Homogeneous Azeotropic Distillation: Comparing Entrainers z L. LAROCHE*, N. BEKIARIS, zyxwvu H. W. ANDERSENt and M. MORARIS Chemical Engineering 210-41, California Institute of Technology, Pasadena, CA 91125, U.S.A. In this article, we present practical solutions (in the case of entrainers which add no azeotropes) to two problems of industrial relevance: Given a binary azeotrope which we want to separate into pure components, and a set of candi- date entrainers, how do we determine which one is the best? zyxwvu Also, for each of these entrainers, what is the flowsheet of the feasible separation sequence(s)? We obtain these solutions by analyzing in details the mechanisms by which heavy, intermediate and light entrainers make separation feasible, using the new notions of equivolatility curves, of isovola- tility curves and of local volatility order. We show that the second question finds an easy solution from the volatility order diagram. This analysis shows that a good entrainer is a component which “breaks” the azeotrope easily (i.e., even when its concentration is small) and yields high relative volatilities between the two azeotropic constituents. Because these attributes can be easily identified in an entrainer from the equivolatility curve diagram of the ternary mixture azeotropic compo- nent #1 zyxwvutsrqponm - azeotropic component zyxwvutsr #2 - entrainer, we can easily compare entrainers by examining the corresponding equivolatility curve diagrams. Finally, we demonstrate the validity and limits of this method with examples. Dans cet article, nous prCsentons des solutions pratiques (dans le cas de composants d’entrainement n’ajoutant pas d’azCotropes) zyxwvutsrqpo B deux probltmes d’importance industrielle: considCrant un azkotrope binaire voulant se separer en deux composants purs, et un ensemble d’Nentraineurs. possibles, comment dCterminer le meilleur? Par ailleurs, pour chacun de ces entraineurs, quel est le diagramme de procCdC de la sCquence ou des sequences de separation possibles? Nous obtenons ces solutions en analysant en dttail les m6canismes par lesquels des entraineurs lourds, intermdiaires et ICgers rendent la sCparation possible, en utilisant les nouvelles notions de courbes CquivolatilitC, des courbes d’isovolatilit6 et de I’ordre de volatilitC local. Nous montrons que la deuxikme question trouve une rCponse facile dans le diagramme de I’ordre de volatilid. Cette analyse montre qu’un bon entraineur est un composant qui rbrise* I’aztotrope facilement (c.a.d. mCme lorsque sa concentration est petite) et produit des volatilitCs relatives ClevCes entre les deux constituants azCotropes. Etant donnC que ces qualitks peuvent Ctre facilement reconnues dans un entraineur B partir du diagramme d’CquivolatilitC du mClange ternaire composant azCotropique #1 - composant azkotropique #2 - entraineur, nous pouvons facilement comparer les entraineurs en examinant les diagrammes d’CquivolatilitCcorrespondants. Enfin, nous dCmontrons la validiti et les limites de cette mtthode et donnons de nombreux exemples. Keywords: distillation, azeotropic distillation, entrainer selection, distillation design. eparating azeotropic mixtures into pure components is S a task commonly encountered in the chemical industry. If pressure-swing distillation cannot be used (because the azeotrope composition does not vary much with pressure or because the required pressure leads to product degradation), there are four basic methods to separate a binary azeotrope through distillation. These techniques have in common the addition of a third component, but the action of this entrainer depends on the considered type of distillation: Homogeneous azeotropic distillation: The entrainer alters the relative volatility of the two azeotropic constituents without inducing liquid-liquid separation. Heterogeneous azeotropic distillation: The entrainer alters the relative volatility of the two azeotropic constituents and induces liquid-liquid separation. Reactive distillation: The entrainer reacts reversibly and preferentially with one of the azeotropic constituents. “Salted” distillation: The entrainer dissociates ionically in the solution and changes the azeotropic composition. Heterogeneousazeotropic distillation is often preferred indus- trially because the decantation involved in the condenser makes the scheme attractive economically, but it suffers from a major drawback: Operating such columns can be very difficult, because upsets can induce phase separation inside the column, leading to severe efficiency losses (Kovack and Seider, 1987). Because homogeneous azeotropic distillation columns are much easier to operate (Jacobsen et al., 1990) *Current Affiliation: Procter & Gamble Canada. ?Current Affiliation: Technical University of Denmark, Lyngby . $Author to whom correspondence should be addressed. and because they can outperform heterogeneous azeotropic columns which separate the same azeotrope (Knapp and Doherty , 1990), homogeneous azeotropic distillation represents an economically attractive way of separating azeotropes. Given a binary azeotrope which we want to separate into two pure components, the design of homogeneous azeotropic sequence* performing the desired separation is usually car- ried out in two steps: First, we screen potential entrainers, then synthesize a separation sequence for each selected entrainer. The first step is critical, since an economically optimal design made with an average entrainer can be much more costly than an average design using the best entrainer. Over the years, several authors (Benedict and Rubin, 1945; Hoffman, 1964; Doherty and Caldarola, 1985; Stichlmair et al., 1989)have developed necessary conditions for separa- bility, i.e., conditions that a candidate entrainer must satisfy in order to make separation feasible. In a previous article (Laroche et al., 1990a), we have shown that these criteria contradict one another and cannot be used reliably in prac- tice. We have demonstrated that these criteria are incorrect because they do not take into account the unusual behavior of homogeneous azeotropic distillation (Laroche et al., 199Oa). More precisely, they fail to recognize that separations *A separation sequence contains usually two columns. The first column called extractive column, breaks the azeotrope and yields one azeotropic constituent as a pure product. The second column, zyx called entrainer recovery column, separates the other azeotmpic constituent from the entrainer, which is recycled to the extractive column. Note that, in some cases, separation may be done with only one column or may require three columns or more (Laroche et al., 1990b). I302 THE CANADIAN JOURNAL OF CHEMICAL ENGINEERING, VOLUME 69, DECEMBER, 1991