原文1--Electro-coagulation of reactive textile dyes and textile wastewater

Chemical Engineering and Processing44(2005)461–470Electro-coagulation of reactive textile dyes and textile wastewaterA.Alinsafia,b,M.Khemis a,M.N.Pons a,∗,J.P.Leclerc a,A.Yaacoubi b,A.Benhammou c,A.Nejmeddine da Laboratoire des Sciences du G´e nie Chimique,CNRS-ENSIC-INPL,1,rue Grandville,BP451,F-54001Nancy Cedex,Franceb D´e partement de Chimie,Semlalia Faculty of Sciences,Cadi Ayyad University,Boulevard Prince Moulay Abdellah,BP2390,40000Marrakech,Moroccoc Laboratoire d’Automatique et d’Etudes des Proc´e d´e s,Semlalia Faculty of Sciences,Cadi Ayyad University,Boulevard Prince Moulay Abdellah,BP2390,40000Marrakech,Moroccod D´e partement de Biologie,Semlalia Faculty of Sciences,Cadi Ayyad University,Boulevard Prince Moulay Abdellah,BP2390,40000Marrakech,MoroccoReceived20November2003;received in revised form17February2004;accepted10June2004Available online25August2004AbstractElectro-coagulation of a blue reactive dye(Drimarene K2LR CDG Blue)solution has been optimised by experimental design and surface response analysis in terms of colour removal and chemical oxygen demand(COD)decrease.The optimal conditions(pH,current density,reac-tion time)have then been applied to other reactive dyes solutions as well as synthetic and real textile wastewater samples.The biodegradability before and after electro-coagulation has been assessed by short-term respirometry and is increased by this type of treatment.©2004Elsevier B.V.All rights reserved.Keywords:Electro-coagulation;Reactive dye;Colour removal;Experimental design;Biodegradability1.IntroductionWastewater from textile dyeing andfinishing factories is a significant source of environmental pollution[1].Reactive dyes are extensively used in textile industry,fundamentally due to the ability of their reactive groups to bind to textile fibres by covalent bonds formation[2].These characteristics facilitate the interaction with thefibre and reduce energy con-sumption[3].The major environmental problem associated with the use of the reactive dyes is their loss in the dyeing process.Thefixation efficiency is in the range60–90%[3]. Consequently,substantial amounts of unfixed dyes are re-leased in wastewater.Textile wastewater is characterized by high chemical oxygen demand(COD),low biodegradability, high-salt content and is the source of aesthetic pollution re-lated to colour.EU directive91/271imposes limits on colour, as it reduces light penetration in receiving water bodies.∗Corresponding author.Tel.:+33383175277;fax:+33383175326.E-mail address:marie-noelle.pons@ensic.inpl-nancy.fr(M.N.Pons).A wide range of wastewater treatment techniques have been tested.Biological processes(aerobic and anaerobic) [4–6]are probably the most inexpensive ones but dyes are inhibiting bacterial development[7].A pre-treatment is of-ten necessary to increase the biodegradability.A large range of physico-chemical processes have been proposed:coag-ulation with alum,ferric chloride,magnesium chloride and lime or polymers[4],adsorption on activated carbon,polymer and mineral sorbents or biosorbents[8,9],chemical oxidation [1,10,11],photolysis[11,12],suspended[13,14]or supported photocatalysis[15],electrophotocatalysis[16]but,except in the two last cases,they require the addition of chemicals.Electro-coagulation has been successfully used to treat a variety of industrial wastewaters[17–22].The goal is to form flocs of metal hydroxides within the effluent to be cleaned by electro-dissolution of soluble anodes.Three main processes occur during electro-coagulation:electrolytic reactions at the surface of electrodes,formation of coagulants in aqueous phase,adsorption of soluble or colloidal pollutants on co-agulants,and removal by sedimentation andfloatation.The0255-2701/$–see front matter©2004Elsevier B.V.All rights reserved. doi:10.1016/j.cep.2004.06.010462 A.Alinsafiet al./Chemical Engineering and Processing44(2005)461–470main reactions at the electrodes are:Al→Al3++3e−(at anode)(1) 3H2O+3e−→32H2+3OH−(at cathode)(2) The destabilized particles then aggregate to formflocs.In the meantime,the tiny hydrogen bubbles produced at the cath-ode induce thefloatation of mostflocs,helping to effectively separate particles from wastewater.In addition,the cathode may be chemically attacked by OH−ions generated together with H2at high pH values[23]. 2Al+6H2O+2OH−→2Al(OH4)−+3H2(3) Al(aq)3+and OH−ions generated by electrode reactions(1) and(2)react to form various monomeric species which trans-formfinally into Al(OH)3according to complex precipitation kinetics[24].Several interaction mechanisms are possible be-tween dye molecules and hydrolysis by-products.Two major mechanisms are generally considered:precipitation(for pH lower than6.5)and adsorption(for higher pH)[24,25].Precipitation:Dye+monomeric Al→[Dye-monomeric Al](s)pH4–5(4)Dye+polymeric Al→[Dye-polymeric Al](s)pH5–6(5)Adsorption:Dye+Al(OH)3(s)→→[particle](6) [Dye-polymeric Al](s)+Al(OH)3(s)→→→[particle](7) Theseflocs polymerise as:n Al(OH)3→Al n(OH)3n(8) Compared with traditionalflocculation and coagulation, electro-coagulation has,in theory,the advantage of removing small colloidal particles;they have a larger probability of be-ing coagulated because of the electricfield that sets them in motion.Addition of excessive amount of coagulants can be avoided,due to their direct generation by electro-oxidation of a sacrificial anode.Electro-coagulation equipment is simple and easy to operate.Short reaction time and low sludge pro-duction[26]are two other advantages of the technique.There are,however,several parameters such as size,shape and dis-tance between electrodes,current density,conductivity,pH, reaction time that should be selected with care to optimise the process efficiency.G¨u rses[24]have investigated the effect of electrode nature,mixing,cell voltage,electrolysis time and current density on aqueous solutions of reactive dyes.The present study aims at evaluating the influence of the key process variables such as current density,electrolysis time and initial pH on the efficiency of treatment(or pre-treatment)of textile wastewater containing reactive dyes. Textile wastewater exhibit usually high pH and contains large amounts of substances such as sizing agents,surfac-tants,volatile organic compounds,salts that can interfere with electro-coagulation.The experiments were carried out according to a23full factorial experimental design.Then, the process was optimised according to a22factorial exper-imental design to improve the efficiency of decolourisation and the abatement of organic matter.Finally the effect of electro-coagulation performed under optimal conditions on the biodegradability of the effluents has been investigated.2.Materials and methodsAluminum(composition:C:0.45–0.5%;Mn:0.5–0.8%; Si:0.4%;Al:98.3–98.61%)flat electrodes of rectangular shape(height=100mm;width=50mm and distance be-tween electrodes=20mm)were used in a100ml electro-coagulator made out of Pyrex glass.The solution to be treated is stored in a magnetically stirred reservoir.The experimental set-up is given in Fig.1.Sodium chloride was added in all runs so to have a conductivity of the solution of4.7mS/cm.Cell voltage and current were measured by digital voltmeter and ampmeter,respectively.The solution(total volume=500ml) was continuously circulated in the system with the help of a peristaltic pump at aflowrate of370ml/min.This corre-sponds to a Reynolds number of about220in the reaction chamber.Theflow regime is laminar and favours the growth of largeflocs that are easier to remove.The current efficiency(φ)for the production of dissolved Al3+was calculated by comparison of the weight loss of the aluminium electrodes during the experiment( m exp)with the theoretical amount of aluminium consumed according to Faraday’s law( m theo).φ=m expm theo(9) where m theo=ItM Al/3F with I(A)is the current intensity, t(s)the experiment duration,M Al(g/mol)the molar mass of aluminium and F the Faradayconstant.Fig.1.Experimental set-up.1:DC power supply,2:Digital Ampermeter, 3:Digital V oltmeter,4:Anode and cathode,5:Sample of wastewater,6: Peristaltic pump,7:Magnetic stirring controller.A.Alinsafiet al./Chemical Engineering and Processing 44(2005)461–470463The specific electrical energy consumption (seec)was cal-culated as a function of the applied cell voltage (U in volts)[27].seec (kWh /kg Al)=3FU3.6×10M Al φ(10)Solutions of reactive textile dyes (at a concentration of 50mg/l)were prepared by dissolution in de-ionised water of Drimarene (Clariant,Mutenz,Switzerland)K2LR CDG Blue,Drimarene Black and Yellow Procion Hexl (Dys-tar,Frankfurt,Germany).Synthetic textile wastewater was obtained by dissolution of hydrolysed reactive dyes (Dri-marene K2LR CDG Blue (2.78mg/l);Drimarene KG Orange (3mg/l);Drimarene K8B CDG Red (24.3mg/l)),hydrolysed starch (2.78mg/l),(NH 4)2SO 4(5.56mg/l)and Na 2HPO 4(5.56mg/l)in de-ionised water.Hydrolysis was performed by heating the solutions at 80◦C for 1.5h after adjustment at pH 12.Synthetic textile wastewater pH was adjusted at 10and conductivity at 4.7mS/cm.Two real textile wastewater samples from a knit cotton factory (Marrakech,Morocco)(global effluent),which is using the same type of dyes were also tested.Dyes concentrations were estimated from their absorbance characteristics in the UV–vis range (200–800nm).A SECO-MAM (Domont,France)Anthelie Light spectrophotometer connected to a PC was used.Each dye is associated to a main absorbance band at a characteristic wavelength:in the con-centration range used throughout this study,linear relations between dye concentration and absorbance at the dye charac-teristic wavelength were obtained.Dissolved organic carbon (DOC)measurements were carried out on an APOLLO 9000total organic carbon analyser after filtration (10␮m).Chem-ical oxygen demand (COD)was measured on a Hach 2400(Loveland,Colorado,USA)(Method 8000).pH and conduc-tivity were measured using respectively a PHM 220pHmeter and a CDM 210conductimeter (Radiometer Analytical SAS,Villeurbanne,France).To compare the efficiency of electro-coagulation with re-spect to wastewater biodegradability short-termrespirometryFig.2.Typical protocol of biodegradability analysis by respirometry.batch tests were performed [28].Activated sludge (1.6l )from the local domestic wastewater treatment plant was placed into the 2l-respirometer,equipped with an air-sparger and a mechanical stirrer.The dissolved oxygen concentration was monitored with Orbisphere (Marin,Switzerland)electrodes connected to a PC for data logging.Activated sludge was suc-cessively spiked with a sodium acetate solution,which acts as a reference carbon source,the wastewater sample to be tested and again the sodium acetate solution (Fig.2).The dissolved oxygen (S 0)mass balance in the liquid phase is given by:d S 0d t=K L a (S ∗0−S 0)−OUR exo −OUR end (11)where S ∗0is the dissolved oxygen concentration at saturation,K L a is the oxygen mass transfer coefficient,OUR end ,the oxygen uptake rate for slowly biodegradable substances and OUR exo the oxygen uptake rate for rapidly biodegradable substances.When recalcitrant and/or toxic substances (with respect to the activated sludge bacteria)are present in the wastewater sample,the oxygen uptake rates decrease.The respirograms were characterized by the maximal value taken by OUR (OUR max )and the volume of oxygen consumed for 20min after the injection of the wastewater sample.Fig.2presents a typical respirogram.3.Experimental designStatistical calculations were done using NEMRODW ver-sion 2000(LPRAI,Marseille,France).In a first stage,the effect of three electrochemical variables (current density,electrolysis time and initial pH of the solution)on the de-colourisation was investigated according to a 23full factorial experimental design with five replicates of the centre point.Before statistical analysis,the original variables X 1(elec-trolysis time),X 2(current density)and X 3(initial pH of the solution)were reduced.The levels for the three main vari-ables were chosen from previous know-how.The original and reduced levels are given in Table 1.464 A.Alinsafiet al./Chemical Engineering and Processing44(2005)461–470 Table1Original and reduced levelsParameters Originalvariable(X)Reduced variable(x)−10+1Electrolysis time(min)X16090120 pH X278.510 Current density(mA/cm2)X3102540 The choice of pH in the range7–10is on one hand due to the good efficiency of the electrocoagulation treatment at pH between4and9with aluminium electrodes[19,29,30]and on the other hand to the alkaline character of the real textile wastewater discharged on the industrial site related to the present study.The choice of an electrolysis time in the range 60–120min may appear large compared to what is described in literature.It is due mainly to the important sample volume (500ml)compared to the capacity of cell(100ml).In a second stage,the main objective was to select the current density and the electrolysis time in order to achieve optimal COD removal and decolourisation.We considered a22Doehlert matrix for investigating the effect of current density and electrolysis time,while the initial pH of the so-lution was set to10,as this value is close to the real textile wastewater pH.Uniform Doehlert networks were generated from a Simplex(Table2).4.Results and discussion4.1.Decolourisation kineticsColour is effectively removed by electro-coagulation,as depicted in Fig.3for a solution of Drimarene K2LR CDG Blue.A global decrease of the UV–vis absorbance is ob-served.Simultaneously DOC decreases.DOC abatement in this case is about46%and can reach up54%in some exper-iments.4.2.Statistical analysis and modellingThe objective is to provide a predictive model able to explain the influence of operational parameters on product quality.The dimension of the model is very much reduced compared to the dimension of the concerned data group.The results obtained for the23full factorial design are given in Table3for Drimarene K2LR CDG Blue,which decolourisa-tion yield varies between72and98%(average value92%). Table2The coordinates of the initial SimplexNumber X1(electrolysis time)X2(current density) 10021030.50.866Fig.3.Absorbance spectra(A)and kinetic curves of colour and dissolved organic carbon(B)of Drimarene K2LR CDG Blue dye vs.reaction time for an initial pH10and current density=10mA/cm2.The experimental error(2.8%)due to uncontrolled factors is calculated from the replicates of the centre point.The decolourisation yield(Y1)can be predicted by Eq.(12)using the coefficients listed in Table4.Y1=b0+b1X1+b2X2+b3X3+b12(X1X2)+b13(X1X3)+b23(X2X3)+b123(X1X2X3)(12) From the experimental values taken by the Student coeffi-cient t exp,an experimental significance level is calculated. b0,b1and b3are significant at a level less than5%,when the interaction coefficient b13is significant at a level less than 10%.Therefore the reaction time(coefficient b1),the cur-rent density(coefficient b3)and their interaction(coefficient b13)are the most influential factors.In fact,the decolourisa-tion is occurring by adsorption[25]on Al(OH)3(s)or on the monomeric anion Al(OH)4−depending upon the dye chemi-cal structure,which is unfortunately unknown for most com-mercial dyes.In the present case no adjustment is needed for pH:on the contrary to what is observed for the current density and the reaction time,its significance level is very high(85%), which means that pH is not an influential parameter.The ef-fects of operational variables on process performance may be explored with wastewater as discharged from the textile dyeing plant.The analysis of variance(Table5)shows thatA.Alinsafiet al./Chemical Engineering and Processing44(2005)461–470465 Table323full factorial design with5repetitions of the centre pointRun x1(electrolysis time)x2(initial pH)x3(current density)Y1(decolourisation)(%) 1−1−1−172.1621−1−197.833−11−180.58411−188.185−1−1194.1361−1194.647−11196.26811196.26900095.541000095.141100094.661200094.351300098.41Table4Model coefficientsCoefficient Value Standarddeviation t exp Significancelevel(%)b092.1651.28671.64<0.01b14.2231.6402.574.92b20.3151.6400.1984.9b35.3171.6403.242.31b12−2.3221.640−1.4221.5b13−4.0951.640−2.505.4b230.6231.6400.3871.8b1232.1951.6401.3423.8the predictive model insured a representativity of the experi-mental data of about92%,as the significance level calculated from the ratio of the mean square errors due to the regression and to the residues is7.7.4.3.Process optimisationThe surface response methodology helps to develop a sta-tistical model of a reaction by performing the minimum num-ber of well-chosen experiments and to determine the optimal values of process parameters.This approach is particularly suitable when a given variable depends on the settings of an-other one(interaction effects).This methodology is here ap-plied to the decolourisation yield and the abatement of COD. The considered factors are the current density in the cell and the electrolysis time.A22full factorial design with three replicates of the centre point(Table6)was applied.Figs.4and5show the contour plots obtained from the lin-ear models built from the experimental results.These plotsTable5Analysis of variance for decolourisation percentageSum of square errors Degrees offreedomMeansquareerrorRatio of meansquare errorsSignificancelevel(%)Regression588784 3.97.7 Residues107521Total69612Table6Doehlert matrixRun X1(electrolysistime)X2(currentdensity)Y1(decolourisation)Y2(CODremoval) 11097.7716.222−1094.79+16.22 30.50.86697.2311.364−0.5−0.86693.8027.27 50.5−0.86690.9836.366−0.50.86696.0713.64 70.094.6618.18 80094.3516.22 90098.4112.12Table7Coefficients of the predictive model of Y1(%decolourisation)Coefficient Value Standarddeviationt exp Significancelevel(%)b095.807 1.23877.36<0.01b10.717 1.2380.5860.5b22.460 1.2381.9914.0b110.473 1.9580.2481.7b22−1.873 1.958−0.9641.1b122.298 2.4770.9342.4were obtained by calculating the coefficients of Eq.(13) and drawing subsequently the contour for the electrolysis time—density current variable pair for Y1(Table7)and Y2 (Table8):Y=b0+b1X1+b2X2+b11X12+b22X22+b12X1X2(13)Table8Coefficients of the predictive model of Y2(%COD removal)Coefficient Value Standarddeviationt exp Significancelevel(%)b015.507 1.7278.980.212b11.135 1.7270.6656.0b2−11.152 1.727−6.460.622b110.713 2.7310.2680.4b228.631 2.7313.164.95b12−6.565 3.455−1.9015.3466 A.Alinsafiet al./Chemical Engineering and Processing 44(2005)461–470Fig.4.Two dimensional contour plot obtained from the experimental data of COD removal vs.electrolysis time (X 1)and current density (X 2).Fig.4shows that COD removal is maximal in the region where the electrolysis time is between 90and 120min and for a current density of about 12mA/cm 2.In fact,when the current density increases,there are an increase of the amount of aluminium dissolved in the liquid phase and an increase of the production of hydroxide Al(OH)3.For long electrolysis times,the structure of the sludge may change,altering the efficiency of pollution removal and the settle-ability and float-ability properties of the flocs.In Fig.5,the decolourisation yield is up to 98%in the global experimental plan:highvalues Fig.5.Two dimensional contour plot obtained from the experimental data of decolourisation yield vs.current (X 2)and electrolysis time (X 1).of both variables maximize the decolourisation yield due to the good adsorption of dye on sludge.The objective is to determine the optimal pair current density—electrolysis time to achieve both a good decolouri-sation yield and a good abatement of organic matter expressed as COD.Several scenarios can be examined in order to find the optimal conditions.The objectives for each scenario are transformed into functions d i ,which give the percentage of satisfaction obtained with respect to the fullfillness of the tar-gets that have been set in terms of decolourisation and CODA.Alinsafiet al./Chemical Engineering and Processing44(2005)461–470467Fig.6.Desirability function.removal(Fig.6).The values of d i are combined in a geometric mean:D=(d1×d2···d m)1/mwith m is the number of functions d i involved in the compu-tation of the total desirability D to be optimised.A minimal efficiency in terms of decolourisation(90%)and COD re-moval(30%)is required.Two scenarios were tested(Table9).In Scenario1,the target for decolourisation was set at95%and for COD re-moval at36%.In Scenario2,the target for decolourisation was increased but the target for COD was decreased.In both cases,COD removal targets are almost reached but colour removal cannot be achieved as efficiently as expected.To select between both scenarios,another criterion should be incorporated.It could be the cost of electrodes or the energy consumption.Energy consumption could be estimated by the product X1×X2.In the present case Scenario1was preferred for its lower energy requirement.The optimum is obtained for X1= 105min and X2=12mA/cm2that insures92%on decolouri-sation and35%on COD removal.It is characterised by an average desirability about55.3%and maximum desirability about88.3%as shown in Table10.In this case,the specific electrical energy consumption is about13.5kWh/kg Al and the apparent current efficiency is around1.1,slightly above1 as in the experiments of Jiang et al.[27],who explained such a discrepancy by the occurrence of secondary reactions.Table9Optimal sets for the two scenariosScenario1Scenario2Minimum Target Minimum Target Decolourisation(%)90959098 COD removal(%)30363035 Optimal electrolysistime(min)104.95105.68Optimal currentdensity(mA/cm2)1212.21Y1,theor91.7391.82Y2,theor35.2735Table10Characteristics of optimal pair(D=desirability)Response Value D(%)Weight D min(%)D max(%) Y1(decolourisation)91.753510.078Y2(COD removal)35.2387137.3100 Global desirability55.3088.3 4.4.Treatment efficiency at optimal conditionsThe optimal electro-coagulation conditions found in the previous section(electrolysis time=105min and current den-sity=12mA/cm2)were applied to other reactive dyes as well as synthetic and industrial textile wastewater samples. In industrial environment it is difficult to determine the opti-mal values of the operation parameters,as the characteristics of the wastewater are changing daily in function of the tex-tile production program.The results show that the electro-coagulation efficiency differs very much according to the dyes.For the blue dye the experimental responses Y1and Y2 (Table11)are close to the theoretical ones(Table9)ob-tained by the desirability study and validate the choice of the electrolysis time and the current density.For the two simple dye solutions(Yellow Procion and Drimarene Black)high COD removal yields were achieved,in spite of very different initial COD concentration.High decolouri-sation yields were also observed.In the case of the syn-thetic wastewater,COD removal was medium in spite of a low initial COD concentration.This synthetic wastewa-ter attempts to mimic the effect of textile dyeing,which is performed at a relatively high temperature,on the dyes and other additives(starch,salts).The two industrial wastewa-ter samples have higher initial COD than the other tested solutions.However the electro-coagulation remains very ef-ficient with respect to COD removal and decolourisation. Those two samples were collected on two successive days, giving another example of the high variability degree of tex-tile wastewater.In any case large variations of DOC are also observed.4.5.Biodegradability assessmentTo estimate the level of biodegradability of textile wastew-ater by the activated sludge before and after the electro-coagulation treatment,batch respirometry tests with activated sludge have been performed.Table12shows a summary of the estimated parame-ters.K L a and OUR end were calculated for each experi-ment.Depending upon the type of sample,the observed behaviour differed.The Drimarene Black sample was slightly biodegradable before electro-coagulation as shown in Fig.7a by the reduced oxygen uptake rate in spite of a high COD(255mg/l).Its biodegradability increased after electro-coagulation,which decreases also strongly its COD content. For Textile I wastewater much larger values of oxygen uptake468 A.Alinsafiet al./Chemical Engineering and Processing 44(2005)461–470Table 11Efficiency of treatment at optimal conditions (COD 0=initial solution COD)SampleDOC removal (%)Y 2(%COD removal)Y 1(%decolourisation)Drimarene Blue DR K2LR CDG (COD 0=154mg O 2/l)72.838.290.7Yellow Procion hexl (COD 0=95mg O 2/l)61.688.292.6Drimarene Black (COD 0=255mg O 2/l)18.681.394.3Real wastewater I (COD 0=450mg O 2/l)22.865.283.3a Real wastewater II (COD 0=620mg O 2/l)34.274.898.6a Synthetic wastewater (COD 0=70mg O 2/l)51.276.083.9aAverage value at three wavelength 620,525and 436nm.rates were observed before than after electro-coagulation (Fig.7b),although this treatment decreases strongly also the COD content.Finally in the case of Textile II wastewater,almost no change can be seen in the respirometry patterns (Fig.7c).Similar behaviours occurred for the yellow and the blue dyes solutions as well as for the syntheticwastewater.Fig.7.Oxygen uptake rate vs.time for (a)Drimarene Black solution,(b)Textile I wastewater and (c)Textile II wastewater,before and after electro-coagulation.To conclude about an eventual biodegradability improvement after pre-treatment the biodegradability yield is calculated as:Y B =(V O 2/COD)after electro-coagulation (V O 2/COD)before electro-coagulationA.Alinsafiet al./Chemical Engineering and Processing44(2005)461–470469 Table12Summary of the estimated parameters during respiromety testsBefore electro-coagulation After electro-coagulation Y BOUR exomax(mg/l/s)V O2(mg/l)OUR exomax(mg/l/s)V O2(mg/l)Yellow Procion hexl0.00523.00.0055 3.09 Drimarene K2LR CDG Blue0.00463.30.0043 2.51.5 Drimarene Black0.00372.0.0066 2.99.5 Synthetic textile wastewater0.00543.10.0054 3.24.4 Textile wastewater I0.01613.80.013 6.52 Textile wastewater II0.00950.009 4.24In all tested cases,an increase of the biodegradability was observed but,as for COD and colour removal efficiency,it depends very much upon the sample composition.5.ConclusionsElectro-coagulation is an efficient process,even at high pH,for the removal of colour and total organic carbon in reactive dyes textile wastewater.The efficiency of the pro-cess is influenced strongly by the current and the time of the reaction.Optimal electrolysis time and current density were de-termined to achieve a decolourisation yield between90and 95%and COD removal between30and36%for a reactive blue dye.Although these operational parameters were ap-plied to other reactive dyes solutions as well as synthetic and real textile wastewater and led to satisfactory colour removal and increase of biodegradability,the effect of the wastewater nature,which is highly time-variable in an industrial envi-ronment,could be pointed out.AcknowledgementsThe authors are thankful to the textile industry Tenmar and to the French-Moroccan Committee(project MA/02/49). They wish also to thank Professor J.P.Corriou for his helpful comments.References[1]E.G.Solozhenko,N.M.Soboleva,V.V.Goncharut,Decolourizationof azo dye solutions by Fenton’s oxidation,Water Res.29(1995) 2206–2210.[2]J.Weber,V.C.Stickney,Hydrolysis kinetics of reactive Blue19-vinylsulfone,Water Res.27(1993)63–67.[3]R.Camp,P.E.Sturrock,The identification of the derivatives ofCI reactive Blue19in textile wastewater,Water Res.24(1990) 1275–1278.[4]M.T.Kennedy,J.M.Morgan,L.K.Benefield,A.F.McFadden,Colorremoval from textile dye wastewater:a case study,in:Proceedings of the47th Ind.Waste Conference,West Lafayette,IN,Lewis Pub, Chelsea,MF,1992,pp.727–741.[5]Y.K.Park, C.H.Lee,Dyeing wastewater treatment by activatedsludge process with a polyurethanefluidised bed biofilm,Water Sci.Technol.34(5–6)(1996)193–200.[6]C.B.Shaw,C.M.Carliell,A.D.Wheatley,Anaerobic/aerobic treat-ment of coloured textile effluents using sequencing batch reactors, Water Res.36(2002)1193–2001.[7]D.Georgiou,P.Melidis,A.Aivasidis,Use of a microbial sensor:inhibition effect of azo-reactive dyes on activated sludge,Bioprocess Biosyst.Eng.25(2002)79–83.[8]T.Robinson,B.Chandran,P.Nigam,Removal of dyes from a syn-thetic textile dye effluent by biosorption on apple pomace and wheat straw,Water Res.(2002)36.[9]A.Pala,E.Tokat,Color removal from cotton textile industry wastew-ater in an activated sludge system with various additives,Water Res.36(2002)2920–2925.[10]S.F.Kang,H.M.Chang,Coagulation of textile secondary effluentswith Fenton’s reagent,Water Res.36(12)(1997)215–222. [11]M.P´e rez,F.Torrades,X.Dom`e nech,J.Peral,Fenton and photo-Fen-ton oxidation of textile effluents,Water Res.36(2002)2703–2710.[12]W.Chu,S.M.Tsui,Modeling of photodecoloration of azo dye in acocktail photolysis system,Water Res.36(2002)3350–3358. [13]B.Zielinska,J.Grzechuslka,A.W.Morawski,Photocatalytic decom-position of textile dyes on TiO2Tytanpol Al1and TiO2-Degussa P25, J.Photochem.Photobiol.A Chem.157(1)(2003)65–70.[14]B.Neppolian,S.Sakthivel,B.Arabindoo,V.Murugesan,Solar/UV-induced photocatalytic degradation of three commercial textile dyes, J.Hazard.Mater.89(2002)303–317.[15]C.Hachem,F.Bocquillon,O.Zahraa,M.Bouchy,Decolourizationof textile ionudstry wastewater byt the photocatalytic degradation process,Dyes Pigments49(2001)117–125.[16]M.V.B.Zanoni,J.Sene,M.A.Anderson,Photoelectrocatalytic degra-dation of Remazol Brilliant Orange3R on titanium dioxide thin-film electrodes,J.Photochem.Photobiol.A Chem.157(1)(2003)55–63.[17]E.Dobolyi,Experiments aimed at the removal of phosphate by elec-trochemical methods,Water Res.12(1978)1113–1116.[18]L.M.Balmer,A.W.Foulds,Separation oil from oil-in water emul-sions by electroflocculation/electroflotation,Filtr.Sep.23(1986) 366–369.[19]J.S.Do,M.L.Chen,Decolorization of dye-containing solutions byelectrocoagulation,J.Appl.Electrochem.24(1994)785–790. [20]S.H.Lin, C.F.Peng,Continuous treatment of textile wastewaterby combined coagulation–electrochemical oxidation and activated sludge,Water Res.30(1996)587–592.[21]X.M.Chen,G.H.Chen,P.L.Yue,Separation of polluants fromrestaurant wastewater by electrocoagulation,Sep.Purif.Technol.19 (2000)65–76.[22]X.M.Chen,G.H.Chen,P.L.Yue,Electrocoagulation and electroflo-tation of restaurant wastewater,J.Environ.Eng.126(2000)858–863.[23]T.Picard,G.Cathalifaud-Feuillade,M.Mazet,C.Vandensteendam,Cathodic dissolution in the electrocoagulation process using alu-minium electrodes,J.Environ.Monit.2(2000)77–80.[24]A.G¨u rses,M.Yalc¸in,C.Do˘g an,Electrocoagulation of some reactivedyes:a statistical investigation of some electrochemical variables, Waste Manage.22(2002)491–499.[25]M.Rebhun,M.Lurie,Control of organic matter by coagulation andfloc separation,Water Sci.Technol.27(1993)1–20.。