SynthesisofLargeDendrimerswiththeDimensionsofSmallVirusesJongdooLim,Â†MauriKostiainen,Â‡JanMaly,Â§VivianaC.P.daCosta,Â†OnofrioAnnunziata,Â†GiovanniM.Pavan,andEricE.Simanek *, Â†Â†DepartmentofChemistry,TexasChristianUniversity,FortWorth,Texas76129,UnitedStatesÂ‡DepartmentofAppliedPhysics,AaltoUniversity,Helsinki,FinlandÂ§DepartmentofBiology,J.E.Purkyne University,U stÃnadLabem,CzechRepublicDepartmentofInnovativeTechnologies,UniversityofAppliedScienceofSouthernSwitzerland,Manno6928,Switzerland*SSupportingInformationABSTRACT: Thedendrimerchemistryreportedo ersa routetosynthetictargetmoleculeswithsphericalshape, well-de nedsurfacechemistries,anddimensionsthat matchthesizeofvirusparticles.Thelargesttarget,a generation-13dendrimercomprisingtriazineslinkedby diamines,isstableacrossrangesofconcentration,pH, temperature,solventpolarityandinthepresenceof additives.Thisdendrimertheoreticallypresents16384 surfacegroupsandhasamolecularweightexceeding8.4 MDa.Transmissionelectronandatomicforcemicroscopies,dynamiclightscattering,andcomputationsreveala diameterof 30nm.Thetargetwassynthesizedthrough aniterativedivergentapproachusingamonochlorotriazine macromonomerprovidingtwogenerationsofgrowthper syntheticcycle.Fidelityinthesynthesisissupportedby evidencefromNMRspectroscopy,massspectrometry,and high-pressureliquidchromatography.Viruses withdiametersrangingfrom20to400nm occupyalengthscalethatrepresentstheheartofallthatis nano .1Thelowerboundsofthislengthscaleareaconvergent pointforbothtop-downandbottom-upsynthesis.Existing organicbuildingblocksprovidingaccesstothissizeregime derivefromalldisciplines,includingchemistry,molecular biology,andvirology,andincludemicelles,liposomes,andviral capsids,respectively.Thesematerialsrelyonself-assembly.Selfassemblyreducestheburdenofsynthesis,buttypicallyatthe costofstability,asre ectedinsensitivitytoconcentration,pH, temperature,solvents,andadditivessuchasdenaturantsand detergents.Covalentsynthesiso ersanalternativetoselfassemblythatprovidessigni cantopportunitiesforcompositionalcontrol.Virus-sized,covalentorganicmoleculeshave remainedelusivetargets. FirstdescribedbyTomalia,2Newkome,3andVo gtle4inthe late1970sandearly1980s,dendrimersarehighlybranched, globularmacromoleculesthatpresentmultiplesurfacegroups ontheperiphery.Thesearchitectureshavebeenthefocusof experimentalandtheoreticalinvestigationsinareasincluding materialsscienceandmedicinalchemistry.5Whileamyriadof platformshavebeendescribed,ahandfulhavebeenthefocusof intensescrutinybecauseoftheireaseofsynthesisand/or commercialavailability.Thesematerialsincludethepolylysines introducedbyDenkewalter,6thepoly(propyleneimine)(PPI) dendrimersofVo gtle4andMeijer,7thepolyamidoamine (PAMAM)dendrimersofTomalia,2thepolyarylethersof Fre chet,8thepolyestersadvancedbyGilliesandFre chet,9and thephosphorus-containingdendrimersofMajoraland Caminade.10Still,thesynthesisoflarge-generationdendrimers israre.ReportsoflargedendrimersincludePAMAM11and phosphorus-containingdendrimers,10,12bothofwhichare commerciallyavailableuptogeneration10.Phosphorus-based dendrimershavelongoccupiedthebenchmarkforlarge molecules,butatgeneration12theyencounterasolubility challengethatprecludesadditionalpursuits.12bAtthese generations,however,boththisplatformandPAMAM dendrimersreachdiametersofonly14 ÂŠ 15nm.Assuch, thesematerialspopulatethelengthscaleofproteinsbutfall shortofthelengthscalesofviruses. Formanyyears,wehavebeeninterestedintriazine dendrimers.13Synthetically,thesematerialsarederivedfrom cyanuricchlorideanddiaminesthatcanbeelaboratedinto dendrimercoresandmonomers.Asaplatform,thisclassof dendrimerso ersanumberofbene ts,including(i)lowcost ofreagents,(ii)compositionaldiversitystemmingfromthe wealthofavailablediaminesandthestepwisesubstitutionofthe triazinenucleus,(iii)easeoflarge-scalesynthesis,(iv)stability inhighlyacidic(pH0)orbasic(pH14)solutions,and(v)long shelflife.Thesematerialscanbepreparedusingeither convergentordivergentapproaches,althoughtheonlylatter providesaccesstolargemolecules. Toreachthetargetgeneration-13dendrimer, G13 ,the synthesisstartedwithageneration-1aminecore, G1 ,thatwas reactedwithmacromonomer M (Figure1).Reactionof G1 with M providestheprotected G3 dendrimer.Upon deprotection,theprocessisrepeatediteratively.Bydesign, macromonomer M hadasinglereactivesite,amonochlorotriazine,whichprecludestheformationofcovalentdimers, therebylimitingsideproducts.Previously,inorderto synthesizelargedendrimers,weemployedtwodi erent diaminesaslinkers:piperazineandanoligoetherdiamine.13cThissyntheticapproachwassuccessfuluptogeneration9,but poorsolubilitywasencounteredatgeneration11.Theformer diamine,piperazine,providesamorereactive,rigidlinking groupwithouthydrogen-bonddonors.Thelattersacri ces Received: January14,2013 Published: February12,2013 Communication pubs.acs.org/JACS Â©2013AmericanChemicalSociety4660dx.doi.org/10.1021/ja400432e | J.Am.Chem.Soc. 2013,135,4660 ÂŠ 4663
reactivityfor exibilityandlengthandintroduceshydrogenbonddonorsandacceptors.Here,onlythelatter exible diaminewasadopted.Dendrimerscomprisingthisdiamine showedexcellentsolubility(>100mg/mL)atallgenerations explored.Theprotecteddendrimersarereadilysolublein organicsolvents.Theamine-terminateddendrimersshowgood solubilityinbothwaterandorganicsolvents. Target G13 isthelargestdendrimerreportedtodatewith13 branchingpoints(generations)betweenthecoreandthe periphery(Figure1).Thistargettheoreticallypresents16384 aminegroupsonthesurfaceandhasamolecularweight(MW) exceeding8.4MDa.However,becauseofthesizeof G13 ,the extenttowhichtheactualtarget(s)ofthissynthesisrepresent theidealizedstructureisunknown.Evidenceforsuccessis derivedfrommultipletechniques,includingthebehaviorof smaller-generationintermediatesaswellasthecharacterization ofthe nalproduct.Thatis,theiterativesynthesisrevealed trendsthatmatchedexpectationswithregardtoboth di erencesinpolarityoftheprotectedanddeprotected materialsandtheappearanceanddisappearanceofcharacteristiclinesderivedfromtheperipheralgroupsinthe1HNMR spectra(Figure2a).Speci cally,removingthe tert -butoxycarbonyl(BOC)protectinggroupledtoachangeinthechemical shiftofthevicinal ÂŠ CH2ÂŠ groupfrom3.2to2.75ppm.While thespectrasuggestacleanreaction,con denceinthis interpretationiscompromisedbythelimitsofdetection a ordedby1HNMRspectroscopy.Inviewofthelargenumber ofprotonsinthe nalproduct(737240intheory),1HNMR spectroscopycannotprovideevidenceof complete successin conversionatthe nalproductstageorevenforintermediates beyond G3 becauseofinherentlimitsinthesignal-to-noise ratio. High-pressureliquidchromatography(HPLC)showedthat larger-generationmaterialselutelaterthansmaller-generation ones(Figure2b).Onthebasisoftheamountofreadily discernedimpuritiesinthetraces,wecanassignthepurityof the G13 targetandothersmaller-generationintermediatesto be 95%.Thispuritylabelre ectspurityinsizeandnotatomic compositionofthetheoreticalstructuredepicted.Weinferthat theseimpuritiesarelower-MWspeciesthatmayarisefrom incompletedeprotectionorincompletereactionsof M withthe deprotecteddendrimer. Themostcompellingevidenceforsuccessinthepreparation ofvirus-sizedparticleswasprovidedbycryogenictransmission electronmicroscopy(cryo-TEM).Figure3showscryo-TEM imagesof G13 andcowpeachloroticmottlevirus(CCMV),an icosahedralRNAplantvirus.X-raycrystallographyandTEM revealedthatthevirushasadiameterof 28nm.The microscopyimagesshowsimilarsizesandshapesforthevirus and G13 .Bothappearsphericalwhenimagedeitherasdried samplesstainedtoprovidehighcontrastorasvitri edaqueous solutions.Becausethestainingisimperfect,arangeofsizes ratherthanasinglevalueisobtained.Ifthestaindoesnot penetratethedendrimersigni cantly(thestandardassumption),thediameterof G13 is24.4 Â± 2.3nm.Ifthedyelayeris includedwiththebeliefthatpartialpenetrationoccursbecause ofthenatureoftheperiphery,thedimensionincreasesto31.9 Â± 1.1nm.Unstainedsamplesderivedfromvitri edsolutions providedadiameterof25.2 Â± 2.3nm,althoughthismay slightlyunderrepresentthesizebecausetheimagecontrastat theedgesofthe G13 particleislow. Atomicforcemicroscopy(AFM)corroboratedthemeasurementsofsizeandshapeandsuggestedahydration-statedependentsize.Figure4showstheAFManalysisof G13 on atomically athydrophilicmicainwater.Theaverageheightof Figure1. Structuresofthe G1 dendrimerwithanalkynecoreandthe monomer M thatisreactedtoinstalltwoadditionalgenerationsper iteration. Â“ Cmpd Â” liststheintermediatedendrimercompounds accessedinrouteto G13 .Thenumberofendgroups( Â“ Ends Â” ),the theoretical( Â“ MWtheoÂ” )andobserved( Â“ MWobsÂ” )molecularweights(in Da,asdeterminedbyESI ÂŠ TOFMS),andthediameters(innm) obtainedbyDLS( Â“ dDLSÂ” )andTEM( Â“ dTEMÂ” )areindicated.ThedTEMdataincludethestaininthemeasurementandprovideanupperlimit onthesize. Figure2. (a)1HNMRspectraofthelarge-generationdendrimers ( G9 ÂŠ G13 )inthe ngerprintregionformonitoringtheiterative additionof M anddeprotection.Thevicinalprotonsignalsofthe NHBocgroupsappearat3.2ppm,whilethevicinalprotonsignalsof NH2groupsappearat2.75ppm.(b)HPLCtracesfor G3 ÂŠ G13 . Figure3. (a)TEMimageof G13 (sampledriedandnegativelystained withuranylacetate).(b)Close-upviewofthe G13 dendrimerand comparisontoCCMV:(left)drysamplespreparedbynegative staining;(middle)colored3Dintensitypro les;(right)samples derivedfromvitri edaqueoussolutions.Imagesizes:100nm Ã— 100 nm. JournaloftheAmericanChemicalSociety Communicationdx.doi.org/10.1021/ja400432e | J.Am.Chem.Soc. 2013,135,4660 ÂŠ 46634661
G13 ( Zmax),whichcorrespondstothedendrimerdiameter assumingtheidealsphericalshape,was31.5 Â± 1.9nm.AFM analysisofdrysamplessuggestedthatthedendrimerscollapse signi cantlyinthe Z dimension[ Zmax=9.8 Â± 1.9nm;seethe SupportingInformation(SI)].Thiscollapsesuggestsahigh degreeofhydrationoftheinternalvoidspaceofdendrimer,or atleastahighersensitivitytotheforceattheAFMtipwhenthe dendrimersaredehydrated.Theresultsaresimilartothose observedbyHaagwithhydrogelsthatcollapsedfrom 100to 20nmondrying.14TheAFMsizedistributionalsoprovides insightsintothedispersityofthesamples.Thehistogramin Figure4cshowsevidenceforthepresenceoftwosmaller materials( 20and 25nmdiameter)alongwith G13 . Representingalmost35%ofthesample,theseentitiesappearto bediscretepopulationsofmacromoleculesandmaycorrespond tomaterialswithsizessimilartothoseof G7 and G9 / G11 ,as suggestedbytheHPLCtraces.Theirappearanceisconsistent withfailuresofpuri cationusingconventionalchromatographicmethodsandincompletereactionsthroughoutthe synthesis.Clearly,roomforimprovementexists,butthishas notproventrivialtodate.Filtrationduringsamplepreparation andaggregationof G13 appearedtoenrichthepopulationsof smallerparticles.Accordingly,theamountofsmallerparticlesis likelytobeclosertotheHPLC-basedestimatesof 5%rather thantheAFM-basedestimatesof 35%.TheHPLCtraces suggestthattheseimpuritiesareelaboratedthroughoutthe synthesis,perhapsasaresultoffaileddeprotectionor macromonomeradditionearlyintheprocess.Theirexistence, however,suggestsopportunitiesfor nercontroloverthesize ofparticlesbymanipulatingthevalencyofthedendrimercore. Dynamiclightscattering (DLS)measurementswere performedondendrimersinaqueoussaltsolutionsatneutral pH.Withtheexceptionof G13 ,resultswereobtainedin phosphate-bu eredsaline(PBS)atpH7.0.Aggregationof G13 inPBSrequiredtheuseof0.01MNaCl(aq)atpH7.Withthe exceptionof G3 ,thedistributionoflight-scatteringintensities wasfoundtobebimodal.Thefast-di usionmodewasrelated tothedendrimersizeusingtheStokes ÂŠ Einsteinequation.The extractedhydrodynamicdiameters( dDLS)increasedwith dendrimergenerationingoingfrom G3 to G13 (Figure1). Theslow-di usionmodecanberelatedtothepresenceoflarge dendrimeraggregateswithdiametersof50 ÂŠ 150nm.As hydrationcontributesto dDLS,thevaluesareconsistentwith therangesprovidedbyTEMandAFM.DLSmeasurements performedon G13 intheabsenceofNaClyieldeddendrimer di usioncoe cientsthatwere 10%higher.Thise ectcanbe relatedtothedendrimerchargeandtheabsenceofelectrostatic screening.15Computationalmodelsprovideadditionalstructuralinsights andanchoredourintuitivepictureofthesemolecules.Afully atomisticsimulationinexplicitsolventwasnotpossiblefor G13 ,asinclusionofexplicitmoleculesofwaterwouldhave exceededthecomputationalinfrastructurenecessaryforthis1.3 millionatommolecule( G13 theoreticallyhastheformula C376812H737240N114682O98298).Moreover,theuseofcoarsegrainedsimpli edmodelsfor G13 wasincompatiblewiththe linearandextremely exiblecharacterofthemonomers constitutingthedendriticsca old.Theconstructionofthe G13 atomisticmodelwasextremelychallengingbecauseofthe highstructuralcomplexityemergingfromthelargenumberof monomerconnections.Moleculardynamics(MD)simulations revealedthat G13 reachesa Â“ hardsphere Â” limit(Figure5).The peripheralaminegroupsextendnotonlyoutwardbutalso inwardbyback-foldingintothedendrimer,resultinginhigh structuraldensity(seetheradialdistributionfunctionplotsin theSI).Themeasuredradiusofgyrationofthismoleculeis Rg=11.0 ÂŠ 11.4nm.Thisvalueisingoodagreementwiththe diametersmeasuredusingDLS,as Rgandtheradiusof hydration( Rh)arerelatedbytheequation Rh 1.29 Rgfor sphericalmoleculesofuniformdensity.16Accordingly,these diametersof28.6 ÂŠ 29.4nmmatchtheestimatesderivedfrom TEMandAFM. Thesuccessingeneratingvirus-sizedmaterialsobservedhere andnotyetseeninPAMAMorphosphorusplatformsmay derivefrommorethanjustthesolubilityofthelinkingdiamine. Clearly,thegenerationunitofthetriazinedendrimersis approximatelytwiceaslongasthosefortheotherplatforms: whilePAMAMrelieson ÂŠ N ( R)CH2CH2C(O)NHCH2CH2ÂŠ withsevenatomsandphosphorusdendrimersrelyon ÂŠ P ( R)(S)O ÂŠ ( p -C6H4) ÂŠ C NNH ÂŠ withnineatoms,the triazineplatformutilizes18atomspergeneration.Intermsof thenumberof exibleatomspermonomer( P ),deGennes predictedthemaximumsizeofadendrimerwithperfect branchingbeforedefectsarenecessitatedbystericcrowdingof theperiphery(eq1):17 =+ P l imiting generation2.88(ln1.5) (1)With P =18,themaximumgenerationachievedbeforedefects is12.6. Figure4. (a)AFMimageof G13 inwateronamicasurface.Inset: close-upview(230nm Ã— 230nm)ofthecircled G13 dendrimer.(b) Cross-sectionpro lesoftwodendrimers(averageofresultsforthree scanlines).(c)Histogramshowingthepresenceoftwo subpopulationsofsmallerparticlesinadditionto G13 . Figure5. (a)Computationalmodelof G13 derivedfromMD simulations.Terminalaminegroupsareshowninblue.(b)CCMV particlecoloredoliveandbluetoillustratetheicosahedralstructure. JournaloftheAmericanChemicalSociety Communicationdx.doi.org/10.1021/ja400432e | J.Am.Chem.Soc. 2013,135,4660 ÂŠ 46634662
Thenatureofthebranchingunitmaybelessrelevant.While branchinginbothPAMAMandphosphorusdendrimers (indicatedby Rintheaboveformulas)occursfromasingle atom(atertiaryamineNorpentavalentP,respectively,shown inboldabove),bothtriazinesandcyclophosphazene dendrimersbranchfromalargerrigidring. Thecovalentsingle-moleculenanomaterialdescribedherein constitutesanewplatformfornanosynthesis.Itsviral dimensionsandsphericalshapearecomplementedbyawellde ned(althoughnotuniquelyperfect)composition.Extendingopportunitiesforfunctionalgroupvariationbothwithinand ontheperipheryofthesebuildingmaterialsisprecedentedin smaller-generationdendrimersofallclasses.Polymerization chemistryo ersanotherroutetosuchtargets.Indeed,linear andhyperbranchedpolymersofsimilarmolecularweightcan beachieved,althoughcontrolovertheshapeinlinearsystemsis notasreadilyguaranteedasitiswithdendrimers,noriscontrol overthecompositioninhyperbranchedsystems.14Selfassembly18hasachievedthesedimensions,includingPercec Â’ s dendrisomes19andothersystemsthatcanbesubsequently cross-linked,20althoughsuchmaterialsareusuallymore sensitivetofactorsincludingtemperature,detergents,and concentration.Theprimarydisadvantageofthepresentsystem, theburdenofsynthesis,hasbeenreducedsigni cantlybyusing aniterativeroutewithacommonmacromonomer.Accordingly, webelievethattherematerialsfurtherexpandthenanoperiodic table,ando erviruses.21Thesepropertiesinclude(i) attachmenttobiologicalsurfacessuchasbacteria,cells,and bone;(ii)deliveryofcargo,includingnucleicacidsorsmall molecules;and(iii)useasbuildingblocksforhybrid materials.22ASSOCIATEDCONTENT*SSupportingInformationDetailsofsynthesis,characterization,andcomputation.This materialisavailablefreeofchargeviatheInternetathttp:// pubs.acs.org.AUTHORINFORMATIONCorrespondingAuthore.firstname.lastname@example.orgNotesTheauthorsdeclarenocompeting nancialinterest.ACKNOWLEDGMENTSTheauthorsacknowledgethefollowing:NIH(NIGMSR01 65460,E.E.S.),theRobertA.WelchFoundation(A-0008, E.E.S.),theAcademyofFinland(13137582,M.K.),theEmil AaltonenFoundationandAaltoStartingGrant(M.K.),the CzechNationalCOSTProject(OC10053,J.M.),theCzech ScienceFoundation(13-06609S,J.M.),andtheACSPetroleum ResearchFund(47244-G4,O.A.).Thisworkmadeuseofthe AaltoUniversityNanomicroscopyCenterwithassistancefrom J.Seitsonenoncryo-TEM.REFERENCES(1)Levine,A.J. Viruses ;Scienti cAmericanLibrarySeries;W.H. Freeman:NewYork,1992;p240. (2)Tomalia,D.A.;Baker,H.;Dewald,J.;Hall,M.;Kallos,G.; Martin,S.;Roeck,J.;Ryder,J.;Smith,P. Polym.J. 1985 , 17 ,117. (3)Newkome,G.R.;Yao,Z.Q.;Baker,G.R.;Gupta,V.K. J.Org. Chem. 1985 , 50 ,2003. (4)Buhleier,E.;Wehner,W.;Vo gtle,F. Synthesis 1978 ,155. (5)(a)Helms,B.;Meijer,E.W. Science 2006 , 313 ,929.(b)Hu,J.; Xu,T.;Cheng,Y. Chem.Rev. 2012 , 112 ,3856.(c)Percec,V.;Peterca, M.;Dulcey,A.E.;Imam,M.R.;Hudson,S.D.;Nummelin,S.; Adelman,P.;Heiney,P.A. J.Am.Chem.Soc. 2008 , 130 ,13079. (6)Denkewalter,R.G.;Kolc,J.;Lukasavage,W.J.Macromolecular HighlyBranchedHomogeneousCompoundBasedonLysineUnits. U.S.Patent4,289,872,1981. (7)deBrabander-vandenBerg,E.M.M.;Meijer,E.W. Angew. Chem.,Int.Ed.Engl. 1993 , 32 ,1308. (8)Grayson,S.M.;Fre chet,J.M.J. Chem.Rev. 2001 , 101 ,3819. (9)Gillies,E.R.;Fre chet,J.M.J. J.Am.Chem.Soc. 2002 , 124 ,14137. (10)Caminade,A.M.;Majoral,J.P. Acc.Chem.Res. 2004 , 37 ,341. (11)(a)Maiti,P.K.;C ag in,T.;Wang,G.;Goddard,W.A.,III. Macromolecules 2004 , 37 ,6236.(b)Li,J.;Piehler,L.T.;Qin,D.; Baker,J.R.,Jr.;Tomalia,D.A. Langmuir 2000 , 16 ,5613. (12)(a)Caminade,A.M.;Laurent,R.;Majoral,J.P. Adv.Drug DeliveryRev. 2005 , 57 ,2130.(b)Majoral,J.P.;Caminade,A.M. Top.Curr.Chem. 1998 , 197 ,79. (13)(a)Simanek,E.E.;Abdou,H.;Lalwani,S.;Lim,J.;Mintzer,M. A.;Venditto,V.J.;Vittur,B. Proc.R.Soc.A 2010 , 466 ,1445.(b)Lim, J.;Simanek,E.E. Adv.DrugDeliveryRev. 2012 , 64 ,826.(c)Lim,J.; Pavan,G.M.;Annunziata,O.;Simanek,E.E. J.Am.Chem.Soc. 2012 , 134 ,1942. (14)Zhou,H.;Steinhilber,D.;Schlaad,H.;Sisson,A.L.;Haag,R. React.Funct.Polym. 2011 , 71 ,356. (15)Schmitz,K.S. AnIntroductiontoDynamicLightScatteringby Macromolecules ;AcademicPress:SanDiego,1990;pp205 ÂŠ 214. (16)Jensen,L.B.;Mortensen,K.;Pavan,G.M.;Kasimova,M.R.; Jensen,D.K.;Gadzhyeva,V.;Nielsen,H.M.;Foged,C. Biomacromolecules 2010 , 11 ,3571. (17)deGennes,P.G.;Hervet,H. J.Phys.,Lett. 1983 , 44 ,L351. (18)Whitesides,G.M.;Mathias,J.P.;Seto,C.T. Science 1991 , 254 , 1312. (19)(a)Peterca,M.;Percec,V.;Leowanawat,P.;Bertin,A. J.Am. Chem.Soc. 2011 , 133 ,20507.(b)Percec,V.;Wilson,D.A.; Leowanawat,P.;Wilson,C.J.;Hughes,A.D.;Kaucher,M.S.; Hammer,D.A.;Levine,D.H.;Kim,A.J.;Bates,F.S.;Davis,K.P.; Lodge,T.P.;Klein,M.L.;DeVane,R.H.;Aqad,E.;Rosen,B.M.; Argintaru,A.O.;Sienkowska,M.J.;Rissanen,K.;Nummelin,S.; Ropponen,J. Science 2010 , 328 ,1009. (20)O Â’ Reilly,R.K.;Hawker,C.J.;Wooley,K.L. Chem.Soc.Rev. 2006 , 35 ,1068. (21)(a)Tomalia,D.A. SoftMatter 2010 , 6 ,456.(b)Tomalia,D.A. NewJ.Chem. 2012 , 36 ,264.(c)Tomalia,D.A.;Christensen,J.B.; Boas,U. Dendrimers,DendronsandDendriticPolymers ;Cambridge UniversityPress:Cambridge,U.K.,2012. (22)Kostiainen,M.A.;Hiekkataipale,P.;Laiho,A.;Lemieux,V.; Seitsonen,J.;Ruokolainen,J.;Ceci,P. Nat.Nanotechnol. 2013 , 8,52. JournaloftheAmericanChemicalSociety Communicationdx.doi.org/10.1021/ja400432e | J.Am.Chem.Soc. 2013,135,4660 ÂŠ 46634663
AN EFFICIE NT APPROACH TO FUNCTIONAL MOLECULES : APPLICATION OF CLICK CHEMISTRY TO THE BENZOTRIFURANONE SCAFFOLD By ASHTON NIKOLE BARTLEY A THESIS PRESENTED TO THE GRADUATE SCHOOL OF THE UNIVERSITY OF FLORIDA IN PARTIAL FUL FILLMENT OF THE REQUIREMENTS FOR TH E DEGREE OF MASTER OF SCIENCE UNIVERSITY OF FLORIDA 2014
Â© 2014 Ashton Nikole Bartley
To those who believe
4 ACKNOWLEDGMENTS It is without a doubt that I would not be where I am today without the unconditional love and support of my family. My mom and dad have willingly sacrificed so much in order for me to have every opportunity possible, some of which they never had. I thank them for teaching me the meaning of hard work and their endless desire to make sure that I was happy and healthy. I am truly blessed to have parents that have encouraged me throughout my life, even when that meant sitting through 8 hour swim meets or listening to the same song on the piano a hu ndred times. And to my little brother Austin, thank you for being the best smallish bear any sister could ever hope for! I would also like to thank my advisor, Dr. Ron Castellano. Without his enthusiasm, patience, and encouragement, I would certainly not have made it to this point. He always puts the needs of his students above his own and encourages us to never stop asking questions. For his tireless pursuit of excellence and knowledge, I am grateful. Finally, I would like to thank all of my Castellan o lab mates, past and present, for their endless contributions to my education. I would like to thank Reggie, Davita, and Renan for patiently answering my seemingly endless number of random questions. I alking science with me (even after he graduated) and for always being up for a trip to Burrito Brothers. Other thanks go to Matt Baker, Anthony Quartararo, Tural Akhmedov, Bill Zhu, Danielle Fagnani, Ania Sotuyo, Lei Li, Asme Weldeab, Kyle Chesney, and Jo se Figueroa all of whom I have had the privilege to work beside and learn from.
5 TABLE OF CONTENTS page ACKNOWLEDGMENTS ................................ ................................ ................................ .. 4 LIST OF FIGURES ................................ ................................ ................................ .......... 7 ABSTRACT ................................ ................................ ................................ ................... 10 CHAPTER 1 DESIGN OF FUNCTIONAL MATERIALS BY EFFICIENT TRANSFORMATIONS ................................ ................................ ............................ 12 Construction of Designer Molecules ................................ ................................ ....... 12 Introduction to Click Chemistry ................................ ................................ ............... 17 Marriage of Heterofunctional Scaffolds and Click Chemistry ................................ .. 20 Scope of The sis ................................ ................................ ................................ ...... 23 2 APPLICATION OF CLICK CHEMISTRY TO THE BENZOTRIFURANONE SCAFFOLD ................................ ................................ ................................ ............. 25 Reactivity of BTF (2 1) ................................ ................................ ............................ 25 Initial Click Reaction: Testing Compatibility ................................ ............................ 29 CuAAC with a monosubstituted BTF derivative ................................ ...................... 30 Post click Functionalizations ................................ ................................ ................... 37 Attempts at Triple Clicked Motifs ................................ ................................ ............ 44 Overview and Discussion ................................ ................................ ........................ 46 Experimental ................................ ................................ ................................ ........... 48 Methods ................................ ................................ ................................ ............ 48 Procedures ................................ ................................ ................................ ....... 49 3 CONCLUSIONS AND FUTURE DIRECTIONS ................................ ...................... 64 Functionalization of BTF via CuAAC ................................ ................................ ....... 64 Future Directions ................................ ................................ ................................ .... 65 APPENDIX A INVESTIGATION OF NOVEL, MULTIFUNCTIONAL SCAFFOLDS ....................... 68 Proposed Future Symmetrical Targets ................................ ................................ ... 72 Discussion ................................ ................................ ................................ .............. 73 B APPROACHES TO DENDRIMERS THROUGH FUNCTIONALIZATION OF BTF . 74 Synthesis of Dendrimer Precursors ................................ ................................ ........ 74
6 Use of Alternative Diamines ................................ ................................ .................... 77 Final Attempts at Dendrimer Synthesis ................................ ................................ ... 78 Di scussion ................................ ................................ ................................ .............. 79 Experimental ................................ ................................ ................................ ........... 80 C SUPPLEMENTARY INFORMATION FOR CHAPTER 2 ................................ ......... 88 REFERENCES ................................ ................................ ................................ ............ 109 BIOGRAPHICAL SKETCH ................................ ................................ .......................... 115
7 LIST OF FIGURES Figure page 1 1 Metho ds of functionalization.. ................................ ................................ ............. 14 1 2 Ugi multicomponent reaction and Passerini multicomponent reaction. ............... 15 1 3 Statistical functionali zation of a symmetric core. ................................ ................ 16 1 4 Trifunctionalization of cyanuric chloride ................................ .............................. 17 1 5 Functionalization of BTF through sequential am inolysis. ................................ .... 17 1 6 A fe w examples of click reactions . ................................ ................................ ...... 19 1 7 Example of the 1,4 regioisomer formed by CuAAC. ................................ ........... 19 1 8 Triple FRET cascade synthesized by the Romieu group). ................................ . 22 2 1 Potential products via aminolysis of 2 1 ................................ ............................. 26 2 2 A few examples of nucleophilic amines that have been used for BTF ................ 26 2 3 One pot functionalization yielding a trifunctional molecule. ................................ 27 2 4 X ray crystal structure of BTF ................................ ................................ ............. 28 2 5 Benzotripyranone ( BTP ) shows littl e to no bond length alternation . ................... 29 2 6 Trifunctionalization of BTF to yield 2 4 . ................................ .............................. 30 2 7 Aminolysis of 2 1 with propargylamine to yield alkyne 2 6 . ................................ 31 2 8 Synthesis of benzylazide 2 7 from benzylbromide. ................................ ............. 32 2 9 S N 2 substitution of 2 ethylhexylbromide with NaN 3 . . ................................ ........... 33 2 10 Substituti on of 1 bromoheptane with NaN 3 . ................................ ........................ 34 2 11 Synthesis of 1 azidod odecane. . ................................ ................................ .......... 36 2 12 Synthesis of azide 2 15 : tosylation followed by sub stitution with NaN 3 . ............. 36 2 13 Synthesis of triazole 2 16 . ................................ ................................ .................. 37 2 14 Synthesis of 2 17 by aminolysis of remaining two lactone rings. ........................ 38
8 2 15 Functionalization through aminolysis with heptylamine. ................................ ..... 38 2 16 Synthesis of a difunctional molecule with a clickable alkyne handle. .................. 39 2 17 Aminolysis with heptylamine to yield 2 20 . ................................ ......................... 40 2 18 Addition of furfurylamine to yield the trisubstituted product. ............................... 41 2 19 Synthesis of a functional molecule ................................ ................................ ..... 42 2 20 Synthesis a trifunctional system through sequential aminolysis of BTF . ............ 42 2 21 Synthesis of a trifunctional molecule through aminolysis of 2 11 . ....................... 43 2 22 General synthesis of a trifunctional molecule ................................ ..................... 44 2 23 Synthesis of a fully substituted phloroglucinol derivative. ................................ ... 44 2 24 Attempted synthesis of tristriazoles 2 30 and 2 31 . ................................ ............ 45 2 25 Two tristriazo les used in catalysis for CuAAC . ................................ ................... 46 3 1 Proposed phloroglucinol derivative ................................ ................................ ..... 66 A 1 Proposed sequential functionalization of MTA . ................................ ................... 68 A 2 C=O stretching frequencies (cm 1 ) and b ond orders ................................ ........... 70 A 3 Synthesis of a benzyltriimide. ................................ ................................ ............. 71 A 4 Transimidation study resulting in a complex mixture of imides. .......................... 71 A 5 Reaction of N ac e tylisatin with 4 bromoaniline. ................................ .................. 72 A 6 Sandmeyer, Stolle, and Gassman synthesis of isatin. ................................ ........ 73 B 1 Synthesis of a first generation dendrimer from BTF . ................................ .......... 75 B 2 Synthesis of first generation dendrimer core. ................................ ..................... 76 B 3 Synthesis of di substituted dendron. ................................ ................................ ... 77 B 4 Synthesis of dendrimer with 4 AMP. ................................ ................................ ... 77 B 5 Sonogashira coupling toward first generation dendrimer synthesis. ................... 78 B 6 Attempted synthesis of linked compound bearing four lactone rings. ................. 79 B 7 1 H NMR (Top) and 13 C NMR (Bottom) of B 2 . ................................ .................... 84
9 B 8 1 H NMR (Top) and 13 C NMR (Bottom) of B 3 . ................................ .................... 85 B 9 1 H NMR of B 4 . ................................ ................................ ................................ ... 86 B 10 1 H NMR (Top) and 13 C NMR (Bottom) of B 6 . ................................ .................... 87 C 1 (Top) 1H NMR and (Bottom) 13 C NMR spectrum of 2 4 . ................................ .... 89 C 2 (Top) 1H NMR and (Bottom) 13 C NMR spectrum of 2 5 . ................................ .... 90 C 3 (Top) 1H NMR and (Bottom) 13 C NMR spectrum of 2 6. ................................ .... 91 C 4 (Top) 1H NMR and (Bottom) 13 C NMR spectrum of 2 8 . ................................ .... 92 C 5 (Top) 1H NMR and (Bottom) 13 C NMR spectrum of 2 10 . ................................ .. 93 C 6 Crude 1 H NMR spectrum of 2 11 . ................................ ................................ ....... 94 C 7 (Top) 1H NMR and (Bottom) 13 C NMR spectrum of 2 11 . ................................ .. 95 C 8 (Top) 1H NMR and (Bottom) 13 C NMR spectrum of 2 14 . ................................ .. 96 C 9 ( Top) 1H NMR and (Bottom) 13 C NMR spectrum of 2 15 . ................................ .. 97 C 10 (Top) 1H NMR and (Bottom) 13 C NMR spectrum of 2 16 . ................................ .. 98 C 11 (Top) 1H NM R and (Bottom) 13 C NMR spectrum of 2 17 . ................................ .. 99 C 12 (Top) 1H NMR and (Bottom) 13 C NMR spectrum of 2 18 . ................................ 100 C 13 (Top) 1 H NMR and (Bo ttom) 13 C NMR spectrum of 2 19 . ................................ . 101 C 14 (Top) 1 H NMR and (Bottom) 13 C NMR spectrum of 2 20 . ................................ . 102 C 15 (Top) 1 H NMR and (Bottom) 13 C NMR spectrum of 2 21 . ................................ . 103 C 16 (Top) 1 H NMR and (Bottom) 13 C NMR spectrum of 2 23 . ................................ . 104 C 17 (Top) 1 H NMR and (Bottom) 13 C NMR sp ectrum of 2 24 . ................................ . 105 C 18 (Top) 1 H NMR and (Bottom) 13 C NMR spectrum of 2 25 . ................................ . 106 C 19 (Top) 1 H NMR and (Bottom) 13 C NMR spectrum o f 2 29 . ................................ . 107 C 20 (Top) 1 H NMR and (Bottom) 13 C NMR spectrum of 2 34 . ................................ . 108
10 Abstract of Thesis Presented to the Graduate School of the University of Florida in Partial Fulfillment of the Requirements for th e Degree of Master of Science AN EFFICIENT APPROAC H TO FUNCTIONAL MOLECULES : APPLICATION OF CLICK CHEMISTRY TO THE BENZOTRIFURANONE SCAFFOLD By Ashton Nikole Bartley August 2014 Chair: Ronald Castellano Major: Chemistry Highly functional, designer molecules are desirable because they combine the properties of several, individual units to provide a material with specific behavior and performance characteristics. Benzotrifuranone ( BTF ) is a C 3h symmetric trilactone compound that can quickly lead to a trifunctional target in good yield by the sequential addition of three different amines at a convenient reaction temperature without purification between each amine addition. The three lactone ri ngs of BTF are strain coupled and display useful kinetic differentiation between reactive sites, allowing aminolysis reactions to proceed sequentially and selectively along a strain release gradient. Furthermore, BTF downstream functionalization through copper catalyzed azide alkyne cycloaddtion (CuAAC). While simple, the efficiency and precision of this 1,3 dipolar cycloaddition makes it an ideal reaction to integrate into the BTF functionalization scheme. Aminolysis of BTF with propargylamine provides a handle for CuAAC. CuAAC of BTF derivatives works efficiently with linear azides such as 1 azidoheptane, 1 azidododecane, and an oligoethylene glycol azide. The reactions always reach full convers ion, and the product is obtained in 50% yield (on average) after purification.
11 Clicked difuranone BTF derivatives were then functionalized through aminolysis to yield difunctional and trifuctional molecules; several primary amines like heptylamine, furfur ylamine, and isopropylamine were used for aminolysis. Pufication of these fully functionalized phloroglucinol derivatives afforded the product in 50 60% yield. Thus, BTF can be predictably and sequentially functionalized through two efficient reactions with few limitations.
12 CHAPTER 1 DESIGN OF FUNCTIONAL MATERIALS BY EFFICIENT TRANSFORMATIONS With the vast number of synthetic tools available for molecular manipulation, there is a growing desire to access complex and highly functional materials. In deed, design of well defined and highly functional molecules requires combining the properties of several, individual units in order to deliver a material with specific behavior and performance characteristics. Since structure and function go hand in hand , system design is directed toward innovative and sophisticated materials, where even a small change in structure can result in noticeable differences in properties. These so called iomedical imaging, complex polymer topologies, and molecular machines. 1 5 Careful design of a heterofunctional target allows the chemist to fine tune the chosen properties in order to achieve a high performance system. These diverse systems often have pro perties very different from their simpler, homofunctional counterparts. Building designer systems is a way for the chemist to wholly control the outcome by modifying each element until the desired complexity and fu nctionality is achieved. C hapter 1 will assess some of the methods to access heterofunctional, designer molecular systems. The methods include classical functionalization approaches, such as stepwise functionalization, as well as the advent of click chemistries for streamlined synthesis of comp lex molecular architectures. Construction of Designer Molecules To take the most advantage of such systems, it is desirable to access customizable materials in the simplest and most predictable manner as possible. Employing systematic and practical c hemistries is the main goal in elaborating
13 molecules with a high degree of functionalization. However, many of the available methods still require lengthy, highly specific transformations as well as taxing purification. These methods are often time consu ming and afford the product in low overall yield, slowing discovery of desired targets. The synthesis of complex architectures is accomplished in one of several ways. Shown in Figure 1 1 is a schematic of four major modes of multifunctionalization; th ese include (1) orthogonal functional group manipulation, (2) cyclization of linear precursors, (3) selective protection/deprotection strategies, and (4) stepwise functionalization. While these are certainly not the only means of functionalization, they d o represent a large majority. Orthogonal functional group manipulation (Figure 1 1A), used extensively in the accelerated synthesis of dendrimers, 6 involves selectively transforming different functional groups found within a single molecule. Some example s of such manipulations include Diels Alder cycloaddition, thiol ene addition, and Michael addition. Although these reactions generally proceed with good yield, execution of this method is difficult because of the difficulty in synthesizing the core and t he constraints of subsequent functionalization. Cyclization of linear precursors (Figure 1 1B), which finds significant utility in the porphyrin and peptide communities, is limited only to those structures that can be synthesized from a linear compound. 7 10 Selective protection/deprotection strategies (Figure 1 1C) are a safe and deliberate way to synthesize multifunctional domains. This method entails the selective deprotection and subsequent transformation of a particular reactive site within a molecul e containing multiple protected sites. 11,12 While useful, given careful selection of protecting groups and deprotection methods, the process is overall tedious and low yielding. Conversely,
14 stepwise functionalization (Figure 1 1D) starts with a symmetric compound that is desymmetrized through sequential chemical transformations. This strategy has found use in the synthesis of carbohydrate receptors. 13 Despite their interesting, individual characteristics, building these complex architectures is often a ccompanied by a laborious synthesis and low overall yield. 9, 13, 14, 15 A) B) C) D) Figure 1 1. Methods of functionalization. A) Orthogonal functional group manipulation. B) Cyclization of linear precursor. C) Selective protection/deprotection. D) Stepwise functionalization. In addition to these modes o f multifunctionalization, there are a few multicomponent reactions that have been used to link several building blocks together in order to access a larger system (Figure 1 2). The Ugi reaction is such a multicomponent reaction involving an isocyanate, al dehyde, carboxylic acid, and an amine where the resulting structure is a bis amide through a single, synthetic step. This reaction is frequently used to generate dipeptides for combinatorial libraries but has also found use in macrocycle construction. 16,1 7 Another multicomponent reaction is the Passerini re action which couples an isocyanate, a carboxylic acid, and an aldehyde or
15 acyloxycarboxamides. 18 It has also been used in the creation of combinatorial libraries as well as natural product and dendrimer synthesis. 19 21 Figure 1 2. a) Ugi multicomponent reaction. b) Passerini multicomponent reaction. Stepwise functionalization has found the most wide spread utility in the construction of multivalent domains. This method typically starts with a sy mmetrical and shape persistent scaffold where three or more identical reactive sites serve as a starting point for a divergent synthesis. 9 In a statistical system, there is little to no kinetic differentiation between reactive sites of the starting materi al and the intermediates, and thus a statistical distribution of products can arise (Figure 1 3a). This often leads to complex reaction mixtures, laborious purification steps, and low overall yield. However, useful kinetic differentiation among identical , competing reactive sites would offer a more appealing approach. This would result in a gradual decrease in reactivity upon each transformation event. This ideal system would allow desymmetrization through a high yielding, one pot synthesis with little t o no purification required while avoiding any complex protection/deprotection strategies and boosting atom economy (Figure 1 3b).
16 Figure 1 3. a) Statistical functionalization of a symmetric core illustrating complex reaction mixture. b) Ideal functio nalization of a symmetric core with useful kinetic differentiation between reactive sites. Symmetric scaffolds that meet all the criteria for this ideal case are especially rare, with cyanuric chloride one of the only known examples in the literature as i ts chemistry dates back to the 1880s (Figure 1 4). The chlorine atoms of cyanuric chloride act as leaving groups in a S N Ar mechanism with up to three different nucleophiles in a one pot procedure at convenient reaction temperatures and times. 22,23 It has found many applications, including use as a synthetic scaffold for synthetic dyes, pesticides, and optical brighteners. 24 For over 100 years, cyanuric chloride has remained one of the only examples of a highly symmetrical but differentially functionaliz able scaffold until the introduction of benzotrifuranone by the Castellano group in 2010. 25 Benzotrifuranone ( BTF ), a symmetric scaffold with three inductively coupled lactones,
17 Figure 1 4. Trifunctionalization of cyanuric chloride by three different amine nucleophiles. can undergo selective and sequential ring opening events through aminolysis. This type of sequential functionalization yields a heterotrifunctional molecule in good yield and predictable fashion (Figure 1 5). Addition of one equivale nt of an amine provides the aminolysis product in a non statistical yield. Furthermore, the sequential addition of three amines (one equivalent each) affords a tri functional molecule in a one pot synthesis in over 85% yield. BTF became an exciting prosp ect in our research group and the idea of using it as a tool for the rapid construction of unique and interesting macromolecules quickly emerged. The chemistry of BTF will be further discussed in Chapter 2. Figure 1 5. Functionalization of BTF through sequential aminolysis. Introduction to Click Chemistry Even with the extensive amount of synthetic reactions available and previously mentioned heterofunctionalization strategies, designer molecules are still difficult to
18 access as these approaches cannot be easily applied to all desired systems. Furthermore, it is advantageous to employ a few, robust reactions in order to quickly realize the advanced functions of complex structures. Because of their robust and orthogona l nature, click reactions have been extensively integrated in to many areas of science, particularly since Sharpless first reported and outlined the click chemistry paradigm. 26 Applying this methodology allows easy functionalization and rapid access to c omplex molecules. The idea of click chemistry was first introduced by Sharpless and co workers in an influential 2001 paper. 26 The purpose of the paper was to turn the focus of chemical modification towards identifying the most optimized and simplifie d manner as possible in which to access the target. In that paper, Sharpless defined a set of strict principles that a reaction must meet in order to be considered a click reaction. The reaction must be wide in scope, proceed with high yields, and genera te few byproducts. The reaction should be conducted with readily available starting materials and benign solvents (like water). Finally, the final product should require little to no purification; if required, nonchromatographic methods are preferred. 26 There are a number of reactions that meet these strict criteria and fall under the umbrella of click chemistry. These reactions include, but are not limited to, Diels Alder cycloaddition, thiol ene addition, nitroxide radical coupling, strain promoted az ide alkyne cycloaddition (SPAAC), and copper catalyzed azide alkyne cycloaddition (Figure 1 6). 27,28
19 Figure 1 6. A few examples of click reactions. 1. Diels Alder cycloaddition. 2. Thiol ene addition. 3. Strain pr omoted azide alkyne cycloaddition. 4. Nitroxide radical coupling. 5. Copper catalyzed azide alkyne cycloaddition. Copper catalyzed azide alkyne cycloaddition (CuAAC) has become synonymous with the term click chemistry. Huisgen first introduced 1,3 d ipolar cycloadditions in the 1960s as a way to synthesize 5 membered heterocycles. Synthesis of 1,2,3 triazoles at the time did not occur under mild conditions or in good regioselectivity. In 2002, Sharpless and Meldal individually reported that the addi tion of a copper catalyst yielded exclusively the 1,4 regioisomer from the reaction of an azide and a terminal alkyne (Figure 1 7). 29,30 Figure 1 7. Example of the 1,4 regioisomer formed by CuAAC. What makes this reacti on so versatile is that it can be conducted under a variety of mild conditions. The reaction usually takes place in a mixture of water and a partially soluble organic solvent which can be tert butanol, tetrahydrofuran, dimethylsulfoxide,
20 ethanol, etc. 31 The active copper species in the reaction is Cu(I) and can be supplied by copper metal, Cu(I) salts, and Cu(II) salts. Many choose to use a Cu(II) source such as CuSO 4 along with a reducing agent like sodium ascorbate to afford the active copper species. 31 This reaction is highly specific and can be used in the presence of sensitive functional groups. CuAAC typically works best with an electron rich azide and an electron deficient terminal alkyne. 32 A fairly recent paper provides a more detailed insight in to why the copper catalyzed reaction yields exclusively the 1,4 regioisomer of the 1,2,3 triazole. 26 First, coordination of copper to the acetylene greatly reduces the acidity of the terminal alkyne proton and allows for the formation of the activated alkyne. Another copper ion complex which directs attack of the electrophilic, terminal azide nitrogen carbon of the acetylene complex. Oxidative coupling followed by reductive elimination yields the pure regioisomer. 33 Marriag e of Heterofunctional Scaffolds and Click Chemistry Because of its orthogonality and robust nature, CuAAC has been implemented in to many areas of science including biological, materials, and polymeric systems. It has been used to modify DNA strands, labe l nucleosides, PEGylate drug molecules, and make derivatives of carbohydrates. 34 36 On the materials end, click chemistry has eased surface modification and functionalization of carbon nanotubes and accelerated the synthesis of supramolecular systems. 37 4 0 These can be viewed as simply illustrative examples to the growing number of materials applications as reflected in the over 1500 papers using click chemistry for materials science between 2010 and 2012 alone. 27 Finally, click reactions have been use d for various polymer modifications: formation of block co polymers, hyperbranched polymers, and polymer end groups. 41 43
21 With the call to simplify and quicken discovery of designer molecules, click reactions have been incorporated in to multifunctional sc affolds. This incorporation can streamline synthesis, boost overall yield, and ease target identification. There are several examples of this combination of strategies including a cross linker for conjugation from the Romieu group. A triple FÃ¶rster reso nance energy transfer (FRET) cascade and a luminescent peptide oligonucleotide conjugate were constructed using this method (Figure 1 8A). By first synthesizing an organic scaffold with three clickable handles, the authors were able to attach cyanine base d dyes in either a sequential or one pot reaction. 44 Similarly, Boturyn and co workers synthesized a sophisticated system that features several biomolecules such as peptides, sugars, and nucleic acids (Figure 1 8B). They prepared a cyclodecapeptide with three different clickable sites for later functionalization: CuAAC, thiol ene, and oxime ligation. Through sequential reactions, the authors were able to use biomolecular friendly reactions while avoiding extensive purification. 45 As a final example of th e marriage between multifunctional scaffolds and click chemistry, the Li group fashioned a multitailed giant surfactant caprolactone), and 2 mercaptoacetic acid to a trifunct ional polyhedral oligomeric silsesquioxane through sequential SPAAC, CuAAC, and thiol ene transformations. 46
22 Figure 1 8. A) Triple FRET cascade synth esized by the Romieu group starting with benzenic building block 1 8a . B) Synthesis of a cyclodecapeptide starting with a serine derivative ( 1 8d ) and 4 pentynoic acid ( 1 8c ). Although these systems use efficient click reactions to yield diverse and hig hly functional compounds, the overall yield can still be rather low. The synthesis of the core is usually very lengthy as it is synthesized through conventional stepwise functionalization. For example, the starting cyclodecapeptide was synthesized by fun ctionalizing the amino acid building blocks followed by peptide synthesis and cyclization for a total of 12 steps. 45,47 The atom economy for these transformations were low; protecting groups were used and coupling reagents were often necessary to form pep tide bonds. Additionally, the core used in the triple FRET cascade was synthesized through stepwise functionalization in 6 steps (excluding synthesis of the PEG linkers) in 10% overall yield. 46 Although the starting material for the core was commercially available, each functional group on the benzene core had to be transformed before actual functionalization. Factoring in the functionalization of the A) B )
23 core and subsequent yields and purification steps makes these strategies less appealing and overall time consuming and expensive. Another downfall of these systems (and multifunctionalization strategies in general) is that they are not always tailorable or versatile. The core is not always made from cheap or commercially available materials, requiring ext ra synthetic steps and purification to install specific functional groups. Some of the functional groups on a particular substrate may not be amenable to harsh reaction conditions that are used for scaffold synthesis, i.e. base, acid, or heat. The atom e conomy of each reaction is often times low as protection/deprotection methods and multi step transformations are required. Positional modification can also be arduous as incomplete or excessive transformations diminish yield. Extra care must be given in choosing both the reaction sequence and the molecules to be added. Therefore, a tailorable, versatile, and easily accessible scaffold coupled with efficient click reactions could significantly accelerate the synthesis of designer molecules. Scope of Th esis Although there are a vast number of synthetic strategies that can be applied to building intricate architectures, there is still a need for more efficient and streamlined methods. Classical multifunctionalization methods are often time consuming and laborious which impedes discovery. The implementation of robust click reactions has helped to simplify the overall process, but accessing the multifaceted frameworks is often through classical, stepwise techniques. Thus, a simple molecular scaffold capab le of accessing heteromultifunctional systems through highly efficient, orthogonal, and flexible methods would be a valuable tool for synthesizing designer molecules. BTF is such a scaffold where the symmetric molecule can undergo positional modification via
24 amidation with a variety of amines. Furthermore, BTF can be functionalized with a dipolar cycloaddition is a simple reaction, its efficiency and precision make it an ideal reaction to integrate into the BTF functionalization scheme. In this way, BTF can be limitations as there are a considerable number of substrates that can be appended to the BTF core. This type of simple and streamlined synthesis shifts the focus from general multifunctionalization to accessing complex (heterofunctional) structures using a few, reliable techniques, the embodiment of click chemistry. Combinin g these two robust manipulations not only broadens the scope of BTF but opens the door for easy target identification in future applications. Chapter 2 will present experimental work done toward synthesis of trifunctional molecules using CuAAC. Several mo nofunctionalized BTF compounds were synthesized through CuAAC and azide scope was studied. These compounds were then further functionalized with a variety of amines to yield trifunctional systems. Synthetic observations will be discussed and reaction out comes will be compared.
25 CHAPTER 2 APPLICATION OF CLICK CHEMISTRY TO THE BENZOTRIFURANONE SCAFFOLD Reactivity of BTF (2 1) Previous work in our group involved developing the chemistry for the BTF scaffold. This work included exploring the sequential ri ng opening events of BTF as well as amine nucleophile scope and order of amine addition. Kinetic studies were performed in order to better understand the unique reactivity of this system. Additionally, crystal structure analysis helped to give insight in to the strain within the molecule. 25,48 Upon aminolysis of BTF ( 2 1 ), three products along with unreacted starting material are imaginable (Figure 2 1A). Using purely statistical methods, product distribution of a statistical system can be modeled (Figu re 2 1B). 25,48 A statistical system, in this case, is one that assumes equal reactivity among the lactone rings of 2 1 , 2 1A (with two different lactones), and 2 1B where controlling the desired degree of functionalization is difficult. With exact stoich iometric control, addition of 1.0 equivalent of an amine gives the product with the desired degree of substitution in about 37% yield (entry 1). Likewise, addition of 2.0 equivalents of an amine gives in the greatest yield undesired trisubstituted 2 1C pr oduct which is unable to undergo further functionalization (entry 2). However, experimental results show a much different product distribution (Figure 2 1B). Adding 1.0 equivalent of an amine gives the expected product 2 1A in 96% yield (entry 3) while t he addition of 2.0 equivalents of an amine gives product 2 1B in the greatest yield (entry 4). These positive initial studies prompted further investigation of amine scope.
26 A) B) Entry SM Amine equiv. 2 1 yield (%) 2 1A yield (%) 2 1B yie ld (%) 2 1C yield (%) 1 2 1 1.0 35.8 36.8 19.0 8.4 2 2 1 2.0 9.4 22.7 26.5 41.5 Entry SM Amine equiv. 2 1 yield (%) 2 1A yield (%) 2 1B yield (%) 2 1C yield (%) 3 2 1 1.0 <1 96 <3 0 4 2 1 2.0 0 Trace 92 <8 5 2 1 3.0 0 0 0 99 Figure 2 1. A) Potent ial products via aminolysis of 2 1 . B) Entries 1 & 2: Statistical product distribution assuming equal reactivity among 2 1 , 2 1A , and 2 1B . Entries 3 5: Experimental product distributions. Yields are reported for reactions using heptylamine as the nuc leophile under general reaction conditions as follows: Entry 3, 41 Â°C (< 30 min); Entry 4, 41 Â°C (3 h); Entry 5, 41 Â°C to rt (16 h). Many types of amines were found suitable to act as nucleophiles for the ring opening events of 2 1 (Figure 2 2). Pri mary, secondary, cyclic, and benzylic amines as well as those with weaker competitive nucleophiles ( 2 2e ) have all been shown to proceed with good reliability in lactone ring openings of BTF as dansyl amine ( 2 2f ) and chiral amin o acid derivatives ( 2 2g ) also work well. Figure 2 2. A few examples of nucleophilic amines that have been used for BTF aminolysis.
27 Desymmetrization of 2 1 through sequential addition of three different amine nucleophiles to yield a trifunctional molecule was found to work efficiently. In fact, a trifunctional molecule could be synthesized through a one pot procedure to yield the desired compound 2 3 in nearly 85% yield (Figure 2 3). It was found that neither the identity of the amine nor the order of amine addition significantly altered the yield. The ease of synthesis and overall high yield would not have been possible using a symmetric system with statistical reactivity. Figure 2 3. One pot functionalization yielding a trifunctional molecule bearing alkene, alkyne, and furan functionalities. Kinetic studies were later performed to quantify the reactivity of BTF . The ring opening events of BTF sho w expected differences in reaction rates. The pseudo first order rate constant for the first ring opening of BTF was found to be 6.94 M 1 s 1 (at 24 Â°C in MeCN) while the rate constants for the second and third ring openings were 0.125 and 0.0126 M 1 s 1 , respectively. The last two ring openings are much slower than the first; the second ring opening is an order of magnitude slower than the first and an order of magnitude faster than the third. 2 Of course, this difference in reactivity between the first and second and second and third ring openings does increase with a decrease in temperature. Selectivity factors (ratio of rate constants, k / k 1 ) double for the first and second ring openings upon lowering the temperature from 24 Â°C to 40 Â°C.
28 It seems that the non statistical functionalization of BTF arises from both the strain associated with the system and inductive effects. Analysis of the crystal structure of 2 1 reveals bond length alternation of the bonds within the central benzene core (Figure 2 4). This bond length alternation speaks to the strain associated with having three 5 membered rings fused to the core. There is a noticeable difference of Â±0.021 Ã… (average) among the bond lengths, where the bonds comprising the furanones are longer tha n the bonds between rings. Important to note here is that there is little Figure 2 4. X ray crystal structure of BTF showing alternating carbon carbon bond lengths within the benzene core. Bond lengths are given in Ã…. Atom colors: C, gray; H, white; O, red. alternation found in the similar compound, benzotripyranone ( BTP ) (Figure 2 5). The first ring opening relieves much of the stress associated with the system as the kinetic data suggests, given that the difference in rates between aminolys es 1 and 2 is larger than between aminolyses 2 and 3. Additionally, inductive effects presumably play a role in the change in reactivity of the lactone rings. After the first ring opening, there is a change in electron density within the benzene ring as the inductively withdrawing lactone becomes a less inductively withdrawing phenol. This is corroborated by deshielded resonances of the lactone carbonyls in the 13 C NMR spectra and shifts to ring. 49,50 For example, the three lactone carbonyl carbons of BTF appear at 171.9 ppm 1.396 1.371 1.397 1.372 1.386 1.374
29 while the chemical shifts of the two lactone carbonyl c arbons of a compound such as 2 1A appear at approximately 173.8 and 174.1 ppm. Figure 2 5. Benzotripyranone ( BTP ) shows little to no bond length alternation for the carbon carbon lengths within the benzene core. Init ial Click Reaction: Testing Compatibility As discussed in Chapter 1, in searching for ways to expand the functional scope of BTF , click chemistry emerged as an attractive approach. Coupling the efficient ring openings of BTF with CuAAC would offer a simpl e and streamlined synthesis of novel compounds by applying two reliable techniques. By functionalizing BTF with a clickable handle, we could expand the types of substrates to be appended to the scaffold. Moreover, BTF could then be affixed to larger syst ems such as polymers or surfaces. To test the applicability of this approach, an initial click reaction was performed using a trifunctional BTF molecule bearing three alkyne handles. Synthesis of the test substrate began with the addition of 3.5 equivalen ts of propargylamine to 2 1 to yield trifunctional compound 2 4 in 63% yield after column chromatography (Figure 2 6A). For the subsequent click reaction, a mixture of THF and H 2 O was chosen as the solvent system along with a catalytic amount (15 mol %) o f CuSO 4 as the copper source and sodium ascorbate (NaAsc) as the sacrificial reducing agent. Trialkyne 2 4 was treated with 1 azidohexane under the previously mentioned click conditions (Figure 2 6B). The desired product 2 5 could be filtered from the or ganic phase during the work up to yield 57% of the expected mass. Both 1 H and 13 C
30 NMR of the filtered solid showed excellent purity and no further purification was performed. Although TLC indicated full conversion of the starting material, Figure 2 6. A) Trifunctionalization of BTF to yield 2 4 with three clickable alkyne handles. B) Triple click reaction with 2 4 to yield the tristriazole 2 5 . the remainder of the mass could not be isolated. The material recovered from the mother liquor indicated that 2 5 was present although not in good purity (impurities were not characterized) and isolation of any remaining 2 5 was not pursued. CuAAC with a monosubstituted BTF derivative Giv en the positive results of the previous reaction, performing click reactions on a mono substituted BTF derivative was considered. It was important to understand whether the remaining two lactone rings could be subjected to the aqueous click conditions and remain intact. Additionally, this type of sequential functionalization would be valuable for utilizing BTF in future applications. For this study, a mono alkyne 2 6 A) B)
31 was first synthesized and then azide scope and reaction conditions were explored in subs equent click reactions. Mono alkyne 2 6 was synthesized by adding 1.0 equivalent of propargylamine to 2 1 to yield the product in 77% yield after column chromatography (Figure 2 7). Using Figure 2 7. Aminolysis of 2 1 with propargylamine to yield alkyne 2 6 . 1.0 equivalent of the amine and careful control of the temperature allows only the formation of 2 6 with no detectable disubstituted product. Although using 1.0 equivalent works well, 0.98 equivalents of propargy lamine is often used to compensate for any measuring issues. Due to the difference in solubility between 2 1 and 2 6 in DCM, it is not necessary to run a column as DCM can be added to the crude mixture and 2 6 filtered off. Since 2 6 is partially soluble in DCM, some mass is lost in the filtration. The yields from filtration and column chromatography are comparable. A primary azide was chosen for the first click reaction of 2 6 . Benzylazide 2 7 (Figure 2 8A) was chosen because it is readily made, well reported, and easy to track via TLC ( BTF derivatives must be stained to be viewed by TLC). The same reaction conditions were chosen for this CuAAC as used in the synthesis of 2 5 (Figure 2 6). Rewardingly, the click reaction proceeded efficiently in aqu eous conditions and the lactone rings remained intact as was observed by the characteristic alpha methylene peak pattern in 1 H NMR (Scheme 2 8B). The alpha methylene protons are observed at approximately 3.9, 3.8, and 3.5 ppm for monosubstituted BTF deriv atives (e.g. 2 1a in
32 Figure 2 1). The desired clicked product 2 8 was isolated in 65% yield after column chromatography. Figure 2 8. A) Synthesis of benzylazide 2 7 from benzylbromid e. B) CuAAC of 2 6 to afford 2 8 . The crude 1 H NMR of 2 8 showed complete conversion of alkyne 2 6 with a few impurities (unidentified) that were removed by column chromatography. A different set of reaction conditions was attempted in order to compare the crude purity of the reactions and determine whether aqueous conditions were the cause of the unidentified impurity. The reaction was run in dry THF with 3.0 equivalents of CuI and a stoichiometric amount of diisopropylethylamine (DIPEA). The outcome of the reaction was not much different than that of the reaction run in aqueous medium as crude yields were comparable (> 90%). The crude 1 H NMR spectrum showed complete conversion of the starting alkyne 2 6 , but there were small, unidentified peaks in th e 1.0 3.0 ppm region. Column chromatography resulted in a lower recovered yield (26%) presumably due to the change in reaction conditions. It is well known that a variety of triazole systems chelate to copper so using a large excess of copper was certa inly not advantageous. 51 53 B) A)
33 It seemed plausible that increasing the solubility of the triazole substituent would help in purification of the target and increase overall yield. Our group extensively uses the 2 ethylhexyl chain to increase the solubility of conjugated materials. Since 2 ethylhexylbromide can easily undergo substitution with azide to afford 2 ethylhexylazide ( 2 9 ), it was the next azide of choice and prepared easily (Figure 2 9A). Again, 2 6 was subjected to aqueous click conditions (15% CuSO 4 , NaAsc, THF, H 2 O) with 2 ethylhexylazide 2 9 (Figure 2 9B). This reaction was slower compared to that with benzylazide 2 7 and had to be stirred overnight. We were once again pleased to find that the two lactone rings were still intact, but the cru de 1 H NMR spectrum showed a small amount of the starting alkyne 2 6 . Nonetheless, the pure product 2 10 could be Figure 2 9. A) S N 2 substitution of 2 ethylhexylbromide with NaN 3 . B) CuAAC of 2 6 with 2 ethylhexylazide. isolated in 37% yield after column chromatography. Organic solvent reaction conditions (like previously stated for 2 8 ) were tried as well but did not provide a better outcome. Full conversion of the starting materia l was not observed and problems eluting the product during column chromatography resulted in low overall yield (17%). As mentioned before, using an excess of copper (3.0 equivalents) is not advantageous given the coordination between triazoles and copper and likely was a reason for difficult purification; use of organic solvent conditions was no longer attempted. 51 53 B) A)
34 We were still disappointed by the outcome of the click reactions so far from the diminished yield to the long reaction times. In comparison to 2 5 (Figure 2 6) it seemed that 2 ethylhexylazide should work very well. We thought that maybe there were some steric issues associated with the azide. Although secondary and even tertiary azides have been shown to work well in CuAAC, with this syste m the branching seemed to hinder the reaction. 29,31 A linear alkyl chain was chosen for the next study. Synthesis of 2 11 (Figure 2 10) began with the synthesis and characterization of 1 azidoheptane 2 12 (Figure 2 10A). This conversion was very efficien t and yielded the desired azide in nearly quantitative yield and with high purity. Reaction of 2 6 and 2 12 was conducted under aqueous CuAAC conditions (10 mol% Cu, THF/H 2 O, NaAsc) as shown in Figure 2 10B. Full conversion of the starting material was n oticed in Figure 2 10. A) Substitution of 1 bromoheptane with NaN 3 . B) Formation of triazole 2 11 under CuAAC conditions. approximately 1 hour by TLC, much faster than any of the p revious click experiments. The work up proved especially easy and yielded 90% of the expected product 2 11 as a white solid. Crude 1 H NMR studies indicated full conversion of the starting material and the product was the only material present (see Append ix C for crude spectrum). In comparison to earlier CuAAC attempts ( 2 8 and 2 10 ), the 1 H NMR spectrum of 2 11 did not show any unreacted starting material or minor unidentifiable by products in the alkyl A) B)
35 region. These results were what we had been hoping for to prove the successful marriage of click chemistry and BTF aminolysis chemistry. Purification of 2 11 was conducted by silica gel column chromatography to study the stability of the target under these conditions. Previous studies in our group have noted sensitivity of the lactones to silica with a mass loss around 20%. Using a mixture of hexanes and ethyl acetate, 40% of the mass loaded on the column was recovered. It is evident that some of the mass remained on the column which can be attributed to the sensitivity of the lactones to silica gel. Also, it is likely there are interactions between the silica gel and triazole ring as N2 and N3 of the triazole are slightly basic causing the compound to coordinate to the silica. The successful preparat ion of 2 11 spurred an interest to create a complementary structure through a homologated azide. The complementary structure would give a better understanding of the reactivity between linear azides and 2 6 , crude purity, and ease of purification. 1 Azido dodecane ( 2 13 ) was prepared (Figure 2 11A) 2 6 to yield triazole 2 14 (Figure 2 11B). Crude 1 H NMR analysis revealed good purity as the only other identifiable peaks belonged to unreacted azide. Presumably using a smaller amount of azide (1.2 vs. 5 equivalents) would give a cleaner crude product. Triazole 2 14 was purified by column chromatography in a mixture of hexanes and ethyl acetate to yield the pure product in 80 % yield. The increased recovery compared to 2 11 (4 0%) could be due to the longer alkyl chain present on the triazole affecting interactions with the silica gel.
36 Figure 2 11. A) Synthesis of 1 azidododecane. B) CuAAC to generate 2 14 . Finally, in the exploration of suitable azides, the use of a more water soluble unit was considered. Triethylene glycol monomethylether ( TGME ) was chosen because it has a manageable length and can be readily converted to the corresponding azide. T he ether was transformed to azide 2 15 as shown in Figure 2 12. The alcohol of TGME Figure 2 12. Synthesis of azide 2 15 : tosylation followed by substitution with NaN 3 . Was converted to a tosylate ( TsTGME ) to create a g ood leaving group. The tosylate was purified by column chromatography before being subjected to azide substitution (65% yield). Potassium iodide was added to the reaction to aid in the substitution process. Azide 2 15 was formed in good yield and no fur ther purification was performed. With the azide in hand, cycloaddition of 2 6 and 2 15 was conducted (Figure 2 13). The reaction proceeded smoothly and full conversion of the starting alkyne was noted after 2 hours. Compound 2 16 was recovered in 83% y ield after the work up and crude 1 H NMR showed only expected product and unreacted azide 2 15 . Attempts at purification proved unsuccessful; the product showed an R f of approximately 0.4 on TLC (10% MeOH/EtOAc) but did not elute from the column. The reac tion was run again A) B)
37 using only 1.5 equivalents of the azide relative to 2 6 and the crude product was spectroscopically clean ( 1 H and 13 C NMR). We are pleased by the results of this experiment as it proves that this methodology is applicable to a variety o f azides, with linear azides working exceedingly well. Figure 2 13. Synthesis of triazole 2 16 with a water soluble oligoethylene glycol chain. Post click Functionalizations With aspects of azide scope fleshed out, we remained curious about functionalization of the remaining two lactone rings. Along these lines, triazoles 2 11 , 2 14 , and 2 16 were functionalized with amines to yield the corresponding phloroglucinol products. It is important to note that the crude pur ity of 2 11 , 2 14 , and 2 16 was good enough for the compounds to be used in subsequent transformations as described in this section. Reactions where the triazole was used crude will be noted. Functionalization of 2 11 (used as crude) was carried out b y adding 4.0 equivalents of propargylamine to the difuranone compound (Figure 2 14). The reaction was allowed to stir overnight, and the crude product was recovered in 65% yield (the remaining mass was lost during work up). The crude purity was good as i ndicated by 1 H NMR but column chromatography was performed to yield the pure product 2 17 in 47%. However, comparison of the crude and pure 1 H NMR showed very few differences. A key point here is that these ring openings have excellent atom economy as th ere are no by products. Since CuAAC produces no by products and uses only
38 catalytic amounts of a Cu salt and NaAsc that are removed during work up, combining it with the ring openings makes for an overall very efficient synthetic sequence. Figure 2 14. Synthesis of 2 17 by aminolysis of remaining two lactone rings. A second example of functionalization of 2 11 was demonstrated through aminolysis with heptylamine (Figure 2 15). Addition of 3.0 equivalents of heptyl amine to 2 11 afforded the trisubstituted product 2 18 in 78% yield after column chromatography. Presumably, the additional solubilizing alkyl chains aided in purification as reflected in the higher yield in comparison to 2 17 . Figure 2 15. Functionalization through aminolysis with heptylamine. Compound 2 18 was alternatively synthesized in a reverse manner where CuAAC followed aminolysis (Figure 2 16). Synthesis began by adding 0.98 equivalents of propargylamine to BTF at 41 Â°C and allowing the reaction to stir for approximately one hour.
39 Figure 2 16. Synthesis of a difunctional molecule with a clickable alkyne handle. At this time, an excess (2.5 equivalents) of heptylamine was added to the reaction at low tem perature and stirring was continued while the reaction was warmed to room temperature. This molecule with a clickable handle ( 2 19 ) was purified by column chromatography to give the pure product in 59% yield. The alkyne then underwent cycloaddition with 1 azidoheptane to yield 2 18 in 54% yield after purification. This method does yield the product through a two pot procedure instead of three as in the previous example. Still, the two synthetic routes yield the product with the same high purity and are otherwise identical. The first synthetic route to 2 18 (Figure 2 15) afforded the final product in 54% yield (with no purification of 2 11 before aminolysis) while the second route (Figure 2 16) afforded 2 18 in 32% yield. The synthetic route shown in Fi gure 2 16 was a shorter and more direct method of accessing 2 18 although it did result in lower overall yield. This route would be preferable over the former for tethering the phloroglucinol derivatives to larger systems such as surfaces or polymers. Th e route shown in Figure 2 15 would be desired when constructing the phloroglucinol derivatives to be used as individual, functional units.
40 This same reaction sequence was performed with the dodecyl derivative 2 14 (Figure 2 17). Heptylamine (3.0 equival ents) was added to 2 14 (used without further purification) and allowed to stir overnight. Functional molecule 2 20 was synthesized in good yield and acceptable crude purity; there was no evidence of unreacted lactones. Column chromatography was performe d and the pure product was recovered in 52%. The low yield of the phloroglucinol derivatives is primarily linked to purification and is not a reflection of conversion as crude 1 H NMR spectra of the phloroglucinol derivatives display high percentage (> 95% ) of conversion. The next, complementary reaction in the sequence utilized furfurylamine (Figure 2 18). This amine has been previously shown to work well as a nucleophile for the Figure 2 17. Aminolysis with heptylami ne to yield 2 20 . aminolysis of BTF . In addition, one can imagine using such a scaffold to execute two consecutive click reactions as furan is a well known diene for the Diels Alder cycloaddition. Starting with 2 14 , 3.0 equivalents of furfurylamine we re added to the reaction mixture and allowed to stir overnight (Figure 2 18). The crude material was isolated in 85% yield after work up but column chromatography proved difficult. The pure product 2 21 could only be isolated in 22% yield,
41 Figure 2 18. Addition of furfurylamine to yield the trisubstituted product. about 30% less than other target compounds in this series. This is perhaps due to the nature of furfurylamine as use of this amine for other functionali zed BTF targets resulted in lower yields after purification in comparison to those that are more soluble (e.g. heptylamine). 48 As a final example of a difunctional system, 2 16 was functionalized with heptylamine to form a compound with both hydrophilic a nd hydrophobic moieties (Figure 2 19). Incorporating both components is advantageous because this methodology could then be employed in dendrimer or polymer applications where amphiphilicity is desired. Addition of 2.5 equivalents of heptylamine to 2 16 (used without further purification, see Appendix C for crude spectrum) resulted in the desired target 2 22 in 65% crude yield as the remaining mass was lost during work up. Analysis of the crude proton NMR revealed full conversion of 2 16 as there was no evidence of remaining lactones. Like the parent compound 2 16 , 2 22 could not be purified by column chromatography (only crude NMR given in Appendix C). Current efforts are directed toward better purification strategies of oligoethylene glycol triazole d erivatives.
42 Figure 2 19. Synthesis of a functional molecule with both hydrophilic and hydrophobic chains. Finally, BTF was desymmetrized to yield a trifunctional molecule with an alkyne handle through sequential ring openings. The first example of a trifunctional molecule begins with sequential aminolysis of BTF with propargylamine, heptylamine, and isopropylamine (Figure 2 20). This reaction sequence gave the trifunctional molecule 2 23 in 91% crude yield. There w ere other substituted products observed in the crude 1 H NMR, but these compounds were present in very small amounts. After column chromatography, no other substituted products were observed by NMR spectroscopy and the product was recovered in 48% yield. Figure 2 20. Synthesis a trifunctional system through sequential aminolysis of BTF followed by CuAAC with 1 azidododecane.
43 The trifunctional molecule 2 23 was then subjected to CuAAC con ditions (10% CuSO 4 , NaAsc, THF, H 2 O) to yield 2 24 in quantitative yield. After column chromatography, the pure product was recovered in 52% yield. Alternatively, a trifunctional molecule was prepared through sequential aminolysis of 2 11 (Figure 2 21). Clicked compound 2 11 was functionalized with heptylamine and isopropyl amine to yield trifunctional compound 2 25 . Crude 1 H NMR spectrum revealed minor side products that are attributed to compounds that were substituted with only heptylamine or isopropy lamine. Figure 2 21. Synthesis of a trifunctional molecule through aminolysis of 2 11 . Comparison of the examples of post click functionalization reveals that the final ring openings proceed predictably and give the prod uct in high yield and acceptable crude purity. Adding to the merits of these reactions (ring openings) is that they do not have to be run under inert conditions. The reactions are conducted with dry DMF and under a blanket of argon, but flame drying the reaction flask and other precautions are not necessary. Column chromatography gives the pure phloroglucinol product in 50 60% yield (on average). However, the identity of the amine does seem to make a difference as to the ease of purification and overa ll final yield. Amines that are more solubilizing, such as heptylamine, prove easier to purify. The phloroglucinol derivatives are generally more difficult to purify than their less substituted lactone precursors as has been observed in previous reaction s. Phloroglucinol derivatives that also have triazole
44 units become more difficult to remove from the column so adding solubilizing alkyl chains aids in purification. Nonetheless, a variety of primary amines can be used for this functionalization as the r eaction always nears 100% conversion. Attempts at Triple Clicked Motifs Having made 2 5 (Figure 2 6), we were also interested in synthesizing triple clicked motifs where BTF would be sequentially functionalized with propargylamine and then subjected to Cu AAC conditions resulting in a molecule ( 2 26 ) with three, distinct Figure 2 22. General synthesis of a trifunctional molecule with three different clicked units. units (Figure 2 22). The first (and easiest) approach to test the methodology would be to follow the synthesis of 2 27 by addition of excess propargylamine and CuAAC using a second, different azide to give 2 28 (Figure 2 23). Figure 2 23. S ynthesis of a fully substituted phlor oglucinol derivative bearing two different clicked units.
45 To initiate this sequence, compound 2 8 was reacted with 3.0 equivalents of propargylamine (Figure 2 24) to provide 2 29 in 90% yield and good purity (no further purification was performed). To th e alkyne 2 29 was added a large excess (10 equivalents) of 2 ethylhexylazide to give the final product. This reaction series proceeded smoothly and was used to create two different functionalized phloroglucinol compounds, one using 1 azidohexane ( 2 30 ) an d the other using 2 ethylhexylazide ( 2 31 ) (Figure 2 24). Figure 2 24. Attempted synthesis of tristriazoles 2 30 and 2 31 . For these particular compounds, we were not as lucky as with 2 5 and simple filtration would not yield the desired product. As previously stated, purification of phloroglucinol derivatives is notoriously difficult and several purification techniques were attempted. First, purification using a silica gel plug was unsuccessful. The recovered product was unidentifiable by 1 H NMR; the triazole proton peaks, as well as the alpha methylene peaks, were not readily detected. If the compound is decomposing, it is difficult to say exactly how. A neutral alumina column was attempted next; however, the produ ct could not be recovered from the column. Given that these compounds were so difficult to purify, we looked at similar systems containing three triazole rings to get a better understanding of their reactivity
46 and purification methods. Indeed, there are numerous papers that report using tristriazole ligands (and related species) in CuAAC. 54 57 Shown in Figure 2 25 are two common examples of tristriazoles used in the catalysis of CuAAC. These ligands do coordinate to the aliphatic amine which is a dist inction between these ligands and our systems. It seems likely that having three triazole rings on the phloroglucinol core could Figure 2 25. Two tristriazoles used in catalysis for CuAAC. TBTA 2 32 (left) and water so luble ligand 2 33 (right). aid in copper chelation which inhibits purification, particularly in the case of 2 31 where 3.0 equivalents of CuI were used. Additionally, as stated earlier, the three basic triazole rings could have significant interactions wi th the slightly acidic silica gel. These interactions coupled with the already difficult purification of the phloroglucinol derivatives can likely explain the inability to purify compounds 2 30 and 2 31 . Overview and Discussion A series of compounds wer e synthesized through CuAAC and aminolysis to yield BTF and phloroglucinol derivatives with varying functionalities. Starting with alkyne 2 7 , azide scope was studied using CuSO 4 and NaAsc (sacrificial reducing agent) in a mixture of THF/H 2 O to yield the respective triazoles. While benzylazide and 2 ethylhexylazide gave the desired product, crude purity was low requiring purification and resulting in diminished overall yield. However, linear azides proved to work exceptionally well. Crude purity was ver y high as 1 H NMR showed only the desired
47 target with a few baseline impurities (e.g. residual azide). These compounds ( 2 11 and 2 14 ) could be readily purified by column chromatography in acceptable yield. However, purification was not necessary before f unctionalization of the final two lactone rings for 2 11 , 2 14 , and 2 15 . Functionalization of the difuranone compounds proved to be especially straightforward. A variety of amines proceeded in the aminolysis with good reliability; heptylamine, propargy lamine, and furfurylamine were all used for this step. The trifunctional compounds could easily be purified by column chromatography and recovered in 50 60% yield. Furfurylamine derivative 2 21 was more difficult to purify but was recovered in high pur ity. This reproducibility and predictability in post click functionalization is important and strengthens the case for BTF to be used as a scaffold for multiple applications. Trifunctional derivatives of phloroglucinol were synthesized by two different routes. The first route included trifunctionalization of BTF through sequential aminolysis with propargylamine, heptylamine, and isopropylamine in a one pot procedure. The trifunctional molecule 2 23 was then subjected to CuAAC with 1 azidododecane to yi eld 2 24 in 52% yield after purification. In an opposite manner, 2 11 was further functionalized by sequential aminolysis with heptylamine and isopropylamine. Lastly, attempts at trifunctional molecules with three different triazoles were unsuccessful. The methodology applied worked well as the first steps could be completed and the products purified. However, the final CuAAC was problematic and purification was impossible. Although accessing and studying targets of that nature
48 would be interesting, tr ifunctional targets can still be achieved using the BTF system through the previously described methodologies. Experimental Methods General Procedure . All chemicals were purchased from Aldrich or Acros and used without further purification unless otherwi se noted. DMF was purified using a Glass Contour solvent system (Glass Contour, Inc.), degassed in 20 L drums, and passed through two purification columns (molecular sieves) under an argon atmosphere. Thin Layer Chromatography (TLC) was performed using D ynamic Adsorbents, Inc. aluminum backed plates. The plates were developed using UV light and ninhydrin staining. Flash column chromatography was performed using Purasil SiO 2 60 230 400 mesh silica gel from Whatman. A Varian Inova spectrometer was used t o record 1 HNMR and 13 CNMR spectra; spectra were collected at 500 MHz for 1 HNMR and 125 MHz for 13 CNMR. Chemicals shifts ( ) were referenced to the appropriate deuterated solvent (DMSO d 6 : H 2.50 ppm, C 39.50; CDCl 3 : H 7.26 ppm, C 77.16 ppm). Abbreviations used are as follows: s (singlet), d (doublet), t (triplet), q (quartet), and m (multiplet). Accurate mass experime nts were performed on an Agilent 6220 ESI TOF (Santa Clara, CA) mass spectrometer equipped with an electrospray and DART source operated in positive ion mode. The following compounds were synthesized according to literature procedure: BTF , 48,58 (azidomet hyl) benzene ( 2 7 ), 59 2 ethylhexylazide, 60 1 azidoheptane, 61 1 azidododecane, 62 and TsTGME . 63
49 Procedures 2,2',2'' (2,4,6 Trihydroxybenzene 1,3,5 triyl)tris(N (prop 2 yn 1 yl)acetamide) (2 4): BTF (0.05 g, 0.20 mmo l) and DMF (3 mL) was added to a 25 mL round bottom flask. After placing reaction vessel on a cold bath ( 41 Â°C) for 15 min., propargylamine (0.046 mL, 0.71 mmol) was added via syringe. The reaction was allowed to stir in a cold bath for 3 h at which tim e it was warmed to rt and stirred for 16 h. Ethyl acetate (75 mL) was added and the organic layer was washed with water (3Ã— 15 mL) and brine (1 Ã— 15 mL). The organic layer was dried over Na 2 SO 4 , filtered, and then concentrated under reduced pressure. Th e resulting crude mixture was purified by column chromatography (15% Hex/EtOAc) to give the desired product (0.052 g, 63%). 1 H NMR (DMSO d 6 ) 3.12 (s, 3H), 3.49 (s, 6H), 3.86 (dd, 6H, J = 2.7 and 5.7 Hz), 8.75 (t, 3H, J = 5.0 Hz), 9.97 (s, 3H) ppm. 13 C NMR (DMSO d 6 ) 28.3, 30.9, 73.3, 80.6, 102.7, 153.7, 173.1 ppm. HRMS (ESI) calculated for C 21 H 21 N 3 O 6 [M+Na] + 434.1323, found 434.1335.
50 2,2',2'' (2,4,6 Trihydroxybenzene 1,3,5 triyl)tris(N ((1 hexyl 1H 1,2,3 triazol 4 yl)methyl)acetamide) (2 5) : To a 10 mL RBF charged with a stir bar was added 2 4 (0.057 g, 0.14 mmol) dissolved in THF (2 mL). 1 Azidohexane (0.266 g, 2.09 mmol) was added to the reaction vessel followed by water (2 mL) and the reaction mixture was allow ed to stir for 5 min. At this time, CuSO 4 Â·5H 2 O (0.0052 g, 0.021 mmol) and NaAsc (0.0083 g, 0.042 mmol) was added to the flask. The reaction stirred at rt for 4 h before begin poured in to brine (75 mL). The aqueous layer was extracted with ether (3 Ã— 25 mL). The organic layers were combined and washed with brine (2 Ã— 25 mL) then dried over Na 2 SO 4 . The organic layer was filtered and the product was collected as a light brown solid (0.063 g, 57%). 1 H NMR (DMSO d 6 ) 0.83 (t, 6H, J = 6.5 Hz ), 1.24 (bs, 18H), 1.76 (p, 6H, J = 6.9 Hz), 3.51 (s, 6H), 4.29 (t, 6H, J = 5.1 and 6.8 Hz), 7.89 (s, 3H), 8.78 (t, 3H, J = 5.0 Hz), 10.14 (s, 3H) ppm. 13 C NMR (DMSO d 6 ) 14.3, 22.4, 25.9, 30.2, 31.0, 31.6, 35.1, 49.7, 103.4, 110.0, 123.3, 154.2, 173.8 ppm. HRMS (ES I) calculated for C 39 H 60 N 12 O 6 [M+Na] + 815.4651, found 815.4644.
51 2 (4 Hydroxy 2,7 dioxo 2,3,7,8 tetrahydrobenzo[1,2 b:3,4 b']difuran 5 yl) N (prop 2 yn 1 yl)acetamide (2 6) : To a 25 mL RBF was added BTF (0.060 g, 0.24 mmol) and dry DMF (2 mL). The reaction vessel was subsequently cooled to 41 Â°C and the solution was allowed to stir for 15 min. After this time, propargylamine (0.016 mL, 0.24 mmol) was added to the vessel. The reaction stirred under argon for 1 h at w hich time the TLC showed no starting material. The reaction was allowed to warm to rt and then was poured in to 75 mL of EtOAc. The organic layer was washed with water (5 Ã— 20 mL) and brine (1 Ã— 20 mL) and dried over Na 2 SO 4 . The organic layer was concen trated and purified by flash chromatography (10% acetone/DCM) to yield the pure product (0.0537 g, 77%). 1 H NMR (DMSO d 6 ) 3.09 (t, 1H, J = 4 Hz) , 3.47 (s, 2H), 3.77 (s, 2H), 3.43 (dd, 2H, J = 4.5 and 9.5 Hz), 3.90 (s, 2H), 8.49 (t, 1H, J =8.5 Hz), 10.22 (s, 1H) ppm. 13 C NMR (DMSO d 6 ) 28.6, 30.7, 31.1, 31.5, 73.6, 71.5, 102.9, 105.4, 148.3, 153.9, 169.9, 174.3, 174.6 ppm. HRMS (ESI) calculated for C 15 H 11 NO 6 [M+H] + 302.0659, found 302.0660. N ((1 Benzyl 1H 1,2,3 triazol 4 yl)methyl) 2 (4 hydroxy 2,7 dioxo 2,3,7,8 tetrahydrobenzo[1,2 b:3,4 b']difuran 5 yl)acetamid e (2 8): To a solution of 2 6
52 (0.060 g, 0.19 mmol) in 3 mL THF was added (azidomethyl)benzene (0.39 g, 3.0 mmol) and 2 mL water. The reaction was allowed to stir at rt before adding CuSO 4 Â· 5H 2 O (0.0074 g, 0.029 mmol) and sodium ascorbate (0.012 g, 0.059 mmol). The reaction wa s stirred at rt for 3 h before being poured in to brine (75 mL). The aqueous layer was extracted with EtOAc (4 Ã— 25 mL). The organic layers were combined and washed with brine (2 Ã— 20 mL), dried over Na 2 SO 4 , filtered, and concentrated. The crude product was purified by column chromatography (10% acetone/DCM) to afford the product (0.056 g, 65%). 1 H NMR (DMSO d 6 ) 3.44 (s, 2 H), 3.74 (s, 2 H), 3.87 (s, 2 H), 4.26 (d, 2 H, J = 5.2 Hz), 5.51 (s, 2 H), 7.30 (m, 5 H), 7.92 (s, 1 H), 8.55 (t, 1 H, J = 4.9 Hz), 10.29 (s, 1 H) ppm. 13 C NMR (DMSO d 6 ) 30.4, 30.6, 30.9, 34.5, 42.1, 52.7, 97.1, 102.7, 105.0, 123.0, 127.9, 128.1, 128.7, 136.1, 144.9, 147.8, 151.5, 153.4, 169.9, 173.9, 174.1, 174.1 ppm. HRMS (ESI) calculated for C 22 H 18 N 4 O 6 [M+H] + 435.1299, found 435.1319. N ((1 (2 Ethylhexyl) 1 H 1,2,3 triazol 4 yl)methyl) 2 (4 hydroxy 2 ,7 dioxo 2,3,7,8 tetrahydrobenzo[1,2 b:3,4 b']difuran 5 yl)acetamide (2 10) : To a 10 mL RBF was added 2 6 (0.072 g, 0.24 mmol), THF (2.5 mL), 1 ethylhexylazide 2 9 (0.15 g, 1.2 mmol), and water (1.5 mL). The reaction was allowed to stir for 5 min before adding CuSO 4 Â·5H 2 O (0.009 g, 0.04 mmol) and NaAsc (0.014 g, 0.072 mmol). The reaction
53 stirred at rt for 16 h at which time TLC showed full conversion of starting material. The reaction was poured into brine (100 mL) and extracted with EtOAc (3 Ã— 30 mL). The organic layers were combined, dried over Na 2 SO 4 , and concentrating, The crude material was purified by column chromatography (10% acetone:CH 2 Cl 2 ) to afford the pure material (0.038 g, 37%). 1 H NMR (DMSO d 6 ) 0.86 (m, 6H), 1.20 (m, 8H), 1.80 (m, 1H), 3.49 (s, 2H), 3.78 (s, 2H), 3.92 (s, 2H), 4.22 (d, 2H, J = 6.5 Hz), 4.31 (s, 2H), 7.87 (s, 1H), 8.61 (t, 1H, J = 5 Hz), 10.35 (s, 1H) ppm. 13 C NMR (DMSO d 6 ) 10.2, 13.8, 22.4, 23.1, 27.8, 29.7, 30.5, 30.6, 31.0 , 34.6, 39.6, 52.4, 97.2, 102.7, 105.1, 123.3, 144.4, 147.8, 151.5, 153.5, 169.9, 173.9, 174.1 ppm. HRMS (ESI) calculated for C 23 H 28 N 4 O 6 [M+H] + 457.2082, found 457.2090. N ((1 Heptyl 1 H 1,2,3 triazol 4 yl)methyl) 2 (4 h ydroxy 2,7 dioxo 2,3,7,8 tetrahydrobenzo[1,2 b:3,4 b']difuran 5 yl)acetamide (2 11): To a 10 mL RBF was added 2 6 (0.067 g, 0.22 mmol), THF (2 mL), water (1 mL), and 1 azidoheptane (0.31 g, 2.2 mmol). The reaction was allowed to stir for 5 min before add ing CuSO 4 (1M, 0.023 mL, 0.023 mmol) and NaAsc (1M, 0.068 mL, 0.068 mmol). The reaction stirred for 1 h at which time TLC showed complete conversion of the starting material. The reaction was poured into brine (100 mL) and extracted with EtOAc (3 Ã— 25 mL ). The organic layers were combined, dried over Na 2 SO 4 , and concentrated to afford a crude gray solid (90%) with good purity. The crude solid was purified by column chromatography (25:75 Hex:EtOAc) to yield the pure product (0.016 g, 36%). 1 H NMR (DMSO d 6 ) 0.85 (t, 3H, J = 6.5 Hz), 1.25 (m, 10H), 1.78 (m, 2H, J = 7 Hz), 3.49 (s, 2H), 3.79 (s, 2H), 3.92 (s, 2H), 4.30 (m, 4H), 7.89 (s, 1H), 8.60 (t, 1H, J = 5.3 Hz), 10.3 (s, 1H) ppm. 13 C NMR
54 (DMSO d 6 ) 13.9, 21.9, 25.8, 28.0, 29.8, 30.4, 30.6, 31.0, 31 .7, 34.6, 49.3, 97.2, 102.7, 105.1, 122.7, 144.5, 147.8, 151.5, 153.5, 169.9, 173.9, 174.2 ppm. HRMS (ESI) calculated for C 22 H 26 N 4 O 6 [M+H] + 443.1925, found 443.1924. N ((1 Dodecyl 1 H 1,2,3 triazol 4 yl)methyl) 2 (4 hydr oxy 2,7 dioxo 2,3,7,8 tetrahydrobenzo[1,2 b:3,4 b']difuran 5 yl)acetamide (2 14): To a 10 mL RBF was added 2 6 (0.025 g, 0.083 mmol), 1 azidododecane (0.088 g, 0.42 mmol), THF (1.5 mL), and water (1 mL). The solution stirred for 5 min and then CuSO 4 (0.3 3M, 0.025 mL, 0.0083 mmol) and NaAsc (1.0M, 0.025 mL, 0.025 mmol) were added. The solution stirred for 2 h at which time TLC indicated complete conversion of the starting alkyne. The reaction mixture was poured into brine (100 mL) and extracted with EtOA c (4 Ã— 25 mL). The organics were combined, dried over Na 2 SO 4 , and concentrated to yield a gray solid. The solid was then purified by column chromatography (1:1 Hex:EtOAc) to yield the pure product (0.034 g, 80%). 1 H NMR (DMSO d 6 ) 0.85 (t, 3H, J = 5 Hz), 1.24 (m, 18H), 1.76 (m, 2H, J = 5 Hz), 3.49 (s, 2H), 3.79 (s, 2H), 3.92 (s, 2H), 4.31 (m, 4H), 7.89 (s, 1H), 8.60 (t, 1H, J = 5 Hz), 10.3 (s, 1H) ppm. 13 C NMR (DMSO d 6 ) 13.9, 22.1, 22.4, 23.2, 25.9, 28.7, 28.9, 29.0, 29.8, 30.4 , 30.6, 31.0, 31.3, 34.6, 49.24, 97.1, 102.7, 105.0, 122.7, 144.5, 147.8, 151.5, 153.5, 169.9, 173.8, 174.1 ppm. HRMS (ESI) calculated for C 27 H 36 N 4 O 6 [M+Na] + 535.2527, found 535.2532. 1 Azido 2 (2 (2 methoxyethoxy)et hoxy)ethane (2 15): To a solution of TsTGME (1.25 g, 3.93 mmol) in DMSO (20 mL) was added NaN 3 (0.55 g, 8.5 mmol) and KI (0.080 g, 0.48 mmol). The solution stirred for 36 h at which time the TLC showed full
55 conversion. The reaction was poured into wate r (30 mL) and extracted with ether (4 x 20 mL). The organic layers were combined, dried over MgSO 4 , and concentrated to yield the pure compound (0.71 g, 96%). 1 H NMR (CDCl 3 ) 3.37 (m, 5H), 3.54 (m, 2 H), 3.64 (m, 8H) ppm. 13 C NMR (CDCl 3 ) 50.7, 59.1, 70.1, 70.7, 70.7, 70.8, 72.0 ppm. HRMS (ESI) calculated for C 7 H 15 N 3 O 3 [M+Na] + 212.1006, found 212.1013. 1 H NMR spectrum matched that reported by Ramakrishnan. 63b 2 (4 Hydroxy 2,7 dioxo 2,3,7,8 tetrahydrobenzo[1,2 b:3 ,4 b']difuran 5 yl) N ((1 (2 (2 (2 methoxyethoxy)ethoxy)ethyl) 1H 1,2,3 triazol 4 yl)methyl)acetamide (2 16): To 2 6 (0.025 g, 0.083 mmol) was added THF (2 mL), water (1 mL), and azide 2 15 (0.024 g, 0.13 mmol). The solution was allowed to stir for 5 min before the addition of CuSO 4 (0.33M, 0.025 mL, 0.0083 mmol) and NaAsc (1M, 0.025 mL, 0.025 mmol). The solution stirred at rt until TLC showed full conversion of starting material (~2 hours). The reaction was then poured into brine (75 mL) and extracted with EtOAc (4 Ã— 25 mL). The organic layers were combined, dried over Na 2 SO 4 , and concentrated to yield the desired product. 1 H NMR (DMSO d 6 ) 3.22 (s, 3H), 3.40 (m, 2H), 3.47 (m, 4H), 3.51 (m, 6H), 3.80 (m, 4H), 3.92 (s, 2H), 4.32 (d, 2H, J = 10 Hz), 4.49 (t, 2H, J = 5 Hz), 7.89 (s, 1H), 8.62 (t, 1H, J = 5 Hz), 10.37 (s, 1H) ppm. 13 C NMR (DMSO d 6 ) 30.5, 30.6, 31.0, 34.5, 49.3, 58.0, 68.7, 6 9.5, 69.6, 71.2, 97.2, 102.7, 105.1, 123.2, 144.4, 147.8, 151.5, 153.4, 169.9, 173.9, 174.2 ppm. HRMS (ESI) calculated for C 22 H 26 N 4 O 9 [M+Na] + 513.1592, found 513.1606.
56 2,2' (5 (2 (((1 Heptyl 1 H 1,2,3 triazol 4 yl)meth yl)amino) 2 oxoethyl) 2,4,6 trihydroxy 1,3 phenylene)bis(N (prop 2 yn 1 yl)acetamide) (2 17): To a solution of 2 6 (0.080 g, 0.18 mmol) in dry DMF (2 mL) was added propargylamine (1M, 0.73 mL, 0.73 mmol). The reaction stirred at rt for 20 h. The reactio n was then diluted with EtOAc and washed with water (5 Ã—20 mL) and brine (1 Ã— 20 mL). The organic layer was then dried over Na 2 SO 4 and concentrated to yield a crude yellow solid. The crude material was purified by column chromatography ( R f = 0.75; EtOAc) to yield the pure yellow solid (0.03 g, 31%). 1 H NMR (DMSO d 6 ) 0.85 (t, 3H, J = 5 Hz), 1.23 (m, 8H), 1.78 (m, 2H, J = 5 Hz), 3.12 (t, 2H, J = 5 Hz), 3.49 (s, 4H), 3.51 (s, 2H), 3.86 (dd, 4H, J = 4.5 and 9.0 Hz), 4.31 (m, 4H), 7.91 (s, 1H), 8.73 (t, 2H, J = 5 Hz), 8.82 (t, 1H, J = 5 Hz), 9.96 (s, 1H), 10.1 (s, 2H) pp m. 13 C NMR (DMSO d 6 ) 13.9, 21.9, 25.8, 28.0, 28.3, 29.7, 30.9, 31.0, 34.6, 49.3, 73.3, 80.6, 102.8, 102.8, 122.8, 143.9, 153.7, 173.1, 173.4 ppm . HRMS (ESI) calculated for C 28 H 36 N 6 O 6 [M+H] + 553.2769, found 553.2774. 2,2' (5 (2 (((1 Heptyl 1 H 1,2,3 triazol 4 yl)methyl)amino) 2 oxoethyl) 2,4,6 trihydroxy 1,3 phenylene)bis(N heptylacetamide) (2 18): To a solution of 2 6 (0.016 g, 0.036 mmol) in dry DMF (2 mL) was added heptylamine (0.016 mL, 0.11 mmol). The solution s tirred under a blanket of argon for 20 h. The reaction mixture was diluted with EtOAc, washed with water (5 Ã— 20 mL), and brine (1 Ã— 20 mL). The organic layer was
57 dried over Na 2 SO 4 and then concentrated to yield a pale yellow solid. The crude material w as purified by column chromatography (25:75 Hex:EtOAc) to yield the desired product (0.019 g, 78%). Alternative procedure: 2 19 (0.036 g, 0.069 mmol) and 1 azidoheptane (0.048 g, 0.34 mmol) was dissolved in THF (1.5 mL) and water (1 mL). The solution s tirred for 5 min before CuSO 4 (0.33M, 0.021 mL, 0.0070 mmol) and NaAsc (1.0M, 0.021 mL, 0.021 mL) were added to the reaction mixture. The resulting solution stirred for 1 h before TLC showed full consumption of the starting alkyne. The reaction was poure d into brine (75 mL) and extracted with EtOAc (4 Ã— 25 mL). The organic layers were combined, dried over Na 2 SO 4 , and concentrated to yield the product. The crude material was purified by column chromatography (25:75 Hex:EtOAc) to yield the desired product (0.025 g, 54%). 1 H NMR (DMSO d 6 ) 0.85 (t, 9H, J = 5 Hz), 1.25 (m, 24H), 1.41 (m, 4H, J = 5 Hz), 1.77 (m, 2H, J = 5 Hz), 3.04 (q, 4H, J = 5 Hz), 3.48 (s, 4H), 3.51 (s, 2H), 4.30 (m, 4H), 7.88 (s, 1H), 8.45 (t, 2H, J = 5 Hz), 8.82 (t, 1H, J = 5 Hz), 10.5 (s, 2H), 10.6 (s, 1H) ppm. 13 C NM R (DMSO d 6 ) 13.9, 13.9, 21.9, 22.0, 25.8, 26.3, 28.0, 28.3, 28.7, 29.7, 31.1, 31.2, 31.3, 34.6, 38.9, 49.2, 102.8, 122.8, 144.1, 153.8, 173.4, 173.7 ppm. HRMS (ESI) calculated for C 36 H 60 N 6 O 6 [M+Na] + 695.4467, found 695.4488. 2,2' (2,4,6 Trihydroxy 5 (2 oxo 2 (prop 2 yn 1 ylamino)ethyl) 1,3 phenylene)bis(N heptylacetamide) (2 19): A reaction flask with BTF (0.025 g, 0.10 mmol) dissolved in dry DMF (2 mL) was cooled to 41 Â°C. After the solution was sufficiently cool ed, propargylamine (1.0M, 0.098 mL, 0.098 mmol) was added to the flask. The reaction stirred for 1.5 h at which time TLC showed conversion of BTF . Heptylamine (0.038 g, 0.26 mmol) was added to the reaction vessel and subsequently warmed to rt and stirred overnight. The reaction mixture was diluted with EtOAc then washed with water (5 Ã— 20 mL) and brine (1 Ã— 20 mL). The organic layer was dried over Na 2 SO 4 and concentrated
58 to yield a crude brown material. The crude solid was purified by column chromatogr aphy ( R f = 0.65; 1:1 Hex:EtOAc) to yield the pure compound (0.032 g, 59%). 1 H NMR (DMSO d 6 ) 0.85 (t, 6H, J = 5 Hz), 1.26 (m, 16H), 1.40 (m, 4H, J = 5 Hz), 3.04 (q, 4H, J = 5Hz), 3.11 (t, 1H, J = 5 Hz), 3.48 (s, 4H), 3.49 (s, 2H), 3.86 (dd, 2H, J = 4.5 and 9.5 Hz), 8.46 (t, 2H, J = 5 Hz), 8.77 (t, 1H, J = 5 Hz), 10.4 (s, 2H), 10.7 (s, 1H) ppm. 13 C NMR (DMSO d 6 ) 13.93, 22.0, 26.3, 28.3, 28.4, 28.7, 30.1, 31.2, 31.3, 38.9, 73.3, 80.6, 102.8, 102.9, 153.8, 173.3, 173.7 ppm. HRMS (ESI) calculated for C 29 H 45 N 3 O 6 [M+H] + 532.3391, found 532.3381. 2,2' (5 (2 (((1 D odecyl 1 H 1,2,3 triazol 4 yl)methyl)amino) 2 oxoethyl) 2,4,6 trihydroxy 1,3 phenylene)bis(N heptylacetamide) (2 20): To a solution of 2 14 (0.040 g, 0.078 mmol) in dry DMF (1.5 mL) was added heptylamine (0.035 mL, 0.027 mmol). The solution was allowed to stir at rt overnight at which time TLC showed full conversion. The reaction mixture was diluted with EtOAc (75 mL) and washed with water (5 Ã— 20 mL) and then brine (1 Ã— 20 mL). The crude material was purified by column chromatography (1:1 Hex:EtOAc) to afford the pure product (0.03 g, 52%). 1 H NMR (DMSO d 6 ) 0.84 (t, 9H, J = 5 Hz), 1.23 (m, 34H), 1.40 (m, 4H), 1.76 (m, 2H, J = 5 Hz), 3.04 (q, 4H, J = 5 Hz), 3.48 (s, 4H), 3.50 (s, 2H), 4.29 (m, 4H), 7.88 (s, 1H), 8.45 (t, 2H, J = 5 Hz), 8.81 (t, 1H, J = 5 Hz), 10.5 (s, 2H), 10.6 (s, 1H) ppm. 13 C NMR (DMSO d 6 ) 13.9, 13.9, 22.0, 22.1, 25.9, 26.3, 28.3, 28.4, 28.7, 28.7, 28.9, 29.0, 29.0, 29.7, 31.2, 31.3, 49.2, 102.9, 122.7, 144.1, 153.8, 173.4, 173.6 ppm. HRMS (ESI) calculated for C 41 H 70 N 6 O 6 [M+H] + 7 43.5430, found 743.5427.
59 2,2' (5 (2 (((1 Dodecyl 1 H 1,2,3 triazol 4 yl)methyl)amino) 2 oxoethyl) 2,4,6 trihydroxy 1,3 phenylene)bis(N (furan 2 ylmethyl)acetamide) (2 21): To a solution of 2 14 (0.042 g, 0.082 mmol) in d ry DMF (1.5 mL) was added furfurylamine (0.022 mL, 0.25 mmol). The solution was allowed to stir overnight under a blanket of argon. At that time, the reaction mixture was diluted with EtOAc (75 mL) and washed successively with water (5 Ã— 20 mL) and brine (1 Ã— 20 mL). The organic layer was dried over Na 2 SO 4 and concentrated to yield a yellow, sticky solid. This crude material was purified by column chromatography ( R f = 0.45; 25:75 Hex:EtOAc) to give the pure product as a yellow solid (0.013 g, 22%). 1 H NMR (DMSO d 6 ) 0.85 (t, 3H, J = 5 Hz), 1.23 (m, 18H), 1.77 (m, 2H, J = 5 Hz), 3.51 (s, 4H), 3.52 (s, 2H), 4.28 (m, 8H), 6.25 (dd, 2H, J = 1 and 3.2Hz), 6.38 (dd, 2H, J = 1.8 and 3.2 Hz), 7.57 (s, 2H), 7.90 (s, 1H), 8.80 (m, 3H), 10.1 (s, 1H), 10.2 (s, 2H) ppm. 13 C NMR (DMSO d 6 ) 13.9, 22.1, 22.4, 23.2, 25.9, 28.4, 28.7, 28.9, 28.9, 29.0, 29.7, 31.1, 31.3, 34.6, 35.8, 38.1, 49.3, 102.9, 107.2, 110.5, 122.8, 142.2, 144.0, 142.2, 144.0, 151.6, 153.7, 153.8, 173.3, 173.4 ppm. HRMS (ESI) calculated for C 37 H 50 N 6 O 8 [M+Na] + 729.3582, found 729.3586. N Heptyl 2 (2,4,6 trihydroxy 3 (2 (isopropylamino) 2 oxoethyl) 5 (2 oxo 2 (prop 2 yn 1 ylamino)ethyl)phenyl)acetamide (2 23): To a 10 mL RBF was added BTF (0.10 g, 0.41 mmol) and 4 mL dry DMF . The solution was allowed to cool to 41 Â°C (dry ice/acetonitrile) for 15 min before the addition of propargylamine (1M, 0.41 mL, 0.41
60 mmol). The solution stirred at low temperature for 1.5 h before TLC indicated full conversion of starting material. A t this time, heptylamine (1M, 0.41 mL, 0.41 mmol) was added to the reaction solution and stirred at 41 Â°C for 10 h when TLC indicated full conversion. Isopropylamine (1M, 0.81 mL, 0.81 mmol) was added and the solution was allowed to slowly warm to rt and stir overnight. The reaction was poured in to EtOAc (75 mL) and washed with water (6 Ã— 30 mL) and brine (1 Ã— 25 mL). The organic layers were combined and dried over Na 2 SO 4 , filtered, and concentrated to yield a crude yellow solid. The solid was purifie d by column chromatography ( R f = 0.7, 1:1 EtOAc:Hex) to yield a pure yellow solid (0.092 g, 48%). 1 H NMR (DMSO d 6 ) 0.85 (t, 3H, J = 7 Hz), 1.07 (d, 6H, J = 6.5 Hz), 1.23 (m, 9H), 1.40 (m, 2H, J = 7.2 Hz), 3.04 (q, 2H, J = 6.4 Hz), 3.12 (t, 1H, J = 2.5 H z), 3.47 (s, 2H), 3.48 (s, 2H), 3.49 (s, 2H), 3.82 (m, 1H, J = 7.0 Hz), 3.86 (dd, 2H, J = 5.4, 2.3 Hz), 8.45 (m, 2H), 8.78 (t, 1H, J = 5.4), 10.4 (s, 1H), 10.5 (2, 1H), 10.7 (s, 1H) ppm. 13 C NMR (DMSO d 6 ) 13.9, 22.0, 22.1, 26.3, 28.3, 28.3, 28.6, 30.8, 3 1.2, 31.3, 31.4, 38.9, 40.9, 73.3, 80.6, 102.8, 102.8, 102.9, 153.8, 153.8, 153.8, 172.9, 173.3, 173.7 ppm. HRMS (DART) calculated for C 25 H 37 N 3 O 6 [M+H] + 476.2745, found 476.2755. N ((1 Dodecyl 1 H 1,2,3 triazol 4 yl)me thyl) 2 (3 (2 (heptylamino) 2 oxoethyl) 2,4,6 trihydroxy 5 (2 (isopropylamino) 2 oxoethyl)phenyl)acetamide (2 24): To a 10 mL RBF was added 2 23 (0.025 g, 0.053 mmol), THF (2 mL), H 2 O (1 mL), and 1 azidododecane (0.017 g, 0.079 mmol). The solution stirre d for 10 min before the addition of CuSO 4 (0.33M, 0.016 mL, 0.0050 mmol) and NaAsc (1.0M, 0.016 mL, 0.016 mmol). The solution stirred at room temperature for 2 h at which time TLC indicated full conversion of the starting alkyne. The reaction was poured into brine (75 mL) and extracted with EtOAc (3 Ã— 25 mL). The organic layers were combined, dried over Na 2 SO 4 , filtered, and concentrated to yield a brown solid. The solid was purified by
61 column chromatography ( R f = 0.5, 1:3 Hex:EtOAc) to yield an off whi te solid (0.018 g, 51%). 1 H NMR (DMSO d 6 ) 0.85 (dt, 6H, J = 6.8 Hz), 1.06 (d, 6H, J = 6.6 Hz), 1.23 (m, 28H), 1.40 (m, 2H, J = 6.5 Hz), 1.77 (m, 2H, J = 7.2 Hz), 3.04 (q, 2H, J = 6.3 Hz), 3.47 (s, 2H), 3.48 (s, 2H), 3.51 (s, 2H), 3.82 (m, 1H, J = 6.5 Hz ), 4.30 (m, 4H), 7.88 (s, 1H), 8.46 (m, 2H), 8.83 (t, 1H, J = 5.6 Hz), 10.5 (s, 1H), 10.5 (s, 1H), 10.7 (s, 1H) ppm. 13 C NMR (DMSO d 6 ) 13.9, 13.9, 22.0, 22.1, 22.1, 25.9, 26.3, 28.4, 28.4, 28.7, 28.7, 28.9, 28.9, 29.0, 29.7, 31.1, 31.2, 31.3, 31.4, 34.6 , 38.9, 40.9, 49.2, 102.8, 102.9, 102.9, 122.8, 144.1, 153.8, 153.8, 153.8, 172.9, 173.5, 173.7 ppm. HRMS (ESI) calculated for C 37 H 62 N 6 O 6 [M+Na] + 709.4623, found 709.4627. N heptyl 2 (3 (2 (((1 heptyl 1H 1,2,3 triazol 4 yl)methyl)amino) 2 oxoethyl) 2,4,6 trihydroxy 5 (2 (isopropylamino) 2 oxoethyl)phenyl)acetamide: (2 25) To a 10 mL RBF was added 2 11 (0.019 g, 0.043 mmol) and dry DMF (2 mL). The reaction was cooled to 41 Â°C for 15 min before the addition of heptylam ine (0.042 mL, 0.043 mmol). The reaction stirred at low temperature for 24 h before the addition of isopropylamine (0.051 mL, 0.051 mmol). The reaction solution was allowed to slowly warm to rt and stirred overnight. The reaction was poured into EtOAc ( 75 mL) and washed with water (5 Ã— 20 mL) and brine (20 mL). The organic layers were combined, dried over Na 2 SO 4 , filtered, and concentrated to yield a brown solid. The solid was purified by column chromatography ( R f = 0.5, 1:4 Hex:EtOAc) to yield an off w hite solid (0.015 g, 72%). 1 H NMR (DMSO d 6 ) 0.85 (dt, 6H, J = 6.8 Hz), 1.06 (d, 6H, J = 6.6 Hz), 1.23 (m, 16H), 1.40 (m, 2H, J = 6.5 Hz), 1.77 (m, 2H, J = 7.2 Hz), 3.04 (q, 2H, J = 6.3 Hz), 3.47 (s, 2H), 3.48 (s, 2H), 3.51 (s, 2H), 3.82 (m, 1H, J = 6.5 Hz), 4.30 (m, 4H), 7.88 (s, 1H), 8.46 (m, 2H), 8.83 (t, 1H, J = 5.6 Hz), 10.5 (s, 1H), 10.5 (s, 1H), 10.7 (s, 1H) ppm. 13 C NMR
62 (DMSO d 6 ) 13.9, 13.9, 22.0, 22.0, 22.1, 25.8, 26.3, 28.0, 28.3, 28.7, 29.7, 31.1, 31.2, 31.3, 31.4, 34.6, 38.9, 40.9, 49.2, 10 2.8, 102.9, 102.9, 122.8, 144.1, 153.8, 153.8, 153.8, 172.9, 173.5, 173.7 ppm. 2,2' (5 (2 (((1 benzyl 1H 1,2,3 triazol 4 yl)methyl)amino) 2 oxoethyl) 2,4,6 trihydroxy 1,3 phenylene)bis(N (prop 2 yn 1 yl)acetamide) (2 2 9) : A solution of 2 8 (0.05 g, 0.12 mmol) in DMF (3 mL) was cooled to 0 Â° C for ten minutes. To this reaction was added propargylamine (0.020 g, 0.36 mmol). The resulting solution was allowed to stir at this temperature for two hours before warming to r oom temperature. The solution continued to stir for an additional 22 hours. At this time, the reaction was poured in to EtOAc (75 mL) and washed with water (3 x 25 mL) and brine (1 x 25 mL). The solution was dried over Na 2 SO 4 , filtered, and concentrated to afford the product as a yellow, sticky solid (0.06 g, 94%). No further purification was performed. 1 H NMR (DMSO d 6 3.12 (t, 2 H, J = 2.5 Hz), 3.48 (s, 4 H), 3.50 (s, 2 H), 3.86 (dd, 4 H, J = 5.5 and 2.5 Hz), 4.30 (d, 2 H, J = 5.5 Hz), 5.56 (s, 2 H), 7.33 (m, 5 H), 7.97 (s, 1 H), 8.73 (t, 2 H, J = 5.5 Hz), 8.81 (t, 1 H, J = 5.5 Hz), 9.96 (s, 1 H), 10.04 (s, 2 H). 13 C NMR (DMSO d 6 31.4, 35.1, 53.2, 73.8, 81.1, 103.3, 123.6, 128.5, 128.6, 129.2, 136.5, 144.9, 154.2, 173.6, 173.8. HRMS (ESI) calculated for C 28 H 28 N 6 O 6 [M+Na] + 567.1963, found 567.1963.
63 N ((1 hexyl 1H 1,2 ,3 triazol 4 yl)methyl) 2 (4 hydroxy 2,7 dioxo 2,3,7,8 tetrahydrobenzo[1,2 b:3,4 b']difuran 5 yl)acetamide (16): To a flame dried flask was added 10 (0.052 g, .017 mmol) in dry THF (3 mL). CuI (0.066 g, 0.35), DIPEA (0.03 mL, 0.17 mmol), and 1 azidohexan e (0.20 g, 1.72 mmol) was added to the flask. The reaction mixture was allowed to stir at room temperature under argon for 2 hours. At this time the reaction was poured into brine and extracted with EtOAc (4 x 20 mL). The organic layer was dried over Na 2 SO 4 , filtered, and concentrated. The resulting crude solid was purified via column chromatography (30% hex:EtOAc) to yield the pure, tan solid (0.26 g, 36 %). 1 H NMR (DMSO d 6 J = 7 Hz), 1.24 (m, 6H), 1.77 (p, 2H, J = 7 Hz), 3.49 (s, 2H), 3.79 (s, 2H), 3.92 (s, 2H), 4.30 (m, 4H), 7.89 (s, 1H), 8.60 (t, 1H, J = 5 Hz), 10.3 (s, 1H). 13 C NMR (DMSO d 6 31.4, 35.0, 49.7, 97.6, 103.1, 105.5, 123.2, 145.1, 148.2, 151.9, 170.3, 174.4, 174.6.
64 CHAPTER 3 CO NCLUSIONS AND FUTURE DIRECTIONS This work has explored using CuAAC with a variety of BTF and phloroglucinol derivatives. Azide scope was studied and post click functionalization of difuranone BTF compounds was accomplished through aminolysis. The major m odes of multifunctionalization were outlined and examples were given as to the field where they are relevant. Copper catalyzed azide alkyne cycloaddition was discussed to highlight the efficiency, versatility, and orthogonality of the reaction. Examples where CuAAC has been coupled with classical multifunctionalization strategies in an effort to streamline synthesis were reviewed. The need for new and powerful multifunctionalization strategies has been identified. Functionalization of BTF via CuAAC Az ide scope was investigated with a monofunctionalized BTF derivative 2 6 . Linear azides gave cleaner products, higher degree of conversion, and shorter reaction times in comparison to benzylazide and 2 ethylhexylazide. Clicked compounds using linear azide s could be isolated after column chromatography in 50% yield on average. These compounds were characterized by 1 H NMR, 13 C NMR, and HRMS. Post click functionalized of BTF difuranones was accomplished through aminolysis with a variety of amines: heptylami ne, furfurylamine, and propargylamine. Conversion of the difuranones to phloroglucinol derivatives always neared 100% and the pure products could be isolated after column chromatography in 50% yield on average. Both difunctional and trifunctional phlorog lucinol derivatives were functionalized by sequential aminolysis and CuAAC.
65 Future Directions Future studies with BTF will continue to expand the azide scope for CuAAC with 2 6 . Alkyl azides with branching at the gamma position (e.g. isoamylazide) will b e considered to study the effects of branching on CuAAC with the BTF system. Other azides of interest would be those with biological or fluorescence applications such as biotin and dansyl chloride derivatives, respectively. 64 Another area of interest is applying other click chemistries such as Diels Alder cycloaddition and thiol ene addition to the BTF scaffold. Expanding the number of efficient reactions that are compatible with the BTF core would also expand the applications of BTF . While few in numb er, there are systems that utilize triple click reactions for the one pot synthesis of a functional system. 65 Functionalizing BTF with an alkyne, diene, and alkene would make it possible for BTF to be used as a core for miktoarm star (three arm) polymers or for bioconjugation. 65 Polymer and/or surface modifications can also find utility in this methodology. Functionalizing BTF with a clickable alkyne unit could be a powerful tool for modifying polymers. Since 2 6 could be functionalized with up to two different units, there are a vast number of synthetic targets that can be appended to a polymer chain. Hence, polymers with new and enhanced properties could be synthesized just by tuning how BTF was functionalized. BTF could be functionalized with a cli ckable handle and two hydrophobic polymer backbone appended with azide side chains (Figure 3 1). The amphiphilicity introduced could induce interesting self assembly or biocompatibility properties. 66 Indeed, block co polymers could be synthesized using the BTF system. Take for example a block co polymer constructed from a hydrophobic polymer and a
66 polymer with azide functionalities along the backbone or at the end grou p. CuAAC with these azide groups and a phloroglucinol derivative with hydrophilic or aromatic functionalities allows self assembly of the block co polymer through hydrophilic/hydrophobic interactions or pi/pi stacking, respectively. 67 69 Homopolymers cou ld undergo the same type of functionalization and subsequent self assembly. Figure 3 1. Proposed phloroglucinol derivative that could be appended to an azide grafted polymer by CuAAC to create an amphiphilic polymer. Additionally, nanoparticles could be assembled bearing peripheral azide functionalities. 70 These azide groups could then undergo CuAAC with a BTF molecule bearing a therapeutic drug, a dye, or some type of targeting agent. 71 Since nanoparticles have foun d increased use as drug delivery platforms, BTF would offer the advantage of efficient functionalization of expensive drug substrates to a nanoparticle core while keeping the drug in close proximity to the dye or targeting agent. Likewise, a phloroglucino l derivative possessing a drug and ligand designed for receptor recognition could be simply clicked to a polymer backbone and further assembled in to micelles or nanoparticles for enhanced drug delivery. 71 Finally, use of BTF derivatives as antibody recru iting molecules (ARMs) can be imagined. ARMs are bifunctional molecules that help antibodies bind to entities relevant to the disease in question. Functionalizing BTF with an antibody binding terminus, a target binding terminus, and even a fluorescent ta g would offer a new approach to ARMs. 72,73
67 Offered here is an approach to highly functional molecules that aims to embody the click philosophy for the entire synthetic process. While BTF can only undergo a limited number of transformations, these reac tions are wide in scope and modular. The reactions have excellent atom economy which boosts crude purity and aids in purification. Also, the reaction conditions are flexible in that the reactions do not require inert or special conditions. Work up is ea sy as only general extraction techniques are needed. Targets are readily made from available materials and the reactions proceed predictably and with high levels of conversion. Another advantage is that the clicked targets (such as 2 11 or 2 14 ) do not n eed to be purified in order to undergo final functionalization. This streamlines the synthesis and avoids purification steps thus boosting yield. However, purification techniques and yield still need to be optimized in order to truly embody the click che mistry philosophy. A noticeable portion of material is lost during column chromatography. Lowering the catalyst and azide loading is one possible way to reduce the purification required of the clicked targets. The established approach has great potentia l for the selective attachment of dyes, biological molecules, or other valuable substrates in order for BTF to be applied to designer systems. Applied to macromolecules, such as polymers, new morphologies and self assembly can be envisioned where the func tionalization of BTF controls hydrophobic/hydrophilic or pi pi stacking interactions. While the results reported in Chapter 2 are preliminary, they demonstrate a tailorable method toward streamlined synthesis of designer molecules.
68 APPENDIX A INVESTI GATION OF NOVEL, MULTIFUNCTIONAL SCAFFOLDS In an effort to expand the library of compounds like BTF , investigation of other symmetrical scaffolds capable of sequential multifunctionalization through coupled inductive or strain effects has begun. An obviou s starting point was mellitic trianhydride ( MTA ) with three anhydride rings fused to a benzene core (Scheme A 1). MTA has been studied for its ability to form charge transfer complexes with naphthalene, triphenylene, and 9,10 dimethylanthracene. 74 However , due to rapid hydrolysis in air and overall difficult handling, little work has been done to exploit its reactivity. Exploring the sequential aminolysis of MTA through control of temperature and stoichiometry could provide another route to trifunctional molecules. Additionally, imide formation via condensation of the trifunctional molecule will yield a tri imide which has not been reported (Figure A 1). Tri imides are often used as electron acceptors so studying non symmetric triimides in this role is i nteresting. 75 Figure A 1: Proposed sequential functionalization of MTA . The first task was to complete some data analysis in order to better understand if the MTA system would be a good candidate for multifunctionaliz ation. Investigation of the reported crystal structure data and IR spectra revealed interesting details. MTA takes on a three blade propeller shape which is consistent with D 3 symmetry. 74 The propeller shape is not a consequence of strain or non bondi ng repulsions but of the
69 packing orientation in the crystal form. Non bonding repulsions between the O atoms of the carbonyls can be excluded as a reason for the distortion as the OÂ·Â·Â·O distance is 3.206 Ã…, well above that of the van der Waals radii. MTA is planar when in molecular complexes with donors such a triphenylene. If the distortion were to arise from intramolecular repulsions, they would be evident in the crystal structure of the complexes. In fact, the distortion arises from intermolecular pa cking. 74 There is a distinct edge to face interaction in the crystal between the anhydride oxygen and the carbonyl carbon. This can be ascribed to nucleophile electrophile type interactions, and pyramidalization of the carbonyl carbon toward the anhydrid e oxygen aids in this attraction. We were curious if MTA exhibited the Mills Nixon effect as to speak to strain associated with the system. The reported crystal structure shows a disordered orientation; both orientations could be well resolved save for the carbon atoms of the central benzene ring. Thus, information about bond length alternation could not be gathered despite any distortions the fused five membered ring might impose. The crystal structure does indicate that the bond angles for the sp 2 hy bridized carbons are acute in comparison to the preferred angle. The carbonyl carbons have an internal bond angle of 106.7Â°, far from the preferred 120Â° for sp 2 hybridized atoms. The decreased bond angles are on the scale of those found in BTF . This alo ne can attest to the strain of the system. 74 IR information was studied to compare MTA with relevant anhydrides. There is a noticeable increase in stretching frequencies from monoanhydride to trianhydride (Figure A 2). This shift in frequency is a result of increasing bond order for the carbonyl
70 groups. The bond order reflects the contribution of the resonance structures for the anhydrides. Since MTA has 14 contributing resonance structures, the bond order will be higher and thus the C=O stretches will shift to higher frequencies. 76 Also, cyclization of the anhydrides shifts the stretch to higher frequencies speaking to ring strain as can be seen by comparing the stretches of acetic anhydride (1827 and 1755 cm 1 ). Phthalic Anhydride Pyromellit ic Anhydride Mellitic Trianhydride In phase 1849 1872, 1854 1886, 1872 Out of phase 1769 1790, 1775 1799 Bond Order 1.667 1.800 1.857 Figure A 2. C=O stretching frequencies (cm 1 ) and bond orders of relevant anhydrides. Preliminary studies focused on synthesizing triimides from MTA . Since triimides are typically synthesized from mellitic acid through a condensation process, our approach to triimides was quite different. MTA was first synthesized by refluxing mellitic acid with acetic anhydride overn ight. The solvent was removed by high vacuum and the solid remained under vacuum until ready to use. Attempts to fully characterize MTA were unsuccessful. Since MTA is highly susceptible to hydrolysis from moisture in the air, performing any type of cha racterization was difficult. Attempts at NMR were difficult due to the low solubility of MTA as well as rapid hydrolysis when exposed to any moisture. Furthermore, most of the literature on the preparation of MTA is rather dated and does not include synt hetic or characterization details. To access the triimides, the green, MTA solid was refluxed with benzylamine in DMF (Figure A 3). TLC showed several spots which corresponded to mono , di , and
71 triimides. Column chromatography allowed the isolation of the desired triimide A 4 , and NMR characterization confirmed the desired product. However, the desired product could only be isolated in 25% yield. Figure A 3. Synthesis of a benzyltriimide. With these somewhat disa ppointing results, the ability of these imides to undergo transimidation was considered. To test this hypothesis, a pure triimide A 5 was refluxed with 3.0 equivalents of benzylamine (Figure A 4). This test reaction revealed a crude mixture of products t hat had incorporated the added amine. It appears by 1 H NMR analysis that some of the products were diamides where benzylamine had been incorporated as evident by the presence of new amide peaks in the 6.0 6.5 ppm region. Other products were imides wher e a benzyl substituent had replaced the heptyl chain as was identified by a singlet at 4.8 ppm. The results of these reactions were not promising and pursuit of diverse triimides was abandoned. Figure A 4: Transimidation study resulting in a complex mixt ure of imides and amides of varying degrees of substitution.
72 Proposed Future Symmetrical Targets Another molecule that could potentially be used as a symmetric, molecular scaffold is trisisatin. Isatin has been around for quite some time and has found its place in science as it is often used as a building block in biological targets and dyes. 77 79 Isatin can undergo ring opening events with good nucleophiles, especially if the amide nitrogen is functionalized with an electron withdrawing group (Figure A 5A). To date, bisisatin has been reported, but there are no references for the synthesis of trisisatin (Figure A 5B). 80 The goal is to synthesize the trisisatin target and investigate its reactivity toward sequential functionalizations. Isatins are ty pically Figure A 5: A) Reaction of N acetylisatin with 4 bromoaniline. B) Bisisatin (left) and proposed target trisisatin (right). synthesized through the Sandmeyer reaction of ani lines with chloral hydrate in acidic conditions (Figure A 6). 81a This method is the most widely reported and always proceeds with good yields. Since Sandmeyer first introduced his synthesis in the early 1900s, other methods have become available. Synthet ic routes by Stolle, Gassman, and others will also be considered for this project pending results using Sandmeyer isatin synthesis (Figure A 6). 81a,81b A) B)
73 Figure A 6. Sandmeyer, Stolle, and Gassman synthesis of isatin. Discussion The desymmetrization of MTA was attempted by first investigating its reactivity toward primary amines and subsequent condensation to afford triimides. Unfortunately, MTA could not be synthesized in high fidelity and the desired triimide could not be gathered in good yield. It was also found that the triimides could undergo transimidation with other amines in the reaction solution which complicates the product outcome as well as diminishes the ability of MTA to be a molecular scaffold for trifu nctionalization. In the future, other symmetrical scaffolds will be considered. Trisisatins are of particular interest since they can undergo ring opening events. Because they have never been reported, there is information to be gathered from their synt hesis alone.
74 APPENDIX B APPROACHES TO DENDRIMERS THROUGH FUNCTIONALIZATION OF BTF Since their discovery in 1978, dendrimers have become a great area of interest because of their potential applications in medicine and material science. Dendrimers a re very interesting macromolecules as they are globular and highly branched structures whose function is typically governed by the many functional groups found on the periphery. However, one of the drawbacks of making sophisticated large molecules is the time and effort that goes in to the synthesis of such dendrimers. A stepwise approach to constructing dendrimers is often very expensive, requiring large amounts of supplies and time, and affords low yields of the desired products. Since high yielding re actions are imperative in achieving high generation dendrimers, there is a continuous need to develop synthetic strategies that will accelerate dendrimer construction. 6, 82 Synthesis of Dendrimer Precursors The synthesis of dendrimers is an attractive ap plication area for BTF because dendrimers require high yielding and reliable reactions. 83 88 For this reason, it was decided to use a divergent approach for the initial synthesis of a first generation dendrimer constructed from BTF . At its simplest, this approach would start with tri functionalization of BTF with a mono protected diamine, followed by deprotection to reveal three free and reactive amines. These amines could then be reacted with a di functionalized BTF analog to quickly yield a first genera tion dendrimer (Figure B 1). Removal of the peripheral protecting groups to reveal free, reactive amines followed by coupling with six equivalents of a di functionalized BTF would easily provide a second generation dendrimer with 12 masked reactive sites around the edges. This reaction sequence could be repeated for larger generations.
75 In executing this approach, BTF was reacted with 3.5 equivalents of mono Boc protected ethylene diamine (Figure B 2). The reaction was run in DMF and allowed to st ir for 19 hours which yielded core B 2 in 90% after column chromatography (9:1 DCM/MeOH). Boc protecting groups were chosen because they are stable to many reaction conditions and are easily removed. With the tri substituted product in hand, removal of t he Boc protecting groups was the next step. Since Boc groups are easily removed with acid, the first attempt at removing the protecting groups employed 4 M HCl in EtOAc. After several trials, the desired product B 3 could not be isolated. It appeared by NMR that Boc groups were still present as a peak at 1.4 ppm was evident. There was also evidence of deterioration of the starting material. Another acid approach using HCl/MeOH/dioxane was attempted. This too gave incomplete deprotection as well as unw anted side products. Figure B 1. Synthesis of a first generation dendrimer from BTF .
76 In an effort to avoid using strong acids, a route using SiMe 3 I in MeCN was considered. This approach gave complete conversion to the desired product in less than 3 hours at room temperature. The corresponding iodide salt B 3 was iso lated by crystallizing the resulting residue in acetone. The product is evident via NMR by the disappearance of the Boc methyl groups at 1.4 ppm. Equally as distinct was the shift of the ethylene peaks from 3.08 ppm and 2.99 ppm to 3.29 ppm and 2.88 ppm, respectively. The next step involved synthesizing the peripheral di substituted BTF units (Figure B 3). Addition of 1.95 equivalents of Boc protected ethylene diamine to BTF gives the desired di substituted product B 4 in 85% yield before column chromatography. Using a slight excess of BTF (in relation to amine) helps to lessen the formation of unwanted tri substituted product. Discerning which impurities are in this reaction mixture is straightforward given that t he methylene protons of BTF have characteristic chemical shifts based on the degree of substitution. The presence of tri substituted product makes purification by column chromatography challenging because a solvent system that gives good separation (9:1 EtOAc/Hex) did not remove all of the product from the column. Likewise, a solvent system (9:1 DCM/MeOH) that carries all the product off of the column does not give the Figure B 2. Synthesis of first generation dendrimer core.
77 best separation. In the end, the desired product could be isolated (< 50%) and is pu re spectroscopically ( 1 H NMR). Use of Alternative Diamines In an attempt to avoid the use of protecting groups, the idea of using unsymmetrical diamines was appealing. The Simanek group has done extensive work in buil ding dendrimers using cyanuric chloride as the symmetric core scaffold. They have used 4 aminomethylpiperidine (4 AMP) as a linking unit in many of their dendrimers and received good results. 22,87 When attempting the tri functionalization of BTF with 3.5 equivalents of 4 AMP, a product insoluble in most organic solvents formed (Figure B 4). The structure of the compound could not be confirmed by NMR or MS. It is likely that the reactivity between the primary and cy clic secondary amine was too similar to give selective aminolysis of BTF . On the same note, cross linking of BTF cores to form oligomers is probable. After several attempts, the use of 4 AMP was abandoned for the more reliable Boc protected ethylene diam ine. Figure B 3. Synthesis of di substituted dendron. Figure B 4. Synthesis of dendrimer with 4 AMP.
78 Final Attempts at Dendrimer Synthesis Unfortunately, after many unsuccessful attempts at linking BTF analogs B 3 and B 4 to yield a first generation dendrimer, that plan was put on hold and other routes were investigated. One route considered was u sing Sonogashira coupling to link an aryl halide and a terminal alkyne. Although synthesis of the starting materials was straight forward, the linking proved difficult. With milder conditions, the crosslinking did not occur and both starting materials were identified by NMR. When the temperature was elevated, the reaction gave products that could not be identified and presumably resulted from the decomposition of the starting materials (Figure B 5). With that, it was decided to try using azide alkyne Huisgen cycloaddition because of the mild reaction conditions and vast functional group tolerance. 28,39,88 The first attempt at making dendrimers using this click reaction entailed linking two mono alkyne 2 6 compounds with 1,6 diazidohexane (Figure B 6) . There were clues that the reaction was proceeding, but the reaction never went to completion. This is most likely due to dilute reaction conditions (0.036 mM due to solubility of 2 6 in THF). Additionally, use of aqueous medium (as used in Chapter 2) would give better outcomes but has not been tried in this case. Figure B 5. Sonogashira c oupling toward first generation dendrimer synthesis.
79 Figure B 6. Attempted synthesis of linked compound bearing four lactone rings. Discussion BTF was used to explore first generation dendrimers formation. Several amines and different linki ng methods were utilized in an effort to construct the targets. Initial divergent approaches proved to be difficult due to incomplete deprotection of the Boc groups as well as trouble isolating the pure disubstituted peripheral group. Attempts to use dia mines were also unsuccessful because the kinetics of the ring opening events were too similar between the primary and cyclic secondary amine. Finally, with the failure of Sonogashira coupling, click chemistry was employed to link two monofunctionalized BTF motifs. Because click chemistry has proven to work exceptionally well with BTF , it is likely that concentration issues were the reason for failure. Although a first generation dendrimer was not synthesized, much has been learned about the type of rea ctions the BTF scaffold can endure as well as the types of amines that can be used.
80 Experimental Tert butyl (2 aminoethyl)carbamate : To a solution of ethylenediamine (13.4 g, 223 mmol) in CH 2 Cl 2 (70 mL) was added Boc Anhydride (17.4 g, 79.8 mmol) over 3.5 hours. Saturated Na 2 CO 3 (100 mL) was added to the solution and the two layers were separated. The aqueous layer was extracted with CH 2 Cl 2 (3 Ã— 100 mL) and dried over MgSO 4 . The organic layer was concentrated under reduced pressure. The oil was purified via column chromatography (9:1 CH 2 Cl 2 /MeOH, 1% NEt 3 ). 1 H NMR (CDCl 3 ) 1.43 (s, 9H), 2.78 (t, 2H, J = 6 Hz), 3.16 (q, 2H, J = 6 Hz), 4.90 (s, 2H) ppm. 13 C NMR (CDCl 3 ) 28.5, 41.9, 43.4, 79.4, 156.4 ppm. HRMS (ESI ) calculated for C 7 H 16 N 2 O 2 [M+H] + 161.1285, found 161.1292. Tri tert butyl (((2,2',2'' (2,4,6 trihydroxybenzene 1,3,5 triyl)tris(acetyl))tris(azanediyl))tris(ethane 2,1 diyl))tricarbamate (B 2) : In a reaction vessel eq uipped with a stir bar was added BTF (0.05 g, 0.20 mmol) and dry DMF (5 mL). The reaction flask was cooled in a dry ice/acetonitrile bath ( 41 Â°C). To the cooled solution was added tert butyl (2 aminoethyl)carbamate (0.11 g, 0.71 mmol). The reaction was allowed to stir for 2 h under a blanket of argon at which time the reaction was warmed to rt. The reaction stirred for an additional 14 h. The reaction was poured in to ethyl acetate (75 mL), washed with water (5 Ã— 30 mL), and once with brine (30 mL). The organic layer was dried over Na 2 SO 4 , filtered, and concentrated under
81 reduced pressure. The product was purified via column chromatography (5% MeOH/CH 2 Cl 2 , R f 1 H NMR (DMSO d 6 ) 1.37 (s, 27H), 2.99 (q, 6H, J = 4.9 and 5.7), 3.07 (q, 6H, J = 5.2 and 5.7), 3.47 (s, 6H), 6.81 (t, 3H, J = 4.7 Hz), 8.44 (t, 3H, J = 4.5 Hz), 10.46 (s, 3H) ppm. 13 C N MR (DMSO d 6 ) 28.7, 31.8, 36.2, 39.7, 78.2, 103.3, 154.2, 156.1, 174.4 ppm. HRMS (ESI) calculated for C 33 H 54 N 6 O 12 [M+Na] + 749.3692, found 749.3702. 2,2' ((2,2' (5 (2 ((2 Aminoethyl)amino) 2 oxoethyl) 2,4,6 trihydroxy 1,3 phenylene)bis(acetyl))bis(azanediyl))bis(ethan 1 aminium) (B 3): To a flame dried 25 mL round bottom flask equipped with a stir bar was added B 2 (0.04 g, 0.05 mmol) and 5 mL dry MeCN. The reaction was allowed to stir for 10 min at which time Me 3 SiI (0.028 mL, 0.20 mmol) was added via syringe. After several minutes, a yellow solid appeared. The reaction stirred for 3 h, and then dry methanol (0.020 mL, 0.49 mmol) was added to quench the reaction. The volatiles were removed and the crude solid was crystallized from acetone. NMR confirmed complete removal of protecting groups. 1 H NMR (DMSO d 6 ) 2.87 (q, 6H, J = 6 Hz), 3.29 (q, 6H, J = 6 Hz), 3.50 (s, 6H), 3.72 (bs, 6H), 7.70 (bs, 9H), 8.48 (t, 3H, J = 5.4 Hz), 10.03 (s, 3H) ppm. 13 C NMR (DMSO d 6 ) 31.9, 37.2, 38.8, 103.5, 154.1, 174.7 ppm. HRMS (ESI, 1% AcOH/MeOH) calculated for C 18 H 30 N 6 O 6 [M+H] 427.2300, found 427.2311.
82 Di tert butyl (((2,2' (4,6 dihydroxy 2 oxo 2,3 dihydrobenzofuran 5,7 diyl)bis(acetyl)) bis(azanediyl))bis(ethane 2,1 diyl))dicarbamate (B 4): In a reaction vessel equipped with a stir bar was added BTF (0.050 g, 0.20 mmol) and dry DMF (5 mL). The reaction flask was cooled in a dry ice/acetonitrile bath ( 41 Â°C). To the cooled solution was added tert butyl (2 aminoethyl)carbamate (0.063 g, 0.39 mmol). The reaction was allowed to stir for 2 h under a blanket of argon at which time the reaction was warmed to rt. The reaction stirred for an additional 14 h. The reaction was poured in to eth yl acetate (75 mL), washed with water (5 Ã— 30 mL), and once with brine (30 mL). The organic layer was dried over Na 2 SO 4 , filtered, and concentrated under reduced pressure. The product was purified via column chromatography (10% MeOH/CH 2 Cl 2 , R f yield the product as a light brown solid (0.08 g, 88%). 1 H NMR (DMSO d 6 ) 1.36 (s, 9H), 2.98 (q, 4H), 3.08 (q, 4H), 3.43 (s, 2H), 3.47 (s, 2H), 3.69 (s, 2H), 6.79 (t, 2H), 8.14 (t, 1H, J = 5 Hz), 8.40 (t, 1H, J = 5 Hz), 9.89 (s, 1H), 10.90 (s, 1H) ppm. 13 C NMR (DMSO d 6 ) 28.2, 30.8, 31.2, 31.7, 39.1, 39.4, 77.7, 98.3, 100.7, 106.5, 150.7, 151.9, 155.4, 155.6, 171.9, 172.5, 174.5 ppm. HRMS (ESI) calculated for C 26 H 38 N 4 O 10 [M+Na] + 589.2480, found 589.2469. 2,2',2'' (2,4,6 Trihydroxybenzene 1,3,5 triyl)tris(N (4 bromobenzyl)acetamide) (B 6): To a 25 mL round bottom flask was added 4 bromobenzylamine hydrochloride (0.16 g, 0.71 mmol), triethylamine (0.11 mL, 0.81 mmol), and THF (3 mL). The mixture was stirred for 1 h at rt at which time it was cooled to 41 Â°C. BTF (0.050 g, 0.20 mmol) was
83 added to the flask and reaction mixture was stirred. After 1 h, the reaction was allowed to warm to rt and stirred for 27 h. EtOAc (75 mL) was added and the organic layer was washed with water (3 Ã— 20 mL) and once with brine (20 mL). The organic layer was then dried over Na 2 SO 4 , filtered, an d concentrated under reduced pressure. The resulting crude material was purified by column chromatography (20% Hex/EtOAc, R f 0.3) to give the desired product as an off white solid (0.146 g, 89%). 1 H NMR (DMSO d 6 ) 3.55 (s, 6H), 4.22 (d, 6H, J = 5.3 Hz), 7.21 (d, 6H, 7.9 Hz), 7.49 (d, 6H, J = 7.9 Hz), 8.79 (t, 3H, J = 3.4 Hz), 10.11 (s, 3H) ppm. 13 C NMR (DMSO d 6 ) 31.8, 42 .3, 103.5, 120.6, 130.0, 131.6, 138.8, 154.3, 173.9 ppm. HRMS (ESI) ion calculated for C 33 H 30 Br 3 N 3 O 6 was not found.
84 Figure B 7. 1 H NMR (Top) and 13 C NMR (Bottom) of B 2 .
85 Figure B 8. 1 H NMR (Top) and 13 C NMR (Bottom) of B 3 .
86 Figure B 9. 1 H NMR of B 4 .
87 Figure B 10. 1 H NMR (Top) and 13 C NMR (Bottom) of B 6 .
88 APPENDIX C SUPPLEMENTARY INFORMATION FOR CHAPTER 2
89 Figure C 1. (Top) 1H NMR and (Bottom) 13 C NMR spectrum of 2 4 .
90 Figure C 2. (Top) 1H NMR and (Bottom) 13 C NMR spectrum of 2 5 .
91 Figure C 3 . (Top) 1H NMR and (Bottom) 13 C NMR spectrum of 2 6.
92 Figure C 4. (Top) 1H NMR and (Bottom) 13 C NMR spectrum of 2 8 .
93 Figure C 5. (Top) 1H NMR and (Bottom) 13 C NMR spectrum of 2 10 .
94 Figure C 6. Crude 1 H NMR spectrum of 2 11 .
95 Figure C 7. (Top) 1H NMR and (Bottom) 13 C NMR spectrum of 2 11 .
96 Figure C 8. (Top) 1H NMR and (Bottom) 13 C NMR spectrum of 2 14 .
97 Figure C 9. (Top) 1H NMR and (Bottom) 13 C NMR spectrum of 2 15 .
98 Figure C 10. (Top) 1H NMR and (Bottom) 13 C NMR spectrum of 2 16 .
99 Figure C 11. (Top) 1H NMR and (Bottom) 13 C NMR spectrum of 2 17 .
100 Figure C 12. (Top) 1H NMR and (Bottom) 13 C NMR spectrum of 2 18 .
101 Figure C 13. (Top) 1 H NMR and (Bottom) 13 C NMR spectrum of 2 19 .
102 Figure C 14. (Top) 1 H NMR and (Bottom) 13 C NMR spectrum of 2 20 .
103 Figure C 15. (Top) 1 H NMR and (Bottom) 13 C NMR spectrum of 2 21 .
104 Figure C 16. (Top) 1 H NMR and (Bottom) 13 C NMR spectrum of 2 23 .
105 Figure C 17. (Top) 1 H NMR and (Bottom) 13 C NMR spectrum of 2 24 .
106 Figure C 18. (Top) 1 H NMR and (Bottom) 13 C NMR spectrum of 2 25 .
107 Figure C 19. (Top) 1 H NMR and (Bottom) 13 C NMR spectrum of 2 29 .
108 Figure C 20. (Top) 1 H NMR and (Bottom) 13 C NMR spectrum of 2 34 .
109 REFERENCES 1. Chevalier, A.; Hardouin, J.; Renard, P Y.; Romieu, A. Org. Lett. 2013 , 15 , 6082 6085. 2. Xu, L.; Li, Y.; Li, Y. Asian J. Org. Chem. 2014 , 3 , 582 602. 3. Kamphuis, M. M.; Johnston, A. P. R.; Such, G. K.; Dam, H. H.; Evans, R. A.; Scott, A. A.; Nice, E. C.; Heath, J. K.; Caruso, F. J. Am. Chem. Soc. 2010 , 132 , 15881 15883. 4. Sugai, N.; Heguri, H.; Yamamoto, T.; Tezuka, Y. J. Am. Chem. Soc. 2011 , 133, 19694 19697. 5. Wan, S.; Zhang, G.; Ge, Z.; Narain , R.; Liu, S. Chem. Asian J. 2011 , 6 , 2835 2845. 6. Watler, M.; Malkoch, M. Chem. Soc. Rev. 2012 , 41 , 4593 4609. 7. Dogutan, D. K.; Zaidi, S. H. H.; Thamyongkit, P.; Lindsey, J. S. J. Org. Chem. 2007 , 72 , 7701 7714. 8. Lemon, C. M.; Karnas, E.; Bawendi, M. G.; Nocera, D. G. Inorg. Chem. 2013 , 52 , 10394 10406. 9. Campbell, F.; Kilner, C. A.; Wilson, A. J. Tetrahderon Letters , 2010 , 51 , 1361 1363. 10. Yoshida, K.; Mori, H.; Tanaka, T.; Mori, T.; Osuka, A. Eur. J. Org. Chem. 2014 . DOI: 10.1002/ejoc.201402435. 11. Lehmann, M.; Jahr, M. Org. Lett. 2006 , 41 , 721 723. 12. Li, Y F.; Morcos, P. A. Bioconjugate Chem. 2008 , 19 , 1464 1470. 13. Mazik, M.; Kuschel, M. Chem. Eur. J. 2008 , 14 , 2405 2419. 14. Feng, X.; Pisula, W.; Zhi, L.; Takase, M.; MÃ¼llen, K. Angew. Chem. Int. Ed. 2008 , 47 , 1703 1706. 15. Yuan, MS.; Wang, Q.; Wang, WJ.; Li, TB.; Wang, L.; Deng, W.; Du, Z.; Wang, JR. Dyes and Pigments 2012 , 95 , 236 243. 16. Carey, F. A.; Sundberg, R. J. Multistep Synthesis. Advanced Organic Chemistry, Part B: Reactions and Synthesis , 5 th Ed.; Springer: New York, 2007; 1256.
110 17. Rivera, D. G.; Weejohan, L. A. J. Am. Chem. Soc. 2009 , 131 , 3721 3732. 18. Strategic Applications of Named Reactions in Organic Synthesis ; Elsevier: California, 2005; 330. 19. Owens, T. D.; Araldi, G. L.; Nutt, R. F.; Semple, J. E. Tetrahedron Letters 2001 , 42 , 6271 6274. 20. Banfi, L.; Guanti, G.; Riva, R. Chem. Commun. 2000 , 985 986. 21. Jee, JA.; Spagnuolo, L. A.; Rudick, J. G. Org. Lett. 2012 , 14 , 3292 3295. 22. Simanek, E. E.; Abdou, H.; Lalwani, S.; Lim, J.; Mintzer, M.; Venditto, V. J.; Vittor, B. Proc. Royal Soc. A Math Phys. Eng. Sci. 2010 , 466 , 1445 146 8. 23. Fries, F. H. J. Chem. Soc. 1886 , 49 , 314 316. 24. EVONIK Industries Applications page. http://cyanuric chloride.evonik.com/product/cyanuric chlorid/en/applications/pages/default.aspx (Accessed May 24, 2014). 25. Baker. M. B.; Ghiviriga, I.; Castellano, R. K. Chem. Sci. 2012 , 3 , 1095 1099. 26. Kolb, H. C.; Finn, M. G.; Sharpless, K. B. Angew. Chem. Int. Ed. 2001 , 40 , 2004 2021. 27. Xi, W.; Scott, T. F.; Kloxin, C. J.; Bowman, C. N. Adv. Func. Mater. 2014 , 24 , 2572 2590. 28. Sumerlin, B. S.; Vogt, A. P. Macromo lecules 2010 , 43 , 1 13. 29. Rostovtsev, V. V.; Green, L. G.; Fokin, V. V.; Sharpless, K. B. Angew. Chem. Int. Ed. 2002 , 41 , 2596 2599. 30. Tornoe, C. W.; Christensen, C.; Meldal, M. J. Org. Chem. 2002 , 67 , 3057 3064. 31. Bock, V. D.; Hiemstra, H.; van Maarsev een, J. H. Eur. J. Org. Chem. 2006 , 51 68. 32. Schulze, B.; Schubert, U. S. Chem. Soc. Rev. 2014 , 43 , 2522 2571. 33. Worrell, B. T.; Malik, J. A.; Fokin, V. V. Science 2013 , 340 , 457 460. 34. Xiong, H.; Seela, F. J. Org. Chem. 2011 , 76 , 5584 5597. 35. Kore, A. R.; Charles, I. Curr. Org. Chem. 2013 , 17, 2164 2191.
111 36. Goswami, L. N.; Houston, Z. H.; Sarma, S. J.; Jalisatgi, S. S.; Hawthorne, M. F. Org. Biomol. Chem. 2013 , 11 , 1116 1126. 37. Li, H.; Cheng, F.; Duft, A. M.; Andronov, A. J. Am. Chem. Soc. 2005 , 127 , 14518 14524. 38. El Habnouni, S.; Darcos, V.; Garric, X.; Lavigne, J. P.; Nottelet, B.; Coudane, J. Adv. Func. Mater. 2011 , 21 , 3321 3330. 39. Fahrenbach, A. C.; Stoddart, J. F. Chem. Asian J. 2011 , 6 , 2660 2669. 40. Hua, Y.; Flood, A. H. J. Am. Chem. Soc. 2010 , 132 , 12838 12840. 41. Malke, M.; Barqawi, H.; Binder, W. H. ACS Macro Lett. 2014 , 3 , 393 397. 42. Li, S.; Han. J.; Gao, C. Polym. Chem. 2013 , 4 , 1774 1787. 43. Vandenbergh, J.; Tura, T.; Baeton, E.; Junkers, T. J. Polym. Sci. Pol. Chem. 2014 , 52 , 1263 127 4. 44. Viault, G.; Dautrey, S.; Maindron, N.; Hardouin, J.; Renard, P Y.; Romieu, A. Org. Biomol. Chem. 2013 , 11 , 2693 2705. 45. Galibert, M.; Renaudet, O.; Dumy, P.; Boturyn, D. Angew. Chem. Int. Ed. 2011 , 50 , 1901 1904. 46. Su, H.; Zheng, J.; Wang, Z.; Lin, F .; Dong, X H.; Becker, M. L.; Cheng, S. Z. D.; Zhang, W B.; Li, Y. ACS Macro Lett. 2013 , 2 , 645 650. 47. Foillard, S.; Rasmussen, M. O.; Razkin, J.; Boturyn, D.; Dumy, P. J. Org. Chem. 2008 , 73 , 983 991. 48. Baker, M.B. Molecular Multifunctionalization via E lectronically Coupled Lactones. Ph.D. Thesis. University of Florida, Gainesville, FL, 2012. 49. Neuvonen, H.; Neuvonen, K.; Koch, A.; Kleinpeter, E.; Pasanen, P. J. Org. Chem. 2002 , 67 , 6995 7003. 50. Neuvonen, H.; Neuvonen, K. J. Chem. Soc., Perkin Trans. 2 , 1999 , 1497 1502. 51. Struthers, H.; Mindt, T. L.; Schibli, R. Dalton Trans. 2009 , 39 , 675 696. 52. Jiang, L.; Wang, Z.; Bai, S. Q.; Hor, T. S. A. Dalton Trans. 2013 , 42 , 9437 9443.
112 53. Fleischel, O.; Wu, N.; Petitjean, A. Chem. Commun . 2010 , 46 , 8454 8456. 54. Teyssot, M. L.; Chevry, A.; Traikia, M.; El Ghozzi, M.; Arignant, D.; Gautier, A. Chem. Eur. J. 2009 , 15 , 6322 6326. 55. Rodionov, V. O.; Presolski, S. I.; DÃaz, D. D.; Fokin, V. V.; Finn, M. G. J. Am. Chem. Soc. 2007 , 129 , 12705 12712. 56. Donnelly, P. S. ; Zanatta, S. D.; Zammitt, S. C.; White, J. M.; Williams, S. J. Chem. Commun. 2008 , 2459 2461. 57. Ã–zÃ§ubukÃ§u, S.; Ozkal, E.; Jimeno, C.; PericÃ s, M. A. Org. Lett. 2009 , 11 , 4680 4683. 58. a) Li, Y.; Lampkins, A. J.; Baker, M. B.; Sumpter, B. G.; Huang, J.; Abboud, K. A.; Castellano, R. K. Org. Lett . 2009 , 11 , 4314 4317. b) Baker, M. B.; Yuan, L.; Marth, C. J.; Li, Y.; Castellano, R. K. Supramol. Chem. 2010 , 22 , 789 802. 59. Campbell Verduyn, L.; Mirfeizi, L.; Dierckx, R. A.; Elsinga, P. H.; Feringa, B. L . Chem. Comm. 2009 , 2139 2141. 60. Sinha, J.; Sahoo, R.; Kumar, A. Macromolecules 2009 , 42 , 2015 2022. 61. Liu, M.; Reiser, O. Org Lett . 2011 , 13 , 1102 1105. 62. Yao, Z.; Yang, Y.; Chen, X.; Hu, X.; Zhang, L.; Liu, L.; Zhao, Y.; Wu, H. C. Anal. Chem. 2013 , 85 , 5650 5653. 63. a) VallÃ©e, M. R.; Majkut, P.; Wilkening, I.; Weise, C.; MÃ¼ller, G.; Hackenberger, R. Org. Lett . 2011 , 13 , 5440 5443. b) De, S.; Ramakrishnan, S. Chem. Asian J. 2011 , 6 , 149 156. 64. a) Wan, X.; Zhang, G.; Ge, Z.; Narain, R.; Liu, S. Che m Asian J , 2011 , 6 , 2835 2845. b) Lewis, W. G.; Magallon, F. G.; Fokin, V. V.; Finn, M. G. J. Am. Chem. Soc. 2004 , 126 , 9152 9153. 65. a) Tunca, U. Macromol. Rapid Commun. 2013 , 34 , 38 46. b) Kempe, K.; Hoogenboom, R.; Jaeger, M.; Schubert, U. S. Mac romolecules 2011 , 44 , 6424 6432. 66. Parrish, B.; Breitenkamp, R.B.; Emrick, T. J. Am. Chem. Soc. 2005 , 127 , 7404 7410. 67. van Zoelen, W.; Zuckermann, R. N.; Segalman, R. A. Macromolecules 2012 , 45 , 7072 7082.
113 68. Ji, X.; Dong, S.; Wei, P.; Xia, D.; Huang, F. Adv. Mater. 2013 , 25 , 5725 5729. 69. a) Burattini, S.; Greenland, B. W.; Merino, D. H.; Weng, W.; Sepalla, J.; Colquhouri, H. M.; Hayes, W.; Mackay, M. E.; Hamley, I. W.; Rowan, S. J. J. Am. Chem. Soc. 2010 , 132 , 12051 12058. b) Yin, C. R.; Han, Y.; L i, L.; Ye, S. H.; Mao, W. W.; Yi, M. D.; Ling, H. F.; Xie, L. H.; Zhang, G. W.; Huang, W. Polym. Chem. 2013 , 4 , 2540 2545. c) Blasco, E.; Schmidt, B. V. K. J.; Barner Kowollik, C.; PiÃ±ol, M.; Oriol, L. Macromolecules 2014 , 47 , 3693 3700. 70. Ranjan, R.; B rittain, W. J.; Macromolecules 2004 , 40 , 6217 6223. 71. Nicolas, J.; Mura, S.; Brambilla, D.; Mackiewicz, N.; Couvreur, P. Chem. Soc. Rev. 2013 , 42 , 1147 1235. 72. McEnaney, P. J.; Parker, C. G.; Zhang, A. X.; Spiegel, D. A. ACS Chem. Biol. 2012 , 7 , 1139 11 51. 73. Spiegel, D. A. Expert Rev. Clin. Pharmacol. 2013 , 6 , 223 225. 74. Ermer, O.; NeudÃ¶rfl, J. Helv. Chim. Acta. 2000 , 83 , 300 309. 75. Park, L. Y.; Hamilton, D. G.; McGehee, E. A.; McMenimen, K. A. J. Am. Chem. Soc. 2003 , 125 , 10586 10590. 76. Hase, Y.; Sal a, O. J. Mol. Struct. 1976 , 31 , 23 28. 77. Yang, S.; Li, X.; Hu, F.; Li, Y.; Yang, Y.; Yan, J.; Kuang, C.; Yang, Q. J. Med. Chem. 2013 , 56 , 8321 8331. 78. Karki, S. S.; Kulkarni, A. A.; Kumar, S.; Veliyath, S. K.; De Clercq, E.; Balzarini, J. Med. Chem. Res. 2013 , 22 , 2014 2022. 79. Mei, J.; Graham, K.; Stadler, R.; Reynolds, J. R. Org. Lett . 2010 , 12 , 660 663. 80. Metwally, M. A.; Khalifa, M. E.; Koketsu, M. American Journal of Chemistry 2012 , 2 , 38 51. 81. Gassman, P. G.; Cue, B. W.; Luh, T. Y. J. Org. Chem. 1 977 , 42 , 1344 1348. b) Kurkin, A. V.; Bernovskaya, A. A.; Yurovskaya, M. A. Chem. Heterocyc. Compd. 2011 , 46 , 1208 1214. 82. Chem. Eur. J. 2012 , 18 , 8498 84507. 83. Lehmann, M.; Jahr, M. Chem. Mater. 2008 , 20 , 5453 5456.
114 84. Synthesis 2007 , 11 , 1589 1620. 85. Antoni, P. ; NystrÃ¶m, D.; Hawker, C. J.; Hult, A.; Malkoch, M. Chem. Commun. 2007 , 2249 2251. 86. Montanez, M. I.; Campos, L. M.; Antoni, P.; Hed, Y.; Walter, M. V.; Kroll, B. T.; Khan, A.; Hult, A.; Hawker, C. J.; Malkoch, M. Macromolecules , 2010 , 43 , 1 13. 87. Steffe nsen, M. B.; Simanek, E. E. Angew. Chem. Int. Ed. 2004 , 43 , 5178 5180. 88. Vieyres, A.; Lam, T.; Gillet, R.; Franc, G.; Castonguay, A.; Kakkar, A. Chem. Commun. 2010 , 46 , 1875 1877.
115 BIOGRAPHICAL SKETCH Ashton N. Bartley was born in the beautiful town of Prosperity, South Carolina. Growing up in Small Town, USA, she honed her cooking skills to perfect the delicacies passed around the Sunday dinner table. She attended and graduated from the College of Charleston in 2011 with a B. S. in chemistry. Ther e she fell in love with the city of Charleston, Lilly Pulitzer, and Thursday morning beach trips. In the fall of 2011, she moved to Gainesville, FL, to attend graduate school at the University of Florida under the direction of Dr. Ronald Castellano. In t he years to come, Ashton hopes to take a general and organic chemistry lecture position to inspire the next generation of chemists.