MSE Department Seminar

The static structure-property relationship is central to materials science. What happens to the properties when the structure keeps evolving? In many cases, the mechanical and electronic properties are determined by the instantaneous lattice structure, but this may not always be true. In particular, a qualitative distinction can be made by atomic motions called “chiral phonons”, which breaks time-reversal symmetry on average, and would in principle affect all properties protected by such symmetry.

This talk will review two recent projects that cover: 1) microstructural modeling of Ni-based super alloys during additive manufacturing and 2) image-driven-machine-learning techniques for characterizing nuclear materials during a complex series of processing steps. The microstructure modeling work focuses on linking length scales from individual dendrites to macroscopic heat flows during laser processing.

Calorimetry has long been used to study chemical reactions and phase transitions in materials. The technique finds its origin in the mid-18th century when Scottish physician Joseph Black discovered the notion of latent heat and Lavoisier developed an ice calorimeter to measure the amount of heat given off during combustion of carbon or respiration of living organisms.

Polar crystals display pyroelectricity upon temperature change with two primary and secondary contributions. Primary pyroelectricity, arises directly from the anharmonicity of chemical bonds, leading, upon heating or cooling, to changes of the relative positions of the atoms without changing the overall volume of the crystal. The secondary pyroelectricity arises from the thermal expansion or contraction of the crystal, i.e. a piezoelectric contribution.

Complex-oxide materials possess a range of interesting properties and phenomena that make them candidates for next-generation devices and applications. But before these materials can be integrated into state-of-the-art devices, it is important to understand how to control and engineer their response in a deterministic manner.

Semiconductors are the most important part of the electronics for the modern society. A good semiconductor should have a right band gap, high carrier mobility in both electrons and holes, etc., but semiconductors with larger band gap, higher carrier mobility, higher thermal conductivity, better matched coefficient of thermal expansion, etc. are very much needed for the future. Boron arsenide (BAs) seems to be the ideal semiconductor we are dreaming of.

The continuing need for reduced power requirements for small electronic components, such as wireless sensor networks, has prompted renewed interest in recent years for energy harvesting technologies capable of capturing energy from ambient vibrations and heat. This presentation provides an overview of piezoelectric harvesting system along with the closely related sub-classes of pyroelectrics and ferroelectrics.

Terahertz (THz) technologies hold great promise for the development of future computing and communication systems. The ideal, energy-efficient, and miniaturized future THz devices should consist of lightweight, low-cost, and robust components with synergistic capabilities. However, a paucity of materials systems possessing both these desirable attributes and functionalities has made device realization difficult.

Spin-orbit torque (SOT) is a relativistic quantum effect that allows for energy-efficient manipulation of magnetization by electric current. Several technologies utilizing spin torques such as non-volatile memory (STT-MRAM) and hard disk drives with microwave assisted magnetic recording have recently been commercialized. Spin torque devices are also under active consideration for neuromorphic computing applications.

Over the past decade, advances in machine learning have led to the creation of new approximate density functionals. I will review this area, with an emphasis on very recent developments. How do such functionals compare to those of human design? What are their advantages and their limitations? For example, can they work for strongly correlated systems?