In order to reduce the cost of enzymatic production of biodiesel, the lipase producing whole cells rather than the isolated enzyme has been used. This eliminates the need for isolation and purification steps before immobilization, which results in a considerable reduction in the cost. Air drying immobilization technique of lipase producing Rhizopus oryzae whole cells was developed by Matsumoto et al. (2001). The use of immobilized whole cells to produce biodiesel by three stepwise addition of methanol in solvent-free system was reported to achieve biodiesel yield of 71% after 165 h.

Hama et al. (2004) found that the fatty acid composition affects the activity of the whole cells by influencing their membranes. It was reported that pretreatment of the whole cells with oleic acid and linoleic acid resulted in higher enzymatic activity, whereas palmitic acid pretreated cells showed higher stability. To compensate for both activity and stability, an optimum ratio of unsaturated to total fatty acids of 0.67 was proposed. Using the pretreated whole cells, methanolysis yields were consistently above 55% even after ten repeated cycles. To explain the fatty acids composition effect, Hama et al. (2006) suggested the existence of two types of lipases: one bound to the cell wall, which plays role in stability, and the other to the cell membrane, which plays role in methanolysis activity. The increase in enzyme activity with addition of unsaturated fatty acids was expected to be due to the increase in the production of membrane-bound lipase.

The immobilized lipase producing whole cells from R. oryzae were prepared in cuboidal polyurethane foam biomass support particles in a 20 L air-lift batch cultivation bioreactor and used in a packed-bed reactor for continuous production of biodiesel by methanolysis of soybean oils (Hama et al., 2007). Compared with methanolysis reaction in a shaken bottle, the packed-bed reactor enhanced repeated batch methanolysis by protecting immobilized cells from physical damage and excess amounts of methanol. The flow rate of reaction mixture had to be optimized, as low flow rates resulted in a significant decrease in activity due to the covering of the immobilized whole cells with a hydrophilic layer of high methanol concentration, and high flow rates resulted in cells leaching. A highest biodiesel yield of 90% was achieved at a flow rate of 25 L h-1. The yield dropped to around 80% after the tenth cycle. To overcome the leaching problem, cross­linking treatment with 0.1% glutaraldehyhde has been proposed (Ban et al., 2002). By glutaraldehyde treatment, biodiesel yield of 83% was maintained after six batch cycles in the stepwise methanol addition process, compared to only 50% without glutaraldehyde treatment.

Recently, Tamalampudi et al. (2008) used the same immobilized whole cells prepared by Hama et al. (2007) for the production of biodiesel from relatively low
cost, inedible oil from the seeds of Jatropha curcas in a 50 ml screw-capped vessels with reciprocating shaking at 150 rpm. The activity of immobilized whole cell was compared with that of Novozym 435 and was found to be more efficient. The maximum biodiesel was 80% after 60 h using the former catalyst, whereas using the latter the maximum yield was only 76% after 90 h.