Transient response of catalyst bed temperature in the pulsed reaction of partial oxidation of methane to synthesis gas over supported rhodium and iridium catalysts

Article abstract of DOI:10.1006/jcat.1999.2576

The mechanisms of the partial oxidation of methane to synthesis gas were investigated using a pulsed gas reaction technique and temperature jump measurement. Catalyst bed temperatures were directly measured by introducing a pulse mixture of methane and O₂ (2:1). Synthesis gas production proceeded via a two-step reaction pathway, consisting of highly exothermic methane complete oxidation to give H₂O and CO₂, followed by the endothermic reforming of methane with H₂O and CO₂ over Ir/TiO₂ catalyst. With Ir/TiO₂ catalyst, a sudden temperature increase at the front edge of the catalyst bed was observed upon introduction of the pulse, but the temperature of the rear end of the catalyst bed increased only slightly. With Rh/TiO₂ and Rh/Al₂O₃ catalysts, a different pathway for synthesis gas formation can be proposed. Rh/TiO₂ and Rh/Al₂O₃ catalysts exhibited high catalytic activity in the decomposition of CH₄ to give hydrogen and deposited carbon or CH_x, even in the presence of oxygen. The temperature at the front edge of the catalyst bed decreased upon introduction of a CH₄/O₂ pulse, and increased temperature at the rear. This indicated that H₂ formation and carbon deposition occurred via decomposition of CH₄, and then deposited carbon or CH_x generated on the Rh surface was oxidized into CO_x. However, on Rh/SiO₂, synthesis gas was produced via a two-step path similar to the case of Ir/TiO₂ catalyst. The reaction pathway with Rh-loaded catalysts depended on the support materials.

Full text of DOI:10.1006/jcat.1999.2576

Journal of Catalysis 186, 405–413 (1999)
Article ID jcat.1999.2576, available online at http://www.idealibrary.com on
Transient Response of Catalyst Bed Temperature in the Pulsed Reaction
of Partial Oxidation of Methane to Synthesis Gas over Supported
Rhodium and Iridium Catalysts
�
,
�
Kiyoharu Nakagawa, † Naoki Ikenaga, Yonghong Teng,‡ Tetsuhiko Kobayashi,‡
�
, ,1
and Toshimitsu Suzuki †
�
Department of Chemical Engineering, Faculty of Engineering and †High Technology Research Center, Kansai University, Suita, Osaka 564-8680,
Japan; and ‡Osaka National Research Institute, AIST, MITI, Ikeda, Osaka 563-8577, Japan
Received March 1, 1999; revised May 20, 1999; accepted May 26, 1999
Partial oxidation:
Mechanisms of partial oxidation of methane were studied using
a pulsed reaction technique and temperature jump measurement.
Catalyst bed temperatures were directly measured by introducing a Combustion:
1H0 = � 36 kJ/mol [3]
CH + 1/2O → CO + 2H
4
2
2
298
pulse of a mixture of methane and oxygen (2/1). With Ir/TiO2 cata-
CH4 +2O2 → CO2 +2H2O 1H0 = � 801 kJ/mol. [4]
Since partial oxidation [3] is a slightly exothermic reac-
tion, it might be operated at a lower temperature to improve
thermal efficiency.
Numerous researchers have proposed reaction mecha-
nisms for the partial oxidation of methane (reaction [3]).
Two reaction pathways have been proposed to account for
the catalytic conversion of methane with oxygen to syn-
thesis gas. The methane combustion [4], followed by steam
and CO2 reformings (reactions [1] and [2]), are commonly
298
lyst, a sudden temperature increase at the front edge of the catalyst
bed was observed upon introduction of the pulse, but the tempera-
ture of the rear end of the catalyst bed increased only slightly. The
synthesis gas production basically proceeded via a two-step path
consisting of highly exothermic methane complete oxidation to give
H2O and CO2, followed by the endothermic reforming of methane
with H2Oand CO2 over Ir/TiO2 catalyst. However, with Rh/TiO2 and
Rh/Al2O3 catalysts, the temperature at the front edge of the catalyst
bed decreased upon introduction of the CH4/O2 (2/1) pulse, and a
small increase in the temperature at the rear end was observed. At
first, endothermic decomposition of CH4 to H2 and deposited carbon
or CHx probably took place at the front edge of the catalyst bed and accepted mechanisms.
then deposited carbon or generated CHx species would be oxidized
The first report of the partial oxidation of methane to
into COx. However, on Rh/SiO2, synthesis gas was produced via a synthesis gas by using supported nickel catalyst was pub-
two-step path similar to the case of Ir/TiO2 catalyst. The reaction
pathway of partial oxidation of methane with Rh-loaded catalysts
lished as early as 1946 by Prettre et al. (2). These authors
reported that the temperature at the front of the catalyst
depended strongly on the support materials.
�c 1999 Academic Press
bed was found to be much higher than the furnace temper-
ature, and that a temperature drop occurred in the inner
part of the catalyst bed. Such phenomena have also been
observed by other researchers (3, 4). Nakamura et al. (5) in-
vestigated the same reaction over SiO2-supported rhodium
catalyst and concluded that the primary products of CO2
and H2O found at 600–700 K were further converted into
CO and H2 through the reactions of CH4 with CO2 and H2O
at an elevated temperature.
Key Words: partial oxidation; titania; alumina; pulsed reaction;
non-steady-state reaction; support effect.
1
. INTRODUCTION
Synthesis gas production from methane is indispensable
for the chemical utilization of natural gas. Synthesis gas is
also important for use in Fischer–Tropsch and methanol
syntheses. The following three reactions are of particular
industrial interest (1):
We previously reported that Ir/TiO2 catalysts has pro-
moted the reaction sequence of total oxidation of methane
to CO2 and H2O (reaction [4]), as well as the reforming
reactions to synthesis gas (reactions [1] and [2]). The CH4
2
Steam reforming:
conversion, and the CO and H selectivities decreased with
CH4 + H2O → CO + 3H2 1H0 = +206 kJ/mol [1]
increasing space velocity, but the CO selectivity increased
over Ir/TiO2 catalyst (6, 7).
The other pathway (reaction [3]) is the direct course.
2
298
CO2 reforming:
CH4 + CO2 → 2CO + 2H2 1H0 = +247 kJ/mol [2] Schmidt et al. (8–11) investigated alumina monolith-
298
405
0021-9517/99 $30.00
Copyright �c 1999 by Academic Press
All rights of reproduction in any form reserved.

4
06
NAKAGAWA ET AL.
supported rhodium and platinum catalysts for the partial tions of RhCl3 � H2O (Kishida Chemicals) and IrCl4 � H2O
oxidation of methane. The reactions were carried out un- (Mitsuwa Pure Chemicals) onto a suspended support, fol-
der adiabatic conditions using an extremely short residence lowed by evaporation to dryness. Supported catalysts were
�
4
� 2
time between 10 and 10 s. A very high selectivity to calcined at 873 K for 5 h in air prior to the reaction.
synthesis gas (>90% ) was observed, indicating the direct
pathway of CH4 to CO and H2. Boucouvalas et al. (12, 13) 2.2. Catalytic Activity Measurements
reported that Ru/TiO2 catalyst promotes the direct forma-
The reaction was carried out with a fixed-bed flow type
tion of synthesis gas, and that high CO and H2 selectivities
quartz reactor (i.d. 10 � 350 mm) at atmospheric pressure.
were obtained over Ru/TiO2 catalyst at methane conver-
Using 60 mg of a catalyst, 25 mL/min CH4 and 5 mL/min of
sions approaching zero.
O2 were introduced at temperature ranges of 673–873 K.
Buyevskaya et al. (14, 15) studied the pulse reaction
Prior to the reaction, the catalysts were heated under ni-
of methane and oxygen over Rh/Al2O3 catalyst using
trogen. The reaction temperature was raised in increments
the temporal-analysis-of-product (TAP) reactor. CO2 was
of 50 K from 673 to 873 K, and the reaction of CH4 with O2
formed as a primary product via a redox mechanism with
wasconducted for 30 min at each temperature. At the end of
the surface oxygen of Rh/Al2O3 catalyst. Dehydrogena-
the reaction period, products (H2, CO, CH4, and CO2) were
tion of methane yielding carbon deposits on the surface
analyzed using a gas chromatograph directly connected to
occurred on reduced surface sites.
the outlet of the reaction tube; columns were packed with
In this paper, we perform detailed analysis of the reac-
a Molecular Sieve 5A, 13X and Porapak Q.
tion pathway of the partial oxidation of methane to synthe-
sis gas using a transient response and pulse reaction tech-
nique. Temperature jump measurement was introduced to
2
.3. Transient Response of the Pulsed Reactions
examine the reaction pathway with supported rhodium and
Transient response measurement of the catalyst bed tem-
iridium catalysts, and the effects of supports on the methane perature in the pulsed reactions were carried out using a
activation are discussed.
fixed bed quartz reactor (i.d. 4 � 200 mm) which was set in
a horizontal position in an electric furnace. In the front and
the rear end of the catalyst bed, two thin wall sheathed ther-
mocouples (outer diameter 0.6 mm) were set and 100 mg
of the catalyst was charged, as shown in Fig. 1. A pulse
of CH4 and O2 (2/1) or CO2 (1/1) mixed gas was intro-
2
. EXPERIMENTAL
2
.1. Catalysts
The supports used were Al2O3, SiO2 (JRC-ALO-4 sur- duced with a six-port gas sampling valve equipped with
2
2
face area, 177 m /g; JRC-SIO-4 surface area, 347 m /g; ref- measuring tubes, under a stream of Ar carrier gas. Before
erence catalyst provided by the Catalyst Society of Japan), the reaction, the catalysts were reduced with H flow for
2
2
and TiO2 (Japan Aerosil Co., surface area, 50 m /g). The 60min at 873K. The reaction temperature wascontrolled by
supported rhodium and iridium metal catalysts containing monitoring the outside temperature of the reactor wall by
5
wt% metal were prepared by impregnating aqueous solu- using a programmable controller.
FIG. 1. Schematic representation of the flow reactor measuring temperature jump.

TRANSIENT RESPONSE OF CATALYST BED TEMPERATURE
407
Analyses of the gases during the pulsed reactions
were made using an on-line quadruple mass spectrometer
HAL201, Hiden Analytical Ltd.). The mass spectrometer
TABLE 1
Effect of Space Velocity on the CH4 Conversion
and Product Selectivities
(
scanned the parent peaks of the five compounds, H2, CH4,
CO, O2, and CO2, within 1 s, and repeated scans were col-
lected in a personal computer. Measured intensities were
corrected for the relative sensitivities of the respective ions. Catalyst (h mL/g-cat)
Transmission electron microscope (TEM) wasperformed
using a Hitachi H-9000 instrument on fresh and used cata-
lyst samples so that the morphology of the deposited carbon
could be determined.
Selectivity
Conversion
SV
CH4
(% )
CO
CO2
H2
H2/CO
ratio
�
1
(% ) (% ) (% )
Ir/TiO2
30,000a
0,000
0,000
30,000a
0,000
25.7
21.6
20.5
81.8 18.2 83.1
69.2 30.8 73.6
65.5 34.5 69.4
2.0
2.1
2.1
6
9
Rh/TiO2
26.5
22.3
23.0
81.6 18.4 87.3
64.8 35.2 78.7
65.5 34.5 77.1
2.1
2.4
2.4
6
3
. RESULTS
90,000
Note. Catalyst: 60 mg; metal loading level = 5 wt% ; CH4/O2 = 5.0; reac-
3
.1. Effect of TiO2-Supported Rhodium and Iridium
Metals on the Performance of Partial
Oxidation of Methane
tion temperature, 873 K.
a
Reference (6).
The effect of reaction temperature on the product con- high methane conversion (>90% ) and a high synthesis gas
centrations during partial oxidation of methane with Rh selectivity (>90% ). Such an equilibrium limitation might
5 wt% )/TiO2 and Ir(5 wt% )/TiO2 catalysts are shown in be the reason for the observed lower methane conversion
(
Figs. 2a and 2b. TiO2 was used as a support of Rh and Ir of 25% at 873 K (Fig. 2). At 873 K, reactions [3] and [4] seem
loaded catalysts, since Ir/TiO2 catalyst afforded a high CH4 to occur. If one assumes that the equilibrium concentration
conversion to synthesis gas and the initial activity of this ratio of CO/CO2 is about 4, methane conversion can be cal-
catalyst was maintained for a long period. Both Ir and Rh culated as about 25% . This value agreed with experimental
catalysts showed a high catalytic activity at 873 K for the results.
formation of synthesis gas and afforded almost the same
Table 1 shows the effect of space velocity on the CH4
product concentrations at above 823 K. Below 773 K, how- conversion and product selectivities at 873 K. When Ir/TiO2
ever, Rh/TiO2 catalyst produced a higher concentration of catalyst was used, the CH4 conversion and the CO and H2
H2 than Ir/TiO2 catalyst.
selectivities decreased with increasing space velocity, but
According to the thermodynamics reported for the CO2 selectivity increased. Although the differences in the
methane and oxygen system (CH4 :O2 = 2 : 1) (1), the for- CH4 conversion was not large, change in the CO2 selectivity
mations of CO2 and H2O are dominant at temperatures was significant, and such differences are much larger than
lower than about 823 K and the synthesis gas formation experimental errors. On the other hand, Rh/TiO2 catalyst
becomes favorable at higher temperatures. Temperatures maintained constant CH4 conversions and high H2 selectiv-
higher than 923 K are thermodynamically required for a ities over the range of space velocities examined. Rh/TiO2
FIG. 2. Effect of temperature on the product concentrations over TiO2-supported Ir(5 wt% ) and Rh(5 wt% ) catalysts: ( ) CO2, (ᮀ) CO, ( ) H2,
�
1
(
᭹) CH4. Reaction conditions: CH4 :O2 = 5 : 1, flow rate = 30 mL/min; catalyst: 60 mg, SV = 30,000 h mL/g-catalyst.

4
08
NAKAGAWA ET AL.
catalyst also showed higher H2 selectivity and H2/CO ratio
than Ir/TiO2 catalyst.
3
.2. Decomposition of Methane over Rhodium
and Iridium-Loaded Catalyst
Figure 3 shows the transient response of H2 production
when the flowing gas was switched from Ar to CH4 at 873 K
reaction [5]) over Rh(5.0 wt% )/TiO2 and Ir(5 wt% )/TiO2
(
catalysts. H2 was the only gaseous species observed (no
higher hydrocarbons):
1H0 = +75 kJ/mol.
[5]
→
�
CH4
C + 2H2
298
When CH4 was supplied to Ir/TiO2 catalyst, a slight H2 re-
sponse appeared instantaneously and then decreased to a
low level. On the other hand, methane decomposition con-
tinued to give H2 over Rh/TiO2 catalyst for 2400 s. These
results show that Rh/TiO2 exhibited a much higher activity
toward the decomposition of CH4 than did Ir/TiO2.
In order to investigate the effect of support of Rh loaded
catalysts on the decomposition of CH4, Rh(5.0 wt% )/Al2O3
and Rh(5.0 wt% )/SiO2 catalysts were tested. As shown
FIG. 4. Decomposition of CH4 over supported Rh(5 wt% ) catalysts.
in Fig. 4, Rh(5.0 wt% )/Al2O3 catalyst initially showed a Reaction conditions: reaction temperature 873 K; catalyst, 100 mg; CH4
�
2
flow rate, 10 mL/min; 1 Torr = 133.3 N m
.
high activity toward decomposition of CH4. However, the
concentration of H2 gradually decreased with increasing
time on stream and H2 production was not detected after
4
800 s. Verykios et al. (16) reported similar results over
after 600 s on stream the rate of H2 production decreased
to about one-seventh of the initial rate. Rh(5.0 wt% )/SiO2
catalyst exhibited behavior similar to that of Ir/TiO2 cata-
lyst.
Rh(0.5 wt% )/Al2O3 catalyst at 923 K. They concluded that
Figures 5, 6, and 7 show transmission electron microscope
(
TEM) images of Rh/TiO2, Rh/Al2O3, and Rh/SiO2 cata-
lysts fresh and after the reaction for 2400 s. No specific car-
bon formation was observed on Ir/TiO2 catalyst. On the
other hand, Rh/TiO2 catalyst clearly showed formation of
a “whisker” type carbon (Fig. 5b). This type of carbon has
also been observed on the steam and carbon dioxide re-
forming catalysts, and particularly on nickel catalyst (1, 17,
1
8). Whisker carbon formation did not alter the rate of H2
production. In our case, no deactivation of the Rh/TiO2
catalyst for methane decomposition was observed with car-
bon deposition during the reaction for 2400 s. Whisker type
carbon was not formed on Rh/Al2O3 catalyst (Fig. 6b), but
probably “encapsulate” type carbon would be formed (19),
since formation of H2 gradually decreased with increasing
time on stream and H2 production ceased after 4800 s. The
encapsulate form of carbon has been reported to build up
with time and to deactivate the catalyst by covering the
nickel surface for partial oxidation of CH4 (19). On Rh/SiO2
catalyst, no type of carbon species was observed, due to
the small activity for CH4 decomposition (Fig. 7b). These
results indicate that the catalytic activities of CH4 decom-
position and deposited carbon species over the Rh loaded
catalysts depended strongly on support materials.
FIG. 3. Decomposition of CH4 over TiO2-supported Ir(5 wt% ) and
Rh(5 wt% ) catalysts. Reaction conditions: reaction temperature, 873 K;
catalyst, 100 mg; CH4 flow rate, 10 mL/min; 1 Torr = 133.3 N m�
2
.

TRANSIENT RESPONSE OF CATALYST BED TEMPERATURE
409
ature at the front edge decreased with the introduction of
CH4/CO2 pulse, and a slight decrease in the temperature of
the rear end of the catalyst bed was observed. The decrease
in the catalyst bed temperature increased with increasing
amounts of reactant introduced. Since the total flow rate
was kept constant during the introduction of mixed gas,
this temperature decrease could not be ascribed to the in-
crease in the cold gas flow rate, but could safely be ascribed
to endothermic reaction. Thus, the validity of this measure-
ment for the test of endo- or exothermic reactions was con-
firmed.
FIG. 5. Transmission electron microscope (TEM) showing the de-
posited carbon on the surface of the Rh(5 wt% )/TiO2 catalyst: (a) fresh,
(
b) after reaction.
3
.3. Transient Response of Catalyst Bed Temperature in
the Pulsed Reaction of Partial Oxidation of Methane
To investigate the mechanisms of partial oxidation of
methane over supported rhodium and iridium catalysts, a
pulsed reaction technique was employed to measure tran-
sient temperature changes in the catalysts beds.
First, in order to prove the validity of this method, a sim-
ple endothermic reaction for CO2 reforming was examined.
Fig. 8 shows the transient temperature responses of Ir/TiO2
FIG. 6. Transmission electron microscope (TEM) showing the de-
posited carbon on the surface of the Rh(5 wt% )/Al2O3 catalyst: (a) fresh,
catalyst against a pulsed injection of CH4/CO2. The temper- (b) after reaction.

4
10
NAKAGAWA ET AL.
effluent composition. Difference in the product intensities
varying with pulse size or catalysts (Figs. 8–12) might be
caused by differences in activities of each catalyst. Prod-
uct peak area and temperature response were changed by
varying with pulse size.
When Ir/TiO2 catalyst was used, a sudden rise in tem-
perature at the front edge of the catalyst bed was observed
upon introduction of the pulse, but the temperature of the
rear end of the catalyst bed was only slightly increased. Sim-
ilar phenomena have also been observed in a conventional
fixed-bed continuous flow system by other researchers
(
3, 4).
However, when Rh/TiO2 and Rh/Al2O3 catalysts were
used, the temperature at the front edge decreased upon
introduction of the CH4/O2 pulse, and an increase in the
temperature at the rear end was observed. Reproducibil-
ity of temperature changes of the catalyst bed was con-
firmed by changing carrier gas, Ar, flow rate from 10 to
4
0 mL/min. At each Ar flow rate, similar temperature drops
at the front edge were observed over Rh/TiO2 catalyst.
Since exothermic peaks were observed at the rear edge
of the catalyst beds in the cases of Rh/TiO2 and Rh/Al2O3
FIG. 7. Transmission electron microscope (TEM) showing the de-
posited carbon on the surface of the Rh(5 wt% )/SiO2 catalyst: (a) fresh,
(
b) after reaction.
Amongthe four reactions, [1]and [2]are highlyendother-
mic, [3] is moderately exothermic, and [4] is highly exother-
mic. Observation of the catalyst bed temperature by means
of injecting a pulse of reactants may provide information
about the reaction taking place at the catalyst bed.
Catalyst bed temperature was directly measured by in-
troducing a pulse of a mixture of methane and oxygen.
Figures 9, 10, 11, and 12 show the transient temperature re-
sponses of Ir/TiO2, Rh/TiO2, Rh/Al2O3, and Rh/SiO2 cata-
lysts against a pulsed injection of CH4/O2, together with the 133.3 N m
FIG. 8. Temperature profile at front and rear edges of the catalyst bed
and the responses of products when pulsing CH4 and CO2 over Ir(5 wt% )/
TiO2 catalyst. Reaction conditions: Ar carrier, 10 mL/min; mixed gas,
CH4 :CO2 = 1 : 1, 1 or 3 mL; furnace temperature, 873 K; 1 Torr =
�
2
.

TRANSIENT RESPONSE OF CATALYST BED TEMPERATURE
411
the dependence of catalytic activity on the support mate-
rials, in relation with the partial oxidation of methane and
the CO2 reforming. Furthermore, when Ir/TiO2 catalyst was
used, the CH4 conversion and the CO and H2 selectivities
decreased with increasing space velocity, but CO2 selectiv-
ity increased (Table 1). These results again indicate that the
slower CO2 reforming reaction was suppressed to give a
large amount of unreacted CO2.
When Ir/TiO2 and Rh/SiO2 catalysts were used, a sudden
rise in temperature at the front edge of the catalyst bed was
observed upon introduction of the pulse of CH4 and O2,
but the temperature of the rear end of the catalyst bed was
only slightly increased (Figs. 9 and 12). If the heat loss from
the catalyst bed and thermocouples were nil, the estimated
temperature increase in the complete oxidation would be
ca. 200 K for the 3-mL pulse. In our apparatus, heat loss
could not be eliminated. The observed temperature rise was
about 30 K. As seen in Fig. 8, the temperature decrease in
the CO2 reforming was ca. 20 K. The difference in the heat
of reaction between exothermic combustion and endother-
mic CO2 reforming was significantly large, as compared to
the observed temperature differences. Endothermic CO2
FIG. 9. Temperature profile at front and rear edges of catalyst bed and
the responses of products when pulsing CH4 and O2 over Ir(5 wt% )/TiO2
catalyst. Reaction conditions: Ar carrier, 10 mL/min; mixed gas, CH4 :O2 =
2
: 1, 1 or 3 mL; furnace temperature, 873 K; 1 Torr = 133.3 N m� 2
.
catalysts, it seemed unlikely that exothermic reaction would
occur before the thermocouple at the front edge of the cata-
lyst bed in the cases of Rh/TiO2 and Rh/Al2O3 catalysts. On
the other hand, Rh/SiO2 showed a rise in temperature at
the front edge of the catalyst bed upon introduction of the
pulse.
In each of the catalysts used, H2, CO, and CO2 were ob-
tained. In all cases, temperature changes increased with in-
creasing amount of pulsed reactant gas. During the mea-
surement of temperature changes, an effluent was analyzed
by a Q-mass spectrometer. In all cases, unreacted CH4, CO,
H2, and CO2 were detected. The amount of CO2 was higher
than that in the steady reaction.
4
. DISCUSSION
The reaction pathway of synthesis gas production over
the Ir(5 wt% )/TiO2 catalyst seemed to proceed via the two-
step reactions that Ir/TiO2 catalyst might have promoted
the reaction sequence of total oxidation of methane to CO2
and H2O (reaction [4]) and reforming reactions to synthe-
sis gas (reactions [1], [2]). This was previously supported by O2 = 2 : 1, 1 or 3 mL; furnace temperature, 873 K; 1 Torr = 133.3 N m
FIG. 10. Temperature profile at front and rear edges of catalyst bed
and the responses of products when pulsing CH4 and O2 over Rh (5 wt% )/
TiO2 catalyst. Reaction conditions:Ar carrier, 10mL/min;mixed gas, CH4 :
�
2
.

4
12
NAKAGAWA ET AL.
primary products of the methane partial oxidation reaction
with these catalysts. Buyevskaya et al. (14, 15) studied the
pulse reaction of methane and oxygen over Rh/Al2O3 cata-
lyst using the TAP reactor. CO2 was formed as a primary
product via a redox mechanism with the surface oxygen of
Rh/Al2O3 catalyst. In our apparatus, at first, H2 formation
and carbon deposition probably took place via the decom-
position of CH4, and then deposited carbon or CHx species
generated on the Rh surface were oxidized into COx over
Rh/TiO2 and Rh/Al2O3. H2 and carbon seemed to be the
primaryproduct;however, it wasdifficult to saythat CO was
the primary product in the partial oxidation of methane in
the pulse reaction over Rh/TiO2 and Rh/Al2O3. It seems to
be reasonable that measurement of temperature changes
of catalysts bed reflected only the overall heat change. If
the heat loss from the catalyst bed and thermocouples were
nil, compared with the our present results, the estimated
temperature decrease in the decomposition of methane
would be much larger. It is possible that both exother-
mic and endothermic reactions could occur anywhere in
the catalyst bed. If synthesis gas was formed at the front
of the catalyst bed, a slightly exothermic response could be
FIG. 11. Temperature profile at front and rear edges of catalyst
bed and the responses of products when pulsing CH4 and O2 over Rh
5wt% )/Al2O3 catalyst. Reaction conditions:Ar carrier, 10mL/min;mixed
gas, CH4 :O2 = 2 : 1, 1 or 3 mL; furnace temperature, 873 K; 1 Torr =
33.3 N m
(
�
2
1
.
and H2O reforming reactions proceeded to induce not only
heat loss from the reactor wall, but also a decrease in the
rise in temperature of the catalyst bed. If a direct partial ox-
idation reaction [3] proceeded, due to the small degree of
heat generation such a large rise in temperature could not
be observed. Thus, when Ir/TiO2 and Rh/SiO2 were used,
the temperature increase in the front edge of the catalyst
bed could safely be ascribed to the complete oxidation of
methane. The results for the Rh/SiO2 catalyst are consis-
tent with the proposed mechanism by Nakamura et al. (5),
although these authors did not provide evidence for this
mechanism.
It is surprising that the temperature drop at the front
edge of the catalyst bed was observed for both Rh/TiO2
and Rh/Al2O3, even when methane and oxygen were intro-
duced in a ratio of 2 to 1. In the steady flow, Rh/TiO2 catalyst
maintained high H2 selectivities at the high space veloc-
ity (Table 1). Rh/TiO2 and Rh/Al2O3 catalysts had a high
activity for decomposition of CH4 (Figs. 3 and 4). Under
the high space velocity, Hickman et al. (8–11) investigated
the partial oxidation of methane using Rh and Pt/alumina
monolith catalysts. They proposed that CO and H2 were the 2 : 1, 1 or 3 mL; furnace temperature, 873 K; 1 Torr = 133.3 N m
FIG. 12. Temperature profile at front edge of catalyst bed and the
responses of products when pulsing CH4 and O2 and Rh(5 wt% )/SiO2
catalyst. Reaction conditions: Ar carrier, 10 mL/min; mixed gas, CH4 :O2 =
�
2
.

TRANSIENT RESPONSE OF CATALYST BED TEMPERATURE
413
observed. However, an endothermic response was detected catalysts exhibited a high catalytic activity in the decom-
only at the front edge of the catalyst bed over Rh/TiO2 and position of CH4 to give hydrogen and deposited carbon or
Rh/Al2O3 catalysts. Therefore, in the high methane con- CHx, even in the presence of oxygen. The temperature at
centration, compared to the oxidation of methane, decom- the front edge of the catalyst bed decreased upon introduc-
position of methane to give carbon or CHx would easily tion of a CH4/O2 pulse, and an increase in the temperature
proceed at the front edge of the catalyst bed over Rh/TiO2 at the rear end was observed, indicating that H2 formation
and Rh/Al2O3 catalyst.
and deposition of carbon probably took place via decompo-
Boucouvalas et al. (12, 13) reported that synthesis gas sition of CH4, and then deposited carbon or CHx generated
over Ru/TiO2 catalyst was, to a large extent, formed via the on the Rh surface was oxidized into COx. In the case of Rh-
direct partial oxidation scheme. CO and CO2 were proba- loaded catalysts, the reaction pathway depended strongly
bly formed by parallel routes via two different sites of the on the support materials.
catalyst.
When Ru/TiO2 catalyst was used, a sudden rise in the
temperature at the front edge of the catalyst bed was ob-
ACKNOWLEDGMENTS
served upon introduction of the 3-mL pulse of CH4 and O2.
This work was supported by a Grant-in-Aid on Priority Areas
However, the temperature drop at the front edge of the (09218255) from the Ministry of Education, Science, Sports, and Culture
of Japan. K. Nakagawa is grateful for his Research Fellowship from Japan
Society for the Promotion of Science (JSPS) for Young Scientists.
catalyst bed was observed for 1-mL pulses of CH4 and O2.
These results seemed to exhibit two possibilities of synthe-
sis gas formation routes over Ru/TiO2 catalyst. Catalytic
pathways depended on the reaction conditions such as con-
centrations of the reactant and flow rates.
REFERENCES
1
2
. Tsang, S. C., Claridge, J. B., and Green, M. L. H., Catal. Today 23, 3
The slight temperature increase of the rear edge of the
catalyst bed on Ir/TiO2 catalyst might be ascribed to the heat
conduction from the front and middle part of the catalyst
bed. Synthesis gas formation over Ir/TiO2 catalyst would
proceed basically via the two-step path. Methane combus-
tion would occur in the front edge of the catalyst bed. Since
a large amount of unreacted CO2 was observed, the total
heat of the reaction would be exothermic.
(
1996).
. Prettre, M., Eichner, CH., and Perrin, M., Trans. Faraday Soc. 43,
35 (1946).
3. Dissanayake, D., Rosynek, M. P., Kharas, K. C. C., and Lunsford, J. H.,
J. Catal. 132, 117 (1991).
. Vermeiren, W. J. M., Blomsma, E., and Jacobs, P. A., Catal. Today 13,
27 (1992).
. Nakamura, J., Umeda, S., Kubushiro, K., Kunimori, K., and Uchijima,
3
4
5
4
T., Sekiyu. Gakkaishi 36, 97 (1993).
On the other hand, when Rh/TiO2 and Rh/Al2O3 cata-
lysts were used, the temperature increase of the rear edge of
the catalyst bed might be caused by oxidation of unreacted
methane, CHx, H2 which were made from the decomposi-
tion of methane or oxidation of reduced Rh metal could be
possible due to a high concentration of oxygen. Probably,
hydrogen was produced mainly by decomposition of CH4
on the front edge of the catalyst bed, and then generated
H2 and CHx species were oxidized into H2O and COx in the
Rh/TiO2 and Al2O3 catalysts.
6. Nakagawa, K., Ikenaga, N., Suzuki, T., Kobayashi, T., and Haruta, M.,
Appl. Catal. A 169, 281 (1998).
7
. Nakagawa, K., Anzai, K., Matsui, N., Ikenaga, N., Suzuki, T.,
Kobayashi, T., Teng, Y., and Haruta, M., Catal. Lett. 51, 163 (1998).
. Hickman, D. A., and Schmidt, L. D., J. Catal. 138, 267 (1992).
. Hickman, D. A., Haupfear, E. A., and Schmidt, L. D., Catal. Lett. 17,
223 (1993).
8
9
1
1
0. Hickman, D. A., and Schmidt, L. D., Science 259, 343 (1993).
1. Bharadwaj, S. S., and Schmidt, L. D., Fuel Proc. Technol. 42, 109
(
1995).
2. Boucouvalas, Y., Zhang, Z., and Verykios, X. E., Catal. Lett. 40, 189
1996).
3. Boucouvalas, Y., Zhang, Z. L., Efstathiou, A. M., and Verykios, X. E.,
Stud. Surf. Sci. Catal. 101, 443 (1996).
1
1
1
1
1
1
(
5
. CONCLUSION
4. Buyevskaya, O. V., Wolf, D., and Baerns, M., Catal. Lett. 29, 249
(
1994).
5. Buyevskaya, O. V., Walter, K., Wolf, D., and Baerns, M., Catal. Lett.
8, 81 (1996).
A pulsed reaction technique of measuring the tempera-
ture jump of the catalyst bed afforded strong support that
synthesis gas was formed via a two-step reaction pathway,
which consisted ofmethane complete oxidation to give H2O
and CO2, followed by the reforming of methane with steam
and CO2 for Ir/TiO2 and Rh/SiO2 catalysts.
3
6. Tsipouriari, V. A., Efstathiou, A. M., and Verykios, X. E., J. Catal. 161,
31 (1996).
7. Rostrup-Nilsen, J. R., in “Catalysis, Science and Tecnology” (J. R.
Anderson and M. Boudart, Eds.), Vol. 5, pp. 1–117. Springer, Berlin,
1984.
On the other hand, when Rh/TiO2 and Rh/Al2O3 cata-
lysts were used, a different reaction pathway for synthe-
1
8. Claridge, J. B., Green, M. L. H., Tsang, S. C., York, A. P. E., Ashcroft,
A. T., and Battle, P. D., Catal. Lett. 22, 299 (1993).
sis gas formation can be proposed. Rh/TiO2 and Rh/Al2O3 19. Trimm, D. L., Catal. Rev. Sci. Eng. Chem. 16, 155 (1987).