Electrochemical degradation of crystal violet with BDD electrodes: Effect of electrochemical parameters and identification of organic by-products

Chemosphere 81 (2010) 26–32

Contents lists available at ScienceDirect

Chemosphere journal homepage: www.elsevier.com/locate/chemosphere

Electrochemical degradation of crystal violet with BDD electrodes: Effect of electrochemical parameters and identification of organic by-products Ricardo E. Palma-Goyes a,b, Fernando L. Guzmán-Duque a, Gustavo Peñuela b, Ignacio González c, Jose L. Nava d, Ricardo A. Torres-Palma a,b,* a

Grupo de electroquímica, Instituto de química, Facultad de ciencias exactas y naturales, Universidad de Antioquia, A.A. 1226, Medellín, Colombia Grupo de diagnóstico y control de la contaminación, Facultad de ingeniería, Universidad de Antioquia, A.A. 1226, Medellín, Colombia Universidad Autónoma Metropolitana-Iztapalapa, Departamento de Química, Av. San Rafael Atlixco No. 186, C.P. 09340, México D.F., Mexico d Universidad de Guanajuato, Departamento de Ingeniería Geomática e Hidráulica, Av. Juárez No. 77, C.P. 36000, Guanajuato, Mexico b c

a r t i c l e

i n f o

Article history: Received 24 February 2010 Received in revised form 12 July 2010 Accepted 14 July 2010 Available online 14 August 2010 Keywords: Electrochemical oxidation Triphenylmethane dye Crystal violet BDD electrode Water treatment Dye treatment

a b s t r a c t This paper explores the applicability of electrochemical oxidation on a triphenylmethane dye compound model, hexamethylpararosaniline chloride (or crystal violet, CV), using BDD anodes. The effect of the important electrochemical parameters: current density (2.5–15 mA cm2), dye concentration (33– 600 mg L1), sodium sulphate concentration (7.1–50.0 g L1) and initial pH (3–11) on the efficiency of the electrochemical process was evaluated. The results indicated that while the current density was lower than the limiting current density, no side products (hydrogen peroxide, peroxodisulphate, ozone and chlorinated oxidizing compounds) were generated and the degradation, through OH radical attack, occurred with high efficiency. Analysis of intermediates using GC–MS investigation identified several products: N-methylaniline, N,N-dimethylaniline, 4-methyl-N,N-dimethylaniline, 4-methyl-N-methylaniline, 4dimethylaminophenol, 4-dimethylaminobenzoic acid, 4-(N,N-dimethylamino)-40 -(N0 ,N0 -dimethylamino) diphenylmethane, 4-(4-dimethylaminophenyl)-N,N-dimethylaniline, 4-(N,N-dimethylamino)-40 -(N0 ,N0 dimethylamino) benzophenone. The presence of these aromatic structures showed that the main CV degradation pathway is related to the reaction of CV with the OH radical. Under optimal conditions, practically 100% of the initial substrate and COD were eliminated in approximately 35 min of electrolysis; indicating that the early CV by-products were completely degraded by the electrochemical system. Ó 2010 Elsevier Ltd. All rights reserved.

1. Introduction Triphenylmethane dye compounds, such as crystal violet (CV), are largely used in the paper, leather, cosmetic, and food industries. The textile industry consumes large amounts of these substances for dyeing nylon, wool, cotton, and silk, as well as for colouring oil, fats, waxes, varnish and plastics (Gessner and Mayer, 2001). Furthermore, some of the triphenylmethane dyes are used as medicine and biological stains (Azmi et al., 1998). These compounds not only cause coloration of water, but also pose a serious risk to aquatic life (Nassar and Magdy, 1997) and their presence in drinking water constitutes a possible human health hazard (Pielesz, 1999).

* Corresponding author at: Grupo de electroquímica, Instituto de química, Facultad de ciencias exactas y naturales, Universidad de Antioquia, A.A. 1226, Medellín, Colombia. Tel.: +57 4 219 56 00; fax: +57 4 219 56 66. E-mail addresses: [email protected], [email protected] (R.A. Torres-Palma). 0045-6535/$ - see front matter Ó 2010 Elsevier Ltd. All rights reserved. doi:10.1016/j.chemosphere.2010.07.020

Biological processes are the most economical option to eliminate organic pollutants. However, these methods cannot be applied to many textile wastewaters due to the toxicity of commercial dyes against the microorganisms used in the processes (Robinson et al., 2001). Physicochemical methods, based on the production and use of hydroxyl radical called Advanced Oxidation Processes (AOPs) (e.g. H2O2/UV, UV/O3, H2O2/O3, TiO2 photocatalysis, Fenton’s reagent, photo-Fenton), have been successfully tested for elimination of this kind of compounds in water (Ruppert et al., 1994; Azbar et al., 2004). These processes are all based on the generation and use of a powerful oxidant, the OH radical. However, because of the hydroxyl radical scavenger effect, they have limited applications in waters containing appreciable quantities of inorganic ions. Guillard et al. (2005) demonstrated how the TiO2 photocatalytic efficiency, under acidic conditions, decreased due to the OH radical 2 3  scavenging effect by anions such as NO 3 , Cl , SO4 , PO4 . Siedlecka et al. (2007) studied the effect of selected inorganic anions on the effectiveness of the Fenton advanced oxidative treatment of waters  contaminated with methyl t-butyl ether (MTBE). SO2 and 4 , Cl  H2 PO led to competition between the organics and the OH radicals, 4

R.E. Palma-Goyes et al. / Chemosphere 81 (2010) 26–32

which retarded the MTBE oxidation. Carbonate, chloride and sulphate ions were also found to inhibit the removal of methylene blue through the TiO2 photocatalyst (Lee et al., 1999). Additionally,  HCO 3 ions have been also reported to consume OH radicals effectively, preventing organic compounds from being attacked by OH in processes such as H2O2/UV (Qiao et al., 2005), O3/UV, O3/TiO2/ UV, and O3/V–O/TiO2 (Tong et al., 2005), O3/H2O2 (Acero and Gunten, 2000) and sonolysis/O3 (Olson and Barbier, 1994). Interestingly, the presence of ions in water increases conductivity and therefore wastewater applications based on electrochemical methods are commonly favored by their presence. Electrochemical methods are promising and versatile alternatives that have the potential to replace or complete already existing processes (Brillas et al., 1998, 2000). They make the treatment of liquids, gases and solids possible and are amenable to automation and compatible with the environment because their main reagent, the electron, is clean (Jüttner et al., 2000), offering an alternative solution to many environmental problems in the processing industry (Panizza and Cerisola, 2009). Recently, electrocoagulation has been effectively applied to the degradation of CV in water (Durango-Usuga et al., 2010). However, this electrochemical process only transfers pollutants from an aqueous phase to a solid one. Thus, the final disposal of the generated sludge becomes a major problem. On the contrary, electrochemical oxidation has the prospective to completely mineralize organic pollutants (Brillas et al., 2005; Nava et al., 2008b; Martínez-Huitle and Brillas, 2009) or transform them into biodegradable by-products that can be eliminated in a subsequent biological step (Torres et al., 2003). The use of a synthetic boron-doped diamond thin film electrode (BDD) in anodic oxidation has shown that O2 overvoltage is much higher than that for conventional anodes such as PbO2, doped SnO2 or IrO2, producing larger amounts of adsorbed OH through the reaction (1), and thereby resulting in a faster and more effective destruction of pollutants. Therefore, anodic oxidation with BDD electrodes seems to be a suitable procedure to mineralize organics.

BDD þ H2 O ! BDDð OHÞ þ Hþ þ e

ð1Þ

Sanromán et al. (2004) studied the electrochemical treatment of CV on Pt anodes and reported the suitability of the technology to decolorize wastewaters from dyeing processes containing this pollutant. However, neither evaluation of the oxidation degree (COD), nor identification of intermediates generated during the process was reported. Because the electrochemical oxidation may lead to the formation of substances which are less biodegradable or more toxic than those in the untreated waters (Torres et al., 2003), the applicability of this technology to effectively oxidize this pollutant cannot be established. The aim of this paper is to study the applicability of electrochemical oxidation on a triphenylmethane dye compound model, hexamethylpararosaniline chloride or CV (C25H30ClN3), using BDD anodes. It also evaluates the effect of important electrochemical parameters (current density, dye and supporting electrolyte concentration and pH) on the efficiency of the electrochemical process. In addition, the evolution of COD, the main aromatic by-products generated and the degradation pathways are investigated under optimal conditions. 2. Material and methods

27

2.2. Electrochemical cells Electrochemical experiments for CV oxidation were carried out in a 150 mL, one-compartment electrolytic cell. The anode was a 4 cm2 BDD electrode (provided by Metakem™, with a thickness of 2–7 lm supported on Ti expanded metal) in contact with an aqueous solution of the substrate in Na2SO4 under constant stirring conditions. The cathode was a zirconium spiral electrode. In order to perform reproducible electrochemical treatments, solutions containing the pollutant were placed into the electrolytic cell after conditioning the anodes for 15 min. The electrolysis was performed under controlled current density (2.5–15 mA cm2). As needed, the electrolysis medium was made alkaline by the addition of aqueous NaOH, or acidic by adding aqueous H2SO4. The electrolytic cell was sampled periodically for UV visible, Chemical Oxygen Demand (COD) and Gas Chromatography coupled to Mass Spectrometry (GC/MS) analysis. On the other hand, cyclic voltammetry experiments were carried out in a single-compartment three-electrode cell (70 mL) using an Autolab PGSTAT 30. The counter electrode was a Pt wire, the reference electrode was Hg/Hg2SO4/K2SO4 (SSE), and the BDD was the working electrode. 2.3. Analysis The Zahn–Wellens Test (OECD, 1981) was used with a high bacterial concentration of 1 g L1. Sludge from the San Fernando wastewater treatment facility in Medellin (Colombia) was aerated for 24 h and centrifuged. The Zahn–Wellens Test was run with a 1 g L1 suspension of centrifuge sludge. Correspondence between the extent of CV removal as measured by spectrophotometry and HPLC techniques was evaluated. No significant differences in CV evolution were observed. Thus, due to economical and environmental reasons and the rapidness of measuring, quantitative analysis of CV was done by UV visible in a Janway, Model 6405 UV/Vis, spectrophotometer set at 590 nm. Chemical oxygen demand, COD, was determined by the method reported by Thomas and Mazas (1986), using a dichromate solution as the oxidizer in a strong acid medium. Test solution (2.5 mL) was transferred into the dichromate reagent and digested at 150 °C for 2 h. Optical density for colour change of the dichromate solution was determined at 445 nm with the spectrophotometer. Evolution of side products during the electrochemical methods (hydrogen peroxide, peroxodisulfate, ozone and hypochlorous acid) were determined by iodometry (Kormann et al., 1988): aliquots taken from the reactor were immediately added to the sample in the quartz cuvette of the spectrophotometer (Janway, Model 6405 UV/Vis) containing the reagent (0.1 M potassium iodide and 0.01 M ammonium heptamolybdate), and the absorbance at 2 min was recorded. After pre-concentration of the sample on a Chromabond HLB cartridge, identification of primary CV intermediates was made by GC/MS through an Agilent 10220 series MSD with electrospray ionization following the method proposed by Chen and Lu (2007). The mass transfer coefficient, km, was determined by linear sweep voltammetry (Autolab PGSTAT 30), using an equimolar ferri/ferrocyanide redox couple (20–80 mM) (protocol adapted from Michaud, 2002).

2.1. Reagents

3. Results and discussion

CV was purchased from Aldrich. Anhydrous sodium sulphate, sulphuric acid and sodium hydroxide were provided by Merck. Distilled water was used throughout for the preparation of aqueous solutions.

3.1. Biodegradability assessment of a CV solution In order to evaluate biodegradability of CV, the Zahn–Wellens test (with a high bacterial concentration of 1 g L1) was carried

R.E. Palma-Goyes et al. / Chemosphere 81 (2010) 26–32

out on a solution of 250 mg L1 of the dye (see Supplementary material (SM), Fig. SM-1). In a control experiment, bacteria removed 95% of phenol after 18 d. On the contrary, the CV concentration remained almost unchanged after 28 d. These results indicate the non-biodegradability of this effluent, even under such favourable conditions for biodegradation as neutral pH, high aeration, presence of nutrients and room temperature. 3.2. Characterization of CV electro-oxidation on BDD anodes Fig. 1 shows a typical electrochemical degradation, using BDD anodes, of CV in water solutions containing sodium sulphate. CV concentration decreases with time; after 23 min (0.15 Ah L1), 50% of the initial substrate concentration (137 mg L1) is removed (Fig. 1). As has been previously reported (Guinea et al., 2009; Hamza et al., 2009) and is indicated in Fig. 1, besides the substrate elimination, the high oxidation power of BDD allows the concurrent formation of several side products: hydrogen peroxide, peroxodisulphate, ozone and chlorinated oxidizing compounds (Fig. 1). Hydrogen peroxide comes from the recombination of the hydroxyl radical (reaction (2)) while, peroxodisulphate is formed by the oxidation of sulphate (reaction (3)) and ozone is generated from water oxidation (reaction (4)) (Guinea et al., 2009):

2BDDð OHÞ ! 2BDD þ H2 O2

ð2Þ

2  2SO2 4 ! S2 O8 þ 2e

ð3Þ

3H2 O ! O3ðgÞ þ 6Hþ þ 6e

ð4Þ

The small chloride concentration (0.34 mM), from the pollutant model formulation, can slowly oxidize to form chlorine and then hypochlorous acid, HOCl, (Torres et al., 2003; Panizza and Cerisola, 2005; Hamza et al., 2009). 

2Cl ! Cl2 þ 2e

ð5Þ 

Cl2 þ H2 O ! HOCl þ H þ Cl

ð6Þ



HOCl $ Hþ þ OCl

ð7Þ



The HOCl/OCl pair reacts with organic compounds by addition, substitution or oxidation (Boyce and Hornig, 1983). Even if all of the oxidizing species electro-generated can also degrade the CV,

Time (min) 0

5

10

15

20

25

1

140

0.9

120

C/Co

0.8

80 0.7 60 0.6

40

0.5

20

0.4 0

0.02

0.04

0.06

0.08

0.1

0.12

0.14

3.3.1. Effect of current density One of the parameters that most influences the electrochemical degradation of organic compounds is the current density. In order to investigate the effect of this parameter on the electro-oxidation of the substrate, Fig. 2 shows the CV (137 mg L1) removal at four different current densities (2.5, 5.0, 7.5, and 15 mA cm2). After 540 C L1 (0.15 Ah L1) the percentages of CV removed were 50%, 38%, 35% and 26%, respectively. This behaviour can be explained by considering the formation and accumulation of side products during the process (inset in Fig. 2). Analysis of the inset in Fig. 2 indicates that the lower the formation and accumulation of these species, the higher the degradation rate of CV. Indeed, after 540 C L1 (0.15 Ah L1), the amount of auxiliary oxidants generated were 10, 17, 42 and 60 lM at 2.5, 5.0, 7.5 and 15 mA cm2, respectively. According to Panizza et al. (2001), at the beginning of the 0 electrolysis, t = 0, the initial limiting current density (ilim , A m2) of the electrochemical mineralization of organics can be calculated as follows:

70

[Side products] (μΜ)

100

3.3. Effect of classical parameter on the electrochemical degradation of CV

0 0.16

Electrical charge (Ah L-1)  Fig. 1. Crystal violet and side products (H2O2, S2 O2 8 , O3 and ClO ) evolution during the oxidation of an aqueous solution containing 137 mg L1 of crystal violet and 50 g L1 Na2SO4 at initial pH 6.4. The electrolysis was performed at imposed current density (15 mA cm2) on a BDD electrode (4 cm2). (d) crystal violet; (+) side products in the absence of crystal violet and (s) side products in the presence of crystal violet.

[Side products] (µM)

þ

and/or its degradation by-products, the oxidation power of them is lower than that of the OH radicals (Butro´n et al., 2007). The presence of CV decreases the rate of the oxidizing species formation (side products, Fig. 1). This is because the reaction of hydroxyl radicals with CV competes with reaction (2). Thus, a lower amount of hydrogen peroxide is produced. Direct electrochemical oxidation of CV on the BDD surface could also inhibit reactions (3)–(5) and diminish the formation of other oxidizing species. In order to investigate direct electro-degradation of CV, cyclic voltammetry experiments were carried out (see Supplementary material, Fig. SM-2). A single irreversible anodic peak with peak potential Eap 1.34 V vs. SSE was found for CV oxidation on the BDD electrode. This result states that CV can be directly electrooxidized on BDD surfaces. However, in this work experiments were carried out at relatively high voltages, where OH radicals are formed from water discharge at the anode (reaction (1)). Thus, because of the relative concentrations of water and CV, oxidation of water to produce OH radicals predominates instead of the direct oxidation of CV at the electrode surface. Therefore, the OH radical attack is the main route of CV degradation. Recently, Hamza et al. (2009) reported similar results during the electrochemical degradation of another triphenylmethane dye, the Methyl Violet 6B, on BDD electrodes.

1.2

1

60 50 40 30 20 10

C/Co

28

0 0

0.8

0.02 0.04 0.06 00.8 0.1 0.12 0.14 0.16 Ah L-1

0.6

0.4 0

0.02

0.04

0.06

0.08

0.1

0.12

0.14

0.16

Electrical charge (Ah L-1) Fig. 2. Influence of the current density applied during the electrochemical degradation of crystal violet on BDD anodes. The initial solution contains 137 mg L1 crystal violet and 50 g L1 Na2SO4 at initial pH 6.4. Inset panel: Side oxidation products evolution during the electrolysis. Current density applied: (h) 15 mA cm2; (+) 7.5 mA cm2; (s) 5 mA cm2 and (d) 2.5 mA cm2.

29

R.E. Palma-Goyes et al. / Chemosphere 81 (2010) 26–32 0

ilim ¼ 4Fkm COD0

ð8Þ

where F is the Faraday constant (96 487 C mol1), km is the mass transport coefficient (2  105 m s1) and COD0 is the initial chemical oxygen demand of the solution (mol m3). Taking into account 0 the initial pollutant concentration (137 mg L1) the ilim value of the 2 electrochemical mineralization of CV is 7.8 mA cm . The inset in Fig. 2 also shows that while the current density is lower than the limiting current density, e.g. the initial step of electro-oxidation experiments among 2.5 and 7.5 mA cm2, no side products are generated and the degradation occurs with high efficiency. In other words, when the applied current is higher than the limiting current density, such as when 15 mA cm2 (inset in Fig. 2) is applied, the formation of side products (reactions (2)–(5)) hampers the generation and availability of the powerful oxidant OH radical. Thus, when the applied current is higher than the limiting current, the efficiency of electrochemical degradation of CV is unfavoured and evolution of the side products predominates.

6

6

5 4

5

3 4

2 3

1 2 200

300

400

500

600

15

20

25

30

35

1

COD

0.6

0.4 CV 0.2

0 0

0.5

1

1.5

2

2.5

Electrical charge (Ah L-1)

700

Fig. 3. Initial crystal violet degradation rate (s) and side oxidation products formation rates () for different crystal violet concentrations (33–600 mg L1) during electrochemical treatment on BDD electrodes. Experimental conditions: 50 g L1 Na2SO4, 22.5 mA cm2, initial pH 3.0.

140 120 100 80 60 40

Fig. 5. Crystal violet concentration and chemical oxygen demand evolution during the electrochemical degradation of crystal violet (250 mg L1) on BDD electrodes under the optimal conditions selected in this work: initial pH 6.4, imposed current 2.5 mA cm2 and electrolyte: 35.5 g L1 Na2SO4.

Normalized CV and by-products concentrations

100

10

0.8

[CV] (mg L-1)

[CV] (mg

5

0 0

L-1)

Time (h) 0

C/Co

7

Side products formation rate (µM min-1)

CV degradation rate (µM min-1)

3.3.2. Effect of the initial CV concentration The initial decomposition rate of CV at various concentrations ranging between 33 and 600 mg L1 is shown in Fig. 3. The curve proves that the higher the substrate concentration, the higher the initial decomposition rate (Fig. 3 empty circles). 33 and 150 mg L1 of CV were degraded at 0.86 and 2.00 mg L1 min1, respectively.

In spite of that, there is not much difference observed in the degradation rate of CV when increasing the concentrations from 300 to 600 mg L1 (degradation rates of 2.49 and 2.78 mg L1 min1, respectively). Thus, a linear relationship is not observed as expected for a first-order kinetic law. As is evident from Eq. (8), the limiting current density depends on the initial COD and therefore on the initial concentration of the substrate. Except for the highest concentration tested, the current density applied to the system (22.5 mA cm2) was above the limiting current density; therefore the electrolysis was under mass 0 0 transport control (iappl > ilim ) and evolution of the side products took place. However, as expected, the initial rate of the side products formation associated with CV treatment decreased with an increase in the organic target concentration (Fig. 3 full pics). In the absence of substrate, oxidizing species were produced at a rate of 6.8 lM min1, whereas this rate was practically zero for the 600 mg L1 CV concentration. Degradation of organic compounds in water by BDD electrodes has been the subject of several reports (Iniesta et al., 2001; Brillas et al., 2005; Panizza and Cerisola, 2005; Nava et al., 2008b; Hamza et al., 2009). Reaction rates and chemical transformation pathways

1.2 1 0.8 A, D, E

0.6 F

0.4

I

G

0.2

H

C

0 0

20

B

CV

40

80

120

160

Time (min)

0 0

5

10

15

20

25

Time (min) Fig. 4. Effect of initial pH on the electrochemical degradation of crystal violet on BDD electrodes under imposed current density (22.5 mA cm2). The initial solution contains 137 mg L1 crystal violet and 50 g L1 Na2SO4. The initial pH is (s) 3; () 6 and (d) 11.

Fig. 6. Crystal violet by-products evolution during electrochemical treatment on BDD anodes at 15 mA cm2. Experimental conditions: 137 mg L1 of CV, initial pH 6.4, and 50 g L1 Na2SO4. N-methylaniline (A), N,N-dimethylaniline (B), 4-methylN,N-dimethylaniline (C), 4-methyl-N-methylaniline (D), 4-dimethylaminophenol (E), 4-dimethylaminobenzoic acid (F), 4-(N,N-dimethylamino)-40 -(N0 ,N0 -dimethylamino) diphenylmethane (G), 4-(4-dimethylaminophenyl)-N,N-dimethylaniline (H), 4-(N,N-dimethylamino)-40 -(N0 ,N0 -dimethylamino) benzophenone (I).

30

R.E. Palma-Goyes et al. / Chemosphere 81 (2010) 26–32

ated with first-order kinetics. Consequently, in this work, the comparison between different tests was performed using the degradation extent (% degradation) or the initial degradation rate (mg L1 min1) rather than the pseudo first-order kinetic constant.

depend on the molecular structure of the pollutant, current density, electrode composition and geometry, and electrochemical reactor design, among other factors (Nava et al., 2008a). Furthermore, the presence or electrochemical generation of supplementary oxidants such as hydrogen peroxide, chlorinated active species, peroxodisulphate or ozone, can also contribute to the kinetics of the electrochemical process. Thus, electrochemical CV elimination is associated with the local concentrations of OH radicals, supplementary oxidants and substrate. It is important to mention that all of the electrolyses presented herein were developed in an undivided cell, therefore the degradation of CV may also involve reactions at the cathode. The complex degradation pathway of CV gives nonlinear kinetics. Therefore, it cannot be associ-

+

3.3.3. Effect of supporting electrolyte concentration and initial pH The type and concentration of the supporting electrolyte are also important parameters during the electrochemical treatment of water. The presence of electrolytes increases the conductivity, diminishes the resistance and thus, lessens the energy cost of the process. In several cases, electrochemistry of the electrolytes plays an important role in the process because of the possible formation of side products that can increase or diminish the efficiency of the

CH3

CH3

N

N CH3

H 3C

CH3

CH3

N

N

H3C

CH3

N H 3C

CH3

G

CH3 N CH3

H3C

HO O

F

CH 3

CH 3

N

N

H 3C

N

CH3

CH 3

CH3

I

O

HN

C

CH3

CH3

H 3C N

N

H 3C

CH3

CH3

H

D

H 3C N H 3C

B -CH3

NH CH3

H3C

N

CH3

A

OH

E

Fig. 7. Main reaction pathways during the electrochemical degradation of crystal violet on BDD anodes. The name of detected by-products is given in Fig. 6.

R.E. Palma-Goyes et al. / Chemosphere 81 (2010) 26–32

system. Chloride and sulphate salts are two of the most used electrolytes for electrochemical treatment. The use of chloride salts is controversial because of the susceptible formation of chlorinated organic compounds (Chiang et al., 2000; Lei et al., 2007). Thus, sodium sulphate was selected as the supporting electrolyte in this work. No significant differences were observed when the supporting electrolyte concentration studied was 7.1, 35.5 or 50.0 g L1 (data not shown). In all cases, 20–25% of CV was eliminated after 720 C L1 (0.2 Ah L1) with initial degradation rates of 1.2– 1.6 mg L1 min1, respectively. However, when 7.1 g L1 Na2SO4 were used, higher potential values were required to carry out the experiments. Additionally, no appreciable differences were found between electrochemical tests done with 35.5 g L1 or 50 g L1 of sodium sulphate (data not shown). As a result, 35.5 g L1 Na2SO4 was the supporting electrolyte concentration selected in this work. The effect of the initial pH during the electrochemical degradation of CV was also investigated (Fig. 4). Three different initial pH values were tested: acid (pH 3), natural (pH 6.4) and alkaline (pH 11). Under acidic and natural initial pH values, the evolution of the curves follows a similar trend with time, reaching in both cases, 37% of CV removal in 20 min (720 C L1 or 0.2 Ah L1) with a degradation rate of 2.6 mg L1 min1. A higher degradation rate (6.8 mg L1 min1) was observed for experiments at pH 11. In fact, 90% CV was removed after 20 min of treatment. As is well-known, this could be associated with the chemical instability of CV under alkaline pH (Arias-Estevez et al., 2008). A control experiment (data not shown) indicated that the chemical transformation of CV at pH 11, in the absence of current, does not produce significant changes in the chemical oxygen demand, showing that no appreciable oxidation of CV takes place. Additionally, during the electrochemical treatment, the COD removal showed to be independent of the initial solution pH. Thus, in order to avoid the additional step and cost of adjusting pH, natural pH was selected for the optimal experiments.

31

control (15 mA cm2) using 137 mg L1 of CV, natural pH (6.4) and 50 g L1 Na2SO4 as supporting electrolyte. Nine by-products were found (see Supplementary material, Table SM-1): N-methylaniline (A), N,N-dimethylaniline (B), 4-methyl-N,N-dimethylaniline (C), 4methyl-N-methylaniline (D), 4-dimethylaminophenol (E), 4-dimethylaminobenzoic acid (F), 4-(N,N-dimethylamino)-40 -(N0 ,N0 dimethylamino) diphenylmethane (G), 4-(4-dimethylaminophenyl)-N,N-dimethylaniline (H), 4-(N,N-dimethylamino)-40 -(N0 ,N0 dimethylamino) benzophenone (I). Compounds A, B, F, E, and I were also identified by Fan et al. (2009) during the degradation of CV by Fenton. Likewise, the by-products B, F and H were also reported by Oturan et al. (2008) during the electro-Fenton degradation of the triphenylmethane dye compound, malachite green. These results confirm that during the electrochemical treatment on BDD anodes, CV is mainly degraded by OH radical attack. The evolution of the CV by-products during the electrochemical treatment is shown in Fig. 6. As suggested in this figure and presented in Fig. 7, first the OH radical attacks on the central carbon portion of CV yields F, G and I. Successive attacks of this radical on the central carbon of G produce C, E, I and H. Compound D is generated by the OH attacks on the N,N-dimethyl group (N-demethylation) of C. The OH attacks on the 4-methyl group of compounds (C) also produces E. Reaction of OH radicals with the ketonic group of I generates B, F and H. Compounds B and E are also formed by  OH attacks on H. B is transformed into A and E by N-demethylation and OH attack on the para position of the N,N-dimethyl group, respectively. ~ izares et al., 2004; Li et al., 2005; Flox et al., Several papers (Can 2007) have indicated that further electrochemical oxidation of simple aromatic rings first leads to the formation of phenolic by-products (such as compound E), which are transformed into hydroquinone and benzoquinone. Later, the ring opening leads to the formation of carboxylic acids which finally oxidize to CO2. 4. Conclusions

3.4. Study of the CV degradation under optimal electrochemical conditions 1

The evolution of CV concentration and COD of a 250 mg L CV solution were determined to evaluate the pollutant elimination and oxidation of the organic matter, respectively. The experiment was achieved at natural pH (6.4), 2.5 mA cm2 and 35.5 g L1 Na2SO4 as supporting electrolyte. Fig. 5 reveals that CV concentration decreases and CV is totally removed after 8100 C L1 (2.25 Ah L1). The profile of the COD evolution is similar to the pollutant removal and practically all of the initial COD is eliminated after 2.25 Ah L1. No evolution of the side products was observed during the first 6 h of the process (data not shown). This indicates that, under working conditions, no collateral reactions take place. It is important to point out that moderate low current densities, as shown in Fig. 5 (2.5 mA cm2), favour the complete oxidation of organics via hydroxyl radicals formed by the oxidation of water at the BDD-solution interface. These results prove that, under current control, electrochemical treatment using BDD anodes is a powerful method to eliminate and completely oxidize dye compounds such as CV. This is in accordance with that suggested by Nava et al. (2008a) during the mineralization of p-cresol, o-cresol and indigo on BDD. 3.5. Identification of the main by-products and degradation pathway during the electrochemical treatment of CV In order to investigate the main degradation pathway of CV by anodic oxidation on BDD anodes, identification of the main byproducts formed during the treatment was done by GC/MS in positive electrospray mode. The study was carried out under diffusion

This study shows that CV electrochemical degradation on BDD electrodes is strongly affected by current density, initial substrate concentration and sodium sulphate concentration. Electrolysis studies indicated that the oxidation of CV was carried out via hydroxyl radicals formed by the oxidation of water at the BDD-solution interface with galvanostatic conditions of 2.5 mA cm2 in 35.5 g L1 Na2SO4 and neutral initial pH. Under these conditions, the process was able to completely oxidize the initial substrate, with high efficiency, without the formation and accumulation of side products. Appendix A. Supplementary material Supplementary data associated with this article can be found, in the online version, at doi:10.1016/j.chemosphere.2010.07.020. References Acero, J.L., Gunten, U.V., 2000. Influence of carbonate on the ozone/hydrogen peroxide based advanced oxidation process for drinking water treatment. Ozone Sci. Eng. 22, 305–328. Arias-Estevez, M., Astray, G., Cid, A., Fernandez-Gandara, D., Garcia-Rio, L., Mejuto, J.C., 2008. Influence of colloid suspensions of humic acids upon the alkaline fading of carbocations. J. Phys. Org. Chem. 21, 555–560. Azbar, N., Yonar, T., Kestioglu, K., 2004. Comparison of various advanced oxidation processes and chemical treatment methods for COD and color removal from a polyester and acetate fiber dyeing effluent. Chemosphere 55, 35–43. Azmi, W., Sani, R.K., Banerjee, U.C., 1998. Biodegradation of triphenylmethane dyes. Enzyme Microb. Technol. 22, 185–191. Boyce, S.D., Hornig, J.F., 1983. Reaction pathways of trihalomethane formation from the halogenation of dihydroxyaromatic model compounds for humic acid. Environ. Sci. Technol. 17, 202–211.

32

R.E. Palma-Goyes et al. / Chemosphere 81 (2010) 26–32

Brillas, E., Mur, E., Sauleda, R., Sánchez, L., Peral, J., Domènech, X., Casado, J., 1998. Aniline mineralization by AOP’s: anodic oxidation, photocatalysis, electroFenton and photoelectro-Fenton processes. Appl. Catal. B: Environ. 16, 31–42. Brillas, E., Calpe, J.C., Casado, J., 2000. Mineralization of 2,4-D by advanced electrochemical oxidation processes. Water Res. 34, 2253–2262. Brillas, E., Sirés, I., Arias, C., Cabot, P.L., Centellas, F., Rodríguez, R.M., Garrido, J.A., 2005. Mineralization of paracetamol in aqueous medium by anodic oxidation with a boron-doped diamond electrode. Chemosphere 58, 399–406. Butro´n, E., Jua´rez, M.E., Solis, M., Teutli, M., Gonza´lez, I., Nava, J.L., 2007. Electrochemical incineration of indigo textile dye in filter-press-type FM01-LC electrochemical cell using BDD electrodes. Electrochim. Acta 52, 6888–6894. ~ Canizares, P., Sa´ez, C., Lobato, J., Rodrigo, M.A., 2004. Electrochemical treatment of 4-nitrophenol-containing aqueous wastes using boron-doped diamond anodes. Ind. Eng. Chem. Res. 43, 1944–1951. Chen, C., Lu, C., 2007. Photocatalytic degradation of basic violet 4: degradation efficiency, product distribution, and mechanisms. J. Phys. Chem. C 111, 13922– 13932. Chiang, L.C., Chang, J.E., Wen, T.C.N., 2000. Destruction of refractory humic acid by electromechanical oxidation process. Water Sci. Technol. 42, 225–232. Durango-Usuga, P., Guzman, F., Mosteo, R., Vazquez, M., Peñuela, G., Torres-Palma, R., 2010. Experimental design approach applied to the elimination of crystal violet in water by electrocoagulation with Fe or Al electrodes. J. Hazard. Mater. 179, 120–126. Fan, H., Huang, S., Chung, W., Jan, J., Lin, W., Chen, Ch., 2009. Degradation pathways of crystal violet by Fenton and Fenton-like systems: condition optimization and intermediate separation and identification. J. Hazard. Mater. 171, 1032–1044. Flox, C., Cabot, P.L., Centellas, F., Garrido, J.A., Rodríguez, R.M., Arias, C., Brillas, E., 2007. Solar photoelectro-Fenton degradation of cresols using a flow reactor with a boron-doped diamond anode. Appl. Catal. B: Environ. 75, 17–28. Gessner, T., Mayer, U., 2001. Triarylmethane and diarylmethane dyes, sixth ed.. Ullmann’s Encyclopedia of Industrial Chemistry Wiley-VCH, New York. pp. 27– 179. Guillard, C., Puzenat, E., Lachheb, H., Houas, A., Herrmann, J.M., 2005. Why inorganic salts decrease the TiO2 photocatalytic efficiency. Int. J. Photoenergy 7, 1–9. Guinea, E., Centellas, F., Garrido, J.A., Rodriguez, R.M., Arias, C., Cabot, P.L., Brillas, E., 2009. Solar photoassisted anodic oxidation of carboxylic acids in presence of Fe3+ using a boron-doped diamond electrode. Appl. Catal. B: Environ. 89, 459– 468. Hamza, M., Abdelhedi, R., Brillas, E., Sirés, I., 2009. Comparative electrochemical degradation of the triphenylmethane dye Methyl Violet with boron-doped diamond and Pt anodes. J. Electroanal. Chem. 627, 41–50. Iniesta, J., Michaud, P.A., Panizza, M., Cerisola, G., Aldaz, A., Comninellis, Ch., 2001. Electrochemical oxidation of phenol at boron-doped diamond electrode. Electrochim. Acta 46, 3573–3578. Jüttner, K., Galla, U., Schmieder, H., 2000. Electrochemical approaches to environmental problems in the process industry. Electrochim. Acta 45, 2575– 2594. Kormann, C., Bahnemann, D.W., Hoffmann, M.R., 1988. Photocatalytic production of H2O2 and organic peroxides in aqueous suspensions of TiO2, ZnO, and desert sand. Environ. Sci. Technol. 22, 798–806. Lee, B.N., Liaw, W.D., Lou, J.C., 1999. Photocatalytic decolorization of methylene blue in aqueous TiO2 suspensions. Environ. Eng. Sci. 16, 165–175. Lei, Y., Shen, Z., Huang, R., Wang, W., 2007. Treatment of landfill leachate by combined aged-refuse bioreactor and electro-oxidation. Water Res. 41, 2417– 2426.

Li, X., Cui, Y., Feng, Y., Xie, Z., Gu, J., 2005. Reaction pathways and mechanisms of the electrochemical degradation of phenol on different electrodes. Water Res. 39, 1972–1981. Martínez-Huitle, C.A., Brillas, E., 2009. Decontamination of wastewaters containing synthetic organic dyes by electrochemical methods: a general review. Appl. Catal. B: Environ. 87, 105–145. Michaud, P.A., 2002. Comportement Anodique du Diamant Synthétique Dopé au Bore. PhD thesis, EPFL, No. 2595. Nassar, M.M., Magdy, Y.H., 1997. Removal of different basic dyes from aqueous solutions by adsorption on palm-fruit bunch particles. Chem. Eng. J. 66, 223– 227. Nava, J.L., Butrón, E., González, I., 2008a. Importance of hydrodynamic conditions on the electrochemical incineration of cresols, indigo textile dye and vinasses present in industrial wastewater using filter-press type FM01-LC reactor with BDD electrodes. J. Environ. Eng. Manage. 18, 221–230. Nava, J.L., Recéndiz, A., Acosta, J.C., González, I., 2008b. Electrochemical incineration of vinasse in filter-press-type FM01-LC reactor using 3D BDD electrode. Water Sci. Technol. 58, 2413–2419. OECD, 1981. Guidelines for Testing Chemicals, Reg 329 B. Olson, T.M., Barbier, P.F., 1994. Oxidation kinetics of natural organic matter by sonolysis and ozone. Water Res. 28, 1383–1391. Oturan, M.A., Guivarch, E., Oturan, N., Sirés, I., 2008. Oxidation pathways of malachite green by Fe3+ catalyzed electro-Fenton process. Appl. Catal. B: Environ. 82, 244–254. Panizza, M., Cerisola, G., 2005. Application of diamond electrodes to electrochemical processes. Electrochim. Acta 51, 191–199. Panizza, M., Cerisola, G., 2009. Direct and mediated anodic oxidation of organic pollutants. Chem. Rev. 109, 6541–6569. Panizza, M., Michaud, P.A., Cerisola, G., Comninellis, Ch., 2001. Anodic oxidation of 2-naphthol at boron-doped diamond electrodes. J. Electroanal. Chem. 507, 206– 214. Pielesz, A., 1999. The process of the reduction of azo dyes used in dyeing textiles on the basis of infrared spectroscopy analysis. J. Mol. Struct. 511, 337–344. Qiao, R.P., Ma, Y.M., Qi, X.H., Li, N., Jin, X.C., Wang, Q.S., Zhuang, Y.Y., 2005. Degradation of microcystin-RR by combination of UV/H2O2 technique. Chin. Chem. Lett. 16, 1271–1274. Robinson, T., McMullan, G., Marchant, R., Nigam, P., 2001. Remediation of dyes in textile effluent: a critical review on current treatment technologies with a proposed alternative. Bioresour. Technol. 77, 247–255. Ruppert, G., Bauer, R., Heisler, G., 1994. UV–O3, UV–H2O2, UV–TiO2 and the photoFenton reaction comparison of advanced oxidation processes for wastewater treatment. Chemosphere 28, 1447–1454. Sanromán, M.A., Pazos, M., Ricart, M.T., Cameselle, C., 2004. Electrochemical decolourisation of structurally different dyes. Chemosphere 57, 233–239. Siedlecka, E.M., Wieckowska, A., Stepnowski, P., 2007. Influence of inorganic ions on MTBE degradation by Fenton’s reagent. J. Hazard. Mater. 147, 497–502. Thomas, O., Mazas, N., 1986. La mesure de la demande chimique en oxygène dans les milieux faiblement pollués. Analusis 14, 300–302. Tong, S.-P., Xie, D.-M., Wei, H., Liu, W.-P., 2005. Degradation of sulfosalicylic acid by O3/UV O3/TiO2/UV, and O3/V–O/TiO2: a comparative study. Ozone Sci. Eng. 27, 233–238. Torres, R.A., Sarria, V., Torres, W., Péringer, P., Pulgarín, C., 2003. Electrochemical treatment of industrial wastewater containing 5-amino-6-methyl-2benzimidazolone: toward an electrochemical-biological coupling. Water Res. 37, 3118–3124.