Journal of Environmental Chemical Engineering 2 (2014) 2111–2119

Combined chemical coagulation–flocculation/ultraviolet photolysistreatment for anionic surfactants in laundry wastewater

E.L. Terechova a,b, Guoquan Zhang a,*, Jie Chen a, N.A. Sosnina c, Fenglin Yang a

aKey Laboratory of Industrial Ecology and Environmental Engineering (Ministry of Education, China), School of Environmental Science and Technology, DalianUniversity of Technology, 2 Linggong Road, Dalian 116024, ChinabHydraulics and Water Supply Department, Far Eastern State Transport University, Serishev St., 47, Khabarovsk 680021, RussiacChemistry and Ecology Department, Far Eastern State Transport University, Serishev St., 47, Khabarovsk 680021, Russia

A R T I C L E I N F O

Article history:Received 24 May 2014Received in revised form 31 July 2014Accepted 10 September 2014

Keywords:Anionic surfactantsCoagulation–flocculationUltraviolet photolysisResponse surface methodologyLaundry wastewater

A B S T R A C T

In this work, a combined chemical coagulation–flocculation/ultraviolet photolysis process was used toseparate and oxidative degrade the linar alkylbenzene sulfonate (LAS), an anionic surfactant in laundrywastewater, aiming at making the effluent dischargeable with suitable characteristics. Mineral ash, ZnCl2,and Praestol-650 (P-650) were chosen as the coagulant-sorbent, the complex former and the cationichigh-molecular flocculants, respectively. The dosages of three components were optimized through theresponse surface methodology (RSM). The optimum parameters values obtained from RSM were furtherproved by a successful parallel trial with the actual laundry wastewater. Results showed that themaximum LAS removal efficiency of 71.26% and 74.58% were achieved for the self-made LAS wastewaterand the actual laundry wastewater when the dosages of ZnCl2, ash and P-650 was 29.54, 1936.35 and196.38 mg/L, respectively. The effect of solution pH in LAS ultraviolet photolysis process was alsoinvestigated. Results indicated that the alkaline medium is beneficial to LAS photolysis removal. Theseresults support the applicability of the combined chemical coagulation–flocculation/ultravioletphotolysis process for LAS removal due to its efficient and rapids treatment rate, high adsorption andextraction capacity, and acceptable catalytic oxidation ability using Zn2+ salts and mineral ash as specificcoagulant and Praestol-650 as cationic high-molecular flocculant.

ã 2014 Elsevier Ltd. All rights reserved.

Contents lists available at ScienceDirect

Journal of Environmental Chemical Engineering

journal homepage: www.else vie r .com/ locat e/ jece

Introduction

Surfactants used as surface-active matters can effectivelydecrease the surface tension of liquids. Those surface activitiesof surfactants derive from the amphiphilic structures that possesboth hydrophilic and hydrophobic parts in one molecule [1].Synthetic surfactants are widely used in many industrialapplications such as metal processing, textile, food, pharma-ceuticals and paper industries [2]. Surfactants are can beclassified into four groups depending on the charge of thehydrophilic part: nonionic (0), anionic (�), cationic (+) andzwitterionic (�) [3].

Linar alkylbenzene sulfonate (LAS) is a typical anionic surfac-tant, which are extensively used in household products, detergents,personal care products, industrial processes and pesticide formu-lations [4,5]. It can also be found in the sewages of many enterprises

* Corresponding author. Tel.: +86 411 84706172; fax: +86 411 84706328.E-mail address: guoquanz@126.com (G. Zhang).

http://dx.doi.org/10.1016/j.jece.2014.09.0112213-3437/ã 2014 Elsevier Ltd. All rights reserved.

including laundries, car washing facilities and railway transportfacilities [6]. LAS were referred as water pollutants of the third groupof dispersibility of Kylsky’s classification of water impurities(diameter of particles from 1 to 10 nm) [6]. It is reported that indomestic sewage, the LAS concentration may vary from 1 to 18 mg/L[7], and the concentration in laundry wastewater may vary from17 to 1024 mg/L [8]. Under low concentration they exist in water inthe form of molecules and ions, forming the homogeneous systems.Under high concentration and in the presence of particulate, fine-dispersed pollutants and oil products, however, they form colloidstructures and act simultaneously as the stabilizers of emulsions andsuspensions.

Due to the biotoxicity and non-biodegradability, wastewaterscontaining surfactants need to be treated before discharging intothe aquatic environment, in terms of public and environmentalhealth [2]. Generally, LAS are also considered as the dangerous andundesirable substances in water body. Even accumulated in smallamounts (0.8–2.0 mg/L), LAS would produce a strong toxic effecton flora and fauna, destroy the organoleptic property and preventthe self-purification process of water body [9]. On the other hand, a

Table 1Chemical components of the mineral ash.

Component SiO2 Al2O3 Fe2O3 MgO CaO SO3 TiO2 MnO K2O Na2O

Content (wt%) 50.64 37.57 5.15 0.80 1.72 0.30 1.15 0.068 0.67 0.44

Table 2The initial composition of the laundry wastewater.

Laundry wastewater ingredients Concentration (mg/L)

LAS 19.68Suspension substances/optical density 126.6/0.76Oil products (oil P) 3.70Chemical oxygen demand (COD) 280pH 9.0–9.5

2112 E.L. Terechova et al. / Journal of Environmental Chemical Engineering 2 (2014) 2111–2119

study of the possibility for wastewater reuse is essential because ofits large quantities in the laundering process of industriallaundries. Laundry wastewater possesses the potential forreclamation and reuse. Such reclamation and reuse of laundrydischarge is important to save water supply and significantlyimprove urban environments [10]. In this context, many technol-ogies such as adsorption, ion exchange, membrane filtration,precipitation, coagulation, flocculation, and oxidation treatmentshave been proposed to treat LAS in laundry wastewater [9–21].Among the currently used techniques, precipitation/coagulationhad received considerable attention due to its high removalefficiency and low-cost. A great number of liquid and solid organic/inorganic coagulants and flocculants are available [11,16–18]. Inaddition, the advanced oxidation processes (AOPs) such asozonation, photocatalysis, Fenton and ultraviolet irradiation[11,15,19–21] had been widely utilized to improve the removalefficiency of LAS. However, it is difficult to meet with the nationalenvironmental quality standard of China (GB 3838-88) aftertreated by the conventional methods. Also, it is difficult todevelop a single and an effective treatment method due to thediversity and the unique physicochemical properties of the laundrywastewater.

In present work, a combined chemical coagulation–floccula-tion/photolysis process was proposed. Mineral ash, ZnCl2 andPraestol-650 (P-650) were chosen as coagulant-sorbent, complexprecursor and cationic high molecular flocculants, respectively.The dosages of the three components were optimized through theresponse surface methodology (RSM). The novel mechanism ofcoagulation–flocculation process was also proposed. In thephotolytic process, the influence of the solution pH wasinvestigated and the products after photolysis were also analyzedby gas chromatography. Results indicated that the combinedchemical coagulation–flocculation/ultraviolet photolysis process isan environmentally friendly strategy for laundry wastewatertreatment.

Material and methods

Chemicals and wastewater samples

In RSM experiments, the simulated anionic surfacant waste-water was self-made using LAS (alkyl benzene sulfonates,R-C6H4SO3H (322 g/mol), where R refers to the alkyl chain(hydrophobic region) ranging from 4 to 10 carbon atoms andanother part (hydrophilic region) corresponding to a sulfonatedaromatic ring). Owing to the high concentration of LAS, the laundrywastewater was generally diluted to obtain an influent LASconcentration of approximately 12.0–14.0 mg/L, below the inhibi-tory value of 50 mg/L in biochemical treatment processes [22]. So,the initial concentration of LAS in this work was selected as13.8 mg/L, which is at concentration below its critical micelleconcentration (CMC) of 2.0 mM [23]. Mineral ash obtained fromBeihai thermal power plant, Dalian, was used as the coagulant-sorbent and its composition was listed in Table 1. ZnCl2and Praestol-650 were chosen as complex precursor and cationichigh molecular flocculants, respectively. Solution pH was adjustedby using sulfuric acid and sodium hydroxide.

Experimental procedures

Coagulation–flocculation experimentsCoagulation–flocculation experiments were carried out in a

classical jar test apparatus. ZnCl2, mineral ash and P-650 wereadded successively into 400 mL wastewater. Then, the wastewaterwas vigorously mixed for 5 min at 200 rpm, followed by slowmixing for 30 min at 30 rpm, and allowed to settle for 30 min.Finally, the supernatant was taken out for measurement andsubsequent ultraviolet photolysis. After the optimization of threecomponents dosages through RSM, three parallel coagulation–flocculation/ultraviolet photolysis experiments were carried outwith the actual laundry wastewater obtained from the launderettein Dalian University of Technology, China. The initial compositionof the laundry wastewater is shown in Table 2.

Response surface methodologyThe dosages of ZnCl2, ash and Praestol-650 were optimized by

the RSM in order to obtain the maximal removal efficiency of AS. ABox–Behnken design [21] was chosen to evaluate the combinedeffect of three independent variables. The contents of ZnCl2, ashand Praestol-650 were termed as X1,X2 and X3, respectively. Theminimum and maximum ranges of variables were investigated andthe full experimental plan with respect to their values is listed inTable 2. The coded values of the three independent variablestogether with the responses are shown in Table 3. An empiricalsecond-order polynomial model for three factors was in thefollowing form:

Yi ¼ b0 þXn

i¼1

biXi þXn

i¼1

biiX2i þ

Xn

i<1

bijXiXj þ e (1)

where Y is residual contents of LAS (mg/L); Xi the variableparameters in codes, bo, bi, bii and bij the parameters of theregression model; e the random error associated with this measure[24]. Design-Expert software was used to attain the coefficientparameters estimated by the multiple linear regression analysis,generating response surface contour plots and analyzing the datacollected by the performing analysis of variance (ANOVA).

Photolysis experimentsUltraviolet photolysis experiments were carried out using a high

pressure mercury lamp (quartz tube, power supply 220 V/36 W,frequency = 50 Hz, l = 253.7 nm) as the light source at roomtemperature. The distance between the lamp and wastewaterwas 5 cm. The volume of wastewater was 200 mL, the width of thewastewater layer was 3 cm and the irradiation time was 30 min.

Table 3Level and code of variables for RSM.

Variables Coded Levels

�1 0 1

ZnCl2 (mg/L) X1 4 17 30Ash (mg/L) X2 0 1000 2000Praestol-650 (mg/L) X3 30 115 200

Table 4The design of RSM and its actual and predicted values.

Number X1 X2 X3 R (%)

Experimental Predicated

1 17 1000 115 48.53 48.532 4 2000 115 36.94 36.963 30 2000 115 60.75 60.464 17 0 200 65.57 65.245 30 1000 30 44.65 44.616 17 2000 200 62.97 62.917 30 0 115 55.26 55.248 17 2000 30 34.87 35.29 4 1000 30 23.02 22.67

10 4 1000 200 56.34 56.3811 17 1000 115 48.53 48.5312 30 1000 200 69.6 69.9513 17 1000 115 48.53 48.5314 17 1000 115 48.53 48.5315 4 0 115 42.93 43.2216 17 1000 115 48.53 48.5317 17 0 30 33.86 33.92

Table 5Analysis of variance table.

Source Sum ofsquares

df Meansquare

F value P-value Prob > Fsignificant

Model 2429.97 9 270.00 2951.81 <0.0001X1-ZnCl2 630.66 1 630.66 6894.86 <0.0001X2-ash 0.55 1 0.55 5.97 0.0445

E.L. Terechova et al. / Journal of Environmental Chemical Engineering 2 (2014) 2111–2119 2113

Analytical determinations

The determination of LAS was conducted by the liquidfluorescent spectrophotometry method (RF 5301 PC, Shimadzu).The excitation wavelength is 261 nm and the emission wavelengthis 315 nm. Firstly, LAS ions with the fluorescent stain (acridineyellow) were extracted from sample using chloroform, and thenthe concentration of LAS was measured according to thefluorescence intensity of the extract. The zeta potential value ofthe complex aggregate was measured by MALVERN ZETASIZER(3000HS, UK). The suspension substances in actual laundrywastewater was measured by the gravimetric method, wherethe wastewater sample firstly passed through 0.45 mm filtermembrane and then dried to constant weight at 103–105 �C. Theoptical density and COD of the actual laundry wastewater weredetermined by LS117 (Linshang Technology Ltd., China) andmicrowave digestion method. The oil products in wastewaterwas also measured by the gravimetric method, where the watersample was firstly acidized by sulfuric acid, then it was extractedby petroleum ether, and finally weighting the extractants after thepetroleum ether was evaporated. an p=-meter HANNA 213 wasused to determinate the solution p=. The presence of ketones(acetones), alcohols, fatty acid (E2–E10) and (E8–E22) wasdetermined by the gas chromatography (GC, HP 5890). Thedetector was the flame ionization detector (FID), and the carriergas was nitrogen. The substances concentrations were determinedaccording to the chromatographic peak height as compared withthe standard peak.

Results and discussion

Optimization of coagulation–flocculation experiments

LAS are present as both micelle and ion-molecular states inactual wastewater [2]. Taking into consideration the fact that thecolloid particles of LAS is negatively charged, thus Praestol-650 which contains the positively charged ions (Fig. 1), was chosenas the cationic flocculants due to the occurrence of ion-ioninteraction. The partial hydrolyzations of the coagulant compo-nents can efficient adsorb the charged particles either in colloidstate or dissolved in water environment. Therefore, the industrialmineral ash from the thermoelectrical plant was used as a solidcoagulant. On the other hand, in order to reduce the flocculants

Fig. 1. The molecule structure of Praestol-650.

consumption and improve the treatment quality, it is necessaryintroducing a coagulant to destruct the stable dispersed systems.Based on the classical colloidal chemistry theory, it is alsonecessary to convert LAS from the ion-molecular state intoinsoluble complexes by using the heavy metal compounds. Thus,ZnCl2 was chosen as the complex precursor, since the stableinsoluble complexes with LAS would generate when Zn(II) wasadded.

RSM was used to optimize the addition of ZnCl2, ash andPraestol-650 in coagulation–flocculation process and the removalefficiency of LAS was shown in Table 4. The coefficients of theresponse function for the dependent variable were determined bycorrelating the experimental data, which was expressed by thesecond-order polynomial equation:

Y ¼ 16:51 þ 0:73X1 � 4:15 � 10�3X2 þ 0:21X3 þ 2:21� 10�4X1X2 � 1:89 � 10�3X1X3 � 1:06 � 10 � 5X2X3 � 1:41� 10�3X2

1 þ 6:78 � 10�7X22 þ 1:52 � 10�5X2

3 (2)

where Y is the removal efficiency of LAS.Statistical testing of the model was performed with the Fisher’s

statistical test for analysis of variance (ANOVA). The results ofANOVA for LAS removal efficiency are depicted in Table 5. Thevariance analysis of quadratic regression model demonstrated that

X3 -P-650

1742.86 1 1742.86 19,054.35 <0.0001

X1X2 32.95 1 32.95 360.21 <0.0001X1X3 17.51 1 17.51 191.48 <0.0001X2X3 3.26 1 3.26 35.62 0.0006X1

2 0.24 1 0.24 2.60 0.1511X2

2 1.93 1 1.93 21.13 0.0025X3 2 0.05 1 0.05 0.56 0.4798Residual 0.64 7 0.09Lack offit

0.64 3 0.21

Pureerror

0.00 4 0.00

Cor total 2430.61 16

R-squared: 0.9997; Adj R-squared: 0.9994; Pred R-squared: 0.9958.

2114 E.L. Terechova et al. / Journal of Environmental Chemical Engineering 2 (2014) 2111–2119

this model was highly significant, as is evident from the Fisher’sF-test with a very low probability value (<0.0001). The F value of2951.81 proved that independent variables as well as theinteractions between them had significant influence on theremoval efficiency of LAS. X1, X3, X1X2, X1X3, X2X3 and X2

2 aresignificant terms (P < 0.05), which indicates that the dosages ofZnCl2 and P-650 and their interaction effect are the key factors forLAS removal.

The goodness-of-fit of the model was checked by the determi-nation coefficient (R-squared), a statistical measure providingexplanatory power to the regression model. In this case, the value ofthe determination coefficient (R-squared = 0.9997) indicates thatonly 0.03 % of the total variations are not explained by the model[25]. The adjusted determination coefficient (Adj R-squared) was0.9994, showing the high significance of the model. The “Pre R-squared” of 0.9958 was close to “Adj R-squared”. Adequateprecision was a signal to noise ratio and compared the range ofthe predicted values at the design points to the average predictionerror.

To aid visualization and help in identifying the type ofinteractions between the test variables, the response surfacesfor the removal efficiency of LAS are shown in Figs. 2–4 . Here, eachplot represented the effect of two variables at their studied rangewith the other one maintained at its zero level. The shapes ofcontour plots indicated the nature and extent of the interactions.Fig. 2 clearly showed that the removal efficiency of LAS graduallyincreased with the increasing ZnCl2 dosages. At a lower ZnCl2dosage, the removal efficiency of LAS decreased with the increase

Fig. 2. The effects of ZnCl2 and ash (P-650 = 115 mg/L) on LAS re

of ash dosage, while further increased the ZnCl2 dosage, theremoval efficiency of LAS improved obviously. Fig. 3 demonstratedthat the removal efficiency of LAS apparently increased with theincrease of ZnCl2 and P-650 dosages, thus the effect of ash on LASremoval is unconspicuous (as shown in Fig. 4). This finding couldbe explained according to the coagulation–flocculation mecha-nism described in the next section.

The main objective of the optimization is to determine theoptimum values of variables for LAS removal from the experimen-tal model. In this optimization study, the constrained optimizationprogram supplied in Design-Expert software was used. Themaximum removal efficiency of LAS was ca. 71.31%, whereasthe maximum values of the process variables in coded valuesgiven as follows: ZnCl2 = 29.54 mg/L, ash = 1936.35 mg/L andP-650 = 196.38 mg/L. To confirm the model adequacy for predictingthe maximum removal efficiency of LAS, two sets of additionalexperiments under this optimum operation condition wereperformed with the self-made LAS wastewater and the actuallaundry wastewater. As seen in Table 6, with regard to the self-made LAS wastewater, the average removal efficiency of LAS wasca. 71.26%, while the average removal efficiencies of LAS, COD andsuspension of the actual laundry wastewater in the three replicatecoagulation–flocculation experiments were 74.58%, 70.12% and55.89%, respectively. The good conformity between the predictedand experimental results verified the validity of the model andreflected the existence of an optimal point. On the other hand, thesolution pH in the two cases was slightly decreased from �9.0 to�8.0 after the coagulation–flocculation treatment.

moval efficiency: (a) surface graphs and (b) contour plots.

Fig. 3. The effects of ZnCl2 and P-650 (ash = 1000 mg/L) on LAS removal efficiency: (a) surface graphs and (b) contour plots.

E.L. Terechova et al. / Journal of Environmental Chemical Engineering 2 (2014) 2111–2119 2115

The mechanism of coagulation–flocculation process

Generally, the conventional coagulation–flocculation treatmentprocess requires 2–10 hours depending on the types of coagulantsand flocculants and their dosages. However, the coagulation–flocculation process in present work takes no longer than 30 minindicating the significant increase of the treatment rate. This findingmaybe due to the use of specific coagulant (Zn2+ salts and ash) andcationichigh-molecular flocculant (Praestol-650),whichtransferredthe conventional coagulation–flocculation treatment mechanisminto an extraction mechanism for contaminants preliminarilydestabilized by high-molecular electrolyte.

According to the results obtained in RSM research, the newmechanism for the present coagulation–flocculation treatmentprocess was proposed as follows: in wastewater LAS is present inthe free ion-molecular state, which is generally considered as thestabilizer of emulsion and suspension (Fig. 5a). When ZnCl2 wasadded, two processes could happen: (i) the conversion of LASmolecules to the insoluble complex and (ii) the partial coagulationof emulsion and suspension due to the consumption of hydratedions of Zn(OH)+ (as shown in Eq. (3) and Fig. 5b).

+

R SO3Na

Zn2+ +

R SO3

2Zn 2Na+ (3)

It is known that the coagulation–flocculation of surfactant isthrough adsorption onto ash particles and largely depends on thesurface property of the ash particles and pH of the suspension [26].In this context, the effect of solution pH on the adsorptionperformance of LAS on the mineral ash particles was investigated.As seen in Fig. 6, the mineral ash particles possess good adsorptionability towards LAS molecules in the pH range of 1.0–13.0, however,a much lower adsorption removal rate was observed at ca. pH 6.0.This phenomenon can be attributed to the fact that the mineral ashparticles and LAS molecules have the same electrical property atthis pH value, which results in the poor removal rate of LAS due tothe electrostatic repulsion. In the pH range of above 6.0, theadsorption removal rate increases significantly. The reason for thistendency may be that abundant hydroxyl ions were adsorbed ontothe mineral ash particles, which generate the hydrogen bondingconnection with the ��SO3

� groups of LAS molecules. Thus, thecoagulation of mineral ash could be explained according to thesorption mechanism. On the other hand, the ash particle surface ishydrophilic and LAS molecule also possesses the free hydrophilicgroup (sulfonated aromatic ring) in laundry wastewater, thereforethe suspension and emulsion particles and LAS molecule can bepartially destabilized by Zn(OH)+ adsorbed on the surface ofmineral ash. This resulted in the formation of intermediatecomplex aggregates (Fig. 7a), which are negatively charged withthe zeta potential value of �18.97 mV. After the addition ofcationic–flocculant (P-650), macro-aggregates would generate inwastewater through the ion-ion interaction [27–29], as shown in

Fig. 4. Surface graphs of the effects of ash and P-650 (ZnCl2 = 17 mg/L) on LAS removal efficiency: (a) surface graphs and (b) contour plots.

2116 E.L. Terechova et al. / Journal of Environmental Chemical Engineering 2 (2014) 2111–2119

Fig. 7b . Therefore, the preliminary stabilization with ash and ZnCl2is the necessary step for wastewater treatment, which cansignificantly reduce the consumption of cationic flocculants.

Photolysis by ultraviolet

According to the requirements in urban wastewater treatment,LAS concentration in wastewater should not exceed 5.0 mg/L, whilewhen the treated effluent is directly discharged into an open waterbody [30], LAS concentration is strictly controlled in the range of0.2–0.3 mg/L, depending on the category of the water body [31].Generally, the residual content of LAS in wastewater was 3–10 mg/Lafter the coagulation–flocculation process, which did not meet withthe discharged standard. Therefore, the supernatant was furthertreated using the ultraviolet photolysis to eliminate LAS in the actuallaundry wastewater. Fig. 8 presents the effect of different pH on the

Table 6The average removal efficiencies of LAS, COD and suspension in the three replicate coaguP-650 was 29.54, 1936.35 and 196.38 mg/L, respectively.

No. test LAS COD

Residual concentration (mg/L) Removal rate (%) Residual COD (

1 3.56a, 5.12b 74.21a, 73.98b 89

2 4.05a, 4.83b 70.68a, 75.46b 82

3 4.29a, 5.06b 68.90a, 74.29b 80

Mean result – 71.26a, 74.58b –

a For the the self-made LAS wastewater.b For the actual laundry wastewater.

residual LAS concentration after UV irradiation. At pH 6.0, LASconcentration exhibits a sharp decrease in the first 5 min afterirradiated, and then it was maintained at 2 mg/L. However, LASconcentration exhibited a successive declining tendency in pH 8.0.After irradiated for 30 min, LAS concentrationwas less than 0.5 mg/L,indicating that the alkaline medium is beneficial to the ultravioletphotolysis of LAS. The reason for this phenomenon is that (i) inalkaline media of pH 8.0, the residual Zn2+ in the supernatant willform Zn(OH)2; (ii) under the irradiation of ultraviolet light, theoxygen molecules in the air were photolyzed to forming atomicoxygen, and then the atomic oxygenwas reacted further with oxygenmolecule generating ozone; (iii) the supernatant taken forphotolysis certainly contains the residual Fe2O3 and TiO2 particles,which results in the ultraviolet photocatalytic oxidation and thecatalytic ozonation in photolysis process. Thus, under the combinedeffects of sedimentation, photocatalytic oxidation and the catalytic

lation–flocculation experiments. Conditions: pH �9.0, the dosages of ZnCl2, ash and

Suspensions

mg/L) Removal rate (%) Residual concentration (mg/L) Removal rate (%)

68.21 51.4 57.1670.71 54.8 54.3371.43 52.6 56.1770.12 – 55.89

Fig. 5. (a) Stabilization of emulsion and suspension in presence of LAS; (b) thecoagulation scheme of colloid particles by Zn(OH)+ ions.

Fig. 7. (a) The formation scheme of the intermediate aggregates after theadsorption of contaminants on ash particles; (b) the macro-aggregates inwastewater in the presence of Praestol-650 flocculant.

E.L. Terechova et al. / Journal of Environmental Chemical Engineering 2 (2014) 2111–2119 2117

ozonation, the removal rate of LAS was significantly promotedand the partial organic substances were broken to forming CO2

and H2O.

The products in photolysis process

After the coagulation–flocculation treatment, the supernatantcomponents of laundry wastewater were analyzed by gaschromatography (GC) before and after ultraviolet photolysis tofurther confirm the degradation degree of LAS. As seen in Table 7,the concentration of ketones (acetones), alcohols and fatty acids inwastewater were lower than the maximum permissible concen-tration (MPC) norms [32,33] after the ultraviolet photolysis. Thisfinding testified the fact that the destruction of organic substancesin wastewaters by ultraviolet photolysis process mainly involvedthe formation of non-toxic components (E?2 and =2?) [11,17,18],

Fig. 6. The effect of solution pH on the adsorption performance of LAS on themineral ash particles.

and also affirmed that no secondary pollutants were generated insewage after the coagulation–flocculation/ultraviolet photolysistreatment.

Fig. 8. Residual LAS concentration in coagulation–flocculation supernatant afterthe ultraviolet photolysis under the condition of the different solution pH.

Table 7Results of gas chromatography.

Parameters (mg/L) After coagulation–flocculation treatment After photolysis t = 30 min MPC in water body (mg/L) Standard

Ketones (acetones) 0.162 0.244 0.55AlcoholsMethyl 0.0984 0.174 3 [29,30]Ethyl 0.135 0.099 –

Isopropyl 0.0198 – 0.25

Fatty acids E2–E10

Acetic (E2) – – 0.55 [30]Butyric (E4) – – 0.55Valeric (E5) 0.0367 0.0377 0.55Caprylic acid (E8) 0.127 0.126 1.0

Fatty acids E8–E22

Capric (E10) 0.00494 0.034 1.0 [30]Palmitic (E16) 0.049 0.24 1.0Stearin (E18) 0.315 0.138 1.0

2118 E.L. Terechova et al. / Journal of Environmental Chemical Engineering 2 (2014) 2111–2119

Conclusions

In this study, a combined chemical coagulation–flocculation/photolysis process was proved to be a suitable treatment strategyfor dealing with LAS in laundry wastewater. Mineral ash, ZnCl2,and P-650 were chosen as coagulant-sorbent, complex formerand cationic high molecular flocculants, respectively, in thecoagulation–flocculation process. Their dosages were optimizedby RSM. Under the condition of the optimal parameters ofZnCl2 = 29.54 mg/L; ash = 1936.35 mg/L and P-650 = 196.38 mg/L,71.26% and 74.58% of LAS were removed from the self-made LASwastewater and the actual laundry wastewater, respectively. Thecoagulation–flocculation process was consistent with the desta-bilization and the extraction of LAS by the high-molecularflocculants through the combined mechanisms of chargeneutralization and bridging. The alkaline environment is benefi-cial to LAS photolysis removal. Results indicated that thecombined chemical coagulation–flocculation/ultraviolet photol-ysis process is an environmentally friendly strategy for laundrywastewater treatment due to its efficient and rapid treatment rateand novel charge neutralization/extraction mechanism.

Acknowledgements

We acknowledge the financial support from the NationalNatural Science Foundation of China (No. 21177017) and theFundamental Research Funds for the Central Universities (No.DUT13LK50).

References

[1] C.A. Basar, A. Karagunduz, A. Cakici, B. Keskinler, Removal of surfactantsby powdered activated carbon and microfiltration, Water Res. 38 (2004)2117–2124, doi:http://dx.doi.org/10.1016/j.watres.2004.02.001. 15087193.

[2] Y. Kaya, C. Aydiner, H. Barlas, B. Keskinler, Nanofiltration of single and mixturesolutions containing anionics and nonionic surfactants below their criticalmicelle concentrations (CMCs), J. Membr. Sci. 282 (2006) 401–412, doi:http://dx.doi.org/10.1016/j.memsci.2006.05.047.

[3] G.G. Ying, Fate, behavior and effects of surfactants and their degradationproducts in the environment, Environ. Int. 32 (2006) 417–431, doi:http://dx.doi.org/10.1016/j.envint.2005.07.004. 16125241.

[4] I. Ciabattia, F. Cesaro, L. Faralli, E. Fatarella, F. Tognotti, Demonstration ofa treatment system for purification and reuse of laundry wastewater,Desalination 245 (2009) 451–459, doi:http://dx.doi.org/10.1016/j.desal.2009.02.008.

[5] I. Kowalska, M.K. Korbutowicz, K.M. Nowak, T. Winnicki, Separation of anionicsurfactants on ultrafiltration membranes, Desalination 162 (2004) 33–40, doi:http://dx.doi.org/10.1016/S0011-9164(04)00024-4.

[6] G.S. Fomin, Water: Control of Chemical, Bacterial and Radiation Safety toInternational Standards (An Encyclopedic Reference), Goskomstandart,Moscow, 2000, pp. 750.

[7] A.K. Mungray, P. Kumar, Mass balance of anionic surfactants through up-flow anaerobic sludge blanket based sewage treatment plants, ProcessSaf. Environ. Prot. 87 (2009) 254–260, doi:http://dx.doi.org/10.1016/j.psep.2009.03.004.

[8] J.K. Braga, M.B.A. Varesche, Commercial laundry water characterization foranaerobic treatment in fluidized bed reactor, X Oficina e Seminário LatinoAmericano de Digestão Anaeróbia (DAAL), Ouro Preto, MG, 2011.

[9] T. Cserháti, E. Forgács, G. Oros, Biological activity and environmental impact ofanionic surfactants, Environ. Int. 28 (2002) 337–348, doi:http://dx.doi.org/10.1016/S0160-4120(02)00032-6. 12437283.

[10] S. Šostar-Turk, I. Petrini�c, M. Simoni9c, Laundry wastewater treatment usingcoagulation and membrane filtration, Resour. Conserv. Recycl. 44 (2005)185–196, doi:http://dx.doi.org/10.1016/j.resconrec.2004.11.002.

[11] M. Rosu, A. Marlina, A. Kaya, A. Schumpe, Surfactant adsorption onto activatedcarbon and its effect on absorption with chemical reaction, Chem. Eng. Sci. 62(2007) 7336–7343, doi:http://dx.doi.org/10.1016/j.ces.2007.08.027.

[12] A.H. Mahvi, A. Maleki, B. Roshani, Removal of anionic surfactants in detergentwastewater by chemical coagulation, Pakistan J. Biol. Sci. 7 (2004) 2222–2226,doi:http://dx.doi.org/10.3923/pjbs.2004.2222.2226.

[13] W. Kong, B. Wang, H. Ma, L. Gu, Electrochemical treatment of anionicsurfactants in synthetic wastewater with three-dimensional electrodes, J.Hazard. Mater. 137 (2006) 1532–1537, doi:http://dx.doi.org/10.1016/j.jhaz-mat.2006.04.037. 16730897.

[14] Y.Y. Eng, V.K. Sharma, A.K. Ray, Photocatalytic degradation of nonionic surfactant,Brij 35 in aqueous TiO2 suspensions, Chemosphere 79 (2010) 205–209, doi:http://dx.doi.org/10.1016/j.chemosphere.2010.01.042. 20171712.

[15] I. Kowalska, Surfactant removal from water solutions by means of ultrafiltra-tion and ion-exchange, Desalination 221 (2008) 351–357, doi:http://dx.doi.org/10.1016/j.desal.2007.01.094.

[16] F. Aloui, S. Kchaou, S. Sayadi, Physicochemical treatments of anionicsurfactants wastewater: effect on aerobic biodegradability, J. Hazard. Mater.164 (2009) 353–359, doi:http://dx.doi.org/10.1016/j.jhazmat.2008.08.009.18799262.

[17] D. Prats, F. Ruiz, B. Vazquez, M. Rodriguez-Pastor, Removal of anionic and non-ionic surfactants in a wastewater treatment plant with anaerobic digestion. Acomparative study, Water Res. 8 (1997) 1925–1930.

[18] J. Kaleta, M. Elektorowicz, The removal of anionic surfactants from water incoagulation process, Environ. Technol. 34 (2013) 999–1005, doi:http://dx.doi.org/10.1080/09593330.2012.733415. 23837351.

[19] T. Chatterjee, S. Chatterjee, D.S. Lee, M.W. Lee, S.H. Woo, Coagulation of soilsuspensions containing nonionic or anionic surfactants using chitosan,polyacrylamide, and polyaluminium chloride, Chemosphere 75 (2009)1307–1314, doi:http://dx.doi.org/10.1016/j.chemosphere.2009.03.012.19345394.

[20] A.M. Amat, A. Arques, M.A. Miranda, S. Seguí, Photo-Fenton reaction for theabatement of commercial surfactants in a solar pilot plant, Solar Energy 77(2004) 559–566, doi:http://dx.doi.org/10.1016/j.solener.2004.03.028.

[21] A.M. Amat, A. Arques, M.A. Miranda, R. Vincente, S. Seguí, Degradation of twocommercial anionic surfactants by means of ozone and/or UV irradiation,Environ. Eng. Sci. 24 (2007) 790–794, doi:http://dx.doi.org/10.1089/ees.2006.0030.

[22] T.P. Delforno, A.G.L. Moura, D.Y. Okada, M.B.A. Varesche, Effect of biomassadaptation to the degradation of anionic surfactants in laundry wastewaterusing EGSB reactors, Bioresour. Technol. 154 (2014) 114–121, doi:http://dx.doi.org/10.1016/j.biortech.2013.11.102. 24384318.

[23] C.A. Basar, A. Karagunduz, A. Cakici, B. Keskinler, Removal of surfactants bypowdered activated carbon and microfiltration, Water Res. 38 (2004) 2117–2124,doi:http://dx.doi.org/10.1016/j.watres.2004.02.001. 15087193.

[24] G.E.P. Box, D.W. Behnken, Some new three level designs for the study ofquantitative variables, Technometrics 2 (1960) 455–475, doi:http://dx.doi.org/10.1080/00401706.1960.10489912.

E.L. Terechova et al. / Journal of Environmental Chemical Engineering 2 (2014) 2111–2119 2119

[25] T.K. Trinh, L.S. Kang, Response surface methodological approach to optimize thecoagulation–flocculation process in drinking water treatment, Chem. Eng. Res.Des. 89 (2011) 1126–1135, doi:http://dx.doi.org/10.1016/j.cherd.2010.12.004.

[26] A. Srinivasan, M.W. Grutzeck, The adsorption of SO2 by zeolites synthesized byfly ash, Environ. Sci. Technol. 33 (1999) 1464–1469, doi:http://dx.doi.org/10.1021/es9802091.

[27] K. Zhu, M.G. El-Din, A.K. Moawad, D. Bromley, Physical and chemical processesfor removing suspended solids and phosphorus from liquid swine manure,Environ. Technol. 25 (2004) 1177–1187, doi:http://dx.doi.org/10.1080/09593332508618385. 15551832.

[28] F. Sher, A. Malik, H. Liu, Industrial polymer effluent treatment by chemicalcoagulation and flocculation, J. Environ. Chem. Eng. 1 (2013) 684–689, doi:http://dx.doi.org/10.1016/j.jece.2013.07.003.

[29] K.E. Lee, N. Morad, T.T. Teng, B.T. Poh, Development, characterization and theapplication of hybrid materials in coagulation/flocculation of wastewater: areview, Chem. Eng. J. 203 (2012) 370–386, doi:http://dx.doi.org/10.1016/j.cej.2012.06.109.

[30] Discharge standard of pollutants for municipal wastewater treatment plant,the national standard of the People’s Republic of China, GB 18918–2002.

[31] Environmental quality standards for surface water, the national standard ofthe People’s Republic of China, GB 3838–2002.

[32] World Health Organization Guidelines for Drinking-water Quality thirdedition. Vol. 1, 2004, Geneva.

[33] EU’s Drinking Water Standards. Council Directive 98/83/EC on the quality ofwater intended for human consumption. Adopted by the Council, on3 November 1998.