Reversible anion exchange and catalytic properties of two cationic metal-organic frameworks based on Cu(I) and Ag(I)

Article abstract of DOI:10.1021/ja102134c

We report the synthesis and characterization of two Ag(I)/Cu(I)-based cationic metal-organic frameworks and their application in both heterogeneous catalysis and anion exchange. The Cu(I)-based material was designed from our previously reported Ag(I) cationic topology. Both structures consist of cationic layers with π-π stacked chains of alternating metal and 4,4′-bipyridine. α,Ω-Alkanedisulfonate serves as an anionic template, electrostatically bonding to the cationic layers. Due to weak interaction between the sulfonate template and cationic extended framework, both materials display reversible anion exchange for a variety of inorganic species. Indeed, the Ag(I)-based material exhibits highly efficient uptake of permanganate and perrhenate anion trapping, a model for pertechnetate trapping. The materials also display heterogeneous Lewis acidity, likely due to the coordinatively unsaturated metal sites which only bind to two bipy nitrogens and a weak interaction with one sulfonate oxygen. A comparative study on the influence of structure versus size selectivity and reusability for both exchange and catalysis is discussed.

Full text of DOI:10.1021/ja102134c

Published on Web 04/28/2010
Reversible Anion Exchange and Catalytic Properties of Two
Cationic Metal-Organic Frameworks Based on Cu(I) and Ag(I)
Honghan Fei, David L. Rogow, and Scott R. J. Oliver*
UniVersity of California, Santa Cruz, Department of Chemistry and Biochemistry, 1156 High
Street, Santa Cruz, California 95064
Received March 12, 2010; E-mail: soliver@chemistry.ucsc.edu
Abstract: We report the synthesis and characterization of two Ag(I)/Cu(I)-based cationic metal-organic
frameworks and their application in both heterogeneous catalysis and anion exchange. The Cu(I)-based
material was designed from our previously reported Ag(I) cationic topology. Both structures consist of cationic
layers with π-π stacked chains of alternating metal and 4,4′-bipyridine. R,ω-Alkanedisulfonate serves as
an anionic template, electrostatically bonding to the cationic layers. Due to weak interaction between the
sulfonate template and cationic extended framework, both materials display reversible anion exchange for
a variety of inorganic species. Indeed, the Ag(I)-based material exhibits highly efficient uptake of
permanganate and perrhenate anion trapping, a model for pertechnetate trapping. The materials also display
heterogeneous Lewis acidity, likely due to the coordinatively unsaturated metal sites which only bind to
two bipy nitrogens and a weak interaction with one sulfonate oxygen. A comparative study on the influence
of structure versus size selectivity and reusability for both exchange and catalysis is discussed.
8-10
Introduction
organic species are able to reversibly exchange into LDHs.
In many cases, however, the anion either cannot be removed
Anionic pollutants such as perchlorate (ClO
4
-), arsenate
or will deintercalate in the presence of common anions such
3
-
-
(
AsO
4
), and chromate (CrO
4
) are becoming increasingly
11-15
as carbonate.
The need for materials qualified to trap
problematic and are listed as EPA (U.S. Environmental
or exchange both inorganic and organic anions remains. A
recent development is a nanoporous cationic metal borate
that on initial studies reversibly traps anions including
pertechnetate, though the material is based on slightly
1
Protection Agency) priority pollutants. Perchlorate contami-
nation of water by spent rocket fuel was reported in the lower
2
Colorado River in 2002. At the same time, organic anion
1
6
pollution from pharmaceuticalsswhich are also intrinsically
anionicsis now a worldwide problem, as most wastewater
radioactive thorium and anion removal percentage was 72%.
New heterogeneous catalysts with greater reusability,
efficiency, and ease of synthesis are continuously sought for
environmental concerns. Micro/mesoporous materials have
been extensively studied for their high internal surface area
where the pore/channel catalytic sites lead to improved
3
treatment ineffectively removes pharmacological anions.
Resins with cationic groups and exchangeable counteranions
4
,5
are still considered the standard ion exchanger. These are
of limited thermal and chemical stability, however, due to
6
17
their organic polymer nature. Layered double hydroxides
efficiency. Microporous inorganic-organic hybrid materials
(
LDHs) are an isostructural group of structures based
with coordinatively unsaturated metal sites continue to be
2
+
3+
18-20
on brucite with a general formula [M 1-x
M
x
(OH)
2
]-
discovered, with Lewis acid catalytic properties.
There
n-
2+
3+
A
x/n · mH
2
O, where M and M are a range of metals (e.g.,
Mg and Al ) and A are interlamellar anions (e.g.,
). They are considered to be an ideal alternative to
anion exchange materials. A variety of both inorganic and
have been few systematic investigations, however, on the
2
+
3+
n-
2
-
CO
3
(
8) Prasanna, S. V.; Kamath, P. V. Solid State Sci. 2008, 10, 260–266.
7
(9) Ma, R. Z.; Liu, Z. P.; Li, L.; Iyi, N.; Sasaki, T. J. Mater. Chem. 2006,
1
6, 3809–3813.
(
10) Goh, K. H.; Lim, T. T.; Dong, Z. L. EnViron. Sci. Technol. 2009, 43,
(
1) Perchlorate enVironmental contamination: toxicological reView and
risk characterization; Second external review draft, NCEA-1-0503;
U.S. EPA, Office of Research and Development, National Center for
Environmental Assessment, U.S. Government Printing Office: Wash-
ington, DC, 2002.
2537–2543.
(11) Chibwe, K.; Jones, W. J. Chem. Soc., Chem. Commun. 1989, 926–
927.
(12) Mackenzie, K. J. D.; Meinhold, R. H.; Sherriff, B. L.; Xu, Z. J. Mater.
Chem. 1993, 3, 1263–1269.
(
(
2) Urbansky, E. T. EnViron. Sci. Pollut. Res. Int. 2002, 9, 187–192.
3) Castiglioni, S.; Bagnati, R.; Fanelli, R.; Pomati, F.; Calamari, D.;
Zuccato, E. EnViron. Sci. Technol. 2006, 40, 357–363.
(13) Hibino, T.; Tsunashima, A. Chem. Mater. 1998, 10, 4055–4061.
(14) Stanimirova, T. S.; Kirov, G.; Dinolova, E. J. Mater. Sci. Lett. 2001,
20, 453–455.
(
4) Praharaj, S.; Nath, S.; Panigrahi, S.; Ghosh, S. K.; Basu, S.; Pande,
S.; Jana, S.; Pal, T. Inorg. Chem. 2006, 45, 1439–1441.
5) Jana, S.; Praharaj, S.; Panigrahi, S.; Basu, S.; Pande, S.; Chang, C. H.;
Pal, T. Org. Lett. 2007, 9, 2191–2193.
(15) Lima, E.; Lasperas, M.; de Menorval, L. C.; Tichit, D.; Fajula, F. J.
Catal. 2004, 223, 28–35.
(
(16) Wang, S.; Alekseev, E. V.; Diwu, J.; Casey, W. H.; Phillips, B. L.;
Depmeier, W.; Albrecht-Schmitt, T. E. Angew. Chem., Int. Ed. 2010,
49, 1057–1060.
(
(
6) Mark, R.; Findley, W. N. Polym. Eng. Sci. 1978, 18, 6–15.
7) Slade, D. G. E. a. R. C. T. Layered Double Hydroxides; Duan, X.,
Evans, D. G., Eds.; Springer-Verlag: New York, 2006.
(17) Joo, S. H.; Choi, S. J.; Oh, I.; Kwak, J.; Liu, Z.; Terasaki, O.; Ryoo,
R. Nature 2001, 412, 169–172.
7
202 9 J. AM. CHEM. SOC. 2010, 132, 7202–7209
10.1021/ja102134c  2010 American Chemical Society

Ag(I)/Cu(I)-Based Cationic Metal-Organic Frameworks
A R T I C L E S
rational design of extended frameworks with coordinatively
unsaturated metal sites and controlled pore/channel size for
size-selective Lewis acidity.
acid disodium salt (NaO
3
SCH
2
CH
2
SO
3
Na, TCI Inc., 95%) were
used as-purchased for the reversible anion exchange reactions.
-Butanone (CH COCH CH , Acros Organics, 99%), 2-pentanone
CH COCH CH CH TCI America, 97%), benzophenone
COC , Alfa-Aesar, 99%), ethylene glycol (HOCH CH OH,
CH , Fisher, 99.7%) were
used as-purchased for the catalytic studies.
Synthesis. Yellow-brown crystals of Cu
SCH CH SO O (which we denote SLUG-22: University
2
3
2
3
(
(
3
2
2
3
,
Metal-organic frameworks (MOFs) have received remark-
C
6
H
5
6
H
5
2
2
able attention in recent years because of not only their intriguing
2
1,22
Acros Organics, 99%), and toluene (C
6
H
5
3
structures but also their use in gas absorption/storage,
gas
2
3,24
25,26
separation,
and ion exchange.
As a small subgroup of
2
2
(4,4′-bipy) -
MOFs, cationic MOFs occur when the positive charge on the
metal ions outnumber the negative (or neutral) charge on the
organic linkers. The net positive charge on the framework
(
O
3
2
2
3
)·3H
2
of California, Santa Cruz, structure no. 22) were synthesized under
hydrothermal conditions. A mixture of Cu(CH COO) ·H O (0.27
3
2
2
2
7-33
necessitates charge-balancing extra-framework anions.
Among this group of cationic MOFs, only a few have been
3 2 2 3
g, 1.35 mmol), HO SCH CH SO H (0.29 g, 1.52 mmol), 4,4′-
bipyridine (0.21 g, 1.34 mmol), and 10 mL of water was stirred at
room temperature for 10 min and then transferred to a 15 mL Teflon
lined autoclave to 2/3 filling. The autoclaves were heated at 175
28,30,32
investigated for reversible anion exchange.
There has been
no investigation of anion pollutant trapping by exchange of
organic species for inorganic molecules.
°
C for 4 days under autogenous pressure, followed by slow cooling
at a rate of 6 °C/h to room temperature. During the reaction, the
pH slightly decreased from 4.1 to 3.5. Yellow-brown block crystals
were isolated after filtration and rinsed with water and acetone
Herein, we report the first cationic MOFs that possess both
chemical stability for recyclable heterogeneous catalysis and
structural flexibility for reversible anion exchange. The first
is a Cu(I)-based cationic MOF similar to our recently reported
(
3
yield: 0.46 g, 95% based on copper acetate). IR (KBr pellets):
467s (O-H stretch), 3050m (aromatic C-H stretch); 1605s, 1535s,
1490s, 1418s (aromatic CdC and CdN stretch); 1328s (CH2
3
4
Ag
2
(4,4′-bipy)
2
(O
3
SCH
2
CH
2
SO
3
2
) · 4H O (SLUG-21). The
materials were designed to have weak electrostatic interaction
between the interlamellar anions and cationic MOF layers.
As a result, both display reversible anion exchange between
organosulfonate and various inorganic species. SLUG-21 has
stretch); 1200m, 1070m (SO
bending).
Colorless crystals of SLUG-21 were synthesized under hydro-
thermal conditions or at room temperature as previously reported.
A reactant solution with a molar ratio of 1:1:1:400 for AgNO3/
3
stretch); 863s, 817s (aromatic C-H
34
-
-
also been applied for anion trapping of MnO
4
and ReO
4
3
5
EDSA/4,4′-bipy/H O were placed into a 15 mL Teflon-lined
2
as a model for pertechnetate, a problematic pollutant. We
also report a detailed comparative study of the two materials
toward size-selective heterogeneous Lewis acid catalysis.
autoclave to 2/3 filling and heated at 150 °C for 5 days under
autogenous pressure. For both compounds, the product was filtered,
rinsed with acetone, and allowed to air-dry.
Anion Exchange. 100 mg of either SLUG-21 (0.126 mmol)
or SLUG-22 (0.147 mmol) solid were placed into 20 mL of 0.1
M NaNO3 or NaClO4 solution and allowed to react either
statically for 7 days or with mild stirring for 1 to 3 days. The
exchange solution and solid were monitored at various time
intervals to follow the exchange progress. The crystal products
were isolated by filtration and rinsed with water/acetone. The
crystals after NaNO3 or NaClO4 exchange were placed into
excess 0.1 M EDS disodium salt solution. The mixture was
gently stirred for 3 days, followed by filtration to isolate the
solid product and exchange solution after the second exchange.
The entire process was repeated several times.
Experimental Section
Reagents. Silver nitrate (AgNO
3
, Fisher, 99%), copper acetate
O, Alfa-Aesar, 98%], 1,2-ethanedis-
SO H, TCI Inc., 95%), and 4,4′-
monohydrate [Cu(CH
ulfonic acid (HO SCH
bipyridine [(C N)
for the synthesis. Sodium nitrate (NaNO
perchlorate monohydrate (NaClO ·H O, Fluka Analytical, 98%),
potassium permanganate (KMnO , Fisher, 99.8%), potassium
perrhenate (KReO , Acros Organics, 99%), and 1,2-ethanedisulfonic
3
COO)
2
·H
2
2
3
2
CH
3
5
H
4
2
, Acros Organics, 98%] were used as-received
, Fisher, 99%), sodium
3
4
2
4
4
(
(
(
(
18) Lee, J.; Farha, O. K.; Roberts, J.; Scheidt, K. A.; Nguyen, S. T.; Hupp,
J. T. Chem. Soc. ReV. 2009, 38, 1450.
Heterogeneous Catalysis. Ketal formation with various sub-
strates was used to characterize the performance of both SLUG-21
and SLUG-22 as a Lewis acid catalyst. 100 mg of the as-synthesized
catalyst, 70 mmol of ketone (2-butanone, 2-pentanone or benzophe-
none), and 70 mmol of ethylene glycol were introduced into 80
mmol of toluene, the latter used as a solvent. The reaction was
refluxed at 110 °C under Dean-Stark conditions for 12 h and/or
19) Swanson, C. H.; Shaikh, H. A.; Rogow, D. L.; Oliver, A. G.; Campana,
C. F.; Oliver, S. R. J. J. Am. Chem. Soc. 2008, 130, 11737–11741.
20) Horike, S.; Dinc aˇ , M.; Tamaki, K.; Long, J. R. J. Am. Chem. Soc.
2
008, 130, 5854–5855.
21) Chae, H. K.; Siberio-Perez, D. Y.; Kim, J.; Go, Y.; Eddaoudi, M.;
Matzger, A. J.; O’Keeffe, M.; Yaghi, O. M. Nature 2004, 427, 523–
5
27.
(
(
(
22) Murray, L. J.; Dinc aˇ , M.; Long, J. R. Chem. Soc. ReV. 2009, 38, 1294–
2
4 h. The catalyst was isolated by filtration and reused on
1
314.
subsequent runs without further treatment. All product yields were
determined by H NMR (Supporting Information).
23) Wang, B.; C oˆ t e´ , A. P.; Furukawa, H.; O’Keeffe, M.; Yaghi, O. M.
Nature 2008, 453, 207–211.
1
Instrumental Details. Samples for powder X-ray diffraction
(PXRD) were measured on a Rigaku Americas Miniflex Plus
diffractometer and were scanned from 2° to 60°(2θ) at a rate of 2°
per minute and 0.04° step size under Cu KR radiation (λ ) 1.5418
Å). Single-crystal X-ray diffraction data for SLUG-22 were
collected at Beamline 11.3.1 at the Advanced Light Source (ALS),
Lawrence Berkeley National Laboratory (λ ) 0.77490 Å). The
structure was solved by direct methods and refined with SHELX-
24) Banerjee, R. P., A.; Wang, B.; Knobler, C.; Furukawa, H.; O’Keeffe,
M.; Yaghi, O. M. Science 2008, 319, 939–943.
(
(
(
25) Oliver, S. R. J. Chem. Soc. ReV. 2009, 38, 1868–1881.
26) Custelcean, R.; Moyer, B. A. Eur. J. Inorg. Chem. 2007, 1321–1340.
27) Yaghi, O. M.; Li, H. L.; Groy, T. L. Inorg. Chem. 1997, 36, 4292–
4
293.
(
28) Min, K. S.; Suh, M. P. J. Am. Chem. Soc. 2000, 122, 6834–6840.
(29) Sudik, A. C.; Cote, A. P.; Yaghi, O. M. Inorg. Chem. 2005, 44, 2998–
3
000.
36
(
(
30) Du, M.; Zhao, X. J.; Guo, J. H.; Batten, S. R. Chem. Commun. 2005,
TL. The models were refined by full-matrix least-squares analysis
4
836–4838.
2
of F against all reflections. All non-hydrogen atoms were refined
31) Bao, S. S.; Ma, L. F.; Wang, Y.; Fang, L.; Zhu, C. J.; Li, Y. Z.; Zheng,
L. M. Chem.sEur. J. 2007, 13, 2333–2343.
with anisotropic thermal displacement parameters. Crystal structure
views were obtained using Diamond v3.2 and rendered by POV-
Ray v3.6. Thermogravimetric analysis (TGA) was performed using
(
(
32) Michaelides, A.; Skoulika, S. Cryst. Growth Des. 2009, 9, 2039–2042.
33) Chen, X. D.; Wan, C. Q.; Sung, H. H. Y.; Williams, I. D.; Mak,
T. C. W. Chem.sEur. J. 2009, 15, 6518–6528.
(34) Fei, H.; Paw U, L.; Rogow, D. L.; Bresler, M. R.; Abdollahian, Y. A.;
Oliver, S. R. J. Chem. Mater. 2010, 22, 2027–2032.
(36) SHELXTL Crystal Structure Determination Package; Bruker Analytical
X-ray Systems Inc.: Madison, WI, 1995-99.
(
35) Wang, Y.; Gao, H. J. Colloid Interface Sci. 2006, 301, 19–26.
J. AM. CHEM. SOC. 9 VOL. 132, NO. 20, 2010 7203

A R T I C L E S
Fei et al.
Table 1. Crystallographic Information, Data Collection, and
Refinement Parameters for SLUG-22
empirical formula
2 22 26 9 4 2
Cu C H O N S
formula weight (g· mol-1) 681.71
temperature (K)
wavelength (Å)
crystal system
space group
150(2)
0.774 90 (synchrotron)
triclinic
j
P1
unit cell dimensions
a ) 8.9031(9) Å
b ) 10.9120(11) Å
R ) 103.6150(10)°
γ ) 97.5290(10)°
c ) 14.4304(10) Å
ꢀ
) 102.272(2)°
volume (Å3)
1307.0(2)
Z
2
crystal descriptions
crystal size (mm )
index range
yellow block
3
0.05 × 0.04 × 0.04
-11 e h e 11, -13
e k e 13, -18 e l e 18
density (calculated g· cm- ) 1.732
3
reflections collected
unique reflections
14355
5316
final R indices [I > 2σ(I)]
R
R
1
) 0.0371, wR
2
2
) 0.0957
R indices (all data)
1
) 0.0500, wR
) 0.1021
-
-3
Figure 1. ORTEP diagram and atomic labeling of SLUG-22. Thermal
ellipsoids are shown at 50% probability.
largest diff. peak and hole -0.63 and 0.67 e ·Å
2
goodness of fit on F
1.037
1
). The Cambridge Crystal Structure Database indicates the
3
8
median bond length of Cu(I)-O is 1.978 Å. The literature
bond length values between Cu(I) and a sulfonate oxygen are
a TA Instruments 2050 TGA by heating from 30 to 700 °C under
a N purge with a gradient of 15 °C/min. In situ mass spectra
2
3
9-41
2
.03 to 2.22 Å,
further supporting only weak electrostatic
coupled to the TGA were collected on a Pfeiffer Vacuum Ther-
moStar GSD 301 T3 mass spectrometer with a 70 eV ionization
potential. UV-vis spectroscopic studies were collected on a
Hewlett-Packard Model 8452A spectrophotometer to monitor the
exchange progress. Fourier transform infrared (FTIR) spectroscopy
of the materials was collected on a Perkin-Elmer Spectrum One
interaction between the cationic Cu(I)-bipy framework and the
charge-balancing EDS anion. This feature is important for the
anion exchange and catalytic properties of the material that will
be discussed below.
Thermal Properties. The thermal stability of SLUG-22 was
1
spectrophotometer with KBr pellets. H NMR spectra were collected
investigated by TGA under a N
2
purge and ex situ PXRD
with a Varian Oxford 600 MHz spectrometer by dissolving the
sample in 700 µL of deuterated choloroform with tetramethylsilane
as an internal standard.
(
Figures 3 and S3, Supporting Information, respectively). The
TGA trace of SLUG-22 indicates a slight mass loss of ∼3% at
ca. 70 °C, likely due to partial loss of the interlamellar water
(observed: 3.1%; calculated: 7.9%, Figure 3). In agreement with
Results and Discussion
the loss of the crystallographic water is a slight structural
transformation observed by ex situ PXRD. This structure is
stable to ca. 325 °C, when both 4,4′-bipyridine and ethanedis-
ulfonate ligands in the structure start to decompose. This gradual
mass loss leads to metallic copper at ca. 450 °C (observed:
Synthesis. Yellow block crystals of SLUG-22 were obtained
hydrothermally as a pure phase and in high yield at 175 °C and
pH ≈ 4.1. A lower temperature (125 or 150 °C) or slight change
in the pH led to lower crystallinity and an unidentified impurity
in the product that could be observed by optical microscopy.
During the hydrothermal synthesis, the Cu(II) reagent was
reduced to Cu(I), as observed by the color change between the
copper precursor and product. Unlike SLUG-21, the material
cannot be synthesized at room temperature or reflux conditions,
and nonaqueous solvents were unsuccessful. The synthesized
product was insoluble in water, acetone, ethanol, and other
common organic solvents.
1
7.8%; calculated: 18.8%). Ex situ PXRD under a N
2
flow also
confirms the formation of pure phase Cu (PDF# 71-3761) above
4
50 °C. The slight mass increase from 450 to 490 °C is
presumably due to nitrogen uptake.
Anion Exchange. Considering that both SLUG-22 and our
previously reported SLUG-21 have only one oxygen of the EDS
weakly interacting with the cationic metal-bipy layers, the
materials were investigated systematically for reversible anion
exchange. Many previously reported cationic inorganic materials
can only exchange its extra-framework anions for smaller or
similar sized inorganic anions. For example, Yu et al. reported
Structure Characterization. SLUG-22 crystallizes in a tri-
j
clinic system with a P1 space group (Table 1), which is the
same as our previous reported SLUG-21 structure. Synchrotron
single-crystal X-ray diffraction studies reveal that the structure
of SLUG-22 is also a cationic one-dimensional MOF with
infinite chains composed of alternating copper(I) centers and
-
Cl located in a microporous aluminoborate framework can be
-
42
partially exchanged by Br . A similar phenomenon is observed
4
3
for the 2-D metal oxychloride francisite and its derivatives.
As mentioned earlier, coordination polymers have been inves-
4
,4′-bipy organic linkers (Figures 1 and 2). The Cu(I)-bipy
chains are arranged into close-packed layers by π-π stacking
between adjacent 1D chains (Figure 2, right). The Cu(I) atoms
and its full d shell form an almost linear two-coordinate
environment, with N(5)-Cu(3)-N(6) and N(7)-Cu(4)-N(8)
angles between 169.40(11)° and 169.48(11)°.
(
(
37) Lucassen, A. C. B.; Karton, A.; Leitus, G.; Shimon, L. J. W.; Martin,
J. M. L.; van der Boon, M. E. Cryst. Growth Des. 2007, 7, 386–392.
38) Cambridge Crystal Structure Database; 8022 of 8818 (91%)
Cu(I)-O bonds with a bond length less than 2.439 Å, and mean bond
length is 2.090 Å with standard deviation of 0.213 Å.
39) Wu, M. Y.; Yuan, D. Q.; Han, L.; Wu, B. L.; Xu, Y. Q.; Hong, M. C.
Eur. J. Inorg. Chem. 2006, 526–530.
The distance between the adjacent Cu(I)-bipy chains is ca.
(
3
.466 Å, which is in the reasonable range of π-π stacking
37
(40) Ju, J.; Lin, J. H.; Li, G. B.; Yang, T.; Li, H. M.; Liao, F. H.; Loong,
distance, leading to a cationic layer. Only one oxygen of the
interlamellar EDS anions has a weak interaction toward Cu(I),
with distances in the range of 2.434(2) to 2.501(2) Å (Figure
C. K.; You, L. P. Angew. Chem., Int. Ed. 2003, 42, 5607–5610.
(
41) Liu, S. X.; Xie, L. H.; Gao, B.; Zhang, C. D.; Sun, C. Y.; Li, D. H.;
Su, Z. M. Chem. Commun. 2005, 5023–5025.
7
204 J. AM. CHEM. SOC. 9 VOL. 132, NO. 20, 2010

Ag(I)/Cu(I)-Based Cationic Metal-Organic Frameworks
A R T I C L E S
Figure 2. Crystallographic a-projection (left) and b-projection (right) of the SLUG-22 structure (copper: green; carbon: gray; sulfur: yellow; oxygen: red;
nitrogen: blue).
the presence of 4,4′-bipy throughout the exchange, as expected.
-
1
The strong, broad band at ∼1200 cm is characteristic of the
-
SO
3
group, confirming the presence of EDS anions in the initial
SLUG-21 framework (Figure 4a, black spectrum). By simply
immersing as-synthesized SLUG-21 crystals in 0.1 M NaNO
3
solution at room temperature with mild stirring, both FTIR and
PXRD indicate the EDS anions in SLUG-21 were completely
-
exchanged for NO
3
after 3 days (Figure 4b, red spectra, 1330
-
1
-1
cm ). The broad ∼1200 cm band has disappeared, with a
-1
-
new peak at ∼1330 cm characteristic of NO
3
(NsO stretch).
PXRD further supports a complete, reversible exchange
between EDS and nitrate. Although the exchanged product could
not be solved by X-ray crystallography due to lower crystal
quality (Figure S2), the resultant PXRD pattern after 3 days
corresponds exactly to the theoretical pattern of the previously
46
3
reported Ag(4,4′-bipy)NO . The first peak is at a higher angle,
as expected for replacement of the larger EDS anions for the
smaller nitrate molecules. The layer-to-layer distance decreased
from ca. 4.6 Å to ca. 3.0 Å, with long Ag-Ag contact between
the layers [ca. 2.976(2) Å] and the nitrates between adjacent
Ag-bipy chains The bipyridine rings have thus “shuttered” upon
nitrate intercalation, favoring interlayer π-π stacking. Simply
immersing the nitrate exchanged product into a 0.1 M EDS
disodium salt aqueous solution reforms the original SLUG-21,
confirmed by both FTIR and PXRD (Figure 4c, blue spectra)
The reversibility implies that the bulky alkanesulfonate anions
favor a more open framework with stronger intralayer π-π
stacking. Indeed, SLUG-21 forms during the synthesis rather
Figure 3. Thermogravimetric analysis of SLUG-22 in the range of 30 to
00 °C.
7
tigated for reversible/irreversible anion exchange, including
-
-
-
-
-
extra-framework NCS , N
ClO .
4
3
, NO
3 3 3
, Cl , CF SO , and/or
-
28,30,32
Meanwhile, there have been no reports of
exchange of bulky organic anions for inorganic species aside
from the LDHs/hydrotalcites, which have limited reusability and
anion uptake.
As initial examples of exchange, we have focused on several
inorganic anions for two reasons: (i) the oxo-anions of many
metals are EPA priority pollutants; (ii) nitrate and perchlorate
were successful templates for our previously reported cationic
than Ag(4,4′-bipy)NO
the synthetic reagent. In fact, SLUG-21-NO
1 when placed into a mixed aqueous solution of 0.1 M EDS
3
despite the fact that silver nitrate was
3
reforms SLUG-
2
44,45
disodium salt and 0.1 M NaNO , an analogous anionic mixture
3
lead fluoride structures.
Figure 4 shows the FTIR and PXRD
compared to the synthetic procedure. Based on these observa-
tions, we conclude that the SLUG-21 exchange reaction with
of the solid products after exchange of the EDS in SLUG-21
for nitrate. The four singlet bands at 1605, 1535, 1490, and
1
1
418 cm- are aromatic CdC and CdN bending, confirming
(
44) Tran, D. T.; Zavalij, P. Y.; Oliver, S. R. J. J. Am. Chem. Soc. 2002,
1
24, 3966–3969.
(
(
42) Yu, J. H.; Xu, R. R.; Chen, J. S.; Yue, Y. J. Mater. Chem. 1996, 6,
65–468.
43) Ok, K. M.; Halasyamani, P. S. Inorg. Chem. 2002, 41, 3805–3807.
(45) Rogow, D. L.; Russell, M. P.; Wayman, L. M.; Swanson, C. H.; Oliver,
A. G.; Oliver, S. R. J. Cryst. Growth Des. 2010, 10, 823–829.
(46) Yaghi, O. M.; Li, H. L. J. Am. Chem. Soc. 1996, 118, 295–296.
4
J. AM. CHEM. SOC. 9 VOL. 132, NO. 20, 2010 7205

A R T I C L E S
Fei et al.
exchange between an individual crystal of SLUG-21 and
stoichiometric excess 1.26 mM aqueous permanganate solution.
A visible color change of both crystals (more were present
outside the image area) and surrounding solution occurred in
1
0 min. The permanganate anions can be seen diffusing into
the structure from the left after 15 min of immersion (Figure
). No apparent morphology change occurred after immersion
5
of the crystals in solution for 1 h. The crystals after exchange
again could not be solved by single crystal X-ray diffraction
due to reduction in crystallinity. Nevertheless, the crystals retain
their morphology throughout, and the anion exchange reaction
is therefore a solid-state process.
A series of UV-vis patterns were collected on the perman-
ganate solution during the anion exchange process (Figure 6).
In this case, a stoichiometric excess of 150 mg of as-synthesized
SLUG-21 crystals was placed into 50 mL of 1.26 mM
permanganate solution; the mixture was stirred mildly for 5 min
before UV-vis data collection to ensure homogeneity. To
determine percent exchange, UV-vis spectra were also collected
of a saturated KMnO
4
solution, diluting from 1 to 5 mL with
deionized water. Based on Beer’s Law, 36.8%, 68.3%, and
-
1
00% of the MnO
4
from the solution were exchanged into
SLUG-21 after 8, 24, and 96 h, respectively. The absorption
below 280 nm for the 96 h sample is due to EDS, which is
now fully exchanged and present in sufficient concentration to
absorb. Both crystal morphology and size are unchanged
throughout the 4 day anion exchange. SLUG-21 remains
heterogeneous at the bottom of the solution and can be readily
recovered after exchange.
PXRD before and after the exchange reaction confirms that no
AgMnO
4
phase formed (bars in Figure 7). Instead, the new SLUG-
phase is highly crystalline, with a shift
(change 22 to 21)-MnO
4
of both (010) and (001) peaks to a higher angle (the d-spacing
decreases from ca. 9.63 and 8.50 Å to 7.49 and 8.25 Å,
respectively). Since the ethanedisulfonate anions are oriented in
the [101] direction, the lattice has contracted along both the b and
c axes, presumably due to the difference in anion size/shape
between permanganate and ethanedisulfonate. No known structure
exists based on Ag-bipy and permanganate, nor was single crystal
analysis possible. FTIR before and after anion exchange (Figure
S4, Supporting Information), however, confirms the presence of
the Ag-bipy framework, while the sulfonate peaks were com-
pletely replaced by permanganate peaks. Although exchange of
Figure 4. FTIR (top) and PXRD (bottom) of (a) SLUG-21 (black); (b)
the solid product after 3 days exchange reaction between SLUG-21 and
0
.1 M NaNO
days exchange reaction between “SLUG-21-NO
salt solution (blue). Theoretical PXRD patterns of SLUG-21 (black) and
SLUG-21-NO (red) are shown as bars in the bottom figure.
3
(“SLUG-21-NO
3
”, red); and (c) the solid product after 3
3
” and 0.1 M EDS sodium
3
nitrate is concentration induced rather than a stronger interaction
between the cationic framework and anion. The large hydration
energy of nitrate also likely aids in the reversibility.
In addition to reversible exchange with nitrate, SLUG-21
exhibits analogous chemistry with perchlorate. FTIR and PXRD
4
Ag-bipy-MnO for EDS was unsuccessful, the permanganate
anions can be trapped, with selectivity: SLUG-21 in a mixed
solution of both permanganate and nitrate (0.1 M each) preferably
reacts with permanganate, leading to the same phase as that above
using pure permanganate solution. These data indicate that the
permanganate pollutant trapping is based on a stronger interaction
(
Figure S2, Supporting Information) indicate that perchlorate
anions intercalate into the cationic layered framework, releasing
-
EDS and resulting in the previously reported phase Ag(4,4′-
between MnO4 and the SLUG-21 cationic framework, which
47
bipy)ClO
4
.
The structure is again a cationic layer where again
would also account for the lack of reversibility for permanganate
exchange. In addition to solid-state anion pollutant trapping with
permanganate, SLUG-21 display analogous chemistry with per-
rhenate, as determined by FTIR and PXRD (Figure S5, Supporting
Information). The material may therefore find application as a
pertechnetate trapping material.
SLUG-22 was also employed in the same exchange reaction
with nitrate and perchlorate to obtain structure-property insight.
The structure is similar to SLUG-21, so we wished to understand
if this anion exchange property is common for the bipy polymers
of d10 group 11 metals. The characteristic absorption of nitrate
in the UV region at 305 nm was followed since EDS has no
absorption at wavelengths longer than 240 nm. Progression of
the interlayer distance decreased due to the smaller anion size,
closer interaction to the anion [ca. 3.913(4) Å], and shuttering
of the Ag-bipy chains to give interlayer π-π stacking.
As a final illustration of anion exchange capability, we
investigated SLUG-21 for permanganate and perrhenate trap-
ping. These anions were chosen as models for pertechnetate, a
problematic radioactive pollutant. The trapping was monitored
versus time by optical microscopy and UV-vis spectroscopy.
Figure 5 shows the in situ optical microscopy of the anion
(
47) Wang, L. S.; Zhang, J. F.; Yang, S. P. Acta Crystallogr., Sect. E 2004,
E60, m1484–m1486.
7
206 J. AM. CHEM. SOC. 9 VOL. 132, NO. 20, 2010

Ag(I)/Cu(I)-Based Cationic Metal-Organic Frameworks
A R T I C L E S
Figure 5. Optical micrographs of SLUG-21 crystals in permanganate solution versus time (scale bar: 500 µm).
Figure 7. PXRD of (a) as-synthesized SLUG-21 and (b) SLUG-21 after
exchange with excess permanganate after 4 days. Theoretical pattern of
AgMnO is shown as bottom bars, indicating no structural decomposition
4
occurred during the anion exchange.
Figure 6. UV-vis spectra of the permanganate solution during anion
exchange with SLUG-21 at various time intervals, showing complete
exchange for EDS.
UV-vis absorption spectra for anion exchange of SLUG-22 in
sodium nitrate solution was carried out for 3 days (Figure 8).
Nitrate was added in 50% molar excess (1.0 mmol SLUG-22
in 1.5 mmol nitrate solution), and SLUG-22 exhibits highly
efficient anion exchange with over 90% exchange (based on
SLUG-22) after 1 day, forming the previously reported struc-
cm-1 (sulfonate stretching) after 3 days of exchange reaction.
-1
The appearance of a strong broad band at ∼1100 cm confirms
49
that the perchlorate has exchanged with EDS, again forming a
previously reported structure, with complete reversibility after
subsequent exchange in EDS solution.
4
8
Size-Selective Heterogeneous Catalysis. In addition to the
excellent anion exchange properties of SLUG-21 and SLUG-
ture. EDS could be reintroduced, giving rise to the partially
dehydrated material mentioned above, obtained after heating
SLUG-22 to 70 °C (Figure S3, Supporting Information). In
addition to nitrate, SLUG-22 also displayed reversible anion
exchange with perchlorate (Figure 9). Similar to the SLUG-21
exchange reaction above, FTIR shows the disappearance of the
2
2, both compounds are highly stable in common organic
solvents. We then found these two materials are catalytically
active in heterogeneous ketal formation. Ketalization is an
important method to protect carbonyl groups in organic synthesis
-1
broad absorbance peak at ∼1200 cm and singlet band at 1020
(
49) Zhang, J.; Kang, Y.; Wen, Y.; Li, Z.; Qin, Y.; Cheng, J.; Yao, Y.
(
48) Yaghi, O. M.; Li, H. J. Am. Chem. Soc. 1995, 117, 10401–10402.
Acta Crystallogr., Sect. E 2004, E60, m504–m505.
J. AM. CHEM. SOC. 9 VOL. 132, NO. 20, 2010 7207

A R T I C L E S
Fei et al.
Scheme 1. Acid Catalyzed Ketal Formation between 2-Butanone
and Ethylene Glycol
Table 2. Conversion Yields for the Ketalization of 2-Butanone
a
catalyst
number of run
time (h)
conv. (%)
SLUG-21
run #1
run #2
run #3
run #1
run #2
run #3
12
12
12
12
12
12
97
96
95
93
72
71
SLUG-22
a
1
Conversion yields were determined by H NMR.
5
0
and drug design. The reaction requires a Lewis acid catalyst
to activate the oxygen of the carbonyl group, allowing glycol
to substitute the ketone group. The reaction typically uses
Figure 8. UV-vis absorption spectra of the anion exchange solution for
SLUG-22 and nitrate.
5
0
homogeneous iodine or toluenesulfonate, in 56-94% yield.
Both structures as-prepared display heterogeneous Lewis
acidity and reusability in activating 2-butanone to form 2-ethyl-
2
-methyl-[1,3]-dioxolane (Scheme 1, Table 2). Unlike the
conventional homogeneous catalyst, both SLUG-21 and SLUG-
2 are heterogeneous and easily separated from the product.
2
The catalyst can be reused without further treatment after
catalytic runs. Removal of either catalyst by filtration after a
6
h reaction time gave only a 2% increase in yield after an
additional 6 h of reflux. This observation confirms that the
catalysis is heterogeneous, with no leaching of a homogeneously
active species into the reaction solution. In addition, the sulfonate
is retained by the material, as confirmed by postcatalysis FTIR
of the solid. SLUG-21 shows almost no decrease in yield after
additional runs, while SLUG-22 showed a decrease in yield from
93% to 72% for the second run and 71% for the third. Extending
the reaction time from 12 to 24 h did not enhance the conversion
efficiency. PXRD of the postcatalysis SLUG-22 material shows
a transformation occurred, with a shift of the (10-1) and (20-2)
peaks to a lower angle (Figure 10). This expansion of the lattice
is possibly due to reorientation of the EDS anions (Figure 2)
and bonding to the Ag centers to reduce catalytic activity.
Nevertheless, this activated form of SLUG-22 is stable under
these conditions, retaining its crystallinity in spite of the lower
ketal yield.
We also investigated the size selectivity of the catalyst for
ketone precursor, to determine whether the reactions are
occurring in the channels of the relatively open interlayer space
(Figure 2) or on the crystal surface. Ketalization of 2-butanone,
2
-pentanone, and benzophenone by ethylene glycol were
performed for both SLUG-21 and SLUG-22 (Table 3). A
significant size selectivity was observed for SLUG-21: the more
bulky benzophenone led to a yield of only 31%, less than one-
third that of 2-butanone. This result demonstrates that ketal
formation by SLUG-21 is likely occurring primarily in the 1D
3
4
pore channels along the b-axis. The relative dimensions
indicate that the cross-sectional channel area of SLUG-21 (ca.
4
.1 Å × 3.7 Å) is large enough to accommodate both 2-butanone
(
ca. 3.5 Å × 1.6 Å × 5.1 Å) and 2-pentanone (ca. 3.8 Å × 1.8
Figure 9. PXRD (top) and FTIR (bottom) of (a) SLUG-22 (black); (b)
the solid product after 3 days exchange reaction between SLUG-22 and
Å × 6.8 Å). In contrast, benzophenone (ca. 7.8 Å × 5.2 Å ×
0
4 4
.1 M NaClO (SLUG-22-ClO , red); and (c) the solid product after 3 days
exchange reaction between SLUG-22-ClO
solution (blue).
4
and 0.1 M EDS sodium salt
(50) Banik, B. K.; Chapa, M.; Marquez, J.; Cardona, M. Tetrahedron Lett.
2005, 46, 2341–2343.
7
208 J. AM. CHEM. SOC. 9 VOL. 132, NO. 20, 2010

Ag(I)/Cu(I)-Based Cationic Metal-Organic Frameworks
A R T I C L E S
3
4
contains open 1D channels whereas the intercalated water
molecules of SLUG-22 are distributed throughout the inter-
lamellar region and should restrict access. Without accessible
channels for diffusion of the organic precursor into the structure,
surface catalysis would be expected. It is possible that surface
Lewis acid sites more easily lose their catalytic activity
compared to interior sites due to long-term exposure to the
chemical solvent, which would account for the drop in yield
versus runs for SLUG-22.
Conclusions
Two Ag(I)/Cu(I)-based cationic metal-organic frameworks
have been rationally synthesized using both an organic metal
linker and organic structure directing agent. Both materials
display efficient anion exchange as well as heterogeneous Lewis
acidity. SLUG-21 exhibits reversible anion exchange of its
charge-balancing organosulfonate anions for nitrate/perchlorate,
as well as a potentially applicable selective solid-state trapping
of group(VII) oxometal anions. Although SLUG-22 is not useful
for permanganate trapping, likely due to the lower redox stability
of Cu(I) versus Ag(I), nitrate and perchlorate can be reversibly
exchanged in high percentage. Cationic MOFs are potentially
useful for heterogeneous catalysis due to their positive charge
compared to the majority of reported MOFs. Our comparative
study into the structure-property behavior of two similar
extended frameworks on varying substrate size shows that 1-D
channels are important in obtaining size-selective, high yield,
recyclable heterogeneous catalysts. We are further exploring this
two-pronged approach of organic MOF linker and organosul-
fonate anionic template with other d-block metals to tune the
charge density as well as openness, which should further
enhance the anion exchange and catalytic properties of cationic
MOFs.
Figure 10. PXRD of SLUG-22: (a) as-synthesized; postcatalysis after one
(
b) and three (c) catalytic cycles of ketal formation.
Table 3. Conversion Yields for the Ketalization of Larger
Substrates
a
catalyst
ketone precursor
time (h)
conv. (%)
no catalyst
2-butanone
12
12
12
12
12
12
12
12
12
1
0
0
97
87
31
89
91
91
2
-pentanone
benzophenone
2-butanone
SLUG-21
SLUG-22
2
-pentanone
benzophenone
2-butanone
2
-pentanone
benzophenone
a
1
Conversion yield is determined by H NMR.
Acknowledgment. This research was supported by an NSF
Career Award (DMR-0506279). Crystallographic data were col-
lected through the SCrALS (Service Crystallography at Advanced
Light Source) program at the Small-Crystal Crystallography Beam-
line 11.3.1 of the Advanced Light Source (ALS), Lawrence
Berkeley National Laboratory. The ALS is supported by the U.S.
Department of Energy, Office of Energy Sciences, Materials Science
Division, under Contract DE-AC02-05CH11231.
4
.6 Å) is too bulky to diffuse into the 1D channel. Meanwhile,
no size selectivity was observed for SLUG-22: all three ketone
precursors formed in similar yields, indicating that SLUG-22
catalysis likely occurs on the crystal surface. Even so, these
yields are still higher than in the case of the conventional
5
0,51
homogeneous iodine.
Based on these heterogeneous catalytic results, we investi-
gated whether the structural differences between the two
frameworks might be responsible for the varying ketalization
mechanism and yields. Although both structures possess a
similar cationic charge and topology, there are slight structural
differences in the EDS anion orientation as well as the position
of the crystallographic water. In addition, the silver atoms are
larger. The percent solvent/guest accessible void volume was
Supporting Information Available: Ex situ PXRD, additional
anion exchange optical micrographs, FTIR spectra and PXRD
pattern of anion exchange between SLUG-21 and perrhenate
1
and permanganate and additional discussion, H NMR spectra,
and crystallographic information file (CIF) including refinement
parameters and full bond distances. This material is available
free of charge via the Internet at http://pubs.acs.org.
52
estimated by PLATON. SLUG-21 possesses 16% void space,
compared with only 12% for SLUG-22. Moreover, SLUG-21
JA102134C
(
52) Spek, A. L. PLATON, A Multipurpose Crystallographic Tool: Utrecht,
(
51) Ren, Y. M.; Cai, C. Tetrahedron Lett. 2008, 49, 7110–7112.
The Netherland, 2007.
J. AM. CHEM. SOC. 9 VOL. 132, NO. 20, 2010 7209