Convex envelope method for T , p T,p flash calculations for mixtures with an arbitrary number of components and arbitrary aggregate states

Quirin Göttl, Natalie Rosen, Jakob Burger · 2025

Read Convex envelope method for T , p T,p flash calculations for mixtures with an arbitrary number of components and arbitrary aggregate states by Quirin Göttl, Natalie Rosen, Jakob Burger (2025) in the Tower of Babel, a free online library of public-domain works preserved in clean, searchable text. No sign-up. The full text is served as a plain-text file at a predictable URL, where text is present: https://antarctic-labs.com/tower-of-babel/arxiv-2502-09402.txt.

Introduction

Quirin Göttl, Natalie Rosen, and Jakob Burger extend a geometry-based method for phase equilibrium math, in this 2025 paper. Flash calculations tell engineers which phases a mixture splits into at a set temperature and pressure, and how much of each forms. The convex envelope method handles liquids of any blend without being told the phase count in advance.

The extension now covers vapor and solid phases too, across the whole range of mixtures. The authors show vapor-liquid and solid-liquid examples with up to four components. They also use the method to fit model parameters and pair it with a learning tool that predicts properties.

Catalog record

CREATORQuirin Göttl, Natalie Rosen, Jakob Burger
YEAR2025
COLLECTIONPAPERS
FORMATTXT
FILE SIZE43 KB
VERSION1.0
LICENSEhttp://creativecommons.org/licenses/by/4.0/
RIGHTSCC BY
SOURCEarXiv

Source: https://arxiv.org/abs/2502-09402

Browse the full library · About the Tower of Babel

Full text

The opening of the work, from the library's hosted text:

Convex envelope method for T , p T,p flash calculations for mixtures with an arbitrary number of components and arbitrary aggregate states Quirin Göttl, Natalie Rosen, Jakob Burger Abstract T , p T,p flash calculations determine the correct number of phases at phase equilibrium and their compositions for fixed temperature and pressure. They are essential for chemical process simulation and optimization. The convex envelope method (CEM) is an existing approach that employs the tangent plane criterion to determine liquid phase equilibria for mixtures with an arbitrary number of components without providing the number of phases beforehand. This work extends the CEM to include also vapor and solid phases. Thus, any phase equilibrium of a given mixture with an arbitrary number of components and phases can be calculated over the whole composition space. The CEM results are presented for various vapor-liquid and solid-liquid phase equilibria examples of up to four components. We show how the CEM can be used for parameter fitting of g E g^{E} -models. As an outlook, we demonstrate how the CEM can be combined with a machine learning-based tool for property prediction to construct phase equilibria. Journal: Computers & Chemical Engineering † † graphicalabstract: † † highlights: Generalization of the convex envelope method to phase equilibria containing solid, liquid, and gaseous aggregate states. Numerical results for vapor-liquid equilibria and solid-liquid equilibria for mixtures containing up to four components. An open source implementation of the framework in Python. 1 Introduction T,p flash calculations are important in practice to determine all types of phase equilibria ( Lucia et al., 2000 ) . Calculation of phase equilibria, in turn, is crucial for chemical process simulation and optimization ( Zhang et al., 2011 ) . Generally, the numerical calculation of phase equilibria is an optimization problem. For a T,p flash (i.e., the calculation of the phase equilibrium at constant temperature and pressure) the Gibbs free energy G of the mixture has to be minimized ( Gibbs, 1873 ; Gibbs, 1876 ) . Thus, a pre-requisite to do a T,p flash is a thermodynamic model of G or other potentials from which it can be derived. There are two main approaches for the respective thermodynamic models: equations of state (EOS) models such as Peng-Robinson ( Peng and Robinson, 1976 ) , Redlich–Kwong-Soave ( Soave, 1972 ) , SAFT ( Chapman et al., 1989 ) , or g^{E} models such NRTL and UNIQUAC ( Prausnitz et al., 1999 ) . While g^{E} models are well suited for describing liquid phase behavior, EOS models are applicable to both liquid and vapor phases and provide a continuous transition between them. A detailed discussion of the theoretical foundations of these models is beyond the scope of this work, and interested readers are referred to the sources cited above. We will simply assume that such thermodynamic models are given. The calculations of the T,p flash are greatly simplified, if the types and number of coexisting phases in equilibrium are known. In this case, the solution of the minimization problem can be transformed into an algebraic equation problem, resulting in well known equations such as the (extended) Raoult’s law for the vapor-liquid equilibrium (VLE) or the iso-activity criterion for the liquid-liquid equilibrium (LLE). In the present work, we will consider however the general case, in which the type and number of phases is not known a priori and the general optimization problem has to be solved. We classify the approaches to solve the general optimization problem into two classes. On the one hand, there are mathematical programming approaches which are either deterministic (e.g., ( Mitsos and Barton, 2007 ; Pereira et al., 2010 ; Perevoshchikova et al., 2012 ; Wasylkiewicz et al., 2013 ) ) or stochastic (e.g., ( Bonilla-Petriciolet et al., 2011 ) ). For a more thorough review of such optimization algorithms for phase equilibrium we refer to ( Zhang et al., 2011 ; Piro and Simunovic, 2016 ; Domanska, 2019 ) . One the other hand, there are geometrically motivated approaches which find stable phases based on tangents (or more generally convex hulls in case of higher dimensions), on the G state surface (or graph of G ) in the composition space, resulting in the tangent plane criterion ( Baker et al., 1982 ; Michelsen, 1982a ; Michelsen, 1982b ) . Examples for such approaches are given in ( Sorensen et al., 1979 ; Swank and Mullins, 1986 ; Teh and Rangaiah, 2002 ) . Another geometrically motivated approach is the class of so-called terrain methods ( Lucia and Feng, 2002 ; Lucia and Feng, 2003 ) . These methods are designed to solve chemical process simulation problems, such as nonlinear continuous stirred tank reactors or T,p flash calculations, by intelligently following specific integral curves. In ( Lucia et al., 2005 ) , terrain methods are applied to a range of multiphase equilibrium problems, including solid–solid equilibria and solid–liquid equilibrium (SLE). Building on the tangent plane criterion ( Baker et al., 1982 ) , numerous approaches have been developed based on the convex envelope method (CEM) ( Ryll, 2009 ; Ryll et al., 2012 ; Göttl et al., 2023 ) . Similar methods are often referred to as the method of the convex hull ( Lee et al., 1992 ) , which has been applied to the calculation of binary and ternary SLEs. Other works have employed convex hull approaches for the calculation of binary and ternary VLE ( Hildebrandt and Glasser, 1994 ; Afanasyev, 2012 ; Joseph et al., 2017 ) . The CEM is fundamentally different from other T,p flash methods. While most of them have the goal to solve the T,p flash problem for a given material feed as efficienctly as possible, the CEM follows a two-step approach. In the first step, which can come with quite some computational effort, the graph of the Gibbs free energy G is computed for the whole discretized composition space of a given mixture. By constructing the convex envelope (or convex hull) for this graph and employing the tangent plane criterion ( Baker et al., 1982 ) , an approximation (due to the discretization) of the phase equilibrium diagram can be stored in a piecewise linear representation. In the second step, which is in principle decoupled from the first step, T,p flash calculations can be done for any given feed composition extremely fast by solving a small linear problem. Both steps are robust in the sense that they avoid the solution of nonlinear equations or a nonlinear optimization problem. The CEM is a method for determining phase equilibria with an arbitrary number of phases, i.e., the number of phases is determined by the CEM. In ( Ryll, 2009 ; Ryll et al., 2012 ) , the CEM was shown to work for liquid phase equilibria for mixtures with up to four components. A graphical evaluation scheme was implemented that distinguishes several cases for the number of existing phases and makes the decision on a case-by-case basis. This work was continued in ( Göttl et al., 2023 ) , where a general mathematical framework was provided that allows the application of the CEM to mixtures with an arbitrary number of components and an arbitrary number of phases

Read the complete text