Choice of Equipment
Up to now it was assumed that reaction and distillation can favorably be combined in a column: in a normal distillation column in the case of homogeneous catalysis and in a column with special internals or an additional exterior volume in the case of heterogeneous catalysis. This was discussed in the previous chapter under the aspect of scale-up in connection with separation and reaction performance. However columns are an appropriate solution only for reactions that are so fast as to achieve considerable conversions in the residence time range of such columns. The question is whether the full potential for combining reaction and distillation can be found and industrially implemented using columns only.
At BASF, there has been some research on this point in the last couple of years, and the results will be included in the following chapters . For reasons of simplicity the focus will be on equilibrium reactions where the advantages of combining reactions and distillation are obvious. The aspects influencing the choice of the equipment will become particularly clear.
2 Reactive Distillation Process Development in the Chemical Process Industries | 39 In simple equilibrium reactions, the reaction equation can be described like this h
The rate constants for the forward and the reverse reaction may be different. The equilibrium state (when the reaction velocity goes to zero) is described by the law of mass action
The conversion of the stoichiometrically limiting reactant, for example component 1, in the equilibrium state is
Vo c1 c2 Kc
The indices 0 indicate the initial state, the indices * the final state (i. e., the equilibrium state). This last equation was developed using some additional simple balance equations.
The equation shows that, the most efficient way of enhancing the conversion is to reduce the concentration of one of the reaction products. That is the principle of superimposing a distillation on a reaction: the theme of this contribution. In the equation the volume effect of a superimposed distillation has been taken into account.
Fig. 2.11 (equilibrium lines) is a graphical representation of the equation above that shows more clearly the possibilities of such a combination: The fractional conversion of component 1 approaches 1 only if the concentration of one reaction product is significantly reduced. This holds true even for extremely unfavorable equilibrium constants. If, for example, the equilibrium constant is 0.01, in the equilibrium state only 1 % of component 1 has been converted. However if component 3 is removed to x = 0.0001 mole/mole, conversions of more than 90% are possible for component 1.
The real conversion may be defined in analogy to an equilibrium conversion but without the equilibrium values:
The reaction velocity of reactant 1 can, in most cases, be described by the simple equation:
T=feiCiC2 - fcminlC3C4
Fig. 2.11 Equilibrium lines and kinetic line
10 5 10" 10 J 10 2 10-1 Mole fraction of component 3, x3! [mol/mol]
Assuming stoichiometric input, with the volume correction term taking into account the distillation and with some simple balance equations, the equations above can be rearranged in the following way:
k1C1C2 - feminlC3C4
Hence the real conversion is increased if the reverse reaction is suppressed by removing one of the reaction products, for example by decreasing the concentration of component 3. The real conversion is also influenced by the residence time r or by the product rk . This can be seen in Fig. 2.11, kinetic line. It is obvious that the suppression of the reverse reaction influences conversion only up to a certain limit. A further drop in the concentration of the reaction product does not increase the conversion. The contribution of the reverse reaction has become so low that the conversion depends on the forward reaction rate only, which is a function of the residence time r and of the reaction constant k. Thus two operating conditions can be distinguished.
• The range in which the conversion is influenced mainly by the concentration of the component to be separated; this range is called 'controlled by distillation'.
• The range in which the conversion is influenced mainly by the residence time and the reaction constant; this range is called 'controlled by kinetics'.
Industrial process design should aim at operating conditions within these two ranges: just the sufficient residence time and only the necessary expenditure for the distillation.
So we have a second scale-up problem: what is the suitable equipment for complying with these demands? That means more precisely: How can the reactor performance be reached over a broad range of reaction velocities ? As preparation for discussion on this question, an alternative configuration is considered (Fig. 2.12): A reaction can be run within a column, that normally is understood as RD, but can also be run in an outside reactor with a pump recycle. Such a sequential arrangement exhibits the same conversion as the simultaneity of reaction and separation as can be seen from Fig. 2.13, where a reactor with pump recycle system and a RD configuration are compared.
So different equipment may be chosen to combine reaction and distillation within the limiting conditions of reaction velocity, relative volatility, and catalytic mechanism as is indicated in Fig. 2.14.
The equipment in question includes:
- stirred vessels,
- cascades of stirred vessels, both with or without columns, and
- reaction columns.
Additional volume can be provided for all of them, examples are listed in Fig. 2.15.
The next considerations concentrate on homogeneous catalysis. Similar considerations apply to heterogeneous catalysis, and this will be commented on later.
At first, a slow reaction is considered. 'Slow' means, that the reaction time is slow, compared with the residence times typical for separation equipment such as distillation trays. For residence time reasons a stirred vessel or, better, a cascade of stirred vessels is needed. Each vessel is supplied with energy to evaporate the component to be separated. If the relative volatility of this component is very high, a one stage evaporation is sufficient. At a lower relative volatility the separation requires more stages, so a column has to be put on top of the vessel. If the separation is even more difficult, a stripping section must be added to the column and a reboiler is necessary. In the limiting case of a very low relative volatility, each stage of the cascade can be operated as a countercurrent stage. The first stage is additionally provided with a fractionating column to enrich the component to be separated. Such a setup is equivalent to a reaction column with a large holdup on each reaction stage.
Fast reactions do not demand long residence times, 'fast' meaning that the reaction reaches equilibrium in the residence time range that is typical for column internals. The equipment therefore may be selected under the aspect of separation efficiency . If the relative volatility of the component to be removed is low, a considerable number of stages is necessary. The only appropriate equipment is a column. Depending on the required residence time it may be a packed or a tray column. A relative volatility in the medium range allows the number of stages to be reduced, though the total hold up has to be kept constant. A tray column, perhaps with special bubble cap trays is possible. At a certain (low) relative volatility the equilibrium reactor with pumparound
reactive distillation with one reactive stage
Fig. 2.13 Comparison of reactor with pumparound with RD
30 40 50 60 70 80 Pumparound / acid-feed stream, [mcl/mol]
Fig. 2.13 Comparison of reactor with pumparound with RD
short residence time focus on separation efficiency little separation efficiency focus on residence time
Was this article helpful?