Thermodynamics and Machine Learning Based Approaches for Vapor–Liquid–Liquid Phase Equilibria in n-Octane/Water, as a Naphtha–Water Surrogate in Water Blends

Abstract

The prediction of phase equilibria for hydrocarbon/water blends in separators, is a subject of considerable importance for chemical processes. Despite its relevance, there are still pending questions. Among them, is the prediction of the correct number of phases. While a stability analysis using the Gibbs Free Energy of mixing and the NRTL model, provide a good understanding with calculation issues, when using HYSYS V9 and Aspen Plus V9 software, this shows that significant phase equilibrium uncertainties still exist. To clarify these matters, n-octane and water blends, are good surrogates of naphtha/water mixtures. Runs were developed in a CREC vapor–liquid (VL_ Cell operated with octane–water mixtures under dynamic conditions and used to establish the two-phase (liquid–vapor) and three phase (liquid–liquid–vapor) domains. Results obtained demonstrate that the two phase region (full solubility in the liquid phase) of n-octane in water at 100 °C is in the 10-4 mol fraction range, and it is larger than the 10-5 mol fraction predicted by Aspen Plus and the 10-7 mol fraction reported in the technical literature. Furthermore, and to provide an effective and accurate method for predicting the number of phases, a machine learning (ML) technique was implemented and successfully demonstrated, in the present study.

Integrated Project as Innovative Assessment to Enhance Learning Experience in Thermodynamics Class

“…The project covers the vapor/liquid equilibrium of binary mixture and calculation related to it by involving HYSYS as a simulation tool. HYSYS itself has been widely used to explore the simulation of thermodynamics concepts related to ideal and non-ideal mixtures (Lopez-Zamora et al, 2021). Through this project, students can relate the concepts taught in class and HYSYS as simulation software can be used to produce the graphs related to vapor/liquid equilibrium variables such as the composition of liquid and vapor at different temperatures and pressures, as well as to perform the necessary calculation (Figure 3).…”

Vapour–liquid–liquid and vapour–liquid equilibrium of paraffinic aromatic synthetic naphtha/water blends: Prediction of the number of phases

“…CREC‐VL‐Cell: (1) stirring head, (2) isolation shells, (3) stirring hot plate, (4) aluminium vessel, (5) VL cell, (6) thermofield, and (7) hydrocarbon/water blend. (This figure was originally presented as part of our previous work [ 9 ] )…”

“…The stability analysis for binary systems can be extended to multicomponent mixtures, by minimizing the Gibbs Free energy of mixing, as reported in our previous work. [ 9 ] In the case of a multicomponent mixture, the can be defined by a surface instead of a curve, and the tangent line expressed as a tangent plane. [ 17 ] This tangent plane distance (TPD) approach was originally proposed by Baker et al [ 18 ]…”

“…In binary systems, such as n ‐octane/water mixtures, an activity coefficient model can be used, as we proposed in our previous work. [ 9 ] However, classical activity coefficient models are limited to low pressures (≤1 MPa) with no C 7+ being contained. Therefore, for water and heavy hydrocarbon blends such as naphtha or bitumen (C 7+ ) in water streams, as in the case of this study, a cubic EoS is strongly suggested.…”

“…[ 8 ] In this work, analyses of (i) the phase stability using Soave‐Redlich‐Kwong EoS with a Kabadi‐Danner modification (SRKKD EoS), (ii) the experimental data reported by Kong [ 8 ] for this mixture, and (iii) a machine learning approach used to predict the number of phases in PASN/water systems, are presented. The machine learning (ML) methodology developed in our previous work, [ 9 ] for n ‐octane/water systems, was adapted for PASN/water mixtures.…”