This article was downloaded by: [University of Colorado - Health Science Library] On: 10 April 2015, At: 11:15 Publisher: Taylor & Francis Informa Ltd Registered in England and Wales Registered Number: 1072954 Registered office: Mortimer House, 37-41 Mortimer Street, London W1T 3JH, UK
Journal of Environmental Science and Health, Part A: Toxic/Hazardous Substances and Environmental Engineering Publication details, including instructions for authors and subscription information: http://www.tandfonline.com/loi/lesa20
Mathematical modeling of COD removal via the combined treatment of domestic wastewater and landfill leachate based on the PACT process a
b
b
Ángel S. Fernández Bou , Alexandre Lioi Nascentes , Barbara Costa Pereira , Leonardo b
c
d
Duarte Batista Da Silva , João Alberto Ferreira & Juacyara Carbonelli Campos a
Biosystems Engineering, Fluminense Federal University, Niterói (RJ), Brazil
b
Click for updates
Institute of Technology, Department of Engineering, Federal Rural University of Rio de Janeiro, Seropédica (RJ), Brazil c
Center of Technology and Science, Department of Sanitary and Environmental Engineering, Rio de Janeiro State University, Rio de Janeiro (RJ), Brazil d
School of Chemistry, Department of Inorganic Processes, Federal University of Rio de Janeiro, Rio de Janeiro (RJ), Brazil Published online: 27 Feb 2015.
To cite this article: Ángel S. Fernández Bou, Alexandre Lioi Nascentes, Barbara Costa Pereira, Leonardo Duarte Batista Da Silva, João Alberto Ferreira & Juacyara Carbonelli Campos (2015) Mathematical modeling of COD removal via the combined treatment of domestic wastewater and landfill leachate based on the PACT process, Journal of Environmental Science and Health, Part A: Toxic/Hazardous Substances and Environmental Engineering, 50:4, 378-384, DOI: 10.1080/10934529.2015.987533 To link to this article: http://dx.doi.org/10.1080/10934529.2015.987533
PLEASE SCROLL DOWN FOR ARTICLE Taylor & Francis makes every effort to ensure the accuracy of all the information (the “Content”) contained in the publications on our platform. However, Taylor & Francis, our agents, and our licensors make no representations or warranties whatsoever as to the accuracy, completeness, or suitability for any purpose of the Content. Any opinions and views expressed in this publication are the opinions and views of the authors, and are not the views of or endorsed by Taylor & Francis. The accuracy of the Content should not be relied upon and should be independently verified with primary sources of information. Taylor and Francis shall not be liable for any losses, actions, claims, proceedings, demands, costs, expenses, damages, and other liabilities whatsoever or howsoever caused arising directly or indirectly in connection with, in relation to or arising out of the use of the Content. This article may be used for research, teaching, and private study purposes. Any substantial or systematic reproduction, redistribution, reselling, loan, sub-licensing, systematic supply, or distribution in any form to anyone is expressly forbidden. Terms & Conditions of access and use can be found at http:// www.tandfonline.com/page/terms-and-conditions
Journal of Environmental Science and Health, Part A (2015) 50, 378–384 Copyright © Taylor & Francis Group, LLC ISSN: 1093-4529 (Print); 1532-4117 (Online) DOI: 10.1080/10934529.2015.987533
Downloaded by [University of Colorado - Health Science Library] at 11:15 10 April 2015
Mathematical modeling of COD removal via the combined treatment of domestic wastewater and landfill leachate based on the PACT process ANGEL S. FERNANDEZ BOU1, ALEXANDRE LIOI NASCENTES2, BARBARA COSTA PEREIRA2, e ALBERTO FERREIRA3 LEONARDO DUARTE BATISTA DA SILVA2, JOAO 4 and JUACYARA CARBONELLI CAMPOS 1
Biosystems Engineering, Fluminense Federal University, Niter oi (RJ), Brazil Institute of Technology, Department of Engineering, Federal Rural University of Rio de Janeiro, Serop edica (RJ), Brazil 3 Center of Technology and Science, Department of Sanitary and Environmental Engineering, Rio de Janeiro State University, Rio de Janeiro (RJ), Brazil 4 School of Chemistry, Department of Inorganic Processes, Federal University of Rio de Janeiro, Rio de Janeiro (RJ), Brazil 2
The experiments performed in this study consisted of 16 batch reactors fed different mixtures of landfill leachate combined with synthetic wastewater treated using the Powdered Activated Carbon Treatment (PACT) process. The objective was to measure the COD mass removal per liter each day for each reactor using two models: the first model combined the variables PAC concentration (0 g¢L¡1, 2 g¢L¡1, 4 g¢L¡1, and 6 g¢L¡1) and leachate rate in the wastewater (0%, 2%, 5%, and 10%), and the second model combined the PAC concentration and the influent COD. The Response Surface Methodology with Central Composite Design was used to describe the response surface of both models considered in this study. Domestic wastewater was produced under controlled conditions in the laboratory where the experiments were performed. The results indicated that the PAC effect was null when the influent did not contain leachate; however, as the concentration of leachate applied to the mixture was increased, the addition of a higher PAC concentration resulted in a better COD mass removal in the reactors. The adjusted R2 values of the two models were greater than 0.95, and the predicted R2 values were greater than 0.93. The models may be useful for wastewater treatment companies to calculate PAC requirements in order to meet COD mass removal objectives in combined treatment. Keywords: Combined treatment of landfill leachate and sewage, powdered activated carbon treatment, mathematical model of COD removal from leachate, COD mass removal with PACT treatment.
Introduction In recent years, the management of municipal solid waste in Brazil has improved, and more than 58% of wastes are disposed in sanitary landfills.[1] This improvement occurred due to recent sanitary and environmental laws, which require all of the municipalities to treat their solid waste properly. However, in 2011, there were still 2,810 cities and towns producing solid wastes which were disposed in 2,906 dumps[2] Nevertheless, due to the new solid waste law, more sanitary landfills are being created, Address correspondence to Angel S. Fernandez Bou, Rua Passo da Patria, 156, bl D, sl 236, S~ ao Domingos, Campus da Praia Vermelha, Niter oi (RJ) 24210-240, Brazil; E-mail: [email protected] Received July 30, 2014. Color versions of one or more of the figures in the article can be found online at www.tandfonline.com/lesa.
thereby increasing the control over traditional environmental problems. Moreover, Brazil has the highest values of long-term average precipitation in the world, with 14,995 km3 of rainfall per year according to AQUASTAT, the FAO’s global water information system. The country holds approximately 12.8% of all freshwater in the world.[3] Large urban nuclei, such as the states of S~ao Paulo or Rio de Janeiro, where the sanitary landfills treat a notorious percentage of all Brazilian solid waste, have an average precipitation in the range of 1,400 to 1,500 mm per year.[4] Landfill leachate generation depends directly on the rainfall amounts. Thus, the new law for controlling solid waste has created the following unintended consequence: much more leachate from sanitary landfills must be treated. As a result, landfill leachate treatment has become one of the most relevant environmental challenges in Brazil. The characteristics of the sanitary landfill leachate depend on different factors, such as rainfall, temperature,
379
Mathematical modeling of COD removal based on the PACT process Table 1. Composition of the synthetic domestic wastewater used in this study. Concentration (mg¢L¡1)
Components
Downloaded by [University of Colorado - Health Science Library] at 11:15 10 April 2015
Peptone from casein Meat extract Urea Monopotassium phosphate Sodium chloride Calcium chloride dihydrate Magnesium sulfate heptahydrate
Table 2. Characteristics of Carbomafra powdered activated carbon (Type: 118 CB AS n 170). Parameters
Result 726.68 m2¢g¡1 560.59 m2¢g¡1 166.08 m2¢g¡1 0.266 cm3 ¢g¡1 25.6 A
BET area Micropore area External surface area Micropore volume Micropore size
360 250 100 26 14 8 4
Source: Campos et al.[19]
Source: Adaptation from the procedures reported by Holler and Tr€ osch[17] to those presented by Von Sperling.[18]
run-off, infiltration, evaporation, air humidity and type of waste.[5] Water is the main component, with a high concentration of dissolved organic matter (normally expressed in COD or BOD), inorganic macro-components, heavy metals and organic xenobiotic components.[6] Several studies indicate the presence of high concentrations of heavy metals in leachate.[7] However, according to some researchers, the concentration of heavy metals is not an issue of concern if the leachate does not contaminate the soil.[8] Giordano analyzed the Gramacho Landfill leachate, located in Rio de Janeiro, Brazil, and found that the heavy metal concentrations were significantly lower than the levels allowed by law.[9] These low concentrations of heavy metals were a consequence of the adsorption and precipitation of metallic ions as hydroxides and sulfides caused by the alkaline and anaerobic environment inside the landfill.[10,11] The landfill age and waste type can result in differences in heavy metal concentrations between countries; e.g., the iron concentration in an old landfill in Brazil was 5.5 mg¢L¡1, whereas that in France was 26 mg¢L¡1, and that in South Korea was 76 mg¢L¡1 (in a medium age landfill in the latter); in addition, the copper concentrations were found to be 0.08 mg¢L¡1 in Brazil, 0.04 mg¢L¡1 in France and 0.78 mg¢L¡1 in South Korea.[5] This result reinforces the variable nature of the leachate composition.
Those values of Brazilian leachates are within the limits allowed by the Brazilian law regarding the discharge of heavy metals into natural water bodies.[12] Sanitary landfill leachates are commonly treated through biological processes, such as activated sludge or stabilization ponds, and using membranes, although other processes can be found in the literature, such as adsorption, coagulation and flocculation. Nevertheless, because the characteristics of the leachate are notably variable, there is not a clear consensus regarding the best treatment system.[13] The Powdered Activated Carbon Treatment (PACT) process was developed by DuPont in the 1970s.[14] The process is a sludge aerobic treatment using powdered activated carbon. The process consists of aerobic digestion of the organic matter, which is synergistic with the use of PAC, thereby improving the results of the system.[15] The objective of this study was to model the COD mass removal using Response Surface Methodology (RSM) based on the results of experiments that studied the combined treatment of wastewater mixed with sanitary landfill leachate via the PACT process, using different settings of leachate mixture and PAC concentration.
Materials and methods The experiments were performed at the Environmental Monitoring Laboratory I – Water and Effluents, in the Federal Rural University of Rio de Janeiro. The analysis
Table 3. Structure of the experimental batteries. Battery I
Battery II
From 02/12/2013 to 03/01/2014
From 15/01/2014 to 07/02/2014
Leachate Percentage in the Mixture
PAC concentration
0 2 4 6
g¢L¡1 g¢L¡1 g¢L¡1 g¢L¡1
0%
2%
5%
10%
R1I R2I R3I R4I
R5I R6I R7I R8I
R1II R2II R3II R4II
R5II R6II R7II R8II
380
Fernandez Bou et al.
Table 4. Leachate addition to the wastewater mixture recipient. Leachate mixture
Leachate Volume Added to 2.5 L of Wastewater
0% 2% 5% 10%
Downloaded by [University of Colorado - Health Science Library] at 11:15 10 April 2015
Table 5. PAC reposition in each reactor.
0 mL 51 mL 132 mL 278 mL
followed the APHA recommendations.[16] Method 5220-D was used to obtain the COD values from the influents and effluents using a Hach 3900 spectrophotometer.
Sampling The sludge seed was grown in the laboratory and its initial concentration was 5,780 mg¢L¡1. The leachate sample originated from the sanitary landfill Dois Arcos, located in the city of S~ ao Pedro da Aldeia (RJ). The leachate was collected in two 10-L plastic receptacles. The synthetic wastewater was produced at the laboratory under controlled conditions, based on the recommendations from Holler and Tr€ osch,[17] with some alterations to adapt the influent to the physicochemical characteristics described by Von Sperling for Brazilian domestic wastewater.[18] The recipe of the synthetic domestic wastewater is presented in Table 1. The powdered activated carbon used in the experiment was Carbomafra Type: 118 CB AS no. 170. The characteristics of this carbon were studied by Campos et al.[19] at the Federal University of Rio de Janeiro and are presented in Table 2.
Experimental structure The experiment was performed in two batteries, each of which had eight reactors, as described in Table 3. There were 16 different configurations of the leachate mixture and the PAC concentration. The first reactor of the first battery (R1I) was the control and thus did not contain any leachate and PAC. The first battery was run with lower leachate concentrations (0% and 2%), compared with those used in the second battery (5% and 10%), whereas the PAC addition was maintained at the same levels in both batteries (0 g¢L¡1, 2 g¢L¡1, 4 g¢L¡1 and 6 g¢L¡1).
Reactor
[PAC]
R1 and R5 R2 and R6 R3 and R7 R4 and R8
0 2 4 6
g¢L¡1 g¢L¡1 g¢L¡1 g¢L¡1
PAC Reposition 0.00 0.07 0.14 0.21
g of PAC g of PAC g of PAC g of PAC
Experimental preparation The sludge age was defined previously as 28 days, i.e., 3.57% of the sludge was removed each day, or 35 mL of the liquor was removed from the reactors. The dissolved oxygen (DO) level was maintained at a level greater than 2 mg¢L¡1 using eight air pumps connected to two uninterruptible power supplies. The hydraulic retention time, HRT, was 40 h, and this time was maintained by removing 60% of the treated effluent every 24 h. The reactors consisted of eight graduated glass cylinders. The working volume of each of the reactors was 1,000 mL. The sludge collected at the treatment plant was carried directly to the laboratory. After sedimentation in 1,850 mL of solution, the sludge was homogenized and divided into eight volumes of 200 mL, using the remaining for analysis. Because the initial concentration of the sludge was 5,780 mg¢L¡1, each reactor received 2,499.45 mg of sludge. The sludge was fed synthetic wastewater over a period of 2 weeks for acclimatization before the application of different mixtures of leachate and PAC was initiated.
System operation For each group of four reactors, 2.5 L of synthetic wastewater were produced each day, and different volumes of sanitary landfill leachate were added to the produced wastewater as detailed in Table 4. To maintain the sludge age, 35 mL of liquor were removed each day, resulting in PAC removal from each reactor (except for the ones without PAC). The PAC reposition was performed according to Table 5. Aeration was operated 23 h per day using conventional aquarium air pumps (one for each reactor) with the goal of maintaining the DO at a level greater than 2 mg¢L¡1 to ensure that it was not a limiting factor.[20,21] Before stopping the aeration, the pH, temperature and DO were measured, and 35 mL of liquor were removed from each
Table 6. COD removal per day and per liter. Leachate % Influent COD
0%
2%
5%
10%
530.5 mg¢L¡1
794.5 mg¢L¡1
1262.8 mg¢L¡1
2001.1 mg¢L¡1
381
Mathematical modeling of COD removal based on the PACT process Table 7. COD removal efficiency. Leachate Rate ! # PAC
Downloaded by [University of Colorado - Health Science Library] at 11:15 10 April 2015
0 2 4 6
g¢L¡1 g¢L¡1 g¢L¡1 g¢L¡1
Table 8. COD mass removal per day and per liter.
0%
2%
5%
10%
90.3% 92.4% 89.1% 83.4%
62.2% 76.3% 74.6% 77.2%
54.3% 70.7% 70.5% 73.9%
47.9% 53.5% 55.0% 59.2%
reactor. The aeration was then stopped, and when the sludge settled at the bottom of the reactors, normally after 45 min, the volume from the upper part was removed from each reactor up to the 400-mL mark, taking care to avoid removing any solid flakes from the bottom. Then 50 mL of treated water from each reactor were collected to perform the COD analysis. Subsequently, PAC reposition was performed, and 600 mL of the wastewater and leachate mixture were added according to the settings of each reactor. Next, 50 mL of each mixture were collected for the COD analysis and then we reconnected the aeration system measuring the pH of the reactors and set the pH to greater than 8.00 with KOH, if necessary.
Leachate Rate ! # PAC 0 2 4 6
g¢L¡1 g¢L¡1 g¢L¡1 g¢L¡1
287.5 294.1 283.7 265.4
Response Surface Methodology (RSM) and Central Composite Design (CCD) were used to understand the COD mass removal. Two models were designed: the first used the PAC and the leachate rate, and the second used the PAC and the COD as independent variables. In both models, the response was the mean COD mass removal obtained for each of the 16 reactors. The equation
mg mg mg mg
2% 296.4 363.9 355.4 368.2
mg mg mg mg
5% 411.5 536.0 534.5 559.8
mg mg mg mg
10% 575.5 642.5 660.9 710.9
mg mg mg mg
intended to describe the behavior of this treatment was a second-order polynomial equation with two independent variables (Eq. 1).[22-24] The efficiency of the treatment was calculated according to Eq. 2. The COD mass removal per day was calculated according to Eq. 3. Z D a1 X2 C a2 Y2 C a3 XY C a4 X C a5 Y C a0 C e
(1)
where Z is the response, X and Y are the relevant independent variables, a1, a2, a3, a4, a5 and a0 are coefficients of the model, and e represents the error. COD Removal Efficiency .%/ D
Data treatment and modeling
0%
COD influent ¡ COD effluent 100 COD influent
(2)
where COD influent is the COD (in mg¢L¡1) of the mixture of wastewater with leachate, and COD effluent is the COD (in mg¢L¡1) of the treated effluent. COD mass removal per day and reactor D .COD influent ¡ COD effluent/ 0:6
L (3) reactor day
where COD influent is the COD (in mg¢L¡1) of the mixture of wastewater with leachate, COD effluent is the COD
Fig. 1. Response surface of model 1: Leachate rate and PAC as the independent variables.
Fig. 2. Plot of the predicted versus actual data for COD mass removal using model 1.
382
Fernandez Bou et al.
Table 9. ANOVA analysis of the COD mass removal response. Response
P-value Prob > F
Std. Dev.
C.V.%
PRESS
R2
Adj. R2
Pred. R2
Adeq. Precision
Model 1 Model 2
< 0.0001 < 0.0001
32.65 30.59
7.31 6.85
23507 21265
0.9667 0.9708
0.9546 0.9602
0.9333 0.9397
25.429 27.033
Downloaded by [University of Colorado - Health Science Library] at 11:15 10 April 2015
Std. Dev. is the standard deviation; C.V.% is the coefficient of variance; PRESS is the predicted residual error sum of squares; R2 is the coefficient of determination; Adj. R2 is the adjusted coefficient of determination; Pred. R2 is the predicted coefficient of determination; Adeq. Precision is the adequate precision.
(in mg¢L¡1) of the treated effluent, and the volume treated per day in each reactor is 0.6 L. The quality of the data fitting to the polynomial models was represented by the coefficient of determination, R2, and the statistical significance was verified using Fisher’s F-test. The non-significant terms were discarded (when “Prob > F” was greater than 0.1). ANOVA was used to completely analyze the results using the Design Expert software.
Results and discussion COD removal efficiency Table 6 shows the COD influent values for the different concentrations of leachate added. The lowest value of 530.5 mg¢L¡1 corresponds to the synthetic wastewater without leachate addition, which was used as the influent for reactors R1I to R4I in the first battery. Table 7 shows the treatment efficiency results. The efficiency of the treatment was greater than 80% in the four reactors without leachate (R1I to R4I) and greater than 90% in R1I and R2I, where the addition of PAC did not improve the treatment. The results presented in Table 7 indicate that an increase in the leachate concentration decreased the efficiency of the COD removal in all of the reactors. The results also indicate that the presence of PAC increased the efficiency in the reactors with leachate. The best result was obtained in reactor R8I, which contained 2% of leachate and 6 g¢L¡1 PAC. The efficiency results obtained for the mixture containing 10% leachate were the worst of the set of experiments for all reactors. None of the reactors achieved a 60% efficiency rate.
COD mass removal models The COD mass removals are presented in Table 8. The results indicate that higher influent COD was associated with greater COD mass removal. At higher PAC concentrations, a greater COD mass tended to be removed with some exceptions: for 0% leachate, the best behavior was obtained for 2 g¢L¡1 and 0 g¢L¡1 PAC; when the leachate concentration was 2%, the best behavior was obtained for 6 g¢L¡1 PAC, but the results were very similar to the COD
mass removal results obtained with 2 g¢L¡1 and 4 g¢L¡1 PAC; when the leachate concentrations were 5% and 10%, the behaviors of the reactors with 2 g¢L¡1 and 4 g¢L¡1 PAC were very similar, but the best COD mass removal was obtained with the reactors with 6 g¢L¡1 PAC. Model 1: PAC and leachate rate as the independent variables. Model 1 describes the behavior of the system when the independent variables used were the leachate rate and the PAC concentration in the reactor. The recommended boundaries for the application of this model are the following: PAC concentration from 0 g¢L¡1 to 6 g¢L¡1 and leachate rate from 0 to 0.1 (0% to 10%). The model is represented by Eq. 4. Figure 1 presents the response surface of this model, and Figure 2 shows the predicted versus actual data obtained from the Design Expert statistical analysis. CODe D 263:61 C 2:61 ¢PAC C 4476:70 ¢Leachate C 232:17 ¢PAC ¢Leachate ¡ 13848:76 ¢Leachate2
(4)
where CODe is the COD mass removal (in mg¢L¡1), PAC is the PAC concentration (in g¢L¡1), and Leachate is the leachate rate (dimensionless). The interval of application of this model is 0 g¢L¡1 PAC 6 g¢L¡1 and 0 Leachate 0.1. “Prob > F” < 0.0001 means that there is a probability lower than 0.01% that this fitting occurs due to noise (Table 9). A third-order polynomial equation was also tested: although the adjusted R2 was slightly better, the PAC term was not significant according to Fisher’s F-test. Thus, the third-order polynomial equation was discarded. For the first second-order polynomial equation calculated, the PAC2 term was not significant and was thus discarded. Removing this term improved the adjusted R2, and the rest of the terms remained significant. Model 2: PAC and COD as independent variables. Model 2 describes the behavior of the system when the independent variables used were the COD of the mixture and the PAC concentration in the reactor. Because the sanitary leachate can be highly variable in composition, this model aims to be more accurate when it is possible to determine the COD of the leachate and the wastewater mixture. The
Downloaded by [University of Colorado - Health Science Library] at 11:15 10 April 2015
Mathematical modeling of COD removal based on the PACT process
383
where CODe is the COD mass removal (in mg¢L¡1), PAC is the PAC concentration (in g¢L¡1) and CODi is the influent COD of the leachate and wastewater mixture (in mg¢L¡1). The interval of application of this model is 0 g¢L¡1 PAC 6 g¢L¡1 and 530 mg¢L¡1 CODi 2,000 mg¢L¡1. “Prob > F” < 0.0001 means that there is a probability lower than 0.01% that this fitting occurs due to noise (Table 9). A third-order polynomial equation was also tested; the adjusted R2 was slightly worse, and the PAC term was not significant according to Fisher’s F-test. Similar to model 1, the PAC2 term for the first second-order polynomial equation calculated, was not significant and was thus discarded. Removing this term improved the adjusted R2, and the rest of the terms remained significant.
Fig. 3. Response surface model 2: influent COD and PAC as the independent variables.
recommended boundaries for the application of this model are the following: PAC concentration from 0 g¢L¡1 to 6 g¢L¡1 and COD of the influent mixture from 530 mg¢L¡1 to 2,000 mg¢L¡1. The model is represented by Eq. 5. Figure 3 presents the response surface of model 2, and Figure 4 shows the predicted versus actual data obtained through the statistical analysis using the Design Expert software. CODe D 76:54 ¡ 5:38 ¢PAC C 0:39 ¢CODi C 0:0156 ¢PAC ¢CODi ¡ 7:262E ¡ 05 ¢CODi2
Conclusions The addition of a higher PAC concentration resulted in a better COD mass removal in the reactors. However, when the leachate rate was zero (no leachate in the mixture), the application of PAC did not improve the COD mass removal. The experimental results indicated that a greater leachate rate was associated with a greater COD mass removal, despite a decrease in the efficiency (inside the intervals of application). Because the models correctly represented the efficiency of the system inside the studied intervals, they may be useful for wastewater treatment companies to calculate the PAC requirements based on the leachate ratio or the COD of the mixture they are going to treat to determine the final COD mass removal.
(5)
Acknowledgments The authors express their gratitude to the interns from the Environmental Monitoring Laboratory I – Water and Effluents at Federal Rural University of Rio de Janeiro, and to the sanitary landfill Dois Arcos.
Funding The authors gratefully acknowledge the financial support given by CAPES.
References
Fig. 4. Plot of the predicted versus actual data for COD mass removal using model 2.
[1] Associa¸c ~ ao Brasileira de Empresas de Limpeza P ublica e Resíduos Especiais (Special Waste and Public Cleaning Brazilian Companies’ Association). Panorama dos Resíduos S olidos no Brasil 2012 (Solid Waste Panorama in Brazil 2012); S~ ao Paulo, 2013, in Portuguese.
Downloaded by [University of Colorado - Health Science Library] at 11:15 10 April 2015
384 [2] Instituto de Pesquisa Econ^ omica Aplicada (Applied Economics Research Institute). Plano Nacional de Resíduos S olidos: diagnostico dos resíduos urbanos, agrosilvopastoris e a quest~ ao dos catadores (Solid Waste National Plan), IPEA official reports no. 145; IPEA: Brasília, 2012. Available at: http://www.ipea.gov.br/ portal/images/stories/PDFs/comunicado/120425_comunicadoi pea0145.pdf (accessed May 2014), in Portuguese. [3] AQUASTAT, FAO. Freshwater Availability - Precipitation and Internal Renewable Water Resources (IRWR); AQUASTAT online database. Available at: http://www.fao.org/nr/water/aqua stat/data/query/index.html?lang=en (accessed July 2014). [4] Ramos, A.M.; Dos Santos, L.A.R.; Fortes, L.T.G. Normais climatologicas do Brasil 1961–1990: edi¸c ~ ao revista e ampliada; Instituto Nacional de Meteorologia: Brasília, 2009, in Portuguese. [5] Renou, S.; Givaudan, J.G.; Poulain, S.; Dirassouyan, F.; Moulin, P. Landfill leachate treatment: Review and opportunity. J. Hazard. Mater. 2008, 150(3), 468–493. [6] Christensen, T.H.; Kjeldsen, P.; Bjerg, P.L.; Jensen, D.L.; Christensen, J.B.; Baun, A.; Albrechtsen, H-J; Heron, G. Biogeochemistry of landfill leachate plumes. Appl. Geochem. 2001, 16(7), 659– 718. [7] Al-Muzaini, S.; Thiem, L. A comparison of leachate treatment between Rhode Island and Kuwait. J. Environ. Sci. Health A 2004, 39(2), 507–514. [8] Kang, K.-H.; H.S. Shin; Park, H. Characterization of humic substances present in landfill leachates with different landfill ages and its implications. Water Res. 2002, 36(16), 4023–4032. [9] Giordano, G. An alise e formula¸c ~ ao de processos para tratamento dos chorumes gerados em aterros de resíduos s olidos urbanos; Material Science and Metalurgy Department, Pontifical Catholic University: Rio de Janeiro, 2003, Doctoral thesis in Portuguese. [10] Castrill on, L.; Fern andez-Nava, Y.; Ulmanu, M.; Anger, I.; n, E. Physico-chemical and biological treatment of MSW no Mara~ landfill leachate. Waste Manage. 2010, 30(2), 228–235. [11] Tatsi, A.; Zouboulis, A. A field investigation of the quantity and quality of leachate from a municipal solid waste landfill in a Mediterranean climate (Thessaloniki, Greece). Adv. Environ. Res. 2002, 6(3), 207–219. [12] BRASIL. Law no. 12305; Brasília, 2010, in Portuguese.
Fernandez Bou et al. [13] Campos, J.C.; Machado, B.D.S.; Blonski, M.E.D.; Bila, D.M..; Ferreira, J.A. Evaluation of coagulation/flocculation process in the landfill leachate treatment at the Municipal Wastewater Treat ment Plant. Rev. Ambient. Agua. 2013, 8(3), 43–53. [14] Flynn, B.P.; Stadnik, J.G. Start-up of a powdered activated carbon-activated sludge treatment system. J. Water Pollut. Con. F. 1979, 51(2), 358–369. [15] Schultz, J.R.; Keinath, T.M. Powdered activated carbon treatment process mechanisms. J. Water Pollut. Con. F. 1984, 56(2), 143–151. [16] APHA. Standard Methods for the Examination of Water and Wastewater, 21st Ed.; APHA/AWWA/WEF: Washington, DC, 2005. [17] Holler, S.; Tr€ osch, W. Treatment of urban wastewater in a membrane bioreactor at high organic loading rates. J. Biotechnol. 2001, 92(2), 95–101. [18] Von Sperling, M. Introdu¸c ~ ao a qualidade das aguas e ao tratamento de esgotos, 2nd Ed.; Editora UFMG: Belo Horizonte, 1996; Vol. 1, 452 pp, in Portuguese. [19] Campos, J.C.; Machado, C.R.A.; Couto, J.M.S.; Florido, P.L.; Cerqueira, A.C.F.P.; Santiago, V.M.J. Evaluation of an activated sludge process combined with powdered activated carbon for the treatment of oil refinery wastewater. J. Int. Environ. Appl. Sci. 2014, 9(1), 24–36. [20] Metcalf and Eddy Inc. Wastewater engineering, treatment, disposal and reuse, 4th Ed.; McGraw-Hill: New York, 2003; 1848 pp. [21] Eckenfelder, W.W. Industrial Water Pollution Control; McGrawHill: Singapore, 1989. [22] Mojiri, A., Aziz, H.A.; Zaman, N.Q., Aziz, S.Q.; Zahed, M.A. Powdered ZELIAC augmented sequencing batch reactors (SBR) process for co-treatment of landfill leachate and domestic wastewater. J. Environ. Manage. 2014, 139, 1–14. [23] Aghamohammadi, N.; Aziz, H.B.A.; Isa, M.H.; Zinatizadeh, A.A. Powdered activated carbon augmented activated sludge process for treatment of semi-aerobic landfill leachate using response surface methodology. Bioresource Technol. 2007, 98(18), 3570–3578. [24] Aziz, S.Q.; Aziz, H.A.; Yusoff, M.S.; Bashir, M.J. Landfill leachate treatment using powdered activated carbon augmented sequencing batch reactor (SBR) process: Optimization by response surface methodology. J. Hazard. Mater. 2011, 189(1), 404–413.
Journal of Environmental Science and Health, Part A: Toxic/Hazardous Substances and Environmental Engineering Publication details, including instructions for authors and subscription information: http://www.tandfonline.com/loi/lesa20
Mathematical modeling of COD removal via the combined treatment of domestic wastewater and landfill leachate based on the PACT process a
b
b
Ángel S. Fernández Bou , Alexandre Lioi Nascentes , Barbara Costa Pereira , Leonardo b
c
d
Duarte Batista Da Silva , João Alberto Ferreira & Juacyara Carbonelli Campos a
Biosystems Engineering, Fluminense Federal University, Niterói (RJ), Brazil
b
Click for updates
Institute of Technology, Department of Engineering, Federal Rural University of Rio de Janeiro, Seropédica (RJ), Brazil c
Center of Technology and Science, Department of Sanitary and Environmental Engineering, Rio de Janeiro State University, Rio de Janeiro (RJ), Brazil d
School of Chemistry, Department of Inorganic Processes, Federal University of Rio de Janeiro, Rio de Janeiro (RJ), Brazil Published online: 27 Feb 2015.
To cite this article: Ángel S. Fernández Bou, Alexandre Lioi Nascentes, Barbara Costa Pereira, Leonardo Duarte Batista Da Silva, João Alberto Ferreira & Juacyara Carbonelli Campos (2015) Mathematical modeling of COD removal via the combined treatment of domestic wastewater and landfill leachate based on the PACT process, Journal of Environmental Science and Health, Part A: Toxic/Hazardous Substances and Environmental Engineering, 50:4, 378-384, DOI: 10.1080/10934529.2015.987533 To link to this article: http://dx.doi.org/10.1080/10934529.2015.987533
PLEASE SCROLL DOWN FOR ARTICLE Taylor & Francis makes every effort to ensure the accuracy of all the information (the “Content”) contained in the publications on our platform. However, Taylor & Francis, our agents, and our licensors make no representations or warranties whatsoever as to the accuracy, completeness, or suitability for any purpose of the Content. Any opinions and views expressed in this publication are the opinions and views of the authors, and are not the views of or endorsed by Taylor & Francis. The accuracy of the Content should not be relied upon and should be independently verified with primary sources of information. Taylor and Francis shall not be liable for any losses, actions, claims, proceedings, demands, costs, expenses, damages, and other liabilities whatsoever or howsoever caused arising directly or indirectly in connection with, in relation to or arising out of the use of the Content. This article may be used for research, teaching, and private study purposes. Any substantial or systematic reproduction, redistribution, reselling, loan, sub-licensing, systematic supply, or distribution in any form to anyone is expressly forbidden. Terms & Conditions of access and use can be found at http:// www.tandfonline.com/page/terms-and-conditions
Journal of Environmental Science and Health, Part A (2015) 50, 378–384 Copyright © Taylor & Francis Group, LLC ISSN: 1093-4529 (Print); 1532-4117 (Online) DOI: 10.1080/10934529.2015.987533
Downloaded by [University of Colorado - Health Science Library] at 11:15 10 April 2015
Mathematical modeling of COD removal via the combined treatment of domestic wastewater and landfill leachate based on the PACT process ANGEL S. FERNANDEZ BOU1, ALEXANDRE LIOI NASCENTES2, BARBARA COSTA PEREIRA2, e ALBERTO FERREIRA3 LEONARDO DUARTE BATISTA DA SILVA2, JOAO 4 and JUACYARA CARBONELLI CAMPOS 1
Biosystems Engineering, Fluminense Federal University, Niter oi (RJ), Brazil Institute of Technology, Department of Engineering, Federal Rural University of Rio de Janeiro, Serop edica (RJ), Brazil 3 Center of Technology and Science, Department of Sanitary and Environmental Engineering, Rio de Janeiro State University, Rio de Janeiro (RJ), Brazil 4 School of Chemistry, Department of Inorganic Processes, Federal University of Rio de Janeiro, Rio de Janeiro (RJ), Brazil 2
The experiments performed in this study consisted of 16 batch reactors fed different mixtures of landfill leachate combined with synthetic wastewater treated using the Powdered Activated Carbon Treatment (PACT) process. The objective was to measure the COD mass removal per liter each day for each reactor using two models: the first model combined the variables PAC concentration (0 g¢L¡1, 2 g¢L¡1, 4 g¢L¡1, and 6 g¢L¡1) and leachate rate in the wastewater (0%, 2%, 5%, and 10%), and the second model combined the PAC concentration and the influent COD. The Response Surface Methodology with Central Composite Design was used to describe the response surface of both models considered in this study. Domestic wastewater was produced under controlled conditions in the laboratory where the experiments were performed. The results indicated that the PAC effect was null when the influent did not contain leachate; however, as the concentration of leachate applied to the mixture was increased, the addition of a higher PAC concentration resulted in a better COD mass removal in the reactors. The adjusted R2 values of the two models were greater than 0.95, and the predicted R2 values were greater than 0.93. The models may be useful for wastewater treatment companies to calculate PAC requirements in order to meet COD mass removal objectives in combined treatment. Keywords: Combined treatment of landfill leachate and sewage, powdered activated carbon treatment, mathematical model of COD removal from leachate, COD mass removal with PACT treatment.
Introduction In recent years, the management of municipal solid waste in Brazil has improved, and more than 58% of wastes are disposed in sanitary landfills.[1] This improvement occurred due to recent sanitary and environmental laws, which require all of the municipalities to treat their solid waste properly. However, in 2011, there were still 2,810 cities and towns producing solid wastes which were disposed in 2,906 dumps[2] Nevertheless, due to the new solid waste law, more sanitary landfills are being created, Address correspondence to Angel S. Fernandez Bou, Rua Passo da Patria, 156, bl D, sl 236, S~ ao Domingos, Campus da Praia Vermelha, Niter oi (RJ) 24210-240, Brazil; E-mail: [email protected] Received July 30, 2014. Color versions of one or more of the figures in the article can be found online at www.tandfonline.com/lesa.
thereby increasing the control over traditional environmental problems. Moreover, Brazil has the highest values of long-term average precipitation in the world, with 14,995 km3 of rainfall per year according to AQUASTAT, the FAO’s global water information system. The country holds approximately 12.8% of all freshwater in the world.[3] Large urban nuclei, such as the states of S~ao Paulo or Rio de Janeiro, where the sanitary landfills treat a notorious percentage of all Brazilian solid waste, have an average precipitation in the range of 1,400 to 1,500 mm per year.[4] Landfill leachate generation depends directly on the rainfall amounts. Thus, the new law for controlling solid waste has created the following unintended consequence: much more leachate from sanitary landfills must be treated. As a result, landfill leachate treatment has become one of the most relevant environmental challenges in Brazil. The characteristics of the sanitary landfill leachate depend on different factors, such as rainfall, temperature,
379
Mathematical modeling of COD removal based on the PACT process Table 1. Composition of the synthetic domestic wastewater used in this study. Concentration (mg¢L¡1)
Components
Downloaded by [University of Colorado - Health Science Library] at 11:15 10 April 2015
Peptone from casein Meat extract Urea Monopotassium phosphate Sodium chloride Calcium chloride dihydrate Magnesium sulfate heptahydrate
Table 2. Characteristics of Carbomafra powdered activated carbon (Type: 118 CB AS n 170). Parameters
Result 726.68 m2¢g¡1 560.59 m2¢g¡1 166.08 m2¢g¡1 0.266 cm3 ¢g¡1 25.6 A
BET area Micropore area External surface area Micropore volume Micropore size
360 250 100 26 14 8 4
Source: Campos et al.[19]
Source: Adaptation from the procedures reported by Holler and Tr€ osch[17] to those presented by Von Sperling.[18]
run-off, infiltration, evaporation, air humidity and type of waste.[5] Water is the main component, with a high concentration of dissolved organic matter (normally expressed in COD or BOD), inorganic macro-components, heavy metals and organic xenobiotic components.[6] Several studies indicate the presence of high concentrations of heavy metals in leachate.[7] However, according to some researchers, the concentration of heavy metals is not an issue of concern if the leachate does not contaminate the soil.[8] Giordano analyzed the Gramacho Landfill leachate, located in Rio de Janeiro, Brazil, and found that the heavy metal concentrations were significantly lower than the levels allowed by law.[9] These low concentrations of heavy metals were a consequence of the adsorption and precipitation of metallic ions as hydroxides and sulfides caused by the alkaline and anaerobic environment inside the landfill.[10,11] The landfill age and waste type can result in differences in heavy metal concentrations between countries; e.g., the iron concentration in an old landfill in Brazil was 5.5 mg¢L¡1, whereas that in France was 26 mg¢L¡1, and that in South Korea was 76 mg¢L¡1 (in a medium age landfill in the latter); in addition, the copper concentrations were found to be 0.08 mg¢L¡1 in Brazil, 0.04 mg¢L¡1 in France and 0.78 mg¢L¡1 in South Korea.[5] This result reinforces the variable nature of the leachate composition.
Those values of Brazilian leachates are within the limits allowed by the Brazilian law regarding the discharge of heavy metals into natural water bodies.[12] Sanitary landfill leachates are commonly treated through biological processes, such as activated sludge or stabilization ponds, and using membranes, although other processes can be found in the literature, such as adsorption, coagulation and flocculation. Nevertheless, because the characteristics of the leachate are notably variable, there is not a clear consensus regarding the best treatment system.[13] The Powdered Activated Carbon Treatment (PACT) process was developed by DuPont in the 1970s.[14] The process is a sludge aerobic treatment using powdered activated carbon. The process consists of aerobic digestion of the organic matter, which is synergistic with the use of PAC, thereby improving the results of the system.[15] The objective of this study was to model the COD mass removal using Response Surface Methodology (RSM) based on the results of experiments that studied the combined treatment of wastewater mixed with sanitary landfill leachate via the PACT process, using different settings of leachate mixture and PAC concentration.
Materials and methods The experiments were performed at the Environmental Monitoring Laboratory I – Water and Effluents, in the Federal Rural University of Rio de Janeiro. The analysis
Table 3. Structure of the experimental batteries. Battery I
Battery II
From 02/12/2013 to 03/01/2014
From 15/01/2014 to 07/02/2014
Leachate Percentage in the Mixture
PAC concentration
0 2 4 6
g¢L¡1 g¢L¡1 g¢L¡1 g¢L¡1
0%
2%
5%
10%
R1I R2I R3I R4I
R5I R6I R7I R8I
R1II R2II R3II R4II
R5II R6II R7II R8II
380
Fernandez Bou et al.
Table 4. Leachate addition to the wastewater mixture recipient. Leachate mixture
Leachate Volume Added to 2.5 L of Wastewater
0% 2% 5% 10%
Downloaded by [University of Colorado - Health Science Library] at 11:15 10 April 2015
Table 5. PAC reposition in each reactor.
0 mL 51 mL 132 mL 278 mL
followed the APHA recommendations.[16] Method 5220-D was used to obtain the COD values from the influents and effluents using a Hach 3900 spectrophotometer.
Sampling The sludge seed was grown in the laboratory and its initial concentration was 5,780 mg¢L¡1. The leachate sample originated from the sanitary landfill Dois Arcos, located in the city of S~ ao Pedro da Aldeia (RJ). The leachate was collected in two 10-L plastic receptacles. The synthetic wastewater was produced at the laboratory under controlled conditions, based on the recommendations from Holler and Tr€ osch,[17] with some alterations to adapt the influent to the physicochemical characteristics described by Von Sperling for Brazilian domestic wastewater.[18] The recipe of the synthetic domestic wastewater is presented in Table 1. The powdered activated carbon used in the experiment was Carbomafra Type: 118 CB AS no. 170. The characteristics of this carbon were studied by Campos et al.[19] at the Federal University of Rio de Janeiro and are presented in Table 2.
Experimental structure The experiment was performed in two batteries, each of which had eight reactors, as described in Table 3. There were 16 different configurations of the leachate mixture and the PAC concentration. The first reactor of the first battery (R1I) was the control and thus did not contain any leachate and PAC. The first battery was run with lower leachate concentrations (0% and 2%), compared with those used in the second battery (5% and 10%), whereas the PAC addition was maintained at the same levels in both batteries (0 g¢L¡1, 2 g¢L¡1, 4 g¢L¡1 and 6 g¢L¡1).
Reactor
[PAC]
R1 and R5 R2 and R6 R3 and R7 R4 and R8
0 2 4 6
g¢L¡1 g¢L¡1 g¢L¡1 g¢L¡1
PAC Reposition 0.00 0.07 0.14 0.21
g of PAC g of PAC g of PAC g of PAC
Experimental preparation The sludge age was defined previously as 28 days, i.e., 3.57% of the sludge was removed each day, or 35 mL of the liquor was removed from the reactors. The dissolved oxygen (DO) level was maintained at a level greater than 2 mg¢L¡1 using eight air pumps connected to two uninterruptible power supplies. The hydraulic retention time, HRT, was 40 h, and this time was maintained by removing 60% of the treated effluent every 24 h. The reactors consisted of eight graduated glass cylinders. The working volume of each of the reactors was 1,000 mL. The sludge collected at the treatment plant was carried directly to the laboratory. After sedimentation in 1,850 mL of solution, the sludge was homogenized and divided into eight volumes of 200 mL, using the remaining for analysis. Because the initial concentration of the sludge was 5,780 mg¢L¡1, each reactor received 2,499.45 mg of sludge. The sludge was fed synthetic wastewater over a period of 2 weeks for acclimatization before the application of different mixtures of leachate and PAC was initiated.
System operation For each group of four reactors, 2.5 L of synthetic wastewater were produced each day, and different volumes of sanitary landfill leachate were added to the produced wastewater as detailed in Table 4. To maintain the sludge age, 35 mL of liquor were removed each day, resulting in PAC removal from each reactor (except for the ones without PAC). The PAC reposition was performed according to Table 5. Aeration was operated 23 h per day using conventional aquarium air pumps (one for each reactor) with the goal of maintaining the DO at a level greater than 2 mg¢L¡1 to ensure that it was not a limiting factor.[20,21] Before stopping the aeration, the pH, temperature and DO were measured, and 35 mL of liquor were removed from each
Table 6. COD removal per day and per liter. Leachate % Influent COD
0%
2%
5%
10%
530.5 mg¢L¡1
794.5 mg¢L¡1
1262.8 mg¢L¡1
2001.1 mg¢L¡1
381
Mathematical modeling of COD removal based on the PACT process Table 7. COD removal efficiency. Leachate Rate ! # PAC
Downloaded by [University of Colorado - Health Science Library] at 11:15 10 April 2015
0 2 4 6
g¢L¡1 g¢L¡1 g¢L¡1 g¢L¡1
Table 8. COD mass removal per day and per liter.
0%
2%
5%
10%
90.3% 92.4% 89.1% 83.4%
62.2% 76.3% 74.6% 77.2%
54.3% 70.7% 70.5% 73.9%
47.9% 53.5% 55.0% 59.2%
reactor. The aeration was then stopped, and when the sludge settled at the bottom of the reactors, normally after 45 min, the volume from the upper part was removed from each reactor up to the 400-mL mark, taking care to avoid removing any solid flakes from the bottom. Then 50 mL of treated water from each reactor were collected to perform the COD analysis. Subsequently, PAC reposition was performed, and 600 mL of the wastewater and leachate mixture were added according to the settings of each reactor. Next, 50 mL of each mixture were collected for the COD analysis and then we reconnected the aeration system measuring the pH of the reactors and set the pH to greater than 8.00 with KOH, if necessary.
Leachate Rate ! # PAC 0 2 4 6
g¢L¡1 g¢L¡1 g¢L¡1 g¢L¡1
287.5 294.1 283.7 265.4
Response Surface Methodology (RSM) and Central Composite Design (CCD) were used to understand the COD mass removal. Two models were designed: the first used the PAC and the leachate rate, and the second used the PAC and the COD as independent variables. In both models, the response was the mean COD mass removal obtained for each of the 16 reactors. The equation
mg mg mg mg
2% 296.4 363.9 355.4 368.2
mg mg mg mg
5% 411.5 536.0 534.5 559.8
mg mg mg mg
10% 575.5 642.5 660.9 710.9
mg mg mg mg
intended to describe the behavior of this treatment was a second-order polynomial equation with two independent variables (Eq. 1).[22-24] The efficiency of the treatment was calculated according to Eq. 2. The COD mass removal per day was calculated according to Eq. 3. Z D a1 X2 C a2 Y2 C a3 XY C a4 X C a5 Y C a0 C e
(1)
where Z is the response, X and Y are the relevant independent variables, a1, a2, a3, a4, a5 and a0 are coefficients of the model, and e represents the error. COD Removal Efficiency .%/ D
Data treatment and modeling
0%
COD influent ¡ COD effluent 100 COD influent
(2)
where COD influent is the COD (in mg¢L¡1) of the mixture of wastewater with leachate, and COD effluent is the COD (in mg¢L¡1) of the treated effluent. COD mass removal per day and reactor D .COD influent ¡ COD effluent/ 0:6
L (3) reactor day
where COD influent is the COD (in mg¢L¡1) of the mixture of wastewater with leachate, COD effluent is the COD
Fig. 1. Response surface of model 1: Leachate rate and PAC as the independent variables.
Fig. 2. Plot of the predicted versus actual data for COD mass removal using model 1.
382
Fernandez Bou et al.
Table 9. ANOVA analysis of the COD mass removal response. Response
P-value Prob > F
Std. Dev.
C.V.%
PRESS
R2
Adj. R2
Pred. R2
Adeq. Precision
Model 1 Model 2
< 0.0001 < 0.0001
32.65 30.59
7.31 6.85
23507 21265
0.9667 0.9708
0.9546 0.9602
0.9333 0.9397
25.429 27.033
Downloaded by [University of Colorado - Health Science Library] at 11:15 10 April 2015
Std. Dev. is the standard deviation; C.V.% is the coefficient of variance; PRESS is the predicted residual error sum of squares; R2 is the coefficient of determination; Adj. R2 is the adjusted coefficient of determination; Pred. R2 is the predicted coefficient of determination; Adeq. Precision is the adequate precision.
(in mg¢L¡1) of the treated effluent, and the volume treated per day in each reactor is 0.6 L. The quality of the data fitting to the polynomial models was represented by the coefficient of determination, R2, and the statistical significance was verified using Fisher’s F-test. The non-significant terms were discarded (when “Prob > F” was greater than 0.1). ANOVA was used to completely analyze the results using the Design Expert software.
Results and discussion COD removal efficiency Table 6 shows the COD influent values for the different concentrations of leachate added. The lowest value of 530.5 mg¢L¡1 corresponds to the synthetic wastewater without leachate addition, which was used as the influent for reactors R1I to R4I in the first battery. Table 7 shows the treatment efficiency results. The efficiency of the treatment was greater than 80% in the four reactors without leachate (R1I to R4I) and greater than 90% in R1I and R2I, where the addition of PAC did not improve the treatment. The results presented in Table 7 indicate that an increase in the leachate concentration decreased the efficiency of the COD removal in all of the reactors. The results also indicate that the presence of PAC increased the efficiency in the reactors with leachate. The best result was obtained in reactor R8I, which contained 2% of leachate and 6 g¢L¡1 PAC. The efficiency results obtained for the mixture containing 10% leachate were the worst of the set of experiments for all reactors. None of the reactors achieved a 60% efficiency rate.
COD mass removal models The COD mass removals are presented in Table 8. The results indicate that higher influent COD was associated with greater COD mass removal. At higher PAC concentrations, a greater COD mass tended to be removed with some exceptions: for 0% leachate, the best behavior was obtained for 2 g¢L¡1 and 0 g¢L¡1 PAC; when the leachate concentration was 2%, the best behavior was obtained for 6 g¢L¡1 PAC, but the results were very similar to the COD
mass removal results obtained with 2 g¢L¡1 and 4 g¢L¡1 PAC; when the leachate concentrations were 5% and 10%, the behaviors of the reactors with 2 g¢L¡1 and 4 g¢L¡1 PAC were very similar, but the best COD mass removal was obtained with the reactors with 6 g¢L¡1 PAC. Model 1: PAC and leachate rate as the independent variables. Model 1 describes the behavior of the system when the independent variables used were the leachate rate and the PAC concentration in the reactor. The recommended boundaries for the application of this model are the following: PAC concentration from 0 g¢L¡1 to 6 g¢L¡1 and leachate rate from 0 to 0.1 (0% to 10%). The model is represented by Eq. 4. Figure 1 presents the response surface of this model, and Figure 2 shows the predicted versus actual data obtained from the Design Expert statistical analysis. CODe D 263:61 C 2:61 ¢PAC C 4476:70 ¢Leachate C 232:17 ¢PAC ¢Leachate ¡ 13848:76 ¢Leachate2
(4)
where CODe is the COD mass removal (in mg¢L¡1), PAC is the PAC concentration (in g¢L¡1), and Leachate is the leachate rate (dimensionless). The interval of application of this model is 0 g¢L¡1 PAC 6 g¢L¡1 and 0 Leachate 0.1. “Prob > F” < 0.0001 means that there is a probability lower than 0.01% that this fitting occurs due to noise (Table 9). A third-order polynomial equation was also tested: although the adjusted R2 was slightly better, the PAC term was not significant according to Fisher’s F-test. Thus, the third-order polynomial equation was discarded. For the first second-order polynomial equation calculated, the PAC2 term was not significant and was thus discarded. Removing this term improved the adjusted R2, and the rest of the terms remained significant. Model 2: PAC and COD as independent variables. Model 2 describes the behavior of the system when the independent variables used were the COD of the mixture and the PAC concentration in the reactor. Because the sanitary leachate can be highly variable in composition, this model aims to be more accurate when it is possible to determine the COD of the leachate and the wastewater mixture. The
Downloaded by [University of Colorado - Health Science Library] at 11:15 10 April 2015
Mathematical modeling of COD removal based on the PACT process
383
where CODe is the COD mass removal (in mg¢L¡1), PAC is the PAC concentration (in g¢L¡1) and CODi is the influent COD of the leachate and wastewater mixture (in mg¢L¡1). The interval of application of this model is 0 g¢L¡1 PAC 6 g¢L¡1 and 530 mg¢L¡1 CODi 2,000 mg¢L¡1. “Prob > F” < 0.0001 means that there is a probability lower than 0.01% that this fitting occurs due to noise (Table 9). A third-order polynomial equation was also tested; the adjusted R2 was slightly worse, and the PAC term was not significant according to Fisher’s F-test. Similar to model 1, the PAC2 term for the first second-order polynomial equation calculated, was not significant and was thus discarded. Removing this term improved the adjusted R2, and the rest of the terms remained significant.
Fig. 3. Response surface model 2: influent COD and PAC as the independent variables.
recommended boundaries for the application of this model are the following: PAC concentration from 0 g¢L¡1 to 6 g¢L¡1 and COD of the influent mixture from 530 mg¢L¡1 to 2,000 mg¢L¡1. The model is represented by Eq. 5. Figure 3 presents the response surface of model 2, and Figure 4 shows the predicted versus actual data obtained through the statistical analysis using the Design Expert software. CODe D 76:54 ¡ 5:38 ¢PAC C 0:39 ¢CODi C 0:0156 ¢PAC ¢CODi ¡ 7:262E ¡ 05 ¢CODi2
Conclusions The addition of a higher PAC concentration resulted in a better COD mass removal in the reactors. However, when the leachate rate was zero (no leachate in the mixture), the application of PAC did not improve the COD mass removal. The experimental results indicated that a greater leachate rate was associated with a greater COD mass removal, despite a decrease in the efficiency (inside the intervals of application). Because the models correctly represented the efficiency of the system inside the studied intervals, they may be useful for wastewater treatment companies to calculate the PAC requirements based on the leachate ratio or the COD of the mixture they are going to treat to determine the final COD mass removal.
(5)
Acknowledgments The authors express their gratitude to the interns from the Environmental Monitoring Laboratory I – Water and Effluents at Federal Rural University of Rio de Janeiro, and to the sanitary landfill Dois Arcos.
Funding The authors gratefully acknowledge the financial support given by CAPES.
References
Fig. 4. Plot of the predicted versus actual data for COD mass removal using model 2.
[1] Associa¸c ~ ao Brasileira de Empresas de Limpeza P ublica e Resíduos Especiais (Special Waste and Public Cleaning Brazilian Companies’ Association). Panorama dos Resíduos S olidos no Brasil 2012 (Solid Waste Panorama in Brazil 2012); S~ ao Paulo, 2013, in Portuguese.
Downloaded by [University of Colorado - Health Science Library] at 11:15 10 April 2015
384 [2] Instituto de Pesquisa Econ^ omica Aplicada (Applied Economics Research Institute). Plano Nacional de Resíduos S olidos: diagnostico dos resíduos urbanos, agrosilvopastoris e a quest~ ao dos catadores (Solid Waste National Plan), IPEA official reports no. 145; IPEA: Brasília, 2012. Available at: http://www.ipea.gov.br/ portal/images/stories/PDFs/comunicado/120425_comunicadoi pea0145.pdf (accessed May 2014), in Portuguese. [3] AQUASTAT, FAO. Freshwater Availability - Precipitation and Internal Renewable Water Resources (IRWR); AQUASTAT online database. Available at: http://www.fao.org/nr/water/aqua stat/data/query/index.html?lang=en (accessed July 2014). [4] Ramos, A.M.; Dos Santos, L.A.R.; Fortes, L.T.G. Normais climatologicas do Brasil 1961–1990: edi¸c ~ ao revista e ampliada; Instituto Nacional de Meteorologia: Brasília, 2009, in Portuguese. [5] Renou, S.; Givaudan, J.G.; Poulain, S.; Dirassouyan, F.; Moulin, P. Landfill leachate treatment: Review and opportunity. J. Hazard. Mater. 2008, 150(3), 468–493. [6] Christensen, T.H.; Kjeldsen, P.; Bjerg, P.L.; Jensen, D.L.; Christensen, J.B.; Baun, A.; Albrechtsen, H-J; Heron, G. Biogeochemistry of landfill leachate plumes. Appl. Geochem. 2001, 16(7), 659– 718. [7] Al-Muzaini, S.; Thiem, L. A comparison of leachate treatment between Rhode Island and Kuwait. J. Environ. Sci. Health A 2004, 39(2), 507–514. [8] Kang, K.-H.; H.S. Shin; Park, H. Characterization of humic substances present in landfill leachates with different landfill ages and its implications. Water Res. 2002, 36(16), 4023–4032. [9] Giordano, G. An alise e formula¸c ~ ao de processos para tratamento dos chorumes gerados em aterros de resíduos s olidos urbanos; Material Science and Metalurgy Department, Pontifical Catholic University: Rio de Janeiro, 2003, Doctoral thesis in Portuguese. [10] Castrill on, L.; Fern andez-Nava, Y.; Ulmanu, M.; Anger, I.; n, E. Physico-chemical and biological treatment of MSW no Mara~ landfill leachate. Waste Manage. 2010, 30(2), 228–235. [11] Tatsi, A.; Zouboulis, A. A field investigation of the quantity and quality of leachate from a municipal solid waste landfill in a Mediterranean climate (Thessaloniki, Greece). Adv. Environ. Res. 2002, 6(3), 207–219. [12] BRASIL. Law no. 12305; Brasília, 2010, in Portuguese.
Fernandez Bou et al. [13] Campos, J.C.; Machado, B.D.S.; Blonski, M.E.D.; Bila, D.M..; Ferreira, J.A. Evaluation of coagulation/flocculation process in the landfill leachate treatment at the Municipal Wastewater Treat ment Plant. Rev. Ambient. Agua. 2013, 8(3), 43–53. [14] Flynn, B.P.; Stadnik, J.G. Start-up of a powdered activated carbon-activated sludge treatment system. J. Water Pollut. Con. F. 1979, 51(2), 358–369. [15] Schultz, J.R.; Keinath, T.M. Powdered activated carbon treatment process mechanisms. J. Water Pollut. Con. F. 1984, 56(2), 143–151. [16] APHA. Standard Methods for the Examination of Water and Wastewater, 21st Ed.; APHA/AWWA/WEF: Washington, DC, 2005. [17] Holler, S.; Tr€ osch, W. Treatment of urban wastewater in a membrane bioreactor at high organic loading rates. J. Biotechnol. 2001, 92(2), 95–101. [18] Von Sperling, M. Introdu¸c ~ ao a qualidade das aguas e ao tratamento de esgotos, 2nd Ed.; Editora UFMG: Belo Horizonte, 1996; Vol. 1, 452 pp, in Portuguese. [19] Campos, J.C.; Machado, C.R.A.; Couto, J.M.S.; Florido, P.L.; Cerqueira, A.C.F.P.; Santiago, V.M.J. Evaluation of an activated sludge process combined with powdered activated carbon for the treatment of oil refinery wastewater. J. Int. Environ. Appl. Sci. 2014, 9(1), 24–36. [20] Metcalf and Eddy Inc. Wastewater engineering, treatment, disposal and reuse, 4th Ed.; McGraw-Hill: New York, 2003; 1848 pp. [21] Eckenfelder, W.W. Industrial Water Pollution Control; McGrawHill: Singapore, 1989. [22] Mojiri, A., Aziz, H.A.; Zaman, N.Q., Aziz, S.Q.; Zahed, M.A. Powdered ZELIAC augmented sequencing batch reactors (SBR) process for co-treatment of landfill leachate and domestic wastewater. J. Environ. Manage. 2014, 139, 1–14. [23] Aghamohammadi, N.; Aziz, H.B.A.; Isa, M.H.; Zinatizadeh, A.A. Powdered activated carbon augmented activated sludge process for treatment of semi-aerobic landfill leachate using response surface methodology. Bioresource Technol. 2007, 98(18), 3570–3578. [24] Aziz, S.Q.; Aziz, H.A.; Yusoff, M.S.; Bashir, M.J. Landfill leachate treatment using powdered activated carbon augmented sequencing batch reactor (SBR) process: Optimization by response surface methodology. J. Hazard. Mater. 2011, 189(1), 404–413.
