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, G: Bacterial regrowth phenomena 185 bactena concentration could have an effect on initial biofilm formation rate. The transport of bactena to the substratum effects bacterial colonization and attachment (Wolfaardt, 1992; Mueller, 1996), which implies the role of bactena concentrations m the bulk Bois et al. (1997) modeled the effect of the increase of the inlet free active bactena concentration on the distribution system. Increasing the bactena concentration from 1 log bacteria/ml to 2 log bacteria/ml resulted in a proportional increase for free and attached bactena in the first loop (24 h residence time), steady-state was achieved after 50 days. In the second and third loop almost no effect was noticed. 5.4. Chlorine residuals To mitigate public health nsk, during its distribution drinking water should contain biocide agents in an adequate concentration. Chlorine is the most widely used disinfectant in drinking water. In distribution systems the possible applied chlorine residuals are the free chlorine and the monochloramine. Since the thorough chemical kinetics of chlorine decay is not known, chlorine decay models are generally based on pseudo first-order kinetics, assuming, that the concentration of the species reacting with the chlorine are much grater than that of chlorine (Chambers et al., 1995). In the distribution system chlorine can react with - suspended and attached biomass, - suspended organic and inorganic matter in the bulk, - organic and inorganic deposits, - pipe and system matenal. The relatíve contribution of the biofilm and pipe walls to the chlonne decay is generally unknown, it can be associated to the surface to volume (S/V) ratio (Parent et al., 1996; Vasconcelos, 1997). Large S/V ratios increase the influence of the biofilm and pipe walls. Although the generally applied chlorine residual concentrations in the distribution system do not eliminate biofilm activity (LeChevallier et al., 1988), they can effectively control HPC bacteria concentrations in the bulk (Neden et al., 1992; Block et al., 1995; Bailey and Thompson, 1995). 6. Modeling bacterial regrowth in the distribution system Due to the complexity of the bacterial regrowth phenomena in the distribution system, the process analysis requires modeling approach. In a bacterial regrowth model hydrodynamic modeling has to be coupled with biofilm kinetics, to describe the propagation and reaction of the water quality parameters (Dombay and öllős, 1998). A successful biofilm model has to take into account - the attachment of bacteria, - the proliferation due to BDOC consumption, - the biofilm detachment, and - the natural and chlorine induced mortality The CIRSEE research institute developed a biofilm model for dnnking water distribution systems, based on plug-flow reactor kinetics (Dukan et al., 1996). By the application of the model and a modified experimentál design, a study on quantifying the influence of environmental parameters and residence time on microbial drinking water quality in distribution systems was published elsewhere (Dombay et al., 1999). In order to understand the basic behavior of the biofilm reactor and the bacterial regrowth phenomena, this article shows the evolution of free active bactena in residence time, simultaneously with BDOC and attached biomass. The modeling was carried out under steady-state conditions, in the absence of chlorine residual. The water quality parameters taken mto account are - free totál bacteria, X; - free active bacteria, XH; - substrate (BDOC) concentration, S; - temperature, T. The initial values applied in the 4 model calculations are shown in Table 1 Table 1. Parameter values applied in model calculations No. 1-4. Parameter Model calculation, No. 1 2 3 4 S, mg c r' 0.25 1.00 0.25 1.00 T, °C 16 16 20 20 X, ml"

1 10

4 XH, ml"

1 10 1E»2 1E»1 . 1 . . • 1 . 1 1 1 rnidtnc* tim*, day Figure 2. Active attached bacteria rtfldanc* tkns, day Figure 3. Substrate (BDOC) concentration The evolution of fixed active bactena (Bib, the active fraction of the biofilm) in residence time is shown in Figure 2. High BDOC concentrations result in more fixed active bactena. The amount of Bib decreases in residence time (2-4), which is accounted to the reduction of BDOC concentration due to consumption, which can be seen in Figure 3. The calculation (1) results in a very limited biofilm activity. There is no BDOC consumption, hence this case represents the ideál biologically stable drinking wa-