Chemosphere 117 (2014) 193–199

Contents lists available at ScienceDirect

Chemosphere

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

Technical Note

Evaluation of Fenton method and ozone-based processes for colour andorganic matter removal from biologically pre-treated swine manure

http://dx.doi.org/10.1016/j.chemosphere.2014.06.0470045-6535/� 2014 Elsevier Ltd. All rights reserved.

⇑ Corresponding author. Tel.: +34 983317188.E-mail addresses: bertariano@yahoo.es (B. Riaño), gargonmi@itacyl.es

(M.C. García-González).

Berta Riaño a,⇑, Mónica Coca b, Mari Cruz García-González a

a Agricultural Technological Institute of Castilla y Léon, Ctra. Burgos, km. 119, 47071 Valladolid, Spainb Department of Chemical Engineering and Environmental Technology, University of Valladolid, C/Doctor Mergelina s/n, 47011 Valladolid, Spain

h i g h l i g h t s

� Advanced oxidation processes are used to treat biologically pre-treated swine manure.� The highest efficiencies (78% for TCOD and 96% for colour) are obtained with Fenton.� Coagulation are elucidated as an important removal mechanism in Fenton method.� Direct reactions with ozone are the main responsible of removals in ozonation.� Fenton is the most cost-effective technology for the post-treatment.

a r t i c l e i n f o

Article history:Received 28 March 2014Received in revised form 10 June 2014Accepted 11 June 2014

Handling Editor: O. Hao

Keywords:Biologically pre-treatedSwine manureFentonOzoneHydrogen peroxide

a b s t r a c t

This work evaluates the efficiency of different advanced oxidation processes (Fenton method, O3, H2O2

and O3/H2O2) for removing total COD (TCOD) and colour from biologically pre-treated swine manure.The Fenton process with a dosage of 100 mg L�1 of Fe2+ and 800 mg L�1 of H2O2 resulted in about 78%TCOD and 96% colour reductions at an initial pH = 3 after a reaction time of 30 min. Coagulation, ratherthan oxidation process, was identified as a crucial mechanism for removing pollutants. Otherwise, singleozonation achieved only 27–30% TCOD and 53–88% colour removals for ozone dosages ranging between0.7 and 4.3 g O3 h�1 at the original wastewater pH (pH = 8.1) after 30 min reaction time. The combinedtreatment with O3/H2O2 at pH = 8.1 did not produce any significant TCOD or colour reduction improve-ment. Therefore, direct reactions with ozone rather than radical reactions were elucidated as the mainremoval mechanisms in the ozone-based processes. Finally, a rough estimation of the operational costsinvolved in each process was also performed to compare their economic feasibility. The findings sug-gested that the Fenton process was more suitable than ozonation for reducing TCOD and colour fromthe biologically pre-treated swine manure.

� 2014 Elsevier Ltd. All rights reserved.

1. Introduction

Among the alternatives to improve manure management, bio-logical processes are often seen as a solution due to their implica-tions in nitrogen, carbon and phosphorous cycles (Bernet andBéline, 2009). After a multistage biological treatment, most of theorganic and nutrient load is removed (Vanotti et al., 2009; Riañoand García-González, 2014). However, livestock wastewater pre-sents a dark brown colour that remains after complete secondarytreatment, which can be attributed to the presence of humic-typesubstances and carbon-rich stable residues (Tanaka, 2002; Hong

et al., 2005). Therefore, further challenges should be faced up tomeet the requirements of discharge imposed by the legislationand to develop innovative technologies for reusing and/or recy-cling livestock wastewater to minimize its environmental impact.In particular, new treatment strategies should aim to removenon-biodegradable organic compounds and colour after biologicalprocesses.

Advanced oxidation processes (AOP) involving Fenton’s reagent,ozone and hydrogen peroxide in single or combined action, havebeen reported to be appropriate for the treatment of relativelylow strength industrial wastewater containing non-biodegradableorganic pollutants and dark colour after a biological treatment(Coca et al., 2005; Lee and Shoda, 2008; Yetilmezsoy and Sakar,2008). Indeed, the application of AOPs to industrial effluents mightbe only economically attractive in combination with other

194 B. Riaño et al. / Chemosphere 117 (2014) 193–199

wastewater treatment technologies, particularly considering theyusually imply a high energetic demand, or an excessive reagentconsumption (Merayo et al., 2013).

Fenton process can have the dual functions of oxidation andcoagulation and, therefore, this technology is capable to removealmost all parts of the organics which consists of both solubleand particulate fractions of COD formed during the biological treat-ment (Aydin and Sarikaya, 2012). In recent years, various studieshave been addressed for the application of Fenton method on theremoval of organic matter and colour from different industrialwastewater such as anaerobically treated leachate (Lau et al.,2001), anaerobically pre-treated poultry manure wastewater(Yetilmezsoy and Sakar, 2008) and livestock wastewater pre-treated by means of a biological activated sludge process (Honget al., 2005).

Chemical oxidation with ozone has also been reported to showan efficient reduction of colour and organic matter of differenttypes of wastewaters. However, wastewater composition shows agreat influence on AOP efficiency (Coca et al., 2005; Merayoet al., 2013). To the best of our knowledge, studies on biologicallypre-treated livestock wastewaters using AOPs are rather limited tofew investigations. The study of the mechanisms involved inorganic matter and colour removal using real wastewater willimprove the application of combined biological systems/AOPs.Moreover, information regarding costs is also required to success-fully introduce new treatment alternatives (Oller et al., 2011;Nolan et al., 2012).

The aim of the present work was to evaluate the effectiveness ofFenton method and ozone-based processes as post-treatments forbiologically pre-treated swine manure. The performance of eachmethod was evaluated in terms of organic matter and colourremoval. The molecular weight distribution of colorants in waste-water before and after each treatment was also studied. Finally, abasic economic assessment of these technologies was carried out.

2. Materials and methods

2.1. Biologically pre-treated swine manure

Wastewater came from an on-farm swine manure treatmentplant located in Segovia (Spain) that consists of a solid–liquid sep-aration phase using screw pressing followed by a coagulation-floc-culation process, and biological nitrification–denitrification of theliquid fraction (Riaño and García-González, 2014). Wastewaterwas collected after the biological treatment, homogenized bymechanical agitation and stored at 4 �C for further use. Character-istics of the wastewater used in the present work are summarizedin Table 1.

2.2. Fenton method

A stock solution of 10 g L�1 of Fe2+ was prepared by dissolvingFeSO4�7H2O (Panreac) in 0.2 N H2SO4. In addition to iron sulphatereagent, 30% H2O2 solution (Panreac) was also used. Experimentalruns were performed in 1 L glass beakers with 0.5 L of sample.Firstly, the pH of the sample was adjusted to 3.0 (Yetilmezsoyand Sakar, 2008) by the addition of 1 N H2SO4. The FeSO4�7H2Oand H2O2 solutions were then added to the wastewater that was

Table 1Physicochemical characteristics of biologically pre-treated swine manure.

pH Coloura (AU) TCOD (mg L�1)

8.1 ± 0.1 0.8 ± 0.2 769 ± 127

a AU: absorbance units measured at 475 nm.

rapidly stirred for 5 min at 90 rpm using a Jar Test Equipment(Fisher Bioblock Scientific, Flocculater 11197). The wastewaterwas then stirred at 10 rpm for 25 min. After that, the stirring wasceased and the flocs formed were allowed to settle for 1 h. About50 mL of the supernatant sample was then collected for totalCOD (TCOD) and colour determination. To evaluate Fenton process,different experimental runs were carried out. First, the initial pH,the H2O2 dose and the reaction time were kept constant whilethe Fe2+ dose was varied between 100 and 400 mg L�1. Then, theinitial pH, the Fe2+ dose and the reaction time were kept constantwhile the H2O2 dose was varied between 200 and 1200 mg L�1. Allexperimental runs were carried out in duplicate and results areexpressed as means.

2.3. Oxidation processes with ozone and hydrogen peroxide

Ozonation experiments were conducted in a 0.75 L glass reactorconnected to an ozone generator (Model Sander 301) that pro-duced ozone from oxygen (60 L h�1) by electrical discharge. Thereactor was filled with 0.4 L of wastewater and then agitated witha magnetic stirrer at 250 rpm. The ozone-containing gas was con-tinuously supplied for 30 min through a gas diffuser at the bottomof the reactor at different rates (0.7–4.3 g O3 h�1). Two ozone trapscontaining 2% potassium iodide solution were connected in serieswith the reactor to determine ozone concentration in the outletgas stream. Samples aliquots (15 mL) were taken from the reactorat regular intervals for TCOD and colour determination.

Single hydrogen peroxide oxidation experiments were carriedout in batch mode by adding different concentrations (200–1200 mg L�1) of 30% H2O2 (Panreac) to 0.5 L of wastewater. AfterH2O2 addition, samples were agitated for 5 min at 90 rpm and thenstirred at 10 rpm for 25 min using a Jar Test Equipment (Fisher Bio-block Scientific, Flocculater 11197). After the oxidation process,about 50 mL of sample was collected for TCOD and colourdetermination.

In the experimental runs combining ozone with hydrogen per-oxide, the H2O2 was initially added to the wastewater to set initialconcentrations ranging from 200 to 800 mg L�1 and, immediatelyafter, ozone at a flow rate of 0.7 g O3 h�1 was continuously fedup for 30 min.

All these experimental runs were performed in duplicate, at theoriginal pH of the wastewater (pH = 8.1) and at room temperature.Results are expressed as means.

2.4. Analytical methods

TCOD and alkalinity were performed in accordance with APHAStandard Methods (2005). pH was measured in a Crison Basic 20pH-meter (Crison Instruments, SA, Barcelona, Spain). Total Fe wasanalysed using an atomic absorption spectrometer (AA 240 FS, Var-ian). Colour content of samples was obtained with a SpectronicHelios Gamma UV–Visible Spectrophotometer (Thermo Fisher Sci-entific, USA). The visible light absorption at 475 nm that character-izes brown colour was measured with 1 cm pathlength cells (Peñaet al., 2003). TOC was measured with a Shimadzu TOC-V CSH TotalOrganic Carbon Analyser. The ozone concentration in the feed gasstream and outlet gas stream was determined by the iodometricmethod (APHA, 2005).

TOC (mg L�1) Alkalinity (mg L�1) Fe (mg L�1)

235 ± 55 3197 ± 37 <2.0

B. Riaño et al. / Chemosphere 117 (2014) 193–199 195

The molecular weight distribution of colorants in the wastewa-ter was determined using an Ultrahydrogel 250 column (exclusionlimit 1–80 kDa). Deionized water was used as mobile phase at aflow rate of 0.7 mL min�1 at 35 �C. A Waters 996 Photodiode ArrayDetector and a Waters 2414 Refractive index were used asdetectors. Dextrans were used as standards for the estimation ofmolecular weights.

2.5. Statistical analysis

Data were compared using ANOVA with 95% confidence level(p < 0.05) to determine whether there were significant differencesin removal efficiencies obtained from the different experimentalruns. Statistical analysis was carried out using StatgraphicsCenturion XVI.

3. Results and discussion

3.1. Fenton process

3.1.1. Effect of Fe2+ dosageThe addition of Fe2+ concentrations higher than 100 mg L�1 for a

fixed dosage of 200 mg H2O2 L�1 at initial pH of 3.0 reduced thecolour removal efficiencies (Fig. 1a). The same effect was observedwith the TCOD removal, obtaining the highest removal efficiency atFe2+ dosage of 100 mg L�1 (75%). Likewise, TOC removal efficien-cies at Fe2+ dosages of 100 and 400 mg L�1 ranged between 60%and 57%, respectively (data not shown). This result can beexplained by the capture of hydroxyl radicals by the reaction withFe (II) in excess. In addition, the Fe (III) formed can react with H2O2

to generate Fe (II) and hydroperoxyl radicals (HO2. ) in solution, con-

suming H2O2 (Benitez et al., 2001; Meric et al., 2004; Lee and

0

10

20

30

40

50

60

70

80

90

100

0 100 200 300 400 500

Rem

oval

eff

icie

ncy

(%)

Fe2+ dose (mg L-1)

H2O2 = 200 mg L-1

Reaction time = 30 minpH = 3.0

Fe2+ = 100 mg L-1

Reaction time = 30 minpH = 3.0

0

10

20

30

40

50

60

70

80

90

100

0 200 400 600 800 1000 1200 1400

Rem

oval

eff

icie

ncy

(%)

H2O2 dose (mg L-1)

(a)

(b)

Fig. 1. Removal efficiency of TCOD (j), and colour (s) from the biologically pre-treated swine manure after Fenton process: (a) influence of Fe2+ dosage (Reactiontime = 30 min, H2O2 dose = 200 mg L�1, initial pH of 3.0) and (b) influence of H2O2

dosage (Reaction time = 30 min; initial pH of 3.0; Fe2+ = 100 mg L�1).

Shoda, 2008). It is worth mentioning that experimental runs add-ing separately 100 mg Fe2+ L�1 and 200 mg H2O2 L�1 to the waste-water at pH = 3.0 were also performed (data not shown), butneither colour or TCOD removals were detected under these condi-tions. In fact, it was observed that the coagulation of particles inthe pre-treated swine manure began after the addition of H2O2 ifFe (II) was previously added, whereas formation of flocs was notobserved only with the addition of Fe (II). Therefore, TCOD andcolour reductions can be attributed to the dual mechanisms ofcoagulation and chemical oxidation caused by the combined actionof Fe (II) and H2O2.

3.1.2. Effect of H2O2 dosageIt is important to optimize the amount of H2O2 in the Fenton

method because excessive doses of H2O2 cause side effects (Duttaet al., 2001; Lee and Shoda, 2008). Colour removal efficiencyslightly increased from 83% to 96% when H2O2 dosage increasedfrom 200 to 800 mg L�1 for a fixed dosage of 100 mg Fe2+ L�1

(Fig. 1b). Dosages higher than 800 mg L�1 did not result in a signif-icant (p = 0.79) increase of removal efficiencies. Moreover, nosignificant (p = 0.10) differences were found for TCOD removalfor H2O2 concentrations higher than 800 mg L�1. Indeed, a slightdecrease in TCOD removals was detected when increasing H2O2

concentrations. This slight decrease might be explained by the factthat residual H2O2 could affect TCOD reduction, as H2O2 acts as ascavenger of the highly potent hydroxyl radical to produce thehydroperoxyl radical, which has much lower oxidation potentialthan the hydroxyl radical (Lee and Shoda, 2008). TOC removal effi-ciencies varied between 58% and 63% regardless the H2O2 dosage(data not shown).

The highest removals corresponded to a H2O2/Fe2+ molar ratioof 4. Fenton process removed 736 mg L�1 of TCOD from the biolog-ically pre-treated swine manure with the dosages of 100 mg Fe2+

L�1 and 800 mg H2O2 L�1 at initial pH of 3.0 for a reaction time of30 min. Therefore, to remove 1 g of TCOD in the wastewater,0.14 g of Fe2+ and 0.33 g of H2O2 (pure) were consumed. Underthese conditions, the H2O2 dose (mg L�1)/TCOD concentration(mg L�1) ratio was 0.90. This calculated ratio was similar to that(1.05) reported by Lee and Shoda (2008) and may be used as a cri-terion to determine the optimum H2O2 dose needed according tothe initial TCOD concentration in the wastewater. Besides, itshould be stressed that optimum dosages determined in the pres-ent study were considerably lower than those reported in previousworks. Thus, Hong et al. (2005) found that the optimum dosages ofFe2+ and H2O2 in Fenton oxidation of livestock wastewater were4000 and 1000 mg L�1, respectively. With these dosages, COD val-ues decreased from 895 mg L�1 to 76 mg L�1. Lee and Shoda (2008)applied Fenton method for high-strength livestock wastewatertreatment and obtained colour removal reductions lower than60% with dosages of 6554 mg Fe2+ L�1 and 4000 mg H2O2 L�1.Yetilmezsoy and Sakar (2008) investigated the treatment of anaer-obic poultry manure wastewater with Fenton process and foundthe highest values of COD and colour removals (95% and 96%,respectively) at Fe2+ and H2O2 dosages of 100 and 1200 mg L�1,respectively. The reduction in the use of reagents in Fenton methodwould result in lowering operational costs. Nevertheless, differ-ences in performances could be due to the characteristics of waste-waters, initial pH values and, also, reaction times (Yetilmezsoy andSakar, 2008).

3.2. Oxidation processes with ozone and hydrogen peroxide

3.2.1. Effect of ozone dosage on colour and organic matter removalefficiencies

Single ozonation results show that colour removal efficienciesincreased from 59% to 89% after a reaction time of 30 min when

0102030405060708090

100

0 10 20 30 40

Col

our r

emov

al (%

)

0

10

20

30

40

50

0 10 20 30 40

TC

OD

rem

oval

(%)

Reaction time (min)

(a)

(b)

Fig. 2. Effect of ozone dosage (0.7 g h�1 (�), 1.4 g h�1 (h), 2.6 g h�1 (N) and 4.3 g h�1

(�) on (a) colour and (b) TCOD removal efficiency from the biologically pre-treatedswine manure at the original wastewater pH.

Table 2Average ozone uptake and ozone consumed per TCOD removed ratios in the differentozone-based processes after 30 min reaction time.

Experiment O3

dosage(g h�1)

H2O2

dosage(mg L�1)

O3

uptake(%)

Ozoneconsumed/TCOD removed(g O3 g�1 TCOD)

Single ozonationoriginal pH (pH = 8.1)

0.7 0 30 1.51.4 0 26 2.62.6 0 22 4.24.3 0 15 12.8

Ozonation with H2O2

addition original pH(pH = 8.1)

0.7 200 26 1.50.7 400 31 1.90.7 800 31 2.2

196 B. Riaño et al. / Chemosphere 117 (2014) 193–199

ozone flow rate increased from 0.7 to 4.3 g O3 h�1 (Fig. 2a). Ozoneflow rate also exerted a positive effect in colour removal rates,wastewater decolourization occurring mainly at the initial stageof the reaction (20 min). Colour in pre-treated livestock wastewa-ter is mainly due to the presence of humic-type substances(Tanaka, 2002; Hong et al., 2005). These compounds have an aro-matic structure. As ozone is able to react with compounds withhigh electron density groups, direct reactions may be responsiblefor colour removal reactions. Likewise, the initial pH of the biolog-ically pre-treated swine manure was about 8.1 (Table 1). This valueslightly increased to 8.2–8.5 throughout ozonation. Under basicpH, hydroxide ions catalyse the decomposition of ozone to formhighly oxidising and non-selective hydroxyl radicals which havean oxidation potential higher than that for ozone (2.80 V). How-ever, the biologically pre-treated swine manure contained a highconcentration of bicarbonate ions (3200 mg L�1 as CaCO3) whichare strong inhibitors of indirect reactions between hydroxyl radi-cals and organic matter (Glaze and Kang, 1987). If the inorganiccarbon acts as a scavenger for hydroxyl radicals, the reaction rateof organic compounds with hydroxyl radicals is often negligibleand the direct oxidation reactions with the organic substratesbecome predominant.

Chemical oxidation with ozone provided as much as 27–30%TCOD removal after 30 min reaction time regardless the dosageapplied, considerably lower than colour removal percentages(Fig. 2b). TOC variations were also independent of the appliedozone flow rate (7–10%; data not shown). Accordingly, colourreduction may be due to direct reactions between ozone and chro-mophore functional groups responsible for pre-treated swine man-ure colour. Ozone transformed the structure of colouredcompounds rather than completely oxidize them (Rehman et al.,2012).

Ozone uptake, defined as the percentage of ozone fed into thereactor that is consumed after 30 min, decreased from 30% to

15% when ozone flow rate increased from 0.7 to 4.3 g h�1 (Table 2)that corresponded to a ratio of 0.25 and 0.77 g O3 consumed L�1 ofwastewater, respectively. The ratio ozone consumed g�1 TCODremoved was also calculated (Table 2). An increase in the appliedozone dosage from 0.7 to 4.3 g h�1 yielded a sharply increase inthe ratio from 1.5 to 12.8 g O3 g�1 TCOD. For that reason, low ozonedosages (0.7 g h�1) are enough to reduce TCOD from the biologi-cally pre-treated swine manure. Although higher ozone dosagesincreased colour removal efficiencies, TCOD removal remainedconstant and most ozone left the reactor without being consumedby organic compounds.

3.2.2. Effect of the combination of ozone and hydrogen peroxideHydrogen peroxide has an oxidant potential (1.77 V) lower than

ozone (2.07 V). In aqueous solutions, hydrogen peroxide dissoci-ates to give perhydroxyl ion and hydroxyl radicals. The perhydr-oxyl ion has a strong nucleophilic activity and can react withchromophore groups reducing colour (Coca et al., 2005). However,in the experimental runs carried out with hydrogen peroxidealone, neither changes in colour nor TCOD values were observedregardless of the concentration added (data not shown). Therefore,oxidation with hydrogen peroxide alone was not effective indegrading colourants in the biologically pre-treated swine manure.This could be due to the high alkalinity of wastewater that acts asscavenger of the hydroxyl radicals formed. Moreover, it has alsobeen pointed out that, under alkaline conditions, hydrogen perox-ide can be decomposed to form molecular oxygen. Therefore, lowconcentrations of perhydroxyl ions and hydroxyl radicals may beavailable in the reaction mixture under these experimentalconditions.

The combination of ozone with H2O2 at the original waste-water pH did not result in a considerably (p > 0.05) increasein the removal efficiency of colour and organic matter incomparison to those obtained with ozone (Fig. 3). Even, itwas observed a slight TCOD decrease from 28% to 24% whenincreasing H2O2 dosage. This could indicate that TCOD reduc-tion was mainly due to direct reactions with ozone rather thanradicals reactions. The addition of hydrogen promoted ozonedecomposition into hydroxyl radicals, so lower ozone moleculeswere available for direct oxidation reactions. As it was previ-ously indicated, radical scavengers in wastewater (bicarbonateions) compete with coloured organics for hydroxyl radicalsformed from the decomposition of molecular ozone. The ratioozone consumed/TCOD removed slightly increased from 1.5 to2.2 g O3 g�1 TCOD when increasing the H2O2 dosage, whichcan be explained by the decomposition of ozone catalysed bythe hydrogen peroxide (Table 2). It was also noticed that TOCvalues remained virtually constant (4–5% TOC removed; datanot shown).

0

10

20

30

40

50

60

70

80

90

100

0 200 400 600 800 1000

Rem

oval

effic

ienc

y (%

)

Initial pH = 8.1

Fig. 3. Influence of H2O2 dosage on the removal efficiency of TCOD (j), and colour(s) from the biologically pre-treated swine manure treated with O3/H2O2 after30 min of reaction time at an initial pH of 8.1.

Table 3Effectiveness of the different AOPs in removing colour and organic matter.

Fenton method Ozonation

Reagent dosages 100 mg Fe2+ L�1,800 mg H2O2 L�1

0.7 g O3 h�1

Initial pH 3.0 8.1Reaction time (min) 30 (1 h settling) 30Oxygen-equivalent chemical-

oxidation capacity (OCC)0.51 0.88

Colour removed (AU g�1 OCC added) 1.78 0.40TCOD removed (g TCOD g�1 OCC) 1.44 0.20

AU

0.000

0.001

0.002

0.003

0.004

0.005

0.006

0.007

0.008

0.009

T2.00 4.00 6.00 8.00 1

AU

0.000

0.001

0.002

0.003

0.004

0.005

0.006

0.007

0.008

0.009

T2.00 4.00 6.00 8.00 1

O3 + H2O2

O3

(a)

>80

kDa

(b)

Fig. 4. Molecular weight distributions of (a) the biologically pre-treated swine manure anpH 8.1 (200 and 800 mg H2O2 L�1) and (b) the biologically pre-treated swine manure at pat different H2O2 dosages (1) 200 mg H2O2 L�1 and (2) 800 mg H2O2 L�1. Chromatogram

B. Riaño et al. / Chemosphere 117 (2014) 193–199 197

3.3. Comparison of performance of AOPs

3.3.1. Oxygen-equivalent chemical-oxidation capacityThe parameter oxygen-equivalent chemical-oxidation capacity

has been used to quantify in arbitrary units the oxidants addedto the wastewater, in order to compare the performance of the dif-ferent AOPs evaluated (Cañizares et al., 2009). This parameter wasobtained from stoichiometric calculations, taking into account thenumber of electrons exchanged in the reduction of hydrogen per-oxide and ozone, resulting that 1 kg O3 m�3 is equivalent to1 kg O2 m�3 and 0.471 kg H2O2 m�3 is equivalent to 1 kg O2 m�3.According to this parameter, Fenton method seems to be the mostefficient process for removing colour and TCOD under the optimumconditions determined in the present study (Table 3). It must bementioned that the maximum value of this ratio should be 1. Thehigher value obtained in Fenton process can be explained in termsof non-oxidative side reactions, mainly coagulation processescaused by the action of iron ions added to the solution and/or sed-imentation processes (Cañizares et al., 2009). Therefore, these highefficiencies can be attributed to these mentioned mechanismsrather than oxidation processes.

3.3.2. Molecular weight distributions of colorantsThe effectiveness of the oxidation processes was also analyzed

from the molecular weight distribution chromatograms corre-sponding to the colorants in wastewater. Colorants in raw waste-water were divided into three groups with different molecularweights corresponding to their peak apices: > 80, 9.6 and 4.9 kDa(Fig. 4a). As a result of the oxidation processes, a considerablereduction in absorbance at 475 nm was observed, that is relatedwith a decrease in the chromophore group concentrations. The

ime (min)0.00 12.00 14.00 16.00 18.00 20.00

3.9

kDa

(1)

(2)

ime (min)0.00 12.00 14.00 16.00 18.00 20.00

4.9

kDa9.

6 kD

a

2kD

a

d the effluents after single ozonation (0.7 g h�1) and combined O3/H2O2 processes atH 3.0 and the effluents after Fenton process (100 mg Fe2+ L�1, 30 min reaction time)s run at 475 nm.

Table 4Operational costs associated to the two studied technologies.

Consumption Unitarycost

Operatingcost

Fenton methodH2SO4 (1 N)a 0.07 m3 m�3 8.77 € m�3 0.61 € m�3

Fe2+ 0.1 kg m�3 0.33 € kg�1(c) 0.03 € m�3

H2O2 0.8 kg m�3 0.23 € kg�1(c) 0.18 € m�3

Sludge treatment and disposal 1 kg m�3 0.50 € kg�1(d) 0.50 € m�3

Total treatment cost 1.32 € m�3

OzonationO3 productionb 0.88 kg m�3 2.56 € kg�1 2.25 € m�3

Total treatment cost 2.25 € m�3

a The amount of 1 N H2SO4 needed to change the pH to the desired value (3.0 inthe present study) was experimentally determined. A H2SO4 price of 0.179 € kg�1

for H2SO4 95–97% (w/w) was considered (Cortez et al., 2011).b Power (W) was measured directly on the lab-scale set up (16 kWh m�3). Esti-

mation was carried out considering an electricity price in Spain of 0.14 € kWh�1.c From Cañizares et al. (2009).d From Anfruns et al. (2013).

198 B. Riaño et al. / Chemosphere 117 (2014) 193–199

9.6 kDa peak completely disappeared, while the reduction ofthe > 80 kDa colorant slightly increased when the combined O3/H2O2 process was used. The chromatograms do not show breakingof the colorants into molecules of lower molecular weight.

The molecular weight distribution of the biologically pre-trea-ted swine manure at pH 3.0 was compared with those after Fentonprocess varying hydrogen peroxide dosage. The acidification ofwastewater led to changes in the distribution of colorants(Fig. 4b). Peaks were eluted at retention times longer than thoseof the raw wastewater. The peaks corresponded to molecularweights of 3.9 and 2 kDa. Therefore, the acidification process forreducing pH resulted in the polymerization of colorants but theabsorbance at 475 nm remained constant. Chromatograms alsoconfirmed that the reduction of colorants increased when H2O2

dose rose from 200 to 800 mg L�1.

3.3.3. Operational costsAn evaluation of the two studied technologies as post-treat-

ment for the removal of colour and organic matter from the biolog-ically pre-treated swine manure was carried out based on theoperational costs associated to these processes (i.e. reagentsrequired, energy costs and sludge production). It has to be empha-sized that the cost calculations were only approximate as a rigor-ous economic analysis should consider, among others, the initialinvestment, prices at plant scale and maintenance and labourcosts. The results show that the Fenton process offers the lowestoperational costs for the post-treatment of biologically pre-treatedswine manure (Table 4). In addition to lower operational costs, thesimplicity of operation and design of Fenton treatment are veryattractive compared with other advanced oxidation processes(Cortez et al., 2011). Also, it must be emphasized that ozonationalways requires significantly higher initial investment than theFenton process (Cortez et al., 2011). However, one of the main dis-advantages of Fenton method is that the sludge produced after thesedimentation is high in iron, and, therefore, this leads to the prob-lem of disposing the sludge (Hsueh et al., 2005; Yetilmezsoy andSakar, 2008; Anfruns et al., 2013). Although ozonation does notproduce sludge, the low TCOD removal achieved using this processlimited considerably the application of this technology as post-treatment for the biologically pre-treated swine manure.

4. Conclusions

Fenton process performed successfully in removing colour andorganic matter from the biologically pre-treated swine manure,with colour and TCOD removal efficiencies up to 96% and 78%,

respectively. Coagulation was elucidated as an important removalmechanism in Fenton method. For the ozone-based processes,maximum colour removal efficiency of 89% for a dosage of4.3 g O3 h�1 after 30 min reaction time was obtained, whereasTCOD removals were lower than 30% regardless the dosageapplied. The addition of hydrogen peroxide to promote the forma-tion of hydroxyl radicals did improve neither colour nor TCODremovals due to the presence of bicarbonate ions that competewith organic coloured compounds for hydroxyl radicals. As aresult, direct reactions between ozone and organic compound werethe main mechanism for colour and TCOD reduction in this type ofwastewater.

To conclude, the application of Fenton method to biologicallypre-treated swine manure would enhance the reusing and/or recy-cling of swine manure, allowing the reduction of the environmen-tal impact of intensive pig farming. Hence, further researchesshould be addressed to reduce operational costs that allowpresenting the combined biological process/Fenton method as acost-competitive technology.

Acknowledgements

This research has been funded by the European Union under theProject Life+ MANEV (LIFE9-ENV-ES-0453) ‘‘Evaluation of manuremanagement and treatment technologies for environmentalprotection and sustainable livestock farming in Europe’’. JesúsRodríguez is gratefully acknowledged for his analytical support.

References

Anfruns, A., Gabarró, J., Gonzalez-Olmos, R., Puig, S., Balaguer, M.D., Colprim, J.,2013. Coupling anammox and advanced oxidation – based technologies formature landfill leachate treatment. J. Hazard. Mater. 258–259, 27–34.

Aydin, A.F., Sarikaya, H.Z., 2012. Fenton’s oxidation for advanced treatment of highstrength opium alkaloid industry effluents treated with biological processes.ITU Dergisi 1.

Benitez, F.J., Acero, J.L., Real, F.J., Rubio, F.J., Leal, A.I., 2001. The role of hydroxylradicals for the decomposition of p-hydroxy phenylacetic acid in aqueoussolutions. Water Res. 35, 1338–1343.

Bernet, N., Béline, F., 2009. Challenges and innovations on biological treatment oflivestock effluents. Bioresour. Technol. 100, 5431–5436.

Cañizares, P., Paz, R., Sáez, C., Rodrigo, M.A., 2009. Cost of the electrochemicaloxidation of wastewaters: a comparison with ozonation and Fenton oxidationprocesses. J. Environ. Manage. 90, 410–420.

Coca, M., Peña, M., González, G., 2005. Chemical oxidation processes fordecolorization of brown colored molasses wastewater. Ozone Sci. Eng. 27,365–369.

Cortez, S., Teixeira, P., Oliveira, R., Mota, M., 2011. Evaluation of Fenton and ozone-based advanced oxidation processes as mature landfill leachate pre-treatments.J. Environ. Manage. 92, 749–755.

Dutta, K., Mukhopadhyay, S., Bhattacharjee, S., Chaudhuri, B., 2001. Chemicaloxidation of methylene blue using a Fenton-like reaction. J. Hazard. Mater. B 84,57–71.

Glaze, W.L., Kang, J.W., 1987. The chemistry of water treatment processes involvingozone, hydrogen peroxide and ultraviolet radiation. Ozone Sci. Eng.,335–352.

Hong, S.W., Choi, Y.S., Kwon, G., Park, K.Y., 2005. Performance evaluation ofphysicochemical processes for biologically pre-treated livestock wastewater.Water Sci. Technol. 52, 107–115.

Hsueh, C.L., Huang, Y.H., Wang, C.C., Chen, C.Y., 2005. Degradation of azo dyes usinglow iron concentration of Fenton and Fenton-like systems. Chemosphere 58,1409–1414.

Lau, I.W.C., Wang, P., Fang, H.H.P., 2001. Organic removal of anaerobically treatedleachate by Fenton coagulation. J. Environ. Eng. 127, 666–669.

Lee, H., Shoda, M., 2008. Removal of COD and color from livestock wastewater bythe Fenton method. J. Hazard. Mater. 153, 1314–1319.

Merayo, N., Hermosilla, D., Blanco, L., Cortijo, L., Blanco, A., 2013. Assessing theapplication of advanced oxidation processes, and their combination withbiological treatment, to effluents from pulp and paper industry. J. Hazard.Mater. 262, 420–427.

Meric, S., Kaptan, D., Ölmez, T., 2004. Color and COD removal from wastewatercontaining Reactive Black 5 using Fenton‘s oxidation process. Chemosphere 53,435–441.

Nolan, T., Troy, S.H., Gilkinson, S., Frost, P., Xie, S., Zhan, X., Harrington, C., Healy,M.G., Lawlor, P.G., 2012. Economic analyses of pig manure treatment options inIreland. Bioresour. Technol. 105, 15–23.

B. Riaño et al. / Chemosphere 117 (2014) 193–199 199

Oller, I., Malato, S., Sánchez-Pérez, J.A., 2011. Combination of advanced oxidationprocesses and biological treatments for wastewater decontamination – areview. Sci. Total. Environ. 409, 4141–4166.

Peña, M., Coca, M., González, G., Rioja, R., García, M.T., 2003. Chemical oxidation ofwastewater from molasses fermentation with ozone. Chemosphere 51, 893–900.

Rehman, M.S.U., Khaid, A., Liaquat, S., Mehmood, C.T., Mahmood, T., Mehmood, S.,2012. Color and COD removal from poultry leachate using an ozonation process.Environ. Eng. Manag. J. II 8, 1467–1474.

Riaño, B., García-González, M.C., 2014. On-farm treatment of swine manure basedon solid-liquid separation and biological nitrification-denitrification of theliquid fraction. J. Environ. Manage. 132, 87–93.

Standard Methods for the Examination of Water and Wastewater, APHA, 2005.American Water Works Association and Water Environment Federation, 21sted. American Public Health Association, Washington, DC.

Tanaka, Y., 2002. A dual purpose packed-bed reactor for biogas scrubbing and methane-dependent water quality improvement applying to a wastewater treatment systemconsisting of UASB reactor and trickling filter. Bioresour. Technol. 84, 21–28.

Vanotti, M.B., Szogi, A.A., Millner, P.D., Loughrin, J.H., 2009. Development of asecond-generation environmentally superior technology for treatment of swinemanure in the USA. Bioresour. Technol. 100, 5406–5416.

Yetilmezsoy, K., Sakar, S., 2008. Improvement of COD and color removal from UASBtreated poultry manure wastewater using Fentons oxidation. J. Hazard. Mater.151, 547–558.