Conductive Characteristics of Indium Tin Oxide Thin Film on Polymeric Substrate Under Long-Term Static Deformation

coatings

Article Conductive Characteristics of Indium Tin Oxide Thin Film on Polymeric Substrate under Long-Term Static Deformation

Dinh-Phuc Tran, Hung-I Lu and Chih-Kuang Lin *

Department of Mechanical Engineering, National Central University, Jhong-Li District, Tao-Yuan 32001, Taiwan; [email protected] (D.-P.T.); [email protected] (H.-I.L.) * Correspondence: [email protected]; Tel.: +886-3-426-7340; Fax: +886-3-425-4501

Received: 17 April 2018; Accepted: 29 May 2018; Published: 1 June 2018

Abstract: The objective of this study is to investigate the effect of long-term static bending on the conductive characteristics of indium tin oxide (ITO) thin film in flexible optoelectronics. Two types of substrate are considered, namely ITO on polyethylene naphthalate (ITO/PEN) and ITO on polyethylene terephthalate (ITO/PET). Electrical properties of the ITO/PEN and ITO/PET sheets are measured in situ under static bending at various radii of curvature. Experimental results indicate that no significant change in electrical resistance of the ITO/PEN and ITO/PET sheets is found for compressive bending after 1000 h at a curvature radius of 10 mm or larger. However, the ITO/PEN and ITO/PET sheets are seriously damaged under a tensile bending of 10 mm radius and 5 mm radius, respectively. The given ITO/PET sheet exhibits a greater resistance to long-term mechanical bending than the ITO/PEN one, which is attributed to the effect of stiffness and thickness of substrate. As the given PET substrate has a lower stiffness and thickness than the PEN one, ITO thin film in the ITO/PET sheet has a smaller stress given a bending radius. Consequently, a smaller extent of change in the electrical conductance of ITO thin film is found in the ITO/PET sheet.

Keywords: indium tin oxide; electrical conductance; static bending; polyethylene terephthalate; polyethylene naphthalate

1. Introduction Flexible organic optoelectronics has recently received much attention due to several advantages, such as flexibility, impact resistance, light weight, low cost, and possible roll-to-roll mass production. It is a promising subject for continuous investigation and development. Organic materials are commonly used as semiconductor components in flexible optoelectronics, thanks to their extremely flexible and ductile properties. Applications of organic semiconductor components in flexible electronics include organic light emitting diode (OLED), organic photovoltaic (OPV), and organic thin film transistor (OTFT) [1]. Flexible optoelectronics are expected to create more innovative applications in the future. In order to make highly efficient organic semiconductor components, transparent conductive layers are needed. The most common transparent conductive layers for flat panel displays, electroluminescent and electrochromic displays, OPV, and OLED are indium tin oxide (ITO) thin films [2]. They can be deposited by various techniques such as magnetron sputtering, reactive evaporation, radio frequency sputtering, spray pyrolysis, pulsed laser deposition, and screen printing process [3–5]. Manufacturing of ITO thin films needs precious raw materials and expensive tools, which limits their compatibility with mass production and low-cost devices [3]. Although alternatives for transparent conductive film are available, such as zinc-doped indium oxide, gallium-doped zinc oxide, and aluminum-doped zinc oxide [4], due to its excellent electrical characteristics and high

Coatings 2018, 8, 212; doi:10.3390/coatings8060212 www.mdpi.com/journal/coatings Coatings 2018, 8, 212 2 of 15 transparency, ITO is still the best candidate for the organic semiconductor components at present [6–11]. However, the natural brittle characteristic of ITO thin film results in damage after mechanical bending, and causes challenges for implementation in flexible organic optoelectronics. Therefore, its bending behavior needs to be studied for better applications in flexible optoelectronics. The most common chemical compositions considered for polymer substrate in flexible organic optoelectronics are polyethylene terephthalate (PET), polyethylene naphthalate (PEN), and polyethersulfone (PES). A material with a greater Young’s modulus is usually more brittle than a low one. Note that PET has a smaller Young’s modulus, such that it has better flexibility [12–14]. On the other hand, heat-resistance also needs to be considered in the roll-to-roll fabrication process of flexible electronics. PET has the lowest glass transition temperature (Tg) among the aforementioned three materials, while PES has the highest one [12–14]. Therefore, PES is more suitable for high-temperature manufacturing processes. The material properties of PEN are in-between, and it is also considered as a favorite substrate for flexible optoelectronics. In this study, PET and PEN are selected to study the substrate effect on the bending performance of ITO thin film. Currently, many studies have been conducted to find ways to improve conductive property and/or mechanical performance or to investigate failure mechanism of ITO thin film under bending [7,12,15–42]. Lim et al. [12] studied the flexural ability of ITO on PES, PET, and PEN substrates at different radii of curvature under cyclic bending [12]. Their results show that Young’s modulus of the substrate significantly affects the conductive stability of ITO electrode and mechanical properties of ITO/substrate film [12]. A cyclic bending test was also conducted to investigate the electrical conductivity of ITO thin film under various conditions of temperature and humidity [20,42]. Their results show that cyclic bending deformation is the dominant factor for the electrical resistance increase in those ITO films. That cyclic bending method was also applied to addressing the failure behavior of aluminum-doped zinc oxide thin film in harshly environmental conditions [43–46]. In addition to cyclic bending test, Hamasha et al. also investigated the conductivity behavior of ITO film under various stretching rates [47]. A three-point bending test combined with dynamic stiffness measurement was used to study deformation change and brittle failure mechanism of ITO thin film by Kirubanandham and Basu [15]. Their results show that bending causes buckling, delamination, and cracking in ITO thin film, and leads to an electrical resistance rise [15]. Lim et al. [38] studied the bending reliability of an ITO thin film on a 125 µm-thick PET substrate, by decreasing the bending radius, to define the critical bending radius. Electrical resistance of that ITO thin film was synchronously measured as the bending radius continued to decrease from 40 mm to 3 mm [38]. Their results show that the critical bending radii of 20, 50, 100, and 150 nm-thick ITO electrodes are 4.5, 4.5, 6, and 6 mm, respectively [38]. Leppänen et al. [40] investigated the mechanical and electrical behavior of an ITO thin film under static bending at different curvatures. Samples were bent from 53 mm to 15 mm radius of curvature [40]. Electrical resistance measurement and cracking observation were immediately conducted after 0, 2, and 24 h of bending [40]. The results show that cracks appear at 20 mm-radius cylinder and smaller, in conjunction with slight degradation in the conductivity of ITO thin film [40]. The exemplified results, described above, indicate that the electrical performance of ITO films is degraded under certain bending conditions. As flexible optoelectronic devices can be curved like paper in applications, the most attractive feature is their flexibility. For prolonged use of such devices, long-term failure mechanisms related to electrical characteristics of conductive layers of flexible organic optoelectronics should be considered. Although investigations on electrical conductivity of ITO thin films under bending have been conducted in previous studies, the change of electrical conductivity is mostly assessed after cyclic bending [12,15–17,20,22,31,42,48–51] or static bending [40]. However, it is necessary to study the concurrent change of electrical conductivity of ITO thin film under long-term mechanical bending, and its correlation with the mechanical failure mechanism. In our prior work [41], the effect of cyclic bending on the in situ variation of electrical conductance of ITO thin film has been investigated. In this study, changes of electrical conductance in ITO thin film on different polymer substrates are investigated by conducting long-term static bending tests with various radii of curvature. Particularly, Coatings 2018, 8, 212 3 of 15

Coatings 2018, 8, x FOR PEER REVIEW 3 of 15 electrical conductivity of the given ITO thin ﬁlms is monitored by in situ electrical resistance help assess lifetime of flexible optoelectronics and clarify the relevant failure mechanisms for long- measurement during static bending up to 1000 h. Hopefully, such results could help assess lifetime of term static bending. ﬂexible optoelectronics and clarify the relevant failure mechanisms for long-term static bending.

2.2. Experimental Procedures

2.1.2.1. Materials andand SpecimenSpecimen PreparationPreparation TwoTwo types of of ITO ITO thin thin film film are are investigated, investigated, one one using using PET PET as the as substrate the substrate and the and other the otherusing usingPEN. PEN. The ITO The ITOthin thin film film on on PET PET substrate substrate (ITO/PET) (ITO/PET) sheets sheets were were purchased purchased from from Win-Optical Win-Optical TechnologyTechnology Co., Co., Ltd., Ltd. (Tao-Yuan,, (Tao-Yuan, Taiwan). Taiwan). According Accord toing product to product specifications specifications [52], surface [52 resistance], surface ofresistance the ITO thinof the film ITO is 150thinΩ film/sq afteris 150 annealing Ω/sq after at 150annealing◦C for 90at min.150 °C The for ITO/PET 90 min. sheetsThe ITO/PET were received sheets inwere a form received of A4 in sheets a form with of A a4 thickness sheets with of 133 a thicknessµm. Rectangular of 133 μm. samples Rectangular in physical samples dimensions in physical of 36,dimensions 51, 83, 114, of 36, and 51, 145 83, mm 114, (length) and 145× mm10 mm (length) (width) × 10 are mm cut (width) out to conductare cut out static to conduct bending static tests atbending various tests radii at ofvarious curvature, radii namelyof curvature, 5, 10, namely 20, 30, and5, 10, 40 20, mm, 30, respectively.and 40 mm, respectively The ITO thin. The film ITO on PENthin substrate film on PEN(ITO/PEN) substrate sheets (ITO/PEN) were fabricated sheets by were Peccell fabricated Technologies, by Peccell Inc., (Yokohama,Technologies, Japan) Inc., and(Yokohama, purchased Japan) through and Ruilong purchased Co., Ltd., through (Miao-Li, Ruilong Taiwan). Co., According Ltd., (Miao to-Li, product Taiwan). specifications According [13 to], itsproduct surface specifications resistance is less[13] than, its surface 15 Ω/sq. resistance The as-fabricated is less than ITO/PEN 15 Ω/sq. sheets The were as-fabricated received inITO/PEN a form ofsheets 20 cm were× 25 received cm × 200 in aµ formm [13 of]. 20 Rectangular cm × 25 cm samples × 200 μm are [13 cut]. Rectangular out of the ITO/PEN samples are sheets cut out for of static the bendingITO/PEN tests sheets and for have static the bending same dimensions tests and have of the the ITO/PET same dimensions samples. of the ITO/PET samples.

2.2.2.2. Static Bending Test ForFor investigatinginvestigating thethe effecteffect ofof long-termlong-term staticstatic bendingbending onon electricalelectrical conductanceconductance ofof thethe givengiven ITOITO thinthin films,films, aa fixturefixture ofof staticstatic bendingbending testtest isis designeddesigned andand made in-house,in-house, as shown in Figure 11.. EachEach ITOITO thinthin filmfilm samplesample isis firstlyfirstly clampedclamped byby twotwo stainlessstainless steelsteel clipsclips atat bothboth ends,ends, asas shownshown inin FigureFigure1 1.. TheThe twotwo clipsclips are are then then fixed fixed on on a a flat flat plate. plate. As As the the sample sample is is bent bent in in a a concave concave downward downward positionposition (Figure(Figure1 1),), ITO ITO thin thin layer layer is is under under tension tension when when placed placed at at the the top top position. position. ITO ITO thin thin layer layer is is subjectedsubjected toto compressivecompressive stressstress whenwhen itit isis placedplaced atat thethe bottombottom ofof thethe sample.sample. FiveFive radiiradii ofof curvature (5,(5, 10,10, 20,20, 30, 30, and and 40 40 mm) mm) are are applied applied for for static static bending bending tests, tests and, and the specimensthe specimens are continuously are continuously bent untilbent 1000 until h.1000 During h. During the bending the bending test, all the test, specimens all the specimens are placed are in a placed closed inbox a at closed room box temperature at room totemperature avoid dust to accumulating avoid dust accumulat on the specimens.ing on the specimens.

FigureFigure 1. 1. Schematic diagram of static bending test setup and cross cross sectionsection of of samplesample structurestructure of of indiumindium tintin oxideoxide (ITO)(ITO) thinthin ﬁlm.film.

During the static bending test, the electrical resistance of ITO film is monitored using a source During the static bending test, the electrical resistance of ITO ﬁlm is monitored using a measurement unit (SMU-Keithley 2400, Keithley Instruments, Inc., Cleveland, OH, USA). For source measurement unit (SMU-Keithley 2400, Keithley Instruments, Inc., Cleveland, OH, USA). measuring electrical resistance of several ITO film samples at the same time, the source measurement For measuring electrical resistance of several ITO ﬁlm samples at the same time, the source unit is combined with a homemade switch system controlled by a LabVIEW program, as shown in measurement unit is combined with a homemade switch system controlled by a LabVIEW program, Figure 2. This homemade switch system, which consists of a relay module and a microcontroller as shown in Figure2. This homemade switch system, which consists of a relay module and a (Arduino Mega 2560, Arduino LLC, Turin, Piedmont, Italy), is in contact with an array of eight ITO microcontroller (Arduino Mega 2560, Arduino LLC, Turin, Piedmont, Italy), is in contact with an array specimens through electrical wires. The electrical wires and relay contacts are used as electrical current entrances to measure the electrical resistance, thanks to their high conductivity and negligible resistance. During the measurement, the electrical current is continuously recorded by the given

Coatings 2018, 8, 212 4 of 15 of eight ITO specimens through electrical wires. The electrical wires and relay contacts are used as electrical current entrances to measure the electrical resistance, thanks to their high conductivity and Coatings 2018, 8, x FOR PEER REVIEW 4 of 15 negligible resistance. During the measurement, the electrical current is continuously recorded by the givensource source measurement measurement unit under unit undera constant a constant source source voltage voltage of 20 mV. of 20 The mV. resistance The resistance change change of each of eachITO sample ITO sample is then is thenin situ in determined situ determined throughout throughout the bending the bending test. test.

Figure 2.2. Schematic diagram of in situ measurement of electrical resistance during static bending test of ITO thin ﬁlm.film.

2.3. Microstructural Analysis In order to to determine determine coating coating thickness thickness of of the the ITO ITO thin thin films films studied, studied, some some specimens specimens are arecut cutalong along the thickness the thickness direction direction to observe to observe the cross the-sectional cross-sectional microstructure microstructure using a usingscanning a scanning electron electronmicroscope microscope (SEM) (Hitachi (SEM) (Hitachi S-800, S-800, Hitachi, Hitachi, Ltd., Ltd., Tokyo, Tokyo, Japan). Japan). Pt Ptfilms films are are deposited deposited on on the specimens toto preventprevent accumulation accumulation of of electrostatic electrostatic charge charge at theat the surface. surface. After After mechanical mechanical bending bending test, fracturetest, fracture surfaces surfaces of the ITO/PET of the ITO/PET and ITO/PEN and ITO/PEN specimens specimens are observed are using observed an optical using microscope an optical (OM)microscope and SEM (OM) to analyzeand SEM the to failure analyze mechanism. the failure In mechanism. addition, the In surface addition, morphology the surface and morphology roughness profilesand roughness of ITO/PEN profiles and of ITO/PET ITO/PEN sheets and ITO/PET in as-received sheets and in as after-bending-received and states after are-bending characterized states are by atomiccharacterized force microscopy by atomic (AFM) force (XE microscopy 7, Park Systems, (AFM) Suwon,(XE 7, ParkKorea). Systems, The microstructural Suwon, Korea). changes, The interfacialmicrostructural cracks, changes, and bonding interfacial defects cracks, observed and are bonding correlated defects with observed the electrical are resistancecorrelated change with the of theelectrical mechanically resistance bent change ITO/PET of the and mechanically ITO/PEN samples.bent ITO/PET and ITO/PEN samples.

3. Results and Discussion

3.1.3.1. Bending Effect on ITO Thin Film The electricalelectrical resistanceresistance change change of of ITO/PEN ITO/PEN and and ITO/PET ITO/PET sheets sheets with with time time under under static static bending bending at variousat various radii radii of curvatureof curvature are shownare shown in Figures in Figures3 and 34 and, respectively. 4, respectively. Two specimens Two specimens are repeatedly are repeatedly tested fortested a given for a bending given bending condition condition to check to the check repeatability the repeatability of data. In of addition, data. In eachaddition, figure each shows figure the resultsshows ofthe four results samples of four bent samples at a similar bent radiusat a similar of curvature, radius of namely curvature, two under namely tension two under (Tensile tension 1 and 2)(Tensile and the 1 otherand 2) two and under the other compression two under (Compressive compression 1 and (Compressive 2). Note that in 1 Figuresand 2).3 Noteand4 ,that change in Figures of the resistance 3 and 4, ischange normalized of the resistance by its original is normalized value and by plotted its original in the ordinate.value and As plotted shown in inthe Figures ordinate.3 and As4 ,shown a slight in reductionFigures 3 and of electrical 4, a slight resistance reduction of ITOof electrical thin film resistance is generally of ITO observed thin film at the is beginninggenerally observed of compressive at the bendingbeginning for of both compressive ITO/PEN bending and ITO/PET for both sheets. ITO/PEN The and initial ITO/PET reduction sheets. of electrical The initial resistance reduction under of compressiveelectrical resistance bending under for ITO/PEN compressive sheet isbending about 20% for ITO/PEN and 2% at sheet a 10 mm is about and a 20% 20 mm and curvature 2% at a 10 radius, mm respectively.and a 20 mm For ITO/PET curvature sheet, radius, the initial respectively. reduction For under ITO/PET compressive sheet, bending the initial is about reduction 50%, 7%, and under 1% compressive bending is about 50%, 7%, and 1% at a radius of 5 mm, 10 mm, and 20 mm, respectively. For a change from neutral state to compressive state, a reduction of electrical resistivity occurs because of a net decrement in the space between ITO aggregates, which produces more conduction paths [53,54]. When the ITO/PET sheet is bent in tension, a slight reduction of electrical resistance is also observed at the beginning for bending radius of 10 mm or 20 mm. The extent of initial reduction

Coatings 2018, 8, 212 5 of 15 at a radius of 5 mm, 10 mm, and 20 mm, respectively. For a change from neutral state to compressive state, a reduction of electrical resistivity occurs because of a net decrement in the space between ITO aggregates, which produces more conduction paths [53,54]. When the ITO/PET sheet is bent in tension, a slight reduction of electrical resistance is also observed at the beginning for bending radius of 10 mm or 20 mm. The extent of initial reduction is about 6% and 1% for the 10 mm and 20 mm curvature radius, respectively. This isCoatings attributable 2018, 8, x toFOR the PEER fact REVIEW thata rotation of ITO aggregates may cause additional connections5 of 15of ITO aggregatesis about in 6% response and 1% to for the the longitudinal 10 mm and 20 expansion mm curvature of specimen radius, respectively. under tension. This is This attributable phenomenon to is calledthe an orientation fact that a rotation effect of of non-uniformly ITO aggregates spaced may cause aggregates additional of nanostructured connections of substanceITO aggregates [53]. inA lower electricalresponse resistance to the in longitudinal a compressive expansion or tensile of specimen state than under that tension. in a neutral This pheno statemenon is also is observed called an for the given ITOorientation thin films effect under of non cyclic-uniformly bending spaced in our aggregates previous of study nanostructured [41]. substance [53]. A lower Aselectrical shown resistance in Figure in3 aa, compressive the electrical or tensile resistance state than changes that in of a ITO/PENneutral state under is also bothobserve tensiled for and compressivethe given bending ITO thin withfilms under a 40 mmcyclic curvature bending in radiusour previous are about study 2%[41].after 1000 h of static bending. The electricalAs shown resistance in Figure change 3a, the under electrical 1000 resistance h of tensile changes or of compressive ITO/PEN under bending both tensile with anda 30 mm compressive bending with a 40 mm curvature radius are about 2% after 1000 h of static bending. The curvature radius is also about 2% (Figure3b). No signiﬁcantly detrimental effect on the electrical electrical resistance change under 1000 h of tensile or compressive bending with a 30 mm curvature conductanceradius is of also ITO/PEN about 2% sheet(Figure is 3 foundb). No significantly for bending detrimental with a radius effect of on curvature the electrical of 30conductance mm or 40 mm. As shownof ITO/PEN in Figure sheet3c, is the found electrical for bending resistance with changea radius graduallyof curvature increases of 30 mm with or 40 time mm. under As shown both in tensile and compressiveFigure 3c, the bending electrical with resistance a 20 mm change curvature gradually radius. increases Compared with time with under the both initial tensile reduction and of electricalcompressive resistance bending at beginning with a 20 of bending,mm curvature the electrical radius. Compared resistance with changes the initial after reduction 1000 h of of tensile and compressiveelectrical resistance bending at beginning are about of 3–5% bending, and the 3%, electrical respectively. resistance The changes electrical after resistance 1000 h of oftensile ITO/PEN signiﬁcantlyand compressive increases bending with time are about under 3– tensile5% and bending 3%, respectively. with a 10The mm electrical curvature resistance radius, of ITO/PEN and a change significantly increases with time under tensile bending with a 10 mm curvature radius, and a change of 800–900% is found after 1000 h of bending (Figure3d). However, for a similar bending radius of of 800–900% is found after 1000 h of bending (Figure 3d). However, for a similar bending radius of 10 mm,10 the mm, electrical the electrical resistance resistance of of ITO/PEN ITO/PEN onlyonly increases increases by by 3% 3% after after 1000 1000 h of hcompressive of compressive bending. bending.

(a) (b)

Figure 3. Electrical resistance change of ITO/PEN (polyethylene naphthalate) sheet under static Figure 3. Electrical resistance change of ITO/PEN (polyethylene naphthalate) sheet under static bending of various radii: (a) 40 mm; (b) 30 mm; (c) 20 mm; and (d) 10 mm. bending of various radii: (a) 40 mm; (b) 30 mm; (c) 20 mm; and (d) 10 mm. An overallAn overall comparison comparison of of the the results results presented presented in in Figure Figure 33 indicates indicates that that a detrimental a detrimental tensile tensile bendingbending effect effect is found is found for for a curvature a curvature of of 20 20 mmmm and smaller. smaller. In In particular, particular, as the as bending the bending radius radius is is decreaseddecreased to 10 to mm, 10 mm, the the electrical electrical resistance resistance signiﬁcantlysignificantly increases increases to toan anextent extent of about of about 500% 500%at the at the beginning, and 800–900% after 1000 h. On the other hand, a slight effect of compressive bending is beginning, and 800–900% after 1000 h. On the other hand, a slight effect of compressive bending is found for bending radius of 20 mm and smaller. The electrical resistance slightly decreases by 2% at the beginning, and increases by 3% after 1000 h of bending with a radius of 20 mm. When the bending radius is reduced to 10 mm for more severe bending, no further detrimental effect is observed, and

Coatings 2018, 8, 212 6 of 15 found for bending radius of 20 mm and smaller. The electrical resistance slightly decreases by 2% at the beginning, and increases by 3% after 1000 h of bending with a radius of 20 mm. When the bending Coatings 2018, 8, x FOR PEER REVIEW 6 of 15 radius is reduced to 10 mm for more severe bending, no further detrimental effect is observed, and the electricalthe electrical resistance resistance change change is similar is similar to thatto that of 20of mm.20 mm. These These results results reveal reveal that that tensile tensile bending bending is muchis much more more detrimental detrimental than than compressive compressive bending bending to the electrical to the electrical conductance conductance of the given of theITO/PEN given sheetITO/PEN under sheet a long-term under a long static-te flexuralrm static deformation. flexural deformation. As shownshown inin FigureFigure4 4a,b,a,b, the the electrical electrical resistance resistance changes changes of of ITO/PET ITO/PET underunder bothboth tensiletensile andand compressive bending with 30 mm and 20 mm curvature radii are about 2% after 1000 h of bending. Therefore, nono significantlysignificantly detrimentaldetrimental effecteffect onon thethe electricalelectrical conductanceconductance ofof ITO/PETITO/PET sheet is found for a bending radius ofof 20 mm or 30 mm. As As shown shown in in Figure Figure 44c,c, thethe electricalelectrical resistanceresistance changechange underunder tensile bending with with a a 10 10 mm mm curvature curvature radius radius starts starts to to increase increase at at100 100 h, h,and and reaches reaches a change a change of 6 of– 6–20%20% at at1000 1000 h. h.For For a bending a bending radius radius of of 5 5mm, mm, the the electrical electrical resistance resistance significantly significantly increases increases from the beginning of tensile bending,bending, and has a change of 60,000% after 1000 h (Figure(Figure4 4d).d). However,However, forfor aa compressive bending state with a radius of curvature of 10 mm or 5 mm,mm, nono significantsignificant change of electrical resistance isis found,found, asas shownshown inin FigureFigure4 4c,d.c,d. As As the the ITO/PET ITO/PET sheetsheet isis seriouslyseriously damageddamaged atat the beginning of tensile bending with a 5 mm curvature radius in a repeated test, no long-termlong-term static bending is conducted onon thisthis repeatedrepeated sample.sample.

(a) (b)

Figure 4. 4. Electrical resistance change of of ITO/PET ITO/PET ( (polyethylenepolyethylene terephthalateterephthalate)) sheet under static static bending of various radii: (a) 3030 mm;mm; ((b)) 2020 mm;mm; ((cc)) 1010 mm;mm; andand ((dd)) 55 mm.mm.

Comparison of the electrical resistance changes of ITO/PET sheet presented in Figure 4 indicates Comparison of the electrical resistance changes of ITO/PET sheet presented in Figure4 indicates that a detrimental tensile bending effect is found for curvature radius of 10 mm and smaller. In that a detrimental tensile bending effect is found for curvature radius of 10 mm and smaller. particular, for a bending radius of 5 mm, tensile bending increases, extremely, the electrical resistance In particular, for a bending radius of 5 mm, tensile bending increases, extremely, the electrical resistance to an extent of 50,000% at beginning, and 60,000% after 1000 h of bending. Such a great change of to an extent of 50,000% at beginning, and 60,000% after 1000 h of bending. Such a great change of electrical resistance makes the damaged ITO/PET sheet almost nonconductive. On the contrary, the electrical resistance makes the damaged ITO/PET sheet almost nonconductive. On the contrary, electrical conductance of the ITO/PET sheet remains stable for all the compressive states of long-term static bending. This indicates, again, that tensile bending is much more detrimental than compressive bending to the electrical conductance of the given ITO/PET sheet under a long-term static flexural deformation. In comparison of ITO/PEN and ITO/PET sheets under tensile bending (Figures 3 and 4), electrical resistance of the ITO/PEN sheet slightly increases at a bending radius of 20 mm, but the ITO/PET

Coatings 2018, 8, 212 7 of 15 the electrical conductance of the ITO/PET sheet remains stable for all the compressive states of long-term static bending. This indicates, again, that tensile bending is much more detrimental than compressive bending to the electrical conductance of the given ITO/PET sheet under a long-term static flexural deformation. In comparison of ITO/PEN and ITO/PET sheets under tensile bending (Figures3 and4), electrical resistance of the ITO/PEN sheet slightly increases at a bending radius of 20 mm, but the ITO/PET sheet remains stable up to 1000 h. When the bending radius is reduced to 10 mm for moreCoatings severe 2018, 8, x bending, FOR PEER REVIEW electrical resistance of the ITO/PEN sheet significantly increases,7 of 15 but electrical resistance of the ITO/PET sheet just slightly increases. For a counterpart comparison sheet remains stable up to 1000 h. When the bending radius is reduced to 10 mm for more severe under compressive bending, these two ITO sheets show no significant change in electrical resistance at bending, electrical resistance of the ITO/PEN sheet significantly increases, but electrical resistance of a bendingthe ITO/PET radius sheet of 30 just mm slightly or larger. increases. For more For a severecounterpart bending comparison of 20 mm under and compressive 10 mm radii, bending, electrical resistancethese two of the ITO ITO/PET sheets show sheet no significant is still stable change under in electrical compressive resistance bending at a bending after 1000 radius h, of but 30 electricalmm resistanceor larger. of theFor ITO/PENmore severe sheetbending slightly of 20 mm increases. and 10 mm Based radii, on electrical the comparisons resistance of described the ITO/PET above, the givensheet ITO/PETis still stable sheet under exhibits compressive a greater bending resistance after 1000 to long-term h, but electrical static resistance bending of than the theITO/PEN ITO/PEN sheetsheet in terms slightly of changeincreases. in Based electrical on the conductance. comparisons described above, the given ITO/PET sheet exhibits a greater resistance to long-term static bending than the ITO/PEN sheet in terms of change in electrical 3.2. Failureconductance. Analysis

As3.2.described Failure Analysis in Section 3.1, given a bending radius of 10 mm, the electrical resistance of ITO/PEN sheet increases significantly by 800–900% under tensile bending, and increases only by 3% under As described in Section 3.1, given a bending radius of 10 mm, the electrical resistance of ITO/PEN compressive bending after 1000 h. OM micrographs of ITO/PEN sheet after 1000 h of compressive and sheet increases significantly by 800–900% under tensile bending, and increases only by 3% under tensilecompressive bending with bending a 10 after mm curvature1000 h. OM radius micrographs are shown of ITO/PEN in Figures sheet5 andafter6 ,1000 respectively. h of compressive Cracks are foundand in tensile both specimens,bending with and a 10 mm are presumablycurvature radius initiated are shown at thein Figures edge of5 and the 6, specimen,respectively. as Cracks shown in Figuresare5 found and6 .in However,both specimens cracks, and develop are presumably on the PENinitiated substrate at the edge when of the ITO specimen, is under as compressive shown in bending,Figures and 5 on and the 6. ITOHowever, thin film cracks when develop ITO is on under the PEN tensile substrate bending, when as ITO shown is under in Figures compressive5a and 6b, respectively.bending, and As shownon the ITO in thin Figure film6 b,when wrinkles ITO is under are also tensile found bending, in the as shown ITO/PEN in Figures sheet 5a after and 6b, tensile bending.respectively. Existence As of shown wrinkles in Figure is attributed 6b, wrinkles to shrinkage are also found deformation in the ITO/PEN caused bysheet Poisson’s after tensile effect in the underlyingbending. Existence film [55 of]. Aswrinkles deformability is attributed of theto shrinkage underlying deformation film is much caused higher by Poisson’s than that effect of the in ITO film,the part underlying of the ITO film film [55 is]. wrinkledAs deformability up [55 ].ofThese the underlying results reveal film is thatmuch static higher bending than that of of 10 the mm ITO radius film, part of the ITO film is wrinkled up [55]. These results reveal that static bending of 10 mm radius seriously damages the ITO/PEN sheet in both compressive and tensile states, but the ITO thin film is seriously damages the ITO/PEN sheet in both compressive and tensile states, but the ITO thin film is damageddama onlyged only under under tensile tensile bending. bending.

(a) (b)

Figure 5. OM micrographs of ITO/PEN sheet after compressive bending with a 10 mm curvature Figure 5. OM micrographs of ITO/PEN sheet after compressive bending with a 10 mm curvature radius for 1000 h: (a) viewed from PEN side; (b) viewed from ITO side. radius for 1000 h: (a) viewed from PEN side; (b) viewed from ITO side. For a larger bending radius of 20 mm, the electrical resistance of ITO/PEN sheet increases by 3– For15% aunder larger tensile bending bending radius of 1000 of 20h. An mm, OM the micrograph electrical of resistance ITO/PEN ofin this ITO/PEN bending sheet case is increases shown by 3–15%in underFigure tensile7. Cracks bending of a smaller of 1000 size h. are An only OM found micrograph at edge of of the ITO/PEN specimen, in as this shown bending in Figure case 7 is. As shown in Figuredescribed7. Cracks above, of tensile a smaller bending size of are ITO/PEN only found at a radiusat edge of of 20 the mm specimen, exhibits a assmaller shown change in Figure of 7. As describedresistance above, (3–15%) tensile in comparison bending of with ITO/PEN that of at a a 10 radius mm of radius 20 mm bending exhibits (800 a– smaller900%). These change of fractography observations are consistent with the electrical resistance measurements. An exemplified resistance (3–15%) in comparison with that of a 10 mm radius bending (800–900%). These fractography SEM micrograph of ITO/PEN sheet after 1000 h of tensile bending is shown in Figure 8. Cracks are found in the ITO thin film of this sample. These cracks are perpendicular to the direction of tensile stress, roughly parallel to each other, and appear as straight lines. They could degrade the electrical conductance of ITO thin film. It is confirmed, again, that cracks are formed to deteriorate the ITO/PEN sheet’s electrical performance.

Coatings 2018, 8, 212 8 of 15 observations are consistent with the electrical resistance measurements. An exemplified SEM micrograph of ITO/PEN sheet after 1000 h of tensile bending is shown in Figure8. Cracks are found in the ITO thin film of this sample. These cracks are perpendicular to the direction of tensile stress, roughly parallel to each other, and appear as straight lines. They could degrade the electrical conductance of ITO thin film. It is confirmed, again, that cracks are formed to deteriorate the ITO/PEN sheet’sCoatings electrical 2018, 8 performance., x FOR PEER REVIEW 8 of 15

Coatings 2018, 8, x FOR PEER REVIEW 8 of 15 Coatings 2018, 8, x FOR PEER REVIEW 8 of 15

(a) (b)

Figure 6. OM micrographs of ITO/PEN sheet after tensile bending with a 10 mm curvature radius for Figure 6. OM micrographs(a of) ITO/PEN sheet after tensile bending with a(b 10) mm curvature radius for 1000 h: (a) viewed from PEN side; (b) viewed from ITO side. (a) (b) 1000 h:Figure (a) viewed 6. OM micrographs from PEN side; of ITO/PEN (b) viewed sheet fromafter tensile ITO side. bending with a 10 mm curvature radius for 1000Figure h: 6(a. )OM viewed micrographs from PEN of side;ITO/PEN (b) viewed sheet after from tensile ITO side. bending with a 10 mm curvature radius for 1000 h: (a) viewed from PEN side; (b) viewed from ITO side.

Figure 7. OM micrograph of ITO/PEN sheet after tensile bending with a 20 mm curvature radius for 1000 h, as viewed from the ITO side. Figure 7. OM micrograph of ITO/PEN sheet after tensile bending with a 20 mm curvature radius for Figure 7. OM micrograph of ITO/PEN sheet after tensile bending with a 20 mm curvature radius for 1000Figure h, 7.as OM viewed micrograph from the of ITO ITO/PEN side. sheet after tensile bending with a 20 mm curvature radius for 1000 h, as viewed from the ITO side. 1000 h, as viewed from the ITO side.

Figure 8. SEM micrograph of ITO/PEN sheet after tensile bending with a 10 mm curvature radius for 1000 h, as viewed from the ITO side. Figure 8. SEM micrograph of ITO/PEN sheet after tensile bending with a 10 mm curvature radius for For the worst case in ITO/PET sheets, the electrical resistance significantly increases from the 1000Figure h, 8.as SEM viewed micrograph from the ofITO ITO/PEN side. sheet after tensile bending with a 10 mm curvature radius for Figurebeginning 8. SEM of micrographtensile bending of ITO/PEN up to a change sheet after of 60,000% tensile bendingat a 5 mm with radius. a 10 mmAn OM curvature micrograph radius of for 1000 h, as viewed from the ITO side. 1000ITO/PET h,For as viewedthe sheet worst after from case 1000 the in ITOhITO/PET of side.tensile sheet bendings, the with electrical a 5 mm resistance curvature significantly radius is shown increases in Figurefrom the 9. beginningSimilarFor tothe of the worsttensile ITO/PEN case bending in sheet, ITO/PET up cracks to asheet change ands, the wrinkles of electrical 60,000% are resistanceat found a 5 mm in thesignificantly radius. ITO/PET An OM sheetincreases micrograph after f rom tensile the of bending, as initiated at the edge of the specimen. For a larger bending radius in ITO/PET sheet, the ForITO/PETbeginning the worst sheet of tensile caseafter in1000bending ITO/PET h of up tensile to sheets, a bendingchange the of with electrical 60,000% a 5 mm at resistance acurvature 5 mm radius. signiﬁcantlyradius An is OMshown micrograph increases in Figure from of9. the failure mechanism is similar to that described above. An exemplified SEM micrograph of ITO/PET SimilarITO/PET to sheet the IafterTO/PEN 1000 sheet,h of tensile cracks bending and wrinkles with a are5 mm found curvature in the radius ITO/PET is shown sheet afterin Figure tensile 9. beginningsheet ofafter tensile 1000 h bending of tensile upbending to a with change a 5 mm of 60,000%curvature atradius a 5 mmis shown radius. in Figure An OM10. Cracks micrograph and of bending,Similar to as the initiated ITO/PEN at the sheet, edge cracksof the andspecimen. wrinkles For area larger found bending in the radius ITO/PET in ITO/PET sheet after sheet, tensile the wrinkles are again found in the ITO thin film of this sample. These straight cracks are perpendicular failurebending, mechanism as initiated is atsimilar the edge to that of the described specimen. above. For aAn larger exemplified bending SEM radius mic inrograph ITO/PET of sheet,ITO/PET the to the loading direction and parallel to each other. sheetfailure after mechanism 1000 h of is tensile similar bending to that withdescribed a 5 mm above. curvature An exemplified radius is shown SEM inmic Figurerograph 10. ofCracks ITO/PET and wrinklessheet after are 1000 again h of found tensile in bending the ITO withthin filma 5 mm of this curvature sample. radius These isstraight shown cracks in Figure are 10perpen. Cracksdicular and towrinkles the loading are again direct foundion and in theparallel ITO thinto each film other. of this sample. These straight cracks are perpendicular to the loading direction and parallel to each other.

Coatings 2018, 8, 212 9 of 15

ITO/PET sheet after 1000 h of tensile bending with a 5 mm curvature radius is shown in Figure9. Similar to the ITO/PEN sheet, cracks and wrinkles are found in the ITO/PET sheet after tensile bending, as initiated at the edge of the specimen. For a larger bending radius in ITO/PET sheet, the failure mechanism is similar to that described above. An exempliﬁed SEM micrograph of ITO/PET sheet after 1000 h of tensile bending with a 5 mm curvature radius is shown in Figure 10. Cracks and wrinkles are again found in the ITO thin ﬁlm of this sample. These straight cracks are perpendicular to Coatingsthe loading 2018, 8 direction, x FOR PEER and REVIEW parallel to each other. 9 of 15 Coatings 2018, 8, x FOR PEER REVIEW 9 of 15

Figure 9. OM micrograph of ITO/PET sheet after tensile bending with a 5 mm curvature radius for Figure 9. OM micrograph of ITO/PET sheet after tensile bending with a 5 mm curvature radius for 1000Figure h, 9.as OMviewed micrograph from the ofITO ITO/PET side. sheet after tensile bending with a 5 mm curvature radius for 1000 h,h, asas viewedviewed fromfrom thethe ITOITO side.side.

Figure 10. SEM micrograph of ITO/PET sheet after tensile bending with a 5 mm curvature radius for Figure 10. SEM micrograph of ITO/PET sheet after tensile bending with a 5 mm curvature radius for 1000Figure h, 10.as viewedSEM micrograph from the ITO of ITO/PET side. sheet after tensile bending with a 5 mm curvature radius for 1000 h, as viewed from the ITO side. 1000 h, as viewed from the ITO side. Moreover, surface morphology of the ITO/PEN and ITO/PET sheets after bending is also examinedMoreover, by AFM surface, and shown morphology in Figures of the 11 ITO/PENand 12. The and surface ITO/PET roughness sheets profiles after bending along selected is also Moreover, surface morphology of the ITO/PEN and ITO/PET sheets after bending is also scanningexamined paths by AFM in the, and investigated shown in areasFigures are 11 also and shown 12. The in Figuressurface 11roughness and 12. Ifprofiles cracks arealong present selected on examined by AFM, and shown in Figures 11 and 12. The surface roughness proﬁles along selected thescanning outer pathssurface in ofthe ITO investigated layer, it is areas expected are also that shown certain in changes Figures in11 the and surface 12. If cracks morphology are present would on scanning paths in the investigated areas are also shown in Figures 11 and 12. If cracks are present on bethe observedouter surface through of ITO AFM. layer, Therefore, it is expected scanning that certain paths i nchanges the AFM in the are surface selected morphology to pass across would an the outer surface of ITO layer, it is expected that certain changes in the surface morphology would be apparentbe observed crack through which is AFM. visible Therefore, by OM. Although scanning the paths selected in the scanning AFM are paths selected in the to AFM pass pass across across an observed through AFM. Therefore, scanning paths in the AFM are selected to pass across an apparent anapparent apparent crack crack, which both is visibleFigure bys 11b OM. and Although 12b reveal the selectedthat the scanningITO outer paths surface in the has AFM no remarkablepass across crack which is visible by OM. Although the selected scanning paths in the AFM pass across an apparent changean apparent in the crack, roughn bothess Figure profiles 11b along and the 12b scanning reveal that line. the Note ITO that outer the surface ITO layer has isno transparent.remarkable crack, both Figures 11b and 12b reveal that the ITO outer surface has no remarkable change in the Therefore,change in the cracks roughn observedess profile in alongOM might the scanning be initiated line. at Notethe interface that the of ITO ITO/PET layer isand transparent. ITO/PEN roughness proﬁle along the scanning line. Note that the ITO layer is transparent. Therefore, the cracks sheetsTherefore,, instead the cracksof at theobserved outer surfacein OM might of ITO be layer. initiated The at AFM the resultsinterface indicate of ITO/PET that cracksand ITO/PEN are not observed in OM might be initiated at the interface of ITO/PET and ITO/PEN sheets, instead of at initiatedsheets, instead at the ITOof at outer the outersurface. surface A similar of ITO observation layer. The has AFM been results reported indicate by Li that and cracks Lin [ 16 are] th notat the outer surface of ITO layer. The AFM results indicate that cracks are not initiated at the ITO outer cracksinitiated are at preferentially the ITO outer initiated surface. at A the similar interface observation between hasITO beenlayer reportedand substrate by Li during and Lin bending. [16] th at surface. A similar observation has been reported by Li and Lin [16] that cracks are preferentially cracksCross are -preferentiallysectional SEM initiated micrographs at the ofinterface ITO/PEN between and ITO/PET ITO layer sheets and substrateare shown during in Figure bending. 13. As initiated at the interface between ITO layer and substrate during bending. shownCross in Figure-sectional 13, SEMthe ITO micrographs layer thickness of ITO/PEN is about and 200 ITO/PET–250 nm sheets for both are cases.shown According in Figure 13to . theAs productshown in specifications, Figure 13, the the ITO overall layer thickness thickness of is the about ITO/PEN 200–250 and nm ITO/PET for both sheets cases. is According 200 μm [13 to] and the 133product μm [specifications,52], respectively. the Asoverall the thicknessthickness of theITO ITO/PEN layer is comparableand ITO/PET between sheets isthe 200 ITO/PEN μm [13] and ITO/PET133 μm [ 52sheets,], respectively. the bending As effect the thickness on the ITO of thinITO filmlayer is is also comparable related to between the substrate the ITO/PEN structure. and In thisITO/PET regard, sheets, stiffness the bendingand thickness effect ofon thethe substrateITO thin alsofilm playis also a rolerelated in theto thebending substrate behavior structure. of IT InO thinthis regard,film. As stiffnesslisted in andTable thickness 1, the Young of the’s modulussubstrate of also PEN play is greater a role inthan the that bending of PET. behavior Moreover, of IT theO substratethin film. As thickness listed in in Table the ITO/PEN1, the Young sheet’s modulus is also greaterof PEN is than greater that than in the that ITO/PET of PET. Moreover, sheet. As the Youngsubstrate’s modulus thickness of inPEN the is ITO/PEN greater than sheet that is alsoof PET, greater it is thanexpected that that in the flexibility ITO/PET of sheet.PET is Asbetter the thanYoung that’s modulusof PEN. of PEN is greater than that of PET, it is expected that flexibility of PET is better than that of PEN.

Coatings 2018, 8, 212 10 of 15

Cross-sectional SEM micrographs of ITO/PEN and ITO/PET sheets are shown in Figure 13. As shown in Figure 13, the ITO layer thickness is about 200–250 nm for both cases. According to the product specifications, the overall thickness of the ITO/PEN and ITO/PET sheets is 200 µm [13] and 133 µm [52], respectively. As the thickness of ITO layer is comparable between the ITO/PEN and ITO/PET sheets, the bending effect on the ITO thin film is also related to the substrate structure. In this regard, stiffness and thickness of the substrate also play a role in the bending behavior of ITO thin film. As listed in Table1, the Young’s modulus of PEN is greater than that of PET. Moreover, the substrate thickness in the ITO/PEN sheet is also greater than that in the ITO/PET sheet. As the Young’s modulus of PEN is greater than that of PET, it is expected that flexibility of PET is better than Coatingsthat of 2018 PEN., 8, x FOR PEER REVIEW 10 of 15 Coatings 2018, 8, x FOR PEER REVIEW 10 of 15

(a) (b) (a) (b) Figure 11. AFM images of ITO/PEN sheet: (a) as-received; (b) after tensile bending with a 10 mm Figure 11.11. AFM images ofof ITO/PENITO/PEN sheet: sheet: ( (aa)) as as-received;-received; ( (bb)) after after tensile tensile bending with with a 10 mm curvature radius for 1000 h.h. curvature radius for 1000 h.

(a) (b) (a) (b) Figure 12. AFM images of ITO/PET sheet: (a) as-received; (b) after tensile bending with a 5 mm Figure 12. AFM images of ITO/PET sheet: (a) as-received; (b) after tensile bending with a 5 mm curvatureFigure 12. radiusAFM for images 1000 ofh. ITO/PET sheet: (a) as-received; (b) after tensile bending with a 5 mm curvature radius forfor 10001000 h.h.

(a) (b) (a) (b) Figure 13. Cross-sectional SEM micrographs of (a) ITO/PEN sheet and (b) ITO/PET sheet. Figure 13. Cross-sectional SEM micrographs of (a) ITO/PEN sheet and (b) ITO/PET sheet.

Coatings 2018, 8, x FOR PEER REVIEW 10 of 15

(a) (b)

Figure 11. AFM images of ITO/PEN sheet: (a) as-received; (b) after tensile bending with a 10 mm

curvature radius for 1000 h.

(a) (b)

CoatingsFigure2018, 8 12., 212 AFM images of ITO/PET sheet: (a) as-received; (b) after tensile bending with a 5 mm11 of 15 curvature radius for 1000 h.

(a) (b)

FigureFigure 13.13. Cross-sectionalCross-sectional SEMSEM micrographsmicrographs ofof ((aa)) ITO/PEN ITO/PEN sheet andand ((bb)) ITO/PETITO/PET sheet. sheet.

Table 1. Material properties and parameters used in the 3D FEA modeling [12–14].

Material Young’s Modulus (GPa) Poison Ratio Thickness (µm) ITO 116 0.35 0.25 PET 3.0 0.37 133 PEN 6.1 0.38 200

In order to correlate the flexible deformation with bending stress, a three-dimensional (3D) finite element method (FEM) model is established to quantitatively determine the corresponding stress values in the bending test. The material properties and thicknesses of thin films used in the FEM modeling are listed in Table1. Distributions of the maximum principal stress ( σ1) in the ITO/PET and ITO/PEN sheets are then obtained for a similar bending radius of 10 mm, as shown in Figure 14. As shown in Figure 14, the ITO/PET or ITO/PET specimen is bent in a concave downward position, and the ITO thin layer on the top is under tension. In the comparison of ITO/PEN and ITO/PET sheets under tensile bending at a bending radius of 10 mm, σ1 in the ITO/PEN and ITO/PET sheets is 1309 MPa and 857 MPa, respectively (Figure 14). Note that the critical tensile fracture stress for ITO thin film is of 1.2 GPa [56]. Accordingly, the ITO/PEN sheet is expected to experience severe cracking right after applying a tensile bending of 10 mm curvature radius, as the calculated σ1 = 1309 MPa is greater than the critical tensile fracture stress. This is consistent with the experimental observation in which the electrical resistance significantly increases by 500% at the beginning of applying a tensile bending of 10 mm radius on the ITO/PEN sheet (Figure3d). These calculations also show that ITO thin film in the ITO/PET sheet has a smaller stress at a given bending radius, compared to the ITO/PEN one. Based on the mechanics calculation of multilayer thin film under flexural deformation [57] and a critical tensile fracture stress of 1.2 GPa for ITO thin film [56], a limited bending radius of 9.6 mm and 6.1 mm is estimated for the given ITO/PEN and ITO/PET sheets, respectively, for an instantaneous bending failure. Therefore, the ITO/PEN sheet is seriously damaged under a tensile bending of 10 mm radius, and a significantly detrimental effect on the ITO/PET sheet is found for bending with a radius of curvature of 5 mm, as observed in the experiment. Based on the comparisons described above, the given ITO/PET sheet exhibits a greater resistance to long-term flexural deformation than the ITO/PEN sheet, in terms of degradation in the electrical conductance of ITO. Coatings 2018, 8, x FOR PEER REVIEW 11 of 15

Table 1. Material properties and parameters used in the 3D FEA modeling [12–14].

Material Young’s Modulus (GPa) Poison Ratio Thickness (μm) ITO 116 0.35 0.25 PET 3.0 0.37 133 PEN 6.1 0.38 200

In order to correlate the flexible deformation with bending stress, a three-dimensional (3D) finite element method (FEM) model is established to quantitatively determine the corresponding stress values in the bending test. The material properties and thicknesses of thin films used in the FEM modeling are listed in Table 1. Distributions of the maximum principal stress (σ1) in the ITO/PET and ITO/PEN sheets are then obtained for a similar bending radius of 10 mm, as shown in Figure 14. As shown in Figure 14, the ITO/PET or ITO/PET specimen is bent in a concave downward position, and the ITO thin layer on the top is under tension. In the comparison of ITO/PEN and ITO/PET sheets under tensile bending at a bending radius of 10 mm, σ1 in the ITO/PEN and ITO/PET sheets is 1309 MPa and 857 MPa, respectively (Figure 14). Note that the critical tensile fracture stress for ITO thin film is of 1.2 GPa [56]. Accordingly, the ITO/PEN sheet is expected to experience severe cracking right after applying a tensile bending of 10 mm curvature radius, as the calculated σ1 = 1309 MPa is greater than the critical tensile fracture stress. This is consistent with the experimental observation in which the electrical resistance significantly increases by 500% at the beginning of applying a tensile bending of 10 mm radius on the ITO/PEN sheet (Figure 3d). These calculations also show that ITO thin film in the ITO/PET sheet has a smaller stress at a given bending radius, compared to the ITO/PEN one. Based on the mechanics calculation of multilayer thin film under flexural deformation [57] and a critical tensile fracture stress of 1.2 GPa for ITO thin film [56], a limited bending radius of 9.6 mm and 6.1 mm is estimated for the given ITO/PEN and ITO/PET sheets, respectively, for an instantaneous bending failure. Therefore, the ITO/PEN sheet is seriously damaged under a tensile bending of 10 mm radius, and a significantly detrimental effect on the ITO/PET sheet is found for bending with a radius of curvature of 5 mm, as observed in the experiment. Based on the comparisons described above,Coatings 2018the ,given8, 212 ITO/PET sheet exhibits a greater resistance to long-term flexural deformation12 than of 15 the ITO/PEN sheet, in terms of degradation in the electrical conductance of ITO.

(a) (b)

Figure 14. DistributionsDistributions of of maximum maximum principal principal stress stress in inITO ITO thin thin film ﬁlm on on(a) (PETa) PET substrate substrate and and (b) (PENb) PEN substrate. substrate.

4. Conclusions 4. Conclusions (1) Tensile bending is much more detrimental than compressive bending to the electrical conductance (1) Tensile bending is much more detrimental than compressive bending to the electrical conductance of the given ITO/PEN and ITO/PET sheets under a long-term static flexural deformation. of the given ITO/PEN and ITO/PET sheets under a long-term static ﬂexural deformation. (2) Under tensile and compressive bending of a curvature radius of 30 mm or larger, electrical (2) Under tensile and compressive bending of a curvature radius of 30 mm or larger, conductance of both ITO/PEN and ITO/PET sheets remain stable for up to 1000 h of static loading. electrical conductance of both ITO/PEN and ITO/PET sheets remain stable for up to 1000 h of (3) In 1000 h of tensile bending at a 20 mm curvature radius for the ITO/PEN sheet and at a 10 mm static loading. curvature radius for the ITO/PET sheet, the electrical conductance of ITO is degraded by 3–15% (3) In 1000 h of tensile bending at a 20 mm curvature radius for the ITO/PEN sheet and at a 10 mm and 6–20%, respectively. curvature radius for the ITO/PET sheet, the electrical conductance of ITO is degraded by 3–15% and 6–20%, respectively. (4) In 1000 h of tensile bending at a 10 mm curvature radius for the ITO/PEN sheet and at a 5 mm curvature radius for the ITO/PET sheet, the electrical conductance of ITO is signiﬁcantly degraded by 800–900% and 60,000%, respectively. (5) The given ITO/PET sheet has a greater resistance to long-term mechanical bending than the ITO/PEN sheet in terms of change in electrical conductance. This is attributed to a smaller stress in the ITO layer because the ITO/PET sheet has a smaller stiffness and thickness in the substrate than the ITO/PEN sheet.

Author Contributions: C.-K.L. and H.-I.L. conceived and designed the experiments; H.-I.L. performed the experiments; D.-P.T. performed the numerical analysis; H.-I.L., C.-K.L., and D.-P.T. analyzed the data and wrote the paper. Funding: This research was funded by the Ministry of Science and Technology (Taiwan) under contract no. MOST 105-2221-E-008-017-MY3. Conﬂicts of Interest: The authors declare no conﬂict of interest.

References

1. Greiner, M.T.; Lu, Z.-H. Thin-ﬁlm metal oxides in organic semiconductor devices: Their electronic structures, work functions and interfaces. NPG Asia Mater. 2013.[CrossRef]

2. Indium Tin Oxide (ITO) (In2O3):(SnO2), Indium Corporation. Available online: http://www.indium.com/ inorganic-compounds/indium-compounds/indium-tin-oxide/ (accessed on 31 May 2018). 3. Cao, W.; Li, J.; Chen, H.; Xue, J. Transparent electrodes for organic optoelectronic devices: A review. J. Photonics Energy 2014, 4, 040990. [CrossRef] 4. Canhola, P.; Martins, N.; Raniero, L.; Pereira, S.; Fortunato, E.; Ferreira, I.; Martins, R. Role of annealing environment on the performances of large area ITO ﬁlms produced by rf magnetron sputtering. Thin Solid Films 2005, 487, 271–276. [CrossRef] Coatings 2018, 8, 212 13 of 15

5. Gonçalves, G.; Grasso, V.; Barquinha, P.; Pereira, L.; Elamurugu, E.; Brignone, M.; Martins, R.; Lambertini, V.; Fortunato, E. Role of room temperature sputtered high conductive and high transparent indium zinc oxide film contacts on the performance of orange, green, and blue organic light emitting diodes. Plasma Process. Polym. 2011, 8, 340–345. [CrossRef] 6. Alam, M.J.; Cameron, D.C. Optical and electrical properties of transparent conductive ITO thin films deposited by sol–gel process. Thin Solid Films 2000, 377, 455–459. [CrossRef] 7. Ding, X.; Yan, J.; Li, T.; Zhang, L. Transparent conductive ITO/Cu/ITO films prepared on flexible substrates at room temperature. Appl. Surf. Sci. 2012, 258, 3082–3085. [CrossRef]

8. Yu, S.; Yang, W.; Li, L.; Zhang, W. Improved chemical stability of ITO transparent anodes with a SnO2 buffer layer for organic solar cells. Sol. Energy Mater. Sol. Cells 2016, 144, 652–656. [CrossRef] 9. Gil, Y.; Kim, H. Hybrid ITO transparent conductive electrodes embedded with Pt nanoclusters for enhanced output efficiency of GaN-based light-emitting diodes. Thin Solid Films 2016, 603, 307–312. [CrossRef] 10. Tan, N.-N.; Hung, D.-T.; Anh, V.-T.; Kang, B.-C.; Kim, H.-C. Improved patterning of ITO coated with gold masking layer on glass substrate using nanosecond fiber laser and etching. Appl. Surf. Sci. 2015, 336, 163–169. [CrossRef] 11. Park, S.-H.; Lee, S.-J.; Lee, J.-H.; Kal, J.; Hahn, J.; Kim, H.-K. Large area roll-to-roll sputtering of transparent ITO/Ag/ITO cathodes for flexible inverted organic solar cell modules. Org. Electron. 2016, 30, 112–121. [CrossRef] 12. Lim, C.-Y.; Park, J.-K.; Kim, Y.-H.; Han, J.-I. Mechanical and electrical stability indium-tin-oxide coated polymer substrates under continuous bending stress condition. J. Int. Counc. Electr. Eng. 2012, 2, 237–241. [CrossRef] 13. Peccell Product Information, Peccell Technologies, Inc. Available online: http://www.peccell.com/products/ PECF/ (accessed on 31 May 2018). 14. Ibeh, C.C. Thermoplastic Materials: Properties, Manufacturing Methods, and Applications; CRC Press: Cleveland, OH, USA, 2011. 15. Kirubanandham, A.; Basu, S. On Characterization of Mechanical Deformation in Flexible Electronic Structures; Agilent Technologies, Inc.: Santa Clara, CA, USA, 2012. 16. Li, T.-C.; Lin, J.-F. Fatigue life study of ITO/PET specimens in cyclic bending tests. J. Mater. Sci. Mater. Electron. 2014, 26, 250–261. [CrossRef] 17. Li, T.-C.; Han, C.-F.; Chen, K.-T.; Lin, J.-F. Fatigue life study of ITO/PET specimens in terms of electrical resistance and stress/strain via cyclic bending tests. J. Display Technol. 2013, 9, 577–585. [CrossRef] 18. Leterrier, Y.; Médico, L.; Demarco, F.; Månson, J.A.E.; Betz, U.; Escolà, M.F.; Kharrazi Olsson, M.; Atamny, F. Mechanical integrity of transparent conductive oxide films for flexible polymer-based displays. Thin Solid Films 2004, 460, 156–166. [CrossRef] 19. Lin, Y.-C.; Shi, W.-Q.; Chen, Z.-Z. Effect of deflection on the mechanical and optoelectronic properties of indium tin oxide films deposited on polyethylene terephthalate substrates by pulse magnetron sputtering. Thin Solid Films 2009, 517, 1701–1705. [CrossRef] 20. Hamasha, M.M.; Alzoubi, K.; Lu, S.; Desu, S.B. Durability study on sputtered indium tin oxide thin film on poly ethylene terephthalate substrate. Thin Solid Films 2011, 519, 6033–6038. [CrossRef] 21. Lee, C.-C. Modeling and validation of mechanical stress in indium tin oxide layer integrated in highly flexible stacked thin films. Thin Solid Films 2013, 544, 443–447. [CrossRef] 22. Jung, S.; Lim, K.; Kang, J.-W.; Kim, J.-K.; Oh, S.-I.; Eun, K.; Kim, D.-G.; Choa, S.-H. Electromechanical properties of indium–tin–oxide/poly(3,4-ethylenedioxythiophene): Poly(styrenesulfonate) hybrid electrodes for flexible transparent electrodes. Thin Solid Films 2014, 550, 435–443. [CrossRef] 23. Machinaga, H.; Ueda, E.; Mizuike, A.; Takeda, Y.; Shimokita, K.; Miyazaki, T. Effects of annealing temperature on mechanical durability of indium-tin oxide film on polyethylene terephthalate substrate. Thin Solid Films 2014, 559, 36–39. [CrossRef] 24. Kim, E.-H.; Yang, C.-W.; Park, J.-W. Improving the delamination resistance of indium tin oxide (ITO) coatings

on polymeric substrates by O2 plasma surface treatment. Curr. Appl. Phys. 2010, 10, 510S–514S. [CrossRef] 25. Hauger, T.C.; Zeberoff, A.; Worfolk, B.J.; Elias, A.L.; Harris, K.D. Real-time resistance, transmission and ﬁgure-of-merit analysis for transparent conductors under stretching-mode strain. Sol. Energy Mater. Sol. Cells 2014, 124, 247–255. [CrossRef] Coatings 2018, 8, 212 14 of 15

26. Saleh, M.N.; Lubineau, G. Understanding the mechanisms that change the conductivity of damaged ITO-coated polymeric films: A micro-mechanical investigation. Sol. Energy Mater. Sol. Cells 2014, 130, 199–207. [CrossRef] 27. Van der Sluis, O.; Abdallah, A.A.; Bouten, P.C.P.; Timmermans, P.H.M.; den Toonder, J.M.J.; de With, G. Effect of a hard coat layer on buckle delamination of thin ITO layers on a compliant elasto-plastic substrate: An experimental–numerical approach. Eng. Fract. Mech. 2011, 78, 877–889. [CrossRef] 28. Neerinck, D.G.; Vink, T.J. Depth profiling of thin ITO films by grazing incidence X-ray diffraction. Thin Solid Films 1996, 278, 12–17. [CrossRef] 29. Chang, R.-C.; Tsai, F.-T.; Tu, C.-H. A direct method to measure the fracture toughness of indium tin oxide thin films on flexible polymer substrates. Thin Solid Films 2013, 540, 118–124. [CrossRef] 30. Yang, C.-W.; Park, J.-W. The cohesive crack and buckle delamination resistances of indium tin oxide (ITO) films on polymeric substrates with ductile metal interlayers. Surf. Coat. Technol. 2010, 204, 2761–2766. [CrossRef] 31. Kim, Y.-S.; Hwang, W.-J.; Eun, K.-T.; Choa, S.-H. Mechanical reliability of transparent conducting IZTO film electrodes for flexible panel displays. Appl. Surf. Sci. 2011, 257, 8134–8138. [CrossRef] 32. Abdallah, A.A.; Bouten, P.C.P.; den Toonder, J.M.J.; de With, G. Buckle initiation and delamination of patterned ITO layers on a polymer substrate. Surf. Coat. Technol. 2011, 205, 3103–3111. [CrossRef] 33. Boehme, M.; Charton, C. Properties of ITO on PET film in dependence on the coating conditions and thermal processing. Surf. Coat. Technol. 2005, 200, 932–935. [CrossRef] 34. Can, M.; Havare, A.K.; Aydın, H.; Yagmurcukardes, N.; Demic, S.; Icli, S.; Okur, S. Electrical properties of SAM-modified ITO surface using aromatic small molecules with double bond carboxylic acid groups for OLED applications. Appl. Surf. Sci. 2014, 314, 1082–1086. [CrossRef] 35. Han, D.; Lee, S.; Kim, H.; Jeong, S.; Yoo, S. Cathodic multilayer transparent electrodes for ITO-free inverted organic solar cells. Org. Electron. 2013, 14, 1477–1482. [CrossRef] 36. Kim, D.-H.; Park, M.-R.; Lee, H.-J.; Lee, G.-H. Thickness dependence of electrical properties of ITO film deposited on a plastic substrate by RF magnetron sputtering. Appl. Surf. Sci. 2006, 253, 409–411. [CrossRef] 37. Kim, E.-H.; Kim, G.; Lee, G.-H.; Park, J.-W. Nucleation and growth of crystalline indium tin oxide (ITO) coatings on polyethylene terephthalate (PET). Surf. Coat. Technol. 2010, 205, 1–8. [CrossRef] 38. Lim, K.; Jung, S.; Kim, J.-K.; Kang, J.-W.; Kim, J.-H.; Choa, S.-H.; Kim, D.-G. Flexible PEDOT: PSS/ITO hybrid transparent conducting electrode for organic photovoltaics. Sol. Energy Mater. Sol. Cells 2013, 115, 71–78. [CrossRef] 39. Minami, T. Present status of transparent conducting oxide thin-film development for indium-tin-oxide (ITO) substitutes. Thin Solid Films 2008, 516, 5822–5828. [CrossRef] 40. Leppänen, K.; Augustine, B.; Saarela, J.; Myllylä, R.; Fabritius, T. Breaking mechanism of indium tin oxide and its effect on organic photovoltaic cells. Sol. Energy Mater. Sol. Cells 2013, 117, 512–518. [CrossRef] 41. Tran, D.-P.; Lu, H.-I.; Lin, C.-K. Effects of cyclic deformation on conductive characteristics of indium tin oxide thin film on polyethylene terephthalate substrate. Surf. Coat. Technol. 2015, 283, 298–310. [CrossRef] 42. Alkhazaili, A.; Hamasha, M.M.; Choi, G.; Lu, S.; Westgate, R. Reliability of thin films: Experimental study on mechanical and thermal behavior of indium tin oxide and poly(3,4-ethylenedioxythiophene). Microelectron. Reliab. 2015, 55, 538–546. [CrossRef] 43. Peng, C.-Y.; Hamasha, M.M.; VanHart, D.; Lu, S.; Westgate, C.R. Electrical and optical degradation studies on AZO thin films under cyclic bending conditions. IEEE Trans. Device Mater. Reliab. 2013, 13, 236–244. [CrossRef] 44. Hamasha, M.M.; Alzoubi, K.; Switzer, J.C.; Lu, S.; Poliks, M.D.; Westgate, C.R. Reliability of sputtered aluminum thin film on flexible substrate under high cyclic bending fatigue conditions. IEEE Trans. Compon. Packag. Technol. 2012, 2, 2007–2016. [CrossRef] 45. Peng, C.-Y.; Dhakal, T.P.; Garner, S.M.; Cimo, P.; Lu, S.; Westgate, C.R. Strained growth of aluminum-doped zinc oxide on flexible glass substrate and degradation studies under cyclic bending conditions. IEEE Trans. Device Mater. 2014, 14, 121–126. [CrossRef] 46. Hamasha, M.M.; Dhakal, T.P.; Vasekar, P.; Alzoubi, K.; Lu, S.; Vanhart, D.; Westgate, C.R. Reliability of sputter deposited aluminum-doped zinc oxide under harsh environmental conditions. Sol. Energy 2013, 89, 54–61. [CrossRef] Coatings 2018, 8, 212 15 of 15

47. Hamasha, M.M.; Alzoubi, K.; Lu, S. Behavior of sputtered indium–tin–oxide thin film on poly-ethylene terephthalate substrate under stretching. J. Display Technol. 2011, 7, 426–433. [CrossRef] 48. Kim, J.-H.; Park, J.-W. Improving the flexibility of large-area transparent conductive oxide electrodes on polymer substrates for flexible organic light emitting diodes by introducing surface roughness. Org. Electron. 2013, 14, 3444–3452. [CrossRef] 49. Park, J.-M.; Gu, G.-Y.; Wang, Z.-J.; Kwon, D.-J.; Devries, K.L. Interfacial durability and electrical properties of CNT or ITO/PVDF nanocomposites for self-sensor and micro actuator applications. Appl. Surf. Sci. 2013, 287, 75–83. [CrossRef] 50. Tseng, S.-F.; Hsiao, W.-T.; Huang, K.-C.; Chiang, D.; Chen, M.-F.; Chou, C.-P. Laser scribing of indium tin oxide (ITO) thin films deposited on various substrates for touch panels. Appl. Surf. Sci. 2010, 257, 1487–1494. [CrossRef] 51. Yu, Z.; Li, Y.; Xia, F.; Zhao, Z.; Xue, W. Properties of indium tin oxide films deposited on unheated polymer substrates by ion beam assisted deposition. Thin Solid Films 2009, 517, 5395–5398. [CrossRef] 52. Product Specification of WT155hh, Win-Optical Technology Co., Ltd. Available online: http://www.win- optical.com.tw/products_cf.html (accessed on 31 May 2018). 53. Busfield, J.; Thomas, A.; Yamaguchi, K. Electrical and mechanical behavior of filled rubber. III. Dynamic loading and the rate of recovery. J. Polym. Sci. Part B Polym. Phys. 2005, 43, 1649–1661. [CrossRef] 54. Yamaguchi, K.; Busfield, J.; Thomas, A. Electrical and mechanical behavior of filled elastomers. I. The effect of strain. J. Polym. Sci. Part B Polym. Phys. 2003, 41, 2079–2089. [CrossRef] 55. Shimamura, H.; Nakamura, T. Mechanical properties degradation of polyimide films irradiated by atomic oxygen. Polym. Degrad. Stab. 2009, 94, 1389–1396. [CrossRef] 56. Crawford, G. Flexible Flat Panel Displays; Wiley: New York, NY, USA, 2005. 57. Kim, N.; Graham, S. Development of highly flexible and ultra-low permeation rate thin-film barrier structure for organic electronics. Thin Solid Films 2013, 547, 57–62. [CrossRef]

© 2018 by the authors. Licensee MDPI, Basel, Switzerland. This article is an open access article distributed under the terms and conditions of the Creative Commons Attribution (CC BY) license (http://creativecommons.org/licenses/by/4.0/).