Statistical modeling of copper slag heat capacity for thermal storage

Abstract

This study presents a statistical approach for the analysis of the specific heat capacity (cp) of Chilean copper slag, a by-product of the copper mining industry, as a filler material for packed-bed thermal energy storage (PBTES). The approach addresses the challenge of using by-products “as-received” by considering material variability, which affects the prediction of the packed-bed performance. A statistical model was developed using the Akaike Information Criterion to select the mathematical model of the cp, based on the 5%, 50%, and 95% percentiles of the data. Such percentiles were then used to run a validated numerical model of a PBTES, which is contrasted with experimental data and a simulated case study. Results showed rapid thermocline advancement in the P5 model and slower propagation in the P95 model. The contrast with the PBTES experimental data suggest that the statistical representation of cp can effectively predict the temperature range where the thermocline develops during the charging process. However, minor changes in roundtrip efficiency suggest that PBTES efficiency is more influenced by design than it is by cp variability of its filler materials. These findings mark an initial step toward using high-variability materials and providing cost-effective energy storage solutions through the valorization of industrial residues.