Reusable oxidation catalysis using metal-monocatecholato species in a robust metal-organic framework

Article abstract of DOI:10.1021/ja411627z

An isolated metal-monocatecholato moiety has been achieved in a highly robust metal-organic framework (MOF) by two fundamentally different postsynthetic strategies: postsynthetic deprotection (PSD) and postsynthetic exchange (PSE). Compared with PSD, PSE proved to be a more facile and efficient functionalization approach to access MOFs that could not be directly synthesized under solvothermal conditions. Metalation of the catechol functionality residing in the MOFs resulted in unprecedented Fe-monocatecholato and Cr-monocatecholato species, which were characterized by X-ray absorption spectroscopy, X-band electron paramagnetic resonance spectroscopy, and ⁵⁷Fe M?ssbauer spectroscopy. The resulting materials are among the first examples of Zr(IV)-based UiO MOFs (UiO = University of Oslo) with coordinatively unsaturated active metal centers. Importantly, the Cr-metalated MOFs are active and efficient catalysts for the oxidation of alcohols to ketones using a wide range of substrates. Catalysis could be achieved with very low metal loadings (0.5-1 mol %). Unlike zeolite-supported, Cr-exchange oxidation catalysts, the MOF-based catalysts reported here are completely recyclable and reusable, which may make them attractive catalysts for 'green' chemistry processes.

Full text of DOI:10.1021/ja411627z

Article
pubs.acs.org/JACS
Reusable Oxidation Catalysis Using Metal-Monocatecholato Species
in a Robust Metal−Organic Framework
Honghan Fei,† JaeWook Shin,‡ Ying Shirley Meng,‡ Mario Adelhardt,§ Jorg Sutter,§ Karsten Meyer,§
̈
and Seth M. Cohen*,†
‡
†Department of Chemistry and Biochemistry and Department of NanoEngineering, University of California, San Diego, La Jolla,
California 92093, United States
§Department of Chemistry and Pharmacy, Inorganic Chemistry, Friedrich-Alexander University Erlangen − Nurnberg (FAU),
̈
Egerlandstrasse 1, 91058 Erlangen, Germany
S
* Supporting Information
ABSTRACT: An isolated metal-monocatecholato moiety has
been achieved in a highly robust metal−organic framework
(MOF) by two fundamentally different postsynthetic strat-
egies: postsynthetic deprotection (PSD) and postsynthetic
exchange (PSE). Compared with PSD, PSE proved to be a
more facile and efficient functionalization approach to access
MOFs that could not be directly synthesized under
solvothermal conditions. Metalation of the catechol function-
ality residing in the MOFs resulted in unprecedented Fe-
monocatecholato and Cr-monocatecholato species, which
were characterized by X-ray absorption spectroscopy, X-band
electron paramagnetic resonance spectroscopy, and 57Fe
Mossbauer spectroscopy. The resulting materials are among the first examples of Zr(IV)-based UiO MOFs (UiO = University
̈
of Oslo) with coordinatively unsaturated active metal centers. Importantly, the Cr-metalated MOFs are active and efficient
catalysts for the oxidation of alcohols to ketones using a wide range of substrates. Catalysis could be achieved with very low metal
loadings (0.5−1 mol %). Unlike zeolite-supported, Cr-exchange oxidation catalysts, the MOF-based catalysts reported here are
completely recyclable and reusable, which may make them attractive catalysts for ‘green’ chemistry processes.
species.27,28 Immobilized catechol ligands in MOFs provide a
INTRODUCTION
■
platform for isolating highly unsaturated mono(catecholato)
metal complexes. Recently, polymeric materials bearing metal-
catecholate groups have been studied in porous organic
polymers using a cobalt-catalyzed acetylene trimerization
strategy.29−31 However, studies of such catalytically active
metalated-catechol functionalities in MOFs are very rare.
Nguyen and co-workers used PSM to incorporate single-site
vanadyl(monocatecholate) moieties into two MOFs; however,
the efficiency of these catalysts was limited, largely due to low
metal loadings and framework instability.32,33
Previously, photochemically driven PSD of 2-nitrobenzyl
protection groups was used to generate free catechol groups in
a UMCM (UMCM = University of Michigan Crystalline
Material) framework.16 Herein, we employed two fundamen-
tally different approaches (i.e., PSD and PSE) to synthesize the
first highly robust MOF (UiO-66, UiO = University of Oslo)
bearing isolated monocatecholato metal sites on the strut of
organic linkers. Both strategies result in good loading of metal
species,32 while maintaining the high porosity of the MOFs and
accessibility to the open metal sites. In particular, PSE allows
Metal−organic frameworks (MOFs) are a class of microporous
crystalline materials gathering substantial attention not only for
their vast array of structures1 but also due to their applications
in gas storage,2 separations,3 catalysis,4 molecular sensing,5 and
drug delivery.6 Accessible and coordinatively unsaturated metal
sites residing in MOFs are desirable for gas or substrate binding
within the pores, opening up a range of technological
applications from improved gas storage to catalysis.7−9 In
most cases, it is challenging to obtain free metal-binding sites
via direct solvothermal synthesis, though some sites (e.g.,
carboxylic groups) have been obtained for a small number of
MOFs.10,11 Postsynthetic methods are synthetic alternatives to
obtain metal-binding sites in MOFs. These include methods
such as postsynthetic modification (PSM),12−15 postsynthetic
deprotection (PSD),16,17 and postsynthetic exchange (PSE)
(also termed SALE = solvent-assisted linker exchange).18−25
Despite the increased use of these synthetic methods, to the
best of our knowledge, no report has described the use of PSE
to introduce metal-chelating groups into MOFs.
Catechol is one of the most widely used chelators in
coordination chemistry.26 The standard solution coordination
chemistry of catechol ligands generally leads to coordinatively
saturated bis(catecholato) or tris(catecholato) homoleptic
Received: November 14, 2013
Published: March 5, 2014
© 2014 American Chemical Society
4965
dx.doi.org/10.1021/ja411627z | J. Am. Chem. Soc. 2014, 136, 4965−4973

Journal of the American Chemical Society
Article
Scheme 1. Synthesis of UiO-66-CAT via PSD and PSE
good control over the number of metal-binding sites and for a
higher density of catalytic sites when compared to PSD. The
metal-catecholate coordination environment (Fe, Cr) was
characterized by X-ray absorption spectroscopy (XAS) and
diffuse reflectance UV−vis electronic spectroscopy. To the best
of our knowledge, this is the first example of these Zr(IV)-
based MOFs decorated with a coordinatively unsaturated and
catalytically active metal site. The functionalized UiO-66
platform containing metal-catecholate species was further
employed as an oxidation catalyst. The immobilized Cr(III)-
monocatecholato UiO materials exhibit efficient catalysis of
secondary alcohol oxidation with several substrates using t-butyl
hydroperoxide (TBHP) or hydrogen peroxide (H2O2) as a
sacraficial oxidant. This is a rare example using MOFs to
produce a nonprecious metal catalyst that is highly recyclable,
thereby holding promise for “green” chemistry applications.
1
RESULTS AND DISCUSSION
Figure 1. H NMR of UiO-66-(OBnNO2)2 (top) and UiO-66-CAT
(via PSD) (bottom). (OBnNO2)(OH)-bdc with a single nitrobenzyl
group cleaved is labeled with squares, catbdc is labeled with circles, and
DMF solvent is labeled with asterisks. The bdc ligand is also labeled in
the spectra.
■
Synthesis of UiO-66-CAT via PSD. 2,3-Bis((2-
nitrobenzyl)oxy)-1,4-benzene dicarboxylic acid ((OBnNO)2-
bdc) was synthesized as previously described (Scheme 1).16
UiO-66-(OBnNO2)2 was prepared using solvothermal con-
ditions containing a 1:1 molar ratio mixture of (OBnNO2)2-bdc
and unfunctionalized 1,4-benzene dicarboxylic acid (bdc) with
ZrCl4 and acetic acid (as a modulator) at 100 °C in DMF for
24 h. The resulting off-white crystalline solids were washed with
copious amounts of DMF and methanol, followed by activation
25% were completely cleaved to catbdc (Scheme 1). Powder X-
ray diffraction (PXRD) confirmed the resulting MOFs adopt
the UiO-66 framework (Figure 2). Irradiated with 365 nm light
for 24 h, UiO-66-(OBnNO2)2 was found to achieve nearly
quantitative deprotection to yield UiO-66-CAT (via PSD).16,34
1H NMR of digested MOFs show a molar ratio of 2.2:1
(bdc:catbdc), indicating that (OBnNO2)2-bdc is less effectively
incorporated into UiO-66 than bdc (Figure 1). Electrospray
ionization-mass spectrometry (ESI-MS) of digested UiO-66-
1
under vacuum. Examination by H NMR spectroscopy upon
digestion of MOFs with HF in d6-DMSO revealed bdc,
(OBnNO2)2-bdc, and a mix of mono- and fully deprotected
(OBnNO2)2-bdc (Figure 1). Within the resulting solids, ∼70%
of (OBnNO2)2-bdc had a single nitrobenzyl group cleaved, and
4966
dx.doi.org/10.1021/ja411627z | J. Am. Chem. Soc. 2014, 136, 4965−4973

Journal of the American Chemical Society
Article
Figure 2. PXRD of UiO-66-(OBnNO2)2 and UiO-66-CAT (via PSD) (left). SEM image of UiO-66-CAT (via PSD) (right).
CAT (via PSD) further confirmed the presence of both bdc and
catbdc in PSD product (Figure S2). PXRD suggested UiO-66-
CAT (via PSD) maintained the same underlying framework
structure as UiO-66-(OBnNO2)2 (Figure 2). Scanning electron
microscopy (SEM) suggested high phase purity with
monodisperse, spherical particles (∼150−200 nm, Figure 2).
Synthesis of UiO-66-CAT via PSE. Recently, PSE has
become a facile and efficient strategy to functionalize MOFs
under mild conditions.18−22,35 More importantly, using PSE to
introduce a metal-chelating group into a robust MOF has not
been investigated. Herein, we expanded PSE as a general
functionalization method to introduce catechol groups into the
UiO-66 lattice.
UiO-66 was synthesized in DMF containing a 1:1 molar ratio
mixture of bdc and ZrCl4 with acetic acid (as a modulator) at
120 °C for 24 h. The resulting crystalline solids were washed
with copious amounts of MeOH and activated under vacuum.
PSE was performed by exposing UiO-66 powder to a DMF/
H2O solution of catbdc for 2 d at 85 °C. After separating the
solids by centrifugation and washing with fresh MeOH, the
1
presence of catbdc in UiO-66 was confirmed by H NMR of
the MOFs digested with dilute HF in CD3OD (Figure 3). The
degree of functionalization was controllable based on the
amount of catbdc present in the starting solution (from 18% to
75%, using 0.5 to ∼5 equiv of catbdc, Figure S3). ESI-MS of
digested UiO-66-CAT (via PSE) further confirmed catbdc in
the PSE product (Figure S4). Further increasing the ratio of
catbdc in the precursor solution did not significantly enhance
the incorporation; however, incubating the 75% functionalized
UiO-66-CAT (via PSE) with a fresh catbdc solution for a
second round of PSE gave >90% incorporation of catbdc.
Importantly, the high crystallinity of exchanged UiO-66-CAT
(via PSE) was confirmed by PXRD (Figure 3). As shown by
SEM, UiO-66-CAT containing 75% catbdc (via PSE) exhibited
a homogeneous UiO particle morphology, with no evidence of
any other crystalline or amorphous phase (Figure S5).
1
Figure 3. H NMR (top) and PXRD (bottom) of digested UiO-66-
CAT synthesized by PSE with 1 equiv of UiO-66 exposed to 0.5
(black), 1 (red), 2 (blue), 3 (magenta), or 5 (cyan) equiv of catbdc.
During digestion of the MOFs for 1H NMR analysis, 1 equiv of Br-bdc
(denoted with circles) was added as a quantitative internal standard.
Attempts to directly synthesize UiO-66-CAT under numerous
solvothermal conditions were unsuccessful, including attempts
using a mixed ligand strategy with bdc and catbdc in an
equimolar ratio (Table S3).
To confirm that catbdc was incorporated into the UiO
framework (and not just included in the pores), experiments
were performed examining the supernatant of the reaction
mixture. PSE between UiO-66 and catbdc was performed in
D2O, and H NMR of the supernatant indicated the presence
of released bdc after 24 h of PSE at 85 °C (Figure S6). UiO-66
bdc under the same experimental conditions. In addition, UiO-
66-CAT (via PSE) exhibited Brunauer−Emmett−Teller (BET)
surface areas in the range of 968 27 m2/g to 1206 11 m2/g
(depending on the degree of catbdc incorporation), which is
close to the BET surface area of pristine UiO-66 (1403 28
m2/g, N2 gas at 77K) (Table S1), again excluding the possibility
1
in D2O in the absence of catbdc did not show any leaching of
4967
dx.doi.org/10.1021/ja411627z | J. Am. Chem. Soc. 2014, 136, 4965−4973

Journal of the American Chemical Society
Article
that the catbdc is included in and blocking the pores of the
MOF. During digestion analysis of these MOFs, 1 equiv (with
respect to the total bdc content of the MOFs prior to PSE) of
internal standard (2-bromoterephthalic acid, Br-bdc) was
added. The NMR analysis (Figure 3) shows that the combined
amount of all bdc linkers in the MOF is unchanged throughout
PSE process. Based on these NMR and BET observations,
along with high degree of catbdc incorporation, the data
support a ligand metathesis process and argue against simple
inclusion of catbdc into the MOF pores or attaching onto the
particle surfaces. Finally, it is also important to note that the
PSE process could also be carried out at room temperature,
with an expectedly lower degree of catbdc functionalization,
which suggests a solid-state exchange mechanism and argues
against a dissolving/reforming process.
molecules that were best fit as three methanol (gave a better fit
than water) and one methoxide ligand (based on required
charge balance for Fe3+, Figure 4). This appears to be a
Both PSD and PSE appear to be viable strategies to
incorporate catechol species into the robust UiO-66 framework.
Compared with PSD, PSE exhibited two advantages for
functionalization. First, as demonstrated above, PSE could be
readily controlled to modulate the degree of catechol
functionalization. However, PSD only allowed for <30%
functionalization of catbdc into the UiO-66 lattice, perhaps
due to the steric influence of the (OBnNO2)2-bdc precursor.
Increasing the (OBnNO2)2-bdc:bdc ligand ratio in the
solvothermal synthesis led to incomplete PSD or even
appearance of an amorphous phase from the reaction mixture.
Second, the PSD strategy required a three-step synthesis of
(OBnNO2)2-bdc, a MOF synthesis step, and a photochemical
PSD step as well with an overall yield of ∼35% (Scheme 1). In
contrast, UiO-66-CAT using PSE was prepared in only two
steps (i.e., UiO-66 synthesis and the PSE reaction) giving >90%
yield. Overall, the PSE strategy to incorporate the catechol
units into UiO-66 proved to be a more convenient, efficient,
and controllable approach.
Metalation of UiO-66-CAT with Fe. UiO-66-CAT
prepared via the PSD approach described above (Scheme 1)
was metalated in a postsynthetic manner with Fe. The
metalation of the catechol units residing in the framework
was performed by incubating UiO-66-CAT (via PSD) in an
aqueous solution of Fe(ClO4)3, whereupon the particles
became dark-brown within several minutes. After 24 h, UiO-
66-FeCAT was washed extensively with fresh MeOH for 3 d,
followed by vacuum activation. Energy-dispersed X-ray spec-
troscopy (EDX) confirmed an atomic ratio of 1:0.21 (Zr:Fe) in
Figure 4. Fourier transformed EXAFS spectrum of UiO-66-FeCAT
(top) and UiO-66-CrCAT (bottom). The black solid line is the
experimental data, the dashed line is the best fit to data, and the dotted
line is fitting window. Inset in top figure: XANES of UiO-66-FeCAT
metalated by Fe(ClO4)3 (a, solid line), UiO-66-FeCAT metalated by
Fe(CF3SO3)3 (b, dashed line), Fe(ClO4)3 (c, dotted line), and
Fe(CF3SO3)3 (d, dashed dotted line).
1
these metal-loaded samples (Figure S7). Based on H NMR
integration showing ∼31% of the organic linkers are catbdc
(overall formula Zr6O4(OH)4(bdc)4.1(catbdc)1.9, Figure 1) the
measured metal ratio suggests ∼68% of the catbdc ligands are
metalated. Importantly, no Cl was detected by EDX, indicating
that free or weakly bound Fe(ClO4)3 species were removed by
washing. Again, PXRD and SEM confirmed retention of the
UiO-66 crystallinity and phase purity (Figures S1 and S8). A
BET surface area of 1134 67 m2/g was obtained using N2 gas
at 77 K, which supports that the Fe(III) ions are bound to the
framework and not forming particles or otherwise blocking the
pores of the MOF.
Fe K-edge extended X-ray absorption fine structure spec-
troscopy (EXAFS) was performed on UiO-66-FeCAT to
investigate the coordination environment of the Fe(III) centers
on the strut of UiO-66 framework. The data were best fit using
an octahedral geometry at the Fe3+ center, with bidentate
coordination from the catecholato ligand and the remainder of
the coordination sphere being comprised of four solvent
reasonable formulation as a variety of Fe3+-based complexes
with a methoxide ligand have been studied and crystallo-
graphically characterized.36 The data fitting indicated Fe3+
bonds to two oxygen atoms from the catecholato ligand at a
distance of 1.98(3) Å and two nearest-neighbor carbon atoms
from the catecholato ligand at a distance of 2.60(5) Å. These
distances are in good agreement with crystallographic data of
Fe3+-catecholato species.37,38 In order to further characterize
the binding of Fe3+ by catechol, we employed another metal
precursor Fe(CF3SO3)3 to metalate UiO-66-CAT (via PSD)
under identical conditions as used for Fe(ClO4)3. X-ray
absorption near-edge structure (XANES) shows both Fe-
(ClO4)3 and Fe(CF3SO3)3 treated UiO-66-CAT (via PSD)
display similar chemical environments around Fe3+, even
4968
dx.doi.org/10.1021/ja411627z | J. Am. Chem. Soc. 2014, 136, 4965−4973

Journal of the American Chemical Society
Article
though the iron precursors are readily distinguishable by
XANES (Figure 4 top). Also, EDX of metal-loaded samples via
Fe(CF3SO3)3 suggest negligible remaining sulfur but a similar
atomic ratio of Zr:Fe (1:0.18) as Fe(ClO4)-treated UiO-66-
CAT (via PSD). Finally, formation of an isolated mono-
catecholato Fe3+ species in UiO-66-FeCAT was confirmed by
diffuse reflectance solid-state UV−vis spectroscopy (Figure S9).
UiO-66-FeCAT exhibited a characteristic broad absorption
band in the range of ∼700−750 nm, characteristic of
monocatecholato Fe3+ species.39,40 In contrast, both Fe3+
precursors and UiO-66-CAT (via PSD) did not show any
significant absorption >600 nm.
species in UiO-66-FeCAT (Figure 5). Data fitting results in an
isomer shift (δ) of 0.54(1) mm·s−1 and a quadrupole splitting
parameter, ΔEQ, of 0.91(1) mm·s−1. These values are
characteristic of high-spin Fe(III) species and are in good
agreement with published values for octahedral high-spin
Fe(III) catecholate complexes.42−44 Accordingly, EPR and
Mossbauer spectroscopy supports the proposed structural motif
̈
determined by EXAFS and diffuse reflectance solid-state UV−
vis spectroscopy.
Metalation of UiO-66-CAT with Cr. UiO-66-CAT
prepared via the PSE approach described above (Scheme 1)
was metalated in a postsynthetic manner with Cr. Metalation of
the catechol unit in UiO-66-CAT (51% catbdc) was carried out
in aqueous K2CrO4 solution under weakly acidic condition (pH
= 3). After 24 h, the resulting dark-brown solids were isolated
by centrifugation and washed by copious amount of deionized
water until the supernatant was colorless. Followed by
immersion in fresh MeOH for 3 d and vacuum drying, UiO-
66-CrCAT was found to be highly crystalline and phase pure by
PXRD and SEM (Figures S12 and S13). EDX suggests an
1:0.55 atomic ratio of Zr:Cr, indicating a nearly quantitative
metalation of catechol units in the framework with overall
formula as Zr6O4(OH)4(bdc)3(Crcatbdc)3. The slightly higher
Cr incorporation is likely due to trapped K2CrO4 in the UiO-66
lattice. However, a very low potassium (atomic ratio Cr:K =
1:0.07) was observed in EDX, indicating a vast majority of Cr
was strongly bound with the catechol sites. A control
experiment was performed by incubation of unfunctionalized
UiO-66 in aqueous K2CrO4 solution under the same condition
(Figure S14). After extensive washing with fresh H2O and
MeOH, EDX of resulting solids confirms the presence of
trapped potassium and chromium, with an atomic ratio of
1:0.04 (Zr:Cr). The BET surface area of UiO-66-CrCAT was
708 24 m2/g (N2 at 77K), which is somewhat lower than the
The iron centers in UiO-66-FeCAT were further charac-
terized by electron paramagnetic resonance (EPR) and 57Fe
Mossbauer spectroscopy. Variable-temperature X-band EPR
̈
measurements (9−289 K) on solid-samples of UiO-66-FeCAT
(Figures 5, S10, and Figure S11) show a rhombic S = 5/2 signal
pristine UiO-66 (1403
28 m2/g), likely due to the higher
degree of functionality and metalation than the iron samples
described above.
X-ray photoelectron spectroscopy (XPS) of UiO-66-CrCAT
consists of two Cr peak contributions 587 and 577.2 eV
corresponding to 2p1/2 and 2p3/2 energy levels, respectively.
These values compare well with binding energies in Cr(NO3)3
(2p1/2 587 eV and 2p3/2 577.3 eV); however, they are distinctly
different from K2CrO4 (2p1/2 589 eV and 2p3/2 579.9 eV,
Figure S15). This suggested that Cr(VI) is reduced to Cr(III)
during the acidic metalation reaction conditions. Therefore, we
tentatively assigned the metalated species as Cr3+ centers with
an octahedral coordination environment, bound to the
catecholato ligand with four weakly bound solvent molecules
and/or counteranion. In order to confirm this coordination, Cr
K-edge EXAFS was performed on UiO-66-CrCAT to study the
coordination environment of the Cr3+ sites. A search in
Cambridge structure database does not give a deposited crystal
structure with a free methanol solvent molecule bound to a Cr-
catechol complex. Instead, we used a structure with a bidentate
catecholato ligand along with three solvent water molecules and
one hydroxide anion (required for charge balance) to perform
the EXAFS data fitting. Fitting with this model gives a
reasonable R-factor of 0.0170 (Figure 4 bottom). The data
fitting gave atomic distance between Cr3+ and two oxygen
atoms of 1.97(6) Å and an extended bond length between Cr3+
to solvent water molecules at 2.03(4) Å. Both distances are in
good agreement with crystallographic data of known Cr3+-
catecholato structures.45 The slight disagreement between the
Figure 5. EPR spectrum of UiO-66-FeCAT (top) recorded in solid
state at 9 K with simulation. Frequency 8.987052 GHz, power 0.3
mW, modulation width 2 mT, time constant 0.1 s. Parameters for
simulation: Lorentzian line shape; species 1 (99.8%): g1 = 2.02, g2 =
4.25, g = 8.80; species 2 (0.2%): g = 1.997. Mossbauer spectrum of
̈
3
UiO-66-FeCAT (bottom), recorded (for 8 days) at 77 K: δ = 0.54(1)
mms−1, ΔEQ = 0.91(1) mms−1, Γfwhm = 0.54(1) mms−1.
with g-values at 8.80, 4.25, and 2.02. This signal accounts for
99.8% of the total signal intensity. In addition, a sharp isotropic
signal is observed at g = 1.997 (0.2% intensity), which may
originate from a minute semiquinone radical impurity (S = 1/2)
from the catbdc ligand. The major S = 5/2 signal confirms the
expected Fe(III) d5 high-spin state of the metal centers
incorporated in UiO-66-FeCAT.41 In addition to EPR, a zero-
field 57Fe Mossbauer spectrum of UiO-66-FeCAT recorded at
̈
77 K in the solid-state of an identical sample (containing
naturally abundant 57Fe) shows a single, well-resolved quadru-
pole doublet, further proving the existence of one distinct iron
4969
dx.doi.org/10.1021/ja411627z | J. Am. Chem. Soc. 2014, 136, 4965−4973

Journal of the American Chemical Society
Article
experimental data and fit may be due to a small amount of
chromate trapped inside the MOFs (e.g., advantageous binding
not bound to the catecholato centers, see EDX data above) as
well as the possible presence of MeOH as solvent molecules
bound to some of the Cr3+ centers. The hypothesis that a trace
amount of chromate may be contributing to the data is also
evidenced by XANES, which shows a characteristic Cr6+ pre-
edge feature from 1s−3d electronic transition (Figure S16).46
Oxidation Catalysis. Ketones are among the most
important chemicals as final products or as intermediates in
the pharmaceutical and chemical industry.47 The liquid-phase
oxidation of secondary alcohols is the most widely used
synthetic method to produce ketones. However, the classical
approach for this reaction requires stoichiometric amount of
Cr6+ or Mn7+ oxidants (e.g., Jones oxidation), thus generating
equal amounts of metal waste that is environmentally
disadvantageous.48 In a constant search for greener methods,
many catalytic systems based on expensive noble metals, with
high catalyst loadings, and costly/toxic additives have been
explored.49,50 Both homogeneous and heterogeneous catalytic
systems have been developed as well as metal nanoparticles;
however, heterogeneous catalysts are preferable in industry due
to their easy separation and reusability.51 High-valent Cr-
species are among the most extensively studied catalysts for
oxidation reactions.52 Efforts were made to prepare heteroge-
neous catalysts with Cr active sites via Cr-exchanged molecular
sieves (i.e., aluminophosphates and silicates);53 however,
leaching of Cr from the zeolite matrix during liquid-phase
oxidation presented a problem.54
Utilizing MOFs as heterogeneous alcohol oxidation catalysts
with hydroperoxides or molecular oxygen has gained increasing
attention for green chemistry processes.55,56 Noble-metal
nanoparticles encapsulated within MOFs also exhibit efficient
aerobic oxidation of alcohol.57−59 HKUST-1 (with the formula
Cu3(BTC)2, where BTC = 1,3,5-benzene tricarboxylate) in the
presence of TEMPO under basic conditions gives moderate
activity for the oxidation of benzylic alcohols but shows poor
activity for oxidation of aliphalic, secondary, and cyclic
alcohols.60 PSM of open bipyridine sites in MOF-253 with an
organometallic Ru complex gives efficient alcohol oxidation
with PhI(OAc)2 as the oxidant.61 Another Cu-based MOF,
constructed with paddlewheel secondary-building units (SBUs)
and a 1,4-cyclohexanedicarboxylate linker, was found to have
moderate activity (10 to ∼55% yield) for both primary and
secondary alcohol oxidation with H2O2.62 Herein, UiO-66-
CrCAT is shown to be a rare example of a heterogeneous
catalysts containing nonprecious metal active centers that are
efficient, clean, and recyclable for secondary alcohol oxidation
with less polluting sacrificial oxidants (e.g., TBHP or H2O2) for
“green” chemistry catalysis.
K2CrO4 or other Cr species is not the source of the high
catalytic activity of UiO-66-CrCAT. Moreover, Cr(acac)3 (acac
= acetylacetonate) and K2CrO4 were used as homogeneous
catalysts; however, 1.5 mol % Cr of either compound gave
<40% yield after 24 h. Another homogeneous control was
carried out using a combination of 1.5 mol % of K2CrO4 and
1.5 mol % catbdc as catalyst, which also produced <40% yield
under the same reaction condition.
a
Table 1. Cr-Catalyzed Oxidation of 2-Heptanol
a
Reaction conditions: 1 mmol 2-heptanol; 1.3 mmol TBHP in 1 mL
b
c
chlorobenzene (or neat) at 70 °C. Based on GC-MS analysis. UiO-
66 was treated with K2CrO4 at pH = 3 and rinsed extensively with
fresh H2O and MeOH before testing as a catalyst.
Eight secondary alcohols, including aliphatic, cyclic, benzylic,
and a steroid alcohol were used as substrates for oxidation
(Table 2). Among them, the majority of substrates gave nearly
quantitative yields in 8 to 24 h using 1 mol % Cr in the form of
UiO-66-CrCAT. Catalysis under solvent-free conditions gave
comparable yields with the same catalyst loadings. Control
reactions of catalytic amount of homogeneous Cr catalysts
(Cr(acac)3 or K2CrO4) and UiO-66-CAT (via PSE without
metalation) under the same conditions gave significantly lower
yields (<50%). Compared to other Cr-based heterogeneous
catalysts, the MOF system described here is significantly
improved. For example, a Cr-exchanged zeolite catalyst53 gave
yields of 79−95% with benzylic alcohols but required 4 equiv of
TBHP.57 Moreover, UiO-66-CrCAT shows no leaching of Cr,
which was also a limitation for the zeolite catalyst. When
compared to other MOF systems, the turnover frequencies
(TOFs) of UiO-66-CrCAT are lower than systems that use
precious-metal nanoparticles embedded in MOFs. Also, when
compared to a system employing Ru and a more toxic oxidant
(PhI(OAc)2), UiO-66-CrCAT gives comparable yields but at
∼10-fold less TOF.61 However, the yields and TOFs of UiO-
66-CrCAT are significantly higher than other MOFs that use
nonprecious metal catalysts, such as Cu or Co.62 Although
some of nonprecious metal homogeneous catalysts (mostly
copper based)63 can give good yields using air or O2 as the
oxidant, they often require much higher metal loadings (5−
10%)64 and toxic/costly additives (e.g., TEMPO65 or
dialkylazodicarboxylate)66 as cocatalysts. In addition, efforts to
produce heterogeneous versions of these copper catalysts are
rather limited.51,67
Secondary alcohol oxidation catalysis using UiO-66-CrCAT
was initially studied with 2-heptanol as the substrate using 1.3
equiv of TBHP as an oxidant and chlorobenzene as the solvent
at 70 °C. Using UiO-66-CrCAT (only 1 mol % Cr) gave nearly
quantitative oxidation to 2-heptanone in 24 h. More
importantly, performing the catalysis in a “green,” solvent-free
environment significantly enhanced the rate of reaction, with
the same oxidation complete in 8 h using only 0.5 mol % Cr
loading. Meanwhile, the same mol % of unfunctionalized UiO-
66 and UiO-66-CAT (via PSE without metalation) only gave
8% and 12% yield of 2-heptanone after 24 h, respectively. In
addition, UiO-66 treated with K2CrO4 solution, rinsed, and
used as a catalyst only gave ∼12% yield, indicating that trapped
To test UiO-66-CrCAT with a complex substrate, 5α-
cholestan-3β-ol was oxidized to the corresponding ketone in 12
h in nearly quantitative yield. The molecular size of this
cholesterol derivative is approximately 16.4 × 8.1 × 6.5 Å as
determined by space-filling models (using the software package
Diamond v3.1e). Although two dimensions of the substrate can
fit within the pores of UiO-66 (pore diameters of ∼8 Å for the
4970
dx.doi.org/10.1021/ja411627z | J. Am. Chem. Soc. 2014, 136, 4965−4973

Journal of the American Chemical Society
Article
encouraging, giving a yield of ∼39% for 2-heptanol. However,
control reactions with H2O2 or TBHP alone (no MOF catalyst)
gave yields of 0 and 8%, respectively, showing that UiO-66-
CrCAT is still promoting this reaction.
Table 2. Cr-Catalyzed Oxidation of Secondary Alcohols
Using TBHP (equiv listed) As Oxidants
a
Table 3. Cr-Catalyzed Oxidation of Secondary Alcohols
a
Using H2O2 (equiv listed) As Oxidants
a
Reaction conditions: 1 mmol sec-alcohol; 2 mmol H2O2 in 1 mL
b
c
CH3CN at 70 °C. Based on GC-MS analysis. TOF calculated as
mol(product)/[mol(catalyst) × (reaction time)].
Heterogeneity of UiO-66-CrCAT was confirmed by a hot
filtration of the catalysts after 2 h of reaction with 2-heptanol,
which resulted in negligible additional yield of product up to 22
h after filtration (Figure S18). This suggested that UiO-66-
CrCAT was a true heterogeneous catalyst, and no catalytically
active species were released into solution. The chemical
robustness of UiO-66-CrCAT was confirmed by its incubation
in aqueous solution for 24 h and its thermal stability up to
∼250 °C was also evidenced by thermogravimetric analysis
(Figures S19 and S20). Leaching of Cr into the solution was
examined by inductive coupled plasma-optical emission spec-
troscopy (ICP-OES, Quantitative Technologies Inc., White-
house, NJ), which showed that <0.1 ppm Cr was present in the
reaction solution. Overall, the robust UiO-66 material exhibited
clean catalysis when compared to Cr-containing zeolites, which
are often susceptible to solvolysis by polar solvents and
extensive leaching of the metal species.54,68,69
To examine recyclability, experiments were performed using
the same batch of 1 mol % UiO-66-CrCAT for the oxidation of
2-heptanol over five runs. Between each run, the catalysts were
simply separated by centrifugation, washed with MeOH, and
dried under vacuum at room temperature. No decrease of yield
(∼99%) was observed between the five runs, highlighting the
robust nature of the Cr-grafted MOF catalyst. Characterization
of UiO-66-CrCAT after catalysis showed the MOF still
possessed high crystallinity as evidenced by PXRD and SEM
(Figure S21). The disappearance of Cr6+ pre-edge features in
the XANES spectrum after one round of catalysis indicated that
the residual Cr6+ observed in freshly prepared UiO-66-CrCAT
was removed or reduced to Cr3+ (Figure S20). Based on XPS
and EXAFS analysis, no notable change in the coordination
environment of the Cr centers (other than the aforementioned
disappearance of trace Cr6+) was observed after one-cycle of 2-
heptanol oxidation (Figures S22 and S23), again suggesting that
the catalytic species is highly robust.
a
Reaction conditions: 1 mmol sec-alcohol, 1.1 (or 1.3) mmol TBHP in
b
1 mL chlorobenzene (or neat) at 70 °C. Based on GC-MS analysis.
c
TOF calculated as mol(product)/[mol(catalyst) × (reaction time)].
d
TOF in parentheses was calculated as TOF for the first 2 h catalysis
reaction.
tetrahedral cages and ∼11 Å for the octahedral cages), the
aperture size connecting these pore cages is only ∼6 Å, making
interior access by 5α-cholestan-3β-ol unlikely. In light of this,
absorption experiment was carried out between UiO-66-CAT
(via PSE) and several substrates (2-heptanol, cyclooctanol, and
1-phenylethanol) at 70 °C. After washing the surface of
particles with methaol, digestion of the MOFs with dilute HF in
CD3OD or d6-DMSO revealed 2-heptanol and cyclooctanol
inclusion in the MOFs (Figure S17). However, incubating the
same MOF in saturated solution of 5α-cholestan-3β-ol in
chlorobenzene under the same conditions did not show any
substrate inclusion. We anticipate that 5α-cholestan-3β-ol is too
large to diffuse into the pore cage and that catalysis with this
substrate likely occurs on the surface of UiO particles. The
minimal difference in yields and TOFs (Table 2) observed with
different substrates suggests that catalysis maybe be occurring
in both the interior and on the surface of the MOF with
comparable efficiencies.
Compared with t-butyl hydroperoxide as an oxidant, catalysis
with H2O2 is a more sustainable alternative, giving water as the
sole byproduct. Alcohol oxidation was performed using 2 equiv
of 30% H2O2 in aqueous solution as the oxidant with
acetonitrile as the solvent. Excellent yields (>95%) were
achieved for three examples of benzylic oxidation. However,
attempts to expand to aliphatic alcohol oxidations were less
In summary, the catalysis presented here are the first
examples in MOFs exhibiting efficient, recyclable, and “green”
alcohol oxidation with nonprecious metal active species.
Excellent yields were achieved with eight different secondary
alcohols, including aliphatic, cyclic, and benzylic alcohols. The
oxidations were performed using a very low metal loading
4971
dx.doi.org/10.1021/ja411627z | J. Am. Chem. Soc. 2014, 136, 4965−4973

Journal of the American Chemical Society
Article
Zhong, C.; Serre, C.; Weireld, G. D.; Maurin, G. Angew. Chem., Int. Ed.
2013, 52, 10316.
(12) Hwang, Y. K.; Hong, D.-Y.; Chang, J.-S.; Jhung, S. H.; Seo, Y.-
K.; Kim, J.; Vimont, A.; Daturi, M.; Serre, C.; Ferey, G. Angew. Chem.,
Int. Ed. 2008, 47, 4144.
(0.5−1 mol %), in some cases under neat conditions and also in
some cases using H2O2 as an environmentally benign oxidant.
Moreover, the robustness of the UiO-66 framework allowed for
easy recovery and reuse without leaching of Cr.
(13) Tanabe, K. K.; Cohen, S. M. Angew. Chem., Int. Ed. 2009, 48,
7424.
(14) Cohen, S. M. Chem. Rev. 2012, 112, 970.
(15) Demessence, A.; D’Alessandro, D. M.; Foo, M. L.; Long, J. R. J.
Am. Chem. Soc. 2009, 131, 8784.
(16) Tanabe, K. K.; Allen, C. A.; Cohen, S. M. Angew. Chem., Int. Ed.
2010, 49, 9730.
(17) Shultz, A. M.; Farha, O. K.; Adhikari, D.; Sarjeant, A. A.; Hupp,
J. T.; Nguyen, S. T. Inorg. Chem. 2011, 50, 3174.
(18) Karagiaridi, O.; Bury, W.; Sarjeant, A. A.; Stern, C. L.; Farha, O.
K.; Hupp, J. T. Chem. Sci. 2012, 3, 3256.
(19) Burnett, B. J.; Barron, P. M.; Hu, C. H.; Choe, W. J. Am. Chem.
Soc. 2011, 133, 9984.
(20) Karagiaridi, O.; Lalonde, M. B.; Bury, W.; Sarjeant, A. A.; Farha,
O. K.; Hupp, J. T. J. Am. Chem. Soc. 2012, 134, 18790.
(21) Kim, M.; Cahill, J. F.; Su, Y.; Prather, K. A.; Cohen, S. M. Chem.
Sci. 2012, 3, 126.
CONCLUSION
■
Our study employs two fundamentally different strategies
(PSD, PSE) to synthesize the first highly robust MOF bearing
isolated monocatecholato metal sites on the strut of the organic
linkers. The results strongly suggest that postsynthetic
approaches are facile and important functionalization methods
to access MOFs that cannot be directly synthesized. The PSE
strategy presented here is a rare postsynthetic example to
introduce open metal sites, in a one-step reaction. Metalation of
these catecholato groups resulted in unprecedented metal-
monocatecholato species, and both Fe-monocatecholato and
Cr-monocatecholato moieties were evaluated by EXAFS. UiO-
66-CrCAT proved to be an efficient and “green” alcohol
oxidation catalyst for a range of substrates. Complete
heterogeneity and recyclability of our MOF catalysts overcome
the problem of leaching in Cr-exchanged molecular sieves.
(22) Kim, M.; Cahill, J. F.; Fei, H.; Prather, K. A.; Cohen, S. M. J. Am.
Chem. Soc. 2012, 18082.
(23) Fei, H.; Cahill, J. F.; Prather, K. A.; Cohen, S. M. Inorg. Chem.
2013, 52, 4011.
(24) Li, T.; Kozlowski, M. T.; Doud, E. A.; Blakely, M. N.; Rosi, N. L.
J. Am. Chem. Soc. 2013, 135, 11688.
(25) Jeong, S.; Kim, D.; Song, X.; Choi, M.; Park, N.; Lah, M. S.
Chem. Mater. 2013, 25, 1047.
ASSOCIATED CONTENT
* Supporting Information
Experimental details of synthesis and catalysis and additional
characterizations. This material is available free of charge via the
Internet at http://pubs.acs.org.
■
S
(26) Pierpont, C. G.; Lange, C. W. Prog. Inorg. Chem. 2007, 41.
(27) Dhungana, S.; Heggemann, S.; Heinisch, L.; Mollmann, U.;
Boukhalfa, H.; Crumbliss, A. L. Inorg. Chem. 2001, 40, 7079.
(28) Powell, H. K. J.; Taylor, M. C. Aust. J. Chem. 1982, 35, 739.
(29) Weston, M. H.; Farha, O. K.; Hauser, B. G.; Hupp, J. T.;
Nguyen, S. T. Chem. Mater. 2012, 24, 1292.
(30) Tanabe, K. K.; Siladke, N. A.; Broderick, E. M.; Kobayashi, T.;
Goldston, J. F.; Weston, M. H.; Farha, O. K.; Hupp, J. T.; Pruski, M.;
Mader, E. A.; Johnson, M. J. A.; Nguyen, S. T. Chem. Sci. 2013, 4,
2483.
(31) Weston, M. H.; Peterson, G. W.; Brown, M. A.; Jones, P.; Farha,
O. K.; Hupp, J. T.; Nguyen, S. T. Chem. Commun. 2013, 49, 2995.
(32) Nguyen, H. G. T.; Weston, M. H.; Farha, O. K.; Hupp, J. T.;
Nguyen, S. T. CrystEngComm 2012, 14, 4115.
(33) Nguyen, H. G. T.; Weston, M. H.; Sarjeant, A. A.; Gardner, D.
M.; An, Z.; Carmieli, R.; Wasielewski, M. R.; Farha, O. K.; Hupp, J. T.;
Nguyen, S. T. Cryst. Growth. Des. 2013, 2013, 3528.
(34) Allen, C. A.; Cohen, S. M. J. Mater. Chem. 2012, 22, 10188.
(35) Pullen, S.; Fei, H.; Orthaber, A.; Cohen, S. M.; Ott, S. J. Am.
Chem. Soc. 2013, 135, 16997.
(36) Jonas, R. T.; Stack, T. D. P. J. Am. Chem. Soc. 1997, 119, 8566.
(37) Merkel, M.; Schnieders, D.; Baldeau, S. M.; Krebs, B. Eur. J.
Inorg. Chem. 2003, 783.
(38) A, L.-T.; Duan, L.-M.; Xu, X.-T.; Liu, S.-Q. Chem. Res. Chin.
Univ. 2009, 25, 273.
(39) Sever, M. J.; Wilker, J. J. Dalton Trans. 2004, 1061.
(40) Abu-Nawwas, A. H.; Cano, J.; Christian, P.; Mallah, T.;
Rajaraman, G.; Teat, S. J.; Winpenny, R. E. P.; Yukawa, Y. Chem.
Commun. 2004, 314.
(41) Weisser, J. T.; Nilges, M. J.; Sever, M. J.; Wilker, J. J. Inorg.
Chem. 2006, 45, 7736.
AUTHOR INFORMATION
Corresponding Author
scohen@ucsd.edu
■
Notes
The authors declare no competing financial interest.
ACKNOWLEDGMENTS
■
This work was supported by a grant from the National Science
Foundation, Division of Materials Research (DMR-1262226).
The synchrotron X-ray absorption spectroscopy data were
collected at Advanced Photon Source in Argonne National
Laboratory on beamline 9-BM through the general user
proposal program. We thank Dr. Y. Su (UC San Diego) for
assistance with the mass spectrometry data and helpful
discussion.
REFERENCES
■
(1) O’Keeffe, M.; Yaghi, O. M. Chem. Rev. 2012, 112, 675.
(2) Sumida, K.; Rogow, D. L.; Mason, J. A.; McDonald, T. M.; Bloch,
E. D.; Herm, Z. R.; Bae, T.-H.; Long, J. R. Chem. Rev. 2012, 112, 724.
(3) Li, J.-R.; Sculley, J.; Zhou, H.-C. Chem. Rev. 2012, 112, 869.
(4) Yoon, M.; Srirambalaji, R.; Kim, K. Chem. Rev. 2012, 112, 1196.
(5) Kreno, L. E.; Leong, K.; Farha, O. K.; Allendorf, M.; Van Duyne,
R. P.; Hupp, J. T. Chem. Rev. 2012, 112, 1105.
(6) Rocca, J. D.; Liu, D.; Lin, W. Acc. Chem. Res. 2011, 44, 957.
(7) Morris, R. E.; Wheatley, P. S. Angew. Chem., Int. Ed. 2008, 47,
4966.
(42) Hider, R. C.; Howlin, B.; Miller, J. R.; Mohd-Nor, A. R.; Silver, J.
Inorg. Chim. Acta 1983, 80, 51.
(8) Bloch, E. D.; Queen, W. L.; Krishna, R.; Zadrozny, J. M.; Brown,
C. M.; Long, J. R. Science 2012, 335, 1606.
(43) Hider, R. C.; Mohd-Nor, A. R.; Silver, J.; Morrison, I. E. G.;
Rees, L. V. C. J. Chem. Soc., Dalton Trans. 1981, 609.
(44) von Nussbaum, F.; Spiteller, P.; Ruth, M.; Steglich, W.; Wanner,
G.; Gamblin, B.; Stievano, L.; Wagner, F. E. Angew. Chem., Int. Ed.
1998, 37, 3292.
(9) Horike, S.; Dinca, M.; Tamaki, K.; Long, J. R. J. Am. Chem. Soc.
2008, 130, 5854.
(10) Biswas, S.; Zhang, J.; Li, Z.; Liu, Y. Y.; Grzywa, M.; Sun, L.;
Volkmer, D.; Van Der Vorrt, P. Dalton Trans. 2013, 42, 4730.
(11) Yang, Q.; Vaesen, S.; Ragon, F.; Wiersum, A. D.; Wu, D.; Lago,
A.; Devic, T.; Martineau, C.; Taulelle, F.; Llewellyn, P. L.; Jobic, H.;
(45) Shiren, K.; Tanaka, K. Inorg. Chem. 2002, 41, 5912.
4972
dx.doi.org/10.1021/ja411627z | J. Am. Chem. Soc. 2014, 136, 4965−4973

Journal of the American Chemical Society
Article
(46) Peterson, M. L.; Brown, G. E.; Parks, G. A.; Stein, C. L. Geochim.
Cosmochim. Acta 1997, 61, 3399.
(47) Smiley, R. A.; jackson, H. L. Chemistry and the Chemical
Industry; CRC Press: Boca Raton, FL, 2002.
(48) Tojo, G.; Fernandez, M. Oxidation of Alcohols to Aldehydes and
Ketones; Springer: New York, 2010.
(49) Corma, A.; Garcia, H. Chem. Soc. Rev. 2008, 37, 2096.
(50) Mallat, T.; Baiker, A. Chem. Rev. 2004, 104, 3037.
(51) Parmeggiani, C.; Cardona, F. Green Chem. 2012, 14, 547.
(52) Muzart, J. Chem. Rev. 1992, 92, 113.
(53) Lounis, Z.; Riahi, A.; Djafri, F.; Muzart, J. Appl. Catal., A 2006,
309, 270.
(54) Lempers, H. E. B.; Sheldon, R. A. J. Catal. 1998, 175, 62.
(55) Maksimchuk, N. V.; Kovalenko, K. A.; Fedin, V. P.; Kholdeeva,
O. A. Chem. Commun. 2012, 48, 6812.
(56) Dhakshinamoorthy, A.; Alvaro, M.; Garcia, H. Catal. Sci.
Technol. 2011, 1, 856.
(57) Liu, H.; Liu, Y.; Li, Y.; Tang, Z.; Jiang, H. J. Phys. Chem. C 2010,
114, 13362.
(58) Ishida, T.; Nagaoka, M.; Akita, T.; Haruta, M. Chem.Eur. J.
2008, 14, 8456.
(59) Chen, G.; Wu, S.; Liu, H.; Jiang, H.; Li, Y. Green Chem. 2013,
15, 230.
(60) Dhakshinamoorthy, A.; Alvaro, M.; Garcia, H. ACS Catal. 2011,
1, 48.
(61) Carson, F.; Agrawal, S.; Gustafsson, M.; Bartoszewicz, A.;
Moraga, F.; Zou, X.; Martin-Matute, B. Chem.Eur. J. 2012, 18,
15337.
(62) Kato, C. N.; Hasegawa, M.; Sato, T.; Yoshizawa, A.; Inoue, T.;
Mori, W. J. Catal. 2005, 230, 226.
(63) Hoover, J. M.; Stahl, S. S. J. Am. Chem. Soc. 2011, 133, 16901.
(64) Davi, M.; Lebel, H. Org. Lett. 2009, 11, 41.
(65) Figiel, P. J.; Sibaouih, A.; Ahmad, J. U.; Nieger, M.; Raisanen, M.
T.; Leskela, M.; Repo, T. Adv. Synth. Catal. 2009, 351, 2625.
(66) Marko, I. E.; Gautier, A.; Dumeunier, R.; Doda, K.; Philippart,
F.; Brown, S. M.; Urch, C. J. Angew. Chem., Int. Ed. 2004, 43, 1588.
(67) Herbert, M.; Montilla, F.; Galindo, A. Dalton Trans. 2010, 39,
900.
(68) Sheldon, R. A.; Wallau, M.; Arends, I. W. C. E.; Schuchardt, U.
Acc. Chem. Res. 1998, 31, 485.
(69) Arends, I. W. C. E.; Sheldon, R. A. Appl. Catal., A 2001, 175.
4973
dx.doi.org/10.1021/ja411627z | J. Am. Chem. Soc. 2014, 136, 4965−4973