Langmuir 2006, 22, 4985-4991
4985
Surface-Deformation Characteristics of Uniaxially Oriented Poly(ethylene terephthalate) Film as Evaluated from Nanoscratch Tests with Scanning Probe Microscopy Hiroki Uehara,* Tatsuki Asakawa, Masaki Kakiage, Takeshi Yamanobe, and Tadashi Komoto Department of Chemistry, Gunma UniVersity, Kiryu, Gunma 376-8515, Japan ReceiVed January 17, 2006. In Final Form: March 17, 2006 The surface-deformation characteristics of uniaxially drawn poly(ethylene terephthalate) (PET) film were successfully evaluated with multiline scratch tests using scanning probe microscopy (SPM) on a nanometer scale. The PET film was prepared by compression molding from the melt, followed by quenching in ice water. The obtained amorphous film was drawn uniaxially below its glass-transition temperature, and the resultant surface roughness could be reduced to within 5 nm. A multiline scratch with the Si3N4 tip of an SPM on the oriented PET surface was made parallel and perpendicular to the drawing axis under applied loads of 5-30 nN. The perpendicular scratching generated a characteristic periodic pattern on the film surface, but the parallel scratching induced a tearing of the surface. These results suggest that surface-deformation mechanisms were dominated by molecular anisotropy. The surface-deformation properties, as evaluated from scratch-angle dependences on morphological changes on a nanometer scale, were similar to the mechanical properties of the bulk.
Introduction Recently, it has been recognized that the morphology of the material surface dominates the surface properties (e.g., friction, lubrication, and wear) on a nanometer scale.1 For polymeric materials in particular, various structures and morphologies can easily be introduced onto the surfaces because of their high process ability. Therefore, the relationship between the structure and properties of the polymeric surface is clarified at the nanoscale level (i.e., nanometer and nanonewton). Scanning probe microscopy (SPM) measurement is a typical method that enables such nanoscale analysis. SPM not only observes surface morphologies, but it also evaluates surface properties by applying various measurement modes, including indentation,2-4 lithography,5 and tapping.6-11 Tanaka et al.11 analyzed the viscoelasticity of polystyrene (PS) surfaces by using SPM to estimate its glass-transition temperature (Tg). The obtained surface Tg was lower than the bulk Tg estimated from differential scanning calorimetry (DSC) measurement. Such a low surface Tg decreases with decreasing sample molecular weight (MW). For PS having a MW of 3 × 105, the surface Tg is equal to room temperature. Similar results have been confirmed by other methods, including thermal probe,12 fluorescence labeling,13 DSC,14 and ellipsometric measurements.15-17 * Corresponding author. E-mail:
[email protected]. (1) Assender, H.; Bliznyuk, V.; Porfyrakis, K. Science 2002, 297, 973. (2) Beake, B. D.; Leggett, G. J. Polymer 2002, 43, 319. (3) Du, B.; Zhang, J.; Zhang, Q.; Yang, D.; He, T.; Tsui, O. K. C. Macromolecules 2000, 33, 7521. (4) Du, B.; Tsui, O. K. C.; Zhang, O.; He, T. Langmuir 2001, 17, 3286. (5) Lyuksyutov, S.; Vaia, R.; Paramonov, P. B.; Juhl, S.; Waterhouse, L.; Ralich, R. M.; Sigalov, G.; Sancaktar, E. Nat. Mater. 2003, 2, 468. (6) Tanaka, K.; Taura, A.; Ge, S.; Takahara, A.; Kajiyama, T. Macromolecules 1996, 29, 3040. (7) Kajiyama, T.; Tanaka, K.; Satomi, N.; Takahara, A. Macromolecules 1998, 31, 5150. (8) Satomi, N.; Takahara, A.; Kajiyama, T. Macromolecules 1999, 32, 4474. (9) Satomi, N.; Tanaka, K.; Takahara, A.; Kajiyama, T. Macromolecules 2001, 34, 6423. (10) Tanaka, K.; Hashimoto, K.; Kajiyama, T.; Takahara, A. Langmuir 2003, 19, 6573. (11) Tanaka, K.; Takahara, A.; Kajiyama, T.; Macromolecules 1997, 30, 6626. (12) Fischer, H. Macromolecules 2002, 35, 3595. (13) Ellison, C. J.; Torkelson, J. M. Nat. Mater. 2003, 2, 695.
In contrast, we18-20 evaluated the surface-deformation properties of PS by nanoscratch tests21-25 applying SPM tip friction. The obtained results indicated that surface deformation is also dominated by such molecular characteristics as MW and chemical compositions.19,20 The SPM scratch tests on a series of PS surfaces having different MWs exhibited debris formation caused by brittle breaking of the lower MW surface and periodic ridges due to plastic deformation of the higher MW surface.19 Transformation of these surface-deformation morphologies occurred at a critical MW of 3-5 × 104, which corresponded to the entanglement MW estimated for bulk PS. These results confirm that surface deformation is similar to tribological properties in the bulk state.18,20 Rather than focusing on entanglement, this study examined the effect of the anisotropy of polymeric chains, which is another important characteristic of polymeric materials on surface deformation. The polymeric molecules are composed of chain propagations of coherent bonds and thus have anisotropy at the molecular level. When this molecular anisotropy is assembled parallel in a given direction, the characteristic properties are induced in the bulk state.26 Such molecular anisotropy also influences the surface-deformation properties on the nanometer level. (14) Efremov, M. Y.; Olson, E. A.; Zhang, M.; Zhang, Z.; Allen, L. H. Phys. ReV. Lett. 2003, 91, 85703. (15) Sharp, J. S.; Forrest, J. A. Phys. ReV. Lett. 2003, 91, 235701. (16) Tsui, O. K. C.; Zhang, H. F. Macromolecules 2001, 34, 9139. (17) Forrest, J. A.; Veress, K. D.; Stevens, J. R.; Dutcher, J. R. Phys. ReV. Lett. 1996, 77, 2002. (18) Aoike, T.; Uehara, H.; Yamanobe, T.; Komoto, T. Langmuir 2001, 17, 2153. (19) Aoike, T.; Yamamoto, T.; Uehara, H.; Yamanobe, T.; Komoto, T. Langmuir 2001, 17, 5688. (20) Aoike, T.; Ikeda, T.; Uehara, H.; Yamanobe, T.; Komoto, T. Langmuir 2002, 18, 2949. (21) Leung, O. M.; Goh, M. C. Science 1992, 255, 64. (22) Jing, J.; Henriksen, P. N.; Wang, H. J. Mater. Sci. 1995, 30, 5700. (23) Schmidt, R. H.; Haugstad, G.; Gladfelter, W. L. Langmuir 2003, 19, 898. (24) Schmidt, R. H.; Haugstad, G.; Gladfelter, W. L. Langmuir 2003, 19, 10390. (25) Beake, B. D.; Leggett, G. J.; Shipway, P. H. Polymer 2001, 42, 7025. (26) Wunderlich, B. Macromolecular Physics; Academic Press: New York, 1973.
10.1021/la0601612 CCC: $33.50 © 2006 American Chemical Society Published on Web 05/02/2006
4986 Langmuir, Vol. 22, No. 11, 2006
Uehara et al.
Table 1. Intrinsic Viscosity and Estimated MW for the Samples Used in This Study
a
sample
i.v.a (dL/g)
Mn b (∼10-3)
lower MW higher MW
0.48 0.84
13 25
Intrinsic viscosity. b Number average MW. Table 2. Density and Crystallinity of Oriented Films
a
sample
DRa
Fb (g/cm3)
χcc (%)
lower MW higher MW
4.5 4.7
1.364 1.367
27 30
Tensile draw ratio. b Measured density. c Crystallinity.
SPM nanoscratch tests were performed on the surface of the oriented film of polymeric material. Here, a high chain orientation is difficult to introduce for amorphous polymers such as PS, but is easy for semicrystalline polymers by tensile drawing. Furthermore, a flat surface is required for accurate SPM nanoscratch tests. However, the oriented films of semicrystalline polymers often contain a fibril structure with a priori unevenness beyond the nanometer scale. Therefore, poly(ethylene terephthalate) (PET) was selected in this study because it retains a flat surface, even after a drawing procedure. Experimental Section Materials. A PET pellet supplied from Aldrich Chemical Co. was used as the starting material. This material is called the “lower MW sample” in this study. The as-received PET was polymerized in the solid state by vacuuming at 220 °C for 36 h. This study refers to the obtained material as the “higher MW sample.” The intrinsic viscosities (i.v.) of these two PET materials were evaluated at 30 °C by using a solution viscosity measurement apparatus (Shibayama Scientific Co., Ltd.) equipped with a Ubbelohde viscometer. The polymer solution was prepared by dissolving these PET materials into a mixture of phenol/1,1,2,2-tetrachloroethane ) 1/1. The number average MW (Mn) was calculated from the obtained intrinsic viscosity assuming i.v. ) KMna
(1)
where K is 2.10 × 10-4 and a is 0.82.27 The resultant values are listed in Table 1. Film Preparation. Each sample was dried in a vacuum at 100 °C and subsequently sandwiched between two polyimide films to be compression molded at 270 °C, which exceeds the sample melting temperature (Tm) of 245 °C. Next, the assembly was quenched in iced water, and the obtained amorphous film was 0.5 mm thick. This film was uniaxially tensile-drawn at 70 °C, which is below the glasstransition temperature (Tg) of PET (∼100 °C), up to a draw ratio (DR) of ∼5 using an Orientec Tensilon RTC-1325A tensile tester. The initial strain rate on the tensile drawing was 1 min-1. The initial unevenness of the drawn film was limited to within a 5 nm height, as evaluated by noncontact SPM measurement. The densities of the prepared films (F) were estimated by using a density gradient column composed of a mixture of n-heptane and tetrachloroethylene. The sample crystallinities (χc) were calculated from the measured density assuming the following equation for the crystalline/amorphous twophase model: χc ) Fc(F - Fa)/F(Fc - Fa)
(2)
where the crystalline density (Fc) was 1.455 g/cm3, and the amorphous density (Fa) was 1.333 g/cm3.28 As presented in Table 2, the χc values of the prepared films were close to each other; thus, the χc dependence on surface deformation was negligible in this study. (27) Conix, A. Makromol. Chem. 1958, 26, 226. (28) Huang, B.; Ito, M.; Kanamoto, T. Polymer 1994, 35, 1210.
SPM Nanoscratch Tests. The SPM used in this study was a Seiko Instruments SPA 400, equipped with an Olympus triangle cantilever having a spring constant of 0.57 N/m. A 30 nm radius Si3N4 tip was provided at the rear of the top of the cantilever. All SPM measurements were made at room temperature under a flow of dried nitrogen, which minimized the humidity to less than 20%. As a first scan, the SPM tip was pushed on the film surface with applied loads of 5-30 nN and subsequently scratched by 256 lines within a 1 × 1 µm area. The scratch direction was varied between 0 (parallel) and 90° (perpendicular) to the predrawing axis of the film, but always perpendicular to the longitudinal axis of the cantilever. After this multiline scratch, a second scan was made under a low applied load below 1 nN without any deformation within a 1.5 × 1.5 µm area, including the first scratched area. The scan speed was 1 µm/s for both scans. Tensile Tests. Tensile tests were conducted at room temperature with an Orientec Tensilon RTC-1325A tensile tester. The stressstrain curves were recorded under an initial strain rate of 1 min-1. The tensile modulus was estimated from the initial slope of the stress-strain curve. X-ray Diffraction. Wide-angle X-ray diffraction (WAXD) patterns of the oriented films were recorded on a Fuji Film imaging plate BAS-SR. Cu-Ka radiation monochromatized with a Ni filter was generated at 40 kV and 200 mA by a Rigaku RU-300 rotating anode X-ray generator. The incident beam was radiated perpendicular to the surface of the oriented PET film. The recorded imaging plate pattern was read out using a Rigaku R-AXIS DS3.
Results and Discussions Molecular Orientation Analysis by WAXD. WAXD measurements were made for analyses of the molecular orientation of the uniaxially drawn PET films. Figure 1 compares the WAXD patterns for a DR ) 4.5 film prepared from the lower MW sample (a) and a DR ) 4.7 film from the higher MW sample (b). The draw direction was vertical. For both films, strong scatterings on the equator were ascribed to the oriented amorphous halo. Weak diffractions attributed to the triclinic crystalline form29 were also observable. These features were almost identical for both films, indicating that similar chain orientation was obtained, independent of the sample MW. Surface Deformation for the Lower MW Sample. A series of SPM nanoscratch tests under different conditions were performed on these oriented PET films to analyze surfacedeformation behavior. Figure 2 presents the topographical images observed after nanoscratch tests for the oriented lower MW film with a DR of 4.5. Nanoscratch tests were made parallel (0°) and perpendicular (90°) to the predrawing axis under applied loads from 5-30 nN. Even for the nanoscratch under the lowest applied load of 5 nN, wear debris surrounding the scratched area was recognizable. Such wear debris formation induced by the SPM nanoscratch was also confirmed for the lower MW PS surface.19 As the applied load was increased, the height difference, indicated by color gradation, became significant. However, a periodic ridge arranged perpendicular to the scratch direction was always obtained under all loads examined in Figure 2. In terms of this wear debris formation and periodic ridge arrangement, the morphologies induced by 0 and 90° scratches are quite similar. Therefore, the anisotropy of the surface deformation was less pronounced for the lower MW sample, compared to the higher MW sample shown later in Figure 3. The origin of the gradual increase of the period of the ridge arrangement will be discussed later. Surface Deformation for the Higher MW Sample. Correspondingly, the surface of the oriented higher MW film with (29) Ran, S.; Wang, Z.; Burger, C.; Chu, B.; Hsiao, B. C. Macromolecules 2002, 35, 10102.
Surface Deformation of Uniaxially Drawn PET Films
Langmuir, Vol. 22, No. 11, 2006 4987
Figure 1. WAXD patterns for the uniaxially drawn PET films used in this study. (a) DR ) 4.5 film prepared from the lower MW sample (i.v. ) 4.8 dL/g). (b) DR ) 4.7 film from the higher MW sample (i.v. ) 0.84 dL/g).
a DR of 4.7 was nanoscratched along the 0 and 90° directions to the predrawing axis, and the obtained surface-deformation patterns were compared (Figure 3). For a series of 0° scratches, the characteristic fibrillar structure extending along the predrawing axis was observed when the applied load was increased beyond 20 nN. In contrast, the periodic patterns were obtained for all 90° scratches made under all loads examined in Figure 3. Therefore, the nanoscratch tests for the oriented higher MW film exhibited significant molecular anisotropy. Furthermore, wear debris was not observed, even at the highest applied load of 30 nN; this result was quite different from the result for the lower MW sample (Figure 2). This finding suggests that the mechanical properties of the film surface increase with increasing sample MW. It could be assumed that these periodic and fibrillar patterns formed through different deformation mechanisms. First, the origin of the fibrillar structure was considered from the series of images observed after the 0° scratches. Figure 4 depicts the cross-sectional profiles extracted perpendicular to the fibrillar ridges on the images in Figure 3. With an increasing applied load, the channels dividing the fibrillar ridges deepened. These results implied that the tip moving at the nanoscratch caused the tearing of the surface region along the drawing axis, giving the resultant fibrillar structure. To confirm this assumption, 0° scratches with reduced line numbers of 128, 64, 32, and 16 lines were made under an applied load of 30 nN, and the obtained results were compared to those with the usual
Figure 2. Topographical images of the scratched surfaces for the lower MW film uniaxially drawn up to DR ) 4.5. The scratches were made parallel (0°) and perpendicular (90°) to the oriented axis denoted by the arrows. The indexes on the left of the images indicate the applied loads. Scan size is 1.5 × 1.5 µm2.
256 lines. Figure 5 presents the sets of the images observed after these nanoscratches with the reduced line numbers and their cross-sectional profiles extracted perpendicular to the scratch direction. The image obtained after a 128-line scratch was almost identical to that observed after a 256-line scratch, but further decreases in the number of scratch lines gave a reduction in the number of fibrillar ridges. For a 16-line scratch in particular, the same number of channels (16) was recognized in both the image and the cross-sectional profile. These results confirmed the assumption that the tearing of the surface region along the predrawing axis forms such a fibrillar structure during the SPM nanoscratching. The origin of the periodic pattern obtained for a 90° scratch was also considered. Figure 6 exhibits the sets of 90°-scratched images shown in Figure 3 and their cross-sectional profiles extracted along the scratch direction. For comparison, the image of the original film surface observed before the nanoscratch and its cross-sectional profile are also shown. With an increasing applied load, the height difference within the scratched area increased, but the number of periodic ridges decreased. A similar phenomenon was observed even for the oriented lower MW film, as illustrated in Figure 2. These experimental results indicate
4988 Langmuir, Vol. 22, No. 11, 2006
Uehara et al.
Figure 4. Comparison of cross-sectional profiles (right) extracted along the blue dotted line in the set of left images observed after 0° scratches (same as shown in the left column of Figure 3) from top to bottom. Dotted vertical lines in the right columns indicate the 0°-scratched region in the left images. Narrow horizontal lines correspond to the original heights estimated from unscratched regions outside the dotted lines.
Figure 3. Topographical images for the scratched surfaces of the higher MW film uniaxially drawn up to DR ) 4.7. Left (right) column corresponds to the 0° (90°) scratched surface. The arrows indicate the predrawing axes. Scan size is 1.5 × 1.5 µm2.
that the tip pushing the oriented molecular chains at the film surface causes their bundle division, which is similar to the movement of the strips of blind when they were laterally flipped. Bulk Deformation Behavior. These surface deformations were compared to the bulk deformations evaluated by tensile tests. Figure 7 compares the stress-strain curves recorded when the oriented films of the higher MW sample were tensile stretched parallel (0°) and perpendicular (90°) to the predrawing axis. The testing specimens were cut as indicated in the inset sketch. The tensile deformation along the 0° direction exhibited a higher yielding stress and a shorter breaking strain. In contrast, the necking behavior induced a longer breaking strain for the tensile test along the 90° direction. These results indicate that the bulk film easily deforms perpendicular to the predrawing axis as well as the film surface. The stress-strain curves for the lower MW sample similarly showed the higher yielding stress along 0° direction and the lower one along 90° direction. However, the breaking strain approached up to 800% for both directions, which is coincident with the less pronounced anisotropy estimated from the surface deformations induced by nanoscratch tests, as shown in Figure 2. Scratch Angle Dependence on Surface Deformation. Quite different morphologies (i.e., fibrillar structure and periodic pattern)
were observed for 0 and 90° scratches of the oriented higher MW film. This result implied that the transition from the former to the latter morphologies would occur at a critical scratch angle to the predrawing axis. Therefore, surface deformations were recorded with a gradual change in the scratch direction from 0 to 90°. Figure 8 compares the images observed after these nanoscratches were made every 15° to the drawing axis. In these images, the scratch direction in each image was adjusted to the horizontal direction. Periodic patterns arranged perpendicular to the scratch direction and the fibrillar structure along the predrawing axis coexisted for the 15° scratch. When the scratch angle was increased to 30 and 45°, the fibrillar alignment gradually appeared along the scratch direction. Such scratch angle dependence on surface deformation suggested that the tearing of the surface region caused the fibrillar structure as well as the scratch line number dependence, depicted in Figure 5. However, this fibrillar structure almost disappeared for the 60° scratch, and the other characteristic surface deformation of the periodic pattern appeared instead. Because of a transition of surface deformation morphologies, the height difference within the scratched area became less pronounced at the middle range of the scratch angle. Further increase of the scratch angle emphasized the height difference because it was the only remaining periodic pattern. A corresponding series of nanoscratches made every 15° to the drawing axis were also made under the lower applied load of 5 nN (Figure 9). In this case, the 0° scratch did not exhibit any recognizable surface deformation, as indicated in Figure 3. For the 15° scratch, no surface deformation was found, but a periodic pattern appeared beyond the 30° scratch.
Surface Deformation of Uniaxially Drawn PET Films
Figure 5. Influence of the number of scratches on surface deformation for the oriented higher MW film. Left (right) column indicates the set of topographical images of the 0° scratched surface (corresponding cross-sectional profiles). The lines in the images and profiles have the same meanings as in Figure 4. In the bottom profile, the arrows denote the channels produced by multiline scratches. The top sets (256-line scratch) are the same as those in Figure 4.
Such scratch angle dependence on surface deformation was compared to stretch angle dependence on bulk deformation. Figure 10 summarizes the tensile modulus and yield stress recorded every 15° to the predrawing axis for the oriented higher MW film. From 0 to 15°, both the yield stress and the modulus decreased significantly. In contrast, constant values were obtained beyond 45°. A similar significant change in deformation behavior between 0 and 15° was observed for nanoscratches. This coincidence also suggests similarity between surface and bulk deformation behaviors. Mechanism of Surface Deformation Induced by SPM Nanoscratching. Various deformation mechanisms have been proposed for polymeric surfaces (i.e., stick-slip,30 peeling,31 and crack-opening23,32). However, discussions of these mechanisms have been based on the results of nanoscratch tests for the isotropic surfaces of PS. The effect of molecular anisotropy on surface deformation has not been considered. In contrast, our nanoscratch tests were performed for oriented films. Additionally, the (30) Beake, B. D.; Shipway, P. H.; Leggett, G. J. Wear 2004, 256, 118. (31) Elkaakour, Z.; Aime, J. P.; Bouhacia, T.; Odin, C.; Masuda, T. Phys. ReV. Lett. 1994, 73, 3231. (32) Khurshudov, A.; Kato, K. Wear 1997, 205, 1.
Langmuir, Vol. 22, No. 11, 2006 4989
Figure 6. Comparison of cross-sectional profiles (right) extracted along the blue dotted line in the set of left images observed after 90° scratches (same as in the right column of Figure 3) from top to bottom. The lines in the images and profiles have the same meanings as in Figure 4. For comparison, the image and corresponding profiles obtained before scratching are also presented as the top sets.
Figure 7. Stress-strain curves for the oriented higher MW film. Tensile tests were made parallel (open triangle) and perpendicular (filled circle) to the predrawing axis. The inset illustration indicates the preparation of the tested specimens cut out from the oriented films.
mechanism with deformation-direction dependences was considered for the oriented higher MW PET film exhibiting the characteristic surface deformation. For a 0° scratch, the molecular chains were aligned parallel to the scratch direction, producing higher yield stress that restricts the molecular movement during SPM nanoscratching. Therefore, surface deformation under the lower applied load was unclear.
4990 Langmuir, Vol. 22, No. 11, 2006
Figure 8. Dependence of scratch direction to the predrawing axis on surface deformation under the higher applied load of 30 nN. The indexes in the images represent the angle between the scratch direction and the predrawing axis indicated by the arrows. The 256-line scratches were always made by repeating left-to-right and rightto-left scans.
In contrast, the higher applied load drove the SPM tip deeper into the film surface. Here, higher tensile properties in the bulk were obtained for the 0° direction compared to those obtained for the 90° direction to the predrawing axis. Thus, molecular interaction parallel to the chain direction was lower than the coherent bonding energy along the chain direction. When the SPM tip was moved during the nanoscratch test, the deformation energy was used to tear the surface region, resulting in the formation of a fibrillar morphology. In contrast, the tensile test for the 90° direction revealed that the molecules easily moved along the scratch direction. Its lower yield stress allowed the formation of the periodic pattern, even at lower applied loads. With increasing applied load, the SPM tip drove deeper into the film surface, giving molecular movement and emphasizing height difference within the deformed area. When the 256-line scratch was made within a 1 µm width of the scratch area, the spacing of the adjacent two scratches was calculated to be 4 nm. However, the radius of the SPM tip used
Uehara et al.
Figure 9. Dependence of scratch direction to the predrawing axis on surface deformation under the lower applied load of 5 nN. The indexes and the arrows have the same meanings as those given in Figure 8.
in this study was 30 nm, which means that the same position on the surface was scratched several times. This is the reason the periodic arrangement of the ridgeline is duplicated (continued) from the start to the end. Thus, the repeating of the slides of the SPM tip over these periodic ridges results in a succeeded period of ridgelines during the nanoscratching. These mechanisms of surface deformations induced by SPM nanoscratching are quite different from those previously proposed. The most significant feature of our surface deformation of the oriented PET film lies in the anisotropy of the deformation for the higher MW sample. Recently, Leggete et al.25,30 reported the results of SPM nanoscratch tests for the surfaces of uniaxially and biaxially drawn PET films. In their study, periodic patterns were always observed parallel to the scratch direction, independent of the scratch angle to the predrawing axes and therefore indicating no anisotropy of surface-deformation characteristics. Considering that our nanoscratch tests of the lower MW sample exhibited only the periodic pattern under all applied loads of 5-30 nN (Figure 2), their PET sample perhaps had the same level of MW (not described in refs 25 and 30) as our lower MW sample. Also,
Surface Deformation of Uniaxially Drawn PET Films
Figure 10. Yield stress (open triangle) and tensile modulus (filled square) for the oriented higher MW film as a function of tensile-stretch angle to the predrawing axis. The 0 and 90° angles respectively correspond to the parallel and perpendicular relationships between tensile-stretch and the predrawing axes.
the less pronounced anisotropy for the lower MW sample (see Figure 2) could be ascribed to the crack-opening mechanism previously reported by Leggete et al.30
Conclusions A series of SPM nanoscratch tests for uniaxially drawn PET film were performed under various conditions. The molecular anisotropy effect on surface deformation characteristics was
Langmuir, Vol. 22, No. 11, 2006 4991
evaluated from a series of resultant surface morphologies. For the lower MW sample, a periodic ridge arranged along the scratched direction was obtained, independent of scratch conditions, indicating that there was no anisotropy for lower MW surface. In contrast, for the higher MW sample, significantly different surface deformations were obtained with varying scratch directions. For a 0° scratch, a fibrillar structure ascribed to the surface tearing induced by the tip scratch was formed parallel to the predrawing axis. However, a periodic ridge arrangement was clearly observed for a 90° scratch. Considering that a similar periodic pattern was observed for the lower MW film, the periodic ridge formation could be attributed to the molecular movement on the surface region induced by SPM nanoscratching. These characteristic surface deformations were compared to bulk deformations evaluated by tensile tests. For a 0° stretch, a higher yield stress and a lower breaking strain were observed. In contrast, a lower yield stress with a necking phenomenon induced a significantly higher breaking strain for a 90° stretch. These results indicated that the molecular chains were easily moved perpendicular to the predrawing axis, even in the bulk state. This bulk deformation behavior corresponded well to the surface deformation evaluated by a series of SPM nanoscratch tests. Therefore, the deformation within the surface region is similar to that of bulk deformation. Acknowledgment. This work was partly supported by The Shimadzu Science Foundation. We would also like to express our appreciation for the discussions with Dr. Taku Aoike concerning this study. LA0601612
