Computational Modeling of Electrical and Phonon Properties of Skutterudites and Two-Dimensional Transition Metal Dichalcogenides

摘要

This dissertation documents the computational modeling of skutterudite and two-dimensional transition metal dichalcogenide (2D TMD) materials for energy and electronic applications by analyzing the effects of materials doping and heterostructure formation on structural, energetic, electrical, phonon, and thermal properties. These topics remain largely unexplored and can accelerate materials development by providing insight on structure-properties-performance relationships.Skutterudites are commonly studied for thermoelectric applications because they are low-cost, easy to process, and offer good intrinsic transport properties. They also exist as large, open structures which can be altered through filler atoms or substitutional dopants. A density functional theory (DFT)-based investigation of dopant effects on skutterudite compounds provided insight to advance the understanding of electrical and phonon properties that experiments could not measure. This also offered a good benchmark material for developing a modeling scheme that was employed for 2D TMD materials. 2D TMD nanosheets also exhibit large variability in structure type, dimensionality, and composition and have attracted much interest for their magnetic, electronic, optoelectronic, catalytic, and thermoelectric properties. Their low dimensionality makes them promising candidates for field-effect transistor (FET) device applications and introduces quantum confinement effects and diffusive boundary scattering, potentially improving their electrical and transport properties. The exploration of composition, substitutional doping, and heterostructure effects is needed for further 2D TMD materials development and property improvement.This dissertation offers an analysis of the structure-property relationships for a wide range of properties on bulk skutterudite and 2D TMD materials. The three key outcomes of this work are: (1) a high throughput approach to compute and analyze electrical and phonon properties, (2) a screening method for investigating 2D TMD materials and highlighting preferred compositions, and (3) design principles for predicting structures and properties to guide experiments. The optimized high throughput approach encompasses: DFT-based total energy minimization calculations to investigate the geometric, energetic, and electronic structure data; Boltzmann transport theory, in combination with electronic band energies, to estimate electrical conductivity (σ), Seebeck coefficient (S), and power factor (S2σ) values; density functional perturbation theory (DFPT)-based second-order force constant calculations to determine phonon dispersion and density of states (DoS) spectra; and the atomistic Green’s function (AGF) method, using force constants as input, to compute interfacial heat flux, phonon transmission coefficients, and thermal boundary conductance (TBC). Error mitigation was handled by optimizing model parameters and validating results through comparison with literature and experimental data. Through the optimized high throughput approach, dozens of 2D TMD structures can now be analyzed within days, whereas initial optimization calculations for each structure took up to one week to compute. Overall, these materials offer great potential for materials-by-design exploration and understanding their structural, electrical, and phonon properties are essential for advancement towards commercial applications.