Estimation of Heterotrophic Biokinetic Parameters in Wastewater Biofilms from Oxygen Concentration Profiles by Microelectrode

Abstract

Knowledge of the biokinetic parameters of heterotrophic biofilm is important for reliable modeling and optimization of biofilm wastewater treatment systems. In this article, an approach to quantify heterotrophic biokinetic parameters, including the maximum specific oxygen uptake rate ( \documentclass{aastex}\usepackage{amsbsy}\usepackage{amsfonts}\usepackage{amssymb}\usepackage{bm}\usepackage{mathrsfs}\usepackage{pifont}\usepackage{stmaryrd}\usepackage{textcomp}\usepackage{portland, xspace}\usepackage{amsmath, amsxtra}\pagestyle{empty}\DeclareMathSizes{10}{9}{7}{6}\begin{document} $$\stackrel{\smallfrown}{q}_{\max}$$ \end{document} ), the Monod half saturation coefficient of oxygen (k_o), and the maintenance decay coefficient for oxygen (K_m) in wastewater biofilms, was presented and evaluated. With the oxygen microprofiles measured in the biofilm by using the oxygen microelectrode, the heterotrophic biokinetic parameters were determined through fitting the data to the diffusion-reaction models. The biomass used in the model was represented in terms of the average volatile suspended solids (VSS) measured by weight and the spatial distribution of VSS by phospholipid analysis. Results showed that with an addition of an external substrate at a concentration of 2000 mg COD/L into biofilms, oxygen became the growth limiting factor. The biokinetic parameters derived from the average VSS possessed the smaller \documentclass{aastex}\usepackage{amsbsy}\usepackage{amsfonts}\usepackage{amssymb}\usepackage{bm}\usepackage{mathrsfs}\usepackage{pifont}\usepackage{stmaryrd}\usepackage{textcomp}\usepackage{portland, xspace}\usepackage{amsmath, amsxtra}\pagestyle{empty}\DeclareMathSizes{10}{9}{7}{6}\begin{document} $$\stackrel{\smallfrown}{q}_{\max}$$ \end{document} compared with the layered VSS. However, the heterogeneity of biomass has a slight impact on the k_o and K_m. Sensitivity analyse showed that the proposed model was a good \documentclass{aastex}\usepackage{amsbsy}\usepackage{amsfonts}\usepackage{amssymb}\usepackage{bm}\usepackage{mathrsfs}\usepackage{pifont}\usepackage{stmaryrd}\usepackage{textcomp}\usepackage{portland, xspace}\usepackage{amsmath, amsxtra}\pagestyle{empty}\DeclareMathSizes{10}{9}{7}{6}\begin{document} $$\stackrel{\smallfrown}{q}_{\max}$$ \end{document} estimator.

Introduction

Modeling plays an important role in the design and control of wastewater biotreatment process. The constants used in these models are termed as “biokinetic parameters.” It is extremely important to have accurate and reliable estimation of these biokinetic parameters, as they are often used in the design of field-scale biotreatment systems. The biokinetic parameters of microorganisms in nature of suspended growth (Ekama et al., 1986; Miloslav et al., 1993; Kim et al., 2001) was commonly qualified based on the interpretation of the oxygen uptake rate (OUR) profiles, which were measured through the batch experiments with additives of a readily biodegradable substrate at different concentrations to the endogenous activated sludge sample. Later, Ellis et al. (1996) and Goudar and Ellis (2001) modified the method by feeding a continuous feed substrate instead of intermittent substrate injection, and the solution to the nonlinear equation in models was also simplified. However, the biokinetic parameters in the attached growth systems are particularly difficult to be measured, because diffusional resistance within the biofilm would possibly mask the true reaction kinetics (Riefler and Smets, 2003). Methods of extracting the biokinetic parameters in the attached biofilm systems and their disadvantages were summarized in our previous study (Zhou et al., 2008a).

Microelectrodes as a powerful tool have been developed and used over the last several decades to investigate the microbial processes in the biofilm (Gillam et al., 2005). Yurt et al. (2003) developed a method of extracting growth parameters based on the in situ measurement of oxygen microprofiles in the colony biofilm of L. discophora; however, only the colony biofilm composed of single microbial species has been inspected. Further, the biomass density used in the model is usually assumed to be constant represented by the mean volatile suspended solids (VSS), despite the fact that the density varies tremendously across the biofilm depth (Zhang and Bishop, 1994; Beyenal and Lewandowski, 2000). Phospholipids, present on bacterial membrane up to 90%–98%, do not form part of cell reserves and are easily degraded during bacteria lysis, which make it reasonable to use the phospholipids concentration to represent the living biomass in biofilms (Findlay et al., 1989).

In the study, an approach is established and evaluated to quantify the growth biokinetic parameters of biofilm in wastewater treatment by means of oxygen microelectrodes. The layered VSS was obtained by using the spatial distribution of living biomass determined by the phospholipids analysis and micro-slicing technique. Results of biokinetic parameters derived from the average and layered VSS were compared to investigate the impact of biomass heterogeneity on the biokinetic parameters of biofilm.

Materials and Methods

Biofilm systems

The support material for biofilm was a piece of flat polymerized ethylene resin with an area of 30 mm×30 mm. One side of the resin was rubbed coarse to make the bacterial inoculation easier. The introduced activated sludge was taken from a bench-scale oxidation ditch. After 1 day of inoculation, the introduced activated sludge was completely withdrawn. Subsequently, eight pieces of support material were immersed in a 2.5 L reactor, and the biofilm was statically cultured with domestic wastewater from the drainage of Tsinghua University Campus (Table 1) at a room temperature of 18°C±2°C. Air from the gas blower was introduced to the reactor to maintain an oxygen concentration of about 5–6 mg/L. The domestic wastewater in the reactor was shifted every 12 h. After 2–3 months of operation, the biofilm was ready for experiment.

Table 1.

Water Quality of Domestic Wastewater

Items	Concentration [mg/L ^a ]
COD	300–400
NH₃-N	25–40
TN	30–45
SS	7–82
pH	6.5–8.0

Except pH.

Microelectrode measurements

The schematic set up for the in situ microelectrode measurement of oxygen microprofiles in the biofilm was shown in Fig. 1. One piece of media was taken from the reactor for microelectrode measurement, from which the biofilm grown on one side of the substratum was wiped off. The nonbiofilm side was fixed onto a measuring vessel. A metal-alloy-filled solid-state microelectrode with gold electroplated on the electrode tip was fabricated with a tip diameter of about 10 μm as reported by Yu (2000). The 90% response time of the microelectrodes was in most cases <15 s with an insignificant stirring sensitivity (<5%). The manipulating of microelectrode as well as how to minimize the ambient interference can be found in our previous work (Zhou et al., 2008b, 2008c).

FIG. 1.

Experimental set up for measurement of oxygen in biofilm by microelectrode.

Phospholipids analysis

The spatial distribution of layer phospholipids concentration in the biofilm was acquired by using microslicing technique as similar as reported by Taylor et al. (1997). A biofilm sample with an area of ∼0.5 cm×0.5 cm cut from the support media with a blade was attached to a cutting stage with the freezing compound (Tissue-Tek O.C.T. 4583; Miles, Inc.). Biofilm was sliced parallel to its surface by using a (LEICA CM1900) cryostatic microtome with 40 μm thickness. Four or five adjacent slices were accumulated and then transferred to carry out the phospholipids analysis. The detailed procedures of phospholipids analysis are described by Zhang and Bishop (1994).

Quantification of the biokinetic parameters in the biofilm

The basic assumptions will be made in the following kinetic analysis:

1. The rates of bacterial growth and substrate removal can be represented by the Monod kinetics, and the rate of reaction is not affected by the presence of reaction products;

2. Oxygen is the growth limiting factor;

3. Oxygen in the biofilm is transported by diffusion in the form of Fick's first law.

Based on the assumptions just mentioned, mass balance for dissolved oxygen through biofilm in one dimension is expressed as follows by taking into consideration the maintenance process. \documentclass{aastex}\usepackage{amsbsy}\usepackage{amsfonts}\usepackage{amssymb}\usepackage{bm}\usepackage{mathrsfs}\usepackage{pifont}\usepackage{stmaryrd}\usepackage{textcomp}\usepackage{portland, xspace}\usepackage{amsmath, amsxtra}\pagestyle{empty}\DeclareMathSizes{10}{9}{7}{6}\begin{document} \begin{align*} D \frac {d^2 S_ {\rm o}} {dx^2} = \stackrel{\smallfrown}{q}_{\max} \cdot \frac {S_ {\rm o}} {k_ {\rm o} + S_ {\rm o}} \cdot X - K_ {\rm m} \cdot X \tag {1} \end{align*} \end{document}

where S_o is the oxygen concentration, mg O₂/L; D is the effective diffusion coefficient of oxygen in biofilms, assumed to be 1.43×10⁻⁵ cm²/s (25°C), 32% lower than the value in pure water as reported by Beyenal and Tanyolac (1994a); \documentclass{aastex}\usepackage{amsbsy}\usepackage{amsfonts}\usepackage{amssymb}\usepackage{bm}\usepackage{mathrsfs}\usepackage{pifont}\usepackage{stmaryrd}\usepackage{textcomp}\usepackage{portland, xspace}\usepackage{amsmath, amsxtra}\pagestyle{empty}\DeclareMathSizes{10}{9}{7}{6}\begin{document} $$\stackrel{\smallfrown}{q}_{\max}$$ \end{document} is the maximum specific OUR, mg O₂/(g VSS·h); k_o is the Monod half saturation coefficient of oxygen, mg O₂/L; X is the biofilm density presented by the average VSS (g VSS/L) or the spatial phospholipids concentrations (g VSS/L) as described by Zhang and Bishop (1994); and K_m is the maintenance decay coefficient for oxygen, mg O₂/(g VSS·h). The boundary conditions for the solution of Eq. (1) are as follows: \documentclass{aastex}\usepackage{amsbsy}\usepackage{amsfonts}\usepackage{amssymb}\usepackage{bm}\usepackage{mathrsfs}\usepackage{pifont}\usepackage{stmaryrd}\usepackage{textcomp}\usepackage{portland, xspace}\usepackage{amsmath, amsxtra}\pagestyle{empty}\DeclareMathSizes{10}{9}{7}{6}\begin{document} \begin{align*} x = 0, \quad S_{\rm o} = S_{{\rm ob}} \tag{2} \end{align*} \end{document} \documentclass{aastex}\usepackage{amsbsy}\usepackage{amsfonts}\usepackage{amssymb}\usepackage{bm}\usepackage{mathrsfs}\usepackage{pifont}\usepackage{stmaryrd}\usepackage{textcomp}\usepackage{portland, xspace}\usepackage{amsmath, amsxtra}\pagestyle{empty}\DeclareMathSizes{10}{9}{7}{6}\begin{document} \begin{align*} x = x_{\rm L}, \quad {\rm d}S_{\rm o} / {\rm d}x = 0 \tag{3} \end{align*} \end{document}

where S_ob is the oxygen concentration at the interface of bulk solution/biofilm, mg O₂/L; x_L is the depth of biofilm, mm. A numerical solution algorithm was applied to compute the biokinetic parameters in Eq. (1). All computations were performed by MATLAB program as follows:

1. Reading the experimental data (S_o with respect to x);

2. Assuming \documentclass{aastex}\usepackage{amsbsy}\usepackage{amsfonts}\usepackage{amssymb}\usepackage{bm}\usepackage{mathrsfs}\usepackage{pifont}\usepackage{stmaryrd}\usepackage{textcomp}\usepackage{portland, xspace}\usepackage{amsmath, amsxtra}\pagestyle{empty}\DeclareMathSizes{10}{9}{7}{6}\begin{document} $$\stackrel{\smallfrown}{q}_{\max}$$ \end{document} , k_o and K_m. According to the reported literatures (Jørgensen et al., 1992; Horn and Hempel, 1997; Chu et al., 2003), let \documentclass{aastex}\usepackage{amsbsy}\usepackage{amsfonts}\usepackage{amssymb}\usepackage{bm}\usepackage{mathrsfs}\usepackage{pifont}\usepackage{stmaryrd}\usepackage{textcomp}\usepackage{portland, xspace}\usepackage{amsmath, amsxtra}\pagestyle{empty}\DeclareMathSizes{10}{9}{7}{6}\begin{document} $$\stackrel{\smallfrown}{q}_{\max}$$ \end{document} be from 0 to 100 mg O₂/(g VSS·h), k_o be from 0 to 1.1 mg O₂/L, and K_m be from 0 to 1 mg O₂/(g VSS·h).

3. Using minimization searching procedure to find the optimized biokinetic parameters.

4. Calculating the sum of squares of the difference (SSD) between the pairs of computed and measured values as

\documentclass{aastex}\usepackage{amsbsy}\usepackage{amsfonts}\usepackage{amssymb}\usepackage{bm}\usepackage{mathrsfs}\usepackage{pifont}\usepackage{stmaryrd}\usepackage{textcomp}\usepackage{portland, xspace}\usepackage{amsmath, amsxtra}\pagestyle{empty}\DeclareMathSizes{10}{9}{7}{6}\begin{document} \begin{align*} {\rm SSD} = \sum_{{\rm i} = 1}^{\rm n} \big( S_{{\rm o_i}, {\rm computed}} - S_{{\rm o_i}, {\rm measured}} \big) ^2 \tag{4} \end{align*} \end{document}

Minimization search procedure continues until it finds \documentclass{aastex}\usepackage{amsbsy}\usepackage{amsfonts}\usepackage{amssymb}\usepackage{bm}\usepackage{mathrsfs}\usepackage{pifont}\usepackage{stmaryrd}\usepackage{textcomp}\usepackage{portland, xspace}\usepackage{amsmath, amsxtra}\pagestyle{empty}\DeclareMathSizes{10}{9}{7}{6}\begin{document} $$\stackrel{\smallfrown}{q}_{\max}$$ \end{document} , k_o, and K_m, which minimize Eq. (4).

O₂ profiles in the biofilm

The trustworthiness of the biokinetic parameters depends on how closely the experimental system and the experimental procedures adhere to the assumptions of the model we used. Oxygen is the growth limiting substrate that was assumed in this study. To make sure that this assumption applied, the following experiment was conducted. First, biofilm in the measuring vessel was immersed in the tap water without adding substrate for over 24 h. Then, the external substrate at concentrations of 500, 1000, and 2000 mg COD/L (in form of glucose), respectively, was supplied into the vessel to replace the pure water. After 2 h of cultivation, O₂ profiles in the biofilm were sequentially measured at each concentration to determine the condition for which microbial growth in the biofilms was limited by oxygen. After that, O₂ profiles were re-measured at the different locations in the biofilm. Note that the bulk solution over the biofilm maintained stable growth without influent and effluent during the whole experiment.

All measurements were conducted at a room temperature of 18°C±2°C.

Results and Discussion

Biofilm

After 2 months' cultivation, biofilms with thickness of 1.2–1.5 mm was observed growing on the supporting media. The spatial distribution of living biomass in forms of phospholipids concentrations was measured as depicted in Fig. 2, as well as the average biofilm density in forms of VSS was measured to be 27.1 g VSS/L (n=10, S.D.=3.4 g VSS/L). The phospholipids values were in the range of 803–1892 mg P/L with a maximum value occurring at the depth of 0.64 mm, indicating the highest concentration of the living cell in the biofilm. Previous studies have shown that the phospholipids values have a linear relationship with biomass in a unit of TSS or VSS (Findlay et al., 1989; Zhang and Bishop, 1994); therefore, the layered VSS were obtained according to the spatial distribution of phospholipids and used in the following model of biokinetic parameters.

FIG. 2.

Spatial distribution of phospholipids concentrations measured in the biofilm.

Oxygen microprofiles in the biofilm

Figure 3 demonstrated the measured oxygen microprofiles in the biofilms grown at the feed substrate concentrations of 500, 1000, and 2000 mg COD/L. It was clear that the oxygen microprofiles changed little when the substrate concentrations exceeded 1000 mg COD/L; therefore, to make sure that oxygen is the limiting factor in microbial growth in our measurements, we used 2000 mg COD/L as the feed concentration.

FIG. 3.

Measured oxygen microprofiles at different feed substrate concentrations in the biofilm (▪ 500 mg COD/L, ● 1000 mg COD/L, ▴ 2000 mg COD/L).

Even though lower substrate concentrations gave increased oxygen concentration in the biofilm, the oxygen was completely depleted before it reached the substratum. Oxygen could not be transported >950 μm into the biofilm under any of the three substrate concentrations, which is consistent with the results that the mass transport greatly influences the rate of biotransformation in attached biofilm systems in contrast to the dispersed bacterial growth (Lu, 2001; Boessmann et al., 2004; Seo and Bishop, 2008). The oxygen penetration depths in the biofilm were in the range of 750–950 μm. Compared with the results reported earlier [(usually in the range of 240–600 μm measured by Seo and Bishop (2008), Boessmann et al. (2004), and Lu (2001)], the oxygen penetrations are really deeper in the study. It indicated the slow bacterial activities or the little active biomass in the biofilm, which would result in the decrease in the oxygen consumption. In addition, the loose structure caused by static cultivation or other growth conditions would increase the diffusivity of oxygen in the biofilm, which is another factor attributed to the deep oxygen penetration.

To confirm uniformity of the biofilm, four oxygen microprofiles were measured at different locations in the biofilm at the concentration of 2000 mg/L, as shown in Fig. 4. The measuring points were located at the intersection points of square's sides with a distance of about 500 μm (see the insert in Fig. 4). Compared with the results measured in colony biofilms reported in the literature by Yurt et al. (2003), the change in oxygen concentrations in water immersed biofilm was more irregular due to its more heterogeneous structure, especially the oxygen concentrations measured above the biofilm. This phenomenon was partly due to the function of the suspended bacteria released in the bulk solution from the biofilm. Figure 4 also demonstrated that the oxygen distributions in the biofilm at different locations were similar although they were not indeed identical, which was thought to accord with the assumption of the uniform distribution of microorganisms in the biofilm.

FIG. 4.

Measured oxygen microprofiles at different locations in the biofilm. The length of square's side was ∼500 μm.

Sensitivity analysis

A sensitivity analysis of biokinetic parameters in the model was conducted to test whether the oxygen profiles were sufficiently sensitive to the biokinetic parameters. It was conducted by varying one model parameter while holding the remaining parameters constant as shown by Riefler et al. (1998). Sensitivity was judged by the effects that these perturbations had on the resulting dissolved oxygen profiles. Figure 5 showed the predicted oxygen profiles for different values of the biokinetic parameters including \documentclass{aastex}\usepackage{amsbsy}\usepackage{amsfonts}\usepackage{amssymb}\usepackage{bm}\usepackage{mathrsfs}\usepackage{pifont}\usepackage{stmaryrd}\usepackage{textcomp}\usepackage{portland, xspace}\usepackage{amsmath, amsxtra}\pagestyle{empty}\DeclareMathSizes{10}{9}{7}{6}\begin{document} $$\stackrel{\smallfrown}{q}_{\max}$$ \end{document} , k_o, and K_m. It clearly illustrated the expected strong sensitivity of the model to the parameter \documentclass{aastex}\usepackage{amsbsy}\usepackage{amsfonts}\usepackage{amssymb}\usepackage{bm}\usepackage{mathrsfs}\usepackage{pifont}\usepackage{stmaryrd}\usepackage{textcomp}\usepackage{portland, xspace}\usepackage{amsmath, amsxtra}\pagestyle{empty}\DeclareMathSizes{10}{9}{7}{6}\begin{document} $$\stackrel{\smallfrown}{q}_{\max}$$ \end{document} . The changes in the parameter \documentclass{aastex}\usepackage{amsbsy}\usepackage{amsfonts}\usepackage{amssymb}\usepackage{bm}\usepackage{mathrsfs}\usepackage{pifont}\usepackage{stmaryrd}\usepackage{textcomp}\usepackage{portland, xspace}\usepackage{amsmath, amsxtra}\pagestyle{empty}\DeclareMathSizes{10}{9}{7}{6}\begin{document} $$\stackrel{\smallfrown}{q}_{\max}$$ \end{document} caused shifts in the dissolved oxygen profile that exceeded shifts caused by the changes in K_m and k_o. This indicated that the model is a better \documentclass{aastex}\usepackage{amsbsy}\usepackage{amsfonts}\usepackage{amssymb}\usepackage{bm}\usepackage{mathrsfs}\usepackage{pifont}\usepackage{stmaryrd}\usepackage{textcomp}\usepackage{portland, xspace}\usepackage{amsmath, amsxtra}\pagestyle{empty}\DeclareMathSizes{10}{9}{7}{6}\begin{document} $$\stackrel{\smallfrown}{q}_{\max}$$ \end{document} estimate than a K_m and k_o estimate. In other words, the K_m and k_o estimation was not more important than \documentclass{aastex}\usepackage{amsbsy}\usepackage{amsfonts}\usepackage{amssymb}\usepackage{bm}\usepackage{mathrsfs}\usepackage{pifont}\usepackage{stmaryrd}\usepackage{textcomp}\usepackage{portland, xspace}\usepackage{amsmath, amsxtra}\pagestyle{empty}\DeclareMathSizes{10}{9}{7}{6}\begin{document} $$\stackrel{\smallfrown}{q}_{\max}$$ \end{document} in the model.

FIG. 5.

Sensitivity of dissolved oxygen profile to the biokinetic parameters of (a) \documentclass{aastex}\usepackage{amsbsy}\usepackage{amsfonts}\usepackage{amssymb}\usepackage{bm}\usepackage{mathrsfs}\usepackage{pifont}\usepackage{stmaryrd}\usepackage{textcomp}\usepackage{portland, xspace}\usepackage{amsmath, amsxtra}\pagestyle{empty}\DeclareMathSizes{10}{9}{7}{6}\begin{document} $$\stackrel{\smallfrown}{q}_{\max}$$ \end{document} , (b) K_m, and (c) k_o.

Growth biokinetic parameters

Table 2 demonstrated the computed biokinetic parameters and SSD from the four measured oxygen microprofiles in Fig. 4, by using the average VSS and layered VSS obtained from the phospholipids analysis. As shown in Table 2, the biokinetic parameters derived from the average VSS possessed the smaller \documentclass{aastex}\usepackage{amsbsy}\usepackage{amsfonts}\usepackage{amssymb}\usepackage{bm}\usepackage{mathrsfs}\usepackage{pifont}\usepackage{stmaryrd}\usepackage{textcomp}\usepackage{portland, xspace}\usepackage{amsmath, amsxtra}\pagestyle{empty}\DeclareMathSizes{10}{9}{7}{6}\begin{document} $$\stackrel{\smallfrown}{q}_{\max}$$ \end{document} compared with the layered VSS. However, the heterogeneity of biomass has a slight impact on the k_o and K_m. The computed \documentclass{aastex}\usepackage{amsbsy}\usepackage{amsfonts}\usepackage{amssymb}\usepackage{bm}\usepackage{mathrsfs}\usepackage{pifont}\usepackage{stmaryrd}\usepackage{textcomp}\usepackage{portland, xspace}\usepackage{amsmath, amsxtra}\pagestyle{empty}\DeclareMathSizes{10}{9}{7}{6}\begin{document} $$\stackrel{\smallfrown}{q}_{\max}$$ \end{document} was in the range from 9±0.7 to 11±0.7 mg O₂/(g VSS·h) (mean value±S.D.), which was smaller than the value measured in suspended growth microorganisms (Temmink et al., 1993; Liu et al., 2006). To the best of our knowledge, a few studies have reported \documentclass{aastex}\usepackage{amsbsy}\usepackage{amsfonts}\usepackage{amssymb}\usepackage{bm}\usepackage{mathrsfs}\usepackage{pifont}\usepackage{stmaryrd}\usepackage{textcomp}\usepackage{portland, xspace}\usepackage{amsmath, amsxtra}\pagestyle{empty}\DeclareMathSizes{10}{9}{7}{6}\begin{document} $$\stackrel{\smallfrown}{q}_{\max}$$ \end{document} in the attached growth systems. However, \documentclass{aastex}\usepackage{amsbsy}\usepackage{amsfonts}\usepackage{amssymb}\usepackage{bm}\usepackage{mathrsfs}\usepackage{pifont}\usepackage{stmaryrd}\usepackage{textcomp}\usepackage{portland, xspace}\usepackage{amsmath, amsxtra}\pagestyle{empty}\DeclareMathSizes{10}{9}{7}{6}\begin{document} $$\stackrel{\smallfrown}{q}_{\max}$$ \end{document} can be obtained based on the stoichiometric relationship between the microbial growth rate, substrate removal, and O₂ consumption rate in a bio-chemical reaction. In this way, we found that the \documentclass{aastex}\usepackage{amsbsy}\usepackage{amsfonts}\usepackage{amssymb}\usepackage{bm}\usepackage{mathrsfs}\usepackage{pifont}\usepackage{stmaryrd}\usepackage{textcomp}\usepackage{portland, xspace}\usepackage{amsmath, amsxtra}\pagestyle{empty}\DeclareMathSizes{10}{9}{7}{6}\begin{document} $$\stackrel{\smallfrown}{q}_{\max}$$ \end{document} measured in this method was close to but still smaller in most cases than that in other biofilm systems reported by Plattes et al. (2007) (converting to be 18.3 mg O₂/(g SS·h) heterotrophic biomass in a pilot scale moving bed bioreactor) and Rittmann et al. (1986) (converting to be 1.75 mg O₂/(g SS·h) with regard to sodium salicylate, 17.5 mg O₂/(g SS·h) with regard to sodium acetate, and 19.8 mg O₂/(g SS·h) with regard to phenol if the yield coefficient Y_H (g biomass/g O₂) be 0.72). The low \documentclass{aastex}\usepackage{amsbsy}\usepackage{amsfonts}\usepackage{amssymb}\usepackage{bm}\usepackage{mathrsfs}\usepackage{pifont}\usepackage{stmaryrd}\usepackage{textcomp}\usepackage{portland, xspace}\usepackage{amsmath, amsxtra}\pagestyle{empty}\DeclareMathSizes{10}{9}{7}{6}\begin{document} $$\stackrel{\smallfrown}{q}_{\max}$$ \end{document} was accompanied with a low maintenance decay rate of ∼0.95 g O₂/(g SS·h), which was even one order of magnitude smaller than the results in the literatures reported by Horn and Hempel (1997), Beyenal and Tanyolac (1994b), and Tsuneda et al. (2002).

Table 2.

Computation of the Biokinetic Parameters and Sum of Squares of the Difference in Biofilm Derived from the Average and Layered Volatile Suspended Solids

	\documentclass{aastex}\usepackage{amsbsy}\usepackage{amsfonts}\usepackage{amssymb}\usepackage{bm}\usepackage{mathrsfs}\usepackage{pifont}\usepackage{stmaryrd}\usepackage{textcomp}\usepackage{portland, xspace}\usepackage{amsmath, amsxtra}\pagestyle{empty}\DeclareMathSizes{10}{9}{7}{6}\begin{document} $$\stackrel{\smallfrown}{q}_{\max} \ mg \ O_2 / (g \ VSS \cdot h)$$ \end{document}		k_o mg O₂/L		K_m mg O₂/(g VSS·h)		SSD
Measurement location	Average	Layered	Average	Layered	Average	Layered	Average	Layered
1#	9	11	0.3	0.4	1	0.9	0.1531	0.1116
2#	8	10	0.4	0.5	1	1	0.0460	0.0748
3#	9	11	0.4	0.4	1	1	0.1152	0.0813
4#	10	12	0.7	0.7	0.8	1	0.1739	0.2336
Mean value	9	11	0.45	0.5	0.95	0.975	0.1221	0.1253

VSS, volatile suspended solids; SSD, sum of squares of the difference.

The half saturation coefficient k_o had a close relationship with the mass transport limitation in attached growth systems (Kappeler and Gujer, 1992; Chen et al., 2003; Pérez et al., 2005). The computed k_o was 0.45–0.5 mg O₂/L on an average (n=4). It was larger than the values measured for microorganisms grown in suspension (0.25 mg O₂/L, Kappeler and Gujer, 1992; 0.20 mg O₂/L recommended by IWA Task Group, 2000), which was in well agreement with the conclusion reported by Pérez et al. (2005) that the effect of mass transport limitations is reflected by half-velocity coefficients in the Monod expression.

Conclusions

From the work presented, the following conclusions can be obtained:

1. With the help of microelectrodes, the biokinetic parameters in the heterotrophic biofilms for wastewater treatment were thoroughly investigated. The biokinetic parameters derived from the average VSS possessed the smaller \documentclass{aastex}\usepackage{amsbsy}\usepackage{amsfonts}\usepackage{amssymb}\usepackage{bm}\usepackage{mathrsfs}\usepackage{pifont}\usepackage{stmaryrd}\usepackage{textcomp}\usepackage{portland, xspace}\usepackage{amsmath, amsxtra}\pagestyle{empty}\DeclareMathSizes{10}{9}{7}{6}\begin{document} $$\stackrel{\smallfrown}{q}_{\max}$$ \end{document} compared with the layered VSS. However, the heterogeneity of biomass has a slight impact on the k_o and K_m.

2. A diffusion-reaction model was established to simulate the oxygen consumption and distribution in the biofilms. With the sensitivity analysis, the \documentclass{aastex}\usepackage{amsbsy}\usepackage{amsfonts}\usepackage{amssymb}\usepackage{bm}\usepackage{mathrsfs}\usepackage{pifont}\usepackage{stmaryrd}\usepackage{textcomp}\usepackage{portland, xspace}\usepackage{amsmath, amsxtra}\pagestyle{empty}\DeclareMathSizes{10}{9}{7}{6}\begin{document} $$\stackrel{\smallfrown}{q}_{\max}$$ \end{document} were simulated to be more sensitive than the k_o and K_m in the Monod kinetic model, thus indicating that the model is a good \documentclass{aastex}\usepackage{amsbsy}\usepackage{amsfonts}\usepackage{amssymb}\usepackage{bm}\usepackage{mathrsfs}\usepackage{pifont}\usepackage{stmaryrd}\usepackage{textcomp}\usepackage{portland, xspace}\usepackage{amsmath, amsxtra}\pagestyle{empty}\DeclareMathSizes{10}{9}{7}{6}\begin{document} $$\stackrel{\smallfrown}{q}_{\max}$$ \end{document} estimator.

Footnotes

Acknowledgments

This research work was supported by the China Postdoctoral Science Foundation funded project (20080430046, 200801093) and Tsinghua University Initiative Scientific Research Program (20101081834).

Author Disclosure Statement

No competing financial interests exist.

References

Beyenal

, Lewandowski

2000. Combined effects of substrate concentration and Mow velocity on effective diffusivity in biofilms. Water Res., 34:528.

Beyenal

, Tanyolac

1994a. The calculation of simultaneous effective diffusion coefficients of the substrates in a fluidized bed biofilm reactor. Water Sci. Technol., 29:463.

Beyenal

, Tanyolac

1994b. A mathematical model for hollow fiber biofilm reactors. Chem. Eng. J., 56:B53.

Boessmann

, Neu

T.R.

, Horn

, Hempel

D.C.

2004. Growth, structure and oxygen penetration in particle supported autotrophic biofilms. Water Sci. Technol., 49:371.

Chen

K.-C.

, Wu

J.-Y.

, Yang

W.-B.

, Hwang

S.-C.J.

2003. Evaluation of effective diffusion coefficient and intrinsic kinetic parameters on azo dye biodegradation using PVA-immobilized cell beads. Biotechnol. Bioeng., 83:821.

Chu

K.H.

, van Veldhuizen

H.M.

, van Loosdrecht

M.C.M.

2003. Respirometric measurement of kinetic parameters: effect of activated sludge floc size. Water Sci. Technol., 48:61.

Ekama

G.A.

, Dold

P.L.

, Marais

G.v.R.

1986. Procedures for determining influent COD fractions and the maximum specific growth rate of heterotrophs in activated sludge systems. Water Sci. Technol., 18:91.

Ellis

T.G.

, Barbeau

D.S.

, Smets

B.F.

, Grady

C.P.L.

Jr.

1996. Respirometric technique for determination of extant kinetic parameters describing biodegradation. Water Environ. Res., 68:917.

Findlay

R.H.

, King

G.M.

, Watling

1989. Efficacy of phospholipid analysis in determining microbial biomass in sediments. Appl. Environ. Microbiol., 55:2888.

10.

Gillam

D.E.

, Biship

P.L.

, Love

N.G.

2005. A study of glutathione-gated potassium efflux in biofilms using potassium microelectrodes. Environ. Eng. Sci., 22:489.

11.

Goudar

C.T.

, Ellis

T.G.

2001. Explicit oxygen concentration expression for estimating extant biodegradation kinetics from respirometric experiments. Biotechnol. Bioeng., 75:74.

12.

Horn

, Hempel

D.C.

1997. Growth and decay in an auto-/heterotrophic biofilm. Water Res., 31:2243.

13.

IWA Task Group. 2000. Mathematical modeling for design and operation of biological wastewater treatment. Activated Sludge Models ASM1, ASM2, ASM2D and ASM3. London: IWA Publishing.

14.

Jørgensen

P.E.

, Eriksen

, Jensen

B.K.

1992. Estimation of viable biomass in wastewater and activated sludge by determination of ATP, oxygen utilization rate and FOD hydrolysis. Water Res., 26:1495.

15.

Kappeler

, Gujer

1992. Estimation of kinetic parameters of heterotrophic biomass under aerobic conditions and characterization of wastewater for activated sludge modeling. Water Sci. Technol., 25:125.

16.

Kim

I.S.

, Kim

, Jang

2001. Activity monitoring for nitrifying bacteria by fluorescence in situ hybridization and respirometry. Environ. Monit. Assess., 70:223.

17.

Liu

Y.-S.

, Wu

J.-Y.

, Ho

K.-P.

2006. Characterization of oxygen transfer conditions and their effects on Phaffia rhodozyma growth and carotenoid production in shake-flask cultures. Biochem. Eng. J., 27:331.

18.

2001. A field study on oxygen penetration into wastewater biofilms [Master's thesis] Edmonton, Canada: University of Alberta.

19.

Miloslav

, Nemeth

, Bodik

1993. Kinetic constants of nitrification. Water Res., 27:35.

20.

Pérez

, Picioreanu

, van Loosdrecht

2005. Modeling biofilm and floc diffusion processes based on analytical solution of reaction-diffusion equations. Water Res., 39:1311.

21.

Plattes

, Fiorelli

, Gille

, Girard

, Henry

, Minette

, O'Nagy

, Schosseler

P.M.

2007. Modelling and dynamic simulation of a moving bed bioreactor using respirometry for the estimation of kinetic parameters. Biochem. Eng. J., 33:253.

22.

Riefler

R.G.

, Ahlfeld

D.P.

, Smets

B.F.

1998. Respirometric assay for biofilm kinetics estimation: parameter identifiability and retrievability. Biotechnol. Bioeng., 57:35.

23.

Riefler

R.G.

, Smets

B.F.

2003. Comparison of a type curve and a least-squared errors method to estimate biofilm kinetic parameters. Water Res., 37:3279.

24.

Rittmann

B.E.

, Crawford

, Tuck

C.K.

, Namkung

1986. In situ determination of kinetic parameters for biofilms: Isolation and characterization of oligotrophic biofilms. Biotechnol. Bioeng., XXVIII:1753.

25.

Seo

, Bishop

P.L.

2008. The monitoring of biofilm formation in a mulch biowall barrier and its effect on performance. Chemosphere, 70:480.

26.

Taylor

D.G.

, Breen

, Bishop

P.L.

1997. Determination of phenol-degrader distribution in biofilms using gene probes. Water Res., 31:119.

27.

Temmink

, Vanrolleghem

, Klapwijk

, Verstraete

1993. Biological early warning systems for toxicity based on activated sludge respirometry. Water Sci. Technol., 28:415.

28.

Tsuneda

, Auresenia

, Morise

, Hirata

2002. Dynamic modeling and simulation of a three-phase fluidized bed batch process for wastewater treatment. Process Biochem., 38:599.

29.

2000. Stratification of microbial processes and redox potential changes in the biofilm [Ph.D. thesis] Cincinnati, OH: University of Cincinnati.

30.

Yurt

, Beyenal

, Sears

, Lewandowski

2003. Quantifying selected growth parameters of Leptothrix discophora SP-6 in the biofilm from oxygen concentration profiles. Chem. Eng. Sci., 58:4557.

31.

Zhang

T.C.

, Bishop

P.L.

1994. Density, porosity and pore structure of biofilms. Water Res., 28:2267.

32.

Zhou

X.-H.

, Shi

H.-C.

, Cai

, He

, Wu

Y.-X.

2008a. Function of self-forming dynamic membrane and bio-kinetic parameters' determination by microelectrode. Water Res., 42:2369.

33.

Zhou

X.-H.

, Shi

H.-C.

, Qiu

Y.-Q.

2008b. Inner-profiles of a spherical bio-carrier determined by microelectrodes. Biochem. Eng. J., 39:28.

34.

Zhou

X.-H.

, Shi

H.-C.

, Yu

2008c. Characterization of a spherical packing medium for biological nitrogen removal by microelectrodes. Environ. Eng. Sci., 25:375.

Estimation of Heterotrophic Biokinetic Parameters in Wastewater Biofilms from Oxygen Concentration Profiles by Microelectrode

Abstract

Abstract

Introduction

Materials and Methods

Biofilm systems

Microelectrode measurements

Phospholipids analysis

Quantification of the biokinetic parameters in the biofilm

O2 profiles in the biofilm

Results and Discussion

Biofilm

Oxygen microprofiles in the biofilm

Sensitivity analysis

Growth biokinetic parameters

Conclusions

Footnotes

Acknowledgments

Author Disclosure Statement

References

O₂ profiles in the biofilm