journal of electrochemical science and engineering...issn: 1847-9286 open access journal journal of...

ISSN: 1847-9286Open Access Journalwww.jese-online.org

Journal of Electrochemical

Science and Engineering

J. Electrochem. Sci. Eng. 2(4) 2012, 155-236

Volume 2 (2012) No. 04 pp. 155-236

IAPC

J. Electrochem. Sci. Eng. 2(4) (2012) 155-236 Published: November 10, 2012

Open Access : : ISSN 1847-9286

www.jESE-online.org

Contents

SVETLOZAR IVANOV, IOANNA MINTSOULI, JENIA GEORGIEVA, STEPHAN ARMYANOV, EUGENIA VALOVA, GEORGIOS KOKKINIDIS, IOANNIS POULIOS, SOTIRIS SOTIROPOULOS Platinized titanium dioxide electrodes for methanol oxidation and photo-oxidation Dedicated to 75th birthday of Professor Sergio Trasatti ................................................................................................. 155

LUO JIANCHENG, YANG JIE, LI WEISHAN, HUANG QIMING, XU HONGKANG Electrochemical degradation of Reactive Brilliant Red K-2BP on Ti/RuTiIrSnMn oxide anode in a batch cell .............................................................................................................................................. 171

JOHN GUSTAVSSON, GONGZHUO LI, CHRISTINE HUMMELGÅRD, JOAKIM BÄCKSTRÖM, ANN CORNELL On the suppression of cathodic hypochlorite reduction by electrolyte additions of molybdate and chromate ions ....................................................................................................................................... 185

ALBANA VESELI, AHMET HAJRIZI, TAHIR ARBNESHI, KURT KALCHER A new amperometric glucose biosensor based on screen printed carbon electrodes with rhenium(IV) - oxide as a mediator ......................................................................................................... 199

BOULBABA ELADEB, CAROLINE BONNET, ERIC FAVRE and FRANÇOIS LAPICQUE Electrochemical extraction of oxygen using PEM electrolysis technology .............................................. 211

FADHIL M. MOHAMMED, EDWARD P. L. ROBERTS, ANDREW K. CAMPEN, NIGEL W. BROWN Wastewater treatment by multi-stage batch adsorption and electrochemical regeneration .................. 223

doi: 10.5599/jese.2012.0020 155

J. Electrochem. Sci. Eng. 2 (2012) 155-169; doi: 10.5599/jese.2012.0020


www.jESE-online.org Original scientific paper

Platinized titanium dioxide electrodes for methanol oxidation and photo-oxidation§ SVETLOZAR IVANOV, IOANNA MINTSOULI*, JENIA GEORGIEVA, STEPHAN ARMYANOV, EUGENIA VALOVA, GEORGIOS KOKKINIDIS*, IOANNIS POULIOS*, SOTIRIS SOTIROPOULOS*

Rostislaw Kaischew Institute of Physical Chemistry, Bulgarian Academy of Sciences, Sofia 1113, Bulgaria

*Department of Chemistry, Aristotle University of Thessaloniki, Thessaloniki 54124, Greece Corresponding Author: E-mail: [email protected]; Tel.: +30-2310-997742; Fax: +30-2310-997709

Received: September 3, 2012; Published: November 10, 2012 § Dedicated to 75th birthday of Professor Sergio Trasatti

Abstract Platinized deposits have been formed on TiO2 particulate films supported on Ti substrates, by means of galvanic replacement of pre-deposited metallic Cu and subsequent immersion of the Cu/TiO2 coatings into a chloroplatinic acid solution. The spontaneous replacement of Cu by Pt results in Pt(Cu)/TiO2/Ti electrodes. Both the platinized and the precursor TiO2/Ti electrodes have been characterized by SEM micro-scopy/EDS spectroscopy, their surface electrochemistry has been assessed by cyclic voltammetry in the dark and their photoelectrochemical properties by photovoltam-metry under UV illumination. It has been found that, although platinized rutile-rich electrodes exhibit typical Pt surface electrochemistry, the anatase-rich electrodes show only traces of oxide formation and stripping. The latter has been translated to a suppression of methanol oxidation at anatase-rich electrodes. On the contrary, methanol oxidation at platinized rutile-rich electrodes occurs at significant rates and can be further enhanced upon UV illumination, as a result of Pt and TiO2 synergism in the photoelectrochemical oxidation of methanol.

Keywords Titanium dioxide; platinum; galvanic replacement; electrocatalysis; photoelectrocatalysis

J. Electrochem. Sci. Eng. 2(4) (2012) 155-169 PLATINIZED TiO2 FOR METHANOL OXIDATION

156

Introduction

Platinized Ti or TiO2 finds a number of practical applications that include industrial anodes, photocatalysis/photoelectrocatalysis and fuel cell catalysts. In more detail, Ti (with native, thermally or electrochemically grown TiO2 on it) is the basis of dimensionally stable anodes (DSAs) which are coated with precious metal and other oxide catalysts and are used in many industrial processes (e.g. chlorine production, water electrolysis, electrowinning etc) [1-4]. Also, modifying the semiconductor oxides of TiO2 by precious metals (especially Pt) is a well-established practice to improve their photocatalytic efficiency by reducing photogenerated electron-hole recombination rates [5-7]. Finally, Pt/TiO2 bi-component powder catalysts mixed with high surface area carbons have recently been tested as alternative anode or cathode catalysts in fuel cells [8-10].

Methods for Pt deposition/incorporation onto/into TiO2 supported layers include electro-deposition [11-14], vacuum deposition [15], thermal decomposition followed by reduction [16] and photodeposition [17,18]. Platinization of TiO2 powders is almost exclusively carried out by chemical reduction of Pt complexes in the presence of TiO2 particles or simultaneously with TiO2 sol-gel preparation [19]. Electrodeposition allows for accurate control of precious metal loading but requires specialized equipment, uses concentrated precious metal solutions and it is difficult to apply to powder substrates; chemical methods usually involve reducing agents and almost always involve a high temperature annealing in a reducing atmosphere; photodeposition is a lengthy process and can also lead to precious metal losses due to uncontrolled deposition at reactor components other than the substrate [17].

During the last decade an alternative method for the introduction of the noble metal onto the electrode support has been developed. The method is based on the spontaneous galvanic replacement of surface layers of a non-noble metal M (M: Pb, Cu, Fe, Co, Ni) by a noble metal (e.g. Pt, Au, Pd, Ir) upon immersion of the former in a solution of metal ions of the latter (the resulting bimetallic system is denoted as Pt(M), Au(M) etc.). For example, in the case of Cu and Pt:

2 Cu + PtCl62- → Pt + 6 Cl- + 2 Cu2+ (1).

The method (also known as transmetallation) was first applied by Adžić and co-workers to underpotentially deposited (UPD) Cu monolayers [20-22] and to Cu and Pb bulk deposits by Kokkinidis and co-workers [23-24] on flat electrode substrates. It has since been extended to other transition metals (Fe, Co, Ni) (see for example the works of Sotiropoulos and co-workers [25-30]) as well as to carbon powder supports (see for example [31,32]). Advantages of the new technique include the fact that it is a fast and room temperature processes, it employs low concentration solutions of the precious metal and can lead to the formation of thin precious metal deposits that may decrease its loading.

Despite the rapid evolution of the application of the technique on conducting substrates, the formation of metal/semiconductor (e.g. Pt/TiO2) catalysts by means of transmetallation has yet to be established. These systems should offer the possibility of enhancing anodic reactions occurring at Pt by means of simultaneous photooxidations at n-type semiconducting supports and to that direction, methanol oxidation at UV-illuminated Pt/TiO2 electrodes has recently been reported [33].

The aim of this work has been to investigate the feasibility of applying the galvanic replacement technique to the metallization of TiO2 powders. Specific objectives have been: i. The preparation and microscopic as well as compositional characterization of Pt(Cu)/TiO2

deposits on Ti substrates ;

S. Ivanov et al. J. Electrochem. Sci. Eng. 2(4) (2012) 155-169

doi: 10.5599/jese.2012.0020 157

ii. The electrochemical characterization of the deposits by means of Pt surface electrochemistry and methanol oxidation;

iii. The study of the effect of UV illumination on the electrochemical behavior of these deposits.

Experimental

Preparation of TiO2/Ti, Cu/TiO2/Ti and Pt(Cu)/TiO2/Ti electrodes

Ti rectangular substrates (0.5 mm thick) had typical dimensions of 1 cm × 0.5 cm and were etched in a HF/HNO3 3:1 mixture and then washed with doubly distilled water. Known volumes of a 1 g L-1 methanol dispersion of Degussa P-25 TiO2 were pipetted onto one side of the Ti substrate and the coating was dried at 50°C for 10 min. TiO2 loadings were in the 2-3 mg cm-2 range. These coated substrates were sintered in an air atmosphere at 700°C (for 5 h) or 500°C (for 1.5 h) using a Carbolite CWF 1100 oven, resulting in particulate TiO2 layers. The back side of the samples had been etched again before it was insulated with epoxy resin glue (RS).

Electrodeposition of Cu on the TiO2/Ti electrode was performed at constant potential (in the mixed control potential regime and more specifically at the half-wave potential of exploratory deposition voltammetry) from 0.1 M HClO4 + 0.01 M CuSO4 deaerated solutions. The total charge density passed was 400 mC cm-2 for rutile-rich electrodes and 400 or 1200 mC cm-2 for anatase-rich electrodes.

The Cu/TiO2/Ti electrodes were immersed in 2 ml of a 0.1 M HCl + 10-3 M K2PtCl6 solution for 30 min so that replacement of Cu by Pt could take place:

2 Cu/TiO2/Ti + PtCl62-

→ Pt (Cu)/ TiO2/Ti + 2 Cu2+

+ 6 Cl− (2)

Microscopic and spectroscopic characterisation of coatings

Scanning Electron Microscopy (SEM) was carried out using a JEOL JSM-5510 microscope and elemental analysis of the coatings was performed by the accompanying EDS (EDAX) system. X-Ray Diffraction (XRD) deposit characterisation was performed with the help of a Rigakou Miniflex diffractometer.

Electrochemical and photoelectrochemical characterisation of coatings

Cyclic voltammetric and constant potential experiments on TiO2 electrodes in the dark and under UV illumination were carried out with the Autolab 30 (EcoChimie) system in a three-electrode cell equipped with a flat quartz window opposite the working electrode. A saturated calomel electrode (SCE) was used as the reference electrode and a Pt foil as the counter electrode. Voltammograms were run for at least three consecutive full cycles since preliminary experiments showed that a near-steady state response was observed only after the second run; all results reported correspond to the stabilized voltammetric picture.

A Radium Ralutec 9W/78 UVA lamp (λ=350-400 nm, λmax=366 nm), placed at a distance of 2.5 cm from the sample, was used for front face electrode illumination. The power density on the sample surface position was measured as 3 mW cm-2 with a photometer.

Electrode materials and chemicals

Ti plates 0.5 mm thick were from Alfa Aesar (99.5 %, metals basis). TiO2 powder was P-25 Degussa® from Degussa (now Evonic). HClO4 from Riedel, (puriss p.a., ACS reagent, ≥ 70 %) and CuSO4.6H2O from Sigma-Aldrich (ACS reagent) were used in the preparation of Cu deposition solutions. H2PtCl6 hexahydrate from Sigma-Aldrich (ACS reagent, ≥37.50 % as Pt) was employed for


158

the Pt exchange solution. MeOH was from Riedel (Chromasolv, for HPLC, gradient grade, ≥ 99.9 %).

Results and Discussion

TiO2 particulate films on Ti electrode substrates: microscopic, electrochemical and photoelectrochemical characterization

Figure 1 shows the SEM micrograph of the surface of a typical Ti/TiO2 (700 °C) electrode. This is characterized by a reticulated network of micrometer-sized aggregates into which the much smaller anatase and rutile particles (22 and 31 nm, as determined by XRD) of the original P-25 TiO2 material have been merged after annealing. The picture of Ti/TiO2 (500 °C) electrodes was very similar, indicating that sintering at 500° (for 1.5 h) or 700 °C (for 5h) has the same effect on the overall morphology of the particulate films. However, as confirmed by XRD [34], the 500 °C-treated samples retain the 75÷25 % w/w anatase-to-rutile composition (and are termed anatase-rich or TiO2-500 samples hereafter) while the 700°C-treated samples are transformed to rutile-rich samples (3÷97 % w/w anatase-to-rutile; TiO2-700).

Figure 1. SEM micrograph of a TiO2 particulate film prepared on a Ti substrate by annealing

solution-cast Degussa P-25® TiO2 particles at 700 °C for 5h. (Scale bar length: 30 μm.)

Electrochemical characterisation of TiO2 coatings in the dark

Figure 2 presents fast (100 mV s-1) cyclic voltammograms in the dark of an anatase-rich/500 °C electrode (with a 4.3 mg cm-2 TiO2 loading) and a rutile-rich/700 °C electrode (with a 2.4 mg cm-2 TiO2 loading), in a 1 M HClO4 solution. Prior to hydrogen evolution, a couple of cathodic and anodic main peaks are observed (together with traces of another couple at more positive potentials), attributed to the surface transformation of the Ti(IV)/Ti(III) couple at potentials negative to the flat band potential of TiO2 and just prior to hydrogen evolution [17,35-37].

The charge density (per geometric substrate area) under the cathodic peaks was found to be 7.5 mC cm-2 for the anatase-rich/500°C electrode and 2.8 mC cm-2 for the rutile-rich/700°C one. Since the value of Ti surface density in a TiO2 crystal has been reported as approximately 9 Ti4+ ions /nm2 [38], the Ti(IV)/Ti(III) transformation translates to an expected charge density of 144 μC cm-2. Then the sintered electrodes presented in Figure 2 have roughness factors of 52 and 19 electroactive TiO2 cm2 per substrate cm2. If we further take into account the TiO2 loading of these particular electrodes (4.3 mg cm-2 and 2.4 mg cm-2 respectively) then their mass-specific TiO2


doi: 10.5599/jese.2012.0020 159

electroactive area can be calculated as 1.2 m2 g-1 and 0.8 m2 g-1 for the anatase-rich and rutile-rich electrodes, respectively. The similar values of electroactive areas are in line with the similarities observed in the morphology by SEM analysis.

Figure 2. Cyclic voltammograms of TiO2/Ti electrodes (anatase-rich/annealed at 500 °C for

1.5 h and rutile-rich/ annealed at 700 °C for 5h) in 1 M HClO4, in the dark.

Figure 3(A) shows photovoltammograms (at 10 mV s-1) of two TiO2/Ti electrodes with similar

loadings (2.2-2.4 mg cm-2 ) at anatase- and rutile-rich TiO2 electrodes under UV illumination (360 nm, 3 mW cm-2). If there are no other oxidizable species in solution, the observed photocurrent is due to water oxidation by photogenerated holes to OH• or other primary oxidation products such as O2

•-and H2O2. The limiting photocurrent observed at high positive potentials is controlled by the diffusion of photogenerated electrons through the particles of the particulate semiconductor layer [39-41]. The onset potential of the photocurrent (which can be approximated to the flat band potential) for the anatase-rich electrode is ca. -0.150 V vs. SCE, lower by some 200 mV from that of the rutile-rich electrode, in line with anatase reported to have a more negative flatband potential than that of rutile [42]. Also, the limiting photocurrent in the plain electrolyte solution is found to be higher at the rutile-rich electrode, indicating its higher photoelectrocatalytic activity for water oxidation. Figure 3(B) shows the effect of the addition of 0.5 M MeOH to the photovoltammetry. It is seen that, although the shift of the flatband potential remains the same, the limiting photocurrent of the anatase-rich electrode is now higher, most likely due a higher adsorption affinity for methanol on the anatase surface where it is oxidized by the photogenerated holes.


160

Figure 3. Cyclic voltammograms of TiO2/Ti electrodes (anatase-rich/annealed at 500 °C for 1.5 h and rutile-rich/ annealed at 700 °C for 5h) in the dark and under continuous (or occasionally

shut) UV illumination, in (A) 1 M HClO4 and in (B) 0.5 M MeOH + 1 M HClO4 solutions. (Arrows indicate potential scan direction; the photocurrent fall for the TiO2-500 electrode

corresponds to a temporary illumination shut down.)

Cu electrodeposition on TiO2/Ti electrodes: deposition voltammetry, microscopic (SEM) and spectroscopic (EDS) characterisation of Cu/TiO2 precursors

Figure 4 presents the cyclic voltammograms of the TiO2/Ti electrodes in a Cu++ containing solution to obtain an exploratory electrodeposition picture. The cathodic electrodeposition peaks are observed for both types of electrodes (at potentials corresponding to the onset of the Ti(IV)/Ti(III) transformation shown in Figure 2) but interestingly no stripping peaks are recorded throughout the positive potential range. This means that the Cu/TiO2 interface behaves as a Schottky diode with dark current (including that of electrodeposition) starting to flow only at


doi: 10.5599/jese.2012.0020 161

potentials negative enough for TiO2 metallization and/or the onset of Ti surface reactions to occur. The fact that the onset of Cu electrodeposition should be related to the Ti(IV)/Ti(III) transformation (possibly via a catalytic cycle) is also supported by both phenomena exhibiting the same trend when one passes from an anatase-rich to a rutile-rich electrode.

Figure 4. Cyclic voltammograms of TiO2/Ti electrodes (anatase-rich/annealed at 500 °C for 1.5

h and rutile-rich/ annealed at 700 °C for 5h) in the dark, in the electrodeposition solution of 0.01 M CuSO4 + 1 M HClO4 solution. (Arrows indicate potential scan direction.)

Constant potential Cu electrodeposition was carried out at the half-wave cathodic peak potential for each type of electrode and (in initial experiments) the electrodeposited Cu charge density was 400 mC cm-2 for both electrodes. Figure 5 presents the SEM images (at different resolution) of such a Cu/TiO2/Ti rutile-rich electrode. Numerous Cu particles of sub-micron dimen-sions can be seen to have been deposited onto/into the reticulated matrix of TiO2. A similar pictu-re has been obtained for anatase-rich electrodes and, for typical TiO2-loadings of 2.2-2.4 mg cm-2 This procedure led to Cu/TiO2 coating having a 17-18 % w/w Cu content (w.r.t. TiO2, as confirmed by EDS).

Figure 5. SEM micrographs of a Cu/TiO2 particulate film prepared by electrodeposition of 400 mC cm-2

on a TiO2/Ti electrode (rutile-rich/ annealed at 700 °C for 5h). (Scale bar lengths: (A) 30 μm and (B) 7.5 μm.)


162

Galvanic replacement of Cu in Cu/TiO2/Ti electrodes by Pt: microscopic, spectroscopic and electrochemical/photoelectrochemical characterization of Pt(Cu)/TiO2/Ti electrodes in acid

Upon immersion of the Cu/TiO2 electrodes prepared as described in the previous section in a 10-3 M chloroplatinate solution for 30 min, Cu was replaced according to (2) by Pt. Figures 6 show SEM images of the thus produced Pt(Cu)/TiO2 coatings, revealing the existence of much smaller Pt nanoparticles (compared to those of the Cu precusrsor shown in Figures 5) and an increased porosity (intermediate of that of the Cu/TiO2 precursor and of the TiO2 starting material), due to significant etching of Cu into the solution.

Figure 6. SEM micrographs of a Pt(Cu)/TiO2 particulate film prepared from the Cu/TiO2 film

shown in Figure 4 above by immersion in a 10-3 M chloroplatinic acid + 0.1 M HCl solution for 30 min. (Scale bar lengths: (A) 30 μm and (B) 7.5 μm.)

The EDS analysis of the Pt(Cu)/TiO2 deposits indicates that almost all Cu has been replaced or etched (only traces of Cu were measured) and that their Pt content is 6% w/w (w.r.t. TiO2) for rutile-rich electrodes and 2% w/w for anatase-rich electrodes. An insight into the mechanism of Pt deposition and the differences observed for the two substrates can be acquired by the following reasoning. The absence of significant quantities of Cu suggests that only Pt deposited directly onto TiO2 locations adjacent to Cu deposits survives (whereas Pt deposited on Cu locations collapses, as layers beneath it corrode from uncovered locations). Since it has already been reported that Pt nuclei are preferably formed on rutile (which has a larger concentration of oxygen vacancies) [43], it follows that larger quantities of Pt are deposited directly onto rutile-rich substrates than on anatase-rich ones. To be able to produce similar Pt loadings on both types of substrates, larger quantities of Cu had to be deposited on anatase-rich substrates. A 3-fold increase in pre-deposited Cu (plating density of 1200 mC cm-2) resulted in a 5 % w/w Pt content. The rutile-rich and anatase-rich electrodes for which results are shown hereafter had Pt loadings of 0.15 mg cm-2 and 0.12 mg cm-2, respectively.

Figures 7 present the surface electrochemistry of the Pt(Cu)/TiO2/Ti electrodes in deaerated acid.


doi: 10.5599/jese.2012.0020 163

Figure 7. Cyclic voltammograms of Pt(Cu)/TiO2/Ti electrodes in the dark, in a deaerated

1 M HClO4 solution. (A) Rutile-rich electrode/annealed at 700 °C for 5 h; (B) Anatase-rich electrode/annealed at 500 °C for 1.5 h.

Although the rutile-rich electrode shows the typical polycrystalline Pt picture (with the hydrogen adsorption/desorption and oxide formation/stripping peaks), there are only traces of oxide formation and stripping for the anatase-rich electrode. In this case, the formation of a Schottky diode between the metal (Pt) and the n-type semiconductor (TiO2) which would allow for currents at potentials negative to the flat band potential and cut them off at more positive potentials, is not a plausible scenario. This is because, as shown in Figures 8, hydrogen evolution occurs at Pt(Cu)/TiO2 at potentials far more positive than the TiO2 metallization region;


164

furthermore some anodic reactions (ferrocyanide oxidation and hydrazine-not shown here) did take place on the anatase-rich electrodes in the potential range of Pt oxide formation and stripping.

Figure 8. Slow potential scan voltammograms of Pt(Cu)/TiO2/Ti and TiO2/Ti electrodes in the dark, in a deaerated 1 M HClO4 solution. (A) Rutile-rich electrode/annealed at 700 °C for 5 h;

(B) Anatase-rich electrode/annealed at 500 °C for 1.5h.

All this means that electron transfer reactions are possible at the Pt overlayer via resonance tunneling from Pt to the Ti base of the electrode through the TiO2 semiconductor, with the help of an impurity band introduced by Pt [44]. Therefore, the supression of Pt oxide/formation is to be attributed to strong Pt-anatase interactions. It should be noted that the extraordinary picture of Figure 7(B) has also been reported for platinized reduced anatase [45], monocrystalline rutile [46] or low Pt coverage/small particle size anatase-rich samples [47], and can been interpreted via the formation of a strong Pt-Ti bond [48], inhibiting the formation of PtO. The reason why these interactions are higher for anatase substrates is due to the fact that smaller nuclei (thus more affected by the TiO2 in contact) and stronger Pt adsorption are observed at anatase [43] (despite the formation of larger Pt clusters at rutile locations). From the charge under the cathodic hydrogen adsorption peak the Pt electroactive area per electrode substrate geometric area


doi: 10.5599/jese.2012.0020 165

(roughness factor) can be calculated as 10 for rutile-rich and 81 for anatase-rich electrodes, close or within the lower part of the 50-1000 range of Pt black or carbon-supported Pt fuel cell catalysts [17]. Taking into account the loading of each electrode, these values of the roughness factor are translated to ca 6.5 and 67.5 m2 g-1 of electroactive Pt (note that a typical fuel cell catalyst has a value of 65 m2 g-1 [49]). The higher surface area/dispersion of Pt in anatase-rich electrodes is in line with the above-mentioned reasoning for stronger Pt-TiO2 interactions at these substrates.

Figures 9 (A)-(B) show the effect of UV illumination on the cyclic voltammetry of the Pt(Cu)/TiO2/Ti electrodes in acid. A comparison with the results of the corresponding TiO2/Ti electrodes of Figure 3(A), reveals that there is approximately a 3-fold decrease of the net photocurrent at both electrodes (from 116 to 40 μA cm-2 for the rutile-rich electrode and from 61 to 22 μA cm-2 for the anatase-rich electrode, at +0.80 V).

Figure 9. Cyclic voltammograms of TiO2/Ti electrodes in the dark and under UV illumination, in 1 M HClO4 . (A) Rutile-rich electrode/annealed at 700 °C for 5 h; (B) Anatase-rich

electrode/annealed at 500 °C for 1.5h.


166

This is opposite to the beneficial effect that Pt has on plain photocatalysis but it is to be expected since first, Pt loading of the photoelectrocatalysts prepared in this work (8-10 % w/w) is much higher than that of Pt-doped TiO2 photocatalysts (0.1-1 % w/w) resulting to partial shielding of TiO2 from light and second, the addition of a metal should not have a key role in decreasing charge recombination during photoelectrocatalysis since this is mainly achieved by the electrical bias.

Methanol oxidation at Pt(Cu)/TiO2/Ti electrodes in the dark and under UV illumination

Figure 10 presents slow potential sweep cyclic voltammograms (at 10 mV s-1; third cycle) of the Pt(Cu)/TiO2/Ti electrodes, in a 0.5 M MeOH + 0.1 M HClO4 solution. Although the typical voltam-metric picture for methanol oxidation on Pt is observed (an oxidation peak showing distinct hysterisis), the current densities are rather low and the anatase-rich electrode is almost ten times inferior to the rutile-rich electrode. The latter trend can be interpreted by the absence of significant Pt oxide formation at the anatase-rich electrode (see Figure 7(B)) and the crucial role of these oxides in oxidizing/removing the carbonaceous intermediates/poisons of methanol oxidation. Note that a similar correlation has been found for methanol oxidation at Pt/TiO2 electrodes in [47].

Figure 10. Cyclic voltammograms of Pt(Cu)/TiO2/Ti electrodes (anatase-rich/annealed at 500°C for 1.5 h and rutile-rich/ annealed at 700 °C for 5h) in the dark, in a 0.5 M MeOH + 0.1 M HClO4

solution. (Arrows indicate potential scan direction.)

The Pt mass-specific current density of the rutile-rich Pt(Cu)/TiO2 catalyst can be estimated as 3.6 mA mg-1 (at the peak potential) which is lower than commonly observed values at Pt/C catalysts (see for example the value of ca 15 mA mg-1 of a commercial Pt/C(E-Tek) catalyst under same methanol concentration and similar scan rate conditions [50]). This is to be expected for the rutile-rich electrode which (based on Pt surface electrochemistry) was found to have a small Pt electroactive surface area (see discussion above). However, when compared to the performance of the Pt/TiO2 catalyst prepared by photodeposition in [17] and resulting in a massive 10 mg cm-2 Pt loading and/or losses (depending on the amounts deposited onto the substrate or the reactor


doi: 10.5599/jese.2012.0020 167

walls), our catalyst (with a 0.15 mg cm-2 Pt loading) is superior. This is because the photodeposited catalyst of [17] gives methanol oxidation currents in the 1.5-3 mA mg-1 range (after correction for different concentrations and depending on scan rate).

Figure 11 shows the effect of UV illumination on methanol oxidation at the rutile-rich electrode. The enhancement factor of the peak current is calculated as 1.66 and the net increase is some 344 μA cm-2. The latter is much higher than the net photocurrent of 40 μΑ cm-2 recorded at the same Pt(Cu)/TiO2/Ti electrode in the acid supporting electrolyte (Figure 9(A)) and even higher than the limiting photocurrent of 260 μΑ cm-2 observed on the plain TiO2/Ti electrode in the presence of methanol (Figure 3(B)). This in turn indicates that the oxidation current enhancement observed at the Pt(Cu)/TiO2/Ti electrode upon illumination in the presence of methanol is not merely due to the superposition of methanol oxidation on Pt and photooxidation on TiO2 but due to a synergistic effect, most likely the photooxidative removal of poisonous intermediates from Pt sites at neighbouring TiO2 sites.

Figure 11. Voltammograms of a Pt(Cu)/TiO2/Ti electrode (rutile-rich/ annealed at 700°C for 5h) in the dark and under UV illumination, in a 0.5 M MeOH + 1 M HClO4 solution. (Arrows indicate

potential scan direction.)

Conclusions

i. It has been demonstrated that the two-step electrodeposition-galvanic replacement technique (previously applied to metallic substrates) can be successfully applied to semiconductor oxides too. Hence, following Cu electrodeposition on particulate TiO2 electrode coatings and its subsequent replacement by Pt, stable Pt(Cu)/TiO2 catalytic layers can be formed.

ii. Depending on whether the TiO2 substrate was rutile-rich or anatase-rich, the formation and stripping of Pt surface oxides could or could not be seen in cyclic voltammetry experiments. This indicates stronger interactions between Pt and anatase than between Pt and rutile.

iii. As a direct consequence of the above point, methanol oxidation in the dark is largely suppressed at platinized anatase-rich electrodes. (Since the suppression of Pt oxide formation is known to have a beneficial effect on oxygen reduction, such electrodes are worth testing in that reaction.)


168

iv. The significant methanol oxidation current enhancement observed upon the UV illumination of the platinized rutile-rich electrodes (due to Pt and TiO2 synergism) makes more research into their optimisation worth pursuing.

Acknowledgements: This research has been co-financed by the European Union (European Social Fund – ESF) and Greek national funds through the Operational Program "Education and Lifelong Learning" of the National Strategic Reference Framework (NSRF) - Research Funding Program: Heracleitus II. Investing in knowledge society through the European Social Fund.

References

[1] S.Trasatti (Ed.), Electrodes of Conductive Metallic Oxides Parts A and B, Elsevier, Amsterdam , 1980-1981

[2] A.C.C.Tseung,, S.Yeh, X.Liu, G.H.Kelsall and P.Dykstra, Novel Acid Resistant Oxygen Evolution Anodes, Commission of the European Communities, Brussels, (1990) p.9

[3] Ulman's Encyclopedia of Industrial Electrochemistry, vol. A6, VCH, Weinheim(1986) [4] H.I.Warren, Anodes for Electrowinning, D.J.Robinson (Ed.), The Metallurgical Society of

AIME, (1984) p.79 [5] V. Subramanian, E. Wolf, P.V. Kamat, J. Phys. Chem. B 105 (2001) 11439-11446 [6] T. Ohsawa, Y. Yamamoto, M. Sumiya, Y. Matsumoto, H. Koinuma, Langmuir 20(2004) 3018 [7] B. Wen, C. Liu, Y. Liu, J. Phys. Chem. B 109 (2005) 12372-12375 [8] J.-M. Lee, S.-B. Han, J.-Y. Kim, Y.-W. Lee, A.-R. Ko, B. Roh, I. Hwang, K.-W. Park, Carbon 48(8)

(2010) 2290-2296 [9] H. Song, P. Xiao, X. Qiu, W. Zhu, J. Power Sources 195(6) (2010) 1610-1614

[10] G. Selvarani, S. Maheswari, P. Sridhar, S. Pitchumani, A.K. Shukla, J. Electrochem. Soc. 156(11) (2009) B1354-B1360

[11] C.S. Chen, F.-M. Pan, Appl. Catal. B-Environ. 91(3-4) (2009) 663-669 [12] E. Mahe, D. Devilliers, Electrochim. Acta 46 (2001) 629-636 [13] J.V. Macpherson, J.-P. Gueneau de Mussy, Electrochem. Solid State Let. 4(9) (2001) E33-E36 [14] J.V. Macpherson, J.-P. Gueneau de Mussy, J. Electrochem. Soc. 149(7) (2002) B306-313 [15] G.A. Hope, A.J. Bard, J. Phys .Chem. 87 (1983) 1979-1984 [16] K. Tammeveski, T. Tenno, A. Rosental, P. Talonen, L.-S. Johansson, L. Niinisto, J. Electro-

chem. Soc. 146(2) (1999) 669-676 [17] B.E. Hayden, D.V. Malevich, D. Pletcher, Electrochem. Commun. 3 (2001) 395-399 [18] X.-Z. Li, F.-B. Li, J. Appl. Electrochem. 32 (2002) 203-210 [19] L. Xiong, A. Manthiram, Electrochim. Acta 49 (2004) 4163-4170 [20] S.R. Branković, J.X. Wang, R.R. Adžić, Surf. Sci. 474 (2001) L173-L179 [21] S.R. Branković, J. McBreen, R.R. Adžić, J. Electroanal. Chem. 503 (2001) 99-104 [22] S.R. Branković, J. McBreen, R.R. Adžić, Surf. Sci. 479 (2001) L363-368 [23] M. Van Brussel, G. Kokkinidis, I. Vandendael, C. Buess-Herman, Electrochem. Commun.

4(10) (2002) 808-813 [24] M. Van Brussel, G. Kokkinidis, A. Hubin, Cl. Buess-Herman, Electrochim. Acta 48 (2003)

3909-3919 [25] S. Papadimitriou, A. Tegou, E. Pavlidou, G. Kokkinidis, S. Sotiropoulos, Electrochim. Acta 52

(2007) 6254-6260 [26] A. Tegou, S. Papadimitriou, E. Pavlidou, G. Kokkinidis, S. Sotiropoulos, J .Electroanal. Chem.

608 (2007) 67-77 [27] S. Papadimitriou, A. Tegou, E. Pavlidou, G. Kokkinidis, S. Sotiropoulos, Electrochim. Acta 53

(2008) 6559-6567


doi: 10.5599/jese.2012.0020 169

[28] A. Tegou, S. Papadimitriou, S. Armyanov, E. Valova, G. Kokkinidis, S. Sotiropoulos, J. Electro-anal. Chem. 623(2) (2008) 187-196

[29] A. Tegou, S. Papadimitriou, G. Kokkinidis, S. Sotiropoulos, J. Solid State Electrochem. 14(2) (2010) 175-184

[30] S. Papadimitriou, S. Armyanov, E. Valova, A. Hubin, O. Steenhaut, E. Pavlidou, G. Kokkinidis, S. Sotiropoulos, J. Phys. Chem. C 114(11) (2010) 5217-5223

[31] Y. Ando, K. Sakaki, R. Adzic, Electrochem. Commun. 11 (2009) 1135-1138 [32] B.I. Podlovchenko, T.D. Gladysheva, A.Yu. Filatov, L.V. Yashina, Russ. J. Electrochem. 46(10)

(2010) 1189-1197 [33] K.-W. Park, S.-B. Han, J.-M. Lee Electrochem. Commun. 9 (2007) 1578–1581 [34] S. Bakardjieva, J. Subrt, V. Stengl, M.J. Dianez. J. Sayagues, Appl. Catal. B-Environ. 58 (2005)

193-202 [35] V.B. Baez, J.E. Graves, D. Pletcher, J. Electroanal. Chem. 340 (1992) 273-286 [36] V.B. Baez and D. Pletcher, J. Electroanal. Chem. 377 (1994) 231-240 [37] V.B.Baez and D.Pletcher, J. Electroanal. Chem. 382 (1995) 59-64 [38] D.N. Furlong, D.E. Yates, T.W. Healy, in S. Trasatti (Ed.), Electrodes of Conductive Metallic

Oxides, Elsevier, Amsterdam, 1981, p.375 [39] B. O’Regan, J. Moser, M. Gratzel, J. Phys. Chem. 94 (1990) 8720-8726 [40] K. Vinodgopal, S. Hotchandani, P.V. Kamat, J. Phys. Chem. 97 (1993) 9040-9044 [41] S. Sodergren, A. Hagfeldt, J. Olsson, S.-E. Lindquist, J. Phys. Chem. 98 (1994) 5552-5556 [42] J. Augustynski, in A.Wieckowski (Ed.), Interfacial Electrochemistry: Theory, Experiment and

Applications, Marcel Dekker Inc., NY, 1999, p.707 [43] H.Iddir, V. Skavysh, S. Agut, N.D. Browning, M.M. Disko, Physical Review B - Condensed

Matter and Materials Physics, 73(4) (2006) art. no. 041403 [44] L. Avalle, E. Santos, V.A. Macagno, Electrochim. Acta 39 (1994) 1291-1295 [45] J. Sanchez, M. Koudelka, J. Augustynski, J. Electroanal. Chem. 140 (1982) 161-166 [46] K. Uosaki, R. Yoneda, H. Kita, J. Phys. Chem. 89 (1985) 4042-4046 [47] B.E. Hayden, D. Pletcher, J.-P. Suchsland, L.J. Williams Phys. Chem. Chem. Phys. 11 (2009)

1564-1570 [48] J.A. Horsley, J. Am. Chem. Soc. 101(11) (1979) 2870-2874 [49] M. Hampden-Smith, P. Atanassova, P. Atanassov, T. Kodas, in W. Vieltich, A. Lamm, H.A.

Gasteiger (Eds.), Handbook of Fuel Cells: Fundamentals Technology and Applications, Vol.3, Wiley, Chichester, England, 2003, p.506

[50] M. Chen, Z.-B. Wang, Y. Ding, G.-P. Yin, Electrochem. Commun. 10 (2008) 443-446

© 2012 by the authors; licensee IAPC, Zagreb, Croatia. This article is an open-access article distributed under the terms and conditions of the Creative Commons Attribution license

(http://creativecommons.org/licenses/by/3.0/)

doi: 10.5599/jese.2012.0022 171




Electrochemical degradation of Reactive Brilliant Red K-2BP on Ti/RuTiIrSnMn oxide anode in a batch cell LUO JIANCHENG*, YANG JIE*, LI WEISHAN*,**,***, HUANG QIMING*,**,*** and XU HONGKANG****

*School of Chemistry and Environment, South China Normal University, Guangzhou 510006, China

**Key Laboratory of Electrochemical Technology on Energy Storage and Power Generation of Guangdong Higher Education Institutes, South China Normal University, Guangzhou 510006, China

***Engineering Research Center of Materials and Technology for Electrochemical Energy Storage (MOE), South China Normal University, Guangzhou 510006, China

****Dongguan Hongjie Environmental Technologies Ltd, Dongguan 523039, China Corresponding Author: E-mail: [email protected]; Tel.: +86-020-39310256; Fax: +86-020-39310256

Received: July 06, 2012; Published: November 10, 2012

Abstract Electrochemical degradation of Reactive Brilliant Red K-2BP on Ti/RuTiIrSnMn oxide anode in chloride containing solution was investigated by voltammetry and electrolysis in a batch cell. It is found that the degradation mechanism of K-2BP on Ti/RuTiIrSnMn oxide anode involves an indirect electrocatalytic oxidation, in which K-2BP is oxidized by the electrochemically generated active chlorine. This degradation reaction follows pseudo-first order reaction kinetics. Ti/RuTiIrSnMn oxide exhibits excellent electrocata-lytic activity toward the generation of active chlorine from chloride. Hence, K-2BP can be electrochemically degraded effectively in chloride containing solution. The decolorization efficiency was found to increase with the decrease in pH and with the increase in current density, NaCl concentration, temperature, and flow rate of the solution.

Keywords Decolorization; Electrooxidation; Active chlorine

Introduction

Dye effluent in textile industry is an environmental concern due to its huge quantity, dark color, low biodegradability, and potential toxicity to aquatic life. Various treatment methods have been proposed and/or tested to effectively degrade the effluent, including biological treatment [1-7],

J. Electrochem. Sci. Eng. 2(4) (2012) 171-183 DEGRADATION OF REACTIVE BRILLIANT RED K-2BP

172

photocatalysis [8-10], ozonation [11,12], wet oxidation [13,14], Fenton reaction [15,16], electro-chemical coagulation [17-19], and electrochemical oxidation [20-31].

Compared with other methods for the treatment of the textile effluent, electrochemical tech-nologies are easy to operate, highly efficient and environmentally friendly. Electrooxidation has been widely used to treat various industrial wastewaters, including the wastewaters containing dye, urea, phenol, molasses, microcystins, tannery, olive oil, pesticide, etc. [20,26-32].

There are two mechanisms for the electrooxidation treatment of wastewater, direct and indirect oxidation. In the direct oxidation mechanism, pollutants are oxidized directly on the anode surface. In that case, there is no any other substance involved, except the electrons in the electrooxidation of pollutants. However, pollutants and their degraded products are usually adsorbed on the anode, leading to the low degradation efficiency and the complicated treatment technology for the anode cleaning. In the indirect mechanism, pollutants are oxidized in a bulk solution by a medium, such as active chlorine and hydroxyl radicals electrogenerated in-situ on the anode surface, and thus the adsorption of pollutants and their degradation products can be avoided. The medium is utilized immediately after its electrogeneration. Therefore, indirect electrooxidation can lead to a high efficiency of degradation.

Dimensionally stable anode (DSA) was originally designed for chlor-alkali industry [33]. DSA has long lifetime and it is much cheaper than noble metal anodes such as platinum. It has been widely used in a wastewater treatment research and development. Ruthenium- and iridium- oxides are the main composition of DSAs. They exhibit good electrocatalytic activity for active chlorine generation and oxygen evolution. Based on these two oxides, many composite oxides have been developed for DSA use. For example, ruthenium-based oxides, especially Ti/Ti0.7Ru0.3O2, have been reported to treat the refractory effluent [26,28,30,31].

Reactive Brilliant Red K-2BP (K-2BP) is an azo-dye which is extensively used in the textile industry. Its molecule structure is shown in Fig. 1. K-2BP is resistant to either biological or photo-degradation. The solution containing K-2BP is brilliant red and can be distinguished even in trace concentration by naked eye.

Figure 1. Molecule structure of Reactive Brilliant Red K-2BP

In this work, a coating on titanium substrate of composite oxides based mainly on Ru and Ir, Ti/RuTiIrSnMn oxide, was used as an anode to degrade K-2BP by the indirect electrooxidation with electrochemically generated active chlorine as the medium. The parameter effects of electrolytic conditions on the degradation efficiency were investigated and the degradation mechanism was discussed.

Experimental

Reagents and solutions

Deionized water was used for the preparation of all solutions. Sodium hydroxide and sulfuric acid were used to adjust pH of the solution. Sodium chloride was used for the generation of active

SO3Na

N N

NaO3S SO3Na

OH HN NH

N

N

N

Cl

Cl

L. Jiancheng et al. J. Electrochem. Sci. Eng. 2(4) (2012) 171-183

doi: 10.5599/jese.2012.0022 173

chlorine. Sodium sulfate and sodium nitrate were also used as electrolytes for comparison. All the reagents were analytical reagents and purchased from Sinopharm Chemical Reagent Co., Ltd, China. The commercially used textile azo dye, mainly containing Reactive Brilliant Red K-2BP, was purchased from the local market. All the chemicals were used as received without further purification.

Electrolysis cell

The experiments were carried out in an un-divided filter-press cell (Fig. 2) with constant current at fixed temperature of 30 °C except otherwise stated. The dye-containing solution was supplied from the feed tank (250 mL) to the electrolysis cell by a peristaltic pump. The treated water can be recycled back to the solution tank if necessary. Constant current was provided by a D. C. power supply (363, EG&G). A magnetic stirrer was used to keep the solution mixed and a thermostatic water bath was employed to control the temperature. The solution volume was 200 mL with 100 mg/L K-2BP. The anode (Ti/RuTiIrSnMn oxide) was supplied by Guangzhou Etsing Plating Research Institute, which was prepared by coating the composites (the compounds containing Ru, Ti, Ir, Sn, and Mn) on Ti plate and treated under high temperature. The cathode was made from stainless steel (SUS304). Both electrodes, anode and cathode were of the same size of geometric area having 21 cm2 exposed to the solution. The gap between the two electrodes was 10 mm. Before each set of experiments, the anode was immersed in absolute alcohol for 5 min and then rinsed with deionized water. The cathode was polished with sandpaper (2000), then immersed in 1.0 M HCl solution for 5 min, and finally rinsed with deionized water.

1. D. C. supply 2. magnetic stirrer 3. peristaltic pump 4. un-divided filter-press cell 5. feed tank

51.sampling 52. inlet 53. Outlet

Figure 2. Schematic diagram of experimental setup for the degradation of K-2BP

Analytical procedure

At desired time interval, the dye solution was sampled and filtered through a 0.45 μm millipore membrane. Several drops of saturated sodium sulfite solution were added into the sample to prevent further oxidation of the dyestuffs and their intermediates by residual chlorine. The samples were analyzed immediately after their collection using a Shimadzu UV-1700 double-beam spectrophotometer.

A VON O FF

- +

Tem Stir

Tem Ctrl Stir Ctrl

2

1

3

45

5152

53

A VON O FF

- +

Tem Stir

Tem Ctrl Stir Ctrl

2

1

3

45

A VON O FF

- +A VA VON O FFON O FF

- +- +

Tem Stir

Tem Ctrl Stir Ctrl

Tem Stir

Tem Ctrl Stir Ctrl

TemTem StirStir

Tem CtrlTem Ctrl Stir CtrlStir Ctrl

2

1

3

45

5152

53


174

The concentration of K-2BP in the samples was determined from the UV-Vis absorbance recorded at 520 nm (max) and the degradation efficiency (A / %) was calculated according to [28]:

0 t

0

( )/ % 100

A AA

A−=

where A0 and At are the absorbance of the solution before and after electrolysis, respectively.

Voltammetric measurements

In order to understand the degradation mechanism of K-2BP on Ti/RuTiIrSnMn oxide anode, linear and cyclic voltammograms were obtained by using a three-electrode cell (Solartron 1470E, Solartron Co., Ltd.). A Pt plate was used as the counter electrode, a saturated Hg/Hg2Cl2 electrode (SCE) was used as the reference electrode, and Ti/RuTiIrSnMn oxide disk with a diameter of 1 cm was used as the working electrode. The potentials reported are with respect to the SCE. All the electrochemical measurements were carried out at room temperature. Linear voltammetry was performed in 0.5 M H2SO4 solutions with and without 1.0 M NaCl from 0.5 V to 1.35 V at a can rate of 1 mV s-1. Cyclic voltammetry was performed in 0.5 M H2SO4 + 1.0 M NaCl solutions with and without 200 mg L-1 K-2BP between 0.5 V and 1.15 V at 1 mV s-1. The cyclic voltammograms reported in this paper are the results of the 5th cycle. Before the experiments, the working electrode was immersed in absolute alcohol for 5 min, rinsed by deionized water and kept in the experimental solutions for 30 min.


Effect of salts

In order to enhance the diffusion and adsorption of dyestuffs onto the textile, abundant inorganic salts, such as NaCl, Na2SO4 and NaNO3, are added into the dyestuff slurries. This technology results in the real textile effluent containing various electrolytes. On the other hand, the existence of salts improves the conductivity of the effluent and reduces the electric energy needed for the electrochemical treatment of the effluent. Therefore, in this paper the effect of various salts on degradation efficiency of K-2BP on Ti/RuIrSnMnTi anode was investigated.

As shown in Fig. 3, the decolorization efficiency of K-2BP in both Na2SO4 and NaNO3 solutions increases slowly and linearly with treatment time. The decolorization efficiency of K-2BP is only 50.9 % after one hour electrolysis in both Na2SO4 and NaNO3 solutions. However, the decolori-zation efficiency of K-2BP in NaCl solution increases drastically in the first 20 min, then increases slowly in the later stage and reaches almost 100 % after 40 min. This indicates that NaCl favors the degradation of K-2BP. The mechanism can be explained as follows.

In the NaCl-free solution, water is decomposed on the surface of metal oxide anode, producing hydroxyl radicals, and the dye is degraded into CO2 by hydroxyl radicals subsequently (Eqs. 1 and 2):

H2O – e- →·OH + H+ (1)

·OH + dye – e- → H2O + H+ + CO2↑ (2)

The reaction of water decomposition on anode to form hydroxyl radicals is usually accompanied by the oxygen evolution reaction (Eq. 3), which reduces formation efficiency of the hydroxyl radicals and thus leads to the low degradation efficiency of K-2BP in both Na2SO4 and NaNO3 solutions.

2H2O – 4e- → 4H+ + O2↑ (3)


doi: 10.5599/jese.2012.0022 175

0 10 20 30 40 50 60

0

20

40

60

80

100

Dec

olor

izat

ion

, %

Time, min.

NaCl NaNO3

Na2SO4

Figure 3. Effect of supporting electrolyte (SE) on the decolorization efficiency of K-2BP.

(j = 30 mA cm-2, cSE = 4.0 g L-1, flux = 50 mL min-1, initial pH = 7.0)

The composite of Ru, Ir, Sn, Mn, and Ti oxides exhibits high activity for the oxidation of chloride ions to form active chlorine. In the NaCl-containing solution, chloride ions are oxidized on this kind of composite oxide anode, forming active chlorine [22,27,29] (Eq. 4).

2Cl- – 2e- → Cl2↑ (4)

The active chlorine, electrogenerated in-situ on an anode surface, transforms into a strong oxidant, such as HClO, which oxidizes dyes into small molecules or even carbon dioxide (Eq. 5). Therefore, higher degradation efficiency of the dye on the anode in the NaCl-containing solution can be expected than that in NaCl-free solution.

HClO + dye (and its oxidation intermediates) → H2O + Cl- + CO2↑ (5)

Fig. 4 presents the linear voltammograms of Ti/RuIrSnMnTi oxide electrode in 0.5 M H2SO4 solution with and without NaCl.

0.6 0.8 1.0 1.2 1.4

0

5

10

15

20

b

Cur

ren

t, m

A

Potential, V vs. SCE

a: without NaCl b: with NaCl

a

Figure 4. Linear voltammograms of Ti/RuIrSnMnTi oxide electrode in the solutions

with and without NaCl


176

In the NaCl-free solution, the decomposition potential of water is 1.231 V. However, when chloride ions are added, the decomposition potential shifts negatively to 1.142 V and the current increases significantly. This indicates that chloride ions can be oxidized more easily on Ti/RuIrSnMnTi oxide electrode than the water decomposition, thus the Ti/RuIrSnMnTi oxide shows a good electrocatalytic activity toward the oxidation of chloride ions.

Fig. 5 presents the voltammograms of Ti/RuIrSnMnTi oxide electrode in 0.5 M H2SO4 + 1.0 M NaCl solutions with and without 200 mg/L K-2BP. For the electrode in the solution without K-2BP, the current increases sharply at the potential higher than 1.1 V during the forward scanning. Based on the observation of Fig. 4, it follows that the oxidation current is corresponding to the formation of active chlorine from oxidation of chloride ions. Active chlorine can be detected during the backward scanning, as shown by the current peak at about 1.09 V. For the electrode in the solution with K-2BP, the current for the oxidation of chloride ions is similar to that in the solution without K-2BP, but the peak current for the reduction of active chlorine is reduced. This suggests that K-2BP is not oxidized directly on the Ti/RuIrSnMnTi anode, but indirectly by active chlorine.

0.4 0.6 0.8 1.0 1.2-0.5

0.0

0.5

1.0

1.5

2.0

2.5

b

Cu

rren

t, m

A

Potential, V vs. SCE

a: without K-2BP b: with K-2BP

a

Figure 5. Cyclic voltammograms of Ti/RuIrSnMnTi oxide electrode in 0.5 M H2SO4 + 1.0 M NaCl solutions with and without 200 mg L-1 K-2BP

Effect of current density

Charge transfer step is a critical process in electrocatalysis, where the current density at the anode surface determines the rate of the electrode reaction. The effect of current density on the decolorization efficiency was investigated. Fig. 6 presents the obtained results. It can be seen from Fig. 6 that in the first 15 min, when the applied current density rises from 5 to 20 mA cm-2, the decolorization efficiency increases drastically from 50 to 85 %. There is an optimal decolorization efficiency at 20 mA cm-2. A further increase of current density makes no obvious improvement in the decolorization efficiency. This phenomenon can be explained as follows. When the applied current density is low, the production rate of active chlorine is less than its consumption rate for the oxidation of K-2BP, resulting in the low decolorization efficiency. When the applied current density is high, the production rate of active chlorine increases and the decolorization efficiency is improved. As the applied current density increases further, the concentration of chloride ions on anode surface is deficient and active chlorine is no more available for oxidizing the K-2BP.


doi: 10.5599/jese.2012.0022 177

0 10 20 30 40 50 60

0

20

40

60

80

100

Dec

olor

izat

ion

, %

Time, min.

5mA/cm2

10mA/cm2

20mA/cm2

30mA/cm2

40mA/cm2

Figure 6. Effect of current density on the decolorization efficiency of K-2BP (CNaCl=4.0g/L, flux=50mL/min, initial pH=7.0)

Effect of NaCl concentration

Fig. 7 presents the effect of NaCl concentration on the decolorization of K-2BP. It can be seen from Fig. 7 that in the broad range of NaCl concentration, from 0.5 to 5.0 g L-1, K-2BP can be removed completely, even at low NaCl concentration. For instance, when the dose of NaCl is 0.5 g L-1, the decolorization efficiency reaches 90 % in the first 25 min, and in the later 15 min the decolorization efficiency reaches 100 %. By increasing the dose of NaCl from 0.5 to 2.0 g L-1 the decolorization rate is improved. However, a further increase of the NaCl dose does not improve the decolorization rate any further.

0 10 20 30 40 50 60

0

20

40

60

80

100

Dec

olor

izat

ion

, %

Time, min.

0.5g/L 1.0g/L 2.0g/L 3.0g/L 4.0g/L 5.0g/L

Figure 7. Effect of NaCl concentration on the decolorization efficiency of K-2BP(j=30 mA/cm2, flux=50 mL/min, initial pH=7.0)

The electrooxidaton process of K-2BP on Ti/RuIrSnMnTi anode is determined not only by the charge transfer step, but also by the mass transfer step. At low NaCl concentration, the production rate of active chlorine is less than its consumption rate with K-2BP, leading to low decolorization


178

rate. The increase of NaCl dose can reduce the effect of the mass transfer process and increases the formation rate of active chlorine. When the dose of NaCl is 2.0 g L-1, the formation rate of active chlorine matches the reaction rate between active chlorine and K-2BP. Further increasing of the dose of NaCl does not increase the decolorization rate, because the formation of active chlorine is controlled by the charge transfer step.

Effect of initial pH

Fig. 8 presents the effect of the initial pH of the solution on the decolorization efficiency. As shown in Fig. 8, the initial pH of the solution affects the decolorization efficiency of K-2BP signifi-cantly. In the pH range between 1.0 and 11.0, the decolorization efficiency of K-2BP increases with the decrease of the initial pH of the solution.

0 10 20 30 40 50 600

20

40

60

80

100

Dec

olor

izat

ion

, %

Time, min.

pH1.0 pH3.0 pH5.0 pH7.0 pH9.0 pH11.0

Figure 8. Effect of initial pH on the decolorization efficiency of K-2BP(j=30 mA/cm2, cNaCl=4.0 g/L, flux=50 mL/min)

In aqueous solution, the active chlorine in-situ electrogenerated is in the form of hypochlorous acid and hypochlorite ion, which reach an equilibrium (Eq. 6):

HClO ⇄ ClO- + H+ (6)

According to the Le Chatelier’s principle, the increase of the hydrogen ion concentration shifts the equilibrium to the left in favor of forming hypochlorous acid. The standard formation potential of hypochlorous acid and hypochlorite ion is 1.49 V and 0.81 V, respectively, and the potential is a function of hydrogen ion concentration in solution (Eqs. 7 and 8):

φ(HClO/Cl-) = φθ(HClO/Cl-) + RT/2F ln([H+][HClO]/[Cl-]) (7)

φ(ClO-/Cl-) = φθ(ClO-/Cl-) + RT/2F ln([ClO-]/[OH-]2[Cl-]) (8)

Therefore, oxidative ability of hypochlorous acid is much stronger than of hypochlorite ion, and the smaller the value of pH is, the stronger is the oxidative ability of hypochlorous acid [30,34-37].

Effect of flow rate

The effect of solution flow rate on decolorization efficiency of K-2BP was considered in the range from 10 to 70 ml min-1. Fig. 9 presents the obtained results. It can be seen from Fig. 9 that decolorization efficiency increases as the flow rate increases from 10 to 50 ml min-1. Further incre-ase in the flow rate does not affect the decolorization efficiency significantly. Apparently, the


doi: 10.5599/jese.2012.0022 179

formation process of active chlorine is also controlled by the diffusion of chloride ions. The diffusion layer thickness is proportional to the flow rate, i.e. the faster flow rate favors the diffusion of chloride ions to the electrode surface. However, at high flow rate, turbulent flow appears and the effect of flow rate cannot be observed.

0 10 20 30 40 50 600

20

40

60

80

100

Dec

olor

izat

ion

, %

Time, min.

10mL/min 30mL/min 50mL/min 70mL/min

Figure 9. Effect of flow rate on the decolorization efficiency of K-2BP

(j = 30 mA cm-2, cNaC l= 4.0 g L-1, initial pH = 7.0）

Effect of temperature

Fig. 10 presents the effect of temperature on the decolorization efficiency of K-2BP. As shown in Fig. 10 the increasing temperature leads to the enhancement of the decolorization efficiency. The time of complete decolorization is 35 min at 303 K, while it reduces to 20 min at 343 K. The enhancement of decolorization efficiency of K-2BP can be ascribed to the improvement of the charge transfer rate and the reaction rate of the active chlorine with K-2BP in the solution.

0 10 20 30 400

20

40

60

80

100

Dec

olor

izat

ion

, %

Time, min.

303K 313K 323K 333K 343K

Figure 10. Effect of temperature on the decolorization efficiency of K-2BP

( j= 30 mA cm-2, cNaC l= 4.0 g L-1, flux = 50 mL min-1, pH = 7.0)


180

Reaction kinetics

The variation of K-2BP concentration (absorbance) with time at different temperatures was modeled by the zero order, the first order and the second order reaction. The obtained results are shown in Table 1. It can be seen from Table 1 that the correlation coefficient (R2) of the first order kinetics model is about 0.999, far higher than those of zero and second order models. This indicates that the decolorization process of K-2BP follows the pseudo-first-order kinetics. Therefore, the decolorization rate equation can be expressed by (Eq. 9):

t tobs

0 0

ln lnc A

K tc A

= = − (9)

where t is the reaction time (min), c0 and ct are the concentrations (mg L-1) of K-2BP at the initial and at t, A0 and At are the absorbance of K-2BP at the initial and at t, and Kobs is the pseudo-first-order kinetic constant (min-1).

Table 1. Results of fitting of the kinetics equation for zero-order, first-order and second-order of Reactive Brilliant Red K-2BP at various temperatures

Temperature, K zero order second order first order

R2 R2 R2 Kobs/min-1

303 0.84361 0.91002 0.99933 0.15119 313 0.79883 0.87378 0.99946 0.16941 323 0.76557 0.79324 0.99987 0.19533 333 0.71530 0.71310 0.99903 0.23650 343 0.67154 0.71861 0.99574 0.28096

Fig. 11 shows the linear relationship between the decolorization time t and ln (At/A0). Fig. 12

presents the relationship between rate constant Kobs and temperature. It can be seen that the rate constant increases with temperature.

0 5 10 15 20-6

-4

-2

0

ln (

A/A 0)

Time, min.

303k 313k 323k 333k 343k Linear Fit of 303K Linear Fit of 313K Linear Fit of 323K Linear Fit of 333K Linear Fit of 343K

Figure 11. First-order kinetic fitting results of decolorization rate of K-2BP at various

temperatures


doi: 10.5599/jese.2012.0022 181

300 310 320 330 340 350

0.15

0.20

0.25

0.30

K obs /

min

-1

Temperature, K Figure 12. Relationship between rate constant Kobs and temperature

UV-visible spectroscopy

In K-2BP there are aromatic rings, naphthalene rings, triazine rings and -N=N- groups. -N=N- is the chromophore group and the bridge between the aromatic ring and naphthalene to form a big bond conjugate system. Fig. 13 presents the variation of K-2BP UV-visible spectrum with time. K-2BP is characterized by three absorbance peaks of 235 nm, 285 nm and 370 nm in ultraviolet region and only one peak of 520 nm in visible region. The absorbance in the visible region should be ascribed to the chromophore group of -N=N-, and the absorbance in ultraviolet region should be ascribed to the aromatic ring, triazine ring and naphthalene ring, respectively [38-40]. It can be seen from Fig. 13 that the intensity of all the absorbance peaks decreases rapidly with time and these peaks disappear completely in 30 min. This indicates that in the NaCl-containing solution, Ti/RuIrSnMnTi anode can remove the color of K-2BP completely, and the triazine ring, aromatic ring and naphthalene ring can be cleaved into aliphatic acid with low molecular weight or even completely mineralized.

200 300 400 500 600 700 8000

1

2

3

4

Ab

sorb

ance

, a. u

.

Wavelength, nm

0min5min10min15min20min25min30min

Figure 13. UV-Vis spectrum variation of K-2BP during electrolysis in NaCl solution


182

Conclusions

Ti/RuIrSnMnTi oxide anode can be used to degrade Reactive Brilliant Red K-2BP effectively in the solution containing chloride ions. The degradation mechanism of K-2BP is an indirect electrocatalytic oxidation, in which K-2BP is oxidized by the electrogenerated active chlorine. This degradation reaction follows pseudo-fist-order-reaction kinetics and can be improved by increasing current density and NaCl concentration, enhancing the temperature and flow rate, and lowering the pH of the solution.

Acknowledgements: This work was financially supported by Natural Science Foundation of Guangdong Province (Grant No. 10351063101000001) and Combination Project of Enterprise, University and Scientific Research of Guangdong Province (Grant No. 2010B090400019).

References

[1] H. Salvado, M.P. Gracia, J.M. Amigó, Water Res. 29 (1995) 1041-1050. [2] W. Fuchs, C. Resch, M. Kernstock, M. Mayer, P. Schoeberl, R. Braun, Water Res. 39 (2005)

803-810. [3] K.C.A. Bromley-Challenor, J.S. Knapp, Z. Zhang, N.C.C. Gray, M.J. Hetheridge, M.R. Evans,

Water Res. 34 (2000) 4410-4418. [4] S. Chinwetkitvanich, M. Tuntoolvest, T. Panswad, Water Res. 34 (2000) 2223-2232. [5] A.B. Dos-Santos, F.J. Cervantes, R.E. Yaya-Beas, J.B. van Lier, Enzyme Microb. Tech. 33 (2003)

942-951. [6] T. Panswad, K. lamsamer, J. Anotai, Bioresource Technol. 76 (2001) 151-159. [7] C.T.M.J. Frijters, R.H. Vos, G. Scheffer, R. Mulder, Water Res. 40 (2006) 1249-1257. [8] J.F. Lei, W.S. Li, Acta Phys. Chim. Sin. 25 (2009) 1173-1178. [9] I. Peternel, N. Koprivanac, H. Kušić, Water Res. 40 (2006) 525-532.

[10] A. Vijayabalan, K. Selvam, R. Velmurugan, M. Swaminathan, J. Hazard. Mater. 172 (2009) 914-921.

[11] J.N. Wu, T.W. Wang, Water Res. 35 (2001) 1093-1099. [12] T.Y. Chen, C.M. Kao, A. Hong, C.E. Lin, S.H. Liang, Desalination 249 (2009) 1238-1242. [13] G.H. Chen, L.C. Lei, P. Yue, Ind. Eng. Chem. Res. 38 (1999) 1837-1843. [14] A. Santos, P. Yustos, S. Rodriguez, F. Garcia-Ochoa, M. de Garcia, Ind. Eng. Chem. Res. 46

(2007) 2423-2427. [15] I. Gulkaya, G.A. Surucu, F.B. Dilek, J. Hazard. Mater. 136 (2006) 763-769. [16] C.T. Wang, J.L. Hu, W.L. Chou, Y.M. Kuo, J. Hazard. Mater. 152 (2008) 601-606. [17] A.H. Essadki, M. Bennajah, B. Gourich, C. Vial, M. Azzi, H. Delmas, Chem. Eng. Process. 47

(2008) 1211-1223. [18] X.D. Zhang, J.D. Hao, W.S. Li, H.J. Jin, J. Yang, Y.M. Huang, D.S. Lu, H.K. Xu, J. Hazard Mater.

170 (2009) 883-887. [19] W. Balla, A.H. Essadki, B. Gourich, A.Dassaa, H. Chenik, M. Azzi, J. Hazard. Mater. 184

(2010) 710-716. [20] M. Panizza, G. Cerisola, Water Res. 40 (2006) 1179-1184. [21] M. Panizza, G. Cerisola, Environ. Sci. Technol. 38 (2004) 5470-5475. [22] Z.C. Wu, M.H. Zhou, Environ. Sci.Technol. 35 (2001) 2698-2703. [23] M.H. Zhou, Q.Z. Dai, L.C. Lei, C.Ma, D. Wang, Environ. Sci.Technol. 39 (2005) 363-370. [24] M. Li, C.P. Feng, W.W. Hu, Z. Zhang, N. Sugiura, J. Hazard. Mater. 162 (2009) 455-462. [25] J.B. Parsa, M. Rezaei, A.R. Soleymani, J. Hazard. Mater. 168 (2009) 997-1003. [26] S.S. Vaghela, A.D. Jethva, B.B. Mehta, S.P. Dave, S. Adimurthy, G. Ramachandraiah, Environ.

Sci. Technol. 39 (2005) 2848-2855.


doi: 10.5599/jese.2012.0022 183

[27] S. Song, J.Q. Fan, Z.Q. He, L. Chan, Z. Liu, J. Chen, X. Xu, Electrochim. Acta 55 (2010) 3606-3613.

[28] B.J. Hernlem, Water Res. 39 (2005) 2245-2252. [29] P. Canizares, J.A. Dominguez, M.A. Rodrigo, J. Villasenor, J. Rodriguez, Ind. Eng. Chem. Res.

38 (1999) 3779-3785. [30] H.X. Shi, J.H. Qu, A.M. Wang, J.T. Ge, Chemosphere 60 (2005) 326-333. [31] C.R. Costa, P. Olivi, Electrochim. Acta 54 (2009) 2046-2052. [32] C.A. Martinez-Huitle, A. Battisti, S. Ferro, S. Reyna, M. Cerro-Lopez, M.A. Quiro, Environ. Sci.

Technol. 42 (2008) 6929-6935. [33] S. Trasatti, Electrochim. Acta 45 (2000) 2377-2385. [34] A. Maljaei, M. Arami, N.M. Mahmoodi, Desalination 249 (2009) 1074-1078. [35] O. Scialdone, S. Randazzo, A. Galia, G. Silvestri, Water Res. 43 (2009) 2260-2272. [36] S. Raghu, C.A. Basha, Ind. Eng. Chem. Res. 47 (2008) 5277-5283. [37] L. Szpyrkowicz, C. Juzzolino, S.N. Kaul, S. Daniele, M.D. De Faveri Ind. Eng. Chem. Res. 39

(2000) 3241-3248. [38] X.N. Wang, J.P. Jia, Y.L. Wang, Ultrason. Sonochem. 17 (2010) 515-520. [39] J. Yang, J. Wang, J.P. Jia, Environ. Sci. Technol. 43 (2009) 3796-3802. [40] X.J. Lu, B. Yang, J.H. Chen, R.Sun, J. Hazard. Mater. 161 (2009) 241-245.



doi: 10.5599/jese.2012.0021 185




On the suppression of cathodic hypochlorite reduction by electrolyte additions of molybdate and chromate ions JOHN GUSTAVSSON, GONGZHUO LI, CHRISTINE HUMMELGÅRD*, JOAKIM BÄCKSTRÖM* and ANN CORNELL

Applied Electrochemistry, School of Chemical Science and Engineering, KTH Royal Institute of Technology, SE 100 44 Stockholm, Sweden

*Department of Natural Sciences, Engineering and Mathematics, Mid Sweden University, SE 851 70 Sundsvall, Sweden Corresponding Author: E-mail: [email protected]; Tel.: +46-87908172

Received: June 29, 2012; Revised: September 15, 2012; Published: November 10, 2012

Abstract The goal of this study was to gain a better understanding of the feasibility of replacing Cr(VI) in the chlorate process by Mo(VI), focusing on the cathode reaction selectivity for hydrogen evolution on steel and titanium in a hypochlorite containing electrolyte. To evaluate the ability of Cr(VI) and Mo(VI) additions to hinder hypochlorite reduction, potential sweep experiments on rotating disc electrodes and cathodic current efficiency (CE) measurements on stationary electrodes were performed. Formed electrode films were investigated with scanning electron microscopy and energy-dispersive X-ray spectroscopy. Cathodic hypochlorite reduction is hindered by the Mo-containing films formed on the cathode surface after Mo(VI) addition to the electrolyte, but much less efficient compared to Cr(VI) addition. Very low levels of Cr(VI), in the μM range, can efficiently suppress hypochlorite reduction on polished titanium and steel. Phosphate does not negatively influence the CE in the presence of Cr(VI) or Mo(VI) but the Mo-containing cathode films become thinner if the electrolyte during the film build-up also contains phosphate. For a RuO2-TiO2 anode polarized in electrolyte with 40 mM Mo(VI), the anode potential increased and increased molybdenum levels were detected on the electrode surface.

Keywords Current efficiency; hydrogen evolution; in-situ additives; cathode; electrolysis; EDS; SEM; potential sweeps; galvanostatic polarization

J. Electrochem. Sci. Eng. 2(4) (2012) 185-198 SUPPRESSION OF CATHODIC HYPOCHLORITE REDUCTION

186

Introduction

Sodium chlorate is industrially produced from sodium chloride in an electrolytic process according to reaction 1 below.

NaCl + 3H2O → NaClO3 + 3H2 (1)

Chlorine is produced on the anode and forms hypochlorite (here defined as ClO- + ClOH) (reactions 2-4) [1], and chlorate is formed in a chemical reaction (reaction 5) that shows an optimal reaction rate at pH 6-7. To prevent that the intermediate hypochlorite and the product chlorate are reduced on the cathode (reactions 6, 7 and 8) hexavalent chromium, Cr(VI), is added to the electrolyte. The hexavalent chromium is reduced on the cathode surface to a film of chromium (III)hydroxide [2]. The Cr(OH)3-film formed has been shown to be very thin and efficient at inhibiting the reduction of hypochlorite, chlorate [3], oxygen and ferricyanide [2].

2Cl- ↔ Cl2 + 2e- (2)

Cl2 + H2O ↔ HClO + Cl- + H+ (3)

HClO ↔ ClO- + H+ (4)

2HClO + ClO- → ClO3- + 2H+ + 2Cl- (5)

ClO- + H2O + 2e- → Cl- + 2OH- (6)

HClO + H+ + 2e- → Cl- + H2O (7)

ClO3- + 3H2O + 6e- → Cl- + 6OH- (8)

Unfortunately Cr(VI) is toxic and carcinogenic, which motivates investigations of possible alternatives.

Molybdenum is in the same group as chromium in the periodic table and the two elements have similarities. Molybdate (MoO4

2-) is opposed to chromate not toxic and is used as an environmentally friendly alternative to Cr(VI) in corrosion applications [4]. Mo-containing cathode films are formed during electrolysis of an electrolyte with Mo(VI), and these films can hinder the cathodic reduction of oxygen on copper [5]. This inhibition was shown to be efficient in electrolytes of neutral pH, where films of MoO2 are stable. Mo(VI) also buffers in a pH region close to that of Cr(VI) [6] and can activate the desired cathode reaction of hydrogen evolution [6,[7]. Not surprisingly Mo(VI) has been considered as a possible alternative to Cr(VI) in the chlorate process [6],[8]. A potential problem observed with Mo(VI) electrolyte additions to chlorate electrolyte is an increased oxygen level in the hydrogen cell gas [6], as this would affect the overall current efficiency (CE) for chlorate production as well as being a safety risk if the oxygen/hydrogen mixture reaches an explosive composition. Other results showed though that at low Mo(VI) concentrations (typically < 1 mM Mo(VI) compared to 39 mM Mo(VI) in [6]) no increased oxygen levels were reached and still the hydrogen evolution reaction on MAXTHAL 312 (Ti3SiC2) was substantially activated [8], by the Mo(VI) addition. Whether the increased oxygen levels related to homogeneous reaction in the electrolyte or to anode reactions is not clear, and no measurements of the anode potential were made in [6] or [8].

At low levels of Mo(VI) or Cr(VI) an additional buffer may be needed to stabilize the electrolyte pH in the chlorate process at 6-7. Phosphate is a possible buffer alternative, which may also form cathode films that can affect the CE [9] and thus the CE in electrolytes containing combinations of Mo(VI), Cr(VI) and phosphate are of interest.

J. Gustavsson et al. J. Electrochem. Sci. Eng. 2(4) (2012) 185-198

doi: 10.5599/jese.2012.0021 187

The goal of this study is to gain a better understanding of the feasibility of replacing Cr(VI) in the chlorate process by Mo(VI), focusing on the cathode reaction selectivity in the presence of a Mo-containing film. Reaction 6 or 7, the reduction of hypochlorite, was chosen as an indicator of the ability of different films to increase the CE for hydrogen evolution. Experiments with Cr(VI) additions to the electrolyte were made to compare the effects of Mo(VI) and Cr(VI). Cathodes of low carbon steel and titanium were used, as these are the most common cathode materials in chlorate electrolysis today. Both rotating disc electrodes (RDE) and stationary electrodes were used. The possibility of effects on the anode reactions of electrolyte additives were studied on a dimensionally stable anode (DSA)-type RDE.

Experimental

Water purified in a Millipore Direct-Q system was used in the preparation of all electrolytes, and to rinse the electrodes prior to experiments. Sodium hydroxide, sodium chloride, hydrochloric acid, sodium molybdate dihydrate, sodium dihydrogen phosphate all from Merck and of Pro Analysi grade; sodium hypochlorite solution 0.5 M in 0.1 M NaOH from VWR BDH Prolabo. In the text the term “hypochlorite” refers to the sum of [ClO-] and [HClO]. If not stated otherwise, the given hypochlorite concentration was the initial concentration. The terms “phosphate concentration” or “total phosphate concentration” here refers to the sum of phosphate species [PO4

3-]tot=[H3PO4]+ +[H2PO4-]+ [HPO4

2-]+ [PO43-].

For potential sweeps and polarization experiments RDEs were used and the measurements were performed using a PAR 273A potentiostat and an electrode rotator Model 616 from Pine Instrument Company. The Fe-RDE (5 mm diameter) and Ti-RDE (diameter 4 mm) were shielded by Teflon sheaths, polished by grit 4000 SiC grinding paper and rinsed in water before all experi-ments. The Fe-RDE was prior to the measurements corroded by immersion at OCP for 3 min with no rotation, in an electrolyte containing 5.2 M NaClO3, 1.9 M NaCl and 10 mM NaClO, pH 6.5 at 70°C. The potential sweep and polarization experiments were performed in a jacketed glass cell with 200 ml electrolyte. The temperature was controlled by a water bath and the electrolyte was purged with nitrogen for at least 15 min prior to and during the experiments. The reference electrodes connected to the electrolyte by an ion bridge filled with 5 M NaCl and a Luggin capillary. As the counter electrode a platinum grid was used.

For investigation of effects on the anode reactions polarization experiments were performed on a DSA-type anode. The anode was produced by spin-coating to produce a well-defined and relati-vely smooth coating of Ti0.7Ru0.3O2. A solution of 5.351 g RuCl3·nH2O (Ru content 35.51 % by mass) and 14.929 g Ti(IV)n-butoxide in n-propanol to a final volume of 50 ml was prepared. The solution was thoroughly stirred for several days. A disc of commercially pure titanium with smooth surface and a diameter of 59 mm was pickled in an aqueous solution of 27 ml concentrated HF per liter and 71 ml concentrated HNO3 per liter for two min at room temperature. The disc was placed in a spin coater (Electronic Micro Systems), an amount of solution was applied to the surface of the disc, and the solution was evenly distributed by spinning at 550 min-1. The disc was then dried at 80 °C for 10 min and calcined at 470 °C for 10 min. Three layers of coating were applied the same way. The final layer was dried for 40 min and calcined for one hour. The mass difference between the bare disc and the final sample was 0.0883 g corresponding to a Ru loading of 10.2 g m-2. To make DSA-RDEs, small discs (diameter 11.3 mm) were punched from the larger disc and put into ti-tanium holders. Exposed titanium was shielded by silicon tubing and epoxy. For the anodic polari-zation experiments the electrodes were pre-polarized for 30 min at 3 kA m-2 in a separate


188

2 M NaCl electrolyte at 70 °C. The recorded potentials were IR-corrected using a current interrupt technique [10].

Measurements of the CE for hydrogen evolution were performed in a divided cell [11] separated by a Gore Select ion-exchange membrane. The cell was immersed in a water bath for temperature control and a magnetic stirrer, set at the same level for all experiments, was used for the cathode compartment. The length of time needed to produce a given volumes of hydrogen gas (between 5 and 20 ml) were recorded. The CE was determined as by calculation from a calibration curve where hydrogen was evolved on a Pt-wire in Na2SO4 solution under the assumption of 100 % CE for hydrogen evolution. For the CE measurements stationary electrodes of titanium and low carbon steel were used. The electrodes were polished with grit 4000 SiC grinding paper and rinsed in water before all experiments. For CE measurements in neutral and alkaline electrolyte the hypochlorite level was measured using arsenite titration. The arsenite titration was performed with 5 ml electrolyte diluted by 200 ml water and pH adjusted to neutral pH. Mo(VI) was found to interfere with the titration, giving unreasonable high “hypochlorite levels”. Therefore hypochlorite levels in the electrolyte were only measured in experiments where no Mo(VI) had been added to the electrolyte. The hypochlorite concentration typically decreased from 80 to 40 mM after 20 min of electrolysis at -3 kA m-2, and the corresponding pH increased from 6.5 to 9-11.

X-Ray Fluorescence (XRF) measurements were performed using a Niton XLT898 with an accelerator voltage of 35 kV and a silver anode. Two different instruments for Scanning Electron Microscopy (SEM) were used; a Zeiss EvO 50 SEM with EDS for the elemental analyses and a Hitachi S-4800 field emission SEM for the micrographs presented.

Results and discussion

Current efficiency for hydrogen evolution in the presence of molybdate

Potential sweep experiments on a Ti-RDE with and without additions of Mo(VI) and hypochlorite to the electrolyte are shown in Figure 1. The sweeps were made in anodic direction after 5 min of pre-polarization at -1.5 V vs. Ag/AgCl, which enabled the formation of Mo-contain-ing films on the cathode whenever the electrolyte contained Mo(VI). In the presence of 15 mM hypochlorite, the dotted line, a cathodic limiting current for hypochlorite reduction of about 350 A m-2 can be seen. This limiting current was suppressed by the addition of Mo(VI), and the higher the Mo(VI) concentration the more efficient the hindering effect – see curves with addition of 1, 10 and 100 mM Mo(VI), respectively. From the Levich equation and data for viscosity and diffusivity from Byrne et al. [12] a limiting current density for hypochlorite reduction of 340 A m-2 was calculated, which agrees well with the experimental result.

Looking at the cathodic end of the graph an activation of hydrogen evolution about 300 mV is seen for the cases of 1 and 10 mM Mo(VI). As high Mo(VI) level as 100 mM did not give a similar decrease in overpotential, likely caused by resistive properties of the Mo(VI) film formed [7].

In Figure 2 potential sweeps on slightly corroded iron are shown, and the electrode had been pre-polarized as above. Again a limiting current for hypochlorite reduction is visible at -300 to -400 A m-2, and addition of 100 mM Mo(VI) only partly suppressed this current. A comparison with the result in Figure 1 indicates that it is easier to form a hindering film on a titanium substrate than on corroded iron. The high currents appearing at potentials > -0.6 V vs. Ag/AgCl in the Mo(VI) free case likely relate to oxidation of the iron electrode, which is hindered by the presence of a molybdenum containing film. Similar as in the case of titanium cathodes (Figure 1) the addition of


doi: 10.5599/jese.2012.0021 189

as high concentration as 100 mM Mo(VI) was less effective at activating the HER than lower Mo(VI) concentration.

Figure 1. Potential sweeps from -1.5 V to 0 V vs. Ag/AgCl at a sweep rate of 50 mV/s in

5 M NaCl, 70 °C, pH 6.5 with a Ti-RDE at a rotation rate of 3000 rpm. Pre-polarization at -1.5 V vs. Ag/AgCl for 5 min.

Figure 2. Potential sweeps from -1.5 V to 0 V vs. Ag/AgCl at a sweep rate of 50 mV s-1 in

5 M NaCl, pH 6.5, 70 °C with a corroded Fe-RDE at a rotation rate of 3000 rpm. Pre-polarization at -1.5 V vs. Ag/AgCl for 5 min.

Measurements of the CE for hydrogen evolution were made to evaluate how hypochlorite reduction as a side reaction could be suppressed at an industrially relevant current density of -3 kA m-2, which corresponds to a potential region where hydrogen evolution prevails. In pH neutral electrolyte the CE on titanium was around 80 % in the presence of 80 mM hypochlorite (Figure 3). When adding 80 mM Mo(VI) the CE increased to 91-94 %. Thus, similar as shown in the potential sweeps (Figure 1), addition of Mo(VI) to a pH neutral electrolyte can hinder hypochlorite reduction on a titanium cathode. In alkaline pH the CE was lower (Figure 3), and some possible explanations for this are (i) a higher and more stable hypochlorite concentration of 80 mM,


190

(ii) a better mass transport of hypochlorite to the cathode surface due to a lower viscosity of the sodium hydroxide electrolyte compared to 5 M NaCl and (iii) at alkaline pH hypochlorite is present as the ion ClO-, which is possibly more easily reduced compared to HClO [12]. The CE was still low after addition of Mo(VI), 55-60%, although there seems to be some minor hindrance of hypochlorite reduction.

Figure 3. Current efficiency measurements for hydrogen evolution on stationary titanium electrode

polarized at -3 kA m-2, 70 °C. Initially: [hypochlorite] ≈80 mM, pH 6.5 respectively pH ≈13

The relatively high concentration of 80 mM Mo(VI) was chosen to see an immediate response of the addition, though at a realistic process a lower concentration is probably needed to avoid high cathode overpotentials and high oxygen levels in the off gas [8]. An experiment with 4 mM Mo(VI) was made, showing that such low concentrations are required over 30 min to give any effect at all on the CE, see Figure 4. The effect would probably be increased if the experiment was continued, giving the Mo-containing cathode film a longer time to build up.

Figure 4. Current efficiency measurements for hydrogen evolution on stationary titanium electrode polarized at -3 kA m-2 1.6 M NaCl, 70 °C. Initially: [hypochlorite] ≈80 mM, pH 6.5


doi: 10.5599/jese.2012.0021 191

As mentioned in the introduction the addition of Cr(VI) to industrial chlorate electrolyte has several functions, one of which is to buffer the electrolyte pH. If replacing Cr(VI) by low levels of Mo(VI), or if lowering considerably the levels of Cr(VI), an additional buffer such as phosphate would be needed. Measurements of the CE on titanium in the presence of phosphate were made to see if possible phosphate films would interfere with the Mo-containing cathode films. As seen in Figure 5 the addition of phosphate had no negative effect on the CE. In contrast, it showed a small positive effect. It has earlier been found that molybdate and phosphate have a synergistic effect on preventing corrosion of an Mg alloy [13].

Figure 5. Current efficiency measurements for hydrogen evolution on a stationary titanium

electrode at -3 kA m-2 in 1.6 M NaCl, 70 °C. Initially: [hypochlorite]≈80 mM, pH 6.5

Results from measurements of CE on a steel electrode can be seen in Figure 6. Lower CE-values were obtained on steel compared to on titanium, which is consistent with our results from the potential sweeps in Figures 1 and 2. Furthermore, also on steel the phosphate additions had no negative effects on the CE, but could possibly increase it a little. The CE values in Figure 6 are relatively low in the beginning of the experiments, and then gradually increase – a pattern not seen as clearly for the titanium electrodes.

Current efficiency for hydrogen evolution at low chromate concentrations

In the chlorate process the product chlorate salt is separated from the electrolyte by crystal-lization, and as the electrolyte contains Cr(VI), small amounts of chromate will co-precipitate or adsorb onto the chlorate crystals and end up in the chlorate product. A CrO4

- concentration of as low as 9 mg dm-3 (80 μM Cr) in the electrolyte has been shown to be enough to form a chromium hydroxide film that hindered the reduction of nitrate and nitrite [14].

Potential sweep experiments were made on titanium in electrolytes with varying Cr(VI) concentrations (Figure 7) to study the effect of low Cr(VI) concentrations in chlorate electrolyte. In the 5 M NaCl electrolyte with 15 mM hypochlorite there was a limiting current for hypochlorite reduction. This electrolyte was Cr(VI) free and there was no inhibition of hypochlorite reduction. A similar sweep was made in chlorate electrolyte made from commercial grade NaClO3 and p.a. grade NaCl, and showed that hypochlorite reduction was inhibited in the presence of 550 g dm-3 NaClO3.


192

Figure 6. Current efficiency measurements for hydrogen evolution on a stationary steel

electrode at -3 kA m-2 in 1.6 M NaCl, 70 °C. Initially: [hypochlorite] ≈80 mM, pH 6.5

From ICP analyses of the chlorate crystals it was calculated that this electrolyte contained also about 20 μM Cr(VI) and this inhibition was reproduced by adding 20 μM Cr(VI) to 5 M NaCl, see Figure 7. Thus, the inhibition of hypochlorite reduction on titanium in chlorate electrolyte without added Cr(VI) was caused by the Cr(VI) traces in non-recrystallized chlorate salt. Experiments were also made with an electrolyte made from NaClO3 that had been recrystallized by cooling crystallization in the lab. For electrolytes with recrystallized chlorate the limiting current for hypochlorite reduction was higher, but the chromate effect was still there. This makes it difficult to evaluate Cr(VI) free alternatives using chlorate salt produced from a conventional industrial process with Cr(VI) containing electrolyte and the present study thus only considers hypochlorite reduction (reactions 6 and 7) and not chlorate reduction (reaction 8).

Figure 7. Potential sweeps (anodic direction) using a Ti-RDE at 3000 rpm and 70 °C in

electrolytes containing 15 mM hypochlorite performed with a sweep rate of 50 mV s-1. Pre-polarization at -1.5 V vs. Ag/AgCl for 5 min.

70

80

90

100

0 10 20 30 40 50

Curr

ent

effc

ienc

y, %

t / min.

80mM Mo(VI), 32mM phosphate

32mM phosphate

80mM Mo(VI)

No addition

-600

-500

-400

-300

-200

-100

0

100

-1,5 -1,2 -0,9 -0,6 -0,3 1,6E-15

i/ A

m-2

E / V vs. Ag/AgCl

5 M NaCl

550 g/dm3 recrystallised NaClO3, 110 g/dm3 NaCl

550 g/dm3 not recrystallised NaClO3, 110 g/dm3 NaCl

5 M NaCl, 20 microM Cr(VI)

5 M NaCl

5.2 M recrystallisedNaClO3, 1.9 M NaCl

5.2 M not recrystallisedNaClO3, 1.9 M NaCl

5 M NaCl, 20 µM Cr(VI)


doi: 10.5599/jese.2012.0021 193

The effect of low levels of Cr(VI) was also studied in CE measurements on a steel electrode, see Figure 8 where a curve representing Mo(VI) additions has been added for comparison. The addition of 2.7 μM Cr(VI) increased the CE, as measured after 20 min of electrolysis, from about 80 % to over 94 %. This can be compared to 80 mM Mo(VI), which only increased the CE to about 83 % despite being present in a concentration about 30000 times higher. Increasing the Cr(VI) level to 27 μM increased the CE to 97 % after 20 min of electrolysis. With the lower Cr(VI) concentration it took over 10 min to reach steady state CE values, whereas for 27 μM Cr(VI) a high CE was reached instantaneously (in less than 1 min). Adding 80 mM Mo(VI) to the Cr(VI) containing electrolyte did not have a negative effect on the CE, but instead a small positive effect. Note that the high CE values obtained in these experiments relate to a steel cathode that had been polished prior to experiments and submerged into the electrolyte under cathodic polarization to avoid steel corrosion. In an earlier study it was found that the selectivity for hydrogen evolution on steel in chlorate electrolytes was highly dependent on whether the steel surface had been allowed to corrode prior to the measurements [11]. Hydrogen evolution was suppressed on corroded steel, in favor of electrochemical chlorate reduction. Compared to non-corroded steel, much higher concentrations of Cr(VI) was needed in the electrolyte to obtain high current efficiencies for hydrogen evolution. Thus, in the present study, if the electrode had been allowed to corrode on open circuit in hypochlorite containing electrolyte, the CE values would have been lower and probably the very low Cr(VI) levels in the present study would have had a negligible effect on the CE. Not only corroded steel but also RuO2 prepared by thermal decomposition have shown a low selectivity for hydrogen evolution in chlorate electrolytes [15]. However, if a smooth and relative stable substrate material, free from oxides catalyzing chlorate reduction, is used the Cr(VI) level in the chlorate process may be lowered substantially while maintaining a high CE. This seems true also for Mo(VI) addition and therefore it will be difficult to replace Cr(VI) in the existing process without replacing the steel cathodes with a more dimensionally stable material. We have earlier shown that Mo(VI) addition can activate a substrate so it will have a similar activity for HER as iron [7], thus less active materials can be used instead of iron/low alloyed carbon steel.

Figure 8. Current efficiency measurements for hydrogen evolution on a stationary steel

electrode at -3 kA m-2 in 1.6 M NaCl at 70 °C. Initially: [hypochlorite]≈80 mM, pH 6.5

70

80

90

100

0,00 10,00 20,00 30,00 40,00

Curr

ent

effic

ienc

y, %

t /min.

27 μM Cr(VI), 80 mM Mo(VI)

27 μM Cr(VI)

2.7 μM Cr(VI), 80 mM Mo(VI)

2.7 μM Cr(VI)

80 mM Mo(VI)

No addition


194

Surface analysis by SEM and EDS

In Figure 9 SEM micrographs of films obtained on a titanium cathode are shown. Figure 9 a, shows the bare titanium surface prior to any contact with the electrolyte and is given for comparison. The other micrographs show the cathode surface after 20 min of electrolysis at -3 kA m-2 in 5 M NaCl with different additives. With 80 mM Mo(VI) in the electrolyte a dense Mo-containing film, covering the cathode surface, was formed (Figure 9 b). The cracks seen are most probably due to drying of the film. With 32 mM phosphate in the electrolyte a totally different surface appears with needle-like crystals, see Figures 9 c and 9 d (larger magnification). In the presence of both Mo(VI) and phosphate in the electrolyte there is again a dense film (figures 9 e and 9 f), now with a spotted appearance.

Figure 9: SEM images of titanium electrodes polarized at -3 kA m-2 for 20 min at 70 oC in 1.6 M NaCl with

100 mM hypochlorite and the following electrolyte additions: 80 mM molybdate (9 b); 32 mM phosphate (9 c-d); 80 mM molybdate + 32 mM phosphate (9 e-f).

The sample in 9 a shows, for comparison, a polished titanium surface not exposed to electrolyte.


doi: 10.5599/jese.2012.0021 195

Elemental analyses with EDS showed that the films formed in the presence of Mo(VI) all contained Mo to a large extent (Table 1). The film formed in the presence of Mo(VI) and phosphate also contained P and was thinner than the film formed when only Mo(VI) was present. Looking at the sample with only phosphate addition, Figures 9 c,d and Table 1 the compounds phosphorous and oxygen were present in a molar ratio of 1:4 and thus phosphorous was likely present as phosphate in the cathode film. For comparison a sample was just dipped, with no cathodic polarization, into the electrolyte with 32 mM phosphate. No needle-like crystals were observed when examining the sample in the SEM and no phosphorous could be detected by EDS (not shown in Table 1).

Table 1. Elemental analysis with EDS on titanium electrodes polarized at -3 kA m-2 for 20 min at 70 oC in 1.6 M NaCl with 100 mM hypochlorite and electrolyte additions according to table.

The values are given in atomic %.

Element *80 mM Mo(VI) *80 mM Mo(VI)+32 mM PO43- **32 mM PO4

3-

O 56.6 61.7 64.9

Ti 14.0 18.4 -

Na 12.7 10.7 5.5

Cl 8.8 2.3 -

Mo 7.5 5.1 -

P - 1.8 15.9

* Acceleration voltage: 15 kV **Acceleration voltage: 5 kV

In an earlier publication we found that if phosphate was added to a Mo(VI)-containing electrolyte, no molybdenum containing films could be detected with EDS on the cathode surface [7]. In that case the phosphate concentration was 2.5-10 times higher than the molybdate concen-tration. In the CE measurements in the present work, 20 times higher molybdate concentration (80 mM Mo(VI)) was used than in the previous study (4 mM Mo(VI), Ti-RDE 3000 rpm, and no added hypochlorite [7]) and therefore molybdate can be favored in a competitive adsorption between molybdate and phosphate.

In the presence of both molybdate and phosphate in the electrolyte there were also some larger circular spots on the electrode surface (Figure 10). The spots were around 10 μm in diameter and darker than the surrounding film (image taken in back-scatter mode). EDS analysis showed that the molybdenum level was higher in the spots that in other areas of the film. The spot can be an elevated structure, dramatically increasing film thickness.


196

Figure 10. SEM image in back-scatter mode of a titanium electrode polarized at -3 kA m-2 for

20 min at 70 oC in 1.6 M NaCl with 100 mM hypochlorite, 80 mM Mo(VI) and 32 mM phosphate.

Effects of molybdate on the anode potential

Since chlorate cells are undivided, the anode will encounter the same electrolyte as the cathode. Thus any effects on the anode reactions that the electrolyte additives may have are important to study, and Mo(VI) at high concentrations have shown to increase the oxygen levels in the off gas [6],[8]. Oxygen may form from several different reactions on the anode or in homogeneous reactions in the electrolyte [1] and experiments were made to evaluate if the high oxygen values are accompanied by an increase in anode potential. The anions Cr2O7

2-, PO43- and

NO3- have been observed to increase the oxygen levels in the off gas in chlorate electrolysis,

probably caused by anodic oxygen evolution and the likely mechanism suggested was an adsorption of the anions on the anode surface [16]. In another study an increased anode potential of about 15 mV was observed when increasing the Cr(VI) concentration from 3 to 9 g dm-3 Na2Cr2O7 for a DSA in an electrolyte of pH 2 [10].

To study anode effects of Mo(VI), RuO2/TiO2 electrodes produced by spin coating and shaped into RDEs were polarized at 3 kA m-2 for 30 min with and without 40 mM Mo(VI) in the electrolyte. The addition of Mo(VI) resulted in an increase in anode potential around 40 mV, see Figure 11. Using SEM, no clear morphology change due to film formation or deposits could be seen on the RuO2/TiO2 surfaces, although Mo/Na ratios 2-10 times larger than expected from electrolyte composition were measured with EDS. Also XRF measurements indicated increased molybdenum content on the surface. Furthermore predominance diagrams show MoO3·2H2O(s) as the dominant phase in 10 mM Mo(VI) solution if the conditions are acidic and oxidizing [17], such as close to the anode. It is thus likely that the increased oxygen is caused by precipitation or adsorption of Mo-containing species on the anode surface.


doi: 10.5599/jese.2012.0021 197

Figure 11. Galvanostatic polarizations of DSA-RDEs (spincoated 3-layers) at 3 kA m-2 in

2 M NaCl, pH 7, 70°C at a rotation rate of 3000 rpm.

Conclusions

The presence of Mo(VI) ions in an electrolyte of near neutral pH can increase the selectivity for hydrogen evolution as the Mo-containing film formed on the cathode surface hinders the side reaction of hypochlorite reduction.

Potential sweeps and current efficiency experiments in neutral electrolytes show that on a smooth substrate such as polished titanium the suppression of hypochlorite reduction by Mo(VI) addition is more efficient than on steel or on corroded iron.

Even very low levels of Cr(VI), in the μM range, can efficiently suppress hypochlorite reduction on polished titanium and steel. Thus if using smooth and stable cathode materials the chromate concentration on the chlorate process may be substantially reduced compared to the levels used today (1-6 g dm-3 Na2Cr2O7 corresponding to about 10-50 mM Cr(VI)). Note though that the reaction of electrochemical chlorate reduction has not been addressed in the present study. The inhibiting action of very low Cr(VI) concentrations also means that when studying alternatives to Cr(VI) in the chlorate process possible Cr(VI) present in the chlorate salt used to make the electrolyte in the experiments should be considered. Mo(VI) on the other hand needed higher concentrations and longer polarization times to show an effect on the current efficiency.

Phosphate, a possible buffer needed in the chlorate process if removing, or substantially lowering the concentration of, Cr(VI) does not negatively influence the current efficiency in the presence of Cr(VI) or Mo(VI). Instead a slight increase in current efficiency was noticed when adding 32 mM NaH2PO4 to the electrolyte. The Mo-containing cathode films formed during cathodic polarization become thinner if the electrolyte during the film build-up also contained phosphate, but the films still appeared as effective at hindering hypochlorite reduction.

The presence of Mo(VI) in the electrolyte can increase the potential of a RuO2-TiO2 anode. The anode potential increased 40 mV after adding Mo(VI) corresponding to 40 mM, a concentration that in a previous study [6] resulted in increased oxygen levels in the off gas. Thus the increased oxygen production is likely produced in an anodic side reaction promoted by the adsorption of Mo(VI) species on the anode.


198

Considering increased oxygen levels and increased cathode and anode potentials, if Mo(VI) is to be used in the chlorate process it should be used at a low concentration to minimize the negative side effects.

Acknowledgements: The Swedish Energy Agency, the Swedish Research Council, Permascand, Eka Chemicals and the European Regional Development Fund through the project “EnergyWise” are acknowledged for financing this study.

References

[1] J.E. Colman, American Institute of Chemical Engineers Symposium Series 77 (1981) 244-263 [2] G. Lindbergh and D. Simonsson, Electrochim. Acta 36 (1991) 1985-1994 [3] A. Cornell, G. Lindbergh and D. Simonsson, Electrochim. Acta 37 (1992) 1873-1881 [4] M.S. Vukasovich, Polyhedron 5 (1986) 551-559 [5] F.J. Presuel-Moreno, M.A. Jakab and J.R. Scully, J. Electrochem. Soc. 152 (2005) B376-B387 [6] M. Li, Z. Twardowski, F. Mok, N. Tam, J. Appl. Electrochem 37 (2007) 499–504 [7] J. Gustavsson, C. Hummelgård, J. Bäckström, I. Odnevall Wallinder, S. Mohammed Habibur

Rahman, G. Lindbergh, S. Eriksson and A. Cornell, J. Electrochem. Sci. Eng. 2(3) (2012) 105-120

[8] M. Rosvall, K. Hedenstedt, A. Sellin, J. Gustavsson and A. Cornell, (Akzo Nobel Chemicals International B.V.), WO 2010/130546A1 (2010)

[9] N. Krstajic, V. Jovic and G.N. Martelli, (Industrie De Nora S.p.A.), WO 2007/063081A2 (2007) [10] A. Cornell, B. Håkansson and G. Lindbergh, J. Electrochem. Soc. 150 (2003) D6-12. [11] J. Wulff and A. Cornell, J. Appl. Electrochem. 37 (2007)181. [12] P. Byrne, E. Fontes, O. Parhammar and G. Lindbergh, J. Electrochem. Soc. 148 (2001) D125-

D132 [13] Z. Yong, J. Zhu, C. Qiu and Y. Liu, Appl. Surf. Sci. 255 (2008) 1672–1680 [14] H.A. Duarte, K. Jha and J.W. Weidner, J. Appl. Electrochem. 28 (1998) 811–817 [15] A. Cornell and D. Simonsson, J. Electrochem. Soc. 140 (1993) 3123-3129 [16] M.M. Jakšić, A. Despić, B. Nikolić and S. Maksić, Croat. Chem. Acta 44 (1972) 61-66 [17] P. Wang, L. L. Wilson, D. J. Wesolowski, J. Rosenqvist and A. Anderko, Corr. Sci. 52 (2010)

1625-1634



doi: 10.5599/jese.2012.0023 199




A new amperometric glucose biosensor based on screen printed carbon electrodes with rhenium(IV) - oxide as a mediator

ALBANA VESELI, AHMET HAJRIZI*, TAHIR ARBNESHI, KURT KALCHER*

Department of Chemistry, Faculty of Natural Science, University of Prishtina, Nëna Terezë, 10000, Prishtina Kosovo

*Institute of Chemistry, Analytical Chemistry, Karl-Franzens University of Graz, Universitätsplatz 1, 8010 Graz, Austria Corresponding Author: E-mail: [email protected]

Received: July 09, 2012; Revised: September 14, 2012; Published: November 10, 2012

Abstract Rhenium(IV)-oxide, ReO2, was used as a mediator for carbon paste (CPE) and screen printed carbon (SPCE) electrodes for the catalytic amperometric determination of hydro-gen peroxide, whose overpotential for the reduction could be lowered to -0.1 V vs. Ag/AgCl in flow injection analysis (FIA) using phosphate buffer (0.1 M, pH=7.5) as a carrier. For hydrogen peroxide a detection limit (3σ) of 0.8 mg L-1 could be obtained. ReO2-modified SPCEs were used to design biosensors with a template enzyme, i.e. glucose oxidase, entrapped in a Nafion membrane. The resulting glucose sensor showed a linear dynamic range up to 200 mg L-1 glucose with a detection limit (3σ) of 0.6 mg L-1. The repeatability was 2.1 % RSD (n = 5 measurements), the reproducibility 5.4 % (n = 5 sensors). The sensor could be applied for the determination of glucose in blood serum in good agreement with a reference method.

Keywords Biosensor; glucose; hydrogen peroxide; rhenium(iv)-oxide; blood serum

Introduction

Nowadays, sensors and biosensors research, an ever expanding field of analytical chemistry, has been attracting scientists from many related disciplines such as biology, material sciences etc. [1]. Chemical sensors and biosensors are offering alternative solutions, capable of satisfying the increasing demand for precise and fast analytical information through devices that require relatively simple instrumentation [2-4].

J. Electrochem. Sci. Eng. 2(4) (2012) 199-210 AMPEROMETRIC GLUCOSE BIOSENSOR

200

Electrochemical biosensors often rely on the use of oxidases or dehydrogenases as biological recognition element [5-8]. When using oxidases, hydrogen peroxide is often formed as an electrochemically active intermediate, whose detection is the crucial point of the design of corresponding biosensors.

H2O2 can be oxidized or reduced at rather high positive or negative potentials with inherent risk to co-oxidize or co-reduce components of complex matrices, such as blood, urine or other biological fluids or extracts. In order to overcome the problem, mediators are used which react chemically with hydrogen peroxide and reduce high over potentials [6-9].

In many cases transition metal compounds are used which may change easily their oxidation states such as hexacyanoferrates [8-17], platinum metals or their oxides [18], and manganese dioxide [6-9,19-21].

Rhenium shows quite a few oxidation states which could be involved in a reaction cycle where hydrogen peroxide is involved [22]. The main goal of this work was to investigate the electroanalytical behavior of ReO2 in terms of acting as a catalyst for the determination of hydrogen peroxide as a basis for oxidase - based biosensors. In particular, glucose oxidase was used as a template enzyme.

Experimental

Apparatus

Batch cyclic voltammetry and chronoamperometric measurements were performed using a potentiostat (PalmSens, Electrochemical Sensor Interface) connected to a laptop computer. Carbon paste electrodes CPEs (modified and unmodified) were used as working electrodes. As a reference electrode an Ag/AgCl electrode was used (3M KCl). All potentials referred to in this paper are against this reference electrode. A platinum wire served as a counter electrode.

The flow injection system consisted of a high performance liquid chromatographic pump (510 Waters, Milford MA, USA) in connection with a system controller (Waters 600E), a sample injection valve (5020 Rheodyne, Cotati, CA, USA), and a thin layer electrochemical detector (LC 4C, BAS, West Lafayette, Indiana, USA) with a flow through cell (spacer thickness 0.19 mm; CC-5, BAS) in combination with an electrochemical workstation (BAS 100B). SPCE was used as the working electrode and an Ag/AgCl-electrode (3 M KCl) as the reference. The steel back plate of the thin layer cell served as the auxiliary electrode.

Reagents and solutions

Phosphate buffer (0.1 mol L-1) was prepared by mixing aqueous solutions (0.1 mol·L-1) of sodium dihydrogen phosphate and disodium hydrogen phosphate to produce solutions of the required pH (7.5).

A stock solution of hydrogen peroxide (1000 mg L-1) was prepared freshly every day and solutions of lower concentrations were prepared immediately before use.

Also a stock solution of glucose (1000 mg L-1) was prepared with the corresponding working buffer solution, kept at room temperature overnight to facilitate α-ß-mutarotation, and stored at 4 oC when not in use. Lower concentrations were prepared immediately before use. All chemicals were of analytical purity grade (p.a., Fluka).

A. Vesel et al. J. Electrochem. Sci. Eng. 2(4) (2012) 199-210

doi: 10.5599/jese.2012.0023 201

Preparation of working electrodes

Carbon paste electrodes (unmodified); 1 g graphite powder and 360 µL paraffin oil were mixed in an agate mortar by gently stirring with a pestle until uniformity and compactness. For modified CPE 50 mg of ReO2 were added per gram of graphite powder.

Screen printed carbon electrodes (modified): 0.05 g of rhenium(IV)-oxide was added to 1 g carbon ink (Electrodag 421 SS, Acheson); the oxide-ink mixture was stirred for 10-30 minutes with a glass-rod or stainless steel spatula and finally sonicated for 30 minutes in an ultrasonic bath (Transsonic 700/H, Elma®). The resulting mixture was immediately used for electrode fabrication using the semi-automatic screen printer (SP-200, MPM, MA-USA) and aluminum oxide support. The printed electrodes were dried at room temperature overnight before using for measurements.

To prepare enzyme casting solutions, required volumes of ethanol (80 µL), Nafion (5 % w/w in lower alcohols, 40 µL), water (80 µL) and GOD solution (5 % w/w, 40 µL) were mixed in the order listed in a plastic vial (1.5 mL micro centrifuge tubes, 616.201, Greiner Labor Technic). Five μL (unless otherwise specified) of the resulting mixture were directly applied onto the active area of the SPCE surface (≈0.40 cm2 area) and air dried. The electrode was inserted into the thin layer cell; electric contact was made with a crocodile clamp.

Procedures

Cyclic voltammograms were recorded from an initial potential of 0.80 V to a vertex potential of -1.00 V. The scan rate was 20 mV s-1; usually three cycles were recorded.

Hydrodynamic amperograms were recorded at potentials from -500 mV to 100 mV in increments of 100 mV. Flow injections analyses were done at a potential of -100 mV if not stated otherwise. The flow rate of the pump was 0.40 mL min-1 and the injection volume of hydrogen peroxide and glucose solutions was 200 µL.

Analyses of samples

Blood samples were taken manually with a syringe and put in the plastic tubes containing EDTA (1.9 mg per mL of blood) as an anticoagulant. 1 mL of each sample was transferred into a 10 mL sterile PP-tube, diluted 10 times and centrifuged for 20 minutes at 4500 rpm. From the centrifuged supernatant serum the analyte solutions were prepared by diluting 40 and 100 times with phosphate buffer (0.1 M, pH 7.5). Measurements were done in FIA mode using the standard addition method by sequentially adding three times an amount of glucose standard (100 mg L-1) in phosphate buffer solution (0.1 M, pH 7.5), which corresponds roughly to half of the original glucose amount in the sample. After each addition the sample was re-measured. A commercial glucometer Ascensia BRIO was used for glucose reference measurements using whole blood.

Results and discussion

The studies were done with a few working electrodes: unmodified and ReO2 - modified carbon paste electrodes (CPE) and screen printed electrodes (SPCE) modified with ReO2 as a mediator and glucose oxidase as a biocomponent (entrapped in a Nafion film).


202

Studies with Hydrogen peroxide

Cyclic Voltammetry

H2O2 is electrochemically active with carbon paste electrodes, but shows high overpotentials for its reduction and also its oxidation as can be seen in Fig. 1. At positive potentials H2O2 does not show any particular electrochemical activity at unmodified carbon paste electrodes.

Figure 1. Cyclic voltammograms of H2O2 at an unmodified CPE; phosphate buffer solution 0.1 M,

pH 7.5;Fill curve blank, broken curve with 200 mg L-1 H2O2; conditions: Einitial = 0.80V, Efinal =-1.00V, scan rate 20 mV s-1.

At negative potentials, similar behavior can be noticed. Direct reduction of hydrogen peroxide occurs at rather negative potentials only.

According to our knowledge, rhenium (ReO2) has not yet been used as a modifier for carbon pastes or carbon inks.

Figure 2.Cyclic voltammogram of a ReO2-modified CPE; Einitial = 0.80 V, Efinal = -1.0 V; scan rate 20 mV s-1; support electrolyte 0.1 M phosphate buffer, pH 7.5.


doi: 10.5599/jese.2012.0023 203

Figure 2 shows the cyclic voltammogram of a rhenium(IV)-oxide - modified carbon paste electrode. No distinct peaks can be discerned in the investigated potential range. Reduction occurs at positive potentials already and dominates at more negative values. Re-oxidation on the other hand begins at slightly negative potentials already, but in fact does not get dominant even up to 0.7 V. The reduction current can be assigned to the reduction of Re(IV) to Re(III) (probably present as Re2O3 in slightly alkaline medium or even as RePO4). Re-oxidation in phosphate buffer solution seems electrochemically rather irreversible because it is much smaller than reduction showing a very broad signal extending from -0.3 V onward.

In the presence of hydrogen peroxide a small additional reduction current occurs in cyclic voltammetry (Fig.3).

Figure 3 Cyclic voltammograms of a CPE modified with ReO2 in the absence and in the presence of hydrogen peroxide; H2O2 200 mg L-1; conditions: E initial =0.80V, E final = -1.00V, scan rate 20

mV/s, support electrolyte 0.1 M phosphate buffer, pH 7.5.

Although the effect of the modifier is small in CV (but more pronounced in amperommetry, see below) it may be noted that in fact a catalytic effect on the reduction of H2O2 exists.

Basically two mechanisms can be assumed: rhenium(IV)-oxide (in the oxidation state +4) is reduced to trivalent rhenium (Re2O3) which is oxidized to ReO2 by hydrogen peroxide again (mechanism A, Fig.1). Thus, virtually more modifier (ReO2) is present at the surface creating a higher reduction current.

ReO2

Re O2 3 H O2 2

H O2e-

reductioncurrent A

ReO3

4

orReO-

ReO2 H O2 2

H O2e-

reductioncurrent B

Fig.4. Models for the catalytic action of ReO2 on the reduction of H2O2; catalytic redox cycle of rhenium(IV)-oxide via trivalent rhenium and the action of H2O2 (A); catalytic redox cycle of

rhenium(IV)-oxide via hexa- or heptavalent rhenium and the action of H2O2 (B).


204

On the other hand it is basically also possible that ReO2 is oxidized to ReO3 or even perrhenate, ReO4

-, a reaction which is known to proceed rather slowly in aqueous solution. In this case hydrogen peroxide would react with the original modifier, ReO2, first, and then the reduction of the Re(VI) or Re(VII)-species would occur (mechanism B, Fig.4). According to the fact that at negative potentials, where catalytic action can be noticed, reduction of ReO2 is observable at the modified electrode alone, mechanism A seems more probable.

Hydrodynamic Amperommetry

Figure 5 summarizes the hydrodynamic voltamperogram of the catalytic current caused by ReO2 as a modifier compared with an unmodified CPE. The measerments were performed with stirred solutions as batch experiments, and the current change after addition of hydrogen peroxide was recorded. High signals can be obtained at potentials beyond -400 mV, but at such values the risk of co-reduction of other compounds in complex matrices is also high. Nevertheless, at a potential of -100 mV there are significant current responses observable already, which can be exploited for quantitative analytical purposes. In fact, even higher operation potentials up to +100 mV could be used, but the current is significantly lower compared to the signal at -100 mV. With unmodified electrodes reduction currents of H2O2 can be observed under the given experimental conditions only at -400 mV and below (Fig.5 curve b).

Figure 5. Reduction current of H2O2 in hydrodynamic voltamperometry at a CPE modified with ReO2 (a) compared to an unmodified CPE (b); supporting electrolyte sodium phosphate buffer

(0.1 M, pH 7.5); hydrogen peroxide concentration 50 mg L-1.

Flow Injection Analysis (FIA)

FIA was used to study the effect of ReO2 on the reduction of H2O2 in detail and to optimize the experimental parameters.

As screen printed carbon electrodes show higher robustness towards mechanical and chemical stress, but are otherwise comparable to carbon paste sensors due to their heterogeneous charac-ter concerning the electrode material, this type of electrode was used for further studies. Figure 6 shows a typical amperogram with injection of hydrogen peroxide using a ReO2-SPCE as a detector.


doi: 10.5599/jese.2012.0023 205

Figure 6. Amperogram obtained by FIA with a SPCE modified with ReO2; injection volume of H2O2 solution 200 µL; operating potential -100 mV; flow rate 0.4 mL min-1; carrier phosphate

buffer (0.1 M, pH 7.5).

The rhenium(IV)-oxide modified electrode responds very well and reproducible to hydrogen peroxide concentrations even in the low mg L-1-range.

For optimization of the operating potential a hydrodynamic voltammperogram was recorded under FIA conditions (Fig.7).

Figure 7. Dependence of the peak height of H2O2 on the working potential in FIA with SPCEs

modified with ReO2 as detector, flow rate 0.4 mL min-1, carrier 0.1 M phosphate buffer, pH 7.5, H2O2-solution 100 mg·L-1; injection volume 200 µL.

Reduction currents can be observed up to even 200 mV. When decreasing the potential to more negative values the signal (peak height of the reduction current) increases. Apart from the reasons discussed already (increased risk of interference with increasing negative potentials) there is another disadvantage connected with highly negative operating potentials. The background


206

current strongly increases below -200 mV (not show) and even tends to vary during operation that the repeatability is significantly deteriorated. The signal-to-background ratio gets unfavorably low because the latter exceeds even the height of the signal. For these reasons -100 mV was chosen as the most favorable operating potential.

Figure 8. Calibration curve for H2O2 within concentration 1-1000 mg·L-1at modified SPCE at -

100 mV. Supporting electrolyte phosphate buffer 0.1 M, pH = 7.5.

A corresponding calibration curve for hydrogen peroxide is displayed in Fig. 8. The linear dynamic range of the sensor (concentration range with linear relation between signal and concentration) is restricted to the lower concentrations of the analyte. Linearity between concentration of hydrogen peroxide and signal exists from 0.6 to 20 mg L-1, with a correlation coefficient over 0.99.

The detection limit (3σ), estimated from the standard deviation of FIA-peaks at 5 mg L-1 H2O2, is 0.2 mg L-1.

The repeatability of measurements for 100 mg L-1 is 3 % RSD (n = 5 measurements), and the reproducibility for 100 mg L-1 at different working electrodes (n = 5 electrodes) is 5.7 %.

Studies with glucose

In order to check if the ReO2 modified electrode may serve as a basis for the design of first generation- biosensors with oxidases producing H2O2, glucose oxidase was used as a template enzyme. It was immobilized on the surface of the ReO2-modified SPCEs using a Nafion membrane [23]. The sensor responds well to injections of glucose under FIA conditions.

A typical calibration curve of the biosensor is shown is shown in Figure 9. The range with a linear relation between current and concentration of glucose was found to

extend up to 100 mg L-1 (Fig. 9 insert) with a correlation factor R2=0.9999. The linear regression function is given in eqn. (1).

i / nA= 0.4867 c / mg L-1 + 2.4184 (1)


doi: 10.5599/jese.2012.0023 207

Figure 9. Calibration of glucose with a glucose oxidase/ReO2-modified SPCE; working potential:

-100 mV; flow rate: 0.4 mL/min; 100 µL injection volume.

The detection limit (3σ) estimated from the standard deviation of FIA-peaks for 20 mg L-1 glucose concentration is 0.6 mg L-1. The repeatability of measurements for 200 mg L-1 glucose was 2.1 % RSD (n = 5 measurements), and the reproducibility was 5.4 % RSD (n = 5 electrodes).

Table 1. Comparison of limit of detection of different used modifiers

Modifiers Detection limit of H2O2 Detection limit of glucose

References mg L-1 mg L-1

IrO2 0.24 0.9 [26]

Fe3O4 0.2 0.5 [23]

MnO2 0.045 0.087 [24]

SnO2 15 6.8 [27]

PtO2 0.03 2.0 [26]

CuO - 0.03 [23, 25]

ReO2 0.2 0.6 this work

PdO 0.8 0.83 [26]

Table 1 compares the LOD values for sensors and biosensors modified with different metal oxides for the detection of H2O2 and glucose. It may be stated that ReO2 is capable of detecting very low concentrations of hydrogen peroxide and glucose; it is comparable to Fe3O4 [23] and super cedes platinum metal oxide [26].

Interferences

Four common interferences, vitamin C, paracetamol, gentisic acid, and uric acid were tested if they interfere with the detection of glucose using the ReO2-based biosensor. All investigated interferents give significant responses in the investigated concentration ratio. Nevertheless, their concentration in blood may be expected to be much smaller so that the extent of influence on the signal of the analyte, glucose, is expected to be negligible (Table 2).


208

Table 2. Relative signals of possible interferences in the current response of the biosensor in the absence of glucose; 100 % corresponds to the current of 200 mg L-1 glucose. Flow rate: 0.40 mL min-1, operating

potential -100 mV vs. Ag/AgCl; carrier: phosphate buffer (pH=7.5; 0.1 M), injection volume 100 µL

Interferent 50 mg L-1 100 mg L-1 500 mg L-1

% % %

Ascorbic acid 46.1 95.6 304

Uric acid 38.0 45.4 166.0

Paracetamol 21.8 21.5 34.7

Gentisic acid 19.2 26.7 61.6

Analysis of blood plasma samples

The glucose biosensor described in this work was used to determine the glucose level in human serum. Two samples (ST1, SA1) were investigated. Analyses were done with the standard addition method basically for two reasons: first to exclude effects of the matrix and interferences, and second to investigate if the linear relation between signal and different concentrations of glucose in the plasma is maintained.

Glucose concentrations obtained with the biosensor (blood plasma) were multiplied with a factor of 1.15 in order to convert them to whole blood concentrations [28, 29].

The results between the method employing the new biosensor and the reference are in very good agreement (Table 3). Sample SA1 is blood from a healthy person in the morning, sample ST1 from the same person after eating a piece of sweet cake.

Table 3 Comparison of glucose concentrations of two human serum samples measured using the new biosensor and the pocket Glucometer Ascensia Brio.

Sample ReO2 –based biosensor Glucometer Ascensia Brio

SA1 993 ± 26 mg L-1 980 ± 10 mg L-1 mL

ST1 1396 ± 32 mg L-1 1360 ± 20 mg L-1 mL

Conclusion

The work presented here has clearly demonstrated that heterogeneous carbon sensors (carbon paste, screen printed carbon electrodes) with rhenium(IV)-oxide as a mediator exhibit good performance for the determination of hydrogen peroxide because the modifier lowers the overpotential of the analyte.

A biosensor based on thick film technology with glucose oxidase as a template enzyme was developed. When using the screen-printed biosensor with flow injection analysis using phosphate buffer (0.1 M, pH 7.5) as a carrier with a flow rate of 0.4 mL min-1, a detection limit of 0.2 mg L-1 glucose could be achieved with a linearity range up to 20 mg L-1. Thus, glucose can be determined in blood with this method.

The influence of possible interferences (ascorbic acid, uric acid, paracetamol, and gentisic acid) on the determination of glucose was estimated. The extent of all investigated interferences is not fatal for the determination of glucose in human blood plasma.


doi: 10.5599/jese.2012.0023 209

Acknowledgements: A.V. wishes to acknowledge support for this work by the ÖAD-Büro für Austauschprogramme, Österreichische Rektorkonferenz and World University Service-Graz, Austria, for a financial support. The authors are thankful to CEEPUS project CII-CZ-0212-02-0809-M-28727 for mobility grants.

References

[1] X. Zhang, H. Ju, J. Wang, Electrochemical Sensors, Biosensors and Their Biomedical Applications, Academic Press, New York, USA, 2008.

[2] V. K. Rao, M. K. Sharma, P. Pandey, K. Sekhar, World J. Microbiol. Biotechnol. 22 (2005) 1135-1143.

[3] K. Kalcher, Electroanal. 2 (1990) 419-433. [4] K. Kalcher, J. M. Kauffmann, J. Wang, I. Svancara, K. Vytras, C. Neuhold, Z. Yang,

Electroanal. 7 (1995) 5-22. [5] L. Gorton, Electroanal. 7 (1995) 23-45. [6] K. Kalcher, I. Svancara, R. Metelka, K. Vytras, A. Walcarius, In Encyclopedia of Sensors, C. A.

Grimes, E. C. Dickey and M. V. Pishko Ed(s)., American Scientific Publishers, Vol 4, (2006) 283-430.

[7] I. Svancara, K. Vytras, K. Kalcher, A. Walcarius, J. Wang, Electroanal. 21 (2009) 7-28. [8] I. Svancara, K. Kalcher, A. Walcarius, K. Vytras, Electroanalysis with carbon paste electrodes,

Taylor & Francis/CRC Press, Boca Raton, FL, USA, 2012. [9] ETurkusic, K. Kalcher, E. Kahrovic, N. Beyene, H. Moderegger, E. Sofic, S. Begic, K. Kalcher,

Talanta 65 (2005) 559-564. [10] R. Garjonyte, A. Malinauskas, Sens. Actuators B 46 (1998) 236–241. [11] K. Itaya, I Uchida, S. Toshima, Nippon Kagaku Kaishi 11 (1984) 1849–1853. [12] A. Boyer, K. Kalcher, R. Pietsch, Electroanal. 2 (1990) 155–161. [13] M. Weissenbacher, K. Kalcher, H. Greschonig, W. Ng, W.-H. Chan and A. N. Voulgaropoulos,

Fresen. J. Anal. Chem. 344 (1992) 87–92. [14] M. I. Gomez de Rio, C. De la Fuente, J. A. Acuna, M. D. Vazquez Barbado, M. Tascon Garcia,

S. de Vicente Perez, P. Sanchez Batanero, Quim. Anal. (Barcelona) 14 (1995) 108–111. [15] D. Moscone, D. D’Ottavi, D. Compagnone, G. Palleschi, A. Amine, Anal. Chem. 73 (2001)

2529–2535. [16] V. M. Ivama, Sh. H. Serrano, J. Brazil Chem. Soc. 14 (2003) 551–555. [17] J. Li, X. Wei, Y. Yuan, Sens. Actuators B 139 (2009) 400–406. [18] M. S. Lin, J. S. Lai, J. Wang, Huaxue 60 (2002) 483–493. [19] G. A. P. Zaldivar, Y. Gushikem, J. Electroanal. Chem. 337 (1992) 165–174. [20] J. Wang, N. Naser, L. Angnes, H. Wu, L. Chen, Anal. Chem. 64 (1992) 1285–1288. [21] N. W. Beyene, P. Kotzian, K. Schachl, H. Alemu, E. Turkusic, A. Copra, H. Moderegger, I.

Svancara, K. Vytras, K. Kalcher, Talanta 64 (2004) 1151–1159. [22] N. N. Greenwood, A. Earnshaw, Chemistry of the Elements, 2nd edition, Butterworth-

Heinemann, Oxford, UK, 1997. [23] T. T. Waryo, S. Begic, E. Turkusic, K. Vytras, K. Kalcher, Scientific Papers of University of

Pardubice, Series A 11 (2005) 265-279. [24] K. Schachl, H. Alemu, K. Kalcher, J. Jezkova, I. Svancara, K. Vytras, Analyst 122 (1997) 985-

989. [25] L. Liu, J. F. Song, P. F. Yu, B. Cui, Talanta 71 (2007) 1842–1848. [26] P. Kotzian, P. Brazdilova, K. Kalcher, K. Handlir, K. Vytras, Sens. Actuators B 124 (2006) 297-

302. [27] L. Berisha, Master Thesis, University of Prishtina, Kosovo, 2009.


210

[28] American Diabetes Association Diabetes Care. Standards of Medical Care, Vol. 29 Supplement 1, (2006) 51–580.

[29] K. L. Tieszen, 1st German Diabetes Diagnostics Conference, diabetes Stoffwechsel and Herz, Kohn, Germany, May 5-7, 2006.



doi: 10.5599/jese.2012.0016 211




Electrochemical extraction of oxygen using PEM electrolysis technology BOULBABA ELADEB, CAROLINE BONNET, ERIC FAVRE and FRANÇOIS LAPICQUE

Laboratoire Réactions et Génie des Procédés, CNRS- Université de Lorraine, ENSIC, 1 rue Grandville, F-54001 Nancy, France Corresponding Author: E-mail: [email protected]; Tel.: 33 (0) 383 17 52 66

Received: March 13, 2012; Revised: July 12, 2012; Published: November 10, 2012

Abstract Electrochemical extraction of oxygen from air can be carried out by chemical reduction of oxygen at the cathode and simultaneous oxygen evolution by water anode oxidation. The present investigation deals with the use of an electrolysis cell of PEM technology for this purpose. A dedicated 25 cm2 cell provided with a commercial water electrolysis MEA and titanium grooved plates has been designed for continuous operation at pressures close to the ambient level. The MEA consisted of a Nafion 117 membrane sandwiched between a Pt/C cathode and a non-supported Pt-Ir anode. Oxygen partial consumption in long-term runs was evaluated by analysis of the outlet air by gas chromatography, depending on the cell voltage - or the current density - and the excess in air oxygen fed to the cathode. Runs over more 50 hours indicated the relative stability of the components used for current densities ranging from 0.1 to 0.2 A cm-2 with high efficiency of oxygen reduction. Higher current density could be envisaged with more efficient MEA’s, exhibiting lower overpotentials for oxygen evolution to avoid too significant degradation of the anode material and the membrane. Interpretation of the data has been carried out by calculation of the cathode current efficiency.

Keywords Oxygen extraction; PEM electrolysis; electrode stability; Pt-Ir anode; current efficiency

1. Introduction

Extraction of oxygen from air can be used for various applications, as follows. For instance oxygen impoverished air reduces the oxidation content from the atmosphere over food for the sake of higher preservation in its packing. Lower oxygen contents in air also reduce the corrosion rate at non-noble metals pieces. Another potential application is the preparation of the gas phase to be injected to biological reactors – upon addition of carbon dioxide – for the possible growth of

J. Electrochem. Sci. Eng. 2(4) (2012) 211-221 EXTRACTION OF OXYGEN USING PEM TECHNOLOGY

212

microorganisms such as bacteria, fungi or microalgae or undifferentiated cells of higher-grade vegetal organisms in dedicated bioreactors.

Various techniques for oxygen removal from a gas phase to be used for biological applications or food processing can be listed as follows. The easiest technique, which can be qualified as indirect, simply consists of adding of pure nitrogen into the air stream: the cost of pressured nitrogen renders the technique somewhat expensive for practical applications. Conventional, non electrochemical techniques for oxygen removal rely upon the following various principles: (i) oxygen combustion by a gas microburner, however other gases of significant safety or environment impact such as carbon monoxide or nitrogen oxides can be formed; (ii) selective reaction of oxygen upon addition of reducing agents e.g. sulfites that have been used for decades for food preservation: however the sulfate produced by oxygen oxidation of sulfite have to be removed from the medium; (iii) selective adsorption on a suitable compound which has nevertheless to be regenerated for its possible reuse; (iv) membrane separation of oxygen from carbon dioxide and nitrogen at temperatures close to ambient: however, because of the intermediate behaviour of oxygen in the three-species, this technique cannot be used for biological applications.

Besides, various electrochemical techniques can be used for oxygen extraction from an air flow, as reviewed by Winnick [1]. In a pioneering paper, Langer and Haldermann [2] showed that oxygen could be extracted from air and recovered in the form of pure oxygen in a two-compartment electrolytic cell, in the presence of acidic or alkaline solutions and separated by a conventional cationic membrane:

Cathode O2 + 4H+ + 4e → 2 H2O (1)

Anode 2 H2O → O2 + 4H+ + 4e (2)

The equilibrium voltage of such a cell is nil, assuming similar conditions of H+ activity and oxygen pressure in the two compartments; the cell voltage is the sum of the two overpotentials –in absolute value- and the ohmic drop of the cell. The two reactions involved are nevertheless slow, even at platinum surfaces, and the cell voltage attains easily 1 V even at low current densities. Since then, the technique was improved by Wynween and Montgomery [3]. In the eighties, Fujita et al. [4] improved the efficiency of the technique by using a Nafion membrane, an air cathode whereas the anode was platinum particles deposited on the membrane surface: this technology derived from PEM fuel cell technology allowed current density up to 200 mA cm-2 for cell voltage at 1.4 V for continuous operation: the corresponding energy consumption for the oxygen extraction was near 4.7 kWh kg-1 O2. It can be observed that the process consumes less energy than the conventional electrolysis, with specific energy consumption near 6.7 kWh kg-1 O2 for a cell voltage of 2 V, even though the current densities of the two processes differ quite a lot. More recently, General Electrics [5] developed an oxygen supply system for high-altitude aircraft relying upon similar technology: oxygen at high pressures could be produced with cell voltage claimed to be near 1 V at 125 mA cm-2.

The electrochemical separation of oxygen can be carried out upon specific diffusion in mixed, non stoichiometric oxides exhibiting oxygen vacancies in the lattice, as described in several patents [6-8]. The technology used derived from solid oxide cells: oxygen is reduced to O2- ions at the cathode and migrates to the anode for back oxidation to oxygen; temperatures higher than 500 °C are required for sufficient diffusivity of O2- ions in the non-stoichiometric oxide mixture. However the current density is usually below 100 mA cm-2.

B. Eladeb et al. J. Electrochem. Sci. Eng. 2(4) (2012) 210-221

doi: 10.5599/jese.2012.0016 213

Oxygen extraction by its reduction to hydrogen peroxide on a graphite-based cathode through exchange of two electrons only was imagined by Tseung and Jasem [9]

Cathode O2 + H2O + 2e → HO2- + OH- (3)

As expected form the electron number, the energy consumption could be reduced to 2.7 kWh kg-1 O2 in a concentrated KOH solution. Peroxide decomposes chemically to oxygen and peroxide on the surface of NiCo2O4 mesh, whereas oxygen is formed at the NiCo2O4 anode after

Anode 2 OH- → ½ O2 + H2O + 2e (4)

Brillas et al. [10] developed the above technique by using carbon-polytetrafluorethylene air-fed cathode. Peroxide anions HO2

- are formed at the carbon-based cathode, whereas oxygen evolution occurs at the anode by oxidation of both HO2

- and OH- anions. Pure oxygen is produced with a voltage lower than in conventional water electrolyzers, because of the lower standard potential of OH-/ H2O2, which further reduces the energy consumption. However, the current density is limited below 0.2 A cm-2 by the finite concentration of peroxide in the solution.

Recently, we demonstrated the use of a PEMFC for oxygen extraction, the anode reaction being hydrogen oxidation [11]. The technique was shown to be of relatively high efficiency, with current density up to 0.6 A cm-2, i.e. 0.015 mol O2 s

-1 per m2 membrane. Long-term tests confirmed the validity of the method, which however requires the presence of hydrogen to feed the fuel cell. To avoid this drawback, it was preferred to replace hydrogen oxidation at the anode by oxygen evolution. The present investigation deals with the use of an electrolysis cell of PEM technology involving reactions (1) and (2) at the electrodes. A dedicated 25 cm2 cell provided with a commercial water electrolysis MEA has been designed for operation at pressures close to the ambient level. Oxygen partial consumption in long term tests was evaluated by analysis of the outlet air, depending on the cell voltage - or the current density - and the excess in air oxygen fed to the cathode. Interpretation of the data has been conducted by calculating the cathode current efficiency.

2. Experimental section

2.1. Electrochemical cell and measurement devices

The 25 cm2 cell has been designed and built up from regular PEM technology. The cell consisted of two Ti bipolar plates being 10 mm thick: the flow pattern of serpentine type with five 1×1 mm2

parallel channels with 1mm broad edge was designed and machined by the mechanical workshop of the lab. The grooved part of the plate was covered by a 0.5 µm gold layer by ion sputtering. Gas diffusion layer at the cathode was of carbon material (SGL 30 BC, UBZM, Germany) including a macroporous carbon fiber paper and a carbon black microporous layer (MPL), whereas a one mm thick high porosity Ti fleece was used at the anode. The MEA (Fumatech) designed for PEM water electrolysis, consisted in a 117 Nafion membrane, a Pt/C (0.4 mg cm-2) cathode and a non-supported Pt/Ir (0.45 mg cm-2) anode (Figure 1).

All experiments were conducted at temperature levels ranging from 50 to 80°C. Liquid water was driven to the anode compartment by a peristaltic pump at 5 cm3 min-1 and preheated before entering the cell. Purified compressed dry air at flow rate ranging from 20 to 200 STP cm3 min-1 was humidified in lab-made trickled bed at RH=62 %. The cell was operated either at fixed voltage or at controlled current density using a PGSTAT 30 Autolab potensiostat connected to a 20 Amp. Autolab booster.


214

Figure 1. Schematic view of the 25 cm2 cell for oxygen extraction

Prior to its use for oxygen extraction, the MEA was run up by carrying out water electrolysis at 80°C: the anode was fed with hot water and nitrogen at 40 STP cm3 min-1 was fed to the cathode. As recommended by the supplier the MEA was conditioned by 24 hour long operation at 2.0 V for maturation of the anode Ir/Pt catalyst.

2.2. Current versus voltage measurements

Current density vs. voltage curve was established either in potensiostatic or in galvanostatic modes, so that the cell voltage was below 1.4 V for long term runs. Because the reactions of interest are oxygen reduction and evolution, the two electrode potentials are distributed around the equilibrium potential of H2O/O2 couple. Moreover the two reactions are relatively slow processes and exhibit absolute overpotentials in the same order of magnitude, it can therefore be estimated that the maximal cell voltage corresponds to anode potentials in the order of 1.8 V vs. NHE i.e. in a domain where corrosion phenomena at the anode can become significant and affect the stability of the anode catalyst. Most measurements were carried out with air but oxygen was also used for comparison.

2.3. Oxygen extraction runs and gas analysis

Oxygen extraction was achieved through numerous runs conducted at fixed current density for which the cell voltage was below 1.4 V as justified above. Taking into account Faraday’s law and the oxygen content in air (

2

inOy =0.21), the inlet oxygen flow rate was large enough for all runs. The

cell voltage at fixed current and air flow rate was monitored for lapses of time ranging from 4 to 24 hours: in most cases, the voltage attained a steady value within ten to thirty minutes.

The gas leaving the cathode was stored in a 400 cm2 polymeric sampling bag. This bag has been thoroughly emptied first, the outlet was then sampled for a couple of minutes and the bag was purged again. The second, longer sampling procedure for analysis was then achieved. After sufficient filling, the sampling gas was shut. The decrease in temperature to the ambient level was


doi: 10.5599/jese.2012.0016 215

the cause of water condensation and the bag slightly shrunk. The ambient temperature, Tamb, recorded for all runs ranged from 19 to 27°C. Analysis of the gas containing in the bag, i.e. nitrogen, oxygen and slight amount of water vapor was carried out off line by gas chromatography using a molecular sieve microcolumn with a TCD detector in the µGC equipment (Varian 490 GC). Three replicate analyses were conducted for each sample. The TCD detector was calibrated by analysis of six synthetic air-nitrogen mixtures of oxygen concentration perfectly known and ranging from 5 to 21 vol. %. The vapor amount in the gas mixture at Tamb was estimated using Antoine’s law:

2H Oamb

1737.93log 5.20389

39.485P

T= −

− (5)

where the partial pressure of water is expressed in Atm and Tamb is in K. The composition of the dry outlet could then be deduced from OHy

2.

3. Experimental results and interpretation

3.1. Electrochemical behavior of the cell

Steady values of the current or the voltage – depending on the electrical mode - were usually obtained within 15 minutes or so and for voltammetric measurements the voltage or current level was changed every 30-60 minutes. The MEA response could slightly depend on the long-term run carried out immediately before the measurement. Temperature was shown to exert a moderate effect (data not shown) and the results presented here were obtained at 60°C.

At a given voltage gas flow rate exerted also a moderate influence on the cell current at a given cell voltage, as exemplified by Figure 2. Over a given flow rate depending on the current density, the effect is however of reduced importance. The observed phenomenon probably expresses the partial control by oxygen diffusion from the air fed to the cathode.

Figure 2. Effect of the air flow rate on the current density recorded at fixed cell voltage.

As expected, current density is an increasing function of the cell voltage (Figure 3). Feeding the cell with pure oxygen largely increases the performance of the cell, with for instance a current density twofold larger with pure oxygen than upon air feed at 1V. In comparison to what occurs with air pure oxygen accelerates charge transfer rates in both oxygen reduction and evolution,


216

while suppressing the gas-side mass transfer occurring in the presence of air nitrogen. Current density up to 0.34 A cm-2 could be measured at 1.2 V with pure oxygen. The voltammetric data were compared to those obtained by Langer [2] or by Fujita [4] in Figure 3. In spite of the different i-V profiles reported, the available current densities are in the same order of magnitude over 1 V, also with better performance with pure oxygen. The differences between the three sources of data can be explained by the different features of the electrochemical cells: in addition to the likely different ohmic resistances, the electrodes differed a lot, in particular the anode, consisting of deposited Pt in the quoted papers and a Pt-Ir mixture in the present investigation.

Figure 3. Current density versus cell voltage at steady state. Present work with air at 200 STP cm3 min-1 and oxygen at 120 STP cm3 min-1; comparison with literature data.

Besides, the cell behaviour in long term runs was examined during oxygen extraction runs. As shown by Figure 4 for fixed values of the cell voltage, the current monitored remained at a stable level for periods up to 50 hours.

Figure 4. Current density in the oxygen extraction cell operated at fixed voltage.


doi: 10.5599/jese.2012.0016 217

3.2. Model of the cell oxygen extraction

A simple zero-D model has been written for the cathode compartment of the cell. The inlet variables are the molar flow rate of dry air in

airF and the cell current I, whereas outlet variables are the molar flow rate of oxygen out

O2F and the temperature of the sampling gas, Tamb: water was

assumed to be at vapor-liquid equilibrium at Tamb. Faraday’s law is written for oxygen consumption at the cathode according to reaction (1):

2 2

out inO O 4

IF F

Φ= −ℑ

(6)

where Φ is the current efficiency of oxygen reduction and ℑ is the Faraday’s constant. The inlet oxygen flow rate is related to the inlet air flow as follows:

2 2 2

in in in in inO air N air OF F F F y= − = (7)

where fraction inO2

y refers to the oxygen content in air. The oxygen molar fraction in the sampling bag is defined as:

2

2

outOout

O outt

Fy

F=

(8)

where subscript t refers to the total number of moles in the gas, covering nitrogen, oxygen and water vapor. Finally the total molar flow rate of gas is related to the flow of dry gas and the vapor molar fraction in the bag:

2 2

inout air

out airt

H O amb H O amb

41 1

IFF

Fy T y T

Φ−ℑ= =

− − (9)

For the sake of simplicity, the ambient temperature is no more mentioned in the following. The fraction of vapor in the sampled gas was calculated by Antoine’s law. Finally, the stoichiometric factor λ of fed air oxygen related to the flow of oxygen consumed by reaction (1) with a current efficiency at unity is defined by the relation:

2 2

in in inO O air 4

IF y F λ= =

ℑ (9)

From relations (6) to (9) the current efficiency of oxygen reduction Φ is expressed as follows:

2

2

2

2 2

outO

H O inO

outO H O

1

1

yy

y

y yλ

− −Φ =

− − (10)

For current efficiency equal to unity, the molar fraction of the oxygen in the sampling bag can be directly expressed as a function of operating parameter λ as:

( )2 2

2

outO H O

inO

11

1y y

y

λλ

−= −−

(11)


218

3.3. Experimental data and comparison with model predictions

The oxygen molar fractions determined by gas chromatography are compared to the values predicted by eq. (11) in Figure 5, considering the actual Tamb value. Good agreement between theory and practice is generally observed, even though the predicted line is slightly below the experimental data, indicating that the current efficiency could be somewhat lower than unity. For very large λ values, corresponding to large excess in oxygen, air is only slightly depleted in oxygen and the outlet gas fraction of oxygen is little different from the inlet value. Conversely, with low excess of oxygen, high abatement of oxygen from the fed air can be obtained, as expected.

Figure 5. Oxygen molar fraction at the outlet of the cell versus the stoichiometric

factor of air oxygen λ

Finally, because the outlet molar fractions of oxygen are in the same order of magnitude as the inlet fraction, in particular for large excess in fed air, the uncertainty in the determination of the current efficiency had to be estimated. The related calculations reported in the appendix show that high values of the current and moderate values for stoichiometric factor are favorable for higher accuracy in the determination of Φ: although the inlet flow rate of air could be moderate, the outlet molar fraction of oxygen is noticeably different from the inlet fraction, which prevails on the overall estimate for the uncertainty. On the contrary, for large λ values, the outlet molar fraction of oxygen is little below the inlet fraction and, from the expressions of terms A and B, the determination of the current efficiency is little accurate, with uncertainty exceeding 30 %.

The variations of the current efficiency with the stoichiometric factor of air oxygen and current density are shown in Figure 6 and 7 respectively, in the form of the maximum and the minimum estimates for each experiment. In spite of a significant dispersion, the data clearly show that the current efficiency is somewhat lower than unity.

The separate analysis of the effect of operating conditions on the current efficiency is rendered uneasy because of the large dispersion of the data in the representations (Φ vs. λ) or (Φ vs. i) in Figure 6 and 7. This apparent dispersion is mainly caused by the uncertainty in the determination of the current efficiency but also by the slight interaction between the two parameters. Nevertheless the stoichiometric factor of air oxygen exerts a negative influence on the current efficiency (Figure 6) whatever the current density applied. The positive effect of current density on Φ appears clearly with very little effect from λ (Figure 7): for current density over 100 mA cm-2, the current efficiency is near or larger than 0.9, which is promising for the investigated process.


doi: 10.5599/jese.2012.0016 219

Figure 6. Current efficiency for oxygen reduction in the cell versus stoichiometric factor λ: maximum and minimum estimates

Figure 7. Current efficiency for oxygen reduction in the cell versus the current density: maximum and minimum estimates

4. Conclusions and significance

The validity of the oxygen extraction using PEM electrolysis technology has been validated in long term runs for current density up to 200 mA cm-2, with a cell voltage near 1.3 V. Assuming that the reductive extraction is carried out with a current efficiency of 90 %, the corresponding energy consumption can be estimated to be near 4.4 kWh kg-1 O2. Although the state of health of the PEM


220

electrodes could not be evaluated by measurements of the electrode active surface [12], as done for PEMFC, no real decay in activity could be observed after more than 400 hour tests with the same MEA.

Considering the current progresses in water electrolysis technology, the current efficiency can be expected to attain 100 %, which would result in energy consumption near 4 kWh kg-1 O2. Besides, the current improvement in the electrodes of PEM electrolysers is to allow substantial reduction in cell voltage. However, to our point of view, the main advantage offered by PEM electrodes is that the cathode reaction involving air oxygen should be carried out with current density to 1 A cm-2, provided that the cell voltage does not exceed 1.4 V to limit degradation issues at an acceptable level: progress in electrolysis MEA has to be achieved for this purpose. In contrast, the reduction from a liquid phase as done through the peroxide route cannot be conducted for current densities over 100 or 200 mA cm-2. Although the PEM technique is not to allow lower energy consumption, higher production rates can be expected.

Long-term tests will be carried out at larger current densities, i.e. higher cell voltages, together with investigation of ageing phenomena, to evaluate the highest current density offered by the technique. Further work has to be done by improving developing electrode materials exhibiting faster kinetics for both oxygen reduction and evolution.

Acknowledgements: The authors are indebted from Region Lorraine and Institut Carnot ICEEL which funded the investigation cost and B. Eladeb’s PhD grant.

Appendix: Uncertainty in determination of the current efficiency

Relation (10) shows that efficiency Φ is a function of parameters or variables λ, outO2

y and OH2y .

The differential of the current efficiency was expressed as a function of partial derivatives with respect to the three above variables:

2 2

2 2

outO H Oout

O H O

d d d dy yy y

λλ

∂Φ ∂Φ ∂Φ Φ = + + ∂ ∂ ∂ (A1)

Derivation of the current efficiency with respect to λ, outO2

y and OH2y followed by algebraic

rearrangement led to:

2 2

2 2

2 2 2

outO H Oout

O H Oout inO O H O

d dd d 1 1 1 1y yy y

y B Ay y B Aλ

λ Φ = + − + − Φ

(A2)

with

2

2

2

outO

H O inO

1y

A yy

= − − (A3)

and

2 2

outO H O1B y y= − −

(A4) Factor λ is linked to the molar flow rate of fed oxygen, which is proportional to the volume flow

rate: the relative uncertainty of l can therefore be estimated by that for the inlet air flow rate. The flow meters employed for the measurements had a 1 % relative uncertainty in a large range of flow rates, i.e. for flow rates larger than 2 % of the full scale flow rate max

airQ . Below this threshold value, larger uncertainty can be expected. The following expressions were then considered:


doi: 10.5599/jese.2012.0016 221

in maxair air

in inair air

0.021%

Q QQ Q

λλ

ΔΔ = = for in maxair air0.02Q Q< (A5)

inair

inair

1%Q

Qλ

λΔΔ = = for in max

air air0.02Q Q>

In practice, because of the flow rate scale considered, the relative uncertainty varied from 3 to 10 %.

Moreover, in spite of the replicate analysis in the gas chromatograph and the thorough calibration procedure, it can be estimated that the relative error in the analytical determination is near 1 %:

2

2

outO

outO

1%y

y

Δ= (A6)

Finally, the uncertainty in OH2y is due to uncertainty in the ambient temperature. Considering

that this temperature was estimated within 1 °C and using Antoine’s law (5), the relative uncertainty in the molar fraction of vapour could be estimated in the temperature range considered:

2

2

H O

H O

4.5%y

y

Δ= (A7)

Applications of relations (A2)-(A7) yielded estimates for the relative uncertainty on the current efficiency, in the range 2-30 % for most cases.

References

[1] J. Winnick, “Electrochemical separation of Gases”, in “Advances in Electrochemical Science and Engineering”, edited by H. Gerischer and C.W. Tobias, Vol. 1, VCH, New-York (1990), 210-248.

[2] S.H. Langer and R.G. Haldemann, J. Phys. Chem. 68 (1964) 962-963. [3] R.A. Wynveen and K.M. Montgomery, J. Electrochem. Soc. 114 (1967) 589-592. [4] Y. Fujita, H. Nakamura, T. Muto, J. Appl. Electrochem. 16 (1986) 935-940. [5] J.W. Harrison, ASME 75-ENAs-51 (1975). [6] W.N. Lawless, US Patent N° 4,462,891 (July 31, 1984); 4,547,277 (Oct. 15, 1985). [7] L.G. Marianowski and R.J. Remick, US Patent N° 4,589,296 (1989). [8] M. Riecke, US Patent N° 6,632,400 B2 (2003). [9] A.C. Tseung, S.M. Jasem, J. Appl. Electrochem. 11 (1981) 209-215.

[10] E. Brillas, A. Maestro, M. Moratalla, J. Appl. Electrochem. 27 (1997) 83-92. [11] S. Altmeyer, E. Favre, C. Bonnet, P. Carré, F. Lapicque « Etude de procédés pour application

au flux gazeux d’un photobioréacteur de culture d’algues », Report 08 GPS 155, Nancy (Jan. 2009) [in French].

[12] A. Pozio, M. De Franceco, A. Cemmi,F. Cardellini and L. Giorgi, J. Power Sources 105 (2002) 13-19.



doi: 10.5599/jese.2012.0019 223




Wastewater treatment by multi-stage batch adsorption and electrochemical regeneration FADHIL M. MOHAMMED‡, EDWARD P. L. ROBERTS, ANDREW K. CAMPEN* and NIGEL W. BROWN*

School of Chemical Engineering and Analytical Science, University of Manchester. The Mill, Oxford Road, Manchester M13 9PL, UK; ‡Current address: Ministry of Science and Technology, Baghdad, Iraq

*Arvia Technology Ltd, Daresbury Innovation Centre, Keckwick Lane, Daresbury, Cheshire WA4 4FS, UK. Corresponding Author: E-mail: [email protected]; Tel.: +44-161-306-8849; Fax: +44-161-306-9321

Received: May 23, 2012; Revised: August 18, 2012; Published: November 10, 2012

Abstract The removal and destruction of a tri-phenyl methane dye, Acid Violet 17 (AV17), from aqueous solution by adsorption and electrochemical regeneration was studied using a graphite intercalation compound (GIC) adsorbent. It was demonstrated that the adsor-bent could be regenerated by anodic oxidation of the adsorbed dye in a simple electro-chemical cell. The GIC adsorbent recovered its initial adsorption capacity after 40 to 60 min of treatment at a current density of 10 mA cm−2, corresponding to a charge of 12 to 18 C g−1 of adsorbent. The charge passed is consistent with that expected for minerali-sation of the dye, suggesting that the dye was removed and destroyed with high charge efficiency. The energy cost of the regeneration was found to be around 120 J per g of adsorbent regenerated or 115 J per mg of the AV17 dye removed and destroyed. A model describing the process of wastewater treatment by multiple cycles of adsorption and electrochemical regeneration, based on adsorption isotherm data, has been deve-loped and validated. It was found that relatively modest improvements in the adsorption capacity of the adsorbent material could significantly improve the process performance.

Keywords Adsorption; electrochemical regeneration; graphite intercalation compound; tri-phenyl methane dye; acid violet.

Introduction

Dyestuff removal has been one of the most persistent problems in wastewater treatment in the last thirty years. Recent methods which have been investigated for dye removal include adsorp-

J. Electrochem. Sci. Eng. 2(4) (2012) 223-236 WASTEWATER TREATMENT ADSORPTION AND REGENERATION

224

tion, ion exchange, chemical oxidation, precipitation and biological treatment [1,2]. There are ever more dyes available commercially, and due to their complex structure and synthetic origin, most are difficult to decolorize [3]. Therefore, it is necessary to remove them from liquid wastes at least to a limit accepted by national and international regulatory agencies before discharge. Adsorption is a widely used technique for the removal of dyes from wastewater and can be effective for overall treatment, particularly if the sorbent is cheap, does not require a pre-treatment step before application and is easy to regenerate [4]. Adsorption on activated carbon is a technology which has been widely studied to remove dyes from wastewater but the high capital and regene-ration cost of activated carbon has led to the search for low cost materials. In the last twenty years, many investigators have studied the feasibility of inexpensive, commercially available mate-rials, that are easy to regenerate and re-utilized as many times as possible [5]. In recent years many inexpensive, widely available materials have been investigated as adsorbents to remove dyes from contaminated water. Wood, fly ash, coal, zeolite, silica, and agricultural wastes have all been tried with varying degrees of success [6-14].

The adsorption of dyes from aqueous solutions onto these inexpensive adsorbent materials focused on the determination of the capacity, kinetics, equilibrium isotherms and the effect of different parameters such as thermodynamics, pH, bonding mechanisms, and desorption [14-16]. However, only limited application of such data has been directed to the modelling of the batch adsorption / regeneration process for wastewater treatment.

Electrochemical regeneration of activated carbon adsorbents was demonstrated by Narbaitz and Cen in 1994 [17] and there have been several recent studies of this process [18-20]. However, the process is not economic as regeneration is slow and the energy cost is high [21], due to the poor conductivity of the activated carbon bed. Recent work has shown the potential of graphite intercalation compound (GIC) adsorbent materials that can be electrochemically regenerated very rapidly and cheaply [21-24].

Conventional adsorption processes are designed based on the single use of a batch of adsorbent. The rapid electrochemical regeneration process discussed above opens up the possibi-lity of using a batch of adsorbent for multiple cycles of adsorption with regeneration carried out on-site. In this way the adsorbent can be reused many times. In addition, a treatment process can be envisaged where the water is treated by the same batch of adsorbent through several cycles of adsorption and regeneration. This approach has been used to treat radioactive organic waste in the nuclear industry [25]. In this study we aim to carry out the first detailed study of this multi-stage adsorption and electrochemical regeneration process.

Experimental

Materials

The adsorbate used in this study was Acid Violet 17, a powdered, anionic tri-phenyl methane dye, which has three substituent groups and is a mono sodium salt. It was supplied by KEMTEX Educatio-nal Supplies Ltd under the trade name Kenanthrol Violet 2B at a dye content of about 22 %. Analysis is by UV/Vis spectroscopy at a λmax of 542 nm, as described in our earlier paper [26]. The adsorbent used was a graphite intercalated compound (GIC), NyexTM 1000, supplied by Arvia Technology Ltd., which contains around 94 wt. % carbon and has a particle size range of 100 – 700 µm [26].

F. M. Mohammed et al. J. Electrochem. Sci. Eng. 2(4) (2012) 223-236

doi: 10.5599/jese.2012.0019 225

Adsorption/electrochemical regeneration methodology

A laboratory scale sequential batch rig (see Fig. 1) was used for both adsorption and electrochemical regeneration, which were carried out at the ambient laboratory temperature of 20°C. In each experiment 100 g of GIC adsorbent was mixed with 1 L of water in the adsorption zone by sparging air into the bottom of the rig.

After adsorption, the adsorbent was allowed to settle into the anodic compartment of the electrochemical regeneration zone. The adsorbent bed was in contact with a graphite anode and was separated from the perforated stainless steel cathode (316L with open area 33%, 3 mm holes) by a microporous polyethylene separator (Daramic 350, Grace GmbH). The separator acts as a barrier to minimize the transport of electrolyte through the membrane and to ensure that the GIC bed does not contact the cathode, as this would short circuit the electrochemical cell. The cathode compartment was filled with 0.3 wt% NaCl solution to provide good conductivity, and the pH of this solution was adjusted to below 2 using hydrochloric acid to ensure that the membrane was stable. The anode, cathode and membrane of the electrochemical cell had dimensions of 10 cm by 7 cm, and the gap between the anode current feeder and the membrane was 2.2 cm. The 100 g of adsorbent used formed a bed of depth 5 cm in the anode compartment. A DC power supply was used to apply a current of 0.5 A to the cell, corresponding to a current density (from studies by Brown, 2005) of 10 mA cm−2 (based on the membrane area).

Ideally electrochemical regeneration should occur throughout the bed depth in the anode compartment. Although it is difficult to measure the current distribution in the packed bed electrode, previous studies with GIC adsorbents have examined the effect of bed thickness [22] or separated and tested layers of adsorbent in the regeneration bed [24]. These studies have indicated that complete regeneration can be achieved with bed depths of up to around 20 mm.

Prior to each experiment 100 g of fresh adsorbent was mixed with 1 litre of clean water for 30 min before being allowed to settle into the electrochemical regeneration cell. The water was drained off and a current of 0.5 A was applied for 30 min in order to oxidise any organic impurities present on the surface of the adsorbent. A volume of 1 L of solution containing 120 mg L−1 of AV17 was then added to the cell. This concentration was selected in order to saturate the adsorbent and to give a significant equilibrium concentration. This was necessary to ensure that the equilibrium concentration achieved with the regenerated adsorbent can be accurately measured.

The adsorbent and AV17 solution were mixed by sparging air into the cell for 120 min. This adsorption time was found to be sufficient to ensure that equilibrium was reached at these conditions. Once the adsorption stage was complete, the air was switched off and the adsorbent particles settled into the anodic compartment of the electrochemical cell. The treated liquid was drained off and a sample was taken and analysed by UV/visible spectroscopy in order to determine the loading of AV17 dye on the adsorbent.

The bed was regenerated for a period of between 10 and 120 min. After regeneration the supernatant was drained from above the bed of GIC adsorbent and the adsorption stage was repeated using a fresh AV17 solution (120 mg L−1) which was added to the cell. The adsorption stage was repeated by sparging with air for a period of 120 min. A sample of the solution obtained after adsorption was analysed by UV/vis spectroscopy, as described above. The performance of the regeneration was characterised using the ‘regeneration efficiency’, obtained by comparing the equilibrium adsorbent loading achieved before and after regeneration: / % 100 (1)


226

where qi and qr are the equilibrium loading (qe) of AV17 on the adsorbent (mg g−1) obtained before and after regeneration, respectively, calculated using Equation (2).

(2)

where C0 and Ce are the initial and equilibrium concentration of AV17 in solution respectively (mg L−1), V is the volume of solution (L), and W is the mass of adsorbent used (g).

Figure 1. Laboratory scale sequential batch rig for electrochemical regeneration of the GIC

adsorbent: schematic diagram of the side (a) and front (b) elevation of the rig, and (c) schematic diagram showing a cross section of the electrochemical regeneration zone.

(c) Electrochemical regeneration zone

Catholyte

Adsorbent bed

Anode

Membrane

Cathode

Hydrogen

Air in

Regeneration Zone

Adsorption Zone

(b) Front View

10 cm

37 cm

30 cm

12.3 cm

2.2 cm

11.5 cm

Air

in

(a) Side View


doi: 10.5599/jese.2012.0019 227

A similar technique was used for multi-stage batch adsorption and electrochemical regeneration, but in this case the solution was not drained off after the adsorption cycle. In this case, the initial concentration of AV 17 was much higher, 668 mg L−1 as this solution was treated by a series of adsorption/regeneration cycles. For these experiments, the amount of GIC adsorbent was increased to 125 g, which formed a bed of 7 cm depth in the anode compartment. In addition, the adsorption time was reduced to 60 min and the regeneration time was 30 min.


Adsorption isotherm

The design of a batch adsorber system requires knowledge of the equilibrium isotherm to understand the adsorption process behaviour and provide fundamental physiochemical data for evaluating the maximum capacity of the adsorbent [27]. Batch sorption studies have been perfor-med previously to investigate the kinetic and equilibrium isotherm of the adsorption of the Acid violet 17 [26]. The experimental data were found to fit the Langmuir isotherm model, Equation (3).

L ee

e1bk C

qbC

=+

(3)

where kL and b are the Langmuir constants related to the capacity of the adsorbent (mg g−1) and the intensity of adsorption, (L mg−1), respectively.

There is little data available in the literature on the adsorption of AV17 by activated carbon, but there is some data on other dyes such as methylene blue [28]. The comparison of Langmuir isotherm constants for adsorption of organic dye calculated in this work with those determined by [28] for activated carbon are shown in Table 1. The low surface area of the adsorbent leads to a relatively low adsorption capacity, as indicated by the relatively low value of kL. However, if the adsorptive capacity (kL) is normalised with the specific surface area, it is found that the GIC adsorbent is able adsorb a higher mass of dye per unit area than the activated carbons studied (albeit for a different adsorbate). This suggests that either the GIC has more adsorption sites per unit area or (more likely) that some of the activated carbon surface area in the micropores and is not accessible to the dye molecules.

Table 1. Comparison of Langmuir constants and surface area for adsorption of methylene blue (MB), and AV17 onto activated carbons [28] and the GIC adsorbent NyexTM 1000 [29], respectively, at room

temperature (23°C) and normal pH.

Adsorbent Adsorbate Langmuir constant Surface area

(BET), m2 g−1 Adsorptive

capacity, mg m−2 kL / mg g−1 b / L mg−1

GIC (NyexTM 1000) AV17 0.987 0.31 1.0 0.987

Activated carbon (PAC1) MB 307 0.12 863.50 0.36

Activated carbon (PAC2) MB 345 0.15 857.14 0.4

Activated carbon (F400) MB 455 0.2 1216.4 0.37

The essential characteristic of the Langmuir isotherm shape on whether adsorption is

‘favourable’ or ‘unfavourable’ can be classified in terms of dimensionless separation factor or equilibrium parameter, RL , Equation (4) [30].


228

Lmx

1 1

RbC

=+

(4)

where Cmx is the maximum solution concentration studied. The separation factor RL indicates the isotherm shape according to Table 2.

Table 2. Effect of separation factor on isotherm shape [30].

Value of RL Type of isotherm

RL>1 Unfavourable

RL=1 Linear

0<RL<1 Favourable

RL=0 Irreversible

The separation factor for the GIC adsorbent (b = 0.31 L mg−1) has been calculated at range of

maximum concentrations of AV17 as shown in Figure 2. It can be observed from this figure that the value of RL lies between 0 and 1 at all initial dye concentrations, confirming the favorable uptake of the AV17 by the GIC adsorbent.

Figure 2. Separation factor for Acid Violet 17 onto the GIC adsorbent NyexTM 1000 at 20°C.

Using the Langmuir model to determine the solid phase loading qmx in equilibrium with Cmx:

L mxmx

mx1bk C

qbC

=+

(5)

Thus:

e mx

mx mx e

11

eq C bCq C bC

+=+

(6)

Defining a dimensionless solid phase concentration as q = (qe/qmx), and a dimensionless liquid phase concentration as C = (Ce/Cmx), we obtain:


doi: 10.5599/jese.2012.0019 229

( )L L

1C

qR R C

=+ −

(7)

Measured values of q for a range of values of C are plotted in Figure 3, where the maximum equilibrium concentration was Cmx = 46 mg L−1. The experimental data was fitted to Equation (7) by finding the value of RL that gave a minimum in the sum of the absolute error (SAE) between the (q, C) data and Equation (7), indicating a value of RL = 0.065. Similar results have been reported for separation factors for the adsorption of the dye AV17 onto activated carbon prepared from sunflower seed hull [16] and orange peel [14].

Figure 3. Isotherm shape for adsorption of AV17 onto the GIC adsorbent NyexTM 1000 at room

temperature (23°C) and the curve obtained using Equation (7) with RL = 0.065.

Adsorption and electrochemical regeneration

The liquid phase concentration of AV17 dye after adsorption was found to be in the range 13.2 to 15.4 mg L−1, corresponding to a loading of AV17 on the GIC adsorbent of 1.07 to 1.05 mg g−1. This was slightly higher than the value of kL (0.987 mg g−1) from the Langmuir adsorption isotherm [29], possibly due to natural variability of the adsorbent, attrition or surface modification of the adsorbent during the washing procedure.

It was found that the full adsorption capacity could be recovered after 40 min of electrochemical regeneration, corresponding to a charge passed of 12 C g−1 (Figure 4). Further increases in the regeneration time led to an increase in the adsorption capacity to a value around 10 % higher than the original capacity. There are a number of possible reasons for this increase in adsorption capacity after regeneration. Previous studies [24] have suggested that the electrochemical regeneration can lead to a slight roughening of the adsorbent surface, and thus an increase in the specific surface area available for adsorption. Another possibility is the presence of oxidising species formed in the adsorbent bed during regeneration which may react with the organic dye leading to an apparent increase in adsorption capacity.

For regeneration times less than 40 min, the full capacity of the adsorbent was not recovered. This suggests that some AV17 or its breakdown products remained adsorbed on the surface of the GIC adsorbent. When the full adsorption capacity is recovered, the AV17 dye is either fully minera-


230

lised, or breakdown products are formed which do not adsorb on the adsorbent and are thus released into the water. Chemical oxygen demand (COD) analysis of treated water (unpublished data from Arvia Technology Ltd) have not found organic breakdown products in solution, suggesting that breakdown products remain adsorbed until full mineralisation is achieved.

It is possible to estimate the charge required to fully mineralise the dye; however this will depend on the products formed. The maximum charge required can be estimated by assuming complete anodic oxidation of the AV17 to carbon dioxide, sulphate and nitrate:

C41H44N3NaO6S2 + 93H2O → 41CO2 + 230H+ + Na

+ + 2SO4

2− + 3NO3

− + 224e− (8)

Alternatively a lower estimate of the charge required for mineralisation can be obtained by assuming that the products were carbon monoxide, sulphide and nitrogen:

C41H44N3NaO6S2 + 35H2O → 41CO + 114H+ + Na

+ + 2S

2− + 1.5N2 + 111e− (9)

The theoretical charge, Q, for mineralisation of the AV17 dye can be estimated from:

e

w

q nFQ

M= (10)

where n is the number of electrons required per molecule of dye oxidised, F is Faraday’s constant (96487 C mol−1), and Mw is the molecular weight of the dye (761.9 g mol−1). The value of qe was 1.05 mg g−1, and using n = 111 and 224 [based on Equations (4) and (5) respectively] the theoretical charge required for mineralisation of the AV17 would be between 15 and 30 C g−1. Comparison of these values with the data in Fig. 4 suggests that the charge efficiency for the anodic oxidation is high, but further work is needed to determine whether any breakdown products are present in the treated water.

Figure 4. Regeneration efficiency as a function of charge passed during electrochemical

regeneration of the GIC adsorbent NyexTM 1000 loaded with 1.06 mg g−1 of the organic dye AV17, using a current density of 10 mA cm−2 and a bed depth 2.2 cm.


doi: 10.5599/jese.2012.0019 231

The cell voltage was not continuously monitored, but was observed to be relatively stable at around 6 V throughout regeneration. The cell voltage is dominated by ohmic losses in the membrane and in the water in the anode compartment. Slight variations in the cell voltage were observed, presumably associated with changes in the conductivity of the water in the anode compartment (for example due the decrease in pH caused by the electrochemical processes) or the formation of gas bubbles leading to a decrease in effective conductivity. It was thus not possible to use the cell voltage to determine the completion of the regeneration. Based on a cell voltage of 6 V, the energy cost of the regeneration was 120 J per g of adsorbent regenerated or 115 J per mg of AV17 removed and destroyed.

Multi-stage batch design model

Batch adsorption and regeneration process are usually carried on small volumes of wastewater. However, the efficiency of dye removal can be improved by carrying out the treatment using multi-stage adsorption/regeneration system. An adsorption isotherm can be used to predict the design of a single stage batch adsorption system [31-36]. A wastewater treatment process consisting of a series of adsorption and regeneration stages can be considered to be a multi-stage equilibrium operation. The adsorption isotherm can thus be used to design a multi-stage adsorption / regeneration process as shown in Figure 5.

Figure 5. Schematic diagram of two stages in a multi-stage batch adsorption and regeneration system, considered as a multi-stage equilibrium process.

The design objective is to reduce the dye concentration in the feed (of volume V) from C0 to Cn (mg L−1), reusing the adsorbent (of mass W) in each adsorption stage after regeneration. The dye mass balance for batch adsorber system (n) in Figure 5 can be written as:

V Ce,n−1 + W qr = V Ce,n + W qe,n (11)

For the electrochemical regeneration process used in this study, we can assume that the regeneration efficiency was 100% (based on the data shown in Figure 4), so qr = q0 = 0 , and Equation (11) thus becomes:

V (Ce,n−1 − Ce,n) = W qe,n (12)

The experimental adsorption data were found to fit the Langmuir isotherm model (Equation 3) for adsorption of AV17 on the GIC adsorbent. Thus the Langmuir equation can be used to substitute for qe,n, giving:

qe,n W

Ce,n-1

V

qe,n-1

W

Batch adsorption

n-1

Ce,n

V

Water

Ce,n-2 V

qr

W

ElectrochemicalRegeneration

Sorbent qr NyexTM1000 W

Batchadsorption

n


232

, 1 ,

,

,

( )

1

e n e n

L e n

e n

C CWV bk C

bC

− −= +

(13)

Equation (13) can be rearranged to a quadratic form and solved to give:

2

L , 1 L e,n 1 e,n 1

,

1 1 4

2

e n

e n

W Wbk bC bk bC bC

V VC

b

− − − − − + − + + =

(14)

For the first adsorption cycle (n = 1), Ce,n−1 in Equation (14) is replaced by the initial concentration C0.

Equation (14) was used to determine the concentration after each stage of treatment for a given feed concentration C0 and mass of adsorbent (W) using the Langmuir isotherm constants for the GIC / AV17 system: b = 0.31 L mg−1 and kL = 0.987 mg g−1. For a given feed concentration, number of adsorption stages and required percentage dye removal, the solution was iterated numerically using a forward derivative Newton’s method with a convergence limit of 10−4 in order to calculate the amount of adsorbent, W (g), required.

The amount of adsorbent W was estimated for a five stage adsorption regeneration system to achieve AV17 dye removals ranging from 10 to 99 %, for a range of initial concentrations (55 - 680 mg L−1), and the results obtained are plotted in Figure 6.

The results indicate that the amount of adsorbent required increases linearly with the percentage dye removal, up to removals of around 90 to 95 %. This can be explained by observation that the adsorbent approaches its maximum loading in each of the adsorption cycles, so the amount of dye removed per gram of adsorbent is approximately constant. However, to achieve very high removals the dye concentration in the final adsorption stage will be low, so that the adsorbent used in this stage will not be fully saturated. The amount of adsorbent required thus begins to increase more rapidly as the final concentration decreases towards zero.

Figure 6. The adsorbent dose required as a function of the dye removal for a range of initial dye

concentrations using a five stage batch adsorption and regeneration system.


doi: 10.5599/jese.2012.0019 233

For validation of the model, an experimental study of a five stage adsorption – regeneration process was carried out for the adsorption of AV17 (with an initial concentration of C = 668 mg L−1) onto the GIC adsorbent (with an adsorbent dose of 125 g L−1). The values of the percentage removal achieved (Rn) after each stage obtained from the model and the experiment (Equations 14 and 15) were compared, as shown in Figure 7. The percentage removal of AV17 was observed to increase from 19 to 93 % in stages 1 to 5, and there was good agreement between the experimental data and the model. From the experimental data, the adsorbent loading after each adsorption cycle was found to be approximately 1±0.05 mg g−1, consistent with the capacity of adsorbent obtained from the isotherm study [24]. This confirms that for the conditions used in this experiment the adsorbent was approaching saturation with AV17 after each adsorption cycle.

0 nn

0

0% 10/C C

RC−= (15)

The adsorbent material investigated in this study has a very low adsorbent capacity, and there is significant potential to develop new adsorbent materials suitable for electrochemical regene-ration but with increased adsorptive capacity compared to the GIC adsorbent NyexTM 1000. Recent studies have shown that higher specific surface area adsorbent materials similar to the GIC used in this study can offer improved adsorption capacity [37,38]. In order to evaluate the benefits of in-creased adsorptive capacity, the effect of the value of the Langmuir kL on the multi-stage adsorp-tion-regeneration process was investigated. In this case, the number of adsorption/regeneration stages required to achieve 99.9 % AV17 removal (with an initial concentration of 1 g L−1) was calcu-lated for an adsorbent dose of 125 g L−1 and for a range of values of kL (see Figure 8). As kL was in-creased, it was found that stage number of stages of adsorption-regeneration required decreased such that a doubling of the adsorbent capacity leads to a reduction in the number of stages required from 10 to 5. A doubling of the adsorbent capacity has been demonstrated [37,38] and an order of magnitude increase in kL can readily be envisaged, which would reduce the number of stages required to only 2. This would have a significant impact on the economics of the process.

Figure 7. Performance of batch adsorption and electrochemical regeneration for removal of

AV17 using the GIC adsorbent NyexTM 1000 with five stages adsorption and regeneration. The initial AV17 concentration was 680 mg L−1 and the adsorbent dose was 125 g L−1.


234

Figure 8. Effect of adsorptive capacity (kL) on the number of stages required for 99.9% removal

of AV17 (with an initial concentration of 1 g L−1) using multi-stage adsorption-regeneration with an adsorbent dose of 125 g L−1. 125 g L−1.

Conclusions

It has recently been shown that low capacity unexpanded GIC adsorbents can be regenerated rapidly and cheaply, so that they can be regenerated and reused on-site [23,25,26]. This study describes the first detailed study of the adsorption/regeneration characteristics of a graphite intercalation compound. The laboratory scale batch regeneration tests demonstrated that the GIC adsorbent loaded with the organic dye Acid Violet 17 could be fully regenerated with a charge of around 15 C g−1. This charge is of the correct order of magnitude to suggest that the dye was mineralised during regeneration. The energy cost of the regeneration was found to be around 120 J per g of adsorbent regenerated or 115 J per mg of AV17 removed and destroyed.

The effective treatment of wastewater containing a dissolved organic contaminant by multi-stage batch adsorption with electrochemical regeneration has been demonstrated for the first time. Multistage adsorption / regeneration was shown to be effective for the removal of high concentrations of AV17 dye using the GIC adsorbent, exploiting its rapid electrochemical regeneration. Removal of 93 % of the dye from a solution containing 668 mg L−1 of AV17 was achieved after five cycles of treatment with 125 g L−1 of the GIC adsorbent. A design model for the multi-stage adsorption and electrochemical regeneration process based on the Langmuir equilibrium isotherm has been developed and validated for removal of a dye. This model was used to predict the dose of adsorbent required to achieve a range of dye removals for a given number of adsorption / regeneration cycles. The model was found to be in good agreement with the experimental results obtained for five stages of adsorption / regeneration. It was found that the predicted number of stages of batch adsorption / regeneration required to achieve 99.9 % AV17 removal was halved when the adsorptive capacity of the adsorbent was doubled. This finding confirms that development of adsorbents which can be regenerated electrochemically with increased capacity compared to the GIC used in this study could significantly improve the economics of the process.


doi: 10.5599/jese.2012.0019 235

Acknowledgements: The authors would like to thank the UK Engineering and Physical Sciences Research Council, ARVIA Technology Ltd and the Iraqi Ministry of Higher Education and Scientific Research for their financial support for this research.

References

[1] P.A. Pekakis, N.P. Xekoukoulotakis, D. Mantzavinos, Water Res. 40 (2006) 1276-1286. [2] J. Mo, J.-E Hwang, J. Jegal, J. Kim, Dyes Pigments 72 (2007) 240-245. [3] P. Nigam, G. Armour, I.M. Banat, D. Singh, R. Marchant, Bioresource Technol. 72 (2000)

219-226. [4] S.B. Wang, Y. Boyjoo, A. Choueib, Z.H. Zhu, Water Res. 39 (2005) 129-138. [5] G. San Miguel, S.D. Lambert, N.J.D. Graham, Water Res. 35 (2001) 2740-2748. [6] N. Kannan, M.M. Sundaram, Dyes Pigments 51 (2001) 25-40. [7] Y. Guo, S. Yang, W. Fu, J. Qi, R. Li, Z. Wang, H. Xu, Dyes Pigments 56 (2003) 219-229. [8] C. Namasivayam, M. Dinesh Kumar, K. Selvi, R. Ashruffunissa Begum, T. Vanathi, R.T.

Yamuna, Biomass Bioenerg. 21 (2001) 477-483. [9] C. Namasivayam, D. Kavitha, Dyes Pigments 54 (2002) 47-58.

[10] G. McKay, M. El Geundi, M.M. Nassar, Water Res. 21 (1987) 1513-1520. [11] M. Özacar, I.A. Sengil, J. Hazard. Mater. 98 (2003) 211-224. [12] G. Annadurai, R.-S. Juang, D.-J. Lee, J. Hazard. Mater. 92 (2002) 263-274. [13] G.S. Gupta, G. Prasad, V.N. Singh, Water Res. 24 (1990) 45-50. [14] R. Sivaraj, C. Namasivayam, K. Kadirvelu, Waste Manage. 21 (2001) 105-110. [15] R.A. Shawabkeh, M.F. Tutunji, Appl. Clay Sci. 24 (2003) 111-120. [16] N. Thinakaran, P. Baskaralingam, M. Pulikesi, P. Panneerselvam, S. Sivanesan, J. Hazard.

Mater. 151 (2008) 316-322. [17] R.M. Narbaitz, J.Q. Cen, Water Res. 28 (1994) 1771-1778. [18] P. Canizares, J. Lobato, J. Garcia-Gomez, M.A. Rodrigo, J. Appl. Electrochem. 34 (2004) 111-

117. [19] R.M. Narbaitz, A. Karimi-Jashni, Environ. Technol. 30 (2009) 27-36. [20] R. Berenguer, J.P. Marco-Lozar, C. Quijada, D. Cazorla-Amoros, E. Morallon, Carbon 48

(2010) 2734-2745. [21] N.W. Brown, E.P.L. Roberts, A. Chasiotis, T. Cherdron, N. Sanghrajka, Water Res. 38 (2004)

3067-3074. [22] N.W. Brown, E.P.L. Roberts, A.A. Garforth, R.A.W. Dryfe, Electrochim. Acta 49 (2004) 3269-

3281. [23] N.W. Brown, E.P.L. Roberts, A.A. Garforth, R.A.W. Dryfe, Water Sci. Technol. 49 (2004) 219-

225. [24] N.W. Brown, Adsorption and electrochemical regeneration for wastewater treatment:

development of a process, PhD Thesis, UMIST, Manchester, UK, 2005. [25] N.W. Brown, D.A. Wickenden, E.P.L. Roberts, Waste Management Symposium, Proceedings

of WM2012, Phoenix, Arizona, USA, 2012. [26] F.M. Mohammed, E.P.L. Roberts, A. Hill, A.K. Campen, N.W. Brown, Water Res. 45 (2011)

3065-3074. [27] S. Ismadji, S.K. Bhatia, Can. J. Chem. Eng. 78 (2000) 892-901. [28] E.N. El Qada, S.J. Allen, G.M. Walker, Chem. Eng. J. 135 (2008) 174-184. [29] F.M. Mohammed, Modelling and design of water treatment processes using adsorption

and electrochemical regeneration, PhD Thesis, University of Manchester, Manchester UK, 2011.

[30] M. Gordon, J. Chem. Technol. Biot. 32 (1982) 759-772. [31] G. McKay, M.S. Otterburn, J.A. Aga, Water Air Soil Poll. 24 (1985) 307-322.


236

[32] M. Dogan, M. Alkan, Y. Onganer, Water Air Soil Poll. 120 (2000) 229-248. [33] M. Özacar, I.A. Sengil, Environ. Geol. 45 (2004) 762-768. [34] M. Özacar, I.A. Sengil, Biochem. Eng. J. 21 (2004) 39-45. [35] V. Vadivelan, K.V. Kumar, K. V. J. Colloid Interf. Sci. 286 (2005) 90-100. [36] E.I. Unuabonah, K.O. Adebowale, A.E. Ofomaja, Water Air Soil Poll. 200 (2009) 133-145. [37] H.M.A. Asghar, Development of graphitic adsorbents for water treatment using adsorption

and electrochemical regeneration, PhD Thesis, University of Manchester, Manchester, UK, 2011.

[38] H.M.A. Asghar, E.P.L. Roberts, S.N. Hussain, A.K. Campen, N.W. Brown, Submitted to J. Appl. Electrochem. (2012).



journal of electrochemical science and engineering...issn: 1847-9286 open access journal journal of...

Documents