Effect of Thermal Treatment on the Properties of ... - ECS Transactions

ECS Transactions, 50 (2) 951-959 (2012) 10.1149/05002.0951ecst ©The Electrochemical Society

Effect of Thermal Treatment on the Properties of Ultra-Thin Nafion Film Devproshad K. Paul1,2 and Kunal Karan1,2* 1

Dept of Chemical Engineering, Queen’s University, Kingston, ON, Canada, 2Queen’sRMC Fuel Cell Research Centre,Kingston, ON, Canada. E-mail addresses: [email protected], Telephone number: 1-613-533-3095. Fax number: 1-613-533-6637.

Recently, we reported results for ultra-thin Nafion films ranging 4 nm - 300 nm self-assembled on SiO2 terminated silicon surface. Very distinct wettability for these films were noted; thinner films (100 nm) films showed hydrophobic surface nature. In this study, we report on the effect of annealing (at 146 oC for 1 hour) on film properties. Water contact angle measurement showed that thinner films turned into hydrophobic resulting similar contact angle to thicker film. This phenomenon indicated that significant surface rearrangement of thinner film happened during annealing period. Moreover, phase contrast correspond to the hydrophilicity/hydrophobicity gave a distinguishable surface properties of unannealed vs annealed and thinner vs thicker films.

Introduction Recently, we reported the characteristics of self-assembled films of Nafion on thermally grown SiO2 surface of silicon wafer (1). The film thickness ranged from ~4 nm to ~300 nm. The surface wettability and proton conductivity of these films were measured. Interestingly, two different behaviors with respect to contact angle were noted. Thinner films (100 nm) films showed hydrophobic contact angles. These films were not subjected to any heat treatment steps. During the fabrication of fuel cell membrane electrode assemblies, the catalystcoated membranes are hot-pressed with the carbon paper backing (2). Heat treatment of polymer films, including Nafion films (3), is known to affect wettability of films on the substrate. Thermal treatment of polymers near glass transition temperature helps relax the polymer and induces ordering. Thus, we were interested in studying the effect of heat treatment on ultra-thin Nafion films. There are very few studies of annealing effect on thin Nafion film. Maeda et al. (4) prepared thin film on patterned glass substrate by dropping 50 μL of Nafion solution onto the front side and dried in air. When the film was heated at 130 oC with a hot plate for 5 mins, the movement of Nafion molecules at the surface caused by the annealing and destroyed the nanostructures several nanometers in vertical size. Hill et al. (3) studied thermal effect on self assembled thin ionomer film (200 nm and 400 nm) supported by silicon wafer. Atomic Force Microscope (AFM) revealed that even prolonged (more than 20 days) heat treatment at the temperature above Tg, did not dewet or break up the thin film which is not very common to other thin diblock copolymer films. STM study

951

ECS Transactions, 50 (2) 951-959 (2012)

showed that ionomer exist as 3-5 nm diameter nano-aggregated structure inside of 20 nm diameter hexagonally packed cluster which led to form 200 nm circular domain but 200nm bundles start to break up into individual micelles with temperature annealing. Recently, Medestino et al. (5) characterized spin cast100 nm film (annealed at 200 oC) properties both on SiO2 and Octadecyltrichloro-silane (OTS) modified substrate by Grazing Incident Small Angle X-ray Scattering (GISAXS). They found that with thermal annealing of thin film, the crystalline domains formed with in the film matrix and restricted water uptake and reduced conductivity as well. In this study, we have investigated the thermal effect on the surface characteristics of Nafion thin films. Water contact angle measurement technique was used to examine the change in surface wetting characteristics before and after annealing. In addition, AFM in the tapping mode was employed to study the surface morphologies, phase contrast and roughness of the films.

Experimental Thin film preparation Nafion thin film preparation involves three steps. Firstly, commercially available 5 wt% Nafion solution in 20% water and 80% alcohol mixture, EW 1100 (Ion Power) was diluted to a desired solution concentration, 0.1 wt% to 3 wt%. For dilution, isopropyl alcohol was used where ionomer to water ratio was maintained constant. The diluted solution was allowed to be equilibrated at least 24 hours after initial sonication. Second step involves substrate cleaning, silicon wafer terminated with 300 nm SiO2 film was used a substrate. The substrate was cleaned according to the procedure as described in our previous report [1]. Firstly, substrate was kept into Piranha solution (H2SO4:H2O2;7:3) at 80-90 oC for half an hour to remove the contamination on the surface. The wafer was rinsed up with deionized water and dried with dry air blow. Further the wafer was placed into the solution mixture (DI H2O: NH4OH: H2O2; 5:1:1), called RCA (Radio Corporation of America) cleaning, at 70-80 oC for another half an hour to remove organic contaminants from the surface. The cleanness of wafer was confirmed by the following AFM and XPS analysis. Third step is the sample preparation step. A range of thin film was prepared by selfassembled or adsorption of Nafion on the substrate. The pre-cleaned model silicon substrate was exposed to Nafion solution/suspension of known range of concentrations. After 12 hours exposure, substrate was took out from the solution and, subsequently, dried by dry air blow. We have found that 12 hours is sufficient to prepare fully covered homogeneous film on the substrate. Thermal treatment The thin films were classified into two categories based on further heat treatment and termed as unannealed and annealed films. Unannealed films involved two drying steps, (i) 30 min drying at N2 blow right after film preparation (ii) dry at vacuum (-30 inch Hg) oven at 40 oC for 17 to 20 hours. Annealed films were involved additional one step where

952


unannealed film was heat treated at 146 oC under vacuum for 1 hour. Before placing the sample in the oven for annealing the oven was equilibrated at 146 oC for 3 to 4 hours period and well maintained temperature with little fluctuation ±0.5 oC during the annealing. Both annealed film was exposed to room temperature after the thermal treatment step and equilibrated with outside atmosphere before any characterization. All the characterization was conducted in ambient temperature and relative humidity. Thin film Characterization

Ellipsometry Measurements: Thickness and optical parameters of self-assembled Nafion thin films were determined for both types of films by VASE® Ellipsometer (J. A. Woollam Co., Inc.). Ellipsometry measurement was carried out at three different angles (55, 65 and 75 o) over the 300 to 800 nm range. The data complex reflectance ratio, ȡ, of a system, which is parameterized by Ȍ and ǻ.

U

rp rs

tan(\ )ei'

[1]

where, tan (Ȍ) is the amplitude ratio upon reflection, and ǻ is the phase shift (difference). The thickness and refractive index was obtained fitting the data of Ȍ and ǻ at a range of wavelength (300 to 800 nm) with Cauchy model using WVASE32® software. The thickness and optical parameters have been summarized in the previous report (1). Different thicknesses, 4 nm, 55 nm, 160 nm and 307 nm thickness was obtained from 0.1 wt%, 1.0 wt%, 3 wt% and 5 wt% respectively.

Contact angle measurement: Water contact angle was measured for both unannealed and annealed films by Goniometer. The contact angle analyzer equipped with a video camera and software for image grabbing and analysis. A sessile drop of water about 0.75 μL large was placed on the top of the surface of both type of unannealed and annealed thin film using a micro syringe. The imaged was captured within 5 sec of the water drop placement on the surface. Several images were also taken every 30 sec interval for 2 to 3 min to check the stability of the droplet. It was observed that the contact angle of sessile drop decreases with time, this might happen due to adsorption of water by the polymer or water evaporation. The reported values are the ones obtained within the first approximately 5 seconds after placing the drop. For annealed film, the film was further exposed to 100% RH for one week. Again the contact angle was measured following the above measuring protocol. Before contact angle measurement, the humidified films were dried in the vacuum oven similar to unannealed film (at 40 oC for 15 h) and equilibrated in the ambient condition.

AFM measurement: The Nafion film morphologies and phase contrast before and after annealing were determined by AFM imaging. The AFM (Di Digital Instrument, Nanoscope IIIa) was performed with silicon tip (Tip radius < 10 nm) mounted on the cantilevers in tapping mode at ambient laboratory condition. Further image analysis and

953


roughness calculation were performed using software. For thickness measurement, the film was scratched with sharp needle gently without damaging the substrate. The thickness was obtained from the image of scratched thin film image.

Results and Discussion The characteristics of supported ionomer films can be classified into three separate properties (a) substrate-ionomer interfacial properties (b) bulk properties and (c) gasionomer surface properties. Interfacial properties are largely dominated by the substrate surface characteristics. Substrate surface microstructure (6) and wettability (7, 8) greatly influence on the nanostructure/orientation of the film, which may determine the bulk properties. Depending on the extent of the substrate influence and the film thickness, substrate might play role in determining the free surface (ionomer-gas) properties too. It is a most important question to answer, how far the substrate may affect on nanostructure and corresponding film properties. However, thermal annealing adds a new dimension of molecular rearrangement/reorientation of corresponding bulk and free surface properties determination.

Contact angle result analysis: Water contact angle was measured to investigate the ionomer surface wettability. Contact angle is calculated from three-phase contact line (TCL) where water and vapor phase make an angle with the substrate or solid phase. Depending on contact angle, it can be characterized the surface is as completely wetting, intermediate wetting or completely non wetting. When the contact angle is zero, it indicates that the surface is completely wetting. On the other hand, completely nonwetting corresponds to the contact angle 180o. Contact angle in between zero to 180o represents intermediate wetting, which can be classified into two terms - hydrophilic or hydrophobic. If the water contact angle of a surface is less than 90o then it is called hydrophilic surface, and if it is more than 90o the surface is termed as a hydrophobic surface. Fig.1 represents water droplet image on 4 nm and 55 nm films before and after annealing. Before annealing, the film was dried at 40 oC in vacuum for 16 to 20 hours. This vacuum dried film is denoted as the unannealed film. On the other hand, after this unannealed film was heat treated at 146 oC for 1 hour, it is referred to as the annealed film. First of all, 4 nm unannealed film showed very low water contact angle, indicating that the surface is very hydrophilic (Fig. 1a). Contact angle was very low or almost spreading for the 55 nm film also (Fig. 1c). Interestingly, whereas the 55 nm annealed film surface turned into hydrophobic, the 4 nm film did not (Fig. 1b and 1d). For 4 nm film, contact angle was found to be ~40o, still showing a hydrophilic characteristic whereas more than 100o contact angle was elevated for 55 nm film. We also investigated the intermediate film thickness which also behaved similar to 55 nm film. It indicates that there is an enormous surface rearrangement occurred during annealing period. The surface rearrangement of 4 nm film might not occur to a similar extent or perhaps some de-wetting may be occurred exposing the underlying SiO2 surface to the water droplet.

954


4 nm a

L:21.0

55 nm

Before annealing

Before annealing

c

L:10.1

R:20.0

d b

L:56.5

R:10.2

After annealing

After annealing

R:59.2

L:107.0

R:107.2

Figure 1. Water contact angle of thinner films before and after annealing, (a) 4 nm unannealed film (b) 4 nm annealed film (c) 55 nm unannealed film (d) 55 nm annealed film. We also examined films prepared from higher Nafion concentration, 3wt% and 5 wt%, corresponding to the film thicknesses of 160 nm and 307 nm, respectively. It was found that both the unannealed film from higher Nafion concentration show high contact angle, 106o (Fig. 2a and 2c), which is completely different from unannealed films of thickness of 55 nm and lower. Therefore, two type of surface characteristic is clearly evident. Self-assembled Nafion thin film prepared from lower concentration (1wt %) is hydrophilic whereas film prepared by the same method from Nafion concentration higher than 1 wt% give hydrophobic surface nature before high temperature treatment. When both 106 nm film and 307 nm film were heat treated via the same annealing procedure, both of these annealed films showed very little contact angle change (2o) which is indicative to very less surface rearrangement of those comparatively thicker films (Fig. 2b and 2d). The annealed film was hydrated with 100% relative humidity at room temperature for a week. Then contact angle was measured after similar treatment done for unannealed film (dry at 40 oC in vacuum oven for 15 hours) but still the surface was hydrophobic. It gave the idea that the surface stability is too high and even prolonged high relative humidity exposure cannot be able to get back the surface properties similar to unannealed state. However, the whole phenomenon generates some questions. First of all, why the surface properties of unannealed films are different based on preparatory Nafion solution concentration and their corresponding thicknesses? Secondly, what exactly happens to the ultra-thin unannealed film which might change the surface properties highly hydrophilic to hydrophobic and why not for high concentration film? Thirdly, what is the

955


reason of 4 nm film to be an exception in terms of surface switchibility? To find out the question answer we adopted AFM as a further characterization tool.

160 nm film a

Before annealing

L:106.7

b

307 nm film

R:106.0

After annealing

L:108.0

c

R:108.3

Before annealing

L:106.6

d

R:106.1

After annealing

L:107.5

R:108.2

Figure 2. Water contact angle of thicker films before and after annealing, (a) 160 nm unannealed film (b) 160 nm annealed film (c) 307 nm unannealed film (d) 307 nm annealed film.

AFM image analysis: Fig. 3 represents AFM topographic, phase image and section analysis of 55 nm films before and after annealing. The surface morphology of unannealed film was measured by 512 scanning point which showed that the surface aggregated ionomer features with around 100 nm diameter (light color) (Fig. 3a). The surface roughness was calculated 0.31 nm which indicate that the surface was smooth. The section analysis also supported around 100 nm feature which made of several bundle where maximum surface height was found 1.5 to 2 nm (Fig. 3c). When the film was heat treated, the surface became smoother where the surface aggregated features had been sufficiently dispersed (mostly light color). This phenomenon further supported by surface roughness calculation and section analysis of the topographic image. The surface roughness went down to 0.21 nm. The section analysis resulted very smooth and planner shape having maximum surface height 0.5 nm to 0.8 nm (Fig. 3e and 3g). More details of the surface feature were observed from the phase image before and after annealing. Usually phase can be differentiated by surface specific properties instead of corresponding height.

956


Figure 3. AFM topographic and phase images of 55 nm film before and after annealing, (a) Topographic image of unannealed film (b) Phase image of unannealed film (c) Section analysis of image (a), (d) Section analysis of image (b), (e) Topographic image of annealed film (f) Phase image of annealed film (g) Section analysis of image (e), (h) Section analysis of image (f).

957


There might be several reasons that are responsible for generating phase contrast but three main reasons are very prominent in this particular case. Three reasons are (i) heterogeneity in terms of surface materials and roughness, (ii) surface softness and hardness, (iii) surface hydrophilicity and hydrophobicity. First of all, the phase image of unannealed film show rod-like features (lighter part) parallel to the surface and section analysis of the corresponding phase image also show the periodicity of 20 to 30 nm feature diameter in the section analysis (Fig. 3b and 3d). Moreover, we can easily estimate the feature length from the image which would be 70 to 100 nm. We think that the surface consists solely of ionomer and surface roughness of corresponding topographic image is very negligible. Thus, the first of the three reasons for phase contract change identified above might be ruled out or contributions due to this effect can be considered to be negligible. Nafion belongs to the class of graft copolymers, which are known to undergo phase segregation. To support this argument, Nafion might be considered to be aggregated through the hydrophobic backbone to form rigid cluster whereas sulfonate side chain might be outward of the cluster surface. Therefore, the surface might be classified into harder region attributed by hydrophobic cluster and softer region originated by hydrophilic sulfonic acid side chain and associated water. If lighter color represents harder hydrophobic part and darker color indicates the hydrophilic part, it indicates that a significant amount of sulfonate group presence in the film surface. This finding might explain the lower contact angle found for 55 nm as well as lower thickness films. Interestingly, those particular features were almost disappeared in the phase image of annealed film. The section analysis of the corresponding film showed comparatively flatter distribution. This might be due to drastic surface rearrangement of the surface molecules. During annealing, the side chain might gain sufficient mobility and hide inside of the bulk materials to minimize the surface energy. This phenomenon might explain the conversion of film surface from highly hydrophilic to hydrophobic. If we assign the phase contrast as a distinguishing feature for the identification of surface hydrophobicity and hydrophobicity, it was noticeable that unannealed corresponding films, prepared from higher Nafion concentration (3 wt% to 5 wt%) showed significantly lower phase contrast. For unannealed 55 nm film, root mean square (RMS) of phase angle was found to be 5.3o whereas that of unannealed 160 nm film was 2.0o. The RMS values for annealed 55 nm film was similar to that of the unannealed 160 nm film. Though corresponding films of low and high Nafion concentration showed hydrophobic after annealing (contact angle more than 100o), only thinnest, 4 nm film showed still hydrophilic after similar heat treatment. It can be explained by surface roughness and associated effects. It was observed that surface roughness of annealed 4 nm film went to double compare to unannealed counterpart whereas roughness of other films went down after annealing. The increased surface roughness might be associated with the dewetting effect that might be responsible for lower contact angle. Though the surface molecule might reorient during annealing, a large portion of sulfonate group still might expose to the surface due to dewetting effect.

Impedance measurement: Thermal annealing might alter the bulk properties too. As we previously reported (9) proton conductivity of self-assembled ionomer thin film is lower and associated with high activation energy. The proton conductivity measurement of annealed films is currently underway. Preliminary results indicate that the thermal annealing significantly depresses conductivity.

958


Conclusion We have investigated the thermal treatment effect on ultra thin ionomer film. It was found that, unannealed thinner films (highly hydrophilic) turned into hydrophobic due to thermal annealing. While unannealed thicker film (hydrophobic) does not change much in terms contact angle. Both thinner and thicker annealed films showed similar high contact angle, i.e., hydrophobicity. Significant change in surface roughness between unannealed and annealed films supports the idea of surface rearrangement. Moreover, we have found a strong correlation between phase contrast and surface wettability which explain and support the switchibility of film surface too.

Acknowledgements Financial assistance from Early Researcher Award (Ontario Ministry of Research and Innovation) and Natural Sciences and Engineering Research Council of Canada (NSERC) is acknowledged.

References 1. D. K. Paul, A. Fraser, J. Pearce, K. Karan, ECS Trans., 41, 1393, (2011). 2. T. Frey, M. Linardi, Electrochimica Acta, 50, 99, (2004). 3. A. T. Hill, L. D. Carroll, R. Czerw, W. C. Martin, D. J . Perahia, J. of Polymer Sci.: Part B: Polymer Phy., 41, 149, (2003). 4. Y. Maeda, Y. Gao, M. Nagai, Y. Nakayama, T. Ichinose, R. Kuroda, K. Umemura, Ultramicroscopy, 108, 529, (2008). 5. A. M. Modestino, A. Kusoglu, A. Hexemer, Z. A. Weber, A. R. Segalman, Macromolecules, 45, 4681, (2012) 6. T. Masuda, H. Naohara, S. Takakusagi, P. R. Singh, K. Uosaki, Chem. Lett., 38, 884, (2009) 7. M. Bass, A. Berman, A.Singh, O. Konovalov, and V.Freger, The J. of Phys. Chm. B , 114, 3784, (2010). 8. M. Bass, A. Berman, A.Singh, O. Konovalov, and V.Freger, Macromolecules, 44, 2893, (2011). 9. D. K. Paul, A. Fraser, K. Karan, Electrochem. Comm.13, 774, (2011).

959