FHRs have the potential to safely and reliably generate large quantities of power at lower cost than any other nuclear reactor class. The United States has attempted to develop non-light-water-cooled reactors since the 1950s. None have succeeded on a sustained basis in the commercial market despite long-term, substantial government funding. Light-water-cooled reactors (LWRs) are a reliable, mature reactor technology with an established cost and performance basis. Even LWRs, however, supply only ~20% of US electricity and only ~8% of total US energy. The most significant issue inhibiting the growth of nuclear energy in the United States is the high cost of available LWR reactor options (compared to the current cost of natural gas) combined with an output temperature considerably lower than that needed to support many industrial process heat uses.
Lower cost through passive safety
Today’s LWRs have an outstanding safety record. Multiple layers of expensive safety-related SSCs are employed to achieve their safety performance. LWR designers have recognized the cost of complexity, and modern LWR designs decrease safety costs by increasing passive safety (e.g., the Gen III+ designs with enhanced passive safety and the passive small modular reactors). For example, Westinghouse’s AP1000 has 50% fewer safety-related valves and 85% less control cable than earlier units.[1] FHRs represent a much larger opportunity for lowering costs than incrementally improved LWRs, as FHRs lack the high pressure and heat transfer threshold phenomena that underlie the requirements for the complex, expensive safety-related, safety-significant SSCs at LWRs. The high degree of passive safety afforded by FHRs provides the opportunity for a dramatic break from the cost vs. safety dilemma inherent in LWRs.
The primary coolant of FHRs will be a liquid fluoride salt. While several fluoride salts could be used in FHRs, the combination of desirable neutronic properties, good hydraulic performance, and very low activation of 27LiF‑BeF2 (FLiBe) makes it the preferred primary coolant. The boiling point of FLiBe is more than 1400°C, and its volumetric heat capacity at 700°C (4.67 J/cm3-K) is comparable to that of water at 100°C (4.04 J/cm3-K).
The three primary safety functions of nuclear reactors are to 1) control the reactivity, 2) cool the fuel, and 3) prevent the release of radionuclides. FHRs can potentially perform these functions at substantially lower cost than LWRs through reliance on the inherent characteristics of the fuel and coolant. FHRs exhibit no cliff-like phenomena (such as departure from nucleate boiling or substantial coolant pressure rise with temperature) in their temperature or heat transfer characteristics. Consequently, their requirements for reactivity control are much less stringent than those of LWRs. The primary coolant has ~700 K of margin between its operating temperature and its boiling point; the fuel has an even larger thermal margin to damage. The fuel also has a strongly negative temperature reactivity coefficient. As the fuel heats, more neutrons are absorbed parasitically in the fuel without causing fission. Also, an FHR’s core power density will be between a fifth and a tenth of that of a modern LWR, greatly decreasing the rate at which thermal transients progress and thereby providing more time for mitigating the transient. The ability of FHRs to tolerate substantial thermal excursions allows the use of passive, high reliability melt-point mechanisms to trigger insertion of negative reactivity.
Passively continuing to cool the fuel following reactor shutdown is also substantially less complex for an FHR. As Fukushima so dramatically demonstrated, nuclear fuel requires active cooling for many days following reactor shutdown. Gen III+ LWRs have significantly mitigated this issue by including large reservoirs of cooling water above the reactor vessel, passively maintaining cooling for several days. FHRs, in contrast, rely upon natural circulation to transfer heat from their fuel for as long as necessary following reactor shutdown. The fuel heats the liquid salt in the core, making it less dense. The hotter salt then rises to the top of the vessel. The hot salt is cooled by flowing through a heat exchanger located near the top of the reactor vessel. The dense, cooler salt sinks to the bottom of the reactor vessel, from where it is drawn up into the core again and heated by the fuel. The primary salt transfers heat to an auxiliary coolant salt on the opposite side of the heat exchanger. The auxiliary coolant salt in turn becomes less dense as it is heated and rises through piping out of the reactor containment, where it transfers its heat to the outside air, becomes denser, and flows back to the in-vessel heat exchanger. This type of natural-circulation-based cooling is called direct reactor auxiliary cooling.
FHRs are a low-pressure reactor class and consequently lack the pressure differential (driving force) that could cause radionuclides to leave the reactor. FHRs will employ multiple layers of containment and source-term reduction mechanisms beginning with the robust tristructural isotropic (TRISO) fuel currently being developed for gas-cooled reactors. Additionally, liquid fluoride salts chemically bind the most radiologically significant fission product radionuclides. In the highly unlikely event of substantial or complete fuel leakage from the multi-layered TRISO fuel particles and carbon fuel matrix, FHRs would behave like dissolved fuel MSRs. Prior experience with operating and evaluating dissolved fuel MSRs provides strong evidence that the radionuclides can be effectively contained. Moreover, an FHR’s low pressure substantially decreases the cost of leak-tight containment. An FHR’s containment will be a thin metal wall truss structure as compared to the meter-thick reinforced concrete and steel structures characteristic of LWRs. In order to meet the external missile (e.g., aircraft impact) requirements that containment also provides at an LWR, an FHR will be partially below grade with its above-grade nuclear island structures covered by an earthen berm (similar to a munitions storage bunker).
The TRISO fuel particles selected for FHR use consist of a microsphere (i.e., kernel) of uranium oxycarbide material encapsulated by multiple layers of pyrocarbon and a silicon carbide layer. This multiple-coating-layer system has been engineered to retain the fission products generated by fission of the nuclear material in the kernel during normal operation and all licensing basis events over the design life of the fuel. TRISO is a well-characterized fuel form originally developed for high temperature gas-cooled reactors.
FHRs will generate substantially more tritium than LWRs. Tritium is the only radionuclide with a significant potential for release into the environment. Tritium in FHRs is generated largely by neutron interactions with the fluoride salt constituent elements. At FHR operating temperatures tritium readily diffuses through structural alloys. Tritium can be stripped from fluoride salts and captured, thus precluding its release from the plant. A recent invention that applies gaseous hydrogen separation technology to flowing fluoride salts appears to be a key tritium removal technology.[2] The most problematic location for tritium escape is through the thin-walled piping of the primary heat exchanger. If tritium stripping is insufficient, escape through the heat exchanger tubing can be blocked by a double-walled heat exchanger employing either a sweep gas or a chemical trap between the walls.
Higher efficiency
A chief advantage of higher temperature operation is higher thermal efficiency. A 700°C output temperature enables a thermal cycle efficiency of 45% as compared with the 33% efficiency typical for LWRs. Moreover, unlike gas-cooled reactors, nearly all of the energy is available at the high temperature, improving the capability of FHRs to support high temperature thermal processes.
The higher thermal efficiency makes more useful energy available and less waste heat rejected. FHRs will fission roughly same fraction of fissile nuclei as other thermal spectrum reactors; however, the higher thermal efficiency increases fissile resource use by 35–40% over LWRs with a corresponding decrease in waste radionuclide generation.
Large power output
FHRs can maintain their passive safety at high power output. The low pressure and full passive safety of FHRs largely negate the difficulties associated with scaling up LWRs, which require more active and redundant safety systems. Large single plants also avoid the complexities related to sharing systems between multiple smaller plants. FHRs do not require large, thick-walled forgings or massive containment structures, which are the key elements whose prices rise nonlinearly with scale at LWRs. Further, the accident source term for an FHR will not vary significantly as reactor power scales up, owing to the radionuclide retention of the liquid salt combined with the multiple containment layers.
Rapid refueling
The combination of low pressure, robust fuel, and a transparent coolant makes refueling potentially more rapid than at LWRs. As a low-pressure system with mechanically and thermally robust fuel, FHRs avoid most of the time-consuming, nonfuel manipulation steps required in LWR refueling. Refueling can begin within minutes after reactor shutdown, as the coolant temperature does not need to be lowered below its boiling point before opening the vessel; neither do massive bolts need to be removed, a multi-ton vessel head moved, or the reactor cavity flooded and later drained. Moreover, FHRs will have little in-core instrumentation, substantially reducing the complexity of removing the upper vessel structures. Furthermore, a robust containment layer (the silicon carbide layer within the TRISO) stays with the fuel during movement, reducing the potential source term for a fuel-handling accident. The thermally robust nature of coated-particle fuel significantly eases the design requirements for fuel transfer safety. Even just a few hours after shutdown, fuel can withstand more than 15 minutes of removal from cooling without exceeding its allowable maximum temperature. Moreover, once refueling has been completed, an FHR can be rapidly brought back to power, as the refueling is performed hot and TRISO fuel does not have a phenomenon equivalent to the pellet–clad interaction that limits the rate of power ascension in LWRs.
Improved grid support
With the introduction of progressively larger amounts of nondispatchable power onto the grid, the stability margins of the US grid are decreasing as the dispatchable power sources must handle larger and faster changes in power output. Modern LWRs are designed to operate in modes other than steady-state 100% power output. In France LWRs operate in load-following mode with typical rates of change of 3–5% of rated power per minute. The robust nature of TRISO fuel affords FHRs greater operational flexibility and effectively eliminates their having any rate limits on power maneuvering.
Current LWRs are dependent upon a generally stable, reliable grid owing to their dependence on powered components to reject decay heat. LWRs have safety-grade local backup power sources to minimize the risk associated with loss of off-site power. Even so, existing LWRs are required to shut down upon loss of off-site power. The ability to passively reject decay heat indefinitely without connection to the grid essentially obviates the safety issue of grid connection for FHRs. Consequently, FHRs are anticipated to be able to help restart a black grid.
Low water use
Water availability is a significant siting limitation in much of the world. As FHRs are classified as high temperature reactors, they reject smaller amounts of heat to the environment than lower temperature systems. Further, the efficiency penalty of raising the heat rejection temperature is smaller for high temperature systems since the efficiency of a thermal power generation system depends on the ratio of the absolute temperature difference within the cycle. Raising the heat rejection temperature is key to enabling dry cooling with reasonable capital costs.
Licensability
Nuclear power is a regulated industry with extensive reliance upon precedent. Securing licensing approval from the US Nuclear Regulatory Commission (NRC) is based upon demonstrating compliance with the general design criteria (GDCs) for commercial nuclear power plants established in the Code of Federal Regulations. Although US licensing procedures focus on large LWRs, the GDCs are intended to be extendable to non-LWR reactor classes. The Fort Saint Vrain high temperature gas-cooled reactor (HTGR) was granted an operating license, and the sodium-cooled, fast spectrum Clinch River Breeder Reactor was granted a construction permit under a licensing framework similar in many respects to that existing today.
Currently, a joint initiative between the US Department of Energy (DOE) and the NRC is under way to develop advanced reactor GDCs that would allow the NRC to develop evaluation criteria for non-LWRs. While the effort appears to be extendable to FHRs, the focus of the work is on sodium fast reactors (SFRs) and HTGRs. FHRs are quite similar to SFRs in terms of their regulatory requirements (i.e., solid fueled, liquid cooled, natural-circulation-based decay heat removal, low pressure). Consequently, the SFR GDCs are anticipated to provide an almost complete map for developing FHR-specific GDCs. Further, the American Nuclear Society is developing a design/safety standard for FHRs (ANS 20.1), which will provide more specific design guidance.
Robust waste form
TRISO fuel is more chemically and thermally robust than LWR fuel rods. Further, the first layer of containment remains with the fuel, so it less likely to fail in the ex-core environment. The mechanically robust, non-chemically reactive nature of FHR used fuel makes it compatible with any form of storage. Much of the expense involved in planning for geologic waste disposal of used fuel is based upon preventing water transport of radionuclides from corroded fuel. FHR fuel is not water soluble, largely avoiding water-based radionuclide transport issues.
TRISO has a lower fuel density than LWR fuel pellets. Consequently, FHRs will have a larger high level waste volume than LWRs. The carbon fuel structures, if not separated from the TRISO particles, will increase the high level waste volume.
[1] http://www.ap1000.westinghousenuclear.com/ap1000_glance.html.
[2] David E. Holcomb and Dane F. Wilson, Apparatus and Method for Stripping Tritium from Molten Salt,US Patent Application 14333627, filed July 17, 2014.
Contacts
David Holcomb, HolcombDE@ornl.gov, Salt Reactor Technical Lead
Jess Gehin, GehinJC@ornl.gov, ORNL Reactor Technology Lead
Gary Mays, MaysGT@ornl.gov, Advanced Reactor Systems and Safety Lead