adhesion improvement of PDMS

Transcription

1 Remote atmospheric pressure DC glow discharge treatment for adhesion improvement of PDMS Nathalie De Geyter 1 *, Rino Morent 1, Tinneke Jacobs 1, Fabrice Axisa 2, Léon Gengembre 3, Christophe Leys 1, Jan Vanfleteren 2 and Edmond Payen 3 1 Department of Applied Physics, Research Unit Plasma Technology, Faculty of Engineering, Ghent University, Jozef Plateaustraat 22, 9000 Ghent, Belgium Nathalie.DeGeyter@ugent.be, Fax : Department of Electronics and Information Systems, CMST, Technologiepark 914, Grote Steenweg Noord, 9052 Ghent, Belgium 3 Unité de Catalyse et Chimie du Solide, UMR CNRS 8181, Université des Sciences et Technologies de Lille, Bât. C3, Cité Scientifique, Villeneuve d Ascq, France Keywords. adhesion, non-thermal plasma, remote plasma processes, silastic, surface modification Summary. In this paper, a remote DC glow discharge at atmospheric pressure is employed for surface modification of polydimethylsiloxane (PDMS) aimed at improvement of its adhesive properties. The effects of the discharge on the surface properties of PDMS are probed using contact angle measurements, X-ray Photoelectron Spectroscopy (XPS) and T-peel tests. Results show that the DC glow discharge transforms the initially hydrophobic PDMS surface into a hydrophilic one due to the incorporation of silanol groups at the expense of methyl groups. Moreover, T-peel tests confirm that the remote DC glow plasma is able to remarkably enhance the adhesion between 2 PDMS layers, but only when both layers are plasma-treated. 1

2 1. Introduction It is believed that in the near future citizens will carry along more and more electronic systems near or even inside the body. Ideally, these electronics should be non-noticeable to the user: it is clear that in this perspective the electronic systems should be soft, compact, light-weighted and preferably take the shape of the object in which they are integrated. [1] Therefore, there is currently a strong tendency to replace common rigid electronic assemblies by mechanically flexible or even stretchable equivalents. This emerging technology can be applied for biomedical electronics (implantable devices and electronics on skin) and for wearable electronics in clothes. [2] Stretchable electronics consist of a moulded matrix of polydimethyldisiloxane (PDMS) with embedded electronic components connected by electronic interconnections. Since little or no stretchable electronic components are available, it is vital to engineer electronic interconnections which are stretchable. Today, metals like cupper, nickel and gold are the best option to realize the interconnections with high performance and low cost. [2] However, metals are not intrinsically stretchable, therefore, a suitable design such as a meander-shaped horseshoe structure is necessary. This structure can be reproducibly stretched and released many times without breaking. [2] Among the identified stretchable substrate materials appropriate for the fabrication of implantable electronic systems, polydimethylsiloxane (PDMS) is one of the most promising due to its mechanical properties (compliance, soft) and its good biocompatibility. [3] In a first step of the production process, metallic meander-shaped interconnections are electroplated on a cupper foil. The components are then assembled and connected to the interconnections by gluing or soldering. In a next step, the sample is overmoulded with viscous PDMS and thermally cured. Then, the cupper film is removed resulting in a thin film of PDMS with the interconnections and components at the surface. Finally, the electronic structures are encapsulated by overmoulding the sample with viscous PDMS. [1-2] This technology has been developed in the frame of the Bioflex programme ( IWT SBO-Bioflex 04101). [1-3] The metals inside the electronic circuit should not leak out in order to obtain a highly biocompatible system. Therefore, an excellent adhesion between the two PDMS layers is of great importance. However, PDMS has a very low surface energy and as a result poor bonding properties. [4] Traditional 2

3 surface treatments to improve bonding characteristics often include liquid chemical processes. In comparison with these methods, dry plasma treatment forms an economical and environmentally friendly alternative. [5] Non-thermal plasma treatment of polymers at low pressure is already a well established technology. However, the necessity of expensive vacuum systems is the biggest shortcoming of these processes in industrial applications besides the limitation to the batch system processes. [6] Therefore, considerable efforts are made in developing atmospheric pressure non-thermal plasmas. [7-9] Today, the so-called corona treaters based on AC dielectric barrier discharges are widely used for plasma processing of various materials. [5] However, in this paper, a remote atmospheric pressure DC glow discharge that operates in ambient air will be used to modify the surface properties of PDMS. [5,7] This plasma system, based on the use of steady-state direct current discharges is developed to avoid thermal effects that could cause degradation of heat-sensitive materials when placed in direct contact with the plasma. Moreover, the design of the reactor ensures scalability to industrial roll widths. Moreover, an additional advantage of the proposed remote plasma source is the absence of limitations on the thickness of the treated substrates as the substrate is not placed between the electrodes. The effects of the remote DC glow discharge on the PDMS films will be studied using contact angle measurements and X-ray Photoelectron spectroscopy (XPS). This work also investigates whether the changes induced by plasma treatment contribute to an improvement in the adhesion properties of the PDMS surfaces. 2. Experimental Part a. Materials 250 µm thick PDMS sheets are manufactured from a biomedical-grade silicone rubber kit (Silastic - MDX Dow Corning) consisting of an elastomeric component and a curing agent. The elastomer component consist of a dimethylsiloxane polymer, a reinforcing silica and a platinum catalyst, while the curing agent component consists of a dimethylsiloxane polymer, an inhibitor and a siloxane crosslinker. The elastomeric component and the curing agent are thoroughly blended in a 10:1 ratio after which the air bubbles are removed by vacuum. Afterwards, the silicone mix is placed into a mould (to produce 3

4 250 µm thick sheets) and is then cured in air by placing the mould in an oven of 150 during 30 minutes. PDMS consists of an inorganic backbone of alternating silicon and oxygen atoms and methyl groups are attached to the silicon atoms forming the repeating unit in the polymer. [3] b. Remote DC glow set-up The experimental set-up used for the remote DC glow plasma treatment at atmospheric pressure is similar to that reported in [5,7] and is shown in Figure 1. The atmospheric pressure plasma source consists of an aluminium plate anode and a single row of 28 cathode electrode elements oriented in parallel to the treated surface. Each of these electrode elements is a stainless steel pin, ballasted with a resistor of 1.5 MΩ (Caddock MX440). The aluminium plate anode is connected to a DC high voltage power supply with output voltage of 18.5 kv, while the cathode electrode elements are connected to earth. The electrode gap is 10 mm and the discharge power is kept constant at W, leading to a surface power density of approximately 9 W/cm 2. The non-thermal plasma is generated between the two electrodes and a fast airflow of 40 m/s produced by a fan (Ventomatic CMT31M) transports the plasma-produced reactive particles towards the PDMS sample to be treated. To diminish any losses in reactive species during their transportation to the treated surface, the distance between the output of the plasma source and the PDMS sample is only 5 mm. The sample is mounted on a rotating drum of 0.40 m in diameter in order to simulate in-line processing at variable line speeds. During the experiments, the speed of the rotating drum is kept constant at 0.74 m/s. The results presented below are for so-called cyclical treatment: this means that the sample passes once or a number of times through the exit region of the plasma source, where it is exposed to the flowing afterglow. The DC glow treatment time, shown in this paper, is the effective exposure time to the non-thermal plasma. c. Contact angle measurements Contact angle measurement is the ideal method to characterize the surface wettability of polymers. Contact angles are obtained on the PDMS surfaces using a commercial contact angle goniometer (EasyDrop Krüss). This system is equipped with a software operated high-precision liquid dispenser 4

5 to precisely control the drop size of the used liquid. The drop image is then stored, via a monochrome interline CCD video camera, using PC-based acquisition and data processing. Using the computer software provided with the instrument, measurement of the static contact angles is fully automated. Distilled water is used as working liquid and the volume of the water drops is maintained at 2 µl. The values of the contact angles, shown in this paper, are the average of at least 8 measured values and the standard deviation on the average contact angles is smaller than 2 %. d. XPS analysis The chemical composition of the PDMS films is investigated by X-ray Photoelectron Spectroscopy (XPS), which is performed in a VG Escalab 220 XL system, using non-monochromatic Mg K α - radiation (hν= ev) operated at 15 kv and 20 ma. The pressure in the analysing chamber is maintained at 10-7 Pa or lower during analysis and the size of the analyzed area is 8 mm x 8 mm. Spectra are acquired at a take-off angle of 90 relatively to the sample surface, leading to a sampling depth of approximately 10.8 nm. [10,11] The high-resolution spectra are taken in the constant analyser energy mode with a 40 ev pass energy and the value of ev of the C1s core level is used for calibration of the energy scale. e. Adhesion test: T-peel test To test the adhesion between 2 PDMS layers, a T-peel test is performed according to the international standard ISO [12] PDMS samples are carefully cut into specimens with a width equal to 10 mm and a length of 60 mm. In a next step, 2 PDMS films are bonded together without any adhesive over a length of 50 mm. The 2 unbounded ends of the flexible films are then bent in opposite directions until each end is perpendicular to the bonded assembly to form a T-shaped specimen. After this, the 2 unbounded ends are mounted into the jaws of an Instron 5543 mechanical tester. The mechanical tester is set in motion with a separation rate of 1 mm/min and the applied force versus the distance of the grip separation is recorded. 5

6 3. Results a. Contact angle measurements The modification of the PDMS surfaces by the remote atmospheric pressure DC glow discharge is first studied using contact angle measurements, since these measurements represent the easiest and quickest method for examining surfaces properties. Figure 2 shows the static water contact angles on the PDMS samples as a function of plasma treatment time. The untreated PDMS surface exhibits a hydrophobic nature characterized by the high value of the water contact angle (107.6 ). As shown in Figure 2, remote DC glow discharge treatment is effective in rendering the silicone samples more hydrophilic, characterized by a reduction in contact angle with increasing plasma treatment time. At plasma treatment times longer than approximately 22 seconds, no further changes in contact angles are observed, which means that a saturation level (14 ) is reached. Different other authors have reported similar decreases in water contact angles using various plasma discharges such as corona discharges, RF discharges, MW discharges, [13-16] In a previous study by Morent et al. [3], PDMS samples, identical to the ones in this paper, have been plasma-treated using a dielectric barrier discharge (DBD) operated at medium pressure (5.0 kpa). This medium pressure DBD treatment also decreases the water contact angle values on PDMS to approximately 14, however, this saturation value is already obtained at a plasma treatment time of 3 seconds despite the lower surface power density (0.17 W/cm 2 ) of the DBD compared to the DC glow discharge. This behaviour can be explained by the fact that the DBD treatment is an active plasma treatment, meaning that the PDMS sample is placed between the electrodes leading to a high concentration of active species near the sample surface. In contrast, the DC glow treatment is a remote plasma treatment: the PDMS sample is located outside the plasma region, but passes in the gas stream that runs through the plasma zone. This gas stream is loaded with radicals and other active species, however, the concentration of these active species decreases with increasing distance to the plasma region due to collision losses. Compared to the active DBD treatment, less plasma species are present near the sample surface, leading to a higher plasma energy density necessary to obtain the lowest water contact angle value. 6

7 b. XPS measurements XPS can be used to characterize the chemical changes that take place on the surface of the PDMS samples due to the DC glow plasma treatment. XPS survey scans obtained with a take-off angle equal to 90 can provide information on the elemental composition of the top 10.8 nm of the PDMS samples. Table 1 shows the elemental composition of an untreated PDMS sample and a PDMS film after 11.0 and 24.4 seconds of plasma treatment. As seen in Table 1, the carbon content on the PDMS samples decreases, whereas the oxygen content increases with increasing plasma treatment time. The silicon content remains essentially constant between 30 and 33 at%. Further information on the chemical groups present in the top 10.8 nm layer of the PDMS sheets can be obtained by resolving the Si2p peak according to the method described by O Hare et al. [17] The full width at half maximum (FWHM) of all the fitted Si2p components is kept constant at ± 1.6 ev. Figure 3 shows the resolved Si2p peaks of (A) the untreated PDMS sample, (B) the PDMS sample after 11.0 seconds of DC glow discharge treatment and (C) the PDMS film after 24.4 seconds of plasma treatment. The Si2p peak of the untreated PDMS sample can be fitted with 2 peaks: a peak at ev (C1) which can be associated with silicon bound to two oxygen atoms (unoxidized PDMS) and a peak at ev (C3) due to silicon bound to four oxygen atoms (SiO 2 ). This latter peak is probably the result of the presence of reinforcing silica in the elastomer component of the rubber kit. After 11.0 seconds of plasma treatment, one can see that the peak at ev (C1) decreases, while the peak at ev (C3) increases. Moreover, a new peak at ev (C2) appears, which can be associated with silicon bound to three oxygen atoms (partially oxidized PDMS). After 24.4 seconds of remote DC glow treatment, the peak at ev (C1) decreases even further, while the peak at ev (C3) further increases. Moreover, the peak at ev (C2) disappears after 24.4 seconds of remote DC glow discharge treatment. The concentration of the different silicon bounds before and after plasma treatment can be seen in Table 2. After 11.0 seconds of plasma treatment, the number of oxygen atoms bounded to silicon increases from two to three and even four. At higher plasma treatment time (24.4 seconds), the oxidation process proceeds resulting in the presence of a large amount of silicon atoms bounded to four oxygen atoms. The above results suggest the formation of linkages between Si and O atoms due to the remote DC glow discharge treatment in the top 10.8 nm of the PDMS samples. It is believed that the plasma treatment 7

8 leads to the incorporation of silanol (Si-OH) groups at the expense of methyl (CH 3 ) groups. [18] These silanol groups make the exposed PDMS surface highly hydrophilic (as observed in section 3.a) due to their polar nature. c. Mechanical peel testing The adhesive behaviour of the plasma-treated PDMS surfaces can be tested using a T-peel test, as described in detail in section 2.e. Different test specimens are prepared within 30 seconds after plasma treatment: a sample where 2 untreated PDMS samples are bounded (sample A), a sample where 1 untreated and 1 plasma-treated (treatment time = 11.0 s) PDMS sample are bounded (sample B), a sample where 1 untreated and 1 plasma-treated (treatment time = 24.4 s) PDMS sample are bounded (sample C), a sample where both PDMS sheets are plasma-treated with a treatment time equal to 11.0 s (sample D) and 24.4 s (sample E). Figure 4 shows the load-displacement curves of the test specimens D and E. The curves of the assemblies A, B and C are not shown in Figure 4 since the load remains at approximately 0 N with increasing displacement, suggesting that there is no adhesion. However, Figure 4 shows that for the test specimens D en E, the load gradually increases with increasing displacement, however, at a certain point, the load remains constant with increasing displacement, suggesting that peeling of the 2 PDMS layers occurs. This constant load F (N) can be used to calculate the peel strength (N/m) using the following equation: [19] peel strength = 2F/W (1) where W is the specimen width in meter, which is equal to m in this paper. The calculated peel strengths are shown in Table 3 for the samples A to E, showing that no adhesion enhancement is observed when only one PDMS layer is treated with the remote DC glow discharge. However, when both PDMS sheets are plasma-treated, a high peel strength can be observed and this without the use of any adhesive. As stated before, the principal chemical changes induced by the remote DC glow discharge is the replacement of pendant CH 3 groups by silanol (Si-OH) groups. When 2 plasma-treated PDMS surfaces are brought into contact, the silanol groups condense with those on another surface resulting in the formation of Si-O-Si bonds. These covalent bonds form the basis of a tight, irreversible seal between 2 PDMS layers. When 8

9 only 1 PDMS films is plasma-treated the formation of these Si-O-Si bonds is impossible, explaining the poor adhesion. [18] 4. Conclusion This paper studies the surface treatment of PDMS by a remote DC glow discharge aimed to improve the adhesive properties of the material. Several techniques have been employed to investigate the effects of the discharge on the PDMS surfaces. Results reveal that the DC glow discharge strongly enhances the PDMS surface wettability: the contact angle can be decreased from for the untreated sample to 14.4 after plasma treatment. From XPS results, it has emerged that this wettability increase can be attributed to the incorporation of silanol groups at the expense of methyl groups. Moreover, T-peel tests have shown that the employed glow discharge significantly enhances the adhesion between 2 PDMS surfaces, but only when both PDMS layers are plasma-treated. When 2 plasma-treated PDMS surfaces are brought into contact, the silanol groups on one surface condense with the ones on the other surface resulting in the formation of strong Si-O-Si bonds. Currently, it is investigated whether this adhesion enhancement is sufficient to prevent leakage of metals from stretchable electronics. 9

10 5. References [1] F. Axisa, D. Brosteaux, E. De Leersnijder, F. Bossuyt, M. Gonzalez, M. Vanden Bulcke, J. Vanfleteren, Proceedings of the 16th European Microelectronics and Packaging Conference (EMPC) 2007, 691. [2] D. Brosteaux, F. Axisa, M. Gonzalez, J. Vanfleteren, IEEE Electr. Device L. 2007, 28, 552. [3] R. Morent, N. De Geyter, F. Axisa, N. De Smet, L. Gengembre, E. De Leersnyder, C. Leys, J. Vanfleteren, M. Rymarczyk-Machal, E. Schacht, E. Payen, J. Phys. D: Appl. Phys. 2007, 40, [4] J. L. Fritz, M. J. Owen, J. Adhes. 1995, 54, 33. [5] E. Temmerman, C. Leys, Surf. Coat. Technol. 2005, 200, 686. [6] T. P. Kasih, S. Kuroda, H. Kubota, Plasma Process. Polym. 2007, 4, 648. [7] E. Temmerman, Y. Akishev, N. Thrushkin, C. Leys, J. Verschuren, J. Phys.D: Appl. Phys. 2005, 38, 505. [8] G. Borcia, N. M. D. Brown, J. Phys.D: Appl. Phys. 2007, 40, [9] G. Borcia, C.A. Anderson, N. M. D Brown, Plasma Sources Sci. Technol. 2005, 14, 259. [10] H. Hillborg, U. W. Gedde, IEEE Trns. Dielectr. Electr. Insul. 1999, 6, 703. [11] S. Tamuna, C. J. Powell, D.R. Penn, Surf. Interface Anal. 1994, 21, 165. [12] International Standard ISO Adhesives-T-peel test for flexible-to-flexible bonded assemblies [13] Y. Zhu, K. Haji, M. Otsubo, C. Honda, IEEE Trns. Plasma Science 2006, 34, [14] H. Hillborg, J. F. Ankner, U. W. Gedde, G. D. Smith, H. K. Yasuda, K. Wikström, Polymer 2000, 41, [15] H. Hillborg, M. Sandelin, U. W. Gedde, Polymer 2001, 42, [16] S. Befahy, S. Yunus, V. Burguet, J.-S. Heine, M. Troosters, P. Bertrand, J. Adhes. 2008, 84, 231. [17] L.-A. O Hare, B. Parbhoo, S. R Leadly, Surf. Interface Anal. 2004, 36, [18] S. Bhattarcharya, A. Datta, J.M. Berg, S. Gangopadhyay, J. Microelectromech. S. 2005, 14, 590. [19] M. J. Shenton, M. C. Lovell-Hoare, G.C. Stevens, J. Phys. D: Appl. Phys. 2001, 34,

11 6. Figures Figure 1. Experimental set-up of the remote DC glow discharge (C=cathode pins, A=anode plate) 11

12 Figure 2. Evolution of the contact angle on PDMS as a function of plasma treatment time 12

13 Figure 3. Si2p peaks of (A) the untreated sample, (B) the PDMS film after 11.0 seconds of plasma treatment and (C) the PDMS film after 24.4 seconds of plasma treatment 13

14 Figure 4. Load-displacement plots for T-peel assemblies D and E 14

15 7. Tables Table 1. Atomic composition of untreated and plasma-treated PDMS films Treatment time (s) O (at%) C (at%) Si (at%) Table 2. Concentration of the different silicon bonds on PDMS before and after plasma treatment Treatment time (s) (%) (%) (%) Table 3. Calculated peel strengths of different T-peel assemblies Sample Peel strength (N/m) A 0 B 0 C 0 D 613 E