Exploiting Diffusion Barrier and Chemical Affinity of Metal-Organic Frameworks for Efficient Hydrogen Isotope Separation

Article abstract of DOI:10.1021/jacs.7b07925

Deuterium plays a pivotal role in industrial and scientific research, and is irreplaceable for various applications such as isotope tracing, neutron moderation, and neutron scattering. In addition, deuterium is a key energy source for fusion reactions. Thus, the isolation of deuterium from a physico-chemically almost identical isotopic mixture is a seminal challenge in modern separation technology. However, current commercial approaches suffer from extremely low separation efficiency (i.e., cryogenic distillation, selectivity of 1.5 at 24 K), requiring a cost-effective and large-scale separation technique. Herein, we report a highly effective hydrogen isotope separation system based on metal-organic frameworks (MOFs) having the highest reported separation factor as high as ～26 at 77 K by maximizing synergistic effects of the chemical affinity quantum sieving (CAQS) and kinetic quantum sieving (KQS). For this purpose, the MOF-74 system having high hydrogen adsorption enthalpies due to strong open metal sites is chosen for CAQS functionality, and imidazole molecules (IM) are employed to the system for enhancing the KQS effect. To the best of our knowledge, this work is not only the first attempt to implement two quantum sieving effects, KQS and CAQS, in one system, but also provides experimental validation of the utility of this system for practical industrial usage by isolating high-purity D₂ through direct selective separation studies using 1:1 D₂/H₂ mixtures.

Full text of DOI:10.1021/jacs.7b07925

Subscriber access provided by UNIVERSITY OF ADELAIDE LIBRARIES
Article
Exploiting Diffusion Barrier and Chemical Affinity of Metal-
Organic Frameworks for Efficient Hydrogen Isotope Separation
Jin Yeong Kim, Rafael Balderas-Xicohtencatl, Linda Zhang,
Sung Gu Kang, Michael Hirscher, Hyunchul Oh, and Hoi Ri Moon
J. Am. Chem. Soc., Just Accepted Manuscript • DOI: 10.1021/jacs.7b07925 • Publication Date (Web): 23 Sep 2017
Downloaded from http://pubs.acs.org on September 24, 2017
Just Accepted
“Just Accepted” manuscripts have been peer-reviewed and accepted for publication. They are posted
online prior to technical editing, formatting for publication and author proofing. The American Chemical
Society provides “Just Accepted” as a free service to the research community to expedite the
dissemination of scientific material as soon as possible after acceptance. “Just Accepted” manuscripts
appear in full in PDF format accompanied by an HTML abstract. “Just Accepted” manuscripts have been
fully peer reviewed, but should not be considered the official version of record. They are accessible to all
readers and citable by the Digital Object Identifier (DOI®). “Just Accepted” is an optional service offered
to authors. Therefore, the “Just Accepted” Web site may not include all articles that will be published
in the journal. After a manuscript is technically edited and formatted, it will be removed from the “Just
Accepted” Web site and published as an ASAP article. Note that technical editing may introduce minor
changes to the manuscript text and/or graphics which could affect content, and all legal disclaimers
and ethical guidelines that apply to the journal pertain. ACS cannot be held responsible for errors
or consequences arising from the use of information contained in these “Just Accepted” manuscripts.
Journal of the American Chemical Society is published by the American Chemical
Society. 1155 Sixteenth Street N.W., Washington, DC 20036
Published by American Chemical Society. Copyright © American Chemical Society.
However, no copyright claim is made to original U.S. Government works, or works
produced by employees of any Commonwealth realm Crown government in the course
of their duties.

Page 1 of 13
Journal of the American Chemical Society
1
2
3
4
5
6
7
8
9
Exploiting Diffusion Barrier and Chemical Affinity of Metal-
Organic Frameworks for Efficient Hydrogen Isotope Separa-
tion
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
Jin Yeong Kim,† Rafael BalderasꢀXicohténcatl,‡ Linda Zhang,‡ Sung Gu Kang,┴ Michael Hirscher,*,‡
Hyunchul Oh,*,§ and Hoi Ri Moon*,†
†Department of Chemistry, Ulsan National Institute of Science and Technology (UNIST), Ulsan 44919, Republic of Korea
‡Max Planck Institute for Intelligent Systems, Heisenbergstr. 3, 70569 Stuttgart, Germany
┴School of Chemical Engineering, University of Ulsan, Ulsan 44610, Republic of Korea
§Department of Energy Engineering, Gyeongnam National University of Science and Technology (GNTECH), Jinju 52725,
Republic of Korea
ABSTRACT: Deuterium plays a pivotal role in industrial and scientific research, and is irreplaceable for various applications such
as isotope tracing, neutron moderation, and neutron scattering. In addition, deuterium is a key energy source for fusion reactions.
Thus, the isolation of deuterium from physicoꢀchemically almost identical isotopic mixture is a seminal challenge in modern sepaꢀ
ration technology. However, current commercial approaches suffer from extremely low separation efficiency (i.e. cryogenic distilꢀ
lation, selectivity of 1.5 at 24 K), requiring a costꢀeffective and largeꢀscale separation technique. Herein, we report a highly effecꢀ
tive hydrogen isotope separation system based on metalꢀorganic frameworks (MOFs) having the highest reported separation factor
as high as ~26 at 77 K by maximizing synergistic effects of the chemical affinity quantum sieving (CAQS) and kinetic quantum
sieving (KQS). For this purpose, MOFꢀ74 system having high hydrogen adsorption enthalpies due to strong open metal sites is choꢀ
sen for CAQS functionality and imidazole molecules (IM) are employed to the system for enhancing the KQS effect. To the best of
our knowledge, this work is not only the first attempt to implement two quantum sieving effects, KQS and CAQS, in one system
but also provides experimental validation of the utility of this system for practical industrial usage by isolating highꢀpurity D2
through direct selective separation studies using 1:1 D2/H2 mixtures.
sive, and, furthermore, possess low separation efficiencies (i.e.
cryogenic distillation, with a selectivity of 1.5 at 24 K).8 Reꢀ
cently, Geim et al. reported that the twoꢀdimensional memꢀ
branes of graphene or boron nitride could efficiently separate
hydrogen ion isotopes.9 In this work, hydrogen ions (H+) difꢀ
fuse 10 times faster than deuterium ions (D+) through the twoꢀ
dimensional crystal membrane at room temperature yielding a
purer hydrogen gas. However, from an industrial viewpoint a
product predominantly containing the heavier isotope would
be more favorable. Thus, although it is still challenging, deꢀ
veloping a costꢀ and energyꢀefficient system for directly exꢀ
tracting high purity deuterium is very important to meet indusꢀ
trial and scientific demand.
1. INTRODUCTION
Highly pure deuterium (D) is an irreplaceable raw material in
industrial and scientific research, used for isotope tracing,1ꢀ3
neutron scattering,4ꢀ6 and fuel in future nuclear fusion reacꢀ
tors.7 Thus to ensure a ready supply of this key material, a
costꢀeffective and largeꢀscale separation technique is required.
However, the nearly identical physicochemical properties of
deuterium and hydrogen make it difficult to extract deuterium
from isotopic mixtures using conventional molecular sieving
methods. The current separation technologies for purifying
heavy hydrogen isotopes, such as liquid hydrogen distillation
and the Girdler sulfide process, are extremely energy intenꢀ
ACS Paragon Plus Environment

Journal of the American Chemical Society
As a promising system for D2/H2 separation, nanoporous
Page 2 of 13
1
materials have attracted significant research interest, as they
can directly capture heavier isotope gas molecules via kinetic
quantum sieving (KQS)10ꢀ26 or chemical affinity quantum sievꢀ
ing (CAQS).27ꢀ30 Heavier isotopes, having a shorter de Broglie
wavelength at cryogenic temperatures, experience a lower
diffusion barrier and can diffuse faster than lighter isotopes in
a confined space, resulting in enrichment of the product with
heavier isotope materials using the KQS effect.10 Hirscher et
al. enforced KQS effects by decorating the internal surface of
MOFs and covalent organic frameworks (COFs) with Cl− and
pyridyl groups, respectively, in order to optimize pore aperꢀ
ture, leading to significantly enhanced isotope selectivity
(SD2/H2) of 7.5 at 60 K and 9.7 at 22 K, respectively.11,12 Howꢀ
ever, due to the weak binding energies of hydrogen isotopes
on the internal surface of porous materials, this approach efꢀ
fectively operates at only very low cryogenic temperatures. In
order to overcome this limitation, CAQS exploits the preferenꢀ
tial adsorption of heavier isotopes using strongly interacting
sites introduced into the nanoporous materials.27ꢀ30 As proꢀ
posed theoretically as well as experimentally, the generation
of open metal sites (OMSs) is a top priority for generating
strong substrate interactions with H2 and D2.31ꢀ33 FitzGerald et
al. first explored the CAQS effect of OMSs in D2/H2 separaꢀ
tion showing a selectivity of 5 at 77 K, which demonstrated
that efficient selectivity can be achieved even at high temperaꢀ
tures using strong binding sites.27 Nonetheless, a D2/H2 separaꢀ
tion system comprising only OMSs is still far from meeting
global demand and thus there is a critical need for an effecꢀ
tively designed of D2ꢀselective system.
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
Based on previous findings, the ideal deuterium separation
and storage system should have a high density of OMSs that
can selectively adsorb deuterium even at high temperatures
and a narrow pore size for further increasing the separation
effect through the selective diffusion of D2 into the pores via
KQS. In this sense, to realize very effective isotope separation,
intelligently designed high capacity porous materials posꢀ
sessing both KQS and CAQS effects are required. Herein, we
report a highly effective isotope separation system based on
MOFs having a high uptake (2.8 mmol gꢀ1) with the highest
reported separation factor as high as ~26 at 77 K by maximizꢀ
ing the synergistic effects of CAQS and KQS. To the best of
our knowledge, this is the first study to implement two quanꢀ
tum sieving effects, KQS and CAQS, in one system to sepaꢀ
rate isotopic mixtures. For this purpose, MOFꢀ74ꢀNi was choꢀ
sen as the host material, providing CAQS functionality due to
its high hydrogen adsorption enthalpies (~13 kJ molꢀ1) at
OMSs as reported in the literature.27,34 Simultaneously, imidꢀ
azole molecules (IM) were introduced into the oneꢀ
dimensional (1D) channel of MOFꢀ74ꢀNi as a diffusion barriꢀ
er, effectively reducing the aperture size and repeatedly blockꢀ
ing H2 diffusion, resulting in the KQS effect. Above all, the
system can be implemented using a simple postꢀmodification
strategy by introducing the imidazole molecules into the
MOFꢀ74ꢀNi channel without complex structural design or a
new MOF synthesis.
Figure 1. FTꢀIR and UVꢀVisꢀNIR characterization of MOFꢀ74ꢀ
ac and MOFꢀ74ꢀIMs. (a) FTꢀIR spectra, and (b) diffuse reflecꢀ
tance UVꢀVisꢀNIR spectra of MOFꢀ74ꢀac (black), MOFꢀ74ꢀIMꢀ
10 (red), MOFꢀ74ꢀIMꢀ38 (blue), and MOFꢀ74ꢀIMꢀ70 (green).
imidazole in dichloromethane (DCM) with three different
concentrations in order to prepare different amounts of imidꢀ
azole containing MOFꢀ74ꢀNi. As calculated based on the eleꢀ
mental analyses results, the resultant MOF samples contained
10, 38, and 70% of all Ni2+ OMSs in MOFꢀ74ꢀNi occupied by
imidazole molecules, respectively, denoted MOFꢀ74ꢀIMꢀ10,
MOFꢀ74ꢀIMꢀ38, and MOFꢀ74ꢀIMꢀ70 (Table S1). The Xꢀray
powder diffraction (XRPD) patterns of MOFꢀ74ꢀac, MOFꢀ74ꢀ
IMꢀ10, MOFꢀ74ꢀIMꢀ38, and MOFꢀ74ꢀIMꢀ70 indicate that the
framework structures of samples were maintained after postꢀ
synthetic treatments (Figure S1). In order to examine the inꢀ
corporation of imidazole molecules in MOFꢀ74ꢀNi, Fourier
transformꢀinfrared (FTꢀIR) spectra were obtained. As shown
in Figure 1a, compare with MOFꢀ74ꢀac, MOFꢀ74ꢀIMꢀ10
shows weak additional peaks at 3434, 3150, 1058, and 729 cmꢀ
1, which correspond to ν(NꢀH), ν(CꢀH), δ(CꢀH), and δ(CꢀH) of
the imidazole molecule, respectively.35,36 The same peaks were
observed in MOFꢀ74ꢀIMꢀ38 and MOFꢀ74ꢀIMꢀ70 with higher
intensities due to the increase in the amount of imidazole. The
appearance of the NꢀH stretching vibration around 3430 cmꢀ1
indicates that the incorporated imidazole molecules are neutral
and not in the deprotonated form as imidazolate. The coordiꢀ
nation of incorporated imidazole molecules to the Ni2+ OMSs
was verified by ultraviolet visible nearꢀinfra red
2. RESULTS AND DISCUSSION
Synthesis and Characterization. In order to develop a sysꢀ
tem containing OMSs as well as diffusion barriers, first, asꢀ
prepared MOFꢀ74ꢀNi was completely activated at 250 °C in
vacuum for 5 h (designated as MOFꢀ74ꢀac). MOFꢀ74ꢀac
which has fully exposed OMSs was soaked in the solutions of
ACS Paragon Plus Environment

Page 3 of 13
Journal of the American Chemical Society
1
2
3
4
5
6
7
8
(a)
(b)
350
300
250
400
300
200
100
0
400
MOF-74-ac
MOF-74-IM-10
300
200
MOF-74-IM-38
100
50
0
9
100
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
MOF-74-IM-70
0
0.0 0.2 0.4 0.6 0.8 1.0
P/P0
10-5 10-4 10-3 10-2 10-1 100 10-2
10-1
P/P0 (log scale)
(c)
(d)
12
10
8
12
10
8
5
4
MOF-74-ac
MOF-74-IM-10
MOF-74-IM-38
MOF-74-IM-70
3
2
6
6
4
4
1
0
2
2
0
0
0.0 0.2 0.4 0.6 0.8 1.0
Pressure (bar)
10-5 10-4 10-3 10-2 10-1 100 10-3
10-2
bar (log scale)
Figure 2. Gas sorption isotherms on the MOFꢀ74ꢀac and MOFꢀ74ꢀIMs. N2 sorption isotherms represented in (a) linear and (b) log scale,
and H2 sorption isotherms represented in (c) linear and (d) log scale at 77 K of MOFꢀ74ꢀac (black), MOFꢀ74ꢀIMꢀ10 (red), MOFꢀ74ꢀIMꢀ38
(blue), and MOFꢀ74ꢀIMꢀ70 (green). Filled and open symbols correspond to adsorption and desorption, respectively. The grey colored
graphs to the right are the enlargement of the lowꢀpressure region in the sorption isotherms in log scale.
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
presence of free imidazole molecules in pore cannot be comꢀ
pletely ruled out. In order to exclude the possibility of H2O
molecules coordination instead of imidazole molecules in an
(UVꢀVisꢀNIR) spectroscopy via diffuse reflectance spectroꢀ
ambient atmosphere, the UVꢀVisꢀNIR spectra for intentionally
scopic measurements conducted with solid powders (Figure
airꢀexposed samples were examined. For MOFꢀ74ꢀIM series
1b). In MOFꢀ74ꢀac, fiveꢀcoordinated Ni2+−OMSs species in a
as well as MOFꢀ74ꢀac, a new peak at 5556 cmꢀ1 was clearly
squareꢀpyramidal geometry show absorption bands at 14144,
shown, which is a characteristic H2O overtone peak originatꢀ
10040 and 5555 cmꢀ1, which correspond to the electronic tranꢀ
ing from the coordination of H2O molecules upon exposure to
3
3
3
3
3
sitions, B1(3F) → A2(3F), B1(3F) → B2(3F), and B1(3F) →
3A2(3P), respectively.37 Upon inclusion of imidazole molecules
into MOFꢀ74ꢀac, new absorption peaks appear at 13141 and
9930 cmꢀ1, which clearly show the dꢀd transitions of an octaꢀ
air (Figure S2).38 The decreasing intensity of the H2O overtone
peak with increasing amount of imidazole molecules in MOFꢀ
74ꢀNi indicates that the
hedral sixꢀcoordinated Ni2+ species, A2g(3F) → T1g(3F) and
3
3
3A2g(3F) → T2g(3F), respectively.37 As the amount of coordiꢀ
3
imidazole molecules occupy Ni2+ꢀOMSs. Overall, via simple
postꢀsynthetic modification we successfully prepared a series
of MOFꢀ74ꢀIM materials possessing 1D channels decorated
with different amounts of imidazole molecules as well as
OMSs.
nating imidazole molecules increases, absorption bands related
to fiveꢀcoordinate Ni2+ decrease, while bands corresponding to
sixꢀcoordinate Ni2+ species increase. Therefore, a major part of
included imidazole molecules were coordinated to Ni2+ centers
as confirmed by UVꢀvisꢀNIR and FTꢀIR spectra, although
ACS Paragon Plus Environment

Journal of the American Chemical Society
Examination of Aperture Modification. In order to underꢀ
Page 4 of 13
1
stand the effect of imidazole addition on the pore structure of
MOFꢀ74ꢀNi, we conducted N2 adsorption/desorption experiꢀ
ments at 77 K for MOFꢀ74ꢀac and imidazole coordinated samꢀ
ples. BrunauerꢀEmmettꢀTeller (BET) surface area is increasꢀ
ingly reduced from 1406 m2 gꢀ1 for MOFꢀ74ꢀac to 1094 and
425 m2 gꢀ1 for MOFꢀ74ꢀIMꢀ10 and MOFꢀ74ꢀIMꢀ38, respecꢀ
tively, as the number of coordinated imidazole molecules inꢀ
creased. Despite this, these samples were still microporous
(Figure 2a and Table S2). After loading 70% of Ni2+ OMSs
with imidazole molecules, the surface area decreased to 51 m2
gꢀ1 showing a nonporous nature, which indicated that the pores
of MOFꢀ74ꢀIMꢀ70 were mostly blocked by imidazole moleꢀ
cules and were thus inaccessible to N2 molecules. H2 sorption
experiments at 77 K with N2 sorption studies imply that our
synthetic strategy can be potentially exploited as a D2/H2 sepaꢀ
ration system. For N2 and H2 gases, MOFꢀ74ꢀac does not limit
the diffusion due to its large channel diameter (~1.1 nm) and
thus exhibits fully reversible isotherms. On the other hand,
MOFꢀ74ꢀIMꢀ10 and MOFꢀ74ꢀIMꢀ38 show large hysteresis
loops due to the increasing amount of coordinated imidazole,
implying slow diffusion kinetics (Figure 2). Moreover, despite
a 22% difference in the N2 BET surface area between MOFꢀ
74ꢀac and MOFꢀ74ꢀIMꢀ10, their similar H2 uptake, 10.5 and
10.0 mmol gꢀ1 respectively, indicates that aperture modiꢀ
fication by imidazole molecule coordination generates selecꢀ
tive accessibility for H2 over N2 due to the existence of numerꢀ
ous apertures smaller than the N2 probe. For H2 and D2 gas
sorption isotherms obtained at various exposure temperatures
(25, 40, 60, and 77 K) (Figures S3ꢀ6), MOFꢀ74ꢀIMꢀ10 and
MOFꢀ74ꢀIMꢀ38 show hysteresis loops (Figures S4 and S5),
revealing the existence of a diffusion barrier indicating that the
KQS effect exists over the entire measured temperature range.
Due to the nonporous nature of MOFꢀ74ꢀIMꢀ70 (Figure S7), it
was excluded for further measurements. Moreover, higher
uptake of D2, which possesses a smaller de Broglie waveꢀ
length and lower zero point energy in confinement compared
to H2, may be ascribed to its favorable diffusion in this pore
system. Therefore, the present aperture modification strategy
may effectively form a kinetic diffusion barrier in MOFꢀ74,
indicating that the MOFꢀ74ꢀIM system may be a potential
advanced material for applications in D2/H2 separation.
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
Figure 3. Thermal desorption spectra of MOFꢀ74ꢀIMꢀ10. (a)
Pure gas H2 (black, filled circle) and D2 (red, open circle) and
(b) 1:1 D2/H2 mixture gas loading on MOFꢀ74ꢀIMꢀ10 at an exꢀ
posure temperature, Texp, of 25 K. The site III clearly shows
preferential adsorption of D2 over H2 from the isotope mixture.
peak corresponding to desorption from site I (benzene ring)
appeared at 32 K for H2 and 34 K for D2 while desorption
from site II (triangle of oxygen atoms) occurred at 41 K and
45 K for H2 and D2, respectively. Significantly, the desorption
maximum of D2 from site III (OMS) occurred at 85 K, which
is much higher than that of hydrogen by ~7 K, indicating a
huge difference between H2 and D2 adsorption enthalpy at
OMSs. We attribute this to the presence of a strong CAQS
effect. Considering the desorption temperature maxima, we
classify sites I and II as weak binding site (WBS) and site III
as a strong binding site (SBS).
TDS studies of adsorption sites and preferential isotope
binding on the sites. In order to verify the adsorption sites on
the MOFꢀ74ꢀNi basedꢀsamples and their adsorption affinity to
H2 and D2, cryogenic thermal desorption spectroscopy
(TDS)11,28 measurements were performed for pure H2 and D2
gas. MOFꢀ74ꢀIMꢀ10 was exposed to a pure D2 or H2 atmosꢀ
phere (10 mbar) at room temperature. Afterwards, the sample
was rapidly cooled to the target exposure temperature, Texp, of
25 K. After 3 min exposure, the sample chamber was evacuatꢀ
ed, and cooled quickly down to below 20 K. A heating ramp
of 0.1 K sꢀ1 was applied and the TDS spectrum was obtained
between 20 ꢀ 120 K. As shown in Figure 3a, the spectra for
both H2 and D2 show three desorption peak maxima. This pheꢀ
nomenon has been similarly observed in previous studies usꢀ
ing MOFꢀ74ꢀ Co.28 Along with wellꢀknown neutron powder
diffraction results,39 those three peaks were assigned as three
energetically different adsorption sites, namely, adsorption at a
benzene ring (site I), at a triangle of oxygen atoms (site II),
and at an OMS (site III). For both H2 and D2, the three desorpꢀ
tion peak maxima appeared at different temperature shifts. The
The preference of SBSs for D2 over H2 is clearly visible in
the TDS spectra obtained for the isotopic mixture of H2 and
D2. As shown in Figure 3b, the SBSs dominantly adsorb D2
over H2 at Texp = 25 K. Despite increasing Texp from 25 to 77 K
(Figure S8), D2 adsorption still dominantly occurs at SBSs,
leading to potentially efficient isotope separation even at high
temperatures. Further examination of the performance of
MOFꢀ74ꢀac and MOFꢀ74ꢀIMꢀ38 using a 1:1 isotope mixture
also clearly demonstrates the existence of SBSs and the prefꢀ
erential adsorption of D2 over H2 on these sites (Figures S7 and
S9), indicating a strong CAQS effect. However, the number of
SBSs is dramatically reduced with increasing imidazole conꢀ
tent, which in turn reduces the CAQS effect. Thus, optimizaꢀ
tion of the amount of imidazole in the system is crucial to efꢀ
fectively merge both the KQS and CAQS effects.
ACS Paragon Plus Environment

Page 5 of 13
Journal of the American Chemical Society
Separation performance of MOF-74-IM systems. Inspired MOFꢀ74ꢀIMꢀ38 show large differences while similar selectiviꢀ
1
by the coꢀexistence of both quantumꢀsieving effects in MOFꢀ
74ꢀIM systems, we further investigated the direct D2/H2 selecꢀ
tivity of MOFꢀ74ꢀIM systems and MOFꢀ74ꢀac with TDS exꢀ
periments. All samples were exposed to an equimolar D2/H2
isotope mixture at various exposure temperatures (Texp = 25,
40, 60, and 77 K). The overall selectivity (SD2/H2) and the corꢀ
responding amount of adsorbed D2 of MOFꢀ74ꢀac, MOFꢀ74ꢀ
IMꢀ10 and MOFꢀ74ꢀIMꢀ38 are presented in Figure 4. One
should note that since the area under the desorption curves in
TDS are proportional to the amount of desorbed gas, the ratio
of these areas directly reflects the selectivity of D2 over H2. In
addition, since all samples are exposed to the gas mixture first
and then evacuated at the same target exposure temperature
before the heating ramp is applied, any gas phase originally
present in the chamber can be excluded and thus the obtained
values suggest the actual gas amount originates only from
adsorbed species and thus is a direct indication of the selectiviꢀ
ty of WBSs and/or SBSs.
ty is obtained below 60 K. With increasing temperature, the
selectivity of MOFꢀ74ꢀIMꢀ10 is greatly enhanced compared to
that of the MOFꢀ74ꢀac, resulting in SD2/H2 = 26 at a Texp of 77
K, whereas the selectivity MOFꢀ74ꢀIMꢀ38 is significantly
reduced to SD2/H2 = 5 at a Texp of 77 K. The greatly enhanced
selectivity of MOFꢀ74ꢀIMꢀ10 can be ascribed to the synergisꢀ
tic effect of KQS and CAQS, as was our aim. Introducing a
diffusion barrier by incorporating small amounts (10%) of
imidazole magnifies the difference in diffusion rates between
H2 and D2 leading to improved accessibility of D2 to the sorpꢀ
tion site (KQS effect). Moreover, a relatively high density of
SBSs in the MOFꢀ74ꢀIMꢀ10 channels result in additional seꢀ
lectivity for D2 adsorption due to the CAQS effect. One should
note that this diffusion limitation, leading to the KQS effect, is
more clearly observed with an increase in imidazole content.
Reducing the exposure time from 120 min to 3 min for MOFꢀ
74ꢀIMꢀ38, enhances the selectivity both at 60 K and 77 K
(Figure S10), implying that the kinetic effect still exists even
at high temperature.
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
For MOFꢀ74ꢀac, selectivity increases linearly from SD2/H2
=
7 at a Texp of 25 K to SD2/H2 = 19 at a Texp of 77 K. At higher
Texp (above 60 K), hydrogen isotopes can be only adsorbed on
SBSs (see Figure S7), and these SBSs are preferentially occuꢀ
pied by deuterium, resulting in high selectivity. It is worth to
note that the selectivity (SD2/H2 = 19 at 77 K) is different from
the reported result (SD2/H2 = 5 at 77 K) by FitzGerald et al.,27
which suggests that the separation results based on ideal adꢀ
sorption solution theory (IAST) obtained from pure gas adꢀ
sorptions may lead to erroneous conclusions and therefore an
appropriate performance evaluation using mixed gas is necesꢀ
sary. On the other hand, at lower Texp (below 60 K), hydrogen
isotopes can be adsorbed on both WBSs and SBSs, thus weakꢀ
ening the CAQS effect, resulting in a lower selectivity. MOFꢀ
74ꢀIMꢀ10 and MOFꢀ74ꢀIMꢀ38 show two distinct regions beꢀ
low and above Texp = 60 K. Above 60 K, where gas adsorption
only occurs at SBSs, the selectivity of MOFꢀ74ꢀIMꢀ10 and
The computational activation energy of H2/D2 in MOFꢀ74ꢀ
IM was found by density functional theory (DFT) calculations.
(Figures S11ꢀ14). The relaxed MOFꢀ74ꢀIM structure was calꢀ
culated (Figure S11), and then the nudged elastic band (NEB)
methods were utilized to explore the transition states of H2/D2
migration. Since there is no difference in the transition states
of H2 and D2 using the classical method, the zeroꢀpoint energy
and formation energy of both H2 and D2 on the metal center
had to be considered (Figures S13 and S14). Due to the system
limitation for designing MOFꢀ74ꢀIM periodic systems for gas
diffusion studies, our model was simplified to characterize a
gas molecule hopping circularly from one Ni2+ OMS to an
adjacent site, rather than molecular transport into the MOF
(Figure S12). It should be noted that D2 selectivity calculation
in the previous report typically considered coupling between
the translation and orientation,40 whereas our calculation deꢀ
scribed in the present study exhibits only activation energy
difference between isotopes. Nevertheless, this model still can
indirectly provide insight into the effect of coordinated imidꢀ
azole on gas migration. The calculated results exhibit a differꢀ
ence in the activation energy of 1.43 kJ molꢀ1, implying that
higher activation energy was observed in H2 than in D2 (Figure
S13 and S14). This difference is even higher than previous
theoretical calculations performed with pristine MOFꢀ74ꢀNi in
which the difference of isosteric heats of adsorption was found
to be 0.92 kJ/mol,27 consistent with our experimental observaꢀ
tions. Thus, the diffusion of heavier isotopes in channels of
MOFꢀ74ꢀIM is more energetically favorable than the diffusion
of lighter isotopes. Although imidazole is useful to increase
the activation energy difference between H2 and D2, increasing
the amount of imidazole diminishes the CAQS effect due to
the reduction of OMSs leading to a low selectivity and low D2
separation capacity (see MOFꢀ74ꢀIMꢀ38 in Figure 4). This
indicates that the optimum imidazole amount should be conꢀ
sidered for both selectivity and the amount of D2 adsorbed. To
the best of our knowledge, the experimental SD2/H2 value of 26
for MOFꢀ74ꢀIMꢀ10 (2.8 mmol/g of D2 uptake) obtained at
liquid nitrogen temperature is one of the highest reported
SD2/H2 experimental values among various porous adsorbents
(Figure S15 and Table S3).
(a)
30
MOF-74-ac
MOF-74-IM-10
25
MOF-74-IM-38
20
15
10
5
20
0
(b)
Total uptake
SBS
MㅌOF-74-ac
MOF-74-IM-10
MOF-74-IM-38
WBS
15
10
5
0
20
30
40
50
60
70
80
Texp (K)
3. CONCLUSIONS
Figure 4. Adsorptive separation for an equimolar D2/H2 isotope
mixture. (a) Selectivity for D2 compared to H2 and (b) D2 uptake
at strong binding sites (dashed line) and weak binding sites (dotꢀ
ted line) and total D2 uptake (solid line) of MOFꢀ74ꢀac (black),
MOFꢀ74ꢀIMꢀ10 (red), and MOFꢀ74ꢀIMꢀ38 (blue) as a function of
ACS Paragon Plus Environment
Texp. At higher Texp (above 60 K), D2 adsorption only occurs at
SBSs.

Journal of the American Chemical Society
Page 6 of 13
[Ni2(dobdc)(IM)0.2(H2O)1.3] (Ni2C8.6H5.4N0.4O7.3): C, 29.60; H,
1.56; N, 1.61%; Found: C, 30.00; H, 1.52; N, 1.59%. FTꢀIR
(ATR, cmꢀ1): νNꢀH(imidazole)
3434, νCꢀH(imidazole) 3150,
νO−C=O(carboxylate, asym) = 1552(s), νO−C=O(carboxylate, sym) = 1415(s), νCꢀO
= 1235(s), δCꢀH(imidazole) = 1058, δCꢀH(imidazole) = 729. Anal. calcd. for
In summary, we demonstrated that deuterium could be effiꢀ
ciently separated and stored inside the pores of the MOFꢀ74ꢀ
IM system by implementing two quantumꢀsieving effects in
one system. The simple post modification strategy of introducꢀ
ing imidazole molecules into the highly dense OMS channels
of MOFꢀ74 eliminates the need for complex design and synꢀ
thesis of new MOF structures to optimize the aperture size and
increase the internal binding energy, simultaneously, and is
also applicable to other gas mixture separation such as tritium,
deuterium and hydrogen mixtures or He isotopic mixtures.
Imidazole molecules are able to repeatedly block the diffusion
of lighter isotopes into the 1D channels of MOFꢀ74, acting as
a diffusion barrier, while the strong binding sites of the MOF
can selectively adsorb heavier isotopes. By optimizing the
amount of coordinated imidazole molecules via a post synthetꢀ
ic strategy, the synergistic effect of KQS arising from the kiꢀ
netic diffusion barrier and CAQS arising from the presence of
open metal sites yields an impressively high selectivity of ~26
even at high temperatures (77 K) with high D2 uptake (2.84
mmol gꢀ1). We believe that this strategy of combining KQS
and CAQS in one system will provide new opportunities for
the intelligent design of porous materials leading to the develꢀ
opment of other highly efficient isotope and gas separation
systems.
1
2
3
4
5
6
7
8
=
=
MOFꢀ74ꢀIMꢀ38,
(Ni2C10.28H6.24N1.52O6.6): C, 33.01; H, 1.68; N, 5.69%; Found: C,
33.52; H, 1.73; N, 5.65%. FTꢀIR (ATR, cmꢀ1): νNꢀH(imidazole)
3430, νCꢀH(imidazole) = 3146, νO−C=O(carboxylate, = 1552(s),
[Ni2(dobdc)(IM)0.76(H2O)0.60]
=
asym)
νO−C=O(carboxylate, sym) = 1411(s), νCꢀO = 1239(s), δCꢀH(imidazole) = 1058,
δCꢀH(imidazole) 741. Anal. calcd. for MOFꢀ74ꢀIMꢀ70,
9
=
[Ni2(dobdc)(IM)1.4(H2O)0.5] (Ni2C12.2H8.6N2.8O6.5): C, 35.24; H,
2.08; N, 9.43%; Found: C, 35.00; H, 2.17; N, 9.42%. FTꢀIR
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
(ATR, cmꢀ1): νNꢀH(imidazole)
= 3426, νCꢀH(imidazole) = 3140,
νO−C=O(carboxylate, asym) = 1556(s), νO−C=O(carboxylate, sym) = 1411(s), νCꢀO
= 1238(s), δCꢀH(imidazole) = 1061, δCꢀH(imidazole) = 739. Since the samꢀ
ple preparation of elemental analyses was conducted in air, adꢀ
sorbed water molecules were detected despite careful handling.
TDS studies of H2 and D2. A laboratoryꢀdesigned experiꢀ
mental setꢀup for TDS measurements was previously described in
detail28. For each measurement, 3–4 mg of sample was used. A
highꢀvacuum (HV) chamber contains the sample holder that is
screwed tightly to a Cu block, which is surrounded by a resistive
heater. This Cu block is connected to a cold finger comprising a
flowing helium cryostat, which allows for cooling below 20 K.
All samples were activated at 400 K for 5 h. For pure gas thermal
desorption experiments, the sample was exposed to either a pure
D2 or H2 atmosphere (10 mbar) at room temperature. Afterwards,
the sample was rapidly cooled to the 25 K. After 3 min exposure,
the sample chamber was evacuated, and cooled quickly down to
below 20 K. Finally, a linear heating ramp (0.1 K sꢀ1) was applied.
In case of mixed D2/H2 thermal desorption experiments, the samꢀ
ple was exposed to a 10 mbar 1:1 D2/H2 mixture at a given expoꢀ
sure temperature between 25 K and 77 K for 3 min. After 3 min,
the gas molecules that had not been adsorbed were mildly pumped
off until HV conditions were reached again. Afterwards, the samꢀ
ple chamber was cooled quickly down to below 20 K. For desorpꢀ
tion experiments, the sample was heated at a rate of 0.1 K sꢀ1 from
20 K to 150 K and the desorbing gas was continuously detected
by a quadrupole mass spectrometer. In order to quantify the mass
spectrometer signal, the TDS device was calibrated by a hydroꢀ
gen/deuterium loaded Pd95Ce5 alloy.
4. MATERIALS AND METHODS
All chemicals and solvents were of reagent grade and were used
as received without further purification. Fourierꢀtransform (FT)ꢀ
NMR spectra were recorded on a Varian 600 MHz spectrometer.
Infrared (IR) spectra were recorded with a ThermoFisher Scienꢀ
tific Nicolet 6700 FTꢀIR spectrometer. Elemental analyses were
performed at the UNIST Central Research Facilities Center
(UCRF) in the Ulsan National Institute of Science and Technoloꢀ
gy (UNIST). UVꢀVisꢀNIR spectra were recorded with a Cary
5000 UV/vis spectrophotometer. XRPD patterns were collected
on a Bruker D8 advance diffractometer at 40 kV and 40 mA for
Cu Kꢀ (ꢁ = 1.54050 Å), with a step size of 0.02° in 2ꢂ. The niꢀ
trogen adsorptionꢀdesorption isotherms were obtained using a
BELSORPꢀmax at 77 K. Prior to the adsorption measurements, all
samples (~100 mg) were evacuated (p < 10ꢀ5 mbar) at 393 K for 5
h. The specific surface area was determined from the linear part of
the BrunauerꢀEmmettꢀTeller (BET) equation, and the total pore
volume was calculated from the amount adsorbed at a relative
pressure of 0.99. The micropore volume was computed using the
HorvathꢀKawazoe (HK) method. The hydrogen and deuterium
adsorption isotherms of all samples were measured at 20 K to 77
K using a fully automated Quantachrome AutosorbꢀiQ with a
controlled cryoꢀcooler. All samples (~70 mg) were activated unꢀ
der ultraꢀhigh vacuum at 400 K for 5 h prior to each measurement.
Theoretical calculation. All theoretical calculations based on
Density Functional Theory (DFT) were done with the Vienna ab
initio Simulation Package (VASP).42,43 The PBEꢀGGA functional
was used.44 The Grimme D3 (DFTꢀD3) correction was included in
our calculations for dispersion correction.45 An energy cutoff and
the convergence criterion of the energy were set to 600 eV and
2×10ꢀ4 eV, respectively. Each atom was completely relaxed until
forces were below 0.03 eV Å−1. The spin polarization was includꢀ
ed in our calculations. A 1×1×4 MonkhorstꢀPack kꢀpoint grid was
chosen.46 The Hubbard U correction with Ueff = 6.4 eV for Ni was
applied using DFT+U method by Dudarev.47,48 The climbing imꢀ
age nudged elastic band (CIꢀNEB) methods were employed to
calculate the classical migration energies of deuterium and hydroꢀ
gen molecules.49 The corrections of zero point energies (ZPEs) to
the classical migration energies of H2 and D2 were applied in our
calculations.
Preparation
of
MOF-74-IM.
2,5ꢀDioxidoꢀ1,4ꢀ
benzenedicarboxylic acid (H4dobdc) and MOFꢀ74ꢀNi were preꢀ
pared using a reported method41 with minor modifications as deꢀ
scribed in the supporting information. Asꢀsynthesized MOFꢀ74ꢀNi
was activated by heating at 250 °C in vacuum for 5 h, and is hereꢀ
after referred to as MOFꢀ74ꢀac. In an Arꢀfilled glove box, the
yellow powder of MOFꢀ74ꢀac (0.100 g, 0.640 mmol based on
Ni2+) was suspended in anhydrous DCM (8.00 mL) under vigorꢀ
ous stirring. To this solution 2.00 mL of imidazole solution in
DCM was added, which corresponds to a concentration of 3.20 x
10ꢀ2, 1.28 x 10ꢀ1, and 2.24 x 10ꢀ1 M for MOFꢀ74ꢀIMꢀ10, MOFꢀ74ꢀ
IMꢀ38, and MOFꢀ74ꢀIMꢀ70, respectively. After stirring for 15 h at
room temperature, the greenish resultant solid was isolated by
centrifugation and dried in an Arꢀfilled glove box. For further
measurements and experiments, the product was completely dried
at 120 °C under vacuum for 5 h. Anal. calcd. for MOFꢀ74ꢀIMꢀ10,
ASSOCIATED CONTENT
Supporting Information.
ACS Paragon Plus Environment

Page 7 of 13
Journal of the American Chemical Society
(19) Noguchi, D.; Tanaka, H.; Fujimori, Tl.; Kagita, H.; Hattori,
Experimental details, Table S1ꢀ3, Figures S1ꢀ15, DFT calculation
Y.; Honda, H.; Urita, K.; Utsumi, S.; Wang, Zꢀm,; Ohba, T.; Kanoh,
H.; Hata, K.; Kaneko, K. J. Phys.: Condens. Matter 2010, 22, 334207.
(20) Kumar, A. V. A.; Bhatia, S. K. Phys. Rev. Lett. 2005, 95,
245901.
(21) Kumar, A. V. A.; Jobic, H.; Bhatia, S. K. Adsorption 2007, 13,
501.
(22) Niimura, S.; Fujimori, T.; Minami, D.; Hattori, Y.; Abrams,
L.; Corbin, D.; Hata, K.; Kaneko, K. J. Am. Chem. Soc. 2012, 134,
18483.
(23) Chen, B.; Zhao, X.; Putkham, A.; Hong, K.; Lobkovsky, E.
B.; Hurtado, E. J.; Fletcher, A. J.; Thomas, K. M. J. Am. Chem.
Soc. 2008, 130, 6411.
(24) Noguchi, D.; Tanaka, H.; Kondo, A.; Kajiro, H.; Noguchi, H.;
Ohba, T.; Kanoh, H.; Kaneko, K. J. Am. Chem. Soc. 2008, 130, 6367.
(25) Liu, D.; Wang, W.; Mi, J.; Zhong, C.; Yang, Q.; Wu, D. Ind.
Eng. Chem. Res. 2012, 51, 434.
(26) Oh, H.; Park, K. S.; Kalidindi, S. B.; Fischer, R. A.; Hirscher,
M. J. Mater. Chem. A 2013, 1, 3244.
(27) FitzGerald, S. A.; Pierce, C. J.; Rowsell, J. L.; Bloch, E. D.;
Mason, J. A. J. Am. Chem. Soc. 2013, 135, 9458.
(28) Oh, H.; Savchenko, I.; Mavrandonakis, A.; Heine, T.; Hirꢀ
scher, M. ACS nano 2014, 8, 761.
(29) Savchenko, I.; Mavrandoaski, A.; Heine, T.; Oh, H.; Teufel,
J.; Hirscher, M. Micropor. Mesopor. Mater. 2015, 216, 133.
(30) Weinrauch, I.; Savchenko, I.; Denysenko, D.; Souliou, S. M.;
Kim, H. ꢀH.; Tacon, M. L.; Daemen, L. L.; Cheng, Y.; Mavrandonaꢀ
kis, A.; RamirezꢀCuesta, A. J.; Volkmer, D.; Schütz, G.; Hirscher, M.;
Heine, T. Nat. Commun. 2017, 8, 14496.
(31) Suh, M. P.; Park, H. J.; Prasad, T. K.; Lim, D. W. Chem.
Rev. 2011, 112, 782.
(32) Cheon, Y. E.; Suh, M. P. Chem. Commun. 2009, 2296.
(33) Yang, Q.; Zhong, C. J. Phys. Chem. B 2006, 110, 655.
(34) Zhou, W.; Wu, H.; Yildirim, T. J. Am. Chem. Soc. 2008, 130,
15268.
(35) González, M.; LemusꢀSantana, A. A.; RodríguezꢀHernández,
J.; Knobel, M.; Reguera, E. J. Solid State Chem. 2013, 197, 317.
(36) Ramasamy, R. Armen. J. Phys. 2015, 8, 51.
(37) Chavan, S.; Bonino, F.; Valenzano, L.; Civalleri, B.; Lamberꢀ
ti, C.; Acerbi, N.; Cavka, J. H.; Leistner, M.; Bordiga, S. J. Phys.
Chem. C 2013, 117, 15615.
(38) Gallo, E.; Lamberti, C.; Glatzel, P. Inorg. Chem. 2013, 52,
5633.
(39) Liu, Y.; Kabbour, H.; Brown, C. M.; Neumann, D. A.; Ahn,
C. C. Langmuir 2008, 24, 4772.
(40) Qu, Y.; Li, F.; Zhou, H.; Zhao, M. Sci. Rep. 2016, 6, 19952.
(41) Cadot, S.; Veyre, L.; Luneau, D.; Farrusseng, D.; Quadrelli, E.
A. J. Mater. Chem. A 2014, 2, 17757.
(42) Kresse, G.; Furthmüller, J. Phys. Rev. B 1996, 54, 11169.
(43) Kresse, G.; Furthmüller, J. Comp. Mater. Sci. 1996, 6, 15.
(44) Perdew, J. P.; Burke, K.; Ernzerhof, M. Phys. Rev. Lett. 1996,
77, 3865.
1
2
3
4
5
6
7
8
details, and related references. This material is available free of
charge via the Internet at http://pubs.acs.org.
AUTHOR INFORMATION
Corresponding Author
* hirscher@is.mpg.de
* oh@gntech.ac.kr
* hoirimoon@unist.ac.kr
9
Notes
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
The authors declare no competing financial interest.
ACKNOWLEDGMENT
This work was supported by the National Research Foundation of
Korea (NRF) Grant funded by the Korean Government
(MSIP)(No. 2016R1A5A1009405, 2016M1A7A1A02005285,
and 2017R1C1B5015469). J.Y.K. acknowledges the Global PhD
Fellowship (NRFꢀ2014H1A2A1020670).
REFERENCES
(1) Stiopkin, I. V.; Weeraman, C.; Pieniazek, P. A.; Shalhout. F.
Y.; Skinner, J. L.; Benderskii, A. V. Nature 2011, 474, 192.
(2) Povinec, P. P.; Bokuniewicz, H.; Burnett, W. C.; Cable, J.;
Charette, M.; Comanducci, J. –F.; Kontar, E. A.; Moore, W. S.; Oberꢀ
dorfer, J. A.; Oliveira, J. de.; Peterson, R.; Stieglitz, T.; Taniguchi,
M. J. Environ. Radioactiv. 2008, 99, 1596.
(3) Keppler, F.; Hamilton, J. T. G.; McRoberts, W. C.; Vigano, I.;
Braß, M.; Röckmann, T. New Phytol. 2008, 178, 808.
(4) Zaccai, G. Science 2002, 288, 1604.
(5) Büldt, G.; Gally, H. U.; Seelig, A.; Seelig, J.; Zaccai, G.; Naꢀ
ture 1978, 271, 182.
(6) Machida, A.; Saitoh, H.; Sugimoto, H.; Hattori, T.; Sanoꢀ
Furukawa, A.; Endo, N.; Katayama, Y.; Lizuka, R.; Sato, T.; Matsuo,
M.; Orimo, Sꢀi,; Aoki, K. Nat. Commun. 2014, 5, 5063.
(7) Miller, A. I. Can. Nuclear Soc. Bull. 2001, 22, 1.
(8) Rae, H. K. Selecting Heavy Water Processes. In Separation of
Hydrogen Isotopes; American Chemical Society: Washington, DC,
1978; Vol. 68, pp 1ꢀ26.
(9) LozadaꢀHidalgo, M.; Hu, S.; Marshall, O.; Mishchenko, A.;
Grigorenko, A. N.; Dryfe, R. A. W.; Radha, B.; Grigorieva, I. V.;
Geim, A. K. Science 2016, 351, 68.
(10) Beenakker, J. J. M.; Borman, V. D.; Krylov, S. Y. Chem.
Phys. Lett. 1995, 232, 379.
(11) Teufel, J.; Oh, H.; Hirscher, M.; Wahiduzzaman, M.;
Zhechkov, L.; Kuc, A.; Heine, T.; Denysenko, D.; Volkmer, D. Adv.
Mater. 2013, 25, 635.
(12) Oh, H.; Kalidindi, S. B.; Um, Y.; Bureekaew, S.; Schmid, R.;
Fischer, R. A.; Hirscher, M. Angew. Chem. Int. Ed. 2013, 52, 13219.
(13) Cai, J.; Xing, Y.; Zhao, X. RSC Adv. 2012, 2, 8579.
(14) Kowalczyk, P.; Terzyk, A. P.; Gauden, P. A.; Furmaniak, S.;
Pantatosaki, E.; Papadopoulos, G. K. J. Phys. Chem. B 2015, 119,
15373.
(15) Hattori, Y.; Tanaka, H.; Okino, Fl.; Touhara, H.; Nakahigashi,
Y.; Utsumi, S.;; Kanoh, H.; Kaneko, K. J. Phys. Chem. B 2006, 110,
9764.
(16) Zhao, X.; VillarꢀRodil, S.; Fletcher, A. J.; Thomas, K. M. J.
Phys. Chem. B 2006, 110, 9947.
(17) Tanaka, H.; Kanoh, H.; Yudasaka, M.; Lijima, S.; Kaneko K.
J. Am. Chem. Soc. 2005, 127, 7511.
(45) Grimme, S.; Antony, J.; Ehrlich, S.; Krieg, H. J. Chem.
Phys. 2010, 132, 154104.
(46) Monkhorst, H. J; Pack, J. D. Phys. Rev. B 1976, 13, 5188.
(47) Dudarev, S. L.; Botton, G. A.; Savrasov, S. Y.; Humphreys, C.
J.; Sutton, A. P. Phys. Rev. B 1998, 57, 1505.
(48) Wang, L.; Maxisch, T.; Ceder, G. Phys. Rev. B 2006, 73,
195107.
(49) Henkelman, G.; Uberuaga, B. P.; Jónsson, H. J. Chem.
Phys. 2000, 113, 9901.
(18) Nguyen, T. X.; Jobic, H.; Bhatia, S. K. Phys. Rev. Lett. 2010,
105, 085901.
ACS Paragon Plus Environment

Journal of the American Chemical Society
Page 8 of 13
Table of Contents
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
8
ACS Paragon Plus Environment

Page 9 of 13
Journal of the American Chemical Society
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
Figure 1
142x225mm (300 x 300 DPI)
ACS Paragon Plus Environment

Journal of the American Chemical Society
Page 10 of 13
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
Figure 2
276x219mm (300 x 300 DPI)
ACS Paragon Plus Environment

Page 11 of 13
Journal of the American Chemical Society
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
Figure 3
147x240mm (300 x 300 DPI)
ACS Paragon Plus Environment

Journal of the American Chemical Society
Page 12 of 13
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
Figure 4
146x196mm (300 x 300 DPI)
ACS Paragon Plus Environment

Page 13 of 13
Journal of the American Chemical Society
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
Graphical Abstract
143x133mm (300 x 300 DPI)
ACS Paragon Plus Environment