Phytoremediation of contaminated ground water using Typha angustifolia

Organic pollutants in water are actually hazardous to public health. The treatment of organic pollutants from industrial wastewaters using traditional physico-chemical methods such as adsorption, oxidation and reduction and chemical precipitation are generally not economic at a larger scale (Yadav et al. 2012). Phytoremediation is a green technology that uses plants and their associated rhizobacteria to remediate soils, surface waters or ground waters from organic or inorganic contaminants (Lyubenova et al. 2013). Phytoremediation is a natural interaction technology between plants and microbes to remediate contaminated sites (Kabra et al. 2013). Lately, phytoremediation has been applied to remediate industrial wastewater contaminated by organochlorines, including chlorobenzenes, chlorophenols, chlorinated hydrocarbons and chlorinated olefins (Rouvière et al. 2012). The phytoremediation process was carried out for contaminated ground water with pilot scale where cattail plants are growing. Constructed wetlands (CWs) have been adopted as an alternative to the conventional wastewater treatment systems due to their economy in energy requirements, easy operation and maintenance (Chen et al. 2012). CWs are engineered wastewater treatment systems, in which different physico-chemical and biological processes contribute to the net elimination of contaminants from the wastewater, when wastewater passes through the media of a subsurface flow CW (Saeed & Sun 2011; Huang et al. 2012; Al-Baldawi et al. 2013a). CW can be classified into surface and subsurface-flow systems. In this study, sub-surface flow system was selected, in which water flows horizontally (horizontal subsurface flow CWs, HSSFCW) through the porous filter medium (usually sand and/or gravel) (Truu et al. 2009). The traditional CWs are considered for treating municipal, industrial and agricultural wastewater as cost-effective systems (Zhao et al. 2012). Use of aquatic macrophyte plant to assimilate and remove nutrients from water is a friendly environment phytoremediation approach (Xu et al. 2011). The study investigated the wastewater purification efficiency via nutrient removal and its relationships with the constructional and operational parameters of wetlands.

The decontamination performance of CWs is based on the interaction between plants, microbes and filter medium. The microbes are involved in mineralization of organic matter in both aerobic and anaerobic conditions. Nitrogen removal in CWs has mostly been assumed to be a result of the combination of nitrification–denitrification (Sundberg et al. 2007), but newly discovered pathways such as the anaerobic oxidation of ammonium could have potential significance in certain conditions as well (Paredes et al. 2007). The selection of suitable plant species has been shown to improve contaminant removal from conventional CWs and this seems an important opportunity to explore the role of selected plant in contaminant removal efficiency (Gagnon et al. 2012). According to Vymazal (2011), the common macrophyte species were often used in horizontal subsurface flow system CWs: Phragmites, Typha, Scirpus, Phalaris arundinacea, and Iris species. To enhance the pollutant removal efficiency of a subsurface flow wetland, recent studies have investigated the effects of major environmental and operation factors and alternative arrangement of wetland media in multiple stage wetland systems (Sundberg et al. 2007; Saeed & Sun 2012). The aim of this study was to assess the impact of plant presence as well as the performance of different substrate matrix form on the phytoremediation efficiency. In this study, in order to enhance the contaminant removal efficiency of a subsurface flow wetland, two different arrangements of wetland media have been examined.

Four mesocosm-scale CWs were set up in open air inside a greenhouse at the Universiti Kebangsaan Malaysia (UKM), which has a tropical climate, with a temperature ranges between 25 and 33 °C (Figure 1(a)). Each mesocosm was with square shape (height: 0.6 m; width and length: 0.9 m) representing industry treatment wetlands and planted with 30 common cattail (Typha angustifolia). Three of the mesocoms (W1, W3 and W4) composed of three filter layers from the top layer to the bottom layer: sand (⁠ 1–2 mm) with two layers of different granular sizes (⁠ 10–20 mm; 20–50 mm); the fourth one (W2) composed of mixture of soil, sand and compost (3:2:1) on the top layer and different granular sizes for another two layers below, similar to the rest mesocoms (Figure 1(b)) to investigate the effect of compost on contaminant removal. W3 was set up as a contaminant control that contained only contaminated water without plants to investigate any effect of microbial activity and sand on contaminant removal. In addition, W4 acted as a plant control in which plants were only exposed to tap water to compare the plant growth in normal condition with the ones in other systems that were exposed to contaminated water. The tested water in this experiment was obtained from an organichlorine contaminated groundwater area in Peninsular-Malaysia. All mesocosm were operated batchwise with a capacity of 200 L and was made of fiber-glass. The height of wastewater in the CW was maintained within the sand layer surface (up to 50 cm height of the CW) to simulate a sub-surface flow system as normally used in a CW (USEPA 2000).

Wastewater quality parameters, including (temperature, pH, dissolved oxygen concentration (DO), Redox potential (ORP), chemical oxygen demand (COD), total suspended solids (TSS), electrical conductivity, nitrates (NO₃-N), ammonium nitrogen (NH₄-N) and Turbidity were recorded every week. Table 1 shows the groundwater characteristics before proceeded for phytoremediation. The analyses were done directly after sampling. Analyses for all parameters were conducted according to standard methods for the examination of water and wastewater (APHA (American Public Health Association) 2001). COD was analyzed using commercial kits (HACH COD solution digestive with low measurement range of 3–150 mg/L) complying with DR3900-Gerrmany. TSS were determined by filtering 50 mL samples of wastewater through pre-weighed glass-fibers with 0.45 μm pore size, then dried at 105 °C before weighing the dried filters (NFT90–105). The concentrations of COD, and were measured by a reactor digestion method (Method 8000), Nesslerization method (Method 8038) and Cadmium reduction method (Method 8039), respectively, using a HACH DR3900-Gerrmany spectrophotometer.

T. angustifolia in seedling were transplanted from a pound in Dengkil, Malaysia. The same size samples (50 ± 5 cm in total plant length) were selected and planted in the mesocoms at the greenhouse. Each sampling day, duplicate plants were harvested to record stem height and root length to observe the growth progress.

The microbial population was counted in this study for the roots and the sand attached to the rhizosphere zone. The serial dilution method was used same as used by Al-Baldawi et al. (2013a) to evaluate the availability of microbial through 95 days of treatment.

Experimental results were statistically assessed using the software SPSS version 16 (SPSS Inc., USA). Differences were considered significant when p < 0.05.

Table 2 summarized the physical conditions through assessment of mesocosm wetlands. The low pH values of the ground water in treatments of W1, W2 and W3 (Table 2) was due to acidic of the raw ground water prior to treatment. Minimum pH value was between 4.47 and 4.62 for treatment wetlands while for mesocosm of plant control was 5.9. The results were shown after 95 days of treatment with Typha angustifolia in which the pH increased during phytoremediation process. Table 2 summarizes DO concentrations and ROP conditions in the four mesocoses and they ranged between 0.79–3.71 mg/L and −59–129 mV, respectively. It is worth mentioning there is no significant different between mesocosm wetlands in DO concentrations (p > 0.05) while there is significant different for ROP conditions (p < 0.05). The DO concentrations and ROP conditions are important for microbiological to enhance degradation process (Hijosa-Valsero et al. 2010). According to Hijosa-Valsero et al. (2010), the removal efficiency of organic contaminants in CW was more effectively with higher ROP which is associated to oxygen and electron acceptor concentrations in treatment medium.

The nitrate concentrations ranged between (W1: 2.95–0.6, W2: 2.73–0.4 and W3: 2.83–0.8 mg/L) in the three mesocosms CWs unit through 95 days of treatment (Figure 2). Phytoremediation using Typha angustifolia enhanced denitrification with nitrate removal efficiency of 79.6%, 85.5% and 71.8% for wetlands unit W1, W2 and W3, respectively. Owing to adequate rhizosphere zone for growth of nitrifying bacteria which had enhanced decreases in nitrate concentration. As presented, cattail of Typha angustifolia supported the removal efficiency.

The gradual accumulation of NH₄-N detected throughout 95 days of treatment especially in W2 may due to some of the processes that occur in the wetlands. At the end of the operation period, the NH₄-N concentration gradual accumulation detected in the effluent (W1: 3.5–5.3, W2: 1.26–44.8 and W3: 2.82–3.47 mg/L) may be clarified by some of the developments that occur in these wetlands (Figure 3). The high accumulation is 97.2%, which was occurring with W2 (soil + sand + compost) and the increase lesser with W1: 33.9% and W3: 18.9%. The reason for NH₄-N increase is the process of ammonification (mineralization) due to the organic matter from the bacteria biomass. In addition, it may be due to the reduction of NO₃-N to NH₄-N (nitrate ammonification) which is achieved by fermenting bacteria under anaerobic conditions that do not depend on the presence of NO₃-N to grow (Vymazal 2007). In this study, the accumulation of NH₄-N can be observed in the effluent due to change of NO₃-N to NH₄-N obviously with decreasing ROP in systems to anaerobic condition (Vymazal & Kröpfelová 2011). In CWs, the NH₄-N removal is done not only by bacteria, but as well by plant uptake and substrate adsorption (Akratos & Tsihrintzis 2007). On the other hand, oxygen was a limiting factor in horizontal subsurface CWs for enhancing nitrification, due to larger consumption of the bounded available oxygen due to faster organics degradation.

As summarized in Table 3, all the treatment systems were able to significantly reduce TSS and COD, with range between 73.3–95.5% and 35.2–86% for all systems, respectively. Much of the TSS and COD removal occurred in the first 28 days (Figures 4 and 5). There was statistically significant difference of TSS removal values between the three treatments stages (ANOVA: F = 4.964, p = 0.01). Post hoc comparison using the Duncan test indicated that TSS removal of the control contaminant (W3) was significantly greater than that of the CWs (W1 and W2) (Table 3). Gikas & Tsihrintzis (2012) concluded that TSS were removed to 66.8% in the settling tank while 41.7% in the CWs. Konnerup et al. (2009) found that the suspended solids are mainly removed by a physical filtration mechanism and settling which find no difference in removal of TSS between planted and unplanted CWs. Typha angustifolia was able to reduce the COD of groundwater as illustrated in Figure 4 for the W1, W2 and W3 wetland units. As shown generally, for the three treatments tanks (W1, W2 and W3), COD concentrations decreases with high removal rates of 81.2%, 86% and 71.5%, respectively. The results showed that there is no significant difference between treatment W2 (soil + sand + compost) and W1 (sand only) (p > 0.05) but there is significant differences when comparing the two treatments with control contaminant (W3) (p < 0.05). The removal in wetlands was due to the sedimentation of suspended solids and by rapid degeneracy processes in the water and upper soil layers (Verhoeven & Meuleman 1999). The matrix of sand, small gravel and big gravel with the helophyte roots and rhizomes are together configuring a layer of attached micro-organisms (biofilm) (Rousseau et al. 2004) which enhance COD removal. The application of infiltration wetlands in Europe to treatment domestic wastewater has shown that systems able to achieve 80% COD removal (Verhoeven & Meuleman 1999).

Typha angustifolia grew very well in the control mesocoms unit (W4) without contaminants than the mesocoms units with contaminants (W1 and W2). The height of the cattails in the two treatments of mesocoms was reduced after day 28 and 72, respectively, than that in the plant control (W4) due to the high concentration of organic contaminants in the ground water. Figure 6 shows the increase of stem height and root length through 95 days and clearly observed plant growth significantly faster (p < 0.001) in treatment W2 (soild + sand + compost) than in treatment W1 (sand only) during the 95 days period. Throughout the exposure period of 95 days, regarding mesocom W2, plant was growing well and new plant grow, while in W1 the plant inhibited after 28 days of exposure, due to the insufficient nutrients available.

The microbial population in T. angustifolia zone was evaluated at different treatments CW configuration (W1, W2, W3 and W4), as shown in Figure 7. It was found that the contaminated groundwater with organic pollutants enhanced the microbial population and increases its diversity. During the experiment, microbial populations in the contaminant control treatment (W3) was significantly lower than those in the treatment with T. angustifolia. As shown in Figure 7, the percentage of increase in microbial population is 6.6, 13.7, 1.7 and 10.2 for W1, W2, W3 and W4, respectively. The microbial population in W1 is lower compared to W4 due to the inhibition by the contaminated compounds in the tested water. The highest increment was in W2 treatment due to two factors, the used medium of sand, soil, and compost which enhanced the plant growth and high nutrient from wastewater. According to Saeed & Sun (2012), wetland matrix often plays an important role in contaminant removal from wastewater, which the medium provides attachment surfaces for microbial communities, and ingredients for biodegradation.

The results indicated that the removal of organic pollutant was enhanced by the presence of T. angustifolia. It is possible that presence of T. angustifolia roots with rhizobacteria had an advantageous effect on improving the organic pollutant degradation. The established design of mesocoms CWs (W1 and W2) to treat groundwater contaminated with volatile organic compound showed a similar performance in both units the for removal efficiency of COD and NO₃-N after 95 days of operation. The usage of sand only (W1) and sand, soil with addition compost (W2) was shown not different significant in organic nutrient removal. These shows the ability of the plant T. angustifolia to enhance removal efficacy in both medium but the biomass was better with medium of sand, soil with addition compost (W2) due to the availability of nutrient for plant growth.

The authors would like to thank Universiti Kebangsaan Malaysia (UKM-KK-03-FRGS0119–2010) and the Tasik Chini Research Centre for supporting this research project. They also acknowledge with thanks to Al-khwarizmi College of Engineering, University of Baghdad for support the first author.