Biosensors and Bioelectronics 18 (2003) 23 /29 www.elsevier.

com/locate/bios

Development and characterization of a novel immobilized microbial membrane for rapid determination of biochemical oxygen demand load in industrial waste-waters
Shikha Rastogi a, Anil Kumar a, N.K. Mehra b, S.D. Makhijani c, A. Manoharan c, V. Gangal c, Rita Kumar a,*
a

Center for Biochemical Technology, Delhi University campus, Mall Road, New Delhi 7, India b Department of Zoology, Delhi University, New Delhi 7, India c Central Pollution Control Board, Parivesh Bhavan, East Arjun Nagar, New Delhi 32, India Received 19 July 2001; accepted 28 May 2002

Abstract The rapid determination of waste-water quality of waste-water treatment plants in terms of pollutional strength, i.e. biochemical oxygen demand (BOD) is difficult or even impossible using the chemical determination method. The present study reports the determination of BOD within minutes using microbial BOD sensors, as compared to the 5-day determination using the conventional method. Multiple criteria establish the basis for the development of a BOD biosensor useful for rapid and reliable BOD estimation in industrial waste-waters. Of these, preparation of a suitable novel immobilized microbial membrane used in conjunction with an apt transducer is discussed. As a result, a microbial biosensor based on a formulated, synergistic, pre-tested microbial consortium has been developed for the measurement of BOD load of various industrial waste-waters. The sensor showed maximum response in terms of current difference, when a cell concentration of 2.25 )/1010 CFU, harvested in their log phase of growth were utilized for microbial membrane construction. The sensor showed a stability of 180 days when the prepared membranes were stored at a temperature of 4 8C in 50 mM phosphate buffer of pH 6.8. The reusability of the immobilized membranes was up to 200 cycles without appreciable loss of their response characteristics. A linear relationship between the current change and a glucose /glutamic acid (GAA) concentration up to 60 mg l(1 was observed (r 0/0.999). The lower detection limit was 1.0 mg l (1 BOD. The sensor response was reproducible within 9/5% of the mean in a series of ten samples having 44 mg l (1 BOD using standard a GGA solution. When used for the BOD estimation of industrial waste-waters, a relatively good agreement was found between the two methods, i.e. 5-day BOD and that measured by the developed microbial sensor. # 2002 Elsevier Science B.V. All rights reserved.
Keywords: Biochemical oxygen demand biosensor; Immobilized microbial membrane; Biochemical oxygen demand; Microbial consortium; Industrial waste-waters

1. Introduction Concern over the pollution risk to the environment from industrial manufacturing processes and intensive agriculture has highlighted the need for rapid, easy-tooperate, low cost screening procedures which can operate in the field (Rawson et al., 1989). Among all the parameters used to assess the pollutional load of

* Corresponding author. Tel.: '/91-11-7666-156; fax: '/91-11-7667471 E-mail address: rita@cbt.res.in (R. Kumar).

water bodies/waste-waters, the biochemical oxygen demand is one of the most important and widely used parameter in the measurement of organic pollution (Karube et al., 1977). The 5-day biochemical oxygen demand (BOD) test has remained a standard pollution monitoring tool since 1936. The conventional BOD test requires a 5-day incubation period at 20 8C and demands skill in determination, thereby, making it unsuitable for process control. Therefore, a more rapid and reproducible method is desirable for assessing BOD. Rapid BOD estimations are possible using microbial sensors; which are a combination of microorganisms

0956-5663/02/$ - see front matter # 2002 Elsevier Science B.V. All rights reserved. PII: S 0 9 5 6 - 5 6 6 3 ( 0 2 ) 0 0 1 0 8 - 2

24

S. Rastogi et al. / Biosensors and Bioelectronics 18 (2003) 23 /29

immobilized in a membrane with transducer like electrochemical devices (Karube and Suzuki, 1990). The first report of such a microbial BOD sensor was published in 1977 by Karube et al., using a variety of microorganisms in pure cultures obtained from activated sludge (Karube et al., 1977; Strand and Carlson, 1984). Thereafter, many workers have reported BOD sensors using microorganisms in pure cultures isolated from activated sludge, viz. Trichosporon cutaneum (Hikuma et al., 1979; Harita et al., 1985; Riedel et al., 1988; Karube and Suzuki, 1990), Hansensula anamola (Kulys and Kadziauskiene, 1980), Clostridium butyricum (Karube et al., 1977), Escherichia coli (Kawabata and Nakamura, 1986), Bacillus subtilis (Riedel et al., 1988), Gluconobacter oxydans (Reshetilov et al., 1998) and Serratia marcescens LSY4 (Kim and Kwon, 1999). The major difficulty experienced by these workers was the application of these BOD biosensors in limited types of waste-waters. Some workers tried to formulate BOD sensors utilizing a mixed population of microorganisms isolated from activated sludge (Galindo et al., 1992; Riedel et al., 1988; Li et al., 1994; Heim et al., 1999; Liu et al., 2000). However, their applicability with respect to accuracy and reproducibility seems to be limited to a few selected waste-waters and cannot be utilized as an ideal device for all types of effluents. This may be due to a random selection of microorganisms used for the construction for BOD biosensor. Whereas, pre-testing of selected microbes in BOD analysis is necessary for the construction of a suitable and reliable BOD biosensor. Moreover, for a rapid and reliable BOD sensor, the microorganisms employed should have a wide substrate spectrum (Riedel et al., 1990). The overall characteristics of such sensors are determined by the characteristics of the immobilized microbial membranes used in conjunction with the physical transducer (Galindo et al., 1992). In all the published reports, the overall characteristics of the BOD sensor are described, however, no studies have been reported regarding the optimization of the microbial membrane component of the BOD biosensor. For the attainment of a fast and reproducible response from a BOD sensor, the immobilization support needs to be mechanically resistant and should preserve the microorganisms as long as possible. In the present study, the response characteristics of a novel immobilized microbial membrane based on mixed culture were examined. The reproducibility in construction and the shelf-life of the said membrane were evaluated. As the respiratory activity of the microorganisms depended on the immobilization conditions as well as storage conditions, the effect of storage temperature, storage pH, phase of cell growth and cell number over the response characteristics of the sensor, using synthetic and industrial waste-water samples were assessed.

2. Materials and methods

2.1. Chemicals Charged nylon membrane (Sigma) with a pore size of 0.45 mm was used throughout the investigation. DGlucose and D-glutamic acid were obtained from Sigma, Germany. Whereas, the other chemicals used to prepare the growth medium were procured from Hi-Media, India.

2.2. Microorganisms and culture conditions Bacteria were isolated from sewage samples as per the procedure of Sharma et al., 2000, with slight modification. The microorganisms were cultured in Erlenmeyer flasks kept on a rotating shaker at 37 8C at 150 rpm for a period of up to 16 h. The growth medium consisted of 0.5% sodium chloride, 0.15% beef extract, 0.15% yeast extract and 0.5% of peptic digest of animal tissue with 0.01% Tween-80. The pH of the medium was maintained at 6.89/0.2 U. The contents of the medium were sterilized by autoclaving at 121 8C for 15/20 min. The microorganisms were maintained at 4 8C in the same medium prepared with 2% agar. The microbial consortium consisted of bacteria belonging to the following genera, viz. Enterobacter cloaca , Citrobacter amalonaticus , Pseudomonas aeruginosa , Yersinia enterocolitica, Klebsiella oxytoca , Enterobacter sakazaki and Serratia liquefaciens .

2.3. Preparation of the cell suspension The cells were harvested at different phases of growth from the broth culture by centrifugation at 8000 rpm for 20 min at 4 8C. Cells were washed twice with 50 mM phosphate buffer, pH 6.8. The resulting cell pellet was suspended in 50 mM phosphate buffer, pH 6.8 and stored at 4 8C until use.

2.4. Immobilization of the bacteria Different concentrations of the cell suspension corresponding to 0.45 )/1010 /4.5 )/1010 CFU ml(1 were placed drop wise under moderate vacuum on a porous nylon membrane. The cells were entrapped in the pores of the membrane as well as adsorbed on the surface through the charge. The immobilized microbial membrane thus prepared was stored under different conditions of temperature and pH. The storage temperature ranged from 4 to 45 8C, whereas, the storage pH ranged between 6.0 and 8.0.

S. Rastogi et al. / Biosensors and Bioelectronics 18 (2003) 23 /29

25

2.5. Assembly of the microbial sensor The immobilized microbial membrane was coupled to the electrode by holding the porous nylon membrane against the Teflon gas permeable membrane by means of a nylon net (140 mesh) and attached firmly using a rubber ring. 2.6. Measurement of the response For measuring the response of the immobilized microbial membrane 100 ml of phosphate buffer; 50 mM pH 6.8 was placed in a thermostated cell at 359/ 2 8C under constant and moderate stirring. Aliquots of stock synthetic/industrial samples were added after a stable current was attained for 30 min (initial steadystate current). The current change (decrease) was observed after addition of the samples until a final steady state was reached. The response was calculated on the basis of current difference between the initial steady state current and the final steady-state current. 2.7. Standard solution A solution containing glucose (150 mg l (1) and glutamic acid (150 mg l (1) was used as a standard solution (GGA) for the calibration of the sensor. The BOD of the standard GGA solution was assumed to be 220 mg l (1 according to the Japanese Industrial Standard Committee (1986). 2.8. Determination of the 5-day biochemical oxygen demand The 5-day BOD (BOD5) of samples was determined using a range of dilutions of the standard GGA solution and industrial waste-waters (APHA, 1998).

A pre-defined microbial consortium when incorporated as biocatalyst in a microbial BOD sensor may give more reliable and reproducible BOD results as compared to sensors based on single or random combination of microorganisms. Moreover, the BOD values are a reflection of the combined metabolic state of the cell population. Keeping this in view, a formulated microbial consortium developed by our group and extensively tested with a wide range of synthetic as well as industrial waste-waters was used as biocatalyst for the construction of the BOD biosensors in the present study (Kumar et al., 1999; Manoharan et al., 2000). The overall characteristics of a BOD biosensor are determined by the characteristics of the microbial membranes used in combination with the electrochemical sensor. The use of such well defined formulated microbial consortium for the preparation of the immobilized microbial membrane, to be used for constructing a BOD biosensor might result in the development of a more ideal and reliable BOD sensor. To date, very few reports have been published regarding the optimization of the microbial membrane for the construction of a BOD biosensor (Galindo et al., 1992). For the best performance of a microbial BOD sensor, the support to be used for the immobilization of microorganisms has to be mechanically resistant and should preserve the microorganisms as long as possible. Further, parameters such as the state of microorganisms (their phase of growth at the time of harvesting) and their concentration (CFU ml (1) on the membrane may also play a vital role in determining the reproducibility of a microbial BOD sensor.

3.2. Response of the electrode The initial steady-state current at time zero is a measure of the endogenous respiration level of the immobilized microorganisms. This is due to the diffusion of oxygen from the air-saturated solution through the composed membrane. After attaining the initial steady-state current, when the standard GGA solution is added to the measuring cell; comprising the developed BOD biosensor, the substrate permeates through the novel selected membrane and is assimilated by the microbial cells, resulting in an increase in the respiration rate of the microorganisms, thereby causing a decrease in the dissolved oxygen signal (in terms of current in nA) as measured by the oxygen electrode. The current of the electrode decreases markedly with time until a final steady-state current is reached, within 5 /7 min. Therefore, a response time of 10 min was employed for further work. The final steady-state current depended on the BOD of the sample solution.

3. Results and discussion 3.1. Effect of the cell population on BOD values The BOD values as observed by the developed BOD sensor reflect the concentration of the dissolved organic substances which are assimilated/metabolized by the immobilized microorganisms (Riedel et al., 1990). The assimilation of organic substances in turn depends on the metabolic and physiological state of the immobilized biocatalysts, i.e. the microorganisms in use and their types. A number of BOD biosensors have been developed using single organisms or a random combination of organisms, but mixed cultures are shown to be an efficient biodegrading agent for organic compounds in aqueous solutions, with good kinetics, sensitivity, stability and reproducibility (Karube and Nakanishi, 1994).

26

S. Rastogi et al. / Biosensors and Bioelectronics 18 (2003) 23 /29

Fig. 1. Effect of the phase of cell growth on sensor response. Fig. 2. Response prole of the sensor with different cell loadings.

3.3. Effect of the phase of cell growth on sensor response Fig. 1 depicts the response of the sensor in terms of current (nA) using the membranes prepared from cells harvested a different phases of growth, i.e. lag, log, stationary and late stationary phase. For optimizing the response with respect to the phase of cell growth, the response was assessed with a range of GGA concentrations (5 /120 mg l (1). It was observed that the immobilized microbial membrane prepared using the cells harvested in their early log phase (up to 4 h of growth) exhibited maximum response with different GGA concentrations. Whereas, the membranes loaded with microbes in their stationary or late stationary phase showed very low response. This can be attributed to the fact that cells in their early log phase are operating at their maximum rate and efficiency. This is because of the intense metabolic capacity exhibited by the log-phase cells, which renders them more sensitive as well as responsive to physiological changes than cells from other growth phases (Caldwell, 1995). 3.4. Effect of cell loading or cell concentration on sensor response Different cell loadings of cells in their early log phase of growth were utilized for the construction of the microbial membrane. Fig. 2 shows the response of the sensor with a range of GGA concentrations, when cell concentrations ranging from 0.45 )/1010 /4.5 )/1010 CFU ml(1 were utilized for microbial membrane construction. When used in conjunction with a dissolved oxygen probe to assess the response, it was observed that as the cell loading on the membrane increases till a particular concentration, the response of the sensor increases, beyond which it remained all the more constant. As illustrated in Fig. 2, maximum response

was observed with cell loadings nearing 2.25 )/1010 CFU ml(1. Cell loadings lower than 2.25 )/1010 CFU ml (1 did not gave appreciable response. This is also evident from other published reports which state that the biofilm prepared for a BOD biosensor should not contain less than 107 /108 cells ml (1 (Li et al., 1994). 3.5. Inuence of storage temperature and storage pH on the efcacy of the BOD biosensor The physiology and metabolic state of the microorganisms depends on their storage conditions. Extremes of temperature(s) probably results in the inactivation of enzymes or other functional cell structures, such as cell membranes. On the other hand, pH exerts a marked control on enzyme activity and the pH of the medium may affect cell permeability and other physiological activities (Moat and Foster, 1995). Therefore, in the present study, the effect of the storage temperature and pH on the efficacy of the microbial membrane was studied (Figs. 3 and 4). The influence of storage temperature on the microbial activity and thus the response of the sensor is shown in Fig. 3. It was observed that the prepared microbial membranes stored at a temperature of 4 8C showed maximum response as compared to the membranes stored at other three temperatures, i.e. 4, 15 and 37 8C. Simultaneously, the influence of pH on the current output of the sensor employing the prepared microbial membranes was studied. Fig. 4 illustrates the effect of storage pH on the microbial activity and thus the sensor response. It is evident from the graph that maximum microbial activity and therefore the resulting current change was observed using the microbial membranes stored at a pH of 6.8 followed by 7.2 and 6.4. At pH values 6.0, 7.6 and 8.0 appreciable current change was not observed. This

S. Rastogi et al. / Biosensors and Bioelectronics 18 (2003) 23 /29

27

Fig. 6. Stability of the developed BOD sensor.

decrease in the stability. In addition, the membrane was reusable for around 200 cycles.
Fig. 3. Inuence of storage temperature on the microbial activity and thus the sensor response.

3.7. Response of the sensor with a range of GGA concentrations and correlation with the conventional BOD method The developed BOD sensor was used to analyze the BOD values of different concentrations of the standard GGA solution. Fig. 5 shows a calibration curve of the microbial electrode sensor when the diluted standard solution of GGA was employed for the experiments. A linear relationship was observed between the current difference (between initial steady-state current and final steady-state current) and the 5-day BOD of the standard solution up to a concentration of 60 mg l (1. The linear range of the sensor is defined as the substrate range that gives a signal directly proportional to the concentration (Chan et al., 2000). The lower limit of detection was 1 mg l (1 BOD, by the developed sensor. The current was reproducible within 9/5% of the mean in a series of ten samples having 44 mg l(1 BOD, using standard GGA solution.

Fig. 4. Effect of storage pH on the sensor response.

shows that a pH of 6.8 was conducive for the storage of microorganisms in use. Literature also reveals that a pH of 6.8 /7.0 and a temperature of 4 8C are more conducive for the growth and survival of microorganisms (Galindo et al., 1992).

3.6. Storage stability study of the immobilized microbial membrane Fig. 6 depicts the response characteristics of the prepared microbial membrane when stored at a temperature of 4 8C in 50 mM phosphate buffer, pH 6.8. As evident from the figure, the developed immobilized microbial membrane used in the BOD biosensor was stable up to a period of 180 days with only 8.3%

Fig. 5. Calibration curve depicting the correlation of sensor BOD with conventional BOD (BOD5).

28

S. Rastogi et al. / Biosensors and Bioelectronics 18 (2003) 23 /29

Table 1 Comparison of BOD values estimated by the sensor with those determined by the 5-day BOD method for various industrial wastewaters Waste-water sample Observed BOD values (mg l (1)a BOD5 method Food industry Tannery Pulp and paper industry
a

BOD sensor 29,041 1098 305

29,352 1164 414

All values are average of three experiments.

3.8. Estimation of the BOD of industrial waste-waters using the developed BOD sensor The developed BOD sensor was used to analyze real waste-water samples. For the same samples, 5-day BOD was also determined by the conventional method for comparison with the BOD values estimated by the sensor. The waste-water samples tested were from food, tannery and pulp & paper industry. Each of these waste-waters were diluted appropriately with buffer depending on their BOD load. Table 1 shows the comparative BOD values for different industrial wastewaters as examined by both the methods. As depicted in Table 1, relatively good agreement between the two methods was obtained for the test samples obtained from the food industry. On the other hand, for the industrial waste-water samples procured from tannery and pulp & paper industry, the BOD values as obtained by the sensor were slightly lower as compared to those obtained using the 5-day BOD method. This can be explained because of the fact that the rate of oxidation for such substrates was slower than that of the standard substrates, glucose and glutamic acid (Hyun et al., 1993). Further, these differences might arise from high-molecular-weight substrates, which are difficult for the immobilized bacteria to assimilate when sandwiched between artificial membranes (Riedel et al., 1988). Liu et al. (2000) have also reported that proper selection of the calibration solution is a key issue for obtaining good agreement between results of the sensor BOD measurement and the conventional BOD analysis. There might not be a universal standard solution that would be suitable for the calibration of real waste-water samples of different compositions.

method and the support adopted for the construction of an immobilized microbial membrane should be such as to provide maximum stability and reproducibility to the sensor. The phase of cell growth at which the microorganisms are to be harvested for microbial membrane construction was standardized and it was found that the cells harvested in their log phase of growth were exhibiting their maximum metabolic potential. A cell loading of 2.25 )/1010 CFU/membrane was found to give optimum response with a wide range of GGA concentrations. A storage pH of 6.8 and a temperature of 4 8C were found to be conducive for longer stability, viability and shelf-life of the immobilized microbes. The microbial membranes prepared, characterized and selected in the present invention, provide a stability of 180 days to the developed sensor and were reusable for approximately 200 cycles. A linear relationship between the current change and GGA concentration up to 60 mg l (1 was observed. The lower detection limit was 1 mg l (1 BOD. The developed sensor was found to be suitable for the estimation of BOD of industrial wastewaters having low-moderate-high biodegradable organic matter. The present study emphasised the novelty of the immobilized microbial membrane w.r.t. the support (charged nylon membrane), the biocatalyst (formulated microbial consortium) and the mode of immobilization (entrapment as well adsorption due to the charge). This has led to a more stable membrane with increased shelflife exhibiting higher reusability. A good correlation (r0/0.999) was observed between the BOD values estimated by the conventional method and that by the developed sensor up to a GGA concentration of 60 mg l (1 (having BOD of 44 mg l (1). A wide range of industrial waste-waters having low /moderate /high biodegradable organic matter could be assessed for their BOD load within a short time period (5 /10 min) using the developed BOD biosensor.

Acknowledgements We acknowledge the Ministry of Environment and Forests, New Delhi for nancial assistance. The author S.R. gratefully acknowledges the CSIR, New Delhi for Senior Research Fellowship.

4. Conclusion References A number of BOD biosensors have been developed using single organisms but mixed cultures are shown to be an efficient biodegrading agent for organic compounds in aqueous solutions, with good kinetics, sensitivity, stability and reproducibility. Further, the
APHA, 1998. Standard Methods for the Examination of Waters and Waste-Waters, 20th ed., American Public Health Association, Washington, D.C. Caldwell, D.R., 1995. Microbial Physiology and Metabolism. WCB Publishers, England.

S. Rastogi et al. / Biosensors and Bioelectronics 18 (2003) 23 /29 Chan, C., Lehmann, M., Chan, K., Chan, P., Chan, C., Gruendig, B., Kunze, G., Renneberg, R., 2000. Designing an amperometric thick lm microbial BOD sensor. Biosens. Bioelectron. 15, 343 /353. Galindo, E., Garcia, J.L., Torres, L.G., Quintero, R., 1992. Characterization of microbial membranes used for the estimation of biochemical oxygen demand with a biosensor. Biotechnol. Tech. 6, 399 /404. Harita, K., Otani, Y., Hikuma, M., Yasuda, T., 1985. BOD quick estimating system utilizing microbial electrode. Instrumental Control. Water, Waste-Water Treatment Transport System. Proc. IAWPRC Workshop 4, 529 /532. Hikuma, M., Suzuki, H., Yasuda, T., Karube, I., Suzuki, S., 1979. Amperometric estimation of BOD by using living immobilized yeast. Eur. J. Appl. Microbiol. Biotechnol. 8, 289 /297. Heim, S., Schnieder, I., Binz, D., Vogel, A., Bilitewski, U., 1999. Development of an automated microbial sensor system. Biosens. Bioelectron. 14, 187 /193. Hyun, C.K., Tamiya, E., Takeuchi, T., Karube, I., 1993. A novel BOD sensor based on bacterial luminescence. Biotechnol. Bioeng. 41, 1107 /1111. Japanese Industrial Standard Committee, 1986. Testing methods for industrial waste-waters. JIS K 0120. Japanese Standard Association, Tokyo, pp. 47 /51. Kawabata, N., Nakamura, N., 1986. New BOD sensor utilizing a functional polymer which captures microorganisms alive. In: Abstracts of the International Symposium on New Sensors and Methods for Environmental Characterization, Kyoto, Japan. Karube, I., Nakanishi, K., 1994. Immobilised cells used for detection and analysis. Curr. Opin. Biotechnol. 5, 54 /59. Karube, I., Suzuki, M., 1990. Microbial biosensors. In: Cass, E.A.G. (Ed.), Biosensor */A Practical Approach. IRL Press, Oxford, pp. 155 /169. Karube, I., Matsunaga, T., Mitsuda, S., Suzuki, S., 1977. Microbial electrode BOD sensors. Biotechnol. Bioeng. 19, 1535 /1547. Kim, M.N., Kwon, H.S., 1999. Biochemical oxygen demand sensor using Serratia marcescens LSY4. Biosens. Bioelectron. 14, 1 /7.

29

Kulys, J.J., Kadziauskiene, K.V., 1980. Yeast BOD sensor. Biotechnol. Bioeng. 22, 221 /226. Kumar, R., Kumar, A., Sharma, A., Gangal, V., Makhijani, S.D., 1999. Formulation and standardization of microbial composition useful for reproducible BOD estimations. J. Environ. Sci. Health Part A 34, 125 /144. Li, F., Tan, T.C., Lee, Y.K., 1994. Effects of pre-conditioning and microbial composition on the sensing efcacy of a BOD biosensor. Biosens. Bioelectron. 9, 197 /205. Liu, J., Bjornsson, L., Mattiasson, B., 2000. Immobilized activated sludge based biosensor for biochemical oxygen demand measurement. Biosens. Bioelectron. 14, 883 /893. Manoharan, A., Gangal, V., Makhijani, S.D., Sharma, A., Kumar, A., Kumar, R., 2000. Validation of the use of microbial consortium as standard seeding material in BOD determination. Hydrobiologia 430, 77 /86. Moat, A.G., Foster, J.W., 1995. Microbial Physiology, 3rd ed.. Wiley / Liss, New York. Reshetilov, A.N., Efremov, D.A., Iliasov, P.V., Boronin, A.M., Kukushskin, N.I., Greene, R.V., 1998. Effects of high oxygen concentration on microbial biosensor signals: hyperoxygenation by means of perourodecalin. Biosens. Bioelectron. 13, 795 /799. Rawson, D.M., Willmer, A.J., Turner, A.P.F., 1989. Whole-cell biosensors for environmental monitoring. Biosensors 4, 299 /311. Riedel, K., Renneberg, R., Kuhn, M., Scheller, F., 1988. A fast estimation of biochemical oxygen demand using microbial sensors. Appl. Microbiol. Biotechnol. 28, 316 /318. Riedel, K., Lange, K.P., Stein, H.J., Kuhn, M., Ott, P., Scheller, F., 1990. A microbial sensor for BOD. Water Res. 24, 883 /887. Sharma, A., Kumar, R., Kumar, A., Gangal, S.V., 2000. Application of dened co-immobilized microbial consortium as a ready-to-use seeding material in BOD analysis. Environ. Monit. Assess. 60, 251 /260. Strand, S.E., Carlson, D.A., 1984. Rapid BOD measurement for municipal waste-water samples using a biolm electrode. J. Water Pollut. Control Fed. 56, 464 /467.