Separation and Purification Technology 36 (2004) 33–39

New bipolar electrocoagulation–electroflotation process

for the treatment of laundry wastewater
Jiantuan Ge, Jiuhui Qu∗ , Pengju Lei, Huijuan Liu
State Key Laboratory for Environmental Aquatic Chemistry, Research Center for Eco-Environmental Sciences,
Chinese Academy of Sciences, Beijing 100085, China

Received 11 January 2003; received in revised form 9 April 2003; accepted 22 April 2003

Abstract

A new bipolar electrocoagulation and electroflotation process was developed to treat laundry wastewater. In this new process,
electrocoagulation and electroflotation were carried out simultaneously in a single reactor. The operating parameters such as
initial pH, hydraulic residence time (HRT) and current density were investigated. The unique design of the reactor made it
possible for the effective removal of turbidity, COD, phosphate and surfactant (MBAS) in a wide pH range (5–9) at a short HRT
(5–10 min). The pilot scale tests (1.5 m3 /h) were carried out successfully in three different places suggested that the bipolar
electrocoagulation–electroflotation process was feasible for the treatment of laundry wastewater.
© 2003 Elsevier B.V. All rights reserved.

Keywords: Laundry wastewater; Electrocoagulation; Electroflotation; Electrochemical reactor; Field test

1. Introduction lation, electroflotation and electrochemical oxidation,

have been developed for the treatment of wastewater.
Traditional methods for dealing with laundry Amongst them, electrocoagulation has been used to
wastewater consist of various combinations of bio- treat urban wastewater [4], oil wastes [5], dyes [6],
logical, physical and chemical methods [1–3]. Due to suspended particles [7], chemical and mechanical pol-
the large variability of the composition of this kind ishing waste [8], organic contaminants from landfill
of wastewater, most of these traditional methods are leachates [9], defluoridation [10] and heavy metal-
becoming inadequate, especially for the simultane- containing solutions [11]. It was found that anodized
ous removal of high content of suspended solids, aluminum was more effective than the aluminum ion
surfactants and phosphate. Hence a search for more introduced in the form of aluminum sulfate solution
cost-effective treatment methods is in need. [12]. However, preliminary experiment in our labo-
A variety of very promising techniques based on ratory showed that the hydrogen gas produced at the
electrochemical technology, including electrocoagu- aluminum cathode in an electrocoagulation cell was
not so fine that could float only about 60% of the total
∗ Corresponding author. Tel.: +86-10-6284-9137; solids.
fax: +86-10-6284-9160. In this study, a combinative bipolar electrocoagul-
E-mail address: [email protected] (J. Qu). ation–electroflotation process was developed to treat

1383-5866/$ – see front matter © 2003 Elsevier B.V. All rights reserved.
doi:10.1016/S1383-5866(03)00150-3
34 J. Ge et al. / Separation and Purification Technology 36 (2004) 33–39

laundry wastewater. The primary objectives of using cheapness. Excellent materials for oxygen evolution
the new reactor were to neutralize pollutant charge, have also been reported [14–16].
generate ultra-fine bubbles and separate the coagulated Since some amount of NaCl always exists in the
flocs from water. Finally, pilot scale field tests (1.5 laundry wastewater, other reactions may be also en-
m3 /h) were carried out in three different places. countered in anodic compartment:
2Cl− → Cl2 + 2e (3)

2. Description of bipolar Cl2 + H2 O → HClO + H+ + Cl− (4)

electrocoagulation–electroflotation
HClO → H+ + ClO− (5)
In the bipolar electrocoagulation–electroflotation While the main cathodic reaction on the Ti cathode
unit (Fig. 1), three aluminum plates placed between and negative side of aluminum is as follows:
two titanium electrodes having opposite charges will
undergo anodic reactions on the positive side while 2H2 O + 2e → H2 + 2OH− (6)
on the negative side cathodic reactions take place. Other reactions could be observed at the cathode
The main anodic reactions are as follows: surface and bring about the precipitation of the car-
bonate salt.
Ti anode : 2H2 O → O2 + 4H+ + 4e (1)
HCO− − 2−
3 + OH → CO3 + H2 O (7)
Al positive side : Al → Al 3+
+ 3e (2)
Ca2+ + CO2−
3 → CaCO3 (8)
The hydrolysis and polymerization of Al3+
un-
der appropriate pH conditions subsequently give Mg2+ + CO2−
3 → MgCO3 (9)
rise to the formation of such species as Al(OH)2+ ,
These will increase the ohmic resistance of the elec-
Al2 (OH)2 4+ , Al(OH)3 and charged hydroxo cationic
trochemical cell. However, alternating current direc-
complexes [13], which can effectively remove pollu-
tion (ACD) at due time can be applied to alleviate the
tants by adsorption to produce charge neutralization,
precipitation of carbonate salt on the electrode surface.
and by enmeshment in a precipitate. In this study, tita-
nium metal is chosen as anode material mainly for its
3. Experimental

The experimental setup is schematically shown in

Fig. 2. The system includes mainly a 2.8-dm3 electro-
chemical reactor and an 11.2-dm3 separator. The elec-
trochemical reactor is composed of four sets of bipolar
electrocoagulation–electroflotation unit. The Ti elec-
trodes are connected to the D.C. power only. There are
eight pieces of Ti plates and 21 pieces of Al plates.
The dimensions of the Ti electrode and Al electrode
are 100 × 60 × 2 mm and 100 × 60 × 3 mm, respec-
tively. The effective area of each electrode is 50 cm2
and the spacing between electrodes is 8 mm. The sep-
arator consists of an outer cylinder with a diameter of
100 mm and an inner tube with a diameter of 20 mm.
The wastewater collected from Huaqing Laundry
House in Tsinghua University was used for all the
Fig. 1. Schematic representation of the bipolar electrocoagula- experiments except for pilot-scale field tests. The
tion–electroflotation. wastewater entered the electrochemical reactor from
 J. Ge et al. / Separation and Purification Technology 36 (2004) 33–39 35

Fig. 2. Flowchart of the experimental system: (1) DC power supply, (2) pump, (3) electrochemical cell (solid bar: Ti electrode and space
bar: Al electrode), (4) polymer electrolyte injection, (5) separator and (6) effluent.

the bottom, moved upwards in a zig-zag pattern from current direction was automatically alternated every
the spacing between electrodes. It flowed out the re- 15 min to alleviate passivation of the electrodes.
actor from the top. After adding desired amount of
polyelectrolytes (PAM), the coagulated wastewater 4.1.1. pH change during the bipolar
was introduced from the bottom of the inner tube electrocoagulation–electroflotation process
of the separator. The treated wastewater exited from It has been established that the influent pH is of
the lower part of the separator. The floated sludge vital importance in the performance of many elec-
could either be removed from the top of the separa- trochemical process and chemical coagulation oper-
tor (laboratory tests) or recycled back to the influent ations. To examine the effect of pH on the bipolar
wastewater tank (pilot-scale field tests). electrocoagulation–electroflotation process, the laun-
Chemical oxygen demand (COD), surfactants dry wastewater was adjusted to the desired pH with
(MBAS) and phosphate (P-PO4 3− ) were measured diluted aqueous sodium hydroxide or sulfuric acid.
according to the Standard Methods [17]. Turbidity The results illustrated in Fig. 3 demonstrated the
was recorded on a 2100N IS Turbidimeter (Hach). removal efficiencies of MBAS, turbidity, COD and
The pH of the wastewater was adjusted with H2 SO4 phosphate as a function of the influent pH. With 1
and NaOH and the measurement was performed on a mg/l PAM alone, the removal efficiencies were negli-
720A pH Meter (Orion). gible. It was found that the optimum pH for turbidity,
MBAS, COD and phosphate was around 5–9. How-
ever, the turbidity removal dropped dramatically at
4. Results and discussion pH < 4 or >9. A slight drop in COD, phosphate and
MBAS removal was observed at pH>10. These re-
4.1. Laboratory test of laundry wastewater sults can be explained by the species distribution of
aluminum ions. In the pH range 5–9 the hydrolysis
First we carried out laboratory experiments of and polymerization of Al3+ give rise to the formation
laundry wastewater to investigate the variables such of such species as Al(OH)2+ , Al2 (OH)2 4+ , Al(OH)3
as hydraulic residence time (HRT), initial pH and and charged hydroxo cationic complexes such as
current density that may influence the processes of Al13 (OH)32 7+ , which are efficient for coagulation
electrocoagulation–electroflotation. The electrical [18,19]. When pH was greater than 10, the main
36 J. Ge et al. / Separation and Purification Technology 36 (2004) 33–39

Fig. 3. Removal efficiencies of COD, MBAS, phosphate and tur-

bidity as a function of initial pH (HRT, 10 min; I, 1.2A; PAM, 1
mg/l). Fig. 4. pH change during bipolar electrocoagulation–electro-
flotation process (HRT, 10 min; I, 1.0 A).

hydrolysis product is Al(OH)4 − , which does not fa- In acidic medium, CO2 is over saturated and re-
vor the formation of anodized aluminum species and leased from the wastewater due to H2 and O2 bub-
the adsorption of dispersed solids. At low pH where ble ‘purge’, which causes a pH increase. In addition,
one has only aluminum ion, the adsorption effect is the chemical dissolution of Al (Reaction 10) will con-
insignificant [20]. sume H+ and gives rise to the pH increase. Moreover,
The removal of surfactant in this process may be reaction (11) will shift towards the left side and again
assumed that the surfactant is first adsorbed on the results in an increase in pH. At high pH condition,
particulate surface, making the surface hydrophobic, reactions (7)–(9) will proceed readily and reaction (12)
and thus forcing the particulate to the surface of the may take place, both lead to a pH decrease.
rising bubble. Finally the surfactant-particulate-bubble The pH neutralization effect made this process ef-
composite is floated to the surface. The reduction of fective in a much wider pH range, which made it su-
COD may be mainly accomplished by electrocoag- perior to traditional chemical coagulation. The latter
ulation, electroflotation, direct anodic oxidation and is highly sensitive to pH change and effective coagu-
indirect oxidation by chloride ions. Electrochemically lation is achieved at pH 6–7 [22].
produced aluminum has been found to be more effi-
cient in removing phosphate than the same amount 4.1.2. Effect of HRT in the electrochemical reactor
of aluminum contained in an aluminum sulfate so- The influence of HRT on the process was studied
lution [21]. In brief, the bipolar electrocoagulation– on the COD removal at the desired HRT where other
electroflotation process includes electroflotation of parameters were kept unchanged. Here HRT should
suspended pollutants, electrocoagulation of colloidal be defined as the ratio of the total volume of the elec-
pollutants, electrodestruction of dissolved organics trochemical reactor to wastewater flow rate. Fig. 5
and electrodeposition of dissolved inorganics. showed that the removal of COD increased with in-
It was found that after electrocoagulation–electro- creasing HRT. But when HRT was greater than 10
flotation process, the pH of the aqueous medium min, the removal of COD reached a plateau. This is
could be neutralized to some extent (Fig. 4). This phe- a very attractive feature for densely populated areas
nomenon can be explained according to reactions (1)– where the compactness of the treatment facility is
(9) and the following chemical reactions: particularly concerned. The mechanism of the elec-
trochemical process in aqueous system is quite com-
Al + 3H+ → Al3+ (10)
plex [23]. It is generally believed that there are three
Al3+ + 3H2 O → Al(OH)3 + 3H+ (11) possible mechanisms involved in the process: elec-
trocoagulation, electroflotation and electrooxidation.
Al(OH)3 + OH− → Al(OH)−
4 (12) However, it can be suggested that, during the bipolar
 J. Ge et al. / Separation and Purification Technology 36 (2004) 33–39 37

Fig. 5. COD removal efficiencies as a function of HRT (I, 1.0 A).

Fig. 6. Removal of turbidity, COD and MBAS as a function of
current density (HRT, 5 min).
electrocoagulation–electroflotation process, bipolar
aluminum electrodes are mainly responsible for elec- that the removal of turbidity, COD and MBAS was
trocoagulation. A relatively large amount of fine H2 increased with increasing current. The influence of
and O2 gas evolution from Ti electrodes satisfies the current density should be correlated with HRT in the
needs for electroflotation to float the flocs formed by continuous operating system. According to Faraday’s
electrocoagulation and other pollutants to the surface. law, the amount of aluminum dissolved electrochem-
The conditions present in the electrocoagulation– ically is proportional to charge loadings. When 1 F
electroflotation cell favored formation of the micro- (26.8 Ah) passes through the electric circuit, 0.0224
bubble-pollutant aggregate composite after the Nm3 hydrogen gas will be evolved, which is much
collision and union of the two occurred in the sub- greater than the volume of gas released in traditional
sequent separation unit. On the whole, the configu- DAF. Consequently, increasing current density will
ration of the reactor can be considered as plug flow. give rise to an increase in charge loading and leading
But the plug flow reactor (PFR) was comprised of to increased removal of pollutants [25]. Furthermore,
many micro continuous stirred tank reactors (CSTR) better collection efficiencies could be reached during
present in two electrodes. Typical bubble sizes in electroflotation by obtaining smaller bubbles with
electroflotation always fall in the range of 20–70 ␮m increasing current density [26].
[24], smaller than those by electrocoagulation or con-
ventional flotation, providing both sufficient sites for 4.2. Pilot-scale field tests (1.5 m3 /h)
gas–liquid–solid interface and mixing effect to ac-
complish aggregation of tiny destabilized particulates. Based on the laboratory test of laundry wastew-
The PFR made it possible for the high removal of ater, 1.5 m3 /h electrocoagulation–electroflotation
pollutants at a short HRT. Basically, the capture of the equipment was manufactured and mounted on a mo-
micro-bubble by the pollutant was rather irreversible bile truck. The equipment mainly consisted of an
under the operating conditions of an electrocoagula- electrocoagulation–electroflotation tank and a sep-
tion cell. This fundamental mechanism distinguished arator. The corresponding dimensions of are D450
the electrocoagulation–electroflotation process from × 1600 mm and D400 × 1600 mm, respectively.
the combination of electrocoagulation and dissolved The electrode gap was 20 mm. The floated sludge
aeration flotation (DAF). flowed out from the separator and was discharged
into the feed tank, which helped pre-coagulation of
4.1.3. Effect of electrical current the wastewater. The flocs could be skimmed from
Fig. 6 showed the removal of turbidity, COD and the feed tank when an operation cycle (usually 4 h)
MBAS as a function of electric current. It was found finished. The D.C. input was 30–35 V and the corre-
38 J. Ge et al. / Separation and Purification Technology 36 (2004) 33–39

Table 1
Pilot scale field test results
Harbin, Heilongjiang Wuxi, Jiangsu Shunyi, Beijing
(24 January, 2002) (13 November, 2001) (16 March, 2002)
Influent Effluent Influent Effluent Influent Effluent

Conductivity (␮S) 1904–786 956–675 830–712 719–655 1159–897 735–589

pH 9.56–7.83 8.15–7.24 10.68–9.77 8.73–7.92 9.35–8.45 8.09–7.77
Turbidity (NTU) 583–471 12.3–8.26 78.0–32.2 11.2–10.5 810–225 4.25–1.41
COD (mg/l) 1090–785 45.6–32.4 65.9–31.8 11.2–5.3 890–628 71.2–59.3
MBAS (mg/l) 72.3–64.5 11.2–9.6 14.6–8.3 2.9–1.7 72.5–57.6 8.9–5.8
Number of samples: 6.

sponding current was about 30A. ACD was employed References

to alleviate the passivation of the electrodes and the
time interval was 15 min. [1] T. Matsuo, T. Nishi, Carbon 38 (2000) 709.
The field test results are shown in Table 1. As for [2] T. Poiger, T.M. Field, H. Siegrist, W. Giger, J.A. Field, Water
the laundry wastewater of Shunyi (Beijing), when Res. 32 (1998) 1939.
[3] T. Matsuo, T.J. Nishi, Nucl. Sci. Technol. 38 (2001)
poly aluminum chloride (10% Al2 O3 ) was used, the
127.
dosage was 1200 mg/l, equal to 63.5 mg/l Al3+ , [4] M.-F. Pouet, A. Grasmick, Water Sci. Technol. 31 (3–4)
to achieve the same residual turbidity. The settled (1995) 275.
sludge amounted to nearly 40% of the total vol- [5] U.B. Ogutveren, S. Koparal, J. Environ. Sci. Health Part A
ume. For the electrochemical process (32V/30A, 32 (9–10) (1997) 2507.
1.5 m3 /h), the Al3+ consumption was 6.72 mg/l. [6] U.B. Oeguetveren, N. Goenen, S. Koparal, J. Environ. Sci.
Table 1 also indicated that the pH of the treated Health Part A 27 (5) (1992) 1237.
[7] M.J. Matteson, R.L. Dobson, R.W. Glenn Jr, N.S. Kukunoor,
wastewater was neutralized and the electrical con-
W.H. Waits, E. Clayfield, J. Colloids Surf. Physicochem. Eng.
ductivity reduced. The latter distinguished the bipolar Aspects 104 (1) (1995) 101.
electrocoagulation–electroflotation process from other [8] B.M. Belongia, P.D. Haworth, J.C. Baygents, S. Raghavan,
coagulation/flotation processes. J. Electrochem. Soc. 146 (11) (1999) 4124.
[9] D. Mills, J. Am. Water Works Assoc. 92 (6) (2000)
34.
5. Conclusion [10] N. Mameri, H. Lounici, D. Belhocine, H. Grib, D.L. Piron,
Y. Yahiat, Sep. Purif. Technol. 24 (2001) 113.
A new bipolar electrocoagulation–electroflotation [11] H. Melcer, D.T. Mrrill, M.B. Gerhardt, C. Van Maltby, Tappi
Proc. Environ. Conf. Exhib. 1 (1998) 291.
process was developed for the treatment of laundry
[12] H.C. Lee, T.-C. Chou, J. Chin. Inst. Chem. Eng. 25 (4) (1994)
wastewater. The process successfully removed turbid- 239.
ity, COD, phosphate and MBAS simultaneously in [13] A.M. Yousuf, S. Robert, R.P. Jose, L.C. David, J. Hazard
the pH range of 5–9. The laboratory results showed Mater B84 (2001) 29.
that the removal of COD was greater than 70%. [14] G. Chen, X. Chen, P.L. Yue, J. Phys. Chem. B106 (2002)
And the removal efficiencies of MBAS, turbidity and 4364.
P-phosphate could be reached above 90%. The 1.5 [15] X. Chen, G. Chen, P.L. Yue, Environ. Sci. Technol. 36 (2002)
778.
m3 /h pilot scale tests proved that the process was
[16] X. Chen, G. Chen, P.L. Yue, J. Phys. Chem. B105 (2001)
competitive for the treatment of laundry wastewater. 4623.
[17] APHA, Standard Methods for Examination of Water and
Wastewater, 17th ed., Washington, DC, 1992.
Acknowledgements [18] W. Stumm, J.J. Morgan, J. Am. Water Works Assoc. 54
(1962) 971.
The research was financially supported by National [19] J.Y. Bottero, M. Axelos, D. Tchoubar, J. Colloid Interf. Sci.
Outstanding Youth Fund (No. 50225804). 117 (1987) 47.
 J. Ge et al. / Separation and Purification Technology 36 (2004) 33–39 39

[20] X. Shen, P. Gao, X. Chen, G. Chen, Chem. Eng. Sci. 58 [24] O.J. Murphy, S. Srinivasan, B.E. Conway, Electrochemistry
(2003) 995. in Transition: From the 20th to the 21st Century, Plenum,
[21] E. Dobolyi, Water Res. 12 (1978) 1113. New York, 1992.
[22] R.L. Sanks, Water Treatment Plant Design, Butterworth– [25] G. Chen, X. Chen, P.L. Yue, J. Environ. Eng. 126 (2000)
Heinemann, Stoneham, MA, 1978. 858.
[23] D. Pletcher, F.C. Walsh, Industrial Electrochemistry, [26] A.I. Zouboulis, K.A. Matis, Environ. Technol. Lett. 10 (1989)
Chapman and Hall, London, 1990. 601.