Air-lift fermentors have often been used instead of the traditional stirred-tank fermentor for production of penicillin(21), a-amylase(22), xanthan(23) and single cell protein(24). As the air-lift fermentor does not require mechanical agitation, the energy consumption is lower than that of a stirred-tank fermentor. The effect of the change in medium viscosity by adding carboxymethylcellulose (CMC) in an air-lift type fermenter on gas-hold up, bubble formation, flow pattern and mass transfer of oxygen has been reported by many researchers(25-27). We investigated the application of an air-lift type fermentor and the effect of change in medium viscosity by addition of CMC. Figure 5 shows the air-lift type fermentor used in this study, which was assembled as described by Okabe et al(28). To obtain high mass transfer achieved through the formation of small bubbles, a sintered stainless steel sparger (pore size, 10 mm; diameter, 12 mm; length, 20 mm) was installed at the bottom of the reactor. The feeding rate of the substrate gas mixture in the air-lift fermentor was 2 dm3 •min-1, which is equivalent to a superficial gas velocity of 2.62 cmes_1. Figure 6 shows the changes in medium viscosity and gas hold-up at various concentrations of CMC in the air-lift fermentor. Gas hold-up increased in proportion to the increase in CMC concentration up to 0.1% (w/v) but the gas hold-up decreased above 0.1% (w/v) of CMC. Deducing from this result, addition of CMC up to 0.1% (w/v) was expected to increase oxygen transfer rate with resultant increase in P(3HB) productivity. Figure 7 shows the time courses of autotrophic culture of A. eutrophus in the air-lift fermentor with addition of various concentrations of CMC. The productivity of P(3HB) in the culture with addition of 0.05 %(w/v) CMC (shown in Fig.7b) was increased to twice as that of the control culture with no addition of CMC(shown in Fig.7a). In the culture with addition of 0.1 % CMC, P(3HB) productivity was about 1.5 times higher than that of the control culture(Fig.7c). However, there was no apparent effect of the addition of CMC on the productivity of P(3HB) in the cultivation using the stirred-tank fermentor. A comparison was made for the effect of CMC addition on the mass transfer of oxygen in the air-lift and stirred-tank fermenters. When measurements were done by the static method, maximum KLa value for the air-lift fermenter was obtained at 0.05% of CMC concentration. The values of KLa for the air-lift fermentor measured by the sulfite oxidation method was observed to decrease with an increase in CMC concentration. For stirred-tank fermentor on the other hand, there was no increase in KLa values by addition of CMC into the culture medium. In the measurement by the sulfite oxidation method, the KLa of the stirred-tank fermenter was lowest at 0.1 % CMC. It is generally known that some kinds of surfactants, such as CMC, affect the sulfite-oxidation reaction. P(3HB) production rate observed in the fermentation experiments using the air-lift fermentor, correlated to the KLa measured by the static method but did not correlated to the KLa measured by the sulfite oxidation method. The KLa value measured by sulfite oxidation method was larger than those measured by the static method. These results mean that the static method is more reliable for the measurement of KLa than the sulfite oxidation method in autotrophic culture for P(3HB) production using air-lift fermentor. However, the relationship between the

The effect of CMC concentration on viscosity and gas-hold up of culture medium.

Cultivation time (h) Cultivation time (h) Cultivation time (h) Figure 7 Time course of autotrophic cultivation of A. eutrophus in air-lift type fermenter under various concentrations of CMC. (a): no addition of CMC, (b): 0.05 %(w/v) CMC, (c): 0.1 % (w/v) CMC.

KLa measured by the static method and the gas hold-up was not close, especially for the case where 0.1% CMC was used. This cannot be easily explained at present.