UK MAST Upgrade Sets Record Fusion Plasma Pressure

Fusion can feel like the sort of science story that always belongs to the future. That is why this update matters. According to the UK Government and the UK Atomic Energy Authority, the MAST Upgrade machine at Culham in Oxfordshire has just finished its fifth experiment campaign after running through 2025 and 2026 and producing more than 1,100 fusion plasmas. The headline result is not simply that a record was broken. Researchers say they reached the highest plasma pressure yet on the machine while also dealing with the stability problems that have long stood between fusion research and a working power station. UKAEA has presented the campaign as one of the major scientific deliveries in its 2026-2030 strategy to make fusion a deployable low-carbon energy source.

To see why that matters, it helps to start with the fuel. In a fusion machine, forms of hydrogen are heated until they become plasma, a superhot electrically charged gas. That plasma then has to be squeezed and held in place by magnetic fields at extreme temperatures and pressures. Higher plasma pressure matters because it lets scientists pack more fusion performance into the same space. In plain English, if you can keep a higher-pressure plasma steady, you are working with conditions that look more like the ones a commercial plant would need. **What this means:** this was not a record for the trophy cabinet. It was a test of whether fusion can be run in a way that is useful outside the lab.

The biggest problem facing the MAST Upgrade team had a strange name but a very practical effect. Edge Localised Modes, or ELMs, are sudden bursts at the outer edge of the plasma. When they hit, they can throw away pressure, dump up to a tenth of the plasma's stored energy in one go, and slowly wear down the inside of the machine. That is why ELMs matter so much. A power plant that keeps damaging its own walls would be expensive to repair and hard to run day after day. **Why this matters:** fusion is not only about making a reaction happen. It is about making it happen repeatedly without the machine punishing itself every time.

UKAEA says the team tackled that problem in several ways. They used Resonant Magnetic Perturbations, meaning carefully applied 3D magnetic fields, to calm the plasma edge. They also worked in Quasi-Continuous Exhaust mode and reached other stable operating states known as QH-mode and I-mode, all of which aim to keep performance high without triggering the damaging bursts scientists are trying to avoid. You do not need to memorise the names to understand the achievement. Think of these modes as different ways of keeping a fast, powerful system steady enough to be useful. What mattered here was that MAST Upgrade accessed these stable conditions under circumstances that differ from those on other fusion machines, giving researchers fresh evidence about what future plants might be able to do.

One of the most interesting advances was not about brute force at all. The team says it developed a world-first method for tracking the plasma's position by measuring visible light from deuterium in the machine's upper and lower outer divertors. Those light signals reveal tiny imbalances in real time, which lets operators spot when the plasma is drifting before it turns into a larger control problem. If you are wondering why that is exciting, here is the simple answer: future fusion plants will need far more automatic control than today's experiments. **What this means:** if scientists can use light to keep the plasma centred and stable, they are building the sort of real-time control system a power station would depend on rather than a machine that needs constant manual correction.

Heat exhaust was another major part of the campaign. The MAST Upgrade team found that adding small amounts of nitrogen around the plasma edge made the exhaust shed a large share of its power as light before that heat slammed into the machine's inner surfaces. That spreads out the heat load and lowers wear on the divertor, the part built to handle the plasma's exhaust. This matters because even clever machine design does not make the heat problem disappear. MAST Upgrade uses a Super-X divertor, built to spread intense exhaust over a larger area, yet UKAEA says a power plant would still need extra help. In these experiments, nitrogen provided some of that help. UKAEA also says this was the first detailed study of that heat-handling mix inside MAST Upgrade's tightly baffled Super-X setup on a spherical tokamak. The team also explored 'negative triangularity' plasma shapes, an approach watched closely across fusion research because it may support high-power operation without ELMs.

Taken together, these results help explain why UKAEA's James Harrison says the experiments are shaping how future fusion plants will be designed. The findings were presented at the European Physical Society Plasma Physics Conference 2026 in Edinburgh, and the data is now being shared to inform STEP, the planned UK prototype at West Burton in Nottinghamshire, as well as ITER, the international fusion project. It is worth keeping one thing in view. This does not mean homes will be powered by fusion next year. What it does mean is that some of the awkward engineering problems, keeping plasma stable, controlling its position, and surviving its heat, are being answered with much more confidence than before. MAST Upgrade is due more enhancements, including two new neutral beam injectors that will double its neutral beam heating capacity and an Electron Bernstein Wave system adding 1.6 megawatts of heating. UKAEA expects that upgrade programme to finish in 2027, with a sixth experiment campaign aimed at STEP-relevant research planned for 2028.