Inside a Fusion Reactor: The Physics of Bottling a Star

The Energy Source of the Stars

Every second, the Sun converts 600 million tonnes of hydrogen into helium through nuclear fusion, releasing energy equivalent to 4 million tonnes of matter via Einstein’s E = mc². This process has powered the Sun for 4.6 billion years and will continue for another 5 billion.

On Earth, physicists and engineers are attempting to replicate this process — not with the Sun’s size and gravitational compression, but with machines that use magnetic fields and extreme temperatures to confine a fuel far hotter than the centre of any star.

The Fusion Reaction

The most accessible fusion reaction for terrestrial reactors combines two isotopes of hydrogen:

Deuterium (one proton, one neutron) + Tritium (one proton, two neutrons) → Helium-4 (two protons, two neutrons) + Neutron + 17.6 MeV

This D-T reaction has the highest cross-section (probability of occurring) at the lowest temperature of any fusion reaction — “only” about 100 million °C, roughly seven times hotter than the core of the Sun. The Sun can fuse ordinary hydrogen at lower temperatures because its immense gravitational pressure compensates with enormous density and confinement time.

At 100 million °C, all atoms are fully ionised. Electrons are stripped from nuclei, creating plasma — the fourth state of matter. The challenge is containing this plasma long enough and densely enough for a significant fraction of the fuel to fuse.

The Lawson Criterion

British physicist John Lawson derived in 1957 the minimum conditions for a fusion reactor to produce net energy. The product of three quantities must exceed a threshold:

Temperature × Density × Confinement time ≥ a critical value

For D-T fusion, the plasma must reach about 100 million °C at a density of roughly 10²⁰ particles per cubic metre, confined for several seconds. This triple product defines the goal that every fusion experiment pursues.

No material container can withstand 100 million °C — any wall would vaporise instantly. Two fundamentally different approaches have been developed to confine the plasma without touching it.

Magnetic Confinement: The Tokamak

Since plasma consists of charged particles, it responds to electromagnetic forces. A sufficiently strong magnetic field forces ions and electrons to spiral along field lines, preventing them from escaping to the walls.

The tokamak — invented in the Soviet Union in the 1950s — shapes the magnetic field into a torus (doughnut). A set of external coils creates a toroidal field (running the long way around the doughnut), while a current driven through the plasma itself creates a poloidal field (running the short way around). The combination produces helical field lines that wind around the torus, confining the plasma in a stable magnetic bottle.

Modern tokamaks use superconducting magnets cooled to 4 K (-269 °C) to produce fields of 5–13 Tesla. The plasma inside is at 100 million °C. Separated by mere metres, the coldest and hottest places in the solar system coexist inside the same machine.

JET (Joint European Torus) in the UK held the record for fusion energy production: 59 megajoules in a single pulse in 2021, with a peak power of about 11 MW. JET ceased operations in 2023 after decades of pioneering work.

ITER (International Thermonuclear Experimental Reactor), under construction in Cadarache, France, is the world’s largest tokamak. It is designed to produce 500 MW of fusion power from 50 MW of heating input — a gain factor (Q) of 10. ITER will be the first fusion device to produce net energy, though it will not generate electricity; it is a physics and engineering experiment.

The Stellarator Alternative

A stellarator confines plasma using external coils alone, without relying on a plasma current. The coils are twisted into complex, computationally optimised shapes that produce the helical field geometry needed for confinement.

Stellarators avoid a major weakness of tokamaks: disruptions — sudden, violent losses of plasma confinement caused by instabilities in the plasma current. Because stellarators have no plasma current, they are inherently more stable and can potentially operate in steady state rather than in pulses.

The Wendelstein 7-X stellarator in Germany, which began operation in 2015, is the world’s most advanced stellarator. Its coils were designed using supercomputer optimisation and manufactured to millimetre precision. Recent results have demonstrated record plasma confinement for a stellarator, suggesting the concept is viable for a power plant.

Inertial Confinement: Compressing Fuel with Lasers

The second approach to fusion abandons sustained confinement. Instead, a tiny pellet of D-T fuel (about 2 mm in diameter) is compressed to extreme density by an array of powerful lasers firing simultaneously from all directions.

The compression must be extraordinarily uniform — any asymmetry and the pellet squirts sideways rather than imploding. At peak compression, the fuel reaches densities 100 times that of lead and temperatures of hundreds of millions of degrees. Fusion occurs in a brief flash (nanoseconds) before the pellet flies apart.

The National Ignition Facility (NIF) at Lawrence Livermore National Laboratory achieved a historic milestone in December 2022: ignition. The 192 laser beams delivered 2.05 MJ of energy to a target, and the fusion reactions produced 3.15 MJ — more energy from fusion than the laser energy delivered to the target. This was the first time ignition had been achieved in a laboratory, confirming the fundamental physics of inertial confinement fusion.

However, the lasers themselves consumed about 300 MJ of electrical energy, so the overall system efficiency remains far from break-even. NIF was designed primarily for weapons physics research, not power generation.

Plasma Instabilities: The Persistent Enemy

The central difficulty of fusion is not reaching high temperatures — that is achievable with neutral beam injection, radiofrequency heating, and ohmic heating. The difficulty is keeping the plasma confined, stable, and well-behaved.

Plasmas are inherently unstable. They develop kinks, bulges, and turbulent eddies that transport heat to the walls far faster than simple collisional diffusion would predict. These instabilities are driven by pressure gradients, current distributions, and the complex interplay of magnetic fields and plasma flows.

Understanding and controlling these instabilities requires advanced plasma physics, real-time feedback systems, and sophisticated magnetic field shaping. Every major advance in fusion performance has come from better understanding and management of plasma instabilities.

Materials Under Extreme Conditions

Even with magnetic confinement, some plasma escapes and strikes the reactor walls. The divertor — the component that handles exhaust plasma — must withstand heat fluxes comparable to the surface of the Sun, bombardment by 14.1 MeV neutrons, and the erosion caused by energetic plasma particles.

No existing material can survive these conditions indefinitely. Tungsten is the current leading candidate for plasma-facing components due to its extremely high melting point (3,422 °C) and resistance to sputtering. But neutron irradiation makes tungsten brittle over time, and the development of fusion-grade materials remains a major engineering challenge.

The Private Fusion Revolution

Since approximately 2015, private companies have entered the fusion race with alternative approaches and aggressive timelines:

Commonwealth Fusion Systems (CFS) — A MIT spin-off developing compact tokamaks using high-temperature superconducting (HTS) magnets that produce stronger fields (20 T) in smaller devices. Their SPARC tokamak aims to demonstrate net energy gain.

TAE Technologies — Developing a field-reversed configuration (FRC) approach that confines plasma in a cigar-shaped magnetic structure, aiming to use a proton-boron fuel cycle that produces no neutrons.

Helion Energy — Pursuing a pulsed field-reversed configuration approach that directly converts fusion energy to electricity via magnetic compression, bypassing the need for a thermal cycle.

These companies have attracted billions in private investment, reflecting growing confidence that fusion energy may be achievable within a commercially relevant timeframe.

The Promise

Fusion fuel is virtually unlimited. Deuterium is abundant in seawater (33 grams per cubic metre). Tritium can be bred from lithium inside the reactor. A single kilogram of D-T fuel produces as much energy as 10,000 tonnes of coal, with no carbon emissions, no long-lived radioactive waste, and no risk of meltdown.

The physics works — the Sun proves it every second. The engineering is the remaining challenge: building machines that confine stellar plasma, withstand the resulting heat and radiation, and operate reliably for decades. After seventy years of research, fusion is closer to reality than it has ever been. The star in a bottle is almost within reach.