- Email: [email protected]
A cost–benefit analysis of methods for the determination of biomass concentration in wastewater treatment
ARTICLE IN PRESS
Anaerobe 12 (2006) 254–259 www.elsevier.com/locate/anaerobe
Ecology/environmental microbiology
A cost–benefit analysis of methods for the determination of biomass concentration in wastewater treatment J.E. Hernandez, R.T. Bachmann, R.G.J. Edyvean Department of Chemical and Process Engineering, The University of Sheffield, Mappin Street, S1 3JD, UK Received 24 May 2006; received in revised form 26 September 2006; accepted 27 September 2006
Abstract The measurement of biomass concentration is important in biological wastewater treatment. This paper compares the accuracy and costs of the traditional volatile suspended solids (VSS) and the proposed suspended organic carbon (SOC) methods. VSS and SOC values of a dilution system were very well correlated (R2 ¼ 0:9995). VSS and SOC of 16 samples were determined, the mean SOC/VSS ratio (0.52, n ¼ 16, s ¼ 0:01) was close to the theoretical value (0.53). For costing analysis, two hypothetical cases were analysed. In case A, it is assumed that 108 samples are analysed annually from two continuous reactors. Case B represents a batch experiment to be carried out in 24 incubated serum bottles. The savings, when using the SOC method, were £11 987 for case A and £90 for case B. This study suggests the use of SOC method as a time saving and lower cost biomass concentration measurement. r 2006 Elsevier Ltd. All rights reserved. Keywords: Biomass concentration; Cost analysis; Suspended organic carbon; Volatile suspended solids; Wastewater
1. Introduction Biomass concentration is a fundamental parameter used to assess, monitor, compare and model biological treatment processes. It is usually reported as volatile suspended solids (VSS) and sometimes as total solids (TS) or volatile solids (VS). VSS, which account for suspended organic and inorganic matter, are determined by measuring the loss of weight of the ignited (550 1C) total suspended solids (TSS) retained on a 1.2 mm grade glass-fibre filter oven-dried at 103–105 1C as described in standard methods [1]. When the performance of several anaerobic bioreactors is being evaluated quantitatively, the number of samples needs to be high as they have to be analysed at least in triplicate. The process involves a thermal pre-treatment cycle to calibrate the crucibles, i.e. oven drying, igniting, cooling, desiccating, and weighing. Thereafter, the filter and sample are placed in the crucible and the analysis is carried out repeating the same cycle until weight change is less than 4%. As a result, the determination of VSS normally Corresponding author. Tel.: +44 114 275 4697; fax: +44 114 222 7501.
E-mail address: cpp02jeh@sheffield.ac.uk (J.E. Hernandez). 1075-9964/$ - see front matter r 2006 Elsevier Ltd. All rights reserved. doi:10.1016/j.anaerobe.2006.09.005
requires more than one working day. Another constraint is that high bias can be introduced when filtering, handling and preserving the sample as well as losses of the filter paper at 550 1C. Therefore, a robust, simple, economic and faster method is required. Attempts to overcome the economic-operational inconveniences of the VSS method have been made, and the chemical analysis of suspended organic carbon (SOC), which represents all carbon atoms with covalent bonds in organic molecules, may fulfil the requirements. Again, there is standard methodology available involving the spectrophotometric measurement of CO2 released from the conversion of organic carbon [1] Xing et al. [2] reported a method to determine VSS from such SOC data and propose a correlation between VSS and chemical oxygen demand (COD). The difference between the experimental and theoretical data was 2.1% when bacteria were formulated as C5H7NO2. Assuming their complete oxidation, this gives a COD/VSS ratio of 1.2. More recently, Lee and Pavlostathis [3] have developed a calibration curve correlating VSS with particulate organic matter retained on a 1.2 mm pore-size filter. However, the experimental data correlating these values was not reported.
ARTICLE IN PRESS J.E. Hernandez et al. / Anaerobe 12 (2006) 254–259
In spite of this, the use and correlation of SOC has not found widespread usage due to the cost, the hazardous nature of the chemicals involved (mercury, hexavalent chromium, sulphuric acid and silver), the expensive and complex nature of the analytical instruments used and the potential for errors in the methodologies (in the hightemperature method salts can promote inhibition over the catalytic surface by melting while in the persulphate method a low pH can result in incomplete oxidation). However, low cost, relatively simple technologies are now available for determining total organic carbon (TOC). The technology, which satisfies EN 1484:1997 [4], is based on the collection of CO2 that passes trough a membrane into a cell containing a colour indicator, and gives comparable results to an automatic organic carbon analyser [5]. This work compares the data and costs of VSS and SOC used in determining biomass concentration in an anaerobic digester. 2. Methods 2.1. Correlation standards A correlation between VSS and SOC was performed in triplicate for a series of dilutions of methanogenic sludge whose original biomass concentration was 3.88 g VSS L1. The sludge was diluted at a volumetric ratio of 1:10 with organic carbon free distilled water, and then diluted again at ratios of: 0.9:9.1, 1.8:8.2, 3.6:6.4, 5.4:4.6 and 7.2:2.8, respectively. A second correlation between experimental data of VSS and SOC was performed with 16 samples, by triplicate, of acidogenic sludge obtained from an acidogenic reactor [6]. The measurement of VSS was carried out in triplicate following a standard method [1]. SOC was determined using the TOC cuvette test (50–500 mg TOC/L) from Dr. Lange, Germany. The TOC was analysed for the raw sample and for the dissolved fraction filtered by using 1.2 mm pore diameter GF/C Whatman paper, and was termed dissolved organic carbon (DOC). The SOC was calculated by subtracting the DOC from the TOC. The
255
VSS and SOC values were compared with the stoichiometric SOC/VSS ratio of 0.53. This theoretical value was calculated from the elemental composition of bacteria, C5H7NO2, proposed by Hoover and Porges [7]. Thus, 5 atoms of carbon are contained per molecular weight. Therefore, for 1 g of bacteria, the amount of organic carbon is 0.53 g. Consequently, upon complete oxidation the organic carbon is converted to carbon dioxide, which is measured by the SOC method, according to the following reaction [7]: C5 H7 NO2 þ 5O2 ! 5CO2 þ NH3 þ 2H2 O: 2.2. Cost analysis The cost analysis and comparison for the SOC and VSS methods were carried out using the activity-based costing system (ABC). The ABC system refines costing systems by focusing on individual activities as the fundamental cost objects and assigns costs to products and services. It takes into account designing products, setting up machines, operating machines and distributing products [8]. Two hypothetical cases were analysed. It was assumed that both scenarios involve an initial investment to buy the necessary items for a new laboratory from commercial laboratory suppliers (Table 1). Some companies may buy on credit and others may use one single cash payment. This affects directly the cash inflows and outflows and may increase or decrease the investment in a period of time, thus altering the accounts. For case A it was assumed that 108 samples are analysed per year. This is a likely situation when monitoring the performance of two reactors sampled at weekly intervals. Storage and processing of samples was not included in the costing. The cost of labour is taken for one person for 365 days with a minimum national wage of £4.85 an hour [9] and electricity costs of £1.91 per kWh [10]. For case B it was assumed that a total of 24 samples are analysed over 4 days. This includes one set of blanks and controls and sets of six samples analysed in triplicate.
Table 1 Initial investment costs excluding VAT for a new laboratory to carry out VSS or SOC analysis VSS
SOC
Equipment
Units
Price £
Equipment
Price £
Units
Price £
Furnace (550 1C, 30 L) Air-circulating Oven (105 1C) Balance (LoD: 0.1 mg) Filtration apparatus Vacuum pump Desiccator (300 mm diameter) Pipettor (5 mL) Crucibles Tongue
1 1 1 1 1 1 1 3 1
2500 1000 1500 160 575 250 150 27 13
Spectrophotometer Thermostat Magnetic stirrer Powder dispenser Pipettor
1060 380 60 32 150
1 1 1 1 1
1060 380 60 32 150
Total initial investment
6175
1682
ARTICLE IN PRESS J.E. Hernandez et al. / Anaerobe 12 (2006) 254–259
256
2.3. Scanning electron microscopy As the matter capable of volatilization at 550 1C may be organic and inorganic, a sample was examined under the SEM microscope to observe the structure of the microbial aggregate and the space relationship between microorganisms and EPS and other materials. Specimens were fixed in Karnovsky’s fixative in 0.1 M phosphate buffer for 3 h at 4 1C, and then washed in sucrose (10%) in 0.1 M phosphate buffer. Afterwards, they were washed three times with sucrose solution 30 min intervals at 4 1C. Secondary fixation was carried out in 2% osmium tetroxide aqueous for 1 h at room temperature. Dehydration was carried out by a graded series of ethanol at room temperature. The specimens were placed in a 50/50 mixture of 100% hexamethyldisilazane and ethanol for 30 min, followed by 30 min in 100% hexamethyldisilazane and then they were allowed to air-dry overnight. After drying, they were mounted on 12.5 mm diameter stubs and attached with sticky tabs and then coated in an Edwards S150B sputter coater with approximately 25 nm of gold. Specimens were examined in a Philips PSEM 501B Scanning Electron Microscope at an accelerating voltage of 30 kV. 3. Results and discussion 3.1. VSS and SOC correlations The results of the VSS and SOC analysis of serially diluted methanogenic sludge correlate very well (R2 ¼ 0:9995, Fig. 1). The SOC/VSS ratio for the calibration curve is 0.52 (standard deviation s ¼ 0:02), which is close to the theoretical value of 0.53. The average SOC/ VSS ratio for the 16 samples from a two-stage anaerobic digester was found to be 0.52 (n ¼ 16, s ¼ 0:01), 1.89% less than the theoretical value of 0.53 (Fig. 1). The experimental data reported here are in good agreement with the theoretical stoichiometric SOC/VSS value of 0.53 showing that VSS and SOC data for
wastewater sludge are interchangeable. It should be noted however, that neither the SOC nor the VSS values truly represent microbial concentration as the fraction of extracellular polymeric substance (EPS) as well as other non-microbial organic suspended compounds such as fibrous cellulose cannot be distinguished from microbial organic carbon (Fig. 2). Therefore, this method cannot accurately assess the concentration of microbes forming aggregates. Also, other materials might not be quantified, e.g. organic acids, which might be lost owing to the stripping out of inorganic carbon. Empirical relationships between SOC and VSS might be valid for specific source water and are independently established for a particular matrix taken at different points of the treatment process [1]. However, with the determination of SOC as a unique biomass concentration parameter, such correlations will no longer be necessary. 3.2. Cost analysis of Case A The results of the cost analysis of VSS and SOC methods for case A, which assumes the analysis of 108 samples over 1 year, are shown in Table 2. After applying the ABC costing system, the yearly cost of the VSS method for 108 samples is £14 544 (Table 2) and the total cost in the first year is £20 719 with initial capital investment (from Table 1). The yearly cost of the SOC method for 108 samples is £2557 (Table 2) and a total of £4239 in the first year including initial investment (from Table 1). Therefore, the savings on total cost per year, without considering initial investment, would be £11 987 if the SOC method is used. The determination of one triplicate might take 20.7 and 3.2 h for VSS and SOC, respectively, thus the time savings are also considerable. 3.3. Cost analysis of case B The results of the cost analysis of VSS and SOC methods for case B are shown in Table 3. The cost of the VSS
-1
SOC (mg L )
150
100
50
0 0
100
200
300
-1
VSS (mg L ) Fig. 1. Linear relationship between volatile suspended solids (VSS) and suspended organic carbon (SOC) of serially diluted methanogenic biomass (&). Samples from acidogenic reactor (J).
Fig. 2. SEM micrograph of methanogenic sludge degrading phenol. A group of bacteria is located inside the dark circle and surrounded by EPS and other substances.
ARTICLE IN PRESS J.E. Hernandez et al. / Anaerobe 12 (2006) 254–259
257
Table 2 Cost analysis of VSS and SOC methods under ABC costing system for case A VSS Direct costs Direct material Pipette tip Filter paper b Direct cleaning and maintenance costs Balance Total direct costs
Units 108 324
Cost ea. £ 0.02 0.6
1
25
Indirect costs Overheads Filtration: electricity 185 W pumpc Filtratione Calibration of crucible: 103–105 1C setting upg Calibration of crucible: 103–105 1C Calibration of crucibre: 550 1C setting upg Calibration of crucibre: 550 1Ch Calibration of crucible: working timei SS determination: 103–105 1C setting upg SS determination: 103–105 1C dryingj SS determination working timek VSS determination: 550 1C setting upg VSS determination: 550 1C ignitionj VSS determination working timek Total indirect costs Total costs per year
Subtotal £ 2.16 194.4
Hours
Allocation base £
kWh
1.5 1.5 1 1 1 0.5 8 1 2 6.1 1 1 5.1
1.91d 4.85f 1.91d 1.91d 1.91d 1.91d 4.85f 1.91d 1.91d 4.85f 1.91d 1.91d 4.85f
0.093 1.4 1.4 4 4 1.4 1.4 4 4
Total £a (1)
Per sample £ (2) ¼ (1)/108
196.56
1.82
25 221.56
0.23 2.05
Total £
Per sample £
(1) 0.48 785.70 288.79 288.79 825.12 412.56 4,190.40 288.79 577.58 3,177.72 825.12 7.64 2,653.92 14,322.62 14,544.18
(2) ¼ (1)/108 0.00 7.28 2.67 2.67 7.64 3.82 38.80 2.67 5.35 29.42 7.64 0.07 24.57 132.62 134.67
SOC Direct costs Direct materials Cuvette (25 units pack)l Pipette tip Direct cleaning and maintenance costsm Total direct costs
Units
Cost ea. £
(1)
(2) ¼ (1)/108
9 432 1
95.55 0.02 25
859.95 8.64 25.00 893.59
7.96 0.08 0.23 8.27
Total £
Per sample £
(1) 2.06 2.29 1658.70 1663.05 2556.64
(2) ¼ (1)/108 0.019 0.021 15.358 15.399 23.67
Indirect costs
Hoursg
Allocation base £
KWh
Stirring of samplen Heating of sample Working timeo Total indirect costs Total costs per year
0.2 2 3.2
1.91d 1.91d 4.85f
0.06 0.6
a
This cost is for 108 samples. The total filter papers needed are calculated by multiplying 108 samples times a triplicate. c The filtration might last 90 min including washings of retained material. d This is the cost of electricity per kWh. e The operator must be present during the 90 min of filtration. f This is the minimum national wage per hour. g The oven and furnace must be previously set up at the desired temperatures, this might take 1 h. h The ignition is carried out twice for 15 min, assuming that two cycles are enough to get less than 4% difference between weight readings. i Calibrating the crucibles might take 8 working hours. The crucibles must be cleaned and identified also the operator must be in the laboratory during drying, igniting, cooling and weighing. j It is assumed that in the second cycle, TSS (1 h) and VSS (0.5 h) can be determined with less than 4% difference between weight readings. k The operator must be in the laboratory during the drying or igniting–cooling–weighing process. l Two cuvettes are needed to analyse separately DOC and TOC in every sample. m Maintenance for spectrophotometer once a year. n Four cuvettes are stirred by pairs of TOC and DOC for a total of 10 min. o The operator must be in the laboratory during the SOC determination. b
method is £326 (Table 3) for the samples and £6501 in the first year including the initial investment (from Table 1). For SOC, the total cost is £236 for the samples (Table 3)
and a first year total of £1918 (from Table 1). Thus, the SOC method results in savings of £90 on total costs without considering initial investment. The determination
ARTICLE IN PRESS 258
J.E. Hernandez et al. / Anaerobe 12 (2006) 254–259
Table 3 Cost analysis of VSS and SOC methods under ABC costing system for case B VSS Direct cost Direct Material Pipette tip Filter paper Direct cleaning and maintenance costs Balance Total direct costs
Units 24 24
Cost ea. £ 0.02 0.6
1
25
Indirect costs Overheads Filtration: electricity 185 W pumpb Filtrationd Calibration of crucible: 103–105 1C setting upf Calibration of crucible: 103–105 1Cg Calibration of crucibre: 550 1C setting upf Calibration of crucibre: 550 1Ch Calibration of crucible: working timei SS determination: 103–105 1C setting upg SS determination: 103–105 1C dryingj SS determination working timek VSS determination: 550 1C setting upf VSS determination: 550 1C ignitionj VSS determination working timek Total indirect costs Total costs per year
Subtotal £ 0.48 14.4
Hours
Allocation base £
kWh
0.5 0.5 1 1 1 0.5 16 1 2 12.1 1 1 10.1
1.91c 4.85e 1.91c 1.91c 1.91c 1.91c 4.85e 1.91c 1.91c 4.85e 1.91c 1.91c 4.85e
0.093 1.4 1.4 4 4 1.4 1.4 4 4
Total £a (1)
Per sample £ (2) ¼ (1)/24
14.88
0.620
25 39.88
0.23 0.85
Total £
Per sample £
(1) 2.13 58.20 2.67 2.67 7.64 3.82 77.60 2.67 5.35 58.85 7.64 7.64 49.15 286.03 325.91
(2) ¼ (1)/24 0.09 2.43 0.11 0.11 0.32 0.16 3.23 0.11 0.22 2.45 0.32 0.32 2.05 11.92 12.77
SOC Direct costs Direct materials Cuvette (25 units pack)l Pipette tip Direct cleaning and maintenance costsm Total direct costs Indirect costs Overheads Stirring of samplen Heating of sample Working timeo Total indirect costs Total costs
Units
Cost ea. £
(1)
(2) ¼ (1)/24
2 96
95.55 0.02
191.10 1.92 25 218.02
7.96 0.08 1.04 9.08
Total £
Per sample
(1) 0.23 2.29
(2) ¼ (1)/24 0.01 0.10
17.88 235.90
0.11 9.19
Hours
Allocation base £
KWh
0.1 2 3.2
1.91c 1.91c 4.85e
0.06 0.6 15.36
a
This cost is for 24 samples. Filtration of one triplicate might last 30 min per crucible including washing of retained solids; this is performed for 24 samples. c This is the cost of electricity per kWh. d The operator must be present during 30 min of filtration process per sample. e This is the minimum national wage per hour. f The oven and furnace must be previously set up at the desired temperatures, this might take 1 h. g It is assume that both oven and furnace have the capacity to allocate 24 samples in one run for 1 h. h The ignition is carried out twice for 15 min, assuming that two cycles are enough to get less than 4% difference between weight readings. i The operator must work at least two periods of 8 h to calibrate 24 crucibles and must be present during drying, igniting, cooling and weighing. j It is assumed that in the second cycle TSS (1 h) and VSS (0.5 h) can be determined with less than 4% difference between weight readings. k The determinations are carried out in two sessions of 6.05 h and 5.05 for TSS and VSS respectively; this is also under supervision of operator. l Two cuvettes are needed to analyse DOC and TOC in every sample. m Maintenance for spectrophotometer once a year. n Two cuvettes are stirred simultaneously to determine TOC and DOC, this is for preventing CO2 diffusion into the sample. o The operator must be present during the SOC determination. b
of biomass in 24 samples might take 38.7 and 3.2 h for VSS and SOC, respectively, again giving considerable time saving.
This shows that the determination of SOC of a batch of samples is slightly cheaper than applying the VSS method. This can be explained as in this case, the working time of the
ARTICLE IN PRESS J.E. Hernandez et al. / Anaerobe 12 (2006) 254–259
operator can be efficiently utilized to process a big number of samples at the same time. This situation does not occur when a continuous process is monitored regularly. 4. Conclusion The results show that the SOC/VSS ratio from both a serially diluted methanogenic sludge (0.52, s ¼ 0:02) and sludge samples taken from an acidogenic reactor (0.52, n ¼ 16, s ¼ 0:01) are close to the theoretical value of 0.53 for complete oxidation of bacteria formulated as C5H7NO2. This suggests that such a formula can be applied for active anaerobic biomass as well as pure cultures. For both scenarios, cost analysis shows the SOC method to be the lower cost and time-saving option. The savings are £11 987 for case A and £90 for case B. The initial investment required in a new laboratory does not affect the outcome in either case. These experimental and economical results suggest that the traditional VSS method can be replaced by the more economic and time-saving SOC method for the measurement of microbial biomass. Acknowledgements We would like to thank the sponsorship of CONACyT and help of N. Coutelle and N. Garcia.
259
References [1] Clesceri LS, Greenberg AE, Eaton AD. Standard methods for the examination of water and wastewater, 8th ed. Washington: American Public Health Association, American Water Works Association; 1998. [2] Xing CH, Wen XH, Qian Y, Tardieu E. Microfiltration–membranecoupled bioreactor for urban wastewater reclamation. Desalination 2001;141:63–73. [3] Lee YH, Pavlostathis SG. Decolorization and toxicity of reactive anthraquinone textile dyes under methanogenic conditions. Water Res 2004;38:1838–52. [4] CEN. EN 1484:1997: Water analysis. Guidelines for the determination of total organic carbon (TOC) and dissolved organic carbon (DOC). European Standarization Committee; 1997. [5] Kraatz C, Wochnik P. Dr. Lange TOC cuvette test conforms to European standard. In: Application report no. 60, Du¨sseldorf: Dr. Bruno Lange GmbH & Co; 1998. [6] Hernandez JE, Edyvean RG. Anaerobic treatment of phenol in a two-stage anaerobic reactor. In: IchemE, editor, Seventh world congress of chemical engineering. SECC, Glasgow, Scotland. Rugby, UK: IChemE; 2005. [7] Tchobanoglous G, Burton FL, Stensel HD. Wastewater engineering: treatment and reuse, 4th ed. New York: McGraw-Hill; 2003. [8] Horngren C, Bhimani A, Datar SM, Foster G. Management and cost accounting. 2nd ed. Essex: Prentice-Hall; 2002. [9] Anon. National minimum wage. UK: Department of Trade and Industry; 2004. [10] Anon. Electricity Tariff Finder. /http://www.powergen.co.uk/Pub/ Sme/A/Ui/Quotation/ElecQuoteStart.aspxS; 2005.
Anaerobe 12 (2006) 254–259 www.elsevier.com/locate/anaerobe
Ecology/environmental microbiology
A cost–benefit analysis of methods for the determination of biomass concentration in wastewater treatment J.E. Hernandez, R.T. Bachmann, R.G.J. Edyvean Department of Chemical and Process Engineering, The University of Sheffield, Mappin Street, S1 3JD, UK Received 24 May 2006; received in revised form 26 September 2006; accepted 27 September 2006
Abstract The measurement of biomass concentration is important in biological wastewater treatment. This paper compares the accuracy and costs of the traditional volatile suspended solids (VSS) and the proposed suspended organic carbon (SOC) methods. VSS and SOC values of a dilution system were very well correlated (R2 ¼ 0:9995). VSS and SOC of 16 samples were determined, the mean SOC/VSS ratio (0.52, n ¼ 16, s ¼ 0:01) was close to the theoretical value (0.53). For costing analysis, two hypothetical cases were analysed. In case A, it is assumed that 108 samples are analysed annually from two continuous reactors. Case B represents a batch experiment to be carried out in 24 incubated serum bottles. The savings, when using the SOC method, were £11 987 for case A and £90 for case B. This study suggests the use of SOC method as a time saving and lower cost biomass concentration measurement. r 2006 Elsevier Ltd. All rights reserved. Keywords: Biomass concentration; Cost analysis; Suspended organic carbon; Volatile suspended solids; Wastewater
1. Introduction Biomass concentration is a fundamental parameter used to assess, monitor, compare and model biological treatment processes. It is usually reported as volatile suspended solids (VSS) and sometimes as total solids (TS) or volatile solids (VS). VSS, which account for suspended organic and inorganic matter, are determined by measuring the loss of weight of the ignited (550 1C) total suspended solids (TSS) retained on a 1.2 mm grade glass-fibre filter oven-dried at 103–105 1C as described in standard methods [1]. When the performance of several anaerobic bioreactors is being evaluated quantitatively, the number of samples needs to be high as they have to be analysed at least in triplicate. The process involves a thermal pre-treatment cycle to calibrate the crucibles, i.e. oven drying, igniting, cooling, desiccating, and weighing. Thereafter, the filter and sample are placed in the crucible and the analysis is carried out repeating the same cycle until weight change is less than 4%. As a result, the determination of VSS normally Corresponding author. Tel.: +44 114 275 4697; fax: +44 114 222 7501.
E-mail address: cpp02jeh@sheffield.ac.uk (J.E. Hernandez). 1075-9964/$ - see front matter r 2006 Elsevier Ltd. All rights reserved. doi:10.1016/j.anaerobe.2006.09.005
requires more than one working day. Another constraint is that high bias can be introduced when filtering, handling and preserving the sample as well as losses of the filter paper at 550 1C. Therefore, a robust, simple, economic and faster method is required. Attempts to overcome the economic-operational inconveniences of the VSS method have been made, and the chemical analysis of suspended organic carbon (SOC), which represents all carbon atoms with covalent bonds in organic molecules, may fulfil the requirements. Again, there is standard methodology available involving the spectrophotometric measurement of CO2 released from the conversion of organic carbon [1] Xing et al. [2] reported a method to determine VSS from such SOC data and propose a correlation between VSS and chemical oxygen demand (COD). The difference between the experimental and theoretical data was 2.1% when bacteria were formulated as C5H7NO2. Assuming their complete oxidation, this gives a COD/VSS ratio of 1.2. More recently, Lee and Pavlostathis [3] have developed a calibration curve correlating VSS with particulate organic matter retained on a 1.2 mm pore-size filter. However, the experimental data correlating these values was not reported.
ARTICLE IN PRESS J.E. Hernandez et al. / Anaerobe 12 (2006) 254–259
In spite of this, the use and correlation of SOC has not found widespread usage due to the cost, the hazardous nature of the chemicals involved (mercury, hexavalent chromium, sulphuric acid and silver), the expensive and complex nature of the analytical instruments used and the potential for errors in the methodologies (in the hightemperature method salts can promote inhibition over the catalytic surface by melting while in the persulphate method a low pH can result in incomplete oxidation). However, low cost, relatively simple technologies are now available for determining total organic carbon (TOC). The technology, which satisfies EN 1484:1997 [4], is based on the collection of CO2 that passes trough a membrane into a cell containing a colour indicator, and gives comparable results to an automatic organic carbon analyser [5]. This work compares the data and costs of VSS and SOC used in determining biomass concentration in an anaerobic digester. 2. Methods 2.1. Correlation standards A correlation between VSS and SOC was performed in triplicate for a series of dilutions of methanogenic sludge whose original biomass concentration was 3.88 g VSS L1. The sludge was diluted at a volumetric ratio of 1:10 with organic carbon free distilled water, and then diluted again at ratios of: 0.9:9.1, 1.8:8.2, 3.6:6.4, 5.4:4.6 and 7.2:2.8, respectively. A second correlation between experimental data of VSS and SOC was performed with 16 samples, by triplicate, of acidogenic sludge obtained from an acidogenic reactor [6]. The measurement of VSS was carried out in triplicate following a standard method [1]. SOC was determined using the TOC cuvette test (50–500 mg TOC/L) from Dr. Lange, Germany. The TOC was analysed for the raw sample and for the dissolved fraction filtered by using 1.2 mm pore diameter GF/C Whatman paper, and was termed dissolved organic carbon (DOC). The SOC was calculated by subtracting the DOC from the TOC. The
255
VSS and SOC values were compared with the stoichiometric SOC/VSS ratio of 0.53. This theoretical value was calculated from the elemental composition of bacteria, C5H7NO2, proposed by Hoover and Porges [7]. Thus, 5 atoms of carbon are contained per molecular weight. Therefore, for 1 g of bacteria, the amount of organic carbon is 0.53 g. Consequently, upon complete oxidation the organic carbon is converted to carbon dioxide, which is measured by the SOC method, according to the following reaction [7]: C5 H7 NO2 þ 5O2 ! 5CO2 þ NH3 þ 2H2 O: 2.2. Cost analysis The cost analysis and comparison for the SOC and VSS methods were carried out using the activity-based costing system (ABC). The ABC system refines costing systems by focusing on individual activities as the fundamental cost objects and assigns costs to products and services. It takes into account designing products, setting up machines, operating machines and distributing products [8]. Two hypothetical cases were analysed. It was assumed that both scenarios involve an initial investment to buy the necessary items for a new laboratory from commercial laboratory suppliers (Table 1). Some companies may buy on credit and others may use one single cash payment. This affects directly the cash inflows and outflows and may increase or decrease the investment in a period of time, thus altering the accounts. For case A it was assumed that 108 samples are analysed per year. This is a likely situation when monitoring the performance of two reactors sampled at weekly intervals. Storage and processing of samples was not included in the costing. The cost of labour is taken for one person for 365 days with a minimum national wage of £4.85 an hour [9] and electricity costs of £1.91 per kWh [10]. For case B it was assumed that a total of 24 samples are analysed over 4 days. This includes one set of blanks and controls and sets of six samples analysed in triplicate.
Table 1 Initial investment costs excluding VAT for a new laboratory to carry out VSS or SOC analysis VSS
SOC
Equipment
Units
Price £
Equipment
Price £
Units
Price £
Furnace (550 1C, 30 L) Air-circulating Oven (105 1C) Balance (LoD: 0.1 mg) Filtration apparatus Vacuum pump Desiccator (300 mm diameter) Pipettor (5 mL) Crucibles Tongue
1 1 1 1 1 1 1 3 1
2500 1000 1500 160 575 250 150 27 13
Spectrophotometer Thermostat Magnetic stirrer Powder dispenser Pipettor
1060 380 60 32 150
1 1 1 1 1
1060 380 60 32 150
Total initial investment
6175
1682
ARTICLE IN PRESS J.E. Hernandez et al. / Anaerobe 12 (2006) 254–259
256
2.3. Scanning electron microscopy As the matter capable of volatilization at 550 1C may be organic and inorganic, a sample was examined under the SEM microscope to observe the structure of the microbial aggregate and the space relationship between microorganisms and EPS and other materials. Specimens were fixed in Karnovsky’s fixative in 0.1 M phosphate buffer for 3 h at 4 1C, and then washed in sucrose (10%) in 0.1 M phosphate buffer. Afterwards, they were washed three times with sucrose solution 30 min intervals at 4 1C. Secondary fixation was carried out in 2% osmium tetroxide aqueous for 1 h at room temperature. Dehydration was carried out by a graded series of ethanol at room temperature. The specimens were placed in a 50/50 mixture of 100% hexamethyldisilazane and ethanol for 30 min, followed by 30 min in 100% hexamethyldisilazane and then they were allowed to air-dry overnight. After drying, they were mounted on 12.5 mm diameter stubs and attached with sticky tabs and then coated in an Edwards S150B sputter coater with approximately 25 nm of gold. Specimens were examined in a Philips PSEM 501B Scanning Electron Microscope at an accelerating voltage of 30 kV. 3. Results and discussion 3.1. VSS and SOC correlations The results of the VSS and SOC analysis of serially diluted methanogenic sludge correlate very well (R2 ¼ 0:9995, Fig. 1). The SOC/VSS ratio for the calibration curve is 0.52 (standard deviation s ¼ 0:02), which is close to the theoretical value of 0.53. The average SOC/ VSS ratio for the 16 samples from a two-stage anaerobic digester was found to be 0.52 (n ¼ 16, s ¼ 0:01), 1.89% less than the theoretical value of 0.53 (Fig. 1). The experimental data reported here are in good agreement with the theoretical stoichiometric SOC/VSS value of 0.53 showing that VSS and SOC data for
wastewater sludge are interchangeable. It should be noted however, that neither the SOC nor the VSS values truly represent microbial concentration as the fraction of extracellular polymeric substance (EPS) as well as other non-microbial organic suspended compounds such as fibrous cellulose cannot be distinguished from microbial organic carbon (Fig. 2). Therefore, this method cannot accurately assess the concentration of microbes forming aggregates. Also, other materials might not be quantified, e.g. organic acids, which might be lost owing to the stripping out of inorganic carbon. Empirical relationships between SOC and VSS might be valid for specific source water and are independently established for a particular matrix taken at different points of the treatment process [1]. However, with the determination of SOC as a unique biomass concentration parameter, such correlations will no longer be necessary. 3.2. Cost analysis of Case A The results of the cost analysis of VSS and SOC methods for case A, which assumes the analysis of 108 samples over 1 year, are shown in Table 2. After applying the ABC costing system, the yearly cost of the VSS method for 108 samples is £14 544 (Table 2) and the total cost in the first year is £20 719 with initial capital investment (from Table 1). The yearly cost of the SOC method for 108 samples is £2557 (Table 2) and a total of £4239 in the first year including initial investment (from Table 1). Therefore, the savings on total cost per year, without considering initial investment, would be £11 987 if the SOC method is used. The determination of one triplicate might take 20.7 and 3.2 h for VSS and SOC, respectively, thus the time savings are also considerable. 3.3. Cost analysis of case B The results of the cost analysis of VSS and SOC methods for case B are shown in Table 3. The cost of the VSS
-1
SOC (mg L )
150
100
50
0 0
100
200
300
-1
VSS (mg L ) Fig. 1. Linear relationship between volatile suspended solids (VSS) and suspended organic carbon (SOC) of serially diluted methanogenic biomass (&). Samples from acidogenic reactor (J).
Fig. 2. SEM micrograph of methanogenic sludge degrading phenol. A group of bacteria is located inside the dark circle and surrounded by EPS and other substances.
ARTICLE IN PRESS J.E. Hernandez et al. / Anaerobe 12 (2006) 254–259
257
Table 2 Cost analysis of VSS and SOC methods under ABC costing system for case A VSS Direct costs Direct material Pipette tip Filter paper b Direct cleaning and maintenance costs Balance Total direct costs
Units 108 324
Cost ea. £ 0.02 0.6
1
25
Indirect costs Overheads Filtration: electricity 185 W pumpc Filtratione Calibration of crucible: 103–105 1C setting upg Calibration of crucible: 103–105 1C Calibration of crucibre: 550 1C setting upg Calibration of crucibre: 550 1Ch Calibration of crucible: working timei SS determination: 103–105 1C setting upg SS determination: 103–105 1C dryingj SS determination working timek VSS determination: 550 1C setting upg VSS determination: 550 1C ignitionj VSS determination working timek Total indirect costs Total costs per year
Subtotal £ 2.16 194.4
Hours
Allocation base £
kWh
1.5 1.5 1 1 1 0.5 8 1 2 6.1 1 1 5.1
1.91d 4.85f 1.91d 1.91d 1.91d 1.91d 4.85f 1.91d 1.91d 4.85f 1.91d 1.91d 4.85f
0.093 1.4 1.4 4 4 1.4 1.4 4 4
Total £a (1)
Per sample £ (2) ¼ (1)/108
196.56
1.82
25 221.56
0.23 2.05
Total £
Per sample £
(1) 0.48 785.70 288.79 288.79 825.12 412.56 4,190.40 288.79 577.58 3,177.72 825.12 7.64 2,653.92 14,322.62 14,544.18
(2) ¼ (1)/108 0.00 7.28 2.67 2.67 7.64 3.82 38.80 2.67 5.35 29.42 7.64 0.07 24.57 132.62 134.67
SOC Direct costs Direct materials Cuvette (25 units pack)l Pipette tip Direct cleaning and maintenance costsm Total direct costs
Units
Cost ea. £
(1)
(2) ¼ (1)/108
9 432 1
95.55 0.02 25
859.95 8.64 25.00 893.59
7.96 0.08 0.23 8.27
Total £
Per sample £
(1) 2.06 2.29 1658.70 1663.05 2556.64
(2) ¼ (1)/108 0.019 0.021 15.358 15.399 23.67
Indirect costs
Hoursg
Allocation base £
KWh
Stirring of samplen Heating of sample Working timeo Total indirect costs Total costs per year
0.2 2 3.2
1.91d 1.91d 4.85f
0.06 0.6
a
This cost is for 108 samples. The total filter papers needed are calculated by multiplying 108 samples times a triplicate. c The filtration might last 90 min including washings of retained material. d This is the cost of electricity per kWh. e The operator must be present during the 90 min of filtration. f This is the minimum national wage per hour. g The oven and furnace must be previously set up at the desired temperatures, this might take 1 h. h The ignition is carried out twice for 15 min, assuming that two cycles are enough to get less than 4% difference between weight readings. i Calibrating the crucibles might take 8 working hours. The crucibles must be cleaned and identified also the operator must be in the laboratory during drying, igniting, cooling and weighing. j It is assumed that in the second cycle, TSS (1 h) and VSS (0.5 h) can be determined with less than 4% difference between weight readings. k The operator must be in the laboratory during the drying or igniting–cooling–weighing process. l Two cuvettes are needed to analyse separately DOC and TOC in every sample. m Maintenance for spectrophotometer once a year. n Four cuvettes are stirred by pairs of TOC and DOC for a total of 10 min. o The operator must be in the laboratory during the SOC determination. b
method is £326 (Table 3) for the samples and £6501 in the first year including the initial investment (from Table 1). For SOC, the total cost is £236 for the samples (Table 3)
and a first year total of £1918 (from Table 1). Thus, the SOC method results in savings of £90 on total costs without considering initial investment. The determination
ARTICLE IN PRESS 258
J.E. Hernandez et al. / Anaerobe 12 (2006) 254–259
Table 3 Cost analysis of VSS and SOC methods under ABC costing system for case B VSS Direct cost Direct Material Pipette tip Filter paper Direct cleaning and maintenance costs Balance Total direct costs
Units 24 24
Cost ea. £ 0.02 0.6
1
25
Indirect costs Overheads Filtration: electricity 185 W pumpb Filtrationd Calibration of crucible: 103–105 1C setting upf Calibration of crucible: 103–105 1Cg Calibration of crucibre: 550 1C setting upf Calibration of crucibre: 550 1Ch Calibration of crucible: working timei SS determination: 103–105 1C setting upg SS determination: 103–105 1C dryingj SS determination working timek VSS determination: 550 1C setting upf VSS determination: 550 1C ignitionj VSS determination working timek Total indirect costs Total costs per year
Subtotal £ 0.48 14.4
Hours
Allocation base £
kWh
0.5 0.5 1 1 1 0.5 16 1 2 12.1 1 1 10.1
1.91c 4.85e 1.91c 1.91c 1.91c 1.91c 4.85e 1.91c 1.91c 4.85e 1.91c 1.91c 4.85e
0.093 1.4 1.4 4 4 1.4 1.4 4 4
Total £a (1)
Per sample £ (2) ¼ (1)/24
14.88
0.620
25 39.88
0.23 0.85
Total £
Per sample £
(1) 2.13 58.20 2.67 2.67 7.64 3.82 77.60 2.67 5.35 58.85 7.64 7.64 49.15 286.03 325.91
(2) ¼ (1)/24 0.09 2.43 0.11 0.11 0.32 0.16 3.23 0.11 0.22 2.45 0.32 0.32 2.05 11.92 12.77
SOC Direct costs Direct materials Cuvette (25 units pack)l Pipette tip Direct cleaning and maintenance costsm Total direct costs Indirect costs Overheads Stirring of samplen Heating of sample Working timeo Total indirect costs Total costs
Units
Cost ea. £
(1)
(2) ¼ (1)/24
2 96
95.55 0.02
191.10 1.92 25 218.02
7.96 0.08 1.04 9.08
Total £
Per sample
(1) 0.23 2.29
(2) ¼ (1)/24 0.01 0.10
17.88 235.90
0.11 9.19
Hours
Allocation base £
KWh
0.1 2 3.2
1.91c 1.91c 4.85e
0.06 0.6 15.36
a
This cost is for 24 samples. Filtration of one triplicate might last 30 min per crucible including washing of retained solids; this is performed for 24 samples. c This is the cost of electricity per kWh. d The operator must be present during 30 min of filtration process per sample. e This is the minimum national wage per hour. f The oven and furnace must be previously set up at the desired temperatures, this might take 1 h. g It is assume that both oven and furnace have the capacity to allocate 24 samples in one run for 1 h. h The ignition is carried out twice for 15 min, assuming that two cycles are enough to get less than 4% difference between weight readings. i The operator must work at least two periods of 8 h to calibrate 24 crucibles and must be present during drying, igniting, cooling and weighing. j It is assumed that in the second cycle TSS (1 h) and VSS (0.5 h) can be determined with less than 4% difference between weight readings. k The determinations are carried out in two sessions of 6.05 h and 5.05 for TSS and VSS respectively; this is also under supervision of operator. l Two cuvettes are needed to analyse DOC and TOC in every sample. m Maintenance for spectrophotometer once a year. n Two cuvettes are stirred simultaneously to determine TOC and DOC, this is for preventing CO2 diffusion into the sample. o The operator must be present during the SOC determination. b
of biomass in 24 samples might take 38.7 and 3.2 h for VSS and SOC, respectively, again giving considerable time saving.
This shows that the determination of SOC of a batch of samples is slightly cheaper than applying the VSS method. This can be explained as in this case, the working time of the
ARTICLE IN PRESS J.E. Hernandez et al. / Anaerobe 12 (2006) 254–259
operator can be efficiently utilized to process a big number of samples at the same time. This situation does not occur when a continuous process is monitored regularly. 4. Conclusion The results show that the SOC/VSS ratio from both a serially diluted methanogenic sludge (0.52, s ¼ 0:02) and sludge samples taken from an acidogenic reactor (0.52, n ¼ 16, s ¼ 0:01) are close to the theoretical value of 0.53 for complete oxidation of bacteria formulated as C5H7NO2. This suggests that such a formula can be applied for active anaerobic biomass as well as pure cultures. For both scenarios, cost analysis shows the SOC method to be the lower cost and time-saving option. The savings are £11 987 for case A and £90 for case B. The initial investment required in a new laboratory does not affect the outcome in either case. These experimental and economical results suggest that the traditional VSS method can be replaced by the more economic and time-saving SOC method for the measurement of microbial biomass. Acknowledgements We would like to thank the sponsorship of CONACyT and help of N. Coutelle and N. Garcia.
259
References [1] Clesceri LS, Greenberg AE, Eaton AD. Standard methods for the examination of water and wastewater, 8th ed. Washington: American Public Health Association, American Water Works Association; 1998. [2] Xing CH, Wen XH, Qian Y, Tardieu E. Microfiltration–membranecoupled bioreactor for urban wastewater reclamation. Desalination 2001;141:63–73. [3] Lee YH, Pavlostathis SG. Decolorization and toxicity of reactive anthraquinone textile dyes under methanogenic conditions. Water Res 2004;38:1838–52. [4] CEN. EN 1484:1997: Water analysis. Guidelines for the determination of total organic carbon (TOC) and dissolved organic carbon (DOC). European Standarization Committee; 1997. [5] Kraatz C, Wochnik P. Dr. Lange TOC cuvette test conforms to European standard. In: Application report no. 60, Du¨sseldorf: Dr. Bruno Lange GmbH & Co; 1998. [6] Hernandez JE, Edyvean RG. Anaerobic treatment of phenol in a two-stage anaerobic reactor. In: IchemE, editor, Seventh world congress of chemical engineering. SECC, Glasgow, Scotland. Rugby, UK: IChemE; 2005. [7] Tchobanoglous G, Burton FL, Stensel HD. Wastewater engineering: treatment and reuse, 4th ed. New York: McGraw-Hill; 2003. [8] Horngren C, Bhimani A, Datar SM, Foster G. Management and cost accounting. 2nd ed. Essex: Prentice-Hall; 2002. [9] Anon. National minimum wage. UK: Department of Trade and Industry; 2004. [10] Anon. Electricity Tariff Finder. /http://www.powergen.co.uk/Pub/ Sme/A/Ui/Quotation/ElecQuoteStart.aspxS; 2005.