Tritium is extraordinarily rare. Only about 7.5 kilograms of it exist naturally on Earth at any given time, produced in tiny amounts when cosmic rays strike gases in the upper atmosphere. Including all human-made supplies, the entire global stockpile sits at roughly 20 to 25 kilograms. To put that in perspective, it would fit in a single backpack, and it’s all that exists on the planet.
Why So Little Exists Naturally
Tritium is a radioactive form of hydrogen with a half-life of about 4,500 days, or 12.3 years. That means half of any given amount decays away in just over a decade. Nature does produce fresh tritium constantly: cosmic rays slam into nitrogen and oxygen in the upper atmosphere, generating roughly 0.25 to 0.5 tritium atoms per square centimeter of Earth’s surface every second. But the decay rate nearly matches the production rate, so the total natural inventory hovers around 7.5 kilograms at any point in time.
In water, natural tritium concentrations are vanishingly small. Scientists measure it in “tritium units,” where one TU equals a single tritium atom per quintillion (10¹⁸) hydrogen atoms. Natural precipitation ranges from about 0.5 TU in tropical regions to around 60 TU in Antarctica. Those numbers are so low that detecting tritium in water requires specialized lab equipment.
The Nuclear Testing Spike
Tritium wasn’t always this scarce in the environment. Thermonuclear bomb tests in the 1950s and 1960s injected massive amounts into the atmosphere, mostly into the stratosphere. At the peak, precipitation in the Northern Hemisphere reached around 10,000 TU, roughly 20,000 times the natural tropical baseline. Even in Tokyo, monthly rainfall in March 1963 measured 1,680 TU. Since those tests stopped, the “bomb pulse” tritium has been steadily decaying. Today, environmental levels have returned close to their natural baseline.
Where the World’s Supply Comes From
Nearly all usable tritium today is a byproduct of a specific type of nuclear fission reactor. Canada’s CANDU heavy-water reactors are the world’s only significant commercial source. Each reactor produces about 0.5 kilograms per year as a waste product. As of 2024, global civilian tritium stores total roughly 25 kilograms, most of it owned by Canada.
This supply is already under pressure. About half of the CANDU reactors are due to retire this decade, and much of the existing stockpile is earmarked for the International Thermonuclear Experimental Reactor (ITER), the massive fusion research project under construction in France. Private fusion companies hoping to buy Canadian tritium are largely out of luck.
One of the Most Expensive Materials on Earth
Tritium’s scarcity makes it staggeringly expensive. It costs around $30,000 per gram, putting it in the same price range as diamonds. That price reflects both the difficulty of production and the constant loss to radioactive decay. Any tritium you store today will be half gone in 12 years, which means stockpiling it for the future is a losing game. You’re essentially paying for a material with a built-in expiration date.
Everyday Products That Use It
Despite its rarity, tritium does show up in consumer products. Self-illuminating exit signs, the kind that glow without electricity, typically contain about 25 curies of tritium gas. That sounds like a lot, but in mass terms it’s a tiny fraction of a gram (roughly 2.5 milligrams). Tritium is also used in watch dials and gun sights, where small glass tubes filled with tritium gas glow continuously for years without batteries. The amounts involved are minute, and the radiation tritium emits is so weak it can’t penetrate skin.
The Fusion Energy Problem
Tritium’s rarity matters most for the future of fusion energy. The leading fusion reactor designs burn a mixture of two hydrogen isotopes: deuterium and tritium. Deuterium is easy to get, extractable from ordinary seawater in virtually unlimited quantities. Tritium is the bottleneck. A single 1-gigawatt fusion power plant would need roughly 55 kilograms of tritium per year to operate. That’s more than double the entire global civilian supply.
Fusion engineers plan to solve this by having reactors breed their own tritium internally, using lithium blankets surrounding the reactor core that generate new tritium when struck by neutrons from the fusion reaction. But this technology hasn’t been proven at scale, and the math is tight. A reactor would need to produce slightly more tritium than it consumes just to keep running, a feat no one has demonstrated yet. If breeding ratios fall even slightly short, fusion reactors could burn through the world’s tritium supply before the technology becomes self-sustaining.
In short, tritium is one of the rarest practical materials on Earth: naturally produced in trace amounts, constantly disappearing through decay, expensive to obtain, and potentially the single biggest constraint on whether fusion energy ever reaches the power grid.