Nuclear Fusion: Tokamak and Magnetic Confinement

3D Simulations

Article

Energy & Renewables · Nuclear Fusion · ⏱ ~13 min read · Last updated: 9 July 2026

Nuclear Fusion: Tokamak and Magnetic Confinement

Inside a tokamak, a wisp of hydrogen gas becomes a plasma hotter than the core of the Sun — over 100 million degrees Celsius — and is held in place by nothing but magnetic fields, floating a few centimetres from walls it would instantly vaporise on contact. Fusion promises near-limitless, low-carbon energy, but confining a star-hot gas long enough to extract more energy than it takes to heat it remains one of the hardest engineering problems ever attempted.

Written by MySimulator Team · Reviewed by MySimulator Editorial Review

1. Why Fusion, and Why It's Hard

Fusion powers every star, including the Sun, by combining light atomic nuclei into heavier ones and releasing the mass difference as energy, following Einstein's E = mc². Unlike fission, fusion produces no long-lived high-level radioactive waste and carries no risk of runaway chain reaction — if confinement is lost for any reason, the plasma simply cools and the reaction stops within milliseconds.

The catch is that fusing nuclei requires overcoming their mutual electrostatic (Coulomb) repulsion, which only happens reliably at extreme temperatures where nuclei move fast enough to get close enough for the short-range strong nuclear force to take over. At these temperatures matter exists as a plasma — a soup of free ions and electrons — which cannot be held in any solid container. Magnetic confinement fusion (MCF) solves this using the fact that charged particles spiral around magnetic field lines rather than crossing them freely.

2. The Deuterium-Tritium Reaction

Nearly every operating and planned tokamak targets the deuterium-tritium (D-T) reaction, which has the largest fusion cross-section at the lowest achievable temperature of any practical fuel combination — a critical advantage given how hard high temperatures are to reach and sustain.

D-T fusion reaction: ²H (deuterium) + ³H (tritium) \to ⁴He (3.5 MeV) + n (14.1 MeV) Total energy released per reaction: 17.6 MeV (about 1 million times more energy per unit mass than burning coal or oil, per nucleon reacting) Fuel sourcing: Deuterium: extracted from seawater, essentially unlimited Tritium: radioactive (12.3-year half-life), scarce in nature — must be bred inside the reactor from lithium: n + ⁶Li \to ⁴He + ³H (in a surrounding "breeding blanket")

The neutron carries away 80% of the reaction's energy and, because it has no electric charge, escapes the magnetic confinement entirely — this is intentional, since that neutron flux is what a future power plant would capture in a blanket to generate heat for a steam turbine, and simultaneously breed the tritium fuel needed to sustain the reaction.

3. Toroidal and Poloidal Fields

A tokamak's magnetic geometry is its defining feature. Confining a hot plasma in a simple straight magnetic tube would let particles leak out the ends; bending that tube into a torus (doughnut shape) removes the ends, but introduces a new problem the tokamak's twisted-field design specifically solves.

Toroidal field

Generated by external coils looping around the torus the "long way" — the dominant confining field, typically several tesla, strongest on the inboard side.

Poloidal field

Generated by a large electric current driven through the plasma itself (up to millions of amps) — circles the "short way" around the torus's minor cross-section.

Combined helical field

Adding the two fields twists magnetic field lines into a helix, so any given particle's orbit samples both the top and bottom of the torus, cancelling out drift.

Why twisting matters

A purely toroidal field alone is stronger on the inboard side than the outboard side, causing grad-B and curvature drifts that would push the whole plasma outward within microseconds without the twist.

Superconducting magnets (increasingly using high-temperature superconductors in newer designs like SPARC) generate the toroidal field with far less resistive power loss than conventional copper coils, which is essential because a commercial reactor must spend as little energy as possible maintaining its own confinement.

4. The Lawson Criterion

In 1955, physicist John Lawson derived the minimum combination of conditions needed for a fusion reactor to break even energetically — a benchmark still used to compare every confinement scheme, from tokamaks to laser inertial confinement, on a common footing.

Lawson triple product: n \cdot τ_E \cdot T \geq threshold value n = plasma density (particles/m³) τ_E = energy confinement time (how long heat stays in before leaking out) T = plasma temperature (keV) For D-T fuel, practical ignition threshold: n \cdot τ_E \cdot T ≳ 3 \times 10²¹ keV\cdots/m³ Physical intuition: you can compensate for lower density with longer confinement time, or for shorter confinement time with higher density — but the PRODUCT of all three must clear the bar

No experimental reactor has yet demonstrated the full triple product simultaneously with net electrical energy gain, though individual devices have separately achieved record density, record temperature, or record confinement time. JET (Joint European Torus) in the UK produced a record 59 megajoules of fusion energy in a single 2021 pulse — a major milestone, but still consuming more energy than it released overall.

5. Plasma Instabilities and Disruptions

A tokamak plasma is a magnetohydrodynamic fluid under enormous pressure and temperature gradients, and like any such system it is prone to instabilities that can grow and destroy confinement if uncontrolled.

Kink and tearing modes: large-scale distortions of the plasma current channel that can reconnect field lines and trigger sudden confinement loss.
Edge-localised modes (ELMs): periodic bursts that expel plasma and heat toward the reactor's divertor, requiring careful engineering to avoid eroding plasma-facing components.
Disruptions: the most severe failure mode — confinement collapses within milliseconds, dumping the plasma's full thermal and magnetic energy onto the vessel walls and inducing large mechanical forces from the sudden current change.

Modern tokamaks use real-time feedback control (adjusting external coil currents and injecting pellets or gas) to actively suppress or mitigate these instabilities, and ITER's design includes dedicated disruption mitigation systems to protect its plasma-facing components from the extreme heat loads a disruption can deliver.

6. Stellarators and Other Approaches

Approach	Confinement time	Key advantage	Key challenge
Tokamak	Seconds to minutes (pulsed)	Best-studied, highest achieved triple product	Plasma current risks disruptions
Stellarator	Potentially continuous	No plasma current → no current-driven disruptions	Extremely complex 3D coil manufacturing
Inertial confinement (lasers)	Nanoseconds	Achieved ignition (NIF, 2022)	Repetition rate far too low for a power plant
Field-reversed configuration	Milliseconds (experimental)	Compact, potentially lower cost	Confinement performance still far below tokamaks

Stellarators like Germany's Wendelstein 7-X shape their entire confining field using intricately twisted external coils rather than relying on a driven plasma current, trading away tokamak-style disruptions for far greater coil engineering complexity — a trade many researchers consider worthwhile for eventual continuous, steady-state power production.

7. ITER and the Road to a Power Plant

ITER, under construction in Cadarache, France, by a consortium of 35 countries, is designed to be the first tokamak to produce significantly more fusion power than the power used to heat its plasma — a target energy gain factor of Q=10 (500 MW fusion power from 50 MW input heating).

ITER is not a power plant. It will not generate electricity for the grid; its purpose is to validate the physics and engineering of a burning, largely self-heated plasma at reactor scale, paving the way for a subsequent demonstration plant (DEMO) that would.

In parallel, private companies (Commonwealth Fusion Systems, TAE Technologies, Helion Energy, and others) are pursuing smaller, faster, and often differently-configured reactors using newer high-temperature superconducting magnets, betting that a more compact design can reach net energy gain sooner and more cheaply than the ITER-DEMO government pathway — a genuine race with several credible but unproven approaches competing simultaneously.

Frequently Asked Questions

What is a tokamak and how does it confine plasma?

A tokamak is a doughnut-shaped (toroidal) magnetic confinement device that traps a hydrogen isotope plasma heated to over 100 million degrees Celsius so it never touches the reactor walls. It uses a combination of a strong toroidal field from external coils and a poloidal field induced by a plasma current to twist the magnetic field lines into helical paths, preventing particles from simply drifting outward and hitting the vessel.

Why does the plasma need to be twisted with two magnetic fields instead of one?

A purely toroidal field alone causes particle drifts (grad-B and curvature drift) that push ions and electrons in opposite vertical directions, building up an electric field that would rapidly eject the whole plasma outward. Adding a poloidal field twists the total field into a helix, so each particle's guiding-centre orbit samples both the top and bottom of the torus, cancelling out the drift on average.

What is the Lawson criterion?

The Lawson criterion is the minimum combination of plasma density, confinement time, and temperature (the fusion triple product) required for a fusion reactor to produce more energy than it consumes. For deuterium-tritium fuel the practical target is roughly n·τ_E·T greater than about 3×10^21 keV·s/m³, and no experimental reactor has yet sustained this simultaneously with net electrical energy gain.

Why is deuterium-tritium the preferred fusion fuel?

The D-T reaction has by far the largest fusion cross-section at the lowest achievable temperatures of any fusion fuel combination, making it the easiest reaction to ignite with current technology, even though tritium is radioactive, rare, and must be bred inside the reactor from lithium in a surrounding blanket.

How hot does tokamak plasma actually get?

Core plasma temperatures in modern tokamaks reach 100-150 million degrees Celsius, roughly six to ten times hotter than the core of the Sun, because the much lower particle density in a tokamak compared to the Sun's gravitationally confined core requires a correspondingly higher temperature to achieve a useful fusion reaction rate.

What is ITER and when will it produce net energy?

ITER is a large international tokamak under construction in France, designed to produce 500 MW of fusion power from 50 MW of input heating power (Q=10), demonstrating sustained burning plasma rather than net electricity to the grid. First deuterium-tritium plasma operations are planned for the 2030s, with the goal of validating physics and engineering for a subsequent demonstration power plant.

What are plasma instabilities and disruptions?

Plasma instabilities are collective oscillation modes (kink modes, tearing modes, edge-localised modes) that can grow and distort the confined plasma; a disruption is a sudden, uncontrolled loss of confinement that dumps the plasma's stored thermal and magnetic energy onto the reactor walls within milliseconds, potentially damaging components and inducing large forces on the vessel.

How does a stellarator differ from a tokamak?

A stellarator generates its entire twisted confining field using complex external coil shapes alone, without needing a driven plasma current, which avoids current-driven disruptions and allows continuous steady-state operation, at the cost of far more complex coil engineering and historically lower confinement performance than tokamaks.

What is the difference between fusion ignition and net energy gain?

Ignition means the fusion reaction becomes self-sustaining from the heat of its own alpha particles without any external heating input; net energy gain (breakeven) more loosely means the fusion power produced exceeds the power put in, which can be measured at the plasma level or, more strictly, at the whole-facility level including all auxiliary systems.

Could magnetic confinement fusion realistically power the grid this century?

Most roadmaps, including ITER's, target a demonstration fusion power plant delivering electricity to a grid sometime in the 2040s-2050s, contingent on ITER successfully validating sustained burning plasma physics and on parallel progress in tritium breeding, materials that survive neutron bombardment, and superconducting magnet technology.