Cobalt phthalocyanine as a biomimetic catalyst in the amperometric quantification of dipyrone using FIA

ContentslistsavailableatScienceDirect

Talanta

jo u r n al h om ep a g e :w w w . e l s e v i e r . c o m / l o c a t e / t a l a n t a

Cobalt

phthalocyanine

as

a

biomimetic

catalyst

in

the

amperometric

quantification

of

dipyrone

using

FIA

Ana

Carolina

Boni

_,

_Ademar

_Wong

_,

_Rosa

_Amália

_Fireman

_Dutra

_,

Maria

Del

Pilar

Taboada

Sotomayor

a_{DepartmentofAnalyticalChemistry,InstituteofChemistry,StateUniversityofSãoPaulo(UNESP),14801-970Araraquara,SP,Brazil} b_{LaboratoryofDiagnosticResearch(LAPED),UniversityofPernambuco,50100-130Recife,PE,Brazil}

a

r

t

i

c

l

e

i

n

f

o

Articlehistory: Received6April2011

Receivedinrevisedform6July2011 Accepted9July2011

Available online 20 July 2011

Keywords: Biomimeticcatalyst Cobaltphthalocyanine Dipyrone

Flowinjectionsystem

a

b

s

t

r

a

c

t

Abiomimeticsensorforthedeterminationofdipyronewaspreparedbymodifyingcarbonpastewith cobaltphthalocyanine(CoPc),andusedasanamperometricdetectorinaflowinjectionanalysis(FIA) sys-tem.TheresultsofcyclicvoltammetryexperimentssuggestedthatCoPcbehavedasabiomimeticcatalyst intheelectrocatalyticoxidationofdipyrone,whichinvolvedthetransferofoneelectron.Theoptimized FIAprocedureemployedaflowrateof1.5mLmin−1_,a75

␮Lsampleloop,a0.1molL−1phosphatebuffer

carriersolutionatpH7.0andamperometricdetectionatapotentialof0.3Vvs.Ag/AgCl.Underthese conditions,theproposedmethodshowedalinearresponsefordipyroneconcentrationsintherange 5.0×10−6_–6.3×10−3_molL−1_{.Selectivityandinterferencestudieswerecarriedoutinordertovalidate}

thesystemforusewithpharmaceuticalandenvironmentalsamples.Inadditiontobeing environmen-tallyfriendly,theproposedmethodisasensitiveandselectiveanalyticaltoolforthedeterminationof dipyrone.

© 2011 Elsevier B.V. All rights reserved.

1. Introduction

Greaterpublicawarenesshasresultedinheightenedinterest intheenvironmentalimpactscausedbythepresenceofindustrial wastesanddomesticsewageinriversandotheraquaticsystems. Dischargesofuntreatedindustrialeffluentscontaminatenatural waters, which can render them unsuitable for use, cause dis-easeand even destroy entireecosystems [1–3]. Pharmaceutical drugsexcretedbyhumansandlivestockcanreachwaterbodies indischargesfromsourcesincludingineffectivesewagetreatment plants.

Manymethodsareavailableforthemeasurementof environ-mentalpollutants.Mostoftheseinvolvecollectionofthematerial tobeanalyzedanditsretrievaltoalaboratoryinordertoconduct therequiredanalyses.Degradationofthesamplecanoccurduring transport,affectingtheanalyticalresults.Itisthereforedesirableto developmethodologiesthatenablemeasurementstobeperformed

insitu,and thatprovide rapidand reliablereal-timeresponses. Portableflowinjectionanalysis(FIA)systemsemploying electro-chemicalsensors[4,5]canbedevelopedforthispurpose,andoffer

∗_{Correspondingauthorat:IQ/CAr–UNESP,RuaProf.FranciscoDegniS/N,} Qui-tandinha,Araraquara14800-900,Brazil.Tel.:+551633019620;

fax:+551633019692.

E-mailaddress:mpilar@iq.unesp.br(M.D.P.T.Sotomayor).

advantages includingexcellent sensitivity aswellas speed and reproducibility.

Biomimeticsensors[6,7]areaclassofelectrochemical detec-tors thatare preparedbythe modificationofelectrodes [8–12] withmetal complexes thathave similar chemical structuresas the activesites of oxide-reductase enzymes.Such sensors pos-sesscharacteristicsthatmakethemmoredurableandstablethan theiranaloguesconstructedusingenzymes,includinghigh stabil-ityovertimeaswellaswithdrasticchangesofpHandtemperature (whichcoulddenatureenzymes).Theyarecheap,simpleto manu-factureandmayevenbedisposable(asinthecaseofscreen-printed electrodes).Biomimeticsensorsaremoresensitive,comparedto enzymaticbiosensors,duetoasmallerdiffusionbarrierandgreater electrontransferbetweenthebiomimeticcomplexandthe ana-lyte[6,7].Thisnewandincreasinglypopularclassofsensorshas beenusedfortheidentificationandquantificationofvariousdrugs [13–17],andhasalsobeencoupledtoflowsystems[18,19].

Amongtheoxide-reductaseenzymesthathavebeenfrequently imitated are theP450, which have a common activesite, pro-toporphyrin IX (orprotoheminIX, aniron porphyrin) [20].The reasontomimictheP450enzymes,ofwhichapproximately450 differenttypesarecurrentlyknown[21],is thattheseenzymes catalyzeawiderangeofchemicalreactionsinorganisms, produc-ingmetabolitesthatarephysiologicallyessentialorbeneficialto them.Inaddition,hydroxylation,oxidationandreductionreactions candegradexenobioticsincludingdrugs,pesticidesandendocrine disruptorssuchassynthetichormonesandsunscreens[20,21].

N

N

N

N

N

Co

N

N

Fig.1.Chemicalstructureofcobaltphthalocyanine.

Table1

Typesofelectrodesmodifiedwithcobaltphthalocyaninethathavebeenreported forthedeterminationofrelevantanalytes.

Typeofelectrode Analyte Carbonpaste Hydrazine[24]

Carbonpaste Guanine[25]

Glassycarbon Dopamine[26]

Screen-printed Thiocholine[27]

Screen-printed Citricacid[28]

Pyrolyticgraphite Glutathione[29]

Boron-dopeddiamond Hydrogenperoxide[30]

AlthoughalloftheP450enzymescontainacommonactivesite, namelytheironporphyrincomplex,theselectivityofeachP450isa consequenceofthechemicalenvironmentsurroundingthehemin group[20].Asaresult,compoundsderivedfromphthalocyanines andporphyrinsofiron[13,18,19,22,23],orothermetalssuchas manganese[14],nickel[15]andcopper[17],havebeen success-fullyusedintheconstructionofbiomimeticchemicalsensorsfora varietyofanalyticalpurposes.

Sensorsbasedonelectrodesmodifiedwithcobalt phthalocya-nine(Fig.1)havealsobeenreportedintheliteraturefordifferent analytesandapplications,usingelectrodematerialssuchascarbon paste[24,25]orboron-dopeddiamond[30](Table1).Thesedevices havebeenshowntobechemicallyandmechanicallystable, how-evernotallofthemeitherpresentelectrocatalyticbehavior[26] orgiveresponsesatveryhighpotentials[25].These characteris-ticsareessentialinbiomimeticsensorsinordertoachieveahigh degreeofsensitivityandselectivity.

Dipyrone(Fig.2)isadrugthatiswidelyused,notonlyinBrazil butalsoinothercountriesincludingFrance,Germany,Hungary, Israel,Spain,Swedenand a numberof developingnations. It is stilloneofthemostpopularandpowerfulanalgesics[31]. How-ever,theuseofdipyronewasabolishedmore thanthirty years agointheUnitedStatesofAmerica,duetoitscontroversialeffect [32]onbonemarrowfunction.Itsusagehasbeenassociatedwith aplasticanemiaandagranulocytosis[32–34].Despiteitspossible

O

H

O

O

N

N

N

S

CH

CH

C

H

O

Na

Fig.2. Chemicalstructureofdipyrone.

ronments.Itisthereforeevidentthattherecontinuestobeaneed foranalyticalmethodsabletoquantifydipyroneinvariousdifferent matrices.

This work describes the development of a sensor prepared withcobalt (II)phthalocyanine [CoPc], used as a P450enzyme biomimeticcatalyst in thesensitiveand selective detectionsof dipyrone.Theobjectivewastodevelopananalyticaltoolthatwas faster,cheaperandmorereliablethanexistingtechniques.

2. Experimental

2.1. Chemicalsandsolutions

All the chemicals used were analytical or HPLC grade. Cobalt (II) phthalocyanine [CoPc], dipyrone, mineral oil and graphite powder were purchased from Sigma–Aldrich. NaH2PO4 and NaOH were purchased from Synth-Brazil. TRIS [tris(hydroxymethyl)aminomethane], HEPES [4-(2-hydroxyethyl)piperazine-1-ethanesulfonicacid]andPIPES[saltof sodium 1,4-piperazinediethanesulfonate] buffers were obtained fromMerck.Methanolandphosphoricacidwerepurchasedfrom Ecibra-Brazil. Thedipyrone and buffersolutions wereprepared withwaterpurified usingaMilli-Q(Direct-0.3)system,andpH wasmeasuredusingaThermoScientificpHmeter(Orion3Star Benchtop)fittedwithaglasselectrode.

2.2. Biomimeticsensorconstruction

Themodifiedpastewaspreparedbyhomogenizing12.5mgof CoPcwith87.5mgofgraphitepowderand1.0mLof0.1molL−1 phosphatebuffersolution(atpH7.0).Thematerialobtainedwas driedatroomtemperature,and thenmixedwithmineraloilto obtainahomogenouspaste.Thepastewasplacedintothecavityof aglasstube(4mminternaldiameter,1mmdepth),andaplatinum slidewasinsertedforelectricalcontactwiththepaste.The influ-enceoftheamountofCoPcemployedinpastepreparationonthe responseofthesensorwasevaluatedusingthreedifferentpastes, preparedwithweightpercentagesofCoPcvaryingfrom7.5%to 20.0%.

2.3. Electrochemicalmeasurementsandflowmanifold

Thecyclicvoltammetryexperimentswereconductedatroom temperatureinaconventionalthree-electrodecell,withthe mod-ified carbon pasteelectrode used as theworkingelectrode. An Ag/AgCl(KClsat)electrodeanda platinumwirewereusedasthe referenceandcounterelectrodes,respectively.Themeasurements wereperformedusingaPalmSensepotentiostat(PalmInstruments BV,TheNetherlands),whichwasinterfacedwithamicrocomputer runningPSLite(v.1.7.3)software(PalmInstrumentsBV)forcontrol ofpotentialandacquisitionofdata.

85 (2011) 2067–2073 2069

WE

AE

RE KClsat

Outlet

Carrier

Sample

Waste

L

Pump

L

Fig.3.Schematicdiagramoftheflowinjectionsystemwithamperometricdetection.L:sampleloop;WE:workingelectrode(biomimeticsensor);AE:auxiliaryelectrode (platinum);RE:homemadereferenceelectrode(Ag/AgCl(KClsat)).

2.4. HPLCanalyses

ChromatographicanalyseswereperformedusingaShimadzu Model20Aliquidchromatograph,coupledtoanSPD-20AUV/Vis detector,a SIL-20Aautosamplerand a DGU-20A5 degasser.The chromatographysystemwascontrolledbyamicrocomputer.AC8 column(250mm×4.6mm,Shim-PackCLCODS)waspositioned insideaShimadzuCTO-10ASoveninordertomaintainaconstant temperature.

Theprocedureemployedwasinaccordancewiththereference methodforthequantificationofdipyrone[36].Themobilephase wascomposedofamixtureof280mLofmethanoland720mLof 0.1molL−1_{NaH2PO4,atpH7.0.Theflowratewas1.0mLmin}−1_and thesampleinjectionvolumewas10␮L.Thedetectionwavelength was254nm.

2.5. Applicationofthebiomimeticsensorusingthe pharmaceuticalformulations

Dipyrone was quantified in four commercial samples using the external standards procedure. Samples of two commercial solutions(1and 2inTable5)wereanalyzedafterdilutionwith deionizedwater,withoutanyadditionaltreatment.Twosamples oftablets(3and4inTable5)werepreparedforanalysisbygrinding andhomogenizinginamortar,afterwhichaportionofthe mate-rialwasweighedoutanddissolvedindeionizedwater.Thesolution wasfilteredinordertoremoveanyinsolublesubstancespresentin thetablets.

2.6. Applicationofthebiomimeticsensorforanalysisofaquatic samples

FivewatersamplesfromriversinSãoPauloState(Brazil)were enrichedwithdipyroneattwoconcentrationlevels(9.76×10−5

and 3.00×10−4_molL−1_{)and then analyzedusing theproposed} sensorinordertoevaluatetheinfluenceofthematrixonrecoveries. TheenrichedsampleswerealsoanalyzedusingtheHPLCreference method[36],inordertodeterminethenominalconcentrationsof dipyroneinthesamples.Alloftheriverwatersampleswerefiltered priortoenrichmentinordertoremoveanyinsolublesubstances present.

3. Resultsanddiscussion

3.1. Electrochemicalandbiomimeticcharacteristicsofthesensor

Cyclicvoltammetryexperiments(usingascanfrom0to450mV

vs.Ag/AgCl,atascanrateof20mVs−1_{)werecarriedoutinorder} toevaluatetheeffectof[CoPc]ontheresponseofthesensorto dipyrone.Theresults(Fig.4)showedthatthepeakcorresponding totheoxidationofcobalt(II)inthecomplexdisappearedinthe presenceofdipyrone,atapotentialof250mVvs.Ag/AgCl(Fig.4, curvea).Atthesametime,ahighcurrentintensityanodicpeakwas obtained,withouttheappearanceofanycathodicpeak,showing thatdipyronewasirreversiblyoxidizedonthesensorsurfaceata potentialof340mVvs.Ag/AgCl(Fig.4,curveb).Theresponseprofile ofthesensorwasasexpectedforanenzymaticcatalysisprocessin whichanincreaseofcurrentoccursinonlyonedirection(either oxidationorreduction)[23].

Experiments were performed using different scan rates in ordertodeterminewhethertheoxidation ofdipyronewasdue toanelectrocatalyticprocess.Theresultsobtainedareshownin Figs.5–7.

Incyclicvoltammetry,itisknownthatwhenthereaction kinet-icsarecontrolledbydiffusionofthespeciesfromthebulksolution totheelectrode surface,thepeak currentisproportionaltothe squarerootofthescanrate,inagreementwiththeRandles–Sevcík equation[37,38].Cyclicvoltammogramswerethereforeobtained

Table2

ParametersevaluatedduringtheoptimizationofthebiomimeticsensorfordipyronedeterminationusingtheFIAsystem.

Parameter Value

Amountofthecomplexinthepaste(%,w/w) 7.5 12.5a _{20.0 – – –}

Buffer(electrolyte) PHOSPHATEa _{HEPES PIPES TRIS – –}

BufferpH 6.0 6.5 7.0a _{7.5 8.0 8.5}

Appliedpotential(mVvs.Ag/AgCl) 200 250 300a _{350 400 –}

0.0 0.1 0.2 0.3 0.4 0.5 -1

0 1 2 3

i (µA)

b

a

Potential vs Ag/AgCl (V)

250 mV

Fig.4. Catalyticprofilefordipyroneoxidationontheproposedsensorbasedon carbonpastemodifiedwithcobaltphthalocyanine:(A)0.1molL−1_{phosphatebuffer}

solution(pH7.0);(B)buffersolutioncontaining1.4× 10−3_molL−1_ofdipyrone.

atdifferentscanrates(Fig.5)inordertoinvestigatewhetherthe dipyroneoxidationprocesswascontrolledbymasstransport.The graphshowninFig.5insetwasplottedusingthese voltammo-grams.Thestraightlinefitindicatedthattheprocesswascontrolled bydiffusion.

Theexistenceofanelectrocatalyticprocesswasconfirmedby plottingthescanrate-normalizedcurrent(i·−1/2_{)againstthescan} rate().Theformofthecurveobtained(Fig.6)indicatedthatthe oxidationofdipyroneoccurredbyelectrocatalysis,andinvolved anelectrochemicalprocesscoupledwithachemicalstep[39].In addition,alinearrelationshipwasobtainedbetweentheanodic peakpotential(Ep)andlog(Fig.7),indicativeofaproportional electrontransferflowintheelectrode.

Using the information obtained from the linear regression (shownintheinsetofFig.5),togetherwiththetheoreticmodel describedbyAndrieuxandSavèant(Eq.(1))[40],itwaspossible tocalculateadiffusioncoefficient(D0)of5.9×10−7cm2s−1,

cor-Fig.5.Cyclicvoltammogramsrecordedatdifferentscanrates:(a)0.010Vs−1_;(b)

0.020Vs−1_;(c)0.050Vs−1_;(d)0.100Vs−1_{and(e)0.200Vs}−1_{.Theinsetshowsthe}

linearprofileofthevariationofthecurrent(i)asafunctionofthesquarerootof thescanrate(1/2_{).Measurementswerecarriedoutin0.1molL}−1_{phosphatebuffer}

solution(atpH7.0)containing5.0× 10−4_molL−1_ofdipyrone.

Fig.6.Atypicalcatalyticelectrooxidationprofile.A5.0×10−4_molL−1_solutionof

dipyronewaselectrooxidizedbythecobalt(II)compleximmobilizedonthecarbon paste.Measurementswerecarriedoutin0.1molL−1_{phosphatebuffersolution(at}

pH7.0),usedastheelectrolyte.

respondingtotheaveragediffusionofdipyroneanditsoxidation productsduringthecatalyticprocess.

ip=0.496FACSD10/2

RT

1/2 (1)

In Eq. (1), CS is the dipyrone concentration (5.0×10−7_molcm−3_{), D}

0 is the diffusion coefficient of dipy-rone,FistheFaradayconstant,R=8.31447Jmol−1_K−1_,T=298K andA=0.126cm2_{.Thedipyronediffusioncoefficientcould} there-forebeobtainedfromtheslope(×10−6_{)ofthelineshowninFig.5} insetandthefactor0.496FACSD01/2F1/2(RT)

−1/2 .

Assumingthattheelectrocatalyticoxidationofdipyroneisan irreversibleprocess,theRandles–Sevcíkrelationship(Eq.(2))can beusedtodeterminethenumberofelectronsinvolved.

jp=(2.99×105)n[(1−˛)na]1/2D1₀/2CS1/2 (2)

InEq.(2),jp=ip/A=114.6␮Acm−2V−1/2s1/2,nisthetotal num-berofelectronsinvolved in thereaction,˛is thecoefficientof electrontransfer,narepresentsthenumberofelectronsinvolvedin thedeterminingstep,D0(cm2s−1)isthediffusioncoefficientofthe

Fig.7.Linearvariationoftheoxidationpotential(Ep)vs.log.Scanswerecarried

85 (2011) 2067–2073 2071

Fig.8. Lineweaver–Burkplotforcatalyzeddipyroneelectrooxidationonthe CoPc-basedsensor.Thefirst25pointsareshowninthefigureinset.Measurementswere carriedoutin0.1molL−1_{phosphatebuffersolution(atpH7.0),applyingapotential}

of340mVvs.Ag/AgCl.

electroactivespecies,CS(molcm−3)isthedipyroneconcentration and

Inordertoestimatethevalueofn,itisfirstnecessaryto cal-culatethevalueof(1−_{˛)na.Incatalyticorirreversiblesystems,} an approximate calculation is based on the linear relationship betweentheanodicpeakpotential(Ep)andlog

0.03536= 1.15RT

[(1−˛)na]F (3)

Aftercalculating thevalueof(1−˛)na,usingEq.(2)andthe previouslyestimateddipyronediffusioncoefficient,itwaspossible tocalculatethenumberofelectronsinvolvedinthecatalytic elec-trooxidationofdipyroneonthesurfaceoftheCoPc-basedsensor. Thevalueobtainedwas1.09,whichwascloseto1×e−_.

In order to explore whether the behavior of the proposed sensorwassimilartothatofenzymatic biosensors,itsresponse was monitored until the concentration reached the saturation region.The results obtainedshowed a hyperbolic profile, indi-cating that the cobalt compound in the carbon paste followed theMichaelis–Mentenkineticmodel.TheLineweaver–Burkgraph (Fig.8)enabledcalculationoftheapparentMichaelis–Menten con-stant(KMMapp).Avalueof4.4×10−4molL−1indicatedstrongaffinity betweenthecobalt complexand thedipyrone substrate.These resultsalsostronglysuggestthatthedetectorpossessedthe char-acteristicsofabiomimeticsensor.

Basedonthefindingsdescribedabove,itwaspossibleto pro-poseaplausiblemechanismforthesensorresponse(Fig.9).Inthis mechanism,Co3+_{iselectrochemicallygeneratedontheelectrode} surface(electrochemicalstep),andisthespeciesresponsiblefor theoxidationofdipyrone(chemicalstep).Afterthereductionof theCo3+_{ion,thecurrentrequiredforitsre-oxidationwillbe} pro-portionaltotheconcentrationoftheanalyteinthestandardsor samples.

3.2. CouplingofthebiomimeticsensortotheFIAsystem

Followingconfirmation,usingbatchmodevoltammetry,that thesensorpossessedgenuinebiomimeticcharacteristics,itwas coupledtoaflowinjectionanalysissystem.Optimizationofthe sys-teminvolvedconsiderationofthefollowingparameters:(i)amount

Fig.9. Proposedmechanismforthesensorresponse,basedonexperimental evi-dence.Dip:dipyrone;red:reduceddipyrone;ox:oxidizeddipyrone.

Table3

Selectivityofthebiomimeticsensor.Theslopesoftheanalyticalcurveswereused tocalculatetheresponseforeachcompound,relativetotheresponseobtainedfor dipyrone.

Drug Relativeresponse(%) Dipyrone 100.0

Captopril 14.5

4-AAP 12.5

Ascorbicacid 11.0 Paracetamol 3.2

ofcomplexinthepaste;(ii)appliedamperometricpotential;(iii) composition,concentrationandpHofthecarrierbuffersolution (electrolyte).TheseparametersaresummarizedinTable2,with theselectedvaluesindicatedinboldtype.

Flow rate optimization (Fig. 10) employed a range of 0.7–2.1mLmin−1_{.Aflowrateof1.5mLmin}−1_{wasselectedsince} thisprovidedthebestcompromisebetweenanalysisfrequencyand sensitivity.Theinjectionvolumewasinvestigatedusingloop vol-umesof25,50,75and100␮L(Fig.8).Thesignalincreasedwith

Fig.10.InfluenceoftheparametersintheproposedFIAsystem:(a)effectofinjected samplevolume(Vi)ontheresponsetodipyrone,usingaflowrateof1.5mLmin−1;

(b)effectofflowrateontheresponsetodipyrone,usingaViof50␮L.The experi-mentswerecarriedoutwithacarriersolutionof0.1molL−1_{phosphatebufferatpH}

7.0,andapplyingapotentialof300mVvs.Ag/AgCl(KClsat).Theerrorbarscorrespond

Fig.11.FIAsignals obtainedforthebiomimetic sensorunderoptimizedflow conditionsfordifferentdipyroneconcentrations.[Dipyrone]:a=5.0× 10−5_molL−1_;

b=9.5 × 10−4_molL−1_{; c=3.9× 10}−3_molL−1_{, d=1.5× 10}−3_molL−1_;

e=3.0 × 10−3_molL−1_{; f=5.0 × 10}−2_molL−1_{; g=6.25× 10}−3_molL−1_{. Left inset:}

Thethreelowerdipyroneconcentrations.Rightinset:Typicalanalyticalcurvesfor increasinganddecreasingdipyroneconcentrations.

increasingVi,reachingamaximumat100␮L.However,avolume of75␮Lwasselectedsincethisprovidedgoodsensitivitywithout undulyaffectingthefrequencyofsampling.

3.3. AnalyticalcharacteristicsoftheFIAsystem

Acalibrationcurve(Fig.11)wasgeneratedaftertheFIAsystem parametershadbeenoptimized.Nomemoryeffectswereobserved, andlinearfitswereobtainedforbothincreasinganddecreasing concentrationsofdipyrone(Eqs.(4)and(5),respectively).

i/␮A=−0.09(±0.07)+16062(±647)[Dipyrone]/mol L−1

(R=0.9960,n=7) (4)

i/␮A=_1.5(±_0.1)+₁₅₈₇₆₍±_{407)[Dipyrone]/molL}−1

(R=0.9987,n=7) (5)

Thelimitsofdetectionandquantificationwere1.5×10−5_and 4.5×₁₀−5_molL−1_{,respectively,calculatedaccordingtoIUPAC} rec-ommendations[41]fromthestandarddeviationoftenseparate measurementsoftheblank.

Therepeatabilityofthesensor wasinvestigated usingseven consecutiveinjectionsof2.5×10−3_molL−1_{dipyronesolution,for} whicharelativestandard deviation(RSD)of0.5%wasobtained. Undertheconditionsemployed,thesensorcouldbeoperatedatan analyticalfrequencyof36injectionsperhour.Thebetween-sensor reproducibilitywas2.5%,calculated astheRSDoftheanalytical curvesforsixseparatesensors.Itshouldbepointedoutthat excel-lentreproducibilityintheconstructionofthesensorswasachieved duetotheeaseofpreparationofthemodifiedcarbonpaste.

3.4. Selectivityandinterferencestudies

Goodselectivityisoneofthemostdesirablecharacteristicsof analyticaltechniques.Hence,inadditiontodipyrone,theresponse ofthesensorwastestedusingsixteenotherdrugs:acetylsalicylic acid,4-aminoantipyrine(4-AAP),amitriptyline,ascorbicacid, cap-topril,chloramphenicol,ciprofloxacin,clarithromycin,diclofenac, ketoprofen,naproxen,paracetamol,piroxicam,prednisone,

raniti-Interferent Relativeresponse(%)

Caffeine 102

Cellulose 102

Lactose 100

Magnesiumstearate 99 Stearicacid 102 Sodiumbisulfite 103 Sodiumphosphate 101

dineandtenoxicam.Table3liststhosedrugstowhichthesensor showeda response. The response to4-AAP was12.5%, relative totheresponse to dipyrone,but this wasexpected because 4-aminoantipyrineisoneoftheactivemetabolitesofdipyroneand hasasimilarchemicalstructure.Unlikeothersensorsbasedon car-bonpaste,thepresentsensorgaveasmallsignalforascorbicacid (11.0%).Theresultsindicatedthattheproposedsensorwashighly selective,asexpectedforadevicedesignedtomimicanenzymatic biosensor.

Theinterference studywasperformedconsidering themain applicationofthesensortobeinanalysesofcommercial phar-maceuticalformulations.Table4showstherelativeresponses(%) obtainedinthepresenceofeachinterferingsubstance(atan inter-ferent/analytemolarratioof10:1),comparedtotheresponsefor dipyronealone.Theresultsindicatedthatthesensordidnot suf-ferfrominterferences due tothepresenceof thesesubstances. Matrixinterferencesarethereforenotexpectedtooccurinanalyses involvingthisclassofsubstance.

3.5. Sensorapplication

Theapplicationoftheproposedsensorwastestedusingtwo differenttypesofsample.Thefirstinvolvedfourcommercial phar-maceutical formulations. The second employed water samples collectedfromfiveriversinSãoPauloState,whichwereenriched withdipyroneattwoconcentrationlevelsinordertodemonstrate theapplicabilityoftheproposedFIAsystemtoenvironmental sam-ples.

Table5showstheresultsobtainedforthecommercial formu-lations.TheconcentrationsweredeterminedusingtheFIAsystem, andthencompared withtheresultsobtainedbythe chromato-graphicreferencemethod.Therewerenosignificantdifferences betweenthetwotechniques(ataconfidencelevelof95%).

UsingtheFIAtechnique,theaverageerrorfortheriverwater analyseswas1.2%,relativetotheHPLCreferencemethod(Table6). Theresultsindicate that dipyronewasalwayspresent inthese rivers,sincethemeasuredconcentrationsexceededthoseexpected fromknowledgeoftheamountsofdipyroneaddedtothewater samples.

Table5

Determinationofdipyroneinpharmaceuticalformulationsusingthebiomimetic sensorFIAtechniqueandthechromatographicreferencemethod.

Sample Dipyroneamount Error(%) Declaredvalue Reference

method(HPLC)

Proposedmethod

1 500a _{438.0 435.0± 16.5}c _0.7

2 50a _{52.0 49.5± 0.7}c _3.8

3 300b _{296.4 297.0± 8.6}c _0.3

4 500b _{520.0 535.0}± 5.6c _2.9

a_mgL−1_.

b_mgtablet−1_.

85 (2011) 2067–2073 2073

Table6

Comparisonbetweenthevaluesobtainedfordipyroneinwaterfromrivers,using theproposedFIAsystemandthereferencemethod.Thesampleswereenrichedat twoconcentrationlevels(9.76× 10−5_{and3.00× 10}−4_molL−1_).

River Measuredconcentration(molL−1_{) Error(%)}

HPLCmethod ProposedFIAmethoda

TietêriverattheUsina daBarraclub

9.8× 10−5 _9.9× 10−5± 2.6× 10−6 _1.0

3.4× 10−4 _3.4× 10−4± 1.4× 10−5 _0.0

JacaréGuac¸ú 9.9× 10−5 _1.0× 10−4± 4.2× 10−6 _1.0

3.4× 10−4 _{3.2× 10}−4_{± 1.1× 10}−5 _5.9

Jaú 1.2× 10−4 _{1.2× 10}−4_{± 3.6× 10}−6 _0.0

3.0× 10−4 _{3.0× 10}−4_{± 7.2× 10}−6 _0.0

JacaréPepira 1.0×10−4 _9.9×10−5_{± 3.5× 10}−6 _1.0

3.3×10−4 _3.2×10−4± 1.2× 10−5 _3.0

Tietêriverbythebridge inBarraBonitacity

1.0× 10−4 _1.0× 10−4± 3.5× 10−6 _0.0

3.6× 10−4 _3.6× 10−4± 1.1× 10−5 _0.0

a_{Standarddeviationfor3replicates.}

4. Conclusions

Thispaperpresentsanalternativeandenvironmentallyfriendly methodologyforthedeterminationofdipyrone,usingaflow injec-tionsystemwithamperometricdetectionbasedonabiomimetic sensormodified withtheCoPccomplex.ThenewFIAsystemis fast,sensitiveandinexpensive,andcanbeusedfortheanalysisof dipyroneinmediaincludingpharmaceuticaldrugformulationsand riverwatersamples.

Acknowledgements

Theauthorsgratefullyacknowledgethefinancialsupport pro-vided by the São Paulo State Research Foundation (FAPESP, processes2008/00303-5and2009/04598-2).

References

[1]T.A.Ternes,M.Stumpf,J.Mueller,K.Haberer,R.D.Wilken,M.Servos,Sci.Total Environ.225(1999)81–90.

[2]M.E.Balmer,H.R.Buser,M.D.Muller,T.Poiger,Environ.Sci.Technol.39(2005) 953–962.

[3]M.Schlumpf,B.Cotton,M.Conscience,V.Haller,B.Steinmann,W. Lichten-steiger,Environ.HealthPerspect.109(2001)239–244.

[4]J. Ruzicka,E.H. Hansen,FlowInjectionAnalysis, JohnWiley, NewYork, 1988.

[5]B.F.Reis,M.F.Giné,E.A.M.Kronka,Quim.Nova12(1989)82–91. [6]M.D.P.T.Sotomayor,L.T.Kubota,Quim.Nova25(2002)123–128.

[7]M.D.P.T.Sotomayor,A.A.Tanaka,R.S.Freire,L.T.Kubota,C.A.Grimes,in:e.c. Dickey,M.V.Pishko(Eds.),EncyclopediaofSensors,AmericanScientific Pub-lishers,California,2006,pp.195–209.

[8]P.R.Moses,L.Wier,R.W.Murray,Anal.Chem.47(1975)1882–1886. [9]J.A.Cox,M.E.Tess,T.E.Cummings,Rev.Anal.Chem.15(1996)173–223. [10]M.A.T.Gilmartin,J.P.Hart,Analyst120(1995)1029–1045.

[11]K.Kalcher,J.M.Kauffmann,J.Wang,I.Svancara,K.Vytras,C.Neuhold,Z.Yang, Electroanalysis7(1995)5–22.

[12]J.Wang,Electroanalysis3(1991)255–259.

[13] I.V.Batista,M.R.V.Lanza,I.L.T.Dias,S.M.C.N.Tanaka,A.A.Tanaka,M.D.P.T. Sotomayor,Analyst133(2008)1692–1699.

[14] W.J.R.Santos,A.L.Sousa,F.S.Damos,S.M.C.N.Tanaka,L.T.Kubota,A.A.Tanaka, M.D.P.T.Sotomayor,J.Braz.Chem.Soc.20(2009)1180–1187.

[15] L.F.Moreira,M.R.V.Lanza,A.A.Tanaka,M.D.P.T.Sotomayor,Analyst134(2009) 1453–1461.

[16] A.C.Boni,M.D.P.T.Sotomayor,M.R.V.Lanza,S.M.C.N.Tanaka,A.A.Tanaka,J. Braz.Chem.Soc.21(2010)1377–1383.

[17] D.F.Gobatto,A.Wong,M.R.V.Lanza,M.D.P.T.Sotomayor,OpenChem.Biomed. MethodsJ.3(2010)98–107.

[18] M.C.Q.Oliveira,M.R.V.Lanza,A.A.Tanaka,M.D.P.T.Sotomayor,Anal.Methods 2(2010)507–512.

[19]A.Wong,M.R.V.Lanza,M.D.P.T.Sotomayor,Comb.Chem.HighThroughput Screen.13(2010)666–674.

[20]M. Sono, M.P. Roach, E.D. Coulter, J.H. Dawson, Chem. Rev. 96 (1996) 2841–2888.

[21]A.S.Kalgutkar,R.S.Obach,T.S.Maurer,Curr.DrugMetab.8(2007)407–447. [22]F.S.Damos,M.D.P.T.Sotomayor,L.T.Kubota,S.M.C.N.Tanaka,A.A.Tanaka,

Ana-lyst128(2003)255–259.

[23]A.Sigoli,M.D.P.T.Sotomayor,M.R.V.Lanza,A.A.Tanaka,L.T.Kubota,J.Braz. Chem.Soc19(2008)734–743.

[24]C.D.C.Conceic¸ão,R.C.Faria,O.Fatibello-Filho,A.A.Tanaka,Anal.Lett.41(2008) 1010–1021.

[25]I.Balan,I.G.David,V.David,A.-I.Stoica,C.Mihailciuc,I.Stamatin,A.A.Ciucu,J. Electroanal.Chem.654(2011)8–12.

[26]F.C.Moraes,M.F.Cabral,S.A.S.Machado,L.H.Mascaro,Electroanalysis20(2008) 851–857.

[27]E.Jubete,K.Zelechowsk,O.A.Loaiza,P.J.Lamas,E.Ochoteco,K.D.Farmer,K.P. Roberts,J.F.Biernat,Electrochim.Acta56(2011)3988–3995.

[28]K.C. Honeychurch, L. Gilbert, J.P. Hart,Anal. Bioanal. Chem. 396 (2010) 3103–3111.

[29]N.Sehlotho,S.Griveau,T.Nyokong,F. Bedioui,Electroanalysis19(2007) 103–106.

[30]T.Kondo,A.Tamura,T.Kawai,J.Electrochem.Soc.156(2009)145–150. [31] F.A.Fehintola,Afr.J.Med.Sci.34(2005)403–405.

[32] P.Danieli,M.B.Leal,Rev.Bras.Farm.84(2003)17–20. [33]I.M.Bense ˜nor,Rev.Paul.Med.119(2001)190–191. [34]V.J.Dorr,J.Cook.South.Med.J.89(1996)612–614.

[35] F.H.Bergamin,J.X.Medeiros,B.F.Reis,E.A.G.Zaggatto,Anal.Chim.Acta101 (1978)9–16.

[36]BritishPharmacopoeia,MentalHealthAct1,2007. [37]J.E.B.Randles,Trans.FaradaySoc.44(1948)327–338. [38] A.Sevcík,Collect.Czech.Chem.Commun.13(1948)349–377.

[39] A.J.Bard,L.Faulkner,ElectrochemicalMethods:Fundamentalsand Applica-tions,2nded.,Wiley,NewYork,2001.