Profile

I use computer programs to simulate interactions between neutrons and different materials.  If you fire neutrons into a freight container you can look at what comes out and identify the location and make up of everything inside.

The type of program I use most is called a Monte Carlo simulation.  Monte Carlos (named after the gambling city) are used to explore problems where huge numbers of interactions happen in sequence all with a different probability.  Imagine wanting to know the most probable outcome of rolling 100 dice where the number of sides on each dice is dependant upon the results of all the preceding rolls and you get an idea of how complicated the problem is.

I use Monte Carlos to understand how neutrons travel through material and what they do along the way.  When a neutron enters a block of material the average distance it travels before interacting, and how it interacts, depends on the material and the energy of the neutron.  In the sort of physics I’m working on we give the probability of an interaction in units called “barns”.

Fun fact:  The “barn” became a unit when physicists working on nuclear bombs tried to work out how big a Uranium nucleus is.  They got an answer of 10^-24 cm^2, to put that into perspective a human hair has an area of about 10^-5 cm^2 and a bacterium is about 10^-8 cm^2.  In fact the nucleus is so small that about 99.9% of an atom is empty space, if you took that space out you could fit the entire human race into a sugar cube.  Even though a Uranium nucleus is so small in Nuclear Physics it’s huge, so when they first measured the cross section they said it’s as big as a barn!

After doing some more work they discovered that the size of the Uranium nucleus seemed to depend on the energy, or speed, of the neutron, it might seem bigger to a faster neutron and smaller to a slower neutron!  We now use barns as an indication of the apparent size of the target for the reaction we’re looking at.

In what I’m doing the number of barns doesn’t just depend on the energy of the neutron and the target but also what’s going to happen after the interaction, the cross section might be x-barns for the neutron to just bounce off, y-barns for it to bounce off but excite the nucleus and z-barns for it to be captured by the nucleus.  When that interaction has happened it affects what happens next, new particles can be created and lost, they can scatter around inside the target and eventually escape and for every neutron that goes in a different chain of events will happen.  The number of reactions that can happen, and the order in which they happen affects the end result and there are such a huge number of options, each dependant on those preceding them that it would be impossible to do without a computer.

It’s important to study things on the computer first because for a lot of things the real life version might be expensive and/or dangerous, especially if you get it wrong.  By looking at things on the computer I can work out what will happen for a variety of options so when it comes to the real thing I only need to do the ones which are most likely to work.

There are 3 programs I do most of my work in:

Geant4
MCNPX
EASY-II

Geant4 is a Monte-Carlo code (a type of program for simulating particle interactions) written in the programming language C++.  Because it’s written in C++ it’s possible to decide exactly how you want the program to run and change what it does.  But because there’s so much you can do it’s REALLY easy to have it go horribly horribly wrong and sometimes the reason is not at all obvious, it’s these times I yell…

Most of the time spent on Geant4 is writing new code for some simulations I’m working on.  I want to look at what happens when I particle called a deuteron (a nucleus of deuterium – heavy hydrogen) breaks up.  The neutron and the proton, the particles that make up a deuteron, can be ripped apart really easily so I want to be able to simulate the result to see what affect it will have on a system.

MCNPX works in a similar way to Geant4 but has some big differences.  I use MCNPX to look at how neutrons behave when they pass through different materials.  A lot of my day is spent setting up MCNPX simulations and then analysing the results, sometimes the results are good, other times I have to yell at the computer some more, though more often than not I have to yell at me because it’s my fault it didn’t work 🙁

EASY-II is rather badly named.  It’s not easy at all, but it’s really really useful.  So useful that despite the large amount of yelling that came about when learning to use it I persevered and without it I would’ve been unable to do some really interesting work.  I use EASY to look at what happens when particles enter a material and interact with the nuclei making it radioactive and how often that happens.

When I’m not using computer programs like the ones above I’m generally reading and writing.  It’s very important when writing a paper that you know what other people have done, partly so you don’t repeat what’s already known and partly so you can make sure you understand all the foundations for your own work.