Hidrológiai Közlöny 1999 (79. évfolyam)

3. szám - Dombay Gábor: Bacterial regrowth phenomena in the drinking water distribution system. A bakteriális vízminőségromlás jelensége az ivóvízelosztó hálózatban

IXJMBAY, (j. Bacterial regrowth phenomerui 183 4. Transport processes and biofilm kinetics Biofilms are influenced by hydraulic conditions, and in tum, biofilms may mfluence the hydrodynamic conditions close to the surface of and within biofilms (Wilderer et al., 1995). In tubular reactors increasing water velocity leads to - an increase of the convective transport of bactena, substrate, chlorine and other constituents in the reactor, - an increase in turbulent diffusion, - a decrease of the thickness of the boundary layer causing a reduction in the mass transfer resistance towards the biofilm, - an increase of shear stress at the pipe wall. Transport processes in the bulk are controlled by convection and turbulent diffusion, in the biofilm by molecular diffusion. Under turbulent conditions in the bulk virtually there is no concentration gradient in radial direction. The concentration gradient evolves in the concentration boundary layer (CBL) and reaches íts maximum at the biofilm surface (Lewandowski et al., 1994). In the CBL diffusion is the predominant mass transfer mechanism. The velocity profilé in the hydrodynamic boundary layer (HBL), where the flow becomes unidirectional, is uncertain. Bishop et al. (1997) showed, that there is a significant difference between the hydrodynamic boundary layer and the concentration boundary layer above a biofilm. CBL vaned between 200-400 |im depending on the water velocity in the bulk. HBL and flow velocity had only minimál relationship, the thickness is around 4000-5500 ^m. Increasing fluid velocity decreases the thickness of the CBL, causing a decrease in the mass transfer resistance (Characklis and Marshall, 1990). Christiansen et al. (1995) showed, that the influence of liquid film diffusion on reaction rate in submerged biofilter can be ignored. In drinking water distribution systems the biofilm thickness is low (around 10-30 ^m, L'Hostis, 1996) and likely there is no transport limitation in the biofilm. Consequently ít might be assumed, that vanation in substrate transport due to different flow velocities does not have a significant effect on biofilm behavior. According to Characklis and Marshall (1990), hydrodynamic shear forces might play a major role in the enhanced biofilm removal when chlorine is being added. Based on RotoTorque experiments, a higher rate of chlorine uptake and a greater amount of detachment was observed when the rotational velocity increased Although the applied chlorine dose was 5 mg/l, hence in drinking water with a very low chlorine concentration this experience might not apply. Donlan et al. (1994), based on in situ measurements, expenenced flow-related effects on biofilm formation (HPC on R2A) in chloraminated drinking water network. In the pipe with the highest flow velocity a biofilm with Iower HPC value was förmed, which the authors hypothesize as a result of thinner CBL hence more rapid diffusion of monochloramine into the biofilm. At this site suspended HPC was the lowest, which they think could be an other reason for lower biofilm activity DeBeer et al (1994) experienced, that chlorine microprofiles showed the presence of a significant CBL of 100400 pm. The chlorine concentration at the biofilm-bulk water interface was typically 20-30% of the bulk concentration, hence external mass transfer resistance was significant under the experimentál conditions. As CBL thickness decreases by increasing fluid velocity, the efficacy of chlorination could be strongly improved by increasing fluid velocity. Experiments performed in an annular reactor, with fluid velocities of ca. 1 m/s showed much higher chlorine efficacy. 5. The influence of water quality parameters on bacterial dynamics in the network 5.1. Substrate For its growth, heterotrophic bacteria require carbon, nitrogén, and phosphorus in a ratio of approximately C:N:P=100:10:L Consequently in most cases the growthliiting nutnent in drinking water is the organic carbon (LeChevallier, 1990). Organic carbon is present in drinking water in the form of vanous compounds, mostly complex macrooeues, generally in the order of a few mg carbon/1 concentration (Anderson et al., 1997). Due to the diversity and its weak concentration, only about 1/3 of the natural organic matter (NOM) is identified in drinking water (Mathieu, 1992). The amount of organic matter can be expressed in totál organic carbon (TOC). The dissolved fraction of NOM (dissolved organic matter, DOM) in drinking water is generally characterized by a molecule size less than 0.45 (im (Buffle et al., 1982). DOM can be expressed in dissolved organic carbon (DOC). In drinking water for its metabolism HPC bacteria can utilize a fraction of DOM (LeChevallier et al., 1991). This fraction is called biodegradable organic matter (BOM). BOM is composed by different organic compounds, of which biodegradability might be different. The amount of BOM can be expressed by the biodegradable fraction of DOC, called the biodegradable dissolved organic carbon (BDOC). For the determination of BDOC several techniques are available (Joret and Levi, 1986; Servais et al., 1987; Prévost et al., 1992; Frias et al., 1995; Maclean et al., 1996; Dubreuil et al., 1997). Vanances in BDOC measurement can be m the rangé of 5070% (Block et al., 1992). The amount of biodegradable organic matter can alsó be expressed by an indirect parameter, called the assimilable organic carbon (AOC, van der Kooij et al., 1982; van der Kooij, 1992). AOC is not an absolute measure of organic carbon concentration, but an indirect measure of biogrowth potential (or of the biologically available energy of BOM) of the particular drinking water. AOC values may be considerably lower than BDOC values (Huck, 1990), consequently ít is argued that AOC can be regarded as a surrogate for the overall BOM (Zhang and Huck, 1996). Based on the assumption that the concentration of one single growth limiting nutrient determines the growth rate of bacterial proliferation, the model of Monod (1942) is