DECEMBER 2025 U.S. Fusion Supply Chain Report CALEB BARNES SPECIAL COMPETITIVE STUDIES PROJECT Supply chain challenges could stall a burgeoning fusion industry, and addressing them effectively will be vital for the industry to rapidly grow. As the Special Competitive Studies Project’s (SCSP) Commission on the Scaling of Fusion Energy outlined in their report Fusion Forward: Powering America's Future, securing a future for fusion energy is of utmost importance. Lessons from other high-tech sectors, such as semiconductors and advanced batteries, underscore the risks of supply chain vulnerabilities and over-reliance on foreign sources, particularly from strategic competitors like China. In this report, we determine a ranking of the most widely used supply chain components within the fusion industry, then provide detailed analysis of each major system and its subsystems, including current locations of manufacture, dominant players in the space, and potential remedies for the most vulnerable components. A few fusion ecosystem inputs in particular stand out. Tritium, which makes up half of the fuel that most fusion companies plan to use, is a scarce material, with only 25-30 kg available worldwide. To breed more tritium and ensure a closed fusion fuel cycle, companies will likely require lithium enriched in lithium-6, for which there is no domestic commercial supply. Laser diodes are crucial to a subset of inertial fusion companies, and China leads in producing the necessary gallium and germanium, and is making strong progress in the diodes themselves. High-temperature superconductors, necessary for compact magnetic fusion approaches, lack reliable American producers, and China has recently taken the lead in global production volume with plans to massively scale up. Both neutral beam injectors and radiofrequency heaters, used to heat plasma in magnetic fusion devices, are specialized machines that rely on scarce materials and unique expertise. Our analysis of the fusion supply chain also found some areas where China is potentially vulnerable relative to the United States and Europe. For example, the United States is the world’s leading supplier of beryllium, which will likely be important as a neutron multiplier for breeding tritium, and as a component of FLiBe molten salts. Cryogenics, particularly the helium cryogenics needed for low-temperature superconductors, stand out as another area where the dominant manufacturers reside not in China, but in Europe. This method was to observe fourteen major fusion supply chain components, sort them by the number of companies that use them and those companies’ raised funding, and present recommendations for addressing the most vulnerable components. Further details of the ranking methodology and risk level analysis can be found in the Methodology section. For a table of fusion supply chain inputs, ranked by their threat level, see Appendix 1. For a table of nations and their relevant fusion supply chain production, see Appendix 2. Select policy mitigations from the full report are presented here. Key Recommendations: Create a Stable Demand Market for Fusion. Federal support to accelerate deployment of a fusion pilot plant through the National Fusion Goal and the Milestone-based commercial demonstration program would provide a powerful demand signal for the supply chain. • Elevate fusion to a national security priority, as recommended in the final report of SCSP’s Commission on the Scaling of Fusion Energy. This would allow fusion companies to bypass traditional roadblocks in procurement and signal demand, as happened with the CHIPS and Science Act. Onshore and Friendshore Manufacturing Capacity. Manufacturing components for fusion power plants on friendly shores would shelter the industry from foreign influence and provide markets for our allies. The United States can both partner with existing companies and support novel industrialists to accomplish this goal. • Tax credits (such as 45X and 48C) are a valuable tool for the U.S. government (USG) to encourage domestic development of manufactured parts. • The Department of Energy’s Office of Energy Dominance Financing (formerly the Loan Programs Office), a multi-hundred-billion-dollar government lending program, stands out as another avenue for the government to facilitate a fusion supply chain, particularly for applications like critical minerals development, capacitor and switch manufacturing critical for pulsed fusion applications, an isotope separation facility for lithium-6, or domestic steel forging. • Including fusion as a priority technology for investment in existing and potentially new place-based innovation organizations, such as the National Institute of Standards and Technology (NIST)-led Manufacturing USA Institutes, NIST Manufacturing Extension Partnerships, and the Economic Development Administration’s Regional Tech Hubs, would help grow innovation ecosystems that can contribute to the emerging fusion supply chain. • In the near and medium term, strategically partnering with like-minded nations like the United Kingdom, Japan, and Germany and including specialized fusion components like cryogenics and gyrotrons in trade deals and tariff exemptions would help stave off potential chokepoints. Recent bilateral tech partnerships with the UK and Japan have already begun this process. • Create a third-party data hub for creating training sets for the robotics that will be used in fusion machines. Prioritize R&D to Mitigate Challenges. Funding American researchers can help American supply chain leverage by both spawning industries and innovating our way out of dilemmas. Fusion suppliers can be the next in a long history of successful spinouts from research labs. • Federal research funding should prioritize fusion energy projects. • Research into fusion fuel cycles and tritium breeding can create self-sustaining fuel cycles, eliminating the pressure to supply tritium from external suppliers. • Research into radiation-resistant materials and robotics, as well as molten salt and liquid metal systems, can bring the United States early-mover advantages. Strategic Management of Critical Materials. A fusion machine will require a number of specialized materials with unique properties. Securing an adequate supply, domestically or from allies, will be essential to building the necessary components. • Ensure domestic and friendly nation production of maximally threatened raw materials. o Leverage domestic capabilities of to expand American production of fusion-relevant critical minerals (including gallium, germanium, gadolinium, and yttrium). MP Materials is the only domestic rare earths producer, and the Department of War is its largest shareholder, giving the United States Government unique influence. o Create an American supply of enriched lithium-6 by helping to establish a domestic commercial supply chain. • Canada looks to be the world’s top tritium producer until fusion machines begin to breed their own tritium, or other countries build more heavy water fission reactors. The United States should ensure continued flexible trade to maintain access to the tritium supplies that will be needed for fusion-plant startup. • For materials like beryllium, the United States should ensure that it maintains and expands its current lead in production. The USG should consider leveraging the Defense Production Act, as it has done so in the past, for production of beryllium and beryllium fluoride as well as other minerals and metals important for fusion energy. • A coordinated fusion fuel strategy from the United States government would provide a structured plan for availability of fusion fuels. Executive Summary .............................................................................................................3 Commercial Fusion Ecosystem Map .....................................................................................8 Research Methodology ...........................................................................................................9 Fusion Supply Chain Component Analysis ...........................................................................12 Electronics ..................................................................................................................12 Conventional Chips ..........................................................................................12 High Power Switches ........................................................................................13 Capacitors .......................................................................................................15 High Strength Forgings................................................................................................ 17 Advanced Robotics......................................................................................................19 Deuterium-Tritium ...................................................................................................... 21 Sidenote: Beryllium.......................................................................................... 23 Sidenote: Enriched Lithium .............................................................................. 25 RF Heating ................................................................................................................. 26 Molten Salts................................................................................................................ 28 High-Temperature Superconductors .......................................................................... 30 Sidenote: Rare Earth minerals ......................................................................... 33 Sidenote: Cryogenics....................................................................................... 34 Alternative Fuels ........................................................................................................ 35 Deuterium-Deuterium..................................................................................... 35 Proton-Boron-11 ............................................................................................. 36 Deuterium-Helium-3 ....................................................................................... 37 Neutral Beam Heating ................................................................................................ 39 Targets .......................................................................................................................41 Copper....................................................................................................................... 43 Liquid Metals .............................................................................................................. 45 Laser Diodes and Optics ............................................................................................. 47 Appendix 1: Supply Chain Risks Chart ...........................................................................50 Appendix 2: Fusion Supply Chain by Nation...................................................................52 Commercial Fusion Ecosystem Map: A map of the fusion technology ecosystem, with various supply chain needs overlaid. Included in the graphic are all fusion companies with funding of $50M USD or higher, plus members of the DOE Milestone program. As the first step in our analysis, we have developed an ecosystem map showing the various supply chain components (or fuel inputs) for fusion machines along different technology pathways. The diagram represents every global fusion power company with a known amount of funding $50M USD or higher, plus the members of the Department of Energy (DOE) Milestone Program. It displays a tree diagram that begins by distinguishing the underlying plasma management technology––inertial vs. magnetic vs. magneto-inertial––and branches into more specific technical approaches. Overlaid on top of the flow chart is a color-coded map displaying specific supply chain components of each company. The full data set of components and companies includes all companies with known valuations, not just those under $50M. Next, we determined the most widely used components in the supply chain based on all of the American companies within the data set. Given the wide variety of fusion energy approaches underway, something more sophisticated than a simple percentage of companies that use a component is needed. Therefore, we first use the total amount of funding raised by each company as a proxy for each company’s significance to the U.S. commercial fusion ecosystem. To determine the most critical supply chain components, the fraction of companies that use a component is weighted by the total funding raised by those companies. Forgings, advanced robotics, and electronics are considered universal components that any company will require, placing them at the top of the list. Similarly, copper should be used in all approaches (including in the fabrication of high-temperature superconducting tape), but only the companies that use it as their primary electromagnetic plasma control system are counted in this ranking. In a different methodology with no such threshold, the importance of copper would be higher. This ranking system values breadth of use, rather than depth. Due to the weighting mechanism, any component that is specialized to an approach will fall down the ranking list here, but this does not mean that a component would not be needed in a future fusion ecosystem. We still don’t know which technical approaches will work for an advanced fusion economy. If one approach ends up being the technological and economic winner, its components will end up being the most important ones, even if they appear relatively low in this ecosystem-level analysis. In most cases, it’s true that the technological approach will live and die with the availability of a component. For example, there could be no laser fusion companies without access to optics. There are also some cases where this methodology fails to take into account possible alternatives. Companies might have some degree of choice between banks of capacitors and banks of batteries, a choice between neutral beam heating and radio frequency heating, and a choice between liquid metals, molten salts, and solid walls. For the purpose of this analysis, it’s assumed that companies will continue to pursue the approaches that they currently have planned. What this list favors is universality, and manufacturing and maintenance-related components therefore take the top position. Based on this weighting, the most important supply chain components for American fusion companies are (in order from most to least used, weighted by the total funding raised by the company): 1(t). Electronics/Forgings/Robotics 4. Deuterium-Tritium 5. RF Heating 6. Molten Salts 7. High-Temperature Superconductors 8. Alternative Fuels 9. Neutral Beam Heating 10. Targets 11. Copper 12. Liquid Metals 13. Laser Diodes and Optics We then evaluate the current supply chain picture for the United States, with a particular eye on Chinese reliance, for all components. Additional consideration is given to the volume of a component needed for a fusion power plant. The risk level for a component (or subcomponent) is rated by the following metric: None: The component is readily available, has a nearly unconstrained natural supply, and has multiple domestic producers. Low: The component has significant natural resources and multiple domestic producers or a very small amount of the component will be needed. Medium: The United States may be mostly reliant on imports of a component, but is dependent on allies or neutral countries, with multiple providers. There may be some American industry of the component, but it is not currently capable of meeting the industry’s needs alone. Medium-High: The United States is largely reliant on a single non-allied country to provide a component, perhaps with a small American commercial industry. The technology to produce the component is known, but not at scale in the United States. High: The United States is reliant exclusively on non-allied countries for a component, or the technology does not yet exist to produce a component. We offer examples of policy levers that could be explored to improve the supply chain for components with a supply chain risk level at or above “Medium.” Notably, the risk levels do not account for the ubiquity of use; that is left up to the ranking system. As such, a component that only one company uses could still be called high-risk, and it would appear towards the bottom of the list. Electronics Every fusion company (much like any industrial facility of a similar size) will use a considerable number of electronic components. Both conventional semiconductors and high power electronics (HPE) will feature heavily throughout a fusion machine, particularly in the implementation of diagnostic and control software. Conventional Chips Conventional computer chips will likely be a major part of the success of a commercial fusion plant. A number of major advances in plasma control have been due to the development of advanced artificial intelligence (AI) systems. The planning and construction of future fusion plants will also be highly dependent upon simulations, perhaps with the utilization of AI to construct device parameters, or creating digital twins of components of fusion machines (and, in time, of the full machines themselves). These chip applications will be fairly conventional for large amounts of data processing. Advanced AI chips will likely be used for this capacity, putting the demands of this supply chain similar to a data center (or perhaps requiring the renting of time on data centers and/or supercomputers, potentially including those at our national labs). Fortunately, the United States is the global leader in advanced computing chip design and development (but not production, which currently comes predominantly from Taiwan). Reshoring domestic chip manufacturing was identified as a need in the CHIPS and Science Act, and has been further cemented by the federal government’s purchase of a 9.9% stake in Intel, the only major company that both designs and manufactures advanced chips in the United States. There are real threats that extend far beyond fusion from not having a domestic chip manufacturing ecosystem. The problem has been taken seriously over the last several years, and the computing industry’s demand represents a much larger market than that which will be needed for fusion energy. Semiconductor Chips Risk Level: Medium. Advanced chips are still generally manufactured overseas, but several mechanisms are in place to reshore domestic construction as-is. On the whole, it’s worth noting that this is not a fusion-specific component, and a chip shortage would be felt by a large share of the American economy. Possible Mitigation(s): Continue the buildout of domestic chip manufacturing proscribed in the CHIPS and Science Act. Fusion companies can partner with National Labs and lease out their supercomputing infrastructure for specific projects. Fusion companies can also negotiate for access to hyperscalers’ compute resources as part of Power Purchase Agreements (PPAs). High Power Switches The area most requiring fusion-specific electronic components is in high power electronics. Pulsed fusion power companies, which form roughly half of all U.S. fusion companies, rely on pulses surges of energy to initiate fusion reactions and are dependent upon power electronics. These are the driving costs for many of these machines. High power switches, in this context, includes semiconductors and spark-gap switches for the control and conversion of electric power, specifically those capable of withstanding extraordinarily large currents. These can be used in both pulsed and steady-state confinement approaches, although the pulsed fusion requires significantly higher requirements for current and peak voltage. For pulsed power (like lasers, Z-pinch, or FRC), these high power electronics are needed to survive the significant surge of electricity that comes through with each pulse or shot. For magnetic confinement approaches like tokamaks, these switches are used when activating an individual magnet–the abrupt nature of plasma whips necessitates significant surges of electricity be delivered to the controlling magnet, and the power electronics for confinement magnets will need to withstand tens of kiloamperes per AC/DC converter, and be able to quickly turn off and on. They will also be used to power the plasma heating systems that create fusion conditions. Traditional power electronics are usually made of silicon, but a new generation of wide-bandgap power electronics, made of materials like silicon carbide (SiC) or gallium nitride (GaN), offer the promise of higher operating temperatures and switching voltage. There may be additional components, such as copper and tungsten, that could be challenging to supply at scale. Most high power switch providers are based in Europe, although the most prominent supplier, Dynex, is now owned by a Chinese company. Other suppliers include OCEM Power Electronics (Italy), JEMA (Spain), Siemens (Germany), and Hitachi Energy (Japan). The United States’ contribution to the International Thermonuclear Experimental Reactor (ITER) project’s power supplies came courtesy of Princeton Plasma Physics Lab, and Princeton Satellite Systems (doing business as Princeton Fusion Systems) attempts to commercialize that technology. Generally, high power electronics for fusion are not a mature industry, and instead are a small, bespoke field driven by individual projects. High Power Switches Risk Level: Medium-High. The engineering expertise and facilities for HPE are currently spread globally, and no one country has a dominant share of the niche and nascent market. Depending on actions taken in the next few years, this could become a supply chain chokepoint for U.S. fusion companies, or could serve as an opportunity to position U.S. innovators at the forefront of an emerging industry. Possible Mitigation(s): Public-private partnerships, such as the Advanced Research Project Agency-Energy’s (ARPA-E) collaborations with the private sector, are already helping to cultivate an American HPE industry. These programs should continue and can provide an example of future programs to create American vendorship of HPE technology. Capacitors Capacitors are a standard electronic that can release a significant charge very quickly. For fusion machines, capacitors will be an integral part of nearly all pulsed-power approaches. They will also see use in some magnetic confinement designs to quickly turn on a controlling magnet. Capacitors are in constant use in modern devices, but those for a fusion machine will require stringent operating standards, tolerating higher voltages and currents than in consumer electronics. Large pulsed power facilities will require a world-class capacitor bank. The National Ignition Facility (NIF) power bank, for example, delivers more than 330 MJ of energy for each laser shot fired, making it the most energetic capacitor bank in existence. Private companies’ needs are more modest, but still noteworthy. Zap plans on using a ~2 MJ capacitor bank for a power plant, Pacific Fusion plans on using 80 MJ for a demonstration plant, and Helion plans on using a 50 MJ capacitor bank for their Polaris facility, requiring 150 shipping containers’ worth of capacitors, and estimates capacitors to be the second-largest component by cost in their system. Capacitor banks often have long lead times on the order of 18-24 months, and a fully functional power plant will require more stringent operational specs (the rapid sequence of firing multiple shots per second will likely require active cooling, and last for hundreds of millions of shots). In Helion’s case, only one supplier was capable of producing capacitors at the scale, quality, and cost required for even their preliminary machines, and that supplier was based in China. As a result, Helion set up their own capacitor factory in Washington, and have been capable of producing 5 MJ worth of capacitors per week, when operating. Zap, similarly, has acquired the machines to facilitate their own capacitor production. Capacitor Risk Level: Medium. The current supply of fusion-relevant capacitors is quite limited, and mostly overseas, but does not feature specialized materials or components that cannot be recreated here. Possible Mitigation(s): Tax credits for capacitor manufacturing would increase the economic viability of domestic production. Vertically integrating capacitor production is not economically feasible or strategically wise for every company, especially using venture capital, but has been appropriate for some. If companies who have vertically integrated their capacitor supply chain decided to sell them to other fusion companies with similar needs, this risk level would decrease. Like many other advanced industries, steel will be a major structural component of fusion machines. The steels used in fusion plants will likely need to survive extreme conditions, including very high temperatures, very low temperatures, neutron irradiation, corrosion, and significant mechanical stress. Steels used near the plasma chamber will likely need to be reduced-activation steel, which can better survive a lifetime of neutron bombardment. The other industry that shares most of these requirements, the fission power industry, is also in a difficult position for sourcing their steel—only Framatome (France) and Japan Steel Works are capable of producing reactor vessel forgings for American companies. Fusion power plants should be able to suffice with smaller forges, but still require very rigorous materials standards and a workforce capable of producing them. A 50% tariff was added on foreign steel imports, increasing the necessity of creating a domestic supply of steel. Major advances have happened overseas, with the UK’s Neurone Consortium (a collaboration of the UK Atomic Energy Authority, and academic and industrial partners both in the UK and internationally) capable of high-temperature-resistant steel production at scale, and Chinese scientists have formulated a steel capable of withstanding much higher temperatures and magnetic field strengths. Supply Chain Risk Level: Medium. U.S. fusion companies are not currently dependent on China for high strength steel production, but the lack of a domestic steel manufacturing industry and workforce, especially for specialty steels such as those needed for fusion, could significantly extend lead times and costs for the basic building blocks of fusion power plants as the industry scales. Possible Mitigation(s): The materials vulnerability could be partially alleviated through government incentives to reshore American high strength steel production, along with tariff exemptions for fusion companies (and others) as a national security priority. However, this alone will not address the forging skill and infrastructure gap. To mitigate this, nearshoring or friendshoring with key allies and partners should be considered as a short- to medium-term mitigation. To sustain the long-term growth of a U.S. fusion industry alongside those of many other advanced technologies, the United States must invest in workforce development for the domestic steel industry. One approach is to designate steel production as a priority area within the Manufacturing USA Institutes, while another is to explore foreign direct investment as a strategy to address steel shortages. Expanding Manufacturing USA’s funding and eliminating restrictive five-year term limits, along with modernizing the Manufacturing Extension Partnership (MEP) program to help small manufacturers adopt advanced technologies, could further strengthen domestic steel production and supply chain resilience. Further, the tax code can be leveraged to benefit the fusion industry, with the addition of 45X and 48C tax credits for fusion-capable high-strength forgings. The Office of Energy Dominance Financing could be a valuable tool for funding the initial stages of reshoring the kind of steel manufacturing needed. While not strictly necessary for a fusion machine to exist, advanced, radiation-resistant robotics will be highly likely to be integrated in all fusion approaches. Any human interaction in a tokamak requires a difficult cleaning process. For example, the recently decommissioned Joint European Torus (JET) required a three-month process for any human activity in the chamber, not including the time of performing the upgrades or repairs. As such, and considering the radioactive activation of fusion, no human entered the chamber for thirty years. This led JET to become a leader in advanced robotics, creating mechanical arms sensitive enough to allow an operator to feel if a bolt is correctly threaded. As a result, the UK is a major leader in fusion-specific robotics applications. Robotics would also incur a design advantage. A fusion machine built around robotic maintenance wouldn’t need human-sized spaces or human-amenable environments, significantly freeing up design criteria. After a fusion machine’s operational life ends, radiation-resistant robotics will be of use in site cleanup. The United States is a global leader in advanced robotics technology today, but China’s significant investments, strategic initiatives, and lower manufacturing costs pose a serious competitive challenge as the industry scales. An example of this can be seen in the Boston Dynamics ‘robot dog,’ Spot, which has been used for high-radiation work in the UK. The Chinese equivalent, the Unitree B2, is available for half the price (pre-tariff), with greater payload capacity, battery life, speed, temperature range, and climb angle. Another major facet of competition will be robotics adoption. China leads in installations of industrial robots, with a majority of all new installations coming in China. That industrial experience is paying off in the fusion sector, as China has revealed the successful testing of a class-leading robotics system at the CRAFT campus, capable of lifting 60 metric tons, with positioning accurate to a fifth of an inch. Other countries have made investments in radiation-hardened robotics. Russia recently announced the development of Rosatom’s “nuclear spider”, which analyzes reactor welds. Mitsubishi Heavy Industries also has an inspection robot, the A-UT, for underwater operation. In the United States, Idaho National Laboratory has done some work in developing radiation-resistant robotics. Advanced Robotics Risk Level: Medium. China and Russia have both made notable strides in radiation-resistant robotics, but the United States has significant experience in robotics overall, and has close ties with the overall industry leader in the United Kingdom. Possible Mitigation(s): Policy recommendations to accelerate U.S. leadership in advanced robotics broadly would have downstream positive impacts on the fusion industry. For example, the creation of voluntary, third-party dataset hubs would help offset China’s ability to compel data sharing across its innovation ecosystem. For fusion, such a hub could enable companies to access data related to the training and deployment of robots in fusion facilities and for associated manufacturing processes, accelerating adoption readiness. Additionally, radiation-hardened robotics should be an area of priority in government-funded fusion R&D. Almost every fusion company plans to use deuterium-tritium (D-T) fuel in some capacity, whether as the intended final fuel source (like Commonwealth Fusion Systems (CFS), Zap, and most other companies), or as a bridge in the phase before their final plant design is realized (like Helion and TAE). Deuterium is plentiful, making up one out of every 6500 hydrogen atoms in seawater, and costing about $13 per gram. There are at least three different suppliers of deuterium based in the United States, but none currently enrich their deuterium from natural water. Current suppliers like Cambridge Isotope Laboratories and Isowater don’t currently enrich deuterium from natural water, but instead purify existing stocks of deuterium oxide. Isowater has proposed an intention to make a plant to produce 20 metric tons of D2O per year from natural water. Deuterium Supply Chain Risk Level: Low-Medium. The natural supply of deuterium in water is enough to last for over a billion years of energy production, so that is of no concern, but the United States currently lacks the capability to produce deuterium from water. Tritium, however, is a different story, and has been consistently cited by industry as a future supply constraint. Within a fusion machine, tritium can be bred from enriched lithium, and commercial tritium exists as a byproduct from CANada Deuterium-Uranium (CANDU) fission reactors, which operate using heavy (deuterium-rich) water. Commercial tritium is expensive ($30,000–50,000 per gram) and scarce, with only around 25–30 kg currently available, mostly sold through Canada’s Ontario Power Generation (OPG). ITER plans to consume 12 of those kilograms in its experimental regime when it comes online, intended to be in the late 2030s. Fusion is also not the only industry use for tritium, which is also used in medical tracing, nuclear weapons, and luminescent devices. It has a half-life of around 12 years, so new tritium needs to be bred as extant supplies decay. Once fusion machines are operating, the industry expects blankets of lithium inside the machines to produce tritium above replacement levels, helping to close the fusion fuel cycle as well as create startup tritium for new fusion power plants. Should this work as planned, the major tritium-related hurdle will be in acquiring the startup supply for initial fusion machines and the early stages of commercial scale-up. A tritium fuel cycle may also require specialized oil-free pumps, which are not currently commonly available. Private fusion companies do not expect to need as much tritium as ITER requires to set up a machine. An ARC-class power plant, for example, has a startup inventory “on the order of kilograms.” If a mature fusion industry pursues plant designs of a similar size, then tritium inventory would affect the rate of scale-up, as tritium-doubling times limit how fast you can increase the number of new constructions. Even so, those fusion plants could run on deuterium-deuterium (D-D) fuel for a few months to create their tritium supply. While doing so would be expensive, a lack of available tritium capacity would not be a technical showstopper for the industry. The current method that the United States uses to produce tritium is the use of Tritium Producing Burnable Absorber Rods (TPBARs), which have been installed at Watts Bar Nuclear Plant in Tennessee (a conventional fission reactor) and processed at the Savannah River National Lab. While this concept theoretically could be used in a purely commercial application, TPBARs are not available for commercial use today and would require significant policy changes to become so. Tritium Supply Chain Risk Level: Medium. Tritium is currently quite scarce and is commercially supplied by only one foreign company, but can be created in existing machines. Functional breeding blankets will allow future fusion machines to create their own supply of tritium fuel. Possible Mitigation(s): The most important action the DOE could take is to derisk tritium breeding and closing the fusion fuel cycle through a mixture of research and the construction of high priority test stands or R&D facilities. Setting up a new facility exclusively to create a commercial tritium supply would likely be far more expensive than the combined opportunity and electricity cost of simply running a fusion machine on D-D fuel to breed a startup supply of tritium or the use of TPBARs or similar technology, for a relatively limited need. Alternatively, the USG could signal interest in companies building U.S.-based heavy water reactors of, or similar to, the CANDU design, which would not only supply tritium, but also be a new source of electricity to the grid. Sidenote: Beryllium A lithium blanket should produce more tritium than is consumed by the fusion machine. A neutron multiplier is required for this process. Beryllium is the best low-Z candidate material for this, and also is one of the few materials where the United States has a clearly defined advantage in mining and production. Building fusion power plants will require a dramatic increase in beryllium production. One estimate suggests that 1 GW of fusion electricity could require almost 500 tons of beryllium, higher than the 360 tons currently produced worldwide per year (but much lower than total identified resources of 100,000 tons). It is a difficult material to work with, as it is known to cause an increase in lung cancer. The American company Materion, the world’s most notable beryllium supplier, recently signed a deal to provide beryllium fluoride for fusion power plants. Only a few nations (the United States, China, Brazil, and Kazakhstan) have meaningful beryllium production capabilities. Of these, Kazakhstan sources their ore from a shrinking Cold War stockpile, Brazil has only recently scaled up production, and China mines a fraction of its production, but the United States creates by far the largest volume of production from fresh ores. While China is a net importer of beryllium, their imports today are generally Kazakh and other foreign sources rather than American. While they do have natural reserves, China will need to look further than their current supply chain to meet the demands of a fusion industry. Beryllium Supply Chain Risk Level: Low-Medium. In tallies of American leverage and policy options, beryllium is one critical component from which we have an advantage in both import and export today, and possess 60% of the world’s known beryllium reserves for future production. The world does not yet produce enough beryllium that a fusion ecosystem will need, but the United States can maintain and expand its leadership if it appropriately increases production. Sidenote: Enriched Lithium The lithium required to breed tritium fuel would be enriched in lithium-6 (6Li). At present, the United States lacks any commercial lithium enrichment capabilities. The Y-12 facility at Oak Ridge National Lab has begun construction of a lithium processing facility, projected to be completed in the early 2030s for $0.96-1.6 billion, but this would be targeted for non-commercial use. Chile and other South American countries have considered installing lithium enrichment facilities, but have not yet. Some advanced fission companies require lithium highly enriched in the other naturally occurring lithium isotope, lithium-7. Creating this, as Kairos Power is doing at their Molten Salt Purification Plant in New Mexico, will inherently create a feedstock of lithium enriched in the 6Li that fusion reactors will need, although likely not to the enrichment levels required. As of November 2025, Kairos has no plans to sell their 6Li-heavy feedstock externally. Support for domestic commercial lithium enrichment capacity could leverage a number of newer technologies, which could potentially be more cost-effective and reduce the environmental wastes of the standard COLEX process. The U.S. Government should incentivize avenues to make a supply of enriched lithium available to private fusion companies, and leveraging incentives such as tax credits and/or the Office of Energy Dominance Financing to facilitate construction of one or more modern lithium enrichment facilities. A coordinated fusion fuel strategy from the U.S. Government would provide a structured plan for availability of fusion fuels. Until commercially accessible lithium enrichment facilities are constructed and completed, 6Li access is very limited. Enriched Lithium Supply Chain Risk Level: High. Raw lithium has more exporters who produce a much greater volume (as is required for global battery production), but China is still a major lithium exporter. The amount of lithium required for a fusion plant (10 to 100 tonnes to start and sustain a machine) is significantly less than will be needed in the battery and EV industries, for example, and there are over a hundred million tons of lithium reserves worldwide. Raw Lithium Supply Chain Risk Level: Low. To achieve the more than 100-million-degree temperatures needed in a fusion machine, most fusion energy approaches require heat to be delivered into a plasma without physical contact with walls. For magnetic methods, the most common approach is radio-frequency heating, akin to how a microwave oven works. The machines used to generate radiofrequencies are typically gyrotrons, which are used to heat the electrons or ions inside the plasma of the fusion chamber. Gyrotrons for RF heating were the most-noted present-day supply chain constraint noted in the industry survey of the FIA 2025 Supply Chain Report. In addition to being a component themselves, gyrotrons rely on critical materials like tungsten, copper, graphite, boron, and molybdenum for specialized vacuum components, as well as superconducting materials and advanced ceramics. Gyrotrons demonstrate both a level of complexity and a reliance upon other nations (including China) that may create significant vulnerabilities for the fusion supply chain later down the line. Russia has historically dominated the gyrotron industry, due to decades of sustained investment in high-power microwave technologies. In recent years, Japan’s Kyoto Fusioneering has also become a major supplier in the gyrotron space. Separate from fusion, gyrotrons also play a critical role in industrial heating, enabling ceramic sintering, glass processing, and semiconductor manufacturing, and are also being explored for directed-energy weapons and as tools for deep borehole drilling for geothermal plants. RF Heaters Supply Chain Risk Level: Medium-High. The gyrotron industry is still in early stages of growth relative to other markets within the fusion ecosystem. With key dependencies on other nations for both materials and expertise, the United States will need to invest in domestic gyrotron capacity and/or deepen strategic partnerships with friendly nations that can meet future supply chain needs. Possible Mitigation(s): Elevating fusion energy to a national security priority and articulating key components in the fusion supply chain could provide an early demand signal for a domestic gyrotron industry. To mitigate potential vulnerabilities in the gyrotron supply chain, the United States could, in support of the National Fusion Goal, leverage the Defense Production Act to secure enough supplies of tungsten, molybdenum, and graphite for the construction of the first fusion plant. This would also help shore up supply chains for other industries that use these components, including aerospace, defense, and medical industries. Molten salts, which are also under consideration for advanced fission reactors, are a solution to a number of problems inside a fusion machine. A primary purpose, in both fusion machines and fission reactors, would be for heat extraction. In a typical fusion plant design, neutrons would deposit their energy in the molten salt layer, which flows out of the fusion chamber and to a heat exchanger, which eventually leads to making steam to spin a turbine. For D-T fusion, the molten salt blanket can serve as a form of shielding for other components of the machine, which will need protection from the 14 MeV neutrons. These molten salt blankets will also be used as the location to breed additional tritium. The most commonly considered salt would be FLiBe, a combination of fluoride (Supply Chain Risk Level: None), lithium, and beryllium, which would include both a neutron multiplier and a source for tritium to be bred from (see the “Enriched Lithium” and “Beryllium” sidenotes above). Other molten salts, like FLiNaK, are under consideration, but FLiBe has a longer research history, exceptional heat capacity, and contains a neutron multiplier that makes tritium breeding easier. Most fusion companies plan to have a liquid component in their machines, whether it be molten salts or liquid metals. Similarly to liquid metals (see “Liquid Metals”), molten salts are also under consideration for some designs of advanced nuclear fission reactors. Current advances in handling, pumping, and engineering molten salt systems for fission reactors will assist the fusion industry. The most critical challenges around molten salts regard salt chemistry and materials. Molten salts will need to be of a very high purity, as is the case with many materials in a fusion machine, to prevent radioactive activation and corrosion. One particular corrosion concern for FLiBe salts (or any fluoride-containing salt) is the creation of tritium fluoride, which is highly corrosive. Some open research questions will affect the operation of a tritium stripping and gas supply system within the blanket. Designing around molten salts carries a number of difficulties, including high operating temperatures. There is still work to be done to ensure that a molten salt system would flow smoothly, but this is more of a matter of research and development than it is of sourcing of parts and raw materials. As long as that R&D continues, molten salt pumps and equipment should be relatively easy to scale. Molten Salts Supply Chain Risk Level: Low-Medium. Molten salt systems are still in early stages of development, but are more likely to depend on the manufacturing of advanced pumping systems and corrosion-resistant piping than on scarce metals. High-temperature superconductors (HTS), which were discovered in the 1980s, can deliver uniquely strong magnetic fields with no resistance, which can enable fusion reactors to be much smaller. For fusion, HTS will most likely be used to its greatest degree in magnetic confinement approaches for plasma control, but are also often found in particle accelerators, as would be seen in a neutral beam heater. HTS could also be important for future data centers and electric transmission lines. The standard form of superconducting wire/tape used by fusion companies is Rare-Earth Barium Copper Oxide (REBCO). REBCO wire/tape features a major advantage in its critical temperature, which can be over 77 K. High-temperature superconductors make the challenges of cryogenics much easier by operating at temperatures above that of liquid nitrogen, which is relatively common and cheap, whereas low-temperature superconductors require liquid helium, which is much more expensive. The potential difficulties of acquiring HTS are more than just the sum of the materials. Superconducting tape and superconducting wire are complicated and expensive, needing to withstand high thermal stress and electromagnetic forces and a commercial fusion power plant will require more than has ever been produced before. A number of stellarator companies seek to have delicate superconductor shapes surrounding their vessels in a design to facilitate steady-state operation. Tokamaks will generally have a simpler HTS geometry, needing circular coils of superconductor. In stellarators, the complexities are more likely to be resolved in the design and construction process, and advanced manufacturing techniques (like additive manufacturing) had to be innovated for stellarators to become a viable concept. The world’s top consumer of REBCO magnets is the U.S. fusion company Commonwealth Fusion Systems, which has already acquired ‘nearly all’ of the 10,000 km of superconducting tape required by their upcoming SPARC demonstration machine. CFS produces REBCO magnets primarily for their own machines, but has delivered magnets to the University of Wisconsin and has inked a deal to exclusively license their superconducting magnet technology to Type One Energy. CFS generally sources their HTS tape through Japanese suppliers like Faraday Factory, which have historically led the world in HTS production and produce about 3,000 km of HTS tape per year. Faraday Factory has also signed further agreements to become the HTS provider for additional companies like Proxima Fusion, despite international concerns about its majority owner and chairman, Andrey Vavilov. There are a few companies in the United States producing HTS tape/wire, but in small volumes and with long lead times. Two of the three American HTS companies pursue the Metal-Organic Chemical Vapor Deposition (MOCVD) approach, which has so far struggled to achieve the price point, quality, and consistency of the Pulsed Laser Deposition (PLD) approach pursued by overseas manufacturers. The last, High Temperature Superconductors, Incorporated, has yet to produce complete tape, but is positioning for completely American supply sourcing. Recently, China has begun to significantly ramp up its efforts in the production of HTS tape. There are now four notable Chinese HTS suppliers, and they have set a price target under half of what Japan’s tape sells for. The largest of these companies, Shanghai Superconductor, is undergoing an expansion that will bring their annual capabilities to double that of Faraday Factory. These Chinese companies claim similar levels of product quality as their Japanese competitors, and aim to quintuple their output by 2027. HTS Risk Level: Medium-High. HTS production is difficult and expensive, and American companies currently lag behind manufacturers in the East in volume, price, and consistency. China has taken steps to try to dominate this industry, and the Japanese industry leader is owned by a Russian oligarch. Accelerating American production of HTS tape would benefit a number of magnetic fusion companies, particularly as they scale. Possible Mitigation(s): Ensure that HTS tape is eligible for tax credits. Additionally, to ensure short- and medium-term needs are met, HTS tape should be a priority area in supply chain cooperation between the United States and Japan. Sidenote: Rare Earth minerals Rare Earth minerals are definitionally required to create REBCO magnets, and some, like gadolinium, are used in laser diodes as well (see the “Laser Diodes” subsection below). Most extant REBCO magnets use yttrium, but lanthanum, samarium, neodymium, gadolinium, and europium can also be used. Yttrium production is almost entirely dependent on China, which provides 94% of the United States’ inventory, as well as 74% of the total rare earths inventory. Over the course of the past year, China has exerted additional controls over supplies of rare earths, but as of November 2025, has suspended a number of export control regulations on rare earths and other fusion-relevant critical minerals for the coming year. There is some domestic rare earth mining through the Mountain Pass mine in California, the only rare-earths mine in the United States. MP Materials, the company in charge of Mountain Pass, initially had to send their mined ores for processing in China, but now processes their own neodymium and praseodymium domestically. In July 2025, the Department of Defense acquired $400M in stock, becoming the largest shareholder, and provided $150M in loans for upgrading MP’s processing facilities. Fortunately, the amount of yttrium in HTS tape is very small, with the Y2O3 layer being less than a micron thick. To build an entire fusion plant, the volume of yttrium would be on the order of just 100 kg. That consumption would be less than a ten-thousandth of 2022 yttrium oxide imports. As a result, threats of crackdowns on rare earth exports are likely not an existential threat to the nascent fusion industry, and HTS producers should be able to scale to meet growing industry needs. Yttrium/Rare Earths Supply Chain Risk Level: Low-Medium. Sidenote: Cryogenics Despite the name, high-temperature superconductors need to be quite cold, but to different degrees. Superconductors with a critical temperature above 77 K, like REBCO, can be cooled by liquid nitrogen, which is cheaper and more available than the liquid helium needed to reach lower temperatures. Cryogenics are used not only for reaching superconductor temperatures in magnetic fusion, but also for freezing the fusion fuel in inertial fusion approaches. Cryogenics is one of the few fields where China clearly lags behind American allies. The leading companies in cryogenics are mostly European, including the French Air Liquide and the Swiss-German Linde Kryotechnik. China has now begun to work on their own cryogenic development, and have in fact partnered with Air Liquide to develop their own helium refrigeration capable of cooling to LTS temperatures. Cryogenics are one of the 19 research experiment subjects at China’s new CRAFT campus in Hefei, the center of Beijing’s fusion R&D program. Cryogenics Supply Chain Risk Level: Low-Medium. As long as the United States maintains its trade relationships with European allies, cryogenics are an opportunity for our axis of influence to advance. “Alternative Fuels” refers to non-D-T fusion fuels. The promise of alternative fuels is fusion that generates fewer, lower-energy neutrons, which would require less radiation shielding and material requirements. In addition, these fuels reduce proliferation concerns and eliminate the need for tritium breeding. That promise is paired with problems, however, as it is more difficult to achieve fusion reactions with advanced fuels than with D-T. Ideal Temperature Estimates for a Fusion Plant Using Different Fuel Types: - Deuterium-Tritium (D-T): 150 Million °C - Deuterium-Deuterium (D-D): 400 - 500 Million °C - Deuterium-Helium-3 (D-3He): 500 - 600 Million °C - Proton-Boron-11 (p-11B): 1.5 Billion °C Deuterium-Deuterium Deuterium-deuterium fusion combines two deuterium (hydrogen-2) atoms (See “Deuterium-Tritium”). D-D fusion requires temperatures higher than D-T fusion does, in the range of 400-500 million degrees for a suitable power plant. Neutrons in D-D fusion have similar energies to those in a fission reactor, avoiding some materials challenges found in D-T fusion, but at the cost of much higher required reaction temperatures. For the first generations of fusion plants, D-D may find its use not as the main fuel cycle, but instead as an initial fuel to breed fuels more economical to use. D-D fusion can also occur as an incidental side reaction for other deuterium-based fuel cycles, particularly the D-3He reaction detail later in this section. Proton-Boron-11 Proton-Boron-11 (p-11B) fusion is the combination of a single proton (a common hydrogen ion) with a boron atom, turning into three alpha particles (Helium-4). The p-11B reaction itself will not produce neutrons, but various side reactions will, albeit several orders of magnitude fewer than D-T. The process is slightly less energetic than other relevant fusion reactions, but is still more energy-dense than any current form of power production. Unfortunately, the incredible promise is paired with significant challenges. Proton-boron fusion will require temperatures almost ten times higher than D-T fusion. The reduced level of neutrons, while limiting radiation, also means that new methods of energy capture will be required. Still, a number of companies consider the promise of p-11B enough to pursue it as their primary approach. Looking at the constituent elements of p-11B, hydrogen is the most abundant element in the universe. It can be produced from water, and has been a clean energy target for decades. Pure hydrogen is very cheap, costing about 0.1 cents per gram. The demand for deuterium in other fusion approaches would also inherently create a supply of undeuterated hydrogen. A 1 GW power plant might need 100 kg of fuel or less per year. Hydrogen Supply Chain Risk Level: None. Boron is also not a major supply chain risk. The United States is a top exporter of boron (and China is a top importer). Unlike tritium and deuterium, 11B is the most common natural isotope of boron, comprising about 80% of natural boron. Boron-11 is also cheap, with 99% enriched boron oxide available for about $70 per gram. Finally, since a fusion power plant would not need very much boron-11, boron availability is unlikely to be a problem in the foreseeable future. Boron-11 Supply Chain Risk Level: None-Low. Deuterium-Helium-3 Deuterium-Helium-3 (D-3He) has some advantages and some drawbacks relative to p-11B fusion. As an advantage, D-3He requires a much lower temperature than p-11B and produces over twice as much energy per reaction. As a downside, D-He3 will inherently produce D-D fusions in its process, meaning that D-3He will still be a neutronic approach, although measurably less than D-T. Helion is the only major company attempting to commercialize D-3He fusion. Helium-3 is the only ‘advanced’ fusion fuel material that is not naturally abundant on Earth, but it can be bred. While helium-3 is stable, it is a miniscule fraction of natural helium. 3He is also the product of tritium decay, so sources of tritium are also inherently sources of 3He with time and processing. The Canadian company OPG (See “Deuterium-Tritium”) is the only current civilian seller of 3He through its subsidiary, Laurentis Energy Partners. The most abundant known source of 3He is in the regolith found on the Moon, and the Department of Energy has made an agreement to purchase lunar-derived 3He for a delivery date by 2029. Helium-3 is similarly expensive to tritium, now costing over $30,000 USD per gram, and a D-3He fusion machine will likely require a similar mass of 3He as a D-T machine would use tritium. Fortunately, since Helium-3 can also be bred from D-D reactions, a company intending to use this fuel cycle could run their machine with D-D fuel until they create startup inventory, and continue to breed some additional fuel during normal operations. Helium-3 Supply Chain Risk Level: Medium-High. Like tritium, helium-3 is highly scarce, but can be bred from more commonplace elements in regular fusion machine operation. Possible Mitigations: Any efforts to increase tritium production would also increase production of 3He, if processed. Lunar helium mining would represent a giant leap forward in effort, but may be a natural step as we build our space travel capabilities in advance of a Martian mission. The primary alternative to RF Heating (see “RF Heating”) is Neutral Beam Heating (NBH) / Neutral Beam Injection (NBI), where particle accelerators play a crucial role by injecting fast neutral particles into the plasma. NBH is generally used in magnetic fusion, and is being pursued by companies designing tokamaks, stellarators, magnetic mirrors, and more. Most companies pursuing neutral beam heating as an approach are also considering RF heating as well, though the inverse is not necessarily true. As magnetically confined fusion machines have very strong magnetic fields, a charged particle could not be reliably injected into the medium without being repelled on the way in. To inject neutral beams, a particle accelerator brings high-speed deuterium ions through a deuterium gas (see “Deuterium-Tritium”). Electrons are stripped off from the gas, and pair with the high-energy deuterium atoms. The atom, now electrically neutral, can pass through the magnetic fields into the plasma. Some companies consider NBI a much cheaper method, citing cost savings of up to 50% over RF heating. The complexity of particle accelerators, their reliance on specialized materials, and intricate existing global supply chains make them vulnerable to supply chain gaps. From a materials perspective, particle accelerators in fusion rely on scarce, high-performance materials. Key components may require high-purity tungsten, of which China provides 80% of the world’s supply (Supply Chain Risk Level: Medium-High). High-temperature superconducting (HTS) tapes and RF power systems are also used within particle accelerators. With China controlling some aspects of the supply chain for these materials, the U.S. fusion industry faces potential bottlenecks and foreign dependence. There are a few American companies that could act as suppliers of NBIs. General Atomics and TAE Technologies stand out as two likely vendors currently established in the space. As the space is currently rather niche, neither have the current capacity to provide NBIs for the entire fusion industry, but would be likely to be the foremost beneficiaries of expanded demand. Beyond material scarcity, particle accelerators for fusion rely on a fragmented and highly specialized supply chain. Many critical components, such as RF cavities and vacuum chambers, require precision manufacturing concentrated in Germany, Japan, and China, leading to years-long lead times (Supply Chain Risk Level: Medium). A shortage of accelerator physicists and engineers further strains production. Additionally, cross-sector competition from industries like medical imaging, semiconductor fabrication, and defense could increase costs and limit availability. Neutral Beam Heating Supply Chain Risk Level: Medium-High. Like many other components on this list, neutral beam injectors are a specialized piece of equipment without the current industrial base to support a full fusion industry. Neutral beams also happen to use a number of somewhat vulnerable components found elsewhere on this list, like high-temperature superconductors, along with tungsten. Those vulnerabilities could compound as the industry scales. Possible Mitigation(s): Creating a robust neutral beam injector industry in the United States will require workforce training and education, and securing appropriate supplies of critical materials will be necessary. The elevation of fusion to a national security priority would be instrumental in clearing the way for fusion industry access to these devices. Targets, which consist of the frozen fusion fuel and sometimes a shaped structure around it, are an essential component in multiple fusion approaches, and they take different forms where necessary. Laser fusion can be separated into two approaches: direct drive, where the laser strikes the fuel pellet, and indirect drive, where the laser strikes a hohlraum around the pellet, which emits X-rays that compress the fuel pellet inside. Targets are also used in pulsed inertial or magneto-inertial approaches. In some variations, a liner surrounding the fusion fuel is compressed by strong magnetic fields. For laser-driven inertial fusion, the most notable difficulty in acquiring targets today is simply cost, as each is a bespoke, ultra-high tolerance part. At the National Ignition Facility, the world’s most successful inertial fusion machine, targets can cost upwards of $10,000 apiece. It’s virtually guaranteed that fusion target costs will significantly decrease with mass production of any specific target design, but it’s yet unclear if targets can reach the ~$1/target threshold many consider to be necessary for economic viability. For pulsed inertial fusion, the targets can reach substantially lower costs (even much less than a $1/target) but still need to be manufactured in the millions at scale. A system will need to be manufactured in all cases to fill these targets. The materials in a target are generally not those with major supply chain constraints, but some are expensive or difficult to work with. The hohlraum of an indirect drive laser target will be made of a high-Z material, like gold or depleted uranium, that may be expensive, but will be accessible. Supply Chain Risk Level: Low-Medium. The question regarding fusion targets is likely not going to be if they can be manufactured using accessible materials, but if they can be made under the strict confines of cost and quality control for a fusion plant’s financial viability. All fusion companies will likely use copper to some extent, as a component in electronics, wiring, and REBCO superconductors. However, for the purpose of this analysis, only the companies that plan on using copper magnets for plasma control (i.e., not using HTS tape) are counted in this paper’s ranking methodology. Those companies have exceptional needs for copper. For example, Helion used over 600 tons of copper for Polaris, their pre-commercial test machine. Copper is also arguably the most threatened constituent material in REBCO high-temperature superconducting tape. Relative to high-temperature superconductors, copper is weaker and can handle less current, but is more durable, malleable, can function without cryogenics, significantly cheaper, and can turn on much faster. Most importantly, copper wires have been manufactured for decades and don’t require significant growth of a new manufacturing industry. Without counting the demand from fusion, copper needs are generally rising, with an additional 61 mines expected to be needed by 2030. Developing new mines is lengthy and laborious, taking an average of 29 years to complete in the United States. In November 2025, copper was added to the United States’ critical minerals list, reflecting its importance not just to the fusion industry, but also to the broader technology sector. Currently, copper production is predominantly centered in Latin America (but with other significant production in Africa, Asia, and North America), and China is the dominant importer of raw copper and producer of finished copper products. This dominance is a result of over a decade of investment into third-world infrastructure through the Belt and Road initiative’s loan programs and direct ownership. In the Democratic Republic of Congo, the second-largest copper producer, Chinese firms own over 80% of copper mines. In recent years, the United States has backed a major railway in Africa to facilitate access to copper, but still lags behind China’s influence. Supply Chain Risk Level: Medium. Copper will be used near-ubiquitously in fusion machines, but only a few companies (those who use magnetic fields but not superconducting magnets) will have truly exceptional volume needs. Copper is in high demand, but there is a robust global supply chain already, and no singular reliance on any competitor state. Possible Mitigation(s): Additional investments in copper-rich places, like with the Lobito Corridor railway, would serve to increase American access to copper. Liquid metal walls have been a proposed solution to the problem of high-energy neutrons from D-T fusion irradiating structural components. The difficulties surrounding liquid metals are not about obtaining metals, but rather the complicated engineering of pumping metal through a system without corrosion, melting structural components, or clogging. Liquid metals are not just limited to fusion machine use, but also appear in designs of next-generation fission reactors. The purpose of liquid metals in a fusion power plant is similar to the role that molten salts would fill (see the Molten Salts section). Liquid metals would serve as a heat conduit, as a medium for multiplying neutrons and breeding tritium, and as a protective layer for more sensitive equipment. Most fusion companies pursuing this approach plan on using lead-lithium as their liquid metal of choice (see the Enriched Lithium sidenote of “Deuterium-Tritium”). In this formulation, lead takes the role of neutron multiplier, avoiding the limited supply of beryllium that could limit the scaling of the FLiBe molten salt approach. Research into all properties of lead-lithium is still limited, and new experiments are recommended to enable use in fusion machines. Some remaining challenges include liquid metal surface characteristics and the magnetohydrodynamic effect. Liquid Metals Supply Chain Risk Level: Low-Medium. This segment of the industry is too new to have a fully developed supply chain, but there are unlikely to be any material constraints that significantly limit the ability to scale in the United States. Just like with molten salts, the challenges remaining are mostly related to research and development, and not the current manufacturing and materials landscape. Additional research is required, which creates the opportunity for the United States to lead in production and sale of liquid metal systems. While not necessary for other approaches, laser diodes are absolutely critical for many laser fusion companies, and are those companies’ current greatest bottleneck in the supply chain. One analysis indicates that laser diode emitters would make up 60 percent of the cost of a laser fusion power plant. Even magnetic fusion companies may have some dependence on laser diodes, as excimer lasers are used in the production of Pulsed-Laser Deposition HTS magnets. For a profitable diode laser power plant, pump diodes will likely need a price point of only a few cents per watt, an order-of-magnitude price reduction. Diodes will also need longer lifetimes, to extend from surviving approximately 2 gigashots to 15-30. China produces the largest volume of laser diode exports of any country, but the highest-quality laser diodes, capable of withstanding a fusion machine’s beamline, have been made in the United States, Japan, or Europe (with Germany standing out in particular). This is beginning to change, as major laser manufacturer Coherent recently opened an office in Shenzhen. Dilas, a German subsidiary of the American company Coherent, has manufactured in China since 2008, and allegedly has recently had its staff poached by Chinese corporations. Potential supply chain disruptions could occur not just at the level of manufactured laser diodes, but also regarding a diode’s constituent materials. Gallium and germanium are necessary for current laser diode formulations, and China recently banned exports of the two metals before reversing course. China has produced 90 percent of the world’s gallium over the last decade. If needed, American companies may still be able to import both metals through third-country trade loopholes, but this could potentially be shut off. Indium is another necessary material for diode production, on which the United States is completely import-dependent, with China as the world’s largest producer. Other rare earths, like gadolinium and scandium, can be used in certain diode formulations. Optics, the mirrors, lenses, and fibers that control the path of a laser beam, will also be a necessity for a laser fusion facility. A laser fusion facility will also have significant constraints in terms of optical purity and robustness. Considering the conditions of a fusion power plant, these optics will likely need to meet optical, mechanical, and thermal specifications that have never before been combined in one optic. Alternative optical constructions, including gas and plasma optics, have been proposed, but do not exist at any commercial scale. Laser Diodes and Optics Risk Level: Medium-High. While the United States and its allies produce most of the high-quality laser diodes that would be used in a fusion power plant, our production capacity would need to be significantly increased, and the cost of diodes significantly lowered. The raw materials behind laser diode production also have notable Chinese dependencies, and manufactured optics will need to be of a yet-unique quality for production. Possible Mitigation(s): Provide tax credits for the manufacture of advanced laser diodes. Secure a supply of relevant rare-earths through the Department of War’s stake in MP Materials. Component, Subcomponent, or System | Risk Level Enriched Lithium | High Helium-3 | Medium-High RF Heaters | Medium-High High Power Switches | Medium-High Tungsten | Medium-High Neutral Beam Injectors | Medium-High Laser Diodes and Optics | Medium-High High-Temperature Superconductors | Medium-High Capacitors | Medium Advanced Robotics | Medium Tritium | Medium Semiconductor Chips | Medium Precision Manufacturing | Medium Copper | Medium Molten Salts | Low-Medium Deuterium | Low-Medium Rare Earths / Yttrium | Low-Medium Cryogenics | Low-Medium Targets | Low-Medium Liquid Metals | Low-Medium Beryllium | Low-Medium Raw Lithium | Low Boron-11 | None-Low Flouride | None Hydrogen | None Nation | Fusion Input Brazil | Beryllium Canada | Tritium, Helium-3 China | Steel Forgings, High Power Switches, Capacitors, Robotics, Lithium, High-Temperature Superconductors, Rare Earth and related minerals (yttrium, gallium, gadolinium, germanium), Tungsten, Precision Manufacturing, Copper, Laser Diodes, Indium Democratic Republic of Congo | Copper France | Cryogenics Germany | High Power Switches, Cryogenics, Precision Manufacturing, Laser Diodes Italy | High Power Switches Japan | Steel Forgings, High Power Switches, Robotics, Gyrotrons, High-Temperature Superconductors, Precision Manufacturing, Laser Diodes Kazakhstan | Beryllium Peru | Copper Russia | Robotics, Gyrotrons, HTS* South Korea | HTS Spain | High Power Switches Switzerland | Cryogenics Taiwan | Computing Chips United Kingdom | Steel Forgings, Radiation-resistant Robotics United States of America | Capacitors, Robotics, Beryllium, HTS Magnet Systems, limited Rare Earths, Neutral Beam Injectors, Laser Diodes * mostly through ownership of Faraday Factory, rather than local production