Desalination 285 (2012) 188–197

Contents lists available at SciVerse ScienceDirect

Desalination

j ourna l homepage: www.e lsev ie r .com/ locate /desa l

Electrochemical degradation of textile dyeing industry effluent in batch and flowreactor systems

C. Ahmed Basha a,⁎, J. Sendhil b, K.V. Selvakumar c, P.K.A. Muniswaran b, Chang Woo Lee a,⁎a Department of Chemical Engineering, College of Engineering, Kyung Hee University 1 Seochun, Gihung, Yongin, Gyeonggi, 446-701, Republic of Koreab Department of Chemical Engineering, S.R.M University, Kattankulathur-603 203, Tamil Nadu, Indiac Department of Chemical Engineering, Adhiyamaan College of Engineering, Hosur-635 109, Tamil Nadu, India

⁎ Corresponding authors. Tel.: +82 31 201 3825; faxE-mail addresses: basha@khu.ac.kr (C.A. Basha), cwl

0011-9164/$ – see front matter © 2011 Elsevier B.V. Alldoi:10.1016/j.desal.2011.09.054

a b s t r a c t

a r t i c l e i n f o

Article history:Received 6 April 2011Received in revised form 19 June 2011Accepted 29 September 2011Available online 24 October 2011

Keywords:Batch reactorBatch recirculation reactorRecycle reactorProcion blueCOD reductionSpecific energy consumption

Electrochemical oxidation of organic pollutants present in the dye-bath and wash water effluents from thetextile industry was carried out in batch, batch recirculation and recycle reactor configurations under differ-ent conditions of current density, treatment duration, effluent flow rate and electrode specific surface. CODreduction of 52.63% to 82.61% could be obtained when the Procion blue dye-bath effluent was treated inthe batch reactor for 8 h. In batch recirculation reactor, the reduction was 94.3% for dye-bath effluent and91.4 for wash water effluent after 6 h of operation at a current density of 5.0 Adm−2 and flow rate of100 L h−1. The specific energy consumption was found to be 4.32 kWh (kg COD)−1 for dye-bath effluentand 83.8 kWh (kg COD)−1 for wash water effluent. The results for wash water effluent under continuous op-eration of recycle reactor conditions showed 52.86% of COD removal at recycle flow rate of 100 L h−1 withdischarge flow rate of 3 L h−1. The specific energy consumption was found to be 11.9 kWh (kg COD)−1.

© 2011 Elsevier B.V. All rights reserved.

1. Introduction

Textile processing industries consume large volumes of water andchemicals generating effluents which when discharged cause unde-sirable water pollution. Even the presence of very low concentrationsof dyes in effluent interferes with the penetration of light in waterbodies and may affect the aquatic biota [1]. Some dyes and their deg-radation products have been shown to be toxic, mutagenic and carci-nogenic in nature [2]. Physicochemical processes such as adsorption,biosorption, membrane filtration, coagulation by salts, biological oxi-dation, chemical degradation etc., are some of the generally employedmethods for treating textile industry wastewater. The physicochemi-cal methods have many disadvantages in terms of efficiency, cost, re-generation or secondary pollution. Conventional activated sludge andother types of bioreactors fail to remove sufficient color and othermicro-pollutants present in textile effluents [1]. Novel, low cost effi-cient processes are needed in view of the more stringent environ-mental regulations and in this context, processes such as,electrochemical techniques [3–8], ozonization [9], photo-catalyticmethods [10–11], sonication [12], enzymatic treatments [13], andengineered wetland systems [14] are being tried by researchers forthe treatment of textile wastewater.

: +82 31 204 8114.ee@khu.ac.kr (C.W. Lee).

rights reserved.

The choice of electrode material is important in any electrochemicalstudy. Dimensionally stable anode (DSA) material has found wide appli-cation in electrochemical wastewater treatment, the most commonbeing the Ti/Ru0.3Ti0.7O2 (Cl2-evolution) anode, which has been exten-sively used in the chlor-alkali industry for long. DSA has been classifiedas “active” or “non-active”, depending on its chemical nature [14–16].When active electrodes are present, theymediate the oxidation of organ-ic species by the formation of higher oxidation state oxides of themetals.When non-active electrodes are present, there is no higher oxidationstate and the organic species are directly oxidized by an adsorbed hy-droxyl radical, generally resulting in complete combustion of the organicmolecule (e.g., SnO2, PbO2, boron doped diamond, and Ti/Pt).

The biodegradability index (BI) is defined as the ratio of BOD to CODand is a useful indicator of the total oxidation that occurs during elec-trolysis [16–19].Morais et al. [17] reported that effluentswith BI smallerthan 0.3 are not suitable for biological degradation. According to Cha-marro et al. [18] the effluent must have a BI of at least 0.4 for completebiodegradation. It is reported that awastewith BI of >0.4 can be treatedby a biological system while a BI of b0.25 indicates a toxic waste[16–19].

When an active electrode is used for mineralization, BOD initiallymarginally increases and COD decreases thus enhancing the BI. Whenan inactive electrode is employed, the organic species is directly oxi-dized by an adsorbed hydroxyl radical, resulting in complete combus-tion of the organic molecule thereby decreasing the BOD, COD and BIor marginally enhancing the BI [20–22]. Sakalis et al. [23] studied theelectrochemical degradation of azo dyes in batch and continuous flow

Fig. 1. A schematic view of batch experimental set-up (1. Regulated power supply, 2.Reference electrode, 3. Noble oxide coated catalytic anode, 4. Cathode 5.Magnetic stirrer.).

189C.A. Basha et al. / Desalination 285 (2012) 188–197

experiments using Ti/Pt anode for both real and synthetic wastewatersamples and reported an increase in BI value from 0.23 to 0.28 due topartial combustion of the organic molecule. When commercial pesti-cides were treated in a laboratory scale electrolysis using Ti/Pt asanode and stainless steel 304 as cathode the BI ratio was found to in-crease considerably from 0.125 to 0.36 [24] indicating that the electro-chemical oxidation can be a possible pretreatment method forpesticide detoxification.

Electrochemical oxidation of refractory organic contaminants pre-sent in effluents by noble oxide coated anode tends to increase BOD(by forming group –OH, –CHO, –CO, and –COOH) decrease COD andmarginally increase BI. The dye house of a textile processing industryreleases two types of wastewater, viz., dye-bath wastewater andwash wastewater in volume ratios of 15% and 85% respectively. Thedye-bath wastewater causes more pollution and its treatment ismore difficult due to its intense color, high COD, lowbiodegradability and high-salt content.

In the present study, the dye-bath and wash water effluents of atextile industry were subjected to electrochemical oxidation inbatch reactor, batch recirculation flow reactor and continuousrecycle reactor with the objective of reducing the color and theCOD. Experiments were conducted under different conditions ofcurrent density, flow rate, specific electrode surface and electrolysistime and the performance was analyzed in terms of specific powerconsumption. The dye-bath and wash water effluents were treatedseparately instead of treating them after mixing them as generallypracticed. Ultimately a sort of pinch technology can be devised, to in-tegrate electrochemical technique (the knowledge of specific energy/charge required), with the conventional biological treatment methodfor complete mineralization of the effluent.

Table 1Characteristics of selected dye bath and wash water.

Characteristics Dye-bath effluent Wash water

Color Purple blue Dark olive greenInitial COD (mgL−1) 5800 560Initial BOD (mg L−1) 181 30BI (=BOD/COD) 0.03 0.05Chloride content (mgL−1) 38995 1500Total dissolved solids (mgL−1) 52,000 3200Suspended solids (mgL−1) 2000 800pH 9 8

2. Materials and methods

All the reagents, used in this study, were of analytical grade. Pro-cion Blue hydrolyzed dye-bath wastewater and wash water sampleswere collected from a dyeing industry located in Tiruppur, Tamilnadu,India. The effluent samples were analyzed according to AmericanPublic Health Association (APHA) procedures [25] and their charac-teristics are given in Table 1. Experiments were carried out using aRuOx–TiOx coated [26] titanium substrate insoluble anode (TSIA)and a stainless steel cathode.

2.1. Batch reactor

Fig. 1 gives the schematic diagram of the batch reactor. It consistsof a cylindrical glass container closed with a lid in which electrodes ofsize 7.7 cm×6.5 cm are fitted with constant gap between them. Thelid is designed so as to facilitate sample collection and to enable theconnection of reference electrode through the salt bridge. The electri-cal energy for the system was drawn from a regulated power supplyand a magnetic stirrer was employed for stirring the effluent. Exper-iments were conducted at room temperature under galvanostaticconditions with a mixture of reactive hydrolyzed dyes effluent of ini-tial COD 5800 mg L−1 and BOD 181 mg L−1 under various operatingconditions of reactor holdup volumes (300 mL, 400 mL, and500 mL), and current density (1.0, 2.0, 3.0, 4.0, and 5.0 Adm−2). Theelectrode area exposed to the effluent varied with the volume of ef-fluent thereby varying the current passed to give the required currentdensity. All experiments were carried out for 8 h and the cell voltage,anode potential and cathode potential were noted down periodically.The samples were collected every hour, kept at acidic conditions andanalyzed for COD. After electrolysis, chlorides, hypochlorites andchlorates concentrations were determined and percentage reductionof COD, energy consumption and rate constant were computedexperimentally.

2.2. Batch recirculation reactor

The electrolytic flow reactor set-up, operated in batch recircula-tion mode is shown in Fig. 2a and b. It consists of an expandedmesh of titanium (coated with TiOx–RuOx) anode and stainless steelflat plate cathodes in between a PVC frame. The anode mesh is fixedin between two parallel cathode plates (7 cm×7 cm) with an inter-electrode gap of 1.25 cm. The available effective electrode area is39.2 cm2 for anodic reactions. The electrodes are connected to a 5 A,30 V DC regulated power supply, an ammeter and a voltmeter inorder to supply a constant current. The reactor with a holdup volumeof 0.588 dm3 has an inlet and outlet and is connected using siliconrubber tubes to a reservoir of 2.5 L volume, a magnetically drivenself-priming centrifugal pump and a rotameter. The set-up can be op-erated either in batch recirculation or in continuous recycle modewherein streams 10 and 12 are in closed state for the former andare in open state for the latter.

1.5 L of dye-bath effluent having initial COD of 5800 mg L−1 andBOD of 181 mg L−1 was taken and experiments were carried out attwo different current densities of 2.5 A dm−2 and 5.0 A dm−2. Ateach current density, experiments were conducted for five differentflow rates of 20, 40, 60, 80 and 100 L h−1. The electrode plates werecleaned manually by washing with distilled water before every run.All the experiments were carried out for 6 h in batch recirculationmode under galvanostatic conditions during which the cell voltagewas noted and the samples were collected periodically for CODanalysis. The dye-bath effluent feed and the samples obtained whentreated under the current density of 5.0 A dm−2 and flow rate of80 L h−1 were subjected to spectral analysis.

Fig. 2. a) Schematic diagram of batch recirculation/continuous recycle flow reactor system (1. Reservoir, 2. Pump, 3. Rotameter, 4. Flow cell, 5. Anode, 6. Cathode 7. Digital Voltmeter, 8.Digital Ammeter, 9. D.C. Regulated Power Supply, 10. Treatedwastewater outlet, 11. Recirculation Stream, 12.Wastewater inlet to system. (For batch recirculation system, streams 10 and12 will be absent). b) Experimental set-up.

190 C.A. Basha et al. / Desalination 285 (2012) 188–197

191C.A. Basha et al. / Desalination 285 (2012) 188–197

The experiments were repeated with untreated wash water efflu-ent of COD 560 mg L−1 and BOD 30 mg L−1 under the same operatingconditions of current density and flow rate.

2.3. Recycle reactor

The untreated wash water of COD 560 mg L−1 and BOD30 mg L−1 was subjected to electrolysis at the current density of5.0 A dm−2 and a recycle flow rate (Q) of 80 L h−1. 2 L of treatedwash water was taken initially in the reservoir to which untreatedwash water was fed at different flow rates (q): 10, 20, 30, 40 and50 mL min−1. The treated wastewater outlet flow rate was adjustedto attain steady state reservoir level and to establish the required re-cycle flow rate Q and flow rate through the reactor (Q+q). DCpower was supplied to the electrodes to maintain a constant currentof 1.96 A, corresponding to a current density of 5 A dm−2. All the ex-periments were carried out till steady state was reached and main-tained for 30 min and after treatment samples collected wereanalyzed for COD.

2.4. Analysis

COD was determined by the dichromate closed reflux methodusing thermo reactor TR620-Merck. The interference of Cl− noticedin the range of its concentration (up to 4 mg L−1) during the estima-tion of COD was overcome by the addition of HgSO4 along with AgSO4

as recommended [25]. BOD was estimated using the Winkler's meth-od, following the American Public Health Association (APHA) proce-dures [25]. Experiments were repeated until the error found wasless than 3%. The initial BI values of dye-bath and wash watereffluents were 0.03 and 0.05 respectively which were noticeablylow. Absorbance of untreated and treated samples of dye-batheffluent was measured at appropriate wavelengths using UV–VisibleSpectrophotometer (Spectroquant NOVA 60, Merck). Thetemperature of the effluent during electrolysis was constant at35 °C. The concentrations of hypochlorite and chlorate were5 mg L−1 and 15.25 mg L−1 respectively.

3. Electro-oxidation — theoretical approach

In the presence of chloride, oxidation of organic compounds is me-diated by active chloro species. The main reactions at the electrodesare:

anode : 2Cl− →k1 Cl2 þ 2e− ð1Þ

cathode : 2H2Oþ 2e−→H2 þ 2OH−: ð2Þ

Cl2 gas is discharged on the anode rapidly as per reaction (1) inthe aqueous medium. Hydrolysis and ionization reactions are givenby Eqs. (3) and (4) respectively.

Cl2 þ H2O→k2 HOClþHþ þ Cl− ð3Þ

HOCl↔k3

k;3Hþ þ OCl−: ð4Þ

The side-reactions at the anode which evolve O2 and form chlorate,compete with Cl2 evolution as per the reactions,

H2O→2Hþ þ §O2 þ 2e− ð5Þ

6OCl− þ 3H2O→3=2O2 þ 6Hþ þ 4Cl− þ 2ClO−3 þ 6e− ð6Þ

Hypochlorous acid (HOCl) is replenished in themediumwith simul-taneous electrochemical reaction of NaCl in the reaction solution. HOCl

is a strong oxidant, which can oxidize the wastewater while OCl− isconsumed by reaction (6). In solutions of neutral tomoderate pH valuesa cycle of chloride–chlorine–hypochlorite–chloride occurs that causesthe initial concentration of chlorides to remain stable. However, instrong alkaline solutions, the cycle of chloride–chlorine–chloride isblockeddue to the production of stable ClO3

−. At lowpH chlorides are re-duced with the production of free chlorine while at high pH the chlo-rides are oxidized and chlorates are produced.

Considering the electrochemical oxidation mechanism proposedby the Comninellis group [27], which assumed the action of the elec-trochemical oxidation extended to a reaction cage in the vicinity ofthe electrode surface, rather than limited at the surface itself, the dis-tinction between direct and mediated oxidation becomes even lessstringent [28]. The Cl2-mediated mineralization has been shown togive good results at low-oxygen-overvoltage electrodes [29, 30]. Theaddition of chloride ions in the electrolyte allows an increase of theremoval efficiency, and a degradation of pollutants can be obtaineddue to the participation of active chlorine. Different authors have al-ready published the possible “direct” and “indirect” roles for the chlo-ride anion in the electrochemical reaction [27–33].

The generated gaseous Cl2 dissolves to produce hypochlorite ions re-peatedly. The hypochlorite ions can mineralize the dye waste, therebyimproving the biodegradability to an extent, especially during theearly stage of the treatment.

Dyeþ OCl− →k4 R þ CO2 þ N2 þ H2Oþ Cl−: ð7Þ

The indirect electrooxidation rate of organic pollutants depends on thediffusion rate of the oxidants into the solution, flow rate of the effluent,temperature and pH. In bulk, a reaction sequence of chloride–chlorine–hypochlorite–chloride takes place. The pseudo-steady state theory canbe applied to each of the intermediate product (HOCl and OCl−) asgiven in Eqs. (3), (4) and (7). The rates of reaction in the bulk solutioncan be written as:

−rCl2 ¼ k2 Cl2½ � ð8Þ

rHOCl ¼ k2 Cl2½ �−k3 HOCl½ � þ k3′ Hþh i

OCl−½ �−k4 Dye½ � OCl−½ � ¼ 0 ð9Þ

rOCl− ¼ k3 HOCl½ �−k3′ Hþ½ � OCl−½ �−k4 Dye½ � OCl−½ � ¼ 0 ð10Þ

−rOrganic¼ k4 Dye½ � OCl−½ �: ð11Þ

From the equation of the chemical mineralization of the organic com-pound (Eq. (12)),

CxHyNwOz þ4xþ y−2z

4

� �O2→xCO2 þ

w2N2 þ

y2H2O ð12Þ

it is possible to obtain the relation between the dye [dye] concentration inmol CxHyNwOz L−1 and the chemical oxygen demand [COD] in mol O2

L−1:

Dye½ � ¼ 44xþ y−2z

� �COD½ �: ð12Þ

From the above equations, rate equation can be expressed as:

−rCl2 ¼ −rDye ¼ k4 Dye OCl−½ � ¼ kCOD½ � OCl−½ �: ð13Þ

3.1. Design and performance of batch reactor

During electrolysis, if a constant operating current density is ap-plied, the rate of degradation of organic contaminant will remain con-stant under a given set of experimental conditions, but it will vary if

192 C.A. Basha et al. / Desalination 285 (2012) 188–197

the applied current density is changed. The reaction rate for the re-moval of COD in a batch reactor can be expressed as:

−dCdt

¼ izF

¼ khaC ð14Þ

CCo

¼ exp −khatð Þ ð15Þ

where C represents COD of the effluent in the batch reactor, i is thecurrent passed during electrolysis time t, a is specific electrode area(Ae/VR), Ae electrode area, VR reactor volume, kh is pseudo firstorder heterogeneous rate constant, F is the Faraday constant(96485.3 Cmol−1) and z is the number of electrons transferred permole of reaction. The unconverted species (1−X) or (C/Co) [by de-fining the extent of degradation, X=(Co−C)/Co], decreases expo-nentially with time, according to Eq. (15).

The rate of degradation is controlled by the operating current densi-ty and it is assessed in terms of the heterogeneous rate constant kh(cm s−1) by monitoring the extent of COD removal. In electrochemicalconversion, the high molecular weight aromatic compounds and ali-phatic chains are broken to intermediate products for further proces-sing. In electrochemical oxidation, the organics are oxidized to CO2

and H2O. The current efficiency of the electrolysis can be calculatedusing the following expressions:

Current Efficiency CEð Þ ¼ ΔCOD� VR

ð16It=2FÞ ¼ Co � X� VR

ð16It=2FÞ : ð16Þ

Process performance is defined in two ways, one with respect tothe extent of reaction completion and the other with respect to theenergy consumption, E (kW h kg−1).

Specific Energy Consumption; E tð Þ ¼ VIt� 103

ΔCOD� VR¼ VIt� 103

Co � X� VR: ð17Þ

Computation of the current efficiency (CE) indicates the path andcourse of the targeted reaction by monitoring the extent of COD re-moval (Eq. (16)). The utilization of energy in processing a unit quan-tity of the targeted reaction is found by Eq. (17).

3.2. Design and performance of batch recirculation reactor

In batch recirculation operation valves are closed to prevent theflow of streams 10 and 12 (Fig. 2a). There is a gradual depletion ofthe COD of wastewater in the reservoir of batch recirculation system.In order to design and control the reactor, the development of amodel is essential, which enables the prediction of the variation ofCOD of wastewater with time in the reservoir. The basic assumptionmade in deriving the model equation is that the COD of wastewateris changing along the flow direction in the reactor which was basedon the residence time distribution experiments conducted using atracer. A plug flow model may be considered for the reactor inwhich these reactions take place.

The concentration variation of COD of wastewater in the electro-chemical reactor can be written as:

∂C∂t ¼ − Q

Ac

∂C∂x−khaC ð18Þ

where Ac is the cross sectional flow area of the electrochemical reac-tor, L is the length of the reactor, Ae is the electrode area, VR=AcL,volume of the reactor, a is the specific surface area [Ae/VR] for reac-tion, Q is the volumetric flow rate, C is the COD of wastewater andkh is the heterogeneous rate constant. Left hand side represents therate of change of COD of wastewater in the reactor. Two terms ofthe right hand side of the Eq. (18) represent the rate of depletion of

COD of wastewater due to bulk movement of effluent and electro-chemical oxidation reactions. Since there is no accumulation of CODof wastewater in the liquid phase of the reactor, it can be assumedthat the reactor is under steady state, ∂C/∂t=0, and C=C for CODof wastewater entering the reactor [at x=0] and C=C for COD ofwastewater leaving the reactor [at x=L]. The steady state solutionof Eq. (18) yields

C′

C¼ exp −khaACL

Q

� �¼ exp − khAe

Q

� �: ð19Þ

Considering the reservoir as a perfect backmix system, the massbalance equation for the reservoir is

−VresdCdt

¼ QðC−C′Þ ð20Þ

where Vres is the volume of the reservoir tank.The mass balance Eq. (20) is solved after substitution of the ex-

pression for C from Eq. (19), knowing the initial COD of wastewater,C=Co at t=0 in the reservoir. The following resultant equation thengives the variation of COD of wastewater in the reservoir.

CCo

¼ exp − tτ

1−exp − khAQ

� �� �� �ð21Þ

where τ is the residence time (Vres /Q).The unconverted species (1−X) or (C/Co) decreases exponentially

with time, according to Eq. (21). The slope of the plot of −ln (1−X)versus t, gives the value 1−exp −khAe

Q

� =τ from which the value of kh,

the heterogeneous rate constant, can be computed. Experiments haveshown that the electrolysis time, current density and recirculationflow rate considerably influence the performance of the process. The ac-tual utilization of energy for processing a unit quantity of the targetedreaction is specific power consumption and it is found by monitoringthe cell voltage and extent of COD removal and can be written as:

Specific Energy Consumption;EðtÞ ¼ VIt� 103

ΔCOD� Vres

¼ VIt� 103

Co � XðtÞ � Vresð22Þ

where X(t) is the extent of conversion, Co (initial COD in mg L−1), dueto passing current I (A) for time t (h); Vres is volume of the reservoir(L); and ‘V’ represents the cell voltage in volts.

3.3. Design and performance of recycle reactor

In the recycle operation valves are opened to allow the flow ofstreams 10, 11 and 12 (Fig. 2a). The material balance around the res-ervoir gives concentration of the reactant entering the reactor as

C ¼ qC0 þ QC′

qþ Q¼ RC0 þ C′

R þ 1ð23Þ

where Q is the recycle flow rate of the treated effluent, q is the inlet(fresh)/discharge(treated) effluent volumetric flow rate, and ‘R’ is theratio, defined as R=q/Q. The material balance around the reactor is

C′ ¼ Cexpð−khaτRÞ ¼ Cexp − khAe

Q þ q

� �ð24Þ

where Q+q is the total flow rate of the effluent passing through the re-actor. From Eqs. (23) and (24) we get,

C ¼RCo þ Cexp − kLAe

Qð1þRÞ�

R þ 1: ð25Þ

Table 2Effect of reactor holdup and current density on the performance of the batch electrochemical reactor: initial COD 5800±20 mg L−1; electrolysis time 8 h; active electrode area50 cm2.

VR

mLCurrent densityA dm−2

CurrentA

Cell voltageV

Potential COD reduction%

Current efficiency%

Power consumptionk Wh (kg COD)−1

kh×104

cm s−1

AnodicV

CathodicV

300 1 0.5 2.9 1.13 1.73 71.71 – 9.29 2.632 1 3.3 1.62 1.73 89.66 65.33 16.92 4.733 1.5 3.58 1.77 1.84 96.73 46.98 25.52 7.124 2 3.84 1.83 1.98 99.15 36.12 35.61 9.925 2.5 4.31 2.09 2.24 99.83 29.09 49.62 13.3

400 1 0.5 2.78 1.23 1.51 56.95 – 8.41 2.342 1 3.26 1.75 1.52 80.53 78.23 13.95 4.543 1.5 3.53 1.805 1.71 91.93 59.54 19.86 6.994 2 3.87 1.87 1.99 97.01 47.12 27.51 9.735 2.5 4.22 2.13 2.11 99.03 38.48 36.73 12.87

500 1 0.5 2.86 1.18 1.63 46.86 – 8.41 2.192 1 3.21 1.38 1.83 73.26 88.97 12.08 4.573 1.5 3.69 1.58 2.19 87.38 70.74 17.47 7.184 2 3.83 1.62 2.22 94.49 57.37 22.36 10.065 2.5 4.21 1.78 2.27 97.81 47.51 29.68 13.27

193C.A. Basha et al. / Desalination 285 (2012) 188–197

The above equation can be rearranged as

khAe

Qð1þ RÞ� �

¼ ln1−X

ðR þ 1Þð1−XÞ−R

� �: ð26Þ

The slope of the plot ln 1−XðRþ1Þð1−XÞ−R

� versus 1/(1+R)Q, gives the

value khAe as per Eq. (26) from which kh, the rate constant, can becomputed for the given conversion. The specific power consumptioncan be found by

Specific Energy Consumption; E ¼ VI� 103

ΔCOD� RQ¼ VI� 103

Co � X� RQð27Þ

where X is the extent of conversion, Co is the initial COD in mg L−1, Iis the passing current and V is the applied cell voltage.

VR=300 mL, a=0.16

1

1.5

2

2.5

3

3.5

4

4.5

0 60 120 180 240

Time (min)

Vo

ltag

e (V

)

Fig. 3. Variation of cell voltage, and anode and cathode potentials in batch reactor at diffe50 cm2, reactor holdup 300 mL.

4. Results and discussion

4.1. Electrolysis of dye-bath effluent in batch reactor

The results of the experiments conducted in the batch reactor arereported in Table 2 and shown in Figs. 3 and 4a, b and c.

Fig. 3 shows the effect of current density on cell voltage, anodeand cathode potentials with time for reactor holdup volume of300 mL. The cell voltage, anode and cathode potentials varied re-markably with the current density and marginally with time at allcurrent densities. However, there was no significant change in themwith the current density and time when reactor holdup volumes are400 mL and 500 mL (not shown).

Fig. 4(a), (b) and (c) depicts the % COD reduction with electrolysistime for various current densities at the three different reactor holdupvolumes of 300, 400 and 500 mL respectively. In all cases it can beseen that the % removal of the organic matter has increased exponen-tially with time. The increase in pollutant removal beyond 5 h oftreatment is marginal when cell is operated at current densities

68

300 360 420 480

1.0 A/dm, C.V

1.0 A/dm, A.P

1.0 A/dm, C.P

2.0 A/dm, C.V

2.0 A/dm, A.P

2.0 A/dm, C.P

3.0 A/dm, C.V

3.0 A/dm, A.P

3.0 A/dm, C.P

4.0 A/dm, C.V

4.0 A/dm, A.P

4.0 A/dm, C.P

5.0 A/dm, C.V

5.0 A/dm, A.P

5.0 A/dm, A.P

rent current densities. Conditions: initial COD 5800 mg L−1; active electrode surface

VR =300 mL, a=0.1668 cm-1

0

20

40

60

80

100

0 60 120 180 240 300 360 420 480

Time (min)

0 60 120 180 240 300 360 420 480

Time (min)

0 60 120 180 240 300 360 420 480

Time (min)

CO

D R

edu

ctio

n (

%)

0

20

40

60

80

100

CO

D R

edu

ctio

n (

%)

0

20

40

60

80

100

CO

D R

edu

ctio

n (

%)

1.0 A/dm2 2.0 A/dm2 3.0 A/dm24.0 A/dm2 5.0 A/dm2

VR=400 mL, a= 0.1251 cm-1

1.0 A/dm2 2.0 A/dm2 3.0 A/dm2

4.0 A/dm2 5.0 A/dm2

VR=500 mL, a=0.1001 cm-1

1.0 A/dm2 2.0 A/dm2 3.0 A/dm2

4.0 A/dm2 5.0 A/dm2

a

b

c

Fig. 4. Effect of electrolysis time and current density on percentage removal of COD inthe batch system. Conditions: initial COD 5800 mg L−1; active electrode surface50 cm2. Reactor holdup a) 300 mL; b) 400 mL; c) 500 mL.

Dye-bath Effluent

Current Density = 2.5 A dm-2

Co = 5800 mg L-1

kL=4.3x10-3u0.1213

0

20

40

60

80

100

0 1 2 3 4 5 6Time (h)

0 1 2 3 4 5 6Time (h)

CO

D R

edu

ctio

n (

%)

CO

D R

edu

ctio

n (

%)

20 L/h40 L/h60 L/h80 L/h100 L/h

Dye bath effluent

Current Density = 5.0 A dm-2

Co = 5800 mg L-1

kL=4.6x10-3u0.1

0

20

40

60

80

100

20 L/h40 L/h60 L/h80 L/h100 L/h

a

b

Fig. 5. Variation of COD reduction with electrolysis time at different flow rate in batchrecirculation system for dye-bath effluent at the applied current density of a)2.5 A dm−2 b) 5.0 A dm−2.

194 C.A. Basha et al. / Desalination 285 (2012) 188–197

above 3 Adm−2. It can also be seen that with increase in current den-sity COD reduction is increased but the current efficiency is de-creased. Increasing the current density from 1 to 5 Adm−2

increased the COD removal from 71.71% to 99.83% after 8 h of treat-ment but the current efficiency dropped to 29% when the reactorholdup was 300 mL. Comparison of Fig. 4(a), (b) and (c) shows thatan increase in reactor holdup has not increased the COD removal sig-nificantly but it has resulted in higher current efficiency and lesserpower consumption. A more noticeable fact in Table 2 is that an in-crease in current density decreases the current efficiency and in-creases power consumption. The possibility of loss of electricalenergy in the form of heat and unwanted reactions is greater athigher levels of current density. Thus, in general, a higher currentdensity operation will give high capacity utilization, at the expenseof higher energy loss. The results also show that COD removal de-creases with a decrease in the specific electrode area.

The variation of heterogeneous rate constant kh (cm s−1) withcurrent density and reactor holdup (VR) was also studied for thebatch operation. kh is found to increase with current density, proba-bly due to the generation of more oxidizing agents at higher currentdensity values. Also kh (cm s−1) is found to increase with the de-crease in the specific electrode surface area. This may be due topoorer extent of reaction though the amount of reactant availablefor reaction is more. The operation is more energy efficient whenthe specific electrode surface is low. In general operation at lowerspecific electrode surface is advantageous because of the high capac-ity of operation and/or degradation of higher amounts of pollutants atbetter energy figures, but with poor extent of completion of theprocess.

4.2. Electrolysis of dye-bath effluent in batch recirculation reactor

Electrolysis of dye-bath effluent of 5800 mg L−1 initial COD wascarried out in a batch recirculation reactor at five different flowrates of 20, 40, 60, 80 and 100 L h−1 and at the current density of2.5 Adm−2 or 5.0 Adm−2. It is seen from Fig. 5 that the percentageCOD reduction increases as the electrolysis time increases. Thedegradation rate was higher during the initial 2 h of electrolysis forall flow rates. The increase in the rate and the subsequent increasein the percentage COD reduction are larger in the beginning of the

Dye-bath Effluent Current Density=2.5 A dm-2

Co = 5800 mg L -1

0.8

1

1.2

1.4

1.6

1.8

2

2.2a

b

1 2 3 4 5 6

Time (h)

1 2 3 4 5 6

Time (h)

E(t

), k

W h

(kg

CO

D)-1

E(t

), k

W h

(kg

CO

D)-1

20 L/h 40 L/h 60 L/h

80 L/h 100 L/h

Dye-bath EffluentCurrent Density = 5 A dm-2

Co = 5800 mg L -1

1.7

2.2

2.7

3.2

3.7

4.2

4.7

20 L/h 40 L/h 60 L/h

80 L/h 100 L/h

Fig. 6. Variation of specific energy consumption with electrolysis time at different flowrate in batch recirculation system for dye-bath effluent at the applied current density ofa) 2.5 A dm−2 b) 5.0 A dm−2.

Washwater Effluent

Current Density = 5 A/dm-2

Co = 560 mg L-1

kL=3.1x10-3u0.27

0

20

40

60

80

100

0 1 2 3 4 5 6

Time (h)

CO

D R

edu

ctio

n (

%)

20 L/h40 L/h60 L/h80 L/h100 L/h

Fig. 8. Variation of COD reduction with electrolysis time at different flow rate in batchrecirculation system for wash water effluent at the applied current density of5.0 A dm−2.

195C.A. Basha et al. / Desalination 285 (2012) 188–197

process which decreased gradually to approach zero increase in thelater hours of electrolysis.

Fig. 5a and b shows that the rate of degradation of organic con-taminants increases with the increase in flow rate and the percentageCOD reduction increases to reach a constant value at the current den-sities 2.5 Adm−2 and 5.0 Adm−2 respectively. The rate of organiccontaminant degradation is dependent on its bulk concentration. Athigher flow rates the rate of destruction of organic contaminant is sig-nificantly more and this may be due to the increased production ofoxidants in the process. As the electrolyte circulation rate increases,the transport of the Cl− ion from the bulk to the electrode surfaces in-creases. This would facilitate Cl2 generation, as well as the dissolutionof Cl2 to form OCl− for reaction with the organic contaminant. Theheterogeneous rates constant were computed for the oxidation ofthe organic contaminant. It is 4.286×10−3 u0.1213 at the current den-sity of 2.5 A dm−2 and 4.551×10−3 u0.0971 at the current density of5.0 A dm−2. Here ‘u’ represents the velocity in cm s−1.

Fig. 7. Effect of electrolysis on color removal for dye-bath effluent at the appliedcurrent density of 5 A dm−2.

The COD removal in 6 h of operation was increased from 88.66 to93.93% as the flow rate was increased from 20 to 100 L h−1 when thecurrent density was 2.5 A dm−2. At a current density of 5.0 A dm−2

the increase was from 79.87 to 91.38% when the flow rate was in-creased from 20 to 100 L h−1. The cell voltages did not vary signifi-cantly with the circulation flow rates at both current densities,around 2.7 V for the current density 2.5 A dm−2 and 3.1 V for the cur-rent density 5 A dm−2. The improvement in voltage (2.79–2.75 V forthe current density 2.5 A dm−2 and 3.13–3.04 V for the current den-sity 5.0 A dm−2) and in COD removal considerably affected the powerconsumption when the flow rate ‘Q’ was increased from 20 to100 L h−1.

When flow rate was increased from 20 to 100 L h−1, at the currentdensity 2.5 A dm−2 the specific energy consumption decreased from2.1 to 1.98 kWh (kg COD)−1 and at current density 5 A dm−2 it de-creased from 4.63 to 4.33 kWh (kg COD)−1. The enhanced currentdensity not only decreases the reduction of pollutant removal(93.93% for the current density 2.5 A dm−2 and 91.38% for the currentdensity 5 A dm−2 in 6 h) but also largely increases the specific energyconsumption (from 1.98 kWh kg−1 for the current density2.5 A dm−2 to 4.33 kWh kg−1 for the current density 5 A dm−2 atflow rate is 100 L h−1 as depicted in Fig. 6a and b). From the results,it was clear that the applied current density and effluent initial COD con-centration influence the electrolysismore than the circulation flow rates.The increase in heterogeneous rate constant due to the increase incurrent density is observed to be marginal (from 4.982×10−3 cm s−1

Washwater Effluent Current Density = 5 A dm-2

Co = 560 mg L-1

35

45

55

65

75

85

95

1 2 3 4 5 6

Time (h)

E(t

), k

W h

(kg

CO

D)-1

20 L/h 40 L/h 60 L/h

80 L/h 100 L/h

Fig. 9. Variation of specific energy consumption with electrolysis time at different flowrate in batch recirculation system for wash water effluent at the applied current densityof 5.0 A dm−2.

Table 3Effect of discharge flow rate on the performance of continuous recycle flow reactor for wash water with the initial concentration 560 mgL−1under current density 5.0 A dm−2 andrecycle flow rate 80 L h−1.

Discharge flow rateq (Lh−1)

Cell voltage(V)

COD reduction(%)

BI Power consumption[kWh (kg COD)−1]

Heterogeneous rate constantcm s−1

0.6 5.3 75.71 0.63 40.83 0.0135121.2 5.3 67.86 0.61 22.78 0.0185141.8 5.3 61.07 0.59 16.88 0.0208282.4 5.4 56.79 0.52 13.87 0.0234823.0 5.4 52.86 0.51 11.92 0.025262

196 C.A. Basha et al. / Desalination 285 (2012) 188–197

for the current density 2.5 A dm−2 to 5.107×10−3 cm s−1 for thecurrent density 5 A dm−2) and in all cases BI increased (0.48) after 1 hof electrolysis.

Raw and treated samples of dye-bath effluent were subjected toabsorbance measurements for different wavelengths using UV–Visible Spectrophotometer. Fig. 7 exhibits the absorption behaviorof both untreated and treated dye-bath effluent for differentwavelengths varying from 340 nm to 865 nm. The absorption of theuntreated effluent shows a peak at wavelength 605 nm, indicatingthat color is present in the sample due to the selected dye. Theabsorption of the treated effluent at the wavelength 605 nm isnearly zero. This indicates the absence of coloring substance (dye)in the effluent, which has disappeared due to electrolysis.

4.3. Electrolysis of wash water in batch recirculation reactor

Electrolysis of wash water effluent of 560 mg L−1 initial COD wascarried out at the current density of 5.0 A dm−2 for five different flowrates of 20, 40, 60, 80 and 100 L h−1. It is seen from Fig. 8 that the CODdecreased exponentially with electrolysis time. Higher degradationrate was observed during the initial period at all flow rates. The %COD removal increased from 79.87 to 91.38% in 6 h of operation asthe flow rate was increased from 20 to 100 L h−1. The value of theheterogeneous rate constant kh at the current density of 5.0 A dm−2

was determined as 3.135×10−3 u0.2747 using Eq. (21).The observed cell voltage was around 5.5 V for the current density

5 A dm−2. When the flow rate was increased from 20 to 100 L h−1

the cell voltage and COD removal were found to improve, resultingin decreased power consumption. This may be due to the improvedionic conductivity of bulk movement and the reduction of resistanceon the electrode surface resulting in the enhancement of heteroge-neous rate constant at higher flow rates. It can be seen from Fig. 9that the specific energy consumption decreased from 97.3 to83.8 kWh (kg COD)−1 in 6 h of operation when the flow rate in-creased from 20 to 100 L h−1.

4.4. Electrolysis of wash water in recycle reactor

The results obtained when wash water effluent was treated in acontinuous recycle flow reactor at the circulation flow rate of80 L h−1 and current density of 5 A dm−2 are shown in Table 3. Thepercentage COD reduction decreased from 75.71 to 52.86% whenthe discharge flow rates increased from 0.6 to 3 L h−1 at a constantrecycle flow rate of 80 L h−1 indicating that an increase in withdraw-al flow rate decreases the % COD reduction. This may be is due to theshorter residence time at the higher withdrawal flow rates. The CODof the effluent was found to decrease from its initial value of560 mg L−1 to a value of 240 mg L−1 at the withdrawal flow rate of2.4 L h−1 and below this rate the COD limit is below the limits of tol-erance. It is seen that with increased withdrawal flow rates, powerconsumption decreases and the value of rate constant kh increases.In general, the operation of the reactor at higher withdrawal flowrates provides good energy figures but poor reaction performancewith an enhancement in biodegradability (see Table 3). The perfor-mance of recycle reactor is poorer when compared to the batch

recirculation reactor on power consumption figures and rate con-stant. The study of variation of rate constant with the effluent circula-tion flow rate can be of much interest in this system.

5. Conclusion

Batch studies provide a good number of combinations (Table 2) ofthe operating parameters of current density and reactor holdup forgood the overall system performance. For dye bath, at the currentdensity of 4.0 A dm−2, 300 mL reactor holdup, electrolysis duration:8 h yield 99.17% COD removal with specific energy consumption of35.6 kWh (kg COD)−1. For dye bath, at the current density of4.0 A dm−2, 300 mL reactor holdup, electrolysis duration: 8 h yield99.17% COD removal with specific energy consumption of 35.6 kWh(kg COD)−1.

Batch recirculation mode of operation was superior for dye-batheffluent, for COD removal with the specific energy consumption(93.93% and 1.98 kWh (kg COD)−1) in 6 h of operation at the flowrate of 100 L h−1 and the current density 2.5 A dm−2. For washwater the % COD removal was 91.38% in 6 h of operation (flow rate100 L h−1, current density 5.0 A dm−2) with the specific energyconsumption of 83.8 kWh (kg COD)−1.

The recycle system of operation has the best performance of theconventional reactor configurations studied. The capacity utilizationof the system improved considerably under continuous recirculationcondition than the batch recirculation. The best operating point (Q80 L h−1, q 2.4 L h−1) of the best reactor configuration (recycle reac-tor) was further investigated by estimating the biodegradabilityindex, resulting in a considerable improvement to 0.52±0.01. Thisshows that the remaining 40% of the COD can be easily digested bio-logically. Thus, the advantage of the process is two-fold; it not onlyremoves the bulk quantity of organic matter (57%), but also makesthe remaining matter more easily biodegradable.

It is concluded that the complete removal of COD could beachieved in batch reactor and in recirculation reactor if the electroly-sis time is long enough, but the specific energy will be very large com-pared with continuous operation. The purpose behind thisinvestigation is to estimate the energy demand for the electrochemi-cal mineralization that enhances biodegradability and the resultswould be helpful for optimization of electrochemical oxidation tech-nology. Electrochemical oxidation of intermediates to carbon dioxideand water is difficult and less energy efficient, because the rate ofelectrochemical C–C bond scission seems to decrease with decreasingmolecular size. Present studies indicate that electrochemical degrada-tion may not be a viably economic proposal if employed alone be-cause of the high energy consumption and low conversion. Effortscan be nowmade to devise a technique in which a conventional tech-nique such as biological treatment is integrated with electrochemicaloxidation with the knowledge of specific energy/charge required forcomplete mineralization of effluent.

Acknowledgements

One of the authors (CAB) thanks the Korean Federation of Scienceand Technology Societies Grant funded by Korea Government (MEST,

197C.A. Basha et al. / Desalination 285 (2012) 188–197

Basic Research Promotion Fund) (No. 101S-4-3-0270) for the finan-cial support of this work.

References

[1] T. Robinson, G. McMullan, R. Marchant, P. Nigam, Remediation of dyes in textile efflu-ent: a critical review on current treatment technologies with a proposed alternative,Bioresour. Technol. 77 (2001) 247–255.

[2] H. Keharia, D. Madamwar, Textile and dye effluent, in: A. Pandey (Ed.), ConciseEncyclopedia of Bioresource Technology, The Haworth Press, New York, 2004,pp. 167–175.

[3] M. Panizza, G. Cerisola, Direct and mediated anodic oxidation of organic pollutants,Chem. Rev. 109 (2009) 6541–6569.

[4] C.A. Martinez-Huitle, E. Brillas, Decontamination of wastewaters containing syntheticorganic dyes by electrochemicalmethods: a general review, Appl. Catal. B: Environ. 87(2009) 105–145.

[5] G. Chen, Electrochemical technologies in wastewater treatment, Sep. Purif. Technol.38 (2004) 11–41.

[6] D. Rajkumar, J.G. Kim, Oxidation of various reactive dyeswith in situ electrogeneratedactive chlorine for textile dyeing industry wastewater treatment, J. Hazard. Mater. B136 (2006) 203–212.

[7] S. Raghu, C. Ahmed Basha, Electrochemical treatment of Procion Black 5B usingcylindrical flow reactor—a pilot plant study, J. Hazard. Mater. B139 (2007)381–390.

[8] A.S. Koparal, Y. Yavuz, C. Gurel, U.B. Ogutveren, Electrochemical degradation andtoxicity reduction of C.I. Basic Red 29 solution and textile wastewater by using di-amond anode, J. Hazard. Mater. 145 (2007) 100–108.

[9] L.W. Lackey, R.O. Mines Jr., P.T. Mc Creanor, Ozonation of acid yellow 17 dye in asemi-batch bubble column, J. Hazard. Mater. B138 (2006) 357–362.

[10] A. Aguedach, S. Brosillon, J. Morvan, E.K. Lhadi, Photocatalytic degradation of azo-dyesreactive black 5 and reactive yellow 145 inwater over a newly deposited titanium di-oxide, Appl. Catal., B 57 (2005) 55–62.

[11] E. Bizani, K. Fytianos, I. Poulios, V. Tsiridis, Photocatalytic decolorization and degrada-tion of dye solutions and wastewaters in the presence of titanium dioxide, J. Hazard.Mater. 136 (2006) 85–94.

[12] S. Vajnhandl, A.M.L. Marechal, Ultrasound in textile dyeing and the decolouration/mineralization of textile dyes, Dyes Pigm. 65 (2005) 89–102.

[13] K. Murugesan, A. Dhamija, I. Nam, Y. Kim, Y. Chang, Decolourization of reactiveblack 5 by laccase: optimization by response surface methodology, Dyes Pigm.75 (2007) 176–184.

[14] L.C. Davies, I.S. Pedro, J.M. Novais, S. Martins-Dias, Aerobic degradation of acid orange7 in a vertical-flow constructed wetland, Water Res. 40 (2006) 2055–2063.

[15] G.R.P. Malpass, A.J. Motheo, Galvanostatic oxidation of formaldehyde-methanol solu-tions on Ti/Ru0.3Ti0.7O2 electrodes using a filter-press cell, J. Appl. Electrochem. 31(2001) 1351–1357.

[16] O. Simond, V. Schaller, C. Comninellis, Theoretical model for the anodic oxidationof organics on metal oxide electrodes, Electrochim. Acta 42 (1997) 2009–2012.

[17] J.L. Morais, P.P. Zamora, Use of advanced oxidation processes to improve the biode-gradability of mature landfill leachates, J. Hazard. Mater. 123 (2005) 181–186.

[18] E. Chamarro, A. Marco, S. Esplugas, Use of fenton reagent to improve organicchemical bio-degradability, Water Res. 35 (2001) 1047–1051.

[19] R.C. Loehr, Pollution Control for Agriculture, Academic Press, Inc., New York, 1977.[20] K.V. Selvakumar, C. Ahmed Basha, H.J. Prabhu, P. Kalaichelvi, N. Sudha, The potential of

free cells of Pseudomonas aeruginosa on textile dye degradation, Bioresour. Technol.101 (2010) 2678–2684.

[21] K.V. Selvakumar, C. Ahmed Basha, H.J. Prabhu, N. Anantharaman, N. Jayanthi,Electro oxidation and biodegradation of textile dye effluent containing ProcionBlue 2G using fungal strain Phanerochate chrysosporium MTCC 787, Int. J. Chem.React. Eng. 8 (2010) A147.

[22] C. Ahmed Basha, K.V. Selvakumar, H.J. Prabhu, P. Sivashanmugam, Lee ChangWoo, Degradation studies for textile reactive dye by combined electrochemical,microbial and photocatalytic methods, Sep. Purif. Technol. 79 (2011) 303–309.

[23] A. Sakalis, K. Mpoulmpasakos, U. Nickel, K. Fytianos, A. Voulgaropoulos, Evaluation of anovel electrochemical pilot plant process for azodyes removal from textilewastewater,Chem. Eng. J. 111 (2005) 63–70.

[24] A. Vlyssides, D. Arapoglou, S.Mai, E.M. Barampouti, Electrochemical detoxification offour phosphorothioate obsolete pesticides stocks, Chemosphere 58 (2005) 439–447.

[25] L.S. Clesceri, A.E. Greenberg, A.D. Eaton, Standard Methods for the Examination ofWater and Wastewater, 20th ed APHA, Washington DC, 1998.

[26] P. Subbiah, S. Krishnamurthy, K. Asokan, K. Subramanian, V. Arumugam, An improvedprocess for the preparation of insoluble non-precious metal oxide anode doped withplatinum group metal oxide to be used in electrochemical processes. Indian Patent178184 (1990).

[27] P. A. Michaud, Thesis No. 2595, EPFL, Switzerland, 2002.[28] C.A. Martı´nez-Huitle, M.A. Quiroz, Ch. Comninellis, S. Ferro, A. De Battisti, Electro-

chemical incineration of chloranilic acid using Ti/IrO2, Pb/PbO2 and Si/BDDelectrodes,Electrochim. Acta 50 (2004) 949–956.

[29] F. Bonfatti, A. De Battisti, S. Ferro, G. Lodi, S. Osti, Anodicmineralization of organic sub-strates in chloride-containing aqueous media, Electrochim. Acta 46 (2000) 305–314.

[30] C.A. Martı´nez-Huitle, S. Ferro, A. De Battisti, Electrochemical incineration in thepresence of halides, Electrochem. Solid-State Lett. 8 (2005) D35–D39.

[31] M. Panizza, G. Cerisola, Electrochemical oxidation of 2-naphthol with in situ electro-generated active chlorine, Electrochim. Acta 48 (2003) 1515–1519.

[32] Ch. Comninellis, A. Nerini, Anodic oxidation of phenol in the presence of NaCl forwastewater treatment, J. Appl. Electrochem. 25 (1995) 23–28.

[33] M. Panizza, G. Cerisola, Influenceof anodematerial on theelectrochemical oxidationof 2-naphthol: Part 2. Bulk electrolysis experiments, Electrochim. Acta 49 (2004) 3221–3226.