Development of novel modeling techniques

A need for high quality large volume materials data requires basic research into thedevelopment of novel modeling techniques. This work concerns method development with increased accuracy and efficiency, including dynamical mean-field theory (DMFT), spin- dynamics, time-dependent response theory (TDRSP), and molecular dynamics (MD).

  1. Large-scale DMFT simulation in data-driven computational materials design. The most accurate state-of-art technique to include many-body effects for strongly correlated materials beyond is the density functional theory (DFT) is its combination with the DMFT, the so-called DFT + DMFT framework. The technique has been used by us with success (see Figure). However, at present it is limited to low-throughput calculations for chemically simple systems. We will develop an implementation for large-scale simulations. This will allow us to consider significantly larger and more realistic systems than previously possible. In particular, using DFT + DMFT calculations for supercells with lattice positions from DFT MD simulations we will be able for the first time consider explicitly effects of lattice vibrations on properties of correlated materials. Therefore, we will enhance the capability of data-driven materials design beyond state-of-the-art data collected at the DFT level. At this point, Dr. Ekholm (LiU) will be supported by the MCP and will be responsible for this task.

  2. Development of the atomistic spin dynamics modeling tool. We will continue the development of a computational tool for magnetization studies, in particular finite temperature effects such as spin-lattice interaction and fluctuations. Better integration into workflow frameworks for automated calculations will be provided. Specifically, full integration of combined MD and spin dynamics in relevant computational software will be carried out first, followed by development of easy-to-use tool that from input structure predicts finite- temperature magnetic properties, e.g., critical temperature, specific heat, magnon softening, magnon heat transport, orbital momentum transfer between the spin and lattice subsystems. Prof. Delin will be responsible for this task. Salaries for permanent faculty will be covered through other means, in accordance with emerging KTH policy on funding of faculty salaries. Instead, the MCP funds will be used for project-specific funding of 2 postdocs for developing spin-lattice simulation software. Both positions will be co-funded by Delin.

Fig: Electronic topological transitions in Os discovered in our DFT+DMFT calculations [Nature 525, 226 (2015); New J. Phys. 19, 033020 (2017)