Does the tube become less transparent because of contaminants? Over what time scale? Is this issue exacerbated before the system is operating at full temperature (e.g., coking)? Is the sodium vapor kept in the closed cavity or is it a consumable? If a consumable how much is needed? How is it stored and dispensed?
we don't see any degradation in sapphire tubes, though quartz, which is more convenient to work with because it almost completely resists thermoshock, does degrade slowly. there is a layer of salt on the tube which becomes transparent when melted, above 800 C. sodium vapor is provided to the reaction tube via direct evaporation -- melted sodium has a high surface tension and surface affinity for alumina, and wicks into the chamber. after combustion as it cools, it reforms into sodium chloride. for all fuels we've explored, sodium-chlorine is the maximum bond energy, but you can have some swaps if you have for some reaction alkali or fluorine in your fuel (don't!), the sodium chloride condenses from 800-1400C in the heat exchanger, and then wicks itself back along the surface to where it is evaporating. We hope to drive this process to some number of 99.99..% recovery, and just add granular salt (or could be a solution) to replenish. There is only a few % of salt needed in the flame, and if you recover 99.9% of the salt then you would have hundreds of total refuelings before you need to replenish a salt vessel of about 1%.