The global nuclear energy landscape is shaped by a variety of reactor designs, each engineered to balance safety, efficiency, and resource availability. Among the most deployed power reactor types, the CANada Deuterium Uranium (CANDU) and the Pressurized Water Reactor (PWR) represent two distinct technological paths that have been refined over decades. While both harness the heat of nuclear fission to produce steam for electricity generation, they differ profoundly in their choice of moderator, coolant, fuel cycle, and operational philosophy. This article examines the engineering principles, operational characteristics, safety features, and economic considerations that set CANDU and PWR reactors apart, providing a comprehensive comparison for educators, students, and energy professionals.

The Fundamentals of Nuclear Fission Energy

At the heart of any thermal nuclear reactor is a controlled fission chain reaction. Heavy atomic nuclei, typically uranium-235 or plutonium-239, absorb a neutron and split into lighter fragments, releasing immense energy, additional neutrons, and radioactive fission products. The energy is carried away by a coolant and ultimately converted into electricity via a steam turbine. One of the most critical design choices is the moderator—a material that slows fast neutrons to thermal energies where they are much more likely to induce further fissions. Without moderation, the chain reaction would fail to sustain itself in a natural or low-enrichment fuel core. The moderator’s neutron-scattering properties fundamentally influence the fuel requirements, core geometry, and safety behavior of the reactor.

Coolants serve a dual role: they remove heat from the reactor core and, in some designs, also participate in neutron moderation. The interplay between moderator and coolant defines the reactor’s neutron economy—the balance of neutron production and loss. A high neutron economy allows a reactor to operate on fuels with low fissile content, such as natural uranium, while a lower neutron economy demands enriched uranium to compensate for parasitic absorptions. The CANDU and PWR are textbook examples of how this neutron balance drives radically different engineering solutions. In the broader context of clean energy, both reactors provide firm, dispatchable baseload power with near-zero greenhouse gas emissions during operation, a key advantage as countries decarbonize their electricity grids.

Inside the CANDU Reactor: Design and Operation

The CANDU reactor, pioneered by Atomic Energy of Canada Limited (AECL) in the 1950s and 1960s, is a pressurized heavy-water reactor that uses deuterium oxide (D2O) as both moderator and primary coolant. Its most distinctive feature is the horizontal fuel channel design. Instead of a single large pressure vessel, a CANDU calandria—a large cylindrical tank containing the heavy-water moderator at near-atmospheric pressure and relatively low temperature (about 70 °C)—houses several hundred horizontal pressure tubes. A typical 700 MWe CANDU-6 unit contains 380 to 480 such tubes, each about 6 meters long. Each pressure tube contains a string of 12 fuel bundles and is surrounded by an outer calandria tube, with an insulating gas annulus between them to keep the moderator cool while the fuel is cooled by the pressurized heavy-water coolant flowing through the tubes. This arrangement allows the moderator to remain separate from the coolant, a crucial safety advantage that provides a large, passive heat sink.

Fuel for a CANDU is natural uranium dioxide (UO2) in the form of short bundles about 0.5 meters long, each containing 28 or 37 elements. Because heavy water absorbs very few neutrons compared to ordinary water, CANDU’s neutron economy is exceptionally high. This enables the reactor to sustain a chain reaction in unenriched uranium—a fuel that would be impossible to use in a light-water reactor. The horizontal, individual fuel channels also make on-power refueling feasible. Two robotic fueling machines latch onto opposite ends of a fuel channel; one pushes fresh bundles in while the other receives spent bundles, all while the reactor remains at full power. This capability allows a CANDU to reach very high capacity factors, often exceeding 90%, because shutdowns for refueling are eliminated. The Canadian fleet at Bruce, Darlington, and Pickering routinely post capacity factors above 85% as reported by the Canadian Nuclear Safety Commission.

Heavy water is expensive and requires a capital-intensive production infrastructure—the cost of D2O can exceed $300 million for a single plant. However, the simplicity of natural uranium fuel and the ability to use different fuel cycles—including thorium, recovered uranium from reprocessing, or even spent LWR fuel directly—provides strategic flexibility. Several countries, including Canada, South Korea, Romania, Argentina, and previously Pakistan, adopted CANDU-based designs, often valuing the independence from foreign enrichment services. India’s pressurized heavy-water reactors (PHWRs), which evolved from CANDU technology, now form the backbone of its nuclear program with over 20 units in operation.

Inside the Pressurized Water Reactor: Design and Operation

Pressurized Water Reactors are the most widely deployed nuclear reactor type, accounting for roughly two-thirds of the world’s operating nuclear capacity. The PWR design originated in the United States as a naval propulsion technology and was later scaled up for commercial electricity generation by Westinghouse and other vendors. In a PWR, ordinary (light) water serves as both coolant and moderator, but it is maintained at high pressure—typically around 15.5 MPa (2,250 psi)—to keep it from boiling at the operating temperatures of approximately 300–325 °C. A typical 1000 MWe PWR core contains between 150 and 250 fuel assemblies, each consisting of a square array of fuel rods containing enriched uranium dioxide pellets (usually with uranium-235 concentrations between 3% and 5%).

Because light water absorbs more neutrons than heavy water, enrichment is mandatory to maintain criticality. The high pressure of the primary coolant loop transfers heat from the core to steam generators, which are large heat exchangers. In the steam generators, heat from the primary water is passed to a secondary loop where water boils at lower pressure, producing the steam that drives the turbine. The primary and secondary loops remain physically separated, which confines radioactive materials to the primary circuit and reduces contamination in the turbine and condenser. Modern PWRs, such as the Westinghouse AP1000 and the EPR by EDF, have increased thermal efficiency and added passive safety features.

PWRs have evolved through multiple generations, with standardized designs, well-characterized materials, and extensive operational experience. Their safety approach heavily relies on redundancy and diversity: multiple emergency core cooling systems, large containment structures, and passive or active decay heat removal features. Refueling typically occurs every 12 to 24 months during planned outages, during which one-third to one-half of the core is replaced. This batch refueling requires the reactor to be shut down and its pressure vessel lid unbolted, adding to outage duration but allowing for integrated maintenance. The average capacity factor for the global PWR fleet has steadily improved, exceeding 90% in many countries for the past decade according to IAEA Power Reactor Information System data.

Comparing Moderators and Coolants: Heavy Water vs. Light Water

The choice between heavy water and light water is the most fundamental differentiator between CANDU and PWR. Heavy water’s low neutron absorption grants CANDU a vast advantage in neutron economy, which directly enables natural uranium fuel and exceptional fuel utilization. According to World Nuclear Association data, a CANDU reactor extracts about 30% more energy per tonne of mined uranium than a typical PWR, partly because of its ability to use fuel that has already been discharged from light-water reactors. However, the low neutron absorption comes at a cost: heavy water production, maintenance of high isotopic purity (typically >99.75% D2O), and the need to minimize leaks are significant operational burdens. Tritium, a radioactive hydrogen isotope with a 12.3-year half-life, is produced in heavy water by neutron capture on deuterium, posing additional radiological hazards and requiring careful management through detritiation systems.

In contrast, light water is inexpensive, widely available, and presents fewer tritium control challenges under normal operation. PWRs achieve acceptable neutron economy by enriching uranium, relying on an established enrichment industry that has grown technologically diverse—from gaseous diffusion and centrifuge enrichment to laser-based methods like SILEX. The compressed pressure boundary in a PWR’s primary circuit demands thick-walled components and sophisticated metallurgy, particularly to address boric acid corrosion and primary water stress corrosion cracking over decades of service. The use of soluble boron in PWR coolant for reactivity control adds complexity in chemistry management and waste treatment, whereas CANDU relies more on mechanical control absorbers and adjuster rods with minimal chemical shim.

Fuel Cycles: Natural Uranium vs. Enriched Uranium

The fuel cycle implications are far-reaching. CANDU’s natural uranium fuel simplifies the front end: no enrichment plants are needed, reducing capital and proliferation-sensitive infrastructure. Mining and milling uranium ore, converting it to UO2, and manufacturing bundles are sufficient. This has historically attracted nations seeking energy independence. Additionally, the CANDU fuel cycle can adapt to alternative fissile sources, including mixed oxide (MOX) fuel, recovered uranium from reprocessing, and thorium-based cycles. Direct fuel fabrication for CANDU is simpler due to the short bundle geometry and lower burnup compared to PWR fuel. The DUPIC (Direct Use of spent PWR fuel In CANDU) process is particularly attractive, enabling reuse of spent LWR fuel without reprocessing, thereby reducing waste volumes and proliferation risks.

PWR fuel requires enriched uranium, tying operators to the enrichment market but allowing higher energy output per fuel assembly. The higher burnup in modern PWRs—commonly approaching 60 GWd/tU—means fuel stays in the core longer, but batch refueling introduces reactivity management challenges, such as the need for soluble boron in the coolant for long-term reactivity control. This boric acid absorption must be balanced by chemical shim and burnable absorbers to avoid positive moderator temperature coefficients under certain conditions. Advanced PWRs are now optimizing fuel assembly designs, including accident-tolerant fuels like chromium-coated cladding and doped uranium dioxide, to further improve safety and economics. For more details on fuel cycle technologies, the IAEA resources on nuclear fuel cycle provide extensive data.

On-Power Refueling and Operational Flexibility

One of the most visible operational distinctions is CANDU’s on-power refueling. The ability to replace individual fuel bundles while the reactor is generating electricity leads to very flat neutron flux distribution over time and better fuel burnup uniformity. It also enhances operational flexibility: if a defective fuel bundle is detected, it can be removed without shutting down the unit. On-power refueling also reduces the need for large reactivity reserves; fresh fuel is added gradually, so excess reactivity remains low, improving control and safety margins. The Bruce Nuclear Generating Station in Ontario, the largest operating nuclear facility in the world by total output, relies heavily on this capability to maintain high availability.

PWRs, which refuel during planned outages, must schedule fuel changes in batches and accept a more complex reactivity management curve over the fuel cycle. However, the batch refueling outage can be combined with major maintenance and testing, effectively concentrating downtime. Modern PWR fleets have optimized these outages to as little as two to three weeks every 18 months, achieving capacity factors comparable to CANDU over the long term. The refueling outage also allows for inspection and replacement of critical components such as steam generator tubes and reactor vessel head penetrations. Both approaches have proven economically viable, and the choice often depends on grid requirements and regulatory frameworks.

Safety Systems and Accident Mitigation

Both reactor types integrate multiple layers of defense-in-depth. CANDU stations feature two independent fast-acting shutdown systems, typically shut-off rods and liquid neutron poison injection (e.g., gadolinium nitrate), each capable of independently terminating the reaction. The moderator system acts as a passive heat sink; in a loss-of-coolant accident (LOCA), the calandria vessel and its heavy water can remove decay heat for an extended period even if the cooling water to the fuel channels is lost. The CANDU design also employs a large vault filled with light water as a secondary containment and heat removal pool. Some newer CANDU designs, such as the Enhanced CANDU 6, incorporate additional passive features like natural circulation-driven cooling.

PWRs rely on high-pressure injection systems, accumulators, and low-pressure recirculation to keep the core covered with water in accident sequences. The large containment building is designed to withstand internal pressure and collect any leaked fission products. Station blackout scenarios, where all alternating current power is lost, have driven the incorporation of passive cooling systems in newer designs, such as the AP1000’s passive containment cooling system or the VVER-1200’s passive heat removal circuits. The Fukushima Daiichi accident in 2011 led to a global reassessment of defense-in-depth, and both CANDU and PWR operators implemented enhanced severe accident management guidelines, hardened emergency equipment, and upgraded seismic and flooding protection.

For a detailed comparison of accident progression, the International Atomic Energy Agency provides comprehensive reports on nuclear safety, which outline the deterministic safety analysis for various reactor types. CANDU’s horizontal fuel channels and separated moderator give it a unique grace period in a station blackout; however, heavy water-related tritium release risks demand robust ventilation systems and operational controls. PWRs, by contrast, must manage the potential for boric acid precipitation in a degraded core.

Nuclear Proliferation and Safeguards Considerations

From a non-proliferation perspective, both reactor types are subject to rigorous international safeguards administered by the IAEA. CANDU’s use of natural uranium and frequent on-power refueling means that access to individual fuel bundles is continuous, and irradiated bundles contain plutonium. However, the plutonium in high-burnup CANDU fuel contains a high proportion of Pu-238 and other heat-generating isotopes, making it less attractive for weapons purposes due to heat dissipation and increased spontaneous neutron emission. Still, the ability to produce plutonium in a CANDU-like reactor operated without safeguards is a proliferation concern that has been addressed in export controls and bilateral agreements. The heavy water inventory itself is a sensitive technology requiring oversight as it can be used in non-power reactors to produce weapons-grade plutonium if diverted.

PWRs, with their low-enriched uranium fuel and long irradiation cycles, produce plutonium with a higher quality for weapons usage if operated in a dedicated mode, but the entire core is contained in a sealed vessel under safeguards surveillance. The IAEA’s checks and seals, combined with remote monitoring, effectively deter diversion. As both designs can be adapted for thorium or closed fuel cycles, future proliferation resistance will depend more on the fuel cycle infrastructure than on the reactor type alone. The OECD Nuclear Energy Agency analysis of proliferation resistance metrics notes that both CANDU and PWR can meet IAEA requirements when operated under appropriate safeguards agreements.

Economic and Construction Factors

The economics of CANDU vs. PWR have long been debated. CANDU’s absence of a large pressure vessel and reliance on numerous individual pressure tubes of smaller diameter historically allowed for manufacturing in countries without heavy forging capabilities, potentially reducing dependency on a few global suppliers. The concrete calandria vault and on-power refueling contribute to high capacity factors. However, the heavy water cost—initially a significant capital expense of up to $400 million per unit—and the ongoing tritium removal and heavy water make-up require continuous operational investment. Early CANDU plants experienced delays and overruns similar to other nuclear projects; recent CANDU builds such as China’s Qinshan Phase III units were completed on schedule and budget in the early 2000s, showcasing the maturity of the technology.

PWRs benefit from enormous global standardization: the supply chain for large forgings, steam generators, and reactor internals is well established. As a result, recent Gen III+ PWR designs have emphasized modular construction and reduced onsite labor. Nevertheless, new-build PWRs in the West have experienced significant cost escalations, particularly when first-of-a-kind engineering challenges emerge. The nuclear industry’s analysis of levelized costs suggests that overnight capital costs and financing rates dominate the economics, often overshadowing the specific reactor technology chosen. Both CANDU and PWR require substantial infrastructure, and their economic viability is highly country- and project-specific. Levelized cost estimates from recent studies indicate that CANDU can be competitive in markets with high uranium prices and low enrichment costs, while PWRs benefit from economies of scale in large fleets.

Waste Management and Decommissioning

Spent nuclear fuel from CANDU and PWR is high-level waste, though the volumes and characteristics differ. CANDU’s lower fuel burnup (typically 7–10 GWd/tU compared to 45–60 GWd/tU for PWR) means it produces a larger volume of spent fuel per unit of electricity generated—roughly twice the number of fuel bundles compared to PWR assemblies for the same energy output. However, each bundle is smaller and simpler. The radiotoxicity profile of spent CANDU fuel is similar to that of PWR fuel, with long-lived transuranics and fission products. Both require interim storage in water pools or dry casks and eventual deep geological disposal, for which extensive research programs exist in Canada (the Deep Geological Repository concept for Ontario Power Generation), Finland (Onkalo), Sweden, and other nations.

Decommissioning strategies also diverge slightly. The multi-channel CANDU design allows for phased dismantling, as individual tubes can be defueled and the heavy water drained relatively easily. The calandria’s low-pressure components result in less activated structural material than a thick-walled pressure vessel. PWR pressure vessels contain induced radioactivity from neutron exposure, necessitating careful segmentation and remote handling. Nevertheless, the global consensus remains that both technologies are manageable with current decommissioning techniques, as demonstrated by completed projects in the United States (e.g., Yankee Rowe PWR) and Europe (e.g., Greifswald PWR units). The cost of decommissioning is estimated to be somewhat lower for CANDU due to less activated material, but site-specific factors dominate.

PWRs dominate global markets, especially through designs from Westinghouse, EDF, Rosatom, and Chinese vendors. New reactor builds overwhelmingly choose evolutionary PWRs because of familiarity, licensing precedents, and established regulatory frameworks. CANDU reactors, though fewer in number (about 50 units in operation or under construction), operate reliably in multiple countries, and their technology continues to evolve. India’s pressurized heavy water reactors (PHWRs), which drew from CANDU experience, form the backbone of its nuclear program and are being upgraded to higher capacities (e.g., the 700 MWe PHWR type). The potential for CANDU variants to utilize thorium or spent PWR fuel—sometimes called DUPIC—offers intriguing waste reduction pathways and resource extension.

Advanced heavy-water concepts integrate passive safety features and aim for even greater fuel cycle flexibility. Canada’s interest in small modular reactors (SMRs) includes heavy-water-cooled designs such as the Integral Molten Salt Reactor, but also light-water SMRs. Meanwhile, PWR-based SMRs like the NuScale Power Module have garnered huge investment and licensing progress. The technology landscape is moving toward hybrid energy systems, and both CANDU and PWR must adapt to demands for load-following capabilities, hybrid heat applications (e.g., hydrogen production, district heating), and compatibility with renewable grid integration. Recent proposals for co-located nuclear-renewable energy parks highlight the versatility of both reactor types.

Conclusion

The CANDU and Pressurized Water Reactor are not competitors so much as complementary solutions born from different national strategies, natural resource endowments, and industrial capabilities. CANDU’s heavy-water moderator and natural uranium fuel cycle provide neutron efficiency and fuel flexibility that reduce dependence on enrichment, while its on-power refueling yields high availability. PWRs leverage abundant light water, enrichment infrastructure, and a vast operational fleet to deliver proven, standardized power plants with a well-understood safety basis. Understanding these design philosophies illuminates the broader challenges of nuclear energy: balancing neutron physics, safety, economics, and sustainability in a world that increasingly demands clean, reliable base-load electricity. As next-generation reactors enter the market, insights drawn from CANDU and PWR experience will remain central to achieving safe and sustainable nuclear power for decades to come. The choice between them ultimately depends on a country’s specific energy policy, industrial capacity, and long-term fuel cycle strategy.