Heating fusion fuel before squeezing it could lower the bar for ignition, Princeton theorists say

Three physicists at Princeton Plasma Physics Laboratory say the order in which a fusion plasma is prepared could change how hard ignition is to reach: heat it first, raise its density afterward. Their calculation, by Luis Delgado-Aparicio, Masayuki Ono and Jonathan Menard, is a piece of theory and has not been run on any machine. It also carries a warning, because a trace of tungsten at one part in 10,000 roughly doubles the pressure the plasma would need.

The work appeared in Physical Review Letters, volume 137, issue 11, as “Generalized Lawson-Cordey-Mills Accessibility of Fusion Ignition.” The laboratory announced the work on Oct. 5, 2026.

Heat-first versus density-first routes to ignition

The laboratory, a U.S. Department of Energy facility managed through Princeton University, announced it on Oct. 5, 2026, and a ScienceDaily release carried it to a wider audience later that week. Ignition is the point where a plasma’s own fusion reactions supply enough heat to keep it burning without constant help from outside. The standard yardstick is the Lawson criterion, an equation from the 1950s that links density, temperature and how long the plasma holds its energy. In its ideal form it assumes pure fuel and few losses, which is a generous picture of a real machine.

The PPPL team rebuilt that criterion with four real-world effects added: helium “ash” that builds up and dilutes the fuel, impurities from the vessel walls, synchrotron radiation from the electrons, and heat leaking out of the plasma at a rate that climbs with temperature. According to the laboratory’s own account, the resulting map has a mountain-pass feature called the Cordey saddle. Conventional thinking raises density first and climbs toward it. The proposal is to go around.

A Q of about 5 at the Cordey saddle, with no savings figure

In the idealized case of pure fuel, the saddle sits where fusion output is about five times the external heating power, a gain written as Q of roughly 5. The authors argue that reaching ignition by heating first and densifying afterward needs much less energy than the density-first route. Delgado-Aparicio put it this way: “Go around the peak instead,” he said, adding that the same place is reached in a much smarter way, using far less energy.

Neither the release nor the summary of the paper gives a percentage saving, so no such figure can be responsibly attached to the claim. The comparison baseline is the conventional density-first path, and the only quantities stated are the Q of about 5 at the saddle and the doubling of required pressure under tungsten contamination. “Lower the bar” is the authors’ direction of travel, supported by calculation, rather than a measured reduction.

Everything reported is computed. No experiment was run, and the release states that no existing machine reaches the temperatures where the Cordey saddle sits. Plasma temperatures must exceed those in the core of the sun, and current devices hold only seconds of fuel at a time. Nothing here amounts to an ignition milestone, a working reactor, or a result demonstrated in a facility.

The tungsten penalty and the stability problem

The same added realism that opens the shortcut also closes part of it. When tungsten makes up about one part in 10,000 of the plasma, the calculation says the pressure needed for ignition roughly doubles. That matters because more than a dozen next-generation machines have chosen tungsten walls, which tolerate heat well but shed atoms into the plasma, where heavy ions radiate energy away.

Ono framed the point as a design lesson. Fusion experiments cost a great deal of money, he said, and mistakes that could have been caught beforehand are ones to avoid. Leaving the extra effects out, he added, makes a design look fine; putting them in changes the picture. The first calculations were two-dimensional. Extending them to three dimensions, the authors found the required pressure can rise past the plasma’s stability limit, meaning the plasma could be pushed beyond what it can physically hold together. The team also notes that energy losses may help by counteracting thermal runaway, which could keep a burning plasma steadier.

Testing the heat-first path on tokamaks and stellarators

A real test needs a device that can reach Cordey-saddle temperatures, a start that heats before it densifies, and impurity levels low enough that tungsten stays below the damaging range. The analysis applies to tokamaks, the doughnut-shaped magnetic machines, and to stellarators. It does not depend on one particular facility, which is part of why experimentalists will have to decide where to try it.

Mitigations are on the table but unproven in this context. The release points to liquid lithium wall coatings to suppress tungsten contamination, following earlier PPPL work on lithium walls, and to spin-polarized fuel, which could raise the fusion reaction rate, drawing on a 2024 study of spinning fuel. The release does not evaluate how either would perform along the heat-first path.

The next step is also computational. The team plans further “digital experiments,” simulations that test whether heating first delivers the predicted advantage once more physics is included. Menard, the third author, said that more study is needed but that the team is excited by these results. Only after that would a case exist for spending machine time, and the paper itself is available through its journal record, with author background on Delgado-Aparicio’s laboratory profile.

This article was produced with the assistance of AI and reviewed by Morning Overview editors prior to publication.

More from Morning Overview