212
J. Alca˜niz-Monge et al. / Journal of Molecular Catalysis A: Chemical 394 (2014) 211–216
Following these results, several polyoxometalates have been used
as effective catalysts for homogeneous peroxidation with hydro-
gen peroxide [8–10]. It must be pointed out that, for this reaction,
polyoxometalates act as catalyst precursors: monomeric, dimeric,
and tetrameric peroxo-type species are generated by the reaction
of the precursor with hydrogen peroxide, being those the catalysts
for the reaction of epoxidation.
Then, it was washed several times with an acidic water solution
until reaching pH 3. Subsequently, the catalyst was dried in a vac-
uum oven at 100 ◦C for 24 h. Finally, prior to its reuse in a new
cycle, it was weighted. The loss weights were used to estimate the
amount of catalyst solubilised during the reaction. In addition, for
Cs-containing heteropolysalts, the importance of a previous step of
calcination in air at 375 ◦C (1 h) was analysed.
Regarding the adipic acid production, Noyori et al. [4] have
obtained excellent results in the cyclohexene conversion to adipic
acid, under homogeneous phase, replacing nitric acid by hydrogen
peroxide as oxidant agent and using sodium tungstate (Na2WO4)
as catalyst. Based on these results, attempts have been made of dis-
Unfortunately, the catalytic tests indicate that only a small portion
of the acidic and redox sites available are active for the reaction,
and some leaching (13–14%) or surface coverage of the catalysts
after their use takes place [11].
With these premises, the objective of the present work is to ana-
lyse the uselfulness of heteropolysalts that are active towards the
conversion of cyclohexene to adipic acid under the experimental
conditions mentioned above, for which they should be insoluble in
the biphasic reaction medium.
2.2. Characterisation
The elemental analyses of POM, the fresh, calcined and used
catalysts are listed in Table S1. The results obtained show that the
elemental contentss are closed to the expected theoretical values
on all POM. The experimental P/Mo and P/W ratios are in agreement
with the exclusive presence of the Kegging compound, indicat-
ing that no significant decomposition takes place either during the
calcination, or during the oxidation process.
The porous texture of the catalysts was analysed by physi-
cal adsorption of gases (N2 at −196 ◦C and CO2 at 0 ◦C), using
two different volumetric equipments (Autosorb-B and Autosorb-
6, respectively). Previously, the samples were degasified for 4 h at
150 ◦C.
The identification of the reaction products of the oxidation of
cyclohexene into adipic acid was performed by taking an aliquot of
the solution (20 L) at the end of the reaction time and analysing
by a GC (Agilent 6890 provided with an HP-1 column of size
30 m × 250 m × 0.25 m) coupled to a mass spectrometer Agilent
5975 MS. Once identified all the reaction products, the conver-
sion and the evolution of the reaction products were quantitatively
estimated by a FID-GC (Agilent 6890), after proper calibration pro-
cedures. Regarding the amount of hydrogen peroxide decomposed,
it was calculated from the subtraction of initial amount set of the
reacted and the non-reacted quantities.
With the aim of determining if the structural or chemical prop-
erties of the catalysts were modified after the reaction, the fresh,
calcined and used catalysts were analysed by different techniques,
i.e. FT-Raman and UV–vis-spectroscopy. It is important to outline
that other characterisation techniques such as: elemental anal-
ysis, XRD, SEM, TEM, DRIFT and TG techniques were also used,
although these did not provide any significant differences among
the fresh, calcined and used catalysts (the results obtained with
the above mentioned techniques are summarised in the Supple-
mentary Information). FT-Raman spectra were carried out with a
FT-Raman (Bruker RFS/100) spectrophotometer using a laser with
an excitation source (1064 nm, Nd-YAG). The UV–vis spectra were
conducted using a UV–vis/NIR spectrophotometer (Jasco V-670),
ranging wavelength values from 190 to 600 nm.
2. Experimental
2.1. Materials and methods
The catalysts used in this work, based on Keggin’s structures,
−3
−3
anions, POM), were prepared in
([PMo12O40
]
and [PW12O40
]
amine cations [13]. Due to their solubility in the reaction medium,
heteropolyacids were supported onto an activated carbon fiber
A20 (PMo/A20) whose preparation procedure is described in previ-
but replacing one of the addenda atoms (M = W or Mo) by other
−5
transition metal (M = W or Mo): [FePM11O39
]
−5, [CoPM11O39
following the procedures found in
]
,
−5
] ]
−5, [PM10V2O40
[NiPM11O39
the literature [15–17].
The oxidation of cyclohexene was conducted in a glass reactor
(27 cm3 of inner volume) located inside of a stainless steel vessel
(45 cm3 of inner volume) coupled to a manometer to monitor the
system pressure. For the reaction, 1 cm3 of cyclohexene, 5 cm3 of
hydrogen peroxide (30%), 1 cm3 of solvent (acetonitrile), 0.2 cm3
of anhydrous acetic acid and a certain amount of catalyst, ranging
from 10 to 120 mg, were mixed. A small magnet of teflon was added
to the mixture and the whole system was immersed in a polyeth-
ylene glycol thermostatic-bath, at different temperatures, ranging
from 40 to 90 ◦C. Preliminary results confirmed that 75 ◦C and a
stirring speed of 800 rpm were the most suitable conditions. A reac-
tion time of 6 h was selected as the optimum value, thus allowing
complete conversion but minimising hydrogen peroxide decompo-
sition. In addition, other reaction times, both shorter (0.5–3 h) and
longer (until 24 h), were tested for comparative purposes.
solvents such as: methanol, anhydride acetic, toluene, dimethyl
sulfoxide, acetonitrile. The solvent plays a role both and assisting
POM stability towards its solvolysis against the hydrogen perox-
ide [18]. Complementary, acetic acid was added to the reaction
medium as a source of protons, which seems to be essential for
the oxidation reaction based on different authors’ studies [19–22].
The reutilisation of the catalysts was approached, using the
catalyst in a consecutive reaction. For that, upon completion of
reaction time, the mixture was filtered, recovering the catalyst.
3. Discussion results
3.1. Study of the efficiency of the catalyst towards H2O2
consumption
Although the air is the most economic oxidant agent, for the
reaction under study is not efficient enough. As already mentioned
in the introduction, H2O2 is a clean and environmentally-friendly
oxidant (only generates H2O as by-product) and it is relatively
cheap. However, it presents the disadvantage of its instability
against decomposition. Many substances catalyse its decomposi-
tion, including most of the transition metals and their compounds.
For this reason, it is important to pay attention to this issue. Not
only because of economic reasons (not all the amount of oxidant
takes part into the reaction) but also because of safety reasons (O2
generated due to hydrogen peroxide decomposition can originate
overpressure inside the reactor). As far as these authors concern,