Fundamentals of Reactive Distillation

It is a well-known and accepted fact that complicated interactions between chemical reaction and separation make difEcult the design and control of RD colunms. These interactions originate primarily from VLL equilibria, VL mass transfer, intra-catalyst diffusion and chemical kinetics. Moreover, they are considered to have a large influence on the design parameters of the unit e.g. size and location of (non)-reactive sections, reflux ratio, feed location and throughput) and to lead to multiple steady states (Chen et al, 2002 Jacobs and Krishna, 1993 Giittinger and Morari, 19996,a), complex dynamics (Baur et al., 2000 Taylor and Krishna, 2000) and reactive azeotropy (Doherty and Malone, 2001 Malone and Doherty, 2000). 

To provide a clear insight on the physical and chemical phenomena that take place within a RD unit, a systems approach is proposed according to the following classification features, 

These four features can be smartly combined leading to 10 different system instances, as hsted in table 2.1. Instances 2, 5 and 8 are not further considered in this explanatory chapter because of their reduced practical interest. As there is no interaction between 

This analysis results in instances 1 (lumped one-stage, closed and equilibrium), 3 (lumped one-stage, open and non-equilibrium), 6 (lumped multistage, open and nonequilibrium) and 10 (distributed, open and non-equilibrium) for further consideration in the course of this chapter. A first classification of these instances is performed based on their spatial structure, aiming to cover the whole range of physical and chemical phenomena that occur in a RD unit. The following levels are then defined. 

Level A. One-stage level where the system is represented by a single lumped stage [embracing instances Ii AND I3], 

---

In this dissertation we addressed the fundamentals of reactive distillation process design and attempt to generate scientific novelty. These contributions to the knowledge domain are grouped in the following categories. 

As mentioned in the Preface, a small number of industrial applications of reactive distillation have been around for many decades. One of the earhest was a DuPont process in which dimethyl terephthalate was reacted with ethylene glycol in a distillation column to produce methanol and ethylene terephthalate. The reactants were fed into the middle of the column where the reversible reaction occurred. The more volatile, low-boiling methanol product was removed from the top of the column, and the high-boiling ethylene terephthalate product was removed from the bottom. The removal of the products from the reaction zone drove the reversible reaction toward the product side. This is one of the fundamental advantages of reactive distillation. Low chemical equilibrium constants can be overcome and high conversions achieved by the removal of products from the location where the reaction is occurring. 

The applicability of reactive distillation is highly dependent on the chemical system at hand, and there are stiU limited applications of this process. A fundamental difference between... 

The variation of efficiencies is due to interaction phenomena caused by the simultaneous diffusional transport of several components. From a fundamental point of view one should therefore take these interaction phenomena explicitly into account in the description of the elementary processes (i.e. mass and heat transfer with chemical reaction). In literature this approach has been used within the non-equilibrium stage model (Sivasubramanian and Boston, 1990). Sawistowski (1983) and Sawistowski and Pilavakis (1979) have developed a model describing reactive distillation in a packed column. Their model incorporates a simple representation of the prevailing mass and heat transfer processes supplemented with a rate equation for chemical reaction, allowing chemical enhancement of mass transfer. They assumed elementary reaction kinetics, equal binary diffusion coefficients and equal molar latent heat of evaporation for each component. 

Knowledge of the equilibrium is a fundamental prerequisite for the design of non-reactive as well as reactive distillation processes. However, the equilibrium in reactive distillation systems is more complex since the chemical equilibrium is superimposed on the vapor-liquid equilibrium. Surprisingly, the combination of reaction and distillation might lead to the formation of reactive azeotropes. This phenomenon has been described theoretically [2] and experimentally [3] and adds new considerations to feasibility analysis in RD [4]. Such reactive azeotropes cause the same difficulties and limitations in reactive distillation as azeotropes do in conventional distillation. On the basis of thermodynamic methods it is well known that feasibility should be assessed at the limit of established physical and chemical equilibrium. Unfortunately, we mostly deal with systems in the kinetic regime caused by finite reaction rates, mass transfer limitations and/or slow side-reactions. This might lead to different column structures depending on the severity of the kinetic limitations [5], However, feasibility studies should identify new column sequences, for example fully reactive columns, non-reactive columns, and/or hybrid columns, that deserve more detailed evaluation. 

Qiemical and Petrochemical Industries. Distillation is one of the fundamental unit operations of chemical engineering and is an integral part of many chemical manufacturing processes. Modern industrial chemistry in the twentieth century was based on the numerous products obtainable from petrochemicals, especially when thermal and catalytic cracking is applied. Industrial distillations are performed in lai e, vertical distillation towers that are a common sight at chemical and petrochemical plants and petroleum refineries. These range from about 2 to 36 feet in diameter and 20 to 200 feet or more in height Chemical reaction and separation can be combined in a process called reactive distillation, where the removal of a volatile product is used to shift the equilibrium toward completion. 

A fundamental difference between the two flowsheets is the ability in the conventional process to adjust reactor temperature and distillation column temperamres completely independently, which is not possible in the reactive distillation process. In the conventional system, the reactor temperature can be set at an optimum value and distillation temperatures can be independently set at their optimum values by adjusting column pressures. In reactive distillation, these temperatures are not independent because only one pressure can be set in the vessel. Therefore, the design of a reactive distillation requires a tradeoff between temperatures conducive for reaction (kinetics and equilibrium constants) and temperatures favorable for vapor-liquid separation. The temperature dependency of the relative volatilities will illustrate this important difference between the two processes. 

These results demonstrate a fundamental difference between a two-product reactive column (quaternary system) and a one-product reactive column (ternary system). This is a good example of the complexity and challenges associated with reactive distillation. 

Before delving into the feed rearranging control structure, we first construct the fundamental control configuration for the reactive distillation with two feeds. Recall that, unlike the control of conventional distillation systems, we need to control the internal composition (or temperature) to maintain stoichiometric amounts of the two fresh feeds. For the purpose of illustration in this work, we choose to control the composition of reactant A on tray 13 where a large change in the composition of A is observed (Fig. 18.5b). Thus, we have three compositions to be controlled top composition of C, bottoms composition of D, and composition A on tray 13. For the manipulated variables, the ratio scheme is used these three ratios are reflux ratio, boilup ratio, and feed ratio. Figure 18.12 shows the control structure. 

The first objective has been accomplished by the development of an HPLC procedure as reported by Spalik et al. ( 5) and GC/NPD procedures developed by Lemley and Zhong ( ). The second and third objectives are being accomplished by fundamental solution studies and reactive ion exchange experiments conducted in parallel. Lemley and Zhong (7) determined recently the solution kinetics data for base hydrolysis of aldicarb and its oxidative metabolites at ppm concentrations and for acid hydrolysis of aldicarb sulfone. They have since ( ) reported similar results for ppb solutions of aldicarb and its metabolites. In addition, the effect on base hydrolysis of temperature and chlorination was studied and the effect of using actual well water as compared to distilled water was determined. Similar base hydrolysis data for carbofuran, methomyl and oxamyl will be presented in this work. 

Impurities are a concern in ionic liquids electrochemistry. Whereas even considerable amounts of impurities, like different metal ions, water or organic impurities, might not disturb a technical process (e.g. extractive distillation, organic synthesis) the wide electrochemical windows of an ionic liquid ( 3 V vs. NHE) allow the electrodeposition of even reactive metals like lithium and potassium, as well as the oxidation of halides to the respective gases. In the best case this codeposition only leads to a low level of impurities, in the worst case fundamental physicochemical studies are made impossible as the impurities are adsorbed onto the electrode surface and subsequently reduced. Furthermore, passivation or activation effects at the counter electrode have to be expected. 

Earlier chapters use simplified and binary models to analyze in a very informative manner some fundamentals such as the effect of reflux ratio and feed tray location, and to delineate the differences between absorption/stripping and distillation. Following chapters concentrate on specific areas such as complex distillation, with detailed analyses of various features such as pumparounds and side-strippers, and when they should be used. Also discussed are azeotropic, extractive, and three-phase distillation operations, multi-component liquid-liquid and supercritical extraction, and reactive multistage separation. The applications are clearly explained with many practical examples. 

---