Electronic structure and quantum transport in controlled impurity systems

Hoon Ryu, Purdue University

Abstract

Due to a continuous device downscaling, a precise control of dopant placements has become a critical factor to determine device performances. Recent progresses in the Scanning Tunneling Microscope lithography control dopant positions within a few atomic layers and have led experimentalists to propose various prototypes of planar patterned densely phosphorous δ-doping silicon (Si:P) devices. Theoretical understanding of electronic properties in such systems based on a realistic modeling approach is critical for potential device designs. Si:P devices are studied with the atomistic tight-binding (TB) approach coupled to charge-potential self-consistent simulations. The dispersion of a 1/4 mono-layer (ML) doped Si:P doping plane is simulated and compared to the previous literatures to validate our methodology. Upon the methodological validation, dispersions of 1/4ML doped ultra-narrow nanowires (NWs) are studied to explain experimentally observed metallic properties. Predicted channel conductances agree well with measured values. Then, a single donor quantum dot device with Si:P NW leads, is modeled to confirm the experimentally realized device is indeed a single atom transistor. Predicted charging energy and gate-control over the channel ground state establish strong connections to the experimental results. Finally, the numerical practicality of the Contact Block Reduction (CBR) method in simulating electron resonance tunneling features, is examined using Si:P NWs as examples. Based on a proof of principles on small TB systems, we show the CBR method can be practical on supercomputing clusters, due a better scalability than the one observed from the Recursive Greens Function and Wavefunction algorithm.

Degree

Ph.D.

Advisors

Klimeck, Purdue University.

Subject Area

Electrical engineering|Quantum physics

Off-Campus Purdue Users:
To access this dissertation, please log in to our
proxy server
.

Share

COinS