DOI: 10.1002/chem.

201303610 Full Paper

& Metal–Organic Frameworks

Flexible and Rigid Amine-Functionalized Microporous Frameworks

Based on Different Secondary Building Units: Supramolecular
Isomerism, Selective CO2 Capture, and Catalysis
Ritesh Haldar,[a] Sandeep K. Reddy,[b] Venkata M. Suresh,[b] Sudip Mohapatra,[b]
Sundaram Balasubramanian,[b] and Tapas Kumar Maji *[a, b]

Abstract: We report the synthesis, structural characteriza- lectivity for CO2 among other gases (N2, O2, H2, and Ar),
tion, and porous properties of two isomeric supramolecular which has been correlated to the specific interaction of CO2
complexes of ([Cd(NH2bdc)(bphz)0.5]·DMF·H2O}n (NH2bdc = with the NH2 and =NN= functionalized pore surface. DFT
2-aminobenzenedicarboxylic acid, bphz = 1,2-bis(4-pyridyl- calculations for the oxo-bridged isomer unveiled that the
methylene)hydrazine) composed of a mixed-ligand system. NH2 group is the primary binding site for CO2. The high
The first isomer, with a paddle-wheel-type Cd2(COO)4 secon- heat of CO2 adsorption (DHads = 37.7 kJ mol1) in the oxo-
dary building unit (SBU), is flexible in nature, whereas the bridged isomer is realized by NH2···CO2/aromatic p···CO2 and
other isomer has a rigid framework based on a m-oxo- cooperative CO2···CO2 interactions. Further, postsynthetic
bridged Cd2(m-OCO)2 SBU. Both frameworks are two-fold in- modification of the NH2 group into NHCOCH3 in the
terpenetrated and the pore surface is decorated with pend- second isomer leads to a reduced CO2 uptake with lower
ant NH2 and =NN= functional groups. Both the frame- binding energy, which establishes the critical role of the
works are nonporous to N2, revealed by the type II adsorp- NH2 group for CO2 capture. The presence of basic NH2
tion profiles. However, at 195 K, the first isomer shows an sites in the oxo-bridged isomer was further exploited for effi-
unusual double-step hysteretic CO2 adsorption profile, cient catalytic activity in a Knoevenagel condensation reac-
whereas the second isomer shows a typical type I CO2 pro- tion.
file. Moreover, at 195 K, both frameworks show excellent se-

Introduction tion,[4] catalysis,[5] ion exchange,[6] and as opto-electronic mate-

rials.[7] A proper combination of metal ion and organic linker(s)
Inorganic–organic hybrid porous solids represent a new class initiates the self-assembly process and by simply tuning the re-
of crystalline porous material, commonly known as metal–or- action conditions (temperature, pH, solvent) for the self-assem-
ganic frameworks (MOFs) or porous coordination polymers bly process, a variety of structures with different network top-
(PCPs).[1] These materials possess the merits of both organic ologies and dimensionalities can be obtained. Use of the same
and inorganic building units, and surmount all the limitations set of metal ions and linkers may yield compounds of identical
that other contemporary materials, such as zeolites[2] or activat- stoichiometry but with different network superstructures; they
ed carbon compounds,[3] could not. A hefty catalogue of metal are known as structural supramolecular isomers.[8] Such supra-
ions and organic linkers makes it feasible to synthesize hybrid molecular isomers are expected to exhibit different porous
materials for versatile applications in gas storage and separa- properties based on their different network superstructures,
which are yet to be properly explored.[9] Furthermore, to the
[a] R. Haldar, Prof. T. K. Maji best of our knowledge, the synthesis of catenated supramolec-
Molecular Materials Laboratory ular isomers based on two different secondary building units
New Chemistry Unit (SBUs), and their porous properties as a result of structural flex-
Jawaharlal Nehru Centre for Advanced Scientific Research
ibility and rigidity in each isomer, are yet to be reported.
Jakkur, Bangalore 560064 (India)
In theory, the use of long linkers can enhance the surface
[b] Dr. S. K. Reddy, V. M. Suresh, Dr. S. Mohapatra, Prof. S. Balasubramanian,
Prof. T. K. Maji area of the MOF; in reality, this tactic often results in frame-
Chemistry and Physics of Materials Unit work interpenetration, which significantly reduces the porosity
Jawaharlal Nehru Centre for Advanced Scientific Research of the MOF. However, recent results suggest that interpenetrat-
Jakkur, Bangalore 560064 (India)
ed frameworks show better interactions with small molecules,
Fax: (+ 91) 80-22082766
E-mail: [email protected] such as H2 and CO2, through an entrapment mechanism, in
Supporting information for this article is available on the WWW under which the molecules are in close proximity with several aro-
http://dx.doi.org/10.1002/chem.201303610. matic rings from different interpenetrating nets.[10] Further-

Chem. Eur. J. 2014, 20, 4347 – 4356 4347  2014 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
 15213765, 2014, 15, Downloaded from https://chemistry-europe.onlinelibrary.wiley.com/doi/10.1002/chem.201303610 by Indian Institute Of Technology, Wiley Online Library on [26/08/2025]. See the Terms and Conditions (https://onlinelibrary.wiley.com/terms-and-conditions) on Wiley Online Library for rules of use; OA articles are governed by the applicable Creative Commons License
Full Paper

more, selective capture of CO2 from flue gases or natural gas

purification is of paramount importance in energy and environ-
mental contexts. The current amine-based technology of CO2
capture from flue gas has the major drawback of a high
energy cost for regeneration. Other related issues are the cor-
rosive nature, volatility, and occasional decomposition of
amines. An alternative approach is to functionalize the pore
surface of a solid adsorbent with less basic amines, for exam-
ple, aryl amines, which would provide strong physiosorption
(30–50 kJ mol1) with CO2 rather than chemisorption. An amine
(NH2) tethered/decorated MOF could give easy-on/easy-off
reversible CO2 uptake balanced with selectivity.[11] However,
such functionalization is not straightforward in practice be-
cause these groups can easily interfere with the coordination
chemistry associated with the assembly of nodes. To date,
there are only few reports of NH2-containing MOFs achieved Scheme 1. Schematic representation of the synthesis of structural (catenane)
supramolecular isomers 1 and 2 based on different SBUs.
by direct synthesis, among more than 10 000 MOFs reported.[12]
In addition to the incorporation of functionality, a stimulating
extension of such work is to increase the binding strength of
CO2 by modification of the pore surface. Extensive reports of Table 1. Crystal data and structure refinement parameters for 1 and 2.
postsynthetic modification of MOFs by Cohen et al. are good
Parameters 1 2
motivation to approach such challenges.[13] Inclusion of NH2
functionality on the pore surface also opens up the prospect empirical formula C17H19CdN4O6 C17H19CdN4O6
Mr 487.767 487.767
of base catalysis.[14] Knoevenagel condensation, a base-cata- crystal system orthorhombic orthorhombic
lyzed reaction, can be studied with the help of such materials space group Cmcm Cmca
and a literature survey reveals very few reports of this kind.[14] a [] 15.5918(15) 15.939(5)
Moreover, a multifunctional material apposite for storage as b [] 15.3726(14) 20.857(5)
c [] 18.2802(16) 14.374(5)
well as catalysis would be one of the rare examples of its V [3] 4381.5(7) 4779(3)
class.[15] To design materials with all the above-mentioned Z 8 8
measures is a stupendous task and a critical choice is impor- T [K] 293 293
tant. We envisioned that a framework that contained linkers 2- l (MoKa) 0.71073 0.71073
1calcd [g cm3] 1.287 1.183
aminobenzenedicarboxylic acid (NH2bdc) or 1,2-bis-(4-pyridyl- m [mm1] 1.022 0.937
methylene)hydrazine (bphz) could generate MOFs with a back- qmax [8] 24 25.8
bone decorated with basic functional groups (NH2 and total reflns 18 132 20 830
=NN=), which will afford interaction sites for CO2 and also unique reflns 1732 2084
Rint 0.139 0.092
create a platform for base catalysis. In this work, we have as- data [I > 2s(I)] 1201 1081
sembled NH2bdc and bphz ligands with CdII under two differ- R[a] 0.0781 0.0883
ent set of reaction conditions to yield two catenane supra- Rw[b] 0.1987 0.2349
molecular isomers, {[Cd(NH2-bdc)(bphz)0.5]·DMF·H2O}n (1 and 2). GOF 1.272 1.76

Both the frameworks are two-fold interpenetrated and show [a] R =  j j F0 j  j Fc j j / j F0 j , [b] Rw = [{w(F 20F 2c)2}/{w(F 20)2}]1/2.
SBU-modulated structural flexibility and rigidity (Scheme 1).
Compound 1 shows a double-step hysteretic CO2 adsorption
profile, whereas 2 exhibits a type I profile at 195 K. Both the groups Cmcm and Cmca, respectively (Table 1). Compound 1,
frameworks exhibit excellent selectivity for CO2 among other synthesized at 90 8C, has one CdII metal center, half a bphz and
gases (H2, N2, O2, and Ar). The high isosteric heat of adsorption one NH2bdc linker, along with one DMF and one H2O mole-
(DHads = 37.7 kJ mol1) in 2 is supported by DFT calculations, cule in the asymmetric unit (Figure 1 a). A paddle-wheel
which suggest that the NH2 group is the primary binding site Cd2(COO)4 SBU is the key component of the framework and
for CO2 and also demonstrate a cooperative interaction be- this is further connected by NH2bdc in the ab plane to furnish
tween the CO2 molecules. Further, 2 was exploited as a base a 2D square sheet-like structure (Figure 1 b). The 2D sheets are
catalyst in a Knoevenagel condensation reaction and very pillared by bphz linkers to extend the network in three dimen-
good efficiency and size selectivity were realized with different sions and provide a 3D a-polonium-type pillared-layer porous
substrates. framework (Figure 1 c). The octahedral CdII center deviates
slightly from a perfect octahedral geometry and this is reflect-
ed in the cisoid and transoid angles (see Table S1 in the Sup-
Results and Discussion
porting Information). The CdIIO bond length is 1.995  and
Single-crystal X-ray diffraction analysis illustrates that 1 and 2 CdIIN bond lengths are in the range of 2.022–2.036 . The
both crystallize in an orthorhombic crystal system with space pore size of the single net along the c axis is 8.4  8.4 2, which

Chem. Eur. J. 2014, 20, 4347 – 4356 www.chemeurj.org 4348  2014 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
 15213765, 2014, 15, Downloaded from https://chemistry-europe.onlinelibrary.wiley.com/doi/10.1002/chem.201303610 by Indian Institute Of Technology, Wiley Online Library on [26/08/2025]. See the Terms and Conditions (https://onlinelibrary.wiley.com/terms-and-conditions) on Wiley Online Library for rules of use; OA articles are governed by the applicable Creative Commons License
Full Paper

reduces to 4  4 2 due to a two-

fold interpenetration (see
Figure S1 in the Supporting In-
formation).[16] These channels are
occupied by two different guest
molecules, H2O and DMF. The
presence of these guests was
confirmed by elemental analysis,
TGA, IR, and 1H NMR spectrosco-
py (see Figures S2–S4 in the Sup-
porting Information). Removal of
the guest molecules creates
a void volume of 2129 3 (49 %
of the unit cell volume).[17] Along
with high void space, the pore
surfaces are decorated with
pendant NH2 groups.
Unlike 1, 2 is made of a differ-
ent SBU, Cd2(m-OCO)2, in which
two CdII centers are m-oxo-
bridged. The NH2bdc linkers
connect the SBUs along the bc
plane to furnish 2D sheets of
{Cd2(NH2bdc)2}n (Figure 2 a and
b). These 2D sheets are pillared
Figure 1. a) Coordination environment of CdII in 1 with a paddle-wheel SBU, b) 2D sheet of the {Cd2(NH2-bdc)2}n by bphz to generate a 3D bipil-
unit in the crystallographic ab plane, c) 3D framework of 1 with an a-polonium-type network topology. The two lared framework, in which the
NH2 groups in the NH2bdc linker are due to its positional disorder.
NH2 group remains pendant,

Figure 2. a) Coordination environment of CdII in 2 with a m-oxo bridge SBU, b) 2D sheet of the {Cd2(NH2-bdc)2}n unit in the crystallographic bc plane, c) 3D bi-
pillared layer framework 2.

Chem. Eur. J. 2014, 20, 4347 – 4356 www.chemeurj.org 4349  2014 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
 15213765, 2014, 15, Downloaded from https://chemistry-europe.onlinelibrary.wiley.com/doi/10.1002/chem.201303610 by Indian Institute Of Technology, Wiley Online Library on [26/08/2025]. See the Terms and Conditions (https://onlinelibrary.wiley.com/terms-and-conditions) on Wiley Online Library for rules of use; OA articles are governed by the applicable Creative Commons License
Full Paper

similar to framework 1 (Figure 2 c and Scheme 1). The CdII

metal center is seven coordinate; the CdIIO bond lengths are
in the range 2.282–2.658  and the CdIIN bond length is
2.468  (see Table S2 in the Supporting Information). A twofold
interpenetration creates rectangular channels along the c axis
with a dimension of 3  5 2 and one H2O and one DMF mole-
cule occupy the free space, as found for 1 (Scheme 1). The
presence of solvent molecules was confirmed by elemental
analysis, TGA, IR, and 1H NMR spectroscopy (see Figures S5–S7
in the Supporting Information). The calculated solvent-accessi-
ble volume was found to be about 28 %, which is less than for
1. A striking similarity between both structures was the pres-
ence of pendant NH2 groups along the pore surfaces. Both
the compounds have an identical molecular formula but have
different network superstructures comprised from different
SBUs, hence, they are structural supramolecular isomers
(Scheme 1). They are two-fold catenated and can also be
termed as catenane supramolecular isomers.
TGA of 1 (see Figure S3 in the Supporting Information)
shows a mass loss in the temperature range 150–200 8C, which
indicates the loss of DMF and water. The desolvated frame-
work {[Cd(NH2bdc)(bphz)0.5]}n (1’) is stable up to 250 8C with-
out further loss of mass. Upon further heating, a rapid mass
loss indicates decomposition of the compound. In the case of
2, the guest molecules are released at 200 8C (see Figure S5 in
the Supporting Information) and the desolvated structure (2’)
is stable up to 280 8C. Upon desolvation at 200 8C under
vacuum, 1 undergoes structural transformation, which is evi- Figure 3. PXRD patterns of a) 1 and b) 2.
dent from the differences between the PXRD patterns of 1’
and 1 (Figure 3 a). First, two peaks that correspond to reflec-
tions (110) and (111) are almost absent in the desolvated state (N3H···N1 = 3.253 ). Two other weak hydrogen bonds oper-
1’, which suggests contraction in the framework. When the ate between the CH=N hydrogen atom of one net and a car-
pores were refilled exclusively with DMF further change was boxylate oxygen atom of a second net (C14H···O1 = 3.216 ,
noted in the PXRD pattern but it does return to its as-synthe- C14H···O1 = 3.402 ). Due to these interactions, the two 3D
sized phase (Figure 3 a). However, the PXRD pattern of desol- nets are stabilized and do not move upon desolvation. But in
vated framework 2’ does not show any change and also re- the case of 1, no such interactions exist between the nets and
tains its structural integrity upon exposure to solvent
vapors (MeOH, EtOH, CHCl3, and CH3CN) [Figure 3 b;
Figure S8 in the Supporting Information]. The struc-
tural change associated with framework 1 upon de-
solvation and the rigid nature of 2 can be explained
on the basis of the different SBUs and their orienta-
tions in the interpenetrated nets. In 1, the axis that
passes through the metal centers of the paddle-
wheel SBU is parallel to the bphz pillars and NH2
bdc plane. This orientation leaves space along the
c axis (Figure 4 a and b), hence, there is a possibility
of net movement upon desolvation, which results in
structural transformation of 1. In the case of 2 this
sliding is not possible due to perfect fitting of the
nets along the a axis; the axis that passes through
the metal centers in the m-oxo bridged SBU is per-
pendicular to the pillars and, in addition, the bipillar
nature leaves no space along the a axis (Figure 4 c
and d). Furthermore, in 2 the pendant NH2 (donor)
of one net and =NN= (acceptor) of a second net are Figure 4. View of 1 along the c axis: a) single net, b) interpenetrated net. View of 2 along
connected by hydrogen bonding interactions the a axis: c) single net, d) interpenetrated net.

Chem. Eur. J. 2014, 20, 4347 – 4356 www.chemeurj.org 4350  2014 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
 15213765, 2014, 15, Downloaded from https://chemistry-europe.onlinelibrary.wiley.com/doi/10.1002/chem.201303610 by Indian Institute Of Technology, Wiley Online Library on [26/08/2025]. See the Terms and Conditions (https://onlinelibrary.wiley.com/terms-and-conditions) on Wiley Online Library for rules of use; OA articles are governed by the applicable Creative Commons License
Full Paper

space for sliding of the nets is available. Hence, the structural

flexibility in 1 and rigidity in 2 are realized.
Large void spaces and the amine-functionalized pore in
both 1 and 2 led us to perform different gas adsorption meas-
urements. Prior to measurement, 1 was activated to 1’ by re-
moval of the guest molecules at 190 8C. At 77 K, 1’ shows a typ-
ical type II N2 isotherm (see Figure S9 in the Supporting Infor-
mation), which indicates surface-adsorption phenomenon. The
calculated BET surface area is 82 m2 g1. To our surprise, at
195 K 1’ shows a two-step commensurate uptake profile for
CO2 and the saturated uptake amount is approximately
118 mL g1, which corresponds to 23 wt % (2  CO2 molecules/
formula; Figure 5). MOF 1’ also shows unique hysteresis in the

Figure 6. a) Adsorption isotherm for 2’ at 195 K: CO2 adsorption (half-filled

diamonds), CO2 desorption (open diamonds), N2 (stars), H2 (filled diamonds),
Ar (squares), O2 (squares). b) CO2 adsorption profile at 273 and 293 K. Inset:
plot of heat of adsoption (DHads) versus CO2 loading.

Figure 5. Adsorption isotherms for 1’ at 195 K: CO2 adsorbtion (half-filled

diamonds), CO2 desorption (open diamonds), N2 (stars), H2 (circles), Ar (trian-
gles), O2 (squares); Inset: CO2 adsorption profile at 195 K (log scale). was found to be approximately 298 m2 g1. The enthalpy of ad-
sorption of CO2 was obtained by measuring two isotherms at
T = 273 and 293 K, followed by a fit of the data to virial equa-
adsorption–desorption isotherm, withholding about 9 wt % of tion.[21] At zero loading, DHads = 37.7 kJ mol1, which clearly in-
CO2 in the framework. With a flexible framework this kind of dicates moderate interaction between framework 2 and CO2.
unusual adsorption behavior is not uncommon.[18] The first Remarkably, at maximum loading, DHads = 39.6 kJ mol1. After
step of the uptake (until pressure P/P0  0.5) equates to one initial loading, DHads falls slowly and rises again to a value that
molecule of CO2 per formula and in the next step another CO2 is higher than that at zero loading (Figure 6 b and Figure S11 in
molecule is adsorbed. Initial steep uptake (type I) is probably the Supporting Information). This is likely due to a cooperative
due to interaction with the pendent NH2 groups present at interaction between the adsorbed CO2 molecules. Similar to
the pore surface and is reflected in the heat of adsorption compound 1, 2 is also found to be selective for CO2 at 195 K
(qst,F) value of approximately 28.5 kJ mol1, calculated by using among other gases (H2, Ar, O2, and N2).
the Dubinin–Radushkevich (DR) equation.[19] The second step is After removal of the guest molecules, 1 undergoes structural
probably due to sliding of the 3D nets, which opens up more transformation. We do not have any structural information for
void space for CO2 molecules to diffuse in. From the CO2 the desolvated phase 1’, therefore DFT calculations would not
uptake profile (until the first step, P/P0  0.5) the calculated BET provide accurate insight into the CO2 adsorption phenomena.
surface area is 185 m2 g1. Interestingly, 1 showed negligible Rather, we chose the rigid framework 2, to carry out DFT calcu-
uptake for other gases [kinetic diameter: H2 (2.83 ),[20] Ar lations to find favorable binding sites. To find these sites, gas
(3.5 ), O2 (3.46 ), N2 (3.64 )] at 195 K and, thus, is very selec- molecules were initially placed randomly inside 2’, the desol-
tive for CO2 gas. Compound 2, like 1, shows no uptake of N2 at vated framework of 2. Geometry optimization of the gas-
77 K (see Figure S10 in the Supporting Information), but at loaded framework was carried out. Two favorable locations for
195 K, it shows a type I adsorption profile for CO2 and the final the CO2 molecules, CO2-I and CO2-II, were obtained. The former
uptake amount is 101 mL g1, which corresponds to approxi- is near the NH2 group and its molecular axis passes through
mately 19 wt % of CO2 (1.8 molecules/formula) [Figure 6 a]. The the bisector of the amine at a distance of 3.10  (Figure 7). The
BET surface area, calculated from the CO2 adsorption profile, backbone angle of CO2-I marginally deviates from linearity

Chem. Eur. J. 2014, 20, 4347 – 4356 www.chemeurj.org 4351  2014 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
 15213765, 2014, 15, Downloaded from https://chemistry-europe.onlinelibrary.wiley.com/doi/10.1002/chem.201303610 by Indian Institute Of Technology, Wiley Online Library on [26/08/2025]. See the Terms and Conditions (https://onlinelibrary.wiley.com/terms-and-conditions) on Wiley Online Library for rules of use; OA articles are governed by the applicable Creative Commons License
Full Paper

tron-deficient carbon atom of CO2-II loses some electronic

charge to the aromatic ring, and 2) weak hydrogen bonding
between the hydrogen atoms of the aromatic ring and oxygen
atoms of CO2-II. Recently, such aromatic p···CO2 interactions in
MOFs have been established by single-crystal X-ray diffraction
studies.[22]
The two CO2 molecules are also stabilized by a cooperative
effect, evident from the short distance (3.76 ) between the
oxygen atom of CO2-I and carbon atom of CO2-II. This distance
is comparable to that observed between two neighboring CO2
molecules in supercritical CO2.[23] The angle between the back-
bone vectors of the two CO2 molecules in the MOF is 59.58,
whereas it is 708 in crystalline CO2. This observation suggests
that the dimer inside the MOF is in a distorted T-shaped con-
figuration, similar to that found in the crystalline phase and su-
percritical CO2.[23] The interaction energy of the CO2 molecules
Figure 7. a) Cooperativity between two adsorbed CO2 molecules within MOF with 2’ was compared to the experimental heat of adsorption
2’. Carbon atom of CO2-I interacts with the nitrogen lone pair of the amino values. The CO2-I and CO2-II molecules interact with the frame-
group and CO2-II interacts with the p cloud of the benzene ring. C (cyan), H
(gray), N (blue), O (red); b) View of the CO2 molecules inside one pore of 2’.
work individually and give binding energies of 36.03 and
30.03 kJ mol1, respectively. When both these two molecules
are present in 2’, the mean binding energy per molecule is cal-
(177.78), which suggests a specific interaction between NH2 culated to be 33.94 kJ mol1. At low pressures, from the Boltz-
and CO2-I. Electron-density-difference maps (i.e., the difference mann ratio, the most preferable binding site is the one that
in electron density between the MOF with the guest gas mole- provides a strong affinity for CO2. As a result CO2 binds prefer-
cule and that of the individual components minus pure MOF entially near an amino group (CO2-I; E = 36.03 kJ mol1), which
and isolated gas molecule) over the nitrogen atom of NH2 is comparable to that calculated from adsorption isotherms by
and the carbon atom of CO2-I indicates a moderate interaction using the virial equation (E = 37.7 kJ mol1). Notably, although
of the quadrupole moment of CO2-I with the lone pair of the there is evident proof of NH2···CO2 interaction and a moderate
NH2 group (Figure 8 a). A positive-difference map over one of DHads value, we do not observe any distinct hysteresis in the
the two oxygen atoms of CO2 indicates a weak hydrogen bond adsorption isotherm. As it is a case of mere physiosorption,
formed with hydrogen atoms of the bphz ligand. The distances hysteresis is unlikely to occur[11b, d, e] unless there is any structur-
between O and H are 2.59, 2.79, 2.89, and 3.49 . The second al transformation (rearrangement), which can act as barrier for
CO2 molecule (CO2-II) is found to be present on top of bphz gas molecules to diffuse out of the pores. In this case, 2 is
with its center at the midpoint between the aromatic ring and rigid in nature and the observations found here align with the
amino group, which suggests aromatic p interaction with CO2 literature reports.[11b, d, e]
(Figure 7). The distance between this midpoint and the carbon We could visualize the role of the NH2 functional group
atom of CO2-II is 3.63 . CO2-II also interacts electrostatically from the DFT calculations; therefore, we also carried out an in-
with the hydrogen atoms of the aromatic ring of bphz. The direct experiment to establish its immense importance. We
distances between the oxygen atom of CO2-II and the hydro- modified the NH2 group to a NHCOCH3 group in a postsyn-
gen atoms are 2.64, 2.89, and 3.13 . The electron-density-dif- thetic method and as a result the polarity of the pore surface
ference map for this molecule (Figure 8 b) reveals two types of was altered. We obtained about 90 % conversion of 2 to
interaction: 1) Lewis acid–base interaction, in which the elec- {[Cd(NH2bdc)0.1(NHCOCH3bdc)0.9(bphz)0.5]·CHCl3)}n (2 a), con-
firmed by 1H NMR spectroscopy (see Figure S12 in the Support-
ing Information). PXRD patterns of 2 a showed no distinct
change relative to the parent compound 2, which suggests
that the structure remains intact (Figure 9 a). The CO2 adsorp-
tion isotherm at 195 K showed interesting changes relative to
the parent compound 2 (Figure 9 b). The adsorption curve
clearly indicates that up to a relative pressure (P/P0) of approxi-
mately 0.1 the uptake is not as sharp as for 2, but in the high
pressure range CO2 molecules enter the pores and the uptake
reaches a saturation of about 9 wt %. Heat of adsorption calcu-
Figure 8. Electron-density-difference maps for a) CO2-I and b) CO2-II interact- lations from the DR equation show qst,F is approximately
ing with MOF 2’, presented at isosurface values of 0.005 and 0.004 a.u., re- 28 kJ mol1 and the BET surface area is 98 m2g1; both values
spectively. Only a portion of the MOF is shown for clarity. CO2 molecules
are lesser than for 2. Evidently absence of the strong Lewis
form weak hydrogen bonds with the MOF. Electron-rich and -deficient re-
gions are shown in green and pink, respectively. C (cyan), H (gray), N (blue), base NH2 and less available space accounted for the low
O (red). heat of adsorption and smaller uptake.

Chem. Eur. J. 2014, 20, 4347 – 4356 www.chemeurj.org 4352  2014 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
 15213765, 2014, 15, Downloaded from https://chemistry-europe.onlinelibrary.wiley.com/doi/10.1002/chem.201303610 by Indian Institute Of Technology, Wiley Online Library on [26/08/2025]. See the Terms and Conditions (https://onlinelibrary.wiley.com/terms-and-conditions) on Wiley Online Library for rules of use; OA articles are governed by the applicable Creative Commons License
Full Paper

Table 2. Substrates and conditions for reactions catalyzed by 2’.[a]

Entry R1 R2 T Solvent t Conversion[b]

[8C] [h] [%]
1 NO2 CN 60 CH3CN 6 99
2 NO2 CN 60 CH3OH 6 85
8 98
3 NO2 CN 40 CH3CN 6 81
9 99
4 NO2 CH3CO2 60 CH3OH 6 41
12 52
5 Cl CN 60 CH3CN 6 99
6 NO2 CN 60 CH3CN[c] 12 0

[a] In all cases, catalyst (0.05 equiv) relative to nitrile substrate (1 equiv)
was used. [b] Calculated by GCMS. [c] No catalyst was used.
Figure 9. a) PXRD patterns of compounds 2 and 2 a, b) CO2 adsorption–
desorption isotherm of postsynthetically modified 2 (2 a) at 195 K.

The presence of pendent NH2 groups, structural rigidity,

and high thermal stability of 2 encouraged us to carry out
a base-catalyzed Knoevenagel condensation reaction. Prior to
the catalytic reaction, we activated the catalyst by removal of
the guest DMF and water molecules from the pore, then car-
ried out the condensation of 4-nitro- and 4-chloro-benzalde-
hyde derivatives with malononitrile and a range of solvents at
60 8C (Table 2; see Figures S13 and S14 in the Supporting Infor-
mation). When malononitrile and 4-nitrobenzaldehyde were
used, we observed about 99 % conversion within 6 h in aceto-
nitrile (Table 2, entry 1) but only 85 % conversion was observed
after 6 h in methanol (Table 2, entry 2). A change from aprotic
to protic solvent hampered the reaction rate. The reaction ki-
netics of malononitrile and 4-nitrobenzaldehyde show very fast
conversion within 2 h and about 99 % conversion was reached Figure 10. Conversion [%] versus time [min] for Knoevenagel condensation
in 6 h (Figure 10). A decrease in the reaction temperature to reaction of 4-nitrobenzaldehyde and malononitrile at 60 8C in acetonitrile.
Inset: PXRD patterns of 2 a) before and b) after six catalytic cycles.
40 8C decreased the conversion rate (Table 2, entry 3). We also
observed excellent size selectivity by the catalyst; when the
larger substituent methyl cyanoacetate was present, the con- nitrile and 4-nitrobenzaldehyde were mixed without any cata-
version was only 41 or 52 % after 6 and 12 h, respectively lyst at 60 8C in acetonitrile no product was formed (Table 2,
(Table 2, entry 4; see Figure S15 in the Supporting Information). entry 6).
We have performed six catalytic cycles to check the efficiency
of the catalyst and we observed no change in the conversion
Conclusion
rate. Similar catalytic reactions with (CONH)-functionalized
MOFs have been reported by Kitagawa et al.[14a] and Zhou We have synthesized two isomeric catenane supramolecular
et al.[14c] These reactions show much slower conversion time complexes of CdII with 3D porous structures. Compound 1 con-
(12 h) than our results. This is due to better diffusion of the tains a paddle-wheel SBU and shows structural flexibility upon
reactants into the pores but this higher diffusion of reactants guest removal. Compound 2 is rigid in nature with a m-oxo
also decreases the size selectivity in the present case com- bridged SBU. The origin of rigidity and flexibility arises from
pared with the report by Kitagawa et al. The PXRD pattern of the orientation of the two interpenetrated nets in the frame-
the catalyst after six catalytic cycles was found to remain the work, controlled by the different SBUs. Both frameworks show
same as that of the as-synthesized catalyst, which suggests selective CO2 uptake. Compound 1 shows a two-step uptake
high thermal and chemical stability of the framework profile with large hysteresis, which emphasizes the flexible
(Figure 10, inset). Notably, a reaction performed in the pres- nature of the framework. In contrast, 2 shows a type I uptake
ence of NH2bdc resulted in a negligible yield. When malono- profile with moderate heat of adsorption (DHabs =

Chem. Eur. J. 2014, 20, 4347 – 4356 www.chemeurj.org 4353  2014 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
 15213765, 2014, 15, Downloaded from https://chemistry-europe.onlinelibrary.wiley.com/doi/10.1002/chem.201303610 by Indian Institute Of Technology, Wiley Online Library on [26/08/2025]. See the Terms and Conditions (https://onlinelibrary.wiley.com/terms-and-conditions) on Wiley Online Library for rules of use; OA articles are governed by the applicable Creative Commons License
Full Paper

37.7 kJ mol1). DFT calculations further describe the NH2···CO2 C17H19CdN4O6 : C 41.86, H 3.92, N 11.48; found: C 42.11, H 4.17, N
and CO2···CO2 cooperative interactions in the framework. The 11.01.
importance of the NH2 group towards CO2 capture was fur-
ther realized through a postsynthetic modification of NH2 to Synthesis of 2 a
NHCOCH3. This modification drastically reduces the uptake of
Activated (desolvated) 2 (2’;  50 mg) was stirred with a solution
CO2 and also the binding energy. Compound 2 also acts as of acetic anhydride (5 mL) in chloroform (0.5 mmol) for 6–7 d at rt.
a very efficient base catalyst for a Knoevenagel condensation Compound 2 a was filtered and washed several times with CHCl3,
reaction and shows size selectivity. The present study is signifi- then dried in air. Compound 2 a was characterized by PXRD and
1
cant because it displays an elegant example of structural H NMR spectroscopy (Figure 9 b; Figure S12 in the Supporting In-
supramolecular isomerism with multifunctional properties, formation). FTIR (KBr pellet): ñ = 3498 (br), 3240 (w), 3091 (w), 2329
such as selective CO2 uptake and catalysis. (s), 2241 (m), 2133 (m), 1615 (s), 1569 (s), 1507 (s), 1425 (s), 1311
(w), 1232 (s), 1069 (s), 1021 (s), 915 (w), 877 (w), 831 cm1 (s); ele-
mental analysis calcd (%) for C16.8H12.8CdN4O4.9 : C 36.41, H 2.31, N
10.11; found: C 37.51, H 3.01, N 10.51.
Experimental Section
Materials X-ray crystallography
All the reagents employed were commercially available and used X-ray single-crystal structural data of 1 and 2 were collected with
as provided without further purification. [Cd(NO3)2·6H2O] was ob- a Bruker Smart-CCD diffractometer equipped with a normal focus,
tained from Spectrochem. 2-Aminobenzenedicarboxylic acid, acetic 2.4 kW sealed-tube X-ray source with graphite monochromated
anhydride, malononitrile, methyl cyanoacetate, p-nitro- and p- MoKa radiation (l = 0.71073 ), operating at 50 kV and 30 mA. The
chloro-benzaldehyde were obtained from Sigma Aldrich. 1,2-Bis-(4- program SAINT[25] was used for integration of diffraction profiles
pyridylmethylene)hydrazine was synthesized by a literature and absorption correction was made with the SADABS[26] program.
method.[24] All the structures were solved by SIR 92[27] and refined by the full-
matrix least-squares method by using SHELXL-97.[28] All the hydro-
gen atoms were fixed by HFIX and placed in ideal positions. Due
Physical measurements to the weakly diffracting nature of the single crystals we could not
Elemental analysis was carried out with a Thermo Fischer Flash locate the guest molecules and carried out 1H NMR and IR spec-
2000 Elemental Analyzer. IR spectra were recorded as KBr pellets troscopy and TGA and elemental analysis to confirm the presence
with a Bruker IFS 66v/S spectrophotometer in the region ñ = 4000– of guest molecules. Potential solvent-accessible areas or void space
400 cm1. TGA was carried with on a Metler Toledo instrument was calculated by using the PLATON multipurpose crystallographic
under a nitrogen atmosphere (flow rate = 50 mL min1) in the tem- software.[17] All crystallographic and structure-refinement data for
perature range 30–500 8C (heating rate = 3 8C min1). PXRD patterns 1 and 2 are summarized in Table 1. Selected bond lengths and
of 1 and 2 were recorded with a Bruker D8 Discover instrument angles for 1 and 2 are given in Tables S1 and S2 (see the Support-
(40 kV, 30 mA) by using CuKa radiation. 1H NMR spectra were mea- ing Information), respectively. All calculations were carried out by
sured on a Bruker AV-400 spectrometer with chemical shifts report- using PLATON and the WinGX system, Ver 1.70.01.[29] CCDC-941967
ed as ppm (in DCl/[D6]DMSO or CDCl3/CD3CN; TMS as internal stan- (1) and 941968 (2) contain the supplementary crystallographic data
dard). for this paper. These data can be obtained free of charge from The
Cambridge Crystallographic Data Centre via www.ccdc.cam.ac.uk/
data_request/cif.
Synthesis of 1 and 2
[Cd(NO3)2·6H2O] (0.2 mmol, 0.049 g) was added to a solution of Adsorption study
NH2bdc (0.2 mmol, 0.035 g) and bphz (0.2 mmol, 0.042 g) in DMF The adsorption isotherms of CO2, H2, N2, Ar, and O2 (T = 195 K) mea-
(14 mL) and the mixture was sonicated for 10–15 min. This solution sured from the dehydrated samples of 1 (1’), 2 (2’), and 2 a were
was then equally divided into two glass vials; one was kept at measured with a QUANTACHROME QUADRASORB-SI analyzer. The
90 8C and one at 120 8C, for 24 h. After cooling to rt, rectangular adsorbent sample (100–150 mg), which had been prepared at
block-shaped yellow crystals were found at the walls of the vials. 453 K under a 1  101 Pa vacuum for about 6 h prior to measure-
We obtained 1 from the reaction at 90 8C and 2 from the reaction ment of the isotherms, was placed in the sample tube. Helium gas
at 120 8C. Crystals were separated from the solution and then thor- (99.999 % purity) at a known pressure was introduced into the gas
oughly washed with DMF. Crystals of both compounds were found chamber and allowed to diffuse into the sample chamber by open-
to be weakly diffracting and, hence, for further characterization we ing the valve. The amount of gas adsorbed was calculated readily
used 1H NMR spectroscopy, elemental analysis, and TGA. from the pressure difference (PcalPe); Pcal is the calculated pressure
Compound 1: Yield: 67 %, relative to Cd. FTIR (KBr pellet): ñ = 3490 with no gas adsorption, Pe is the observed equilibrium pressure. All
(br), 3234 (w), 3072 (w), 2326 (s), 2238 (m), 2128 (m), 1614 (s), 1562 operations were computer controlled and automatic.
(s), 1502 (s), 1428 (s), 1310 (w), 1228 (s), 1068 (s), 1020 (s), 912 (w),
874 (w), 820 cm1 (s); elemental analysis calcd (%) for
C17H19CdN4O6 : C 41.86, H 3.92, N 11.48; found: C 41.21, H 4.41, N Computational details
10.97. All periodic calculations were carried out by DFT as implemented
Compound 2: Yield: 58 %, relative to Cd. FTIR (KBr pellet): ñ = 3490 in the QUICKSTEP module in CP2K.[30] A mixed basis set was used,
(br), 3234 (w), 3072 (w), 2326 (s), 2238 (m), 2128 (m), 1614 (s), 1562 in which Kohn–Sham orbitals were expanded in an atom-centered
(s), 1502 (s), 1428 (s), 1310 (w), 1228 (s), 1068 (s), 1020 (s), 912 (w), Gaussian basis set and electronic charge density was described by
874 (w), 820 cm1 (s); elemental analysis calcd (%) for using an auxiliary-plane wave basis set. Initial coordinates were

Chem. Eur. J. 2014, 20, 4347 – 4356 www.chemeurj.org 4354  2014 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
 15213765, 2014, 15, Downloaded from https://chemistry-europe.onlinelibrary.wiley.com/doi/10.1002/chem.201303610 by Indian Institute Of Technology, Wiley Online Library on [26/08/2025]. See the Terms and Conditions (https://onlinelibrary.wiley.com/terms-and-conditions) on Wiley Online Library for rules of use; OA articles are governed by the applicable Creative Commons License
Full Paper

taken from the experimental structure and then the cell parame- [5] a) J. Y. Lee, O. K. Farha, J. Roberts, K. A. Scheidt, S. T. Nguyen, J. T. Hupp,
ters and geometry were optimized iteratively until the lowest- Chem. Soc. Rev. 2009, 38, 1450; b) R. K. Das, A. Aijaz, M. K. Sharma, P.
energy structure was found. A super cell of size 1  1  1 was used. Lama, P. K. Bharadwaj, Chem. Eur. J. 2012, 18, 6866.
[6] a) M. Kim, J. F. Cahill, H. Fei, K. A. Prather, S. M. Cohen, J. Am. Chem. Soc.
Goedecker–Teter–Hutter-type norm-conserving pseudo potentials
2012, 134, 18082; b) T. K. Maji, R. Matsuda, G. Mostafa, S. Kitagawa, Nat.
were used to represent the combined effect of the core electrons
Mater. 2007, 6, 142; c) T. K. Prasad, D. H. Hong, M. P. Suh, Chem. Eur. J.
and nucleus.[31] Exchange and correlation interactions were ac- 2010, 16, 14043.
counted for by using the generalized gradient approximation [7] a) B. Chen, S. Xiang, G. Qian, Acc. Chem. Res. 2010, 43, 1115; b) M. D. Al-
(GGA) of Perdew, Burke, and Ernzerhof (PBE).[32] Long-range interac- lendorf, C. A. Bauer, R. K. Bhakta, R. J. T. Houk, Chem. Soc. Rev. 2009, 38,
tions were taken into account by using Grimme’s dispersion cor- 1330; c) R. Haldar, K. V. Rao, S. J. George, T. K. Maji, Chem. Eur. J. 2012,
rections for this functional.[33] Double-zeta single-polarized molecu- 18, 5848; d) K. Jayaramulu, P. Kanoo, S. J. George, T. K. Maji, Chem.
larly optimized basis sets (DZVP) were used to describe the valence Commun. 2010, 46, 7906; e) C. A. Kent, B. P. Mehl, L. Ma, J. M. Papaniko-
electrons. Because this basis set is not available for cadmium, the las, T. J. Meyer, W. Lin, J. Am. Chem. Soc. 2010, 132, 12767.
shorter-range version of the same type was used. Structural relaxa- [8] a) P. Kanoo, K. L. Gurunatha, T. K. Maji, Cryst. Growth Des. 2009, 9, 4147;
b) P. Kanoo, K. L. Gurunatha, T. K. Maji, J. Mater. Chem. 2010, 20, 1322;
tion was carried out until forces on all atoms were less than
c) J. J. Perry IV, J. A. Perman, M. J. Zaworotko, Chem. Soc. Rev. 2009, 38,
104 a.u. Binding energy (DE) of CO2 with the MOF was calculated 1400; d) J. P. Zhang, S. Kitagawa, J. Am. Chem. Soc. 2008, 130, 907.
by using the formula given in [Eq (1)], [9] a) P. K. Thallapally, J. Tian, M. R. Kishan, C. A. Fernandez, S. J. Dalgarno,
P. B. McGrail, J. E. Warren, J. L. Atwood, J. Am. Chem. Soc. 2008, 130,
16842; b) K. L. Mulfort, O. K. Farha, C. D. Malliakas, M. G. Kanatzidis, J. T.
DE ¼ EðMOFþCO2 ÞEðMOFÞEðCO2 Þ ð1Þ Hupp, Chem. Eur. J. 2010, 16, 276; c) J. H. Wang, M. Li, D. Li, Chem. Sci.
2013, 4, 1793.
in which, E(MOF+CO2), E(MOF), and E(CO2) represent the energy of [10] a) R. Haldar, T. K. Maji, CrystEngComm 2012, 14, 684; b) W. Yang, A. J.
the MOF with CO2, the MOF, and CO2, respectively. All structures Davies, X. Lin, M. Suyetin, R. Matsuda, A. J. Blake, C. Wilson, W. Lewis,
J. E. Parker, C. C. Tang, M. W. George, P. Hubberstay, S. Kitagawa, H. Sa-
were visualized by using visual molecular dynamics (VMD).[34]
kamoto, E. Bichoutskaia, N. R. Champness, S. Yang, M. Schroder, Chem.
Sci. 2012, 3, 2993; c) P. Kanoo, R. Matsuda, M. Higuchi, S. Kitagawa, T. K.
Maji, Chem. Mater. 2009, 21, 5860; d) H. L. Jiang, T. Makal, H. C. Zhou,
Acknowledgements Coord. Chem. Rev. 2013, 257, 2232.
[11] a) K. Sumida, D. L. Rogow, J. A. Mason, T. M. McDonald, E. D. Bloch, Z. R.
We thank Professor C. N. R. Rao for his support and encourage- Herm, T. H. Bae, J. R. Long, Chem. Rev. 2012, 112, 724; b) R. Vaidhyana-
ment. R.H. acknowledges DST, Government of India and than, S. S. Iremonger, G. K. H. Shimizu, P. G. Boyd, S. Alavi, T. K. Woo, Sci-
ence 2010, 330, 650; c) C. M. Nagaraja, R. Haldar, T. K. Maji, C. N. R. Rao,
JNCASR, Bangalore. M.V.S., S.M., and S.K.R. acknowledge CSIR, Cryst. Growth Des. 2012, 12, 975; d) W. Lu, J. P. Sculley, D. Yuan, R. Krish-
India for senior research fellowships. S.B. acknowledges the na, Z. Wei, H. C. Zhou, Angew. Chem. 2012, 124, 7598; Angew. Chem. Int.
Centre for Development of Advanced Computing (C-DAC), Ed. 2012, 51, 7480; e) T. M. McDonald, W. R. Lee, J. A. Mason, B. M. Wiers,
Bangalore for provision of computational resources. T.K.M. ac- C. S. Hong, J. R. Long, J. Am. Chem. Soc. 2012, 134, 7056; f) T. M. McDo-
nald, D. M. D’Alessandro, R. Krishna, J. R. Long, Chem. Sci. 2011, 2, 2022.
knowledges Sheikh Saqr fellowship. [12] a) T. Ahnfeldt, N. Guillou, D. Gunzelmann, I. Margiolaki, T. Loiseau, G.
Frey, J. Senker, N. Stock, Angew. Chem. 2009, 121, 5265; Angew. Chem.
Int. Ed. 2009, 48, 5163; b) X. Y. Chen, B. Zhao, W. Shi, J. Xia, P. Cheng,
Keywords: adsorption · catenanes · flexible framework ·
D. Z. Liao, S. P. Yan, Z. H. Jiang, Chem. Mater. 2005, 17, 2866; c) A. L.
Knoevenagel condensation · metal–organic frameworks Grzesiak, F. J. Uribe, N. W. Ockwig, O. M. Yaghi, A. J. Matzger, Angew.
Chem. 2006, 118, 2615; Angew. Chem. Int. Ed. 2006, 45, 2553; d) X. Si, C.
[1] a) S. Kitagawa, R. Kitaura, S. I. Noro, Angew. Chem. 2004, 116, 2388; Jiao, F. Li, J. Zhang, S. Wang, S. Liu, Z. Li, L. Sun, F. Xu, Z. Gabelica, C.
Angew. Chem. Int. Ed. 2004, 43, 2334; b) O. M. Yaghi, Nat. Mater. 2007, 6, Schick, Energy Environ. Sci. 2011, 4, 4522; e) Y. Zhao, H. Wu, T. J. Emge,
92. Q. Gong, N. Nijem, Y. S. Chabal, L. Kong, D. C. Langreth, H. Liu, H. Zeng,
[2] a) A. Zukal, I. Dominguez, J. Mayerov, J. Cezka, Langmuir 2009, 25, J. Li, Chem. Eur. J. 2011, 17, 5101; f) Z. Zhang, Y, Zhao, Q, Gong, J. Li,
10314; < lit b > R. V. Siriwardane, M. S. Shen, E. P. Fisher, J. Losch, Energy Chem. Commun. 2013, 49, 653.
Fuels 2005, 19, 1153. [13] a) S. M. Cohen, Chem. Soc. Rev. 2012, 112, 970; b) K. K. Tanabe, C. Allen,
[3] a) Y. Belmabkhout, A. Sayari, Adsorption 2009, 15, 318; b) S. Kim, J. Ida, S. M. Cohen, Angew. Chem. 2010, 122, 9924; Angew. Chem. Int. Ed. 2010,
V. V. Guliants, J. Y. S. lin, J. Phys. Chem. B 2005, 109, 6287; c) S. N. Kim, 49, 9730; c) K. K. Tanabe, S. M. Cohen, Chem. Soc. Rev. 2011, 40, 498.
W. J. Son, J. S. Choi, W. S. Ahn, Microporous Mesoporous Mater. 2008, [14] a) S. Hasegawa, S. Horike, R. Matsuda, S. Furukawa, K. Mochizuki, Y. Ki-
115, 497. noshita, S. Kitagawa, J. Am. Chem. Soc. 2007, 129, 2607; b) Y. K. Hwang,
[4] a) R. Banerjee, A. Phan, B. Wang, C. Knobler, H. Furukawa, M. O’Keeffe, D. Y. Hong, J. S. Chang, S. H. Jhung, Y. K. Seo, J. Kim, A. Vimont, M.
O. M. Yaghi, Science 2008, 319, 939; b) H. Hayashi, A. P. Ct, H. Furuka- Daturi, C. Serre, G. Frey, Angew. Chem. 2008, 120, 4212; Angew. Chem.
wa, M. O’Keeffe, O. M. Yaghi, Nat. Mater. 2007, 6, 501; c) J. A. R. Navarro, Int. Ed. 2008, 47, 4144; c) J. Park, J. R. Li, Y. P. Chen, J. Yu, A. A. Yakoven-
E. Barea, A. Rodrguez-Diguez, J. M. Salas, C. O. Ania, J. B. Parra, N. Mas- ko, Z. U. Wang, L.-B. Sun, P. B. Balbuena, H. C. Zhou, Chem. Commun.
ciocchi, S. Galli, A. Sironi, J. Am. Chem. Soc. 2008, 130, 3978; d) F. Deba- 2012, 48, 9995.
tin, A. Thomas, A. Kelling, N. Heddin, Z. Bacsik, I. Senkovska, S. Kaskel, [15] a) A. Hazra, P. Kanoo, G. Mostafa, T. K. Maji, Chem. Commun. 2011, 47,
M. Junginger, H. M ller, U, Schilde, C. J ger, A. Friedrich, H.-J. Holdt, 538; b) C. D. Wu, A. Hu, L. Zhang, W. Lin, J. Am. Chem. Soc. 2005, 127,
Angew. Chem. 2010, 122, 1280; Angew. Chem. Int. Ed. 2010, 49, 1258; 8940; c) S. Su, Y. Zhang, M. Zhu, X. Song, S. Wang, S. Zhao, S. Song, X.
e) J. An, S. J. Geib, N. L. Rosi, J. Am. Chem. Soc. 2010, 132, 38; f) R. Vaid- Yang, H. Zhang, Chem. Commun. 2012, 48, 11118; d) R. Q. Zou, H. Sakur-
hyanathan, S. S. Iremonger, K. W. Dawson, G. K. H. Shimizu, Chem. ai, Q. Xu, Angew. Chem. 2006, 118, 2604; Angew. Chem. Int. Ed. 2006, 45,
Commun. 2009, 5230; g) C. A. Fernandez, J. Liu, P. K. Thallapally, D. M. 2542.
Strachan, J. Am. Chem. Soc. 2012, 134, 9046; h) L. Pan, D. H. Olson, L. R. [16] The sizes of the channels were calculated by considering the van der
Ciemnolonski, R. Heddi, J. li, Angew. Chem. 2006, 118, 632; Angew. Waals radii of the atoms.
Chem. Int. Ed. 2006, 45, 616; i) J. Y. Lee, D. H. Olson, L. Pan, T. J. Emge, J. [17] A. L. Spek, J. Appl. Crystallogr. 2003, 36, 7.
Li, Adv. Funct. Mater. 2007, 17, 1255; j) J. R. Li, R. J. Kuppler, H.-C. Zhou, [18] a) S. Yang, L. Liu, J. Sun, K. M. Thomas, A. J. Davies, M. W. George, A. J.
Chem. Soc. Rev. 2009, 38, 1477; k) L. J. Murray, M. Dincă, J. R. Long, Blake, A. H. Hill, A. N. Fitch, C. C. Tang, M. Schrçder, J. Am. Chem. Soc.
Chem. Soc. Rev. 2009, 38, 1294. 2013, 135, 4954; b) M. K. Sharma, P. K. Bharadwaj, Inorg. Chem. 2011, 50,

Chem. Eur. J. 2014, 20, 4347 – 4356 www.chemeurj.org 4355  2014 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
 15213765, 2014, 15, Downloaded from https://chemistry-europe.onlinelibrary.wiley.com/doi/10.1002/chem.201303610 by Indian Institute Of Technology, Wiley Online Library on [26/08/2025]. See the Terms and Conditions (https://onlinelibrary.wiley.com/terms-and-conditions) on Wiley Online Library for rules of use; OA articles are governed by the applicable Creative Commons License
Full Paper

1889; c) D. Tanaka, K. Nakagawa, M. Higuchi, S. Horike, Y. Kubota, T. C. [26] G. M. Sheldrick, Siemens area detector absorption correction program;
Kobayashi, M. Takata, S. Kitagawa, Angew. Chem. 2008, 120, 3978; University of Gçtingen, Germany, 1994.
Angew. Chem. Int. Ed. 2008, 47, 3914. [27] A. Altomare, G. Cascarano, C. Giacovazzo, A. Gualaradi, J. Appl. Crystal-
[19] M. M. Dubinin, Chem. Rev. 1960, 60, 235. logr. 1993, 26, 343.
[20] C. E. Webster, R. S. Drago, M. C. Zerner, J. Am. Chem. Soc. 1998, 120, [28] G. M. Sheldrick, SHELXL-97, Program for the crystal structure solution
5509. and refinement; University of Gçttingen, Germany, 1997.
[21] a) O. Talu, Adv. Colloid Interface Sci. 1998, 227, 76; b) J. Purewal, Hydro- [29] WinGX-A, Windows program for crystal-structure analysis L. J. Farrugia,
gen Adsorption by Alkali Metal Graphite Intercalation Compounds, Ph. D. J. Appl. Crystallogr. 1999, 32, 837.
Dissertation, California Institute of Technology, Pasadena, CA, 2010. [30] J. van de Vondele, M. Krack, F. Mohamed, M. Parrinello, T. Chassaing, J.
[22] A. M. Plonka, D. Banerjee, W. R. Woerner, Z. Zhang, N. Nijem, Y. J. Hutter, Comput. Phys. Commun. 2005, 167, 103.
Chabal, J. Li, J. B. Parise, Angew. Chem. 2013, 125, 1736; Angew. Chem. [31] S. Goedecker, M. Teter, J. Hutter, Phys. Rev. B 1996, 54, 1703.
Int. Ed. 2013, 52, 1692. [32] J. P. Perdew, K. Burke, M. Ernzerhof, Phys. Rev. Lett. 1996, 77, 3865.
[23] M. Saharay, S. Balasubramanian, J. Phys. Chem. B 2007, 111, 387. [33] S. J. Grimme, Comput. Chem. 2006, 27, 1787.
[24] A. R. Kennedy, K. G. Brown, D. Graham, J. B. Kirkhouse, M. Kittner, C. [34] W. Humphrey, A. Dalke, K. Schulten, J. Mol. Graphics 1996, 14, 33.
Major, C. J. McHugh, P. Murdoch, E. E. Smith, New J. Chem. 2005, 29,
826. Received: September 12, 2013
[25] SMART (V 5.628), SAINT (V 6.45a), XPREP, SHELXTL; Bruker AXS Inc.: Revised: December 4, 2013
Madison, Wisconsin, USA, 2004. Published online on March 3, 2014

Chem. Eur. J. 2014, 20, 4347 – 4356 www.chemeurj.org 4356  2014 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim