Prediction and optimization of nanoclusters-based thermal conductivity of nanofluids: Application of Box–Behnken design (BBD)

Abstract

ABSTRACTThe existing models to predict the thermal conductivity of nanofluids are based on single particle diameter, whereas, in actual solutions, nanoparticles mostly exist in a cluster form. Experiments are carried out to observe the effects of various surfactants on stability, nanocluster formation, and thermal conductivity of Al2O3–H2O nanofluid, which is found to be improved considerably with SDS surfactant. The prolonged sonication was not adequate to break the clusters of Al2O3 nanoparticles, into an average size of less than 163 nm, indicating the tendency of Al2O3 nanoparticles to remain in the form of clusters instead of individual nanoparticles of primary size of 20 nm. Response surface methodology has been employed to design and optimize the experimental strategy by taking volumetric concentration, temperature, and surfactant amount as the contributing factors. The developed model has been validated against the experimental data and the existing models with an accuracy level of ± 8% in the former case. Analysis reveals about the formation of nanoclusters and enhancement in thermal conductivity. The results confirmed that the model can predict thermal conductivity enhancement with an accuracy level of R square value of the order of 0.9766.