The role of coolant is to transfer heat from the nuclear core to the thermal application, be it steam generation for an electrical turbine or direct process heat. The main considerations in choosing the coolant have historically been heat transfer effectiveness, neutronic moderation for thermal reactors, and availability. In the case of water, designers have chosen a highly effective and available fluid that also moderates neutrons (allowing for smaller cores or more fuel). Unfortunately, water also introduces significant chemical energy sources (hydrogen generation), material corrosion challenges, and high-pressure requirements. 

Are water’s excellent heat transfer capabilities worth it? In terms of hazard reduction, the answer is “no” as there are ready alternatives. Water is directly implicated in all three past nuclear accidents, either causing or escalating the accident. At Fukushima, loss of grid and emergency power left operations without cooling ability for the reactors during decay heat shutdown. Without flowing water to cool the fuel, temperatures increased, and water reacted with Zircaloy cladding to produce hydrogen gas, increasing pressures, and leaked out of the containment and eventually exploding violently in the reactor building.

In terms of cost, one would expect water to offer clear benefits, but the fact is that heat transfer in the primary loop is a small component of the LCOE. A 10x reduction in heat transfer related costs could be negligible. And unfortunately for water, lifecycle heat transfer costs related to OPEX and extra CAPEX to deal with activated water and water related accidents could be higher. The higher corrosion and pressure necessitate a more involved maintenance and inspection burden, and more numerous and more expensive subsystems and components. Water cooling components and systems may be cheaper up-front today only because of their ubiquity in other manufactured systems including engines, air conditioners, and conventional power plants.


Using Helium, a designer sacrifices significant volume in the core for coolant channels and has to use larger pumps and heat exchangers to accomplish the heat transfer task. But the advantages are an incredibly simplified system and access to higher temperatures. It is wasted volume because helium as a gas, has very poor heat transfer properties and is also neutronically transparent and does not moderate the neutrons as water does. The coolant is clean and totally non-corrosive thereby reducing the maintenance, inspection, and radioactivity problem in the reactor peripherals. The helium has very little chemical reaction potential to produce other compounds or release chemical energy. It can basically leak out at any time with minor effects, though it may become costly to replace it if wide deployment of helium cooled reactors take place. Combined with all-ceramic cores, using helium coolant is like using an underpowered electric motor while massively, upgrading the brakes and traction control. Since helium is a poor heat transfer fluid, it has to be pressurized and combined with its large volume, HTGRs feature enormous reactor pressure vessels per power generated. However, if the core inlet temperature is kept low enough (not using salt storage or SCO2 power cycle), the use of helium allows the pressure vessel to be manufactured with low alloy steel and last 40-60 years with minimal corrosion. 

Another consideration is phase change. Helium is already a gas at all encountered conditions, while other coolants (sodium, molten salt, water) will have phase changes at different temperatures and pressures. Having only a single phase in a typical nuclear reactor operating space limits the design and system burden of dealing with multiple phases and limits the possibility of some accidents. Under HTGR conditions, a gas cannot phase change into a more chaotic phase. This is especially a concern during accident conditions when liquid coolants begin to boil, providing a dispersive energy source and increasing pressure and corrosion and can result in loss of coolant or rapid degradation of components as in the case of water, sodium, or molten salt boiling.

Compared to water or molten salts, helium is less effective at heat transfer at the relevant temperatures and pressures. Greater electrical power must be used to pump the coolant and the heat exchangers are larger. As mentioned before, while the heat transfer fluid is small cost component of the LCOE, the heat transfer equipment can be a significant part depending on the design LCOE target.  Nevertheless, the heat transfer equipment with helium should reduce the maintenance and inspection burden and not complicate the chemistry and neutronics of the core. The helium poor heat transfer properties are actually leveraged in HTGRs as it allows them to operate with such low power densities that most rare postulated accidents can be passively overcome.

Liquid Metals

Liquid metals including sodium, lead, and tin and various eutectics can be used as reactor coolant. They offer efficient heat transfer due to the high thermal conductivity of the fluid and do not require high pressure boundaries like water or gas coolants. They are typically used in fast reactors, but can be also be used with thermal reactors, such as the graphite-moderated sodium-cooled reactor concept.

The only challenges with liquid metals are the highly energetic reactions with water and air, activation of the coolant, and the greater expected complexity in cleanup and maintenance for Liquid Metal components compared to gas coolants.

Liquid metals can be packaged as pool type reactors, where the reactor core sits in a pool of the fluid. Alternatively, liquid metals can be packaged as heat pipes.