Crystal Flexibility Design through Local and Global Motility Cooperation

Article abstract of DOI:10.1002/anie.202015257

Incorporating local mobility into a flexible framework promises to create cooperative properties unattainable in a conventional soft porous crystal. In this study, we propose a design strategy that integrates substituent moieties and a flexible porous crystal framework via intra-framework π–π interactions. This integration not only facilitates framework structural transitions but also gives the porous coordination polymers (PCPs) different guest-free structures that depend on the activation conditions. The incorporated flexibility gives the material the ability to discriminate C6 alkane isomers based on different gate-opening behaviors. Thus, the PCP has potential applications in C6 isomer separation, a critical step in the petroleum refining process to produce gasoline with high octane rating. This strategy, based on ligand designability, offers a new approach to flexible PCP structural and functional design.

Full text of DOI:10.1002/anie.202015257

Angewandte
Communications
Chemie
How to cite: Angew. Chem. Int. Ed. 2021, 60, 7030–7035
International Edition:
German Edition:
doi.org/10.1002/anie.202015257
doi.org/10.1002/ange.202015257
Molecular Recognition
Hot Paper
Crystal Flexibility Design through Local and Global Motility
Cooperation
Ping Wang, Ken-ichi Otake,* Nobuhiko Hosono, and Susumu Kitagawa*
Abstract: Incorporating local mobility into a flexible frame-
work promises to create cooperative properties unattainable in
a conventional soft porous crystal. In this study, we propose
a design strategy that integrates substituent moieties and
a flexible porous crystal framework via intra-framework p–p
interactions. This integration not only facilitates framework
structural transitions but also gives the porous coordination
polymers (PCPs) different guest-free structures that depend on
the activation conditions. The incorporated flexibility gives the
material the ability to discriminate C6 alkane isomers based on
different gate-opening behaviors. Thus, the PCP has potential
applications in C6 isomer separation, a critical step in the
petroleum refining process to produce gasoline with high
octane rating. This strategy, based on ligand designability,
offers a new approach to flexible PCP structural and func-
tional design.
properties, other than an altered gate-opening pressure, have
not been demonstrated. This lack of progress might be
attributed to a lack of interactions that integrate motion of the
substituents and framework and regulate the behavior of
structural transitions.
Herein, we proposed a rational incorporation strategy
that connects a flexible framework and mobile fragments
through intra-framework p–p interactions (Figure 1). The
stacking arenes adjust position synergistically during a struc-
tural transition, endowing this PCP substituent-framework
cooperative flexibility. The synchronized motion confines the
ligandsꢀ spatial position, preventing the framework pores
from collapsing upon solvent removal, and facilitates the
framework structural transition at a relatively small gate-
opening pressure. Furthermore, the flexibility integration
recasts the system energy landscape, enabling the PCP to
adopt various guest-free structures depending on the activa-
tion conditions, either thermal or supercritical drying. In
contrast, isostructural PCPs without p–p interactions deform
without restraint into a compact closed structure with gate-
opening behavior that is triggered with difficulty by guest
molecules regardless of the activation approach.
F
lexible porous coordination polymers (PCP) are charac-
terized by reversible structural transitions that originate from
a global lattice change triggered by guest uptake and
release.[1–7] This reversible structural transformation, which
is termed gate-opening behavior, occurs at a certain threshold
pressure (a gate-opening pressure).[8–18] This feature can
endow flexible PCPs with an exclusive molecular discrim-
ination capability, which thus makes them of interest as
separation media.[19–25] However, a rational control strategy of
flexible PCP remains elusive. Recently, local motion of the
dynamic entities in rigid PCP pores has also drawn interest.
These moieties can form rigid PCPs with an adaptive nature
and selectivity by regulating the guest diffusion inside the
pore.[26–29]
To introduce the dynamic aromatic substituents, we first
designed functionalized 1,4-benzenedicarboxylic acid
(H2bdc) ligands, decorated with mobile aromatic substituents.
Incorporation of local dynamics into flexible frameworks
is a promising method for creating cooperative flexibility that
is unattainable in conventional PCPs.[30] Efforts to this end
have been attempted by introducing dynamic sidearms into
the pillar of flexible PCPs;[31–33] however, novel cooperative
[*] Dr. P. Wang, Prof. Dr. K. Otake, Prof. Dr. N. Hosono,
Prof. Dr. S. Kitagawa
Institute for Integrated Cell-Material Sciences (WPI-iCeMS)
Kyoto University Institute for Advanced Study, Kyoto University
Yoshida Ushinomiya-cho, Sakyo-ku, Kyoto 606-8501 (Japan)
E-mail: ootake.kenichi.8a@kyoto-u.ac.jp
kitagawa@icems.kyoto-u.ac.jp
Figure 1. PCP structural transformation. a) Conventional flexible PCPs
undergo unrestricted deformation after solvent removal, transforming
to a specific close phase; if the framework deforms severely, the
compact close phase cannot be reopened by guest molecules. b) PCP
with local and global flexibility integrated by p–p interaction. The
interaction restricts the PCP deformation, leading to various porous
guest-free structures that can be reopened by guest molecule inclu-
sion.
Prof. Dr. N. Hosono
Department of Advanced Materials Science, Graduate School of
Frontier Sciences, The University of Tokyo
5-1-5 Kashiwanoha, Kashiwa, Chiba 277-8561 (Japan)
Supporting information and the ORCID identification number(s) for
the author(s) of this article can be found under:
https://doi.org/10.1002/anie.202015257.
7030
ꢀ 2020 Wiley-VCH GmbH
Angew. Chem. Int. Ed. 2021, 60, 7030 –7035

Angewandte
Communications
Chemie
The aromatic substituents were connected by short alkane
ligand (dihedral angle = 86.68; Supporting Information, Fig-
chains, which facilitated stacking of the substituentꢀs stacking
with neighbor ligands. It has previously been theoretically[34]
and experimentally[35,36] shown that electron-withdrawing
groups relieve the electrostatic repulsion between arenes
and facilitate p–p stacking. For this reason, we synthesized
two ligands for comparison: 2,5-bis[(2-nitrobenzyl)oxy]-1,4-
benzenedicarboxylic acid (H2bdc-NO2) with an electron-
withdrawing nitro group and 2,5-bis(benzyloxy)-1,4-benzene-
dicarboxylic acid (H2bdc-H) with a benzyl substituent (Fig-
ure 2a).
ure S2). This vertical conformation eliminated the possibility
of p–p interactions between benzyl substituents and bdc-H by
minimizing stacking. Thus, the 1-H 3D channels had a void
space as high as 34.3%. We observed CH···p interactions
between benzyl substituents with a dihedral angle of central
benzene and CH···p distance of 88.58 and 2.9 ꢁ, respectively.
Solvent molecules accommodated in the as-synthesized
PCPs were exchanged with acetonitrile before 808C thermal
activation in a vacuum, which was confirmed by thermogra-
vimetric analyses (TGA; Supporting Information, Figures S3,
S4). After activation, the guest-free 1-NO2 (1-NO2-vac) had
a sharp powder X-ray diffraction (PXRD) pattern distinct
from that of the as-synthesized one, as confirmed by the
disappearance of the peak at 6.288 (Figure 3a). Conversely,
the thermally activated 1-H (1-H-vac) had a broad PXRD
signal with a single intense peak at 6.528 (Figure 3b). The
structural transformation of both PCPs was reversible, as
indicated by the recovery of the PXRD patterns after soaking
in N,N-dimethylformamide (DMF) at room temperature.
The N2 (77 K) and CO2 (195 K) adsorption isotherms
were measured to test the PCP flexibility and porosity.
Though both adsorbed trace amounts of N2 because of its
small pore size (Supporting Information, Figures S5, S6),
distinct uptake behavior of 1-NO2-vac and 1-H-vac was
observed in CO2 sorption. 1-NO2-vac had a typical gate-
opening type adsorption isotherm for CO2: an initial uptake
of 28 mLgÀ1 before the gate-opening pressure at P/P0 = 0.23
and a saturated capacity of 172 mLgÀ1 after the structural
transformation (Figure 3c). The desorption isotherm does not
trace the adsorption curve and sharply decreased at P/P0 =
0.1. The sharp adsorption increase and desorption decrease
Figure 2. a) 3D pillared-layer PCPs synthesized by aromatic substitu-
ents bearing ligand, Zn(NO3)2·6H2O, and bipyridine. Top view of as-
synthesized b) 1-NO2 and c) 1-H. Aromatic parts involved in the p–p
interaction are highlighted in orange. Zn purple, C gray, N blue, O red.
Hydrogen atoms and solvent molecules are omitted for clarity. The
dashed line and numbers mark the center distance of arenes that form
p–p dimers.
Single crystals were obtained by a solvothermal reaction
in the presence of Zn(NO3)2·6H2O, 4,4’-bipyridine (bpy), and
the aromatic substituent bearing a 1,4-benzenedicarboxylic
acid. Single crystal X-ray diffraction (SXRD) analyses
revealed that the two PCPs were isostructural and charac-
terized by a non-interpenetrated 3D pillared-layer structure
with the chemical formula of [Zn2(bdc-R)2(bpy)]·(solvent)n}
(1-R) (Figures 2; R = NO2 or H). bdc-R ligands and Zn2
paddlewheel clusters formed 2D square grid-like layers,
further connected by bpy pillar ligands to form 3D frame-
works.
According to SXRD analysis, parallel p–p stacking
occurred for 1-NO2. The nitro-substituted arenes stack with
the central benzene of bdc-NO2 ligand, forming a p–p
interactive dimer with a center distance of 3.6 ꢁ and
a dihedral angle of 8.68 (Figure 2b). The aromatic substituents
aligned in channels along the b axis with the nitro groups
pointing toward the channel center, dividing the PCP
channels into isolated cages with a void space of 10.8%
(Supporting Information, Figure S1). On the other hand, in 1-
H, all the appended benzenes anchored at the corner of the
channel and remained perpendicular to the nearest bdc-H
Figure 3. a) Powder X-ray diffraction patterns (PXRD) of 1-NO2 simu-
lated from as-synthesized single crystal (indigo), 1-NO2 experimental
(blue), 1-NO2-vac simulated from thermal activated single crystal
(red), 1-NO2-vac experimental (orange), and 1-NO2-vac regenerated
from DMF (green). b) PXRD of 1-H simulated from as-synthesized
single crystal (indigo), 1-H experimental (blue), 1-H-vac experimental
(orange), and 1-H-vac regenerated from DMF (green). Adsorption
isotherms of 1-NO2-vac (red circles) and 1-H-vac (black squares) for
c) CO2 at 195 K and d) toluene at 298 K. Closed symbols denote
adsorption and empty symbols denote desorption.
Angew. Chem. Int. Ed. 2021, 60, 7030 –7035
ꢀ 2020 Wiley-VCH GmbH
www.angewandte.org
7031

Angewandte
Communications
Chemie
with the hysteresis indicate that the structural transformation
was triggered by CO2 adsorption, as further supported by
changes of the in situ PXRD patterns under various CO2
pressures (Supporting Information, Figure S7). Conversely, 1-
H-vac did not have gate-opening behavior for CO2 at
atmospheric pressure, resulting in a small capacity of
33 mLgÀ1 (Figure 3c; Supporting Information, Figure S8).
Adsorption behaviors of other gas and vapor species were
also measured. 1-NO2-vac had gate-open type adsorption for
C2H4 (170 K, P/P0 = 0.06; Supporting Information, Fig-
ure S9), C2H6 (185 K, P/P0 = 0.20; Supporting Information,
Figure S10), and toluene (298 K, P/P0 = 0.08; Figure 3d).
Conversely, 1-H-vac had only a small uptake of these guests
without a gate-open transformation (Figure 3c,d; Supporting
Information, Figures S9, S10).
To uncover the reason for the distinctive sorption
behaviors of the isostructural PCPs, we used variable-temper-
ature powder X-ray diffraction (VTPXRD) to monitor the
guest solvent removal process. The PXRD pattern of as-
synthesized 1-H was maintained up to 808C but transformed
gradually from 808C to 1208C. At 1208C, the structural
transformation was completed, and the resulting pattern had
only one broad peak at 6.528, which is assigned to the PCP
inter-layer distance (Supporting Information, Figure S11).
Diffraction signals from 2D grid layers all faded and
disappeared during thermal activation, as exemplified by
the (200) peak relating to the distance between bdc-H ligands
(Supporting Information, Figure S12). The changes of the
VTPXRD indicate that upon DMF removal, 1-H lost its long-
range order in the a and b directions, and the ordered grids of
the [Zn (bdc-H)]n 2D layer collapsed.
Unlike the gradual transition of 1-H, the VTPXRD of 1-
NO2 underwent an abrupt structural change at 908C (Sup-
porting Information, Figure S13). To elucidate the reason for
the rapid change of 1-NO2, we examined single crystals of 1-
NO2-vac by SXRD. Diffraction analyses of 1-NO2-vac con-
firmed that the framework connectivity was the same as that
of 1-NO2; however, the space group changed from monoclinic
ꢀ
P2/c to triclinic P1 (Supporting Information, Table S1). 1-
Figure 4. Structure of guest-free 1-NO2. a) Side and b) top view of
vacuum activated 1-NO2-vac. c) Side and d) top view of supercritical
CO2 activated 1-NO2-SCO2. The coordination environment of the Zn2
paddlewheel clusters is shown in the inset. Aromatic parts involved in
the p–p interactions are highlighted in orange and green. Zn purple,
C gray, N blue, O red. Hydrogen atoms are omitted. The dashed line
and numbers mark the center distance of arenes that formed p–p
dimers.
NO2-vac underwent slippage between layers and the inter-
layer distance decreased from 14.0 ꢁ (1-NO2) to 10.8 ꢁ (1-
NO2-vac) (Figure 4a). Notably, despite the considerable
interlayer shrinkage, the p–p stacking dimers were well
preserved in the grid of [Zn(bdc-NO2)]n layers (Figure 4b). In
1-NO2-vac, 50% of the dimers were almost identical to those
of the as-synthesized 1-NO2 with the same focal distance
(3.6 ꢁ) (Figure 4b, orange part); the other 50% rear-ranged,
such that the bdc-NO2 ligand and nitrobenzyl substituent
rotated along the c axis, yielding a stacking distance of 3.7 ꢁ
(Figure 4b, green part). Because the two-types of p–p dimers
were distributed in two neighboring grids separately, the 2D
layer was corrugated.
The VTPXRD results confirmed that in the absence of
intra-framework p–p interactions (1-H), the long-range order
of the 2D grid layers was easily disrupted upon guest removal;
whereas in the case of intra-framework p–p interactions (1-
NO2), the crystal-to-crystal structural transition occurred
when p–p stacking dimers were maintained even after
thermal activation. Therefore, we attribute the differences
between the isostructural PCPs to the substituent-framework
p–p interaction. These interactions confined the ligands
spatial position and maintained the p–p stacking configura-
tion, thus restraining distortion of the 2D-grid. Consequently,
the p–p interaction-controlled deformation created a porous
guest-free framework that could be reopened by guest
molecules. Conversely, unrestrained deformation yielded
compact-close 1-H-vac disordered 2D layers that required
an unpractically high gate-opening pressure (Figure 3c,d;
Supporting Information, Figures S9, S10). Our assumption
was further tested by investigating isostructural PCP with the
7032
www.angewandte.org
ꢀ 2020 Wiley-VCH GmbH
Angew. Chem. Int. Ed. 2021, 60, 7030 –7035

Angewandte
Communications
Chemie
bdc-Me ligand (1-Me; Supporting Information, Figures S14–
resulting in 1-NO2-SCO2. Thus, we assumed the deformation
S17). The as-synthesized 1-Me had weak p–p interactions
with a 3.8 ꢁ stacking distance, which is longer than that those
of 1-NO2 (3.6 ꢁ). As expected from the stabilization effect of
p–p stacking dimers, for CO2 and toluene adsorption, guest-
free 1-Me-vac had gate-opening behavior but a large gate-
opening pressure compared with 1-NO2-vac (Supporting
Information, Figures S18, S19).
When we activated 1-NO2 by supercritical CO2, we
obtained an activated phase (1-NO2-SCO2) unlike that of
the thermally activated one (1-NO2-vac; Supporting Informa-
tion, Figures S20, S21). The SXRD structure determination of
1-NO2-SCO2 shows that the framework connectivity changed
because zinc ions varied in their coordination environment
(Figure 4c). In 1-NO2-SCO2, one carboxylate out of four
became monodentate and the Zn2 paddlewheel deformed.
The cluster deformation compressed the space between the
2D layers, resulting in two kinds of 1D channels along the c
axis. In the cluster distorted grids, p–p stacking between the
nitrobenzyl substituents and the bdc-NO2 ligand was retained
(Figure 4d). The stacking distance of the 50% p–p inter-
difference of PCP 1-NO2 might result from structural stress
induced by varied activation approaches.[37]
C6 alkane isomers are the main components of gasoline,
including linear n-hexane (HEX), mono-branched 3-methyl-
pentane (3MP), and di-branched 2,2-dimethylbutane
(2,2DMB). C6 sorting is a vital process in industrial gasoline
upgrading because only 2,2DMB is desirable for refined
gasoline because of its high research octane number
(RON).[38] Several PCPs have been investigated for C6
isomer sorting; however, all suffer from the poor selectiv-
ity.[39–41] We posit that the aromatic substituents that crowd the
1D channels are able to form CH-p interactions with C6
molecules. The soft nature of the PCPs can discriminate
against differences in the host-guest interactions, leading to
varied gate-opening behavior. To test this hypothesis, we
performed C6 isomer vapor adsorption measurements.
Single-component vapor adsorption at 308C indicated the
perfect discrimination performance of 1-NO2-vac: HEX was
adsorbed by the gate-opening process (Pgo/Psat = 0.09) with
a saturated capacity of 145 mggÀ1; 3MP induced no gate-
opening adsorption but instead had a linear isotherm with
41 mggÀ1 capacity; 2,2DMB was completely size-excluded
with no adsorption (Figure 5; Supporting Information, Figur-
es S26, S27). On the contrary, 1-NO2-SCO2 had gate-opening
adsorption to HEX, but was unable to uptake either 3MP or
2,2DMB (Supporting Information, Figure S28); 1-H-vac pre-
vented entrance of all isomers into its pore (Supporting
Information, Figures S29, S30). Therefore, the low porosity of
1-NO2-SCO2 and the large gate-opening pressure of 1-H-vac
rendered these materials unsuitable for isomer discrimina-
tion.
À
action increased from 3.6 to 3.7 ꢁ because of Zn O bond
cleavage. Unlike 1-NO2-vac, two stacking types coexisted in
every grid of 1-NO2-SCO2. 1-NO2-SCO2 was readily opened
by CO2 at 195 K, as characterized by a gate-opening pressure
of P/P0 = 0.37 (Supporting Information, Figure S22). Con-
versely, the supercritical CO2 activated 1-H (1-H-SCO2) had
a broad PXRD pattern and flat CO2 uptake similar to that of
1-H-vac (Supporting Information, Figures S23–S25). The
SXRD results confirm that 1-NO2-SCO2 maintained its
structure even at 808C when activated for 12 h and did not
transform to 1-NO2-vac.
To uncover the reason for the 1-NO2 structural depend-
ence on the activation approach, crystal structures of guest-
free PCPs were compared. Despite the differently deformed
frameworks of 1-NO2-vac and 1-NO2-SCO2, their p–p stack-
ing configurations were similar. In the absence of intra-
framework p–p interactions (1-H), the 2D grid layers,
composed of bdc and Zn2 paddlewheel clusters, deformed
during the activation (Supporting Information, Figures S11,
S12). In 1-NO2, the p–p attraction locked the layer distortion,
recasting the soft system energy landscape. To stabilize the
guest-free frameworks, different deformation routes were
followed: the interlayer slippage in 1-NO2-vac and the cluster
switch in 1-NO2-SCO2. After thermal activation, the angles
between the bpy pillar and the 2D grid layer changed from 908
in 1-NO2 to 468 in 1-NO2-vac, and the Zn2 clusters from
neighbor layers linked by each bpy shifted to 7.6 ꢁ. While
after supercritical drying, the bpy pillars in 1-NO2-SCO2 tilted
with respect to the 2D grid layer only at an angle of 678 in
accompany with the shift of the Zn2 cluster to 6.5 ꢁ. The most
striking change in 1-NO2-SCO2 is that the Zn2 paddlewheel
In summary, we have developed a crystal flexibility design
strategy by integrating substituent and framework motility via
intra-framework p–p interactions of ligands. The flexible and
synchronous movement confined framework distortion upon
À
cluster changes its structure: one out of eight Zn O bonds
cleaved. The conditions such as solvents, temperature, and
pressure play a crucial role in flexible PCP structural changes.
Unlike thermal activation in vacuo, supercritical CO2 activa-
tion applied high pressure to liquefy CO2, which could cause
the structural change of Zn2 cluster in 1-NO2. Subsequently,
in-pore CO2 solvents were removed without surface tension,
Figure 5. 1-NO2-vac adsorption isotherms at 303 K for HEX (red), 3MP
(yellow), and 22DMB (blue). Closed symbols denote adsorption, and
empty symbols denote desorption. Kinetic diameter (blue number, in
ꢀ) and RON (red number) of each isomer is marked below the
chemical structure.
Angew. Chem. Int. Ed. 2021, 60, 7030 –7035
ꢀ 2020 Wiley-VCH GmbH
www.angewandte.org
7033

Angewandte
Communications
Chemie
[12] S. Kusaka, A. Kiyose, H. Sato, Y. Hijikata, A. Hori, Y. Ma, R.
Matsuda, J. Am. Chem. Soc. 2019, 141, 15742 – 15746.
[13] A. X. Zhu, Q. Y. Yang, S. Mukherjee, A. Kumar, C. H. Deng,
A. A. Bezrukov, M. Shivanna, M. J. Zaworotko, Angew. Chem.
Int. Ed. 2019, 58, 18212 – 18217; Angew. Chem. 2019, 131, 18380 –
18385.
[14] M.-H. Yu, B. Space, D. Franz, W. Zhou, C. He, L. Li, R. Krishna,
Z. Chang, W. Li, T.-L. Hu, X.-H. Bu, J. Am. Chem. Soc. 2019,
141, 17703 – 17712.
solvent removal, giving PCPs with different porous guest-free
structures that depend on the activation approach. These
PCPs discriminated between C6 alkane isomers by a distinct
gate-opening behavior. These features suggest potential for
applications as an isomer splitter for gasoline upgrading. The
availability of various aromatic substituents and designability
of PCPs enables the creation of an aromatic p–p interactions
between local mobility and soft porous crystals. This structure
in turn provides a new chemical approach for flexible PCP
structural and functional design.
[15] M. Souto, J. Romero, J. Calbo, I. J. Vitꢂrica-Yrezꢃbal, J. L. Zafra,
J. Casado, E. Ortꢄ, A. Walsh, G. Mꢄnguez Espallargas, J. Am.
Chem. Soc. 2018, 140, 10562 – 10569.
[16] Y. Hui, G. Feng, L. Prem, G. Wen-Yang, W. Hui, J. B. Leonard,
Z. Wei, Z. Jian, A. Briana, M. Shengqian, ACS Cent. Sci. 2018, 4,
1194 – 1200.
Acknowledgements
[17] Y. T. Zheng, H. Sato, P. Y. Wu, H. J. Jeon, R. Matsuda, S.
Kitagawa, Nat. Commun. 2017, 8, 100.
[18] Q. Gao, J. Xu, D. Cao, Z. Chang, X.-H. Bu, Angew. Chem. Int.
Ed. 2016, 55, 15027 – 15030; Angew. Chem. 2016, 128, 15251 –
15254.
[19] Y. Yan, E. J. Carrington, R. Pꢅtuya, G. F. S. Whitehead, A.
Verma, R. K. Hylton, C. C. Tang, N. G. Berry, G. R. Darling,
M. S. Dyer, D. Antypov, A. P. Katsoulidis, M. J. Rosseinsky, J.
Am. Chem. Soc. 2020, 142, 14903 – 14913.
[20] Y. Gu, J.-J. Zheng, K.-i. Otake, K. Sugimoto, N. Hosono, S.
Sakaki, F. Li, S. Kitagawa, Angew. Chem. Int. Ed. 2020, 59,
15517 – 15521; Angew. Chem. 2020, 132, 15647 – 15651.
[21] H. Sato, W. Kosaka, R. Matsuda, A. Hori, Y. Hijikata, R. V.
Belosludov, S. Sakaki, M. Takata, S. Kitagawa, Science 2014, 343,
167.
This study was supported by KAKENHI Grants-in-Aid for
Scientific Research (S) (JP18H05262), Early-Career Scien-
tists (JP19K15584), Scientific Research (B) (JP18H02072),
and Scientific Research on Innovative Areas “Molecular
Engine” (JP19H05381) from the Japan Society for the
Promotion of Science (JSPS). NMR spectroscopy was per-
formed at the iCeMS Analysis Center, Kyoto University
Institute for Advanced Study. We thank Nanae Shimanaka at
iCeMS, Kyoto University, for her contributions to SXRD
analysis.
Conflict of interest
[22] D.-D. Zhou, P. Chen, C. Wang, S.-S. Wang, Y. Du, H. Yan, Z.-M.
Ye, C.-T. He, R.-K. Huang, Z.-W. Mo, N.-Y. Huang, J.-P. Zhang,
Nat. Mater. 2019, 18, 994 – 998.
The authors declare no conflict of interest.
[23] S. Sen, N. Hosono, J. J. Zheng, S. Kusaka, R. Matsuda, S. Sakaki,
S. Kitagawa, J. Am. Chem. Soc. 2017, 139, 18313 – 18321.
[24] L. Li, R.-B. Lin, R. Krishna, X. Wang, B. Li, H. Wu, J. Li, W.
Zhou, B. Chen, J. Am. Chem. Soc. 2017, 139, 7733 – 7736.
[25] M. L. Foo, R. Matsuda, Y. Hijikata, R. Krishna, H. Sato, S.
Horike, A. Hori, J. Duan, Y. Sato, Y. Kubota, M. Takata, S.
Kitagawa, J. Am. Chem. Soc. 2016, 138, 3022 – 3030.
[26] S. Horike, R. Matsuda, D. Tanaka, S. Matsubara, M. Mizuno, K.
Endo, S. Kitagawa, Angew. Chem. Int. Ed. 2006, 45, 7226 – 7230;
Angew. Chem. 2006, 118, 7384 – 7388.
[27] D. I. Kolokolov, H. Jobic, A. G. Stepanov, V. Guillerm, T. Devic,
C. Serre, G. Fꢅrey, Angew. Chem. Int. Ed. 2010, 49, 4791 – 4794;
Angew. Chem. 2010, 122, 4901 – 4904.
[28] K. Zhu, C. A. OꢀKeefe, V. N. Vukotic, R. W. Schurko, S. J. Loeb,
Nat. Chem. 2015, 7, 514 – 519.
Keywords: activation method · flexibility ·
molecular recognition · motility cooperation ·
porous coordination polymers
[1] S. Kitagawa, R. Kitaura, S. Noro, Angew. Chem. Int. Ed. 2004, 43,
2334 – 2375; Angew. Chem. 2004, 116, 2388 – 2430.
[2] S. Horike, S. Shimomura, S. Kitagawa, Nat. Chem. 2009, 1, 695 –
704.
[3] A. Schneemann, V. Bon, I. Schwedler, I. Senkovska, S. Kaskel,
R. A. Fischer, Chem. Soc. Rev. 2014, 43, 6062 – 6096.
[4] Z. Chang, D.-H. Yang, J. Xu, T.-L. Hu, X.-H. Bu, Adv. Mater.
2015, 27, 5432 – 5441.
[29] C. Gu, N. Hosono, J.-J. Zheng, Y. Sato, S. Kusaka, S. Sakaki, S.
Kitagawa, Science 2019, 363, 387.
[5] T. D. Bennett, A. K. Cheetham, A. H. Fuchs, F.-X. Coudert, Nat.
Chem. 2017, 9, 11 – 16.
[30] S. Krause, N. Hosono, S. Kitagawa, Angew. Chem. Int. Ed. 2020,
59, 15325 – 15341; Angew. Chem. 2020, 132, 15438 – 15456.
[31] J. Seo, R. Matsuda, H. Sakamoto, C. Bonneau, S. Kitagawa, J.
Am. Chem. Soc. 2009, 131, 12792 – 12800.
[32] S. Henke, A. Schneemann, A. Wutscher, R. A. Fischer, J. Am.
Chem. Soc. 2012, 134, 9464 – 9474.
[33] S. Jeoung, S. Lee, J. H. Lee, S. Lee, W. Choe, D. Moon, H. R.
Moon, Chem. Commun. 2019, 55, 8832 – 8835.
[34] M. O. Sinnokrot, C. D. Sherrill, J. Am. Chem. Soc. 2004, 126,
7690 – 7697.
[6] S. K. Elsaidi, M. H. Mohamed, D. Banerjee, P. K. Thallapally,
Coord. Chem. Rev. 2018, 358, 125 – 152.
[7] F. Bigdeli, C. T. Lollar, A. Morsali, H. C. Zhou, Angew. Chem.
Int. Ed. 2020, 59, 4652 – 4669; Angew. Chem. 2020, 132, 4680 –
4699.
[8] B.-Q. Song, Q.-Y. Yang, S.-Q. Wang, M. Vandichel, A. Kumar, C.
Crowley, N. Kumar, C.-H. Deng, V. GasconPerez, M. Lusi, H.
Wu, W. Zhou, M. J. Zaworotko, J. Am. Chem. Soc. 2020, 142,
6896 – 6901.
[9] P. H. Dominik, A. K. Ryan, A. B. Michael, A. T. Benjamin,
M. B. Craig, R. L. Jeffrey, Chem. Sci. 2020, 11, 6709 – 6716.
[10] T. Kundu, M. Wahiduzzaman, B. B. Shah, G. Maurin, D. Zhao,
Angew. Chem. Int. Ed. 2019, 58, 8073 – 8077; Angew. Chem. 2019,
131, 8157 – 8161.
[35] F. Cozzi, M. Cinquini, R. Annuziata, J. S. Siegel, J. Am. Chem.
Soc. 1993, 115, 5330 – 5331.
[36] F. Cozzi, M. Cinquini, R. Annunziata, T. Dwyer, J. S. Siegel, J.
Am. Chem. Soc. 1992, 114, 5729 – 5733.
[37] A. P. Nelson, O. K. Farha, K. L. Mulfort, J. T. Hupp, J. Am.
Chem. Soc. 2009, 131, 458 – 460.
[38] H. Wang, J. Li, Acc. Chem. Res. 2019, 52, 1968 – 1978.
[11] S. Suginome, H. Sato, A. Hori, A. Mishima, Y. Harada, S.
Kusaka, R. Matsuda, J. Pirillo, Y. Hijikata, T. Aida, J. Am. Chem.
Soc. 2019, 141, 15649 – 15655.
7034 www.angewandte.org
ꢀ 2020 Wiley-VCH GmbH
Angew. Chem. Int. Ed. 2021, 60, 7030 –7035

Angewandte
Communications
Chemie
[39] P. S. Bꢃrcia, F. Zapata, J. A. C. Silva, A. E. Rodrigues, B. Chen, J.
Proserpio, Y. J. Chabal, T. Thonhauser, J. Sun, Y. Han, J. Li, Nat.
Phys. Chem. B 2007, 111, 6101 – 6103.
Commun. 2018, 9, 1745.
[40] Z. R. Herm, B. M. Wiers, J. A. Mason, J. M. van Baten, M. R.
Hudson, P. Zajdel, C. M. Brown, N. Masciocchi, R. Krishna, J. R.
Long, Science 2013, 340, 960.
[41] H. Wang, X. Dong, J. Lin, S. J. Teat, S. Jensen, J. Cure, E. V.
Alexandrov, Q. Xia, K. Tan, Q. Wang, D. H. Olson, D. M.
Manuscript received: November 15, 2020
Accepted manuscript online: December 29, 2020
Version of record online: February 18, 2021
Angew. Chem. Int. Ed. 2021, 60, 7030 –7035
ꢀ 2020 Wiley-VCH GmbH
www.angewandte.org
7035