Researchers from the Plasma Physics Group talk about how 2022’s biggest fusion breakthrough affects their research at Imperial College London, and how their simulations may one day help scientists achieve commercial fusion energy.
By Aidan Crilly and Brian Appelbe
Nuclear fusion dominated headlines around the world last week, but our plasma physics work at Imperial College London has been a source of excitement for scientists interested in fundamental physics as well as those hoping for a breakthrough in fusion energy.
A fusion first
The US Department of Energy recently announced a significant breakthrough in nuclear fusion research. On 5 December 2022, an experiment using the National Ignition Facility (NIF) laser at Lawrence Livermore National Laboratory (LLNL) in California produced more energy from fusion reactions than the energy input to the experiment by lasers that initiated them. This is referred to as ‘scientific energy gain’, and this is the first time that it has been achieved in any laboratory fusion experiment.
The experiment at LLNL is based on an approach to fusion called ‘inertial confinement fusion’ (ICF). The fusion fuel comprises of the deuterium and tritium isotopes of hydrogen. The inner nucleus of hydrogen is usually composed of only one proton, but deuterium has one additional neutron and tritium has two. These hydrogen isotopes were contained in a small spherical fuel pellet, approximately 1 mm in radius. In a few billionths of a second, the NIF laser delivered 2.05 MJ of energy – approximately the amount of energy consumed by your kettle if you boil it five times – and compressed the deuterium-tritium fuel down to a sphere with a radius similar to that of a human hair.
This compression caused the fuel to heat up to temperatures significantly greater than those found in the centre of the Sun. These high temperatures ignited a wave of fusion reactions where the hydrogen isotopes merged together, which released 3.15 MJ of energy in the form of neutrons – superfast particles that were discharged from the reaction at velocities over 5000 km per second (about 20 per cent the speed of light). This resulted in a scientific energy gain factor of approximately 1.54.
This experiment is a momentous breakthrough for ICF. It demonstrates a ‘proof of principle’ that more energy can be extracted from a fusion experiment than is inputted into it – an important step in developing a fusion energy source. However, it should be emphasised that there are still many technical challenges to be overcome to finally develop a commercially viable fusion energy source. Not least, it should be noted that the energy gain here doesn’t include the energy consumed by the laser energy, which is several hundred MJ. To name but a few of these challenges: an energy gain factor of over 100 is required; the laser would need to fire about once per second (whilst the NIF laser is designed to fire about once per day); and there are significant engineering challenges in converting energy from fast neutrons to electricity.
For scientists like us working on ICF, however, there are additional causes for excitement from this experiment. Energy gain is due to a runaway process occurring in the hot fusion fuel in which some of the energy released from fusion reactions causes further heating of the fuel, resulting in more reactions. We refer to the fusion fuel in this state as a ‘burning plasma’. The success of ICF experiments at LLNL mean that we can now study burning plasmas in the laboratory and develop an understanding of the fundamental physical processes occurring in this exotic state, which we believe will aid efforts to increase the energy gain factor in future ICF experiments.
Describing the extreme conditions in burning plasmas draws on many fields of physics, such as hydrodynamics, plasma, atomic and nuclear physics. At Imperial, we build state of the art simulation codes to try and capture this physics and describe what we observe in real-life experiments. These simulation codes can then be used to study not just burning plasmas but also fundamental physics in other fields, such as astrophysics and space weather, which contain systems that follow similar rules of hydrodynamics and thermodynamics. For us, the interdisciplinary nature of our research makes it particularly exciting – when we learn something new about burning plasmas in ICF we immediately ask if it is relevant to stars.
Our studies of burning plasmas have recently focused on the measurement and analysis of the energies of the fast neutrons emitted from ICF experiments. The spectrum of neutron energies gives us information on the behaviour of the deuterium and tritium ions just prior to them fusing. We have collaborated with scientists from LLNL to identify the curious behaviour of these ions in recent ICF experiments leading up to the breakthrough experiment.
Far from equilibrium
In physics, the concept of a distribution is used to describe the behaviour of systems such gases and plasmas (including the extremely hot, dense, burning plasma). These systems are assumed to be composed of individual particles moving around at high velocity and interacting randomly. The distribution defines how many particles can be found at any given velocity.
A fundamental law of nature describes a concept known as ‘entropy’, which can be simplistically understood as a measure of disorder. For example, a broken glass shattered into hundreds of pieces would have higher entropy than a glass sitting on a countertop.
Systems of gases or plasma that have relaxed to a steady-state of equilibrium always have maximum entropy because they’re just composed of free-moving particles with random interactions. The velocities of those particles is described by the Maxwell-Boltzmann distribution. Therefore, we expect the ion velocities in our burning plasma to have a distribution that is Maxwell-Boltzmann, or very close to it.
However, our analysis of the neutron energy spectra indicates the ions are far from the Maxwell-Boltzmann distribution. This is surprising, since our standard theories suggest that the burning plasma should have sufficient time to relax to equilibrium – the burning plasma only exists for one-tenth of one-billionth of a second (100 picoseconds), but it should relax to equilibrium in approximately 0.01 picoseconds.
This discrepancy may seem like a somewhat obscure theoretical point but many of the computer models used to design and analyse ICF experiments are based on the assumption that ions obey a Maxwell-Boltzmann distribution. Therefore, explaining the discrepancy is necessary to ensure our confidence in these models which will be used to design experiments aiming for a larger energy gain factor.
Discovering this discrepancy required a close collaboration between us and scientists at LLNL who measure neutron energy spectra using detectors. Our collaboration involved many Zoom calls where we ensured our theoretical models were consistent with the methods used to analyse the detector signals.
At the College, we look for the cause of the unexpected behaviour of ions by analysing the neutron energy spectra data in new ways. The recent experiment is particularly exciting for us as it produced far more neutrons than any previous ICF experiment from a much hotter burning plasma and so, analysis of the neutron energy spectrum could rule out or validate some of the hypotheses that have been developed to explain the ion behaviour.
There are strong motivations for explaining this ion behaviour. We need computer models to be as accurate as possible for designing good fusion experiments and so we need to ensure that the physical assumptions on which those models are based is sound.
In addition, if the burning plasma is in a state that is far from the Maxwell-Boltzmann distribution, then this could open up new strategies for increasing the energy output from ICF experiments. This is because the number of fusion reactions that occur is highly sensitive to the distribution of ion velocities.
ICF research happens on a large scale. NIF is the world’s largest laser. Thousands of scientists, technicians and engineers at LLNL and around the world contribute to ICF experiments at NIF. For us, a major benefit of working at Imperial College is the institutional reputation which helps us to network and collaborate with researchers around the world. During our research, we even had the opportunity to travel to the NIF to see our the very reactions we simulate here at the College.
These experiments are designed and analysed using some of the world’s fastest computers (like the ones we have in the College). Being a part of this endeavour is thrilling and recent experiments have shown that we can continue to learn about fundamental physical processes while taking (small, but significant!) steps towards a fusion energy source.