Remediation of a winery wastewater combining aerobic biological oxidation and electrochemical advanced oxidation processes

Accepted Manuscript Remediation of a winery wastewater combining aerobic biological oxidation and electrochemical advanced oxidation processes Francisca C. Moreira, Rui A.R. Boaventura, Enric Brillas, Vítor J.P. Vilar PII:

S0043-1354(15)00102-5

DOI:

10.1016/j.watres.2015.02.029

Reference:

WR 11165

To appear in:

Water Research

Received Date: 18 November 2014 Revised Date:

26 January 2015

Accepted Date: 16 February 2015

Please cite this article as: Moreira, F.C., Boaventura, R.A.R., Brillas, E., Vilar, V.J.P., Remediation of a winery wastewater combining aerobic biological oxidation and electrochemical advanced oxidation processes, Water Research (2015), doi: 10.1016/j.watres.2015.02.029. This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and review of the resulting proof before it is published in its final form. 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.

ACCEPTED MANUSCRIPT BIODEGRADABLE FRACTION + RECALCITRANT FRACTION

EFFLUENT DISCHARGE INTO THE ENVIRONMENT OR FURTHER BIOLOGICAL OXIDATION

BIOLOGICAL OXIDATION SYSTEM

ELECTROLYTIC SYSTEM

IMMOBILIZED BIOMASS REACTOR

CONDITIONER TANK

ELECTROCHEMICAL CELL

H2SO4 NaOH AIR

AIR

RECALCITRANT FRACTION

RI PT

pH

AIR

BORONDOPED DIAMOND (BDD) ANODE

O2

CPCs STRUCTURE

AC C

EP

TE D

M AN U

SC

FI

CARBONPTFE AIRDIFFUSION CATHODE

(+) (-)

ACCEPTED MANUSCRIPT

Remediation of a winery wastewater combining aerobic biological

2

oxidation and electrochemical advanced oxidation processes

3

Francisca C. Moreira a, Rui A.R. Boaventura a, Enric Brillas b, Vítor J.P. Vilar a,*

RI PT

1

4

a

5

Departamento de Engenharia Química, Faculdade de Engenharia, Universidade do Porto, Rua Dr.

6

Roberto Frias, 4200-465 Porto (Portugal)

7

b

8

Facultat de Química, Universitat de Barcelona, Martí i Franquès 1-11, 08028 Barcelona (Spain)

9

*

SC

LSRE - Laboratory of Separation and Reaction Engineering - Associate Laboratory LSRE/LCM,

AC C

EP

TE D

M AN U

Laboratori d’Electroquímica dels Materials i del Medi Ambient, Departament de Química Física,

10

Corresponding author. Tel.: +351 918257824; Fax: +351 225081674; E-mail address:

[email protected] (Vítor J.P. Vilar) 1

ACCEPTED MANUSCRIPT Abstract

12

Apart from a high biodegradable fraction consisting of organic acids, sugars and alcohols, winery

13

wastewaters exhibit a recalcitrant fraction containing high-molecular-weight compounds as

14

polyphenols, tannins and lignins. In this context, a winery wastewater was firstly subjected to a

15

biological oxidation to mineralize the biodegradable fraction and afterwards an electrochemical

16

advanced oxidation process (EAOP) was applied in order to mineralize the refractory molecules or

17

transform them into simpler ones that can be further biodegraded. The biological oxidation led to

18

above 97% removals of dissolved organic carbon (DOC), chemical oxygen demand (COD) and 5-

19

day biochemical oxygen demand (BOD5), but was inefficient on the degradation of a bioresistant

20

fraction corresponding to 130 mg L-1 of DOC, 380 mg O2 L-1 of COD and 8.2 mg caffeic acid

21

equivalent L-1 of total dissolved polyphenols. Various EAOPs such as anodic oxidation with

22

electrogenerated H2O2 (AO-H2O2), electro-Fenton (EF), UVA photoelectro-Fenton (PEF) and solar

23

PEF (SPEF) were then applied to the recalcitrant effluent fraction using a 2.2 L lab-scale flow plant

24

containing an electrochemical cell equipped with a boron-doped diamond (BDD) anode and a

25

carbon-PTFE air-diffusion cathode and coupled to a photoreactor with compound parabolic

26

collectors (CPCs). The influence of initial Fe2+ concentration and current density on the PEF

27

process was evaluated. The relative oxidative ability of EAOPs increased in the order AO-H2O2 <

28

EF < PEF ≤ SPEF. The SPEF process using an initial Fe2+ concentration of 35 mg L-1, current

29

density of 25 mA cm-2, pH of 2.8 and 25 ºC reached removals of 86% on DOC and 68% on COD

30

after 240 min, regarding the biologically treated effluent, along with energy consumptions of 45

31

kWh (kg DOC)-1 and 5.1 kWh m-3. After this coupled treatment, color, odor, COD, BOD5, NH4+,

32

NO3- and SO42- parameters complied with the legislation targets and, in addition, a total dissolved

33

polyphenols content of 0.35 mg caffeic acid equivalent L-1 was found. Respirometry tests revealed

34

low biodegradability enhancement along the SPEF process.

35

Keywords: Winery wastewater; Biological oxidation; EAOPs; Solar photoelectro-Fenton

AC C

EP

TE D

M AN U

SC

RI PT

11

2

ACCEPTED MANUSCRIPT 36

1. Introduction Winery wastewaters are generated by the different activities carried out during processing and

38

cleaning operations in wineries. The production of this kind of effluent is seasonal, which leads to

39

significant variations in volume and organic load produced throughout the year, according to the

40

type of wine (red, white, rosé, sparkling, etc.), the phase of production (grape harvesting, crushing,

41

fermentation, aging, filtration, bottling, etc.), the processing operations and the cleaning practices.

42

Typically, winery wastewaters are characterized by pH values from 2.5 to 6.0 and chemical oxygen

43

demand (COD) values of 0.8–70 g L-1 (Petruccioli et al., 2002; Silva et al., 2011; Ioannou and

44

Fatta-Kassinos, 2013; Orescanin et al., 2013; Souza et al., 2013). The major constituents of such

45

effluents are organic contaminants like organic acids (tartaric, lactic and acetic), sugars (glucose

46

and fructose) and alcohols (ethanol and glycerol) and also recalcitrant high-molecular weight

47

compounds like polyphenols, tannins and lignins (Chapman et al., 2001). The release of winery

48

wastewaters into natural aquatic environments without adequate treatment can cause negative

49

effects on the oxygen balance, bad odors and decrease of natural photoactivity due to color and

50

turbidity.

TE D

M AN U

SC

RI PT

37

Fig. 1 shows the evolution of research articles on winery wastewaters treatment from 1995 to

52

2013. The proposals for the treatment of this kind of effluent were introduced in 1995 but the

53

interest of the scientific community on this topic rose only since 2005. Nevertheless, only 88

54

publications along all years have been carried out. The major focus was on biological treatments

55

(e.g. Petruccioli et al., 2002; López-Palau et al., 2009; Ganesh et al., 2010; Silva et al., 2011) since

56

the high biodegradability of this kind of effluents can often justifies this choice. However, such

57

treatments may not be able to degrade the bioresistant fraction of these wastewaters and hence

58

alternative treatment strategies have been investigated. Among them, the application of advanced

59

oxidation processes (AOPs) as a single stage treatment acquired significance. The main applied

60

AOPs have been (i) ozone and ozone based processes (O3, O3/UV, O3/UV/H2O2) (Lucas et al.,

AC C

EP

51

3

ACCEPTED MANUSCRIPT 2010); (ii) catalysis with titanium dioxide (TiO2) and combined with H2O2 (TiO2/H2O2) and S2O82-

62

(TiO2/S2O82-) (Lucas et al., 2009a); and (iii) Fenton’s reaction based processes like photo-Fenton

63

(Fe2+/H2O2/UV) (Mosteo et al., 2006), solar photo-Fenton (Fe2+/H2O2/UV-Vis) (Lucas et al., 2009a)

64

and ferrioxalate-induced solar photo-Fenton (Monteagudo et al., 2012). In contrast, the combination

65

of biological oxidation and AOPs was not extensively studied, only in 8 published research articles

66

(Sigge et al., 2005; Mosteo et al., 2007; Mosteo et al., 2008; Anastasiou et al., 2009; Lucas et al.,

67

2009b; Ioannou et al., 2013; Ioannou and Fatta-Kassinos, 2013; Souza et al., 2013). The AOPs have

68

been used both as pretreatment and post-treatment steps, although the recommended treatment

69

strategy for high biodegradable wastewaters, like winery effluents, combines (i) a biological

70

pretreatment to remove the biodegradable compounds; (ii) a further AOP to convert the bioresistant

71

molecules into simpler ones that are able to be further biodegraded; and (iii) a final biological

72

polishing step (Oller et al., 2011). This integrated system may lead to the total mineralization of

73

organics along with the minimization of the total treatment cost. Regarding the electrochemical

74

processes, few studies have been performed, including only electrocoagulation and electrochemical

75

oxidation methods (Kirzhner et al., 2008; Kara et al., 2013; Orescanin et al., 2013). To the best of

76

our knowledge, no reports on electrochemical AOPs (EAOPs), alone or in combination with other

77

processes, were disclosed. In fact, the application of EAOPs based on Fenton´s reaction chemistry

78

to the treatment of real wastewaters is reduced and mainly refers to electro-Fenton (EF) technology

79

(Wang et al., 2010; Zhu et al., 2011; Da Pozzo and Petrucci, 2013).

AC C

EP

TE D

M AN U

SC

RI PT

61

80

EAOPs with H2O2 electrogeneration may reduce the cost related to H2O2 consumption, which

81

is the main cost factor regarding consumables for traditional AOPs based on Fenton’s reaction

82

(Malato et al., 2009). In these EAOPs, H2O2 is directly electrogenerated at the cathode of the

83

electrochemical cell from the two-electron reduction of injected O2 via Eq. (1) (Brillas et al., 2009):

84

O2 (g) + 2 H+ + 2 e
→ H2 O2

(1)

4

ACCEPTED MANUSCRIPT 85

Good efficiencies for H2O2 generation from Eq. (1) have been described for various carbonaceous

86

cathodes, with highlighting on carbon felts (Oturan et al., 2011; Mhemdi et al., 2013) and carbon-

87

PTFE gas (O2 or air) diffusion electrodes (Brillas et al., 2003; Moreira et al., 2013; Garcia-Segura et

88

al., 2014).

RI PT

In the EAOPs, •OH adsorbed at the anode (M) surface, denoted M(•OH), are formed as

89 90

intermediate of O2 evolution from water oxidation through Eq. (2) (Brillas et al., 2008):

91

M + H2 O → M(• OH) + H+ + e


92

When a one-compartment cell is utilized, the simplest EAOP is the anodic oxidation with

93

electrogenerated H2O2 (AO-H2O2), where organic compounds are mainly oxidized by M(•OH) with

94

much higher oxidizing ability than generated H2O2. The preferred anodes are boron-doped diamond

95

(BDD) thin-film electrodes since they produce much greater amounts of weakly physisorbed

96

BDD(•OH) from Eq. (2) than other common anodes (Panizza and Cerisola, 2005).

M AN U

SC

(2)

EAOPs based on Fenton’s reaction chemistry (3) like EF, UVA photoelectro-Fenton (PEF) and

98

solar photoelectron-Fenton (SPEF) methods have shown high efficiency for contaminants removal

99

from water solutions (Brillas et al., 2009). In EF, low amounts of Fe2+ ion are added as catalyst to

100

react with electrogenerated H2O2, yielding Fe3+ along with •OH in the bulk from Fenton’s reaction

101

(3) with optimum pH of 2.8 (Brillas et al., 2009).

102

Fe2+ + H2 O2 → Fe3+ + • OH + OH


(3)

AC C

EP

TE D

97

103

The PEF and SPEF processes involve the irradiation of the contaminated solution by UVA light

104

and sunlight, respectively, which can accelerate the pollutants degradation by (i) the photoreduction

105

of photoactive Fe(III)-hydroxy complexes like FeOH2+, regenerating more Fe2+ and producing more

106

•

107

formed between Fe3+ and some organic intermediates, especially those containing the carboxylate

108

group as shown in the general Eq. (5), together with Fe2+ ion regeneration (Horváth and Stevenson,

109

1992; Zuo and Hoigne, 1992; Faust and Zepp, 1993).

OH according to Eq. (4) (Sun and Pignatello, 1993); and (ii) the direct photolysis of complexes

5

ACCEPTED MANUSCRIPT 110

Fe(OH)2+ + hv → Fe2+ + • OH

(4)

111

Fe(OOCR)2+ + hv → Fe2+ + CO2 + R•

(5)

The aim of this work was to evaluate the performance of a winery wastewater treatment

113

comprising (i) an initial biological oxidation accomplished in an immobilized biological reactor

114

(IBR) to remove the biodegradable fraction of the effluent and (ii) a further EAOP such as AO-

115

H2O2, EF, PEF or SPEF to degrade the refractory compounds. The biodegradability enhancement

116

along the SPEF process was assessed to appraise the suitability of the application of a final

117

biological oxidation. Moreover, the influence of initial Fe2+ concentration and current density (j) on

118

the PEF process was evaluated.

119

2. Experimental

120

2.1. Chemicals

M AN U

SC

RI PT

112

Iron(II) sulfate heptahydrate, used as catalyst, was of analytical grade purchased from Panreac.

122

Concentrated sulfuric acid and sodium hydroxide, both of analytical grade and used for pH

123

adjustment, were supplied by Pronalab and Merck, respectively. All the other chemicals were either

124

of HPLC grade or analytical grade supplied by VWR-Prolabo, Sigma-Aldrich, Panreac, Merck,

125

Fisher Chemical and Pronalab. Ultrapure and pure water used for analyses were obtained by a

126

Millipore® Direct-Q system (18.2 MΩ cm resistivity at 25 ºC) and a reverse osmosis system

127

(Panice), respectively.

128

2.2. Winery wastewater

AC C

EP

TE D

121

129

The winery wastewater was collected in May 2013 at a Port wine company located in the

130

northeast of Portugal. This company includes only bottling activities and the wastewater resulted

131

from the washing of a container of ruby Port wine, which was produced from different grape

132

varieties grown in the Douro Demarcated Region. Table 1 collects the main physicochemical

133

characteristics of the raw winery wastewater.

134

2.3. Experimental set-up

6

ACCEPTED MANUSCRIPT 135

2.3.1. Biological oxidation system The biological oxidation system was equipped with an IBR of 45 L capacity and a conditioner

137

tank of 50 L capacity. The IBR was a flat-bottom container packed with 62 units of propylene rings

138

(nominal diameter of 50 mm) colonized by activated sludge from an urban wastewater treatment

139

plant (WWTP) of Northern Portugal and equipped with a Hailea V-20 air pump providing an air

140

flow rate of 20 L min−1 through a ceramic air diffuser. The conditioner tank was a flat-bottom

141

vessel equipped with a mechanical stirrer (Timsa) and control units for dissolved oxygen (Crison,

142

electrode and OXI49P controller) and pH (Crison, electrode and PH27P controller) in order to keep

143

these parameters in a selected range. H2SO4 or NaOH was added by means of two metering pumps

144

(Dosapro Milton Roy, series GTM A).

145

2.3.2. EAOPs system

M AN U

SC

RI PT

136

Electrochemical experiments were performed in a lab-scale flow plant with 2.2 L of total

147

capacity mainly composed of: (i) a thermostatically controlled 1.5 L cylindrical glass vessel under

148

vigorous magnetic stirring at 400 rpm; (ii) a photoreactor consisting of a borosilicate tube allocated

149

in the focus of two stainless steel reflectors, i.e. double compound parabolic collector (CPC), with

150

694 mL of irradiated volume; and (iii) an electrochemical filter-press MicroFlowCell reactor from

151

ElectroCell (Tarm, Denmark) with a 10 cm2 BDD anode and a 10 cm2 carbon-PTFE air-diffusion

152

cathode. A detailed description of the characteristics of these components has been previously

153

reported in Moreira et al. (2014). In the PEF trials, the irradiation was provided by a Philips TL

154

6W/08 fluorescent blacklight blue lamp, which emits UVA light in the wavelength region between

155

350 and 410 nm with λmax at 360 nm, allocated in the middle of the borosilicate tube and protected

156

by a concentric inner quartz tube. In the SPEF runs, the photoreactor was tilted 41º (local latitude),

157

the top reflector was removed and the solar radiation was measured by a global UV radiometer

158

(Kipp & Zonen B.V., model CUV5) placed at the same angle, which provides the incident energy in

159

W m-2 from 280 to 400 nm. Moreira et al. (2014) determined the photonic flux reaching the system

AC C

EP

TE D

146

7

ACCEPTED MANUSCRIPT in PEF trials by actinometry based on 2-NB concentration and similar actinometric measurements

161

were here accomplished under different solar UV radiations between 18.5 and 50.0 W m-2. A

162

photonic flux of 0.65±0.04 J s-1 was determined in the PEF system (Moreira et al., 2014) and values

163

from 0.44±0.02 to 0.82±0.08 J s-1 were achieved under SPEF conditions. Additionally, a linear

164

correlation between the solar UV radiation (IUV, in W m-2) and the photonic flux (Ep, in J s-1)

165

reaching the system could be established (Ep = (0.0121±0.0001)×IUV + (0.216±0.002); coefficient of

166

determination (R2) = 1.00; residual variance (S2R) = 1.27×10-6 J2 s-2). Taking into account the

167

previous calculations, the accumulated UV energy (QUV,n, in kJ L-1) inside the reactor in a time

168

interval ∆t per unit of volume of solution was determined via Eqs. (6) and (7) for the PEF and SPEF

169

systems, respectively:

170

QUV,n = 0.65

171

G,n + 0.216) QUV,n = QUV,n-1 + (0.0121UV

172

where 0.65 is the photonic flux reaching the PEF system (in J s-1), tn is the time corresponding to the

173

G,n is the n sample (in s), Vs is the solution volume (in L), 1000 is a conversion factor (in J kJ-1), UV

174

G,n + average solar UV radiation (in W m-2) measured during the period ∆tn (in s) and (0.0121UV

175

0.2161) is the correlation between the solar UV radiation (in W m-2) and the photonic flux (in J s-1)

176

valid for 18.5-50.0 W m-2.

177

2.4. Experimental procedure

178

2.4.1. Biological oxidation procedure

tn

∆tn

Vs ×1000

;

∆tn = tn -tn-1

(6) (7)

AC C

EP

TE D

Vs ×1000

M AN U

SC

RI PT

160

179

A volume of 40 L of raw winery wastewater was added to the aerobic biological system and

180

recirculated in batch mode for 10 days at a flow rate of 6.6 L min-1. The pH was maintained

181

between 6.5 and 7.5, the dissolved oxygen in a 2-4 mg O2 L−1 range and temperature values from 20

182

to 30 ºC were registered.

183

2.4.2. EAOPs procedure

8

ACCEPTED MANUSCRIPT The temperature controller was switched on at a temperature set-point that permitted to

185

preserve the wastewater temperature at 25 °C. A volume of 1.295 L of biologically treated winery

186

wastewater was added to the glass vessel and homogenized by recirculation during 10 min in the

187

dark (a 15 mL first control sample was taken for further characterization). The pH was adjusted to

188

2.8 and the solution was homogenized during 10 min in the dark (a 15 mL second sample was

189

taken). Afterwards, 20-70 mg Fe2+ L-1 were added and the solution was homogenized for another 10

190

min in the dark (a 15 mL third sample was taken). j was set at 10-100 mA cm-2 and in light assisted

191

EAOPs the radiation was simultaneously provided (in PEF experiments the UVA lamp was

192

switched on and in SPEF trials the CPCs structure was uncovered). Samples of 20 or 25 mL were

193

taken at different time intervals to evaluate the degradation process. To remove impurities of the

194

BDD surface and activate the cathode, the electrodes were previously polarized in a 7.0 g L-1

195

Na2SO4 solution at 100 mA cm-2 for 180 min.

196

2.5. Analytical determinations

M AN U

SC

RI PT

184

Total dissolved carbon (TDC) and dissolved inorganic carbon (DIC) were separately

198

determined by catalytic combustion at 680 °C and acidification, respectively, both using a non-

199

dispersive infrared detector (NDIR) in a TOC-VCSN analyzer equipped with an ASI-V autosampler

200

(Shimadzu). From these data, DOC was given by the difference between TDC and DIC. Total

201

dissolved nitrogen was determined in the same analyzer coupled with a TNM-1 unit. The energy

202

consumption per unit DOC mass (ECDOC, in kWh (kg DOC)-1) and per unit volume (EC, in kWh m-

203

3

204

ECDOC =

205

EC =

AC C

EP

TE D

197

) were obtained from Eqs. (8) and (9), respectively (Flox et al., 2007): 1000 Ecell I t

Vs ∆(DOC)exp

Ecell I t Vs

(8) (9)

9

ACCEPTED MANUSCRIPT 206

where 1000 is a conversion factor (in mg g-1), Ecell is the average applied cell voltage (in V), I is the

207

applied current (in A), t is the electrolysis time (in h), Vs is the solution volume (in L) and

208

∆(DOC)exp is the experimental DOC concentration decay (in mg L-1). The pH and temperature were determined by a WTW inoLab 730 laboratory meter.

210

Conductivity was measured by a HANNA Instruments HI 9828 Multiparameter meter. Alkalinity,

211

turbidity, COD, 5-day biochemical oxygen demand (BOD5), total suspended solids (TSS), volatile

212

suspended solids (VSS), total nitrogen and total phosphorous were measured according to the

213

Standard Methods for the Examination of Water and Wastewater (Clesceri et al., 2005). H2O2

214

concentration was measured by the colorimetric metavanadate method (Nogueira et al., 2005).

215

Fe2+, Fe3+ and total dissolved iron concentration were determined according to the colorimetric

216

1,10-phenantroline standardized procedure (ISO6332:1998, 1998). Total dissolved polyphenols

217

concentration, expressed in terms of caffeic acid equivalent concentration (mg caffeic acid

218

equivalent L-1), was determined by spectrophotometry using the Folin-Ciocalteu reagent (Merck)

219

(Folin and Ciocalteu, 1927). UV-Vis measurements were carried out using a VWR UV-6300PC and

220

a Merck Spectroquant® Pharo 100 spectrophotometers.

TE D

M AN U

SC

RI PT

209

A 28-day biodegradability Zahn-Wellens test was performed according to the Test Guideline

222

no. 302 B (OECD, 1992). (i) A volume of 240 mL of sample at neutral pH, (ii) activated sludge

223

from an urban WWTP of Northern Portugal previously centrifuged and (iii) mineral nutrients

224

(KH2PO4, K2HPO4, Na2HPO4, NH4Cl, CaCl2, MgSO4 and FeCl3) were added to an open glass

225

vessel magnetically stirred and kept in the dark at 25 ºC. Reference and blank experiments were

226

prepared using the highly biodegradable glucose and pure water, respectively, instead of sample.

227

The percentage of biodegradation (Dt) was calculated from Eq. (10) (OECD, 1992):

228

Dt = 1 − C T
C B  ×100

AC C

EP

221

C
C A

(10)

BA

10

ACCEPTED MANUSCRIPT 229

where CT and CB are sample and blank DOC concentrations (in mg L-1) determined at the sampling

230

time t, respectively, and CA and CBA are sample and blank DOC concentrations (in mg L-1)

231

measured 3 h after the beginning of the test, respectively. Respirometry tests were carried out with a BM-Advance analyzer from Surcis, S.L. (Barcelona,

233

Spain). The reactor vessel was loaded with 1000 mL of activated sludge from an urban WWTP of

234

Northern Portugal previously aerated for 24 h and with the addition of 2 mg N-allylthiourea per g of

235

VSS to stop the nitrification process. The activated sludge was subjected to continuous agitation,

236

aeration and recirculation by means of a peristaltic pump and the temperature and pH were

237

maintained at 20 ºC and 7.0±0.2, respectively, during all the trial. Firstly, the heterotrophic biomass

238

yield coefficient (YH) was calculated via Eq. (11) by the addition of an acetate solution with known

239

COD (CODacetate, in mg O2 L-1) and a R test was performed to determine the total consumed oxygen

240

(COT, in mg O2 L-1) to biodegrade the acetate solution. Afterwards, 34-50 mL of sample (winery

241

wastewater) were added to the reactor vessel and another R test was accomplished to determine the

242

COT to biodegrade the sample. Then, the biodegradable fraction of COD (bCOD, in mg O2 L-1) was

243

calculated from Eq. (12) using this COT value and the previously determined YH . The ratio between

244

bCOD and the COD measured according to Clesceri et al. (2005) gives information on the sample

245

biodegradability: values below 0.05 indicate that the sample is not biodegradable, values between

246

0.05 and 0.1 correspond to low biodegradable samples, for ratios higher than 0.1 up to 0.3 the

247

samples are considered as biodegradable and values higher than 0.3 point to very biodegradable

248

samples (Ballesteros Martín et al., 2010).

249

YH = 1 − COT /CODacetate

(11)

250

bCOD = COT /(1 − YH )

(12)

AC C

EP

TE D

M AN U

SC

RI PT

232

251

Low-molecular-weight carboxylic acids (LMCA) were identified and quantified by ion-

252

exclusion HPLC using a VWR Hitachi ELITE LaChrom (Merck-Hitach, Tokyo, Japan) fitted with a

253

diode array detector (DAD) and a RezexTM ROA-Organic Acid H+ (8%) 300 mm × 7.8 mm

11

ACCEPTED MANUSCRIPT (Phenomenex) column at ambient temperature (25 ºC). The mobile phase was 0.0025 M sulfuric

255

acid at a flow rate of 0.5 mL min-1. Samples of 10 µL were injected into the HPLC and the DAD

256

was set at λ = 210 nm. Before HPLC analysis, 1 M methanol, a well-known •OH scavenger (k•OH =

257

9.7×108 M-1 s-1) (Buxton et al., 1988), was added to samples in order to stop the mineralization

258

process.

RI PT

254

Inorganic anions released during the mineralization process were analyzed by ion

260

chromatography using a Dionex ICS-2100 LC equipped with a IonPac® AS11-HC 250 mm × 4 mm

261

column at 30 ºC and an anion self-regenerating suppressor (ASRS® 300, 4 mm) under isocratic

262

elution of 30 mM NaOH at a flow rate of 1.5 mL min-1. Inorganic cations were also determined by

263

ion chromatography but using a Dionex DX-120 LC equipped with a IonPac® CS12A 250 mm × 4

264

mm column at ambient temperature and a cation self-regenerating (CSRS® Ultra II, 4 mm)

265

suppressor under isocratic elution of 20 mM methanesulfonic acid at a flow rate of 1.0 mL min-1.

266

Analyses were performed by injecting 10 and 25 µL aliquots for anions and cations, respectively.

M AN U

SC

259

Before DOC, total dissolved nitrogen, total dissolved iron, total dissolved polyphenols,

268

inorganic ions and carboxylic acids analysis, the samples were filtered with 0.45 µm Nylon filters

269

from Whatman.

270

2.6. Model parameters estimation

EP

TE D

267

A pseudo-first-order kinetic model was fitted to the DOC data as a simple mathematical model

272

from which properly kinetic constants could be calculated to quantitatively compare the DOC decay

273

under distinct conditions. The kinetic model was adjusted by a nonlinear regression method using

274

Fig.P software for Windows from Biosoft. The pseudo-first-order kinetic constant (kDOC), in min−1,

275

was calculated as follows:

276

[DOC]t = [DOC]0 × e-kDOC × t

277

where [DOC]0 and [DOC]t are DOC concentrations at time 0 and after t time, respectively.

AC C

271

(13)

12

ACCEPTED MANUSCRIPT The resulting kinetics parameters were found by minimizing the sum of the squared deviations

279

between experimental and predicted values. The goodness of all these parameters was assessed by

280

calculating the relative standard deviations, R2 and S2R.

281

3. Results and discussion

282

3.1. Characterization of the raw winery wastewater

RI PT

278

The raw winery wastewater exhibited the following main characteristics: (i) dark violet color;

284

(ii) strong odor; (iii) acid pH of 3.7; (iv) high organic content; (v) high biodegradability; (vi) low

285

nitrogen content; (vii) moderate total dissolved polyphenols content of 41 mg caffeic acid

286

equivalent L-1; (viii) moderate alkalinity of 545 mg CaCO3 L-1; (ix) moderate conductivity of 3178

287

μS cm-1 and moderate ionic content corresponding to a calculated ionic strength of 8.5 × 10-3 M; (x)

288

absence of phosphate; (xi) low total phosphorous concentration of 8.2 mg L-1; and (xii) absence of

289

dissolved iron. The complete characterization can be accessed in Table 1.

290

3.2. Aerobic biological oxidation

M AN U

SC

283

Fig. 2 depicts a removal of an organic load corresponding to 96% of DOC after 7 days of

292

aerobic biological oxidation and afterwards the mineralization rate remained almost null, achieving

293

a value of 97% after 10 days. This DOC decay came along with 97 % of COD and 98% of BOD5

294

abatements after 10 days of biological oxidation (see complete wastewater characterization after the

295

biological treatment in Table 1). The high biodegradability here attained is in agreement with the

296

99% biodegradability determined by the Zahn-Wellens test. Compounds easily biodegradable like

297

organic acids, sugars and alcohols might be mineralized throughout the treatment (Chapman et al.,

298

2001) and, in addition, the air stripping of ethanol might occur (Colin et al., 2005). Nevertheless,

299

the biological oxidation was inefficient to mineralize an organic fraction comprising 130 mg L-1 of

300

DOC, 380 mg O2 L-1 of COD and a total dissolved polyphenols content of 8.2 mg caffeic acid

301

equivalent L-1. A BOD5 value of 150 mg O2 L-1 was registered at the end of the biological

302

oxidation, which can be associated to the excretion of metabolites and/or release of products from

AC C

EP

TE D

291

13

ACCEPTED MANUSCRIPT cell lysis of the IBR microorganisms. The abovementioned results advise the application of a

304

further oxidation process like an EAOP to degrade the remaining non-biodegradable compounds

305

and hence allow to discharge this wastewater into the environment as a final effluent from a WWTP

306

according to the Portuguese legislation (Decree-Law no. 236/98) and the European Directive no.

307

91/271/CEE.

RI PT

303

While some investigations have achieved very much alike DOC, COD and BOD5 values after

309

biological treatment (Petruccioli et al., 2002; Ioannou et al., 2013), other studies have handled with

310

more biodegradable winery wastewaters. For example, Souza et al. (2013) have achieved DOC and

311

COD values of 30 mg L-1 and 84 mg O2 L-1, respectively, after 7 days of treatment also in an IBR;

312

and Ioannou et al. (2014) and Ioannou and Fatta-Kassinos (2013) have operated with winery

313

wastewaters treated in a membrane bioreactor (MBR) displaying final DOC, COD and BOD5 values

314

of 30-60 mg L-1, 120-210 mg O2 L-1 and < 5 mg O2 L-1, respectively. Consequently, the variable

315

composition of the winery effluents may allow to occasionally omit the application of a further

316

oxidation process in order to comply with the legislation targets, though the removal of the

317

bioresistant fraction of the effluent should always be taken into account due to its environmentally

318

friendly character.

TE D

M AN U

SC

308

Color and odor were also reduced along the biological oxidation, reaching standards in

320

agreement with the Portuguese discharge limits for final effluents from WWTPs. Due to the

321

addition of high amounts of NaOH to increase the pH from acidic to neutral values, an increment on

322

Na+ of ca. 360 mg L-1 took place at the end of the biological treatment whereas the concentrations

323

of the other ions remained practically constant.

AC C

EP

319

324

Neither nitrification nor denitrification occurred throughout the biological treatment (see Table

325

1). As a result, the conditions of the biological oxidation treatment must be adjusted and optimized

326

to provide the nitrogen removal. In particular, the sludge age should be increased to ensure the

327

development of nitrifying/denitrifying microorganisms (Wang et al., 2009).

14

ACCEPTED MANUSCRIPT 328

3.3. EAOPs degradation

329

3.3.1. General To apply an EAOP to a wastewater, it has to possess a conductivity large enough to transport

331

the electric charge and minimize the power consumption. In some cases, it is necessary to add a salt

332

like Na2SO4, NaCl or HClO4 to increase the conductivity. The biologically treated winery

333

wastewater exhibited a moderate conductivity of 3.0 mS cm-1 (see Table 1) that increased to ca. 4.6

334

mS cm-1 after acidification to pH 2.8, which allowed to directly apply an EAOP. However, this

335

conductivity value was lower than the 8.6 mS cm-1 exhibited by the 7.0 g Na2SO4 L-1 solution

336

commonly applied as background electrolyte in EAOPs (Flox et al., 2007; Moreira et al., 2014).

SC

RI PT

330

All trials were performed at 25 ºC (ambient temperature) and pH of 2.8 since this pH value is

338

often assumed as optimal because iron precipitation does not take place yet and the dominant iron

339

species in solution is Fe(OH)2+, which is the most photoactive ferric iron–water complex

340

(Pignatello, 1992; Safarzadeh-Amiri et al., 1996). For all the assays, the pH adjustment from 8.3 to

341

2.8 consumed high amounts of H2SO4 due to the high alkalinity of the biologically treated

342

wastewater (1561 mg CaCO3 L-1) and induced (i) the reduction of 93-99% of DIC; (ii) the decrease

343

of 16-28% of DOC; (iii) the formation of high amounts of foam and slight amounts of precipitate;

344

and (v) no modification on TSS and turbidity values along with a color change to only a slightly

345

lighter brown. From these results, one can suggest the following occurrences during the

346

acidification step: (i) conversion of carbonates and/or bicarbonates into CO2; (ii) retention of

347

dissolved organic compounds (and perhaps some inorganic dissolved matter) into the foam; and (iii)

348

precipitation in low extent of dissolved and/or suspended organic and/or inorganic compounds.

349

Furthermore, the addition of H2SO4 came along with a large increase of SO42- concentration from

350

46 mg L-1 to 1.6 g L-1. Although this SO42- content was below the discharge limit for WWTPs final

351

effluents according to the Portuguese legislation (2.0 g SO42- L-1), the disadvantages arising from

352

the presence of high SO42- contents are well-known: (i) the establishment of complexes of SO42-

AC C

EP

TE D

M AN U

337

15

ACCEPTED MANUSCRIPT with Fe3+ (FeSO4+ and Fe(SO4)2-), thus affecting the distribution and reactivity of the iron species

354

(Benkelberg and Warneck, 1995; Safarzadeh-Amiri et al., 1996; De Laat et al., 2004); (ii) the

355

scavenging of •OH by hydrogensulfate ion (HSO4-) with the formation of sulfate radical (SO4•−)

356

from Eq. (14), a weaker oxidant than •OH (Neta et al., 1988); and (iii) the decomposition of H2O2

357

by its reaction with SO4•− via Eqs. (15) and (16) (Neta et al., 1988). Table 2 shows the calculated

358

concentrations of some Fe3+ and SO42- species in the current systems calculated by the chemical

359

equilibrium modelling system MINEQL+ (Schecher and McAvoy, 2007). For larger SO42- contents,

360

it was theoretically predicted the formation of lower amounts of FeOH2+ (exception for 70 mg Fe3+

361

L-1) and remarkable higher quantities of HSO4-, along with the prevention or, at least, reduction of

362

iron precipitation as Fe(OH)3 (s), which can be interpreted as a benefit arising from SO42- addition.

363

Besides that, the presence of more ions in solution increased the wastewater conductivity as

364

abovementioned, reducing the power consumption.

365

•
• HSO
4 + OH → SO4 + H2 O

366

2
• + SO•
4 + H2 O2 → SO4 + H + HO2

367

• 2
+ SO•
4 + HO2 → SO4 + H + O2

TE D

M AN U

SC

RI PT

353

(14) (15) (16)

No active chlorine species may be generated at the anode because of the low Cl− concentration

369

in the winery wastewater, ca. 20 mg L-1. In a first glance this can be regarded as a drawback, but the

370

formation of undesirable toxic chloro-organic derivatives and chlorine-oxygen by-products with a

371

high health-risk in living beings might be avoided (Martínez-Huitle and Ferro, 2006).

372

3.3.2. Influence of initial Fe2+ concentration on PEF process

AC C

EP

368

373

Initial Fe2+ concentrations from 20 to 70 mg Fe2+ L-1 were tested for the PEF process using j of

374

100 mA cm-2. This range of Fe2+ concentrations was selected considering other studies on photo-

375

Fenton’s processes applied to wastewaters with similar DOC values (e.g. Anastasiou et al., 2009;

376

Soares et al., 2014). Fig. 3a shows a progressively faster DOC removal at higher Fe2+ content that

377

can be related to the increasing amount of Fe2+ initially available and regenerated both from (i) the

16

ACCEPTED MANUSCRIPT photolysis of Fe(OH)2+ through Eq. (4) and (ii) cathodic reduction of Fe3+ to Fe2+ from Eq. (17),

379

which enhances the •OH production from Fenton’s reaction (3) and subsequently its reaction with

380

the organic compounds of the winery wastewater. Nevertheless, the difference between the two

381

highest initial Fe2+ concentrations was not very emphasized, with a kDOC value only 1.2 times higher

382

for 70 mg Fe2+ L-1 in comparison with 35 mg Fe2+ L-1 (see Table 3). Taking the aforementioned

383

results into consideration, an initial Fe2+ dose of 35 mg Fe2+ L-1 can be chosen as the best initial

384

Fe2+ dose in this study.

385

Fe3+ + e
→ Fe2+

SC

RI PT

378

(17)

Fig. 3b displays a drop on the total dissolved iron concentration during the first 20-40 min of

387

reaction in an extent of 55-75%, 34-60% and 33-50% for initial iron concentrations of 20, 35 and 70

388

mg Fe2+ L-1, respectively, along with a visible formation of iron sludge and TSS increase from 72 to

389

110-172 mg L-1. The Fe3+ precipitation was not associated to sharp decays on the DOC content as

390

illustrated on Fig. 3b and, moreover, an additional test on the addition of 70 mg L-1 of Fe3+ to the

391

biologically treated effluent revealed an iron precipitation of ca. 50% and null DOC abatement.

392

From these results, one can suggest the formation Fe3+ complexes exclusively with non-dissolved

393

inorganic and/or organic compounds. Note that null/almost null Fe(OH)3 (s) formation is predicted

394

from a theoretical point of view (see Table 2).

395

3.3.3. Influence of current density on PEF process

AC C

EP

TE D

M AN U

386

396

A j range from 10 to 100 mA cm-2 was employed to assess the oxidation ability of the PEF

397

treatment using an initial Fe2+ concentration of 35 mg L-1. These j values were chosen based on

398

other EAOPs applied to wastewaters with similar DOC values (e.g. Flox et al., 2007; Moreira et al.,

399

2013). Fig. 4a exhibits increasing DOC decays for raising j values with DOC removals of 73%,

400

84% and 89% after 240 min of electrolysis for 10, 25 and 100 mA cm-2, respectively. The

401

corresponding kDOC values reported in Table 3 were 1.4 and 2.2 times higher for 100 mA cm-2 in

402

comparison with 25 and 10 mA cm-2, respectively. The quasi-steady DOC removal for times higher

17

ACCEPTED MANUSCRIPT 403

than 150 min at 100 mA cm-2 suggests the presence of compounds hardly oxidized by BDD(•OH),

404

•

405

persistent compounds, the reactions performed at 25 and 100 mA cm-2 attained an almost similar

406

DOC removal after 240 min.

OH in the bulk and/or photodecomposed by UVA radiation. Considering the presence of these

Despite the faster DOC decays with the rise in j, very high energy consumptions of 97-434

408

kWh (kg DOC)-1 for 100 mA cm-2 were obtained, which were 30-61 and 5-13 times superior than

409

the ones found at 10 and 25 mA cm-2, respectively (see Fig. 4b). In terms of energy consumptions

410

per unit volume, values of 1.4, 4.9 and 48 kWh m-3 were attained after 240 min of PEF process for

411

10, 25 and 100 mA cm-2, respectively.

SC

RI PT

407

H2O2 was accumulated in excess during all reaction time for all systems, thus guaranteeing the

413

maximum production of •OH from Fenton’s reaction (3). Large concentrations of 73-591 mg L-1

414

were available in the 100 mA cm-2 system, whereas much lower contents of 5-13 and 30-61 mg L-1

415

were accumulated for 10 and 25 mA cm-2, respectively (see Fig. 4c).

M AN U

412

Fig. 4d reveals that a j of 10 mA cm-2 was not able to degrade the total dissolved polyphenols,

417

whereas these compounds attained concentrations up to 2.6 and 0.31 mg caffeic acid equivalent L-1

418

for 25 and 100 mA cm-2 after 180 and 120 min of reaction, respectively.

TE D

416

Taking into account the above outcomes, 25 mA cm-2 can be selected as a pertinent j value for

420

the degradation of the winery wastewater since (i) a high DOC removal of 84% was achieved at 240

421

min of PEF treatment; (ii) moderate energy consumptions of 61 kWh (kg DOC)-1 and 4.9 kWh m-3

422

were spent at 240 min; (iii) the moderate content of accumulated H2O2 along the reaction may

423

ensure the presence of this species under SPEF conditions since Fe2+ regeneration is favored

424

accordingly to Eqs. (4) and (5) and hence the H2O2 consumption is expected to be greater compared

425

to that of PEF; and (iv) the total dissolved polyphenols can reach almost null values.

426

3.3.4. Comparative application of AO-H2O2, EF, PEF and SPEF processes

AC C

EP

419

18

ACCEPTED MANUSCRIPT AO-H2O2, EF, PEF and SPEF treatments were applied to the degradation of the biologically

428

treated winery wastewater using j of 25 mA cm-2 and an initial Fe2+ concentration of 35 mg L-1 in

429

EAOPs based on Fenton’s reaction chemistry. Fig. 5a shows that the relative oxidation ability to

430

remove DOC increased in the sequence AO-H2O2 < EF < PEF ≤ SPEF. As can be seen in Table 3,

431

quite similar kDOC values for SPEF and PEF were obtained and the much slower degradations for

432

EF and AO-H2O2 were not able to be described by a pseudo-first order kinetic model. The low DOC

433

abatement achieved in AO-H2O2 can be mainly related to a small reaction rate of the organic matter

434

with BDD(•OH) generated through Eq. (2). In EF, the presence of •OH in the bulk formed from

435

Fenton´s reaction (3) and their high potential to destroy organic compounds improved DOC

436

removal. The faster DOC decay in PEF suggests a crucial effect of the extra •OH production under

437

UVA light from the photolysis of Fe(OH)2+ via Eq. (4) along with the possible direct photolysis of

438

complexes formed between Fe3+ and some organic intermediates, especially those containing the

439

carboxylic acid group, according to the general Eq. (5). Surprisingly, the more potent UV intensity

440

supplied by sunlight in SPEF only had a little improvement on DOC removal, both in terms of time

441

and accumulated UV energy (see inset panel of Fig. 5a), in contrast to previous studies applied to

442

aromatic compounds (Garcia-Segura and Brillas, 2011; Moreira et al., 2013; Moreira et al., 2014).

443

This points to the formation of persistent organic intermediates hardly degraded by BDD(•OH), •OH

444

in the bulk and/or photodecomposed even under potent UV intensity from sunlight.

EP

TE D

M AN U

SC

RI PT

427

On the other hand, H2O2 was always available in all EAOPs and was accumulated in larger

446

extent in the order AO-H2O2 > EF > PEF > SPEF (see Fig. 5b), as predicted by the increasing rate

447

of Fe3+ to Fe2+ regeneration of these EAOPs. During the SPEF process, the H2O2 concentration

448

diminished to very low values, 4-9 mg L-1, when more pronounced DOC abatement was patent, 60-

449

120 min, which confirms the operation at 25 mA cm-2 as a good option since a low j might not

450

ensure the H2O2 occurrence during all the SPEF process.

AC C

445

19

ACCEPTED MANUSCRIPT 451

Fig. 5c outlines that the AO-H2O2 and EF processes were ineffective on the total dissolved

452

polyphenols degradation. In contrast, the SPEF treatment was able to reduce these compounds to

453

values as low as 0.4 mg caffeic acid equivalent L-1 after a short time of 120 min, whereas the PEF

454

process reduced polyphenols concentration up to 2.6 mg caffeic acid equivalent L-1 at 180 min. The SPEF process under the best conditions considered in the current study reached DOC and

456

COD removals of 86% and 68%, respectively, in relation to the biologically treated effluent after

457

240 min of treatment, with energy consumptions of 45 kWh (kg DOC)-1 and 5.1 kWh m-3. Fig. 5d

458

reveals that the COD decay was similar (for times below 90 min) or lower (for times above 90 min)

459

than the DOC abatement, thereby suggesting a contribution of the particulate organic compounds to

460

the COD value. A decrease of 54% in the total nitrogen content was observed during the SPEF

461

process (see Table 1), which can be linked with the retention of some non-dissolved N-compounds

462

into the precipitate during the EAOP since the total dissolved nitrogen remained unaffected. Table 1

463

also shows that at 240 min of SPEF, COD and BOD5 complied with the Portuguese and the

464

European legislation limits for discharge of WWTPs final effluents (Decree-Law no. 236/98 and

465

Directive no. 91/271/CEE, respectively), in contrast with TSS, total nitrogen and total phosphorous

466

parameters (total phosphorous only exceeded the European limit). Moreover, color, odor, NH4+,

467

NO3- and SO42- parameters were in agreement with the Portuguese targets, but pH and total

468

dissolved iron surpassed the limits. In this context, the final SPEF solution was neutralized to pH

469

6.2 with subsequent sedimentation for 30 min with a resultant supernatant effluent displaying a total

470

dissolved iron below the detection limit (0.13 mg L-1), a total phosphorous concentration of 0.35 mg

471

L-1 and 55 mg TSS L-1 but without change on total nitrogen. Therefore, besides the application of a

472

further step comprising the effluent neutralization and subsequent precipitation of the settleable

473

compounds, the initial biological oxidation should be enhanced and include nitrifying and

474

denitrifying

AC C

EP

TE D

M AN U

SC

RI PT

455

bacteria

to

promote

total

nitrogen

release

as

nitrogen

gas.

The

20

ACCEPTED MANUSCRIPT 475

neutralization/precipitation step led to the formation of 26 mL of sludge per L of treated winery

476

wastewater that require further adequate treatment. Regarding the combination of the biological oxidation and SPEF processes, very high

478

abatements on DOC, COD and BOD5 parameters of ca. 99% were attained, which is in agreement

479

with the great COD removals of 95% and above 99% accomplished by Anastasiou et al. (2009) and

480

Lucas et al. (2009b) for the combination of a biological oxidation and a further photo-Fenton or

481

Fenton process, respectively.

482

3.3.5. Evolution of generated carboxylic acids and inorganic ions during EAOPs

SC

RI PT

477

The ion-exclusion HPLC analysis on LMCA revealed the formation of oxalic (tr = 8.5 min) and

484

malic (tr = 13.2 min) acids during the AO-H2O2, EF, PEF and SPEF treatments. These acids are

485

expected to be formed from the oxidative cleavage of the benzenic ring of aromatic intermediates

486

(Brillas et al., 2009). Malic acid can be subsequently transformed into oxalic acid, which is an

487

ultimate acid that can be directly mineralized to CO2 (Vel Leitner and Doré, 1997; Oturan et al.,

488

2008; Garcia-Segura and Brillas, 2011). Malic and oxalic acids may be primordially present in

489

solution as Fe(III)-carboxylate complexes because the iron is mainly available as Fe3+ during the

490

processes. Fig. 6a illustrates that in the AO-H2O2 process the LMCA were accumulated in very low

491

amounts with a contribution never superior than 1.2% for DOC, suggesting inefficiency of this

492

process to convert high-molecular aromatic compounds into the simple LMCA, in corroboration

493

with the low mineralization attained. In contrast, in EF both acids were accumulated in larger extent

494

with maximum concentrations of ca. 10 mg C L-1 for oxalic acid (see Fig. 6b) and ca. 5 mg C L-1 for

495

malic acid (see Fig. 6c), corresponding to 24% of DOC due to LMCA at the end of the process (see

496

Fig. 6a). In PEF and SPEF, both oxalic and malic acids were accumulated in low extent, below 3.0

497

mg C L-1 (Figs. 6b and c), which might be related to a fast photolysis rate of Fe(III)-oxalate and

498

Fe(III)-malate complexes.

AC C

EP

TE D

M AN U

483

21

ACCEPTED MANUSCRIPT The mineralization of organic compounds is expected to be followed by the loss of their

500

nitrogen and sulfur atoms in the form of inorganic ions such as NH4+, NO3-, NO2- and SO42-. In the

501

SPEF process under the best conditions, NO2- was not found as expected due to their instability in

502

strong oxidant media. NH4+, NO3- and total dissolved nitrogen concentrations remained almost

503

unaffected, suggesting the inability of SPEF to degrade dissolved recalcitrant organic N-

504

compounds. As aforementioned, total nitrogen decreased in an extent of 54%, pointing to the

505

retention of non-dissolved N-compounds into the precipitate. On the other hand, SO42- was

506

gradually released into the solution up to ca. 150 mg L-1 at 240 min, which can be mainly attributed

507

to the dissolution of SO42- retained in the foam and also to a possible degradation of organic

508

compounds containing sulfur in their structure.

509

3.4. Biodegradability enhancement during SPEF process

M AN U

SC

RI PT

499

As abovementioned, alternatively to the application of a SPEF process to achieve organic loads

511

in agreement with the discharge limits, the electrochemical process can be applied as a pre-

512

treatment to transform recalcitrant compounds into simpler ones that can be subsequently

513

biodegraded, thereby reducing the overall cost of the treatment. In this context, the biodegradability

514

of the winery wastewater was assessed along the SPEF process by means of respirometry assays.

515

Fig. 7 reveals the bCOD/COD ratios attained throughout the treatment. Maximal bCOD/COD of 0.1

516

were achieved for times of 40 and 90 min, corresponding to low biodegradable samples according

517

to Ballesteros Martín et al. (2010). However, the short times employed in the respiration tests, ca.

518

30 min, could be insufficient to degrade the total content of slowly biodegradable organic matter

519

(Kümmerer et al., 2004), and, furthermore, the applied biomass was not adapted to degrade the

520

organic matrix of the winery wastewater, probably reaching lower efficiencies than an adapted

521

biomass test as Zahn-Wellens.

522

4. Conclusions

AC C

EP

TE D

510

22

ACCEPTED MANUSCRIPT The raw winery wastewater exhibited a high biodegradability and, as a consequence, the

524

biological oxidation treatment attained high DOC, COD and BOD5 removals above 97%. However,

525

the biologically treated effluent was constituted of a bioresistant fraction comprising 130 mg L-1 of

526

DOC, 380 mg O2 L-1 of COD and 8.2 mg caffeic acid equivalent L-1 of total dissolved polyphenols

527

and a subsequent EAOP was then employed to mineralize the recalcitrant compounds or transform

528

them into simpler ones. The degradation of the winery effluent by EAOPs could be efficiently

529

carried out using 25 ºC and pH of 2.8 at a current density of 25 mA cm-2 and, for Fenton’s reaction

530

based processes, an initial Fe2+ concentration of 35 mg L-1. The relative oxidative capability of

531

EAOPs increased in the order AO-H2O2 < EF < PEF ≤ SPEF, with DOC removals on the

532

biologically treated effluent of 36%, 54%, 84% and 86%, respectively, after 240 min of reaction.

533

The poor DOC removal attained in AO-H2O2 revealed a small ability of BDD(•OH) generated at the

534

anode surface to react with recalcitrant winery wastewater compounds. In EF, the production of

535

•

536

additional •OH production induced by UVA or solar radiation, respectively, along with the possible

537

direct photolysis of complexes formed between Fe3+ and some organic intermediates, led to the

538

fastest reaction rates. The SPEF process under the best conditions chosen in the present study

539

attained removals of 86% for DOC and 68% for COD regarding the biologically treated effluent

540

after 240 min of treatment, with energy consumptions of 45 kWh (kg DOC)-1 and 5.1 kWh m-3. At

541

this time of SPEF, a total dissolved polyphenols content of 0.35 mg caffeic acid equivalent L-1 was

542

found and color, odor, COD, BOD5, NH4+, NO3- and SO42- parameters complied with the European

543

and/or Portuguese legislation limits for discharge of WWTPs final effluents. However, to achieve

544

total nitrogen, total phosphorous, pH, total dissolved iron and TSS targets to discharge the winery

545

wastewater into the environment, the biological oxidation treatment must be optimized to provide

546

the removal of nitrogen and additional neutralization and precipitation steps should succeed the

M AN U

SC

RI PT

523

AC C

EP

TE D

OH in the bulk increased the mineralization process and, in PEF and SPEF processes, the

23

ACCEPTED MANUSCRIPT SPEF process. The respirometry assays revealed low biodegradability enhancement along the SPEF

548

process.

549

Acknowledgements

550

Financial support was partially provided by (i) PEst-C/EQB/LA0020/2013 project, co-financed by

551

FCT (Fundação para a Ciência e a Tecnologia) and FEDER (Fundo Europeu de Desenvolvimento

552

Regional) under COMPETE program (Programa Operacional Fatores de Competitividade) of

553

QREN (Quadro de Referência Estratégico Nacional); (ii) NORTE-07-0162-FEDER-000050

554

project, co-financed by FEDER, QREN and ON2 program (Programa Operacional Regional do

555

Norte); and (iii) funds of the Pluridisciplinar project SOLARVIN (PP-IJUP2011-46) from the

556

University of Porto. F.C. Moreira acknowledges her Ph.D. fellowship SFRH/BD/80361/2011

557

supported by FCT. V.J.P. Vilar acknowledges the FCT Investigator 2013 Programme

558

(IF/01501/2013).

AC C

EP

TE D

M AN U

SC

RI PT

547

24

ACCEPTED MANUSCRIPT References

560 561 562

Anastasiou, N., Monou, M., Mantzavinos, D., Kassinos, D., 2009. Monitoring of the quality of winery influents/effluents and polishing of partially treated winery flows by homogeneous Fe(II) photo-oxidation. Desalination 248(1–3), 836-842.

563 564 565 566

Ballesteros Martín, M.M., Casas López, J.L., Oller, I., Malato, S., Sánchez Pérez, J.A., 2010. A comparative study of different tests for biodegradability enhancement determination during AOP treatment of recalcitrant toxic aqueous solutions. Ecotoxicology and Environmental Safety 73(6), 1189-1195.

567 568 569

Benkelberg, H.-J., Warneck, P., 1995. Photodecomposition of iron(III) hydroxo and sulfato complexes in aqueous solution: wavelength dependence of OH and SO4- quantum yields. The Journal of Physical Chemistry 99(14), 5214-5221.

570 571 572

Brillas, E., Baños, M.A., Garrido, J.A., 2003. Mineralization of herbicide 3,6-dichloro-2methoxybenzoic acid in aqueous medium by anodic oxidation, electro-Fenton and photoelectroFenton. Electrochimica Acta 48(12), 1697-1705.

573 574 575

Brillas, E., Garrido, J.A., Rodríguez, R.M., Arias, C., Cabot, P.L., Centellas, F., 2008. Wastewaters by electrochemical advanced oxidation processes using a BDD anode and electrogenerated H2O2 with Fe(II) and UVA light as catalysts. Portugaliae Electrochimica Acta 26(1), 15-46.

576 577

Brillas, E., Sirés, I., Oturan, M.A., 2009. Electro-Fenton process and related electrochemical technologies based on Fenton’s reaction chemistry. Chemical Reviews 109(12), 6570-6631.

578 579 580

Buxton, G.V., Greenstock, C.L., Helman, W.P., Ross, A.B., 1988. Critical review of rate constants for reactions of hydrated electrons, hydrogen atoms and hydroxyl radicals (.OH/.O-) in aqueous solution. Journal of Physical and Chemical Reference Data 17(2), 513-886.

581 582

Chapman, J., Baker, P., Wills, S., 2001. Winery Wastewater Handbook: Production, Impacts and Management, Winetitles, Adelaide, Australia.

583 584 585

Clesceri, L.S., Greenberg, A.E., Eaton, A.D., 2005. Standard Methods for Examination of Water & Wastewater, 21st ed., American Public Health Association (APHA), American Water Works Association (AWWA) & Water Environment Federation (WEF).

586 587

Colin, T., Bories, A., Sire, Y., Perrin, R., 2005. Treatment and valorisation of winery wastewater by a new biophysical process (ECCF). Water Science & Technology 51(1), 99-106.

588 589 590

Da Pozzo, A., Petrucci, E., 2013. Sequential use of Fenton and electro-Fenton process for the oxidation of an effluent-containing hypophosphite and phosphite. Desalination and Water Treatment.

591 592 593

De Laat, J., Truong Le, G., Legube, B., 2004. A comparative study of the effects of chloride, sulfate and nitrate ions on the rates of decomposition of H2O2 and organic compounds by Fe(II)/H2O2 and Fe(III)/H2O2. Chemosphere 55(5), 715-723.

594 595 596

Faust, B.C., Zepp, R.G., 1993. Photochemistry of aqueous iron(III)-polycarboxylate complexes: roles in the chemistry of atmospheric and surface waters. Environmental Science & Technology 27(12), 2517-2522.

AC C

EP

TE D

M AN U

SC

RI PT

559

25

ACCEPTED MANUSCRIPT Flox, C., Cabot, P.L., Centellas, F., Garrido, J.A., Rodríguez, R.M., Arias, C., Brillas, E., 2007. Solar photoelectro-Fenton degradation of cresols using a flow reactor with a boron-doped diamond anode. Applied Catalysis B: Environmental 75(1-2), 17-28.

600 601

Folin, O., Ciocalteu, V., 1927. On tyrosine and tryptophane determinations in proteins. Journal of Biological Chemistry 73(2), 627-650.

602 603 604

Ganesh, R., Rajinikanth, R., Thanikal, J.V., Ramanujam, R.A., Torrijos, M., 2010. Anaerobic treatment of winery wastewater in fixed bed reactors. Bioprocess and Biosystems Engineering 33(5), 619-628.

605 606 607

Garcia-Segura, S., Brillas, E., 2011. Mineralization of the recalcitrant oxalic and oxamic acids by electrochemical advanced oxidation processes using a boron-doped diamond anode. Water Research 45(9), 2975-2984.

608 609 610

Garcia-Segura, S., Cavalcanti, E.B., Brillas, E., 2014. Mineralization of the antibiotic chloramphenicol by solar photoelectro-Fenton: From stirred tank reactor to solar pre-pilot plant. Applied Catalysis B: Environmental 144(0), 588-598.

611 612

Horváth, O., Stevenson, K.L., 1992. Charge transfer photochemistry of coordination compounds, VCH, New York.

613 614 615

Ioannou, L., Michael, C., Kyriakou, S., Fatta-Kassinos, D., 2014. Solar Fenton: from pilot to industrial scale application for polishing winery wastewater pretreated by MBR. Journal of Chemical Technology & Biotechnology 89(7), 1067-1076.

616 617 618

Ioannou, L., Velegraki, T., Michael, C., Mantzavinos, D., Fatta-Kassinos, D., 2013. Sunlight, iron and radicals to tackle the resistant leftovers of biotreated winery wastewater. Photochemical & Photobiological Sciences 12(4), 664-670.

619 620

Ioannou, L.A., Fatta-Kassinos, D., 2013. Solar photo-Fenton oxidation against the bioresistant fractions of winery wastewater. Journal of Environmental Chemical Engineering 1(4), 703-712.

621 622

ISO6332:1998, 1998. Water Quality - Determination of iron - Spectrometric Method Using 1,10Phenanthroline.

623 624

Kara, S., Gürbulak, E., Eyvaz, M., Yüksel, E., 2013. Treatment of winery wastewater by electrocoagulation process. Desalination and Water Treatment 51(28-30), 5421-5429.

625 626

Kirzhner, F., Zimmels, Y., Shraiber, Y., 2008. Combined treatment of highly contaminated winery wastewater. Separation and Purification Technology 63(1), 38-44.

627 628

Kümmerer, K., Alexy, R., Hüttig, J., Schöll, A., 2004. Standardized tests fail to assess the effects of antibiotics on environmental bacteria. Water Research 38(8), 2111-2116.

629 630

López-Palau, S., Dosta, J., Mata-Álvarez, J., 2009. Start-up of an aerobic granular sequencing batch reactor for the treatment of winery wastewater. Water Science and Technology 60(4), 1049-1054.

631 632 633

Lucas, M.S., Mosteo, R., Maldonado, M.I., Malato, S., Peres, J.A., 2009a. Solar photochemical treatment of winery wastewater in a CPC reactor. Journal of Agricultural and Food Chemistry 57(23), 11242-11248.

AC C

EP

TE D

M AN U

SC

RI PT

597 598 599

26

ACCEPTED MANUSCRIPT Lucas, M.S., Mouta, M., Pirra, A., Peres, J.A., 2009b. Winery wastewater treatment by a combined process: Long term aerated storage and Fenton's reagent. Water Science and Technology 60(4), 1089-1095.

637 638 639

Lucas, M.S., Peres, J.A., Li Puma, G., 2010. Treatment of winery wastewater by ozone-based advanced oxidation processes (O3, O3/UV and O3/UV/H2O2) in a pilot-scale bubble column reactor and process economics. Separation and Purification Technology 72(3), 235-241.

640 641 642

Malato, S., Fernández-Ibáñez, P., Maldonado, M.I., Blanco, J., Gernjak, W., 2009. Decontamination and disinfection of water by solar photocatalysis: Recent overview and trends. Catalysis Today 147(1), 1-59.

643 644

Martínez-Huitle, C.A., Ferro, S., 2006. Electrochemical oxidation of organic pollutants for the wastewater treatment: direct and indirect processes. Chemical Society Reviews 35, 1324-1340.

645 646 647

Mhemdi, A., Oturan, M.A., Oturan, N., Abdelhédi, R., Ammar, S., 2013. Electrochemical advanced oxidation of 2-chlorobenzoic acid using BDD or Pt anode and carbon felt cathode. Journal of Electroanalytical Chemistry 709, 111-117.

648 649 650

Monteagudo, J.M., Durán, A., Corral, J.M., Carnicer, A., Frades, J.M., Alonso, M.A., 2012. Ferrioxalate-induced solar photo-Fenton system for the treatment of winery wastewaters. Chemical Engineering Journal 181–182(0), 281-288.

651 652 653 654

Moreira, F.C., Garcia-Segura, S., Boaventura, R.A.R., Brillas, E., Vilar, V.J.P., 2014. Degradation of the antibiotic trimethoprim by electrochemical advanced oxidation processes using a carbonPTFE air-diffusion cathode and a boron-doped diamond or platinum anode. Applied Catalysis B: Environmental 160–161(0), 492-505.

655 656 657 658

Moreira, F.C., Garcia-Segura, S., Vilar, V.J.P., Boaventura, R.A.R., Brillas, E., 2013. Decolorization and mineralization of Sunset Yellow FCF azo dye by anodic oxidation, electroFenton, UVA photoelectro-Fenton and solar photoelectro-Fenton processes. Applied Catalysis B: Environmental 142–143(0), 877-890.

659 660 661

Mosteo, R., Lalinde, N., Ormad Maria, P., Ovelleiro, J.L., 2007. Winery wastewater treatment by heterogeneous photo-Fenton process and activated sludges. Depuración de efluentes vinícolas mediante tratamientos Foto-Fenton en fase heterogénea y lodos activos 27(280), 55-59.

662 663 664

Mosteo, R., Ormad, P., Mozas, E., Sarasa, J., Ovelleiro, J.L., 2006. Factorial experimental design of winery wastewaters treatment by heterogeneous photo-Fenton process. Water Research 40(8), 1561-1568.

665 666 667

Mosteo, R., Sarasa, J., Ormad, M.P., Ovelleiro, J.L., 2008. Sequential solar photo-Fenton-biological system for the treatment of winery wastewaters. Journal of Agricultural and Food Chemistry 56(16), 7333-7338.

668 669

Neta, P., Huie, R.E., Ross, A.B., 1988. Rate constants for reactions for inorganic radicals in aqueous solution. Journal of Physical and Chemical Reference Data Reprints 17(3), 1027-1247.

670 671

Nogueira, R.F.P., Oliveira, M.C., Paterlini, W.C., 2005. Simple and fast spectrophotometric determination of H2O2 in photo-Fenton reactions using metavanadate. Talanta 66(1), 86-91.

AC C

EP

TE D

M AN U

SC

RI PT

634 635 636

27

ACCEPTED MANUSCRIPT OECD, 1992. Guideline for testing of chemicals 302 B, Inherent biodegradability: Zahn Wellens/EMPA test, Organization of Economic Cooperation and Development, Paris.

674 675 676

Oller, I., Malato, S., Sánchez-Pérez, J.A., 2011. Combination of advanced oxidation processes and biological treatments for wastewater decontamination-A review. Science of the Total Environment 409(20), 4141-4166.

677 678 679 680

Orescanin, V., Kollar, R., Nad, K., Mikelic, I.L., Gustek, S.F., 2013. Treatment of winery wastewater by electrochemical methods and advanced oxidation processes. Journal of Environmental Science and Health - Part A Toxic/Hazardous Substances and Environmental Engineering 48(12), 1543-1547.

681 682 683

Oturan, M.A., Oturan, N., Edelahi, M.C., Podvorica, F.I., Kacemi, K.E., 2011. Oxidative degradation of herbicide diuron in aqueous medium by Fenton's reaction based advanced oxidation processes. Chemical Engineering Journal 171(1), 127-135.

684 685 686

Oturan, M.A., Pimentel, M., Oturan, N., Sirés, I., 2008. Reaction sequence for the mineralization of the short-chain carboxylic acids usually formed upon cleavage of aromatics during electrochemical Fenton treatment. Electrochimica Acta 54(2), 173-182.

687 688

Panizza, M., Cerisola, G., 2005. Application of diamond electrodes to electrochemical processes. Electrochimica Acta 51(2), 191-199.

689 690

Petruccioli, M., Cardoso Duarte, J., Eusebio, A., Federici, F., 2002. Aerobic treatment of winery wastewater using a jet-loop activated sludge reactor. Process Biochemistry 37(8), 821-829.

691 692

Pignatello, J.J., 1992. Dark and photoassisted Fe3+-catalyzed degradation of chlorophenoxy herbicides by hydrogen peroxide. Environmental Science & Technology 26(5), 944-951.

693 694

Safarzadeh-Amiri, A., Bolton, J.R., Cater, S.R., 1996. Ferrioxalate-mediated solar degradation of organic contaminants in water. Solar Energy 56(5), 439-443.

695 696

Schecher, W.D., McAvoy, D.C., 2007. MINEQL+: A Chemical Equilibrium Modeling System, Version 4.6 for Windows, Environmental Research Software, Hallowell, Maine, United States.

697 698

Sigge, G.O., Britz, T.J., Fourie, P.C., Barnardt, C.A., 2005. The efficacy of ozone as a pre- and post-treatment option for UASB-treated food processing wastewaters, pp. 167-173.

699 700 701

Silva, F., Pirra, A., Sousa, J., Arroja, L., Capela, I., 2011. Biodegradation kinetics of winerywastewater from port wine production. Chemical and Biochemical Engineering Quarterly 25(4), 493-499.

702 703 704

Soares, P., Silva, T.C.V., Manenti, D., Souza, S.A.G.U., Boaventura, R.R., Vilar, V.P., 2014. Insights into real cotton-textile dyeing wastewater treatment using solar advanced oxidation processes. Environmental Science and Pollution Research 21(2), 932-945.

705 706 707

Souza, B.S., Moreira, F.C., Dezotti, M.W.C., Vilar, V.J.P., Boaventura, R.A.R., 2013. Application of biological oxidation and solar driven advanced oxidation processes to remediation of winery wastewater. Catalysis Today 209, 201-208.

708 709

Sun, Y., Pignatello, J.J., 1993. Photochemical reactions involved in the total mineralization of 2,4-D by Fe3+/H2O2/UV. Environmental Science & Technology 27(2), 304-310.

AC C

EP

TE D

M AN U

SC

RI PT

672 673

28

ACCEPTED MANUSCRIPT Vel Leitner, N.K., Doré, M., 1997. Mecanisme d'action des radicaux OH⋅ sur les acides glycolique, glyoxylique, acetique et oxalique en solution aqueuse: Incidence sur la consammation de peroxyde d'hydrogene dans les systemes H2O2/UV et O3/H2O2. Water Research 31(6), 1383-1397.

713 714 715

Wang, C.-T., Chou, W.-L., Chung, M.-H., Kuo, Y.-M., 2010. COD removal from real dyeing wastewater by electro-Fenton technology using an activated carbon fiber cathode. Desalination 253(1–3), 129-134.

716 717

Wang, L.K., Shammas, N.K., Hung, Y.-T.E., 2009. Advanced biological treatment processes, Handbook of enviromental engineering, Vol. 9, Humana Press, New York.

718 719 720

Zhu, X., Tian, J., Liu, R., Chen, L., 2011. Optimization of Fenton and electro-Fenton oxidation of biologically treated coking wastewater using response surface methodology. Separation and Purification Technology 81(3), 444-450.

721 722 723

Zuo, Y., Hoigne, J., 1992. Formation of hydrogen peroxide and depletion of oxalic acid in atmospheric water by photolysis of iron(III)-oxalato complexes. Environmental Science & Technology 26(5), 1014-1022.

AC C

EP

TE D

M AN U

SC

RI PT

710 711 712

29

ACCEPTED MANUSCRIPT Figure captions

725

Fig. 1. Research articles on winery wastewater remediation per year arranged by treatment process

726

(source: http://www.scopus.com/, December 2013, search for “winery wastewater treatment” with

727

further manual refinement to exclude misclassified articles and distribute them by treatment

728

process).

729

Fig. 2. Assessment of the biological oxidation efficiency in the treatment of the raw winery

730

wastewater in terms of (
) DOC removal, () total dissolved polyphenols and (
) pH.

731

Fig. 3. Influence of initial Fe2+ concentration on (a) normalized DOC removal and (b) total

732

dissolved iron concentration as a function of time in PEF degradations of the winery wastewater

733

after biological oxidation using a BDD anode, 25 ºC, j = 100 mA cm-2 and pH = 2.8. [Fe2+]0: (
)

734

20, () 35 and (
) 70 mg L-1.

735

Fig. 4. Effect of current density (j) on (a) normalized DOC removal, (b) energy consumption per

736

unit DOC mass, (c) H2O2 concentration and (d) total dissolved polyphenols as a function of time in

737

PEF degradations of the winery wastewater after biological oxidation using a BDD anode, 25 ºC,

738

[Fe2+]0 = 35 mg L-1 and pH = 2.8. j: (
) 10, () 25 and (
) 100 mA cm-2.

739

Fig. 5. Evolution of (a) normalized DOC removal, (b) H2O2 concentration and (c) total dissolved

740

polyphenols as a function of time in the degradation of the winery wastewater after biological

741

oxidation under different EAOPs using a BDD anode, 25 ºC, j = 25 mA cm-2, pH = 2.8 and [Fe2+]0

742

= 35 mg L-1 in EF, PEF and SPEF. EAOP: (
) AO-H2O2, () EF, (
) PEF and () SPEF. The

743

inset panel of Fig. 5a depicts the normalized DOC removal in PEF and SPEF systems as a function

744

of accumulated UV energy per L of solution. (d) SPEF process assessment in terms of ()

745

normalized DOC removal and () normalized COD removal.

746

Fig. 6. Time course of (a) percentage of [LMCA]/DOC ratio and (b) oxalic and (c) malic acids

747

during the (
) AO-H2O2, () EF, (
) PEF and () SPEF processes of Fig. 5.

AC C

EP

TE D

M AN U

SC

RI PT

724

30

ACCEPTED MANUSCRIPT Fig. 7. Biodegradability of samples collected at different times of a SPEF treatment under the

749

conditions of Fig. 5 assessed by respirometry. The inset panel depicts the biodegradable character of

750

a sample according to its bCOD/COD ratio. A.B.O.: Winery wastewater after biological oxidation.

AC C

EP

TE D

M AN U

SC

RI PT

748

31

ACCEPTED MANUSCRIPT Table 1. Physicochemical characterization of the winery wastewater samples (raw, after 10 days of biological oxidation and after 240 min of SPEF process using a BDD anode, [Fe2+]0 = 35 mg L-1, j = 25 mA cm-2, pH = 2.8 and 25 ºC) and discharge limits for WWTPs final effluents according to

RI PT

Portuguese legislation (Decree-Law no. 236/98) and European Directive no. 91/271/CEE. Winery wastewater Raw

Color

Dark violet

Light brown

d

e

n.d. Weak n.d.e 8.3 20 2998 1561 30 480 350 130 (97%) 380 (97%) 150 (98%) 0.4 -g <0.1h 0.21 72 68 59 18 6.8 7.6 3.6 <0.01h <0.008h 19 <0.04h 46 20 <0.004h 5.4 370 413 10 <0.02h

M AN U

d. Strong d.d 3.7 20 3178 545 38 4425 128 4298 12000 7950 0.7 99 <0.1h 0.25 81 70 62 18 6.0 7.7 4.1 0.2 <0.008h 20 <0.04h 30 20 <0.004h 5.2 365 52 8.2 <0.02h

AC C

EP

TE D

Color (diluted 1:20) Odor Odor (diluted 1:20) pH Temperature (ºC) Conductivity (μS cm-1) Alkalinity (mg CaCO3 L-1) Turbidity (NTU) Total dissolved carbon - TDC (mg L-1) Dissolved inorganic carbon - DIC (mg L-1) Dissolved organic carbon - DOC (mg L-1) Chemical oxygen demand - COD (mg O2 L-1) 5-day biochemical oxygen demand - BOD5 (mg O2 L-1) BOD5/COD Biodegradability – Zahn Wellens test (%) Total dissolved iron (mg L-1) Absorbance at 254 nm (AU) (diluted 1:5) Total suspended solids - TSS (mg L-1) Volatile suspended solids - VSS (mg L-1) Total nitrogen (mg L-1) Total dissolved nitrogen (mg L-1) Dissolved organic nitrogen (mg L-1) Ammonium – N-NH4+ (mg L-1) Nitrite – N-NO2- (mg L-1) Nitrate – N-NO3- (mg L-1) Bromide – Br- (mg L-1) Chloride – Cl- (mg L-1) Fluoride – F- (mg L-1) Sulfate – SO42- (mg L-1) Calcium – Ca2+ (mg L-1) Lithium – Li+ (mg L-1) Magnesium – Mg2+ (mg L-1) Potassium – K+ (mg L-1) Sodium – Na+ (mg L-1) Total phosphorous (mg L-1) Phosphate – PO43- (mg L-1) Total dissolved polyphenols (mg caffeic acid equivalent L-1) a

b

After SPEF (Removalb)

SC

Parameter (units)

After biological oxidation (Removala)

41

8.2 (80%) c

ELVc for DecreeLaw no. 236/98 or Directive no. 91/271/CEE

Very light yellow n.d.e Very weak n.d.e 2.8 20 4614 n.d.e 54 18 0.2 18 (86%) 120 (68%) 16 (89%) 0.1 -g 12/<0.13 h, j 0.08 158/55 j 80 27/27 j 18 9.5 8.5 <0.01h <0.01h <0.008h 17 <0.04h 1729 15 <0.004h 4.2 347 396 10/0.35 j <0.02h

n.d. or n.d.e or 6.0 - 9.0 3 ºC increasef or 150 or 125 40 or 25 2.0 or 60 or 35i 15 or 10 7.8 or 11 or 2000 or 10 or 1 -

0.35 (96%)

-

e

d

From raw to after biological oxidation; From after biological oxidation to after SPEF; ELV – Emission Limit Value; d. – detected; n.d. – not detected; f Comparatively to the receptor medium; g – not determined; h Limit of detection value; i Facultative; j Clarified effluent after neutralization.

e

32

ACCEPTED MANUSCRIPT Table 2. Concentration of FeOH2+, HSO4- and Fe(OH)3 (s) in various systems using 46 mg SO42- L-1 or 1.6 g SO42- L-1 and 20, 35 or 70 mg Fe3+ L-1 (ionic strength of 1.84-5.49 ×10-2 M). Data were calculated by the chemical equilibrium modelling system MINEQL+ (Schecher and McAvoy, 2007) using its

RI PT

equilibrium constants and considering the amounts of NH4+, Cl-, Ca2+, Mg2+, K+ and Na+ determined by ion chromatography after biological oxidation (Table 1).

Species concentration at pH 2.8 (mg L-1) FeOH2+

HSO4-

[SO42-] = 46 mg L-1 and [Fe3+] = 20/35/70 mg L-1

4.6/4.6/4.6

4.3/4.3/4.3

15/44/111

[SO42-] = 1.6 g L-1 and [Fe3+] = 20/35/70 mg L-1

1.6/2.9/5.7

148/144/141

0/0/1.7

Fe(OH)3 (s)

AC C

EP

TE D

M AN U

SC

System

33

ACCEPTED MANUSCRIPT Table 3. Pseudo-first-order kinetic constants for DOC removal (kDOC) along with the corresponding coefficient of determination (R2) and residual variance (S2R), obtained for the treatment of the winery wastewater under conditions of Figs. 3, 4 and 5. S2R (mg2 L-2) 9.0 1.9 21.8 1.8 7.9 1.9 -a -a 7.9 10.4

SC

RI PT

R2 0.976 0.997 0.985 0.994 0.984 0.997 -a -a 0.984 0.990

EP

TE D

No fitting of a pseudo-first-order kinetic model to experimental data.

AC C

a

kDOC (×10-3 min-1) 7.0±0.6 11.2±0.2 13.0±1.0 5.0±0.2 8.3±0.5 11.2±0.2 -a -a 8.3±0.5 8.9±0.5

M AN U

System Initial iron 20 concentration 35 2+ -1 (mg Fe L ) 70 Current 10 density 25 -2 (mA cm ) 100 AO-H2O2 Various EF processes PEF SPEF

34

ACCEPTED MANUSCRIPT

M AN U

SC

RI PT

Figure 1

Fig. 1. Research articles on winery wastewater remediation per year arranged by treatment process (source: http://www.scopus.com/, December 2013, search for “winery wastewater treatment” with

AC C

EP

TE D

further manual refinement to exclude misclassified articles and distribute them by treatment process).

ACCEPTED MANUSCRIPT Figure 2

20

2000

10

1000

0

1

2

3

4 5 6 7 time (days)

8

9 10

0

8 6 4 2 0

-1

SC

0

10

pH

3000

12

RI PT

-1

DOC (mg L )

30

Total dissolved polyphenols

40 4000

(mg caffeic acid equivalent L )

14

Fig. 2. Assessment of the biological oxidation efficiency in the treatment of the raw winery

AC C

EP

TE D

M AN U

wastewater in terms of () DOC removal, () total dissolved polyphenols and () pH.

ACCEPTED MANUSCRIPT Figure 3

1.0

RAD-ON

(a)

RAD-ON

(b)

0.6 0.4 0.2

SC

70 60 50 40 30 20 10 0

60

120 180 time (min)

TE D

0

M AN U

-1

[Total Dissolved Iron] (mg L )

0.0

RI PT

DOC/DOC0

0.8

240

Fig. 3. Influence of initial Fe2+ concentration on (a) normalized DOC removal and (b) total dissolved iron concentration as a function of time in PEF degradations of the winery wastewater after biological

AC C

70 mg L-1.

EP

oxidation using a BDD anode, 25 ºC, j = 100 mA cm-2 and pH = 2.8. [Fe2+]0: () 20, () 35 and ()

1.0

(a) ACCEPTED MANUSCRIPT RAD-ON

0.8

DOC/DOC0

Figure 4

0.6 0.4

RAD-ON

400 300

SC

200 100 0 700

RAD-ON

600 500 400 300 200

(c)

TE D

-1

[H2O2] (mg L )

(b)

M AN U

-1

ECDOC (kWh (kg DOC) )

0.0 500

RI PT

0.2

100

(mg caffeic acid equivalent L )

RAD-ON

(d)

10 8

AC C

EP

-1

Total dissolved polyphenols

0

6

4 2 0 0

60

120 time (min)

180

240

Fig. 4. Effect of current density (j) on (a) normalized DOC removal, (b) energy consumption per unit DOC mass, (c) H2O2 concentration and (d) total dissolved polyphenols as a function of time in PEF degradations of the winery wastewater after biological oxidation using a BDD anode, 25 ºC, [Fe2+]0 = 35 mg L-1 and pH = 2.8. j: () 10, () 25 and () 100 mA cm-2.

1.0

(a)

ACCEPTED MANUSCRIPT

0.6 0.4

1.0

RAD-ON

0.8

DOC/DOC0

Figure 5

DOC/DOC0

0.8

RAD-ON

0.6

0.2

0.4 0.2 0.0

0

2

4

6

-1

8

10

QUV (kJ L )

0.0 RAD-ON

(b)

RI PT

-1

[H2O2] (mg L )

400 300 200

SC

100

RAD-ON

(c)

M AN U

10 8 6 4 2

TE D

(mg caffeic acid equivalent L )

-1

Total dissolved polyphenols

0

0

DOC/DOC0; COD/COD0

1.0

RAD-ON

(d)

EP

0.8

0.6

AC C

0.4

0.2 0.0

0

60

120 180 time (min)

240

Fig. 5. Evolution of (a) normalized DOC removal, (b) H2O2 concentration and (c) total dissolved polyphenols as a function of tim e in the degradation of the winery wastewater after biological oxidation under different EAOPs using a BDD anode, 25 ºC, j = 25 mA cm-2, pH = 2.8 and [Fe2+]0 = 35 mg L-1 in EF, PEF and SPEF. EAOP: () AO-H2O2, () EF, () PEF and () SPEF. The inset panel of Fig. 5a depicts the normalized DOC removal in PEF and SPEF systems as a function of accumulated UV energy per L of solution. (d) SPEF process assessment in terms of () normalized DOC removal and () normalized COD removal.

ACCEPTED MANUSCRIPT Figure 6 30 RAD-ON

(a)

RAD-ON

(b)

20 15 10

10 8 6 4 2 0 RAD-ON

(c)

TE D

-1

[Malic acid] (mg C L )

5

M AN U

-1

[Oxalic acid] (mg C L )

0 12

SC

5

RI PT

[LMCA]/DOC (%)

25

4

3

2

EP

1

0

AC C

0

60

120 180 time (min)

240

Fig. 6. Time course of (a) percentage of [LMCA]/DOC ratio and (b) oxalic and (c) malic acids during the () AO-H2O2, () EF, () PEF and () SPEF processes of Fig. 5.

ACCEPTED MANUSCRIPT Figure 7

1.0

0.6

bCOD/COD > 0.3 > 0.1 – 0.3 0.05 – 0.1 < 0.05

Character Very biodegradable Biodegradable Low biodegradable Not biodegradable

RI PT

bCOD/CODkit

0.8

0.4

0.0

A.B.O.

40

90 120 time (min)

SC

0.2

240

M AN U

Fig. 7. Biodegradability of samples collected at different times of a SPEF treatment under the conditions of Fig. 5 assessed by respirometry. The inset panel depicts the biodegradable character of

AC C

EP

TE D

a sample according to its bCOD/COD ratio. A.B.O.: Winery wastewater after biological oxidation.

ACCEPTED MANUSCRIPT Highlights - First study on EAOPs for winery wastewater remediation - Biodegradable fraction of the winery effluent removed by initial biological oxidation - Order of EAOPs efficiency on the recalcitrant fraction reduction: AO-H2O2
RI PT

- Up to 86% DOC, 68% COD and 89% BOD5 removals with 5.1 kWh m-3 cost for SPEF process

AC C

EP

TE D

M AN U

SC

- Color, odor, COD, BOD5, NH4+, NO3- and SO42- in agreement with legislation at the end