Features of propane conversion in the presence of SmVO3 and SmVO4

Article abstract of DOI:10.1134/S0036024416090193

Features of propane conversion in the presence of samarium vanadite and samarium vanadate, both produced via solid-phase synthesis, are studied. It is shown that SmVO₃ catalyzes mainly the propane cracking process to form methane and ethylene, while SmVO₄ equally accelerates both cracking and the dehydrogenation of propane. Based on the results from catalytic experiments, energies of activation are calculated for the thermal cracking of propane (104 kJ/mol) and the conversion of propane in the presence of SmVO₃ (39 kJ/mol) and SmVO₄ (42 kJ/mol). The thermal stability of SmVO₄ in a hydrogen atmosphere is studied via temperature-programmed reduction, while SmVO₃ stability in an oxidizing environment is studied by DTA. Energies of activation for the reduction of SmVO₄ (75 kJ/mol) and the oxidation of SmVO₃ (244 kJ/mol) are calculated using the Kissinger method.

Full text of DOI:10.1134/S0036024416090193

ISSN 0036-0244, Russian Journal of Physical Chemistry A, 2016, Vol. 90, No. 9, pp. 1754–1759. © Pleiades Publishing, Ltd., 2016.
Original Russian Text © E.B. Markova, A.S. Lyadov, V.V. Kurilkin, 2016, published in Zhurnal Fizicheskoi Khimii, 2016, Vol. 90, No. 9, pp. 1336–1341.
CHEMICAL KINETICS
AND CATALYSIS
Features of Propane Conversion in the Presence
of SmVO and SmVO
3
4
a
b
a
E. B. Markova , A. S. Lyadov , and V. V. Kurilkin
a
Peoples’ Friendship University of Russia, Moscow, 117198 Russia
b
Topchiev Institute of Petrochemical Synthesis, Russian Academy of Sciences, Moscow, 119991 Russia
e-mail: lyadov@ips.ac.ru
Received October 22, 2015
Abstract—Features of propane conversion in the presence of samarium vanadite and samarium vanadate,
both produced via solid-phase synthesis, are studied. It is shown that SmVO catalyzes mainly the propane
3
cracking process to form methane and ethylene, while SmVO equally accelerates both cracking and the
4
dehydrogenation of propane. Based on the results from catalytic experiments, energies of activation are cal-
culated for the thermal cracking of propane (104 kJ/mol) and the conversion of propane in the presence of
SmVO (39 kJ/mol) and SmVO (42 kJ/mol). The thermal stability of SmVO in a hydrogen atmosphere is
3
4
4
studied via temperature-programmed reduction, while SmVO stability in an oxidizing environment is stud-
3
ied by DTA. Energies of activation for the reduction of SmVO (75 kJ/mol) and the oxidation of SmVO
244 kJ/mol) are calculated using the Kissinger method.
4
3
(
Keywords: samarium vanadate, samarium vanadite, propane cracking, propane dehydrogenation, DTA, tem-
perature-programmed reduction, Kissinger method, energy of activation
DOI: 10.1134/S0036024416090193
INTRODUCTION
ning of the 20th century [6]. In the Oleflex process
developed by UOP, propane dehydrogenation pro-
ceeds in the presence of the Pt/Sn/Al O catalyst in an
There has recently been a great increase world-
wide in the production of propylene, an important
large-scale intermediate product for subsequent syn-
thesis of polypropylene, acrylonitrile, and propylene
oxide [1]. There are several methods for the indus-
trial production of propylene; most prominent
among these are the cracking of naphta and the FCC
process [2]. Substantial disadvantages of these pro-
cesses are the variety of the products that form,
which requires the creation of a complex system of
separation, and a low yield of propylene. Obtaining
propylene through the dehydrogenation of propane
is devoid of these shortcomings, and the process pro-
ceeds according to the reaction
2
3
adiabatic reactor with a moving bed at atmospheric
pressure and temperatures of 580–650°C. These cata-
lysts have several disadvantages, however, making the
search for new catalytic systems worthwhile.
Several studies [7–10] have shown that com-
pounds with vanadium-containing anions are capa-
ble of catalyzing the dehydrogenation of light hydro-
carbons. It is also known that rare-earth elements
can be used as the components of dehydrogenation
catalysts [11–13] and can increase their durability.
The development of catalytic systems containing
both vanadium and rare-earth elements is thus very
promising. Note that rare-earth vanadates have zir-
con-like structures [14], while rare-earth vanadites
have perovskite-like structures [15]. These types of
structures have high thermal stability, while per-
ovskite-like structures are permeable to oxygen.
C H → C H + H , ΔH° = 124 kJ/mol.
3
8
3
6
2
Despite the apparent simplicity of the reaction, it
occurs only at high temperatures (over 600°C) due to
its high endothermicity [3], which leads to rapid cata-
lyst deactivation resulting from surface carbonization
The aim of this work was to investigate features of
[
4]. In addition to the main reaction, a side reaction is propane conversion in the presence of samarium
also possible:
C H → CH + C H , ΔH° = –56 kJ/mol [5].
vanadate and samarium vanadite.
3
8
4
2
4
EXPERIMENTAL
Analytically pure vanadium(V) and samarium(III)
Another large-scale intermediate product, ethylene, is
a product of the side reaction. It is known that the
most active dehydrogenation catalysts are platinum- oxides were used as the initial materials for the synthe-
containing systems that were developed at the begin- sis of samarium vanadate SmVO
754
. The initial compo-
4
1

FEATURES OF PROPANE CONVERSION
1755
nents were preliminarily calcined in a muffle furnace
Features of the reduction of samarium vanadate
at 450°C for 5 h to remove moisture and adsorbed was studied via temperature-programmed reduction
CO . SmVO was prepared via high-temperature solid- (TPR) on a Micromeritics Auto Chem II device. A
2
4
phase synthesis, according to the reaction
fine sample (a fraction smaller than 0.05 mm) was
placed in a U-shaped quartz reactor equipped with a
thermocouple and was heated in a stream of a gas mix-
ture to a temperature of 950°C. The gas mixture used
in this work contained 5 vol % of hydrogen in argon.
The feed rate of the gas mixture was 50 mL/min and
the heating rates were 0.083, 0.167, and 0.333 K/s. The
temperature was measured inside the reactor (the
thermocouple was placed 5 mm above the surface of
the sample) and the reducing agent absorbance was
recorded by a thermal conductivity detector.
Sm O + V O → 2SmVO .
2
3
2
5
4
This required thorough homogenization of stoichio-
metric amounts of vanadium(V) and samarium(III)
oxides by grinding in the presence of ethyl alcohol in
an agate mortar, compression into tablets, and subse-
quent step annealing in a muffle furnace at 550°C for
3
h, at 650°C for 3 h, and at 750°C for 3 h. The step-
wise temperature rise was due to the possibility of V O
2
5
melting (T = 670°C). The annealed samples were
mp
Samarium vanadite oxidation was studied on a
MOM Q-1500D derivatograph (Hungary) in the tem-
perature range of 298–1298 K at heating rates of
grinded again in order to homogenize them and then
annealed for another 4 h at a temperature of 950°C to
obtain a well-formed structure.
0
.083, 0.167, and 0.333 K/s in air in a platinum cruci-
Samarium vanadite SmVO was prepared from
3
ble. The sample weight was 100 mg.
SmVO via high-temperature reduction. A sample of
4
The energies of activation of the reduction of
samarium vanadate, and of the oxidation of samarium
vanadite were calculated using the Kissinger method,
based on the TPR and DTA data obtained at different
rates of sample heating and from the inclination of a
2
finely-ground SmVO was placed in a molybdenum
4
boat in a tube furnace, heated in hydrogen to a tem-
perature of 900°C for 1 h, and allowed to stand at this
temperature for a further 4 h. The volumetric hydro-
−1
gen feed rate was 500 h .
line built in Arrhenius coordinates ln(b/T ) –
max
All of the synthesized samples were studied via
X-ray powder diffraction (XPD). XPD was performed
on a Rigaku X-ray diffractometer (Japan) equipped
3
1
0 /T [16, 17].
max
Features of propane cracking in the presence of
samarium vanadate and samarium vanadite were stud-
ied in a flow apparatus with a U-shaped quartz reactor
under stationary conditions at atmospheric pressure
and over a wide range of temperatures. High-purity
propane (99.98 wt %) was used as the initial reagent.
The feed rate was constant in all experiments:
with an Ultima IV theta-theta goniometer (CuK radi-
α
ation) at a scanning step of 0.02° and a scanning time
1 s. The range of diffraction angle measurement was
2θ = 10°–100°.
The X-ray powder diffraction patterns were
indexed by means of homology, based on data from
the international ICDD PDF-4 database. The crystal
lattice parameters were refined using the RTP (X-ray
Table Processor) program.
5
5.8 μmol/s. The propane cracking temperature was
raised from 298 to 1143 K in steps of 50 K. Instant cok-
ing of the catalyst occurred at temperatures higher
than 1143 K as a result of the condensation reactions of
the formed unsaturated and aromatic hydrocarbons,
minimizing the formation of olefins.
The parameters of the samples’ porous structure
were determined from adsorption isotherms of nitro-
gen at 77 K, measured on a Micromeritics ASAP 2020
The catalyst load was equal in all experiments. The
MR automatic high-vacuum unit (United States) in rate of propane conversion was calculated using the
the 0.001 to 0.98 range of relative vapor pressures. The equation
samples were preliminarily evacuated to a residual
−7
Kw s
Vloop
vacuum of less than 10 mmHg without heating. The
ex
w =
[μmol/(g s)],
following parameters were determined: S
was the
BET
specific surface area, calculated by means of BET; W
o
where K is a correction coefficient, w is the exit
was volume; S was the relative surface of micropores
ex
mic
velocity of the reaction mixture, s is the area of the
in the samples, calculated using the t-plot comparative
peak, and V is the loop volume.
was the mesopore volume, calculated
loop
method; W
BJH
using the Barrett–Joyner–Halenda method (BJH);
The selectivity of each component was calculated
2
2
x was the pore size (two half-widths for slit model according to the formula
о
о
x ) determined in the BET region and the area of
nx
nin − n
capillary condensation on the desorption branch using
the Barrett–Joyner–Halenda model (BJH); and
Sx =
×100%,
rest
2
x
was the average pore size in capillary conden- where n is the amount of propane involved in the
о BJH
in
sation, determined using the Barrett–Joyner–Hal- reaction (μmol), and n is the amount of unreacted
rest
enda method.
propane (μmol).
RUSSIAN JOURNAL OF PHYSICAL CHEMISTRY A Vol. 90 No. 9 2016

1756
MARKOVA et al.
Table 1. Parameters of the porous structures of SmVO and SmVO
4
3
3
3
W , cm /g
2x , nm
BET
2xо BJH, nm
des/ads
о
WBJH, cm /g
о
2
Catalysts
SBET, m /g
2
Smic, m /g
des/ads
SmVO3
SmVO4
2.5
3.1
0.01/0.5
0.047/0.045
0.053/0.052
12.3
18.2
13.2/13.3
19. 6/20.2
''
Table 2. Results from propane conversion (S is selectivity)
Catalysts
–
T, K
α
S(CH ), %
S(C H ), %
S(C H ), %
S(C H ), %
4
2
6
2
4
3
6
923
0.20
0.21
0.23
0.50
0.60
0.95
0.55
0.95
1.00
75
71
73
30
23
21
24
15
11
25
–
–
–
–
–
–
–
9
73
023
923
73
023
923
73
023
29
27
–
–
–
–
–
–
1
SmVO3
SmVO4
70
69
72
67
43
44
9
8
1
7
9
9
42
45
1
The results obtained via catalytic conversion were
The catalytic activity of samarium vanadate and
compared to the thermal cracking data. The degree of samarium vanadite (SmVO and SmVO ) in propane
4
3
propane conversion was determined from the amount conversion was studied for the first time. For compar-
of reacted propane when a steady state was achieved:
ison, we conducted the noncatalytic conversion of
propane. It was shown that propane conversion started
at 773 K, was only 2% at 873 K, and raising the tem-
perature to 923 K increased propane conversion to
nin − n
rest
α =
.
nin
2
(
1%; methane and ethane were the major products
Table 2). Propane dehydrogenation thus does not
RESULTS AND DISCUSSION
occur under conditions of noncatalytic thermal crack-
ing. In the presence of SmVO and SmVO , a consid-
3
4
The synthesized samarium vanadate and samarium
vanadite were analyzed via X-ray powder diffraction to
determine the structure and refine the parameters of
erable increase in propane conversion was observed
that reached 50% at 923 K. In addition to shifting the
degree of half-conversion toward lower temperatures,
the selectivity with regard to olefin changed with
respect to thermal cracking (Table 2). In the presence
the unit cells. SmVO crystallized in the zircon struc-
4
tural type (space group I4 /amd, a = 7.272 ± 0.006 Å,
1
and b = 6.391 ± 0.006 Å), while SmVO had a per-
3
of SmVO , the cracking of propane to form ethylene
3
ovskite-like structure (space group Pbnm, a = 5.384 ±
(
9
selectivity 70%) predominates at a temperature of
0
0
.008 Å, b = 5.580 ± 0.005 Å, and c = 7.667 ±
73 K. A further increase in temperature leads to a
slight rise in selectivity with respect to ethylene (up to
2%). Over the investigated range of temperatures,
selectivity to propylene does not exceed 8%. SmVO
.008 Å). No impurity phases were found in the syn-
thesized samples, and the unit cell parameters corre-
sponded to the literature data [17, 18].
7
3
with perovskite-like structure catalyzes the propane
cracking process to form ethylene, while dehydroge-
nation in the presence of the catalyst proceeds slowly.
In contrast, the selectivity toward propylene in the
presence of SmVO is 9% at 923 K and reaches 45% at
4
The parameters of the pore structure were deter-
mined for these samples (Table 1). The SmVO and
3
SmVO synthesized using the solid-phase method had
4
2
low specific surface areas of about 3 m /g and low vol-
umes of pores.
RUSSIAN JOURNAL OF PHYSICAL CHEMISTRY A
Vol. 90
No. 9
2016

FEATURES OF PROPANE CONVERSION
1757
lnW
2
Degree of conversion
2
1
0.9
0.7
0.5
0.3
0.1
1
2
3
1
0
0
.85 0.95 1.05 1.15 1.25 1.35 1.45 1.55 1.65 1.75
3
1/T, K × 10
−
1
2
3
4
−
−
−
0
20
40
60
t, h
Fig. 1. Reaction rate as a function of reciprocal tempera-
Fig. 2. Degrees of propane conversion as a function of on-
ture for (1) thermal cracking, (2) SmVO , and (3) SmVO .
stream time for the (1) SmVO and (2) SmVO catalysts at
3
4
3
4
1023 K.
0.20
0.16
0.12
0.08
0.04
0
1 2
3
(b)
(
а)
1
6.2
5.8
1
1
5.4
5.0
y = 9.0401x + 7.3381
2
R = 0.9983
1
7
00
900
T, °С
1100
0.85
0.90
0.95
1.00
3
10 /Tmax
2
3
1
8
4
16.2
15.8
15.4
15.0
(d)
(c)
0
900
1000
1100
T, °С
−
4
8
y = 29.34x − 13.752
2
R = 0.9894
−
0
.96 0.98
1.00
1.02
1.04
3
1
0 /Tmax
Fig. 3. Results from SmVO TPR and our thermal analysis of SmVO : (a) TPR profile of SmVO at heating rates of (1) 0.083, (2)
4
3
4
0
.167, and (3) 0.333 K/s; (b) Arrhenius dependence for the reduction of SmVO ; (c) DTA profiles for SmVO at heating rates of
4
3
(
1) 0.083, (2) 0.167, and (3) 0.333 K/s; (d) Arrhenius dependence for the oxidation of SmVO .
3
1
023 K (Table 2). It is clear that at 100% propane con- a zircon-like structure, the cracking process and the
version, the yield of targeted products ethylene and dehydrogenation of propane proceed equally.
propylene was maximal, with only 11% selectivity with
respect to methane. On the SmVO catalyst, which has tion [18], the energies of activation for the process
Since the cracking of propane is a first order reac-
4
RUSSIAN JOURNAL OF PHYSICAL CHEMISTRY A Vol. 90 No. 9 2016

1758
MARKOVA et al.
were calculated for all of the studied catalysts. This can oxidize to form vanadate. During thermal analy-
required constructing the patterns of the reaction rate sis, it was found that SmVO is stable in air up to 900 K
3
logarithm as a function of the reciprocal temperature (Fig. 3c). The energy of activation of vanadite samar-
(
Fig. 1). The experimental results were approximated ium oxidation, determined using the Kissinger
by a straight line, allowing us to calculate the process’s method (Fig. 3d), was 244 kJ/mol. Even though the
energy of activation.
temperatures of SmVO reduction and SmVO oxida-
4
3
The energy of activation calculated from these data tion were very close, the energies of activation of these
was found to be 104 kJ/mol for the thermal cracking of processes differed by a factor of exactly four.
propane, which was close to the literature data [19];
for the reactions using the SmVO and SmVO cata-
3
4
CONCLUSIONS
lysts, it was 39 and 42 kJ/mol, respectively.
For the catalysts, the energy of activation over the
investigated range of temperatures remained constant,
indicating that the process proceeded according to a
carbene mechanism, and the reaction did not transit
to the gas phase even at high temperatures.
The conversion of propane was studied in the
presence of samarium vanadate and samarium vana-
dite synthesized by solid-phase method. It was
shown that SmVO catalyzes mainly the process of
3
propane cracking to form methane and ethylene,
During the cracking of propane, the formation of and SmVO equally accelerates both cracking and
4
CH radicals is accompanied by that of free carbon,
x
the dehydrogenation of propane. Based on the
which in turn can react with the forming hydrogen to results from experiments to determine catalytic
form methane and other hydrocarbons. The interac- activity, the energies of activation were calculated for
tion between the free carbon atoms begins to dominate the thermal cracking of propane (104 kJ/mol) and
as the temperature rises, resulting in carbonization.
the conversion of propane in the presence of SmVO
3
Our study of catalyst stability (Fig. 2) showed that (39 kJ/mol) and SmVO
(42 kJ/mol). The thermal
4
SmVO exhibits high catalytic activity for 30 h, while stability of SmVO in a hydrogen atmosphere was
4
4
propane conversion was 90%. This was followed by studied by means of TPR, and the stability of SmVO
3
rapid catalyst deactivation; after 50 h, propane conver- in an oxidizing environment was studied via DTA;
sion was only 20%. Deactivation was much faster with the energies of activation were calculated for these
SmVO , and it was observed after 20 h of operation at processes.
3
a rate that was slightly slower than the one for SmVO .
4
These results agree with the data on the selectivity of
product formation. It was found that SmVO better
REFERENCES
4
catalyzes the dehydrogenation process, as a result of
which the reaction mixture contains hydrogen. To
some extent, this prevents the formation of condensa-
tion products on the catalyst’s surface, which in turn
leads to a loss of catalytic activity.
1
2
. Market Study: Propylene (UC-1705) (Ceresana
Research, 2011).
. L. M. Wolschlag and K. A. Couch, UOP FCC Innova-
tions Developed Using Sophisticated Engineering
Tools (2010). http://www.uop.com/?document=uop-
fcc-innovation-tech-paper&download=1.
Since hydrogen is produced during the dehydroge-
nation of propane, and the process takes place at high
temperatures, the ability of samarium vanadate to be
reduced in a hydrogen stream was studied by means of
3. L. Låte, J.-I. Rundereim, and E. A. Blekkan, Appl.
Catal. A 262, 53 (2004).
TPR. The TPR profile of SmVO was characterized by
4. S. Sahebdelfar and F. Tahriri Zangeneh, Iran. J. Chem.
4
one well-formed hydrogen absorption peak (Fig. 3a)
Eng. 7 (2), 51 (2010).
+
5
that corresponded to vanadium reduction (V
→
5. Y. Zhang, Y. Zhou, A. Qiu, et al., Ind. Eng. Chem. Res.
+3
V ). The reduction process began at ~900 K, and the
maximum hydrogen absorption depended on the rate
of heating. At a heating rate of 0.167 K/s, maximum
absorption occurred at 1097 K. Figure 3b shows the
rate of heating as a function of the reciprocal tempera-
ture in Arrhenius coordinates, from which we can eas-
ily determine the energy of activation for samarium
vanadate reduction 75 kJ/mol. It should thus be noted
45, 2213 (2013).
6. L. Duck Hyun, L. Won Jun, L. Won Jong, et al., Phys.
Rev. Lett. 106, 170401 (2011).
7. M. C. Abello, M. F. Gomez, and L. E. Cadus, Ind. Eng.
Chem. Res. 35, 2137 (1996).
8
9
. W. Ueda, K. H. Lee, Y. S. Yoon, and Y. Moro-oka,
Catal. Today 44, 199 (1998).
that SmVO can be reduced during the dehydrogena-
tion process, which could adversely affect the selectiv-
4
. J. E. Miller, N. B. Jackson, L. Evans, et al., Catal. Lett.
58, 147 (1999).
ity with respect to the target products.
10. J. D. Pless, et al., J. Catal. 223, 419 (2004).
The thermal stability of samarium vanadite was
also studied in this work, since it can be used in some 11. R. Zhou, Y. Cao, S. Yan, and K. Fan, Appl. Catal. A:
cases as a catalyst for oxidation processes [20, 21], and
Gen. 236, 103 (2002).
RUSSIAN JOURNAL OF PHYSICAL CHEMISTRY A
Vol. 90
No. 9
2016

FEATURES OF PROPANE CONVERSION
1759
1
2. Y. Zhang, Y. Zhou, J. Shi, et al., Fuel Process. Technol. 18. J. E. Germain, Catalytic Conversion of Hydrocarbons
1
11, 94 (2013).
3. B. K. Vu et al., Catal. Today 164, 214 (2011).
4. Yoshio Oka, Takeshi Yao, and Naoichi Yamamoto,
J. Solid State Chem. 152, 486 (2000).
(Academic Press, London, 1969).
1
1
19. A. H. Kootenaei, J. Towfighi, A. Khodadadi, and
Y. Mortazavi, Appl. Surf. Sci. 298, 26 (2014).
20. S. Keav, S. Kumar Matam, D. Ferri, and A. Weiden-
15. D. Saha, S. Mahapatra, T. N. Guru Row, and
kaff, Catalysts 4, 226 (2014).
G. Madras, Ind. Eng. Chem. Res. 48, 7489 (2009).
21. N. M. Panich, G. V. Pirogova, R. I. Korosteleva, and
16. H. E. Kissinger, Anal. Chem. 29, 1702 (1957).
Yu. V. Voronin, Russ. Chem. Bull. 48, 694 (1999).
1
7. A. Y. Krylova, A. S. Lyadov, S. A. Sagitov, et al., Russ.
J. Phys. Chem. A 85, 55 (2011).
Translated by V. Avdeeva
RUSSIAN JOURNAL OF PHYSICAL CHEMISTRY A Vol. 90 No. 9 2016