Relativistic effects, including scalar relativistic effects and spin-orbit couplings (SOC), plays an important role in electronic structure theory, as 2/3 of the elements in the periodic table are heavy elements in which the velocities of electrons are non-negligible – compared to the speed of light. While traditional four-component methods based on directly solving the Dirac-Kohn-Sham equation provide exact treatments for full relativistic effects, the large computation heavily impedes their application, thus leaving many challenges unsolved for us in both physics and chemistry (such as SOC effect). In this talk, I will introduce a more efficient exact two-component (X2C) method that adopts a two-component algorithm but does not sacrifice the accuracy: the dimension of the Hamiltonian matrix in X2C is only half of that in four-component methods, and thus the diagonalization process is 8 times faster. I will also talk about the all electron full potential ab initio code FHI-aims that we are developing, and how we are currently implementing the X2C algorithm in it. X2C is expected to make many formerly unaffordable computations become realistic for us in the future, thus contributing for strong SOC systems such as topological insulators, perovskites, etc.

Rundong Zhao obtained his B.S. in Applied Physics from Shandong University in 2010, and his PhD in Physical Chemistry from Peking University in 2015. He then visited Beijing Computational Science Research Center from 2015-2016 and moved to Hong Kong Baptist University as a postdoctoral research fellow from 2016-2018. He is currently a postdoctoral associate at Duke University. His research interest is electronic structure theory, including both method/code development and application in bulk systems and surface science. He is the author of X2C-BAND (an DFT code for relativistic effects calculations in periodic systems) and an active developer of FHI-aims code.