Solid State Ionics 172 (2004) 135 – 138
www.elsevier.com/locate/ssi
Performance of vanadium–molybdenum mixed oxide catalysts in selective
oxidation of hydrogen sulfide containing excess water and ammonia
Bong-Guk Kim a, Wol-Don Ju a, Il Kim a, Hee-Chul Woo b, Dae-Won Park a,*
a
Division of Chemical Engineering, Department of Chemical Engineering, College of Engineering, Pusan National University,
Janjun-Dong, Kumjung-ku, Busan 609-735, South Korea
b
Department of Chemical Engineering, Pukyung National University, Busan 608-739, South Korea
Received 9 November 2003; received in revised form 30 January 2004; accepted 5 February 2004
Abstract
The selective oxidation of hydrogen sulfide containing excess water and ammonia was studied over V2O5/TiO2 and Mo – V – O/TiO2
catalysts. Ammonia reacted either with H2S or SO2, produced from the oxidation of H2S. Water vapor promoted the reaction of ammonia and
SO2. Mo – V – O/TiO2 catalysts showed very high H2S conversion without any considerable emission of SO2. Temperature-programmed
studies [temperature-programmed reduction (TPR) and temperature-programmed oxidation (TPO)], X-ray diffraction (XRD), and X-ray
photoelectron spectroscopy (XPS) analyses revealed that the high catalytic performance of Mo – V – O/TiO2 catalysts originated from the high
redox capacity of vanadium molybdate phase.
D 2004 Elsevier B.V. All rights reserved.
PACS: 81.07.Bc; 81.16.Hc; 81.20.Fw; 82.33.Pt
Keywords: Hydrogen sulfide; Selective oxidation; Mo – V – O/TiO2; Ammonia
1. Introduction
In recent years, a great deal of effort has been put into
research on new processes to treat hydrogen sulfide in the
tail gas from a Claus plant or from other emission sources.
Commercially developed procedures include titanium-based
catalysts in MODOP process [1] and iron-based catalysts in
Super Claus process [2 –4]. In our previous works [5,6], we
reported very high activities of TiO2 and V2O5 catalysts in
the selective oxidation of hydrogen sulfide to elemental
sulfur. Some binary metal oxides, such as Bi – V –O [7] or
Fe –Cr – O [8], have also been reported as catalysts for the
gas phase conversion of H2S to sulfur. Li et al. [9,10]
reported V –Mo, V – Bi, V –Mg, Fe – Sn, and Fe– Sb mixed
oxide catalyst systems for this reaction.
Although vanadium – molybdenum catalyst has recently
been studied for the oxidation of hydrogen sulfide to
elemental sulfur [11], no information is available about
the use of this catalyst system for the selective oxidation
of H2S containing NH3 and excess water. A mixed gas of
* Corresponding author. Tel.: +82-51-510-2399; fax: +82-51-512-8563.
E-mail address: [email protected] (D.-W. Park).
0167-2738/$ - see front matter D 2004 Elsevier B.V. All rights reserved.
doi:10.1016/j.ssi.2004.02.043
H2S, NH3, and water vapor is released from steel smelting
processes where the H2S from coke ovens is generally
scrubbed and concentrated using aqueous ammonia solution. However, the separation of H2S from the solution is
not perfect and the remaining aqueous ammonia stream
contains about 2% H2S, which in turn causes the SOx
emission problem during incineration.
In this study, we report a new vapor phase catalytic process
for the selective conversion of H2S in the stream containing
both ammonia and water. We examined the performance of
Mo –V – O supported on TiO2 for this reaction system.
2. Experimental
V2O5/TiO2 and Mo –V –O/TiO2 catalysts were prepared
by an evaporation method. The precursors of vanadium and
molybdenum were ammonium metavanadate and ammonium molybdate, respectively. The support was TiO2 (JRCTiO2, anatase, 17.1 m2/g), supplied by the Japan Reference
Catalyst Committee. Oxalic acid, 5 wt.%, (Tedia) was used
to easily dissolve the vanadium precursor. After evaporation
at 80 jC, samples were dried at 110 jC overnight and
calcined at 500 jC for 5 h.
136
B.-G. Kim et al. / Solid State Ionics 172 (2004) 135–138
The surface area of the catalyst was measured by N2
adsorption method using the BET technique (ASAP 2000;
Micromeritics). The phase analysis was performed by using
an X-ray diffraction (XRD) crystallography with Cu-Ka
radiation (DMAX 2400; Rigaku). X-ray photoelectron spectroscopy (XPS) analyses were carried out by using an X-ray
photoelectron spectrometer (ESCALAB 220; VG) with
monochromatic Al-Ka radiation. In order to investigate
the phase cooperation mechanism, temperature-programmed
reduction (TPR) and temperature-programmed oxidation
(TPO) were also carried out. Reaction tests were carried
out in a continuous flow fixed-bed reactor. The content of
effluent gas was analyzed by a gas chromatograph (HP
5890) equipped with a thermal conductivity detector and a
1.8-m Porapak T column (80 –100 mesh) at 100 jC. The
conversion of H2S and the selectivity to SO2, S, and ATS
are defined as follows:
Conversion of H2 S ðX Þ ¼
½H2 Sinlet ½H2 Soutlet
100ð%Þ
½H2 Sinlet
Selectivity ðSÞ to a special product ðSO2 ; S; ATSÞ
¼
½Productoutlet
100ð%Þ
½H2 Sinlet ½H2 Soutlet
For the calculation of ATS selectivity, moles of ATS were
multiplied by a factor of 2 because 1 mol of ATS can be
obtained from 2 mol of H2S.
Table 1
Conversion of H2S and the selectivity of products for V2O5/TiO2 and Mo –
V – O/TiO2 catalysts at 260 jC
Catalyst
X, H2S (%)
S, SO (%)
S, S (%)
S, ATS (%)
V2O5/TiO2
Mo – V – O/TiO2
(Mo/V = 0.3)
Mo – V – O/TiO2
(Mo/V = 1.0)
80.0
83.5
0.0
0.0
63.5
78.4
36.5
21.6
86.0
0.0
82.3
17.7
selective oxidation of H2S containing excess water and
ammonia. Fig. 1 shows the H2S conversion (X) and SO2
selectivity (S) at temperature ranges of 240 –340 jC with a
feed composition of H2S/O2/NH3/H2O/He = 5/2.5/10/60/
22.5 in volume percent and GHSV of 30,000 h-1. These
data were obtained after 6 h of the reaction. Mo – V – O/TiO2
catalysts show higher H2S conversion than V2O5/TiO2, and
Mo – V –O/TiO2 (Mo/V = 1) shows the highest conversion
over all the temperature ranges. The addition of molybdenum oxide into V2O5/TiO2 improves the performance of the
catalyst. The selectivity to SO2 is very low under 300 jC.
As reported in other works [5,12], higher temperature
yielded more production of SO2 and less production of
elemental sulfur than low temperature did in the H2S
oxidation without NH3. With the presence of NH3, the
produced SO2 can react to form (NH4)2SO3, then finally
produce ATS by the reactions of Eqs. (1) and (2):
SO2 þ 2NH3 þ H2 O ! ðNH4 Þ2 SO3
ð1Þ
3. Results and discussion
ðNH4 Þ2 SO3 þ S ! ðNH4 Þ2 S2 O3
ð2Þ
Mo –V – O/TiO2 catalysts with different Mo/V ratios (0.3
and 1.0) and 5 wt.% V2O5/TiO2 catalyst were tested for the
Table 1 shows the H2S conversion and selectivity to the
products with Mo – V –O/TiO2 and V2O5/TiO2 catalysts at
260 jC. The addition of molybdenum oxide into the V2O5/
TiO2 increased the H2S conversion and sulfur selectivity.
However, the ATS formation decreased as the added amount
of molybdenum oxide increased. SO2 was not produced for
all the catalysts.
XRD analyses were carried out for the Mo – V – O/TiO2
catalysts; however, only the characteristic peaks of TiO2
support could be observed. V2O5, MoO3, and vanadium
molybdate (Mo6V9O40) could not be detected by the XRD
probably because they were highly dispersed on the TiO2
surface. Mo6V9O40 was observed for unsupported Mo – V –O
catalysts, and it was converted to an amorphous compound
during the selective oxidation of H2S to elemental sulfur [9].
In order to investigate the performance of Mo – V – O/
TiO2 catalysts, TPR/TPO experiments for V2O5/TiO2 and
Mo – V –O/TiO2 catalysts were carried out, and the TPO
results are shown in Fig. 2. It is evident that the reoxidation
of the Mo –V – O/TiO2 catalysts after TPR starts at a lower
temperature than does the reoxidation of the V2O5/TiO2
catalyst. This implies that the addition of molybdenum
oxide increases the reoxidation property of V2O5/TiO2 due
to a high oxygen dissociation capacity of MoO3. Table 2
Fig. 1. Conversion of H2S and selectivity to SO2 for V2O5/TiO2 and Mo –
V – O/TiO2 catalysts at different reaction temperatures (5 vol.% H2S, 2.5
vol.% O2, 5 vol.% NH3, 60 vol.% H2O, 27.5 vol.% He, GHSV = 30,000
h1).
B.-G. Kim et al. / Solid State Ionics 172 (2004) 135–138
137
Table 3
Bulk and surface composition of Mo – V – O/TiO2 catalysts before and after
the reaction at 260 jC for 6 h
Fig. 2. TPO profiles for V2O5/TiO2 and Mo – V – O/TiO2 catalysts.
summarizes the amount of H2 and O2 consumption during
TPR and TPO, respectively. Since the amounts of H2 and O2
consumption were highest for Mo –V – O/TiO2 (Mo/V = 1),
one can see that this catalyst has the highest redox capacity
among the tested catalysts. This may be the reason why the
Mo –V – O/TiO2 (Mo/V = 1) catalyst shows the highest catalytic activity in the selective oxidation of H2S. The
reducibility of the active site VOx was reported to be an
important factor for determining product selectivity in
alkane oxidation [13 –15]. For the selective oxidation of
H2S, the VOx are the major active centers that abstract
hydrogen atoms from the chemisorbed H2S on the catalyst
surface [12,16].
XPS analyses were performed for V2O5/TiO2 and Mo –
V – O/TiO2 catalysts. The surface V/Ti ratio of the V2O5/
TiO2 catalyst before and after the reaction was 0.090 and
0.086, respectively. These were much higher than the bulk
value (0.0462) calculated from the used amount of precursors and TiO2 in the preparation of the catalyst. It means that
vanadium oxide is highly dispersed on the surface of TiO2.
Table 3 shows the surface concentrations of Mo –V –O/TiO2
catalysts. As we can see in the Mo/Ti values of the Mo – V –
O/TiO2 catalysts, surface concentration of molybdenum was
also much higher than the corresponding calculated bulk
Mo/V
State
V/Ti
Mo/Ti
V/(Ti +
Mo + V)
Mo /(Ti +
V + Mo)
0.3
0.3
0.3
1.0
1.0
1.0
Bulk
BR
AR
Bulk
BR
AR
0.0474
0.079
0.123
0.0504
0.066
0.055
0.0142
0.015
0.127
0.0504
0.157
0.034
0.0447
0.0663
0.0987
0.0458
0.0539
0.0406
0.0133
0.0889
0.1019
0.0458
0.1280
0.2239
values. For Mo – V –O/TiO2 (Mo/V = 1) catalyst, the surface
was enriched with molybdenum because Mo/(V + Mo) for
the fresh catalyst was 0.704. The surface enrichment of
molybdenum was also reported by Matralis et al. [17] for the
V2O5 –MoO3/TiO2 catalyst. Total surface coverage of the
titania by both active elements (Mo + V)/(Ti + Mo + V) was
0.182 for Mo –V – O/TiO2 (Mo/V = 1), much higher than the
calculated value of 0.0916.
The oxidation state of vanadium in V2O5/TiO2 and Mo –
V – O/TiO2 catalysts is studied by XPS. Fig. 3 shows XPS
spectra of V 2p3/2 for these catalysts before and after 6 h of
reaction at 260 jC with the standard reactant mixture and
GHSV of 30,000 h 1. The standard XPS peaks of V 2p3/2
for V5 + and V4 + are located at 517.2 and 515.9 eV,
respectively. After the reaction, the XPS spectra were
broadened and shifted to lower binding energy. It means
that some of the fresh VOx having higher oxidation states
are reduced after the reaction. The increase in full width at
half maximum (FWHM) value after the reaction was 30.0%,
18.3%, and 7.8% for V2O5/TiO2, Mo – V – O/TiO2 (Mo/
V = 0.3), and Mo – V – O/TiO2 (Mo/V = 1), respectively.
Therefore, Mo – V – O/TiO2 (Mo/V = 1) showed the least
reduction of vanadium oxide phase during the reaction.
Table 2
The consumption amount of hydrogen and oxygen in TPR and TPO
experiments
Catalyst
TPR
TPO
H2 (Amol/gcat)
O2 (Amol/gcat)
V2O5/TiO2
Mo – V – O/TiO2
(Mo/V = 0.3)
Mo – V – O/TiO2
(Mo/V = 1.0)
22.3
26.0
41.3
65.5
34.1
77.6
Fig. 3. XPS analyses of V 2P3/2 for V2O5/TiO2 and Mo – V – O/TiO2
catalysts before and after the reaction at 260 jC for 6 h.
138
B.-G. Kim et al. / Solid State Ionics 172 (2004) 135–138
4. Conclusions
The selective oxidation of hydrogen sulfide in the presence of excess water and ammonia was investigated in this
study. V2O5/TiO2 and Mo – V –O/TiO2 catalysts with different Mo/V ratios are tested in this reaction. Mo – V –O/TiO2
showed more improved catalytic performance than V2O5/
TiO2 did. Mo –V – O/TiO2 (Mo/V = 1) showed the highest
oxygen consumption in TPO and the least degree of
vanadium oxide reduction in XPS analysis. Therefore, the
high catalytic activity of Mo –V – O/TiO2 can be concluded
from the increased reoxidation capacity. However, the
formation of vanadium molybdate phase cannot be excluded
for the reason of its high activity.
Acknowledgements
The authors wish to acknowledge the financial support of
the Korea Science and Engineering Foundation (R 05-2003000-10050-0), and the Brain Korea 21 and Brain Busan 21
programs.
References
[1]
[2]
[3]
[4]
[5]
[6]
[7]
[8]
[9]
[10]
[11]
[12]
[13]
[14]
[15]
[16]
[17]
R. Kettner, T. Lubcke, N. Liermann, Europe Patent 0078690 (1983).
P.J. Brink, J.W. Geus, Europe Patent 0409353 (1991).
P.H. Berban, J.W. Geus, US Patent 5037629 (1991).
P.J. van den Brink, J.W. Geus, US Patent 5352422 (1994).
S.W. Chun, J.Y. Jang, D.W. Park, H.C. Woo, J.S. Chung, Appl. Catal.,
B 16 (1998) 235.
D.W. Park, S.W. Chun, J.Y. Jang, H.S. Kim, H.C. Woo, J.S. Chung,
Catal. Today 44 (1998) 73.
R.H. Haas, J.W. Ward, US Patent 4528277 (1985).
P.J. Brink, J.W. Geus, US Patent 5286697 (1994).
K.T. Li, M.Y. Huang, W.D. Cheng, Ind. Eng. Chem. Res. 35 (1996)
621.
K.T. Li, C.S. Yen, N.S. Shyu, Appl. Catal., A 156 (1997) 117.
K.T. Li, N.S. Shyu, Ind. Eng. Chem. Res. 36 (1997) 1480.
M.Y. Shin, D.W. Park, J.S. Chung, Appl. Catal., B 30 (2001) 409.
L.T. Weng, B. Delmon, Appl. Catal., A 81 (1992) 141.
N. Arora, G. Deo, I.E. Wachs, A.M. Hirt, J. Catal. 159 (1996) 1.
P.M. Michalakos, M.C. Kung, I. Jahan, H.H. Kung, J. Catal. 140
(1993) 226.
Y.G. Cho, B.H. Hwang, D.W. Park, H.C. Woo, J.S. Chung, Korean
J. Chem. Eng. 19 (2002) 616.
H.K. Matralis, Ch. Papadopoulou, Ch. Kordulis, A.A. Elguezabal,
V.C. Corberan, Appl. Catal., A Gen. 126 (1995) 365.