Porous coordination-polymer crystals with gated channels specific for supercritical gases

Article abstract of DOI:10.1002/anie.200390130

Flexible and dynamic coordination polymers based on interpenetration and interdigitation have been synthesized and characterized. Reversible crystal-to-crystal transformation from a nonporous structure (24 m²g^-1) to a microporous structure (at least 320 m²g^-1) is triggered by adsorption of various supercritical gases (see picture).

Full text of DOI:10.1002/anie.200390130

Communications
Flexible Coordination Polymers
Porous Coordination-Polymer Crystals with
Gated Channels Specific for Supercritical Gases**
Ryo Kitaura, Kenji Seki, George Akiyama, and
Susumu Kitagawa*
Considerable effort has been devoted to the synthesis and
characterization of new crystalline nanosized porous materi-
als, such as coordination polymers and inorganic zeolites,
because of their versatile applicability to gas storage, molec-
ular sieves, size- or shape-selective catalysis, and ion ex-
7]
change.[1 For successful performance of these functions,
robustness and stability of the porous frameworks are
essential. In contrast, flexible host frameworks that respond
to guest molecules, for example, induced-fit recognition
between a protein and its substrate, have high selectivity for
guest inclusion. Coordination polymers, in spite of their
crystalline form, are suitable for creating flexible and dynamic
porous frameworks, so-called third-generation compounds,[8]
because the wide variety of architectures and topologies are
based not only on coordinative bonds, but also on hydrogen
Figure 1. Schematic representation of dynamic frameworks. Above: In-
terdigitated framework of layers whose gap depends on the guest mol-
ecules. Below: Interpenetratedframework whose interunit space de-
pends on the guest molecules.
[Cu(dhbc)2(4,4’-bpy)]¥H2O (1a), was prepared from CuII
nitrate, 2,5-dihydroxybenzoic acid (Hdhbc), and 4,4’-bipyr-
idine (4,4’-bpy). The copper ions are connected by 4,4’-bpy to
produce straight chains, and linked by dhbc to give a 2D sheet
motif (Figure 2a and b, respectively). The motif has inter-
locking ridges and hollows constructed by the dhbc benzene
planes in an upright fashion. The distance of 3.443 ä between
the planes of the nearest-neighbor dhbc ligands indicates the
presence of p p stacking interactions, and the motifs are
mutually interdigitated to create 1D channels along the a axis
with a cross section of 3.6 î 4.2 ä (Figure 2c), in which one
water molecule is included per copper ion. The type B
coordination polymer 2 was obtained by using CuII, ben-
zene-1,4-dicarboxylate (bdc), and 4,4’-bpy. Two-dimensional
grid-type sheets composed of CuII and bdc are linked by 4,4’-
bpy to produce a 3D jungle-gym-like framework, similar to
the structure of Prussian Blue. Two independent 3D struc-
tures interpenetrate each other to form a 3D framework with
cylindrical channels with approximate dimensions of 3.4 î
3.4 ä along the c axis (Figure 3).[14]
ꢀ
bonds and/or p p stacking, which are weaker than the Si O
[9 13]
ꢀ
and Al O bonds in zeolites.
Hence, we focused on the
preparation of two new types of flexible and dynamic
nanoporous frameworks created by coordination polymers
with interdigitation and interpenetration, and examined the
gas-adsorption properties (CO2, CH4, O2, and N2) at high
pressure and ambient temperature.
Our strategy for functional coordination polymers that
respond to guest molecules is to assemble rigid motifs with a
degree of freedom in displacement to provide two types of
integrated frameworks (Figure 1): framework A with mutu-
ally interdigitated two-dimensional (2D) motifs, and frame-
work B with mutually interpenetrated three-dimensional
(3D) motifs. Both the rigidity of a structural motif and a
degree of freedom in displacement could give rise to
flexibility of the whole framework and thus improve porous
properties. A coordination polymer of type A, namely,
Thermogravimetric analysis of 1a shows the release of
water molecules of crystallization with increasing temper-
ature up to 1208C to give guest-free anhydrous 1b. No further
weight loss was observed between 1208C and 1708C, that is,
the guest-free phase is stable. Since the X-ray powder
diffraction (XRPD) pattern of 1b (Figure 4) shows sharp
diffraction peaks similar to those of 1a, the porous framework
is largely maintained, even without water molecules. Detailed
comparison of the two XRPD patterns reveals that only small
shifts were observed for the peaks (010) and (100), which are
attributed to the ac and bc planes, respectively. Therefore, the
2D layer motif is retained during dehydration. On the other
hand, peaks corresponding to the relative position of the layer
motifs show an apparent shift to the higher angle region
between 15 and 208. This result clearly indicates shrinking of
the layer gap along the c axis, which is attributable to a gliding
motion of the two p-stacked rings, which decreases the
channel cross section.
[*] Prof. Dr. S. Kitagawa, R. Kitaura, G. Akiyama
Department of Synthetic Chemistry andBiological Chemistry
Graduate School of Engineering
Kyoto University
Yoshida, Sakyo-ku, Kyoto 606-8501 (Japan)
Fax: (þ81)75-753-4979
E-mail: kitagawa@sbchem.kyoto-u.ac.jp
K. Seki
Energy Conversion andStorage Technology
AppliedResearch Centre
Research andDevelopment Department
Osaka Gas Co., Ltd.
6-19-9, Torishima, Konohana-ku, Osaka 554-0051 (Japan)
[**] This work was supportedby Grant-in-Aidfor Creative Scientific
Research (No. 13GS0024), Ministry of Education, Culture, Sports,
Science, andTechnology, Japan.
Supporting information for this article is available on the WWW
under http://www.angewandte.org or from the author.
428
¹ 2003 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
1433-7851/03/4204-0428 $ 20.00+.50/0
Angew. Chem. Int. Ed. 2003, 42, No. 4

Angewandte
Chemie
Figure 2. Crystal structure of 1a. a) 1D chain structure constructedby Cu II, 4,4’-bpy, anddhbc along the b axis (Cu, green; O, red; C, gray; N,
blue). b) 2D layer structure along the ab plane. c) 3D p-stackedpillaredlayer structure of 1a. Hydrogen atoms and water molecules are omitted
for clarity.
Compound 1b shows a type II nitrogen adsorption
isotherm at 77 K, and the calculated BET specific surface
area is 24 m2 gꢀ1 (see Supporting Information). This low
specific surface area leads to the conclusion that nitrogen
molecules (molecular size 3.3 ä) could not diffuse into the
micropores at all. If the original cross section of 1a were
retained, the nitrogen molecules would readily diffuse into
the channels. This also indicates pore contraction on dehy-
dration, as was demonstrated by XRPD. Figure 5a shows the
N2 adsorption isotherm of 1b in the pressure range between 1
and 120 atm at 298 K. Unlike the adsorption profile at 77 K,
the isotherm shows a profile unlike the usual adsorption
patterns–a flat curve indicative of no adsorption in the lower
pressure region, with an abrupt increase at a pressure of
50 atm (point a), as if channels large enough for the molecule
already existed. Langmuir plots are almost linear in the higher
pressure region and give a specific surface area of 320 m2 gꢀ1.
This high specific surface area clearly shows that nonporous
phase 1b was transformed into microporous phase 1a. This
onset pressure at which the channels of 1b change into open
gates is referred to as the gate-opening pressure.[15] Interest-
ingly, the desorption isotherm shows an abrupt decrease at
30 atm (gate-closing pressure, point b), indicative of hyste-
resis in the adsorption and desorption isotherms. This
hysteretic profile could be repeated many times. This
hysteretic sorption behavior is well reproduced by the sum
of the isotherms for nonporous phase 1b and porous phase
1a. Red and blue dashed lines in Figure 5a represent the
estimated adsorption isotherms for nonporous phase and
microporous phase, respectively. Below the gate-opening
pressure, the adsorption isotherm coincides with the red line
for the nonporous phase. Above the gate-opening pressure,
the adsorption isotherm starts to approach the blue line
(porous phase), and they coincide above 100 atm. At the gate-
opening pressure, nitrogen molecules result in transformation
Figure 3. Space-filling model of the crystal structure of 2. The structure
model was optimized by molecular mechanics (MM) with the program
package Cerius2.
Angew. Chem. Int. Ed. 2003, 42, No. 4
¹ 2003 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
1433-7851/03/4204-0429 $ 20.00+.50/0
429

Communications
was confirmed by an XRPD measurement on 1b containing
adsorbed methanol (molecular size 4.2 ä; see Supporting
Information). To the best of our knowledge, this is the first
example of hysteretic sorption behavior of supercritical
nitrogen gas at ambient temperature under high pressure.
Structural transformation triggered by a supercritical gas,
which only exhibits weak dispersive interactions in this
system, is facilitated by the unusual framework flexibility
provided by displacement of p p stacked moieties. Figure 5b
shows the adsorption isotherms for various gases on 1b in the
pressure range between 1 and 100 atm at 298 K. All of the
isotherm profiles are essentially similar to that for N2 at
298 K, but a clear difference is observed for the gate-opening
and gate-closing pressures. The gate-opening pressures for
nitrogen, oxygen, and carbon dioxide are 50, 35, and 0.4 atm,
and the gate-closing pressures 30, 25, and 0.2 atm, respec-
tively. This difference should be attributable to differences in
the intermolecular interaction force of the guest molecules.
Similar adsorption behavior was observed for 2 (see Support-
ing Information).
Many coordination polymers have been synthesized with
the aim of achieving stable microporous structures like those
of zeolites; however, the goal here is not to mimic zeolites but
to create new properties characteristic of coordination
polymers. A coordination polymer is the best candidate for
realizing of the unique nanostructure reported here. Guests
are permitted to pass the gate at a specific gate-opening
pressure that depends on the strength of the intermolecular
interaction. We expect that such coordination polymers will
find applications in gas separation, sensors, switches, and
actuators. This flexible and dynamic coordination network
opens up a new field in porous coordination materials.
Figure 4. a) ObservedXRPD pattern for the as-synthesizedsample of
1a. b) XRPD pattern for the anhydrous sample 1b (dried at 1108C un-
der reduced pressure for 5 h). Estimated channel size based on the
peak shift of (10ꢀ2) is 3.0î3.6 ä.
Experimental Section
Synthesis of 1a: A diethyl ether solution (20 mL) containing a
mixture of 4,4’-bpy (0.13 g, 0.8 mmol) and Hdhbc (0.50 g 3.2 mmol)
was carefully layered on the top of an aqueous solution (20 mL) of
Cu(NO3)2¥3H2O. Platelike green crystals began to form over a few
days. One of these crystals was used for X-ray crystallography. The
green crystals were collected by filtration, washed with ethanol, and
dried under reduced pressure for 2 h. Yield: 0.18 g (0.32 mmol, 41%).
Elemental analysis (%) calcd for C24H14N2O10Cu: C 54.81, H 3.45, N
5.33; found: C 54.64, H 4.04, N 5.12. IR (KBr pellet): n˜ ¼ 2810m,
1613s, 1492s, 1383w, 1279w, 1241m, 1131w, 1073w, 820s, 785s, 688m,
648w 554w and 493m cmꢀ1
.
Single-crystal X-ray structure determination on 1a: intensities
were measured at 298 K with a graphite-monochromatized MoKa
radiation source (l ¼ 0.71069 ä). 2198 independent reflections
(8955 total measured) were analyzed by a direct method using the
SIR92 program and expanded by Fourier techniques. Hydrogen
atoms of hydroxy groups were located in the Fourier difference map.
Hydrogen atoms except for those of hydroxy groups were placed in
idealized positions, and their parameters were not refined. Crystal
data: C24H22CuN2O10, Mr ¼ 561.99, monoclinic, space group P2/c
(No. 13), a ¼ 8.167(4), b ¼ 11.094(8), c ¼ 15.863(2) ä, b ¼ 99.703(4)8,
V¼ 1416(1) ä3, Z ¼ 2, R ¼ 0.068, wR ¼ 0.088. CCDC-191853 contains
the supplementary crystallographic data for this paper. These data
can be obtained free of charge via www.ccdc.cam.ac.uk/conts/
retrieving.html (or from the Cambridge Crystallographic Data
Centre, 12, Union Road, Cambridge CB21EZ, UK; fax: (þ 44)
1223-336-033; or deposit@ccdc.cam.ac.uk).
Figure 5. a) Nitrogen adsorption (filled circles) and desorption iso-
therms (open circles) for 1b at 298 K. Blue andreddashedlines were
determined by fitting the linear parts of the Langmuir plots in the high-
er (from 83 to 122 atm) andlower (from 8 to 34 atm) pressure ranges,
respectively. b) Adsorption (filled circles) and desorption (open circles)
isotherms of N2, CH4, CO2 andO 2 at 298 K.
of crystalline ™closed∫ form 1b into crystalline ™open∫ form
1a.[15,16] Once nitrogen gas is adsorbed into the micropores of
1a, nitrogen molecules stabilize the porous phase. Therefore,
in the desorption isotherm, the porous phase was maintained
until the pressure reached 30 atm (gate-closing pressure). This
crystal structure transformation triggered by guest molecules
430
¹ 2003 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
1433-7851/03/4204-0430 $ 20.00+.50/0
Angew. Chem. Int. Ed. 2003, 42, No. 4

Angewandte
Chemie
Coordination polymer 2 was synthesized according to a published
procedure. Elemental analysis (%) calcd for C13H8NO4Cu: C 51.06, H
2.62, N 4.58; found C 50.62, H 2.45, N 4.60.
Glycopeptide Synthesis
Toward Fully Synthetic N-Linked
Glycoproteins**
Gas adsorption measurements: Sorption isotherms were meas-
ured at 298 K on an FMS-BG (BEL inc.) automatic gravimetric
adsorption measurement system with Rubotherm magnet coupling
balance incorporated in a SUS steel pressure chamber. A known
weight (200 300 mg) of the as-synthesized sample was placed in the
aluminum sample cell in the chamber, and the sample was dried under
high vacuum at 373 K for 5 h to remove the host water molecules. The
adsorbate was dosed into the chamber, and the change in weight was
monitored. After correction for buoyancy, the absorbed amount was
determined.
Justin S. Miller, Vadim Y. Dudkin, Gholson J. Lyon,
Tom W. Muir, and Samuel J. Danishefsky*
The structural and biological consequences of cellular protein
modification through posttranslational glycosylation are cen-
tral issues in the rapidly growing field of glycobiology.[1] The
availability of homogeneous glycopeptides, both O-linked
(serine, threonine, or tyrosine a-glycosides) and N-linked
(asparagine b-glycosides), could greatly enhance insight into
glycobiology.[2] It became our view that the prospect of total
synthesis of homogeneous glycoproteins provides the best
chance for gaining such access.
Numerous methods exist for glycopeptide synthesis:
glycans have been introduced into peptides by means of
amino acid ™cassettes∫ with pendant protected saccharides,[3]
through enzymatic manipulations of glycopeptides,[4] or by
conjugation of fully elaborated, complex saccharides to short
synthetic peptides.[5] Larger O-linked glycopeptides have
been synthesized by using ligation techniques[6] such as
expressed protein ligation.[7] Bertozzi and co-workers ex-
tended the scope of the ™cassette∫ approach by applying
native chemical ligation to the synthesis of a biologically
active glycoprotein with two single-residue O-linked gly-
cans.[8] Tolbert and Wong described the ligation of a 392-
residue intein-generated peptide thioester and a dipeptide
functionalized with a single N-acetylglucosamine residue.[7c]
Using a different fragment condensation protocol, Hojo et al.
reported the synthesis of a glycopeptide domain of Emmprin
that contains an N-linked chitobiose unit, but the saccharide
was not entirely stable to the conditions required for resin
cleavage in their solid-phase synthesis.[9]
Received: August 27, 2002 [Z50052]
[1]G. A. Ozin, A. Kuperman, A. Stein, Angew. Chem. 1989, 101,
373 390; Angew. Chem. Int. Ed. Engl. 1989, 28, 359 376.
[2]A. Corma, Chem. Rev. 1997, 97, 2373 2419.
[3]M. Eddaoudi, D. B. Moler, H. Li, B. Chen, T. M. Reineke, M.
O©Keeffe, O. M. Yaghi, Acc. Chem. Res. 2001, 34, 319 330.
[4]M. Eddaoudi, J. Kim, N. Rosi, D. Vodak, J. Wachter, M.
O©Keeffe, O. M. Yaghi, Science 2002, 295, 469 472.
[5]B. Moulton, M. J. Zaworotko, Chem. Rev. 2001, 101, 1629 1658.
[6]J. S. Seo, D. Whang, H. Lee, S. I. Jun, J. Oh, Y. J. Jeon, K. Kim,
Nature 2000, 404, 982.
[7]S. Noro, S. Kitagawa, M. Kondo, K. Seki, Angew. Chem. 2000,
112, 2162 2164; Angew. Chem. Int. Ed. 2000, 39, 2082 2084.
[8]S. Kitagawa, M. Kondo, Bull. Chem. Soc. Jpn. 1998, 71, 1739
1753.
[9]R. Robson, J. Chem. Soc. Dalton Trans. 2000, 21, 3735 3744.
[10]A. J. Fletcher, E. J. Cussen, T. J. Prior, M. J. Rosseinsky, C. J.
Kepert, K. M. Thomas, J. Am. Chem. Soc. 2001, 123, 10001
10011.
[11]D. V. Soldatov, J. A. Ripmeester, S. I. Shergina, I. E. Sokolov,
A. S. Zanina, S. A. Gromilov, Y. A. Dyadin, J. Am. Chem. Soc.
1999, 121, 4179 4188.
[12]L. C. Tabares, J. A. R. Navarro, J. M. Salas, J. Am. Chem. Soc.
2001, 123, 383 387.
[13]M. Albrecht, M. Lutz, A. L. Spek, G. van Koten, Nature 2000,
406, 970 974.
[14]K. Seki, Phys. Chem. Chem. Phys. 2002, 4, 1968 1972.
[15]D. Li, K. Kaneko, Chem. Phys. Lett. 2001, 335, 50 56.
[16]R. Kitaura, K. Fujimoto, S. Noro, M. Kondo, S. Kitagawa,
Angew. Chem. 2002, 114, 141 143; Angew. Chem. Int. Ed. 2002,
41, 133 135.
[*] Prof. S. J. Danishefsky, Dr. J. S. Miller, Dr. V. Y. Dudkin
Laboratory for Bioorganic Chemistry
Sloan-Kettering Institute for Cancer Research
1275 York Avenue, New York, NY 10021 (USA)
Fax: (þ1)212-772-8691
E-mail: s-danishefsky@ski.mskcc.org
Prof. S. J. Danishefsky
Department of Chemistry, Columbia University
Havemeyer Hall, New York, NY 10027 (USA)
G. J. Lyon, Prof. T. W. Muir
Laboratory of Synthetic Protein Chemistry
The Rockefeller University
1230 York Avenue, New York, NY 10021 (USA)
[**] This work was supportedby the NIH (AI16943). The receipt of a
Pfizer Awardto S.J.D. for Creative Work in Organic Synthesis is
gratefully acknowledged. We thank Drs. Andrzej Zatorski and Ulrich
Iserloh for the preparation of starting materials andfor helpful
discussions, andDr. George Sukenick andMs. Sylvi Rusli (NMR
Core Facility, CA-02848) for mass spectral analyses.
Supporting information for this article is available on the WWW
under http://www.angewandte.org or from the author. Experimental
details include the preparation of and mass spectral characteristics
for 2 6, 8, and 10 12; andNMR spectra for 5, 6, 8, and 10.
Angew. Chem. Int. Ed. 2003, 42, No. 4
¹ 2003 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
1433-7851/03/4204-0431 $ 20.00+.50/0
431