Low-dimensional porous coordination polymers based on 1,2-bis(4-pyridyl) hydrazine: From structure diversity to ultrahigh CO2/CH4 selectivity

Article abstract of DOI:10.1021/ic300097y

Solvothermal reactions of metal salts, benzenedicarboxylic acids, and 4,4′-azopyridine (azpy) in different conditions produced four coordination polymers, namely, [Zn₃(bdc)₃(bphy)₃] ·2DMF·10H₂O (3; H₂bdc = 1,4- benzenedicarboxylic acid, bphy = 1,2-bis(4-pyridyl)hydrazine, and DMF = N,N-dimethylformamide), [Ni(bdc)(bphy)]·DMF·3.5H₂O (4), [Zn(nipa)(bphy)]·EtOH (5; H₂nipa = 5-nitroisophthalic acid), and [CoBr(bdc)_0.5(bphy)]·2DMA·H₂O (6; DMA = N,N-dimethylacetamide), in which the azpy ligand was in situ reduced. Structural determination reveals that 3-5 consist of the same metal/ligand ratio and similar coordination modes, as well as similar two-dimensional square-grid networks, but differ from their packing/interpenetration modes. 3 consists of alternately arranged single layers and interweaved double layers. Single layers in 4 directly stack in an offset fashion, while 5 is constructed of interdigitated double layers. 6 is a one-dimensional ladderlike structure, which could be regarded as that half of the bridging benzenedicarboxylate ligands in 3-5 are replaced by monodentate bromide ions. Interestingly, the crystal structures of these low-dimensional coordination polymers contain considerable solvent-accessible voids. Thermogravimetric curves, powder X-ray diffraction, and gas sorption experiments were used to study the potential porosity of these structures, which indicated that they can all reversibly desorb and adsorb solvent molecules. In particular, 4 showed gated sorption behavior and high CO₂/CH₄ selectivity because of its flexible structure.

Full text of DOI:10.1021/ic300097y

Article
pubs.acs.org/IC
Low-Dimensional Porous Coordination Polymers Based on 1,2-Bis(4-
pyridyl)hydrazine: From Structure Diversity to Ultrahigh CO2/CH4
Selectivity
Xiao-Min Liu, Rui-Biao Lin, Jie-Peng Zhang,* and Xiao-Ming Chen*
MOE Key Laboratory of Bioinorganic and Synthetic Chemistry, State Key Laboratory of Optoelectronic Materials and Technologies,
School of Chemistry and Chemical Engineering, Sun Yat-Sen University, Guangzhou 510275, China
S
* Supporting Information
ABSTRACT: Solvothermal reactions of metal salts, benzenedicarboxylic acids, and 4,4′-azopyridine (azpy) in different
conditions produced four coordination polymers, namely, [Zn3(bdc)3(bphy)3]·2DMF·10H2O (3; H2bdc = 1,4-
benzenedicarboxylic acid, bphy = 1,2-bis(4-pyridyl)hydrazine, and DMF = N,N-dimethylformamide), [Ni(bdc)-
(bphy)]·DMF·3.5H2O (4), [Zn(nipa)(bphy)]·EtOH (5; H2nipa = 5-nitroisophthalic acid), and [CoBr-
(bdc)0.5(bphy)]·2DMA·H2O (6; DMA = N,N-dimethylacetamide), in which the azpy ligand was in situ reduced. Structural
determination reveals that 3−5 consist of the same metal/ligand ratio and similar coordination modes, as well as similar two-
dimensional square-grid networks, but differ from their packing/interpenetration modes. 3 consists of alternately arranged single
layers and interweaved double layers. Single layers in 4 directly stack in an offset fashion, while 5 is constructed of interdigitated
double layers. 6 is a one-dimensional ladderlike structure, which could be regarded as that half of the bridging
benzenedicarboxylate ligands in 3−5 are replaced by monodentate bromide ions. Interestingly, the crystal structures of these
low-dimensional coordination polymers contain considerable solvent-accessible voids. Thermogravimetric curves, powder X-ray
diffraction, and gas sorption experiments were used to study the potential porosity of these structures, which indicated that they
can all reversibly desorb and adsorb solvent molecules. In particular, 4 showed gated sorption behavior and high CO2/CH4
selectivity because of its flexible structure.
crystal engineering.16−18 Also, they usually show structural
INTRODUCTION
Porous coordination polymers (PCPs) have attracted great
attention in recent years for their intriguing architectures and
potential applications as functional materials in catalysis,
■
changes in packing modes during guest adsorption/desorption,
which is the base of flexible PCPs.19,20 It has been shown that
flexible PCPs are especially useful for selective sorption and
separation applications.21−24 Nevertheless, it is still a great
challenge to control the dynamic nature of a crystalline
coordination network, which can be brittle, flexible, or rigid.25,26
Theoretically, flexible PCPs could be obtained by means of a
flexible ligand, framework entanglement, weak interaction
between multiple frameworks, etc.27−30 In practice, choosing
or designing a suitable ligand is the most important strategy in
constructing the coordination network with targeted structures
and properties. Recently, we found that 4,4′-azopyridine (azpy)
magnetism, nonlinear optics, gas storage, separation, etc.1−7
To generate porosity, three-dimensional (3D) frameworks have
been largely studied because high-dimensional structures
usually possess stronger 3D framework rigidity, which facilitates
the retention of pores after removal of the template
molecules.8−13 In contrast, the 3D supramolecular structures
of low-dimensional frameworks sustained by weak interactions
such as hydrogen-bonding, π−π-stacking, and van der Waals
interactions generally collapse when the template guests are
removed.14,15 On the other hand, low-dimensional structural
motifs can produce packing and/or structure diversity easier
during crystallization, which has been a research interest in
Received: January 14, 2012
Published: May 7, 2012
© 2012 American Chemical Society
5686
dx.doi.org/10.1021/ic300097y | Inorg. Chem. 2012, 51, 5686−5692

Inorganic Chemistry
Article
Table 1. Crystal Data and Structure Refinement for Compounds 3−6
3
4
5
6
formula
fw
C60H76N14O24Zn3
1573.55
P2/c
C21H28N5O8.5Ni
545.17
C20H19N5O7Zn
506.80
C22H32N6O5BrCo
599.36
space group
a/Å
P21/n
P21/n
P21/n
10.9425(13)
21.354(2)
18.840(2)
105.218(2)
4247.8(9)
2
10.0803(11)
15.8190(18)
16.3208(18)
101.034(3)
2554.4(5)
4
10.1848(10)
11.1036(11)
19.1820(18)
94.397(2)
2162.9(4)
4
14.1464(16)
10.7476(12)
18.141(2)
97.938(2)
2731.7(5)
4
b/Å
c/Å
β/deg
V/Å3
Z
Dc/g cm−3
μ/mm−1
GOF
0.975
1.064
1.415
0.990
0.888
0.782
1.178
2.097
1.003
1.002
1.006
1.002
a
R1 [I > 2σ(I)]
0.0480
0.0593
0.0405
0.0532
b
wR2 (all data)
0.1387
0.1090
0.1077
0.1601
a
b
2
2
R1 = ∑||Fo| − |Fc||/∑|Fo|. wR2 = [∑w(Fo − Fc2)2/∑w(Fo )2]1/2
.
with Belsorp-Max and Belsorp-HP automatic volumetric adsorption
apparatuses.
could be in situ reduced into 1,2-bis(4-pyridyl)hydrazine
(bphy) during a solvothermal reaction.31 The new bphy ligand
is advantageous for constructing flexible networks. On the one
hand, bphy possesses backbone flexibility arising from the long
C−N−N−C single bonds. On the other hand, the hydrazo
group (−NH−NH−) moiety can serve as hydrogen-bonding
sites for internetwork interactions and/or guest molecules. For
example, we have synthesized two new 3D PCPs, [Zn(bdc)-
(bphy)] (1) and [Zn(bdc)(bphy)0.5] (2), with uninodal 4-
connected dmp and pcu topologies, by reacting Zn(NO3)2, 1,4-
benzenedicarboxylic acid (H2bdc), and azpy with correspond-
ing metal/ligand molar ratios in N,N-dimethylformamide
(DMF)/N,N-dimethylacetamide (DMA)/ethanol (EtOH) at
160 °C, both of which demonstrate dramatic framework
flexibility upon inclusion of different solvent molecules.31 For
example, 1 was synthesized as two guest-induced framework-
distortional isomers, [Zn(bdc)(bphy)]·DMF·H2O (1a) and
[Zn(bdc)(bphy)]·EtOH·H2O (1b), by using DMF and EtOH
as solvents, respectively.
Syntheses of [Zn3(bdc)3(bphy)3]·2DMF·10H2O (3). A mixture of
Zn(NO3)2·6H2O (0.5 mmol, 148 mg), H2bdc (1 mmol, 40 mg), and
azpy (0.5 mmol, 58 mg) in 15 mL of DMF was stirred at 60 °C for 30
min, then sealed in a Teflon-lined autoclave, and heated to 160 °C for
2 days. After the autoclave was cooled to room temperature at 5 °C
h−1, the mother liquid was decanted, and the crystals were rinsed three
times with DMF (8 mL × 3) and dried in air for 1 h (yield ∼80%
based on Zn). Anal. Calcd for C60H76N14Zn3O24: C, 45.80; H, 4.87; N,
12.46. Found: C, 45.71; H, 4.83; N, 12.41. IR (KBr, cm−1): 3731 (w),
3627 (w), 2360 (s), 1623 (s), 1559 (m), 1480 (w), 1389 (s), 1104
(w), 1024 (w), 848 (m), 748 (s), 720 (s), 669 (m), 594 (w).
[Ni(bdc)(bphy)]·DMF·3.5H2O (4). The synthesis of 4 was similar to
that of 3 except that Ni(NO3)2·6H2O (0.5 mmol, 90 mg) was used
instead of Zn(NO3)2·6H2O (yield ∼70% based on Ni). Anal. Calcd for
C21H28N5O8.5Ni: C, 46.27; H, 5.18; N, 12.85. Found: C, 46.21; H,
5.12; N, 12.79. IR (KBr, cm−1): 3284 (m), 2490 (w), 1668 (w), 1619
(s), 1541 (m), 1439 (w), 1402 (s), 1353 (w), 1305 (w), 1211 (m),
1057 (w), 1016 (s), 842 (m), 817 (m), 751 (m), 537 (m).
[Zn(nipa)(bphy)]·EtOH (5). A mixture of Zn(NO3)2·6H2O (0.5
mmol, 148 mg), H2nipa (1 mmol, 40 mg), and azpy (0.5 mmol, 58
mg) in 15 mL of EtOH was stirred at 60 °C for 30 min, then sealed in
a Teflon-lined autoclave, and heated to 160 °C for 2 days. After the
autoclave was cooled to room temperature at 5 °C h−1, the mother
liquid was decanted, and the crystals were rinsed three times with
EtOH (8 mL × 3) and dried in air for 0.5 h (yield ∼78% based on
Zn). Anal. Calcd for C20H19N5O7Zn: C, 47.40; H, 3.78; N, 13.82.
Found: C, 47.45; H, 3.71; N, 13.89. IR (KBr, cm−1): 3236 (m), 2459
(w), 1681 (m), 1621 (s), 1569 (m), 1535 (w), 1496 (m), 1367 (s),
1344 (m), 1214 (s), 1058 (s), 824 (m), 731 (s), 659 (w), 612 (w), 527
(w).
[CoBr(bdc)0.5(bphy)]·2DMA·H2O (6). The synthesis of 6 was similar
to that of 3 except that CoBr2 (0.5 mmol, 108 mg) and DMA were
used instead of Zn(NO3)2·6H2O and DMF (yield ∼50% based on
Co). Anal. Calcd for C22H32N6O5BrCo: C, 44.09; H, 5.38; N, 14.02.
Found: C, 44.12; H, 5.35; N, 14.09. IR (KBr, cm−1): 3424 (w), 2360
(m), 1614 (s), 1501 (w), 1382 (s), 1207 (m), 1054 (w), 1019 (s), 827
(s), 748 (s), 589 (w).
X-ray Crystallography. Diffraction data were collected on a
Bruker Apex CCD area-detector diffractometer with Mo Kα radiation
at 93(2) K. The structures were solved by direct or Patterson methods
and refined by the full-matrix least-squares method using SHELXTL.
All H atoms were placed geometrically, and anisotropic thermal
parameters were used to refine all non-H atoms of the frameworks.
Most disordered guest molecules in 3−6 could not be modeled, and
their electron density peaks were removed by the SQUEEZE route in
PLATON. Their amounts were determined by TG results and
In this work, we further investigated the application of this
new solvothermal in situ ligand reaction in the construction of
new coordination flexible PCPs by varying the reaction
conditions and/or using other aromatic acids as the coligands.
Four new low-dimensional coordination polymers showing
interesting structures and gas sorption properties have been
successfully synthesized and characterized.
EXPERIMENTAL SECTION
■
Materials and General Methods. Commercially available
reagents were used as received without further purification. The
ligand azpy was prepared according to a reported method.32
Electrospray ionization mass spectrometry (ESI-MS) spectra were
measured on a Shimadzu LCMS-2010A apparatus using an ESI source
with methanol as the mobile phase. IR spectra were recorded with a
Bruker TENSOR 27 Fourier transform infrared spectrophotometer on
KBr pellets in the range of 4000−400 cm−1. Elemental analyses (C, H,
and N) were performed on a Perkin-Elmer 240 elemental analyzer.
Thermogravimetric (TG) analyses were performed on a Netzsch TG
209 instrument in flowing N2 with a heating rate of 10 °C min−1.
Powder X-ray diffraction (PXRD) measurements were performed on a
Bruker D8 ADVANCE X-ray diffractometer with Cu Kα radiation. For
variable-temperature (VT)PXRD measurements, the diffraction
patterns at different temperatures were recorded after the sample
had stayed at the respective temperature for 30 min in a N2
atmosphere. CO2 and CH4 sorption measurements were performed
5687
dx.doi.org/10.1021/ic300097y | Inorg. Chem. 2012, 51, 5686−5692

Inorganic Chemistry
Article
elemental analyses. The crystal data and structure refinement results
are listed in Table 1.
RESULTS AND DISCUSSION
■
Synthesis. The azpy ligand was in situ reduced into bphy in
the presence of metal salts and benzenedicarboxylic acids,
giving rise to four new compounds, viz., 3−6. It should be
noted that the framework composition of 3 is identical with
those of 1a, 1b, and 1 with the same dmp topology.31 1a and
1b were synthesized using different solvents, resulting in
different guest inclusions and framework distortions, and they
can be converted to the same guest-free phase 1 upon guest
removal. Interestingly, the synthetic conditions, including the
solvent, reactant, and reaction temperature, are basically
identical for 1a and 3. The only difference is the pretreatment
and reaction time of solvothermal syntheses. We found that
stirring the suspension of starting materials at 60 °C before
solvothermal treatment and shortening the reaction time was
beneficial to producing 3. Otherwise, the product would be 1a
or a mixture of 1a and 3. For reference, pure 1a and 3 were
synthesized at 160 °C for 3 and 2 days, respectively. These
observations indicate that 1a and 3 are thermodynamically and
kinetically favored phases, respectively.
Crystal Structures. X-ray crystallographic analyses revealed
that the N−N bond lengths [1.383(3)−1.399(7) Å] in 3−6 are
obviously longer than the typical NN bond lengths (e.g., 1.25
Å in azpy) yet similar to those of the N−N bonds (e.g., 1.36−
1.44 Å in bphy). Besides, the absolute values of C−N−N−C
torsion angles [73.9(2)−95.7(7)°] are distinctly deviated from
the planar configuration of C−NN−C moieties in azpy
(180°) yet similar to that reported for free bphy [84.37(3)°].
All of these structural data confirm in situ reduction of the
starting material azpy into bphy. There are two independent
Zn2+ ions (Zn1 and Zn2), two bdc2− ligands, and two bphy
ligands in the asymmetric unit of 3, although the coordination
modes and/or molecular geometries of the two sets of metal
ions/ligands are similar to each other. Each Zn2+ ion is four-
coordinated with two carboxylate O atoms [Zn−O 1.923(3)−
1.957(4) Å] from two bdc2− ligands and two pyridyl N atoms
[Zn−N 2.009(4)−2.025(4) Å] from two bphy ligands in a
distorted tetrahedral geometry (Figure 1a,c). The bdc2− ligands
bridge two Zn2+ ions by two monodentate carboxylate groups,
and the bphy ligands also link two Zn2+ ions by the two pyridyl
ends. Consequently, the Zn ions are connected by bdc2‑ and
bphy into two sets of crystallographically independent two-
dimensional (2D) wavy square-grid-like layers with the
common sql topology (Figure 1b,d). Although the structures
of these 2D networks are quite similar, their interpenetration
modes are different. The networks containing Zn1 do not
interpenetrate with each other, while those containing Zn2
undergo parallel 2-fold interpenetration to form double layers,
with π−π-stacking interactions between the aromatic rings of
bdc2− from one layer and bdc2− or bphy from another layer
(Figures 1d and S3 in the Supporting Information). These
double and single layers stack alternately to furnish the 3D
structure of 3 (Figure 1e). The closest interactions between the
double and single layers are the N−H···O hydrogen bonds
between the hydrazo group of bphy and the carboxylate group
of bdc2− [N···O 2.794(3)−2.824(1) Å; N−H···O 160.4(2)−
166.2(1)°] and π−π-stacking interactions between the pyridyl
rings of bphy ligands (Figures S2 and S4 in the Supporting
Information). The wavy square-grid structures and missing
interpenetration in half of the packing layers in 3 result in a
Figure 1. Perspective views of (a) the coordination environment of
Zn1, (b) the corresponding single-layer coordination network, (c) the
coordination environment of Zn2, (d) the corresponding parallel
interpenetrated double-layer coordination network, and (e) the
alternate stacking of the single and double layers in 3 (pink and
green dotted lines represent hydrogen-bonding and π−π-stacking
interactions, respectively).
large solvent-accessible void of 45.6%, which is occupied by
DMF and water molecules, as confirmed by TG and elemental
analyses.33 The hydrazo groups of bphy ligands not only form
interlayer hydrogen bonds to sustain the 3D stacking structure
but also behave as hydrogen-bonding donors to fix the guest
water molecules [N−H···O 2.865(1) Å].
Similar to 3, 4 is also composed of square-grid-like
coordination networks with sql topology. However, there is
only one Ni2+ ion, one bdc2− ligand, and one bphy ligand in the
asymmetric unit. Being somewhat different from the Zn2+ ions
in 3, the Ni2+ ion is six-coordinated with four carboxylate O
atoms [Ni−O 1.950(4)−2.195(7) Å] from two bdc2− ligands
and two N atoms [Ni−N 2.011(4)−2.052(4) Å] from two
bphy ligands in distorted octahedral geometries (Figure 2a). In
contrast to the complicated structure of 3, the 2D sql networks
are not interpenetrated in 4 (Figure 2b,c). These layers stack
via interlayer N−H...O hydrogen-bonding [N···O 2.872(1) Å;
N−H···O 158.2(2)°] and π−π-stacking interactions to form the
3D supramolecular structure (Figures S5 and S6 in the
Supporting Information). By virtue of the wavy shapes of the
layers and directional interlayer supramolecular interactions,
there is a very small offset between adjacent layers, which is
different from the staggered stacking fashion commonly
observed for 2D layer structures (offset of adjacent layers is
half of the grid size). Consequently, 4 also possesses large
solvent-accessible voids of 44.1%, which is occupied by DMF
and water molecules, forming hydrogen bonds with the hydrazo
groups [N···O 2.794(3)−2.824(1) Å; N−H···O 110.1(2)−
150.2(1)°].
Although the dicarboxylate nipa2− is bent, the local
coordination configuration [Zn−O 1.950(4)−2.195(7) Å;
5688
dx.doi.org/10.1021/ic300097y | Inorg. Chem. 2012, 51, 5686−5692

Inorganic Chemistry
Article
layer structure, leaving the bphy ligands in the outside of the
double layer (Figure 3c). The closest contact between the two
layers is the very strong π−π interaction between the
nitrobenzene moiety of nipa2− (3.4 Å). These double layers
further stack to form the 3D structure via N−H···O hydrogen
bonding between the hydrazo and carboxylate groups [N···O
2.851(3)−2.862(1) Å; N−H···O 152.3(1)−171.2(3)°] and the
π−π-stacking interactions between the pyridyl rings of the bphy
ligands (Figures S7 and S8 in the Supporting Information).
Compared to the structures of 3 and 4, the staggered stacking
mode of 5 is more similar to those of common 2D structures.
Owing to the quite small lattice size, 5 possesses a small
solvent-accessible void of 21.6%. Because all of the hydrazo
groups are employed to form interlayer hydrogen bonds, there
is no strong hydrogen bond [C···O 3.558(7) Å] between the
guest molecules and the framework.
In 6, there are one Co2+, half of a bdc2− ligand, one bphy, and
one Br− in the asymmetric unit. The Co2+ ion is coordinated by
a monodentate carboxylate [Co−O 1.971(4) Å] and two
pyridyl groups [Co−N 2.020(5)−2.025(5) Å] from two bphy
ligands and a Br− ion [Co−Br 2.396(1) Å] to furnish a
distorted tetrahedral geometry (Figure 4a). These Co2+ ions are
Figure 2. Perspective views of (a) the coordination environment of the
Ni atom, (b) the 2D coordination network with sql topology, and (c)
the stacking fashion of the 2D layers (pink and green dotted lines
represent the interlayer hydrogen-bonding and π−π-stacking inter-
actions, respectively) for 4.
Zn−N 2.011(4)−2.052(4) Å] and the network topology in 5
are all similar to 3 constructed by the linear bdc2− (Figure
3a,b). It is worth mentioning that the sql network of 5 is quite
Figure 4. Perspective views of (a) the coordination environment of the
Co atom, (b) the ladderlike 1D coordination network, (c) the
hydrogen-bonding interaction of the ladder, and (d) the 3D
supramolecular structure sustained by hydrogen-bonding (pink dotted
lines) and π−π-stacking (green dotted lines) interactions in 6.
Figure 3. Perspective views of (a) the coordination environment of the
Zn atom, (b) the 2D coordination network, and (c) the stacking of the
2D coordination layers (pink and green dotted lines represent the
interlayer hydrogen-bonding and π−π interactions, respectively) in 5.
connected by bidentate bphy ligands to form a 1D wavy chain.
Two chains as side pieces are bridged together by ditopic bdc2−
ligands as rungs into a molecular ladder (Figure 4b). The
adjacent Co···Co separations bridged by the bphy and bdc2−
ligands are 10.75 and 10.95 Å, respectively. Each ladder forms
strong hydrogen bonds with two neighbors using the hydrazo
and carboxylate groups [N···O 2.795(7) Å; N−H···O 171.5(2)
°], giving a 2D hydrogen-bonding-layered structure (Figure
4c). Adjacent layers are then packed via π−π-stacking
interactions to form a 3D supramolecular architecture (Figures
4d and S2 and S4 in the Supporting Information). By virtue of
regular (the grid looks like a square), while those of 3 and 4 are
distorted as rhombus. Because the bridging distance of the
bending dicarboxylate nipa2− is shorter than that of bdc2−, the
grid size of 5 is smaller than those of 3 and 4. The different
shapes of the 2D layers result in different packing modes. In 5,
two layers interdigitate by the nipa2− side to form a double-
5689
dx.doi.org/10.1021/ic300097y | Inorg. Chem. 2012, 51, 5686−5692

Inorganic Chemistry
Article
the directional hydrogen-bonding and π−π-stacking interac-
tions, the ladders do not interdigitate with each other.
Consequently, the whole framework possesses a large solvent-
accessible void of 45.9%, in which the guest DMA molecules
are anchored to the host framework by hydrogen bonds with
the hydrazo groups [N···O 2.735(4) Å; N−H···O 169.7(5)°].
Framework Stability. Thermogravimetric (TG) curves
show that 3−6 release all of the guests at 190, 180, 150, and
240 °C and decompose at 300, 370, 390, and 350 °C,
respectively. Their amounts of released guests match well with
their formulas (see the Supporting Information).
VTPXRD patterns show that 3 and 4 can be converted to
new phases after guest removal, indicating flexibility of these
frameworks (Figures S11 and S14 in the Supporting
Information). Because 1a/1b/1 and 3 have the same
framework composition but different topologies, we have
checked whether a structural transformation between 1a/1b/1
and 3 takes place upon removal of the solvents. We have
demonstrated that 1a and 1b could convert to the same phase 1
after their guests are removed.31 However, the PXRD patterns
of 1a, 1b, 1, 3, and guest-free 3 are different with each other
(Figure S11 in the Supporting Information). In other words, a
structural transformation between 1a/1b/1 and 3 would not
take place. The PXRD pattern of the guest-free sample of 5
matches well with its as-synthesized phase, suggesting that its
framework is rigid, probably related to its relatively close-
packing structure (Figures S17 and S19 in the Supporting
Information). For 6, the PXRD pattern of its guest-free sample
became weaker and broader, indicating that the host framework
lost long-range order or collapsed partially (Figure S20 in the
Supporting Information). In order to verify the reversibility in
the structure conversions of 3, 4, and 6, their activated samples
were exposed in DMF and/or DMA vapors. After several hours,
PXRD patterns show that the activated samples of 3 and 4
transform back to their as-synthesized structures. However,
guest-free 6 can hardly be recovered to the good crystallinity of
the as-synthesized 6 even after several weeks.
Sorption Studies. Gas sorption studies were carried out to
evaluate the porosity of guest-free 3, 4, and 6 because their
crystal structures contain relatively large solvent-accessible
voids. The activation conditions were optimized based on the
above-mentioned TG and VTPXRD results. The as-synthesized
samples of 3, 4, and 6 were heated at different temperatures
under high vacuum for about 8 h and then further checked by
PXRD and TG measurements to confirm the complete removal
of template molecules and retention of high crystallinity. The
results indicated that the activation temperatures of 3 and 4
should be 190 and 180 °C, as indicated by TG analyses, while
the framework of 6 could be activated at 200 °C. The CO2
sorption isotherms of 3, 4, and 6 measured at 195 K showed
type I isotherm characteristics, illustrating the microporous
nature of these samples (Figure 5). The CO2-saturated uptakes
of 3 and 6, only 24 and 35 cm3 g−1, respectively, are
significantly lower than the theoretical values estimated from
the crystal structures (262 and 301 cm3 g−1). Besides, the
isotherms of 3 and 6 represent obvious hysteresis, illustrating
that there is a great barrier for the gas diffusion. All of these
phenomena indicate that the structures of 3 and 6 severely are
deformed into relatively condensed phases after guest removal,
which can accommodate relatively small amounts of gas
molecules and also hamper the gas diffusion. In contrast, 4
exhibits a relatively high saturated adsorption at 195 K (131
cm3 g−1), corresponding to a pore volume of 0.15 cm3 g−1,
Figure 5. CO2 isotherms measured at 195 K.
although it is only one-third of the calculated amount (0.44 cm3
g−1) from its single-crystal structure. As mentioned above, the
structure of 4 also undergoes phase transition after guest
removal, just similar to that of 3. Compared to guest-free 3,
guest-free 4 can be obviously opened by CO2 at 195 K.
Moreover, the isotherm for guest-free 4 is very steep at low
pressure. The onset pressure is very low (if any), and there is
only very small hysteresis, implying that the framework of
guest-free 4 is very flexible and can be opened by CO2 at very
low pressure. Of course, the onset pressure of adsorption
should be changed along with the temperature and/or guests.
For example, we have demonstrated that 2, showing flexible
behavior similar to that of 4, could be used to efficiently
separate CO2 and CH4. At 298 K, guest-free 2 started to adsorb
in a little CO2 and totally opened its pore at about 22 bar, while
it could hardly adsorb in CH4 up to 40 bar. In short,
considering the lower onset pressure and the more dramatic
step of the CO2 isotherm, 4 may exhibit a better CO2/CH4
separation performance.
CO2 and CH4 sorption isotherms were measured for 4 at 298
K with pressure up to 40 bar (Figure 6a). The CO2 uptake is
very low below 5 bar (3.8 cm3 g−1). After that, the uptake
suddenly increases to 81 cm3 g−1 at 6 bar and further increases
to 147 cm3 g−1 at 40 bar gradually. Although at 298 K and 40
bar the pressure is far away from the saturated vapor pressure
64 bar (P/P0 = 0.6), the adsorption amount is obviously higher
than that of P/P0 = 1 at 195 K, suggesting that the increasing
temperature is beneficial to increasing framework flexibility,
which has been observed in other PCPs.34−37 In contrast,
almost no CH4 could be adsorbed up to 40 bar (2.0 cm3 g−1).
Consequently, the CO2/CH4 uptake ratio changes basically
according to the CO2 isotherm shape (Figure 6b). Between 0
and 5 bar, the CO2/CH4 uptake ratio gradually increases from 2
to 36 and then sharply increases to a maximum of 466 at 6 bar.
Although the ratio gradually decreases at higher pressures, it is
still 75 at 40 bar. For comparison, we also calculated the
pressure-dependent CO2/CH4 uptake ratio for 2, which has a
shape similar to that for 4 because the two frameworks both
undergo a gate-opening process during CO2 adsorption.
However, the maximum CO2/CH4 uptake ratio for 2 is only
22, which locates at a relatively high pressure of 22 bar. The
adsorption isotherms of the strong adsorbate CO2 reveal the
different framework flexibility and the location of the maximum
CO2/CH4 uptake ratios of 2 and 4. The host framework of 2
can be completely opened, but the phase transition occurs
gradually between 10 and 22 bar. In contrast, the host
framework of 4 can be only opened to one-third of the as-
synthesized phase, but its phase transition occurs suddenly
5690
dx.doi.org/10.1021/ic300097y | Inorg. Chem. 2012, 51, 5686−5692

Inorganic Chemistry
Article
flexibility plays an important role in determining the gas
separation performance. Although rational design and control
of the framework structure and flexibility is still of great
challenge, systematic structure modulation of highly similar
coordination framework structures could serve as a viable way
for the discovery of excellent gas separation materials.
ASSOCIATED CONTENT
■
S
* Supporting Information
Additional plots, TG analyses, and PXRD patterns, as well as X-
ray data files in CIF format. This material is available free of
charge via the Internet at http://pubs.acs.org.
AUTHOR INFORMATION
■
Corresponding Author
*E-mail: zhangjp7@mail.sysu.edu.cn (J.-P.Z.), cxm@mail.sysu.
edu.cn (X.-M.C.).
Notes
The authors declare no competing financial interest.
ACKNOWLEDGMENTS
■
This work was supported by the “973 Project” (Grant
2012CB821706), NSFC (Grants 21121061 and 21001120),
and Chinese Ministry of Education (Grants NCET-10-0863 &
ROCS).
Figure 6. High-pressure CO2 and CH4 (a) adsorption isotherms and
(b) CO2/CH4 uptake ratios of 2 and 4 measured at 298 K (inset:
magnified CO2/CH4 uptake ratios of 4).
REFERENCES
■
(1) Xie, Z. G.; Ma, L. Q.; deKrafft, K. E.; Jin, A.; Lin, W. B. J. Am.
Chem. Soc. 2010, 132, 922.
between 5 and 6 bar. For adsorbents based on exclusion
mechanisms such as the size selectivity and gate opening, the
uptake of weaker adsorbate is most important in determining
the adsorption selectivity. The guest-free phase of 2 contains
more accessible pores than those of 4, although they both can
be described as nonporous. Therefore, 2 can adsorb more CH4
than 4 in the whole measured pressure region, which
significantly reduces the CO2/CH4 selectivity of 2. In summary,
because of the different framework flexibility from 2, 4 can
separate CO2/CH4 more efficiently because of not only the
high adsorption ratios but also the higher ratio appearing at
lower pressure, which is beneficial for energy savings. This
interpretation may also be suitable for understanding why
guest-free 3 cannot show gated sorption behavior. Although the
frameworks of 2−4 all show structural flexibility upon removal
of solvent/template molecules (high boiling points and ability
to strongly interact with the frameworks), the different
thermodynamic characteristics, i.e., energy difference or energy
barrier between multiple thermodynamically stable states, can
be revealed by the gas molecules CO2/CH4 (low boiling points
and ability to weakly interact with the frameworks).
(2) Xie, Y. M.; Yu, R. M.; Wu, X. Y.; Wang, F.; Chen, S. C.; Lu, C. Z.
CrystEngComm 2010, 12, 3490.
(3) Chu, Q.; Su, Z.; Fan, J.; Okamura, T.; Lv, G. C.; Liu, G. X.; Sun,
W. Y.; Ueyama, N. Cryst. Growth Des. 2011, 11, 3885.
(4) Zhang, X. M.; Hao, Z. M.; Zhang, W. X.; Chen, X. M. Angew.
Chem., Int. Ed. 2007, 46, 3456.
(5) Zhou, X.-P.; Li, M.; Liu, J.; Li, D. J. Am. Chem. Soc. 2011, 134, 67.
(6) Alkordi, M. H.; Belof, J. L.; Rivera, E.; Wojtas, L.; Eddaoudi, M.
Chem. Sci. 2011, 2, 1695.
(7) Ohtani, R.; Yoneda, K.; Furukawa, S.; Horike, N.; Kitagawa, S.;
Gaspar, A. B.; Munoz, M. C.; Real, J. A.; Ohba, M. J. Am. Chem. Soc.
2011, 133, 8600.
(8) Yuan, D. Q.; Lu, W. G.; Zhao, D.; Zhou, H. C. Adv. Mater. 2011,
23, 3723.
(9) Wu, S. T.; Ma, L. Q.; Long, L. S.; Zheng, L. S.; Lin, W. B. Inorg.
Chem. 2009, 48, 2436.
(10) Chang, Z.; Zhang, D. S.; Chen, Q.; Li, R. F.; Hu, T. L.; Bu, X. H.
Inorg. Chem. 2011, 50, 7555.
(11) Zeng, M. H.; Wang, Q. X.; Tan, Y. X.; Hu, S.; Zhao, H. X.;
Long, L. S.; Kurmoo, M. J. Am. Chem. Soc. 2010, 132, 2561.
(12) Takashima, Y.; Martinez, V. M.; Furukawa, S.; Kondo, M.;
Shimomura, S.; Uehara, H.; Nakahama, M.; Sugimoto, K.; Kitagawa, S.
Nat. Commun. 2011, 2, 168.
(13) Seeber, G.; Cooper, G. J. T.; Newton, G. N.; Rosnes, M. H.;
Long, D. L.; Kariuki, B. M.; Kogerler, P.; Cronin, L. Chem. Sci. 2010, 1,
62.
CONCLUSIONS
■
In summary, four new low-dimensional coordination polymers
with similar coordination structures have been synthesized by
adopting a new solvothermal in situ ligand reaction with
different secondary ligands and reaction conditions. The
solvothermal in situ reduction of azpy to bphy is further
proven to be an effective strategy for the crystal engineering of
coordination polymers. As observed in the diverse inter-
penetration/packing structures and supramolecular interactions
of these low-dimensional coordination frameworks, bphy
represents a unique ligand for the construction of flexible
coordination networks. More importantly, the framework
(14) Kitagawa, S.; Kitaura, R.; Noro, S. Angew. Chem., Int. Ed. 2004,
43, 2334.
(15) Kitagawa, S.; Uemura, K. Chem. Soc. Rev. 2005, 34, 109.
(16) Zhang, J. P.; Qi, X. L.; He, C. T.; Wang, Y.; Chen, X. M. Chem.
Commun. 2011, 47, 4156.
(17) Wang, C. C.; Yang, C. C.; Yeh, C. T.; Cheng, K. Y.; Chang, P.
C.; Ho, M. L.; Lee, G. H.; Shih, W. J.; Sheu, H. S. Inorg. Chem. 2011,
50, 597.
(18) Chen, D.; Liu, Y. J.; Lin, Y. Y.; Zhang, J. P.; Chen, X. M.
CrystEngComm 2011, 13, 3827.
5691
dx.doi.org/10.1021/ic300097y | Inorg. Chem. 2012, 51, 5686−5692

Inorganic Chemistry
Article
(19) Kondo, A.; Nakagawa, T.; Kajiro, H.; Chinen, A.; Hattori, Y.;
Okino, F.; Ohba, T.; Kaneko, K.; Kanoh, H. Inorg. Chem. 2010, 49,
9247.
(20) Arslan, H. K.; Shekhah, O.; Wieland, D. C. F.; Paulus, M.;
Sternemann, C.; Schroer, M. A.; Tiemeyer, S.; Tolan, M.; Fischer, R.
A.; Woll, C. J. Am. Chem. Soc. 2011, 133, 8158.
(21) Inubushi, Y.; Horike, S.; Fukushima, T.; Akiyama, G.; Matsuda,
R.; Kitagawa, S. Chem. Commun. 2010, 46, 9229.
(22) Xiang, S. C.; Zhang, Z. J.; Zhao, C. G.; Hong, K. L.; Zhao, X. B.;
Ding, D. R.; Xie, M. H.; Wu, C. D.; Das, M. C.; Gill, R.; Thomas, K.
M.; Chen, B. L. Nat. Commun. 2011, 2, 204.
(23) Demessence, A.; Long, J. R. Chem.Eur. J. 2010, 16, 5902.
(24) Jin, Z.; Zhao, H. Y.; Zhao, X. J.; Fang, Q. R.; Long, J. R.; Zhu, G.
S. Chem. Commun. 2010, 46, 8612.
(25) Uemura, T.; Yanai, N.; Watanabe, S.; Tanaka, H.; Numaguchi,
R.; Miyahara, M. T.; Ohta, Y.; Nagaoka, M.; Kitagawa, S. Nat.
Commun. 2010, 1, 83.
(26) Colombo, V.; Galli, S.; Choi, H. J.; Han, G. D.; Maspero, A.;
Palmisano, G.; Masciocchi, N.; Long, J. R. Chem. Sci. 2011, 2, 1311.
(27) Cheng, Y.; Kajiro, H.; Noguchi, H.; Kondo, A.; Ohba, T.;
Hattori, Y.; Kaneko, K.; Kanoh, H. Langmuir 2011, 27, 6905.
(28) Ma, S.; Sun, D.; Wang, X.-S.; Zhou, H.-C. Angew. Chem., Int. Ed.
2007, 46, 2458.
(29) He, Y. B.; Xiang, S. C.; Chen, B. L. J. Am. Chem. Soc. 2011, 133,
14570.
(30) Kanoo, P.; Mostafa, G.; Matsuda, R.; Kitagawa, S.; Maji, T. K.
Chem. Commun. 2011, 47, 8106.
(31) Liu, X. M.; Xie, L. H.; Lin, J. B.; Lin, R. B.; Zhang, J. P.; Chen, X.
M. Dalton Trans. 2011, 40, 8549.
(32) Launay, J. P.; Tourrel-Pagis, M.; Lipskier, J. F.; Marvaud, V.;
Joachim, C. Inorg. Chem. 1991, 30, 1033.
(33) Spek, A. L. J. Appl. Crystallogr. 2003, 36, 7.
(34) Choi, H. J.; Dinca, M.; Long, J. R. J. Am. Chem. Soc. 2008, 130,
7848.
(35) Walton, K. S.; Millward, A. R.; Dubbeldam, D.; Frost, H.; Low,
J. J.; Yaghi, O. M.; Snurr, R. Q. J. Am. Chem. Soc. 2008, 130, 406.
(36) Zhang, J. M.; Wu, H. H.; Emge, T. J.; Li, J. Chem. Commun.
2010, 46, 9152.
(37) Matsuda, R.; Tsujino, T.; Sato, H.; Kubota, Y.; Morishige, K.;
Takata, M.; Kitagawa, S. Chem. Sci. 2010, 1, 315.
5692
dx.doi.org/10.1021/ic300097y | Inorg. Chem. 2012, 51, 5686−5692