Liquid-phase microextraction–gas chromatography–mass spec(4)

148K.Reddy-Nooneetal./J.Chromatogr.A1148(2007)

145–151

Fig.1.Thederivatizationreactionsusedforiodideandbromide.

2,6-Dimethylanilinewasconsideredasahalogenscavengerofchoicesince,contrarytootherreagents[32–34],itrespondedforbothiodinationandbrominationreaction,andthatonlyoneisomerofeachhaloanilinewasformed.ThehaloanilineswereamenableforSDME/LPMEduringpreconcentrationandcleanupstepofanalysis.

Determinationofbromateandiodateisbasedontheirreductiontohalidesbyascorbicacid.Beforereduction,anyfreebromideandiodidepresentinthesamplewasquantitativelyremovedbyanionexchangewithAgCl,areactionthatisbasedondifferencesinsolubilityproduct(Ksp)ofsilversalts,beingAgCl,1.8×10 10,AgBr,5.0×10 13,AgI,8.3×10 17,AgBrO3,5.5×10 5,andAgIO3,3.1×10 8[39].BothAgBrandAgIhaveKsplowerthanthatofAgClandthusbromideandiodideareprecipitatedastheirsilversaltsontreatmentwithAgCl.BromateandiodateremaininsolutionastheirsilversaltshaveKsphigherthanthatofAgCl.InsolutionsofhalidestreatedwithAgCl,atequilibriumAgL=Ag+the[Ag++L ]=√,1.8×10 10=1.34×10 5(AgL=AgCl);[Br ]=√ 13=7.07×10 7(AgL=AgBr);and[I ]=√5.0×108.3×10 17=9.0×10 9(AgL=AgI)(allvaluescal-culatedwithoutinvolvingcouplingofequilbria).Thus,[Ag+][Br ]=9.47×10 12and[Ag+][I ]=1.21×10 13,bothvaluesarehigherthantheKspofAgBrandAgI,respectively,andunderthisconditionbromideandiodidearecompletelyprecipitatedupontreatmentwithAgCl.Whenasolutionof10mgl 1eachofbromideandiodidewastreatedwithAgClandthesupernatantsolutionwassubjectedtohalidesdeter-minationbytheproposedmethod,nopeakforhalo-organiccompoundswasobtained(Fig.2).At[Ag+]=1.34×10 5(fromAgCl),precipitationofAgBrO3required[BrO3 ]=K 5spAgBrO3/[Ag+]=5.5×10 5/1.34×10=4.10moll 1 ;

andprecipitationofAgIO3required[IO3]=K+spAgIO3/[Ag]=3.1×10 8/1.34×10 5=2.31×10 3moll 1.BothconditionsareunrealisticwithpracticalsamplesandthusthereisnopossibilityofAgBrO3orAgIO3precipitationontreatmentwithAgCl.Thiswascon rmedbyexperimentsinwhichaliquotsofstandardscontaining10mgl 1eachofbromateandiodateweretreatedwithAgCl,andoxyhalidesweredeterminedinthesupernatantsolutionbythepresentmethod;therecoveryofbromateandiodatewas

99.6–100.5%.

Fig.2.GC–MSof10mgl 1eachofbromideandiodideafterderivatizationto4-bromo-2,6-dimethylanilineand4-iodo-2,6-dimethylaniline(uppertrace),andaftertreatmentwithAgClandtheirdeterminationinthesupernatantsolution(lowertrace).LPMEin50 loftoluene.IS,theinternalstandard2,4,6-tribromoaniline.Inset,magni edbaselineoflowertrace.

3.2.Extractionofhaloanilines

SDME[40,41]usestypically1–3 loforganicsolventandcombinessamplingandpreconcentrationinthesamestep,andsincethewholevolumeofextractedanalyte(s)istransferredtothechromatograph,thesensitivityattainedissimilartosolid-phasemicroextraction(SPME).Asalwaysafreshdropofsolventisusedforextractionthereisnochanceofanalytecarry-overinSDME,whichisacommonlimitationofSPME.ThereareseveralparameterswhichcontroltheoptimumperformanceofSDMEandincludethenatureofsolvent,dropsizeofsolvent,extractiontime,stirring,ionicstrengthofsolution,etc.Theseparameterswereseparatelyevaluatedtooptimizetheextraction.4-Iodo-2,6-dimethylaniline,4-bromo-2,6-dimethylanilineandtheinternalstandard2,4,6-tribromoanilineexistmostlyinunprotonatedforminthe nalsolutionofpH7.5–8.SDMEofhaloanilinesformedinthedeterminationof1mgl 1eachofiodideandbromidewascarriedoutwith1 lofhexane,cyclohexane,isooctaneandtoluene,theoptimumextractionwasobtainedwithtolueneina15minequilibrationtime(Fig.3).Theextractionalsoincreasedwithincreasingdropvolumeinrange1–4 l;a2 ldropoftoluenewasusedinsubsequentexperimentssinceitgavesuf cientsensitivityandwasstableduringextraction.Theextractionreachedoptimumvaluein15min(rangestudied1–25min).Highsaltconcentrationaffectstheextractionef ciencyandforthisitisnecessarytodilutethesamples.Thisbehaviourhasbeenreportedearlier[42,43].Saltinwaterperhapschangedthenatureof

Nernst

Fig.3.Effectofsolvent,1 ldrop,ontheSDMEof1mgl 1ofiodideandbromideafterderivatizationto4-halo-2,6-dimethylaniline.Equilibrationtime,15min.