Spatially Resolved Optical Measurements of Electron and Hole Spins in Monolayer Tungsten Diselenide

Abstract

The development of spintronics devices over the last 30 years has necessitated study into a large number of unique semiconductors. In the past 8 years, one emerging focus of study has been into two dimensional materials, with a focus on layered Van der Waals materials such as graphene and transition metal dichalcogenides (TMDs). In this dissertation, we utilize optical methods to study the spatial dependence of the spin valley polarization in the monolayer TMD tungsten diselenide (WSe2). When reduced to a single monolayer, TMDs are a direct band gap semiconductor with large spin splitting in both conduction and valence bands. These materials also display valley-dependent optical selection rules, allowing for carefully tuned laser pulses to selectively excite electron and hole spins in one direction. We use time-resolved Kerr rotation and other similar optical methods to excite spin valley polarizations in monolayer MOCVD-grown WSe2 and probe their dynamics. Our Kerr signal reveals bi-exponential decay with time constants of 100 ps and 3 ns. Measurements are repeated on multiple flakes with radii of approximately 9 to 13 micrometers, as well as more continuous sample areas with many overlapping grain boundaries. These measurements also reveal larger spin valley polarizations at the edges of individual flakes than near the centers. Both photoluminescence and reflectivity measurements are also performed on the same flakes, but do not exhibit a similar spatial dependence. In order to perform the measurements described here, a time-resolved Kerr microscopy setup was constructed in our lab. This setup is capable of performing Kerr rotation measurements while varying multiple parameters including pump-probe separation, pump-probe time delay, magnetic field, and spatial position on the sample. The pump and probe spot sizes on the sample using this experimental setup are under 3 microns FWHM, and the pulse width of the laser allows for time delay scans to be taken with 10 ps resolution. I was heavily involved in the assembly and calibration of this experimental setup; as such, its workings and calibration procedures are described in this dissertation. Efforts were also taken to create electrically contacted samples in the Lurie Nanofabrication Facility. Multiple iterations of lithography procedures were performed in order to properly align source and drain contacts across sub-10 micron flakes. Unfortunately, once these devices were fully realized in the clean room, the lift-off procedure for the photoresist caused a significant drop in signal in the samples and therefore complete realization of these devices remains a promising subject of future work.