Laccase enzyme from Trametes versicolor was immobilized using calcium-alginate entrapment method to decolorize triphenylmethane dye. Selected variables namely initial dye concentration (ppm), agitation speed (rpm) and process time (day) were optimized using response surface methodology (RSM) Box-Behnken design protocol. Under optimal conditions, good percentage of crystal violet decolorization was observed. Experiments were scaled up to a one litre (1 L) stirred tank reactor (STR) to examine the effects of impeller speed and geometry. In both the shake flask and stirred vessel study, lower initial dye concentration showed greater removal of dye. The dye removal achieved in the stirred tank ranged from 14 percent to 36 percent. A 6-blade mixed flow impeller (6MFD) showed better removal than a radial 6-curved blade (6CB) impeller, at the same agitation speeds, albeit with higher power demand. The effect of impeller speed depended on level of dye concentration and impeller type; speed was important at low concentration for the 6MFD while the exact opposite was true for the 6CB.