Applied Catalysis B: Environmental 33 (2001) 137–148
Microwave assisted catalytic reduction of sulfur dioxide with
methane over MoS
2
catalysts
Xunli Zhang
1
, David O. Hayward, Colleen Lee, D. Michael P. Mingos
∗
Department of Chemistry, Imperial College of Science, Technology and Medicine, South Kensington, London SW7 2AY, UK
Received 25 November 2000; received in revised form 6 January 2001; accepted 18 March 2001
Abstract
The catalytic reduction of sulfur dioxide with methane to form carbon dioxide and sulfur has been studied over MoS
2
/Al
2
O
3
catalysts. The reaction has been found to occur with microwave (2.45 GHz) heating at recorded temperatures as much as 200
◦
C
lower than those required when conventional heating was used. An activation energy of 117 kJ mol
−1
has been calculated for
the conventionally heated reaction, but an Arrhenius analysis of the data obtained with microwave heating was not possible,
probably because of temperature variations in the catalyst bed. The existence of hot spots in the catalysts heated by microwave
radiation has been verified by the detection of
␣
-alumina at a recorded temperature some 200
◦
C lower than the temperature
at which the
␥-to␣
-alumina phase transition is normally observed. Among four catalysts prepared in different ways, a
mechanically mixed catalyst showed the highest conversion of SO
2
and CH
4
for microwave heating at a given temperature.
Supported catalysts, sulfided either by conventional heating or under microwave conditions, showed little difference in the
extent of SO
2
and CH
4
conversions. The highest conversions to carbon dioxide and sulfur, combined with low production
of undesirable side products, was obtained when the molar ratio of SO
2
to CH
4
was equal to two, the stoichiometric ratio.
© 2001 Elsevier Science B.V. All rights reserved.
Keywords: Sulfur dioxide; Methane; Catalytic reduction; Microwave irradiation; MoS
2
catalyst
1. Introduction
Sulfur dioxide is generally accepted to be the most
important precursor to acid rain. It is produced in vast
quantities by large fuel-consuming installations such
as non-ferrous smelters and the need to find efficient
and economical methods of reducing the amount of
sulfur dioxide emissions has become ever more urgent,
as evidence demonstrating the detrimental effects of
∗
Corresponding author. Present address: St. Edmund Hall, Uni-
versity of Oxford, Oxford OX1 4AR, UK. Fax: +44-1865-279030.
E-mail address: michael.mingos@seh.ox.ac.uk (D.M.P. Mingos).
1
Present address: Department of Chemistry, University of Hull,
Hull HU6 7RX, UK.
both sulfur dioxide and acid rain on the environment
has accumulated [1,2].
A number of processes for the removal of sulfur
dioxide have been developed and evaluated [3–6], and
they can be classified into three groups.
1. Conversion of sulfur dioxide into solid salts such
as calcium sulfate and calcium sulfide by the fol-
lowing reactions.
CaO(s) + SO
2
(g) +
1
2
O
2
= CaSO
4
(s) (1)
4CaO(s) + 4SO
2
(g) = 3CaSO
4
(s) + CaS(s) (2)
2. Oxidation of sulfur dioxide to sulfur trioxide, which
may then be converted into sulfuric acid.
0926-3373/01/$ – see front matter © 2001 Elsevier Science B.V. All rights reserved.
PII: S0926-3373(01)00171-0
138 X. Zhang et al. / Applied Catalysis B: Environmental 33 (2001) 137–148
SO
2
(g) +
1
2
O
2
(g) = SO
3
(g) (3)
SO
3
(g) + H
2
O(l) = H
2
SO
4
(l) (4)
3. Reduction of sulfur dioxide to elemental sulfur with
a number of reducing agents such as hydrogen, hy-
drogen sulfide, carbon, carbon monoxide and hy-
drocarbons.
SO
2
(g) + reductants → [S](s) (5)
In method (1), large amounts of essentially worth-
less materials are produced, which require disposal
and subsequently result in other environmental prob-
lems. Sulfuric acid produced in method (2) is a prod-
uct which is difficult to store and transport. As an end
product, elemental sulfur has several advantages over
sulfuric acid, because it can be handled, stored and
transported safely, with no danger to the environment.
Therefore, method (3), which involves reducing sulfur
dioxide to elemental sulfur, especially with methane
as the reductant, appears the most promising and has
attracted increasing attention from researchers in re-
cent years because of the relative availability and low
price of methane.
The principal reaction for the reduction of SO
2
by
CH
4
can be represented as
2SO
2
(g) + CH
4
(g) = 2[S](g) + CO
2
(g) + 2H
2
O(g)
(6)
where [S] denotes the various sulfur species (S
1
,
S
2
, ... S
8
) in the gas phase. From thermodynamic
data it can be calculated that most of the sulfur will
exist as diatomic molecules at temperatures >600
◦
C
[7].
Sarlis and Berk [8] have shown that the homoge-
neous reaction occurs at a satisfactory rate at temper-
atures between 650 and 700
◦
C, but there is significant
formation of unwanted by-products such as H
2
S,
COS, CO and H
2
. To achieve both a high conversion
of sulfur dioxide and a high product selectivity it is
necessary to work with heterogeneous catalytic sys-
tems, which operate in a lower temperature range.
Much research has been conducted in this area, the
catalysts used being activated alumina [8] and metal
sulfides such as MoS
2
and CoS
2
[9–12].
Recent studies [13–15] have shown that the rates of
many heterogeneous catalytic reactions can be consid-
erably enhanced by using microwave radiation to heat
the catalyst. This effect is thought to arise because the
reaction sites are at a considerably higher tempera-
ture than the bulk of the catalyst. This paper reports
an investigation into the effect of microwave heat-
ing on the catalytic reduction of SO
2
with CH
4
over
MoS
2
catalysts, which were prepared in a variety of
ways.
2. Experimental
2.1. Materials
The following chemicals were purchased from
Aldrich Chemicals: sulfur dioxide SO
2
, purity
>99.9%; methane CH
4
, purity >99.0%; molybdenum
disulfide MoS
2
, purity >99%.
Aluminum oxide (activated, 99% Al
2
O
3
, surface
area 90 m
2
g
−1
) and ammonium heptamolybdate
(NH
4
)
6
[Mo
7
O
24
]·4H
2
O were supplied by Alfa, John-
son Matthey plc.
2.2. Catalyst preparation
The catalysts used in this study were prepared either
by mechanical mixing or by impregnation.
2.2.1. Mechanically mixed catalysts (MC)
MoS
2
was purchased as a powder (particle diame-
ter 2 m) and activated alumina as 3.2 m pellets. The
activated alumina was ground and sieved to give the
desired size before use. The MoS
2
powder was com-
pressed into pellets in a stainless steel die before also
being ground and sieved to the desired size.
Catalyst MC-1 was made by mixing 30 wt.% MoS
2
with 70 wt.% alumina, both in the same particle size
range (152–178 m).
Catalyst MC-2 was made by first mixing 30 wt.%
MoS
2
(2 m) with 70 wt.% alumina (66 m), then
compressing the mixture into pellets, followed by
grinding and sieving to the desired size (152–178 m).
2.2.2. Supported catalysts (SC)
The supported molybdenum catalysts were prepared
by impregnating the activated alumina support with
ammonium heptamolybdate [(NH
4
)
6
[Mo
7
O
24
]·4H
2
O]
solution. An appropriate amount of solution, calcu-
lated to produce 30 wt.% MoS
2
on the support, was
X. Zhang et al. / Applied Catalysis B: Environmental 33 (2001) 137–148 139
first mixed with alumina and allowed to stand at room
temperature for a period of 6 h. The paste was then
dried in a furnace with gradual increase of temperature
to 105
◦
C over a period of 3 h, followed by isothermal
heating at 105
◦
C for 18 h. Once dried, the sample
was calcined in a flow of air at 500
◦
C for a period of
2 h. The catalyst precursor was then sulfided at a tem-
perature of 400
◦
C for 2 h, using either conventional
or microwave heating. The sulfiding gas consisted
of a mixture of H
2
S/H
2
(15 mol% H
2
S), which was
flowed over the catalyst at a rate of 90 ml min
−1
. Cat-
alyst SC-CH was sulfided in the conventional furnace
and catalyst SC-MW was sulfided in the microwave
cavity.
2.3. Characterization of catalysts
The catalysts used in this study were characterized
by using a combination of several analytical meth-
ods. The surface area was measured, before and after
reaction, by the BET method using a Micromeritics
ASAP-2000 analyzer. The composition of the catalysts
before and after reaction was determined by X-ray
powder diffraction using a PW1710 X-ray diffrac-
tometer (Phillips Electronic Instruments) with Cu
K␣ radiation. The surface structures of the catalysts
were investigated by scanning electron microscopy
(SEM) using a JEOL (JSM-T 200) scanning micros-
cope.
2.4. Experimental procedures
All reactions were carried out in a laboratory-scale,
continuous-flow reaction system with a tubular
packed-bed quartz reactor (i.d. 10 mm) which could
be placed either in a cylindrical microwave cavity or
in a conventional furnace [14]. A directional coupler
was inserted into the microwave guide system so that
the amount of microwave power reflected from the
cavity could be measured. This was minimized by
tuning the cavity with two adjustable stubs.
The temperature of the catalyst was measured with
an Accufiber optical fiber thermometer (Model 10,
Luxtron), placed at the center of the catalyst bed. This
was calibrated against a chromel/alumel thermocouple
before the catalytic studies commenced.
The product mixture from the reactor outlet passed
through a heated tube to an ice–water trap, where sul-
fur and water were condensed out. This trap was emp-
tied after each reaction. Products remaining in the gas
phase were analyzed using a quadruple mass spec-
trometer (QMS-200D, European Spectrometer Sys-
tems).
All the catalytic reactions were carried out under at-
mospheric pressure at temperatures ranging from 500
to 800
◦
C. The molar ratio of SO
2
to CH
4
in the feed
gas was varied between 1 and 3, keeping the total flow
rate constant at 40 ml min
−1
. The weight of catalyst
used in all experiments was 0.50 g.
The reaction products leaving the reactor consisted
mainly of SO
2
,CH
4
,CO
2
and CO, although, some
traces of H
2
S and H
2
were observed occasionally. The
percentage conversion of sulfur dioxide and methane
are given in terms of the percentage of sulfur dioxide
and methane that have reacted. Thus,
%SO
2
conversion = 100 ×
[(SO
2
)
in
− (SO
2
)
out
]
(SO
2
)
in
(7)
%CH
4
conversion = 100 ×
[(CH
4
)
in
− (CH
4
)
out
]
(CH
4
)
in
(8)
The products yields and selectivities are defined as
follows:
%CO
2
yield = 100 ×
(CO
2
)
out
(CH
4
)
in
(9)
%CO yield = 100 ×
(CO)
out
(CH
4
)
in
(10)
%H
2
S yield = 100 ×
(H
2
S)
out
(SO
2
)
in
(11)
%Sulfur yield = 100
×
[(SO
2
)
in
− (SO
2
)
out
− (H
2
S)
out
]
(SO
2
)
in
(12)
%Sulfur selectivity = 100 ×
sulfur yield
SO
2
conversion
(13)
The effectiveness of the various catalysts were primar-
ily judged by the conversion/yield criteria.
140 X. Zhang et al. / Applied Catalysis B: Environmental 33 (2001) 137–148
Table 1
Surface area analysis of the catalyst (m
2
g
−1
)
Catalyst Before
reaction
After
reaction (CH
a
)
After reaction
(MW
b
)
MC-1
c
79.98 51.68 39.22
MC-2 74.08 42.90 12.35
SC-CH 55.45 42.21 36.51
SC-MW
d
58.46 43.42 35.63
MoS
2
5.01 – –
␥-Al
2
O
3
97.04 – –
a
CH: conventional heating.
b
MW: microwave heating.
c
MC: mixed (mechanically) catalyst.
d
SC: support catalyst.
3. Results and discussion
3.1. Characterization of the catalysts
3.1.1. Surface area measurements
The surface areas of each of the catalysts, measured
before and after reaction, are presented in Table 1. In
each case, the reaction was carried out for a period
of one and a half hours at a measured temperature
of 800
◦
C. It is apparent that the surface areas of the
mechanically mixed catalysts were much smaller after
reaction when microwave heating was used than they
were after conventional heating. For example, the sur-
face area of catalyst MC-2 was 57% of the original
area after conventional heating, but only 17% of the
original area after microwave heating. This indicated
that microwave heating caused a considerable reorga-
nization of the catalyst structure.
The similarities in the areas for the SC-CH and
SC-MW catalysts, both before and after reaction, show
that the use of microwave or conventional heating dur-
ing the catalyst sulfiding process has little effect on
the catalyst structure.
3.1.2. X-ray diffraction analysis
The mechanically mixed catalyst MC-2 and the
supported catalyst SC-CH were both analyzed by
X-ray diffraction after reaction at a measured tempera-
ture of 800
◦
C using both conventional and microwave
heating. Three crystal structures were identified from
X-ray data [16]: the MoS
2
layer structure (with prin-
cipal peaks at 2θ = 14.3, 32.6, 33.5, 39.6 and 49.8
◦
),
␥-Al
2
O
3
(with broad, ill-defined peaks at 2θ = 37.4,
45.7 and 67.2
◦
) and ␣-Al
2
O
3
(with well defined peaks
at 2θ = 25.5, 35.1, 37.6, 43.3, 52.6, 57.4 and 68.3
◦
).
MoS
2
and ␥-Al
2
O
3
were the only species found to
be present after reaction with conventional heating,
while significant amounts of ␣-Al
2
O
3
were observed
after reaction under microwave conditions for both
catalysts examined. The peaks for ␣-alumina were
quite sharp, showing a high degree of crystallinity.
Since the phase change from ␥-alumina to ␣-alumina
is known to occur at temperatures above 1273 K [16],
and the maximum average temperature recorded in
the microwave experiments was 1073 K, it is con-
cluded that microwave heating caused some regions
of the catalyst to reach temperatures at least 200
◦
greater than the average temperatures measured by
the optical probe.
Molybdenum sulfide has a layer structure and is
known to form slabs which tend to be anchored to
the alumina particles via their edge planes [17,18].
A reasonable value for the thickness of these molyb-
denum sulfide slabs can be obtained by applying the
Debye–Scherrer line-broadening equation to the XRD
peak occurring at 2θ = 14.3
◦
[19]. This peak rep-
resents the (0 0 2) reflection and the line broadening
gives the coherence length of the crystallites along
the c-axis, L
c
. The latter has been calculated from the
equation
L
c
=
0.9λ
β cos 7.15
(14)
where λ is the wavelength of the X-rays and β the
width of the peak at half-maximum in radians. The av-
erage number of layers of MoS
2
in a slab can then be
calculated by dividing L
c
by 0.619 nm, the thickness of
a single layer. The results of this calculation are given
in Table 2, where it can be seen that the thickest slabs
were found with the mechanically mixed catalyst, for
Table 2
Average number of layers of MoS
2
in slab calculated from (0 0 2)
XRD line broadening
Catalyst
sample
Method of
heating
Width of (0 0 2) peak
at half-maximum (
◦
)
Number of
MoS
2
layers
MC-2 CH 0.50 26
MC-2 MW 0.40 32
SC-CH CH 0.90 14
SC-CH MW 0.65 20
X. Zhang et al. / Applied Catalysis B: Environmental 33 (2001) 137–148 141
which well-developed MoS
2
crystals already existed
in the starting material. It is also noticeable that, for
a given catalyst, the average thickness of the slabs
is greater after microwave heating than after conven-
tional heating, showing that more crystal growth had
occurred.
For MoS
2
powders it has been found that the width
at half-maximum of the (0 0 2) XRD peak decreases
with increasing annealing temperature from about 1.8
◦
at 450
◦
C to about 0.55
◦
at 900
◦
C [20]. On this basis
the maximum temperature reached during microwave
heating of the MC-2 catalyst must have been consid-
erably in excess of 900
◦
C, a conclusion which is in
keeping with the observed formation of ␣-alumina.
3.1.3. Scanning electron microscopy examination
Scanning electron micrographs were taken of cata-
lysts MC-2 and SC-CH, both before and after reaction.
It was found that conventional heating in the furnace
during reaction did not cause any significant change
in the appearance of the surface or the state of disper-
sion of the MoS
2
particles for both catalysts investi-
gated. However, after microwave heating some of the
molybdenum disulfide, which was initially evenly dis-
tributed as 152–178 m amorphous particles, formed
hexagonal crystals. The results for the mechanically
mixed catalyst MC-2, are shown in Fig. 1(a). For the
supported catalyst the effect of microwave heating was
even more marked with extensive formation of hexag-
onal crystals of MoS
2
being observed all over the cat-
alyst, as shown in Fig. 1(b).
Compared with the state of dispersion of the MoS
2
before reaction it was notable that a considerable mi-
gration of MoS
2
had occurred. The melting point of
MoS
2
is 1458 K, which is some 385
◦
higher than the
maximum average temperature recorded in the mi-
crowave experiments. This temperature difference is
significantly larger than that required to explain the
formation of ␣-alumina. However, complete melting
may not be necessary to explain the formation of the
hexagonal platelets, since this could equally well come
about by a rapid rate of migration of MoS
2
moieties
across the surface.
The extent of the migration occurring during mi-
crowave heating was also demonstrated by the for-
mation of spheres of loosely coagulated material with
diameters as large as 2 mm. These spheres were found
to contain both Al
2
O
3
and MoS
2
.
In summary, X-ray diffraction, scanning electron
microscopy and surface area measurements have
shown that marked changes occurred in both the MoS
2
and ␥-Al
2
O
3
components of the catalysts during mi-
crowave heating. This is attributed to the formation
of hot-spots, which reach a much higher temperature
than that measured for the catalyst bed as a whole.
This, in turn, resulted in a considerable acceleration
in the reaction rate, as described below.
3.2. Conversion of SO
2
and CH
4
as a function
of temperature with different heating methods
Fig. 2 compares the SO
2
and CH
4
conversion
efficiencies for catalyst MC-2 as a function of tem-
perature for conventional and microwave dielectric
heating at an SO
2
/CH
4
molar feed ratio of two. It can
be seen that significant conversion of SO
2
and CH
4
to products does not occur with conventional heating
until the temperature reaches 600
◦
C, whereas this can
be achieved with microwave heating at temperatures
as low as 450
◦
C. For conversions between 10 and
70%, it is found that the same percentage conver-
sion of SO
2
or CH
4
can be achieved with microwave
heating at temperatures which are about 200
◦
C lower
than those required when conventional heating was
used. This temperature difference is almost exactly
the same as that required to explain the formation
of ␣-alumina under microwave heating, although, it
would be insufficient to cause melting of the MoS
2
crystallites. There is, therefore, clear evidence from
a number of sources that some regions of the catalyst
reach temperatures which are approximately 200
◦
C
in excess of the measured temperature.
This conclusion is in keeping with that of a previ-
ous study of the catalytic decomposition of H
2
S over
MoS
2
catalysts [14], where microwave heating caused
an acceleration in reaction rates and an apparent shift
in the position of equilibrium. As in the current study,
the results could be explained only by the formation of
hot-spots, which were estimated to have temperatures
100–200
◦
C higher than the average bulk temperature
measured in the catalyst bed.
With conventional heating the percentage conver-
sion was observed to increase approximately expo-
nentially with temperature throughout the temperature
range used. Fig. 3 shows an Arrhenius plot of the data,
from which an activation energy of 117 kJ mol
−1
was
