Multiscale theory of memristive materials systems

点击翻译按钮获取中文摘要

TECHNICAL SUMMARYThis award supports computational and theoretical research and education on memristive materials. Memristive materials exhibit intrinsic nonlinear conduction behavior due to the coupling between electronic and ionic conduction, the latter being mediated by defects, such as vacancies and interstitials. A subclass of materials, usually referred to as resistive-switching materials, can be switched between a high-resistance state and a low-resistance state and may be ideal for constructing resistive random access memory. More generally, memristive materials feature a hysteresis loop in their I-V characteristics. Besides memory applications, they can lead to new analog applications and even bio-inspired neuromorphic circuits, both of which utilize a gradual 'switch' in the I-V characteristics. However, the lack of an accurate description of defect dynamics and its interplay with electronic conduction has become a major obstacle for realizing the potential of the materials. The PI will focus on revealing the underlying physics and chemistry in memristive phenomena.The PI will establish a multiscale theoretical framework based on first-principles quantum-mechanical calculations using realistic atomic models and apply it to investigate the memristive switching in various materials systems. The atomic-scale investigation will include first-principles calculations of defects, their interactions, and pertinent electron energy levels. The effects of the defects on the conductivity will be probed using first-principles transport codes. Defect energetics will be incorporated in a kinetic Monte-Carlo model to obtain realistic I-V behavior of the materials. Since the focus is on the microscopic mechanisms of the memristive switching, the proposed research will identify the fundamental properties of the materials that affect their switching performance. The proposed research will also construct physics-based state equations that will be the foundation of device and circuit design. Interactions with experimental and engineering groups at universities and industry will be pursued. Extensive education outreach will be made to high schools and the broader society.NON-TECHNICAL SUMMARYThis award supports computational and theoretical research and education on memristive materials. Memristive materials conduct electricity in an interesting way that can lead to devices for memory applications, be a basis for new types of analog circuits, and even introduce neuromorphic computers that mimic the biological brain functions. In many materials, electric current flowing through the material is linearly related to the voltage across the material. In memristive materials, the relation between current and voltage is nonlinear. While conduction of electrons through metals originates from the flow of electrons alone, the non-linearity in memristive materials originates from the coupled conduction by electrons and conduction by ions in the materials, the latter being mediated by defects in the regular arrangement of atoms in the materials, such as missing atoms or vacancies. The PI aims to develop a fundamental understanding of the way electric current flows through these materials by coupled electron and ion motion and the mechanisms that underlie the memristive behavior. The PI aims to conduct theoretical and computational investigations in order to illuminate the conduction process at the microscopic level, identify the origin of the memristive behavior, and establish correlations between atomic-scale properties of materials and their implications for conductive phenomena. Broad interactions with experimentalists and engineers will be a part of the program. Extensive education outreach will make advances accessible to high schools and the community.