The primary stages of polyoxomolybdate catalyzed cyclohexanone oxidation by hydrogen peroxide as investigated by in situ NMR. Substrate activation and evolution of the working catalyst

Article abstract of DOI:10.1016/j.apcata.2018.05.017

The catalytic process of cyclohexanone oxidation by hydrogen peroxide was investigated using in situ NMR spectroscopy in real working conditions. The behavior of the Keggin heteropolyacid H₃PMo₁₂O₄₀, used as a model catalyst, was explored before and after adding the oxidant agent. This study revealed the evolution pathways to different reduced states of H₃PMo₁₂O₄₀ and its reversible transformation into peroxomolybdate complexes. These latter were identified as the active species for the adipic acid formation, while the acid function of the catalyst was found important for the substrate activation via ketonic-enolic tautomerism. The oxidative mechanism of the cyclohexanone was described through three successive steps to produce adipic acid.

Full text of DOI:10.1016/j.apcata.2018.05.017

Accepted Manuscript
Title: The primary stages of polyoxomolybdate catalyzed
cyclohexanone oxidation by hydrogen peroxide as
investigated by in situ NMR. Substrate activation and
evolution of the working catalyst
Authors: Dahbia Amitouche, Mohamed Haouas, Tassadit
´
Mazari, Sihem Mouanni, Romain Canioni, Cherifa Rabia,
Emmanuel Cadot, Catherine Marchal-Roch
PII:
S0926-860X(18)30239-4
DOI:
Reference:
https://doi.org/10.1016/j.apcata.2018.05.017
APCATA 16666
To appear in:
Applied Catalysis A: General
Received date:
Revised date:
Accepted date:
11-3-2018
18-5-2018
19-5-2018
Please cite this article as: Amitouche D, Haouas M, Mazari T, Mouanni S, Canioni
R, Rabia C, Cadot E, Marchal-Roch C, The primary stages of polyoxomolybdate
catalyzed cyclohexanone oxidation by hydrogen peroxide as investigated by in situ
NMR. Substrate activation and evolution of the working catalyst, Applied Catalysis A,
General (2018), https://doi.org/10.1016/j.apcata.2018.05.017
This is a PDF file of an unedited manuscript that has been accepted for publication.
As a service to our customers we are providing this early version of the manuscript.
The manuscript will undergo copyediting, typesetting, and review of the resulting proof
before it is published in its final form. Please note that during the production process
errors may be discovered which could affect the content, and all legal disclaimers that
apply to the journal pertain.

The primary stages of polyoxomolybdate catalyzed cyclohexanone
oxidation by hydrogen peroxide as investigated by in situ NMR.
Substrate activation and evolution of the working catalyst.
Dahbia Amitouche,a,b,c Mohamed Haouas,a,* Tassadit Mazari,b,c Sihem Mouanni,b,c Romain
Canioni,a Chérifa Rabia,b,c Emmanuel Cadot,a Catherine Marchal-Rocha
a
Institut Lavoisier de Versailles, CNRS UMR 8180, Univ. Versailles Saint Quentin,
Université Paris-Saclay, 45 av. des Etats-Unis, 78035 Versailles cedex (France).
b Laboratoire de Chimie du Gaz Naturel, Faculté de Chimie, Université des Sciences et de la
Technologie Houari Boumediene (USTHB), BP 32, El-Alia, 16111 Bab-Ezzouar, Alger,
Algeria.
c Département de Chimie, Faculté des Sciences, Université Mouloud Mammeri (UMMTO),
15000 Tizi Ouzou, Algeria
Graphical Abstract
1

Highlights
 In situ NMR spectroscopy was employed to monitor oxidative conversion of
cyclohexanone by hydrogen peroxide and Keggin-type phosphomolybdates.
UV-Vis spectroscopy was used to identify the various reduced and oxidized
phosphomolybdate species.


Activation of the substrate via ketonic-enolic tautomerism was evidenced as the
primary step.
 Peroxomolybdate-type complexes were identified as active species for oxidation
process.
ABSTRACT
The catalytic process of cyclohexanone oxidation by hydrogen peroxide was investigated
2

using in situ NMR spectroscopy in real working conditions. The behavior of the Keggin
heteropolyacid H3PMo12O40, used as a model catalyst, was explored before and after adding
the oxidant agent. This study revealed the evolution pathways to different reduced states of
H3PMo12O40 and its reversible transformation into peroxomolybdate complexes. These latter
were identified as the active species for the adipic acid formation, while the acid function of
the catalyst was found important for the substrate activation via ketonic-enolic tautomerism.
The oxidative mechanism of the cyclohexanone was described through three successive steps
to produce adipic acid.
Keywords:
Nuclear magnetic resonance
Operando spectroscopy
Multinuclear NMR
Phosphomolybdate polyanions
1. Introduction
The worldwide demand for clean chemical processes has expanded greatly in recent
decades [1]. Intense efforts have therefore been devoted to the development of catalytic
systems employing benign reagents with respect to the environment [2]. The utilization of
green oxidants, such as oxygen (or air) or hydrogen peroxide (H2O2), together with
environmentally friendly solvents, such as water, is of particular interest in catalytic oxidation
processes [3-9]. H2O2 is particularly attractive reagent not only due to its high contents of
active oxygen species, but also because its reduction leads only to water as by-product [10].
Adipic acid is of a great interest in the manufacture of various commercially valuable
products such as nylon-6,6 and polyamide [11-13]. The current industrial procedure of its
production is based on two-step process [14], where cyclohexane is first oxidized in presence
of air to a mixture of cyclohexanol and cyclohexanone (first step), which is further oxidized to
adipic acid by nitric acid (second step). During this process, the used nitric acid causes the
N2O emission, which is obviously an undesired byproduct [15, 16]. Others efficient
ecologically friendly catalysts are therefore needed to replace nitric acid.
Polyoxometalates (POMs) consisting of a large family of anionic metal–oxygen
nanoscopic clusters [17-20] have especially received much attention in the area of the
oxidation catalysis because of their strong oxidative effeciency beside their strong Brønsted
and Lewis acidities [21, 22]. In particular, molybdenum based POMs have been shown to be
effective catalysts for green oxidations using H2O2 or O2 [23-27]. The remarkable catalytic
3

activities of POMs could be related to the multifunctional character of the active sites at their
surface. For instance, their ability to activate simultaneously the substrate and oxidant is well
recognized by stabilizing reaction intermediates, and facilitating oxygen or multi-electron
transfer [26-31]. Although catalytic performance of POMs is widely studied and in some
cases well established, there is still a lack of rational and systematic use of knowledge. The
need for kinetic and mechanistic studies is obvious to shed light on the intimate mechanisms
of oxidative transformation.
Previous studies revealed different mechanisms of organic substrates oxidation catalyzed
by POMs and reaction pathways were found to be dependent on reaction conditions and
catalytic systems [32-34]. In this respect, diverse experimental approaches for the
investigation of the catalytic behavior of material under working conditions were developed,
in particular, spectroscopic methods [35-37]. Among these techniques, NMR spectroscopy is
considered to be one of the most informative, since it allows following the fate of both
reactant and catalyst during the time course of the reaction [38-40]. Furthermore, it offers the
possibility to obtain quantitative results on the transformations of the substrates and thus
permits to study reaction kinetics in situ [41].
In this study, the early stages of cyclohexanone oxidative conversion, in presence of the
Keggin type phosphomolybdate anion PMo12O403-, are investigated in situ by means of
multinuclear (95Mo, 31P, 17O, and 1H) NMR spectroscopy. The evolution of the chemical state
3-
of PMo12O40 during the catalytic cycle is revealed as well as the activated form of the
substrate allowing to suggest a general mechanism of the catalytic process. The initial
experimental conditions (substrate and catalyst concentrations, solvent, amount of H2O2, etc.)
have found to affect greatly the stability of the catalyst and its evolution.
2. Experimental section
2.1 Chemicals and materials
1,4-Dioxane (99.5%, Sds), N2H4•H2O (98%, Alfa Aesar), H3PO4 (85wt% in water,
Merck), Na2MoO4•2H2O (99.5% Sigma-Aldrich), H2O2 (30wt% in water, Sigma-Aldrich),
D2O (99.90% D, euriso-top), cyclohexanone (99.0%, Prolabo), and tetrabutylammonium
4

bromide (99.0%, Acros Organics), were used as obtained from commercial suppliers.
Hydrogen peroxide concentration was verified by potassium permanganate titration.
2.2 Physical methods
2.2.1 Fourier Transform Infrared (FT-IR) and UV-Vis spectroscopies
FT-IR spectra were recorded on a 6700 FT-IR Nicolet spectrophotometer, using diamond
ATR technique. UV-Vis spectra were measured on a Lambda 19 Perkin Elmer
spectrophotometer in 0.1 cm quartz cell. To monitor formation of the reduced species as a
function of time, a solution of the initial oxidized H3[PMo12O40] in 50/50 (v/v) water/dioxane
mixture is prepared and then one equivalent of hydrazine is added. The electronic spectra
were recorded every 2 h over 66 h of total measurement time.
2.2.2 Nuclear magnetic resonance (NMR) spectroscopy
Spectra were recorded on a Bruker Avance 500 MHz spectrometers using standard 10 mm
1
NMR tubes and fixing sample volume to 3 mL. Typically, H NMR spectra were obtained
accumulating 8 scans and using 1 s acquisition time, 15 s relaxation delay, and 24 s pulse
length (/2 flip angle). 31P NMR spectra were recorded with 32 numbers of scans, 1 s
acquisition time, 51 s relaxation delay, and 15 s pulse length (/2 flip angle). The 17O NMR
spectra were recorded with an accumulation of ca. 2000000 scans, 0.02 s acquisition time, 0.1
s relaxation delay, and 10 s pulse length (/2 flip angle). 95Mo NMR spectra were run using
ca. 16000 scans, 0.2 s acquisition time, 0.1 s relaxation delay, and 3 s pulse length (/12 flip
angle). Chemical shifts were referenced to external standards (δ = 0 ppm) that are
1
Tetramethylsilane for H, 85wt% H3PO4 for 31P, H2O for 17O, and 2 M Na2MoO4 aqueous
95
solution for Mo. The variable temperature experiments were conducted using a Eurotherm
temperature unit within the range 30-70 °C, and the actual temperature in the tube was
calibrated using ethylene glycol protocol reported in the literature [42]. Various in situ
experimental conditions were investigated and some typical experiments were as follows: i) m
PMo12 + 3 mL D2O/H2O2; ii) m PMo12 + 3 mL cyclohexanone/H2O2; iii) m PMo12 + 0.5 mL
cyclohexanone + 2.5 mL D2O/H2O; where m = 30 or 300 mg. These experiments where
conducted for two different temperatures, namely 27 °C or 55 °C. D2O was introduced only
when it is possible in aqueous phase, while in neat cyclohexanone system no lock was used.
5

2.3 Syntheses of the Keggin phosphomolybdate and its reduced derivatives
The compound H3[PMo12O40].13H2O was prepared as described by Courtin [43]. FT-
IR/cm-1 (Diamond ATR, ATR correction applied): 1064 (P-Oa), 961 (Mo=Ot), 869 (Mo-Ob-
Mo), 786 (Mo-Oc-Mo). The control of kinetic parameters (temperature and time) is essential
for the selective isolation of the reduced derivatives. These parameters are determined
according to preliminary kinetics studies (see section 3.2.1). The optimal conditions for
formation and isolation of the corresponding tetrabutylammonium (TBA) salts are as follow.
2.3.1 TBA4H[-PMo12O40], abbreviated TBA4H{-II}
In a flask, 5 mL of a 50/50 (v/v) water/dioxane solution of H3[PMo12O40] (C = 6.0 10-2
M), is mixed with 600 μL of a hydrazine solution (C = 0.53 M). The molar ratio
H3[PMo12O40]:N2H4 was fixed to 1:1. The mixture was stirred for 24 h at room temperature
and the product was then isolated as a TBA salt precipitate after adding 587 mg of TBABr
(1.8 mmol). FT-IR/cm-1 (Diamond ATR, ATR correction applied): 1059 (P-Oa), 954 (Mo=Ot),
857 (Mo-Ob-Mo), 794 (Mo-Oc-Mo).
2.3.2 TBA3H4[-PMo12O40] abbreviated TBA3H4{-IV}
In a flask, 5 mL of a 50/50 (v/v) water/dioxane solution of H3[PMo12O40] (C = 6.0 10-2
M), is mixed with 600 μL of a hydrazine solution (C = 0.53 M). The molar ratio
H3[PMo12O40]:N2H4 was fixed to 1:1. The mixture was stirred for 45 h at room temperature
and the product was then isolated as a TBA salt precipitate after adding 440 mg of TBABr
(1.35 mmol). FT-IR/cm-1 (Diamond ATR, ATR correction applied): 1059 (P-Oa), 951
(Mo=Ot), 876 (Mo-Ob-Mo), 799 – 768 (Mo-Oc-Mo).
2.3.3 TBA3H4[-PMo12O40] abbreviated TBA3H4{-IV}
In a flask, 5 mL of a 50/50 (v/v) water/dioxane solution of H3[PMo12O40] (C = 3.0 10-2
M), is mixed with 109 μL of a hydrazine solution (C = 2.06 M). The molar ratio
H3[PMo12O40]:N2H4 was fixed to 1:1.5. The mixture was heated at 70 °C under stirring for 3 h
and the product was then isolated as a TBA salt precipitate after adding 220 mg of TBABr
(0.68 mmol). FT-IR/cm-1 (Diamond ATR, ATR correction applied): 966 – 952 (Mo=Ot).
6

In the solid state, the derivatives TBA4H{-II} and TBA3H4{-IV} are not stable and
undergo slow reoxydation, as shown by their color change to green after few weeks. The {-
IV} compound appears however much more stable over time.
2.4 Catalytic reaction
The adipic acid synthesis was carried out according to procedure of previous studies [44,
45]. The oxidation of cyclohexanone in homogeneous phase was performed under reflux at 90
°C. A POM color change from yellow to blue green (characteristic color for intervalence
transfert between MoV and MoVI) is observed. Then, 0.5 mL of hydrogen peroxide (30%) was
added to restore the MoVI oxidation state of the catalyst, characterized by yellow color. This
sequence is repeated after each color change until there is no more change of color. At that
moment, the reaction end is reached and the POM catalyst is no longer reduced. The resultant
homogeneous mixture was cooled at 4 °C overnight to isolate adipic acid by cold
crystallization and its identification was made by means of FT-IR and NMR spectroscopy and
also its characteristic melting point at 152 °C.
3. Results and discussion
3.1 Catalytic tests
Some catalytic tests were conducted in order to search the reaction conditions of the
liquid-phase cyclohexanone oxidation to adipic acid by hydrogen peroxide. It is emphasized
that the adipic acid synthesis requires both presence of substrate and catalyst, and hydrogen
peroxide must be added gradually after each reduction of the catalyst by the substrate. In this
work, the effects of catalyst/substrate molar ratio, solvent presence and reaction time on
adipic acid yield were examined.
Table 1 summarizes the adipic acid yield obtained for the different reaction conditions
studied. The highest yield recorded was obtained after 20 h reaction at 90 °C using 30 mg of
catalyst in 3.16 mL (30 mmol) cyclohexanone. Increase of the catalyst loading or decrease of
the substrate amount reduced the product yield. However, an optimum catalyst/substrate ratio
of 30 mg for 30 mmol was found for the highest adipic acid yield. Finally, presence of organic
solvent inhibited the catalytic reaction.
7

Table 1: Adipic acid yield (%) as a function of reaction conditions.a
Reaction time
(h)
Catalyst weight
(mg)
Substrate amount
Solvent
Adipic acid yield
(%)b
(mmol)
reaction time effect
8
30
30
30
30
30
30
30
30
none
none
none
none
23
31
43
36
14
20
26
catalyst weight effect
20
20
20
20
30
60
90
30
30
30
30
none
none
none
none
43
36
41
39
120
substrate amount effect
20
20
20
20
20
30
30
30
30
30
15
30
45
60
none
none
none
none
none
21
43
35
33
22
75
solvent effectc
20
20
20
20
20
30
30
30
30
30
30
30
30
30
30
none
CH3COOH
CH3CN
CH3OH
CHCl3
43
17
32
0
0
a)
Fixed reaction conditions: reaction temperature = 90 °C, agitation rate = 1000 rpm, amount of successive
addition of H2O2 (30wt%) = 0.5 ml. b) based on weight of isolated product. c) volume of solvent = 3.16 mL (equal
to volume of substrate for 30 mmol).
In summary, these preliminary catalytic tests allowed identifying the optimum reaction
conditions with H3PMo12 catalyst. The best results were obtained using
H3PMo12/cyclohexanone molar ratio of 5 10-4 corresponding to a catalyst mass of 30 mg and
substrate amount of 30 mmol, after 20 hours of reaction under agitation in solvent free
medium. The next sections will be devoted to determine the evolution of the catalyst and to
study the course of the reaction at molecular level.
3.2 The chemical states of the catalyst
In order to identify the different states of the catalyst, a preliminary study of the reduction
of the [PMo12O40]3- anion was carried out in solution combining NMR and UV-Vis
spectroscopies. Furthermore, the solution behavior of the POM catalyst in presence of H2O2
8

was also studied by means of multinuclear NMR spectroscopy. These studies aim to identify
and characterize the different forms of the POM for the catalytic oxidation of cyclohexanone
by hydrogen peroxide.
3.2.1 Reduced species of POMs
The most stable isomer of the anion [PMo12O40]3-, in its oxidized state, is the  form
(denoted {-0}) of Td symmetry. A preliminary electrochemical study showed that it was
possible to successively form a 2-electron reduced derivative (denoted {-II}), followed by a
4-electron reduced derivative (denoted {-IV}). This latter then underwent a structural
evolution toward a more stable isomer (denoted {-IV}) of C3v symmetry. This isomerization
results from the formal rotation of 60° of a trimeric group {Mo3O13}, probably associated
with a minimization of the coulombic repulsions within the polyanionic structure.
Experimentally, during the electrolysis, we observed a decomposition of the reduced POM
above 4 electrons in pH conditions of the study.
In 2000, Yamase et al. have described the photoreduction process of an aqueous solution
of H3[-PMo12O40] at pH = 2 by methanol [46]. They were able to isolate and record the
structures of the derivatives {-II} and {-IV} by single crystal X-ray diffraction and to
31
follow the evolution of the reaction by P NMR. More recently, Maksimovskaya studied the
31
electron-transfer reactions in molybdophosphate heteropoly blues using P NMR, revealing
one-, two-, and four-electron reduced [PMo12O40]n- species in aqueous solutions [47].
Based on the 31P NMR chemical shifts of the compounds {-II} and {-IV}, their
syntheses were optimized in order to isolate these reduced derivatives. In this work, the
reduced derivatives syntheses were carried out by solubilizing the H3PMo12O40 acid in a
water/dioxane medium in the presence of hydrazine, used as a reducing agent capable of
supplying four electrons according to the following equation: N2H4  N2 + 4 H+ + 4 e-. The
reduced derivatives were isolated from the reaction medium by precipitation with
tetrabutylammonium cation. The saturated salt solutions in water/dioxane before precipitation
were analyzed by 31P NMR. These spectra are presented in Fig. 1.
9

Fig. 1: Room temperature (27 °C) 31P NMR spectra in 50:50 D2O:dioxane solutions of a) the oxidized Keggin
anion [PMo12O40]3- {-0}, and reduced derivatives b) [PMo12O40]5- {-II}, c) [PMo12O40]7- {-IV}, and d)
[PMo12O40]7- {-IV}: a) 60 mM of H3[PMo12O40]; b) after addition of 1 eq. N2H2 to the POM solution in a) and
24 h room temperature stirring ; c) after addition of 1 eq. N2H2 to the POM solution in a) and 45 h room
temperature stirring; d) after addition of 1.5 eq. N2H2 to the POM solution in a) and 3 h stirring at 70 °C.
The oxidized ion, [PMo12O40]3- {-0}, has a characteristic resonance at -3.1 ppm. Its
reduction to 2 electrons causes a significant change in chemical shift at -5.5 ppm,
characteristic of a shielding of the 31P nucleus. The introduction of two other additional
electrons (-IV) results in phosphorus deshielding ( = -4.0 ppm), while the isomerization of
the {-IV} ion to {-IV} has an opposite effect resulting in the appearance of a peak at -12.5
ppm. In conclusion, the reduction of the ion [PMo12O40]3- causes a shielding of the 31P nucleus
and an increase in the charge of the polyanion that generates weak acidities responsible of the
POM protonation. Thus, the variation in the chemical shift of the phosphorus core could be
due to two antagonistic effects: i) a shielding associated with the introduction of electrons into
the POM framework, and ii) a deshielding resulting from the protonation and therefore the
change in the acidity strength. Moreover, the isomerization that results in a significant
variation of the chemical shift towards negative values, causes besides a structural
modification associated with a lowering of the symmetry, a modification of the 4d electrons
distribution and therefore acidity modification.
UV-Vis study was carried out to further characterize the different phosphomolybdate
species formed in solution in presence of reducing agent. The curve of t = 0 h in Fig. 2
corresponds to the electronic spectrum of the starting acid H3[PMo12O40] {-0} before
addition of hydrazine. It does not absorb in the visible. The bands at 200 and 312 nm are
attributed to O  MoVI charge transfer. However, these bands are wide enough and intense to
allow a weak absorption at the visible limit, thus giving a yellow color. On the other hand, the
10

broad absorptions observed in the visible and near-infrared domains, between 400 and 1200
nm during the reduction of the [PMo12O40]-3 ion, are due to MoVI  MoV intervalence
electronic transitions responsible for the dark blue color of the solution. This is therefore a
direct evidence of the reduction of POM. The addition of hydrazine to the starting yellow
solution of the POM instantly causes a blue-green color characteristic of the reduction of the
polyanion. The reduction results in the rapid growth of the bands at 520 and 785 nm. The
maximum observed at 785 nm is typical of the reduced two-electron derivative {-II}. The
formation of the four-electron derivative {-IV} takes longer time and results in a decrease in
the absorbance of the 785 nm band in favor of two new bands at 710 and 900 nm. The
formation of the isomerized four-electron derivative {-IV} at room temperature is too slow
and was not observed in this experiment. Its electronic spectrum could nevertheless be
recorded from the synthesis solution described in experimental part 2.2.3 after verification by
31P NMR.
Fig. 2: UV-Vis kinetics monitoring the transformation of [PMo12O40]3- {-0} into their reduced derivatives,
[PMo12O40]5- {-II} and [PMo12O40]7- {-IV} at room temperature under the effect of hydrazine. The
measurements start immediately after addition of 1 eq. N2H2 to 0.6 mM of H3[PMo12O40] in 50:50 H2O:dioxane
solution over a period of 66 h at 27 °C. The arrows indicate the direction of evolution of spectra with time.
UV-Vis of the oxidized parent {-0} and the isolated reduced derivatives, as two-
electrons {-II}, four electrons {-IV}, and isomer four electrons {-IV}, are also recorded
and the spectra are superimposed in Fig. 3. The isomerization to {-IV} results in the
decrease of the absorption located at 710 nm, the growth of the maximum observed at 900
nm, as well as the lowering of the wavelengths of these characteristic bands by approximately
35 nm relative to that observed with its -isomer counterpart {-IV}.
11

Fig. 3: Room temperature (27 °C) UV-Vis spectra in 50:50 H2O:dioxane solutions of the parent [PMo12O40]3-
{-0} and the isolated reduced derivatives, [PMo12O40]5- {-II}, [PMo12O40]7- {-IV}, and [PMo12O40]7- {-IV}:
a) 0.6 mM of H3[PMo12O40]; b) after addition of 1 eq. N2H2 to the POM solution in a) and 24 h room temperature
stirring ; c) after addition of 1 eq. N2H2 to the POM solution in a) and 45 h room temperature stirring; d) after
addition of 1.5 eq. N2H2 to the POM solution in a) and 3 h stirring at 70 °C.
The evolution of the POM species in the presence of hydrazine, as a function of time
at 70 °C, was then monitored by means of 31P NMR spectroscopy. Just before 31P NMR
analysis, 1.5 equivalents of hydrazine was added to 60 mM H3[PMo12O40] in D2O/dioxane
(50/50) solution. Spectra were recorded every 10 min during 5 h, total reaction time. The
integration of the different peaks, now identified, gives an overview of the evolution of these
species over time and allows determination of their distribution as a function of time (Fig. 4).
Fig. 4: Evolution of the oxidized Keggin anion [PMo12O40]3- {-0} toward reduced derivatives [PMo12O40]5- {-
II} and [PMo12O40]7- {-IV} at 70 °C as a function of time as measured by 31P NMR. The measurements start
immediately after addition of 1.5 eq. N2H2 to 60 mM of H3[PMo12O40] in 50:50 D2O:dioxane solution.
12

The obtained results show disappearance of the signal at -3.1 ppm, characteristic of the
oxidized species {-0} during the first two hours and the appearance of a signal at -5.45 ppm
assigned to the two electrons reduced species {-II} followed by that of the isomerized four-
electron derivative {-IV} ( = -12.5 ppm). This latter appears clearly as the final product
after 4.5 h of reaction, whereas the {-II} species is an intermediate with a maximum amount
at ca. 2 h. In this study, the non-isomerized four-electron derivative {-IV} ( = -4.0 ppm)
was not observed. Under the conditions of the experiment (T = 70 °C), the isomerization
process is too fast and proceeds quasi simultaneously as soon as the compound {-IV} is
formed. Reduction of H3[PMo12O40] by hydrazine is a relatively slow process. Heating
significantly accelerates the reaction leading to the most stable 4-electron derivative that is
{-IV} anion. The latter no longer accepts electrons and when left in the presence of
hydrazine, it decomposes as indicated by the appearance of a resonance peak due to phosphate
ions.
3.2.2 Peroxo species
In the presence of hydrogen peroxide, POMs usually decompose into peroxo-based
metalate (tungstate or molybdate) complexes [21, 23, 29, 32, 48-50]. In the case of
31
phosphotungstate-peroxo complexes, the three main species identified by P NMR using the
2
scalar coupling J(31P-183W) were {PO4[WO(O2)2]4}3-, {PW3Om}n-, and {PO4[WO(O2)2]2}2-,
denoted PW4, PW3, and PW2 respectively [48, 49, 51, 52]. Such species are known to play a
key role in many oxidation reactions of organic substrates [53, 54]. With molybdenum-based
POM catalysts, similar peroxomolybdate complexes were also suggested as intermediate
active species [24].
In order to investigate the hydrogen peroxide action on the catalyst, H2O2 was
progressively added to an aqueous solution of H3[PMo12O40] (58 mM). After each H2O2
31
31
addition, the solution was analyzed by P NMR. The series of P NMR spectra is presented
in Fig. 5. The spectrum of the initial POM exhibits the characteristic signal of the Keggin
phosphomolybdate at -2.7 ppm. Upon addition of two equivalents of H2O2, two new
resonance peaks appeared at -0.5 and 0.6 ppm assigned to lacunary POMs [PMo11O39]7- and
[PMo9O34]9-, respectively [47]. Further addition of H2O2 (26 equivalents) led to the
disappearance of all the previous signals and the appearance of three main broad resonances
13

at 3.9, 3.0, and 1.8 ppm that could be attributed to peroxomolybdate complexes, respectively,
{PO4[MoO(O2)2]4}3- (PMo4), {PMo3Om}n- (PMo3), and {PO4[MoO(O2)2]2}2- (PMo2), similar
to PW4, PW3, and PW2 respectively. These latter showed peaks at 3.5, ~2, and 0.4 ppm in
acetonitrile solutions [51]. Such results are fully consistent with progressive decomposition of
the Keggin POM in presence of hydrogen peroxide and formation of novel peroxo complexes
as discussed in the literature [24, 53, 55]. Nonetheless, peroxo tungstate complexes also
showed ability to transform into Keggin derivative during oxidation of cyclopentene with
aqueous H2O2 [56]. This would indicate reversible structural rearrangement of the catalyst
during the catalytic processes.
Fig. 5: 31P NMR spectra of the Keggin anion [PMo12O40]3- in H2O/D2O (58 mM) in the presence of 0, 2, 6, and
26 equivalents of H2O2.
To further characterize the species present in the aqueous solution of [PMo12O40]3- in
presence of hydrogen peroxide, 95Mo NMR spectra were recorded for the same solutions used
in 31P NMR and the related results are shown in Fig. 6. The 95Mo NMR spectrum of
[PMo12O40]3- exhibits a characteristic broad signal at 21 ppm [57]. After addition of few
equivalences of H2O2, a shoulder at -14 ppm and a new resonance peak at -191 ppm appeared.
The former could be assigned to lacunary Keggin [PMo11O39]7-, and the second one to the
peroxomolybdate complex MoO(O2)2(H2O)2, according to Talsi et al. [58]. In the presence of
95
large excess of H2O2 (26 equivalents), all Mo resonances were completely transformed into
peroxomolybdate species with a main signal at -191 ppm and another one less intense at -242
ppm. These two signals should correspond to quite similar environments around the Mo
14

center. The signal at -242 ppm is therefore tentatively assigned to PMon peroxo complexes (n
31
= 3, 4) as observed in P NMR. It should be mentioned that such peroxo complexes would
adopt a heptavalent configuration where the coordination sphere is completed by solvent
31
95
molecules [36, 58, 59]. From the P and Mo NMR data, it appears clearly that the main
molybdenum based peroxo species produced are MoO(O2)2(H2O)2, PMo4, PMo3, and PMo2.
Fig. 6: 95Mo NMR spectra of the Keggin anion [PMo12O40]3- in H2O/D2O (58 mM), before and after adding 2, 6,
and 26 equivalents of H2O2.
17
Fig. 7 shows the O NMR spectra of the phosphomolybdate species before and after
addition of 6 and 26 equivalents of hydrogen peroxide. The spectrum of the starting
compound exhibits the four expected resonances typical of the Keggin structure at 70, 551,
571, and 946 ppm assigned to (P-Oa), (Mo-Ob-Mo), (Mo-Oc-Mo), and (Mo=Ot) oxygen types,
respectively [60]. These signals disappear upon addition of large excess of hydrogen peroxide
and novel resonance peaks are observed, corresponding to the peroxo species formed. The
signal at 872 ppm and the very broad resonance at ca. 440 ppm, assigned to oxo Mo=O and
peroxo groups respectively, are characteristic of the peroxo molybdenum complex
95
MoO(O2)2(H2O)2 in agreement with Mo NMR results [58]. The intermediate spectrum after
addition of 6 eq. H2O2 showed loosely defined signals of the starting Keggin anion and
peroxo species present as minor products while signatures of prominent species, the low-
31
symmetrical lacunary Keggin [PMo11O39]7-, according to P and 95Mo NMR, could not be
17
observed at O natural abundance level due to the dispersion of O sites. The signal at 181
ppm is assigned to H2O2, visible when this alter is used in large excess.
15

Fig. 7: 17O NMR spectra of the Keggin anion [PMo12O40]3- in H2O/D2O (58 mM), before and after adding 6, and
26 equivalents of H2O2.
This preliminary study allows the identification of the main reduced phosphomolybdate
species produced from the Keggin H3[PMo12O40] under reduction conditions and the
molybdate and peroxomolybdate species formed in oxidation condition with H2O2. Table 2
summarizes their NMR characterization in comparison with the literature. These
spectroscopic data will serve as references for identifying the main catalytic active species in
the sequential oxidation process of cyclohexanone by POM and hydrogen peroxide that will
be presented in next sections.
Table 2
NMR data ( / ppm) in aqueous solution of main species formed upon transformation of
H3[PMo12O40] in presence of reducing or oxidant agent.
Compound
31P
95Mo 17O
Initial POM
-3.1a 21
70; 551; 571; 946
514; 525; 900b
[PMo12O40]3- {-0}
Reduced species
[PMo12O40]5- {-II}
[PMo12O40]5- {-II}
[PMo12O40]7- {-IV}
[PMo12O40]7- {-IV}
Peroxo species
-5.5a
-6.6c
-4.0a
-12.5a 307c 70; 453; 515; 548; 555; 906; 929c
MoO(O2)2(H2O)2
-191 440; 872
-242
3-
PO4[MoO(O2)2]4 {PMo4} 3.5
PMo3Omn- {PMo3}
2.0
2-
PO4[MoO(O2)2]2 {PMo2} 0.4
a) in 50:50 water:dioxane solution; b) from reference [60]; c) from reference [47].
16

3.3 Monitoring the state of the working catalyst
3.3.1 The first reduction stages
It is known that H3[PMo12O40] acid is highly soluble in many organic solvents. Thus, its
behavior toward the cyclohexanone was examined by 31P NMR analysis. A solution
containing 30 mg H3[PMo12O40] in 3 mL of cyclohexanone was maintained at 55 °C over a
period of 20 h during the NMR measurements. The evolution of the spectra as a function of
time is showed in Fig. 8. The spectrum of the solution as prepared within the first hour
exhibits a main signal at -2.3 ppm corresponding to the starting oxidized POM {-0}, and a
secondary resonance at ca. -4.1 ppm that could be due to the two electron reduced species {-
II}. This signal increased quickly during increasing temperature and then decreased until
complete disappearance after a prolonged heating time. The main signal at -2.3 ppm gradually
moved to ca. -3.0 ppm after 20 h of heating. This observation is interpreted as a result of fast
interconversion exchange between two distinct species, most probably [PMo12O40]3- {-0}
and [PMo12O40]5- {-IV} based on the observed chemical shifts.
Fig. 8: Time-dependent 31P NMR of a mixture of 30 mg H3[PMo12O40] + 3 mL cyclohexanone at 55 °C over a
period of 21 h. Inset: Change in chemical shifts of main resonance (-2.3 - -3.0 ppm).
The differences in chemical shifts of the species, {-0}, {-II}, and {-IV}, i.e.,
systematic 1-1.5 ppm low-field shift compared to previous study in 3.2 section are explained
by the solvent effect, cyclohexanone medium versus pure aqueous or mixed aqueous-dioxane
solutions. Furthermore, the sequential appearance of these signals is also consistent with the
17

previously established shielding effect order, {-0} < {-IV} < {-II}. The reduction of
the POM is confirmed by the color change of the sample from yellow (starting oxidized
POM) to dark blue at the end of the experiment, after 20 h at 55 °C. Such a phenomenon
appeared much faster in cyclohexanone than in aqueous medium confirming the interaction
between the catalyst and the substrate.
The second remarkable difference with the behavior of the aqueous solution is the
intriguing fast exchange process between {-0} and {-IV}. This four-electron reduced
species is favored at high temperature and should be obtained through reduction from the two-
electron reduced species {-II}. In an additional experiment without heating, one can see first
almost complete conversion of {-0} to {-II} after 6 h at 27 °C (Fig. S1) and then only after
heating at 55 °C that {-II} is transformed into {-IV} after 15 h at 55 °C (Fig. S2). Thus it is
reasonable to assign the final signal at –3 ppm to {-IV}.
3.3.2 Effect of adding H2O2
31
The behavior of POM under reduction and oxidation cycles was monitored by P NMR
analysis as a function of time. The first period consists of reducing [PMo12O40]3- {-0} (30
mg) by cyclohexanone (3 mL), during which a color change from yellow to blue is observed.
Then, 0.5 mL of H2O2 was added twice, first in the beginning of the second period of again 20
h at 55 °C and in the beginning of the third period of 44 h at 55 °C. Fig. 9 shows the 31P NMR
spectra at the beginning and end of each period.
Fig. 9: Representative 31P NMR of a mixture of 30 mg H3[PMo12O40] + 3 mL cyclohexanone at 55 °C over a
first period of 20 h, followed by a subsequent addition of 0.5 mL H2O2 over a period of 20 h at 55 °C, and then a
18

second addition of 0.5 mL H2O2 over a period of 44 h at 55 °C. Note that t = 0 h refers to NMR measurement
starting after ca. 1 h of sample preparation. Asterisks (*) indicates unknown decomposition products, possibly
lacunary Keggin POMs.
The complete series of spectra are shown in Fig. 8 for the first period, and in Supporting
Information for the second and third periods (Figs. S3, and S4). The spectrum at the
beginning of the first period is characterized as seen previously by the signals of {-0} and
{-II} (-2.3 and -4.1 ppm, respectively). After 20 h at 55 °C, the main species left is the four-
electron reduced {-IV} (-3.0 ppm). Addition of H2O2 solution restores the initial signal at -
2.3 ppm of the starting oxidized form {-0}. After further reaction at 55 °C for 20 h, reduced
species {-II} and {-IV} reappeared again, but accompanied also by the formation of {-
IV}. Interestingly enough, this latter species appeared only after intervention of water in the
cyclohexanone oxidation. Furthermore, in presence of water the two species {-0} and {-
IV} will coexist in slow chemical exchange. The second H2O2 addition at the end of this
second period, allowed conversion of all reduced species, not only into the starting oxidized
POM {-0}, but also to some decompositions products and peroxo complexes characterized
by low-field signals in the range 2-3 ppm. These results would also indicate that peroxo
complexes play a key role in catalytic oxidative cyclohexanone conversion.
Quantitative analysis of the NMR data offers a way to follow the evolution of catalytic
species. Fig. 10 displays the distribution of the different forms of the catalyst in their evolving
medium during the three successive periods studied, i) the reduction in cyclohexanone, ii) first
and iii) second subsequent reoxidation by H2O2. The proportions of species are determined
from NMR integration of the corresponding signals. In the case of fast chemical exchange
between species {-0} and {-IV} in the first period, their relative proportion are determined
from the observed chemical shift (obs = x { -0} + (1 - x){ -IV}). During the first period (the


reduction by cyclohexanone), one can see the continuous decrease of both oxidized and two-
electron reduced species {-0} and {-II} in favor of the four-electron reduced species {-
IV} that becomes the prominent form after 20 h of reaction at 55 °C. This observation
confirms that this latter is formed most probably from conversion of both {-II} and species
{-0} directly. Furthermore, significant NMR signal loss is noticed (up to 25%) that should
account for other non-detected phosphorus-based species. Such species could represent
paramagnetic intermediates with one or three electron reduction. In period II, H2O2 addition
19

led to the oxidized species {-0} but not quantitatively, where less than 20% of initial
quantity is only observed. The remaining undetected fraction probably corresponds to
paramagnetic species. After 5 h at 55 °C, the first reduced species formed are {-II} followed
few hours later by {-IV} and {-IV}. Note the formation of peroxo species in very few
amounts at prolonged reaction time. After the second addition of H2O2 (period III), similar
observations took place, i.e., formation of large amount of undetected species, regeneration of
oxidized species {-0}, and appearance of peroxo species in significant amount. This latter
decreased quickly with time at the expense of the {-0} species, that in its turn decreased few
hours later in favor of four-electron reduced species {-IV} and {-IV}. Note that the {-IV}
continued to increase and {-IV} continued to decrease. This clearly indicates that the final
stable four-electron reduced species should be the -isomer as it appears much later.
20

Fig. 10: Phosphorus species distribution based on 31P NMR in a mixture of 30 mg H3[PMo12O40] + 3 mL
cyclohexanone at 55 °C over a period of 20 h (period I), followed by a subsequent addition of 0.5 mL H2O2 over
a period of 20 h at 55 °C (period II), and then a second addition of 0.5 mL H2O2 over a period of 44 h at 55 °C
(period III).
3.3.3 Effect of water
Another similar series of experiments have been performed with D2O. 300 mg of
H3[PMo12O40] are dissolved in a mixture of 2.5 mL D2O and 0.5 mL cyclohexanone. The
system is now biphasic because of the immiscibility of cyclohexanone with water. Fig. 11
shows some selected spectra of 31P NMR analysis of reaction mixture. The complete series of
spectra corresponding to the reduction of [PMo12O40]3- by cyclohexanone at 55 °C over a
period of 16.5 h and subsequent addition of 0.5 mL H2O2 and reaction at 55 °C for 17.5 h are
21

shown respectively in Figs. S5 and S6 of Supporting Information. In this system, the same
species are observed with D2O, except the signal of species {-IV} that was not detected. The
signals of species {-0}, {-II}, and {-IV} appeared doubled because of two phases,
aqueous and organic. It is interesting to note the formation of the four-electron reduced
species {-IV} already in the first reduction period in absence of intermediate species {-
IV}. This observation has already been seen in pure aqueous system (see Fig. 4), where the
transition {-IV}  {-IV} was too fast to be detected. One could conclude, therefore, the
isomer {-IV} is stabilized to some extent in the organic phase. The second interesting
remark concerns the ease of peroxo species formation upon H2O2 addition. This would
indicate peroxo species are more favored in aqueous medium than in organic substrate
environment.
Fig. 11: Representative 31P NMR of a mixture of 300 mg H3[PMo12O40] + 2.5 mL H2O + 0.5 mL cyclohexanone
at 55 °C over a first period of 16.5 h, and then after addition of 0.5 mL H2O2 over a second period of 17.5 h at 55
°C.
In Fig. 12, one can see the distribution of H3[PMo12O40] species in the system D2O-
cyclohexanone (v/v : 5/1) at 55 °C, before and after H2O2 addition. The curves indicate that
[PMo12O40]3- {-0} transforms to [PMo12O40]5- {-II} that transforms later to [PMo12O40]7-
{-IV}. This sequence is similar to that observed previously in water (Fig. 4). Note that here
again ca. 20% of the initial signal is missing, that is indicative of some non-detected
paramagnetic species. Addition of hydrogen peroxide leads mainly to peroxo species which
decrease slowly with time to produce the starting oxidized POM, [PMo12O40]3- {-0}. This
proves the reversible transformation of Keggin POMs to peroxomolybdate complexes under
22

the catalytic conditions. No reduced species have been observed in these conditions up to 18 h
of reaction, but a quite large amount of undetectable species was measured.
Fig. 12: Phosphorus species distribution based on 31P NMR in a mixture of 300 mg H3[PMo12O40] + 2.5 mL H2O
+ 0.5 mL cyclohexanone at 55 °C over a period of 16.5 h (period I), and after subsequent addition of 0.5 mL
H2O2 over a period of 17.5 h at 55 °C (period II).
3.4 Activation of the substrate
1
The substrate behavior during the catalytic reaction was checked by H NMR analysis in
parallel to 31P NMR. A stack plot of 1H NMR spectra of a solution of 30 mg H3[PMo12O40] in
3 mL cyclohexanone is presented in Fig. 13 showing the spectral evolution with time at 55
°C. The spectra are characterized by the three dominant resonances of cyclohexanone in the
range 2.2-2.8 ppm and a signal of residual water from the catalyst. This latter underwent a
significant high-field shift from ca. 5 to 3.8 ppm when the temperature increases from 27 to
55 °C. Very early just after heating, new resonances appeared and grew with time at 2.0, 2.3,
2.4, 3.3, and 5.8 ppm. The resonance at 5.8 ppm falls in typical ethylenic protons resonance
range and allows assigning the new resonance lines to tautomeric enol-form of
cyclohexanone. Such a tautomerism between cyclohexanone and the corresponding enolic
23

form has been reported and it is suggested that it represents the activation step for the
oxidation of cyclohexanone [33, 61].
Fig. 13: Time-dependent 1H NMR of a mixture of 30 mg H3[PMo12O40] + 3 mL cyclohexanone at 55 °C over a
period of 20 h.
1
No other H resonances than those of keto-enol tautomers of cyclohexanone has been
detected in presence of H3[PMo12O40] over a period of 20 h at 55 °C. Addition of H2O2 had
however led to the appearance of novel signals at 1.2, 1.6, 3.9 and 10 ppm, indicating
products formation of the first oxidation (Fig. 14). In particular, the weak deshielded
resonance at ca. 10 ppm that could represent carboxylic acid protons (observable in organic
phase). This would indicate that formation of adipic acid is initiated only after addition of
hydrogen peroxide, and the POM alone is not able to cleave the C-C bound of the organic
substrate in these conditions.
24

Fig. 14: Time-dependent 1H NMR of a mixture of 30 mg H3[PMo12O40] + 3 mL cyclohexanone at 55 °C over a
period of 20 h, after adding 0.5 mL H2O2 subsequently to a first period of 20 h at 55 °C (shown in Fig. 13).
To further evidence the tautomerism reaction an H/D exchange experiment has been
1
conducted and monitored in situ by H NMR. A mixture constituted of 30 mg H3[PMo12O40]
+ 0.5 mL cyclohexanone + 2.5 D2O is heated at 55 °C over a period of 17.5 h. 1H NMR does
not show the appearance of new resonance lines, but a change in integration of the initial
signals as a result of H/D substitution (Fig. 15).
Fig. 15: Time-dependent 1H NMR of a mixture of 30 mg H3[PMo12O40] + 2.5 mL D2O + 0.5 mL cyclohexanone
at 55 °C over a period of 20 h. Inset: Evolution in signal integration of H1, H2, H3, H2O, and H1+H2O. Note that
this latter is nearly constant during all experimental time.
The initial spectrum showed the three resonance peaks of the cyclohexanone, from low
to high-field H1, H2, and H3 for -, -, and-position respectively (see Fig. 15), as well as
the signal of water at ca. 4.8 ppm. A temperature increase leads to a shift of all the
cyclohexanone signals to a low field domain, a sharp decrease in the signal integration of the
resonance H1 (-position), and simultaneously to an increase of the water signal of water.
The signals intensities of the other protons vary a little. This is a clear indication of H/D
exchange between the deuterium of water and the H1 of cyclohexanone. Such a process needs
necessarily a tautomeric equilibrium to explain the H1 lability in the molecule according to
the process drawn in Scheme 1. The signal loss of H1 reaches below 10% of its initial
integration (Fig. 16). This process occurs also at room temperature but with much slower rate,
i.e., one order on magnitude lower (Figs. S7 and S8). The tautomerism equilibrium should be
catalyzed by acidic protons provided by the POM. Indeed, when the same experiment is
25

conducted in the absence of the POM, i.e., 0.5 mL cyclohexanone + 2.5 D2O (Fig. S9), no
H/D exchange reaction occurs throughout the period of 18 h at 55 °C. This result clearly
indicates the need of the POM presence as acidic proton source to activate the substrate.
H
D
O
O
O
O
H
D
D2O
Scheme 1: Ketonic-enolic tautomerism in cyclohexanone and regioselective H/D exchange.
Fig. 16: Left: Selected 1H NMR spectra of cyclohexanone in the mixture of 30 mg H3[PMo12O40] + 2.5 mL D2O
+ 0.5 mL cyclohexanone. The number on the top of each signal indicates the relative signal area. Right:
Evolution of H1 signal integration as a function of time at 55 °C.
3.5 Mechanism of cyclohexanone activation and oxidative conversion
Based on current obtained results, a proposed mechanism of the cyclohexanone activation
and conversion in presence of water and H2O2, after the third addition of H2O2, is shown in
Scheme 2. The first step consists of activation of the cyclohexanone molecule through
ketonic-enolic tautomerism and the enolic form undergoes readily an oxidation by a
peroxomolybdate complex, leading to 2-hydroxy cyclohexanone. The tautomerism
equilibrium is accelerated by the catalytic acid action of the heteropolyacid, H3[PMo12O40],
that facilitates the subsequent oxidation reaction through an epoxy derivative intermediate.
The peroxo catalyst is regenerated in its most stable form by the hydrogen peroxide. As it has
been shown by our in situ NMR experiments, peroxomolybdate species are formed in situ
during decomposition of the Keggin [PMo12O40]3- by H2O2 in aqueous solution, and had also
26

been suggested as active species in previous studies [62]. In the last steps, intermediate
species would undergo further oxidation to probably lead to adipic anhydride, which can be
readily hydrolyzed. Similar mechanism has also been proposed recently [63]. Further
investigations are needed to determine the nature of the intermediates and to reveal the precise
structure of the peroxo-type catalyst. Unfortunately, in contrast to aqueous solutions, 95Mo
NMR failed in cyclohexanone media and no signal could be measured probably due to severe
linewidth broadening. Also, 17O NMR showed only the signal of cyclohexanone ( = 560
ppm), while the catalyst was difficult to detect because it was too dilute.
H
H
O
O
O
O
O
Mo-peroxo/H2O2
H
"H+"
OH
HO
O
O
O
O
O
O
O
O
O
O
O
O
O
H2O2
- H2O
Mo
Mo
Mo
OH2
O
O
OH2
OH2
X
X
X
Mo-peroxo
X = H2O or PO4(Mo(O2)2)n
Scheme 2: Proposed mechanism of oxidative conversion of cyclohexanone by peroxomolybdate species. The
formal charge on the peroxo Mo complex is omitted for simplicity.
4. Conclusion
In this report, the catalytic behavior of the Keggin H3[PMo12O40] in oxidation of
cyclohexanone by H2O2 has been investigated in situ by multinuclear NMR spectroscopy.
Time-dependent NMR spectra of the working catalyst were recorded at moderate temperature
in order to monitor the early stages of the catalytic processes. In the absence of oxidant agent,
i.e., H2O2, the POM is reduced to -[PMo12O40]5- first then to -[PMo12O40]7- and finally
isomerizes into -[PMo12O40]7- only in presence of water at 55 °C. During these processes
cyclohexanone did not undergo substantial oxidation, but was subjected to keton-enol
tautomerisation catalyzed by the Brønsted acidity of the POM. This tautomeric equilibrium
can be seen as a key step for activation of the substrate for oxidation reaction. Under the
effect of hydrogen peroxide the Keggin structure decomposes progressively into lacunary
27

species first, and then to peroxo-type molybdenum complexes, including
peroxophosphomolydate PMon complexes (n = 3, 4). These species are responsible of the
catalytic oxidative conversion of cyclohexanone into carboxylic acids. A three-step
mechanism is proposed to explain the successive oxidation of the cyclohexanone by the
peroxomolybdates species. The current studies provided some useful insights about the
catalyst states change during the process, but further investigation is still needed to provide
the complete scenario in particular concerning the non-observed NMR fraction of
paramagnetic species.
Acknowledgments
This research is supported by the University of Versailles St-Quentin en Yvelines and the
Centre National de la Recherche Scientifique.
References
[1]
[2]
[3]
M. Freemantle, Chem. Eng. News 78 (2000) 7.
J.A. Glaser, Clean Technol. Environ. Policy 18 (2016) 983.
N. d'Alessandro, L. Liberatore, L. Tonucci, A. Morvillo, M. Bressan, New J. Chem.
25 (2001) 1319.
[4]
N.V. Maksimchuk, K.A. Kovalenko, V.P. Fedin, O.A. Kholdeeva, Chem. Commun.
48 (2012) 6812.
[5]
[6]
[7]
[8]
[9]
R. Noyori, M. Aoki, K. Sato, Chem. Commun. (2003) 1977.
I.Q. Penate, G. Lesage, P. Cognet, M. Poux, Chem. Eng. J. 200 (2012) 357.
B.M. Trost, Angew. Chem.-Int. Edit. Engl. 34 (1995) 259.
M. Vafaeezadeh, M.M. Hashemi, Chem. Eng. J. 221 (2013) 254.
Y.Q. Wen, X.Y. Wang, H.J. Wei, B.J. Li, P. Jin, L.M. Li, Green Chem. 14 (2012)
2868.
[10] J.M. Campos-Martin, G. Blanco-Brieva, J.L.G. Fierro, Angew. Chem.-Int. Edit. 45
(2006) 6962.
[11] A. Castellan, J.C.J. Bart, S. Cavallaro, Catal. Today 9 (1991) 255.
[12] A. Rahman, M. Mupa, C. Mahamadi, Catal. Lett. 146 (2016) 788.
[13] S. Van de Vyver, Y. Roman-Leshkov, Catal. Sci. Technol. 3 (2013) 1465.
[14] A. Castellan, J.C.J. Bart, S. Cavallaro, Catal. Today 9 (1991) 285.
[15] A.V. Leont'ev, O.A. Fomicheva, M.V. Proskurnina, N.S. Zefirov, Uspekhi Khimii 70
(2001) 107.
[16] J. Perez-Ramirez, F. Kapteijn, K. Schoffel, J.A. Moulijn, Appl. Catal. B-Environ. 44
(2003) 117.
[17] A. Banerjee, B.S. Bassil, G.V. Roschenthaler, U. Kortz, Chem. Soc. Rev. 41 (2012)
7590.
[18] X.J. Feng, Y.G. Li, Z.M. Zhang, E. Wang, Acta Chim. Sin. 71 (2013) 1575.
[19] K. Nomiya, Y. Sakai, S. Matsunaga, Eur. J. Inorg. Chem. (2011) 179.
[20] A. Proust, B. Matt, R. Villanneau, G. Guillemot, P. Gouzerh, G. Izzet, Chem. Soc.
Rev. 41 (2012) 7605.
[21] D.E. Katsoulis, Chem. Rev. 98 (1998) 359.
28

[22] K. Nomiya, K. Ohta, Y. Sakai, T. Hosoya, A. Ohtake, A. Takakura, S. Matsunaga,
Bull. Chem. Soc. Jpn. 86 (2013) 800.
[23] J. Alcaniz-Monge, G. Trautwein, A. Garcia-Garcia, J. Mol. Catal. A-Chem. 394
(2014) 211.
[24] S. Benadji, T. Mazari, L. Dermeche, N. Salhi, E. Cadot, C. Rabia, Catal. Lett. 143
(2013) 749.
[25] S.B. Jing, Z.L. Wang, W.C. Zhu, J.Q. Guan, G.J. Wang, React. Kinet. Catal. Lett. 89
(2006) 55.
[26] H.Y. Lu, W.Z. Ren, P.F. Liu, S.X. Qi, W.H. Wang, Y.M. Feng, F.X. Sun, Y.W.
Wang, Appl. Catal. A-Gen. 441 (2012) 136.
[27] M. Vazylyev, D. Sloboda-Rozner, A. Haimov, G. Maayan, R. Neumann, Top. Catal.
34 (2005) 93.
[28] B.G. Donoeva, T.A. Trubitsina, G.M. Maksimov, R.I. Maksimovskaya, O.A.
Kholdeeva, Eur. J. Inorg. Chem. (2009) 5142.
[29] O.A. Kholdeeva, R.I. Maksimovskaya, J. Mol. Catal. A-Chem. 262 (2007) 7.
[30] A. Lesbani, Sumiati, Mardiyanto, N.A. Fithri, R. Mohadi, Makara J. Sci. 19 (2015) 85.
[31] J.L. She, Z.H. Fu, J.W. Li, B. Zeng, S.P. Tang, W.F. Wu, H.H. Zhao, D.L. Yin, S.R.
Kirk, Appl. Catal. B-Environ. 182 (2016) 392.
[32] J.M. Bregeault, M. Vennat, L. Salles, J.Y. Piquemal, Y. Mahha, E. Briot, P.C. Bakala,
A. Atlamsani, R. Thouvenot, J. Mol. Catal. A-Chem. 250 (2006) 177.
[33] F. Cavani, L. Ferroni, A. Frattini, C. Lucarelli, A. Mazzini, K. Raabova, S. Alini, P.
Accorinti, P. Babini, Appl. Catal. A-Gen. 391 (2011) 118.
[34] R. Neumann, A.M. Khenkin, Chem. Commun. (2006) 2529.
[35] S.A. Chavan, D. Srinivas, P. Ratnasamyi, J. Catal. 212 (2002) 39.
[36] M. Haouas, J. Trébosc, C. Roch-Marchal, E. Cadot, F. Taulelle, C. Martineau-Corcos,
Magn. Reson. Chem. 55 (2017) 902.
[37] H. Shi, J.A. Lercher, X.Y. Yu, Catal. Sci. Technol. 5 (2015) 3035.
[38] T. Blasco, Chem. Soc. Rev. 39 (2010) 4685.
[39] Ivanova, II, Y.G. Kolyagin, Chem. Soc. Rev. 39 (2010) 5018.
[40] L. Zhang, Y.H. Ren, B. Yue, H.Y. He, Chem. Commun. 48 (2012) 2370.
[41] S. Walspurger, A. Goeppert, M. Haouas, J. Sommer, New J. Chem. 28 (2004) 266.
[42] C. Ammann, P. Meier, A.E. Merbach, J. Magn. Reson. 46 (1982) 319.
[43] P. Courtin, Rev. Chim. Min. 8 (1971) 75.
[44] A. Tahar, S. Benadji, T. Mazari, L. Dermeche, C. Marchal-Roch, C. Rabia, Catal. Lett.
145 (2015) 569.
[45] K. Nomiya, M. Miwa, Y. Sugaya, Polyhedron 3 (1984) 607.
[46] E. Ishikawa, T. Yamase, Bull. Chem. Soc. Jpn. 73 (2000) 641.
[47] R.I. Maksimovskaya, Polyhedron 65 (2013) 54.
[48] A.C. Dengel, W.P. Griffith, B.C. Parkin, J. Chem. Soc.-Dalton Trans. (1993) 2683.
[49] D.C. Duncan, R.C. Chambers, E. Hecht, C.L. Hill, J. Am. Chem. Soc. 117 (1995) 681.
[50] I.D. Ivanchikova, N.V. Maksimchuk, R.I. Maksimovskaya, G.M. Maksimov, O.A.
Kholdeeva, ACS Catal. 4 (2014) 2706.
[51] L.I. Kuznetsova, N.I. Kuznetsova, R.I. Maksimovskaya, O.S. Koshcheeva, V.A.
Utkin, Kinet. Catal. 54 (2013) 420.
[52] L. Salles, C. Aubry, R. Thouvenot, F. Robert, C. Doremieuxmorin, G. Chottard, H.
Ledon, Y. Jeannin, J.M. Bregeault, Inorg. Chem. 33 (1994) 871.
[53] L.I. Kuznetsova, N.I. Kuznetsova, R.I. Maksimovskaya, G.I. Aleshina, O.S.
Koscheeva, V.A. Utkin, Catal. Lett. 141 (2011) 1442.
29