Influence of the Type of Adhesive on the Properties of the GFRP Composite Adhesive Joint, Determined on the Basis of the Static T-Peel Test

ADVANCES IN MATERIALS SCIENCE, Vol. 21, No. 3 (69), Sep embe 2021 DOI: 10.2478/adms-2021-0018 And zej Kubi 1,*, Tomáš Ka ňák2, Tomasz Py lowany3 1 Depa men o Manu ac u ing and P oduc ion Enginee ing, Rzeszow Uni e si y o Technology, Poland 2 Ins i u e o Ae ospace Enginee ing, B no Uni e si y o Technology, B no, Czech Republic 3 EKOLOT, ul. P zemysłowa 3A, 38-457 Szczepańcowa, Poland * [email p o ec ed] INFLUENCE OF THE TYPE OF ADHESIVE ON THE PROPERTIES OF THE GFRP COMPOSITE ADHESIVE JOINT, DETERMINED ON THE BASIS OF THE STATIC T-PEEL TEST ABSTRACT The a icle p esen s he esul s o expe imen al s udies de e mining he in luence o he ype o adhesi e on he s a ic s eng h p ope ies o he Glass Fibe Rein o ced Polyme (GFRP) composi e join de e mined on he basis o he T-peel es . As pa o he s a ic es s on peeling join s, a compa ison o peak load and s i ness o indi idual join s was made. The ac u e su aces we e also analyzed, showing a ious ailu e mechanisms. I was shown ha he a ian o join s made wi h he Engua d BP72A polyes e adhesi e was cha ac e ized by he highes s eng h p ope ies wi h a mean peak load o 836.73 N. Keywo ds: GFRP composi es; adhesi e join s; T-peel es ; ailu e analisys INTRODUCTION Composi e s uc u es based on ibe ein o ced polyme s (FRP) a e mo e and mo e widely used in he cons uc ion o s uc u es in a ia ion, au omo i e, shipbuilding, as well as, among o he s, in wind ene gy [1, 2]. Composi e ma e ials a e also e y o en used o epai exis ing s uc u es [3, 4]. GFRP composi es a e widely used in a chi ec u al applica ions, among o he s oo ing and pa i ion walls a e made o hem [5,6]. This is due o he e y a ac i e p ope ies o his ype o composi es, and abo e all, i is abou high speci ic s eng h. The e o e, composi es a e o en used in he cons uc ion o s uc u es which equi e low weigh while main aining high s eng h [7]. The conside ed composi es a e composed o wo main componen s, namely he plas ic ma ix and he ein o cemen ibe s ha ca y he load [8-10]. The p inciple o ope a ion o FRP composi es is based on he ans e o loads by a ious ypes o ibe s. The ma ix se es only as a binde o he ibe s and di ec p o ec ion agains ex e nal ac o s. The ibe s used o hei p oduc ion can be con inuous (con inuous o mono ilamen ) o discon inuous (s aple ibe s, whiske s). Nume ous p oduc s made o single 64 ADVANCES IN MATERIALS SCIENCE, Vol. 21, No. 3 (69), Sep embe 2021 ibe s can also be used as ein o cemen : o ings, ma s, ab ics, p e-imp egna es, shaped elemen s [11,12]. Cu en ly, a signi ican limi a ion in he use o GFRP composi es in mass indus y is he echnology ha equi es a lo o wo k, which makes he p oduc ion cos s ela i ely high. Ano he p oblem is he disposal o used composi e p oduc s based on i e e sibly ha dened epoxy o polyes e esins [1]. Howe e , despi e ce ain limi a ions, composi es a e consis en ly inc easingly used in esponsible, high-s eng h and ligh weigh s uc u es. Wi h he de elopmen o he use o polyme - ibe composi es, design solu ions, including combina ions o such ma e ials, de elop. Gene ally, in hin-walled composi e s uc u es simila echnologies a e used as in his ype o s uc u es made o me als, i.e. i e ing and adhesi e bonding a e used, while composi es wi h a he moplas ic ma ix a e esis i ely welded [13]. Rega ding he i e ing echnology, which wo ks well in me al s uc u es, i is less e ec i e o composi es, because by making i e holes, he con inui y o he ibe s is b oken, which educes he load capaci y o he s uc u e. The i e ed holes make he s uc u e highly suscep ible o he ini ia ion o a igue p ocesses, and inally, me al i e s lead o an inc ease in he s uc u e mass [14,15]. Resis ance welding canno be used o joining composi es wi h a ma ix o epoxy o polyes e esins. Adhesi e join s seem o be a a o able o m o joining elemen s o composi e s uc u es. These ypes o join s do no in oduce holes, s ess concen a o s, ensu e uni o m dis ibu ion o s esses, and due o he hin laye o adhesi e, in p ac ice hey do no inc ease he weigh o he s uc u e. Ensu ing a high-quali y adhesi e bond equi es p ope p epa a ion o he su aces o he joined ma e ials. The au ho s o he wo ks [16,17] ha e ca ied ou a se ies o s udies on he in luence o he su ace p ope ies o polyme - ibe composi es on he e ec i eness o adhesi e bonds. The s eng h o he s uc u e in which he e a e adhesi e join s is ex emely in luenced by he con igu a ion o hese join s. Resea ch on he in luence o a ious con igu a ions o adhesi e bonding composi e s uc u es was s udied by Adams e al. [18] and Godzimi ski e al. [19]. Fo FRP composi e join s, ASTM D5573 [20] dis inguishes se en common ailu e cases. They a e: adhesi e ailu e, cohesi e ailu e, hin-laye cohesi e ailu e, ibe - ea ailu e, ligh - ibe - ea ailu e, s ock-b eak ailu e, o mixed ailu e. The mechanism o ailu e o he adhesi e join is in luenced by a numbe o ac o s, anging om he quali y o he composi e, he adhesi e used, he joining echnology, su ace p epa a ion, e c. The au ho s o [21-28] aised he issue o he in luence o a ious ac o s, incl. edge cham e ing, in luence o a ious en i onmen al condi ions, su ace con amina ion, su ace ea men wi h plasma on he way o join ailu e. Due o he mul i ude o ac o s in luencing he ailu e mode o composi e join s, i is di icul o p edic he o m o ailu e. The au ho s o he wo ks [29-33] made a emp s o p edic he manne o ailu e o he adhesi e join s o GFRP composi es. This pape desc ibes expe imen al s udies in which he in luence o a ious adhesi es on he ailu e mechanism and he s eng h o he join s o GFRP composi es in he s a ic peel es was de e mined. Floa ing olle peel es s desc ibed in he ASTM D3167 s anda d [34] a e commonly used o es he s eng h o adhesi e join s unde peel condi ions in ela ion o FRP composi es. Riul e al. [35] desc ibed he expe imen al es s ca ied ou wi h his me hod, which we e aimed a de e mining he s eng h o he join s o composi es p oduced wi h a ious me hods. An equally common me hod o es ing he peel s eng h o adhesi e join s is he T-peel es . I is cha ac e ized by he ac ha i does no equi e he use o addi ional de ices, and he p epa a ion o samples is no complica ed, hence i is a simple me hod, and he e o e o en used [36,37]. This me hod was adop ed in he pape as an e ec i e way o de e mine he p ope ies o GFRP composi es join s. And zej Kubi , Tomáš Ka ňák, Tomasz Py lowany: In luence o he ype o adhesi e on he 65 p ope ies o he g p composi e adhesi e join , de e mined on he basis o he s a ic -peel es MATERIALS AND METHODS This pape deals wi h he issue o de e mining he pa ame e s o adhesi e o he wing pa s made o GFRP composi es o he ai c a p oduced by he manu ac u e o ligh -spo ai c a (LSA), he Ekolo company (K osno, Poland). The esea ch was ca ied ou o he join s o GFRP composi es wi h a ma ix o inyl es e esin. Fo he p epa a ion o he composi es, ou laye s o In e glas 92125 glass ab ic we e used, a double 2/2 wea e wi h a g ammage o 280 g/m2, addi ionally, a delamina ing ab ic was used o he i s su ace. The a angemen o he ibe s was +/- 45°. The ma ix o he lamina ing composi e was inyl es e esin, howe e , he de ailed composi ion o he esin is a ade sec e o he manu ac u e . Lamina ion was ca ied ou in a mold gi ing an angle shape app oxima ely 180 mm wide (Fig. 1a). In he ha dening p ocess, he acuum p essu e me hod was used, a e ha dening, a composi e 1.4 mm hick was ob ained. A e cu ing unde acuum condi ions o 24 hou s and a e seasoning o 10 days, he composi e was glued wi h ou a ian s o adhesi es: Va ian Adhesi e_1: Engua d BP72A bonding pas e polyes e adhesi e. Va ian V2: Engua d BP06AW bonding pas e polyes e adhesi e. Va ian V3: inyl es e esin, he same composi ion as used o lamina ion. Va ian V4: adhesi e p epa ed on he basis o inyl es e esin, wi h p opo ions iden ical o he use used o lamina ion, and illed wi h CAB-O-SIL TS-720 (syn he ic amo phous silica) powde o ob ain a uni o m adhesi e. A e cu ing, he join was cu open wi h a band saw o samples ha ing a wid h o 30 mm, he eby ob aining T-peel es samples ha ing he shape and dimensions shown in Figu e 1b. Fig. 1. Geome y o T-peel es specimen: (a) be o e cu ing in o indi idual specimens; (b) specimen dimensions (dimensions in mm) T-peel es s we e conduc ed acco ding o he ecommenda ions o he ASTM D 1876 s anda d [36]. The es s o he s a ic peel s eng h we e ca ied ou on he Zwick/Roell Z100 es ing machine, he es a e was 100 mm/min. The s a ic s eng h es s we e ca ied ou a oom empe a u e. Fi e epe i ions we e ca ied ou o each o he conside ed a ian s. 66 ADVANCES IN MATERIALS SCIENCE, Vol. 21, No. 3 (69), Sep embe 2021 The mo phologies o he ac u e su aces o he adhe ends we e examined using an scanning elec on mic oscope (SEM) Phenom P oX (Nanoscience Ins umen s, Phoenix, AZ, USA). RESULTS Figu e 2 shows a g aph con aining ep esen a i e s a ic peel es cu es o he conside ed join a ian s. On he basis o he peel cu es, i can be concluded ha in p ac ically e e y a ian o he join , he peak load alue is signi ican ly highe han he basic, a e age alue o he peel o ce. This is due o, in e alia, he ac ha he samples did no emo e he lash in he on pa o he join , which signi ican ly s eng hens hem. When analyzing he peel cu es o he T-peel es , di e ences in he s i ness o indi idual join s can be obse ed. To isualize he onse o he peel cu es, Figu e 3 shows he same cu es bu wi hin a na owed displacemen ange, i.e. 0-2.5 mm. On his basis, i can be obse ed ha he join made wi h Adhesi e_2 is cha ac e ized by he lowes s i ness among he conside ed adhesi es. The o he join a ian s ha e compa able s i ness. Ano he conclusion om he peel cu e analysis is ha each o he adhesi e ille s used, i.e. in he case o Engua d BP06AW i was a ligh mine al ille ( a ian Adhesi e_2), while in he case o inyl es e esin, syn he ic amo phous silica ( a ian Adhesi e_4), lead o educe he s eng h o he join , as demons a ed he e on he basis o ela i ely low peak load alues. Fig. 2. Typical o ce-displacemen cu es o T-peel es o adhesi es s udied 0 100 200 300 400 500 600 700 800 900 0 5 10 15 20 25 30 Peel o ce (N) Displacemen (mm) Adhesi e_1 Adhesi e_2 Adhesi e_3 Adhesi e_4 And zej Kubi , Tomáš Ka ňák, Tomasz Py lowany: In luence o he ype o adhesi e on he 67 p ope ies o he g p composi e adhesi e join , de e mined on he basis o he s a ic -peel es Fig. 3. Ini ial agmen o T-peel es cu es showing di e ences in s i ness o indi idual adhesi es in he ini ial phase o peel Mo ing on o he s a is ical analysis o he ob ained esul s, he Figu e 4 shows a ba cha which summa izes he a e age peak load alues o indi idual a ian s wi h he s anda d de ia ion alue ma ked. I has been shown ha he a ian o join s made wi h Adhesi e_1 is cha ac e ized by he highes epea abili y o he ob ained esul s. Wi h an a e age peak load alue o 836.73 N, he s anda d de ia ion was 18.06 N. The Adhesi e_3 a ian is also sa is ac o y, wi h an a e age peak load alue o 651.29 N wi h a s anda d de ia ion o 25.64 N. De ini ely he lowes epea abili y o he peel s eng h was demons a ed o he join made o inyl es e esin illed wi h silica ( a ian Adhesi e_4). In his case, he a e age peak load alue was 409.72 N and he s anda d de ia ion was 72.23 N. The e ec o such a low epea abili y o join s in he Adhesi e_4 a ian may be une en dis ibu ion o he ille in he esin s uc u e. Fig. 4. A e age alues o peak load o adhesi es s udied 0 100 200 300 400 500 600 700 800 900 0 0,5 1 1,5 2 2,5 Peel o ce (N) Displacemen (mm) Adhesi e_1 Adhesi e_2 Adhesi e_3 Adhesi e_4 68 ADVANCES IN MATERIALS SCIENCE, Vol. 21, No. 3 (69), Sep embe 2021 a) b) c) d) Fig. 5. View o he ypical ac u e su aces o adhesi e join s o a ian s: (a) Adhesi e_1; (b) Adhesi e_2; (c) Adhesi e_3; (d) Adhesi e_4 And zej Kubi , Tomáš Ka ňák, Tomasz Py lowany: In luence o he ype o adhesi e on he 69 p ope ies o he g p composi e adhesi e join , de e mined on he basis o he s a ic -peel es By analyzing he su ace o ailu e o join s on a mac oscopic scale, i is possible o indica e he mode o ailu e ypical o gi en a ian s o join s. Figu e 5 shows he iews o he ac u e su aces o he join o bo h adhe ends and he enla ged iew o he on pa o he join . In he case o join s made wi h bo h Engua d adhesi es ( a ian s Adhesi e_1 and Adhesi e_2), a ypical cohesi e ailu e was demons a ed (Figs. 5a, b). Which in he case o adhesi e join s is he desi ed mode o ailu e. Fo he a ian whe e he join was made wi h he unmodi ied esin used o lamina ion ( a ian Adhesi e_3), ligh - ibe - ea ailu e was achie ed by de aching he ma ix om he su ace o he ibe s (Fig. 5d). Thus, he e is a s ong bond be ween he p e iously ha dened ma ix esin and he esin used as a adhesi e wi h iden ical p ope ies.. This join was so s ong ha he adhesi e be ween he lamina e ma ix and ibe s ob ained in he echnological lamina ion p ocess was damaged. In he case o a esin-based adhesi e ha was illed (Adhesi e_4 a ian ), adhesi e ailu e was ob ained on he majo i y o he join su ace, while a small a ea o ligh - ibe - ea ailu e was shown in he on pa o he join (Fig. 5c). Join ailu e su aces we e also obse ed on a mic oscopic scale. Fo his pu pose, SEM mic oscopy was used o make ac og aphs. On he basis o he ac u e su ace iew in he case o samples joined wi h Adhesi e_1, i was obse ed ha despi e he dominan cohesi e ailu e (Fig. 6a), small a eas o adhesi e ailu e can be seen on he ailu e su ace, as shown in Figu e 6b. The composi e su ace is isible he e om he adhesi e was sepa a ed. This ype o ailu e o he adhesi e join s o CFRP (Ca bon Fibe Rein o ced Polyme ) composi es was also demons a ed by he au ho s o he wo k [33]. a) b) Fig. 6. SEM images o selec ed a eas o ac u e su aces o o he specimen joined wi h Adhesi e_1: (a) cohesi e ailu e; (b) adhesi e ailu e Mo ing on o he SEM ac og aphy analysis o join s p oduced wi h he use o he Adhesi e_2 a ian , he ully cohesi e na u e o join ailu e was con i med. On he bo h su aces o adhe ends he cha ac e is ic ille was obse ed (Figs. 7a, b). The isible ille is hollow mine al objec s aimed a educing he mass o adhesi e s uc u es. I can be obse ed ha he ille has a high adhesion o he adhesi e. As can be seen in he ac og aphy, he ille pa icles ha e di e en sizes and when he join ailu es, hey also c ack. Since he ailu e o 70 ADVANCES IN MATERIALS SCIENCE, Vol. 21, No. 3 (69), Sep embe 2021 he adhesi e s uc u e also occu s h ough c acking o he ille pa icles, and i s pa icles a e hollow, i can be concluded ha o some ex en he ille leads o a dec ease in he adhesi e s eng h, which was con i med by he esul s o s a ic s eng h es s. These ypes o ille s wi h pa icle sizes in he o de o 10-6 m usually ha e a nega i e e ec on he s eng h o he join s [38-39], bu a e ne e heless used due o o he desi able p ope ies, such as educ ion o join mass, elec ical and he mal conduc i i y o he join , e c. a) b) Fig. 7. SEM ac og aphs o he Adhesi e_2 a ian showing cohesi e ailu e wi h isible pa icles o adhesi e ille a di e en magni ica ion (a) 690x and (b) 2550x The adhesi e wi h mine al ille used in he wo k is usually used whe e a hick laye o he join is equi ed o as a pu y o ill in une enness, he e o e i is impo an ha i s mass is as low as possible. Hollow, hin-walled sphe ical s uc u es c acking du ing he ailu e o he join , cause he cohesi e cha ac e o c acking in he adhesi e join . I is also caused by he ac ha in he join plane due o he hollow hin-walled ille pa icles, he e is less ac i e adhesi e su ace. A simila na u e o he ailu e o adhesi e join s wi h ille s was demons a ed by he au ho s o [39], who in es iga ed he e ec o he ille con en on he s eng h o join s o aluminium alloy shee s. In he case o join s made wi h non- illed inyl es e esin, iden ical o he one used in he lamina ion p ocess, which is he ma ix o he composi e, SEM mic og aph analysis showed ligh - ibe - ea ailu e, which con i ms he conclusions ob ained on he basis o he mac oscopic analysis. The su ace ac og aphy o he Adhesi e_3 a ian showed comple ely exposed ibe s on he en i e su ace o one o he adhe ends. Figs. 8a, b show he ac u e su aces o he beginning and end po ions o he join , espec i ely. The isible, exposed ibe s p o e ha he ailu e ook place he e by b eaking he adhesi e connec ion be ween he ibe s and he ma ix. The esin used o bond he p e iously lamina ed composi es was hus cha ac e ized by high adhesion, c ea ing a s uc u ally uni o m mass wi h he ma ix. A simila na u e o ailu e was demons a ed o he adhesi e in he o m o illed inyl es e esin (Adhesi e_4), bu only in he on al pa o he join (Fig. 8c). Fu he om he ace o he join , he cohesi e na u e o he ailu e was demons a ed (Fig. 8d), which p o es And zej Kubi , Tomáš Ka ňák, Tomasz Py lowany: In luence o he ype o adhesi e on he 71 p ope ies o he g p composi e adhesi e join , de e mined on he basis o he s a ic -peel es ha he powde used o hicken he esin weakens he ha dened s uc u e. This is con i med by he esul s o s a ic s eng h es s, which showed ha he join s p oduced wi h he conside ed adhesi e, a e cha ac e ized by he lowes s eng h among he conside ed a ian s. The on o join was de ined as a dis ance o up o 10 mm om he leading edge o he join , and he end o join was de ined as he a ea app oxima ely 90 mm om he ace o he join . a) b) c) d) Fig. 8. SEM images o ac u e su aces o adhe ends o : (a) on pa o join – a ian Adhesi e_3; (b) end pa o join – wa ian Adhesi e_3; (c) on pa o join – a ian Adhesi e_4; (d) end pa o join – a ian Adhesi e_4