Basic Features Design and Operating Variables for Packed Bed Bioreactors

The basic design features of a packed-bed bioreactor have been already presented in Sect. 3.3.1. Figures 7.1 and 7.2 show these features in more detail. Some possible variations in the design include:

• the column may have a cross section other than circular.

• the column may lie horizontally, or for that matter, at any angle. This alters the relative directions of the forces due to gravity and air pressure.

• the column may be aerated from either end. For a vertical column, the air may enter the bed from either the top or the bottom. Aerating from the top avoids the fluidization of particles at high air velocities, but will contribute to bed compaction since the air flow is in the same direction as gravity.

• the column may have a perforated tube inserted along its central axis, allowing an extra air supply in addition to the end-to-end aeration (Fig. 7.1(b)). However, this will only be effective for very small bioreactor diameters.

• the column may be water-jacketed or heat transfer plates may be inserted into the bed. In this chapter, packed-bed bioreactors with internal heat transfer plates will be referred to as "Zymotis packed-beds", using the name coined by Roussos et al. (1993), while those lacking such plates will be referred to as "traditional packed-beds" (Fig. 7.2).

Fig. 7.1. Basic design features of packed-bed bioreactors and possible design variations. (a) A simple "traditional" packed-bed design. (b) A packed bed with a perforated tube inserted along its central axis: The benefits of this will only be apparent if the bed is relatively thin or, in a wide bed, if many perforated tubes are inserted. This is due to the fact that the forced aeration in the axial direction will tend to force the radial flow to follow the axial direction also. (c) Radial flow packed-bed: The advantage of this design is that, compared to a column of the same dimensions, the distance of flow through the bed is decreased. It is similar to the use of a wider "traditional" packed bed with a lower bed height. (d) A "short-wide" packed-bed

Plafractor
Fig. 7.2. Basic design and operating features of (a) traditional packed beds and (b) the Zy-motis packed-bed with internal heat transfer plates of Roussos et al. (1993)

Taking the most common design, namely a vertical column in which the bed is aerated from the bottom and without any internal perforated tubes, the available design variables for a packed-bed are (Fig. 7.2):

• the presence or absence of a cooling jacket or internal heat transfer plates;

• the height and width of the bioreactor. The height to diameter ratio can vary over quite a wide range;

• if internal cooling plates are used, their height and the spacing between them.

The available operating variables are (Fig. 7.2):

• the temperature of the inlet air;

• the temperature of the "surroundings" (which might be cooling water).

In a static bed, the relative humidity of the inlet air is not a useful operating variable. The problem is that it is not practical to add water into an unmixed bed in such a way as to distribute it evenly amongst the substrate particles; therefore evaporative water loss must be minimized. If the air entering the bed were not saturated with water, this unsaturated air would promote evaporation and dry out the bed, eventually decreasing the water activity to values unfavorable for growth and product formation. In order to minimize evaporation, saturated air must be supplied at the air inlet, which removes manipulation of the inlet air humidity as an available operating variable. Note that the use of saturated air does not prevent evaporation from occurring within the bed (see Fig. 4.3), but it does minimize evaporation compared to the use of unsaturated air. As will be discussed in Chap. 10, it is possible to replenish water during the mixing events of intermittently mixed beds, in which case unsaturated air can be used to aerate the bed.

At large scale, water-jacketing of the side walls of the bioreactor is not a good idea for the traditional design, since the water jacket will influence only the outer 20 cm or so of the bed. If cooling surfaces are to be used, then the internal cooling plates used in the Zymotis bioreactor will be more effective, as long as they are reasonably closely spaced. Optimum spacing of the plates will be discussed later. Another option for cooling surfaces is given by the "Prophyta" and "PlaFractor" designs (Fig. 7.3), two bioreactors that use a number of thin beds coupled with cooling plates oriented normal to the air flow (Luth and Eiben 1999; Suryanarayan and Mazumdar 2000; Suryanarayan 2003). The difference between the two biore-actors is that in the Prophyta design the same air passes through each successive bed while in the PlaFractor design the air is introduced separately into each bed.

Important phenomena that are affected by the values chosen for the design and operating variables are:

• the axial and radial temperature gradients in the bed. In packed-beds it is impossible to prevent temperature gradients from arising within the bed, so the aim is generally to minimize the size of any temperature gradients.

• the evaporation of water from the bed. Efforts must be made to minimize evaporation in order to prevent the bed or parts of the bed from drying out.

• the pressure drop through the bed. This will depend on the bed height and the degree to which the organism fills the inter-particle spaces, with the resulting pressure drop affecting the design of the aeration system and its operating costs, and maybe placing a limit on the bed height that can be used.

In general, O2 supply to the particle surface will not be considered in the selection of design and operating variables. The aeration rates that are chosen on the basis of heat removal considerations will typically be high enough that sufficient O2 supply is ensured. However, note that problems such as channeling are possible, in which O2 transport to large parts of the bed can be limiting (Fig. 7.4). Channeling is discussed in more detail in Sect. 7.3.3.5.

This chapter explains what is known, on the basis of experimental studies, about how these design and operating variables influence bioreactor operation. Later, Chap. 24 will show how mathematical models can be used to explore further the design and operation of packed-beds.

Was this article helpful?

0 0

Post a comment