This article was downloaded by: [University of Windsor]On: 19 November 2014, At: 08:14Publisher: Taylor & FrancisInforma Ltd Registered in England and Wales Registered Number: 1072954 Registered office: Mortimer House,37-41 Mortimer Street, London W1T 3JH, UK

Environmental TechnologyPublication details, including instructions for authors and subscription information:http://www.tandfonline.com/loi/tent20

Removal and transformation of effluent organic matter(EfOM) in biotreated textile wastewater by GAC/O3 pre-oxidation and enhanced coagulationFeiyue Qian a , Xianbo Sun a , Yongdi Liu a & Hongyong Xu aa School of Resources and Environmental Engineering , East China University of Science andTechnology , Shanghai , ChinaAccepted author version posted online: 14 Dec 2012.Published online: 08 Jan 2013.

To cite this article: Feiyue Qian , Xianbo Sun , Yongdi Liu & Hongyong Xu (2013) Removal and transformation of effluentorganic matter (EfOM) in biotreated textile wastewater by GAC/O3 pre-oxidation and enhanced coagulation, EnvironmentalTechnology, 34:12, 1513-1520, DOI: 10.1080/09593330.2012.758662

To link to this article: http://dx.doi.org/10.1080/09593330.2012.758662

PLEASE SCROLL DOWN FOR ARTICLE

Taylor & Francis makes every effort to ensure the accuracy of all the information (the “Content”) containedin the publications on our platform. However, Taylor & Francis, our agents, and our licensors make norepresentations or warranties whatsoever as to the accuracy, completeness, or suitability for any purpose of theContent. Any opinions and views expressed in this publication are the opinions and views of the authors, andare not the views of or endorsed by Taylor & Francis. The accuracy of the Content should not be relied upon andshould be independently verified with primary sources of information. Taylor and Francis shall not be liable forany losses, actions, claims, proceedings, demands, costs, expenses, damages, and other liabilities whatsoeveror howsoever caused arising directly or indirectly in connection with, in relation to or arising out of the use ofthe Content.

This article may be used for research, teaching, and private study purposes. Any substantial or systematicreproduction, redistribution, reselling, loan, sub-licensing, systematic supply, or distribution in anyform to anyone is expressly forbidden. Terms & Conditions of access and use can be found at http://www.tandfonline.com/page/terms-and-conditions

Environmental Technology, 2013Vol. 34, No. 12, 1513–1520, http://dx.doi.org/10.1080/09593330.2012.758662

Removal and transformation of effluent organic matter (EfOM) in biotreated textile wastewaterby GAC/O3 pre-oxidation and enhanced coagulation

Feiyue Qian, Xianbo Sun, Yongdi Liu∗ and Hongyong Xu

School of Resources and Environmental Engineering, East China University of Science and Technology, Shanghai, China

(Received 30 June 2012; final version received 1 December 2012 )

GAC/O3 (ozonation in the presence of granular activated carbon) combined with enhanced coagulation was employedto process biotreated textile wastewater for possible reuse. The doses of ozone, GAC and coagulant were the variablesstudied for optimization. The effects of different treatment processes on effluent organic matter (EfOM) characteristics,including biodegradability, hydrophobic and hydrophilic nature, and apparent molecular weight (AMW) distribution werealso investigated. Compared with ozonation, GAC/O3 not only presented a higher pre-oxidation efficiency, but also improvedthe treatability of hydrophobic and high molecular weight compounds by enhanced coagulation. After treatment by GAC/O3pre-oxidation (0.6 mg O3·mg−1 COD and 20 g·L−1 GAC) and enhanced coagulation (25 mg·L−1 Al3+ at pH 5.5), the removalefficiencies of chemical oxygen demand (COD), dissolved organic carbon (DOC) and colour were higher than those forcoagulation alone by 17.3%, 12.0% and 25.6%, respectively. Residual organic matter consisted mainly of hydrophobic acidsand hydrophilic compounds of AMW <1 kDa, which were colourless and of limited biological availability. The combinationof GAC/O3 and enhanced coagulation was proved to be a simple and effective treatment strategy for removing EfOM frombiotreated textile wastewater.

Keywords: biotreated textile wastewater; effluent organic matter; ozone; granular activated carbon; enhanced coagulation

1. IntroductionBiologically treated (biotreated) wastewater can be anaccepted source of water for process reuse in arid regionswith a highly concentrated textile industry. To achievethis goal, a necessary part of wastewater reclamation isthe removal of effluent organic matter (EfOM), whichrepresents a variety of recalcitrant organic compoundsranging from low to high molecular weight, including nat-ural organic matter (NOM), synthetic organics and solublemicrobial products (SMP) [1]. Most of the EfOM is inthe soluble fraction, which is more difficult to removethan particulate matter. In particular, residual dyestuffswith additives in textile wastewater cannot be degradedeffectively by conventional biological processes because ofchromophores (such as azo or anthraquinone) and complexpolyaromatic structures, which are specific contaminants inbiotreated effluent and serve as the source of colour andturbidity [2].

Enhanced coagulation (EC) is a common and econom-ical wastewater treatment process, which can achieve notonly the effective removal of suspended solids and colloids,but also a reduction of dissolved organic carbon (DOC) byaround 30–60% [3]. However, previous studies have shownthat EC tends to remove hydrophobic organic matter with ahigh molecular weight, while having only a slight effect on

∗Corresponding author. Email: ydliu@ecust.edu.cn

small molecules [4,5]. Thus, it is difficult for coagulationalone to meet the needs of wastewater reclamation.

Ozonation (O3) has gained attention for its higheffectiveness in achieving biodegradability improvement,decolourization and microbial disinfection [6,7]. However,its use is also limited by selective oxidation of organic mat-ter and the relative high cost of ozone generation. Froman economic point of view, O3 is better to be employed asa pre-oxidation step rather than the main treatment. Thecombination of pre-ozonation and enhanced coagulation(O3 + EC) has been widely used in drinking water treat-ment, where an appropriate dose of ozone is applied notonly as a coagulation aid to destabilize and aggregate smallparticles and colloids, but is also beneficial in reducing theformation of disinfection by-products (DBPs) during chlo-rination [8,9]. Furthermore, it was recently observed that inaddition to its excellent adsorption capacity, activated car-bon also acts as a catalyst in promoting ozonation [10].Depending on the operational conditions and the natureof target pollutants, the removal efficiencies of dissolvedorganic carbon (DOC) or chemical oxygen demand (COD)increased by 10–35% after activated carbon was involved inthe ozonation (AC/O3) of recalcitrant organic compoundssolution [11–13], textile wastewater [14] and pulp milleffluents [15]. Compared with the powdered form, granular

© 2013 Taylor & Francis

Dow

nloa

ded

by [

Uni

vers

ity o

f W

inds

or]

at 0

8:14

19

Nov

embe

r 20

14

1514 F. Qian et al.

activated carbon (GAC) is more suitable for use on an indus-trial scale in continuous AC/O3 system because of its easysolid-liquid separation. To our knowledge, the efficacy ofcombined AC/O3 pre-oxidation and enhanced coagulationhas not been studied in wastewater treatment.

The main objective of this study was to present theresults of the combined application of GAC/O3 pre-oxidation and enhanced coagulation (GAC/O3 + EC) inthe reclamation of typical biotreated textile wastewa-ter. The doses of ozone, GAC and coagulant were thevariables studied for optimization. The effects of differ-ent treatment processes on EfOM characteristics, includ-ing biodegradability, hydrophobic/hydrophilic nature andapparent molecular weight (AMW), were also evalu-ated to elucidate the behaviours of EfOM removal andtransformation.

2. Materials and methods2.1. Biotreated textile wastewaterSamples used in this study were collected daily for fiveconsecutive days from the same effluent point of acti-vated sludge reactors (sludge retention time (SRT) = 15 d;hydraulic residence time (HRT) = 18 h) in a wastewatertreatment plant located in the east of mainland China havinga total flow rate of 10,000 m3·d−1, more than 90% of whichoriginated from textile industries. After settling overnight,the supernatant was refrigerated at 4 ◦C. Water quality infor-mation for the biotreated textile wastewater is provided inTable 1.

2.2. Pre-oxidation and coagulation treatmentPre-oxidation treatment was carried out in a gas–liquid con-tacting column (cross-sectional diameter 60 mm, effectivedepth 200 mm), having an external cooling jacket for tem-perature control at 25 ± 1 ◦C. An approximately completemix flow pattern was achieved by magnetic stirring. Thevolume of the wastewater sample was 500 mL for eachbatch reaction of 2 min. Ozonated air at a flow rate of2.0 L·min−1 (at 101.5 MPa and 20 ◦C) was generated froma CF-G-3-010G ozone generator (GUOLIN Inc., China)and injected through a titanium diffuser (mean pore size

Table 1. Water quality characteristics of biotreated textilewastewater.

Average ±Parameter standard deviation Range

pH 8.2 ± 0.1 8.0–8.3COD (mg·L−1) 67 ± 4 61–75BOD5 (mg·L−1) 8 ± 2 6–10DOC (mg·L−1) 15.73 ± 0.24 15.34–16.12Colour (ADMI units) 317 ± 4 310–325UV254 (cm−1) 0.877 ± 0.020 0.860–0.907Alkalinity (mg·L−1 as CaCO3) 105.3 ± 2.1 102.5–109.2Turbidity(NTU) 48 ± 10 40–61

5 μm) at the bottom of the reactor. O3 concentrations werecontrolled at 5.0 ± 0.2 mg·L−1 and 11.0 ± 0.5 mg·L−1 ofdried air by power adjustment, corresponding to doses ofapplied ozone per initial COD of 0.6 mg O3·mg−1 COD and1.3 mg O3·mg−1 COD, respectively. In the AC/O3 system,coal-based GAC (Shanghai AC Inc., China) with a particlesize of 0.45–0.90 mm was employed at low O3 concentra-tion. According to the manufacturer, the activated carbonhas a BET surface area of 970 m2·g−1, micro/meso/macroporous volumes of 0.35/0.09/0.01 mL·g−1 and a pHpzc of7.4. Gaseous ozone concentrations were determined by theiodometric method [16]. For a comparison, activated car-bon adsorption within the same time was also conducted byinjecting dried air into the reactor instead.

Coagulation experiments were performed using a JJ-4Asix-paddle programmable jar testing apparatus (GUOHUAInc., China). Considering the potential colour impartedby residual iron, aluminium sulphate octadecahydrate(Al2(SO4)3.18H2O; Sinopharm Chemical Reagent Co.,China; analytical grade) was used as coagulant in this study,although the performance of ferric-based coagulants forDOC removal has been found to be better than that ofaluminium-based ones for some given waters, or in coldand low turbidity conditions [3,4]. The stock solution wasprepared at 10 g·L−1 Al3+ and renewed every two weeks.After adding the required dose of Al3+ at the optimal pH5.5 determined in previous studies, a 500 mL sample under-went rapid mixing at 220 rpm (G = 500 s−1) for 1 min,followed by slow mixing at 40 rpm (G = 40 s−1) for 15 min,and was then allowed to settle for 1 h. Temperature was25 ± 2 ◦C during the process. The supernatants were sam-pled for analysis of pH, COD and turbidity directly, and ofDOC, ultraviolet (UV) absorbance at 254 nm wavelength(UV254, cm−1) and colour after being filtered through a0.45 μm membrane (XINYA Inc., China).

2.3. Analytical methodsCOD was measured using a digital reactor DRB200 with aDR2800 spectrometer (HACH Co., USA) following HACHMethod 8000 (Cr). DOC was analysed using a Liquid TOCanalyser (Elementar Co., Germany). Both UV254 and colourwere quantified using the DR5000 UV/VIS spectropho-tometer. Colour was represented by American Dye Man-ufacturers Institute (ADMI) value based on HACH Method10048. The pH and turbidity were measured using PB-10(Sartorius Co., Germany) and Orion AQ2010 (Electron Co.,USA) meters, respectively. Special UV absorbance (SUVA)was calculated as 100 UV254 × DOC−1 (m−1 · mg−1 · L).

The test of biodegradable dissolved organic carbon(BDOC) was based on the procedure described in the lit-erature [17,18]. The pH values of samples after EC wereadjusted to 7.0 ± 0.2 by addition of 1.0 mol·L−1 NaHCO3before being filtered through a 0.45 μm membrane. A100 mL filtered sample with a known initial DOC was inoc-ulated with biotreated textile wastewater (filtered through

Dow

nloa

ded

by [

Uni

vers

ity o

f W

inds

or]

at 0

8:14

19

Nov

embe

r 20

14

Environmental Technology 1515

a 2 μm membrane) at 1% volume ratio, then incubatedunder aeration in the dark at 20 ± 1 ◦C for seven days.The blank group, which was Milli-Q water (Millipore Co.,USA) inoculated at the same ratio, was included. In thisstudy, the blank group DOC during the incubation was0.15 ± 0.10 mg·L−1. The measured BDOC was calculatedas the difference between the DOC reductions of the samplesand blank group during this time.

In this study, EfOM was fractionated into four quasi-homogeneous organic classes: hydrophobic acids (HPOA),non-acid hydrophobic (NAHPO), transphilic (TPI) andhydrophilic (HPI) substances, using Amberlite XAD-8 andXAD-4 resins (Rohm and Haas Co., USA), as outlined byGong et al. [19]. The apparent molecular weight (AMW)cut-offs were performed in an ultra-filtration cup (SINAPCo., China) with a magnetic stirrer at 25 ± 2 ◦C. Cellu-lose membranes were prewashed with Milli-Q water untilthe DOC level of effluents became negligible. Highly purenitrogen provided the applied pressure of 0.20–0.25 MPa fordividing the EFOM into four AMW fractions:>50 kDa, 10–50 kDa, 1–10 kDa and <1 kDa. These divisions are basedon DOC and ADMI analysis of filtrates.

3. Results and discussion3.1. EfOM removal in pre-oxidationOur results relating to EfOM removal in the pre-ozonationof biotreated textile wastewater are summarized in Table 2.The increase in O3 dose caused efficient decompositionof chromophores and unsaturated structures in EfOM withlimited mineralization. The reduction of organic parameterswas in the following order: ADMI>UV254 >COD>DOC.The DOC level of a sample ozonated at 1.3 mg O3·mg−1

COD was 15.4 ± 0.2 mg·L−1, which is higher than theDOC level of 15.0 ± 0.2 mg·L−1 at 0.6 mg O3·mg−1 COD(t = 2.7550; p = 0.0283 <0.05). Meanwhile, the residualturbidity under the former condition is lower. These couldbe attributed to some particulate organic matter (or organiccoatings on particles) resolving into dissolved ones at a highO3 dose [17].

In terms of DOC removal, activated carbon adsorption,especially with 20 g GAC·L−1, was more effective thanpre-ozonation, although the former exhibited a much lower

efficiency of decolourization (Table 2). When GAC/O3 pre-oxidation was used at 0.6 mg O3·mg−1 COD, each 10 g·L−1

increment in GAC caused varying degrees of reduction inDOC, UV254 and ADMI of the samples respectively, inwhich its effect on DOC removal was the most significant(F = 115.03; p < 0.0005).

In this study, a synergistic effect was obvious inGAC/O3, since its DOC reduction was around 1.6 timesthat of the sum of pre-ozonation and GAC adsorption. Inaddition, a break-up of GAC was observed and substantialquantities of fine activated carbon particles left in the aque-ous phase could not be separated via free settling, resultingin a potential increase of COD in effluents. When the quan-tity of GAC was greater than 10 g·L−1, COD removal ofGAC/O3 pre-oxidation appeared to decline.

3.2. Effects of pre-oxidation on coagulationDecolourization and extensive elimination of EfOM aretwo major objectives in the reclamation of biotreated tex-tile wastewater. The coagulation results with and withoutpre-ozonation are shown in Figure 1. In all cases, resid-ual DOC and ADMI declined with increasing coagulantdose, and their values tended to be stable at more than25 mg·L−1 Al3+. Thus, pre-ozonation did not have an appar-ent effect on the point of diminishing returns, which wasdefined as the dose for which the last 10 mg·L−1 Al3+ addi-tion resulted in a less than 0.3 mg·L−1 reduction in DOC [3].However, pre-ozonation at 1.3 mg O3·mg−1 COD was moredetrimental to DOC removal in subsequent EC, especiallywith the low coagulant dose, despite resulting in excel-lent decolourization. Similarly, in drinking water treatment,ozone dosage in pre-ozonation is better controlled belowabout 0.7 mg O3·mg−1 TOC to avoid having an adverseeffect on physical removal of organic matter [20]. Thiscould be attributed to the percentage of hydrophilic andlow molecular weight EfOM, which increased significantlyin O3 at a high ozone dose [21]. Therefore, only the ozonedose of 0.6 mg O3·mg−1 COD was used in GAC/O3.

The same coagulation tendencies were shown afterGAC/O3 pre-oxidation (Figure 2), with initial and finalDOC and ADMI in coagulation becoming lower at higherGAC quantities. This suggests that part of the EfOM

Table 2. Biotreated textile wastewater treated by pre-ozonation, GAC adsorption and GAC/O3 pre-oxidation.

Samples

O3 GAC Turbidity COD DOC UV254 ADMI(mg·mg−1 COD) (g·L−1) removal (%) removal (%) removal (%) removal (%) removal (%)

0.6 — 3.7 ± 1.8 9.4 ± 0.0 4.4 ± 1.1 26.7 ± 2.7 65.0 ± 2.81.3 — 10.8 ± 1.5 14.1 ± 4.2 2.0 ± 1.3 44.4 ± 3.1 80.2 ± 1.3— 10 1.4 ± 0.6 4.5 ± 1.5 6.2 ± 1.2 11.3 ± 3.4 10.4 ± 0.9— 20 1.2 ± 0.6 6.0 ± 2.2 11.8 ± 1.0 16.2 ± 2.1 20.3 ± 0.60.6 10 6.5 ± 1.6 11.3 ± 2.8 16.9 ± 3.0 38.3 ± 2.4 70.2 ± 2.50.6 20 4.1 ± 1.0 −9.9 ± 2.8 27.3 ± 0.0 47.8 ± 1.1 76.8 ± 1.0

Dow

nloa

ded

by [

Uni

vers

ity o

f W

inds

or]

at 0

8:14

19

Nov

embe

r 20

14

1516 F. Qian et al.

Figure 1. Effects of pre-ozonation on DOC (a) and ADMI (b)removal in Al3+ coagulation.

Figure 2. Effects of GAC/O3 pre-oxidation of 0.6 mg O3·mg−1

COD on DOC (a) and ADMI (b) removal in Al3+ coagulation.

removed by GAC/O3 could be resistant to both ozona-tion and coagulation. For a clear comparison, the resultsof enhanced coagulation at 25 mg·L−1 Al3+ with and with-out pre-oxidation are presented in Table 3. By consideringthe O3 dose as the only variable, the F statistic and pvalue of the DOC removal in pre-ozonation combinedwith enhanced coagulation were 10.102 and 0.006, respec-tively. Furthermore, a higher significance of GAC quantityin GAC/O3 + EC(F = 125.343; p < 0.0005) was found.With 20 g·L−1 GAC at 0.6 mg O3·mg−1 COD, GAC/O3 +EC provided comparable decolourization to EC with pre-ozonation at 1.3 mg O3·mg−1 COD, while COD, DOC and

UV254 reductions in the former were greater by 9.1%, 16.7%and 4.3%, respectively. In other words, GAC/O3 was foundto achieve the equivalent treatment targets with a lowerozone dose than O3.

Comparison with the data in Table 2 shows that theimprovements in pre-oxidation performance reduced thecontribution of coagulation to EfOM removal, suggestingthat there were common target pollutants shared by twoprocesses. When pre-ozonation was at 1.3 mg O3·mg−1

COD, subsequent EC removed around 36.2% of DOC,26.6% of UV254 and 13.4% of ADMI in biotreated tex-tile wastewater – values much lower than those for ECalone. Moreover, with the increase of GAC from 0 g L−1

to 20 g L−1 in GAC/O3, the percentages of original DOC,UV254 and ADMI removed via EC declined from 37.0%,36.3% and 24.4% to 27.6%, 27.5% and 16.9%, respectively.In addition, it is worth noting that residual fine activatedcarbon particles from GAC/O3 were entirely eliminated bycoagulation.

Very often, a SUVA value <2 generally indicates a highpredominance of hydrophilic and biodegradable organicmatter, while a SUVA value of ≥4 indicates mainly humicsubstances with large molecules [22]. Since ozone decom-posed the activated aromatic structures preferentially, areduction in the SUVA value of treated samples is quitepronounced, especially at high O3 dose (Table 4). Giventhat the increase of EfOM biological availability may pro-mote microbial regrowth in distribution systems and affectreclaimed water quality [23], it was necessary to evaluatethe amount of BDOC during different treatments.

Biodegradable organic matter in bodies of natural waterhad been considered to include not only low molecu-lar weight and hydrophilic compounds, but also a fewhumic substances, while short-chained aliphatic aldehy-des, ketones and carboxylic acids were common BDOCcomponents produced by ozonation [3,23]. After effectivebiological treatment, the original BDOC concentration wasonly 0.71 ± 0.15 mg·L−1, as seen in Table 4. In this case, ECalone proved to be an effective treatment method for BDOC,since its removal of 83.1% is much higher than 42.9%for DOC. BDOC produced by pre-ozonation was markedlydependent on ozone doses (F = 271.211; p < 0.0005), andits increase resulted in a progressively higher ratio of BDOCremoval to DOC for EC. In contrast to the report by Li et al.[24], GAC/O3 produced less BDOC than O3 in this study(F = 11.613; p = 0.009), probably as a result of the adsorp-tion of oxidation by-products by GAC. Although a lowerratio of BDOC removal to DOC for EC following GAC/O3was observed, the residual BDOC was close to the level ofcoagulated sample with pre-ozonation at 0.6 mg O3·mg−1

COD (F = 4.362; p = 0.068).

3.3. Treatment operating costsBased on the operating parameters obtained from the batchstudy, including the energy consumption and the chemicals

Dow

nloa

ded

by [

Uni

vers

ity o

f W

inds

or]

at 0

8:14

19

Nov

embe

r 20

14

Environmental Technology 1517

Table 3. Biotreated textile wastewater treated by pre-ozonation or GAC/O3 pre-oxidation combined with enhance coagulation.

Samples

O3 GAC Al3+ Turbidity COD DOC UV254 ADMI(mg·mg−1 COD) (g·L−1) (mg·L−1) removal (%) removal (%) removal (%) removal (%) removal (%)

— — 25 91.6 ± 0.2 54.5 ± 3.3 42.9 ± 1.2 48.1 ± 0.6 68.1 ± 0.30.6 — 25 92.7 ± 0.7 58.6 ± 1.1 41.4 ± 2.2 63.0 ± 1.0 89.4 ± 1.11.3 — 25 93.0 ± 1.2 61.9 ± 3.3 38.2 ± 1.2 71.0 ± 2.0 93.6 ± 0.90.6 10 25 91.8 ± 1.2 62.0 ± 1.1 48.4 ± 0.5 69.3 ± 1.0 90.9 ± 0.50.6 20 25 93.7 ± 0.9 71.8 ± 1.0 54.9 ± 0.8 75.3 ± 1.0 93.7 ± 0.5

Table 4. Evolution of SUVA and BDOC of samples in different treatment processes.

Samples

O3 GAC Al3+ SUVA BDOC BDOC removal/DOC removal(mg·mg−1 COD) (g·L−1) (mg·L−1) (m−1·mg−1·L) (mg·L−1) via EC (mg·mg−1)

Raw sample 5.58 ± 0.15 0.71 ± 0.15 0.09— — 25 5.04 ± 0.12 0.12 ± 0.020.6 — — 4.17 ± 0.14 2.24 ± 0.07 0.220.6 — 25 3.51 ± 0.08 0.95 ± 0.131.3 — — 3.15 ± 0.10 3.18 ± 0.16 0.251.3 — 25 2.60 ± 0.14 1.79 ± 0.130.6 10 — 4.18 ± 0.09 1.89 ± 0.22 0.150.6 10 25 3.39 ± 0.04 1.14 ± 0.070.6 20 — 4.04 ± 0.09 1.58 ± 0.18 0.100.6 20 25 3.16 ± 0.01 1.16 ± 0.07

dose, the economic feasibility of GAC/O3 + EC could beassessed for wastewater reclamation.

The energy requirement for O3 production from airwas around 20 kWh·kg−1 O3 with the energy cost ofEUR 0.12 kWh−1 and the average current price of acti-vated carbon was EUR 0.60 kg−1 in Shanghai, China.The estimated cost for coagulant (Al2O3 > 15.6%) andhydrochloric acid (HCl > 36%) was taken to be EUR0.14 kg−1 and EUR 0.09 kg−1 respectively. Treatment oper-ating costs in EUR·m−3 are listed in Table 5. After GAC/O3pre-oxidation, activated carbon, with the loss ratio less than1% for each batch, was separated from aqueous phase andrecycled in the system. The increase amount of chemicalsludge from fine activated carbon particles was negligi-ble, while the sludge generation (water content >99%) wasaround 66.7 kg·m−3 in EC.

The cost of GAC/O3 + EC could be reasonable takinginto account its high treatment efficacy, although coagula-tion alone turned out to be the cheapest method.

3.4. Characterization of EfOM in different treatmentprocesses

EfOM fractionations, on the basis of hydrophobic/hydrophilicnature and apparent molecular weight distribution, providedinsight into their removal and transformation behaviours inpre-oxidation and enhanced coagulation.

3.4.1. Effects on EfOM hydrophobic/hydrophilicfractions

The hydrophobic/hydrophilic fractions of EfOM inbiotreated textile wastewater (raw sample) are shown inFigure 3. The DOC composition consisted of HPOA38.9 ± 0.5%, NAHPO 21.4 ± 2.6%, TPI 16.0 ± 1.7%, HPI23.7 ± 1.6%, while HPOA (42.6 ± 2.5%) and NAHPO(46.8 ± 3.8%) fractions predominated in total ADMI.

With limited mineralization and efficient decolouriza-tion, pre-ozonation changed the EfOM characteristics to bemore hydrophilic (Figure 3(a)), in which the HPI percentagein DOC increased to 34.3%. In contrast, the most sensi-tive fraction to ozone was NAHPO, while HPOA was morerecalcitrant. After ozonation at 0.6 mg O3·mg−1 COD, DOCand ADMI in NAHPO decreased by 57.6% and 72.5%,respectively, compared with their removal levels in HPOAof 8.3% and 57.4%, respectively. It appeared that the pres-ence of GAC enhanced EfOM removal of ozonation, inwhich the residual DOC concentrations of all organic frac-tions were lower than those in pre-ozonation. In terms ofdecolourization, GAC/O3 helped in reducing the ADMI val-ues of hydrophobic fractions, which were the main colouredorganic matter in raw sample (Figure 3(b)).

As shown in Figure 3(a), it was always more diffi-cult to remove hydrophilic than hydrophobic DOC via EC,whether pre-oxidation was set up or not. For EC alone, therewas only a 15.1% of DOC reduction from HPI, while the

Dow

nloa

ded

by [

Uni

vers

ity o

f W

inds

or]

at 0

8:14

19

Nov

embe

r 20

14

1518 F. Qian et al.

Table 5. Operating cost calculations of different treatment processes.

Chemicals dose

Treatment Coagulant HCl GAC Required energy Total costsprocessa (kg·m−3) (kg·m−3) (kg·m−3) (kWh·m−3) (EUR·m−3)

(1) 0.30 0.15 — — 0.06(2) 0.30 0.15 — 0.80 0.16(3) 0.30 0.15 — 1.76 0.27(4) 0.30 0.15 <0.20 0.80 0.28

a(1) EC: 25 mg·L−1 Al3+ (pH 5.5); (2) O3 + EC: 0.6 mg O3·mg−1 COD, 25 mg·L−1 Al3+ (pH 5.5);(3) O3 + EC: 1.3 mg O3·mg−1 COD, 25 mg·L−1 Al3+ (pH 5.5); (4) GAC/O3 + EC: 0.6 mg O3·mg−1

COD, 20 g·L−1 GAC, 25 mg·L−1 Al3+ (pH 5.5).

Figure 3. Hydrophobic/hydrophilic fractions of DOC (a) andADMI (b) in different treatment processes: (1) Raw sample;(2) EC: 25 mg·L−1 Al3+; (3) O3: 0.6 mg O3·mg−1 COD; (4)O3 + EC: 0.6 mg O3·mg−1 COD, 25 mg·L−1 Al3+; (5) GAC/O3:0.6 mg O3·mg−1 COD, 20 g·L−1 GAC; (6) GAC/O3 + EC:0.6 mg O3·mg−1 COD, 20 g·L−1 GAC, 25 mg·L−1 Al3+.

major removed fractions were HPOA (45.0%) and NAHPO(31.1%). With a pre-ozonation, the removed DOC of HPIbecame higher than that of NAHPO on account of the EfOMtransformation. In comparison, although the treatability ofhydrophobic DOC was improved by GAC/O3, HPI and TPIfractions became more difficult to remove by EC. This isconsistent with the low removal efficiency of BDOC inthe same process, since hydrophilic materials dominate theBDOC produced by O3 [25].

After treatment by GAC/O3 + EC, the percentages ofhydrophobic/hydrophilic fractions in DOC were in the fol-lowing order: HPI > HPOA > TPI > NAHPO, while theresidual colour was mainly contributed by HPOA.

3.4.2. Effects on EfOM apparent molecular weightdistribution

The molecular weight distributions of EfOM in biotreatedtextile wastewater are shown in Figure 4. Our results

Figure 4. Apparent molecular weight distributions of DOC (a)and ADMI (b) in different treatment processes: (1) Raw sam-ple; (2) EC: 25 mg·L−1 Al3+; (3) O3: 0.6 mg O3·mg−1 COD; (4)O3 + EC: 0.6 mg O3·mg−1 COD, 25 mg·L−1 Al3+; (5) GAC/O3:0.6 mg O3·mg−1 COD, 20 g·L−1 GAC; (6) GAC/O3 + EC:0.6 mg O3·mg−1 COD, 20 g·L−1 GAC, 25 mg·L−1 Al3+.

deviated slightly from the summary of Barker et al. [26]concerning molecular weight distributions in biotreatedeffluents, which are considered in general to follow bimodalpatterns, even though influents exhibit a skewed non-normaldistribution with a predominance of the very low molecularweight fraction. The results from raw sample showed thatintermediate compounds of 1–10 kDa (34.1 ± 2.0%) rep-resented a higher DOC percentage than the lower AMWfraction (16.9 ± 1.9%), while water colour was highlyconcentrated in the AMW >50 kDa (41.3 ± 1.3%) and 10–50 kDa (38.1 ± 1.3%) fractions, and the ADMI percentageof AMW <1 kDa was only 4.4 ± 0.9%. Thus, there couldhave been some small molecules that were bound to largeones (such as humic-like substances) or aggregated mutu-ally by hydrogen bonding, non-polar force and polyvalentcation interactions [27], considering the relative low molec-ular weight (usually <1 kDa) and high tinctorial strength ofthe dyestuff.

Dow

nloa

ded

by [

Uni

vers

ity o

f W

inds

or]

at 0

8:14

19

Nov

embe

r 20

14

Environmental Technology 1519

Pre-ozonation decomposed large molecules into smallerones, since the DOC in AMW of 1–10 kDa and <1 kDaclearly increased (Figure 4(a)). Meanwhile, the effectivedecolourization was observed in all the fractions except forAMW <1 kDa. GAC/O3 was found to be more efficientthan pre-ozonation at reducing DOC of AMW <10 kDaand ADMI of all AMW fractions, which was beneficial inimproving the quality of reclaimed water.

As shown in Figure 4, EC alone was more effective atremoving EfOM with AMW >10 kDa than smaller ones.However, pre-ozonation remarkably improved the treatabil-ity of intermediate molecules of 1–10 kDa, which becamethe majority of DOC removed in the coagulation process.This is in accordance with some reports on NOM treatment[28,29], since some low molecular weight organic matterwith hydrophilic and weakly hydrophobic nature producedby pre-ozonation could complex with Al3+ species to beremoved.

In addition, the DOC percentage of AMW <1 kDa aftertreatment by GAC/O3 + EC reached around 63.9%, whichis higher than 58.8% of EC following pre-ozonation. Itwas obvious that a pre-oxidation step enhanced the break-ing of complexation and aggregation between low AMWcompounds and other EfOM in EC caused by the chargeneutralization and dramatic change of pH.

4. ConclusionsPre-ozonation at a low dose (0.6 mg O3·mg−1 COD) wascapable of improving wastewater reclamation efficacy interms of reducing ADMI and UV254, and limiting BDOCformation and the adverse effects of overdosing on subse-quent coagulation. Moreover, GAC/O3 pre-oxidation waseasy to implement on the basis of the present ozonation sys-tem, which gave a higher performance for decolouring andremoving DOC, and improved the coagulation treatabilityof hydrophobic and high AMW compounds. After treat-ment by GAC/O3 + EC, residual organic matter consistedmainly of hydrophobic acids and hydrophilic substances ofAMW<1 kDa. In addition, treatment operating cost of EUR0.28 m−3 would be reasonable taking into account its hightreatment efficacy, while the increase amount of chemicalsludge from fine activated carbon particles was negligible.In brief, the combination of GAC/O3 and enhanced coagu-lation was a simple and effective treatment strategy for thereclamation of biotreated textile wastewater.

AcknowledgementsThis study was funded by National High Technology Researchand Development Program (863) of China (Grant NO.2009AA063904).

References[1] H.K. Shon, S. Vigneswaran, and S.A. Snyder, Effluent

organic matter (EfOM) in wastewater: Constituents, effects

and treatment, Crit. Rev. Environ. Sci. Technol. 36 (2006),pp. 327–374.

[2] O.J. Hao, H. Kim, and P.C. Chiang, Decolorization ofwastewater, Crit. Rev. Environ. Sci. Technol. 30 (2000),pp. 449–505.

[3] C. Volk, K. Bell, E. Ibrahim, D. Verges, G. Amy, and M.Lechevallier, Impact of enhanced and optimized coagula-tion on removal of organic matter and its biodegradablefraction in drinking water, Water Res. 34 (2000), pp. 3247–3257.

[4] A. Matilainen, N. Tuhkanen, and T. Tuhkanen, Compari-son of the effiency of aluminium and ferric sulphate in theremoval of natural organic matter during drinking watertreatment process, Environ. Technol. 26 (2005), pp. 867–876.

[5] M.Q. Yan, D.S. Wang, J.R. Ni, J.H. Qu, Christopher W.K.Chow, and H.L. Liu, Mechanism of natural organic matterremoval by polyaluminum chloride: Effect of coagulationparticle size and hydrolysis kinetics, Water Res. 42 (2008),pp. 3361–3370.

[6] O. Karahan, H. Dulkadiroglu, I. Kabdasli, S. Sozen, F. Ger-mirli Babuna, and D. Orhon, Effect of ozonation on thebiological treatability of a textile mill effluent, Environ.Technol. 23 (2002), pp. 1325–1336.

[7] R. Trequer, R. Tatin, A. Couvert, D. Wolbert, and A. Tazi-Pain, Ozonation effect on natural organic matter adsorptionand biodegradation – application to a membrane bioreactorcontaining activated carbon for drinking water production,Water Res. 44 (2010), pp.781–788.

[8] M.Q. Yan, D.S. Wang, B.Y. Shi, Q.S. Wei, J.H. Qu, and H.X.Tang, Transformations of particles, metal elements and nat-ural organic matter in different water treatment processes,J. Environ. Sci. 19 (2007), pp. 271–277.

[9] Y.S. Pei, J.W. Yu, Z.H. Guo, Y. Zhang, M. Yang, J.S. Zhang,and H. Junji, Pilot study on pre-ozonation enhanced drinkingwater treatment process, Ozone Sci. Eng. 29 (2007), pp. 317–323.

[10] B. Kasprzyk-Hordern, M. Ziolek, and J. Nawrocki, Cat-alytic ozonation and methods of enhancing molecular ozonereactions in water treatment, Appl. Catal., B 46 (2003),pp. 639–669.

[11] J. Rivera-Utrilla and M. Sanchez-Polo, Ozonation of 1,3,6-naphthalenetrisulphonic acid catalysed by activated carbonin aqueous phase, Appl. Catal., B 39 (2002), pp. 319–329.

[12] F.J. Beltran, P. Pocostales, P. Alvarez, and A. Oropesa,Diclofenac removal from water with ozone and activatedcarbon, J. Hazard. Mater. 163 (2009), pp. 768–776.

[13] L. Gu, X.W. Zhang, and L.C. Lei, Degradation of aqueous p-nitrophenol by ozonation integrated with activated carbon,Ind. Eng. Chem. Res. 47 (2008), pp. 6809–6815.

[14] P.C.C. Faria, J.J.M. Orfao, and M.F.R. Pereira, Miner-alisation of coloured aqueous solutions by ozonation inthe presence of activated carbon, Water Res. 39 (2005),pp. 1461–1470.

[15] C.H. Ko, P.H. Hsieh, M.W. Chang, J.M. Chern, S.M. Chi-ang, and C.J. Tzeng, Kinetics of pulp mill effluent treatmentby ozone-based processes, J. Hazard. Mater. 168 (2009),pp. 875–881.

[16] K. Rakness, G. Gordon, B. Langlais, W. Masschelein,N. Matsumoto, Y. Richard, and C. Michael, Guideline formeasurement of ozone concentration in the process gas froman ozone generator, Ozone Sci. Eng. 18 (1996), pp. 209–229.

[17] H. Woo, W.N. Kim, E. Shoto, and M. Okada, Pilot plantstudy on ozonation and biological activated carbon processfor drinking water treatment, Water Sci. Technol. 35 (1997),pp. 21–28.

Dow

nloa

ded

by [

Uni

vers

ity o

f W

inds

or]

at 0

8:14

19

Nov

embe

r 20

14

1520 F. Qian et al.

[18] E. Khan, S. King, R.W. Babcock, and M.K. Stenstrom, Fac-tors influencing biodegradable dissolved organic carbonmeasurement, J. Environ. Eng. 125 (1999), pp. 514–521.

[19] J.L. Gong, Y.D. Liu, and X.B. Sun, O3 and UV/O3 oxidationof organic constituents of biotreated municipal wastewater,Water Res. 42 (2008), pp. 1238–1244.

[20] J.K. Edzwald, Coagulation in drinking water treatment:Particles, organics and coagulants, Water Sci. Technol. 27(1993), pp. 21–25.

[21] P. Bose and D.A. Reckhow, The effect of ozonation on naturalorganic matter removal by alum coagulation, Water Res. 41(2007), pp. 1516–1524.

[22] J.K. Edzwald and J.E. Toblason, Enhanced coagulation: USrequirements and a broader view, Water Sci. Technol. 40(1999), pp. 63–70.

[23] J. Swietlik, A. Dabrowska, U. Raczyk-Stanislawiak, andJ. Nawrocki, Reactivity of natural organic matter fractionswith chlorine dioxide and ozone, Water Res. 38 (2004),pp. 547–558.

[24] L.S. Li, W.P. Zhu, P.Y. Zhang, Z.L. Zhang, H.H. Wu,and W.Y. Han, Comparison of AC/O3-BAC and O3-BAC

processes for removing organic pollutants in secondaryeffluent, Chemosphere 62 (2006), pp. 1514–1522.

[25] T. Ratpukdi, S. Siripattanakul, and E. Khan, Mineralizationand biodegradability enhancement of natural organic matterby ozone-VUV in comparison with ozone, VUV, ozone-UV,and UV: Effects of pH and ozone dose, Water Res. 44 (2010),pp. 3531–3543.

[26] D.J. Barker and D.C. Stuckey, A review of soluble microbialproducts (SMP) in wastewater treatment systems, Water Res.33(1999), pp. 3063–3082.

[27] J.A. Leenheer and J.P. Croue, Characterizing aquatic dis-solved organic matter, Environ. Sci. Technol. 37 (2003),pp. 18A–26A.

[28] J.A. Davis and R. Gloor, Adsorption of dissolved organ-ics in lake water by aluminium oxide: Effect of molecu-lar weight, Environ. Sci. Technol. 15 (1981), pp. 1223–1229.

[29] B. Brian, A.B. Gudrun, D. David, E. Rob, F. Fritz, H. Sebas-tian, K. Simon, and T. Melissa, Experimental evaluation ofcationic matter in conventional treatment with alum, WaterSci. Technol. 40 (1999), pp. 71–79.

Dow

nloa

ded

by [

Uni

vers

ity o

f W

inds

or]

at 0

8:14

19

Nov

embe

r 20

14