This page contains a summary of my research interests, which lie broadly within statistical mechanics, more specifically within systems with unusual collective behaviour and more exactly in the areas of glassy physics and self assembly. The three areas at the top of the page are where my current research is focussed.
As the temperature of supercooled liquids is reduced, their relaxation times increase dramatically and they become incredibly stable without developing an ordered structure. This dramatic slowing down makes studying glasses at low temperatures to determine the source of their stability extremely difficult. Recently, the swap algorithm has been used to equilibrate supercooled liquids to temperatures below the experimental glass transition temperature. If you know anything about simulating glasses, this is a remarkable achievement.
Having used the swap algorithm to rapidly thermalise a supercooled liquid, the question remains as to what you can do with it. To make any measurements involving real dynamics you cannot use the swap algorithm so this leaves you with a thermalised glass with a typical relaxation time many decades longer than you can measure. Various forms of destructive testing are ideal for testing the properties of these unique systems - we can learn a lot about glasses by watching how they fall apart.
Inspired by measurements of the kinetic stability of thin films produced by vapor deposition experiments, I have melted hard sphere glasses produced using the swap algorithm. Doing this in the correct ensemble (at constant pressure) reveals the role of a stable glass's high density (compared to the liquid it is trying to melt to) in making it so stable. I measured kinetic stabilities for these hard sphere glasses comparable to the largest kinetic stabilities measured in experiments, a world record for simulated glasses. The melting process of ultrastable glasses remains to be understood, and doing so will reveal much about the nature of glasses themselves. To explore this further, I have begun producing Lennard Jones glasses using the swap algorithm.
To better explore Lennard Jones glasses with large system sizes, I have implimented the swap algorithm as a fix in LAMMPS. This presents an interesting problem where molecular dynamics and monte carlo dynamics are hybridised, and using the fix efficiently means optimising many parameters. I hope to make the code containing the fix public in the near future, along with instructions for using it effectively.
More to come
More to come!