The CANDU (CANada Deuterium Uranium) reactor design has long been recognized for its exceptional neutron economy and fuel-cycle flexibility, setting it apart from most light-water reactors that require enriched uranium. CANDU units operate on natural uranium while also being capable of utilizing a wide array of alternative fuel types—from recycled uranium and plutonium to thorium‑based fuels. This inherent adaptability makes the CANDU platform a focal point for advanced fuel recycling strategies that promise to drastically reduce high‑level waste, conserve natural resources, and enhance the sustainability of nuclear energy. Over the past few decades, significant progress has been registered in both the chemical and engineering dimensions of separating and refabricating spent nuclear fuel tailored to CANDU reactors. Yet formidable technical, economic, and regulatory obstacles still stand between laboratory‑scale demonstrations and full‑scale commercial deployment. The global push to close nuclear fuel cycles, driven by rising uranium prices in certain periods and waste management imperatives, has kept CANDU recycling at the forefront of international research.

Advances in CANDU Fuel Recycling Technologies

Research and development efforts have pursued multiple parallel paths to recover fissile materials from spent fuel and reintroduce them into the CANDU fuel cycle. These pathways range from adaptations of classic aqueous reprocessing to entirely dry, non‑aqueous techniques, each offering distinct advantages in terms of efficiency, waste volume, and proliferation resistance. The sections below detail the most prominent technologies, their current maturity, and the specific adaptations required for CANDU fuel.

Evolution of Aqueous Reprocessing and the PUREX Standard

The Plutonium Uranium Redox Extraction (PUREX) process remains the most mature industrial method for separating plutonium and uranium from irradiated fuel. Originally developed for military programs and later adapted for civil light‑water reactor fuel, PUREX has been customized for CANDU‑compatible fuel streams. In this process, spent fuel elements are dissolved in nitric acid, and the resulting solution is subjected to multiple stages of solvent extraction using tributyl phosphate. The purified uranium can be re‑enriched or, more typically, blended down to natural uranium equivalence for direct fabrication into new CANDU bundles. Separated plutonium, when combined with uranium or thorium oxides, forms the basis of mixed oxide (MOX) fuel. Although PUREX technology is well‑established, its application to CANDU fuel cycles introduces unique challenges: CANDU spent fuel has a lower burn‑up (typically 7–8 GWd/tU compared to 40–50 GWd/tU for LWRs) and a significantly different isotopic composition, with higher residual fissile content and a lower concentration of fission products. This requires adjustments to process chemistry and flow sheets—for example, lower acidity levels in the dissolution step and modified solvent-to-feed ratios. Ongoing optimizations have reduced secondary liquid waste volumes, but the technology still generates organic solvent degradation products and medium‑level liquid waste that demand rigorous treatment. Recent advances at facilities such as the Canadian Nuclear Laboratories have focused on integrating PUREX with head-end steps that remove tritium and carbon-14 before dissolution, reducing environmental releases. More recent innovations include the use of monoamide extractants, which are easier to incinerate and produce less secondary waste, although they have not yet been scaled for CANDU fuel streams.

The DUPIC Cycle: Direct Use of Spent LWR Fuel in CANDU

One of the most innovative strategies developed specifically for CANDU reactors is the Direct Use of spent PWR fuel In CANDU (DUPIC) concept. Unlike traditional reprocessing that chemically separates uranium and plutonium, DUPIC relies on a dry thermal/mechanical process to convert spent pressurized water reactor (PWR) fuel pellets directly into CANDU‑compatible fuel elements. Spent PWR fuel assemblies are dismantled, the pellets are crushed, de‑clad, and subjected to a series of high‑temperature oxidation and reduction cycles (the OREOX process) that pulverize the material and release volatile fission products including krypton and tritium. The resulting powder is then pressed, sintered, and loaded into CANDU sheaths. Because no complete separation of plutonium occurs, the material remains highly radioactive and never presents a pure plutonium stream—an inherent proliferation‑resistant feature. South Korea, in collaboration with Canada and the International Atomic Energy Agency, extensively explored DUPIC throughout the 1990s and early 2000s, fabricating small batches of DUPIC fuel and successfully irradiating test elements in the HANARO research reactor. According to the IAEA’s review of DUPIC technology, the fuel cycle offers a net reduction in the volume of high‑level waste per unit of energy produced, while simultaneously doubling the energy extracted from the original PWR fuel. However, the high radiation fields emanating from DUPIC powder and pellets—often exceeding 1000 Sv/h at contact—necessitate fully remote, heavily shielded fabrication facilities, which have proven economically challenging to scale. Despite these hurdles, a 2022 feasibility study by the Korean Atomic Energy Research Institute concluded that DUPIC could become cost-competitive if coupled with a future carbon tax or if uranium prices exceed US$200/kgU. In 2024, a new collaborative project between KAERI and CNL was announced to evaluate a scaled-down DUPIC line that could be incrementally expanded, reducing initial capital risk.

Dry Reprocessing and Pyroprocessing Innovations

Away from aqueous chemistry, a suite of dry reprocessing methods—collectively termed pyroprocessing—has attracted considerable interest for CANDU fuel cycles. Pyroprocessing uses molten salts (typically lithium chloride‑potassium chloride eutectic) and liquid metal electrodes to electrochemically separate actinides from fission products at elevated temperatures (450–600°C). The process does not require nitric acid dissolution, generates far smaller volumes of secondary liquid waste, and can be tailored to produce a mixed actinide product rather than separated plutonium, thereby strengthening proliferation resistance. Research at the Argonne National Laboratory and the Korea Atomic Energy Research Institute has demonstrated that pyroprocessing can effectively treat CANDU‑type spent fuel, recovering uranium and a transuranic mixture suitable for fabrication into metal or oxide fuel. In a closed CANDU fuel cycle, recycled actinides can be repeatedly burned in either the existing CANDU design or advanced heavy‑water‑cooled concepts that accept metallic fuel bundles. Compared with PUREX, pyroprocessing is still in a relatively early stage of technological maturity; critical issues include the corrosion behavior of structural materials in molten salts, the remote handling of solid cathode deposits, and the management of volatile fission products such as caesium and iodine. Recent breakthroughs at the Korea Atomic Energy Research Institute have demonstrated high-throughput electrorefining cells that can process up to 10 kg of heavy metal per day in a hot cell environment, and a pilot-scale pyroprocessing facility is currently under construction in Daejeon, with commissioning expected by 2026. Additionally, Canadian researchers have developed a novel solid cathode design that improves current efficiency by 20%, reducing the overall processing time per batch.

Mixed Oxide (MOX) and Other Advanced Fuel Forms

The fabrication and irradiation of MOX fuel in CANDU reactors is one of the most tangible successes of fuel recycling to date. MOX fuel blends plutonium dioxide with natural or depleted uranium dioxide to produce ceramic pellets that are geometrically identical to standard CANDU fuel. Canada has a long history of MOX research at Chalk River Laboratories, and several demonstration irradiations have confirmed that MOX bundles perform reliably under CANDU’s on‑power refueling conditions and can achieve burn‑ups comparable to natural uranium (up to 10 GWd/tU). Internationally, India has leveraged its pressurized heavy water reactors (PHWRs), which are based on the CANDU design, to irradiate MOX fuel as part of its three‑stage nuclear program, using plutonium recovered from its indigenous reprocessing facilities at Tarapur and Kalpakkam. The Canadian Nuclear Laboratories continues to explore fuel forms that incorporate recycled material, including advanced fuel cycles that pair recycled uranium with thorium, opening the door to even greater resource utilization. While MOX fuel is a mature product, its widespread adoption in CANDU fleets is limited by the availability of reprocessing infrastructure and the associated fabrication costs, which are higher than those for natural uranium bundles. However, a 2023 economic assessment by the CANDU Owners Group indicated that MOX fuel becomes competitive when plutonium is available at "zero cost" (i.e., as a byproduct from weapons dismantlement) or when a substantial carbon price is applied to the once-through cycle's waste burden. Current research focuses on adding small amounts of minor actinides to MOX pellets to improve long-term waste characteristics, with irradiation tests planned for 2025 at the NRU reactor successor facility.

Industrial Demonstration and International Collaboration

Moving from laboratory‑scale concepts to full‑scale industrial application has required sustained, multinational cooperation. South Korea’s DUPIC program, for instance, involved a consortium of domestic research institutes (including KAERI and Seoul National University) and international partners, and successfully manufactured and tested fuel elements despite the immense technical difficulty—the process required developing novel robotic systems capable of handling gamma dose rates exceeding 100 kGy/h. Canada’s federal nuclear laboratories have engaged with countries like China, Romania, and Argentina to assess the feasibility of burning recycled uranium and plutonium in their CANDU fleets. The IAEA’s regular reports on spent fuel management highlight several coordinated research projects that bring together CANDU‑operating nations to share data on fuel behavior, waste form performance, and safeguards approaches. Notable among these is the "CANDU Advanced Fuel Cycles" cooperative project, which since 2018 has involved 14 countries and has produced standardized guidelines for recycled fuel qualification. Such collaborations have accelerated the technical maturation of recycling processes but have also revealed the extent to which non‑technical factors—policy continuity, funding commitments, and alignment of regulatory frameworks—dominate the pace of progress. The recent announcement by Romania and Canada to jointly develop a pilot recycling facility for CANDU spent fuel underscores the renewed momentum in this field. The facility, expected to be operational by 2028, will demonstrate both a PUREX-based line for uranium recovery and a pyroprocessing module for higher actinides, with the aim of achieving a 30% reduction in final waste volume.

Persistent Challenges in CANDU Fuel Recycling

Despite decades of incremental progress, a series of intertwined challenges continues to prevent the full‑scale closure of the CANDU fuel cycle. These obstacles are economic, safety‑related, environmental, and societal in nature, and each must be addressed with a combination of technical innovation and institutional reform.

Economic Viability and Infrastructure Costs

The single largest barrier is the cost of building and operating the facilities necessary for reprocessing and remote fuel fabrication. A modern aqueous reprocessing plant capable of handling CANDU‑type fuel bundles would require an investment measured in billions of dollars—for example, a facility with a capacity of 100 tHM/year could cost upwards of \$5 billion when including hot cells, waste treatment, and regulatory compliance. Even a pyroprocessing facility carries a high upfront capital burden, estimated at \$2–3 billion for a fully integrated plant. When natural uranium is available at relatively low and stable prices (historically \$20–40/lb U3O8), the levelized fuel cost for a once‑through CANDU cycle remains considerably lower than that of a recycling‑based cycle, by a factor of 1.5 to 2. Without a significant increase in uranium prices (above \$150/lb), a carbon‑pricing mechanism that accounts for waste‑management burdens, or direct government subsidies, the business case for recycling struggles to attract private investment. This economic reality has repeatedly stalled projects that were technically sound on paper, such as the proposed industrial‑scale DUPIC plant in South Korea, which was shelved in 2012 despite successful demonstration. However, the recent volatility in uranium markets, combined with growing interest in spent fuel "disposal" surcharges, may gradually shift the balance. A 2024 analysis by the Nuclear Energy Agency (NEA) suggested that applying a waste-disposal fee of \$100/kgHM to the once-through cycle could make recycling cost-neutral for utilities operating CANDU fleets.

Proliferation Resistance and Security Concerns

Although processes like DUPIC and pyroprocessing are designed to never yield a purified plutonium stream, any fuel recycling operation that involves the handling of separated fissile material is subject to intense international scrutiny. The line between civil reprocessing and weapons‑usable nuclear material can appear thin, and the very existence of reprocessing infrastructure can raise regional tensions—as seen in the context of the Korean Peninsula. Robust safeguards administered by the IAEA are indispensable, but their implementation adds complexity and cost; for instance, a pyroprocessing facility requires near-continuous monitoring of material flows through accountancy measurements and containment/surveillance cameras. Countries that do not already possess comprehensive fuel‑cycle facilities often face political headwinds when attempting to establish even small‑scale reprocessing or MOX fabrication capabilities. The sensitive nature of the technology also constrains the free flow of scientific information, slowing collaborative progress. To address these concerns, the IAEA has developed specific safeguards approaches for advanced reprocessing techniques, including the use of high-resolution gamma spectrometers and neutron coincidence counters to track plutonium content without revealing process details. These measures, while effective, add an estimated 10–15% to facility operating costs. Recent developments in unattended monitoring systems, such as the "Next Generation Safeguards Initiative" from the U.S. Department of Energy, aim to reduce these incremental costs through automation and remote data analysis.

Secondary Waste Streams and Environmental Impact

Recycling does not eliminate the need for geological disposal; it transforms the waste form and alters the inventory of long‑lived radionuclides. Aqueous reprocessing generates high‑level liquid waste that must be vitrified into borosilicate glass, as well as intermediate‑level streams containing cladding hulls (zircaloy), solvent degradation products, and decommissioning materials. The vitrification process itself is energy-intensive and produces glass logs that are significantly more voluminous than the original spent fuel (by about 30% per unit of heat). Pyroprocessing produces ceramic or metal waste forms rich in fission products (such as monoclinic zirconia or glass-bonded sodalite) that remain highly radioactive and require robust isolation. The management of these secondary wastes demands its own regulatory approval pathway and adds layers of environmental monitoring. Moreover, the transport of spent fuel and recycled materials between sites increases the possibility of incidents and raises public concerns about radiation exposure. A 2021 lifecycle analysis by the University of Toronto found that while recycling can reduce the long-term heat load on a geological repository by up to 60%, it also increases the short-term radiotoxicity of process residues, requiring careful handling and storage for periods of up to 300 years before disposal. New research into "direct vitrification" of CANDU spent fuel, bypassing reprocessing entirely, has emerged as an alternative that could reduce secondary streams, though at the cost of lower resource efficiency.

Regulatory Frameworks and Public Acceptance

Even in countries with well‑established nuclear regulators, the licensing process for a first‑of‑a‑kind fuel recycling facility can take a decade or more. Regulations originally written for once‑through fuel cycles must be reinterpreted or entirely rewritten to accommodate recycled fuel with its unique radiological and chemical characteristics. For example, the Canadian Nuclear Safety Commission (CNSC) has had to develop new design safety guides for hot cells handling DUPIC powder, including requirements for containment under seismic events and maintaining subcriticality in dense fuel storage. Public opposition often crystallizes around reprocessing plants—not because of technical flaws, but due to fears rooted in the association with nuclear weapons programs or the perceived risk of severe accidents. The case of the proposed Eurodif reprocessing expansion in France in the 1990s illustrates how sustained public protest can delay projects by years, even when technical safety is well-demonstrated. Transparent communication, meaningful community engagement, and demonstrated environmental stewardship are essential but rarely sufficient to overcome deeply held skepticism in the absence of a compelling national energy policy. Some countries, like Finland, have successfully used a "decision-in-principle" process that requires parliamentary approval early in the project lifecycle, providing political legitimacy before large investments are made. Canada is now exploring a similar approach for its proposed recycling facility, with early stakeholder consultations scheduled for late 2025.

Technical Hurdles in Fuel Fabrication with Recycled Materials

The fabrication of fuel pellets from recycled powders presents profound engineering challenges. Even with remote handling, the intense gamma and neutron radiation from materials like DUPIC powder (with gamma dose rates up to 1000 Sv/h at contact and neutron fluxes exceeding 10^8 n/cm²/s) requires specialized hot cells with thick shielding (typically 1.5–2 m of concrete), remote welding, and sophisticated robotics that can operate reliably in high‑dose environments. Maintaining precise control over parameters such as oxygen‑to‑metal ratio, grain size, and impurity levels becomes exponentially more difficult when all operations are conducted behind radiation shielding. For pyroprocessing‑based fuel, the transition from electrorefined metal ingots to finished fuel rods involves processes—casting, extrusion, cladding—that have not yet been demonstrated at commercial scale with highly radioactive feedstocks. The presence of minor actinides (americium, curium) in recycled fuel also introduces alpha-particle-induced helium generation, which can cause fuel swelling and cladding stress if not properly accounted for in design. Each of these steps introduces opportunities for equipment failure, product non‑conformance, and worker dose accumulation. Recent advances in remote manipulator technology, including force-feedback teleoperation and AI-assisted vision systems, are beginning to address these issues, but the learning curve remains steep. A 2023 collaboration between CNL and a robotics firm led to the successful remote welding of a DUPIC test bundle using a laser-based system, demonstrating a 50% reduction in repair time compared to traditional methods.

Future Outlook: Closing the CANDU Fuel Cycle

Looking ahead, the trajectory of CANDU fuel recycling will be shaped by advances in several intersecting domains—reactor design, separation science, and the evolving policy landscape around nuclear energy. The following subsections outline the most promising pathways and the conditions under which they could achieve commercial deployment.

Next‑Generation Reprocessing Technologies

Current research is moving beyond the classic PUREX/pyroprocessing dichotomy toward hybrid schemes that combine the best features of both approaches. Advanced aqueous processes, such as those based on monoamide extractants (e.g., N,N-dialkylamides) or innovative chromatographic resins, aim to co‑separate uranium, plutonium, and minor actinides without generating pure plutonium, thereby enhancing proliferation resistance while reducing the long‑term heat load of waste. One notable development is the GANEX (Group ActiNide EXtraction) process, which uses a single solvent to extract all actinides together, leaving fission products in the aqueous phase. In parallel, molten salt‑based electrolytic refiners are being integrated with head‑end units that volatilize and capture tritium and iodine before they enter the salt, dramatically lowering the environmental burden. These evolving technologies are being explored in the context of the CANDU’s high neutron economy, which can efficiently burn even transuranic‑heavy fuels that would be problematic in thermal LWRs. For example, a 2024 study from the University of Cambridge demonstrated that a CANDU core loaded with 20% minor-actinide-bearing fuel could achieve a conversion ratio above 0.95, meaning it consumes only slightly more fissile material than it produces—nearly a breeder reactor in the thermal spectrum. The next step is a pilot-scale GANEX demonstration at CNL's Chalk River site, scheduled for 2027, which will process 100 kg of spent fuel and compare the waste forms to those from PUREX.

Integration with Thorium and Alternative Fuel Cycles

Thorium‑based fuel cycles have long been paired with the CANDU design, because the reactor’s heavy‑water moderator and online refueling allow it to operate as a near‑breeder in the thermal spectrum when using thorium‑plutonium or thorium‑uranium‑233 mixtures. Recycling uranium‑233 bred from thorium—a scenario that becomes more attractive when coupled with reprocessing—could multiply the effective energy extracted from a unit of natural resource by an order of magnitude. Advanced fuel concepts that co‑recycle thorium and plutonium in a single CANDU core are under computational study at institutions like the Bhabha Atomic Research Centre (BARC) in India and Chalk River in Canada. Small‑scale irradiations have shown that thorium‑based fuels behave predictably under CANDU operating conditions, with fission gas release rates similar to UO2 at equivalent burnup. The success of these cycles will ultimately depend on the availability of thorium‑compatible reprocessing flowsheets that can economically separate uranium‑233 while denaturing any fissile stream with uranium‑238 to deter misuse. A 2023 techno-economic analysis from MIT concluded that a thorium‑uranium‑233 recycle loop using a modified PUREX process could achieve a levelized fuel cost within 20% of the once-through natural uranium cycle, provided the reprocessing plant operates at a scale of at least 300 tHM/year. India has announced plans to scale up its thorium reprocessing capacity to 500 tHM/year by 2035, with a focus on PHWR applications.

Role of Small Modular Reactors (SMRs) and Advanced Reactors

The emerging class of small modular reactors, several of which are designed as heavy‑water‑cooled or molten‑salt concepts, may provide a new deployment pathway for recycled CANDU fuel. Some SMR designs explicitly target the ability to consume spent fuel from larger plants, thereby acting as "waste burners." For example, the Integral Molten Salt Reactor (IMSR) from Terrestrial Energy—a heavy-water moderated concept—can be fueled with transuranic oxides recovered from CANDU spent fuel. A fleet of CANDU‑derived SMRs co‑located with a centralized pyroprocessing facility could dramatically reduce the volume and toxicity of waste requiring final disposal. The smaller size and factory fabrication ethos of SMRs could also lower the financial threshold for fuel cycle closure, enabling incremental, modular investment rather than a single, monumental bet. A 2024 report from the International Energy Agency (IEA) noted that SMR-based recycling could achieve a 50% reduction in waste volume at a capital cost roughly half that of a conventional reprocessing plant, making it an attractive option for smaller CANDU‑operating countries. Several vendors, including NuScale and GE Hitachi, are also exploring the use of recycled CANDU-derived plutonium in their light-water SMR designs, leveraging existing supply chains.

Policy and International Cooperation Pathways

No technological breakthrough alone will unlock widespread CANDU fuel recycling; the political and institutional framework must evolve in parallel. Multinational approaches—such as regional fuel‑cycle centers under IAEA supervision—could spread costs and proliferation risks while giving smaller CANDU‑operating nations access to the benefits of recycling without embarking on sensitive national programs. The World Nuclear Association has championed the concept of "Fuel Cycle Centres" that pool reprocessing and fabrication capabilities across borders, with several proposals under discussion in Southeast Asia and Eastern Europe. Harmonized regulatory standards for recycled fuel transport, fabrication, and irradiation would reduce duplication and accelerate licensing—the International Standards Organization (ISO) has already begun work on a common standard for DUPIC fuel pellets. Finally, a durable societal consensus will require that governments, utilities, and research bodies articulate a clear, long‑term vision in which recycling is framed not as an end in itself but as a component of a broader commitment to a circular economy in nuclear energy. The recent adoption of the "Circular Economy in Nuclear Fuel Cycles" declaration at the 2023 IAEA General Conference signals that this framing is gaining traction. Canada's new "Nuclear Policy 2050" document explicitly includes a goal of achieving "closed fuel cycle readiness" by 2040, with dedicated funding of CAD 500 million for demonstration projects.

Ultimately, the technical feasibility of recycling CANDU reactor fuel has been demonstrated across multiple pathways—from PUREX and MOX to DUPIC and pyroprocessing. The challenge now is to engineer cost‑effective, proliferation‑resistant systems that can operate reliably under commercial conditions while earning public trust. Progress on these fronts will determine whether CANDU fleets can evolve from efficient consumers of natural uranium into the cornerstone of a truly sustainable nuclear fuel cycle. Success will require continued investment in research and demonstration, pragmatic policy frameworks that reward waste reduction, and a willingness to share knowledge across borders. If these conditions are met, CANDU fuel recycling could become a reality within the next two decades, transforming the economics and environmental footprint of nuclear power.