To achieve high thermal-to-electric conversion efficiency and to make solar technologies cost-competitive with conventional electric power generation, the U.S. Department of Energy (DOE) launched the SunShot Initiative in 2011. Unlike photovoltaic (PV) systems, concentrating solar power (CSP) technology captures and stores the sun’s energy in the form of heat, using materials that are low cost and materially stable for decades. This allows CSP with thermal energy storage (TES) to deliver renewable energy while providing important capacity, reliability and stability attributes to the grid, thereby enabling increased penetration of variable renewable electricity technologies. Today’s most advanced CSP systems are towers integrated with 2-tank, molten-salt TES, delivering thermal energy at 565°C using molten nitrates for integration with conventional steam-Rankine power cycles.

Higher efficiencies are obtained integrating CSP plants with a supercritical CO₂ (sCO₂) Brayton power cycle. To achieve this integration the next generation CSP (Gen3 CSP) needs to operate at temperatures above 550°C requiring high-temperature advanced fluids in the range of 550°C to 750°C. New salts are required to operate in this higher temperature range because nitrates are unstable at temperatures above 620°C. Selection of a high-temperature molten salt is needed, especially with regard to its compatibility with containment materials with acceptable mechanical strength, durability, and cost targets at these high temperatures. Chloride and carbonate salt blends have been proposed and tested, but each brings new challenges. The corrosion mechanism differs among candidate salts and information is needed for CSP component designers.

Because of their low cost, and high decomposition temperatures, molten chlorides are the top candidates. However, these molten salts introduce a set of technological and engineering challenges because of their very corrosive characteristics for typical materials. Corrosion mitigation approaches are been investigated to obtain degradation of containment materials around 20 µm/year or lower. From the salt handling, and thermal energy density point of views molten ternary eutectic carbonate Na₂CO₃/K₂CO₃/Li₂CO₃ is the best heat transfer fluid and TES for Gen3CSP, but its cost is extremely high.

Corrosion in molten chlorides is controlled in atmospheres without oxygen and water. If these impurities are present, molten chlorides become very corrosive in the liquid and vapor phases. Catastrophic mechanical failure is then possible because intergranular attack is the corrosion mode. Some researchers have proposed the redox potential control using active-metals such as Mg to reduce corrosion rates to below 10 µm/year at 800°C but the use of no-oxygen/water in the atmosphere is also required. Other corrosion mitigation approach is the use of surface treatments such as in-situ passivation, pre-oxidation, coatings, and diffusive coatings such as boronizing and aluminizing.

Several nickel-based (NiCo)CrAl(Y,Ta,Hs,Si) coatings have been tested in molten carbonates and chlorides from 650°C to 750°C using electrochemical techniques to determine corrosion rates, and mechanisms. Several high-alloyed stainless steels, nickel superalloys, and alumina forming alloys have been tested.

Untreated In800H and 310SS alloys corroded rapidly (~2,500 to 4,500 µm/year) in molten chlorides and carbonates. The lowest corrosion rate in molten chlorides of 190 µm/yea was obtained for atmospheric plasma spray NiCoCrAlY coatings pre-oxidized in zero air (ZA) at 900°C for 24 h with a heating/cooling rate of 0.5°C/min. The corrosion of the alloys exposed to molten carbonate at 700°C in bone-dry CO₂ atmosphere was reduced from ~2,500 µm/year to 34 µm/year when coated with high-velocity oxyfuel NiCoCrAlHfSiY and pre-oxidized (ZA, 900°C, 24 h, 0.5°C/min). Metallographic characterization of the corroded surfaces showed that the formation of a uniform thin alumina scale before exposure to the molten salts considerably reduced the corrosion of the alloy. However, the rates of corrosion determined herein are still large, highlighting the relevance of testing materials durability in solar power applications.

Further electrochemical impedance spectroscopy tests and metallographic characterization showed that the best performing alumina forming alloy was In702 pre-oxidized in ZA at 1050 °C for 4 h due to the formation of protective, dense and continuous alumina layers. But these layers were unstable when argon was used as the carrier gas during corrosion evaluations. Corrosion results in static ZA are promising for next-generation CSP applications using molten chlorides because alumina scales were stable after 185 h of immersion in the oxygen-containing atmosphere. Alumina layers in pre-oxidized AFA In702 grew from 5 µm (before immersion) to 13 µm (after 185 h of immersion). The use of these alloys could be commercial feasibility and cost-effective because of the possibility of using oxygen-containing atmospheres instead of keeping enclosed systems with inert atmospheres to protect alloys from corrosion in molten chlorides.