The production of L-DOPA immobilized on chemically modified nylon 6,6 membranes was studied in a batch reactor. Tyrosinase was immobilized on nylon using glutaraldehyde as a crosslinking agent. The effects of membrane pore size and glutaraldehyde concentration upon enzyme uptake and L-DOPA production were investigated. Enzyme uptake was unaffected by glutaraldehyde concentration; approximately 70% uptake was observed when 25% w/v (group 1), 5% (group 2), and 3% (group 3) glutaraldehyde were used, indicating that glutaraldehyde was in excess. Similarly, uptake was the same for membranes with 0.20 and 10 mu m pore sizes. Membranes produced using different levels of glutaraldehyde exhibited dramatically different capacities for L-DOPA production, despite the fact that enzyme uptake was equivalent. Membranes from groups 2 and 3 (5% and 3% glutaraldehyde) produced L-DOPA at a rate of 1.70 mg L(-1) h(-1) over 170 h in a 500-mL batch reactor. However, no free L-DOPA was detected when group 1 membranes were used. Experimental evidence suggests that L-DOPA was produced, but remained bound to these membranes via excess glutaraldehyde left over from the immobilization process. Membrane pore size also effected L-DOPA production; less production was observed when 10-mu m membranes were used, despite equivalent enzyme uptake. The observed difference in production may be due to differences in the pore density on the two types of membranes which could affect the access of the substrate to the immobilized enzyme. The results of these studies indicate that tyrosinase can be effectively immobilized on nylon 6,6. L-DOPA production was optimal when 0.20-mu m-pore-size membranes were activated with 3-5% glutaraldehyde. Stability studies indicated a 20% reduction in activity over 14 days when the immobilized enzyme was used under turnover conditions.