Clean alternative for adipic acid synthesis via liquid-phase oxidation of cyclohexanone and cyclohexanol over H3-2xCoxPMo 12O40 catalysts with hydrogen peroxide

Article abstract of DOI:10.1007/s10562-013-1025-3

Adipic acid (AA) was synthesized from the oxidation of cyclohexanol (-ol), cyclohexanone (-one) and cyclohexanol/cyclohexanone (-ol/-one) mixture over a series of H_3-2xCo_xPMo₁₂O₄₀ (x: 0.0-1.5) Keggin-type polyoxometalates as catalysts in the presence of hydrogen peroxide (30 %) under mild conditions (90 C, 20 h) and in solvent free. The heteropolysalts were prepared from H₃PMo₁₂O₄₀ and Ba(OH)₂ using the cationic exchange method and characterized by Infrared and diffuse reflectance UV-Vis spectroscopies, X-ray diffraction and TG analyses. The catalytic tests showed that salts are more efficient than the corresponding heteropolyacid and among the H_3-2xCo _xPMo₁₂O₄₀ (x: 0.25-1.5) series, H ₁Co₁PMo₁₂ exhibits the highest AA yield for whole substrates, with ca. 76, 53 and 43 % from oxidation of -one, -ol and -ol/-one, respectively.

Full text of DOI:10.1007/s10562-013-1025-3

Catal Lett (2013) 143:749–755
DOI 10.1007/s10562-013-1025-3
Clean Alternative for Adipic Acid Synthesis Via Liquid-Phase
Oxidation of Cyclohexanone and Cyclohexanol Over
H322xCoxPMo12O40 Catalysts with Hydrogen Peroxide
•
S. Benadji T. Mazari L. Dermeche
•
•
•
N. Salhi E. Cadot C. Rabia
•
Received: 27 February 2013 / Accepted: 11 May 2013 / Published online: 24 May 2013
Ó Springer Science+Business Media New York 2013
Abstract Adipic acid (AA) was synthesized from the
oxidation of cyclohexanol (-ol), cyclohexanone (-one) and
cyclohexanol/cyclohexanone (-ol/-one) mixture over a series
of H3-2xCoxPMo12O40 (x: 0.0–1.5) Keggin-type polyoxo-
metalates as catalysts in the presence of hydrogen peroxide
(30 %) under mild conditions (90 °C, 20 h) and in solvent
free. The heteropolysalts were prepared from H3PMo12O40
and Ba(OH)2 using the cationic exchange method and char-
acterized by Infrared and diffuse reflectance UV–Vis spec-
troscopies, X-ray diffraction and TG analyses. The catalytic
tests showed that salts are more efficient than the corre-
sponding heteropolyacid and among the H3-2xCoxPMo12O40
(x: 0.25–1.5) series, H1Co1PMo12 exhibits the highest AA
yield for whole substrates, with ca. 76, 53 and 43 % from
oxidation of -one, -ol and -ol/-one, respectively.
corresponds either to cyclohexane oxidation in the presence
of air to a cyclohexanone/cyclohexanol mixture, phenol
hydrogenation to cyclohexanol in the presence of molecular
hydrogen or cyclohexene hydration to cyclohexanol (Asahi
process). Cyclohexane, phenol or cyclohexene can be
obtained from oxidation or hydrogenation of benzene. In the
second step, the cyclohexanone/cyclohexanol mixture or
cyclohexanol is oxidized to AA in the presence of nitric acid
as oxidant and Cu–V–O as catalyst. In addition to AA, there
is formation of succinic, glutaric and oxalic acids as
by-products and nitrogen oxides gases (NOx) as co-prod-
ucts. Among these gases, N2O constitutes the major source
of polluting species.
Currently, the development of a green protocol, as
alternative non-HNO3 pathway for the AA synthesis seems
to be the objective of the research. Many attempts have
been made to substitute the traditional process, using nitric
acid, by catalytic systems more friendly using oxidants,
respectful of environment, such as air, molecular oxygen or
hydrogen peroxide.
Keywords Keggin-type polyoxometalates ꢀ
Cyclohexanol ꢀ Cyclohexanone ꢀ Hydrogen peroxide ꢀ
Adipic acid
The AA synthesis from cyclohexane using only one eco-
friendly and cost-effective step was examined by several
authors. With a radical catalyst as N-hydroxyphthalimide
(NHPI) combined with Mn(acac)2 or Co(OAc)2 and
molecular oxygen as oxidant, the best catalytic performance
was obtained in the presence of the cobalt (73 % of AA
selectivity at 73 % of cyclohexane conversion) [8]. Bonnet
et al. have proposed a method for the AA production by air
oxidation of cyclohexane, using different lipophilic car-
boxylic acids (3,5-di-tert-butylbenzoic, 4-tertbutylbenzoic
or 4-trifluoromethylbenzoic acids) as catalytic systems and
metallic precursors (cobalt, manganese) as co-catalysts. In
this case also, the best catalytic results were obtained with
the cobalt element (*70 % of AA selectivity) [9]. Among
tested Anderson-type polyoxometalates (POMs) as catalysts
1 Introduction
Adipic acid (AA) is the main precursor for nylon-6,6, its
production involves two steps [1–7]. The first one
S. Benadji ꢀ T. Mazari ꢀ L. Dermeche ꢀ N. Salhi ꢀ C. Rabia (&)
´
Laboratoire de Chimie du Gaz Naturel, Faculte de Chimie,
Universite des Sciences et de la Technologie Houari
´
`
Boumediene (USTHB), BP 32, El-Alia, 16111 Bab-Ezzouar,
Algeria
e-mail: c_rabia@yahoo.fr; crabia@usthb.dz
E. Cadot
ILV-UMR 8180 CNRS, Universite de Versailles Saint-Quentin-
´
en-Yvelines, Batiment Lavoisier, 45 Avenue des Etats-Unis,
ˆ
78035 Versailles Cedex, France
123

750
S. Benadji et al.
for aerobic oxidation of cyclohexane to AA, Lu¨ et al. [10]
have found that [(C18H37)2N(CH3)2]6Mo7O24 is the
most effective with 87 % of AA selectivity at 10 % of
conversion.
substitution of protons of the HPA (H3PMo12O40) by a
transition metal could be also more attractive, because the
oxidizing power can be enhanced.
In a previous study [22], we have used Keggin-type
heteropolysalts of formula MxPMo12O40 (M: Fe, Ni, Co
and x: 1 or 1.5) as catalysts in the AA synthesis from
liquid-phase oxidation of cyclohexanol (-ol), cyclohexa-
none (-one) or a cyclohexanol/cyclohexanone mixture
(-ol/-one) in the presence of hydrogen peroxide (30 %) and
in the absence of solvent. It was shown that Co-based
catalyst exhibits the highest AA yield for whole substrates
(-ol, -one and -ol/-one). In the present work, the AA
synthesis using the same conditions will be studied over
a series of soluble polyoxometalates of formula H3-2xCox
PMo12O40, noted CoxPMo12, (x: 0.0–1.5), in order to
examine the effect of x value on AA yields.
The hydrogen peroxide as an oxidizing agent was used
for the AA synthesis from cyclohexene oxidation on
materials based on tungsten [11] and peroxotungstates
peroxomolybdates [12], mesoporous silicates doped with
titanium [13–15] and based phosphomolybdic and phos-
photungstic heteropolyacids (HPAs) complexes [16] as
catalysts. High AA yields have been obtained with these
processes.
The AA synthesis from the cyclohexanone oxidation in
the presence of hydrogen peroxide has also been the sub-
ject of several studies. With tetrabutylammonium salts of
P/W or P/Mo HPAs, it was found that HPAs react cata-
lytically in the presence of aqueous hydrogen peroxide;
whereas, they react only stoichiometrically without it [17].
Since hydrogen peroxide convert the reduced-form (blue)
of the HPA produced by the substrate oxidation, to the
oxidized-form (yellow). H2O2 plays the role of a promoter
or a reoxidizing reagent for the HPA after substrate oxi-
dation. The AA production by the liquid-phase oxidation of
cyclohexanone with molecular oxygen was examined in
presence of Mn(OAc)2/Co(OAc)2 catalysts [18] and l3-
Oxo-bridged Co/Mn cluster complexes, CoMn2(O) [19], in
acetic medium. High catalytic activity and AA selectivity
were observed with these catalytic systems. With HPAs of
composition H3?xPMo12-xVxO40 (x: 1, 2), it was found
that the use of acetic acid as co-solvent led to a strongly
enhanced cyclohexanone conversion with respect to the
reaction carried out with water-only solvent [20]. It was
also reported in this study that the cyclohexanone oxidation
proceeds via a redox mechanism, in presence of water-only
solvent and a radical-chain autoxidation mechanism pre-
vailed, in presence of acetic acid as co-solvent. With H2O2
as oxidant and Dawson-type heteropolytungstate as cata-
lyst, the AA yield could reach 14.0 % from the cyclohex-
anone oxidation in absence of organic solvents, acidic
additives and phase transfer agents [21].
2 Experimental
2.1 Catalysts Preparation
H3PMo12O40 (noted Co0PMo12) was prepared in accord with
classic method [23] and
H3-2xCoxPMo12O40 (noted
CoxPMo12, with x = 0.25–1.5) heteropolysalts were synthe-
sized intwo steps. Inthe first step, BaxPMo12O40 was prepared
byadditionofBa(OH)2 toanaqueoussolutionofH3PMo12O40
in stoichiometric ratios (according to the Eq. (1)) and in the
second step BaSO4 was precipitated by cobalt sulfate via
cationic exchange between Ba2? and Co2? (according to
Eq. (2)). After filtration, the resulting heteropolysalt solution
was evaporated to dryness at 50 °C under vacuum.
H3PMo12O40 þ x BaðOHÞ2! H3ꢁ2xBaxPMo12O40
þ 2x H2O
ð1Þ
ð2Þ
H3ꢁ2xBaxPMo12O40 þ xCoSO4
! H3ꢁ2xCoxPMo12O40 þ x BaSO4 #
In the liquid phase oxidation processes, the use of
hydrogen peroxide requires an acidic environment that can
come from the organic or inorganic acids. However, the
use of the liquid acids can have a negative impact on the
environment by generating more waste water volume. To
overcome these drawbacks, it would be interesting to
introduce the Keggin-type phosphomolybdic POMs as
catalysts for liquid phase oxidative reactions. They have
the advantage of possessing acid and oxidative properties
that can be controlled according to the constituent elements
and the requirements of the reaction. Thus, the acidity as
well as the oxidizing power needed for the reactions could
come from the POMs only. On the other hand, the partial
2.2 Characterization
Infrared spectroscopy of KBr pellets was recorded on a
PERKIN-ELMER (Spectrum One) apparatus. The resolu-
tion was fixed at 2 cm-1 and the number of scans at 32.
Before analysis, the diluted solids (1.5 mg) in KBr (150 mg)
were pressed into pellets at 4 tons/cm2. Diffuse reflectance
UV–Vis spectra were recorded at room temperature using
two apparatus: (1) a Perkin-Elmer Lambda19 spectrometer,
equipped with a 60 mm integrating sphere coated with
barium sulphate, the spectra were recorded with a sweep
speed in wavelength of 240 nm/min, between 170 and
800 nm; and (2) a Jasco V-650 spectrophotometer, with a
123

Clean Alternative for Adipic Acid Synthesis
751
sweep speed in wavelength of 200 nm/min, between 190
and 900 nm. X-ray diffraction patterns (XRD) were recor-
ded at room temperature using a Siemens D5000 diffrac-
tometer equipped with a copper anticathode Cu Ka
Figure 2 shows the UV–Vis spectra of H3PMo12O40 and
CoxPMo12 salt. The broad charge transfer band from oxy-
gen to Mo(VI) appears between 250 and 500 nm. It was
reported that this is the only ligand–metal charge transfer
band that is observed in the case of POMs [25–27]. In
presence of CoxPMo12, another band of lower intensity is
observed between 500 and 600 nm, in particular for
x C 0.75. For the other x values, the cobalt content in the
POM don’t exceed 2 %, therefore it is very difficult to
observe this charge transfer band. This latter can be
attributed either to d–d transition band of d1 Mo(V) species
in octahedral coordination and/or to d6 Co(III) species in
octahedral coordination [19]. Same observations have been
made by Mazari et al. [28, 29] in the presence of Co based
salts of P/Mo HPA and by Dermeche et al. [26] in the
presence of Sb or Sn based salts of P/Mo HPA. The exis-
tence of this band can be explained by a charge transfer
between molybdenum and cobalt cations according to the
following equation:
˚
(k = 1.5418 A), the indexing of the diffraction lines and
phase identification were performed using the EVA soft-
ware including a ICDD-JCPDS database. TGA analysis was
performed on a TGA 2050, 626 Polarecord Metrohm
apparatus with a rate of heat of 5 °C/min from 25 to 600 °C
(to prevent the MoO3 sublimation) under nitrogen.
2.3 Catalysis
The synthesis method is based on that described in the
literature [17]. The liquid-phase oxidation of cyclohexanol
(-ol), cyclohexanone (-one) or mixture of -ol and -one was
carried out at 90 °C, using a 100 mL round-bottomed flask
equipped with a magnetic stirring bar and a reflux con-
denser. The whole is stirred at 800 rpm. The reaction
mixture is constituted by a calculated amount of catalyst
and substrate (15 mmol); where into one drop of hydrogen
peroxide (30 %) was added whenever the polyoxometalate
showed a color change (from yellow or light green to blue).
The reaction time was determined to be a time in that an
appropriate amount of hydrogen peroxide in the dropping
funnel had been consumed that corresponds to oxidation of
substrate. The reaction time was found to be of 20 h. It is
noteworthy that beyond 90 °C, hydrogen peroxide may
decompose. The resultant homogeneous mixture was
cooled at 0 °C overnight. AA, one of oxidation products,
was isolated as white crystals and identified by FT-IR
spectroscopy and melting point (*151 °C). The AA yield
is given by the following relationship: R % = AA mass
formed/theoretical AA mass.
MoðVIÞ þ CoðIIÞ ! MoðVÞ þ CoðIIIÞ
Figure 3 shows that the XRD patterns of CoxPMo12 are
similar to that of the parent acid, indicating that the salts
crystallize in a triclinic structure as in the case of the acid.
The partial or total substitution of the protons of H3PMo12
by Co2? ion does not affect the crystal system of the POM.
Thermal analysis (TGA) was used in our case to deter-
mine the x value. The departure of water molecules of
constitution resulting from the combination of protons with
oxygen atoms of the solid is observed in the 300–500 °C
range (figures not shown). The stoichiometric coefficients
derived from TGA are close to those calculated theoreti-
cally (Table 1). These results confirm the reliability of the
synthesis method (cation exchange) used in this work.
3.2 Catalytic Tests
3 Results and Discussion
The catalytic properties of soluble Keggin type salts,
CoxPMo12, together with that of Co0PMo12, were exam-
ined in the liquid-phase oxidation of cyclohexanol (-ol),
cyclohexanone (-one) and mixture of -ol and -one (-ol/-one)
to AA, in the presence of hydrogen peroxide (30 %) under
solvent-free conditions at 90 °C, using 15 mmol of
substrate.
3.1 Catalysts Characterization
The FT-IR characteristic bands of the Keggin structure of
both HPA and CoxPMo12 series were observed in the low
wavenumber region (500–1,100 cm-1) (Fig. 1). According
to Rocchiccioli-Deltcheff et al. [24], the bands at 1,066–
1,065, 964–962, 872–868, 788–782 and 596–594 cm-1 are
assigned to mas(P–Oa), mas(Mo–Od), mas(Mo–Ob–Mo) (vertex
junction), mas(Mo–Oc–Mo) (edges junction) and d(P–O)
vibrations, respectively. The comparison of the FT-IR fre-
quency bands of CoxPMo12 salts, with those of the parent
acid (H3PMo12O40) did not reveal significant difference
within 1,500–250 cm-1 range. This suggests that the Keggin
primary structure remains intact after the partial substitution
of protons of H3PMo12 acid by the Co2? ions.
It is worth noting that the substrate oxidation did not
proceed when the reaction mixture is constituted of sub-
strate, catalyst and hydrogen peroxide simultaneously or
when no catalyst was used. On the other hand, the oxi-
dizing power of POM is generated only after addition of
hydrogen peroxide (transition from blue to yellow). The pH
values of the solution after reaction, tested by pH meter are
around 2 for all catalytic tests, indicating that the reaction
medium is acid.
123

752
S. Benadji et al.
Fig. 1 FT-IR spectra of
Co0PMo12 a, Co1PMo12 b and
Co1.5PMo12 c solids
c
b
a
3850 3650 3450 3250 3050 2850 2650 2450 2250 2050 1850 1650 1450 1250 1050 850
650
450
Wavenumber (cm−1
)
2
Fig. 2 UV–Vis DR spectra of
Co0PMo12 a, Co0.25PMo12 b,
Co0.5PMo12 c, Co0.75PMo12 d,
Co1PMo12 e, Co1.25PMo12 f and
Co1.5PMo12 g materials
1.2
1
a
g
g
f
1.8
1.6
1.4
1.2
1
0.8
0.6
0.4
0.2
0
e
d
c
b
a
200
300
400
500
600
700
800
Wavenumber (nm)
0.8
0.6
0.4
0.2
200
300
400
500
600
700
800
900
Wavenumber (nm)
In this work, only AA formation was taken into account.
The other reaction products observed are the succinic and
glutaric acids.
decreases the AA yield, in particular for x [ 1, from ca. 38 to
ca.32 % for Co1.25PMo12 and from ca. 34 to ca.30 % for
Co1.5PMo12. This decrease of the AA yield is probably
caused by an increase of yields of other acids, due to a
competition between the different acids. As, the highest AA
yield was obtained with a mass of 0.0625 g, the study of the
other parameters will be examined with this mass value.
Figure 4 shows obtained AA yields with CoxPMo12
(x: 0.0–1.5) catalysts from cyclohexanol oxidation. The
cobalt salts are more effective than the corresponding HPA
with AA yields of ca. 23–53 % against ca. 16 %. The effect
of the x value variation on AA production shows that the
AA yield increases from ca. 23 to ca. 53 % when x passes
from 0.25 to 1. Beyond this value (x = 1), it decreases to
ca. 34 %. These results demonstrate the influence of cobalt
3.2.1 Cyclohexanol Oxidation
In order to optimize the reaction conditions, the mass effect
of catalyst on AA yield was examined over the following
POMs: Co1PMo12, Co1.25PMo12 and Co1.5PMo12. The data
of Table 2 show that regardless the x value, an increase of the
catalyst mass from 0.0313 to 0.0625 g leads to an increase in
the AA yield, from ca. 39 to 53 %, ca. 23 to 38 % and ca. 20
to 34 % in the presence of Co1PMo12, Co1.25PMo12 and
Co1.5PMo12, respectively. Whereas, a second increase of
catalyst mass by a factor of 2 (from 0.0625 to 0.1250 g)
123

Clean Alternative for Adipic Acid Synthesis
753
80
70
60
50
40
30
20
10
0
g
f
e
d
c
b
a
0
0.25
0.5
0.75
1
1.25
1.5
x value
Fig. 4 Adipic acid yields as a function of the Co atoms number in
H3-2xCoxPMo12O40 solids in cyclohexanone (blue square) and
cyclohexanol (red square) oxidation (Treact: 90 °C, mcat = 0.0625 g,
n-ol or -one = 15 mmol (V (-ol) = 1.63 mL and V (-one) = 1.58 mL),
agitation rate: 800 rpm, reaction time: 20 h)
5
10
20
30
40
50
60
2 Theta (Degree)
Fig. 3 X-ray patterns of Co0PMo12 a, Co0.25PMo12 b, Co0.5PMo12 c,
Co0.75PMo12 d, Co1PMo12 e, Co1.25PMo12
g materials
and appeared to be independent of the nature of the sub-
strate (Fig. 4). While, with cobalt salts, whatever the x
value, the AA yields obtained from the -one oxidation are
much higher than those obtained from the -ol oxidation
with ca. 32–76 against ca. 23–53 %. These results seem
predictable, since the ketone (-one) is oxidized directly to
the acid products, while the alcohol (-ol) is oxidized ini-
tially to -one and subsequently, to the acid products.
As in the case of the -ol oxidation, it is apparent from
Fig. 4 that the AA yield increases from ca.16 to 76 % when
x rises from 0 to 1. Beyond this value (x = 1), it decreases
until ca. 53 %. The most effective catalyst, in the cyclo-
hexanone oxidation to AA is also Co1PMo12.
f and Co1.5PMo12
Table 1 Theoretical and experimental formulas of CoxPMo12 solids
Theoretical formula
Experimental formula
nH2O*
H3.00Co0.00PMo12O40
H3.08Co0.00PMo12O40
H2.35Co0.32PMo12O40
H1.91Co0.55PMo12O40
H1.45Co0.78PMo12O40
H0.87Co1.07PMo12O40
H0.52Co1.24PMo12O40
H0.04Co1.48PMo12O40
13
12
12
12
12
14
16
H2.50Co0.25PMo12O40
H2.00Co0.50PMo12O40
H1.50Co0.75PMo12O40
H1.00Co1.00PMo12O40
H0.50Co1.25PMo12O40
H0.00Co1.50PMo12O40
* Number of H2O molecules of crystallization
3.2.3 Oxidation of Cyclohexanone/Cyclohexanol Mixture
Table 2 AA yields (%) as a function of catalyst mass in cyclohex-
anol oxidation
It was reported in the literature that AA can be produced
from a mixture constituted of cyclohexanone (50 %) and
cyclohexanol (50 %) [1–5]. Hence, in this work, the
influence of the addition of cyclohexanol (10–50 %) to the
reaction mixture constituted of cyclohexanone and catalyst
on the AA yield was studied. Co1PMo12O40, the most
effective catalyst, was tested in this reaction. The catalytic
activity data are presented in Table 3. In the absence of -ol,
the AA yield is very high (ca. 76 %). In contrast, an
increase of the percentage of -ol from 10 to 50 % in the
reaction mixture caused a decrease of AA yield from ca. 76
to 43 %. The latter value (ca. 43 %) is lower than that
obtained with the -ol alone (ca. 53 %). This observation of
the decrease of AA yield has already been reported by
other authors [19]. This result can be explain by the low
rate of alcohol oxidation to AA compared with that of
the ketone oxidation (-ol ? -one ? AA). After 20 h of
CoxPMo12
Catalyst mass (g)
0.0313
0.0625
0.125
Co1PMo12
38.8
23.3
19.6
52.5
37.9
33.8
51.6
32.1
29.6
Co1.25PMo12
Co1.5PMo12
Reaction conditions: Treact: 90 °C, n-ol = 15 mmol (V (-ol) =
1.63 mL), agitation rate: 800 rpm, reaction time: 20 h
atom number in the POM on AA yields. The highest AA
yield (ca. 53 %) is obtained with Co1PMo12.
3.2.2 Cyclohexanone Oxidation
The AA yields obtained in the presence of the acid form,
Co0PMo12, from the -one or -ol oxidation, are of ca. 16 %
123

754
S. Benadji et al.
with molecular oxygen in presence of metal oxide as cat-
alyst, there is an oxygen atom insertion from the crystal
lattice to organic molecule. So, there is reduction of the
catalyst. Then, its reoxidation will occur by the molecular
oxygen of the gas phase. In liquid-phase oxidation, the
substrate (alcohol or ketone) consumes oxygen that would
come from POM, and then reduced POM is oxidized by
H2O2 according to the following mechanism:
Table 3 Adipic acid yields as a function of the compositon of
cyclohexanone/cyclohexanol mixture over Co1PMo12O40
Substrate
(-one/-ol)
100 / 90 /
10
80 /
20
70 /
30
60 /
40
50 /
50
0 /
100
0
AA yield (%) 75.5
69.5 61.6 57.8 50.2 42.9 52.5
Reaction conditions: Treact: 90 °C, mcat = 0.0625 g, n-olone
15 mmol, agitation rate: 800 rpm, reaction time: 20 h
=
Substrate þ POMox ! oxygenated products þ POMred
POMred þ H2O2 ! ‘‘peroxo-POMox’’ þ H2O
reaction, the amount of AA obtained from the ketone
oxidation is therefore greater than that obtained from the
alcohol.
It was also reported that the presence of H2O2, in addition
to its function to oxidize the POM, lead to the formation of
‘‘peroxo-POMox’’ intermediate species, that would probably
be the active species in the AA formation [17, 30].
It appears also from this study, that the efficiency of
cobalt salts, compared to that of parent acid, in both -ol and
-one oxidation to AA, is probably due to an increase of the
oxidizing power of POM after Co introduction and to the
simultaneous presence of Mo(VI)/Mo(V) and Co(III)/
Co(II) couples as observed by UV spectroscopy. Thus, the
tow oxidative forms, Mo(VI) and Co(III) are involved in
the substrate oxidation.
The obtained results indicate that the presence of cobalt
in the catalyst system increases strongly the AA yield.
Some studies have already reported that cobalt-based sys-
tems are effective in the oxidation of cyclohexanone and
cyclohexanol to AA [8, 9, 18, 19, 22].
In this study, the highest AA yield (ca. 53 % from -ol
oxidation and ca. 76 % from -one oxidation) is obtained with
one proton and one cobalt atom per Keggin unit (H1Co1
PMo12O40). This chemical composition allowed to have an
acidic environment necessary for both reduction of H2O2 and
cleavage of C–C bond of substrate and an oxidative envi-
ronment favoring AA formation, with the oxidation of
terminal carbon atoms. Therefore, a bifunctional catalytic
system, with appropriate acidic and oxidizing properties, is
very important for optimizing AA production from -ol, -one
or -ol/-one mixture, in the presence of hydrogen peroxide and
in the absence of solvent or additive acid. These results are in
agreement with the literature works [11, 18–20], reporting
that the AA synthesis from oxidation of cyclohexene, cyclo-
hexanone or cyclohexanol in presence of H2O2, is not only an
oxidation reaction, but also an acid-catalyzed reaction.
The efficiency of these catalysts is related to the fact that
the reduced blue POM, obtained after the substrate oxi-
dation reaction, can be reused for the same reaction by
addition of one drop of hydrogen peroxide that converted it
immediately to the oxidized form (yellow POM). The
catalytic cycle is finished when the POM catalyst is no
longer reduced, indicating that the substrate was com-
pletely consumed. These observations have already been
noted by Nomiya et al. [17] which have also shown the
reoxidizing effect of hydrogen peroxide on the POM.
The importance of gradual addition of H2O2 after each
reduction of the catalyst by the substrate is confirmed by
preliminary tests. It was shown that the reaction did not
take place when the initial mixture composed of catalyst,
H2O2 and substrate (-ol or -one) was stirred vigorously at
90 °C for 20 h.
4 Conclusion
In this work, a series of POMs of formula H3-2xCox
PMo12O40 (x: 0.0–1.5) was prepared, characterized and
tested in the AA synthesis in the presence of H2O2, under
mild conditions (90 °C, 20 h) and in solvent free.
The study of the physico-chemical characterization of
CoxPMo12 (x = 0.0–1.5) solids showed that the cationic
cobalt position does not disturb the Keggin anion sym-
metry or the triclinic system of H3PMo12O40 acid, what-
ever the number of introduced Co atoms by POM. The UV
spectroscopy evidenced the presence of both Mo(VI)/
Mo(V) and Co(III)/Co(II) couples and the TGA, the proton
number per Keggin unit.
This study showed that cobalt salts are more efficient
than parent acid in the direct catalytic oxidation of cyclo-
hexanone, cyclohexanol and cyclohexanone/cyclohexanol
mixture to AA with hydrogen peroxide, a clean oxidant,
without any solvent. The POMs act as both an acidifying
and an oxidant agent. Using this green method, a ca. 76 %
yield of AA has been achieved with H1Co1PMo12O40 from
cyclohexanone oxidation.
These results suggest that the substrate oxidation in the
presence of hydrogen peroxide would involve a similar
mechanism to that of type Mars-van Krevelen. In this
mechanism, it is admitted that during an alkane oxidation
References
1. Lindsay AF (1954) Spec Suppl Chem Eng Sci 8:78
2. Castellan A, Bart JCJ, Cavallaro S (1991) Catal Today 9:255
123

Clean Alternative for Adipic Acid Synthesis
755
3. Castellan A, Bart JCJ, Cavallaro S (1991) Catal Today 9:285
4. Castellan A, Bart JCJ, Cavallaro S (1991) Catal Today 9:301
5. Kapteijn F, Rodriguez-Mirasol J, Moulijn JA (1996) Appl Catal
B Environ 9:25
6. Davis DD, Kemp DR (1991). In: Kroscwitz JI, Howe-Grant M
(ed) Kirk-Othmer encyclopedia of chemical technology, vol 1,
4th edn. Wiley, New York, p 446
17. Nomiya K, Miwa M, Sugaya Y (1984) Polyhedron 3:607
18. Shimizu A, Tanaka K, Ogawa H, Matsuoka Y, Fujimori M,
Nagamori Y, Hamachi H (2003) B Chem Soc Jpn 76:1993
19. Chavan SA, Srinivas D, Ratnasamy P (2002) J Catal 212:39
20. Cavani F, Ferronia L, Frattini A, Lucarelli C, Mazzinia A,
Raabova K, Alini S, Accorinti P, Babini P (2011) Appl Catal A
Gen 391:118
7. Davis DD (1985). In: Gerhartz W (ed) Ullman’s encyclopedia of
industrial chemistry, vol A1, 5th edn. Wiley, New York p 270
8. Iwahama T, Syojyo K, Sakaguchi S, Ishii Y (1998) Org Process
Res Dev 2:255
21. Chen D, Yang X, Zhou H, He J (2010) Shiyou Huagong/Petro-
chemical Technology 39:656
22. Mazari T, Benadji S, Tahar A, Dermeche L, Rabia C (2013) J
Mat Sci Eng B3:146
9. Bonnet D, Ireland T, Fachea E, Simonato JP (2006) Green Chem
8:556
10. Lu¨ H, Ren W, Liu P, Qi S, Wang W, Feng Y, Sun F, Wang Y
(2012) Appl Catal A Gen 441–442:136
11. Jin P, Zhao Z, Dai Z, Wei D, Tanga M, Wang X (2011) Catal
Today 175:619
12. Zhu W, Li H, He X, Zhang Q, Shu H, Yan Y (2008) Catal
Commun 9:551
13. Lapisardi G, Chiker F, Launay F, Nogier JP, Bonardet JL (2004)
Catal Commun 5:277
14. Lapisardi G, Chiker F, Launay F, Nogier JP, Bonardet JL (2005)
Microporous Mesoporous Mater 78:289
15. Timofeeva MN, Kholdeeva OA, Jhung SH, Chang JS (2008)
Appl Catal A Gen 345:195
16. Ren S, Xie Z, Cao L, Xie X, Qin G, Wang J (2009) Catal
Commun 10:464
23. Tsigdinos GA (1974) Ind Eng Chem Prod Res Dev 13:267274
24. Rocchiccioli-Deltcheff C, Fournier M, Franck R, Thouvenot R
(1983) Inorg Chem 22:207
`
25. Cavani F, Mezzogori R, Pigamo A, Trifiro F, Etienne E (2001)
Catal Today 71:97
26. Dermeche L, Thouvenot R, Hocine S, Rabia C (2009) Inorg Chim
Acta 362:3896
27. Dermeche L, Salhi N, Hocine S, Thouvenot R, Rabia C (2012) J
Mol Catal A: Chem 356:29
28. Mazari T, Roch-Marchal C, Hocine S, Salhi N, Rabia C (2009) J
Nat Gas Chem 18:319
29. Mazari T, Roch-Marchal C, Hocine S, Salhi N, Rabia C (2010) J
Nat Gas Chem 19:54
30. Mizuno M, Nozaki C, Kiyoto I, Misono M (1998) J Am Chem
Soc 120:9267
123