The present invention relates coolant systems for nuclear reactors. In particular, the invention relates to molten salt coolants for use in nuclear reactors.
Nuclear reactors using molten salts as fuel or reactor coolant have been known for many years. There are several vital properties for such salts which were reviewed in depth by Williams et al (ORNL/TM-2006/12) who, like many other authors recommended that salt mixtures containing LiF, BeF2, NaF, ZrF4, RbF and KF should be considered (see tables A and B in the above reference).
Molten salt reactors utilising graphite as moderator are known to suffer problems of interaction of the molten salt with graphite. They suffer a particular problem when the redox state of the molten salt is maintained sufficiently strongly reducing so as not to extract chromium from steels. Many suitable molten salts react with graphite under such conditions. ZrF4 for example, which is particularly easy to maintain in a reducing condition as described by Scott (PCT/GB2016/053861) reacts readily with graphite forming stable carbides, as does the widely proposed FLIBE salt with BeF2.
There is a need therefore for a molten salt for use in contact with graphite in nuclear reactors which can be maintained readily in a strongly reducing condition without causing degradation of the graphite.
According to a first aspect of the present invention, there is provided the use of a molten salt comprising aluminium trifluoride and sodium fluoride as a primary coolant for a fission reactor, wherein the molten salt is in contact with graphite and with aluminium metal during operation of the fission reactor.
According to a second aspect of the invention, there is provided a nuclear fission reactor. The nuclear fission reactor comprises a reactor core and a primary coolant system. The reactor core comprises one or more containment units containing fissile fuel. The primary coolant system comprises a primary coolant and is configured such that the primary coolant is in contact with the containment units and removes heat from the reactor core. The primary coolant is a molten salt comprising aluminium trifluoride and sodium fluoride. The nuclear fission reactor comprises aluminium metal and graphite arranged such that the aluminium metal and the graphite are in contact with the primary coolant during operation of the reactor.
According to a third aspect of the invention, there is provided a nuclear fission reactor. The nuclear fission reactor comprises a heat exchanger and a fuel salt system. The fuel salt system comprises a reactor core and a fuel salt, and is configured such that the fuel salt flows between the reactor core and the heat exchanger. The fuel salt is a molten salt comprising aluminium trifluoride, sodium fluoride, and a chloride or fluoride of an element having a fissile isotope.
A nuclear reactor in which a molten salt coolant is in contact with graphite is cooled with a molten salt comprising aluminium trifluoride (AIF3) and sodium fluoride (NaF) with the coolant further in contact with metallic aluminium in the bulk molten or dispersed states. The molten salt may be used as is, where the coolant salt carries heat from immobile nuclear fuel structures, or may be mixed with fluorides of elements having fissile isotopes, e.g. actinide fluorides, where the coolant is also the fissile fuel of the reactor.
The metallic aluminium may be added as such or formed in situ by addition of a reducing agents such as sodium metal, by electrolytic processes, or by other known processes.
The eutectic mixture of AIF3 and NaF (approx. 55% mol% NaF and 45 mol% AIF3) has a particular desirable combination of properties as shown in
The eutectic AIF3/NaF salt has a low viscosity falling from 1.2 cP at 750° C. to 1 cP at 830° C. making it a close to ideal coolant with flow properties similar to liquid water.
Its neutron absorbance is particularly low as Na, F and Al have respectively thermal neutron cross sections of 531 mb, 9.6 mb and 230 mb.
A particularly important property of these salt systems is that they can dissolve over 1% of aluminium oxide without increasing their viscosity significantly. In the context of nuclear reactor coolants it allows the salt to absorb substantial amounts of oxygen without major changes in its properties provided that the resulting fluorine produced is neutralised. The presence of substantial quantities of aluminium metal in the system therefore allows the molten salt to remain non corrosive to metals despite substantial leakage of water or oxygen into the system. Active processes to maintain the low redox state can thus be replaced with simple passive presence of aluminium metal, which will effectively neutralise the fluorine produced. Since aluminium melts at a lower temperature than the AIF3/NaF eutectic, the aluminium will normally be in the liquid state which is desirable as it prevents formation of a passivating layer of aluminium oxide on the surface of the metal.
However, it has been found that this use of aluminium metal within the system means that, in contrast to suggestions in the literature, NaF is not interchangeable with other alkali metal salts when graphite is also present. Replacement with lithium, potassium or rubidium fluoride results in the production of the alkali metal in metallic form, which intercalates into the graphite, rapidly damaging its structure. While not wishing to be bound by theory, we believe that the particular resistance of graphite to the NaF salt in this scenario is a result of the thermodynamic barrier to sodium intercalation in graphite preventing formation of damaging concentrations of sodium in the molten salt. In contrast, the positive thermodynamic driving force for intercalation of the other alkali metals results in the equilibrium between aluminium and those metals shifting towards production of the metals with the result that high levels of intercalation take place.
LiF, KF or RbF can be tolerated in the salt provided that they are not the primary component (i.e. they are present in lower amounts than NaF). At such low concentrations, the ratio of those contaminating salts to graphite surface is such that they are depleted before significant graphite damage occurs.
There is a second factor in stability of graphite in reducing molten salts. This is formation of carbides. In the case of zirconium fluoride the reaction of the reducing zirconium species with graphite destroys the graphite structure. Aluminium fluoride, in contact with aluminium metal, does form a carbide, however experience in aluminium smelters has shown that this carbide forms a stable surface coating which does not degrade the graphite.
The combination of AIF3 and NaF is thus uniquely suited to the application in graphite containing nuclear reactors as both the Al and Na elements interact benignly with the graphite while replacement of either with comparable fluorides results in graphite damage.
While AIF3 has been mentioned in passing in existing documents as a potential constituent of a coolant salt, these previous discussions have not addressed the issues described above of ensuring adequate oxygen absorption while also preventing contamination of graphite structures.
Some properties of AIF3 are mentioned in the Williams et al (ORNL/TM-2006/12) paper mentioned earlier, but only in tables of properties and with no suggestion that they are appropriate salts for nuclear use. Specifically, the system NaF/AlF3 is mentioned in table 8 of this paper as having a composition of 75% NaF/25% AIF3 with a melting point of 1000° C., without any discussion of the need to mitigate production of sodium gas. Additionally, there is no disclosure of contact between the coolant and graphite and aluminium..
The definitive report regarding molten salts for nuclear reactors (Thoma, ORNL-2548) lists a large number of potential salt systems but includes no AIF3 salts.
Holcomb et al (ORNL/TM-2010/156) while mentioning the use of AIF3 in aluminium smelting (p3) where it is extensively use, mention it only once in their paper (table 9, p42) only to record the thermal conductivity of the same 75 mol% NaF/25 mol% AIF3 salt mentioned in Williams, with no mention of its suitability as a nuclear reactor coolant, or the other particular issues which arise from that use (e.g. the interaction with graphite).
AIF3 based salts have been proposed for use in reprocessing of spent nuclear fuel, for example by Carr et al (ORNL 4574) and Thoma (ORNL-3594) in the fluoride volatility process for recovery of uranium. No mention in either paper was made of AIF3 use in nuclear reactor cooling.
The World Nuclear Association (https://www.world-nuclear.org/information-library/current-and-future-generation/molten-salt-reactors.aspx) mention AIF3 salt as secondary coolants thus
“In industrial applications molten fluoride salts (possibly simply cryolite - Na-Al fluoride) are a preferred interface fluid in a secondary circuit between the nuclear heat source and any chemical plant. The aluminium smelting industry provides substantial experience in managing them safely.”
but does not mention them as primary reactor coolants.
Laurenty (The LM-LS experiment: investigating corrosion control in Liquid Fluoride Salts by Liquid alkali Metal, Report UCBTH-06-002, 2006) mentions Al and AIF3 as part of a corrosion control system for molten salt coolant in the Advanced High Temperature Reactor but rejects them as unsuitable (p30), which is clearly teaching away from their use.
Benson et al (WO2019/231971, WO2020/123509, 2020/123513) list AIF3 as one of a large number of salts that could be incorporated into molten salt nuclear reactors but ascribe no particular advantages of the use of that salt. Benson does not consider the use of aluminium metal to maintain the reduction state of the salt, and therefore does not consider the issue described above of the reaction of the salt with graphite.
Number | Date | Country | Kind |
---|---|---|---|
2014181.8 | Sep 2020 | GB | national |
2020536.5 | Dec 2020 | GB | national |
Filing Document | Filing Date | Country | Kind |
---|---|---|---|
PCT/EP2021/074301 | 9/2/2021 | WO |