Defects are ubiquitous in materials. Despite the deceptive name “defects”, they have far-reaching significance to energy materials beyond degradation, from improved mechanical properties in metals to enhanced energy conversion efficiency in waste heat recovery, from driving classical and quantum phase transitions to functional properties in nanomaterials. We can even say that understanding defects are like having a set of whole new dimensions for materials improvement.

However, the vast opportunities associated with defects come with a price: the extreme complexity. Even for 0D point defects there are already different types (vacancies, interstitials and substitutions), not to mention the extended defects, such as 1D line defects (dislocations and disclinations), 2D planar defects (interfaces, surfaces, grain boundaries, twin boundaries, stacking faults etc.) and 3D bulk defects (voids, precipitates, inclusions etc.). Each single type of defect can cause profound change to materials. To see the difficulty, even the resistance R=V/I in a dislocated crystal is still not possible to be computed microscopically from first principles:

In this sense, to attack the highly challenging defect problems, we free our methodologies and always seek the best solutions for a given problem:  

To provide a high-quality platform for defect studies, we grow single crystalline materials. This includes both energy materials such as thermoelectrics and materials with fundamental interests such as heavy fermions. To create a reliable source of defects, we utilize nanoscale radiation method at near field.

To measure the properties change in defective materials, we use radiation based technologies, including state-of-the-art neutron and X-ray scattering, and electron spectroscopies and microscopies. These technologies give high energy and/or spatial resolution in understanding the materials structural, dynamics and excitations with great flexibility, i.e. with a large parameter space.  In addition, we also design new spectroscopic methods to see as much as information possible aided by machine learning algorithms.

To provide an ultraclean platform to study the interplay between materials defects and functional properties, we use nano fabrication technique to coherently measure the properties of one single defect. This goes far beyond the conventional approach where defects are bulk averaged properties.

To understand the defects, the clean measurement will also be compared theories: we develop novel quantum theories for functional defects, at a microscopic, quantum mechanical level at nanoscale. This could help understand that how defects can really change the materials energy transport and conversion processes, and provide guidelines for future materials design. Our recent work includes “dislon”, a theoretical framework of quantized dislocations, which aims at understanding dislocation’s influence on materials' multiple functional properties - electrical, optical, magnetic, thermoelectric, superconducting properties -with one single unified analytical theory, with no empirical fitting parameter.