Multiscale Computational Characterization

Abstract

To understand, predict, and control matter and energy at the electronic, atomic, molecular, microstructural, and continuum levels, scientists need to investigate materials at a combination of length and time scales that are characteristic to relevant physical and chemical phenomena. As a consequence, experimental, theoretical, and computational methods must cover a wide range of space and time scales, starting with the nucleus and the electronic structure of individual or clustered atoms (Å), to nano/microstructural features, all the way to continuum properties of the sample (cm). Along the time scale, the investigation domain ranges from excitations (ps) to nucleation of new phases (ns), all the way to diffusion (minutes, hours) and aging characteristic times (months, years). The article briefly reviews the major theoretical and computational methods used in multiscale characterization of materials, including atomistic (DFT, MD, AIMD, kMC), mesoscale (PF, DD, DDD), and continuum (FEM, FDM, FVM, CALPHAD) methods. Method coupling schemes are also discussed, from sequential to loose and tight concurrent coupling and ending with hybrid coupling methods. The multiscale characterization methodology is illustrated with results of simulations phenomena in Li‐ion and UO 2 materials. The chapter ends with a discussion of challenges and opportunities in the multiscale characterization of materials.