Ion Diffusion and Exchange for the design of Advanced Energy MATerials – IDEA-MAT

Thus, three families of compounds have been studied: 1) Ca3Co4O9+d derivatives, initially studied at UCCS because of possible mixed ionic electronic conductivity due to their crystalline structure, 2) double perovskite REBaCoO5+d with RE = Gd, Nd, for which the possibility of proton conduction was proposed but unproven and 3) materials with the Ruddlesden-Popper structure derived from La2-xSrxCoO4+d and (Ln, Sr)n+1MnnO3n+1-d (n = 1, 2), the latter being likely relevant as fuel cell anode. A description of the transport of oxygen from the macroscopic scale to the atomic scale was obtained by combining techniques such as isotopic exchange (UCCS), relaxation (SPMS), the precise analysis of the surface ( UCCS), diffraction and total neutron scattering (SCR) and modeling (SPMS / CEA). The study initially focused on ion diffusion oxide before being extended to the proton diffusion. Since isotopic exchange experiments with heavy water did not confirm the insertion of protons in these phases, the study was extended to materials for which the proton diffusion was proven, barium zirconate and barium stannate, with particular emphasis on the application of the Maximum Entropy Method (MEM) for the detection of proton diffusion path (deuterons) from neutron diffraction data and taking into account the quantum effects of protons using molecular dynamics.

This study confirmed the excellent transport properties of Ca3Co4O9+d derivatives and double perovskite REBaCoO5+d. The presence of calcium at the outermost surface of Ca3Co4O9+d implies an oxygen exchange at the Ca-O layers of the NaCl layers of the structure when the modeling shows that the oxygen vacancies are formed very preferentially in the CoO of these NaCl layers, suggesting diffusion of oxide ions through these layers. The treatment of neutron diffraction data with MEM and molecular dynamics modeling allowed the identification of elementary mechanisms of diffusion in double perovskite, in particular the fact that the oxide diffusion was limited primarily by the formation of oxygen vacancy in the cobalt plans. By modeling, it was also possible to demonstrate the possibility of water up-take in these materials under certain conditions and identify preferential sites for protons. The demonstration of proton conduction being difficult in cobaltites, the study was extended to barium zirconate and barium stannate. Analysis of neutron diffraction data collected at 300°C on the BaZr0.8Y0.2O3-d compound by MEM suggests the localization of the protons. Molecular dynamics calculations show a significant increase in the diffusion of protons when the structure is compressed, letting consider a new approach to easily increase the ionic conduction. The study of the diffusion of protons in BaSnO3 allowed for the first time to quantify the effect of the quantum nature of the proton on its transport properties. A study by neutron diffraction under hydrogen/air flow showed the reversible reduction of manganese for compositions LaSr2Mn2O7 and La1.2Sr1.8Mn2O7-d, making these materials good candidates as fuel cell anode.

Overall most of the project objectives were achieved, the characterization and modeling tools are now controlled. The study of the transport properties of lanthanum manganites could not be performed by isotope exchange and SIMS analysis because of difficulties in obtaining sufficiently dense ceramics. The first sintering tests by SPS are nevertheless promising and we do not despair be able to characterize these materials. Another difficulty has been the characterization of the proton diffusion by isotope exchange. Having failed to confirm the insertion of protons in cobaltites, we decided to consider initially the BaZr0.8Y0.2O3-d compound, for which the proton diffusion has been proved. If the isotopic exchange experiments allowed us to confirm the proton insertion in the structure, we encounter an evolution of the exchanged material at room temperature which renders impossible the treatment of the deuterium diffusion profiles. The treatment of neutron diffraction data by MEM opens new perspective for the study of anionic and protonic conductors. Similarly, the contribution of modeling in understanding the mechanisms involved must be emphasized. The methods, molecular dynamics and DFT taking into account quantum effects or not, make it possible to predict the most common defects, characterize their dynamics and thus the transport properties. In general, the proposed theoretical approaches will allow designing new materials according to the characteristics required for the application: dynamics of ion, insertion of water, types of dopants.

This work has already led to the publication of 10 articles in journals with peer review, 3 conference proceedings and was the subject of 7 invitations to international conferences including a Faraday Discussion over 15 other oral communications in international conferences.

The IDEA-MAT project ‘Ion Diffusion and Exchange for the Design of Advanced Energy MATerials’ aims at directing the research of new Mixed Ionic Electronic Conductors materials with enhanced transport properties, for use in Solid Oxide Cells to provide society with a clean energy technology. While empirical route has been the main approach for finding new materials, we propose here to combine both experimental and theoretical studies in order to identify important features to direct the design of efficient electrode materials. Whereas most of research programmes deal with Solid Oxide Cells with an emphasis on the elaboration of systems and their durability, here, we propose a fundamental approach which could considerably impact the target application and the scientific community. By combining and developing ionic diffusion techniques, neutron diffraction and total scattering with structural modelling, we aim at identifying key parameters in the oxide ion and proton diffusion in order to propose some strategy for designing new electrode materials with enhanced properties. In parallel to the development of robust tools for the determination of transport parameters in electrode materials, novel compositions of layered oxides will be explored with emphasis on proton diffusion, a topic which has been scarcely studied. The project is a new one and will rely on a unique consortium with complementary skills:- Partner 1 – UCCS (Unité de Catalyse et de Chimie du Solide, Lille, team: Materials for Energy): coordination of the project, isotope exchange, extreme surface composition analysis by Low Energy Ion Scattering, thin film preparation, in situ X-ray diffraction under controlled atmosphere; - Partner 2 – SCR (Sciences Chimiques de Rennes, team “Chimie du Solide et Matériaux”): Materials synthesis, in situ neutron diffraction under controlled atmosphere at high temperature and use of the MEM and PDF to evidence oxygen and possibly proton diffusion pathways; - Partner 3 – SPMS (Laboratoire Structures, Propriétés et Modélisation des Solides, Ecole Centrale Paris) : oxygen content, electrical conductivity relaxation, diffusion mechanisms through Molecular Dynamics; - Partner 4 - CEA (Group of Condensed Matter Physics, CEA-DAM-IdF): Ab initio modelling of oxygen/proton diffusion/exchange. Coupling of 18O/16O isotope exchange (UCCS), relaxation techniques (SPMS), accurate surface (UCCS) and bulk (SCR) structural characterization and modelling (CEA), will allow a description of the oxygen transport from the macroscopic scale to the atomic one. Since Proton Ceramic Fuel Cell also appears as a promising technology, these techniques will be extended to the characterisation of proton diffusion in PCFC cathodes. We will focus here on layered oxide materials, which have shown to be among the most promising since the double functionality of electron/ion conduction might be tuned through an adequate control of the layers composition. Namely, 3 types of compounds will be studied 1) Ca3Co4O9 evidenced as a promising SOFC cathode by UCCS, 2) Double perovskites LnBaCo2O5+? and Ruddlesden-Popper (RP) with n = 1 Re2-xSrxCoO4+? cobaltites as potential PCFC cathodes whose proton diffusion has been scarcely studied, 3) Ruddlesden-Popper manganites (Ln,Sr)n+1(Mn1-xCrx)nO3n+1-d (n = 1, 2), possibly doped with chromium, as novel SOFC anode materials.