Fusion's fuels are among the most abundant elements on Earth. Here is where they come from, how much we have, and which one is the bottleneck.
The most technically mature fusion reaction — and the one nearly all first-generation plants will use — fuses deuterium (D) with tritium (T). Understanding fusion fuel availability means understanding three elements: deuterium, lithium (from which tritium is bred), and tritium itself.1
Deuterium is a naturally occurring isotope of hydrogen. About 1 in every 6,420 hydrogen atoms in seawater is deuterium, which translates to roughly 33 grams per cubic meter of ocean water. Since the oceans contain about 1.4 billion cubic kilometers of water, Earth's deuterium reserves total approximately 46 trillion tonnes.2
Deuterium extraction from water is straightforward and commercially practiced today. The Girdler sulfide process and other isotope separation methods produce heavy water (D2O) at industrial scale, primarily for use in fission reactor moderators. Scaling up for fusion fuel supply would be trivial compared to the energy the fuel would produce.
Tritium does not occur naturally in useful quantities because its 12.3-year half-life means it decays away. Instead, fusion plants will breed their own tritium by surrounding the reactor with a lithium-containing blanket. When neutrons from the fusion reaction strike lithium-6 or lithium-7, they produce tritium and helium.3
Known lithium reserves are estimated at roughly 22 million tonnes (in conventional mineral deposits), with identified resources several times larger. At projected fusion energy consumption rates, mined lithium alone could sustain global energy production for thousands of years. But lithium is also dissolved in seawater at concentrations of about 0.17 parts per million — totaling some 230 billion tonnes, enough for millions of years of fusion power.4
The lithium supply chain is currently under pressure from the electric vehicle battery industry. However, fusion requires far less lithium per unit of energy than batteries do — a fusion plant consuming a few hundred kilograms of lithium per year would produce gigawatts of continuous power.
Tritium is the scarcest and most challenging of the three fuels. Global tritium inventory is estimated at only 20 to 25 kilograms, mostly produced as a byproduct in Canadian CANDU heavy-water fission reactors. Each kilogram is worth roughly $30 million.5
A commercial fusion reactor burning D-T fuel would consume approximately 55 kilograms of tritium per year per gigawatt of fusion power. Since the global supply is measured in tens of kilograms, the first fusion plants will need to breed virtually all of their own tritium from lithium. Achieving a tritium breeding ratio (TBR) greater than 1.0 — producing slightly more tritium than the reactor burns — is one of the critical unsolved engineering challenges for commercial fusion.
Future fusion systems may use deuterium-helium-3 (D-He3) or even proton-boron-11 reactions. Helium-3 is extremely rare on Earth (a few hundred kilograms exist) but may be abundant on the lunar surface, embedded in regolith by the solar wind over billions of years. These advanced-fuel cycles remain decades away but could further expand fusion's fuel horizon.
Fusion's primary fuels — deuterium and lithium — are so abundant that fuel supply will never limit the technology. The near-term bottleneck is tritium, which must be bred on-site, making tritium breeding blanket engineering one of the most important challenges on the road to commercial fusion power.