Molecular Dynamics Hard-Sphere Simulation


I won’t claim that I know anything about Molecular Dynamics, or even Physics - I scraped an A last semester in Physics Mechanics - but here I am, doing an independent study involving Sonoluminescence. I’m not even entirely clear on what that is. It’s about gas bubbles, and producing sound and light which may have applications in generating energy.

What I’m doing right now is easy enough, I suppose: simulate an n-body system of hard-sphere collisions with a bubble wall, assigning a velocity to each atom according to the Maxwell distribution and calculating collisions according to an elastic collision model. In layman’s terms, a bunch of balls bouncing around in a bigger ball with no gravity. For this project, it’s back to Dark GDK for me, for the reasons mentioned in the very first post in the blog.

My other assignment is to make an interface to complement a program that already does what I mentioned in the paragraph above. This program, however, is extremely slow and inefficient at lower numbers of atoms, “lower” being 50000 vs. “higher” being billions. Yay Big O notation! For this, I’m using SDL and Agar.

I haven’t had time to post, but I guess it doesn’t really matter; no one follows this, anyway. Oh well. Need to keep an account of things, for posterity.

MD Simulation Project Page Link