Nuclear reactor that uses organic liquids for cooling and neutron moderation
Organic-Cooled Nuclear Reactors
Executive Summary
Organic nuclear reactors were initially both organically-moderated and organically-cooled, but, organically-moderated reactors were abandoned despite being technically feasible because organic moderation was found to be inferior to water and heavy water moderation. As a result, these organically-moderated reactors were not pursued beyond early research programs.
Organic-cooled reactors (OCR) is a type of nuclear reactor that uses an organic heat-transfer fluid, typically a hydrocarbon-based liquid, as the primary coolant. Organic coolants were investigated during the 1950s–1980s as an alternative to water and liquid-metal coolants, offering the potential for high operating temperatures at low system pressure. Several experimental and demonstration OCRs were built in the United States, Canada, and the Soviet Union. Operational experience across these reactors showed that organic coolants provided stable heat transfer but were susceptible to radiolysis and thermal degradation producing gases, acidic compounds, and polymeric residues when operated outside well-defined thermal limits and performance envelope. The WR-1 successfully demonstrated the potential and key advantages of the organic-cooled reactors using a well-managed coolant chemistry control program through purification, filtration, and chemistry monitoring.[1][2]
Organic coolants[1][3] have high boiling points and low vapor pressures, allowing reactor operation at temperatures comparable to Pressurized Water Reactors (PWRs) and Boiling Water Reactors (BWRs) without requiring high pressure. In most practical designs, the organic liquid served only as coolant, while moderation was provided by light water, heavy water, or graphite. Early experiments also evaluated organic moderation, but this approach was later abandoned.
Using an organic fluid had a major advantage over conventional designs using water as the coolant. Water tends to corrode and dissolve metals, both the nuclear fuel and the reactor as a whole. To avoid corrosion of the fuel, it is formed into cylindrical pellets and then inserted in zirconium tubes or other "cladding" materials. The rest of the reactor has to be built out of materials that are both corrosion resistant and resistant to the effects of neutron embrittlement. In contrast, many common organic fluids are less corrosive to metals, allowing the fuel assemblies to be much simpler and the coolant pipes to be built of normal carbon steels instead of more expensive corrosion-resistant metals. Some organics also have the advantage that they do not flash into gas in the same fashion as water, which may reduce or eliminate the need for a containment building. Another benefit is the low pressure operations with significantly reduced pressure-boundary stresses, low energy release during loss-of-coolant events, and minimal, operationally insignificant neutron activation of the coolant. More on this later.
Demonstration and Experimental Organic-Cooled Reactors
Below is a summary of demonstration and/or experimental organic-cooled reactors built[1][2][3][4][5]
United States
Organic Moderated Reactor Experiment (OMRE): OMRE was an early experimental reactor operated in the 1950s in Idaho National Engineering Laboratory. It uniquely employed organic liquid as both coolant and moderator, demonstrating basic feasibility but also exposing issues with radiolysis, gas generation, and chemistry control problems. OMRE was strictly experimental and was shut down after fulfilling its research objectives.
Piqua Nuclear Generating Station: Piqua was a small demonstration power reactor using organic coolant with light-water moderation. It was erroneously operated in a manner like a light-water-reactor without a practical organic coolant chemistry control program which led to chronic contamination, coolant degradation, and maintenance challenges resulting in poor reliability and early shutdown.
Canada
WR-1: An organic-cooled, heavy-water-moderated demonstration reactor operated by Atomic Energy of Canada Limited (AECL) at Whiteshell Laboratories from 1965 to 1985. The reactor was designed around the organic coolant operational envelope regarding coolant chemistry controls and thermal performance which led to WR-1 achieving reliable sustained operations for nearly 20 years with >80% availability factor. It is generally regarded as the most successful organic-cooled reactor.
Soviet Union / Russia
ARBUS(А́РБУЗ): The Soviet ARBUS reactor (the name derived from the Russian acronym for an aromatic-hydrocarbon-cooled experimental reactor) was an experimental organic-cooled research reactor operated in the Research Institute of Atomic Reactors USSR during the 1960s. ARBUS employed organic coolant with solid moderation and was used to investigate coolant stability, radiolysis, and materials compatibility under irradiation. Like Western counterparts, the program identified coolant degradation and chemistry control as limiting factors. (IAEA TRS-61; IAEA TECDOC-432)
Physics
Fission basics
Conventional fission power plants rely on the chain reaction caused when nuclear fission events release neutrons that cause further fission events. Each fission event in uranium releases two or three neutrons, so by careful arrangement and the use of various absorber materials, you can balance the system so one of those neutrons causes another fission event while the other one or two are lost. This careful balance is known as criticality.[6]
Natural uranium is a mix of several isotopes, mainly a trace amount of U-235 and over 99% U-238. When they undergo fission, both of these isotopes release fast neutrons with an energy distribution peaking around 1 to 2MeV. This energy is too low to cause fission in U-238, which means it cannot sustain a chain reaction. U-235 will undergo fission when struck by neutrons of this energy, so it is possible for U-235 to sustain a chain reaction, as is the case in a nuclear bomb. However, there is too little U-235 in a mass of natural uranium, and the chance any given neutron will cause fission in these isolated atoms is not high enough to reach criticality. Criticality is accomplished by concentrating, or enriching, the fuel, increasing the amount of U-235 to produce enriched uranium,[7] while the leftover, now mostly U-238, is a waste product known as depleted uranium.[8]
U-235 will undergo fission more easily if the neutrons are of lower energy, the so-called thermal neutrons. Neutrons can be slowed to thermal energies through collisions with a neutron moderator material, the most obvious being the hydrogen atoms found in water. By placing the fission fuel in water, the probability that the neutrons will cause fission in another U-235 is greatly increased, which means the level of enrichment needed to reach criticality is greatly reduced. This leads to the concept of reactor-grade enriched uranium, with the amount of U-235 increased from less than 1% to between 3 and 5% depending on the reactor design. This is in contrast toweapons-grade enrichment, which increases the U-235 enrichment to, commonly, over 90%.[8]
Coolants and Moderators
When a neutron is moderated, its kinetic energy is transferred to the moderator material. This causes it to heat up, and by removing this heat, energy is extracted from the reactor. Water makes an excellent material for this role, both because it is an effective moderator, as well as being easily pumped and used with existing power generation equipment similar to the systems developed for steam turbines in coal fired power plants. The main disadvantage of water is that it has a relatively low boiling point, and the efficiency in extracting the energy using a turbine is a function of the operational temperature.
The most common design for nuclear power plants is the pressurized water reactor (PWR), in which the water is held under pressure, on the order of 150 atmospheres, in order to raise its boiling point. These designs may operate at temperatures as high as 345 °C, which greatly improves the amount of heat that any unit of water can remove from the core, as well as improving the efficiency when it is converted to steam in the generator side of the plant. The main downside to this design is that keeping water at this pressure adds complexity, and if the pressure drops, it can flash into steam and cause a steam explosion. To avoid this, reactors generally use a strong containment building or some form of active steam suppression.[9]
A number of alternative designs have emerged that use alternative coolants or moderators. For instance, the UK's program concentrated on the use of graphite as the moderator and carbon dioxide gas as the coolant. These reactors, the Magnox and AGR operated at roughly twice the temperature as conventional water-cooled plants. This not only increases the efficiency of the turbomachinery, but is designed to allow it to run with existing coal-fired equipment that runs at the same temperature. However, they had the disadvantage of being extremely large, which added to their capital costs.[10]
In contrast, the Canadian CANDU designs used two separate masses of heavy water, one acting as the moderator in a large tank known as the calandria, and another acting solely as the coolant in a conventional pressurized loop. This design did not have the entire coolant mass under pressure, which simplified the construction of the reactor. The primary advantage was that the neutron moderation of heavy water is superior to normal water, which allowed these plants to run on natural, unenriched, uranium fuel. However, this was at the cost of using expensive heavy water.[7]
Organic Coolants and Moderators
In conventional water-cooled designs, a significant amount of effort is needed to ensure that the materials making up the reactor do not dissolve or corrode into the water. Many common low-corrosion materials are not suitable for reactor use because they are not strong enough to withstand the high pressures being used, or are too easily weakened by exposure to neutron damage. This includes the fuel assemblies, which in most water-cooled designs are cast into a ceramic form and clad in zirconium to avoid them dissolving into the coolant.[11]
Selected organic-based coolants avoid this problem because they are hydrophobic and generally do not corrode metals. This is why they are often used as anti-corrosion agents and rustproofing. Greatly reducing corrosion allows the complexity of many of the reactor parts to be simplified, and fuel elements no longer require exotic formulations. In most examples the fuel was refined uranium metal in pure form with a simple cladding of stainless steel or aluminum.[12]
In the simplest organic reactor designs, one simply replaces just the coolant with the organic fluid. This is most easily accomplished when the moderator was originally separate, as is the case in the UK and Canadian designs. In this case, one can modify the existing designs to become the 'graphite moderated, organic cooled reactor' and 'heavy water moderated, organic cooled reactor', respectively. Possible moderators other than graphite or organic fluid include beryllium, beryllium oxide, and zirconium hydride.[13]
However, the US program, by far the largest, concentrated on the 'organic moderated and cooled reactor' design, which is conceptually similar to the pressurized water reactor, simply replacing the water with a suitable organic material. In this case the organic material is both the coolant and moderator, which places additional design limitations on the layout of the reactor. However, this is also the simplest solution from a construction and operational point of view, and saw significant development in the US, where the PWR design was already common.[14]
Another common design in US use is the boiling water reactor (BWR). In this design the water is placed under less pressure and allowed to boil in the reactor core. This limits the operational temperature, but is simpler mechanically as it eliminates the need for a separate steam generator and its associated piping and pumps. One can adapt this design to an organic moderated and cooled reactor cycle as well, which is aided by the fact that suitable organic fluids superheat on their own when they expand into the gas state, which can simplify the overall design.[15]
This last issue also has a significant safety benefit; in contrast to water, oils do not flash into steam, and thus there is no real possibility of a steam explosion. Other potential explosion sources in water-cooled designs also include the buildup of hydrogen gas caused when the zirconium cladding heats; lacking such a cladding, or any similar material anywhere in the reactor, the only source of hydrogen gas in an oil-cooled design is from the chemical breakdown of the coolant. This occurs at a relatively predictable rate, and the possibility of a hydrogen buildup is extremely remote. This greatly reduces the required containment systems.[16]
Disadvantages
Organic-based coolants have several disadvantages as well. Among these is their relatively low heat transfer capability, roughly half that of water, which requires increased flow rates to remove the same amount of energy.[12] Another issue is that they tend to decompose at high temperatures, and although a wide variety of potential materials were examined, only a few appeared to be stable at reasonable operational temperatures, and none could be expected to operate for extended periods above 530 C.[17] Most are also flammable, and some are toxic, which presents safety issues.[12]
Another issue, when the oil is also the moderator, is that the moderating capability of the fluid increases as its temperature cools. This means that as the moderator heats up, it has less moderating capacity, which causes the overall reaction rate of the reactor to slow and further cool the reactor. Normally this is an important safety feature, in water-moderated reactors the opposite may occur and reactors with positive void coefficients are inherently unstable. However, in the case of an oil moderator, the temperature coefficient is so strong that it can rapidly cool. This makes it very difficult to throttle such designs for load following.[12]
But by far and away the largest problem for hydrocarbon coolants is that they decompose when exposed to radiation, an effect known as radiolysis. In contrast to heat-based decomposition, which tends to make lighter hydrocarbons, the outcome of these reactions is highly variable and results in many different reaction products. Water also undergoes decomposition due to radiation, but the output products are hydrogen and oxygen, which are easily recombined into water again. The resultant products of the decomposition of oils are not readily recombined, and have to be removed.[17]
One particularly worrying type of reaction occurred when the resulting products polymerized into long-chain molecules. The concern was that these would form large masses within the reactor, and especially its cooling loops, and might "exert significant deleterious effects on the operation of a reactor."[17] It was polymerization of the coolant sticking to the fuel cladding that led to the shutdown of the Piqua reactor after only three years of operation.[18]
History
Early experiments
Early theoretical work on the organic cooled concept was carried out at the Argonne National Laboratory between 1953 and 1956. As part of this work, Mine Safety Appliances studied a variety of potential biphenyl coolants. In 1956-75, Aerojet conducted studies on the rate of "burnout" of polyphenyl coolants, and in the following two years, Hanford Atomic Products carried out several studies of polyphenyl irradiation.[19]
Monsanto began operating a single coolant loop in the Brookhaven Graphite Research Reactor beginning in 1955 to study heat transfer, and in 1958 began to consider coolant reclamation and studies on boiling diphenyl coolant loops.[20]Atomic Energy of Canada Limited (AECL) began similar studies around the same time, with an eye to the design of a future test reactor.[20]
A similar program began in the UK at Harwell in the 1950s. This soon focused on radiation damage to organic compounds, specifically polyphenyls. Around 1960, Euratom began studies of such designs as part of their ORGEL project.[20][21][22] A similar but separate project began in Italy under the direction of the Comitato nazionale per l'energia nucleare, but their PRO design was never built. Likewise, a major study carried out in Denmark considered the heavy water-moderated reactor.[20][23]
Major experiments
The first complete organically cooled and moderated reactor design was the Organic Moderated Reactor Experiment (OMRE), which began construction at the Idaho National Laboratory in 1955 and went critical in 1957. This used biphenyl and Santowax (commercial name of an isomeric mixture of terphenyl) for coolant and moderation and operation was generally acceptable. The reactor was a very low-energy design, producing 15 MW thermal, and operated for only a short period between 1957 and 1963. During this time the core was rebuilt three times to test different fuels, coolants and operating conditions from 260 to 370 C. It was planned that a larger 40 MW design, the terphenyl-cooled Experimental Organic Cooled Reactor (EOCR), would take over from the OMRE. It began construction at Idaho in 1962, but was never loaded with fuel when the AEC shifted their focus mostly to light water reactors.[18]
The next major reactor was a commercial prototype built as a private/public venture, the Piqua Nuclear Generating Station, which began construction in 1963 at Piqua, Ohio. This used the same Santowax coolant as the original OMRE, but was as large as the EOCR, producing 45 MW thermal and 15 MW electrical. It ran on 1.5% enriched fuel formed into annular tubes that were clad in finned aluminum casings. It ran only for a short time until 1966, when it was shut down due to films building up on the fuel cladding, formed from radiation degraded coolant.[18]
The most powerful ONR was the Canadian 60 MW thermal WR-1. It began construction at the newly formed Whiteshell Laboratories in Manitoba in 1965 and went critical late that year. WR-1 used heavy water as the moderator and terphenyls as the coolant, and did not suffer from the problems with coolant breakdown seen in the US designs. It operated until 1985, by which time AECL had standardized on using heavy water for both the moderator and the coolant, and the organic cooled design was no longer being considered for development.[24]
Although various European nations did development work on organic reactor designs, only the Soviet Union built one. Work on the 5 MW thermal Arbus NPS began in Melekess, Russia in 1963 and it ran until 1979. It produced a maximum of 750 kW of electricity.[25] In 1979 it was rebuilt as the AST-1, this time to deliver 12 MW of process heat instead of electrical power. It ran in this form until 1988.[18]
WR-1 employed an organic heat-transfer fluid as it's primary coolant which avoided the mechanical stresses, corrosion, and embrittlement concerns seen in high-pressure PWRs and BWRs. This coolant allowed WR-1 to achieve near-steam-cycle temperatures without PWR/BWR pressure boundaries and demonstrated that an organic coolant could be integrated into a pressure-tube, heavy-water moderated reactor design with stable heat removal characteristics.
During sustained operations, WR-1 experienced some radiolysis and thermal degradation of the organic coolant caused by neutron and gamma irradiation and prolonged high-temperature exposure. Neutron activation of the WR-1 organic coolant was minimal and operationally insignificant while corrosion rates in the primary system remained very low with proper materials selection. As a result, operational issues were dominated by coolant chemistry management and maintenance rather than structural degradation of pressure boundaries. These processes produced lighter hydrocarbons, dissolved gases, acidic species, and heavier polymeric by-products, leading to coolant darkening and particulate buildup in pumps, valves, and heat-exchange surfaces. AECL mitigated these effects through online purification systems, filtration, and periodic coolant reconditioning, maintaining system operability within operational limits.
Operational radiation surveys at WR-1 confirmed that dose rates in the organic coolant system were driven mainly by trace fission product contamination and surface plate-out, rather than by neutron activation of the coolant itself. This significantly reduced radiation exposure during maintenance and inspection activities compared with water-cooled reactor primary systems, where activated coolant is a dominant contributor to occupational dose. Unlike light-water reactor coolants, the organic fluid did not contain oxygen or nitrogen as intrinsic constituents, eliminating dominant activation pathways such as the production of ¹⁶N that drives high gamma dose rates in PWR and BWR primary systems. Furthermore, the measured tritium levels and other activation products associated with the organic coolant were orders of magnitude lower than those observed in light-water reactor primary systems.
While WR-1 did experience three documented coolant leaks over its near 20-year operating life, these incidents behaved like traditional hydrocarbon releases rather than radiological emergencies, as the coolant did not become significantly radioactive. Corrective actions included coolant containment or recovery, sediment cleanup in affected areas, and mechanical and procedural upgrades to prevent recurrence. In all cases, the incidents were handled as chemical release issues rather than nuclear safety events, demonstrating a key distinction between organic coolant behavior and traditional water-cooled accident scenarios.
WR-1 proved that organic coolants can be operated safely, effectively, and with stable heat transfer when the reactor design is purpose-built for them. Piqua failed not because the coolant concept was inherently flawed, but because the reactor system architecture was mismatched to the coolant’s operational performance requirements. In modern reactor design terms, WR-1 provided the proof-of-concept, while Piqua demonstrated the misapplication of the concept.
WR-1 operated well because the reactor design, operational envelope, and coolant performance through chemistry controls were aligned from the start. By combining a low-pressure organic coolant with a CANDU-like pressure-tube configuration and heavy-water moderation, AECL paired the coolant’s strengths—low pressure, thermal stability, and minimal corrosion—with a system that did not depend on high mechanical loads. It was a purpose-built match of coolant operational performance to reactor system architecture.
Piqua[31][32], by contrast, used an inferior organic coolant and moderator with poor melting/pour point characteristics into a reactor designed much more like a light-water reactor (LWR) than like WR-1, resulting in chronic contamination issues, coolant handling risks, and maintenance burdens the plant was never engineered to support. These design choices is the key reason why Piqua performed poorly. Where WR-1 validated the feasibility of organic cooling as a highly reliable nuclear generation concept, Piqua illustrated that the coolant could not simply be treated as a substitute for water in an existing reactor architectural layout.
Criteria
WR-1 (Canada)
Success Case
Piqua (USA)
Failure Case
Design Fit
CANDU-like design matched coolant with operational envelope
Inferior organic coolant choice mismatched to operational envelope
Poor organic coolant/moderator choice in a LWR inspired design yielded persistent operational issues
Coolant Leakage
There were three leakage events that took place at WR-1 over its 20-year lifetime[33][34][35]:
August 1967: A pinhole leak developed in a heat exchanger tube releasing approximately 300 litres of organic coolant into the Winnipeg River via the plant’s outfall drainage system. This incident prompted design and operational changes such as additional valves and notification systems to help prevent future events.
January–May 1977: A prolonged, low-volume leak occurred over several months, estimated to release about 1,450 kg of organic coolant into the Winnipeg river. AECL calculated that 900–1,100 kg of this product settled into the riverbed sediments up to about 1 km downstream of the plant’s discharge point, with small amounts found farther downstream on both banks.The cause was attributed to slow leaks from primary cooling circuits rather than a single rupture. This led to corrective measures to address the affected seals in the coolant system to prevent future leaks.
November 1978 (No Release): A pump failure caused an organic coolant leak estimated at 3,270 kg within plant systems. Unlike the previous events, no coolant was directly discharged into the Winnipeg River. Most of the leaked coolant was recovered, contained, and stored on site, and operating procedures were revised to help prevent recurrence.
Despite sediment detection, river water samples generally showed organic coolant levels below analytical detection limits, indicating limited aqueous dispersion of dissolved contaminants. Biological sampling during follow-up monitoring did not show internal contamination or elevated mortality attributable to WR-1 organic coolant residues. Furthermore, multiagency reviews — including Environment Canada and Health Canada participation — concluded that river water downstream of the WR-1 site met Canadian drinking water guidelines, with no evidence that the coolant releases posed measurable human health risks. In 2006, AECL analyzed river sediment core samples at areas downstream of the site where deposits from the outfall were found. AECL concluded that there was no contamination of the river sediments that would have an ecological impact or affect human health.[33][34][35]
Key WR-1 Operational Takeaways
The WR-1 organic coolant contained only carbon and hydrogen and was not the neutron moderator, hence, neutron activation products were minimal, short-lived, and operationally insignificant—making coolant behavior chemical rather than radiological in both routine operation and leak scenarios. Because of this, the WR-1 leakage events behaved as chemical releases rather than radiological incidents. Environmental assessments following the coolant leaks concluded that the primary hazards were chemical, not radiological, and emergency response actions typical of activated water releases in PWRs/BWRs were not required. This distinction was repeatedly cited by AECL as a favorable operational characteristic of organic-cooled reactor systems.
Renewed interest
Indian officials have periodically expressed interest in reviving the concept. They initially received CANDU design materials during the period of the WR-1 experiment. To further lower operational costs, there have been several revivals of the WR-1-like concept. It is believed that an organic coolant purification system can be developed to handle the decomposition of the organic coolant, and research has begun to this effect. However, as of 2018[update], no experimental system has been constructed.[16]
↑"WR-1". Manitoba Branch of the Canadian Nuclear Society. 2005-03-18. Archived from the original on 2005-03-18. Retrieved 2016-11-07.
↑Tsykanov, V. A.; Chechetkin, Yu. V.; Kormushkin, Yu. P.; Polivanov, I. F.; Pochechura, V. P.; Yakshin, E. K.; Makin, R. S.; Rozhdestvenskaya, L. N.; Buntushkin, V. P. (1981). "Experimental nuclear heat supply station based on the arbus reactor". Soviet Atomic Energy. 50 (6): 333–338. doi:10.1007/bf01126338. ISSN0038-531X. S2CID93462910.
↑Tegart, D. R., Operation of the WR-1 Organic-Cooled Research Reactor, AECL-3523, AECL, 1970.
↑Webb, J. G., “Development of Organic-Cooled Reactor Systems in Canada,” Nuclear Engineering Review, Vol. 7, No. 3, 1968
↑Whiteshell Laboratories Chemistry Division, Coolant Degradation and Purification Performance in WR-1, 1982.
↑Review of Operating Experience with Organic-Cooled Reactors, IAEA-TECDOC-432, 1987.
↑AECL WR-1 Reactor Operations and Experimental Program Summary. Whiteshell Laboratories, 1986.
↑U.S. Atomic Energy Commission, Operating Experience and Shutdown Analysis of the Piqua Nuclear Power Facility, 1969.
↑Starr, C. “Organic Coolants in Power Reactors.” Transactions of the American Nuclear Society, Vol. 15, 1971.
This page is based on this Wikipedia article Text is available under the CC BY-SA 4.0 license; additional terms may apply. Images, videos and audio are available under their respective licenses.
