Path integral Monte Carlo simulations of solid H2 surfaces and thin 4He films on H2 substrates

Based on Richard P. Feynman's formulation of quantum mechanics, Path Integral Monte Carlo is a computational ab-initio method to calculate finite temperature equilibrium properties of quantum many-body systems. As input, only fundamental physical constants and pair-potentials are required. We carry out the first ab-initio particle simulations of three related physical systems. First, the bare H$\sb2$ substrate is simulated between 0.5 and 1.3K, because a liquid H$\sb2$ film is a candidate for a new superfluid. We find evidence of quantum exchange in surface terraces for up to 1K. Second, the melting of the H$\sb2$ surface between 3 and 15K is examined since this is the cleanest example of quantum surface melting. Third, atomically thin superfluid $\sp4$He films on H$\sb2$ surfaces are simulated, calculating binding energies per $\sp4$He atom and third sound, an important experimental probe for superfuid $\sp4$He films. For all systems we compute density profiles perpendicular and parallel to the surface and compare to experiment. We treat both H$\sb2$ molecules and $\sp4$He atoms on the same footing, as spherical particles. For simulations of bulk/vapor interfaces and surface adsorption, a realistic representation of the macroscopic surface is crucial. Therefore, we introduce an external potential to account for arbitrarily layered substrates and long-range corrections. Two algorithms for parallel computers with independent processors are introduced, one to manage concurrent simulations of entire phase-diagrams, and one to improve input/output speed for files shared by all processors.

Name thesis:degree_name

Ph.D.

Level thesis:degree_level

Dissertation

Discipline thesis:degree_discipline

Physics

Grantor

University of Illinois at Urbana-Champaign

Year dc:date

2011

Author dc:creator

• Wagner, Marcus

Contributors dc:contributor

• Ceperley, David M.

• Physics, Condensed Matter
• Engineering, Materials Science
• Computer Science

Statement dc:rights

• Copyright 1994 Wagner, Marcus

Language dc:language

eng

Identifier

AAI9512584
(UMI)AAI9512584

OAI identifier oai:identifier

oai:www.ideals.illinois.edu:2142/20316