Summary
Mixed protonic-electronic conductors can simultaneously transport protons and electrons through their structure and they can be applied as hydrogen separation membranes and catalytic membrane reactors. Hydrogen separation at high temperature is a key step in power plants applying pre-combustion strategies, e.g. integrated gasification combined cycle (IGCC), where CO2 and H2 are separated after fuel gasification and shift, resulting in a final exhaust stream consisting of humid CO2, which can readily be liquefied and stored. Another challenging use of hydrogen-permeable membranes is the integration in new catalytic reactors, e.g. non-oxidative coupling of methane and aromatization, hydrocarbon reforming and water gas shift reaction. These membrane reactors allow the continuous and controlled hydrogen removal or feeding and consequently enable to surpass equilibrium conversion or increase product selectivity by avoiding the direct contact between reactants.
Hydrogen separation does not require the application of any electrical field or external current, main advantage of dense proton conducting membranes. When a hydrogen chemical potential gradient exists, hydrogen permeation occurs due to the ambipolar diffusion of protons and electrons. The proton conductor oxide requirements are: to have a deficient oxygen structure, to incorpote water in their structure and to allow the fast transport of protons when they are incorporated in the structure. Specifically, the present thesis is based in the development and characterization of compounds based on zirconates (BaZrO3) and more widely on tungstates (Ln6WO12).
The aim of the study about zirconate based compounds was: the improvement of the protonic and electronic conductivity, the reduction of grain boundary resistance (that limit the conductivity in this kind of compounds) and the improvement of the stability in CO2 containing gases. The improvement of the total conductivity was achieved by the partial substitution in the B-position (Zr) in the selected reference compound, BaZr0.9Y0.1O3-d, with internal and external transition elements and by the increase of the Y quantity in the sample. The used elements were: Mn, Fe and Pr. 
In the case of tungstates based compounds, the optimization of electrochemical properties and stability was achieved by following different strategies:
(a) Synthesis and development of compounds based on the system Ln5.5WO11.25-d for Ln: La, Er, Eu and Nd obtaining a pure phase at temperatures lower than 1000 ºC with crystal size in the nanometer range. Nanometer range materials allow to reduce production cost and to avoid the possible evaporation of some elements that can produce the degradation of the materials.
(b) Selection and study of the compounds Nd5.5WO11.25-d and La5.5WO11.25-d and later optimization by partial substitution of both, lanthanide and W. Lanthanides used for the A position substitution were: La, Ce, Pr, Nd, Sm, Eu, Gd, Tb and Yb and Mo, Re and U were used in the B position.
An exhaustive study of the crystalline structure as a function of the sintering temperature and the effect of the dopants was performed for all described compounds. Due to the multi-species transport (oxygen ion, protons, electrons and electron holes) that these compounds present depending on temperature, pO2 and la pH2O, electrochemical properties were analysed systematically by total conductivity measurements in different atmospheres and temperatures in order to discern the predominant charge carrier species. The incorporation of H+/D+ and H2O/D2O in the structure was also studied. For the most promising materials, hydrogen permeation measurements were performed and the influence of the different factors was studied: (a) hydrogen concentration in the feed, (b) humidification degree and (c) operation temperature. Finally, it was evaluated the stability of these compounds in CO2 and H2S containing atmospheres. All the analysed compounds were stable in the studied conditions. 
Nd5.5WO11.25-d and La5.5WO11.25-d hydrogen flow was improved by partial substitution of Ln and W. In fact, by partial substitution of La5.5WO11.25-d compound, a hydrogen flow of 0.095 mL•min-1•cm-2 was reached at 700 ºC, value five times higher than the obtained with the undoped compound and one of the highest reported nowadays in bulk membranes.