Craig A. J. Fisher

Oxide-ion and proton conducting electrolyte materials for clean energy applications: structural and mechanistic features

This critical review presents an overview of the various classes of oxide materials exhibiting fast oxide-ion or proton conductivity for use as solid electrolytes in clean energy applications such as solid oxide fuel cells. Emphasis is placed on the relationship between structural and mechanistic features of the crystalline materials and their ion conduction properties. After describing well-established classes such as fluorite- and perovskite-based oxides, new materials and structure-types are presented. These include a variety of molybdate, gallate, apatite silicate/germanate and niobate systems, many of which contain flexible structural networks, and exhibit different defect properties and transport mechanisms to the conventional materials. It is concluded that the rich chemistry of these important systems provides diverse possibilities for developing superior ionic conductors for use as solid electrolytes in fuel cells and related applications. In most cases, a greater atomic-level understanding of the structures, defects and conduction mechanisms is achieved through a combination of experimental and computational techniques (217 references).

Surface structures and crystal morphologies of LiFePO4: relevance to electrochemical behaviour

Advanced simulation techniques are used to provide atomic-scale insight into the surface structures and crystal morphologies of the lithium battery cathode material LiFePO4. Relaxed surface structures and energies are reported for 19 low index planes. The calculated equilibrium morphology takes on a rounded, isometric appearance, with {010}, {201}, {011}, and {100} faces prominent. Almost all of the low energy surfaces are lithium-deficient relative to the bulk lattice, requiring Li vacancies at the surface. The calculated growth morphology exhibits the {010}, {100} and {101} faces, with an elongated hexagonal prism-like shape; this morphology is more consistent with experimentally observed LiFePO4 particles. The exposure of the (010) surface in our calculated equilibrium and growth morphologies is significant since it is normal to the most facile pathway for lithium ion conduction (along the [010] channel), and hence important for the reversible insertion/de-insertion of lithium ions. SEM images of plate-like crystallites from hydrothermal synthesis are also simulated by our methods, and exhibit large (010) faces.

Tersoff Potential Parameters for Simulating Cubic Boron Carbonitrides

We have developed Tersoff potential parameters for boron in order to simulate cubic boron carbonitride systems by molecular dynamics. Combined with parameters for C and N available from the literature, our parameters are shown to reproduce the lattice parameters and bulk moduli of boron nitride and boron carbonitride (C0.33(BN)0.67) with good accuracy. By simulating several systems of formula (Cx(BN)1-x) over a wide range of carbon contents (x=0 to 1), we observed the same trends in the deviation from ideal mixing as found experimentally. We attribute this deviation to the relatively longer C-N bonds distributed randomly throughout the intermediate C content systems.

First‐Principles Calculations of Lithium‐Ion Migration at a Coherent Grain Boundary in a Cathode Material, LiCoO2

Results of theoretical calculations are reported, examining the effect of a coherent twin boundary on the electrical properties of LiCoO(2) . This study suggests that internal interfaces in LiCoO(2) strongly affect the battery voltage, battery capacity, and power density of this material, which is of particular concern if it is used in all-solid-state Li-ion batteries.

Real-time direct observation of Li in LiCoO2 cathode material

The direct observation of light elements such as Li is a challenge even for state-of-the-art electron microscopy techniques because such elements scatter electrons only weakly. Using the annular bright field scanning transmission electron microscopy imaging technique, we have simultaneously visualized columns of Li, O, and Co ions in the lithium-ion battery cathode material LiCoO2, which is one of the most important cathode materials for industrial applications. The annular bright field image exhibits a good signal-to-noise ratio and the image contrast is not reversed as the specimen thickness changes. The direct visualization of light elements in real time with this method represents an important breakthrough in characterizing the active materials in solid-state electrochemical devices.

Mixed ionic/electronic conductors Sr2Fe2O5 and Sr4Fe6O13: atomic-scale studies of defects and ion migration

Atomistic simulations of the structures and defect energetics of two mixed-conducting strontium ferrite materials, Sr2Fe2O5 and Sr4Fe6O13, are reported. Oxygen Frenkel defects are found to be the predominant intrinsic defects in both materials, with the Frenkel energy in the intergrowth structure, Sr4Fe6O13, being much less than that in Sr2Fe2O5 or other known oxide ion conductors. Formation of electronic defects under oxidizing and reducing conditions is calculated to be more favourable in the Sr4Fe6O13 intergrowth structure than in Sr2Fe2O5. Comparison of solution energies for cobalt incorporation shows Sr4Fe6O13 has a slight preference for Co2+ being located on lower coordination sites, whereas in Sr2Fe2O5, Co2+ ions located on tetrahedral sites are most favourable. Binding energy calculations suggest the possible formation of Co2+–vacancy clusters with increasing Co2+ concentration. The rapid oxide ion conductivity in Sr4Fe6−xCoxO13±δ membranes is thought to arise from a combination of factors: a low oxygen Frenkel energy (to produce an intrinsic population of mobile interstitial oxide ions and vacancies), low migration energy barriers, and ease of distortion of polyhedra.

Domain boundary structures in lanthanum lithium titanates

Perovskite-type lanthanum lithium titanate (LLTO) is attracting extensive interest because of its high intrinsic ionic conductivity. The material exhibits a complex microstructure with domains of various sizes and orientations that vary with the lithium content. Based on a systematic examination of both Li-poor and Li-rich LLTO compounds using state-of-the-art scanning transmission electron microscopy (STEM), we reveal the structures and composition of the domain boundaries (DBs) and consider their effect on Li-ion mobility and ionic conductivity, in the process positing the origin of the microstructural variations. DBs in this material are shown to consist essentially of two types: frequently occurring 90° rotation DBs and a much less common antiphase-type boundary. It is found that the 90° DBs are coherent interfaces consisting of interconnected steps that share La sites, with occupancies of La sites higher than in the domain interiors. The origin of microstructural variations in the two compounds is associated with different degrees of lattice mismatch strain at DBs in Li-poor and Li-rich materials. The lattice strain and associated O vacancies, as well as the high La occupancies, at DBs are expected to result in lower interdomain Li-ion mobility, which will have a deleterious effect on the overall ion conductivity.

Atomic-Scale Observations of (010) LiFePO4 Surfaces Before and After Chemical Delithiation

The ability to view directly the surface structures of battery materials with atomic resolution promises to dramatically improve our understanding of lithium (de)intercalation and related processes. Here we report the use of state-of-the-art scanning transmission electron microscopy techniques to probe the (010) surface of commercially important material LiFePO4 and compare the results with theoretical models. The surface structure is noticeably different depending on whether Li ions are present in the topmost surface layer or not. Li ions are also found to migrate back to surface regions from within the crystal relatively quickly after partial delithiation, demonstrating the facile nature of Li transport in the [010] direction. The results are consistent with phase transformation models involving metastable phase formation and relaxation, providing atomic-level insights into these fundamental processes.

First-Principles Calculations of Electronic Structure and Solution Energies of Mn-Doped BaTiO3

Doping with 3d transition metals, particularly Mn, is thought to play an important role in determining the reliability of dielectrics used in multi-layer ceramic capacitors (MLCCs). However, a detailed examination of the electronic structure, solution energies and compensation mechanisms of these systems is lacking. In this paper, the quantitative analysis of the substitution of Mn in perovskite-type BaTiO3 using first-principles calculations in combination with chemical thermodynamics is reported. The solution energies of dopants with vacancy and n-type and p-type charge compensations have been systematically calculated. Substitution onto the two crystallographically different cation sites in cubic BaTiO3 under four different thermodynamic conditions with different chemical potentials is also examined. Mn is found to be stable on Ti sites under all conditions examined, although its charge state varies. In the oxidizing limit, Mn substitutes for Ti as a Mn4+ ion, but in the reducing limit, Mn substitutes for Ti as a Mn2+ ion compensated by the formation of an O vacancy. Depending on the Fermi level of the system, the valence state of Mn varies from Mn4+ under p-type conditions, to Mn2+ under n-type conditions. Mn3+ is not found to be stable. These results agree well with the experimentally determined site preferences and valence states of Mn, and help to further elucidate the features of Mn-doped BaTiO3 at the atomic level.

Quantitative Evaluation of Electrochemical Potential Windows of Electrolytes for Electric Double-Layer Capacitors Using Ab Initio Calculations

The electrochemical potential windows of seven organic liquid electrolytes for electric double-layer capacitors calculated using ab initio molecular orbital theory are reported. Four types of models were used to investigate the effect of intermolecular interactions: (i) a single-ion in vacuo model, (ii) a single-ion-in-solvent model, (iii) an ion-pair in vacuo model, and (iv) an ion-pair-in-solvent model. For all the calculations, the Hartree-Fock level of theory using the 6-31 + G(d,p) basis set was used. Solute-ion interactions were treated by considering several cation-anion pair confirmations, and solute-solvent interactions were introduced by applying the isodensity polarizable continuum model. The ion-pair-in-solvent model quantitatively reproduced the experimental electrochemical potential windows with high accuracy. This demonstrates that in actual electrolytes intermolecular interactions, particularly cation―anion and solute-solvent, play an important role in determining electrochemical potential windows.