Parameterization And Relevant Application Of Structural Distortions In Perovskites And Their Derivatives

A perovskite or its structural derivative composed of coordination polyhedra can transit into many phase structures which have different orientations in optics,electrics,and magnetism,etc.The orientations and their combinations generate some excellent properties such as multiferroicity,ionic conductivity,and catalytic behavior,but equally these properties are only possessed by a very few structure-specified phases.The common synthetic methods for obtaining these phases can be classified as: 1)The manipulations based on outside physical conditions at the macroscopic level,such as phase transitions induced by changes in composition,temperature,pressure,etc.2)The manipulations based on polyhedral geometry and topology at the microscopic level,such as phase constructions according to the tolerance-factor adjustment by doping and substrate proximity effects on thin-film structures.As a result,we find that the common mechanism of all phase manipulations is making crystals transit into the structure-specified phases by inducing polyhedral distortions.Thus,the core issue guiding the synthesis of desired phases is which distortion should be manipulated and what state the distortion should be manipulated to.To solve the issue,our work includes the following three parts:1.Proving that the phase structure is completely formed by the coupling of polyhedral distortions.We construct normalized structures by introducing all possible couplings of dominant distortions into a 2 × 2 × 2 cubic supercell and then compare them with variously shaped primitive/conventional cells known in the database.The consistency in structure comparison demonstrates that polyhedral distortions are the only cause for the phase and property variations.This confirms the feasibility of constructing a strict quantitative/functional relation between distortion,phase,and property.And this also confirms the feasibility of identifying phase structures(solely)accordingly to distortions rather than traditional structural information on lattice parameters and atomic/ion positions.2.Proposing a process of parameterizing polyhedral distortions for measuring distortion manipulations.Based on the theory of structural evolution from aristotype to hettotype,polyhedral distortions are divided into the polyhedral motions and characteristic changes of the coordination bonding between B and X.They are parameterized using the corresponding mathematical models and polyhedral geometry,respectively.The obtained parameters can be used to measure distortion manipulations.In addition,their set can be used to make a full account of structural characteristics of crystal models.The parameters contain the information on positional relations and interactions of distortions,and thus can provide a more accurate measurement standard for polyhedral scale-based researches,compared to other structural descriptors.3.Proposing a universal paradigm for distortion parameter applications.Our proposed paradigm can effectively associate the parameter values of the distortion types,amplitudes,and positional relations with the structural characteristics and property levels of the phases by searching for the corresponding relations among phases/properties,atomic/ionic motions,and polyhedral distortions.The strict quantitative/functional relations obtained from this can be used to explain experimental phenomena related to phase evolutions,and to guide the design and synthesis of high-performance materials.The target of the paradigm is to simplify and translate traditional material development logic route into a mathematical problem solving process,which will inspire researchers with different backgrounds to participate,especially mathematicians and computer scientists,and thus significantly reduce the relevant threshold.At the same time,with the help of mathematics and computer related technologies,the increase in the possibility of solving functions also means a decrease in the corresponding research and development difficulty.We also see good prospects for combining the paradigm with AI technology in the future.