Highly selective H2O2-based oxidation of alkylphenols to p-benzoquinones over MIL-125 metal-organic frameworks

Article abstract of DOI:10.1002/ejic.201301098

The titanium-based metal-organic framework MIL-125 and its amine-functionalized analog, MIL-125-NH₂, have been synthesized and characterized by elemental analysis, XRD, SEM, TEM, N₂ adsorption measurements, and spectroscopic techniques, including FTIR, Raman, and DR UV/Vis spectroscopy. Catalytic properties of MIL-125 and MIL-125-NH₂ were evaluated in the selective oxidation of two representative alkyl-substituted phenols, 2,3,6-trimethylphenol and 2,6-di-tert-butylphenol, with the clean oxidant H₂O₂. With both MIL-125 and MIL-125-NH ₂, the selectivity toward the corresponding p-benzoquinones was 100 %. Samples of MIL-125 with different sizes of crystallites (0.5, 1.5, and 5 μm) demonstrated similar reaction rates, thus indicating the absence of diffusion limitations. The efficiency of the oxidant utilization and stability of the MIL-125 structure increased upon decreasing the amount of water in the reaction mixture. Even if the structural integrity of MIL-125 was destroyed by the reaction medium, the metal-organic framework acted as a precursor for the highly active, selective, and recyclable catalyst. The MIL-125-derived materials were stable toward titanium leaching, behaved as true heterogeneous catalysts, and could easily be recovered by filtration and reused several times without the loss of the catalytic properties. Copyright

Full text of DOI:10.1002/ejic.201301098

FULL PAPER
DOI:10.1002/ejic.201301098
Highly Selective H2O2-Based Oxidation of Alkylphenols
to p-Benzoquinones Over MIL-125 Metal–Organic
Frameworks
Irina D. Ivanchikova,[a] Ji Sun Lee,[b,c] Nataliya V. Maksimchuk,[a]
Alexander N. Shmakov,[a] Yurii A. Chesalov,[a] Artem B. Ayupov,[a]
Young Kyu Hwang,[b] Chul-Ho Jun,[c] Jong-San Chang,*[b,d] and
Oxana A. Kholdeeva*[a]
Keywords: Heterogeneous catalysis / Oxidation / Hydrogen peroxide / Metal–organic framework / Titanium
The titanium-based metal–organic framework MIL-125 and
its amine-functionalized analog, MIL-125_NH2, have been
synthesized and characterized by elemental analysis, XRD,
SEM, TEM, N2 adsorption measurements, and spectroscopic
techniques, including FTIR, Raman, and DR UV/Vis spec-
troscopy. Catalytic properties of MIL-125 and MIL-125_NH2
were evaluated in the selective oxidation of two representa-
tive alkyl-substituted phenols, 2,3,6-trimethylphenol and 2,6-
di-tert-butylphenol, with the clean oxidant H2O2. With both
MIL-125 and MIL-125_NH2, the selectivity toward the corre-
sponding p-benzoquinones was 100%. Samples of MIL-125
with different sizes of crystallites (0.5, 1.5, and 5 μm) demon-
strated similar reaction rates, thus indicating the absence of
diffusion limitations. The efficiency of the oxidant utilization
and stability of the MIL-125 structure increased upon
decreasing the amount of water in the reaction mixture. Even
if the structural integrity of MIL-125 was destroyed by the
reaction medium, the metal–organic framework acted as a
precursor for the highly active, selective, and recyclable cata-
lyst. The MIL-125-derived materials were stable toward tita-
nium leaching, behaved as true heterogeneous catalysts, and
could easily be recovered by filtration and reused several
times without the loss of the catalytic properties.
An important feature of MOFs is a high content of metal
sites (most often, well-defined clusters), which are uni-
formly spatially distributed and are accessible for reagents.
Titanium is a very attractive candidate for construction of
MOFs due to its low toxicity, photocatalytic activity, and
potential ability to activate hydroperoxides. Ti-containing
solids, in particular, titanium-silicates are well known as ef-
fective catalysts for a number of selective oxidations with
aqueous hydrogen peroxide and alkylhydroperoxides.[3] Re-
cently, Férey and coworkers reported on the synthesis of
a titanium-oxo-hydroxo-cluster-based MOF designated as
MIL-125 using terephthalic acid (H2bdc) and titanium(IV)
isopropoxide.[4] The quasi-cubic tetragonal structure of
MIL-125 is built up from cyclic octamers of edge- and cor-
ner-sharing TiO5(OH) octahedra connected to 12 other cy-
clic octamers through 1,4-benzenedicarboxylate (bdc) link-
ers. The structure of MIL-125 has two types of cages,
namely, an octahedral one (12.5 Å diameter) and a tetra-
hedral one (6 Å diameter), accessible through triangular
windows of 5–7 Å. An amine-functionalized analog MIL-
125_NH2, isostructural to MIL-125, has been prepared by
replacing H2bdc with 2-amino benzene dicarboxylic acid
(H2bdc-NH2).[5] Photocatalytic properties of MIL-125 and
MIL-125_NH2 were studied by a few research groups.[4–6]
Kim et al. demonstrated the catalytic activity of both Ti-
Introduction
Metal–organic frameworks (MOFs) are functional po-
rous coordination polymers with a crystalline structure
comprised of metal ions or clusters linked by polydentate
organic ligands. Because of their unique combination of
properties, such as large surface areas and pore volumes,
diverse topologies, and possibilities of functionalization,
MOFs have generated a great deal of interest in recent years
as prospective materials for gas or liquid adsorption and
separation, molecular recognition, and heterogeneous catal-
ysis.[1] Moreover, MOFs can be applied as precursors for
the preparation of various nanomaterials.[2]
[a] Boreskov Institute of Catalysis,
Pr. Akademika Lavrentieva 5, Novosibirsk 630090, Russia
E-mail: khold@catalysis.ru
www.catalysis.ru
[b] Korea Research Institute of Chemical Technology (KRICT),
P. O. Box 107, Gajeongro 141, Yuseong, Daejeon 305-600,
Republic of Korea
E-mail: jschang@krict.re.kr
http://rgn.krict.re.kr/index.sko
[c] Department of Chemistry and Center for Bioactive Molecular
Hybrid, Yonsei University, Seoul 120-749, Republic of Korea
[d] Department of Chemistry, Sungkyunkwan University,
Suwon 440-476, Republic of Korea
Supporting information for this article is available on the
WWW under http://dx.doi.org/10.1002/ejic.201301098.
Eur. J. Inorg. Chem. 2014, 132–139
132
© 2014 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

www.eurjic.org
FULL PAPER
MOFs in high-pressure cycloaddition of CO2 to epi- MIL-125-S) while the addition of water [molar ratio of
chlorohydrin.[7] They also studied the oxidative desulf- H2O/Ti(OiPr)4 = 1] favors the formation of larger particles
urization of heterocyclic aromatic sulfur compounds with (ca. 5 μm, MIL-125-L).
cumene hydroperoxide over MIL-125.[7] Recently,
McNamara et al. reported similar results using tert-butyl
hydroperoxide (TBHP) as the oxidant.[8] To the best of our
knowledge, no examples have been reported to date on the
application of MIL-125 in H2O2-based liquid-phase oxi-
dation.
The selective oxidation of substituted phenols is a topic
of great interest since quinones bearing different functional
groups are intermediates in the production of fine chemi-
cals, including the synthesis of vitamins E and K, coenzyme
Q, and other valuable products.[9] Recently, some of us
found that Ti,Si catalysts prepared by grafting titanium(IV)
onto commercial mesoporous silicas are efficient catalysts
for oxidation of 2,3,6-trimethylphenol (TMP) and 2,6-di-
methylphenol with H2O2 to produce 2,3,5-trimethyl-1,4-
benzoquinone (TMBQ, vitamin E key intermediate) and
2,6-dimethyl-1,4-benzoquinone, respectively, with a nearly
quantitative yield.[10] A correlation between the selectivity
Figure 1. SEM images of (a) MIL-125-S, (b) MIL-125-M, (c) MIL-
toward quinone and Ti surface concentration has been re-
vealed.[10b,10c] The key point to achieve the high quinone
selectivity was found to be the presence of TiIV dimers or
125-L, and (d) MIL-125_NH2.
Figure S1 (see SI) shows TG profiles of MIL-125-M and
small clusters homogeneously distributed on the silica sur- MIL-125_NH2. Virtually identical TG profiles were ob-
face.[10b,10c] Taking into account that the structure of MIL- tained for the two samples. On the whole, two weight-loss
125 is composed of titanium oxo-hydroxo clusters, one steps were observed: the first one corresponds to the depar-
might expect that highly selective oxidation of alkylphenols ture of the guest molecules (H2O, MeOH, and DMF) over
to p-benzoquinones could be realized over this MOF. Fol- the range 25–300 °C (ca. 35%) and the second one may be
lowing this idea, we explored the catalytic performance of from the phase transition or structure reorganization above
MIL-125 and MIL-125_NH2 in the oxidation of two repre- 300 °C in MIL-125_NH2 and 450 °C in MIL-125-M.[11]
sentative alkyl-substituted phenols, TMP and 2,6-di-tert- The first weight loss below 100 °C (ca. 30%) is attributed
butylphenol (DTBP), using aqueous H2O2 as the oxidant to the departure of guest solvents, mainly water, because
and acetonitrile as the solvent. The results obtained have the samples were fully saturated at 70% humidity before the
been compared with the results acquired using Ti,Si cata- TGA measurements.
lysts and anatase. Special attention was drawn to the crucial
issues of the catalyst stability and recyclability.
XRD patterns of MIL-125-M and MIL-125_NH2 are
given in Figure S2 in the SI and confirm the structure of
MIL-125 reported by Férey and coworkers.[4] Low-tem-
perature nitrogen adsorption revealed type I isotherms
characteristic of microporous solids (Figure S3 in SI, curve
A). Textural parameters acquired from the N2 adsorption
measurements are presented in Table 1 along with the ele-
mental analysis data and the average particle size estimated
Results and Discussion
Catalyst Preparation and Characterization
MIL-125 was synthesized by a solvothermal method as by SEM. Specific BET surface areas exceeded (ca. 10–15%)
reported elsewhere.[11] Detailed procedures for the synthesis the values reported in the literature for MIL-125-M and for
of MIL-125 and MIL-125_NH2 are given in the Supporting MIL-125_NH2.[4,5,6b] Micropore volumes estimated by the
Information (SI). Samples of MIL-125 with three different αS-plot method were close to the value of 0.65 cm3 g–1 re-
sizes of crystallites (0.5, 1.5, and 5.0 μm) have been desig- ported for MIL-125.[4]
nated as MIL-125-S, MIL-125-M, and MIL-125-L, respec-
FTIR spectra of MIL-125-M and MIL-125_NH2 (Fig-
tively. Figure 1 shows SEM images of the MIL-125 and ure S4 in SI, curves A) reveal strong characteristic bands
MIL-125_NH2 samples. MIL-125-M, which was prepared arising from asymmetric (1500–1700 cm–1) and symmetric
by a conventional procedure, was rather disk-like in shape (1350–1450 cm–1) vibrations of carboxylate groups along
with crystallites of ca. 1.5 μm, while MIL-125_NH2 was with O–Ti–O vibrations over the range 400–800 cm–1. No
rather rectangular in shape with particles of ca. 0.7 μm. It band at 1710 cm–1, characteristic of free H2bdc, was present
was found that benzoic acid and water can be used as a corroborating the absence of the free linker within the
modulator to affect the particle size of MIL-125. The ad- micropores. Raman spectra of MIL-125-M and MIL-125_
dition of benzoic acid [molar ratio of benzoic acid/ NH2 (Figure 2a and 2b, respectively, curves A) are also con-
Ti(OiPr)4 = 10] leads to smaller crystallites (ca. 0.5 μm, sistent with the literature.[6b,11] It is well known that the
Eur. J. Inorg. Chem. 2014, 132–139
133
© 2014 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

www.eurjic.org
FULL PAPER
Table 1. Physicochemical properties of Ti-MOFs studied.
ening of the DRS UV/Vis band.[3d,3f,9b,12a,13,14] Figure S5 in
the SI shows the DR UV/Vis spectra of the three samples
of MIL-125 in comparison with the spectrum of a silica-
grafted titanium catalyst (the latter contained well-dis-
persed oligomeric TiO2 species).[10b] One can see that the
spectra of both types of the Ti catalysts are quite similar.
Hence, we might tentatively ascribe the adsorptions at 240
and 290 nm in the DR UV/Vis spectra of MIL-125 to li-
gand-to-metal charge-transfer bands from the presence of,
at least, two types of octahedrally coordinated clusterized
TiIV species.
[a]
[b]
Catalyst
Ti
Particle size SBET
Vmicro
Vmeso
[wt.-%] [μm]
[m2g–1] [cm3 g–1] [cm3 g–1]
MIL-125-S
MIL-125-M
23.8
0.5
1.5
1713
1868
320[c]
364[d]
1683
1575
0.64
0.76
0[c]
0
0
20.0
25.9[c]
30.0[d]
24.6
0.25[c]
0[d]
0.23[d]
MIL-125-L
MIL-125_NH2
5.0
0.7
0.64
0.54
0
0
23.5
[a] Micropore volume. [b] Mesopore volume. [c] After one run in
TMP oxidation with H2O2; reaction conditions as in Table 2, entry
5. [d] After four runs in TMP oxidation with H2O2; reaction condi-
tions as in Table 2, entry 5.
Raman technique enables the early detection of the emer-
gence of anatase microcrystallites in Ti-containing solids by
the presence of
a strong absorption band at 140–
145 cm–1.[3c,3f,12] The lack of this absorption (see inserts in
Figure 2) in the Raman spectra of both Ti-MOFs confirms
that no anatase-like admixtures are present.
Figure 3. DR UV/Vis spectra of (a) MIL-125-M and (b) MIL-125_
NH2: (A) initial, (B) and (C) after 1 and 4 runs of TMP oxidation
with H2O2, respectively. Reaction conditions as in Figure 2.
Catalytic Properties
Catalytic properties of the MIL-125 materials were first
assessed in TMP oxidation with aqueous H2O2 in aceto-
nitrile solvent. The main results are collected in Table 2.
Both MIL-125 and MIL-125_NH2 favored the formation
of TMBQ with 100% selectivity (Table 2, entries 1 and 2).
With another substituted phenol, DTBP, the corresponding
Figure 2. Raman spectra of (a) MIL-125-M and (b) MIL-125_
NH2: (A) initial, (B) and (C) after 1 run and 4 runs of TMP oxi-
dation with H2O2, respectively. Reaction conditions: TMP (0.1 m),
H2O2 (0.4 m), MIL-125 (21 mg), MeCN (7 mL), 80 °C, 10 min.
DR UV/Vis spectra of MIL-125-M and MIL-125_NH2 p-benzoquinone was also the sole oxidation product found
1
are shown in Figure 3a and 3b, respectively (curves A). The by means of both GC and H NMR spectroscopy, thereby
spectra resemble those reported in the literature.[6,7] While indicating excellent selectivity.
MIL-125 shows two main adsorptions with maxima cen-
Earlier, some of us established, for titanium-silicate cata-
tered at ca. 240 and 290 nm, MIL-125_NH2 reveals an ad- lysts, that the crucial point to achieve excellent benzoquin-
ditional band at ca. 390 nm with the absorption edge ex- one selectivity in the H2O2-based oxidation is the presence
tending to around 500 nm (Figure 3, b) associated with the of TiIV dimers or small clusters on the silica surface. Such
chromophore of bdc-NH2 units.[6a] It is well documented species are manifested by broad DRS UV/Vis bands in the
in the literature that solids with site-isolated tetrahedrally range 240–290 nm (see, for example, curve D in Figure S5
coordinated TiIV centers exhibit a narrow absorption band in the SI).[10b,10c] Catalysts that contain isolated TiIV sites
at 208–210 nm attributed to an oxygen-to-metal charge (DRS UV maximum at 210–220 nm) and/or anatase-like
transfer,[3c,3f,12a,13] while the increase in the TiIV coordina- microcrystallites (DRS UV/Vis: 330–340 nm) produce a sig-
tion number and the appearance of dimeric and oligomeric nificant amount of dimeric byproducts.[10c,12b] Hence, we
TiO2 species are manifested by a long-wave shift and broad- can suggest that the outstanding selectivity toward TMBQ
Eur. J. Inorg. Chem. 2014, 132–139
134
© 2014 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

www.eurjic.org
FULL PAPER
Table 2. TMP oxidation with H2O2 over MIL-125.[a]
Entry Catalyst
[H2O2] [m]
TMP conversion [%] H2O2 efficiency [%]
TOF[b] [min–1]
TMBQ selectivity [%]
1
2
3
4
5
6
7
8
MIL-125-M
0.4
0.4
0.4
0.4
0.4
56
50
57
54
57
40
70
35
50
50
49
50
50
23
28
25
29
27
29
32
18
10
24
38
37
39
32
9
1.4
1.0
1.5
1.3
1.1
100
100
100
100
100
100
100
57[h]
94
MIL-125_NH2
MIL-125-M[c]
MIL-125-M[d]
MIL-125-M[e]
MIL-125-M
MIL-125-M
MIL-125-M
MIL-125-M
MIL-125-M
MIL-125-S
MIL-125-L
0.25
0.8
0.7
1.5
0.4[f]
0.4
n.d.[g]
0.1
9[i]
10
11
12
13
14
0.25[j]
0.25[j]
0.25[j]
0.25[j]
0.4
0.9
1.0
1.0
0.8
96
96
98
MIL-125_NH2
89
TiO2 (Degussa P25)[k]
0.03
80[h]
[a] Reaction conditions: TMP (0.1 m), MIL-125 (3 mg), H2O2 (used as 35% aq. solution, 0.25–0.8 m, 44–140 μL), MeCN (2 mL), 80 °C,
10–15 min. [b] TOF = (mol of TMP consumed)/[(mol of Ti)ϫtime]; determined from the initial rates of TMP consumption under the
assumption that all Ti centers are accessible. [c] Before the catalytic reaction the catalyst was pre-heated at 130 °C under vacuum for 6 h.
[d] MIL-125 (1.5 mg); reaction time 20 min. [e] MIL-125 (6 mg). [f] H2O2 was added dropwise over 30 min; reaction time 1 h. [g] Not
determined. [h] The main byproduct was 2,2Ј,3,3Ј,5,5Ј-hexamethyl-4,4Ј-biphenol. [i] 40 °C, 4 h. [j] 88% H2O2 was used. [k] TiO2 (3 mg)
contained mostly anatase.
Previously, it was found that microporous titanium silicalite
TS-1 (5.4ϫ5.6 Å) is, at least, an order of magnitude less
active in this reaction than mesoporous Ti,Si catalysts.[15]
In order to elucidate the impact of internal diffusion in
the case of MIL-125, samples with different sizes of crystal-
lites (0.5, 1.5, and 5 μm) have been examined. As one can
judge from the kinetic curves shown in Figures S6a and b
in the SI, the initial rate of both TMP and DTBP oxidation
was not affected by the size of the crystallites. Therefore,
we might assume that no diffusion limitation takes place in
the course of the catalytic oxidation of TMP and DTBP
over MIL-125, although the smallest dimensions of the
TMP (5.4ϫ6.7 Å) and DTBP (5.6ϫ9.4 Å) molecules are
close to the dimensions of the MOF cage windows. We
might suppose that π–π interactions between aromatic rings
of phenols and of the MOF linker (bdc) could favor ef-
ficient adsorption of the organic substrates within MIL-
125, thus ensuring its high catalytic activity. Moreover, it
has been recently established for hydrocarbon–zeolite sys-
tems that penetration of guest molecules into host pores
can be accelerated when dimensions of guest and host are
close to each other (the so-called levitation effect).[16] Sim-
ilar effects were suggested for encapsulation of iron tetra-
sulfophthalocyanine (ca. 17ϫ17 Å) into cages of Cr-MIL-
101 (15ϫ16 Å windows).[17] Finally, we should also take
into account that destruction of the MIL-125 framework
under the conditions of liquid-phase oxidation with aque-
ous H2O2 (see below) could be, at least, partially responsi-
ble for the unusually high catalytic activity of the Ti-MOFs
in the selective oxidation of the alkyl-substituted phenols.
Regardless of the high activity, only moderate TMP con-
and DTBQ realized over MIL-125 is from the specific struc-
ture of MIL-125 being composed of well-defined titanium
oxo-hydroxo clusters. The lack of isolated Ti species or ana-
tase microcrystallites in the Ti-MOFs ensures the 100% p-
benzoquinone selectivity.
Previously, some of us found that turnover frequencies
(TOFs) for silica-grafted Ti catalysts with di(oligo)meric Ti
centers usually vary over the range 1.4–2.1 min–1 for TMP
oxidation.[10b] If we compare these values with TOFs char-
acteristic of MIL-125 under the same reaction conditions
(e.g., Table 2, entry 1), we can conclude that the rate of
TMP oxidation, referring to one Ti center, is quite similar
for both types of catalysts. Coupled with the 100% selec-
tivity toward TMBQ, this indicates a similar structure and
accessibility of the active Ti centers, at least, at the initial
stage of the reaction. The presence of amine groups in the
MOF linker did not affect the quinone selectivity but
slightly reduced the catalytic activity (TOF 1.0 min–1 vs. versions were attained for both MIL-125 and its amine-
1.4 min–1 for MIL-125). The fact that microporous Ti-MIL- functionalized analog (56 and 50%, respectively) when a
125 (5–7 Å windows) and mesoporous Ti,Si catalysts with twofold excess of 35% H2O2 was employed. With DTBP,
pores in the range of 2.6–15.4 nm[10b] revealed close TOF the substrate conversion was even lower (ca. 20%). Neither
values in the TMP oxidation was quite surprising for us. phenol conversion nor the quinone-product yield could be
Eur. J. Inorg. Chem. 2014, 132–139
135
© 2014 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

www.eurjic.org
FULL PAPER
improved by increasing the reaction time. Preliminary treat-
ment of MIL-125 under vacuum at 130 °C had almost no
effect on the catalytic performance of the coordination
polymer (Table 2, compare entries 1 and 3), which is not
surprising if we remember that the structure of MIL-125
does not allow formation of coordinatively unsaturated
metal sites.[5]
In principle, two main reasons for the low phenol conver-
sions over the Ti-MOFs could be envisaged. The first
reason might be blockage of the MOF pores or active sites
by the reaction products during the reaction course. To in-
vestigate such a possibility, we studied the effect of the cata-
lyst amount on the reaction course. If pore filling by oxi-
dation products occurred, one might expect an increase of
the substrate conversion upon increasing the catalyst load-
ing. However, this parameter turned out to produce practi-
cally no effect on the resulting phenol conversion (Table 2,
compare entries 1, 4, and 5).
Another reason for the low phenol conversions could be
high rates of the unproductive decomposition of H2O2 rela-
tive to the rates of the TMP oxidation over the Ti-MOFs,
i.e. a rather low efficiency of the oxidant utilization. Indeed,
Figure S7 (see SI) shows that H2O2 decomposition occurs
with a rate comparable to the rate of TMP oxidation. Iodo-
metric titration confirmed that no H2O2 was present at the
end of the catalytic reactions (Table 2, entries 1–6). Import-
antly, TMP conversion could be increased up to 97% with-
out loss of TMBQ selectivity by using an additional portion
of the oxidant (Figure 4), which proves that decomposition
Figure 4. Effect of the oxidant amount on TMP oxidation with
H2O2 over MIL-125-M (filled and open symbols correspond to
substrate consumption and product yield, respectively). Reaction
conditions: TMP (0.1 m), H2O2 (0.4–0.8 m, 70–140 μL), MIL-125-
M (3 mg), MeCN (2 mL), 80 °C.
sion under the conditions of entry 1, Table 2). In turn, the
use of TBHP (4.7 m solution in decane) as the oxidant re-
sulted in the reduction of both substrate conversion (18%
after 1 h) and quinone selectivity (56% vs. 100% with
H2O2).
Catalyst Stability and Recycling
Catalyst stability toward leaching of active species and
of H2O2 to produce O2 and water is the main reason for the nature of catalysis are indispensable properties of solid
the incomplete phenol conversions in the presence of MIL- catalysts, which should be carefully investigated in liquid-
125. However, the higher the concentration of H2O2 that phase oxidation processes.[18] First of all, we performed a
was employed the lower the efficiency of the oxidant utiliza- hot filtration test and found no further substrate conversion
tion (Table 2, compare entries 1, 6, and 7). A dropwise ad- or growth of the product yield in the filtrate after removal
dition of the oxidant (this procedure is often employed to of the catalyst (Figure S8 in the SI). This proved that the
improve the oxidant efficiency) resulted in a significant de- observed catalysis in the presence of MIL-125 is truly
crease of both TMBQ selectivity and H2O2 efficiency heterogeneous, i.e. occurs on the catalyst surface rather than
(Table 2, entry 8). Reduction of the reaction temperature in solution as a result of leached active metal species. More-
produced a similar effect, although not so pronounced over, only a trace amount of titanium (0.1 ppm) was deter-
(Table 2, entry 9). In general, the oxidant efficiency for the mined in the filtrate by ICP-AES.
Ti-MOFs was inferior to that observed for Ti,Si catalysts,
The MIL-125 catalyst could be recovered by simple fil-
for which this parameter attained ca. 57%, i.e. a twofold tration and reused without loss of selectivity to the target
excess of H2O2 was sufficient to convert 100% of TMP.[10b] quinone product (Figure 5). Importantly, no catalyst regen-
On the other hand, both MIL-125 and MIL-125_NH2 eration was needed before reuse. In the second run, the
showed much better oxidant efficiency and selectivity to- maximal conversion of TMP and the oxidant efficiency
ward TMBQ than anatase (Table 2, entry 14).
even increased and reached 68 and 35%, respectively, while
We found that the efficiency of the oxidant utilization the catalytic activity expressed in TOF values determined
over MIL-125 could be improved by reducing the amount from the initial rates of TMP consumption (the correspond-
of water in the reaction mixture. With more concentrated ing kinetic curves are given in Figure S9) decreased. How-
H2O2 (ca. 88%), the oxidant efficiency achieved 37–39%, ever, TOFs stabilized after the second reuse (Figure 5), thus
but the selectivity toward TMBQ and TOF slightly de- indicating that the MOF catalyst, most likely, converted to
creased (Table 2, entries 10–12). Again, the size of catalyst another, more stable form. It is noteworthy that the recycl-
particles produced almost no effect on the reaction rate and ing behavior of the Ti-MOF catalyst in TMP oxidation is
selectivity, indicating an insignificant role of diffusion significantly better than that of mesoporous Ti,Si catalysts,
limitations during the oxidation process.
for which a progressive loss of both the catalytic activity
We also attempted to use an anhydrous H2O2–urea ad- and selectivity readily occurs with 35% aqueous H2O2 that
duct instead of aqueous H2O2. However, this led to a dras- causes hydrolysis of Ti–O–Si bonds and agglomeration of
tic decrease in the catalyst activity (only 4% TMP conver- Ti active sites on the surface.[10b]
Eur. J. Inorg. Chem. 2014, 132–139
136
© 2014 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

www.eurjic.org
FULL PAPER
ture of smaller crystallites (ca. 0.5–0.8 μm). The N2 adsorp-
tion study on recycled MIL-125-M revealed a significant
decrease of the specific surface area and appearance of
mesopores instead of the initially present micropores (see
Table 1; the corresponding N2 adsorption isotherms are
shown in Figure S3). We assume that the increase in TMP
conversion observed in the second run (Figure 5; Figure S9)
might be due to the appearance of the mesopores that favor
adsorption of the phenolic substrate. Importantly, the tex-
tural parameters of the catalyst after the first and fourth
reuse were very close to each other, which is in line with the
observed catalytic results (Figure 5; Figure S9).
A significant broadening of signals in the FTIR spectra
of recycled MIL-125 and MIL-125_NH2 also indicated
structural alterations (see Figures S4a and b in the SI,
curves B–C). The IR spectra exhibited a broad absorption
below 1000 cm–1, which is characteristic of ionic oxides, in-
cluding TiO2 polymorphs.[14] However, it did not coincide
precisely with the spectrum of any TiO2 crystalline modifi-
cation. In turn, the intensity of signals in the Raman spectra
of both MIL-125 and MIL-125_NH2 decreased markedly
after the catalyst reuse (Figure 2, a and b, curves B–C),
which also corroborates a MOF destruction process. How-
ever, no strong absorption bands around 140 and 160 cm–1
typical of TiO2 polymorphs (anatase and brookite, respec-
tively)[19] emerged, as one can judge from the inserts in Fig-
ure 2.
Finally, DR UV/Vis spectra of MIL-125-M recovered af-
ter the catalytic runs showed a redshift in the absorption
edge and a new absorption in the visible light range (Fig-
ure 3, a, curves B and C), indicating some further agglomer-
ation of the Ti clusters. Importantly, the DR UV/Vis spec-
tra runs practically coincided after the first and fourth cata-
lyst use, which, in line with the other results, implies that
the main structural modification occurs during the first
catalytic run to produce a new solid form that is fairly stable
to further modifications. For MIL-125_NH2, the intensity
of the absorption band at 390 nm decreased after catalysis,
accompanied by a redshift in the absorption edge (Figure 3,
b, curve B). Note that a redshift in the absorption edge and
appearance of absorption in the visible range have recently
been observed for graphene-wrapped TiO2 nanoparticles.[20]
One might expect that agglomeration of TiO2 clusters dur-
ing the MOF destruction process would result in worsening
the accessibility of the Ti active centers. Indeed, the initial
rates normalized per one Ti atom (TOF) decreased after
the first catalyst use (Figure 5).
Figure 5. Reuse of MIL-125-M in TMP oxidation with 35% H2O2.
Reaction conditions: TMP (0.1 m), H2O2 (0.4 m), MIL-125 (21 mg),
MeCN (7 mL), 80 °C, 10 min.
Stability of the Ti-MOFs under the conditions of TMP
oxidation was probed by XRD. Figure 6 shows low-angle
XRD patterns of MIL-125-M before and after catalysis.
The catalyst sample recovered after TMP oxidation with
35% H2O2 (in this case, the reaction mixture contained
1.5 m of H2O) exhibited no XRD reflections (Figure 6,
curve B), thereby indicating that the crystal structure is
completely destroyed. Additionally, elemental analysis of
the recycled catalyst revealed an increase of the Ti content
in the MIL-125-M sample after reuse in TMP oxidation
(see Table 1). A similar picture was observed for MIL-125_
NH2 although, on the basis of recent results of Ahn and
coworkers,[7] one might expect a higher hydrolytic stability
for the amine-modified Ti-MOF. On the other hand, after
the reaction with concentrated H2O2 (0.1 m H2O in the re-
action mixture), the characteristic XRD reflections of MIL-
125 were still present (Figure 6, curve C), even though their
reduced intensity and broadening pointed at a partial de-
composition and/or some local distortions of the Ti-MOF
crystal structure. Obviously, structural stability of MIL-125
in liquid-phase oxidation depends on the nature of oxidant
and water content in the reaction medium. Indeed,
McNamara et al. demonstrated that MIL-125 was stable in
the presence of TBHP even under harsh reaction conditions
(24 h at 100 °C).[8]
Bearing in mind all the results collectively, we may sup-
pose that the transformation of the MIL-125 metal–organic
framework, which evidently takes place under turnover con-
ditions of the liquid-phase oxidation with H2O2, has led to
a stable, highly active and selective form of the catalyst,
which contained amorphous fine-dispersed TiO2, most
likely, stabilized by carboxylate ions. Such stabilization did
not allow the formation of anatase-like species, which are
Figure 6. Low-angle XRD patterns for MIL-125-M: (A) initial, (B)
and (C) after 1 run of TMP oxidation with 35 and 88% H2O2,
respectively. Reaction conditions: TMP (0.1 m), H2O2 (0.4 m of
35% or 0.25 m of 88% H2O2), MIL-125 (21 mg), MeCN (7 mL),
80 °C, 10 min.
Both SEM and TEM images (Figures S10 and S11, detrimental to the selective formation of benzoquinones,
respectively) showed that MIL-125-M after treatment with and might be responsible for the superior recycling per-
H2O2 still consisted of well-defined particles with an admix- formance of the MIL-125-derived catalytic material.
Eur. J. Inorg. Chem. 2014, 132–139
137
© 2014 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

www.eurjic.org
FULL PAPER
surface areas (SBET) were evaluated using adsorption data over a
relative pressure range of 0.005 to 0.2 (P/P0; P0: saturation vapor
pressure of N2 at –196 °C) with a specific choice of experimental
points to satisfy the consistency criteria given by Walton and
Conclusions
The Ti-containing metal–organic frameworks MIL-125
and MIL-125_NH2 demonstrate superior selectivity toward
p-benzoquinones in the oxidation of alkyl-substituted phen-
ols with aqueous H2O2, MIL-125 being slightly more active
than its amine-modified analog. With both Ti-MOFs nearly
100% selectivity was reached in the oxidation of TMP to
TMBQ (vitamin E key intermediate). The efficiency of the
oxidant utilization with MIL-125 and MIL-125_NH2 is su-
perior to that found for anatase (besides, the latter was sig-
nificantly less selective toward the formation of benzoquin-
ones) but inferior to the oxidant efficiency achieved pre-
viously with mesoporous titanium-silicate catalysts. The
XRD, N2 adsorption, and spectroscopic studies revealed
that, under turnover conditions of the liquid-phase oxi-
dation process, the structure of the Ti-MOFs undergoes
transformation into an X-ray amorphous mesoporous solid
that contains fine-dispersed amorphous TiO2, most likely,
stabilized by carboxylate ions. Such stabilization prevents
the formation of anatase microcrystallites, the presence of
which is detrimental to selective oxidations with hydrogen
peroxide. As a result, the recycling performance of the
Snurr.[21] The volumes of micro- and mesopores (Vmicro and Vmeso
)
were estimated by the αS-plot method.[22] XRD measurements were
performed with a high-precision X-ray diffractometer mounted on
beamline No. 2 of the VEPP-3 storage ring at the Siberian Synchro-
tron Radiation Center (SSRC). Radiation wavelength was λ =
0.15393 nm. High natural collimation of the synchrotron radiation
beam, the flat perfect crystal analyzer, and the parallel Soller slit
on the diffracted beam provided an extremely high instrumental
resolution of the diffractometer over a small angle region of 2θ =
0.5–10° and higher. The crystal size and morphology of the cata-
lysts were confirmed by a scanning electron microscope (SEM,
Philips, XL-30S FEG). Thermogravimetric (TG) analysis was car-
ried out with a thermogravimetric analyzer (TA Instruments, Uni-
versal V4.5A). The TG analysis was performed in a dry nitrogen
flow of 100 mLmin–1 while the temperature was increased from 25
to 700 °C with a heating rate of 10 °min–1. To get reproducible TG
data, prior to TG measurements the samples were hydrated in a
chamber with 70% relative humidity at 30 °C for 1 d.
Synthesis of MIL-125 and MIL-125_NH2: MIL-125 has been syn-
thesized by a solvothermal method using terephthalic acid and tita-
MIL-125-derived catalytic materials is superior to that of nium isopropoxide in a solvent mixture of N,NЈ-dimethylform-
amide (DMF) and methanol as reported elsewhere.[11] MIL-125_
mesoporous Ti,Si catalysts, which usually lose their activity
NH2 was prepared using 2-aminoterephthalic acid and titanium
and selectivity in the title reaction due to gradual aggrega-
isopropoxide in a mixture of solvents including DMF and dry
tion of TiO2 active species on the silica surface. Both the
methanol. The detailed preparation method is described in the SI.
structural stability of MIL-125 and efficiency of H2O2 utili-
The particle size of samples was controlled by the addition of ben-
zation over MIL-125 can be improved by reducing the con-
zoic acid as a modulator as used for the synthesis of UiO-66[23] and
centration of water in the reaction mixture. An alternative
route could be the use of a more hydrophobic organic linker
for synthesis of Ti-cluster-containing MOFs.
the addition of water [H2O/Ti(OiPr)4 molar ratio = 1] at the initial
mixing of the precursors.
Catalytic Experiments: Catalytic runs were carried out in thermo-
statted glass vessels under vigorous stirring at 80 °C. Typically, the
reactions were initiated by adding 35% aqueous H2O2 (0.4 m,
70 μL) to a mixture containing TMP (0.1 m, 28 mg), MIL-125
(3 mg), internal standard (biphenyl), and acetonitrile (2 mL) as a
solvent. The reaction time was 7–20 min. Samples of the reaction
mixture were withdrawn periodically during the reaction course by
Experimental Section
Materials: Hydrogen peroxide was used as a 35% or 88% solution
in water; its concentration was determined by iodometric titration.
Acetonitrile was dried and stored over activated molecular sieves
(4 Å). All other reactants were obtained commercially and used
a syringe through a septum. The reaction products were identified
without further purification.
1
by GC–MS and H NMR spectroscopic techniques and quantified
Instrumentation: GC analyses were performed using a gas chroma-
tograph “Tsvet-500” equipped with a flame ionization detector and
a quartz capillary column (30 mϫ0.25 mm) Supelco MDN-5S.
GC–MS analyses were carried out using an Agilent 7000B system
with triple-quadrupole mass-selective detector Agilent 7000 and
GC Agilent 7890B (quartz capillary column 30 mϫ0.25 mm/HP-
5ms). ATR-FTIR spectra (4000–250 cm–1, 32 scans, resolution
4 cm–1) were collected using a Cary 660 FTIR spectrometer (Ag-
ilent Technologies) and a PIKE Technologies GladiATR accessory
(diamond crystal). FT-Raman spectra (3600–100 cm–1, 300 scans,
resolution 4 cm–1, 180° geometry) were recorded using a RFS 100/
S spectrometer (Bruker). Excitation of the 1064-nm line was pro-
vided by a Nd-YAG laser (100 mW power output). DRS UV/Vis
measurements were performed with a Shimadzu UV/Vis 2501PC
by GC. Each experiment was reproduced at least three times. The
experimental error in the determination of the substrate conver-
sions and product yields normally did not exceed 2%. After reac-
tion, the catalyst was filtered off, washed with hot acetonitrile and
acetone, dried in air at room temperature, and then reused. Catalyst
reusability was studied in 4–6 time-scaled experiments.
2,3,5-Trimethyl-1,4-benzoquinone (TMBQ): GC–MS (EI): m/z (rel-
ative int.) = 150 (100) [M]+, 122 (47) [M – CO]+, 107 (64) [M –
CO, – CH3]+, 79 (48) [M – 2CO, – CH3]+. 1H NMR (CDCl3,
400.13 MHz, 25 °C): δ = 6.59 (s, 1 H, CH), 2.05–2.10 (m, 9 H, 3
CH3) ppm.
2,6-Di-tert-butyl-1,4-benzoquinone (DTBQ): 1H NMR (CD3CN,
400.13 MHz, 25 °C): δ = 6.48 (s, 2 H), 1.26 (s, 18 H) ppm.
1
spectrophotometer. H NMR spectra were recorded with a Bruker
Avance-400 spectrometer (400.13 MHz). Nitrogen adsorption at
77 K was studied by using an Autosorb-6B-Kr volumetric adsorp-
tion analyzer (Quantachrome Instruments, Boynton Beach, FL).
Before measurements, samples were degassed at 130 °C overnight
under vacuum (the final pressure was 5–6 mTorr). Specific BET
Supporting Information (see footnote on the first page of this arti-
cle): TG profiles, XRD patterns, N2 adsorption isotherms, FTIR
and DR UV/Vis spectra, SEM and TEM images, and kinetic curves
for catalytic reactions.
Eur. J. Inorg. Chem. 2014, 132–139
138
© 2014 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

www.eurjic.org
FULL PAPER
Patai, Z. Rappoport), Wiley-Interscience, New York, 1988; c)
W. Bonrath, M. Eggersdorfer, T. Netscher, Catal. Today 2007,
121, 45–57.
Acknowledgments
The authors thank Dr. I. E. Soshnikov for 1H NMR measurements
and Dr. M. S. Melgunov for discussions. Korean authors are grate-
ful to Global Frontier Center for Hybrid Interface Materials
(GFHIM) and ISTK for their financial support through the Global
Frontier R&D Program and the Institutional Collaboration Re-
search Program (SK-1301). The research was partially supported
by the Russian Foundation for Basic Research (grant number 13-
03-00413).
[10] a) O. A. Kholdeeva, I. D. Ivanchikova, M. Guidotti, N. Rava-
sio, Green Chem. 2007, 9, 731–733; b) O. A. Kholdeeva, I. D.
Ivanchikova, M. Guidotti, C. Pirovano, N. Ravasio, M. V. Bar-
matova, Y. A. Chesalov, Adv. Synth. Catal. 2009, 351, 1877–
1889; c) O. A. Kholdeeva, I. D. Ivanchikova, M. Guidotti, N.
Ravasio, M. Sgobba, M. V. Barmatova, Catal. Today 2009, 141,
330–336.
[11] F. Vermoortele, M. Maes, P. Z. Moghadam, M. J. Lennox, F.
Ragon, M. Boulhout, S. Biswas, K. G. M. Laurier, I. Beurroies,
R. Denoyel, M. Roeffaers, N. Stock, T. Düren, C. Serre, D. E.
De Vos, J. Am. Chem. Soc. 2011, 133, 18526–18529.
[12] a) X. Gao, I. E. Wachs, Catal. Today 1999, 51, 233–254; b)
N. N. Trukhan, V. N. Romannikov, A. N. Shmakov, M. P. Van-
ina, E. A. Paukshtis, V. I. Bukhtiyarov, V. V. Kriventsov, I. Yu.
Danilov, O. A. Kholdeeva, Microporous Mesoporous Mater.
2003, 59, 73–84.
[1] a) S. Kitagawa, R. Kitaura, S. Noro, Angew. Chem. 2004, 116,
2388; Angew. Chem. Int. Ed. 2004, 43, 2334–2375; b) G. Férey,
Chem. Soc. Rev. 2008, 37, 191–214; c) A. U. Czaja, N. Trukhan,
U. Müller, Chem. Soc. Rev. 2009, 38, 1284–1293; d) A. Corma,
H. García, F. X. Llabrés i Xamena, Chem. Rev. 2010, 110,
4606–4655; e) A. Dhakshinamoorthy, M. Alvaro, H. García,
Catal. Sci. Technol. 2011, 1, 856–867; f) Metal–Organic Frame-
works Applications from Catalysis to Gas Storage (Ed.: D. Far-
russeng) Wiley-VCH, Weinheim, Germany, 2011; g) M. Ranoc-
chiari, J. A. van Bokhoven, Phys. Chem. Chem. Phys. 2011, 13,
6388–6396.
[13] a) L. Marchese, T. Maschmeyer, E. Gianotti, S. Coluccia, J. M.
Thomas, J. Phys. Chem. B 1997, 101, 8836–8838; b) L. Mar-
chese, E. Gianotti, V. Dellarocca, T. Maschmeyer, F. Rey, S.
Coluccia, J. M. Thomas, Phys. Chem. Chem. Phys. 1999, 1,
585–592.
[2] M. Y. Masoomi, A. Morsali, Coord. Chem. Rev. 2012, 256,
[14] B. J. Aronson, C. F. Blanford, A. Stein, Chem. Mater. 1997, 9,
2921–2943.
2842–2851.
[3] a) A. Corma, H. García, Chem. Rev. 2002, 102, 3837–3892;
b) M. G. Clerici, in: Metal Oxide Catalysis, vol. 2 (Eds.: S. D.
Jackson, J. S. J. Hargreaves), Wiley-VCH, Weinheim, Germany,
2009, p. 705–754; c) B. Notari, Adv. Catal. 1996, 41, 253–334;
d) A. Tuel, L. G. Hubert-Pfalzgraf, J. Catal. 2003, 217, 343–
353; e) M. G. Clerici, Top. Catal. 2000, 13, 373–386; f) P. Rat-
nasamy, D. Srinivas, H. Knozinger, Adv. Catal. 2004, 48, 1–
169.
[15] O. A. Kholdeeva, M. S. Melgunov, A. N. Shmakov, N. N. Tru-
khan, V. V. Kriventsov, V. I. Zaikovskii, M. E. Malishev, V. N.
Romannikov, Catal. Today 2004, 91–92, 205–209.
[16] H. Jobic, B. J. Borah, S. Yashonath, J. Phys. Chem. B 2009,
113, 12635–12638.
[17] O. V. Zalomaeva, K. A. Kovalenko, Yu. A. Chesalov, M. S.
Mel’gunov, V. I. Zaikovskii, V. V. Kaichev, A. B. Sorokin, O. A.
Kholdeeva, V. P. Fedin, Dalton Trans. 2011, 40, 1441–1444.
[4] M. Dan-Hardi, C. Serre, T. Frot, L. Rozes, G. Maurin, C. San- [18] R. A. Sheldon, M. Wallau, I. W. C. E. Arends, U. Schuchardt,
chez, G. Férey, J. Am. Chem. Soc. 2009, 131, 10857–10859.
[5] C. Zlotea, D. Phanon, M. Mazaj, D. Heurtaux, V. Guillerm,
C. Serre, P. Horcajada, T. Devic, E. Magnier, F. Cuevas, G.
Férey, P. L. Llewellyn, M. Latroche, Dalton Trans. 2011, 40,
4879–4881.
[6] a) Y. Horiuchi, T. Toyao, M. Saito, K. Mochizuki, M. Iwata,
H. Higashimura, M. Anpo, M. Matsuoka, J. Phys. Chem. C
2012, 116, 20848–20853; b) Y. Fu, D. Sun, Y. Chen, R. Huang,
Z. Ding, X. Fu, Z. Li, Angew. Chem. 2012, 124, 3420–3423.
[7] S.-N. Kim, J. Kim, H.-Y. Kim, H.-Y. Cho, W.-S. Ahn, Catal.
Today 2013, 204, 85–93.
[8] N. D. McNamara, G. T. Neumann, E. T. Masko, J. A. Urban,
J. C. Hicks, J. Catal. 2013, 305, 217–226.
[9] a) P. Schudel, H. Mayer, O. Isler, in: The Vitamins, vol. 5 (Eds.:
W. H. Sebrell, R. S. Harris), Academic Press, New York, 1972,
p. 165; b) The Chemistry of Quinonoid Compounds (Eds.: S.
Acc. Chem. Res. 1998, 31, 485–493.
[19] a) G. Busca, G. Ramis, J. M. Gallardo Amores, V. S. Escribano,
P. Piaggio, J. Chem. Soc. Faraday Trans. 1994, 90, 3181–3190;
b) G. A. Tompsett, G. A. Bowmaker, R. P. Cooney, J. B. Met-
son, K. A. Rodgers, J. M. Seakins, J. Raman Spectrosc. 1995,
26, 57–62.
[20] J. S. Lee, K. H. You, C. B. Park, Adv. Mater. 2012, 24, 1084–
1088.
[21] K. S. Walton, R. Q. Snurr, J. Am. Chem. Soc. 2007, 129, 8552–
8556.
[22] S. J. Gregg, K. S. W. Sing, Adsorption, surface area and porosity,
Academic Press, New York, 1982.
[23] A. Schaate, P. Roy, A. Godt, J. Lippke, F. Waltz, M. Wiebcke,
P. Behrens, Chem. Eur. J. 2011, 17, 6643–6651.
Received: August 26, 2013
Published Online: December 5, 2013
Eur. J. Inorg. Chem. 2014, 132–139
139
© 2014 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim