China
Africa
Explore chapters and articles related to this topic
Multi-Phase Systems
Published in Marc J. Assael, Geoffrey C. Maitland, Thomas Maskow, Urs von Stockar, William A. Wakeham, Stefan Will, Commonly Asked Questions in Thermodynamics, 2022
Marc J. Assael, Geoffrey C. Maitland, Thomas Maskow, Urs von Stockar, William A. Wakeham, Stefan Will
In the absence of specific measurements, the parameters of the activity-coefficient model can be estimated using a group-contribution method which assumes that groups of atoms within a molecule contribute in an additive manner to the overall thermodynamic property for the entire molecule. Thus, a methyl group may make one kind of contribution while a hydroxyl group makes another contribution. Once the contributions to the property from each group of the molecule have been determined the activity coefficient of the molecule can be obtained from the contributions of the groups it contains. Schemes of this type ultimately rely on (vapor + liquid) equilibria measurements that are used with definitions of the groups within molecules to determine the parameters of a model for the molecular group by regression. Examples of this approach are the Analytical Solution of Groups (ASOG) (Wilson and Deal 1962; Wilson 1964; Kojima and Toshigi 1979) and the Universal Functional Group Activity Coefficients (UNIFAC) (Fredenslund et al. 1975; 1977) models; the UNIFAC method is widely used (Kontogeorgis and Folas 2010).
Ethyl trifluoroacetate formation as a means to recover trifluoroacetic acid from dilute aqueous mixture: reaction, separation and purification
Published in Journal of the Chinese Institute of Engineers, 2023
Reshma R. Devale, Amit Katariya, Yogesh S. Mahajan
kf is the forward rate constant, expressed in Arrhenius form. kf0 is the pre-exponential factor, and Ef as the activation energy. KM is equilibrium constant, and is autocatalysis constant. Objective function is the sum of squares of error (SSE). Ordinary differential equations (ODEs) representing batch reactor model (Equations 3,4) were implemented in MATLAB®. Using initial values of reactant concentrations and guess values of kinetic constants, the ODEs were solved with the help of ‘ode15s’ routine of MATLAB®. Simulated TFA concentrations were compared with the experimental data and SSE between these values was calculated (Equation 5). Nc is the number of components, Nexp is the number of experiments, and NTime is the number of time points. Kinetic constants were varied so as to minimize SSE. ‘Fminsearch’ code of MATLAB® was used for finding optimum values of parameters. Table 2 provides values of constants in the rate expression. ‘UNIQUAC’ parameters were estimated using group contribution method (‘UNIFAC’) of ASPEN PLUS and are given in Table 3. Figures 4(a-c) show comparison between experimental data and model prediction. Figure 4(d) shows that a good fit was obtained (parity plot, R2 = 0.9822). Equations 3–5 pertain to rates and SSE:
Design and optimization of an acetic acid recovery system via extraction–distillation using an isopropyl acetate + isopropanol mixed solvent
Published in Chemical Engineering Communications, 2020
Peng Fang, Chao Yu, Zuoxiang Zeng, Weilan Xue
For HAc recovery, IPA-01 containing a certain amount of IPA-02 was used as an extractant to remove water from its aqueous solution. The nonrandom two-liquid (NRTL) model was widely used to predict liquid activity coefficients (Renon et al., 1968). In this study, this model was selected as a thermodynamic model for the design of the whole process, which was conducted by Aspen Plus V9.0. All the NRTL parameters of this system are shown in Table 1, in which the parameters of water–HAc, HAc–IPA-01, HAc–IPA-02 were estimated by the UNIFAC group-contribution method (Fredenslund et al. 1975). The Hayden–O'Connell second virial coefficients are used to predict the vapor phase of systems (Hayden et al., 1975).
Selection of eco-efficient downstream separation configuration for isopropanol–butanol–ethanol purification process
Published in Chemical Engineering Communications, 2023
Ilayda N. Oksal, Devrim B. Kaymak
The simulations are carried out using Aspen Plus (V10.0). All distillation columns are represented with rigorous RadFrac module, while extractor and decanter are represented by Extract and Decanter modules, respectively. The internals of the columns are represented as equilibrium trays. NRTL is selected as the thermodynamic model of simulations for both process configurations. The NRTL binary interaction parameters (BIP), which do not exist in Aspen Plus, are estimated with UNIFAC group contribution method. Supporting Information Table S2 includes the BIP values estimated with UNIFAC group contribution method.