FULL PAPER WWW.POLYMERPHYSICS.ORG Substrates for Flexible Electronics: A Practical Investigation on the Electrical, Film Flexibility, Optical, Temperature, and Solvent Resistance Properties Valerio Zardetto, Thomas M. Brown, Andrea Reale, Aldo Di Carlo C.H.O.S.E. (Centre for Hybrid and Organic Solar Energy), Department of Electronic Engineering, University of Rome-Tor Vergata, via del Politecnico 1, 00133 Rome, Italy Correspondence to: T. M. Brown (E-mail: Thomas.Brown@uniroma2.it) Received 21 January 2011; accepted 2 February 2011; published online 1 March 2011 DOI: 10.1002/polb.22227 ABSTRACT: Designing and developing flexible electronics requires a thorough investigation of the substrates available for the fabrication of devices. Here, we present a practical study on a variety of significant substrates: polyethylene terephthalate (PET), its heat-stabilized (HS) derivative, HS-PET, and polyethylene naphthalate (PEN) plastic insulating films; indium tin oxide (ITO)-coated ITO/PEN and ITO/PET transparent conducting films; rigid ITO/glass and FTO/glass substrates; stainless steel and titanium foils. We put the substrates through a range of tests these actually undergo during device fabrication to determine their optical, mechanical flexibility (under different types of tensile and compressive stress bending with and without a PEDOT:PSS conducting polymer layer), solvent resistance, stability to temperature treatment (conductivity and deformation), and to UV irradiation. We highlight issues and propose solutions to improve substrate response. The results and thresholds extracted reveal limitations and windows of opportunity useful for the designer of flexible optoelectronics in determining manufacturing processes and the final applications under everyday C 2011 Wiley Periodicals, Inc. J Polym Sci Part B: operation. V Polym Phys 49: 638–648, 2011 INTRODUCTION In the past decade, the interest from both the industry and consumers toward flexible electronic and optoelectronic devices has risen enormously due to the unique properties of these being lightweight, bendable, conformable, rugged, and not easily breakable. The enthusiasm is also related to the fact that the films used as substrates enable web roll-to-roll processing leading to very substantial gains in manufacturing throughput and economy.1 These features promise the future spread of flexible electronics in many applications such as photovoltaics,2–7 organic lightemitting diodes,8–10 displays,11–13 lighting,14,15 e-paper,16,17 radiofrequency identification systems (RFID),18,19 thin-film transistor electronics (TFT),19–23 pressure, image and chemical sensors,24–26 logic memories,27 and a raft of yet-to-bedevised new applications. The technologies are being developed with a great variety of materials, ranging from solution-processed polymers, evaporated organic molecules, thin-film inorganic semiconductors, and oxides to hybrids materials and multilayers. Thin polymeric foils are extremely attractive, because, first, they can be made lightweight and inexpensive making it possible to target high-volume and low-cost commercial products. Processing of these materials being usually carried out at temperatures below 180–120  C makes them particularly attractive for organic technology based on low-temperature printing or evaporation techniques. Secondly, they can be made transparent: an essential prerequisite for applications where light has to either enter (e.g., photovoltaics and photodetection) or exit (displays, lighting) the device. Finally, these polymer foils are inherently plastic enabling a high degree of mechanical flexibility and conformability.28 A variety of foils is being developed as substrates for flexible electronics espousing the mechanical/optical properties of these with the requirements arising from the various manufacturing processes and operation conditions of the finished applications. In general, two classes of substrates are considered: plastic foils and thin metal foils. C 2011 Wiley Periodicals, Inc. V 638 JOURNAL OF POLYMER SCIENCE PART B: POLYMER PHYSICS 2011, 49, 638–648 KEYWORDS: flexible films; indium tin oxide (ITO); polyethylene naphthalate (PEN); polyethylene teraphthalate (PET); substrates A number of plastic films are available including thermoplastic amorphous polymers such as polycarbonate, polyethersulfone, and high-transition glass temperature materials (Tg) such as polyimide.29 The latter shows excellent thermal properties but due to its low transparency, it is usually used as an opaque substrate (e.g., on the back electrode side for inorganic CIGS and a-Si solar cells, TFTs, and light-emitting and reflective displays).30–32 The other aforementioned transparent amorphous polymers do not always exhibit the desirable mechanical and solvent resistance properties exhibited instead by semicrystalline polyethylene terephthalate (PET) and polyethylene naphthalate (PEN) polymers. It is FULL PAPER WWW.POLYMERPHYSICS.ORG these materials (PET and PEN) that constitute the two main transparent flexible substrates used today in the development of flexible electronics. Not only they provide good resistance to solvents and a discrete tolerance to temperature but also their intrinsic clarity makes them apt as transparent substrate electrodes when coated with indium tin oxide (ITO) for all those optoelectronic applications that require transmission of light and electrical conduction. Thin metal foils represent an alternative to plastic films where higher temperatures are required during manufacturing (for instance, in the fabrication of inorganic thin film a-Si, CIS, CIGS solar cells, or silicon transistors). They also provide excellent conductivity, good chemical resistance, and a better barrier to moisture or oxygen ingress compared to plastic sheets, which often require additional special encapsulation where long-lasting operation is a must.33 Stainless steel (SS) is a common choice,34–38 or even titanium foil for chemical inertness.39 For many high-volume applications such as the aforementioned displays and photovoltaics, however, at least one substrate has to be transparent. For all these ‘‘front interfaces,’’ PET, and/or PEN films have often become the substrate of choice. Building a quantitative and comprehensive knowledge of the characteristics of these substrates becomes crucial for ongoing research and development into the manufacturing processes one can use and for designing applications that can withstand every day use. Some studies have been carried out focusing on particular aspects such as those relating to the bare nonconductive polymer substrates,28 or on the mechanical stability of ITO,40–42 or on the thermal stability and optical properties of conductive plastic foils.43,44 The aim of this work is to carry out a broad, comprehensive systematic investigation encompassing mechanical flexibility, thermal and electrical stability, solvent-resistance, and optical properties of PET and PEN substrates, both in their bare form but importantly also when coated with ITO. The investigation consists in putting the substrates through a range of reproducible tests, which the substrates actually undergo during device fabrication in the laboratory. Furthermore, we have also endeavored to extend the practical investigation to other situations that simulate possible manufacturing processes and operation conditions of the final device. First, we investigated the effect of a temperature treatment (from room temperature to 250  C) on the sheet resistance of flexible foils (PET/ITO, PEN/ITO, Ti, and SS). We compared these with the behavior of ITO and fluorine tin oxide (FTO)-coated glass, which were able to withstand higher temperatures. ITO and FTO glass represent the transparent substrates usually used for ‘‘rigid’’ optoelectronics. Subsequently, we quantified the deformation arising from temperature treatments in bare PET, its heat stabilized (HS) derivative (HS-PET), and HS-PEN (here listed in inverted order of stability according to our experiments). PET/ITO and PEN/ ITO films (sheet resistance of 15 X/h) showed poorer temperature stability compared to their bare counterparts. As an example, to highlight the importance of temperature treat- WWW.MATERIALSVIEWS.COM ments on an optoelectronic device that is also being developed on flexible substrates,45 experimental data of the conversion efficiencies of dye solar cells, as a function of the sintering temperature of the TiO2 layer will be presented. For binder-free pastes, we found that the power conversion efficiency g increases monotonically with T. We evaluated the mechanical stability of substrates monitoring how the sheet resistance of both PET/ITO and PEN/ITO conductive substrates changed under compressive and tensile stress for different radii of curvature r (down to 3 mm). Because ITO is rigid and brittle, it cracked when it was bent and stretched, leading to a dramatic decrease in its conductivity. Cracking occurred under tensile stress for bending radii <14 mm, depending on flexing conditions. Tensile stress was more damaging than compressive stress. We spin-coated poly(3,4-ethylenedioxythiopene) poly(styrenesulfonate) (PEDOT:PSS) layer over ITO and observed improvement in the electrical stability of the system under mechanical stress with the conductive polymer layer also furnishing an upper limit to the overall sheet resistance of ITO/PEDOT:PSS even after the most demanding bending tests. Optical transmission data for all the substrates are also reported, highlighting the differences between PET and PEN and their conductive ITO counterparts (including ITO and FTO glass). Optical degradation was monitored after UV treatment leading to a loss of transparency and some yellowing in particular in PEN. Finally, we show that the PET/ITO and PEN/ITO films show remarkable stability to treatments under the most common polar and a-polar solvents used in the fabrication of plastic electronic devices. EXPERIMENTAL Substrates PET (Melinex506), heat-stabilized HS-PET (Melinex ST506), and HS-PEN (Teonex Q65FA) films (125-lm thick) were supplied by Dupont-Tejiin Films; PET/ITO (LR15 175 lm thick) and PEN/ITO (125 lm thick) by Solutia Performance Films (sheet resistances of 15 X/h); glass/ITO (10 X/h thick 1.1 mm) by Kintec and glass/FTO (8 X/h 3.1 mm and 15 X/h 3 mm) by Pilkington. Stainless steel (FE220270) and titanium (TI000380) foils by GoodFellows; additional PET/ITO by Sigma Aldrich (639303). Before all experiments, the substrates were rinsed in beakers containing acetone and ethanol. For resistance measurements, the substrates were cut to a dimension of 2  2.5 cm2 for plastic and metal samples and 2  2 cm2 for glass. Silver paste (RS 186-3600) was applied precisely to two sides. The resistance, RCAL, was measured between the two silver strips. The sheet resistance (RSHEET) was estimated using RCAL
(W/L), and the values obtained were equivalent to those provided on the specification sheets by the suppliers. Thermal Stability Plastic films, metal foils and conductive glass were processed in a precision hotplate with closed lid to increase uniformity (Titan Typ PZ 28-3TD) for 30 min at a particular temperature. JOURNAL OF POLYMER SCIENCE PART B: POLYMER PHYSICS 2011, 49, 638–648 639 FULL PAPER WWW.POLYMERPHYSICS.ORG The oven used was Lenton VHT6/120. Sheet resistance value was always measured at room temperature (after the 30-min treatment), when the temperature returned back to RT. Temperature-induced plastic deformation was gauged estimating the radius of curvature r. r was calculated by using the trigonometric Chord Theorem and Law of sines based on two experimentally measured dimensions: projection of the substrate length on the ground plane (measured with a caliper) and the substrate height from the plane (measured by monitoring the focus distance difference between the curved surface and a reference flat one with an Olympus BX51M Optical microscope with a 50 magnification lens). For the most severe deformation (i.e., 2pr < substrate length), the diameter of the rolled up substrate was measured directly with the caliper. Mechanical Flexibility To gauge the bending effects on the electrical characteristics of the conductive substrates, the samples were applied conformally to cylinders of different radii ranging from 30 to 3 mm. For tensile (compressive) stress, the ITO side was facing externally (internally) from the cylindrical surface. Resistance was monitored in two situations: in the first, measurements were carried out flat after bending cycles (1, 50, 100, and 150); in the second, when the sample was kept fixed on the cylinders. For practicality, the substrate lengths and width were p*r and 0.75
p*r, respectively. Micrographs were taken with an Olympus BX51M optical microscope with a 50 magnification lens. PEDOT:PSS CLEVIOS P HC-V4 was deposited from a 2.7-lm filtered solution by spin-coating yielding (110 6 10) nm thick films. The samples were annealed at 120  C for 10 min in a Nitrogen glove box. Optical Investigation Transmittance spectra were measured with a spectrophotometer (Shimadzu UV–vis 2550, with integrating sphere, MPC220). The effect of UV irradiation on the plastic samples was carried out by subjecting these to a 150-min UV treatment under a (5000-EC by Dymax) lamp (with a power of 225 mW/cm2). Total luminous AM1.5G light transmittance (sTOT) was calculated with the following equation: sTOT ¼ Z760 380 , Z760 TðkÞ
DðkÞ
VðkÞ
dðkÞ DðkÞ
VðkÞ
dk 380 where T(k) is transmittance data measured, D(k) is solar irradiance spectra,46 and V(k) is human eye sensitivity.47 Solvent Resistance PET/ITO and PEN/ITO samples were immersed in an ultrasonic bath of acetone (10 min) and ethanol (10 min). Samples were dipped also for 3 h in ethanol (414608 by Carlo Erba), acetone (406974 by Carlo Erba), and 2 h in xylene, toluene (488555 by Carlo Erba), and chlorobenzene (270644 Sigma Aldrich). Preparation of Dye Solar Cells Preparations of the devices were those described in ref. 48, except for dye (N719 from Dyesol), electrolyte (EL-HSE from 640 JOURNAL OF POLYMER SCIENCE PART B: POLYMER PHYSICS 2011, 49, 638–648 Dyesol), and for the TiO2 layer. TiO2 layer was prepared by a binder-free paste containing TiO2 nanoparticles of 25-nm diameter (P25 Degussa) and 100-nm diameter (Sigma Aldrich 634662V) mixed in a ratio of (7:3) in water and ethanol (at 20% wt) solution (1:1.5). After deposition via blade coating (5  5 mm2 cm2) on FTO-glass, the TiO2 was sintered in the hotplate at different temperatures (RT to 600  C) before completing the device fabrication. All the devices were tested under a sun simulator (Solar constant 1200 KHS) at AM 1.5 1000 W/m2 calibrated with a Skye SKS 1110 sensor. R RESULTS AND DISCUSSION The substrates investigated were of four types: (1) bare polymer films, PET, HS-PET, HS-PEN, provisioned from the same manufacturer (Dupont Teijin), to highlight the difference between the substrates rather than variations that can arise from different suppliers; (2) PET/ITO and PEN/ITO films (N.B. where not specifically mentioned all PET/ITO and PEN/ITO data consisted in Solutia Performance Films foils. Any additional data from different suppliers will be specified in the text); (3) Stainless steel and titanium foils and (4) glass/ITO and glass/FTO. The results of each data point represent the average over three different samples, and the error bars correspond to the standard deviation. Thermal Properties of Substrates Two important characteristics to gauge when choosing a substrate for temperature stability arising from thermal processes during device fabrication and/or operation are the deformation (for flexible substrates) and electrical degradation (for conductive substrates). Electrical Stability After Thermal Treatment Thermal stability of the conductive plastic (PET/ITO, PEN/ ITO), metal foil (Ti, SS), and glass (glass/ITO and glass/FTO) substrates was evaluated by monitoring the sheet resistance as a function of treatment temperature (30 min) as shown in Figure 1. Starting with glass substrates, the sheet resistance (RSHEET) of ITO stays roughly constant (10 X/h) till 250  C, after which it starts to increase, doubling at 325  C and reaching four times its original value at 600  C. Interestingly instead, the sheet resistance of glass/FTO does not increase but stays constant till about 450  C and then actually decreases to 7 X/h at 600  C making it a more viable option for those applications, which require stability for temperatures above 250–300  C. The behavior for these conductive glass substrates is consistent with work by Kawashima et al.49 PEN/ITO shows good electrical stability up to 235  C up to which the sheet resistance value remains stable (15 X/h). At 250  C, however, the resistance increases almost 20-fold due to partial melting of the substrate. For PET/ITO, the sheet resistance remains stable (15 X/h) only up to 150  C, increasing by 20% at 200  C (18.1 X/h). Above 200  C, RSHEET increases dramatically until the substrate has melted substantially at 250  C. The higher thermal stability of PEN/ ITO compared to PET/ITO can be mainly ascribed to the WWW.POLYMERPHYSICS.ORG FULL PAPER measurements were carried out: in the first case, the probes were carefully placed on the Ti surface; for the second curve, the probes were applied with force, thereby, (partially) boring through the thermal oxide. This effect was not present in stainless steel. Thermal Deformation We found that the films curve when heated. The deformation of plastic substrates is reported in Figure 2(a) in terms of radius of curvature (and its reciprocal) of the deformed substrate as a function of treatment temperature (30 min). Obviously, the smaller the radius, the larger the temperatureinduced deformation undergone by the substrate. Results are presented for PET, HS-PET, HS-PEN, PET/ITO, and PEN/ITO. For the glass/ITO, glass/FTO, and the Ti and SS metal foils, we did not observe any appreciable deformation. At 115  C, the PET films bend to a radius of curvature of r ¼ 1 m. The more lenient threshold of r ¼ 0.5 m is attained at 152  C. Improvement was observed for heat stablized HSPET for which the thresholds are 180  C (r ¼ 1 m) and 183  C (r ¼ 0.5 m). HS-PEN, whose thresholds are 180  C (r ¼ 1 m) and 193  C (r ¼ 0.5 m), shows a less pronounced deformation above 180  C. At 250  C, both types of PET melt loosing all mechanical consistency, whereas PEN still retains some of its consistency albeit signifcantly deformed. FIGURE 1 (a) Room temperature sheet resistance of conductive PET/ITO, PEN/ITO foils and glass/ITO, glass/FTO substrates after undergoing a thermal treatment at different temperatures for 30 min; (b) resistance as a function of treatment temperature (30 min) for stainless steel (SS) and titanium (Ti). For Ti (1), the electrical probes were gently placed on the Ti surface; for Ti (2), these were applied with significant pressure (see text). The higher temperature stability of heat stablized HS-PET, in comparison to PET, is due to the heat-stabilization process that increases the dimensional stability of polymer films, because the internal strain in the film itself is relaxed by exposure to high temperature while under minimum line thermal stability of the underlying polymer films, as we will show below. Stainless steel shows a constant resistance (R ¼ 0.4 X) for the entire range of temperatures (RT to 600  C). With respect to the resistance measured on the PET/ITO films (R ¼ 16.5 X, equivalent to RSHEET ¼ 15 X/h), the resistance R measured for SS of the same dimension is, thus, a factor of 41 lower (useful for applications where high current densiies are present). It is interesting to note that the resistance of the Ti of the same dimension (0.3 X at RT) only remains stable up to 325  C and then increases considerably. Two curves are reported in Figure 1(b). The first increases steeply with T (R > 2 M X at 600  C). The other has a less pronounced slope (R ¼ 5 X at 600  C). The increase of resistance is likely due to the growth of a thin layer of insulating oxide whose thickness depends on annealing temperature (and time) when the process is carried out in air.50 The difference between the two curves lies in the way the two WWW.MATERIALSVIEWS.COM FIGURE 2 (a) Deformation of plastic substrates (PET, HS-PET, HS-PEN, PET/ITO, PEN/ITO) in terms of radius of curvature (and its reciprocal) as a function of treatment temperature (30 min). The ‘‘PET/ITO constr.’’ samples consisted in PET/ITO mechanically constrained on two sides of the film onto rigid substrates; (b) image of PET/ITO substrates after the thermal treatment. JOURNAL OF POLYMER SCIENCE PART B: POLYMER PHYSICS 2011, 49, 638–648 641 FULL PAPER WWW.POLYMERPHYSICS.ORG tension.51,52 In addition, HS-PEN is more stable than HS-PET due to the presence of naphthalene rings instead of the benzenes, which increases the Tg of the material.53 The behavior in temperature changes when one considers the same substrates coated with an ITO layer. First, the substrates show some (albeit not significant) curvature even at room temperature. This may be due to residual stresses contained in the foils arising from the deposition process of the ITO on the polymer.40 The films start to bend more rapidly (i.e., at lower temperatures) than their neat PET and PEN counterparts. The presence of ITO leads to more pronounced deformation as the radii of curvature were about an order of magnitude greater than the neat polymer film counterparts. Above 150  C, PET/ITO shows a more pronounced deformation compared to PEN/ITO, which shows a dramatic increase in curvature only above 180  C. PET/ITO films actually tended to completely roll up at 220  C (and to melt at 250  C) as shown in Figure 2(b). Although in most laboratories, the substrates are processed in hotplates or ovens as sheets (RSHEET show the same values in oven and hotplate, whereas thermal deformation is slightly higher in the oven), it is likely that for high-volume production, the substrates will be processed in roll-to-roll continuous web form. Here, tensile forces along the substrate axis will constrain the film (along the ends of the two rolls). We simulated the application of these tensile forces by constraining two sides of the flexible films to rigid substrates using tape. As can be noted, the constraint improves the ruggedness of the flexible substrates by helping to reduce significantly the deformation caused by the thermal treatments: for PET/ITO, the onset of significant deformation is delayed to 150  C. We observed that the ITO films always bend convexly both in the oven and hotplates [in Fig. 2(b), the ITO is on the top side of the substrates], as a result of internal stresses and differences in the coefficient of linear thermal expansion of polymer and ITO.40 Finally, to highlight the importance of this temperature study, we report the conversion efficiencies (g) of dye solar cells as a function of sintering temperature for the binder-free TiO2 layer on FTO-glass substrates in Figure 3. For these binderfree pastes, differently from pastes that include binders,54 g increases monotonically with T because higher temperatures lead to improved electromechanical bonding between the TiO2 nanoparticles. Thus, the availability of substrates that withstand higher temperatures can be important for improved performance of devices (and ultimately of cost/KWp). At the moment, until R&D provides even more dimensionally stable alternatives, opaque metal foils have to be used for temperatures above 180  C,2,55 or alternative sintering methods identified, such as mechanical compression,6 microwave,56 UV,57,58 or raster scanning laser processing.48,59 Flexibility and Resistivity One of the most valuable properties of polymer films is their inherent flexibility. This is particularly appealing for electronic applications, which can be made flexible or conformal to an underlying curved surface but also during manufacturing (the films are wound-up in rolls). Flexibility becomes 642 JOURNAL OF POLYMER SCIENCE PART B: POLYMER PHYSICS 2011, 49, 638–648 FIGURE 3 Power conversion efficiencies of dye solar cells as a function of sintering temperature of the binder-free nanocrystalline TiO2 layer on glass/FTO substrates. Also reported vertically are our estimates for the upper treatment temperature (for 30 min) applicable for some of the substrates investigated in our study without leading to significant degradation of the substrates. For PET/ITO this depends on heating procedure as explained in the text; for stainless steel and FTO, this may lay even higher than the 600  C range of our studies. appealing only when there is little loss in the conductance of the conductive films on bending or flexing the substrates. We monitored the sheet resistance of PET/ITO and PEN/ITO films under compressive and tensile stress for varying radii of curvature (from 30 to 3 mm) in two different situations: in the first, the samples underwent bending cycles and were then measured when flat; in the second, the substrates were measured when still fixed on the curved cylindrical surface. Naturally, the results depend greatly on the properties of the conductive film used. ITO is a brittle inorganic electrode. PEDOT:PSS, first used in OLEDs as anode60 leading to improved device performance in combination with ITO61 mainly due to a lowering of the injection barrier (increase of the built in potential)62 has become a widely used conductive polymer,8,60–63 for a great variety of other optoelectronic devices including solar cells and TFTs.23,64 We will show that PEDOT:PSS films on curved flexible foils, both alone and in combination with ITO, dramatically improves the electrical stability of the substrates on flexing. ITO In Figure 4(a), the sheet resistance of PET/ITO is reported as a function of the radius of curvature, r, during the bending process. Under all our bending conditions, the conductivity of the TCO remained unaltered for r > 14 mm. For smaller radii of curvature, we observed a big difference in behavior between the substrates that underwent tensile stress (i.e., ITO on the outer surface) to those that underwent compressive stress (i.e., ITO on the inner surface). Immediately below WWW.POLYMERPHYSICS.ORG FULL PAPER a surface). For one cycle, degradation of the conductance only occurs at r ¼ 5 mm. We also measured RSHEET for 50, 100 bending cycles. RSHEET increased monotonically under repeated flexing (from 1 to 150 cycles). The degradation of ITO conductivity when increasing curvature is related to the fact that the ITO film is brittle. The optical microscopy images of Figure 5 (relating to 150 cycle tests) show that the ITO coating has damaged considerably after undergoing those tensile and compressive stresses that degrade its conductance. The number of visible cracks increases for smaller radii of curvature (thus, leading to an increase in resistance). For tensile stresses (Fig. 5), the cracks are less visible although more numerous per cm2 compared to those arising from compressive stress. Furthermore, degradation not visible at the optical microscopy level may also occur. The fact that ITO substrates are more sensitive to tensile stresses is also evident in Figure 5 and confirmed in other studies.65,66 It has been reported that for tensile stress, the main failure is usually due to cracking/ channeling, whereas for compressive stress, debonding may become important.67,68 Debonding will depend also on the quality of adhesion between ITO and the polymer films.69 FIGURE 4 Sheet resistance of PET/ITO (a) and PEN/ITO (b) as a function of radius of curvature under tensile (in black) and compressive (in red) stress for three different situations: measured on a curved cylindrical surface (squares-line); after 150 (triangles-dashed) and 1 (diamonds-dotted) bending cycles measured flat. r ¼ 14 mm, both the PET/ITO substrates that underwent fixed tensile stress (i.e., the film was measured when still bent—useful for conformal applications) and bending for 150 cycles (measured flat), show a marked degradation in conductance. At r ¼ 11 mm, the resistance has already increased by about an order of magnitude. The situation aggravates at even smaller r. Compressive stress is much more lenient on the substrates: degradation of the conductivity occurs around r ¼ 8 mm for fixed compressive and around r ¼ 11 mm for 150 cycles stress. Resistance values measured in fixed tensile stress are higher than after 150 cycles flat, because in the former case, the failures or cracks are stretched under measurement. The stretching is not as present under fixed compressive stress as can be noted from the different behavior. Figure 4(a) also shows the data measured on substrates that underwent only one bending cycle as it consists in a useful data point for applications or manufacturing processes where the substrate finds itself bent only once (e.g., in a production or storage roll before being applied flat or curved to WWW.MATERIALSVIEWS.COM PEN/ITO substrates have a similar behavior to that of PET/ ITO films in that tensile stress degradation occurs for r < 14 mm, whereas for compressive stress, degradation occurs for r < 8 mm (at r ¼ 5 mm RSHEET increased by 50% after 150 cycles); for one cycle degradation occurs around r ¼ 5 mm. However, it is evident that the electrical degradation on flexing is less pronounced. Although different adhesion properties cannot be excluded, the different behavior (particularly evident under tensile stress) can be ascribed to the fact that the PEN films (125 lm) were thinner than the PET films (175 lm) because strain is related to film thickness.65,69 We also found that the same flexing tests carried out on Sigma Aldrich PET/ITO, which (presumably) has a thinner ITO film (due to its higher sheet resistance of 60 X/h), improves the flexibility versus resistivity behavior of the films in all conditions (no variations of RSHEET observed till r ¼ 7.5 mm). Thus, for electronic devices for which high current densities are not developed, thinner ITO coatings lead to better mechanical properties (at the expense of conductivity) as also reported by Chen et al.65 Our investigations included different types of measurements, which may be useful in designing the structure and the position of a conductive layer in a flexible device and the maximum bending that can be tolerated. Bending cycles simulate all applications in which the device will be rolled up or flexed during operation. A sample fixed on a curve surface simulates a (conformal) device, which remains curved on operation. We note that, in some applications, where two opposite conductive substrates are used (e.g., plastic dye solar cells) one electrode will undergo tensile stress at the same instant when the opposite electrode will undergo a compressive one. In this case, to avoid any type of failure mechanisms in the conductive layer, our data show that the radius of curvature must be limited to a value 14 mm. JOURNAL OF POLYMER SCIENCE PART B: POLYMER PHYSICS 2011, 49, 638–648 643 FULL PAPER WWW.POLYMERPHYSICS.ORG FIGURE 5 Optical microscopy images (50) of the ITO surface (on PET) after 150 bending cycles under compressive (C) and tensile (T) stresses for different radii of curvature: (a) r ¼ 14 mm; (b) r ¼ 7.5 mm; (c) r ¼ 3 mm. PEDOT:PSS and ITO/PEDOT:PSS In Figure 6, the sheet resistance of PEDOT:PSS on PET, PEDOT:PSS on PET/ITO are compared to that of PET/ITO as a function of the radius of curvature, r, during the fixed tensile stress experiment (i.e., the film was measured when still curved). No appreciable variation in the sheet resistance (5.6  103 X/h) was observed for PET/PEDOT:PSS substrates for the whole range of curvatures investigated down to r ¼ 3 mm). This is in contrast with the PET/ITO sample, which although starting from a greatly lower value, shows a dramatic increase in resistance for all r < 14 mm. The polymer film shows its obvious ‘‘plastic’’ properties as opposed to the degree of britleness inherent in the ITO coatings.70,71 the latter case, this is because the overall resistance of the ITO/PEDOT:PSS system is a parallel of the two contributions (the one due to ITO becoming very large). Note that smaller cracks may appear for the larger r (e.g., r > 10 mm) and that PEDOT:PSS may also fill these acting as an electric bridge. Furthermore, if the crack size is comparable to the length of the polymer chains, the conductance enhancement In many applications, PEDOT:PSS is deposited onto ITO, using the latter for its better conductive properties and the former for its electronic properties (that lead to better hole injection or collection).61 Figure 6 shows that degradation of the sheet resistance in PET/ITO/PEDOT:PSS films is delayed compared to the bare PET/ITO films and starts only for r < 11 mm (instead of r < 14 mm) and also is less marked. Although the resistance for PET/ITO continues to monotonically increase when reducing r (reaching values of 106 X/h for r ¼ 3 mm), the resistance of the PET/ITO/PEDOT:PSS does not exceed the intrinsic value of the PEDOT:PSS resistance even for r ¼ 3 mm. The introduction of PEDOT:PSS layer over the ITO coating has three positive effects. Firstly, it delays the onset of degradation to lower radii of curvature; secondly, the rate of resistance increase is reduced; thirdly, it limits the highest resistance value to that of the intrinsic PEDOT:PSS film. In 644 JOURNAL OF POLYMER SCIENCE PART B: POLYMER PHYSICS 2011, 49, 638–648 FIGURE 6 Sheet resistance of PEDOT:PSS on PET (squares) and PEDOT:PSS on ITO/PET (empty triangles) and ITO/PEN (empty circles) compared with the bare PET/ITO (full triangles) and PEN/ITO (full circles) as a function of the radius of curvature during the fixed tensile stress test (i.e., the film was measured when curved). WWW.POLYMERPHYSICS.ORG FULL PAPER added by the PEDOT:PSS overlayer will be more effective (high r) that in the case, where the cracks become much wider (lower r). If this particular PEDOT:PSS layer is replaced with a different and more conductive variety, it is likely that the asymptotic value would decrease to the intrinsic sheet resistance achievable.72–75 The closer this value to that of the more brittle ITO, the more appealing it would become for flexible electronics. Optical Investigation Transparency of the substrate is one of the crucial requirements for flexible optoelectronics where light has to either penetrate or exit the device (e.g., displays, photodiodes). Here, we study the transmittance spectra of the materials and estimate light transmission in the visible range (380– 760 nm).76 Also, we investigated the optical stability under UV exposure, important for outdoor applications or for particular manufacturing processes (e.g., photolithography or UV-ozone treatments). Transmittance Spectra Transmittance spectra for HS-PET and HS-PEN and their ITO-covered counterparts are reported in Figure 7(a) in the 330–800 nm range. Neat HS-PET films show the higher transmittance over all the visible range is flat and around 90% (sTOT¼90.5%), only decreasing appreaciably in the UV (k < 380 nm). HS-PEN films show a lower transmittance (sTOT¼88.7%) especially in the blue region of the visible spectrum (transmittance drops decisevely below 400 nm). The difference between the two materials is ascribed to the naphthalene rings in the PEN main chain, which start absorbing at longer wavelengths compared to the benzene rings in the PET.77,78 Note that nonheat stablized films are more transparent than the HSPET. The addition of ITO to the PET and PEN foils brings about an overall reduction of sTOT (85.2 and 82.3%, respectively) particularly evident in the blue and especially in the UV part (k < 420 nm) of the visible specrtum. The spectra show a maximum at 560 nm (87.3%) and 576 nm (84.6%), respectively (N.B., the peak sensitivity of the photopic human eye is at 555 nm).79 FIGURE 7 (a) UV–vis trasmittance spectra for PET/ITO and PEN/ ITO and also HS-PET and HS-PEN (before and after UV exposure). (b) UV–vis trasmittance spectra for PET/ITO (15 X/h), PEN/ITO (15 X/h), glass/ITO (10 X/h), glass/FTO (8 X/h and 15 X/h) conductive susbtrates. Figure 7(b) shows a comparison of the transmittance spectra of conductive plastic (PET/ITO PEN/ITO), with those of conductive glass (glass/ITO 10 X/h and glass/FTO 8 X/h and 15 X/h). The glass substrate covered with ITO shows a total light transmittance of 86%, while glass covered with FTO 78.6% and 83% in the visible range, for the more and less conductive versions respectively (note: however, that the glass substrates covered with different conductive layers are also different in thickness, see Experimental section). The spectra also show that glass substrates are more transparent in the UV range compared to the plastic films and also that FTO shows a flatter transmittance in the visible range. exposure (150 min). When HS-PEN is exposed to a significant flux of UV light, transparency decreases (sTOT from 88.7 to 87.7%) especially in the blue range of wevelengths, 380 nm < k < 500 nm.76 PET is more resistant to UV exposure (sTOT from 90.6 to 90%). Indeed, we noticed a discernible yellowing of PEN after UV irradiation, not appreacible with PET (even if the transmission spectra in the violet range is slightly lower post-treatment even for PET). Yellowing is due to photo-oxidation, triggered by energetic UV photons leading to scission of bonds in the backbone of polymer, formation of new complexes that can lead to yellowing, loss of elasticity, and even cracks and material loss.80–82 The difference is mainly due to the presence in PEN of the naphthalene structures (instead of the benzenes in PET), which are more easily affected by UV radiation.80 Processes for UV stabilization of polymers and UV-stabilized films can be considered where increased UV stability is sought.83 Stability Under UV Exposure Figure 7(a) shows a comparison between the transmission spectra of HS-PET, and HS-PEN before and after a UV-light Solvent Resistance Before fabricating the various electronic devices over the flexible substrates, these typically undergo a cleaning step to WWW.MATERIALSVIEWS.COM JOURNAL OF POLYMER SCIENCE PART B: POLYMER PHYSICS 2011, 49, 638–648 645 FULL PAPER remove any contaminants.84 A common wet-cleaning procedure involves subjecting the substrates to ultrasonic baths in acetone and ethanol (or isopropanol). We monitored the PET/ITO and PEN/ITO samples resistivity before and after an ultrasonic bath (10 min each) and a longer immersion (3 h each) in these two solvents, and no visible damage and variation in the conductivity of ITO was observed. The substrates need to also withstand the action of more a-polar solvents that are often used to deposit the active layers over these substrates (e.g., via printing techniques). After 2 h of complete immersion in xylene, toluene, and chlorobenzene, the samples showed no visible damage or any variation in the sheet resistance for either substrate (PET/ITO, PEN/ITO). CONCLUSIONS A practical investigation was carried out on the electrical, thermal and mechanical flexibility, thermal deformation, optical transmittance, and solvent resistance properties of plastic film substrates (PET, PEN and ITO/PET, ITO/PEN). Where applicable, these properties were compared with rigid transparent ITO/glass and FTO/glass substrates, flexible titanium and stainless steel (SS) metal foils. The results aimed to provide useful information and insights for the design and fabrication of flexible electronic devices. After a thermal treatment of 30 min, the PET films spontaneously bent to a radius of curvature of 1 m at 115  C. The more lenient threshold of r ¼ 0.5 m was attained at 152  C. Improvement was observed for high stability HS-PET and HS-PEN for which the thresholds were 180 and 180  C (r ¼ 1 m) and 183 and 193  C (r ¼ 0.5 m), respectively. For the ITO-coated films, the deformation was much more pronounced compared to neat substrates and at considerably lower temperatures. We found that by constraining two sides of these films (i.e., simulating the tensile forces of web processing) deformation can be contained yielding good mechanical and electrical stability up to higher temperatures (150  C for PET/ITO). We observed that the limiting factor was not the electrical stability (PET/ITO showed no variations of sheet resistance up to 180  C while PEN/ITO until 235  C) but the physical deformation of the plastic samples. Stainless steel guaranteed a constant resistance in the whole range of T we studied (RT to 600  C), whereas thermal oxidation of Ti needs to be taken into account for temperatures above 325  C. No appreciable thermal deformation of these opaque metal foils occurred in T. For ITO/glass, the sheet resistance stayed roughly constant (10 X/h) till 250  C, after which it started to increase, doubling at 325  C. For FTO/ glass no increase in RSHEET was observed up to 600  C. Flexibility tests for PET/ITO and PEN/ITO showed that tensile stress degrades the conductivity more significantly than compressive stress both when the substrates were fixed on a curved surface and after 150 bending cycles measured flat (electrical degradation increased monotonically with number of bending cycles). In all cases, no electrical degradation was seen when flexing the substrates down to r ¼ 646 JOURNAL OF POLYMER SCIENCE PART B: POLYMER PHYSICS 2011, 49, 638–648 WWW.POLYMERPHYSICS.ORG 14 mm, which, thus, constitutes a ‘‘safe’’ bending radius. For compressive tests, no changes in RSHEET occurred down to r ¼ 11 mm for PET/ITO, and r ¼ 8 mm for PEN/ITO. Optical microscopy highlighted the influence of failures such as cracks appearing in the ITO coating upon different degrees of flexing. When applying a PEDOT:PSS film over the ITO/ PET and ITO/PEN substrates, the electrical flexibility of the device improved also giving an upper limit to RSHEET in the cases where dramatic degradation (r < 5 mm) of the ITO occurred. Optical transmittance spectra showed that PET is more transparent than PEN (with and without ITO coating layer). Application of a 15 X/h ITO overlayer reduces absolute transmission in the visible by 5%. UV-induced degradation was evident in the yellowing of PEN, but not noticeable in PET. Common solvents of different polarity used for either cleaning procedures (acetone and ethanol) or processing (xylene, toluene, and chlorobenzene) did not affect the electrical characteristics of the ITO-coated plastic foils. ACKNOWLEDGMENTS The authors thank Girolamo Mincuzzi, Luigi Salamandra, Gianpaolo Susanna, and Gabriele De Angelis for assistance in laboratory equipments and materials, Solutia Performance Films (Seube Thierry) and Dupont Tejiin Films (Angela Gardner), for providing plastic substrates, This work was financially supported by a grant from Regione Lazio ‘‘Polo Solare Organico Regione Lazio.’’ T. M. Brown thanks MIUR for a ‘‘Incentivazione alla mobilità di studiosi residenti all’estero’’ fellowship. REFERENCES AND NOTES 1 Forrest, S. R. Nature 2004, 428, 911–918. 2 Ito, S.; Ha, N. C.; Rothemberger, G.; Liska, P.; Comte, P.; Zakeeruddin, S. M.; Pechy, P.; Nazeeruddin, M. K.; Graetzel, M. Chem. Commun. 2006, 4004–4006. 3 Lungenschmied, C.; Dennler, G.; Neugebauer, H.; Sariciftci, S. N.; Glatthaar, M.; Meyer, T.; Meyer, A. Sol. Energy Mater. Sol. Cells 2007, 91, 379–384. 4 Brabec, C. J.; Padinger, F.; Hummelen, J. C.; Janssen, R. A. J.; Sariciftci, N. S. Synth. Metals 1999, 102, 861–864. 5 Krebs, F. C. Sol. Energy Mater. Sol. Cells 2009, 93, 394–412. 6 Yamaguchi, T.; Tobe, N.; Matsumoto, D.; Arakawa, H. Chem. Commun. 2007, 4767–4769. 7 Shah, A.; Torres, P.; Tscharner, R.; Wyrsch, N.; Knepper, H. Science 1999, 285, 692–698. 8 Gustafsson, G.; Cao, Y.; Treacy, G. M.; Klavetter, F.; Colaneri, N.; Heeger, A. J. Nature 1992, 357, 477–479. 9 Friend, R. H.; Gymer, R.; Holmes, A.; Kim, J.; Burroughes, J. H.; Marks, R.; Taliani, C.; Bradley, D.; Dos Santos, D.; Bredas, J.; Loglund, M.; Salaneck, W. Nature 1999, 397, 121–128. 10 Hebner, T. R.; Wu, C. C.; Marcy, D.; Lu, M. H.; Sturm, J. C. Appl. Phys. Lett. 1998, 72, 519–521. 11 Zhou, L.; Wanga, A.; Wu, S. C.; Sun, J.; Park, S.; Jackson, T. N. Appl. Phys. Lett. 2006, 88, 1–3. 12 Gelinck, G. H.; Huitema, H.; van Veenendaal, E.; Cantatore, E.; Schrijnemarkers, L.; van der Putten, J.; Geuns, T.; Beenhakkers, M.; Giesber, J. B.; Huisman, B. H.; Meijer, E. J.; Mena WWW.POLYMERPHYSICS.ORG Benito, E.; Touwslager, F. J.; Marsman, A. W.; van Rens, B. J. E.; de Leeuw, D. M. Nat. Mater. 2004, 3, 106–110. 13 Sugimoto, A.; Ochi, H.; Fujimura, S.; Yoshida, A.; Miyadera, T.; Tsuchida, M. J. Sel. Top. Quantum Electron. 2004, 10, 107– 114. 14 Bao, J.; Zimmler, M. A.; Capasso, F.; Wang, X.; Ren, Z. F. Nano Lett. 2006, 6, 1719–1722. 15 Könenkamp, R.; Word, R. C.; Schlegel, C. Appl. Phys. Lett. 2004, 85, 6004–6006. 16 Chen, Y.; Au, J.; Kaslas, P.; Ritenour, A.; Gates, H.; McCreary, M. Nature 2003, DOI: 10.1038/423136a. 17 Burns, S. E.; Reynolds, K.; Reeves, W.; Banach, M.; Brown, T.; Chalmers, K.; Cousins, N.; Etchells, M.; Hayton, C.; Jacobs, K.; Menon, A.; Siddique, S.; Too, P.; Ramsdale, C.; Watts, J.; Cain, P.; von Werne, T.; Mills, J.; Curling, C.; Sirringhaus, H.; Amundson, K.; McCreary, M. D. J. Soc. Inf. Display 2005, 13, 583–586. FULL PAPER 40 Leterrier, Y.; Medico, L.; Demarco, F.; Manson, J. A. E.; Betz, U.; Escola, M. F.; Olsson, M. K.; Atamny, F. Thin Solid Films 2004, 460, 156–166. 41 Cairns, D. R.; Witte, R. P., II; Sparacin, D. K.; Sachsman, S. M.; Paine, D. C.; Crawford, G. P. Appl. Phys. Lett. 2000, 76, 1425–1427. 42 Cotterell, B.; Chen, Z. Int. J. Fract. 2000, 104, 169–179. 43 Cairns, D. R.; Paine, D. C.; Crawford, G. P.; Kukureka, S. N. SID Symp. Dig. Tech. Pap. 2001, 32, 654–657. 44 Yang, C.; Lee, S.; Chen, S. Thin Solid Films 2008, 516, 1984–1991. 45 Pichot, F.; Pitts, J. R.; Gregg, B. A. Langmuir 2000, 16, 5626–5630. 46 ASTM G173–03. Tables: Extraterrestrial Spectrum, Terrestrial Global 37 deg South Facing Tilt & Direct Normal þ Circumsolar. 47 Human Eye Sensitivity and Photometric Quantities, http://www.ecse.rpi.edu/~schubert/Light-Emitting-Diodes-dot-org/ Sample-Chapter.pdf. 18 Vaillancourt, J.; Zhang, H.; Vasinajindakaw, P.; Xia, H.; Lu, X.; Han, X.; Janzen, D. C.; Shih, W. S.; Jones, C. S.; Stroder, M.; Chen, M. J.; Subbaraman, H.; Chen, R. T.; Berger, U.; Renn, M. Appl. Phys. Lett. 2008, 93, 243301. 19 Redinger, D.; Molesa, S.; Yin, S.; Farschi, R.; Subramanian, V. IEEE Trans. Electron Devices 2004, 51, 1978–1983. 20 Nomura, K.; Ohta, H.; Takagi, A.; Kamiya, T.; Hirano, M.; Hosono, H. Nature 2004, 432, 488–492. 50 Chappel, S.; Grinis, L.; Ofir, A.; Zaban, A. J. Phys. Chem. B Lett. 2005, 109, 1643–1647. 21 Sirringhaus, H. Adv. Mater. 2005, 17, 2411–2425. 22 Artukovic, E.; Kaempgen, M.; Hetch, D. S.; Roth, S.; Gruner, G. Nanoletters 2005, 5, 757–760. 51 MacDonald, W. A.; Mace, J. M.; Polack, N. P. 45th Annual Technical Conference Proceedings of the Society of Vacuum Coaters, 2002, 482. 23 Sirringhaus, H.; Kawase, T.; Friend, R. H.; Shimoda, T.; Inbasekaran, M.; Wu, W.; Woo, E. P. Science 2000, 290, 2123–2126. 52 MacDonald, W. A.; Rollins, K.; Eveson, R.; Rustin, R. A.; Handa, M. Soc. Inf. Display Dig. Tech. Pap. 2003, 34, 264–267. 24 Someya, T.; Sekitani, T.; Iba, S.; Kato, Y.; Kawaguchi, H.; Sakurai, T. Nas 2004, 101, 9966–9970. 48 Mincuzzi, G.; Vesce, L.; Reale, A.; Di Carlo, A.; Brown, T. M. Appl. Phys. Lett. 2009, 95, 103312 1–3. 49 Kawashima, T.; Matsui, H.; Tanabe, N. Thin Solid Films 2003, 445, 241–244. 25 McAlpine, M. C.; Ahmad, H.; Wang, D.; Heat, J. R. Nat. Mater. 2007, 6, 379–384. 26 Zhou, J.; Gu, Y.; Fei, P.; Mai, W.; Gao, Y.; Yang, R.; Bao, G.; Wang, Z. L. Nano Lett. 2008, 8, 3035–3040. 53 MacDonald, W. A.; Looney, M. K.; MacKerron, D.; Eveson, R.; Adam, R.; Hashimoto, K.; Rakos, K. J. Soc. Inf. Display 2007, 15, 1075–1083. 54 Zardetto, V.; De Angelis, G.; Gasiorowski, J.; Reale, A.; Di Carlo, A.; Brown, T. M. Manuscript unpublished. 27 Yang, Y.; Ouyang, J.; Ma, L.; Jia-Hung Tseng, R.; Chu, C. Adv. Funct. Mater. 2006, 16, 1001–1014. 55 Kang, M. G.; Park, N. G.; Ryu, K. S.; Chang, S. H.; Kim, K. J. Sol. Energy Mater. Sol. Cells 2006, 90, 574–581. 28 MacDonald, W. A. J. Mater. Chem. 2004, 14, 4–10. 56 Uchida, S.; Tomiha, M.; Takizawa, H.; Kawaraya, M. J. Photochem. Photobiol. A: Chem. 2004, 164, 93–96. 29 MacDonald, W. A.; Rollins, K.; Eveson, R.; Rakos, K.; Rustin, B. A.; Handa, M. Mater. Res. Soc. Symp. Proc. 2003, 769, Paper H9.3. 30 Gleskova, H.; Wagner, S.; Soboyejo, W.; Suo, Z. J. Appl. Phys. 2002, 92, 6224–6229. 31 Kessler, F.; Herrmann, D.; Powalla, M. Thin Solid Films 2005, 480–481, 491–498. 32 Park, J. S.; Kim, T. W.; Stryakhilev, D.; Lee, J. S.; An, S. G.; Pyo, Y. S.; Lee, D. B.; Mo, Y. G.; Jin, D. U.; Chung, H. K. Appl. Phys. Lett. 2009, 95, 013503-1–3. 33 Burrows, P. E.; Graff, G. L.; Gross, M. E.; Martin, P. M.; Shi, M. K.; Hall, M.; Mast, E.; Bonham, C.; Bennett, W.; Sullivan, M. B. Display 2001, 22, 65–69. 34 Troccoli, M. N.; Roudbari, A. J.; Chuang, T.; Hatalis, M. K. Solid-State Electron. 2006, 50, 1080–1087. 35 Xie, Z.; Hung, L.; Zhu, F. Chem. Phys. Lett. 2003, 381, 691–696. 36 Kessler, F.; Rudmann, D. Sol. Energy 2004, 77, 685–695. 37 Fang, X.; Ma, T.; Akiyama, M.; Guan, G.; Tsunematsu, S.; Abe, E. Thin Solid Films 2005, 472, 242–245. 38 Toivola, M.; Ahlskog, F.; Lund, P. Sol. Energy Mater. Sol. Cells 2006, 90, 2881–2893. 39 Onoda, K.; Ngamsinlapasathian, S.; Fujieda, T.; Yoshikawa, S. Sol. Energy Mater. Sol. Cells 2007, 91, 1176–1181. WWW.MATERIALSVIEWS.COM 57 Longo, C.; Freitas, J.; De Paoli, M. A. J. Photochem. Photobiol. A: Chem. 2003, 159, 33–39. 58 Murakami, T.; Kitijori, Y.; Kawashima, N.; Miyasaka, T. J. Photochem. Photobiol. A: Chem. 2004, 164, 187–191. 59 Mincuzzi, G.; Vesce, L.; Liberatore, M.; Reale, A.; Di Carlo, A.; Brown, T. M. Manuscript submitted. 60 Carter, S. A.; Angelopoulos, M.; Karg, S.; Brock, P. J.; Scott, J. C. Appl. Phys. Lett. 1997, 70, 2067–2069. 61 Brown, T. M.; Cacialli, F. J. Polym. Sci. Part B: Polym. Phys. 2003, 41, 2649–2664. 62 Brown, T. M.; Kim, J. S.; Friend, R. H.; Cacialli, F.; Daik, R.; Feast, W. J. Appl. Phys. Lett. 1999, 75, 1679–1681. 63 Groenendaal, L.; Jonas, F.; Freitag, D.; Pielartzik, H.; Reynolds, J. R. Adv. Mater. 2000, 12, 481–484. 64 Granstrom, M.; Petritsch, K.; Arias, A. C.; Lux, A.; Andersson, M. R.; Friend, R. H. Nature 1998, 395, 257–260. 65 Chen, Z.; Cotterell, B.; Wang, W.; Guenther, E.; Chua, S. J. Thin Solid Films 2001, 394, 202–206. 66 Chen, Z.; Cotterell, B.; Wang, W. Eng. Fract. Mech. 2002, 69, 597–603. 67 Hutchinson, J. W.; Suo, Z. Adv. Appl. Mech. 1992, 29, 63–191. 68 Cotterell, B.; Chen. Z. Int. J. Fract. 2000, 7, 104–169. JOURNAL OF POLYMER SCIENCE PART B: POLYMER PHYSICS 2011, 49, 638–648 647 FULL PAPER 69 Lewis, J. Mater. Today 2006, 6, 38–45. 70 Yong, Q.; Lian, D.; Liduo, W. Chin. Sci. Bull. 2002, 47, 1979– 1982. 648 WWW.POLYMERPHYSICS.ORG 71 Na, S. I.; Kim, S. S.; Jo, J.; Kim, D. Y. Adv. Mater. 2008, 20, 4061–4067. 77 Tonelli, A. E. Polymer 2002, 43, 637–642. 78 Han, H.; Adams, D.; Meyer, J. W.; Alford, T. L. J. Appl. Phys. 2005, DOI: 10.1063/1.2106013. 79 Vos, J. J. Color Res. Appl. 1978, 3, 125–128. 72 Wang, G. F.; Tao, X. M.; Wang, R. X. Nanotechnology 2008, 19, DOI: 10.1088/0957-4484/19/14/145201. 80 Scheirs, J.; Gardette, J. L. Polym. Degrad. Stab. 1997, 56, 339– 350. 73 Wu, J.; Becerill, H. A.; Bao, Z.; Liu, Z.; Chen, Y.; Peumans, P. Appl. Phys. Lett. 2008, 92, 263302 1–3. 74 Li, J.; Hu, L.; Wang, L.; Zhou, Y.; Gruner, G.; Marks, T. J. Nano Lett. 2006, 6, 2472–2477. 75 Ashlswede, E.; Muhleisen, W.; Wahinuddin bin Moh Wahi, M.; Hanisch, J.; Powalla, M. Appl. Phys. Lett. 2008, DOI: 10.1063/1.2907564. 81 Allen, N.; Edge, M.; Mohammadian, M.; Jones, K. Polym. Degrad. Stab. 1994, 43, 229–237. 82 Fechine, G. J. M.; Souto-Maior, R. M.; Rabello, M. S. J. Appl. Polym. Sci. 2007, 104, 51–57. 83 Stabilization of Polymeric Materials State-of-the-Art, Scope and Limitations, http://academic.sun.ac.za/UNESCO/PolymerED 2000/Conf2000/ZweifelC%28s%29.pdf. 76 Space Environment Technologies - ISO 21348, http://www. spacewx.com/Docs/ISO_PRF_21348_e.pdf. 84 Kim, J. S.; Friend, R. H.; Cacialli, F. J. Appl. Phys. 1999, 86, 2774–2778. JOURNAL OF POLYMER SCIENCE PART B: POLYMER PHYSICS 2011, 49, 638–648