Yiyang Li

My research interests are to develop new electrochemical materials for energy storage, brain-inspired computing, and information processing. Many ionic materials canhosta high density of point defects, which are mobile and act as dopants.Whereas silicon and other semiconductors usually have dopant concentrations below 10¹⁸cm^-3, ionic materials can have dopants concentrations above 10²²cm^-3, on the order of one formula unit.When integrated into an electrochemical redox system, such point defect dopants can be inserted and removed dynamically upon the application ofelectrochemicalcurrent and voltage. The high concentration ofthese point defectscombined with the ability to dynamically tune the doping concentration enable, alongside other applications, a high density of energy storage anda high density of information storage. My work focuses on three mobile ions, the lithium ion, the proton, and the oxygen ion, and their interactions in a variety ofhostmaterials. For more background, please see Y. Li &amp; W. Chueh, Electrochemical and Chemical Insertion for Energy Transformation and Switching.<i>Ann. Rev. Mater. Res.</i>48, 137-65 (2018) My first thrust is to understand Li-ion batteries by studying individual particles. Li-ion battery electrodes contain a large numberofnano- and micron-sized battery particles. My goal is to understand the electrochemical and materialspropertiesof individual battery particle building blocks, as opposed to the averagepropertyof a large number of particles, analogous to how biologists study individual cells in order to understand the behavior of an organism. My second thrust aims to develop new materials forlow-power,memory-based computing. While machine learning and artificial neural networks have made significant impacts in a number of fields,theyare extremely energy intensive due to the need to shuttle information between memory and processor. Matrix multiplications are especially energy intensive for digital processors. My research aims to develop materials for analogue non-volatile memory elements based on electrochemical redox processes, and to understand their physical mechanisms.Using analogue memory is projected to decrease systems-level energy consumption by ~ 2 orders of magnitude by conducting operations in a highly parallel manner.