Sustainable Oxidative Cleavage of Vegetable Oils into Diacids by Organo-Modified Molybdenum Oxide Heterogeneous Catalysts

Article abstract of DOI:10.1007/s11746-017-3047-2

Abstract: Exploiting vegetable oils to produce industrially valuable diacids via an eco-friendly process requires an efficient and recyclable catalyst. In this work, a novel catalytic system based on organo-modified molybdenum trioxide was synthesized by a green hydrothermal method in one simple step, using Mo powder as precursor, hydrogen peroxide, and amphiphilic surfactants cetyltrimethylammonium bromide (CTAB) and tetramethylammonium bromide (TMAB) as capping agents. The synthesized materials were first characterized by different techniques including XRD, SEM, TGA, and FT-IR. Interestingly, various morphologies were obtained depending on the nature of the surfactants and synthetic conditions. The synthesized catalysts were employed in oxidative cleavage of oleic acid, the most abundant unsaturated fatty acid, to produce azelaic and pelargonic acids with a benign oxidant, H₂O₂. Excellent catalytic activities resulting in full conversion of initial oleic acid were obtained, particularly for CTAB-capped molybdenum oxide (CTAB/Mo molar ratio of 1:3) that gave 83 and 68% yields of production of azelaic and pelargonic acids, respectively. These are the highest yields that have been obtained for this reaction by heterogeneous catalysts up to now. Moreover, the CTAB-capped catalyst could be conveniently separated from the reaction mixture by simple centrifugation and reused without significant loss of activity up to at least four cycles. Graphical Abstract: [Figure not available: see fulltext.].

Full text of DOI:10.1007/s11746-017-3047-2

J Am Oil Chem Soc
DOI 10.1007/s11746-017-3047-2
ORIGINAL PAPER
Sustainable Oxidative Cleavage of Vegetable Oils into Diacids
by Organo‑Modifed Molybdenum Oxide Heterogeneous Catalysts
Aimé Serge Ello1,2 · Amir Enferadi‑kerenkan1 · Albert Trokourey2 · Trong‑On Do1
Received: 26 June 2017 / Revised: 28 August 2017 / Accepted: 7 September 2017
© AOCS 2017
Abstract Exploiting vegetable oils to produce industrially
valuable diacids via an eco-friendly process requires an ef-
cient and recyclable catalyst. In this work, a novel catalytic
system based on organo-modifed molybdenum trioxide was
synthesized by a green hydrothermal method in one simple
step, using Mo powder as precursor, hydrogen peroxide, and
amphiphilic surfactants cetyltrimethylammonium bromide
(CTAB) and tetramethylammonium bromide (TMAB) as
capping agents. The synthesized materials were frst char-
acterized by diferent techniques including XRD, SEM,
TGA, and FT-IR. Interestingly, various morphologies were
obtained depending on the nature of the surfactants and syn-
thetic conditions. The synthesized catalysts were employed
in oxidative cleavage of oleic acid, the most abundant unsat-
urated fatty acid, to produce azelaic and pelargonic acids
with a benign oxidant, H2O2. Excellent catalytic activi-
ties resulting in full conversion of initial oleic acid were
obtained, particularly for CTAB-capped molybdenum oxide
(CTAB/Mo molar ratio of 1:3) that gave 83 and 68% yields
of production of azelaic and pelargonic acids, respectively.
These are the highest yields that have been obtained for this
reaction by heterogeneous catalysts up to now. Moreover, the
CTAB-capped catalyst could be conveniently separated from
the reaction mixture by simple centrifugation and reused
without signifcant loss of activity up to at least four cycles.
Keywords Oleic acid · Oxidative cleavage · Azelaic
acid · Surfactant · Molybdenum oxide · Heterogeneous
catalysis
Introduction
Biomass-derived feedstock is one of the most promising
candidates that would substitute the crude oil-based fuels
and chemicals. The vast abundance and renewable nature of
vegetable oils have attracted growing interest in both aca-
demic and industrial researches. Triglyceride is the main
component of oleaginous feedstock, i.e. vegetable oils, of
which its unsaturated fatty acids can be chemically modi-
fed, mainly through their double bonds, to be converted into
value-added chemicals. For instance, oleic acid, the most
abundant monounsaturated fatty acid [1], can produce di-
and mono-carboxylic, azelaic and pelargonic acids through
oxidative cleavage reaction, which are valuable materials for
diferent industrial applications like production of polymers,
plasticizers, adhesives, lubricants, cosmetics, herbicides, and
fungicides, etc. [2–6]. Currently, the oxidative cleavage of
oleic acid is performed in industry via ozonolysis. However,
using ozone is not in line with the principles of sustainable
chemistry due to its inevitable hazardous problems. Attempt-
ing to employ a benign oxidant, several catalyst/oxidant sys-
tems, in homogeneous and heterogeneous forms, have been
reported in the literature for oxidative cleavage of unsatu-
rated fatty acids, which have been thoroughly reviewed in
our recent review paper [7].
Electronic supplementary material The online version of this
article (doi:10.1007/s11746-017-3047-2) contains supplementary
material, which is available to authorized users.
Aimé Serge Ello and Amir Enferadi-kerenkan contributed
equally to this work.
* Trong-On Do
trong-on.do@gch.ulaval.ca
1
Department of Chemical Engineering, Laval University,
Quebec G1V 0A6, Canada
2
Laboratoire de Chimie Physique, Université Félix
Houphouët-Boigny de Cocody, 22 bp 582 Abidjan,
Côte d’Ivoire
Vol.:(0123456789)
1 3

J Am Oil Chem Soc
Homogeneous catalysts have generally shown excellent
performances in oxidation of unsaturated fatty acids (e.g.
production yield of azelaic acid up to 82% from oleic acid
by peroxo complex, oxoperoxo (pyridine-2,6-dicarboxy-
lato) molybdenum (VI) hydrate {Mo(O2)[C5H3N (CO2)2]
(H2O)} catalysts [6]), but such catalytic systems have
always been associated with a lack of catalyst recovery. On
the other hand, activities of heterogeneous catalysts, with
recoverability, reported so far, are not as high as homoge-
neous ones [7, 8]. Various strategies have been reported
in the literature to improve the efciency of the reaction,
like reaction in supercritical fuids with dual oxidants and
microwave irradiations and ultrasound-assisted reactions
[9–12]. Works should be encouraged to fnd a suitable
reaction strategy that enjoys both high activity and ef-
cient recovery. Current eforts in this domain can be gener-
ally classifed into two groups. The frst group is involving
diferent transition metal oxide catalysts, which are inher-
ently solid, along with a benign oxidant such as hydrogen
peroxide or sodium hypochlorite, and the second group is
based on heterogenization of highly active homogeneous
catalysts, like polyoxometallates, via immobilization or
solidifcation methods [13–16]. In these two groups, direct
or indirect selective oxidative cleavage mechanisms have
been proposed [17]. Nevertheless, an efcient heterogene-
ous catalytic system that can aford the excellent reaction
yields obtained by the homogeneous catalysts has not been
reported so far.
(99.9%, Alfa Aesar), hydrogen peroxide (30%, Fischer Sci-
entifc), and surfactants; CTAB (98%, Fischer Scientifc) and
TMAB (98%, Aldrich). In a typical synthesis, 1 g of metal-
lic molybdenum powder was added to 10 ml of deionized
water and the mixture was stirred for 30 min at room tem-
perature. Then, the mixture was placed in an ice bath under
a well-ventilated fume hood. To this mixture, an aqueous
solution of hydrogen peroxide was added dropwise under
vigorous stirring until the color of the solution changed to
yellow (attention: the reaction is very exothermic and must
be done in an ice bath under skilled supervision). The deter-
mined amounts of surfactants (CTAB or TMAB) to obtain
surfactant/molybdenum molar ratios of 1:2, 1:3, and 1:4
were added to the yellow solution. The resultant slurry was
refuxed at 100 °C for 4 h. Finally, the solid product was
separated by centrifugation at high speed and dried in air at
70 °C for 10 h.
Characterization Equipment
The physicochemical properties of the materials were char-
acterized by X-ray difraction (XRD), scanning electron
microscopy (SEM), thermogravimetric analysis (TGA), and
Fourier transform infrared spectroscopy (FT-IR). Powder
XRD patterns of the samples were obtained on a Bruker
SMART APEXII X-ray difractometer equipped with a Cu
Kα radiation source (λ = 1.5418 Å) with steps of 0.020 and
step time of 1 s. SEM images were taken on a JEOL 6360
instrument at an accelerating voltage of 3 kV. Materials were
spread on a carbon tape prior to analysis. The FT-IR spec-
tra were recorded on an FTS 45 infrared spectrophotometer
with the KBr pellet technique. TGA was performed in the
temperature range of 30–800 °C with a TGA Q500 V20.13
Build 39 thermogravimetric analyzer at a heating rate of
5 °C min−1 under an air fow or argon fow of 50 ml min−1.
In this work, we have tried to push the proved potential of
molybdenum oxide as an oxidizing solid catalyst [18–20] to
oxidative cleavage reaction of unsaturated fatty acids. Curi-
ously, employing MoO3 heterogeneous catalysts in oxidative
cleavage of oleic acid has not been reported so far, to our
knowledge. Herein, a series of molybdenum oxides were syn-
thesized via simple oxidative dissolution of Mo powder in
H2O2, with and without organic surfactants. Two well-known
quaternary ammonium surfactants, cetyltrimethylammo-
nium bromide (CTAB) and tetramethylammonium bromide
(TMAB) with diferent alkyl chain lengths were employed as
capping agents. Efects of CTAB and TMAB with diferent
concentrations in the synthesis medium on physicochemi-
cal properties and morphology of the products, as well as
enhancement of catalytic efciency in the liquid phase oxi-
dative cleavage of oleic acid by organo-modifcation of the
surface of molybdenum oxide have been investigated.
Catalytic Test
Catalytic reactions were carried out in a glass reactor
equipped with an oil bath, magnetic stirrer, and refux con-
denser. Typically, the reactor was charged with a suspension
of 0.4 g catalyst and 4 ml aqueous H2O2. Then, 1 g oleic acid
(≥99%, Sigma-Aldrich) was added dropwise under stirring
followed by addition of 8 ml of tert-butanol as solvent. The
reaction mixture was heated to 85 °C and kept at this tem-
perature for a reaction time of 3 h 30 min. After the reaction,
the solution was allowed to cool to room temperature. Dur-
ing the reaction, the system showed homogeneous catalysis
properties, while upon cooling at the end of reaction, the
catalyst particles were precipitated, enabling self-separation.
The precipitated catalyst was separated using a centrifuge
at 8000 rpm and recovered via washing with ethanol and
water several times and drying at 70 °C for 4 h, in order to be
Experimental
Synthetic Details
Surfactant-capped molybdenum oxide particles were synthe-
sized by hydrothermal method using molybdenum powder
1 3

J Am Oil Chem Soc
used in the next catalytic cycle. The resultant solution then
underwent a derivatization process prior to gas chromatog-
raphy–mass spectrometry (GC–MS) analysis.
using any surfactant, denoted as MO (Fig. 1a), confrmed
that it was composed of MoO3–H2O, in accordance with
PDF no. 26-1449 of the ICDD library of spectra. The pattern
exhibits fve well-resolved peaks indexed to the (010), (100),
(120), (020), and (030) planes of triclinic molybdenum tri-
oxide hydrate material, and no peaks corresponding to other
phases implies the single triclinic phase of this material.
Presence of surfactants CTAB and TMAB in the syn-
thesis medium dramatically influenced the crystalline
phase of molybdenum oxide. Figure 1b, c show XRD pat-
terns of the samples prepared in the presence of these two
surfactants with the surfactant/molybdenum molar ratio
of 1:2 (denoted as MO–TMAB 1:2 and MO–CTAB 1:2).
The XRD pattern of MO–TMAB (Fig. 1b) indicated that
the quaternary ammonium cation could incorporate well
into the crystalline structure of the fnal product. TMAB,
also, afected the oxidation state of the formed molybdenum
oxide. The peaks could be precisely indexed on monoclinic
tetramethylammonium tetramolybdate, [N(CH3)4]Mo4O12
(PDF no. 50-1901), and ammonium octamolybdate tet-
rahydrate (NH4)6Mo8O27·4H2O (PDF no. 50-0607). The
surfactant with a longer organic chain, CTAB, directed the
crystalline phase toward a multiphase structure. As seen in
Fig. 1c, it has many more peaks than the previous samples
in 2θ less than 15°, which could be mainly ascribed to the
crystalline phases of MoO3 (PDF no. 21-0569), ammonium
molybdenum oxide hydrates (NH4)6Mo7O24·4H2O (PDF no.
11-0071) and (NH4)4(Mo8O24.8(O2)1.2(H2O)2)(H2O)4 (PDF
no. 88-1326), and tetramethylammonium tetramolybdate
[N(CH3)4]Mo4O12 (PDF no. 50-1901). This multiphase is
probably due to the cationic nature of CTAB during the syn-
thesis reaction, which could be adsorbed by molybdenum
oxide nuclei and afterward it would detach at diferent times
to form multiple phases.
Quantitative Analysis of the Product
GC–MS was used for separation and quantifcation of methyl
esters of fatty acids. Since fatty acids in their free forms are
difcult to analyze with GC (due to their adsorption issues
on stationary phase in GC columns), the reaction products
were esterifed before GC–MS analysis via Metcalfe et al.
derivatization procedure [21, 22] using boron trifuoride
solution in methanol. Briefy, BF3-methanol (5 ml; 10%
w/w, Sigma-Aldrich Co.) was added to the solution of
products of the oxidation reaction. Then, the solution was
heated to 80 °C for 15 min, followed by cooling at room
temperature for about 20 min. The esterifed products were
extracted by adding 3 ml of petroleum ether and 2 ml of
water. The extraction was repeated for the aqueous phase
(lower layer) twice. Obtained organic phase, after dehydra-
tion by sodium sulfate and removal of the solvent by passing
a steady stream of dry air, was ready for injection into the
GC–MS instrument.
A typical derivatized sample including methyl esters of
the involved fatty acids (expectedly dimethyl azelate, methyl
pelargonate, and possibly methyl oleate) was injected into
the GC–MS instrument which had been previously cali-
brated by analytical standards of these methyl esters. The
GC–MS system included a Hewlett-Packard HP 5890 series
GC system and an MSD Hewlett-Packard model 5970. The
GC system was equipped with Zebron ZB-5MS capillary
column (30 m × 0.25 mm × 0.25 mm). Helium was used as
a carrier gas with the fow rate of 30 ml/min. A split ratio of
15:1 was fxed. The front inlet temperature was 280 °C. The
oven temperature program consisted of maintaining at 50 °C
for 2 min, then a ramp rate of 10 °C/min to 160 °C following
by a hold-up time of 1 min, and further increase of the rate
of 5 °C/min to 290 °C. Direct injection was employed with
a 1-μl injection amount for each run, and the injection was
repeated at least four times for each sample to be averaged.
HP Chemstation software was used to analyze data.
Figure 2 shows SEM images of the synthesized materi-
als prepared with and without the surfactants. Apart from
1:2, two other molar ratios of surfactant/molybdenum (1:3
and 1:4) were used in the synthesis to investigate efects of
the surfactants amounts on fnal morphology. The prepared
samples are denoted as MO–CTAB 1:4, MO–CTAB 1:3,
MO–TMAB 1:4, and MO–TMAB 1:3. The SEM image of
MO, synthesized without surfactant (Fig. 2a), demonstrates
one-dimensional nanorods of a few micrometers in length.
Obviously, different sizes and morphologies were
observed according to concentration as well as nature of
the surfactant used (Fig. 2b–g). A lower amount of CTAB
in MO–CTAB 1:4 (Fig. 2b) did not change the morphology
signifcantly; similar nanorods were obtained. Increasing
CTAB content in MO–CTAB 1:3, however, a heterogeneous
mixture of nanorods and microplatelet-like morphology was
obtained (Fig. 2c), which became homogeneous microplate-
lets in MO–CTAB 1:2. The efect of a high concentration of
CTAB is in good agreement with the previous works, where
Results and Discussion
Characterization of the Catalysts
Crystalline structures of the synthesized samples were
analyzed by XRD. Figure 1 represents the XRD patterns
of the synthesized molybdenum oxides. Presence of sharp
peaks indicates good crystallinity of the products, which
were synthesized even without using an autoclave and in a
short time. The XRD pattern of the sample prepared without
1 3

J Am Oil Chem Soc
formation of nanofber, as shown in Fig. 2a. Although using
a low amount of CTAB in MO–CTAB 1:4 did not signif-
cantly change the morphology (Fig. 2b), a further increase of
CTAB in MO–CTAB 1:3 and MO–CTAB 1:2 led to forma-
tion of an emulsion, highly likely because the concentration
of CTAB reached the critical micelle concentration (CMC).
The lipophilic groups of micelles tend to move inward and
hydrophobic groups outward. The concentration of CTAB up
to 1:3 led to a stable spherical micelle and still generated thin
fbers. However, the further increase of CTAB concentration
beyond the CMC increased the deformation of micelles [25].
It has been reported that at higher concentrations of CTAB,
the shape of micelles changes from a sphere to a prolate
ellipsoid, and then the preferred orientation growth of MoO3
is inhibited, resulting in the micro-ellipsoid structure [25].
The concentration of CTAB up to 1:3 led to both spherical
and deformed micelles; thus, the material possesses both a
fber and a plate morphology in Fig. 2c. The highest concen-
tration of CTAB generated homogeneous deformed micelles,
leading to only large plates in MO–CTAB 1:2 (Fig. 2b). The
use of TMAB as a surfactant apparently could not produce
enough micelles, and the concentration of TMAB appears
to have no efect on the size of particles. According to the
literature, the Ostwald ripening mechanism [27] leads to the
formation of hexagonal rods after growth, and TMAB has
not shown much efect; this was confrmed by XRD analysis
(Fig. 1c).
FT-IR analysis has been employed to identify the pres-
ence of surface functional groups on the catalyst. Figures 3
and 4 show the FT-IR spectra of the samples prepared with
CTAB and TMAB, respectively. In both fgures, FT-IR spec-
tra of the corresponding surfactant, as well as the MO sam-
ple (prepared without surfactant), have been also included
to gain better interpretations.
Fig. 1 XRD patterns of the synthesized molybdenum oxides: a MO
(without surfactant), b MO–TMAB 1:2, and c MO–CTAB 1:2
Figure 3 shows two bands at 2920 and 2850 cm−1 for the
CTAB-containing samples (Fig. 3a–c), which are attributed
to the characteristic peaks of symmetric and asymmetric
C–H stretching vibrations of methylene groups of CTAB.
Moreover, the characteristic peak of angular deforma-
tion vibrations of a methylene group was also observed at
1475 cm−1 in these samples. These results were verifed by
the spectrum of pure CTAB (Fig. 3e).
a stacked micro-size of MoO3 fbers and a micro-ellipsoid
structure of MoO3 were predominant [23, 24].
Using TMAB in the synthesis considerably increased the
size of the product particles, leading to a heterogeneous mix-
ture of microrods, which turned to agglomerated rods when
the concentration of was TMAB increased (Fig. 2e–g). By
contrast to CTAB, the size of these hybrid particles contain-
ing TMAB did not change considerably by changing the
amount of TMAB. From these studies, it can be observed
that morphology of the particles has been infuenced signif-
cantly by the nature of the surfactants.
As seen in Fig. 4, asymmetric and symmetric deforma-
tion modes pertaining to (CH3)3N+ of the head group of
TMAB appear at around 1490 and 1390 cm−1, according to
the literature [23]. A pure TMAB spectrum (Fig. 4e) shows
the deformation vibrations of methyl groups at 1490 and
1397 cm−1 and all other samples containing TMAB confrm
these characteristic peaks (Fig. 4a–c). Detailed analysis of
the spectrum of the MO sample (Figs. 3d or 4d) reveals
that the characteristic peak of stretching vibrations of Mo–O
appears at 540 cm−1 [23, 27, 28], which was shifted to 563
and 556 cm−1 in the presence of CTAB and to 580 cm−1 in
Generally, the mechanism of MoO3 formation follows the
anisotropic growing of MoO6 octahedral crystal nuclei as a
basic building unit of MoO3 [25, 26]. In the absence of any
surfactant in the synthesis medium, the growth along the
[010] direction is much more favored, and thus, results in
1 3

J Am Oil Chem Soc
Fig. 2 SEM images of the synthesized molybdenum oxides: a MO, b MO–CTAB 1:4, c MO–CTAB 1:3, d MO–CTAB 1:2, e MO–TMAB 1:4, f
MO–TMAB 1:3, and g MO–TMAB 1:2
the presence of TMAB (Figs. 3 and 4). The stretching mode
of Mo=O bands was located at 965 and 934 cm−1 [29, 30],
and no peak was observed around 995 cm−1, which is gener-
ally assigned to stretching vibrations of Mo–O–Mo. It is a
characteristic of the orthorhombic phase of MoO3.
Quantitative analysis of the water content of catalysts was
performed by TGA.
Figures 5 and 6 show TGA curves of the samples syn-
thesized with CTAB and TMAB, respectively. The TGA
curve of MO (without surfactant; Figs. 5d or 6c) shows
a primary weight loss (~12%) observed in the range of
70–200 °C, which corresponds to desorption of physically
adsorbed water on MoO3–H2O. After 200 °C, the weight of
MO remained constant up to 700 °C, confrming the ther-
mal stability of molybdenum oxide without any surfactant
The peaks observed at 1630 and 3550 cm−1 were attrib-
uted to stretching and bending vibration of hydrogen bonded
–OH groups, qualitatively confrming the presence of OH
groups on the surface and water molecules adsorbed in the
catalysts, which concur with the results of XRD analysis.
1 3

J Am Oil Chem Soc
attached on its surface. Further heating after 700 °C led to
a great weight loss, not only for the MO sample but also for
the other samples, which is due to sublimation of molybde-
num oxide that has been reported before [31].
TGA curves of the samples containing CTAB (Fig. 5a–c)
generally show lower weight losses compared to MO in the
low-temperature region of ˂200 °C, which was attributed to
the hydrogen-bonded water molecules present in the crystal-
line phase; this assumption was confrmed by FT-IR analysis.
The second weight loss observed at the range of 200–300 °C
could be ascribed to decomposition of nitrates and ammo-
nia compounds obtained from the surfactant. In addition, a
third weight loss was obtained in the high-temperature region
(300 ˂ T ˂ 500 °C), which further diferential thermal analysis
(DTA) confrmed that it belongs to an exothermic interaction.
This weight loss could be ascribed to removal or decompo-
sition of CTA+ and elimination of bromide species during
oxidation in air. The total amount of weight losses had a posi-
tive relationship with CTAB content in the samples; 50, 40,
and 37% weight losses were obtained for MO–CTAB 1:2,
1:3, and 1:4, respectively. TGA curves of the samples with
TMAB (Fig. 6) and their DTA analysis showed exothermic
peaks between 300 and 450 °C, corresponding to elimination
of +N(CH3)4 group and bromide species. The weight losses
during oxidation were between 16 and 17% which are much
lower compared to those of the CTAB-capped samples. It
could be due to the small molar weight of TMAB (210.16 g/
mol) compared to that of CTAB (364.45 g/mol).
Fig. 3 FT-IR spectra of the molybdenum oxides prepared with
CTAB at various concentrations: a MO–CTAB 1:2, b MO–CTAB
1:3, c MO–CTAB 1:4, d MO, and e CTAB
Catalytic Test Results
Fig. 4 FT-IR spectra of the molybdenum oxides prepared with
TMAB at various concentrations: a MO–TMAB 1:2, b MO–TMAB
1:3, c MO–TMAB 1:4, d MO, and e TMAB
Catalytic activities of the surfactant-capped molybdenum
oxide catalysts in oxidative scission of oleic acid were
Fig. 5 TGA curves of the
molybdenum oxides prepared
with diferent amounts of
CTAB: a MO–CTAB 1:2, b
MO–CTAB 1:3, c MO–CTAB
1:4, and d MO
1 3

J Am Oil Chem Soc
Fig. 6 TGA curves of the
molybdenum oxides prepared
with diferent amounts of
TMAB: a MO–TMAB 1:2, b
MO–TMAB 1:3, and c MO
investigated in a glass reactor equipped with condenser, ther-
mocouple, and magnetic stirrer. A reaction was conducted
without oleic acid to determine H2O2 loss via catalytic or ther-
mal decomposition and the loss was calculated by titrimetry.
For a 4-h reaction at 85 °C, the loss of the oxidant was 26%.
Therefore, excess amounts of H2O2 were used to complete the
oxidation reactions (the molar ratio of H2O2/oleic acid used in
the reaction was 11.1, which shows an excess amount of H2O2
of about 180%, based on the stoichiometric reaction, Eq. 1).
The frst step of the reaction involves protonation of a per-
oxo moiety on the catalyst to give surface peroxo groups. A
yellow suspension was observed in the reaction solution, due
to the formation of metal-peroxo complexes, –MoO(O2), on
surfaces of the surfactant-capped catalysts. Time-lapse obser-
vations of the reaction mixture containing catalyst, oleic acid,
hydrogen peroxide, and tert-butanol was recorded during the
reactions over the catalysts to track changes that take place
slowly over time (Figure S1 and S2, Supporting Information).
Intriguingly, the heterogeneity behavior of MO–CTAB 1:3
catalyst was changed during the reaction; in the beginning
of the reaction, the catalyst was solid powders observable in
the reaction mixture, which, upon heating to 85 °C, turned
into a homogeneous system. Afterwards, the reaction mix-
ture remined as a clear solution until the end of the reac-
tion. As soon as the reaction mixture was cooled to room
temperature, the catalyst started to precipitate (Figure S1),
enabling self-separation and easy recovery of the catalyst. By
contrast, all of the TMAB containing catalysts retained their
solid and heterogeneous natures during the reaction (Figure
S2). This may be due to the structural diferences between
these two surfactants; CTAB has a long hydrophobic chain of
cetyl groups, which improves the hydrophobic–hydrophobic
interaction between the surface of molybdenum oxide and
oleic acid molecules and traps these molecules in the reaction
medium, resulting in enhancement of contacts between the
reactants and active sites on the catalyst. High dispersion of
MO–CTAB 1:3 catalyst in the reaction mixture, which arose
from the surface organo-modifcation, created conditions
similar to homogeneous catalysis. Moreover, a high concen-
tration of hydrogen peroxide during the reaction results in
rapid formation of metal-peroxo complexes and reinforce-
ment of the emulsion, while at the end of the reaction, when
a majority of H2O2 has been consumed, concentration of such
complexes on the catalyst’s surface drops, resulting in pre-
cipitation of the catalyst.
Table 1 shows conversion of oleic acid and yields of pro-
duction of the desired products, azelaic and pelargonic acids,
over diferent synthesized catalysts with H2O2 as oxidant
after 3.5 h of reaction at 85 °C. The catalytic test results
presented are the average of at least three runs over each
catalyst. The frst catalytic test was done without any catalyst
(Table 1, entry 1), that gives very low conversion (26%) and
yields of azelaic and pelargonic acids (6.4 and 6.8%, respec-
tively). Using the catalyst without surfactant, MO, signif-
cantly increased the conversion (97%) and yields of azelaic
and pelargonic acids (60 and 50%, respectively; entry 2).
An overwhelming majority of the organo-modifed catalysts
resulted in better efciencies (entries 3–8). Ideally, oxidative
cleavage of oleic acid should give similar yields of azelaic
and pelargonic acids (1 mol of oleic acid into 1 mol of each
product) based on the stoichiometric equation of the reac-
tion (Eq. 1). In practice, however, diferences were obtained
in the yields of these two products, which arise from non-
ideality of the catalytic system, diferent decomposition rates
of azelaic and pelargonic acids in the reaction medium, and
presence of by-products.
H3C(CH2)7CH
Catalyst
→
= CH(CH2)7COOH + 4H2O2
H3C(CH2)7COOH
HOOC(CH2)7COOH + 4H2O
(1)
1 3

J Am Oil Chem Soc
Table 1 Catalytic tests results
(conversion of oleic acid and
yields of production of desired
products)
Entry
Catalyst
Surfactant:Mo
molar ratio
Conversiona Yieldb of azelaic Yieldb of pelar-
(%)
acid (%)
gonic acid (%)
1
2
3
4
5
6
7
8
Without catalyst
MO
–
0
26
97
6.4
60
78.2
83
6.8
50
61
68
55
59
60
46
MO–CTAB 1:4
MO–CTAB 1:3
MO–CTAB 1:2
MO–TMAB 1:4
MO–TMAB 1:3
MO–TMAB 1:2
1:4
1:3
1:2
1:4
1:3
1:2
100
100
100
100
100
100
70
68
67.4
58
a
Reaction conditions: time: 3.5 h, temperature: 85 °C, solvent: tert-butanol, initial amounts of oleic acid:
1 g, t-butanol: 8 ml, H2O2: 4 ml, catalyst: 0.4 g
Yield, in this work, is defned as the number of moles of a product formed per mole of oleic acid con-
b
sumed
Use of the surfactants as capping agent increased yields
of both azelaic and pelargonic acids. The maximum yields
were achieved by MO–CTAB 1:3 catalyst (entry 4), which
produced 83% azelaic and 68% pelargonic acids. Com-
pared to the literature, these obtained yields are superior
to what have been obtained by heterogeneous catalysts and
are comparable with what homogeneous ones have shown.
Table 2 lists the results of other works (all the heterogene-
ous catalysts reported so far, to the best of our knowledge,
and some of the best homogeneous catalysts). The excel-
lent catalytic activity of MO–CTAB 1:3, as Table 2 implies,
could be ascribed to the presence of multiple alkyl chains
on the catalyst’s surface, which improve the interactions of
oleic acid with peroxo-molybdenum complexes formed by
the reaction of H2O2 with the particles’ surfaces.
MO–CTAB 1:3 before and after each cycle (Fig. 7) does not
show any notable diference, which proves the stability of
the CTAB capped on the catalyst’s surface. Even after the
fourth cycle, CTAB groups were still attached to the catalyst;
all the characteristic peaks of the alkyl group and molybde-
num oxide remained after the cycles. The slight decrease
in the yields may be due to the progressive detachment of
CTAB and leaching of molybdenum oxide from the particle
surface to the solution (leaching of Mo oxide: ~10% after 4
cycles of reaction). Furthermore, to verify that the reaction
is truly heterogeneous and to investigate the possibility of
leaching of Mo species in the reaction solution after removal
of the catalyst, a catalytic test was performed with MoCTAB
1:3 catalyst and stopped after 1 h of reaction, which showed
44% conversion and 28 and 21% yields of azelaic and pel-
argonic acids, respectively. After cooling to room tempera-
ture, the catalyst was removed, and the reaction solution was
exposed to the reaction conditions for the remaining reaction
time (2.5 h, at 85 °C). Although the conversion increased
to 61%, the yields were negligibly changed (29 and 23%
for azelaic and pelargonic acids, respectively), which could
be well ascribed to the lack of catalyst. However, since the
oxidant could cause the conversion of oleic acid indepen-
dently (see Table 1, entry 1), the conversion was increased
somewhat. From these results, it can be concluded that this
new catalytic system displays high activity, good selectiv-
ity, and environmentally benign properties (using hydrogen
peroxide instead of hazardous ozone) and the efcient reus-
ability of the catalyst makes the process cost-efective and
eco-friendly.
Recyclability of the Catalysts
Stability and, consequently, reusability of the catalyst, which
are important parameters to scale up the process, were also
investigated for the catalyst that gave the best efciency
(MO–CTAB 1:3). The catalyst was conveniently recovered
from the mixture of products after the reaction by simple
centrifugation at high rpm. No signifcant loss of catalyst (or
no signifcant leaching of Mo species to the reaction solu-
tion) was found, even after four cycles of catalytic reaction.
Yield of weight recovery for all the catalysts after the frst
cycle was more than ~97% and loss of the CTAB-capped
catalyst after four cycles obtained by ICP analysis was
9.4 wt%. This good recovery could be due to the protec-
tive role of the capping surfactant on the catalyst surface.
Table 3 represents catalytic efciencies obtained by cata-
lyst MO–CTAB 1:3 in the four performed reaction cycles,
which shows insignifcant activity loss. Yields of desired
products slightly decreased during the cycles while the con-
version was always the same. Comparing FT-IR spectra of
Conclusions
In summary, we have reported successful synthesis of sur-
factant-capped molybdenum oxide catalysts by a simple
1 3

J Am Oil Chem Soc
Table 2 Comparison of
catalytic efciency of MO–
CTAB 1:3 with recent reported
works in the literature
Catalyst/oxidant system
Reaction time
Conversion (%)
–
Yielda (%)
References
[2]b
and temperature
Peroxo-tungsten complex PTA/H2O2
5 h, 80 °C
PA: 82
AA: 79
AA: 82
Polyoxomolybdate
RuCl3/NaIO4
5 h, 90 °C
8 h, RT
–
–
[6]b
PA: 98
[11]b
AA: 62
H2WO4/H2O2
8 h, 100 °C
1 h, 130 °C
1 h, 130 °C
8 h, 80 °C
–
PA: 69
AA: 92
[17]b
Tungsten oxide-SiO2/H2O2
Tungsten oxide/H2O2
79
PA: 36
[32]c
AA: 32
56
PA: 29
AA: 30
[32]c
Chromium supported on MCM-41/O2
Tungsten oxide/H2O2
>95
95
AA: 32.4
[9]c
PA: 32.2
5 h, 120 °C
3.5 h, 85 °C
AA: 58
PA: 24
[33]c
H2O2/MO–CTAB 1:3
100
PA: 68%
This workc
AA: 83%
a AA azelaic acid, PA pelargonic acid
b Homogeneous catalysis
c Heterogeneous catalysis
Table 3 Catalytic efciencies obtained by catalyst MO–CTAB 1:3 in
diferent reaction cycles
Catalysts
Reac-
tion
Conver- Yieldb of
Yieldb of pelar-
siona
(%)
azelaic acid gonic acid (%)
(%)
cycle
MO–CTAB 1:3
MO–CTAB 1:3
MO–CTAB 1:3
MO–CTAB 1:3
1
2
3
4
100
100
100
99
83
80
78
77
68
65
63
61
a
Reaction conditions: time: 3.5 h, temperature: 85 °C, solvent: tert-
butanol, initial amounts of oleic acid: 1 g, t-butanol: 8 ml, H2O2:
4 ml, catalyst: 0.4 g
b
Yield, in this work, is defned as the number of moles of a product
formed per mole of oleic acid consumed
Fig. 7 FT-IR spectra of MO–CTAB 1:3 catalyst after diferent reac-
tion cycles: a after cycle 1, b after cycle 2, c after cycle 3, and d after
cycle 4
oxidative dissolution method at 100 °C, which were char-
acterized by diferent techniques including XRD, FT-IR,
TGA, and SEM analyses. CTAB and TMAB were used as
surfactants to organo-modify surfaces of the molybdenum
oxides. The physicochemical analysis revealed that morphol-
ogy of the products strongly depends on nature and con-
centration of capping agents. The synthesized surfactant-
capped catalysts showed excellent catalytic efciency in
oxidative cleavage of oleic acid to mono- and dicarboxylic
acids. The highest efciency was obtained for the CTAB-
capped molybdenum oxide (with CTAB/Mo molar ratio of
1/3), resulting in full conversion of oleic acid and 83 and
68% yields of production of azelaic and pelargonic acids,
respectively. This catalyst exhibited convenient recovery,
good stability, and steady reusability over recycling experi-
ments without signifcant activity loss up to four cycles.
Employing an environmentally benign oxidant, hydrogen
peroxide, this catalytic system would open a new pathway
for production of diacids and fuel components from renew-
able feedstock.
Acknowledgements ASE thanks the Department of Chemical Engi-
neering at Laval University for welcoming him as a visiting scientist
during his stay in Canada. The authors would like to thank the indus-
trial partners (Oleotek and SiliCycle Inc.) for stimulating discussions
and comments.
1 3

J Am Oil Chem Soc
a peroxomolybdate ionic liquid. RSC Adv 6(1):140–147.
doi:10.1039/C5RA23272D
Compliance with Ethical Standards
14. Du D-Y, Qin J-S, Li S-L, Su Z-M, Lan Y-Q (2014) Recent
advances in porous polyoxometalate-based metal-organic frame-
work materials. Chem Soc Rev 43(13):4615–4632. doi:10.1039/
C3CS60404G
Confict of interest The authors declare that they have no conficts
of interest.
Funding This study was funded by the "Programme Canadien de
Bourses de la Francophonie” (PCBF) and the Nature Sciences and
Engineering Research Council of Canada (NSERC) through an
INNOV-UC grant.
15. Song Y-F, Tsunashima R (2012) Recent advances on polyoxomet-
alate-based molecular and composite materials. Chem Soc Rev
41(22):7384–7402. doi:10.1039/C2CS35143A
16. Zhou Y, Chen G, Long Z, Wang J (2014) Recent advances in
polyoxometalate-based heterogeneous catalytic materials for liq-
uid-phase organic transformations. RSC Adv 4(79):42092–42113.
doi:10.1039/C4RA05175K
References
17. Benessere V, Cucciolito ME, De Santis A, Di Serio M, Esposito
R, Rufo F, Turco R (2015) Sustainable process for production of
azelaic acid through oxidative cleavage of oleic acid. J Am Oil
Chem Soc 92(11):1701–1707. doi:10.1007/s11746-015-2727-z
18. Chandra P, Doke DS, Umbarkar SB, Biradar AV (2014) One-pot
synthesis of ultrasmall MoO3 nanoparticles supported on SiO2,
TiO2, and ZrO2 nanospheres: an efcient epoxidation catalyst. J
Mater Chem A 2(44):19060–19066. doi:10.1039/C4TA03754E
19. Li Z, Li Y, Zhan E, Ta N, Shen W (2013) Morphology-controlled
synthesis of [small alpha]-MoO3 nanomaterials for ethanol
oxidation. J Mater Chem A 1(48):15370–15376. doi:10.1039/
C3TA13402D
1. Rhead MM, Eglinton G, Drafan GH, England PJ (1971) Conver-
sion of oleic acid to saturated fatty acids in Severn estuary sedi-
ments. Nature 232(5309):327–330
2. Godard A, De Caro P, Thiebaud-Roux S, Vedrenne E, Mouloungui
Z (2013) New environmentally friendly oxidative scission of oleic
acid into azelaic acid and pelargonic acid. J Am Oil Chem Soc
90(1):133–140. doi:10.1007/s11746-012-2134-7
3. Köckritz A, Blumenstein M, Martin A (2010) Catalytic cleavage
of methyl oleate or oleic acid. Eur J Lipid Sci Technol 112(1):58–
63. doi:10.1002/ejlt.200900103
4. Kulik A, Janz A, Pohl M-M, Martin A, Köckritz A (2012) Gold-
catalyzed synthesis of dicarboxylic and monocarboxylic acids.
Eur J Lipid Sci Technol 114(11):1327–1332. doi:10.1002/
ejlt.201200027
20. Ma Z, Wu Y, He Y, Wu T (2013) A novel protocol for the oxida-
tive degradation of chitosan with hydrogen peroxide catalyzed by
peroxomolybdate in aqueous solution. RSC Adv 3(30):12049–
12051. doi:10.1039/C3RA40424B
5. Noureddini H, Rempe ML (1996) Pelargonic acid in enhanced
oil recovery. J Am Oil Chem Soc 73(7):939–941. doi:10.1007/
bf02517999
21. Metcalfe LD, Schmitz AA (1961) The rapid preparation of
fatty acid esters for gas chromatographic analysis. Anal Chem
33(3):363–364. doi:10.1021/ac60171a016
6. Turnwald SE, Lorier MA, Wright LJ, Mucalo MR (1998) Oleic
acid oxidation using hydrogen peroxide in conjunction with transi-
tion metal catalysis. J Mater Sci Lett 17(15):1305–1307. doi:10.
1023/a:1006532314593
22. Metcalfe LD, Schmitz AA, Pelka JR (1966) Rapid preparation
of fatty acid esters from lipids for gas chromatographic analysis.
Anal Chem 38(3):514–515. doi:10.1021/ac60235a044
23. Song R-Q, Xu A-W, Deng B, Fang Y-P (2005) Novel multilamel-
lar mesostructured molybdenum oxide nanofbers and nanobelts:
synthesis and characterization. J Phys Chem B 109(48):22758–
22766. doi:10.1021/jp0533325
7. Enferadi Kerenkan A, Beland F, Do T-O (2016) Chemically
catalyzed oxidative cleavage of unsaturated fatty acids and their
derivatives into valuable products for industrial applications:
a review and perspective. Catal Sci Technol 6(4):971–987.
doi:10.1039/C5CY01118C
24. Wang S, Zhang Y, Ma X, Wang W, Li X, Zhang Z, Qian Y (2005)
Hydrothermal route to single crystalline α-MoO3 nanobelts and
hierarchical structures. Solid State Commun 136(5):283–287.
doi:10.1016/j.ssc.2005.08.002
8. Spannring P, Bruijnincx PCA, Weckhuysen BM, Klein Gebbink
RJM (2014) Transition metal-catalyzed oxidative double bond
cleavage of simple and bio-derived alkenes and unsaturated
fatty acids. Catal Sci Technol 4(8):2182–2209. doi:10.1039/
C3CY01095C
25. Gong J, Zeng W, Zhang H (2015) Hydrothermal synthesis of con-
trolled morphologies of MoO3 nanobelts and hierarchical struc-
tures. Mater Lett 154:170–172. doi:10.1016/j.matlet.2015.04.092
26. Li Y, Liu T, Li T, Peng X (2015) Hydrothermal fabrication of con-
trolled morphologies of MoO3 with CTAB: structure and growth.
Mater Lett 140:48–50. doi:10.1016/j.matlet.2014.10.153
27. Chithambararaj A, Chandra Bose A (2014) Role of synthesis vari-
ables on controlled nucleation and growth of hexagonal molyb-
denum oxide nanocrystals: investigation on thermal and optical
properties. CrystEngComm 16(27):6175–6186. doi:10.1039/
C4CE00418C
9. Dapurkar SE, Kawanami H, Yokoyama T, Ikushima Y (2009)
Catalytic oxidation of oleic acid in supercritical carbon diox-
ide media with molecular oxygen. Top Catal 52(6):707–713.
doi:10.1007/s11244-009-9212-6
10. Sparks DL, Antonio Estevez L, Hernandez R (2009) Supercrit-
ical-fuid-assisted oxidation of oleic acid with ozone and potas-
sium permanganate. Green Chem 11(7):986–993. doi:10.1039/
B816515G
11. Rup S, Sindt M, Oget N (2010) Catalytic oxidative cleavage
28. Chithambararaj A, Sanjini NS, Bose AC, Velmathi S (2013)
Flower-like hierarchical h-MoO3: new fndings of efcient visible
light driven nano photocatalyst for methylene blue degradation.
Catal Sci Technol 3(5):1405–1414. doi:10.1039/C3CY20764A
29. Chithambararaj A, Bose AC (2011) Hydrothermal synthesis of
hexagonal and orthorhombic MoO3 nanoparticles. J Alloy Compd
509(31):8105–8110. doi:10.1016/j.jallcom.2011.05.067
30. Masteri-Farahani M, Mahdavi S, Rafzadeh M (2013) Microemul-
sion-mediated synthesis and characterization of monodispersed
nickel molybdate nanocrystals. Ceram Int 39(4):4619–4625.
doi:10.1016/j.ceramint.2012.11.059
of olefns by RuO4
organic solvent-free under ultrasonic irra-
diation. Tetrahedron Lett 51(23):3123–3126. doi:10.1016/j.
tetlet.2010.04.040
12. Rup S, Zimmermann F, Meux E, Schneider M, Sindt M, Oget N
(2009) The ultrasound-assisted oxidative scission of monoenic
fatty acids by ruthenium tetroxide catalysis: infuence of the
mixture of solvents. Ultrason Sonochem 16(2):266–272.
doi:10.1016/j.ultsonch.2008.08.003
13. Dai B, Wu P, Zhu W, Chao Y, Sun J, Xiong J, Jiang W, Li
H (2016) Heterogenization of homogenous oxidative des-
ulfurization reaction on graphene-like boron nitride with
1 3

J Am Oil Chem Soc
31. Mao Y, Li W, Sun X, Ma Y, Xia J, Zhao Y, Lu X, Gan J, Liu Z,
Chen J, Liu P, Tong Y (2012) Room-temperature ferromagnetism
in hierarchically branched MoO3 nanostructures. CrystEngComm
14(4):1419–1424. doi:10.1039/C1CE05700F
33. Enferadi Kerenkan A, Ello Aimé S, Echchahed B, Do T-O (2016)
Synthesis of mesoporous tungsten oxide/γ-alumina and surfactant-
capped tungsten oxide nanoparticles and their catalytic activi-
ties in oxidative cleavage of oleic acid. Int J Chem Reactor Eng
14(4):899–907
32. Noureddini H, Kanabur M (1999) Liquid-phase catalytic oxidation
of unsaturated fatty acids. J Am Oil Chem Soc 76(3):305–312.
doi:10.1007/s11746-999-0236-7
1 3