The New Jersey Institute of Technology's
Electronic Theses & Dissertations Project

Title:	Programming for the prediction of thermodynamic phenomena
Author:	Rim, One Kwon
View Online:	njit-etd1988-045 (199 pages ~ 4.5 MB pdf)
Department:	Department of Chemical Engineering, Chemistry and Environmental Science
Degree:	Master of Science
Program:	Chemical Engineering
Document Type:	Thesis
Advisory Committee:	Knox, Dana E. (Committee chair) Roche, Edward Charles, Jr. (Committee member) McCormick, John E. (Committee member)
Date:	1988-01
Keywords:	Thermodynamics -- Data processing Vapor-liquid equilibrium -- Data processing Liquid-liquid equilibrium -- Data processing Heat of mixing -- Data processing
Availability:	Unrestricted
Abstract:	A Thermodynamic Calculation Program was developed. The program contains the calculation procedure for VAPOR/LIQUID EQUILIBRIUM(Isothermal P-X data and T-X data), HEAT-OF-MIXING(liquid mixture data) and LIQUID/LIQUID EQUILIBRIUM SEPARATION data. The program may optionally use the Knox Molecular(GAM), UNIQUAC, NRTL, Wilson and Redlich-Kister equations, and is prepared to accept any kind of model which will be developed in the future. This program also performs parameter file creation, parameter input, updating of parameters, and is structured to allow implementation of data regression when program is updated. Comparison and analysis of each function were executed using this program. It was revealed that VLE(Vapor-Liquid Equilibrium) predictions are most easily made, while LLES (Liquid-Liquid Equilibrium Separation) predictions can be performed in limited ranges and Heat-of-mixing (Liquid mixture) predictions showed relatively poor results. The models which were used to predict the above were also compared and analyzed with each other. The Knox molecular(GAM) and modified UNIQUAC equations showed the most best performance, but each model has its advantages and disadvantages. It was concluded that all systems can not be treated by only one model. Each system must select the most suitable model which fits well for its system. Eventually the model which can perform all predictions using the same parameter values must be developed. Also, there is a need for a model which can also treat electrolyte systems.