Submitted to the University of New South Wales in Partial Fulfillment Requirements for the Degree of Doctor of Philosophy in Chemical Engineering. Australia. April 2010. 281 р.ABSTRACT
A process for the production of clean hydrogen from methane based upon the sequential
reduction and oxidation of metal oxides has been studied. The original process, based on
iron oxide, suffers from significant disadvantages including deactivation by sintering and
coke deposition. Improvement of the iron based system and identification and
development of alternative metal oxides for hydrogen production has formed the basis of
this study.
The literature review outlines current methods for hydrogen production, followed by a
review of the Steam-Iron Process as an improved and simpler method for clean hydrogen
production. Thermodynamic assessment shows Fe3O4/FeO/Fe, WO3/WO2/W and
SnO2/SnO/Sn to be the most prospective systems for the Steam-Metal Process.
Experimental testing showed that Fe and W based systems were suitable for hydrogen
production, but Sn based systems were unsuitable due to poor reducibility using methane.
Attention was then focused on the addition of CeO2/ZrO2 promoters to Fe and W based
systems in order to improve reactivity and prevent catalyst deactivation. CeO2/ZrO2
promoted Fe2O3 showed improved redox reactivity and increased stability, with
formation of FeO. This aided in mitigation of sintering and introduced the possibility of
prevention of coking, as catalysed by methane decomposition over fully reduced Fe
metal.
Although WO3 was found to be a suitable oxide, complete reduction to tungsten metal
resulted in the formation of tungsten carbide and contamination of hydrogen produced.
The formation of 31mol% [CeO2/ZrO2] / 69 mol% WO3 showed stabilised reduction
using methane, allowing for redox cycling of the WO3-WO2 couple and preventing
complete reduction to W metal. The use of the doped metal oxide showed the best
performance of all the metal oxides tested, with clean hydrogen production over multiple
redox cycles and high metal oxide stability.
iii
Further kinetic studies of both the reduction and oxidation reactions show reduction is
chemical reaction controlled process (WO3/WO2.9 WO2) with an apparent activation
energy of 142 ± 3 kJ/mol. Oxidation is also fitted to a chemically controlled process, with
a reaction rate expression derived as:
2 2
[0.064 ( 0.00038)] ( 108750/8.314 T ) [ ]0.75
H HO r = +F× ⋅e− × ⋅P
The apparent activation energy for oxidation was calculated as 109 ± 1 kJ/mol.
