Batch Distillation: Simulation and Experimental Validation

Batch industries, specially the fine chemical industry, need rapid and robust mathematical models that can easily predict the degree of separation of mixtures like volatile organic compounds. Otherwise, in the study of on line production scheduling and planning of fine chemical plants, there is a lack of quick but reliable and more accurate than black box models in batch distillation. In this work, a robust and rapid mathematical model has been developed, based on mass balance and relative volatility. The model has been extensively tested in a variety of scenarios using a batch distillation pilot plant that has been constructed adhoc. The satisfactory results obtained allows to use the model in real time for a better control and management of batch distillation, resulting in an improved production scheduling/rescheduling of batch processes. INTRODUCTION Environmental constraints force industries to have rapid and contrasted models in batch distillation for solving the recovery of volatile organic compounds. The European Community has developed a directive in order to restrict the industrial emissions of this kind of solvents [1]. The fine chemical industry works with a high variety of chemical products with a high added value, that are changing continuously following market fluctuations and have a small lifetime. This implies that batch distillation must be a flexible unit operation and must be able to separate the changing mixtures efficiently. This is an important challenge in batch distillation. Rapid but reliable mathematical models contribute to this objective. Luyben [2] and Al-Tuwain and Luyben [3] studied a short-cut method based on mass balances and mathematical correlations to design batch distillation columns. Diwekar and Madhavan [4] also presented a rapid method to design batch distillation columns, based on the equations of Fenske-Underwood-Gilliland (FUG). Barolo and Botteon [5] worked with a column at infinite reflux ratio in order to obtain theoretically pure components in a binary mixture. Galindez and Fredenslund [6] studied a model based on the program Unidist applied to continuous columns. Their results are compared with the experimental data of Domenech and Enjalbert [7]. Stewart et al.[8] and Hitch and Rousseau [9] presented a parametric study based on mass and energy balances that was experimentally validated. Diwekar et al.[10], Diwekar [11] and Noda et al.[12] have studied optimization problems in batch distillation. Their work conclude that optimal reflux ratio is comparable with constant reflux ratio for batch distillation columns. In this work a mathematical model based on mass balances and the equilibrium vapor-liquid equations has been developed. The industry needs rapid models like this one presented in this work. The vapor-liquid equilibrium will be studied with constant and variable relative volatility. This is a new aspect presented in this work for the first time. All the work in the past has been based on the use of constant relative volatility. In batch distillation only a few research works are experimentally validated. Most of them uses the experimental works of Domenech and Enjalbert [7] and Nad and Spiegel [13]. There is a lack of experimental studies in batch distillation. In this work the model has been extensively validated experimentally. Firstly, it has been applied to the methanol-water mixture, that has served for the start-up and fine tuning of the pilot plant. Then, the ternary mixture cyclohexane-toluene-chlorobenzene has been studied with the model and experimentally validated. This mixture hasn’t been referenced in the bibliography, and only experimental data of a ternary mixture have been published for Nad and Spiegel [13]. Azeotropic mixtures present a barrier in the recovery of solvents in batch chemical plants. The rapid model developed in this work has also been applied to the study of binary azeotropic mixtures. The interest of this part of the work is that only a few short-cut methods for azeotropic mixtures are published, and also there is a lack in experimental data. Toluene-n-butanol, with a minimum boiling point is the mixture studied. The last part of the work emphasizes in the resolution of this binary azeotropic mixture by adding a solvent, n-octanol. All the work presented has an important objective which is the application to the scheduling of batch chemical plants, specially fine chemical plants. In production scheduling, batch distillation is treated like a black box model, where the actual distillation time is not contemplated. The proposed model permits to accurate modeling at the scheduling level and offers an improved on line operation facilitating eventual rescheduling. MATHEMATICAL MODEL The mathematical model presented in this work is based on mass balances and the equilibrium vapor-liquid equations. It can be applied to multicomponent mixtures and also binary azeotropic mixtures. Figure 1 represents a schematic representation of the batch distillation column used. Figure 1. Schematic representation of a batch distillation column for the simplified mathematical model. The model is based on the assumptions of constant flowrate for the liquid and the vapor, constant holdup for the liquid, negligible vapor holdup, total condenser, adiabatic column and theoretical trays. The mathematical model can be described by the following equations: Overall mass balance around the column D dt dM B − = (1) N n Dj x D,