Investigating Fusion Bombardment: Making progress

Investigating Fusion Bombardment: Making progress

Hey there!

Little over halfway into the project I’ve finally gotten to grips with it. In this piece I’m going to introduce my project titled ‘Computational atomic-scale modelling of materials for fusion reactors.’

Its all comes down to energy generation. Climate change presents a growing global issue and requires a complete shift in the current energy production systems which rely on fossil fuels. Nuclear fusion was one of the ideal options to solve this problem, as it offers a way to generate carbon free power. It has been one of the ongoing international goals of science since the 1970s (Knaster et al, 2016) as a replacement for fission. The big advantage of fusion is that its major waste product is harmless helium gas. As well as this it is a controlled process without the dangerous chain reaction seen in nuclear fission. The process occurs in the sun, in which hydrogen atoms are brought together with enough force to fuse , releasing large amounts of energy (Figure 1).

Figure 1 – Simple equation for nuclear fusion using deuterium and tritium with mass numbers shown. Other fusion routes exist, for example using lithium.

A practical reactor (i.e. one which produces more energy then it consumes) has so far remained out of reach. While fusion has been performed, it is not sustainable, and as the world lacks a comparable high energy neutron source, there has been only limited work to investigate material performance under prolonged fusion (Chapman, 2021). Of course here a HPC can provide great predictive work.

Our project aims to support these efforts. The temperature and pressure conditions of such industrial reactors will be extreme and require high performance materials. Evaluation of such materials cannot currently take place as there is a lack of sufficiently high energy neutron sources (Knaster et al, 2016), limiting testing. Here computational modelling offers a way to perform these investigations in advance of real world testing, helping to speed development. Computational modelling can occur at a number of levels of scale; from ‘coarse’ representations presenting materials as continuous blocks, to full depth representations using quantum mechanical calculations to include the effects of electrons in calculations and therefore include chemical bonding. Our work takes place on the atomic level using so-called molecular dynamics. This scale allows atomic level resolution while avoiding computationally costly electronic structure calculations. It does this by representing atoms using classical physics, viewing atoms as a series of spheres which have a number of parameters controlling the forces they exert on each other, which are then calculated to give trajectories over a series of time steps. This allows large systems (Millions of atoms or micrometer scale) relevant to radiation damage to be represented.

One of the favored materials for fusion reactor components is crystalline tungsten. It has the highest melting point of any element and a high resistance to sputtering (breaking apart) due to energetic particles. Molecular dynamics will allow investigation of damage cascades under neutron bombardment; the results of atoms receiving energy from oncoming neutrons and flying out of position in the crystal disrupting and mobilizing increasing numbers of adjacent atoms. This gives cascades similar to those seen in Figure 2.

Figure 2
Figure 2 – Evolution of two cascades in tungsten at given energies with time. Sourced from, J. Knaster, A Moeslang & T. Muroga Nature Physics, 12, 424–434 (2016)

My colleague Paolo is working on thermal conductivity measurements of such systems, and its been really interesting to see his work and perspective on things. My side of things has been to generate cascades such as those shown in the header. Its been a challenging but enjoyable project. My personal experience has been one dominated by problem solving, attempting to find out how a simulation set-up has gone wrong. Though at times this is frustrating, it is rewarding to find solutions and to expand my knowledge of the HPCs I am using. Especially as I’ve finally been able to execute defect cascades to the million atom scale.

Moving forward, Paolo and I hope to further our analysis of these reactor components and defect cascades. I will focus on comparing cascades at different energies and using various potentials (determiners of tungsten interactions), to be presented at the end of the project. This will be an opportunity to really show how our work has developed and contributes to the existing research on the massive undertaking of developing nuclear fusion. Overall, I am happy to be contributing to this area of investigation and I am hopeful for the future of nuclear fusion and how it can improve the sustainability of our world.


J. Knaster, A Moeslang & T. Muroga, Nature Physics, 12, 424–434 (2016)

I. Chapman, Putting the sun in a bottle: the path to delivering sustainable fusion power | The Royal Society,, 2021.

Tagged with: , , , ,

Leave a Reply

Your email address will not be published. Required fields are marked *


This site uses Akismet to reduce spam. Learn how your comment data is processed.