Propane ammoxidation on Bi promoted MoVTeNbOx oxide catalysts: Effect of reaction mixture composition

Article abstract of DOI:10.1016/j.apcata.2015.08.034

MoVTeNbO catalysts were characterized with XRD, XPS, and FTIR techniques and tested in the ammoxidation of propane. Bismuth-modified MoVTeNbO catalysts showed a higher acrylonitrile yield than the base four-component system. The effect of the reaction mixture composition (C₃H₈, NH₃ and O₂) on selectivity towards different products was studied at propane conversion above 80%. The favorable effect of bismuth promoter on the selectivity towards acrylonitrile was explained by suppression of acrylonitrile transformation connected with decreasing acidity of the catalyst.

Full text of DOI:10.1016/j.apcata.2015.08.034

Accepted Manuscript
Title: Propane ammoxidation on Bi promoted MoVTeNbOx
oxide catalysts: Effect of reaction mixture composition
Author: Tamara V. Andrushkevich Galina Y. Popova Yuriy A.
Chesalov Evgeniya V. Ischenko Mikhail I. Khramov Vasily V.
Kaichev
PII:
S0926-860X(15)30129-0
DOI:
Reference:
http://dx.doi.org/doi:10.1016/j.apcata.2015.08.034
APCATA 15528
To appear in:
Applied Catalysis A: General
Received date:
Revised date:
Accepted date:
27-5-2015
31-7-2015
27-8-2015
Please cite this article as: Tamara V.Andrushkevich, Galina Y.Popova, Yuriy
A.Chesalov, Evgeniya V.Ischenko, Mikhail I.Khramov, Vasily V.Kaichev,
Propane ammoxidation on Bi promoted MoVTeNbOx oxide catalysts:
Effect of reaction mixture composition, Applied Catalysis A, General
http://dx.doi.org/10.1016/j.apcata.2015.08.034
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.

Propane ammoxidation on Bi promoted MoVTeNbO oxide catalysts: Effect of
x
reaction mixture composition
a*
a
a,b
Tamara V. Andrushkevich andrushk@catalysis.ru, Galina Y. Popova , Yuriy A. Chesalov ,
a,b
c
a,b
Evgeniya V. Ischenko , Mikhail I. Khramov , Vasily V. Kaichev
a
Boreskov Institute of Catalysis SB RAS, Lavrentieva str, 5, Novosibirsk, 630090, Russia
b
Novosibirsk State University, Pirogova str., 2, Novosibirsk, 630090, Russia
c
Ascend Performance Materials LLC., P.O. Box 97, Gonzalez, FL 32560-0097, USA
*
Corresponding author. fax +7 383-3308056.
1

Graphical Abstract
1
2
- Mo V0.3Te0.14Nb0.12Ox
1
- Mo V0.28Te0.18Nb0.11Bi0.01O
1
C H N
x
3
3
6
0
0
2
4
1
1
COx
4
0
0
2
2
C H
3
6
20
0
1,2
1
0
C H N+HCN
2 3
1
2
5
4
2
1
2
C H O
3
4
2
10
20
30
40
50
60
70
80
90 100
Conversion of propane, %
2

Highlights
►
►
►
►
MoVTeNb catalyst promoted by Bi was studied in propane ammoxidation.
Bi-modified MoVTeNb catalyst demonstrates increased selectivity to acrylonitrile.
The transformation of acrylonitrile is lowered due to decreased catalyst acidity.
The effect of propane, ammonia, and oxygen concentrations was investigated.
3

ABSTRACT
MoVTeNbO catalysts were characterized with XRD, XPS, and FTIR techniques and tested in the
ammoxidation of propane. Bismuth-modified MoVTeNbO catalysts showed a higher acrylonitrile
yield than the base four-component system. The effect of the reaction mixture composition (C
3 8
H ,
NH and O ) on selectivity towards different products was studied at propane conversion above 80%.
3
2
The favorable effect of bismuth promoter on the selectivity towards acrylonitrile was explained by
suppression of acrylonitrile transformation connected with decreasing acidity of the catalyst.
Keywords: MoVTeNb oxide catalyst; Bi-doped, ammoxidation; propane; acrylonitrile.
4

1
. Introduction
Direct oxidation of propane to acrylic acid or propane ammoxidation to acrylonitrile (ACN) is a
promising alternative to their production from propylene. MoVTeNbO mixed oxide catalysts
developed in the 90-s by Mitsubishi Chemical Corp. are efficient in the conversion of propane to
acrylic acid and ACN [1-3]. The studies have been focused mainly on the chemical and phase
composition of the catalysts, the structure of M1 and M2 phases, and the preparation methods [4-17].
Recent publications provide data on the nature of the active sites, the reaction mechanism and
kinetics [4, 18-30].
The pathways of ACN formation via propylene have been established for catalysts of various phases
and stoichiometric compositions [21-25]. More detailed kinetic studies of the ammoxidation of
propane on MoVTeNb were described elsewhere [30-32]. According to Asakura et al. [30], ACN is
formed directly from propane rather than through intermediate propylene. The rate of propane
ammoxidation to ACN is described by an empiric power equation; the reaction rate increases at
raising the concentrations of oxygen and propane while ammonia influences negatively. The effect of
1 x
the reaction mixture composition on the rate of reaction on Mo V0.4Nb0.1Te0.2O catalysts was studied
earlier [31]. The acrylonitrile yield of 53% was achieved at 703 K using a feed containing propane,
ammonia and air at 1/1.2/15 ratio. The parallel and consecutive pathways of the reaction products
formation on the MoV0.3Nb0.12Te0.17
O
x
catalyst according to previous studies [32] are shown on
Scheme 1. The first order rate constants were calculated in accordance with the proposed mechanism.
The slow step in the ACN formation is the propane activation (k = 0.81). There are two routes of
1
ACN formation; ACN is formed directly from propane and via propylene.
The rate of propylene conversion to ACN is more than an order of magnitude higher than that of
propylene formation; the reaction rate constant ratio is k
constants of CO formation from ACN, propylene and propane (k
that the complete propylene oxidation is the fastest of the three. Comparison of the rate constants of
parallel propylene conversion to CO and ACN, k /k = 0.4, suggests that substantial share of ACN
selectivity is lost due to the deep propylene oxidation.
For the majority of MoVTeNb catalyst systems, initial selectivity towards propylene can achieve
2
/k
1
= 13. The comparison of the rate
x
3
= 0.46, k = 4.20, k = 0.02) shows
4
5
x
4
2
8
0% or greater value. The ACN selectivity in the middle conversion range up to 80% is usually 65-
0%. If the selectivity does not decrease at higher conversions, the ACN yield would be close to its
7
selectivity. Apparently, the oxidation of ACN plays a significant role in the yield decrease. In the
work cited above [32] the experiments were performed at conversions lower than 70%; at these
conversions the ACN oxidation is not prominent. The propane ammoxidation should be reliably
studied at considerably higher propane conversions to elucidate the reaction route of ACN
5

transformation. Increased ACN yield can be achieved preserving high selectivity by suppression of
these routes.
According to the literature [33-35] and our opinion [36], the direction of hydrocarbons
transformation to the selective oxidation or to the deep oxidation depends on the bond strength of the
surface compounds of reactants and products. Weak products binding provides a high rate of the
desorption preventing the deep oxidation or degradation. The bond strength of surface complexes
depends on the acid-base properties of the reactant and the catalyst. The conception of the acid-base
interactions between gaseous reactants and surface sites of the solid catalysts was developed by Stair
[
37]. Doping is an effective way to change the acid-base properties of the catalyst. Introduction of
electronegative elements onto the surface should produce an inductive effect making the nearby
metal cations less basic or more acidic. Introduction of electropositive species such as alkali elements
should produce a more basic or less acidic sites [37]. Grasselli hypothesized that acidic elements (e.g.
1
P, B, W) should enhance the acidic properties of Mo V0.31Nb 0.12Te0.2 catalysts whereas basic
elements (e.g. Cs) should enhance basic properties. The elevation of acidity in the first case should
enhance the selectivity towards acrylic acid. Contrariwise, the enhancement of basic properties in the
second case should decrease the selectivity towards acrylic acid. Indeed, the addition of P results in
the increase the selectivity towards acrylic acid, whereas the addition of Cs depresses significantly
the formation of acrylic acid. [38]. The incorporation of potassium in Mo–V–Sb–O catalysts result in
a significant decrease in the number of acid sites on the catalyst surface favoring the formation of
acrylic acid on the redox sites and decreasing the formation of acetic acid on the oxyhydrative
scission acid sites in the oxidation of propane [39-41]. Baca et al. [42] found a correlation between
acidity of the M1 phase promoted with Nb in MoVTe(Sb)O catalysts and selectivity in the oxidation
of propane. Surface acidity strongly affects the binding and reactivity of propene molecules which
have a basic character. Lowering the amount of strong Lewis acid sites (LAS) or Brønsted acid sites
(
BAS) increases the selectivity towards acrylic acid due to a reduction in the deep oxidation of
propylene to CO on LAS or due to a decline in the acetic acid and acetone formation on BAS.
3
Lopez-Medina et al. [22] demonstrated that Te in nanostructured MoVNbTe/Al O catalyst
x
4
+
2
decreases the amount of strongly acidic sites. It results in increasing the selectivity towards acrylic
acid due to reducing the number of strongly bound carboxylate complexes. The latter are
intermediates of deep oxidation products. The higher selectivity towards acrylic acid over Ga-
containing MoVTeO catalysts has been related to modifying acid properties of the catalysts surface.
3
+
The presence of Ga cations on the catalysts surface decreases the number and the strength of acid
sites favoring the selective oxidation of propane to acrylic acid [43]. The acidity of the active sites
has a strong influence on selectivity of the MoVTeNbO catalysts in both the oxidation and the
ammoxidation of propane. Centers with moderate acidity favor to the partial oxidation where as
6

strong acid sites advantage the deep oxidation. Steam or ammonia during the oxidation or
ammoxidation reactions, respectively, block the strong LAS and promote selective course of the
reactions [17].
In general, we can conclude that the strong acidity of active centers leads to the formation of side
products. Deep oxidation products are formed on LAS, whereas oxyhydrative scission products are
generated on BAS.
Cited above works associate acidity and selectivity towards acrylic acid through the influence on the
transformation of the intermediate propylene. The effect of the acidity on the transformation of final
products is equally important. But such data both on acrylic acid in the oxidation of propane and on
acrylonitrile in the ammoxidation of propane are absent.
ACN is organic base due to the presence of unshared electron pair on the nitrogen atom. It interacts
with acid sites by donation of the electron density from the nitrogen atom [37]. The MoVTeNb oxide
catalyst contains highly electronegative elements in the highest oxidation state ensuring the catalyst
acidity. Reducing the acidity of active centers by introduction electropositive supplements should
lead to a decrease in the bond strength of the adsorbed acrylonitrile preventing the oxidation and the
oxidative scission of surface species.
As a promoter, we have chosen Bi, firstly as a more electropositive element in comparison with the
basic composition of MoVTeNb catalyst and secondly as an indispensable component of industrial
catalysts for the ammoxidation of propylene and the oxidation of propylene to acrolein in a two-stage
production of acrylic acid.
In this study, we attempted to promote the MoVTeNb oxide catalyst with Bi in an effort to control of
acrylonitrile conservation. The object of our study was the reaction pathways of propane
ammoxidation on the Mo1.2
x
V0.3Te0.23Nb0.12O and Bi-modified catalysts, and the effect of the reaction
mixture composition at propane conversion above 80%.
2
. Experimental
2
.1. Catalyst preparation
The Mo1.2
V
0.3Te0.23Nb0.12
O
x
catalyst was prepared using ammonium heptamolybdate
O, ammonium metavanadate NH VO (chemicals from commercial suppliers in
Russia) and telluric acid H TeO (Aldrich); the purity of each reagent was 99% or higher. Niobium
(
NH
4
)
6
Mo
7
O
24×4H
2
4
3
6
6
oxalate was prepared by hydrolysis of niobium pentachloride (NbCl
water and neutralization with ammonium hydroxide (NH OH). White precipitate was filtered,
washed with water and dissolved in a solution of oxalic acid (C
5
, Acros Organics, 99.8%) in
4
2
-
2
4
O /Nb = 3.0).
The catalysts were synthesized from slurry according to the patented procedure [1]. First, the aqueous
solution of molybdenum, vanadium and tellurium compounds was prepared. Then, niobium oxalate
7

was added to the ternary solution. After addition of niobium oxalate solution pH of the catalyst slurry
was 2.9. Solid precursor powder was prepared by spray-drying of the slurry in a laboratory spray-
dryer (Buchi-290), the inlet temperature was 220 C and the outlet temperature was 110 C. The
resulting powder was calcined for a short period in air at 300 C and in a He flow at 600 C for 2 h.
1 n x
The Mo V0.3Te0.23Nb0.12Bi O (n = 0.01 - 0.05) catalyst was prepared by mixing powders of
Mo V Te0.23Nb0.12O and Bi O . The product was calcined at 550 C in He flow for 2 h. The
1
0.3
x
2
3
calcined powders were pressed into tablets, crushed and then sieved. The 0.25-0.50 mm fraction was
used in the experiments.
2
.2. Catalyst characterization
The specific surface areas of the catalysts were determined by low temperature nitrogen adsorption at
7 K using an automatic ASAP-2400 apparatus.
7
The phase composition of the catalysts was studied using a Siemens D500 diffractometer. Scanning
was performed in the 2θ range from 5 to 70 with a step of 0.028 using Cu Kα radiation (λ = 1.5418
Å).
The chemical composition of the catalysts was determined by X-ray photoelectron spectroscopy
(
XPS) using a SPECS’s Surface Nano Analysis GmbH spectrometer equipped with a PHOIBOS-150
hemispherical electron energy analyzer, a FOCUS-500X-ray monochromator and an XR-50M X-ray
source with a double Al/Ag anode. The core-level spectra were obtained using the monochromatic Al
Kα radiation under ultrahigh vacuum conditions. Binding energies of the photoemission peaks were
corrected to the C1s peak (284.8 eV) from the adventitious contaminated layer. Relative element
concentrations were determined from the integral intensities of XPS peaks using the cross-sections
according to Scofield.
The thermal programmed desorption (TPD) of ammonia in combination with IR spectroscopy was
applied to characterize the acidity of the samples. FTIR spectra were obtained with a BOMEM MB-
1
02 spectrometer. The spectrometer was operated in the transmission mode using a specially
2
designed quartz cell-reactor with BaF windows. The catalysts were pressed into a thin self-
2
supporting pellet (15 mg/cm ) and placed into the cell-reactor. The FTIR experiments were
performed at atmospheric pressure using a feed of 5 vol% NH
3
in a He flow. The spectra were
-1
-1
measured at 25–300 C in a range of 2000–1000 cm at a resolution of 4 cm . The samples were
preliminarily activated in a flow of air for 60 min at 300 C. Subsequently, the cell-reactor and the
catalyst sample were cooled to 25 C, and the air flow was replaced by the ammonia containing
mixture flow. The IR spectra were measured during the adsorption. After constant surface covering
was achieved, the ammonia containing mixture flow was replaced by a He flow. Then the
8

temperature was increased at a rate of 1 С/min. The IR spectra were collected at selected
temperatures during the TPD experiments.
2
.3. Catalytic tests
The catalytic tests were carried out at atmospheric pressure in a tubular flow fixed bed reactor (i.d. 12
mm; length 50 mm) with a coaxial thermocouple pocket (i.d. 4 mm). Samples (0.25–0.50 mm
particle size) were diluted with an inert material and charged into the reactor. The reactor was kept in
an air heated furnace. The temperature change along the furnace height was 2 C. The reaction
temperature was measured using a thermocouple placed in the thermocouple pocket directly in the
catalyst bed. Before operation, the catalyst was heated in a helium flow to the required temperature.
The catalysts were compared at a constant composition of the reaction mixture with a molar ratio
3 8 3
C H /NH /air = 1/1.2/15 at 420 C. Kinetic studies were performed at different reaction mixture
compositions and contact times. The later was changed by variation of the catalyst weight or the flow
rate.
3
3
. Results and discussion
.1. The composition and structure of the catalysts
Table 1 summarizes the chemical composition of the samples under study before and after
calcination as well as their specific surface area. Tellurium loss due to sublimation is the main reason
for the difference in composition of calcined and uncalcined samples. According to the XRD analysis
all samples consist of the orthorhombic phase M1, the hexagonal phase M2, TeMo
5
O16 and
(
V0.07Mo ) O .
0.93 5
14
The complexity of the chemical composition and small amounts of additives hinder the identification
of structural differences of promoted and unpromoted catalysts. To study the phase composition, we
used XRD and IR spectroscopy. XRD patterns are typical for M1 and M2 phases (Fig. 1). IR spectra
obtained (not shown) are also not original [44] and entirely confirm the XRD data.
Fig. 1 shows the XRD patterns of the calcined catalysts. The 2θ peaks at 6.6, 7.9, 9.0, 22.1 and 27.2º
are assigned to the orthorhombic phase M1 and the 2θ peaks at 22.1, 28.2, 36.2 and 45.15º are
assigned to the M2 phase. In addition, the weak reflections of TeMo
6.7, 30.5 and 34.9ºJCPDS, 31-874] and (V0.07Mo0.93 14 (peaks at 22.2, 23.4, 25.0 and 31.5º)
JCPDS, 31-1437] are also observed. Introduction of bismuth in the Bi/Mo ratio of 0.01–0.02 results
5
O16 (peaks at 26.2,
2
5
) O
[
in a minor increase in the amount of the M2 phase. The sample with the atomic ratio Bi/Mo of 0.05
comprises a lot of the hexagonal phase M2.
9

According to the XPS analysis, the samples contain Mo, V, Nb, Te, Bi, and O. Corresponded core-
level spectra of the Mо
1 1
V0.28Te0.14Nb0.13 and Mо V0.28Te0.18Nb0.11Bi0.01 catalysts are presented in Fig.
S1 of supporting material. The Mo3d5/2, Te3d5/2, Nb3d5/2, V2p3/2, Bi4f7/2 and O1s binding energies
are summarized in Table 2. In both the cases the Mo3d spectra consist of two symmetric peaks
corresponded to the Mo3d5/2 – Mo3d3/2 spin-orbit doublet. The Mo3d5/2 binding energy is in the range
6
+
2
33.1-233.2 eV that is typical of molybdenum in the Mo state [45]. The Te3d5/2 spectra consist of a
6
+
sharp peak at 576.8-577.0 eV that is also typical of Te in the highest oxidation state Te [46]. The
Nb3d core-level spectra contain well-defined Nb3d5/2 – Nb3d3/2 spin-orbit doublet. The Nb3d5/2
5
+
binding energy is in the range 207.3-207.4 eV that can be attributed to the Nb state. Bulk Nb O is
2
5
characterized by the Nb3d5/2 binding energy equal to 207.8 eV [47]. In the V2p3/2 spectrum of
Mо 0.28Te0.14Nb0.13 the wide peak at 517.0 eV is observed. The doping by Bi leads to a shift the
1
V
5
+
peak to 517.2 eV. Both these values correspond to the V state [48]. The Bi4f spectrum consist of
two sharp peaks at 159.9 and 165.2 eV which correspond to the Bi4f7/2 – Bi4f5/2 spin-orbit doublet
3
+
(
not shown). According to the literature [49], Bi is characterized with the similar values of the
Bi4f7/2 binding energy in the range of 159.3-159.9 eV. The atomic ratio [Bi]/[Mo] is equal to 0.011
that is in good agreement with the results of chemical analysis (Table 1).
Table 3 demonstrates the surface content of the elements with respect to the surface content of Mo.
One can see that on the surface of the promoted sample, the V content is significantly higher, the Te
content is considerably reduced, whereas the Nb content is virtually unchanged. XPS results show
also the correspondence between surface and bulk Bi content of promoted sample. It confirms clearly
the introduction of Bi in M1 phase structure.
Our high-resolution transmission electron microscopy studies [50] identified no new phases induced
by the bismuth addition. Bismuth addition also does not change the structure and morphology of the
M1 phase.
Previously, Grasselli et al. [38] studied the promotion of MoVNbTe catalyst by small additives of
phosphorus (P/Mo = 0.01, 0.001). They emphasized that the dopants concentration must be kept
critically low for the success of the catalysis. The authors also found no change in the structure of the
phases and suggested that phosphorus is located in the heptagonal or hexagonal channels of the M1
and M2 phases. Taking into account our XPS data, we [50] also assume that Bi is localized in the six-
and seven-membered channels of the M1 structure similar to its chemical analog Sb [12] or P [38].
3
.2. Catalytic performance
Catalytic characteristics of the samples containing different amounts of bismuth are shown in Table
. Samples with atomic ratio Bi/Mo = 0.01–0.02 are more active than the standard sample. We
4
1
0

associate the decrease in the overall catalysts activity after introduction of a larger amount of bismuth
Bi/Mo = 0.05) with the formation of M2 phase, which is more than an order of magnitude less active
(
than the M1 phase both in the propane and in the propylene oxidation [51]. In the region of propane
conversions above 90%, the selectivity towards acrylonitrile obtained on the Bi-modified catalysts is
much higher than that obtained on the non-modified samples. The modified samples also show a
lower activity in the oxidation of ammonia and in the oxidation of propane into carbon oxides. As
mentioned above, the introduction of a larger amount of bismuth (Bi/Mo = 0.05) decreases the
overall catalysts activity as well as the selectivity towards ACN. Let us note that the M2 phase
contains more reactive weakly bound oxygen than the M1 phase [52], and this contributes to the
reduction of a selectivity toward ACN in the case of the sample with Bi/Mo = 0.05.
The selectivities towards C
conversion on the Mo
Fig. 2. The major products of the selective conversion of propane on both catalysts are acrylonitrile
and propylene. By-products include acetonitrile, HCN, acrylic acid, and CO . The dependence of
selectivity on conversion is the same for both catalysts when conversion is in the range of 20-80%.
As the propane conversion increases from 20 to 40%, selectivity towards C decreases from about
2% to 10%, and the selectivity towards acrylonitrile increases from 53% to 67%.
3
H
3
N, C
3
H
6
, CO
x
, C
2
H
3
N+HCN, and C
3
H
4
O
2
as a function of the propane
1
V0.28Te0.14Nb0.13
O
x
and Mо
V0.28Te0.18Nb0.11Bi0.01O catalysts are shown in
1 x
x
3
H
6
2
x
Selectivity towards CO continuously increases with the increase of the conversion of propane.
Selectivity towards acetonitrile and HCN are fairly constant at 20-80% propane conversion. At the
conversion of propane higher than 80%, a notable difference between the unmodified and modified
catalysts is observed. In the case of the M0.1
conversion to 93% leads to a dramatic decline in the selectivity towards acrylonitrile from 67% to
0%. On the modified catalyst, the selectivity towards acrylonitrile decreases only to 56%. The
decline in selectivity towards acrylonitrile is accompanied by the increase in selectivities towards
CO , acrylic acid and acetonitrile.
x
V0.3Te0.14Nb0.13O catalyst, an increase in the propane
4
x
Therefore, the selectivity towards acrylonitrile on the modified catalyst is comparable to that on the
non-modified sample at the propane conversion up to 80% and is much higher in the region of the
propane conversions above 80%.
3
.3. Effect of the reaction mixture composition
Reaction products in the ammoxidation of propane are acrylonitrile, oxidation products (acrylic acid
and COx) and oxidative scission products (HCN and C N). With increasing the propane
2
H
3
conversion, an intensive conversion of ammonia takes place simultaneously, and the direction of the
reaction is shifted towards the oxidation and the oxidative scission. In Oliver et al. opinion [17], in
the propane ammoxidation, the optimal amount of ammonia should depend on conversion level.
1
1

Thereby, the maintenance of the optimal composition of the reactant mixture is particularly important
at the high propane conversion.
The effect of the reactant mixture composition on the catalytic performance was studied over the
1 x
Mо V0.28Te0.18Nb0.11Bi0.01O catalyst at the high propane conversion level (higher than 90%).
Monoparametric experiments were performed by changing the content of a certain component in the
reactant mixture while the concentrations of the other components were held constant. Figures 3a, b,
and c show the results of the experiments.
3.3.1. Effect of propane concentration
The concentration of propane in the reactant mixture was varied from 4 to 7% at the constant
concentrations of oxygen and ammonia (18.5% and 7%, respectively) and the contact time equal to
3
.0 s. The experimental results are shown in Fig. 3a.
As the concentration of propane increases from 3.5 to 7% the propane conversion decreases from
3.8 to 89.2%. The selectivity towards ACN passes through a small maximum (58.6–58.1%) at the
9
propane concentration 5–5.3%. At the propane concentration below 5%, the selectivity towards ACN
decreases due to increased formation of deep oxidation products, while at the propane concentration
above 5.3%, this is occurred due to elevated formation of acetonitrile. An increase in the selectivity
towards propylene from 0.1 to 0.7% in the region of high propane concentrations is due to lower
conversion of propane. As the propane concentration increases, the selectivity towards hydrocyanic
acid decreases steadily from 1.7 to 0.9%. The selectivity towards acrylic acid varies only slightly in
the range of propane concentrations under consideration.
3
.3.2. Effect of ammonia concentration
The concentration of ammonia in the reactant mixture was varied from 4.5 to 10.2% at the constant
concentrations of propane and oxygen (5.0% and 19%, respectively) and the constant contact time of
3
.4 s. The experimental results are shown in Fig. 3b.
As the ammonia concentration increases from 4.5 to 10%, the propane conversion decreases from 96
to 90%. The selectivity towards ACN at the propane conversion 95–96% changes from 46.1 to 56%
as the ammonia concentration increases from 4.5 to 6%, and then declines to 54.8% as the ammonia
concentration further increases to 10.2%. A decrease in the selectivity towards ACN in the region of
high ammonia concentrations is accompanied by an increase in the acetonitrile concentration.
At high concentration of ammonia, there is a considerable (approximately 4-fold) decrease in the
selectivity towards acrylic acid and carbon oxides. The selectivity towards HCN remains virtually
constant (1.5%) in a range of the ammonia concentrations 4.5–6.8% and decreases to 0.3% at further
1
2

increase in the ammonia concentration. In the region of high ammonia concentrations, the propane
conversion declines, and the selectivity towards propylene increases from 0.3 to 0.8 %.
3.3.3. Effect of oxygen concentration
The concentration of oxygen in the feed mixture was varied from 14.3 to 23.6% at the constant
concentrations of propane and ammonia (5.0% and 6.0%, respectively) and the contact time 3.2 s).
The experimental results are presented on Fig. 3c.
The elevation of the oxygen concentration from 14.3 to 23.6% leads to an increase in the propane
conversion from 85.9 to 94%. The selectivity towards acrylonitrile passes through a maximum (about
5
6%) in a range of the oxygen concentrations 17–19%.
As the oxygen concentration decreases from 17 to 14.3%, the selectivity towards ACN also decreases
and the selectivity towards acetonitrile increases. When the oxygen concentration exceeds 18.7%, the
selectivity towards carbon oxides and acrylic acid starts to increase. The selectivity towards hydrogen
cyanide changes slightly in a range of the oxygen concentrations 16–23.6%. The selectivity towards
propylene decreases as the oxygen concentration changes from 14.3 to 16% and remains constant at
higher oxygen concentration.
Processing of all series of monoparametric experiments in “acrylonitrile yield versus the ratio of
component concentrations” coordinates allows determining the condition that provides the maximum
yield of the target product (fig. 4).
The yield of acrylonitrile passes through a maximum in a range of the O
2 3 8
/C H ratio 3.4–3.6. At
lower values of the О /С ratio, the yield of acrylonitrile decreases due to increased formation of
2
3 8
Н
acetonitrile, while at higher values this is due to increased formation of acrylic acid and carbon
oxides.
2 3
The maximum yield of acrylonitrile is observed in a range of the О /NH ratio between 2.7 and 3.1
and decreases at higher values of the О /NH ratio due to increased formation of propane oxidation
2
3
products, acrylic acid and carbon oxides.
The yield of acrylonitrile passes through a maximum in a range of NH
.4. At the NH /C ratio below 1.2, the yield of acrylonitrile decreases due to increased formation
of acrylic acid and CO . At the NH /C ratio above 1.4, the yield of acrylonitrile declines due to
3 3 8
/C H ratio between 1.2 and
1
3
3 8
H
x
3
3 8
H
increased formation of acetonitrile and carbon oxides.
Therefore, the kinetic experiments determined quite narrow intervals for optimal ratios of the feed
mixture components at high conversion level.
1
3

Typically, in the academic literature, air is used as an oxidizer; the rest is in accordance with the
objectives of the study. To compare the samples, we traditionally used in our work the mixture with
ratio of C
ratio corresponds to C
NH /O ratios. The C
3
H
8
/NH
3
/air = 1/1.2/15. This composition is the most common in the literature also. The
/NH /O ratio = 1/1.2/3, which is within the range of optimal C /NH and
/O ratio appears significantly below the optimum (3 vs 3.5). Therefore, at
3
H
8
3
2
3
H
8
3
3
2
3
H
8
2
high propane conversion, it is necessary to dose oxygen apparently due to the high degree of the
catalyst reduction under reaction conditions. The optimal components ratios are the concentration
ranges. This gives a certain freedom in the independent variation of reactant mixture components.
Concerning the industrial process [53], the composition of the reaction mixture is not reported, but
our recommendations are the same. We believe that these recommendations are useful also for
process with recycle [54].
4
. Discussion
Dependence of selectivities towards different products in the propane conversion (Fig. 2) can be
described by the parallel-consequential network of the ammoxidation reaction mechanism. The fact
that the selectivity towards acrylonitrile increases as the selectivity towards propylene decreases,
suggests a consecutive (via propylene) mechanism of acrylonitrile formation. Products of total
oxidation (CO
and acrylonitrile. A drastic decrease in the selectivity towards acrylonitrile and increase in the
selectivity to CO and acetonitrile at the propane conversions greater than 80% indicates the
predominant reaction pathway of these products formation is acrylonitrile transformation. The
formation of CO from propylene is less significant, which is evidenced by only a small change in the
x
) form directly from propane (selectivity ca.15% at X  0%) and also from propylene
x
x
selectivity towards CO at intermediate propane conversions.
x
The promotion of the MoVTeNbO catalysts with Bi does not affect the proportion of formation of
products from the parallel route from propane and propylene, however decreases acrylonitrile
transformation.
Fig. 5 illustrates a positive effect of bismuth addition on the acrylonitrile yield. The modification of
the catalyst with Bi results in increasing the yield of acrylonitrile as compared to the four-component
catalyst at high conversions. Therefore, the incorporation of bismuth leads to an increase of the
acrylonitrile yield due to decline in the ACN transformation.
It is impossible to explain the improvement in the catalytic performance by structural features of the
promoted catalyst. Reduced transformation of ACN on the MoVTeNbBi catalyst could be caused by
the modification of the surface acid properties. The thermal programmed desorption (TPD) of
ammonia in combination with IR spectroscopy was applied to characterize the acidity of the samples.
1
4

The adsorption of ammonia at room temperature on the samples both with and without bismuth
-1
-1
results in the appearance of the single band at about 1410 cm in the 2000-1000 cm region of
+
infrared spectra. The band could be assigned to the δas(NH
4
) mode of adsorbed ammonium ions
formed by the interaction of ammonia molecules with Brønsted acid sites. The intensity, position of
maximum and width of the band under steady state are only slightly dependent on the presence of
bismuth in the samples. Fig. 6 demonstrates the dependence of the relative coverage of the surface
with ammonium ions on the temperature during the TPD experiments. The surface covering
+
measured at room temperature is adopted as 1. Decreasing the intensity of the δas(NH
4
) band is
observed with increasing temperature. The bismuth incorporation reduces the temperature of
desorption start at approximately 50 C and increases the desorption rate, particularly at 120-180 C.
In the case of Bi containing sample, complete desorption of ammonium proceeds also at a lower
temperature. Temperature shift of the desorption is a measure of the strength of the acid sites [55]. In
-1
the presence of Bi, approximately a 10 cm low-frequency shift of the band is also observed during
temperature increase. Thus, in this case the band position depends on the surface covering. In the
absence of Bi, the maximum of the band does not depend on the surface coating. Thus, the presence
of bismuth in the catalyst lowers the strength of the Brønsted acid sites.
It is important to note that despite the high electronegativity of elements constituting the catalyst, the
strength of Lewis acid sites is small, since we have failed to detect any bands characterizing surface
species of pyridine and ammonia with Lewis acid sites in IR spectra of both quaternary catalyst and
that of the catalyst promoted with bismuth. We have failed to measure the spectrum of adsorbed
pyridine (relatively weak base). This is in agreement with the work [41]. Thus, the surface of the
catalysts is insufficient acidic to adsorb a weak base. However, ACN has a basic nature due to the
unshared electron pair of nitrogen [37]; and an interaction of ACN and a catalyst is an indicator of
active center acidity. In accordance with [17, 22], highly acidic LAS are sites of adsorption of
strongly bound surface species which are intermediates of the deep oxidation products. The decrease
in the selectivity towards products of the deep oxidation of acrylonitrile is an evidence of itself of
decline LAS acidity.
The promotion of MoVTeNb oxide catalyst composed of highly electronegative elements by less
electronegative bismuth decreases the acidity of active sites and even more weakens the binding of
acrylonitrile with active centers reducing its degradation and oxidation. At the same time, the reduced
selectivity toward acetonitrile and HCN is a result of the decrease of the strength of Brønsted acid
sites responsible for oxyhydrative scission [17, 22, 55].
It should be noted that the promotion of the catalyst with bismuth does not affect the transformation
of propylene. We attribute this fact to relatively low basicity of the Bi, which is insufficient to
adequately reduce the acidity of the active sites binding propylene strongly. Analyzing the value of
1
5

the ratio of the rate constants of parallel propylene conversion routs to CO
32], we can conclude that substantial share of propylene is strongly bound. To sufficiently reduce
the acidity of these centers, the catalyst must be doped with the element, which is more basic than the
x 4 2
and ACN (k /k = 0.4)
[
+
3
Bi. The Ga can be an example of such a promoter. On the samples doped with Ga, a sharp increase
in selectivity toward acrylic acid is occurred at low propane conversions (X < 10%) [43]. This region
is characterized by the conversion of propylene into acrylic acid and COx [32]. The increase in
selectivity toward acrylic acid can be attributed to inhibition of propylene parallel conversion into
COx caused by a decrease in strength of propylene binding by less acidic active centers.
Unfortunately, the introduction of Ga results in an increase in the acrylic acid conversion into the by-
products. The maximum of selectivity toward acrylic acid is observed at 20-40% propane conversion
for different samples. Increased conversion is accompanied by COx and acetic acid formation.
Apparently, Ga reduces the acidity (increase the basicity [37]) too strongly. It leads to stronger
acrylic acid binding and as result to degradation of the surface intermediates. In summary, “the nature
and strength of acid sites should be tailored in order to optimize allylic oxidation and to avoid
oxidative breaking of CC bonds” [29]. The promotion of well-known multicomponent MoBiCoFeNi
PKO catalyst of propylene oxidation into acrolein and propylene ammoxidation into ACN with both
acidic and basic additives is a good illustration.
The second effect of bismuth addition consists in a lowered activation of ammonia. Direct
experiments have demonstrated a higher activity of the MoVTeNb oxide catalyst in the ammonia
oxidation as compared to the bismuth modified catalyst (Table 4). This corresponds to reduced
activity of the BiMo catalyst as compared to the activity of the MoVTeNb oxide catalysts in the
direct oxidation of ammonia [31]. Along with the data of TPD of ammonia, this also supports the
idea that the bismuth incorporation decreases the acidity of active sites.
Introduction of small amounts of Bi increases an activity in the propane ammoxidation (Table 4). We
attribute this fact to a change in the surface composition of the catalyst. By XPS, we have established
the correspondence between the bulk and surface bismuth content in promoted sample. We also
found an increase in vanadium, a decrease in tellurium, and constancy of niobium content on the
surface of promoted sample as compared to unpromoted one (Table 3).
A comparison of the catalytic properties and XPS data shows the general trend in the change of
vanadium surface content and an activity of the catalysts in the conversion of propane. The ratio of
5+
5+
surface concentration V Bi/V (1.27) is very close to the ratio of rate constants kBi/k (1.33).
The formation of ACN on MoVTeNbO M1 phase proceeds through intermediate oxidative
dehydrogenation of propane. In this reaction, the limiting rate-determining step is the H abstraction.
1
6

5
+
5+
The V =O acid-base pair participates in the activation of propane by V centers and in the H
abstraction by lattice oxygen according to a Mars-van Krevelen redox mechanism [56].
5
+
Our correlation between V surface content and activity confirms that H abstraction from propane is
5
+
the rate-limiting step and that V on the surface of MoVTeNbO catalysts is the active site in propane
activation.
The Mo-O-Te moieties are the centers of the allyl oxidation of propylene [51]. Since MoTe oxide
system is active in the oxidation of propylene at a ratio of Te/Mo≤0.1 [57], the decrease of tellurium
content (Table 3) is not critical.
Molybdenum is the main component in the M1 phase. In the Grasselli mechanism [12], Mo is the
center where the interaction of C3 fragments and ammonia is occurred resulting in AСN formation.
In accordance with the principles of acid-base interaction of the reactants and catalyst cations [37],
6
+
Mo must be a center of the strong binding of ACN; the reduction in the Mo oxidation state (acidity)
weakens the binding. Indeed, in the Grasselli mechanism, ACN is desorbed unchanged from Mo in
the reduced state [12]. Thus, the weak binding of the surface acrylonitrile intermediate provides a
high selectivity of MoVTeNb catalyst in the reaction under consideration.
The target product preservation is a common problem of selective oxidation. A strong re-absorption
of a target product under a reaction conditions leads to its transformation. A modification of catalysts
should be directed to the active site of a possible re-adsorption of the product in order to weaken its
bond strength. In case of the formation of acidic product, promoter should increase the acidity of the
active center (see, e.g. [58]). Contrariwise, in the case of a basic product, promoter should reduce the
acidity (increase the basicity) of the active center. The main effect of bismuth is to decrease the
acidity of the active sites. This prevents strong binding of re-adsorbed acrylonitrile and subsequent
degradation of the carbon skeleton. The latter results in COx, HCN, and acetonitrile formation.
Let us note that it is reasonable to use the target product transformations reactions as test ones when
selecting promoter.
5
. Conclusion
The ammoxidation of propane on the MoVTeNb catalyst proceeds with a high selectivity at
conversions less than 80%. At higher conversions, the selectivity towards acrylonitrile decreases due
to its further oxidation and oxyhydrative scission. Undesired reactions may be suppressed by the
addition of elements with a basic character to the catalyst.
All the components of the MoVTeNb oxide catalyst are highly electronegative elements that
determine acidic properties of the catalyst active sites. The promotion by bismuth decreased the
1
7

acidity of active sites of acrylonitrile activation and slows down its further transformation. The
second key factor is a low catalyst activity in the oxidation of ammonia. The addition of bismuth
3 3 8
decreases the activation and suppresses the oxidation of ammonia. As result, the NH /C H ratio
retains close to desired values over the entire range of the propane conversions. This favors the target
product formation and a high selectivity at the propane conversion above 90%. The promotion by
bismuth results in increasing the yield of acrylonitrile as compared to that of the four-component
catalyst.
A high conversion of propane is a crucial parameter for industrial processes. The composition of
reactant mixture at high conversions strongly effects on the yield of acrylonitrile. The influence of
propane, ammonia, and oxygen on the activity and selectivity of the MoVTeNbBi oxide catalyst was
studied. The ratios of components concentration are crucial for the ACN yield. The optimal ratios
3 3 8 2 3 8 2 3
appear to be in narrow ranges: NH /C H = 1.2–1.4, О /C H = 3.4–3.6, О /NH = 2.7–3.1.
Maintenance of these ratios in the reactant mixture via feeding appropriate components in a industrial
reactor [53] or into a gas recycle stream [54] maybe an effective technique for the enhancing
productivity of the process.
Acknowledgment. This work was partially supported by the Ministry of Education and Science of
the Russian Federation. The authors gratefully acknowledge the support by the Federal Special
Program "Scientific and Educational Cadres of Innovative Russia" (2009-2013) and Russian
Foundation for Basic Research (the research project No. 11-03-00584-а). We thank Dr. Aleshina for
help in the preparation of catalysts and Prof. Plyasova for the XRD measurements.
1
8

References
1
.
M. Hatana, A. Kayou, US Patent 5,049,692 (1991); T. Ushikubo, K. Oshima, T. Ihara, H.
Amatsu, US Patent 5,380,933 (1995), Assigned to Mitsubishi Chemical Co.
T. Ushikubo, K. Oshima, A. Kayou, M. Vaarkamp, M. Hatano, J. Catal. 169 (1997) 394-396
M. Lin, M. Linsen, US Patents 6,180,825 (2001); 6,514,901 (2003); 6,514,903 (2003) and
Patent Application EP 962,253 (1999).
2
.
.
3
4
.
.
T. Ushikubo, K. Oshima, A. Kayo, M. Hatano, Stud. Surf. Sci. Catal. 112 (1997) 473- 480.
P. De Santo Jr., D.J. Buttrey, R.K. Grasselli, C.G. Lugmair, A.F. Volpe, B.N. Toby, T. Vogt, Z.
Kristallogr. 219 (2004) 152-165.
5
6
7
8
.
.
.
J. Holmberg, R.K. Grasselli, A. Andersson, Top. Catal. 23 (2003) 55-63.
H. Tsuji, K. Oshima, Y. Koyasu, Chem. Mater. 15 (2003) 2112-2114.
P. Botella, E. García-González, A. Dejoz, J.M. López Nieto, M.I. Vazquez, J. M. González-
Calbet, J. Catal. 225 (2004) 428-438.
9
1
.
H. Murayama, D. Vitry, W. Ueda, G. Fuchs, M. Anne, J.L. Dubois, Appl. Catal. A 318 (2007)
1
37-142.
0. P. Botella, E. García-González, J.M. López Nieto, J. M. González-Calbet, Solid State Sci. 7
2005) 507-519.
(
1
1
1. D. Vitry, Y. Morikawa, J.L. Dubois, W. Ueda, Top. Catal. 23 (2003) 47-53.
2. R.K.Grasselly, J.D. Burrington, D.J. Buttrey, P. Desanto Jr., C.G. Lugmair, A.F. Volpe Jr., T.
Weingand, Top. Catal. 23 (2003) 5-22.
1
1
1
3. M. Baca, M. Aouine, J.L. Dubois, J.M.M. Millet, J. Catal. 233 (2005) 234-241.
4. J. Holmberg, R.K. Grasselli, A. Andersson, Appl. Catal. A 270 (2004) 121-134.
5. P. Korovchenko, N.R. Shiju, A.K. Dossier, U.M. Gram, M.O. Guerrero-Pérez, V.V. Guliants,
Top. Catal. 50 (2008) 43-51.
1
1
1
1
2
2
2
6. W. Ueda, D. Vitry, T. Katou, Catal. Today 96 (2004) 235-240.
7. J.M. Oliver, J.M. López Nieto, P. Botella, Catal. Today 96 (2004) 241-249.
8. .P. Concepción, P. Botella, J.M. López Nieto, Appl. Catal. A 278 (2004) 45-56.
9. M.M. Lin., T.B. Desai, F.W. Kaiser., P.D. Klugherz, Catal. Today 61 (2000) 223-229.
0. W. Ueda and K. Oshihara, App. Catal. A 200 (2000) 135-143.
1. W. Ueda, K. Oshihara, D. Vitry, T. Hisano, Y. Kayashima, Catal. Surv. Japan 6 (2002) 33-44.
2. R. Lopez-Medina, I. Sobczak, H. Golinska-Mazwa, M. Ziolek, M.A. Ba˜nares, M.O. Guerrero-
Perezc, Catal. Today 187 (2012) 195– 200
2
3. P. Botella, J.M. López Nieto, B. Solsona, A. Mifsud, F. Marquez, J. Catal. 209 (2002) 445-455.
4. J.M. López Nieto, B. Solsona, P. Concepción, F. Ivars, A. Dejoz, M.I. Vázquez, Catal. Today
2
1
57 (2010) 291-296.
1
9

2
5. W.A. Goddard III, K. Chenoweth, S. Pudar, A.C.T. van Duin, M.-J. Cheng, Top. Catal. 50
2008) 2-18.
(
2
2
6. E. Balcells, F. Borgmeier, I. Grißtede, H.-G. Lintz, Catal. Lett. 87 (2003) 195-199.
7. E. Balcells, F. Borgmeier, I. Grißtede, H. -G. Lintz, F. Rosowski, App. Catal. A 266 (2004)
2
11-221.
2
2
3
3
3
3
3
3
3
3
3
3
8. R. K. Widi, S. Bee Abd Hamid, R. Schlogl, React. Kinet. Catal. Lett. 98 (2009) 273-286.
9. P. Concepcion, S. Hernandez, J.M. Lopez Nieto, Appl. Catal. A 391 (2011) 92–101.
0. K. Asakura, K. Nakatani, T. Kubota, Y. Iwasava, J. Catal. 194 (2000) 309-317.
1. M. Vaarkamp, T. Ushikubo, Appl. Catal. A 174 (1998) 99-107.
2. N. Watanabe, W. Ueda, Ind. Eng. Chem. Res. 45 (2006) 607-614
3. V.D. Sokolovskii, Catal. Rev.-Sci. Eng. 32 (1990) 1-49.
4. B. Grzybowska-Swierkosz, Top. Catal. 11/12 (2000) 23-42.
5. J. Haber, Catal. Today 142 (2009) 100-113.
6. T.V. Andrushkevich, Catal. Rev.-Sci. Eng. 35 (1993) 213–259.
7. P.C. Stair, J. Am. Chem. Soc. 104 (1982) 4044-4052.
8. R.K. Grasselli, C.G. Lugmair, A.F. Volpe Jr., Top Catal. 50 (2008) 66–73.
9. B. Solsona, V.A. Zazhigalov, J.M. Lopez Nieto, I.V. Bacherikova, E.A. Diyuk, Appl. Catal. A
2
49 (2003) 81–92.
0. F. Ivars, B. Solsona, P. Botella, M.D. Soriano, J.M. Lopez Nieto, Catal. Today 141 (2009)
94–299.
4
2
4
4
4
1. P. Botella, P. Concepcion, J.M. Lopez Nieto, B. Solsona, Catal. Lett. 89 (2003) 249-253.
2. M. Baca, A. Pigamo, J.L. Dubois, J.M.M. Millet, Catal. Commun. 6 (2005) 215-220.
3. S. Hernández-Morejudo, A. Mass, E. García-González, P. Concepción, J.M. López Nieto,
Appl. Catal. A (2015) In Press doi:10.1016/j.apcata.2014.12.039
4
4. G.Ya. Popova, T.V. Andrushkevich, L.S. Dovlitova, G.A. Aleshina, Yu.A. Chesalov, A.V.
Ishenko, E.V. Ishenko, L.M. Plyasova, V.V. Malakhov, M.I. Khramov, Appl. Catal. A: Gen.
3
53 (2009) 249–257
5. R.G. Kukushkin, O.A. Bulavchenko, V.V. Kaichev, V.A. Yakovlev, Appl. Catal. B 163 (2015)
31-538.
6. H. Hayashi, N. Shigemoto, S. Sugiyama, N. Masaoka, K. Saitoh, Catal. Lett. 19 (1993) 273-
77.
4
4
4
5
2
7. D. Morris, Y. Dou, J. Rebane, C.E.J. Mitchell, R.G. Egdell, D.S.L. Law, A. Vittadini, M.
Casarin, Phys. Rev. B 61 (2000) 13445.
2
0

4
8. V.V. Kaichev, G.Ya. Popova, Yu.A. Chesalov, A.A. Saraev, D.Y. Zemlyanov, S.A.
Beloshapkin, A. Knop-Gericke, R. Schlögl, T.V. Andrushkevich, V.I. Bukhtiyarov, J. Catal.
3
11 (2014) 59-70.
9. C. Wang, C. Shao, L. Wang, L. Zhang, X. Li, Y. Liu, J. Colloid Interface Sci. 333 (2009) 242-
48.
4
5
5
5
5
2
0. E.V. Ishchenko, A.V. Ishchenko, V.M. Bondareva, T.Yu. Kardash, V.I. Sobolev, T.V.
Andrushkevich, Kinet. Catal. (2015) In Press
1. E.V. Ishchenko, G.Ya. Popova, T.Yu. Kardash, A.V. Ishchenko, L.M. Plyasova, T.V.
Andrushkevich, Catal. Sust. Energ. 1 (2013) 75–81.
2. E. Sadovskaya, V. Goncharov, G. Popova, E. Ishchenko, D. Frolov, A. Fedorova, T.
Andrushkevich, J. Molec. Catal. A: Chem. 392 (2014) 61–66
3. http://www.chemicals-technology.com/projects/-ptt-asahi-chemical-acrylonitrile-methyl-
methacrylate/
5
5
4. J.F. Brazdil, Top. Catal. 38 (2006) 290-294.
5. M. Hävecker, S. Wrabetz, J. Kröhnert, L.-I. Csepei, R. Naumann d’Alnoncourt, Yu.V.
Kolen’ko, F. Girgsdies, R. Schlögl, A. Trunschke, J. Catal. 285 (2012) 48–60.
6. V.V. Guliants, Catalysis 27 (2015) 33–61
5
5
7. Т.V. Andrushkevich, G.К. Boreskov, L.L. Kuznetsova, L.М. Plyasova, Yu.N. Tyurin, Yu.М.
Schekochihin, Kinet. Catal. 15 (1974) 424-429.
58. Z. Song, T. Matsushita, T. Shishido, K. Takehira, J. Catal. 218 (2003) 32–41.
2
1

Figure Captions
CH- CH- CH
3
2
3
k1
k5
k6
СН =С Н- СН
2
3
k2
CH =C H- CN
k4
k3
2
COx
Scheme 1. Mechanism of propane ammoxidation [32].
2
2

1 x x
Fig. 1. XRD patterns of Mо V0.28Te0.14Nb0.13O and MоVTeNbBiO catalysts.
2
3

Fig. 2. The selectivity towards propylene (C
acetonitrile and hydrocyanic acid (C N+HCN) as a function of propane conversion on the
Mo (curve 1) and Mо (curve 2) catalysts.
3 6 3 3 x 3 4 2
H ), acrylonitrile (C H N), CO , acrylic acid (C H O ),
2
H
3
1
V0.28Te0.14Nb0.13
O
x
1 x
V0.28Te0.18Nb0.11Bi0.01O
2
4

Fig. 3. Propane conversion (1) and selectivity towards reaction products (2-8) versus concentration of
C
3
H
6
(a); NH
3
(b) and O
2
(c). 2 - C
3
H
3
N, 3 - CO
2
, 4 - CO, 5 - C
2
H
3
N, 6 - C
3
H
4
O
2
, 7 - HCN, 8 - C
3
H
6
.
Mо V0.28Te0.18Nb0.11Bi0.01O catalyst was tested.
1 x
2
5

1 x
Fig. 4. Acrylonitrile yield versus the ratio of component concentrations. Mо V0.28Te0.18Nb0.11Bi0.01O
catalyst was tested.
2
6

Fig. 5. Dependence of acrylonitrile yield on propane conversion at the reaction temperature 420 C on
the Mo (curve 1) and Mо (curve 2) catalysts.
1
V
0.28Te0.14Nb0.13
O
x
1 x
V0.28Te0.18Nb0.11Bi0.01O
2
7

+
Fig. 6. Dependence of NH
4
relative covering upon temperature obtained in the TPD experiment on
the Mo
1
V
0.28Te0.14Nb0.13
O
x
1 x
(curve 1) and Mо V0.28Te0.18Nb0.11Bi0.01O (curve 2) catalysts.
2
8

Tables
Table 1. Physicochemical properties of the samples under study
Chemical composition of the catalysts
2
before calcinations
after calcination
S
BET, m /g
Mо
Mо
Mо
1
1
1
V
V
V
0.3Te0.23Nb0.12
Mо
Mо
Mо
1
1
1
V
V
V
0.28Te0.14Nb0.13
7.3
0.3Te0.23Nb0.12Bi0.01
0.3Te0.23Nb0.12Bi0.02
0.28Te0.18Nb0.11Bi0.01
0.3Te0.17Nb0.12Bi0.02
7.0
6.1
Mо V Te0.23Nb0.12Bi0.05
Mо V0.28Te0.23Nb0.14Bi0.05
5.0
1
0.3
1
2
9