Transcript
Insights of ibuprofen electro-oxidation on metal-oxide-coated Tianodes: Kinetics, energy consumption and reaction mechanisms Chong Wang, Yanxin Yu, Lifeng Yin, Junfeng Niu * , Li-An Hou State Key Laboratory of Water Environment Simulation, Beijing Normal University, Beijing 100875, PR China h i g h l i g h t s g r a p h i c a l a b s t r a c t Electrochemical degradation of ibuprofen (IBP) was investigated. The degradation ef ciency was gov-erned by applied current and platedistance. The economic feasibility was evalu-ated by energy consumption per or-der ( E EO ). The electrochemical mineralizationmechanism of IBP was proposed. a r t i c l e i n f o Article history: Received 27 June 2016Received in revised form11 August 2016Accepted 12 August 2016Available online 25 August 2016Handling Editor: E. Brillas Keywords: IbuprofenElectrochemical oxidationElectrodesEnergy costMineralizationMechanisms a b s t r a c t Electrochemical degradation of ibuprofen (IBP) was performed on three types of Ti-based metal oxideelectrodes. The degradation of IBP followed pseudo- rst-order kinetics and the electrochemical degra-dation rate constant ( k ) over Ti/SnO 2 -Sb/Ce-PbO 2 (9.4 10 2 min 1 ) was 2.0 and 1.7 times of the valuesover Ti/Ce-PbO 2 (4.7 10 2 min 1 ) and Ti/SnO 2 -Sb (5.6 10 2 min 1 ), respectively. The removal of totalorganic carbon and the energy consumption per order for IBP degradation were 93.2% and 13.1 Wh L 1 ,respectively, under the optimal conditions using Ti/SnO 2 -Sb/Ce-PbO 2 anode. Six aromatic intermediateproducts of IBP were identi ed by ultra-high-performance liquid chromatography coupled with aquadrupole time-of- ight mass spectrometer. The electrochemical mineralization mechanism of IBP wasproposed. It was supposed that OH radicals produced on the surface of anode attacked IBP to formhydroxylated IBP derivatives that werethen followed bya series of hydroxylation, loss of isopropanol andisopropyl,decarboxylation and benzene ring cleavage processes to form simple linearcarboxylic acids. Bysuccessive hydroxylation, these carboxylic acids were then oxidized to CO 2 and H 2 O, achieving thecomplete mineralization of IBP. © 2016 Elsevier Ltd. All rights reserved. 1. Introduction Ibuprofen (IBP) is a non-prescription, non-steroidal anti-in ammatory drug which has been widely used in the treatmentof fever, migraine, musculature pain and in ammatory rheumaticdiseases. More than 15,000 tons of IBP are produced worldwideeach year, but a signi cant percentage is excreted by patients in itsoriginalformorasmetabolites(Ghauchetal.,2012;Markovicetal.,2015).IBPhasbeendetectedinsurfacewaterandwastewaterattherange from nanogram to low microgram amounts per litter, and iscontinuously being discharged into the environment (Eslami et al.,2015; Khan et al., 2014; Nebot et al., 2015; Wang et al., 2011). Theubiquityof IBP is particularlyalarming due tothe potential adverseimpact on the reproduction of aquatic organisms and human * Corresponding author. E-mail address:
[email protected] (J. Niu). Contents lists available at ScienceDirect Chemosphere journal homepage: www.elsevier.com/locate/chemosphere http://dx.doi.org/10.1016/j.chemosphere.2016.08.0570045-6535/ © 2016 Elsevier Ltd. All rights reserved. Chemosphere 163 (2016) 584 e 591 endocrine system even at low concentrations (Han et al., 2010;Kang et al., 2015). Hence it is of great importance to developeffective technology to eliminate IBP from aquatic environment.Various techniques have been studied to prevent the accumu-lation of IBP in the aquatic environment. Some physical methods,such as adsorption (Guedidi et al., 2013; Iovino et al., 2015) ormembrane nano ltration (Narbaitz et al., 2013), did not lead todegradation either, which just involved phase transfer of IBP andthus induced another pollution problem. The drawbacks of bio-logical processes were generally inef cient and the subsequentdisposal of sludge (Kosjek et al., 2007; Londo ~ no and Pe ~ nuela, 2015;Quintana et al., 2005). Chemical methods such as ozonation(Quero-Pastor et al., 2014) and ionizing radiation (Ill es et al., 2013),as well as some advanced oxidation processes (AOPs) such asphotocatalytic systems and Fenton reaction based processes, couldprovide effective degradation of IBP (Ambuludi et al., 2013b; Liet al., 2015; Molinari et al., 2006; Xiang et al., 2016). Neverthe-less,strictoperationalconditions(e.g.,oxygensupplyandpH)wererequired, or more toxic intermediates, like 1,3-dihydroxylibuprofen and 1-(4-isobutylphenyl)ethanone, formed in somemethods (Ill es et al., 2013; Quero-Pastor et al., 2014). The devel-opment of cost-effective technologies to eliminate IBP with highmineralization rate is thus of great challenge.The electrochemical advanced oxidation process (EAOP) hasrecently demonstrated its great ability to ef ciently abate envi-ronmental recalcitrant pollutants (García-G omez et al., 2014; Linet al., 2013a, 2013b; Niu et al., 2012; Shmychkova et al., 2015). In-formations about the removal of chemical oxygen demand (COD)and total organic carbon (TOC) for IBP solutions were providedusing Pt, Ti/Pt/PbO 2 and boron-doped diamond (BDD) electrodes(Ambuludi et al., 2013a; Ciríaco et al., 2009; Lima et al., 2013). TheBDD anode had better characteristics for the TOC removal of IBPsolution, which could achieve more than 92% with the currentdensity of 20 e 30 mA cm 2 after 360 e 480 min of electrolysis in200 mL IBP solution (Ambuludi et al., 2013a; Ciríaco et al., 2009).Unfortunately, intermediate products were not followed, or just afew small molecular carboxylic acids were identi ed in thesestudies, while explicit mineralization pathway of IBP by electro-chemical treatment has not been fully understood yet. Besides, thehigh production cost makes BDD not feasible for large-scaleapplication (Lin et al., 2013a; Panizza and Cerisola, 2009). Con-cerning over the commercial application of electrochemicaldegradation of IBP, it is essential to search anodes with highmineralization rate, mild condition and low cost.The explored Ti-based metal oxide electrodes (e.g., Ti/RuO 2 , Ti/SnO 2 and Ti/PbO 2 ) have proven to be ef cient electrode materialsfor the treatment of recalcitrant pollutants in aqueous solution(García-G omez et al., 2014; Shmychkova et al., 2015; Santos et al.,2013; Xu et al., 2015; Yang et al., 2015; Zhao et al., 2014; Zhouet al., 2005). In previous researches, we found that Ti/SnO 2 -Sband Ti/SnO 2 -Sb/Ce-PbO 2 electrodes can rapidly and almostcompletely remove various organic contaminants, such as penta-chlorophenol, per uorocarboxylic acids (PFCAs) and sulfamethox-azole from aqueous solution (Lin et al., 2013a, 2013b; Niu et al.,2013). Even the performance of Ti/SnO 2 -Sb/Ce-PbO 2 electrodewas comparable with that of BDD electrode for the degradation of PFCAs (Lin et al., 2013b). Thus Ti-based metal oxide electrodes mayhave great potentiality to completely mineralize IBP to harmlessbyproducts.The aim of this study was to investigate the performance of theelectrochemical oxidation process for the degradation of IBP usingthree Ti-based metal oxide anodes. The mineralization of IBP waselucidated by assessing the removal of the TOC. The energy con-sumptionperorder( E EO )ofIBPdegradationwasevaluatedtoassessthe application potential of the electrochemical oxidationtechnology. The intermediate products of IBP formed during theelectrochemical degradation were identi ed to reveal the electro-chemical mineralization mechanism of IBP. 2. Materials and methods 2.1. Chemicals Analytical grade IBP, Salicylic Acid (SA), 2,3-dihidroxybenzoicAcid (2,3-DHBA) and 2,5-dihydroxibenzoic Acid (2,5-DHBA) werepurchased from Sigma Aldrich. All other chemicals were analyticalgrade and were purchased from Sinopharm, China. All aqueoussolutions used in our experiments were prepared with Milli-Q deionized water (18.2 M U cm). The pH values of the solutionwere adjusted by H 2 SO 4 (5%, v v 1 ) and NaOH (5%, m m 1 ). 2.2. Electrode fabrication and characterization Titanium sheets (purity 99.9%, 50 mm 50 mm 1 mm) werepreviously polished with different grades papers to a mirror-likesurface, then immersed in NaOH solution (10%, m m 1 ) andetched with boiling oxalic acid solution (10%, m m 1 ). Ti/SnO 2 -Sbelectrode and the middle layer of the Ti/SnO 2 -Sb/Ce-PbO 2 wereprepared using the sol-gel technique with a coating solution con-taining ethylene glycol, citric acid, SnCl 4 $ 4H 2 O and SbCl 3 with amolar ratio of 140:30:9:1. The sol-gel solution was used to coat onthe Ti sheets by the dip-coating method. Then, the Ti sheets weredried at 145 C for 10 min and sintered at 500 C for 10 min in amuf e furnace. This procedure was repeated 20 times, and the lastbaking was annealed for 2 h at 500 C to obtain Ti/SnO 2 -Sb elec-trode (Lin et al., 2013a).Ti/Ce-PbO 2 and the top layer of Ti/SnO 2 -Sb/Ce-PbO 2 electrodewereprepared byelectro-deposition. The Ce-doped PbO 2 layer wasgenerated coating on the Ti and Ti/SnO 2 -Sb, respectively, with anacidic electrolyte (68 C) consisting of 0.1 M HNO 3 , 200 g L 1 Pb(NO 3 ) 2 ,0.4gL 1 Ce(NO 3 ) 3 ,and0.5gL 1 NaFataconstantcurrentdensity of 20 mA cm 2 for 60 min (Lin et al., 2013b).The microstructure and morphology of the prepared electrodeswere observed by scanning electron microscopy (SEM; S4800,Hitachi, Japan) with an accelerating voltage of 10 kV and X-raydiffraction (XRD; Thermo ARL SCINTAG XTRA, Netherlands) usingCu K a radiation, respectively. The oxygen evolution potential (OEP)of the electrodes were measured with the linear sweep voltam-metry (LSV) technique conducted in conventional three-electrodeelectrochemical cells on an electrochemical workstation (CHI660D, Chenhua, Shanghai, China). The prepared electrodes servedastheworkelectrode(1cm 2cm),aplatinumfoilandasaturatedcalomel electrode (SCE) as the counter electrode and the referenceelectrode, respectively. The characterized results are shown inFigs. SM-1 e 3 of the Supplementary Material (SM). 2.3. Electrochemical experiments The electrochemical oxidation experiments were conducted inundivided electrolytic cells made of organic glass. The Ti/SnO 2 -Sb/Ce-PbO 2 , Ti/Ce-PbO 2 and Ti/SnO 2 -Sb electrodes were used as an-odes,andTisheetsasthecathodes.Beforeelectrolysisexperiments,controlledtrialswithnopowersupplywereconducted. Noobviouslosses of IBP were detected after 240 min, indicating that degra-dation by sorption or hydrolysis was negligible. For most experi-ments,a relative high initial concentration of IBP(20.0 mg L 1 ) wasused to investigate the effects of main operating factors on thedegradation ef ciency and the monitoring of major degradationintermediates during the treatment. When investigating the effectofoperatingparametersonIBPelectrochemicaldegradation,30mL C. Wang et al. / Chemosphere 163 (2016) 584 e 591 585 of IBP solutionwas used to conduct the experiments with differentcurrent density (2 e 40 mA cm 2 ), initial pH value (4.0 e 12.0) with10mMNa 2 SO 4 assupportingelectrolyte.30 e 100mLofIBPsolutionwas used toinvestigate the effectof electrode distance (5 e 20 mm).Besides, relative low initial IBP concentrations (1.0 e 10.0 mg L 1 )were also used to better simulate the degradation ef ciency of thereal wastewater. The solution was not stirred due to the shortelectrodes distance. The electrolysis was conducted with a powersupply (DH1715A, Beijing Dahua Electronic Co., China) at constantcurrent. The solution pH was detected by a microprocessor pHmeter (pH211, EUTECH Co., USA). All the electrochemical experi-ments were conducted at least in triplicate and operated at(25 ± 1) C. 2.4. Instrumental analysis The concentrations of IBP were determined by a high-performance liquid chromatography (HPLC, Dionex U3000, USA)equipped with a uorescence detector and an Athena C18-WPcolumn (4.6 mm 250 mm, 5 m m particle size). The productionof hydroxyl radical ( OH) was determined with SA trapping andanalyzed byHPLC-UV (Peralta et al., 2014). The details areprovidedinTextSM-1andSM-2oftheSM.TOCwasmeasuredusingaTOC-L-CPN analyzer (Shimadzu Co, Japan) to evaluate the mineralizationof IBP.The intermediate products formed during the electrochemicalprocess of IBP were identi ed using an ultra-high-performanceliquid chromatography coupled with a quadrupole time-of- ightmass spectrometer (UPLC-Q-TOF-MS, Xevo G2, Waters Corp, USA)equipped with an ACQUITY UPLC BEH C18 column (2.1mm 100 mm; 1.7 m m particle size; Waters, USA). The columnovenwas kept at 35 C. The mobile phase Awas Milli-Q water with0.1% formic acid (v v 1 ), and the mobile phase B was acetonitrile.The ow rate and gradient condition are showninText SM-3 of theSM. Before analysis, solution sample was prepared by an enrich-ment and clean-up step with a solid phase extraction (SPE) systemwith HLB cartridges (6 mL, 500 mg, Waters, Watford, UK). 2.5. Energy consumptionE EO is a powerful scale-up parameter of merit forcomparing theef ciency and electric cost of the treatment system with otherAOPs. The E EO for operating system can be calculated in Eq. (1) (Xia et al., 2015): E EO ¼ SjUt V log ð C 0 = C Þ (1) where E EO is the electric energy consumed to degrade the con-centration of IBP by one order of magnitude in 1 L solution(Wh L 1 ), S is the anode surface area (cm 2 ), j is the applied currentdensity (mA cm 2 ), U is the recorded average voltage (V) duringelectrolysis in each experimental condition, and V is the volume of the reaction solution (mL), C 0 is the initial concentration of IBP(mg L 1 ), and C is the nal concentration of IBP (mg L 1 ), t is thetime (min) needed to degrade IBP from the initial concentration tothe nal concentration. 3. Results and discussion 3.1. Electrochemical degradation kinetics of IBP 3.1.1. Effect of anode materials Since OH played a signi cant role in electrochemical degrada-tion of recalcitrant pollutants over “ non-active ” anodes (Carter andFarrell, 2008; Lin et al., 2013a; Liao and James, 2009; Zhao et al.,2010), the generation capacity of OH on the anodes under a cur-rent density of 5 mA cm 2 was determined based on their reactionwithSAtoformhydroxylatedproducts.Inourstudy,only2,5-DHBAwas detected during the electrolysis process, so the OH concen-tration was equal to the 2,5-DHBA concentration according to thereaction stoichiometry. Fig. 1 shows that the concentration of OHgenerated on the Ti/SnO 2 -Sb/Ce-PbO 2 anode was higher than thaton Ti/Ce-PbO 2 and Ti/SnO 2 -Sb anodes. The production of OH fol-lowed pseudo-zero-order kinetics ( R 2 > 0.99) before 15 min on allanodes due tothe excess of SA. The production rate constants ( k ) of OH were 17.2, 12.3, and 14.5 mM (min m 2 ) 1 on Ti/SnO 2 -Sb/Ce-PbO 2 , Ti/Ce-PbO 2 and Ti/SnO 2 -Sb anodes, respectively, at the cur-rent density of 5 mA cm 2 . (It is noted that these calculated valuesmight be lower than actual generation rate due to the short lifetime of OH (Peralta et al., 2014) and their incompletely reactionwith SA.) This suggested that Ti/SnO 2 -Sb/Ce-PbO 2 might have thebest electrochemical oxidation performance.To verify above speculation, the degradation of IBP on thesethree metal-oxide-coated Ti based electrodes was carried out at5mAcm 2 withtheinitialIBPconcentrationof20.0mgL 1 (Fig.2).The degradation ratios of IBP on Ti/SnO 2 -Sb/Ce-PbO 2 , Ti/Ce-PbO 2 and Ti/SnO 2 -Sb anodes were 92.4%, 65.2% and 78.6%, respectively,after 30 min of electrolysis. The degradation of IBP followedpseudo- rst-order kinetics on all anodes. The electrochemicaldegradation rate constant ( k ) over Ti/SnO 2 -Sb/Ce-PbO 2 (9.4 10 2 min 1 ) was 2.0 and 1.7 times of the values over Ti/Ce-PbO 2 (4.7 10 2 min 1 ) and Ti/SnO 2 -Sb (5.6 10 2 min 1 ),respectively. Besides, the degradation of IBP were signi cantly Fig.1. Concentrations of hydroxyl radicals produced on Ti/Ce-PbO 2 , Ti/SnO 2 -Sb and Ti/SnO 2 -Sb/Ce-PbO 2 electrodes at the applied current density of 5 mA cm 2 . Initialtrapping agent SA: 7 mM, Na 2 SO 4 : 10 mM, electrode distance: 5 mm, pH: withoutadjusted, T: (25 ± 1) C. Fig. 2. Electrochemical degradation of 20.0 mg L 1 IBP without (solid lines) and with(dotted lines) isopropanol (IPA) (5%, v v 1 ) on IBP removal by electrochemical degra-dation by Ti/Ce-PbO 2 , Ti/SnO 2 -Sb and Ti/SnO 2 -Sb/Ce-PbO 2 electrodes at the appliedcurrent density of 5 mA cm 2 . Na 2 SO 4 :10 mM, electrode distance: 5 mm, pH: withoutadjusted, T: (25 ± 1) C. C. Wang et al. / Chemosphere 163 (2016) 584 e 591 586 inhibitedonallanodes(see Fig.2)byadding5% (vv 1 )isopropanol(IPA), one scavenger of OH (Li et al., 2009), to the solution. Thisproved indirectly that the OH generation ability of the electrodematerial dominated the electrochemical degradation of IBP. ThusTi/SnO 2 -Sb/Ce-PbO 2 , which had the highest generation capacity of OH, was used to conduct the following experiments. 3.1.2. Effect of applied current density Theeffectof currentdensityon thedegradationef ciencyof IBP(20.0 mg L 1 ) with Ti/SnO 2 -Sb/Ce-PbO 2 electrode was determinedand the result is shown in Fig. 3a. The results showed that theincreasing current density could signi cantly accelerate thedegradation rate of IBP. The k values increased from 5.8 10 2 to2.1 10 1 min 1 as the current density increased from 2 to40 mA cm 2 (see Table 1). A much quicker degradation rate wasobserved at a higher applied current density ( 20 mA cm 2 ), withalmost complete degradation of IBP after 30 min, which wasprobably due to the higher OH production with higher currentdensities (Lin et al., 2012). AsshowninTable 1, the t 1/2 values of IBPwere11.9,7.3,5.4,4.4,3.7and3.4minforthecurrentdensityvaluesof 2, 5, 10, 20, 30, and 40 mA cm 2 , respectively. The results indi-cated that the t 1/2 values decreased with limited as the appliedcurrentdensity was higher than 20 mAcm 2 , implying that furtherincrease current density ( 20 mA cm 2 ) was not as ef cient. The Fig. 3. The degradation of IBP in 10 mM Na 2 SO 4 at (25 ± 1) C. (a) Effect of the applied current density on degradation of 20 mg L 1 IBP at electrode distance 5 mm, pH: withoutadjusted. (b) Effect of the electrode distance on degradation of 20 mg L 1 IBP at applied current density 10 mA cm 2 , pH: without adjusted. (c) Effect of the initial pH on degradationof 20 mg L 1 IBPat electrode distance 5 mm at applied current density 10 mA cm 2 . (d) Effect of the initial concentration on degradation of IBPat electrode distance 5 mm at appliedcurrent density 10 mA cm 2 , pH: without adjusted. Table 1 The kinetics and energy cost for the IBP degradation by Ti/SnO 2 -Sb/Ce-PbO 2 electrode.Parameters Average voltage (V) Rate constants ( k , min 1 ) a Half-lives ( t 1/2 , min) R 2 E EOb (Wh L 1 )Current density (mA cm 2 ) 2 3.9 5.8 10 2 11.9 0.995 4.35 4.2 9.4 10 2 7.3 0.993 7.110 5.2 1.3 10 1 5.4 0.991 13.120 5.8 1.6 10 1 4.4 0.986 23.330 6.5 1.8 10 1 3.7 0.989 33.640 6.8 2.1 10 1 3.4 0.995 42.1Plate distance (mm) 5 5.2 1.3 10 1 5.4 0.991 13.110 5.5 7.6 10 2 9.1 0.994 16.615 6.1 4.3 10 2 15.9 0.993 26.920 6.8 3.1 10 2 22.7 0.987 30.6pH 4 4.5 1.3 10 1 5.2 0.989 10.86 4.8 1.2 10 1 5.3 0.990 13.08 4.6 1.2 10 1 5.9 0.997 12.410 4.7 9.1 10 2 7.5 0.994 16.012 4.8 7.7 10 2 9.0 0.996 20.1Initial concentration ( C 0 , mg L 1 ) 1 6.8 2.1 10 1 3.2 0.990 4.72 5.9 1.8 10 1 3.8 0.987 6.85 4.9 1.6 10 1 4.3 0.991 8.510 5.2 1.4 10 1 5.0 0.986 11.620 5.2 1.3 10 1 5.4 0.991 13.1Other operating conditions are given in Fig. 3. a Pseudo- rst-order rate constants of electrochemical degradation. b Electrical energy cost per order of magnitude in 1 L. C. Wang et al. / Chemosphere 163 (2016) 584 e 591 587
