Use of response surface methodology to optimise the hydrolysis of whey protein isolate in a tangential flow filter membrane reactor

Response surface methodology (RSM) was used to investigate the effects of substrate concentration (S), enzyme concentration (E) and initial permeate flow rate, (J(i)), on permeate flux behaviour in a 10 kDa nominal molecular weight cut-off (NMWCO) tangential flow filter (TFF) enzyme membrane reactor (EMR) during 3 h hydrolysis of whey protein isolate (WPI) using Protease N Amano (IUB 3.4.24.28, Bacillus subtilis) at pH 7.0 and 45 degrees C. The average residual permeate flow rate (J(residual)), residual enzyme activity (A(residual)) and product recovered in permeate designated as apparent sieving (S-apparent) were monitored. The quadratic model regression equations obtained revealed that all the three factors had significant but dissimilar influences on permeate flux behaviour. Jresidual, Sapparent and A(residual) increased with increasing E, A(residual) decreased with increasing J(i) and there was substrate inhibition at low E. The optimised factors were S = 4.72% (w/v), E = 0.055% (w/v; hence E/S approximate to 1% w/w) and J(i) = 6.91 mL/min (approximately 0.7% reactor volume per minute). The optimised values were 87.24%, 52.37% and 35.08% for J(residual) A(residual) and S-apparent respectively. The actual values for the responses agreed well with the predicted values implying that RSM is suitable for EMR optimisation. Covariance values showed that J(residual) and S-apparent increased concomitantly while A(residual) decreased with increasing S-apparent and J(residual). (c) 2006 Elsevier Ltd. All rights reserved.