A New Fast Estimation Method for Critical Properties of Mixtures Based on the Modified Redlich–Kister Method-Part 2: Prediction of Critical Properties for Binary and Ternary Mixtures

The
vapor–liquid critical properties of mixtures,
which
represent the end points of vapor–liquid equilibrium curves,
are crucial for the development of next-generation environmentally
friendly working fluids and advancements in supercritical fluid technology.
Experimental measurement and theoretical prediction are the main means
to obtain the critical properties. However, the existing theoretical
prediction methods have the problems of low accuracy, especially when
predicting critical volumes of binary mixtures and critical properties
of ternary mixtures. Moreover, in most prediction methods, all the
data are used to fit the adjustable parameters. No prediction data
set was added to test its prediction ability. In this work, new prediction
models for critical properties, including critical temperature, critical
pressure, and critical volume of binary mixtures and ternary mixtures,
were proposed. New methods extend our previous work (Tang et al.’s
model) to the prediction of critical volumes and critical properties
of ternary mixtures and accurately evaluate the extrapolation ability.
New prediction models inherit many advantages of Tang et al.’s
model, including considering the effect of molecular polarity to some
extent, possessing four fewer adjustable parameters, a simpler form
in use, and not requiring the critical volume data of pure substances
when predicting critical temperatures and critical pressures of mixtures.
Moreover, three new methods possess fast prediction abilities for
critical properties of mixtures, showing higher accuracy in predicting
critical properties of binary and ternary mixtures in the fitting
data set and prediction data set. When predicting critical temperatures
and critical pressures, new models can be applied to binary and ternary
mixtures consisting of methane-free alkanes, alkenes, alkynes, alicyclic
hydrocarbons, benzene and its derivatives, NH3, CO2, halogenated hydrocarbon, N2O, Kr, Xe, sulfur-compounds,
and oxygen-containing organic compounds. Notably, new methods are
applicable only to class I (continuous curve of the critical point
of two pure components) of the vapor–liquid critical locus,
classified by Van Konynenburg and Scott. Class I covers many systems
and most of the available critical property experimental data. Almost
all systems can be covered in predicting critical volumes. About 9000
critical data points including critical temperatures, critical pressures,
and critical volumes for binary mixtures and ternary mixtures are
collected. About 80% of the data for binary mixtures were used to
fit the empirical parameters. The remaining data were used to test
the prediction ability for binary and ternary mixtures for three new
prediction methods. The prediction model 1 shows the highest correlation
and prediction accuracy among the three prediction models. The average
absolute relative deviations are 1.13, 3.91, and 5.90 when correlating
critical temperatures, critical pressures, and critical volumes; 1.43,
4.94, and 6.93 when predicting critical temperatures, critical pressures,
and critical volumes of binary mixtures; and 1.12 and 3.99 when predicting
critical temperature and critical pressures of ternary mixtures.