Skip to main content
Fusion EnergyTokamaks· 7 min read· in Science

Physicists Solve 40-Year Fusion Puzzle, Predicting the Power Needed to Ignite Tokamak Reactors

Researchers have developed a unified mathematical theory that calculates the exact heating power required for fusion plasmas to enter a high-confinement state. The breakthrough replaces decades of empirical guesswork with a physics-based model for future reactors.

By Ishani Patel

Share this story

Listen

For over four decades, physicists have known that pumping enough heat into a fusion reactor causes the plasma to suddenly lock in its energy, but they could not mathematically predict the exact trigger point. Now, researchers at the Swiss Plasma Center at EPFL have finally solved the 40-year puzzle.[1][2]

Published in volume 137 of the journal Physical Review Letters in November 2026, their unified theory calculates the precise heating power required to push a tokamak reactor into this high-confinement state. The breakthrough replaces decades of empirical guesswork with a rigorous, physics-based model that will directly inform future reactor designs.[1][4]

The phenomenon at the center of the mystery is known as the L-H transition, marking the shift from a low-confinement mode to a high-confinement mode. It was first discovered by accident in 1982 during experiments at the ASDEX tokamak in Germany, fundamentally altering the trajectory of fusion research.[2][6]

When operators increased the heating power beyond a certain threshold, the chaotic turbulence at the plasma's edge abruptly plummeted. The plasma began retaining heat much more effectively, a crucial requirement for sustaining the extreme temperatures needed to fuse atomic nuclei and generate net positive energy.[1][2]

The accidental discovery of high confinement

Since that 1982 discovery, the H-mode has become the standard operating regime for all major tokamak reactors. However, because the underlying fluid dynamics were too complex to model from first principles, reactor designers had to rely on empirical scaling laws to estimate the power required to trigger the transition.[4][6]

These empirical laws were essentially statistical regressions, extrapolating the heating requirements of future, massive reactors based on the performance of smaller, existing machines. While useful, this approach left a dangerous margin of uncertainty for multi-billion-dollar projects that operate at extremes of magnetic field strength and plasma density.[6]

The EPFL model predicts that ITER has a wide margin for achieving H-mode, while the compact SPARC reactor faces much tighter engineering tolerances.

To crack the mystery, the EPFL team, led by doctoral assistant Brenno De Lucca, turned to advanced computer simulations. They used a global flux-driven code known as GBS to solve the drift-reduced Braginskii equations, modeling the behavior of plasma in a diverted lower-single-null geometry as heating power increased.[1][6]

Starting from a state of quasi-steady L-mode turbulence, the researchers raised the core heating by a factor of about 1.4 while keeping the density source constant. The simulated power was split equally between ions and electrons, allowing the team to observe the exact physical mechanisms that trigger the confinement shift.[4][6]

Electromagnetic effects proved essential in these simulations, because identical power ramps conducted at zero beta produced no confinement transition. The authors identified two distinct mechanisms at play: the linear stabilization of drift waves through an enhanced electron adiabatic response, and the nonlinear generation of shear flow.[6]

Simulating the invisible plasma barrier

The simulations revealed that electromagnetic drift-wave turbulence spontaneously generates a sheared flow of particles at the plasma's edge, known as an E×B flow. This flow acts like an invisible barrier, suppressing the turbulent eddies that normally carry heat away from the reactor's core and steepening the edge pressure profile.[4][6]

The researchers translated these simulated physical processes into a comprehensive set of analytical equations. When they tested their mathematical model against experimental measurements from existing tokamaks around the world, the predictions aligned perfectly with the real-world data, validating the theoretical framework across multiple different machine configurations.[1][2]

"The theory explains why the required power increases with density in one regime but decreases in another, producing a density at which the heating requirement is lowest," the EPFL team noted in a statement. This density minimum is a well-documented experimental quirk that previous models had struggled to explain.[1][5]

Combining the transition criteria with a mixing-length estimate of turbulent heat flux, the team derived closed-form expressions for the threshold power. In the high-density branch, the required power scales mathematically with the minor and major radii, the magnetic field strength, and the plasma density.[4][6]

Crucially, every exponent in the new scaling law comes directly from the theoretical derivation, rather than from statistical curve-fitting. Experiments are now only required to fix the order-one constants, drastically reducing the uncertainty when scaling the math up to the size of a commercial power plant.[6]

The new theory successfully explains the density minimum, a well-documented phenomenon where the heating power required to reach high confinement drops before rising again.

Deriving the mathematical scaling laws

The new framework also explains the toroidal-field asymmetry of the transition, detailing why the required power nearly doubles when the magnetic field is reversed at high densities. The simulations revealed that this asymmetry arises from time-reversal symmetry breaking caused by particle collisions.[4][6]

Previous time-independent models had missed this collisionality factor entirely. By accurately capturing how the radial electric-field well deepens during the transition, the EPFL equations provide the first unified analytic prediction of the power threshold, its asymmetry, and the density minimum.[4][6]

The breakthrough arrives at a critical moment for the fusion industry, which is currently building the next generation of massive experimental reactors. Applying their equations to ITER, the massive international fusion project under construction in France, the EPFL team calculated a highly reassuring power threshold.[1][5]

The model predicts that ITER will require a minimum of 43 megawatts of heating power to trigger the L-H transition and achieve high confinement. This figure sits comfortably below the facility's planned 73-megawatt heating capacity, providing a wide margin of error for the flagship international project.[1][2]

However, the math paints a much tighter picture for the commercial sector, specifically for SPARC. Developed by the United States company Commonwealth Fusion Systems, SPARC is a compact, high-field tokamak designed to demonstrate net energy gain and pave the way for commercial fusion power.[1][6]

Predicting thresholds for ITER and SPARC

SPARC is expected to sit at the absolute extremes of magnetic field strength and plasma density covered by existing devices. Because empirical scaling laws become highly unreliable at these extremes, having a first-principles mathematical model is essential for predicting the reactor's behavior before it is switched on.[6]

Illustration: Researchers used the global flux-driven code GBS to simulate the behavior of plasma turbulence as core heating was increased.

For the SPARC reactor, the EPFL equations predict a minimum heating requirement of 26 megawatts to reach H-mode. This theoretical threshold sits just above the facility's planned 25-megawatt heating capacity, highlighting the incredibly tight engineering margins involved in compact reactor design.[1][2]

While the 26-megawatt prediction carries an error margin of roughly 7 megawatts based on the fit of the scaling constants, the calculation leaves commercial developers with considerably less room for uncertainty. It underscores the importance of precise theoretical models as private companies race to finalize their reactor blueprints.[1][6]

The ability to predict these thresholds mathematically, rather than empirically, fundamentally changes how engineers will approach future reactor designs. It allows them to optimize the magnetic field configurations and plasma densities to minimize the heating power required, improving the overall efficiency of the power plant.[1][2]

As the fusion industry transitions from purely scientific research to commercial power generation, reducing the energy input required to sustain a reaction is paramount. Every megawatt saved on heating the plasma is an additional megawatt of net electricity that can be delivered to the grid, dramatically improving the commercial viability of the technology.[2]

A calculated certainty for commercial fusion

The EPFL study stands as a landmark achievement in plasma physics, successfully translating the chaotic behavior of superheated gas into a reliable predictive tool. By replacing 40 years of guesswork with rigorous mathematics, the researchers have provided a vital foundation for the entire fusion energy sector.[1][4]

The EPFL study stands as a landmark achievement in plasma physics, successfully translating the chaotic behavior of superheated gas into a reliable predictive tool.

As engineers and physicists worldwide continue their pursuit of limitless, clean energy, they now have a physics-based criterion to guide their work. The exact power needed to ignite the stars on Earth is no longer a mystery, but a calculated certainty.[1][2]

Key points

  1. Researchers at EPFL have developed a unified mathematical theory predicting the heating power required for fusion plasmas to enter a high-confinement state.
  2. The breakthrough solves a 40-year-old mystery surrounding the L-H transition, where edge turbulence suddenly drops and energy retention improves.
  3. The new model predicts the ITER reactor will need at least 43 megawatts of heating power to achieve this state, well within its 73-megawatt capacity.
  4. For the compact SPARC reactor, the predicted 26-megawatt threshold sits just above its planned 25-megawatt capacity, highlighting tight commercial engineering margins.

What we don’t know

  • Whether the theoretical predictions will perfectly match the physical performance of SPARC when it begins operations, given it sits at the extremes of magnetic field strength.
  • How additional variables, such as specific wall materials or impurities in the plasma, might subtly alter the power threshold in real-world commercial reactors.

How we got here

  1. 1982

    Researchers at the ASDEX tokamak in Germany accidentally discover the L-H transition, observing a sudden drop in plasma turbulence at higher heating powers.

  2. 1997

    Construction begins on the National Ignition Facility, highlighting the parallel pursuit of inertial confinement fusion alongside magnetic tokamaks.

  3. 2026

    EPFL researchers publish the first unified analytical theory predicting the power threshold for the L-H transition in Physical Review Letters.

  4. 2027

    Commonwealth Fusion Systems plans to achieve first plasma in the SPARC reactor, providing a real-world test for the new theoretical predictions.

  5. 2034

    The ITER experimental reactor in France is scheduled to begin research operations, utilizing its 73-megawatt heating capacity.

Theoretical Plasma Physicists 40%Commercial Fusion Developers 35%International Reactor Operators 25%
Theoretical Plasma Physicists
Emphasize the importance of deriving scaling laws from first principles rather than statistical regression.
Commercial Fusion Developers
Focus on the tight engineering margins and the financial necessity of minimizing heating power.
International Reactor Operators
Emphasize the importance of reliable scaling laws for massive, multi-billion-dollar experimental facilities operating at unprecedented scales.

Perspectives this story doesn't cover

  • Grid operators planning for the integration of future fusion power plants

Sources

Source coverage

6 outlets

3 viewpoints surfaced

Theoretical Plasma Physicists 40%Commercial Fusion Developers 35%International Reactor Operators 25%
  1. [1]EPFLTheoretical Plasma Physicists

    Solving a 40-year puzzle in fusion energy

    Read on EPFL →
  2. [2]GizmodoCommercial Fusion Developers

    Physicists Finally Solve a 40-Year-Old Mystery About Fusion Plasma

    Read on Gizmodo →
  3. [3]Phys.orgCommercial Fusion Developers

    Theory explains how fusion plasmas enter a state of improved energy confinement, solving a 40-year puzzle

    Read on Phys.org →
  4. [4]Physical Review LettersTheoretical Plasma Physicists

    Theory for the Power Threshold of the L-H Transition in Tokamaks

    Read on Physical Review Letters →
  5. [5]Mirage NewsCommercial Fusion Developers

    Theory explains how fusion plasmas enter a state of improved energy confinement, solving a 40-year puzzle

    Read on Mirage News →
  6. [6]Fusion FutureInternational Reactor Operators

    EPFL theory predicts the power a tokamak needs to enter H-mode

    Read on Fusion Future →
Listen

Comments

Stay informed

Every angle. Every day.

Get Science stories with full source coverage and perspective breakdowns, free every day.