Karl P. Travis


Department of Materials Science and Engineering,
Sir Robert Hadfield Building
Mappin Street
University of Sheffield
Sheffield, UK


The Molecular Dynamics (MD) simulation method, invented by Berni Alder in the mid 1950s, and underpinned by Boltzmann’s formulation of statistical mechanics, provides an essentially exact description of matter at the atomic level once a forcelaw has been specified. Smooth Particle Applied Mechanics (SPAM) - invented by Lucy, Monaghan and Gingold in the late 1970s originally to solve problems in astrophysics, also uses particles. In this case, the underlying equations being solved are those of continuum mechanics.

Both these particle-based methods are powerful computational tools which can be applied to a diverse range of problems ranging from chaos theory to planetary motion.

MD is at its most powerful when used to generate pseudo-experimental data for testing new statistical mechanical theories or discovering new constitutive laws. It can also be used to yield mechanistic information. SPAM is useful for modelling matter at the scale of interest to the engineer. When the two methods are used together, MD can provide the material properties required as input to SPAM simulations, allowing true multiscale capability.

The present-day interest in nano-engineering raises questions on the realm of applicability of continuum mechanics upon which most models are based. The standard textbook treatment of diffusion, heat and viscous flows is now being called into question based on results obtained from non-equilibrium molecular dynamics simulations of nano-confined fluids, shock waves and Joule-Thomson flows. The constitutive laws of nature require modification when the strain rates, thermal and concentration gradients vary over time and length scales comparable with molecular dimensions. This will have significant implications in fields such as nano-fluidics, for example.

Following a brief introduction to these two particle-based methods, examples will be presented of their application to engineering problems, to extract mechanistic information, and to learn new physics. These applications include: highly confined fluid flow, Joule-Thomson throttling, radiation damage in ceramic wasteforms, filtration, and fragmentation.


Biographic Details

Name: Karl Patrick Travis

Title: Reader in Computational Physics and Nuclear Waste Disposal

Uinversity of Sheffield, UK:

Phone: +44(0)114 222 5483, E-mail: k.travis@sheffield.ac.uk

Research interests: Deep borehole disposal of nuclear waste, statistical mechanics and thermodynamics.