| Summary: | 碩士 === 東海大學 === 化學工程學系 === 91 === Abstract
In the design of separation processes involving high-purity chemicals or design of waste water treatment equipment, it is oftentimes necessary to have dilute-end vapor liquid equilibrium data available in advance. On the other hand, data of activity coefficient at infinite dilution is an important part of dilute-end vapor liquid equilibrium data. When available, it can be used to solve for parameters in many activity coefficient models which may in turn determine the vapor liquid equilibrium behavior for the entire concentration range. In addition, key information, such as Henry’s constant, partition coefficient, selectivity of mass separating agents being used in extractive distillation, azeotropic distillation, or extraction processes, and partial molar excess enthalpy at infinite dilution, needed in designing and developing separation processes can be subsequently obtained.
Binary acetone-H2O system has been used as the test system for this study. Specifically, three headspace sampling-based experimental methods — equilibrium headspace sampling method (EHS), phase ratio variation method (PRV), and multiple headspace extraction method (MHE) were employed to measure dilute-end vapor liquid equilibrium data and infinite-dilution activity coefficient data. They have been compared with literature data. Advantages and disadvantages of these methods were also looked into.
Our findings showed that it was feasible to use such methods to properly measure dilute-end and infinitely dilute vapor liquid equilibrium data. The infinite-dilution activity coefficient values taken at three temperature levels by using the PRV and MHE methods turned out to be very consistent. Even though there were some differences between the experimental and scattered literature data, the dependence of experimental infinite-dilution activity coefficient data for three different methods with respect to temperature was in very satisfactory agreement. It is clear that the dilute-end equilibrium data generated by the methods, when combined with finite-concentration vapor liquid equilibrium data, will aid in a full understanding of the vapor liquid equilibrium behavior.
|
|---|
