| Summary: | 碩士 === 東海大學 === 化學工程學系 === 92 === We often rely heavily on vapor-liquid equilibrium (VLE) data to properly design separation processes. Yet, vapor-liquid equilibrium data at the dilute end in particular is critical in assessing separation processes involving high-purity chemicals or close-boiling mixtures. Nevertheless, much of the emphasis has been conventionally placed on non-dilute end or so-called finite concentration region. Dilute-end phase equilibrium behavior is usually generated from the extrapolation of finite-concentration vapor-liquid equilibrium data. In this way, the regressed data cannot accurately represent true phase equilibrium behavior especially in the dilute region. In fact, many separation processes, such as wastewater treatment, extractive distillation, and azeotropic distillation processes, frequently use the phase equilibrium data in this particular region as the basis for process design and development purposes.
In this work, the finite-concentration vapor-liquid equilibrium data for the propylene glycol monomethyl ether (PM)- propylene glycol monomethyl ether acetate (PMA)- acetic acid (HOAc) system were first measured with a dynamic recirculation still. Specifically, isobaric VLE measurements were made for the binary systems PM+HOAc, PMA+HOAc, and PM+PMA at 101.3KPa. In regressing the data using the NRTL model, vapor phase nonidealities due to association of acetic acid were accounted for in the first two binary systems. In order to compare infinite dilution activity coefficient data from other methods, VLE data for the PM+PMA system were also measured at six sub-ambient pressures. Satisfactory regression results were obtained for the above data, and optimized NRTL parameters were also found. Additionally, all of the measured data passed the integral test for thermodynamic consistency.
Three experimental methods involving headspace gas chromatography were used to carry out isothermal dilute-end VLE measurements (at 60, 80, and 100℃) - equilibrium headspace sampling method, phase ratio variation method, and multiple headspace extraction method. Infinite dilution activity coefficients of PM in PMA and of PMA in PM, and , at these three temperatures were obtained, along with partial molar excess enthalpies at infinite dilution, and . The values of and from all of the preceding methods were compared and the agreement was less than satisfactory. However, the values at 100 oC measured by the phase ratio variation method were found to be comparable to those calculated using the low-pressure VLE data. Likely errors and deficiencies of the methods were discussed.
|
|---|
