CANDU Reactors and the Decentralization of Nuclear Power Generation

The global energy landscape is shifting toward systems that are more resilient, distributed, and aligned with local economic and environmental priorities. Within this transformation, nuclear power is being reimagined not as a monolithic, centrally planned asset but as a flexible component of a diversified energy portfolio. CANDU reactors, originally developed in Canada, have long embodied principles that are now central to the concept of decentralized nuclear generation. Their unique heavy-water design, modular architecture, and fuel-cycle versatility offer lessons and capabilities that extend far beyond Canada's borders, providing a blueprint for how nuclear technology can adapt to a more distributed energy future.

The Unique Engineering of CANDU Reactors

CANDU stands for Canada Deuterium Uranium, and its defining characteristic is the use of heavy water (deuterium oxide) as both moderator and primary coolant. Unlike light-water reactors that require enriched uranium, CANDU units operate on natural uranium, a fuel that does not need to pass through enrichment facilities. This fundamental difference shapes everything from fuel supply chains to plant construction and operational safety, and it creates a technology platform that is inherently suited to decentralized deployment.

Heavy Water as the Key Differentiator

Heavy water has an extremely low neutron absorption cross-section, which means it moderates neutrons efficiently without stealing them from the fission chain reaction. As a result, a CANDU reactor can sustain criticality using unenriched uranium, which eliminates the need for a domestic enrichment industry. For many nations adopting nuclear power, this factor has proven decisive: they can advance their programs while adhering to international non-proliferation norms. The heavy water moderator is kept cool and separate from the primary coolant loop, which adds a distinct thermal buffer that contributes to plant safety and enables a high neutron economy.

The production of heavy water itself is an energy-intensive process, typically using a combination of the Girdler sulfide (GS) process or later electrolytic methods. Canada has historically been a world leader in heavy water production, with plants at Bruce, La Prade, and Port Hawkesbury. Although some production facilities have been mothballed, Canada retains the technical capability to restart or scale up production if new CANDU builds require it. This domestic capacity reduces supply chain vulnerabilities and ensures that countries deploying CANDU technology can rely on a stable source of heavy water, further reinforcing the decentralized model where fuel and moderators are sourced from a trusted partner.

Modular Pressure Tubes and Scalability

Instead of a single large pressure vessel, each CANDU reactor contains hundreds of horizontal pressure tubes that house the fuel bundles and transport the high-temperature heavy-water coolant. This tube-based design is inherently modular: individual channels can be inspected, maintained, or replaced without an outage of the entire core. From a manufacturing standpoint, the pressure tubes and associated components are produced in factories and shipped to the site, enabling a construction approach that parallels modern small modular reactor (SMR) philosophies.

While typical CANDU units have an electrical output around 700 megawatts, the pressure-tube concept can be scaled up or down. The Enhanced CANDU 6 (EC6), for example, refines this 700 MWe class with modern safety systems, while earlier design studies explored smaller variants suitable for grids with more modest demand. This scalability means that a utility can deploy a single unit in a region with moderate load growth or cluster multiple units at one site for larger markets. The modular design also reduces on-site construction time because pressure tubes, headers, and feeders are prefabricated and assembled in sequence rather than requiring massive field-welded vessels. In a decentralized network, this standardization allows multiple sites to share a common supply chain and workforce training, driving down costs through repetition.

Online Refueling and Operational Efficiency

One of the most distinctive features of CANDU reactors is online refueling. Remotely operated fueling machines push fresh fuel bundles into one end of a pressure tube while receiving spent fuel from the other, all while the reactor remains at full power. This avoids the extended refueling outages typical of light-water reactors, boosting capacity factors consistently above 90 percent. For a decentralized grid where multiple smaller plants may share the load, high and predictable availability is a powerful asset. It also means that fuel can be shuffled continuously to optimize burnup, and defected bundles can be removed individually without shutting down the unit, which minimizes unplanned downtime.

The fueling machines themselves are complex but proven systems that operate under computer control. They travel on a movable gantry on the reactor face, engaging with each pressure tube in sequence. This approach creates a highly flexible fuel cycle: bundles with different burnups or compositions can be placed strategically throughout the core to flatten the power distribution and extend fuel life. Operators can also insert experimental fuel bundles on a small scale without affecting the rest of the core, which is valuable for testing new fuel concepts such as thorium-based or recycled fuels.

Decentralized Nuclear Power: A Paradigm Shift

Traditional nuclear power deployment has often relied on a small number of large, centralized stations tied to major transmission corridors. Decentralization, by contrast, distributes generation assets across more sites, each connected closer to load centers. CANDU technology facilitates this model in several practical ways that go beyond mere plant size.

First, the pressure-tube architecture enables a plant to be built in stages or replicated across locations with a high degree of design standardization. A country or province can deploy multiple EC6 units in different regions, each operating independently yet sharing a common supply chain, workforce training, and regulatory framework. This spreads risk: a maintenance event or an extreme weather impact at one site does not threaten a large fraction of total grid capacity. In Ontario, Canada's nuclear fleet — consisting of 18 CANDU reactors across the Bruce, Darlington, and Pickering stations — already functions as a distributed backbone, supplying roughly 60 percent of the province's electricity without relying on a single geographic point of failure. The stations are spaced along the Great Lakes shoreline, each with its own transmission corridors, creating a resilient network that has withstood ice storms, heat waves, and transmission line outages.

Second, smaller and more numerous generating stations reduce the need for extensive high-voltage transmission. By locating a CANDU plant nearer to industrial parks or population centers, grid losses are cut, and local voltage support improves. This also aligns with the growth of renewable energy, where nuclear provides steady baseload power to balance variable solar and wind production. A decentralized nuclear fleet can be dispatched in harmony with regional renewable portfolios, enhancing overall system resilience. For example, a 700 MWe CANDU unit can be sited near a wind farm cluster, allowing the nuclear plant to back down during high-wind periods while maintaining grid stability.

Third, decentralized deployment creates economic spillovers across multiple communities. Rather than one massive construction project concentrating jobs temporarily, a series of modular CANDU projects can sustain manufacturing, engineering, and skilled trades employment over decades. The localization of fuel fabrication and heavy-water production further distributes the economic benefits. In Ontario, the nuclear supply chain supports thousands of jobs across the province, from fuel bundle manufacturing in Toronto to pressure tube production in Mississauga and tooling in smaller towns. This distributed economic impact strengthens political support for nuclear energy and reduces the social disruption associated with boom-and-bust construction cycles.

Grid Integration and Load Following

While CANDU reactors have traditionally operated as baseload plants, modern control systems allow them to participate in load-following operations within limits. The online refueling capability means that fuel burnup can be adjusted on the fly by modifying the refueling rate, but the heavy-water inventory and thermal inertia limit rapid power changes. Improvements in digital control and regulatory allowances have enabled CANDU units to reduce power to 60 percent of full output for several hours per day, which is valuable for grids with high renewable penetration. As decentralized grids become more complex, this flexibility enhances the value of a CANDU fleet.

Global Deployment and Proven Adaptability

CANDU reactors have been exported and operated successfully on several continents, demonstrating that a heavy-water design can meet diverse regulatory, climatic, and seismic conditions. Each deployment has contributed to a growing body of operational experience that supports new builds in a wide range of settings.

  • South Korea: The Wolsong Nuclear Power Plant hosts four CANDU units that have provided reliable electricity for decades. The Korean nuclear industry's experience with CANDU technology contributed to its domestic pressurized heavy-water reactor designs and enriched the country's overall nuclear engineering expertise, now embodied in its APR-1400 exports. Korean engineers have also collaborated on DUPIC fuel research, leveraging CANDU's fuel cycle flexibility.
  • Romania: The Cernavodă Nuclear Power Plant operates two CANDU 6 reactors, with plans to complete two additional units. This project has become a cornerstone of Romania's energy independence and a case study in international cooperation, combining Canadian technology with European financing and local workforce development. The plant's performance has met or exceeded expectations, with capacity factors regularly above 90 percent.
  • China: The Qinshan Phase III station comprises two CANDU 6 units that were built on schedule and have achieved strong operational performance. They introduced heavy-water reactor technology to China's emerging nuclear fleet and provided an alternative to the dominant pressurized water reactor designs. The project also facilitated technology transfer, with Chinese engineers gaining expertise in heavy-water technology that could inform future domestic designs.
  • Argentina: The Embalse nuclear power plant, a single CANDU 6 unit, has operated since the 1980s and underwent a successful life extension that added 30 years to its design life. This mid-life refurbishment program replaced pressure tubes, feeders, and steam generators, demonstrating that CANDU units can be economically revitalized after decades of service.
  • India: India's pressurized heavy-water reactor program, based on earlier CANDU cooperation, now encompasses a large fleet of indigenous PHWRs, underscoring how the CANDU approach can be adapted and standardized for a nation's unique energy growth trajectory. Indian PHWRs have evolved to use thorium fuel in some units, building on CANDU's fuel cycle versatility.

In each case, the use of natural uranium removed the need for enrichment services, simplifying fuel acquisition and allowing these nations to advance their nuclear programs while remaining compliant with international safeguards administered by the International Atomic Energy Agency. The openness of the pressure-tube design, with its many individual channels, also facilitates sophisticated inspection and monitoring, which enhances transparency and builds trust with regulators and the public.

Safety, Sustainability, and Fuel Cycle Flexibility

The safety case for CANDU reactors rests on several built-in characteristics that have been validated through decades of operation and analysis. The large volume of heavy water inside the calandria provides an enormous heat sink that slows temperature transients. In the event of a loss-of-coolant accident, the moderator remains low-pressure and can perform an emergency cooling function even if the primary cooling circuit is compromised. The horizontal orientation of the pressure tubes ensures that the core geometry is maintained, and natural circulation of the moderator can remove decay heat passively under many scenarios. Each pressure tube is surrounded by a gas annulus and the calandria tube, creating a robust defence in depth barrier.

Beyond safety, CANDU technology offers unusual fuel cycle flexibility. Because the neutron economy is so high, the reactor can consume alternative fuels that would otherwise require complex processing. Natural uranium can be replaced or supplemented with recycled uranium from light-water reactor spent fuel, a direct use of spent PWR fuel in CANDU known as the DUPIC (Direct Use of Spent PWR Fuel in CANDU) concept. DUPIC fuel fabrication involves shearing, milling, and pelletizing spent PWR fuel without separating plutonium, which avoids proliferation concerns. Research at the Korea Atomic Energy Research Institute and Canadian laboratories has shown that DUPIC fuel can achieve acceptable burnup in CANDU reactors, reducing the volume of high-level waste by up to 70 percent compared to direct disposal of PWR spent fuel.

Thorium, a more abundant fertile material, can also be introduced into CANDU fuel bundles. The high neutron economy of heavy water allows thorium to be converted to fissile uranium-233 in situ, potentially extending fuel resources for thousands of years. India has already demonstrated the use of thorium in its PHWRs, and Canadian research has explored thorium fuel cycles for CANDU. While challenges remain in reprocessing and fuel fabrication, the option to shift toward thorium adds strategic depth to the long-term sustainability of CANDU reactors.

The production of heavy water does create tritium as a byproduct within the moderator and coolant systems. CANDU operators manage tritium through dedicated detritiation facilities that remove and safely store it, with ongoing innovation to minimize releases. The Tritium Removal Facility at Darlington, for example, uses cryogenic distillation to extract tritium from heavy water, producing a high-purity product that can be stored in metal hydrides. Public perception of tritium remains a challenge for all heavy-water reactor operators, yet independent environmental monitoring has consistently shown that emissions are managed well within regulatory limits. The Canadian Nuclear Safety Commission publishes annual reports on tritium releases, confirming that doses to the public are a small fraction of the regulatory limit.

Economic and Policy Dimensions

The economics of CANDU reactors deviate from those of light-water reactors in important ways. Because fuel enrichment is unnecessary, the fuel fabrication chain is simpler and, for many nations, cheaper per unit of energy produced. Natural uranium fuel pellets are pressed and sintered directly from uranium dioxide powder, then loaded into zircaloy tubes to form bundles. This process can be performed in relatively modest facilities, reducing capital costs for a domestic fuel supply. The offset is the initial investment in heavy water, which is expensive to produce. A 700 MWe CANDU 6 requires roughly 265 tonnes of heavy water, a significant capital outlay that can be recovered over the plant's long operational life through reduced fuel costs. Canada maintains domestic heavy-water production capability, which lowers the supply risk for new builds and supports an indigenous nuclear supply chain.

From a policy perspective, CANDU exports have long been a vehicle for peaceful nuclear cooperation. Canada's bilateral nuclear cooperation agreements require recipient countries to adopt comprehensive safeguards and forgo nuclear explosive devices, aligning with global non-proliferation objectives. The absence of enrichment infrastructure on the recipient's side naturally reinforces these goals. The modular and distributed nature of CANDU fleets fits squarely into contemporary energy policies that prize regional development and climate resilience. As governments look to replace coal-fired plants with clean baseload power, the ability to site a standard 700 MWe unit on the ash pond of a retired coal plant — with minimal transmission upgrades — offers a compelling transition pathway.

Policy makers also recognize the value of nuclear energy in meeting decarbonization targets. The Canadian government has included nuclear in its net-zero by 2050 strategy, and provinces like Ontario are proceeding with refurbishment of existing CANDU units at Bruce and Darlington. The financial models for these projects demonstrate that a fleet of standardized, distributed CANDU reactors can be cost-competitive with combined-cycle gas plants when carbon pricing is factored in, especially given their long operating lives and high capacity factors.

The Future: Modern CANDU Concepts and Distributed Generation

The nuclear industry is evolving rapidly, and Candu Energy Inc., a subsidiary of AtkinsRéalis, continues to refine the technology. The EC6 design incorporates enhanced passive safety features, a 60-year design life, and digital control systems that align with the latest Generation III+ standards. While not a small reactor by SMR definitions, the EC6's standardized, factory-fabricated components allow serial construction, which can reduce project timelines and costs through repetition. The approach retains the hallmark CANDU attributes — natural uranium fuel, online refueling, and modular pressure tubes — while modernizing the auxiliary systems, steam turbines, and instrumentation.

Digitalization of control systems has improved operator interface and diagnostic capabilities. Modern CANDU plants use distributed control systems with redundant safety channels, providing better monitoring of individual fuel channels. Advanced analytics can predict fuel bundle performance and optimize refueling schedules, further enhancing capacity factors. These digital tools also support predictive maintenance, reducing forced outage rates and improving overall plant economics.

Farther into the future, Canada's SMR Action Plan provides a framework for deploying new reactor technologies, including those informed by heavy-water experience. Though the first wave of Canadian SMRs is based on light-water or advanced non-water technologies, the pressure-tube concept remains relevant. A small, CANDU-derived reactor could serve remote communities, mining operations, or off-grid industrial sites, exactly the settings where decentralization yields the greatest benefits. The institutional knowledge gathered from decades of CANDU operation and maintenance will be valuable for any future heavy-water SMR, particularly in areas such as online refueling, tritium management, and neutron economy optimization.

Challenges and a Realistic Appraisal

No nuclear technology is without challenges, and CANDU is no exception. The heavy-water inventory represents a significant upfront financial and environmental investment. Tritium management, while technologically mature, adds regulatory and public-communication burdens. The positive void coefficient of reactivity, present in many pressure-tube designs, requires careful engineered safeguards, though CANDU's independent shutdown systems and large moderator mass have proven effective in mitigating this feature. Construction costs, like those of all large nuclear projects, remain subject to supply-chain and labour-market fluctuations. The operational track record of CANDU reactors is strong: consistently high capacity factors, accepted safety performance, and a design life that can be extended past 60 years through mid-life refurbishment programs, as demonstrated at Embalse and Point Lepreau.

Another challenge is the aging workforce in nuclear engineering. As experienced CANDU operators and engineers retire, utilities must invest in knowledge transfer and training. The modular, distributed nature of the fleet helps here because multiple sites can share training simulators and resources. Industry partnerships with universities, such as the University of Ontario Institute of Technology's nuclear engineering program, ensure a pipeline of new talent familiar with CANDU technology.

An Enduring Blueprint for Distributed Nuclear Power

CANDU reactors have demonstrated that nuclear power does not have to be synonymous with gigawatt-scale, one-of-a-kind megaprojects. The use of natural uranium, the modular pressure-tube core, and the ability to refuel online create a technology platform that inherently supports the decentralization of generation. By enabling multiple standardized units to be distributed across regions, CANDU reduces single-point failure risks, strengthens local economies, and extends the benefits of nuclear energy to grids that would otherwise be too small to justify a large plant.

As the global conversation around clean energy intensifies, the lessons embedded in the CANDU experience — modular construction, fuel cycle versatility, and resilient, distributed operation — continue to influence how nations envision their nuclear future. Whether through the EC6, the potential for small heavy-water reactors, or the application of CANDU principles to next-generation designs, this uniquely Canadian innovation remains a vital part of the world's nuclear toolkit. The path forward is not about replicating the large centralized plants of the past, but about deploying flexible, safe, and efficient reactors that are closer to the people they serve. CANDU's legacy is that it has already paved that path.