Simulation of Cr(VI) Removal in a Biofilm Reactor System and Optimisation Using an Ecological Algorithm

Abstract

A one-dimensional diffusion-reaction model was developed to simulate the removal of the carcinogenic transitional metal, Cr(VI), in a biofilm reactor with glucose and phenolic compounds as sole supplied carbon and energy sources in different runs. Substrate utilization and Cr(VI) reduction in the biofilm was best represented by a system of (second-order) partial differential equations (PDEs). Organic acid metabolic intermediates were measured and included in the dynamic model. The PDE system was solved by the (fourth-order) Runge-Kutta method adjusted for mass transport resistance using the (second-order) Crank-Nicholson and Backward Euler finite difference methods. A heuristic procedure, genetic search algorithm (GSA), was used to find global optimum values of Cr(VI) reduction and substrate utilization rate kinetic parameters. The fixed-film bioreactor system yielded higher values of the maximum specific Cr(VI) reduction rate coefficient and Cr(VI) reduction capacity (kmc = 0.062 1/h, and Rc = 0.13 mg/mg, respectively) than previously determined in batch reactors (kmc = 0.022 1/h and Rc = 0.012 mg/mg). The model predicted effluent Cr(VI) concentration with 98.9% confidence ((y2 = 2.37 mg2/L2, N = 119) and effluent glucose and phenol concentration with 96.4% and 99.3% confidence ((y(w)2 = 5402 mg2/L2, N = 121, w = 100) over a wide range of Cr(VI) loadings (10–500 mg Cr(VI)/L/d).