Effects of initial molecular weight on removal rate of dextran in biofilms

Abstract

Degradation kinetics of different size dextrans in a biofilm reactor were evaluated. Degradation rates of dextran standards, measured as time series of oxygen utilisation rates, decreased with increasing initial molecular weight. Removal of bulk phase total organic carbon with time was highly correlated (R²>0.99) and could be modelled with variable half-order degradation rate expressions. A power correlation between initial molecular weight and the variable half-order degradation rate coefficient was found for polymers in the range 6–500 kDa. Degradation of dextran in the colloid size range (M_W>1 Mda) did not follow the same kinetics. Reductions in the observed removal rate with polymer size can be explained by the effect of reduced diffusivities of the substrate, without assuming reaction rate effects.

Introduction

More than 95% of the available organic matter in natural aquatic environments is composed of high molecular weight compounds (Münster and Chróst, 1990). In municipal wastewater a significant fraction of biological oxygen demand is found as suspended organic particles and dissolved polymeric material (Levine et al., 1985). Direct bacterial uptake of substrates from the surrounding media is limited to molecules of molecular weight (M_W) less than 600–1000 Da. Thus degradation of macromolecules (here defined by that limit) depends on extracellular enzymatic depolymerisation followed by uptake and mineralisation (Chróst, 1991; Confer and Logan, 1998a; White, 2000). In the wastewater engineering literature degradation of macromolecules and particulate organic matter is often lumped into the term hydrolysis. A quantitative understanding of hydrolysis kinetics is essential for the understanding of local electron donor utilisation rates, nutrient removal performance and microbial population dynamics as discussed in a review by Morgenroth et al (2002). Even though hydrolysis is regarded as a central process in wastewater systems, uncertainties still prevail regarding mechanisms stoichiometry and rate (kinetics) relations (Henze et al., 2000).

Early models of particulate substrate degradation were based on direct growth (Stenstrom, 1975) or adsorption followed by direct growth (Ekama and Marais, 1979; Dold et al., 1980; Frigon et al. 2001). Hydrolysis of slowly biodegradable into easily biodegradable substrates was adopted by the IAWPRC task group on mathematical modelling for design and operation of biological wastewater treatment processes, as a one-step hydrolysis process (Henze et al., 1987). Several authors have expanded this lumped substrate model to separately describe the kinetics of slow, intermediate, and rapidly hydrolysable substrates in order to reflect the chemical heterogeneity and molecular weights of particulate substrates. Sollfrank and Gujer (1991), Orhon et al. (1998), Janning (1998) and Vollertsen and Hvitved-Jacobsen (1999) defined parallel hydrolysis into easily biodegradable substrates, while Novak et al (1995), Bjerre (1996), Confer and Logan, 1997a, Confer and Logan, 1997b and Spérandio and Paul (2000) applied sequential hydrolysis. Separation into substrate classes based on degradability reflects the complexity of particle and polymeric substances degradation.

Models and parameters from studies of suspended systems (e.g. Henze and Mladenovski, 1991; Kappeler and Gujer, 1992; Eliosov and Argaman, 1995) may not be directly applicable in biofilm systems where there is less biomass–liquid interface area per biomass, causing less efficient mass transfer from the bulk liquid to the biomass in the biofilm. Hydrolysis of particulate and colloidal organic matter in biofilm systems has been studied by several groups (Sprouse and Rittmann, 1990; Larsen and Harremoës, 1994; Janning et al., 1998; Janning, 1998; Mosquera-Corral et al., 2003). Confer and Logan, 1997a, Confer and Logan, 1997b; 1998a) studied depolymerisation of model polymeric substrates in biofilm systems and suspended cultures. Using the polysaccharide dextran, they showed how intermediates may form, and discussed how intermediate formation and hydrolysis are affected by mass transfer limitations in biofilms and other bioaggregates with diffusion gradients.

Kinetics of biodegradation in biofilm systems have been mainly evaluated using dissolved substrates. The behaviour of readily diffusive low molecular weight compounds like acetate and glucose in biofilms has successfully been described by a set of diffusion-reaction equations (Harremoës, 1978). However, mass transfer in gel like structures like biofilms may be restricted by physical size. Even though large polymers, or particles, may not be able to penetrate into the biofilm matrix, these molecules are thought to adsorb to the biofilm surface (Bouwer, 1987; Guiot et al., 2002; Thurnheer et al., 2003). Following biofilm surface deposition, surface extracellular enzymes would act upon adsorbed polymers and particles, and release smaller intermediates, which would be able to penetrate into the biofilm for further depolymerisation and oxidation (Haldane and Logan, 1994; Confer and Logan, 1998b). Based on this conceptual model, the molecular weight and the geometry of a polymeric substrate will influence degradation kinetics due to possible sorption effects, diffusion restrictions and mechanisms of depolymerisation. A mathematical description of polymer degradation must take into account all of these potentially limiting processes.

In this paper, we investigate how the initial molecular weight of an added model substrate affects observed degradation rates in a mixed population batch operated biofilm reactor. Batch experiments were performed with different size dextrans and observed degradation rates were evaluated using an analytical solution of a diffusion-reaction model.