Analysis and optimization of bipolar nonvolatile random access memory cells in 6H silicon carbide

Youlong Wang, Purdue University

Abstract

This thesis documents the design optimization of the nonvolatile random access memory (NVRAM) cells in 6H silicon carbide (SiC). The process of cell optimization encompasses a broad range of topics including: characterization of electron-hole generation mechanisms in SiC pn junctions, discovery of the defect-related carrier generation, direct measurement of charge storage time of the pn junction storage cells in the temperature range from 23$\sp\circ$C to 350$\sp\circ$C, experimental and theoretical study of the bipolar transistor, and analysis of transient behaviors of the bipolar NVRAM cell by using numerical simulations. Defect generation is found to be the dominant generation mechanism at room temperature. A strongly field enhanced generation rate and a low activation energy are observed when defect generation is present. At room temperature, storage times are demonstrated to be in excess of 100 years. Such a long charge storage time is suitable for nonvolatile memory applications. In order to optimize the bipolar access transistor, the inverse gain is critically examined for both npn and pnp transistors. The npn transistor exhibits lower current gain at high currents (although higher gain is shown at low currents) due to current crowding, which is caused by high spreading resistance of the p-type base region. In contrast, the pnp transistor shows a uniform current gain at high currents. This would eliminate the requirement for a slowly rising word line waveform which presently limits the speed of the npn access transistor. In addition, the transient behaviors are analyzed by computer simulation. The analysis shows important 2-D and transient effects in the memory cell. Further cell design optimization is conducted based on this analysis.

Degree

Ph.D.

Advisors

Cooper, Purdue University.

Subject Area

Electrical engineering

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

Share

COinS