titanium dioxide composite adsorbent loaded with an organic dye

Journal Pre-proof Anodic electrochemical regeneration of a graphene/titanium dioxide composite adsorbent loaded with an organic dye Farbod Sharif, Edward P.L. Roberts PII:

S0045-6535(19)32259-3

DOI:

https://doi.org/10.1016/j.chemosphere.2019.125020

Reference:

CHEM 125020

To appear in:

ECSN

Received Date: 8 July 2019 Revised Date:

25 September 2019

Accepted Date: 29 September 2019

Please cite this article as: Sharif, F., Roberts, E.P.L., Anodic electrochemical regeneration of a graphene/titanium dioxide composite adsorbent loaded with an organic dye, Chemosphere (2019), doi: https://doi.org/10.1016/j.chemosphere.2019.125020. This is a PDF file of an article that has undergone enhancements after acceptance, such as the addition of a cover page and metadata, and formatting for readability, but it is not yet the definitive version of record. This version will undergo additional copyediting, typesetting and review before it is published in its final form, but we are providing this version to give early visibility of the article. Please note that, during the production process, errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain. © 2019 Published by Elsevier Ltd.

1 2

Anodic Electrochemical Regeneration of a Graphene / Titanium Dioxide Composite Adsorbent Loaded with an Organic Dye

3

Farbod Sharif1, Edward P. L. Roberts1*

4

5 6

1

7

AB T2N 1N4, Canada

8

*Corresponding author. Tel: (403) 220-4466. E-mail: [email protected] (Edward Roberts)

University of Calgary, Department of Chemical and Petroleum Engineering, 2500 University Drive NW, Calgary,

9

10

Abstract

11

A nanocomposite of graphene and titanium dioxide (G/TiO2) was prepared using the sol-gel

12

method for use in an electrochemical adsorption / regeneration process. The effect of annealing

13

temperature on electrochemical characteristics of the nanocomposites was investigated by cyclic

14

voltammetry and constant current electrochemical regeneration, using methylene blue (MB) as

15

the adsorbate. The G/TiO2 could be regenerated more rapidly and with less corrosion than the

16

bare graphene. The G/TiO2 annealed at 400 °C had a higher proportion of anatase phase TiO2

17

(ca. 7% rutile TiO2) compared to that annealed at 500 °C (ca. 40% rutile TiO2). Cyclic

18

voltammetry indicated that the G/TiO2 annealed at 400 °C had a higher activity for MB

19

oxidation than the nanocomposite annealed at 500 °C. Similarly, the regeneration of MB loaded

20

G/TiO2 annealed at 400 °C was much faster than for the nanocomposite annealed at 500 °C.

21

Complete regeneration of the G/TiO2 annealed at 400 °C was obtained after an electrochemical

22

charge of 21 C per mg of adsorbate. The G/TiO2 annealed at 400 °C was regenerated in half the

23

time required for the bare graphene. TEM studies showed that the bare graphene was rapidly

24

corroded, while corrosion was not observed for the G/TiO2 nanocomposites.

25

Keywords: graphene, titanium dioxide, nanocomposite, adsorbent, electrochemical regeneration.

26 27 28

1. Introduction

29

Graphene is a two dimensional, one atom thick arrayed carbon with an sp2 hybridized

30

carbon structure (Allen et al., 2010). Owing to their high surface area and good chemical stability

31

(Zhao et al., 2012), graphene-based materials may be considered as a good candidate for removal

32

of contaminants through adsorption in water treatment (Chandra and Kim, 2011; Gao et al.,

33

2012; Madadrang et al., 2012; Xu et al., 2012) . However, the application of graphene-based

34

materials to adsorption processes is limited by their high production and regeneration costs.

35

Several methods have been used to regenerate the loaded graphene-based materials, namely

36

thermal regeneration at high temperature (Yanyan et al., 2013), Fenton-like reactions (Qin et al.,

37

2014), solvent extraction (Zhang et al., 2014) and electrochemical regeneration (Flores et al.,

38

2018; Sharif et al., 2017; Sharif and Roberts, 2015). However, each of these methods has

39

shortcomings including high cost, incomplete regeneration, adsorbent losses, secondary

40

pollution, or long regeneration times. Among the aforementioned methods, electrochemical

41

regeneration showed that it can be a promising technique in which minimum adsorbent loss, in-

42

situ regeneration, and high regeneration efficiencies can be achieved (Dai et al., 2017; Hussain et

43

al., 2015; Narbaitz and Cen, 1994; Narbaitz and Karimi
Jashni, 2009), with no harmful reagents

44

required.

2

45

The main mechanism for anodic electrochemical regeneration is the oxidation of organic

46

pollutants on the surface of the absorbents (Brown et al., 2004a). Ideally, the adsorbate should be

47

mineralized completely during regeneration by electrochemical oxidation, converting them into

48

carbon dioxide, water and salts.

49

Anode materials used for electrochemical oxidation can be classified as active or inactive

50

(Comninellis, 1994). While active anodes tend to chemisorb hydroxyl radicals and have a low

51

overpotential for oxygen evolution, inactive anodes physisorb the hydroxyl radicals and have

52

higher oxygen evolution overpotential (Brillas and Martínez-Huitle, 2011; Gautam and

53

Chattopadhyaya, 2016). Depending upon the anode material and the mechanism of the oxidation,

54

the pollutants may not be oxidized completely, resulting in the production of undesirable

55

breakdown products. Active electrodes such as graphite, platinum, and ruthenium and iridium

56

oxides can only partially mineralize the organics. Inactive electrodes, however, such as tin oxide,

57

lead oxide and boron-doped diamond can completely mineralize organics to carbon dioxide and

58

water (Boye et al., 2002; Panizza and Cerisola, 2009, 2003).

59

In the electrochemical oxidation process, a part of the charge passed will usually be

60

consumed by side reactions, in particular, oxygen evolution, reducing the cell’s current

61

efficiency. It is, therefore, preferable to use an adsorbent material that can act as an inactive

62

anode, in order to decrease side reactions and increase the current efficiency (Yun-Hai et al.,

63

2012). This will result in reduced energy costs and reduced regeneration times, making

64

adsorption and electrochemical regeneration processes more economically feasible.

65

As graphite is categorized as an active anode (Rueffer et al., 2011), it needs to be

66

functionalized to produce more hydroxyl radicals. Among the inactive electrodes, boron-doped

67

diamond (BDD) has been shown to be very effective for hydroxyl radical production (Iniesta et 3

68

al., 2001; Marselli et al., 2003). This is mainly due to the large band gap of diamond (5.45 eV)

69

which make the BDD suitable to produce a large amount of hydroxyl radical (Beck et al., 2000;

70

Kraft, 2007). However, the preparation of BDD is not simple, and BDD electrodes are

71

consequently a relatively high-cost material (Xie et al., 2017). Ntsendwana et al. (2013) reported

72

the modification of exfoliated graphite (EG) using diamond. Their findings revealed that

73

exfoliated graphite/diamond composites have a higher removal and current efficiency and faster

74

degradation kinetics than exfoliated graphite electrodes for electrochemical oxidation of

75

trichloroethylene. The effect of adding diamond particles onto expanded graphite electrodes for

76

the electrochemical degradation of anthraquinonic dye was also investigated by Peleyeju et al. (

77

2016). Their findings confirm the results obtained by Ntsendwana, where the incorporation of

78

diamond increased the electro-catalytic activity and consequently the degradation rate of

79

anthraquinonic dye in comparison with bare EG. It has been proposed that the enhancement in

80

organic oxidation is mainly influenced by high production rates of hydroxyl radicals and electron

81

trapping on the anode surface. In several studies, TiO2 has been shown to act as an efficient

82

electrocatalyst owing to its large bandgap of 3.2 eV and increase the hydroxyl radical formation

83

for organic oxidation (Yang and Hoffmann, 2016). The versatility and effectiveness of TiO2 as a

84

catalyst for organic oxidation was demonstrated by Jasmann et al. (2016), who showed that

85

anatase TiO2 is catalytically active in the dark even in the absence of an applied potential. This is

86

probably due the generation of electron–hole pairs, eCB− and hVB+ in the TiO2, excited by the

87

applied potential. Kesselmanet et al. (1997) showed that polycrystalline TiO2 can generate

88

produce hydroxyl radicals at its surface where the valence band (hVB+) is very positive and

89

capable of producing hydroxyl radicals and the conduction band (eCB−) potential is highly

90

negative and capable of reducing oxygen to hydrogen peroxide.

4

91

An improved electrochemical oxidation of Rhodamine B using SnO2-Sb-doped TiO2-coated

92

granular activated carbon was reported by Li et al. (2016). In another study, in order to produce

93

more active sites and hydroxyl radicals, Guo et al. (2015) prepared a TiO2/nano-graphite

94

composite anode. The TiO2/nano-graphite composite showed that it can generate large amounts

95

of hydroxyl radical and degrade the organics efficiently. This result is consistent with the study

96

by Wang (2008) who observed that the addition of TiO2 particles to a packed bed

97

electrochemical reactor can enhance the production of hydroxyl radicals and, consequently, the

98

organic oxidation. It has also been reported that composite electrodes with TiO2 demonstrate a

99

high catalytic activity and anticorrosive behavior which can increase the durability of the

100

material (Evdokimov, 2002; Lin et al., 2013)

101

In our previous work, the adsorption and electrochemical oxidation of organics using

102

reduced graphene oxide / iron oxide adsorbents was studied (Sharif et al., 2017). The result

103

showed 100% regeneration efficiency and good adsorptive capacity compared to a flake graphite

104

intercalation compound (GIC) adsorbent. This graphene nanocomposite also demonstrated

105

significantly lower specific charge for regeneration than has previously been reported for

106

graphite adsorbents (Brown and Roberts, 2007) . However, the graphene nanocomposite

107

adsorbent was oxidized in the course of the regeneration process, similar to what was observed

108

for GIC adsorbents (Nkrumah-Amoako et al., 2014). Oxidation of large diameter (a few hundred

109

micrometers) GIC flakes led to an increase in specific surface area (presumably due to surface

110

roughening) along with changes in the surface functional groups. However, the graphene / iron

111

oxide nanocomposite was found to severely corrode during electrochemical regeneration, so that

112

it was not possible to use the adsorbent for more than a few cycles of adsorption and

113

electrochemical regeneration (Sharif et al., 2017) .

5

114

In the present study, graphene/TiO2 nanocomposites (G/TiO2) calcined at different

115

temperature were synthesized, characterized and used in an adsorption and electrochemical

116

regeneration process. Previous work on the electrochemical regeneration of adsorbents materials

117

focused on carbon derivatives, namely activated carbon (Berenguer et al., 2010; Wang and

118

Balasubramanian, 2009; Weng and Hsu, 2008) or graphite intercalation compound (Asghar et al.,

119

2014, 2013), which both are categorized as active anodes. The aim of this research was to offer

120

new adsorbents with high adsorptive capacity for removal of organics, combined with rapid

121

electrochemical regeneration and stability for many cycles in order to overcome the challenges

122

of conventional adsorbents. Methylene blue (MB) solution was used as a model contaminant in a

123

synthetic wastewater. Along with the adsorption characteristics of MB on these materials, the

124

voltammetric behavior and electrochemical regeneration performance were investigated,

125

including the regeneration time and specific charge required for complete regeneration. The

126

stability of the G/TiO2 nanocomposite was also evaluated over multiple cycles of adsorption and

127

regeneration.

128 129

2. Materials and Methods

130

2.1. Preparation of graphene / titanium dioxide nanocomposites

131

The sol-gel method was used to prepare the TiO2 sol (Falamaki and Veysizadeh, 2007). A

132

volume of 15 mL of tetraisopropylorthotitanate was dissolved in 200 mL of isopropanol. In order

133

to hydrolyze the mixture, 150 mL of deionized water was added to the solution under vigorous

134

stirring. The precipitate was washed using a centrifuge in order to eliminate the alcohol traces,

135

after which 330 mL of deionized water was added. A volume of 24 mL of 1 mol L−1 nitric acid

136

was then added as a peptizing agent and the resulting mixture was stirred for 72 h to form the sol. 6

137

A known volume of TiO2 sol with a concentration of 13 g L-1 was mixed with 0.4 g of

138

commercial few-layer graphene (XG Sciences grade M5). The few layer graphene particles have

139

a thickness of approximately 6 - 8 nm, a surface area ~62 m2 g-1, and an average particle diameter

140

of 5 µm). The mixture was stirred for 24 h after which the obtained nanocomposite was dried at

141

80 °C for 12 h, and calcined at either 400 °C or 500 °C for 2h in air. The transformation

142

temperature from anatase to rutile can fall in the range of 400–1200 °C depending on the

143

materials used, and processing methods (Hanaor and Sorrell, 2011). Based on the methods used

144

in this study anatase titanium dioxide is formed at 400 °C, and at 500 °C the transition from the

145

anatase to the rutile phase of titanium dioxide has begun. The nanocomposite adsorbents

146

annealed at 400 °C and 500 °C are referred to as G/TiO2 400 °C and G/ TiO2 500 °C

147

respectively.

148 149

2.2. Characterization

150

Transmission electron microscopy (TEM) and scanning electron microscopy (SEM) were

151

used to characterize the morphology and nanostructure of the adsorbent. The TEM used was a

152

Tecnai TF20 G2 FEG-TEM (FEI, Hillsboro, Oregon, USA) instrument operating at 200 kV

153

acceleration voltage with a standard single-tilt holder, while SEM was performed with a Zeiss

154

supra55 field-emission system. The particle size of the TiO2 sol was measured by laser

155

diffraction particle size analysis (Malvern Mastersizer 3000). Raman spectra were recorded on a

156

WITec alpha 300 R Confocal Raman Microscope (WITec GmbH, Germany) using a 532-nm

157

laser.

158

XPS spectra were recorder using Kratos Axis ULTRA photoelectron spectrometer (Kratos

159

analytical limited, UK) with an Al-Alpha excitation source (1436 eV). A very dilute solution of 7

160

the nanocomposite was drop-casted on the silicon wafer.. A thermogravimetric analyzer (TA

161

Instruments, Q500) was used to obtain TGA data under nitrogen atmosphere from room

162

temperature to 1073 K, with a rate of heating at 10 K min−1. UV-Vis spectroscopy (UV-2600,

163

Shimadzu) was used to measure the concentration of MB.

164

165

2.3. Adsorption and electrochemical regeneration

166

Kinetic adsorption experiments were conducted to determine the equilibrium time required

167

to reach equilibrium. Adsorption experiments were carried out using fresh adsorbents at room

168

temperature, with an aqueous solution of methylene blue (MB) as a model wastewater.

169

Typically, 0.1 g of nanocomposite was mixed with 150 mL of MB solutions with different

170

concentrations in 250 mL flasks. The flasks were shaken for 30 min at 200 rpm until equilibrium

171

was established. Subsequently, the adsorbent was separated using a filter paper and the sample

172

was analyzed via UV-Vis spectrophotometry.

173

The electrochemical regeneration of adsorbents was conducted in an electrochemical cell, as

174

described in our previous paper (Sharif et al., 2017). A mass of 0.1 g of adsorbent was mixed

175

with 0.15 mL of solution containing 25 ppm MB. Thus, under these conditions the loading of

176

MB on the adsorbent is close to the maximum loading for all three adsorbents. Briefly, the

177

loaded adsorbent was filtered using a HTTP polycarbonate filter with 0.45-micron pore size. The

178

filtered solids collected on the filter were saturated with a few drops of 2% w/w sodium chloride

179

electrolyte. The filter cake, with an area of 11 cm2 and filter paper were then pressed between a

180

graphite plate current feeder and a stainless steel cathode, with the adsorbent in contact with the

181

graphite plate. A Metrohm Autolab potentiostat operating in galvanostatic mode was used to

8

182

supply a constant current density of 10 mA cm˗2 to the current feeders for a regeneration time in

183

the range 1 to 20 min. In order to measure the regeneration efficiency (RE), regenerated

184

adsorbent was used for re- adsorption of MB under the same conditions as the initial adsorption.

185

The regeneration efficiency was calculated as follows (Brown et al., 2004b):

186

RE =

qr ×100 qi

(1)

187

where: qr is the loading of the regenerated adsorbent and qi is the initial adsorbate loading

188

measured under the same adsorption conditions.

189

190

2.4. Electrochemical properties

191

The electrochemical characteristics of the G/TiO2 nanocomposites were studied by cyclic

192

voltammetry using a conventional three electrode system with a 1 mol L-1 NaCl electrolyte, a

193

platinum wire counter electrode with a surface area of 4.7 cm2, silver/silver chloride reference

194

electrode and modified glassy carbon working electrode. Working electrodes were prepared by

195

drop casting 20 µL of a suspension of the nanocomposite onto the glassy carbon electrode

196

(diameter of 5 mm, 0.2 cm2). Suspensions were prepared by mixing the nanocomposite (before

197

or after MB adsorption) with a suspension Nafion to act as a binder (Nafion-to-adsorbent mass

198

ratio of ~0.1) (Zehtab Yazdi et al., 2016, Brown et al., 2004b). The cyclic voltammetry

199

experiments were performed using an Autolab PGSTAT (Metrohm, UK), with a potential range

200

from −0.6 and 0.4 V (versus Ag/AgCl) and a scan rate of 5 mV s-1.

201

9

202

3. Result and discussion

203

3.1. Characterization

204

The Raman spectra of the G/TiO2 nanocomposites is shown in Figure 1a. The main Raman

205

features of graphene based materials are the so-called D band (ca.1350 cm−1), G band (ca.1580

206

cm−1) and 2D band (ca. 2670 cm−1). The G band arises due to in-plane vibration of sp2 carbon

207

atoms, while the D band originates from disorder in the sp2-hybridized carbon systems. The

208

degree of disorder is a function of the D/G intensity ratio (Stankovich et al., 2007). Raman

209

spectra shows that the D/G intensity ratio remained approximately constant which implies that

210

the coating of the TiO2 and the heat treatment at 400 °C and 500 °C did not affect the structure of

211

the graphene.

212

The formation of the anatase phase of TiO2 in the G/TiO2 400 °C nanocomposite was confirmed

213

by observing the Eg mode (at 145 cm−1) (Ohsaka et al., 1978; Ren et al., 2013). However

214

increasing the calcination temperature in G/ TiO2 500 °C leads to increase the rutile phase which

215

is confirmed by observing the Bg (at 427 cm−1), A1g (at 620 cm−1) modes and the second-order

216

effect at 237 cm−1 (Swamy et al., 2006). The phases of TiO2 present in the nanocomposites was

217

also confirmed by powder x-ray diffraction (see Supporting Information Figure A.1). It was

218

found that the G/TiO2 400 °C nanocomposite contained 93 % anatase phase TiO2 and only 7% of

219

the less active rutile phase. When the nanocomposite was annealed at 500 °C the rutile content

220

increased to around 40%.

221

SEM images of the bare graphene and G/TiO2 nanocomposites (Supporting Information

222

Figure A.2) illustrate that bare graphene is composed of smooth, free-standing thin sheets, while

223

the SEM image of 6 wt% G/TiO2 nanocomposites revealed that the surface of graphene was

224

covered with TiO2 nanocrystals. The morphology of G/TiO2 nanocomposites can also be seen in 10

225

high resolution using TEM (Figure 1b). The morphology of the nanocomposites is consistent

226

with the bare graphene in the range of several micrometers. TEM image of the nanocomposites

227

confirm that the graphene acts as a substrate and support for the TiO2 nanomaterials. It can also

228

be seen that size of the of the TiO2 particles is around 5-10 nanometers which is in good

229

agreement with the DLS measurement of the particle size in the original sol, as shown in the

230

supporting information (Figure A.3).

231 232

The precise surface chemistry and the chemical status of elements present in the

233

synthesized G/ TiO2 nanocomposite were examined using XPS. Figure 2 depicts XPS survey and

234

high-resolution spectra of C, O and Ti. Figure 2a demonstrates that the TiO2/Nano-G composite

235

was constituted of C, O and Ti. Figure 2b illustrates the XPS spectrum of C1s, this peak was

236

deconvoluted into two peaks with Gaussian-Lorentzian rules with binding energies of 284.6 eV,

237

285.3 eV respectively. The peak at 284.6 eV can be attributed to sp2 carbon, and the 285.3 eV

238

peak corresponds to sp3 carbon (Wang et al., 2015; Yu and Chua, 2010). Figure 2c shows the

239

Ti2p spectra of the G /TiO2 nanocomposite. Ti2p3/2 can be detected at a binding energy of

240

459.6 eV and Ti2p1/2 can be observed at a binding energy 465.1 eV (Cheng et al., 2012; Ming et

241

al., 2012). A small shift of 0.5 eV was observed in for both the Ti2p3/2 and Ti2p1/2 peaks

242

compared to those for the bare TiO2 nanoparticles. This phenomenon can be attributed to the

243

doping the carbon in the TiO2 structure (Lei et al., 2015). The high-resolution XPS spectra of

244

O1s of the nanocomposite was verified and depicted in Figure 2d. Two peaks could be identified

245

at 530.8 eV and 533.7 eV, suggesting the formation of two types of O states, which was

246

recognized as Ti-O bond at 530.8 eV (Ramqvist et al., 1969) and Si-O bond coming from the

247

substrate , respectively. The XPS spectra confirm the formation of a graphene TiO2 composite.

11

248 249

3.2. Adsorption Isotherm

250

Figure 3a shows the adsorption isotherm of the bare graphene and the G/TiO2

251

nanocomposites. The isotherms were found to be a good fit to the Langmuir adsorption model.

252

The data were fitted to the Langmuir model using nonlinear regression and the results are

253

summarized in Table A.1. A small reduction in adsorptive capacity of the G/TiO2 400 °C

254

nanocomposite (6 wt% TiO2) was observed when compared to bare graphene, from around 25

255

mg g−1 to 21 mg g−1. The adsorptive capacity of the G/TiO2 500 °C nanocomposite was

256

significantly lower, at around half the capacity of the bare graphene. These results suggest that

257

the TiO2 particles have a lower adsorption capacity than the graphene. All three types of

258

adsorbent, the bare graphene, and the G/TiO2 nanocomposites, were able to remove organics to

259

low concentrations (below 1 mg L−1).

260

The lower adsorptive capacity of the G/TiO2 500 °C nanocomposites may be explained by

261

the thermal behavior of the graphene. The TGA results (Supporting Information Figure A.4)

262

show that the graphene starts losing mass at around 350-400 °C, probably due to the presence of

263

amorphous carbon. One possible explanation for the significant reduction in the adsorptive

264

capacity after annealing at 500 °C, is that a significant proportion of amorphous carbon may be

265

lost during annealing, leading to loss of adsorption sites.

266

A control experiment was conducted to verify this hypothesis. The bare graphene was

267

calcined at 400 °C and 500 °C (G400 °C and G500 °C respectively) and the adsorption was

268

conducted using 0.1 g of the adsorbent and 150 mL of 25 ppm MB solution. The results showed

269

a drop in adsorptive capacity of the bare graphene after annealing at 400 °C and 500 °C, from

270

around 23 mg g−1 to 21 mg g−1 and 15 mg g−1 respectively. In addition, MB adsorption 12

271

experiments were carried out with the as prepared TiO2 (in the absence of graphene). The

272

adsorption capacities for MB on the TiO2 annealed at both 400 °C and 500 °C were found to be

273

negligible compared to the bare graphene and the G/TiO2 nanocomposites. It can thus be

274

concluded that the reduction in the adsorption capacity of the nanocomposites when compared to

275

the bare graphene was most likely due to the loss of amorphous carbon during annealing,

276

reducing the number of adsorption sites.

277

278

3.3. Electrochemical Regeneration

279

The effect of the electrochemical treatment time on regeneration efficiency is presented in

280

Figure 3b for bare graphene, G/TiO2 400 °C and G/TiO2 500 °C nanocomposites. For all

281

adsorbents, the regeneration efficiency was found to increase to 100% regeneration or greater

282

with increasing the regeneration time. For the bare graphene, the regeneration efficiency

283

increased steadily with regeneration time up to around 12 minutes, and then more rapidly to

284

greater than 100% RE. A possible explanation for this observation is that initially most of the

285

increase in regeneration efficiency is associated with the oxidation of the MB adsorbate, but as

286

the proportion of the surface loaded with MB decreases, corrosion of the graphene accelerates.

287

Corrosion of the graphene will lead to an increase in surface area, thus increasing the

288

regeneration efficiency by creating new adsorption sites. Similar behavior has been seen

289

observed in our previous study of the electrochemical regeneration of reduced graphene oxide /

290

iron oxide composite adsorbents (Sharif et al., 2017).

291

For the G/TiO2 nanocomposites electrochemical regeneration the rapid increase in

292

regeneration efficiency above 100% was not observed, presumably due to the stability of the

293

nanocomposites. In addition to having the lowest adsorptive capacity, the G/TiO2 500 °C 13

294

nanocomposite also requires a longer regeneration time to achieve a regeneration efficiency of

295

100%. This may be due to the higher content of rutile phase TiO2 and hence lower catalytic

296

activity of the nanocomposite annealed at 500 °C (see Supporting Information Figure A.1).

297

Oxidation breakdown products may remain adsorbed on the graphene and/or the rutile TiO2,

298

blocking adsorption sites and slowing oxidation reactions, leading to the long regeneration

299

time. In contrast, the G/TiO2 400 °C nanocomposite showed more rapid regeneration, with

300

more than 80 % of the nanocomposite’s adsorptive capacity restored after only 3 minutes and

301

complete regeneration obtained after around 7 minutes. These results suggest that the higher

302

proportion of anatase phase TiO2 in the nanoparticles catalyzes the oxidation of MB, possibly

303

through the generation of hydroxyl radicals, so that a higher proportion of the current is used for

304

the degradation rate of MB (Kusmierek and Chrzescijanska, 2015). It has been shown that

305

electrochemical regeneration of organics in the presence of chloride can result in undesirable by-

306

products (Hussain et al., 2013a, 2013b; Hussain et al., 2015). Chlorinated by-products can be

307

formed due to oxidation by active chlorine (Comninellis and Nerini, 1995), however Hussain et

308

al. (2013b) have shown that these breakdown products can also be adsorbed on graphite. The

309

investigation of any by-products (such as chlorinated products) formed during electrochemical

310

regeneration of G/TiO2 nanocomposites has not yet been performed, and will be the subject of

311

future studies.

312

As the by-products of the regeneration have not been identified in this study, two different

313

pathways can be suggested for mechanism of electrochemical oxidation of the MB. One

314

possibility is direct electrochemical oxidation by hydroxyl radicals formed on the electrode

315

surface, which can be described by the following reactions (De Moura et al., 2016):

316

M + H2O → M(•OH) + H+ + e−

[1] 14

317

MB + •OH → Intermediates →→→ CO2+ H2O

[2]

318

where M is the electrode or active site. Alternatively the MB oxidation may occur by indirect

319

electrooxidation, in particular by active chlorine generated from the chloride ions present. The

320

reactions of chlorine, hypochlorous acid and hypochlorite ion formation are as follows (Bonfatti

321

et al., 2000):

322

2Cl‒ → Cl2+2e−

[3]

323

Cl2+ H2O → HOCl + H++ Cl‒

[4]

324

Cl‒+ H2O → HOCl + H++ 2e‒

[5]

325

HOCl ↔ H++ OCl‒

[6]

326

The regeneration was conducted at neutral pH, so active chlorine would be expected to be

327

present as both hypochlorous acid and hypochlorite ion. However, as the regeneration progressed

328

the pH in the bed is expected to decrease, which may shift the equilibrium reaction [6] towards

329

hyphochlorous acid. Active chlorine species, particularly hypochlorite (OCl−) and hypochlorous

330

acid (HOCl), can oxidize the MB:

331

MB+ OCl‒→ breakdown products →→→ CO2+ H2O + Cl‒

[7]

332

MB + HOCl → breakdown products →→→ CO2+ H2O + HCl

[8]

333

In addition, TiO2 has been shown to produce highly oxidative hydroxyl and oxygen radicals

334

(Shen et al., 2006), •OH and •O2, by the oxidation of water by the valence band (hVB+) at the

335

surface of TiO2. Thus the MB oxidation may occur by a combination of both indirect and direct

336

mechanisms.

337

Adsorption and regeneration of graphene annealed at 400 °C and 500 °C (G400 °C and

338

G500 °C) was also evaluated. The adsorption-regeneration was conducted using 150 mL of 25 15

339

ppm of MB solution and 0.1 g of adsorbent regenerated for 7 min (same conditions for which

340

complete regeneration was achieved for G/TiO2 400 °C). The regeneration efficiency of both

341

G400 °C and G500 °C was found to be similar to the bare graphene after 7 min of regeneration ( ̴

342

60%). It can be concluded that calcination of the bare graphene did not affect the electrochemical

343

regeneration behavior of the graphene.

344

The feasibility of the adsorption and electrochemical regeneration process, and the stability

345

of the adsorbent material, can be evaluated by carrying out successive adsorption and

346

regeneration cycles. The performance of the bare graphene and nanocomposite adsorbents was

347

determined over five successive adsorption and electrochemical regeneration cycles, as shown in

348

Figure 3c. The regeneration time for each adsorbent was based on the time required to achieve

349

100 % regeneration, using the data shown in Figure 3b. The regeneration efficiency for the bare

350

graphene increased to around 150% after the second cycle. This suggests that the number of

351

adsorption sites was increased by the electrochemical regeneration, presumably due to the

352

corrosion of the graphene. The adsorptive capacity and regeneration efficiency remain

353

approximately constant for the subsequent cycles (three to five), possibly due to the loss of

354

adsorbent mass. Table A.4 shows the mass of adsorbent recovered after each cycle for all three

355

adsorbents. For the bare graphene there is a significant loss of adsorbent, amounting to around

356

30% of the initial mass lost during 5 cycles. The volume of solution used for the adsorption

357

cycles was adjusted to maintain a constant adsorbent dose in all cycles. For the G/TiO2

358

nanocomposite adsorbents, the mass loss was much lower, corresponding to around 8% of the

359

initial mass during 5 cycles. For the bare graphene, the additional loss was presumably due to

360

corrosion, while for the G/TiO2 composite adsorbents the mass loss was attributed to physical

361

losses as it is difficult to recover all of the adsorbent after filtration. A control experiment was

16

362

carried out to estimate the physical losses from the filtration process, and these were found to be

363

around 9±1%, confirming that there was no significant mass loss due to corrosion for the G/TiO2

364

composites.

365

The regeneration efficiency of the G/TiO2 nanocomposites does not show a significant

366

change throughout the five consecutive cycles of adsorption and regeneration, remaining at

367

around 100%. These results indicate that the G/TiO2 nanocomposites are relatively stable under

368

electrochemical regeneration.

369

The amount of charge required for 100% regeneration efficiency, normalized by the mass of

370

adsorbent or MB, can be used to compare the energy consumption for the regeneration of the

371

adsorbents. In other words, a lower charge passed per unit mass indicates lower energy

372

requirement for regeneration or contaminant removal and oxidation, since the cell voltage was

373

about the same (around 3 V) in all cases. Table A.5 shows a comparison of the Graphene, G/

374

TiO2 400 °C and G/ TiO2 500 °C nanocomposites in terms of the specific charge required for

375

complete regeneration, based on the mass of MB. By considering the MB loading on the

376

adsorbents and the required time for complete regeneration, the amount of the charged passed for

377

the G/TiO2 500 °C nanocomposite is 4.3 times higher than the bare graphene and the charged

378

passed for the G/TiO2 400 °C nanocomposite is almost half of the bare graphene. This

379

demonstrates the low electrocatalytic activity of the G/TiO2 500 °C nanocomposite relative to

380

that of the G/TiO2 400 °C. The high electrocatalytic activity of the G/TiO2 400 °C

381

nanocomposite leads to much lower energy required to achieve complete regeneration compared

382

to bare graphene.

383

17

384

3.4. Electrochemical properties of the nanocomposite

385

To study the electrocatalytic activity of the nanocomposites, cyclic voltammetry studies

386

were carried out. Cyclic voltammograms were obtained for the bare graphene, MB adsorbed bare

387

graphene, G/TiO2 400 °C, G/TiO2 400 °C loaded with MB, and G/TiO2 500 °C loaded with MB,

388

and the obtained results are depicted in Figure 4a. Compared to the blank samples (bare graphene

389

and G/ TiO2 400 °C), a pair of poorly defined redox peaks can be observed for the bare graphene

390

loaded with MB at a potential of around −0.35 V versus Ag/AgCl. These redox peaks are most

391

likely associated with reduction of MB to leucomethylene on the surface of the G/TiO2 (Barou et

392

al., 2012). The peak currents for the G/TiO2 nanocomposites were different to those on bare

393

graphene, which can be explained through the difference in the electrocatalytic activity of the

394

TiO2 annealed at different temperatures. For the MB loaded G/TiO2 500 °C nanocomposite,

395

containing a higher proportion of the less active (Fuentes et al., 2008; García et al., 2007) rutile

396

phase TiO2, the peak currents was around half of that for the MB loaded bare graphene. In

397

contrast, the magnitude of the well-defined redox peak currents for MB loaded G/TiO2 400 °C

398

(with a higher proportion of the more active anatase phase of TiO2) was three times larger than

399

for the MB loaded bare graphene, due to the high catalytic activity of the anatase phase TiO2.

400

The onset potential for oxygen evolution is a significant criterion in evaluating active and

401

inactive electrodes. Figure 4b shows the linear sweep voltammetry behavior obtained for bare

402

graphene and the G/TiO2 nanocomposites. The sweep voltammetry was carried out in 0.1 M

403

Na2SO4, in order to avoid overlapping currents for chlorine and oxygen evolution. The oxygen

404

evolution potential (vs. Ag/AgCl) was estimated using extrapolations of the linear portion of the

405

current-potential curves (dotted lines) to 0 A. The onset potential for oxygen evolution was

406

estimated to be 1.65, 1.76 and 1.78 V for bare graphene, G/TiO2 400 °C, G/TiO2 500 °C (both 6 18

407

wt% TiO2), respectively. Similar results have been reported by Zhao et al. (2009) and Wang et

408

al. (2013), who found that TiO2 increased the onset potential for oxygen evolution in TiO2-SnO2

409

composites. A high onset potential for oxygen evolution can promote the oxidation efficiency of

410

organic contaminants (Xie et al., 2017).

411

412

4. Conclusions

413

G/TiO2 nanocomposites were successfully prepared using the sol-gel method and utilized in

414

an adsorption and electrochemical regeneration process. The nanocomposites had a lower

415

adsorptive capacity than bare graphene. When annealed at 500 °C, the adsorption capacity

416

decreased significantly, probably due to the loss of the amorphous carbon during the annealing.

417

TEM, analysis of the COD of rinse waters for electrochemically treated adsorbent, regeneration

418

efficiency data consistently showed that the bare graphene was rapidly corroded during

419

electrochemical oxidation, while almost no corrosion was evident in the case of the annealed

420

G/TiO2 nanocomposites.

421

In addition to reduction of corrosion rate, it was found that the G/TiO2 400 °C showed a

422

high electrocatalytic activity for oxidation of the MB. Complete regeneration of 22 mg g−1

423

adsorbed MB on the G/TiO2 400 °C nanocomposite was achieved by passing 10 mA cm−2 for 7

424

min which was much faster in comparison with the bare graphene. From cyclic voltammetry,

425

while the G/TiO2 500 °C nanocomposite was less catalytically active than bare graphene, the

426

G/TiO2 400 °C nanocomposite demonstrated excellent electrocatalytic activity. It can be inferred

427

that the addition of nanoparticles of anatase TiO2 increases the overpotential for graphene

428

oxidation and oxygen evolution, catalyzes the direct oxidation of MB, and perhaps also the

19

429

generation of hydroxyl radicals, leading to rapid regeneration by oxidation of MB adsorbed on

430

the surface of the graphene. The more rapid regeneration can be attributed to the electrocatalytic

431

activity of the TiO2 nanoparticles, particularly for the material annealed at 400 °C which

432

contained a higher proportion of anatase phase TiO2. Further work is needed to investigate the

433

detailed mechanism for MB oxidation on G/ TiO2 nanocomposites.

434

The stability of the G/TiO2 nanocomposite was investigated through successive cycles of

435

adsorption and regeneration, demonstrating that the G/TiO2 nanocomposites were more stable

436

than the bare graphene, maintaining their performance through five cycles of adsorption and

437

regeneration. The stability of the composites is more difficult to explain than the rapid

438

regeneration. The observed stability may be due to removal of the unstable amorphous carbon

439

during the annealing, or reduced overpotential during electrochemical regeneration.

440

Acknowledgement

441

This research has received financial support from the Natural Sciences and Engineering

442

Research Council of Canada (NSERC 435634-2013) and the Canada Foundation for Innovation

443

(CFI 32613).

444

20

445 446 447 448 449

Figure 1. (a) Raman spectra of Bare graphene and 6 wt% G/TiO2 nanocomposites (b) Morphological observation TEM image of G/TiO2 400 °C (6 wt% TiO2) nanocomposite (i) Low magnification, (ii) High maginification

21

450 451 452

Figure 2. XPS survey spectra of G/ TiO2; (b) High resolution C1s; (c) High resolution Ti2p; (d) High resolution O1s

453 454

22

455 456 457 458 459 460 461 462 463 464 465 466 467

Figure 3. (a) Adsorption isotherm of MB adsorption on bare graphene and the G/TiO2 nanocomposites (6 wt% TiO2). (b) Effect of regeneration time on regeneration efficiency of MB on G/TiO2 nanocomposites (6 wt % TiO2) by applying the current density of 10 mA cm-2. (C) Regeneration efficiency for a series of adsorption and electrochemical regeneration cycles for MB adsorption on bare graphene, G/TiO2 400 °C and G/TiO2 500 °C nanocomposites (both 6 wt% TiO2). Adsorption was carried out under conditions that give close to the maximum loading of MB on the adsorbent. Regeneration of the bare graphene, , G/TiO2 400 °C and G/TiO2 500 °C adsorbents were carried out for 14, 7, and 20 min respectively at a current density of 10 mA cm−2.

23

468 469 470 471 472 473 474 475

Figure 4. (a) Cyclic voltammetry of bare graphene, bare graphene loaded with MB, G/TiO2 400 °C, G/TiO2 400 °C loaded with MB and G/TiO2 500 °C loaded with MB. The TiO2 loading on the nanocomposites was 6 wt% at a scan rate of 5 mV s-1 in 1 mol L-1 NaCl. The inset is the voltammogram for bare graphene and G/TiO2 400 °C in the absence of adsorbed MB. (b) Linear sweep voltammograms of (a) bare graphene, (b) G/TiO2 400 °C, and (c) G/TiO2 500 °C in 0.1 mol L-1 Na2SO4. The loading of TiO2 was 6 wt %. Experiments were carried out at a scan rate of 100 mV S-1.

476 477 478

Supporting Information. Supporting data, including particle size distribution (determined by

479

laser diffraction), X-ray diffraction, thermogravimetric analysis, and linear sweep voltammetry

480

are provided in the supporting information document.

481 482

24

References Allen, M.J., Tung, V.C., Kaner, R.B., 2010. Honeycomb carbon: a review of graphene. Chem. Rev. 110, 132–145. Asghar, H.M.A., Hussain, S.N., Roberts, E.P.L., Campen, A.K., Brown, N.W., 2013. Pretreatment of adsorbents for waste water treatment using adsorption coupled-with electrochemical regeneration. J. Ind. Eng. Chem. 19, 1689–1696. Asghar, H.M.A., Hussain, S.N., Sattar, H., Brown, N.W., Roberts, E.P.L., 2014. Electrochemically synthesized GIC-based adsorbents for water treatment through adsorption and electrochemical regeneration. J. Ind. Eng. Chem. 20, 2200–2207. Barou, E., Bouvet, M., Heintz, O., Meunier-Prest, R., 2012. Electrochemistry of methylene blue at an alkanethiol modified electrode. Electrochim. Acta 75, 387–392. Beck, F., Kaiser, W., Krohn, H., 2000. Boron doped diamond (BDD)-layers on titanium substrates as electrodes in applied electrochemistry. Electrochim. Acta 45, 4691–4695. \ Berenguer, R., Marco-Lozar, J.P., Quijada, C., Cazorla-Amorós, D., Morallón, E., 2010. Electrochemical regeneration and porosity recovery of phenol-saturated granular activated carbon in an alkaline medium. Carbon. 48, 2734–2745. Bonfatti, F., Ferro, S., Lavezzo, F., Malacarne, M., Lodi, G., De Battisti, A., 2000, Electrochemical Incineration of Glucose as a Model Organic Substrate. II. Role of Active Chlorine Mediation J. Electrochem. Soc. 147, 592-596. Boye, B., Dieng, M.M., Brillas, E., 2002. Degradation of herbicide 4-chlorophenoxyacetic acid by advanced electrochemical oxidation methods. Environ. Sci. Technol. 36, 3030–3035.

Brillas, E., Martínez-Huitle, C.A., 2011. Synthetic diamond films: preparation, electrochemistry, characterization, and applications. John Wiley & Sons. Brown, N.W., Roberts, E.P.L., Garforth, A.A., Dryfe, R.A.W., 2004a. Electrochemical regeneration of a carbon-based adsorbent loaded with crystal violet dye. Electrochim. Acta 49, 3269–3281. Brown, N.W., Roberts, E.P.L., Garforth, A.A., Dryfe, R.A.W., 2004b. Treatment of dyehouse effluents with a carbon-based adsorbent using anodic oxidation regeneration. Water Sci. Technol. 49, 219–225. Brown, N.W., Roberts, E.P.L., 2007. Electrochemical pre-treatment of effluents containing chlorinated compounds using an adsorbent. J. Appl. Electrochem. 37, 1329–1335. Chandra, V., Kim, K.S., 2011. Highly selective adsorption of Hg2+ by a polypyrrole-reduced graphene oxide composite. Chem. Commun. 47, 3942–3944. Cheng, X., Yu, X., Xing, Z., 2012. One-step synthesis of visible active CNS-tridoped TiO2 photocatalyst from biomolecule cystine. Appl. Surf. Sci. 258, 7644–7650. Comninellis, C., 1994. Electrocatalysis in the electrochemical conversion/combustion of organic pollutants for waste water treatment. Electrochim. Acta 39, 1857–1862. Comninellis, C., Nerini, A., 1995. Anodic oxidation of phenol in the presence of NaCl for wastewater treatment. J. Appl. Electrochem. 25, 23–28. Dai, Y.-D., Yuan, C., Huang, C.P., Chiang, P.-C., 2017. Regeneration of spent carbon nanotubes by electrochemical oxidation over RuO2/Ti electrode. Sep. Purif. Technol. 178, 207–214.

26

De Moura, D.C., Quiroz, M.A., Da Silva, D.R., Salazar, R., Martínez-Huitle, C.A., 2016. Electrochemical degradation of Acid Blue 113 dye using TiO2-nanotubes decorated with PbO2 as anode. Environ. Nanotechnology, Monit. Manag. 5, 13–20. Evdokimov, S. V, 2002. Electrochemical and corrosion behavior of electrode materials based on compositions of ruthenium dioxide and base-metal oxides. Russ. J. Electrochem. 38, 583– 588. Falamaki, C., Veysizadeh, J., 2007. A new inorganic membrane: TiO2 ultrafilter based on functionally fraded porous alumina. Mater. Sci. Forum 553, 239–244. Flores, N., Sharif, F., Yasri, N., Brillas, E., Sirés, I., Roberts, E.P.L., 2018. Removal of tyrosol from water by adsorption on carbonaceous materials and electrochemical advanced oxidation processes. Chemosphere 201, 807–815. Fuentes, R.E., Garcia, B.L., Weidner, J.W., 2008. A Nb-doped TiO2 electrocatalyst for use in direct methanol fuel cells. ECS Trans. 12, 239–248. Gao, Y., Li, Y., Zhang, L., Huang, H., Hu, J., Shah, S.M., Su, X., 2012. Adsorption and removal of tetracycline antibiotics from aqueous solution by graphene oxide. J. Colloid Interface Sci. 368, 540–546. García, B.L., Fuentes, R., Weidner, J.W., 2007. Low-temperature synthesis of a PtRu⁄ Nb0.1Ti0 .9O2

electrocatalyst for methanol oxidation. electrochem. Solid-State Lett. 10, B108–B110.

Gautam, R.K., Chattopadhyaya, M.C., 2016. Advanced nanomaterials for wastewater remediation. CRC Press.

27

Guo, X., Li, D., Wan, J., Yu, X., 2015. Preparation and electrochemical property of TiO2/Nanographite composite anode for electro-catalytic degradation of ceftriaxone sodium. Electrochim. Acta 180, 957–964. Hanaor, D.A.H., Sorrell, C.C., 2011. Review of the anatase to rutile phase transformation. J. Mater. Sci. 46, 855–874. Hussain, S.N., Asghar, H.M.A., Campen, A.K., Brown, N.W., Roberts, E.P.L., 2013a. Breakdown products formed due to oxidation of adsorbed phenol by electrochemical regeneration of a graphite adsorbent. Electrochim. Acta 110, 550–559. Hussain, S N, Asghar, H.M.A., Sattar, H., Brown, N.W., Roberts, E.P.L., 2015. Removal of tartrazine from water by adsorption with electrochemical regeneration. Chem. Eng. Commun. 202, 1280–1288. Hussain, S.N., Asghar, H.M.A., Sattar, H., Brown, N.W., Roberts, E.P.L., 2015. Chlorinated breakdown products formed during oxidation of adsorbed phenol by electrochemical regeneration of a graphite intercalation compound. J. Ind. Eng. Chem. 30, 212–219. Hussain, S.N., Roberts, E.P.L., Asghar, H.M.A., Campen, A.K., Brown, N.W., 2013b. Oxidation of phenol and the adsorption of breakdown products using a graphite adsorbent with electrochemical regeneration. Electrochim. Acta 92, 20–30. Iniesta, J., Michaud, P.A., Panizza, M., Cerisola, G., Aldaz, A., Comninellis, C., 2001. Electrochemical oxidation of phenol at boron-doped diamond electrode. Electrochim. Acta 46, 3573–3578.

28

Jasmann, J.R., Borch, T., Sale, T.C., Blotevogel, J., 2016. Advanced electrochemical oxidation of 1, 4-dioxane via dark catalysis by novel titanium Dioxide (TiO2) Pellets. Environ. Sci. Technol. 50, 8817–8826. Kesselman, J.M., Weres, O., Lewis, N.S., Hoffmann, M.R., 1997. Electrochemical production of hydroxyl radical at polycrystalline Nb-Doped TiO2 electrodes and estimation of the partitioning between hydroxyl radical and direct hole oxidation pathways. J. Phys. Chem. B 101, 2637–2643. Kraft, A., 2007. Doped diamond: a compact review on a new, versatile electrode material. Int. J. Electrochem. Sci 2, 355–385. Kusmierek, E., Chrzescijanska, E., 2015. Application of TiO2–RuO2/Ti electrodes modified with WO3 in electro- and photoelectrochemical oxidation of Acid Orange 7 dye. J. Photochem. Photobiol. A Chem. 302, 59–68. Lei, X.F., Xue, X.X., Yang, H., Chen, C., Li, X., Pei, J.X., Niu, M.C., Yang, Y.T., Gao, X.Y., 2015. Visible light-responded C, N and S co-doped anatase TiO2 for photocatalytic reduction of Cr(VI). J. Alloys Compd. 646, 541–549. Li, X., Wu, Y., Zhu, W., Xue, F., Qian, Y., Wang, C., 2016. Enhanced electrochemical oxidation of synthetic dyeing wastewater using SnO2-Sb-doped TiO2-coated granular activated carbon electrodes with high hydroxyl radical yields. Electrochim. Acta 220, 276–284. Lin, W.-C., Chen, C.-H., Tang, H.-Y., Hsiao, Y.-C., Pan, J.R., Hu, C.-C., Huang, C., 2013. Electrochemical photocatalytic degradation of dye solution with a TiO2-coated stainless steel electrode prepared by electrophoretic deposition. Appl. Catal. B Environ. 140–141, 32–41. 29

Madadrang, C.J., Kim, H.Y., Gao, G., Wang, N., Zhu, J., Feng, H., Gorring, M., Kasner, M.L., Hou, S., 2012. Adsorption behavior of EDTA-graphene oxide for Pb (II) removal. ACS Appl. Mater. Interfaces 4, 1186–1193. Marselli, B., Garcia-Gomez, J., Michaud, P.-A., Rodrigo, M.A., Comninellis, C., 2003. Electrogeneration of hydroxyl radicals on boron-doped diamond electrodes. J. Electrochem. Soc. 150, D79–D83. Ming, H., Huang, H., Pan, K., Li, H., Liu, Y., Kang, Z., 2012. C/TiO2 nanohybrids co-doped by N and their enhanced photocatalytic ability. J. Solid State Chem. 192, 305–311. Narbaitz, R.M., Cen, J., 1994. Electrochemical regeneration of granular activated carbon. Water Res. 28, 1771–1778. Narbaitz, R.M., Karimi
Jashni, A., 2009. Electrochemical regeneration of granular activated carbons loaded with phenol and natural organic matter. Environ. Technol. 30, 27–36. Nkrumah-Amoako, K., Roberts, E.P.L., Brown, N.W., Holmes, S.M., 2014. The effects of anodic treatment on the surface chemistry of a graphite intercalation compound. Electrochim. Acta 135, 568–577. Ntsendwana, B., Mamba, B.B., Sampath, S., Arotiba, O.A., 2013. Synthesis, characterisation and application of an exfoliated graphite-diamond composite electrode in the electrochemical degradation of trichloroethylene. RSC Adv. 3, 24473–24483. Ohsaka, T., Izumi, F., Fujiki, Y., 1978. Raman spectrum of anatase, TiO2. J. Raman Spectrosc. 7, 321–324.

30

Panizza, M., Cerisola, G., 2009. Direct And Mediated Anodic Oxidation of Organic Pollutants. Chem. Rev. 109, 6541–6569. Panizza, M., Cerisola, G., 2003. Influence of anode material on the electrochemical oxidation of 2-naphthol: Part 1. Cyclic voltammetry and potential step experiments. Electrochim. Acta 48, 3491–3497. Peleyeju, M.G., Umukoro, E.H., Babalola, J.O., Arotiba, O.A., 2016. Electrochemical degradation of an anthraquinonic dye on an expanded graphite-diamond composite electrode. Electrocatalysis 7, 132–139. Qin, Y., Long, M., Tan, B., Zhou, B., 2014. RhB Adsorption performance of magnetic adsorbent Fe3O4/RGO composite and its regeneration through a fenton-like reaction. Nano-Micro Lett. 6, 125–135. Ramqvist, L., Hamrin, K., Johansson, G., Fahlman, A., Nordling, C., 1969. Charge transfer in transition metal carbides and related compounds studied by ESCA. J. Phys. Chem. Solids 30, 1835–1847. Ren, F., Zhou, X.-D., Liu, Y.-C., Wang, Y.-Q., Cai, G.-X., Xiao, X.-H., Dai, Z.-G., Li, W.-Q., Yan, S.-J., Wu, W., 2013. Fabrication and properties of TiO2 nanofilms on different substrates by a novel and universal method of Ti-ion implantation and subsequent annealing. Nanotechnology 24, 255603. Rueffer, M., Bejan, D., Bunce, N.J., 2011. Graphite: An active or an inactive anode? Electrochim. Acta 56, 2246–2253.

31

Sharif, F., Gagnon, L.R., Mulmi, S., Roberts, E.P.L., 2017. Electrochemical regeneration of a reduced graphene oxide/magnetite composite adsorbent loaded with methylene blue. Water Res. Sharif, F., Roberts, E., 2015. Study of graphene FOAM characteristics: adsorption and electrochemical regeneration, in: Meeting Abstracts. The Electrochemical Society, p. 1197. Shen, Z.M., Wu, D., Yang, J., Yuan, T., Wang, W.H., Jia, J.P., 2006. Methods to improve electrochemical treatment effect of dye wastewater. J. Hazard. Mater. 131, 90–97. Stankovich, S., Dikin, D.A., Piner, R.D., Kohlhaas, K.A., Kleinhammes, A., Jia, Y., Wu, Y., Nguyen, S.T., Ruoff, R.S., 2007. Synthesis of graphene-based nanosheets via chemical reduction of exfoliated graphite oxide. Carbon N. Y. 45, 1558–1565. Swamy, V., Muddle, B.C., Dai, Q., 2006. Size-dependent modifications of the Raman spectrum of rutile Ti O 2. Appl. Phys. Lett. 89, 163118. Wang, L., Balasubramanian, N., 2009. Electrochemical regeneration of granular activated carbon saturated with organic compounds. Chem. Eng. J. 155, 763–768. Wang, L., Zhang, R., Jansson, U., Nedfors, N., 2015. A near-wearless and extremely long lifetime amorphous carbon film under high vacuum. Sci. Rep. 5, 11119. Wang, L., Zhao, Y., Fu, J., 2008. The influence of TiO2 and aeration on the kinetics of electrochemical oxidation of phenol in packed bed reactor. J. Hazard. Mater. 160, 608–613. Wang, Q., Jin, T., Hu, Z., Zhou, L., Zhou, M., 2013. TiO2-NTs/SnO2-Sb anode for efficient electrocatalytic degradation of organic pollutants: Effect of TiO2-NTs architecture. Sep. Purif. Technol. 102, 180–186.

32

Weng, C.-H., Hsu, M.-C., 2008. Regeneration of granular activated carbon by an electrochemical process. Sep. Purif. Technol. 64, 227–236. Xie, R., Meng, X., Sun, P., Niu, J., Jiang, W., Bottomley, L., Li, D., Chen, Y., Crittenden, J., 2017. Electrochemical oxidation of ofloxacin using a TiO2-based SnO2Sb/polytetrafluoroethylene resin-PbO2 electrode: Reaction kinetics and mass transfer impact. Appl. Catal. B Environ. 203, 515–525. 7 Xu, J., Wang, L., Zhu, Y., 2012. Decontamination of bisphenol A from aqueous solution by graphene adsorption. Langmuir 28, 8418–8425. Yang, Y., Hoffmann, M.R., 2016. Synthesis and stabilization of blue-black TiO2 nanotube arrays for electrochemical oxidant generation and wastewater treatment. Environ. Sci. Technol. 50, 11888–11894. Yanyan, G., Daobo, Z., Dafeng, Z., Xipeng, P., Xin, S., Changhua, S., Xianxian, Y., Songsong, L., 2013. Thermal regeneration of recyclable reduced graphene oxide/Fe3O4 composites with improved adsorption properties. J. Chem. Technol. Biotechnol. 89, 1859–1865. Yu, J., Chua, D.H.C., 2010. Enhanced field emission properties of hydrogenated tetrahedral amorphous carbon/carbon nanotubes nanostructures. Electrochem. Solid-State Lett. 13, K80–K82. Yun-Hai, W., Qing-Yun, C., Guo, L., Xiang-Lin, L., 2012. Anodic materials with high energy efficiency for electrochemical oxidation of toxic organics in waste water. INTECH Open Access Publisher.

33

Zehtab Yazdi, A., Roberts, E.P.L., Sundararaj, U., 2016. Nitrogen/sulfur co-doped helical graphene nanoribbons for efficient oxygen reduction in alkaline and acidic electrolytes. Carbon N. Y. 100, 99–108. Zhang, Y., Cheng, Y., Chen, N., Zhou, Y., Li, B., Gu, W., Shi, X., Xian, Y., 2014. Recyclable removal of bisphenol A from aqueous solution by reduced graphene oxide–magnetic nanoparticles: Adsorption and desorption. J. Colloid Interface Sci. 421, 85–92. Zhao, G., Cui, X., Liu, M., Li, P., Zhang, Y., Cao, T., Li, H., Lei, Y., Liu, L., Li, D., 2009. Electrochemical degradation of refractory pollutant using a novel microstructured TiO2 nanotubes/Sb-doped SnO2 electrode. Environ. Sci. Technol. 43, 1480–1486. Zhao, J., Ren, W., Cheng, H.-M., 2012. Graphene sponge for efficient and repeatable adsorption and desorption of water contaminations. J. Mater. Chem. 22, 20197–20202.

34

•

Graphene TiO2 nanocomposites is an effective adsorbent for methylene blue

•

Graphene/anatase TiO2 adsorbent rapidly regenerated by anodic treatment

•

Charge required for regeneration of G/TiO2 adsorbent around half that of graphene

•

Graphene/TiO2 composites stable over 5 cycles of adsorption and regeneration

Declaration of interests ☒ The authors declare that they have no known competing financial interests or personal relationships that could have appeared to influence the work reported in this paper. ☐The authors declare the following financial interests/personal relationships which may be considered as potential competing interests: