(363e) An Experimental and Thermodynamic Investigation On the Effect of Metal Oxide Dopants On Ceria-Based Solar Thermochemical Cycles

Two-step solar
thermochemical cycles using ceria-based oxides as a reaction intermediate
provide the potential to produce solar fuels directly from concentrated solar
energy, as indicated by the redox reactions shown below

where M is the dopant type
(M=Gd, Sm, Fe, etc.), x is the dopant concentration, and d is
the degree of oxygen nonstoichiometry. In the first, high temperature reaction
(T>1400 ^oC), ceria is thermally reduced, evolving O₂,
and in the second lower temperature reaction (T<1400 ^oC), the
reduced ceria is oxidized with H₂O and/or CO₂ to produce
H₂ and/or CO.

Dopants of an oxidation
state of +3, (i.e. Gd, Sm, Fe) have been shown to have a positive effect on the
O₂ capacity and reduction temperature of ceria based materials,
especially in solid oxide fuel cell applications where ceria is reduced
electrochemically. However, the effect of reducing these materials with thermal
energy at much higher temperatures, specifically for applications of solar fuel
production is not as well understood. For example, higher temperatures may
result in enhanced sintering and the formation of un-desired oxide phases, both
of which may affect fuel production rates, yields, and repeatability.

We have experimentally explored
the effect of metal oxide dopants (Sm, Gd, Fe) on the ceria (CeO₂)
thermochemical cycle using a high temperature thermogravimetric analyzer (TGA).
Materials have been cycled repetitively at oxidation and reduction temperatures
of 1000 to 1450 ^oC, respectively, and the effect of cycling, dopant
type and dopant concentration (x=0.1 and 0.2) on fuel production yields and
rates has been quantified. Preliminary results indicate that the morphological
stability of both Sm and Gd doped ceria are superior to pure ceria. However,
extensive sintering was observed when doped with Fe due to the formation of
mixed Ce-Fe oxides confirmed by powder x-ray diffraction analysis.
Unexpectedly, the O₂ capacity of pure ceria was greater than all
doped materials explored. We have also considered how the incorporation of
doped ceria into a redox cycle such as this affects the thermodynamically
limited efficiency of the process. Thermodynamic parameters have been
extrapolated from the literature and incorporated into a second law analysis
and preliminary results indicate that efficiency gains compared to pure ceria can
be realized for both Sm and Gd doped ceria due to decreased enthalpic
requirements necessary to drive the reduction.