Kinetic and structural understanding of bulk and supported vanadium-based catalysts for furfural oxidation to maleic anhydride

Article abstract of DOI:10.1039/d1cy01060c

The kinetics of gas-phase furfural partial oxidation to maleic anhydride (MA) was studied over bulk vanadium-phosphorus-based catalysts obtained by aqueous (VPAq) and organic (VPOr) methods and compared to a supported V2O5/Al2O3 catalyst. The solids were characterized by N2 adsorption-desorption, XRD and UV-vis DRS. Results showed a higher specific surface area on VPOr compared with VPAq materials, with a well-defined (VO)2P2O7 crystalline structure. UV-vis analysis showed mainly V(v) on VPAq and an intermediate state between V(iv) and V(v) on VPOr. A detailed kinetic study demonstrated that furfural can be oxidized to MA or COx through parallel paths. At high oxygen partial pressures MA oxidation is inhibited on VPO catalysts but favored on V2O5/Al2O3. A Langmuir-Hinshelwood kinetic model with negligible site occupancy fits the experimental data with a 16% mean error. It also shows a higher apparent activation energy for furfural partial oxidation than for complete oxidation, highlighting the favored selectivity to maleic anhydride at higher temperatures on VPO catalysts.

Full text of DOI:10.1039/d1cy01060c

View Article Online
View Journal
Catalysis
Science &
Technology
Accepted Manuscript
This article can be cited before page numbers have been issued, to do this please use: O. Gomez-
Capiro, L. Bravo, P. Lagos, P. Santander, G. Pecchi and A. Karelovic, Catal. Sci. Technol., 2021, DOI:
10.1039/D1CY01060C.
Volume
Number
21 January 2017
Pages 313-534
7
2
This is an Accepted Manuscript, which has been through the
Royal Society of Chemistry peer review process and has been
accepted for publication.
Catalysis
Science &
Technology
rsc.li/catalysis
Accepted Manuscripts are published online shortly after acceptance,
before technical editing, formatting and proof reading. Using this free
service, authors can make their results available to the community, in
citable form, before we publish the edited article. We will replace this
Accepted Manuscript with the edited and formatted Advance Article as
soon as it is available.
You can find more information about Accepted Manuscripts in the
Information for Authors.
Please note that technical editing may introduce minor changes to the
text and/or graphics, which may alter content. The journal’s standard
Terms & Conditions and the Ethical guidelines still apply. In no event
shall the Royal Society of Chemistry be held responsible for any errors
or omissions in this Accepted Manuscript or any consequences arising
from the use of any information it contains.
ISSN 2044-4761
PAPER
Chaoqiu Chen, Yong Qin et al.
Highly dispersed Pt nanoparticles supported on carbon
nanotubes produced by atomic layer deposition for hydrogen
generation from hydrolysis of ammonia borane
rsc.li/catalysis

Page 1 of 13
Catalysis Science & Technology
Please do not adjust margins
Catalysis Science & Technology
ARTICLE
View Article Online
DOI: 10.1039/D1CY01060C
Kinetic and structural understanding of bulk and supported
vanadium-based catalysts for furfural oxidation to maleic
anhydride†
Received 00th January 20xx,
Accepted 00th January 20xx
Oscar Gómez-Cápiroa,b, Luis Bravoa, Patricio Lagosa, Paola Santandera,b, Gina Pecchib,c, Alejandro
Karelovica,b,d
DOI: 10.1039/x0xx00000x
The kinetics of gas-phase furfural partial oxidation to maleic anhydride (MA) was studied over bulk vanadium-phosphorous-
based catalysts obtained by aqueous (VPAq) and organic (VPOr) methods and compared to a supported V2O5/Al2O3 catalyst.
The solids were characterized by N2 adsorption-desorption, XRD and UV-vis DRS. Results showed a higher specific surface
area on VPOr compared with VPAq materials, with a well-defined (VO)2P2O7 crystalline structure. UV-vis analysis showed
mainly V(V) on VPAq and an intermediate state between V(IV)-V(V) on VPOr. A detailed kinetic study demonstrated that
furfural can be oxidized to MA or COx through parallel paths. At high oxygen partial pressures MA oxidation is inhibited on
VPO catalysts but favored on V2O5/Al2O3. A Langmuir-Hinshelwood kinetic model with negligible sites occupancy fits the
experimental data with a 16% mean error. It also shows a higher apparent activation energy for furfural partial oxidation
than for complete oxidation highlighting the favored selectivity to maleic anhydride at higher temperatures on VPO catalysts.
through the so-called redox Mars-van Krevelen (MvK) mechanism 29
.
In this mechanism, the structure of the catalyst and the nature of the
Introduction.
30,31
support have a strong influence on the catalytic activity
.
Maleic Anhydride (MA) is used as raw material for the production of
resins, agrochemicals, and coatings 1,2. It is currently produced from
the partial oxidation of petrochemicals, mainly n-butane 3–5. The n-
butane partial oxidation has been extensively studied in terms of
reaction conditions
promotors and selectivity to maleic anhydride 11,12, and kinetics 13–16
According to their location in the solid, the O atoms have different
probabilities of reacting 32. However, the oxidation states of V are not
equally active, being V(IV) the most selective to the formation of MA
30. The oxidation state of the V phase is strongly affected by the
nature of the precursor, preparation method and thermal treatment
after the synthesis of the catalysts. Consequently, calcination
temperature and media (air, O2, N2, etc.) are parameters to be taken
into account during the synthesis 33,34. Additionally, it has been
reported that the nature of the solvent, aqueous or organic, has a
6,7
8–10
,
catalysts structure dependence
,
.
However, the possibility of obtaining MA by similar reactions from
17
furfural has been little investigated. The use of furfural obtained
18–20
from biomass
contributes to reducing the carbon footprint of
the products that use MA as raw material, hence it is a real possibility
to join biorefineries with current technologies in the chemical
industry. The furfural partial oxidation in the aqueous phase has
been studied always in batch process 21–25. However, continuous
processes are more efficient and have a better chance of becoming
commercially established 26. In this sense, furfural partial oxidation
in gas phase presents a clear advantage, which is not free of
obstacles. One of the main limitations of this alternative is the
formation of polymerization products under the normally used
strong influence on the crystalline structure obtained after the
8,35
activation of the catalyst
that define the activity and selectivity
towards MA 29,36,37. It has been also reported that, even though P do
not catalyse the MA formation 23, it improves the thermal resistance
of the solids 38 and limits the total oxidation of the surface V atoms,
maintaining the selectivity to MA 39. Several authors reported
(VO)2P2O7 phase as the most active to the selective partial furfural
oxidation towards MA 35,36,40, being detected by X-ray diffraction
(XRD), UV-vis diffuse reflectance spectroscopy, Raman spectroscopy,
reaction conditions (240°C-360°C) 27
.
among others 28,41
.
Catalysts based on vanadium are the most used for partial oxidation
of organic molecules, reported as supported and bulk catalysis 24,27,28
It has been reported that furfural partial oxidation reaction proceeds
Kinetic models of partial catalytic oxidation reactions over metal
transition oxides catalysts has been developed mostly for n-butane
as raw material 4,5,42,43. There is a consensus that the oxygen involved
in the reaction is provided by the support 10,15,27. In the case of
oxidation of n-butane to MA over partially reduced vanadyl
pyrophosphate catalysts, it has been reported that the catalyst
reoxidation step displays a reaction order of 0.55 with respect to the
oxygen pressure, in a range of 2.1 and 21 %vol. O2 and 345°C to
400°C, implying dissociative O2 adsorption under the experimental
conditions 44. On the other hand, Murthy and Rajamani studied
vanadium pentoxide catalysts for furfural partial oxidation to MA.
.
a. Carbon and Catalysis Laboratory (CarboCat), Department of Chemical
Engineering, University of Concepción, Concepción, Chile.
b. Millennium Nuclei on Catalytic Process towards Sustainable Chemistry (CSC),
Chile.
c. Physical Chemistry Department, Faculty of Chemical Sciences, University of
Concepción, Chile.
d. Unidad de Desarrollo Tecnológico (UDT), Universidad de Concepción, Chile.
* Corresponding author akarelov@udec.cl (A. Karelovic)
†Electronic Supplementary Information (ESI) available. See
DOI: 10.1039/x0xx00000x
This journal is © The Royal Society of Chemistry 2021
Catal. Sci. Technol., 2021, 00, 1-3 |1
Please do not adjust margins

Please do not adjust margins
Catalysis Science & Technology
Page 2 of 13
ARTICLE
Catalysis Science & Technology
They proposed a two-stage redox mechanism with a first reaction the method described previously 50. The obtained solids were dried
View Article Online
DOI: 10.1039/D1CY01060C
and calcined as described before. Both series, VPOr and VPAq
catalysts were ground to procure a particle size between 150 and 380
μm.
order to furfural and oxygen partial pressure, in the range 200-280°C
45. The same authors suggested the re-oxidation of the support as the
rate-determining step under the experimental conditions of the
study 46. Regarding supported VOX catalysts, Alonso-Fagundez et al.
report a Langmuir-Hinshelwood model with first-order reaction for
For a better comparison of the catalytic performance of the VPAq
and VPOr catalysts, a supported V2O5/γ-Al2O3 catalyst was also
both furfural and O2 based on experimental results 29
.
studied, as it is has been shown to be active for this reaction 27,29,31
.
We used a 21 wt.% V2O5/γ-Al2O3 (ca. 12 wt.% V) catalyst synthesized
in our last work 31. Briefly, the synthesis procedure was as follows: γ-
Al2O3 (Saint-Gobain Norpro SA 6173) was impregnated with NH4VO3
in presence of oxalic acid (oxalic acid/NH4VO3 molar ratio of 2) in
distilled water at 80°C under continuous stirring at 700 rpm. The
metavanadate was gradually added until a dark blue dissolution was
obtained and, once the oxalate-vanadium complex was formed, the
dissolution under agitation was cooled down to room temperature.
Finally, the dissolution was mixed with the support and dried under
vacuum in a rotary evaporator at 58°C for 2 h. The solid as obtained
was dried overnight at 105°C and calcined in static air at 500°C for 3
h using a heating rate of 10 °C/min.
In this work, we report the effect of the nature of the preparation
method (aqueous or organic) and the P/V molar ratio in the structure
of V-P-O catalysts to be used in the gas phase partial oxidation of
furfural to maleic anhydride. We performed a kinetic study varying
the residence time, temperature and PO2/Pfur feed ratio. A model
developed using a simplified Langmuir-Hinshelwood formalism
assuming O2 molecular dissociation and second O atom addition as
the rate determining steps for both total and selective oxidation,
fitted the experimental data with a mean error of 16.3%. We also
present the kinetic parameters obtained for V2O5/Al2O3 for
comparison, and show the higher activity of the latter catalyst, for
both selective and unselective paths, highlighting the suitability of
VPO catalysts at higher oxidizing conditions. Additionally, the
negligible coverage predicted by the model shows that the reactions
present apparent first-order kinetics.
Catalysts characterization
Adsorption-desorption of N2 at 77K was used to analyze the
morphological structure of the solids.
A Tristar II 3020
(Micromeritics) device was used. After degassing under vacuum at
200°C for 4 h, the samples were analyzed to get the adsorption and
desorption isotherms, following the suggestions of De Lange et al. for
sample preparation and analysis 51. The obtained data were
processed using the BET 52 model to estimate the specific surface and
Experimental.
Catalysts preparation.
53
the BJH method or the hydraulic radii using the Gurvitsch rule to
Catalysts were prepared using two methods, one based on aqueous
(VPAq) and the other on organic (VPOr) medium. It is well known that
both methods allow obtaining solids with marked structural and
morphological differences, whose influence, although studied in the
oxidation of n-butane, has not been studied in depth in the selective
oxidation of furfural. For the VPAq method 47, three catalysts with a
nominal P/V molar ratio of 0.43, 0.85, 1.0 were prepared. The most
validated selective phase for n-butane selective oxidation to MA is
the (VO)2P2O7 8,37,48, which has a P/V molar ratio of 1.0, however, as
this work deals with furfural instead of n-butane, we studied
different P/V molar ratios to determine its impact on selectivity to
MA. The procedure required 25 mL of distilled water heated at 60°C,
H3PO4 (99% purity, Merck) added according to the nominal P/V molar
ratio and 2.89 g of NH4VO3(s) (99% purity, Merck), as V source, mixed
slowly, while the solution was kept at 400 rpm for 2h. This method
was previous described by several authors 40,47. The dissolution was
concentrated by rota-evaporation for 6h and the precursor dried at
105°C for 8h. The calcination was performed in a tubular reactor
inside a cylindrical furnace at 500°C for 4h, with a heating rate of
2°C/min under a constant flow (40 mL/min) of N2 (99.999% purity,
Air Liquide).
obtain the pore volume 54,55. Elemental surface distribution of V and
P
was observed trough Scanning Electron Microscopy-Energy
Dispersive Spectroscopy technique using an EOL JSM-6610LV 6610LV
(BES detector) device at an acceleration voltage of 20 kV. Samples
were coated with a thin layer of gold for 30 s at 0.06 mbar using a
DENTON VACUUM apparatus. The crystalline structure was analyzed
by X-ray diffraction using a Bruker Endeavor D4 diffractometer with
CuKα radiation (λ=0.154059 nm) using 0.02°/step and 1 s/step
between 3° and 90°. Particle size and d-spacing were calculated
according to the Scherrer equation (Eq. 1) and Bragg correlation (Eq.
2), respectively:
퐾휆
퐿 = 훽푐표푠 (휃)
(1)
푛휆 = 2푑푠푖푛(휃)
(2)
Where L is the particle size (nm), K is a constant equal to 0.94 for all
solids, λ is the wavelength (nm), β is the full width at half-maximum
intensity (FWHM) (°), θ is the plane angle (°), n is the order of
reflection and d is the interplanar spacing (nm).
A Thermo Scientific Evolution 260 Bio spectrometer equipped with a
Spectralon-coated integrating sphere ISA 220 was used to perform
UV-vis diffuse reflectance spectroscopy (DRS). The spectra were
recorded in the range of 190 and 1100 nm using a Spectralon disc as
standard.
For the VPOr organic phase method 49, the same nominal P/V molar
ratios were used in isobutanol (14.1 mL) as the dissolution medium.
V2O5 (1.18 g) obtained by the calcination of NH4VO3(s) (99% purity,
Merck) was added with the appropriate amount of H3PO4 (99%
purity, Merck). The dissolution was kept in a reflux heating at 110°C
for 6h until a sky-blue color solid was obtained. The reaction products
were filtered and washed with isobutanol and acetone according to
Catalytic tests.
2 | Catal. Sci. Technol., 2021, 00, 1-3
This journal is © The Royal Society of Chemistry 2021
Please do not adjust margins

Page 3 of 13
Catalysis Science & Technology
Please do not adjust margins
Catalysis Science & Technology
ARTICLE
Catalysts were tested at three temperatures (280°C, 300°C and
320°C) in a tubular reactor heated by a furnace controlled by an
Omega CN7200 device and the gas flow was controlled by mass flow
controllers (Kofloc D8500). We used a 3/8” diameter stainless steel
reactor in which the catalysts was loaded sandwiched by quartz
wool. The catalyst (40 mg) was mixed with quartz sand (400 mg) in
order to prevent the formation of hotspots due to the exothermic
nature of oxidation reactions. Before each experiment, the catalysts
were activated using synthetic air at 350°C for 1h heating at
10°C/min. After that, the furfural/O2/N2 flow was introduced to the
Kinetic modeling of furfural partial oxidation.
View Article Online
DOI: 10.1039/D1CY01060C
The experimental data were used to fit a kinetic model to a simplified
reaction scheme, with partial oxidation of furfural towards MA (R. 1),
complete oxidation in a parallel reaction (R. 2) and subsequently the
total oxidation of MA (R. 3).
푘푝
퐶5퐻4푂2 +2푂2 퐶4퐻2푂3 + 퐻2푂 + 퐶푂2
(R. 1)
(R. 2)
푘푐
퐶5퐻4푂2 +5푂2 2퐻2푂 + 5퐶푂2
푘푐 푀퐴
퐶4퐻2푂3 +3푂2
system and the temperature descended to 320°C to start the
퐻2푂 + 5퐶푂2
(R. 3)
27
reaction at the highest temperature as suggest previous reports
.
Afterward, to check a deactivation process, the temperature was
decreased to 300°C, then to 280°C, and finally increased to 320°C.
Each temperature was maintained for 4 h. Furfural (Merck-Millipore)
was fed using a saturator kept at 30°C in a thermostated bath, with
the constant bubbling of 5 mL/min of N2 to reach a partial pressure
of 0.0804 kPa. The parameters, in the first round of experiments,
were: PO2/Pfur ratio=20 (Synthetic Air, Air Liquide), 2 mL/min of Ar
(99.999% purity, Air Liquide) as an internal reference, and N2
(99.999% purity, Air Liquide) to complete a total flow of 22 mL/min.
As a detector, a mass spectrometer device was used (Thermostar
Omnistar GSD 320). The furfural conversion was determined
according to equation 3 (Eq. 3). The equations 4 and 5 (Eq. 4 and Eq.
5) were used to calculate the yield (for MA and CO2) and the space-
time yield (STY) (for MA).
As will be shown in section 3.5, the total oxidation of MA (R.3) was
negligible for V-P-O catalysts, therefore the model was simplified to
two main reactions (R.1 and R.2). To these simplified models, a
Langmuir-Hinshelwood based model was studied. A 1-D pseudo-
homogeneous model was assumed for the fixed bed reactor (Eq. 6),
that was operated at integral conditions.
푑푃푖
=
푅푇푟푖
(6)
푑푊
푣
Where Pi is the partial pressure for compound i (kPa), W is the
catalyst bed mass (g), R is the universal gas constant (J mol-1 K-1), T is
the temperature (K), v is the volumetric flow rate of feed (ml min-1)
and r is the net reaction rate for the compound i (mol min-1 gcat-1).
The absence of mass transfer limitations was checked using the
recommended criteria (Section S1 of Supplementary Information).
An optimization method based on genetic algorithms, using the
optimization toolbox of Matlab®, was used to determine the model
parameters, by minimizing the error between the experimental and
calculated data. The objective function and method of calculation are
detailed in Section S3.
푃0푓푢푟 ― 푃푓푢푟
푥푓푢푟
=
100
(3)
푃0푓푢푟
푛푖
푎푖 푛0푓푢푟
푌푖 =
100
(4)
(5)
푆푇푌 = 푚푚
푀퐴
푐푎푡
Where xfur is the furfural conversion, P0fur is the initial partial pressure
of furfural, Pfur is the furfural partial pressure at the exhaust, Yi is the
yield of the product i, ṅi represent the molar flow of the product i in
the exhaust, ai represent the stoichiometric coefficient of product i
Results and Discussion.
Morphological and elemental characteristics.
(stoichiometric coefficient of furfural is one for all reactions), ṅ0
The VPAq and VPOr catalysts with different P/V molar ratios were
characterized using N2 adsorption-desorption isotherms at 77K
(Figure S1 in Supplementary Information). The isotherms were
classified as type II according to the IUPAC classification 56, typically
of solid with low porosity or macro-porosity. Figure 1 shows the
specific surface area as a function of the P/V molar ratio for the
synthesized catalysts, and Table S3 summarizes the specific surface
and the mean pore size values. It can be seen for VPOr and VPAq
fur
represent the initial molar flow of furfural, ṁMA is the MA mass flow
in the exhaust and mcat is the mass of the catalyst bed inside the
reactor.
Once the performance of the catalysts was established, the VPOr1.0
and the V2O5/γ-Al2O3 catalysts were used in further analysis to
perform a detailed kinetic study, changing the PO2/Pfur ratio,
residence time (W/F ratio) and reaction temperature. Therefore, the
PO2/Pfur ratios of 10, 20 and 40 were studied, and the W/F ratio was
changed keeping constant the amount of catalyst, using total flows
40
catalysts the same trend reported by Horowitz i.e., a decrease in
the surface area while increase the P/V molar ratio. Concerning the
of 10, 30, and 40 mL/min, in addition to the 22 mL/min previously effect of the preparation method, VPOr catalysts presented higher
mentioned. The reagents and reaction products were analyzed in-
line with a GC (SRI 8610C) with a capillary column (30m x 0.53mm
I.D. 1.0U MXT-WAX, Restek 70655-273) and three packed columns
(6' x 1/8" Molecular sieve 5A, 18" x 1/8" Hayesep D and 6' x 1/8"
Hayesep D). The system was equipped with a thermal conductivity
detector (TCD), a flame ionization detector (FID) and a FID detector
equipped with a methanizer.
surface area than VPAq, in all of the studied range of P/V molar ratio.
Although the specific surface area values are very low and may be
within the error range of the measuring device, the differences
between the two groups of catalysts are clear. Since all
measurements were performed on the same device under similar
environmental conditions, the differences between VPAq and VPOr
can be assumed as significant. The pore volume in the VPAq group
This journal is © The Royal Society of Chemistry 2021
Catal. Sci. Technol., 2021, 00, 1-3 | 3
Please do not adjust margins

Please do not adjust margins
Catalysis Science & Technology
Page 4 of 13
ARTICLE
Catalysis Science & Technology
Crystalline structure.
View Article Online
DOI: 10.1039/D1CY01060C
Figure 2 shows the X-ray diffraction patterns associated with the
crystalline structure of the synthesized catalysts. The observed
features correspond to phases previously described 35,57–59, and it is
observed that for higher P/V molar ratios VPOr catalysts display
notorious diffraction peaks, indicative of more crystalline structures.
On the contrary, VPAq catalysts are mainly amorphous. The differe-
Figure 1. The specific surface area as a function of the P/V molar ratio
for VPAq and VPOr (P/V=0.43, 0.85, 1.00). Data for V-P-O catalysts
reported by Horowitz 40 is included. The lines are added to show the
trend.
maintains the same tendency to decrease as the P/V ratio increases.
However, this morphological parameter in the VPOr group shows an
opposite behavior. On the other hand, the supported V2O5/Al2O3
catalyst presents high specific surface area (158 m2/g), whereas
showing a decrease of 30% of the initial γ-Al2O3 specific surface area
(227 m2/g) upon vanadium impregnation. The V2O5/Al2O3 and γ-Al2O3
isotherms and their pore size distribution are shown in Figure S1.
The elements distribution on the bulk catalysts surface were studied
using SEM-EDS mapping. Figure S2 show the screenshots
corresponding to each element founded in the catalysts. It is
observed an homogeneous distribution in all catalysts, which is
important to form V-P phases declared as active for organic
molecules oxidation 30. In addition, Table 1 show the elemental
composition determined by this technique. The V diminish as well
the P increase observed into each catalysts group correspond with
the changes expected according P/V molar ratio, although it may not
represent the elemental composition of every solid.
Table 1. Elemental composition of VPAq and VPOr catalysts obtained
by SEM-EDS analysis.
Elements (wt. %)
Catalyst
VPAq0.43
VPAq0.85
VPAq1.0
VPOr0.43
VPOr0.85
VPOr1.0
V
P
O
43.28
42.39
41.6
17.95
21.2
38.75
36.4
21.92
18.13
20.28
21.75
36.48
35.08
37.17
38.49
46.79
42.55
39.76
Figure 2. XRD patterns for (a) VPAq and (b) VPOr catalysts (P/V=0.43,
0.0.85, 1.00), and (c) V2O5/Al2O3 supported catalyst.
4 | Catal. Sci. Technol., 2021, 00, 1-3
This journal is © The Royal Society of Chemistry 2021
Please do not adjust margins

Page 5 of 13
Catalysis Science & Technology
Please do not adjust margins
Catalysis Science & Technology
ARTICLE
nce observed between both groups of catalysts coincide with
previous results reported by Batis et al. 57, who prepared V-P-O
catalysts by similar synthesis methods as used in this work. In the
latter group of catalysts, VPAq0.43 catalyst show significant diffraction
peaks (Fig. 2a) which correspond to VOPO4 compounds, mainly γ-
UV-vis DRS and band gap energy determination.
View Article Online
DOI: 10.1039/D1CY01060C
UV-vis DRS experiments were carried out for a better understanding
of the oxidation state of vanadium in the VPOr and VPAq catalysts
(Fig. S3). The maximum of the main peak is presented in Table 3.
Martin et al. reported a charge transfer transition band for V(IV) at
220 nm and the corresponding V(V) at 280 nm and 340-350 nm
VOPO4. In the case of VPAq0.85, the plane (040) is maintained, and a
70
.
57
new peak, attributed to γ-VOPO4 (plane (230)) was identified
.
The spectra indicate that the main peaks of the VPOr catalysts (243-
256 nm) are closer to 220 nm than the VPAq signals (260 to 280 nm).
Regarding VPOr catalysts (Fig. 2b), the main difference with respect
to the aqueous route of preparation is the appearance of (VO)2P2O7
signals coexisting with γ-VOPO4 in the case of VPOr0.43. The (VO)2P2O7
phase has been reported as the active catalytic phase for n-butane
partial oxidation specifically the plane (200) 8,37,48, which is not
detected in the VPAq catalysts. However, the evidence regarding the
activity of this phase is much more scarce 59. These authors achieved
high MA selectivities and yields, which is the focus of most studies in
the field to the detriment of elucidating the kinetics governing the
reaction.
This result is in line with XRD, which showed that (VO)2P2O7, with
V(IV) is the predominant crystalline phase in the VPOr catalysts. The
shift towards higher wavelength for VPOr indicates an amorphous
phase not identified by XRD with V(V) in its structure since the
influence of γ-VOPO4 (with V(V)) is negligible as P/V molar ratio
increases, as show in the XRD pattern for these catalysts (Fig. 2b).
Table 3. The wavelength of the main peak of DRS as a function of
the P/V molar ratio for the VPAq and VPO catalysts.
The crystallite size, consistent with the planes (200), (032) and (024),
corresponding to (VO)2P2O7, were calculated by Scherrer equation
and are shown in Table 2 for VPOr catalysts, together with the
changes in the d-spacing related with the P/V molar ratio. The P-
induced defects cause changes in the interplanar distance since the
d-spacing increases progressively with the P/V molar ratio on plane
(024) while it is still constant for plane (032). It has been reported d-
spacing dimensions of 0.387 nm, 0.314 nm, and 0.298 nm for the
planes (200), (024), and (032), respectively 37. The results show a
minimum in the planes (200) and (024) (0.371nm and 0.301 nm) for
P/V molar
ratio
VPAq Peak (nm)
VPOr Peak (nm)
0.43
0.85
1.00
280
275
260
243
256
252
The amorphous phase with V(V) is also evidenced for the VPAq
where the wavelength values shifted towards 280 nm. Wavelength
where the DRS maximum occurs (280 nm) for VPAq0.43 matches
exactly with literature reports for pure V(V) phases 70. This was to be
expected since this catalyst presents exclusively crystalline phases of
vanadium phosphates (β-VOPO4, δ-VOPO4 and γ-VOPO4). VPAq
catalysts with P/V≥0.85 display a larger amount of amorphous phase
with a large amount of V(V) species, but with presence of V(IV) as
well, since the wavelength values are lower than 280 nm. The phase
responsible for the decrease in the wavelength for the maximum in
the DRS of VPAq catalysts with P/V≥0.85 could be a disorganized no
crystalline structure of (VO)2P2O7, impossible to detect in XRD
VPOr0.85. This means that increasing P concentration affects
differently the (VO)2P2O7 crystallite formation, favoring the growth in
the direction of one plane over the others. Regarding the V2O5/Al2O3
pattern, only the peaks attributed to the crystalline V2O5 phase 60 and
the support 61 were detected (Fig. 2c).
These kind of patterns (Fig. 2a and 2b), with inhomogeneous line
broadening of the diffraction lines of a crystalline phase, should not
be analyzed with the Williamson-Hall methods because would
provide wrong results 62. Others authors use Scherrer equation to
estimate the crystallite size of V-P-O catalysts with similar XRD
patterns 63–69
.
analysis 57
.
Determination of the catalytic activity of VPAq and VPOr.
Table 2. Crystallite particle size calculated by Scherrer equation, and
d-spacing from Bragg correlation for different (VO)2P2O7 planes for
the VPOr catalysts.
The catalytic performance of VPAq and VPOr catalysts for the
oxidation of furfural, was studied in the range of 280°C and 320°C.
Figure 3 shows the MA yield as a function of the furfural conversion
over the VPAq and VPOr catalysts. As expected, the increase in the
reaction temperature produces an augmentation in the furfural
conversion. Thus, as more furfural is converted, MA yield
proportionally increases, with similar behavior regardless of the P/V
molar ratio or the catalyst preparation method. The continuous
increase in the MA yield with furfural conversion means that, even at
the highest conversion rates (30-40%), MA does not suffer significant
further oxidation reactions over the bulk catalysts. On the other
hand, Figure 4 shows that for VPAq catalysts a decrease in the furfu-
Particle size (nm) | d-spacing (nm)
P/V molar
ratio
Plane (200)
22.0 | 0.372
14.7 | 0.371
14.7 | 0.375
Plane (024)
17.9 | 0.302
8.9 | 0.301
8.9 | 0.307
Plane (032)
14.9 | 0.291
14.9 | 0.290
22.4 | 0.290
0.43
0.85
1.00
This journal is © The Royal Society of Chemistry 2021
Catal. Sci. Technol., 2021, 00, 1-3 | 5
Please do not adjust margins

Please do not adjust margins
Catalysis Science & Technology
Page 6 of 13
ARTICLE
Catalysis Science & Technology
experimentally confirm this, since the highest MA yieldVsieawreAroticbletaOinnleinde
DOI: 10.1039/D1CY01060C
over VPOr with the best defined (VO)2P2O7 crystalline structure
(VPOr0.43) as the reaction temperature increases. The fact that (200)
plane favors MA formation from furfural partial oxidation has been
observed by Li et al. who prepared plate V-P-O catalysts 59
.
A quick comparison between temperature extremes over catalysts
with molar ratio P/V=1 is evidence of this. At 320°C the furfural
conversion over VPOr1.0 almost doubles the conversion over VPAq1.0
(33% vs 17%, respectively), while the MA yield is more than 20%
higher for VPOr1.0 with respect to VPAq1.0 at the same reaction
conditions (9.7% vs 8%, respectively). At 280°C VPAq1.0 is not able to
significantly oxidize furfural, however VPOr1.0 presents a furfural
conversion of 8.2% reaching an MA yield of 1.1%.
Figure 3. MA yield vs. Furfural conversion over VPAq (empty symbols)
and VPOr (filled symbols) catalysts. (P/V molar ratios=0.43, 0.85,
1.00; T=280°C, 300°C, 320°C; W/F=1.81 min mgcat mL-1; PO2/Pfur=20).
The lines are added to show the trend.
ral conversion is observed with the increase in the P/V molar ratio,
whereas no significant effect is observed for VPOr catalysts.
Considering the previous characterization results, the lower furfural
conversion for the VPAq catalysts can be attributed to their lower
specific surface area (Fig. 1) and/or their lower crystallinity (Fig. 2a).
Even taking into account that the specific surface areas of both
groups of catalysts are low and that the crystallinity is not completely
defined, it can be observed that the VPOr have a higher surface area
for all the P/V ratios, while the VPAq do not show signs of the phase
(VO)2P2O7 in the XRD patterns.
Evidence of that is the high furfural partial oxidation activity and MA
yield of VPAq0.43 catalyst (Fig. 4a), which has the higher specific
surface area and the best defined crystalline structure in the group
of VPAq. These results suggests a combined effect of both crystalline
structure and specific surface area on the improvement of the
catalysts activity. A similar behaviour shows the VPOr0.43 catalyst,
which has the highest specific surface area and the best-defined
(VO)2P2O7 crystal structure among the VPOr catalysts.
Regarding the stability of the catalysts, since the furfural conversion
and MA yield returned to the initial level when the reaction
temperature was increased to the highest value, no significant
deactivation was detected for all of the studied catalysts (Fig. S4).
The slight difference evidenced on furfural conversion between the
initial and final segments at 320°C is due the steady state is not yet
reached when the temperature change to 280°C.
Figure 4. Furfural conversion vs. P/V molar ratio (0.43, 0.85, 1.00) at
different reaction temperatures of 280°C, 300°C and 320°C on (a)
VPAq and (b) VPOr catalysts. (W/F=1.81 min mgcat ml-1, PO2/Pfur=20).
The lines are added to show the trend. Error bars indicate the
standard deviation in furfural conversion (20%).
Figure 5 shows the effect of the P/V molar ratio on the MA yield for
the VPAq and VPOr catalysts. On n-butane oxidation to MA the
crystalline (VO)2P2O7 phase is not only the active phase for the
oxidation reaction but also selective in producing MA from n-butane
Influence of the reaction conditions on VPOr1.0 catalytic activity.
Comparison with V2O5/Al2O3.
34
partial oxidation
. Therefore, in the VPOr catalysts, the
preferentially exposed plane (200) of the (VO)2P2O7 phase (Fig. 2b),
which is absent in the VPAq catalysts (Fig. 2a), could be responsible
for the MA formation from furfural. The results shown in Figure 5b
The kinetics of the reaction was studied by varying the PO2/Pfur ratio
in the feed as well as the space velocity. Since VPOr catalysts display
6 | Catal. Sci. Technol., 2021, 00, 1-3
This journal is © The Royal Society of Chemistry 2021
Please do not adjust margins

Page 7 of 13
Catalysis Science & Technology
Please do not adjust margins
Catalysis Science & Technology
ARTICLE
accounted for 86.1% (Calculations detailed in sectionVSie2w).ArTtihcliesOmnlianye
DOI: 10.1039/D1CY01060C
be due to secondary reactions that polymerize the furfural,
generating resin deposits on the surface of the catalysts. This has
27
been observed on V2O5/Al2O3 and reported previously
.
Nevertheless, the activity of the catalysts was not significantly
decreased during the 20h duration catalytic test. Longer run times
may be required to determine if the polymerization affects the
activity of the catalysts.
To evaluate the effect of the residence time, four W/F values were
tested at different reaction temperatures, maintaining constant the
inlet furfural partial pressure (Fig. S6). Figure 6b shows the W/F ratio
influence at 300°C. CO2 formation still almost constant with the
increase in the W/F ratio, indicating that the total oxidation is not
favored and the increase of furfural conversion is related to the
increase of the MA yield. Furthermore, implies that for VPOr1.0
catalyst, larger contact time favor MA formation without suffering
subsequent oxidation or other side reactions.
The catalytic performance of the previously studied V2O5/Al2O3
catalyst was also analyzed. Figure 7a shows the effect of the PO2/Pfur
ratio and Figure 7b, the effect W/F residence time on furfural
conversion and MA and CO2 yields at 300°C. The full temperature
range is show in Figure S6 for PO2/Pfur ratio influence and in Figure S7
for W/F ratio influence. Compared to VPOr1.0 catalyst, it can be seen
that, under the same experimental conditions, furfural conversion
and MA yield tend to be higher for the V2O5/Al2O3 catalyst. This result
could be attributed to the higher surface area, and hence a larger
amount of surface V sites available on the V2O5/Al2O3 catalyst,
compared to the bulk V-P-O catalysts. The increase in the MA yield
with the increase of PO2/Pfur ratio (Fig. 7a) is in line with previous
reports 27,71,72, while the CO2 yield shows the opposite trend. Figure
7b shows the effect of contact time on the furfural conversion and
product yields. At higher temperatures (300°C and 320°C) the
furfural conversion is almost complete. Under these conditions, the
CO2 yield and the MA yield each remain at a respective range, both
with little change. Since that, the conversion does not change; this
was expected for both yields.
Figure 5. MA yield vs. P/V molar ratio at different reaction
temperature on (a) VPAq and (b) VPOr catalysts. (W/F=1.81 min mgcat
ml-1, PO2/Pfur=20). The lines are added to show the trend. Error bars
indicate the standard deviation in MA yield (20%).
larger furfural conversion and MA yield compared with VPAq, the
VPOr1.0 catalyst was choose for the kinetics assays. It should be
emphasized that the aim was the kinetics study of the furfural
oxidation process instead the optimization of the MA yield. The
higher furfural conversion and MA yield of VPOr0.43 was no required
in this case. Three PO2/Pfur ratios were studied in the range of 280°C
to 320°C (Fig. S5), and the CO2 yield was included as an indication of
total oxidation extent. Figure 6a shows that the furfural conversion
and MA yield increase proportionally with the increase of the PO2/Pfur
ratio at 300°C, evidencing the importance of reactions R. 1 and R. 2.
The MA yield did not decrease with the increase of the oxidizing
potential, indicating that transformation of MA via total oxidation is
not significant (R. 3). It is important to highlight that other products
(not identified) were detected albeit in small amount. In this sense,
the carbon balance was performed for the furfural partial oxidation
in the whole range of reaction conditions. Over VPOr1.0, the
identified products and unconverted furfural detected at the reactor
outlet accounted for 92.5% of the carbon in the inlet as average of
all experiments. In the case of V2O5/Al2O3, the same compounds
Figure 8 shows the STY of MA as a function of the reaction
temperature on V2O5/Al2O3 and bulk VPOr1.0 catalysts. Even though
we evidenced a higher furfural conversion and MA yield on
V2O5/Al2O3 catalyst, it is worth noting that under the stronger
reaction condition i.e., 320°C and PO2/Pfur=40, the MA yield on
V2O5/Al2O3 decreases, while on VPOr1.0 catalyst it keeps increasing.
The combination of both effects, reaction temperature and PO2/Pfur
ratio defines the oxidative potential 27 as the conditions in which the
catalyst exhibits its maximum oxidative performance and the
maximum selectivity to MA production is reached. Although the
V2O5/Al2O3 catalyst is inherently more active for oxidation reactions,
under the same oxidative conditions, the complete furfural oxidation
is favored. Figure 8 shows that the V2O5/Al2O3 catalyst presents the
maximum performance at PO2/Pfur=40 and 280°C and that the VPOr1.0
catalyst does not reach the maximum MA yield during the
experimental conditions. This result means that the VPOr1.0 catalyst
require higher oxidizing conditions (T>320°C and/or PO2/Pfur >40) to
This journal is © The Royal Society of Chemistry 2021
Catal. Sci. Technol., 2021, 00, 1-3 | 7
Please do not adjust margins

Please do not adjust margins
Catalysis Science & Technology
Page 8 of 13
ARTICLE
Catalysis Science & Technology
View Article Online
DOI: 10.1039/D1CY01060C
Figure 6. Effect of (a) the PO2/Pfur ratio at 300°C on furfural Figure 7. Effect of: (a) the PO2/Pfur ratio at 300°C on furfural
conversion, MA yield, and CO2 yield over VPOr1.0 catalyst (W/F=1.81 conversion, MA yield, and CO2 yield over V2O5/Al2O3 catalyst
min mgcat mL-1); (b) the W/F ratio at 300°C on furfural conversion, MA (W/F=1.81 min mgcat mL-1); (b) the W/F ratio at 300°C on furfural
yield and CO2 yield over VPOr1.0 catalyst (PO2/Pfur=20). The lines are conversion, MA yield and CO2 yield over V2O5/Al2O3 catalyst
added to show the trend. Error bars indicate the standard deviation (PO2/Pfur=20). The lines are added to show the trend. Error bars
in furfural conversion; MA and CO2 yield (20%).
indicate the standard deviation in furfural conversion, MA and CO2
yield (20%).
achieve the same conversion than V2O5/Al2O3 catalyst.
A comparison with STY previously reported show similar trends. Li et
al. 59 tested a V-P-O catalyst prepared by organic synthesis at 340°C
reaching 116 gMAkgcat-1h-1, which is the expected because the higher
temperature used by these authors. In the same work, a V-P-O
catalyst activated specially to reach a better defined (VO)2P2O7
crystalline structure reach 234 gMAkgcat-1h-1 at the same temperature
59. In the case the V2O5/Al2O3, others authors tested the furfural
oxidation to MA between 300°C and 345°C, reaching a maximum STY
of 145 gMAkgcat-1h-1 at 315°C 27. Higher temperatures values cause a
-
decrease in the MA selectivity with the minimum STY of 51.8 gMAkgcat
1h-1 at the upper temperature of the range studied 27
.
Kinetic modelling of the partial oxidation of furfural
Figure 8. Effect of the reaction temperature on the STY for the
supported V2O5/Al2O3 and bulk VPOr1.0 catalysts at 280°C, 300°C and
320°C.(PO2/Pfur=10, 40; W/F=1.81 min mgcat mL-1). The lines are
added to show the trend.
Previous studies reported by Fagúndez et al. 29 showed a first-order
kinetics for both furfural and O2 pressure on MA formation. Their
results suggest thst the reaction occur via a Langmuir-Hinshelwood
8 | Catal. Sci. Technol., 2021, 00, 1-3
This journal is © The Royal Society of Chemistry 2021
Please do not adjust margins

Page 9 of 13
Catalysis Science & Technology
Please do not adjust margins
Catalysis Science & Technology
ARTICLE
mechanism. If the surface reaction between adsorbed furfural and Partial oxidation and complete oxidation kineVtiiecw AcrtoicnlesOtannlintes
DOI: 10.1039/D1CY01060C
oxygen is the rate determining step, one can derive equations 7a and correspond to equations 8a and 8b, respectively. (Eq. 8a and Eq. 8b),
7b for furfural partial oxidation and furfural complete oxidation, the expressions were referred to the lower reaction temperature
respectively. It should be noted that the kinetic constants implicitly (280°C).
푬풂
incorporate the kinetic constant of the rate determining step and the
adsorption equilibrium constants of furfural and O2 adsorption. The
equations for partial and complete oxidation are thus the following:
풑
ퟏ
ퟏ
푻
―
(
)
(
)
푹
ퟐퟖퟎ
풌풑 = 풌풑 ퟐퟖퟎ
풆
(8a)
(8b)
푬풂
풄
ퟏ
ퟏ
―
(
)
(
)
푹
ퟐퟖퟎ
푻
―풓풑 = 풌풑 푷풇풖풓 푷푶
풌풄 = 풌풄 ퟐퟖퟎ
풆
(7a)
ퟐ
Where Eai is the apparent activation energy (J mol-1 K-1), kp 280 is the
kinetic constant for partial oxidation at 280°C (mol min-1 gcat-1 kPa-2),
kc 280 is the kinetic constant for complete oxidation at 280°C (mol min-
1 gcat-1 kPa-2), kp is the kinetic constant for partial oxidation (mol min-
―풓풄 = 풌풄 푷풇풖풓 푷푶
(7b)
ퟐ
Where kc is the kinetic constant for complete oxidation (mol min-1
gcat-1 kPa-2) and kp is the kinetic constant for partial oxidation (mol
min-1 gcat-1 kPa-2)
gcat kPa-2), and kc is the kinetic constant for complete oxidation
1
-1
The absence of adsorption terms in the denominator of equations 7a
and 7b implies a negligible surface coverage by furfural, O2, or their
intermediates in the reactions. Furan and compounds with similar
structure such as furanone has been identified as surface
intermediates 72,73. Transient studies of furan, 2(5H)-furanone and
MA adsorption over vanadyl pyrophosphate catalyst were carried
out using FTIR spectroscopy and reported previously 14. These
authors inform a progressively coverage loss of all the compounds
mentioned and their intermediates in a range of 200°C-300°C when
the steady-state was reached (Experiment duration>3h). The clean
surface observed under the mentioned temperature conditions are
in line with equations 7a and 7b.
(mol min-1 gcat-1 kPa-2).
Table 4 summarizes the parameters required to fit the kinetic model
and the average error of the model (ΣErrfur and ΣErrMA). The apparent
activation energy for partial oxidation is higher than that for
complete oxidation, which indicates that the former is favored at
higher temperatures, as it has been shown in Figure 8 for the VPO1.0
catalyst. Schneider et al. reported an activation energy of 72.14
kJ/mol for the partial oxidation of n-butane to MA, and 72.02 kJ/mol
for the complete oxidation to CO2 74. Bej et al. reported slightly higher
values (108 kJ/mol) for the partial oxidation of the same molecule 75
.
Table 4. Parameters obtained from the Langmuir-Hinshelwood model for both catalysts (VPOr1.0 and V2O5/Al2O3).
Mean error (%)
Catalyst
Reaction
k280 (mol min-1 gcat-1 kPa-2)*
Eapp (kJ/mol)*
For Pfur
For PMA
Furfural partial
oxidation (R.1)
7.67±0.23×10-4
4.66±0.05×10-3
3.87±0.08×10-2
7.85±0.005×10-2
4.19±0.42×10-6
156±2.3
90±0.9
n.d.
VPOr1.0
1.7
30.9
Furfural complete
oxidation (R.2)
Furfural partial
oxidation (R.1)
Furfural complete
oxidation (R.2)
V2O5/Al2O3
n.d.
17.6
10.4
MA complete oxidation
(R.3)
n.d.
*95% confidence interval calculated based on individual error.
The obtained apparent activation energy of 156 and 90 kJ/mol (Table used here as was demonstrated previously in this document (Fig. 8).
4), for the partial and complete oxidation, respectively, over the Then, the higher activity observed in the supported catalyst could be
VPOr1.0 catalyst are greater for both reactions; however, the explained by this activation energy difference.
molecule oxidized was different (n-butane vs furfural). The furfural
Figure S8 shows the parity plot for furfural partial pressure (Fig. S9a)
molecule oxidation (partial or complete) require to break the furan
and MA partial pressure (Fig. S9b) predictions. The model represent
ring demanding higher activation energy. Fagúndez et al. 27 reports a
very accurately the experimental furfural partial pressure with a
lower activation energy (98.3 kJ/mol) than the reported in this work,
mean error of only 1.7%. On the other hand, the MA partial pressure
for furfural partial oxidation to MA over V2O5/γ-Al2O3. In this case,
present a higher average error, with a trend to under-predict the MA
the catalyst used by these authors is quite more active than VPOr1.0
This journal is © The Royal Society of Chemistry 2021
Catal. Sci. Technol., 2021, 00, 1-3 | 9
Please do not adjust margins

Please do not adjust margins
Catalysis Science & Technology
Page 10 of 13
ARTICLE
Catalysis Science & Technology
production. The fact that the kinetic models predict more accurately
the reactant pressure than MA partial pressure, has been reported
previously by Lorences for partial oxidation of n-butane 15. This can
be rationalized taking into account that MA errors are amplified by
the fact that MA predictions also take into account the furfural
prediction errors.
View Article Online
DOI: 10.1039/D1CY01060C
Figure 9 show the model fit at the extremes of the temperature work
range as a function of the change in PO2/Pfur ratio (Fig. 9a) and W/F
ratio (Fig. 9b). The kinetic model accurately predicts exit furfural
partial pressure related with the PO2/Pfur ratio and the W/F ratio as
well. MA yields are also accurately predicted although errors
increase at the conditions where the experiments reach higher MA
yields. These conditions are the extremes of the temperature
(320°C), PO2/Pfur ratio (40) and W/F ratio (4 min/gcat mL) work range.
It could be argued that the assumptions made to propose the kinetic
model begin to be less accurate, probably related to the occurrence
of side reactions or the crowding of surface sites which the model
does not predict. Future studies must be developed under similar
experimental condition using techniques such as DRIFTS capable to
determine the functional groups present on the catalyst surface
during the reaction.
The developed reaction model was used to represent the
experimental values obtained over V2O5/Al2O3 catalyst using the
results at 280°C when the furfural conversion was not complete. In
this case, the reaction R.3 was included in the system since the
decrease in the MA yield at higher furfural conversion indicated the
occurrence of a consecutive path of complete MA oxidation (Fig. 8).
A similar rate equation was used here (Eq. 9), namely assuming that
surface reaction is the rate determining step and that the surface of
the catalyst is empty.
Figure 9. Model fit to the experimental data when occur a variation
on (a) PO2/Pfur ratio (W/F=1.81 min mgcat mL-1) and (b) W/F ratio
(PO2/Pfur=20) in the extremes of the temperature work range (280°C
and 320°C), over VPOr1.0. Lines correspond to model prediction for
MA and furfural partial pressures at the exit of the reactor at 280°C
and 320°C.
―풓풄 푴푨 = 풌풄 푴푨 푷푴푨 푷푶
(9)
ퟐ
The utilization of only one reaction temperature value prevents the
obtaining of activation energies. However, it allowed to compare the
kinetic constants for both VPOr1.0 and V2O5/Al2O3. Table
4
summarizes the kinetic constants for furfural partial and complete
oxidation, as well as for MA complete oxidation, furthermore show
the errors between the calculated and experimental partial
pressures for both compounds.
Conclusions
Two preparation methods, based on aqueous (VPAq) and organic
(VPOr) media, were used to synthesize bulk vanadium phosphorous
oxide (V-P-O) catalysts with different P/V molar ratios (0.43, 0.85,
and 1.00). The preparation method exhibited a strong influence on
the structure of the catalysts. VPOr catalysts present a more
crystalline structure with a marked presence of crystalline (VO)2P2O5
phase, considered responsible for the selective MA formation. On
the other hand, an amorphous phase was detected in the VPAq
catalysts, with a predominance of V(V), reported as a non-selective
site for furfural partial oxidation to MA. The differences in the
structure were reflected in the catalytic performance, being VPOr
catalyst more selective for the partial oxidation of furfural to MA
than VPAq catalysts. Both series of V-P-O catalysts were stable under
the experimental conditions with no significant deactivation obser-
The k(V2O5/Al2O3): k(VPOr1.0) ratio for furfural partial oxidation
indicate that the furfural partial oxidation is 50.5 times faster over
the supported catalyst. Similarly, the furfural complete oxidation is
16.9 faster indicating that this catalyst favor the unselective
oxidation as well.
Figure 10 represent the model fit to the experimental data obtained
over the V2O5/Al2O3 catalyst at 280°C as a function of the change in
PO2/Pfur ratio (Fig. 10a) and W/F ratio (Fig. 10b). The model was
capable to represent the change in furfural and MA partial pressure
as function of PO2/Pfur ratio and W/F ratio over this catalyst as well.
Similarly to what happened with VPOr1.0, the higher deviation occur
at the extreme conditions evaluated where the assumptions made
for propose the model become less accurate.
10 | Catal. Sci. Technol., 2021, 00, 1-3
This journal is © The Royal Society of Chemistry 2021
Please do not adjust margins

Page 11 of 13
Catalysis Science & Technology
Please do not adjust margins
Catalysis Science & Technology
ARTICLE
Hinshelwood in which the rate determining step of both main
View Article Online
DOI: 10.1039/D1CY01060C
reactions (partial and total oxidations) corresponds to the surface
reaction of furfural and O2. The model accurately predicts the
furfural consumption and MA production with a mean error of
16.3%. Moreover, the surface of the catalysts seems not to be
occupied significantly by surface intermediates which is in line with
previous reports. The apparent activation energy for furfural partial
oxidation reaction is higher than for complete oxidation (156 kJ/mol
vs 90 kJ/mol) which agrees with the increase in STY of MA at higher
oxidizing potential for the studied V-P-O catalyst.
Conflicts of interest
There are no conflicts to declare.
Acknowledgements
CONICYT PIA/APOYO CCTE AFB170007 and ANID Millennium Science
Initiative grant NCN17_040 are acknowledged for financing this
work. Saint-Gobain NorPro is acknowledged for kindly providing the
alumina support.
References
1
S. Shi, H. Guo and G. Yin, Catal. Commun., 2011, 12, 731–
733.
2
3
X. Li, P. Jia and T. Wang, ACS Catal., 2016, 6, 7621–7640.
Y. Rodenas, R. Mariscal, J. L. G. Fierro, D. Martín Alonso, J.
A. Dumesic and M. López Granados, Green Chem., 2018,
20, 2845–2856.
M. Müller, M. Kutscherauer, S. Böcklein, G. Mestl and T.
Turek, Chem. Eng. J., 2020, 401, 126016.
M. Müller, M. Kutscherauer, S. Böcklein, G. Mestl and T.
Turek, Ind. Eng. Chem. Res., 2021, 60, 218–229.
K. C. Waugh and Y. H. Taufiq-Yap, Catal. Today, 2003, 81,
215–225.
G. S. Patience and M. J. Lorences, Int. J. Chem. React. Eng.,
2006, 4, 1–18.
G. Busca, F. Cavani, G. Centi and F. Trifirò, J. Catal., 1986,
99, 400–414.
S. K. Bej and M. S. Rao, Ind. Eng. Chem. Res., 1991, 30,
1829–1832.
M. J. Lorences, G. S. Patience, R. Cenni, F. Díez and J. Coca,
Catal. Today, 2006, 112, 45–48.
S. K. Bej and M. S. Rao, Ind. Eng. Chem. Res., 1991, 30,
1819–1824.
S. K. Bej and M. S. Rao, Appl. Catal. A, Gen., 1992, 83, 149–
163.
J. S. Buchanan and S. Sundaresan, Appl. Catal., 1986, 26,
211–226.
Z. Y. Xue and G. L. Schrader, J. Catal., 1999, 184, 87–104.
M. J. Lorences, G. S. Patience, F. V. Díez and J. Coca, Ind.
Eng. Chem. Res., 2003, 42, 6730–6742.
J. Gascón, R. Valenciano, C. Téllez, J. Herguido and M.
Menéndez, Chem. Eng. Sci., 2006, 61, 6385–6394.
Figure 10. Model fit to the experimental data when occur a variation
on (a) PO2/Pfur ratio (W/F=1.81 min mgcat mL-1) and (b) W/F ratio
(PO2/Pfur=20) at 280°C, over V2O5/Al2O3. Lines correspond to model
prediction for MA and furfural partial pressures at the exit of the
reactor at 280°C.
4
5
6
ved. A higher P/V molar ratio led to a decrease in the specific surface
area, particularly on VPAq. Besides, it causes a loss of crystallinity in
the VPAq solids structure. In the VPOr catalysts, the increase in P/V
ratio changes the crystallite size of (VO)2P2O5 differently according to
the plane ((200), (024), (032)), which means that the P/V ratio does
not affect all planes equally. Future work should study lower P/V
molar ratios in order to better understand the influence of this
parameter on the selective oxidation of furfural.
7
8
9
10
11
12
13
Comparatively, a supported V2O5/Al2O3 catalyst was more active for
furfural conversion (50.5 times faster than VPOr1.0 at 280°C),
however, the selectivity decreased at higher reaction temperature
and PO2/Pfur ratios due to its also higher rates of total oxidation paths.
On the contrary, the bulk VPOr1.0 catalyst showed a constant increase
of activity and selectivity to MA at 320°C and PO2/Pfur=40. This result
indicates that the bulk VPOr1.0 catalyst is stronger and more resistant
to higher oxidizing potential with no significant loss in MA yield than
the V2O5/Al2O3 catalyst. Kinetic modelling was used to estimate
fundamental kinetic parameters for both, partial and complete
furfural oxidation. The experimental data was fitted to a Langmuir-
14
15
16
This journal is © The Royal Society of Chemistry 2021
Catal. Sci. Technol., 2021, 00, 1-3 | 11
Please do not adjust margins

Please do not adjust margins
Catalysis Science & Technology
Page 12 of 13
ARTICLE
Catalysis Science & Technology
17
18
19
20
R. Mariscal, P. Maireles-Torres, M. Ojeda, I. Sádaba and M.
43
44
45
46
M. J. Lorences, G. S. Patience, F. V. Díez and J. Coca, Appl.
Catal. A Gen., 2004, 263, 193–202.
D. Wang and M. A. Barteau, Appl. Catal. A Gen., 2002, 223,
205–214.
M. S. Murthy and Rajamani, Chem. Eng. Sci., 1974, 29, 601–
609.
K. Rajamani, P. Subramanian and M. S. Murthy, Ind. Eng.
Chem. Process Des. Dev., 1976, 15, 232–234.
E. Bordes and P. Courtine, J. Catal., 1979, 57, 236–252.
B. Chen and E. J. Munson, J. Am. Chem. Soc., 2002, 124,
1638–1652.
F. Cavani, D. De Santi, S. Luciani, A. Löfberg, E. Bordes-
Richard, C. Cortelli and R. Leanza, Appl. Catal. A Gen., 2010,
376, 66–75.
B. K. Hodnett, Catal. Rev. - Sci. Eng., 1985, 3, 373–424.
M. F. De Lange, T. J. H. Vlugt, J. Gascon and F. Kapteijn,
Microporous Mesoporous Mater., 2014, 200, 199–215.
S. Brunauer, P. H. Emmett and E. Teller, J. Am. Chem. Soc.,
1938, 60, 309–319.
E. P. Barrett, L. G. Joyner and P. P. Halenda, J. Am. Chem.
Soc., 1951, 73, 373–380.
K. K. Unger, Porous silica—Its properties and use as support
in column liquid chromatography, Elsevier Scientific
Publishing Company, Amsterdam, 1st Editio., 1979, vol.
179.
H. Marsh and F. Rodríguez-Reinoso, Activated Carbon,
Elsevier B.V, 1st edn., 2006.
M. Thommes, K. Kaneko, A. V. Neimark, J. P. Olivier, F.
Rodriguez-Reinoso, J. Rouquerol and K. S. W. Sing, Pure
Appl. Chem., 2015, 87, 1051–1069.
N. H. Batis, H. Batis, A. Ghorbel, J. C. Vedrine and J. C.
Volta, J. Catal., 1991, 128, 248–263.
W. Weng, R. Al Otaibi, M. Alhumaimess, M. Conte, J. K.
Bartley, N. F. Dummer, G. J. Hutchings and C. J. Kiely, J.
Mater. Chem., 2011, 21, 16136–16146.
X. Li, J. Ko and Y. Zhang, ChemSusChem, 2018, 11, 612–618.
A. Riaz, M. U. Ali, T. G. Enge, T. Tsuzuki, A. Lowe and W.
Lipiński, Research, 2020, 2020, 1–12.
K. Inumaru, M. Misono and T. Okuhara, Appl. Catal. A Gen.,
1997, 149, 133–149.
View Article Online
DOI: 10.1039/D1CY01060C
López Granados, Energy Environ. Sci., 2016, 9, 1144–1189.
N. R. Reed and E. S. C. Kwok, ed. P. B. T.-E. of T. (Third E.
Wexler, Academic Press, Oxford, 2014, pp. 685–688.
G. Shen, B. Andrioletti and Y. Queneau, Curr. Opin. Green
Sustain. Chem., 2020, 26, 100384.
P. L. Arias, J. A. Cecilia, I. Gandarias, J. Iglesias, M. López
Granados, R. Mariscal, G. Morales, R. Moreno-Tost and P.
Maireles-Torres, Catal. Sci. Technol., 2020, 10, 2721–2757.
N. Alonso-Fagúndez, I. Agirrezabal-Telleria, P. L. Arias, J. L.
G. Fierro, R. Mariscal and M. L. Granados, RSC Adv., 2014,
4, 54960–54972.
A. C. Alba-Rubio, J. L. G. Fierro, L. León-Reina, R. Mariscal, J.
A. Dumesic and M. López Granados, Appl. Catal. B Environ.,
2017, 202, 269–280.
T. Soták, M. Hronec, M. Gál, E. Dobročka and J. Škriniarová, 51
Catal. Letters, 2017, 147, 2714–2723.
47
48
21
22
49
50
23
24
25
26
27
28
29
30
P. M. Malibo, P. R. Makgwane and P. G. Baker,
ChemistrySelect, 2020, 5, 6255–6267.
A. Hu, X. Wang, X. Wang, Q. Peng and H. Wang, J. Theor.
Comput. Chem., 2020, 19, 1–13.
52
53
54
R. Gérardy, D. P. Debecker, J. Estager, P. Luis and J. C. M.
Monbaliu, Chem. Rev., 2020, 120, 7219–7347.
N. Alonso-Fagúndez, M. Ojeda, R. Mariscal, J. L. G. Fierro
and M. López Granados, J. Catal., 2017, 348, 265–275.
N. Pethan Rajan, G. S. Rao, V. Pavankumar and K. V. R.
Chary, Catal. Sci. Technol., 2014, 4, 81–92.
N. Alonso-Fagúndez, M. L. Granados, R. Mariscal and M.
Ojeda, ChemSusChem, 2012, 5, 1984–1990.
F. Ivars-Barceló, G. J. Hutchings, J. K. Bartley, S. H. Taylor, P.
Sutter, P. Amorós, R. Sanchis and B. Solsona, J. Catal.,
2017, 354, 236–249.
55
56
57
58
31
32
33
34
35
P. Santander, L. Bravo, G. Pecchi and A. Karelovic, Appl.
Catal. A Gen., 2020, 595, 117513.
B. M. Weckhuysen and D. E. Keller, Catal. Today, 2003, 78,
25–46.
59
60
F. Ben Abdelouahab, R. Olier, N. Guilhaume, F. Lefebvre
and J. C. Volta, J. Catal., 1992, 134, 151–167.
J. C. Volta, R. Olier, M. Roullet, F. B. Abdelouahab and K.
Bere, Stud. Surf. Sci. Catal., 1993, 75, 1531–1534.
C. J. Kiely, A. Burrows, S. Sajip, G. J. Hutchings, M. T.
Sananes, A. Tuel and J. C. Volta, J. Catal., 1996, 162, 31–47.
E. Bordes, Catal. Today, 1993, 16, 27–38.
V. V. Guliants, J. B. Benziger, S. Sundaresan, I. E. Wachs, J.
M. Jehng and J. E. Roberts, Catal. Today, 1996, 28, 275–
295.
61
62
G. Ertl, H. Knözinger, F. Schüth and J. Weitkamp, Eds.,
Handbook of Heterogeneous Catalysis, WILEY-VCH,
Second., 2008.
Y. H. Taufiq-Yap, C. S. Yuen, N. Mohamed Nurul Suziana
and R. Irmawati, Cuihua Xuebao/Chinese J. Catal., 2014, 35,
270–276.
A. Iglesias-Juez, M. V. Martínez-Huerta, E. Rojas-García, J.
M. Jehng and M. A. Banares, J. Phys. Chem. C, 2018, 122,
1197–1205.
36
37
63
64
38
M. Rezaei, A. Najafi Chermahini, H. A. Dabbagh, M. Saraji
and A. Shahvar, J. Environ. Chem. Eng., ,
DOI:10.1016/j.jece.2018.102855.
39
40
41
42
L. M. Cornaglia and E. A. Lombardo, Appl. Catal. A, 1995,
127, 125–138.
H. S. Horowitz, C. M. Blackstone, A. W. Sleight and G.
Teufer, Appl. Catal., 1988, 38, 193–210.
N. P. Rajan, G. S. Rao, B. Putrakumar and K. V. R. Chary, RSC 67
Adv., 2014, 4, 53419–53428.
Y. Zhang-Lin, M. Forissier, J. C. Volta and J. C. Védrine, J.
Catal., 1994, 145, 267–275.
65
66
Y. H. Taufiq-Yap, N. M. Nurul Suziana and M. Z. Hussein,
Catal. Letters, 2011, 141, 136–148.
A. A. Rownaghi, Y. H. Taufiq-Yap and F. Rezaei, Catal.
Letters, 2009, 130, 504–516.
L. Pérez-Moreno, S. Irusta, J. Soler, J. Herguido and M.
Menéndez, Chem. Eng. J., 2009, 147, 330–335.
V. Mahdavi and H. R. Hasheminasab, J. Taiwan Inst. Chem.
Eng., 2015, 51, 53–62.
68
12 | Catal. Sci. Technol., 2021, 00, 1-3
This journal is © The Royal Society of Chemistry 2021
Please do not adjust margins

Page 13 of 13
Catalysis Science & Technology
Please do not adjust margins
Catalysis Science & Technology
ARTICLE
69
70
71
72
73
74
75
X. Feng, Y. Yao, Q. Su, L. Zhao, W. Jiang, W. Ji and C. T. Au,
View Article Online
DOI: 10.1039/D1CY01060C
Appl. Catal. B Environ., 2015, 164, 31–39.
A. Martin, C. Janke and V. N. Kalevaru, Appl. Catal. A Gen.,
2010, 376, 13–18.
N. A. Milas and W. L. Walsh, J. Am. Chem. Soc., 1935, 57,
1388–1393.
V. A. Slavinskaya, D. R. Kreile, e. E. Dzilyuma and D. e. Sile,
Chem. Heterocycl. Compd., 1977, 13, 710–721.
M. Rogojerov, G. Keresztury and B. Jordanov, Spectrochim.
Acta - Part A Mol. Biomol. Spectrosc., 2005, 61, 1661–1670.
P. Schneider, G. Emig and H. Hofmann, Ind. Eng. Chem.
Res., 1987, 26, 2236–2241.
S. K. Bej and M. S. Rao, Ind. Eng. Chem. Res., 1991, 30,
1824–1828.
This journal is © The Royal Society of Chemistry 2021
Catal. Sci. Technol., 2021, 00, 1-3 | 13
Please do not adjust margins