Personal tools
You are here: Home Science & Projects Deisa Extreme Computing Initiative Projects 2006 - 2007 First-principles statistical mechanics for molecular switches at surfaces

First-principles statistical mechanics for molecular switches at surfaces

Project MolSwitch
Scientific Discipline Solid State Physics, Materials Science
Principal Investigator(s) Karsten Reuter
Institution(s)
  • Fritz-Haber-Institute, Max Planck Society, Germany

Abstract

This DEISA initiative aims at the switching properties of organic molecules adsorbed at metal surfaces, which could be exploited in a future “molecular electronics” as a fundamental building block for storage and logic. Employed will be a modern multi-scale modelling approach, which suitably combines first-principles electronic structure calculations with concepts from thermodynamics and statistical mechanics. First-principles density-functional theory (DFT) calculations will be employed for an accurate atomic-scale characterization of the molecule-substrate system, while the transition from a single adsorbed molecule to a molecular ensemble or network is treated by means of statistical Monte Carlo methods. Already individual DFT calculations at the involved system sizes (up to 250 transition metal atoms) form a grand computational challenge by themselves, but are furthermore required for a large number of adsorption geometries and systematically varied coverages to parameterize the statistical models. Based on the data provided in the DEISA run we will achieve an entirely first-principles, i.e. parameter-free, description of the molecular switch at the surface, from the microscopic implications in the electronic and geometric structure to mesoscopic order and disorder phenomena within a network of adsorbed switches. The proposed quantitative multi-scale calculations offer therefore the unique potential to obtain unprecedented insight and understanding into the molecular switching function, induced e.g. by coverage or temperature changes or by mechanical manipulation.

Document Actions