The Distinctive Engineering of CANDU Reactors

The CANDU reactor, an acronym for CANada Deuterium Uranium, represents a distinctive pressurized heavy-water reactor (PHWR) design that has supplied electricity to grids in Canada and six other countries for over half a century. Its core principle centers on using heavy water (deuterium oxide, D2O) as both the neutron moderator and the primary coolant. This choice fundamentally alters reactor physics and unlocks a fuel efficiency profile unmatched by light-water reactors (LWRs). Because heavy water absorbs far fewer neutrons than ordinary water, the neutron economy inside a CANDU core is exceptionally high. This abundance of thermal neutrons allows the reactor to sustain a fission chain reaction using natural uranium fuel that has not undergone isotopic enrichment, bypassing the entire upstream enrichment infrastructure that LWRs require. The simplicity of this fuel supply chain reduces both capital and operating costs for new builds, as enrichment facilities represent a major expense and energy sink in the LWR fuel cycle.

Using natural uranium dioxide (UO2) in collapsible pressure tubes, rather than a single large pressure vessel, defines the CANDU architecture. Each fuel channel is a separate horizontal assembly that can be accessed remotely during full-power operation. This on-power refueling capability is a cornerstone of the fleet’s fuel efficiency, enabling operators to replace small batches of fuel while the reactor remains at full electrical output, thus eliminating the scheduled refueling outages that force other reactor types offline. The continuous addition of fresh fuel and removal of spent fuel allows the reactor to operate with very low excess reactivity, which simplifies control and reduces parasitic neutron absorption in control materials. Over decades, these engineering fundamentals have been refined into a fleet that consistently achieves world-class capacity factors and fuel utilization metrics. The unique pressure tube design also localizes component aging and facilitates life extension strategies that have proven effective in recent refurbishment projects at Darlington and Bruce.

Neutron Economy and Natural Uranium Utilization

The link between heavy water moderation and fuel efficiency is best understood through neutron balance. In a light-water reactor, the moderator absorbs enough thermal neutrons that only enriched uranium—typically 3–5% uranium-235—can compensate. CANDU’s heavy water moderator absorbs roughly 1/70th as many neutrons as light water, so the reactor can maintain criticality with natural uranium containing only 0.7% U-235. The remaining uranium-238, which is fertile, captures neutrons to create plutonium-239, which fissions and contributes a growing share of total power as the fuel bundle resides in the core. This in-situ plutonium breeding can eventually supply more than half of the energy extracted from a CANDU fuel bundle, a phenomenon that dramatically improves the system’s burnup potential. Unlike LWRs that require a precise enrichment level, CANDU reactors can accommodate modest variations in incoming uranium ore grade, giving utilities greater procurement flexibility.

Burnup, measured in megawatt-days per tonne of uranium (MWd/tU), is the standard yardstick for fuel efficiency. Early CANDU units at Pickering achieved burnup levels around 7,500 MWd/tU in the 1970s, already respectable for natural uranium. Today, standard 37-element natural uranium bundles operating in the Darlington and Bruce B stations routinely exceed 8,500 MWd/tU, while advanced bundle designs have demonstrated the ability to surpass 20,000 MWd/tU in research irradiations and demonstration campaigns. The movement from a 28-element to a 37-element bundle geometry, combined with improvements in pellet density and cladding integrity, has progressively extracted more energy from each kilogram of uranium without sacrificing safety margins. The key to this progress has been the systematic reduction of parasitic neutron capture inside the core, achieved through precise control of heavy water purity and the minimization of structural materials.

Achievements in Fuel Utilization

Canadian utilities and their research partners have systematically improved CANDU fuel performance over several decades. Three headline achievements deserve particular attention because they illustrate how deep operational experience can compound into substantial gains in fuel utilization and overall plant economy.

Pushing Burnup Beyond Historical Limits

The development of the CANFLEX (CANDU Flexible Fuelling) bundle was a turning point. This 43-element bundle, developed by Atomic Energy of Canada Limited (AECL) and the Korea Atomic Energy Research Institute (KAERI), uses thinner fuel elements arranged in a configuration that reduces the maximum linear power rating and enhances coolant mixing. By lowering the peak element temperature, CANFLEX bundles can safely accommodate higher burnup without breaching fuel integrity limits. In-pile tests at the NRU reactor and subsequent commercial irradiation campaigns at Point Lepreau and Wolsong have validated burnup levels above 20,000 MWd/tU, roughly two and a half times the value typical of first-generation CANDU fuel. When combined with slightly enriched uranium (SEU) containing around 0.9–1.2% U-235, the CANFLEX bundle becomes a powerful platform for even higher discharge burnups, making it the reference fuel carrier for advanced fuel cycles. The reduced peak power density also allows operators to increase overall reactor power output without exceeding safety limits, a feature exploited in some units to boost capacity.

Extended Fuel Cycles and Capacity Factor Gains

On-power refueling inherently enables CANDU reactors to avoid cycle-length constraints. Instead of operating in fixed 12- to 24-month cycles, a CANDU unit undergoes hundreds of small refueling operations each year, with one or two bundles replaced per channel pass. The ability to tailor the refueling schedule to the actual burnup state of each channel has allowed utilities to extend the effective “fuel cycle” indefinitely, limited only by the need for periodic maintenance of turbines and balance-of-plant. The result is a fleet that regularly achieves annual capacity factors above 90%, with Bruce Power’s units frequently exceeding 95%. This operational pattern reduces the per-kilowatt-hour impact of fixed capital costs and directly improves the lifetime economics of the plant. Furthermore, continuous refueling provides opportunities to flatten the axial power distribution, reducing the likelihood of localized fuel failures and extending the life of pressure tube materials.

Minimizing Spent Fuel Volume

Fuel efficiency is also measured in waste arisings. Because CANDU reactors extract more thermal energy from each tonne of uranium than LWRs do from enriched fuel (after accounting for enrichment tails assay), they produce considerably less spent fuel per terawatt-hour of electricity generated. A typical 700 MW CANDU 6 unit discharges roughly 100–130 tonnes of spent fuel per year, while an equivalent LWR of similar output would discharge a larger mass of spent enriched fuel, albeit with a different isotopic composition. This lower spent fuel volume simplifies interim storage at station sites and reduces the long-term burden on any future geological repository. The Nuclear Waste Management Organization (NWMO) in Canada has repeatedly highlighted the compactness of CANDU used fuel as an advantage in repository design, and the planned deep geological repository benefits from the smaller footprint of CANDU waste packages.

The CANDU Owners Group (COG) has compiled extensive operational data confirming that fuel defect rates in the Canadian fleet remain below 0.01%, a world-leading metric that directly supports high burnup ambitions. Such low failure rates are the product of rigorous quality control during bundle manufacturing, improvements in cladding metallurgy (Zircaloy-4 and newer alloys), and conservative operational management. Together, these achievements provide a robust foundation for the next generation of fuel innovation. The feedback loop between operational data and fuel design has been critical: failure modes identified in early bundles led to changes in pellet geometry, end-cap welding techniques, and bearing pad designs that are now standard across the fleet.

Future Roadmap: Fuels, Materials, and Digital Integration

The Canadian nuclear industry, supported by Natural Resources Canada and the broader international community, is actively pursuing several technology streams designed to further improve fuel efficiency, sustainability, and safety. The following goals define the near- to medium-term evolution of CANDU fuel cycles, focusing on incremental improvements that can be introduced without major reactor modifications while also exploring transformative longer-term options.

Advanced Fuel Cycles: MOX, DUPIC, and Thorium

One of the most promising near-term strategies is the introduction of mixed-oxide (MOX) fuel, in which plutonium recovered from spent LWR fuel is blended with natural or depleted uranium oxide. CANDU’s flexible neutron balance makes it an ideal partner reactor for plutonium disposition, as it can achieve high burnup with MOX without requiring major hardware modifications. The DUPIC (Direct Use of spent PWR fuel in CANDU) cycle, pioneered by KAERI in collaboration with Canada, goes further. DUPIC employs a dry thermal process to convert spent PWR fuel assemblies directly into CANDU-compatible pellets, avoiding chemical separation of plutonium and addressing proliferation concerns. Irradiation tests have shown that DUPIC fuel bundles can achieve burnup levels comparable to natural uranium while effectively destroying long-lived actinides. Full-scale deployment would transform CANDU reactors into an integral component of a symbiotic LWR-CANDU fuel cycle, significantly reducing the total volume and radiotoxicity of high-level waste.

Parallel investigations into thorium-based fuels are underway. Thorium-232 is fertile, and when irradiated in a CANDU core it captures neutrons to produce fissile uranium-233. A thorium-uranium or thorium-plutonium blend could further improve sustainability. Canada’s extensive thorium reserves, documented by the International Atomic Energy Agency (IAEA), provide a strategic rationale for this research. Reactor physics simulations indicate that a seed-blanket configuration, with a central driver region of enriched uranium or plutonium surrounded by thorium bundles, could boost burnup while generating enough U-233 to partially or fully replace the initial fissile inventory over time. The potential to achieve a near-breeder fuel cycle in a pressure tube reactor could significantly extend global uranium resources and reduce long-term waste radiotoxicity.

Accident-Tolerant Fuel and Advanced Cladding

Higher burnup and advanced fuel compositions place greater demands on the cladding that forms the primary barrier against fission product release. Traditional Zircaloy-4 has performed well, but its oxidation resistance at elevated temperatures—especially under beyond-design-basis conditions—can be improved. Research efforts within the Advanced Fuel CANDU Reactor (AFCR) project and the broader international accident-tolerant fuel (ATF) community are focusing on cladding materials such as silicon-carbide composites (SiC/SiC), iron-chromium-aluminum (FeCrAl) alloys, and coated zirconium-based concepts. These materials offer significantly slower oxidation kinetics in high-temperature steam, maintaining structural integrity for longer periods during accidents. For CANDU, ATF cladding adoption would directly widen the safety envelope for higher linear heat rates and enable safe achievement of burnup targets beyond 30,000 MWd/tU with advanced bundle geometries. In addition to improved accident performance, ATF cladding can reduce hydrogen pickup, a known aging mechanism for current pressure tube materials.

Digital Twins and Real-Time Fuel Performance Optimization

The fleet is increasingly moving toward a digitally-enabled operating model where every fuel bundle’s irradiation history is tracked by high-fidelity core-following software. Modern codes such as RFSP (Reactor Fuelling Simulation Program) and advanced coupled neutronics-thermalhydraulics solvers, validated against billions of hours of operation, allow station physicists to predict the burnup state of every fuel channel in near real time. By coupling these simulations to digital twin platforms, operators can optimize refueling schedules to flatten the channel power distribution, reduce peak channel power, and extend the interval between refueling visits. Candu Energy Inc., a member of the SNC-Lavalin Group, has demonstrated that such tools can increase average discharge burnup by several percent without any hardware changes, simply by optimizing the sequence and timing of fuel pushes. The integration of machine learning algorithms trained on historical operational data promises to further refine these strategies, detecting subtle patterns that human planners might overlook. For example, predictive models that account for channel-specific corrosion rates can help schedule refueling to minimize the risk of pressure tube leaks.

Small Modular Reactors and the CANDU Legacy

The fuel efficiency lessons learned from the large-scale CANDU fleet are being directly transplanted into the next generation of small modular reactors (SMRs) that build on heavy-water technology. Concepts such as the Enhanced CANDU 6 (EC6) and various advanced PHWR designs under development in Canada and abroad retain the core principle of on-power refueling and natural uranium capability, but incorporate modern passive safety systems, compact steam generators, and design-for-decommissioning features. The EC6, for instance, targets a 60-year plant life with a design burnup of up to 8,500 MWd/tU using standard natural uranium bundles and the flexibility to adopt CANFLEX and SEU without licensing modifications. Even in a small modular package, the neutron economy of heavy water ensures that fuel efficiency remains a defining competitive advantage. Smaller units also benefit from a simpler fuel supply chain, as they do not require enriched uranium—an advantage for countries with limited enrichment infrastructure or non-proliferation commitments.

Economic and Environmental Implications

Continuous improvements in fuel efficiency resonate through both the balance sheet and the environmental footprint of nuclear power plants. Higher burnup means each tonne of mined uranium produces more electricity, reducing the upstream uranium mining and milling impacts per megawatt-hour. According to the Canadian Nuclear Safety Commission (CNSC), the greenhouse gas emissions of nuclear electricity are already among the lowest of any available generation source—approximately 4–6 grams of CO2-equivalent per kilowatt-hour—and improved fuel utilization can chip away at the mining and transport components of that figure. The emissions from uranium mining and milling account for a small but non-negligible portion of the lifecycle carbon footprint; each incremental burnup improvement reduces that portion further.

Economically, the ability to stretch fuel resources reduces the variable cost of generation and shields operators from volatility in uranium spot prices. For jurisdictions that import uranium, a 10% increase in average discharge burnup translates into a 10% reduction in annual fresh fuel purchases, saving millions of dollars per reactor-year. The deployment of DUPIC or MOX cycles would further decouple fuel costs from virgin uranium mining, offering a more stable economic profile. In refurbished reactors like Darlington and Bruce, the adoption of 37-element or CANFLEX bundles with slightly higher U-235 content can extend the period between major outages, improving revenue predictability. Moreover, reduced spent fuel volume directly translates into lower costs for interim storage and eventual disposal. A study by the OECD Nuclear Energy Agency found that a 20% improvement in burnup could reduce the total back-end fuel cycle cost by roughly 15–20%.

International Collaboration and Regulatory Progress

Advancing fuel efficiency goals relies on sustained international cooperation. The CNSC has established a licensing framework for advanced fuel forms and higher burnup envelopes, conducting safety reviews that encompass normal operation, transient conditions, and severe accidents. Collaborative projects through the IAEA’s Coordinated Research Programme (CRP) on PHWR fuel performance provide a forum for sharing irradiation data between Canadian operators, Argentina, China, India, Pakistan, Romania, and South Korea. These joint efforts have produced open-access databases on fuel defect mechanisms, fuel-cladding interaction, and fission gas release, accelerating the qualification of novel fuel designs without requiring each country to repeat expensive integral tests. The IAEA’s Coordinated Research Projects have been instrumental in harmonizing experimental protocols and creating a common language for fuel performance modeling.

Standards developed by the CSA Group for fuel design and manufacturing (such as CSA N285.0 and N287 series) are being updated to incorporate higher burnup limits and advanced cladding materials, ensuring that code predictions remain conservative and that the safety case is built on robust experimental evidence. The alignment of regulatory frameworks across member states of the CANDU community is expected to shorten the time-to-market for ATF concepts and MOX-based cycles. Additionally, the Nuclear Energy Agency (NEA) has facilitated several benchmarking exercises that compare computer codes for CANDU fuel performance, building confidence in the predictive capabilities needed for high-burnup licensing.

Overcoming Remaining Challenges

While the trajectory is promising, several technical and economic hurdles remain. Prolonged exposure to high neutron fluxes and aggressive coolant chemistry can accelerate cladding corrosion and hydrogen pickup, potentially limiting the practical burnup ceiling unless new alloys or coatings are deployed. The production scale-up of CANFLEX and ATF bundles requires investment in manufacturing facilities and quality verification infrastructure. Moreover, commercial adoption of DUPIC fuel hinges on demonstrating a fully licensed and proliferation-resistant remote fabrication facility, a multi-billion-dollar undertaking that no single operator can fund alone. Public-private partnerships and government support, such as those outlined in Canada’s Strategic Innovation Fund, are essential to bridge the gap between laboratory success and fleet-wide deployment.

Another challenge lies in managing increased fission gas release at higher burnups. As burnup increases, the volume of fission gases trapped within the fuel pellet rises, increasing internal rod pressure and stressing the cladding. Innovative fuel designs, such as annular pellets and burnable neutron absorbers, are being explored to mitigate this issue. The long-term irradiation behavior of advanced cladding materials must be validated under prototypical conditions, including the effects of irradiation creep and growth. Despite these challenges, the foundational physics of heavy-water moderation ensures that the CANDU platform remains uniquely suited to absorb these advances. The design’s inherent tolerance for a wide range of fissile materials, combined with its proven high availability, creates a compelling case for continued investment in fuel efficiency improvements. As global energy systems seek to decarbonize while maintaining grid stability, the ability to extract more clean energy from each kilogram of uranium will only grow in significance. The continuous evolution of CANDU fuel technology stands as a testament to the value of incremental innovation built on a robust scientific foundation.