Since 2000, fusion has evolved from a single global project, ITER (the International Thermonuclear Experimental Reactor), to dozens of competing designs. According to the Fusion Industry Association’s 2024 Global Fusion Industry Report there are at least 45 companies working to commercialise fusion around the world.
ITER identifies the deuterium-tritium (D-T) reaction as “the most efficient for fusion devices‚ and among the competing designs 30 projects plan to use a DT fuel
source.
Compared to the size of the reactor, the fusion fuel load is miniscule. The ITER reactor is a huge piece of engineering: its tokamak will be as heavy as three Eiffel Towers and the central vacuum vessel, with ancillary equipment and a shielding ‘blanket’, which also acts as a source of new fuel, weighs 8,000 tonnes. The reactor building stands 23m high. But at any given moment less than 1 g of fuel will be present in the superheated and compressed plasma where the fusion reactions will take place.
Nor is that fuel load fully ‘burnt’. The effective burn rate in the plasma chamber is estimated at only 1%, and fuel that is not consumed gets pumped out as part of torus plasma exhaust and recycled for reuse.
ITER says that a 1 GW fusion plant will require 250 kg of fuel per year, half of it deuterium and half of it tritium, compared with the 2.7 million tonnes of fuel that would required for a 1 GW coal plant. Other estimates of the tritium requirement for a 1 GW reactor put it at around 55 kg per year.
Of those two fuel components, deuterium is relatively easy to produce by distillation, for example from seawater, which contains around 33 g of deuterium per tonne, and it is routinely used for scientific and industrial applications. In contrast, tritium is much more difficult. It occurs only in trace quantities in nature, formed when some gases interact with cosmic rays. It is unstable, and it decays into helium 3, with a half-life of around 12.3 years. Tritium has uses outside the fusion industry, as an element of glow-in-the-dark lighting and in biomedical research. As a gas it is colourless, odourless and tasteless. It has a high coefficient of diffusion and readily diffuses both through porous substances such as rubber and through metals. In an interview with Science Business magazine in February 2024, Stephen Wheeler, executive director of fusion technology at the UK Atomic Energy Authority (UKAEA), said prices for tritium were currently $30,000-40,000 per gramme.
US Tariffs are shifting - will you react or anticipate?
Don’t let policy changes catch you off guard. Stay proactive with real-time data and expert analysis.
By GlobalDataThat could change. The current global inventory of tritium, according to ITER, is presently around 20 kg. But for a globally significant source of energy this is clearly insufficient and even at the current stage of development, for an industry with a number of projects in the pipeline the availability of tritium.
In the long term it is hoped that the reactors themselves will be the tritium source. It is planned that in most operating projects the walls around the tokamak will be lined with a so-called ‘blanket’ containing lithium, an element whose nucleus contains three protons and four neutrons. When neutrons are released in the fusion reaction and absorbed by the lithium atoms in the blanket, the lithium atom recombines into an atom of tritium and an atom of helium. The tritium can then be removed from the blanket and recycled into the plasma as fuel.
ITER notes that a future fusion plant will be required to “breed‚ all of its own tritium and further research and development will be necessary to demonstrate the feasibility of large-scale tritium production and recycling.
That is fine in a steady operating state. But what about the plants’ early days, before sufficient lithium conversions have taken place in the blanket to produce enough tritium to feed the reaction? Until then, the nascent fusion industry has to have access to enough tritium from other sources to fuel its reactors, and where necessary this will have to be serviced by a tritium transport and storage industry.
Tritium sources
In a paper, Fusion reactor start-up without an external tritium source, S Zheng, D B King, L Garzotti, E Surrey and T N Todd of the UK’s Culham Science Centre list five mechanisms to produce tritium in civil nuclear fission reactors:
- Fissioning of uranium
- Neutron capture reactions with boron and lithium added to the reactor coolant
- Neutron capture reactions with boron in control rods
- Activation of deuterium in water
- High energy neutron capture reactions within structural materials
Of these, they say that “in the civil tritium market, the principal source of tritium is fission reactors with heavy water cooling and moderation.
So one source of tritium is reactors, such as the Canadian Candu design, that use heavy water as a moderator. AECL says a Candu reactor will typically produce 130 g of tritium per year. Tritium removal facilities at Ontario Power Generation’s 3.5 GW Darlington plant, in operation since 1989, process up to 2,500 tonnes of heavy water a year.
In December 2023 the European Investment Bank signed off on a €145m loan to Romania’s Nuclearelectrica to build a tritium removal facility at the Cernavoda plant. In this context both safety and fuel security are on offer. EIB badged the loan as a radiation safety measure, saying it would “reduce the volume of radioactive wastes, prioritising the well-being of the workers and allowing for the coolant and moderator to be re-used, after tritium removal. Successful completion of the project will allow for regular maintenance, refurbishments, and eventual decommissioning to be undertaken more easily, safely and efficiently. But in addition, “the removed tritium as a rare radioactive isotope has significant strategic value as a critical material needed in the development process of another promising low-carbon power generation technology, nuclear fusion.
The Cernavoda tritium removal facility is due to enter operation at the start of 2028. It will use technology developed by the Romanian National Research and Development Institute for Cryogenic and Isotopic Technologies.
A tritium extraction facility is already in use at the Wolsung plant in Korea and in June 2023 KHNP signed an engineering, procurement and construction contract with Nuclearelectrica for the new facility.
The USA has a defence industry use for tritium. Its supply, along with other key materials, is addressed by the National Nuclear Security Administration (NNSA, part
of the Office of the US Department of Energy) in its Enterprise Blueprint, a 25-year plan, published in October 2024, to align the delivery of specialised infrastructure
with demands across the nuclear stockpile, global security and naval nuclear propulsion missions. It says, “robust supply chains, increased processing efficiency, and mitigation of single-point failures are critical for the tritium mission.
It said resilient and efficient tritium processing and the loading of gas transfer systems (GTS) are needed to sustain the defence stockpile. The Savannah River Site hosts tritium extraction, isotope separation, and storage, in addition to GTS loading and finishing capabilities. A new facility will replace key capabilities for GTS surveillance, packaging, and shipping.
The USA uses neutron capture reactions with boron in (tritium-producing burnable absorber rods or TPBARS) control rods for tritium production in Tennessee Valley Authority’s Watts Bar reactors. NNSA’s tritium capabilities rely on other entities. Pacific Northwest National Laboratory is the TPBAR design authority. Columbia Fuel Fabrication carries out TPBAR assembly and the Idaho National Laboratory is responsible for TPBAR testing.
TVA entered into an interagency agreement with NNSA in 2000 under The Economy Act to provide irradiation services to produce tritium in its light water reactors in the period to November 2035. It has been producing tritium at Watts Bar 1 since 2003 but the production could increase.
In a 2024 article for the USA Federation of American Scientists, ‘Promoting fusion energy leadership with U.S. tritium production capacity’, author Taylor Loy highlighted the USA’s “proven and scalable tritium production supply chain‚, but noted it is “largely reserved for nuclear weapons‚. Loy called for tritium production capacity to be leveraged to ensure US leadership in fusion energy. Loy highlights that one goal of NNSA is to “Demonstrate enhanced tritium production capability‚ for 2025, although that is for ‘nuclear deterrent’ purposes, and said a new programme would extend this production capacity into a longer-term effort directed toward a fusion energy future and provide a framework for repurposing tritium from decommissioned warheads.
Loy wants to double the production goal from 2.8 kg, by operating the reactors at their maximum licensed limit. Loy says, “the NNSA and DOE could leverage production capacities in excess of defence requirements to promote the deployment of FOAK reactors and support US leadership in fusion energy.
After an environmental impact statement in 2016, TVA increased irradiation of TPBARs at Watts Bar 1 under License Amendment 107 (July 2016) and at Watts Bar 2 under License Amendment 27 (May 2019). In April 2024, Watts Bar Units 1 and 2 were further authorised to increase their tritium production to 2,496 TPBARs in each unit.
TVA does not plan to follow a previously assessed option to produce tritium at its Sequoyah site, so it and NNSA plan to follow up on a different previously assessed option, which would allow for the irradiation of up to 5,000 TPBARs every 18 months at Watts Bar 1 and 2.
This way, the NNSA and TVA expect to produce up to 4 kg of tritium over the next fuel cycles (18-month cycles offset by 6 months) from the two Watts Bar reactors.
The UK tritium plan
Cernavoda’s tritium removal facility claims to be the first in Europe but the UK wants to be close behind. In May 2024 the UK Atomic Energy Authority (UKAEA) appointed AtkinsRäalis to complete detailed design of an Isotope Separation System to form part of UKAEA’s Hydrogen-3 Advanced Technology (H3AT) Facility. The contract is part of the UK’s attempt to strengthen its research into sustainable fusion delivery.
The tritium fuel cycle research facility will include a prototype-scale process plant and experimental platform (a scaled version of the design for ITER).
AtkinsRäalis has already completed the concept and detailed process design of the main H3AT facility, which is under construction at UKAEA’s site at Culham, Oxfordshire, and the concept and preliminary design of the Isotype Separation System. It will now deliver detailed process and mechanical designs for the system, including cryogenic and ambient temperature equipment to collect, process and recycle the tritium.
Jason Dreisbach, Head of Advanced Energy Technologies at AtkinsRäalis, said: “The H3AT Facility will be a first-of-a-kind research facility to strengthen UK and international efforts to advance tritium fuel cycle technology. The Isotype Separation System is a key element to demonstrate fusion fuel cycle performance at scale, and we look forward to contributing our significant experience in fusion engineering and tritium to help realise UKAEA’s ambitions.
Stephen Wheeler, UKAEA Executive Director, said: “This system will be the first industrial-scale Tritium facility for fusion in the world and will enable industry and academia to study how to process, store and recycle tritium.
In the chamber
With such small amounts of fuel involved, careful management will be required to make the most of any tritium within the fusion chamber.
Again ITER has led the way on this. Fuelling the fusion reactions in ITER is not a “once-through‚ process. Instead, fuel that is not consumed gets pumped out as part of the torus plasma exhaust, together with helium ‘ash’ and impurity gases. umps in the divertor region are designed to continuously exhaust to the Tritium Plant where it is then recycled and the deuterium and tritium are extracted for reinjection and reuse. Although the effective burn rate in the plasma chamber is estimated at only 1%, that is enough for helium ash to begin accumulating and the core plasma to dilute.
ITER anticipates a multi-stage extraction and separation process to recover ‘unburned’ tritium from its plasma. The Tritium Plant will have several sub-systems: tokamak exhaust processing to separate out impurities from the hydrogen isotopes; separation of deuterium from tritium in the exhaust gas; storage and delivery; recovery of tritium from impurity gases such as water vapour; and return to the fuelling stream.
Alongside tritium-breeding fusion reactors when they emerge, there are multiple sources of tritium currently operating worldwide as well as efforts underway to carefully conserve this precious resource where possible. However, with a global inventory of just 20 kg the question over whether there will be enough tritium available to sustain this rapidly growing industry before breeder reactors can come on-line remains a pressing one. For the fusion industry looking at the years of research needed to support the emergence of a commercial power play it is another question that needs an answer.
