with regard to the activity of the biocatalyst and the chemical stability of the substrate
and product molecules. Where the aqueous phase is not the continuous phase, “hot” spots
of extreme pH may arise which can only be made homogeneous by continual coalescence
and breakage with other droplets i.e. by inter-droplet mixing. An alternative approach,
and the most common, is to operate with the aqueous phase as the continuous phase
where the phase volume ratio is typically less than 0.3–0.4. The phase volume ratio also
affects both the absolute and specific interfacial areas as will be discussed later.
Table 5.3 Organic solvent and phase volume ratio
selection criteria
Factors to consider Phase volume ratio effects
Solubility limit of S and P in organic phase Ease of pH control
S and P distribution coefficients Nature of dispersed phase
Biocompatibility (e.g. Log P values) Possibility of phase inversion
Aqueous solubility limit of solvent Ease of phase separation
Emulsion formation Interfacial area available for mass transfer
Toxicity and flammability Inter-droplet mixing
Solvent recycling options
Environmental impact
Cost and availability
Table 5.3 summarises the factors to be considered when screening a range of solvents and
indicates effects which will be a function of the phase volume ratio used. A more detailed
discussion of the phase volume ratio affects can be found in a review by Lilly and
coworkers (1990).
TWO-LIQUID PHASE REACTORS
Types of ReactorThe two key requirements in selecting an appropriate reactor are (1) to establish
sufficient interfacial area between the aqueous and organic phases to enable adequate
substrate or product mass transfer, and (2) to be able to readily control the interfacial
area. To date these requirements have been achieved in three types of reactor; stirred
tanks, liquid-impelled loops and membrane reactors.
The stirred-tank reactor is by far the most popular since it enables excellent control of
the interfacial area via agitator speed. It is also the design of reactor most readily
available in both research laboratories and industrial facilities. For such reactors it has
been shown that the rate of a particular biotransformation will be a function of the reactor
geometry (Collins and Woodley, 1993) and the power input per unit volume (Woodley,
1990a). Based on the principles of the gas fluidised bed and the air-lift reactor an
Advances in the selection and design of two-liquid phase biocatalytic reactors 129