TY - JOUR
T1 - Entropy Scaling of Viscosity - III
T2 - Application to Refrigerants and Their Mixtures
AU - Yang, Xiaoxian
AU - Xiao, Xiong
AU - May, Eric F.
AU - Bell, Ian H.
PY - 2021/3/11
Y1 - 2021/3/11
N2 - The residual entropy scaling of viscosity was applied to pure refrigerants, including natural refrigerants, hydrofluoroolefins, hydrochlorofluoroolefins, perfluorocarbons, hydrofluorocarbons, chlorofluorocarbons, and hydrochlorofluorocarbons and their mixtures. Experimental temperature, pressure, and viscosity data of 39 pure refrigerants, including more than 15,000 experimental data values from more than 400 literature sources, were used to build a univariate correlation function between the reduced residual viscosity and the dimensionless residual entropy. The correlation function contains only four fitted parameters and a fluid-specific scaling factor. Approximately, 80.0% of the experimental data are predicted within 5.0% when the fluid-specific fitted parameters are used. About 80.0% of the experimental data collapse onto one single curve within 7.9% when the global fitted parameters and the fluid-specific scaling factor were adopted for the correlation function. The correlation function is able to predict mixture viscosity without any additional empirical parameters. Approximately, 80.0% of the experimental data of 27 binary or multi-component mixtures composed by the investigated pure components, encompassing 2890 experimental values from more than 20 literature sources, agree with the correlation function within 7.9%, which is as good as the comparison for pure fluids. The commonly used extended corresponding states model has as many as four more parameters for each pair of components and has been optimized for some of the binaries; therefore, it generally yields better agreement than the proposed correlation function for binary mixtures but similar performance for multi-component mixtures.
AB - The residual entropy scaling of viscosity was applied to pure refrigerants, including natural refrigerants, hydrofluoroolefins, hydrochlorofluoroolefins, perfluorocarbons, hydrofluorocarbons, chlorofluorocarbons, and hydrochlorofluorocarbons and their mixtures. Experimental temperature, pressure, and viscosity data of 39 pure refrigerants, including more than 15,000 experimental data values from more than 400 literature sources, were used to build a univariate correlation function between the reduced residual viscosity and the dimensionless residual entropy. The correlation function contains only four fitted parameters and a fluid-specific scaling factor. Approximately, 80.0% of the experimental data are predicted within 5.0% when the fluid-specific fitted parameters are used. About 80.0% of the experimental data collapse onto one single curve within 7.9% when the global fitted parameters and the fluid-specific scaling factor were adopted for the correlation function. The correlation function is able to predict mixture viscosity without any additional empirical parameters. Approximately, 80.0% of the experimental data of 27 binary or multi-component mixtures composed by the investigated pure components, encompassing 2890 experimental values from more than 20 literature sources, agree with the correlation function within 7.9%, which is as good as the comparison for pure fluids. The commonly used extended corresponding states model has as many as four more parameters for each pair of components and has been optimized for some of the binaries; therefore, it generally yields better agreement than the proposed correlation function for binary mixtures but similar performance for multi-component mixtures.
UR - http://www.scopus.com/inward/record.url?scp=85100775448&partnerID=8YFLogxK
U2 - 10.1021/acs.jced.0c01009
DO - 10.1021/acs.jced.0c01009
M3 - Article
AN - SCOPUS:85100775448
SN - 0021-9568
VL - 66
SP - 1385
EP - 1398
JO - Journal of Chemical and Engineering Data
JF - Journal of Chemical and Engineering Data
IS - 3
ER -