Correlating the strain distribution and plasticity in a layered chiral magnet using symbolic regression
ORAL
Abstract
Quasi‐two‐dimensional (quasi‐2D) materials host mechanical, electrical, and magnetic properties useful for novel microelectronic devices. Of particular interest are quasi-2D chiral magnets which are a much‐desired platform for new topological spin structures and related transport phenomena. Therefore, mapping and understanding the mechanical strain inherent or induced in quasi‐2D materials accurately is critical to enhancing their functionality in devices.
We focus on the distribution of strain and sheer in an intercalated transition metal dichalcogenide (TMD) Cr1/3TaS2. Recent work on Cr1/3TaS2 demonstrated emergent spiral magnetic superstructures that can be directly affected by both the shear strain and the application of an external field as low as 100 Oe. Despite these observations, the connection between the magnetic spin and the lattice is yet to be experimentally explored.
We performed scanning x-ray nanodiffraction measurements of exfoliated Cr1/3TaS2 flakes. We collected five-dimensional dataset that was transformed into the intensity distribution in reciprocal space around the lattice vector at each probed location and used it to spatially resolve the distribution of εyz and εxz as the leading terms of the strain tensor. We then employed a symbolic regression methodology to select the outliers in terms of strain magnitude. Using a model trained on similar data, the model is used to quantify the plasticity of strain in the Cr1/3TaS2.
We focus on the distribution of strain and sheer in an intercalated transition metal dichalcogenide (TMD) Cr1/3TaS2. Recent work on Cr1/3TaS2 demonstrated emergent spiral magnetic superstructures that can be directly affected by both the shear strain and the application of an external field as low as 100 Oe. Despite these observations, the connection between the magnetic spin and the lattice is yet to be experimentally explored.
We performed scanning x-ray nanodiffraction measurements of exfoliated Cr1/3TaS2 flakes. We collected five-dimensional dataset that was transformed into the intensity distribution in reciprocal space around the lattice vector at each probed location and used it to spatially resolve the distribution of εyz and εxz as the leading terms of the strain tensor. We then employed a symbolic regression methodology to select the outliers in terms of strain magnitude. Using a model trained on similar data, the model is used to quantify the plasticity of strain in the Cr1/3TaS2.
*The work at Argonne National Laboratory was supported by the U.S. Department of Energy, Office of Science, Basic Energy Sciences, Materials Science and Engineering Division, through the Early Career Research Program.
–
Presenters
-
Justin S Wood
- Argonne National Laboratory