Thermal conductivity prediction of pure refrigerants and mixtures based on entropy-scaling concept

Abstract

The residual entropy scaling (RES) concept builds up connection between the transport properties and thermodynamic equation of state directly. However, the low-density divergence issue of the reduced dimensionless thermal conductivity might not be fixed properly through current literature models. This study proposes a new RES model coupled with the original perturbed-chain statistical associating fluid theory (PC-SAFT) equation of state (EoS) for the thermal conductivities of 11 typical refrigerants (including hydrocarbons, hydrofluorocarbons, hydrofluoroolefins and carbon dioxide) and their multicomponent mixtures. Model parameters are adjusted to saturated vapor and liquid data obtained from REFPROP. The combination of a modified reference thermal conductivity term as an extension of the original work by Rosenfeld et al. and the new formulation of the reduced term could provide reasonable predictive results for the investigated pure and mixture systems with AAD% of 3.95% and 5.76%, respectively. As expected, quasi-linear behaviors in both dilute and dense fluid regions could be observed for the refrigerant systems under study. The extended corresponding states (ECS) and the other two RES approaches reported in the literatures are used as reliable tools for assessing the predictive capability of the proposed model. However, the critical enhancement of thermal conductivity and the correction of the reference EoS especially near the critical point will be the priority in the future.