Return to search

Temporal and spatial modeling of analog memristors

As silicon meets its performance limits, new materials and methods for advancing computing and electronics as a whole are being intensely researched, as described in Chapter 1. Memristors are a fusion of these two research areas, with new materials being pursued concurrently to development of novel architectures to take advantage of these new devices. A background of memristors and an overview of different memristive developments in the field are reviewed in Chapter 2.

Chapter 3 delves into the physical mechanisms of analog memristors. To investigate and understand the operation of analog memristors, a finite element method model has been developed.

More specifically, the devices simulated include a simple memristor simulation where the lithium ions (dopants) are confined to the device, but allowed to move in response to a voltage applied across the device. To model a more physical memristor, charge carrier mobility dependence on dopant levels was added to the device, resulting in a simulated device that operates similarly to the first simulation. Thereafter, the effect of varying geometries was modeled, and it was determined that both the speed and the resistance change of the device were improved by increasing the ratio of the top and bottom metal contact lengths in a restrictive flow geometry. Finally, the effect of dopant removal was investigated. It was determined that if the greatest change in resistance is required, then the removal of dopants is the optimal operating regime for an analog memristor.

Through a greater understanding of analog memristors developed by the simulation described herein, researchers will be able to better harness their power and implement them in bio-inspired systems and architectures.

Identiferoai:union.ndltd.org:GATECH/oai:smartech.gatech.edu:1853/41215
Date08 July 2011
CreatorsGreenlee, Jordan
PublisherGeorgia Institute of Technology
Source SetsGeorgia Tech Electronic Thesis and Dissertation Archive
Detected LanguageEnglish
TypeThesis

Page generated in 0.0102 seconds