Guest-Induced Breathing Effect in a Flexible Molecular Crystal

Article abstract of DOI:10.1002/anie.201510637

By introducing a flexible component into a molecular building block, we present an unprecedented alkyl-decorated flexible crystalline material with a breathing behavior. Its selective adsorption is derived from the breathing effect induced by a guest trig

Full text of DOI:10.1002/anie.201510637

Angewandte
Communications
Chemie
International Edition: DOI: 10.1002/anie.201510637
German Edition: DOI: 10.1002/ange.201510637
Host–Guest Systems
Guest-Induced Breathing Effect in a Flexible Molecular Crystal
Yujie Sheng, Qibin Chen,* Junyao Yao, Yunxiang Lu, Honglai Liu,* and Sheng Dai*
Abstract: By introducing a flexible component into a molec-
ular building block, we present an unprecedented alkyl-
decorated flexible crystalline material with a breathing behav-
ior. Its selective adsorption is derived from the breathing effect
induced by a guest triggered alkyl transformation. This feature
allows the crystal to take up 2.5 mmolgÀ1 of chloroform with
high adsorption selectivity (CHCl3/EA > 2000 for example),
implying a potential application in sorption separation and
chemical sensors.
in one assembly unit, where the former are responsible for
retaining the integral scaffold, while the latter are responsible
for a structural transformation response to external stimuli.
Recently, we found that a Gemini surfactant composed of
two alkyl chains, a typical flexible group, and a rigid biphenyl
spacer, formed a crystal in which water was included.[8]
Unfortunately, stability of the crystal was lost upon guest
removal. Nevertheless, because such Gemini molecules
possess both rigid and flexible groups, we hypothesized that
construction of a novel PMC with an external stimulus-
responsive behavior arising from the structural transforma-
tion of alkyl chains was reasonable.
The uptake and encapsulation of guest species by micro-
porous materials has great potential in a wide range of
applications, such as selective molecular separations, chemical
sensing, heterogeneous catalysis, and gas storage.[1] Over the
past few years porous molecular crystals (PMCs) have
evolved as a promising alternative to traditional adsorbents,
such as zeolites, activated carbon, metal–organic frameworks
(MOFs), covalent organic frameworks, or network polymers,
because of their distinguishing features.[2] Generally, PMCs
should be fabricated by removal of solvent molecules from
their inclusion crystals.[3] However, a major challenge is that
PMCs do not commonly retain their incipient porosity upon
guest removal, but rather collapse to form a dense phase.
Thus, a more general design strategy would incorporate
molecular rigidity into building blocks to prevent precursors
from close-packing during evacuation.[4] In other words,
molecular flexibility has to be avoided. In contrast, to control
the pore size and achieve switchable processes in MOFs,
flexible components (namely metal ions and organic linkers)
are usually employed.[5] For example, a reversible open-dense
framework transformation was successfully facilitated by
distorting nodes and bending struts.[6] Moreover, MOFs can
enable enhanced selective guest sorption in response to
external stimuli by a gate-opening effect resulting from the
structural transformation of a flexible architecture.[7] Inspired
by PMC and MOF design concepts, we propose construction
of a new bimodal PMC containing rigid and flexible moieties
The primary objective of this work was to test this
hypothesis by changing the spacer and the crystallizing
method, with a view to enhancing the crystal stability during
guest removal. With this purpose in mind, we synthesized
a new Gemini surfactant, N,N’-((diazene-1,2-diylbis(4,1-phe-
nylene))bis(methylene))bis(N,N-dimethyl-dodecan-1-ami-
Scheme 1. Molecular structure of a Gemini molecule employed as
a PMC (porous molecular crystal) building block.
nium) bromide (Scheme 1), by introducing an azobenzene
spacer into the structure (see Supporting Information). A
flexible PMC was grown by liquid phase diffusion of tetra-
chloroethane into an ethyl acetate solution of the Gemini
surfactant, producing PMC-1. Intriguingly, PMC-1 only
responds to chloroform vapor, exhibiting a reversible adsorp-
tion–desorption process after exposure to chloroform vapor
and nitrogen atmosphere, thus indicating a better selectivity
for chloroform compared to other organic vapors. This
preferential sorption property probably stems from the
formation of transient pores, which originate from the
bending of terminal chains in PMC-1 as they accommodate
guest molecules. Simultaneously, the structural transforma-
tion of flexible moieties and adsorption–desorption of chloro-
form lead to a reversible expansion/shrinkage of the PMC-
1 unit cell.
[*] Y. Sheng, Prof. Q. Chen, J. Yao, Dr. Y. Lu, Prof. H. Liu
State Key Laboratory of Chemical Engineering, Department of
Chemistry, East China University of Science and Technology
Shanghai, 200237 (P.R. China)
E-mail: qibinchen@ecust.edu.cn
Initially, we examined the N2 adsorption–desorption
isotherm of PMC-1. The isotherm plot showed that PMC-
1 takes up extremely low volumes of N2 gas at low relative-
pressures at 77 K (Supporting Information, black scatter in
Figure S1). Thus, PMC-1 was deemed to be non-porous. A
low 8.1 m2 gÀ1 Brunauer–Emmett–Teller surface area may be
attributed to the solid surface and piled pores of crystal
particles. Despite this, we investigated the properties of the
material at 298 K with the aim of understanding whether
thermal stimuli could “activate” the transition of alkyl chains,
hlliu@ecust.edu.cn
Prof. S. Dai
Chemical Sciences Division, Oak Ridge National Laboratory
Oak Ridge, TN 37831 (USA)
and
Department of Chemistry, University of Tennessee
Knoxville, TN 37996-1600 (USA)
E-mail: dais@ornl.gov
Supporting information for this article is available on the WWW
under http://dx.doi.org/10.1002/anie.201510637.
3378
ꢀ 2016 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Angew. Chem. Int. Ed. 2016, 55, 3378 –3381

Angewandte
Communications
Chemie
and in so doing generate void spaces. Unfortunately, this
out because the molecules employed are of a size less than
hypothesis was not substantiated (Supporting Information,
red scatter in Figure S1).
that of CHCl3, and have a low CH2Cl2, CH3OH, and C2H5OH
uptake, and a negligible N2 uptake. It is worth noting that
PMC-1 crystal particles agglomerated above the relative-
pressure mid-range when a routine gas-adsorption measure-
ment method was used to determine adsorption–desorption
characteristics of the four VOCs (CHCl3, CH2Cl2, CH3OH,
and C2H5OH). Therefore, we focused on the sorption proper-
ties in the low relative-pressure region in case of the
occurrence of capillary condensation. These results suggest
that PMC-1 responds exclusively to CHCl3 vapor as a result of
structural transition. Furthermore, 10 repeated sorption
cycles clearly demonstrate that PMC-1 can adsorb/desorb
CHCl3 with good reproducibility, indicating that structural
transformations are reversible (Figure 1b).
Surprisingly, PMC-1 showed a significant uptake of CHCl3
when exposed to chloroform vapor in the low relative-
pressure region at 298 K (Figure 1a). The isotherm curve of
CHCl3 revealed a steep rise in the range of p/p0 from about
0.02–0.06. A plateau after saturation suggested that PMC-1 is
porous to CHCl3. Desorption of CHCl3 only starts at p/p0 <
0.055, resulting in a hysteresis loop with a width of 0.66 kPa.
Hysteresis can be attributed to significant adsorbent–adsor-
bate interactions arising from hydrogen-bonding or dipole–
dipole contacts.[9] In contrast, it is surprising that for other
volatile organic compounds (VOCs) PMC-1 reveals a slight
linear or level uptake, indicating that this material appears to
be essentially non-porous to such VOCs. These results are
consistent with the N2 adsorption–desorption isotherms.
Here, a molecular sieving effect by PMC-1 should be ruled
In Figure 1a, CHCl3 uptake is close to 2.5 mmolgÀ1,
comparable to that of other porous networks.[10] PMC-
1 demonstrates slight adsorption of the other VOCs, con-
sequently enabling a high selective adsorption of CHCl3 over
the other VOCs. For example, the ratio of CHCl3/EA
selectivity exceeds 2000 (Supporting Information, Table S1).
Even the low CHCl3/CH2Cl2 selectivity (ca. 14) determined by
our work is an order of magnitude higher than that described
in previous reports.[11] For practical applications, adsorption–
desorption rates were evaluated using mass variation by
alternatively exposing PMC-1 to a N2 stream saturated with
CHCl3 vapor and a dry N2 stream. The adsorption process
took only six minutes to reach saturation in a CHCl3/N2
mixture, while the desorption process took one hour under
N2 purging (Supporting Information, Figure S2). Moreover,
agglomeration of PMC-1 particles did not take place in the
CHCl3/N2 mixture, unlike the situation described above using
routine methods. Although we did not understand the reason
for this behavior, avoidance of the agglomeration phenom-
enon renders PMC-1 a suitable material for practical appli-
cations.
Remarkably, the inclusion single crystal PMC-2 was
obtained after exposure of PMC-1 to a CHCl3/N2 mixture
atmosphere for about an hour at room temperature.[12] The
unique CHCl3 induced response of PMC-1 prompted us to
elucidate structural transitions using single-crystal structure
analysis. Both PMC-1 and PMC-2 were characterized by
thermogravimetric analysis (TGA), X-ray diffraction (XRD),
and single-crystal X-ray diffraction (SC-XRD). TGA results
indicate that PMC-1 is a completely desolvated crystal, while
PMC-2 is solvated with two CHCl3 molecules per host
molecule (Supporting Information, Figure S3). According to
the XRD patterns, PMC-2 is able to recover to PMC-1 by
means of a thermal treatment (< 758C) or N2 sweeping
(Supporting Information, Figure S4). Moreover, a reversible
transition is confirmed by the single-crystal diffraction results
from desolvated PMC-2 (Supporting Information, Figure S5
and Table S3).
Single-crystal XRD patterns indicate that both PMC-
1 and PMC-2 are in the same Pbcn space group (crystal details
are summarized in the Supporting Information, Table S2).
Structural analysis of PMC-1 in Figure 2a revealed a number
of key features: 1) all of the Gemini molecules adopt a zigzag
conformation and their alkyl chains align adjacent to benzene
Figure 1. The adsorption–desorption isotherms of PMC-1 for various
VOCs at 298 K. a) Adsorption (closed symbols)–desorption (open
symbols) isotherms for VOCs. b) Adsorption–desorption cycles for
CHCl3. P0 for VOCs: chloroform 26.54 kPa, dichloromethane 53.33 kPa,
methanol 8.205 kPa, ethanol 16.98 kPa, ethyl acetate 13.33 kPa, ben-
zene 12.69 kPa.
Angew. Chem. Int. Ed. 2016, 55, 3378 –3381
ꢀ 2016 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.angewandte.org
3379

Angewandte
Communications
Chemie
À
rings to form a 1D column along the c axis as a result of C
H···p interactions (a 4.62 Š distance from the antepenulti-
mate carbon atom of terminal alkyl chains to the center of the
adjacent benzene ring; Supporting Information, Fig-
ure S6);[13] 2) two such columns stack face to face to form
a pair of columns connected by electrostatic attractions
between quaternary ammonium head groups and counter-
ions (Supporting Information, Figure S7), p–p stacking and
ing Information, Figure S9), rendering the PMC an imprac-
tical host for atoms or molecules. The framework cannot even
accommodate H+ ions. Unremarkable adsorption of VOC
vapors are a likely consequence of the space restrictions in the
crystal, except in the case of CHCl3.
In sharp contrast, PMC-2 is formed after exposure to
chloroform with a unit cell volume increase of about 15%,
from 4513 (PMC-1) to 5185 (PMC-2) Š3. The size of the unit
cell in b and c directions increases by 11.22% and 2.88%,
respectively, with near retention of length along the a axis
(Figure 2b). Overall, the assembly motif of PMC-2 is similar
to that of PMC-1. However, closer inspection of substructures
shows several points of difference. Firstly, the localized
amplified images reveal that the terminal alkyl chain
(brown) transforms from an all-trans conformation to parti-
ally gauche upon guest adsorption. Secondly, the inter-plane
distance between the two neighboring azo planes, which
belong to two adjacent column pairs, increased from 11.46 Š
to 13.38 Š. Thirdly, alkyl chains are pushed towards the rigid
À
C H···p interactions (a 3.88 Š distance from the middle of the
alkyl chain to the center of the benzene ring of the
neighboring layer; Supporting Information, Figure S8); and
3) column pairs align in a staggered lattice because of
electrostatic attractions in the aob plane, thereby forming
the PMC-1 host scaffold. The space-filling model of the
crystal is densely packed without substantial voids (Support-
À
spacer, compressing intermolecular C H···p interactions
from 4.62 Š to 3.71 Š (Supporting Information, Figure S6).
These transformations in structure lead to an expanded
scaffold and generate many larger transient pores for
accommodation of CHCl3.[14] Consequently, the adsorption–
desorption of guests endows PMC-1 with a “breathing”
characteristic, closely related to the expansion–shrinkage of
its scaffold (Figure 2c).
Breathing effects can be attributed to the unique molec-
ular structure of the Gemini molecule and its packing motif.
As expected, azobenzene spacers facilitate fabrication of
a rigid scaffold which is immovable in both PMC-1 and PMC-
2, while alkyl chains are the origin of structural transforma-
tion. When Gemini molecules assemble along one direction,
rigid spacers adopt the same orientation and alkyl chains
orient to one side. Thus, the resulting molecular column can
be regarded as a skeleton covered with a spongy cushion on
one side (represented by blue and green boxes in Figure 2a).
Columns of this type stack in pairs to form a column pair
À
because of electrostatic, p–p stacking and C H···p interac-
tions, thereby formulating the building unit (“brick”) of each
PMC. The skeleton moieties of these bricks are connected
through electrostatic interactions (Supporting Information,
Figure S10) and form a stable scaffold in which spongy
cushions (flexible alkyl chains) are filled. During the adsorp-
tion process, guest molecules ingress into the spongy moieties,
leading to a structural transformation. Adjacent bricks are
compressed and transient pores are formed to accommodate
guest molecules (Supporting Information, Figure S11). Upon
guest evacuation, the spongy cushion recovers to its original
state and the PMC converts into a dense staggered arrange-
ment, akin to a brick wall.
When guest molecules squeeze into voids, fully extended
alkyl chains are compelled to adopt a gauche conformation
because of their flexibility. As a result the scaffold expands
simultaneously. This thermodynamically unfavorable process
requires additional energy, which can only stem from affinity
between the guest molecule and the scaffold. In PMC-2, host–
guest affinity mainly comes from hydrogen bonding inter-
actions, which are summarized in Table S4 (Supporting
Figure 2. Stacking process of PMC-1 and structural variation before
and after guest encapsulation. a) Representation of the PMC-1 assem-
bly process. b) Unit cell expansion/shrinkage and alkyl transformation
during adsorption–desorption processes. C gray, N blue, Cl green,
Br olive. Hydrogen atoms are omitted for clarity. c) A representation of
transitions in framework topology between PMC-1 and PMC-2.
3380
www.angewandte.org
ꢀ 2016 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Angew. Chem. Int. Ed. 2016, 55, 3378 –3381

Angewandte
Communications
Chemie
[2] a) J. R. Holst, A. Trewin, A. I. Cooper, Nat. Chem. 2010, 2, 915 –
Information). The interaction between scaffold and CHCl3
has a large binding affinity energy of À14.8 kcalmolÀ1
according to DFT calculations[15] and non-covalent interac-
tion (NCI) analysis[16] (Supporting Information, Figure S12),
which facilitates the structural transition of alkyl chains and
volume expansion of the framework. For comparison, the
binding energy affinity of other VOC molecules for CHCl3
was also calculated (Supporting Information, Figure S13).
Calculations indicate that none of the other VOCs possess an
affinity for CHCl3 as great as that shown by PMC-1. Logically,
molecules larger in size than CHCl3 require a higher addi-
tional energy to induce a more expansive transition in the
PMC-1 structure. The affinity between bigger molecules and
the scaffold was not determined in this case, but it is
anticipated that variations in binding affinity of different
host–guest complexes will signal propensities for guest
sorption.
In summary, we have confirmed that flexible components
contribute to enhance the selectivity of porous molecular
crystals. More importantly, we have described the first
example of an alkyl decorated PMC for specific adsorption
of VOCs. The fast adsorption rate and low-pressure satura-
tion of PMC-1 lends the material to gas sensor applications
for specific detection of chloroform. Our findings indicate
that introduction of flexible components may facilitate the
design of new functional PMCs.
920; b) P. Li, Y. He, Y. Zhao, L. Weng, H. Wang, R. Krishna, H.
Wu, W. Zhou, M. OꢀKeeffe, Y. Han, B. Chen, Angew. Chem. Int.
Ed. 2015, 54, 574 – 577; Angew. Chem. 2015, 127, 584 – 587; c) P.
Li, Y. He, J. Guang, L. Weng, J. C.-G. Zhao, S. Xiang, B. Chen, J.
Am. Chem. Soc. 2014, 136, 547 – 549; d) J. Lü, C. Perez-Krap, M.
Suyetin, N. H. Alsmail, Y. Yan, S. Yang, W. Lewis, E. Bichout-
skaia, C. C. Tang, A. J. Blake, R. Cao, M. Schrçder, J. Am. Chem.
Soc. 2014, 136, 12828 – 12831.
[3] a) N. B. McKeown, J. Mater. Chem. 2010, 20, 10588 – 10597; b) O.
Buyukcakir, Y. Seo, A. Coskun, Chem. Mater. 2015, 27, 4149 –
4155.
[4] M. Mastalerz, Chem. Eur. J. 2012, 18, 10082 – 10091.
[5] S.-i. Noro, D. Tanaka, H. Sakamoto, S. Shimomura, S. Kitagawa,
S. Takeda, K. Uemura, H. Kita, T. Akutagawa, T. Nakamura,
Chem. Mater. 2009, 21, 3346 – 3355.
[6] B. Chen, C. Liang, J. Yang, D. S. Contreras, Y. L. Clancy, E. B.
Lobkovsky, O. M. Yaghi, S. Dai, Angew. Chem. Int. Ed. 2006, 45,
1390 – 1393; Angew. Chem. 2006, 118, 1418 – 1421.
[7] a) I. M. Hauptvogel, R. Biedermann, N. Klein, I. Senkovska, A.
Cadiau, D. Wallacher, R. Feyerherm, S. Kaskel, Inorg. Chem.
2011, 50, 8367 – 8374; b) N. Nijem, H. Wu, P. Canepa, A. Marti,
K. J. Balkus, T. Thonhauser, J. Li, Y. J. Chabal, J. Am. Chem. Soc.
2012, 134, 15201 – 15204; c) C. Wang, L. Li, J. G. Bell, X. Lv, S.
Tang, X. Zhao, K. M. Thomas, Chem. Mater. 2015, 27, 1502 –
1516.
[8] a) Y. Sheng, J. Yao, Q. Chen, H. Liu, CrystEngComm 2015, 17,
1439 – 1447; b) Q. Chen, J. Yao, X. Hu, J. Shen, Y. Sheng, H. Liu,
J. Appl. Crystallogr. 2015, 48, 728 – 735.
˘
[9] H. J. Choi, M. Dinca, J. R. Long, J. Am. Chem. Soc. 2008, 130,
Supporting information for this article may be accessed
using the link at the end of the document. CCDC 1437194 and
1437195 contain the supplementary crystallographic data for
this paper. Data can be obtained free of charge from The
Cambridge Crystallographic Data Centre.
7848 – 7850.
[10] a) Y.-C. Chiang, P.-C. Chiang, C.-P. Huang, Carbon 2001, 39,
523 – 534; b) E. Ahmed, J. Khanderi, D. H. Anjum, A. Roth-
enberger, Chem. Mater. 2014, 26, 6454 – 6460; c) S. Van der
Perre, T. Van Assche, B. Bozbiyik, J. Lannoeye, D. E. De Vos,
G. V. Baron, J. F. M. Denayer, Langmuir 2014, 30, 8416 – 8424;
d) H. Yamagiwa, S. Sato, T. Fukawa, T. Ikehara, R. Maeda, T.
Mihara, M. Kimura, Sci. Rep. 2014, 4, 6247; e) K. Yang, Q. Sun,
F. Xue, D. Lin, J. Hazard. Mater. 2011, 195, 124 – 131.
[11] a) M. Eddaoudi, H. Li, O. M. Yaghi, J. Am. Chem. Soc. 2000, 122,
1391 – 1397; b) F. Yang, Q.-K. Liu, J.-P. Ma, Y.-A. Li, K.-X.
Wang, Y.-B. Dong, CrystEngComm 2015, 17, 4102 – 4109.
[12] It should be noted that PMC-2 shows localized uncertainty at the
tip of the alkyl chains. This is because the transformation of
flexible alkyl chain may undergo multiple metastable states with
tiny difference. This situation also occured when we soaked
PMC-1 in a saturated chloroform solution. Localized disorder
does not affect our conlusions because we discuss only one
position throughout the manuscript.
Acknowledgements
Y.J.S., Q.B.C., J.Y.Y. , Y.X.L., and H.L.L. thank the National
Natural Science Foundation of China (No.21273074,
91334203, 21576079), the 111 Project of China (No.B08021)
and the Fundamental Research Funds for the Central
Universities of China. SD was sponsored by the Division of
Chemical Sciences, Geosciences, and Biosciences, Office of
Basic Energy Sciences, U.S. Department of Energy.
[13] M. J. Plevin, D. L. Bryce, J. Boisbouvier, Nat. Chem. 2010, 2,
466 – 471.
Keywords: chloroform · host–guest systems ·
porous molecular crystals · preferential adsorption ·
self-assembly
[14] a) S. A. Herbert, A. Janiak, P. K. Thallapally, J. L. Atwood, L. J.
Barbour, Chem. Commun. 2014, 50, 15509 – 15512; b) L. J.
Barbour, Chem. Commun. 2006, 1163 – 1168; c) A. Kondo, N.
Kojima, H. Kajiro, H. Noguchi, Y. Hattori, F. Okino, K. Maeda,
T. Ohba, K. Kaneko, H. Kanoh, J. Phys. Chem. C 2012, 116,
4157 – 4162.
How to cite: Angew. Chem. Int. Ed. 2016, 55, 3378–3381
Angew. Chem. 2016, 128, 3439–3442
[15] Y. Liu, J. Zhao, F. Li, Z. Chen, J. Comput. Chem. 2013, 34, 121 –
131.
[16] E. R. Johnson, S. Keinan, P. Mori-Sµnchez, J. Contreras-García,
A. J. Cohen, W. Yang, J. Am. Chem. Soc. 2010, 132, 6498 – 6506.
[1] a) S. Horike, S. Shimomura, S. Kitagawa, Nat. Chem. 2009, 1,
695 – 704; b) L. Chen, P. S. Reiss, S. Y. Chong, D. Holden, K. E.
Jelfs, T. Hasell, M. A. Little, A. Kewley, M. E. Briggs, A.
Stephenson, K. M. Thomas, J. A. Armstrong, J. Bell, J. Busto, R.
Noel, J. Liu, D. M. Strachan, P. K. Thallapally, A. I. Cooper, Nat.
Mater. 2014, 13, 954 – 960; c) M. Zhao, S. Ou, C.-D. Wu, Acc.
Chem. Res. 2014, 47, 1199 – 1207; d) R. E. Morris, P. S. Wheatley,
Angew. Chem. Int. Ed. 2008, 47, 4966 – 4981; Angew. Chem. 2008,
120, 5044 – 5047.
Received: November 16, 2015
Revised: December 23, 2015
Published online: February 2, 2016
Angew. Chem. Int. Ed. 2016, 55, 3378 –3381
ꢀ 2016 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.angewandte.org
3381