Fuel salt

Fuel salt Composition EiF-BeFj-ZrF4-UF4 (65.0-29.1-5.0-0.9 mole... 

The Dow Chemical Company Explosive Composition Comprising a Salt Component Contiguous to an Over-Fueled Salt Component... 

Paint Coating, Ship Fuel, Salt Tanks... 

L3. Lindauei, R. B. Processing of the MSRE Flush and Fuel Salts, Report ORNL-TM-2578, Aug. 1969. 

The calculated Loss-of-Flow and reactivity accidents were described and estimates of the behaviour of a lead-cooled fast system and that of thermal ADSs with a circulating fuel/salt mixture were described in the chapter on ADS safety in the IAEA State of the Art (SOAR) report on accelerator-driven systems which will soon be published. 

Fission products either quickly form stable fluorides that will stay within the salt during any leak or accident or are volatile or insoluble and can be continuously removed. For example, cesium forms stable CsF and iodine forms an iodide which is relatively stable in the fuel salt. 

While the fuel is considered mobile, many layers of contaiiunent are employed at low-pressure operation. The fuel salt is typically run at a lower pressure than intermediate loops such that any leak is inward. 

A fwo fluid design uses a fuel salt that carries the fissile UF4, but no fertile material and a separate blanket salt for fhe fertile ThF4. As is produced in the blanket, it is transferred to the fuel salf by a relatively simple fluorination process. Fluorination is accomplished by bubbling fluorine gas through the salt, thereby converting UF4 to volatile UFg. Gaseous UFg is collected and converted back to UF4 by a well-established process before being added to fuel salf. 

The main advantage of a two fluid system is that the processing-out of fission products from the fuel salt is simplified by the absence of fhorium. The prime method is known as vacuum distillation (ORNL 3791, 1966) and was developed in 1964. After removal of all UF4, the carrier salt would be evaporated off af low pressure and high temperature (1000°C) and recycled, leaving most fission products behind in the still bottoms. 

Most proposed MSR designs call for the use of enriched lithium-7 and/or beryllium such as in the 70.7%T ,iF-17%BeF2-12%ThF4-0.3%UF4 fuel salt of the 1970s MSBR program. This carrier salt is often termed FLiBe. The presence of either Li or Be leads to significant production of tritium, on par with production levels in heavy water reactors such as CANada Deuterium Uranium (CANDU). For example in the 1970s, the 1000 MWe MSBR projected 2420 curies per day (ORNL 4541,1971), 98.3% from lithium and beryllium, and the remainder of 0.4% from fluorine and 1.2% from ternary fissions. For comparison, CANDU operations typically experience a tritium release rate of less than 24 curies per day (CNSC INFO-0793,2009). 

Noble metal fission products will not form stable fluorides in the fuel salt but will tend to plate out on surfaces in the primary loop. The main complication that arises from this is in terms of the primary heat exchangers. Work with the MSRE indicated that upward of 40% of noble metals plated-out on the walls of heat exchanger. If shell and tube designs are proposed, then the heat generation by noble metals attached to tube surfaces can be a concern if both primary and secondary salts are drained from the heat exchangers, a rare but plausible event. ORNL determined this a manageable but concerning situation (ORNL TM-3145,1971). 

The issue of corrosion of Hastelloy N on exposure of fluoride-based fuel salts is one that has attracted a perhaps xmwarranted concern in the past. Numerous long-durahon salt loops emplo)dng various fuel salt and/or coolant salts performed very well (ORNL-TM-4286, 1972), and also several stainless steels such as 316SS (ORNL-TM-5782, 1977), while somewhat inferior to Hastelloy N, showed significant promise. 

During operation of the MSRE however, an issue of modest concern did surface. That being that the fission product tellurium was depositing within the grain boundaries of Hastelloy N ". Upon observation of test samples and reactor components post-shutdown, shallow surface cracking, less than 15 mils deep, was evident on surfaces exposed to the fuel salt. 

A program by ORNL later showed this issue to be manageable by a combination of changing the alloy makeup and reduction potential of the fuel salt (ORNL 4812,1972). The tellurium cracking however was widely cited in the WASH 1200 (Shaw, 1972) report used to justify the cancellation of the MSBR program in the early 1970s. 

Ta within a fuel salt also adds an interesting effect during long-term shutdown of the reactor. In this situation, there is potential increase in reactivity as Pa decays to fissile U. If it is intended that the reactor be kept in hot standby, shutdown systems must take this factor into account. 

For breeder MSR versions, on-site continuous processing is typically proposed. In the early work of ORNL, as rapidly as the entire fuel salt on a 10-day cycle. In more recent proposed designs with harder spectrums, this time can be extended to several months and still allow breeding. With the thorium- U cycle, a factor that greatly complicates processing is that thorium behaves chemically very similar to the lanthanide fission products. 

Mid-1960s efforts at ORNL sought to bypass this issue by using two separate liquids, a fuel salt containing the produced and a blanket salt containing the thorium to produce... 

The blanket salt would require periodic fluorination to remove the produced (although often rapid Ta removal was also suggested). The fuel salt, in a continuous side stream, would be first fluorinated to remove and then salt processed by vacuum distillation where at high temperature (1(XX)°C) the carrier salt would evaporate off and leave behind most fission products in the still bottoms. 

All processing methods however require far more development before commercial operations. While molten salt processes may be far simpler than solid fuel processes, they still represent a large capital and regulatory burden. As an example, the 1970s 1000 MWe MSBR design called for processing the entire fuel salt every 10 days, which would equate to 5600 tonnes processed per annum. 

---