Automated Analysis of Shubnikov-de Haas Oscillations in 2D Material

Abstract

In recent years, some remarkable discoveries, such as superconductivity and ferromagnetism, have been made in few-layer graphene heterostructures. Understanding the new physics relies on analyzing the samples’ magnetoresistance, such as studying the periodic dependence of resistance on the magnetic field and charge carrier density, known as Shubnikov-de Haas (SdH) oscillations. This research aims to automate the analysis of SdH oscillations in 2D materials, specifically focusing on graphene, to accelerate data processing while ensuring accuracy. The project employs machine learning techniques, particularly Density-Based Spatial Clustering of Applications with Noise (DBSCAN), to separate data into clusters, where each cluster corresponds to a local resistance minimum. All the local minima form a Landau fan. The position of the Landau fan is determined by a linear relation between the magnetic field and charge carrier density, which can be found by applying a linear fit to the clusters. From that, the filling factor of a particular cluster can be extracted. This helps determine essential parameters like twist angles and band structure degeneracy in the system. Overall, this project efficiently determines the filling factor of 70% of visible minima, providing sufficient information to comprehend the system's properties.