Ravisankar, Vijay (2016-11). Precipitate Generation and Characterization of Corrosion Source Alumina for GSI-191 Chemical Effects Issue. Master's Thesis.
Thesis
The release of corroded alumina during a Loss of Coolant Accident (LOCA) leads to deposition of these particles on the strainer, which consequently introduces a head loss that can cause shutdown of the cooling system. Salts of alumina have been used as surrogates for generating corrosion source particles to create a possible post-LOCA environment and for head loss testing. However, previous studies have shown that alumina surrogates tend to produce higher artificial head loss in the system compared to representative post-LOCA alumina particles, resulting in over-design of containment systems. It is therefore important to identify exactly what particles are generated, and relate their composition and properties to the corresponding head loss generated in order to develop more accurate predictions of LOCA scenarios. But surrogate formulations employed in head loss testing are likely not representative of actual corrosion products, making it difficult to assess the reliability of predictions based on these experiments. The goal of this study is to address the need for more pertinent data by exploring surrogate formulations produced from corrosion products obtained under chemical and thermal conditions mirroring those encountered during plant operation. The resulting precipitates are characterized physically and chemically using techniques including turbidity analysis, particle size distribution, SEM and to establish the influence of variables like pH and cooling rate. The resulting surrogates are then employed to assess head loss under LOCA conditions in a pilot-scale test facility. Our findings indicate that the quantity of aluminum corroded increased from 44 mg/L to 70 mg/L with an increase in pH from 7.2 to 7.5. The size of the particles increased with a decrease in cooling rate, with rapidly cooled particles having a diameter of approximately 12 nm irrespective of pH, whereas intermediate and slow cooled particles increase in size with increasing pH. The results of these studies lay a foundation for development of improved surrogate formulations that can be used to obtain more accurate predictive capabilities to better mitigate LOCA events.
