Nuclear Fusion: How Stars Burn & Why It's Hard
The Sun fuses 600 million tonnes of hydrogen into helium every second, releasing energy via E = mc². Humans have sought to replicate this clean, nearly limitless energy source since the 1950s. The challenge: sustaining a plasma hotter than the Sun's core in a machine that won't melt.
1. The Fusion Reaction
The most practical fusion reaction for near-term energy production fuses deuterium (²H) and tritium (³H):
Total energy per reaction: 17.6 MeV
The helium nucleus (alpha particle) stays inside the plasma and heats it. The high-energy neutron escapes — it passes through magnetic fields but deposits its 14.1 MeV into a surrounding blanket as heat, which drives steam turbines to generate electricity.
The mass deficit (Δm) between reactants and products converts to energy: ΔE = Δm × c². For D-T fusion, Δm ≈ 0.0188 u, giving 17.6 MeV per reaction — about 10 million times more energy per unit mass than burning coal.
2. The Coulomb Barrier
Both nuclei are positively charged. To fuse, they must overcome or tunnel through the repulsive electrostatic (Coulomb) barrier. The barrier height for D-T at nuclear range (~1 fm) is:
At 100 million °C (10 keV), thermal nuclei have only ~10 keV — far below the 1.4 MeV barrier. Fusion happens because of quantum tunnelling: the wavefunction of the nucleus extends through the barrier, giving a small but nonzero probability of spontaneous penetration.
The D-T reaction has the highest cross-section (reaction probability) at the lowest energy among practical fusion reactions, peaking at ~65 mb at ~100 keV ion kinetic energy. This is why D-T is the preferred mix for first-generation fusion reactors.
3. Lawson Criterion
For a fusion plasma to produce net energy, the product of plasma density (n), temperature (T), and energy confinement time (τ_E) must exceed a threshold — the Lawson criterion:
This means a fusion reactor can reach ignition via three strategies:
- Very high density (inertial confinement) — compress fuel to 1000× liquid density, hold together for ~10 ns
- Long confinement time (magnetic confinement) — hold a moderate-density plasma for seconds
- Extremely high temperature — increases reaction cross-section
4. Magnetic Confinement & Tokamaks
A tokamak (Russian: тороидальная камера с магнитными катушками) uses powerful superconducting magnets to confine plasma in a toroidal (donut-shaped) vessel. The plasma never touches the walls — it floats magnetically suspended at 100–200 million °C.
Two sets of magnetic fields: toroidal (along the donut) and poloidal (around the cross-section). Their combination creates helical field lines that stabilise the plasma via the Lorentz force. Charged particles spiral along field lines and cannot easily escape.
ITER (International Thermonuclear Experimental Reactor) in France is the world's largest tokamak, under construction since 2010. Its 847-tonne central solenoid creates a magnetic field of 13 Tesla. Target: Q = 10 (produce 10× the input heating power as fusion energy). First plasma planned for 2025–2026; D-T experiments around 2035.
5. Inertial Confinement (NIF)
The National Ignition Facility (NIF) at Lawrence Livermore uses 192 laser beams delivering 2.15 MJ to compress a millimetre-scale capsule containing D-T fuel. The capsule implodes to 100 billion atmospheres and ~100 million °C in under 10 nanoseconds.
In December 2022, NIF achieved fusion ignition for the first time in history: 3.15 MJ of fusion energy from 2.05 MJ of laser energy (Q > 1, ignoring the ~300 MJ taken from the grid to power the lasers). In 2023, follow-up shots repeatedly exceeded ignition, with outputs up to 5.2 MJ.
Challenges for ICF power plants: the laser-to-target energy conversion efficiency is only ~1%, requiring shots every few seconds with extreme precision, and capsule targets cost ~$100 each — needing dramatic cost reduction for commercial viability.
6. Key Milestones
7. Remaining Challenges
Materials: 14.1 MeV neutrons from D-T fusion damage reactor materials over years, causing embrittlement and activation. No material currently certified for 40-year neutron exposure at full power. Tungsten and advanced oxide-dispersion steels are candidates.
Tritium breeding: The global tritium inventory is only ~25 kg. Fusion power plants must breed their own in lithium blankets and achieve a tritium breeding ratio (TBR) > 1.05 to sustain operations.
Plasma stability: Tokamak plasmas experience instabilities (ELMs — Edge-Localised Modes, disruptions) that can dump energy onto the vessel wall. Active suppression using magnetic coils and pellet injection is under development.
Economics: Fusion plants are expected to cost several billion euros. Levelised cost of electricity must compete with solar and wind (now <$50/MWh). Private fusion companies are pursuing smaller, cheaper machines using high-temperature superconductors.