Comparison of test methods estimating the stiffness of ultrathin coatings

Comparison of test methods estimating the stiffness of ultrathin coatings J. Coat. Technol. Res., 15 (4) 743–752, 2018 https://doi.org/10.1007/s11998-018-0085-0 Comparison of test methods estimating the stiffness of ultrathin coatings Marcus Vinı´cius Tavares da Costa , Cristian Neagu, Pierre Fayet, Urban Wiklund, Hu Li, Klaus Leifer, E. Kristofer Gamstedt The Author(s) 2018 Abstract A key engineering parameter of thin coatings measurements dominated by the coating. The strain- is their stiffness. Stiffness characterization of ultrathin induced elastic buckling method is simple in practice, coatings with a nanometer scale thickness is experimen- but showed a large scatter due to variation in local tally challenging. In this work, three feasible methods coating thickness and irregular deformation patterns. have been used to estimate the Young’s modulus of The stiffness characterization using atomic force micro- metal coatings on polymer films. The methods are: (1) scopy gave the most consistent results, due to a sharp tip nanoindentation, (2) strain-induced elastic buckling and with a radius comparable to the thinnest coating (3) peak-force measurements integrated in atomic force thickness. All methods gave a higher Young’s modulus microscopy. The samples were prepared by atomic layer for the TiO coating than for the mixed oxide coating, deposition of TiO (6 and 20 nm thick) and mixed oxides with a variation within one order of magnitude between of TiO and Al O (4 and 20 nm thick). The differences the methods. 2 2 3 in estimated Young’s modulus are interpreted in terms of the underlying assumptions and test conditions. Their Keywords Nano coating, Young’s modulus, Polymer specific advantages and drawbacks are also compared substrate and discussed. In particular, the nanoindentation neces- sitates a sufficiently sharp indenter tip to make localized Introduction This paper was presented at the 13th Coatings Science Inter- national Conference (COSI) on June 26–30, 2017, in Noordwijk, Carton food packages are indeed an important inven- The Netherlands tion. Every day, billions of liters of water, milk, juice and other liquid foods are consumed around the world. M. V. Tavares da Costa, E. K. Gamstedt (&) Carton packaging is mostly used to contain and protect Division of Applied Mechanics, Department of Engineering beverages so that they can be transported to the Sciences, Uppsala University, Box 534, 751 21 Uppsala, consumers with increased lifetime. The barrier func- Sweden e-mail: kristofer.gamstedt@angstrom.uu.se tion is today mainly assured by aluminum foils. This material protects food from the outside (intrusion of M. V. Tavares da Costa oxygen, humidity, etc.), as well as prevents leakage by e-mail: marcus.tavares@angstrom.uu.se keeping the nutrients inside. One of the potential candidates which can replace aluminum foil is a thin C. Neagu Tetra Pak AB, DSO Packaging Materials, Ruben Rausings metal oxide brittle coating deposited on a polymer film Gata, 22186 Lund, Sweden substrate. A number of material properties are typi- cally of interest in coatings development. One of these P. Fayet properties is the stiffness, which affects the strain and Adhemon Sarl, Thin Technology, Avenue Edouard-Dapples stress state in coating structures. It is needed in the 20, 1006 Lausanne, Switzerland estimation of interfacial fracture toughness in coatings subject to mechanical loading. U. Wiklund, H. Li, K. Leifer There are some well-known methods to characterize Division of Applied Materials Science, Department of coatings on polymers such as the fragmentation test. Engineering Sciences, Uppsala University, Box 534, This test can be used to calculate the interfacial shear 751 21 Uppsala, Sweden 743 J. Coat. Technol. Res., 15 (4) 743–752, 2018 strength between coating and polymer substrate, based Tensile testing on the observation of crack density of the coating, as an increasing tensile strain is being applied. Analytical The Young’s modulus of the substrate was obtained by models have been developed to determine adhesive tensile test from the slope of the initial linear regime of properties of the interface from the crack density at the stress–strain diagram. Three tests were carried out saturation. Finite element simulation has also been for each barrier coating. The influence of the ultrathin used for the coating/substrate assemblies to determine coating on the film stiffness was substantially smaller 5,6 the cohesive fracture toughness of the interface. To than the measurement scatter. The samples were cut in use these analytical or numerical approaches, it is first small pieces with dimensions 4 mm 9 20 mm using necessary to determine the elastic properties of the scalpel and ruler. The samples were stretched up to coating. Most coatings are assumed to be isotropic, and 40% of nominal strain, the speed was 0.1 mm/min, and the foremost parameter is then the Young’s modulus, the gauge length was > 10 mm. All data were mea- E, since the Poisson ratio, m, does not show as much sured by the Deben ‘‘Microtest’’ tensile stage, and the variation and has less influence on adhesive behavior. scatter was taken as the standard deviation of the test The determination of coating Young’s modulus is of results. particular interest for thin coatings, where it is known to depend on the coating thickness. The focus of this work is therefore to investigate the Nanoindentation Young’s modulus of thin metal oxide coating using The main principle of this technique is the nanoscale independent methods. Three experimental methods indentation of a rigid indenter into the surface of a have been used to obtain values of the Young’s modulus, deformable material. From the load–deflection curve namely nanoindentation, mechanical buckling measure- of the unloading phase, the elastic properties close to ment and Peak-Force Quantitative Nanomechanical the surface can be obtained. The CSM Ultra Nano Property Mapping technique (PF-QNM). Since the ultrathin coatings (thickness in the order of 10 nm) do Hardness Tester with a diamond cube corner indenter not allow for stiffness characterization by in-plane of 40 nm tip radius was used for all ALD samples. At tensile testing, all three methods are based on local the time of testing, this was the sharpest tip available out-of-plane deformations. Using three complementary for this particular nanoindentation device. The inden- characterization methods for the same material also tation depth was up to 30 nm. The loading rate was gives the opportunity to compare the methods in chosen to be 100 lN/min, as a compromise to be slow practice and identify the different advantages and enough to obtain stable data and fast enough to avoid drawbacks of the methods for the present class of viscous effects. A holding time of 10 s before unloading coating materials. Furthermore, this work also seeks to was chosen to let the material set, and thus avoid determine the mechanical properties of coating alterna- viscous effects in the unloading curve. The sample size tives to aluminum and their thickness dependency. was around 5 mm 9 5 mm, and 10 indentations were made for each sample from which the average Young’s modulus of the coating was calculated. Each specimen was attached using double-sided adhesive tape on top Materials and methods of the horizontal sample holder. The influence of the adhesive tape was neglected since the indentation Samples depth was extremely shallow. The contact stiffness was experimentally measured from the initial portion of The materials in this study were selected based on their 10,11 unloading curve barrier function for carton food packages. All coatings were deposited by atomic layer deposition (ALD) on pffiffiffiffi dP 2 polymer film substrate. The materials of these coatings S ¼ ¼ mP ðh  h Þ ¼ bpffiffiffi E A max max p r dh p are TiO (titanium oxide) and MOX (mixed oxide, max composed of even contributions of TiO and Al O ). 2 2 3 ð1Þ The thickness of these coatings was 20 nm for the thicker coatings of both materials, 4 nm for the thin where E is the reduced modulus, A is the projected MOX coating and 6 nm for the thin TiO . The 2 area, h  h is the total penetration depth, P is max p max thickness was determined by the number of deposition the maximum force, and m and b are constants which cycles since only one atomic layer is produced for each depend on the geometry of the indenter. The value of b cycle. The substrate, biaxially oriented polyethylene was set to 1.0 for a hemispherical indenter, and the terephthalate (BoPET) film 120-lm-thick Teijin Meli- projected area A and constant m were experimentally nex ST504 from Dupont, was the same for all barrier calibrated by the nanoindentation device. The coatings. The thickness of the coating is negligible calibration of A accounts for the assumed spherical compared with the thickness of the substrate, and the tip. Once the reduced modulus E is known, the in-plane stiffness of the coated film is therefore entirely Young’s modulus of the coating, E , is obtained 9–12 dominated by the stiffness of the substrate. through 744 J. Coat. Technol. Res., 15 (4) 743–752, 2018 1  m as a function of the thickness of the coating, h , the E ¼ ð2Þ 1 1m Young’s modulus and Poisson’s ratios of the coating E E r i and the substrate using Euler derivative equation. Later, Stafford et al. used this equation to deter- where E is the Young’s modulus of the indenter, 1145 mine the Young’s modulus of the coating using a GPa. The Poisson ratios of the indenter and coating, designed tensile stage with laser diffraction to measure v and v are 0.070 and 0.30, respectively, for the i c the wavelength of the buckled coating. Solving for E diamond indenter and titanium and aluminum oxide gives SIEBIMM main equation coatings. 3 1  m k E ¼ E : ð4Þ c s 3 2 Strain-induced elastic buckling instability 8p 1  m h for mechanical measurement A state of plane strain is assumed. In our case, the 13 1 The phenomenon of elastic buckling instability of a Poisson ratios m and m are 0.30 and 0.44, respec- c s superficial coating subjected to in-plane compressive tively. Equation (4) was also used, e.g., by Cranston strains can be instrumental in estimating the stiffness of et al. to determine the Young’s modulus of thin films the coating. This kind of test is usually abbreviated ( 70 nm) of nanofibrillated cellulose multilayers SIEBIMM in the literature, denoting strain-induced using a tensile microtest stage together with scanning elastic buckling instability for mechanical measure- electron microscopy (SEM). Such studies show that ment. The principle is to measure the buckling wave- SIEBIMM may be effective for elastic characterization length caused by an applied or residual compressive of thin layers. strain on coating/substrate–film. The analytical model- It should be noted that equation (4) is applicable ing typically requires that (1) the stiffness of the only for small deformations. Jiang et al. proposed a coating is much higher than that of the substrate, correction of equation (3) for the case of finite defor- E  E , (2) the substrate is thicker than the coating, c s mation in buckling due to substrate prestraining. By h  h , and (3) the buckling takes place in the small s c use of their approach, a corrected wavelength k can be strain regime. Figure 1 shows schematically how the calculated buckles appear as well as the important dimensions in the cross section. In our case, a tensile strain was 1=3 k ¼ kð1 þ e Þð1 þ nÞ ð5Þ 0 T applied and, due to the Poisson effect (quantified by m and m ), the elastic buckling forms by an effective where lateral contraction force F. A practical approach has been developed by Volyn- n ¼ 5eðÞ 1 þ e =32: 14 T T skii et al., where they reported the mechanism of buckling formation in the gold layer of 10 nm thick on The compressive transverse strain, e ¼m e ,is T s k PET substrate and derived the relation for the wave- determined from the applied longitudinal tensile strain length at buckling onset e . The buckling strain e and the k k sffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffi wavelength k are obtained experimentally. ð1  m ÞE 3 c In this study, four SIEBIMM tests were carried out k ¼ 2ph  ð3Þ 31  m E in situ in a Deben ‘‘Microtest’’ tensile stage installed in a Hitachi TM-1000 table-top SEM operating at 15 kV with a backscattered electron detector. For high magnification, the specimen was subsequently brought to a Zeiss Merlin field emission gun SEM used at an acceleration voltage of 2 kV and the high-efficiency secondary electron (HE-SE2) detector suited for Crack detailed topographic analysis. The considered coatings , , E v h C C C were made from nonconductive materials. Conse- quently, unwanted charging effects were observed in the SEM during the first attempts. In order to reduce the charging effects, all these samples were coated on top of barrier layers with Au–Pd conductive coating by Quorum SC7640 polaron sputter coater. , , E v h S S S The effects of the Au–Pd coating on the mechanical ε experiments are presented in Appendix A. In these tests, the coating properties were examined in the Fig. 1: Buckling formation due to an effective compressive lateral direction, measuring the buckling size before force (Poisson effect) in a coated film caused by an applied and after the conductive layer, and longitudinal direc- tensile strain 745 J. Coat. Technol. Res., 15 (4) 743–752, 2018 tion, observing the cracking progress regarding tensile where F  F is the difference between the force on adh strain in different Au–Pd thicknesses. Essentially, it the cantilever and the adhesion force, R is the tip AFM could not be demonstrated that the conductive coating radius, E is the reduced modulus, and d  d is the had any influence on the estimation of coating stiffness. sample deformation. Just as for nanoindentation, the Figure 2 illustrates how the SIEBMM test was Young’s modulus of the coating is calculated from the performed. A removable rig was designed and manu- reduced modulus by equation (2). factured to fit the Deben tensile stage in the vacuum chamber of the table-top SEM, as shown in Fig. 2a. When the buckling was clearly observed in situ during Results and discussion tensile testing in the SEM, the loading was stopped and the applied strain in the sample was fastened by Tensile testing tightening two fixation screws in the removable rig, as shown in Fig. 2b. The compact rig was then detached The in-plane tensile test confirmed that there was no and taken to the high-resolution SEM. The removable influence of the coating stiffness on the film stiffness, rig can thus swiftly be taken out from the microtensile which was effectively dominated by the substrate. The stage and moved to a high-resolution SEM while obtained value of the Young’s modulus of the substrate maintaining the tensile strain at buckling onset e ,as was comparable for all samples, and no tangible shown in Fig. 2c. difference was found compared with the same sub- strate material without any barrier coating. That also AFM peak-force measurements The Young’s modulus of the coatings was also DTM fit for elastic modulus estimated using a Multimode 8 Atomic Force Micro- scope, Bruker, in the Peak-Force Quantitative 12,18 Nanomechanical Mapping (PF-QNM) mode. A silicon probe was utilized with the tip radius of 8 nm. PF-QNM mode is a technique to quantitatively mea- 0 Deformation sure the surface mechanical properties of the sample by recording the force–separation feedback loop in the AFM as illustrated in Fig. 3. The Young’s modulus, the adhesion force, energy dissipation and the sample deformation can be extracted from this loop, where the Young’s modulus can be estimated by fitting the adh retraction part of the force–separation loop to the Derjaguin-Muller-Toporov (DMT) model : Tip-sample separation qffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffi AFM Fig. 3: Force–separation curve generated in AFM PF-QNM F  F ¼ E Rðd  d Þ ð6Þ adh 0 mode Fig. 2: Steps in the elastic buckling instability analysis: (a) The removable rig attached in tensile stage. (b) The fixation rig which is dismounted at the point of buckling formation. (c) The rig is brought to a high-resolution SEM for measurement of the buckle dimensions Force J. Coat. Technol. Res., 15 (4) 743–752, 2018 shows that the ALD manufacturing process did not nanoindentation tests, from which the coating Young’s change the elastic property of the polymer substrate. moduli were determined using equations (1) and (2)as Figure 4 shows the results from tensile test. an average from 10 measurements. These values are summarized in Table 1 together with those from the other independent test methods. For some indenta- Nanoindentation tions, in particular for the thinner coatings, viscous effects due to the polymer substrate were apparent. The contact stiffness was determined by least squares This resulted occasionally in nearly vertical unloading fitting from unloading curves starting at 98% down to curves in the force–indentation diagrams. 40% of the maximum measured force. With a holding time of 10 s before unloading, obvious viscous effects could be avoided. Figure 5 shows the results from the SIEBIMM The buckling formation was observed in situ by SEM. The strain for onset of buckling could then be determined. It was noticed that the strains for initial 3.5 buckling showed a significant variation, since the elastic instability was controlled by the local thickness and elastic properties. A high tensile strain at buckling 2.5 onset was observed for the thinner TiO and MOX coatings of 0.31 and 0.42, respectively. For 20-nm coatings, the buckling was perceptible at tensile strain of 0.1. Occasionally, cracks were found along the ridges 1.5 of the larger buckles (see Fig. 6). The wavelength of the cracked buckles was, however, found to be the same as with the uncracked ones. If a band of adjacent 0.5 buckles could not be found, the local wavelength of localized buckles was measured. TiO 6nm TiO 20nm MOX 20 nm MOX 4 nm Figure 6 shows the buckles on top of a TiO coating 2 2 of 6 nm strained at 0.31 in the longitudinal direction. Fig. 4: Young’s modulus of the coated substrate This image was taken from the center of the sample, determined from tensile testing where the load was relatively uniform. This region had 25 25 25 25 20 20 20 20 15 15 15 10 10 10 5 5 5 5 0 0 0 0 0 10 20 30 40 0 10 20 30 40 0 10 20 30 40 0 10 20 30 40 Indentation depth (nm) Indentation depth (nm) Indentation depth (nm) Indentation depth (nm) (d) (a) (b) (c) Fig. 5: Force versus indentation depth in nanoindentation of the coatings. (a) TiO 6 nm, (b) TiO 20 nm, (c) MOX 4 nm, 2 2 (d) MOX 20 nm Table 1: Summary of estimated Young’s moduli for the characterization methods Barrier coating h (nm) Tensile test Nanoindentation AFM SIEBIMM Bulk E (GPa) E (GPa) E (GPa) E (GPa) E (GPa) s c c c c TiO 6 3.5 ± 0.4 44 ± 19 45 ± 2 420 ± 140 115 TiO 20 3.2 ± 0.4 35 ± 17 76 ±264 ± 16 115 MOX 4 3.7 ± 0.2 31 ± 21 29 ± 2 240 ± 110 160–180 MOX 20 3.0 ± 0.3 23 ± 09 43 ±352 ± 13 160–180 Young’s modulus of coated polymer (GPa) Force (μN) J. Coat. Technol. Res., 15 (4) 743–752, 2018 a high and relatively uniform concentration of buckles. Comparison of results The wavelength pattern was different that those 15 14 reported by Stafford et al., Volynskii et al. and The results from the various characterization tech- Cranston et al. The buckling process is influenced by niques are summarized in Table 1, where the scatter is the stiffness ratio E /E , according to equation (3). The the standard deviation of these values. Although the c s 3 5 reported values ranged in the order 10 –10 , whereas in values of Young’s moduli of the coatings are all in the this study, the ratio is in the range of 7–25. same order of magnitude, there are considerable All coatings had similar buckling shapes, but showed differences. The TiO coating of 6 nm and MOX different dimensions. The size of all buckles was coating of 4 nm had a relatively good agreement measured in SEM, and then averaged to calculate the between the nanoindentation and AFM values, but coating Young’s modulus E using equation (4).The very high values for the SIEBIMM values. The 20-nm values are reported in Table 1. coatings showed a large variation between the different characterization methods. For comparison, bulk mod- ulus from literature is presented in the last column. AFM peak-force measurement The presented methods showed higher stiffness for the TiO coatings than for the MOX coatings, which was The indentation depth in AFM was controlled within a not the case for the literature values. The buckling few nm by a sharp AFM tip (radius of 8 nm), resulting method, SIEBIMM, showed higher Young’s moduli for both materials, compared with AFM and nanoinden- in a limited influence from the substrate, which is tation. The sensitivity and accuracy of this method is suitable for ultrathin coating measurement. Figure 7 adversely affected by the relatively low coating– shows the effective Young’s modulus maps of an area substrate stiffness ratio and inelastic deformations at approximately 1 lm for each barrier coating. Each high strains. As for the thickness dependency, only the pixel in the AFM images corresponds to a loop AFM test method showed a higher stiffness for the described previously in Fig. 3. Each image is made up of more than 250,000 measurements. The measured thicker coatings. force deformations for all pixels were used to deter- mine the average Young’s modulus of the coating by means of equations (2) and (5). Concluding remarks The average values of each coating are reported in Table 1. Four test methods were experimentally assessed for the determination of the Young’s moduli of ultrathin coatings deposited on polymer films. Tensile testing is generally unsuitable to characterize the coating stiff- ness for thin coated polymer films, since the measured stiffness is dominated by the relatively thick polymer film. Nanoindentation gave crude measures of the coating Young’s modulus, since the radius of the indenter tip was higher than the coating thickness. Viscous effects were observed, in particular for the thinly coated samples. The viscoelastic polymer substrate then affects the estimated coating stiffness. Shallow inden- tations with sharp tips on nonviscous materials would lead to improvements in this method. We tried to use Fig. 6: SEM image of a TiO coating of 6 nm stretched at 0.31 strain approaches that account the substrate effect: Song/ Fig. 7: AFM mapping of the effective Young’s modulus of the coatings. (a) TiO 6 nm, (b) TiO 20 nm, (c) MOX 4 nm, 2 2 (d) MOX 20 nm 748 J. Coat. Technol. Res., 15 (4) 743–752, 2018 22 23 21 Pharr, Hay/Crawford and King models. These Acknowledgments Financial support and supply of models were developed in a range of hundred nanome- the test samples for this study by Tetra Pak Packaging ters of coating thickness and depend of normalized Solutions AB, Development and Service Operations contact area over coating thickness. It turned out that (DSO), are much appreciated. MVTdC is grateful to the values of the coating modulus did not improve. the European Commission within the framework of the The buckling method, SIEBIMM, has the advantage Erasmus Mundus Programme, Action 2-STRAND 1, of being straightforward, not requiring an advanced Lot 9, Brazil for PhD studies. The authors would like apparatus to measure local loads and displacements. In to thank Mr. Luimar Correa for the technical our case, the buckles appeared at relatively high assistance in the SEM experiments and Dr. Jessica strains, where the polymer substrate was subject to Bolinsson for valuable scientific discussions. some inelastic deformation. The moduli estimated using this method were much larger than those from Open Access This article is distributed under the the other methods. With a fully elastic and relatively terms of the Creative Commons Attribution 4.0 compliant substrate, such as an elastomer, the clarity of International License (http://creativecommons.org/lice the buckles and the calculated stiffness values would nses/by/4.0/), which permits unrestricted use, distribu- most likely be improved. tion, and reproduction in any medium, provided you The AFM method showed the most consistent give appropriate credit to the original author(s) and results compared with the other methods. In this case, the source, provide a link to the Creative Commons the tip radius was much smaller than that used in license, and indicate if changes were made. nanoindentation, which provides better local deforma- tion. For nanoindentation, it is likely pushing the coating down rather than being measured from a point load at a pyramidal tip (Oliver Pharr method). Addi- tionally, the indentation depth in AFM is small enough Appendix A: Effects of the conductive layer to have only a limited influence of the viscoelastic in the barrier coating properties polymer substrate. A considerable amount of stiffness data is generated, since the AFM equipment automat- Observation of buckling caused by elastic instability in ically maps a given area. This leads to average values of thin coatings typically requires a conductive coating in higher statistical confidence compared with few point- SEM analysis since the features of interest are in the wise measurements in nanoindentation. sub-micron scale. In order to reduce charging effects, Although the variation was large in absolute num- all the samples were coated on top of the barrier layer bers, all methods showed the same trend with a higher with a Au–Pd conductive coating by Quorum SC7640 Young’s modulus for the TiO coating than for the polaron sputter coater. Au–Pd is one of thinnest MOX coating (TiO and Al O mixture). This is 2 2 3 24 continuous coating materials for SEM purposes. For consistent with Cherneva et al., who measured a the deposition process the following parameters were slightly higher modulus for TiO than for Al O . 2 2 3 used: an accelerate voltage of 2 kV, a plasma discharge The influence of the coating thickness on Young’s current of 20 mA, a pressure between 5 and 8 Pa and a modulus of the coating was not conclusive. The AFM deposition time of 1 min. Using these parameters method showed a higher stiffness for the thicker provides a layer of 8.5 nm thick, which was the coating, whereas the other methods showed the reverse minimum thickness allowing observation of surface order, which was also found by Chen et al. buckling and fracture without excessive charging To estimate Young’s modulus of coatings with a effects. The issue to address here is whether or not thickness of less than 0.1 lm is indeed challenging. The this additional layer would have any influence on the present paper presents a number of candidate meth- buckling formation and size. ods, each with their specific advantages and drawbacks. To investigate the influence of the Au–Pd coating, Both AFM and nanoindentation are based on pushing two different test samples were produced with the TiO in and retracting a rigid tip. AFM allows for more 20-nm coating. In the first one, the sample was shallow indentation, which makes it more suitable for stretched up to a strain level of 0.12, where the buckles very thin coatings. The advantage of SIEBIMM is its were clearly observed and then Au–Pd was sputtered simplicity since only the wavelength of coating buckles on top of coating. In the second one, the Au–Pd was needs to be measured, but it relies on elastic reversible sputtered before the stretching at the same strain. Both deformation at relatively high strains which is not test samples were observed in the SEM using an always feasible. Overall, all presented methods still acceleration voltage of 3 kV and the high-efficiency show potential to rank the stiffness properties of thin secondary electron HE-SE2 detector. The micrographs coatings. Quantitative measures are, however, needed, can be seen in Fig. 8, where (a) the precoated and (b) e.g., in models predicting the fragmentation of strained the post-coated samples show the same deformation brittle coatings. 749 J. Coat. Technol. Res., 15 (4) 743–752, 2018 Fig. 8: Stretched samples coated with 20 nm TiO : (a) and (c) have been deposited with a conductive Au–Pd coating before stretching, and (b) and (d) after stretching behavior. The similarities are also found at higher shown in Fig. 11, obtained in 246 k9 of magnification magnification for (c) the precoated and (d) post-coated with an in-lens secondary electron detector. The Au– samples. Not only qualitatively, but also quantitatively Pd layer has a granular structure on the polymer did the dimensions of the buckles not show any surface in the crack, as well as on the TiO barrier. If significant difference if the conductive coating was these granules are separated from each other in applied before or after the buckling formation. It is tension, the conductive layer cannot expect to carry therefore assumed that the Au–Pd coating did not much stress, which has also been pointed out by influence the mechanical behavior in the SIEBIMM Rochat et al. test of the TiO coated film. Based on the negligible influence of the Au–Pd In addition, the conductive coating did not have coating on the buckling formation and tensile crack- significant influence on the fragmentation process of ing of the investigated metal oxide coatings, it is the brittle barrier coating in tensile loading. Figures 9 assumed that the conductive coating does not have and 10 show that the crack accumulation was not any influence in the mechanical analysis of the influenced by the thickness of the conductive coating SIEBIMM test. The observed microstructure with for the MOX and TiO films, respectively. separable granules in the conductive coating can be a The Au–Pd is not a continuous monolithic material, reason for the very limited mechanical effect of the but shows a granular structure at high magnification as material. 750 J. Coat. Technol. Res., 15 (4) 743–752, 2018 1.5 nm 4 nm 8.5 nm ε = 0.05 ε = 0.10 ε = 0.20 Fig. 9: SEM images showing a comparable accumulation of crack with increasing tensile strain for various thicknesses of conductive coatings on for 20 nm MOX films 1.5 nm 4 nm 8.5 nm ε = 0.05 ε = 0.10 ε = 0.20 Fig. 10: SEM images showing a comparable accumulation of crack with increasing tensile strain for various thicknesses of conductive coatings on for 20 nm TiO films 751 J. Coat. Technol. Res., 15 (4) 743–752, 2018 12. Pittenger, B, Erina, N, Su, C, ‘‘Quantitative Mechanical Property Mapping at the Nanoscale with PeakForce QNM.’’ Bruker Application Note, 128, 2012. 13. Chinmulgund, M, Inturi, RB, Barnard, JA, ‘‘Effect of Ar Gas Pressure on Growth, Structure, and Mechanical Properties of Sputtered Ti, Al, TiAl, and Ti Al Films.’’ Thin Solid Films, 270 (1–2) 260–263 (1995) 14. Volynskii, AL, Bazhenov, S, Lebedeva, OV, Bakeev, NF, ‘‘Mechanical Buckling Instability of Thin Coatings Depos- ited on Soft Polymer Substrates.’’ J. Mater. Sci., 35 (3) 547– 554 (2000) 15. Stafford, CM, Harrison, C, Beers, KL, Karim, A, Amis, EJ, VanLandingham, MR, Kim, H-C, Volksen, W, Miller, RD, Simonyi, EE, ‘‘A Buckling-Based Metrology for Measuring the Elastic Moduli of Polymeric Thin Films.’’ Nat. Mater., 3 545–550 (2004) 16. Cranston, ED, Eita, M, Johansson, E, Netrval, J, Salajkova, Fig. 11: High-resolution SEM image of a crack of TiO ˚ 2 M, Arwin, H, Wagberg, L, ‘‘Determination of Young’s 20 nm showing the granular nanoscale structure of the Au– Modulus for Nanofibrillated Cellulose Multilayer Thin Films Pd coating both on the polymer substrate in the crack and Using Buckling Mechanics.’’ Biomacromolecules, 12 (4) 961– on top of TiO layer 969 (2011) 17. Jiang, H, Khang, DY, Song, J, Sun, Y, Huang, Y, Rogers, JA, ‘‘Finite Deformation Mechanics in Buckled Thin Films on References Compliant Supports.’’ Proc. Natl. Acad. Sci., 104 (40) 15607– 15612 (2007) 1. Fayet, P, Neagu, C, Gamstedt, K, ‘‘Mechanics-Driven 18. Cai, Y, Li, H, Karlsson, M, Leifer, K, Engqvist, H, Xia, W, Material Design for Optimized Barrier Films.’’ Proc. The ‘‘Biomineralization on Single Crystalline Rutile: The AIMCAL Web Coating and Handling Conference, Naples, Modulated Growth of Hydroxyapatite by Fibronectin in a USA, October 2015 Simulated Body Fluid.’’ RSC Adv., 6 (42) 35507–35516 2. Henry, BM, Erlat, AG, McGuigan, A, Grovenor, CRM, (2016) Briggs, GAD, Tsukahara, Y, Miyamoto, T, Noguchi, N, 19. Papadakis, R, Li, H, Bergman, J, Lundstedt, A, Jorner, K, Niijima, T, ‘‘Characterization of Transparent Aluminium Ayub, R, Haldar, S, Jahn, BO, Denisova, A, Zietz, B, Lindh, Oxide and Indium Tin Oxide Layers on Polymer Sub- R, Sanyal, B, Grennberg, H, Leifer, K, Ottosson, H, ‘‘Metal- strates.’’ Thin Solid Films, 382 (1) 194–201 (2001) Free Photochemical Silylations and Transfer Hydrogena- 3. Leterrier, Y, ‘‘Durability of Nanosized Oxygen-Barrier tions of Benzenoid Hydrocarbons and Graphene.’’ Nat. Coatings on Polymers.’’ Prog. Mater. Sci., 48 (1) 1–55 (2003) Commun., 7 12962 (2016) 4. Andersons, J, Leterrier, Y, Tornare, G, Dumont, P, Ma ˚ nson, J- 20. MacDonald, WA, Looney, MK, MacKerron, D, Eveson, R, AE, ‘‘Evaluation of Interfacial Stress Transfer Efficiency by Adam, R, Hashimoto, K, Rakos, K, ‘‘Latest Advances in Coating Fragmentation Test.’’Mech.Mater.,39 (9) 834–844 (2007) Substrates for Flexible Electronics.’’ J. Soc. Inf. Disp., 15 (12) 5. Andersons, J, Tarasovs, S, Leterrier, Y, ‘‘Evaluation of Thin 1075–1083 (2007) Film Adhesion to a Compliant Substrate by the Analysis of 21. Saha, R, Nix, W, ‘‘Effects of the Substrate on the Determi- Progressive Buckling in the Fragmentation Test.’’ Thin Solid nation of Thin Film Mechanical Properties by Nanoinden- Films, 517 (6) 2007–2011 (2009) tation.’’ Acta Mater., 50 (1) 23–38 (2002) 6. Jansson, NE, Leterrier, Y, Medico, L, Ma ˚ nson, J-AE, 22. Rar, A, Song, H, Pharr, GM, ‘‘Assessment of New Relation ‘‘Calculation of Adhesive and Cohesive Fracture Toughness for the Elastic Compliance of a Film–Substrate System.’’ of a Thin Brittle Coating on a Polymer Substrate.’’ Thin MRS Proceedings, 695 (2002) Solid Films, 515 (4) 2097–2105 (2006) 23. Hay, J, Crawford, B, ‘‘Measuring Substrate-Independent 7. Chen, S, Liu, L, Wang, T, ‘‘Investigation of the Mechanical Modulus of Thin Films.’’ J. Mater. Res., 26 (6) 727–738 Properties of Thin Films by Nanoindentation, Considering (2011) the Effects of Thickness and Different Coating–Substrate 24. Cherneva, S, Iankov, R, Radic, N, Grbic, B, Stoychev, D, Combinations.’’ Surf. Coat. Technol., 191 (1) 25–32 (2005) ‘‘Nanoindentation Investigation of Mechanical Properties of 8. Oliver, WC, Pharr, GM, ‘‘An Improved Technique for ZrO , ZrO –Y O ,Al O and TiO Thin Films Deposited on 2 2 2 3 2 3 2 Determining Hardness and Elastic Modulus Using Load and Stainless Steel OC 404 Substrate by Spray Pyrolysis.’’ Mater. Displacement Sensing Indentation Experiments.’’ J. Mater. Sci. Eng. B, 183 12–16 (2014) Res., 7 (6) 1564–1583 (1992) 25. Goldstein, J, Newbury, D, Joy, D, Lyman, C, Echlin, P, Lifshin, 9. Doerner, M, Nix, W, ‘‘A Method for Interpreting the Data E, Sawyer, L, Michael, J, Scanning Electron Microscopy and X- from Depth-Sensing Indentation Instruments.’’ J. Mater. ray Microanalysis, 3rd ed. Springer, New York (2003) Res., 1 (4) 601–609 (1986) 26. Quorum Technologies, ‘‘SC7680 High Resolution Sputter 10. Sneddon, IN, ‘‘The Relation Between Load and Penetration Coater Operating Manual.’’ Document Number OM- in the Axisymmetric Boussinesq Problem for a Punch of SC7680, Issue 2 (08/05). Arbitrary Profile.’’ Int. J. Eng. Sci., 3 (1) 47–57 (1965) 27. Rochat, G, Leterrier, Y, Fayet, P, Ma ˚ nson, J-AE, ‘‘Mechan- 11. Woirgard, J, Cabioc’h, T, Riviere, JP, Dargenton, JC, ical Analysis of Ultrathin Oxide Coatings on Polymer ‘‘Nanoindentation Characterization of SiC Coatings Pre- Substrates In situ in a Scanning Electron Microscope.’’ Thin pared by Dynamic Ion Mixing.’’ Surf. Coat. Technol., 100– Solid Films, 437 204–210 (2003) 101 128–131 (1998) http://www.deepdyve.com/assets/images/DeepDyve-Logo-lg.png Journal of Coatings Technology and Research Springer Journals

Comparison of test methods estimating the stiffness of ultrathin coatings

Free
10 pages
Loading next page...
 
/lp/springer_journal/comparison-of-test-methods-estimating-the-stiffness-of-ultrathin-3FopZOnOA9
Publisher
Springer US
Copyright
Copyright © 2018 by The Author(s)
Subject
Materials Science; Tribology, Corrosion and Coatings; Surfaces and Interfaces, Thin Films; Polymer Sciences; Industrial Chemistry/Chemical Engineering; Materials Science, general
ISSN
1547-0091
eISSN
1935-3804
D.O.I.
10.1007/s11998-018-0085-0
Publisher site
See Article on Publisher Site

Abstract

J. Coat. Technol. Res., 15 (4) 743–752, 2018 https://doi.org/10.1007/s11998-018-0085-0 Comparison of test methods estimating the stiffness of ultrathin coatings Marcus Vinı´cius Tavares da Costa , Cristian Neagu, Pierre Fayet, Urban Wiklund, Hu Li, Klaus Leifer, E. Kristofer Gamstedt The Author(s) 2018 Abstract A key engineering parameter of thin coatings measurements dominated by the coating. The strain- is their stiffness. Stiffness characterization of ultrathin induced elastic buckling method is simple in practice, coatings with a nanometer scale thickness is experimen- but showed a large scatter due to variation in local tally challenging. In this work, three feasible methods coating thickness and irregular deformation patterns. have been used to estimate the Young’s modulus of The stiffness characterization using atomic force micro- metal coatings on polymer films. The methods are: (1) scopy gave the most consistent results, due to a sharp tip nanoindentation, (2) strain-induced elastic buckling and with a radius comparable to the thinnest coating (3) peak-force measurements integrated in atomic force thickness. All methods gave a higher Young’s modulus microscopy. The samples were prepared by atomic layer for the TiO coating than for the mixed oxide coating, deposition of TiO (6 and 20 nm thick) and mixed oxides with a variation within one order of magnitude between of TiO and Al O (4 and 20 nm thick). The differences the methods. 2 2 3 in estimated Young’s modulus are interpreted in terms of the underlying assumptions and test conditions. Their Keywords Nano coating, Young’s modulus, Polymer specific advantages and drawbacks are also compared substrate and discussed. In particular, the nanoindentation neces- sitates a sufficiently sharp indenter tip to make localized Introduction This paper was presented at the 13th Coatings Science Inter- national Conference (COSI) on June 26–30, 2017, in Noordwijk, Carton food packages are indeed an important inven- The Netherlands tion. Every day, billions of liters of water, milk, juice and other liquid foods are consumed around the world. M. V. Tavares da Costa, E. K. Gamstedt (&) Carton packaging is mostly used to contain and protect Division of Applied Mechanics, Department of Engineering beverages so that they can be transported to the Sciences, Uppsala University, Box 534, 751 21 Uppsala, consumers with increased lifetime. The barrier func- Sweden e-mail: kristofer.gamstedt@angstrom.uu.se tion is today mainly assured by aluminum foils. This material protects food from the outside (intrusion of M. V. Tavares da Costa oxygen, humidity, etc.), as well as prevents leakage by e-mail: marcus.tavares@angstrom.uu.se keeping the nutrients inside. One of the potential candidates which can replace aluminum foil is a thin C. Neagu Tetra Pak AB, DSO Packaging Materials, Ruben Rausings metal oxide brittle coating deposited on a polymer film Gata, 22186 Lund, Sweden substrate. A number of material properties are typi- cally of interest in coatings development. One of these P. Fayet properties is the stiffness, which affects the strain and Adhemon Sarl, Thin Technology, Avenue Edouard-Dapples stress state in coating structures. It is needed in the 20, 1006 Lausanne, Switzerland estimation of interfacial fracture toughness in coatings subject to mechanical loading. U. Wiklund, H. Li, K. Leifer There are some well-known methods to characterize Division of Applied Materials Science, Department of coatings on polymers such as the fragmentation test. Engineering Sciences, Uppsala University, Box 534, This test can be used to calculate the interfacial shear 751 21 Uppsala, Sweden 743 J. Coat. Technol. Res., 15 (4) 743–752, 2018 strength between coating and polymer substrate, based Tensile testing on the observation of crack density of the coating, as an increasing tensile strain is being applied. Analytical The Young’s modulus of the substrate was obtained by models have been developed to determine adhesive tensile test from the slope of the initial linear regime of properties of the interface from the crack density at the stress–strain diagram. Three tests were carried out saturation. Finite element simulation has also been for each barrier coating. The influence of the ultrathin used for the coating/substrate assemblies to determine coating on the film stiffness was substantially smaller 5,6 the cohesive fracture toughness of the interface. To than the measurement scatter. The samples were cut in use these analytical or numerical approaches, it is first small pieces with dimensions 4 mm 9 20 mm using necessary to determine the elastic properties of the scalpel and ruler. The samples were stretched up to coating. Most coatings are assumed to be isotropic, and 40% of nominal strain, the speed was 0.1 mm/min, and the foremost parameter is then the Young’s modulus, the gauge length was > 10 mm. All data were mea- E, since the Poisson ratio, m, does not show as much sured by the Deben ‘‘Microtest’’ tensile stage, and the variation and has less influence on adhesive behavior. scatter was taken as the standard deviation of the test The determination of coating Young’s modulus is of results. particular interest for thin coatings, where it is known to depend on the coating thickness. The focus of this work is therefore to investigate the Nanoindentation Young’s modulus of thin metal oxide coating using The main principle of this technique is the nanoscale independent methods. Three experimental methods indentation of a rigid indenter into the surface of a have been used to obtain values of the Young’s modulus, deformable material. From the load–deflection curve namely nanoindentation, mechanical buckling measure- of the unloading phase, the elastic properties close to ment and Peak-Force Quantitative Nanomechanical the surface can be obtained. The CSM Ultra Nano Property Mapping technique (PF-QNM). Since the ultrathin coatings (thickness in the order of 10 nm) do Hardness Tester with a diamond cube corner indenter not allow for stiffness characterization by in-plane of 40 nm tip radius was used for all ALD samples. At tensile testing, all three methods are based on local the time of testing, this was the sharpest tip available out-of-plane deformations. Using three complementary for this particular nanoindentation device. The inden- characterization methods for the same material also tation depth was up to 30 nm. The loading rate was gives the opportunity to compare the methods in chosen to be 100 lN/min, as a compromise to be slow practice and identify the different advantages and enough to obtain stable data and fast enough to avoid drawbacks of the methods for the present class of viscous effects. A holding time of 10 s before unloading coating materials. Furthermore, this work also seeks to was chosen to let the material set, and thus avoid determine the mechanical properties of coating alterna- viscous effects in the unloading curve. The sample size tives to aluminum and their thickness dependency. was around 5 mm 9 5 mm, and 10 indentations were made for each sample from which the average Young’s modulus of the coating was calculated. Each specimen was attached using double-sided adhesive tape on top Materials and methods of the horizontal sample holder. The influence of the adhesive tape was neglected since the indentation Samples depth was extremely shallow. The contact stiffness was experimentally measured from the initial portion of The materials in this study were selected based on their 10,11 unloading curve barrier function for carton food packages. All coatings were deposited by atomic layer deposition (ALD) on pffiffiffiffi dP 2 polymer film substrate. The materials of these coatings S ¼ ¼ mP ðh  h Þ ¼ bpffiffiffi E A max max p r dh p are TiO (titanium oxide) and MOX (mixed oxide, max composed of even contributions of TiO and Al O ). 2 2 3 ð1Þ The thickness of these coatings was 20 nm for the thicker coatings of both materials, 4 nm for the thin where E is the reduced modulus, A is the projected MOX coating and 6 nm for the thin TiO . The 2 area, h  h is the total penetration depth, P is max p max thickness was determined by the number of deposition the maximum force, and m and b are constants which cycles since only one atomic layer is produced for each depend on the geometry of the indenter. The value of b cycle. The substrate, biaxially oriented polyethylene was set to 1.0 for a hemispherical indenter, and the terephthalate (BoPET) film 120-lm-thick Teijin Meli- projected area A and constant m were experimentally nex ST504 from Dupont, was the same for all barrier calibrated by the nanoindentation device. The coatings. The thickness of the coating is negligible calibration of A accounts for the assumed spherical compared with the thickness of the substrate, and the tip. Once the reduced modulus E is known, the in-plane stiffness of the coated film is therefore entirely Young’s modulus of the coating, E , is obtained 9–12 dominated by the stiffness of the substrate. through 744 J. Coat. Technol. Res., 15 (4) 743–752, 2018 1  m as a function of the thickness of the coating, h , the E ¼ ð2Þ 1 1m Young’s modulus and Poisson’s ratios of the coating E E r i and the substrate using Euler derivative equation. Later, Stafford et al. used this equation to deter- where E is the Young’s modulus of the indenter, 1145 mine the Young’s modulus of the coating using a GPa. The Poisson ratios of the indenter and coating, designed tensile stage with laser diffraction to measure v and v are 0.070 and 0.30, respectively, for the i c the wavelength of the buckled coating. Solving for E diamond indenter and titanium and aluminum oxide gives SIEBIMM main equation coatings. 3 1  m k E ¼ E : ð4Þ c s 3 2 Strain-induced elastic buckling instability 8p 1  m h for mechanical measurement A state of plane strain is assumed. In our case, the 13 1 The phenomenon of elastic buckling instability of a Poisson ratios m and m are 0.30 and 0.44, respec- c s superficial coating subjected to in-plane compressive tively. Equation (4) was also used, e.g., by Cranston strains can be instrumental in estimating the stiffness of et al. to determine the Young’s modulus of thin films the coating. This kind of test is usually abbreviated ( 70 nm) of nanofibrillated cellulose multilayers SIEBIMM in the literature, denoting strain-induced using a tensile microtest stage together with scanning elastic buckling instability for mechanical measure- electron microscopy (SEM). Such studies show that ment. The principle is to measure the buckling wave- SIEBIMM may be effective for elastic characterization length caused by an applied or residual compressive of thin layers. strain on coating/substrate–film. The analytical model- It should be noted that equation (4) is applicable ing typically requires that (1) the stiffness of the only for small deformations. Jiang et al. proposed a coating is much higher than that of the substrate, correction of equation (3) for the case of finite defor- E  E , (2) the substrate is thicker than the coating, c s mation in buckling due to substrate prestraining. By h  h , and (3) the buckling takes place in the small s c use of their approach, a corrected wavelength k can be strain regime. Figure 1 shows schematically how the calculated buckles appear as well as the important dimensions in the cross section. In our case, a tensile strain was 1=3 k ¼ kð1 þ e Þð1 þ nÞ ð5Þ 0 T applied and, due to the Poisson effect (quantified by m and m ), the elastic buckling forms by an effective where lateral contraction force F. A practical approach has been developed by Volyn- n ¼ 5eðÞ 1 þ e =32: 14 T T skii et al., where they reported the mechanism of buckling formation in the gold layer of 10 nm thick on The compressive transverse strain, e ¼m e ,is T s k PET substrate and derived the relation for the wave- determined from the applied longitudinal tensile strain length at buckling onset e . The buckling strain e and the k k sffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffi wavelength k are obtained experimentally. ð1  m ÞE 3 c In this study, four SIEBIMM tests were carried out k ¼ 2ph  ð3Þ 31  m E in situ in a Deben ‘‘Microtest’’ tensile stage installed in a Hitachi TM-1000 table-top SEM operating at 15 kV with a backscattered electron detector. For high magnification, the specimen was subsequently brought to a Zeiss Merlin field emission gun SEM used at an acceleration voltage of 2 kV and the high-efficiency secondary electron (HE-SE2) detector suited for Crack detailed topographic analysis. The considered coatings , , E v h C C C were made from nonconductive materials. Conse- quently, unwanted charging effects were observed in the SEM during the first attempts. In order to reduce the charging effects, all these samples were coated on top of barrier layers with Au–Pd conductive coating by Quorum SC7640 polaron sputter coater. , , E v h S S S The effects of the Au–Pd coating on the mechanical ε experiments are presented in Appendix A. In these tests, the coating properties were examined in the Fig. 1: Buckling formation due to an effective compressive lateral direction, measuring the buckling size before force (Poisson effect) in a coated film caused by an applied and after the conductive layer, and longitudinal direc- tensile strain 745 J. Coat. Technol. Res., 15 (4) 743–752, 2018 tion, observing the cracking progress regarding tensile where F  F is the difference between the force on adh strain in different Au–Pd thicknesses. Essentially, it the cantilever and the adhesion force, R is the tip AFM could not be demonstrated that the conductive coating radius, E is the reduced modulus, and d  d is the had any influence on the estimation of coating stiffness. sample deformation. Just as for nanoindentation, the Figure 2 illustrates how the SIEBMM test was Young’s modulus of the coating is calculated from the performed. A removable rig was designed and manu- reduced modulus by equation (2). factured to fit the Deben tensile stage in the vacuum chamber of the table-top SEM, as shown in Fig. 2a. When the buckling was clearly observed in situ during Results and discussion tensile testing in the SEM, the loading was stopped and the applied strain in the sample was fastened by Tensile testing tightening two fixation screws in the removable rig, as shown in Fig. 2b. The compact rig was then detached The in-plane tensile test confirmed that there was no and taken to the high-resolution SEM. The removable influence of the coating stiffness on the film stiffness, rig can thus swiftly be taken out from the microtensile which was effectively dominated by the substrate. The stage and moved to a high-resolution SEM while obtained value of the Young’s modulus of the substrate maintaining the tensile strain at buckling onset e ,as was comparable for all samples, and no tangible shown in Fig. 2c. difference was found compared with the same sub- strate material without any barrier coating. That also AFM peak-force measurements The Young’s modulus of the coatings was also DTM fit for elastic modulus estimated using a Multimode 8 Atomic Force Micro- scope, Bruker, in the Peak-Force Quantitative 12,18 Nanomechanical Mapping (PF-QNM) mode. A silicon probe was utilized with the tip radius of 8 nm. PF-QNM mode is a technique to quantitatively mea- 0 Deformation sure the surface mechanical properties of the sample by recording the force–separation feedback loop in the AFM as illustrated in Fig. 3. The Young’s modulus, the adhesion force, energy dissipation and the sample deformation can be extracted from this loop, where the Young’s modulus can be estimated by fitting the adh retraction part of the force–separation loop to the Derjaguin-Muller-Toporov (DMT) model : Tip-sample separation qffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffi AFM Fig. 3: Force–separation curve generated in AFM PF-QNM F  F ¼ E Rðd  d Þ ð6Þ adh 0 mode Fig. 2: Steps in the elastic buckling instability analysis: (a) The removable rig attached in tensile stage. (b) The fixation rig which is dismounted at the point of buckling formation. (c) The rig is brought to a high-resolution SEM for measurement of the buckle dimensions Force J. Coat. Technol. Res., 15 (4) 743–752, 2018 shows that the ALD manufacturing process did not nanoindentation tests, from which the coating Young’s change the elastic property of the polymer substrate. moduli were determined using equations (1) and (2)as Figure 4 shows the results from tensile test. an average from 10 measurements. These values are summarized in Table 1 together with those from the other independent test methods. For some indenta- Nanoindentation tions, in particular for the thinner coatings, viscous effects due to the polymer substrate were apparent. The contact stiffness was determined by least squares This resulted occasionally in nearly vertical unloading fitting from unloading curves starting at 98% down to curves in the force–indentation diagrams. 40% of the maximum measured force. With a holding time of 10 s before unloading, obvious viscous effects could be avoided. Figure 5 shows the results from the SIEBIMM The buckling formation was observed in situ by SEM. The strain for onset of buckling could then be determined. It was noticed that the strains for initial 3.5 buckling showed a significant variation, since the elastic instability was controlled by the local thickness and elastic properties. A high tensile strain at buckling 2.5 onset was observed for the thinner TiO and MOX coatings of 0.31 and 0.42, respectively. For 20-nm coatings, the buckling was perceptible at tensile strain of 0.1. Occasionally, cracks were found along the ridges 1.5 of the larger buckles (see Fig. 6). The wavelength of the cracked buckles was, however, found to be the same as with the uncracked ones. If a band of adjacent 0.5 buckles could not be found, the local wavelength of localized buckles was measured. TiO 6nm TiO 20nm MOX 20 nm MOX 4 nm Figure 6 shows the buckles on top of a TiO coating 2 2 of 6 nm strained at 0.31 in the longitudinal direction. Fig. 4: Young’s modulus of the coated substrate This image was taken from the center of the sample, determined from tensile testing where the load was relatively uniform. This region had 25 25 25 25 20 20 20 20 15 15 15 10 10 10 5 5 5 5 0 0 0 0 0 10 20 30 40 0 10 20 30 40 0 10 20 30 40 0 10 20 30 40 Indentation depth (nm) Indentation depth (nm) Indentation depth (nm) Indentation depth (nm) (d) (a) (b) (c) Fig. 5: Force versus indentation depth in nanoindentation of the coatings. (a) TiO 6 nm, (b) TiO 20 nm, (c) MOX 4 nm, 2 2 (d) MOX 20 nm Table 1: Summary of estimated Young’s moduli for the characterization methods Barrier coating h (nm) Tensile test Nanoindentation AFM SIEBIMM Bulk E (GPa) E (GPa) E (GPa) E (GPa) E (GPa) s c c c c TiO 6 3.5 ± 0.4 44 ± 19 45 ± 2 420 ± 140 115 TiO 20 3.2 ± 0.4 35 ± 17 76 ±264 ± 16 115 MOX 4 3.7 ± 0.2 31 ± 21 29 ± 2 240 ± 110 160–180 MOX 20 3.0 ± 0.3 23 ± 09 43 ±352 ± 13 160–180 Young’s modulus of coated polymer (GPa) Force (μN) J. Coat. Technol. Res., 15 (4) 743–752, 2018 a high and relatively uniform concentration of buckles. Comparison of results The wavelength pattern was different that those 15 14 reported by Stafford et al., Volynskii et al. and The results from the various characterization tech- Cranston et al. The buckling process is influenced by niques are summarized in Table 1, where the scatter is the stiffness ratio E /E , according to equation (3). The the standard deviation of these values. Although the c s 3 5 reported values ranged in the order 10 –10 , whereas in values of Young’s moduli of the coatings are all in the this study, the ratio is in the range of 7–25. same order of magnitude, there are considerable All coatings had similar buckling shapes, but showed differences. The TiO coating of 6 nm and MOX different dimensions. The size of all buckles was coating of 4 nm had a relatively good agreement measured in SEM, and then averaged to calculate the between the nanoindentation and AFM values, but coating Young’s modulus E using equation (4).The very high values for the SIEBIMM values. The 20-nm values are reported in Table 1. coatings showed a large variation between the different characterization methods. For comparison, bulk mod- ulus from literature is presented in the last column. AFM peak-force measurement The presented methods showed higher stiffness for the TiO coatings than for the MOX coatings, which was The indentation depth in AFM was controlled within a not the case for the literature values. The buckling few nm by a sharp AFM tip (radius of 8 nm), resulting method, SIEBIMM, showed higher Young’s moduli for both materials, compared with AFM and nanoinden- in a limited influence from the substrate, which is tation. The sensitivity and accuracy of this method is suitable for ultrathin coating measurement. Figure 7 adversely affected by the relatively low coating– shows the effective Young’s modulus maps of an area substrate stiffness ratio and inelastic deformations at approximately 1 lm for each barrier coating. Each high strains. As for the thickness dependency, only the pixel in the AFM images corresponds to a loop AFM test method showed a higher stiffness for the described previously in Fig. 3. Each image is made up of more than 250,000 measurements. The measured thicker coatings. force deformations for all pixels were used to deter- mine the average Young’s modulus of the coating by means of equations (2) and (5). Concluding remarks The average values of each coating are reported in Table 1. Four test methods were experimentally assessed for the determination of the Young’s moduli of ultrathin coatings deposited on polymer films. Tensile testing is generally unsuitable to characterize the coating stiff- ness for thin coated polymer films, since the measured stiffness is dominated by the relatively thick polymer film. Nanoindentation gave crude measures of the coating Young’s modulus, since the radius of the indenter tip was higher than the coating thickness. Viscous effects were observed, in particular for the thinly coated samples. The viscoelastic polymer substrate then affects the estimated coating stiffness. Shallow inden- tations with sharp tips on nonviscous materials would lead to improvements in this method. We tried to use Fig. 6: SEM image of a TiO coating of 6 nm stretched at 0.31 strain approaches that account the substrate effect: Song/ Fig. 7: AFM mapping of the effective Young’s modulus of the coatings. (a) TiO 6 nm, (b) TiO 20 nm, (c) MOX 4 nm, 2 2 (d) MOX 20 nm 748 J. Coat. Technol. Res., 15 (4) 743–752, 2018 22 23 21 Pharr, Hay/Crawford and King models. These Acknowledgments Financial support and supply of models were developed in a range of hundred nanome- the test samples for this study by Tetra Pak Packaging ters of coating thickness and depend of normalized Solutions AB, Development and Service Operations contact area over coating thickness. It turned out that (DSO), are much appreciated. MVTdC is grateful to the values of the coating modulus did not improve. the European Commission within the framework of the The buckling method, SIEBIMM, has the advantage Erasmus Mundus Programme, Action 2-STRAND 1, of being straightforward, not requiring an advanced Lot 9, Brazil for PhD studies. The authors would like apparatus to measure local loads and displacements. In to thank Mr. Luimar Correa for the technical our case, the buckles appeared at relatively high assistance in the SEM experiments and Dr. Jessica strains, where the polymer substrate was subject to Bolinsson for valuable scientific discussions. some inelastic deformation. The moduli estimated using this method were much larger than those from Open Access This article is distributed under the the other methods. With a fully elastic and relatively terms of the Creative Commons Attribution 4.0 compliant substrate, such as an elastomer, the clarity of International License (http://creativecommons.org/lice the buckles and the calculated stiffness values would nses/by/4.0/), which permits unrestricted use, distribu- most likely be improved. tion, and reproduction in any medium, provided you The AFM method showed the most consistent give appropriate credit to the original author(s) and results compared with the other methods. In this case, the source, provide a link to the Creative Commons the tip radius was much smaller than that used in license, and indicate if changes were made. nanoindentation, which provides better local deforma- tion. For nanoindentation, it is likely pushing the coating down rather than being measured from a point load at a pyramidal tip (Oliver Pharr method). Addi- tionally, the indentation depth in AFM is small enough Appendix A: Effects of the conductive layer to have only a limited influence of the viscoelastic in the barrier coating properties polymer substrate. A considerable amount of stiffness data is generated, since the AFM equipment automat- Observation of buckling caused by elastic instability in ically maps a given area. This leads to average values of thin coatings typically requires a conductive coating in higher statistical confidence compared with few point- SEM analysis since the features of interest are in the wise measurements in nanoindentation. sub-micron scale. In order to reduce charging effects, Although the variation was large in absolute num- all the samples were coated on top of the barrier layer bers, all methods showed the same trend with a higher with a Au–Pd conductive coating by Quorum SC7640 Young’s modulus for the TiO coating than for the polaron sputter coater. Au–Pd is one of thinnest MOX coating (TiO and Al O mixture). This is 2 2 3 24 continuous coating materials for SEM purposes. For consistent with Cherneva et al., who measured a the deposition process the following parameters were slightly higher modulus for TiO than for Al O . 2 2 3 used: an accelerate voltage of 2 kV, a plasma discharge The influence of the coating thickness on Young’s current of 20 mA, a pressure between 5 and 8 Pa and a modulus of the coating was not conclusive. The AFM deposition time of 1 min. Using these parameters method showed a higher stiffness for the thicker provides a layer of 8.5 nm thick, which was the coating, whereas the other methods showed the reverse minimum thickness allowing observation of surface order, which was also found by Chen et al. buckling and fracture without excessive charging To estimate Young’s modulus of coatings with a effects. The issue to address here is whether or not thickness of less than 0.1 lm is indeed challenging. The this additional layer would have any influence on the present paper presents a number of candidate meth- buckling formation and size. ods, each with their specific advantages and drawbacks. To investigate the influence of the Au–Pd coating, Both AFM and nanoindentation are based on pushing two different test samples were produced with the TiO in and retracting a rigid tip. AFM allows for more 20-nm coating. In the first one, the sample was shallow indentation, which makes it more suitable for stretched up to a strain level of 0.12, where the buckles very thin coatings. The advantage of SIEBIMM is its were clearly observed and then Au–Pd was sputtered simplicity since only the wavelength of coating buckles on top of coating. In the second one, the Au–Pd was needs to be measured, but it relies on elastic reversible sputtered before the stretching at the same strain. Both deformation at relatively high strains which is not test samples were observed in the SEM using an always feasible. Overall, all presented methods still acceleration voltage of 3 kV and the high-efficiency show potential to rank the stiffness properties of thin secondary electron HE-SE2 detector. The micrographs coatings. Quantitative measures are, however, needed, can be seen in Fig. 8, where (a) the precoated and (b) e.g., in models predicting the fragmentation of strained the post-coated samples show the same deformation brittle coatings. 749 J. Coat. Technol. Res., 15 (4) 743–752, 2018 Fig. 8: Stretched samples coated with 20 nm TiO : (a) and (c) have been deposited with a conductive Au–Pd coating before stretching, and (b) and (d) after stretching behavior. The similarities are also found at higher shown in Fig. 11, obtained in 246 k9 of magnification magnification for (c) the precoated and (d) post-coated with an in-lens secondary electron detector. The Au– samples. Not only qualitatively, but also quantitatively Pd layer has a granular structure on the polymer did the dimensions of the buckles not show any surface in the crack, as well as on the TiO barrier. If significant difference if the conductive coating was these granules are separated from each other in applied before or after the buckling formation. It is tension, the conductive layer cannot expect to carry therefore assumed that the Au–Pd coating did not much stress, which has also been pointed out by influence the mechanical behavior in the SIEBIMM Rochat et al. test of the TiO coated film. Based on the negligible influence of the Au–Pd In addition, the conductive coating did not have coating on the buckling formation and tensile crack- significant influence on the fragmentation process of ing of the investigated metal oxide coatings, it is the brittle barrier coating in tensile loading. Figures 9 assumed that the conductive coating does not have and 10 show that the crack accumulation was not any influence in the mechanical analysis of the influenced by the thickness of the conductive coating SIEBIMM test. The observed microstructure with for the MOX and TiO films, respectively. separable granules in the conductive coating can be a The Au–Pd is not a continuous monolithic material, reason for the very limited mechanical effect of the but shows a granular structure at high magnification as material. 750 J. Coat. Technol. Res., 15 (4) 743–752, 2018 1.5 nm 4 nm 8.5 nm ε = 0.05 ε = 0.10 ε = 0.20 Fig. 9: SEM images showing a comparable accumulation of crack with increasing tensile strain for various thicknesses of conductive coatings on for 20 nm MOX films 1.5 nm 4 nm 8.5 nm ε = 0.05 ε = 0.10 ε = 0.20 Fig. 10: SEM images showing a comparable accumulation of crack with increasing tensile strain for various thicknesses of conductive coatings on for 20 nm TiO films 751 J. Coat. Technol. Res., 15 (4) 743–752, 2018 12. Pittenger, B, Erina, N, Su, C, ‘‘Quantitative Mechanical Property Mapping at the Nanoscale with PeakForce QNM.’’ Bruker Application Note, 128, 2012. 13. Chinmulgund, M, Inturi, RB, Barnard, JA, ‘‘Effect of Ar Gas Pressure on Growth, Structure, and Mechanical Properties of Sputtered Ti, Al, TiAl, and Ti Al Films.’’ Thin Solid Films, 270 (1–2) 260–263 (1995) 14. Volynskii, AL, Bazhenov, S, Lebedeva, OV, Bakeev, NF, ‘‘Mechanical Buckling Instability of Thin Coatings Depos- ited on Soft Polymer Substrates.’’ J. Mater. Sci., 35 (3) 547– 554 (2000) 15. Stafford, CM, Harrison, C, Beers, KL, Karim, A, Amis, EJ, VanLandingham, MR, Kim, H-C, Volksen, W, Miller, RD, Simonyi, EE, ‘‘A Buckling-Based Metrology for Measuring the Elastic Moduli of Polymeric Thin Films.’’ Nat. Mater., 3 545–550 (2004) 16. Cranston, ED, Eita, M, Johansson, E, Netrval, J, Salajkova, Fig. 11: High-resolution SEM image of a crack of TiO ˚ 2 M, Arwin, H, Wagberg, L, ‘‘Determination of Young’s 20 nm showing the granular nanoscale structure of the Au– Modulus for Nanofibrillated Cellulose Multilayer Thin Films Pd coating both on the polymer substrate in the crack and Using Buckling Mechanics.’’ Biomacromolecules, 12 (4) 961– on top of TiO layer 969 (2011) 17. Jiang, H, Khang, DY, Song, J, Sun, Y, Huang, Y, Rogers, JA, ‘‘Finite Deformation Mechanics in Buckled Thin Films on References Compliant Supports.’’ Proc. Natl. Acad. Sci., 104 (40) 15607– 15612 (2007) 1. Fayet, P, Neagu, C, Gamstedt, K, ‘‘Mechanics-Driven 18. Cai, Y, Li, H, Karlsson, M, Leifer, K, Engqvist, H, Xia, W, Material Design for Optimized Barrier Films.’’ Proc. The ‘‘Biomineralization on Single Crystalline Rutile: The AIMCAL Web Coating and Handling Conference, Naples, Modulated Growth of Hydroxyapatite by Fibronectin in a USA, October 2015 Simulated Body Fluid.’’ RSC Adv., 6 (42) 35507–35516 2. Henry, BM, Erlat, AG, McGuigan, A, Grovenor, CRM, (2016) Briggs, GAD, Tsukahara, Y, Miyamoto, T, Noguchi, N, 19. Papadakis, R, Li, H, Bergman, J, Lundstedt, A, Jorner, K, Niijima, T, ‘‘Characterization of Transparent Aluminium Ayub, R, Haldar, S, Jahn, BO, Denisova, A, Zietz, B, Lindh, Oxide and Indium Tin Oxide Layers on Polymer Sub- R, Sanyal, B, Grennberg, H, Leifer, K, Ottosson, H, ‘‘Metal- strates.’’ Thin Solid Films, 382 (1) 194–201 (2001) Free Photochemical Silylations and Transfer Hydrogena- 3. Leterrier, Y, ‘‘Durability of Nanosized Oxygen-Barrier tions of Benzenoid Hydrocarbons and Graphene.’’ Nat. Coatings on Polymers.’’ Prog. Mater. Sci., 48 (1) 1–55 (2003) Commun., 7 12962 (2016) 4. Andersons, J, Leterrier, Y, Tornare, G, Dumont, P, Ma ˚ nson, J- 20. MacDonald, WA, Looney, MK, MacKerron, D, Eveson, R, AE, ‘‘Evaluation of Interfacial Stress Transfer Efficiency by Adam, R, Hashimoto, K, Rakos, K, ‘‘Latest Advances in Coating Fragmentation Test.’’Mech.Mater.,39 (9) 834–844 (2007) Substrates for Flexible Electronics.’’ J. Soc. Inf. Disp., 15 (12) 5. Andersons, J, Tarasovs, S, Leterrier, Y, ‘‘Evaluation of Thin 1075–1083 (2007) Film Adhesion to a Compliant Substrate by the Analysis of 21. Saha, R, Nix, W, ‘‘Effects of the Substrate on the Determi- Progressive Buckling in the Fragmentation Test.’’ Thin Solid nation of Thin Film Mechanical Properties by Nanoinden- Films, 517 (6) 2007–2011 (2009) tation.’’ Acta Mater., 50 (1) 23–38 (2002) 6. Jansson, NE, Leterrier, Y, Medico, L, Ma ˚ nson, J-AE, 22. Rar, A, Song, H, Pharr, GM, ‘‘Assessment of New Relation ‘‘Calculation of Adhesive and Cohesive Fracture Toughness for the Elastic Compliance of a Film–Substrate System.’’ of a Thin Brittle Coating on a Polymer Substrate.’’ Thin MRS Proceedings, 695 (2002) Solid Films, 515 (4) 2097–2105 (2006) 23. Hay, J, Crawford, B, ‘‘Measuring Substrate-Independent 7. Chen, S, Liu, L, Wang, T, ‘‘Investigation of the Mechanical Modulus of Thin Films.’’ J. Mater. Res., 26 (6) 727–738 Properties of Thin Films by Nanoindentation, Considering (2011) the Effects of Thickness and Different Coating–Substrate 24. Cherneva, S, Iankov, R, Radic, N, Grbic, B, Stoychev, D, Combinations.’’ Surf. Coat. Technol., 191 (1) 25–32 (2005) ‘‘Nanoindentation Investigation of Mechanical Properties of 8. Oliver, WC, Pharr, GM, ‘‘An Improved Technique for ZrO , ZrO –Y O ,Al O and TiO Thin Films Deposited on 2 2 2 3 2 3 2 Determining Hardness and Elastic Modulus Using Load and Stainless Steel OC 404 Substrate by Spray Pyrolysis.’’ Mater. Displacement Sensing Indentation Experiments.’’ J. Mater. Sci. Eng. B, 183 12–16 (2014) Res., 7 (6) 1564–1583 (1992) 25. Goldstein, J, Newbury, D, Joy, D, Lyman, C, Echlin, P, Lifshin, 9. Doerner, M, Nix, W, ‘‘A Method for Interpreting the Data E, Sawyer, L, Michael, J, Scanning Electron Microscopy and X- from Depth-Sensing Indentation Instruments.’’ J. Mater. ray Microanalysis, 3rd ed. Springer, New York (2003) Res., 1 (4) 601–609 (1986) 26. Quorum Technologies, ‘‘SC7680 High Resolution Sputter 10. Sneddon, IN, ‘‘The Relation Between Load and Penetration Coater Operating Manual.’’ Document Number OM- in the Axisymmetric Boussinesq Problem for a Punch of SC7680, Issue 2 (08/05). Arbitrary Profile.’’ Int. J. Eng. Sci., 3 (1) 47–57 (1965) 27. Rochat, G, Leterrier, Y, Fayet, P, Ma ˚ nson, J-AE, ‘‘Mechan- 11. Woirgard, J, Cabioc’h, T, Riviere, JP, Dargenton, JC, ical Analysis of Ultrathin Oxide Coatings on Polymer ‘‘Nanoindentation Characterization of SiC Coatings Pre- Substrates In situ in a Scanning Electron Microscope.’’ Thin pared by Dynamic Ion Mixing.’’ Surf. Coat. Technol., 100– Solid Films, 437 204–210 (2003) 101 128–131 (1998)

Journal

Journal of Coatings Technology and ResearchSpringer Journals

Published: May 31, 2018

References

You’re reading a free preview. Subscribe to read the entire article.


DeepDyve is your
personal research library

It’s your single place to instantly
discover and read the research
that matters to you.

Enjoy affordable access to
over 18 million articles from more than
15,000 peer-reviewed journals.

All for just $49/month

Explore the DeepDyve Library

Search

Query the DeepDyve database, plus search all of PubMed and Google Scholar seamlessly

Organize

Save any article or search result from DeepDyve, PubMed, and Google Scholar... all in one place.

Access

Get unlimited, online access to over 18 million full-text articles from more than 15,000 scientific journals.

Your journals are on DeepDyve

Read from thousands of the leading scholarly journals from SpringerNature, Elsevier, Wiley-Blackwell, Oxford University Press and more.

All the latest content is available, no embargo periods.

See the journals in your area

DeepDyve

Freelancer

DeepDyve

Pro

Price

FREE

$49/month
$360/year

Save searches from
Google Scholar,
PubMed

Create lists to
organize your research

Export lists, citations

Read DeepDyve articles

Abstract access only

Unlimited access to over
18 million full-text articles

Print

20 pages / month

PDF Discount

20% off