Comparison of gas sorption properties of neutral and anionic metal-organic frameworks prepared from the same building blocks but in different solvent systems

Article abstract of DOI:10.1002/chem.201303086

Two different 3D porous metal-organic frameworks, [Zn₄O(NTN) ₂]×10 DMA×7 H₂O (SNU-150) and [Zn ₅(NTN)₄(DEF)₂][NH₂(C ₂H₅)₂]₂×8 DEF×6 H ₂O (SNU-151), are synthesized from the same metal and organic building blocks but in different solvent systems, specifically, in the absence and the presence of a small amount of acid. SNU-150 is a doubly interpenetrated neutral framework, whereas SNU-151 is a non-interpenetrated anionic framework containing diethylammonium cations in the pores. Comparisons of the N ₂, H₂, CO₂, and CH₄ gas adsorption capacities as well as the CO₂ adsorption selectivity over N ₂ and CH₄ in desolvated SNU-150 (BET: 1852 m² g^-1) and SNU-151 (BET: 1563 m² g^-1) samples demonstrate that the charged framework is superior to the neutral framework for gas storage and gas separation, despite its smaller surface area and different framework structure. Choice selection: A neutral metal-organic framework (MOF) (SNU-150) and an anionic MOF with a cationic guest (SNU-151; see figure) have been prepared from the same molecular building blocks but in different solvents. Despite its smaller surface area, SNU-151 exhibits higher uptake capacities, higher isosteric heats of H₂, CO₂, and CH₄ adsorption, and greater CO₂ adsorption selectivities over N ₂ and CH₄ than SNU-150, because of the charged framework and guests.

Full text of DOI:10.1002/chem.201303086

DOI: 10.1002/chem.201303086
Comparison of Gas Sorption Properties of Neutral and Anionic Metal–
Organic Frameworks Prepared from the Same Building Blocks but in
Different Solvent Systems
Myung-Ho Choi, Hye Jeong Park, Dae Ho Hong, and Myunghyun Paik Suh*[a]
Abstract: Two different 3D porous
metal–organic frameworks, [Zn4O-
interpenetrated neutral framework,
whereas SNU-151 is a non-interpene-
trated anionic framework containing
diethylammonium cations in the pores.
Comparisons of the N2, H2, CO2, and
CH4 gas adsorption capacities as well
as the CO2 adsorption selectivity over
N2 and CH4 in desolvated SNU-150’
(BET: 1852 m2 gꢀ1) and SNU-151’
(BET: 1563 m2 gꢀ1) samples demon-
strate that the charged framework is
superior to the neutral framework for
gas storage and gas separation, despite
its smaller surface area and different
framework structure.
A
(SNU-150)
(DEF)2][NH2-
E
N
ACHTUNGTRENNUNG
ACHTUNGTRENNUNG
synthesized from the same metal and
organic building blocks but in different
solvent systems, specifically, in the ab-
sence and the presence of a small
amount of acid. SNU-150 is a doubly
Keywords: adsorption · carbon di-
oxide · cationic guests · gas separa-
tion · metal–organic frameworks
Introduction
Here, we report two porous MOFs, [Zn4O-
(NTN)2]·10DMA·7H2O (SNU-150) and [Zn5A(NTN)4-
(DEF)2][NH2A(C2H5)2]2·8DEF·6H2O (SNU-151) (H3NTN=
6,6’,6’’-nitrilotri-2-naphthoic acid, DMA=N,N-dimethylacet-
AHCTUNGTREGaNNUN mide, DEF=N,N-diethylformamide), which are synthe-
sized by using the same metal and organic building blocks
but different solvent systems, in particular, in the absence
and presence of a small amount of acetic acid (Scheme 1).
SNU-150 is a neutral framework, whereas SNU-151 is an
anionic framework containing diethylammonium cations as
G
CHTUNGTRENNUNG
Metal–organic frameworks (MOFs) have attracted great at-
tention because of their potential applications in gas stor-
age[1–6] and separation.[7–14] The gas adsorption capacities of
MOFs are enhanced by the increase in surface areas, pres-
ence of accessible metal sites,[6,15] and functional groups.[16]
The gas separation capabilities of MOFs, in particular for
CO2 separation from flue gas and landfill gas, are affected
by the flexibility of the frameworks[7,17] and the ions or mol-
ecules contained in the pores of the MOFs,[18–20] because
CO2 molecules can interact with the MOF much more
strongly than other gases because of the high quadrupole
moment (14.3ꢀ10ꢀ40 Cm2) and polarizability (26.3ꢀ
10ꢀ25 cm3). Recently, it has been reported that anionic
MOFs containing metal cations in their pores show higher
gas sorption capacities and significantly enhanced CO2 ad-
sorption selectivities over other gases than common neutral
MOFs.[18] However, the synthetic method for obtaining
charged MOFs exclusively from neutral MOFs has not yet
been well established. Moreover, simple comparison of the
gas sorption properties between charged MOFs and neutral
MOFs encounters difficulties owing to the large differences
in their framework structures and surface areas, because
they are constructed from different metal ions and organic
building blocks.
R
G
CHTUNGTRENNUNG
ACHTUNGTRENNUNG
guests. Desolvated solid SNU-151’ ([Zn
5ACHTUNTGRENNUG(NTN)4]ACHTUNGTRENNUNG[NH2-
AHCTUNGTRENNUNG
higher adsorption capacities and higher isosteric heats of H2,
CO2, and CH4 gas adsorption, as well as higher CO2 adsorp-
tion selectivities over N2 and CH4 gases, than SNU-150’
([Zn4OACTHNUGTRNE(UNG NTN)2]), because of the stronger interaction of the
gas molecules with the anionic framework and the ammoni-
um cations included in the channels.
Scheme 1. Preparation of SNU-150 and SNU-151.
[a] M.-H. Choi, H. J. Park, D. H. Hong, Prof. M. P. Suh
Department of Chemistry, Seoul National University
Seoul 151-747 (Republic of Korea)
Fax : (+82)2-886-8516
Results and Discussion
E-mail: mpsuh@snu.ac.kr
Synthesis and X-ray crystal structures of neutral (SNU-150)
and charged (SNU-151) frameworks: Greenish truncated-oc-
Supporting information for this article is available on the WWW
under http://dx.doi.org/10.1002/chem.201303086.
17432
ꢁ 2013 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Chem. Eur. J. 2013, 19, 17432 – 17438

FULL PAPER
Figure 1. PXRD patterns of the products formed from the solvothermal
reaction of Zn(OAc)2 2H2O (67.3 mg, 0.3 mmol) and H3NTN (26.3 mg,
ACHTUNGTRENNUNG
0.05 mmol) in DEF (1.5 mL) in the presence of various amount of acetic
acid: SNU-150 is formed in the presence of less than 75 mL (1.3 mmol) of
acetic acid, whereas SNU-151 is formed in the presence of 75 mL
(1.3 mmol) to 300 mL (5.2 mmol) of acetic acid. Reaction conditions
(from top to bottom): 908C, no acid; 908C, acetic acid (20 mL); 908C,
acetic acid (30 mL); 908C, acetic acid (50 mL); 908C, acetic acid (60 mL);
908C, acetic acid (75 mL); 908C, acetic acid (250 mL); 908C, acetic acid
(300 mL).
Figure 2. X-ray crystal structures of SNU-150, SNU-150’, and SNU-151.
a) View of SNU-150 on the ab plane. b) View of SNU-150’ on the ab
plane. c) View of SNU-151 on the ab plane. d) View of SNU-151 on the
ac plane. Tetrahedral Zn, orange; octahedral Zn, yellow.
When single crystals of SNU-150 were activated by using
supercritical CO2 fluid, a desolvated sample of SNU-150’
was obtained, as shown by IR data, elemental analysis
(EA), and thermogravimetric analysis (TGA). Interestingly,
SNU-150 undergoes single-crystal-to-single-crystal transfor-
mation during the activation process (Figure 2b). The
single-crystal X-ray structure of activated SNU-150’ exhibits
a fine structure that differs from that of SNU-150. Many of
the key dihedral angles of SNU-150’ become different from
those of SNU-150, and the pore size is enlarged to approxi-
mately 5.9ꢀ2.7 ꢂ (compared to 4.5ꢀ1.7 ꢂ for as-synthe-
sized SNU-150) (see Table S1 and Figure S6 in the Support-
tahedral crystals of SNU-150 were prepared by heating
a DMA solution of Zn
for 24 h. Brown rhombus-shaped crystals of SNU-151 were
synthesized by heating a DEF solution of Zn(OAc)2·2H2O
ACHTUGNTREN(NUNG OAc)2·2H2O and H3NTN at 908C
AHCTUNGTRENNUNG
and H3NTN in the presence of acetic acid (DEF/acetic
acid=100:3–100:12, v/v) at 908C for 24 h. When the volume
ratio of acetic acid/DEF was less than 2.4%, SNU-150 was
formed instead of SNU-151, as shown by the powder X-ray
diffraction (PXRD) data (Figure 1). That is, the addition of
a small amount of acetic acid
to the reaction mixture leads
Table 1. Crystallographic data of SNU-150, SNU-150’, and SNU-151 (squeezed).
SNU-150 SNU-150’
Zn4C66H36N2O13 Zn4C66H36 N2O13
to
a totally different MOF.
SNU-151
SNU-150 and SNU-151 are in-
soluble in common organic sol-
vents.
In the single-crystal X-ray
structure of SNU-150 (Fig-
formula
crystal system
space group
Mr
Zn5C142H94N6O26
triclinic
cubic
Ia3
cubic
Pa3
¯
¯
¯
P1
1326.45
27.143(3)
27.143(3)
27.143(3)
90
1326.45
27.3095(6)
27.3095(6)
27.3095(6)
90
2627.08
a [ꢂ]
b [ꢂ]
c [ꢂ]
a [8]
16.158(3)
18.171(4)
18.955(4)
99.19(3)
106.28(3)
116.34(3)
4518.8(27)
1
0.965
100(2)
0.69999
0.646
ure 2a), the Zn4OACHTUNGTRENNUNG(CO2)6 clus-
ter acts as an octahedral secon-
dary building unit (SBU), and
NTN3ꢀ acts as a triangular or-
ganic building block, which re-
sults in a (6.3)-connected net
with a PdF2-type net topology,
similarly to those reported pre-
viously.[21–23] SNU-150 is doubly
interpenetrated, and generates
3D channels that extend along
the three orthogonal direc-
tions. The effective window
size of the channels is 4.5ꢀ
1.7 ꢂ2. The void volume of
SNU-150 is 49.5%, as calculat-
ed by PLATON (Table 1).[24]
b [8]
90
90
g [8]
90
19997(4)
8
0.881
100(2)
90
V [ꢂ3]
Z
20367.7(8)
8
0.865
1calcd [gcmꢀ3
T [K]
]
120(2)
l [ꢂ]
0.80003
1.316
1.271
0.71073
0.970
0.999
m [mmꢀ1
]
GOF (F2)
(000)
1.539
1346
16832
F
A
5360
5360
reflections collected
R1, wR2 [I>2s(I)]
3912
11598
0.1107,[a] 0.3329[b]
0.1328,[a] 0.3499[b]
0.289/ꢀ0.350
0.0771,[a] 0.2090[c]
0.0970,[a] 0.2187[c]
4.170/ꢀ1.125
0.1059,[a] 0.3253[b]
0.1120,[a] 0.3371[b]
1.559/ꢀ3.119
R1, wR2 (all data)
largest diff. peak/hole [eꢂꢀ3
]
1
2
[a] R=Sj jFo jꢀjFc j j/SjFo j.
[b] wR(F2)=[Sw(Fo2ꢀFc2)2/Sw(F2o)2] =
where
w=1/[s2(F2o)+(0.2000P)2 +
1
2
(0.0000)P], P=(F2o +2Fc2)/3 for SNU-150 and SNU-151. [c] wR(F2)=[Sw(F2oꢀFc2)2/Sw(F2o)2] = where w=1/
[s2(Fo2)+(0.1469P)2 +(0.0000)P], P=(Fo2 +2F2c)/3 for SNU-150’.
Chem. Eur. J. 2013, 19, 17432 – 17438
ꢁ 2013 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chemeurj.org
17433

M. P. Suh et al.
ing Information), similarly to the case of SNU-77S.[21] The
powder X-ray diffraction patterns indicate that the frame-
work structure of SNU-150’ is similar to that of SNU-150
(Figure 3).
Figure 4. PXRD patterns of a) as-synthesized SNU-151, b) simulated
powder pattern of SNU-151 based on its single-crystal data, c) SNU-151’
obtained from the treatment of SNU-151 with supercritical CO2 fluid.
Figure 3. PXRD patterns of a) as-synthesized SNU-150, b) simulated
powder pattern of SNU-150 based on its single-crystal X-ray data,
c) SNU-150’ obtained by treatment of SNU-150 with supercritical CO2
fluid, and d) simulated powder pattern of SNU-150’ based on its single-
crystal X-ray data.
Gas sorption properties: The gas adsorption isotherms of
SNU-150’ and SNU-151’ were measured for N2, H2, CO2,
and CH4 gases, and the data are summarized and compared
with those of the previously reported SNU-77S[21] and SNU-
100’,[18] which have similar structures to SNU-150’ and SNU-
151’, respectively (Table 2).
The N2 sorption isotherms of SNU-150’ and SNU-151’
show type-I curves, which are characteristic of microporous
materials (Figure 5a). SNU-150’ shows a small hysteresis of
type H-4 in the N2 desorption curve, which may be associat-
ed with the narrow slit-like pores.[25] The BET (Langmuir)
surface areas of SNU-150’ and SNU-151’ are 1852 (1945)
and 1563 (1674) m2 gꢀ1, respectively. The pore volumes esti-
In the X-ray structure of SNU-151, there are three crystal-
lographically independent ZnII centers (Zn1, Zn2, and Zn3),
which are linked in the sequence Zn1–Zn2–Zn3–Zn2–Zn1
to form a Zn5 cluster unit. Zn1 and Zn2 show tetrahedral
(td) coordination geometry, being coordinated with four car-
boxylate oxygen atoms of four different NTN3ꢀ ligands. Zn3
shows octahedral coordination geometry, being coordinated
with four carboxylate oxygen atoms of four different NTN3ꢀ
ligands and two oxygen atoms of DEF solvent molecules
(see Figure S7 in the Supporting Information). The Zn5 clus-
ter units are connected by carboxylate groups to form
a non-interpenetrated anionic 3D framework generating 2D
channels. The effective window sizes of the square channels
extending along the ab plane and of the cylindrical channels
extending along the ac plane are 5.2ꢀ2.5 and 7.7ꢀ6.5 ꢂ2, re-
spectively (Figure 2c and 2d). Because the Zn5 metal cluster
consists of five ZnII ions and twelve carboxylate groups, two
diethylammonium species should be included per formula
unit of the host as the counter cationic guests. The chemical
formula of SNU-151 was determined from the EA data as
well as the IR spectrum showing ammonium peaks at
2875 cmꢀ1, because the diethylammonium cations could not
be refined through single-crystal X-ray diffraction owing to
the severe disorder. The void volume of SNU-151 is 51.5%
without taking the diethylammonium guests into consider-
ation. When SNU-151 was activated by using supercritical
CO2 fluid, two DEF molecules coordinated at the Zn3 atom
were removed together with the DEF and water guest sol-
vent molecules included in the channels. The PXRD pat-
terns indicate that the framework structure of SNU-151’ dif-
fers significantly from that of the as-synthesized SNU-151
(Figure 4).
Table 2. Gas adsorption data of SNU-150’ and SNU-151’ together with
the comparison with SNU-77S and SNU-100’.
T
[K]
P
[atm]
Adsorption capacity [wt%]
SNU-150’ SNU-77S[21] SNU-151’ SNU-100’[18]
N2
H2
77
298
77
0.95
71.1
0.546
1.54
1.00
131
55.3
0.519
2.00
1.24
26.2
0.624
1.81
1.30
1
1
1
0.40
1.79
1.01
87
Qst of H2 adsorption
5.04–4.65 7.05
6.27–5.82 8.14–7.08
[kJmolꢀ1
]
CO2
195
231
273
298
1
1
1
1
78.6
40.5
12.0
6.09
169
130
8.21
3.94
67.8
56.0
22.2
14.1
45.2
37.2
19.9
14.1
Qst of CO2 adsorption 17.1–16.5 19.9–19.4
27.1–21.0 29.3–27.7
[kJmolꢀ1
]
CH4
195
231
273
298
1
1
1
1
11.1
3.15
1.29
0.859
8.70
4.12
1.20
0.62
11.3
7.30
2.00
1.24
10.4
7.57
2.56
1.41
Qst of CH4 adsorption 12.8–12.3 14.3–14.2
18.2–16.6 26.5–22.4
[kJmolꢀ1
]
17434
www.chemeurj.org
ꢁ 2013 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Chem. Eur. J. 2013, 19, 17432 – 17438

Gas Sorption Properties of Metal–Organic Frameworks
FULL PAPER
ure 5b). The isosteric heats (Qst) of the H2 adsorption esti-
mated from the isotherms measured at 77 and 87 K by using
a virial equation[27] are 5.04–4.65 kJmolꢀ1 for SNU-150’ and
6.27–5.82 kJmolꢀ1 for SNU-151’, depending on the degree
of H2 loading.
The CO2 and CH4 gas adsorption capacities of SNU-151’
at 1 atm at T=231 and 273 K are higher than those of SNU-
150’, as shown in Table 2 and Figure 7. At 298 K and 1 atm,
SNU-151’ adsorbs 14.1 wt% of CO2, which is approximately
2.3 times higher than the CO2 uptake capacity (6.09 wt%)
of SNU-150’. The Qst values of the CO2 and CH4 adsorp-
tions were calculated by using the Clausius–Clapeyron equa-
tion, applying the polynomial equation and Langmuir–
Freundlich equation, respectively, to the adsorption iso-
therms measured at 195, 231, 273, and 298 K. The Qst values
of the CO2 and CH4 adsorptions in SNU-151’ are 27.1–
21.0 kJmolꢀ1 and 18.2–16.6 kJmolꢀ1, respectively, which are
also higher than those (17.1–16.5 and 12.8–12.3 kJmolꢀ1, re-
spectively) in SNU-150’ (see Figure S11 in the Supporting
Information).
The higher uptake capacities and higher Qst values for the
H2, CO2, and CH4 adsorptions in SNU-151’ than in SNU-
150’ must be attributed to the stronger interactions of those
gas molecules with the anionic framework and diethylam-
monium cationic guests included in SNU-151’, compared to
their interactions with the neutral framework of SNU-150’.
Figure 5. Gas sorption isotherms for SNU-150’ (green) and SNU-151’
(burgundy). a) N2 at 77 K (squares) and CO2 at 195 K (circles). b) H2 at
77 K (circles) and 87 K (triangles). Filled shapes, adsorption; open
shapes, desorption.
mated by applying the Dubinin–Radushkevich (DR) equa-
tion are 0.717 cm3 gꢀ1 for SNU-150’ and 0.614 cm3 gꢀ1 for
SNU-151’. The pore-size distributions obtained through the
Horvath–Kawazoe (HK) method suggest that SNU-150’ and
SNU-151’ have pore sizes of 6.18 and 6.08 ꢂ, respectively
(Figure 6).[26]
The H2 adsorption capacities of SNU-150’ and SNU-151’
are 1.54 and 2.00 wt%, respectively, at 77 K and 1 atm (Fig-
Figure 7. Gas sorption isotherms for SNU-150’ (green) and SNU-151’
(burgundy). a) CO2 (circles) and CH4 (diamonds) at 273 K. b) CO2 (cir-
cles), CH4 (diamonds), and N2 (squares) at 298 K. Filled shapes, adsorp-
tion; open shapes, desorption.
Figure 6. Pore-size distributions of SNU-150’ (solid line) and SNU-151’
(dotted line), estimated by using the Horvath–Kawazoe (HK) method
using the N2 sorption data measured at 77 K.
Chem. Eur. J. 2013, 19, 17432 – 17438
ꢁ 2013 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chemeurj.org
17435

M. P. Suh et al.
In addition, the pore size (6.08 ꢂ) of SNU-151’ is slightly
smaller than that (6.18 ꢂ) of SNU-150’ as estimated from
the N2 adsorption data (Figure 6), which is also responsible
for the higher Qst values of the gas adsorption in SNU-151’,
even though its activated framework structure is different
from that of SNU-150’. It is quite well recognized that
charged frameworks show superior gas sorption properties
to neutral ones.[18,28–30]
CH4 were also calculated by using IAST with the assumption
that the landfill gas has a CO2/CH4 volume ratio of 50:50 at
298 K. The values are 7.20 for SNU-151’ and 2.26–2.35 for
SNU-150’ depending on the pressure (up to 1 atm) at 298 K.
SNU-151’ shows a higher selectivity for CO2 adsorption over
CH4 than [Cu
A
(SiF6)][32] (bpy-2=1,2-bis(4-pyridyl)e-
2A
(DABCO)] (DMOF) series.[33]
thene) and the [Zn
C
CHTUNGTRENNUNG
When the gas adsorption properties of SNU-150’ are com-
pared with those of SNU-77S,[21] which also has PdF2 topolo-
gy, it is found that SNU-150’ adsorbs much smaller amounts
of N2 and H2 gases at 77 K and 1 atm, as well as a smaller
amount of CO2 gas at 195 K and 1 atm. This must be attrib-
uted to the significantly smaller surface area (1852 m2 gꢀ1) of
SNU-150’ than that (3670 m2 gꢀ1) of SNU-77S. When the gas
adsorption properties of SNU-151’ are compared with those
Conclusion
We have prepared two different porous MOFs, SNU-150
and SNU-151, from the same metal and organic building
blocks but in the presence and absence of acetic acid. SNU-
150 is a neutral framework, whereas SNU-151 is an anionic
framework containing diethylammonium guests. SNU-150
undergoes a single-crystal-to-single-crystal transformation
upon guest removal, which affords SNU-150’ with a slightly
altered structure. SNU-151’, which is formed upon the re-
moval of guest solvent as well as coordinated solvent mole-
cules from SNU-151, has a different structure from SNU-
151. SNU-151’ shows higher uptake capacities for H2, CO2,
and CH4 gases as well as higher isosteric heats of those gas
adsorptions than SNU-150’, despite the smaller surface area
of SNU-151’ (BET: 1563 m2 gꢀ1) than that of SNU-150’
(BET: 1852 m2 gꢀ1). These properties are attributed to the
charged framework and smaller pore size of SNU-151’. Fur-
thermore, SNU-151’ shows a significantly higher CO2/N2 ad-
sorption selectivity at 298 K than SNU-150’. The present re-
sults suggest that for gas storage and gas separation applica-
tions, the synthesis of charged frameworks rather than neu-
tral ones should be recommended, regardless of the frame-
work structures.
of SNU-100’, which has an anionic framework with dieth
ACHTUNGERTNyNUNG l-
ACHTUNGTRENNUNGammonium cations, it is observed that SNU-151’ is superior
to SNU-100’.[18] This is also attributed to the higher surface
area (1563 m2 gꢀ1) of SNU-151’ than that (814 m2 gꢀ1) of
SNU-100’. However, at 298 K and 1 atm, the CO2 and CH4
gas uptake capacities of SNU-150’ and SNU-151’ become
comparable to those of SNU-77S and SNU-100’, respective-
ly.
SNU-150’ and SNU-151’ adsorb CO2 gas selectively at
298 K, with hardly any adsorption of N2 and CH4 gases (Fig-
ure 7b) (see Figure S10 in the Supporting Information). Ac-
cording to the calculation using the ideal adsorbed solution
theory (IAST) with the assumption that the flue gas has
a CO2/N2 volume ratio of 15:85, the selectivity values of the
CO2/N2 adsorption at 298 K are 30.0 for SNU-151’ and 5.44–
6.42 for SNU-150’ depending on the pressure (up to 1 atm)
(Figure 8). The present result is consistent with our previous
report that an anionic framework (SNU-100’) with cationic
guests showed very high CO2 uptake selectivity over N2.[18]
The CO2/N2 adsorption selectivity of SNU-151’ is higher
than those of SNU-100’ (26.5)[18] and H
3ACHTUNGTRNENUG[(Cu4Cl)3ACHTUNGTNER(NUGN BTTri)8]
Experimental Section
(21.0).[31] The selectivity values of the CO2 adsorption over
General methods: Anhydrous 1,4-dioxane was purchased from Sigma Al-
drich, and used without further purification. Other chemicals and sol-
vents used in the syntheses were of reagent grade and were used without
further purification. Infrared spectra were recorded with a Perkin–Elmer
2000 FTIR spectrophotometer. Elemental analyses were recorded with
1
a Perkin–Elmer 2400 Series II CHNS/O Analyzer. H NMR spectra were
measured on Bruker Advance DPX-300 spectrometer (300 MHz) at
room temperature. Thermogravimetric analysis (TGA) and differential
scanning calorimetry (DSC) were performed under N2 (g) at a scan rate
of 58C with a TGA Q50 and DSC Q10, respectively (TA instruments).
Powder X-ray diffraction (PXRD) data were recorded on a Bruker
D5005 diffractometer at 40 kV and 40 mA (CuKa, l=1.54050 ꢂ) with
a scan speed of 0.5 8minꢀ1 and a step size of 0.028 (2q).
Synthesis of H3NTN: H3NTN (6,6’,6’’-nitrilotri-2-naphthoic acid) was pre-
pared as follows through modification of the previously reported meth-
ods.[34]
Synthesis of trimethyl 6,6’,6’’-nitrilotri-2-naphthoate: Under a N2 atmos-
phere, PdACHTUNRGTNE(NUG OAc)2 (129 mg, 0.575 mmol) and rac-2,2’-bis(diphenylphosphi-
Figure 8. Selectivities of SNU-150’ (green) and SNU-151’ (burgundy) for
CO2/N2 (circles) and CO2/CH4 (stars) adsorption depending on the pres-
sure, calculated by using ideal adsorbed solution theory (IAST) with the
assumption that the flue gas has a CO2/N2 volume ratio of 15:85 and
landfill gas has a CO2/CH4 volume ratio of 50:50.
no)-1,1’-binaphthyl) (rac-BINAP) (716 mg, 1.15 mmol) were dissolved in
anhydrous 1,4-dioxane (100 mL). After stirring for 15 min, methyl 6-
bromo-2-naphthoate (8.16 g, 30.8 mmol, 2.8 equiv), methyl 6-amino-2-
naphthoate (2.32 g, 11.5 mmol, 1 equiv), and Cs2CO3 (5.63 g, 29.2 mmol,
2.5 equiv) were added. The solution was heated at reflux for five days,
17436
www.chemeurj.org
ꢁ 2013 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Chem. Eur. J. 2013, 19, 17432 – 17438

Gas Sorption Properties of Metal–Organic Frameworks
FULL PAPER
cooled to room temperature, and diluted with CH2Cl2 (200 mL). The
crude mixture was filtered through celite, concentrated through evapora-
tion of the solvent, and then purified by silica gel column chromatogra-
phy with n-hexane/CH2Cl2 (2:1, v/v) and then CH2Cl2 as eluents. The
product was obtained as a light yellow powder. 1H NMR (CDCl3): d=
3.99 (s, 9H), 7.44 (d, J=8.8 Hz, 3H), 7.60–7.62 (m, 6H), 7.90 (d, J=
8.9 Hz, 3H), 8.03 (d, J=8.6 Hz, 3H), 8.60 ppm (s, 3H).
used for data collection (SNU-150: detector distance, 62 mm; omega
scan, Dw=18; exposure time, 10 s/frame; SNU-151: detector distance,
70 mm; omega scan, Dw=18; exposure time, 10 s/frame), and
HKL3000 sm (Ver. 703r)[36] was used for cell refinement, reduction, and
absorption corrections. The diffraction data of the activated sample
[Zn4OACHTUNTRGNEUNG(NTN)2] (SNU-150’) were collected at 100 K with an Enraf–Nonius
Kappa CCD diffractometer (MoKa, l=0.71073 ꢂ, graphite monochroma-
tor) by coating the crystal with Paratone-N oil. Preliminary orientation
matrices and unit-cell parameters were obtained from the peaks of the
first ten frames, and then refined by using the whole data set. Frames
were integrated and corrected for Lorentz and polarization effects by
using DENZO.[36] The scaling and global refinement of crystal parame-
ters were performed by using SCALEPACK.[36] No adsorption correction
was made. The crystal structures of SNU-150, SNU-151, and SNU-150’
were solved by direct methods[37] and refined through full-matrix least-
squares refinement using the SHELXL-97 program.[38] The hydrogen
atoms were positioned geometrically by using a riding model. The elec-
tron densities of the disordered guest molecules were flattened by using
the SQUEEZE option of PLATON.[39] In SNU-150, the overall frame-
work was statistically disordered over two sites with 1:1 occupancies.
Consequently, all the atoms were given by 0.5 occupancy, except Zn(1),
O(1), O(1)’, N(1), and N(1)’, which sit on threefold crystallographic axes,
and were given by the occupancy of 1/6. In SNU-150’, the site occupancy
factors were given as 0.33333 for the Zn(1), O(1), N(1), and N(2) atoms,
which sit on threefold crystallographic axes. CCDC-917750 (SNU-150),
917751 (SNU-150’), and 917752 (SNU-151) contain the supplementary
crystallographic data 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 6,6’,6’’-nitrilotri-2-naphthoic acid (H3NTN): Trimethyl
6,6’,6’’-nitrilotri-2-naphthalate was dissolved in MeOH, and then a saturat-
ed aqueous solution of NaOH was added until the solution reached
pH 8. The solution was heated at reflux for one day, and then the solvent
was removed by evaporation. The crude product was dissolved with dis-
tilled water, and HCl was added until the solution reached pH 2. The so-
lution was cooled to room temperature, and the solid formed was filtered
off, washed with water, and dried in vacuo. 1H NMR ([D6]DMSO): d=
7.40 (d, J=8.4 Hz, 3H), 7.64 (s, J=35.7 Hz, 3H), 7.76 ꢀ7.90 (m, 6H),
8.00 (d, J=9 Hz, 3H), 8.03 (d, J=9 Hz, 3H), 8.46 ppm (s, J=25.8 Hz,
3H); elemental analysis calcd (%) for C33H21NO6: C 75.13, H 4.01, N
2.66; found: C 73.01, H 4.06, N 2.56.
Synthesis of [Zn4O
ACHTUNGTRENNUNG
tion (1.5 mL) of ZnACHTUNGTRENNUNG
lution (1 mL) of H3NTN (26.3 mg, 0.05 mmol) were added in a serum
glass bottle. The bottle was sealed and heated in an oven at 908C for
24 h, and truncated octahedral green crystals of [Zn4O-
ACHTUNGTRENNUNG(NTN)2]·10DMA·7H2O were obtained. When DEF was used as the sol-
vent instead of DMA with the same reaction mixture, the product was
formed as a polycrystalline powder, the PXRD data of which indicated
that it had the same framework structure as SNU-150. Yield: 42.5 mg
ꢀ
(73.2% based on H3NTN); FTIR for SNU-150 (KBr): n˜ =3401 (O H),
3053 (C H(NTN)), 2931 (C H(DMA)), 1626, 1587 cmꢀ1 (O=C O); elemental
analysis calcd (%) for Zn4C106H140N12O30: C 54.78, H 6.07, N 7.23; found:
C 54.46, H 5.90, N 7.33.
ꢀ
ꢀ
ꢀ
Supercritical CO2 activation method: Before being dried, as-synthesized
crystals, which were still in the mother liquor, were transferred to a vial
(20 mL). The mother liquor was decanted and the crystals were washed
briefly with pure solvent (2ꢀ15 mL). The crystals were placed inside the
supercritical dryer together with the solvent, and the drying chamber was
sealed. The temperature and pressure of the chamber were raised to
458C and 200 bar with CO2, above the critical point (31.88C, 73 atm) of
CO2. The chamber was vented at a rate of 10 mLminꢀ1 and then filled
with CO2 again. The cycles of refilling with CO2, pressurizing, and vent-
ing were repeated for 6 h. After drying, the closed container with the
dried crystals was transferred to a glove bag to avoid exposure of the
crystals to air. The gas sorption isotherms of the samples were measured
without further activation.
Preparation of [Zn4OACHTUNGTRENNUNG(NTN)2] (SNU-150’): Before being dried, crystals
of as-synthesized SNU-150 were transferred into a vial (20 mL) together
with the mother liquor. The mother liquor was decanted and the crystals
were washed briefly with pure DMA (2ꢀ15 mL). The sample was desol-
vated by using supercritical CO2 fluid. FTIR for SNU-150’ (KBr): n˜ =
3052 (C H(NTN)), 1623, 1589 cmꢀ1 (O=C O); elemental analysis calcd
(%) for Zn4C66H36O13N2: C 57.67, H 2.64, N 2.04; found: C 57.36, H 2.66,
N 2.24.
ꢀ
ꢀ
Synthesis of [Zn
(NTN)
E
N
2CATHUNGTERN(UNNG C2H5)2]2·8DEF·6H2O (SNU-151):
The DEF solution (1.5 mL) of ZnAHCNUTGTRENNUNG
mixed with the DEF solution (1 mL) of H3NTN (26.3 mg, 0.05 mmol) in
a serum glass bottle, and then acetic acid (0.1 mL) was added. The reac-
tion bottle was sealed and heated in an oven at 908C for 24 h, and rhom-
Gas sorption measurements: The gas adsorption/desorption experiments
were performed with an automated micropore gas analyzer, Autosorb-3B
(Quantachrome Instruments). All the gases used were of 99.999% purity.
SNU-150’ and SNU-151’ were pre-desolvated by using supercritical CO2
fluid at 458C for 8 h. The predried solid was introduced to a gas sorption
cell, the weight of which was measured exactly, and then the gas sorption
isotherms were measured. The sample weight was measured precisely
after the measurement of gas sorption. Between the experiments with
various gases, the out-gassing procedure was repeated for approximately
1 h. The N2 gas sorption isotherms were monitored at 77 K by using
liquid nitrogen, and the H2 gas sorption isotherms were monitored at 77
and 87 K, at each equilibrium pressure, by the static volumetric method.
The adsorption isotherms for CO2 and CH4 gases were measured at 195,
231, 273, and 298 K. The sorption properties, including pore volume, pore
size, and surface area were analyzed by using Autosorb 1 for Windows
1.24 software.
bus-shaped
brown
crystals
of
[Zn5ACHTUNTGERG(NUNN NTN)4AHCUTNTRGEG(NNUN DEF)2]CAHTNUGTREN[NUGN NH2-
ACHTUNGTRENNUNG
ꢀ
ꢀ
H3NTN); FTIR for SNU-151 (KBr pellet): n˜ =3429 (O H), 3054 (C
ꢀ
ꢀ
H(NTN)), 2975, 2935 (C H(NTN)), 2875 (N H(diethylammonium)), 1659 (C=
O
(DEF)), 1625, 1594 cmꢀ1 (O=C O); elemental analysis calcd (%) for
Zn5C190H208N16O35: C 61.80, H 5.95, N 6.07; found: C 61.39, H 5.77, N
5.95.
ꢀ
Preparation of [Zn
tion, crystals
(C2H5)2]2·8DEF·6H2O were transferred to a vial (20 mL) together with
5A
CATHNUGTREN[GNUN NH2AHCTUNRTGEG(NNUN C2H5)2]2 ACTHUNGTRENNUGN
of as-synthesized
N
N
ACHTUNGTRENNUNG
ACHTUNGTRENNUNG
the mother liquor. The mother liquor was decanted and the crystals were
washed briefly with pure DEF (2ꢀ15 mL). The sample was desolvated
by using supercritical CO2 fluid. FTIR for SNU-151’ (KBr pellet): n˜ =
ꢀ
ꢀ
3050 (C H(NTN)), 2853 (N H(diethylammonium)), 1690 (C=O(NTN)), 1624,
1590 cmꢀ1 (O=C O); elemental analysis calcd (%) for Zn5C140H96O24N6:
ꢀ
Estimation of isosteric heats of H2 adsorption: The isosteric heats (Qst)
of H2 adsorption in SNU-150’ and SNU-151’ were estimated from the H2
sorption data measured at 77 and 87 K. A virial-type expression was used
[Eq. (1)], which is composed of parameters ai and bi, which are indepen-
C 65.34, H 3.76, N 3.27; found: C 65.81, H 3.80, N 3.39.
X-ray crystallography: Crystals of [Zn4OACHTNUGRTEN(NNUG NTN)2]·10DMA·2H2O (SNU-
150) and [Zn5A(NTN)4A(DEF)2][NH2A(C2H5)2]2·8DEF·6H2O (SNU-151)
C
N
G
CHTUNGTRENNUNG
AHCTUNGERTGdNNUN ent of temperature. In Equation (1), P is the pressure (atm), N is the
were coated with Paratone-N oil, and the diffraction data were measured
at 100 K with synchrotron radiation (l=0.80003 ꢂ for SNU-150 and l=
0.69999 ꢂ for SNU-151) on an ADSC Quantum-210 detector at 2D SMC
with a silicon (111) double crystal monochromator (DCM) at the Pohang
Accelerator Laboratory, Korea. The ADSC Q210 ADX program[35] was
amount of adsorbed H2 gas (mggꢀ1), T is the temperature (K), ai and bi
are the virial coefficients, and m and n represent the number of coeffi-
cients required to describe the isotherms adequately. An equation was fit
using the R statistical software package.[40]
Chem. Eur. J. 2013, 19, 17432 – 17438
ꢁ 2013 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chemeurj.org
17437

M. P. Suh et al.
m
n
X
X
1
T
Wiper, J. B. Claridge, M. J. Rosseinsky, Science 2010, 329, 1053–
1057.
[9] T. M. McDonald, D. M. DꢃAlessandro, R. Krishna, J. R. Long,
Chem. Sci. 2011, 2, 2022–2028.
[10] S. R. Caskey, A. G. Wong-Foy, A. J. Matzger, J. Am. Chem. Soc.
2008, 130, 10870–10871.
[11] S. Henke, R. A. Fischer, J. Am. Chem. Soc. 2011, 133, 2064–2067.
[12] M. Higuchi, D. Tanaka, S. Horike, H. Sakamoto, K. Nakamura, Y.
Takashima, Y. Hijikata, N. Yanai, J. Kim, K. Kato, Y. Kubota, M.
Takata, S. Kitagawa, J. Am. Chem. Soc. 2009, 131, 10336–10337.
[13] L.-H. Xie, M. P. Suh, Chem. Eur. J. 2011, 17, 13653–13656.
[14] H. J. Park, Y. E. Cheon, M. P. Suh, Chem. Eur. J. 2010, 16, 11662–
11669.
ln P ¼ ln N þ
aiNi þ
biNi
ð1Þ
i¼0
i¼0
The values of the virial coefficients, a0 through am, were used to calculate
the isosteric heat of adsorption by using Equation (2), in which Qst is the
coverage-dependent isosteric heat of adsorption and R is the universal
gas constant.
m
X
Qst ¼ þR
aiNi
ð2Þ
i¼0
Estimation of isosteric heats of CO2 and CH4 adsorptions: The CO2 ad-
sorption data of SNU-150’ and SNU-151’ at 195, 231, 273, and 298 K
were fit to a polynomial equation by using the Origin 8 program, and the
CH4 adsorption data of SNU-150’ and SNU-151’ at 195, 231, 273, and
298 K were fit to the Langmuir–Freundlich equation, in which N is the
amount of absorbed gas, Nsat is the amount of adsorbed gas at saturation,
P is the pressure, and b and t are constants [Eq. (3)]. The heat of the gas
adsorption was calculated by using the Clausius–Clapeyron equation
[Eq. (4)] at each gas loading point.
[15] X. Kong, E. Scott, W. Ding, J. A. Mason, J. R. Long, J. A. Reimer, J.
Am. Chem. Soc. 2012, 134, 14341–14344.
[16] T. M. McDonald, W. R. Lee, J. A. Mason, B. M. Wiers, C. S. Hong,
J. R. Long, J. Am. Chem. Soc. 2012, 134, 7056–7065.
[17] D. H. Hong, M. P. Suh, Chem. Commun. 2012, 48, 9168–9170.
[18] H. J. Park, M. P. Suh, Chem. Sci. 2013, 4, 685–690.
[19] H. J. Park, M. P. Suh, Chem. Commun. 2012, 48, 3400–3402.
[20] J. J. Ravichandar Babarao, J. Am. Chem. Soc. 2009, 131, 11417–
11425.
[21] H. J. Park, D.-W. Lim, W. S. Yang, T.-R. Oh, M. P. Suh, Chem. Eur.
J. 2011, 17, 7251–7260.
[22] E. Y. Lee, S. Y. Jang, M. P. Suh, J. Am. Chem. Soc. 2005, 127, 6374–
6381.
[23] H. K. Chae, J. Kim, O. D. Friedrichs, M. OꢃKeefe, O. M. Yaghi,
Angew. Chem. 2003, 115, 4037–4039; Angew. Chem. Int. Ed. 2003,
42, 3907–3909.
[24] A. L. Spek, PLATON99, Utrecht University, Utrecht (The Nether-
lands), 1999.
[25] F. Rouquerol, J. Rouquerol, K. Sing, Adsorption by Powders, Porous
Solids Principles, Methodology and Applications, Academic Press,
San Diego, CA, 1999.
[26] G. Horvꢄth, K. Kawazoe, J. Chem. Eng. Jpn. 1983, 16, 470–475.
[27] L. Czepirski, J. Jagiello, Chem. Eng. Sci. 1989, 44, 797–801.
[28] E. Haldoupis, S. Nair, D. S. Sholl, J. Am. Chem. Soc. 2012, 134,
4313–4323.
N
bPt
ð3Þ
ð4Þ
¼
Nsat 1 þ bPt
@ðln PÞ
@ð1=TÞ
Qst
R
¼ ꢀ
Calculation of selectivity of CO2 adsorption: The selectivities of CO2 ad-
sorption over N2 and CH4 were calculated by using the ideal adsorbed so-
lution theory (IAST),[41] which enables the prediction of adsorption equi-
libriums of binary gas mixtures from the related single-component iso-
therms. The CO2, N2, and CH4 adsorption isotherms of SNU-150’ and
SNU-151’ measured at 298 K were fit to the Allometric1 equation (y=
axb) and polynomial equation, respectively, on a logarithmic scale, and
the IAST parameters used in the selectivity calculations were obtained
(see Figures S8, S17, and S18 and Tables S2 and S3 in the Supporting In-
formation).
[29] S. Yang, X. Lin, A. J. Blake, K. M. Thomas, P. Hubberstey, N. R.
Champness, M. Schrçder, Chem. Commun. 2008, 6108–6110.
[30] “Zeolitelike Metal–Organic Frameworks (ZMOFs): Design, Struc-
ture, and Properties”: M. H. Alkordi, M. Eddaoudi, in Supramolec-
ular Chemistry: From Molecules to Nanomaterials, Wiley, 2012.
[31] A. Demessence, D. M. DꢃAlessandro, M. L. Foo, J. R. Long, J. Am.
Chem. Soc. 2009, 131, 8784–8786.
[32] S. D. Burd, S. Ma, J. A. Perman, B. J. Sikora, R. Q. Snurr, P. K. Thal-
lapally, J. Tian, L. Wojtas, M. J. Zaworotko, J. Am. Chem. Soc. 2012,
134, 3663–3666.
Acknowledgements
This work was supported by a National Research Foundation of Korea
(NRF) grant funded by the Korean Government (MSIP) (no. 2012–
055324 and no. 2005–0049412). The authors acknowledge the Pohang Ac-
celerator Laboratory (PAL) for the use of the synchrotron beamline.
[33] N. C. Burtch, H. Jasuja, D. Dubbeldam, K. S. Walton, J. Am. Chem.
Soc. 2013, 135, 7172–7180.
[1] H. Furukawa, N. Ko, Y. B. Go, N. Aratani, S. B. Choi, E. Choi, A. O.
Yazaydin, R. Q. Snurr, M. OꢃKeeffe, J. Kim, O. M. Yaghi, Science
2010, 329, 424–428.
[2] M. P. Suh, H. J. Park, T. K. Prasad, D.-W. Lim, Chem. Rev. 2012,
112, 782–835.
[3] J.-R. Li, R. J. Kuppler, H.-C. Zhou, Chem. Soc. Rev. 2009, 38, 1477–
1504.
[4] 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.
[5] Y. E. Cheon, M. P. Suh, Angew. Chem. 2009, 121, 2943–2947;
Angew. Chem. Int. Ed. 2009, 48, 2899–2903.
[6] Y.-G. Lee, H. R. Moon, Y. E. Cheon, M. P. Suh, Angew. Chem. 2008,
120, 7855–7859; Angew. Chem. Int. Ed. 2008, 47, 7741–7745.
[7] H.-S. Choi, M. P. Suh, Angew. Chem. 2009, 121, 6997–7001; Angew.
Chem. Int. Ed. 2009, 48, 6865–6869.
[34] G. Bordeau, R. Lartia, G. Metge, C. Fiorini-Debuisschert, F. Charra,
M.-P. Teulate-Fichou, J. Am. Chem. Soc. 2008, 130, 16836–16837.
[35] A. J. Arvai, C. Nielsen, ADSC Quantum-210 ADX Program, Area
Detector System Corporation; Poway, CA, USA, 1983.
[36] Z. Otwinowski, W. Minor, in Methods in Enzymology, Vol. 276, Part
A (Eds.: C. W. Carter, Jr., R. M. Sweet, Academic Press, New York,
1997, p. 307.
[37] G. M. Sheldrick, Acta Crystallogr. Sect A 2008, 64, 112–122.
[38] G. M. Sheldrick, SHELEX97, Program for the crystal structure re-
finement, University of Gçttingen: Gçttingen (Germany), 1997.
[39] G. M. Sheldrick, Acta Crystallogr. A 1990, 46, 467–473.
[40] The software package is available online at http://www.r-project.org.
[41] A. L. Myers, J. M. Prausnitz, AIChE J. 1965, 11, 121–127.
[8] J. Rabone, Y. F. Yue, S. Y. Chong, K. C. Stylianou, J. Bacsa, D. Brad-
shaw, G. R. Darling, N. G. Berry, Y. Z. Khimyak, A. Y. Ganin, P.
Received: August 5, 2013
Published online: November 8, 2013
17438
www.chemeurj.org
ꢁ 2013 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Chem. Eur. J. 2013, 19, 17432 – 17438