O·0.61(bpac) (1’): C 42.28, N 15.25, H 2.25, Pt 29.39,
Fe 8.41; found: C 42.28, N 14.54, H 1.73, Pt 28.79, Fe 8.45, Pt/Fe=0.976;
calcd (%) for {Fe(bpac)[Pt(CN) ]}·H O·0.87(bpac) (1’’): C 44.72, N 15.27,
H 2.39, Pt 27.49, Fe 7.87; found: C 44.72, N 15.32, H 2.09, Pt 27.92, Fe
.89, Pt/Fe=1.013; calcd (%) for {Fe(bpac)[Pd(CN) ]}·H O·0.78(bpac)
2): C 50.45, N 17.51, H 2.67; found: C 50.43, N 16.83, H 2.16; calcd (%)
Mçssbauer spectroscopy: Fe Mçssbauer spectra have been recorded
A
C
H
T
U
N
G
T
R
E
N
N
U
N
G
(bpac)[Pt(CN)
4
2
using a conventional constant-acceleration-type spectrometer equipped
5
7
with a 50 mCi Co source and a flow-type, liquid-helium cryostat. Spec-
tra of the powder samples (ca. 30 mg) were recorded between 5 and
300 K. Least-squares fittings of the Mçssbauer spectra have been carried
out with the assumption of Lorentzian line shapes using the Recoil soft-
ware package.
A
C
H
T
U
N
G
T
R
E
N
N
U
N
G
4
2
7
A
H
U
G
R
N
U
G
4
2
(
Chem. Eur. J. 2012, 18, 507 – 516
ꢅ 2012 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chemeurj.org
515

A. Bousseksou, J. A. Real et al.
Acknowledgements
Angew. Chem. 2009, 121, 9106–9109; Angew. Chem. Int. Ed. 2009,
4
8, 8944–8947; c) P. D. Southon, L. Liu, E. A. Fellows, D. J. Price,
G. J. Halder, K. W. Chapman, B. Moubaraki, K. S. Murray, J. F.
Lꢁtard, C. J. Kepert, J. Am. Chem. Soc. 2009, 131, 10998–11009.
6] I. Boldog, A. B. Gaspar, V. Martꢂnez, P. Pardo-IbꢀÇez, V. Ksenofon-
tov, A. Bhattacharjee, P. Gꢈtlich, J. A. Real, Angew. Chem. 2008,
This work was supported by the CHEMOSWITCH project (ANR 2010-
BLAN-1018 01), the BISTABLE project (University Paul Sabatier
PRES), the Spanish Ministerio de Ciencia e Innovaciꢃn (MICINN), and
FEDER funds (CTQ2010-18414) and the Generalitat Valenciana (GVA-
COMP2010-139). N.A.O.V. thanks the Instituto de Ciencia y Tecnologꢂa
del DF (ICyTDF) in Mexico for a postdoctoral fellowship and C.B.M.
thanks the French Ambassy in Spain and the Caixa for a doctoral fellow-
ship.
[
[
1
20, 6533–6537; Angew. Chem. Int. Ed. 2008, 47, 6433–6437.
7] a) F. Volatron, L. Catala, E. Rivieꢉre, A. Gloter, O. Steꢇphan, T.
Mallah, Inorg. Chem. 2008, 47, 6584–6586; b) J. Larionova, L.
Salmon, Y. Guari, A. Tokarev, K. Molvinger, G. Molnꢀr, A. Bous-
seksou, Angew. Chem. 2008, 120, 8360–8364; Angew. Chem. Int. Ed.
2
008, 47, 8236–8240.
8] a) S. Cobo, G. Molnꢀr, J. A. Real, A. Bousseksou, Angew. Chem.
006, 118, 5918–5921; Angew. Chem. Int. Ed. 2006, 45, 5786–5789;
b) G. Molnꢀr, S. Cobo, J. A. Real, F. Carcenac, E. Daran, C. Vieu,
A. Bousseksou, Adv. Mater. 2007, 19, 2163–2167; c) C. Bartual-
Murgui, L. Salmon, A. Akou, C. Thibault, G. Molnꢀr, T. Mahfoud,
Z. Sekkat, J. A. Real, A. Bousseksou, New J. Chem. 2011, 35, 2089–
[
[
1] a) D. Zacher, R. Schmid, C. Wçll, R. A. Fischer, Angew. Chem.
011, 123, 184–208; Angew. Chem. Int. Ed. 2011, 50, 176–199; b) C.
Janiak, J. K. Vieth, New J. Chem. 2010, 34, 2366–2388; c) Special
issue: Metal–Organic Frameworks, J. Long, O. Yaghi, Chem. Soc.
Rev. 2009, 38, Issue 5; d) D. Maspoch, D. Ruiz-Molina, J. Veciana,
Chem. Soc. Rev. 2007, 36, 770–818.
2
2
2
094.
[
2] a) V. Niel, J. M. Martinez-Agudo, M. C. MuÇoz, A. B. Gaspar, J. A.
Real, Inorg. Chem. 2001, 40, 3838–3839; b) G. J. Halder, C. J.
Kepert, B. Moubaraki, K. S. Murray, J. D. Cashion, Science 2002,
[
9] G. Agustꢂ, S. Cobo, A. B. Gaspar, G. Molnꢀr, N. O. Moussa, P. ꢊ.
Szilꢀgyi, V. Pꢀlfi, C. Vieu, M. C. MuÇoz, J. A. Real, A. Bousseksou,
Chem. Mater. 2008, 20, 6721–6732.
2
98, 1762–1765; c) A. Bousseksou G. Molnꢀr, L. Salmon, W. Nico-
[
[
10] H. J. Shepherd, C. Bartual-Murgui, G. Molnꢀr, J. A. Real, M. C.
lazzi, Chem. Soc. Rev. 2011, 40, 3313–3335; d) M. C. MuÇoz, J. A.
Real, Coord. Chem. Rev. 2011, 2068–2095; e) J. A. Real, E. Andrꢁs,
M C. MuÇoz, M. Julve, T. Granier, A. Bousseksou, F. Varret, Sci-
ence, 1995, 268, 265–267.
3] a) C. Bartual-Murgui, N. Ortega-Villar, H. J. Shepherd, M. C.
MuÇoz, L. Salmon, G. Molnꢀr, A. Bousseksou, J. A. Real, J. Mater.
Chem. 2011, 21, 7217–7222; b) T. Tayagaki, A. Galet, G. Molnꢀr,
M. C. MuÇoz, A. Zwick, K. Tanaka, J. A. Real, A. Bousseksou, J.
Phys. Chem. B 2005, 109, 14859–14867.
4] a) S. Bonhommeau, G. Molnꢀr, A. Galet, A. Zwick, J. A. Real, J. J.
McGarvey, A. Bousseksou, Angew. Chem. 2005, 117, 4137–4141;
Angew. Chem. Int. Ed. 2005, 44, 4069–4072; b) S. Cobo, D. Ostrov-
skii, S. Bonhommeau, L. Vendier, G. Molnaꢇr, L. Salmon, K.
Tanaka, A. Bousseksou, J. Am. Chem. Soc. 2008, 130, 9019–9024.
5] a) M. Ohba, K. Yoneda, G. Agustꢂ, M. C. MuÇoz, A. B. Gaspar,
J. A. Real, M. Yamasaki, H. Ando, Y. Nakao, S. Sakaki, S. Kitagawa,
Angew. Chem. 2009, 121, 4861–4865; Angew. Chem. Int. Ed. 2009,
MuÇoz, L. Salmon, A. Bousseksou, New J. Chem. 2011, 35, 1205–
1
210.
11] a) K. Kondo, N. Ohnishi, K. Takemoto, H. Yoshida, K. Yoshida, J.
Org. Chem. 1992, 57, 1622–1625; b) H. L. Anderson, C. J. Walter,
A. Vidal-Ferran, R. A. Hay, P. A. Lowden, J. K. M. Sanders, J.
Chem. Soc. Perkin Trans. 1 (1972–1999) 1995, 2275–2280; c) N. R.
Champness, A. N. Khlobystov, N. Andrei, A. G. Majuga, M.
Schroeder, N. V. Zyk, Tetrahedron Lett. 1999, 40, 5413–5416; d) B. J.
Coe, J. L. Harries, J. A. Harris, B. S. Brunschwig, S. J. Coles, M. E.
Light, M. B. Hursthouse, Dalton Trans. 2004, 2935–2942.
12] N. Inoue, O. Sugimoto, K. Tanji, Heterocycles 2007, 72, 665–671.
13] a) J. S. O. Evans, J. Chem. Soc. Dalton Trans. 1999, 3317–3026; b) T.
Pretsch, K. W. Chapman, G. J. Halder, C. J. Kepert, Chem.
Commun. 2006, 1857–1859.
[
[
[
[
[
[
14] P. Guionneau, M. Marchivie, G. Bravic, J.-F. Lꢁtard, D. Chasseau,
Top. Curr. Chem. 2004, 234, 97–128.
4
8, 4767–4771; b) G. Agustꢂ, R. Ohtani, K. Yoneda, A. B. Gaspar,
Received: July 29, 2011
M. Ohba, J. F. Sꢀnchez-Royo, M. C. MuÇoz, S. Kitagawa, J. A. Real,
Published online: December 6, 2011
516
www.chemeurj.org
ꢅ 2012 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Chem. Eur. J. 2012, 18, 507 – 516