Fusion — real progress, uncertain timing

What has been proven, what is missing and why an experiment is not yet a power plant.

Lighting a star isn't enough to plug it in

The same physical process that powers the Sun

Nuclear fusion joins light atomic nuclei — typically hydrogen isotopes — to form a heavier nucleus, and that process can release a considerable amount of energy, exactly the same physical mechanism that powers the Sun itself and every other star. On Earth, two different experimental approaches have achieved real, verifiable progress toward reproducing that process in a controlled way: magnetic confinement, which uses extraordinarily intense magnetic fields to hold a plasma at temperatures of tens of millions of degrees without it touching the reactor walls, and inertial confinement, which compresses a small fuel capsule using very high-power lasers fired simultaneously from multiple directions. Both approaches have reached genuinely real scientific milestones in recent decades, not empty promises, but both also share the same underlying problem: producing net energy at the scale of an isolated laboratory experiment is a different, and considerably simpler, scientific achievement than producing electricity continuously and economically viably at the scale of a real power plant connected to a grid.

Everything missing between an experiment and a power plant

A commercial fusion plant would need, beyond reproducing the physical reaction itself, to solve a considerably broader set of engineering problems: producing positive net energy accounting for the entire facility's consumption, not just the isolated reaction; repeating or sustaining that process continuously over periods far longer than a single experimental shot; withstanding the constant bombardment of high-energy neutrons on the reactor's own structural materials without them degrading too quickly; continuously manufacturing and regenerating fuels scarce in nature, such as tritium; efficiently extracting the heat generated by the reaction; using that heat to drive a conventional turbine and generate electricity; keeping all these components running reliably for years; and, finally, competing on real cost and availability against other, already mature, already large-scale-deployed electricity-generation sources.

Why the same announcement can produce apparently contradictory headlines

A scientific-breakthrough announcement in fusion can use the phrase "energy gain" referring only to the net energy produced by the fuel or the plasma itself in that specific experiment, without including at all the total energy consumed by the rest of the facility — the lasers, the superconducting magnets, the cooling systems, and every auxiliary system needed to run the experiment. That difference in analysis boundary explains why headlines that sound contradictory can appear almost simultaneously: one genuinely celebrating a real scientific milestone, and another rightly pointing out that the facility as a whole still consumes far more energy than it produces. Both headlines can be correct at once, simply because they are talking about different analysis boundaries.

A promising environmental balance, still to be confirmed at commercial scale

Nuclear fusion does not directly produce carbon dioxide during the reaction itself, nor does it generate the very long-lived radioactive waste equivalent to that of a conventional nuclear fission plant, two genuine environmental advantages over other generation technologies. But the intense neutron bombardment on the reactor's structural materials does activate those materials, generating some radioactivity that will need proper management, and a commercial fusion plant will require, like any large-scale industrial facility, a construction process with its own material and energy cost, plus a final decommissioning that will also need to be planned and funded in advance. None of these considerations invalidates the technology's genuine potential, but they are worth keeping in mind when evaluating its full environmental balance, rather than simply assuming fusion is automatically a technology with no waste and no associated construction cost at all.

Four questions for a fusion news story

  1. How long did the highlighted result actually last, and can it be reliably repeated, or was it a single isolated experimental shot?
  2. Does the "net energy" the story talks about include the consumption of the entire facility, or only the energy of the fuel or plasma itself?
  3. Has real electricity production already been demonstrated, structural materials that withstand long-term neutron bombardment, and a sustainable fuel cycle, or only the isolated physical reaction?
  4. Is it a scientific experiment, an engineering prototype, or a real commercial offer with a date and cost verifiable by third parties?

Supporting scientific fusion research does not at all require counting on it to meet the nearest-term climate targets. It is perfectly consistent to celebrate the genuine scientific progress as it happens and, at the same time, deploy today's already available and mature energy efficiency, electricity grids and generation, without waiting for a technology that is not yet ready to solve the most urgent climate problem.

The numbers, with their source

Real ignition in the lab, still far from a complete electrical balance

On 5 December 2022, the National Ignition Facility at Lawrence Livermore National Laboratory achieved, for the first time in history, a fusion reaction producing more energy than it directly received from the laser: 2.05 megajoules of laser energy delivered to the fuel generated 3.15 megajoules of fusion energy, a net gain over the laser energy of roughly 54%. However, the facility's own lasers need on the order of 300 megajoules of electricity from the grid to be able to deliver those 2.05 megajoules to the fuel — the whole facility's energy balance therefore remains far from reaching breakeven, even though the fuel's own scientific result is absolutely real and historic.

Sources: Lawrence Livermore National Laboratory — Achieving Fusion Ignition; US Department of Energy — DOE National Laboratory Makes History by Achieving Fusion Ignition.

ITER seeks a tenfold energy gain, on a timeline already delayed several times

The central goal of the international ITER project, the world's largest magnetic-confinement fusion experiment, is to generate about 500 megawatts of fusion power for 400 seconds from just 50 megawatts of plasma heating, an energy gain (Q factor) of 10. The project's timeline has been revised several times since its original conception: first plasma is now expected in 2034, full magnetic-power operation is expected around 2036, and operations with the deuterium-tritium fuel mix, the full fusion reaction, are not expected until 2039 — more than a decade after the originally planned dates.

Source: Max Planck Institute for Plasma Physics — New ITER schedule.

What this table is missing

Neither of these two experimental figures predicts when, if ever, fusion will become a commercially competitive electricity source: that final step still depends on solving materials-engineering, fuel-cycle and economic-cost problems that go considerably beyond having already demonstrated the fusion reaction itself in a laboratory.

Sources