Estimation of solar receiver corrosion conditions during operation to aid in the design of receiver-corrosion lab tests

Abstract

Solar power tower plants widely rely on molten nitrite salt as heat transfer fluid and storage media. However, its corrosiveness endangers several metallic components, with central receivers being one of the most compromised. Great efforts are being made to establish molten salt corrosion in different metals. Recently, the importance of the amount of salt used in these experiments has been highlighted. Yet, most tests disregard the hot-receiver-surface/salt ratio that would represent the receiver operation conditions. Commonly, larger metal surfaces are employed, accelerating molten salt thermal decomposition kinetics. Hence, this study seeks to develop basic design conditions for experimental corrosion tests to realistically mimic receiver-tube corrosion. Consequently, the high-temperature operation of two receiver designs, resembling those of Crescent Dunes and Gemasolar plants, with their tubes manufactured in HA230, is analysed. The yearly salt degradation in these plants is found to be nearly negligible due to the large salt volume stored and the relatively limited receiver surface above concerning temperatures, with the whole salt volume being exposed to it for around 20 min during a 25-year operation for large receivers. Moreover, the hot-metal-surface/solar-salt ratio ranges from 5 to 7 mm2/kg, depending on the receiver design, which should be preserved in a corrosion experiment. Metal loss is also estimated: less than 7 % of the receiver would lose at least 50 % of the initial tube thickness. Still, lower metal loss rates are expected since coefficients from static-isothermal tests were employed for the calculation, which also disregarded the metal surface/salt proportion, leading to an overestimation.