Fig. 3 Change in catalytic activity for the decomposition of ethane as a function of time. (a) System = AC Darco KB-B/fixed-bed reactor: (8) C
2
conversion,
.) organic product yield. T = 923 K, W/F = 125 g h mol2 , ethane: N
1
= 2:98. Inset: Plot of H /C molar ratio against time. (b) System = graphite/Ag-Pd
conversion, (.): organic products yield. T = 923 K, W/F = 125 g h mol , ethane: N = 2:98. Inset: Plot of H t/C molar
2 2
(
2
2
2
1
membrane reactor: (8) C
2
ratio against time.
was between that of the PMR with argon and that of the FBR
system.These findings indicate that the combined use of the
carbon catalyst and the PMR system was very effective for
On the other hand, we found that the deactivation did not occur
when graphite catalyst (2) was used, and the ethane conversion
increased [Fig. 3(b)] as carbon was formed. In the initial stages,
C
2
H
6
pyrolysis.
2 2
the H t/C ratio (the molar ratio of total amount of H to carbon)
Fig. 2 shows the effects of some carbon catalyst/PMR
was greater than 1.5 [the value given by stoichiometric ethane
decomposition, see eqn. (1)] because of the formation of
organic products ( > 20%). After 5 h, however, the ratio
approached 1.5, and subsequently remained constant; this
indicated that the decomposition reaction occurred predom-
inantly. This is the first example of a catalyst not being
deactivated even at high ethane conversion. It is possible that
the graphite-like carbon produced by the reaction also effi-
ciently catalyzes the reaction in the membrane reactor§ over the
temperature range 800–1000 K. Further investigations are
currently underway to elucidate the decomposition mechanism
operative in the graphite/Ag-Pd membrane reactor system.
In summary, we have found that graphite is a very good
catalyst for converting ethane to hydrogen in a Ag-Pd
membrane system with high activity and no deactivation. We
believe this to be an excellent procedure for hydrogen
production for PEM fuel cells.
systems on the product yields at 973 K. The efficiency of
pyrolysis (carbon yield) was in the order: 2/graphite > 6/AC
Darco G-60 > 3/AC Darco KB-B > 1/AC Darco G-60/FBR >
5
/C60 soots > 4/quartz sand. In the graphite catalyst, in
particular, organic by-products were formed only in ca. 4%
yield. In the activated carbons (3 and 6), the yield of organic
products was ca. 15%, although their carbon yields were close
to that on graphite. The use of a fixed-bed reactor (FBR)
resulted in the formation of a larger amount of organic products
(
(
(
> 35% yield) than that in a Pd-membrane reactor (PMR)
compare 1 and 6). In the cases of C60 soots (5) and quartz sand
4), the dehydrogenation of ethane to form ethylene predom-
inates over decomposition.
In order to shed more light on the characteristics of carbon
catalyst/PMR systems, the effects of the reaction parameters on
the ethane decomposition were briefly examined. The effect of
the flow rate of the argon on the activity was found not to be
significant under the conditions employed, provided the flow
2
1
Notes and references
rate is above 100 ml min . This suggested that the rate of
hydrogen permeation was, in general, greater than that of ethane
decomposition over the carbon catalyst under the conditions
employed. The ethane conversion at 923 K increased with an
†
In order to ensure a smooth gas flow, a weight ratio of graphite:quartz of
1:5 was used; the results were not affected by the amount of quartz.
2
1
‡ W/F = weight of catalyst (g) over flow rate (mol h ).
2
1
§ In these cases, both graphite and PMR are important. In fact, the minimum
amount of organic products was formed in the graphite catalyst/PMR
system, as can be seen from Fig. 2. In the graphite/PMR system, graphite-
like carbon formed over the graphite surface would not undergo successive
reaction such as methane formation from hydrogen and the carbon, because
almost all of the hydrogen was eliminated from the graphite surface. Thus,
it seems likely that the stable graphite-like carbon product could also
catalyze the reaction. On the contrary, in the AC/FBR system, the carbon
product could undergo many successive reactions, since hydrogen was still
present over the surface. Thus, in this case, it seems unlikely that the carbon
product could catalyze the reaction.
increase in contact time (W/F = 25–125 g h mol ). As
expected thermodynamically, lower ethane concentrations
(
2–20%) gave higher ethane conversions, over the temperature
range from 773 to 973 K.
The mechanism of decomposition over the carbon catalysts is
not clear. However, the reaction is promoted by the removal of
the hydrogen formed (Fig. 1) and it seems likely that a material
with a graphite-like nature is formed during decomposition, as
2
shown by X-ray powder diffraction and H temperature
programmed desorption of the carbon formed, carried out by
analyzing methane formed by hydrogenation of the carbon
product. It is well-known that catalyst deactivation occurs as a
1
N. Dave and G. A. Foulds, Ind. Eng. Chem. Res., 1995, 34, 1037.
5
2 M. L. Cubeiro and J. L. G. Fierro, J. Catal., 1998, 179, 150.
result of the formation of graphite-like carbon when metal-
3
4
5
N. Z. Muradov, Int. J. Hydrogen Energy, 1993, 18, 211; T. Zhang and
M. D. Amiridis, Appl. Catal. A: Gen., 1993, 167, 161.
K. Murata and H. Ushijima, J. Chem. Soc., Chem. Commun., 1994,
oxide catalysts are used in the temperature range 600–800 K. In
fact, even with the activated carbon catalyst (3)/FBR system
without metal, the ethane conversion at 923 K decreased as the
reaction proceeded [Fig. 3(a)]; as a result, the H /C ratio
2
increased due to the reduction of carbon formation and
1
157.
N. Z. Muradov, Energy Fuels, 1998, 12, 41.
predominant formation of organic compounds [Fig 3(a), inset].
Communication 9/00009G
574
Chem. Commun., 1999, 573–574