Time Resolved Momentum Microscopy on Strongly Correlated Materials

Abstract

Intriguing phenomena in strongly correlated electron systems arise from the interplay between multiple degrees of freedom. Time-resolved Momentum Microscopy (trMM) is the ideal tool for studies of such systems. It provides access to copulings between these degrees of freedom (charge, spin and lattice) by studying their dynamics upon optical excitation on the femtosecond timescale. This is crucial for the understanding of many open questions as e.g. the nature of photodoping in antiferromagnetic Mott insulators. To this end we applied trMM to study the dynamics of photodoping in a parent compound of a prototypical high-Tc superconductor: La2CuO4 (LCO). These results suggest that upon excitations in-gap states are formed, in agreement with previous indirect observations from core level spectroscopy. From a wider perspective, Multidimensional Photoemission Spectroscopy (MPES), as the generalization of photoemission spectroscopy beyond the 3 dimensions (energy and parallel momenta), can help shed light on different exotic phenomena. Using High Energy X-Ray Photoelectron Spectroscopy (HAXPES), we investigated the prototypical Kondo compound YbRh2Si2, resolving the full 3D momentum space to track the transition between a large and small Fermi surface, as a function of temperature. In this Thesis, great attention is dedicated to the description of the methods we developed to efficiently process (photo)electron resolved data streams, recorded together with a large set of parameters that identify the experimental conditions in which each electrons was detected. Among these parameters, timing jitters, photon energies, excitation laser fluence and many others can be used to correct artifacts and calibrate physical axes (i.e. convert time-of-flight (ToF) to kinetic energy). This delivers a data structure which makes it possible to exploit fluctuations and isolate outliers and artifacts, to greatly increase information density and signal-to-noise ratio (SNR) ratio of these complex experiments. Furthermore, parametric correction methods allow to track the changes applied to the raw data, enabling full reproducibility of the workflow applied.This is one of the key ingredients for data processing and data analysis tools to be future-proof and ready for the application novel Artificial Intelligence methods.