Thin Solid Films 520 (2012) 18851889

Contents lists available at SciVerse ScienceDirect

Thin Solid Films

journal homepage: www.elsevier.com/locate/tsf

Preparation of polytetrauoroethylene by pulsed electron ablation: Deposition and wettability aspects

Redhouane Henda a,, Graeme Wilson a, Joy Gray-Munro b, Omar Alshekhli a, Andrew M. McDonald c
a b c

School of Engineering, Laurentian University, Sudbury, Canada Dept. Chemistry and Biochemistry, Laurentian University, Sudbury, Canada Dept. Earth Sciences, Laurentian University, Sudbury, Canada

a r t i c l e

i n f o

a b s t r a c t
Polytetrauoroethylene (PTFE) has been prepared by pulsed electron deposition technique on glass and silicon substrates. Deposition of the thin lms has been carried out in the temperature range from room temperature to 300 C, pressure range from 133.32 10 3 Pa to 799.93 10 3 Pa, and discharge voltages between 10 kV and 16 kV. Argon or nitrogen has been used as a background gas during the deposition of the lms. Attenuated Total Reection Fourier Transform Infrared spectroscopy shows absorption peaks in the lms at 644 cm 1, 1154 cm 1 and 1210 cm 1 consistent with those of PTFE target material. Atomic force microscopy and spectroscopic reectometry reveal the clustered nature of the lms and other morphological characteristics. Surface wettability of the lms, expressed via the contact angle, has been measured via static angle goniometry. PTFE lms increase the contact angle from about 32 (bare glass) and 43 (bare silicon) to up to 90 and 110 for PTFE-coated glass and silicon substrates, respectively. The contact angle decreases with an increase in both pressure and temperature, while it increases then decreases as the discharge voltage increases. Crown Copyright 2011 Published by Elsevier B.V. All rights reserved.

Article history: Received 28 June 2010 Received in revised form 10 September 2011 Accepted 14 September 2011 Available online 24 September 2011 Keywords: Pulsed electron ablation Polytetrauoroethylene Deposition Wettability

1. Introduction Polytetrauoroethylene (Teon) possesses a combination of technologically attractive properties including good corrosion resistance, biological compatibility, low dielectric constant, low friction coefcient, high chemical and thermal stability, good mechanical strength, and excellent plasticity. This set of properties has attracted considerable attention to polytetrauoroethylene (PTFE) for applications ranging from lubrication to molecular electronics [13]. PTFE thin lms have been prepared by many techniques such as vacuum evaporation [4], radio frequency sputtering [5], plasma evaporation [6], ion beam sputtering [7], and pulsed laser deposition (PLD) [816]. Many problems exist with these techniques specically in the area of controlling the structure/composition of the deposited lms. Conventional wet processes such as spin-coating methods are not applicable to the formation of thin PTFE lms as the latter exhibit poor solubility in all solvents and have low surface adhesion. Pulsed electron deposition (PED) has been a promising alternative to PLD in thin lm coating applications [17, 18]. PED is conceptually similar to PLD except that a short pulse (~100 ns) of energetic electrons replaces a short pulse of photons to ablate material from a target. Due to a short penetration depth (~1 m) of the electron beam into the target, rapid non-equilibrium heating results, leading to the formation of a

highly forward directed, stoichiometric plasma plume, which in turn facilitates stoichiometric preservation of the target composition in the deposited lm under optimal process conditions. Compared to lasers with similar power densities (~108 W/cm2), advantages of electron beam sources include higher electrical efciency (30%), the ability to process materials that are transparent or highly reective to laser light, and lower capital costs [18]. Nearly all solid-state materials with varying complexity can be deposited in thin lm form with PED. The pressure domain for thin lm deposition in PED is restricted relatively to PLD, due to the narrow working pressure range of the high voltage abnormal discharge. Recently, Chandra and Manoharan [19] have investigated the fabrication of PTFE lms by PED under argon as the background gas. Thin lms have been obtained under a constant gas pressure of 666.61 103 Pa, room temperature, and constant discharge voltage. In this study, we report on the preparation of PTFE by channelspark PED technique on glass and silicon substrates. PTFE thin lms have been deposited under a wide range of process conditions, namely, background gas, temperature, pressure, and discharge voltage. The chemical composition and wetting properties of the deposited lms are investigated. 2. Experimental details The Polytetrauoroethylene thin lms were deposited in a pulsed electron ablation system, PEBS-20 model with pulse duration of 100 ns (Neocera, Inc.). A simplied schematic illustrating the main

Corresponding author. Tel.: + 1 705 675 1151; fax: + 1 705 675 4862. E-mail address: rhenda@laurentian.ca (R. Henda).

0040-6090/$ see front matter. Crown Copyright 2011 Published by Elsevier B.V. All rights reserved. doi:10.1016/j.tsf.2011.09.035

1886

R. Henda et al. / Thin Solid Films 520 (2012) 18851889

features of the PED system is shown in Fig. 1. The pulsed electron beam source based on channel-spark discharge geometry consists of a trigger, a hollow cathode, and a dielectric capillary tube. A two-inch diameter PTFE rod (Small Parts, Inc.) was cut into inch thick targets, and glass microscope slides (Labcraft) were cut into inch2 substrates. Both substrate and target were sonically cleaned at 50 C prior to deposition. The substrate was further rinsed with acetone followed with methanol, while the target was cleaned in methanol. The electron pulse energy was varied from 10 to16 kV with xed pulse frequency of 1 Hz and a constant number of 600 electron pulses throughout the experiments. Prior to lm deposition experiments the deposition chamber was evacuated to 6.6 104 Pa before admitting the background gas into the chamber at a specic pressure for deposition. The gas pressure was varied from 133.32103 Pa to 799.93 103 Pa. The distance between the target and electron gun ceramic tube tip was kept at 15 mm whereas the target to substrate distance was kept at 80 mm. Prior to deposition the target was pre-ablated by the electron beam over 1200 pulses at 1 Hz and 10 kV. The operating conditions of PED system were varied to determine their effect on the wettability of the lm produced. Subsequently, the deposited lms were characterized using Attenuated Total Reection Fourier Transform Infrared (ATR-FTIR) spectroscopy. The spectra were collected using Hyperion infrared microscope (Bruker Optics) with a liquid nitrogen cooled mercury cadmium telluride detector and an ATR objective equipped with a germanium crystal. Each spectrum is the result of 100 scans collected at a resolution of 4 cm 1. All spectra were collected with an air background and were corrected for CO2 and H2O with the atmospheric compensation function of the software. The PTFE lm spectra are presented after subtraction of the spectrum for a bare glass substrate. Static angle goniometry (Model PG2, Fibro Systems AB, Sweden) was used to assess lm hydrophobicity by measuring the contact angle between a drop of de-ionized water and the lm surface under various process conditions. Single water drops were photographed using the goniometer built-in digital camera. Water droplets of volume around 4 L were used in these measurements. The 2 m 2 m atomic force microscopy (AFM) images were taken with a Nanoscope IIIa scanning probe microscope (Veeco, USA) controller operated in tapping mode. Veeco TESPA tapping mode tips with resonant frequency and force constant of approximately 320 kHz and 42 N/m were used. The E scanner head with a maximum scan range of 25 m was used with a scan rate of 0.51.0 Hz. The raw data were attened using the instrument software and all image height scales were set to 25 nm. Film thickness and roughness have been obtained via reectance spectroscopy in the visible range (M-Probe series, Semiconfsoft,

USA), where experimental data were tted by means of a modied MarquardtLevenberg minimization using the reectometer software. 3. Results and discussion The chemical composition of the deposited thin lms was characterized by ATR-FTIR. Fig. 2.a shows the FTIR spectrum of the PTFE target material. This spectrum has two strong vibrational bands at approximately 1154 cm1 and 1210 cm1 that are ascribable to the -CF2- symmetric and asymmetric stretching vibrations of the PTFE molecule [2022]. A third weaker peak assigned to the -CF2- wagging of the PTFE molecule is also observed at approximately 644 cm1. Fig. 2.b shows the FTIR spectra for three different spots on PTFE lms deposited on a glass substrate. The negative peaks at approximately 953 cm1 and 776 cm1 are due to the underlying substrate. The fact that the intensity of the substrate bands is negative upon subtraction of the substrate conrms the presence of a thin lm at the surface. Upon comparison to the PTFE target material, the deposited lm also has two absorption bands at approximately 1154 cm1 and 1210 cm1. This conrms that PTFE is present on the glass surface and that, overall, its chemical structure is similar to PTFE target material. A new band was also observed at approximately 1259 cm1. This peak has previously been attributed to overlapping CF, CF2 and CF3 vibrational modes and may be due to the presence of cross-linked or unsaturated species [21, 22]. The intensity of these bands is very low which limits the detection of the -CF2- wagging peak at 644 cm1, because it is by nature a weak absorber. This low intensity coupled with the presence of strong substrate bands indicates that the coating is much thinner than the overall sampling depth of the instrument (1 m). The FTIR spectra for all three areas on the sample are similar, although the intensities do vary slightly. This conrms that PTFE is uniformly distributed across the surface with slight variations in thickness from spot to spot. Reectance spectroscopy ndings show lms deposited on glass to have a uniform thickness within the range 351 1 nm under various process conditions and background gas, except at 300 where the thickness is about 344 nm. Film roughness, dened as peak-to-valley distance, of 51 1 nm and 60 1 nm has been reported for lms deposited under argon and nitrogen, respectively, except at 300C (argon) where roughness is about 33 nm. No appreciable differences between lm spectra have been noticed, an indication (at least qualitative) that deposited lms have similar composition. Fig. 3 shows AFM topographic images of the deposited lms on glass under various process conditions. The

Fig. 1. Simplied schematic diagram of PED system.

Fig. 2. ATR-FTIR spectra of (a) PTFE target material and (b) PTFE thin lm (T = RT, P = 133.32 103 Pa) on glass substrate at three different positions on the lm.

R. Henda et al. / Thin Solid Films 520 (2012) 18851889

1887

25 nm

25 nm 0 nm

(a)

(b)

25 n nm 0 nm

(c)

(d)

(e)

(f)

25 nm 0 nm

(g)

(h)

(i)

Fig. 3. AFM topographs of PTFE lms on glass under argon as a function of pressure: a) 133.32 103 Pa, b) 266.64 103 Pa, c) 399.96 103 Pa; discharge voltage: d) 12 kV, e) 14 kV, f) 16 kV; and temperature: g) RT, h) 100 C, i) 300 C.

0 nm

lms are composed of clusters or particulates of pyramidal shape and of varying size. Overall, particulate size seems to increase with pressure, to decrease then increase with discharge power, and to increase with temperature as depicted in Fig. 3.ac, df, and gi, respectively. The measured contact angle for water is about 110 for bulk PTFE, whereas the contact angle for bare glass, as depicted in Fig. 4.a, and silicon substrates is around 32 and 43, respectively. The wetting properties of the deposited lms can be qualitatively assessed through the digital snapshots of a drop of water spread on the substrate-supported lm. An example of snapshots for lms deposited at room temperature (RT) on a glass substrate is shown in Fig. 4. PTFE lms were obtained under argon (Fig. 4.bd) and nitrogen (Fig. 4.eg) at a constant discharge voltage and for different gas pressures. Fig. 5 illustrates contact angle data as a function of deposition pressure and background gas. The data indicate an overall decrease (increase) in contact angle (in solidliquid area) as the pressure increases for both gasses. For any given pressure the contact angle is higher for

lms deposited under nitrogen as such lms have a higher roughness, according to reectance data (see above). Based on AFM measurements (Fig. 3.ac) the likely cause for the decrease in contact angle as the pressure is increased is a decrease in air entrapment between the water drop, as it sits on particulates of large size and lower surface density, and the deposited lm [23]. This is in accordance with the predictions of the model of Cassie and Baxter [24]. The latter have derived the following equation to account for the contact angle of water on a non-smooth surface cosa f1 cosf2 1

where a is the apparent/observed contact angle, is the contact angle of the smooth surface, fl is the area fraction of solidliquid interface and f2 is the area fraction of liquidair interface both relatively to the surface of the smooth surface. The parameters f1 and f2 depend on and the size of the nano-particulates. While f1 and f2 do not necessarily

0 nm

n 25 nm 0 nm

25 nm 0 nm n

25 n 0 nm nm

25 nm 0 nm

25 nm

1888

R. Henda et al. / Thin Solid Films 520 (2012) 18851889

1 mm

a)

b)

c)

d)
Fig. 6. Contact angle versus discharge voltage for lms obtained under argon and nitrogen backgrounds. Conditions are: RT; 266.64 103 Pa (argon), 533.28 103 Pa (nitrogen).

e)

f)

g)

Fig. 4. Snapshots of water drops on deposited PTFE lms at RT. a) bare glass substrate, bd) Argon at 133.32 103 Pa, 266.64 103 Pa, 399.96 103 Pa; 14 kV, eg) Nitrogen at 133.32 103 Pa, 533.28 103 Pa, 799.93 103 Pa; 12 kV.

add up to unity, the tally is close to one. For the sake of simplicity, let's assume this to be true and Eq. (1) then becomes cosa f1 cos f1 1: 2

than 300 C, but the surface density of the particulates seems to be larger at 300 C than at 100 C. Also, both thickness and roughness of the lms obtained at 300 C are lower than those of lms deposited at 100 C and RT as per reectometry measurements. Combining both particulates features, it seems that the previous reasoning to explain the dependency of the contact angle on pressure and discharge voltage holds as well for deposition temperature. The reason why lms deposited on silicon are less wettable than those deposited on glass is not yet clear except that lm thickness measurements reveal much thinner lms on silicon (thickness of about 20 nm) than on glass.

4. Conclusions Accordingly, as the solidliquid contact area is increased less air is entrapped between the lm surface and the drop of water, and the observed contact angle will decrease for a given . Overall, for both gasses the measured contact angle increases at rst, albeit the increase is less pronounced for argon, as the discharge voltage increases then it decreases in magnitude, as shown in Fig. 6. At both ends of the discharge voltage interval the deposited lms consist of larger size particulates and lower surface density than at 14 kV as evidenced from AFM images (Fig. 3.df). Finally, the effect of the deposition temperature on lm wettability has been assessed as depicted in Fig. 7. For PTFE lms deposited on glass and silicon (100) substrates, the contact angle decreases as the deposition temperature of the lms increases. From an inspection of AFM images (Fig. 3.gi) lms deposited at higher temperatures than RT present pinholes. The size of particulates is smallest at RT, while at higher temperatures the particulates size is somewhat higher at 100 C In summary, we have applied pulsed electron deposition for the preparation of PTFE thin lms. The lms have been deposited under various process conditions, namely, temperature, pressure, discharge voltage, and background gas. ATR-FTIR has revealed an overall correspondence between the target material and deposited thin lms. Contact angle measurements show declining lm hydrophobicity with increasing deposition temperature and gas pressure, while the contact angle increases with increasing discharge voltage then decreases beyond a threshold value of the voltage. The changes in the contact angle with the various parameters seem to correlate well with the size and surface density of PTFE particulates deposited on the substrate.

Fig. 5. Contact angle versus deposition pressure for lms obtained under argon and nitrogen backgrounds. Same conditions as in Fig. 4.

Fig. 7. Contact angle versus deposition temperature for lms obtained under argon on glass and silicon substrates. Conditions are: 14 kV and 266.64 103 Pa (glass), 12 kV and 533.28 103 Pa (silicon).

R. Henda et al. / Thin Solid Films 520 (2012) 18851889

1889

Acknowledgments The principal author (R.H.) wishes to thank the Natural Sciences and Engineering Research Council of Canada (NSERC) for the nancial support. G.W. gratefully acknowledges NSERC for the Undergraduate Summer Research Award. References
[1] R. Kressman, G.M. Sessler, P. Gnter, IEEE Trans. Dielectr. Electr. Insul. 3 (1996) 607. [2] M.P. Cuminal, F. Touhari, M. Brunet, G. Cohen-Solal, Mol. Cryst. Liq. Cryst. 333 (1999) 181. [3] R.W. Jaszewski, H. Schift, B. Schnyder, A. Schneuwly, P. Groning, Appl. Surf. Sci. 143 (1999) 301. [4] P. Iuff, M. White, Vacuum 18 (1968) 437. [5] D.J. Morrison, T. Robertson, Thin Solid Films 15 (1973) 87. [6] L. Holland, H. Biederman, S.M. Ojha, Thin Solid Films 35 (1976) L19. [7] L. Wang, H. Li, J. He, X. He, W. Li, Y. Wang, H. Li, Mater. Lett. 33 (1997) 77. [8] G.B. Blanchet, C.R. Fincher Jr., Adv. Matter. 6 (1994) 881. [9] M.G. Norton, W. Jiang, J.T. Dickinson, K.W. Hipps, Appl. Surf. Sci. 9698 (1996) 617.

[10] Q. Luo, X. Chen, Z. Liu, Z. Sun, N. Ming, Appl. Surf. Sci. 108 (1997) 89. [11] S.T. Li, E. Arenholz, J. Heitz, D. Buerle, Appl. Surf. Sci. 125 (1998) 17. [12] R. Schwdiauer, J. Heitz, E. Arenholz, S. Bauer-Gogonea, S. Bauer, W. Wirges, J. Polym. Sci. B. 37 (1999) 2115. [13] N. Huber, J. Heitz, D. Buerle, R. Schwdiauer, S. Bauer, H. Niino, A. Yabe, Appl. Phys. A. 72 (2001) 581. [14] T. Smausz, B. Hopp, N. Kresz, J. Phys. D: Appl. Phys. 35 (2002) 1859. [15] W.A. Daoud, J.H. Xin, Y.H. Zhang, C.L. Mak, Thin Solid Films 515 (2006) 835. [16] H.Y. Kwong, M.H. Wong, Y.W. Wong, K.H. Wong, Appl. Surf. Sci. 253 (2007) 8841. [17] V. Matias, J. Talvacchio, X. Xi, Z. Han, H.W. Neumuller (Eds.), Frontiers in Superconducting MaterialsNew Materials and Applications, Mat. Res. Soc. Symp. Proc. EXS-3, 2004, EE1.8.1. [18] A. Goyal (Ed.), Second-Generation HTS Conductors, Springer US, New York NY, 2005, p. 109. [19] V. Chandra, S.S. Manoharan, Appl. Surf. Sci. 254 (2008) 4063. [20] S. Wang, J. Li, J. Suo, T. Luo, Appl. Surf. Sci. 256 (2010) 2293. [21] G.A. Hishmeh, T.L. Barr, A. Sklyarov, S. Hardcastle, J. Vac. Sci. Technol. A14 (1996) 1330. [22] Y. Zhang, G.H. Yang, E.T. Kang, K.G. Neoh, W. Huang, A.C.H. Huan, D.M.Y. Lai, Surf. Interface Anal. 34 (2002) 10. [23] B. Bhushan, Y.C. Jung, K. Koch, Philos. Trans. R. Soc. A367 (2009) 1631. [24] A.B.D. Cassie, S. Baxter, Trans. Faraday Soc. 40 (1944) 546.