Programming Feature Size in the Thermal ... - ACS Publications

Subscriber access provided by EAST TENNESSEE STATE UNIV

Article

Programming Feature size in Thermal Wrinkling of Metal Polymer Bilayer by Modulating Substrate Visco Elasticity Anuja Das, Aditya Banerji, and Rabibrata Mukherjee ACS Appl. Mater. Interfaces, Just Accepted Manuscript • Publication Date (Web): 15 Jun 2017 Downloaded from http://pubs.acs.org on June 16, 2017

Just Accepted “Just Accepted” manuscripts have been peer-reviewed and accepted for publication. They are posted online prior to technical editing, formatting for publication and author proofing. The American Chemical Society provides “Just Accepted” as a free service to the research community to expedite the dissemination of scientific material as soon as possible after acceptance. “Just Accepted” manuscripts appear in full in PDF format accompanied by an HTML abstract. “Just Accepted” manuscripts have been fully peer reviewed, but should not be considered the official version of record. They are accessible to all readers and citable by the Digital Object Identifier (DOI®). “Just Accepted” is an optional service offered to authors. Therefore, the “Just Accepted” Web site may not include all articles that will be published in the journal. After a manuscript is technically edited and formatted, it will be removed from the “Just Accepted” Web site and published as an ASAP article. Note that technical editing may introduce minor changes to the manuscript text and/or graphics which could affect content, and all legal disclaimers and ethical guidelines that apply to the journal pertain. ACS cannot be held responsible for errors or consequences arising from the use of information contained in these “Just Accepted” manuscripts.

ACS Applied Materials & Interfaces is published by the American Chemical Society. 1155 Sixteenth Street N.W., Washington, DC 20036 Published by American Chemical Society. Copyright © American Chemical Society. However, no copyright claim is made to original U.S. Government works, or works produced by employees of any Commonwealth realm Crown government in the course of their duties.

Page 1 of 22

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

ACS Applied Materials & Interfaces

Programming Feature size in Thermal Wrinkling of Metal Polymer Bilayer by Modulating Substrate Visco Elasticity Anuja Das,# Aditya Banerji,# and Rabibrata Mukherjee* Instability and Soft Patterning Laboratory, Department of Chemical Engineering, Indian Institute of Technology Kharagpur, Pin-721302, India. *Author for correspondence: e-mail: [email protected] , Tel:+91-3222 283912 # Authors with equal credit Key words: Wrinkling, Interfacial Stress, Viscoelasticity, Patterns, Elastomer Abstract: We report a novel strategy to create stress induced self-organized wrinkles in a metal polymer bilayer with programmable periodicity (λS) varying over a wide range, from ≈ 20 µm down to ≈ 800 nm by modulating the viscoelasticity of the bottom polymer layer. Substrates with different visco-elasticity are prepared by pre-curing thin films of a thermo-curable poly-dimethylsiloxane (PDMS) elastomer (Sylgard 184) for different durations (tP) prior to deposition of the top aluminum layer by thermal evaporation. Pre-curing of the Sylgard 184 film for different durations leads to films with different degree of viscoelasticity due to variation in the extent of cross linking of the polymer matrix. The λS as well as the amplitude (aS) of the wrinkles progressively decrease with increase in the extent of elasticity of the film, manifested as increase in the storage modulus (G/). Based on the variation in the rate of decay of λS with G/, we identify three clearly distinguishable regimes over predominantly viscous, viscoelastic and elastic bottom layers. While λS and aS drop with increase in G/ for both the first and the third regimes, it remains nearly independent of G/ for the intermediate regime. This is attributed to the difference in the mechanisms of wrinkle formation in the different regimes. We finally show that simultaneous modulation of λS and aS can be used to engineer surfaces with different wettability as well as antireflection properties.

1 ACS Paragon Plus Environment


1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 2 of 22

Introduction: Wrinkle formation is ubiquitous is nature, as it can be seen on human skin with aging, in fruit peels, dried leaves and so on. While it is undesirable in many cases, artificially fabricated wrinkles with micron and meso scale features find wide application in flexible electronics,1 smart adhesives,2 determination of mechanical properties of thin film,3 fabrication of OLED,4 stretchable electronic devices,5 super hydrophobic surfaces,6 anti-fouling coatings,7 unique identity tags,8 etc. Recently, wrinkled surfaces have found wide application as photonic structures in photovoltaics due to enhanced light harvesting capability.9 Typically, such wrinkles appear on a bilayer comprising of a hard skin over a soft compliant bottom layer, and result due to compressive stress generated at the interface between the two layers.10–34 Formation of such self-organized wrinkles was first observed by Bowden et al. in a metal capped cross linked polydimethylsiloxane (PDMS) elastomeric bottom layer.10 While wrinkle formation can be engendered by various means such as solvent vapor induced swelling,12 photo triggered structural transformation,13 chemical surface oxidation,14 plasma oxidation of an elastomeric surface,15 etc., wrinkles based on mismatch in thermal expansion coefficient between a metal and a polymer thin film has received the maximum attention.10, 15–36 The wrinkles appear when the interfacial stress exceeds a critical compressive stress (σ0) and gets relieved. If the metal film is coated on an amorphous or a semi-crystalline polymer film, the wrinkles appear only when the bilayer is thermally annealed beyond the glass transition (Tg) or the melting (Tm) temperature of the polymer, respectively.16-18 In contrast, the patterns appear right after metal deposition on an elastomeric bottom layer.10,29,30,36 The elastomer layer gets heated due to thermal radiation from the heating source and expands, resulting in an in-plane compressive stress at the interface during subsequent cooling.


Page 3 of 22

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


Apart from the formation of anisotropic wrinkles on a flat elastomeric film, various interesting aspects of the wrinkling instability on an elastomer thin film such as morphology of the wrinkles on a topographically pre-patterned substrate, transition from disordered to ordered structures close to an edge, morphology over a step etc. have already been investigated.10 Huck reported the formation of complex ordered structures when the metal film is deposited on a chemically prepatterned elastomer layer having periodically varying domains of stiffness and coefficient of thermal expansion.19 Chua et al. reported that the periodicity of the wrinkles can be controlled by varying the plasma power as well as the duration of exposure.20 Hyun and Jeong showed that λS depends on the thickness (hS) as well as the elastic modulus of the substrate.27 Yoo and coworkers have reported that λS can be progressively reduced if the film is annealed by a combination of ultraviolet/ozone irradiation and oxygen plasma exposure or subject to repeated short-period oxygen plasma exposure.28 Recently, Yu et al. have shown how the morphology of the patterns gradually changes in case there is a continuous thickness gradient of the metal film. While broad cracks appear in the thicker film region, the cracks attenuate along the direction of film thickness reduction with the co-existence of different wrinkle patterns such as stripes, herringbones, labyrinths etc.30 A special class of problem is wrinkle formation on a viscoelastic substrate31-35 which has received far less attention than wrinkling on either glassy or elastomeric bottom layer. Huang et. al. have extensively studied and obtained analytical solution for the system by subjecting it to linear perturbation by modelling the elastic top layer with non-linear von Karmann Plate Theory and the bottom thin layer approximated to exhibit linear viscoelastic response.31-34 They have further highlighted that on a viscoelastic substrate the wrinkle amplitude (aS) and periodicity (λS) depends both on the minimization of the total elastic energy as well as viscosity dependent



1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 4 of 22

kinetics of the substrate.34 In a very recent paper Yu et al. have deposited metal film on a PDMS bottom layer with gradient visco-elasticity.36 Yu et. al. observed that the pattern morphology gradually changes from the liquid like end to the elastic end of the sample and qualitatively identified two morphology regimes comprising of folds over the areas where the PDMS layer is liquid like and wrinkles over the areas where the PDMS layer is cross linked and elastic like. However, the precise dependence of λS and aS on the rheology of the viscoelastic bottom layer is yet totally unexplored. Scientifically, the rheology of the bottom layer is an important parameter as it influences dissipation, which in turn is likely to alter the magnitude of the accumulated interfacial stresses. In this article, we show how the viscoelasticity of the polymer under-layer can be used to tune the morphology of the wrinkles, by thermally evaporating Aluminium on Sylgard 184 films precured for different durations of time.33,34 We vary the degree of crosslinking and mechanical properties of PDMS by varying the amount of crosslinker and curing time. While we observe that λS and aS both continues to drop with increase in G/ of the bottom layer till the film gets fully cross-linked, we identify three clearly distinct regimes based on the nature of variation of λS with G/. We show that the buckling patterns can be tuned over a much broader range compared to the conventional method of varying the thickness of the top skin layer, which suffers from delamination and cracking of the film as the thickness of the skin layer increases. The work is different from those theoretically studied by Huang et. al. and experimentally by Stratford et. al., where the visco-elasticity of the bottom layer changes with time due to annealing and consequently the morphology of the wrinkles also evolve. In contrast, in our case the metal is deposited on a pre-cured viscoelastic elastomer film, where the patterns freeze instantaneously after the metal deposition resulting from rapid cross linking of the bottom elastomer layer due to


Page 5 of 22

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


enhancement of the surface temperature of the film during metal deposition. Our finding provides a control on creating patterns with feature size on demand, which can be effectively used for fabricating surface with tuneable wettability and anti-reflection properties. Experimental Details Thin films of Sylgard-184, a two part polydimethylsiloxane (PDMS) based thermo curable elastomer (Dow Corning, USA) was used to create the polymer bottom layers with different levels of viscoelasticity. The cross-linker concentration (CL), which is the ratio of Part B to Part A (wt. /wt.) was varied between 5% and 15% to obtain films with different levels of stiffness. The films were spin coated (Apex Instruments, India) onto thoroughly cleaned glass slides from dilute solution of varying polymer and cross-linker concentrations in n-heptane (HPLC grade, Merck, India). The bottom layer thickness (hS) was varied between 2 µm and 20 µm by varying the dilution of the pre-polymer mix in solvent, while the RPM and the spinning duration was held constant at 2500 and 60 seconds respectively. Bottom layers with different levels of viscoelasticity was prepared by pre-curing the as cast films in a hot air oven at 80°C for different durations (tP).37 The top Aluminum (Al) layer was deposited on the partially cured Sylgard 184 under layer by thermal evaporation (Advanced Process Technology, India) using Aluminum beads (Sigma-Aldrich India) as source under a vacuum of 5×10-6 mbar. The thickness of the Al film was kept uniform at hM ≈ 40 nm in most experiments. The morphology of the wrinkles was characterized by an Optical Microscope (Leica DM 2500) in the Reflection Mode and an Atomic Force Microscope (Agilent Technologies, model 5100) in intermittent contact mode, using a Silicon Cantilever (PPP-NCL, Nanosensors Inc. USA). The periodicity of the wrinkles were determined by 2-D FFT based on power spectral analysis of the AFM as well as the optical microscope images using GPL software, Gwyddion. The same



1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 6 of 22

software was also used to determine the fractal dimension of the wrinkles using box counting algorithm. The detailed methodology of FFT analysis and fraction dimension calculation is described in sections S1 and S2 of the online supporting information respectively. Optical characterization of the surfaces for reflectance was performed using an UV-Vis-NIR spectrophotometer (Agilent Varian Cary 5000 UV-Vis-NIR) and the wettability of the surfaces were measured by a Contact Angle Goniometer (Ramé-hart, USA, model −290 G1). Result and Discussions:

Figure 1: Typical rheological behavior indicating the progressive change of G/ and G// with tP of the Sylgard 184 bottom layer. The three regimes are marked as R-1, R-2 and R-3 respectively. The demarcation line between individual regimes is qualitative and acts as a guide only. The time axis has been truncated at tP ≈ 15 minutes as the subsequent variation of both G/ and G// with tP is marginal.

The variation of storage (G/) and loss (G//) modulus of films with progressive pre-curing was quantified using an oscillatory Rheometer with a cone and plate arrangement (cone angle 2°, Physica MCR 301, Anton Paar, Austria) operating at an angular frequency of 10 rad/s. A typical representative plot for CL = 10.0% is shown in figure 1. It can be seen that as cast Sylgard 184 film is in a purely viscous state, evident from a higher value of loss modulus (G//) as compared to storage modulus (G/). With progressive pre-curing, gelation or cross-linking reaction proceeds. This gradually enhances G/, which eventually exceeds G//. For a film with CL = 10.0%, Rheometer data shows that crosslinking reaction gets complete after tP ≈ 35 minutes. The cross 6 ACS Paragon Plus Environment

Page 7 of 22

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


linking reaction is nearly completed at the time, as beyond this stage G/ and G// both remain unaltered upon further pre-curing, as the films become predominantly elastic, which is evident from a two order higher final value of G/ (G/F ≈ 106 Pa) as compared to G// (G//F ≈ 104 Pa). However, it must be highlighted that the time required for complete crosslinking in an oven is somewhat longer, as there the mode of heat transfer from the heating coils to the film surface is by convection, in contrast to direct conduction of heat to the Sylgard 184 layer on the hot stage of the Rheometer. Consequently, it takes roughly tP ≈ 75 minutes for a film with CL = 10.0% to crosslink fully inside an air oven, in comparison to 35 minutes on the Rheometer stage. The time required for crosslinking reduces with increase in CL, as we observed tP for complete cross linking coming down from ≈ 110 minutes to ≈ 55 minutes as CL varies from 5.0% to 15.0%. In fact, we used a concept that is discussed later, that in a metal elastomer bilayer once the bottom layer is fully cross linked then λS and aS of the wrinkles does not change with longer tP to assess the precise curing times of the films with different CL. The final value of G/ in fully cross-linked films increase with higher CL of the bottom layer, which implies that the elastic stiffness of the bottom layer increases with increase in CL. Figure 2 shows the gradual variation of λS and aS of the wrinkles on the metal – polymer bilayer films on Sylgard 184 films pre-cured for different durations. The corresponding optical micrographs of some of these samples are shown in figure S3 of the online supporting information. The morphology of the wrinkles formed on partially cured PDMS films remain totally unaltered, even as partially cured PDMS continues to cross link slowly at room temperature. As metal hits the surface at adequately high temperature, the contact zone between the metal film and Sylgard crosslinks rapidly, due to which morphology of the wrinkles does not change further. This is shown in figure S4 of the online supporting information.



1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 8 of 22

Figure 2: Variation in morphology of the wrinkles with progressive pre-curing of the bottom layer Sylgard 184 layer. (A) 1 min, (B) 2 min, (C) 4 min, (D) 7 min, (E) 27 min, and (F) 70 min. CL ≈ 10.0% in all the frames.

Further, though the patterns are random and isotropic, FFT of the images clearly shows the existence of a dominant wavelength (λS) in each case, which is mentioned in the respective frames. The amplitude of the features (aS) is found out from the cross sectional AFM line scans. One can clearly observe that both λS and aS gradually reduce with longer duration of tP vis-à-vis enhancement of G/ of the substrate layer. It can also be seen that the cross section of the wrinkles are much more circular when tP is low (≤ 4 min) and becomes more kinky in films where the bottom layer is pre-cured longer. This is verified by calculating the dimension (D) of samples with different tP (Table S1, supporting information). We note that D increases from 2.19 to 2.64 as tP increases from 1 minute to 75 minutes. While CL = 10% in all the frames of figure 2, the trend that both λS as well as aS reduces with longer tP is same for bottom layer with other CL as well, which can be seen the subsequent figure.


Page 9 of 22

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


Figure 3: (A) Variation of wavelength (λs) of the wrinkles with change in the storage modulus (G/) of PDMS for different percentage of cross linker, Inset A1 shows similar trend in variation of the buckle wavelength for CL≈ 10% cross linker in PDMS; (B) Variation of amplitude (aS) of the buckles with change in storage modulus (G/) of the PDMS, for different cross linker ratio in PDMS, Inset B1 and B2 shows AFM morphology of wrinkles for CL≈ 10% and CL≈ 5% cross linker in PDMS (pre-cured for same duration of time)respectively.

Figure 3A shows the variation of λS of the wrinkles as a function of G/ of the pre-cured bottom layers for different CL. In order to have a greater understanding on the data, the variation of λS with G/ for films with CL = 10.0% is shown as inset to figure 3A. One can clearly observe in figure 3A that for each CL, the variation of λS with G/ can be broadly divided into three distinct regimes that relates to the distinct rheological regimes, as marked in Figure 1. From the log – log plot shown in the inset of figure 3A, we obtain the power law dependence of the form λS ≈G/ m for each regime. Regime 1 is characterized by initial rapid drop of λS with increase of G/. For CL = 10.0%, the regime lasts till G/ attains a value of ≈ 102 Pa, and spans for tP ≈ 2 minutes. The power law dependence for this regime is found to be λS ≈ G/ – 0.158 ± 0.012. In Regime 2, which spans from G/ ≈ 102 to G/ ≈ 105 Pa (tP varying between 3 to 6 minutes), λS becomes nearly invariant of G/, leading to a scaling relation λS ≈ G/ – 0.010±0.001. Figure 3A shows that the near invariance of λS with G/ is observed irrespective of CL of the bottom layer. Subsequent rapid drop of λS with G/ is observed in Regime 3, which spans from G/ ≈ 105 to G/ > 106 Pa and exhibits a functional dependence λS ≈ G/ – 0.415±0.034. In terms of pre-curing time, this regime lasts from tP ≈7 to 75 minutes for CL ≈ 10.0%. Beyond tP ≈ 75 minutes, λS does not change at all with any further



1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 10 of 22

pre-curing, indicating attainment of complete cross linking of the elastomeric bottom layer. Figure 3B shows that similar to λS, aS of the wrinkles also progressively reduce with gradual increase in G/, and the qualitative nature of dependence with G/ is rather similar to that observed for λS in figure 3A. the nature of variation of λS as a function of G/ is qualitatively same for all bottom layers with CL between 5 to 15%, including the values of m (slope of the best fit curves for each regime, inset A1, figure 3A) for each regime.

Figure 4: Dependence of wrinkles morphology on thickness of Sylgard 184 layer (A) Change in wavelength λS with G/; (B) Change in amplitude aS with G/ for CL≈ 10.0%

The variation of λS and aS of the wrinkles as a function of the elastomeric bottom layer thickness (hS) are shown in figure 4A and 4B respectively. It can be seen that for higher values of G/ (Regime 3) both λS and aS are independent of hS. Interestingly, we observe that λS remains independent of hS in Regime 1 and 2 as well. This can be attributed to the stiffening of the upper layer of Sylgard film during metal due to localized heating, which triggers cross linking. The depth of this modified PDMS layer, which varies between 1 and 2 µm, depends only on the coating condition of the metallic film and does not depend on hS. As the compressive stresses responsible for wrinkle formation predominantly remain confined within this layer, therefore the magnitude of the compressive stresses remain nearly constant in films of different hS and consequently in figure 4A, λS is seen to be nearly independent of hS. On the other hand, figure 10 ACS Paragon Plus Environment

Page 11 of 22

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


4B shows that aS of the wrinkles rather strongly depend with hS in Regime 1, only in films with hS ≤ 5 µm. For bottom layers with hS ≥ 7 µm in Regime 1, as well as for any hS in Regimes 2 and 3, aS is seen not to depend on hS. The reason for variation of aS with hS for very thin films in Regime 1 is discussed in the subsequent section. At this point, it becomes evident that it is possible to control the λS of the wrinkles by simply tuning the visco elasticity of the elastomeric bottom layer. However, the most intriguing and non-trivial observation is the non-monotonic decay characteristic of λS with G/, particularly the near invariance of λS with G/ in Regime 2. None of the published papers on wrinkling of a metallic layer on a visco elastic bottom layer report such observation.31-35 This is because all the published articles focus on the evolution of wrinkle morphology as a function of the gradual change in the visco elasticity of the polymeric bottom layer, an example of which is the experimental study by Stafford and co. workers, which reports how the wrinkle morphology of an Aluminum thin films coated on polyhydroxystyrene (PHS) bottom layer with varying degree of chemical cross-linking changes with time, when the system is subject to high temperature annealing.35 In contrast, in the present study the metal film gets deposited on polymer layers with pre-existing levels of visco elasticity, and due to the high temperature of the deposited metal layer, the polymer layer in close proximity of it instantaneously crosslinks further. The patterns evolve during cooling of the sample, where the compressive stresses get released. Beyond this stage there is no evolution of the pattern morphology. While formulating the model on wrinkling over a visco elastic polymer layer, it is considered that the energy required for wrinkle formation is contributed by a combination of bending energy of the metal film and energy of deformation of the polymer film. It is further argued that wrinkle stress is minimized at a particular characteristic wavelength, which is manifested in 11 ACS Paragon Plus Environment


1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 12 of 22

experiments.36 Based on these considerations, the functional dependence of λS with the substrate stiffness for purely elastic and viscous regimes are given as:10,35

M ,

2

for elastic regime

….. (1)

, for viscous regime

….. (2)

E, ν and h are the Young’s Modulus, Poisson’s ratio and film thickness respectively with the subscript P and M denoting polymer and metal respectively. As EM, νM, hM, hP are constants, the above equations eventually predict the power law dependence as λS ≈ G/

-1/6

for the viscous

regime (for a particular hP) and λS ≈ G/ -1/3 for the elastic regime.36 From the inset of figure 3A it can be noted that the value of m in Regime 1 and Regime 3 are ≈ – 0.158 ± 0.012 and ≈ – 0.4158 ± 0.034 respectively, which are close to the theoretically predicted values. However, neither equation 1 nor equation 2 explains the invariance of λS with G/ in Regime 2. We propose a qualitative argument based on the mechanism of pattern formation in Regimes 1 and 3 to explain the behavior observed in Regime 2.

Figure 5: Schematics representation of wrinkle formation mechanism across different regimes.


Page 13 of 22

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


As already stated, the bottom Sylgard layer gets heated up during deposition of the metal film. Further, for a fully cured elastomer bottom layer (Regime 3), the adhesion between the metal and the polymer is very strong with no interfacial slippage. As the sample cools down, compressive stress gets generated at the metal – polymer interface due to significant difference in their coefficients of thermal expansion. The wrinkles form as the compressive stress gets relieved. The physical cross links tries to prevent the deformation of the elastomeric layer, resulting in wrinkles with low periodicity. On the other hand, the adhesion between the metal and a liquid like viscous Sylgard 184 layer, pre-cured for a short duration is weak, resulting in significant interfacial slippage.36 It is argued that in Regime 1, during thermal deposition, the high energy metal atoms in vapor state bombards the polymer surface and directly deforms the free surface of the liquid like film, resulting in surface undulations. The thermal compression further gets released as the liquid can slide below the metal layer, due to interfacial slip. This results in wrinkles with large λS and aS. In Regime 1 as the extent of cross linking enhances with longer tP, the partially cross-linked elastomer matrix tries to resist the direct deformation of the layer during deposition, which is manifested as rapid reduction of λS and aS with G/. However, figure 1 reveals that Regime 1 is entirely dominated by viscous effects as within this regime G// is always > G/. This leads to significant viscous dissipation of energy from a viscous bottom layer and as a result the interfacial stress accumulation during cooling becomes negligible. Further, we argue that in Regime 1 the extent of deformation of the liquid film depends only on the momentum of the deposited metal layer. Ideally this would lead to invariance of aS (≈ 6 µm) with hS, which is indeed the case in thicker films (hS ≥ 7 µm). However, when hS is of the order of 6 µm or lower, the rigid substrate truncates the deformation of the film, and consequently aS drops to lower values in Regime 1, when hS ≤ 5 µm. Thus, as shown in figure 5, almost the entire deformation



1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 14 of 22

takes place in the liquid phase, during deposition itself in Regime 1. This is in total contrast to that in Regime 3, where there is almost no deformation during deposition, and the wrinkles appear during cooling. In Regime 2 on the other hand, both elastic and viscous effects are dominant, as the values of G/ and G// are of the same order. Consequently, within this regime two simultaneous effects happen with increase in G/. Firstly, the film becomes stiffer and therefore the extent of direct deformation during thermal evaporation gradually reduces. However, with increase in G/ the viscous dissipation also reduces, particularly in the later stage of Regime 2, once the value of G/ exceeds G//, and interfacial stress starts to get accumulated during deposition. This stress gets relieved when the sample is cooled. Thus the wrinkle formation in Regime 2 is contributed by two separate effects. The wrinkle formation in the initial stage of Regime 2 is dominated by direct viscous deformation and low stress release. One the other hand, though with increase in G/, the direct deformation gradually reduces, the stress release induced wrinkling enhances. It turns out that a combination of these two effects keeps λS nearly constant over Regime 2. This is schematically shown in figure 5, where we see that the extent of direct deformation gradually reduces and wrinkling due to stress release progressively enhances with increase in G/ in Regime 2. Towards the end of Regime 2, the direct viscous deformation almost tends to zero and subsequent wrinkle formation in Regime 3 gets entirely dominated by interfacial stress release. Our observations reveal several other interesting scientific aspects, particularly about the pattern formation mechanism. Hong Lee, in an earlier paper has argued that wrinkles on a metal film coated on a purely viscous layer eventually flattens out as the liquid underneath slips and fails to pin.16 However, we observed wrinkles with large λS in our system. This happens as the thermo curable elastomer (Sylgard 184) locally crosslinks while deforming once the hot metal atoms 14 ACS Paragon Plus Environment

Page 15 of 22

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


deposit on the film surface, making the patterns permanent. This effect can also explain the occurrence of wrinkles with lower λS with decrease in CL in Regime 1 (figure 3A). We argue that as the CL reduces the rate of local crosslinking during impingement of hot metal atoms to the film surface also gets reduced, which in turn allows some degree on interfacial slippage between the viscous polymer and the metal, before the former is fully cross-linked. This argument fully explains why λS is maximized for CL = 15.0% in Regime 1. On the other hand, Lee et al. used polystyrene bottom layer, which does not crosslink and therefore the wrinkles eventually flattened away.16 We also reheated samples of all the three regimes to observe any variation in pattern morphology. The wrinkles in Regime 3 almost completely disappeared (reappeared) upon heating (cooling) the sample to 160 °C. In contrast the λS and aS of Regime 1 films remains nearly unaltered upon reheating. In Regime 2 films, the aS of the films reduces substantially while λS remains nearly unaffected. This observation also endorses the mechanism for each regime that we have argued above.

Figure 6: (A) Reflectance spectra for buckles of different wavelength and height; Inset shows the reflection image on bare polished silicon wafer (Inset A1), thin PDMS film coated on wafer(Inset A2); aluminum deposited on wafer (Inset A3); wrinkles with λS ~ 10µm (Inset A4); (B) Variation of water contact angle for buckles of different wavelength.

We have so far discussed how wrinkles with programmable values of λS and aS can be obtained by forming a metal polymer bilayer using thermal evaporation, where the visco elastic nature of 15 ACS Paragon Plus Environment


1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 16 of 22

the bottom polymer layer is precisely modulated. We now show that presence of such programmable wrinkles enables fabrication of surfaces with controlled wettability as well as coatings with broadband anti reflecting properties. Figure 6A shows the reflectance spectra of the wrinkled surfaces with different λS (plot group iv), along with the spectra for a flat aluminum layer (plot i), bare glass substrate (plot ii) and a flat Sylgard 184 thin film (plot iii). There is a drastic decrease in % reflectance for all the samples with wrinkled films as compared to the bare substrates. The wrinkled morphology with λS = 10µm showed the best anti reflecting property with a minimum % reflectance of 6.0% at 477 nm, whereas the substrate and the flat metal film has a reflectance of 12.15% and 40.49% respectively at 477 nm. An overall average reflectance of ~ 6.87% was observed for all wrinkled samples with λS ranging from 20µm down to 0.8µm, as compared to 39.0% for metal layer in the entire visible spectra. This reduction in reflectance in the visible region was obtained despite of the fact that none of the samples had sub wavelength structures, as essential for structural anti-reflective property.39 It is argued that the reduced reflectance is associated with localized oxidation of Al layer during deposition on cross-linked PDMS substrate. This results in the formation of an interfacial thin AlOx layer,4 which in turn transforms the obtained patterns into inhomogeneous, graded refractive index (GRIN) layered structures.40 The matching in the optical impedance due to this gradient nature causes the observed reduction in reflection and an increased internal scattering of the incident light. Such surfaces enhance the internal light retention capacity of any device, and thus these wrinkled surfaces when used as a back reflector may lead to enhancement in the performance of optoelectronic devices. For example, Kim et. al. reported improved power conversion efficiency and a 47% increase in photocurrent of polymer solar cells fabricated on wrinkled substrates.9


Page 17 of 22

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


Koo et. al. utilized the wrinkle structures to enhance the light extraction efficiencies of organic light-emitting diodes (OLEDs) by at least doubling the outcoupling of waveguide modes.41 Figure 6B shows that simultaneous modulation of λS and aS also results in variation of wettability of the surfaces. It is well known that surfaces with nano or meso scale structures exhibit either Cassie or Wenzel wetting regimes, depending on the roughness ratio of the surface (RF). A flat cross linked Sylgard 184 film is weakly hydrophobic and exhibits an equilibrium contact angle of θE ≈ 102° with water. On the other hand, a flat layer of Al thin film coated on glass exhibits θE ≈ 62.3° (inset B1, figure 4B). However, figure 4B shows excellent modulation of θE between ≈ 79.4° and 131.2°, which offers the possibility of tailoring the wettability of the surface on demand, including possible modulation between hydrophilic and hydrophobic surfaces. What is most interesting is an Al surface is seen to be hydrophilic, and ideally if the wetting regime is Wenzel, then on a structured surface the effective equilibrium contact angle (θ*) should ideally be lower than θE. However, that is not observed in our results. Of course one can argue as the Al film is very thin, the effective surface tension of the Al- Sylgard bilayer is lower than that of pure Aluminum, which is likely to be responsible for higher values of θ*. The plot also shows that θ* is maximized at intermediate values of λS between 5 µm and 10 µm. The corresponding values of aS between 600 nm to 1.0 µm. The values of RF vary between 1.05 and 1.12, and clearly indicate a Wenzel type regime. Thus, it is indeed bit surprising to obtain significantly high values of θ* (up to ≈ 130°) in these structures. As we mentioned in the context of figure 2C and 2D, the patterns in this range (Regime 2) are kinky, as the wrinkles are not smooth. It is likely that these kinks result in some sort of secondary roughness that might lead to mixed wetting regimes, as has been suggested by Jiang,42 and observed recently by us in the context of high aspect ratio



1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 18 of 22

structures.43 However, a detailed understanding on the precise mechanism of wetting is however beyond the scope of this paper and is being taken up separately. Conclusion: In this article we have shown that the λS and as of the thermal wrinkles on a metal polymer bilayer can be modulated by varying the viscoelasticity of the elastomeric bottom layer. This technique is more versatile than the existing practice of modulating λS and as by tailoring the thickness of the metallic thin film, which allows variation of λS only over a rather narrower range.30 Also thicker metal films are prone to cracking. In contrast, the method reported here allows modulation of λS over a wide range varying from ≈ 20 µm down to ≈ 800 nm. We also identify three clearly distinct regimes of variation of λS with G/, and discuss how the wrinkles forming mechanism varies in each of them. Of particular interest is the identification of a regime, where for rather wide range of G/ (varying between 102 Pa to 105 Pa), λS remains nearly independent of G/, due to combined effect of viscous and elastic components of the bottom layer. The technique is facile, lithography free and does not require any sophisticated hardware and is therefore ideally suited for bulk nano fabrication of wrinkled surfaces which are widely used as templates for various applications. As an example we have demonstrated that wrinkles with different λS and aS can be used for fabrication of surfaces with controlled wettability and reflectance. Supporting Information: Details about the FFT analysis procedure, calculation of Fractal Dimension, Large area optical mircrographs and AFM images to convince that the wrinkle morphology remains unaltered over prolonged period is available in the Online Supporting Information.


Page 19 of 22

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


References 1. Khang, D. Y.; Rogers, J. A.; Lee, H. H. Mechanical Buckling: Mechanics, Metrology, and Stretchable Electronics. Adv. Funct. Mater. 2009, 19, 1526–1536. 2. Chan, E. P.; Smith, E. J.; Hayward, R. C.; Crosby, A. J. Surface Wrinkles for Smart Adhesion. Adv. Mater. 2008, 20, 711–716. 3. Stafford, C. M.; Harrison, C.; Beers, K. L.; Karim, A.; Amis, E. J.; VanLandingham, M. R.; Kim, H-C; Volksen, W.; Miller, R. D.; Simonyi, E. E. A Buckling-based Metrology for Measuring the Elastic Moduli of Polymeric Thin Films. Nat. Mater. 2004, 3, 545–550. 4. Koo, W. H.; Zhe, Y.; So, F. Direct Fabrication of Organic Light-Emitting Diodes on Buckled Substrates for Light Extraction. Adv. Opt. Mater. 2013, 1, 404–408. 5. Rogers, J. A.; Someya, T.; Huang, Y. Materials and Mechanics for Stretchable Electronics. Science 2010, 327, 1603–1607. 6. Li, Y.; Dai, S.; John, J.; Carter, K. R. Superhydrophobic Surfaces from Hierarchically Structured Wrinkled Polymers. ACS Appl. Mater. Interfaces 2013, 5, 11066–11073. 7. Efimenko, K.; Finlay, J.; Callow, M. E.; Callow, J. A.; Genzer, J. Development and Testing of Hierarchically Wrinkled Coatings for Marine Antifouling. ACS Appl. Mater. Interfaces 2009, 1, 1031–1040. 8. Yin, J.; Boyce, M. C. Materials Science: Unique Wrinkles as Identity Tags. Nature, 2015, 520, 164–165. 9. Kim, J. B.; P. Kim, N. C. Pegard, S. J. Oh, C. R. Kagan, J. W Fleischer, H. A. Stone and Y. L. Loo, Wrinkles and Deep Folds as Photonic Structures in Photovoltaics. Nat. Phot. 2012, 6, 327– 332. 10. Bowden, N.; Brittain, S.; Evans, A. G.; Hutchinson, J. W.; Whitesides, G. M. Spontaneous Formation of Ordered Structures in Thin Films of Metals Supported on an Elastomeric Polymer. Nature 1998, 393, 146–149. 11. Yang, B. S.; Khare, K.; Lin, P-C. Harnessing Surface Wrinkle Patterns in Soft Matter. Adv. Func. Mater. 2010, 20, 2550–2564. 12. Kim, H. S.; Crosby, A. J. Solvent-Responsive Surface via Wrinkling Instability. Adv. Mater. 2011, 23, 4188–4192. 13. Chen, C-M.; Reed, J. C.; Yang, S. Guided Wrinkling in Swollen, Pre-patterned Photoresist Thin Films with a Crosslinking Gradient. Soft Matter 2013, 9, 11007–11013. 14. Watanabe, M.; Mizukami, K. Well-Ordered Wrinkling Patterns on Chemically Oxidized Poly(dimethylsiloxane) Surfaces. Macromolecules 2012, 45, 7128–7134. 19 ACS Paragon Plus Environment


1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 20 of 22

15. Bowden, N.; Huck, W. T. S.; Paul, K. E.; Whitesides, G. M. The Controlled Formation of Ordered, Sinusoidal Structures by Plasma Oxidation of an Elastomeric Polymer. Appl. Phys. Lett. 1999, 75, 2557–2559. 16. Yoo, P. J.; Lee, H. H. Morphological Diagram for Metal/Polymer Bilayer Wrinkling: Influence of Thermomechanical Properties of Polymer Layer. Macromolecules 2005, 38, 2820–2831. 17. Vandeparre, H.; Le´opolde`s, J.; Poulard, C.; Desprez, S.; Derue, G.;Gay, C.; Damman, P. Slippery or Sticky Boundary Conditions: Control of Wrinkling in Metal-Capped Thin Polymer Films by Selective Adhesion to Substrates. Phys. Rev. Lett. 2007, 99, 188302-1–188302-4. 18. Sharp, J. S.; Thomas, K. R.; Weir, M. P. Mechanically Driven Wrinkling Instability in Thin Film Polymer Bilayers. Phys. Rev. E 2007, 75, 011601. 19. Huck, W. T. S.; Bowden, N.; Onck, P.; Pardoen, T.; Hutchinson, J. W.; Whitesides, G. M. Ordering of Spontaneously Formed Buckles on Planar Surfaces. Langmuir 2000, 16, 3497–3501. 20. Chua, D. B. H.; Ng, H. T.; Li, S. F. Y. Spontaneous Formation of Complex and Ordered Structures on Oxygen-plasma-treated Elastomeric Polydimethylsiloxane. Appl. Phys. Lett. 2000, 76, 721–723. 21. Yoo, P. J.; Suh, K. Y.; Park, S. Y.; Lee, H. H. Physical Self-Assembly of Microstructures by Anisotropic Buckling. Adv. Mater. 2002, 14, 1383–1387. 22. Yoo, P. J.; Lee, H. H. Evolution of a Stress-Driven Pattern in Thin Bilayer Films: Spinodal Wrinkling. Phys. Rev. Lett. 2003, 91, 154502-1–154502-4. 23. Yoo, P. J.; Lee, H. H. Polymer Elasticity-Driven Wrinkling and Coarsening in High Temperature Buckling of Metal-Capped Polymer Thin Films. Phys. Rev. Lett. 2004, 91, 034301-1–034301-1. 24. Okayasu, T.; Zhang, H. L.; Bucknall, D. G.; Briggs, A. D. Spontaneous Formation of Ordered Lateral Patterns in Polymer Thin-Film Structures. Adv. Funct. Mater. 2004, 14, 1081–1088. 25. Chen, X.; Hutchinson, J. W. Herringbone Buckling Patterns of Compressed Thin Films on Compliant Substrates. J. of Appl. Mech. 2004, 71, 597–603. 26. Kwon, S. J.; Park, J-G.; Lee, S. H. Morphological Dynamics of Swelling-induced Surface Patterns in Metal-capped Polymer Bilayer. J. Chem. Phys. 2005, 122, 031101-1–031101-4. 27. Hyun, D. C.; Jeong, U. Substrate Thickness: An Effective Control Parameter for Polymer Thin Film Buckling on PDMS Substrates. J. Appl. Polym. Sc. 2009, 112, 2683–2690. 28. Yoo, P. J.; Lee, H. H. Complex Pattern Formation by Adhesion-Controlled Anisotropic Wrinkling. Langmuir 2008, 24, 6897–6902. 29. Park, J-Y.; Chae, H. Y.; Chung, C-H; Sim, S. J.; Park, J.; Lee, H. H.; Yoo, P. J. Controlled Wavelength Reduction in Surface Wrinkling of Poly(dimethylsiloxane). Soft Matter 2010, 6, 677–684. 20 ACS Paragon Plus Environment

Page 21 of 22

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


30. Yu, S.; Ni, Y.; He, L.; Ye, Q-L. Quantifying the Stress Relaxation Modulus of Polymer Thin Films via Thermal Wrinkling. ACS Appl. Mater. Interfaces 2015, 7, 5160–5167. 31. Huang, R.; Suo, Z. Wrinkling of a Compressed Elastic Film on a Viscous Layer. J. Appl. Phys. 2002, 91, 1135–1142. 32. Huang, Z.; Hong, W.; Suo, Z. Evolution of Wrinkles in Hard Films on Soft Substrates. Phys. Rev. E 2004, 70, 030601-1–030601-2 33. Huang, R. Kinetic Wrinkling of an Elastic Film on a Viscoelastic Substrate J. Mech. Phys. Solids 2005, 53, 6–89 34. Im, S. H.; Huang, R. Evolution of Wrinkles in Elastic-Viscoelastic Bilayer Thin Films. J. Appl. Mech. 2005, 72, 955–961. 35. Chan, E. P.; Lin, Q.; Stafford, C. M. Quantifying the Elasticity and Viscosity of Geometrically Confined Polymer Films via Thermal Wrinkling. J. Polym. Sci. Part B Polym. Phys. 2012, 50, 1556–1561. 36. Yu, S.; Sun, Y.; Ni, Y.; Zhang, X.; Zhou, H. Controlled Formation of Surface Patterns in Metal Films Deposited on Elasticity-Gradient PDMS Substrates. ACS Appl. Mater. Interfaces 2016, 8, 5706−5714. 37. Bhandaru, N.; Roy, S.; Harikrishnan, G.; Mukherjee, R. Lithographic Tuning of Polymeric Thin Film Surfaces by Stress Relaxation. ACS Macro Lett. 2015, 2, 195–200. 38. Roy, S.; Bhandaru, N.; Das, R.; Harikrishnan, G.; Mukherjee, R. Thermally Tailored Gradient Topography Surface on Elastomeric Thin Films. ACS Appl. Mater. Interfaces 2014, 6, 6579−6588. 39. Walheim, S.; Schäffer, E.; Mlynek, J.; Steiner, U. Nanophase-Separated Polymer Films as HighPerformance Antireflection Coatings. Science. 1999, 283, 520–522. 40. Raut, H. K.; Ganesh, V.; Nair, A. S.; Ramakrishna, S. Anti-reflective Coatings: A Critical, inDepth Review. Energy Environ. Sci. 2011, 4, 3779–3804. 41. Koo H.W.; Jeong, S. M.; Araoka, F.; Ishikawa, K.; Nishimura, S.; Toyooka, T.; Takezoe, H. Light Extraction From Organic Light-Emitting Diodes Enhanced by Spontaneously Formed Buckles. Nat. Phot. 2010, 4, 222–226. 42. Wang, S.; Jiang, L. Definition of Superhydrophobic States. Adv. Mater. 2007, 19, 3423–3424. 43. Bhandaru, N.; Mukherjee, R.; Sharma, A. Programmable Nanopatterns by Controlled Debonding of Soft Elastic Films. ACS Appl. Mater. Interfaces 2016, 10.1021/acsami.6b09127.



1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 22 of 22

TOC Graphic Image


Programming Feature Size in the Thermal ... - ACS Publications

Recommend Documents