composites Steel foil reinforcement for high performance bearing strength in Thin ‐Ply Composites Part C: Open Access

(1)

ContentslistsavailableatScienceDirect

Composites Part C: Open Access

journalhomepage:www.elsevier.com/locate/jcomc

Steel foil reinforcement for high performance bearing strength in Thin ‐Ply composites ^☆

Benedikt Kötter

^a^,^∗

, Kohei Yamada

^b

, Johann Körbelin

^a

, Kazumasa Kawabe

^b

, Masaaki Nishikawa

^c

, Masaki Hojo

^c

, Bodo Fiedler

^a

aInstitute of Polymer and Composites, Hamburg University of Technology, Denickestraße 15, Hamburg 21073, Germany

bIndustrial Technology Center of Fukui Prefecture, 61-10, Kawaiwashizuka, Fukui 910-0102, Japan

cDepartment of Mechanical Engineering and Science, Kyoto University, C3, Kyoto Daigaku-Katsura, Nishikyo-ku, Kyoto 615-8540, Japan

a r t i c le i n f o

Keywords:

Fibre metal laminate (FML) Hybrid material Stress distribution Open hole test Digital Image Correlation

a b s t r a ct

ThisstudyinvestigatestheinfluenceoflocalhybridizationofThinandThick-PlyCFRPlaminatesontheopen-hole andbearingproperties.TheareaweightoftheCFRPunidirectionalprepregsusedis40gsminthecaseofThin-Ply layersand160gsminthecaseofThick-Plylayers.Thesteelusedisa1.4310stainless-steelfoilwiththesame layerthicknessastheprepregs.Inthehybridarea,90^◦layerswerelocallyreplacedbystainlesssteelpatches.The localmetalfoilcontentvariesfrom6.25%,12.5%to25.0%.Fornotchedlaminates,theopenholetensilestrength issignificantlydecreasedwiththinnerlayerthicknesses.Thefailurebehaviorchangesfromcomplexdelamination dominatedfailuretobrittlefailure.Byusingstainlesssteelfoilsintheregionsofstressconcentrations,energycan bedissipatedbyplasticdeformationofthesteelfoilandstressescanbedeflectedtoneighbouringareas.ForThin- Plysampleswithalocalsteelcontentof25%theopenholetensilestrengthincreasesby64%incomparisontothe referenceThin-Plyspecimensandthesensitivitytowardsstressconcentrationsdecreases.Thebearingstrength ofthehybridCFRPlaminatesisincreasedbyupto54.6%incomparisontothereferencematerial,duetothe confinementofthesteelfoilandtheresultinghighercompressivestrengthandtheplasticdeformationathigh stresses.Thestress–straindiagramsandmicrographsofthefibremetalsamplesrevealthatdamageisinitiated beforethemaximumbearingstrength.However,thedamageoffsetbearingstrengthincreaseconcerningthe specificdensityofthematerialsignificantly.

1. Introduction

Duetotheexcellentdensity-specificmechanicalproperties,carbon fibrereinforcedcomposites(CFRPs)areoftenusedasastructuralma- terialforlightweightstructures[1].However,conventionalCFRPlam- inatesorstructuresdonotexploitthefullpotentialoftheusedcarbon fibres.Oneapproachtoimprovetheirperformanceistoreducethelayer thickness.Kawabeetal.andSihnetal.presentedaspread-towprocess toproducethinunidirectionalplies,so-calledThin-Plylayersaslowas 20μm[2,3].Asaresult,thedegreeoffreedomindesignincreaseswith decreasinglayerthicknessduetothelargernumberof usedplies or morepreciseload-dependentdesign.

Thefailurebehaviorishighlydependentonthelayerthickness,a decreasedlayerthicknesssuppressestransversemicrocrackingandfree edgedelaminationduetoanincreaseoftransversetensilestrength(in- situeﬀect)[4].Thefailuremodechangesfromcomplexdelamination

☆Thisworkwasﬁnancialsupportedbythe“DerÜbersee-Clube.V.”,Hamburg,Germany.ThankstoAviatecGobalAviationGmbH&Co.KG,Henstedt-Ulzburg, Germanyforsupplyingsurfacepre-treatmentsystems

∗Correspondingauthor.

E-mailaddress:benedikt.koetter@tuhh.de(B.Kötter).

URL:http://www.tuhh.de/kvweb(B.Kötter)

dominatedfailureto abrittle failurefrom Thick- toThin-Ply[3–7], whichresultsinahigherultimatetensilestrengthforunnotchedquasi- isotropicspecimens[3,6,8].Inadditiontotheimprovementoftheme- chanicalproperties duetothein-situ effect,thelaminatequalityim- proveswithregardtovoidcontent,fibreangledeviationandresinrich areas withdecreasing layerthickness. This can be attributedto the spreading process,whichresultsin amorehomogeneousfibredistri- bution.Duetothesefactors,thecompressivestrengthofunidirectional Thin-Plysamplesincreasescomparedtothickerlayers[6,8].

Incontrasttotheunnotchedsamples,notchedThin-Plyspecimens exhibitaloweropenholetensilestrength.Asaconsequenceofthesup- pressionofinter-ﬁbrefractureanddelamination,nocrackbluntingof thestressconcentrationduetointerlaminardamageoccurs.Thus,no energycanbedissipatedorstressdivertedtoadjacentareasandnopre- damageleadstoprematureﬁbrefailureatthenotchorstressconcentra- tionregion[3,6,7].Inordertoreducethenotchsensitivity,thematerial

https://doi.org/10.1016/j.jcomc.2020.100085

Received13October2020;Receivedinrevisedform7December2020;Accepted8December2020

(2)

Table1

Laminatelayupsandglobalspecimendensitiesfortheopenholetensile(OHT)andbearingtests.

Conﬁguration (SF: Steel Foil) Density in g/cm ³ Layup OHT Bearing

CFRP: 40 gsm; Reference 1.53 1.53 [45/90/-45/0] 12s

CFRP: 40 gsm; SF 6.25 %; ( t = 0.04 mm) 1.62 1.67 [(45/SF/-45/0) 3/(45/90/-45/0) 9] s

CFRP: 40 gsm; SF 12.5 %; ( t = 0.04 mm) 1.70 1.80 [(45/SF/-45/0) ₆/(45/90/-45/0) ₆] _s CFRP: 40 gsm; SF 25.0 %; ( t = 0.04 mm) 1.86 2.07 [45/SF/-45/0] 12s

CFRP: 160 gsm; Reference 1.53 1.53 [45 4/90 4/-45 4/0 4] 3s

CFRP: 160 gsm; SF 25.0 %; ( t = 0.16 mm) 1.88 2.12 [45 4/SF/-45 4/0 4] 3s

canbemodiﬁedinsuchawaythattheentirefailuremechanismchanges and,forexample,micro-damagesoccur.Onepossibilityistointroduce asecondmaterial(hybridization),allowingsynergyeﬀectstobeused incustomizeddesignsdependingontheloadsituationsuchasnotches orloadintroductionareas.Inthisstudy,theapproachistolocallysub- stitute90^◦CFRPlayerswithstainlesssteelfoilsintheareaofstresscon- centration.Stainlesssteelislesssensitivetostressconcentrationsdue totheirisotropicmaterialandstrainhardeningproperties.Incontrastto titaniumoraerospacealuminiumalloys,stainlesssteelsaremoreeco- nomical[9].Duetohybridizationwithstainlesssteel,thefailurebehav- iorchangesfromabrittlefailurebacktoacomplexmulti-modefailure.

PreviousstudieswhichhaveinvestigatedcarbonfibrereinforcedThin- Plylaminateswithstainlesssteelpatchesshowhigheropenholetensile strengths[10,11],whichincreasedfrom390MPato630MPawitha localsteelcontentof25%,anincreaseofabout60.44%.Althoughthe globaldensityofthespecimenincreasesduetohybridization,theden- sityspecificstrengthisnonethelessincreasingandthenotchsensitivity isreduced.Localthickeningorotherdesignfeaturescanbeminimized oreliminated.Previousstudiesencounteredstressconcentrationsinthe transitionzonecausedbythedifferenceinstiffnessofCFRPandstainless steel.Thereforethestainlesssteelpatcheswerearrangedstep-wise.As aresult,theincreasedstressesaredistributedoverawiderarea,andlo- callythestressconcentrationdecreases[11].Thestaircasearrangement hasalsobeenusedinotherworks[12–14].

AnotherfieldofapplicationforhybridCFRPlaminatesisfastener- basedjoints.Boltorrivetjointsarethemostcommonlyusedtypeof joiningtechniquestodayduetotheirsimpleapplication,lowpriceand thepossibilityofrelease[15–17].However,thistypeofjointisnotap- propriatefortheuseofCFRP.AstudybyAmacheretal.showedthat thebearingstrength ofThick-Plysamplesis20%lowerthantheun- notchedtensilestrength.Inthecaseofthin-layerspecimens,thereduc- tionwas31%,wherebytheabsolutebearingstrengthof584MPafor Thin-Plyishigherthanthebearingstrengthof476MPaforThick-Ply samples[4,6].ThebearingstrengthofThin-Plyishigher,howeverthe differencebetweenunnotchedandbearingstrengthisgreater,whichis limitingthepotentialofThin-Ply.Asaresultthejointshavetobeim- proved.Theuseofstainlesssteellayerscouldreducethegapbetween tensilestrengthandbearingstrengthandthuspreventsolutionssuchas localthickeningofthelaminate[18],placinginsertssuchasmetallic rings[16]orZ-pinning[19,20].Thefrequentlyusedlocalthickening hasthedisadvantagethatbendingstressesandpeelstressesareintro- ducedintothematerial,whichcanbepreventedbyhybridisationwhile retainingtheoriginalgeometry.Petersenetal.investigatedthetransi- tionzone(TZ)betweenstainlesssteelandCFRPlayersforunnotched tensileandbendingsamples[21].Thelayersusedhadathicknessof 0.13mm.Fourdifferent arrangementsof themetalin thetransition zonewereinvestigated.Itwasobservedthatdelaminationbetweenma- trixandstainlesssteelisachallengeintheTZ[21].Withreducedlayer thickness,thenumberoflayersincrease,whichdecreasestheinterlam- inarshearstressescausedbythermalandmechanicalloads,whichre- ducesandsuppressedthegrowthofdelaminations.Asaconsequence, therequirementsofthebondingbetweenresinandmetalare,therefore, lesscritical.Well-knownsolutionssuchastheBoeingSol-Gelprocess aresufficientandnofurtherinvestigationsarenecessary[11,22].Other

workhasfocusedontheinfluenceofthefasteningtorqueonthebearing strengthandthefailurebehaviorofthespecimen.Whethertheboltis tightenedornotandwithwhichtorqueitistightenedhasasignificant influenceonthebearingstrength.Ingeneral,highertighteningtorques increasethebearingstrength,aslongasthetighteningdoesnotdamage thespecimen[12,23–25].

2. Materialsandmethods

2.1. Materialsandspecimenpreparation

The prepregsystem used in this studywas manufacturedby the spread towtechnology atthe IndustrialTechnology Centerof Fukui Prefecture,Japan.TR50ScarbonfibresfromMitsubishiChemicalCo., Ltd. and a bisphenol-A based epoxy resin from a combination of jER828:jER1001ina4:6ratiofromMistubishiChemicalCo.,Ltd.was used.Theareaweightoftheprepreglayerswas40gsm.Thick-Plylam- inates witharesultingareaweightof160 gsmwereproducedusing theblock-scaling method.Thefibrevolumecontentof theproduced laminatesis55%.Assubstitutelayers,theausteniticstainlesssteelal- loy1.4310(X10CrNi18-8)fromh+sPräzisionsfolienGmbH,Pirkbei Weiden,Germanywereused.Thealloy1.4310ischaracterizedbyhigh strengthandductility,andexhibitsgoodcorrosionproperties.Thethick- nessesofthefoilsarethesameastheprepreglayers.Thestrengthofthe Thin-PlymetalfoilishigherthanthestrengthoftheThick-Plyfoildue strainhardeningfromthecoldrollingduringmanufacturing.Thefoils arecutusingaprecisioncutterforelectricalboards.Thiscuttingmethod preventswarpingedgesandthereforeadditionalweakpointsinthema- terials.Table1displaysthelocalvolumecontentofstainlesssteelnear theholeandtheglobalspecimendensity.Incontrasttothelocalsteel content,theglobaldensityiscalculatedovertheentiresample,asthe patchesareonlyappliedlocally,eveninpracticaluse.Ifthesteelcon- tentislowerthan25%,theouter90^◦layersarereplacedbysteelfoils symmetricaltothemid-plane,becauseduetotheclearancefitbetween boltandholetheboltinclinesunderloadandtheouterlayersareloaded underpressure.Inordertominimizealocalstiffnessincreasebyinsert- ingstainlesssteelinsteadof90^◦CFRPlayers,thesteelpatcheshavea differentlength,andtheirdesignisstepwise,asshowninFig.1.

Thelaminatesaremanufacturedbyahandlayupandanintermedi- atevacuumwasappliedeveryfourthlayer.Ahigh-performancesol-gel surfacepre-treatmentprocessusing3M’s(Germany)AC-130-2surface pre-treatmentsystem,anaerospacecertiﬁedtwo-partwater-basedsys- temwithoutchromate,wasappliedtoincreasetheadhesionbetween stainlesssteelandmatrix.Thepre-treatmentsystemimprovesthead- hesionasaresultofthechemicalinteractionattheinterfacebetween stainlesssteelandthematrix[22]. Theprocessconsistsof foursteps [22]:

1. Deoxidizingandextendingthesurfacemanuallyofthemetalwith# 500abrasivepaper.

2. Cleaningofthesurfacewithacetonetoremoveallmetalandabra- sivepartsandmixingtheAC-130-2Sol-GelKit.

3. Applyingthesurfacepre-treatmentbyimmersionbathfor2min.

4. Dryingofthecoatedsurfaceforatleast60min.

(3)

Fig.1. Openholetensilespecimengeometrywithalocal steelcontentof25%and12.5%.

Table2

Openholetensileandbearingspecimendimensions[26,27]. Specimen dimension OHT specimen in mm Bearing specimen in mm Hole diameter, d 6 ± 0.06 6 +0.00/ − 0.03

Thickness, t 3.84 3.84

Length, l 250 135

Width, w 36 ± 1 36 ± 1

Edge distance, e 125 ± 0.12 18 ± 1

Thepatchesareinsertedmanuallyduringlaminationandthelami- natesarecuredinanautoclaveat130^◦Cand5barfor2h.Thecured plateswerecutwith thecircularsaw Brillant265 fromATMwitha Corundumbladeandaconstantfeedspeedof1.5mm/s.Theholewas milledusing adouble-edgeddiamondmillingcutterwithadiameter of1.8mm.Thespecimendimensionsaredeterminedaccordingtothe standardsforOHT(ASTMD5766[26])andforBearing(ASTMD5961 [27])tests.AllspecimendimensionsareshowedinTable2.

2.2. Experimentalmethods

Theopenholetension testsaccordingtoASTMD5766 [26]were carriedoutintwodifferentlabs.Ofthe24samplestested,6weretested inJapanatShimadzu’stestlaboratoryinKyoto,Japan.AshimadzuAG- Xplusuniversaltestmachinewithamaximumtensileforceof250kN andmechanicalwedgeclampswasused.Far-fieldstrainrecordingwas performedusingstraingages,near-fieldrecordingusingaDigitalIm- ageCorrelation(DIC)system(5Msystem)fromGOMGmbH,Germany.

Thenear-ﬁelddisplacementsandstrainswererecordedusingaspeckle pattern,whichwasappliedtothesurfaceofthespecimenusingwhite andblackacrylicpaint.Theevaluationofthedatawascarriedoutby thesoftwareARAMISProfessional,whichisdistributedbyGOMGmbH.

Theremainingopenholetensiontestsandthebearingtestsaccording toASTMD5961[27]areconductedinHamburg,Germany.Forthein- vestigationsinHamburgaZwickRoellZ400universaltestingmachine withamaximumloadof400kNwasused.Thespecimenswereclamped usingmechanicalwedgeclamps.Thetestparameterswereadaptedto thecorrespondingstandardsothattheresultsarecomparablewiththe datafromJapanorotherstudies.ThetestsinJapanaswellasinHam- burgwereperformedunderstandardclimateconditionsof20^◦Cand 50%humidity.

Theprocedure A“Double ShearTest” accordingtoASTM D5961 [27]was chosen forthe bearingtests.The fasteningtorque was set to3Nm.Previousinvestigationshavedemonstratedthatifthehigh- strengthboltsarenottightened,theywillfailbefore thehybridFML specimen.Thereforenopredictionofthebearingstrengthwillbepos- sible.Theclampingpressurewasappliedtothesampleusingasteel

Fig.2.Testsetupdouble-shearbearingtest.

washerwithaninnerdiameter of6.3mm andanouterdiameterof 12mm.Inthecaseofbearingtests,thecalculationofthestrainsresults fromthelocaldisplacementsofthesampleandtheupperclampingsys- tem.Thecalculationofthebearingstrain𝜀^br isshowninFormula1.𝛿1

and𝛿2arethelocaldisplacementsasitisshownintheschematictest setupin Fig.2,𝐾 isthecalculationfactortodistinguishsingle-shear fromdouble-sheartests.Forsingle-shearteststhe𝐾is1.0.𝐷isthedi- ameterofthehole.

𝜀^br= (𝛿1+𝛿2)∕2

𝐾⋅𝐷 (1)

Thebearingstrength𝜎^br iscalculatedaccordingtoFormula2.𝑃 is theload,ℎisthespecimenthicknessand𝑘istheloadperholefactor;

1.0forsingle-fastenerorpintestsand2.0fordouble-fastenertests.

𝜎^br= 𝑃

𝑘⋅𝐷⋅ℎ (2)

To be able toassess thefailuremechanisms, micrographsof the testedsampleswereprepared.Samplesuptotheﬁnalfailure,aswellas samplesuptoastressof95%ofthemaximumstress,wereexamined.

Forthispurpose,thesampleswereﬁrstembeddedincolouredepoxy resintopreventfurtherdamage duringsawingofthesamples. After sawing,thespecimenswereembeddedinacold embeddingmaterial (KEM15plusfromATMQnessGmbH)andautomaticallygroundand polished.Thepolishingprocessconsistsofseveralsteps,startingwith

#320abrasivepapertoadiamondsuspensionwithagrainsizeof1μm.

Thepolishedsurfaceswereobservedwiththedigitalmicroscopefrom KEYENCE.

(4)

Fig.3. Openholetensilestrength(blue)anddensityspeciﬁcopenholetensile strength(grey)ofCFRPandHybridCFRPﬁbremetallaminates.

3. Resultsanddiscussion 3.1. Openholetensiletests

Fig.3showstheresultsoftheopenholetensiletests.Thestrengthis illustratedinblue,andthespeciﬁcstrengthingrey.Thespeciﬁcstrength representstheratioofthestrengthandtheglobaldensityofthesample.

Therefore,Thick-Plysampleswithalocalsteelcontentof25%havea densityof1.88g/cm³incontrasttothereferencesampleswithadensity of1.53g/cm³.Thecorrespondingdensitiesofthesamplesaresumma- rizedinTable1.TheopenholetensilestrengthoftheThin-Plysamples recordeda9%lowerstrengththantheThick-Plyspecimens.Thiscorre- spondstootherstudieswhichhadinvestigatedtheOHTstrengthcon- cerningthelayerthickness[3,6,11].Amacheretal.showedthatinitial damageinthematerialshiftstohigherstrainsorstresseswithdecreasing layerthickness.Theonsetofdamageincreasedfrom255MPaforThick- Plyspecimensto352MPaforThin-Plyspecimens.Duetothedelayed onsetofdamage,thelackofstressrelaxationthroughdamageleadsto higherstressconcentrationsandpremature,brittlefailure[3,6,7].

Theultimatestrength increasessigniﬁcantlywithincreasing steel content.Withalocalsteelcontentof25%,thestrengthrisesby64%, andeventhespeciﬁcstrengthrisesbyupto36%.Duetothehybridiza- tion,thenotchsensitivitydecreaseswithanincreaseinsteelcontent.

Thereductioninnotchsensitivityisexpressedinadecreasingnotched strengthreductionratio(NSR).TheNSRindicatestheratiobetweenthe notchedstrengths𝜎UNTandthemiddlestressinthenetsection𝜎net.The stressinthenetsectioniscalculatedfromtheOHTstrength𝜎OHT,the specimenwidth𝑊 andtheholediameter𝐷,seeEq.(3)[7].

𝑁𝑆𝑅=𝜎UNT

𝜎net

= 𝜎UNT

𝜎OHT∕(1−_𝑊^𝐷)

(3)

Fig.4.Stress–strain-diagramoftheopenholetensiletests.

Avalueofnearlyoneindicatesthatthestressconcentrationhasonly aminimalinfluenceonthestrength,whereashighervaluesindicatehigh sensitivitytostressconcentrations.Themeasuredvaluesareshownin Table3.ItcanbeseenthattheNSRdecreaseswithincreasingsteelcon- tent,thusreducingthesusceptibilitytostressconcentrations.Inthecase oftheThin-PlyFMLsampleswithasteelcontentof25%,theNSRvalue is1.18.Asaresult,areaswithhigh-stressconcentrationshaveonlya smallinfluence,andnothickeningorotherdesignchangesneedtobe madelocally.Thelocalhybridisationofthematerialavoidstheprob- lemoftheinfluenceofstressconcentrationsinthin-layerlaminatesso thatthepotentialofthethinlayerscanbeexploited.Therightbarof Fig.3displaystheresultsoftheThick-Plysampleswithalocalsteelcon- tentof25%.AdecreaseinstrengthisshownincontrasttotheThin-Ply sampleswith25%steelcontent.Thelowerstrengthcanbeexplainedby theformationofdelaminationsbetweenthemetallayersandthematrix.

DuetothehighershearstressesbetweenthelayerscomparedtoThin- Ply,thechosensurfacepre-treatmentisnotsufficient,anddelamina- tionsareformed.Highershearstressesresultpartlyfromthemechanical loadsintroducedandpartlyfromtheresidualstressesinthelaminate.In thecaseoftheinterfacebetweenstainlesssteelandCFRP,andthelarge differenceinthecoefficientofthermalexpansion,theresidualstresses betweenthemarehigher.Thisconsiderationfavourstheformationof delamination.Theformationofdelaminationscanbeseeninthestress- straindiagram(Fig.4)byadecreaseinstiffnessatabout340MPa.

On the abscissa, the near-field strain is plotted, which was ob- tainedusingtheDigitalImageCorrelationsystem.NearfieldStrainwas recordedoveralengthof35mm,sothatonlytheareawiththemaxi- mumstainlesssteelcontentdependingonthesampleismeasured.The measuringfieldisontheleftsideoftheholeandwaschosentoensure thatlocaleffectsoftheholehavenoinfluence(Fig.1).Exceptforthe Thick-Plyspecimenwithalocalsteelcontentof25%,allspecimensex- hibitbrittlefailureandnomajorsignificantpre-damage.Thedifference

Table3

NotchedstrengthreductionratiooftheThin-andThick-PlyFMLsamples.

Conﬁguration (SF: Steel Foil) Strength in MPa NSR

Unnotched Open hole tension CFRP: 40 gsm; Reference 911.97 ± 34.43 392.41 ± 3.84 1.94 CFRP: 40 gsm; SF 6.25 %; ( t = 0.04 mm) 460.78 ± 9.23 1.65 CFRP: 40 gsm; SF 12.5 %; ( t = 0.04 mm) 521.41 ± 11.25 1.46 CFRP: 40 gsm; SF 25.0 %; ( t = 0.04 mm) 642.78 ± 20.67 1.18 CFRP: 160 gsm; Reference 857.21 ± 11.06 427.27 ± 1.67 1.67 CFRP: 160 gsm; SF 25.0 %; ( t = 0.16 mm) 428.02 ± 9.70 1.67

(5)

Fig.5.StrainintensiledirectionshortlybeforeﬁnalfailurerecordedbyaDICsystem.

Table4

Theoreticalandmeasuredstiﬀnessofthehybridarea.Thestiﬀnessesofthe referencesamples(italics)weredeterminedbymeasurementsaccordingto ASTMD3039.

Configuration Measured stiffness in GPa Theoretical stiffness in GPa 40 gsm 43.57 ± 0.91 47.00 ± 0.67

40 gsm; 6.25% SF 51.38 ± 0.39 50.23 40 gsm; 12.5% SF 57.00 ± 1.04 56.89 40 gsm; 25.0% SF 69.43 ± 0.35 70.20 160 gsm 42.70 ± 0.37 47.87 ± 1.49 160 gsm; 25.0% SF 74.13 ± 4.06 70.20

inthestiffnesscanbeexplainedbythelocalsteelcontent.Thesteellay- ersreplace90^◦layersofCFRP,whichhavealowertensilestiffnessthan thesteelfoil.Themeasuredstiffnessescorrespondtothetheoretically calculatedstiffnessesiftheruleofmixtureandthelocalsteelcontent areused.Thestiffnessesoftheindividualmaterialsisbasedontensile testsforquasi-isotropicCFRPsamplesaccordingtotheASTMD3039 [28]andtensiletestsforthesteelfoilaccordingtotheASTME345-16 [29].ThemeasuredandcalculatedstiffnessesareshowninTable4.

TheFar-Fieldstrainwasrecordedoutsidetheareaoflocalhybridiza- tionusingstraingauges, andthestiffnesses weredetermined.Asex- pected,no nearfieldstiffness differencesbetween theconfigurations canbedetected,sincethestiffnessisinfluencedbythefibresandresin andnotbythelayerthickness.Thestiffnessistheequivalenttothenear fieldstiffnessofthenotchedreferencesamples.

TheDICimagesin Fig.5show thesurfacestrainsof thetop45^◦ layerintensiledirectionrightbeforeﬁnalfailure.Exceptfortheright specimen(Thick-Ply,25%SF),allsamplesshowatypicalstresspattern neartheholeforquasi-isotropicopenholetensionspecimens.Astress concentrationpropagatesfromtheholein±45^◦,furthermoreaboveand belowthehole,alocalstressminimumappears.

Furthermore,itisvisiblethatwithincreasingsteelcontent,thetran- sitionzonebetweenCFRPandstainlesssteelismoreclearlyvisibleand thestressconcentrationsatthetransitionzoneincrease.Thus,thetran- sitionzoneinadditiontotheholerepresentsasecondcriticalareawith respecttospecimenfailure.Fig.6illustratestwocurves,representing thestrainintensiledirectionasafunctionofposition.Thepositiondata isaverticalsectionthroughthesample,where0mmrepresentsthecen- treofthesample,i.e.thecentreofthehole(seeFig.6topright).The strainsinthediagrambelongtoafarﬁeldtensilestressof350MPaat whichdelaminationgrowthbeginsinthehybridThick-Plyspecimens.

Thestraincurve(orange)oftheThick-Plyspecimenwithalocalsteel contentof25%showstwostrongstresspeakssymmetricaltothemid- dle.These arelocatedatthetransitionbetween theoutermetallay- ersandthecorresponding90^◦CFRPlayers.Thelocalincreaseofstress initiatesdelaminationgrowthandresultsinearlyfailure.Inthecase ofthehybridThin-Plyspecimenswithasteelcontentof25%(black), nostronglocalizedstressincreaseisobservable.Duetothelowlayer thickness,moremetallayers arerequired,which inturn canbe dis- tributedmoresmoothlyinsteps. Thisensuresthatthestressincrease isdistributedoveralargerarea,andthereforethelocalstressconcen- trationsdecreased.Thespecimenfailsatthehole,asdothereference specimens.

Fig.6. Nearﬁeldstrainintensiledirectionasafunctionofthepositiononthe specimen.

InadditiontotheDICimages,micrographsofthefracturesurfacesof thespecimenswereproduced.Theimagesa)tod)ofFig.7showmicro- graphsofThin-Plyspecimensnexttothehole.Asthecurvesofthestress- straindiagramhavealreadyshown,brittlematerialbehaviorcanalso beseenhere.EspeciallytheThin-Plyspecimenwithoutstainlesssteel foil(Fig.7a)showsaverystraightfracturesurface.Areaswithstainless steelfoilsshowarougherfracturesurface.Thelengthofthefracture surfaceincreasessignificantly.Fig.7eandfshowsdetailedimagesof thefractureedges.Anexplanationforthehigherfracturesurfacesin thehybridareasisthehigherfracturestrainandplasticdeformationof thestainlesssteel.Duetothehigherstrains,moreinter-fibrefractures intheCFRPlayersoccurbeforeultimatefailure.Thesefracturesdonot alwaysoccuratthesamelocationbutarestatisticallydistributed.Right beforeultimatefailure,theindividuallayersarepulledout,similarto thepull-outoffibres.Thisensuresthatadditionalenergyisdissipated andtheopenholetensilestrengthincreases.Acharacteristicoflocalized highplasticstrainsinmetalmaterialsisthefractureangleofthemetal.

InFig.7eitcanbeseenaductilefailurebehaviorofthemetalfoil.

Inordertobeabletoexaminetheinter-ﬁbrefracturesmoreprecisely, sampleswereloaded uptoamaximumstressof95%oftheultimate strength,andfurthermicrographswereprepared.Theseareshownin Fig.7gandh.However,nodamagetothematerialhasyetbeende- tected.Nevertheless,itcanbeseenthatnoresin-richregionsorvoids areintroducedintomaterialduetotheintroductionofmetallayers.Fur- thermore,therearenodelaminationsinthelaminatebetweenthemetal andtheCFRPduetothethermalstresses.Delaminationsareonlyonthe mostoutermetallayers.However,sinceDICdetectednodeformation perpendiculartothesurfaceofthespecimen,theoutsidedelaminations arearesultofthecompressivestressesoccurringwithinthespecimen aftertheﬁnalfracture.

(6)

Fig.7.AtoF:Micrographsofthefracturesurfacesoftheopenholetensilespecimens;GandH:Micrographsoftwospecimensloadedto95%ofthemaximum strength.

Fig.8. Resultsofthemaximumbearingstrengthregardinglayerthicknessand steelcontent.

3.2. Bearingstrength

Inthefollowingsection,theresultsfromthebearingtestsarepre- sented.Fig.8demonstratesthebearingstrengthsofthesixconfigura- tions.Aswiththeopenholetensiontests,threesamplesweretestedper configurationsincetheresultsshowedasignificantdifference.Higher bearing strengths areachieved with decreasing layerthickness. The bearingstrengthincreasesby15.3%from849.85MPa(Thick-Ply)to 979.61MPa(Thin-Ply)withareductionoftheareaweightfrom160gsm to40gsm.PreviousstudiesbyMasaniaetal.andAmacheretal.show thattheuseof thinlayersinfluencesthebearingdamagemechanism [6,10].Theonsetoffirstdamageshiftstohigherstressesandstrainsand theinitiationandpropagationofdelaminationaswellasmatrixcracks aresuppressed.Furthermore,theintra-laminarshearstressesarelower duetothethinlayerthicknesses,whichaccordingtoGarboetal.[30], resultsinadelayintheinitiationofdamagemechanismssuchasmatrix breakage,fibrekinkingandthrough-thicknessshearcracking.Which,

asaresult,increasesthebearingstrength.However,thedisadvantageis thatprogressivebehaviorisnolongerpresentandthestressreduction afterinitialdamageisverylarge.Forsafety-relevantcomponents,such asinaircraftstructures,thisisparticularlycritical[6,31,32].

ThemicrographsinFig.9showthedamagepatternoftestedand embeddedsamples.Fig.9aanddshowsoverviewimagesoftheload- ingareaofthinlayer(a–c)andthick (d–f)specimens. Inthecaseof Thin-Plyspecimens,fewerdelaminationsoccur,andthesamplesshow lessdamageafterloading.Noticeableforthethinlayerthicknessesare thefibrebreakagewithinthe0^◦layers(Fig.9b)and(globalkinkbands (Fig.9c),whichincreasefromthecentreoutwards.Theclampingarea ofthewasherisvisibleasaflatareabesidethehole(bluelinesinFig.9, directlybehindthisclampedareasglobalbucklingoccurs,whichresults infibrebreakageandfinalfailure.Fibrebreakswithinthe0^◦layersin- dicatelargecompressivestresses.Duetothelowlayerthickness,the bendingstiffness ofthespecimen isonlyslightly reduced,withfrac- tureofasingle0^◦layer.Duetotheremainingresidualstiffnessandthe clampingbythewasher,noglobalbucklingoccurs.Concerningregions farawayfromstressconcentrations,nodamageis spreadingfurther.

TheareaaﬀectedbydamageresultingfromtheBearingloaddamage isconsiderablysmallerintheThin-Plyspecimen.Incontrast,largede- laminationsbetween thelayerscan beseenin theThick-Plysample.

Althoughthesupportoftheﬁrstarea(bluelinesinFig.9aandd)does notcauseanybucklingeither,thecompletesamplefailswithinthisarea.

Theindividuallayershavefracturedinnumerousfragmentsandhave shiftedduetothedelamination.Thisbehaviorcanalsobeseeninthe twodetailedimagesinFig.9eandf.Duetothefracturesandthedis- placements,noglobalKinkbandscanbeobserved.Regardingthemore distantareas,externaldelaminationhascontinued,andtheouter45^◦ layerhasdetached.

Duetothehybridizationwithstainlesssteelfoils,signiﬁcantlyhigher bearingstrengthscanbeobserved(seeFig.8).IncontrasttotheThin-Ply sampleswithabearingstrengthof979.6MPa,thestrengthofthesam- pleswithstainlesssteelfoilincreasesto1165.4MPainthecaseof6.25%

stainless steel,via1239.5MPa inthecase of12.5%,to1513.9MPa MPaifall90^◦layersarereplacedbystainlesssteel(25%),whichcorre- spondstoanincreaseof54.6%.InthecaseoftheThick-Plyﬁbremetal specimens,thebearingstrengthincreasesto1470.6MPa,anincreaseof 73%.Inadditiontotheincreasedbearingstrength,however,theoccur- ringdamagemechanismchanges.Thecorrespondingstress-straindia- gramsareshowninFig.10.Onerepresentativesampleisshownforeach conﬁguration.Inadditiontothefactdescribedabove thattheThick-

(7)

Fig. 9. Micrographs of bearing Thin- (up- permicrographs)andThick-Ply(lowermicro- graphs)specimen.

Fig.10. Stress–strain-diagramofThin-Ply,Thick-Plyandtheirhybridconﬁgu- rations.

Plyspecimens(magenta)showaprogressivedamagebehaviorandthe Thin-Plyspecimens(red)abrittlefailure,stressreductionscanbeseen withinthecurvesofthehybridspecimenswithinthetest.Forexample, thecurveoftheThin-Plysamplewithasteelcontentof25%(lightblue) showsthreestressreductions.Theﬁrstoccursat31.8%,thesecondat 37%andthethirdjustbeforeultimatefailureat44.9%.

Thisdevelopmentindicatesthatthematerialhasbeendamaged.In ordertoinvestigatethedamageprocessmoreindetail,sampleswere testedandstoppedwhenthefirstdecreaseinstresswasreachedsothat micrographscouldbepreparedandthedamagemechanismresponsi- bleforthefirstdropintensilestresscouldbeobserved.Thesemicro- graphsareshowninFig.11.Fig.11ashowsaThick-Plysamplewith 25%stainlesssteel.Globalbucklinghasformed,wherebythestainless steellayershavealreadydeformedplastically.Fig.11bshowsadetailed viewofthesample.Theradiiofthebucklingaresmallwhichresultsin fibrefracture.Besides,theFMLThick-Plysamplesshowdelamination attheinterfacesbetweenstainlesssteelandmatrix.Ononehand,de- laminationscanbecausedbyhighshearstressesattheinterface due tothehighstrainofthestainlesssteel.Ontheotherhand,thestressin combinationwithinitialdamagecanleadtoout-of-planestressesand

thusenhancedelaminationgrowth.Besidesthedelaminationsandthe fibrebreaksinthe0^◦layers,inter-fibrefracturesoccurinthe±45^◦layers.Largerradiishowanintactfibre-matrixstructure.Fig.11dande showaThin-Plysamplewitha25%stainlesssteelcontent.Again,the firstloaddropindicatesthefirstbuckling.Thisbucklingspreadsout symmetricallyfromthecentreofthesampletotheoutsideandforms akindofV-shapedwedge,asalreadyobservedwiththeFMLThick-Ply samples.AsinFig.11b,fibrebreakscanbeobservedinthe0^◦layers inFig.11e.Also,some±45^◦layersofinter-fibrebreaksshowregions withplasticdeformations.However,nodelaminationbetweenstainless steelandmatrixisvisible.Duetothelowerlayerthicknessesandthe associatedhighernumberofinterfaces,theshearstressesbetweenthe layersdecrease,anddelaminationgrowthissuppressed.Fig.11cand fshowaThick-PlysamplewithoutstainlesssteelandaThin-Plysam- plewith12.5%ofstainlesssteel.Bothsamplesshowthetypicalfailure behaviordescribedabove,globalbucklingexhibitsfromthemid-plane.

WiththeThick-Plysample,itappearsasifthe0^◦layersinthemiddleof thesamplewouldfailundercompression,initiatingintheformationof bucklingtotheoutside.Thisassumptionconﬁrmswiththesymmetrical damage.

Concerningthequestionofwhetheritmakessensetousefibremetal laminatesforstructuralapplications,itisnotsufficientonlytoconsider themaximumbearingstrength.Inthecaseofconventionalconstruction materialssuchasmetals,anoffsetbearingstrengthof2%strainisthe evaluationcriterion.Therefore,threeparametersareofparticularim- portanceinthecaseofbearingtests.Themaximumbearingstrength,the offsetbearingstrengthandthestressatthefirstloaddrop,whichisthe firstsignificantdamage.AllthreeparametersareshowninFig.12.The bluebarsrepresentthestrengthatthetimeofthefirstmeasurabledam- age,thegreybarsrepresenttheoffsetbearingstrength,andtheblack barsrepresentthemaximumbearingstrength.TheThick-Plysamples withoutstainlesssteelserveasreferencematerialsincetheyhavethe thicknesses commonlyused today.However,the160 gsmplies were laminatedbyblock-scaling(four40gsmlayers),sothatthespreading processofthefibresalsoresultsingoodqualityintermsoffibrevol- umecontent,resin-richregionsandfibreangledeviation,whichhasa positive effectoncompressivebehavior[6,8].TheThick-Plysamples show small differences between thefirst damage,theoffsetand the maximumbearingstrength,butThick-Plysamplesexhibitaprogressive damagebehavior,sothatevenaftertheultimatebearingstrength,there isnoriskofsuddenfailure.Incomparison,theThin-Plysampleswith- outstainlesssteelhaveahigheroffsetandmaximumbearingstrength, butthe stressof thefirstdamage is atthe samelevel astheThick- Plysamples. Therefore,ifthedesignguide toleratesno damage,the

(8)

Fig.11. Micrographsofsamplestesteduntil theﬁrstdamage;a)Thick-Plywith25%local steelcontent,b)Detailedviewofmicrograph a),c)Thin-Plyreference,d)Thin-Plywith25%

localsteelcontent,e)Detailedviewsofmicro- graphd),f)Thin-Plywith12.5%local steel content.

Fig.12. Firstfailure(blue),oﬀset(grey)andmaximumbearing(black)strength forneatandﬁbremetallaminates.

maximumallowablestressisequaltotheThick-Ply.However,afterthe firstdamageahighresidualsafetyfactorexists,asthemaximumbear- ingstrengthis26.96%higher.Thehybridlaminatesshowsignificantly higherstrengthsregardlessoftheselectedparameter.Itisnoticeable thatthestressatthepointsofthefirstdamageinallthreeFMLThin-Ply configurationshaveasimilarvalueduetothehighscattering.Also,the offsetstress,whichisabovethestressofthefirstdamage,isthesame forallthreeconfigurations.Adifferenceisvisibleinthemaximumbear- ingstrength,whichincreasessignificantlywithincreasingsteelcontent.

ComparedtotheThin-Plysampleswithoutsteel,themaximumbearing strengthoftheThin-Plysampleswith6.25%steelincreasesby19.0%, with12.5%by26.5%andthethinFMLsampleswith25%by54.6%.

Thedifferencebetweenthesampleconfigurationsresultsintheresidual safetyfactorafterdamage.Withincreasingsteelcontent,thesafetyfac- torincreasessignificantly.Whetherfirstdamagesaretolerablewould havetobecheckedinafurtherstudyundercyclicload.TheThick-Ply samplesreinforcedwithstainlesssteelshowthebestresultsintermsof bearingstrength.Themaximumbearingstrengthisinthesamerange asthe25%Thin-Plysamples,buttheparametersoftheoffsetandthe

firstdamageincreasesignificantly.Thedifferentdamagemechanisms canbeinterpretedbasedonmicrographsinFig.13.

Finalfailureofallconfigurationsisduetobucklingbehindthespeci- men’ssupportBucklingintheareaoftheclampinghasalocalloaddrop asaconsequencebutisnotcriticalregardingthefinalfailure.Theselo- caldamagescanbeassignedtothestressreductionsinthestress-strain diagrams.However,duetotheside-wisesupportbythewashers,the specimensdonotfailatthefirstbucklingandreceivehighercompres- sivestresses.TheThin-PlyFMLspecimensallhaveseveralglobalkink bands.Itisnoticeablethatintheregionswherethestainlesssteelfoils replacedthe90^◦layers,thecarbonlayersaresupportedandshowminor damage.Evenathighdeformationsofthestainlesssteel,thesupporting effectismaintained.

Especiallythemiddlelayersofthe6.25%(Fig.13a–c)andthe12.5%

(Fig.13d–f)samplesshowsigniﬁcantdamagesintheareaofneatCFRP.

Brokenfibresinthe0^◦layersindicatethatthefibreshavefaileddueto highcompressivestresses.Duetothesupportingeffectofthestainless steelaswellasthehigherresidualbendingstiffnessinthecaseofthe thinlayerthicknesses(seeabove),alocalbucklingdoesnotresultin aglobalfailure.TheresultsfortheThick-PlyFMLsamplesshowthat thefinalfailureresults alsoinabucklingaftertheclampedarea.In contrast,thebucklingcannotbeseensymmetricallyfromthemiddle, butthewholespecimenbucklestooneside.Nobucklingisfoundinthe regionoftheclampingbythewashes,asisthecasewiththinnerlayers.It canbeseenthatthefirstmeasurabledamage,asshowninthebarchart, appliesbyhigherstresses.Thisdamageisprobablythebeginningofthe finalbucklingbehindtheclampingarea.However,smallerkinkbands canbefoundinthe0^◦layers,whicharecausedbythehighcompressive loads.Due tothehighbendingstiffnessof thethickerstainlesssteel foils,the0^◦layersarestillsupportedsothatnoglobalbucklingoccurs.

Inthefrontareaofthesample,asshowninfigurek),fibresbrokeand displacedlocallythattheyformakindofcircle.Thesefibresarebroken intosmallpiecesandshiftedintoeachother.Especiallyinterestingare the±45^◦layers,whichinsomelayersshownovisibledamageandthey arestronglydeformed.Thisbehaviorwouldindicatethatthematrixhas deformedplastically,whichisunusualforafullycuredepoxyresin.

Inadditiontothebearingstrength,theweightandthegeometryof thejointisanimportantfactor.Animprovedstrengthtakingintoac- countfourtimes theweight, asis thecasewithpure stainlesssteel, for example, would be practically meaningless. Forthis reason, the strengthsshouldbeviewedinrelationtotheirweight,orinthiscase in relationtotheirdensity. Fig.14 showstheglobalspeciﬁcbearing strength.Theglobalspeciﬁcbearingstrengthisthequotientofthebear- ingstrengthandtheglobaldensityofthetestedsample.Itcanbeseen

(9)

Fig. 13. Micrographs of hybrid ﬁbre metal compositespecimensafterbearingtests;a–c) Thin-Plywith6.25%localsteelcontent,d–f) Thin-Ply12.5%localsteelcontent,g–i)Thin- Plywith25%localsteelcontent,j–l)Thick-Ply with25%localsteelcontent.

thattheparametersofthethreeFMLconfigurationsapproximatetothe maximumspecificbearingstrength,butarestillhigherthanthoseofthe conventionalspecimens.Withregardtothefirstfailure,theresultsof theThick-Plysampleswith25%stainlesssteelexhibitthehighestval- ues,wherebythesecorrespondtothevalueoftheoffsetandshouldbe takenasthemaximumfordesign.Duetotheformationofdelamina- tions,thepropagationofdamageisnotascriticalinstaticbearingtests, butprobablyhaveastrongeffectonthefatiguebearingproperties.This hastobeclarifiedinfurtherinvestigations,relatedtotheThin-PlyFML samples,thesampleswithasteelcontentof6.25%showthehighest specificstrengthsinfirstfailure.Thevaluesare35.5%abovethespe- cificfirstfailurestrengthoftheThin-Plysamplesand26.0%abovethe Thick-Plysamples.Inrelationtotheindustrialapplicationofthehybrid compositesinvestigatedhere,itisshownthathybridisationwith6.25%

stainlesssteelachievesthesamespecificmaximumbearingstrengthand thehighestspecificoffsetbearingstrengthastheconfigurationswitha highersteelcontent.Therefore,althoughthebearingstrengthincreases withahighersteelcontent,thisdoesnotofferanyfurtheradvantage concerningthelightweightdesignduetotheincreasedweight,andthus inpractice.

4. Conclusion

ThisstudyshowsthatthehybridisationofThin-PlyCFRPlaminates withstainlesssteelpatchesinareasofstressconcentrationssigniﬁcantly increases openhole tensile andbearingstrength.The previous limi- tation of thegapbetween unnotchedstrength andopenholetensile strength orratherbearingstrength canbe reduced,allowingthepo- tentialofthinpliestobeappliedinstructurallightweightapplications.

InthecaseoftheThin-Plysampleswithalocalsteelcontentof25%, theopenholetensionstrengthwasincreasedby64%andthebearing strengthby54.6%comparedtotheThin-Plysamplewithoutstainless steel.Ifthestrengthsarenormalisedtotheglobaldensityofthesam- ple,theincreasesare36%forthespecificopenholetensionstrengthand 14%forthespecificbearingstrength.Micrographsofsamplesfromboth loadcasesshowthatthesteeldeformedplasticallyunderhighstrains andwasabletodissipateenergy.Duetotheplasticdeformationand thesupportingeffectofthestainlesssteelagainstbuckling,thespecific strengthisincreased.AnotherlimitingfactorinthecaseofThick-Ply FMLlaminatesisthetransitionzonebetweenstainlesssteelandCFRP.

Duetothedifferentstiffnessesandcoefficientsofthermalexpansion,

(10)

Fig.14. Specificfirstfailure,offsetandmaximumbearingstrength.

stressconcentrationsoccur,whichinitiatedelaminationinthickerlay- ersandleadtoprematurefailure.Theuseofthin-layerstainlesssteel foilsandastep-wisearrangementofthese foilsleadstoasmoothin- creaseinstiffnessoveralargerareasothatthestressconcentrationsare belowthecriticalvaluefortheformationofdelamination.Thushigher openholetensilestrength isachieved.Inthecaseof bearingperfor- mance,itwasshownthatconcerningthedensityofthesample,anideal localsteelvolumecontentof6.25%wasfiguredout.However,thebear- ingstrengthincreaseswithhighersteelcontent,butthedensityspecific bearingstrengthdoesnotincreasebecauseofthehigherweight.Overall, theperformanceofCFRPThin-Plysamplescouldbeimprovedbyusing stainlesssteelfoilsaspatchesatstressconcentrationsandintheareaof loadintroduction,demonstratinghighpotentialforhighperformance structures.

DeclarationofCompetingInterest

Theauthorsdeclarethattheyhavenoknowncompetingﬁnancial interestsorpersonalrelationshipsthatcouldhaveappearedtoinﬂuence theworkreportedinthispaper.

References

[1] M.C.-Y. Niu , Composite Airframe Structures: Practical Design Information and Data, 3rd, Conmilit Press, Hong Kong, 2000 .

[2] K. Kawabe , S. Tomoda , T. Matsuo , A pneumatic process for spreading reinforc- ing ﬁber tow, in: Proceedings of the 42nd International SAMPE Symposium, 1997, pp. 65–76 .

[3] S. Sihn, R. Kim, K. Kawabe, S. Tsai, Experimental studies of thin-ply laminated composites, Compos. Sci. Technol. 67 (6) (2007) 996–1008, doi: 10.1016/j.compscitech.2006.06.008 .

[4] A. Arteiro, C. Furtado, G. Catalanotti, P. Linde, P.P. Camanho, Thin-ply polymer composite materials: a review, Compos. Part A: Appl. Sci. Manuf. 132 (2020) 105777, doi: 10.1016/j.compositesa.2020.105777 .

[5] M.R. Wisnom, B. Khan, S.R. Hallett, Size eﬀects in unnotched tensile strength of unidirectional and quasi-isotropic carbon/epoxy composites, Compos. Struct. 84 (1) (2008) 21–28, doi: 10.1016/j.compstruct.2007.06.002 .

[6] R. Amacher, J. Cugnoni, J. Botsis, L. Sorensen, W. Smith, C. Dransfeld, Thin ply composites: experimental characterization and modeling of size-eﬀects, Compos. Sci.

Technol. 101 (101) (2014) 121–132, doi: 10.1016/j.compscitech.2014.06.027 .

[7] J. Cugnoni, R. Amacher, S. Kohler, J. Brunner, E. Kramer, C. Dransfeld, W. Smith, K. Scobbie, L. Sorensen, J. Botsis, Towards aerospace grade thin- ply composites: eﬀect of ply thickness, ﬁbre, matrix and interlayer toughening on strength and damage tolerance, Compos. Sci. Technol. 168 (2018) 467–477, doi: 10.1016/j.compscitech.2018.08.037 .

[8] T. Yokozeki, Y. Aoki, T. Ogasawara, Experimental characterization of strength and damage resistance properties of thin-ply carbon ﬁber/toughened epoxy laminates, Compos. Struct. 82 (3) (2008) 382–389, doi: 10.1016/j.compstruct.2007.01.015 . [9] J. Studer, A. Keller, F. Leone, D. Stefaniak, C. Dransfeld, K. Masania, Local reinforce-

ment of aerospace structures using co-curing RTM of metal foil hybrid composites, Prod. Eng. 12 (2) (2018) 195–201, doi: 10.1007/s11740-018-0794-3 .

[10] K. Masania, R. Geissberger, D. Stefaniak, C. Dransfeld, Steel Foil Reinforced Compos- ites: Experimental and Numerical Study of Strength, Plasticity and Ply Size Eﬀect, 2015.

[11] B. Kötter, J. Karsten, J. Körbelin, B. Fiedler, Cfrp thin-ply ﬁbre metal laminates:

inﬂuences of ply thickness and metal layers on open hole tension and compression properties, Materials 13 (4) (2020), doi: 10.3390/ma13040910 .

[12] B. Bosbach, M. Baytekin-Gerngross, E. Sprecher, J. Wegner, M.-D. Gerngross, J. Carstensen, R. Adelung, B. Fiedler, Maximizing bearing fatigue lifetime and CAI capability of ﬁbre metal laminates by nanoscale sculptured al plies, Compos. Part A:

Appl. Sci. Manuf. 117 (2019) 144–155, doi: 10.1016/j.compositesa.2018.11.017 . [13] G. Kolks, K.I. Tserpes, Eﬃcient progressive damage modeling of hybrid compos-

ite/titanium bolted joints, Compos. Part A: Appl. Sci. Manuf. 56 (2014) 51–63, doi: 10.1016/j.compositesa.2013.09.011 .

[14] J. Both , M. Wedekind , H. Baier , Simulation and experimental characterization of the bearing behaviour of CFRP-metall laminates, in: Proceedings of the 15th European Conference on Composites, 2012, pp. 1–7 . Venice

[15] P.P. Camanho , F.L. Matthews , Stress analysis and strength prediction of mechanically fastened joints in FRP: a review, Compos. Struct. 28 (1997) 529–547 . [16] P.P. Camanho, C. Tavares, R. de Oliveira, A.T. Marques, A. Ferreira, Increasing the

eﬃciency of composite single-shear lap joints using bonded inserts, Compos. Part B:

Eng. 36 (5) (2005) 372–383, doi: 10.1016/j.compositesb.2005.01.007 .

[17] G. Catalanotti, P.P. Camanho, A semi-analytical method to predict net-tension failure of mechanically fastened joints in composite laminates, Compos. Sci. Technol. 76 (2013) 69–76, doi: 10.1016/j.compscitech.2012.12.009 .

[18] A. Fink, P.P. Camanho, J.M. Andrés, E. Pfeiﬀer, A. Obst, Hybrid CFRP/titanium bolted joints: performance assessment and application to a spacecraft payload adaptor, Compos. Sci. Technol. 70 (2) (2010) 305–317, doi: 10.1016/j.compscitech.2009.11.002 .

[19] A. Crosky, D. Kelly, R. Li, X. Legrand, N. Huong, R. Ujjin, Improvement of bearing strength of laminated composites, Compos. Struct. 76 (3) (2006) 260–271, doi: 10.1016/j.compstruct.2006.06.036 .

[20] R. Li, N. Huong, A. Crosky, A.P. Mouritz, D. Kelly, P. Chang, Improving bearing performance of composite bolted joints using z-pins, Compos. Sci. Technol. 69 (7-8) (2009) 883–889, doi: 10.1016/j.compscitech.2008.12.005 .

[21] E. Petersen, D. Stefaniak, C. Hühne, Experimental investigation of load car- rying mechanisms and failure phenomena in the transition zone of lo- cally metal reinforced joining areas, Compos. Struct. 182 (2017) 79–90, doi: 10.1016/j.compstruct.2017.09.002 .

[22] 3M Aerospace and Aircraft Maintanance Division, 3M Surface Pre-Treatment AC- 130, 2012.

[23] W.-H. Chen , S.-S. Lee , J.-T. Yeh , Three-dimensional contact stress analysis of a composite laminate with bolted joint, Compos. Struct. 30 (1995) 287–297 .

[24] P.P. Camanho, M. Lambert, A design methodology for mechanically fastened joints in laminated composite materials, Compos. Sci. Technol. 66 (15) (2006) 3004–3020, doi: 10.1016/j.compscitech.2006.02.017 .

[25] C. Cooper , G.J. Turvey , Eﬀects of joint geometry and bolt torque on the structural performance of single bolt tension joints in pultruded GRP sheet material, Compos.

Struct. 32 (1995) 217–226 .

[26] D30 Committee, Test Method for Open-Hole Tensile Strength of Polymer Matrix Composite Laminates, ASTM D 5766-02, 2002, 10.1520/D5766_D5766M-11R18 [27] D30 Committee, Test Method for Bearing Response of Polymer Matrix Composite

Laminates, ASTM D 5961-01, 2001, 10.1520/D5961_D5961M-17

[28] D30 Committee, Test Method for Tensile Properties of Polymer Matrix Composite Materials, ASTM D 3039-00, 2000, 10.1520/D3039_D3039M-17

[29] E28 Committee, Test Methods of Tension Testing of Metallic Foil, ASTM e345-16, 2016, 10.1520/E0345-16

[30] S.P. Garbo , J.M. Ogonowski , Eﬀect of variances and manufacturingtolerances on the design strength and life of mechanically fastenedcomposite joints, Tech. Rep.

AFWAL-TR-81-3041, vol. 1, McDonnell Aircraft Company, 1981 . Final Report AFWAL-TR-81-3041.

[31] A. Arteiro, G. Catalanotti, J. Xavier, P.P. Camanho, Notched response of non-crimp fabric thin-ply laminates: Analysis methods, Compos. Sci. Technol. 88 (2013) 165–

171, doi: 10.1016/j.compscitech.2013.09.003 .

[32] P.P. Camanho , Application of Numerical Methods to the Strength Prediction of Me- chanically Fastened Joints in Composite Laminates, Imperial College of Science, Lon- don, 1999 Dissertation .