136 Pages
English

Surface microstructuring for controlled drug release in coronary stents [Elektronische Ressource] / Michael Stöver

Gain access to the library to view online
Learn more

Description

Technische Universität MünchenLehrstuhl für MedizintechnikSurface Microstructuring for controlled Drug Release incoronary StentsMichael StöverVollständiger Abdruck der von der Fakultät für Maschinenwesen der TechnischenUniversität München zur Erlangung des akademischen Grades einesDoktor-Ingenieur (Dr.-Ing.)genehmigten DissertationVorsitzender: Univ.-Prof. Dr.-Ing. Horst BaierPrüfer der Dissertation:1. Univ.-Prof. Dr. med., Dr.-Ing. habil Erich Wintermantel2. Dr. Sannakaisa Virtanen,Friedrich-Alexander-Universität Erlangen-NürnbergDie Dissertation wurde am 13.09.2006 bei der Technischen Universität MüncheneingereichtunddurchdieFakultätfürMaschinenwesenam21.11.2006angenommen.AbstractThis thesis introduces a novel type of electrochemical microstructuring for 316Lstainless steel. The microstructuring methods were developed as a basis for a drugeluting coronary stent system. Coronary stents are used in order to dilate narrowedarteries. In order to prevent inflammation reactions, which lead to an excessiveproliferation of cells and thus to a renarrowing of the artery, the new generation ofstents are coated with drugs. The aim of this thesis was to create three dimensionalmicrostructures on the surface as a basis for drug coatings. The microstructures aresupposed to provide a protection for the drug during the implantation process and,moreover,controlthereleaseofthedrugbyprovidingmicrodepotsretainingacertainamount of the drug.

Subjects

Informations

Published by
Published 01 January 2006
Reads 20
Language English
Document size 10 MB

TechnischeUniversitätMünchen
LehrstuhlfürMedizintechnik

SurfaceMicrostructcouringronaforyrStecontntsrolledDrugReleasein

vStöhaelMicer

VollständigerAbdruckdervonderFakultätfürMaschinenwesenderTechnischen
UniversitätMünchenzurErlangungdesakademischenGradeseines

genehmigtenDissertation

Dr.-Ing.)(enieuroktor-IngD

Vorsitzender:Univ.-Prof.Dr.-Ing.HorstBaier

PrüferderDissertation:

1.Univ.-Prof.Dr.med.,Dr.-Ing.habilErichWintermantel
2.Univ.-Prof.Dr.SannakaisaVirtanen,
Friedrich-Alexander-UniversitätErlangen-Nürnberg

DieDissertationwurdeam13.09.2006beiderTechnischenUniversitätMünchen
eingereichtunddurchdieFakultätfürMaschinenwesenam21.11.2006angenommen.

Abstract

Thisthesisintroducesanoveltypeofelectrochemicalmicrostructuringfor316L
stainlesssteel.Themicrostructuringmethodsweredevelopedasabasisforadrug
elutingcoronarystentsystem.Coronarystentsareusedinordertodilatenarrowed
arteries.Inordertopreventinflammationreactions,whichleadtoanexcessive
proliferationofcellsandthustoarenarrowingoftheartery,thenewgenerationof
stentsarecoatedwithdrugs.Theaimofthisthesiswastocreatethreedimensional
microstructuresonthesurfaceasabasisfordrugcoatings.Themicrostructuresare
supposedtoprovideaprotectionforthedrugduringtheimplantationprocessand,
moreover,controlthereleaseofthedrugbyprovidingmicrodepotsretainingacertain
amountofthedrug.Itwasfoundthatsuitablemicrostructurescanbegenerated
byacombinedetchingmethodcomprisingafirstelectrochemicalgrainboundary
etchingstepfollowedbyanisotropicetchingstep.Thefirststepisperformedwith
nitricacidandcreatesamicrogridofnarrowgrainboundaryfurrows.Thesefurrows
arehollowedoutbyasecondetchingstep,carriedouteitherwithphosphoricacid,
hydrochloricacidoroxalicacid.Thegeneratedstructureswerefoundtobesuperior
orequalintermsofcorrosionpropertiesandcellproliferationbehaviorcompared
toconventionalgritblastedstents.Itwasshownthattheinvitroreleasebehavior
couldbesubstantiallydecelerated,withuptothreetimesmoredrugretentionafter
thefirstweekofreleaseincomparisontogritblastedstents.

I

sungZusammenfas

EswurdeeinneuesVerfahrenzurOberflächenmodifikationvonbiokompatiblem
MikrEdelstahlometerbentwicereickhelt,ausmitdendemintrdurcinsischhenelektroMatchemiscerialstrukheÄturentzungerzeugtMikrwoerstrdenukturkenönnen.im
DasVerfahrendientalsBasisfüreineMedikamentenbeschichtungvonStützenfür
verengteHerzkranzgefässe(Stents).EingroßesProblembeidieserBehandlungist
eineDiesehkoheanndurcWiedervheineerscBeschlussrate,hichtungdiedesdurchStenEntsmitzüntendungsrtzünduneaktgionenshemmendverursacenhtMedikwird.a-
mentenverhindertwerden.MitHilfedesVerfahrenswerdenineinemzweistufigen
ÄtMedikzprozamessentaMikroufzunehkavitätenmenundverzeugt,erzögdieertindfreizuseterzLagene.Essind,wurdeeinenagechwigsseewiesen,Mengedaassn
dieerzeugtenOberflächeninHinblickaufinvitroKorrosions-undZellbesiedelungs-
eigenschaftenbesserodergleichwertigzukonventionellensandgestrahltenOber-
flächensind.DerinvitroMedikamentenReleasekonnteerheblichverzögertwerden,
dieerzeugtenOberflächenwiesennacheinerWocheReleasenocheineimVergleich
zusandgestrahltenStentsbiszudreimalhöhereMengeangespeichertemMedika-
auf.tmen

II

Acknowledgements

IliketothankProf.Dr.Dr.ErichWintermantelforprovidingmetheopportunity
todevelopthisthesis,forhiscordialityandtheproductiveatmosphereinhisinsti-
e.tutMyrefereeProf.Dr.SannakaisaVirtanenalsodeservesmygratidueforherinterest
inthetopic,hersupportandherproductiveinputs.
SpecialthankstoDr.JuliaWillwhoguidedmethroughthefirstandmostdemand-
ingpartofthisthesisandcontributedsubstantiallytothesuccessofthiswork.
IalsowouldliketothankDr.JuliaKunzeforsupervisingtheelectrochemicalanal-
ysis.IamcommittedtogreatthanktoGernotKrobathforthetirelessthoroughand
criticalproofreading.InthiscontextIalsoliketothankDr.MarkusEblenkamp
andMichaelBrosch.
MycolleguesMagdaRenke-GluszkoandDr.TomSchratzenstallerdeservethanks
forthegoodcooperationandsupport.ThanksalsotoDr.BorisBehnischfrom
TransluminaLabsforprovidingthecountlessamountsofstentsandfortheproduc-
ions.discussetivMoreoverIliketothankallotherpersonswhocontributedtothesuccessofthiswork:
SusanneSchnell-Witteczek,RalfRettig,SteffiBlechinger,BarbaraRöhrnbauer,Do-
minikRietzel,FlorianBrunschön,ShadiSabeti,NilsKlink,HannaDressel,Adrian
Frenzel,UliEbnerandJosephHintermair.
NotatleastIwouldliketothankmyparentsandmysisterfortheirconsistent
supportatallmydecisions.

III

tentsCon

1Introduction1
1.1SurfaceModificationofImplants....................1
1.2StentsasaTreatmentofCoronaryHeartDisease...........3
1.2.1PathologyofAtherosclerosis...................3
1.2.2CoronaryStenting........................5
1.2.3PathologyofRestenosisandStrategiesforitsPrevention...6
1.2.4DrugDeliveryStentSystems...................8

2WorkingHypothesisandProceeding

10

3ElectrochemicalTheory16
3.1Basics...................................16
3.2SelectiveEtching.............................18
3.3TheAlloyingElementsof316L.....................19
3.4ElectrochemicalBehaviorofCr-Ni-Steels................21
3.4.1EffectofHydrochloricAcidonStainlessSteel.........22
3.4.2EffectofNitricAcidonStainlessSteel.............25
3.4.3EffectofOxalicAcidonStainlessSteel.............26
3.4.4EffectofPhosphoricAcidonStainlessSteel..........27

4MaterialandMethods29
4.1Material..................................29
4.2TheEtchingAssembly..........................30
4.2.1CalculationofCurrentDensities.................33
4.3TopographicSurfaceAnalysis......................34
4.4CrossSectionalMicrographs.......................36
4.5SoftwareAnalysisoftheDepotStructures...............37
4.6CalculationofDepotProperties.....................37
4.7CoatingProcessandAnalysisofCoatings...............39

IV

5

6

7

8

4.8ChemicalSurfaceAnalysis........................
4.8.1SEM-EDS.............................
4.8.2AugerElectronSpectroscopy...................
4.9ElectrochemicalAnalysis.........................
4.9.1CyclicVoltammetry.......................
4.9.2PotentiostaticTestsforDeterminationofIonRelease.....
4.10CellSeedings...............................
4.10.1SeedingProcess..........................
4.10.2PreparationforSEM-Imaging..................
4.11MechanicalTests.............................

MicrostructuringwithHydrochloricAcid
5.1Results...................................
5.2AnalysisoftheFormationProcess....................

MicrostructuringwithNitricAcid
6.1Results...................................
6.2AnalysisoftheFormationProcess....................
6.3ChemicalAnalysis............................
6.4MathematicalDescriptionoftheEtchingProcess...........

CombinationEtchingsonNitricAcidBasis
7.1PhosphoricAcid..............................
7.1.1ParameterAnalysisonStents..................
7.1.2SoftwareAnalysis.........................
7.2HydrochloricAcid.............................
7.2.1ParameterAnalysisonStents..................
7.2.2SoftwareAnalysis.........................
7.3OxalicAcid................................
7.3.1ParameterAnalysisonStents..................
7.3.2SoftwareAnalysis.........................
7.4EtchingswithoutpreviousElectropolishing...............

PerformanceofthemodifiedStents
8.1ChemicalPerformance..........................
8.1.1CyclicVoltammetry.......................
8.1.2ReleaseTestsofNickel......................
8.1.3AnalysisofthePassivityLayer.................
8.2BiologicalPerformance..........................

04404014424574474884

525275

6060267686

7247747708808248848688

901991949559

V

8.2.1CellSeeding............................95
8.3MechanicalPerformance.........................100
8.3.1RadialStrengthTests.......................100
8.3.2FatigueTests...........................101
8.4CoatingBehavior.............................102
8.5ReleaseKinetics..............................104

9SummaryandOutlook108
9.1Summary.................................108
9.1.1MicrostructuringwithHydrochloricAcid............108
9.1.2MicrostructuringwithNitricAcid................109
9.1.3CombinationEtchings......................110
9.1.4Performance............................111
9.1.5Conclusion.............................112
9.2Outlook..................................112

114dixenApp1010.1DeterminationofGrainSize.......................114
10.2CalculationofPolygonLine.......................114
10.3CellCultures...............................115
10.4CompositionofRingersSolution.....................117
10.5ApproximationofStrain.........................117

VI

onductintroI1

1.1SurfaceModificationofImplants

Apluralityofsubstanceshasbeenestablishedforthepurposeofsubstitutingor
supportingorganicfunctionsofthehumanbody,rangingfrommetalstoceramics
topolymers.Almostallmaterialscurrentlyusedarestillbeingoptimizedinterms
ofcompositionandprocessing.Inrecenttimes,paralleltothedevelopmentinother
fields,thefocushasbeenaimingmoreandmoreatthesurfacepropertiesofthema-
terial.Sincebodytissuenormallyinteractsonlywiththeupmostfewmicrometers
ofanimplantedmaterial,thechemicalandtopographicpropertiesoftendecideover
thesuccessorfailureofanimplantationprocess.Dependingontheintendeduse
ofanimplant,varioustopographiesmayberequired.Thedesiredsurfacemaybe
verysmoothasinthecaseofimplantsdirectlyincontactwiththebloodflow(e.g.
artificialheartvalves),astructuredsurfacewithveryhighroughnessasinthecase
ofpermanentimplantswheregoodadhesionandquickcellingrowthareimportant
(e.g.shaftsofhipimplants)orsurfacestructureswithintermediateroughness.An
examplefortherequirementofintermediateroughnessaretemporaryimplantslike
screwsandnailswhereasufficientadhesionisrequired,withthepossibilityofa
subsequentremovaloftheimplant.Anotherexamplewherethesurfaceproperties
havetobeadjustedcarefullyarepermanentboneimplantsnearsensitiveareaslike
thebrain.Inthiscaseahighsurfaceroughnessmaybebeneficialtopromotecell
ingrowth,howeverifthisprocessisoverdrawntheresultmaybeanaboundingrowth
oftheimplantintothesensitiveareas[1,2,3].

Commonlyusedtechniquesinordertomodifysurfacesincludechemicaltreatments,
laserstructuringaswellasmechanicalsurfacetreatments.Mostofthesetechniques
areconventionaltechniqueswhichhavebeenadaptedforuseinbiomedicalengineer-
ing.Asimple,widespreadsurfacemodificationtechnique,withwhichanincrease
insurfaceroughnesscanbeachieved,isgritblasting.Anincreasedroughnessmay
providebothimprovedadhesionpropertiesandafavoredbasisforcellgrowth[4,5].
Accordingly,gritblastingisappliedforpermanentimplantswhichrequireaconsol-

1

idatedingrowth,e.g.shaftsofhipimplants,dentalimplants[6,5].Thetechniqueis
comparativelyeasytoperformandapplicableforlargequantities.However,forgrit
blastedimplantsitisknownthatariskofremainingparticlesexists[7,8].Another
limitationresultsfromthefactthatmechanicalstressesareimposedonthematerial.
Whenappliedtosensitivestructures,gritblastingcarriestheriskofdeformations
andmayleadtoconsiderablescraprates.

Fortitaniumimplantsalsochemicalandplasmachemicalsurfacemodificationsare
subjectofcurrentresearchwithpromisingresults[9,10,11].Inthiscasehighly
reactive,mostlyfluorinecontainingchemicalsareusedinordertoetchthesurfaces.
Similartogritblasting,withthismethodsthesurfaceroughnesscanbeincreased,
resultinginpromotedcellingrowth.

Anothermicrostructuringmethodisofferedbylaserablation.Withthismethod
threedimensionalstructurescanbecreated.Thesestructuresmaybeusedto
promotecellingrowth,howeveranevenmoreimportantfeatureofferedbythese
structuresisthepossibilitytoloadthesurfacewiththerapeuticsubstances.This
waydrugscanbeapplicateddirectlyintotheaffectedlocation,resultinginahigh
therapeuticefficacyatsmallamountsofdrugsneeded.Examplesfordrugcoatings
areantibiotics,antithromboticagentsaswellascellgrowthstimulants(e.g.[12,13]).

Othermicrostructuringtechniquesthatgivethepossibilitytocreatethreedimen-
sionalstructuresoriginatemainlyfromthefieldofmicrochipfabrication.Photo-
andlaser-lithographictechniquesareemployedbydepositingaprotectionlayerof
lacqueronthesubstrate.Subsequentlythemicrostructuresareetchedelectrochem-
icallyintothematerial.Withmultistepetchingsitispossibletogeneratethree
dimensionalstructuresaswellasundercuts[10,14,15].Thetechniqueistime-
consumingandverydifficulttoapplyforcomplexgeometries.

Inthisthesisanovelmicrostructuringtechniqueforstainlesssteelmaterialsisintro-
duced,whichcombinestheadvantagesofaselectivethreedimensionalstructuring
cwithhemicalthecoetcnhving,enienceinaofwaaymerethatcinhemitrinsiccalmmethateroiad.lstTheructurteceshniqauerebausedsesinonelorderecttro-o
formthreedimensionalstructuresoutofthematerial.Thetechniquewasdeveloped
forcprotectivoronaryepolystentmers.sasaThebasisbasicforrequiranoemenn-sitetwdrasugthecoadevtingelopsystemmentofwitahomiuctrothestruseucturo-f
ingmethodwhichiseconomicalandapplicableonmechanicallysensitivestructures.

2

Themicrostructuresshouldoffersufficientadhesionforadrugcoatingand,after
elutionofthedrug,serveasafavoredsurfaceforcellingrowth.Itwasfoundthat,
beyond,withacombinationofspecialelectrochemicaletchingtechniquesmicrode-
potscanbegeneratedwithinthesurface,whichofferthepossibilitytostoredrugs
withinthesurface.Withthiscombinationtechniquemicrodepotsofdifferentsize
anddifferentshapescanbegenerated,whichallowstotailorthedepotproperties
accordingtotherespectiveuse.

1.2StentsasaTreatmentofCoronaryHeart
eDiseasCoronarystentsareemployedtodilatenarrowedcoronaryarteriesinordertoensure
asufficientbloodflow.Theyconsistofacylindricalwire-meshofbiocompatiblema-
terialwhichisplacedintothearteryandisdilatedbymeansofaballooncatheter
(fig.1.1).Afterwardthestentremainspermanentlyintheartery,actingasame-
chanicalscaffoldwhichkeepsthevesselopen.Theindicationforauseofstentsisthe
occurrenceofcoronaryheartdiseaseasaconsequenceofatherosclerosis.Symptoms
ofthisdiseaseareheartinsufficiency,anginapectorisaswellaspendingorsuffered
cardiacinfarction[16,17].

1.2.1PathologyofAtherosclerosis

Atherosclerosisdescribesadegenerationprocessofarterialbloodvesselswhichleads
toareducedor,intheworstcase,aninterruptedbloodflow.Atherosclerosistrans-
latedfromGreekmeansasmuchasgruel-likerigidification.Atherosclerosisiscaused
byintra-arterialdeposits,originatedbyconglomerationoflipids,complexcarbohy-
drates,bloodandconnectivetissueaswellascalciumdeposits[18].Theprocess
normallystartswiththegenerationofalipidenrichedareaunderneaththeintima.
Thearteryconsistsofthreelayersofwhichtheintimaistheinnermostlayerofthe
artery.Itconsistsofamonolayerofendothelialcells.Themiddlelayer,themedia,
consistsofcircumferrentiallyorientatedsmoothmusclecellsandcollagenfibers.The
outerlayer,theadventitia,comprisesconnectivetissuemadeupoffibroblastsand
associatedcollagenfibers[19].Withinthelipidenrichedareabeneaththeintima

3

Figure1.1:Schematicillustrationofcoronarystenting(withpermissionoftheMedical
UniversityofSouthCarolina).Thestentisplacedintothenarrowedareaonaballoon
catheter.Afterdilatationofthestentthecatheterisremoved,whilethestentremains
permanentlywithinthevessel,preventingarenarrowingofthevessel.

defensecellsdifferentiateintosocalledfoamcellsthroughincorporationoflipiden-
richedmaterial[20].Althoughexcessiveresearchhasbeenperformedoverdecades,
theexactcauseforthisprocessisnotyetfullyunderstood.Asearlyas1976itwas
suggestedthattheoriginisaninflammatoryprocesswhichstartswithminorinjuries
oftheendotheliallayer.Thesemicroinjuries,whichmaybecausedbymorphologic
damages,biochemicalimpairments,bacterialtoxinsetc.,enablethepermeationof
defensecellsthroughtheendotheliallayer[21,22].Inconnectionwiththepresence
ofaspecialprotein(oxidizedlowdensitylipoprotein,LDL),whichisassumedto
playthedecisiveroleinthedevelopmentofatherosclerosis,thesecellsareableto
differentiateintofoamcells.Subsequentlyamigrationofsmoothmusclecellsfrom
themediaintotheintimaconnectedwithanincreaseofextracellularfibrilscanbe
observed.Theresultisacapsulartypestructureofconnectivetissuewithanadipoid
core(plaque)[23](s.fig.1.2).

Intheprogressofthediseasethealteringofthevesselscanleadtoaseverimpair-
mentofthebloodflow.Thiscanbecausedeitherbyagradualnarrowingofthe
arteryorbyaruptureoftheplaque[24].Theresultarethetypicalconsequencesof
thecoronaryheartdiseaselikeanginapectoris,formationofthrombusesandfinally
k.tactaheart

4

Figure1.2:Sectionthroughanarteryaffectedbyatherosclerosis(Source:
www.tmc.edu/thi/cad1.gif).Intra-arterialdepositsconsistingoflipids,complexcarbohy-
drates,bloodandconnectivetissueaswellascalciumdepositsleadtotheformationof
plaquebeneaththeendothelium.Thefinalresultisanimpairmentofthebloodflow.

1.2.2CoronaryStenting

Associatedwiththetendencytowardminimalinvasiveinterventionsandthesteady
improvementofthetechnique,coronarystentinghasbecomethemethodofchoice
forthetreatmentofcoronaryheartdiseasesymptoms.Thestentingprocedureis
carriedoutbyplacingastentcrimpedonaballooncatheterintothenarrowedartery.
Thecatheterisnormallyinsertedthroughthefemoralartery,occasionallythrough
thecarotidartery.Atthesiteofthestenosisthestentisdilatedwithupto16atms
usingRingerssolutionforthedilatationprocess.

Thepreferredstentmaterialismedicalgradestainlesssteelbecauseofitsmechanical
propertiesanditsgoodmachinabilitycombinedwithfairX-rayvisibility.Toalesser
extentNitinolstentsandstentsmadeofcobaltalloysareused,whileNitinolallows
tofabricselfexpandablestentsbecauseofitssuperelasticbehavior.Themajorityof
stentsaremeanwhilefabricatedoutofsmalltubesbymeansoflasercutting.This
techniquemostlyreplacedoldermethodsusingwiremeshesorspirallywoundwires.
Typicaldimensionsofcoronarystentsareonetotwomillimetersdiameter(before
dilatation)and8to25mmlength.Thethicknessandwidthofthesinglestrutsare
intherangeof100µm.

5

1.2.3PathologyofRestenosisandStrategiesforits
tionPreven

Themainproblemoccurringwiththeuseofstentsisarenarrowingoftheartery
throughproliferatingtissue(e.g.[25]).Thisproliferationistoberegardedasanin-
flammatoryresponseagainsttheintroducedforeignbodyaswellasthemicroinjuries
atthearterialwallthatarecausedbythedilatationprocess.Inthefirststagean
activationofcirculatingthrombocytescanbeobserved[26].Theimplantationofa
stentresultsinacompletedestructionoftheendotheliallayerwithintheareaofim-
plantation[27].Asaresultathinthrombusisformedbetweenthestentsurfaceand
thevessel.Inthefirstweeksthisthrombusworksasaborderbetweenthestentma-
terialandthevesselwall.Inordertopreventtheriskofthrombusformationwithin
thevesselhighdosesofacetylsalicylicacidaregivenduringthisstage.Another
processthatisinducedwithinthefirstweeksisaproliferationofsmoothmuscle
cells.Thiseffectisenhancedbytheshearstressesinducedintothearterialwall
thatstimulatethegrowthofsmoothmusclecells[28].Ifthisproliferationpersists
itleadstoaconsiderablelumenlosswhichisreferredtoasrestenosis.Parallelto
thephaseofproliferationofsmoothmusclecellsexpandingendothelialcellclusters
developwhichfinallyescalatetoacoherentcelllayer.Sinceendothelialcellswork
asaregulatorforsmoothmusclecells,thegenerationofanendothelialcelllayeris
themaincriteriaforthepreventionofanexcessiveproliferationofsmoothmuscle
cells[29,30,28].Greweetal.stateatimeintervalofaboutthreemonthforthis
basedontheirexperimentsonhumanarteries[31].Thepropagationofendothelial
cellsdependonthechemicalaswellasthetopographicalpropertiesofthestent
surface.Thepropagationisfavoredbysurfacestructuresintherangeof15-20µm
thatpromoteadhesionandingrowthofthecells[32,33].

Asthemosteffectivemeansagainstrestenosisdrugsareusedthatinhibittheprolif-
erationofsmoothmusclecellsand/orpromotequickreendothelization.Aplurality
ofsubstanceshavebeeninvestigatedforthispurposewhichcanbedividedintofour
classes.Thefirstclassareanti-inflammatorydrugsthatintervenewiththeimmune
response.Amongstthiskindofsubstancesisthewellprovenimmuno-suppressant
Rapamycin[34,35]aswellasthewidespreadanti-inflammatorydexamethasone
[36].Thesecondclassarethedirectlyanti-proliferativesubstancesthatinhibitthe
progenyofsmoothmusclecells.Paclitaxelasanothercommonlyusedstentcoating
drugisanexampleforthisclass[37,38].Lesscommonlyusedareselectivemigration
inhibitors(e.g.Baltimastat)anddrugsthatsupporthealingandreendothelization

6

(e.g.growthfactorsorthehormoneestrogen)[39].

Inordertoguarantyasufficientlyhighdrugconcentrationandtominimizeside
effects,itissensibletodeliverthedruglocallyinsteadofasystemicdelivery.There-
foreacoatingofthestentsisnecessary.Twomainproblemsoccurwiththeuse
ofdrugcoatedstentsintermsofsurfaceproperties.Thefirstproblemarestresses
imposedonthecoatingduringtheimplantationprocesswhichcanleadtoflaking
ofthecoating,accompaniedbyaconsiderabledrugloss.Thesecondproblemisto
provideadeceleratedreleasekineticinordertoguarantyasufficientefficacyofthe
drug.Currentlymostdrugelutingstentsarecoveredwithapolymerinordertopre-
ventflakingandtobeabletocontrollthereleasekinetics[40,41,42].Duetobetter
mechanicalproperties,mostlynondegradablepolymersareused.Howeverrecent
dataindicatethattheuseofnondegradablepolymerscanleadtosubstantiallong
termsideeffectslikelaterestenosis[43]andthromboticstentocclusions[44].These
longtermsideeffectsareattributedtoachronicinflammationeffect,expressedby
alocalaccumulationofdefencecells[45].Moreoveritwouldbebeneficialtoindi-
viduallychoosetypeandconcentrationofthedrug,accordingtothehistoryand
thediseasepatternofthepatient[46].Thisisnotpossibleforthecurrentlyused
polymercoatingssincethestentsarecoateddirectlyafterthefabricationprocess.
Duetothecomplexityoftheprocessthisstepnormallycannotbeperformeddirectly
beforeimplantation.Sincedifferentgeometriesareemployedforeachpatientitis
alsonotpossibletostorecoatedstentswithdifferentdrugconcentrations.

Inrecenttimecoatingsystemshavebeendevelopedwithwhichstentscanbecoated
withinthecatheterlaboratoryimmediatelybeforeimplantation,preferablywithout
theuseofapolymer.Inthiscasehighdemandsaresetonthesurfaceproperties.
Themainrequirementisgoodadhesionofthedrugwhichisrealizedbyincreasing
thesurfaceroughnessofthestent.Forthispurposegritblastingisused,whichis
theonlysurfacemodificationmethod,knowntotheauthor,whichiscommercially
usedforstentswithoutpolymercoatings.Besidesfunctionalbenefits,stentswith
increasedsurfaceroughnessmightofferanotheradvantage.In2005arandomized
studyonhumans(n=200)wasperformedbyDibraetal.comparinguncoatedgrit
blastedandelectropolishedstents[47].Inthisstudynosignificantdifferencecould
beobservedbetweenthetwosurfaces.Neverthelessitwasfoundthattherough
surfacestenttoyieldslightlylowerrestenosisratesthansmoothsurfaceswithout
havingnegativesideeffects.Theresultisattributedtoabetterendothelialization
ofthesurface.Theeffectofcontrolledcellpromotioncanalsobeachievedvery

7

efficientlybycreatingdefinedmicrostructuresonthesurface[48].

Anevenmoreimportantrequirementthathastobeprovidedbythetopographyof
thesurfaceistheachievementofadelayedreleasekinetic.Withouttheuseofan
additionalpolymermostofthedrug(Rapamycin)onagritblastedstentisremoved
withinafewdaysafterimplantation.IninternalstudiesthepeakoftheRapamycin
concentrationinthesurroundingtissuewasfoundonthethirddayafterimplan-
tation.Atthesixthdaytheconcentrationhaddroppedtoonethirdofthepeak
value.AsufficienttimeperiodforRapamycinandPaclitaxeltoeffectivelyreduce
restenosis,however,isassumedtobeintherangeofweeks(e.g.[49,43,50,51,52]).
InthiscontextDudaetal.stateforRapamycin[52]:"Itisnotclearhowlongthe
drugneedstoremaininthearea,butitappearsthat4–6weeksissufficientbasedon
theclinicaldatageneratedtodate".Aninterestingfactinthiscontextisdescribed
byParryetal.[53]whofoundininvitroexperimentsthat"theeffectsofsirolimus
onsmoothmusclecellproliferationwerereversiblesincecellsresumedproliferation
severaldaysafterwashoutofsirolimus".

Adifferentapproachisofferedbylaserablationtechniques.Withshortlaserpulses
smallslotscanbecutintosurfaces[54],thatcanbeusedasdrugdepots.This
methodbringsaboutthelimitationthatonlyratherlargedepotscanbegenerated
sothatadrugcoatingisdistributedinhomogeneouslyandthereleasekineticsare
aonlylargeslighdetlygree,imsoprovthated.aFnuradditthermoreionalptheolymeradhesioconatingpropisertiesstillareneedednot.Aimprpoverspedtec-o
oftivaenapplimmicaunotionsuppforressthisantmethowhicdhcoisfouldllobweaedmbyultian-layerendothelialcoating,cellforgroexawthmplsetaimreleaulator.se

1.2.4DrugDeliveryStentSystems

Differentdrugreleasingstentsystemsarecurrentlyused,themostcommonbe-
ingpolymercoatedRapamycin(BXVelocitystent,Cypherc)orPaclitaxeleluting
stents(NIRxstent,BostonScientific).TheTaxusstentcomprisesanondegradable
polymer(onahydrocarbonbasis)coatingcontainingPaclitaxelintwomodifica-
tions,aslowandafastrelease.Theslowreleasetypefeatures1µgpaclitaxelper
mm2withareleasetimeof28days,thefastrelease2µgpaclitaxelpermm2)witha
24-hourreleasetime[55].TheCyphercstentcoatingconsistsofanondegradable
polymeronametacrylatebasisthatcontainsRapamycininaconcentrationof140

8

µm/cm2andatoplayerasdiffusionbarrier.Thestentreleases80%oftheamount
ofRapamycinwithinthefirstmonth[52,49].Alternativelysurfaceroughenedstents
withRapamycincoatingareusedwithouttheuseofapolymer(Yukonc).This
stentusesapureRapamycincoating,appliedviaspraycoatingofanethanolicRa-
pamycinsolution.AcommerciallyusedpurePaclitaxelcoatingisnotknowntothe
author,howeverDudaetal.statethatitisevenpossibletoapplypaclitaxelasa
simplecoatingwithoutthenecessityofapolymer.[55].

Asignificantreductionoftherestenosisratecouldbeprovedforallthreestents.
TheSIRIUSTrialshowedareductionfrom35.4%to3.2%fortheCyphercstent,
PintheaclitaxelTAXUstenStIV[56].TriaThleareductrestenosisionrefromduction24.4%oftothe5.5Y%ukoncouldcbestenactishievsligedthwitlyhlowtheer
thanthatoftheCyphercwitharestenosisrateof14,7%comparedto25,9%for
cothensibarederablemetalwhenstentrisks[57].factorsThelikerestdiabenoetessis(17rates.6%)oftorhestheesstentents,tinghowofevaer,arebifurcatedstill
vessel(28.0%)arepresent[58].

9

2WorkingHypothesisand
ingeedcPro

Thisthesisaimsatthedevelopmentofmicrostructuredsurfacesoncoronarystain-
lesssteelstentsbymeansofelectrochemicaletching.Thesurfacesaredesignedina
waytoenableacoatingofthestentwiththerapeuticdrugs,preferablywithoutthe
useofanadditionalpolymer.Theprocedureofstentingwithdrugelutingstents
canbeseparatedintothreestages,witheachonehavingdifferentrequirementson
thesurfaceproperties.

Thefirststageaddressestheimplantationofthestent.Duringtheimplantation
processthedrugcoatingissubjectedtofrictionforcesduetocontactwiththear-
teries,mechanicalstressesduetostentdeformationsatturningsandshearstresses
inducedbythebloodflow.Thebasicrequirementinordertoassureaprotectionof
thedrugduringthisphaseisasufficientadhesionofthedrug.Thiscanberealized
byahighsurfacemicroroughnessandbyamechanicalinterlockingofthedruglayer
andthestentsurface.

Oncethestentisplacedwithinthestenosedarterythephaseofthedrugrelease
begins.Inthisstageacontrolled,gradualreleaseisrequiredinordertoassurea
maximumefficiencyofthedrug.Theideafollowedinthisthesisistocreateevenly
distributedmicrodepotsonthestentsurface,thatstorepartofthedrugcoating,
inordertoachieveacontrolledreleasewithoutanadditionalbarrierlayer.The
microdepotsarecreatedonthethreeoutersidesofthestent(s.fig2.1),whichwill
immediatelybeincontactwiththetissueafterdilatation.Untilnowlittleisknown
abouttheidealsizeandshapeofsuchmicrodepots.Smallanddeepdepotswill
retainasmallamountofdrugforalongtime,whilebroadshallowdepotswillstore
alargeramountofdrugforashorttime.Inthisthesisthemicrostructureswere
designedinawaytoyieldadepthtowidthratiointherangeof1to2,inorderto
haveasuitablecompromisebetweenasufficientretainingcapacityandasufficient
amountofstoreddrug.Thedepthofthemicrodepotsislimitedduetomechanical

10

aspects,sincethemicrodepotsactasnotchesandthusweakentheintegrityofthe
material.Whenassumingdepotswitharoundedbase,adepotof10µmdepthand
5µmwidth,wouldhaveanotchradiusof2.5µmandaratioofdepthtonotch
radiusof4.Internstudiesshowedthatconventionallyapproved(e.g.gritblasted)
stentsmayhavenotchnumbersashighas10,sothatthisnotchnumbersarecon-
dimisiderednisheduncritbyicanotlbytmorehethanauthor.20µThemwhiceffectivhecanloabedblevearieledngoutareabyinanthiscaseadaptationwillbofe
thestentdesign.

Thethirdstageofthestentingprocedureisthebarestentphase.Afterreleaseof
thedrug,whichpreviouslyformedabarrierbetweenstentandtissue,thestentisin
directcontactwiththetissue.Inordertoassurealongtermfixationofthestent
withinthetissueandtopreventchronicinflammationprocesses,anoptimalsurface
hastobebiocompatibleafterreleaseofthedrugandshouldprovideasuitablebasis
forcellingrowth.Thelattercanbeachievedbymicrostructureswhichsupportthe
ingrowthofcells.InthiscontextitwasshownbyPalmazandbyFussthatstruc-
turesintherangeofthesizeofcells(about10-20µm)canpromotecoordinatedcell
growth[32,33].

Preliminaryexperimentswithetchantslikeaquaregiashowedthatitwasnotpos-
sibletogeneratesuitablesurfaceswithconventionaletchingmethods.Theseex-
perimentsyieldedsurfaceswithmoderatelyincreasedroughness,howevertheother
requirementscouldnotbemet.Thedecisiveideathatisfollowedupinthisthesis
originsfrommetallographicpractices.Theapproachwastoaggravatetheselectivity
ofetchings,thatareusedinthefieldofmetallographyinordertovisualizegrain
patterns,byelectriccurrentinawaythatthreedimensionalstructuresareworked
outoftheintrinsicstructuresofthematerial.Twoelectrochemicaletchingtech-
niqueshavebeendeveloped,withwhichitispossibletogenerateevenlydistributed
microstructuresofseveralmicrometersdepthonmedicalgradestainlesssteel.The
firstoneisbasedonhydrochloricacidetchingandrendersanevenlyfacetedsurface.
Thisetchingmethodyieldsahighsurfaceroughness,howeverwithoutprovidingde-
pots.Duetobadreproductivityandunexpectedbadadhesionbehaviorthismethod
wasfoundnottobesuitableasabasisfordrugcoatingofstents.Thesecondmethod
isbasedonnitricacidetchingandproducesafinegridoffurrowsonthesurface.The
furrowscreatedwithnitricacidcanhavedifferentshapeswhencontemplatingthe
crosssectionalview.Dependingontheselectivityoftheetchingprocesstheforms
canrangefromratherbroadV-shapestonarrow,crevicelikefurrows.Basingon

11

Figure2.1:Schematicillustrationofamicrostructuredstentstrut.Thecrosssectional
viewbelowillustratesthedistributionofthedepots.Theinner(luminal)sideofthestent
doesnothavedrugdepots,sincethissideissubjectedtothebloodflowafterimplantation,
sothatadrugwouldbewashedoutquickly.

thisobservationamathematicalmodelwasestablishedinordertopredicttheshape
andsizeofthefurrows.Independentoftheshapevariations,themicrostructures
generatedwithnitricacidhavesharptipsandrelativelylowdepotvolume,making
themunsuitableasaonestepetchingmethod.

However,itwasfoundthatespeciallythenarrowercrevicesarewellsuitedasabasis
fortwostepetchingmethodsasdescribedinchapter7.Thetwostepetchingmethod
comprisesthecreationofnarrowfurrowswithnitricacid,whicharehollowedoutby
asecond,isotropicetchingstep.Threeacidsolutionswereinvestigateintermsof
suitabilityforthissecondetchingstep,phosphoricacid,hydrochloricacidandoxalic
acid.Thegeneratedtopographieswereanalyzedbyscanningelectronmicroscopyas
wellasbysoftwareanalysisofcrosssectionalmicrographs.Forthesoftwareanalysis
apolygonlinewasgeneratedaroundthestentsimulatingtheboarderlinebetween
thearterytissueandthestent.Usingthismodeltheaveragedepotvolume,the
averagedepotdepthaswellasthedistributionofthedepotsofthegeneratedstruc-
tureswerecalculated.

Furthermoreitistestedifanelectropolishingstepasanormallyobligatorypretreat-

12

mentcanbewaivedwhenusingelectrochemicaletchtreatments.Thefabricationof
stentsbylasercuttingleadstosharpedgesandburrsthatrequireanafter-treatment.
Forconventionalstentsafirstpicklingstepwithahydrofluoricacidcontainingso-
lutionisperformeddirectlyaftercuttinginordertoremoveburrs.Subsequently
anelectropolishingtreatmentisnecessarytoroundedgesandsmooththesurface.
Whenapplyingthecombinedelectrochemicaletchings,thesecondstepproducesa
smoothingeffect,thereforethepicklingstepmightbespared.

Anotherwelcomesideeffectofelectrochemicaltreatmentsis,thatthesurfaceis
cleanedandchemicallyactivatedbytheetchingprocess,sothattheadhesionprop-
ertiesmightbefurtherimproved.Animplementationofthisprinciplecanbefound
inthepreparationofsurfacesforbondingprocesses[59,60].However,theetching
processmightinfluencethesurfacepropertiesinbothanegativeorapositivewayin
respecttocorrosionproperties.Mostauthorsfoundanincreasedchemicalresistance
of316Lafteretchingtreatmentswithvariousacidssuchasnitric,phosphoricand
hydrochloricacids[61,62,63].Incontrary,sometimesalsonegativeeffectsfrom
nitricacidetchingsarereported,likeanincreasedsusceptibilityagainstcrevicecor-
rosion[64].Thereforechemicalandelectrochemicalanalysisofthemodifiedsurfaces
wasperformedinordertoevaluatethecorrosionbehaviorandpossiblechangesof
thesurfacecompositionofthestents.

Inordertoassesstheefficacyofthesurfaces,drugquantificationswereperformed
andthereleasekineticswereanalyzedincomparisontogritblastedstents.The
mechanicalperformanceofthemodifiedstentswasevaluatedbystaticstrengthand
dynamicfatiguetests.Finallycellseedingtestswereperformedtocontemplatebio-
compatibilityaspectsofthesurfacemodifications.Fibroblasts,whicharerelatively
robust,versatilecells,wereusedinordertoassessthegeneralbiocompatibilityprop-
erties.Inadditionthesurfaceswereseededwithendothelialcellsandwithsmooth
musclecells.Theintentionoftheseexperimentswastoobtaininformationabout
thepossibilityofinfluencingcellsindividuallybythemicrostructures.

13

14

Fiegur.2:2veySur

of

Proceeding

,I

Creation

fo

the

es.facsur

Fiegur2.:3veySur

of

Proceeding

I,I

sinalysA

fo

the

sures.fac

15

3ElectrochemicalTheory

csiasB3.1

Electrochemicaletchingisawellestablishedmethodfordifferentusesintheareaof
metalprocessingandanalysis.Themainapplicationsaresparkerosionprocessing
aswellasthevisualizationofgrainstructuresformetallographicanalysis.Thefun-
damentalsofelectrolyticetchingwillbeexplainedinthefollowingsection.

Forthefollowingcontemplationstwometalelectrodesareassumedwhicharesur-
roundedbyaliquidelectrolyte.Miscellaneoussolvents(e.g.water,methanol)with
mobilechargecarriers(e.g.NaCl,HNO3)canbeusedaselectrolytes.Thelarger
thesolventmolecules,thelargeristhesolvatecoveringaroundtheionsandthusthe
loweristhemobilityoftheions.Theconductivityofanelectrolyteisafunctionof
thenumberofsolvedchargecarriers,theirchargeaswellastheirmobility:
•Atlowerconcentrationstheconductivityincreaseswithincreasingconcentra-
tionsbecauseofanincreasingnumberofchargecarriers.Atfurtherincreased
concentrations,however,themobilityofionsdecreasesduetomutualinter-
ferenceandanincreasingformationofneutralionpairsistheresult,which
donotcontributetotheconductivity(incompletedissociation).Thusathigh
concentrationstheconductivityoftheelectrolytenormallydecreasesagain.

16

•Anincreaseoftemperaturecausesahighermobilityofionsandthusahigher
conductivity.

•Ifacidsorbasesareusedaselectrolytesthechargemovementtakesplace
mainlyviaaspecialmigrationmechanismofprotonsandhydroxideions.This
typeofchargemovementisconsiderablyquickerthanthedirecttransportby
ions,soanespeciallywellconductivityispresent.

Theprocessesattheboundarylayerbetweenelectrodeandelectrolytearedescribed
bytheHelmholtzlayermodel.

Figure3.1:SchematicillustrationoftheHelmholtzlayer,modifiedfrom[65,66].A
indicatesadsorbedorientedwaterdipols,Bmarksspecificallyadsorbedcathions,Cindicates
negativeoverchargeswithinthedoublelayerwhich´scentersbuilttheouterHelmholtzlayer.

Causedbydifferencesinpotentialanelectrochemicaldoublelayerisbuilt.Itcon-
sistsofpositivechargesatthesideoftheelectrodeandofnegativeionsatthesideof
theelectrolyte.Theareapassingthroughthecentersoftheanions,isreferredtoas
theouterHelmholtzArea.AccordingtotheHelmholtztheorytheionsaredirectly
adjacenttotheelectrode,sothattheouterHelmholtzlayerrunsparalleltotheelec-
trodesurfaceinadistanceofhalfadiameteroftheions(solvatecoveringincluded)
[67,68].Morerecentmodels,establishedbyGouyChapmanandadvancedbyStern
[69]assumeastatisticallydistributedioncloud,whichcontainstheoppositecharge
oftheelectrode.

17

Thedissolutionratedm/dt(masspertime)forelectrochemicaletchingscanbe
calculatedasfollows(modifiedfrom[70]and[71]):
MdmAdt=z∙F∙j∙γ(3.1)
M-molarmass,F-Faradayconstant=96485As,z-chargenumber,j-current
density,γ-rateofcurrentyield,A-areasubjectedtoetching

Thechargenumberzdependsonthereactionproductgeneratedandcan,asin
thecaseofiron,bedependentonthecurrentdensity.Thecurrentyieldγinthis
caseindicatesthepercentageofcurrent,thatisusedformaterialdissolution.The
fraction1-γislostthroughdecompositionprocessesoftheetchingmedium.

3.2SelectiveEtching

Whenemployingchemicalorelectrochemicaletchingsinordertoachieveadiffer-
entiationofmolecularstructures,theprocessesbecomemuchmorecomplexthan
describedabove.Materialssuchasstainlesssteelconsistsofgrainswithdifferent
orientation,boundaryareas,precipitations,inclusionsetc.,whichallshowdifferent
electrochemicalbehavior.Dependingonthecorrosivemedium,theappliedpoten-
tial,thetemperatureandotherparameterscertainphasesmaybecorrodedtoa
muchhigherdegreethanothers.Fig3.2illustratesasimplifiedmodelofametal
consistingoftwophasesinacorrosiveenvironment(comp.[72])atagivenpotential.
PhaseAinthiscaserepresentstheanodicsite,PhaseBshowscathodicbehavior.
PhaseAmaybeasiteofhigherdisorderlikeagrainboundaryoradislocationor
anareaofdifferentcompositionlikeaprecipitation.Attheanodicsitethemetal
hasthetendencytooxidize,whereasatthecathodicsitethereisatendencytoward
areductionreaction.Thisreactioncanbee.g.thereductionofH+toH[72].It
hastobenotedthat,ifahighpositiveouterpotentialisapplied,asitisthecase
inallelectrochemicaletchingsdescribedinthisthesis,bothphasesmayexhibitan-
odicbehavior.Inthiscasebothphaseswillcorrode,howeverwithPhaseAbeing
corrodedmuchquicker.

Stansburydescribesanapproachforatheoreticaldescriptionoftheselectiveetching
ofdifferentphasesin[71].Itisassumedthatanymicroregiononthesurfacehas
adissolutionrate(currentdensity)thatisthesameasthatobservedinabulk

18

Figure3.2:Schematicillustrationofthecorrosionofatwophasesystem.PhaseB
representstheanodicphasewhichtendstodissolution,phaseArepresentsthecathodic
phasewhichsupportsthedissolutionofPhaseB.

sampleidenticalincompositiontothemicroregionwhenexposedunderthesame
conditions-e.g.environmentandpotential.Inordertoestimatetheetchingrates
ofdifferentphases,currentdensitydiagramsofthedifferentphasessuchasfig.3.3
canbeemployed,usingtheratioofthecurrentdensitiesjA/jBofthephasesatthe
respectivepotential(comp.also[73]).Whenthecurrentdensityratioisknown,the
ratioofthedissolutionratescanbedeterminedusingformula3.1

dmA∙AB=MA∙zB∙jA
dmB∙AAMB∙zAjB

3.3TheAlloyingElementsof316L

)(3.2

Themainalloyingelementsof316Larenickel,chromiumandmolybdenum.The
mainfunctionoftheseelementsareimportantfortheunderstandingoftheeffects
ofetchingsandareexplainedinthefollowingsection.

19

elNickIfnickelisalloyedtoiron,theareaofcubicface-centeredaustenite,whichoc-
cursatunalloyedsteelsonlyatatemperatureabove723◦C(s.Fe-Cdiagram),
isaugmented.Above8%nickelcontentasteelisausteniticatroomtemper-
ature.Austeniticsteelischaracterizedwithgoodformabilityandincreased
chemicalresistenceespeciallytowardsnonoxidizingacids.

mmiuroChChromiumisaferriteformerwhich,however,increasestheausteniteareain
thepresenceofnickel.Incontentsover10.5%alloyedtoasteelitbuilds,in
atmosphereaswellasinmostotheroxygencontainingenvironments,apro-
tectingsurfacelayerofchromiumoxide,thatdefendsthesteelfromcorrosion,
especiallyagainstoxidizingacids[74,75].Anincreasedchromiumcontentin
thepassivitylayerleadstoachangeinpotentialtowardsmorenoblevaluesand
thustoahigherchemicalresistance[76].Thepassivatinglayerisexplainedin
detailinthefollowingsection.

menudlybMoMolybdenumsupportsthepassivatingeffectofchromium.Itisalloyedtosteel,
ifanexceptionalresistanceagainstpittingcorrosionandcrevicecorrosionis
required[77,78,79,80].Thedifferentefficiencyofmolybdenum,chromium
andnitrogenintermsofpassivationcanbeestimatedbythefollowingformula,
whichintroducesanefficiencyfactorW:W=%Cr+3.3∙%Mo+30∙%N.The
corrosionrepressingpropertiesofmolybdenumareattributedtoanincreased
resistanceofthepassivitylayeragainstdebilitatingelements[81,82]aswell
astoanimprovedrepassivationcharacteristic[83].Itwasfound,thatthe
thicknessofthepassivitylayerincreaseswithincreasingmolybdenumcontent
0].8,[81

nSilicoSilicontendstoincreaseresistancebydispersionofchromiumandsupports
theformationofCr2O3.Howeverathightemperatures(800◦Candabove)
siliconactivatestheformationoftheδphase.Theformationoftheδphase
leadscreateschromiumdenudedzonesandisattributedtoaweakeningofthe
corrosionresistance[84].

20

3.4ElectrochemicalBehaviorofCr-Ni-Steels

Forausteniticsteelsthecorrosionbehaviorinvariousmediaisdeterminedpredom-
inantlythroughtheformationofapassivitylayer.Thepassivitylayerconsistsof
afilmofmetaloxide/hydroxidecomplexeshavingverylowconductivityformetal
ions[85].Sothemetalatomsbeneaththelayerarepreventedfromcorrosion.The
typicalbehaviorofstainlesssteelinanaqueouscorrosivemediumisillustratedin
fig.3.3ontheexampleof316Lindilutedsulfuricacid(comp.[86]).

Figure3.3:Typical/I-curveforstainlesssteelinacorrosivemedium.Afterthepas-
sivitypotentialpthecurrentdropstoaconstantlowvalue,indicatingthepassivationof
thematerial.Afterreachingthepointoftranspassivity(t)thecurrentincreasesagain,
indicatingthebreakdownofpassivation(comp.[87]).Mostetchingsdescribedinthiswork
areperformedinthisarea.

Thegraphstartsattheequilibriumpotentialoftheelectrode,thesurfacebeingnot
passivatedatthattime(leftindiagram).Attheequilibriumpotentialnooutercur-
rentispresent.Neverthelessthereareanodicandcathodicreactionstakingplace.
Thesereactionshavecurrentsofoppositedirectionswhichareinthiscasecom-
mensuratesothattheoutercurrentiszero.Withincreasingpotentialthecurrent
increases,indicatinganactivedissolutionofthematerial.Towardhigherpotentials
thesurfaceisincreasinglycoveredbyhydroxideadsorbates.Abovethepassivitypo-
tentialpdeprotonizationofthehydroxidesleadstotheformationofapassivating
filmmainlyconsistingofmetaloxides.Atthispointacontinuousprotectingpas-
sivitylayerisformed,sothatthecurrentdecreases,reachingaplateaueventually.

21

Thisareaisthepotentialindependentpassivityarea,inwhichalmostnomaterial
dissolutiontakesplace.The(verylow)currentinthisareaoriginsfromaveryslow
decompositionofthepassivitylayer,whichiscompensatedbyarebuildingofthe
layeratthesideofthemetalorfromminordissolutionprocessesatspotswherede-
fectsinthepassivitylayerarepresent,respectively.Accordinglythepassivecurrent
densitygivesinformationabouttheprotectingabilityofthepassivelayer.When
thecurrentisfurtherincreased,differentprocessescantakeplace,dependingonthe
materialandthecorrosivemedium.Inmostcasesapointoftranspassivity(t)is
reachedwherethepassivitylayerchangesincompositionandlosesitspassivation
properties(comp.[88])whichresultsinanotherconsiderableincreaseofcurrent.
Anotherpossibility,especiallywhenchlorideionsarepresent,istheformationof
pittingcorrosionspots,wherethematerialislocallydissolvedataveryhighrate.
Anotherprocessathigherpotentialswhichcanoccurparalleltotheaforementioned
isacurrentincreasecausedbyoxygenevolution.Ifthepotentialisdecreasedagain
fromhighvalues,i.e.thediagramistracedbackwards,anewincreaseofcurrent
canbenoticedafterpassingthepassivityarea.Thispointindicatesthebeginof
depassivationandisreferredtoastheflade-potentialf.

Thepassivitypropertiesarestronglydependentonthecompositionofthepassivity
layer[89].Thecompositionofthepassivitylayercanbechangedbytheexposureto
corrosivemedia.Accordinglythecorrosionresistanceandwithitthebiocompatibil-
itycanbeimprovedconsiderablybycertainpretreatmentslikee.g.electropolishing
ornitricacidtreatments.Theimprovementismainlycausedbyanenrichmentof
chromiumwithinthepassivitylayer[61].Foralloyslike316Lwhichhaveimproved
passivityproperties,theareaofactivedissolutionisshiftedtotheleftwhichiscon-
nectedwithastrongreductionoftheactivedissolutioncurrent,finallyleadingtoa
diagramwithoutanactivearea.

3.4.1EffectofHydrochloricAcidonStainlessSteel
Chlorideionsownthepropertyofattackingthepassivitylayerofstainlesssteel.An
analysisofthelayerrevealsacompositionmainlyofchromiumoxideand-hydroxide
aswellasalowpercentageofironhydroxide.Thethicknessisintherangeoffew
nm[90,91].Interestinglyinmostcasesneithernickelnormolybdenumcanbefound
inpassivitylayer[92,93].ThroughtheinfluenceofCl−onthefilmontheonehand
theintegrityofthefilmisweakened,ontheotherhandatransportofmetalions
tothesurfaceisenabledwhichperishesthepassivityeffectofthefilm[94].Asa

22

consequenceallmainalloyingelements,Fe,NiandCrinpureformatleastslowly
dissolveinHClatroomtemperaturewithoutanypotentialapplied.Aneffective
passivationoccursonlyathighcurrentdensities,incaseofCrintherangeof10
mA/mm2[88].ThepredominantreactionstakingplaceinHClcontainingsolutions
arethefollowing

Me+2H2O−→Me(OH)2+H2(3.3)
wherethehydroxidesfurtherreacttoformoxidesorchlorides,e.g.:

Me(OH)2+2HCl−→MeCl2+2H2O(3.4)
DuringHCletchingstheamountofmetaloxidedecreaseswhiletheoxy-hydroxid
contentincreases[95].Atlowerpotentialstheattackoftenmanifestsitselfaspitting
corrosion,whichisexplainedindetailinthefollowingchapter.

PittingCorrosion
Pittingcorrosionoccurs,ifanattacktakesplaceonlylocallyatcertainspots,whereas
mostofthesurfaceremainsvirtuallyuntouched.Thisformofcorrosioncanbeob-
servedonmetalsoralloyswhichformpassivitylayers.Ifthepassivitylayeris
weakenedatasinglespot,theattackcausestheformationofalocalanode.Accord-
inglythesurroundingsurfaceactsasacathode.Theresultisaquickprogressof
corrosionatthelocalanodeeventuallyleadingtoalocaldestructionofthematerial
(s.fig.3.4).Forstainlesssteelspittingcorrosionismainlycausedinconnection
withhalogenideions.Theseionsareabletoweakenthepassivitylayerlocallyand
thuscreatingalocalanode.Despiteintensiveresearchtheexactprocessesofthe
developmentofpittingcorrosionarenotyetfullyunderstood.Accordingtocurrent
models,comp.[67],adsorbedhalogenideionscausealocalincreaseoftheconduc-
tivityoftheoxidelayeraswellasanincreaseddissolutionofthelayer.Theresult
istheformationofanactivepitbeneaththelayerwhereactivemetaldissolution
takesplace.Withinthepitanincreaseofhalogenideionconcentrationaswellas
adecreasingpH-Valuecanbeobserved.Thesefactorsitselfleadtoanadditional
accelerationofthedissolutionprocess.Criticalfortheformationofpittingcorrosion
isthepresenceofdefectsandinclusions[96].Itcouldbeshownthatpitsinitiate
veryoftenatinclusionsofsulfides[97,98,99].SuterandBoehninoteacriticalsize
ofabout1µmformanganesesulfites[98].Othersourcesrelatethecreationofpits

23

totheformationofthedelta-phase[100].

Figure3.4:Pittingcorrosionof316LafterelectrochemicaletchinginHCl.Thelocal
breakdownofthepassivitylayerleadstotheformationofsphericalpits.

glishinopElectro

Aneffectthatcanoccurwhenetchingwithchlorideionsistheelectropolishingef-
fect.Inthiscasetheselectivityoftheetchingprocessiscompletelypreventedby
adjustingtheetchingparametersinawaythatthemasstransport(ofionsandwa-
termolecules)islimited.Thiswaythereactionisnotcontrolledbythereaction
rateanymore,butthemasstransportbecomesthelimitingfactorandalimiting
currentdensityoccurs.Theresultisalevelingandsmootheningofthesurface.An
explanationforthelevelingeffectisthatatuprisingsandirregularitiesthefieldin-
tensityisincreasedasafunctionof1/r(r-curvatureradius).Astheetchingrateis
dependentonthefieldintensityanacceleratedremovaltakesplaceatthissites[101].

Thepreconditionforanelectropolishingeffectisarelativelyhighcurrentdensity
(inthetranspassivearea).Additionallyahighviscosityoftheetchingsolutionis
advantageous.Thehighertheviscosity,thebetteraboundarylayerofionscan
bemaintainedattheetchingareaandthemorethemasstransportislimited.An
increaseofviscositycanbeachievedforexamplebyaddingglycerintotheetching
tion.luso

24

3.4.2EffectofNitricAcidonStainlessSteel
Nitricacidreactswithmetalsformingnitrates.However316Laswellaspureiron
andchromiumhaveatendencytopassivateinnitricacidsolutionsdependingon
theconcentrationoftheacid.Thehighertheacidconcentrationthehigheristhe
tendencytopassivate.Thompson[102]statesanalmostcompletepassivebehaviour
ofchromiumin68%nitricacidatroomtemperature.Thereactionsaresimilar
tothoseoccuringwithHCl,however,thegeneratedhydroxidsimmediatelyform
nitrates,thereforeinthefollowingtheintermediatehydroxidationstepiswaived.
Concerningchemicaletchingsandelectrochemicaletchingsatlowerpotentialsiron,
chromiumandnickeldissolveinabivalentstate[103,88,104].

Me+2HNO3−→Me(NO3)2+H2(3.5)
Athigherpotentialsandathighercurrentdensities(acc.toGmelin[105]forNi
about0.2mA/mm2)themetalsdissolveinatrivalentstate

2Me+6HNO3−→2Me(NO3)3+3H2(3.6)
Inbothcasesnogasformationattheanodecanbeobserved.Areasonforthisis
theimmediateoxidationofthehydrogentowater(comp.[106]).

H2+HNO3−→HNO2+H2O(3.7)
Nitricacidpreferablyattacksthegrainboundaries.Thereforenitricacidisusedin
metallographicetchingsolutionsforvisualizationofgrainboundaries.Theetching
rateatthegrainboundariesisinthiscasealothigherthenatthegrainsurfaces.
Tworeasonscanbefoundforthis.Firstlyamuchhigherdegreeofdisorderandan
increasednumberofvacanciesandlatticedefectsispresent,sothatthesesitesare
preferablyattacked.Secondlythecompositionatthegrainboundariesoftendiffers
fromthecompositionattheinteriorofthegrains.Thisdifferenceismainlycaused
bysegregationsofalloyingelements.Christienetal.[107]showedthatitispossible
toquantitativelydeterminetheconcentrationofcertainsegregationelementsatthe
grainboundarieswiththehelpofmetallographicetchingmethods.Itwasfound
thatthedepthoftheetchedgrainboundaryindentsaredirectlyproportionalto
thecontentofthesegregationelementphosphor.Amodelforthecalculationof
segregationsatthegrainboundariesisdescribedbySahlaouietal.in[108].Using
microchemicalanalysisofsolutionannealed316Ladepletionofthepassivatingele-
mentMobutnosignificantdepletionofCrcouldbedeterminedbyFurutanietal.
[109].Anotherfactorfavoringthecorrosioneffectisthepresenceofchromiumrich

25

carbidesofthetypeM23C6,thatcanbefoundatthegrainboundaries.Inthedirect
vicinityofthesechromiumrichprecipitationsthereisachromiumdepletionand
thusthepassivitylayerisweakened[110].Chromiumcarbideshoweverdeveloponly
athightemperatures(above700◦C)orduringverylongtemperingcycles(1000h
at600◦C)[111].Allnitratesaresolubleinwater,sothatadepositionofreaction
productsonthesurfacecanbeexcluded.

Figure3.5:316Lslightlyetchedwithnitricacid.Thegrainboundariesbecomevisible
whiletherestofthesurfaceremainsvirtuallyunaffected.

Nitricacidtreatmentsaresometimesusedinordertoincreasethepassivationeffect
forHNO3[62,112].Thisincreaseisexplainedbyanenrichmentofchromiumwithin
thepassivitylayerandbyadissolutionofinclusionslikeMnS(comp.[64]).However
someauthorsalsoreportnegativeeffectsofHNO3onthepassivation[62].Innitric
acidtheamountofmetaloxidedecreaseswhilethemetaloxy-hydroxidescontent
].[95increases

3.4.3EffectofOxalicAcidonStainlessSteel
StainlesssteelsaswellaspureFeslowlycorrodeinoxalicacidcontainingmedium,
whilepureCrandNiactpassiveinoxalicacid.Chromiumandironmainlyreact
withoxalicacidasfollows(comp.[113])

Me+2H2O−→Me(OH)2+H2(3.8)
Me(OH)2+3H2C2O4−→Me(C2O4)3+2H2O+2H2(3.9)
Whenferrousoxideispresentitispreferablyattacked[114].Thereforeoxalicacid
issometimesusedfordecontaminationanddescalingofstainlesssteels[115].In

26

bothcasesmetaloxalatesarecreated.Ferrousoxalateasthemainreactionproduct
fo2*r10−7.stainlessThereforesteel[1it16]has,islikyeellocwinhromiumcolorandandnichakselaosxalaolubilte(itKyspco=1efficien*10t−7of),aKpspo=or
solubilityinwater.Afterchemicaletchingwithoxalicaciddepositionofoxalates
on[117].theIsnutrfahecefieldwasofmobservetaedllog,rawhicphhycouldelectrolythardlyicobxealicremoacidvedetcbyhinultrgisasonicusedincleaniorderng
toalsovisuausedlizeforthemvisualizingicrostdisloructurcaeoftionsaaustndteniticwinsteegrainlsb(fig.ound3ari.6)[11eso8].fIncertainsomesteelscases[1it19].is

Figure3.6:316Letchedwithoxalicacid.Thecrystallographicstructuresarevisible.

3.4.4EffectofPhosphoricAcidonStainlessSteel
316LaswellasallmainalloyingelementsFe,NiandCrpassivateinphosphoric
acidwhennopotentialisapplied.Chromiumandironreactwithphosphoricacid
atlowerpotentialsasfollows(comp.[120,88,104])

Me+2H2O−→Me(OH)2+H2(3.10)
3Me(OH)2+2H3PO4−→Me3(PO4)2+6H2O(3.11)
Similartonitricacidathigherpotentialsthemaindissolutionprocesschangesfrom
abivalentsolutionofthemetaltoatrivalentone

2Me(OH)2+2H3PO4−→2Me(PO4)+4H2O+H2(3.12)
Phosphoricacidfeaturesahighviscosityandisusedasanelectropolishingagent
orasaningredientofelectropolishingsolutions,respectively[121].Incontrastto
27

theotheracidsdescribed,etchingwithphosphoricacidexhibits,undertheapplied
currentsusedinthiswork,exclusivelysmoothingeffects.Theprocesses,thatare
responsibleforthissmoothingbehavioraredescribedindetailinsubsection3.3.1.
Intheliteratureanincreaseofchromiumcontentwithinthepassivitylayerincon-
nectionwithanincreasedchemicalresistanceisreportedforsurfaceselectropolished
withphosphoricacid[63].Allphosphatesoccurringinconnectionwiththephos-
phoricacidetchingof316Larevirtuallyinsolubleinwater(Kspforironphosphate
is1,3*10−21).

Figure3.7:316Letchedwithphosphoricacid.Auniformsmoothsurfacedeveloped.

28

4MaterialandMethods

lMateria4.1

Allexperimentswerecarriedoutonmedicalgrade316Lsteel.316Lisanaustenitic
chromium-nickelsteelwithahighchemicalresistance.Theresistanceisachievedby
ahighchromiumcontent,anadditionofmolybdenumandahighnitrogencontent.
316Lisusedforavarietyofimplantmaterialslikeboneimplants(plates,screws,
nails),vesselimplants,etc.[122]).Therawmaterial(Minitubes,Grenoble,France)
forthestentfabricationaretubesof1.8mmouterdiameterand0.14mmwall
thickness.Thestentsarefabricatedbylasercuttingandreceiveaheattreatment
aftercutting.Boththestentsandtherawmaterial(forevaluationexperiments
andforroughnessdeterminations)wereusedfortheexperiments.Thematerial
compositionisasfollows:

CSiMnSPNiCrMoCuNFe
0,0170,361,80,0020,01814,7217,332,750,080,076bal

Table4.1:Compositionof316Linweight-%accordingtosupplier.

Concerningthemechanicalpropertiesthefollowingdataisgiven:

Sampleno.Rm(N/mm2)Rp0,2(N/mm2)A(%on50mm)
19438296,5
29458116,0
Table4.2:Mechanicalpropertiesof316L.

ItfulfillsthenormsISO5832.1andASTMF138.92.Roughnessofthetubeswasde-
terminedtoRz=0,76,Ra=0,075(s.p.34).Theaverageintersectionalgrainsize
ofthemeasuringmaterialetchedwascrosdsseterminectioednalaccmicrordiongtographsDIN(fig.506041.1).(Flineaorlaindtetaerceptiledmethodescriptid)boyn

29

s.appendix.Thegrainsizeofthetubeswas43.2µm±8.2µminlongitudinaldi-
rectionand21.1µm±5.4µmintransversaldirection.Thestentshaveanaverage
grainsizeof15.9µm±3.2µm,independentofthemeasureddirection.

Unlessotherwisenotedelectropolishedstentswereusedforalletchings.Allofthe
electropolishedstentswereobtainedcrimpedonacatheter.Inordertoremovethe
stentstheyweredilatatedat6atmswithinapolymertubeof1.7mminnerdiame-
ter.Thisprocedureleadstoaslightdistortionofthestent.Thisfactisimportantto
notewhenetchingonapin(s.followingsec.)asitleadstoamuchworseprotection
oftheinnersideofthepin.

Gritblastedstentsservedasreferencefortheadhesiontests.Fig.4.2showsstruts
ofanelectropolishedstentandagritblastedstent.

Figure4.1:Grainstructureofstentmaterial(section),left:stent,right:rawmaterial.
Atextureinlongitudinaldirectionisvisiblefortherawmaterial.

4.2TheEtchingAssembly

Thefollowingacidswereused:nitricacid65%,hydrochloricacid35%,oxalicacid
(crystallinepowder),phosphourusacid85%(p.A.,Merck,Germany).Theacids
weredilutedtotherequiredconcentrationswithdemineralizedwater.Foralletch-
ingsthefollowingconcentrationswereused:nitricacid40%,hydrochloricacid24%,
oxalicacid10%.Incaseofmostetchingsthestentswereimmersedintothesolution
crimpedonastainlesssteelpinof1.4mmdiameter,thecrimpingbeingperformed

30

Figure4.2:SEMimagesofagritblasted(left)andanelectropolishedstent(right).

manually.Incaseofmosthydrochloricacidthestentswereetchedwithoutapin
holdbyathinplatinumwire.Whenusing316Ltubesthematerialwasholddirectly
intotheetchingsolutionbymeansofaclamp.

ThecurrentwassuppliedbyaDCtransformer(PPS3003,Conrad,Hirschau,Ger-
many)withthestentsswitchedastheanode.Thecurrentwasmeasuredwithan
amperemeter.Prioretchingthespecimenswerecleanedinacetonewithultrasonic.
Aftertheetchingprocedurethespecimenswereholdin0.5molNaOHfor2min.
Thisstepassuresthatanyacidremainsareeliminatedand,asasideeffect,the
NaOHpromotesapassivationofthesurface.Afterthisprocedureallspecimens
werecarefullywashedwithdemineralizedwaterandcleanedinanultrasonicbath
.min20rfo

Alletchingswiththeexceptionofthehydrochloricacidetchingswerecarriedout
inastainlesssteelblockwitha32mmholeof16mmdiameterdrilledinit(fig.
4.3).Thesteelblockwaselectricallyconnectedsothatitservedasacontainerfor
theetchingsolutionandatthesametimeasthecathodefortheetchingprocess.
Thestentonthepinwasimmersed20mmintotheetchingsolution.Inmostcases
hydrochloricacidetchingswerecarriedoutwithoutapin.Inthiscasethestentwas
immersedintotheetchingsolutiononathinplatinumwire.

Forthepreliminaryexperimentsinordertoevaluatetheetchingparametersof
thecombinationetchings30mmstentrawmaterialtubes(1.8mmdiameter)were
used.Thefirstetchingstepwasperformedasdescribedabove.Thesecondstep
wasperformedwithaspecialassemblywithinasmallglasscylinder.Thecathode

31

Figure4.3:Schematicdrawingofthesetupoftheetchinginasteelblock.Thestentis
switchedastheanodewhilethesteelblock,whichisfilledwiththeetchingsolution,serves
asthecathode.

consistsofacylindrical316Ltube(stentrawmaterial)with6mmdiameterand
4mmhight.Thetubetobeetchedwasfixedwithoneendatthemiddleofthe
cylindricalcathodebyathinpolymerdiscwithaholeinitscenter(fig.4.4).By
thisadecreasingcurrentdensityfromthetoptotheendofthetubewasachieved.
Whencreatingcrosssectionalmicrographsofthistubesdifferentareasofthetube
canbeinvestigatedrepresentingdifferentetchingstrengths.Aschematicillustration
isgivenin4.5.Sincetheelectricalresistanceofthesolutionaswellasofthetubeis
nearlyconstantoverthewholelength,analmostlinearincreasefromcurrentdensity
jminatthebottomtojmaxatthetopisgenerated,withanaveragecurrentdensity
ja=0.5∙(jmin+jmax).Whentheratioc=jmin/jmaxisknowntheminimum
currentdensitycanbeestimatedby
jajmin=0,5+21c(4.1)
jacanbecalculatedbydividingtheappliedcurrentbythesurfaceareaofthetubes
whichis30mm∙1.8mm∙π.ccanbeestimatedwhencomparingthematerial
removalatthetopwiththematerialremovalatthebottom.Theremovalswere
determinedto8µmand22µm,respectively,sothatccanbeassumedtobeap-
prox.0.37.Acurrentof200mAresultsinthevaluesjmin=0.64mA/mm2and
jmax=2∙ja−jmin=1.57mA/mm2.

32

Figure4.4:Etchingassemblyforthepreliminaryexperiments.Thesetupresultsinahigh
currentdensityatthetopofthetubewhichdecreasesgraduallytothebottom.

ForalletchingswithhydrochloricacidexcepttheHCl-combinationetchingswithni-
tricacidadifferentassemblywithseparatedelectrodeswasemployedbecausethese
experimentswereperformedusingastirrer.Fortheseetchingsanassemblywith
twoseparatedbeakerswasused,illustratedschematicallyinfig.4.6.Thesepara-
tionoftheelectrodesholdstheadvantagethatthegasformationattheelectrodes
isreducedandthereforethedissociationofthesolutionisinhibited.Asastirrer
apolymercoatedmagnetwasusedwhichwasadjustedat270u/minforallexper-
iments.Eachelectrodewassetinaseparatebeakerofthesameetchingsolution
whichwereconnectedwithacellulosemembranesoakedwithetchingsolution.As
thecathodeservedasheetof316L.

4.2.1CalculationofCurrentDensities

Forsomecontemplationstheetchingcurrentsareconvertedintocurrentdensities
inordertoprovidevaluesthatareindependentoftheexperimentaldesign.When
etchingwithoutapinthetotalstentsurfacearea(4∙25.9mm2=103.6mm2for

33

Figure4.5:Schematicillustrationofthepreliminaryexperiments.Thefirststepcreates
smallcavitieswhicharehollowedoutinasecond,isotropicetchingstep.

a16mmstent)isusedforthecalculations.Thesurfaceareaofthepinplusthe
sidesurfaceareasofthe(electropolished)stent(2∙25.9mm2fora16mmstent)
areusedwhenetchingonapin.Sotheareausedforthecalculationsofthecurrent
densitiesis1.6mm∙pi∙20mm+51.9mm2=152.5mm2.However,theexact
areaatwhichetchingtakesplacecannotbedeterminedinthiscasesincepartof
thestentmaynotbeincontactwiththepinandthuscontributestotheetched
surfacearea.Alsowhenetchingwithoutapintheactualareavariesontheone
handwiththematerialremovalandontheotherhandwithchangesinroughness.
Itisimportanttonotethatthecurrentdensitiesandtheetchingtimescanbeused
inordertoqualitativelycomparethestrengthofdifferentetchings.Howeveritis
notanexactlyprecisemeasureoftheactualetchingstrength.Anotherparameter
thatcanbeusedasanestimationfortheetchingstrengthisthematerialremoval
measuredasdiameterloss.

4.3TopographicSurfaceAnalysis

ThesurfacesofallsampleswereanalyzedunderSEMatmagnificationsfrom300X
to2000X.ForthisanalysisanS-3500N,HitachiScienceSystems,Tokyo,Japanwas

34

Figure4.6:Schematicassemblyoftheetchingwithseparatedbeakers.Bothbeakersare
filledwithetchingsolutionandareseparatedwithamembrane.Thissetupenablestheuse
ofastirrerinawaythataninfluenceofthecathodicreactionontheetchingprocessis
d.mizeinmi

employedwith5to20kVaccelerationvoltage.

Forboththedeterminationofthegrainsizeincaseofthegrainboundaryetched
samplesandthecharacterizationofthefacetsincaseofthehydrochloricetched
samplesthelinecutmethodwasused.Tenlinesof100µmlengtheachweredrawn
randomlyonlightmicroscopeimageswith200Xmagnification(Axiovert25,Carl
ZeissGmbH,Jena,Germany)andtheintersectionswiththegrainboundariesorthe
facets,respectively,werecount.Itwasfoundthatforalletchedstentstheresults
ofthecutnumbersareindependentofthelinedirection,soitcanbeassumedthat
thestructuresareisotropic.

Roughnesswasdeterminedbymechanicalperthometry(M2,MahrGmbH,Göttin-
gen,Germany)ontubematerial.Themeasuredparametersweretheaveragerough-
nessdepth(Rz)andthearithmeticaverageroughness(Ra).Themostimportant
roughnessvalueforthegrainboundaryetchingsistheRzvalue.Thisvalueismost
sensitivefordiscreteindentationsasitgivesanaverageofthehighestpeaksandthe

35

lowestvalleys.ForthehydrochloricacidetchingstheRavaluesaregiven,character-
izingtheaverageroughness.Alimitationofroughnessdetermination(mechanical
roughnessmeasurementaswellasconfocalperthometry)isthecorrectmeasurement
ofslotswithveryhighaspectratios(likedeepnarrowcrevices).Inordertovalidate
theresultsthevaluesofthemeasurementswerecomparedtopolishedsectionsof
thesamesurfaces.Thecomparisonshowedgoodcongruencebetweenthesections
andthemeasuredprofilesforthehydrochloricacidetchedtubesamples.Forall
grainboundaryetchedsurfacestheresultingdeepcrevicescouldnotbecapturedby
conventionalmethodsbecauseofthehighaspectratios.Forthesesamplesaswell
asforallexaminationonstents,includingthedeterminationoftheexactdepotvol-
umesofthecombinationetchedsamples,ananalysisofcrosssectionalmicrographs
erformed.pasw

4.4CrossSectionalMicrographs

Fordetailedtopographicanalysiscrosssectionalmicrographsofetchedstentswere
prepared,embeddedintwo-componentepoxyresin(Specifix-40,fa.Struehrs,Wil-
lich,Germany).Forallstentsthesectionswerecreatedwiththestentonapinin
ordertofixthestentwithintheresin.Thesectionswerepolishedwithdiamond
suspensiondownto1µmparticlesizeandfinallysputteredwithgold(approx.10
nm)usingasputter-coater(SCD005,BAL-TECAG,Balzers,Lichtenstein)under
highvacuumwithacurrentrangebetween5-15kV.Besidesanexactanalysisofthe
depotsitispossibletoestimatethematerialremovaloftheetchingswithpolished
sections.Theremovalratecannotbedeterminedbythewidthofthestentsbecause
forpolishedsectionsthisparametervarieswiththeorientationofthestrutswithin
thestentnetting.Besidesthewidthcanvaryitselfbecauseofinaccuraciesofthe
lasercutting.Thethickness,however,doesnotvary,providingthatthestentis
mountedexactlyvertical.

FortheanalysisofRapamycincoatedstents,thecoatedstentsweresputteredwith
goldfromallsidespriormounting(processseeabove).Thisgoldlayerpreventsthe
solutionoftheRapamycininthemountingresin.Besidesthegoldsputteringenables
thecontoursoftheRapamycinlayertobeseenwellunderthelightmicroscopeas
wellasundertheSEM,eveniftheRapamycinhasbeenwashedoutbythepolishing
s.cespro

36

4.5SoftwareAnalysisoftheDepotStructures

TheSEMimagesofthesectionsweretransformedintoimage-filesofacalculation
program(MATLAB)foradetailedanalysisofthedepotproperties.Forthedeter-
minationofthedepotlimitsapolygonwasgeneratedaroundthestentasshownin
fig.4.8.Thepolygonisdefinedthroughtheattributesofhavingalledgepointson
thecontourlineandhavingnoconcaveparts.Thethree-dimensionalanalogtothe
polygonmodelwouldbeafoilinwhichastentstrutisenfolded.(Itisassumedthat
theheightofthemicrostructuresdoesnotvaryconsiderablyindirectionperpendic-
ulartothecrosssection.Thisassumptionhastobemadeinordertoexcludeminor
deviationsbetweenthe2Dmodelandthe3Danalog.)Determinedthroughthis
attributes,thepolygoncoversthestrutsimilarlikethearterytissue.Accordingly
theareasbetweenpolygonandstrutcanbedefinedasdepotsandcanbesoftware
analyzed.Thismodelisusedtodescribeastateatwhichtheouterdruglayerofthe
stentisalreadyusedup.Fig4.7showsacrosssectionalmicrographofacoatedgrain
boundaryetchedstentstrutaftercoatingwith1%Rapamycin.Itcanbeseenthat
thedrugbuildsalayeroffewµmthicknessoverthedepots.Mostofsuchalayeris
normallyremovedwithinafewdaysasalreadydescribedintheintroduction.After
thisperiodthedepotswillpredominatethereleasebehavior.Atthispointthestrut
isalreadyintegratedintothetissue,orcoveredbyathrombus,respectively,sothat
thetissueisadjacenttothestrutsurfaceasassumedinthemodel.Anexamplefor
astentedporcinearteryatthethirddayafterimplantationisillustratedinfig.4.7.
Moredetailsonthereleaseoutofthedepotsaregivenonp.104.

4.6CalculationofDepotProperties

Firstlythecontourlineofthestentstrutwasdeterminedthroughcontrastdifferences
betweenstentandbackground.Thiswasperformedusinganobjectdetermination
function(selectobjectmex)oftheMATLABimageprocessingtoolbox.Itwasfound
thatthecontrastwassufficientforallimagessothatapreprocessingwasnotnec-
essary.Afterwardthepolygonwascreatedaroundthestrut(s.fig.4.8).

Adetaileddescriptionofthecalculationofthepolygonpointscanbefoundinthe
appendix.Fortheanalysisofthedepotsonlythreeofthefoursidesofthestent

37

Figure4.7:Left:coatedgrainboundaryetchedstentstrut,section;right:crosssectional
viewofan(uncoated)stentstrutwithinanarterythreedaysafterimplantation[123].A
thrombushasalreadybeenbuiltaroundthestent.

Figure4.8:Sectionofastentstrutwithpolygonline.Thepolygonlinesimulatesthe
artery,circumscribingthemicrodepots.

areconsidered,waivingtheinnersideofthestent.Thissideisadjacenttothepin
duringtheetchprocessandthusprotectedfromtheetchingattack.Ofeachsingle
depotthearea,thecenteroftheareaanditssmallestdistancefromthepolygonline
isdetermined.Inordertocomparedifferentstentsthetotaldepotareawasdivided

38

bythelengthofthe(threeside)polygonline.Sincetheorientationofthestruts
withinthemountingresinvaries,thecrosssectionalareaandwithitthelengthsof
thepolygonlinesofdifferentstrutsdiverge.Dividingthedepotareasbythepolygon
lengthyieldsanaveragestructuredepthwhichisindependentofthesizeandshape
ofthecrosssection.Thisvaluecorrespondstothetheoreticalstoragevolumeper
areaofamicrostructuredsurface.E.g.iftheaveragestructuredepthis1µm,
thetheoreticalstoragevolumewouldbe0.001mm3/mm2.Anevenmoreimportant
parameterthanthetheoreticalstoragevolumeistheeffectiveaveragedepotdepth.
Thisvalueisobtainedbymultiplyingtheareaofeachsingledepotwiththedistance
ofitscenterpointfromthepolygonlineanddividingthesumofthisvaluesbythe
totaldepotarea.Theproductofthisvalueandtheaveragestoragevolumecanbe
regardedasameasureforthedepoteffect.Foreachetching6to8strutimageswere
.dzelyana

4.7CoatingProcessandAnalysisofCoatings

Theprocedureofcoatingandcoatinganalysisisonlysummarizedinthefollowing
chapter,adetaileddescriptioncanbefoundin[124].Thestentswerecoatedviaa
TransluminaMagicBoxccoatingsystemwith1or2%ethanolicRapamycinsolu-
tion.Subsequentdryingofthestentwasnotnecessarysincethesystemallowsthe
ethanoltoevaporateduringthecoatingprocess.Forthequantificationoftheamount
ofdrugthestentwasdilatatedwith16atmsandsubsequentlytheRapamycinwas
washedoffwithethanolandthetotalamountofRapamycinwasdeterminedby
UV-Visspectroscopy(Specord-210,JenaAnalytikAG,Jena,Germany).Forthe
adhesionteststhestentwasinsertedintoanartificialbloodcircuitanddilatated
inanartificialartery.Afterdilatationthearterywiththestentwascutoutand
therapamycinwasextractedwithethanol.Forthedeterminationofreleasekinetics
thecoatedstentswereplacedinsideacoveredplastictubewithRinger´ssolution.
Inordertoachieveafluxofthesolutionthetubesrevolvedperpendiculartothe
tubeaxiswith20turnsperminute.Atconstanttimeintervalsasmallamountof
solutionwaswithdrawnandanalyzedintermsofRapamycincontentviaUV-Vis
spectroscopy.Duringthereleasethetemperaturewasholdataconstant37◦C.

39

Figure4.9:ExampleforanEDSanalysisspot.

4.8ChemicalSurfaceAnalysis

-EDSSEM.14.8

EDS(energydispersiveX-Rayspectroscopy)analysisisusedinordertogetinforma-
tionsaboutthecompositionofasamplethroughinterpretationofX-Rayradiation.
Theexcitationisperformedbyacceleratedelectrons.InthecaseofaSEM-EDSthe
EDSiscombinedwithanelectronmicroscopewhichsuppliestheelectronbeam.The
informationdepthisdependentontheaccelerationvoltageoftheelectronbeam.For
allanalysisinthisworktheaccelerationvoltagewas20kVwhichyieldsaninfor-
mationdepthofapprox.3to4µm.Thelateralresolutionis1to2µm.Inorderto
visualizethegrainboundariestheanalyzedsampleswereslightlyoxalicacidetched.
Thiscombinationetchinghastheadvantagethatthegrainboundariesaremade
visiblewithoutcreatingcreviceswhichcannotbeanalyzedbyEDS.Fig.4.9shows
anexampleofanimageofameasuredpointatthegrainboundaries.

4.8.2AugerElectronSpectroscopy

AESisusedinordertoanalyzethecompositionofsurfaces.Inthisworkitwasused
inordertoobtaininformationsaboutthecompositionofthepassivitylayers.The
informationdepthislimitedtothetop2to20atomlayersofthesamplesurface.
Sincethepassivitylayerofstainlesssteelisintherangeoffewnmaninfluenceof
thebulkmaterialontheresultscannotalwaysbeexcluded.BesidesAESisquite

40

componentconc.in−3Ringer´sso-normalconc.−in3human
lution(10mol/l)bloodplasma(10mol/l)
Na+143137-148
K+5,44,1-5,6
Ca2+1,82,2-2,7
Mg2+0,80,7-0,9
Cl−125116
SO42−0,80
H2PO4−1,00,8-1,6
−2926HCO3Haematin5,53,3-5,6
030031totalTable4.3:ComparisonofRinger´ssolutionandbloodplasma.
sensitiveagainstsurfacecontaminationssothatevenminorremainscanaffectthe
results.ForsecondaryionmassspectroscopyorX-rayphotoelectronspectroscopy
whicharethebestsuitedmethodsforanalysisofpassivitylayerslarge,plainsurface
areasarenecessary.Thereforethesemethodscouldnotbeusedforstents.
4.9ElectrochemicalAnalysis
AllanalysiswasperformedaccordingtoDIN10993-15.ForallexperimentsRinger´s
solutionwasusedaselectrolyte(compositions.appendix).Ringer´ssolutionisused
asanartificialsubstituteforbloodplasmaorfortissuefluid,whichhasbasically
thesamecompositionexceptasmalleramountofproteins[125].Acomparisonof
thecompositionofhumanbloodplasmacomponentsandRinger´ssolutionisgiven
.3.4tableinDespitethegoodreplicationoftheplasmacompositioninrespecttothemaincon-
stituents,Ringer´ssolutioncanonlyroughlysimulatethecomplexcorrosiveproper-
tiesofbodyfluids.Besidesthelistedconstituentsapluralityofenzymes,proteins,
livingcellsandothercomplexelementsarepresentwithinthelivingbody,which
cannotbesimulatedbyanartificialsolution.Especiallyenzymeshaveagreatim-
pactonthechemicalactivity,sothatitcanbeassumedthatthecorrosivenessin
vivoagainstmetalsisconsiderablyhigherthaninRinger´ssolution.
41

4.9.1CyclicVoltammetry

Cyclicvoltammetryisusedinordertocharacterizetheelectrochemicalbehavior
ofamaterialsurfaceinacertainenvironment.Thesamplewhichisimmersedin
electrolytesolutionisinitiallyholdattheequilibriumpotential.Thentheelectrode
potentialisdecreasedtoamorenegativepotential(incathodicdirection).This
scanisperformedinordertoguarantythesameinitialsituationforallsamples.
Fromthispointthepotentialisscannedinanodicdirectionuptoamaximumpo-
tential.Finallythescanisreverseduntilthepotentialreachestheinitialstarting
point.Theminimumandmaximumpotentialsareselectedinawaythathydrogen
oroxygenevolution,respectively,canbeobservedatthesample.Itwasnotpossible
todeterminetheexactsurfaceareaoftheimmersedpartofthestent.Thereforea
standardizationofthecurrentstocurrentdensitieswaswaived.

Anexampleofameasurementisgiveninfig4.10.Priortoandafterthemeasure-
mentsthespecimenswereanalyzedunderSEMandunderthelightmicroscope.

42

Figure4.10:Exampleforacyclovoltammogramm.

Theassemblyforthecyclicvoltammetryisillustratedinfig4.11(comp.[126]).For
themeasurementsathreeelectrodessetupasillustratedinfig4.12wasused.The
potentialismeasuredviaathirdadditionalreferenceelectrode.Thereasonforthis
setupisthefactthatareferenceelectrodepervadedbyelectricalcurrentexhibits
anovervoltage,makingacorrectpotentialmeasurementimpossible.

Figure4.11:Assemblyforcyclicvoltammetry(B:referenceelectrode,M:measuringelec-
trode,G:counterelectrode).

Thevoltageforthepotentiostat(Jaissle,1030DA)isdeliveredbyafunctiongener-
ator(Prodis1/16l),thefunctiongeneratorbeingcontrolledbyaPC.Thereleased
analogcurrentmeasurementsignalissenttothePCviaavoltmeter(Keithley199
SystemDMM/Scanner)whichservesasanalog-/digitaltransformer.Thesoftware
thatwasusedwaspurpose-writtenfortheexperimentsattheInstitutforCorrosion

43

andSurfaceTechnology,ErlangenwiththeprogrammingsoftwareQuickBasic.

Themeasuringunitforthecyclicvoltammetryisillustratedinfig4.12.Thereference
electrodeisaAg/AgClelectrodewithsaturatedKClsolutionandiscombinedwitha
Haber-Luggin-Capillary.ThepotentialofsuchanelectrodeisnormallyEAg/AgCl/KCl
=+198mVSHE(SHE-standardhydrogenelectrode),howeverfortheemployedelec-
trodeforunknownreasonsastablepotentialof+220mVSHEwasmeasured.For
thecounterelectrodeandthewiretoholdthestentplatinumwasused.Inorderto
avoidaninfluenceoftheplatinumonthecyclicvoltammogramthestentwasnot
completelyimmersedintothesolution,sothattheplatinumhadnocontactwith
theelectrolyte.Thecompleteassemblywassurroundedbyametalcageinorderto
shieldelectromagneticdisturbanceradiation.

Figure4.12:Simplifiedillustrationofthecircuitforcyclicvoltammetrywiththreeelec-
trodes(B:referenceelectrode,M:measuringelectrode,G:counterelectrode).

Prioreachmeasurementtheelectrolytewasdegasedwithnitrogen(99,999%)forap-
prox.onehour.Additionallythegaschamberwasscavengedwithnitrogenduring
themeasurement.OpenCircuitpotentialmeasurementswereperformeddirectlyaf-
terimmersingthestentintotheelectrolytesolution.Fortheseequilibriumpotential
measurementsatimeofapprox.45minhadtobeawaiteduntilastabilizationof
thevalueappeared.Themeasurementswereperformedininertgasatmospherein
ordertoreceiveacurrent-potentialcurvethatisnotsuperposedbyreductioneffects
ofoxygen.Allmeasurementswerecarriedoutatascanrateof1mV/sstartingfrom

44

theequilibriumpotentialincathodicdirectiontill-1050mVAg/AgCl,subsequentlyin
anodicdirectiontill1200mVAg/AgClandfinallyagainincathodicdirectiontillthe
equilibriumpotential.

4.9.2PotentiostaticTestsforDeterminationofIonRelease

Duetothehighamountofnickelandchromiumin316Lariskofnegativesideeffects
causedbyreleasedionsispresent.Chromiumandnickelandalsomolybdenumare
toxicsubstancesthatcancausesevereallergicreactions.Inthecaseofcoronary
stentsallergicreactionhavebeenattributedtoaconsiderablyhigherriskofresteno-
sis.[127,128].Potentiostaticexperimentswereperformedinordertocomparethe
releasebehavioroftheetchedstentswithgritblastedstentsasreference.Ithasto
benotedthatthechosenconditionscannotexactlysimulatetheinvivoconditions.
Neverthelesstheexperimentsgivesomeinformationabouttheexaminedsurfacesin
comparisontoeachother.

Thetestassemblyisanalogtothecyclicvoltammetryassembly,whiletheionrelease
experimentswerecarriedoutinaglasschamberwithalid(fig.4.13)(comp.[126]).
Thefillingvolumeoftheglasschamberwas40ml.Thetestswereperformedata
constantpotentialof150mVabovethebreakthroughpotentialforatimeperiodof
72hours.Thecurrentwasmeasuredcontinuouslyoverthewholeperiod.Allspeci-
mensweresubsequentlyanalyzedunderSEMandlightmicroscope.Theelectrolyte
solutionswiththereleasedionswereanalyzedwiththegraphitetubetechniqueon
anAnalyst800(PerkinElmer,Wellesley,Massachusetts,USA,detectionlimit2
l).g/µ

45

egurFi

easeler

46

.43:1

(:B

maticSche

ationrlustli

referenceelectrode,

M:

of

the

ringasuem

ylassemb

for

electrode,

:G

determination

ounterc

of

ongl

de).otrcele

mter

ion

eedingsSCell4.10

4.10.1SeedingProcess
Endothelialcells(cellline),stromalcells(primarycells)andfibroblasts(cellline)
used:erew

•HUVEC:humanumbilicalveinendothelialcells(Cat#C-015-10C,Cascade
)logicsBio•UCSC:umbilicalcordstromalcells(inhouseproduced,theprocedureisde-
scribedin[129])
•3T3:embryonicfibroblasts(Cat#CRL-1658,ATCC,Manassas,USA)

Allcellseedingswereperformedon316Lplateletmaterial.Roundplateletsof1
mmthicknessand10mmdiameter(correspondingtoasurfaceareaof188,57mm2)
wTheeremacplateletshinedwerfromegrstitentrablastedwormateretcialhebad,rsrespandectivelysubsequen,atlyccordinggroundtotheandpstents.olished.It
wasfoundthathighercurrents(600mAforaplateletroughlycorrespondsto200
mAforastentonapin)andlongeretchingtimeswereneededfortheplatelets
inSEMorderbetoforegenerateseedingthineordersametomicrobserostrvetheuctures.surfaceThetopsurofacesgraphies.wereFexoartheminedseediunderng
apropolcessyuretthehaneplateletstube,winereawayinsertedthatinthteotheplateletsbottomsealstofheapprotubex.co2cmnsistenhightly.sectInioansnextof
stepthesectionswerefilledeachwith5000cellsincellmedium.Thecellnumber
wSyasstemdetermGmbH,inedRusineutgtlingheen,CasyGerman1-Celly).CoFunortertheAnaanalysislyzeraSysmstem,allpartModelofthTTes(Scolutihäorfen
tobetestedwasdilutedonetoonehundredusingaweakelectrolyte(Casyrton,
sterilefiltered,SchärfeSystemGmbH,Reutlingen,Germany)anddrawnthrougha
capillarywithaconstantflowvelocity.

Thecellcultivationtookplaceundersterileconditionsincellcultureincubators
(HeraeusKendroLaboratoryProductsGmbH,Langenselbold,Germany)at37◦C
and5%CO2-atmosphere.Forsubcultivation,culturemediumwasremovedfrom
thePetridishandwashedtwicewithphosphatebufferedsaline(PBS-Dulbecco,
BiochromAG,Berlin,Germany).Thecellmonolayerwasremovedbytrypsiniza-
tion(TrypsinEDTASolution,BiochromAG,Berlin,Germany).Alistofallreagents

47

andculturemediausedcanbefoundintheappendix.Astimepointsfortheanal-
ysisday1,day4andday8afterseedingwerechosen.Ateachtimepointthecells
werefixedandthreeWST-measurementswereperformed.Subsequentlythecells
werepreparedforSEM-Imaging.TheWSTmeasurementswerecarriedout30and
90minutesafteradditionoftheWST-measuringsolution.Adetaileddescription
oftheprocedureisgivenintheappendix.

4.10.2PreparationforSEM-Imaging
Thecellswerewashedtwicewithphosphatebufferedsolutionandfixedfor2days
inglutaraldehyde(VWR,Darmstadt,Germany)at4◦C.Thesampleswerethen
dehydratedusinganincreasingalcohollinefrom10%to100%.Subsequentlythe
sampleswerecriticalpointdriedbysubstitutingthealcoholstepwisebyCO2(CPD-
030,Bal-TecAG,Balzer,Liechtenstein).Afterwardthesamplesweresputteredwith
anapprox.10nmgoldlayerusingasputter-coater(SCD005,BAL-TECAG,Balz-
ers,Liechtenstein)underhighvacuum.SEManalysiswasperformedwith10kV
accelerationvoltageatmagnificationsbetween30Xand500X.

4.11MechanicalTests

Duringtheimplantationacoronarystentissubjectedtohighstressesbythedilata-
tionprocess.Theresultsareplasticdeformationsatpredefinedspotssothatthe
stentremainsinthedilatatedshape.Aftertheimplantationprocessastaticload
actsuponthestent,imposedbythedilatatedarteria.Themagnitudeofthisload
dependsontheconsistenceofthestenosis.Superimposedonthestaticloadisa
cyclicloadwhichiscausedbytheoscillationofthebloodpressure.Thispressure
oscillationhasafrequencyofapprox.1Hz,accordingtotheheartbeat,anda
magnitudeofapprox.40mmHg(s.Fig.4.14).Thepressurechangeactsasacyclic
loadrelievingduringtheheartbeat.Itisassumedthattheworsecaseinterms
offatiguestressoccursatverysoftstenosises.Inthiscasethestaticloadwillbe
smaller,howeverthepressureoscillationisabsorbedleastbytheartery,increasing
thecylclicloadsonthestent.

48

Figure4.14:Bloodpressurediagramm,modifiedfrom[130].Thedifferencebetweendi-
astolicandsystolicpressureis40mmHg.

Themainintentionofthefatiguetestswastoevaluatetheeffectofdifferentstruc-
turesonthefatiguebehaviorandtogiveacomparisonofthedifferentsurfacemod-
ificationsamongsteachotheratconditionsapproximatinginvivoconditionsrather
thantomakeapredictionifthestentswouldendureausewithinthebody.The
experimentswereperformedonetchedconventionalstentsthatwerenotespecially
designedforanetchingprocedure.Foracommercialuseaspecialstentdesign
wouldbenecessarywithlargerstrutthicknessinordertocompensatethematerial
lossthroughtheetchingprocess.

Forthefatiguetestsapressurechamberwasdevelopedwithwhich8stentscanbe
testedsimultaneously.Withtheassemblydescribedinthischapteritispossibleto
simulatepressureoscillationsatafrequencyofupto70Hz.Thepressurechamber
consistsofastainlesssteelhousingfilledwithRinger´ssolution,whichisclosed
withamembraneonthelowerside.Onthelowersidethemembraneisconnected
withapiezobattery(AgPdstack,5.2*5.2mm,AN01/8505b,EPCOSAG,Munich,
Germany)whichheightcanbeadjustedbyadifferentialthread.Thetubeswith
thestentsaremountedontopofthechamberinawaythatthepressurewithin
thechamberistransmittedtotheinteriorsofthetubes.Itisimportanttoassure
thatnogasbubblesgetintotheassemblyduringthefillingprocess.Bymoving
themembranewiththepiezoapressurechangecanbeimposedonthechamber.
Forthefatigueexperimentsthepiezoisactuatedbyanamplifierforlowervoltage
piezoactuators(AVU200/5P,DASSmbH,Saarbrücken,Germany)withthesignal
beingsuppliedbyasinusgenerator(FG200,H-TronicGmbH,Hirschau,Germany).
Thepressurewithinthechamberiscontrolledbyapressuretransmitter(PR-23S,

49

KellermbH,Jestetten,Germany).Thatway,withthehelpofanoscillator,itis
possibletodirectlyvisualizethepressureoscillation.Thismethodwasusedforthe
evaluationoftheactuationparametersforthepiezoinordertogettherequired
pressureamplitude.Thepiezowasactuatedinawaythatanamplitudeof30mbar
wasgeneratedatabasispressureof40mbar.Forthebasispressureaslightlyhigher
valuethantheamplitudewaschoseninordertopreventnegativepressure.Forthe
testingprocedurethestentsweredilatedwithinasiliconetubesectionsof3cm
length,3mminnerdiameterand4mmouterdiameteratapressureof15bar.It
hastobementionedthatminorpressurelossthroughthetubescannotbeavoided,
sothatthebasispressurehadtobeadjustedonadailybasisduringthetestusing
thedifferentialthread.Afterwardthepressurechamberandthesectionswerefilled
withRinger´ssolutionandfixedontheprovidedspouts.Afterthetestingtimeof
52daysthesectionswereremovedfromthestentsandthestentswereinvestigated
underSEM.Atechnicaldrawingoftheassemblycanbefoundintheappendix.

50

Fiegur.45:1awingDr

of

whicharemountedontop

stent

eguiatf

r.etest

heT

tentss

ear

dtemplani

ofthechamber.Tubesandpressurechamber

andfilledwithRinger´ssolution.

into

olyperm

estub

areinterconnected

51

Microstructurin5thwigHydrochloricAcid

tsulRes5.1

Electrochemicaletchingsonbasisofhydrochloricacidcanresultintheformationof
varioussurfacetopographies.Dependingonparameterslikeviscosityoftheetching
solution,appliedcurrent,flowvelocityoftheetchingsolution,surfacecontamina-
tionsetc.theresultcanbeanythingfromanelectropolishedsurfaceoralocalattack
likepittingcorrosiontoanevenlydistributedmicrostructuring.Thelattercaseis
subjecttotheexaminationofthischapter.Thistypeofetchingoffersaninterest-
ingpossibilitytogeneratemicrostructuredsurfacesthatcouldhavegoodresistance
againstwearwhencoatedwithsoftmaterials.Itcouldbeshownthat,undercer-
taincircumstances,itispossibletogenerateanevenlyroughfacetedsurfacewith
HCl.Howeveritwasfoundthatthesetypeofetchingsareverydifficulttoreproduce.

Table5.1showstheresultsofetchingsonapin.Thesamples(fourforeachcurrent)
wereetchedatcurrentsof50mA(0.48mA/mm2)to400mA(3.86mA/mm2)in
stirredetchingsolution.Theetchingtimeswere20min,10min,5minand150
sec(for50mA,100mA,200mAand400mA).Thenumbersgiveninthetable
indicatetheapproximatedaveragedistancebetweentwofacetsinµm,calculatedby
countingtheintersectionsbetweenastraightlineof100µmlengthandthefacets.
Aslashindicatesthatnomicrostructurewaspresent.Whenamicrostructurewas
presentthat,however,didnothavecountablefacetsorthevaluewasbelow2µm,
itwasindicatedwithn.c.

Whenetchingonapin,thefacetedsurfaceswerefoundonlyinfewofthespecimens.
Mostspecimenshadasmoothsurfacewithoutanymicrostructure.Anexplanation
ofthisphenomenamightbefoundinthegenerationofadarkblacklayerconsist-
ingpossiblyofironoxidebetweenstentandpin,whichcansometimesbefound
afteretching.Thelayersmightpreventasufficientcurrentflowthroughthestent

52

Sampleno.50mA100mA200mA400mA
1/1411/
/8//2//n.c.n.c.3////4Table5.1:Averagedistanceinµmbetweenfacetsfordifferentetchingtimeswithpin,
measuredon4stentseach.

inmanycases.Nevertheless,interestinglytheformationofthislayerseemstobe
connectedwiththeformationprocessofthefacetedstructures.Inallcaseswhere
amicrostructurewaspresentthestentsurfacewascoveredwiththeblacklayer.
Soaconditionforthemicrostructuringmightbethepresenceofthelayerwhilea
sufficientcurrentflowisstillsustained.Afteretchingthelayerhastobecarefully
removedwhenexaminingthesamples.EspeciallyundertheSEMitcanbeeasily
mistakenwiththeactualmetalsurface.Fig.5.1showsanSEMimageofastent
wherethelayerwasnotremoved.Fig5.2showsoneofthetwomicrostructured
samplesofthe200mArow(top)andoneofthesampleswherethemicrostructures
couldnotbecounted(bottom).

Table5.2showstheresultsoftheetchingswithoutapin.Thesampleswereetched
atcurrentsof40mA(0.39mA/mm2)to240mA(2.31mA/mm2)instirredetching
solution.Theetchingtimeswere12min,6min,3minand2min.

Sampleno.40mA,12min80mA,6min160mA,3min240mA,2min
/11n.c.31n.c.73/224n.c.n.c.3/54/4Table5.2:Averagedistancebetweenfacetsfordifferentetchingtimeswithoutpin,mea-
suredon4stentseach.

53

Figure5.1:Hydrochloricacidetchedstentcoveredwithblackoxidelayer.Thelayercan
beeasilymistakenformicrostructures.

Figure5.2:Hydrochloricacidetchingwithpin,200mA,5min,twodifferentstents,
surfacesandsections.Thelowerimagesshowastentwherethefacettscouldnotbecounted.

54

Withouttheuseofapin,thefacetedsurfaceswerebetterreproducible.Atacurrent
of160mA(1.54mA/mm2)thefacetedsurfacecouldbegeneratedonallsamples.
Thecreatedmicrostructures,however,differtoaconsiderabledegreeinsizeand
distributionofthefacetedstructuresasshowninfig.5.4.Roughnessvaluesfor
tubesetchedwith1.54mA/mm2werebetweenRa=(0.22µm-0.71µm)and(Rz
=0.87µm-3.29µm).

Figure5.3:Hydrochloricacidetching80mA,6min,withoutpin,twodifferentstents.
Thelowerstentshowsgenerationofmicrostructures,theupperdoesnot.

55

egurFi

Note

56

5:4.

the

Hydrochloric

esencdiffer

in

the

idac

hing,etc

dateenerg

160

mA

thoutwi

es.cturostrurmic

n,pi

3

n,mi

two

different

s.stent

5.2AnalysisoftheFormationProcess

Byinterruptingtheetchingprocesspreliminarilyadetailedpicturecanbedrawn
oftheformationofthestructures.Onpicture5.5theinitialphaseofthisetching
methodcanbeseen.Inthehighestmagnificationthecreationofevenlydistributed
smallpitsofdifferentgeometries(smallblackarrows)canbeobserved.Asdescribed
in[131],triangularpitssuggestthattheattackedsurfaceisorientatedinthe(111)
directionwhereassquaredpitsindicatea(100)orientationandrectangularpitsare
createdat(110)planes.Someofthepitshaveprogressedtocavitiesofseveralµm
sizeshowingaplateletlikesurfaceinside(largewhitearrows).Severalfeaturesdis-
tinguishthisformofcorrosionfromnormalpittingcorrosion.Firstlyasaresultof
highmechanicalstrainduringtheprocessingofthetubesahighdislocationdensity
ispresent,enablingtheformationofahighnumberofevenlydistributedsmallpits.
Secondlytheeffectofthestirrerpreventstheformationofaconsistentetchfilmon
thesurface.Bythistheetchingprocessbecomesactivationcontrolledinsteadof
masstransfercontrolled,sothattheelectro-polishingeffectissuppressed.Further-
morepresumablynoprotectingremnantofapassivitylayercanbemaintainedover
thegrowingpitsasitisthecaseinpittingcorrosion[132].

Interestingly,asmentionedbefore,thegenerationofthemicrostructuresseemto
beinconnectionwithaninsolubleblacklayerofironoxide/hydroxide.Thislayer
mightdisturbboththeformationofaconsistentetchfilmandtheconservationof
thepassivitylayer.Becauseofthesefactorsnogrowinghemisphereswithsmooth
innersurfacescandevelop.Insteadaselectiveetchattacktakesplace,which,in
theinitialphase,createssmalltriangularorrectangularpits.Wherethesurfaceis
etchedfurtheritshowsplateletlikeappearanceaccordingtotheorientationofthe
grains.Whilethehighamountofpitspreventstheinhomogeneousgrowthofsingle
pitsanevenlyfacetedsurfaceiscreated.Sincetheformationofpitsislargelyunpre-
dictableanddependsonthepresenceofslightimpuritiesthismodelcouldexplain
thebadreproductivityoftheresults[133].

Fig.5.6showsaU/I-Diagramofthe316Lmaterialintheetchingsolution.Ithas
tobenotedthattheU/Icurveisdependentonthescanratewhichwasapprox.0.1
V/sec.Thehigherthescanrate,theloweristheresultingcurrentathighvoltages,
probablyduetoprotectioneffectsbythegrowingetchlayer.Thisfactalsohas
tobeconsideredwhenevaluatingapplicableetchingparameters.Theformationof
thestructuresseemstobelimitedtocertaincurrentdensitieswithinthelowerpart

57

ofthegraph.Athighercurrentselectropolishedsurfacesoccur.Atlowercurrent
densitiesthan1.4mA/mm2themicrostructuresoccuronlyoccasionally.

Figure5.5:InitialStageofhydrochloricacidetching.Theblackarrowsmarksmallpits,
indicatingthebeginningoftheformationofthemicrostructures,thewhitearrowsmark
largercavities,generatedthroughprogressedpitting.

58

Fiegur5.:6ammagri-DU/I

of

316L

stent

almateri

in

cchloriohydr

acid.

59

6MicrostructuringwithNitric
dicA

tsulRes6.1

ElectrochemicaletchingwithHNO3withintherangeoftheexaminedetchingparam-
etersleadstoahighlyselectiveattackongrainboundaries.Thegeneratedsurface
roughnesses(n=5,measuredonetchedtubes)werebetweenRz=1,96µmandRz
=8,87µmdependingontheetchtime.Howevertheseroughnessvaluescanonly
givesomevagueindications,aboutpossibleadhesionpropertiesforexample.They
cannotbeusedforacharacterizationofthesurfaces,sincehighaspectratiosand
crevicesarenotcaptured.Thereforecrosssectionalmicrographswereusedasthe
mainsourcefortheanalysis.Ithastobenotedthateventhoughthestentsare
etchedonpinscavitiesarenotonlylimitedtothethreesidesofthestentsexposed
totheacid,butarealsopresentattheinnersideadjacenttothepin.Thisisduetoa
slightdistortionofthestentsbythecrimpinganddilatationprocedureasdescribed
inthematerialsandmethodssectiononp.30.Forthesectionalmicrographsitis
importanttoconsideronlytherelevantoutersides,thereforeonallimagestheinner
sideisseparatedwithablackline.

Fig.6.1showstheetchingstructuresfordifferenttimesatacurrentof200mA
(1.31mA/mm2).The2minetchingshowsaslightgroovingofthegrainboundaries.
bouSectionsndarsies.howAtthe4fominrmaetctiohingnoftimtheinthecrgraevicesinbofo2untodar4ieµsmaredepthalreadyalongtheconsiderablygrain
broaance.dened.Aftera6Sectionsminshoetcwhingtheafovrerymatrioonugh,ofcagravitniesulatwitedhsunarfarrceowiscrV-shaeaptededthapproueagrh-
asdetacasurhedfgaceramoins.Andificationetchinfgorproimplcessantosfbthisecausestrengthoftheisrtiskoboferegdetacardedhedapasrticles.unsuitable

60

Fiegur6.1:ghtiSl

aigrn

min(bottom).The6min

arydnoub

etchising

oftheriskofdetachedparticles.

ng,etchi

200

,mA

2

nmi

(top),

4

toberegardedunsuitableforause

min

le)d(mid

and

6

forimplantsbecause

61

6.2AnalysisoftheFormationProcess
Especiallyforthestrongeretchingsincreasinglyspotsonthegrainsarevisible,where
thesurfacehasbeenattackedlocallywithoutthepresenceofgrainboundaries.In
somecasessupposedlysomesurfaceheterogeneitieswerepresenthere,leadingtoa
kindofpittingcorrosionsimilarasinthecaseofchlorideetching.Ontheother
handitalsoappearedasifinsomerarecasessinglegrainboundariesorparts
ofgrainboundarieswerenotattacked.Formoredetailedinvestigationofthese
phenomenascrosssectionalmicrographsofdifferentgrainboundaryetchedstent
andtubesampleswereinturnmetallographicallygrainboundaryetched(fordetails
s.appendix)tovisualizethepathsonwhichthecavitiespropagate(fig.6.2).The
rightimageshowsanexampleforacavitywithoutthepresenceofagrainboundary.
Withthisimageitcannotbetoldifthecavitywascausedbyacontaminationorby
intrinsicstructures.Thereforeasecondmetallographicetchingstepwasperformed
withoxalicacid.Withthistwostepmethodthegrainboundariesaswellasthe
interiorgrainstructuresaremadevisible.Fig.6.3revealsthatthecavityinthis
casefollowedadislocationoratwinboundary.Interestingly,however,mostofthe
twinsanddislocationswherenotaffectedbytheetching.Theseresultsindicatethat
theetchattackreactsquitesensitivelytowarddifferentpartlyunknownparameters.

Figure6.2:Grainstructureofetchedtubes,survey(left)andenlargedpicture(right).The
whitearrowontherightimagemarksanetchattackwerenograinboundaryispresent.

Inordertogetdetailedinformationabouttheselectivityoftheetchingprocess
independenceoftheappliedcurrent,whichisimportantwhenfindingbestsuited
etchingparameters,polishedsectionsofstentswerecreated,etchedwithdifferent
currentdensitiesjfrom0.2to6.9mA/mm2.Theetchingtimeswerebetween16min
(for0.2mA/mm2)and40sec(for6.9mA/mm2).Thatwayitisprovidedthatthe

62

Figure6.3:Crystallographicstructureofetchedtubes.Atthesiteswhereetchattack
occurredwithoutthepresenceofgrainboundaries,dislocationsortwinboundaries,respec-
tively,becomevisible(whitearrows).Ontheleftimageatwinboundarycanberecognized,
thecutoutontherightimagerevealsadislocation(visibleasafaintlightlinerunningfrom
thecavitytothegrainboundary).

Figure6.4:Crystallographicstructureofanetchedtube.Thewhitearrowsmarkatwin
boundary(leftimage)whichwasnotaffectedbytheetching.

63

etchingstrengthsareinapproximatelythesamerangeforallcurrentdensities.From
eachstentthelengthsofatleast15crevicesaswellasthewidthsofthecrevices
atthesurfaceweremeasuredondifferentstruts.Theratioofhalfthewidthtothe
lengthofthecrevicesisusedasavaluetocomparetheselectivityofgrainboundary
etchingprocesses.Thisselectivityfactorisillustratedindetailonp.69.Whena
highselectivityispresentdeep,narrowcrevicesdevelop.Whenalowselectivityis
presentthebulkmaterialremovalismuchhigher,resultinginshort,widecrevices
6.5.Usingthisindexfortheselectivityitisimportanttoexcludeoutliers,which
mayresultfromtwingrainboundariesorthelike.

Itcouldbeshownthattheselectivitydeclinesgraduallyfromlowcurrentstohigh
currentswithouthavingadistinctpeakatanyoftheanalyzedcurrentdensities
(fig.6.6).Thisobservationisincontrasttoconventionalgrainboundaryetching
modelsasdescribedbyBellin1976wheretheU/Icurveofgrainboundariesand
matrixisassumedasillustratedinfig.6.7[134].Inthisdiagramthegrainboundary
curveisassumedtohaveanoffsettowardbothhighercurrentsandlowerpotentials.
Theoffsettowardlowerpotentialswouldresultinalargedifferencebetweenlocal
currentsandthusahighselectivityattheareawhereahighslopeispresent.This
areaismarkedasthegrainboundaryetchingrangeinthediagram.Anoffsetonly
towardhighercurrents,however,wouldresultingradualdifferencesinselectivity
withoutadistinctpeak(fig.6.8).Anexplanationforthisoffsetisnotgivenin
[134],howeveritcanbeassumedthatacurrentshiftisratherduetothehigher
degreeofdisorderatthegrainboundarieswhileapotentialshiftshouldoriginate
fromdeviationinthematerialcompositionatthegrainboundaries.Accordingto
thistheselectivityvarieswiththeslopeofthecurveonlyiftheetchingprocessis
dominatedbyacompositiondeviation.Inthiscasethehighestselectivitywouldbe
achievedintherangeofvoltagewiththehighestslope.Aconstantcurrentshiftasit
wouldbecausedbyahigherdegreeofdisorderatthegrainboundaries,wouldresult
inagraduallydecliningselectivityoverthewholerange.Sothemeasuredresults
indicatethatchemicaldifferencesbetweengrainboundariesandmatrixmayplaya
lessimportantroleintheetchingprocessthanhithertoassumed.Inordertoobtain
additionalinformationsaboutthisphenomenonchemicalanalysiswasperformedon
grainboundaryetchedspecimens.

64

Figure6.5:Differentcrevicestructures,ontheleftimageahighselectivity(narrow
crevices)canbeseen,therightimageisanexampleforalowselectivity(broadcrevices).

Figure6.6:Selectivityoftheetchingsfordifferentcurrentdensities.

65

egurFi

:76.

U/I-Diagrammof316Linnitricacidmodifiedfromadiagrammin[134].

Figure6.8:U/I-Diagrammof316Linnitricacidbasedonadiagrammin[134](left)and
asmeasuredfortheanalyzedstents.

66

Figure6.9:U/I-Diagrammwithoffsetonlytowardshighercurrents.

6.3ChemicalAnalysis

EDSanalysiswascarriedoutinordertoinvestigatethecompositionatthegrain
boundaryareaandthegrainareaofthestents.Theanalysisgavenoindication
foranydepletionofpassivatingelementsintheareaofthegrainboundaries.Even
aslightenrichmentofchromiumcouldbefoundatthegrainboundaries.The
enrichment,however,couldbecausedbychromiumrichprecipitatesatthegrain
boundaries,whichdonotleadtoahigherresistance.

Ithastobenoticedthatthemaincompositionvariationsatthegrainboundaries
arefoundwithinarangeoflessthan100nmwhilethespotsizeoftheEDSis1
to2µmindiameter.Thereforeitwouldbeonlypossibletodetectmajordevia-
tionsincompositionwithinsuchalimitedzone.Foradetailedanalysisofthegrain
boundariesatransmissionelectronmicroscopeEDSwouldbenecessarywhichwas
notincludedinthiswork.

67

6.4MathematicalDescriptionoftheEtching
sescoPr

Inthefollowingsectionamodeloftheetchingprocessisestablishedwiththeaimof
calculatingthecrevicedepth,thecrevicewidthaswellasthematerialremovalfrom
theetchingparameters.Fig.6.10showsaschematicimageofanetchedsurface
withonegrainboundarywhichisthebasisforthefollowingcontemplations.

Figure6.10:Schematicillustrationofgrainboundaryetchingprocess,topimaget=0,
bottomimaget>0.

Themasslossratedm/dtoccuringatanareaAcanbecalculatedfromthecurrent
densityj(comp.p.18)

68

dtdm=M∙z∙γF∙j∙A

)(6.1

Itisassumedthattheefficiencyfactorisconstantovertime.Thesurfaceareais
notconstant,itincreasesduringtheetchingprocess.Howevertheproductj∙A
canberegardedasconstantwhencontemplatingafixedbaseareaA0oftheoriginal
surfaceasillustratedonthedrawing,sincethetotalcurrentisholdconstantduring
theetchingprocess.Thereforethecurrentj∙Acanbesubstitutedbytheconstantj0
∙A0.Nowthemasslossratecanbeintegratedintoanamountofdissolvedmaterial
ΔmperetchingintervalΔt.

Sincegeometricalaspectsarecontemplated,themasshastobetransformedintoa
:elumov

ΔV=Δm=M∙γ∙j0∙Δt∙A0(6.2)
ρ316Lz∙F∙ρ316L
Withthevolumeofthedissolvedmaterialknown,thespacialdistributionofthe
dissolvedmaterialhastobemodeled.Itisassumedthatthegrainboundaryis
verythinsothatapointPcanbedefinedwhichmarksanintersectionofthegrain
boundaryandthesurface(Fig.6.10,top).Duringtheetchingthispointtravels
downwards,ataconstantfactorfasterthantheremainingsurface.Thisfactor,
whichistherationbetweentheetchingrateatthegrainboundariesandthebulk
material,indicatestheselectivityoftheetchingprocess.Besideamaterialremovalin
verticaldirection,amaterialremovalfrompointPtowardbothhorizontaldirections
occurs.Thelatterdeterminesthewidthofthecrevices.Whenitisassumedthat
thebulkmaterialremovalrateisindependentofthedirection,theratiobetweenthe
materialthicknesslossbatthesurfaceandthecrevicewidth2∙dcanbecalculated
(Fig.6.10,bottom)applyingtheintercepttheorems
2d=2b=>d=a∙b(6.3)
aa+ba+b
abeingthedepthofthecreviceaftertheetching.acanbeexpressedintermsof
dbyapplyingaselectivityfactorsj,whichdescribestheratioofd/aatacurrent
j:ydensit

Solvingtheequationforbyields

d=sj∙d∙b
sj∙d+b

sjb=d∙sj−1

)(6.4

(6.5)69

sjmustbe>1,sinceavaluebelow1wouldmeanthatthegrainboundariesare
etchedslowerthanthebulkmaterial.Nowasecondequationcanbesetupwhich
describesthevolumeΔVindependenceofthecrevicewidthd:
lsΔV=A0∙(sj−j1)∙0.5∙d+Akk∙A0∙d2∙sj(6.6)
Thefirstpartoftheequationisthevolumelossatthesurface,namelythecontem-
platedareaA0timesb.Thesecondpartisanapproximationofthevolumefraction
ofthecrevice.Thispartcomprisesthecrosssectionalareaofthecreviced2∙sj
(=d∙a)aswellasafactorlk/Akwhichindicatesthetotallengthoftheexposed
grainboundariespersurfacearea.Aninfluenceoftheknotpointswheretwograin
boundariestouchisneglected.Thefactorlk/Akcanbemeasuredfrommicroscope
imagesof2slightlyetchedstents.Forthestentmaterialitwasdeterminedto77
.mm/mm

EquatingbothtermsforΔVandsolvingfordfinallyyields
AkAkAk∙M316L∙γ∙j0∙Δt
d=−lk∙(sj−1)±(lk∙(sj−1))2+lk∙sj∙ρ316L∙z∙F(6.7)
Theselectivityfactorsjcanbetakenfromdiagram6.6.Thechargenumberzcanbe
assumedas3forj>0.2mA/mm2(comp.p.25).TheFaradayconstantFis9.6485
∙107mAs.Therateofcurrentyieldγis1forthecontemplatedetchingprocess,the
currentyieldγisassumedas0.95(comp.[135]).Theaveragemolarmassof316L
hastobecalculatedfromthecorrespondingmolarmassesofthealloyingelements
(comp.[106])(MFe=55.85g/mol,MCr=51.99g/mol,MNi58.7g/mol,MMo=
/mol):g.9495

M316L=0.632MFe+0.173MCr+0.147MNi+0.0275MMo=55.46g/mol(6.8)
Thedensityofthematerialwasdeterminedto7.73g/cm3=7.73∙10−3g/mm3.

Calculatingthecrevicewidthforacurrentdensityof1.3mA/mm2withaselectivity
ofwidt2.9hanandd4.7anµetmcforhingthetimecrevΔtice=1depth,80secrespyieldsectivelya.valueTheseofva3.3luesµmarefoinracthecordancecrevice
withmeasuredvalues(6µm±2µmcrevicedepth).

Whenplottingthecrevicewidthagainsttimeitbecomesvisiblethatthecrevices
growthapproximatesarootfunction,whichismostpronouncedforlowselectivities.

70

Forhigherselectivitiesandsmallertimesthegrowthofthecrevicesisalmostlinear.
Thiseffectoriginsfromthedecreasingeffectivecurrentdensityduetotheincreasing
surfacearea.Forlargetimestthedecreasingeffectivecurrentdensitycancausea
changeinselectivity,sothattheformulacannotbeappliedforlargeetchingtimes
wherethesurfaceareaAt>>A0.

Figure6.11:Crevicelengthplottetagainsttime,accordingtoformula6.7fordifferent
currentdensities(j=1.3mA/mm2toj=7mA/mm2))withthecorrespondingselectivities
sj.Forsmalltimesandforlowselectivitiesthegraphresemblesasquarerootfunction,for
largertimesandforhigherselectivitiesthegraphapproximatesalinearfunction.

71

7CombinationEtchingsonNitric
AcidBasis

Thebasicideaofthecombinationetchingsistousemicrocrevicesgeneratedbynitric
acidetchingsinordertocreateagridofevenlydistributeddepots.Thisisrealized
byacombinationofthenitricacidetchingwithasecond,isotropicetchingstep.
Theaimofthesecondstepistobroadenthegrainboundaryfurrowsgeneratedby
thegrainboundaryetchingsothatlarger,widerdepotswithoutsharpcrevicesare
created.Therearetwotypesofetchingsolutionssuitableforthispurpose.Onetype
areelectropolishingsolutions,analyzedontheexampleofphosphoricacid.With
thismethod,smoothsurfacesarecreatedsothattheriskofprogressingcrevicesis
minimized.Thedisadvantageofthismethodisthefactthatthedepotvolumeis
tosomeextentdecreasedbecausethematerialremovalonthesurfaceisrelatively
high.Sothevolumethatisgainedbybroadeningofthedepotsisatleastpartlylost
byadecreasingdepthofthedepots.Thesecondpossibilityistouseanisotropic
microstructuringliketheHCl-etchingsanalyzedinchapter5assecondstep.With
thismethodafinemicrostructureissuperimposedonthegrainboundaryetched
structuresinawaythatthefurrowvolumeisincreasedwhilecrevicesarehollowed
out.ThismethodisanalyzedonetchingswithHCl.Sincethereproductivityofthis
etchingmethodislimitedasitwasalreadyfoundinchapter5analternativeacid
wasanalyzedintermsofsuitability.Forthatpurposeoxalicacidwaschosen.A
schematicillustrationofthethreecombinationetchingsisgiveninfig.7.1.Elec-
trochemicaletchingwithoxalicacidcreatessimilar,butfinermicrostructuresthan
HClwithgoodreproductivity.

Foreachetchingsolutionasuitedfirstgrainboundaryetchingstepwasevaluated
asabasisforfurtherexaminations.Theseevaluationswerecarriedoutonstentraw
materialtubesasdescribedonp.31.Usingthisfirststepbasicetching,different
etchingstrengthsforthesecondstepwereinvestigated.ForafirstevaluationSEM
imagesofsurfacesoftheetchedstentsandcrosssectionalmicrographsofembedded
stentswerecontemplated.Thecrosssectionalmicrographswerethenusedfora

72

Figure7.1:Schematicillustrationofdifferentcombinationetchings.Phosphoricacid
yieldssmoothroundedcavities,HClyieldsfaceted,rathernonuniformcavities,theeffectof
oxalicacidisinbetween.

softwareanalysisofthedepotproperties.Thesoftwarecreatesapolygonaround
thestentsection,determiningthelimitsofthedepots,sothatfeatureslikethedepot
depthandthestoragecapacitycanbedetermined.

Notesonthesectionalimages

Whencontemplatingthecrosssectionalmicrographsitisimportanttoexcludethe
innersideofthestentfromtheanalysis.Thissidewillbewithinthebloodstream
afterimplantation,soaroughsurfaceshouldbeavoided.Whenetchingvirginstents
theinnersideisalmostcompletelyprotectedfromtheetchingprocess.However,
asdescribedonpage30,previouslycrimpedstentswereusedforallexperiments,
unlessotherwisenoted.Soinmostcasesanetchingoftheinnersidecouldnotbe
avoided,whichcanleadtoconfusion.Thereforeonallimagestheinner(luminal)
sideisseparatedwithablackline.Thesectionswereperformedonundilatated
stents.Itwasshownbycrosssectionalmicrographsindirectionsparalleltothe
stentaxis(comp.p.102),thatthesizeofthedepotswithintheareasaffectedby
thestraindoesnotsignificantlyincreaseduringdilatation.Ashortapproximation
ofthetheoreticalstrainvalidatesthisobservation(s.appendix).

73

7.1PhosphoricAcid
PhosphoricAcidwasusedatcurrentsbetween200mA(1.31mA/mm2)and400mA
(2.62mA/mm2)inundilutedform.Phosphoricacidcausesalevelingandsmoothen-
ingofthesurface.Ithastobeassured,thatasufficientcleaningofthestentsis
guarantiedafteretching,sinceetchingwithphosphoricacidproducesinsolublere-
actionproducts.

7.1.1ParameterAnalysisonStents
Asabasisforthephosphoricacidcombinationetchingsafirstetchingstepwith
HNO3with4minat200mA(1.31mA/mm2)wasused.Thisetchingyieldsan
averagegrainboundaryfurrowdepthofabout7to8µm(s.fig7.2).Slightergrain
boundaryetchingswereregardedaslesssuitableinpreliminaryexperimentsbecause
thedepotslostalargeamountoftheirvolumeasaconsequenceofthelevelingeffect
ofthephosphoricacid.Thephosphoricetchingstepwascarriedoutonapinat
etchingstrengthsbetween200mA*120secand400mA*180sec.

Figure7.2:Grainboundaryetching,200mA,4min,section.Thisetchingwasusedas
thebasisforallphosphoricacidcombinationetchings.
Aphosphoricacidetchingat200mAfor90secyieldsaslightsmootheningofthe
crevicetipsasvisibleonthepolishedsections.Theshapeofmostofthedepots
isnotnoticeablychanged(fig.7.3,top).Increasingtheetchingtimeto180sec
produceswidenedandrounded,butstillV-shapeddepots.Etchingat400mAfor
120secproducesmoreevenlyhollowedoutdepots(fig.7.4,top).Thedepthofthe
depots,however,isdiminished.Atthisstagenocrevicesarepresentanymoreatthe
74

relevantsidesofthestent.A180secetchingat400mAfinallylevelsalldepots,so
thatthestorageeffectseemstobeconsiderablydiminished.

Figure7.3:Phosphoricacidcombinationetching,200mA,90sec(top),200mA,180sec
ottom).(b

Atmostoftheillustratedpicturestheinnersideofthestentisalsoaffectedby
theetching.Thiseffectisduetobadprotectionbythepinduringthefirstetching
stepwhichisaconsequenceofusingelectropolishedstents(88).Thisisarather
undesiredsideeffectwhichcancausedamagesontheballooncatheter.Thecrevices,
however,canbeleveledifthesecondetchingstepiscarriedoutwithoutapin.A
pictureofanintermediatephosphoricacidcombinationetchingwithoutapinis
showninfig7.5.

75

egurFi

7.4:

ottom).(bcse

cPhosphori

acid

Figure7.5:phosphoricacid

76

onatinombic

etching,200

ng,etchi

400

mA,

120

cse

mA,180sec,secondstep

,p)o(t

detche

400

A,m

180

in.ptwithou

7.1.2SoftwareAnalysis
Fig7.6showsanexampleforaslightphosphoricacidetchedstentaftersoftware
analysis.Thebeginandtheendofeachdepotaremarkedwithgreencrosses,the
centerpointsaremarkedwithlightbluecrosses.Redcrossesmarkthepolygonedge
points.Infig.7.7theaveragedepotareadividedbythelengthofthethreeside
polygonline(storagecapacity)aswellastheaverageeffectivedepotdepthareillus-
tratedforalletchings.Amaximumoftheaveragestoragecapacityandthedepot
effectcanbefoundwithintherangebetween200mA*120sand200mA*180s.
Contrarytotheobservationonthecrosssectionsbothvaluesdonotseemtogreatly
decreaseathighetchingstrength.Thereasonforthisarefewverylargespaces
thatcontributealargefractionofthecalculatedstoragecapacity.Thesespaces
arecausedbylargeareasspannedbythepolygonduetocoalescenceofcavitiesat
strongeretchings(comp.fig.7.8).Suchlargespacemayalsooccuratslighteretch-
ings,whenacavityisincidentallycutexactlyinitslongitudinaldirection.However
thisoccursrarely,soacumulativeoccurrenceisastrongindicationforcoalescence
s.eviticaof

Figure7.6:Exampleforasoftwareanalysisofanetchedstent,phosphoricacidcombina-
tionetching,200mA,90s.

77

Figure7.7:Depoteffectofphosphoricacidetchedstents.

Figure7.8:Softwareanalysisofanetchedstent,phosphoricacidcombinationetching,
400mA,120s,thewhitearrowmarksalargedepotgeneratedthroughmergingdepots.

78

egurFi

y.ve

7.9:

eSoftwar

ysisanal

fo

dtchee

,stents

icphosphor

daci

inationombc

hing,cte

-sur

79

7.2HydrochloricAcid
Theacidwasusedinformof20%HClatacurrentof200mA(1.31mA/mm2).The
resultsrangefromasmoothingeffectsimilartothephosphoriccombinationetchings
toamicrostructuringeffectasdescribedinchapter4.Reasonablyreproduciblere-
sultscouldonlybeachievedbyetchingwithoutapin.Stillwiththismethodetching
severalsampleswiththesameparametersdidnotyieldexactlyidenticalresults.

7.2.1ParameterAnalysisonStents
Itwasfoundinpreliminaryexperimentsthata3minHNO3etchingat200mA(1.31
mA/mm2)yieldthelargestdepots.Thisetchingyieldsanaveragegrainboundary
furrowdepthofabout6µm(fig.7.10).

Figure7.10:Grainboundaryetching,200mA,3min,section.Thisetchingwasusedas
thebasisforallcombinationetchingsexceptthephosphoricacidetchings.

ForthesecondstepHCletchingswithoutapinbetween100mA(0.96mA/mm2)*
30secand200mA(1.91mA/mm2)*60secwereexamined.Thecombinationetch-
ingobviouslyledtoasuperpositionoftherough,facetedsurfaceoftheHCletching
overthegrainboundaryfurrowstructure.100mAappliedfor30secproducessmall,
evendepotswiththedepottipsalreadybeingrounded(fig.7.11).Increasingthe
currentto200mAleadstomuchlargerdepots.However,itwasfoundthatthe
depotsizesandshapesvariedbetweendifferentstents.At50secetchingtimemost
ofthedepotshavebeendiscreated.Strikinglythetopviewimageseemstoreveal
lessmicroroughnesscomparedtothe30secsample.

80

Figur200

e7,mA

.1150

:oninatCombi

cse

tom).ot(b

etching

,HCl

100

mA,

03

ces

(top),

200

,mA

30

cse

(midd

le),

81

7.2.2SoftwareAnalysis

InFig.7.12theresultsofthesoftwareanalysisofthedepotstructuresareshown.
Amaximumoftheaveragestoragecapacity(3,06µm3/µm2)atalargeeffective
averagedepotdepth(2,54µm)canbefoundatanetchingstrengthof200mA*30
s.Incontrasttothephosphoriccombinationetchingsaconsiderabledropofboth
storagecapacityandeffectivedepotdepthispresentwhenincreasingtheetching
th.gstren

82

Figure7.12:Depoteffectofhydrochloricacidetchedstents.

egurFi

vey.sur

1.73:

eSoftwar

sianalys

fo

detche

stents,

icchlorohydr

cida

inationombc

etching,

83

7.3OxalicAcid

Theeffectofelectrochemicaletchingwithoxalicacid,whenusedasacombination
etching,isinbetweentheeffectofphosphoricacidandhydrochloricacid.Besides
alevelingeffect,afinesuperimposedsurfacemicroroughnesssimilar,butfinerthan
forHClisgenerated.Twinboundaries,dislocationsandgrainboundariesareetched
slightlyfavoredwhenetchingwithoxalicacid.Thisgrainboundaryetchingeffect
isnotnearlyaspronouncedasitiswhenusingHNO3.Howeverifoxalicacidis
usedasacombinationetchantwithHNO3,theeffectofthefirstgrainboundary
etchingstepisslightlysupported.Ithastobeassured,thatasufficientcleaningof
thestentsisprovided,sinceoxalicacidetchingproduceshardlysolublecompounds.
Theetchingswerecarriedoutonapinwith10%oxalicacidatcurrentsfrom200
mA(1.31mA/mm2)to400mA(2.62mA/mm2).

7.3.1ParameterAnalysisonStents
InpreliminaryexperimentsetchingwithHNO3at200mA(1.31mA/mm2)for3min
yieldedevenandreproducibledepotsforoxalicacidcombinationetching.Whenus-
ingstrongerHNO3etchingsthestructureswerelessuniformandthedepotvolume
didnotincreasenoticeably.Thusa3minfirstetchingstepwaschosenasfirstetch-
ep.sting

Bothetchingstepswereperformedonapin.Theoxalicacidetchingstepwascarried
outatetchingstrengthsbetween200mA*150secand400mA*120min.The
etchingsyieldeddepotssimilartotheseofthehydrochloricacidetchingsbutwith
afinerandlesspronouncedfacetedsubmicrostructuresuperimposedonthegrain
boundarystructures.

A150secetchingat200mAproducesrathersmall,butquiteuniformdepots(fig.
7.14).Increasingthecurrentto300mAat120secdoesnotyieldnoticeabledif-
ferencestothepreviousetching.Atatimeof180secfirstgrainsstarttodetach,
creatinglargeshallowdepots.Thedepotdepthseemstobedecreasedincomparison
withthepreviousetching(fig.7.15).At180secat400mAmostofthesingledepots
havemerged,creatinganinhomogeneoussurfacewithoutdistinctdepots.

84

egurFi

(midd

4:17.

le),

300

inatCombion

mA,

180

ces

oxalic

daci

.)omott(b

hing,etc

002

mA,

150

sec

,(top)

300

,mA

120

cse

85

Figure7.15:Combinationoxalicacidetching,400mA,180sec.

7.3.2SoftwareAnalysis

Fig.7.16showstheresultsofasoftwareanalysisofthedepotstructures.Amax-
imumofthestoragecapacityaswellastheeffectivedepotdepthcanbefoundin
therangeofananetchingstrengthof300mA*120s.Theresultisingoodac-
cordancewiththeobservationsontheimages.Atloweretchingstrengththedepots
areonlyslightlybroadened,athigheretchingstrengthmuchofthedepoteffectis
lostthroughmaterialremovalandmergingofdepots.

86

Figure7.16:Depoteffectofcombinationoxalicacidetchedstents.

egurFi

7:1.7

eSoftwar

sianalys

fo

deetch

stents,

oxalic

daci

nationombic

tehing,c

ey.surv

87

7.4EtchingswithoutpreviousElectropolishing

Alloftheaboveetchingswerecarriedoutonelectropolishedstents.Ontheexam-
pleoftheoxalicacidcombinationetchingmethoditwasevaluatedifitispossible
towaivetheelectropolishingprocessbyusingrawstents.Thesestentsareonly
subjectedtoapicklingtreatmentdirectlyafterthelasercuttingprocessinorderto
removeslagandexcrescentoxidelayers.Thestentswereetchedintheasproduced
state.Thatmeansincontrasttoallotherstentsusedinthisworkthestentswere
notmechanicallystressedbeforeetching(comparesec.Materials).

Theupperpicturesonfig.7.18showsastentetchedwith1.HNO3,200mA,4
minand2.oxalicacid300mA,2min,bothstepsetchedonapin.Thesquare
shapedgeometryofthestentwasalmostcompletelymaintainedwithmoreorless
sharpangles.Aroundingoftheedgescanhardlybeobserved.Strikinglytheinner
sideofthestentisalmostcompletelysparedbytheetcheffecthavingonlysmall
microstructures.Thesesmallmicrostructurescouldpromotetheproliferationof
endothelialcellswithouttheriskofdamagingtheballooncatheter.Primarilythe
goodprotectionwillbeduetothemuchbettercontacttothepinoftheasproduced
stents.Howeveranotherimportantfactoraretheunroundededgeswhichprevent
acidofgettingbetweenstentandpin.Thepicturebelowshowsastentwherethe
secondstepwascarriedoutwithoutusingapinat150mAfor2min.Theinnerside
(thebottomsidefortheleftpictureandtherightsidefortherightpicture)shows
muchmoreetchattackthanthefirststents.Howevertheprotectionoftheinnerside
isstillbetterthanfortheelectropolishedstents.Concerningthesharpanglesthe
secondetchingstepwithoutapinresultedinroundingoftheedges.Interestinglynot
onlytheinneranglesbutalsotheouteranglesaremuchmorerounded.Tworeasons
canbefoundforthis.Firstlywithoutthepinabetteracidflowdevelopsaround
thestentstruts.Secondlythedepthofalldepotsindicatethatthecurrentdensity
ofthesecondetchingstepwasprobablyhigherthanforthefirststent.Althoughit
wasattemptedtousethesameparametersforbothstentsthecurrentdensitiesof
etchingswithandwithoutpincannotbedirectlycompared(comp.33).

88

Fiegur:7.18ingrVi

stent

t(withou

pretreatment),

inationombc

setching

twithou

1.step:200mA180sec,2.steponapin:300mA,100sec;bottom1.step:

pin,

top:

mA200

210sec,2.stepwithoutapin:300mA,120sec(bottom).Theslighteretchingsabove

stillshowthesharpedgesgeneratedbythelasercutting.Atthestrongeretchingsbelowthe

edgesareroundedbythesecond,isotropicetchingstep.

89

8Performanceofthemodified
Stents

Inthepreviouschaptersfiveelectrochemicalsurfacemodificationmethodshavebeen
developed,hydrochloricacidetching,grainboundaryetchingandthreecombined
methodswithgrainboundaryetchingasbasis(phosphoric,hydrochloricandoxalic
acidetching).Oneofthemethods,themeregrainboundaryetching,apparently
isnotsuitedfortheuseonstents.Twomethods,thehydrochloricacidetching
andthehydrochloriccombinationetching,turnedouttobenotsufficientlyrepro-
ducible.However,theremightbeotherpossibleapplicationsfortheseetchingmeth-
ods.Imaginableisauseintheareaofotherimplantslikee.g.boneimplants.A
possibleapplicationcouldbesurfacemodificationofanycomplex,sensitivepart
whereahighsurfaceroughnessisrequiredandgritblastingisnotpossible.There-
foreallfivemethodshavebeeninvestigatedintermsoftheirchemicalandbiological
performance.Thephosphoricandoxaliccombinationetchings,withwhichrepro-
duciblesurfacedepotscanbegenerated,werealsoexaminedintermsofcoating
behavior,releaseperformanceandmechanicalproperties.Fortheseinvestigations
thefollowingparametersetswerechosen.TheHCletchingswerecarriedoutwith-
outpin,allotheretchingswereperformedonapin.

90

stepsinglegrainH3PO4oxalicHCl
HClboundarycomb.comb.comb.
1.step,HNO3/200mA200mA200mA200mA
/5min4min3min3min
2.step80mA/200mA300mA200mA
5min/3min2min30sec
Table8.1:Etchingparametersallanalyzedstents.

8.1ChemicalPerformance

8.1.1CyclicVoltammetry

Cylclicvoltammogrammswererecordedforalletchings.Fig8.1givesasurveyofthe
differentetchedsurfacesincomparisonwithgritblastedandelectropolishedstents.
Thefigureshowsonlythemostsignificantpartofthecyclicvoltammogrammswhich
isthepassivityareaoccurringatthescaninanodicdirection.Thefactthatthecur-
rentatthepassivityareadiffersconsiderably(abouttwoordersofmagnitude)can
probablybeattributedtoadifferentsurface(micro-)roughnessandthusadiffer-
enteffectivesurfaceareaofthestents.Accordinglythesmoothelectropolishedand
phosphoriccombinationetchedstentsshowamuchlowerpassivitycurrentthanthe
gritblastedandtheoxalicacidetchedstentswhichhaveaveryhighmicroroughness.

Therestpotentialsofdifferentetchedstentsareshowninfig.8.2.Therestpotential
isthepotentialwhichismeasuredwhennoouterpotentialisapplied.Itgivesan
indicationforthechemicalactivityofthesurface.Arestpotentialclosetozero
indicateschemicalinactivity.Thepotentialswererecordedafteraconstantvalue
wasreached,whichlastaboutonehourformostofthespecimens.Inmostcases
thepotentialvariedtosomeextendbeforereachingaconstantvalue,soitcanbe
assumedthatinitiallysomechemicalsurfacereactionstookplace.Theetchedstents
showarestpotentialsimilartothegritblastedstent.Onlythe(absolute)valueof
theelectropolishedstentisslightlylowerthanthatoftheotherstents.

Fig.8.3showsacomparisonofthepassivityareasofthedifferentsamples.The
passivityareaindicatestherangebetweenthethefladepotentialandthebreak
throughpotential.Thefladepotentialindicatesthepointatwhichpassivationbe-
ginsandisdefinedbytheinflectionpointafterthefirst(prepassive)maximum.The
breakthroughpotentialinthiscaseisdefinedbythepotentialatwhichthecurrent
increasedtotwicethevalueofthe(lowest)passivitycurrent.Thepassivityareasof
allstentsaresimilar,excepttheoxalicacidetchedsample,whichhasamarkedly
rea.aendedext

Besidesthecharacteristiccurvesallsamplesshowsomecurrenttransientswhen
approachingthebreakthroughpotential(s.fig.8.1).Mostpronouncedisthisphe-
reasonomenonnforforthisthecouldhydrboecpittinhloricgacidcorrosioetchedn.andSEMtheimaphgesoofsphoricthespacidetecimenscheddidnotsurfaces.showA

91

92

Fiegur8.:1vesCur

of

tassivipy

aear

of

entreffdi

modfiedi

.serfacsu

Figure8.2:Restpotentialsofdifferentmodifiedsurfaces.
withtheelectropolishedhavingaslightlylowerpotential.

Thesurfacesarecomparable

Figure8.3:Passivityareasofdifferentmodifiedsurfaces.Thepassivityareaofoxalic
acidismarkedlyincreased.

93

anyevidentpittingcorrosiononthesurface.Howeverwithinthemicrostructures
pittingcorrosioncouldstillhaveoccurredwithoutbeingvisible[126].

8.1.2ReleaseTestsofNickel

Allanalyzedsurfaceshavebeensubjecttoananalysisoftheionreleaseproperties.
Theinterestwasfocusedonthereleaseofnickelandchromiumsincethesearethe
mostcriticalionsintermsoftoxicity.Thebreakthroughpotentialwasusedasa
referenceforthistests.Theamountofreleasedionswasdeterminedafter60minin
Ringer´ssolutionat37◦Catapotentialof50mVabovethebreakthroughpoten-
tial.Thechromiumreleasewasbelowthedetectionlimitof2µg/lforallspecimens
exceptonesample,namelythegrainboundaryetchedone(6µg/l).Thereleaseof
nickelisillustratedin8.4.Thevalueforthegrainboundaryetchinghastobecon-
templatedwithcaution,sincewiththissampleproblemsoccurredwiththepotential
control,sothatthetesthadtobecanceledafter40h.Whenthetestwasrepeated
thenickelreleasecouldstillnotbedeterminedduetotechnicalproblemsduringthe
analysis.Howeverthechromiumrelease,whichcouldbedetermined,wasingood
congruencewiththepreviousvalue(5µg/l),affirmingthehighreleaseofthegrain
boundaryetchedsamples.

Figure8.4:Nickelreleaseofdifferentmodifiedsurfaces.

Accordingtothemeasureddatathenitricacidetchingseemstoproducethemost
criticalsurface.HCletchingsproduceslightlybetterresults.Theoxalicacidetching

94

with2,3µg/landthephosphoricacidetching(belowdetectionlimitof2µg/l)had
thelowestreleaseofnickelions.Thereleaseoftheelectropolishedstentswasbelow
thedetectionlimit.Anobviousconclusionoftheseobservationswouldbethatnitric
acidorhydrochloricacidchangethepropertiesofthepassivationlayerinaway
thatthepassivationandreleasepropertiesareeffectedinanegativeway.However
thiswouldbeincontrasttomostexaminationsthatcanbefoundinliterature.
Nitricacidaswellaselectropolishingsolutionsaremostlyreportedtosupportthe
passivationproperties[89]ofstainlesssteelbyincreasingthechromiumcontent
withinthepassivitylayer[62,112].Oxalicacidontheotherhandisratherattributed
toadenudationofchromium[136].(Ananalysisofthepassivitylayerscanbe
foundinthenextsection.)Anexplanationforthiscontradictioncanbefoundin
thetopographiesofthedifferentsurfaces.Astrikingaccordancecanbeseenonthe
occurrenceofnarrowcrevicesorhighmicroroughnesswithahighreleaseofnickel.
Oneexplanationthatsuggestsitselfistheincreasedsurfacearea.Anotheraspectis
thepresenceofacidremainswithincrevicesthatarenotremovedbythecleaning
process.Theseacidremainsmightcauseanacceleratedionrelease.Especially
nitricacidetchingcreatesverynarrowcrevicesinwhichacidcouldremain.The
electropolishedsurfacesaswellasthephosphoricacidetchedonescreateverysmooth
surfaceswhichcouldexplainthelowrelease[126].

8.1.3AnalysisofthePassivityLayer

AnanalysisofthepassivitylayercompositionwithAESforanoxalicacidetched
stentsandforanelectropolishedstentisshownin8.5.Allsamplesshowarelatively
highcontentofironcomparedtoliteratureandthepresenceofnickel[121,89].How-
ever,thenickelpeakmayresultfromaninfluenceofthebulkmaterial.Interestingly
nochangeincompositionofthepassivitylayercouldbefound,neitherbetweenthe
twostentsnorbetweenthesurfaceatthegrainboundariesandthematrix[126].

8.2BiologicalPerformance

8.2.1CellSeeding
Intheprevioussectionstheeffectoftheetchingtreatmentsonthechemicalperfor-
manceofthesurfaceshavebeeninvestigated.Althoughnosignificantdifferences

95

Figure8.5:AugerAnalysisofstentsurfaces.Nodifferencescanbedetectedbetweengrain
boundariesandbulk.

inthecompositionofthepassivitylayercouldbemeasured,theelectrochemical
behaviorofthemodifiedsurfacesdiffered.Inordertogetinformationsaboutthe
effectoftheelectrochemicaldifferencesaswellastheeffectofthesurfacetopogra-
phiesoncells,seedingexperimentswerecarriedout.Alletchingtreatmentswere
analyzed,exceptthemerehydrochloricacidetching.Asurveyofallsurfacesis
givenin8.6.Asreferenceservedgritblastedsurfaces.Threetypesofcellswere
chosen:endothelialcells(primarycells),myo-fibroblasts(USCS,primarycells)and
fibroblasts(cellline).Fordetailss.appendix.Asdescribedonp.6endothelial
cellsconstitutetheinnerlayerofanartery.Therecreationofthislayerafterthe
stentingprocessdecidesabouttheoccurrenceofrestenosis.Myo-fibroblastswere
chosenasprecursorsofsmoothmusclecells.AsRestenosisisaresultofanexces-
siveproliferationofsmoothmusclecellstheproliferationbehaviorofmyo-fibroblasts
maygivesomeindicationsforpossibletoxicreactions.Theendothelialcellsandthe
myo-fibroblastsshowednonoticeableproliferationforallexaminedspecimens.Also
forthegritblastedreferencespecimensthecellsdidnotproliferate.Nevertheless
forallseededspecimenssomecellactivitymaintaineduntiltheendofthetimepe-
riod.Thisindicatesthatallofthesurfacesareatleastnotcytotoxic.Howeverno
informationintermsofacomparisonbetweenthespecimenscouldbeobtainedfrom

96

thesecells.Primarymyo-fibroblastsandendothelialcellsareverysensitive,sothat
adirectseedingonametalsurfacemaybeproblematicingeneral.Inliterature[32]
comparableexperimentswereperformedwithendothelialcellswheremigrationon
316Lsurfacescouldbeobserved.Howeverinthiscasethemetalspecimenswere
implantedintoaconfluentendothelialcellseededsurfaceandmigrationwassup-
portedbysimulatedbloodflow.Ithastobenoticedthatininvivoperformance,
endothelialcellsorsmoothmusclecellswillnotbeindirectcontactwiththestent
material.Normallybetweenstentandcellsathinlayeroffibrinandotherorganic
materialwillbepresent[137].Insteadoffurtherexperimentswithsensitivearterial
cellsmorerobustfibroblastwereusedforthefollowingtestsinordertogivecom-
parableinformationaboutthegeneralbiologicalperformanceofthesurfaces.

Thefollowingetchparameterswereused,analogtotheprevioussections(comp.p.
).90

stepgrainH3PO4oxalicHCl
boundarycomb.comb.comb.
1.step,HNO3600mA600mA600mA600mA
7min5min5min5min
2.step/600mA600mA600mA
/7min7min5min
Table8.2:Etchingparametersfortheseededspecimens.

Fibroblastsproliferatedonallsurfaces,tovaryingextent(fig.8.9).Whilethesand-
blastedaswellasthemeregrainboundaryetchedsurfacesshowedonlymoderate
proliferation(fig8.7),ontheoxalicandhydrochloriccombinationetchedsurfaces
considerablyincreasedproliferationcouldbeobserved(fig8.8).SEManalysisofthe
65fibr%oinblastscaseshoofwtedheagritcoverageblastedofs100amples.%inAcasecomofpartheisonHClbetwcomeenthebinationdifferencomptacomredtbio-
acinationdetcetchingshingsandshoawnedintamarkermediateedlydecresultreafosedrtheprolifeorxalicatioancidforetcthehingsmo.(soth.p8.9)hosphoric

97

Figure8.6:Surfacesusedforseedingtests(fromlefttoright),top:grainboundaryetched
andgritblasted,bottom:hydrochloric,oxalicandphosphoricacidcombinationetchedsur-
.efac

Figure8.7:Fibroblastseedingongritblastedandgrainboundaryetchedsurfaces(small
pictureunseededsurface).Thesurfaceisseededincomplete.

98

Figure8.8:FibroblastceedingonHClcombinationetchedsurfaces(smallpictureunseeded
surface).Thesurfaceiscompletelycovered.

Figure8.9:Comparisonofcellactivitysignalsofdifferentsurfacetreatments.TheWST-
signalsindicatethemetabolicactivityofcellsprovidinganindexfortheproliferationand
theviabilityofcells.Notetheweaksignalofthephosphoricacidetchedsurface.

99

8.3MechanicalPerformance
Forstents,accordingtoDIN-EN12006fornotactivesurgicalimplants,staticand
dynamicstabilityhastobeproved.AccordingtoDIN-prEN14299dynamicsta-
bilityhastobeprovedfor380mio.cyclesundersimulationofinvivoconditions.
Especiallysurfaceetchingscanleadtoachangeinmechanicalpropertiesthrough
variouseffects.Ontheonehandetchingofsurfacesmeansamaterialremovalso
thattheloadbearingcrosssectionalareaisdiminished.Thiseffectcanbecompen-
satedbyaproperadaptionofthestentdesign,namelytheallowingfortheexpected
materialremoval.Amorecriticalproblemisthemodifiedsurfaceinrespectto
thehighcyclefatiguebehavior.Thefurrowsgeneratedthroughtheetchingprocess
workasnotchesandmayrepresentinitialpointsforcrackpropagationduringthe
cyclicstrainoftheheartbeat.Fig8.10showsacomparisonofdifferentmodified
surfacesinrespecttotheirnotcheffects.Shownistheaveragenumberofnotcheson
astentstrut,splitintothreecategoriesaccordingtotheirnotcheffects.Anumber
largerthan10indicatescriticalnotcheswhileanumberlowerthan1indicatesshal-
lowdepotswithlowdepoteffect.Anotchnumberoftenwaschosenasacritical
limitasinternstudiesshowednonotchnumbershigherthantenforconventional
(gritblasted)FDAapprovedstents.Thenotcheffectswereobtainedbydividing
thelengthofthedepotsbythecurvatureradiusofthenotchtip.Themeregrain
boundaryetchedsamplesexhibitalmostonlycriticaldepotswithanotchnumber
of10andabove.Withthephosphorousandtheoxalicacidcombinationetchings
allofthesedepotscanbeconvertedintolesscriticalnotches.Interestinglyalsofor
thegritblastedstentsfewnotcheswithhighnotcheffectcouldbefound.

8.3.1RadialStrengthTests

Radialstrengthtestwereperformedwithtwooxalicacidcombinationetchedstents
andtwophosphoriccombinationetchedstents.Detailsonthisprocedurecanbe
foundin[138].Thetestsshowedthattheradialstrengthisconsiderablyweakened
forbothetchings.Thecollapsepressureofthephosphoricacidetchingswas0.19
and0.20bar,thepressuresmeasuredfortheoxalicacidetchingswas0.35and0.32
bar.Thereferencestent,incomparison,hadacollapsepressureof0.66bar.This
resultswerenotunexpectedsincealletchingproceduresleadtoamaterialremoval
aswellastoadiminishedloadbearingareathroughthemicrostructures.Asacon-
sequence,forcommercialuseofthesurfaceetchedstents,anewstentdesignwould

010

Figure8.10:Averagenumberofnotchesonmodifiedsurfaces(referringtoonestent),
splitintothreecategories,high,intermediateandlownotcheffect.Thecombinationetching
considerablydecreasedthenotcheffectswithoutlevelingthedepots.

havetobedesignedwithwhichthematerialremovalaswellasthediminishedload
bearingareaiscompensated.

8.3.2FatigueTests

Toevaluatethefatiguebehavioroftheetchedstents,atestchamberwasconstructed
whichsimulatestheinvivoconditionswithanacceleratedfrequencyof70Hz.Three
stentseach,oxalicacidandphosphoricacidcombinationetchedweretested,with
anelectropolishedstentasreference.Besidesagrainboundaryetchedstentwas
testedsimultaneouslyasnegativecontrol.Theetchingswithhydrochloricacidwere
nottestedduetotheirinsufficientreproductivity.

Polishedsectionswerepreparedofalltestedstentscreatinghorizontalcuts.Special
interestwaspaidtothedepottipsattheinnerandoutersurfacesofthebends.
Uptoamagnificationof2000nofatiguecrevicescouldbeobservedatanystent.
Surprisinglyevenforthegrainboundaryetchedstentsusedasnegativecontrollno
fatiguecrevicescouldbeobserved(fig.8.11).Thestrainsimposedduringthetests

110

seemtobebeyondrelevantstrainsforthematerial.

Figure8.11:Sectionthroughcurvaturesofatestedstent,top:oxaliccombinationetching,
bottom:grainboundaryetchedstent.Atnoneoftheobservedsectionsanyindicationscould
befoundforcrackspropagatingfromtheetchedstructures.

8.4CoatingBehavior

Inthefollowingsectiontheperformanceofdrugcoatingsonthesurfacemodified
stentsisexaminatedusingtheoriginalcoatingmethodwhichisemployedinthe
catheterlaboratories.

AllcombinationetchedstentswerecoatedwithRapamycinusing1%and2%solu-
tioninethanol.Thecoatingsareuniform,withtheRapamycinapparentlybeing
absorbedbythemicrostructures(s.fig.8.12).Sectionsofthecoatedstents(s.p.

210

Figure8.12:Oxalicacidcombinationetchedstentscoatedwith1%(left)and2%ra-
pamycin,afterdilatation.Thecoatingsappearuniformandintact.

38)revealedanapproximatethicknessofabout10to15µm.Withthismethod,how-
ever,itcouldnotbeshowntowhatextentthedepotswerefilledwithRapamycin,
sincemostofthedrugwaswashedoutduringthepolishingprocess.

Moredetailedinformationcouldbeobtainedbyinvitroperformancetests.Adhesion
testsaswellasreleasekineticsimulationswereperformedwithallthreecombination
etchingsaswellaswiththefacetedhydrochloricacidetchedstents(fig.8.13).The
stentswerecoatedwith1%Rapamycin.Fordetailedinformationsee[124].Very
pooradhesionpropertieswerefoundforthefacetedsurfaces(>50%lossduringthe
simulatedimplantationprocess).Interestinglythepropertiesweremuchworsethan
thoseoftheelectropolishedstents(approx.25%loss),althoughtheroughnessof
theetchedstentsisconsiderablyhigher.Anexplanationcouldbetheedgesofthe
microstructuresactingasbreakpointsforthecrystallinedruglayer.Obviouslythe
layerisweakenedbythestructureswhilethemicroadhesionisnotimproved.The
adhesionpropertiesofthephosphoricandthehydrochloricacidcombinationetched
stentsalsoseemtobeconsiderablydeterioratedcomparedtogritblastedstents.
Thesurfaceofthephosphoricacidetchedstentsisprobablytoosmoothtoprovide
asufficientadhesion,whilefortheHCletchedstentsthesameproblemasdescribed
abovemayoccure.Theoxalicacidcombinationetchedstentsshowthebestadhe-
sionpropertiesbeingsimilartothatofgritblastedstents.

310

Figure8.13:Adhesionofrapamycinondifferentstentsurfaces.

KinetiRelease8.5cs

AsdescribedintheMaterialsandMethodschapter(p.37)thestrutisquickly
coveredbytissuesothatthemainreleasetakesplacethroughtissue.Itisassumed
that,whencoveredwithtissue,theremovedRapamycinwithinthedepotsissub-
stitutedbytissuefluid,whichfurtherassistsinsolvingRapamycin.Tissuefluidis
similarincompositiontobloodplasma(fromwhichitorigins),butunlikeblood
itcontainsfrom2.5to6%lipids[125].AsaconsequencethelipophileRapamycin
shoulddissolvetosomeextent.Thesurroundingtissue/cellsinthiscaseactasa
semipermeablemembranesothatthereleasefromthedepotsisgovernedbydiffu-
sion/permeationprocesses.Theconcentrationofrapamycinwithinthecellsaswell
astheRapamycinevacuationprocessesarenotknownsotheexactreleasekinetics
canonlybeevaluatedbyinvivovasculartissueexaminations.Howeverastatement
aboutthegeneraldepoteffectaswellasanevaluationoftherapamycinlossduring
thefirststatewhenthestentisindirectcontactwiththebloodflowcanbeobtained
byinvitroexperimentsasdescribedbelow.

Forallthreecombinationetchingmethodsdrugreleaseexperimentswerecarried
outinordertoevaluatethestorageeffectofthesurfaces.Forthispurposethree
stentsofeachtypewerecoatedwith1%ethanolicRapamycinsolutionandplaced
inplastictubescontainingRinger´ssolution(comp.p.39).

Itwasfoundthatallthreecombinationetchingsyieldeddifferentreleasekinetics
(fig.8.14).Thereleasekineticsweredeceleratedforallstents,comparedtogrit

410

blastedstentswhichreleased90%withinthefirstweek.Thetotalamountsofre-
leasedRapamycinwerebetween290and370µg/stentcorrespondingto1.9to2.4
µg/mm2.Interestinglythephosphoricacidetchingwasfoundtobethemosteffi-
cientintermsofdeceleration.15%ofthedrugwasstoredforaslongasthreeweeks.
Fig.8.15showsthatonallthreephosphoricacidetchedstentsmorethan30%(on
average40%)ofthedrugstillremainedafterthefirstweek.Thehydrochloricacid
etchedstentsshowedsligthlyfasterreleasewithapprox.25%releasedwithinthe
firstweek(fig.8.16).Surprisinglytheoxalicacidstentsrevealedarelativelyquick
release.Theoxalicacidetchingswereexpectedtobeasefficientasormoreeffi-
cientthanthephosphoricacidetchedstentsduetothehighersurfaceroughnessin
thesubmicrometerrange.However,thereleasewasfoundtobeconsiderablyfaster
thanthereleaseoftheotherstentswithonly15%ofthedrugremainingafterthe
firstweek.Becauseoftheunexpectedresulttheexperimentwiththeoxalicacid
etchedstentswasrepeated,yieldingasimilarresult,whichmakesanexperimental
deficiencyunlikely.Areasonfortherelativelyquickreleaseoftheoxalicacidstents
couldnotbefound.

Figure8.14:Drugreleaseofdifferentsurfacetreatedstents,averagevaluesofthreestents
ach.e

510

610

egurFi

Fiegur5:1.8

.86:1gDru

reasele

of

ricphoosph

daci

cmbinationo

hing.cte

Drugreleaseofhydrochloricacidcombinationetching.

Fiegur.87:1gDru

easeler

of

oxalic

idac

inationombc

ng.etchi

710

9SummaryandOutlook

9.1yrSumma

Inthisworkdifferentelectrochemicaletchingmethodsforstainlesssteelsurfaces
havebeeninvestigatedinrespecttotheirapplicabilityassurfacemodificationsfor
drugcoatings.Themodificationtechniquesareillustratedon316Lcoronarystents.
Thebasicaimwastoprovideasurfacewithsufficientadhesionofthedrug.Itwas
foundthat,beyondthis,itispossibletocreatemicrodepotsonthesurfaceswhich
havethecapacityofstoringacertainamountofsubstance.Throughthisthepossi-
bilityisgiventodeceleratereleaseofadrug.Incontrasttoothermicrostructuring
methodswherepredefinedstructuresareimposedontothesurface,theapproach
wastousetheintrinsicmaterialstructuresinordertocreaterandomlydistributed
microstructures.Thestructuresareworkedoutofthematerialbyelectrochemi-
caletching.Twodifferentetchingprincipleswereinvestigatedinthiswork.Oneis
basedontheselectiveremovalofcrystalplaneswithinthegrainbyhydrochloricacid
etching.Thismethodproducesanincreasedroughnesswithoutcreatingmicrode-
pots.Theotheronebasesontheselectivematerialremovalatgrainboundariesby
nitricacidetching.Theseetchingscreateagridofmicrofurrowsofdifferentdepth,
dependingontheetchingstrength.Bothmicrostructureswerefoundtobenotsuit-
ablefortheuseonstentsassingletreatments.Howeveritwasfoundthatpromising
microstructurescanbegeneratedwhencombiningnitricacidetchingswithasec-
ondetchingstep.Forthesecondstepofthecombinationetchingsthreeacidswere
examined:phosphoricacid,hydrochloricacidandoxalicacid.

9.1.1MicrostructuringwithHydrochloricAcid

Etchingswereperformedusingcurrentdensitiesfrom0.36to2.86mA/mm2.In
somecasesatcurrentdensitiesbetween0.72and1.08mA/mm2evenlyfacetedsur-
facescouldbegeneratedwithsurfaceroughnessesbetween0.22and0.71(Ra)and
between0.87and3.29(Rz),respectively.Thesesurfaces,however,didonlyoccuron

108

fewspecimens.Itwasfoundthatthesuccessrateforgeneratingthemicrostructures
couldbeconsiderablyimprovedwhenthestentwasnotetchedcrimpedonapin,
butholddirectlyintotheetchingsolutionbyaplatinumwire.Thisphenomenon
isattributedtoablackoxidelayergeneratedbytheetchingprocess.Thislayer
probablyisolatesthestentfromthepinandleadstoanadulterationoftheactual
currentdensitybythepin.However,withouttheuseofapinstillnosufficiently
reproducibleresultscouldbeobtained.Inordertoassessthereasonsforthediffi-
cultiesinreproducingthestructures,theformationprocesswasinvestigated,based
onobservationsoftheinitialstageoftheetchingprocess.Itissuggestedthatthe
formationofthestructureisbasedonaformofatypicalpittingcorrosionthatcan
developinstirredetchingsolutionwhenahighamountofinitiationpointsispresent.

Adhesiontestsofhydrochloricacidetchedstentsrevealeddeterioratedadhesion
propertiesincomparisonwithgritblastedandalsoincomparisonwithelectropol-
ishedstents.Thisunexpectedresultisexplainedbythefacetedmicrostructures
probablyactingasbreakinitiatorstowardthedruglayer.Followingtheseresults
theetchingprocessisregardedasunsuitableforauseonstents.Neverthelessa
useinotherapplicationareasisimaginable,providedthereproductivitycanbe
improvedconsiderably.Apossibleusemightbeapplications,whereanincreaseof
surfaceroughnessisrequiredandgritblastingorotherstraininducingtechniques
cannotbeused.

9.1.2MicrostructuringwithNitricAcid

Theeffectofelectrochemicaletchingwithnitricacidon316Lwasexaminedat
currentdensitiesbetween0.1and9.8mA/mm2.Acurrentvoltagediagramwas
recordedandtheselectivitiesatdifferentcurrentdensitiesweredeterminedoncross
sectionalmicrographsofetchedspecimens.Itwasfoundthattheselectivityofthe
etchingprocessdecreasesgraduallywithincreasingcurrentdensities,withouthav-
ingadistinctmaximumwithinthemeasuredrange.Basedonthisobservationitis
assumedthattheselectivityoftheprocessmayberatherdominatedbydifferences
inmolecularstructures(higherdegreeofdisorder,voidsetc.),thanbydifferences
inchemicalcomposition.Supportingthisassumption,observationonetchedcross
sectionsrevealedthat,besidesgrainboundaries,insomecasesalsotwinboundaries
ordislocations,respectively,wereaffectedbytheattacks.

910

Theeffectonthesurfacetopographywasinvestigatedusing1.3mA/mm2atetching
timesbetween2and6min.Thegeneratedsurfacesandcrosssectionalmicrographs
ofthespecimenswereinvestigatedunderSEM.Theresultingstructuresrangedfrom
small,crevicelikefurrowsaroundthegrainsof2to4µmdepthat2mintovery
rough(Rz=15.8µm),granulatedsurfaceresultingfromdetachedgrainsat6min
etchingtime.Thesurfacescannotbeusedforstentsdirectly.Theslighteretched
surfacesexhibitonlysmalldepotvolumes.Forthestrongeretchedsurfacesarisk
ofparticledetachmentduringimplantationispresentwhichcanleadtoclinical
problems.However,thenitricacidetchingscanbeusedasabasisforcombination
etchingsasdescribedinthefollowingsection.

9.1.3CombinationEtchings

Threetypesofacidswereexaminedassecondetchingstepsforacombinationetch-
ingbasedonnitricacidetchings,namelyphosphoricacid,hydrochloricacidand
acid.xalico

Phosphoricacidturnedouttoyieldsuitabledepotswhencombinedwithnitricacid
etchingof1.31mA/mm2for240sec.Thelargestdepotswereobtainedatphos-
2phorictheoreticalacidstoetcrahingegvofolu2.0meof2mA/mm.8∙10−3appliedmm3for/mm1802.sec.Softwareanalysisyieldeda

Hydrochloricacidwascombinedwithnitricacidetchingsat1.3mA/mm2for180
sec.Thelargestdepotswithatheoreticalstoragevolumeof3.1∙10−3mm3/mm2
wereobtainedatacurrentof1.9mA/mm2appliedfor30sec.Incontrasttothe
otheretchingsthisetchingstephadtobecarriedoutwithoutapinwiththestent
beingetchedonaplatinumwiretoimprovethereproductivity.

Oxalicacidetchingswerecombinedwith1.3mA/mm2for180secnitricacidetching.
2Beststorageresultvsolumewereof2.7found∙10for−3anmmetc3hing/mmo2f.2.0ThemoA/xamlicmfoacidr1etc20hingsecwwithasathfoundeorettoicabel
betterreproduciblethanthehydrochloricacidetching.

011

9.1.4Performance

Thecorrosionbehaviorofthesurfacesaftertreatmentwiththedifferentacidswas
thatexaminednitricbyacidelectretcoched,hemicalphosphoaricnalysiscompbierfonartimedoninetcRhedingersandsohydrlutoiocn.hloricItwcoasmfoundbina-
tionetchedstentsurfacesshowverysimilarbehavioraselectropolishedandgrit
blastedreferencestents.Foroxalicacidcombinationetchedstentsanextended
passivityareawasrecorded,indicatingsuperiorpassivationbehavior.Infurther
ainvpoestigatentialtioonsft15he0mlongVabtermoveiontherelbreeaseakintRhrougingherspostenolutitial.onwItaswexaasmfoinedundthatapplyitheng
merenitricacidetchedstentshadthehighestreleaseofnickelions.Thehydrochlo-
etcricheaciddstencomts,binatwhereaionsetcthehedstreleaseentsofhadallaotheslighrsttlyenlotswwererereleabelosewthanormatherginalnitricabaovcide
thedetectionlimit.Itissuggestedthatthisfindingscanratherbeattributedto
thedifferentsurfaceareasaswellastoaninfluenceofpossibleacidremainswithin
crevicesthantochemicaldifferencesofthepassivitylayer.Resultsofaugerelectron
spectroscopywhichrevealedalmostidenticalcompositionofthepassivitylayerssup-
portthisassumption.

Inordertoassessbiocompatibilityaspects,cellseedingexperimentswereperformed
withthecombinationetchedsurfaces.Cellseedingexperimentswithfibroblastcell
linesrevealedgoodproliferationfortheoxalicacidcombinationetched,thehy-
drochloriccombinationetchedandgritblastedreferencesurfaces.Phosphoricacid
combinationetchedsurfacesshoweddecreasedproliferationmostprobablyasaresult
ofthesmoothersurfaceofthemicrostructures.Proliferationofprimaryendothelial
cellsorsmoothmusclecellscouldnotbeaccomplishedonanysurface.

Fortheoxalicandphosphoriccombinationetchingstheradialstrengthandthefa-
tiguepropertiesweretested.Thefatiguetestsgavenoevidencesformaterialfatigue
foranyofthetestedstents.Theradialstrengthtestsshowedupto30%decreased
collapsepressuresforalletchedstents.Thisresultwasexpectedsincethetested
stentdesignwasnotadjustedfortheetchingprocedure.Formedicaluseofthe
stentsaspeciallymodifieddesignwouldberequiredinordertocompensatethe
materiallossandtheslightnotcheffectcausedbytheetchingprocess.

Oxalicacidandphosphoricacidcombinationetchedstentswerecoatedwith1%
Rapamycinsolutionandtheadhesionpropertiesinanartificialbloodcircuitaswell

111

asthereleasekineticsweretested.Itwasshownthattheadhesionpropertiesof
oxalicacidetchedstentsarecomparabletothatofgritblastedreferencestentwhile
thephosphoricacidetchedstentshadslightlydeterioratedproperties.

Alletchedstentsshowedcharacteristicdeceleratedreleasecurveswiththephospho-
ricacidetchedstentsyieldingtheslowestrelease.Withthisetchingmethodmore
than30%ofthecoatingstillremainsafterthefirstweek.Forgritblastedstents,
incomparison,15%ofthedrugremainedafteroneweek.Oxalicacidetchedstents
showedareleasewhichwasonlyslightlysuperiorthanthatofgritblastedstents
withapprox.20%remainingafteroneweek.

nlusioConc.59.1Twomethodsofacombinedetchingmethod,phosphoriccombinationetchingand
oxaliccombinationetchingwerefoundtobesuitabletogenerateevenlydistributed
microdepotsonstainlesssteelstentsurfaceswithwhichdeceleratedreleasekinetics
canbeobtained.Bothwerefoundtobeequalorsuperiorintermsofcorrosion
propertiescomparedtogritblastedstents.Thedrugreleasecouldbesignificantly
reduced.

oOutlo9.2k

Theexperimentsperformedinthisworksuggestthat,withcertaintypesofcombined
etchings,microstructuredstentsurfacescanbegeneratedwithwhichthereleaseki-
neticscanbeconsiderablydecelerated.Thesurfacesmaybesuitablefortheuseof
drugcoatingswithoutapolymer.Chemical,biologicalandmechanicaltestsaswell
asinvitroexperimentsevaluatingthecoatingbehaviorwerecarriedoutinorder
toevaluatethepropertiesincomparisontoconventionalstents.Invivotestsare
currentlyplanedinordertovalidatetheclinicalapplicability.

Iftheinvivoresultswillbepositivethemicrostructurescouldbefurtheroptimized
inordertoyieldanoptimumreleaseandanoptimumdistributionofthedrug.This
optimizationcouldberenderedbyamodulationoftheetchingparametersinterms
ofthereleasekinetics.Thedistribution,ontheotherhandcouldbeadjustedto
acertaindegreebyslightvariationsintheheattreatmentofthestents,i.e.by

211

generatinguniformgrainsofacertainsize.

Anaspectthatmaybesubjectoffutureresearchistheadaptionoftheinner(lu-
minal)sideofthestentinordertoyieldatargetedpromotionofcellingrowth.The
withinnerthesideblodoodesnotstream.contaHoinwevdrugs,er,safterincetheininithetialinitialstagetstaghiseitsideisisinaswdirectellconquicktactly
cocellveredingrowithwth.Atissue,futsourethastentittiscouprldonereceivforaetwomicrostructudifferentsringurfaspceecifitreacatllymentaits,loaredffirstor
onethatcomprisesamicrostructuringoftheinnerside,followedbyasecondstep
forthecreationofmicrodepotsfordrugreleaseonthethreeoutersides.

311

dixenApp10

10.1DeterminationofGrainSize

Stentsandrawmaterialwereelectricallyconnectedeachwithathinwireandem-
beddeduprightinepoxyresin(Specifix-40,Struehrs,Willich,Germany)inaway
thatthewiressticksoutoftheresin.Afterhardeningthespecimenswerepolished
withdiamondsuspensiondownto1µparticlesize.Thespecimenwasthenhold
into40%nitricacid.ThewireswereconnectedwiththepositivpoleofaDCTrans-
formatorandacathodewasimmersedintotheacid.Etchingwascarriedoutat0.9
Vforapprox.2min(requiredetchingtimemaydifferfromstenttostentandhasto
beadjustedempirically).Forthevisualizationofthedislocationsasecondetching
stepwasperformedusingoxalicacidat1.5Vforapprox.2min.Theimageswere
takenusingbrightfieldmodeofalightmicroscope(Axiovert25,CarlZeissGmbH,
Jena,Germany)at200Xmagnification.Forthegrainsizedetermination50lines
of100µmweredrawnrandomlyondifferentcrosssectionalstrutimagesofdifferent
stentsandthenumberofintersectedgrainboundarieswascounted.Accordingly
5000µmdevidedbythesumcountedforallstrutsyieldstheaverageintersectional
grainsize.Itmaybeworthmentioningthatthisvalueisdifferentfromtheactual
threedimensionalaveragesizeofthegrains.Howeverforthesurfacestructuringthe
intersectionalvalueisthemostrelevant.

10.2CalculationofPolygonLine

Thestartingpointwasdefinedasthepointoftheoutlinewiththelargestx-
coordinate.(Theresultisindependentofthechoiceofthestartingpointaslong
asitisanextreme.Sothesmallestx-valueorthey-valuecanalsobeusedand
will,independentoftheorientationoftheimage,yieldthesamepolygon).Firstly
astraightlineiscalculatedrunningthroughthestartingpointandasecondpoint
ofthecontourline,withoutsectioningthestrutarea.(Therearetwopossibilities
bothofwhichleadtothesameresult.)Asecondstraightlineisdrawnfromthis
secondpointtothenexttouchingpointonthecontourline,withoutsectioningthe

411

strutarea.Thisprocessisrepeateduntilthestartingpointisreachedagain.The
areasbetweenthepolygonlineandthestentaredefinedasdepotareas.

CulturesCell3.10

agentseR

•WST-measurement:CellproliferationreagentWST-1(Cat#1644807,ready
touse,RocheDiagnosticsGmbH,Mannheim,Germany)
•celltracking:CellTrackerCM-Dil(Cat#C-7001,MolecularProbes,Karlsruhe,
Germany),diluted1:1000
•celldetachmentagent:Trypsin/EDTA(Cat#L2143,BiochromAG,Berlin,
)yGerman•Phosphatebuffersaline(Cat#L1825,BiochromAG,Berlin,Germany)
•fixingofcells:3%glutardialdehyde(Cat#1.04239.0250,Merck,Darmstadt,
)yGerman

CultureMedia,Additions

•DMEM/phen:Dulbecco‘sModifiedEagleMediumhgwithphenolredandsta-
bleL-glutamine(Cat#FG0435,BiochromAG,Berlin,Germany)
•ECGM:EndothelialCellGrowthMedium(Cat#C-22010,PromoCell,Heidel-
berg,Germany)
•FCS:fetalcalfserum(HyClonevonPerbio,Cat#CH30160.3,Lot#CMK
0190,forUCSC-Medium)(Cat#S0115,Lot#364B,BiochromAG,Berlin,
Germany,forHUVEC-and3T3-Medium)
•Na-P:Na-Pyruvat(Cat#L0473,BiochromAG,Berlin,Germany)
•LAPh:L-Ascorbinsäure-2-phosphat(Cat#A-8960,SigmaAldrichCo.,Mu-
nich,Germany)
•SMix:SupplementMix(Cat#C-39215,PromoCell,Heidelberg,Germany)

511

•Pen/S:Penicillin/Streptomycin(Cat#A2213,BiochromAG,Berlin,Ger-
y)man

•PA:PartricinA(Cat#A2812,BiochromAG,Berlin,Germany)

Thefollowingcompositionsoftheculturemediawereused:

cellsDMEMECGMFCSSMixNa-PLAPhPen/SPA
USCSX/10%/1mM50µMXX
3T3HUVECX//X110%0%/X////XXXX

ests-TWST

10%vol.ofcellproliferationreagentWST-1(RocheDiagnosticsGmbH,Mannheim,
Germany)wasaddeddirectlyintothemedium.After3hoursthebehaviourofthese
monolayerswereanalysed.Controlwastheculturemediumand10%WST-1.The
photometricmeasurementofthecontroltookplaceina96-well-Plattenwith450nm
(referencewavelength620nm)onanELISAreader(Sunrise,TecanGmbH,Crail-
sheim,Germany).Theabsorptionofthecontrolwassubtractedfromthemeasured
samples.Thephotometricmeasurementsofthesampleswereperformedthreetimes
andwiththesameparametersasthecontrol.Anaverageopticaldensity(OD)with
standarddeviationwascalculated.ThedifferenceinODpercentagefromthesam-
pleswascomparedtothecontrolwell,whichwastakentobe100%.WST-1,isthe
standardassayusedincytotoxicitytestandisrecommendedbyENISO10993-5.

611

10.4CompositionofRingersSolution

componentmassconcentration
KNaClCl0,46,800
CaCl2xH2O0,27
MgSO4xH2O0,20
NaH2PO4xH2O0,14
02,2NaHCO3α-D-Glucose(C6H12O6)1,00

10.5ApproximationofStrain

Duringdilatationthestentisexpandedto2.2timestheoriginaldiameter(from
1.8to3.8mmouterdiameter).Thisexpansionleadstoanaverageangleincrease
from65◦toapprox.140◦oneachbending.Giventhatthecenterpointoftheangle
movementisinthemiddleofthestentstrutthisexpansionleadstoanapprox.
prolongationoftheinnercontourlineof110µm.Inordertoestimatethestrainfor
theaffectedpartoftheinnercontourlinethelengthoftheaffectedpartmustbe
known.Assumingthatthestrainisconstantovertheevenlyroundedpartwhich
hasalengthofapprox.1.2mmandalmostzeroatthestraightpartthestraincan
beestimatedtosome9%.Thiscalculationcanonlybearoughestimation,however
itseemsreasonablethatthestrainwillbenotconsiderablyhigherthan6%since
thisisthefractureelongationof316Landdilatatedelectropolishedstentsdonot
showanycrevicesafterdilatation.

711

yographbliBi

[1]R.G.Richards.Surfacestocontrolimplanttissueadhesionforosteosynthesis:
Invitroandinvivoevaluations.19thEuropeanConferenceonBiomaterials,
Sorrento,2005.
[2]K.MustafaandA.Wennerbergetal.Determiningoptimalsurfaceroughness
oftio(2)blastedtitaniumimplantmaterialforattachment,proliferationand
differentiationofcellsderivedfromhumanmandibularalveolarbone.Clinical
OralImplantsResearch,12(5):515–25,2001.
[3]J.LincksandBD.Boyanetal.Responseofmg63osteoblast-likecellstotita-
niumandtitaniumalloyisdependentonsurfaceroughnessandcomposition.
Biomaterials,19(23):2219–32,1998.
[4]K.MustafaandJ.Wroblewskietal.Effectsoftitaniumsurfacesblastedwith
tio2particlesontheinitialattachmentofcellsderivedfromhumanmandibular
bone.ascanningelectronmicroscopicandhistomorphometricanalysis.Clini-
calOralImplantsResearch,11(2):116–28,2000.
[5]M.S.SaderandA.Balduinoetal.Effectofthreedistincttreatmentsof
titaniumsurfaceonosteoblastattachment,proliferation,anddifferentiation.
ClinicalOralImplantsResearch,16(6):667–675,2005.
[6]D.BuserandR.K.Schenketal.Influenceofsurfacecharacteristicsonbone
integrationoftitaniumimplants.ahistomorphometricstudyinminiaturepigs.
JournalofBiomedicalMaterialResearch,25:889–902,1991.
[7]B.W.Darvell,N.Samman,W.K.Luk,R.K.F.Clark,andH.Tideman.
Contaminationoftitaniumcastingsbyaluminiumoxideblasting.Journalof
Dentistry,23(5):319–322,1995.TY-JOUR.
[8]U.Gbureck,A.Masten,J.Probst,andR.Thull.Tribochemicalstructuring
andcoatingofimplantmetalsurfaceswithtitaniumoxideandhydroxyapatite

811

layers.MaterialsScienceandEngineering:C,23(3):461–465,2003.TY-
.JOUR[9]M.A.M.SilvaandA.E.Martinellietal.Surfacemodificationoftiimplantsby
plasmaoxidationinhollowcathodedischarge.SurfaceandCoatingsTechnol-
ogy,200(8):2618–2626,2006.
[10]O.ZingerandG.Zhaoetal.Differentialregulationofosteoblastsbysubstrate
microstructuralfeatures.Biomaterials,26(14):1837–1847,2005.
[11]S.Ban,Y.Iwaya,H.Kono,andH.Sato.Surfacemodificationoftitaniumby
etchinginconcentratedsulfuricacid.DentalMaterials,inpress,2006.
[12]CagiannosC.andO.R.Abul-Khoudoudetal.Rapamycin-coatedexpanded
polytetrafluoroethylenebypassgraftsexhibitdecreasedanastomoticneointi-
malhyperplasiainaporcinemodel.JournalofVascularSurgery,42(5):980–
.05208,98[13]KimD.D.andM.M.Takenoetal.Glowdischargeplasmadeposition(gdpd)
techniqueforthelocalcontrolleddeliveryofhirudinfrombiomaterials.Phar-
maceuticalResearch,15(5):783–786,1998.
[14]R.KassingandR.Käsmaier.Lithographiedernächstengeneration.Physikalis-
cheBlätter,pages31–36,2000.
[15]P.R.Choudhury.HandbookofMicrolithography,MicromachiningandMicro-
fabrication,volume1,2.OpticalEngineeringPress,1997.
[16]R.-M.Beck.UntersuchungvonOberflächenbeschichtungenbeiGefässstützen
zurReduktionderRestenoserate.PhDthesis,UniTübingen,2001.
[17]P.KuukasjarviandA.Malmivaaraetal.Overviewofsystematicreviewson
invasivetreatmentofstablecoronaryarterydisease.InternationalJournalof
TechnologyAssessmentinHealthCare,22(2):219–234,2006.
[18]A.Machraoui,P.Grewe,andA.Fischer.Koronarstenting,page49.Steinkopff,
Darmstadt,2001.
[19]S.SobottaandF.Hammersen.Histologie,FarbatlasderMikroskopischen
Anatomie,page126.UrbanundSchwarzenberg,1975.
[20]M.Gawaz.DasBlutplättchen,page89.GeorgThiemeVerlag,1999.

911

[21]R.RossandJ.A.Glomset.Thepathogenesisofatherosclerosis,partiandii.
NewEnglandJournalofMedicine,295:369–377,420–425,1976.
[22]A.C.Newby.Anoverviewofthevascularresponsetoinjury:atributetothe
laterussellross.Toxicologyletters,112-113:519–529,2000.
[23]H.ItabeandM.Uedaetal.Measurementofoxidizedldlpresentinhuman
plasmaandatheroscleroticlesions.InternationalCongressSeries,1262:87–90,
.0420[24]M.Gawaz.DasBlutplättchen,page94.GeorgThiemeVerlag,1999.
[25]D.R.Jr.Holmes.Stateoftheartincoronaryintervention.TheAmerican
JournalofCardiology,91(3,Supplement1):50–53,2003.TY-JOUR.
[26]M.Gawaz.DasBlutplättchen,page104.GeorgThiemeVerlag,1999.
[27]P.H.Grewe,T.Deneke,A.Machraoui,J.Barmeyer,andK.-M.Muller.Acute
andchronictissueresponsetocoronarystentimplantation:pathologicfindings
inhumanspecimen.JournaloftheAmericanCollegeofCardiology,35(1):162,
2000.TY-JOUR.
[28]S.KinlayandA.U.Coskunetal.Endothelialshearstressidentifiedinvivo
withinthestentisrelatedtoin-stentrestenosisandremodelingofstentedcoro-
naryarteries.JournaloftheAmericanCollegeofCardiology,39(Supplement
1):5,2002.TY-JOUR.
[29]N.KipshidzeandD.Georgeetal.Roleoftheendotheliuminmodulating
neointimalformation:Vasculoprotectiveapproachestoattenuaterestenosis
afterpercutaneouscoronaryinterventions.JournaloftheAmericanCollegeof
Cardiology,44(4):733–739,2004.
[30]P.W.Serruys.HandbookofDrug-ElutingStents,page15.TaylorandFrancis,
.0520[31]P.H.Grewe,T.Deneke,A.Machraoui,J.Barmeyer,andK.-M.Muller.Acute
andchronictissueresponsetocoronarystentimplantation:pathologicfindings
inhumanspecimen.JournaloftheAmericanCollegeofCardiology,35(1):159–
160,2000.TY-JOUR.
[32]J.C.Palmaz,S.Bailey,D.Marton,andE.Sprague.Influenceofstentdesign
andmaterialcompositiononprocedureoutcome.JournalofVascularSurgery,
36(5):1031–1039,2002.TY-JOUR.

012

[33]C.Fuss,E.A.Sprague,S.R.Bailey,andPalmazJ.C.Surfacemicrogrooves
improveendothelializationrateinvitroandinvivo.AmericanJournalof
Cardiology,37(2SupplA):70A,2001.

[34]S.O.MarxandA.R.Marks.Benchtobedsidethedevelopmentofrapamycin
anditsapplicationofstentrestenosis.Circulation,104:852–855,2001.

[35]K.McKeage,D.Murdoch,andK.L.Goa.Thesirolimus-elutingstentareview
ofitsuseinthetreatmentofcoronaryarterydisease.AmericanJournalof
CardiovascularDrugs,31:211–230,2003.

[36]B.C.Berk,J.B.Gordon,D.R.Holmes,andR.W.Alexander.Pharmaco-
logicrolesofheparinandglucocorticoidstopreventrestenosisaftercoronary
angioplasty.JournaloftheAmericanCollegeofCardiology,17:111B–117B,
.9119

[37]D.I.Axel,W.Kunert,andC.Goggelmannetal.Paclitaxelinhibitsarterial
smoothmusclecellproliferationandmigrationinvitroandinvivousinglocal
drugdelivery.Circulation,96:636–645,1997.

[38]E.K.RowinskyandR.C.Donehower.Paclitaxel(taxol).NewEnglandJournal
ofMedicine,332:1004–1014,1995.

[39]A.M.Salam,J.AlSuwaidi,andD.R.Holmes.Drug-elutingcoronarystents.
CurrentProblemsinCardiology,31:8–119,2006.

[40]R.FattoriandT.Piva.Drug-elutingstentsinvascularintervention.The
Lancet,361(9353):247–249,2003.TY-JOUR.

[41]M.-C.Morice,P.Serruys,C.Constantini,andK.Wuelfert.Three-yearfollow-
upoftheravelstudy:Arandomizedstudywiththesirolimus-elutingbxveloc-
ity(tm)stentinthetreatmentofpatientswithdenovonativecoronaryartery
lesions.JournaloftheAmericanCollegeofCardiology,43(5,Supplement
1):A87–A88,2004.TY-JOUR.

[42]R.Wombacher.PolymerealsBeschichtungs-materialienfürStents.PhDthe-
sis,Philipps-UniversitätMarburg,2003.

[43]WesselyR.,KastratiA.,andSchoemigA.Laterestenosisinpatientsreceiving
apolymercoatedsirolimus-elutingstent.AnnInternMed,inpress,2005.

211

[44]McFaddenEP.,StabileE.C.,RegarE.,CheneauE.,andT.Kinnairdetal.Late
thrombosisindrug-elutingcoronarystentsafterdiscontinuationofantiplatelet
therapy.Lancet,364:1519–1521,2004.
[45]R.VirmaniandG.Guagliumietal.Localizedhypersensitivityandlatecoro-
narythrombosissecondarytoasirolimus-elutingstent:shouldwebecautious?
Circulation,109:701–706,2004.
[46]R.Wessely,J.Hausleiter,andC.Michaelis.Inhibitionofneointimaformation
byanoveldrug-elutingstentsystemthatallowsfordose-adjustable,multiple,
andon-sitestentcoating.Arteriosclerosis,Thrombosis,andVascularBiology,
25(4):748–753,2005.
[47]A.DibraandA.Kastratietal.Influenceofstentsurfacetopography
ontheoutcomesofpatientsundergoingcoronarystenting:Arandomized
double-blindcontrolledtrial.CatheterizationandCardiovascularInterven-
tions,65:374–380,2005.
[48]BrunetteD.M.Theeffectofsurfacetopographyoncellmigrationandadhesion.
SurfaceCharacterizationofBiomaterials,1987.
[49]J.E.SousaandM.A.Costaetal.Lackofneointimalproliferationafter
implantationofsirolimus-coatedstentsinhumancoronaryarteries.aquan-
titativecoronaryangiographyandthree-dimensionalintravascularultrasound
study.Circulation,103:192,2001.
[50]P.W.SerruysandG.Sianosetal.Theeffectofvariabledoseandrelease
kineticsonneointimalhyperplasiausinganovelpaclitaxel-elutingstentplat-
form:Thepaclitaxelin-stentcontrolledelutionstudy(pisces).Journalofthe
AmericanCollegeofCardiology,46(2):253–260,2005.
[51]H.WienekeandA.Schmermundetal.Therapeuticpotentialofactivestent
coating.ExpertOpiniononInvestigationalDrugs,12(5):771–779,2003.
[52]S.H.DudaandT.C.Poerneretal.Drug-elutingstents:Potentialapplications
forperipheralarterialocclusivedisease.JournalofVascularandInterventional
Radiology,14:293,2003.
[53]T.J.ParryandR.Brosiusetal.Drug-elutingstents:Sirolimusandpaclitaxel
differentiallyaffectculturedcellsandinjuredarteries.EuropeanJournalof
Pharmacology,524(1-3):19–29,2005.

212

[54]C.Momma,S.Nolte,B.N.Chichkov,F.v.Alvensleben,andA.Tunnermann.
Preciselaserablationwithultrashortpulses.AppliedSurfaceScience,109-
110:15–19,1997.TY-JOUR.
[55]S.H.DudaandT.C.Poerneretal.Drug-elutingstents:Potentialapplications
forperipheralarterialocclusivedisease.JournalofVascularandInterventional
Radiology,14:295,2003.
[56]B.L.vanderHoevenandM.M.Nunoetal.Drug-elutingstents:results,
promisesandproblems.InternationalJournalofCardiology,99(1):9–17,2004.
[57]J.Hausleiter,A.Kastrati,andR.Wesselyetal.Preventionofrestenosisbya
noveldrug-elutingstentsystemwithadose-adjustable,polymer-free,on-site
stentcoating.EuropeanHeartJournal,26:1475–1481,2005.
[58]ColomboA.,MosesJW.,MoriceMC.,andJ.Ludwigetal.Randomizedstudy
toevaluatesirolimus-elutingstentsimplantedatcoronarybifurcationlesions.
Circulation,109:1244–1249,2004.
[59]R.HaakandT.Smith.Surfacetreatmentofam355stainlesssteelforadhesive
bonding.InternationalJournalofAdhesionandAdhesives,3(1):15–23,1983.
JOUR.-TY[60]M.HaidopoulosandS.Turgeon.Surfacemodificationsof316stainlesssteelfor
theimprovementofitsinterfacepropertieswithrfgd-depositedfluorocarbon
coating.SurfaceandCoatingsTechnology,197(2-3):278–287,2005.
[61]E.J.Sutow.Theinfluenceofelectropolishingonthecorrosionresistanceof
316lstainlesssteel.JournalOfBiomedicalMaterialsResearch,14(5):587–595,
1980.TY-JOUR.
[62]J.S.Noh,N.J.Laycock,W.Gao,andD.B.Wells.Effectsofnitricacid
passivationonthepittingresistanceof316stainlesssteel.CorrosionScience,
42(12):2069–2084,2000.TY-JOUR.
[63]S.-J.LeeandJ.-J.Lai.Theeffectsofelectropolishing(ep)processparameters
oncorrosionresistanceof316lstainlesssteel.JournalofMaterialsProcessing
Technology,140(1-3):210,2003.TY-JOUR.
[64]M.A.Barbosa.Thepittingresistanceofaisi316stainlesssteelpassivatedin
dilutednitricacid.CorrosionScience,23:1293–1305,1983.

231

[65]SurtecGMBH.Elektrochemischegrundlagen.script,1997.
[66]H.Kaesche.KorrosionderMetalle,page117.UrbanundSchwarzenberg,
.9019[67]V.Kirchner.MikrostrukturierungvonEdelstahlmitultrakurzenSpan-
nungspulsen,pages10–11.FreieUniversitätBerlin,2001.
[68]W.Schmickler.Modelsfortheinterfacebetweenametalandelectrolyteso-
lution.InJ.LipkowskyandP.N.Ross,editors,StructureofElectrifiedInter-
faces,pages201–238.VCHPubl.Inc.,NewYork,1993.
[69]K.B.OldhamandJ.C.Myland.FundamentalsofElectrochemicalScience.
AcademicPressIncorporated,1994.
[70]J.A.McGeough.PrinciplesofElectrochemicalMachining,pages5–6.Chap-
manandHall,1974.
[71]G.VanderVoort.AppliedMetallography,pages27,28.VanNostrandReinhold
Company,1989.
[72]G.VanderVoort.AppliedMetallography,pages21–39.VanNostrandReinhold
Company,1989.
[73]M.Prazak,V.Cihal,andM.Holinka.Ueberdiedifferenzierungderstruktur-
phasenbeimmetallographischenÄtzen.CollectionofCzechoslovakChemical
Communications,24:9–15,1959.
[74]N.SridharandP.Crook.Designinganalloytoresistcorrosionandwear.
Journalofmetals,41(3):22–26,1989.
[75]KimD.H.andH.H.Leeetal.Chemicaldepthprofileofpassiveoxideon
stainlesssteel.AppliedPhysicsLetters,85(26):6427–6429,2004.
[76]S.HauptandH.-H.Strehblow.Acombinedsurfaceanalyticalandelectroche-
micalstudyoftheformationofpassivelayersonfe/cralloysin0.5mh2so4.
CorrosionScience,37(1):43–54,1995.
[77]J.Sedriks.CorrosionofStainlessSteels.JohnWileyandSons,NewYork,2
edition,1996.
[78]V.Vignal,J.M.Olive,andD.Desjardins.Effectofmolybdenumonpassivity
ofstainlesssteelsinchloridemediausingexsitunearfieldmicroscopyobserva-
tions.CorrosionScience,41(5):869–884,1999.TY-JOUR.

412

[79]J.M.Bastidas,C.L.Torres,E.Cano,andJ.L.Polo.Influenceofmolybdenum
onpassivationofpolarisedstainlesssteelsinachlorideenvironment.Corrosion
Science,44(3):625–633,2002.TY-JOUR.
[80]L.BeaunierandFSalihi.Grainboundariescorrosionofni-moalloysbyelec-
trochemicaletching.JournaldePhysique,C1(1):429–434,1990.
[81]K.SugimotoandY.Sawada.Theroleofmolybdenumadditionstoaustenitic
stainlesssteelsintheinhibitionofpittinginacidchloridesolutions.Corrosion
Science,17(5):425–445,1977.TY-JOUR.
[82]Y.C.Lu,M.Ives,andC.R.B.Clayton.Synergismofalloyingelementsand
pittingcorrosionresistanceofstainlesssteels.CorrosionScience,35(1-4):89–
3.991,96[83]H.Ogawa,H.Omata,I.Itoh,andH.Okada.Corrosion,34:53,1978.
[84]D.LinandT.Chang.Influenceofsicontentontheintergranularcorrosionof
sus309lstainlesssteels.MaterialsScienceandEngineeringA,359(1-2):396–
401,2003.TY-JOUR.
[85]V.Kirchner.MikrostrukturierungvonEdelstahlmitultrakurzenSpan-
nungspulsen,pages55–56.FreieUniversitätBerlin,2001.
[86]G.VanderVoort.AppliedMetallography,page26.VanNostrandReinhold
Company,1989.
[87]H.Kaesche.KorrosionderMetalle,page201.UrbanundSchwarzenberg,
.9019[88]L.Gmelin.HandbuchderanorganischenChemie,chapterCr-A2,pages458–
459.Springer,1978.
[89]C.-C.ShihandC.-M.Shihetal.Effectofsurfaceoxidepropertiesoncorrosion
resistanceof316lstainlesssteelforbiomedicalapplications.CorrosionScience,
46(2):427–441,2004.TY-JOUR.
[90]I.OlefjordandL.Wegrelius.Surfaceanalysisofpassivestate.Corrosion
Science,31:89–98,1990.TY-JOUR.
[91]P.MarcusandJ.M.Grimal.Theanodicdissolutionandpassivationofni—
cr—fealloysstudiedbyesca.CorrosionScience,33(5):805–814,1992.TY-
.JOUR

251

[92]R.GoetzandD.Landolt.Theinfluenceofchromiumcontentandpotentialon
thesurfacecompositionoffe—cr—moalloysstudiedbyaes.Electrochimica
Acta,29(5):667–676,1984.TY-JOUR.
[93]R.J.Galvele,J.B.Lumsden,andR.W.Staehle.Electrochem.Soc.,125:1204,
.7819[94]G.OkamotoandT.Shibata.Passivityandthebreakdownofpassivityof
stainlesssteels.InCorrosionMonographSeries,pages646–677.Electrochem.
.7819c.,So[95]D.Lothongkum,S.Chaikittisilpa,andA.W.Lothongkumb.Xpsinvestigation
ofsurfacefilmsonhighcr-niferriticandausteniticstainlesssteels.Applied
SurfaceScience,218(1-4):203–210,2003.
[96]M.KanekoandH.S.Isaacs.Effectsofmolybdenumonthepittingofferritic-
andaustenitic-stainlesssteelsinbromideandchloridesolutions.Corrosion
Science,44(8):1825–1834,2002.TY-JOUR.
[97]A.J.Sedriks.Roleofsulphideinclusionsinpittingandcrevicecorrosion
ofstainlesssteels.InternationalMetalsReviews,28(5):295–307,1983.TY-
.JOUR[98]T.SuterandH.Boehni.Anewmicroelectrochemicalmethodtostudypit
initiationonstainlesssteels.ElectrochemicaActa,42(20-22):3275–3280,1997.
JOUR.-TY[99]P.Schmuki,A.Hildebrand,FriedrichS.,andS.Virtanen.Thecomposition
oftheboundaryregionofmnsinclusionsinstainlesssteelanditsrelevancein
triggeringpittingcorrosion.CorrosionScience,47:1239–1250,2005.
[100]H.J.DundasandA.P.Bond.Effectsofdeltaferriteandnitrogencontents
ontheresistanceofausteniticstainlesssteelstopittingcorrosion.NACE
Conference/75,159,1975.
[101]P.V.Shigolev.ElectrolyticandChemicalPolishingofMetalls.FreudPublish-
4.971ing,[102]ThompsonJ.G.Chemmetengg,614(6),1926.
[103]S.Fujimoto.Growthandpropertiesofcr-richthickandporousoxidefilms
ontype304stainlesssteelformedbysquarewavepotentialpulsepolarisation.
ElectrochimicaActa,47(4):548,2001.

612

[104]L.Gmelin.HandbuchderanorganischenChemie,chapterFe-A1,page316ff.
Springer,1978.
[105]L.Gmelin.HandbuchderanorganischenChemie,chapterNi-A2,page146.
Springer,1978.
[106]WDegner.ElektrochemischeMetallbearbeitung,pages20–21.VEBVerlag
Technik,1984.
[107]F.Christien,R.LeGall,andG.Saindrenan.Phosphorusgrainboundary
segregationinsteel17-4ph.ScriptaMaterialia,48(3):301–306,2003.TY-
.JOUR[108]H.Sahlaoui,H.Sidhom,andJ.Philibert.Predictionofchromiumdepleted-
zoneevolutionduringagingofni-cr-fealloys.ActaMaterialia,50(6):1383–
1392,2002.TY-JOUR.
[109]G.Furutani,N.Nakajima,T.Konishi,andM.Kodama.Stresscorrosion
crackingonirradiated316stainlesssteel.JournalofNuclearMaterials,288(2-
3):179–186,2001.TY-JOUR.
[110]M.Matula,L.Hyspecka,M.Svoboda,V.Vodarek,C.Dagbert,J.Galland,
Z.Stonawska,andL.Tuma.Intergranularcorrosionofaisi316lsteel.Materials
Characterization,46(2-3):203–210,2001.TY-JOUR.
[111]H.Sahlaoui,K.Makhlouf,H.Sidhom,andJ.Philibert.Effectsofageing
conditionsontheprecipitatesevolution,chromiumdepletionandintergran-
ularcorrosionsusceptibilityofaisi316l:experimentalandmodelingresults.
MaterialsScienceandEngineeringA,372(1-2):98–108,2004.TY-JOUR.
[112]S.Kannan,A.Balamurugan,andS.Rajeswari.Electrochemicalcharacteriza-
tionofhydroxyapatitecoatingsonhno3passivated316lssforimplantappli-
cations.ElectrochimicaActa,50:2065–2072,2005.
[113]E.B.Borghi,S.P.Ali,P.J.Morando,andM.A.Blesa.Cleaningofstainless
steelsurfacesandoxidedissolutionbymalonicandoxalicacids.Journalof
NuclearMaterials,229:122,1996.TY-JOUR.
[114]E.B.Borghi,S.P.Ali,P.J.Morando,andM.A.Blesa.Cleaningofstainless
steelsurfacesandoxidedissolutionbymalonicandoxalicacids.Journalof
NuclearMaterials,229:115,1996.TY-JOUR.

712

[115]H.Jiang,X.Chen,L.Hong,andM.Textor.Mineral-acid-freechemicalpol-
ishingsolutionsforferrousalloys.AppliedSurfaceScience,218(1-4):306–310,
.0320[116]I.Sekine,C.Okano,andM.Yuasa.Thecorrosionbehaviourofferriticstainless
steelinoxalicacidsolutions.CorrosionScience,30(4-5):351–366,1990.TY-
.JOUR

[117]E.B.Borghi,S.P.Ali,P.J.Morando,andM.A.Blesa.Cleaningofstainless
steelsurfacesandoxidedissolutionbymalonicandoxalicacids.Journalof
NuclearMaterials,229:121,1996.TY-JOUR.

[118]G.L.Kehl.ThePrinciplesofMetallographicLaboratoryPractice,page434.
McGraw-HillBookCompany,1949.
[119]H.Schumann.Metallographie.WileyVerlag,2004.

[120]T.R.Giles.Pretreatmentofvarioussubstratespriortoelectrocoating.Metal
Finishing,99(9):10–12,2001.
[121]S.-J.LeeandJ.-J.Lai.Theeffectsofelectropolishing(ep)processparameters
oncorrosionresistanceof316lstainlesssteel.JournalofMaterialsProcessing
Technology,140(1-3):206,2003.TY-JOUR.
[122]J.D.Bronzino.BiomedicalEngineeringHandbook,Ed.2,volume1,page539.
CRCPressandIEEEPress,2000.
[123]A.J.Carter,M.Aggarwal,andKopiaG.A.Long-termeffectsofpolymer-
based,slow-release,sirolimus-elutingstentsinaporcinecoronarymodel.Car-
diovascularResearch,63(4):622,2004.

[124]M.Renke-Gluszko.DrugReleaseofCoatedCoronaryStents.PhDthesis,TU
München,2006.

[125]P.Emschermann.Lymphe.InLexikonderBiologie,page304.HerderVerlag,
.9419

[126]R.Rettig,J.Kunze,andM.Stoeveretal.Electrochemicalbehaviorofsurface
etched316l.JournalofMaterialsScience,MaterialsinMedicine,inpress,
.0620

812

[127]R.Iijima,Y.Ikari,andE.Amiyaetal.Theimpactofmetallicallergyonstent
implantation.metalallergyandrecurrenceofin-stentrestenosis.International
JournalofCardiology,104(3):319–325,2005.
[128]R.Koester,D.Vieluf,andM.Kiehnetal.Nickelandmolybdenumcon-
tactallergiesinpatientswithcoronaryin-stentrestenosis.TheLancet,
356(9245):1895–1897,2000.
[129]M.Eblenkamp,J.Aigner,andJ.Hintermairetal.Umbilicalcordstromal
cells(ucsc).cellsfeaturingosteogenicdifferentiationpotential.DerOrthopäde,
33:1338–1345,2004.
[130]S.SilbernaglandA.Despopoulos.AtlasderPhysiologie.dtvTaschenbücher,
.0120[131]G.VanderVoort.Metallography,PrinciplesandPractice,page194.Materials
ScienceandEngineeringSeries.MacGraw-HillBookCompany,1984.
[132]V.Kirchner.MikrostrukturierungvonEdelstahlmitultrakurzenSpan-
nungspulsen,page57.FreieUniversitätBerlin,2001.
[133]M.Stoever,M.Renke-Gluszko,andT.Schratzenstalleretal.Microstructuring
ofstainlesssteelimplantsbyelectrochemicaletching.JournalofMaterials
Science,MaterialsinMedicine,inpress,2006.
[134]F.C.BellandD.E.Sonon.Improvedmetallographicetchingtechniquesfor
stainlesssteelandforstainlesssteeltocarbonsteelweldments.Metallography,
9(2):105–106,1976.TY-JOUR.
[135]J.A.McGeough.PrinciplesofElectrochemicalMachining,page93.Chapman
andHall,1974.
[136]K.VargaandP.Baradlaietal.Corrosionbehaviourofstainlesssteelsurfaces
formeduponchemicaldecontamination.ElectrochimicaActa,46(24,25):3783–
3790,2001.
[137]P.H.Grewe,T.Deneke,A.Machraoui,J.Barmeyer,andK.-M.Muller.
Rasterelektronenmikroskopischeanalysederreaktivenundreparativenverän-
derungenhumanerkoronararteriennachstentimplantation.Zeitschriftfür
Kardiologie,89(1):21–27,2000.
[138]T.Schratzenstaller.EntwicklungeinesTraumaReduzierendenStentsystems.
PhDthesis,TUMünchen,2006.

912