Looking into the Crystal Ball

In June the Science and Technical Advisory Committee (STAC) met with the EUROfusion Programme Management Unit to review 2018’s progress. Representing the ITER Physics Department was Marie-Line Mayoral, Deputy Head. The highlight of her updates was the predictive simulation results for a new D-T (Deuterium Tritium) campaign at JET with the  ITER-like wall. Fusion in Europe spoke with Marie-Line to learn more about modelling and the results that were shared with STAC.

Fusion in Europe (FiE): How are you able to model something as complex as plasmas?

Marie-Line Mayoral (MLM): It takes very powerful codes which we’ve been developing for years. Codes were initially built to explain what happened during an experiment. With each new experiment we learned something new which we built into the codes. They have evolved to the point where we can now model expected results ahead of an experiment.

FiE: How does this help?

MLM: Modelling helps predict what we will have to deal with. Take runaway electron beams for example. They have a lot of energy and can make holes in the machine’s wall if not properly contained. If the modelling predicts runaway electron beams we can mitigate their effects and even avoid them. This helps us avoid costly repairs and downtime. Modelling will become even more important with ITER as a nuclear facility. Its plasmas will be so powerful that predicting the plasma’s energy, the behaviour of runaway elec­trons and the power load on the wall will be essential.

FiE: Could modelling replace the need to run experiments someday?

MLM: It is not clear yet if it could replace experiments but it can help to optimise experiments. The better a code becomes the more confidently we can predict what will happen to the plasma in different machines under different conditions. We are also extrapolating data from one machine to another and most importantly to ITER.

FiE: Who writes these codes?

MLM: For the most part physicists or engineers are writing codes. My boss, Xavier Litaudon, started a unit of software engineers called the EUROfusion Theory and Advanced Simulation Coordination initiative (ETASC). They write and modify the codes purely to reduce run-time, which saves a lot of money. Before EUROfusion the labs worked independently. By pooling resources, sharing codes and optimizing run-times, member labs can model far more efficiently.

FiE: What progress was made in 2018?

MLM: In 2018 we used modelling to optimise the preparation of the second D-T campaign at JET when factoring in cost, radioactivity and the extra safety required. This was a major decision. A lot has changed since the first D-T campaign back in 1997 (DTE1). JET now has enhanced heating, real-time networks and an ITER ­like wall made of beryllium with a tungsten-clad divertor. Our diagnostics have developed a lot since then, enabling us to capture far more data. And advances made in supercomputing let us do far more with this data than ever before. Based on the model’s predictions, we have simulated the range of fusion powers that could be reached in 2020.

FiE: What did the modelling predict?

MLM: The modelling shows that running a second D-T campaign (DTE2) at the upgraded JET facility is feasible and will help us better understand the neutrons and alpha ­particles produced. These are the key learnings we want to add to our codes and extrapolate to ITER. It also predicts 10 to 17 megawatts of output power that could be sustained for up to 5 seconds producing a record of fusion energy - which is a product of power times duration. In addition to the performance, our aim is to improve our understanding of the burning plasma with a sufficient fraction of alpha­heating. To quote Xavier, “(with DTE2) ...we are aiming for stability rather than peak performance. And we will be operating in ITER-like conditions.”

FiE: How does this support EUROfusion in reaching its goals?

MLM: We have well-defined steps in our roadmap towards the realisation of fusion energy, and are currently in the step from JET towards ITER. Accordingly, our efforts focus on making ITER a success. Keep in mind that ITER’s first D-T campaign is scheduled after it completes its 10-year commissioning period running from 2025 until 2035. Running DTE2 at JET will give us key neutron and alpha-particle data in 2020 – a full 15 years before we could get it from ITER. Using this data, adding it to our codes, will significantly advance the progress made by the EUROfusion Programme during this time. It will help us to optimise ITER’s operation.

Why D-T fusion?

Based on our current knowledge and technology, D-T fusion promises to generate more gross energy – and net electricity – than it consumes. Compared with deuterium-only fusion reactions, D-T fuel releases 60 to 90 times more neutrons. Additionally, these neutrons contain four times more energy. This is why the next experimental reactor, ITER, will run on D-T fuel in 2035 after an initial phase in hydrogen, helium and deuterium. The high-energy neutrons from D-T fusion are both a blessing and a curse. They are wildcards in that their name-giving neutral charge makes them uncontainable by magnets. In future reactors they are needed to breed tritium from the lithium in the blankets and to generate heat that can be converted into electricity. While indispensable from an energy generation perspective, the neutrons will make the wall and blanket of the reactor radioactive. This is the reason why JET uses the D-T mixture in only a very limited number of experiments.

The case for a second D-T campaign

With ITER as the next step in the European Research Roadmap to the Realisation of Fusion Energy, EUROfusion is focussed on making it a success. The more that can be learned about D-T fusion ahead of ITER’s planned operational debut in 2035 the better.