Princeton Physicists Crack Decades-Old Fusion Reactor Mystery by Adding One Missing Variable

Physicists at the Princeton Plasma Physics Laboratory have resolved a long-standing puzzle that has frustrated fusion reactor engineers for decades: why plasma particles in tokamaks consistently strike the inner side of the exhaust system far more heavily than the outer side. The finding, published in Physical Review Letters in April 2026, identifies plasma rotation as the missing variable — a factor that had been overlooked in computational models but that exerts as large an influence on particle behavior as the electromagnetic forces previously considered.

The mystery has practical consequences that extend well beyond academic interest. Tokamak fusion reactors work by confining superheated plasma in a donut-shaped magnetic field, but the plasma must eventually exhaust heat and particles somewhere. That exhaust region, called the divertor, is where the most intense heat loads in the entire reactor occur — comparable, per unit area, to the surface of the sun. Engineers designing reactors like ITER, the international fusion project under construction in southern France, must predict exactly where the hottest material will strike the divertor walls in order to build components that can survive decades of operation.

For years, experiments consistently showed plasma particles striking the inner divertor target at higher intensity than the outer target, but state-of-the-art simulations using the SOLPS-ITER code — the industry standard for this type of modeling — could not reproduce this asymmetry. The mismatch was known and acknowledged, but its cause remained unresolved. Lead researcher Eric Emdee and his team at PPPL, working with collaborators from MIT and North Carolina State University, suspected that plasma rotation was the culprit.

When the team incorporated the measured plasma rotation speed of 88.4 kilometers per second into their SOLPS-ITER simulations, the model's predictions matched the experimental data for the first time. The rotational flow, it turned out, deflects particles asymmetrically — tilting the distribution of heat and particle loads toward the inner divertor in a way that is distinct from but comparable in magnitude to the cross-field drift effects already built into the models. The finding suggests that every prior simulation of divertor heat loads that excluded rotation was systematically miscalibrated.

The implications for fusion engineering are significant. ITER and the commercial fusion reactors being developed by private companies like Commonwealth Fusion Systems and TAE Technologies all rely on computational models to design divertor components. If rotation effects are now incorporated into those models as a standard variable, the designs for heat-exhaust systems may need to be revised before construction is finalized. The PPPL team said their next step is to verify the result in other tokamak devices and to work with reactor design teams to assess what adjustments are needed. For a field that has spent 70 years working toward practical fusion power, eliminating a persistent blind spot in its core computational tools is a step forward that could affect the engineering of every future reactor.