Personal tools

Difference between revisions of "Investigation of Metal Diffusion in Oxides for CBRAM Applications"

From iis-projects

Jump to: navigation, search
Line 41: Line 41:
[[Category:Master Thesis]]
[[Category:Master Thesis]]

Latest revision as of 16:11, 4 September 2019

Short Description

The goal of this project is to develop a framework for the determination of diffusion parameters of metal atoms in amorphous oxides. In the first stage of the project, you will become familiar with state-of-the-art simulation software involving density functional theory (DFT) and/or force-field based techniques. In parallel, you will learn how to deal with large data sets and how to process them efficiently by your own analysis scripts. In the second stage, you will createa framework which allows you to set up, run and analyze a large number of simulations efficiently on our in-house HPC clusters. In a third stage (Master’s thesis), you will apply your framework to a large number of metal/oxide pairs and analyze them regarding their applicability in CBRAM applications

The Big Picture

Conductive bridging RAM (CBRAM)is an emerging(non-)volatile memory storage technology. Its operation principle relies on the repeatable formation and disruption of a metallic filament bridging an oxide layer sandwiched between two electrodes. Therein, the diffusivity of the involved metal ions in the oxide plays a crucial role by governing the rate of growth and dissolution of the filament. The Nano-TCAD group has developed a model in COMSOL to simulate ON/OFF switching in CBRAM. So far, the model depends on a number of free input parameters, among them critical ones such as the diffusion coefficients of the involved species. Therefore, we aim at replacing these by values derived from DFT. Eventually, this refined model should not only help us to understand and confirm results obtained from experimental measurements, but could also be used to predict the behavior of new, not-yet-fabricated devices.

Status: Available

Looking for 1 Master student
Interested candidates please contact: Jan Aeschlimann or Fabian Ducry


We are looking for a candidate with a general interest in molecular modelling techniques(no former experience required). Basic knowledge in MATLAB and/or Python is advantageous.

Type of Work

Theory (20%), model development (30-50%), simulation & analysis (30-50%)


Mathieu Luisier

↑ top