Purdue University Graduate School
Dissertation_Terry_Final.pdf (23.02 MB)
Download file


Download (23.02 MB)
posted on 2019-08-13, 20:19 authored by Yi-Tse HungYi-Tse Hung
Charge degree of freedom has been successfully manipulated in the semiconductor industry over the past few decades. The trend of doubling the number of transistors every two years in each chip was observed by Gordon Moore at 1965 and this observation was named after him, Moores law. People have kept up with the prediction fairly well till very recently when the fundamental physics limitations has reached in the conventional Si-based devices. All variety of materials and different degrees of freedom are being explored intensively to make novel device designs to overcome this challenge. In this dissertation, we will focus on two-dimensional transition metal dichalcogenides (TMDs) materials and explore not only charge but also valley and spin degrees of freedom. 2D TMDs have attracted a lot of attention for many reasons and one of them is their superior electrostatic control due to the lowering of dimensionality from 3D to 2D. Such reduction of the dimensionality besides the easiness of doping, on the other hand, makes good metal contact harder to achieve due to its inert surface comparing to the existing Si technology. To evaluate the possibility of being one of the promising candidates of post-CMOS (complementary metal oxide semiconductor) devices, the access to both electrons (conduction band) and holes (valence band) is required in order to make CMOS devices. Fermi-level pinning in these materials, however, severely limits the tunability of the Fermi level alignment between metal and semiconductor by choosing different metal work functions. In Chapter 2, we will discuss our results on making good contact by lowering the Schottky barrier height and having atomically precise doping layer control and its associated doping level where we also achieved the record high hole branch current at the bias volt- age of -1V. Besides the manipulation of charge degree of freedom, we also explored and demonstrated the unique valley degree of freedom that can be electrically generated and detected for the first time in Chapter 3. Many fascinating properties of valley physics can be analogized to spin physics, such as, zero dissipation pure spin/valley current and binary nature (spin +1/2 and -1/2, valley K and K’). Due to the unique lattice structure in TMDs, monolayer particularly, the inversion symmetry is intrinsically broken which lifts the Kramers degeneracy and leads to non-zero Berry curvature. As a result, it possesses valley Hall effect. Even more interestingly, when the transport carriers are in the valence band of monolayer TMDs, spin and valley are locked and it is called spin-locked valley Hall effect. Owing to the nature of being 2D materials, these spins’ polarization is out-of-plane unlike the conventional spin Hall effect materials, such as Pt, Ta, and W, where spins are polarized in the surface plane. This out-of-plane polarization is particularly favorable in the SOT-magnetic random access memory (SOT-MRAM) applications due to the lowering of critical switching current and consequently the reducing of power consumption. We directly observed this spin-locked valley Hall effect for the first time and we will discuss it in Chapter 4.




Degree Type

  • Doctor of Philosophy


  • Electrical and Computer Engineering

Campus location

  • West Lafayette

Advisor/Supervisor/Committee Chair

Dr. Zhihong Chen

Additional Committee Member 2

Dr. Supriyo Datta

Additional Committee Member 3

Dr. Peide Peter Ye

Additional Committee Member 4

Dr. Joerg Appenzeller