Reversible alteration of CO2 adsorption upon photochemical or thermal treatment in a metal-organic framework

Article abstract of DOI:10.1021/ja209197f

A metal-organic framework (MOF) for reversible alteration of guest molecule adsorption, here carbon dioxide, upon photochemical or thermal treatment has been discovered. An azobenzene functional group, which can switch its conformation upon light irradiat

Full text of DOI:10.1021/ja209197f

Communication
pubs.acs.org/JACS
Reversible Alteration of CO2 Adsorption upon Photochemical or
Thermal Treatment in a Metal−Organic Framework
Jinhee Park, Daqiang Yuan, Khanh T. Pham, Jian-Rong Li, Andrey Yakovenko, and Hong-Cai Zhou*
Department of Chemistry, Texas A&M University, College Station, Texas 77842, United States
S
* Supporting Information
negligible N2 uptake. Lowering regeneration cost thus becomes
a focal point for MOF application in carbon capture.
ABSTRACT: A metal−organic framework (MOF) for
reversible alteration of guest molecule adsorption, here
carbon dioxide, upon photochemical or thermal treatment
has been discovered. An azobenzene functional group,
which can switch its conformation upon light irradiation or
heat treatment, has been introduced to the organic linker
of a MOF. The resulting MOF adsorbs different amount of
CO2 after UV or heat treatment. This remarkable stimuli-
responsive adsorption effect has been demonstrated
through experiments.
Our strategy to introduce an optically or thermally switchable
functional group into the internal surface of an overall rigid
MOF is the synthesis of the MOF with an organic linker having
azobenzene as a dangling group. Since CO2 adsorption on the
MOF is relatively high at room temperature, it has been chosen
to demonstrate the stimuli-responsive effect. As an exper-
imental proof-of-concept, CO2 adsorption has been measured
to study the effect of the trans−cis isomerization of the ligand
on the gas adsorption amount of the MOF. In the MOF, the
selectivity and regeneration efficiency of the sorbent can be
tuned upon light irradiation or with thermal treatment. To the
best of our knowledge, reversible alteration of guest molecule
adsorption triggered by photochemically induced isomerization
of ligands in a MOF has not been reported until this work.
During the preparation of this report, we noticed that a MOF
with an azobenzene moiety was made, albeit no reversible
alteration of guest molecule adsorption was disclosed.26 Herein,
we report the synthesis, structural and spectroscopic character-
ization, and gas adsorption studies on a photoswitchable MOF,
which exhibits a remarkable stimuli-responsive adsorption
effect.
In this work, a photoswitchable MOF has been synthesized
using a ligand, 2-(phenyldiazenyl)terephthalate. The acid form
of the ligand can undergo isomerization photochemically from
trans to cis and thermally from cis to trans, as shown in Figure 1.
An adsorption spectrum of this ligand in its acid form (the
ligand precursor) in N,N-diethylformamide (DEF) confirmed
the reversible isomerization. A fresh sample of the ligand
precursor, in primarily trans form, showed a typical π-to-π*
band around 320 nm and an n-to-π* absorption peak at 430
nm. UV (302 or 365 nm) irradiation decreased the intensity of
the 320 nm band and slightly increased that of the 430 nm
band gradually, indicating the trans-to-cis transformation. When
the ligand precursor solution was placed at room temperature
for a day or at 60 °C for 30 min in the dark, the absorption
intensity was fully recovered, indicating a complete recovery of
the trans form (Figures S1 and S2). The results of the
absorption spectroscopic studies corroborate well with those of
the NMR spectroscopic studies on the ligand precursor.
Because the trans-to-cis conformational change of azobenzene
is relatively slow on the NMR time scale, this isomerization of
the ligand has also been monitored by NMR studies (Figures
S4 and S5). A MOF, designated as PCN-123 (PCN represents
unctional materials structurally switchable by external
F
stimuli such as heat,1,2 redox potential,3,4 pH,5,6 or
light7−10 have captured the imagination of those interested in
stimuli-responsive functional materials. For example, phase
transition via annealing has been reported as a strategy to
obtain robust and thermally stable porous networks.11,12
Among these stimuli, light can be efficiently applied without
the generation of any byproducts and is abundantly available in
the form of sunlight. Optical memory effects with coordination
compounds of the d10 monovalent ions of group 11, which can
be reversed by gentle heating, have been studied for potential
data storage application.13,14 One of the most popular light-
responsive organic groups applied is azobenzene. Triggered by
UV irradiation or heat, an azobenzene group can undergo trans-
to-cis isomerization reversibly.15 This isomerization involves a
distance change between the para-carbon atoms in an
azobenzene molecule from 9 Å in the trans isomer to 5.5 Å
in the cis one.16 Based on this conformational change, devices
constructed with azobenzenes can conduct photomechanical
tasks on a molecular level upon light irradiation or with thermal
treatment.17,18
Metal−organic frameworks (MOFs) have attracted enor-
mous research interest due to their “designable” potential.19,20
The combination of metal or metal-cluster secondary building
units (SBUs) of well-known geometry and ligands prepared
with the aid of organic synthesis leads to almost endless
possibilities. Based on the specific needs of an application, the
overall structure, porosity, and surface functional groups of a
MOF can be designed, synthesized, and tuned.21−23 MOFs
have been intensively explored for potential applications in gas
storage and separation. In particular, they have been considered
as sorbents for carbon capture.24,25 For a sorbent competent in
carbon capture, high selectivity of CO2 over N2 and
energetically efficient sorbent regeneration are two prereq-
uisites. In general, at room temperature and above, MOFs have
Received: October 3, 2011
Published: December 13, 2011
© 2011 American Chemical Society
99
dx.doi.org/10.1021/ja209197f | J. Am. Chem.Soc. 2012, 134, 99−102

Journal of the American Chemical Society
Communication
Figure 1. (Top) Trans-to-cis isomerization of the ligand of PCN-123
(PCN represents porous coordination networks) induced by UV
irradiation and the cis-to-trans isomerization induced by heat
treatment. (Bottom) Schematic illustration showing the suggested
CO2 uptake in MOF-5, PCN-123 trans, and PCN-123 cis.
Figure 2. CO2 adsorption isotherms (at 295 K) of PCN-123 showing
reversible conformational change: pristine sample (red), right after the
first UV irradiation (half-filled blue), 5 h after the first UV irradiation
(fully filled blue), after the first regeneration process at 60 °C for 20 h
in the dark (orange), after the second UV irradiation (green), and after
the second regeneration process at 60 °C for 20 h in the dark (yellow).
In the figure, “UV” represents “UV irradiation”.
porous coordination networks) for the convenience of
discussion, has been synthesized via a solvothermal reaction
between the ligand precursor and Zn(NO3)2·6H2O in DEF.
The color of the resulting crystals, red-orange, indicates the
presence of azobenzene functional groups in PCN-123 (Figure
S3). Because X-ray diffraction analysis cannot distinguish the
difference between randomly oriented azobenzene groups and
free solvates inside the pores, and four-fold positional disorder
is imposed by the space group, crystals of PCN-123 exhibit
pseudo high symmetry; thus, the azobenzene groups cannot be
located in the electron density map derived from even the
single-crystal X-ray diffraction data.27,28 The powder X-ray
diffraction (PXRD) pattern confirmed PCN-123 is isostructural
with MOF-5 in bulk. When a sample of PCN-123 was digested
in an aliquot of 6 M HCl, the acid form of the ligand was
recovered quantitatively, the NMR spectrum of which was
identical to that of the original ligand precursor (Figure S3).29
This confirms the presence of the azobenzene groups in PCN-
123 crystals and excludes any significant ligand decomposition
during the solvothermal reaction procedure. The ligand from
PCN-123 digested by DCl/DMSO-d6 solution can also
undergo trans−cis isomerization. Newly appearing peaks from
cis isomers matched well with the ones from the UV-light-
exposed ligand precursor solution (Figure S5).
Each cubic cavity in PCN-123 is occupied by azobenzene
groups, and there is an enormous amount of azobenzene units
in PCN-123. This leads to optically induced changes in gas
uptake, thanks to the structural rearrangement during the
trans−cis isomerization. Although the crystals are synthesized
with zinc ions (d10), it is still difficult for photons to penetrate
into the core of the crystals because of scattering from defects
and absorption from competing chromophores. Therefore, to
promote isomerization in crystals, UV light was irradiated on
the crystals for 1−2 h in this experiment.
which can be attributed to the steric hindrance among the
azobenzene groups around each cubic cavity. There may exist
some synergy among the conformational changes of the
azobenzene groups. The azobenzene groups in the pristine
PCN-123 sample are in predominantly trans conformation, a
relatively stable form. Right after UV irradiation, only some of
the trans azobenzene groups become cis, and then more trans
forms become cis triggered by the initial conformational
changes, leading to significant decreases in gas uptake.
At this stage, the possibility of a partial structural collapse of
PCN-123 during and after UV irradiation cannot be completely
ruled out. Gas adsorption measurements at different temper-
atures were performed to check whether the porosity of PCN-
123 was maintained if the sample was allowed to return to its
original state after UV irradiation. First, the CO2 adsorptions
were measured at 0 °C and then at room temperature before
UV irradiation. The CO2 uptake drops as temperature
increases, which can be ascribed to the common thermody-
namic effect as the thermal motion overcomes the CO2−
framework interaction. After the sample was exposed to UV
light for 1 h, the CO2 uptake was measured again at room
temperature to test the trans-to-cis conversion. As expected, a
noticeable drop in gas uptake was observed. Subsequently, the
sample was allowed to return to its trans form (heating, visible
light irradiation, or allow it to stay in the absence of UV light
for a prolonged period of time), and the CO2 uptake was
measured at 0 °C once again. Although the two measurements
for the trans form were performed sequentially using the same
sample, the gas uptakes at 0 °C before UV irradiation and after
returning to its original form following UV treatment were
almost identical (Figure 3), demonstrating that the framework
of PCN-123 is stable under the conformational change of the
azobenzene group. The hypothesis of a partial decomposition
of the MOF after UV irradiation can thus be ruled out.
Before UV exposure, the CO2 uptake of a pristine sample of
PCN-123 at room temperature after activation was 22.9 cm3/g
at 1 bar, and the uptake decreased to 16.8 cm3/g after 1 h of
UV irradiation, corresponding to a 26.6% decrease in CO2
uptake. When CO2 adsorption was measured again at the same
temperature, the uptake decreased further to 10.5 cm3/g
(Figure 2), corresponding to an overall decrease of 53.9% in
CO2 uptake when compared to the pristine sample. This delay
of change in gas adsorption isotherms upon UV irradiation can
be ascribed to the slow isomerization of the ligand in PCN-123,
To unambiguously show alteration of guest molecule
adsorption caused by cis-to-trans isomerization, the sample
tube was wrapped with aluminum foil to prevent visible light
exposure and kept under 60 °C for 20 h. The uptake at 1 bar
100
dx.doi.org/10.1021/ja209197f | J. Am. Chem.Soc. 2012, 134, 99−102

Journal of the American Chemical Society
Communication
proximity of the metal clusters, the main adsorption sites for
CO2 molecules.30,31
To evaluate the change of gas-accessible space caused by
isomerization, the pore size distributions were estimated from
N2 adsorption isotherms collected at 77 K. For the fresh PCN-
123 sample, the pore size distribution peaks at 10.0 and 11.8 Å
were almost the same height. After UV irradiation, while the
peak for the cis isomer at 10.0 Å decreased significantly, the one
at 11.0 Å increased. After heat treatment, the pore size
distribution curve of the regenerated PCN-123 was almost
identical to that of the fresh sample (Figure 4). This
Figure 3. Demonstration of the maintenance of porosity throughout
the experiments. CO2 adsorption isotherms of a pristine sample
(before UV exposure) at 273 K (red squares) and at 295 K (red
circles) and the isotherms at 295 K after UV exposure (blue circles)
and at 273 K after the sample was allowed to return to predominantly
trans form (blue squares). Open circles or squares represent
desorption, and filled ones correspond to adsorption.
was 26.4 cm3/g, 13.3% higher than the uptake of the fresh
sample (Figure 2). There are two possible explanations for this
surprising increase. Most likely, the fresh sample may contain
mostly but not exclusively trans forms. Some cis forms may have
been generated during synthesis and sample handling due to
unintentional light exposure. After regeneration of the sample
under heating in the dark, however, these residual cis forms may
change to trans as well. Another plausible reason is that heating
may help the further removal of solvates from PCN-123,
leading to increased gas-accessible volume. Neither of these two
explanations can be ruled out completely at this stage.
To test the first hypothesis, gas adsorption studies of a newly
prepared sample from a different batch have been performed.
In particular, the sample was synthesized in the dark to ensure
the formation of trans products exclusively. When this sample
was regenerated by heating after UV irradiation, its gas uptake
did not exceed that of the fresh sample (Figure S19), indicating
that the sample prepared in the dark contains only trans
isomers. The second hypothesis can thus be excluded.
The second cycle of UV irradiation on the first sample for 1.5
h decreased the gas uptake to 14.2 cm3/g, 46.2% less than that
of the regenerated sample. The second regeneration process
under the same conditions increased the gas uptake to 22.4
cm3/g, comparable to that of the fresh sample (Figure 2).
These results show that PCN-123 can be used as a smart
sorbent for reversible CO2 adsorption.
The decrease of gas uptake upon trans-to-cis transformation
after UV irradiation can be ascribed to the positional change of
the dangling benzene group in PCN-123. The PXRD peaks of
PCN-123 did not shift before and after UV irradiation.
However, the ratios of the diffraction intensities changed after
UV irradiation. This intensity change motivates us to generate
the structure envelope of the PCN-123 from PXRD data. The
electron density of the UV-irradiated PCN-123 (cis isomer)
around the Zn4O cluster increased compared with that of the
heat-treated one (trans isomer) (Figure S21). This means CO2
molecules are relatively difficult to approach and occupy the
Figure 4. Pore size distribution of PCN-123 initially (black), after UV
exposure (red), and after heating (blue). The changes upon UV
irradiation and thermal treatment show reversibility of isomerization in
PCN-123.
reversibility demonstrates that the pore size distribution change
was driven by the isomerization of the ligands in PCN-123.
In conclusion, reversible alteration of CO2 capture upon
photochemical or thermal treatment in a MOF has been
demonstrated experimentally. The freshly made PCN-123
adsorbs a significant amount of CO2 but a negligible amount
of N2. Upon light irradiation, the total uptake of CO2 decreased
readily due to the change of conformation of the azobenzene
groups inside the pores of the MOF. The adsorbent returns to
its original state when allowed to stay at ambient conditions for
a prolonged period of time or with gentle heating if fast
regeneration is needed. Potentially, systems such as sorbent
regeneration or drug delivery which require efficient release of
guest molecules can be designed on the basis of the findings in
this report.
ASSOCIATED CONTENT
■
S
* Supporting Information
Experimental procedures for the syntheses of the ligand, PXRD
and TGA data of PCN-123, isomerization experiments, and
spectral characterization data. This material is available free of
charge via the Internet at http://pubs.acs.org.
AUTHOR INFORMATION
■
Corresponding Author
zhou@chem.tamu.edu
101
dx.doi.org/10.1021/ja209197f | J. Am. Chem.Soc. 2012, 134, 99−102

Journal of the American Chemical Society
Communication
(31) Karra, J. R.; Walton, K. S. J. Phys. Chem. C 2010, 114, 15735−
15740.
ACKNOWLEDGMENTS
■
This work was supported by the U.S. Department of Energy
(DOE DE-SC0001015 and DE-AR0000073), the National
Science Foundation (NSF CBET-0930079), and the Welch
Foundation (A-1725).
REFERENCES
■
(1) Ma, S. Q.; Sun, D. F.; Yuan, D. Q.; Wang, X. S.; Zhou, H. C. J.
Am. Chem. Soc. 2009, 131, 6445−6451.
(2) Zhao, D.; Yuan, D. Q.; Krishna, R.; van Baten, J. M.; Zhou, H. C.
Chem. Commun. 2010, 46, 7352−7354.
(3) Nguyen, T. D.; Liu, Y.; Saha, S.; Leung, K. C. F.; Stoddart, J. F.;
Zink, J. I. J. Am. Chem. Soc. 2007, 129, 626−634.
(4) Nguyen, T. D.; Tseng, H. R.; Celestre, P. C.; Flood, A. H.; Liu,
Y.; Stoddart, J. F.; Zink, J. I. Proc. Natl. Acad. Sci. U.S.A. 2005, 102,
10029−10034.
(5) Park, C.; Oh, K.; Lee, S. C.; Kim, C. Angew. Chem., Int. Ed. 2007,
46, 1455−1457.
(6) Angelos, S.; Yang, Y. W.; Khashab, N. M.; Stoddart, J. F.; Zink, J.
I. J. Am. Chem. Soc. 2009, 131, 11344−11346.
(7) Boyer, J. C.; Carling, C. J.; Gates, B. D.; Branda, N. R. J. Am.
Chem. Soc. 2010, 132, 15766−15772.
(8) Yagai, S.; Kitamura, A. Chem. Soc. Rev. 2008, 37, 1520−1529.
(9) Ferris, D. P.; Zhao, Y. L.; Khashab, N. M.; Khatib, H. A.;
Stoddart, J. F.; Zink, J. I. J. Am. Chem. Soc. 2009, 131, 1686−1688.
(10) Zhu, Q.-L.; Sheng, T.-L.; Fu, R.-B.; Hu, S.-M.; Chen, L.; Shen,
C.-J.; Ma, X.; Wu, X.-T. Chem.Eur. J. 2011, 17, 3358−3362.
(11) Yang, C.; Kaipa, U.; Zhang Mather, Q.; Wang, X.; Nesterov, V.;
Venero, A. F.; Omary, M. A. J. Am. Chem. Soc. 2011, 133, 18094−
18097.
(12) Ohara, K.; Martı-Rujas, J.; Haneda, T.; Kawano, M.; Hashizume,
D.; Izumi, F.; Fujita, M. J. Am. Chem. Soc. 2009, 131, 3860−3861.
(13) Omary, M. A.; Colis, J. C. F.; Larochelle, C. L.; Patterson, H. H.
Inorg. Chem. 2007, 46, 3798−3800.
(14) Hollingsworth, G.; Barrie, J. D.; Dunn, B.; Zink, J. I. J. Am.
Chem. Soc. 1988, 110, 6569−6570.
(15) Dou, Y. S.; Hu, Y.; Yuan, S. A.; Wu, W. F.; Tang, H. Mol. Phys.
2009, 107, 181−190.
(16) Liu, N. G.; Chen, Z.; Dunphy, D. R.; Jiang, Y. B.; Assink, R. A.;
Brinker, C. J. Angew. Chem., Int. Ed. 2003, 42, 1731−1734.
(17) Koshima, H.; Ojima, N.; Uchimoto, H. J. Am. Chem. Soc. 2009,
131, 6890−6891.
(18) Muraoka, T.; Kinbara, K.; Aida, T. Nature 2006, 440, 512−515.
(19) Kitagawa, S.; Kitaura, R.; Noro, S. Angew. Chem., Int. Ed. 2004,
43, 2334−2375.
(20) Zhao, D.; Timmons, D. J.; Yuan, D.; Zhou, H.-C. Acc. Chem. Res.
2010, 44, 123−133.
(21) Wang, C.; Lin, W. J. Am. Chem. Soc. 2011, 133, 4232−4235.
(22) Pramanik, S.; Zheng, C.; Zhang, X.; Emge, T. J.; Li, J. J. Am.
Chem. Soc. 2011, 133, 4153−4155.
(23) Lu, Z.-Z.; Zhang, R.; Li, Y.-Z.; Guo, Z.-J.; Zheng, H.-G. J. Am.
Chem. Soc. 2011, 133, 4172−4174.
(24) Li, J.-R.; Kuppler, R. J.; Zhou, H.-C. Chem. Soc. Rev. 2009, 38,
1477−1504.
(25) Li, J.-R.; Ma, Y.; McCarthy, M. C.; Sculley, J.; Yu, J.; Jeong, H.-
K.; Balbuena, P. B.; Zhou, H.-C. Coord. Chem. Rev. 2011, 255, 1791−
1823.
(26) Modrow, A.; Zargarani, D.; Herges, R.; Stock, N. Dalton Trans.
2011, 40, 4217−4222.
(27) Dugan, E.; Wang, Z.; Okamura, M.; Medina, A.; Cohen, S. M.
Chem. Commun. 2008, 3366−3368.
(28) Wang, Z.; Cohen, S. M. Angew. Chem., Int. Ed. 2008, 47, 4699−
4702.
(29) Deng, H.; Doonan, C. J.; Furukawa, H.; Ferreira, R. B.; Towne,
J.; Knobler, C. B.; Wang, B.; Yaghi, O. M. Science 2010, 327, 846−850.
(30) Pianwanit, A.; Kritayakornupong, C.; Vongachariya, A.;
Selphusit, N.; Ploymeerusmee, T.; Remsungnen, T.; Nuntasri, D.;
Fritzsche, S.; Hannongbua, S. Chem. Phys. 2008, 349, 77−82.
102
dx.doi.org/10.1021/ja209197f | J. Am. Chem.Soc. 2012, 134, 99−102