N-Hydroxyphthalimide Anchored on Diamond Nanoparticles as a Selective Heterogeneous Metal-free Oxidation Catalyst of Benzylic Hydrocarbons and Cyclic Alkenes by Molecular O2

Article abstract of DOI:10.1002/cctc.201700886

N-hydroxyphthalimide (NHPI) derived from trimellitic anhydride has been covalently anchored (4 wt %) through ester bonds to the surface of commercial diamond nanoparticles previously functionalized by chemical and thermal treatments (DH). IR spectroscopy of the NHPI/DH solid shows the presence of a characteristic peak at 1735 cm^?1 attributable to the ester bond, while solid-state MAS ¹³C NMR spectroscopy shows two broad bands corresponding to carbon atoms of carbonyl groups (167 ppm) and aromatic carbons (127 ppm). NHPI/DH promotes autoxidation of isobutylbenzene involving superoxide (O₂ ^.?) and hydroperoxide (HOO?) radicals; NHPI/DH can be reused at least three times and reach a minimum turnover number as high as 20 600, which is probably due to the inertness of the diamond surface, allowing free diffusion of reactive oxygen species. Analogous materials in which NHPI was anchored on activated carbon and multiwall carbon nanotubes were inefficient as autoxidation catalysts. NHPI/DH is able to promote also selective aerobic oxidation of other benzylic hydrocarbons and cycloalkenes.

Full text of DOI:10.1002/cctc.201700886

Accepted Article
Title: N-hydroxyphthalimide anchored on diamond nanoparticles as
selective heterogeneous metal-free oxidation catalyst of benzylic
hydrocarbons and cyclic alkenes by molecular O2
Authors: Juan F. Blandez, Sergio Navalon, Mercedes Alvaro, and
Hermenegildo García
This manuscript has been accepted after peer review and appears as an
Accepted Article online prior to editing, proofing, and formal publication
of the final Version of Record (VoR). This work is currently citable by
using the Digital Object Identifier (DOI) given below. The VoR will be
published online in Early View as soon as possible and may be different
to this Accepted Article as a result of editing. Readers should obtain
the VoR from the journal website shown below when it is published
to ensure accuracy of information. The authors are responsible for the
content of this Accepted Article.
To be cited as: ChemCatChem 10.1002/cctc.201700886
Link to VoR: http://dx.doi.org/10.1002/cctc.201700886
A Journal of
www.chemcatchem.org

ChemCatChem
10.1002/cctc.201700886
FULL PAPER
N-hydroxyphthalimide anchored on diamond nanoparticles as
selective heterogeneous metal-free oxidation catalyst of benzylic
hydrocarbons and cyclic alkenes by molecular O
2
Juan F. Blandez, [a] Sergio Navalón, [a] Mercedes Álvaro, [a] and Hermenegildo García*[a,b]
Abstract: N-hydroxyphthalimide (NHPI) derived from trimellitic
anhydride has been covalently anchored (4 weight %) through ester
bonds to the surface of commercial diamond nanoparticles previously
functionalized by chemical and thermal treatments (DH). IR
spectroscopy of the NHPI/DH solid shows the presence of a
mechanism is known to involve phthalimide N-oxyl (PINO) radical
generated from NHPI, the process being easier in the presence
of first-row transition metals, particularly Co2+ ions that coordinate,
with the hydroxyl group of the diacyloxime and subsequently
generate the PINO radicals with the simultaneous change in the
oxidation state of the metal.[1, 5, 6] In spite of the high efficiency and
applicability together with affordability of NHPI, its implementation
at industrial scale is hampered by the difficulty to recover NHPI
from the reaction mixtures. However, the number of studies
reporting heterogeneous catalyst in with NHPI or its derivatives
has been immobilized on solid supports is relatively scarce.[5, 8]
A few years ago, we reported the catalytic activity of NHPI
adsorbed on porous metal organic frameworks (MOFs), in where
there is a synergism between the adsorbed NHPI and the metal
nodes of the MOF.[9] In other reports, NHPI derivatives have been
anchored to polymers, based on polystyrene[10] or acrylates,[11]
and used to promote oxidations of benzylic hydrocarbons in
combination or in the absence of Co2+ ions. NHPI derivatives have
also been anchored on mesoporous silica again using Co2+ salts
as promoter.[12]
-1
characteristic peak at 1735 cm attributable to the ester bond, while
13
solid-state MAS C NMR spectroscopy shows two broad bands
corresponding to carbon atoms of carbonyl groups (167 ppm) and
aromatic carbons (127 ppm). NHPI/DH promotes autoxidation of
·
-
isobutylbenzene involving superoxide (O
2
)
and hydroperoxide
·
(
HOO ) radicals NHPI/DH can be reused at least three times and
reach a minimum turnover number as high as 20,600, probably due
to the unique inertness of the diamond surface, allowing free diffusion
of reactive oxygen species. Analogous materials in where NHPI was
anchored on activated carbon and multiwall carbon nanotubes were
inefficient as autoxidation catalysts. NHPI/DH is able to promote also
selective aerobic oxidation of other benzylic hydrocarbons and
cycloalkenes.
In spite of the related precedents, it would be still convenient to
explore other heterogeneous catalyst in where NHPI is anchored
to other supports of different properties. Herein we have used
surface-modified diamond nanoparticles (DH) as support to
anchor NHPI. The main advantage of DH as support is the
inertness of its surface against radicals. In early precedents it was
Introduction
Commonly, hydrocarbon oxidation is carried out in the presence
of inorganic oxidizing agents such as chromium and manganese
oxides, nitric and sulfuric acid among others, resulting in
generation of toxic wastes of difficult environmental
found that even highly aggressive hydroxyl radicals do not react
with the surface of modified DH.[13] The data presented here
management.[
friendly oxidations involves the use of organic and/or inorganic
peroxides or hydroperoxides,[3] the use of molecular O
to
1, 2]
Nowadays, development of environmentally-
2+
shows that NHPI/DH in the absence of Co or any other transition
2
metals as co-catalyst is able to promote the aerobic oxidation of
benzylic hydrocarbons and cyclic alkenes behaving as a solid
radical initiator.
promote selective oxidations still being a challenge.[4] In this
context, N-hydroxylphthalimide (NHPI) and its derivatives have
shown a wide scope in promoting the aerobic oxidations of
organic functional groups including benzylic and alicyclic
hydrocarbons, alcohols, oximes, alkenes, etc.[1, 5-7] The reaction
Results and Discussion
In the first part of this section the preparation and characterization
of NHPI/DH will be described. Then, the performance of NHPI/DH
as heterogeneous initiator for autoxidation in benzylic carbons
and cyclic alkenes will be presented.
[a]
Dr. J. F. Blandez, Dr. S. Navalón, Prof. M. Álvaro Prof. H. García
Departamento de Química and Instituto de Tecnología Química
CSIC-UPV, Universitat Politécnica de Valencia, Consejo Superior
de Investigaciones Científicas, Av. de los Naranjos s/n, 46022
Valencia, Spain.
E-mail: hgarcia@qim.upv.es
Center of Excellence for Advanced Materials Research, King
Abdulaziz University, Jeddah, Saudi Arabia
Synthesis and characterization of NHPI/DH
[b]
The synthesis of NHPI/DH started with commercially available
trimellitic anhydride that was reacted with hydroxylamine to form
the 4-substituted carboxylic acid derivative of NHPI. This
Supporting information for this article is given via a link at the end of
the document.
1
13
derivative was characterized by H- and C-NMR spectroscopy
and mass spectrometry (Figures S1-S3). This NHPI acid
This article is protected by copyright. All rights reserved.

ChemCatChem
10.1002/cctc.201700886
FULL PAPER
derivative was reacted with D NPs that were previously purified,
then, surface-modified by Fenton oxidation and subsequently
surface (Figure 1a and S4). Specifically, the weight loss from 250
to 400 ºC is attributable to the decomposition of NHPI. These
measurements give an estimation of NHPI content in the series of
carbon supports of about 4 wt% for NHPI/DH (Figure 1a),
NHPI/ACH (Figure S4a) and NHPI/MWCNTsH (Figure S4b). The
presence of NHPI on the support can be ascertained by FT-IR
spectroscopy, comparing the spectra of NHPI and DH with that of
NHPI/DH (Figure 1b). The most characteristic observations were
the disappearance of OH band in the DH support (about 3300-
3000 cm ) and the appearance of C=O stretching vibration band
about 1735 cm-1 attributable to the formation of the ester groups.
Figure 1b shows a comparison of three FT-IR spectra for NHPI,
DH and NHPI/DH.
hydrogenated at 500 ºC under H
is illustrated in Scheme 1. It is known that commercial D NPs can
suffer an oxidative functionalization by Fenton reaction with H O
2
2
flow (DH). The synthetic route
2
using Fe2+ salt.[13, 14] This Fenton treatment gives rise to water
dispersible D NPs in where the surface of these D NPs contains
a significant population of hydroxylic and carboxylic groups.
2
Treatment of the purified D NPs with H at 500 ºC resulted in the
-1
chemical reduction of surface carboxylic and carbonyl groups that
are converted into hydroxyl group.[15] These surface -OH groups
will form an ester bond with the carboxyl acid substituent of NHPI
forming a covalent linkage NHPI and DH.
O
O
O
O
HCl
,
HO
HO
+
NH2
O
H
O
N
OH
,
º
~ 4 wt %
pyridine 15 h 90
C
a)
b)
O
O
O
O
O
-
p toluensulfonic acid
O
HO
+
OH
N
O
OH
-
N
OH
p xylene reflux overnight
,
,
DH
O
O
NHPI/DH
Scheme 1. Synthetic route for the preparation of the NHPI/DH catalyst.
Figure 1. (a) TGA (black line) and DTGA (red line) for NHPI/DH; (b) FT-IR
spectra for NHPI (1), DH (2) and NHPI/DH (3).
Similar chemical oxidation reactions but using sulfuric or nitric
acid as oxidants followed by reductive hydrogenation and further
NHPI derivative anchoring were also applied to commercial
activated carbon (AC) and multi-walled carbon nanotubes
Comparison of XPS C1s peaks of DH and NHPI/DH shows an
increase in the percentage of carbons bonded to oxygen from 33
to 60 % after anchoring the NHPI on DH. The broadness of N 1s
peak slightly increases from DH to NHPI/DH due to the presence
of imide group in the NHPI that overlaps with that of graphitic N.
This XPS data are in general agreement with the amount of NHPI
anchored on DH surface (Figure 2).
(
MWCNT) samples, leading to the formation of NHPI/ACH and
NHPI/MWCNTsH that would be reference materials to compare
the performance of NHPI/DH.
The materials containing covalently anchored NHPI were
characterized by different techniques including elemental
combustion analysis, thermogravimetry, FT-IR, XPS and solid-
state 13C NMR spectroscopies. Table 1 shows analytical data of
the three supports (DH, ACH and MWCNTsH) compared to the
elemental analysis of the corresponding materials where NHPI is
anchored. As it can be seen in the Table 1, derivatization of DH,
ACH and MWCNTH result in a decrease in the carbon content,
while the amount of nitrogen undergoes minor relative variations.
These data indicate that the percentage of NHPI anchored to the
material should be low. It is also noted that the process introduces
elemental sulphur indicating that some adsorbed p-toluensulfonic
acid must be present in our material.
a)
b)
GraphiticN
Pyridinic
N-oxide
c)
d)
-CO-NH-
Table 1. Elemental analysis data for the solid materials before and after
NHPI anchoring.
Solid material
DH
NHPI/DH
ACH
NHPI/ACH
MWCNTsH
NHPI/MWCNTsH
N (%)
2.07
1.95
1.05
1.48
0.06
0.28
C (%)
88.61
80.32
88.54
69.68
98.53
82.38
H (%)
0.98
1.04
0.82
1.47
0.20
0.94
S (%)
0
0.30
0
0.17
0
0.20
O (%)
8.34
16.39
9.59
27.20
1.21
16.20
Figure 2. XPS spectra for C 1s and N 1s for the DH (top) and NHPI/DH (bottom)
and the percentages of their individual components obtained from the best fit.
Magic-angle spinning (MAS) solid-state 13C NMR spectra of
NHPI/DH shows two broad weak peaks about 167 and 127 ppm
attributable to the carbonyl groups and aromatic carbon atoms of
NHPI. These peaks in MAS 13C NMR spectroscopy were absent
The amount of NHPI present on the support was estimated from
TG analysis (TGA) and TGA derivative (DTGA) profiles of the
carbon supports covalently functionalized with NHPI on the
This article is protected by copyright. All rights reserved.

ChemCatChem
10.1002/cctc.201700886
FULL PAPER
in the case of DH as support (Figure 3), where only two signal at
3
5 and 50 ppm corresponding to carbon atoms of DH bonded or
not to oxygen, respectively, can be observed.[15]
b)
O
O
O
N
OH
O
Carbonyl
group
Aromatic carbon
210 190 170 150 130 110 90 70 50 30 10 0
(ppm)
Figure 4. Time profile of IB conversion in the presence of NHPI/DH (●),
a)
NHPI/ACH (□), NHPI/MWCNTsH (∆) and DH (■) as catalyst. Reaction
conditions: Catalyst (40 mg), IB (20 mmol), O (1 bar), 140 ºC.
2
OH
Comparison of the performance of NHPI/DH solid, with those of
DH, commercial NHPI molecule, and unsupported NHPI
carboxylic acid shows that while DH was unable to promote IB
autoxidation, the two NHPI molecules promote IB autoxidation,
but at much lower reaction rate than for NHPI/DH, even though
the reactions using commercial NHPI and unsupported NHPI
carboxylic acid were carried out using double amount in
comparison with NHPI/DH. Therefore, NHPI/DH was the best
performing catalyst to carry out IB autoxidation of IB of all the
series (Figure 5). This higher catalytic activity of NHPI anchored
to DH is attributed to the combination of a series of factors, easier
generation of the corresponding PINO radicals, their longer
lifetime when they are anchored to DH and the free diffusion of
oxygen-centred radicals to the liquid phase without interaction
with the solid surface.
2
10 190 170 150 130 110 90 70 50 30 10 0
(ppm)
Figure 3. MAS 13C-NMR spectra for DH (a) and NHPI/DH (b).
Catalytic data
Comparison of the performance of NHPI anchored to the three
types of carbon supports shows that only for NHPI/DH products
are formed in significant concentration (Figure 4). In the case of
NHPI/ACH
and
NHPI/MWCNTsH
disappearance
of
isobutylbenzene (IB) was not accompanied by the formation of
significant amount of oxidation products (Figures S5 and S6).
Moreover, in the cases of NHPI/ACH (Figure S5) and
NHPI/MWCNTH (Figure S6), comparison of the percentage of IB
disappearance under oxygen or argon atmosphere shows that
most of the IB is really absorbed on the material while oxidation is
taking place in only a very minor extent as revealed by the
presence of reaction products (Figures S5 and S6). This high
adsorption of IB on NHPI/ACH or NHPI/MWCNTsH it is
agreement with the high adsorption capacity characteristic of
these carbonaceous materials.[16] It is proposed that the higher
efficiency of NHPI/DH respect to the other NHPI/ACH and
NHPI/MWCNTH materials is a reflection of the nature of the
surface. The inertness of the diamond surface after pre-
treatment with Fenton and hydrogenation is favorable for the
generation of reactive oxygen species as free radicals in the
solution, thus, increasing the number of initiation events. In
contrast, the surfaces of ACH and MWCNTH with sp2 carbons
and multiple bonds have functional groups that can interact
with these reactive oxygen species and, therefore, the
availability of these intermediates to promote hydrocarbon
oxidation is lower.
a)
b)
Figure 5. a) Time profile of IB conversion in the presence of commercial NHPI
○), carboxylic acid derivative of NHPI (∆) and NHPI/DH (●). b) Selectivity
(
evolution for 1-hydroperoxyisobutylbenzene (●), phenyl isopropyl ketone (■)
and benzoic acid (∆) during IB conversion using NHPI/DH as catalyst. Reaction
conditions: IB (20 mmol), NHPI/DH (40 mg), commercial NHPI and derived acid
2
NHPI (1.6 mg), O (1 bar), 140 ºC.
In all reactions, the products observed in IB autoxidation were the
corresponding 1-hydroperoxy IB, phenyl isopropyl ketone and
benzoic acid. The 1-hydroperoxy IB cannot be quantified by GC
due to its decomposition. Therefore, the reaction mixture was
treated with triphenylphosphine at 80 ºC to promote
This article is protected by copyright. All rights reserved.

ChemCatChem
10.1002/cctc.201700886
FULL PAPER
decomposition of the hydroperoxide and, then, the formation 1-
hydroxy-1-IB, together with 1-phenyl-1-isobutanol was quantified
together with triphenylphosphine oxide and 1-phenyl-2-methyl-1-
propanol.[ Benzoic acid is a secondary product, coming from
oxidation of phenyl isopropyl ketone. Accordingly, the temporal
profile of benzoic acid formation shows an induction period,
indicating that it is not a primary product. Figure 6b shows the time
argon, was observed as consequence of the occurrence of a
negligible oxidation reaction, in the case of IB autoxidation by
NHPI/DH the initial reaction rate follows an apparent pseudo first
order relationship with oxygen pressure and no IB disappearance
was observed under argon atmosphere (Figure 7).
17]
The activation energy for IB autoxidation promoted by
profile for IB conversion and selectivity towards the main products. NHPI/DH was determined form the Arrhenius plot of the natural
NHPI/DH was reused three times observing that during the first
two uses, the temporal profile of IB conversion remains unaltered,
while a decrease in the reaction rate was observed in the third
use, even though substantially the same conversion was
achieved at final reaction time (Figure 6a). This partial
deactivation can be attributed to the occurrence of
leaching/degradation of NHPI as can be deduced from the
changes observed in the IR spectra of the used NHPI/DH catalyst
logarithm of the initial reaction rate versus the inverse of the
absolute temperature. The initial reaction rate was obtained from
the slope of the temporal profile of IB conversion at temperatures
ranging from 120 to 150 ºC. In this way, an activation energy of
69 kJ·mol-1 was estimated for the process (Figure 8). This Ea is
similar to the values estimated for the aerobic oxidation of
benzylic positions using transition metal-based catalysts such as,
for instance, the metal-organic framework MIL-101(Cr).[
17]
(
Figure S7). The most evident change was the decreased
a)
b)
-
intensity of the characteristic band of the ester groups at 1735 cm
1.
O
2
In addition, TG profiles show that the three times used material
contains a significant amount of organic compounds adsorbed on
its surface that was estimated to be about 12 wt% (Figure S8).
Air
Argon
Figure 7. (a) Time-conversion plot under different oxygen pressure (argon, air;
molecular oxygen) and relationship of the oxygen atmosphere and the initial
reaction rate (b) for the oxidation of IB in the presence of NHPI/DH. Reaction
2
conditions: Catalyst (40 mg), IB (20 mmol), 140 ºC, O atmosphere as indicated
in the graph.
a)
b)
st
Figure 6. (a) Reusability of NHPI/DH during the oxidation of IB. Legend: 1 use
nd
rd
(
■), 2 use (●) and 3 use (○). Reaction conditions: IB (20 mmol), NHPI/DH (20
mg), (1 bar), 140 ºC. (b) Time conversion (■) and selectivity (○:
O
2
hydroperoxide IB; □ phenyl isopropyl ketone; ∆: benzoic acid) plots for the
maximum IB turnover number reaching a value of 20,600. Reaction conditions:
IB (40 mmol), NHPI/DH (10 mg), O (1 bar), 140 ºC.
2
In order to determine the maximum turnover number (TON) of
NHPI/DH as catalyst for autoxidation of IB, an additional
experiment was carried out in where a large amount of IB was
used and the reaction was allowed to proceed for longer times
Figure 8. a) Time-conversion plot for IB oxidation at different temperatures and
b) Arrhenius plot of the natural logarithm of the initial reaction rate vs the inverse
of the absolute temperature. Legend: 150 ºC (■), 140 ºC (●), 130 ºC (▲) and
120 ºC (♦). Reaction conditions: IB (20 mmol), NHPI/DH (40 mg), 140 ºC, O2
(
Figure 6b). Under these conditions, a TON value of 20,600
atmosphere (1 bar).
molecules of IB converted per NHPI molecule in the catalyst was
achieved. This value is several orders of magnitude higher than
those previously reported using NHPI/MOF[9, 18] or even using
homogeneous NHPI as radical initiator in combination with
transition metal ions,[ and is also higher or comparable to the
TON values reported for other catalysts based on transition
metals as active centers.[20] Selectivity about 90 % for benzoic
acid was achieved at full IB conversion (Figure 6b).
The influence of the oxygen content in the atmosphere on the
catalytic activity of IB oxidation using NHPI/DH as catalyst was
also studied. In contrast to the previously commented adsorption
experiment in NHPI/ACH or NHPI/MWCNTsH, in where little
influence of the composition of the atmosphere either oxygen or
The fact that the reaction mechanism involves carbon center
radicals was supported by performing a hot filtration test in where
the reaction is initiated in the presence of solid catalyst and then
the catalyst was removed after 4 h, while the reaction is still hot
at the reaction temperature. Comparison of the temporal profiles
after filtration of the catalyst with that of a twin reaction where the
catalyst was present shows that the reaction does not completely
stop in the absence solid catalyst (Figure S9). It was, however,
determined that IB conversion was smaller after filtration of the
19]
This article is protected by copyright. All rights reserved.

ChemCatChem
10.1002/cctc.201700886
FULL PAPER
catalyst than when the catalyst was present. These results can be
interpreted considering that during the initial 4 h, of reaction
radicals such as hydroperoxide are generated, and these radicals
will continue oxidation through a radical chain mechanism even in
the absence of solid catalyst. Similar results, i.e. evolution of the
reaction after catalyst filtration, have been observed for the
aerobic oxidation of benzylic substrates employing metal-based
Figure 9. a) Experimental (black line) and simulated EPR spectra (red line) of
PBN–OOH. Hyperfine coupling constants of PBN–OOH AG
N
= 13.40 and AG
H
=
1.25. Reaction conditions: NHPI/DH (10 mg), PBN (20 mg), O
2
(1 atm) in n-
dodecane (20 mL) at 140 ºC for 2 h. b) Proposed reaction mechanism for the
oxidation of benzylic substrates using supported NHPI/DH.
catalysts such as MIL-101(Cr),[
17]
Cu-MOF as initiators.
[21]
Although the contribution of some leached NHPI from the solid to
the solution cannot be ruled out, leached NHPI should promote
oxidation in a very minor extent according to the previously
commented experiment in where pure NHPI in a much higher
amount than possibly leached out was used as catalyst, observing
a slow progress of the oxidation reaction (Figure 5a). This radical
chain mechanism was also supported by quenching experiment
The scope of the aerobic oxidation of benzylic compounds using
NHPI/DH as heterogeneous autoxidation catalyst was expanded
to other industrially important substrates including cumene,
indane, 1-phenylbutane and 2-phenylbutane (Table 2). Excellent
to good selectivities to the corresponding ketones or
hydroperoxides were obtained for the studied substrates. Of
particular importance is oxidation of cumene to 1-
hydroperoxycumene with selectivity as high as 93 % at 50 %
conversion under solventless conditions (Figure S10). Cumyl
hydroperoxide is a commodity largely employed in the chemical
industry as primary oxidant and radical initiator for the production
of oxygenated compounds or as polymerization initiator. Currently,
·-
using p-benzoquinone as selective O
2
radical quencher (Figure
S9b). [17, 22, 23] When the reaction is started in the absence of p-
benzoquinone and then, 20 mol % p-benzoquinone was added at
5
h reaction time, an almost complete quenching of the reaction
was observed, supporting the intermediacy of superoxide radical
ions as a radical intermediate. The use of DMSO as selective
hydroxyl radical scavenger also stopped the reaction (Figure S9b). most of the phenol produced worldwide is obtained by the
Additional support for the intermediacy of superoxide radical
anions was obtained by Electron paramagnetic resonance (EPR)
spectroscopy using N-tert-butyl-α-phenylnitrone (PBN) as radical
trapping agent. Heating n-dodecane containing PBN and
NHPI/DH under oxygen led to the observation of an EPR spectra
chemical industry through decomposition of cumyl hydroperoxide
with H SO4 (Hock process).[25] Interestingly, using NHPI/DH
2
acetophenone can be obtained with high selectivity (95 %) at
almost full conversion as secondary product from cumyl
hydroperoxide in the same reaction. Acetophenone is an
important commodity for the production of resins and an important
intermediate for the production of fine chemicals.
·
corresponding to the adduct of PBN with hydroperoxide (HOO )
radicals (Figure 9). Hydroperoxyl is the protonated form of
superoxide (O
2
·-) that should be the primary oxygen reactive
Also indane can be oxidized with good selectivity (82 %) to 2-
indanone at high conversions (81 %) (Figure S11). Indanone is
an industrial organic intermediate for the production of oximes
employed in the production of agrochemicals and
pharmaceuticals. As commented earlier, IB is oxidized to phenyl
isopropyl ketone and benzoic acid with selectivities of 64 and
30 %, respectively. In the case of 2-phenylbutane the
corresponding ketone in benzylic position is obtained with good
selectivity (80 %), even at high conversions (~80 %) (Figure S12).
In the case of 1-phenylbutane also the major selectivity was
achieved for the ketone in benzylic position (66 %) accompanied
by benzoic acid (29 %) at 70 % conversion (Figure S13). It is
important to highlight the good conversions and selectivities
species generated in the system. The good match between the
experimental and simulated EPR spectrum for PBN-OOH adduct
indicated that no other reactive oxidizing species should be
present in comparable amounts to that of O ·
- [24]
.
The fact that
2
hydroxyl radical was not observed by EPR spectroscopy is
interpreted considering that the generated hydroperoxides are
quantitatively trapped by PBN and undergo decomposition to the
corresponding hydroxyl radical in significant extent.[17]
a)
HO
b)
O
O
O
N
O
+
N
O
·
HOO
O
∆
+
HOO
N
OH
2
N
O
O
O
NHPI
PINO
2
achieved using exclusively molecular O as oxidant, respect to the
PINO + R-H
+ O
ROO + R H
NHPI + R
ROO
ROOH + R
HO
use of other transition metal[26] or metal-free catalysts that employ
R
2
[27]
organic or inorganic peroxides as terminal oxidants. Figure 9b
-
+
-
shows the proposed reaction pathway for the oxidation of benzylic
compounds using NHPI/DH as radical initiator in the presence of
ROOH
RO + R H
RO
-
R OH + R
=
+
H
ROOH
R
O
2
O
2
molecular O .
3
455
3480
3505
Magnetic field / G
Table 2. Conversion and selectivity for the benzylic oxidation of hydrocarbons
promoted by NHPI/DH.[
a]
Initial
reaction
rate
Conver
sion
Time
(h)
Main reaction
products
Selectiv
ity (%)
(%)
-
1
(
mmol h )
Cumene
11.2
97
24
Acetophenone
95
This article is protected by copyright. All rights reserved.

ChemCatChem
10.1002/cctc.201700886
FULL PAPER
Cyclooctene[b]
Cycloheptene
Cyclohexene
Cyclopentene
33
58
0
0
0
67.3
57.2
-
100
32.7
0
Benzoic acid
Indanone
5
Indane
4.8
83
96
80
55
5
76
17
3-hydroxy-1-
25
35.5
-
indanone
IB
1.38
1.49
1.46
24
72
24
Phenyl isopropyl
ketone
64
< 3
-
Benzoic acid
2-Phenyl-2-butanol
Acetophenone
31
80
6
[
a] Reaction conditions: Catalyst (40 mg), substrate (20 mmol),
atmosphere (5 bars), 150 °C, 24 h
b] Blank control experiments in the absence of catalyst resulted in a
O
2
2-
[
Phenylbu
tane
conversion of ~6 %
1-
Phenyl propyl
ketone
66
Phenylbu
tane
Benzoic acid
29
[
1
a] Reaction conditions: Catalyst (40 mg), substrate (20 mmol), O
40 °C. Reaction time profiles are presented in figures S10 to S13.
2
atmosphere,
Conclusions
In the present manuscript it has been shown that the nature of the
carbonaceous support in with NHPI is anchored plays a significant
role determining its catalytic activity. While covalent anchoring of
NHPI through ester bonds to the surface OH group of ACH or
MWCNTsH render materials with negligible autoxidation activity,
when NHPI is anchored on the surface of treated D NPs, the
resulting material is able to promote autoxidation of benzylic
hydrocarbons and alicyclic alkenes. In general, good to excellent
selectivity were observed even at conversions higher than 60 %
under solventless conditions. In the case of IB a TON as high as
20,600 was achieved. Quenching experiments and EPR detection
of the adduct of PBN with hydroperoxyl radical indicate that the
most likely reaction mechanism involves the superoxide and
hydroperoxide species as primary reactive oxygen species. The
catalyst can be reused and the activity decay appears to be
related to poisoning of the solid surface. NHPI/DH has a broad
scope to promote autooxidation of benzylic hydrocarbons and
cycloalkenes, following the reactivity pattern already reported for
others heterogeneous NHPI catalysts.
As expected in view of the reported activity of NHPI, NHPI/DH is
also active to promote oxidation of cycloalkenes, the product
distribution depending on the size of the cycloalkene (Table 3).
Aerobic oxidation of cycloalkenes is a challenging reaction for the
production of fine chemicals such as epoxides, alcohols and
ketones. In the case of large cyclooctene, epoxide was the only
product observed with almost complete selectivity at significant
substrate conversion. This selectivity towards the epoxide
decreased as the ring size diminishes. In this way cycloheptene
undergoes aerobic oxidation to afford the allylic ketone and the
cyclic epoxide in a proportion 2:1. In the case of cyclohexene, no
epoxide was observed and the major products were the
corresponding allylic alcohol and ketone. No autoxidation was
observed in the case of cyclopentene. These results and the
product distribution for cycloalkene oxidation parallel well the data
that have been reported in the literature for the autoxidation of
these cyclic substrates by NHPI adsorbed on Fe-BTC[9] or using
a Cu-MOF.[23]
Experimental Section
Materials
Diamond nanopowder (ref: 636444, 95%), activated carbon (Norit SX Ultra,
ref. 53663), multi-walled carbon nanotubes (MWCNTs) and trimellitic
anhydride were supplied by Sigma-Aldrich. The other reagents or solvents
used in this work were of analytical or HPLC grade.
Synthesis of NHPI 4-carboxylic acid molecule
The preparation of NHPI 4-carboxylic acid was carried out according
reported procedures.[28] To a suspension of N-hydroxylamine·HCl (3.06 g,
4
4 mmol) in pyridine (50 mL), trimellitic anhydride (7.69 g, 40 mmol) was
Table 3. Oxidation of cycloalkenes. Catalytic data for the aerobic oxidation
of cycloalkenes by NHPI/DH.[
a]
added. The mixture was stirred magnetically at 90 ºC for 15 h. After this
time, the reaction was quenched by addition of water (50 mL). The mixture
was cooled at room temperature and the pH value brought to 2-3 with HCl.
A white solid precipitated. The solid was filtered using a Nylon filter (0.2
µm), washed with abundant water and dried under vacuum. The
spectroscopic data is provided in Figures S1-S3.
Cycloalkene
Conversion
%)
Selectivity (%)
Ketone
(
Alcohol
Epoxide
This article is protected by copyright. All rights reserved.

ChemCatChem
10.1002/cctc.201700886
FULL PAPER
Preparation of surface-modified D support (DH)
General catalytic procedure
[
15,
Preparation of DH was made following previously reported procedures.
Typically, the catalyst (20 mg) was placed into a round bottom flask (25
mL) having a magnetic stirring bar. The reagent (10 mmol) was added to
the flask and the suspension sonicated for 1h. Then, the mixture was
29]
Briefly, commercial diamond NPs (0.5 g) (D NPs) were added to H
150 mL) in a beaker and, the suspension was sonicated. A solution of
FeSO (2 g) in water at pH 3 (H SO ) was added to this mixture dropwise.
2 2
O
(
4
2
4
purged with O
2
for 15 min. Then, the reaction mixture was stirred at 140
Caution: sudden temperature increase can occur because the high
exothermicity of the reaction. When the reaction is finished, the suspension
was allowed to settle overnight. The supernatant was removed and the
solid exhaustively washed with a diluted sulfuric aqueous solution (0.1 M)
until the iron is absent as evidenced by colorimetric titration using KSCN.
Then, the solid was washed with Milli-Q water until pH 7. The H O was
2
removed by lyophilization. Dry samples of purified D NPs were reduced
with H2 at 500 ºC for 6 h and labelled as DH.
ºC in a preheated oil bath under oxygen atmosphere from a balloon. The
course of the reaction was periodically followed by extracting aliquots (0.1
mL) of the reaction mixture with a syringe that were diluted with acetonitrile.
Then, a known weight of nitrobenzene as calibration standard was added,
the solution was filtered and the mixture was immediately injected in GC
(6890 Network GC system Agilent technologies) to avoid spontaneous
subsequent oxidation. Product quantification was done using calibration
data obtained using commercially available products.
Preparation of surface-modified activated carbon support (ACH)
Reuse experiments for NHPI/DH sample were performed following the
general reaction conditions, but employing the used NHPI/DH sample as
catalyst. At the end of each reaction, the solid catalyst was filtered through
a Nylon membrane filter (0.2 µm) and washed with hot dichloromethane.
Then, the catalyst was dried at 100 ºC for 24 h before next use. The
catalyst was loaded in a new reaction mixture.
Commercially available activated carbon (Norit SX Ultra; 100 mg) was
sonicated in 25 mL of concentrated HNO
3
. The suspension was heated at
85 ºC for 20 h. After this time, the mixture was cooled to room temperature
and the solid recovered by centrifugation. The solid was washed with water
several times until pH 7. Finally, the solid was dried under vacuum. Dry
samples of purified AC were reduced with H
resulting sample labelled as ACH.
2
at 500 ºC for 6 h and the
Selective radical quenching experiments were carried out following the
general procedure, but with the addition of dimethylsulfoxide (DMSO) as
hydroxyl radical quencher[17,
23,
31]
and p-benzoquinone as
superoxide/hydroperoxide radical quencher[17, 22, 23] (20 mol % respect to
substrate) after the reaction started.
Preparation of surface-modified MWCNTs support (MWCNTH)
Commercial MWCNTs were submitted to purification and pretreatment by
adding a sample (150 mg) to a HNO aqueous solution (75 mL, 3 M) in a
3
Electron paramagnetic resonance (EPR) measurements
flask.[30] The reaction mixture was stirred under reflux temperature for 12
h. Then, the solid was collected by filtration, washed until pH 7 and the
sample dried by lyophilization. Subsequently the length of purified
MWCNTs was shortened by adding pretreated MWCNTs (150 mg) to an
NHPI/DH (20 mg) was added to a round bottom flask (25 mL) containing
N-tert-butyl-α-phenylnitrone (PBN, 1150 mg L ) dissolved in n-dodecane
(10 mL). Then, the flask was sonicated (20 min) and purged with a balloon
-1
acid mixture of concentrated H
in H O (20 mL). The suspension was sonicated at 60 ºC for 1 h. The solid
was washed with water until pH 7 and the H O was removed by
lyophilization. Dry samples of purified MWCNTs were reduced with H2 at
00 ºC for 6 h and labelled as MWCNTsH
2
SO
4
(6 mL) and concentrated HNO
3
(2 mL)
containing O
reaction aliquot was filtered (Nylon filter, 0.2 μm) and the sample purged
with N before recording the EPR. EPR spectra were acquired using a
Bruker EMX spectrometer with the following parameters: frequency 9.803
GHz, sweep width 3489.9 G, time constant 40.95 ms, modulation
frequency 100 kHz, modulation width 1 G, microwave power 19.92 mW
2
. Finally, the system was heated at 120 °C for 4 h. Then, a
2
2
2
5
Covalent anchoring of NHPI 4-carboxylic acid to the surface-modified
supports
Acknowledgements
NHPI 4-carboxylic acid was added in at 10 w/w% respect to the support to
a suspension of p-xylene (40 mL) containing the solid support (DH, ACH
or MWCNTsH).[28] Then, p-toluensulfonic acid (0.1 equivalent) was added
to the suspension as catalyst. The suspension was heated overnight at
reflux temperature. The solid was filtered and washed with hot
dichloromethane. The solid samples were labelled as NHPI/DH,
NHPI/ACH or NHPI/MWCNTsH.
Financial support by the Spanish Ministry of Economy and
Competitiveness (Severo Ochoa and CTQ2015-69153-CO2-R1
andCTQ2014-53292-R and Generalitat Valenciana (Prometeo
2013/14) and is grateful acknowledged. S.N. is thankful for
financial support by Fundacion Ramon Areces (XVIII Concurso
Nacional, Ciencias de la Vida y de la Materia, Energıa renovable:
materiales y procesos 2016).
Catalyst characterization
Fourier transformed infrared (FTIR) spectra were collected using a Nicolet
Keywords: Heterogeneous catalysis • aerobic oxidations • N-
hydroxylphtalimide • diamond nanoparticles as support •
benzylic oxidation • cyclic alkene oxidation
6
700 instrument (Thermo scientific, USA) within the range of 4000-400
–1
cm range. Thermogravimetric analyses (TGA) were performed on a
Perkin Elmer Diamond TGA/DTA STA 6000 in the 25-600 ºC temperature
range under at 5 ºC·min–1 scan rate under an O flow of 20 mL·min–1
2
.
Elemental analyses (EA) were carried out with a Euro EA 3000. X-Ray
photoelectronic spectroscopy (XPS) measurements were carried out using
a spectrometer SPECS equipped with a Phoibos 150 9 MCD detector. The
measurements were performed using a monochromatic source of X-Ray
[
[
1]
2]
Y. Ishii, S. Sakaguchi, T. Iwahama, Adv. Synth. Catal.
001, 343, 393-427.
D. G. Lee, V. S. Chang, J. Org. Chem. 1979, 44, 2726-
730.
2
(Al) at 200 W. Solid-state and liquid phase NMR spectra were recorded in
2
a Bruker AV 400 and a Bruker Advance de 400 MHz, respectively.
This article is protected by copyright. All rights reserved.

ChemCatChem
10.1002/cctc.201700886
FULL PAPER
[
3]
Y. Gao, G. Hu, J. Zhong, Z. Shi, Y. Zhu, D. S. Su, J.
Wang, X. Bao, D. Ma, Angew. Chem. Int. Ed 2013, 52,
[18]
[19]
[20]
[21]
[22]
A. Dhakshinamoorthy, M. Alvaro, H. Garcia,
ChemCatChem 2010, 2, 1438-1443.
Y. Yoshino, Y. Hayashi, T. Iwahama, S. Sakaguchi, Y.
Ishii, J. Org. Chem. 1997, 62, 6810-6813.
A. Rezaeifard, A. Khoshyan, M. Jafarpour, M.
Pourtahmasb, RSC Adv. 2017, 7, 15754-11576.
F. X. Llabres i Xamena, O. Casanova, R. G. Tailleur, H.
Garcia, A. Corma, J. Catal. 2008, 255, 220−227.
A. Gómez-Paricio, A. Santiago-Portillo, S. Navalón, P.
Concepción, M. Alvaro, H. Garcia, Green Chem. 2016, 18,
508-515.
2109-2113.
[
4]
a) S. Navalon, A. Dhakshinamoorthy, M. Alvaro, M.
Antonietti, H. García, Chem. Soc.
Rev.(doi:10.1039/c7cs00156h)) 2017; b) P. Tang, G. Hu,
M. Li, D. Ma, ACS Catal. 2016, 6, 6948-6958; c) O. A.
Kholdeeva, Catal. Sci. Technol. 2014, 4, 1869-1889.
S. Coseri, Catal. Rev. 2009, 5, 218-292.
[
[
5]
6]
a) Y. Ishii, S. Sakaguchi, Catal. Today 2006, 117, 105-
113; b) F. Recupero, C. Punta, Chem. Rev. 2007, 107,
3
800-3842.
[23]
P. Cancino, A. Vega, A. Santiago-Portillo, S. Navalon, M.
Alvaro, P. Aguirre, E. Spodine, H. García, Catal. Sci.
Technol. 2016, 6, 3727-3736.
[
[
7]
8]
l. Melone, c. Punta, Beilstein J. Org. Chem. 2013, 9, 1296-
1310.
J. Tong, Z. Li, C. Xia, Prog. Chem. 2005, 17, 96-110; R. A.
Sheldon, I. W. C. E. Arenas, Adv. Synth. Catal. 2004, 346,
[24]
[25]
G. R. Buettner, Free Radic Biol Med. 1987, 3, 259-303.
O. Fukuda, S. Sakaguchi, Y. Ishii, Adv. Synth. Catal.
2001, 343, 809-813.
O. A. Kholdeeva, Catal. Today 2016, 278, 22-29.
J. Zhang, Z. Wang, Y. Wang, C. Wan, X. Zheng, Z. Wang,
Green Chem. 2009, 11, 1973-1978.
1
051-1071.
A. Dhakshinamoorthy, M. Alvaro, H. Garcia, J. Catal.
012, 289, 259-265.
M. Jian, J. Cui, D. Luo, M. Xie, RSC Adv. 2016, 6, 68170-
8177.
[
[
[
[
[
[
[
[
9]
[26]
[27]
2
10]
11]
12]
13]
14]
15]
16]
6
[28]
[29]
N. Sawatari, T. Yokota, S. Sakaguchi, Y. Ishii, J. Org.
Chem. 2001, 66, 7889-7891.
P. Łątka, K. Kasperczyk, B. Orlińska, M. Drozdek, B.
Skorupska, E. Witek, Catal. Lett. 2016, 146, 1991-2000.
F. Rajabi, B. Karimi, J. Mol. Catal. A.-Chem 2005, 232, 95-
a) J. C. Espinosa, S. Navalón, M. Álvaro, H. García,
ChemCatChem 2015, 7, 2682-2688; b) J. C. Espinosa, S.
Navalón, M. Álvaro, H. García, Catal. Sci. Technol. 2016,
6, 7077-7085; c) D. Sempere, S. Navalon, M. Dančíková,
M. Alvaro, H. Garcia, Appl. Catal. B-Environ 2013, 142-
143, 259-267; d) A. Dhakshinamoorthy, S. Navalon, D.
Sempere, M. Alvaro, H. Garcia, ChemCatChem 2013, 5,
241-246.
G. G. Stavropoulos, P. Samaras, G. P. Sakellaropoulos, J.
Hazard. Mater. 2008, 151, 414-421.
J. C. Espinosa, S. Navalón, M. Álvaro, H. García,
ChemCatChem 2016, 8, 2642-2648.
99.
S. Navalon, R. Martin, M. Alvaro, H. Garcia, Angew.
Chem. Int. Ed. 2010, 49, 8403-8407.
A. Dhakshinamoorthy, S. Navalon, D. Sempere, M. Alvaro,
H. Garcia Chem. Commun. 2013, 49, 2359-2361.
S. Navalon, D. Sempere, M. Alvaro, H. Garcia, ACS Appl.
Mater. Interfaces 2013, 5, 7160-7169.
a) F. Villacañas, M. F. R. Pereira, J. J. M. Órfão, J. L.
Figueiredo, J. Colloid Interf. Sci. 2006, 293, 128-136; b) S.
Navalon, A. Dhakshinamoorthy, M. Alvaro, H. Garcia,
ChemSusChem 2011, 4, 1712-1730.
[30]
[31]
[
17]
A. Santiago-Portillo, S. Navalon, F. Cirujano, F. Llabrés i
Xamena, M. Alvaro, H. Garcia, ACS Catal. 2015, 5, 3216–
3224.
This article is protected by copyright. All rights reserved.

ChemCatChem
10.1002/cctc.201700886
FULL PAPER
Entry for the Table of Contents
FULL PAPER
In contrast to analogous active
carbons and multiwall carbon
nanotubes, OH functionalized
diamond nanoparticles are suitable
supports to develop heterogeneous
oxidation catalysts based on
Juan F. Blandez, Sergio Navalón,
Mercedes Álvaro, Hermenegildo
García*
N-hydroxyphthalimide covalently
anchored on diamond nanoparticles
as selective heterogeneous metal-
free oxidation catalyst of benzylic
hydrocarbons and olefins by
anchored N-hydroxyphthalimide.
molecular O
2
This article is protected by copyright. All rights reserved.