Adaptive Semiconductors for Artificial Intelligence
Adapting evolutionary knowledge from nature and the corresponding plasticity into the physical world may lead to paradigm shift in designing neuromorphic computing hardware; haptic interfaces for robots and reconfigurable metamaterials. We are interested in exploring the use of correlated semiconductors as model systems to design tunable electronic states via learning and adaptation. Materials synthesis, discovery of new phase change systems and understanding electronic structure of orbitally non-degenerate crystal lattices under various environmental stimuli form an important aspect of scholarship in this program.
Soft interfaces for discovery of novel semiconductors
Soft (i.e. fluidic) interfaces offer non-thermal routes to design and discover metastable, electronically graded or transient functional materials. Understanding the thermodynamic (with condensed matter theorists) and kinetic (dynamical relaxation) aspects of these processes combined with in-situ diagnostics (with DOE collaborators) is of interest. A fundamental problem is understanding glassy dynamics that manifests at such dissimilar interfaces and may be used to create spin glass-like analogs for information processing and memory in the central nervous system.
Semiconductor Materials Synthesis in Extreme Environments
Experimental realization of synthetic neural circuits and haptic interfaces require advances in semiconductor synthesis and test structures that allow interrogation of the intrinsic properties. Two problems in this regard are of particular interest in our group: (1) experimental techniques to advance crystalline materials synthesis via extreme thermodynamic environments and (2) methods to study phase formation in-operando in a dynamic environment. A significant part of this research is conducted in close collaboration with researchers at national laboratories.
