Three-way crystal-to-crystal reversible transformation and controlled spin switching by a nonporous molecular material

Article abstract of DOI:10.1021/ja411595y

Porous materials capable of hosting external molecules are paramount in basic and applied research. Nonporous materials able to incorporate molecules via internal lattice reorganization are however extremely rare since their structural integrity usually does not resist the guest exchange processes. The novel heteroleptic low-spin Fe(II) complex [Fe(bpp)(H₂L)](ClO ₄)₂·1.5C₃H₆O (1; bpp = 2,6-bis(pyrazol-3-yl)pyridine, H₂L = 2,6-bis(5-(2-methoxyphenyl) pyrazol-3-yl)pyridine) crystallizes as a compact discrete, nonporous material hosting solvate molecules of acetone. The system is able to extrude one-third of these molecules to lead to [Fe(bpp)(H₂L)](ClO₄) ₂·C₃H₆O (2), switching to the high-spin state while experiencing a profound crystallographic change. Compound 2 can be reversed to the original material upon reabsorption of acetone. Single crystal X-ray diffraction experiments on the latter system (1′) and on 2 show that these are reversible single-crystal-to-single-crystal (SCSC) transformations. Likewise, complex 2 can replace acetone by MeOH and H₂O to form [Fe(bpp)(H₂L)](ClO₄)₂·1.25MeOH·0. 5H₂O (3) through a SCSC process that also implies a switch to the spin state. The 3→1 transformation through acetone reabsorption is also demonstrated. Besides the spin switching at room temperature, this series of SCSC transformations causes macroscopic changes in color that can be followed by the naked eye. The reversible exchanges of chemicals are therefore easily sensed at the temperature at which these occur, contrary to what is the case for most of the few existing nonporous spin-based sensors, which feature a large temperature gap between the process monitored and the mechanism of detection.

Full text of DOI:10.1021/ja411595y

Article
pubs.acs.org/JACS
Three-Way Crystal-to-Crystal Reversible Transformation and
Controlled Spin Switching by a Nonporous Molecular Material
Jose Sanchez Costa,*,‡ Santiago Rodríguez-Jimenez,‡ Gavin A. Craig,‡ Benjamin Barth,‡
́
́
́
Christine M. Beavers,§ Simon J. Teat,§ and Guillem Aromí*,‡
‡Departament de Química Inorgan
̀
ica, Universitat de Barcelona, Diagonal, 645, 08028 Barcelona, Spain
§Advanced Light Source, Berkeley Laboratory, 1 Cyclotron Road, Berkeley, California 94720, United States
S
* Supporting Information
ABSTRACT: Porous materials capable of hosting external
molecules are paramount in basic and applied research.
Nonporous materials able to incorporate molecules via internal
lattice reorganization are however extremely rare since their
structural integrity usually does not resist the guest exchange
processes. The novel heteroleptic low-spin Fe(II) complex
[Fe(bpp)(H2L)](ClO4)2·1.5C3H6O (1; bpp = 2,6-bis(pyrazol-
3-yl)pyridine, H2L = 2,6-bis(5-(2-methoxyphenyl)pyrazol-3-
yl)pyridine) crystallizes as a compact discrete, nonporous
material hosting solvate molecules of acetone. The system is
able to extrude one-third of these molecules to lead to [Fe(bpp)(H2L)](ClO4)2·C3H6O (2), switching to the high-spin state
while experiencing a profound crystallographic change. Compound 2 can be reversed to the original material upon reabsorption
of acetone. Single crystal X-ray diffraction experiments on the latter system (1′) and on 2 show that these are reversible single-
crystal-to-single-crystal (SCSC) transformations. Likewise, complex 2 can replace acetone by MeOH and H2O to form
[Fe(bpp)(H2L)](ClO4)2·1.25MeOH·0.5H2O (3) through a SCSC process that also implies a switch to the spin state. The 3→1
transformation through acetone reabsorption is also demonstrated. Besides the spin switching at room temperature, this series of
SCSC transformations causes macroscopic changes in color that can be followed by the naked eye. The reversible exchanges of
chemicals are therefore easily sensed at the temperature at which these occur, contrary to what is the case for most of the few
existing nonporous spin-based sensors, which feature a large temperature gap between the process monitored and the mechanism
of detection.
resulting in alterations to some physical properties. Good
candidates for exploiting these effects are spin crossover (SCO)
complexes.13 In fact, SCO complexes present the ability to
switch between high-spin (HS) and low-spin (LS) electronic
states in response to external stimuli, such as temperature or
pressure variations, light irradiation, or the inclusion of analytes.
In particular, some systems are known to be very sensitive to
solvent exchange through the molecular network.14 The
transition implies distinctly different magnetic, optical,
electrical, and structural outputs.15 Thus, in a limited number
of PCPs,16 mainly the Hoffmann Clathrate-type systems,
indirect analyte−metal interactions can be enough to afford a
spin-state change of Fe(II) centers. On the other hand, to the
best of our knowledge there is only one well characterized
Fe(II) discrete molecular material system that shows
vapochromism as a result of SCO, while maintaining the
integrity of its crystal lattice.17,18
INTRODUCTION
■
The development of chemical sensors capable of detecting
small volatile organic compounds and toxic gases by simple
means is a critical task for environmental and energy science.1 A
principle for the design of chemosensors is that they must be
able to detect substances and produce an easily measurable
response, such as a change of color, of fluorescence, or of any
other macroscopic property.2 One approach to achieve this
kind of material is through the construction of porous
coordination polymers (PCPs)3 (or metal organic frameworks,
MOFs),4 which by their nature contain the voids necessary to
absorb guest molecules, thereby changing their properties.5 In
the absence of pores, such exchange might still take place
through a process of diffusion throughout the crystal lattice.6−8
These phenomena are, however, extremely rare in crystalline
molecular materials9 since the absence of long-range covalent
interactions usually leads to the collapse of the lattice. In some
of the documented examples, the exchangeable solvents can
directly coordinate to the metal center,10−12 causing significant
changes to the d−d absorptions. Otherwise, the guests
intercalate within the crystal lattice, influencing the supra-
molecular interactions within the material. Ideally, in the latter
case the metal center may be sensitive to subtle changes thus
We have been engaged in the design of new molecular SCO
materials through the synthesis and use of derivatives of the
ligand 2,6-bis-(pyrazol-3-yl)pyridine (bpp, Figure 1),19 as we
Received: November 19, 2013
Published: February 20, 2014
© 2014 American Chemical Society
3869
dx.doi.org/10.1021/ja411595y | J. Am. Chem. Soc. 2014, 136, 3869−3874

Journal of the American Chemical Society
Article
Table 1. Crystallographic Data of 1−3
1
2
3
T (K)
250
monoclinic
P21/n
250
100
triclinic
P-1
cryst. system
space group
a (Å)
triclinic
P-1
14.0569(17)
21.606(3)
15.2698(19)
90
13.099(5)
13.355(5)
14.078(5)
71.968(4)
69.923(4)
70.857(4)
2130.8(2)
12.474
12.708(4)
13.214(5)
14.931(5)
71.792 (4)
69.208 (4)
62.660 (4)
2048.2(12)
9.507
b (Å)
c (Å)
α (°)
β (°)
γ (°)
V (Å3)
Oct. volume (Å3)
R1 (all data)
wR2 (all)
av. Fe−N (Å)
solv. accessible voids (Å3)
114.946
90
4205.0(9)
9.515
Figure 1. Ligands bpp and H2L (top) and representation of the cation
[Fe(bpp)(H2L)]2+ of 1. Orange, Fe; red, O; purple, N; gray, C.
Hydrogen atoms not shown.
0.0512
0.1372
1.949
0.1055
0.0781
0.3050
0.2357
2.169
1.947
have recently reported.20−22 The ligand bpp gives rise to a
family of coordination complexes, many exhibiting SCO, with
general formula [Fe(bpp)2]A·xH2O (A = anions), which often
are extremely sensitive to their degree of hydration.23 For
example, the LS, trihydrated phase of the triflate salt becomes a
SCO system upon desorption of two molecules of water,
displaying a huge hysteresis loop (up to 140 K wide).24 The
tetrafluoroborate derivative of a [Fe(Me-bpp)2)]2+ cation (Me-
bpp, a methyl-substituted bpp) exhibits up to five different
crystallographic phases with varying degrees of hydration and
SCO responses.25 The rationalization of the exact influence of
water in the spin active species in this and other related systems
is however hampered by the lack of structural information on
the pertinent individual phases.26
Attaching hydroxyphenyl moieties at the ends of bpp
enhances the intermolecular interactions within the lattice of
the corresponding Fe(II) complexes, yielding a variety of
responses as a function of the solvent of crystallization
present.20−22 We were thus intrigued by the possibility of
combining these properties with the capabilities for exchanging
guest molecules demonstrated by complexes of simpler bpp.
With this in mind, we prepared a novel bpp-based ligand, 2,6-
bis(5-(2-methoxyphenyl)pyrazol-3-yl)pyridine (H2L, Figure 1),
and combining it with bpp, we managed to produce the
heteroleptic coordination complex [Fe(bpp)(H2L)](ClO4)2·
1.5C3H6O (1). This compound has turned out to exhibit a
remarkable three-way single-crystal-to-single-crystal (SCSC)
transformation process involving the reversible exchange of
small volatile molecules at room temperature. These trans-
formations are coupled to dramatic optical, magnetic, and
crystallographic changes, thus converting this nonporous
material into a robust room-temperature switch, sensitive to
methanol and acetone.
0
34
0
(Table S1), indicating that the Fe(II) centers are in the LS
state.22
The complexes of 1 organize in sheets according to the so-
called terpyridine embrace,27 where each cation interacts with
its four first neighbors via π···π contacts reinforced by C−H···π
interactions (Figure S2). The differing dimensions of both
ligands allow interactions with second neighbors that otherwise
would not be possible. This is believed to be one of the driving
forces leading to the heteroleptic compound over any of the
two possible homoleptic species (not observed). The sheets are
then connected to each other through additional π···π contacts,
−
together with hydrogen-bonding motifs involving the ClO4
groups and the N−H groups of the cations (Figure S1 and
Table S2 and S3). The solvents of crystallization are located in
between the sheets. Of these, the full molecule of acetone is
well ordered, forming an H-bond with the N−H moiety of bpp,
whereas the molecule with half occupancy is disordered over
two positions and only loosely bound. It must be noted that the
lattice exhibits no significant pores or channels (Figure S3).
Conversion between 1 and [Fe(bpp)(H2L)](ClO4)2·
C3H6O (2). To test for the desorption of acetone and the
potential consequent physical changes, a single crystal of 1 was
heated to 375 K and kept at this temperature for 2 h under a
dry N2 flow. During the course of this process, the crystal
changed color from dark-red to orange, while maintaining its
structural integrity. After cooling to 250 K, a single crystal X-ray
diffraction experiment was performed to determine the
structure. The new system was found now in the triclinic P-1
space group. The change of space groups was accompanied by a
50% reduction of the cell volume and the disappearance of the
disordered molecules of acetone. The new material thus
exhibits the composition [Fe(bpp)(H2L)](ClO4)2·C3H6O (2,
Figure S4). Other remarkable transformations unveiled by the
new structure are the simultaneous rotation by ∼180° of both
methoxyphenyl groups of H2L in every other cation (thus
maintaining the syn,anti arrangement, Figure 2) together with a
change to the average Fe−N bond distances, to 2.169 Å, which
are the values expected for Fe(II) centers in the HS state22
(Table S4 and Figure S4). The new compound exhibits an
analogous network of intermolecular interactions as its
precursor, while presenting a nonporous character (Figure S5
and Table 1). Thus, a drastic crystallographic phase transition,
together with a dramatic change in optical properties (see SI)
and spin state, has occurred after the desorption of acetone
RESULTS AND DISCUSSION
■
Description of [Fe(bpp)(H2L)](ClO4)2·1.5C3H6O (1). Dark-
red single crystals of 1 were obtained from the reaction of bpp,
H2L, and Fe(ClO4)2 in acetone. The complex crystallizes in the
monoclinic space group P21/n (Table 1), with an asymmetric
unit comprising one [Fe(bpp)(H2L)]2+ cation (Figure 1),
together with two perchlorate anions and a total of 1.5
molecules of acetone (Figure S1). Both bpp-type ligands
coordinate the metal in a mer fashion through their three N-
donor atoms, providing the metal with a distorted octahedral
geometry. The ligand H2L is in a syn,anti configuration of the
methoxyphenyl rings with respect to the central coordination
pocket. The average Fe−N bond length at 250 K is 1.949 Å
3870
dx.doi.org/10.1021/ja411595y | J. Am. Chem. Soc. 2014, 136, 3869−3874

Journal of the American Chemical Society
Article
with all the changes in physicochemical properties could be
reversed entirely without causing damage to the crystal lattice.
This reversible process was monitored on single crystals by
optical reflectivity (Figure S7) and optical microscopy (SI,
movie 2) and in the bulk by thermogravimetric analysis (TGA)
experiments and magnetometric measurements (see below and
SI).
Absorption/Desorption of [MeOH + H2O] Involving
[Fe(bpp)(H2L)](ClO4)2·1.25MeOH·0.5H2O (3). The capacity
of 2 to incorporate acetone vapors into the structure
concomitant to drastic changes in properties stimulated
attempts to detect in a similar manner other small volatile
molecules, such as MeOH. Thus, exposure of 2 to saturated
vapors of MeOH at room temperature for 24 h caused the
recovery of a dark-red color with persistence of the crystallinity.
A single crystal X-ray diffraction study of this material was
possible and revealed a new formulation for the novel system,
[Fe(bpp)(H2L)](ClO4)2·1.25MeOH·0.5H2O (3), as a result of
the complete exchange28 of the acetone molecules by methanol
and atmospheric water (Figure S8). This process does not lead
to a change in space group, which remains as triclinic P-1, while
the unit cell volume is approximately maintained (Table 1).
The Fe(II) cations are organized in the lattice as with the
acetone precursor (Figure S9), now with the methanol
molecules playing an analogous role; one forms a hydrogen
bond with one N−H group from bpp, while the other one, with
25% occupancy, is disordered over two positions and loosely
bound. In addition, a molecule of water with 50% occupancy
forms an H-bond with the ordered MeOH molecule and with a
ClO4− group (Figure S8). Quite remarkably, the average Fe−N
bond distance at 100 K is now 1.947 Å, which means that the
replacement of acetone by MeOH and H2O also causes a
change in magnetic state to LS (Table S7, see below). As with
the acetone solvates, there are no pores inside the crystal lattice
(see Table 1 and SI). This process could also be monitored
through optical reflectivity experiments (Figure S7) and was
Figure 2. Representation of the SCSC reversible transformation
between 1 and 2, emphasizing the ∼180° rotation of the
methoxyphenyl groups from every other complex (only the [Fe(bpp)-
(H2L)]2+ cations are represented).
(one-half molecule per complex cation) in a process that must
occur by diffusion through the nonporous lattice, in a manner
that permits the persistence of the crystallographic order
throughout the bulk of the single crystal (Figure 3). This
process was monitored by optical reflectivity (Figure S6) and
optical microscopy on a single crystal (see SI, movie 1).
Compound 2 could be obtained intact also by heating 1 for
2h in an oven at 390 K, as determined also by single crystal X-
ray diffraction. The reversibility of the process was demon-
strated by exposing single crystals of 2 to a room-temperature
saturated atmosphere of acetone for 24 h. After the exposure,
the dark-red color was recovered, while the quality of the
resulting crystals allowed a new X-ray diffraction determination.
The crystal structure (1′) was found to be virtually identical to
that of the initial phase 1 (Tables 1, S1 and S2), thus
confirming that upon reabsorption of acetone, the entire
crystallographic transition previously demonstrated together
Figure 3. Representation of the molecular structure of 1−3 (only major component of disordered species shown) emphasizing the full conversion
cycle undergone by these three species through subsequent absorption, desorption, or exchange of small molecules. Among the represented
processes are the three-way proven crystal-to-crystal transformations discussed in the text (see SI). Pictures of the corresponding single crystals are
also shown.
3871
dx.doi.org/10.1021/ja411595y | J. Am. Chem. Soc. 2014, 136, 3869−3874

Journal of the American Chemical Society
Article
recorded on a single crystal by optical microscopy experiments
(SI, movie 3).
The effect of the vapors of acetone on the hydro-
methanolated material 3 was also studied. Thus, exposure of
these crystals to a saturated atmosphere of acetone at room
temperature for 24h caused the replacement of the methanol
and water molecules again by acetone molecules to return to
the original compound 1 (Figure 3). The identity of this
material was consistent with microanalysis, TGA analysis, and
magnetic studies (see below and SI). These experiments show
that the lattice of [Fe(bpp)(H2L)](ClO4)2 solvated with
acetone not only reversibly evacuates acetone but also can be
cycled through a process of desorption of acetone, absorption
of methanol and water, followed by full replacement of these
solvents by acetone (Figure 3) back to the original compound.
The cycle could be achieved with the same crystals more than
once as proved again by TGA and magnetic measurements (see
below and SI).
Magnetic Switching Behavior. The changes in spin state
observed for the various transformations described above
provide this material with potential as a spin-based sensor. This
was further investigated through variable-temperature magnetic
susceptibility measurements. Thus the bulk magnetization of a
polycrystalline sample of 1 was measured in the 2−375 K
temperature range under a constant magnetic field of 0.5 T.
The χMT vs T curve (χM is the molar paramagnetic
susceptibility, Figure 4, top) shows that the compound is
diamagnetic (LS) until nearly 300 K, consistent with the
observations from single crystal X-ray diffraction. From that
temperature, the χMT product starts to increase slightly as a
result of a gradual change of Fe(II) from LS to HS. When the
temperature reached 375 K, the heating was stopped, while the
χMT product continued to rise. After 2 h at this temperature,
the magnetic response reached a saturation value close to 3.25
cm3 K mol−1, which is slightly above the figure expected (3.00
cm3 K mol−1) for Fe(II) ions with an S = 2 spin state (HS) and
g = 2. On cooling down the sample back to 2 K (Figure S10),
the χMT value remains practically constant down to near 240 K,
where it experiences an abrupt decrease to a value of 0.3 cm3 K
mol−1. This is consistent with a process of SCO with T1/2↓ of
235 K. After bringing the system to 6 K, the temperature was
raised again to 300 K. At 236 K the product χMT increases fast
up to a value of 2.8 cm3 K mol−1 at 265 K, corresponding to
Fe(II) in the HS state (T1/2↑ = 240 K). Compound 2 is thus a
SCO system featuring a small hysteresis loop 5 K wide. The X-
ray diffraction experiments of 2 at 250 K indeed confirm it to
be in the HS state. The reversibility of the process could also be
monitored by optical reflectivity, which was conducted over at
least three full cycles (Figure S11).
A sample of 2 exposed to saturated acetone at room
temperature for 24 h was measured in the magnetometer in the
same conditions as above proving that it became diamagnetic
again, consistent with the process of reabsorption of acetone to
reform 1, as unveiled by single crystal X-ray crystallography.
These observations represent a direct proof of the spin
switching processes accompanying the absorption or desorp-
tion of acetone molecules by lattices of [Fe(bpp)(H2L)]-
(ClO4)2. The reversibility and robustness of the process could
be monitored by measuring the magnetic response after
successive steps of absorption and desorption of acetone
(Figure S12), and it was verified by measuring the TGA of each
recovered phase of 1 (Figure S13). The absorption of methanol
by 2 was also monitored through magnetic measurements.
Figure 4. (top) Plots of χMT vs T for 1 upon warming and
concomitant desorption of acetone (black circles), for resulting
compound 2 upon cooling (red circles) and for 3 (blue triangles,
formed from 2, after substitution of acetone by MeOH + H2O) upon
warming from room temperature and concomitant depletion of
MeOH + H2O. The crystal-to-crystal transformations described in the
text (1→2, 2→1, and 2→3) are shown with vertical arrows positioned,
for clarity, at arbitrary temperatures. (bottom) TGA graphs for the
desorption of acetone (from 1, black symbols) and MeOH + H2O
(from 2, blue symbols), emphasizing the difference in temperatures for
these two processes, consistent with the observations from magnetic
measurements.
Thus a polycrystalline sample of this system exposed to
methanol vapors as described above was measured through
variable-temperature magnetization at 0.5 T, from 300 to 375
K.
The χMT vs T curve (Figure 4, top) confirms the fact that the
compound 3 is diamagnetic. Upon warming, the value of χMT
slowly increases until the temperature reaches 350 K, where the
rise of magnetic response becomes more abrupt, attaining a
value of 3.00 cm3 K mol−1 at 375 K. This shows that the system
progressively changes from LS to HS until the transition
becomes complete at the highest measured temperature. This
process seems analogous to that observed for 1 and is also due
to a process of desolvation. After this process takes place, the
compound exhibits a different magnetic behavior (Figure S14).
In fact, if the temperature of 375 K is maintained, the χMT
product gradually diminishes down to an equilibrium value of
1.9 cm3 K mol−1, which corresponds to having approximately
two-thirds of the spin centers in the HS state. This is either the
result of a process of relaxation from a metastable state to an
equilibrium situation or the final step of a complete solvent
depletion. Upon cooling from the equilibrium, the χMT value
decreases to vanish almost completely near 200 K, in a process
3872
dx.doi.org/10.1021/ja411595y | J. Am. Chem. Soc. 2014, 136, 3869−3874

Journal of the American Chemical Society
Article
completed, the reaction mixture was heated to reflux, turning from
yellow to orange after a few hours, and left overnight. The mixture was
then left to cool to room temperature, and some drops of EtOH were
added in order to quench any remaining NaH, followed by careful
addition of 100 mL of water. Acetic acid (20 mL) was then added,
which led to a change of color to orange and a change of texture while
a precipitate formed. After adding about 75% of the acetic acid, the
mixture turned suddenly dark brown, and the precipitate dissolved.
The resulting two layers were separated and the organic phase was
extracted with diethyl ether. The combined organic phases were dried
over MgSO4, and the solvent removed in vacuum. The product is a
yellow solid. The yield was 3.91 g (50%). 1H NMR (400 MHz, ppm in
CDCl3): 16.3 (s, 2H), 8.3 (d, 2H), 8.06 (t, 1H), 7.96 (dd, 2H), 7.72
(s, 2H), 7.48 (td, 2H), 7.06 (dd, 2H), 6.95 (dt, 2H), 3.8 (s, 6H). IR
(KBr pellet)/cm−1: 3432 w, 2923 w, 2847 w, 1600 s, 1489 s, 1464 m,
1284 m, 1243 s, 1180 m, 1020 w, 992 m, 814 m, 763 m, 615 m. Mass
(M + H)+ 432.2. EA, calcd (%) for C25H21NO6 (found): C, 69.6
(69.15); H, 4.91 (4.91); N, 3.25 (3.27).
2,6-bis(5-(2-methoxyphenyl)-pyrazol-3-yl)pyridine (H2L).
Solid H2L1 (2 g, 4.64 mmol) was refluxed overnight with hydrazine
monohydrate (0.44 g, 9.27 mmol) in MeOH (120 mL). After cooling
to room temperature, the off-white suspension was filtered and washed
with diethyl ether to afford a white precipitate which was dried in air.
The yield was 1.72g (87%). 1H NMR (400 MHz, ppm in DMSO-d6):
13.5 (broad s, 2H), 7.98 (broad t, 1H), 7.90 (broad m, 2H), 779
(broad m, 2H), 7.36 (m, 4H), 7.14 (dd, 2H), 7.04 (m, 2H), 3.93 (s,
6H). IR (KBr pellet)/cm−1: 3314 s, 3175 s, 2922 s, 1602 m, 1576 s,
1477 s, 1443 m, 1292 s, 1181 s, 1032 m, 1004 m, 958 m, 797 m, 738 s.
Mass (M + H)+ 424.4. EA, calcd (%) for C25H21N5O2 H2O (found):
C, 68.00 (67.60); H, 5.25 (5.23); N, 15.86 (15.42).
Coordination Complexes. [Fe(bpp)(H2L)](ClO4)2·1.5·C3H6O (1).
A suspension of H2L (0.026 g, 0.06 mmol) and bpp (0.013 g, 0.06
mmol) in acetone (10 mL) was added dropwise with stirring to a
solution of Fe(ClO4)2·6H2O (0.034 g, 0.13 mmol) and ascorbic acid
(∼3 mg) in acetone (10 mL). The resulting orange solution was
stirred for 45−60 min at room temperature, before being filtered and
layered with diethyl-ether (volume 1:1). Crystals suitable for X-ray
diffraction formed after 2 days (0.038 g, 65%). IR (KBr pellet)/cm−1:
3404 (m), 3145 (w), 2942 (w), 1615 (w), 1474 (m), 1259 (w), 1121
(s), 1112 (s), 1089 (s), 1015 (m), 775 (m), 625 (m). EA, calcd (%)
for C36H30Cl2FeN10O10·1.5C3H6O (found): C, 49.81 (49.30); H, 4.03
(4.04); N, 14.34 (14.45).
[Fe(bpp)(H2L)](ClO4)2·C3H6O (2). Crystals of compound 1 were
brought to 120 °C in an oven for 2 h, inducing a color change from
dark red to orange. The heating process caused the loss of 0.5 mol of
acetone per formula unit. IR (KBr pellet)/cm−1: 3408 (w), 2948 (w),
1615 (w), 1475 (m), 1255 (w), 1120 (s), 1108 (s), 1092 (s), 1080 (s),
1014 (m), 767 (m), 624 (m). EA, calcd (%) for C36H30Cl2FeN10O10·
0.96C3H6O·0.8H2O (found): C, 48.66 (48.17); H, 3.92 (3.56); N,
14.60 (15.1).
that corresponds to a SCO. On warming again, the χMT vs T
curve superimposes with the cooling branch up to 375 K (and
not with the curve originally seen for 3), which is consistent
with a process of SCO of the system in equilibrium (Figure
S14).
TGA experiments comparing the behavior of 1 and 3 suggest
that indeed both systems experience the expected solvent loss
following the increase in temperature; 0.5 mol of acetone in
one case and to 1 mol of MeOH plus 1 mol of water in the
other (Figures S15 and S16). The difference in desorption
temperatures (Figure 4, bottom) are in agreement with the
effects observed through magnetometry. This difference was
maintained when cycling the system between the three different
structurally characterized phases more than once.
The system depleted of MeOH/H2O by heating, remained
crystalline, however, it could not be characterized by single
crystal X-ray diffraction. Nevertheless, it was exposed to
saturated methanol vapors for 24 h. The magnetic measure-
ment of the resulting system recreated the behavior seen for 3
(Figure 4, top), suggesting that the hydro-methanolated
product was regenerated and that this process of solvent
depletion and reabsorption was also reversible. This was
corroborated through TGA measurements (Figure S17) by
repeating the transformation over two cycles. The fact that both
solvent systems studied (acetone and MeOH/H2O, respec-
tively) exhibit a distinctly different temperature of desorption
provides a fingerprint that enhances the interest of investigating
in the future the response of this lattice to other analytes.
Like the replacement of MeOH and H2O by acetone in the
transformation of 3 into 1 (Figures 2, S18 and S19), the
formation of 1, as obtained from heating 3 and exposing it to
acetone vapors, was also corroborated by EA, TGA and
magnetometry (Figures S20 and S21).
CONCLUSIONS AND OUTLOOK
■
In summary, the lattice of the heteroleptic molecular
coordination complex [Fe(bpp)(H2L)](ClO4)2 has proven to
exhibit remarkable small-molecule exchange abilities accom-
panied by drastic variations on the structure and physicochem-
ical properties. Surprisingly, the observed processes of
desorption/absorption or exchange are fully reversible with
persistence of the crystallinity, despite the fact that the lattice is
nonporous and nonpolymeric. This is the result of a subtle
balance between flexibility of end moieties of H2L in 1,
combined with a robust network of intermolecular interactions.
This opens up perspectives for developing this and related
compounds as potential sensors of small molecules in the gas
phase, by exploiting the robustness of the system and the
possibility of following this process through a variety of
common techniques. We are currently investigating these
promising avenues by exploring the absorption and exchange of
a large variety of other analytes.
[Fe(bpp)(H2L)](ClO4)2·1.25CH4O·0.5H2O (3). Crystals of 2 were
exposed to saturated methanol vapors during 24 h. During the process
the crystals changed colors from orange to dark red. IR (KBr pellet)/
cm−1: 3405 (w), 2928 (w), 1615 (w), 1476 (m), 1260 (w), 1122 (s),
1108 (s), 1100 (s), 1074 (s), 1014 (m), 772 (m), 623 (m). EA, calcd
(%) for C36H30Cl2FeN10O10·1.25CH4O·H2O (found): C, 47.22
(46.66); H, 3.94 (3.67); N, 14.78 (15.26).
X-ray Crystallography Data Collection. Data were collected at
different temperatures for 1−3 and 1′ using a Bruker APEX II CCD
diffractometer on station 11.3.1 of the Advanced Light Source at
Lawrence Berkeley National Laboratory, at λ = 0.7749 Å, from a
silicon 111 monochromator. The structures were solved by direct
methods. The integration of diffraction profiles and absorption
corrections were made with the SAINT30 and SADABS31 programs.
Materials for publication were prepared using SHELXTL,32
PLATON,33 Mercury 1.4,34 and Olex235 programs. Olex2 has been
used for calculating the solvent accessible voids.
EXPERIMENTAL SECTION
■
Synthesis of the Organic Ligands. Ligand 2,6-bis(pyrazol-3-
yl)pyridine (bpp) was synthesized as described in the literature.29
2,6-bis(3-oxo-3-(2-methoxyphenyl)-propionyl)pyridine
(H2L1). 2-Methoxyacetophenone (5.44 g, 36.2 mmol) and ethyl-2,6-
pyridinedicarboxylate (4.04 g, 18.1 mmol) were dissolved in
dimethoxyethane (DME, 200 mL) under a nitrogen atmosphere. A
60% oil dispersion of NaH (8.35 g, 208.8 mmol) was washed for 20
min under N2 with hexane. The solvent was extracted using a filter
cannula, and DME (50 mL) was added. This suspension was added
dropwise with stirring to the above mixture. After the addition was
Physical Measurements. Variable-temperature magnetic suscept-
ibility data were obtained with a Quantum Design MPMS-XL SQUID
magnetometer at the “Unitat de Mesures Magnetiques” of the
̀
3873
dx.doi.org/10.1021/ja411595y | J. Am. Chem. Soc. 2014, 136, 3869−3874

Journal of the American Chemical Society
Article
Universitat de Barcelona. Pascal’s constants were used to estimate
diamagnetic corrections to the molar susceptibility, and a correction
was applied for the sample holder. The optical reflectivity variation was
gauged by optical microscopy, using a MOTIC SMZ-140 optical
stereoscope equipped with a charge-coupled device (CCD) camera
MOTICAM1000 and a white halogen lamp as a light source. Images
were collected in RGB format without any filter with the mean value
from each region of interest (ROI) analyzed under the ImageJ
environment. The temperature was controlled by using a Linkam
THMS600 liquid-nitrogen cryostat. Movies 2 and 3 were recorded at
room temperature with a double vessel assembly; an external vessel
containing the solvents and a small internal pot holding the sample,
allowing filming the color change with the CCD camera. IR spectra
were recorded on KBr pellets, in the range 4000−400 cm−1, with a
Thermo Nicolet Avatar 330 FT-IR spectrometer. Elemental analyses
were performed with a Perkin-Elmer Series II CHNS/O Analyzer
(12) Albrecht, M.; Lutz, M.; Spek, A. L.; van Koten, G. Nature 2000,
406, 970−974.
́
(13) Bousseksou, A.; Molnar, G.; Salmon, L.; Nicolazzi, W. Chem.
Soc. Rev. 2011, 40, 3313−3335.
(14) Halder, G. J.; Kepert, C. J.; Moubaraki, B.; Murray, K. S.;
Cashion, J. D. Science 2002, 298, 1762−1765.
(15) Gutlich, P.; Goodwin, H. A. Spin Crossover in Transition Metal
̈
Compounds; Springer-Verlag: Berlin, 2004; Vol. I−III.
(16) Munoz, M. C.; Real, J. A. Coord. Chem. Rev. 2011, 255, 2068−
̃
2093.
(17) Li, B.; Wei, R.-J.; Tao, J.; Huang, R.-B.; Zheng, L.-S.; Zheng, Z. J.
Am. Chem. Soc. 2010, 132, 1558−1566.
(18) Wei, R.-J.; Tao, J.; Huang, R.-B.; Zheng, L.-S. Inorg. Chem. 2011,
50, 8553−8564.
(19) Halcrow, M. A. Coord. Chem. Rev. 2009, 253, 2493−2514.
́
(20) Craig, G. A.; Costa, J. S.; Roubeau, O.; Teat, S. J.; Aromı, G.
̀
2400 at the “Servei de Microanalisi” of CSIC, Barcelona.
Chem.Eur. J. 2012, 18, 11703−11715.
́
(21) Craig, G. A.; Costa, J. S.; Roubeau, O.; Teat, S. J.; Aromı, G. Eur.
ASSOCIATED CONTENT
* Supporting Information
J. Inorg. Chem. 2013, 745−752.
■
(22) Craig, G. A.; Costa, J. S.; Roubeau, O.; Teat, S. J.; Aromí, G.
Chem.Eur. J. 2011, 17, 3120−3127.
S
The molecular characterization of the crystalline samples and
the experimental details for studies of absorption−desorption
behavior, including the magnetic behavior of all the species.
This material is available free of charge via the Internet at
http://pubs.acs.org
(23) Marcen, S.; Lecren, L.; Capes, L.; Goodwin, H. A.; Letard, J. F.
Chem. Phys. Lett. 2002, 358, 87−95.
(24) Buchen, T.; Gutlich, P.; Sugiyarto, K. H.; Goodwin, H. A.
Chem.Eur. J. 1996, 2, 1134−1138.
(25) Roberts, T. D.; Tuna, F.; Malkin, T. L.; Kilner, C. A.; Halcrow,
M. A. Chem. Sci. 2012, 3, 349−354.
AUTHOR INFORMATION
́ ́
(26) Clemente-Leon, M.; Coronado, E.; Gimenez-Lopez, M. C.;
■
Romero, F. M. Inorg. Chem. 2007, 46, 11266−11276.
Corresponding Authors
josesanchezcosta@gmail.com
guillem.aromi@qi.ub.es
(27) McMurtrie, J.; Dance, I. CrystEngComm 2005, 7, 230−236.
(28) Wu, C. D.; Lin, W. B. Angew. Chem., Int. Ed. 2005, 44, 1958−
1961.
Notes
(29) Camparini, A.; Ponticelli, F.; Tedeschi, P. J. Heterocycl. Chem.
1977, 14, 435−438.
(30) SAINT Plus, Data Reduction and Correction Program, v. 6.01;
Bruker AXS, Inc.: Madison, Wisconsin, 1998.
(31) SADABS, Bruker/Siemens Area Detector Absorption Correction
Program, v.2.01; Bruker AXS, Inc.:Madison, Wisconsin, 1998.
(32) Sheldrick, G. M. Acta Crystallogr., Sect. A 2008, 64, 112−122.
(33) PLATON, A Multipurpose Crystallographic Tool; Utrecht
University: Utrecht, The Netherlands, 2008.
(34) Macrae, C. F.; Edgington, P. R.; McCabe, P.; Pidcock, E.;
Shields, G. P.; Taylor, R.; Towler, M.; van de Streek, J. J. Appl.
Crystallogr. 2006, 39, 453−457.
(35) Dolomanov, O. V.; Bourhis, L. J.; Gildea, R. J.; Howard, J. A. K.;
Puschmann, H. J. Appl. Crystallogr. 2009, 42, 339−341.
The authors declare no competing financial interest.
ACKNOWLEDGMENTS
■
G.A. thanks the ERC for a Starting Grant (258060
FuncMolQIP). The authors thank the Spanish MCI through
CTQ2009-06959 (JSC, GAC and GA) and through CTQ2012-
32247 (GA). The Advanced Light Source is supported by the
Director, Office of Science, Office of Basic Energy Sciences of
the U.S. Department of Energy under contract no. DE-AC02-
05CH11231.
REFERENCES
■
(1) An Introduction to Indoor Air Quality (IAQ) Volatile Organic
Compounds (VOCs); EPA: Washington, DC, 2012; http://www.epa.
gov/iaq/voc.html.
(2) Wenger, O. S. Chem. Rev. 2013, 113, 3686−3733.
(3) Horike, S.; Shimomura, S.; Kitagawa, S. Nature Chem. 2009, 1,
695−704.
(4) Inokuma, Y.; Yoshioka, S.; Ariyoshi, J.; Arai, T.; Hitora, Y.;
Takada, K.; Matsunaga, S.; Rissanen, K.; Fujita, M. Nature 2013, 495,
461−466.
(5) Ohba, M.; Yoneda, K.; Agusti, G.; Munoz, M. C.; Gaspar, A. B.;
̃
Real, J. A.; Yamasaki, M.; Ando, H.; Nakao, Y.; Sakaki, S.; Kitagawa, S.
Angew. Chem., Int. Ed. 2009, 48, 4767−4771.
́ ́
(6) Coronado, E.; Gimenez-Marques, M.; Espallargas, G. M.;
Brammer, L. Nat. Commun. 2012, 3, 828.
(7) Niel, V.; Thompson, A. L.; Munoz, M. C.; Galet, A.; Goeta, A. E.;
̃
Real, J. A. Angew. Chem,. Int. Ed. 2003, 42, 3760−3763.
(8) Galet, A.; Munoz, M. C.; Real, J. A. Chem. Commun. 2006, 4321−
̃
4323.
(9) Huang, Z.; White, P. S.; Brookhart, M. Nature 2010, 465, 598−
601.
(10) Supriya, S.; Das, S. K. J. Am. Chem. Soc. 2007, 129, 3464−3465.
(11) Albrecht, M.; Gossage, R. A.; Lutz, M.; Spek, A. L.; van Koten,
G. Chem.Eur. J. 2000, 6, 1431−1445.
3874
dx.doi.org/10.1021/ja411595y | J. Am. Chem. Soc. 2014, 136, 3869−3874