ΑII-(V1-xWx)OPO4 catalysts for the selective oxidation of n-butane to maleic anhydride

Article abstract of DOI:10.1016/j.cattod.2018.05.040

The vanadyl pyrophosphate (VPP) based catalyst is unique in converting n-butane selectively (60–70%) into maleic anhydride (MAN), whereas a MAN selectivity of 20% may be regarded as high for structurally different catalyst systems. We present novel vanadium phosphorus oxides and mixed metal phosphate solid solutions tested for n-butane oxidation to MAN with a selectivity of >30%. The majority of the catalysts were prepared by solution combustion synthesis. (V_1-xW_x)OPO₄ with α_II structure was found to be more active and selective in the oxidation of n-butane compared to β-VOPO₄. By adjusting the tungsten content the oxidation state of vanadium in (V_1-xW_x)OPO₄ can be tuned between 4.74 and 4.99, which is regarded as a key factor for MAN production. All catalysts were structurally stable, but the specific surface area increased during the reaction, as detected by X-ray diffraction and N₂ physisorption, respectively. (V_1-xMo_x)OPO₄ was also stable, but the MAN selectivity was lower compared to β-VOPO₄. Low conversions result from the low surface area of the screening samples, however, could be overcome by advanced synthesis protocols.

Full text of DOI:10.1016/j.cattod.2018.05.040

Accepted Manuscript
Title: αII-(V1−xWx)OPO4 catalysts for the selective oxidation
of n-butane to maleic anhydride
Authors: C. Schulz, S.C. Roy, K. Wittich, R. Naumann
d’Alnoncourt, S. Linke, V. Strempel, B. Frank, R. Glaum, F.
Rosowski
PII:
S0920-5861(18)30640-0
DOI:
Reference:
https://doi.org/10.1016/j.cattod.2018.05.040
CATTOD 11470
To appear in:
Catalysis Today
Received date:
Accepted date:
19-3-2018
21-5-2018
Please cite this article as: Schulz C, Roy SC, Wittich K, d’Alnoncourt RN, Linke
S, Strempel V, Frank B, Glaum R, Rosowski F, αII-(V1-xWx)OPO4 catalysts for
the selective oxidation of n-butane to maleic anhydride, Catalysis Today (2018),
https://doi.org/10.1016/j.cattod.2018.05.040
This is a PDF file of an unedited manuscript that has been accepted for publication.
As a service to our customers we are providing this early version of the manuscript.
The manuscript will undergo copyediting, typesetting, and review of the resulting proof
before it is published in its final form. Please note that during the production process
errors may be discovered which could affect the content, and all legal disclaimers that
apply to the journal pertain.

αII-(V1-xWx)OPO4 catalysts for the selective oxidation of n-
butane to maleic anhydride
C. Schulza, S. C. Royb, K. Wittichb, R. Naumann d’Alnoncourta, S. Linkea,d, V. Strempela,d,
B. Franka,*, R. Glaumb,*, F. Rosowskia,c
a) BasCat - UniCat BASF JointLab, Technische Universität Berlin, 10623 Berlin, Germany
b) Institute for Inorganic Chemistry, Rheinische Friedrich-Wilhelms-Universität, 53121 Bonn, Germany
c) BASF SE, Process Research and Chemical Engineering, 67056 Ludwigshafen, Germany
d) Present address: BASF SE, Process Research and Chemical Engineering, 67056 Ludwigshafen, Germany
*) corresponding authors: BF: b.frank@bascat.tu-berlin.de; RG: rglaum@uni-bonn.de
Graphical Abstract
aII-V1-xMxOPO4 (M=W,Mo)

Highlights




solution combustion synthesis (SCS) leads to a variety of phase pure catalysts
doping with Mo and W tunes the oxidation state of V in VOPO4 between 4+ and 5+
αII-(V1-xWx)OPO4 reaches selectivities of maleic anhydride of more than 30%
αII type solid solutions are a flexible platform for further research
Highlights

A variety of phase pure catalysts for the selective oxidation of n-butane to maleic anhydride can be prepared via solution
combustion synthesis (SCS)


By doping with Mo and W the average oxidation state of V in VOPO4 can be tuned between 4+ and 5+
αII-(V1-xWx)OPO4 reaches selectivities of maleic anhydride of more than 30%, which is hardly achieved with other non-VPP
based catalysts
•
αII type solid solutions are a flexible platform for further research in selective oxidation catalysis
The vanadyl pyrophosphate (VPP) based catalyst is unique in converting n-butane selectivity (60-70%)
into maleic anhydride (MAN), whereas a MAN selectivity of 10-20% may be regarded as high for
structurally different catalyst systems. We present novel vanadium phosphorous oxides and mixed
metal phosphate solid solutions tested for n-butane oxidation to maleic anhydride with a selectivity of
> 30%. The majority of the catalysts were prepared by solution combustion synthesis. (V1-xWx)OPO4
with II structure was found to be more active and selective in selective oxidation of n-butane
compared to -VOPO4. The oxidation state of vanadium in (V1-xWx)OPO4 can be tuned between 4.74
and 4.99 by the tungsten content, which is regarded as the key for MAN production. All catalysts were
structurally stable, but the specific surface area increased during the reaction, as detected by X-ray
diffraction and N2 physisorption, respectively. (V1-xMox)OPO4 were also stable, but the selectivity for
MAN was lower compared to -VOPO4. Low conversions result from low surface areas in the
screening samples, however, could be overcome by advanced synthesis protocols.

Key words: butane, maleic anhydride, selective oxidation, VPO, vanadium phosphorous oxide, solid
solution, mixed metal phosphates
1. Introduction
The catalytic oxidation of n-butane to maleic anhydride (MAN) is the first and so far only industrial
process of selective alkane oxidation and has an annual world capacity of more than 1 million tons [1].
Although the molar yield of the industrial process has been steadily improved over the past decades,
the yields are still below 65%, which is mainly limited by product-selectivity [1–9]. The industrial state
of the art catalyst is based on crystalline (VIVO)2P2O7 (vanadyl(IV) pyrophosphate, VPP). Due to
several vanadium(V) orthophosphate by-phases, which are formed under reaction conditions, the
catalyst is often referred to as vanadium phosphorous oxide (VPO). Significant improvements in
activity and (by-)product selectivity with this catalyst system were achieved by increasing specific
surface areas and by adding promotors like Nb [10], Co [11], Fe [12], and Sm [13]. As the extensive
research on VPP-based systems missed a significant breakthrough within the past decades, a new
perspective on this reaction is expected from the search for new materials rather than improving the
VPO catalyst. In the literature the number of materials selectively converting n-butane to MAN is limited
to VPO [14–21]. They can be subdivided into a) vanadyl(IV) pyrophosphate and b) vanadyl(V)
orthophosphate with different structure types as I-VOPO4, II-VOPO4, -VOPO4, -VOPO4, and -
VOPO4 [22]. All of them are well discussed in the literature with regard to their catalytic performance
in n-butane oxidation as a by-phase in VPO [23–27].
Less effort was put into synthesis and catalytic testing of single phase bulk VOPO4 catalysts. First
results published by Shimoda et al. showed that single phase II-VOPO4 and -VOPO4 can produce
maleic anhydride from n-butane. Starting from (NH4)2((VO)2C2O4(HPO4)2)·5H2O or NH4HVPO6 single
phase II-VOPO4 and -VOPO4 were synthesized. Selectivities of MAN ranged from 14% to 20% [23].

Under reaction conditions most phases are stable and no indication for by-phases were found by in
situ Raman studies [27].
On the search for novel materials with structural motifs similar for those of VOPO4 and VPP were
targeted. The crystal structure of VPP consists of a chain-ladder-motif of edge- and corner-sharing
VO6-octahedra, which are connected to pyrophosphate anions. Similar motifs are present in II-VOPO4.
Here, the vanadium oxygen polyhedra are only connected by vertices. Pyrophosphate groups are
replaced by orthophosphate anions (Figure 1).
Recently, Roy et al. reported a new class of VIV/VV containing mixed metal vanadium tungsten
phosphate solid solutions. Solution combustion synthesis [30,31] starting from vanadium and tungsten
precursors in nitric acid yields solid solutions V1-xWxOPO4 (0.04 < x < 0.26), which crystallize with II-
VOPO4- and -VOPO4-type structure (0 < x < 0.01), respectively [29]. XPS studies showed that the
substitution of VV by VIV/WVI in the II-VOPO4 lattice sensitively affects the average oxidation state of
V, which can be finely adjusted between 4.74 and 4.96 (II-VOPO4-type) and 4.99 and 5.00 (-VOPO4-
type). NAP XPS showed that on the surface of VPO, VIV and VV are present with an average oxidation
state of 4.3 [32]. The presence of VV on the surface is crucial as it can be correlated with production
of MAN [33]. Therefore, the introduction of bulk catalysts with a tunable vanadium oxidation state is a
promising approach.
In the present study, triggered by this finding, further mixed metal (V, W, Mo, Fe) oxide phosphates in
various oxidation states were synthesized and tested for their performance in the selective oxidation
of n-butane. In particular, the substitution of V by Mo is of interest as Mo is a frequent constituent of
many catalysts for hydrocarbon (amm)oxidation [34,35].
2. Experimental
2.1. Catalyst Preparation
Single phase solid solutions (V1-xWx)OPO4 (II- and -VOPO4-type), (V1-xMox)OPO4 (- and -VOPO4-
type), W0.67Fe0.33OPO4 (WOPO4-type), Fe(WO2)2(P2O7)(PO4), and “(V0.95Mo0.05)2P2O7“ were obtained

by solid state synthesis or the solution combustion technique [29,36,37]. Single phase VPP was
obtained by reducing -VOPO4 in Ar (p(O2) = 10-5 mbar) at 750 °C for 2 days. Polycrystalline WOPO4
was synthesized from W2P2O10, WO3, and P4 in sealed silica tubes at 1000 °C for 7 days [38]. The
reference catalyst VPO was synthesized according to patent via the organic synthesis route using iso-
butanol [12]. Finally, the reduction of -VOPO4 in H2, which was saturated with H2O in a bubbler at
ambient temperature (psat (H2O) = 0.023 bar), at 500 °C for 1 day yields the thermodynamic metastable
phase VPO4-m1, which is structurally related to -VOPO4 [37].
The fine powders of the original catalyst samples were pelletized in a lab press (2.5 min, pressure
equivalent to a weight of 2 t on a 12 mm tablet), respectively. Afterwards the pellets were crushed in
an agate mortar and a 100-200 m particle size fraction was obtained using analytical sieves made of
stainless steel. The estimation of the Weisz-Prater criterion as well as previous pre-tests with more
active catalysts indicate that this sieve fraction is appropriate to exclude mass and heat transfer
limitations under the reaction conditions chosen.
2.2. Catalyst Characterization
Routine XRPD analyses (IP-Guinier technique [39]) were performed for phase identification and purity
control. The patterns were recorded at ambient temperature using an image plate Guinier camera
(HUBER G670, Cu-Kα1 radiation, λ = 154.059 pm) with integrated read-out system [40]. The finely
ground powder samples were spread on Mylar foil (Fluxana GmbH & Co. KG, type TF-160). Powder
pattern were recorded for 15 min in the range of 4° ≤ 2θ ≤ 100°. The cell parameters were determined
with the program SOS [41], α-SiO2 (p. A., Merck) was used as an internal standard. Scherrer diameters
(D) were calculated according to equation 1 with a Scherrer constant (K) of 0.94 for  (FWHM) of
spherical crystals with cubic symmetry. The FWHM  was corrected for instrumental broadening with
Y2O3 standard.
Kλ
D =
(1)
ω cos θ

BET measurements were carried out in a volumetric N2 physisorption setup (Autosorb-6-B,
Quantachrome) at the temperature of liquid N2. The sample was degassed in a dynamic vacuum at
423 K for 2 h prior to physisorption. Full adsorption-desorption isotherms were recorded. The specific
surface area according to the BET method was calculated from the linear range of the adsorption
isotherm (p/p0=0.05-0.3). All catalysts were characterized before and after catalytic testing (Table 1).
2.3. Catalyst Test Setup
Catalytic tests were performed in a fully automated, commercial 8-channel parallel testing setup (hte
GmbH, Figure 2). It consists of a gas dosing unit, a reactor unit, an analytical unit. The gas dosing unit
provides mixed feeds comprising N2, O2 (0-25 %), n-butane (0-10 %), Ar (as an internal standard),
and H2O (0-60 %). The typical range of gas hourly space velocities (GHSV) is 500-5000 h-1. The test
unit consists of 8 reactors for catalyst samples and one blank reactor filled with inert material (calcined
corundum, 400-700 µm, CeramTec). Catalyst volumes of 1 mL were placed inside stainless steel
reactors with inner diameter of 10 mm. Stainless steel thermowells (OD=1/8”) are placed in the center
of all reactors, which are used to guide thermocouples into the catalyst beds. The reactor temperature
can be individually controlled in the range of 250-550 °C. The product gases are analyzed by two gas
chromatographs (GC 7890A, Agilent). The effluent gas of the blank reactor, which corresponds to the
inlet gas of all reactors, is also analyzed. Prior to the GC analysis, all effluent gas streams are diluted
with hot inert gas to avoid the condensation of organic products or steam. Transfer lines between
reactors and gas chromatographs are completely heated (200 °C). The GCs are optimized for the
analysis of the product spectrum of the oxidation of n-butane to MAN.
The conversion (X) of n-butane (Eqs. 1 and 3) and the selectivities (S) of the respective products are
calculated from the concentrations measured with reference to the internal standard Ar (Eqs. 2 and 4).
The latter are averaged over 5 runs measured at each temperature step, respectively. The selectivities
of the (by)products (Eq. 2) were normalized to the number of carbon atoms (Nc) in the (by)products

and n-butane, respectively. Depending on the catalyst, byproducts include butenes, acrylic acid,
propionic acid, acetic acid, acrolein, acetaldehyde, propane, propene, acetylene, ethylene, ethane,
carbon monoxide, and carbon dioxide. There are no unidentified peaks in the gas chromatograms. At
low levels of n-butane conversion, the product-based analysis of selectivities and conversions (Eqs. 3
and 4) is preferred.
2.4. Catalytic Testing
All catalysts were tested in groups of 8 samples following the same testing protocol. Catalyst volumes
of 1 mL were kept in their position in the reactor tubes by two layers of inert corundum particles,
respectively. The GHSV was fixed at 2000 h-1. The catalysts were ramped at 2 K min-1 in a lean air
mix of 5 % O2/N2 to 250 °C. At 250 °C the feed gas was changed to 2 % n-butane, 20 % O2, 3 % H2O,
3 % Ar with N2 balance and subsequently the reactor temperature was further increased to 375 °C.
After an initial equilibration period of 30 min the catalytic measurements were started. The standard
testing procedure includes 5 GC analyses per reactor and temperature with the reactor temperature
ramping up in 25 K steps up to a maximum temperature of 450 °C, followed by ramping down to
400 °C. Finally, the reactors were cooled down to ambient temperature in 5% O2/N2. Thus, information
on activity, selectivity and thermal stability/deactivation of the catalysts is obtained. One testing cycle
takes 8 days including the (dis)charging and cleaning of the reactors.
3. Results and Discussion
According to crystal structure and chemical composition of the catalysts, the results of characterization
and catalytic testing are collected in two categories:
3.1. Vanadium Phosphorous Oxides
The group of vanadium phosphorous oxides investigated in our study comprises vanadium(III)
orthophosphate VPO4-m1 and vanadyl(IV) pyrophosphate (VO)2P2O7. VPO4-m1 is a new polymorph

of VPO4 and a promising model catalyst with V in the low oxidation state +III and absence of vanadyl
oxygen atoms. The latter, however, are often discussed as crucial functional groups for the activation
of n-butane [3,42]. XRPD analysis of the as-prepared sample clearly identifies “single phase” VPO4-
m1 (Figure 3). VPO4-m1 and its structure are part of ongoing investigations and not clarified to its full
extend. Therefore, it is unclear if the additional reflections in the XRPD pattern indicate a
superstructure or an adphase. The main reflections of its XRPD pattern are assigned on the basis of
the -VOPO4 structure (Pnma, with z = 4, lattice parameters a = 7.7863(5) Å , b = 6.1329(3) Å,
c = 6.9673(5) Å). Under reaction conditions VPO4-m1 is oxidized to (VO)2P2O7, -VOPO4. This
oxidation is accompanied by a more than two-fold increase of the specific surface area (Table 2). In
parallel, the conversion of n-butane as well as the selectivity for MAN steadily increases during the
reaction. Likely the significant boost of activity can be correlated to both the increased surface area
and the formation of highly active (VO)2P2O7 and -VOPO4, respectively. This is confirmed by the final
performance matching the reference VPO from solution combustion synthesis with regard to the
selectivity for MAN (Table 2).
In principle (VO)2P2O7 can be synthesized via two different routes [43–49]. In the aqueous route V2O5
is reduced by NH2OH in the presence of H3PO4 yielding highly crystalline (VO)2P2O7 after calcination
of the VOHPO4·0.5H2O precursor [44,48–52]. Instead, the organic route uses iso-butanol as the
reducing agent [44,53,54]. Here, V2O5, H3PO4, and iso-butanol are refluxed for several hours and
converted to VOHPO4·0.5H2O, which similarly forms (VO)2P2O7 in the calcination process. However,
the crystallinity of the “organic” material is significantly lower as shown by XRPD measurements, and
typically leads to a higher surface area enhancing catalyst performance [44]. Single-phase (VO)2P2O7
with a high crystallinity can also be synthesized by the reduction of -VOPO4 in Ar at 750 °C, as done
in the present study. The selectivity for MAN obtained over this material is lower compared to the
industrial reference system VPO from the organic route (Table 2). However, it is in the same range as
VPO catalysts synthesized via the water route with a P:V ratio of 1:1 [55].

For (VO)2P2O7 the effect of Mo doping was tested. The sample with 5 % of V substituted by Mo
(V0.95Mo0.05O)2P2O7 was prepared by solution combustion synthesis (SCS). The influence on
crystallinity and specific surface area is negligible (Table 2). However, in terms of catalyst performance
the surface area-corrected conversion of n-butane normalized is similar as compared to the undoped
material, whereas the selectivity for MAN is reduced to 30 %. Selectivity loss is mainly contributed to
formation of acrylic(S(acrylic acid)=17%) and acetic acid (S(acetic acid)=10%). The formation of COx
is also reduced, which indicates on the one hand inhibition of total oxidation of n-butane or MAN by
Mo. On the other hand Mo enhances C-C cleavage of reaction intermediates. Since the amount of Mo
is small and the results match the undoped material quiet well it cannot be ruled out that Mo is not
incorporated at all.
3.2. Mixed Metal Phosphate Solid Solutions
The most extensively investigated sub-category within this class is based on the substitution of V by
W according to the formula (V1-xWxOPO4) [29]. XRPD patterns of these materials indicate single
phases ( for x ≤ 0.01, II for 0.04 ≤ x ≤ 0.26) with a mean crystallite size in the range of 37 (x = 0.2)
to 54 nm (x = 0.06). These phases are stable and neither phase transformation nor formation of an
adphase was observed after the catalytic testing (Figure 4).
The BET surface area of the as-prepared samples, which are generally very low in the range of 1–4
m2 g-1, occasionally changed during the catalytic process (Table 3). An almost doubled specific surface
area as well as activation is observed for the tungsten-rich V0.8W0.2OPO4. On the one hand our results
could indicate a dramatic reconstruction of the morphologies of particle and surface during the catalytic
process. On the other hand pores might be blocked by residual organic compounds from synthesis.
Calcination of all samples at higher temperatures compared to reaction temperatures should be
sufficient for carbon removal from synthesis. However, the effect on surface area is small. With
decreasing W content within these solid solution-type catalysts the effect of reaction conditions on the
specific surface area is reduced, which is almost absent on V0.99W0.01OPO4 and even negative

(reduction of SBET) for the tungsten-free VOPO4 catalyst. Instead, the industrial reference VPO, owing
to elaborated synthesis protocols, has an approximately 10-fold higher and stable specific surface
area of 26 m2 g-1, which is retained after the catalytic test.
Catalytic tests of samples of compositions V1-xWxOPO4 show a remarkably high selectivity for MAN in
the range of 20-31% (Figure 5). XPS studies showed that W5+ is oxidized to W6+ in V1-xWxOPO4
reducing an equimolar amount of V5+ to V4+ [29]. The comparison of V0.8W0.2OPO4 and VOPO4 clearly
shows an increased selectivity of MAN, which might be attributed to the reduced (average) oxidation
state of vanadium [29]. Activation energy can be correlated with tungsten content or the amount of V4+.
Isotope labeling studies of n-butane showed that C-H activation of methylene groups is the rate
determining step and the activation is correlated with the presence of VV [6]. By substitution of
vanadium by tungsten, the oxidation state of vanadium is reduced, which explains the higher activation
energy. Nevertheless, a direct correlation between substituted V and selectivity for MAN is hardly
possible. Our results indicate a maximum selectivity near 20% substitution of V by W (V1-xWxOPO4
with x = 0.2). On the other hand, vanadium-free WOPO4 does not convert n-butane to maleic anhydride,
which underlines that W plays not an active role in the catalytic cycle. The change in surface area on
the course of the reaction, which is observed for several samples, is typically accompanied by a
corresponding change in activity, whereas selectivity for MAN remains constant. Notably, the activity
of most of these catalysts if referred to the specific surface area is comparable or even higher than for
the reference VPO. For instance, the most selective V0.8W0.2OPO4 provides a conversion and specific
surface area (after reaction) of 42% and 3.4 m2 g-1, however, 80% of butane conversion over VPP can
be considered mainly as a result of the superior surface area (26 m2 g-1) [12]. Thus, a high potential
for improvement of these catalysts is expected from the optimization of synthetic strategies.

The main by-products are COx. The organic by-products of n-butane oxidation over V1-xWxOPO4-type
materials comprise acrylic and acetic acids (< 8%), as well as traces of 1-butene, ethylene, furan, and
propene (< 1%).
Molybdenum is a common constituent of selective oxidation catalysts, such as MoVTeNb mixed oxides
for ammoxidation of propane to acrylonitrile, oxidation of propane to acrylic acid, or oxidative
dehydrogenation of ethane to ethylene[56–60]. Thus, Mo was in our focus for the substitution of V in
VOPO4. We tested the γ- and the β-phase with 20% substitution (V0.8Mo0.2OPO4), respectively (Figure
6)[61].
Neither phase changes, decomposition nor significant changes in specific surface area were observed
after exposure to reaction conditions (Table 2). However, the catalysts were nearly inactive, which
might be explained by the very low surface areas and/or by a negative effect of Mo. Compared to the
mixed-metal phosphates V1-xWxOPO4, the selectivity for MAN drops significantly. Similar to
(V0.95Mo0.05O)2P2O7 selectivity for acetic and acrylic acid is enhanced whereas CO and CO2 formation is
reduced. The surface-corrected net activity is also much lower than for the W-substituted vanadyl
phosphates. However, the level of selectivity for MAN of the mixed-metal phosphates V1-xWxOPO4 and
-VOPO4 is in good agreement with single phase -VOPO4 and -VOPO4 catalysts as shown by
6+
Shimoda.16 Finally, we tested the substitution of W5+ by (Fe3+
W 2/3) in WOPO4 (W1-xFexOPO4 with
1/3
x = 0.33). This sample, which was obtained by SCS, provides a relatively high specific surface area of
6.2 m2 g-1 but is unstable under reaction conditions. This instability under the testing conditions is
rather surprising since zero activity for n-butane activation indicates little interaction with the gas phase.
Although Fe [62] and W [63] are known as efficient promotors for selective oxidation catalysts, this is
further indication for the active role of V in the activation of n-butane as stated by Centi et al.[3]
4. Conclusion

A variety of catalysts were synthesized and tested for selective oxidation of n-butane based on
(VO)2P2O7 and VOPO4. Vanadium was substituted or replaced by other transition metals like tungsten,
molybdenum or iron. Substitution of V by W leads to V1-xWxOPO4 with II structure type, which is a
novel class of catalysts for selective oxidation of n-butane to MAN. The amount of tungsten can be
tuned in the range from 0.04 ≤ x ≤ 0.26. All catalysts are phase stable under reaction conditions,
although the specific surface area increases during the reaction, which needs to be studied in detail.
Reducing the oxidation state of vanadium increase the apparent activation energy for n-butane
conversion. Compared to -VOPO4, the selectivity for MAN could be increased from 20% up to 31%.
Nevertheless, V1-xWxOPO4 catalysts are less selective and active than VPO reference. However, it
could be shown that the oxidation state of vanadium influences the selectivity for MAN. Substitution of
V with Mo in (VO)2P2O7 and VOPO4 showed different results. In VOPO4, V by Mo can be easily
substituted leading to pure - and -phases, which are more stable under reaction conditions
compared to V1-xWxOPO4. In case of (V0.95Mo0.05O)2P2O7 there is no evidence in the XRPD that Mo is
incorporated in the (VO)2P2O7 crystal structure. In both cases, the effect on catalysis is comparable.
Selectivity for MAN and COx is reduced compared to the non-substituted catalysts. Further, addition
of Mo strongly influences the C-C cleavage leading to the formation products like acetic and acrylic
acid. Additionally, we could show that vanadyl pyrophosphate synthesized by solution combustion
synthesis is similar to VPO from aqueous routes. Differences in activity can be explained by low
surface areas. The reduced selectivity for MAN might be correlated with a 1:1 ratio of P:V since an
excess of phosphor enhances the selectivity. VPO4-m1 is a new polymorph of VPO, which was
introduced by removing the vanadyl oxygen atoms of -VOPO4. Under reaction conditions it is oxidized
to (VO)2P2O7 and -VOPO4. The performance is comparable to VPO by SCS and stresses the
importance of stable catalysts as well as the inflexibility of (VO)2P2O7. Complete substitution of V in
(VO)2P2O7 and VOPO4 with iron or tungsten yields phase-stable materials that are not active in the
selective oxidation of n-butane. In general, the combination of vanadium in the oxidation state +4 and
+5 and the ratio of vanadium and phosphate are crucial parameters for selective oxidation of n-butane
to MAN. Although, VPO is the best catalyst known, it has limited options for optimization due to the

inflexible crystal structure and limited effect of promotors [62]. Therefore, mixed transition metal VxM1-
xOPO4 might be a good starting point for novel catalysts with higher potential for improvement. It could
be shown that solid state synthesis methods like solution combustion synthesis might be a good way
to access mentioned new materials/structure motifs for catalysis. However, surface areas are quite
low. Consequently, other synthesis methods like hydrothermal routes should be considered as well as
supporting of the active catalyst. Transition metals should be selected based on their catalytic
performance in oxidation catalysis or by their potential for reduction of vanadium since recent NAP
XPS studies of VPO showed that under working conditions the average oxidation state of surface
vanadium is +4.3 [32].
From a mechanistic point of view, kinetics and reaction networks of the newly found mixed-metal
phosphates V1-xWxOPO4 should be analyzed and compared to VPO as well as -VOPO4, which could
be a crucial step towards understanding the active site. Especially, the surface composition and
oxidation state under reaction conditions needs to be analyzed in detail.
References
[1] N. Ballarini, F. Cavani, C. Cortelli, S. Ligi, F. Pierelli, F. Trifirò, C. Fumagalli, G. Mazzoni, T. Monti, Top.
Catal. 38 (2006) 147–156.
[2] F. Trifirò, R.K. Grasselli, Top. Catal. 57 (2014) 1188–1195.
[3] G. Centi, Catal. Today 16 (1993) 5–26.
[4] C.H. Bartholomew, R.J. Farrauto, Fundamentals of Industrial Catalytic Processes, John Wiley & Sons,
Inc, Hoboken, NJ, USA, 2005.
[5] J.-C. Volta, Comptes Rendus de l'Académie des Sciences - Series IIC - Chemistry 3 (2000) 717–723.
[6] G. Centi, F. Trifiro, J.R. Ebner, V.M. Franchetti, Chem. Rev. 88 (1988) 55–80.
[7] M.A. Pepera, J.L. Callahan, M.J. Desmond, E.C. Milberger, P.R. Blum, N.J. Bremer, J. Am. Chem. Soc.
107 (1985) 4883–4892.

[8] R. Glaum, C. Welker-Nieuwoudt, C.-K. Dobner, M. Eichelbaum, F. Gruchow, C. Heine, A. Karpov, R.
Kniep, F. Rosowski, R. Schlögl, S.A. Schunk, S. Titlbach, A. Trunschke, Chem. Ing. Tech. 84 (2012) 1766–
1779.
[9] M.-J. Cheng, W.A. Goddard, J. Am. Chem. Soc. 135 (2013) 4600–4603.
[10]
[11]
F. Ghelfi, G. Mazzoni, C. Fumagalli, F. Cavani, F. Pierelli, EP Patent (2004) EP1663482 (A1).
V.A. Zazhigalov, J. Haber, J. Stoch, A.I. Pyatnitzkaya, G.A. Komashko, V.M. Belousov, Appl. Catal.
A: Gen. 96 (1993) 135–150.
[12] C. Dobner, M. Duda, A. Raichle, H. Wilmer, F. Rosowski, M. Hoelzle, EP Patent (2006) EP1915210
(A1).
[13] H.-Y. Wu, H.-B. Wang, X.-H. Liu, J.-H. Li, M.-H. Yang, C.-J. Huang, W.-Z. Weng, H.-L. Wan, Appl. Surf.
Sci. 351 (2015) 243–249.
[14] H.-Y. Wu, P. Jin, Y.-f. Sun, M.-H. Yang, C.-J. Huang, W.-Z. Weng, H.-L. Wan, J. Mol. Catal. A: Chem.
414 (2016) 1–8.
[15]
[16]
[17]
[18]
[19]
[20]
[21]
[22]
[23]
[24]
[25]
B. Jordan, C. Calvo, Can. J. Chem. 51 (1973) 2621–2625.
B.D. Jordan, C. Calvo, Acta Crystallogr. B Struct. Crystallogr. Cryst. Chem. 32 (1976) 2899–2900.
R. Gopal, C. Calvo, J. Solid State Chem. 5 (1972) 432–435.
D. Ballutaud, E. Bordes, P. Courtine, Mater. Res. Bull. 17 (1982) 519–526.
M. Tachez, F. Theobald, E. Bordes, J. Solid State Chem. 40 (1981) 280–283.
G. Ladwig, Z. Anorg. Allg. Chem. 338 (1965) 266–278.
R.N. Bhargava, R.A. Condrate, Appl. Spectrosc. 31 (1977) 230–236.
E. Bordes, Catal. Today 1 (1987) 499–526.
T. Shimoda, T. Okuhara, M. Misono, Bull. Chem. Soc. Japan 58 (1985) 2163–2171.
Z.-Y. Xue, G.L. Schrader, J. Phys. Chem. B 103 (1999) 9459–9467.
M. Abon, J. Catal. 156 (1995) 28–36.

[26]
V.V. Guliants, J.B. Benziger, S. Sundaresan, I.E. Wachs, J.-M. Jehng, J.E. Roberts, Catal. Today 28
(1996) 275–295.
[27]
[28]
F. Benabdelouahab, J. Catal. 134 (1992) 151–167.
S. Geupel, K. Pilz, S. van Smaalen, F. Büllesfeld, A. Prokofiev, W. Assmus, Acta Crystallogr. C Cryst.
Struct. Commun. 58 (2002) i9-i13.
[29]
S.C. Roy, R. Glaum, D. Abdullin, O. Schiemann, N. Quang Bac, K.-H. Lii, Z. Anorg. Allg. Chem. 640
(2014) 1876–1885.
[30]
M.S. Hegde, G. Madras, K.C. Patil, Acc. Chem. Res. 42 (2009) 704–712.
[31]
[32]
A. Varma, A.S. Mukasyan, A.S. Rogachev, K.V. Manukyan, Chem. Rev. 116 (2016) 14493–14586.
C. Heine, M. Hävecker, E. Stotz, F. Rosowski, A. Knop-Gericke, A. Trunschke, M. Eichelbaum, R.
Schlögl, J. Phys. Chem. C 118 (2014) 20405–20412.
[33]
[34]
G.W. Coulston, Science 275 (1997) 191–193.
E. Mannei, F. Ayari, C. Petitto, E. Asedegbega–Nieto, A.R. Guerrero–Ruiz, G. Delahay, M. Mhamdi,
A. Ghorbel, Micropor. Mesopor. Mater. 241 (2017) 246–257.
[35]
K. Amakawa, Y.V. Kolen’ko, A. Villa, M.E. Schuster, L.-I. Csepei, G. Weinberg, S. Wrabetz, R.
Naumann d’Alnoncourt, F. Girgsdies, L. Prati, R. Schlögl, A. Trunschke, ACS Catal. 3 (2013) 1103–1113.
[36]
[37]
S.C. Roy, B. Raguž, W. Assenmacher, R. Glaum, Solid State Sci. 49 (2015) 18–28.
Subrata Chandra Roy, Thermodynamically stable and metastable solids: New approaches to the
synthesis of anhydrous phosphates containing vanadium, molybdenum, and/or tungsten. PhD Thesis,
Bonn, 2015.
[38]
[39]
861.
S.L. Wang, C.C. Wang, K.H. Lii, J. Solid State Chem. 82 (1989) 298–302.
S. Fischer, H. Fischer, S. Diele, G. Pelzl, K. Jankowski, R.R. Schmidt, V. Vill, Liq. Cryst. 17 (1994) 855–

[40]
I. Tanaka, M. Yao, M. Suzuki, K. Hikichi, T. Matsumoto, M. Kozasa, C. Katayama, J. Appl. Crystallogr.
23 (1990) 334–339.
[41]
G. Meyer, SOS. Staatsexamenarbeit, Gießen, 1980.
[42]
11th International Congress On Catalysis - 40th Anniversary, Proceedings of the 11th ICC, Elsevier,
1996.
[43]
[44]
[45]
[46]
[47]
[48]
[49]
[50]
J.W. Johnson, D.C. Johnston, A.J. Jacobson, J.F. Brody, J. Am. Chem. Soc. 106 (1984) 8123–8128.
G. BUSCA, J. Catal. 99 (1986) 400–414.
H. E. Bergna (1988) US 4769477 A.
E. Bordes, Catal. Today 16 (1993) 27–38.
H. Morishige, J. Tamaki, N. Miura, N. Yamazoe, Chem. Lett. 19 (1990) 1513–1516.
E.A. Lombardo, C.A. Sánchez, L.M. Cornaglia, Catal. Today 15 (1992) 407–418.
G. Poli, I. Resta, O. Ruggeri, F. Trifirò, Appl. Catal. 1 (1981) 395–404.
G.J. Hutchings, M.T. Sananes, S. Sajip, C.J. Kiely, A. Burrows, I.J. Ellison, J.C. Volta, Catal. Today 33
(1997) 161–171.
[51]
[52]
[53]
[54]
[55]
L.M. Cornaglia, C.A. Sa´nchez, E.A. Lombardo, Appl. Catal. A: Gen. 95 (1993) 117–130.
N. Batis, J. Catalysis 128 (1991) 248–263.
F. Cavani, G. Centi, F. Trifiro, Appl. Catal. 9 (1984) 191–202.
M. O'Connor, F. Dason, B.K. Hodnett, Appl. Catal. 64 (1990) 161–171.
R.I. Bergman, N.W. Frisch, Production of maleic anhydride by oxidation of n-butane, Google
Patents, 1966.
[56]
A. Celayasanfiz, T. Hansen, A. Sakthivel, A. Trunschke, R. Schlögl, A. Knoester, H. Brongersma, M.
Looi, S. Hamid, J. Catal. 258 (2008) 35–43.
[57]
R. Naumann d'Alnoncourt, Y. V. Kolen'ko, R. Schlögl, A. Trunschke, Comb. Chem. High Throughput
Screening 15 (2012) 161–169.

[58]
A.C. Sanfiz, T.W. Hansen, D. Teschner, P. Schnörch, F. Girgsdies, A. Trunschke, R. Schlögl, M.H.
Looi, S.B.A. Hamid, J. Phys. Chem. C 114 (2010) 1912–1921.
[59]
J.S. Valente, R. Quintana-Solórzano, H. Armendáriz-Herrera, G. Barragán-Rodríguez, J.M. López-
Nieto, Ind. Eng. Chem. Res. 53 (2014) 1775–1786.
[60] R.K. Grasselli, J.D. Burrington, D.J. Buttrey, P. DeSanto Jr., C.G. Lugmair, A.F. Volpe Jr., T. Weingand,
Top. Catal. 23 (2003) 5–22.
[61]
[62]
[63]
A. Raminosona, E. Bordes, P. Courtine, J. Solid State Chem. 68 (1987) 1–10.
G.J. Hutchings, R. Higgins, J. Catal. 162 (1996) 153–168.
B.Y. Jo, S.S. Kum, S.H. Moon, Appl. Catal. A: Gen. 378 (2010) 76–82.

Figure 1 – Crystal structure of VPP (left) and II-type VOPO4 (right) with VO6 units (green) and PO4
units (blue)[28,29]. Oxygen atoms from phosphate groups are given as ○, vanadyl oxygen atoms as
●.

Figure 2 – Flow scheme of the laboratory scale test setup (hte GmbH) with 8 parallel fixed-bed reactors
and 1 blank reactor, gas and liquid dosing unit and on-line gas chromatograph for product analysis.

Figure 3 – Guinier photographs of VPO4-m1 before (top) and after catalytic testing (bottom).
Reflections of Al2O3 assigned to corundum, which was used as inert material and could not be
separated from the catalyst.

Figure 4 – Guinier photographs of single phase V0.8W0.2OPO4 after catalytic testing and simulated
XRPD pattern of V0.8W0.2OPO4.[37]

Figure 5 – Selectivity for MAN as a function of conversion of n-butane.

Figure 6 – Selectivity for MAN dependent on conversion of n-butane under GHSV = 2000 h-1
c(C4)/c(O2)/c(H2O)=2 %/20 %/3 %.

Table 1
Catalyst characterization by XRPD and BET measurements and surface before catalytic testing
Catalyst
SBET,fresh Structure
Space
Group
Pnma
Scherrer
Diameter (nm)
>200
(m2/g)
3.5
Typea

VOPO4
V0.99W0.01OPO4
V0.94W0.06OPO4
V0.9W0.1OPO4
V0.8W0.2OPO4
WOPO4
1.3
Pnma
P4/n
103
>200
165
84

2.4
II
4.9
P4/n
II
3.4
P4/n
II
3.9
WOPO4

P21/m
Pnma
Pbam
78
V0.8Mo0.2OPO4
V0.8Mo0.2OPO4
0.8
52
1.1
192

W0.67Fe0.33OPO4
VPO4-m1
6.2
6
WOPO4
VPO4
VPP
VPP
VPP
VPO4
P21/m
C2/c
68
53
b
(VO)2P2O7
31
80
14
15
3
Pca21
Pca21
Pca21
C2/c
c
(VO)2P2O7
>200
>200
>200
(V0.95Mo0.05O)2P2O7
Fe(WO2)2(P2O7)(PO4)
a Greek letters refer to the various polymorphs of VOPO4; b from organic synthesis route; c from SCS

Table 2
Catalyst performance data, stability, specific surface before and after catalytic testing.
Catalyst
X(n-butane)f
S(MAN)f
(%)
21
S(COx)
(%)
72
SBET,fresh SBET,used Structure
(%)
6
(m2/g)
1.9
-
(m2/g)
-
Stability
-
VPO4-m1 a
VPO4-m1 b
Noc
14
80
14
15
3
49
54
4.7
26
d
(VO)2P2O7
(VO)2P2O7
Yes
Yes
Yes
Yes
70
29
26
e
43
45
2.1
1.5
1.1
1.8
1.3
0.8
(V0.95Mo0.05O)2P2O7
30
34
Fe(WO2)2(P2O7)(PO4)
0
68
aTOS = 254 h; bTOS = 480 h; c Formation of (VO)2P2O7 and -VOPO4; d from organic synthesis route; e
material from SCS; f temperature for catalytic testing at 400 °C

Table 3
Catalyst stability, specific surface before and after catalytic testing, and activation energy
b
Structure SBET,fresh
SBET,used
(m2/g)
2.6
Structure
Stability
Yes
Ea (kJ
mol-1)
74
Typea
(m2/g)
3.5
1.3
2.4
4.9
3.4
3.9
0.8
1.1
6.2
VOPO4

V0.99W0.01OPO4
V0.94W0.06OPO4
V0.9W0.1OPO4
V0.8W0.2OPO4
WOPO4
1.2
3.9
6.5
6.3
5.5
0.5
1.3
6.1
Yes
Yes
Yes
Yes
Noc
Yes
Yes
Nod
93

93
II
108
115
24
II
II
WOPO4
V0.8Mo0.2OPO4
V0.8Mo0.2OPO4
W0.67Fe0.33OPO4
98

111
n.a.

WOPO4
aGreek letters refer to the various polymorphs of VOPO4; bActivation energies derived from Arrhenius
plots; cBy-phase related to MPTB. dAfter catalytic testing one additional reflection occures