Mineralization of aniline and 4-chlorophenol in acidic solution by ozonation catalyzed with Fe2+ and

Applied Catalysis B:Environmental29(2001)135–145Mineralization of aniline and4-chlorophenol in acidic solution by ozonation catalyzed with Fe2+and UV A lightRoser Sauleda,Enric Brillas∗Department de Qu´ımica Fisica,Laboratori de Ciència i Tecnologia Electroqu´ımica de Materials,Facultat de Qu´ımica,Universitat deBarcelona,Mart´ıi Franauès1-11,08028Barcelona,SpainReceived8April2000;received in revised form23June2000;accepted23June2000AbstractSolutions with1.07mmol dm−3aniline or with1.38mmol dm−34-chlorophenol at pH ca.3have been treated with ozone and ozonation catalyzed with Fe2+and/or UV A.The initial mineralization rate increases as more oxidizing hydroxyl radical is produced in the medium by the catalyzed ozonations.Direct ozone treatment leads to stable oxidation products,which are quickly destroyed under UV A illumination.In the presence of Fe2+as catalyst,the degradation process is inhibited by the formation of Fe3+complexes with short organic diacids,being photodecomposed by UV A light.Each initial pollutant is destroyed at similar rate in all processes.p-Benzoquinone and nitrobenzene are identified as intermediates of aniline oxidation. The former product is only detected when high amounts of hydroxyl radical are produced by the action of Fe2+.Ammonium ion released during p-benzoquinone formation is also generated in larger extension under the same conditions.Nitrate ion reaches maximum production under UV A irradiation,indicating that generation of nitrobenzene from selective attack of O3on the amino group of aniline is photocatalyzed.Reaction of4-chlorophenol with ozone leads to4-chloro-1,3-dihydroxybenzene and4-chloro-1,2-dihydroxybenzene.The last product is produced in larger extension when high amounts of hydroxyl radical can selectively attack the initial pollutant.Chloride ion is completely lost during the further degradation of both dihydroxylated derivatives.Oxidation of all aromatic intermediates detected during aniline and4-chlorophenol degradation gives maleic acid, which is further mineralized via oxalic acid.A general reaction pathway for the degradation of each pollutant is proposed.©2001Elsevier Science B.V.All rights reserved.Keywords:Aniline;4-Chlorophenol;UV A light;Ozone;Fe(II)1.IntroductionIn the last years,a wide variety of methods are being applied to the mineralization(i.e.conversion to CO2and inorganic ions)of recalcitrant organic con-taminants in wastewaters.Among these technologies, ozonation is widely used for drinking water disinfec-∗Corresponding author.Tel.:+34-93-4021223;fax:+34-93-4021231.E-mail address:e.brillas@qf.ub.es(E.Brillas).tion and its effectiveness towards organics oxidation is well proven[1].The practical use of ozonation for wastewater treatment,however,is limited by its high energy demand.Several approaches have been taken to improve the oxidizing power of this technique lead-ing to the reduction of the required reaction time and hence,decreasing its energy cost.In this way,combi-nation of ozonation with UV light[1],homogeneous catalysts,such as H2O2[1,2]and Fe2+[3–5],and pho-tocatalysis with TiO2[6–9],has shown an important increase in its effectiveness on contaminant removal.0926-3373/01/$–see front matter©2001Elsevier Science B.V.All rights reserved. PII:S0926-3373(00)00197-1136R.Sauleda,E.Brillas/Applied Catalysis B:Environmental29(2001)135–145It is well known[1,2]that in acidic medium ozone selectively attacks organic species,usually follow-ing second-order reactions with small rate constants. When the solution is irradiated with UV light(O3/UV system),the degradation of pollutants becomes faster and non-selective because they mainly react with hydroxyl radical(OH•).For UV irradiation atλ< 300nm,this species is produced by photodissociation of H2O2previously formed from ozone photolysis[1]: O3+H2O+hν→H2O2+O2(1) H2O2+hν→2OH•(2) whereas if UV A light withλ≥300nm is utilized,di-rect reaction of ozone with H2O2produced by reaction (1)is the main source of hydroxyl radicals[1]:O3+H2O2→OH•+HO2•+O2(3) and weaker oxidizing hydroperoxyl radicals(HO2•) are also produced.Several authors[3–5]have shown that Fe2+cat-alyzes O3decomposition to generate hydroxyl radical. In fact,the catalytic O3/Fe2+system involves direct reaction of Fe2+with ozone to give the intermediate FeO2+,species that evolves to OH•[4]:Fe2++O3→FeO2++O2(4) FeO2++H2O→Fe3++OH•+OH−(5) The intermediate FeO2+is also able to oxidize Fe2+ to Fe3+at slower rate[4]:FeO2++Fe2++2H+→2Fe3++H2O(6) thus limiting the generation of hydroxyl radical for high Fe2+concentration[9].In previous work[9],the degradation of herbicide 2,4-D at pH ca.3has been comparatively tested by ozonation catalyzed with Fe2+and/or UV A light.The use of direct ozonation,as well as of the O3/Fe2+ and O3/UV A systems,leads to a partial removal of2,4-D,whereas it is completely mineralized by Fe2+catalyzed ozonation under UV A illumination (O3/Fe2+/UV A system).The highest oxidative capa-bility of the last system has been accounted for by the photoreduction of Fe3+formed in Eqs.(5)and(6)to give Fe2+and an additional OH•via photo-Fenton reaction[10]:Fe3++H2O+hν→Fe2++OH•+H+(7)Fe2+regeneration then propagates catalytic reaction (4)and larger amounts of oxidizing OH•can be formed by reaction(5).Since H2O2is generated in this system from reaction(1),more OH•is also pro-duced from the classical Fenton reaction between this species and Fe2+[11]:Fe2++H2O2→Fe3++OH•+OH−(8) and the resulting Fe3+can be photoreduced again by reaction(7).In addition,the mineralization process can be accelerated by photolysis of complexes of Fe3+with short organic diacid intermediates,such as oxalic acid[12].To gain a better knowledge on the behavior of the O3/Fe2+and O3/Fe2+/UV A systems for contaminant removal,we have undertaken a study on the degra-dation of aniline and4-chlorophenol,as models of biorefractory aromatics in wastewaters,by ozona-tion catalyzed with Fe2+and/or UV A light.Several works have shown the effective mineralization of 4-chlorophenol by chemical[13],photochemical[14], electrochemical[15]and O3/UV[16–22]treatments. Aniline degradation has also been previously stud-ied by advanced electrochemical oxidation methods [23],photocatalysis with TiO2[23]and combining photocatalysis and ozonation[6].The present paper reports the results obtained for the mineralization of 1.07mmol dm−3aniline and1.38mmol dm−3chlorophenol solutions at pH ca.3by applying the O3,O3/UV A,O3/Fe2+and O3/Fe2+/UV A systems under comparable conditions. An initial concentration1mmol dm−3of Fe2+has been chosen for the two last methods,since higher contents of this catalyst in solution do not enhance the degradation of organics due to the increase in rate of competitive reaction(6),such as found previously for 2,4-D mineralization[9].In each process considered, organic intermediates and inorganic ions released have been identified.A general reaction pathway for the degradation of every initial pollutant is proposed.2.Experimental2.1.ReagentsReagent grade aniline from Probus was doubly distilled at178–180◦C under atmospheric pressureR.Sauleda,E.Brillas/Applied Catalysis B:Environmental29(2001)135–145137before use.Reagent grade4-chlorophenol sup-plied by Merck was recrystallized in ethanol-water. p-Benzoquinone,nitrobenzene,maleic acid,oxalic acid and4-chloro-1,3-dihydroxybenzene were reagent grade from Merck and Fluka.4-Chloro-1,2-dihydrox-ybenzene was synthesized by chlorination of pyro-catechol with SO2Cl2at room temperature following a standard method described in the literature[24]. Analytical grade sulfuric acid was purchased from Merck.Ferrous sulfate heptahydrate,puriss.p.a.,was supplied by Fluka.Acidic solutions were prepared with water obtained from a Millipore Milli-Q system, with a conductivity lower than6×10−8 −1cm−1.2.2.Ozonation systemAll degradation experiments were performed with a thermostated cylindrical vessel of150ml capacity. Ozone was generated with an Erwin Sander300.5 ozonizer fed with pure O2as carrier gas at15l h−1.A production of0.5g O3h−1was determined by standard iodometry.The resulting O2+O3mixture was supplied to the solution through a stainless steel diffuser.A6W Philipsfluorescent black light blue tube was used to irradiate the sample with UV A light, which emitted in the range300–420nm with a max-imum centered at360nm.It was placed at the top of the open vessel at4cm over the sample and the intensity of the incident light inside the solution was 8.3×10−7einstein dm−3min−1,as detected with an uranyl actinometer active to photons in the range 250–500nm.Test solutions were prepared with1.07mmol dm−3 aniline or with 1.38mmol dm−34-chlorophenol. Samples of100ml of both solutions were acidified with0.25mol dm−3H2SO4to pH ca.3and fur-ther mineralized by the O3,O3/UV A,O3/Fe2+and O3/Fe2+/UV A systems at25◦C.For the two last treat-ments,1mmol dm−3FeSO4was previously added to each sample.2.3.Product analysisThe solution pH was measured with a Crison2000 pH-meter.Mineralization of pollutants was monitored for total organic carbon(TOC)removal,which was determined on a Shimadzu5050TOC analyzer by the non purgeable organic carbon(NPOC)method.The decay in substrate concentration and the evolu-tion of stable aromatic intermediates were followed by reverse-phase chromatography with a Waters600 HPLC liquid chromatographfitted with a Spherisorb ODS25␮m column(150mm×4.6mm(i.d.))at room temperature,and equipped with a Waters486spec-trophotometric detector at280nm.This instrument was connected to a PS-486DX microcomputer and controlled through a Waters Baseline program.20␮l aliquots were injected into the liquid chromatograph and their components were separated with an eluent composed of a60:40acetonitrile/phosphate buffer(pH 3.1)mixture at1.0ml min−1.The same HPLC chro-matographfitted with an Aminex HPX87H column (300mm×7.8mm(i.d.))at35◦C,in conjunction with the spectrophotometric detector selected at210nm, were used to analyze the presence of short-chain organic acids in treated solutions by ion-exclusion chromatography.In this case,20␮l aliquots were also injected into the liquid chromatograph,using a 4mmol dm−3sulfuric acid solution at0.6ml min−1 as mobile phase.Ammonium ion concentration was measured on a Technicon Autoanalyzer II system fol-lowing the standard colorimetric automated phenate method[25].Treated solutions were diluted several times for measuring NH4+concentration in a linear range from5to100␮mol dm−3.The presence of nitrate and nitrite ions in such samples was tested by ion chromatography using a Kontron600HPLC liquid chromatographfitted with a Waters IC-Pak anion column at30◦C and equipped with a Waters spectrophotometric detector.For this technique,a borate-gluconate buffer of pH8.5was employed as eluent and100␮l aliquots were injected into the liq-uid chromatograph.Chloride ion concentration was determined by potentiometric titration with AgNO3 using a Metrohm Titrino702SM automatic titrator.3.Results and discussion3.1.Aniline mineralization3.1.1.TOC removalSolutions of1.07mmol dm−3aniline were adjusted at pH3.3with0.25mol dm−3H2SO4to be further de-graded during2h as maximum.The TOC abatement for the O3,O3/UV A,O3/Fe2+and O3/Fe2+/UV A138R.Sauleda,E.Brillas /Applied Catalysis B:Environmental 29(2001)135–145Fig.1.TOC removal with time for the mineralization of a 1.07mmol dm −3aniline solution at pH 3.3under the following conditions:(a,᭛)O 3+1mmol dm −3Fe 2+;(b,᭺)O 3;(c,᭝)O 3+1mmol dm −3Fe 2++UV A;(d,ᮀ)O 3+UV A.Aniline concentration decay with time during the same treatments is given as insert.Ozone flow rate 0.5g h −1;temperature 25◦C.systems is depicted in Fig.1.From these data,the min-eralization rate until 15min is found of 59(Fig.1b),101(Fig.1d),132(Fig.1a)and 154(Fig.1c)ppm TOC h −1,respectively.These results indicate that the degradation ability of the O 3system at the beginning of the process is accelerated in the presence of catalytic Fe 2+,as expected if intermediates are more quickly mineralized by OH •produced from reaction (5)than by ozone.Since a greater initial TOC decay is found for the O 3/Fe 2+system than for the O 3/UV A one,it is apparent that OH •is more efficiently generated from reaction (5)than from reaction (3).The O 3/Fe 2+/UV A treatment leads to the highest initial mineralization rate because more oxidizing OH •is produced by the simultaneous action of reactions (3),(5),(7)and (8).The relative oxidation ability of the above processes varies at longer degradation times,although the solu-tion pH slightly decreases to a value ca.3.0at the end of all treatments.Thus,78%of organics are removed after 1h of direct ozonation (see Fig.1b),whereupon no significant mineralization occurs due to the forma-tion of stable products.Under UV A irradiation,the decomposition of such stable species is accelerated by reaction with OH •generated from reaction (3)and the solution TOC is already reduced by 91%after 1h of the O 3/UV A treatment (see Fig.1d).In contrast,a very different behavior is observed in the presence of Fe 2+.While at the first oxidation stages this catalyst causes an increase in degradation rate of contaminants,at times >30min the mineralization becomes very hard (see Fig.1a)because of the formation of stable com-plexes of Fe 3+with oxalate and other short organic diacids [12].At the end of this treatment,only a 80%of TOC is removed.When the solution is also illumi-nated with UV A light using the O 3/Fe 2+/UV A system,the process becomes more efficient,allowing a 91%of TOC removal after 1.5h (see Fig.1c).This find-ing can be explained by the photodecomposition and further mineralization of Fe 3+complexes produced as stable products [12].3.1.2.Aniline decay and product detectionThe kinetics for aniline degradation and the evo-lution of aromatic intermediates for the different processes was followed by reverse-phase chromatog-raphy.A fast decay of aniline by direct attack with O 3is expected due to its high second-order rate constant of 108dm 3mol −1[26].As can be seen in Fig.1,this compound disappears quickly from the medium in a time ranging between 10and 15min in all cases,but its kinetics does not follow a defined reaction order.These results are indicative of a similar rate for the reaction of aniline with both O 3and OH •under the experimental conditions tested,because this com-pound is not photodegraded under UV A illumination.R.Sauleda,E.Brillas /Applied Catalysis B:Environmental 29(2001)135–145139Several aromatic intermediates,such as p -benzo-quinone and nitrobenzene,were also detected and quantified using the same technique.Fig.2a shows that p -benzoquinone is only accumulated when 1mmol dm −3Fe 2+is added to the solution and high amounts of OH •are produced from reactions (5),(7)and/or (8)to react with aniline.Under these condi-tions,maximum concentrations of 0.036mmol dm −3at 8min and 0.13mmol dm −3at 5min for this inter-mediate can be observed in the presence and absence of UV A light,respectively.For the O 3/Fe 2+/UV A system,the degradation rate of p -benzoquinone is then faster,disappearing from the medium in 10min.In contrast,nitrobenzene is accumulated in solution for all methods tested,such as depicted in Fig.2b.This intermediate reaches a maximum concentration of ca.8␮mol dm −3,being more quickly destroyed in the presence of Fe 2+.Ion-exclusion chromatography allowed to detect the presence of small contents of oxalic acid and to fol-low the evolution of maleic acid,which is shown in Fig.2c.This product is slowly accumulated and de-stroyed during direct ozonation,where attains a maxi-mum concentration of 10␮mol dm −3.When the solu-tion is irradiated with UV A light,its degradation rate is accelerated disappearing in <15min.In the pres-ence of Fe 2+,however,maleic acid is more slowly de-graded and until 6␮mol dm −3can be accumulated in the medium.It is noteworthy that possible complexes of Fe 3+with this acid are not detected by the analyt-ical method.Table 1summarizes the concentrations of both ammonium and nitrate ions,as well as the corre-sponding percentage of released nitrogen,after 2h of treatment of the 1.07mmol dm −3aniline solution by the different methods.No nitrite ions were detected by ion chromatography.Ammonium ion is released dur-ing the oxidation of aniline to p -benzoquinone,while nitrate ion is generated when nitrobenzene degradesTable 1Concentration of ammonium and nitrate ions obtained after 2h of treatment of a 1.07mmol dm −3aniline solution at pH 3.3by O 3and catalyzed ozonation processes at 25◦C aO 3O 3/UV A O 3/Fe 2+O 3/Fe 2+/UV A [NH 4+]/mmol dm −30.43(40%)0.49(45%)0.74(69%)0.68(64%)[NO 3−]/mmol dm −30.14(13%)0.38(35%)0.07(7%)0.36(33%)aThe percentage of released nitrogen with respect to that of initial aniline is given in parenthesis.[23].As can be seen in Table 1,NH 4+is produced in larger extension (until 64–69%)in the presence of Fe 2+,just when more p -benzoquinone is formed,in agreement with results of Fig.2a.In contrast,the maximum accumulations of NO 3−(until 33–35%)are reached under UV A irradiation with and without Fe 2+,indicating that generation of nitrobenzene from aniline oxidation is photocatalyzed by UV A light.Results of Table 1also show the release of more percentage of nitrogen as NH 4+in all cases,although the total nitrogen contained in both NH 4+and NO 3−concen-trations is always lower than that initially present in aniline.This suggests nitrogen elimination by evapo-ration of volatile intermediates,such as nitrobenzene,which is promoted by the O 2+O 3flow through the solution.In addition,it is also possible the existence of stable nitrogen-containing compounds in the final degraded solutions,specially after direct ozonation where only a 53%of the initial nitrogen is trans-formed into inorganic ions.3.1.3.Reaction pathways for aniline degradation On the basis of intermediates detected in this work and previous results reported in the literature [12,23,27],a general reaction pathway for aniline mineralization at pH ca.3is proposed in Fig.3.The main oxidizing agents are O 3for direct ozona-tion and O 3and OH •for the other systems.Both species have similar degradation power on aniline under our experimental conditions (see Fig.1),al-though intermediates are more quickly mineralized by the stronger oxidizing OH •.This radical also fa-vors p -benzoquinone and NH 4+production.On the other hand,UV A irradiation enhances nitrobenzene formation and Fe 3+complexes decomposition.Taking into account that at pH ca.3the solution contains a small proportion of the neutral form of aniline (p K a =4.7),which is more easily oxidable than its protonated one,the initial reactant in Fig.3140R.Sauleda,E.Brillas/Applied Catalysis B:Environmental29(2001)135–145Fig.2.Evolution of the concentration of:(a)p-benzoquinone;(b)nitrobenzene;(c)maleic acid,detected as intermediates during the degradation of a1.07mmol dm−3aniline solution at pH3.3by the system:(᭺)O3;(ᮀ)O3/UV A;(᭝)O3/Fe2+/UV A;(᭛)O3/Fe2+.R.Sauleda,E.Brillas/Applied Catalysis B:Environmental29(2001)135–145141Fig.3.Proposed reaction pathway for aniline mineralization by O3 and ozonation catalyzed with Fe2+and/or UV A light at pH ca.3. is assumed to be the unprotonated aniline.The pro-cess is then initiated by the attack of O3and/or OH•on this form to yield either benzoquinonimine or nitrobenzene.p-Benzoquinone is subsequently pro-duced by hydrolytic decomposition of benzoquinon-imine[23,27],leading to the release of NH3,present as NH4+ion in the acidic medium.This way is en-hanced in the O3/Fe2+and O3/Fe2+/UV A systems when more OH•can react with the initial reactant. In contrast,nitrobenzene is generated from the se-lective attack of O3on the amino group of aniline, being photocatalyzed by UV A light in the O3/UV A and O3/Fe2+/UV A systems.Further degradation of p-benzoquinone,as well as of nitrobenzene with re-lease of NO3−,lead to the formation of maleic acid, which is mineralized to CO2in all cases.This process depends on the conditions used.Thus,for the O3and O3/UV A systems,maleic acid is directly mineralized via oxalic acid,being this path accelerated by reaction of such diacids and its intermediates with OH•formed by reaction(3).In the presence of Fe2+,however, stable Fe3+-oxalato complexes are produced[12], which can be further photodecomposed to CO2under the action of UV A light in the O3/Fe2+/UV A system.3.2.4-Chlorophenol mineralization3.2.1.TOC removal4-Chlorophenol was mineralized under compa-rable conditions to those used for aniline degrada-tion.The sample initially contained1.38mmol dm−3 4-chlorophenol(equivalent to100ppm of TOC),be-ing adjusted to pH3.3with0.25mol dm−3H2SO4. In all these trials,a slow acidification of the solution occurs,always reaching afinal pH value close to3. Fig.4shows the TOC decay determined under oxidation by the O3(curve b),O3/UV A(curve d), O3/Fe2+(curve a)and O3/Fe2+/UV A(curve c)sys-tems.The mineralization rates until15min for these processes are68,124,144and152ppm TOC h−1, respectively,values very close to those reported pre-viously for aniline degradation.This trend confirms again that OH•produced in the different catalyzed ozonations has higher oxidizing power than O3,ac-celerating the initial mineralization of intermediates.A more efficient OH•generation is then attained in the presence of Fe2+from reaction(5)than under UV A illumination from reaction(3),although it is en-hanced by reactions(7)and(8)when both catalysts, Fe2+and UV A light,are combined.At longer times than30min,the decreases in TOC for the different oxidative processes of4-chlorophenol in Fig.4are similar to those of aniline in Fig.1.While 75%of organics are destroyed after2h of treatment with O3(see Fig.4b),the resulting stable products are quickly removed by reaction with OH•under UV A ir-radiation,thus,allowing to reduce the solution TOC by 98%in1h(see Fig.4d).The formation of stable Fe3+ complexes causes a fast inhibition in TOC decay for the O3/Fe2+system,where only69%mineralization is possible during2h(see Fig.4a).The photodegra-dation of such complexes in the O3/Fe2+/UV A sys-tem accelerates again the TOC removal and ca.95% of pollutants can be destroyed during thefirst hour of this treatment(see Fig.4c).It is noteworthy that aniline and4-chlorophenol undergo similar TOC abatements(<90%)after1h of the corresponding O3/UV A and O3/Fe2+/UV A142R.Sauleda,E.Brillas /Applied Catalysis B:Environmental 29(2001)135–145Fig.4.TOC abatement with time for a 1.38mmol dm −34-chlorophenol solution at pH 3.3when it is degraded with:(a,᭺)O 3+1mmol dm −3Fe 2+;(b,᭛)O 3;(c,᭝)O 3+1mmol dm −3Fe 2++UV A;(d,ᮀ)O 3+UV A.4-Chlorophenol concentration decay with time during the same treatments is shown in insert.Ozone flow rate 0.5g h −1;temperature 25◦C.treatments,as can be deduced from Figs.1and 4,respectively.This differs from results previously re-ported for 2,4-D,where only the O 3/Fe 2+/UV A sys-tem leads to its fast and complete mineralization [9].Then,the Fe 3+complexes derived from aniline and 4-chlorophenol oxidation are more slowly photode-composed by the O 3/Fe 2+/UV A system,while the O 3/UV A treatment has sufficient oxidizing power to mineralize the pollutant at a similar rate.3.2.2.4-Chlorophenol decay and products time-courseThe 4-chlorophenol concentration decay for all pro-cesses was followed by reverse-phase chromatography and the results obtained are also presented in Fig.4.A similar degradation rate can be observed in all cases,disappearing from the solution in 25–30min.How-ever,its kinetics does not fit with any defined reac-tion order,probably because under our experimental conditions,no steady concentrations for the oxidiz-ing agents O 3and/or OH •are attained.This finding indicates that the increase in TOC abatement for the catalyzed ozonations is due to the faster degradation rate of its intermediates by the action of OH •and/or UV A light.The fact that 4-chlorophenol is more slowly eliminated than aniline under comparable conditions (see Figs.1and 4)can be associated with the lower value of the second-order rate constant of the first pol-lutant with O 3(3.4×106dm 3mol −1s −1[26])than that of the second one (108dm 3mol −1s −1[26]).Two dihydroxylated derivatives,4-chloro-1,3-dihy-droxybenzene and 4-chloro-1,2-dihydroxybenzene,have been identified as primary products of 4-chloro-phenol by reverse-phase chromatography.These compounds are generated by meta -and ortho -hydrox-ylation of the initial pollutant,respectively.Note that 4-chloro-1,2-dihydroxybenzene is the unique interme-diate found when 4-chlorophenol is only oxidized with OH •[14,15],and hence,chloro-1,3-dihydroxybenzene is expected to be produced by its reaction with O 3.The evolution of 4-chloro-1,2-dihydroxybenzene is depicted in Fig.5a.As can be seen,the presence of Fe 2+yields more accumulation of this product.Its maximum concentration reached in solution is 22and 17␮mol dm −3for the O 3/Fe 2+/UV A and O 3/Fe 2+systems,whereas a maximum value of 10␮mol dm −3is attained without Fe 2+.The combined action of Fe 2+and UV A light then enhances the generation of 4-chloro-1,2-dihydroxybenzene.In contrast,Fig.5b shows that 4-chloro-1,3-dihydroxybenzene is accumu-lated in larger extension in the absence of Fe 2+,with maximum concentrations of 0.13and 0.09mmol dm −3for the O 3and O 3/UV A systems,respectively.In the O 3/Fe 2+treatment,a maximum accumulation of only 0.04mmol dm −3can be observed for this product,being undetected for the O 3/Fe 2+/UV A system.The144R.Sauleda,E.Brillas /Applied Catalysis B:Environmental 29(2001)135–145opposite effect shown in Fig.5a and b on the action of Fe 2+confirms that 4-chloro-1,2-dihydroxybenzene formation is favored when 4-chlorophenol can be selectively ortho -hydroxylated with OH •.Ion-exclusion chromatography of treated solutions also allowed the detection of oxalic acid in small concentrations and the quantification of free maleic acid,i.e.without forming complexes with Fe 3+.As can be seen in Fig.5c,the mineralization of this last acid is accelerated by UV A illumination,disappear-ing after 12min without Fe 2+and after 16min in the presence of this catalyst.when the solution is not ir-radiated,maleic acid is slowly accumulated in the O 3/Fe 2+system,probably because its generation from 4-chloro-1,2-dihydroxybenzene is more difficult than from 4-chloro-1,3-dihydroxybenzene.The further degradation of the above dihydroxylated derivatives yields chloride ion,which remains stableinFig.6.General reaction pathway proposed for 4-chlorophenol mineralization by O 3and ozonation catalyzed with Fe 2+and/or UV A light at pH ca.3.the medium.Fig.5d shows that Cl −is completely re-leased (up to 1.38mmol dm −3)in all processes tested.The dechlorination process is faster for the O 3/UV A and O 3/Fe 2+/UV A treatments,in agreement with the faster mineralization rate found under UV A irradiation in Fig.4.3.2.3.Reaction pathways for 4-chlorophenol degradationThe proposed reaction pathway for 4-chlorophenol mineralization at pH ca.3in Fig.6involves its ini-tial reaction with O 3to give 4-chloro-1,3-dihydroxy-benzene and 4-chloro-1,2-dihydroxybenzene.The formation of the second compound is enhanced in the catalyzed ozonations by the selective attack of OH •on the ortho -position of 4-chlorophenol.This path is more favorable for the O 3/Fe 2+/UV A system due to the production of higher amounts of this radical fromR.Sauleda,E.Brillas/Applied Catalysis B:Environmental29(2001)135–145145reactions(3),(5),(7)and(8).Further degradation of both dihydroxylated derivatives with O3and/or OH•leads to maleic acid with loss of Cl−.For the O3and O3/UV A systems,this acid is mineralized to CO2via oxalic acid,although in the presence of UV A light,their destruction is strongly accelerated by OH•radicals produced from reaction(3).For the O3/Fe2+system,however,maleic acid yields stable Fe3+-oxalato complexes[12],which can be quickly photodecomposed to CO2under UV A irradiation in the O3/Fe2+/UV A system.4.ConclusionsThe mineralization of aniline and4-chlorophenol at pH ca.3by O3and ozonation catalyzed with Fe2+ and/or UV A has been examined.Stable products are formed when they are directly oxidized with O3,al-though these species are quickly destroyed under UV A illumination.The presence of Fe2+as catalyst im-proves the initial reaction efficiency,but the degrada-tion processes are inhibited by the formation of Fe3+ complexes with intermediates,which are photode-composed by UV A light.Aniline and4-chlorophenol are then quickly mineralized by the O3/UV A and O3/Fe2+/UV A systems,a different result from previ-ously found for2,4-D,where only the O3/Fe2+/UV A system leads to a fast and complete mineralization [9].p-Benzoquinone,nitrobenzene,maleic acid and oxalic acid are identified as intermediates during ani-line mineralization,and ammonium and nitrate ions are found in thefinal solutions.The production of benzoquinone and ammonium ion is enhanced in the presence of Fe2+,whereas more nitrate ion is accumu-lated in the solution under UV A irradiation,indicating that formation of nitrobenzene from the selective at-tack of O3on aniline is photocatalyzed.The direct reaction of ozone with4-chlorophenol mainly gives 4-chloro-1,3-dihydroxybenzene as primary product. The formation of4-chloro-1,2-dihydroxybenzene is enhanced in the catalyzed ozonations by the selec-tive attack of hydroxyl radical on the ortho-position of4-chlorophenol.Further degradation of such di-hydroxylated intermediates leads to maleic acid with complete release of chloride ion in all processes.For the O3and O3/UV A systems,this acid is mineralized to CO2via oxalic acid,whereas for the O3/Fe2+sys-tem,it yields stable Fe3+-oxalato complexes,which are photodecomposed to CO2under UV A irradiation in the O3/Fe2+/UV A system.References[1]G.R.Peyton,in:N.M.Ram,R.F.Christan,K.P.Canto(Eds.),Significance and Treatment of V olatile Organic Compounds in Water Supplies,Lewis Pu.,1990,pp.313–362.[2]C.Adams,P.A.Scanlan,N.D.Secrist,Environ.Sci.Technol.28(1994)1812.[3]T.J.Conocchioli,E.J.Hamilton,N.J.Sutin,J.Am.Chem.Soc.87(1965)926.[4]T.L␾gager,J.Holeman,K.Sehested,T.Pedersen,Inorg.Chem.31(1992)3523.[5]D.L.Sedlack,J.Hoigné,M.M.David,R.N.Colvile, E.Seyffer,K.Acker,W.Wiepercht,L.Wolfgang,J.A.Lind,S.Fuzzi,Atmos.Environ.31(1997)2515.[6]L.Sánchez,J.Peral,X.Domènech,Appl.Catal.B:Environ.19(1998)59.[7]L.Rivas,I.R.Bellodono,F.Ascari,Chemosphere37(1998)1033.[8]S.Hay,T.N.Obee,J.Adv.Oxid.Technol.4(1999)209.[9]E.Piera,J.C.Calpe,E.Brillas,X.Domènech,J.Peral,Appl.Catal.B:Environ.27(2000)169.[10]B.C.Faust,J.Hoigné,Atmos.Environ.24(1990)79.[11]Y.Sun,J.J.Pignatello,Environ.Sci.Technol.27(1993)304.[12]Y.Zuo,J.Hoigné,Environ.Sci.Technol.26(1992)1014.[13]O.Koyama,Y.Kamagata,K.Nakamura,Water Res.28(1994)895.[14]E.Lipcynska-Kochany,R.James,Environ.Sci.Technol.26(1992)259.[15]E.Brillas,R.Sauleda,J.Casado,J.Electrochem.Soc.145(1998)759.[16]S.Esplugas,P.L.Yue,M.I.Pervez,Water Res.28(1994)1323.[17]M.A.Boncz,H.Bruning,W.H.Rulkens,E.J.R.Sudhoelter,G.H.Harmsen,J.W.Bijsterbosch,Water Sci.Technol.35(1997)65.[18]K.Abe,K.Tanaka,Chemosphere37(1997)2837.[19]M.Trapido,A.Hirvonen,Y.Veressinina,J.Hentunen,R.Munter,Ozone:Sci.Eng.19(1997)75.[20]M.Hautaniemi,J.Kallas,R.Munter,M.Trapido,Ozone:Sci.Eng.20(1998)259.[21]R.Andreozzi,R.Marotta,J.Hazard Mater.69(1999)303.[22]S.W.Kuo,Chemosphere39(1999)1853.[23]E.Brillas, E.Mur,R.Sauleda,L.Sánchez,J.Peral,X.Domènech,J.Casado,Appl.Catal.B:Environ.16(1998)31.[24]J.Frejka, B.Sefránek,J.Zika,Collect.Czech.Chem.Commun.9(1937)238.[25]Methods for Chemical Analysis of Water and Wastes,method350.1,US Environmental Protection Agency,Cincinnati, 1983.[26]J.Hoigné,H.Bader,Water Res.17(1983)185.[27]R.L.Hand,R.F.Nelson,J.Electrochem.Soc.125(1978)1059.。