Carbon Dioxide as Feedstock in Selective Oxidation of Propane

Article abstract of DOI:10.1002/ejic.201701049

Carbon dioxide is a promising nonconventional oxidant for catalytic dehydrogenation of lower alkanes to olefins. The lower catalytic activity of CO₂ compared to oxygen, which is typically used in oxidative dehydrogenation (ODH), also means that more reactive olefins are less likely to undergo undesirable combustion in the presence of CO₂. Supported vanadia catalysts were synthesized by incipient wetness impregnation of silica supports, and by varying the surface coverage of VO_x species by means of sodium on the silica surface, the nature of the catalyst was investigated. Promotion with Na cations significantly improved the dispersion of surface VO_x species but resulted in the formation of surface Na metavanadate or another reduced V³⁺/V⁴⁺ phase, which lacked catalytic activity in propane ODH in the presence of CO₂.

Full text of DOI:10.1002/ejic.201701049

A Journal of
Accepted Article
Title: Carbon Dioxide as Feedstock in Selective Oxidation of Propane
Authors: Jungshik Kang, Andrew Czaja, and Vadim Guliants
This manuscript has been accepted after peer review and appears as an
Accepted Article online prior to editing, proofing, and formal publication
of the final Version of Record (VoR). This work is currently citable by
using the Digital Object Identifier (DOI) given below. The VoR will be
published online in Early View as soon as possible and may be different
to this Accepted Article as a result of editing. Readers should obtain
the VoR from the journal website shown below when it is published
to ensure accuracy of information. The authors are responsible for the
content of this Accepted Article.
To be cited as: Eur. J. Inorg. Chem. 10.1002/ejic.201701049
Link to VoR: http://dx.doi.org/10.1002/ejic.201701049

10.1002/ejic.201701049
European Journal of Inorganic Chemistry
FULL PAPER
Carbon Dioxide as Feedstock in Selective Oxidation of Propane
Jungshik Kang,[a] Andrew D. Czaja[b] and Vadim V. Guliants*[a]
Abstract: Carbon dioxide is a promising nonconventional oxidant for
catalytic dehydrogenation of lower alkanes to olefins. Lower catalytic
activity of CO2 as compared to oxygen that is typically used in
oxidative dehydrogenation (ODH) also means that more reactive
olefins are less likely to undergo undesirable combustion in the
presence of CO2. Supported vanadia catalysts synthesized by
incipient wetness impregnation of silica support, and by varying the
surface coverage of VOx species using sodium on the silica surface,
the nature of the catalyst was investigated. The promotion with Na
cations significantly improved the dispersion of surface VOx species
and resulted in the formation of surface Na-metavanadate or another
reduced V3+/V4+ phase, which lacked catalytic activity in propane ODH
in the presence of CO2.
Selective oxidation of lower alkanes has been a long-standing
fundamental and practical challenge, primarily due to inertness of
C-H bonds (e.g., the secondary C-H bond energy of 96 kcal/mol
in propane) requiring energy-intensive conditions to achieve
practical conversion and limited process selectivity due to greater
reactivity of the partial oxidation products, e.g., propylene.[5,6]
Recent reports indicated that CO2 is a mild oxidant that can
perform oxidative dehydrogenation (ODH) of lower alkanes (C2-
C4) into corresponding olefins (1), all of which are highly valuable
chemical intermediates:
CnH2n+2 + CO2
à
CnH2n + CO + H2O
(1)
Ethane and propane ODH by CO2 is expected to produce a 1:1
molar mixture of olefins with carbon monoxide that can be directly
used in a variety of downstream chemical processes, such as co-
polymerization to produce valuable thermoplastic poly-ketones[7]
Introduction
and hydro-formylation[8] to make
intermediates.
a
variety of aldehyde
Carbon dioxide (CO2) and C2-C3 alkanes are well-known
greenhouse gases that are generated by human. Our planet is
projected to undergo significant environmental changes this
century with potentially devastating consequences for the global
economy and the natural world if humanity fails to reduce CO2
emissions from burning fossil fuels. The electricity and heat
generation combined with the use of hydrocarbon-based fuels in
transportation have been the largest contributors of CO2, creating
almost two-thirds of global emissions (e.g., 5,271 million metric
tons of the total U.S. energy-related CO2 emissions in 2015).[1] On
the other hand, natural gas production continues to increase as
shale gas is rapidly becoming an alternative source of natural gas
and its production is currently one of the fastest growing segments
of the US oil and gas industry.[2] While C2-C3 alkanes are the
major components of shale gas, they currently have no significant
applications as chemical feedstocks due to high chemical stability
of these molecules and the lack of technologies that can transform
them selectively into chemical intermediates. Therefore, it is
highly desirable to develop novel chemical processes that can
jointly employ CO2 and C2-C3 alkanes in chemical synthesis.[1, 3]
The total annual amount of CO2 used in industry as a chemical
feedstock to make urea, inorganic carbonates, pigments,
methanol and salicylic acid, is approximately two orders of
magnitude lower than its annual atmospheric emissions.[4]
Therefore, it is highly desirable to develop new large-scale
chemical processes that utilize CO2 to more fully explore the
potential of its fixation into value-added products.
Supported vanadia and chromia catalysts have been identified
in recent studies as the most promising catalysts for lower alkane
ODH by CO2.[9-15] These catalysts are typically synthesized by
incipient wetness impregnation of oxide supports, such as SiO2,
Al2O3, CeO2, and ZrO2, by solutions of V(V) and Cr(III) precursors,
such as ammonium metavanadate (NH4VO3) and Cr(III) nitrate,
followed by drying and thermal activation at ~350oC in air to
produce well-dispersed, tetrahedrally coordinated molecular VOx
and CrOx species on the surface of oxide supports. The catalytic
activity and selectivity of these supported VOx and CrOx catalysts
in lower alkane ODH is associated with the presence of
monomeric and polymeric VOx and CrOx species at sub-
monolayer surface coverage. Therefore, the catalytic behavior of
these catalysts may be tuned by varying the surface coverage of
VOx and CrOx species, the nature of the oxide support, and
addition of promoter species.[9-15]
In a recent study,[15] the silica-supported vanadia catalysts have
been extensively investigated and found to be highly promising
for the propane ODH using molecular oxygen. However, the
silica-supported catalysts display significantly lower monolayer
coverage of vanadia (ca. 3.3 V/nm2) as compared to other, more
reactive oxide supports, such as alumina (7-9 V/nm2). The
promotion of silica with Na+ cations resulted in significantly
enhanced dispersion of monomeric vanadia species on SiO2.[16]
This effect was found to be optimal at a Na+/V ratio of 0.2, and
above this ratio the formation of sodium metavanadate resulted in
the loss of catalytic activity. According to Grant et al.,[16] the Na+
reacts with surface silanols to form more reactive Si−O−Na+
anchoring sites that bond to several monomeric vanadia species.
Therefore, the use of the Na+ species to promote VOx dispersion
on the silica surface offers new opportunities to design novel 2D
metal oxide catalysts supported on SiO2. According to Grant et
al.,[16] greater surface coverage of 2D vanadia species in Na-
promoted catalysts leads to higher propane ODH rate while
maintaining high propylene selectivity. However, the behavior of
Na-promoted vanadia catalysts in propane ODH by CO2 has not
[a]
[b]
Professor, V. V. Guliants, Dr. J. Kang, Department of Chemical and
Environmental Engineering, University of Cincinnati, Cincinnati, OH
45221, E-mail: vadim.guliants@uc.edu
Associate Professor, A.D., Czaja, Department of Geology,
University of Cincinnati, Cincinnati, OH 45221.
Supporting information for this article is given via a link at the end of
the document.
This article is protected by copyright. All rights reserved.

10.1002/ejic.201701049
European Journal of Inorganic Chemistry
FULL PAPER
been investigated. Therefore, in this study we synthesized Na-
promoted vanadia catalysts supported on silica and investigated
their behavior in two propane ODH reactions employing molecular
oxygen and CO2.
700oC. However, no reactions were observed over the silica
support at relevant temperatures (600oC and below) employed in
this study to investigate propane ODH reactions. Moreover, while
propane ODH in the presence of O2 and CO2 was investigated
over silica-supported vanadia catalysts at 350-600oC, significant
propane conversion was observed only at 550 and 600oC.
Results and Discussion
Table 2 shows the initial propane conversion and product
selectivity of propane ODH over VOx/SiO2 catalysts at 550oC. In
the case of O2, the formation of C1 and C2 products increased
with vanadia content in these catalysts. Also, as expected the
conversion of propane was considerably higher during O2 ODH
as compared to CO2 ODH reaction. However, the propylene yield
was similar for both CO2 and O2 ODH reactions due to a greater
extent of cracking reactions during O2 ODH. Also, the propane
conversion over S1 catalyst (1.9 V/nm2) was 18 mol.% for O2 ODH
and 9.2 mol. % for CO2 ODH, while it increased with vanadia
content, e.g., to 25 mol. % for the S2 catalyst for both O2 and CO2
ODH reactions, until reaching a maximum for the S3 catalyst. The
maximum yield of propylene of ca. 11 mol. % was observed over
S2 catalyst (3.1 V/nm2). It should be noted that the S1 catalyst
corresponded to ca. 50% of theoretical monolayer coverage of
vanadia on silica, whereas the S2 and S3 catalysts are
characterized by ca. 90% monolayer and 120% monolayer
coverages, respectively. Therefore, based on initial reactivity of
VOx/SiO2 catalysts in these 2 ODH reactions, one may conclude
that the presence of a monolayer of surface VOx species is
important for the reactivity in these propane ODH reactions.
Table 1 shows the composition of model supported vanadia
catalysts synthesized in this study. A series of VOx/SiO2 catalysts
were synthesized at different vanadia loadings to study the effect
of vanadia surface coverage on conversion and selectivity in
propane ODH. The vanadia surface coverage[16] in these catalysts
ranged from 54 to 185% of a theoretical monolayer on the silica
support in the absence of Na+, and from 21 to 72% for the Na-
promoted silica.
Table 1. Vanadium content in silica-supported catalysts.
Coverage, % monolayer
Samples
V %*
V/nm2
Silica
Na promoted
S1 or Na-S1
S2 or Na-S2
S3 or Na-S3
S4 or Na-S4
4.2
6.9
1.9
3.1
4.1
6.6
54.0
88.0
21.2
34.4
45.6
72.2
9.1
117.1
185.7
14.7
[a] * The vanadium content (wt. %) was analysed by EDS.
The occurrence of homogeneous reactions was investigated
first by passing the above-mentioned propane/O2 and
propane/CO2 mixtures through an empty reactor and over the
original silica support at 550-750oC. For an empty reactor, trace
amounts of ethylene and methane were detected above 650oC for
the propane/O2 mixture, while no reaction was observed for the
propane/CO2 mixture at any temperature. Similar results were
observed for the ODH reactions over the original silica support,
i.e., trace levels of ethylene and methane above 550oC for the
propane/O2 mixture. Moreover, no reaction products were
observed up to 700oC for the propane/CO2 mixture, while above
700oC, CO and a trace amount of propane cracking products were
observed. However, we observed the occurrence of
homogeneous combustion reaction over silica support in the
presence of O2 above 600oC, as well as cracking reaction
products when the reactor temperature was further increased to
Table 3. Propane ODH reactions over VOx/SiO2 catalysts at 600oC.
C3H8
conv.
(%)
Coverage, % monolayer
Yield
(%)
V/nm2 Oxidant
CO
27.7
50.9
6.8
CO2
CH4
15.5
9.0
C2H4
9.5
C3H6
32.9
40.2
30.4
21.4
10.3
11.3
15.0
12.6
O2
CO2
O2
22.9
13.7
33.4
29.2
52.1
50.7
45.9
31.1
14.6
7.5
5.5
10.1
6.3
5.4
5.7
6.9
3.9
1.9
(S1)
3.1
-
0.0
18.8
25.0
9.7
18.9
0.0
CO2
O2
69.0
28.0
84.7
23.8
71.7
-
10.3
-
(S2)
4.1
47.0
4.0
4.8
CO2
O2
0.0
(S3)
6.6
14.9
39.0
7.0
CO2
-
13.2
2.5
(S4)
Reaction conditions; 600oC, GHSV = 2500 h-1, O2 : C3H8 = 1:2, CO2 : C3H8
= 2:1, He balance
Table 3 shows the initial propane conversion and product
selectivity results for propane ODH over VOx/SiO2 catalysts at
600oC. The propane conversion and formation of C1-C3 cracking
products increased dramatically at 600oC, while the propylene
yield was not significantly affected. The main difference in
reaction products observed for CO2 and O2 ODH reactions was
significant amount of CO produced in the former case and
methane for O2 ODH. C2 product (ethane and ethylene) was not
observed among the reaction products during CO2 ODH, except
over the S4 catalyst (6.6 V/nm2) at 600oC that contained ca. 180%
theoretical monolayers of VOx species. However, ethylene and
CO2 were already observed low temperature (450oC) during O2
ODH. The conversion of propane increased with VOx coverage
reaching a maximum over the S3 catalyst and then declining at
higher VOx coverage.
Table 2. Propane ODH reactions over VOx/SiO2 catalysts at 550oC.
C3H8
conv.
(%)
Coverage, % monolayer
Yield
(%)
V/nm2 Oxidant
CO
CO2
CH4
C2H4
C3H6
O2
CO2
O2
18.7
9.2
28.1
60.6
26.4
51.5
28.0
86.8
18.0
77.6
15.6
16.4
0.0
0.0
0.0
0.0
0.0
1.0
0.0
5.3
0.0
39.3
39.4
41.6
45.1
9.4
1.9
(S1)
3.1
-
0.0
-
7.4
3.6
27.3
25.0
37.2
29.4
38.5
22.3
32.0
3.4
CO2
O2
11.4
11.3
3.5
(S2)
4.1
13.6
-
49.0
3.4
CO2
O2
9.8
(S3)
6.6
8.5
-
40.0
0.0
28.0
2.9
CO2
22.4
10.8
(S4)
Reaction conditions; 550oC, GHSV = 2500 h-1, O2 : C3H8 = 1:2, CO2 : C3H8 =
2:1, He balance
This article is protected by copyright. All rights reserved.

10.1002/ejic.201701049
European Journal of Inorganic Chemistry
FULL PAPER
Turnover Frequencies (TOF) for propane consumption were
calculated by eq. (2):
reactions. Moreover, the S2 and S3 catalysts which showed a
higher conversion for CO2 ODH of propane tended to deactivate
rather quickly with time on stream. Also deactivation of the S2 and
S3 catalysts was accelerated at higher reaction temperature of
600oC, as propane conversion decreased by ca. 50% after 1 h
under catalytic reaction conditions. These results suggested that
more limited extent of surface reoxidation with increased VOx
coverage may be responsible for the decreased ODH activity of
VOx/SiO2 catalysts.
(2)
TOFs report the numbers of propane molecules converted at each
VOx species per second, where ṅ denotes the molar flowrate of
propane (mol/s), X the fractional conversion, Mv the molar mass
of vanadium (g/mol), mcat the catalyst weight (g) and WV the
vanadium content by mass (wt. %). Figure 1 shows the TOFs for
propane consumption as a function of surface VOx coverage at
500oC under different space velocity to maintain low propane
conversion (<15%). All supported vanadia catalysts displayed
very similar TOF values except for the S1 catalyst (at the lowest
V content). Also for the S2 and 3 catalyst (3.1 and 4.1 V/nm2 close
to theoretical monolayer coverage) it showed the maximum TOF
value was observed but the values only varying from 3.8 to 4.1 ×
10−3s−1 among three vanadia catalysts of S2, S3 and S4.
Figure 2.
Figure 2 shows the TGA curves for used VOx/SiO2 catalysts
after CO2 ODH of propane at 600oC for 72 h. All used catalysts
exhibited a weight losses during TGA in air associated with
carbon deposition, whereas used catalysts after O2 ODH showed
very little weight loss features in TGA. In a recent paper, Ascoop
et al.,[17] also reported significant carbon deposition even after a
short time on stream during CO2 ODH of propane over VOx/SiO2
catalysts. The extent of carbon deposition varied with varying
vanadium content, with two catalysts at near monolayer
coverage, i.e., S2 (3.1 V/nm2) and S3 (4.1 V/nm2) showing much
higher weight loss, ca. 45 and 20 wt. %, respectively, while the S1
(1.9 V/nm2, ca. 0.5 monolayer) showed almost no weight loss
during TGA measurement. These observed differences in carbon
deposition behavior may be related to catalytic activity, since the
Figure 1. Propane consumption TOFs as a function of surface
VOx coverage at 500oC determined at propane conversion of
~15%.
All supported vanadia catalysts were further investigated at 550
and 600oC, GHSV=2500 h-1, and feed composition of CO2:C3H8 =
2:1 (Figures S1-S4) as a function of time on stream. The S1
catalyst (1.9 V/nm2) initially displayed low catalytic activity at
550oC, which was subsequently lost after 2 h under catalytic
reaction conditions, while stable catalytic behavior was observed
at 600oC for more than 3.5 h. The S2 catalyst (3.1 V/nm2) initially
showed high catalytic activity, which continuously declined with
time at 550oC. The formation of C1 and C2 products dramatically
increased at 600oC after 3 h on stream, while propane conversion
and propylene selectivity decreased. On the other hand, the
catalytic activity rapidly decreased over the S3 catalyst (4.1
V/nm2) with time on stream. These results indicated that the
VOx/SiO2 catalysts rapidly deactivate with increasing VOx content
above the theoretical monolayer coverage. The catalytic activity
did not increase with reaction temperature, while the selectivity to
propylene decreased at the expense of propane cracking
Table 4. CO2 ODH of propane reaction results over sodium-promoted silica
catalyst with various vanadium content.
Temp.
(oC)
Selectivity (%)
C3H8 conv.
(%)
Samples
CO
100
CH4
C2H4
C3H6
550
600
550
600
550
600
550
600
1.3
3.4
0.9
2.2
0.9
4.3
1.8
5.5
0.0
8.2
0.0
8.3
0.0
8.3
0.0
8.0
0.0
85.5
0.00
86.2
0.00
85. 5
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
1.9
6.3
100
5.5
(Na-S1)
3.1
(Na-S2)
4.1
100
6.3
(Na-S3)
6.6
100
4.0
88.0
(Na-S4)
Reaction conditions; GHSV = 2500 h-1, O2 : C3H8 = 1:2, CO2 : C3H8 = 2:1, He
balance
This article is protected by copyright. All rights reserved.

10.1002/ejic.201701049
European Journal of Inorganic Chemistry
FULL PAPER
catalysts showing higher reactivity in CO2 ODH of propane
displayed greater carbon deposition. These results suggested
that the acidic silica support may also play a role in initiating
carbon deposition by adsorbing hydrocarbon intermediates during
the ODH reaction in the presence of CO2. However, the exposure
of the silica surface to hydrocarbon intermediates is minimized
above the monolayer coverage of the VOx species, e.g., for S4
(6.6 V/nm2), explaining diminished carbon deposition observed for
this catalyst (ca. 2% weight loss).
Na-promoted silica support were employed to improve surface
dispersion of two-dimensional VOx species. Table 4 shows CO2
ODH of propane reaction results over Na-promoted VOx/SiO2
catalysts prepared at the same VOx surface coverage as the Na-
free supported VOx catalysts (Table 1). Unlike Na-free catalysts,
Na-promoted VOx/SiO2 catalysts showed no selectivity to
propylene instead displaying increasing tendency to form
cracking products, ethane and methane, with increasing
vanadium content. The Na-promoted catalysts showed very low
catalytic activity at 550 and 600oC. A high reaction temperature of
600oC, a selectivity more than 85% to ethylene was observed
over all the catalysts while at the reaction temperature of 550oC
only CO was observed with very low propane conversion of less
than 2%. The selectivity to C2 species (ethane and ethylene) was
particularly high and remained relatively constant over all
catalysts investigated, suggesting that these species may
represent reaction intermediates during propane/propylene
degradation to methane.
Raman spectra of the Na-free and Na-promoted silica support
and VOx/SiO2 catalysts were collected in order to establish the
nature of the surface species in these two series of supported
vanadia catalysts. Figure 3 shows the Raman spectra of the Na-
free and Na-promoted silica supports. Raman shift peaks at 1058
and 1023 cm-1 in the spectrum of the Na-bearing silica support
(upper spectrum in Fig. 3) are assigned to sodium silicate.[16]
Figure 4 shows the Raman spectra of Na-free (a) and Na-
promoted S2 (b) catalysts as well as corresponding silica
supports. The Raman spectrum of the Na-free S2 catalyst (upper
spectrum in Fig. 4 (a)) shows the presence of peaks at 993 and
1033 cm-1, which correspond to the bulk V2O5 crystals and V=O
Figure 4.
bond stretching vibration of the monomeric VOx species,
respectively.[17] The addition of Na resulted in the appearance of
Raman peaks at 506, 637, 914 and 950 cm-1 associated with
presence of Na metavanadate instead of 2D or 3D vanadia.[18] Na
metavanadate was previously reported to be inactive in ODH
reactions.[16-19] which may explain the lack of catalytic activity of
our Na-promoted VOx/SiO2 catalysts. According to lrusta et al.,[19]
vanadate species react with Na+ cations present on the surface of
the Na-promoted silica forming surface NaVO3-like compounds
that were inactive for methane oxidation to formaldehyde. Similar
observations were reported by Adamski et al.,[20] that the
conversion of propane during its ODH over alkali-doped
supported 3V/Zr catalysts decreased in the following order: un-
doped > Li+ > Na+ > K+, indicating that alkali dopants were
detrimental for the ODH activity of supported vanadia catalysts.
Furthermore, we confirmed the carbon deposition in used
catalysts by comparing the Raman spectra of fresh and used
catalysts. Figure 5 compares the Raman spectra of the fresh and
used S4 catalysts after propane ODH in the presence of CO2
(GHSV = 2500 h-1 and CO2 : C3H8 = 2 : 1 in He balance at 600oC).
The Raman peaks at 993 and 1033 cm-1 decreased in intensity
due to reduction of VOx species during the reaction, which
represent the stretching mode of monomeric V=O species, while
Figure 3.
This article is protected by copyright. All rights reserved.

10.1002/ejic.201701049
European Journal of Inorganic Chemistry
FULL PAPER
for V, Si, O, C and Na on Na-free and Na-promoted S4
catalysts (6.6 V/nm2) corresponding to the highest V(wt.%)
loading investigated in this study. In the absence of Na+, a
large vanadium oxide particle is visible in the center top
region of the images shown in Figure S5a confirming that
VOx species form a bulk phase above theoretical monolayer
coverage (185.7% in Table 1). In the presence of Na+, the
dispersion of VOx species is improved significantly
manifested in the absence of large vanadium oxide particles
(Figure S5b). Increased theoretical monolayer coverage of
VOx in Na-promoted catalysts reduces the V surface
coverage in the Na-promoted S4 catalyst to ca. 72.2% of the
theoretical monolayer (Table 1). However, while Na
promotion improves the dispersion of VOx species on silica,
it results in the formation of surface Na metavanadate
species that lack catalytic activity in both propane ODH
reactions. The presence of Na+ is believed to adversely affect
the redox behavior of surface VOx species, which prevents
the regeneration of the V5+ oxidation state important for
propane activation.
Figure 5. Raman spectra of fresh and used S4 catalysts after
propane ODH in the presence of CO2. (GHSV = 2500 h-1 and
CO2 : C3H8 = 2 : 1, He balance at 600oC)
new peaks were observed at 1350 and 1598 cm-1 that correspond
to the D and G bands of carbon. The G band represents graphitic
carbon and the D band represents disorder within the sp2 bonded
carbon as in Figure 5.[21]
Conclusions
Figure 6 shows the Raman spectra of the Na-free and Na-
promoted S4 catalysts containing the highest VOx loading (6.6
V/nm2). The Raman V=O stretch is clearly observed at 1028 cm-1
in the absence of Na+ indicating the presence of molecularly
dispersed VOx species. However, this band was broadened
as increasing vanadium contents might be due to a higher
content microcrystalline V2O5 than this feature for the
catalysts at lower VOx coverage. The Raman features of
molecularly dispersed VOx species were absent in the
Raman spectrum of the Na-promoted S4 catalyst. Surface
Na metavanadate was also expected for this catalyst,
although its Raman spectrum lacked strong features
associated with Na metavanadate in agreement with
previous observations.[16] Improved vanadium dispersion in
the presence of Na+ was further confirmed by EDS elemental
mapping. Figure S5 reports the results of elemental mapping
Propane ODH by O2 and CO2 was performed over Na-free and
Na-promoted supported VOx/SiO2 catalysts. The propylene yield
was very similar in both CO2 and O2 ODH reactions with a
maximum value of ca. 11 mol.% observed for the S2 catalyst.
However, the propane conversion increased dramatically with
vanadium content, e.g., 18 % for O2 and 9.2% for CO2 ODH at 1.9
V/nm2 for the S1 catalyst, while it increased to > 25% at higher
VOx loadings for both O2 and CO2 ODH. We observed a trend
through Raman study that as increase vanadium content on silica,
2D vanadia structure was formed at first and then as vanadium
content increases, 3D vanadia crystalline structure is favorably
formed. So that maximum 2D structure of vanadia which shows
high activity for CO2 ODH of propane was obtained in a monolayer
coverage range and we confirmed that by calculating TOF rate of
propane conversion. Considerable carbon deposition was
observed in used catalysts after propane ODH by CO2, the extent
of which correlated with their catalytic activity.
Promotion with Na cations significantly improved the dispersion
of surface VOx species and resulted in the formation of surface
Na metavanadate or another reduced V3+/V4+ phase, which
displayed no catalytic activity in propane ODH in the presence of
CO2.
Experimental Section
A series of VOx/SiO2 catalysts were synthesized at different vanadia
loadings to study the effect of vanadia surface coverage on conversion and
selectivity in propane ODH. Ammonium metavanadate (Sigma-Aldrich,
99% ACS grade) was deposited on a commercial silica support (Saint-
Gobain; SS 61138, BET surface area: 261 m2g−1) by incipient wetness
impregnation method at 4 different vanadia loadings. After impregnation,
the as-synthesized catalysts were dried in a vacuum oven at 60°C for 4 h
and then calcined in air at 550°C for 3 h. Also 4 different Na-promoted
Figure 6.
This article is protected by copyright. All rights reserved.

10.1002/ejic.201701049
European Journal of Inorganic Chemistry
FULL PAPER
VOx/SiO2 catalysts were also synthesized at the same vanadia loadings
(Table 1). First, Na-promoted silica supports were prepared using sodium
nitrate (Fisher Scientific, Certified ACS Grade, >99% purity) by incipient
wetness impregnation of the same silica support by a NaNO3 solution. Na
loadings were adjusted for different supported vanadia catalysts to yield
Na-promoted VOx/SiO2 catalysts corresponding to the atomic ratio of Na/V
= 0~0.3. Prior to VOx deposition, as-synthesized Na-promoted silica were
calcined in air by ramping the temperature at 1.5°C min−1 to 700°C and
holding for 4 h. These catalysts were then synthesized following the same
procedures described above that were used for the Na-free catalysts. The
surface loadings (V/nm2) and coverages (% of the theoretical monolayer)
reported in Table 1 were calculated on the basis of EDS V and Si
compositions in these catalysts. The V content (wt. %) was converted to
surface loadings (V/nm2) according to the following formula:
Acknowledgements
This research was supported by the National Science
Foundation under Grant no.OCDO R-12-14.
Keywords: ODH by CO2 • Selective Oxidation of Propane •
Sodium modification of silica • Extend vanadia coverage • Na-
metavanadate
[1]
[2]
V. Havran, M. P. Dudukovic, Ind. Eng. Chem. Res. 2011, 50, 7089–7100.
The US Energy Information Administration, 2016:
http://www.eia.gov/tools/faqs/faq.cfm?id=77&t=11.
[3]
The US Energy Information Administration’s Annual Energy Outlook
2015 (AEO2015): http://www.eia.gov/forecasts/aeo/.
J. N. Armor, J. Energy Chem. 2013, 22, 21-26.
ꢈꢉꢊꢋꢌ
ꢋꢊꢍ
ꢀ ꢄꢅ.% ∙ꢆ (
ꢇ
)
ꢀ
[4]
[5]
[6]
=
(3)
ꢃ
ꢃ
ꢚꢋ
ꢘꢙ
ꢃ
ꢐ
ꢐ
ꢐ
ꢋ
ꢁꢂ
R. K. Grasselli, Catalysis Today 2005, 99, 23-31.
M. Zboray, A. T. Bell, E. Iglesia, J. Phys. Chem. C 2009, 113, 12380-
12386.
ꢎꢏꢏ
∙ꢑꢒ (
)∙ꢔ
(
ꢕꢖꢗ
)∙ꢎꢏ
(
)
ꢃ
ꢓ
ꢋꢊꢍ
ꢐ
ꢋ
These loadings were referenced to theoretical monolayer coverages
reported by Grant e al.[16] to yield vanadium surface coverages reported in
Table 1.
[7]
E. Drent, W. P. Mul, A. A. Smaardijk, Encyclopedia of Polymer Science
and Technology. John Wiley
&
Sons, Inc. 2001, DOI:
10.1002/0471440264.pst273.
Oxidative dehydrogenation of propane by O2 and CO2 was performed
at atmospheric pressure in a continuous flow fixed bed tubular reactor. 0.5
g of different VOx/SiO2 catalysts placed between quartz wool plugs inside
the reactor (a stainless steel tube of 1/4 inch O.D.) were pre-treated in a
flowing O2/He mixture (10 ml/min each) for 1 h at 450oC prior to the
catalytic tests. The feed gas mixture consisted of helium (UHP grade,
Wright Brothers, Inc.), carbon dioxide (Tech. grade, Wright Brothers, Inc.)
or oxygen (UHP grade, Wright Brothers, Inc.) and propane (CP grade,
Matheson gas) at a total flow rate of 16 ml/min to establish the GHSV of
2500 h-1. The O2 concentration employed in these studies was half of the
CO2 concentration used in ODH reaction to minimize homogeneous
oxidation of propane (propane/oxidant ratio of 2:1 for CO2 and 1:1 for O2,
balance of He), while reaction temperature was varied from 350 to 550oC.
The effluent gas was directly passed through the heated sampling valves
to online GC-MS system (Shimadzu, GCMS QP-5000) equipped with a
capillary column (Supelco, Carboxen 1006 PLOT, Fused silica) of 30 m in
length and 0.32 micron film thickness and using ultrapure helium (Wright
Brother Inc., Lot 9047-1) as the carrier gas. The products were identified
using NIST Mass Spectrum Library 2008 (Shimadzu, No. 225-13290-91).
The catalysts were also characterized by TGA (SDT-Q600 unit from TA
Instruments, New Castle, DE, USA), and Raman spectroscopy (Horiba
T64000, Horiba Scientific, Edison, NJ, USA) equipped with a 514 nm laser
line, before and after the ODH of propane reaction. STEM imaging and
EDS elemental mapping of the two S4 catalysts corresponding to the
highest V surface loading investigated in this study was conducted using
the FEI Probe Corrected Titan3™ 80-300 S/TEM.
[8]
[9]
V. Leeuwen, W. N. M. Piet, C. Claver, Rhodium Catalyzed
Hydroformylation. Springer, 2002, DOI: 10.1007/0-306-46947-2.
N. Mimura, M. Okamoto, H. Yamashita, S. T. Oyama, K. Murata, J. Phys.
Chem. B 2006, 110, 21764-21770.
[10] M. A. Smith, A. Zoelle, Y. Yang, R. M. Rioux, N. G. Hamilton, K.
Amakawa, P. K. Nielsen, A. Trunschke, Journal of Catalysis 2014, 312,
170-178.
[11] Z. W. Liu, C. Wang, W. B. Fan, Z. T. Liu, Q. Q. Hao, X. Long, J. Lu J.
G. Wang, Z. F. Qin, D. S. Su, ChemSusChem 2011, 4 (3), 341-345.
[12] G. Raju, B. M. Reddy, B. Abhishek, Y. H. Mo, S. E. Park, Applied
Catalysis A: General 2012, 423-424, 168-175.
[13] W. Z. Li, F. Gao, Y. Li, E. D. Walter, J. Liu, C. H. F. Peden, Y. Wang, J.
Phys. Chem. C 2015, 119 (27), 15094-15102.
[14] V. V. Guliants, Catalysis Today 1999, 51 (2), 255-268.
[15] C. A. Carrero, R. Schloegl, I. E. Wachs, R. Schomaecker, ACS Catal.
2014, 4 (10), 3357-3380.
[16] J. T. Grant, C. A. Carrero, A. M. Love, R. Verel, I. Hermans, ACS Catal.
2015, 5 (10), 5787-5793.
[17] I. Ascoop, V. V. Galvita, K. Alexopoulos, M. Reyniers, P. V. D. Voort, V.
Bliznuk, G. B. Marin, Journal of Catalysis 2016, 335, 1-10.
[18] S. Seetharaman, H. L. Bhat, P. S. Narayanan, Journal of Raman
Spectroscopy 1983, 14 (6), 401-405.
[19] S. Irusta, A. J. Marchi, E. A. Lombardo, E. E. Miré, Catalysis Letters 1996,
40 (1), 9-16.
[20] A. Adamski, Z. Sojka, K. Dyrek, Langmuir 1999, 15 (18), 5733-5741.
[21] A. C. Ferrari, J. Robertson, Physical Review B 2000, 61 (20), 14095-
14107.
This article is protected by copyright. All rights reserved.

10.1002/ejic.201701049
European Journal of Inorganic Chemistry
FULL PAPER
Entry for the Table of Contents (Please choose one layout)
FULL PAPER
Jungshik Kang,[a] Andrew D. Czaja[b]
and Vadim V. Guliants*
Page No. – Page No.
Carbon Dioxide as Feedstock in
Selective Oxidation of Propane
This article is protected by copyright. All rights reserved.