Generating and optimizing the catalytic activity in UiO-66 for aerobic oxidation of alkenes by post-synthetic exchange Ti atoms combined with ligand substitution

Article abstract of DOI:10.1016/j.jcat.2018.07.032

The catalytic activity for the aerobic epoxidation of cyclooctene of UiO-66 has been introduced by post-synthetic ion exchange of Zr⁴⁺ by Ti⁴⁺ at the nodes and the performance optimized by nitro substitution in the terephthalate ligand. In this way a TON value of 16,600 (1660 considering Zr + Ti content) was achieved, comparing favorably with the highest catalytic activity reported in homogeneous for the same reaction (10,000 for γ-SiW₁₀{Fe³⁺(OH₂)}₂O₃₈^6?). Kinetic studies have shown that the most likely reactive oxygen species involved in the oxidation is superoxide, with hydroxyl radicals also contributing to the reaction. UiO-66(Zr_5.4 Ti_0.6)-NO₂ is stable under catalytic conditions, being used six times without any change in the conversion temporal profile and in the X-ray diffractogram. The scope of UiO-66(Zr_5.4 Ti_0.6)-NO₂ promoted aerobic oxidation of alkenes was expanded by including smaller rings cycloalkenes, as well as acyclic and aryl conjugated alkenes.

Full text of DOI:10.1016/j.jcat.2018.07.032

Journal of Catalysis 365 (2018) 450–463
Contents lists available at ScienceDirect
Journal of Catalysis
journal homepage: www.elsevier.com/locate/jcat
Generating and optimizing the catalytic activity in UiO-66 for aerobic
oxidation of alkenes by post-synthetic exchange Ti atoms combined with
ligand substitution
a
a,b,
Andrea Santiago-Portillo a, Sergio Navalón a, , Mercedes Álvaro , Hermenegildo García
⇑
⇑
a Departamento de Química and Instituto de Tecnología Química CSIC-UPV, Universitat Politècnica de València, Av. de los Naranjos s/n, 46022 Valencia, Spain
b Center of Excellence for Advanced Materials Research, King Abdulaziz University, Jeddah, Saudi Arabia
a r t i c l e i n f o
a b s t r a c t
Article history:
The catalytic activity for the aerobic epoxidation of cyclooctene of UiO-66 has been introduced by post-
synthetic ion exchange of Zr4+ by Ti4+ at the nodes and the performance optimized by nitro substitution in
the terephthalate ligand. In this way a TON value of 16,600 (1660 considering Zr + Ti content) was
Received 16 May 2018
Revised 22 July 2018
Accepted 24 July 2018
achieved, comparing favorably with the highest catalytic activity reported in homogeneous for the same
6À
reaction (10,000 for
c
-SiW10{Fe3+(OH2)}2O ). Kinetic studies have shown that the most likely reactive
38
oxygen species involved in the oxidation is superoxide, with hydroxyl radicals also contributing to the
reaction. UiO-66(Zr5.4 Ti0.6)-NO2 is stable under catalytic conditions, being used six times without any
change in the conversion temporal profile and in the X-ray diffractogram. The scope of UiO-66(Zr5.4
Ti0.6)-NO2 promoted aerobic oxidation of alkenes was expanded by including smaller rings cycloalkenes,
as well as acyclic and aryl conjugated alkenes.
Keywords:
Heterogeneous catalysis
Aerobic oxidation of alkenes
Metal-organic frameworks
UiO-66
Ó 2018 Elsevier Inc. All rights reserved.
1. Introduction
oxidations [7,14–23]. As solid catalysts, MOFs offer various possi-
bilities to introduce the catalytic activity by changing the nature
Transition metals have general catalytic activity to promote aer-
obic oxidations of hydrocarbons through different mechanisms,
some of them involving reactive oxygen species [1–3]. While the
use of oxygen as terminal oxidant, rather than peroxides or other
oxidizing reagents, is very appealing due to availability, greenness
and sustainability of the process, the lack of product selectivity
limits, in most of the cases, the applicability of aerobic oxidations
[3–9].
Metal-organic frameworks (MOFs) are crystalline porous solids
whose lattices are constituted by nodes of metal ions or clusters of
metal ions, most frequently transitions metals, coordinated by
rigid bi- or multidentate organic linkers [10–13]. Due to the high
metal content, high surface area and high porosity, among other
properties, MOFs are attracting much current interest as heteroge-
neous catalysts to promote organic reactions, including aerobic
of the metal ion [24–27] and also subsequent optimization by
introducing substituents on the organic linker that are able to con-
trol the electronic density of the active metal sites by inductive
effects [28–30]. Following the pioneer work of de Vos [30], we
and others [28,29,31–33] have shown that the activity of a given
MOFs can be increased over one order of magnitude in Lewis or
Brönsted acid catalyzed reactions [29] and some oxidations [28]
by introducing electron withdrawing substituents on the aromatic
ring of the terephthalate linker in MIL-101 [28,29,31] and UiO-66
[30,32,33].
Continuing with this line of research aimed at exploiting the
catalytic activity of MOFs, it would be of interest to create active
sites in otherwise inactive MOFs by ion exchange at the nodes
and, subsequently, optimization of their performance by substitu-
tion on the terephthalate linker [34–36]. In this regard, Li and
coworkers have shown that it is possible to replace in UiO-66
(Zr) a certain percentage up to about 30% of Zr4+ metal ions by
Ti4+ ions by post-synthetic modification of parent UiO-66(Zr) with
TiCl4(THF)2 through ion exchange [37]. This strategy has been used,
for instance, to increase the photocatalytic activity of parent UiO-
66(Zr) for CO2 reduction by introducing Ti4+ acting as electron relay
between photoexcited terephthalate ligand and Zr4+ [38].
Abbreviations: MOF, metal-organic framework; TON, turnover number.
Corresponding authors at: Departamento de Química and Instituto de
Tecnología Química CSIC-UPV, Universitat Politècnica de València, Av. de los
Naranjos s/n, 46022 Valencia, Spain.
⇑
E-mail addresses: sernaol@doctor.upv.es (S. Navalón), hgarcia@qim.upv.es
(H. García).
https://doi.org/10.1016/j.jcat.2018.07.032
0021-9517/Ó 2018 Elsevier Inc. All rights reserved.

A. Santiago-Portillo et al. / Journal of Catalysis 365 (2018) 450–463
451
To check this concept, and considering the known activity of
2.3. Catalyst characterization
Ti4+ grafted on mesoporous silicas and zeolites as oxidation cata-
lysts, particularly in the presence of peroxides as oxidizing
reagents, but also using oxygen as terminal oxidant [39–45], in
the present study we have focused on the reactivity of alkenes, par-
ticularly, cycloalkenes, to form the corresponding epoxide or allylic
ol/one mixtures. The novelty of our study derives from the consid-
eration that, as it will be described below, UiO-66(Zr) is devoid of
any significant activity in the aerobic oxidation of cycloalkenes, but
by applying known chemistry in the field of zeolites and aluminum
silicates grafting Ti4+ it is possible to introduce active sites to pro-
mote this aerobic oxidation. Furthermore, it will be shown that the
activity of UiO-66(Zr, Ti) can be increased by a factor of 8 when
NO2 is present on the terephthalate linker. In the area of Ti cata-
lysts grafted on micro- and mesoporous materials, it was found
that isolated titanium sites are active sites for epoxidation
[43,46,47] and similar isolated Ti4+ sites can be easily obtained in
UiO-66 by appropriate post-synthetic partial exchange Zr4+ by
Ti4+ ion. Therefore, up to now, the combination of the flexibility
of MOFs to exchange metal ions and introduce substituents in
the organic linker has not been combined with the concept derived
from Ti-containing zeolites of site isolated Ti atoms.
Powder X-ray diffraction (PXRD) patterns of UiO-66(Zr Ti)-X (X:
NH2, H and NO2) materials were recorded on a Philips XPert
diffractometer equipped with a graphite monochromator (40 kV
and 45 mA) employing Ni filtered Cu Ka radiation. ATR-FTIR spec-
tra of UiO-66(Zr, Ti)-X materials were collected at 20 °C using a
Bruker Tensor27 instrument. Previously, the solid samples were
heated in an oven (100 °C for 20 h) to remove physisorbed water.
N2 adsorption isotherms were recorded at 77 K using
a
Micromeritics ASAP 2010 instrument. Thermogravimetric mea-
surements were performed on a TGA/SDTA851e Mettler Toledo
station. Scanning electron microscopy (SEM, Zeiss instrument,
AURIGA Compact) having incorporated a EDX detector has been
employed to determine the morphology of the solid samples and
obtain the element mapping of selected areas, respectively. The
EDX detector has a limit of detection of about 1 wt%. X-ray photo-
electron (XP) spectra were collected on a SPECS spectrometer with
a MCD-9 detector using a monochromatic Al (Ka = 1486.6 eV) X-
ray source. Spectra deconvolution was performed with the CASA
software using the C 1s peak at 284.4 eV as binding energy [52].
The metal content (Zr and/or Ti) of solid samples have been
determined by ICP-OES analysis. Previously, the solid samples (5
mg) were digested using a HNO3 aqueous solution (3 M, 30 mL)
at 80 °C for 12 h. Then, the filtered aqueous samples were analyzed
by ICP-OES [37].
Although recent reports have confirmed incorporation of Ti in
UiO-66 they have proposed that it occurs at defective sites where
linkers are vacant through Ti attachment rather than ion
exchanged [48].
FTIR spectra of CO adsorption were recorded in a Nexus 8700
FTIR spectrophotometer using an IR cell allowing in situ treatments
at controlled temperature, from À176 °C to 500 °C, and connected
to a high vacuum system with gas dosing facility. For CO adsorp-
tion measurements the samples were pressed into self-supported
wafers and treated under vacuum (10À6 mbar) at 150 °C for 2 h.
After activation, the wafers were cooled down to À176 °C under
dynamic vacuum, followed by CO dosing at increasing pressure
(0.4–6 mbar). IR spectra were collected after each dosage. All IR
spectra corresponding to CO adsorption measurement have been
normalized to the weight of the IR wafer.
2. Experimental section
2.1. Materials
All the reagents and solvents used in this work were of analyt-
ical or HPLC grade and supplied by Merck.
2.2. Catalyst preparation
Isoestructural UiO-66(Zr)-X (X: H, NO2 and NH2) were prepared
following reported procedures [32,37,49,50]. Briefly, the corre-
sponding terephthalic acid derivative (1 mmol) and ZrCl4 (1 mmol)
were added to a Teflon-lined autoclave containing dimethylfor-
mamide (3 mL). The system was heated at the corresponding tem-
perature for the required period of time (Table 1). After this time,
the system was cooled down to room temperature and the result-
ing precipitate was first washed under stirring with DMF (40 mL)
for 2 h (3 times) and, then, the solid was washed with methanol
in a Soxhlet system for 12 h. Finally, the solid was dried in an oven
at 100 °C for 24 h.
The three UiO-66-X (X: NH2, H and NO2) solids were further
submitted to Zr4+ exchange using different percentages of Ti4+
following reported procedures [37,51]. Briefly, UiO-66-X (X: NH2,
H and NO2) solids (200 mg) were suspended in anhydrous DMF
(5 mL) and magnetically stirred for 96 h at 120 °C with freshly
prepared TiCl4(THF)2 complex. TiCl4(THF)2 is prepared immedi-
ately prior to its use by mixing TiCl4 (2.9 mL) in anhydrous
dichloroethane (50 mL) with 8.6 mL of THF in 100 mL of anhydrous
n-hexane.
2.3.1. Catalytic experiments
Briefly, the required amount of UiO-66(Zr Ti)-X (X: NH2, H and
NO2) employed as catalyst (0.016 mmol of total metal Zr + Ti) was
introduced into a reactor vessel (5 mL). Subsequently, the olefin
reagent (2 mmol) dissolved in CH3CN (2.5 mL) was added to the
vessel. The system was pressurized with O2 at the required value
at room temperature (i.e. 5 or 2 atm). The reactions were carried
out under 600 rpm magnetic stirring to ensure that the process is
under kinetic control.
Catalyst reusability was studied for the most active sample
(UiO-66(Zr5.4Ti0.6)-NO2). At the end of the reaction, the solid cata-
lyst was recovered by filtration (Nylon membrane, 0.2 lm) and,
transferred to a round-bottom flask (50 mL) and washed under
magnetic stirring with ethanol (20 mL) at 80 °C for 2 h. This proce-
dure was repeated three times. The washed, used solid catalyst
was recovered by filtration (Nylon membrane, 0.2 lm) and dried
in an oven at 100 °C for 24 h. Before the new catalytic cycle, the
solid catalyst was activated at 150 °C under vacuum for 16 h.
Selective radical quenching experiments were carried out fol-
lowing the general reaction procedure described above, but with
the addition of radical quenchers (20 mol% with respect to the sub-
strate). In particular, dimethylsulfoxide (DMSO) [53–56] or p-
benzoquinone [53,54,56,57] were added as selective hydroxyl or
superoxide/hydroperoxyl radical scavengers, respectively.
Activation energy for cyclooctene was estimated according the
Arrhenius law by plotting the natural logarithm of the initial reac-
tion rate of cyclooctene disappearance versus the reciprocal of the
absolute temperature, fitting the experimental points to the best
Table 1
Temperature and time employed for the preparation of UiO-66-X (X: NH2, H and NO2)
solids [32].
X
Temperature (°C)
Reaction time (h)
NH2
H
NO2
100
220
220
24
12
24

452
A. Santiago-Portillo et al. / Journal of Catalysis 365 (2018) 450–463
linear correlation. The amount of oxygen consumed was estimated
from the difference of the reactor pressure before and after the
reaction.
3. Results and discussion
3.1. Catalyst characterization
2.3.2. Product analysis
The list of catalyst prepared in the present study, the most rel-
evant textural properties and analytical data are summarized in
Table 2. As it can be seen there, three sets of isostructural UiO-66
having on the terephthalate linker a nitro or amino substituents,
as well as the parent terephthalate were prepared at four different
levels of Zr4+ by Ti4+ ion exchange. Preparation of isostructural UiO-
66-H, UiO-66-NO2, UiO-66-NH2 was carried out as reported in the
literature [32,37]. Briefly, the three UiO-66(Zr) were prepared by
solvothermal crystallization of ZrCl4 and terephthalic, 2-
nitroterephthalic and 2-aminoterephthalic ligand in DMF. The con-
ditions and the exact mass of the components are indicated in
Table 1 of the experimental section. The three UiO-66-X (X:H,
NO2, NH2) were further submitted to Zr4+ to Ti4+ exchange in
DMF at 120 °C by increasing the concentration of TiCl4(THF)2 com-
plex to afford for each case of linker substituted UiO-66-X new
mixed-metal materials with increasing Ti content ranging for 0–
10% [37,51].
Previously filtered reaction aliquots were diluted in acetonitrile
containing a known amount of nitrobenzene as external standard.
The aliquots were immediately analyzed by gas chromatography
using a flame ionization detector. Quantification was carried out
by using calibration curves of authentic samples against nitroben-
zene as standard. Mass balances for all the reactions were higher
than 95%. Product yields can be estimated by multiplying conver-
sion by selectivity.
2.3.3. Leaching experiments
At the end of the reaction the solid was removed by filtration
(0.2 lm nylon filter) and the solvent removed under vacuum in a
rotary evaporator. Then, an aqueous solution of HNO3 (3 M, 30
mL) was added to the flask reaction and the system heated at 80
°C for 24 h. Finally, the presence of Zr or Ti in the aqueous phase
was analyzed by ICP-OES.
Since Zr4+ ion forms with carboxylate strong coordinative
bonds, it has been found that is not possible to achieve a high level
of exchange. In the present case [59], the maximum Zr4+ to Ti4+
exchange level achieved was 10%. It has been proposed that Ti4+
exchange in UiO-66(Zr) occurs due to the intrinsic dynamics
around the Zr4+ ions that can change the instantaneous coordina-
tion number [60]. Precedents reporting Ti4+ exchanged in UiO-66
to improve the performance of this MOF for gas permeability
[61] or in photocatalysis can be found in the literature [22,48].
As it can be seen in Figs. 1 and S1–S2, PXRD of all the UiO-66-X
(Zr6-x Tix) samples exhibited identical pattern corresponding to a
highly crystalline material. As expected in view of the reported
data, the presence of Ti4+ in the material causes a gradual shift of
the most intense (1 1 1) peak from the initial 2 h value at 7.39° to
higher diffraction angles reaching 2 h values of 7.43°. This effect
of gradual shift in the position of the diffraction peaks has been
previously reported in the literature and has been rationalized as
a reflection of the somewhat smaller unit cell of UiO-66(Zr6ÀxTix)
due to the contraction caused by the smaller ionic radius of Ti4+
compared to Zr4+ [37,51]. Therefore, according to the literature
the gradual shift of the position of (1 0 0) diffraction peak observed
for each series of UiO-66(Zr Ti)-X (X: H, NO2 or NH2) is considered
here an evidence of substitution of Zr4+ by Ti4+ in the lattice
[37,51].
2.3.4. Colorimetric peroxide quantification of treated UiO-66
(Zr5.4Ti0.6)-H.
Prior to peroxide titrations, an acidic aqueous solution (100 mL)
of titanyl oxalate (2.5 g) in H2SO4(conc) (25 mL) and HNO3(conc) (1
mL) was prepared. Then, 10 mg of UiO-66 (Zr5.4Ti0.6)-H was heated
at 120 °C under 5 atm oxygen pressure in 2 mL of acetonitrile for
24 h. After this time, the solid was dissolved in the acidic titanyl
solution in where Ti@O2+ acts as peroxide colorimetric indicator
[58]. An aliquot sample (0.5 mL) was diluted with water (4.5 mL)
before measuring its absorbance at 420 nm. The difference in
absorbance between this sample and that of an analogous one that
was not submitted to the 24 h treatment was compared. Formation
of peroxide was not detected.
2.3.5. EPR measurements
UiO-66(Zr5.4 Ti0.6)-NO2 or UiO-66(Zr)-NO2 (5 mg) was added to
a two-neck round bottom flask (25 mL) containing the commer-
cially available spin trap N-tert-butyl-a-phenylnitrone (PBN,
1150 mg LÀ1) dissolved in n-dodecane. The system was sonicated
for 30 min and purged with a balloon containing O2. Then, the flask
is heated at 120 °C for 6 h. At this time, a reaction aliquot is filtered
(Nylon filter, 0.2 lm) and the sample was purged with N2 before
recording the EPR. EPR spectrum of each sample was acquired
using a Bruker EMX instrument operating at a frequency of
9.803 GHz with sweep width of 3489.9 G, a time constant of
40.95 ms, a modulation frequency of 100 kHz, a modulation width
of 1 G and a microwave power of 19.92 mW.
Supplementary data associated with this article can be found, in
the online version, at https://doi.org/10.1016/j.jcat.2018.07.032.
The different substitution of the terephthalate linker is con-
firmed by IR spectroscopy of the solids, where the characteristic
Table 2
List of catalysts employed in the present study with their corresponding BET surface area, pore volume, metal content and initial reaction rate for the catalytic aerobic epoxidation
of cyclooctene.a
Catalyst
BET surface area (m2 gÀ1
)
Pore volume (cm3 gÀ1
)
ICP Zr (%)
ICP Ti (%)
r0 (mM hÀ1
)
UiO-66(Zr)-NO2
600
520
580
590
800
770
740
740
780
720
730
770
0.60
0.46
0.52
0.50
0.75
0.78
0.68
0.77
0.85
0.50
0.52
0.51
100.00
99.49
98.40
89.60
100.00
99.42
98.80
89.40
100.00
99.40
98.50
89.60
-
2.02
6.34
7.01
9.83
1.25
3.83
4.98
6.94
1.20
2.38
3.97
6.09
UiO-66(Zr5.97 Ti0.03)-NO2
UiO-66(Zr5.94 Ti0.06)-NO2
UiO-66(Zr5.4 Ti0.6)-NO2
UiO-66(Zr)-H
UiO-66(Zr5.97 Ti0.03)-H
UiO-66(Zr5.94 Ti0.06)-H
UiO-66(Zr5.4 Ti0.6)-H
UiO-66(Zr)-NH2
0.51
1.60
10.40
–
0.58
1.20
10.60
-
UiO-66(Zr5.97 Ti0.03)-NH2
UiO-66(Zr5.94 Ti0.06)-NH2
UiO-66(Zr5.4 Ti0.6)-NH2
0.60
1.50
10.40
a
Catalyst (0.016 mmol of metal), substrate (2 mmol), CH3CN (2 mL), 120 °C and O2 (5 atm).

A. Santiago-Portillo et al. / Journal of Catalysis 365 (2018) 450–463
453
Fig. 1. PXRD patterns of UiO-66(Zr)-H (a), UiO-66(Zr5.97 Ti0.03)-H (b), UiO-66(Zr5.94 Ti0.06)-H (c), UiO-66(Zr5.4 Ti0.6)-H (d).
vibrational peaks associated to NO2 (1500 and 1380 cmÀ1) and NH2
(3350 and 1595 cmÀ1) can be observed (Figs. S3–S5). In contrast,
the partial substitution of Zr4+ by Ti4+ ions in the UiO-66(Zr)-NO2
materials does not cause significant changes in their FT-IR spectra
(Fig. S4).
Isothermal nitrogen adsorption measurements indicate that the
specific BET surface area and pore volumes of the UiO-66-X sam-
ples under study range from 890 to 520 m2 gÀ1 and 0.46 to 0.85
cm3 gÀ1, respectively (Table 2). These values and the general trend
observed, where the surface area of each parent UiO-66-X (X: H,
NO2 or NH2) based material, although similar, decreases along
the extent of the ion exchange, is in agreement with previous
reports [27,51].
Thermogravimetry under aerobic conditions offers important
information related to the presence of defects in the material, by
comparing the experimental residual weight after complete com-
bustion on the organic linker that is assumed to correspond to
Zr6-xTixO2 residue with the theoretical weight of the residue calcu-
lated based on the structural formula [28,29]. Thermogravimetric
profiles (Figs. S6–S8) exhibit an initial weight loss from room tem-
perature to 150 °C that corresponds to the desorption of co-
adsorbed water and solvent molecules. After this step, there is an
exothermic weight loss attributable to the combustion of the
organic linker that takes places in all the samples under study from
250 to 400 °C. The difference between the experimental residual
weight in the thermogravimetric profile at 800 °C with that calcu-
lated based on the molecular formula, either positive and negative,
indicates that excess or defective amount of metal oxides in the
material respect to the ideal value according to the formula. In
the present case, the excessive weight of the residue in the TGA
indicates either an excess of Zr/Ti in the material or vacancies of
terephthalate linkers in the solid as indicated in Table S1. It is
known in the state of the art that UiO-66(Zr) can present a large
density of defects that can be quantified in a simple way by the dif-
ference between the theoretical an experimental chemical analysis
[62]. It should be mentioned that the experimental error associated
to thermogravimetric measurements does not allow to determine
differences in the residual weight for those samples with the low-
est Ti exchange percentage. A summary of the relevant experimen-
tal thermogravimetric data and its comparison with the theoretical
value is presented in Table S1. As it can be seen there, the parent
UiO-66(Zr)-X (X: H, NO2 or NH2) before Ti exchange treatments
exhibit somewhat higher (ꢀ1 wt%) percentages of Zr than calcu-
lated according to the ideal cell formula that range from 0.3 for
UiO-66(Zr)-NH2 to 0.9% for UiO-66(Zr)-NO2, indicating that the
materials contain some defects associated to a larger than ideal
amount of Zr. Those defects can be, for instance, terephthalate
vacancies or the presence of amorphous zirconium oxide clusters

454
A. Santiago-Portillo et al. / Journal of Catalysis 365 (2018) 450–463
in the material. To put into context the data presented in Table S1
showing just a relatively minor higher-than-expected Zr content,
the density of defects in our case should be relatively low in com-
parison with other precedents reported in the literature, where Zr
excess higher than 10% have been determined [32,63]. Upon sub-
mitting UiO-66(Zr)-X to Ti exchange in the largest proportion
(10%), a change in the residual remaining weight in the material
due to the lower atomic mass of Ti (47.9 a.u.) respect to Zr (91.2
a.u.) should be expected according to the cell formula. Specifically,
the decrease in the residual weight due to the Zr to Ti exchange
should theoretically be between 2.4 wt% in the case of UiO-66
(Zr5.4 Ti0.6)-NO2 to 2.8 wt% for UiO-66(Zr5.4 Ti0.6)-H. Importantly
this decrease in the percentage of the residual weight of less than
3% in the residual mass measured was experimentally observed by
TG. Experimental measurements of the decrease of residue mass
range between À1.9 and À2.3% for UiO-66(Zr)-NO2 and UiO-66
(Zr)-NH2, respectively, as indicated in Table S1. These analytical
data are in agreement with the exchange of a heavier metal ele-
ment as Zr by another lighter as Ti in measurable degree.
in the partially substituted UiO-66(Zr6-xTix)-NO2 solids (Figs. S9–
S12). As expected the Ti/Zr atomic ratio increases in all cases with
the percentage of the ion exchanged, in agreement with the suc-
cessful incorporation of Ti4+ in the solid material. Fig. 2b–d also
shows a shift in the binding energies of Zr 3d, Ti 2p and O 1s peaks
as the Ti4+ content increases. There are precedents in the literature
suggesting that these shifts in XPS binding energy reflect the
weaker electronegativity probed by the metallic elements as the
Zr4+ ions are partially exchanged by Ti4+, and the same explanation
should apply in the present case [27,37,49,50,64].
Diffuse reflectance UV–Vis spectra for the series of the UiO-66
materials were recorded (Figs. 3 and S13–S14). As it has been
XPS analysis of UiO-66(Zr/Ti)-X (X: H, NO2 or NH2) also detects
the presence of titanium for those samples submitted to Ti
exchange. In the particular case of the UiO-66(Zr Ti)-NO2 series
the Ti/Zr atomic ratio increases in all cases with the percentage of
the ion exchanged and it was always higher than the values deter-
mined by other analytical methods (Fig. 2a). This is probably due to
the fact that XPS is a surface specific technique and, therefore, there
should probably be a gradient of titanium content that according to
XPS quantification compared to the bulk chemical analysis should
be higher on the external surface of the MOF particles, decreasing
this Ti/Zr ratio towards the interior of the UiO-66-NO2 crystallites.
Fig. 2a shows the survey XPS spectra revealing the presence of
the different elements in the UiO-66(Zr)-NO2 material, as well as
a-d
300
350
400
450
Wavelenght (nm)
250
300
350
400
450
Wavelenght (nm)
Fig. 3. Diffuse reflectance UV–Vis of UiO-66(Zr)-H (a), UiO-66(Zr5.97 Ti0.03)-H (b),
UiO-66(Zr5.94 Ti0.06)-H (c) and UiO-66(Zr5.4 Ti0.6)-H (d).
Ti/Zr atomic
a)
b)
raꢀo
Zr 3d
O 1s
Ti 2p
0.97
C 1s
Zr 3p
Zr 3d
0.34
0.27
d
c
b
a
175
180
185
190
1000 800 600 400 200
0
Binding energy (eV)
Binding energy (eV)
c) Ti 2p
d) O 1s
450
455
460
465
470
525
530
535
540
Binding energy (eV)
Binding energy (eV)
Fig. 2. XPS spectra of survey (a) and the high resolution scan of Zr 3d (b), Ti 2p (c) and O 1s (d) regions for the UiO-66(Zr6-xTix)-H sample.

A. Santiago-Portillo et al. / Journal of Catalysis 365 (2018) 450–463
455
reported [27,32], a red shift of the absorption on set was generally
observed, the partial substitution of Zr4+ ions by Ti4+ resulting in
extending the absorption of the parent UiO-66(Zr)-X (X: H, NO2
or NH2) material at longer wavelengths [27,51].
The morphology of the UiO-66(Zr)-X particles was visualized by
SEM (Fig. 4 and Tables S2–S4). SEM images of the UiO-66(Zr)-X
samples before Ti exchange show that the material is constituted
by crystals with defined shape in the micrometric scale, accompa-
nied by much smaller nanometric particles. EDS mapping of the
individual particles shows the homogeneous distribution of the ele-
ments according to their cell formula. Importantly, the morphology
of the particles changes somewhat during the ion exchange proce-
dure, the change being characterized by a decrease in the popula-
oxidation of cyclooctene in CH3CN. The only product observed in
all cases was the corresponding cyclooctene oxide with complete
selectivity (Fig. 5a) and complete mass balance. It was observed
that the presence of UiO-66 (0.8 mol%) does not introduce any sig-
nificant catalytic activity respect to the blank control (Fig. 6, S26
and S27). In contrast, it was observed that the catalytic activity
for the series of materials with the three differently substituted
terephthalates increases consistently with the percentage of the
Ti exchange and it is higher for the NO2 (Fig. 5), then for H
(Fig. S26) and the NH2 (Fig. S27) terephthalate substituted UiO-
66 series, being UiO-66(Zr)-NH2 the less active of the samples
under evaluation. The performance of the series of UiO-66(Zr Ti)-
X can be quantitatively evaluated by comparing the initial reaction
rates, determined from the slope of the time-conversion plots at
zero time. These values are summarized in Table 2.
Therefore, according to this table, the most active material was
UiO-66(Zr5.4 Ti0.6)-NO2 in where the beneficial influence of the
presence of Ti ions introducing active sites and nitro substituents
optimizing the performance are combined. Regarding the active
sites and the role of Ti4+, it should be noted that, as clearly seen
in Figs. S26 and S27, the catalytic activity of UiO-66(Zr)-H and
UiO-66(Zr)-NH2 is coincident with that of a blank control in the
absence of any solid. This indicates that in the absence of Ti,
UiO-66(Zr)-H and UiO-66(Zr)-NH2 do not exhibit any catalytic
activity. Accordingly, it is proposed that Ti4+ are the active sites
in this aerobic oxidation. Furthermore, as it can be seen in Table 2
and Fig. 6, S26 and S27, the initial reaction rate increases almost
linearly with the Ti content, up to a certain Ti/Zr exchange value
beyond with the increase in the initial reaction rate is less impor-
tant reinforcing the proposal that this metal is the site promoting
oxidation (Fig. S28). Since according to Table S1, the density of
defects, although low, also increases with the percentage of Ti
exchange, the possibility that the defects created in the process
of Ti4+ exchange are the real sites, rather than the Ti4+ ions or asso-
ciated with it, cannot be disregarded with the present catalytic
tion of large crystals and
a concomitant increase in the
percentage of much smaller nanometric particles. This morphology
change suggests that there is some breakdown of large particles
during the ion exchange. Elemental mapping at submicrometric
resolution by EDS shows that Ti is homogeneously distributed in
the particle (Fig. 4 and Figs. S15–S25). It should be, however, com-
mented that the low sensitivity of the EDS compared to XPS does
not allow quantification of the Ti/Zr atomic ratio for the UiO-66
(Zr6-xTix)-X samples with low Ti content and this Ti/Zr atomic ratio
is better estimated by XPS. At large percentages of Ti exchange for
which the Ti Ka1 images are better defined, a coincidence the dis-
tribution of Zr and Ti can be observed, confirming that Ti is present
at the same positions as Zr at submicrometric resolution. Fig. 4
shows selected images to illustrate these two pieces of information
provided by SEM, i.e., the increase in the percentage of smaller par-
ticles with the increase in the percentage of Ti and the coincidence
between the distribution of Zr and Ti in the individual particles.
3.2. Catalytic activity
As commented earlier, evaluation of the catalytic activity of
UiO-66(Zr, Ti)-X samples was initially carried out for the aerobic
Fig. 4. SEM images of UiO-66(Zr)-H (a), UiO-66(Zr5.97 Ti0.03)-H (b), UiO-66(Zr5.94 Ti0.06)-H (c) and UiO-66(Zr5.4 Ti0.6)-H (d). Mapping and spectrum of selected image of UiO-66
(Zr5.94 Ti0.06)-H (e-j).

456
A. Santiago-Portillo et al. / Journal of Catalysis 365 (2018) 450–463
Fig. 5. (a) Aerobic oxidation of cyclooctene by UiO-66 based catalysts; (b) Time–conversion plot for the aerobic oxidation of cyclooctene to cyclooctene oxide using UiO-66-
NO2 as catalyst. Legend: UiO-66(Zr5.4 Ti0.6)-NO2 (j), UiO-66(Zr5.94 Ti0.06)-NO2 (h), UiO-66(Zr5.97 Ti0.03)- NO2 (d) UiO-66(Zr)-NO2 (s) and blank control in the absence of
catalyst (▲). Reactions conditions: Catalyst (0.016 mmol of metal), substrate (2 mmol), CH3CN (2 mL), 120 °C and O2 (5 atm). c) Time-conversion plot for the aerobic oxidation
of cyclooctene to cyclooctene oxide under productivity test conditions using a low amount of UiO-66(Zr5.4 Ti0.6)-NO2. Legend: UiO-66(Zr5.4 Ti0.6)-NO2 (j) and blank control in
the absence of catalyst (h). Reaction conditions: Catalyst (0.006 mmol of metal), substrate (10 mmol), CH3CN (3 mL), 120 °C and O2 (5 atm).
data. However, if defects of other nature unrelated to the presence
of Ti4+ were responsible for the catalytic activity, no direct relation-
ship between the catalytic activity and the percentage of Ti4+
should be expected. Therefore, the present kinetic data rather sup-
ports the role of Ti4+ as active sites.
the stability of UiO-66(Zr5.4 Ti0.6)-NO2 chemical analysis of the liq-
uid phase was carried out to determine the absence of Zr and Ti,
experimental data indicating that the amount of Zr and Ti leached
is less than 0.5%. Moreover, crystallinity of UiO-66(Zr5.4 Ti0.6)-NO2
after its use as catalyst was confirmed by recording PXRD pattern
upon reuse, whereby no significant changes in the crystallinity of
the material were observed. Therefore, all the available data, the
very low Zr4+ and Ti4+ leaching, coincidence of the temporal pro-
files upon reuse, lack of variation of PXRD patterns, indicate the
stability of the material under the reaction conditions.
Additional control experiments were performed using 2-
nitroterephthalic acid as promoter or ZrO2 or TiO2 at the percent-
age contained in UiO-66(Zr Ti)-NO2 or just the amount measured
by leaching. In particular, the presence of 2-nitroterephtalic acid
has not any activity to promote cyclooctene epoxidation. The
results presented in Fig. 7 show that the catalytic activity of UiO-
66(Zr Ti)-NO2 is much higher in terms of initial reaction rate or
final conversion than any of these five controls. Besides, the cat-
alytic of mixed-metal UiO-66(Zr Ti)-NO2 was also notably much
higher than that of MIL-125(Ti), showing the benefits of large
porosity and Ti distribution of UiO-66(Zr Ti)-NO2 respect to MIL-
125(Ti). Comparison of the performance of UiO-66(Zr Ti)-NO2
and MIL-125(Ti) illustrates the importance of Ti site isolation at
the Zr5Ti(O)4(OH)4 nodes and the large cavities of UiO-66.
To understand the role of UiO-66(Zr Ti)-X as catalyst in this aer-
obic oxidation, as well as to determine the heterogeneity of the
catalysis, a hot filtration test in which the temporal profile of
two twin reactions, one of them in the presence of the catalyst
and the other initiated with the solid catalyst and, then, removing
the solid at the reaction temperature and the resulting clear solu-
tion allowed to continue the reaction in the absence of solid were
compared. The results obtained are presented in Fig. 8. As it can be
seen there, in one of the two parallel reactions the catalyst was fil-
tered at 0.5 h reaction time, when cyclooctene conversion was
about 20%. Comparison of the two time-conversion plots shows
that, although the reaction in the absence of catalyst progressed
about 50% less than in the presence of the solid catalyst, there is
a substantial cyclooctene to cyclooctene oxide conversion in the
Using the most active UiO-66(Zr5.4Ti0.6)-NO2 material a produc-
tivity test was performed as well as the sample was submitted to
consecutive reuses. The purpose is to determine the stability of
UiO-66(Zr5.4 Ti0.6)-NO2 under catalytic conditions, by using, in
one case, a large excess of substrate with respect to the amount
of catalyst (productivity test) and, in the other, to establish the
variation of the temporal evolution upon consecutive reuses. For
the productivity test (Fig. 5b) complete cyclooctene conversion
with complete selectivity toward cyclooctene oxide was achieved
in 6 days. In this way a turnover number (TON) of 1,660 consider-
ing the total amount of metal as the active site, or 16,600 if only the
titanium content is considered was estimated. This TON value as
well as the TOF (6640 hÀ1 at 20% conversion) compare favorably
with other reports in homogeneous aerobic oxidation using
heteropolyacids (TON 10,000 and TOF 22 hÀ1 at 20% conversion
6À
for
c
-SiW10{Fe3+(OH2)}2O
;
reaction conditions: catalyst 1.5
38
lmol, solvent 1,2-dichloroethane/acetonitrile 1.5/0.1 mL, substrate
18.5 mmol, O2 1 atm, 83 °C) [65]. And even though others in
heterogeneous catalysis that oxidize cycloalkenes by H2O2 employ-
ing CH3ReO3 (TON 20,000 and TOF 666 hÀ1 at 20% conversion;
reaction conditions: 0.001 mmol CH3ReO3, 20 mmol cyclooctene,
40 mmol H2O2, 2 mmol 3-methylpyrazole in 10 mL CH2Cl2, 25°C)
[66] and Ta2O5/SiO2 (TON 150 and TOF 720 hÀ1 at 40% of conver-
sion; reaction conditions: 30 mg of catalyst, 6 mL CH3CN, 300 mL
DCB, 0.77 M cyclooctene, 0.51 M H2O2, 60 °C) [67].
It should be noted, however, that comparison of the absolute
catalytic activity from literature data is always not fair due to the
different temperature and conditions employed in the oxidation.
Very similar temporal profiles for cyclooctene disappearance
were observed for six consecutive reuses of UiO-66(Zr5.4 Ti0.6)-
NO2 (Fig. 6). The minor changes in the initial reaction rates in the
recycling test are probably due to the incomplete recovery of the
solid in the filtration of the catalyst after the reaction. To confirm

A. Santiago-Portillo et al. / Journal of Catalysis 365 (2018) 450–463
457
Fig. 6. (a) Reusability experiments employing UiO-66(Zr5.4 Ti0.6)-NO2 as catalyst in the aerobic oxidation of cyclooctene to cyclooctene oxide. Legend: 1st use (j), 2nd use
(h), 4th use (d) and 6th use (s). (b) PXRD of fresh UiO-66(Zr5.4 Ti0.6)-NO2 (1), three times used UiO-66(Zr5.4 Ti0.6)-NO2 (2) and six times used UiO-66(Zr5.4 Ti0.6)-NO2 (3).
Reactions conditions: Catalyst (0.016 mmol of metal), substrate (2 mmol), CH3CN (2 mL), 120 °C and O2 (5 atm).
Fig. 7. Time-conversion plots for the aerobic oxidation of cyclooctene to
Fig. 8. Time-conversion plots for the aerobic oxidation of cyclooctene to
cyclooctene oxide using different catalysts. Legend: UiO-66(Zr5.4, Ti0.6)-NO2 (j),
MIL-125 (Ti) (h), ZrO2 (d) TiO2 (s) ZrO2 with the amount of metal leached (▲), TiO2
with the amount of metal leached (D) and 2-nitroterephthalic acid with the amount
of ligand that in MOF (.). Reactions conditions: Catalyst (0.016 mmol of metal),
substrate (2 mmol), CH3CN (2 mL), 120 °C and O2 (5 atm).
cyclooctene oxide using UiO-66(Zr5.4 Ti0.6)-NO2 as catalyst that is filtered at 0.5 h
reaction time or that remains in contact with the substrate for the full reaction time.
Legend: UiO-66(Zr5.4 Ti0.6)-NO2 (j), after catalyst filtration at 0.5 h (h), after
catalyst filtration at 0.5 h and 20 mol% of TEMPO were added (d) and after catalyst
filtration at 3 h and 20 mol% of TEMPO were added(s). Reactions conditions:
Catalyst (0.016 mmol of metal), substrate (2 mmol), CH3CN (2 mL), 120 °C and O2
(5 atm).
clear solution after removal of the catalyst. It is worth to comment
that according to Fig. 5, under the reaction conditions, some
autooxidation of cyclooctene (about 30%) occurs in the absence
of any catalyst. However, in the present case, after filtration of
the catalyst, the conversion of cyclooctene increases from 20 to
50% that is significantly higher than the non-catalytic conversion
expected according to the blank reaction (about 20% for the same
reaction time, Fig. 5). Overall, the catalytic data indicate that after
initiation of the oxidation by UiO-66(Zr5.4 Ti0.6)-NO2 some reactive
oxygen species or leached species or reaction intermediates pre-
sent in the liquid phase may contribute somewhat (ꢀ30%) to the
overall conversion observed for the catalytic experiment without
removal of the solid.
Possible species present in the liquid phase that could con-
tribute to the progress of the reaction after removal of the catalyst
include H2O2, organic hydroperoxides and peroxides as well as col-
loidal ZrOx or TiOx nanoparticles. Since the amount of O2 consumed
in the reaction, as determined from the decrease in the reactor
pressure, is stoichiometric with cyclooctene conversion within
the error limit (0.975 mmol), it can be inferred that the percentage
of this oxygenated intermediates, including any possible reaction
with the solvent, should be low, barely in the detection limit.
To address the origin of the progress of cyclooctene oxidation in
the hot filtration test, analysis of the Zr and Ti content in the liquid
phase after filtration of the catalyst at 0, 5 h in hot was performed.
The results show that there was a negligible amount of Zr and Ti
equivalent of 0.5% of the metal content present in the catalyst. It
should be commented, however, that although the leached Zr
and Ti metals was very small, there are precedents in the literature
in where it has been found that small amounts of transition metals
in the ppm range are sufficient to accelerate autooxidation reac-
tions in a measurable extent [56]. Therefore, some contribution
of these leached metal species cannot be totally disregarded.
To address this possible contribution of leached Zr and Ti spe-
cies, we performed the previously commented control experiments
shown in Fig. 7 in where aerobic oxidation of cyclooctene was pro-
moted using ZrO2 and TiO2, both at the concentration determined
in the leaching analysis and at the concentration corresponding to
the total amount of Zr and Ti present in UiO-66(Zr5.4 Ti0.6)-NO2. In
both cases it was observed that although particularly TiO2 exhibits
some catalytic activity, it was much lower than that of UiO-66(Zr5.4
Ti0.6)-NO2. These catalytic experiments suggest some minor

458
A. Santiago-Portillo et al. / Journal of Catalysis 365 (2018) 450–463
activity due to the leached Zr and Ti species, but far from that mea-
sured for UiO-66(Zr5.4 Ti0.6)-NO2. On the other hand, it is also
known that autoxidation occurs through a chain mechanism, in
where chain propagation may result in the formation of a large
number of product molecules with just a single initiation event
[56]. Normally, the length of this propagation cycle is not relatively
long when radicals are generated in cages and porous materials
due to restrictions to the diffusion of these reaction intermediates
[56]. However, when the catalyst is filtered, these free radicals pre-
sent in the liquid phase could exhibit a much longer propagation
chain length and in that way, they could contribute to the conver-
sion of cyclooctene. In the related precedents in the literature, it
was also observed that hot filtration does not stop aerobic oxida-
tions of benzylic hydrocarbons and the role of the solid material
was proposed to be more as solid initiator of carbon-centred radi-
cals than a real catalyst in where active centers present in the solid
carry out every turnover converting one substrate molecule into a
product in each of them. Similarly here, it seems that, although not
completely, UiO-66(Zr5.4Ti0.6)-NO2 could behave partially as solid
initiator [28]. In the present case of UiO-66(Zr5.4Ti0.6)-NO2 a higher
contribution about 65% of the total conversion appears to corre-
spond to a true catalyst.
addition of TEMPO to the clear solution after removal of the reac-
tion completely stops further cyclooctene conversion under the
reaction conditions (Fig. 8). Moreover, if after the hot filtration test
at 20% conversion, TEMPO is added when conversion was 42% at
4.5 h, then, additional progress of cyclooctene conversion immedi-
ately stops again at this point. These experiments with TEMPO con-
clusively demonstrate that carbon centered radicals derived from
cyclooctene are the reaction intermediates responsible for the pro-
gress of the reaction after filtration of the solid catalyst and that
these carbon centered radicals do survive for hours in the reaction
medium after filtration of the catalyst.
3.2.1. Reaction mechanism
To gain further insights into the reaction mechanism and,
specifically, to obtain some evidence about the nature of the possi-
ble reactive oxygen species involved in the cyclooctene oxidation,
quenching experiments using p-benzoquinone and DMSO were
carried out. p-Benzoquinone is a selective quencher of hydroperox-
ide/superoxide [53,54,56,57], while DMSO quenches selectively
hydroxyl radicals [53–56]. Thus, comparison of the performance
of the cyclooctene oxidation in the absence and in the presence
of quenchers (20% molar proportion respect to substrate) could
provide valuable information on the nature of the active oxygen
species responsible for the reaction. The results of the quenching
tests are presented in Fig. 9. It should be noted that these quench-
ers are added at 0.5 h reaction time, when cyclooctene conversion
was already about 20% and possible intermediates have already
been generated. Also, the amount of quencher was in substoichio-
metric ratio respect to the substrate. Fig. 9 shows that the reaction
stops almost completely upon addition of p-benzoquinone indicat-
To provide some evidence on the nature of the reaction inter-
mediates responsible for the progress of cyclooctene oxidation
after removal of the solid catalyst in the hot filtration test, 2,6,6-
tetramethylpiperidine 1-oxyl (TEMPO) was used as a quencher. It
is well-known in the literature that TEMPO reacts with carbon cen-
tered radicals trapping them and stopping the reaction [56].
Quenching experiments by TEMPO were carried out with 20 mol
% of this quencher respect to cyclooctene. It was observed that
100
a)
c)
80
60
40
20
0
2
1
0
5
10
20
25
Time (1h5)
3490
3500
3510
3520
3530
3540
Magnetic Field (G)
d)
b)
3475
3493
3510
3527
3545
Fig. 9. (a) Time-conversion plots for the aerobic oxidation of cyclooctene to cyclooctene oxide using UiO-66(Zr5.4 Ti0.6)-NO2 as catalyst. Legend: In the absence of any
quencher (j), upon addition of 20% DMSO (d) and upon addition of 20% p-benzoquinone (h). Reactions conditions: Catalyst (0.016 mmol of metal), substrate (2 mmol),
CH3CN (2 mL), 120 °C and O2 (5 atm), quencher addition was made at 20% cyclooctene conversion. (b) Proposed pathways for oxygen activation in the alkene oxidation using
UiO-66(Zr5.4 Ti0.6)-NO2 as catalyst. (c) Experimental EPR spectra obtained by using PBN + O2 (1) and UiO-66(Zr5.4 Ti0.6)-NO2 + PBN + O2 using n-dodecane as solvent at 120 °C
for 4 h. (d) Experimental (black) and simulated (red) EPR spectra of PBN-OOH adduct when using UiO-66(Zr5.4 Ti0.6)-NO2 + PBN + O2 using n-dodecane as solvent at 120 °C for
4 h. Experimental hyperfine coupling constants of PBN-OOH coinciding with those reported in literature AGN = 14.0 and AGH = 2.05.[28,57,71].

A. Santiago-Portillo et al. / Journal of Catalysis 365 (2018) 450–463
459
ing that hydroperoxide/superoxide are likely to be involved in the
process. More surprising was, however, the observation that
DMSO, a selective hydroxyl radical quencher, also decreases signif-
icantly cyclooctene conversion, although it does not stop com-
pletely the reaction. Overall, the combined information of the
kinetic data shown in Fig. 9 can be understood considering that
superoxide would be the primary oxygen radical species, and
therefore, would be involved somehow in the formation of all the
reaction product. In addition, more aggressive hydroxyl radicals
should also be involved, but these hydroxyl radicals should not
be the only reactive oxygen species because they do not account
for the whole product formation. Moreover, they would be formed
as secondary intermediates from superoxide. Fig. 9b proposes sev-
eral pathways in which hydroxyl radicals could be generated.
Accordingly, it seems that the main role of solid UiO-66(Zr5.4
Ti0.6)-NO2 solid material is to open additional pathways for oxygen
activation and generation of reactive oxygen species (Fig. 9b). EPR
measurements using the UiO-66(Zr5.4Ti0.6)-NO2 as solid catalyst
confirm the formation of hydroperoxy radicals by detecting its
adduct with PBN-OOH (Fig. 9d) [68–70]. It should be noted that
although PBN-OOH adduct is observed in the absence as well as
in the presence of UiO-66(Zr5.4Ti0.6)-NO2 catalyst, the intensity of
the PBN-OOH signals was larger in the presence of the solid, sug-
gesting that ÁOOH concentration is higher when UiO-66
(Zr5.4Ti0.6)-NO2 is present. It should be noted, however, that degra-
dation of PBN under oxidative conditions makes not possible a
quantitative estimation of ÁOOH radical.
the detection of organic peroxides generated from the linker, an
additional experiment in which UiO-66(Zr5.4, Ti0.6)-H was submit-
ted to the reaction conditions at 120 °C and 5 bar oxygen pressure
in acetonitrile in the absence of any substrate was performed.
These conditions were considered adequate to form peroxides of
the organic linker, because they are similar to the reaction condi-
tions except that no cyclooctene is present. After 24 h treatment,
the heated UiO-66(Zr5.4, Ti0.6)-H sample was submitted to a stan-
dard colorimetric test to detect peroxides (see experimental sec-
tion), but no evidence of the formation of these species could be
obtained. The failure to detect organic peroxides in UiO-66(Zr5.4
,
Ti0.6)-H is in agreement with the reluctance of electron poor aro-
matics to form peroxides.
As commented before, the catalytic activity of UiO-66(Zr)-Ti
increases both upon partial replacement of Zr4+ ions by Ti4+ atoms
and by the presence of NO2 on the terephthalate organic ligand.
The most active catalyst for the aerobic oxidation of olefins is the
UiO-66(Zr5.4Ti0.6)-NO2 material (Fig. 5a). In the introduction sec-
tion it has been already indicated that isolated Ti4+ ions in a zeolitic
framework have been previously reported to be active to promote
oxidation reactions acting as Lewis acid [39,41]. We speculate that
the Lewis acidity of the parent UiO-66(Zr5.4Ti0.6)-H solid should
increase or decrease due to the presence of electron withdrawing
or donor groups such as NO2 or NH2, respectively [29]. In the case
of Ti-containing zeolites it has been established that the role of Ti4+
in the epoxidation is to act as Lewis acid site forming a Ti-OOH spe-
cies that increases the electrophilicity of the peroxy group, facili-
tating its attack to the C@C double bond [40]. In fact, a linear
The role of molecular O2 in the resulting catalytic activity was
further confirmed by observing a linear relationship between the
oxygen pressure and the initial reaction rate for cyclooctene epox-
idation (Fig. 10a). This observation is in agreement with the pro-
posed radical mechanism in which molecular O2 reacts with
carbon center radicals to form hydroperoxyl radicals as the rate
determining step when using UiO-66(Zr5.4 Ti0.6)-NO2 as catalyst.
The apparent activation energy for the aerobic cyclooctene oxida-
tion was estimated by performing a series of reactions in the range
of temperature between 110 and 140 °C (Fig. 10b). From the Arrhe-
nius plot of the natural logarithm of the initial reaction rate (r0)
against the inverse of the absolute temperature a value for the
apparent activation energy of 36 kJ molÀ1 was determined. This
value is similar, but somewhat lower, to other activation energy
values for aerobic oxidation of aromatic hydrocarbons using MOFs
as catalysts or promoters [56].
relationship between the meta Hammett constant (rm) reported
in the literature [30] and the logarithm of r0 was found
(Fig. 11b). To gain insight about the influence of NO2 substitution
on the acidity, an in situ FT-IR CO adsorption study on the series
of UiO-66(Zr Ti)-X materials was carried out. It was observed that,
the partial substitution of Zr by Ti atoms in the UiO-66(Zr5.4 Ti0.6)-
NO2 material does not change the wavenumber of the Zr4+-CO
interaction. This fact is not unexpected considering the similar
Lewis acidity of Zr4+ and Ti4+ ions. On the other hand, the presence
of electron donor or withdrawing groups such as NH2 or NO2,
respectively, causes a notable shift of the metal-CO frequency to
lower or higher values, respectively. This observation correlates
well with the proposed decrease or increase of the Lewis acidity
of the UiO-66(ZrTi) material when bearing -NH2 or NO2 groups,
respectively. We speculate that an increase in the Lewis acidity is
beneficial for HOOÁ activation, once formed from molecular O2
and, then, epoxidation rates increases due to the higher polariza-
tion of the peroxy O-O bond [39,72].
One additional point of interest was to establish whether or not
the linker plays any role in the generation of reactive oxygen spe-
cies, as for instance, by forming some organic peroxide. Aimed at
Fig. 10. (a) Influence of the oxygen partial pressure on the resulting catalytic activity of UiO-66(Zr5.4 Ti0.6)-NO2 for cyclooctene oxidation. The inset shows the linear
relationship between the initial reaction rates of cyclooctene oxidation versus the oxygen partial pressure. Legend: oxygen (j), air (h), argon for 25 h and after 25 h oxygen
(d). (b) Time-conversion plots for the aerobic oxidation of cyclooctene to cyclooctene oxide using UiO-66(Zr5.4 Ti0.6)-NO2 as catalyst carried out at four different
temperatures. The inset shows the Arrhenius plot for the aerobic oxidation cyclooctene to cyclooctene oxide based on the initial reaction rates r0 obtained from the time-
conversion plot. Legend: 110 °C (h), 120 °C (j), 130 °C (d), 140 °C (s). Reactions conditions: catalyst (0.016 mmol of metal), substrate (2 mmol), CH3CN (2 mL) and O2 (5
atm).

460
A. Santiago-Portillo et al. / Journal of Catalysis 365 (2018) 450–463
Fig. 11. (a) Time-conversion plots for cyclooctene epoxidation by O2 catalyzed by a series of UiO-66(Zr/Ti)-X catalysts. Legend: absence of catalyst (1), UiO-66(Zr)-NO2 (2),
UiO-66(Zr5.4 Ti0.6)-NH2 (3), UiO-66(Zr5.4 Ti0.6)-H (4) and UiO-66(Zr5.4 Ti0.6)-NO2 (5). Reaction conditions: catalyst (0.016 mmol of metal of Zr + Ti), substrate (2 mmol), CH3CN
(2 mL), 120 °C and O2 (5 atm); (b) Plot of the logarithm of the relative rate constants as a function of the meta Hammett constant (rm) of the substituent present on the
terephthalate linker; (c) FTIR spectra of CO adsorption at 6 mbar and À176 °C on UiO-66(Zr)-NO2 (1) and UiO-66(Zr5.4 Ti0.6)-NO2 (2). (d) FTIR spectra of CO adsorption at 6
mbar and À176 °C on UiO-66(Zr5.4 Ti0.6)-H (1) and UiO-66(Zr5.4 Ti0.6)-NO2 (2) and UiO-66(Zr5.4 Ti0.6)-NH2 (3).
3.2.2. Scope of the catalyst
the reaction of cyclohexene takes place predominantly inside the
porous of UiO-66(Zr5.4 Ti0.6)-NO2 near the active sites. The contri-
bution due to free-radical oxidation outside the pores of UiO-66
(Zr5.4 Ti0.6)-NO2 in the solution accounting for the 15% observed
in the hot filtration test (Fig. 12).
The present study about the aerobic oxidation of cyclooctene
was expanded to determine the scope of UiO-66(Zr5.4 Ti0.6)-NO2
as catalyst. It is known in the literature that the size of the
cycloalkene ring has a strong influence on the product distribution
and, particularly, on the ratio between C@C epoxidation versus
allylic oxidation that decreases from cyclooctene as the size of
the ring decreases [53,54,73]. A similar trend was observed herein
(Fig. 12). Thus, while for cyclooctene the corresponding epoxide
was the only product observed with almost complete selectivity,
in the case of cycloheptene formation of the corresponding oxide
was accompanied by formation of 2-cycloheptenone with a selec-
tivity that changes as a function of the conversion Fig. 12 shows the
corresponding time-conversion and conversion-selectivity plots
for cycloheptene oxidation using UiO-66(Zr5.4 Ti0.6)-NO2 as cata-
lyst. As expected, in the case of cyclohexene, selectivity towards
cyclohexene oxide was very low, being the allylic oxidation
towards the corresponding ol/one mixture the major process with
a selectivity value at final reaction time about 85% (Fig. 12). One
specific feature of cyclohexene is the shorter chain length of the
propagation step in the autooxidation process compared to the
case of cyclooctene. This is reflected by the fact that the hot filtra-
tion test for cyclohexene results upon removal of the solid UiO-66
(Zr5.4 Ti0.6)-NO2, in a much lower progress of the conversion (15%)
as compared to the case of cyclooctene (35%). Also, due to the
shorter chain length and low concentration of active species out-
side the solid, in the case of cyclohexene, quenching experiments
by p-benzoquinone and DMSO did not were positive, meaning that
To confirm that similarly to cyclooctene, also cyclohexene epox-
idation occurs through carbon centered radical and that the differ-
ence in the percentage of progress of substrate conversion is due to
the different radical chain length between cyclooctene and cyclo-
hexene, after hot filtration of the catalyst TEMPO (20 mol%) was
added to the clear solution of cyclohexene oxidation in the absence
of solid catalyst and the mixture was allowed to react under the
reaction conditions. No further progress of cyclohexene conversion
was observed in the experiments where TEMPO as carbon radical
quencher was present in comparison to the hot filtration test in
the absence of TEMPO. Therefore, also for cyclohexene oxidation,
carbon centered radicals are the reaction intermediates. Thus, the
different progress of cyclohexene and cyclooctene after filtration
of the catalyst in the hot filtration tests has to be attributed to
the differences in the relative length on the propagation steps of
the radical chain mechanism (Figs. 8 and 12).
UiO-66(Zr5.4 Ti0.6)-NO2 was also able to promote the aerobic
epoxidation of linear terminal alkenes such as 1-octene. In this
case, kinetics is slower than for the cyclic analogues. This can be
easily rationalized considering that diffusion of linear alkenes in
porous materials is always slower than for the corresponding cyclic
analogues and/or alternatively, the intrinsic activity of mono sub-
stituted terminal alkenes is lower than that of disubstituted cyclic

A. Santiago-Portillo et al. / Journal of Catalysis 365 (2018) 450–463
461
Fig. 12. Time-conversion plots for the aerobic oxidation of cycloheptene (a) or cyclohexene (c) using UiO-66(Zr5.4 Ti0.6)-NO2 as catalyst. Selectivity-conversion plots for the
aerobic oxidation of cycloheptene (b) or cyclohexene (d) to the corresponding reaction products as indicated in the plots. Legend c: UiO-66(Zr5.4 Ti0.6)- NO2 (j), after catalyst
filtration at 13 h. (h) and after catalyst filtration at 13 h and 20 mol% of TEMPO were added (d). Reactions conditions: Catalyst (0.016 mmol of metal), substrate (2 mmol),
CH3CN (2 mL), O2 (5 atm) and 120 °C.
compounds. What is remarkable in any case is the high selectivity
of the UiO-66(Zr5.4 Ti0.6)-NO2 catalyst for the aerobic epoxidation
that results in almost complete selectivity towards 1-octene oxide
as it can be seen in Fig. S29.
dimensions of the latter. These observations are in agreement with
previous reports on shape-selectivity on medium-pore size zeolites
that have reported a smaller kinetic diameter for trans-stilbene
(6.5 Å) compared with the larger cis isomer (9.1 Å) [75]. Alterna-
tively, the possibility that the cis isomer exhibits lower reactivity
respect to the trans isomer cannot be ruled out. Thus, in contrast
to the case of cyclooctene, where some degree of conversion occurs
in the liquid phase, the lack of reactivity of cis-stilbene indicates
that, for conjugated alkenes, the contribution of oxidation outside
the pores should negligible. This observation is in accordance with
basic knowledge in organic chemistry about the influence of chem-
ical structure of organic compounds and their reactivity.
The scope of UiO-66(Zr Ti)-X as promoter of the aerobic oxida-
tion of alkenes was expanded to aryl conjugated alkenes by study-
ing the reaction of styrene and stilbene. In the case of styrene,
oxidation gives rise to a mixture in which the product selectivity
changes with conversion, the major product at final conversion
being benzaldehyde with a selectivity of 40%, accompanied by
phenylacetaldehyde (15%), acetophenone (15%) and lesser
amounts of styrene oxide (8%) and benzoic acid (8%) (Fig. S30). For-
mation of all these products can be easily explained as derived
from two competitive pathways. On one hand, the oxidative C@C
bond cleavage results initially in benzaldehyde that subsequently
undergoes further oxidation to benzoic acid. The second pathway
is epoxidation forming initially styrene oxide that undergoes rear-
rangement to the terminal (phenylacetaldehyde) or internal (ace-
tophenone) carbonylic isomer.
In the case of trans-stilbene analogous two competitive path-
ways should take place leading to benzaldehyde (C@C bond cleav-
age) in lower selectivity (25% at final reaction time) and trans-
stilbene oxide (C@C epoxidation) that was the major product at
final reaction time with 75% of selectivity (Fig. S31). Interestingly
cis-stilbene was inert under these conditions and no oxidation
products were observed in the presence of UiO-66(Zr5.4 Ti0.6)-
NO2 under identical conditions. Again, this lack of reactivity of
the cis isomer compared to the reactivity of trans-stilbene is a typ-
ical case of shape-selectivity catalysis [74]. It derives from the fact
that the reaction in the case of conjugated alkenes takes place pre-
dominantly inside the pores where diffusion of cis isomer is
impeded respect to the trans-stilbene due to the lower kinetic
4. Conclusions
The present study has established that while UiO-66(Zr) is
devoid of catalytic activity for oxygen activation, ion exchange of
Zr4+ by Ti4+ introduces catalytic activity for aerobic oxidation of
alkenes. The activity of Ti4+ containing UiO-66 can be further
enhanced by introducing electron withdrawing groups in the
ligand. In that way, the initial reaction rate for the conversion of
cyclooctene to the corresponding epoxide can be optimized for
UiO-66(Zr5.4 Ti0.6)-NO2 compared to the inactive parent UiO-66
(Zr)-H, reaching TON values that are higher than those of the most
active homogeneous catalysts. The most active sample of the series
is also able to promote epoxidation of other cycloalkenes as well as
some linear and aromatic alkenes, following the pattern expected
according to the structure-reactivity of the substrate. Overall, since
MOFs are constituted by nodes and linkers, the present study illus-
trate the strategy to create active sites by ion exchange at both
components should be simultaneously tuned to boost and opti-

462
A. Santiago-Portillo et al. / Journal of Catalysis 365 (2018) 450–463
[17] J. Gascon, A. Corma, F. Kapteijn, X.F. Llabrés
i
Xamena, Metal organic
mize the catalytic activity of each MOF for a particular chemical
reaction. It is proposed that this strategy of combining ligand sub-
stitution with post-synthetic modification of metal nodes should
be general and it can be useful for many other catalytic reactions
involving Lewis acid sites as well as redox mechanism.
framework catalysis: Quo vadis?, ACS Catal 4 (2014) 361–378.
[18] M. Yoon, R. Srirambalaji, K. Kim, Homochiral metal organic frameworks for
asymmetric heterogeneous catalysis, Chem. Rev. 112 (2012) 1196–1231.
[19] Y.-B. Huang, J. Liang, X.-S. Wang, R. Cao, Multifunctional metal-organic
framework catalysts: synergistic catalysis and tandem reactions, Chem. Soc.
Rev. 46 (2017) 126–157.
[20] Y. Li, H. Wang, L. Xie, Y. Liang, G. Hong, H. Dai, MoS2 nanoparticles grown on
graphene: an advanced catalyst for the hydrogen evolution reaction, J. Am.
Chem. Soc. 133 (2011) 7296–7299.
5. Associated content
[21] S.M.J. Rogge, A. Bavykina, J. Hajek, H. Garcia, A.I. Olivos-Suarez, A. Sepúlveda-
Escribano, A. Vimont, G. Clet, P. Bazin, F. Kapteijn, M. Daturi, E.V. Ramos-
Fernandez, F.X. Llabrés i Xamena, V. Van Speybroeck, J. Gascon, Metal-organic
and covalent organic frameworks as single-site catalysts, Chem. Soc. Rev. 46
(2017) 3134–3184.
The following files are available free of charge. Characterization
data by PXRD, FT-IR, TGA, XPS and SEM. Catalytic activity of UiO-66
based materials.
[22] J. Tu, X. Zeng, F. Xu, X. Wu, Y. Tian, X. Hou, L. Long, Microwave-induced fast
incorporation of titanium into UiO-66 metal–organic frameworks for
enhanced photocatalytic properties, Chem. Commun. 53 (2017) 3361–3364.
[23] L. Zhu, X.-Q. Liu, H.-L. Jiang, L.-B. Sun, Metal-organic frameworks for
heterogeneous basic catalysis, Chem. Rev. 117 (2017) 8129–8176.
6. Author contributions
The manuscript was written through contributions of all
authors. All authors have given approval to the final version of
the manuscript.
ˇ
ˇ
[24] G. Gómez-Pozuelo, C.P. Cabello, M. Opanasenko, M. Horácek, J. Cejka, Superior
activity of isomorphously substituted MOFs with MIL-100(M=Al, Cr, Fe, In, Sc,
V) structure in the prins reaction: impact of metal type, ChemPlusChem (2017)
82.
[25] L. Mitchell, P. Williamson, B. Ehrlichov, A.E. Anderson, V.R. Seymour, S.E.
Ashbrook, N. Acerbi, L.M. Daniels, R.I. Walton, M.L. Clarke, P.A. Wright, Mixed-
Metal MIL-100(Sc, M) (M=Al, Cr, Fe) for Lewis acid catalysis and tandem C-C
Bond formation and alcohol oxidation, Chem. Eur. J. 20 (2014) 17185–17197.
[26] Q. Sun, M. Liu, K. Li, Y. Han, Y. Zuo, F. Chai, C. Song, G. Zhang, X. Guo, Synthesis
of Fe/M (M = Mn Co, Ni) bimetallic metal organic frameworks and their
catalytic activity for phenol degradation under mild conditions, Inorg. Chem.
Front. 4 (2017) 144–153.
7. Notes
The authors declare no competing financial interest.
Acknowledgments
[27] R. Zou, P.-Z. Li, Y.-F. Zeng, J. Liu, R. Zhao, H. Duan, Z. Luo, J.-G. Wang, R. Zou, Y.
Zhao, Bimetallic metal-organic frameworks: Probing the Lewis acid site for
CO2 conversion, Small 12 (2016) 2334–2343.
[28] A. Santiago-Portillo, J.F. Blandez, S. Navalón, M. Álvaro, H. García, Influence of
the organic linker substituent on the catalytic activity of MIL-101(Cr) for the
oxidative coupling of benzylamines to imines, Catal. Sci. Technol. 7 (2017)
1351–1362.
[29] A. Santiago-Portillo, S. Navalón, P. Concepción, M. Álvaro, H. García, Influence
of terephthalic acid substituents on the catalytic activity of MIL-101(Cr) in
three lewis acid catalyzed reactions, ChemCatChem 9 (2017) 2506–2511.
[30] F. Vermoortele, M. Vandichel, B.V. de Voorde, R. Ameloot, M. Waroquier, V. Van
Speybroeck, D.E. De Vos, Electronic effects of linker substitution on Lewis acid
catalysis with metal–organic frameworks, Angew. Chem. Int. Ed. 51 (2012)
4887–4890.
Financial support by the Spanish Ministry of Economy and
Competitiveness (Severo Ochoa and CTQ2014-53292-R and
CTQ2015-69563-CO2-14) is gratefully acknowledged. Generalidad
Valenciana is also thanked for funding (Prometeo 2017/018). SN
thanks financial support by the Fundación Ramón Areces (XVIII
Concurso Nacional para la Adjudicación de Ayudas a la Investi-
gación en Ciencias de la Vida y de la Materia, 2016).
References
[31] A. Herbst, A. Khutia, C. Janiak, Brønsted instead of Lewis acidity in
functionalized MIL-101Cr MOFs for efficient heterogeneous (nano-MOF)
catalysis in the condensation reaction of aldehydes with alcohols, Inorg.
Chem. 53 (2014) 7319–7333.
[32] J.F. Blandez, A. Santiago-Portillo, S. Navalon, M. Gimenez-Marques, M. Alvaro,
P. Horcajada, H. Garcia, Influence of functionalization of terephthalate linker
on the catalytic activity of UiO-66 for epoxide ring opening, J. Mol. Catal. A
Chem. 425 (2016) 332–339.
[33] V.N. Panchenko, M.M. Matrosova, J. Jeon, J.W. Jun, M.N. Timofeeva, S.H. Jhung,
Catalytic behavior of metal-organic frameworks in the Knoevenagel
condensation reaction, J. Catal. 316 (2014) 251–259.
[34] S.M. Cohen, Modifying MOFs: new chemistry, new materials, Chem. Sci. 1
(2010) 32–36.
[1] Y. Ishii, S. Sakaguchi, T. Iwahama, Innovation of hydrocarbon oxidation with
molecular oxygen and related reactions, Adv. Synth. Catal. 343 (2001) 393–
427.
[2] R.A.F. Tomás, J.C.M. Bordado, J.F.P. Gomes, P-xylene oxidation to terephthalic
acid:
a literature review oriented toward process optimization and
development, Chem. Rev. 113 (2013) 7421–7469.
[3] T. Punniyamurthy, S. Velusamy, J. Iqbal, Recent advances in transition metal
catalyzed oxidation of organic substrates with molecular oxygen, Chem. Rev.
105 (2005) 2329–2363.
[4] S.E. Allen, R.R. Walvoord, R. Padilla-Salinas, M.C. Kozlowski, Aerobic copper-
catalyzed organic reactions, Chem. Rev. 113 (2013) 6234–6458.
[5] A. Dhakshinamoorthy, M. Alvaro, H. Garcia, Metal–organic frameworks as
heterogeneous catalysts for oxidation reactions, Catal. Sci. Technol. 1 (2011)
856–867.
[6] C.N. Cornell, M.S. Sigman, Recent progress in Wacker oxidations: moving
toward molecular oxygen as the sole oxidant, Inorg. Chem. 46 (2007) 1903–
1909.
[7] A. Dhakshinamoorthy, A. Santiago-Portillo, P. Concepcion, J.R. Herance, S.
Navalon, M. Alvaro, H. Garcia, Room temperature silylation of alcohols
catalyzed by metal organic frameworks, Catal. Sci. Technol. 7 (2017) 2445–
2449.
[8] N. Gunasekaran, Aerobic oxidation catalysis with air or molecular oxygen and
ionic liquids, Adv. Synth. Catal. 357 (2015) 1990–2010.
[9] A.E. Wendlandt, A.M. Suess, S.S. Stahl, Copper-catalyzed aerobic oxidative C-H
functionalizations: trends and mechanistic insights, Angew. Chem. Int. Ed 50
(2011) 11062–11087.
[10] T. Devic, C. Serre, High valence 3p and transition metal based MOFs, Chem. Soc.
Rev. 43 (2014) 6097–6115.
[11] G. Férey, Hybrid porous solids: past, present, future, Chem. Soc. Rev. 37 (2008)
191–214.
[35] S.M. Cohen, Postsynthetic methods for the functionalization of metal-organic
frameworks, Chem. Rev. 112 (2012) 970–1000.
[36] K.K. Tanabe, S.M. Cohen, Postsynthetic modification of metal–organic
frameworks—a progress report, Chem. Soc. Rev. 40 (2011) 498–519.
[37] D. Sun, W. Liu, M. Qiu, Y. Zhang, Z. Li, Introduction of a mediator for enhancing
photocatalytic performance via post-synthetic metal exchange in metal–
organic frameworks (MOFs), Chem. Commun. 51 (2015) 2056–2059.
[38] Y. Lee, S. Kim, J.K. Kang, S.M. Cohen, Photocatalytic CO2 reduction by a mixed
metal (Zr/Ti), mixed ligand metal–organic framework under visible light
irradiation, Chem. Commun. 51 (2015) 5735–5738.
[39] A. Corma, H. García, Lewis acids as catalysts in oxidation reactions: from
homogeneous to heterogeneous systems, Chem. Rev. 102 (2002) 3837–3892.
[40] T. De Baerdemaeker, B. Steenackers, D. De Vos, Ti-substituted zeolite Beta: a
milestone in the design of large pore oxidation catalysts, Chem. Commun. 49
(2013) 7474–7476.
[41] D.E. De Vos, B.F. Sels, P.A. Jacobs, Practical heterogeneous catalysts for epoxide
production, Adv. Synth. Catal. 345 (2003) 457–473.
[42] P. Ratnasamy, D. Srinivas, H. Knözinger, Active sites and reactive intermediates
in titanium silicate molecular sieves, Adv. Catal. 48 (2004) 1–169.
[43] M.G. Clerici, G. Bellussi, U. Romano, Synthesis of propylene oxide from
propylene and hydrogen peroxide catalyzed by titanium silicalite, J. Catal. 129
(1991) 159–167.
[44] G.J. Hutchings, D.F. Lee, A.R. Minihan, Epoxidation of allyl alcohol to glycidol
using titanium silicalite TS-1: effect of the reaction conditions and catalyst
acidity, Catal. Lett. 39 (1996) 83–90.
[12] S. Kitagawa, R. Kitaura, S.-I. Noro, Angew. Chem., Int. Ed. 43 (2004) 2334–2337.
[13] H. Furukawa, K.E. Cordova, M. O’Keeffe, O.M. Yaghi, The chemistry and
applications of metal-organic frameworks, Science 341 (2013) 1230444.
[14] A.H. Chughtai, N. Ahmad, H.A. Younus, A. Laypkov, F. Verpoort, Metal-organic
frameworks: versatile heterogeneous catalysts for efficient catalytic organic
transformations, Chem. Soc. Rev. 44 (2015) 6804–6849.
[15] A. Corma, H. Garcia, Llabrés
i Xamena, F. X., Engineering metal organic
frameworks for heterogeneous catalysis, Chem. Rev. 110 (2010) 4606–4655.
[16] D. Farrusseng, S. Aguado, C. Pinel, Metal-organic frameworks: opportunities
for catalysis, Angew. Chem. Int. Ed. 48 (2009) 7502–7513.
[45] Z.Y. Diao, L.L. Han, Z.X. Wang, C.C. Dong, The adsorption and dissociation of O2
on Cu low-index surfaces, Phys. Chem. B. 109 (2005) 5739–5745.

A. Santiago-Portillo et al. / Journal of Catalysis 365 (2018) 450–463
463
[46] Z. Saedi, S. Tangestaninejad, M. Moghadam, V. Mirkhani, I. Mohammadpoor-
Baltork, MIL-101 metal–organic framework: a highly efficient heterogeneous
catalyst for oxidative cleavage of alkenes with H2O2, Catal. Comm. 17 (2012)
18–22.
[59] C.K. Brozek, M. Dinca, Cation exchange at the secondary building units of
metal–organic frameworks, Chem. Soc. Rev. 43 (2014) 5456–5467.
[60] J. Hajek, C. Caratelli, R. Demuynck, K. De Wispelaere, L. Vanduyfhuys, M.
Waroquier, V. Van Speybreck, On the intrinsic dynamic nature of the rigid UiO-
66 metal–organic framework, Chem. Sci. 9 (2018) 2723–2732.
[47] D. Tantanak, M.A. Vincent, I.H. Hillier, Elucidation of the mechanism of alkene
epoxidation by hydrogen peroxide catalysed by titanosilicates:
computational study, Chem. Commun. (1998) 1031–1032.
a
[61] S.J.D. Smith, B.P. Ladewig, A.J. Hill, C.H. Lau, M.R. Hill, Post-synthetic Ti
exchanged UiO-66 metal-organic frameworks that deliver exceptional gas
permeability in mixed matrix membranes, Sci. Rep. 5 (2015) 7823.
[62] Z. Fang, B. Bueken, D.E. De Vos, R. Fischer, Defect-engineered metal–organic
frameworks, Angew. Chem. Int. Ed 54 (2015) 7234–7254.
[63] G.C. Shearer, S. Chavan, S. Bordiga, S. Svelle, U. Olsbye, K.P. Lillerud, Defect
engineering: Tuning the porosity and composition of the metal–organic
framework UiO-66 via modulated synthesis, Chem. Mater. 28 (2016) 3749–
3761.
[64] T. Iwahama, Y. Yoshino, T. Keitoku, S. Sakaguchi, Y. Ishii, Efficient oxidation of
alcohols to carbonyl compounds with molecular oxygen catalyzed by N-
hydroxyphthalimide combined with a Co species, J. Org. Chem. 65 (2000)
6502–6507.
[65] Y.Y. Nishiyama, N. Nakagawa, Mizuno High turnover numbers for the catalytic
selective epoxidation of alkenes with 1 atm of molecular oxygen, Angew.
Chem. Int. Ed. 40 (2001) 3639–3641.
[48] J.G. Santaclara, A.I. Olivos-Suarez, A. Gonzalez-Nelson, D. Osadchii, M.A.
Nasalevich, M.A. van der Veen, F. Kapteijn, A.M. Sheveleva, S.L. Veber, M.V.
Fedin, A.T. Murray, C.H. Hendon, A. Walsh, J. Gascon, Revisiting the
incorporation of Ti(IV) in UiO-type metalÀorganic frameworks: metal
exchange versus grafting and their implications on photocatalysis, Chem.
Mater. 29 (2017) 8963–8967.
[49] J.H. Cavka, S. Jakobsen, U. Olsbye, N. Guillou, C. Lamberti, S. Bordiga, K.P.
Lillerud, A new zirconium inorganic building brick forming metal organic
frameworks with exceptional stability, J. Am. Chem. Soc. 130 (2008) 13850–
13851.
[50] M. Kandiah, M.H. Nilsen, S. Usseglio, S. Jakobsen, U. Olsbye, M. Tilset, C. Larabi,
E.A. Quadrelli, F. Bonino, K.P. Lillerud, Synthesis and stability of tagged UiO-66
Zr-MOFs, Chem. Mater. 22 (2010) 6632–6640.
[51] A. Santiago Portillo, H.G. Baldoví, M.T. García Fernandez, S. Navalón, P.
Atienzar, B. Ferrer, M. Alvaro, H. Garcia, Z. Li, Ti as mediator in the
photoinduced electron transfer of mixed-metal NH2-UiO-66(Zr/Ti):
Transient absorption spectroscopy study and application in photovoltaic cell,
J. Phys. Chem. C 121 (2017) 7015–7024.
[66] S. Yamazaki, An improved methyltrioxorhenium-catalyzed epoxidation of
alkenes with hydrogen peroxide, Org. Biomol. Chem. 5 (2007) 2109–2113.
[67] N. Morlanes, J.M. Notestein, Kinetic study of cyclooctene epoxidation with
aqueous hydrogen peroxide over silica-supported calixarene–Ta (V), Appl.
Catal. A.-Gen. 387 (1–2) (2010) 45–54.
ˇ
[52] D. Sempere, S. Navalon, M. Dancíková, M. Alvaro, H. Garcia, Influence of
pretreatments on commercial diamond nanoparticles on the photocatalytic
activity of supported gold nanoparticles under natural Sunlight irradiation,
Appl. Catal. B-Environ. 142–143 (2013) 259–267.
[68] M.J. Burkitt, R.P. Mason, Direct evidence for in vivo hydroxyl-radical
generation in experimental iron overload: an ESR spin-trapping
investigation, Proc. Natl. Acad. Sci. USA 88 (1991) 8440–8444.
[53] J.F. Blandez, S. Navalon, M. Alvaro, H. Garcia, N-Hydroxyphthalimide anchored
on diamond nanoparticles as a selective heterogeneous metal-free oxidation
catalyst of benzylic Hydrocarbons and cyclic Alkenes by Molecular O2,
ChemCatChem 10 (2018) 198–205.
[69] E.G. Janzen, R.D. Hinton, Y. Kotake, Substituent effect on the stability of the
hydroxyl radical adduct of
Lett. 33 (1992) 1257–1260.
a-phenyl N-tert-butyl nitrone (PBN), Tetrahedron
[70] E.G. Janzen, Y. Kotake, D.H. Randall, Stabilities of hydroxyl radical spin adducts
of PBN-type spin traps, Free Radical Bio. Med. 12 (1992) 169–173.
[71] G.R. Buettner, Spin trapping: ESR parameters of spin adducts, Free Radic. Biol.
Med. 3 (1987) 259–303.
[54] P. Cancino, A. Vega, A. Santiago-Portillo, S. Navalon, M. Alvaro, P. Aguirre, E.
Spodine, H. García,
A novel copperIJII)–lanthanumIJIII) metal organic
framework as selective catalyst for the aerobic oxidation of benzylic
a
hydrocarbons and cycloalkenes, Catal. Sci. Technol. 6 (2016) 3727–3736.
[55] J.C. Espinosa, S. Navalón, M. Álvaro, H. García, reduced graphene oxide as a
metal-free catalyst for the light-assisted fenton-like reaction, ChemCatChem 8
(2016) 2642–2648.
[56] A. Santiago-Portillo, S. Navalon, F. Cirujano, F. Llabrés i Xamena., M. Alvaro, H.
Garcia, MIL-101 as reusable solid catalyst for autooxidation of benzylic
hydrocarbons in the absence of additional oxidizing reagent, ACS Catal 5
(2015) 3216–3224.
[72] K.A. Joergensen, Transition-metal-catalyzed epoxidations, Chem. Rev. 89
(1989) 431–458.
[73] A. Dhakshinamoorthy, M. Alvaro, H. Garcia, Aerobic oxidation of cycloalkenes
catalyzed by iron metal organic framework containing N-hydroxyphthalimide,
J. Catal. 289 (2012) 259–265.
[74] S. Hu, D. Liu, C. Wang, Y. Chen, Z. Guo, A. Borgna, Y. Yang, Liquid-phase
epoxidation of trans-stilbene and cis-cyclooctene over vanadium-exchanged
faujasite zeolite catalysts, Appl. Catal. A.-Gen. 386 (2010) 74–82.
[75] M.V. Baldovi, A. Corma, H. Garcia, V. Marti, Shape-selective photosensitized
isomerization of stilbene using a benzophenone incorporated within acid
zeolites, Tetrahedron Lett. 35 (1994) 9447–9450.
[57] A. Gómez-Paricio, A. Santiago-Portillo, S. Navalón, P. Concepción, M. Alvaro, H.
Garcia, MIL-101 promotes the efficient aerobic oxidative desulfurization of
dibenzothiophenes, Green Chem. 508–515 (2016).
[58] R.M. Sellers, Spectrophotometric determination of hydrogen peroxide using
potassium titanium(IV) oxalate, Analyst 105 (1980) 950–954.