Optimization and modelling of enzymatic polymerization of ε-caprolactone to polycaprolactone using Candida Antartica Lipase B with response surface methodology and artificial neural network

Recently enzymatic catalysts have replaced organic and organometallic catalysts in the synthesis of bio-resorbable polymers. Enzymatic polymerization is considered as an alternative to conventional polymerization as they are less toxic, environmental friendly and can operate under mild conditions. In this research, the enzymatic ring-opening polymerization (e-ROP) of e-caprolactone (e-CL) using Candida Antartica Lipase B (CALB) as catalyst to produce the Polycaprolactone. Two modelling techniques namely response surface methodology (RSM) and artificial neural network (ANN) have been used in this work. RSM is used to optimize the parameters and to develop a model of the process. ANN is used to develop the model to predict the results obtained from the experiment. The parameters involved are time, reaction temperature, mixing speed and enzyme-solvent ratio. The experimental result is Polydispersity index (PDI) of the polymer. The experimental data obtained was adequately fitted into second-order polynomial models. Simulation was done using artificial neural network model developed with Mean absolute error (MAD) value of 1.65 in comparison with MAD value of 7.4 for RSM. The Regression value (R2) values of RSM and ANN were found to be 0.96 and 0.93 respectively. The predictive models were validated experimentally and were found to be in agreement with the experimental values.