This chapter discusses the performance of HSSF and VSSF constructed wetland system (CWS) in treating landfill leachate (COD, NH3-N, PO4-P, Mn, and Fe) based on the pilot scale reactors and laboratory analysis. In this study, landfill leachate with high COD and ammonia nitrogen concentration was treated by using constructed wetland systems operated in sub-surface flow mode. The experiment with two vertical and two horizontal flow subsurface systems was carried out during December 2009 until February 2010 including planting and acclimation periods.
4.2 Initial leachate characterization study
The initial leachate characterization study was conducted to determine the physical and chemical characteristic of the leachate and to determine the most significant heavy metals that will be the parameter of interest. The results of initial leachate characterization study are summarized in Table 4.1. The initial leachate characterization study was conducted using Atomic Absorption Spectrometry (AAS)
Table 4.1: Results of initial leachate characteristic
Get your grade
or your money back
using our Essay Writing Service!
Ammonia Nitrogen (NH3-N)
Chemical Oxygen Demand (COD)
Note: ND = Not detected
Referring to Table 4.1, it can be clearly observed that the leachate sample was high in nutrient as it is being indicated by high ammonia-nitrogen and phosphate value, with 414mg/l and 30.5mg/l for ammonia-nitrogen and phosphate respectively. The leachate has also exhibit significant value of heavy metals content, among which the highest concentration was recorded for manganese (Mn) and iron (Fe) with 10.6mg/L and 11.6mg/L respectively. However, the leachate sample collected at Padang Siding MSWLF was observed to have COD concentration of 636mg/L which does not represent a new and operating landfill as new and operating landfill should have higher COD concentration (Kjeldsen et al., 2002). This is because the leachate has undergone a pre-treatment process as it has been collected at the aeration pond which has increased the dissolved oxygen level in the leachate.
4.3 COD removal
The results of overall treatment efficiencies of all four microcosms in this study are summarized in Table 4.2. As for organic compound such as COD the biological reactions are supposed to be due to the microbial activity, through bio-film formation on the bed material (Yalcuk & Ugurlu, 2008). The influent concentration of COD in this study was 636mg/L and has been subsequently reduced significantly throughout the treatment period as can be observed in Figure 4.1. Although it has been commonly known that the organic matter in the leachate are degraded by the heterotrophic microorganism in the wetland reactor depending on the concentration of oxygen within the wetland reactor (Kadlec et al., 2000; Ong 2008), it may be degraded in SSF CWs both aerobically and anaerobically and it is difficult to quantify the ratio between aerobic and anaerobic degradation (Vymazal, 2000), since both conditions exists in the HSSF and VSSF systems.
Figure 4.1: Concentration of COD throughout treatment period.
As shown in Figure 4.1, all four (4) reactors managed to significantly reduced the concentration of COD in the leachate to as low as 10 mg/L which recorded for the effluent of HP reactor (refer Table 4.3). The effluent concentration of COD were reduced and varies among four reactors with 42 mg/L, 37mg/L, 18mg/L and 10mg/L for reactor VC, VP, HC, and HP respectively (refer Table 4.3). Whereas the optimum removal was achieved on the 12th day for VP, HC and HP with 38.84%, 45.60% and 51.42% removal respectively, while the optimum removal for reactor VC was only achieved on the 15th day of the treatment period with the removal of 36.95%. This clearly shows that HSSF system which has higher HRT values requires shorter treatment period as compared to VSSF system (Yalcuk & Ugurlu, 2008). A higher removal and shorter treatment period was also apparent for planted system (VP and HP) as compared to the unplanted system (VC and HC) be evidence for the role of macrophytes in expediting the treatment process, which due to its role in providing a large surface area for attached microbial growth and the supply of oxygen into the rhizosphere that can promotes the development of aerobic bacteria (Gersberg et al., 1986; Tanner, 2001; Gagnon et al., 2006; Brisson & Chazarenc, 2008).
Always on Time
Marked to Standard
Table 4.2: Overall treatment performance of the wetland systems
VC (% removal)
VP (% removal)
HC (% removal)
HP (% removal)
Apart from being significant in expediting the treatment period for COD removal there is not much difference in the overall treatment performances (refer Table 4.2) between planted and unplanted systems with 94.18% as compared to 93.40% for VSSF systems and 98.43% as compared to 97.17% for planted(HP) and unplanted (HC) HSSF systems, which coherent with the findings by Akratos and Tsihrintzis (2007) which found that the presence of plant improved the removal of COD only slightly with 89.3% as compared to 87.2% in unplanted units. This is because although emergent plants are the most obvious components of the wetland ecosystem, wastewater treatment is accomplished through an integrated combination of biological, physical and chemical interactions between the plants, the substrata and the inherent microbial community (Wood, 1994; Vymazal, 2000). The excellent COD removal was mainly contributed by the microbial activities both aerobic biological decomposition and denitrification processes. The gravel media placed inside the UFCW reactors allowed the accumulation of immerse amounts of attached microbes, which were very helpful in rapidly catalyzing biochemical reactions (Ong et al., 2008)
Table 4.3: Initial and final concentration of leachate sample
However, the influent and effluent correlation (R2 = 0.989(VC), 0.91(VP), 0.98(HC), 0.80(HP)) has shown a significant difference in the concentration of effluent COD between VSSF and HSSF with 42mg/L and 18mg/L for unplanted control unit (VC and HC), and 37mg/L and 10mg/L for planted systems (VP and HP). It can also be noticed in Figure 4.1 that there were a significant increase in the concentration of COD in both planted system which on day 21 and day 24 for VP and HP respectively, the reason for this event is due to the accumulation of dead roots that starts to degrade, in which it turn into biomass that conserved oxygen within it (G. Tchobanoglous et al., 2003) as VP exhibit a higher increment as compared to HP due to greater number of shoots that died along the treatment period (refer Figure 4.6).
4.4 Nutrients removal
The nitrogen cycle in wetlands is complex and is discussed in detail by Kadlec and Knight (1996) and Reed et al. (1995). Nitrogen transformation in wetlands occurs by five principal biological processes: ammonification, nitrification, denitrification, nitrogen fixation and nitrogen assimilation. For secondary treated wastewater in which the predominant forms of nitrogen are ammonia and nitrate, nitrification and denitrification are generally indicated as the principal processes for nitrogen reduction together with some assimilation by biota (Greenway and Wooley, 1999). Nitrogen removal is achieved not only by bacteria, but also by plant uptake which have been visualized in Figure 4.9, adsorption, where ionized ammonia reacts with the media in SF constructed wetlands (Kadlec and Knight, 1996; Yang et al., 2001; Al-Omari and Fayyad, 2003), and volatilization, where ammonia is transformed to free nitrogen (Kadlec and Knight,1996; Yang et al., 2001; Vymazal, 2002; Tanner et al., 2002; Al-Omari and Fayyad, 2003; Mayo and Mutamba, 2004) The magnitude of the reduction depends on factors such as temperature, pH, alkalinity, organic carbon, dissolved oxygen and biota (Reed et al., 1995; Kadlec and Knight, 1996).
Figure 4.2: Concentration of NH3-N throughout the treatment period
As shown in Table 4.3, the influent concentration of Ammonia Nitrogen (NH3-N) in the leachate sample used in this study was high with the concentration of 414.0mg/L, which was then reduced significantly to 7.200mg/L, 2.240mg/L, 5.720mg/L and 1.540mg/L for reactor VC, VP, HC, and HP respectively as it been demonstrated in Figure 4.2. Whereas, the optimum phosphate removal for the planted system (VP ans HP) was achieved on the 3rd day (HP) and 6th day (VP) of the treatment period with 33.33% and 47.83% respectively, while for the unplanted systems, the optimum removal efficiency was achieved on the 9th day with 50.72% and 52.66% for reactor VC and HC respectively. In this study, all of the reactors performed well in the removal of NH3-N in the landfill leachate, as indicated by the overall treatment performance which does not differ greatly between VSSF system and HSSF system with 98.26%, 99.46%, 98.62% and 99.63% for reactor VC, VP, HC and HP (refer Table 4.2). The higher removal efficiency of the HSSF system as compared to VSSF systems indicates that higher removal efficiencies can be achieved with the increase in HRT, the correlation between NH3-N removal and HRT is consistent with the findings of Sakadevan et al. (1995) and Jing and Lin, (2003).
This Essay is
a Student's Work
This essay has been submitted by a student. This is not an example of the work written by our professional essay writers.Examples of our work
NH3-N in a constructed wetland is removed by volatilization, adsorption, plant uptake and nitrification (IWA, 2000). In this work, the pH in the VSSF and HSSF wetlands were found to be ranging from 8 to 6 throughout the treatment period. NH3-N loss through volatilization was negligible in the study because NH3-N volatilization is generally insignificant at pH below 9.3 (IWA, 2000). NH3-N could be adsorbed on sediment or matrix in wetlands; however, such removal is not considered to be a long-term sink because the adsorbed NH3-N is released easily when water chemistry conditions change (Kadlec and Knight, 1996). Besides, nitrogen could be also taken up by plants, but released back to the water after decomposition process, as it been indicated by the increase in the concentration of NH3-N during the treatment period for both planted system (refer Figure 4.2). Consequently, nitrification played an important and a long-term role in removing NH3-N. Furthermore, the appearance of air bubble in the outlet zone of the reactors may suggest that denitrification may occur in the all four reactors which happen due to the anaerobic or anoxic condition within the microcosms (Jing and Lin, 2003).
Phosphate removal is governed by physical (sedimentation) and chemical (adsorption) process and biological transformations (Yalcuk & Ugurlu, 2008). Phosphorus can be removed directly by macrophytes uptake or chemical storage in the sediments (Bonomo et al., 1997). Sakadevan and Bavor (1998) suggest that the principle long-term phosphate removal mechanism in constructed wetland systems is via substratum, litter and Al/Fe component, while plant uptakes being in smaller extend which was shown by a small difference between the overall treatment performance between planted and unplanted system in this study. The treatment performance also does not differ greatly between VSSF system and HSSF system with 98.95%, 99.21%, 99.05% and 99.87% for reactor VC, VP, HC and HP (refer Table 4.2). The planted and unplanted system shows a small difference in term of overall treatment performance, although the uptake of phosphate as high as 4.313mg/g (refer Figure 4.9) has been determined in the root tissue of Limnocharis flava planted in HSSF system.
Figure 4.3: Concentration of PO4-P throughout treatment period
The influent concentration of phosphate in the leachate sample used in this study was 30.50mg/L, which was then reduced significantly to 0.320mg/L, 0.240mg/L, 0.290mg/L and 0.040mg/L for reactor VC, VP, HC, and HP respectively as it been demonstrated in Figure 4.3. Whereas, the optimum phosphate removal for the planted system (VP ans HP) was achieved on the 3rd day of the treatment with 20.0% and 25.25% respectively, while for the unplanted systems (VC and HC) the optimum removal efficiency was achieved on the 9th day (40.98%) for reactor HC, and 12th day (35.74%) for reactor VC. In this study, it was believed that, most phosphate is believed to be stored in the media bed (soil and gravel) rather than in the macrophytes tissues given that the difference in overall treatment efficiency between planted and unplanted system was relatively small. It can also believe that the adsorption played an important role for phosphate removal as indicated with a significant adsorption capacity of 2.85mg/g for gravel and 3.01mg/g for soil (refer Figure 4.10). The result of this study have indicated that HSSF systems have a higher treatment efficiency as compared to VSSF system, in which it is contrary with the findings from Yalcuk and Ugurlu, (2008) which indicated that VSSF systems have a higher removal percentage than HSSF system. A higher overall treatment performance and shorter period in achieving the optimum removal have shown the significance of macrophytes addition in the wetland systems. Basically, the successful removal of phosphate in CWs are governed the coexistence of various processes such as adsorption, precipitation and uptake by plant or microorganism (Bonomo et al., 1997; Kaldec et al., 1997; Reddy et al., 1997; Sakadevan and Bavor, 1998; Gray et al., 2000 and Ong et al., 2008)
4.5 Heavy metals removal
The metal removal processes in constructed wetlands is very complex and these processes include a combination of biotic and abiotic reactions such as sedimentation, flocculation, adsorption, precipitation, co-precipitation, cation and anion exchange, complexation, oxidation and reduction, microbial activity and plant up-take (Kosopolov et al., 2004; Ujang et al., 2005; Yalcuk & Ugurlu, 2008). The metals cannot be destroyed but their chemical and physical characteristics are modified (Ujang et al., 2005; Yalcuk & Ugurlu, 2008).
Figure 4.4: Concentration of Fe throughout treatment period
The influent concentration of Iron (Fe) in the leachate sample was 11.6mg/L and it was subsequently reduced to a significantly low concentration throughout the treatment period as shown in Figure 4.4. As shown in Table 4.3, the final effluent concentration of Fe was significantly reduced and varies among all four wetland microcosm, with 0.985mg/L, 0.250 mg/L, 0.653 mg/L and 0.093 mg/L for reactor VC, VP, HC, and HP respectively. The optimum removal percentage of the planted systems (VP and HP) was recorded on the 3rd day of the treatment period with 15.517% and 17.241% for VP and HP respectively, while the optimum removal of the unplanted control units (VC and HC) was only achieved on the 21st day of the treatment period with 32.759% for VC and 24.138% for HC reactor. However, it was observed the overall treatment efficiency for all reactors does not varies greatly between each other, with the most efficient system in the removal of Fe from landfill leachate was reactor HP with the removal rate of 99.20%, while the least efficient system was reactor VC with removal rate of 91.51% (refer Table 4.2). These finding was higher than those reported by Jayaweera et al. (2008) Maine et al. (2009) and Khan et al. (2009), which proved the effectiveness of CWs in the removal of Fe from wastewater and concurrent with the findings by Ye et al. (2001a,b) and O'Sullivan et al. (2004) but in contrast with those reported by Lesage et al. (2007).
Figure 4.5: Concentration of Mn throughout treatment period
While, as shown in Table 4.3 the influent concentration of manganese (Mn) of the leachate sample was 10.6mg/L and as it been demonstrated in Figure 4.5, the concentration was reduced to a significant value of 0.561 mg/L, 0.042mg/L, 0.323mg/L and 0.027 mg/L for VC, VP, HC, and HP reactors respectively. Which shows that all of the reactors were efficient in the removal of Mn from the landfill leachate sample, with the highest overall treatment efficiency was recorded for reactor HP with 99.75% removal rate and the least was reactor VC with 94.71% removal at the end of treatment period (refer Table 4.2). The optimum removal of Mn for the planted systems (VP and HP) was recorded on the 3rd day of the treatment period with 18.87% and 20.76% for VP and HP respectively, while the optimum removal of the unplanted control units (VC and HC) was only achieved on the 12th and 15th day of the treatment period with 16.98% for VC and 25.66% for HC reactor.
The obvious difference in achieving the optimum removal for Fe and Mn in both planted and unplanted systems indicates the positive effect of macrophytes in establishing external aerobic microzone around parts of the growing roots (Armstrong and Armstrong, 1990). As results, iron and manganese is precipitated and iron plaque or coating appears (Peverly et al., 1995), where iron plaque can be observed as iron is concentrated in the root sample which caused the rood to exhibit an orange, oxidized iron color on its surfaces (Peverly et al., 1995) .The cause of such depositions is thought to be oxygen movement across the root cortex into the rhizosphere when oxygen supply to the roots from tops is sufficient (Conlin and Crowder, 1989). Jayaweera et al. (2008) and Khan et al. (2009) indicated that chemical precipitation and rhizofilteration are the key mechanisms to remove Fe from Fe-rich wastewater, where in this study it was also enhanced by the uptake of Limnocharis flava (refer Figure 4.9). Although there were not much difference, the lower Mn and Fe concentration in the effluent of HSSF systems as compared to VSSF system also indicates that the higher removal of heavy metals were managed to be achieved by HSSF system which have a higher HRT values compared to VSSF systems (Yalcuk & Ugurlu, 2008). The findings by Ujang et al. (2005); Yalcuk & Ugurlu, (2008), which stated that heavy metals cannot be destroyed but their chemical and physical characteristics are modified are also proven in this study by the increase of Mn and Fe concentration in the effluent of the reactors (refer Figure 4.4 and Figure 4.5), suggesting that these metals were actually washed out of the systems (Yalcuk & Ugurlu, 2008).
4.6 Plant Monitoring and Analysis
The plant monitoring was conducted through physical observation of the plant's (limnocharis flava) adaptability to the polluted environment, density of shoots and stem, suitability of the plant for landfill leachate treatment, and other related issues that arise such as pest and disease attack. During the initial period of the study, the plant managed to acclimatize well in the new environment, as it indicated by the increased number of stem and the healthy grow of stem, leaves, and fluorescence. Shoot counts for both planted reactors were conducted in a weekly basis, where Figure 4.6 shows the dynamics of Limnocharis flava's growth throughout the treatment period.
Figure 4.6: Plant's growth rate throughout treatment period (note that day 0 indicates initial number of limnocharis flava planted into the reactor, while day 0* indicates the addition of leachate into the reactors)
Plants planted in both reactors (VP & HP) shows a significant growth during the acclimatization period with an increase of 7 and 9 stems for VP and HP respectively. However, the density of plants in both reactors starts to decline after the addition of leachate, where tip burn and chlorosis symptom and also plant death were observed. This symptom also has been reported for Phragmites australis and Manchurian wild rice after the addition of acid orange VII (Ong et al., 2008). The death of plants may also catalyzed by insects and disease attack such as from aphids. Aphids attack was observed during the first week after the addition of leachate, as shown in Figure 4.7. It may pose a significant threat to the development of plant, if its growth were somehow unable to be contained. However, in this study, it was observed that, the appearance of ants at the surrounding of the reactors might have become the solutions of the aphid problem. Where, the increased population of ants has significantly reduced the density of aphids per stems. The growth of young stem was also significantly inhibited after the addition of leachate with only a few numbers young stems was observed throughout the experimental period Although, there was insignificant development of new stem, the plant managed to accommodate well with the high COD, nutrient, and heavy metal, during the middle of experimental period as indicated by the healthy stems and leaves. The ability to withstand the highly polluted environment has shows the suitability of Limnocharis flava to be used in treating landfill leachate as low growth or poor health may indicate that the species simply does not withstand well the specific conditions under which it was tested (type of wastewater, matrix, climate, etc.) in which case it is unsuitable for such constructed wetland (Brisson & Chazarenc, 2008).
Figure 4.7: Aphids attack on Limnocharis flava
(a) Aphids on plant's stem; (b) closed-up view of Wingless aphid
Plant size at a specific time is an obvious measure of plant health because it integrates plant performance over a long period of time (Brisson & Chazarenc, 2008). In this study, it was also observed that there is a significant difference in the density and the size of plant in reactor HP and VP as shown in Figure 4.6. The reasons behind this differences is due to the natural habitat of Limnocharis flava itself, where it grows well at ditches and swamps surrounding the paddy field, which provide them with unlimited supply of water. As being an emergent plant, Limnocharis flava need high amount of water available at its surrounding in order to successfully grow well. This need was unable to be fulfilled as it been planted in vertical subsurface flow constructed wetland system, which have lower water level as compared to horizontal subsurface flow constructed wetland system. Apart from being suitable to be used in HSSF system, this study has clearly shown that Limnocharis flava, are more suitable to be used in free water surface (FWS) constructed wetland system. The difference in density of plant for both reactors during the initial and final phased of the experiment as shown in Figure 4.8.
Figure 4.8: Emergent plant during (a) initial and (b) final phase of the experiment
4.6.1 Analysis of Plant Tissue
The analysis of plant tissue was conducted to study the extents of phytoaccumulation or phytoextraction of nutrient (ammonia nitrogen and orthophosphate) and heavy metals (iron and manganese) in the plant tissues which was segregated into three (3) main components which is leaves, stems, and roots. The results of the plant tissue analysis as shown in Figure 4.4 shows that there was an accumulation of nutrients and heavy metals in the tissue of Limnocharis flava planted in both horizontal (HP) and vertical (VP) subsurface flow system. The accumulations of pollutant (nutrients and heavy metals) shows that the contribution of macrophytes in the sense of the uptake of pollutant is significant in this study, apart from providing a large surface area for attached microbial growth, supplying reduced carbon through root exudates and micro-aerobic environment and a via root oxygen release in the rhizosphere, and stabilizing the surface of the bed (Brisson & Chazarenc, 2008; Gersberg et al., 1986; Tanner, 2001; Gagnon et al., 2006).
Figure 4.9: Accumulation of pollutant in plant tissue
As shown in Figure 4.9, it can be observed that Limnocharis flava planted in both reactors (VP and HP) exhibit the highest uptake on phosphate (PO4-P). Where, the highest amount of phosphate was observed to be accumulated within the plant's root, with 3.950mg/g and 4.313mg/g for HSSF and VSSF reactors respectively and decreasing towards the plant's stems and leaves with 3.100mg/g (leaf), 3.050 mg/g (stem) and 3.738mg/g (leaf), 3.900 mg/g (stem) for VSSF and HSSF respectively. The high uptakes of phosphate are coherent with the needs for nutrients in order for it to survive. Apart from being undisputedly great in the uptake of phosphate, Limnocharis flava has shown a momentous result in the uptake of other parameters such as nitrogen (ammonia nitrogen), and iron. Ammonia nitrogen uptake was also high with the uptake of 1.263mg/g and 1.388mg/g for VSSF and HSSF respectively, both observed in the root tissue where nitrogen removal is known to be influenced by the presence of plant directly through assimilation (Brisson & Chazarenc, 2008).
The ability of Limnocharis flava to uptake heavy metals was also proven in this study, which coherent with the findings by Kosopolov et al., 2004; Ujang et al., 2005; Yalcuk & Ugurlu, 2008. Where, the highest amount of heavy metals were determined in the root for both VSSF and HSSF, with 0.223mg/g (VSSF), 0.362mg/g (HSSF) and 0.728mg/g (VSSF), 1.117mg/g (HSSF) for manganese and iron respectively which consistent with the findings by Janet et al. (1990) and Peverly et al. (1995). As it can be noticed in Figure 4.4, HSSF systems exhibit a higher uptake of nutrients and heavy metals as compared to VSSF system due to the higher HRT for HSSF system. These findings have shown the significant and positive effect of macrophytes on pollutant removal (Tanner, 2001) and that the role of macrophytes as an essential component of constructed wetland is well established (Brix, 1997; Stottmeister et al., 2003; Brisson & Chazarenc, 2008).
4.7 Soil Analysis
The wetland media is one of the important components of CWs, as it provides a viable condition for maximum removal of pollutant, since the reduction is said to be accomplished by diverse treatment mechanisms including sedimentation, filtration, chemical precipitation and adsorption, microbial interactions and uptake by vegetation (Watson et al., 1989) which governed by the accurate selection of media type. Therefore in this study, soil analysis was conducted to determine the suitability of the media beds used, as it is indicated by the accumulation of the pollutant (heavy metals) within the soil media used in this study. Whereby, it should correlates with the overall treatment performance of the CWs.
Figure 4.10: Concentration of (a) Fe and (b) Mn in soil at different depth
Figure 4.10 show a concentration of Fe and Mn in the soil media collected at different depth of all four (4) reactors (VC, VP, HC and HP). The unplanted control systems (VC and HC) exhibit a higher concentration of Fe and Mn in the soil samples collected at the bottom of the reactors, with an increase. While, the reactors planted with Limnocharis flava (VP and HP) exhibit a higher concentration of Fe and Mn in the soil samples collected at mid-depth of the reactors, with an increase of 3.856% (VC) and 6.602% (HP) for Fe and 0.226% (VP) and 0.309% (HP) for Mn. The increased concentration of heavy metals (Fe and Mn) in the soil samples collected at the bottom of unplanted control reactors (VC and HC) indicates that the heavy metals were actually precipitated towards the bottom of the reactors (Khan et al., 2009). While, the higher concentration of Fe and Mn at the middle of the planted reactors was due to the rhizofiltrations of these heavy metal in the rhizosphere since precipitation and rhizofiltration are the main mechanism in the removal of heavy metals in CWs (Khan et al., 2009)