Reaction selectivity and reactor choice

It can be seen that complex reactions often produce more than one product. In most industrial processes, one particular product (or group of products) is usually considered more desirable than the rest. Efforts will be made to choose reaction conditions and reactor types which favour the production of the desired material. Also, if more than one reactant is involved, attempts will be made to reduce the relative consumption of the most expensive reactant. In order to make quantitative comparisons between various courses of action, it is convenient to have some way of expressing relative product yields. This may be achieved by defining a reaction selectivity which refers to the comparitive formation rates of reaction products or by relating the appearance of a particular product to the consumption of a specified reactant. Various definitions have appeared in the literature the choice of terms is arbitrary. The use of terms in this chapter can be illustrated by an example. Consider the reactions 

This expression requires some qualification. It must be made clear that yield , for this example, means yield of C with respect to A. Also, it must be recognised that the concentration of a reaction species may change with time or with location within a reactor. Consequently, the relative yield may also change. The symbol (j will be used to denote instantaneous relative yield (for a very small element of space or time) and will be used to denote the overall yield for the whole reactor during its operational period. For the reactions (95) and (96) we have 

The subscripts on 0 and 1 indicate the two species to which 0 and refer. The second subscript, f, refers to a final concentration. The relationship between 0 and 4 depends on the reactor type which is being used. It can be seen that, when Vq Va it is possible for 0c,a to exceed unity. Stoichiometric coefficients may in incorporated in a definition of relative yield so that 0 always has a value between 0 and 1 (with rate being measured in mols m ). This is straightforward for relatively simple reactions but might lead to some confusion when a complicated reaction scheme is involved. 

For a batch reactor in which no density changes occur, we have, from eqns. (98) and (99) 

Equation (100) applies also to a continuous-flow reactor in which the contents experience no back-mixing (equiveilent to plug flow) time has been eliminated from the expressions. When complete back-mixing is achieved in a continuous-flow reactor, concentration gradients are absent and we have 

---