Statistical modeling and optimization of Escherichia coli growth parameters for the biological treatment of phenol

Abstract

Aromatic compounds, including phenols, are a significant source of pollution which need to be treated by environmentally-friendly methods, such as bioprocesses. This work focuses on the biodegradation of phenol in a batch reactor with bacteria, and the optimization of the growth parameters in order to obtain the highest phenol degradation. The model and algorithms fitting the growth data are emphasized. Primary models, applied to monitor the dynamic evolution of the microbial biomass of the selected strain, were fitted to the data by nonlinear regression based on the Levenberg Marquart algorithm. The statistically-validated Baranyi and Roberts equation was used to evaluate the growth parameters: maximum growth rate ( μ max ), latency time ( λ ), and maximum optical density ( OD max ). To improve bacterial growth and phenol degradation performance, physico-chemical conditions, such as initial phenol concentration, pH, and nitrogen source (ammonium sulfate), were optimized using secondary models based on a central composite rotatable design (CCRD). The correlation coefficient, R 2 , for each regression equation is >94%. The optimal values of growth parameters are λ min = 21.08 h, μ max = 8.68 h −1 , and OD max = 0.39 at pH = 6.3 for an initial concentration of phenol = 200 mg/L and initial concentration of ammonium sulfate = 1.33 g/L. Escherichia coli showed an ability to degrade up to 963 mg/L of phenol in 250 h without prior acclimatization of the strain. • Biodegradation process of phenol in a batch reactor using Escherichia coli. • Optimization of the growth parameters thanks to Baranyi and Roberts'equations. • Optimization of experimental conditions using secondary models based on a central composite rotatable design. • Degradation up to 963 mg/L of phenol in 250 h without acclimatization of Escherichia Coli.