A critical study of some representative models has been conducted with the experimental determinations of the viscosity of ten pure substances and three binary systems measured as a function of temperature T(25℃,40℃...A critical study of some representative models has been conducted with the experimental determinations of the viscosity of ten pure substances and three binary systems measured as a function of temperature T(25℃,40℃,60℃,80℃,90℃)and pressure P(≤ 100 MPa). The pure substances were:toluene,p-xylene,m-xylene,o-xylene,methylcyclohexane, methylnaphtalene,decahydronaphtalene,phenyldodecane,heptamethylnonane and tetramethylpentadecane(pristane).The three binaries were:toluene+ tetramethylpentadecane,toluene+ methylnaphtalene and toluene+ heptamethylnonane,for molar fractions x of toluene ranging between 0 and 1.The three binaries are highly'contrasted' systems,i.e.systems in which the viscosities of the pure components are very different for each P,T pair.In all,547 experimental determinations were carried out,the 279 experimental data for viscosity of the pure substances and 268 data concerning the mixtures(x ≠0 and 1) were used to make a critical analysis of the various representative models,These included an empirical representation of variations of versus T on each isobar,a self-referencing model developed in our laboratory,a corresponding states model with one reference,various mixing rules with or without adjustable parameters,a combination of the self-referencing model and mixing rules,a method based on the Bloomfield and Dewan excess viscosity,and a group contribution method,comparison of the results showed the self-referencing method to be of particular interest,Moreover when coupled with mixing rules without adjustable parameters it was able to account satisfactorily for the variationsη(P,T,x) of the mixtures,on the simple basis of knowledge if the viscosity of the components of the system,at 25℃ and at atmospheric pressure,For all the binaries an average absolute deviation of 9.5% was observed.Finally it appears that the introduction of the pressure parameter does not alter the usual performance of each of the models tested,although they were not al equivalent as regard their performances.展开更多
文摘A critical study of some representative models has been conducted with the experimental determinations of the viscosity of ten pure substances and three binary systems measured as a function of temperature T(25℃,40℃,60℃,80℃,90℃)and pressure P(≤ 100 MPa). The pure substances were:toluene,p-xylene,m-xylene,o-xylene,methylcyclohexane, methylnaphtalene,decahydronaphtalene,phenyldodecane,heptamethylnonane and tetramethylpentadecane(pristane).The three binaries were:toluene+ tetramethylpentadecane,toluene+ methylnaphtalene and toluene+ heptamethylnonane,for molar fractions x of toluene ranging between 0 and 1.The three binaries are highly'contrasted' systems,i.e.systems in which the viscosities of the pure components are very different for each P,T pair.In all,547 experimental determinations were carried out,the 279 experimental data for viscosity of the pure substances and 268 data concerning the mixtures(x ≠0 and 1) were used to make a critical analysis of the various representative models,These included an empirical representation of variations of versus T on each isobar,a self-referencing model developed in our laboratory,a corresponding states model with one reference,various mixing rules with or without adjustable parameters,a combination of the self-referencing model and mixing rules,a method based on the Bloomfield and Dewan excess viscosity,and a group contribution method,comparison of the results showed the self-referencing method to be of particular interest,Moreover when coupled with mixing rules without adjustable parameters it was able to account satisfactorily for the variationsη(P,T,x) of the mixtures,on the simple basis of knowledge if the viscosity of the components of the system,at 25℃ and at atmospheric pressure,For all the binaries an average absolute deviation of 9.5% was observed.Finally it appears that the introduction of the pressure parameter does not alter the usual performance of each of the models tested,although they were not al equivalent as regard their performances.
