Treatment of domestic wastewater phosphate by electrocoagulation using Fe and Al electrodes: A comparative study

Treatment of domestic wastewater phosphate by electrocoagulation using Fe and Al electrodes: A comparative study

Process Safety and Environmental Protection 116 (2018) 34–51 Contents lists available at ScienceDirect Process Safety and Environmental Protection j...

2MB Sizes 0 Downloads 140 Views

Process Safety and Environmental Protection 116 (2018) 34–51

Contents lists available at ScienceDirect

Process Safety and Environmental Protection journal homepage: www.elsevier.com/locate/psep

Treatment of domestic wastewater phosphate by electrocoagulation using Fe and Al electrodes: A comparative study P.I. Omwene ∗ , M. Kobya Department of Environmental Engineering, Gebze Technical University, 41400 Gebze, Kocaeli, Turkey

a r t i c l e

i n f o

Article history: Received 15 October 2017 Received in revised form 25 December 2017 Accepted 7 January 2018 Keywords: Phosphorus removal Electrocoagulation Domestic wastewater Al and Fe electrodes

a b s t r a c t In this work, phosphorus removal from domestic wastewater was studied through electrocoagulation (EC) using iron (Fe) and aluminium (Al) anodes in a batch EC reactor. Key parameters investigated include; initial pH, initial phosphorus concentration (Ci ), reaction time, current density (j), metal-tophosphorous ratio, charge loading (q) and electrode type. The Al/P mole ratio at current densities of 10, 20, 30 and 40 A/m2 was obtained as 0.84, 1.68, 0.95 and 0.77, respectively, while the Fe/P mole ratio for these current densities was calculated as 1.81, 3.29, 4.09 and 4.48, respectively. The optimums to obtain <0.01 mg/L effluent P concentration at Ci = 52 mg/L PO4 -P were; pHi = 4, j = 20 A/m2 and EC time = 100 min (q = 954C) for Fe electrode, whereas for Al electrode, the optimums were; EC time = 50 min (q = 372C), pHi = 4, j = 20 A/m2 . Also, the final pHf increased with increase in EC time; when pHi = 4, Ci = 52 mg/L and j = 20 A/m2 , the final pHf was 8.52 at 50 min for Al electrode and 10.62 at 100 min for Fe electrode. The operating costs, energy and electrode consumptions were calculated as 1.032 $/m3 , 1.143 kWh/m3 and 0.218 kg/m3 respectively for Al electrode and 1.343 $/m3 , 4.179 kWh/m3 and 0.884 kg/m3 , respectively for Fe electrodes. Overall, Al anodes provided a higher phosphorus removal efficiency in a shorter EC time and lesser metal-to-phosphorous ratio compared to Fe anodes. © 2018 Institution of Chemical Engineers. Published by Elsevier B.V. All rights reserved.

1. Introduction Phosphorus (P) is one of the essential elements for many living organisms; however, excess amount of phosphorus in lakes and other natural water bodies can increase eutrophication, which deteriorates water quality (Park et al., 2016; Ansari and Gill, 2014). Municipal wastewater treatment plants are one of the major phosphorous pollution sources in many lakes and reservoirs, representing about 30–50% of the total phosphorus discharge (Yang et al., 2010). Wastewater treated by only conventional mechanical–biological methods still contains phosphorus concentration of 0.5–15 mg/L, an adequate concentration to cause eutrophication in the receiving environment (Ansari and Gill, 2014). Thus, controlling phosphorous discharged from municipal and industrial wastewater treatment plants is a key factor in preventing eutrophication of surface waters. Biological processes for phosphorous removal are disadvantaged by longer treatment time, lower removal efficiency and complicated plant configurations. In contrast, chemical precipitation using coagulants such as calcium, aluminium, and iron salts is more favourable because of its simplic-

∗ Corresponding author. E-mail address: [email protected] (P.I. Omwene).

ity, flexibility and cost-effectiveness (Szabo et al., 2008; Georgantas and Grigoropoulou, 2007). Membrane processes such as electrodialysis and reverse osmosis are too expensive and remove only about 10% of the total phosphate (Yeoman et al., 1988). Furthermore, membrane bioreactors have also been used for phosphorus removal from wastewater, but are disadvantaged by their high maintenance cost and membrane fouling (Ngo and Guo, 2009). Phosphate removal from wastewater by adsorption using different materials has also been applied, however, the major disadvantages are low efficiency and high cost (Ramasahayam et al., 2014). Recent research and development in electrocoagulation (EC) process using sacrificial metal anodes for removing phosphorus from aqueous solutions has attracted much attention, with >90% phosphorus removal efficiency at different process conditions (Kim et al., 2010; Stafford et al., 2014; Bouamra et al., 2012; Lacasa et al., 2011; Vasudevan et al., 2009; Irdemez et al., 2006a; Bektas et al., 2004). The EC technique is advantaged by its simplicity, easy operation, less retention time, reduction or absence of chemicals addition, rapid sedimentation of the electrogenerated flocs, less sludge production and environmental compatibility when compared to conventional methods (Mollah et al., 2001). Although many studies on wastewater properties (i.e. Initial pH and initial phosphorus concentration) and operating parameters (i.e. current density and treatment time) for phosphorus removal by EC process

https://doi.org/10.1016/j.psep.2018.01.005 0957-5820/© 2018 Institution of Chemical Engineers. Published by Elsevier B.V. All rights reserved.

P.I. Omwene, M. Kobya / Process Safety and Environmental Protection 116 (2018) 34–51

35

using Fe and Al anodes were reported in the literature, analyses of electrochemically generated metal-to-removed phosphorus ratio (Me/P; Me: Al or Fe) were not studied in detail, despite the metalto-phosphorus (Fe/P or Al/P) molar ratio being one of the main factors determining phosphorus removal efficiency in both chemical coagulation and electrocoagulation processes (Huang et al., 2017; Stafford et al., 2014; Szabo et al., 2008). This article presents research on phosphorus removal from synthetic domestic wastewater using Al and Fe plate electrodes in a batch EC reactor. For this purpose, effect of different process parameters like initial pHi , current density, initial phosphorus concentration on phosphorus removal have been studied as function of the EC time. Also, metal-to-phosphorus ratios, phosphorus adsorption capacity of electrochemically generated coagulant, operating cost, energy and electrode consumptions were analysed at different process conditions.

et al., 2007; Fytianos et al., 1998). When H2 O molecules are present between the adsorbed phosphate ions and oxide surface, its referred as outer surface complexation. The electrolysis of the electrode produces not only precipitates such as ferrous, ferric, and aluminium phosphate or hydroxyl-phosphate, as shown in Eqs. (9)–(11), but also hydroxides such as Fe(OH)2(s) , Fe(OH)3(s) , and Al(OH)3(s) . When iron and aluminium are present in the water, FePO4(s) and AlPO4(s) forms in the low pH range (<6.5) and at higher pH range (>6.5) iron and aluminium increasingly convert to oxides and hydroxides. FePO4(s) has minimum solubility within pH range of 4.5–5.5, but its solubility increases with increasing pH (Nguyen et al., 2016; Lacasa et al., 2011; Zhang et al., 2010; Szabo et al., 2008; Irdemez et al., 2006b; Thistleton et al., 2002; Jiang and Graham, 1998). Moreover, the optimum pH range for phosphate precipitation with ferric iron is between pH 4.0 and 5.0, while that for ferrous iron is close to pH 8.0.

2. Mechanism of phosphate removal with EC

3Fe2+ + 2PO3− 4 → Fe3 (PO4 )2(s)

In EC process, coagulants are generated in-situ by the electrodissolution of a sacrificial anode, which is usually made of aluminium (Al) or iron (Fe). The metal ions generation takes place at the anode; hydrogen gas is released from the cathode. The hydrogen gas would also help to float the flocculated particles out of the water (Can et al., 2003; Mollah et al., 2001). Anode and cathodes reactions for Al electrodes:





Anode : Al → Al3+ + 3e− E 0 = 1.66V

 −

Cathode : 3H2 O + 3e− → 3/2H2(g) + 3OH



E 0 = −0.828 V

(1) (2)

Anode and cathodes reactions for Fe electrodes:



Anode : Fe → Fe2+ + 2e− E 0 = 0.44 V 2+

Fe

→ Fe

3+



+e





(3)



0

E = −0.771 V

 −

Cathode : 2H2 O + 2e− → H2(g) + 2OH



E 0 = −0.828 V

(4) (5)

In addition, the rate of oxidation of Fe2+ depends on the availability of dissolved oxygen (Mollah et al., 2001). O2(g) + 4Fe2+ +2H2 O → 4Fe3+ +4OH− (Fe3+ )

(6)

Al3+

Ferric ions and ions generated by electrochemical oxidation in EC process may form monomeric and polymeric iron and aluminium species, which transform finally into Fe(OH)3(s) and Al(OH)3(s) depending on the pH of the aqueous medium (Palahouane et al., 2015; Mollah et al., 2001; Rebhun and Lurie, 1993). Al3+ + 3H2 O → Al(OH)3(s) + 3H+

(7)

Fe3+ +3H2 O → Fe(OH)3(s) +3H+

(8)

The generated and ions also react with OH− in solution to form amorphous hydroxide flocs. Consequently, pollutant removal mechanism with both electrodes is related to formation of Fe(OH)3(s) , Al(OH)3(s) , monomeric and polymeric iron and aluminium species due to, coagulation, precipitation, co-precipitation, and electro-oxidation (Mollah et al., 2001). Furthermore, freshly formed amorphous Al(OH)3(s) and Fe(OH)3(s) “sweep flocs” in the EC process have large surface areas which are beneficial for a rapid adsorption of soluble organic compounds such as phosphate ions and trapping of colloidal particles. 3x-y Phosphate ion has strong affinity for metal ions like Mex (OH)y(s) Al3+ ,

Fe3+

Fe2+

(Me; Al3+ or Fe3+ ), Al and Fe in aqueous medium always have surface OH groups. In solution, phosphate ion undergoes ligand exchange with the OH and adsorption of phosphate ions on the metal oxide surface leads to inner surface complexation (Karageorgiou

Al3+ or Fe3+ + PO−3 4 → FePO4(s) or AlPO4(s) 3+

+ PO3− 4



(9) (10)

+ (3x − 3) OH → Fex PO4 (OH)3x−3(s)

(11)

− xAl3+ + PO3− 4 + (3x − 3) OH → Alx PO4 (OH)3x−3(s)

(12)

xFe

In general, two major mechanisms have been linked to the coagulation of PO3− 4 with Al or Fe(III) salts: (i) Formation of Al or Fe-hydroxo-phosphate complexes; Me(OH)3−x (PO4 )x(s) . These complexes either adsorb onto positively charged Al/Fe(III) hydrolysis species or act as centres of precipitation for Al/Fe(III) hydrolysis products. (ii) Adsorption of PO3− ions on to the Al or Fe(III) 4 hydrolysis species. Such hydroxides are partially transformed into hydroxyl-complexes depending on the pH of the solution, but these can remove phosphate by adsorption (Nguyen et al., 2016; Lacasa et al., 2011; Zhang et al., 2010; Szabo et al., 2008; Thistleton et al., 2002; Jiang and Graham, 1998). FePO4(s) (strengite) and AlPO4(s) (variscite) are the stable solid phases if phosphate is precipitated in the pH range of 4–7. Minimum solubility of AlPO4 occurs at pH 6 which is one unit higher than that of FePO4(s) (pH 5). The stoichiometric mass ratio of Fe/P for FePO4(s) and Fe3 (PO4 )2(s) is 1.8:1 and 2.7:1, respectively (Karageorgiou et al., 2007; Fytianos et al., 1998). 3. Material and methods 3.1. Material Typical composition of untreated domestic wastewater is pH of 7.0–8.5 (∼7.7), conductivity of 0.403–1.284 mS/cm (∼0.85 mS/cm), chemical oxygen demand (COD) of 250–800 mg/L (∼430 mg/L), total phosphorous of 4–12 mg/L (∼7 mg/L), chloride of 30–90 mg/L (∼50 mg/L), and sulphate of 20–50 mg/L (∼30 mg/L) according to literature (Tchobanoglous et al., 2004). As per this composition, synthetic domestic wastewater for this study was prepared from KH2 PO4 with distilled water for the required concentration (5–52 mg/L PO4 -P). Also, typical chloride, sulphate and alkalinity concentrations found in domestic wastewater were obtained by addition of NaCl, Na2 SO4 , Na2 CO3 and potassium hydrogen phthalate compounds. Characterization of synthetic domestic wastewater used in this study is pH of 7.3, conductivity of 810 ␮S/cm, COD of 520 mg/L, chloride of ∼38.2 mg/L, sulphate of ∼33.2 mg/L, and bicarbonate alkalinity of ∼148.5 mg/L CaCO3 . 3.2. Experimental setup and procedure The EC experiments were carried out in a batch mode using a 385 mL plexiglass reactor (110 mm × 70 mm × 50 mm) using vertically positioned anodes and cathodes electrodes spaced by 11 mm

36

P.I. Omwene, M. Kobya / Process Safety and Environmental Protection 116 (2018) 34–51

(a) Al electrode

45 40 35 30 25 20 15 10 5 0 0

10

20

30

55

2

j (A/m ) 10 20 30 40

40

50

60

70

80

90 100 110

Residual phosphorus concentration (mg/L)

50

(b) Fe electrode

50

2

j (A/m ) 10 20 30 40

45 40 35 30 25 20 15 10 5 0 0

10

20

30

40

50

60

70

80

90 100 110

Fig. 1. Variation of residual phosphorous concentration with EC time at different applied current densities.

and dipped in a domestic wastewater. Two anodes (Fe or Al) and two cathodes (Ti) with dimensions of 44 mm × 47 mm × 2 mm were connected to a digital DC power supply (Agilent 6675A mode, 0–120 V and 0–20 A) in monopolar parallel connection mode and operated at galvanostatic mode. The purity of iron, aluminium and titanium plates were 99.5%, 99.3% and 99.5%, respectively. The total effective anode area was 62 cm2 . All experimental runs were conducted with 350 mL of wastewater and stirred at 300 rpm by a magnetic stirrer (Heidolp MR 3000D) to reduce the mass transport over potential of the EC reactor, and the current density was adjusted to a desired value by a digital DC power supply for all the experimental runs. During the experiment, wastewater samples were withdrawn from the reactor at specified intervals and filtered by 0.45 ␮m microspore membrane filter before analyses. EC experiments for phosphorus removal were carried out at operating conditions of initial pH (4–7), current density (10–40 A/m2 ), and initial phosphorus concentration (5–52 mg/L) as function of the operating time (0–100 min). Furthermore, to reduce the effect of electrode passivation the electrodes were rinsed in the mixer of HCl solution (35%) and hexamethylenetetramine aqueous solution (2.8%) before starting each experiment, dried and re-weighed (Can et al., 2003).

3.3. Analytical methods Wastewater samples collected from the reactor at specified intervals of time were filtered by 0.45 micro meter cellulose acetate membrane filter before analysis. The analysis of phosphate was carried out using the Vanadomolybdophosphoric Acid Method by a single beam spectrophotometer (PerkinElmer Lambda 25) according to the Standard Methods for Examination of Water and Wastewater (APHA, 2005). The pH and conductivity of the samples before and after the EC process was measured by a pH and a conductivity meter (Hach Lange HQ40). The experiments were repeated three times and the average data was reported. All chemicals used in the EC experiments were of analytical grade. In addition, sludge generated after the EC experiment was dried in an oven (Memmert, Germany) at 105 ◦ C for 24 h. The sludge generated after the EC process was characterized by scanning electron microscope (SEM, Philips XL30S-FEG), and Fourier transform infrared spectroscopy (FTIR, Bio Rad FTS 175C spectrophotometer). Crystal phases of the precipitates were characterized using the X-ray diffraction (XRD; Rigaku 2000 D/max with CuK ␣-radiation,  = 0.154 nm at 40 kV and 40 mA).

4. Results and discussion 4.1. Effect of current density on phosphorus removal Current density is important parameter for controlling the reaction rate in the EC process (Attour et al., 2016; Bouamra et al., 2012; Zheng et al., 2009; Irdemez et al., 2006b). Applied current determines the coagulant dosage, bubble production rate, floc size and growth rate; these influence the P removal efficiency of EC process (Mollah et al., 2001). The effect of current density on the phosphorus removal at 10, 20, 30 and 40 A/m2 as function of the EC time were studied at operating conditions of pHi = 4.0 and Ci = 52 mg/L. Fig. 1 illustrates the effect of current density on the residual phosphorus concentration for Al and Fe electrodes. According to the European Environment Agency, the critical phosphorus concentrations for incipient eutrophication are about 0.1–0.2 mg P/L in running water and 0.005–0.01 mg P/L in still water (EEA, 2015). The United States Environmental Protection Agency (US-EPA) water quality criteria state that phosphates should not exceed 0.05 mg/L if streams discharge into lakes or reservoirs, 0.025 mg/L within a lake or reservoir, and 0.1 mg/L in streams or flowing waters not discharging into lakes or reservoirs to control algal growth. Surface waters that are maintained at 0.01–0.03 mg/L of total phosphorus tend to remain uncontaminated by algal blooms (Ramasahayam et al., 2014). As seen in Fig. 1, the EC time required to obtain effluent phosphorus concentration (Cf ) of 0.01 mg/L reduced with increase in the current density for both electrodes. To obtain less than 0.01 mg/L P removal, required EC time (or charge loading: q) for current densities of 10, 20, 30 and 40 A/m2 was 60 min (q = 223.2C), 50 min (q = 372C), 30 min (q = 334.8C) and 20 min (q = 297.6C), respectively, for Al electrode whereas for Fe electrode the required EC time was >100 min, 100 min (q = 954C), 80 min (q = 1144.8C), 60 min (q = 1144.8C), respectively at the above conditions. According to the results obtained in this study, the optimum current density was 20 A/m2 for both Al and Fe electrodes. It was observed that required EC time and charge loading for effluent phosphorus concentration of 0.01 mg/L in the EC process for Al electrode was less than in the case of Fe electrode. As the EC time and current density increased, the amount of oxidized aluminium or iron and the required charge loading increased. In addition, final pHf at 10, 20, 30 and 40 A/m2 current densities was measured as 10.20, 10.62, 11.08 and 10.94, respectively for Fe electrode and 7.89, 8.52, 7.76 and 7.55 for Al electrode, respectively. The pH of the effluents tended to pH value of >9 as the EC phosphorus removal process increased.

P.I. Omwene, M. Kobya / Process Safety and Environmental Protection 116 (2018) 34–51

450

1200 (a) Al electrode

1100

Removal capacity (mg P/g Al)

800

Removal capacity (mg P/g Fe)

j (A/m ) 10 20 30 40

900

700 600 500 400 300 200

2

j (A/m ) 10 20 30 40

350 300 250 200 150 100 50

100 0

(b) Fe electrode

400

2

1000

37

0 0

10

20

30

40

50

60

70

80

90

100

110

0

10

20

30

40

50

60

70

80

90

100

110

EC time (min)

EC time (min)

Fig. 2. Phosphorus removal capacity at different applied current densities.

When processes such as adsorption and precipitation (including co-precipitation and sweep precipitation) are the key mechanisms of phosphorus removal, EC maintains pH values in the range of 4.5–9.5 with insoluble Al(OH)3(s) or Fe(OH)3(s) as the prevailing aluminium or iron species and precipitants such as FePO4(s) , AlPO4(s) , Al1.4 PO4 (OH)1.2(s) , Fe3 (PO4 )2(s) and Fe2.5 PO4 (OH)4.5(s) are formed in solution (Karageorgiou et al., 2007; Irdemez et al., 2006b; Fytianos et al., 1998). The decrease in required EC time with current density is attributed to the increase in dissolution rate of the sacrificial Al and Fe electrodes with current density according to Faraday’s law. As a result, the amount of phosphate adsorption increases with the increase in the amount of iron and aluminium hydroxides (adsorbent concentration), which indicates that the adsorption depends on the availability of binding sites for phosphate ions in the solution by the EC reactions as explained in Eqs. (1)–(2) and (4)–(6), and the solution reactions of Eqs. (3) and (7)–(12).

4.1.1. Effect of charge loading Charge loading or the delivered coagulant dosage by EC process is a combined function of applied current density (or current) and EC time, as these two parameters control the rate of electrode dissolution, according to the Faraday’s law in Eq. (13) (Chen et al., 2000; Vik et al., 1984). However, required charge loading for phosphorus removal by EC process using Al electrode showed an increase with increase in current density from 10 to 20 A/m2 then decreased with increasing current density from 20 to 40 A/m2 . Therefore, some investigators have reported that charge loading (q) too can influence the treatment efficiency of EC process because it is controlled by applied current and EC time. q = i × tEC Celectrode,the =

(13) i × tEC × Mw, Fe or Al zFe or Al × F

(14)

where q is the charge loading (C), Celectrode,the is the amount of electrode consumed theoretically according to the Faraday’s law (g), i is applied current (A), tEC is EC time (s), Mw is molecular mass of electrode (Mw,Al = 26.98 g/mol, Mw,Fe = 55.85 g/mol), z is number of electrons transferred (ZAl = 3, ZFe = 2), and F is Faraday’s constant (96,487 C/mol). On the other hand, real electrode consumption depends on the Faradic yield of EC process. The Faradic yield or current efficiency (ϕFe or Al ) of electrode dissolution (Al or Fe) was calculated as the ratio of the weight loss of the electrodes during the experiments (Celectrode,exp ) to the amount of electrode consumed theoretically

according to Faraday’s law (Celectrode,the ) (Chafi et al., 2011; Merzouk et al., 2009; Chen et al., 2000). ϕFe or Al = (Celectrode, exp /Celectrode, the ) × 100

(15)

Current efficiency is also an important parameter for the EC process because it affects the lifetime of the electrodes. In this study, to achieve effluent phosphorus concentration of 0.01 mg/L, the experimental electrode consumptions and current efficiencies were determined as follows; at 10 A/m2 : 0.0382 g and 183.5% respectively for Al electrode, 0.1657 g and 120% respectively for Fe electrode. At 20 A/m2 : 0.0764 g and 220.3% respectively for Al electrode, 0.3092 g and 112% respectively for Fe electrode. At 30 A/m2 : 0.0430 g and 138% respectively for Al electrode, 0.3843 g and 116% respectively for Fe electrode. At 40 A/m2 : 0.0350 g and 126.2% respectively for Al electrode, 0.4208 g and 127% respectively for Fe electrode. Theoretically, whenever 1 Faraday of charge passes through the circuit, 9 g of aluminium and 28 g of iron is dissolved at each anode of an EC process. In this case, the molar mass of dissolved iron electrode in EC process is about twice that of aluminium electrode. The electrochemical dissolution of 1 mol of Fe requires only 2 mol of electrons whereas the dissolution of 1 mol of Al needs 3 mol of electrons at constant current. Some EC studies using aluminium electrodes have reported current efficiencies ranging from ∼100% to more than 300% (Mansouri et al., 2011; Zongo et al., 2009; Chen et al., 2000; Szynkarczuk et al., 1994). However, some studies using iron electrodes have reported current efficiencies close to 100% (Chafi et al., 2011; Zongo et al., 2009). According to the above results, the experimental (actual) electrode consumptions were greater than the theoretical values, the actual electrode consumption may reduce or increase from the theoretical value due to conditions such as chemical corrosion, hydrodynamics in EC reactor, electrochemical side-reactions, wastewater characteristic and operational condition (Mansouri et al., 2011; Chen et al., 2000). Phosphate ions in solution have strong affinity for metal oxides and hydroxides of Al3+ and Fe3+ . The electrochemically dissolved anodes (Al and Fe) and generated metal hydroxides in solution adsorb a large amount of phosphate on their surface, which is responsible for EC phosphorus removal. Phosphorus removal capacity per amount of electrochemically dissolved aluminium or iron (qe , mg P/g Al or Fe) was calculated from Eq. (16): qe =

(Ci − Ct ) × v Celectrode, exp

(16)

where Ci and Ct are the phosphorous concentrations at initial and time t (mg/L) of EC process, respectively, and v is solution volume in the EC reactor. The results of the phosphorus removal capacity for

38

P.I. Omwene, M. Kobya / Process Safety and Environmental Protection 116 (2018) 34–51

7

(a) Al electrode Iron-to-phosphorous molar ratio (nFe/nP)

3.5 3.0 2

j (A/m ) 10 20 30 40

2.5 2.0 1.5 1.0 0.5 0.0

0

10

20

30

40

50

60

70

80

90

100 110

(b) Fe electrode

6 5

2

j (A/m ) 10 20 30 40

4 3 2 1 0

0

10

20

30

40

50

60

70

80

90

100

110

Fig. 3. Variation of metal-to-phosphorous mole ratio in EC process for the applied current densities.

Al and Fe electrodes are presented in Fig. 2. The P removal capacities for Al and Fe electrodes at different current densities (to obtain effluent P concentration of 0.01 mg/L) were calculated as 477.93 mg P/g and 107.58 mg P/g respectively at 10 A/m2 , 238.86 mg P/g and 59.01 mg P/g respectively at 20 A/m2 , 423.98 mg P/g and 47.47 mg/g respectively at 30 A/m2 , and finally 521.29 mg P/g and 43.36 mg P/g respectively at 40 A/m2 . However, as seen in Fig. 2, phosphate removal capacity for both electrodes generally showed an increasing trend with increase in EC time and current density (except at EC time <5 min). Initially the rate of removal is high, then falls to a nearly constant rate as adsorption of phosphate ions continue. This trend is attributable to changes in the systems as it moves towards attaining equilibrium. Nevertheless, phosphorus removal capacity significantly decreased with increase in EC time as more coagulant dosage was electrochemically generated and residual P concentrations lowered (Fig. 2). This behaviour was likely due to a decrease in adsorption capacity of iron and aluminium hydroxides at low phosphorus concentrations. When we look at EC phosphorus removal results in literature, the maximum removal efficiency and adsorption capacity for Ci = 25–100 mg P/L was found as 99% and 124–505 mg/g respectively for aluminium alloy anode at j = 200 A/m2 , EC time of 50 min and pH of 7.0 (Vasudevan et al., 2009). EC studies for phosphorous removal with low carbon steel electrodes from liquid phase of dairy manure showed 96.7% removal efficiency and adsorption capacity of 54.05–45.25 mg P/g Fe at experimental conditions of: pHi = 7.4, EC time = 100 min, j = 263 A/m2 and Ci = 67.5–101 mg P/L (Zhang et al., 2016). Phosphorus removal efficiency and adsorption capacity for swine wastewater by continuous EC reactor was 96.3% and 48.3–86.2 mg/g Fe respectively for Fe electrode (at Ci = 64.93 ± 19.57 mg P/L, j = 278 A/m2 , hydraulic retention time in reactor: HRT = 70 min and final pHf = 8.5) and 87.1% and 76.7–70.1 mg/g Al respectively for Al electrode (at Ci = 73.41 ± 3.30 mg P/L, j = 389 A/m2 , HRT = 31.8 min and final pHf = 7.8), Mores et al. (2016). On the other hand, the maximum phosphorus adsorption capacity of chemical coagulation sludges from drinking water treatment plants was obtained as 25.5 mg/g at pH 5.5 and Ci = 5–50 mg P/L for the ferric sludge (Song et al., 2011) and 31.6 mg/g at pH 4.0 and Ci = 3425 mg P/L for alum sludge (Li and Zhao, 2010). In another study, Electrocoagulated Al-hydroxides sludge after calcinations at 600 ◦ C was used as an adsorbent for removal of phosphates from aqueous phase and equilibrium adsorption capacity was determined as 17.27 mg/g at initial pHi 3.0 (Golder et al., 2006). The maximum adsorption capacity by

Al-based drinking water treatment sludge (DWTs) were reported as 12.5 mg/g at Ci = 300 mg/L and 24-h equilibration time (Ippolito et al., 2003), 31.9 mg/g at Ci = 360 mg P/L and 48-h equilibration time (Babatunde et al., 2009), and 82.6 mg/g at Ci = 100 mg/L and 6 days equilibration time (O’Rourke et al., 2012). Dayton and Basta (2005) reported a maximum P adsorption capacity of 0.66–37 mg/g for a series of Al-based DWTs (Ci = 100 mg P/L and equilibration time of 17-h). Parfitt et al. (1975) reported phosphate adsorption capacities (mg P/g) of amorphous hydrous iron oxide (FeOOH), akaganeite (␤-FeOOH), lepidocrocite (-FeOOH), goethite (␣-FeOOH) and hematite (␣-Fe2 O3 ) at pH 3.5 as 29.5, 26.7, 16.7, 6.7 and 5.3, respectively. The amorphous iron oxides appeared to have a higher phosphate adsorption capacity than the crystalline iron oxides. In comparison with the results obtained in this study, electrocoagulated Al or Fe-hydroxides illustrated in Eqs. (3), (7) and (8) have demonstrated a higher P adsorption capacity than ferric and alum sludges of chemical coagulation. In order to achieve very low residual phosphorus concentration in wastewater treatment plant effluents, optimum coagulant dosages are required. According to Eq.(10), the theoretical coagulant dose for phosphate removal by both Fe3+ and Al3+ ions is of the molar ratio of 1:1. However, P removal from wastewater is complexed by several competing reactions occurring between the metal ions and the contaminants in the wastewater. The metal-tophosphorous molar ratios at different current densities for Al and Fe electrodes with respect to EC time are presented in Fig. 3. As seen in this figure, the Fe-to-P ratio increased with increase in current density and EC time. At current densities of 10 and 20 A/m2 , EC time of 100 min was required to obtain effluent P concentration of <0.01 mg/L, the Fe/P ratios for these current densities were 1.81 and 3.29 mol/mol respectively, whereas for current densities of 30 and 40 A/m2 , to obtain effluent P concentration of <0.01 mg/L, the Fe/P ratios were 4.48 and 4.09 mol/mol, respectively at their respective EC times of 80 and 60 min. The Al/P ratios varied from 0.44 to 1.40 for 10 A/m2 , 1.35 to 3.37 for 20 A/m2 , 0.52 to 3.16 for 30 A/m2 and 0.75 to 3.87 for 40 A/m2 for EC time variations of 5–100 min For Al electrode, to achieve effluent phosphorus concentration of 0.01 mg/L, the metal/P mole ratios at 10, 20, 30 and 40 A/m2 were obtained as 0.84, 1.68, 0.95 and 0.77 mol/mol, respectively. Phosphorus concentration at start-up of EC process (between 10–30 min) for j > 10 A/m2 rapidly decreased with the increase in the EC time (Fig. 1). The required Metal-to-P ratios to ensure low effluent concentrations for both electrodes were observed to increase with current density as seen in Fig. 3. However, whenever 1 C (i = 1 A and tEC = 1 s, Eq. (13)) of charge passes through the cir-

P.I. Omwene, M. Kobya / Process Safety and Environmental Protection 116 (2018) 34–51

(a) Al electrode 7 3

Energy consumption (kWh/m )

6 2

j (A/m ) 10 20 30 40

5 4 3 2 1 0

0

10

20

30

40

50 60 70 EC time (min)

80

90

100

110

15 (b) Fe electrode 14 13 12 11 10 9 8 7 6 5 4 3 2 1 0 0 10 20

39

2

j (A/m ) 10 20 30 40

30

40

50

60

70

80

90

100

110

Fig. 4. Energy and electrode consumptions for the applied current densities.

cuit according to the Faraday’s law, theoretically, 9.321 × 10−5 g of Al and 2.894 × 10−4 g of Fe is dissolved (at Eq.(14)). Therefore, for 1 C of charge passing through EC process, theoretically dissolved Fe/Al ratio is about 3.11. In this study, the amount of experimentally dissolved metal to achieve effluent P concentration of 0.01 mg/L (99.99% removal efficiency) at 10, 20, 30 and 40 A/m2 and initial P concentration of 52.12 mg/L were calculated as 0.0382, 0.0764, 0.043, and 0.0305 g respectively for Al electrode and 0.1657, 0.3092, 0.3843, and 0.4208 g respectively, for Fe electrode. These sum up to experimental Fe/Al gram ratios of 2.60, 2.02, 3.35 and 3.21 at 10, 20, 30 and 40 A/m2 current densities respectively. Studies have suggested that optimal doses for P removal by EC range from 2:1 to 1:1 mol ratio for Fe: P, but these studies do not directly compare P removal by EC with conventional coagulation, or use synthetic wastewater matrices (Yu et al., 2006). The chemical coagulation results in literature indicated that P removal increased with an increase in Fe/P mole ratio and initial phosphate concentration (Huang et al., 2017; Zhang et al., 2010; Szabo et al., 2008; Georgantas and Grigoropoulou, 2007; Thistleton et al., 2002; Caravelli et al., 2010; De Haas et al., 2000). Reported optimum metal to phosphorus ratio in the precipitate using different coagulants such as aluminium sulphate, poly-Al-chloride and Fe-sulphate have varied from less than 1 to as high as 10 at initial phosphorus concentration of 0.5–12.5 mg/L and initial pH 3–10 (Szabo et al., 2008). The metal to phosphorous molar ratios for effluent phosphorus concentration of 0.10 mg/L by these coagulants at experimental conditions (pH = 4.5–7.5 and Ci = 3.6–4.0 mg/L) were obtained as Fe/P = 2.8 and Al/P = 3.2 (Szabo et al., 2008). Zhang et al. (2010) studied phosphate removal using ferric chloride and obtained a Fe/P molar ratio of 1.5–2.0 at optimum pH range of 4.0–7.0 for initial phosphate concentration of 20 mg/L. Fytianos et al. (1998) obtained the optimum phosphate removal ratio as Fe/P = 1:1 at a pH of 4.5. When the Fe3+ ions are added at the Fe/P molar ratio of 1:1, ferric phosphate (FePO4(s) ) precipitation may be formed, whereas at the Fe/P molar ratio of more than 1:1, Fe2.5 PO4 (OH)4.5(s) precipitates are formed. Caravelli et al. (2010) obtained PO4 -P removals higher than 97% at pH above 6.2 using Fe/P = 1.9. Thistleton et al. (2002) calculated the required Fe/P molar ratios for 80% (initial phosphate concentration of 5 mg/L) of total phosphorus removal as 1.48 for FeCl3 , >4.8 for Fe(OH)3(s) , and 1.86 for FeCl3 /Fe(OH)3(s) , respectively. Narasiah et al. (1991) found that a weight ratio of Fe/P = 2.9 was needed to achieve 90% removal of phosphorus from municipal wastewater. Huang et al. (2017) found that Fe(III)/P and Al/P molar ratios by chemical coagulation was higher than 1.3 and Fe(II)/P ≥ 1.6, the PO4 -P in the wastewater was almost completely removed at experimental conditions (for Fe(II) at pHi 6.5

and 30 min, for Fe(III) at pHi = 4.5 and 30 min, for Al(III) at pHi = 5 and 30 min). Moreover, at the same current intensity and electrolysis time, the molar ratio of Fe/P in the filtered sludge anaerobic supernatant (Ci = 148 ± 6.6 mg/L) during the electrolysis of the Fe electrode was far higher than that during EC process using Al electrode (Fe/P = 0.1–4.5 and Al/P = 0.1–2.5 at j = 1.25–6.25 A/m2 ). 4.1.2. Operating cost, energy and electrode consumptions Electrical energy (Cenergy ) and electrode (Celectrode ) consumptions were calculated using the Eqs. (17) and (18) below: Cenergy (kWh/m3 ) = Celectrode (kg/m3 ) =

i × tEC × U

(17)

v i × tEC × MAl or Fe z×F ×v

(18)

where i is applied current (A), U is cell voltage (V), tEC is EC time (hour for energy consumption or sec for electrode consumption), v is volume (m3 ) of the wastewater in the EC reactor. Fig. 4 shows the energy and electrode consumptions per m3 wastewater treated for aluminium and iron electrodes during EC process. Measured cell voltages between electrodes for current densities of 10, 20, 30 and 40 A/m2 were 2.15, 3.873, 4.923 and 6.260, respectively for Al electrode, meanwhile Fe electrode the cell voltages were 4.464, 5.520, 9.660 and 12.40 V, respectively at the same current density conditions. Although increasing current density and operating time enhances the efficiency of EC process, it causes raise in the cell voltage, applied current, and eventually energy consumption. Energy consumptions at effluent phosphorus concentration of 0.01 mg/L (99.99% removal) at 10, 20, 30 and 40 A/m2 for Al and Fe electrodes were 0.5078 and 1.690 kWh/m3 , 1.1433 and 4.180 kWh/m3 , 1.308 and 11.7024 kWh/m3 , and 1.4786 and 11.266 kWh/m3 , respectively. On the other hand, electrode consumptions per m3 wastewater treated for Al and Fe electrodes were 0.1091 and >0.4733 kg/m3 at 10 A/m2 , 0.2182 and 0.884 kg/m3 at 20 A/m2 , 0.123 and 1.098 kg/m3 at 30 A/m2 , and 0.100 and 1.202 kg/m3 at 40 A/m2 , respectively. Above results showed that energy and electrode consumptions for Fe electrode was higher than Al electrode. Operating cost (OC) of the EC process are mainly material (electrodes) and electrical energy costs. The OC ($/m3 ) was calculated using Eq. (19): 3

OC($/m ) = ˛ × Cenergy + ˇ × Celectrode +  × Cchemicals (kWh/m3 )

(19)

and Celectrode is where Cenergy is energy consumption electrode consumption (kg/m3 ), ˛ is the unit electrical energy cost and ˇ is the cost of a unit electrode. As of August 2017, the price

P.I. Omwene, M. Kobya / Process Safety and Environmental Protection 116 (2018) 34–51

(a) Al electrode

50

Initial pHi 4 5 6 7

45 40 35 30 25 20 15 10 5 0 0

10

20

30

40

50

60

70

80

90

Residual phosphorous concentration (mg/L)

40

(b) Fe electrode

50

Initial pHi 4 5 6 7

45 40 35 30 25 20 15 10 5 0

100 110

0

10

20

30

40

50

60

70

80

90

100 110

Fig. 5. Variation of residual phosphorous concentration in the EC process for different initial pHi .

(a) Al electrode

13

10

(b) Fe electrode

12 11

8

10

7 6

Value of pHi 4 5 6 7

5 4 10

20

30

40

50 60 70 EC time (min)

80

90

9 8 7 Value of pHi

6

4 5 6 7

5 4

3 0

Effluent pHf

f

9

100 110

3

0

10

20

30

40

50

60

70

80

90

100

110

Fig. 6. Variation of effluent pHf during the EC process for different initial pHi .

of electricity in the Turkish market was 0.120 $/kWh and the electrodes prices were 4.10 $/kg and 0.952 $/kg for Al and Fe electrodes respectively. In addition, prices of chemicals (: technical NaOH and H2 SO4 ) were 1.01 $/kg and 0.40 $/kg, respectively. To obtain effluent phosphorus concentration of 0.01 mg/L (99.99% P removal), the operating costs were calculated as 0.493, 1.032, 0.662, and 0.588 $/m3 at 10, 20, 30, and 40 A/m2 respectively for Al electrode whereas for Fe electrodes the operating costs were calculated as 0.654, 1.343, 2.450 and 2.496 $/m3 at 10, 20, 30, and 40 A/m2 respectively.

4.2. Effect of initial pH on phosphorus removal The initial pH (pHi ) of the electrolyte is one of the important factors affecting the performance of EC process (Lacasa et al., 2011; Vasudevan et al., 2009; Irdemez et al., 2006a). The pH affects speciation of Al and Fe, hence a significant influence on the phosphorus removal mechanism. In this study, effect of initial pHi on the removal of phosphorus from the domestic wastewater was explored within the initial pHi range of 4.0–7.0, current density of 20 A/m2 , initial P concentration of 52 mg/L and EC time of 0–100 min. Fig. 5 shows the effect of initial pH on phosphorus removal.

As presented in Fig. 5, initial pHi influenced phosphorus removal by EC process. At EC time of 100 min; for initial pHi 4, residual P of 0.0001 mg/L (>99.99% removal) and 0.001 mg/L (also >99.99% removal) was obtained for Al and Fe electrodes respectively. At pHi 5, residual P of 1.18 mg/L (97.74% removal) and 0.001 mg/L (also >99.99% removal) was obtained for Al and Fe electrodes respectively. At pHi 6, residual P of 1.30 mg/L (97.51% removal) and 1.23 mg/L (97.64% removal) was obtained for Al and Fe electrodes respectively. At pHi 7, residual P of 4.84 mg/L (90.72% removal) and 3.18 mg/L (93.90% removal) was obtained for Al and Fe electrodes respectively. According to these results, the optimum conditions for Al electrode was obtained as initial pHi 4 and EC time of 50 min (q = 372C, j = 20 A/m2 ) whereas for Fe electrode the optimums were initial pHi 4–5 and EC time of 100 min (q = 954C, j = 20 A/m2 ). The effluent pHf values were measured as 8.52 for pHi 4 in the case of Al electrodes while for Fe electrodes, effluent pH was 10.62 and 11.06 for pHi of 4 and 5 respectively (Fig. 6). At these effluent pH values, the dominant phosphate forms are the monovalent H2 PO− 4

and divalent HPO2− 4 forms (Nguyen et al., 2016; Zhang et al., 2010; Szabo et al., 2008; Jiang and Graham, 1998). The formation of aluminium and iron hydroxyl complexes strongly depends on the pH of the solution in the EC reactor. Fig. 6 shows the final pHf variation with EC time for different initial pHi for Al and Fe electrodes. The pH increased with time and with

P.I. Omwene, M. Kobya / Process Safety and Environmental Protection 116 (2018) 34–51

appreciable effect of the initial pHi . The trend shows raise in pH until it approaches a value of approximately 10, then it starts to stabilize. This trend can be attributed to the buffering effect of − Al(OH)3 /Al(OH)− 4 and Fe(OH)3 /Fe(OH)4 in their respective solutions (Attour et al., 2014). The increase in pH during EC process could be interpreted in terms of reactions in the EC reactor such as cathodic water reduction and the chemical dissolution of electrodes. When aluminium electrodes are used in EC process, Al3+ and OH− ions are generated by reactions at anode and cathodes, as illustrated Eqs: (1) and (2) and subsequently form various aluminium species. The predominant species in solution at pH values below 3.5 is the aluminium ion, and above this pH value, monomeric and poly(3x−y)+ meric Al-hydroxyl species of the formAlx (OH)y (H2 O)n (such

4+ 3+ 4+ 4+ as Al(OH)2+ , Al(OH)+ 2 , Al2 (OH)2 , Al6 (OH)15 , Al7 (OH)17 , Al8 (OH)20 ,

5+ Al13 O4 (OH)7+ 12 , and Al13 (OH)34 ) occur, which transform finally into solid amorphous aluminium hydroxide, Al(OH)3(s) according to complex precipitation reactions (Ahmadzadeh et al., 2017; Can et al., 2003; Rebhun and Lurie, 1993). The minimum solubility, 0.03 mg Al/L, occurs at pH 6.3, with solubility increasing as the solution becomes more acidic or alkaline. The positively charged polyhydroxo-complexes such as Al8 (OH)4+ 20 , in the pH range between 4 and 7, are the effective flocculants (Rebhun and Lurie, 1993). These gelatinous charged Al-hydroxo cationic complexes can effectively remove pollutants by adsorption to produce charge neutralization, and by enmeshment in a precipitate. Above pH value of 10 solid aluminium hydroxide dissolves and the predominant species are Al(OH)− 4 ions (Georgantas and Grigoropoulou, 2007). This ion is soluble and directly affects the pollutants removal from wastewaters. On the other hand, the cathode may be chemically attacked by hydroxyl ions generated during H2(g) evolution at high pH values (Mouedhen et al., 2008; Picard et al., 2000):

Al(OH)3(s) + OH− → Al(OH)− 4 −

2Al + 6H2 O + 2OH →

2Al(OH)− 4

(20) + 3H2

(21)

For the case of Fe electrode in EC process, electrochemically generated ferric ions by anode-cathode reactions illustrated in Eqs. (4)–(6) may form monomeric and polymeric species or ferric hydroxyl complexes such as FeOH2+ , Fe(OH)+ 2, + + 4+ 2+ Fe2 (OH)4+ 2 , Fe(H2 O)2 , Fe(H2 O)5 OH , Fe(H2 O)4 OH2 , Fe(H2 O)8 OH2

and Fe2 (H2 O)6 OH2+ which finally transform into Fe(OH)3(s) 4 depending on the pH range, with Fe(OH)− 4 as the dominant species in solution at above pH 10 (Pykhteev et al., 1999). The formed − monomeric species such as Al(OH)− 4 and Fe(OH)4 at above of pH 10 in EC process are soluble and therefore not useful for adsorption of phosphorus. Studies by Karageorgiou et al. (2007) indicated 2− that H2 PO− 4 and HPO4 species are predominant in the pH region

2− 5–10, with H2 PO− 4 and HPO4 species seen in the pH ranges 5–7

and 7–10, respectively. In addition, for pH range 10–12, HPO2− 4 was

predominant over PO3− 4 species meanwhile, for pH above 12.5, the

2− PO3− 4 become of significant concentration HPO4 . The Al(OH)3(s) and Fe(OH)3(s) compounds are less soluble in the pH range 4–7, where their formation is favoured. In addition, their generation promotes formation of positively charged hydroxide solution (on aluminium and iron oxide surfaces) capable − of adsorbing HPO2− 4 and H2 PO4 anions, and enhancing adsorption of anionic phosphate species by electrostatic attraction and ligand exchange. Adsorption of phosphates onto aluminium and iron hydroxides is generally sought to be through ligand exchange between the phosphates and surface hydroxyl groups, because most negatively charged species are easily adsorbed by aluminium and iron hydroxides (Pulkka et al., 2014; Song et al., 2011). However, at higher pH values, the surface charge of the sol (Al or Fe hydroxides) decreases, impairing its ability to adsorb and thus

41

remove phosphate anions. This explains why low pH values are advantageous to phosphate ions adsorption as this promotes anion adsorption and release of hydroxyl anions (Song et al., 2011). Moreover, phosphate adsorption only occurs at the active sites of the Al(OH)3(s) and Fe(OH)3(s) surfaces. At low pH, the surface hydroxyl groups are protonized and (−OH+ ) and is easier to replace at the 2 binding sites compared with the hydroxyl groups. On the other hand, monomeric and polymeric Al and Fe complexes formation at between pH 4–10 could form insoluble compounds such as AlPO4(s) , Al1.4 PO4 (OH)1.2(s) , Al3 (OH)3 (PO4 )2(s) , FePO4(s) , Fe3 (PO4 )2(s) and Fe2.5 PO4 (OH)4.5(s) according to the reactions, Eqs. (9)–(12). During EC process, continuous generation of fresh Al or Fe hydroxides in the EC reactor occurs. This implies an increase in capacity for reactive phosphorus removal. However, some of the formed Al(OH)3(s) and Fe(OH)3(s) during the EC process age out with time and their capacity to remove phosphorus decrease. The two generally considered main mechanisms for phosphorous removal are precipitation (for pH < 6.5) and adsorption (for pH > 6.5). The amount of experimental coagulant from Al electrode at EC time of 100 min were found as 0.1528 g (ϕAl = 220.3%) at pHi 4, 0.1123 g (ϕAl = 162%) at pHi 5, 0.1373 g (ϕAl = 198%) at pHi 6 and 0.1165 g (ϕAl = 168%) at pHi 7. In this case, the amount of removed P per electrochemically generated Al at pHi 4, 5, 6 and 7 were calculated as 119.43, 158.74, 129.57 and 142.07 mg/g, respectively. The phosphorus removal capacity at EC time of 50 min and pHi 4 was 238.86 mg/g Al (amount of removed P of 52.08 mg and coagulant dosage of 0.0764 g Al). On the other hand, the current efficiency and amount of actual coagulant at EC time of 100 min for Fe electrode were found as 112% and 0.3092 g respectively at pHi 4, 120% and 0.3313 g respectively at pHi 5, 124% and 0.3424 g respectively at pHi 6, and 130% and 0.3589 g respectively at pHi 7. The phosphorus removal capacity at these conditions for pHi 4, 5, 6 and 7 were calculated as 59.01, 55.07, 52.04 and 47.73 mg P/g Fe, respectively. As seen in literature, ferric oxides and metal oxides of aluminium have more affinity for the monovalent phosphate forms. Razali et al. (2007) reported that the adsorption capacity of orthophosphate from typical municipal wastewater by drinking-water treatment alum sludge was 10.2 mg/g sludge at pH 4 and initial P concentration of 14.7 mg PO4 /L. Studies by Yang et al. (2008) showed phosphorus adsorption capacity of dewatered alum sludge as 21.4–23.9 mg/g at pH 4.3, 14.3–14.9 mg/g at pH 7, and 0.9–1.1 mg/g at pH 9. In addition, orthophosphates adsorption capacity from synthetic solutions with alum sludge was reported as 4.52 mg/g at pH 4 (Babatunde et al., 2008). Barthelemy et al. (2012) reported a decrease in the phosphate adsorption capacity of carbonated ferric green rust with increase in pH in the pH range 4–9, and the highest adsorption capacity was noted as 64.8 mg/g at pH 4 (and as 30.5 mg/g at pH 9). Peleka and Deliyanni (2009) obtained the maximum sorption capacity of goethite (␣-FeOOH; specific surface area of 316 m2 /g, average size of grains of <70 ␮m, pore volume, 0.22 cm3 /g and mean pore diameter 1.39 nm) at pH 5 and 25 ◦ C as 144 mg/g. Similar studies by other researchers have indicated aluminium oxides as more effective adsorbent than iron oxides, this is attributed to the higher sorption capacity of the former. The interactions of iron hydroxide with phosphorus are mainly electrostatic whereas for aluminium hydroxide interactions are through both electrostatic forces and formation of hydrogen bonds with the orthophosphate anion. The Fe to P mole ratio between EC time of 2.5–100 min varied as 0.48–3.29 at pHi 4, 0.75–3.53 at pHi 5, 1.65–3.73 at pHi 6, and 1.21–4.07 at pHi 7 (Fig. 7). As seen in Fig. 7, Fe-to-P ratio was increased by increasing of EC time (except for EC time of <15 min at pHi 6). Increase in EC time means an increase in both Phosphorous removal and amount of produced coagulant in EC process, this in turn increased the Fe-to-P ratio. For effluent P concentration of 0.01 mg/L (at tEC = 100 min), the Fe-to-P rations at pHi 4 and 5 were 3.29 and 3.53 mol/mol, respectively. Basing on this

42

P.I. Omwene, M. Kobya / Process Safety and Environmental Protection 116 (2018) 34–51

3.5

(b) Fe electrode

(a) Al electrode

4.0 Fe-to-phosphorous molar ratio (nFe/nP)

3.0 2.5 2.0 1.5 Initial pHi

1.0

4 5 6 7

0.5 0.0

0

10

20

30

40

50 60 70 EC time (min)

80

90

100

3.5 3.0 2.5 2.0 1.5

4 5 6 7

1.0 0.5 0.0

110

Initial pHi

0

10

20

30

40

50

60

70

80

90

100

110

Al electrode

50

55

Al electrode

50 Residual phosphorous concentration (mg/L)

45 40 35 30 25 20 15 10

Ci (mg/L) 5 10 25 50

45 40 35 30 25 20 15 10 5 0 0

100

200

5

300

400

500

600

700

800

Charge loading (C)

(b) Fe electrode 55

50

Residual phosphorus concentration (mg/L)

55

Residual phosphorous concentration(mg/L)

Fig. 7. Variation of metal-to-phosphorous mole ratio in EC process for different initial pHi .

45 40 35 30 25 20

Fe electrode

50 40 35 30 25 20 15 10 5 0 0

15

C i (mg/L) 5 10 25 52

45

100

200

300

400

500

600

700

800

Charge loading (C)

10 5 0

0 0

10

20

30

40

50

60

70

80

90

0

10

20

30

40

50

60

70

80

Fig. 8. Variation of residual phosphorous concentration and charge loading for different initial phosphorous concentrations.

study, Fe electrode showed an increase in Fe-to-P molar ratio as initial pHi increased. For attainment of 0.01 mg/L effluent P (>99.99% P removal), Fe-to-P ratio of above 3.53 mol/mol was required. On the other hand, Al-to-P mole ratio varied from 1.35 to 0.71 for change in EC time from 5–20 min at pHi 4, this ratio then increased to 1.68 and 3.37 at EC time of 50 and 100 min respectively (Fig. 7). Generally, for residual P concentration of 0.01 mg/L, Al-to-P ratios were obtained as less than 3.37 mol/mol, and the required Me/P ratio for Fe electrode was about twice higher than from Al electrode, except for initial pHi of 4. Energy and electrode consumptions for Al electrode at initial pHi of 4, 5, 6 and 7 were obtained as 1.1433 kWh/m3 and 0.2183 kg/m3 (tEC = 50 min), 2.486 kWh/m3 and 0.321 kg/m3 (tEC = 100 min), 2.487 kWh/m3 and 0.392 kg/m3 (tEC = 100 min), and 2.699 kWh/m3 and 0.333 kg/m3 (tEC = 100 min), respectively. In the case of Fe electrode, energy and electrode consumptions at initial pHi of 4, 5, 6 and 7 were calculated as 4.1794 kWh/m3 and 0.3092 kg/m3 (tEC = 100 min), 4.893 kWh/m3 and 0.947 kg/m3 (tEC = 100 min), 4.971 kWh/m3 and 0.978 kg/m3 (tEC = 100 min), and 5.799 kWh/m3 and 1.026 kg/m3 at pHi 7 (tEC = 100 min), respectively. At these conditions, operating costs at initial pHi of 4, 5, 6 and 7 were calculated as 1.032, 1.614, 1.908, and 1.688 $/m3 , respectively for Al electrode and 1.343, 1.488, 1.528, and 1.672 $/m3 , respectively for Fe electrode. 4.3. Effect of initial phosphorus concentration on removal Effect of initial phosphorus concentration (Ci , mg/L) on phosphorus removal by EC process was studied at initial P concen-

trations (Ci ) of 5, 10, 25 and 52 mg/L, initial pHi of 4.0 and current density of 20 A/m2 . Variations of residual phosphorus concentration with respect to charge loading and EC time at different initial phosphorus concentrations for both Al and Fe electrodes are illustrated in Fig. 8. The rate of P removal was sharp at the beginning of the EC process (2.5–15 min) and then proceeded gradually. At the beginning (i.e., lag stage) of the EC process, the amount of electrochemically produced coagulant would be less to cause decrease in phosphorus concentration in the domestic wastewater, especially for high initial phosphorus concentrations. However, with increase in the EC time, more coagulant (hydrous aluminium or ferric oxides) was produced and subsequently adsorbed the orthophosphates hence decreasing the effluent P concentration. The curves show a similar trend at the end of the experiment, this is attributed to abundance of generated hydrous metal oxides. It also can be seen that removal efficiency decreased with increase in initial phosphorus concentration (Fig. 8) and the required EC time, also charge loading (q) for sufficient P removal increased with increase in initial P concentration for both electrodes. When Fe electrode was used with Ci of 5 mg/L, phosphorus removal efficiency (Re , %) increased from 2.6% (Cf = 4.87 mg/L) to 64.8% (Cf = 1.76 mg/L) for EC time change from 2.5 min (q = 23.85C) to 7.5 min (q = 95.4C). At these conditions, effluent P concentration of 0.01 mg/L (Re = 99.98% and q = 190.8C) was subsequently reached at EC time of 20 min (Fig. 8). In addition, for initial P concentrations of 10 and 25 mg/L with Fe electrodes, required EC times to obtain Cf = 0.01 mg/L were determined as 60 min (Re = 99.99% and q = 572.4C) and 80 min (Re = 99.996% and q = 763.2C), respectively. For Ci = 52 mg/L, EC time of 100 min (q = 954C) was needed

P.I. Omwene, M. Kobya / Process Safety and Environmental Protection 116 (2018) 34–51

(a) Al electrode

1100

500

Ci (mg/L) 5 10 25 52

900 800

(b) Fe electrode

400

Removal capacity (mg P/g Fe)

1000

43

700 600 500 400 300 200

300

Ci (mg/L) 5 10 25 52

200

100

100 0

0

100

200

300 400 500 600 Charge loading ( Coulomb)

700

0

800

0

100

200

300

400

500

600

700

800

900

Fig. 9. Phosphorus removal capacity for different initial phosphorous concentrations.

5.0

(a) Al electrode

Iron-to-Phosphorous molar ratio (nFe/nP)

4.0 3.5 3.0 2.5 2.0 Ci (mg/L)

1.5

5 10 25 52

1.0 0.5 0.0

0

100

(b) Fe electrode

10

4.5

200

300 400 500 Charge loading ( Coulomb)

600

700

800

Ci (mg/L) 5 10 25 52

9 8 7 6 5 4 3 2 1 0

0

100

200

300

400

500

600

700

800

900

Fig. 10. Variation of metal-to-phosphorous mole ratio for different initial phosphorous concentrations.

to achieve P removal efficiency of above 99.99% for both electrodes. Besides, the charge loading required to achieve over 99% P removal by Al electrodes for Ci of 5, 10, 25 and 52 mg/L was calculated as 148.8C (at 20 min), 223.2C (at 30 min), 223.2C (at 30 min) and 372C (at 50 min), respectively. Al electrode required lesser EC time and charge loading to achieve the same removal efficiency compared to Fe electrode. Higher removal efficiency was observed at lower initial phosphorus concentrations of 5 and 10 mg/L within about 15 min of EC time compared to higher initial P concentrations of 25 and 52 mg/L (>20 min of EC time). Since the analysis of the effect of initial phosphorous concentration was done at constant current density, the generation rate of the metal ions was the same for the experimental runs. Consequently, for higher phosphorous concentrations the generated amount of metal hydroxides was inadequate to coagulate the larger number of phosphate molecules present hence more EC time was required to achieve higher removal efficiency. This agrees with literature studies (Shalaby et al., 2014). In this study, Faradic and experimental electrode consumptions to achieve final phosphorus concentration of 0.01 mg P/L for Fe electrode with Ci of 5, 10, 25, and 52 mg/L were calculated as 0.0552 and 0.0635 g (ϕFe = 115%), 0.1657 and 0.1822 g (ϕFe = 110%), 0.2209 and 0.2341 g (ϕFe = 106%), and 0.2761 and 0.3092 g (ϕFe = 112%), respectively. On the other hand, the Faradic and experimental electrode consumptions to achieve the same P removal by Al electrode with Ci of 5, 10, 25, and 52 mg/L were calculated as 0.0139 and 0.0187 g (ϕAl = 135%), 0.0208 and 0.0342 g (ϕAl = 165%), 0.0208 and 0.02954 g

(ϕAl = 142%), and 0.0347 and 0.0444 g (ϕAl = 128%), respectively. From these results, the experimental (actual) electrode consumptions for both electrodes were greater than the theoretical values, this implies that the current efficiencies were above 100%, with current efficiencies of Al electrode notably higher than iron electrode. As illustrated in Fig. 9, the optimum charge loading for Fe and Al electrode for Ci of 5, 10, 25 and 52 mg/L were determined as 190.8 and 148.8C (at 20 min), 572.4C (at 60 min) and 223.2C (30 min), 763.2C (at 80 min) and 223.2C (at 30 min), and 763.2C (at 80 min) and 372C (at 50 min), respectively. Moreover, phosphorus removal capacity per gram electrochemically dissolved metal electrode for over 99% P removal was obtained as 27.552 mg P/g Fe and 93.446 mg P/g Al for Ci = 5 mg/L; 19.205 mg P/g Fe and 101.952 mg P/g Al for Ci = 10 mg/L; 37.371 mg P/g Fe and 88.858 mg P/g Al for Ci = 25 mg/L; 59.001 mg P/g Fe and 238.59 mg P/g Al for Ci = 52 mg/L. The variation of Al-to-P and Fe-to-P ratios with charge loading at different Ci values of 5, 10, 25 and 52 mg/L is illustrated in Fig. 10. The Al-to-P mole ratio for Ci = 5 mg/L decreased from 4.48 to 2.12 as EC time increased from 2 min (q = 14.88C) to 4 min (q = 29.76C), after this point the ratio increased to 4.30 at 148.8C (at 20 min). The same trends were observed for other initial P concentrations, for example at Ci = 10 mg/L; the Al: P mole ratio reduced from 4.33 to 2.04 for change in EC time from 2.5 min (18.6C) to 15 min (111.6C), and Al: P mole ratio = 3.95 at EC tome of 30 min (223.2C). At residual P concentration of 0.01 mg/L with Fe electrode, energy consumptions per m3 treated wastewater for Ci of 5, 10, 25 and 52 mg/L were calculated as 1.203, 3.608, 4.811 and

44

P.I. Omwene, M. Kobya / Process Safety and Environmental Protection 116 (2018) 34–51

Fig. 11. SEM images and FTIR spectra of the sludge produced in the EC process; (a) Al electrode SEM image, (b) Fe electrode SEM image, and (c) FTIR spectra for both Al and Fe electrodes.

4.180 kWh/m3 , respectively whereas the electrode consumptions were determined to be 0.181, 0.521, 0.669 and 0.884 kg/m3 , respectively. Likewise, for the case of Al electrode at the same conditions, energy consumptions were 1.117, 1.290, 0.945 and 1.143 kWh/m3 , respectively whereas the electrode consumptions were 0.0535, 0.0981, 0.0844, and 0.218 kg/m3 , respectively. The operating costs with Ci of 5, 10, 25 and 52 mg/L were 0.317, 0.929, 0.911 and 1.343 $/m3 , respectively for Fe electrode, whereas for Al electrode operating costs were obtained to be 0.354, 0.460, 0.557 and 1.032 $/m3 , respectively at the same conditions. Residual metal (Al and Fe) concentrations are an important consideration for any wastewater treatment by EC process using Al and Fe electrodes. The residual metal concentrations may decrease carrying capacity of the sewer system due to precipitation of aluminium and iron hydroxides. Moreover, residual metal concentrations cause toxic effects on organisms and increase turbidity in receiving water environments such as river and lake. There are also concerns that residual aluminium in drinking water may cause Alzheimer’s disease. In this study, residual aluminium and iron concentrations in the treated samples by EC phosphorus removal process at current densities of 10, 20, 30, and 40 A/m2 and EC time of 100 min, were obtained as 0.024, 0.029, 0.042, and 0.051 mg/L, respectively for Al electrodes and 0.113, 0.128, 0.164, and 0.211 mg/L, respectively for Fe electrodes. The residual Al and Fe concentrations in the treated effluents at different current densities were lower than the permissible limit for aluminium and iron in drinking waters (0.3 mg/L) set by Turkish Water Pollution Control Regulation. On the other hand, the permissible limits for sewer

system discharging of industrial wastewater after the treatment were 3 mg/L for Al and 5 mg/L for Fe. Electrocoagulation sludge is reported to be of a better quality in terms of floc size, acid-resistance of flocs and dewatering potential than conventional chemical coagulation sludge, making it easy to handle and process (Kuokkanen et al., 2015). Therefore, the resulting electrocoagulation sludges may be processed for application as fertilizers especially for wastewaters with high phosphate concentration. 4.4. Analysis of generated sludge from the electrocoagulation process The colour of the dried generated Fe electrodes precipitate was red–brown whereas for Al electrode, the dried sludge was whitish grey. The results therefore suggest that in the case of Fe electrodes, the formed precipitate consisted of iron phosphate and hydroxyl iron phosphate, whereas for Al electrodes, the formed precipitated consisted of aluminium phosphate and hydroxyl aluminium phosphate. The hydroxyl phosphate groups have good adsorption and coagulation, which help to improve the rate of phosphorus removal from wastewater (Wang et al., 2016). The SEM observation of the sludges from electrocoagulation is shown in Fig. 11(a–b), which indicates that anomalistic floccules exist in the porous surface for Al electrode generated sludge, whereas for Fe electrode, the results did not show any classical well-crystalline appearance on sludge surface. The FTIR spectrum and the scanning electron microscope (SEM) images of the electro-coagulated sludge from both Al and Fe

P.I. Omwene, M. Kobya / Process Safety and Environmental Protection 116 (2018) 34–51

45

Fig. 12. X-ray diffraction pattern of generated EC sludge: (a) Al electode and (b) Fe lectode.

plate electrodes is shown in Fig. 11. It can be seen that the FTIR spectrum is characterized by a broad peak at about 3351.7 1/cm for both sludges, this is attributed the vibration of the OH group in the precipitates (Yan et al., 2010). The strong absorption peak seen at about 1009.8 cm−1 is attributed to the bending vibration of Me–O–P group in the metal phosphate molecule (Parikh et al., 2014). Fig. 12(a–b) illustrates The X-ray diffraction patterns of Al and Fe electrogenerated sludges. It can be seen that the XRD pattern for Al electrogenerated sludge exhibited strong output signals with distinct peaks. However, the XRD results for Fe electrogenerated sludge showed no a quantitative description of the sludge structure, with no sharp characteristic diffraction peaks over a wide

range of d-spacings (15–70◦ , 2), this indicates poor arrangement of particles with ferric sludge (Yang et al. 2006; Berkowitz et al. 2005). 4.5. Comparison with results in the literature In the current literature, many metal electrodes such as iron, aluminium and Al/Fe hybrid have been used to remove phosphorus from waters using the EC process. The results on phosphorus removal from synthetic water solutions and, different industrial and municipal wastewaters in the literature were given in Tables 1–3. Phosphorus removal from wastewater solutions var-

46

P.I. Omwene, M. Kobya / Process Safety and Environmental Protection 116 (2018) 34–51

Table 1 An overview of literature studies on phosphorus removal from wastewaters by EC process using Fe electrodes. Wastewater type

Operating conditions

Optimum values

Re (%)

Energy and electrode consumptions, and OC

Ref.

SW

BR, synthetic human urine, j = 3 A/m2 , pHi = 2–8, tEC = 60 min, Fe/P = 0.3–5, N = 4, d = 0.3 cm

pHi = 8

99

OC = 0.20 D /m3

Inan and Alaydın (2014)

SW

BR, CNaCl = 0.5–2 g/L, i = 0.5–2 A, pHi = 3–9, d = 1.8 cm, Sanode = 32 cm2 (for anode of 1), tEC = 0–70 min, Ci = 10 mg/L, N = 2–4

Fe/P = 2.5 pHi = 6

>96



Bouamra et al. (2012)

i = 1.5 A tEC = 70 min CNaCl = 2 g/L N=4 j = 400 A/m2

98

Cen = 0.0042 kWh/m3

Zheng et al. (2009)

Human urine

SW

SW

j = 100–500 A/m2 , tEC = 0–30 min, Sanode = 160 cm2 , Ci = 550 mg/L, urine-tap water dilution ratio (DR) = 1:1–1:10, N = 2; ω = 150 rpm, pHi = 6–7, pHf = 9.32

BR, j = 1–5 A/m2 , d = 0.5 cm, Sanode = 200 cm2 , tEC = 5–45 min, Ci = 10–50 mg/L, pHi = 2–12, N = 2, T = 305 K

BR, j = 10 A/m2 , d = 0.5 cm, Sanode = 1500 cm2 , tEC = 20 min, Ci = 100 mg/L, pHi = 3–9, N = 6

d = 0.5 cm tEC = 20 min DR = 0.781:1 Fe/P = 1.7–2.5 pHi = 6.5

j = 5 A/m2 tEC = 30 min T = 305 K Ci = 100 mg/L pHi = 3

Cel = 2.4 kg Fe/kg P

98

Cen = 3.75 kWh/m3

Vasudevan et al. (2008)

qe = 10–48.5 mg/g

86

Cen = 0.294–1.135 kWh/m3

Irdemez et al. (2006a)

j = 10 A/m2 tEC = 8 min BR: batch EC reactor, Cen : energy consumption, Cel : electrode consumption, Ci = initial P concentration, CNaCl = amount of NaCl, CR: continuous EC reactor, d: distance between electrodes, HRT: hydraulic retention time, IW: industrial wastewater, j: current density, N: number of electrodes,OC: operating cost, pHi : initial pH, Q: flowrate of wastewater in the continuous reactor, qe : adsorption capacity, Selectrode : effective electrode surface area, SW: synthetic wastewater, T: temperature, tEC : EC time, U: average voltage between electrodes, : conductivity, ω: stirring speed.

ied from 84.7% to 99.7% for Fe electrodes (Tables 1 and 3) and from 87.1% to 100% for Al electrodes (Tables 2A, 2B and 3). As per these results, optimum conditions for phosphorus removal by Fe and Al electrodes were obtained as initial pHi of 3–8 and 3–8.8, current efficiency of 3–469 and 10–389 A/m2 , and EC time of 8–70 and 5–120 min, respectively. These studies showed Al electrodes to be more effective than iron electrodes for phosphorus removal. P removal efficiency by Fe/Al hybrid electrodes was 71–100% (Table 3) at pHi of 4–8, current density of 20–100 A/m2 , and EC time of 15–80 min. In this study, EC process was performed with Al and Fe electrodes; for of Ci = 52 mg/L, to obtain effluent phosphorus concentration of 0.01 mg P/L the optimum conditions were; j = 20 A/m2 , pHi = 4 and EC time = 50 min. At these conditions, energy consumption, electrode consumption, operating cost, Al-to-P mole ratio and amount of removed P per g coagulant were obtained as 1.143 kWh/m3 , 1.143 kg/m3 , 1.032 $/m3 , 1.684 and 239 mg P/g Al, respectively for Al electrode at EC time of 50 min (or charge loading of 372 C). Whereas for Fe electrode, energy consumption, electrode consumption, operating cost, Fe-to-P mole ratio and amount of removed P per g coagulant were obtained as 4.179 kWh/m3 , 0.884 kg/m3 , 1.343 $/m3 , 3.293 and 59 mg P/g Fe, respectively at EC time of 100 min (954 C). Pulkka et al. (2014) reviewed the removal of different anions, including cyanide, fluoride, nitrate, nitrite, phosphate, and sulphate, from water by EC process. The cyanide removal efficiency was achieved as 93% by Fe-Al hybrid electrode at operating conditions of 20 min, 150 A/m2 and initial cyanide concentration of 300 mg/L. Moreover, iron electrodes were more effective in removal

of cyanide ions since oxidation potential of iron electrode is higher than that of aluminium electrodes (Moussavi et al., 2011). According to the literature results, the fluoride removal efficiency was observed to vary from 60 to 100% with Al electrodes at initial F concentration of 2.5–25 mg/L and for initial pHi range 4–7 (Pulkka et al., 2014). The fluoride removal mechanism was attibuted to the competitive adsorption between OH− and F− and fluoride precipitation as cryolite (Al(OH)3−x Fx ) (Palahouane et al., 2015). Studies on treatmnet of synthetic wastewater containing 109 mg/L nitrate ions and electrolyte of 3000 mg/L Na2 SO4 by a batch EC reactor using two iron or aluminium electrodes obtained phosphorus removal efficiency of 100% at 1–50 A/m2 . When aluminium plates were used, the growing flocks of aluminium hydroxide adsorbed the formed ammonium ions (Lacasa et al., 2011). As a result, the main mechanisms in removal of anions like phosphate by EC process using Fe and Al electrodes is adsorption, charge neutralization and coprecipitation. Tanada et al. (2003) proved that the adsorption of phosphate on aluminium hydroxide was up to about 7000 times higher than the adsorption of ions such as chloride, nitrate, carbonate and sulphate. Ions of higher charge in the solution are more easily adsorbed on aluminium hydroxide. 5. Conclusion This research investigated the effects of different operating parameters on phosphorus removal from synthetic domestic wastewater solution by Al and Fe plate anodes, and titanium plate cathode in a batch EC reactor. The obtained results showed

P.I. Omwene, M. Kobya / Process Safety and Environmental Protection 116 (2018) 34–51

47

Table 2A An overview of literature studies on phosphorus removal from wastewaters by EC process using Al electrodes. Wastewater type

Operating conditions

Optimum conditions

Re (%)

Energy and electrode consumptions, and OC

Ref.

SW

CR, Landscape water, U = 1.93–17.2 V, Q = 0.3 L/min, tEC = 20 min, TP = 16 mg/L, Ci = 16–55 mg/L, CNaCl = 0–600 mg/L d = 1.5–6.0 cm

d = 2.5 cm

99.9

Cen = 0.71 kWh/m3

Zhang et al. (2013)

SW

BR, j = 24.5–98.2 A/m2 , pHi = 5–8, tEC = 0–160 min, d = 0.5 cm, fixed bed of Al raschig rings anodes, Ci = 40 mg/L, CNaCl = 0.5–7 g/L, Ci = 5–50 mg/L, T = 25–45 ◦ C

 = 0.77–0.114 mS/cm pHi = 7

100



Nassef (2012)

j = 85.9 A/m2 tEC = 30 min CNaCl = 1 g/L T = 25 ◦ C pHi = 3

100

Cen = 4 kWh/m3

Attour et al. (2014)

j = ≥100 A/m2 tEC = 120 min d = 0.5 cm T ≥ 30 K  = 1 mS/cm pHi = 7

99

Cen = 10 kWh/m3

Vasudevan et al. (2009)

SW

SW

SW

SW

SW

IW Urban wastewater

BR, j = 20–180 A/m2 , d = 0.5–2 cm, Sanode = 50 cm2 , tEC = 0–140 min, Ci = 100 mg/L, pHi = 2–11, N = 2,  = 0.8–3.2 mS/cm

BR, j = 2–20 A/m2 , d = 0.5 cm, Sanode = 200 cm2 , tEC = 5–50 min, Ci = 25–100 mg/L, pHi = 3–10, N = 2, T = 293–333 K

BR, j = 2.5–10 A/m2 , d = 0.5 cm, Sanode = 1500 cm2 , tEC = 20 min, Ci = 100 mg/L, pHi = 3, N = 6

BR, j = 25–100 A/m2 , d = 0.3 cm, Sanode = 1500 cm2 , tEC = 15 min, Ci = 10–200 mg/L, pHi = 4–8, N = 8

BR, N = 2 Al, d = 2.5 cm, pHi = 3–9, tEC = 0–60 min, i = 0.1–2 A, Ci = 0.07–1.75 mg/L, Sanode = 13.32 cm2 ,  = 0.15–0.88 mS/cm BR, N = 2, d = 3 cm, Sanode = 73 cm2 , j = 50–200 A/m2 , pHi = 7.4, Ci = 15 mg/L, initial COD = 1490 mg/L

j = 20 A/m2 tEC = 30 min Ci = 100 mg/L T = 305 K pHi = 3

90

Cen = 14.866 kWh/m3

Irdemez et al. (2006b)

j = 10 A/m2 tEC = 12 min pHi = 6.2

100

Cen = 0.8–1.5 kWh/m3

Bektas et al. (2004)

j = 75 A/m2 tEC = 15 min pHi = 5–8.8

99



Franco et al. (2017)

tEC = 60 min j = 200 A/m2

97

Cen = 0.6 kWh/kg P

Elazzouzi et al. (2017)

j = 20 A/m2

91.1

Cen = 6 kWh/kg COD OC = 0.35 $/kg P OC = 0.90 $/kg COD OC = 0.070 $/m3

HRT = 5 min COD = 274 mg/L j = ≥20 A/m2

99.96

Cen = 2.294 kWh/m3

Ince (2013)

tEC = 35 min pHi = 7

99

Cel = 5.584 kg/m3 –

Mahvi et al. (2011)

qe = 124–505 mg/g

tEC = 50 min

Municipal wastewater

CR, N = 2 (anode) and 2 (cathode), pHi = 7.1–7.9, HRT = 5–15 min (12.96–4.32 L/h), j = 20–80 A/m2 , Sanode = 255 cm2 , Ci = 4.5–6.4 mg P/L, initial COD = 200–350 mg/L,  = 3.1–3.3 mS/cm

IW

BR, j = 2.5–30 A/m2 , d = 1 cm, Sanode = 225 cm2 , tEC = 2.5–60 min, N = 4, Ci = 355 mg/L, pHi = 3.12

IW

CR, N = 2 (RuO2 /Ti, stainless steel cathodes) and 3 Al (anode), pHi = 5–10, tEC = 0–60 min, i = 1–3 A, Ci = 15–100 mg/L,  = 1.6 mS/cm

Makwana and Ahammed (2016)

48

P.I. Omwene, M. Kobya / Process Safety and Environmental Protection 116 (2018) 34–51

Table 2A (Continued) Wastewater type

IW

Operating conditions

BR, j = 10–30 A/m2 , d = 1.3 cm, Sanode = 225 cm2 , tEC = 0–60 min, pHi = 3.04, N = 4, Ci = 32.1 mg/L,  = 6.4 mS/cm

Optimum conditions

Re (%)

Energy and electrode consumptions, and OC

Ref.

i=3A tEC = 60 min Ci = 50 mg/L j = 20 A/m2

99.8

Cel = 0.1432 kg/m3

Kobya and Demirbas (2015)

Cen = 1.366 kWh/m3 OC = 0.366 D /m3

tEC = 40 min

Table 2B An overview of literature studies on phosphorus removal from wastewaters by EC process using Al and Fe electrodes. Wastewater type

Operating conditions

Optimum values

Re (%)

Energy and electrode consumptions, and OC

Ref.

Swine wastewater

CR, the exit of anaerobic up-flow, j = 163–575 A/m2 , HRT = 31.8–88.2 min for Al and 61.8–118.2 min for Fe, d = 2 cm, N = 5 (2 anode and 3 cathode), pHi = 6 (Al) and 7 (Fe)

HRT = 31.8 min (Al)

87.1 (Al)

Cen = 4400 kWh/m3 (Al)

Mores et al. (2016)

HRT = 70 min (Fe) j = 389 A/m2 (Al) j = 278 A/m2 (Fe)

96.3 (Fe)

pHi = 3

100 (Al)

Cen = 4200 kWh/m3 (Fe) Cel = 0.62 g (Al) Ce l = 0.98 g (Fe) pHf = 7.8 (Al) pHf = 8.5 (Fe) Cel,Al = 0.7515 kg/m3

Behbahani et al. (2011)

j = 250 A/m2 Ci = 400 mg/L j = ≥30 A/m2

84.7 (Fe)

Cel,Fe = 1.031 kg/m3

99.6

Cen,Al = 0.06–0.73 kWh/m3

Lacasa et al. (2011)

pHi = 5 (Al),

97.7 (Fe)

Cen,Fe = 0.05–4.38 kWh/m3 Cel,Al = 0.0188–0.0674 kg/m3 Cel,Fe = 0.0274–0.5864 kg/m3 Cel,Al = 0.01–0.35 kg/m3

Kobya et al. (2010)

pHi = 3 (Fe) j = 60 A/m2

99.8 (Al)

Cel,Fe = 0.2–0.62 kg/m3 Cen = 0.18–11.29 kWh/m3 (Al) Cen = 0.24–8.47 kWh/m3 (Fe) OC = 0.046–2.16 $/m3 (Al) OC = 0.149–1.39 $/m3 (Fe) Cen = 8.93 kWh/m3 (Al)

Kobya et al. (2010)

Cen = 11.30 kWh/m3 (Fe) Cel = 0.27 kg/m3 (Al) Cel = 0.48 kg/m3 (Fe) OC = 9.50 $/m3 (Al) OC = 11.50 $/m3 (Fe) –

Gharibi et al. (2010)

SW

SW

BR, j = 83–250 A/m2 , d = 0.3 cm, Sanode = 240 cm2 , tEC = 5–40 min, Ci = 25–400 mg/L, N = 4

BR, j = 10–50 A/m2 , d = 0.9 cm, Sanode = 100 cm2 , tEC = 100–550 min, Ci = 27 mg/L, N = 2

tEC = 120 min (Fe) tEC = 40 min (Al) IW

BR, j = 10–100 A/m2 , d = 1.1 cm, Sanode = 146.3 cm2 , tEC = 10–100 min, Ci = 120 mg/L, N = 4, pHi = 2–9

tEC = 25 min (Al) tEC = 15 min (Fe) IW

CR, j = 60 A/m2 , d = 2 cm, Sanode = 660 cm2 , tEC = 0–80 min, Ci = 120 mg/L, pHi = 3 (Fe), pHi = 5 (Al), N = 4, Q = 50–400 mL/min

Q = 0.10 L/min

99.6 (Fe)

99.9 (Al)

SW

BR, U = 20–40 V, tEC = 0–60 min, pHi = 3–10, Ci = 4–10 mg/L, d = 1.5 cm, N = 2

pHi = 8.5 (Al)

>99 (Fe)

pHi = 5.5 (Fe) U = 40 V, tEC = 40 min

>99 (Al)

that phosphate removal greatly increased with increase of current density and EC time. At initial pHi of 4 and initial phosphorus concentration of 52 mg/L for both electrodes, required EC time to obtain effluent phosphorus concentration of 0.01 mg/L decreased from 60 min to 20 min for Al electrode and from >100 min to 60 min for Fe electrode when current density was increased from 10 to 40 A/m2 . The Al/P mole ratio at current densities of 10, 20, 30 and 40 A/m2 was obtained as 0.84, 1.68, 0.95 and 0.77, respectively, while the Fe/P mole ratio for these current density conditions

was calculated as 1.81, 3.29, 4.09 and 4.48, respectively. Overall, Al electrode required lesser metal-to-P mole ratio compared to Fe electrode for effective phosphorus removal from domestic wastewater. Phosphate removal efficiency decreased with increase in initial phosphate concentration hence more time was necessary to obtain higher removal efficiencies for higher initial phosphate concentration. Required EC time to obtain >99% P removal for Fe electrodes at initial P concentrations of Ci = 5, 10, 25 and 52 mg/L were deter-

P.I. Omwene, M. Kobya / Process Safety and Environmental Protection 116 (2018) 34–51

49

Table 3 An overview of literature studies on phosphorus removal from wastewaters by EC process using Al–Fe hybrid electrodes. Wastewater type

Operating conditions

Optimum values

Re (%)

Energy and electrode consumptions, and OC

Ref.

SW

BR, pHi = 4–7, j = 10–40 A/m2 , tEC = 0–100 min, Ci = 5.01–52.13 mg/L, Sanode = 62 cm2 , ω = 300 rpm, N = 2 (titanium cathode) and 2 (Al and Fe anode)

pHi = 4

100

Cen = 2.738 kWh/m3

Omwene et al. (2017)

96

Cel = 0.749 kg/m3 qe = 66.39 mg/g Me/P = 3.54 (Fe/P = 2.12, Al/P = 1.42) Cen = 0.75 kWh/m3

Kuokkanen et al. (2015)

j = 20 A/m2 tEC = 80 min Ci = 52.13 mg/L SW

SW

SW

BR, pHi = 5–9, j = 25–150 A/m2 , tEC = 0–30 min, Ci = 30 mg/L, CNaCl = 0.1–1 g/L, d = 0.7–1.5 cm, ω = 250 rpm, N = 2

CR, j = 40–160 A/m2 , pHi = 4–10, Sanode = 1920 cm2 , tEC = 10–60 min, N = 4, Ci = 10–160 mg/L, Q = 0.04–0.10 L/min

CR, d = 0.5–2 cm, ω = 150 rpm, Q = 0.05 L/min, j = 80 A/m2 , Ci = 60 mg/L, pHi = 7, tEC = 50 min

pHi = 5

j = 100 A/m2 tEC = 15 min CNaCl = 1 g/L d = 0.7 cm pHi = 8

92

Cen = 0.191 kWh/m3

Chen et al. (2014)

j = 80 A/m2 Ci = 60 mg/L d = 1 cm Q = 0.05 L/min d = 0.5 cm

71 (Al–Fe)

Cen = 0.17 kWh/m3 (Al–Fe)

Gao et al. (2012)

88 (Al) 51 (Fe)

Cen = 0.12 kWh/m3 (Al) Cen = 0.21 kWh/m3 (Fe)

mined as 20, 60, 80 and 100 min, respectively. On the other hand, EC time to achieve >99% P removal for Al electrodes at initial P concentrations of 5, 10, 25 and 52 mg/L was obtained as 20, 30, 30 and 50 min, respectively. Al/P mole ratio for initial phosphorous concentration of 5, 10, 25 and 52 mg/L were obtained as 4.30, 3.94, 1.36 and 1.68, respectively, while Fe/P mole ratios at the same initial phosphorous concentration were 7.05, 10.12, 3.90 and 3.29, respectively. At lower initial pHi values of 4–5, the phosphate ion was removed by precipitation of Al or Fe-hydroxo-phosphate complexes, whereas the adsorption by electrostatic attraction and ligand exchange on Al or Fe hydroxides was predominant when both the electrochemical EC reactions and initially pH values increased. Generally, the final pHf values increased with increase in EC time; at conditions of pHi = 4, Ci = 52 mg/L and j = 20 A/m2 , the final pHf for achievement of <0.01 mg/L effluent PO4 -P (>99.99% removal) was measurement to be 8.52 at 50 min for Al electrode and 10.62 at 100 min for Fe electrode. The optimum conditions to obtain <0.01 mg/L effluent P concentration for Fe electrode were determined to be initial pHi 4–5, current density of 20 A/m2 and EC time of 100 min at Ci = 52 mg/L. At these optimums, metal-to-P mole ratio, operating cost, energy and electrode consumptions were 3.29 mol/mol, 1.343 $/m3 , 4.179 kWh/m3 and 0.884 kg/m3 , respectively; meanwhile optimum conditions for Al electrode were determined as EC time of 50 min, initial pHi of 4, current density of 20 A/m2 at Ci = 52 mg PO4 -P/L. At these optimums; the Al-to-P mole ratio of 1.684 mol/mol, operating cost of 1.032 $/m3 , energy consumption of 1.143 kWh/m3 and electrode consumption of 0.218 kg/m3 were obtained. Amount of milligram removed phosphorus per gram dissolved metal electrode at the above conditions was calculated as 239 mg P/g Al (actual Al dosage of 0.0764 g and current efficiency of 220.3%) for Al electrode and 59 mg P/g Fe (actual Fe dosage of 0.3092 g and current efficiency of 112.0%) for Fe electrode. In con-

OC = 0.17 D /m3

clusion, considering the overall results from both electrodes, Al electrode for phosphorus removal is found to be more efficient than Fe electrode in terms of operating cost and phosphorus removal efficiency.

Acknowledgement This research did not receive any specific grant from funding agencies in the public, commercial, or not-for-profit sectors.

References Ahmadzadeh, S., Asadipour, A., Pournamdari, M., Behnam, B., Rahimi, H.R., Dolatabadi, M., 2017. Removal of ciprofloxacin from hospital wastewaterusing electrocoagulation technique by aluminum electrode: optimization and modelling throughresponse surface methodology. Process Saf. Environ. Prot. 109, 538–547. Ansari, A.A., Gill, S.S., 2014. Eutrophication: Causes, Consequences and Control, vol. 2. Springer, Dordrecht, Heidelberg, London. APHA, 2005. Methods for the Examination of Water and Wastewater, 21st ed. American Public Health Association, American Water Works Association, Water Environment Federation, Washington, DC, USA. Attour, A., Touati, M., Tlili, M., Ben Amor, M., Lapicque, F., Leclerc, J.P., 2014. Influence of operating parameters on phosphate removal from water by electrocoagulation using aluminum electrodes. Sep. Purif. Technol. 123, 124–129. Attour, A., Ben Grich, N., Mouldi Tlili, M., Lapicque, F., Leclerc, J.P., 2016. Intensification of phosphate removal using electrocoagulation treatment by continuous pH adjustment and optimal electrode connection mode. Desalin. Water Treat. 57, 13255–13262. Babatunde, A.O., Zhao, Y.Q., Burke, A.M., Morris, M.A., Hanrahan, J.P., 2009. Characterization of aluminium-based water treatment residual for potential phosphorus removal in engineered wetlands. Environ. Pollut. 157, 2830–2836. Babatunde, A.O., Zhao, Y.Q., Yang, Y., Kearney, P., 2008. Reuse of dewatered aluminium-coagulated water treatment residual to immobilize phosphorus: batch and column trials using a condensed phosphate. Chem. Eng. J. 136, 108–115. Barthelemy, K., Naille, S., Despas, C., Ruby, C., Mallet, M., 2012. Carbonated ferric green rust as a new material for efficient phosphate removal. J. Colloid Interface Sci. 384, 121–127.

50

P.I. Omwene, M. Kobya / Process Safety and Environmental Protection 116 (2018) 34–51

Behbahani, M., Alavi Moghaddam, M.R., Arami, M., 2011. A comparison between aluminum and iron electrodes on removal of phosphate from aqueous solutions by electrocoagulation process. Int. J. Environ. Res. 5, 403–412. Bektas, N., Akbulut, H., Inan, H., Dimoglo, A., 2004. Removal of phosphate from aqueous solutions by electrocoagulation. J. Hazard. Mater. 106, 101–105. Berkowitz, J., Anderson, M.A., Graham, R.C., 2005. Laboratory investigation of aluminum solubility and solid-phase properties following alum treatment of lake waters. Water Res., 3918–3928. Bouamra, F., Drouiche, N., Ahmed, D.S., Lounici, H., 2012. Treatment of water loaded with orthophosphate by electrocoagulation. Procedia Eng. 33, 155–162. Can, O.T., Bayramoglu, M., Kobya, M., 2003. Decolorization of reactive dye solutions by electrocoagulation using aluminium electrodes. Ind. Eng. Chem. Res. 42, 3391–3396. Caravelli, A.H., Contreras, E.M., Zaritzky, N.E., 2010. Phosphorus removal in batch systems using ferric chloride in the presence of activated sludge. J. Hazard. Mater. 177, 199–208. Chafi, M., Gourich, B., Essadki, A.H., Vial, C., Fabregat, A., 2011. Comparison of electrocoagulation using iron and aluminium electrodes with chemical coagulation for the removal of a highly soluble acid dye. Desalination 281, 285–292. Chen, S., Shi, Y., Wang, W., Li, Z., Gao, J., Bao, K., Han, R., Zhang, R., 2014. Phosphorus removal from continuous phosphate contaminated water by electrocoagulation using aluminium and iron plates alternately as electrodes. Sep. Purif. Technol. 49, 939–945. Chen, X., Chen, G., Yue, P.L., 2000. Separation of pollutants from restaurant wastewater by electrocoagulation. Sep. Purif. Technol. 19, 65–76. Dayton, E.A., Basta, N.T., 2005. A method for determining the phosphorus sorption capacity and amorphous aluminum of aluminum-based drinking water treatment residuals. J. Environ. Qual. 34, 1112–1118. De Haas, D.W., Wentzel, M.C., Ekama, G.A., 2000. The use of simultaneous chemical precipitation in modified activated sludge systems exhibiting biological enhanced phosphate removal. Part 1: literature review. Water SA 26, 439–452. EEA (European Environment Agency), Fresh Water Quality-nutrients in River, Briefing Published 18 Feb 2015. Last modified 15 Nov 2016, https://www.eea.europa. eu/soer-2015/countries-comparison/freshwater. Elazzouzi, M., Haboubi, Kh., Elyoubi, M.S., 2017. Electrocoagulation flocculation as a low-cost process for pollutants removal from urban wastewater. Chem. Eng. Res. Des. 117, 614–626. Franco, D., Lee, J., Arbelaez, S., Cohen, N., Kim, J.Y., 2017. Removal of phosphate from surface and wastewater via electrocoagulation. Ecol. Eng. 108, 589–596. Fytianos, K., Voudrias, E., Raikos, N., 1998. Modelling of phosphorus removal from aqueous and wastewater samples using ferric iron. Environ. Pollut. 101, 123–130. Gao, J., Bao, K., Jia, X., Zhou, J., Zhang, R., 2012. Optimization of phosphorus removal from phosphate-contaminated water by electrocoagulation using aluminium and iron plate electrodes. Fresen. Environ. Bull. 21, 2581–2586. Georgantas, D.A., Grigoropoulou, H.P., 2007. Orthophosphate and metaphosphate ion removal from aqueous solution using alum and aluminum hydroxide. J. Colloid Interface Sci. 315, 70–79. Gharibi, H., Mahvi, A.H., Chehrazi, M., Sheikhi, R., Hosseini, S.S., 2010. Phosphorous removal from wastewater effluent using electro-coagulation by aluminum and iron plates. Anal. Bioanal. Electrochem. 2, 165–177. Golder, A.K., Samanta, A.N., Ray, S., 2006. Removal of phosphate from aqueous solutions using calcined metal hydroxides sludge waste generated from electrocoagulation. Sep. Purif. Technol. 52, 102–109. Huang, H., Zhang, D., Zhao, Z., Zhang, P., Gao, F., 2017. Comparison investigation on phosphate recovery from sludge anaerobic supernatant using the electrocoagulation process and chemical precipitation. J. Clean. Prod. 141, 429–438. Inan, H., Alaydın, E., 2014. Phosphate and nitrogen removal by iron produced in electrocoagulation reactor. Desalin. Water Treat. 52, 1396–1403. Ince, M., 2013. Treatment of manganese-phosphate coating wastewater by electrocoagulation. Sep. Sci. Technol. 48, 515–522. Ippolito, J.A., Barbarick, K.A., Heil, D.M., Chandler, J.P., Redente, E.F., 2003. Phosphorus retention mechanisms of a water treatment residual. J. Environ. Qual. 32, 1857–1864. Irdemez, S., Demircioglu, N., Yildiz, Y.S., 2006a. The effects of pH on phosphate removal from wastewater by electrocoagulation with iron plate electrodes. J. Hazard. Mater. 137, 1231–1235. Irdemez, S., Demircioglu, N., Yildiz, Y.S., Bingul, Z., 2006b. The effects of current density and Phosphate concentration on phosphate removal from wastewater by electrocoagulation using aluminum and iron plate electrodes. Sep. Purif. Technol. 52, 218–223. Jiang, J.Q., Graham, N.J.D., 1998. Pre-polymerised inorganic coagulants and phosphorus removal by coagulation: a review. Water SA 24, 237–244. Karageorgiou, K., Paschalis, M., Anastassakis, G.N., 2007. Removal of phosphate species from solution by adsorption onto calcite used as natural adsorbent. J. Hazard. Mater. 139, 447–452. Kim, H.G., Jang, H.N., Kim, H.M., Lee, D.S., Chung, T.H., 2010. Effect of an electro phosphorous removal process on phosphorous removal and membrane permeability in a pilot-scale MBR. Desalination 250, 629–633. Kobya, M., Demirbas, E., 2015. Evaluations of operating parameters on treatment of can manufacturing wastewater by electrocoagulation. J. Water Proc. Eng. 8, 64–74. Kobya, M., Demirbas, E., Dedeli, A., Sensoy, M.T., 2010. Treatment of rinse water from zinc phosphate coating by batch and continuous electrocoagulation processes. J. Hazard. Mater. 173, 326–334.

Kuokkanen, V., Kuokkanen, T., Ramo, J., Lassi, U., Roininen, J., 2015. Removal of phosphate from wastewaters for further utilization using electrocoagulation with hybrid electrodes-techno-economic studies. J. Water Proc. Eng. 8, 50–57. Lacasa, E., Canizares, P., Saez, C., Fernandez, F.J., Rodrigo, M.A., 2011. Electrochemical phosphates removal using iron and aluminium electrodes. Chem. Eng. J. 172, 137–143. Li, W.C., Zhao, Y.Q., 2010. Phosphorus immobilization in Al-drinking water treatment sludge (Al-DWTS) and soil under laboratory conditions. Int. J. Environ. Stud. 67, 747–762. Mahvi, A.H., Ebrahimi, S.J.A., Mesdaghinia, A., Gharibi, H., Sowlat, M.H., 2011. Performance evaluation of a continuous bipolar electrocoagulation/electrooxidationelectroflotation (ECEO-EF) reactor designed for simultaneous removal of ammonia and phosphate from wastewater effluent. J. Hazard. Mater. 192, 1267–1274. Makwana, A.R., Ahammed, M.M., 2016. Continuous electrocoagulation process for the post-treatment of anaerobically treated municipal wastewater. Process Saf. Environ. Prot. 102, 724–733. Mansouri, K., Ibrik, K., Bensalah, N., Abdel-Wahab, A., 2011. Anodic dissolution of pure aluminum during electrocoagulation process: influence of supporting electrolyte, initial pH, and current density. Ind. Eng. Chem. Res. 50, 3362–13372. Merzouk, B., Gourich, B., Sekki, A., Madani, K., Vial, C., Barkaoui, M., 2009. Studies on the decolorization of textile dye wastewater by continuous electrocoagulation process. Chem. Eng. J. 149, 207–214. Mollah, M.Y.A., Schennach, R., Parga, J.R., Cocke, D.L., 2001. Electrocoagulation (EC)science and applications. J. Hazard. Mater. 84, 29–41. Mores, R., Treichel, H., Zakrzevski, C.A., Kunz, A., Steffens, J., Dallago, R.M., 2016. Remove of phosphorous and turbidity of swine wastewater using electrocoagulation under continuous flow. Sep. Purif. Technol. 171, 112–117. Mouedhen, G., Feki, M., Wery, M., Ayedi, H.F., 2008. Behavior of aluminum electrodes in electrocoagulation processes. J. Hazard. Mater. 150, 124–135. Moussavi, G., Majidi, F., Farzakia, M., 2011. The influence of operational parameters on elimination of cyanide from wastewater using the electrocoagulation process. Desalination 280, 127–133. Narasiah, K.S., Morasse, C., Lemay, J., 1991. Nutrient removal from aerated lagoons using alum and ferric chloride-a case study. Water Sci. Technol. 23, 1563–1572. Nassef, E., 2012. Removal of phosphorous compounds by electrochemical technique. Eng. Sci. Technol. Int. J. 2, 403–407. Ngo, H.H., Guo, W., 2009. Membrane fouling control and enhanced phosphorus removal in an aerated submerged membrane bioreactor using modified green bioflocculant. Bioresour. Technol. 100, 4289–4291. Nguyen, D.D., Ngo, H.H., Guo, W., Nguyen, T.T., Chang, S.W., Jang, A., Yoon, Y.S., 2016. Can electrocoagulation process be an appropriate technology for phosphorus removal from municipal wastewater. Sci. Total Environ. 563–564, 549–556. O’Rourke, S.M., Foy, R.H., Watson, C.J., Gordon, A., Doody, D., 2012. Assessment of co-blending water treatment residual with dairy manure to reduce phosphorus concentrations in run-off in Northern Ireland. Soil Use Manag. 28, 157–166. Omwene, P.I., Kobya, M., Can, O.T., 2017. Phosphate removal from domestic wastewater in electrocoagulation reactor using hybrid Al-Fe plate anodes. In: 2nd International Conference on Civil and Environmental Engineering (ICOCEECappadocia 2017), Nevsehir, Turkey, May 8–10. Palahouane, B., Drouiche, N., Aoudj, S., Bensadok, K., 2015. Cost-effective electrocoagulation process for the remediation of fluoride from pretreated photovoltaic wastewater. J. Ind. Eng. Chem. 22, 127–131. Parfitt, L., Atkinson, J., Smart, R.St.C., 1975. The mechanism of phosphate fixation by iron oxides. Soil Sci. Soc. Am. J. 39, 837–841. Parikh, S.J., Mukome, F.N.D., Zhang, X., 2014. ATR-FTIR spectroscopic evidence for biomolecular phosphorus and carboxyl groups facilitating bacterial adhesion to iron oxides. Colloids Surf. B: Biointerfaces 119, 38–46. Park, T., Ampunan, V., Lee, S., Chung, E., 2016. Chemical behavior of different species of phosphorus in coagulation. Chemosphere 144, 2264–2269. Peleka, E.N., Deliyanni, E.A., 2009. Adsorptive removal of phosphates from aqueous solutions. Desalination 245, 357–371. Picard, T., Feuillade, G.C., Mazet, M., Vandensteendam, C., 2000. Cathodic dissolution in the electrocoagulation process using aluminum electrodes. J. Environ. Monit. 2, 77–80. Pulkka, S., Martikainen, M., Bhatnagar, A., Sillanpaa, M., 2014. Electrochemical methods for the removal of anionic contaminants from water—a review. Sep. Purif. Technol. 132, 252–271. Pykhteev, O.Y., Ofimov, A.A., Moskvin, L.N., 1999. Hydrolysis of iron(III) aqua complexes. Russ. J. Appl. Chem. 72, 9–20. Ramasahayam, S.K., Guzman, L., Gunawan, G., Viswanathan, T., 2014. A comprehensive review of phosphorus removal technologies and processes. J. Macromol. Sci. A 51, 538–545. Razali, M., Zhao, Y.Q., Bruen, M., 2007. Effectiveness of a drinking-water treatment sludge in removing different phosphorus species from aqueous solution. Sep. Purif. Technol. 55, 300–306. Rebhun, M., Lurie, M., 1993. Control of organic matter by coagulation and floc separation. Water Sci. Technol. 27, 1–20. Shalaby, A., Nassef, E., Mubark, A., Hussein, M., 2014. Phosphate removal from wastewater by electrocoagulation using aluminium electrodes. Am. J. Environ. Eng. Sci. 1, 90–98. Song, X., Pan, Y., Wu, Q., Cheng, Z., Ma, W., 2011. Phosphate removal from aqueous solutions by adsorption using ferric sludge. Desalination 280, 384–390. Stafford, B., Dotro, G., Vale, P., Jefferson, B., Jarvis, P., 2014. Removal of phosphorus from trickling filter effluent by electrocoagulation. Environ. Technol. 35, 3139–3146.

P.I. Omwene, M. Kobya / Process Safety and Environmental Protection 116 (2018) 34–51 Szabo, A., Takacs, I., Murthy, S., Daigger, G.T., Licsko, I., Smith, S., 2008. Significance of design and operational variables in chemical phosphorus removal. Water Environ. Res. 80, 407–416. Szynkarczuk, J., Kan, J., Hassan, T.A.T., Donini, J.C., 1994. Electrochemical coagulation of clay suspensions. Clays Clay Miner. 42, 667–673. Tanada, S., Kabayama, M., Kawasaki, N., Sakiyama, T., Nakamura, T., Araki, M., Tamura, T., 2003. Removal of phosphate by aluminum oxide hydroxide. J. Colloid Interface Sci. 257, 135–140. Tchobanoglous, G., Burton, F.L., Stensel, H.D., 2004. Wastewater Engineering: Treatment and Reuse, 4th ed. Metcalf and Eddy Inc., McGraw Hill, New York. Thistleton, J., Berry, T.A., Pearce, P., Parsons, S.A., 2002. Mechanisms of chemical phosphorus removal II—iron (III) salts. Chem. Eng. Res. Des. (Trans IChemE) 80, 265–269. Vasudevan, S., Lakshmi, J., Jayaraj, J., Sozhan, G., 2009. Remediation of phosphate contaminated water by electrocoagulation with aluminium, aluminium alloy and mild steel anodes. J. Hazard. Mater. 164, 1480–1486. Vasudevan, S., Sozhan, G., Ravichandran, S., Jayaraj, J., Lakshmi, J., Sheela, S.M., 2008. Studies on the removal of phosphate from drinking water by electrocoagulation process. Ind. Eng. Chem. Res. 47, 2018–2023. Vik, E.A., Carlson, D.A., Eikum, A.S., Gjessing, E.T., 1984. Electrocoagulation of potable water. Water Res. 18, 1355–1360. Wang, S., Li, F., Liu, Z., Liu, G., 2016. Phosphorus removal from wastewater in Johkasou sewage treatment tank by electro-coagulation. Int. J. Environ. Sci. Dev. 7, 797–800. Yan, Z., He, Q., Long, T., Zhang, H., 2010. Mechanisms of phosphorus removal by aerated filtration with porous iron-rich media. Chin. J. Environ. Eng. 4, 1305–1308. Yang, Y., Zhao, Y.Q., Babatunde, A.O., Wang, L., Ren, Y.X., Han, Y., 2006. Characteristics and mechanisms of phosphate adsorption on dewatered alum sludge. Sep. Purif. Technol. 51 (2), 193–200.

51

Yang, K., Li, Z., Zhang, H., Qian, J., Chen, G., 2010. Municipal wastewater phosphorus removal by coagulation. Environ.Technol. 31, 601–609. Yang, Y., Zhao, Y.Q., Kearney, P., 2008. Influence of ageing on the structure and phosphate adsorption capacity of dewatered alum sludge. Chem. Eng. J. 145, 276–284. Yeoman, S., Stephanson, T., Lester, J.N., Perry, R., 1988. The removal of phosphorus during wastewater treatment: a review. Environ. Pollut. 49, 183–233. Yu, M.J., Ku, Y.H., Kim, Y.S., Myung, G.N., 2006. Electrocoagulation combined with the use of an intermittently aerating bioreactor to enhance phosphorus removal. Environ. Technol. 27, 483–491. Zhang, S., Zhang, J., Wang, W., Li, F., Cheng, X., 2013. Removal of phosphate from landscape water using an electrocoagulation process powered directly by photovoltaic solar modules. Ecol. Eng. Solar Energy Mater. Solar Cells 117, 73–80. Zhang, T., Ding, L., Ren, H., Guo, Z., Tan, J., 2010. Thermodynamic modelling of ferric phosphate precipitation for phosphorus removal and recovery from wastewater. J. Hazard. Mater. 176, 444–450. Zhang, X., Lin, H., Hu, B., 2016. Phosphorus removal and recovery from dairy manure by electrocoagulation. RSC Adv. 6, 57960–57968. Zheng, X.Y., Kong, H.N., Wu, D.Y., Wang, C., Li, Y., Ye, H.R., 2009. Phosphate removal from source separated urine by electrocoagulation using iron plate electrodes. Water Sci. Technol. 60, 2929–2938. Zongo, I., Leclerc, J.P., Maiga, H.A., Wethe, J., Lapicque, F., 2009. Removal of hexavalent chromium from industrial wastewater by electrocoagulation: a comprehensive comparison of aluminium and iron electrodes. Sep. Purif. Technol. 66, 159–166.