treating stabilized landfill leachate
TREATING STABILIZED LANDFILL LEACHATE: A CASE OF DYNAMIC MEMBRANE BIOREACTOR (DMBR): STRATEGY FOR START-UP AND PERFORMANCE EVALUATION M. SALEEMA*, A. SPAGNI B, L. ALIBARDIC, R. COSSU D, M. C. LAVAGNOLOD a
Department of Civil, Environmental and Architectural Engineering, University of Padova, via Marzolo 9, 35131 Padova, Italy b Laboratory of Technologies for Waste, Wastewater and Raw Materials Management, Italian National Agency for New Technologies, Energy and Sustainable Economic Development (ENEA), via M.M. Sole 4, 40129 Bologna, Italy c Cranfield Water Science Institute, Cranfield University, Bedford, MK43 0AL, UK d Department of Industrial Engineering, University of Padova, via Marzolo 9, 35131 Padova, Italy * Corresponding author: Mubashir Saleem LISA (Laboratorio di ingegneria sanitaria ambientale) Lungargine Rovetta, 8-35127 PADOVA, Italy Email: [email protected]
SUMMARY: This study aims to evaluate the performance of dynamic membrane (DM) for the treatment of stabilized landfill leachate in two successive lab-scale experimental runs. The main observations of this study highlights the effect of change in feed wastewater characteristics on the formation and performance of DMs including the effect of mesh porosities, the effect of feeding strategy on the biological nitrogen removal performance and NH4+-N conversion efficiency. Furthermore, the study assessed the effect of free ammonia (FA) and free nitrous acid (FNA) on the denitrification activity of the bioreactors. Keywords: Mesh filtration; dynamic membrane; membrane bioreactor, landfill leachate, nitrification, dentrification
1. INTRODUCTION An inevitable consequence of landfilling is the production of landfill leachate (LFL) that contains contaminants originating from variety of biological and physico-chemical processes (Kurniawan et al., 2006; Wang et al., 2016). Due to severe environmental and public health concerns (Kurniawan et al., 2006), the treatment of LFL is indispensable. At the same time it poses a great engineering challenge arising from its long term, unpredictable variability in quality and quantity (Renou et al., 2008). Among the technologies used currently, including physical/chemical treatments, biological leachate treatment is comparatively low cost and well established for organics and ammonia removal in young and mature LFLs (Ahmed and Lan, Proceedings Sardinia 2017 / Sixteenth International Waste Management and Landfill Symposium/ 2 - 6 October 2017 S. Margherita di Pula, Cagliari, Italy / © 2017 by CISA Publisher, Italy
Sardinia 2017 / Sixteenth International Waste Management and Landfill Symposium / 2 - 6 October 2017
2012; Stegmann et al., 2005). In this perspective, membrane assisted biological treatment has shown exceptional results in meeting increasingly stringent LFL discharge standards while successfully accommodating varying loading conditions (Hashisho et al., 2016; Ahmed and Lan, 2012; Robinson, 2017). However, management and control of fouling in membrane bioreactors (MBRs) in general (Judd, 2016), and with leachate treatment in particular, represent a major obstacle in its prevalence (Ahmed and Lan, 2012). To address this issue, an out of the box and cheaper alternative has been proposed as dynamic membranes (DM). DM is a purpose-built, regenerative fouling layer, composed of solids, colloids, microbial cell particles contained in filtering suspension carries out solid-liquid separation (Ersahin et al., 2016; Saleem et al., 2016). As for now, very few studies have been performed to evaluate the performance of DMs in treating LFL (Dong et al., 2007; Xie et al., 2014) as compared to other wastewater streams (Alibardi et al., 2016; Hu et al., 2016; Xiong et al., 2016). Therefore, this study was aimed to increase understanding in this regard while evaluating the formation and performance of DM by integrating it with a pre-anoxic and post-aerobic treatment scheme for the treatment of stabilized LFL. Furthermore, this study also outlined the parameters of concern and the strategy adopted for the successful start-up of the technology under investigation. 2. MATERIAL AND METHODS 2.1.
EXPERIMENTAL SET-UP
This study was conducted on two successive experimental runs lasted for 120 and 220 days respectively. DM filtration was performed in two laboratory scale, continuously mixed, dual chamber bioreactor made up of Plexiglas. The experimental arrangement was consisted of a pre-anoxic tank connected to a post-aerobic tank (Figure 1). The scale of the system was reduced in the second experiment to reduce the demand for influent leachate and thus the frequency of collection of leachate from the landfill site (Table 1).
Figure 1. Schematic diagrams of the experimental setup A modular design was adopted for making DM supports. These modules were constructed by weaving polyamide nylon meshes (10, 52, 85 and 200 µm) over a perforated cylindrical plastic frame. The frame had an external diameter of 15 mm and a height of 70 mm. The perforations (5mm X 3mm) of the supporting cylinder were distributed uniformly along the surface of the frame. The effective filtration area of the DM support was approximately 61 % of the total surface area of the cylindrical frame measuring around 0.0019 m2. Modules were submerged inside the aerobic tank under a constant hydrostatic head of 17 and 8 cm in the first and second experiment respectively. In the first experimental run filtration performance of four different mesh porosities (10, 52, 85 and 200 µm) was tested while in the second experiment 200 and 52
Sardinia 2017 / Sixteenth International Waste Management and Landfill Symposium / 2 - 6 October 2017
µm meshes were mainly tested successively during the experiment in order of their decreasing porosities. Permeate extraction, recirculation flow and leachate feeding was assisted by peristaltic pumps (Watson Marlow series) connected at different locations as shown in figure 1. The study was conducted at ambient temperature averaging around 26.5±0.2 °C throughout the bioreactor operations. Sufficient dissolved oxygen concentration (DO) was maintained inside the aerobic tank by providing aeration through fine air bubble diffusers submerged inside the aeration tank. The two bioreactors were kept completely mixed by using overhead stirrers (LS F201A0151, VELP Scientifica) in the first experiment and with magnetic stirrers (Komet Variomag Maxi) in the second experiments. Table 1. Dimensions of the bioreactors Parameter Volume (Anoxic tank) Volume (Aerobic tank) Recirculation Flow Filtration area
Unit L L Multiple of influent flow m2
Run I 2.8 7.5 4
Run II 0.9 2.5 4-6
0.0057
0.0019-0.0038
Permeate extraction, recirculation flow and leachate feeding was assisted by peristaltic pumps (Watson Marlow series) connected at different locations as shown in figure 1. The study was conducted at ambient temperature averaging around 26.5±0.2 °C throughout the bioreactor operations. Sufficient dissolved oxygen concentration (DO) was maintained inside the aerobic tank by providing aeration through fine air bubble diffusers submerged inside the aeration tank. The two bioreactors were kept completely mixed by using overhead stirrers (LS F201A0151, VELP Scientifica) in the first experiment and with magnetic stirrers (Komet Variomag Maxi) in the second experiments. 2.2.
INNOCULUM AND FEED
The bioreactor was fed with aerobic sludge as initial inoculum taken from the aerobic tank of a conventional activated sludge process of a full-scale municipal wastewater treatment plant located in Padova (Italy). The initial inoculum had a total suspended solids (TSS) concentration of 8.7 and 5.50 g L-1 and volatile suspended solids (VSS) concentration of 5.4 and 3.76 g L-1 in the first and second experiment respectively. Table 2. Characteristics of the leachate samples Parameter BOD TOC NH4+ Average NO3--N Average NO2--N Average TOTAL PHOSPHORUS Average pH Average Alkalinity
Unit mg/L mg/L mg/L mg/L mg/L mg/L
Run I 479 / 325 / 425 1110 / 1154 / 1590 1380 / 1426 / 2272 none 7.65 9.63
Run II 190 / 230 1154 / 585 2275 / 2224 4 0.2 10.3
mg CaCO3/L
8.56 14583
8.56 / 7.66 14583 / 9944
Sardinia 2017 / Sixteenth International Waste Management and Landfill Symposium / 2 - 6 October 2017
In the first experiment, raw (untreated) LFL characterised by high ammonium concentration and low BOD5 values (Table 2) was fed to the bioreactor. However, in the second experiment raw leachate mixed with tap water in gradually increasing concentrations of LFL, starting from 20 % to 100 % over the period of bioreactor operation was fed (Table 2). Raw leachate was collected from a non-hazardous, municipal solid waste (MSW) landfill situated in northern Italy (Sant’Urbano, Padova). Given the lack of bio-available organics in the raw leachate, supplemental readily biodegradable COD was supplied as carbon source to the anoxic tank to support heterotrophic denitrification in the form of tri-hydrated sodium acetate (CH3COONa·3H2O). No pH adjustment was performed in the first experiment however, in the second experiment a 0.5 M HCL solution for pH adjustment was added in the anoxic tank by using a 3-line peristaltic pump operating intermittently and controlled by an electronic timer 2.3.
ANALYTICAL METHODS AND EQUIPMENTS
All the analytical procedures were performed according to Standard Methods (APHA, 2005), otherwise stated elsewhere. Ammonia nitrogen, nitrates and nitrites were measured on the filtered samples (0.45 μm PTFE membranes) taken from the two compartments of the bioreactor and from the effluent. UV-visible Spectrophotometer (Shimatzu UV-1601) was used to analyse all dissolved nitrogen species. Mixed liquor suspended solids and volatile suspended solids concentrations (MLSS and MLVSS) were periodically measured to assess biomass growth inside the system. Dissolved organic content in the influent, effluent and inside the bioreactors were measured as TC and DOC by a Shimadzu TOC-VCSN analyzer. The study was performed at ambient temperature, measured by using an electronic thermometer (Hanna Checktemp °C). Transmembrane pressure (TMP) was measured by means of a U-shaped manometer having water as a manometric fluid. pH measurements were performed by using an electronic pH meter (Crison GLP 22). Effluent turbidity, as a measure of DM solids retention performance, was also measured with the help of (Hach 2100p iso Turbidimeter). 3. RESULTS AND DISCUSSION Excellent solid retention performance of the formed DM was observed during the initial stages (first 26 days) of the first experiment regardless of the difference in mesh porosities and high TMP (Figure 2) and effluent turbidity were less than 5 NTU for all the meshes. However, after the first cleaning operation solids removal performance of the DM was deteriorated due to the change in the feed characteristics (i.e. from municipal wastewater to LFL) because all the other parameters were unchanged. Thereafter, the effect of mesh porosity on the filtration performance became more prominent and soon after that, different trends in, rise in TMP and suspended solids removal were observed for each mesh porosity (200, 85 and 10 μm) (Figure 2). The large size of 200 μm mesh allowed to operate at lower TMP higher flux values and 10 μm mesh exhibited very high suspended solids removal of around 95% followed by 200 and 85 μm meshes with 85 % and 55 % respectively. However, the use of 52 μm mesh provided a considerable trade-off between high filtration fluxes observed for 200 and 85 μm meshes and high effluent quality of 10 μm mesh. The average effluent turbidity values for 10 and 52 μm meshes during the last 16 days of bioreactor operation were 26±3.4 and 12±2.3 NTU respectively. Furthermore, the average filtration flux was also higher than what was observed for 10 μm mesh and it was comparable to 200 and 85 μm meshes averaging around 4.7±0.4 LMH (data not shown). The average TMP during this time interval for 52 μm was 26±4.8 kPa and it was higher than the recorded TMPs for 200 and 85 μm meshes (Figure 2) due to the lower cut-off diameter of 52 μm mesh. In order to examine the previous observation regarding the effect of feed characteristics on the formation and performance of DM, the second experiment was started with gradually increasing concentration of LFL, starting from 20% leachate to 100% leachate during 220 days of
Sardinia 2017 / Sixteenth International Waste Management and Landfill Symposium / 2 - 6 October 2017
continuous bioreactor operation. The results once again suggested that the filtration behavior of the sludge was significantly affected by the change in the characteristics of the feed wastewater (i.e. from municipal wastewater to stabilized landfill leachate) and increase in LFL concertation (p < 10E-7), resulting in higher fouling rate, deteriorated effluent quality and frequent DM cleaning. Turbidity 80
TMP 10000
(a)
60
1000
40
100
20
(b)
60
T M P (k Pa)
10000 1000
40
100
20 10
0 80
10000
(c)
60
1000
40
100
20
T u r b i d i t y (NTU)
10
0 80
10
0 80
10000
(d)
60
1000
40
100
20 10 0 0
20
40
60
80
100
120
Time (d)
Figure 2. Observed TMP and turbidity profiles in the first experimental run for (a) 200 µm, (b) 85 µm, (c) 52 µm and (d) 10 µm mesh A short-term initial start-up phase (lasting for 12 days) characterised by slow and gradual rise in TMP and low effluent turbidity values (Figure 3a) followed by a second, much longer phase with very sharp increase in TMP and lower filtration fluxes (Figure 3b). The solid-liquid separation performance of a 52 µm, achieved greater than 99 % solids retention (data not shown) corresponding to turbidity values less than 10 NTU (Figure 3b). In fact, the strategy allowed operating the system at higher filtration fluxes as compared to the first experiment and an improved retention of suspended solids inside the system (Table 3). (200 µm)
TMP (kPa)
50
10000
(52 µm)
TMP
Turbidity HRT 1000
40 30
100
20 10
10 0
Turbidity (NTU)
60
1 0
5
10 15 20 25 25
50
75
100
125
150
175
200
225
Time (d)
Figure 3. Observed TMP and turbidity profiles in the Second experimental run for (a) 200 µm and (b) 52 µm mesh Since both the experiments were performed on stabilized LFL therefore, only the NH4+-N conversion and total nitrogen removal efficiencies were considered. In both the experimental runs the bioreactors exhibited high NH4+-N conversion of around 84±1.4% and 98.97±0.2% for
Sardinia 2017 / Sixteenth International Waste Management and Landfill Symposium / 2 - 6 October 2017
the first and second experiment respectively (Figure 3). However, in the first experiment due to the presence of high concentrations of free ammonia (FA) and free nitrous acid (FNA), partial nitrification was observed and accumulation of nitrite, as high as 1062 mg L-1 was observed (data not shown). The adverse effect of high concentration of FA and FNA on nitrification and denitrification activity was first studied by Anthonisen et al, (1976). Denitrification performance was also affected by the accumulation of these free species that resulted in a very low total nitrogen removal of only 19±2.6 % due to very low denitrification activity in the first experiment. Table 3. Bioreactor performance during the 2 experimental runs Parameter MLTSS MLVSS ML VSS/TSS Effluent Flux Effluent Total Nitrogen Total Nitrogen removal at steady state (last 20 days) NLR at steady state NH4+-N conversion
Unit g L-1 g L-1 % LMH mg L-1
Run I 6.4 ± 0.2 4.1 ± 0.2 64 4.6 ± 0.2 977 ± 37.7
Run II 10.2 ± 0.7 6.5 ± 0.4 63.7 7.7 ± 0.4 232 ± 19.8
%
19 ± 2.6
83.1 ± 1.8
Kg m-3 d-1 %
0.1 ± 0.005 84 ± 1.4
0.36 ± 0.005 99 ± 0.2
On the other hand, the strategy of gradually increasing the LFL concentration up to 100 % in the feed adopted for second experiment provided enough time for biomass acclimatization. As a result of which the concentration of FA and FNA never reach the toxicity limit in the anoxic tank and on reaching the steady state the bioreactor showed a total nitrogen removal of 98 % corresponding to a nitrogen loading rate of 0.36 kg-Nm-3d-1. Although the system experienced some intervals of poor denitrification whenever the the feed LFL concertation was increased mainly due to high NH4+-N concentration inside the anoxic tank however, the denitrification was always recovered by increasing the recirculation fluxes (from 4 to 6 times of effluent flow) that facilitated their rapid removal from the anoxic tank for further oxidation in the aerobic tank. In short, the solids retention efficiency and biological performance of the system in terms of total nitrogen removal and NH4+-N conversion observed in the second experiment was better than the one observed for first experiment under conditions of higher filtration fluxes and nitrogen loading rates (Table 3). 4. CONCLUSION The study shows the effectiveness of dynamic membrane technology in retaining and enriching slow growing nitrifying bacteria for the treatment of stabilized landfill leachate. During the study the biomass filtration behaviour changed, resulting in higher fouling and rapid rise in TMP due to which the effect of mesh porosity on solid liquid separation performance became prominent. Gradually incresing the concentration of LFL in the influent feed was effective in maintaining steady performance and avoided possible inhibition due to free ammonia and free nitrous acid accumulation. As compared to the first experiment the second experiment showed total nitrogen removal up to 98 % at the end of bioreactor operation similarly high NH4+-N conversion efficiency of > 98% regardless of gradually increasing LFL concentration in the feed up to NLR of 0.36 kg-Nm-3d-1 as observed in the second experiment.
Sardinia 2017 / Sixteenth International Waste Management and Landfill Symposium / 2 - 6 October 2017
1500
2000
80
1600
60
1200
40
800
20
400
0 20
40
60 T i m e (d)
80
100
120
Concentration ( mg N L-1 )
(b) 2400
100
2000
80
1600
60
1200
40 INF NH4+_N Total_N Removal NH4+_N CONVERSION %
400 0
50
100
150 Tiime ( d )
200
Toxicity Limit Free NH3 (Anthonisen et al., 1976)
1200 900 600 300 0
120
2800
800
(c)
20 0 250
0.5
(d)
0.4
First Experiment (Anoxic) Second Experiment (Anoxic)
0.3
Toxicity Limit Free HNO2 (Anthonisen et al., 1976)
0.2 0.1
Free HNO2 (mg N L-1)
0
Free NH3 (mg N L-1)
1800
100
(a)
N H4+_N Conversion and N-removal ( % )
120
2400
N H4+_N Conversion and N-removal ( % )
Concentration ( mg N L-1 )
2800
0.0 0
50
100
150
200
250
T i me ( d)
Figure 4. Biological total nitrogen removal and conversion performance of the bioreactors and influent ammonia nitrogen profile for (a) first experimental run (b) second experimental run (c) free ammonia concentration profile inside the anoxic tank (d) free nitrous acid concentration profile inside the anoxic tank
ACKNOWLEDGMENTS Mubashir Saleem gratefully acknowledges the financial support from Fondazione Cassa di Risparmio di Padova e Rovigo (CARIPARO).
REFERENCES Ahmed, F.N., Lan, C.Q., 2012. Treatment of landfill leachate using membrane bioreactors: A review. Desalination 287, 41–54. Anthonisen, A.C., Loehr, R.C., Prakasam, T.B., Srinath, E.G., 1976. Inhibition of nitrification by ammonia and nitrous acid. Journal - Water Pollution Control Federation 48, 835–52. Alibardi, L., Bernava, N., Cossu, R., Spagni, A., 2016. Anaerobic dynamic membrane bioreactor for wastewater treatment at ambient temperature. Chemical Engineering Journal 284, 130–138. Dong, C.-S., Fan, Y.-B., Li, G., Yang, W.-J., Yuan, D.-D., 2007. Study of a new type of tubular self-forming dynamic membrane bioreactor and its application for treatment of landfill leachate. 28, 747–53. Ersahin, M.E., Tao, Y., Ozgun, H., Spanjers, H., van Lier, J.B., 2016. Characteristics and role of dynamic membrane layer in anaerobic membrane bioreactors. Biotechnology and Bioengineering 113, 761–771. Hashisho, J., El-Fadel, M., Al-Hindi, M., Salam, D., Alameddine, I., 2016. Hollow fiber vs. flat sheet MBR for the treatment of high strength stabilized landfill leachate. Waste Management.
Sardinia 2017 / Sixteenth International Waste Management and Landfill Symposium / 2 - 6 October 2017
Hu, Y., Wang, X.C., Tian, W., Ngo, H.H., Chen, R., 2016. Towards stable operation of a dynamic membrane bioreactor (DMBR): Operational process, behavior and retention effect of dynamic membrane. Journal of Membrane Science 498, 20–29. Judd, S.J., 2016. The status of industrial and municipal effluent treatment with membrane bioreactor technology. Chemical Engineering Journal 305, 37–45. Kurniawan, T.A., Lo, W.-H., Chan, G.Y., 2006. Physico-chemical treatments for removal of recalcitrant contaminants from landfill leachate. Journal of Hazardous Materials 129, 80–100. Renou, S., Givaudan, J.G., Poulain, S., Dirassouyan, F., Moulin, P., 2008. Landfill leachate treatment: Review and opportunity. Journal of Hazardous Materials 150, 468–493. Robinson, T., 2017. Removal of toxic metals during biological treatment of landfill leachates. Waste Management, in press. Stegmann, R., Heyer, K.-U., Cossu°, A.R., 2005. Leachate treatment. Proceedings Sardinia Margherita di Pula 3–7. Wang, H., Wang, Y.-N., Li, X., Sun, Y., Wu, H., Chen, D., 2016. Removal of humic substances from reverse osmosis (RO) and nanofiltration (NF) concentrated leachate using continuously ozone generation-reaction treatment equipment. Xiong, J., Fu, D., Singh, R.P., Ducoste, J.J., 2016. Structural characteristics and development of the cake layer in a dynamic membrane bioreactor. Separation and Purification Technology 167, 88–96. Xie, Z., Wang, Z., Wang, Q., Zhu, C., Wu, Z., 2014. An anaerobic dynamic membrane bioreactor (AnDMBR) for landfill leachate treatment: Performance and microbial community identification. Bioresource Technology 161, 29–39.
Department of Civil, Environmental and Architectural Engineering, University of Padova, via Marzolo 9, 35131 Padova, Italy b Laboratory of Technologies for Waste, Wastewater and Raw Materials Management, Italian National Agency for New Technologies, Energy and Sustainable Economic Development (ENEA), via M.M. Sole 4, 40129 Bologna, Italy c Cranfield Water Science Institute, Cranfield University, Bedford, MK43 0AL, UK d Department of Industrial Engineering, University of Padova, via Marzolo 9, 35131 Padova, Italy * Corresponding author: Mubashir Saleem LISA (Laboratorio di ingegneria sanitaria ambientale) Lungargine Rovetta, 8-35127 PADOVA, Italy Email: [email protected]
SUMMARY: This study aims to evaluate the performance of dynamic membrane (DM) for the treatment of stabilized landfill leachate in two successive lab-scale experimental runs. The main observations of this study highlights the effect of change in feed wastewater characteristics on the formation and performance of DMs including the effect of mesh porosities, the effect of feeding strategy on the biological nitrogen removal performance and NH4+-N conversion efficiency. Furthermore, the study assessed the effect of free ammonia (FA) and free nitrous acid (FNA) on the denitrification activity of the bioreactors. Keywords: Mesh filtration; dynamic membrane; membrane bioreactor, landfill leachate, nitrification, dentrification
1. INTRODUCTION An inevitable consequence of landfilling is the production of landfill leachate (LFL) that contains contaminants originating from variety of biological and physico-chemical processes (Kurniawan et al., 2006; Wang et al., 2016). Due to severe environmental and public health concerns (Kurniawan et al., 2006), the treatment of LFL is indispensable. At the same time it poses a great engineering challenge arising from its long term, unpredictable variability in quality and quantity (Renou et al., 2008). Among the technologies used currently, including physical/chemical treatments, biological leachate treatment is comparatively low cost and well established for organics and ammonia removal in young and mature LFLs (Ahmed and Lan, Proceedings Sardinia 2017 / Sixteenth International Waste Management and Landfill Symposium/ 2 - 6 October 2017 S. Margherita di Pula, Cagliari, Italy / © 2017 by CISA Publisher, Italy
Sardinia 2017 / Sixteenth International Waste Management and Landfill Symposium / 2 - 6 October 2017
2012; Stegmann et al., 2005). In this perspective, membrane assisted biological treatment has shown exceptional results in meeting increasingly stringent LFL discharge standards while successfully accommodating varying loading conditions (Hashisho et al., 2016; Ahmed and Lan, 2012; Robinson, 2017). However, management and control of fouling in membrane bioreactors (MBRs) in general (Judd, 2016), and with leachate treatment in particular, represent a major obstacle in its prevalence (Ahmed and Lan, 2012). To address this issue, an out of the box and cheaper alternative has been proposed as dynamic membranes (DM). DM is a purpose-built, regenerative fouling layer, composed of solids, colloids, microbial cell particles contained in filtering suspension carries out solid-liquid separation (Ersahin et al., 2016; Saleem et al., 2016). As for now, very few studies have been performed to evaluate the performance of DMs in treating LFL (Dong et al., 2007; Xie et al., 2014) as compared to other wastewater streams (Alibardi et al., 2016; Hu et al., 2016; Xiong et al., 2016). Therefore, this study was aimed to increase understanding in this regard while evaluating the formation and performance of DM by integrating it with a pre-anoxic and post-aerobic treatment scheme for the treatment of stabilized LFL. Furthermore, this study also outlined the parameters of concern and the strategy adopted for the successful start-up of the technology under investigation. 2. MATERIAL AND METHODS 2.1.
EXPERIMENTAL SET-UP
This study was conducted on two successive experimental runs lasted for 120 and 220 days respectively. DM filtration was performed in two laboratory scale, continuously mixed, dual chamber bioreactor made up of Plexiglas. The experimental arrangement was consisted of a pre-anoxic tank connected to a post-aerobic tank (Figure 1). The scale of the system was reduced in the second experiment to reduce the demand for influent leachate and thus the frequency of collection of leachate from the landfill site (Table 1).
Figure 1. Schematic diagrams of the experimental setup A modular design was adopted for making DM supports. These modules were constructed by weaving polyamide nylon meshes (10, 52, 85 and 200 µm) over a perforated cylindrical plastic frame. The frame had an external diameter of 15 mm and a height of 70 mm. The perforations (5mm X 3mm) of the supporting cylinder were distributed uniformly along the surface of the frame. The effective filtration area of the DM support was approximately 61 % of the total surface area of the cylindrical frame measuring around 0.0019 m2. Modules were submerged inside the aerobic tank under a constant hydrostatic head of 17 and 8 cm in the first and second experiment respectively. In the first experimental run filtration performance of four different mesh porosities (10, 52, 85 and 200 µm) was tested while in the second experiment 200 and 52
Sardinia 2017 / Sixteenth International Waste Management and Landfill Symposium / 2 - 6 October 2017
µm meshes were mainly tested successively during the experiment in order of their decreasing porosities. Permeate extraction, recirculation flow and leachate feeding was assisted by peristaltic pumps (Watson Marlow series) connected at different locations as shown in figure 1. The study was conducted at ambient temperature averaging around 26.5±0.2 °C throughout the bioreactor operations. Sufficient dissolved oxygen concentration (DO) was maintained inside the aerobic tank by providing aeration through fine air bubble diffusers submerged inside the aeration tank. The two bioreactors were kept completely mixed by using overhead stirrers (LS F201A0151, VELP Scientifica) in the first experiment and with magnetic stirrers (Komet Variomag Maxi) in the second experiments. Table 1. Dimensions of the bioreactors Parameter Volume (Anoxic tank) Volume (Aerobic tank) Recirculation Flow Filtration area
Unit L L Multiple of influent flow m2
Run I 2.8 7.5 4
Run II 0.9 2.5 4-6
0.0057
0.0019-0.0038
Permeate extraction, recirculation flow and leachate feeding was assisted by peristaltic pumps (Watson Marlow series) connected at different locations as shown in figure 1. The study was conducted at ambient temperature averaging around 26.5±0.2 °C throughout the bioreactor operations. Sufficient dissolved oxygen concentration (DO) was maintained inside the aerobic tank by providing aeration through fine air bubble diffusers submerged inside the aeration tank. The two bioreactors were kept completely mixed by using overhead stirrers (LS F201A0151, VELP Scientifica) in the first experiment and with magnetic stirrers (Komet Variomag Maxi) in the second experiments. 2.2.
INNOCULUM AND FEED
The bioreactor was fed with aerobic sludge as initial inoculum taken from the aerobic tank of a conventional activated sludge process of a full-scale municipal wastewater treatment plant located in Padova (Italy). The initial inoculum had a total suspended solids (TSS) concentration of 8.7 and 5.50 g L-1 and volatile suspended solids (VSS) concentration of 5.4 and 3.76 g L-1 in the first and second experiment respectively. Table 2. Characteristics of the leachate samples Parameter BOD TOC NH4+ Average NO3--N Average NO2--N Average TOTAL PHOSPHORUS Average pH Average Alkalinity
Unit mg/L mg/L mg/L mg/L mg/L mg/L
Run I 479 / 325 / 425 1110 / 1154 / 1590 1380 / 1426 / 2272 none 7.65 9.63
Run II 190 / 230 1154 / 585 2275 / 2224 4 0.2 10.3
mg CaCO3/L
8.56 14583
8.56 / 7.66 14583 / 9944
Sardinia 2017 / Sixteenth International Waste Management and Landfill Symposium / 2 - 6 October 2017
In the first experiment, raw (untreated) LFL characterised by high ammonium concentration and low BOD5 values (Table 2) was fed to the bioreactor. However, in the second experiment raw leachate mixed with tap water in gradually increasing concentrations of LFL, starting from 20 % to 100 % over the period of bioreactor operation was fed (Table 2). Raw leachate was collected from a non-hazardous, municipal solid waste (MSW) landfill situated in northern Italy (Sant’Urbano, Padova). Given the lack of bio-available organics in the raw leachate, supplemental readily biodegradable COD was supplied as carbon source to the anoxic tank to support heterotrophic denitrification in the form of tri-hydrated sodium acetate (CH3COONa·3H2O). No pH adjustment was performed in the first experiment however, in the second experiment a 0.5 M HCL solution for pH adjustment was added in the anoxic tank by using a 3-line peristaltic pump operating intermittently and controlled by an electronic timer 2.3.
ANALYTICAL METHODS AND EQUIPMENTS
All the analytical procedures were performed according to Standard Methods (APHA, 2005), otherwise stated elsewhere. Ammonia nitrogen, nitrates and nitrites were measured on the filtered samples (0.45 μm PTFE membranes) taken from the two compartments of the bioreactor and from the effluent. UV-visible Spectrophotometer (Shimatzu UV-1601) was used to analyse all dissolved nitrogen species. Mixed liquor suspended solids and volatile suspended solids concentrations (MLSS and MLVSS) were periodically measured to assess biomass growth inside the system. Dissolved organic content in the influent, effluent and inside the bioreactors were measured as TC and DOC by a Shimadzu TOC-VCSN analyzer. The study was performed at ambient temperature, measured by using an electronic thermometer (Hanna Checktemp °C). Transmembrane pressure (TMP) was measured by means of a U-shaped manometer having water as a manometric fluid. pH measurements were performed by using an electronic pH meter (Crison GLP 22). Effluent turbidity, as a measure of DM solids retention performance, was also measured with the help of (Hach 2100p iso Turbidimeter). 3. RESULTS AND DISCUSSION Excellent solid retention performance of the formed DM was observed during the initial stages (first 26 days) of the first experiment regardless of the difference in mesh porosities and high TMP (Figure 2) and effluent turbidity were less than 5 NTU for all the meshes. However, after the first cleaning operation solids removal performance of the DM was deteriorated due to the change in the feed characteristics (i.e. from municipal wastewater to LFL) because all the other parameters were unchanged. Thereafter, the effect of mesh porosity on the filtration performance became more prominent and soon after that, different trends in, rise in TMP and suspended solids removal were observed for each mesh porosity (200, 85 and 10 μm) (Figure 2). The large size of 200 μm mesh allowed to operate at lower TMP higher flux values and 10 μm mesh exhibited very high suspended solids removal of around 95% followed by 200 and 85 μm meshes with 85 % and 55 % respectively. However, the use of 52 μm mesh provided a considerable trade-off between high filtration fluxes observed for 200 and 85 μm meshes and high effluent quality of 10 μm mesh. The average effluent turbidity values for 10 and 52 μm meshes during the last 16 days of bioreactor operation were 26±3.4 and 12±2.3 NTU respectively. Furthermore, the average filtration flux was also higher than what was observed for 10 μm mesh and it was comparable to 200 and 85 μm meshes averaging around 4.7±0.4 LMH (data not shown). The average TMP during this time interval for 52 μm was 26±4.8 kPa and it was higher than the recorded TMPs for 200 and 85 μm meshes (Figure 2) due to the lower cut-off diameter of 52 μm mesh. In order to examine the previous observation regarding the effect of feed characteristics on the formation and performance of DM, the second experiment was started with gradually increasing concentration of LFL, starting from 20% leachate to 100% leachate during 220 days of
Sardinia 2017 / Sixteenth International Waste Management and Landfill Symposium / 2 - 6 October 2017
continuous bioreactor operation. The results once again suggested that the filtration behavior of the sludge was significantly affected by the change in the characteristics of the feed wastewater (i.e. from municipal wastewater to stabilized landfill leachate) and increase in LFL concertation (p < 10E-7), resulting in higher fouling rate, deteriorated effluent quality and frequent DM cleaning. Turbidity 80
TMP 10000
(a)
60
1000
40
100
20
(b)
60
T M P (k Pa)
10000 1000
40
100
20 10
0 80
10000
(c)
60
1000
40
100
20
T u r b i d i t y (NTU)
10
0 80
10
0 80
10000
(d)
60
1000
40
100
20 10 0 0
20
40
60
80
100
120
Time (d)
Figure 2. Observed TMP and turbidity profiles in the first experimental run for (a) 200 µm, (b) 85 µm, (c) 52 µm and (d) 10 µm mesh A short-term initial start-up phase (lasting for 12 days) characterised by slow and gradual rise in TMP and low effluent turbidity values (Figure 3a) followed by a second, much longer phase with very sharp increase in TMP and lower filtration fluxes (Figure 3b). The solid-liquid separation performance of a 52 µm, achieved greater than 99 % solids retention (data not shown) corresponding to turbidity values less than 10 NTU (Figure 3b). In fact, the strategy allowed operating the system at higher filtration fluxes as compared to the first experiment and an improved retention of suspended solids inside the system (Table 3). (200 µm)
TMP (kPa)
50
10000
(52 µm)
TMP
Turbidity HRT 1000
40 30
100
20 10
10 0
Turbidity (NTU)
60
1 0
5
10 15 20 25 25
50
75
100
125
150
175
200
225
Time (d)
Figure 3. Observed TMP and turbidity profiles in the Second experimental run for (a) 200 µm and (b) 52 µm mesh Since both the experiments were performed on stabilized LFL therefore, only the NH4+-N conversion and total nitrogen removal efficiencies were considered. In both the experimental runs the bioreactors exhibited high NH4+-N conversion of around 84±1.4% and 98.97±0.2% for
Sardinia 2017 / Sixteenth International Waste Management and Landfill Symposium / 2 - 6 October 2017
the first and second experiment respectively (Figure 3). However, in the first experiment due to the presence of high concentrations of free ammonia (FA) and free nitrous acid (FNA), partial nitrification was observed and accumulation of nitrite, as high as 1062 mg L-1 was observed (data not shown). The adverse effect of high concentration of FA and FNA on nitrification and denitrification activity was first studied by Anthonisen et al, (1976). Denitrification performance was also affected by the accumulation of these free species that resulted in a very low total nitrogen removal of only 19±2.6 % due to very low denitrification activity in the first experiment. Table 3. Bioreactor performance during the 2 experimental runs Parameter MLTSS MLVSS ML VSS/TSS Effluent Flux Effluent Total Nitrogen Total Nitrogen removal at steady state (last 20 days) NLR at steady state NH4+-N conversion
Unit g L-1 g L-1 % LMH mg L-1
Run I 6.4 ± 0.2 4.1 ± 0.2 64 4.6 ± 0.2 977 ± 37.7
Run II 10.2 ± 0.7 6.5 ± 0.4 63.7 7.7 ± 0.4 232 ± 19.8
%
19 ± 2.6
83.1 ± 1.8
Kg m-3 d-1 %
0.1 ± 0.005 84 ± 1.4
0.36 ± 0.005 99 ± 0.2
On the other hand, the strategy of gradually increasing the LFL concentration up to 100 % in the feed adopted for second experiment provided enough time for biomass acclimatization. As a result of which the concentration of FA and FNA never reach the toxicity limit in the anoxic tank and on reaching the steady state the bioreactor showed a total nitrogen removal of 98 % corresponding to a nitrogen loading rate of 0.36 kg-Nm-3d-1. Although the system experienced some intervals of poor denitrification whenever the the feed LFL concertation was increased mainly due to high NH4+-N concentration inside the anoxic tank however, the denitrification was always recovered by increasing the recirculation fluxes (from 4 to 6 times of effluent flow) that facilitated their rapid removal from the anoxic tank for further oxidation in the aerobic tank. In short, the solids retention efficiency and biological performance of the system in terms of total nitrogen removal and NH4+-N conversion observed in the second experiment was better than the one observed for first experiment under conditions of higher filtration fluxes and nitrogen loading rates (Table 3). 4. CONCLUSION The study shows the effectiveness of dynamic membrane technology in retaining and enriching slow growing nitrifying bacteria for the treatment of stabilized landfill leachate. During the study the biomass filtration behaviour changed, resulting in higher fouling and rapid rise in TMP due to which the effect of mesh porosity on solid liquid separation performance became prominent. Gradually incresing the concentration of LFL in the influent feed was effective in maintaining steady performance and avoided possible inhibition due to free ammonia and free nitrous acid accumulation. As compared to the first experiment the second experiment showed total nitrogen removal up to 98 % at the end of bioreactor operation similarly high NH4+-N conversion efficiency of > 98% regardless of gradually increasing LFL concentration in the feed up to NLR of 0.36 kg-Nm-3d-1 as observed in the second experiment.
Sardinia 2017 / Sixteenth International Waste Management and Landfill Symposium / 2 - 6 October 2017
1500
2000
80
1600
60
1200
40
800
20
400
0 20
40
60 T i m e (d)
80
100
120
Concentration ( mg N L-1 )
(b) 2400
100
2000
80
1600
60
1200
40 INF NH4+_N Total_N Removal NH4+_N CONVERSION %
400 0
50
100
150 Tiime ( d )
200
Toxicity Limit Free NH3 (Anthonisen et al., 1976)
1200 900 600 300 0
120
2800
800
(c)
20 0 250
0.5
(d)
0.4
First Experiment (Anoxic) Second Experiment (Anoxic)
0.3
Toxicity Limit Free HNO2 (Anthonisen et al., 1976)
0.2 0.1
Free HNO2 (mg N L-1)
0
Free NH3 (mg N L-1)
1800
100
(a)
N H4+_N Conversion and N-removal ( % )
120
2400
N H4+_N Conversion and N-removal ( % )
Concentration ( mg N L-1 )
2800
0.0 0
50
100
150
200
250
T i me ( d)
Figure 4. Biological total nitrogen removal and conversion performance of the bioreactors and influent ammonia nitrogen profile for (a) first experimental run (b) second experimental run (c) free ammonia concentration profile inside the anoxic tank (d) free nitrous acid concentration profile inside the anoxic tank
ACKNOWLEDGMENTS Mubashir Saleem gratefully acknowledges the financial support from Fondazione Cassa di Risparmio di Padova e Rovigo (CARIPARO).
REFERENCES Ahmed, F.N., Lan, C.Q., 2012. Treatment of landfill leachate using membrane bioreactors: A review. Desalination 287, 41–54. Anthonisen, A.C., Loehr, R.C., Prakasam, T.B., Srinath, E.G., 1976. Inhibition of nitrification by ammonia and nitrous acid. Journal - Water Pollution Control Federation 48, 835–52. Alibardi, L., Bernava, N., Cossu, R., Spagni, A., 2016. Anaerobic dynamic membrane bioreactor for wastewater treatment at ambient temperature. Chemical Engineering Journal 284, 130–138. Dong, C.-S., Fan, Y.-B., Li, G., Yang, W.-J., Yuan, D.-D., 2007. Study of a new type of tubular self-forming dynamic membrane bioreactor and its application for treatment of landfill leachate. 28, 747–53. Ersahin, M.E., Tao, Y., Ozgun, H., Spanjers, H., van Lier, J.B., 2016. Characteristics and role of dynamic membrane layer in anaerobic membrane bioreactors. Biotechnology and Bioengineering 113, 761–771. Hashisho, J., El-Fadel, M., Al-Hindi, M., Salam, D., Alameddine, I., 2016. Hollow fiber vs. flat sheet MBR for the treatment of high strength stabilized landfill leachate. Waste Management.
Sardinia 2017 / Sixteenth International Waste Management and Landfill Symposium / 2 - 6 October 2017
Hu, Y., Wang, X.C., Tian, W., Ngo, H.H., Chen, R., 2016. Towards stable operation of a dynamic membrane bioreactor (DMBR): Operational process, behavior and retention effect of dynamic membrane. Journal of Membrane Science 498, 20–29. Judd, S.J., 2016. The status of industrial and municipal effluent treatment with membrane bioreactor technology. Chemical Engineering Journal 305, 37–45. Kurniawan, T.A., Lo, W.-H., Chan, G.Y., 2006. Physico-chemical treatments for removal of recalcitrant contaminants from landfill leachate. Journal of Hazardous Materials 129, 80–100. Renou, S., Givaudan, J.G., Poulain, S., Dirassouyan, F., Moulin, P., 2008. Landfill leachate treatment: Review and opportunity. Journal of Hazardous Materials 150, 468–493. Robinson, T., 2017. Removal of toxic metals during biological treatment of landfill leachates. Waste Management, in press. Stegmann, R., Heyer, K.-U., Cossu°, A.R., 2005. Leachate treatment. Proceedings Sardinia Margherita di Pula 3–7. Wang, H., Wang, Y.-N., Li, X., Sun, Y., Wu, H., Chen, D., 2016. Removal of humic substances from reverse osmosis (RO) and nanofiltration (NF) concentrated leachate using continuously ozone generation-reaction treatment equipment. Xiong, J., Fu, D., Singh, R.P., Ducoste, J.J., 2016. Structural characteristics and development of the cake layer in a dynamic membrane bioreactor. Separation and Purification Technology 167, 88–96. Xie, Z., Wang, Z., Wang, Q., Zhu, C., Wu, Z., 2014. An anaerobic dynamic membrane bioreactor (AnDMBR) for landfill leachate treatment: Performance and microbial community identification. Bioresource Technology 161, 29–39.
