Nuclear Fusion Explained: How Stars Make Energy

How Fusion Works

Fusion occurs when two nuclei come close enough for the strong nuclear force to grab hold and bind them together. The challenge is that both nuclei carry positive electrical charges, which means they repel each other through the Coulomb force. To overcome this electrostatic barrier, the nuclei must approach each other at extremely high speeds, corresponding to temperatures of millions of degrees. At these temperatures, matter exists as plasma, where atoms are fully ionized and electrons move freely.

The probability of two nuclei fusing depends on their kinetic energy, their nuclear charges, and the quantum mechanical probability of tunneling through the Coulomb barrier. Even at the extreme temperatures inside stars, nuclei do not have enough energy to classically overcome the barrier. They rely on quantum tunneling, where there is a small but nonzero probability of passing through the energy barrier rather than over it. Because tunneling probability drops exponentially with barrier height, lighter nuclei (with fewer protons and therefore lower barriers) fuse much more readily than heavier ones.

The energy released in fusion comes from the difference in nuclear binding energy between the reactants and products. When two deuterium nuclei fuse to form helium-3 and a neutron, the helium-3 product is more tightly bound per nucleon than the original deuterium nuclei. This binding energy difference manifests as kinetic energy of the products, since the products weigh slightly less than the reactants. The "missing" mass has become energy according to E=mc2.

Fusion in Stars

The sun fuses approximately 600 million tonnes of hydrogen into helium every second, converting about 4.3 million tonnes of matter into energy in the process. This produces 3.8 x 10^26 watts of power, a luminosity the sun has maintained relatively steadily for 4.6 billion years. The primary mechanism in sun-mass stars is the proton-proton (pp) chain, a multi-step process that ultimately combines four protons into one helium-4 nucleus, two positrons, two neutrinos, and 26.7 MeV of energy.

The first step of the pp chain is the slowest and most improbable: two protons must fuse while simultaneously converting one proton into a neutron through the weak nuclear force, producing deuterium, a positron, and a neutrino. This step is so unlikely that a given proton at the sun's center waits an average of about 9 billion years before undergoing this reaction. Only because the sun contains an incomprehensible number of protons (roughly 10^57) does fusion proceed at a rate sufficient to power the star.

More massive stars burn hotter and access the CNO cycle (carbon-nitrogen-oxygen cycle), where carbon acts as a catalyst for fusing hydrogen into helium. Stars above about 8 solar masses eventually fuse progressively heavier elements in their cores: helium into carbon, carbon into neon, neon into oxygen, oxygen into silicon, and silicon into iron. Iron is the endpoint because it sits at the peak of the binding energy curve, meaning neither fusing iron nor splitting it releases energy. When a massive star accumulates an iron core, fusion can no longer support the star against gravity, and it collapses catastrophically into a supernova.

Fusion Reactions for Earth

The most promising fusion reaction for terrestrial energy production is deuterium-tritium (D-T) fusion. Deuterium (hydrogen-2, one proton and one neutron) reacts with tritium (hydrogen-3, one proton and two neutrons) to produce helium-4 and a fast neutron, releasing 17.6 MeV of energy. This reaction has the lowest ignition temperature of any fusion reaction (about 100 million degrees Celsius) and the highest reaction rate at achievable temperatures, making it the closest to practical realization.

Deuterium is abundant: it constitutes about 0.015% of all hydrogen atoms in seawater, meaning the world's oceans contain enough deuterium to supply fusion energy for billions of years. Tritium is radioactive with a half-life of only 12.3 years and does not exist naturally in significant quantities. For a D-T fusion reactor, tritium must be bred on-site by surrounding the reactor with lithium blankets. When the fusion neutrons strike lithium-6 or lithium-7, they produce tritium and helium, maintaining the fuel supply. Global lithium reserves are sufficient to fuel D-T fusion for thousands of years.

Alternative fusion reactions avoid the need for tritium but require significantly higher temperatures. Deuterium-deuterium (D-D) fusion produces either tritium plus a proton or helium-3 plus a neutron at roughly equal rates. Deuterium-helium-3 (D-He3) fusion produces helium-4 and a proton with no neutrons at all, making it "aneutronic" and potentially much cleaner, but it requires temperatures six times higher than D-T fusion. Proton-boron-11 (p-B11) fusion is completely aneutronic and uses abundant fuel, but requires temperatures 30 times higher than D-T, placing it well beyond current technology.

The Confinement Challenge

No physical container can hold 100-million-degree plasma. Any solid wall touching such plasma would instantly vaporize, and even before that, the contact would cool the plasma below fusion temperatures. Fusion research therefore centers on confining plasma without material walls, using either magnetic fields or inertial compression.

Magnetic confinement fusion uses powerful superconducting magnets to create a magnetic "bottle" that suspends charged plasma particles away from material walls. The most successful design is the tokamak, a doughnut-shaped (toroidal) chamber where helical magnetic field lines keep plasma circulating in endless loops. The ITER tokamak, the world's largest, uses niobium-tin superconducting magnets cooled to -269 degrees Celsius to produce fields of 11.8 Tesla, strong enough to confine plasma at 150 million degrees Celsius in a volume of 840 cubic meters.

Inertial confinement fusion takes the opposite approach: instead of holding plasma steady for long periods, it compresses tiny fuel pellets (typically 1-2 mm diameter) to extreme density using arrays of powerful lasers or ion beams. The compression happens so fast (over nanoseconds) that the fuel's own inertia holds it together long enough for significant fusion to occur before it flies apart. The National Ignition Facility (NIF) at Lawrence Livermore National Laboratory uses 192 laser beams delivering 2.05 megajoules of ultraviolet light to achieve fusion ignition, and in December 2022, NIF achieved net energy gain from fusion for the first time in history.

The Triple Product

Fusion performance is measured by the "triple product" of plasma density (n), temperature (T), and energy confinement time (tau). The Lawson criterion specifies the minimum triple product needed for a fusion plasma to produce more energy from fusion reactions than it loses through radiation and particle escape. For D-T fusion, this requires n x T x tau to exceed approximately 3 x 10^21 keV per cubic meter per second. Meeting this criterion simultaneously, achieving high enough density, temperature, and confinement time all at once, has been the central challenge of fusion research for over 70 years.

Progress has been dramatic despite the difficulty. The fusion triple product achieved in experiments has increased by a factor of roughly 10,000 since the first tokamak experiments in the 1960s. Modern tokamaks routinely achieve temperatures above 100 million degrees, densities above 10^20 particles per cubic meter, and confinement times of several seconds. ITER is designed to push these parameters into the regime where fusion heating dominates over external heating, a condition called a "burning plasma," which has never been achieved in a magnetic confinement device.

Stellarators offer an alternative magnetic confinement geometry. Unlike tokamaks, which rely on an electrical current flowing through the plasma itself to generate part of the confining magnetic field, stellarators use carefully shaped external coils to create the entire confining field. This eliminates plasma current disruptions, a potentially damaging instability that plagues tokamaks. The Wendelstein 7-X stellarator in Germany, currently the world largest, has demonstrated excellent plasma confinement and in 2025 successfully generated high-energy helium-3 ions using radio-frequency heating. The trade-off is engineering complexity: stellarator magnets must be shaped with extraordinary precision in twisted three-dimensional forms that are far more difficult to design and fabricate than a tokamak simpler geometry.

Private fusion companies have attracted over fifteen billion dollars in investment as of 2026, reflecting growing optimism that commercial fusion power may be achievable within the next two decades. Commonwealth Fusion Systems is building SPARC, a compact high-field tokamak using high-temperature superconducting magnets made from rare-earth barium copper oxide. Helion Energy pursues a pulsed field-reversed configuration approach that directly converts fusion energy to electricity. TAE Technologies uses a beam-driven field-reversed configuration targeting proton-boron-11 fusion. Each company represents a distinct approach to the confinement problem, and the diversity of strategies increases the probability that at least one will succeed.