Role of macrophyte and effect of supplementary aeration in up-flow constructed wetland-microbial fuel cell for simultaneous wastewater treatment and energy recovery

Accepted Manuscript Role of Macrophyte and Effect of Supplementary Aeration in Up-Flow Constructed Wetland-Microbial Fuel Cell for Simultaneous Wastewater Treatment and Energy Recovery Yoong-Ling Oon, Soon-An Ong, Li-Ngee Ho, Yee-Shian Wong, Farrah Aini Dahalan, Yoong-Sin Oon, Harvinder Kaur Lehl, Wei-Eng Thung, Noradiba Nordin PII: DOI: Reference:

S0960-8524(16)31493-6 http://dx.doi.org/10.1016/j.biortech.2016.10.079 BITE 17234

To appear in:

Bioresource Technology

Received Date: Revised Date: Accepted Date:

30 August 2016 22 October 2016 25 October 2016

Please cite this article as: Oon, Y-L., Ong, S-A., Ho, L-N., Wong, Y-S., Aini Dahalan, F., Oon, Y-S., Kaur Lehl, H., Thung, W-E., Nordin, N., Role of Macrophyte and Effect of Supplementary Aeration in Up-Flow Constructed Wetland-Microbial Fuel Cell for Simultaneous Wastewater Treatment and Energy Recovery, Bioresource Technology (2016), doi: http://dx.doi.org/10.1016/j.biortech.2016.10.079

This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and review of the resulting proof before it is published in its final form. Please note that during the production process errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.

Title: Role of Macrophyte and Effect of Supplementary Aeration in Up-Flow Constructed Wetland-Microbial Fuel Cell for Simultaneous Wastewater Treatment and Energy Recovery

Authors: Yoong-Ling Oon1, Soon-An Ong*1, Li-Ngee Ho 2, Yee-Shian Wong1, Farrah Aini Dahalan1, Yoong-Sin Oon1, Harvinder Kaur Lehl1, Wei-Eng Thung 1, Noradiba Nordin2

Address: 1

School of Environmental Engineering, Universiti Malaysia Perlis, 02600, Arau, Perlis,

Malaysia 2

School of Materials Engineering, Universiti Malaysia Perlis, 02600, Arau, Perlis,

Malaysia

Corresponding author: Soon-An Ong

Corresponding author’s address: School of Environmental Engineering, University Malaysia Perlis, Perlis 02600, Malaysia

Email: [email protected]

Tel/Fax: +604-9798986

Role of Macrophyte and Effect of Supplementary Aeration in Up-Flow Constructed Wetland-Microbial Fuel Cell for Simultaneous Wastewater Treatment and Energy Recovery

Yoong-Ling Oon1, Soon-An Ong*1, Li-Ngee Ho 2, Yee-Shian Wong1, Farrah Aini Dahalan1, Yoong-Sin Oon1, Harvinder Kaur Lehl1, Wei-Eng Thung1, Noradiba Nordin2

1

School of Environmental Engineering, Universiti Malaysia Perlis, 02600, Arau, Perlis,

Malaysia 2

School of Materials Engineering, Universiti Malaysia Perlis, 02600, Arau, Perlis, Malaysia

*Corresponding author: Soon-An Ong (E-mail: [email protected])

ABSTRACT This study investigates the role of plant (E. nuttallii) and effect of supplementary aeration on wastewater treatment and bioelectricity generation in an up-flow constructed wetland-microbial fuel cell (UFCW-MFC). Aeration rates were varied from 1900 to 0 mL/min and a control reactor was operated without supplementary aeration. 600 mL/min was the optimum aeration flow rate to achieve highest energy recovery as the oxygen was sufficient to use as terminal electron acceptor for electric current generation. The maximum voltage output, power density, normalized energy recovery and Coulombic efficiency were 545.77±25 mV, 184.75±7.50 mW/m3, 204.49 W/kg COD, 1.29 W/m3, 10.28 %, respectively. The variation of air flow rate influenced the NO3- and NH4+ removal differently as nitrification and denitrification involved conflicting requirement. In terms of wastewater treatment performance, at 60 mL/min aeration rate, UFCW-MFC achieved 50 and 81 % of

NO3- and NH4+ removal, respectively. E. nuttallii enhanced nitrification by 17 % and significantly contributed to bioelectricity generation.

Keywords: Up-flow constructed wetland –microbial fuel cell, biological wastewater treatment, aeration, bioelectricity recovery, normalized energy recovery

1. INTRODUCTION Environmental pollution particularly water pollution and energy scarcity are two important challenges that our planet is facing nowadays (Xu et al., 2015). Water is required in energy production and at the same time, energy is needed in water extraction, treatment and distribution (Olsson, 2012). From the understanding of water-energy nexus and the need for clean water and energy, have driven the exploration of sustainable wastewater treatment technology to address both challenges simultaneously.

Constructed wetlands (CWs) appear to be a favourable wastewater treatment option worldwide due to the advantages like low cost in construction and maintenance as well as more sustainable. Lim et al., (2003) showed that it was possible to provide aerobic and anaerobic condition in a single unit sub-surface flow CW tank with 44 rhizomes/m2 plant density. However, there were also some single-stage CWs, which was unable to accomplish high total nitrogen removal (Vymazal, 2007). Therefore, artificial aeration was proposed for systems that lack aerobic condition and could not achieve good nitrogen removal. Higher aeration rate means the higher the consumption of energy. It would be ideal to provide optimum air flow rate to ensure minimum wastage. Aeration can be energy intensive; however, wastewater treatment plant can also be an energy producer.

Lately, microbial fuel cell (MFC) has gained large attention from researchers worldwide for its ability to achieve wastewater treatment and energy recovery. MFC is an emerging technology that can utilize wastewater as fuel for anode for electrical current generation via biocatalyst and this makes MFC a sustainable technology for waste management and merits with energy generation (Logan 2006; Srikanth and Venkata Mohan, 2012). MFC systems have been improving from time to time to achieve better performance in all possible ways. The development of MFC evolved from with the use of mediators, membranes and separators to mediator-less, membrane-less, without separator and the use of biocathode. These are some interesting components in a MFCs system, which are noteworthy, as they are more practical in real applications.

The desire for sustainable wastewater treatment technology has driven the development of microbial fuel cell integrated with constructed wetland, which is becoming a cutting edge hybrid technology (Yadav et al., 2012; Liu et al., 2014; Doherty et al., 2015a; Fang et al., 2015; Oon et al., 2016). This type of MFC based technology was derived from the similarity of both systems. CW possesses the natural stratified redox potential along the wetland bed, where aerobic condition can be developed at upper regions, while anaerobic microenvironment at bottom region. Similarly, in MFC system, the conditions also corresponded to the traits of CW, where anode and cathode can be strategically placed at anaerobic and aerobic conditions, respectively (Xu et al., 2015). CW-MFC studies found that the pollutant removal efficiencies improved compared to treatment by an individual unit (Srivastava et al., 2015). Anyway, CW-MFC studies are limited and still in its infancy, thus further researches on the configurations and optimizations are in great need. A wastewater treatment plant resembles a water-energy nexus, and in this study it is envisioned to serve as an energy producer to partial off-set the energy consumed by the system. The main objectives of this study are to investigate the role of wetland macrophytes and the effect of

supplementary aeration in wastewater treatment and bioelectricity generation performance by using up-flow constructed wetland-microbial fuel cell (UFCW-MFC) reactors. This study also hoped to provide a reference on efficient supplementary aeration to optimize the wastewater treatment and bioelectricity recovery in a cost effective manner.

2. MATERIALS AND METHODS 2.1 Bioreactor configurations and operation Two parallel UFCW-MFC bioreactors were fabricated and set up to study the effect of aeration and role of submerged plant on the performance of nitrification, denitrification, organic degradation and bioelectricity generation. Figure 1 shows the schematic diagram of the two bioreactors. The bioreactors were located at outdoor sheltered by veranda at average temperature of 28 ± 4ºC. Acrylic columns with diameter of 18 cm and height of 75 cm were used to construct the two bioreactors. Glass beats with an average diameter of 1 cm were filled from the bottom of the reactor up to 3 cm height of the reactor to ensure even distribution of wastewater influent. Gravel was used as supporting medium. Five sampling points (S1, S2, S3, S4 and S5) were designed along the wetland bed at 7, 21, 36, 51 and 66 cm from the bottom of the bioreactor, respectively (Oon et al., 2016).

[Figure 1 Schematic diagram of UFCW-MFC Reactor 1 and Reactor 2 (control)]

Activated carbon was used as the electrode in this study. Activated carbon was layered in between gravels and each layer of the anode and cathode electrode consists of 2544.69 cm3. The electrode set up was similar to Oon et al. (2016) to study the bioelectricity performance of the reactors. Three anodes were positioned at 8 cm, 23 cm and 38 cm; while, cathode was placed at 53 cm from the bottom of the reactor. Activated carbon and gravels

were immersed in the mixed cultures sludge for inoculation for a month ahead before reactors’ set up. The activated sludge was obtained from a local glove manufacturing factory’s wastewater treatment plant (Shorubber (M) Sdn. Bhd). Anode and cathode were connected by insulated copper wires across a 1000 Ω external resistant. Stainless steel and carbon rod were used as connects at the electrodes. Voltage output of the UFCW-MFC systems were monitored and recorded by using a data logger (Midi LOGGER GL820 GRAPHTEC).

Reactor 1 (R1) was planted with Elodea nuttallii (waterweed), which is a type of submerged wetland plant. The plant was obtained from Tasik Melati, Perlis. About 200 g of E. nuttallii was strategically placed at the cathodic region of R1. On the other hand, Reactor 2 (R2) was a control reactor, which was without plant and without supplementary aeration. Aeration sparger was installed below cathodic compartment, 45 cm from the bottom of the reactor, in order to produce aerobic microenvironment.

Concomitant aeration was provided to the

reactor. The airflow rates were regulated from 1900, 1200, 600, 60 and 0 mL/min by using an air flow meter with control valve (Model: LZB-3, China). The control valve at the air flow meters was used to adjust the air flow at a desired constant rate.

This study was operated for a total of 276 days. The bioreactors were operated for 70, 73, 46, 38, and 45 days during airflow rate of 1900, 1200, 600, 60 and 0 mL/min, respectively. It was observed that after 30 days of operation on each air flow rate, the voltage output appear stable.

Synthetic wastewater made up of 214.2 mg/L C6H5COONa, 409.8 mg/L CH3COONa, 176.1 mg/L NH4NO3, 7.0 mg/L, NaCl, 3.4 mg/L MgCl2 6H2O, 4.0 mg/L CaCl2·2H2O and 36.7 mg/L K2HPO4 were fed to the bioreactors by a peristaltic pump (Natong BL-100C, China) at flow rate of 4.048 mL/min. Air pump and peristaltic pump were connected to a

timer, which regulated to 3 h on followed by 0.5 h off cycles and maintained hydraulic retention time (HRT) of 1 day.

2.2 Physical and chemical analyses Wastewater samples from influent and all sampling points (S1-S5) were collected to assess the performance of wastewater treatment in the UFCW-MFC systems. The oxidation reduction potential (ORP), dissolved oxygen (DO) and pH were periodically measured by using an ORP probe with platinum electrode (Ag/AgCl reference) connected to an ORP meter (HANNA HI 8424, USA), DO meter (HANNA HI 9146, USA), and pH meter respectively. COD concentration was evaluated by using a HACH DR 2800 colorimeter and prior to analysis, water samples were centrifuged by using a tabletop low speed centrifuge at 4200 rpm (Cence L500, China) for 10 min. Nitrate and ammonium concentration in the wastewater along the bioreactor were analysed by using nitrate electrode and ammonium electrode connected to Bante instruments 920 Precision pH/ORP meter, China and Martini Instrument Mi 151 pH/ORP/Temperature Bench Meter, China, respectively to evaluate the performance of nitrification and denitrification in the systems.

2.3 Calculation The COD reduction efficiency can be calculated by using the formula: CODE = [(Cin-Cef)/Cin ] × 100 %

(1)

Where,Cin is the COD concentration of influent and Cef is the COD concentration of effluent (mg/L)

Current (I) in Amperes (A) was calculated by using Ohm’s law: I=V/R

(2)

Where, V (V) is cell voltage, which obtained from multimeter, R (Ω) is the external resistance.

Power (W) is calculated by using the formula as follows: P=I×V

(3)

Where, I is current (A) and V (V) is cell voltage.

Since the electrons used to power generation was generated in anode, the volumetric power density (Pd,W/m3) is calculated by the effective volume of the anode chamber VA (m3) ( Logan et al., 2006). Pd =V2 / VA

(4)

Coulombic Efficiency, CE (%) was calculated using the formula as follow: CE = M·I / F·q·n· ∆COD

(5)

Where, M is the molecular mass of O2 (32 g/mol); I is current (A); F is Faraday’s constant (C/mol), which is 96,485; q is flow rate (L/s) while n is number of electrons donated per mole of O2 (mol e-/mol O2), which is 4; ∆COD represents the change in COD between influent and sampling point S1; S1 and S2; S2 and S3 (g/L). Normalized Energy Recovery (NER) can be calculated based on two formulae as follows (Ge et al., 2014): NER based on organic removal: NERs = (power × time) / [COD (removed within time)] = (power)/ (wastewater flow rate × ∆COD) kWh/kg COD NER based on volume of water treated: NERv = (power × time) / [wastewater volume (treated within time)]

(6)

= (power)/ (wastewater flow rate) kWh/m3

(7)

3. RESULT AND DISCUSSION 3.1 Plant evaluation Water, microbes, media and vegetation are the four major components of CW (Shelef et al., 2013). Plants play several important roles in CW such as physical effect of root structure to provide filtering effect. Besides that, plants also increase the surface availability for the growth of microorganism biofilm (Brix, 1997). Microbes that are present in the CW are the main contributor in biodegradation treatment in CW. Gas and exudates produced by the plants are also a vital element of plant’s effect in CWs. Oxygen released by the root system could influence the redox potential (Białowiec et al., 2012), which can potentially enhance nitrification process and support heavy-metal sedimentation; however, Vymazal, (2011) found that the oxygen leakage from root is arguably limited in horizontal CW. Besides releasing oxygen, the root system of submerging plant will also release carbon compounds. The root exudates may act as a form of carbon source for denitrifiers and hence improve the removal of nitrate (Vymazal, 2011). In the study of Liu et al., (2013), the root exudates of Ipomoea aquatica was utilized as part of the fuel in photosynthetic MFC. Liu et al., (2016b) reported that wetland plants species have a major effect on the pollutant removal efficiencies as well as on the microbial communities. P. australis had better removal of NH4-N than I. pseudacorus, and can enhance nitrification process in the rhizosphere as a result of stronger radial oxygen loss (ROL). Furthermore, plants may uptake pollutant such as N, P and heavy metals; however, several studies reported that plant uptake is inappreciable (Brix, 1994; Vymazal, 2011).

Table 2 presents the plant species and aeration conditions in CW-MFC studies and the treatment performance of the systems. In the previous studies, emergent plant e.g. cattail and common reed were commonly employed in CW-MFC studies; meanwhile, in this study, Elodea nuttallii (waterweed) was used as the model of submerged macrophyte. The plant was placed at the cathode region of the reactor to provide oxygen to cathode and facilitate nitrification process of the treatment system. The initial plant mass was about 200 g. The growth of the submerged macrophytes was only observed from physical observation. The plant showed positive growth over time as the amount of plant was abundant. Plant mass was not measured as removing the plant from the reactor could disturb the operating conditions. The main reason E. nuttallii was chosen because photosynthesis process can be carried out under water and the oxygen produced will be released to the immediate surrounding (cathode) and remain in the medium (GISD 2010). Macrophytes particularly emergent plants can cause significant water loss in CW through evapotranspiration. The treatment efficiency in CWs could be affected as the volume of wastewater flows through the system declines due to water loss especially when the evapotranspiration rate exceeds 2.5 mm/d (Biolowiec et al., 2014). Since the plant was fully submerged in water; the water loss in 1 day HRT due to evapotranspiration was low (3 %) compared to Oon et al., (2016) with 13 %, which planted with emergent plant, cattail. Moreover, this submerged macrophyte possesses high capability to withstand the cold weather and low light level, ability to survive in the high organic and nutrient loading environments, and ability as a surface for microbial attachment (White and Bishop 1984). It is also known for its effectiveness in nitrogen and phosphorous removal in wastewater during summer (Bishop and Eighmy, 1989).

The role of the plant in the present study was evaluated from wastewater treatment and bioelectricity generation performance and presented in the following sub-sections.

[Table 2 Comparisons of COD removal efficiency and Columbic efficiency (CE) of CWMFC studies with different plant species and aeration conditions.]

3.2 Wastewater treatment performances 3.2.1 Dissolved oxygen (DO) and oxidation reduction potential (ORP) Dissolved oxygen (DO) is one of the core factors that directly affect the effectiveness of pollutant removal in CW (Liu et al., 2016a). From the perspective of MFC, oxygen is used as the terminal electron acceptor for bioelectricity generation and at the same time redox gradient, which creates a potential difference that would greatly influence the performance of energy recovery in the MFC system (Gil et al., 2003). Redox potential and DO are interrelated and are important monitoring parameters in CW-MFC study. In order to study the influence of aeration flow rate on the wastewater treatment and bioelectricity performance, aeration flow rates were varied from 1900 to 0 mL/min. R1 bioreactor was then set to dark condition to evaluate the role of plant in the system. Figure 2 exhibits the full profile of DO and ORP at various air flow rates. Redox potential dynamics were developed along the UFCW-MFC reactor when R1 was supplied with supplementary aeration.

Generally, the DO profile was corresponding to the ORP profile. At high aeration flow rate (1900 and 1200 mL/min), the cathodic region was highly aerobic. The DO at S3 was about 1.0 mg/L as air spargers were located above S3. The strong artificial aeration caused oxygen to diffuse into S3 region. When the aeration flow rate was reduced to 600 mL/min, such phenomena did not occur. This indicates that oxygen diffusion was not significant at air flow rates below 600 mL/min, and it would not change the microenvironment of S3, which was intended to be in anaerobic/anoxic condition. The DO at the cathodic region was mainly used for major processes such as chemical reduction during

the half reaction of MFC, organic matter biodegradation and nitrification. When a comparison is made between the DO of current study planted with E. nuttallii and Oon et al., (2016) planted with T. latifolia, it appears that the DO of current study (at S4 and S5) was 12 – 14 % higher during the provision of artificial aeration at rate of 1900 mL/min. This indicates that E. nuttallii could produce more oxygen than T. latifolia.

The DO at S4 and S5 reduced significantly to 0.44–0.45 mg/L, when artificial aeration was stopped. In the absence of artificial aeration, the oxygen supply mainly relied on the submerged plant; the oxygen produced by the plant was used for biodegradation, nitrification and as electron acceptor, thus the DO level was low. When R1 was operated in the absence of sunlight, the DO was even lower than when it was exposed to sunlight as photosynthesis could not happen. Insufficiency of oxygen commonly occur in conventional CW systems, therefore, supplementary artificial aeration seems to be the most effective alternative to overcome the challenge (Guo et al., 2016). Excessive supply of artificial aeration is not only a form of energy wastage, but could also inhibit denitrification process. Guo et al., (2016) demonstrated that ammonium removal efficiency (90 % (18.1 g/(m2·d)) was greatly enhanced from intermittent aeration (1 h on: 1 h off); however, TN removal efficiency (53 %) was restricted.

The role of plant was evaluated and verified by comparing R1 with R2 (control) when aeration was not provided to R1, and in the absence of sunlight (dark condition). Supplementary artificial aeration out-performed R2 and R1 without artificial aeration in supplying DO at the cathodic region. When R1 was operated without the supply of artificial aeration, the DO of R1 was higher compared to R2 as a result of the presence of submerged macrophytes. During day time, the submerged plants were able to carry out photosynthesis when exposed to sunlight. During photosynthesis, plants use the energy of sunlight to convert

water and carbon dioxide into chemical energy (glucose), and the by product is gaseous O2, which was diffused by plant cells into the immediate surroundings, hence contributing to the DO in R1. Photosynthesis process equation: 6 CO2 + 6 H2O → C6 H12 O6 + 6 O2 On the other hand, in the absence of light, plant cell and microorganism respiration will take place and consume O2. Hence, the DO level in the reactor was reduced as the DO consumption was more than production. The submerged plant’s photosynthesis and respiration altered the oxygen dynamics of the reactor. Supplementary aeration, aeration flow rate and the presence of macrophytes could influence the oxygen dynamics significantly, which can consequently affect the nitrogen transformation and removal of the system (Maltais-Landry et al., 2009). Even though supplementary aeration can effectively enhance the treatment performance; inevitably, it requires energy input, which increases the cost of life cycle. Hence, by optimizing artificial aeration, it would ensure an effective and sustainable design of CW (Wu et al., 2016).

[Fig. 2. DO monitoring (a) and ORP monitoring (b) of R1 and R2 at various aeration flow rates]

3.2.2 Ammonium and nitrate removal Wastewater treatment performance of R1 and R2 during various aeration flow rates and different conditions were summarized and presented in Table 1. Generally, the aerated bioreactor outperformed the non-aerated bioreactor particularly in nitrification. The removal rates for nitrogen in conventional CW was low due to the conflicting requirement of DO for nitrification and denitrification processes in a single wetland unit (Liu et al., 2013a). In order

to overcome such challenges, supplementary intermittent aeration was proposed, which is a reliable alternative to improve the treatment performance (Wu et al., 2016).

The two oxidization steps involved in nitrification are as follows (Tchbanoglous 1991): NH4 + + 3/2 O2 (nitrosomonas)
NO2- + H2O + 2 H+ ; NO2- + 1/2 O2 (nitrobacter)
NO3Ammonium oxidizing bacteria (AOB) nitrosomonas are responsible for the oxidation of ammonium (NH4+) to nitrite (NO2-), then followed by oxidation of the intermediate (NO2- ) to nitrate (NO3-) by nitrobacter species.

Figure 3 (a) shows that ammonium removal efficiencies improved when supplementary aeration was provided. Aeration rate of 1900 mL/min greatly increased the DO at sampling point S4 and above, thus it possesses the highest oxidizing ability for removing NH4+. The supplementary aeration 1900, 1200, 600, 60 mL/min increased the DO to 3.52-4.50, 3.17-4.38, 2.28-3.27,0.56-1.38 mg/L and corresponded to the NH4+ removal of 97, 96, 84, 81 %, respectively. Significant difference was also observed between aerated and non-aerated bioreactor R1. The nitrification efficiency dropped to 47 %, when supplementary aeration was not provided at the cathodic region. The supplementary aeration was located below the cathodic region (at 45 cm from the bottom of the reactor). Low level of DO was observed at cathode (S4), when artificial aeration was terminated. The DO provided by the macrophytes during photosynthesis was used by nitrifiers for nitrification process, biodegradation of organic matter, and also as terminal electron acceptor for electrical current generation. As a result, the NH4+ removal efficiency of UFCW-MFC system was relatively low without supplementary aeration.

The result also showed that planted bioreactor performed better than the control reactor (unplanted). The NH4 + removal efficiency for unplanted control reactor R2 was only 30 %, which was 8 % lower than non-aerated R1. The performance of nitrification R1 at dark condition was slightly better than R2 as the submerged plant was unable to carry out photosynthesis to produce oxygen but consumed oxygen during respiration, consequently, it could not contribute to nitrification. Anaerobic ammonium oxidation (Anammox) was identified as an alternative N removal pathway, which converts ammonium to N2 in anaerobic condition, without requiring carbon. There was about 12 % of ammonium removal at S1, which was in anaerobic condition. It could be due to anammox as the carbon source at S1 was low. Hu et al., (2016) reported that anammox contributed to 55.6-60.0 % of N removal in an integrated vertical-flow CW system. However, anammox was not obvious in the present study.

Microbial denitrification is another major step of N transformation and removal. Denitrifiers such as Pseudomonas species are responsible for the reduction of NO3- to N2 gas (NO3-
NO2-
N2O
N2), which happens in anaerobic/anoxic condition. As shown in Fig. 3 (b), non-aerated R1 and R2 exhibits greater NO3- removal efficiencies than the aerated ones. Non-aerated R1, dark condition R1 and control reactor R2 favoured denitrification process and enabled the system to achieve NO3- removal of 93, 97 and 98 %, respectively. NO3-N removal can proceed and be removed permanently via anaerobic microbial denitrification (Maltais-Landry et al., 2009). Denitrification also occurred at the anodic regions (from S3 and below) of R1, which was in anoxic/anaerobic condition. When supplementary aeration was provided to R1, the system was able to remove 99 % of NO3- up to S3 only. While for airflow rate 1900 mL/min, the removal efficiency of NO3- was 96 %, due to oxygen diffusion into S3. The concentration of NO3- increased at S4 and S5 due to the conversion of NH4+ to NO3- from nitrification process. This result corresponded to the decrease of NH4+

concentration in Fig. 3 (a). Denitrification was restricted when artificial aeration was supplied to the system. The removal efficiency of NO3- at the effluent S5 during air flow rates of 1900, 1200, 600, 60 mL/min were 36, 39, 44 and 50 %, respectively. In terms of wastewater treatment performance, with low aeration rate (60 mL/min) 50 % and 81 % of nitrate and ammonium removal were obtained in a single reactor.

The removal efficiency of NH4 + in the current study during aeration flow rate of 1900 mL/min was 97 %, which was slightly higher than the previous study Oon et al., (2016) (96 %). On the other hand, NO3- removal efficiency was 10 % lower than the previous study. From the comparison, the different in nitrate and ammonium removal efficiencies could be due to the plant species employed by the two systems. E. nuttallii that used in the current study could uptake nutrient. Ozimek et al., (1993) found that E. nuttallii prefers NH4+ over NO3- uptake when both ions present in the water at same concentration. Besides that, in the presence of plants and artificial aeration, current study demonstrates that reactor planted with E. nuttellii had higher DO. Such condition induced the growth of nitrifiers and was more conducive for nitrification; meanwhile, it inhibited denitrification process. Nevertheless, the TN removal of Oon et al., (2016) was higher than the current study, which could relate to the higher biomass of the macrophyte. In the study of Tanner, (1996), TN removal showed significant linear relationship to plant biomass. Brix, (1997) reported that the capacity of nutrient uptake by emergent plant was higher than submerged plant; however, the amount of nutrient that removed by plant uptake is generally inappreciable. The metabolism of macrophytes such as oxygen release and plant uptake could influence the treatment processes to different extents depending on the design (Brix, 1997). This study also concluded that nitrogen removal pathways in UFCW-MFC system were mainly nitrification and denitrification.

[Fig 3. Nitrate (a) and ammonium (b) removal profile of R1 and R2 at various aeration flow rates]

[Table 1 Wastewater treatment performance of UFCW-MFC]

3.2.3 Organic matter degradation The average COD concentration of influent synthetic wastewater was 643 mg/L and the UFCW-MFC reactors were operated under 1 day HRT. Figure 3 (a) shows COD removal profile of UFCW-MFC R1 and R2 and the wastewater treatment performance was summarized in Table 1. It can be seen that there was a sharp decline of COD from influent to S1 and followed by a steady decline from sampling point S1 to S3, which contributed up to 96 % of COD reduction at the anodic region. Carbon sources were mostly oxidized by biocatalysts (anaerobic microbes) and will contribute to bioelectricity generation. Anaerobic microbes contributed to the COD reduction by oxidizing organic substrate as fuel to support microbial activity. The unutilised carbon sources were further degraded at S4 and S5. It is important for the system to achieve high COD removal at anode to prevent high organic matter from entering the cathodic region as this could deteriorate the aerobic condition of cathodic region and affect the redox potential and voltage output (Villaseñor et al., 2013). The COD removal efficiency at anaerobic regions (S1-S3) improved over time, which could be due to the growth of anaerobic microbes. Aerated R1 achieved 98-99 % of COD removal, while non-aerated R1 and control R2 achieved 98% and 97 %, respectively. The COD removal efficiency between aerated and non-aerated systems was not obvious. Artificial aeration did not appear to facilitate biodegradation of organic matter, since most of the organic matter was degraded at S1-S3.

It was found that the COD removal efficiency at the anodic region (up to sampling point S3) of the present study (R1- 1900 mL/min) was 37 % better than Oon et al., (2015). In this study, activated carbon was used as the electrode material instead of carbon felt. Besides providing larger surface availability for biofilm attachment, it also protects them from shock loading of toxic and inhibitory substances (Sublette et al., 1982). Bioregeneration that takes place at activated carbon could also enhance the biological treatment process (Aktaş and Çeçen, 2007). The role of artificial aeration and role of macrophyte may be seen if higher organic loading was provided. Organic loading can be increased in the future. Many studies showed that artificial aeration had significant impact on the reduction of organic matter, as the supplementary aeration in the aerobic zone of the reactor could enhance the activities of aerobic bacteria and facilitate the biodegradation of organic matter (Ong et al., 2010; Zhao et al., 2013; Headley et al., 2013). From Fig. 4 (b), the long-term operation inferred that the system was stable and UFCW-MFC possesses great potential in treating high strength organic wastewater.

[Fig 4. COD removal profile of R1 and R2 (a) and time course monitoring of R1 (b) at various aeration flow rates]

3.3 Bioelectricity Generation 3.3.1 Voltage output and power densities Oxygen, which is a strong electron acceptor, was used as the terminal electron acceptor in this study. In order to study the effect of aeration flow rate on the performance of bioelectricity generation, the aeration flow rates were varied from 1900 to 0 mL/min. The use of catalyst (such as platinum) in the cathodic compartment is known to help in electron transfer from cathode to oxygen and lower the activation overpotential. However, applying

such precious metal as cathode will certainly increase the cost and more importantly, it is not economically viable when up scaling to real application. Studies also showed that microbes could be used as catalyst and/or mediator in the cathode chamber (Rosenbaum et al., 2011), and these biocatalysts could retrieve electrons directly from cathode and then transfer to terminal electron acceptors such as oxygen, sulfer, nitrogen etc. (Clauwaert et al., 2007). Srikanth and Venkata Mohan, (2012) successfully demonstrated that oxygen reduction on the cathode was directly done by the biofilm and the concept was clearly portrayed from bioelectro kinetic analysis. Therefore, biocathode activated carbon was used as the electrode, and oxygen, which is a strong electron acceptor and available in abundance was used as the terminal electron in the present study. Three anode electrodes (A1, A2, and A3) were developed in the UFCW-MFC system to evaluate the effect of electrode spacing on the performance of UFCW-MFC in electricity generation. Distance between A1 to cathode, A2 to cathode, and A3 to cathode were specified as 15 cm, 30 cm and 45 cm, respectively. Performance of UFCW-MFC system in terms of bioelectricity generation was observed in voltage output, power density, Coulombic efficiency (CE) and normalized energy recovery (NER). These results were summarized in Table 3.

Figure 5 (a) and (b) depict the voltage output from each anode connected to cathode across a 1000 Ω resistor in R1 and R2. Highest voltage output was observed between anode 1 (A1) and cathode (15 cm). This can be due to the small electrode spacing between anode and cathode, which consequently led to lower internal resistance (Liang et al., 2007). The voltage output between A1-C increased from the beginning of aeration flow rate 1900 to 600 mL/min. At aeration flow rate of 1900 mL/min, DO at the cathodic region was the highest (Fig 2); the system was expected to generate higher voltage at A1-C. However, due to the higher aeration flow rate, oxygen diffusion occurred at A1, which is located just below the air sparger. When the aeration flow rate was reduced to 600 mL/min, the average voltage output was the highest

(545.77±25 mV). This could be due to sufficient oxygen supply at the cathodic region. Oxygen was used as terminal electron acceptor and at the same time air from aeration supplied did not diffuse to A1, thus the condition of A1 remained reductive; while the cathodic region was oxidative. Therefore, higher (optimum) voltage output was obtained during 600 mL/min aeration flow rate (A1: 546±25 mV; A2: 473±24 mV; A3: 442±22 mV). Voltage output did not vary significantly during airflow rate of 1900, 1200 and 600 mL/min.

From the bioelectricity generation performance, aeration flow rate of 1900 and 1200 mL/min can be regarded as excess of aeration. From the COD removal efficiency profiles, COD removal efficiencies during all air flow rates were rather consistent. The voltage output was less dependent on the electrons and protons transferred from carbon sources reduction at anode, but mostly attributed to the ability of oxygen reduction reaction at the cathode. Voltage output decreased from day 144 onwards, with the decrease of aeration flow rate from 600 to 0 mL/min. The reduced aeration flow rates showed lower DO at the cathodic region, which deprived the system of oxygen as terminal electron acceptor for electricity current generation. Larminie and Dicks (2000) proposed that the electron transfer efficacy tends to increase with the rise of electron acceptor (O2) concentration. The drop in voltage during the reduction of supplementary aeration flow rate was due to the decline of DO concentration, which hindered electron transfer from anode to cathode.

The voltage output of current study during aeration flow rate of 1900 mL/min was 478 % higher when compared with Oon et al., (2016). The higher voltage obtained was corresponding to the higher DO level in the system. This suggests that E. nuttallii could provide more oxygen to the cathodic region than T. latifolia. Since there were more terminal electron acceptors (O2) available at the cathode, more O2 could be accepted for electrical current generation and consequently contributed to higher voltage output.

Even when the aeration of R1 was terminated, voltage output still maintained around 260 mV despite the low DO. Plant will produce more oxygen when the surrounding DO is not in excess or surpass the DO production rate of the plant. The role of macrophyte in contributing to bioelectricity generation was clearly observed by comparing R1 to the control system R2. In this study, nitrate was almost completely removed at S1. Thus, in the absence of oxygen and nitrate; there were no electron acceptors at cathode to receive electron transfer from anode via external circuit, which resulted in poor voltage output. The voltage output for control reactor was very low (0-7 mV) due to both anodic and cathodic regions being in oxidative condition. The cathode condition of the control reactor was very different with R1. The cathodic region of the control reactor was in oxidative condition, which was similar to the condition at the anodic region. This phenomenon occurred because both anode and cathode were undergoing oxidation process, where there were no significant potential between the two electrodes, therefore, very low electrical current was generated. As compared to R1, in R2 there are no electron acceptors (such as O2, NO3-) available at cathode. Srikanth and Venkata Mohan, (2012) also suggested that the ability of electrons to transfer from anode to cathode can be influenced by the absence of terminal electron acceptor conditions and consequently cause inappreciable electrogenesis.

The highest power density obtained was 184.75±7.50 mW/m3 during 600 mL/min (Fig 5(c)). This can be ascribed to sufficient oxygen as electron acceptor. The power densities of 1900 and 1200 mL/min (170.64±5.84 mW/m3, 174.11±3.39 mW/m3) were slightly lower than 600 mL/min as a result of oxygen diffusion into A1. When R1 was switched to nonaerated condition, the power density dropped to 63.68 mW/m3, which was 99.5 % higher compared to R2 (0.34 mW/m3).

Based on the polarization curve, internal resistance between A1, A2 and A3 connected to cathode were identified and recorded in Table 3. The smallest electrode spacing had the lowest internal resistance. The internal resistance of A1 during aeration flow rates 1900- 600 mL/min were 200 Ω, which was lower than the internal resistances of A1 during 60 and 0 ml/min, which were 300 Ω. As the electrode spacing increases, the internal resistance between cathode and anode (A2 and A3) increased. The higher internal resistance reduced the power output of the system; thus, lowering the power density. The maximum power density in the present study was comparable to other similar systems with the same flow regime such as Doherty et al., (2015a) with 168 W/m3 (Table 4). However, the maximum power density was lower than Doherty et al., (2015a), which employed a simultaneous upflow–downflow regime, with smaller electrode spacing (10 cm) and higher organic loading (268 mW/ m3).

[Fig 5. Voltage output of R1 (a), voltage output of R2 (b), power density-current density of R1 (c), NER and CE of R1 and R2 (d) at various aeration flow rates]

[Table 3 Bioelectricity performance of UFCW-MFC]

[Table 4 Performance comparison of CW-MFC studies]

3.3.2 Coulombic efficiency (CE) and Normalized energy recovery (NER) CE is a key parameter to evaluate energy recovery efficiency of MFC technology. The CEs of R1 and R2 (control) were presented in Table 3 and Fig 5 (d). In terms of effect of electrode spacing, the CE and NER results (Table 3) showed similar relation trends. However, in terms of effect aeration flow rates, NERs of A1 at different aeration flow rates were distinctly different compared to CE and NERv (Fig 5 (d)). Vast difference was also observed

in CE between non aerated R1 and control R2, which indicated that there was contribution of macrophytes towards bioenergy recovery. Among the three anodes, the smallest electrode spacing (15 cm) (between A1 and C) obtained the maximum CE as the COD at that region was low and from there more organic matter were able to convert into electrical current. This shows that CE could be improved with smaller electrode spacing (Cheng et al., 2006). The maximum CE obtained from various aeration flow rates were 0.08-10.28 %, which was during 600 mL/m, which suggests that this flow rate can be a reasonable choice. From the CE results, non-aerated R1 was comparable with CE during aeration flow rate of 1200 mL/min. Generally, the CE obtained by UFCW-MFC was low, which is an identified characteristic of such hybrid system.

Previous studies also showed low contribution of MFC within the hybrid system in terms of CE (Table 2). Besides that, the comparison in Table 2 also illustrates that CEs obtained in present study were relatively higher compared to previous studies with similar configurations. From the literatures, it reflects that higher concentration of COD caused low CE. In most CW-MFC studies, only approximately 1 % of the COD removed contributed to electrical current and such set back need to be addressed. According to Liu et al. (2014), increase in current generation was not proportional to the increase in substrate concentration. Thus, it is suggested that the UFCW-MFC is more efficient in electricity generation under low organic loading rate. As reported by Doherty et al., (2015b), bacteria also utilize electrons to breakdown complex compounds to short-chain organics that can be oxidized by electrogenic bacteria. This process reduced the availability of electrons for electricity generation; hence, low CE was obtained in the study. Logan et al. (2006) identified that low CE could be due to the competitive processes and bacterial growth. The competition between electrogenic bacteria and other microorganism for organic matter oxidation as well as other processes e.g. fermentation and/or methanogenesis, resulted in lesser electrons transfer to the

anode. Since mixed culture sludge was used for inoculation in this study, it is believed that mass diversity microorganisms exist in the system, which resulted in competitive processes, and lowered the CE.

NER is a rather new parameter to present and compare the energy performance data in MFC studies (Ge et al., 2014). NER can be used to analyse the energy recovery based on the volume of treated wastewater (NERv: kWh/m3) and the removal of organic substrates (NERs: kWh/kg COD). The results of NER were presented in Table 3. Since NER is expressed in kWh and it is commonly used by the wastewater industry; a more effective communication between academic research and the industry would be established. Moreover, NER involves wastewater characteristics (wastewater flow rate and organic removal efficiency); regardless of the dimension or scale, hence energy performance can be analysed from a different approach.

In terms of effect of electrode spacing on CE and NER, the CE and NER results showed similar relation trend; however, in terms of effect of aeration flow rates, NERs at A1 at various aeration flow rates were distinctly different compared to CE and NERv (Fig. 5 (d)). The maximum NERs recorded in this study was 10.28 Wh/kg COD (600 mL/min); while maximum NERv obtained was 1.29 Wh/m3 (1200 mL/min), which was comparable with NERv at 600 mL/min (1.23 Wh/m3). The NERs at lower aeration (600, 60 and 0 mL/min) were higher than at 1900 and 1200 mL/min. The results suggest that the oxygen reduction rate was optimum at 600 mL/min of aeration supply. This also indicates that the system can achieve better energy recovery and energy saving by lowering the aeration rate. NERv of this study was higher compared to Xu et al., (2016), which was (15-30Wh/kg COD). So far, Liu et al., (2014) achieved the maximum NERs of 0.0473 kWh/Kg COD, which was two-fold higher than the present study; however, it was still lower than many pure MFC systems.

Meanwhile, the NERv of the present study was comparable to other similar CW-MFC studies, such as Yadav et al., (2012) and Zhao et al., (2014) with 0.0014 kWh/m3. Due to the negligible voltage output from R2, the NER was inappreciable. Vast difference was also observed in NER and CE between non-aerated R1 and control R2, which indicated that there was contribution of macrophytes in R1 towards bioenergy recovery. Submerged plant significantly affects the performance of bioelectricity generation and its role was magnified from the evaluation of energy recovery performance. CW-MFC studies are still in its infancy and the publications are still limited; therefore, further optimization of operation conditions and configurations are much needed to improve the CE and NER of the system.

4. CONCLUSION

This study demonstrates the significance of macrophytes and supplementary aeration in achieving simultaneous wastewater treatment and bioelectricity generation. Compared to control reactor, macrophytes and supplementary aeration could improve the DO at the cathodic region and benefit ammonium removal and bioelectricity generation in a single unit UFCW-MFC. Although aeration would increase the operational cost, it could be adjusted to obtain the optimum aeration flow rate to achieve maximum results in COD, nitrate, ammonium removal and energy recovery without excessive wastage. The best performance in bioenergy recovery was achieved at 600 mL/min of aeration flow rate.

ACKNOWLEDGMENT

The authors would like to acknowledge the financial support of Science Fund (Grant No. 02-01-15SF0201) provided by the Ministry of Science, Technology and Innovation (MOSTI), Malaysia.

REFERENCES 1.

Aktaş, Ö., Çeçen, F., 2007. Bioregeneration of activated carbon: A review. Int. Biodeterior. Biodegrad. 59, 257–272.

2.

Białowiec, A., Albuquerque, A., Randerson, P.F., 2014. The influence of evapotranspiration on vertical flow subsurface constructed wetland performance. Ecol. Eng. 67, 89–94.

3. Białowiec, A., Davies, L., Albuquerque, A., Randerson, P.F., 2012. The influence of plants on nitrogen removal from landfill leachate in discontinuous batch shallow constructed wetland with recirculating subsurface horizontal flow. Ecol. Eng. 40, 44–52. 4.

Bishop, P.L., Eighmy, T.T., 1989. Aquatic wastewater treatment using Elodea nuttallii. J. Wat. Poll. Cont. Fed. 61, 641–648.

5. Brix, H., 1994. Functions of macrophytes in constructed wetlands. Water Sci. Technol. 29, 71–78. 6.

Brix, H., 1997. Do macrophytes play a role in constructed treatment wetlands? Water Sci. Technol. 35, 11–17.

7.

Cheng, S., Liu, H., Logan, B.E., 2006. Increased power generation in a continuous flow MFC with advective flow through the porous anode and reduced electrode spacing. Environ. Sci. Technol. 40, 2426–2432.

8.

Clauwaert, P., Rabaey, K., Aelterman, P., De Schamphelaire, L., Pham, T.H., Boeckx, P., Boon, N., Verstraete, W., 2007. Biological denitrification in microbial fuel cells. Environ. Sci. Technol. 41, 3354–3360.

9.

Doherty, L., Zhao, Y., Zhao, X., Hu, Y., Hao, X., 2015a. A review of a recently emerged technology : Constructed wetland e Microbial fuel cells. Water Res. 85, 38–45.

10. Doherty, L., Zhao, Y., Zhao, X., Wang, W., 2015b. Nutrient and organics removal from swine slurry with simultaneous electricity generation in an alum sludge-based constructed wetland incorporating microbial fuel cell technology 266, 74–81. 11. Fang, Z., Song, H., Cang, N., Li, X., 2013. Performance of microbial fuel cell coupled constructed wetland system for decolorization of azo dye and bioelectricity generation. Bioresour. Technol. 144, 165–171. 12. Fang, Z., Song, H., Cang, N., Li, X., 2015. Electricity production from Azo dye wastewater using a microbial fuel cell coupled constructed wetland operating under different operating conditions 68, 135–141. 13. Ge, Z., Li, J., Xiao, L., Tong, Y., He, Z., 2014. Recovery of Electrical Energy in Microbial Fuel Cells. Environ. Sci. Technol. Lett. 1, 137–141. 14. Gil, G.C., Chang, I.S., Kim, B.H., Kim, M., Jang, J.K., Park, H.S., Kim, H.J., 2003. Operational parameters affecting the performannce of a mediator-less microbial fuel cell. Biosens. Bioelectron. 18, 327–334. 15. GISD, 2010. Global Invasive Species Database. Invasive Species Specialist Group of the IUCN. Website: http://www.issg.org/database 16. Guo, L., He, K., Wu, S., Sun, H., Wang, Y., Huang, X., Dong, R., 2016. Optimization of high-rate TN removal in a novel constructed wetland integrated with microelectrolysis system treating high-strength digestate supernatant. J. Environ. Manage. 178, 42–51. 17. Headley, T., Nivala, J., Kassa, K., Olsson, L., Wallace, S., Brix, H., Afferden, M. Van, Müller, R., 2013. Escherichia coli removal and internal dynamics in subsurface flow ecotechnologies : Effects of design and plants. Ecol. Eng. 61, 564–574. 18. Hu, Y., He, F., Ma, L., Zhang, Y., Wu, Z., 2016. Microbial nitrogen removal pathways in integrated vertical-flow constructed wetland systems. Bioresour. Technol. 207, 339–345. 19. Larminie, J., Dicks, A., 2000. In: Fuel cell systems explained. John Wiley & Sons, Chichester, p308.

20. Liang, P., Huang, X., Fan, M.-Z., Cao, X.-X., Wang, C., 2007. Composition and distribution of internal resistance in three types of microbial fuel cells. Appl. Microbiol. Biotechnol. 77, 551–8. 21. Liu, H., Hu, Z., Zhang, J., Ngo, H.H., Guo, W., Liang, S., Fan, J., Lu, S., Wu, H., 2016a. Optimizations on supply and distribution of dissolved oxygen in constructed wetlands: A review. Bioresour. Technol. 214, 797–805. 22. Liu, J., Yi, N.K., Wang, S., Lu, L.J., Huang, X.F., 2016b. Impact of plant species on spatial distribution of metabolic potential and functional diversity of microbial communities in a constructed wetland treating aquaculture wastewater. Ecol. Eng. 94, 564–573. 23. Liu, L., Zhao, X., Zhao, N., Shen, Z., Wang, M., Guo, Y., Xu, Y., 2013a. Effect of aeration modes and influent COD/N ratios on the nitrogen removal performance of vertical flow constructed wetland. Ecol. Eng. 57, 10–16. 24. Liu, S., Song, H., Li, X., Yang, F., 2013b. Power Generation Enhancement by Utilizing Plant Photosynthate in Microbial Fuel Cell Coupled Constructed Wetland System. Int. J. Photoenergy, 2, 1-10. 25. Liu, S., Song, H., Wei, S., Yang, F., Li, X., 2014. Bio-cathode materials evaluation and configuration optimization for power output of vertical subsurface flow constructed wetland - Microbial fuel cell systems. Bioresour. Technol. 166, 575–583. 26. Logan, B.E., Hamelers, B., Rozendal, R., Schröder, U., Keller, J., Freguia, S., Aelterman, P., Verstraete, W., Rabaey, K., 2006. Microbial fuel cells: Methodology and technology. Environ. Sci. Technol. 40, 5181–5192. 27. Maltais-Landry, G., Maranger, R., Brisson, J., Chazarenc, F., 2009. Nitrogen transformations and retention in planted and artificially aerated constructed wetlands. Water Res. 43, 535–545.

28. Olsson, G., 2015. Water and Energy – Threats and Opportunities, first ed. IWA Publishing, London. 29. Ong, S.-A., Uchiyama, K., Inadama, D., Ishida, Y., Yamagiwa, K., 2010. Performance evaluation of laboratory scale up-flow constructed wetlands with different designs and emergent plants. Bioresour. Technol. 101, 7239–44. 30. Oon, Y.-L., Ong, S.-A., Ho, L.-N., Wong, Y.-S., Oon, Y.-S., Lehl, H.K., Thung, W.-E., 2015. Hybrid system up-flow constructed wetland integrated with microbial fuel cell for simultaneous wastewater treatment and electricity generation. Bioresour. Technol. 186, 270–275. 31. Oon, Y.-L., Ong, S.-A., Ho, L.-N., Wong, Y.-S., Dahalan, F.A., Oon, Y.-S., Lehl, H.K., Thung, W.-E., 2016. Synergistic effect of up-flow constructed wetland and microbial fuel cell for simultaneous wastewater treatment and energy recovery. Bioresour. Technol. 203, 190–197. 32. Ozimek, T., van Donk, E., Gulati, R.D., 1993. Growth and nutrient uptake by two species of Elodea in experimental conditions and their role in nutrient accumulation in a macrophyte-dominated lake. Hydrobiologia 251, 13–18. 33. Rosenbaum, M., Aulenta, F., Villano, M., Angenent, L.T., 2011. Cathodes as electron donors for microbial metabolism: Which extracellular electron transfer mechanisms are involved? Bioresour. Technol. 102, 324–333. 34. Shelef, O., Gross, A., Rachmilevitch, S., 2013. Role of plants in a constructed Wetland: Current and new perspectives. Water (Switzerland) 5, 405–419. 35. Srikanth, S., Venkata Mohan, S., 2012. Change in electrogenic activity of the microbial fuel cell (MFC) with the function of biocathode microenvironment as terminal electron accepting condition: Influence on overpotentials and bio-electro kinetics. Bioresour. Technol. 119, 241–251.

36. Srivastava, P., Yadav, A.K., Mishra, B.K., 2015. The effects of microbial fuel cell integration into constructed wetland on the performance of constructed wetland. Bioresour. Technol. 37. Sublette, K.L., Snider, E.H., Sylvester, N.D., 1982. A review of the mechanism of powdered activated carbon enhancement of activated sludge treatment. Water Research. 16, 1075–1082. 38. Tanner, C.C., 1996. Plants for constructed wetland treatment systems - A comparison of the growth and nutrient uptake of eight emergent species. Ecol. Eng. 7, 59–83. 39. Tchbanoglous, G., 1991. Wastewater Engineering: Treatment, Disposal, Reuse, third ed. Metcalf & Eddy, Boston, MA. 40. Villaseñor, J., Capilla, P., Rodrigo, M. a, Cañizares, P., Fernández, F.J., 2013. Operation of a horizontal subsurface flow constructed wetland--microbial fuel cell treating wastewater under different organic loading rates. Water Res. 47, 6731–8. 41. Vymazal, J., 2007. Removal of nutrients in various types of constructed wetlands. Sci. Total Environ. 380, 48–65. 42. Vymazal, J., 2011. Plants used in constructed wetlands with horizontal subsurface flow: A review. Hydrobiologia 674, 133–156. 43. White, M., and Bishop, P., 1984. Evaluation of five species of temperate climate aquatic plants for application in aquaculture treatment of wastewater, Environmental Engineering, Proceedings of the 1984 National Environmental Engineering Conference, Los Angeles, CA, 564-572, 44. Wu, H., Fan, J., Zhang, J., Ngo, H.H., Guo, W., Hu, Z., Lv, J., 2016. Optimization of organics and nitrogen removal in intermittently aerated vertical flow constructed wetlands: Effects of aeration time and aeration rate. Int. Biodeterior. Biodegradation 113, 139–145.

45. Xu, L., Zhao, Y., Doherty, L., Hu, Y., Hao, X., 2016. Promoting the bio-cathode formation of a constructed wetland-microbial fuel cell by using powder activated carbon modified alum sludge in anode chamber. Sci. Rep. 6, 26514. 46. Xu, L., Zhao, Y., Doherty, L., Hu, Y., Hao, X., 2015. The integrated processes for wastewater treatment based on the principle of microbial fuel cells: A review. Crit. Rev. Environ. Sci. Technol. 46, 1–32. 47. Yadav, A.K., Dash, P., Mohanty, A., Abbassi, R., Mishra, B.K., 2012. Performance assessment of innovative constructed wetland-microbial fuel cell for electricity production and dye removal. Ecol. Eng. 47, 126–131. 48. Zhao, Y., Collum, S., Phelan, M., Goodbody, T., Doherty, L., Hu, Y., 2013. Preliminary investigation of constructed wetland incorporating microbial fuel cell: Batch and continuous flow trials. Chem. Eng. J. 229, 364–370.

Figures and Tables

Fig. 1. Schematic diagram of UFCW-MFC Reactor 1 (R1) and Reactor 2 (R2) (Control) 1 peristaltic pump ○ 2 anode (activated carbon) ○ 3 aeration (optional) ○ 4 (not scaled)[○ ○ 5 wetland plant ○ 6 carbon rod ○ 7 resistor] cathode (activated carbon) ○ Fig. 2. DO monitoring (a) and ORP monitoring (b) of R1 and R2 at various aeration flow rates Fig 3. Nitrate (a) and ammonium (b) removal profile of R1 and R2 at various aeration flow rates Fig 4. COD removal profile of R1 and R2 (a) and time course monitoring of R1 (b) at various aeration flow rates Fig 5. Voltage output of R1 (a), voltage output of R2 (b), power density-current density of R1 (c), NER and CE of R1 and R2 (d) at various aeration flow rates Table 1 Wastewater treatment performance of UFCW-MFC Table 2 Comparisons of COD removal efficiency and Columbic efficiency (CE) of CWMFC studies with different plant species and aeration conditions. Table 3 Bioelectricity performance of UFCW-MFC Table 4 Performance comparison of CW-MFC studies

R1

R2

Fig. 1. Schematic diagram of UFCW-MFC Reactor 1 (R1) and Reactor 2 (R2) (Control) 1 peristaltic pump ○ 2 anode (activated carbon) ○ 3 aeration (optional) ○ 4 (not scaled)[○ ○ 5 wetland plant ○ 6 carbon rod ○ 7 resistor] cathode (activated carbon) ○

(a)

6.00

1900 mL/min 5.00

1200 mL/min 600 mL/min

DO (ppm)

4.00

60 mL/min 3.00

0 mL/min 0 mL/min (Dark condition)

2.00

0 mL/min (R2 - Control) 1.00 0.00 Influent

S1

250

(b)

150 100

ORP (mV)

S4

S5

1900 mL/min 1200 mL/min 600 mL/min 60 mL/min 0 mL/min 0 mL/min (Dark condition) 0 mL/min (R2 - Control)

200

50 0 Influent -50

S2 S3 Sampling points

S1

S2 S3 Sampling points

S4

S5

-100 -150 -200 -250

Figure 2 DO monitoring (a) and ORP monitoring (b) of R1 and R2 at various aeration flow rates

160

(a)

Nitrate conc. (mg/L)

140

1900 mL/min

120

1200 mL/min

100

600 mL/min

80

60 mL/min

60

0 mL/min 0 mL/min Dark condition 0 mL/min (R2 Control)

40 20 0 Influent

S1

S2 S3 Sampling points

S4

S5

50

(b)

45

Ammonium conc. (mg/L)

40 35 30

1900 mL/min

25

1200 mL/min

20

600 mL/min

15

60 mL/min

10

0 mL/min 0 mL/min (Dark condition)

5 0 Initial

0 mL/min (R2 - Control) S1

S2 S3 Sampling Points

S4

S5

Figure 3 Nitrate (a) and ammonium (b) removal profile of R1 and R2 at various aeration flow rates

700

(a) 600

COD conc. (mg/L)

500 400 300 200 100 0 Influent

S1 1900 mL/min 600 mg/L 0 mg/L 0 mg/L (R2 Control)

S2 S3 Sampling points

S4 1200 mg/L 60 mg/L 0 mg/L Dark condition

700

100 90

600

COD conc. (mg/L)

80 500

70 60

400

Influent

300

50 40

S5/Effluent 200

% removal 100

30 20

COD removal efficiency (%)

(b)

S5

10

0

0 0

20 40 60 80 100 120 140 160 180 200 220 240 260 280 Time (d)

Figure 4 COD removal profile of R1 and R2 (a) and time course monitoring of R1 (b) at various aeration flow rates

700

(a)

A1-C

600

A2-C

Voltage (mV)

500

A3-C

400 300 200

1200 mL/min

1900 mL/min

600 0 60 mL/min mL/min mL/min

100 0 0

20

40

60

80 100 120 140 160 180 200 220 240 260 280 Time (d)

100

(b)

Voltage (mV)

80 60

C-A1 C-A2

40

C-A3 20 0 0

20

40

60

80 100 120 140 160 180 200 220 240 260 280 Time (d)

200

(c)

180

Power density (mW/m3)

160 140 120 100 80 60 40 20 0 0

400 600 800 Currrent density (mA/m3)

1900 mL/min

1200 mL/min

60 mL/min

0 mL/min

1000

1200

600 mL/min

220 200 180 160 140 120 100 80 60 40 20 0

12.00

NERs CE

10.00 8.00

NERv 6.00 4.00

NERv (wh/m3) or cE (%)

NERs (Wk/kgCOD)

(d)

200

2.00 0.00

1900

1200

600

60

0

0 (R2)

Aeration flow rates (mL/min)

Figure 5 Voltage output of R1 (a), voltage output of R2 (b), power density-current density of R1 (c), NER and CE of R1 and R2 (d) at various aeration flow rates

Table 1 Wastewater treatment performance of UFCW-MFC

Parameters

Reactor

Nitrate

R1

R2 Ammonium R1

R2 COD

R1

R2

Aeration flow rates (mL/min)

Influent (mg/L)

Effluent (mg/L)

% Removal

1900 1200 600 60 0 0 (Dark condition) 0

141.4±3.24 138.3±2.87 140.8±3.36 138.9±3.03 138.7±2.75 139.8±1.03 140.5±2.80

90.06±6.02 84.53±8.43 78.60±7.94 69.80±5.40 9.31±3.71 3.59±3.30 2.1±1.61

36 39 44 50 93 97 98

1900 1200 600 60 0 0 (Dark condition) 0

40.94±2.67 39.5±1.94 39.43±2.22 40.72±2.32 41.86±2.30 41.35±2.51 41.37±2.42

1.05±0.47 1.43±0.56 6.24±1.60 7.72±2.12 22.19±1.56 25.38±1.92 28.98±1.95

97 96 84 81 47 39 30

1900 1200 600 60 0 0 (Dark condition) 0

648±7 643±8 644±3 646±3 642±5 643±2 650±4

8±3 11±4 12±2 15±3 16±3 15±3 22±3

99 98 98 98 98 98 97

Table 2 Comparisons of COD removal efficiency and Columbic efficiency (CE) of CW-MFC studies with different plant species and aeration conditions.

Plant Species

Artificial aeration

Mean influent COD concentration (mg/L)

COD removal efficiency (%)

Coulombic efficiency (%)

R

Canna indica Ipomoea aquatica Phragmites australis Phragmites australis Ipomoea aquatica Ipomoea aquatica Phragmites australis Ipomoea aquatica Typha latifolia Elodea nuttallii

Without Without Without With Without Without With Without With With

1500 180 560 1058 200 250 583 135 624 646

75 85.7 90.0-95.0 76.5 94.8 95 64 85.7 99 98

0.05-0.06 0.58-1.71 0.27 0.1-0.6 0.39-1.29 2.8-3.9 0.25 0.5 0.06-1.42 0.08-10.28

(Y (F (V (Z (L (L (D (F (O P

Table 3 Bioelectricity performance of UFCW-MFC Circuit connection between cathode and anode

Distance between anode and cathode (cm)

Aeration flow rates (mL/min) A1-C A2-C A3-C

15 30 45

Voltage output (

Internal resistant (Ω)

1900

1200

600

60

0

R2

1900

1200

600

200 430 450

200 430 450

200 430 450

300 430 450

300 430 450

430 450

528.86±38 493.25±35 440.55±31

540.07±35 481.78±33 449.66±33

545.77±25 472.79±24 442.26±22

40

42 37 34

Circuit connection between cathode and anode

Power density (mW/m3)

C

Aeration flow rates (mL/min)

1900

1200

600

60

0

R2

1900

1200

A1-C A2-C A3-C

170.64±5.84 135±10.64 98.58±12.31

174.11±3.39 129.46±10.86 107.52±9.1

184.75±7.50 123.31±10.81 104.00±7.65

125.40±14.36 88.78±13.84 67.83±14.12

63.68±3.52 41.46±9.72 34.26±8.69

0.00 0.17±0.07 0.15±0.06

3.14 3.28 0.08

6.78 2.87 0.09

Circuit connection between cathode and anode

Normalized Energy Recovery (NE

Normalized Energy Recovery (NERs) (Wh/kg COD)

Aeration flow rates (mL/min)

1900

1200

600

60

0

R2

1900

1200

600

60

A1-C A2-C A3-C

60.62 58.89 1.37

133.42 50.32 1.42

204.49 51.15 1.36

146.55 30.92 0.83

75.11 16.63 0.45

0.00 0.01 0.00

1.18 0.96 0.83

1.29 1.03 0.90

1.23 0.91 0.81

0.73 0.57 0.49

Table 4 Performance comparison of CW-MFC studies

41

CW-MFC Operation mode

Electrode Material

Vertical upflow-downflow

Anode – graphite plate Cathode – graphite plate Anode – granular activated carbon Cathode – granular activated carbon Anode – granular graphite Cathode – Pt coated carbon cloth Anode – activated carbon Cathode – activated carbon Anode – activated carbon Cathode – activated carbon

Vertical upflow Vertical flow (batch) Vertical upflow Vertical upflow

Electrode spacing (cm) 10.0

950

Maxim power density 268

15.0

1000

852

-

-

320.8

30.0

450

93.0

15.0

200

184.75±

Resistance (Ω)

42

Graphical abstract

43

Highlights

1.

Simultaneous organic matter, nutrient removal and bioenergy recovery by UFCWMFC.

2.

Supplementary aeration contributed to oxygenation of the wetland matrix.

3.

Aerated UFCW-MFC out-performed control reactor in bioelectricity generation.

4.

Macrophyte enhanced nitrification and bioelectricity generation.

5.

Oxygen as terminal electron acceptor for bioelectricity generation.

44