Directing the structural features of N2-phobic nanoporous covalent organic polymers for CO2 capture and separation

Article abstract of DOI:10.1002/chem.201303493

A family of azo-bridged covalent organic polymers (azo-COPs) was synthesized through a catalyst-free direct coupling of aromatic nitro and amine compounds under basic conditions. The azo-COPs formed 3D nanoporous networks and exhibited surface areas up to 729.6 m² g^-1, with a CO₂-uptake capacity as high as 2.55 mmol g^-1 at 273 K and 1 bar. Azo-COPs showed remarkable CO₂/N₂ selectivities (95.6-165.2) at 298 K and 1 bar. Unlike any other porous material, CO ₂/N₂ selectivities of azo-COPs increase with rising temperature. It was found that azo-COPs show less than expected affinity towards N₂ gas, thus making the framework N₂-phobic , in relative terms. Our theoretical simulations indicate that the origin of this unusual behavior is associated with the larger entropic loss of N₂ gas molecules upon their interaction with azo-groups. The effect of fused aromatic rings on the CO₂/N₂ selectivity in azo-COPs is also demonstrated. Increasing the π-surface area resulted in an increase in the CO₂-philic nature of the framework, thus allowing us to reach a CO₂/N₂ selectivity value of 307.7 at 323 K and 1 bar, which is the highest value reported to date. Hence, it is possible to combine the concepts of CO₂-philicity and N ₂-phobicity for efficient CO₂ capture and separation. Isosteric heats of CO₂ adsorption for azo-COPs range from 24.8-32.1 kJ mol^-1 at ambient pressure. Azo-COPs are stable up to 350 °C in air and boiling water for a week. A promising cis/trans isomerization of azo-COPs for switchable porosity is also demonstrated, making way for a gated CO₂ uptake. Copyright

Full text of DOI:10.1002/chem.201303493

DOI: 10.1002/chem.201303493
Full Paper
&
Microporous Materials
Directing the Structural Features of N2-Phobic Nanoporous
Covalent Organic Polymers for CO2 Capture and Separation
Hasmukh A. Patel,[a] Sang Hyun Je,[a] Joonho Park,[a] Yousung Jung,[a] Ali Coskun,*[a, b] and
Cafer T. Yavuz*[a]
Abstract: A family of azo-bridged covalent organic polymers
(azo-COPs) was synthesized through a catalyst-free direct
coupling of aromatic nitro and amine compounds under
basic conditions. The azo-COPs formed 3D nanoporous net-
works and exhibited surface areas up to 729.6 m2 gꢀ1, with
a CO2-uptake capacity as high as 2.55 mmolgꢀ1 at 273 K and
1 bar. Azo-COPs showed remarkable CO2/N2 selectivities
(95.6–165.2) at 298 K and 1 bar. Unlike any other porous ma-
terial, CO2/N2 selectivities of azo-COPs increase with rising
temperature. It was found that azo-COPs show less than ex-
pected affinity towards N2 gas, thus making the framework
“N2-phobic”, in relative terms. Our theoretical simulations in-
dicate that the origin of this unusual behavior is associated
with the larger entropic loss of N2 gas molecules upon their
interaction with azo-groups. The effect of fused aromatic
rings on the CO2/N2 selectivity in azo-COPs is also demon-
strated. Increasing the p-surface area resulted in an increase
in the CO2-philic nature of the framework, thus allowing us
to reach a CO2/N2 selectivity value of 307.7 at 323 K and
1 bar, which is the highest value reported to date. Hence, it
is possible to combine the concepts of “CO2-philicity” and
“N2-phobicity” for efficient CO2 capture and separation. Iso-
steric heats of CO2 adsorption for azo-COPs range from
24.8–32.1 kJmolꢀ1 at ambient pressure. Azo-COPs are stable
up to 3508C in air and boiling water for a week. A promising
cis/trans isomerization of azo-COPs for switchable porosity is
also demonstrated, making way for a gated CO2 uptake.
ganic frameworks, MOFs)[2,5] or the difficulty to tune their struc-
tures (e.g., zeolites, activated carbon) or the lack of sustainable
synthesis, owing to the use of rare earth catalysts (e.g., conju-
gated microporous polymers).[6]
Introduction
Carbon dioxide (CO2) emissions into the atmosphere account
for the majority of energy and environmental challenges, and
its global impact in the form of climate change is well-docu-
mented.[1] The unprecedented increase in the atmospheric
presence of CO2 also endangers the ecosystem by ocean
acidification and other, yet unknown, direct interferences to
the life on Earth.[2] The threat becomes more imminent as the
time passes because no viable solution to effectively capture
and store CO2 is available. In the case of capture and separa-
tion, CO2 scrubbing by ethanolic aqueous amines is yet to be
challenged.[3] Nanoporous materials (pore sizes under 100 nm)
have been studied extensively for CO2 capture and separation
on account of their modular nature, chemical and thermal sta-
bility, and large accessible surface area.[4] Most existing porous
materials, however, suffer from water instability (e.g., metal–or-
There are several examples of crystalline or amorphous
nanoporous polymers reported in the literature;[7] namely,
polymers with intrinsic microporosity (PIM),[8] hypercross-linked
and conjugated microporous polymers (HCP and CMP),[6b,9] or-
ganic cages,[10] benzimidazole-linked polymers (BILP),[11] cova-
lent triazine frameworks (CTF),[12] porous aromatic frameworks
(PAF),[13] porous polymer networks (PPN),[14] covalent organic
frameworks (COF),[15] porous organic frameworks (POF),[16] cova-
lent organic polymers (COP),[17] carbazole-based porous organic
polymer (CPOP),[18] elemental-organic frameworks (EOFs),[19]
porous cross-linked polymers (PCPs),[20] porous organic poly-
mers (POPs),[21] tetrazine-based organic frameworks (TzFs),[22]
porous polymer frameworks (PPF),[23] Schiff base networks
(SNW),[24] and others.[25] Although CO2-philic sorbents that are
decorated with amines have been well-documented,[2,4b] their
energy intensive regeneration on account of their high isoste-
ric heats of adsorption (Q>40 kJmolꢀ1) and low thermal and
chemical stability due to the oxidation of amino groups limit
their application in CO2 capture and separation. For example,
polyamine-tethered PPNs showed CO2 capacity of 4.3 mmolgꢀ1
and CO2/N2 selectivity as much as 442 at 295 K and 1 bar,[26]
however, amine functionalities were prone to oxidative degra-
dation and decomposed below 2008C in air. This observation
is in line with the concept that physisorption of CO2 is more
[a] Dr. H. A. Patel, S. H. Je, Dr. J. Park, Prof. Y. Jung, Prof. A. Coskun,
Prof. C. T. Yavuz
Graduate School of EEWS, Korea Advanced Institute of Science
and Technology (KAIST), Daejeon 305–701 (Republic of Korea)
E-mail: coskun@kaist.ac.kr
yavuz@kaist.ac.kr
[b] Prof. A. Coskun
Department of Chemistry, Korea Advanced Institute of Science and
Technology (KAIST), Daejeon 305–701 (Republic of Korea)
Supporting information for this article is available on the WWW under
http://dx.doi.org/10.1002/chem.201303493.
Chem. Eur. J. 2014, 20, 772 – 780
772
ꢀ 2014 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

Full Paper
preferred over the chemisorption one due to the fact that
there are no chemical bonds formed or broken during adsorp-
tion and regeneration cycles. One proof for the latter is
a report by El-Kaderi et al.[27] in which BILPs are shown to have
exceptional CO2-capture capacities up to 5.34 mmolgꢀ1 at
273 K and 1 bar. Another example is the recent report on PPFs
by Zhu et al.[23] in which CO2 capacities of up to 6 mmolgꢀ1
were reported with imine-linked nanoporous amorphous poly-
mers at 273 K and 1 bar. All these examples showed exception-
al CO2 capacities with moderate-to-low CO2/N2 selectivity,
which decrease with rising temperature, an important limita-
tion for their application in the temperature range (>408C) for
the post-combustion CO2 capture.[1a] We have recently report-
ed a new class of nanoporous polymers called azo-bridged co-
valent organic polymers (azo-COPs), synthesized by catalyst-
free coupling of aromatic nitro and amine compounds under
basic conditions.[28] Azo-COPs revealed an increase in CO2/N2
methanetetrayltetraaniline (azo-COP-1), p-phenylenediamine
(azo-COP-2), benzidine (azo-COP-3), 2,3,5,6-tetramethyl-1,4-
phenylenediamine (azo-COP-4), 2,5-dimethyl-1,4-phenylenedia-
mine (azo-COP-5), 1,3-phenylenediamine (azo-COP-6), 3,3’-di-
methylbenzidine (azo-COP-7), 4,4’-methylenedianiline (azo-
COP-8), 4,4’-oxydianiline (azo-COP-9), 1,5-diaminonaphthalene
(azo-COP-10), and tris(4-aminophenyl)methanol (azo-COP-11),
resulting in a series of azo-COPs with different structural geo-
metries and properties (Figure 1).
To verify the formation of azo-COPs, we have carried out
a set of analyses, including Fourier transform infrared spectros-
copy (FTIR), elemental analysis, cross-polarization magic-angle
spinning (CP/MAS) 13C NMR spectroscopy, and thermogravi-
metric analysis (TGA). The FTIR spectra confirmed the forma-
tion of azo (ꢀN=Nꢀ) linkages as shown by the stretching
bands at 1447 and 1403 cmꢀ1 (Figure S1 in the Supporting In-
formation). Other noticeable bands were observed at 3400 (s,
selectivity with rising temperature due to the fact that azo (ꢀ free NꢀH), 3200 (s, hydrogen-bonded NꢀH), 1610 (s, C=C aro-
N=Nꢀ) functionalities showed less than expected affinity to-
wards N2 gas, thus making the framework “N2-phobic”, in rela-
tive terms. Azo-COP-2 showed moderate CO2/N2 selectivity of
130.6 at 298 K, which is increased to the then highest value
(288.1, at 323 K and 1 bar) with rising temperature. Our previ-
ously reported[28] theoretical simulations suggested that the
origin of N2-phobicity is due to the entropic loss of N2-gas mol-
ecules upon their interaction with the azo-groups, although
the binding process is enthalpically favorable.
matic), 1447 and 1403 (s, N=N), 1520 and 1340 (s, NꢀO), and
1280 cmꢀ1 (s, CꢀN). The high oxygen content of azo-COPs in
the elemental analysis could be attributed (Table S1 in the Sup-
porting Information) to the presence of terminal nitro groups
along with moisture trapped in the framework. Moreover, the
ICP-MS analyses of azo-COPs show that the potassium content
varies in the range from undetectable to 0.036% (w/w;
Table S2 in the Supporting Information), confirming that higher
oxygen content is associated with terminal nitro groups and
moisture. Moreover, the stretching band located at 3600–
3000 cmꢀ1 also indicates the presence of terminal amino
groups. The chemical shifts in the CP/MAS 13C NMR spectra of
azo-COPs, namely, azo-COP-1 to azo-COP-11, located at d=
150.2, 144.7, 129.5, 123.4, and 54.9 ppm confirmed the forma-
tion of the azo-linked aromatic polymers (Figure S2 in the Sup-
porting Information). The signal at d=115 ppm, which is at-
tributed to the carbon atom incorporating the terminal ꢀNH2
moiety, is clearly visible in all samples, confirming the presence
of amine end-groups. The chemical shifts of aromatic moieties
are merged in the region d=120–155 ppm.
Herein, we have described the synthesis of a family of azo-
COPs that show remarkable CO2/N2 selectivities at warm tem-
peratures. To understand the scope of azo-coupling, we have
used a series of monomers with varying steric hindrance, rigidi-
ty, and p-surface area and investigated their effect on the CO2
and N2 gas sorption characteristics. We have also demonstrat-
ed that the increasing p-surface area by using fused aromatic
rings resulted in a significant increase in the CO2/N2 selectivity.
Our simulations indicate that CO2 has a higher affinity towards
aromatic units when compared with N2 gas molecules, thus in-
creasing the p-surface area increases the CO2-philicity of the
framework. These results indicate that it is possible to combine
the concepts of CO2-philicity with N2-phobicity to improve
CO2/N2 selectivity and also CO2-uptake capacity. The formation
of azo-COPs and their structural stability were confirmed by
using a series of analytical techniques. We have also investigat-
ed the switching of azo-COPs and its effect on the surface
area. Azo-COPs are shown to be extremely stable up to 3508C
in air and also in boiling water for a week.
Thermogravimetric analysis (TGA) performed on azo-COPs
up to 8008C with a heating rate of 108Cminꢀ1. Azo-COPs ex-
hibited remarkable thermal stability up to 3508C in air (Fig-
ure S3 in the Supporting Information). The TGA patterns
showed a single-step mass-loss, which is attributed to the deg-
radation of organic frameworks. Interestingly, when we carried
out the TGA experiment under an N2 atmosphere, azo-COPs
decomposed at lower temperatures (e.g., 3008C) when com-
pared with the one under air. We believe that although the
low molecular weight residual impurities in azo-COPs induced
the decomposition at elevated temperatures under an N2 at-
mosphere, those impurities are just oxidized in the presence of
O2 without affecting the network structure. To confirm this be-
havior, we have synthesized a model azo compound (see the
Supporting Information) and studied its thermal stability. The
model azo compound showed (Figure S4 in the Supporting In-
formation) identical weight loss both in N2 and in air, confirm-
ing that the early mass loss of azo-COPs under N2 atmosphere
is due to the decomposition of low molecular weight residual
Results and Discussion
Aromatic azo compounds are generally prepared from oxida-
tion of aromatic amines in the presence of metal-based cata-
lysts.[29] In the search for an environmentally benign process,
we have recently demonstrated[28] the catalyst-free direct cou-
pling of aromatic amines and nitro compounds under basic
conditions to prepare azo-COPs. A family of azo-COPs were
synthesized by the direct coupling of tetrakis(4-nitrophenyl)-
methane with various aromatic amines, including 4,4’,4’’,4’’’-
Chem. Eur. J. 2014, 20, 772 – 780
www.chemeurj.org
773
ꢀ 2014 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

Full Paper
Figure 1. Building blocks of azo-bridged covalent organic polymers (azo-COPs), azo-COPs-1 to -11. Direct coupling of aromatic nitro with amines under basic
conditions at 1508C resulted in the formation of azo-COPs as bright-orange powders on a multigram scale. Azo-COP-1, -2, and -3 are discussed[28] for compar-
ison with other azo-COPs.
impurities. To investigate the water stability of azo-COPs, we
kept the polymers in boiling water for a week and measured
their surface areas. Azo-COPs did not show any noticeable loss
of surface area (Figure S5 in the Supporting Information), indi-
cating that their network structures are highly robust.
To investigate the permanent nanoporous nature of azo-
COPs, N2 adsorption–desorption experiments were conducted
at 77 K (Figure 2). All samples were degassed at 1508C for 5 h
under vacuum prior to the measurement. Azo-COPs showed
typical type I reversible sorption isotherms with Brunauer–
Emmett–Teller (BET) surface areas ranging from 11.1–
729.6 m2 gꢀ1, verifying the porous nature of the synthesized
materials (Table 1). The BET surface areas were calculated from
the slope within 0.01–0.25 relative pressures (Figure S6 in the
Supporting Information). The Langmuir surface area of azo-
COPs is varied from 13.4 to 949.8 m2 gꢀ1 (Table 1). Azo-COPs
Figure 2. N2 adsorption–desorption isotherms of azo-COPs-1 to -11 carried
out at 77 K and the gas uptake and release are indicated by solid and open
symbols, respectively. The BET surface areas of azo-COP-1 to -3 (measured
by Ar adsorption–desorption at 87 K) have been reported[42] and were used
for comparison. The N2 adsorption–desorption isotherms of azo-COPs-1 to -3
are given as Figure S12 (in the Supporting Information).
contain mesoporous network along with microporosity as evi-
dent from the hysteresis at the low pressure region that is as-
cribed to the pore network effects, wherein mesopores accessi-
ble only through micropores.[30] The shape of nitrogen iso-
therm for azo-COP-6 differs at the higher relative pressure
range, that is, more than 0.9, indicating an adsorption on the
outer surface of small particles. The increase in the N2 sorption
may arise in part from inter-particulate voids coupled with the
meso and macrostructures.[31] In addition to the fact that the
enhancement in BET surface area may not be the deciding
factor for gas sorption behavior (especially for CO2 gas sorp-
tion in which nitrogen, electron-rich moieties perform better),
a moderate porosity (e.g. surface areas of 100 m2 gꢀ1 or less)
could lead to complete access to CO2-philic nitrogen-rich func-
tionalities in the framework.[9]
Chem. Eur. J. 2014, 20, 772 – 780
www.chemeurj.org
774
ꢀ 2014 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

Full Paper
tive acts. Whereas the flexibility
brings free rotation and conse-
quent contraction in framework
construction, it also eliminates
the restriction on the in-plane
alignment of biphenylic azo
units. As a result, the more con-
Table 1. BET and Langmuir surface areas, CO2 and N2 capture capacities, CO2/N2 selectivities and isosteric heats
of adsorption (Qst) for azo-COPs.
Compound
BET
[m2 gꢀ1
Langmuir
CO2 adsorption
[mmolgꢀ1
N2 adsorption
[mmolgꢀ1
CO2/N2 (15:85)
at 1 bar
Qst
[kJmolꢀ1
]
]
[m2 gꢀ1
]
]
]
273 K
298 K
273 K
298 K
273 K
298 K
azo-COP-1
azo-COP-2
azo-COP-3
azo-COP-4
azo-COP-5
azo-COP-6
azo-COP-7
azo-COP-8
azo-COP-9
azo-COP-10
azo-COP-11
635.8
729.6
493.1
11.1
127.6
679.1
241.6
472.1
509.6
156.8
259.2
800.7
949.8
632.4
13.4
169.4
887.2
315.3
612.7
649.5
200.2
336.1
2.44
2.55
1.93
1.75
2.04
2.22
1.91
2.02
2.05
1.91
2.13
1.48
1.53
1.22
1.12
1.24
1.31
1.16
1.22
1.23
1.15
1.26
0.11
0.06
0.07
0.13
0.16
0.17
0.16
0.16
0.17
0.15
0.15
0.04
0.03
0.03
0.05
0.05
0.06
0.06
0.07
0.07
0.04
0.06
63.7
109.6
78.6
79.3
73.8
72.7
68.5
70.3
69.6
69.6
77.8
96.6
130.6
95.6
29.3
24.8
32.1
26.8
27.3
25.8
26.1
25.3
25.3
27.9
27.4
jugated,
semi-rigid
oxygen
bridge shows an enhanced sur-
face area, whereas the methyl-
ene bridge reports a reduced
area. Tetrahedral linkers of both
azo-COP-1 and 11 are expected
to produce the highest degree
of porosity, as shown previously
in porous boronic esters, that is,
COF-102.[34] This is, however, not
138.5
144.6
114.8
115.3
103.4
102.8
165.2
128.4
Accessible surfaces can be directed by the structural fea-
tures,[32] despite the challenge of lacking order in azo-COP
frameworks. In an attempt to increase pore-size diameter, we
doubled the linker length (R2 unit in Figure 1), a common strat-
egy in the construction of hybrid porous materials.[33] Azo-COP-
3 (with double linker-length than that of azo-COP-2) showed
a 55% increase in pore size (from 0.58 to 0.9 nm) though with
a loss of 32% in BET surface area (from 729.6 to 493.1 m2 gꢀ1).
Bulky groups are introduced to the ortho positions of the
linker in azo-COP-2, leading to azo-COPs incorporating tetra-
methyl (azo-COP-4) and dimethyl (azo-COP-5) groups, to inves-
tigate the effect of steric hindrance. Surface areas are consider-
ably reduced (>80%), owing to the inability of in-plane exten-
sion of the monomers, a preferable alignment for azo-aromat-
the case for azo-COPs, because azo-COP-1 is the third spacious
among the 11 azo-COPs reported here. Azo-COP-11 also fea-
tures a reactive alcohol at the tetrahedral center, presumably
leading to side redox reactions under highly basic conditions.
The redox effect is most evident in a failed azo-COP architec-
ture (see the Supporting Information), in which a biphenylic
linker with a sulfone (ꢀSO2ꢀ) bridge is used. The reductive cou-
pling of nitro and amino groups are altered substantially with
the interference of the sulfone functionality of the linker.
Given the variation of azo-COPs in their surface area, micro-
porosity, and structural geometry, it is anticipated that each
azo-COP would reveal a different gas sorption behavior. The
gas adsorption and desorption measurements for CO2 and N2
were obtained for all azo-COPs using a volumetric gas analyzer
ics. In the simplest terms, methyl groups force the ꢀN=Nꢀ at 273 and 298 K (Figure 3). All azo-COPs were degassed at
double bond to move outer plane, to at least a tilted geometry
with respect to the aromatic phenyls. The biphenyl linker in
azo-COP-3 also shows a similar loss of surface area (51%)
when ortho-methyl groups are attached on both ends (see
azo-COP-7). The lesser decrease in the latter can be attributed
to the already tilted geometry of biphenyl moieties. Tetrameth-
yl substituents in azo-COP-4 are also responsible for the diffi-
culty in the formation of a porous network structure.
1508C over 5 h to remove the adsorbed water and residual sol-
vent molecules from the nanopores of the framework. The
shapes of the CO2 isotherms are somewhat different at the
two temperatures. At 273 K, the CO2 adsorption amount in
azo-COPs increased rapidly at the lower pressure range before
approaching saturation, and consequently showed a higher
CO2 adsorption capacity when compared with the value found
at 298 K. Because both the high porosity and nitrogen content
contribute to the CO2-philicity of the framework, azo-COP-2
showed the highest CO2 sorption capacity of 2.55 and
1.53 mmolgꢀ1 at 273 and 298 K, respectively. Other azo-COPs
demonstrated CO2 adsorption capacity in the range of 1.75–
2.44 and 1.12–1.48 mmolgꢀ1 at 273 and 298 K, respectively.
Most likely, at low pressures, the nitrogen-rich sites of the azo-
COPs have much higher binding affinity for CO2 because of its
large quadrupolar moment and when the pressure is increased
these binding sites become saturated and therefore less avail-
able for CO2 molecules.
The best measure for the effect of kinks and tilts, perhaps,
would be forcing the linker to have a non-linear connectivity.
We chose a 1,3-diaminobenzene as a linker (azo-COP-6) to
check against azo-COP-2 incorporating 1,4-diaminobenzene as
a linker. To our surprise, the reduction in the surface area was
only 7%. This result might be associated with the uninterrupt-
ed in-plane p-connectivity of azo units with the aromatic rings.
The trans configuration of azo groups also allows a zigzag
alignment, resulting in a relatively high surface area. The small
decrease, however, could be attributed to short monomer dis-
tances with a contracted spring-like monomer extension.
In azo-COP-8 and 9, we assessed the effect of flexible
bridges between biphenylic linkers. When compared to the
rigid azo-COP-3, both structures showed similar surface areas.
This result is rather contrary to the common belief that flexible
linkers yield considerable losses in overall porosity. In this par-
ticular case with azo aromatics, however, we see two competi-
There is a negligible hysteresis in CO2 adsorption–desorption
isotherms at 273 and 298 K (Figure S7 in the Supporting Infor-
mation), confirming the complete regeneration without any
need for an external heat treatment. This property is highly de-
sirable in post-combustion CO2 capture because it eliminates
excessive energy consumption during adsorbent regeneration.
The CO2 scrubbing by well-established industrial processes
Chem. Eur. J. 2014, 20, 772 – 780
www.chemeurj.org
775
ꢀ 2014 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

Full Paper
higher than those previously re-
ported[4e] (Table S3 in the Sup-
porting Information). At more in-
dustrially relevant temperatures
(>408C), CO2/N2 selectivity of
azo-COP-2 is 288.1 at 323 K and
1 bar[28] whereas azo-COP-10
shows an even higher CO2/N2 se-
lectivity of 307.7, which is the
highest value reported to date
at 323 K and 1 bar (Figure S9 in
the Supporting Information). The
IAST selectivity is higher at low
pressure region and subsequent-
ly decreased with the increasing
pressure (Figure 4). We also de-
termined the isosteric heats of
adsorption (Qst) by using the
Clausius–Clapeyron
method
from the CO2 isotherms collected
at 273 and 298 K (Figure 4).[36] At
zero coverage, the Qst values of
CO2 for azo-COPs were found to
be 27.0–34.1 kJmolꢀ1
were decreased with higher
loadings to 24.8–27.9 kJmolꢀ1
,
which
Figure 3. Low-pressure: a,b) CO2, and c,d) N2 adsorption isotherms for azo-COPs-1 to -11 at 273 and 298 K. When
the temperature was increased from 273 to 298 K, the CO2-uptake capacity loss was about 40%, whereas it is ap-
proximately 70% for N2. We believe that this higher loss for N2 is because of the lack of dispersion forces and also
electron–electron repulsion between N2 gas and azo groups.
.
These Qst values are very
common in nitrogen-rich nano-
porous polymers and facilitate
that use aqueous amine solutions (e.g., 30% monoethanola-
mine) require significant amounts of heat for regeneration, re-
sulting in a high-energy penalty, not to mention the accelerat-
ed corrosion.[1a] Azo-COPs adsorb relatively small amount of N2
and exhibit excellent CO2/N2 selectivity. Unlike any porous ma-
terial, azo-COPs reveal an increase in selectivity with rising
temperature. This unusual behavior is associated with the un-
favorable interactions between N2 gas molecules and azo-link-
ages in azo-COPs, allowing us to coin the term “N2-phobicity”
for this new concept. The CO2/N2 (0.15:0.85) selectivities of
azo-COPs were obtained from ideal adsorbed solution theory
(IAST) at 273 and 298 K (Table 1) for a CO2/N2 gas mixture. The
IAST method predicts the adsorption selectivity for gas mix-
tures based on pure component gas isotherms and has been
used to investigate several adsorbents.[26,35] Although break-
through studies with mixture of gases are ultimately needed
for absolute selectivity values, it is well-known[1b,28] that com-
petitive adsorptions at limited sorption sites favor even higher
selectivities than pure component IAST calculations. The abso-
lute component loadings were fitted with a single-site Lang-
muir model for N2 isotherms and dual-site Langmuir model for
CO2 isotherms (Figure S8 in the Supporting Information). The
saturation loading and parameter in the pure component
Langmuir adsorption isotherms were obtained from the fitting
curves and were used to calculate CO2/N2 selectivity for a gas
mixture of 0.15:0.85.
pressure swing desorption (<40 kJmolꢀ1) from the porous net-
works without needing a successive thermal treatment.[25a,37]
In a typical theoretical assessment, the quantum-chemical
calculations using Perdew–Burke–Ernzerhof exchange-correla-
tion functional[38] with dispersion correction[39] in Q-Chem[40]
were performed to understand the changes in Qst and the se-
lectivity with different models of azo-COPs. The fragment azo-
models that are considered in this paper are illustrated along
with the optimized geometries and binding energies (Fig-
ure S10 in the Supporting Information). The binding energies
(BEs) of these modeled azo-COP fragments for CO2 adsorption
range between 14 and 17 kJmolꢀ1 at the PBE-D2/aug-cc-pVTZ
level. A more-or-less uniform underestimation of the calculated
BE’s with respect to the experimental Qst by about 10 kJmolꢀ1
(with the correct trend) is perhaps due to the removal and
simplification of the amorphous polymer network surround-
ings in our model calculations. The multiple interactions with
more than one functional group are possible in amorphous
polymers due to the interpenetration of the framework, which
leads to an increase in the observed binding energies (Fig-
ure S11 in the Supporting Information).
The fact that azo-COP-10 is superior to azo-COP-2 in its CO2/
N2 selectivity, prompted us to investigate these two structures
more closely. Since both structures only differ in the linkers,
benzene (azo-COP-2) versus naphthalene (azo-COP-10), we
chose to compare the respective model constructs. The opti-
mized geometries of CO2 and N2 binding with the bare ben-
zene and naphthalene molecules as well as the same molecu-
The CO2/N2 selectivities of the azo-COPs range from 63.7–
109.6 at 273 K and become 95.6–165.2 at 298 K, which is
Chem. Eur. J. 2014, 20, 772 – 780
www.chemeurj.org
776
ꢀ 2014 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

Full Paper
same (from 6.7–7.4 to 6.0–7.5 kJmolꢀ1), the azo group signifi-
cantly enhanced the CO2 binding (from 10.1–11.6 to 14.5–
15.5 kJmolꢀ1), thereby increasing the selectivity. One interest-
ing observation in these experiments was the change in Qst as
well as the selectivity that occurred by the modification of the
benzene linker (azo-COP-2) with naphthalene (azo-COP-10).
The increase (3.1 kJmolꢀ1) in Qst of CO2 for azo-COP-10 (naph-
thalene) compared with azo-COP-2 (benzene) is qualitatively
consistent with the calculated binding energy difference
(1.0 kJmolꢀ1) shown in Figure 5. The selectivity at low tempera-
ture can be predicted by the difference in calculated binding
energies of CO2 versus N2. The BE difference between CO2 and
N2 for azo-COP-2 (Figure 5c) and azo-COP-10 (Figure 5d) were
8.5 and 8.0 kJmolꢀ1, which means that the selectivity would be
marginally higher for azo-COP-2 at low temperature as is ob-
served in experiments. However, the large temperature-depen-
dent selectivity increase in azo-COP-10 compared with azo-
COP-2 seems to have its origin in the relative entropy differ-
ence between the CO2 and N2 binding configurations.[28] Quali-
tatively, this phenomenon may be due to the larger spatial dis-
tribution of the CO2-bound configurations around naphthalene
compared with benzene because naphthalene has a larger p-
surface area that can interact with CO2.
We have also investigated the trans!cis isomerization of
the azo-COPs by irradiating the samples with UV light to alter
the pore characteristics (on–off porosities)[41] of their network
structure. The absorbance maxima, lmax, recorded around
360 nm for all the azo-COPs represent the p–p* transition of
the relatively stable trans form (Figure S12 in the Supporting
Information). Specifically, we have recorded the solid-state UV/
Vis spectra of powder azo-COPs after irradiating with UV light
(light source: 365 nm; 8 W) overnight and measured the
change in surface area. We did not observe a significant
change in the peak located at 360 nm for azo-COPs, except for
azo-COP-2 and -9, after exposing the samples to UV light for
24 h. It is known[42] that the trans-to-cis isomerization of azo-
benzenes is hampered severely in restricted environments.
Azo-COP-2 and azo-COP-9 showed the formation of a new
broad peak located at 440 nm, which can be attributed to an
n–p*-transition of cis-azobenzene. This result prompted us to
investigate further alterations in its surface area with respect
to UV irradiation. Azo-COP-2 and azo-COP-9 were irradiated by
UV light for 24 h, followed by degassing for 3 h at 298 K, in
which low-temperature activation is preferred to minimize
thermal relaxation of cis-azobenzene back to its trans configu-
ration. The surface area measurement was then carried out
under identical conditions to those for azo-COP-2 and azo-
COP-9 with and without UV irradiation (Figure 6). The BET sur-
face area is reduced from 594.8 to 568.8 and 343.7 to
325.9 m2 gꢀ1 for azo-COP-2 and azo-COP-9, respectively, sug-
gesting that a structural change in the network has occurred
after UV irradiation. As expected, these BET surface areas are
lower than those measured after degassing at 1508C for 5 h.
Although we observed a relatively small change in the porosity
of the material, these findings could lead to the development
of porous polymers with tunable pore size and chemical
nature.
Figure 4. CO2/N2 selectivity of azo-COPs measured by IAST technique for
CO2/N2 gas mixture of 15:85 at: a) 273 K, and b) 298 K. c) Isosteric heats of
adsorption (Qst) for CO2 calculated from the isotherms measured at 273 and
298 K for azo-COPs.
lar moieties with azo extensions are shown in Figure 5. The BE
values of CO2 and N2 for bare benzene are 10.1 and
6.7 kJmolꢀ1, respectively, and these values are increased to
11.6 and 7.4 kJmolꢀ1 for bare naphthalene. When there are
azo-groups, however, the preferential CO2 adsorption com-
pared with N2 adsorption is noticeably increased to 14.5 (CO2)
versus 6.0 kJmolꢀ1 (N2) for azo-benzene, and 15.5 (CO2) and
7.5 kJmolꢀ1 (N2) for azo-naphthalene. In other words, although
the N2 binding with and without the azo group is almost the
Chem. Eur. J. 2014, 20, 772 – 780
www.chemeurj.org
777
ꢀ 2014 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

Full Paper
Conclusion
We have prepared a family of
nanoporous azo-COPs by using
a
catalyst-free chemical route
and show that their azo groups
significantly disfavor N2 by ex-
hibiting an anomalous trend in
which CO2/N2 selectivity is in-
creased with a rise in tempera-
ture. The CO2/N2 selectivity of
azo-COP-10 (165.2 at 298 K) ex-
ceeds previously reported values
for azo-COPs. Azo-COP-2 shows
the highest CO2 uptake capacity
of 2.55 mmolgꢀ1 and surface
area of 729.6 m2 gꢀ1, among
other azo-COPs studied here.
The moderate Qst and negligible
hysteresis for CO2 adsorption iso-
Figure 5. The optimized structures and binding energies using PBE+D2 in Q-Chem for: a) benzene, b) naphtha-
lene, c) azo-benzene (core functionality of azo-COP-2), and d) azo-naphthalene (core functionality of azo-COP-10).
therms are suggesting energy efficient recyclability. We have
also demonstrated the scope of azo-coupling and contribution
of p-surfaces to the CO2/N2 selectivity. Considering the higher
affinity of CO2 towards the azo-naphthalene moiety over the
azo-benzene one, we have proven that the concepts of CO2-
philicity and N2-phobicity can co-exist, thus creating myriad of
possibilities to combine N2-phobic azo-groups with CO2-philic
functionalities. When coupled with porosity engineering, for
example, tuning surface area and pore size, the new chemical
insights that our study brings may lead to custom-designed ar-
chitectures for efficient CO2 capture and separation. In particu-
lar, the unique ability of azo-COPs to retain high CO2/N2 selec-
tivity at elevated temperatures, combined with their chemical
stability, makes them ideal candidates for post-combustion
CO2 capture. We believe that the tailoring of structural features
in azo-COPs with control over both the functionality and sur-
face area will further contribute to the development of superi-
or CO2 sorbents and also to the fight against global warming
and ocean acidification.
Experimental Section
General synthetic procedure for azo-COPs
All organic solvents and materials used for synthesis were of re-
agent grade and used without further purification. In an typical re-
action, tetrakis(4-nitrophenyl)methane (0.5 mmol, 250 mg) was
treated with the aromatic amines, namely, 4,4’,4’’,4’’’-methanete-
trayltetraaniline (190 mg), p-phenylenediamine (108 mg), benzidine
(184 mg), 2,3,5,6-Tetramethyl-1,4-phenylenediamine (164 mg), 2,5-
dimethyl-1,4-phenylenediamine (136 mg), 1,3-phenylenediamine
(108 mg), 3,3’-dimethylbenzidine (212 mg), 4,4’-methylenedianiline
(198 mg), 4,4’-oxydianiline (200 mg), tris(4-aminophenyl)methanol
(214 mg), and 1,5-diaminonaphthalene (158 mg), in N,N-dimethyl-
formamide (25 mL) in the presence of potassium hydroxide
(5 mmol, 281 mg) under N2 atmosphere at 1508C for 24 h. The re-
action mixture was cooled to room temperature and stirred in an
excess of distilled water for 1 h. The precipitate was filtered off and
Figure 6. UV/Visible spectra showing the trans–cis transition of azo-benzene
units while irradiating the solid samples with UV light (365 nm; 8 W) for
24 h. A significant decrease in the trans azo-benzene peak (lmax =360 nm)
and the formation of a new broad cis azo-benzene peak at 440 nm points to
a trans–cis switching of azo-benzene units. Inset: alteration of surface area
with UV irradiation. N2 adsorption–desorption isotherms for azo-COP-2 and
-9 measured at 77 K of azo-COP-2, -9, and azo-COP-2-UV, -9-UV (azo-COP-2,
-9 were irradiated with UV light for 24 h).
Chem. Eur. J. 2014, 20, 772 – 780
www.chemeurj.org
778
ꢀ 2014 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

Full Paper
washed with warm distilled water (ꢁ3), followed by washing with
Me2CO (ꢁ3) and THF (ꢁ2). Subsequently, an orange precipitate
was dried at 1108C under vacuum for 4 h. All azo-COPs were isolat-
ed as orange powders and were reproducible on at least three
separate batches and on multigram scales.
The IAST selectivities for the CO2:N2 (0.15:0.85) gas mixtures were
calculated by using following Equation (3):
q1=q2
ð3Þ
S ¼
p1=p2
in which S is the selectivity factor, q1 and q2 represent the quantity
adsorbed of components 1 and 2, and p1 and p2 represents the
partial pressure of components 1 and 2, respectively.
Measurements
FTIR spectra were recorded on KBr pellets using a PerkinElmer FTIR
spectrometer. Solid-state cross polarization magic angle spinning
(CP/MAS) 13C NMR spectra of azo-COPs were collected by a Bruker
Avence III 400 WB NMR spectrometer. Thermogravimetric analysis
(TGA) was performed on a NETZSCH-TG 209 F3 instrument by
heating the samples to 8008C at a rate of 108Cminꢀ1 under an air
atmosphere. The absorbance of azo-COPs samples was measured
on a UV/Vis spectrophotometer, in which the powder samples
were sandwiched between two quartz glasses and irradiated by
a 365 nm (8 W; 230 V; ꢁ50/60 Hz from Vilbert Lourmat, France) UV
lamp. To evaluate the porosity of azo-COPs, N2 adsorption iso-
therms were obtained with a Micromeritics ASAP 2020 accelerated
surface area and porosimetry analyzer at 77 K, after the samples
had been degassed at 1508C for 5 h under vacuum. The adsorp-
tion–desorption isotherms were obtained to give the pore parame-
ters, including BET (P/P0 =0.01–0.25) and Langmuir (P/P0 =0.1–0.35)
surface area, and pore volume. The pore volume of azo-COPs was
calculated at P/P0 =0.95. The low pressure CO2 and N2 adsorption–
desorption isotherms for azo-COPs were measured at 273 and
298 K by using a static volumetric system (ASAP 2020, Micromerit-
ics Inc.). The temperature during adsorption and desorption was
kept constant by using a circulator. Prior to the adsorption meas-
urements, the samples were dried at 1108C for 24 h. The samples
were further activated in situ by increasing the temperature at
a heating rate of 1 Kminꢀ1 up to 423 K under vacuum (5ꢁ
10ꢀ3 mm Hg) and the temperature and vacuum was maintained for
5 h before taking the sorption measurements. All the adsorption–
desorption experiments were carried out twice to ensure the re-
producibility. There were no noticeable differences in the isotherm
points obtained from both experiments.
Isosteric heats of adsorption were calculated from the adsorption
data by using the Clausius–Clapeyron Equation (4):
ꢂ
ꢃ
.
ꢀ
ꢁ
DadHꢂ ¼ R
1
ð4Þ
@ ln P
@
=
T
q
in which R is the universal gas constant, whereas q is the fraction
of the adsorbed sites at a pressure P and temperature T.
Acknowledgements
C.T.Y. and Y.J. acknowledge the financial support by grants
from Korea CCS R&D Centre, and IWT (NRF-2012-
C1AAA001M1A2A2026588) through the National Research
Foundation of Korea (NRF) funded by the Ministry of Science,
ICT & Future Planning (2013R1A1A1012998), and KAIST EEWS
Initiative. This work was also made possible by an NPRP grant
#5-499-1-088 from the Qatar National Research Fund (a
member of Qatar Foundation). A.C. acknowledges the support
from the KUSTAR-KAIST Institute, Korea, under the R&D pro-
gram supervised by the KAIST and Basic Science Research Pro-
gram through the National Research Foundation of Korea
(NRF) funded by the Ministry of Science, ICT & Future Planning
(2013R1A1A1012282).
Keywords: adsorption · carbon dioxide · gas selectivity ·
microporous materials · polymers
CO2/N2 Selectivity and isosteric heat of adsorption
[1] a) D. M. D’Alessandro, B. Smit, J. R. Long, Angew. Chem. 2010, 122,
6194–6219; Angew. Chem. Int. Ed. 2010, 49, 6058–6082; b) N. Y. Du,
H. B. Park, G. P. Robertson, M. M. Dal-Cin, T. Visser, L. Scoles, M. D.
Guiver, Nat. Mater. 2011, 10, 372–375; c) N. B. McKeown, P. M. Budd,
Chem. Soc. Rev. 2006, 35, 675–683; d) R. Monastersky, Nature 2013, 497,
13–14.
[2] S. Keskin, T. M. van Heest, D. S. Sholl, ChemSusChem 2010, 3, 879–891.
[3] a) G. T. Rochelle, Science 2009, 325, 1652–1654; b) C. T. Yavuz, B. D. Shi-
nall, A. V. Iretskii, M. G. White, T. Golden, M. Atilhan, P. C. Ford, G. D.
Stucky, Chem. Mater. 2009, 21, 3473–3475.
[4] a) Z. Chang, D. S. Zhang, Q. Chen, X. H. Bu, Phys. Chem. Chem. Phys.
2013, 15, 5430–5442; b) R. Dawson, A. I. Cooper, D. J. Adams, Prog.
Polym. Sci. 2012, 37, 530–563; c) P. Kaur, J. T. Hupp, S. T. Nguyen, ACS
Catal. 2011, 1, 819–835; d) A. Thomas, Angew. Chem. 2010, 122, 8506–
8523; Angew. Chem. Int. Ed. 2010, 49, 8328–8344; e) Z. H. Xiang, X.
Zhou, C. H. Zhou, S. Zhong, X. He, C. P. Qin, D. P. Cao, J. Mater. Chem.
2012, 22, 22663–22669; f) Y. G. Zhang, S. N. Riduan, Chem. Soc. Rev.
2012, 41, 2083–2094; g) Y. Jin, Y. Zhu, W. Zhang, CrystEngComm 2013,
15, 1484–1499; h) X. Feng, X. Ding, D. Jiang, Chem. Soc. Rev. 2012, 41,
6010–6022.
Ideal adsorbed solution theory (IAST) data were calculated using
OriginPro v8.5. The calculations for IAST can be derived from the
fitting adsorption isotherms with an appropriate Langmuir model.
The absolute component loadings were fitted with either a single-
site Langmuir model or a dual-site Langmuir model. The single-site
Langmuir model was applied when there were no noticeable iso-
therm inflections. These models were fitted purely on the basis of
giving the best fit with adjusted R2 values exceeding 0.98. The Ori-
ginPro 8.5 was used to solve and calculating following equations.
The single-site Langmuir model can be defined as:
qsatbp
1 þ bp
ð1Þ
q ¼
The dual-site Langmuir model can be defined as:
qsat;AbAp qsat;BbBp
1 þ bAp 1 þ bBp
ð2Þ
q ¼ qA þ qB ¼
þ
[5] H. H. Fei, J. F. Cahill, K. A. Prather, S. M. Cohen, Inorg. Chem. 2013, 52,
4011–4016.
in which q is molar loading of adsorbate; qsat is the saturation load-
ing; b is parameter in the pure component Langmuir adsorption
isotherm, whereas A and B refer to two different sites.
[6] a) D. Gopalakrishnan, W. R. Dichtel, J. Am. Chem. Soc. 2013, 135, 8357–
8362; b) J. X. Jiang, F. Su, A. Trewin, C. D. Wood, N. L. Campbell, H. Niu,
C. Dickinson, A. Y. Ganin, M. J. Rosseinsky, Y. Z. Khimyak, A. I. Cooper,
Chem. Eur. J. 2014, 20, 772 – 780
www.chemeurj.org
779
ꢀ 2014 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

Full Paper
Angew. Chem. 2007, 119, 8728–8732; Angew. Chem. Int. Ed. 2007, 46,
8574–8578.
[7] a) T. A. Makal, J. R. Li, W. G. Lu, H. C. Zhou, Chem. Soc. Rev. 2012, 41,
7761–7779; b) D. C. Wu, F. Xu, B. Sun, R. W. Fu, H. K. He, K. Matyjaszew-
ski, Chem. Rev. 2012, 112, 3959–4015.
[8] a) M. Carta, R. Malpass-Evans, M. Croad, Y. Rogan, J. C. Jansen, P. Bernar-
do, F. Bazzarelli, N. B. McKeown, Science 2013, 339, 303–307; b) N. B.
McKeown, P. M. Budd, Macromolecules 2010, 43, 5163–5176.
[9] R. Dawson, E. Stockel, J. R. Holst, D. J. Adams, A. I. Cooper, Energy Envi-
ron. Sci. 2011, 4, 4239–4245.
[10] a) A. I. Cooper, Angew. Chem. 2011, 123, 1028–1030; Angew. Chem. Int.
Ed. 2011, 50, 996–998; b) M. Mastalerz, Chem. Eur. J. 2012, 18, 10082–
10091.
[11] M. G. Rabbani, H. M. El-Kaderi, Chem. Mater. 2011, 23, 1650–1653.
[12] P. Kuhn, M. Antonietti, A. Thomas, Angew. Chem. 2008, 120, 3499–3502;
Angew. Chem. Int. Ed. 2008, 47, 3450–3453.
[13] T. Ben, H. Ren, S. Q. Ma, D. P. Cao, J. H. Lan, X. F. Jing, W. C. Wang, J. Xu,
F. Deng, J. M. Simmons, S. L. Qiu, G. S. Zhu, Angew. Chem. 2009, 121,
9621–9624; Angew. Chem. Int. Ed. 2009, 48, 9457–9460.
[14] D. Q. Yuan, W. G. Lu, D. Zhao, H. C. Zhou, Adv. Mater. 2011, 23, 3723–
3725.
[29] a) A. Coskun, D. C. Friedman, H. Li, K. Patel, H. A. Khatib, J. F. Stoddart, J.
Am. Chem. Soc. 2009, 131, 2493–2495; b) R. Zhao, C. Y. Tan, Y. H. Xie,
C. M. Gao, H. X. Liu, Y. Y. Jiang, Tetrahedron Lett. 2011, 52, 3805–3809.
[30] P. M. Budd, E. S. Elabas, B. S. Ghanem, S. Makhseed, N. B. McKeown, K. J.
Msayib, C. E. Tattershall, D. Wang, Adv. Mater. 2004, 16, 456–459.
[31] M. Rose, N. Klein, W. Bohlmann, B. Bohringer, S. Fichtner, S. Kaskel, Soft
Matter 2010, 6, 3918–3923.
[32] a) J. W. Colson, W. R. Dichtel, Nat. Chem. 2013, 5, 453–465; b) E. Deniz,
F. Karadas, H. A. Patel, S. Aparicio, C. T. Yavuz, M. Atilhan, Microporous
Mesoporous Mater. 2013, 175, 34–42; c) F. Karadas, H. El-Faki, E. Deniz,
C. T. Yavuz, S. Aparicio, M. Atilhan, Microporous Mesoporous Mater. 2012,
162, 91–97; d) S. Kitagawa, R. Kitaura, S. Noro, Angew. Chem. 2004, 116,
2388–2430; Angew. Chem. Int. Ed. 2004, 43, 2334–2375; e) K. Sumida,
D. L. Rogow, J. A. Mason, T. M. McDonald, E. D. Bloch, Z. R. Herm, T. H.
Bae, J. R. Long, Chem. Rev. 2012, 112, 724–781.
[33] J. R. Li, J. Sculley, H. C. Zhou, Chem. Rev. 2012, 112, 869–932.
[34] a) H. M. El-Kaderi, J. R. Hunt, J. L. Mendoza-Cortes, A. P. Cote, R. E. Taylor,
M. O’Keeffe, O. M. Yaghi, Science 2007, 316, 268–272; b) H. Furukawa,
O. M. Yaghi, J. Am. Chem. Soc. 2009, 131, 8875–8883.
[35] a) T. H. Bae, M. R. Hudson, J. A. Mason, W. L. Queen, J. J. Dutton, K.
Sumida, K. J. Micklash, S. S. Kaye, C. M. Brown, J. R. Long, Energy Environ.
Sci. 2013, 6, 128–138; b) M. G. Rabbani, A. K. Sekizkardes, O. M. El-Kadri,
B. R. Kaafarani, H. M. El-Kaderi, J. Mater. Chem. 2012, 22, 25409–25417.
[36] R. R. Pawar, H. A. Patel, G. Sethia, H. C. Bajaj, Appl. Clay Sci. 2009, 46,
109–113.
[15] F. J. Uribe-Romo, J. R. Hunt, H. Furukawa, C. Klock, M. O’Keeffe, O. M.
Yaghi, J. Am. Chem. Soc. 2009, 131, 4570–4571.
[16] A. P. Katsoulidis, M. G. Kanatzidis, Chem. Mater. 2011, 23, 1818–1824.
[17] a) H. A. Patel, F. Karadas, J. Byun, J. Park, E. Deniz, A. Canlier, Y. Jung, M.
Atilhan, C. T. Yavuz, Adv. Funct. Mater. 2013, 23, 2270–2276; b) H. A.
Patel, F. Karadas, A. Canlier, J. Park, E. Deniz, Y. Jung, M. Atilhan, C. T.
Yavuz, J. Mater. Chem. 2012, 22, 8431–8437.
[18] Q. Chen, M. Luo, P. Hammershoj, D. Zhou, Y. Han, B. W. Laursen, C. G.
Yan, B. H. Han, J. Am. Chem. Soc. 2012, 134, 6084–6087.
[19] M. Rose, W. Bohlmann, M. Sabo, S. Kaskel, Chem. Commun. 2008, 2462–
2464.
[20] Z. G. Xie, C. Wang, K. E. deKrafft, W. B. Lin, J. Am. Chem. Soc. 2011, 133,
2056–2059.
[21] O. K. Farha, Y. S. Bae, B. G. Hauser, A. M. Spokoyny, R. Q. Snurr, C. A.
Mirkin, J. T. Hupp, Chem. Commun. 2010, 46, 1056–1058.
[22] D. S. Zhang, Z. Chang, Y. B. Lv, T. L. Hu, X. H. Bu, RSC Adv. 2012, 2, 408–
410.
[37] M. G. Rabbani, T. E. Reich, R. M. Kassab, K. T. Jackson, H. M. El-Kaderi,
Chem. Commun. 2012, 48, 1141–1143.
[38] J. P. Perdew, K. Burke, M. Ernzerhof, Phys. Rev. Lett. 1996, 77, 3865–
3868.
[39] S. Grimme, J. Comput. Chem. 2006, 27, 1787–1799.
[40] Y. Shao, L. F. Molnar, Y. Jung, J. Kussmann, C. Ochsenfeld, S. T. Brown,
A. T. B. Gilbert, L. V. Slipchenko, S. V. Levchenko, D. P. O’Neill, R. A. DiSta-
sio, R. C. Lochan, T. Wang, G. J. O. Beran, N. A. Besley, J. M. Herbert, C. Y.
Lin, T. Van Voorhis, S. H. Chien, A. Sodt, R. P. Steele, V. A. Rassolov, P. E.
Maslen, P. P. Korambath, R. D. Adamson, B. Austin, J. Baker, E. F. C. Byrd,
H. Dachsel, R. J. Doerksen, A. Dreuw, B. D. Dunietz, A. D. Dutoi, T. R. Fur-
lani, S. R. Gwaltney, A. Heyden, S. Hirata, C. P. Hsu, G. Kedziora, R. Z.
Khalliulin, P. Klunzinger, A. M. Lee, M. S. Lee, W. Liang, I. Lotan, N. Nair,
B. Peters, E. I. Proynov, P. A. Pieniazek, Y. M. Rhee, J. Ritchie, E. Rosta,
C. D. Sherrill, A. C. Simmonett, J. E. Subotnik, H. L. Woodcock, W. Zhang,
A. T. Bell, A. K. Chakraborty, D. M. Chipman, F. J. Keil, A. Warshel, W. J.
Hehre, H. F. Schaefer, J. Kong, A. I. Krylov, P. M. W. Gill, M. Head-Gordon,
Phys. Chem. Chem. Phys. 2006, 8, 3172–3191.
[41] J. T. A. Jones, D. Holden, T. Mitra, T. Hasell, D. J. Adams, K. E. Jelfs, A.
Trewin, D. J. Willock, G. M. Day, J. Bacsa, A. Steiner, A. I. Cooper, Angew.
Chem. 2011, 123, 775–779; Angew. Chem. Int. Ed. 2011, 50, 749–753.
[42] a) C. M. Nagaraja, R. Haldar, T. K. Maji, C. N. R. Rao, Cryst. Growth Des.
2012, 12, 975–981; b) V. Zelenak, Z. Vargova, M. Almasi, A. Zelenakova,
J. Kuchar, Microporous Mesoporous Mater. 2010, 129, 354–359.
[23] Y. Zhu, H. Long, W. Zhang, Chem. Mater. 2013, 25, 1630–1635.
[24] M. G. Schwab, B. Fassbender, H. W. Spiess, A. Thomas, X. L. Feng, K.
Mullen, J. Am. Chem. Soc. 2009, 131, 7216–7217.
[25] a) O. K. Farha, A. M. Spokoyny, B. G. Hauser, Y. S. Bae, S. E. Brown, R. Q.
Snurr, C. A. Mirkin, J. T. Hupp, Chem. Mater. 2009, 21, 3033–3035; b) S.
Wan, F. Gandara, A. Asano, H. Furukawa, A. Saeki, S. K. Dey, L. Liao,
M. W. Ambrogio, Y. Y. Botros, X. F. Duan, S. Seki, J. F. Stoddart, O. M.
Yaghi, Chem. Mater. 2011, 23, 4094–4097; c) L. H. Xie, M. P. Suh, Chem.
Eur. J. 2013, 19, 11590–11597; d) C. H. Lee, M. R. Tsai, Y. T. Chang, L. L.
Lai, K. L. Lu, K. L. Cheng, Chem. Eur. J. 2013, 19, 10573–11579.
[26] W. G. Lu, J. P. Sculley, D. Q. Yuan, R. Krishna, Z. W. Wei, H. C. Zhou,
Angew. Chem. 2012, 124, 7598–7602; Angew. Chem. Int. Ed. 2012, 51,
7480–7484.
[27] M. G. Rabbani, H. M. El-Kaderi, Chem. Mater. 2012, 24, 1511–1517.
[28] H. A. Patel, S. H. Je, J. Park, D. P. Chen, Y. Jung, C. T. Yavuz, A. Coskun,
Nat. Commun. 2013, 4, 1357.
Received: September 5, 2013
Published online on December 11, 2013
Chem. Eur. J. 2014, 20, 772 – 780
www.chemeurj.org
780
ꢀ 2014 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim