147 Original scientific paper  MIDEM Society Thermoplastic - PDMS polymer covalent bonding for microfluidic applications Borut Pečar, Matej Možek and Danilo Vrtačnik University of Ljubljana, Faculty of Electrical Engineering, Laboratory of Microsensor Structures and Electronics, Ljubljana, Slovenia Abstract: Two room-temperature bonding processes for thermoplastic - PDMS polymer covalent bonding based on the organic substrate surface functionalization by means of organofunctional silanes APTES and amine-PDMS linker were developed and applied. The efficiency of covalent bonding was evaluated by measuring water contact angles on oxygen plasma pretreated surfaces and by measuring burst pressure on fabricated test devices. Developed amine-PDMS linker bonding process resulted in bond strength of 5 bar and 2 bar on continuous pressure of air and water respectively, while water initiated the hydrolysis of covalent bonds established via the modified APTES bonding process. Both bonding processes were applied on piezoelectric micropumps where glass substrate was replaced by thermoplastic substrate. Micropumps employing amine-PDMS linker exhibit no deterioration in their performance after eight weeks of continuous operation. Keywords: PDMS; WCA; APTES; thermoplastics; covalent bonding; micropump Kovalentno spajanje termoplasta s PDMS polimerom za mikrofluidne aplikacije Izvleček: Raziskali, razvili in vpeljali smo dva nizko temperaturna postopka kovalentnega spajanja termoplasta in PDMS polimera, ki temeljita na funkcionalizaciji organske površine preko organofunkcionalnega silana APTESa in amino-PDMS povezovalca. Učinkovitost površinske aktivacije, ki je ključna za učinkovit kovalenten spoj, smo ovrednotili z merjenjem omakalnih kotov vodnih kapelj na površini vzorcev pred in po aktivaciji površin v kisikovi plazmi. Za ovrednotenje kvalitete spoja smo na namensko izdelanih testnih čipih izvedli tlačne in porušitvene teste. Amino-PDMS povezovalec je zagotovil obstojnost spoja ob stiku z vodo, medtem ko so vezi vzpostavljene preko APTESa kljub dodatni toplotni obdelavi po nanosu in modifikaciji parametrov plazemske aktivacije površin razpadle. Razvita postopka spajanja smo vpeljali v proces izdelave piezoelektričnih mikročrpalk. Piezoelektrične mikročrpalke izdelane s postopkom amine-PDMS povezovalca po osmih tednih neprekinjenega delovanja ne izkazujejo upada pretočne zmogljivosti. Ključne besede: PDMS; WCA; APTES; termoplast; kovalentno spajanje; mikročrpalka * Corresponding Author’s e-mail: borut.pecar@fe.uni-lj.si Journal of Microelectronics, Electronic Components and Materials Vol. 47, No. 3(2017), 147 – 154 1 Introduction Plastics are indispensable in mass production of mi- crofluidic devices due to their robustness, light weight, optical transparency, simplicity of molding and cost efficiency [1]. Plastics rigidity enables a variety of reli- able external interface options, such as manifold inte- gration, direct barbed tubing connections, and gasket connectors [2]. In addition, thermoplastic (TP)–polydi- methylsiloxane (PDMS) assemblies have a number of advantages over homogeneous assemblies. The com- bined surface properties of the two materials, for ex- ample, could provide an optimal environment for con- ducting cell-based research necessitating precise fluid control for targeted cell or biomolecule immobilization [3]. Many strategies for plastic–PDMS bonding have been previously reported, such as sol–gel coating approach, chemical gluing approach and organofunctional si- lanes approach [4, 5]. First approach requires multiple coating procedures as well as complex technology. Sec- ond approach creates chemically robust amine–epoxy bonds at the interface at room temperature, however, two silane-coupling reagents are required and both 148 B. Pečar et al; Informacije Midem, Vol. 47, No. 3(2017), 147 – 154 surfaces had to be oxidized prior to chemical modifica- tion. Third approach requires only one coupling agent. In this approach, the most widely used organofunc- tional silane is 3-amino propyltriethoxysilane (APTES), aminosilane frequently employed in covalent bonding of organic films to metal oxides [6]. However, TP-PDMS covalent bonds established via or- ganofunctional silanes are prone to degradation over prolonged period in aqueous environment which might limit the use in specific microfluidic applications. Few studies attempted to increase hydrolytic stability of APTES by mixing it with complex agents such as BTI- SPA, BTMSPA, BTESE [2] or GPTMS [7]. Hydrolytic resist- ance of APTES mixed with BTISPA improved to hydro- lytically stable bonds over a range of 0 to 15 pH when BTISPA was prevailing in the mixture. Authors attribut- ed increased hydrolytic resistance to the greater cross- link density for bis-silanes. APTES mixed with GPTMS yielded higher bond strength as compared to APTES, but did not improved hydrolytic stability [8]. Being aware of reported limited hydrolytic resistance of APTES [2, 7], we tried to introduce additional post- deposition heat treatment and modification of post- deposition plasma treatment that might overcome this disadvantage and provide the bond strength sufficient for specific application e. g. in the range of 0.5-0.7 bar. The advantages of employing APTES linking agent in- clude easy availability of the product and well estab- lished bonding process due to substantial popularity in microfluidic community. In this work, approach of employing two organofunc- tional silanes, APTES and poly [dimethyl siloxane-co-(3- aminopropyl) methyl siloxane (amine-PDMS linker) for bonding PDMS elastomer to TP substrates via methanol aligning medium was developed and applied. Amine- PDMS linker incorporates an amine functionality at one terminal and a segment of low molecular weight PDMS at the other, which might provide better hydrolytic bond stability [3]. Surface properties of TPs and PDMS were analyzed by measuring the water contact angles (WCA). The bond strength and bond hydrolytic stabil- ity were evaluated by delamination and burst pressure tests. Both bonding processes were further improved and applied on piezoelectric micropumps, where sup- porting bottom glass was replaced by TP substrate. 2 Experimental 2.1 Materials In the presence of amine in organofunctional ami- nosilane bonding process, thermoplastics undergo aminolysis followed by chain scission of the carbonyl backbone, forming a strong urethane bond. Therefore, not all thermoplastics, but only thermoplastics that can undergo aminolysis are suitable for the purpose. For TP substrates, optically transparent 2 mm thick Polycar- bonate (PC), Acrylonitrile butadiene styrene (ABS) and Poly methyl methacrylate (PMMA) from INEOS Styrolu- tion Group GmbH were employed. In all experiments, a PDMS Sylagard® 184 two-part kit consisting of a pre-polymer (base) and a cross-linker (curing agent) from Dow Corning Corporation mixed at a ratio 10:1 was applied. For surface functionalization of plastic substrates, a commercial solution of APTES (Sigma Aldrich) and amine- PDMS linker (Sigma Al- drich) were used. In bonding processes, methanol and DI of technical purity were applied. 2.2 Surface wettability measurements As argued by Garbassi et al. [9], the oxidation of the surface layer increases the concentration of hydroxyl groups which leads to the formation of strong intermo- lecular bonds. As the silanol groups are polar in nature, they make the exposed surface highly hydrophilic and this can be observed by measuring WCAs [10]. Those WCAs were found in direct correlation with bond strength [11]. For WCA determination a method was developed which included photographing of droplets on inves- Figure 1: Micrograph (camera Nikkon e990) of water droplet on raw PDMS surface (a) and on oxygen plasma treated PDMS surface (b). 149 tigated surfaces. The images of the droplet were ana- lyzed by computer software “ImageJ” and “ContactAn- gle” plugin from points marked along the droplet-air interface to calculate the contact angle at the droplet- surface interface. An example of WCA measurement on raw PDMS and on PDMS treated in oxygen plasma (20 sec, 0.8 mbar, 40W) is shown in Fig. 1. For all plasma treatments, ATTO Low Pressure Plasma Systems Diener electronic GmbH was employed. 2.3 Bonding process In order to overcome reported limited hydrolytic resist- ance of APTES [2, 7], additional post-deposition heat treatment and modification of post-deposition plasma treatment were introduced in the bonding process. The process flow for TP-PDMS sandwich covalent bonding is schematically shown in Fig. 2. First, TP substrates were cleaned in ultrasonic bath, followed by silylation of the surfaces through the use of organofunctional silanes. In order to achieve good adhesion, TPs were pre-activated in oxygen plasma, resulting in the hydro- philization of the TP surfaces. TP sheets were then immersed in the APTES (5% per volume) or coated with amine-PDMS linker and addi- tionally heated to 60 °C for 20 min. Such an additional post-deposition heating step was expected to increase the number of established urethane bonds, especially between TP and APTES where bonds are prone to hy- drolytic decay. Unlinked organofunctional silanes were washed away with isopropyl alcohol in ultrasonic bath. Both PDMS and functionalized TP substrate surfaces were again activated in the oxygen plasma. Post-depo- sition plasma treatment was prolonged to 1 min at in- creased pressure of 2 mbars and increased power of 50 W in order to increase the formation of Si–OH groups on both surfaces (Fig. 2 b). For all further oxygen plas- ma treatments, modified parameters were employed. After plasma activation, the activated surfaces of the two substrates were brought into contact, using meth- anol as an aligning medium (Fig. 2 b). After methanol evaporation, covalent bonds were formed which were then additionally stabilized by curing at 80°C for 1 h in laboratory furnace. 2.4 Characterization of the bond strength The bond strength was evaluated by performing de- lamination and burst pressure tests. Fig. 3 shows PDMS residues on ABS substrate after the PDMS elastomer was delaminated. In this particular case, APTES was employed as a linking agent. It was presumed that the area of PDMS residues left on the TP surface was directly related to bond strength. No obvious correla- tion between TP type and bond strength was found by delamination tests. Figure 3: Photography of PDMS residues on ABS sub- strate after the PDMS elastomer was delaminated. Next, the bond strengths were measured using burst pressure test devices. Top view and lateral cross-sec- tion of designed burst pressure test device is shown in Fig. 4.Figure 2: Process flow for TP-PDMS sandwich covalent bonding. B. Pečar et al; Informacije Midem, Vol. 47, No. 3(2017), 147 – 154 150 Figure 4: Schematics illustration of burst pressure test device. Top view (a) and lateral cross section (b). Devices were fabricated by employing replica molding technique. Silicon mold for PDMS cast was fabricated by one-step photolithography and deep reactive ion etching (DRIE). Casted PDMS elastomer layers with mi- crostructures (square-shaped inlet chambers with mi- crochannels) were thermally cured at 60 °C for 1 hour and bonded to TP substrates applying bonding process described in Sect. 2.3. Pressure regulated air supply was connected to the inlet of the test device and the pressure at which the device failed was determined. Device always failed at the region where the square-shaped inlet chamber narrows into the channel indicated by arrow position in Fig. 6. Here, the structural stress caused by applied fluidic pressure was the largest. The region of failure was further confirmed by employ- ing 3-D numerical simulations in COMSOL Multiphysics software. Test device behavior can be explained consid- ering two different physics models coupled together. Fluid flow is described by the Navier-Stokes equation ( )( ) Tp t ρ ρ µ∂   + ⋅∇ = ∇ − + ∇ + ∇ +   ∂   v v v I v v F (1) where the left hand side represents contribution of the force acting on a differential volume of a fluid and the inertial force. v is the fluid velocity, ρ density, p pressure and µ dynamic viscosity. Equation (1), which is describ- ing conservation of momentum, needs to be solved together with equation of mass continuity which for incompressible fluid reads 0∇ ⋅ =v (2) Deformation of a structure is modeled by structural mechanics equation for displacement vector u 2 2t ρ ∂ = − ∇ ⋅ ∂ u u f σ (3) where fu is a force acting on a differential volume and σ is a stress tensor. Stress σ and strain ε tensors are re- lated through equation σ = cEε, where cE is the elastic- ity matrix (determined with Young modulus of elastic- ity and Poisson ratio). Equations (1) to (3) are solved for fluid velocities and structural deformations together with supporting relations described in the text. After the pressure boundary condition was applied on the inlet, stationary direct fully-coupled solver was em- ployed to solve structural deformation and stress in burst pressure test device. Outlet boundary condition velocity was set to zero, although model tolerates also non-zero outlet boundary condition, due to coupled fluidic module. Simulated magnitude of bond stress at applied inlet fluidic pressure of 50 kPa is shown in Fig. 5. Maximum stress was calculated at the region (pre- sented with dark red color in Fig. 5) where fabricated test devices failed first (pointed out with arrow in Fig. 6), which positively validated simulation model. Figure 5: Simulated magnitude of bond stress at el- evated fluidic pressure. Arrow indicates the point of maximum simulated bond stress. Figure 6: Micrograph of burst pressure test device. Ar- row shows the failure onset region. Simulation results show that the microfluidic channel design should avoid sharp edges throughout flow di- rection where layers delamination might be initiated. B. Pečar et al; Informacije Midem, Vol. 47, No. 3(2017), 147 – 154 151 3 Results Since surfaces wettability is an essential criterion for inspecting the degree of surface activation in covalent bonding procedures, WCAs were measured first on raw and with O2 plasma treated TP and PDMS surfaces. Re- sults are presented in Table 1 (see also Fig. 1). Table 1: WCAs for raw and oxygen plasma treated sam- ples. Polymer raw O2 plasma (1 min) O2 plasma (5 min) ABS 87 27 32 PC 74 38 37 PMMA 69 45 47 PDMS 87 4.5 / Next, APTES (5%v/v in DI) was employed for function- alization of TP surfaces. With respect to others [2, 12] an additional post-deposition heat treatment and modified post-deposition plasma treatment were in- troduced in order to improve hydrolytic stability of co- valent bonds. As expected, oxygen plasma treatment after modified APTES functionalization considerably improved surfaces wettability (see Table 2), which in- dicates high surface concentration of hydroxyl groups. However, wettability still declined considerably over time. In this study, no straightforward correlation be- tween wettability of TP surfaces and TP-PDMS bond strength was found after the TP surfaces were func- tionalized. Table 2: WCAs for APTES functionalized and oxygen plasma treated samples. Polymer APTES (5%v/v in DI) Bond strength [°] 0 min [°] after 30 min [°] after 24 h O2 plasma (20 sek) ABS 9.5 38.7 74 accept. PC 14.5 47 81 accept. PMMA 12.3 36.5 83 accept. In further investigation, 30 burst pressure tests em- ploying water and compressed air were performed on fabricated PDMS test devices where TPs were previ- ously functionalized with APTES or amine-PDMS linker. Again, no direct correlation between TP type (receiving equal surface preparation and bonding process) and bond strength was found. All test devices could sustain air pressures of 5 bars for at least 3 hours. However, when burst pressure tests were performed by pressurized water, APTES devices started to fail imme- diately after the fluidic pressure of 1 bar was applied with a channel edge delamination rate of 7.5µm min-1 in spite of additional post-deposition heat treatment and modification of post-deposition plasma treatment. Similar hydrolysis of covalent bonds established via conventionally treated APTES was reported by Aran K et al. [12]. Results from testing bond strength under applied air pressure showed that the thermoplastics - PDMS bond was able to withstand more than 227.8 kPa, which was the maximum limit of their measuring equipment, without any sign of delamination. In fur- ther tests, the channels of the microdevices were filled with water and stored at room temperature for 72 h or the devices were filled and completely immersed in water for 72 h. Surprisingly, the reported bonding strength for APTES coated membranes remained very strong (over 227.8 kPa) in devices stored at room temperature for 72 h with the device channels filled with water. However, complete immersion of the devices in water for an ex- tended period of time weakened the membrane bond- ing strength for all of their tested bonding methods. In another study, S. Kevin Lee et al. [2] reported bond failure and delamination of PDMS-APTES-TP sandwich structure after subjected to burst pressure test with water compressed above 15 psi. An overview of silane is needed in order to understand the mechanisms for hydrolysis-induced bond failure. An organofunctional silane is a molecule comprising a silicon atom with at least one bond to carbon to en- able organic functionality [6]. The inorganic side of the silane molecule consists of a silicon atom bound to alk- oxy groups through Si—O—C linkages [2]. Hydrolytic instability of these bound alkoxy groups allows silanes to hydrolyze in the presence of water, converting the bound alkoxy groups to hydroxyl groups while liber- ating alcohol molecules. Any contact with water after bond formation will result in Si—O—C bond hydro- lysis and ultimately bond failure [6, 13]. Furthermore, Si—O—C bonds have also been found to form directly between alkoxy groups such as methoxy and surface hydroxyl groups via alcoholysis [14]. Hydrolytic bond failure can occur at three locations in the bonding structure, at the thermoplastics-silane interface, at the PDMS-silane interface, and in the silane network itself [15]. While direct interface hydrolysis is unlikely due to the stability of the amide bond, any hydrophilic groups at the interface can act as nucleation sites for water condensation, allowing the silane network near the interface to be plasticized and weakened [16]. A simi- lar process can occur at the PDMS-silane interface but with the possibility of hydrolysis directly at the inter- B. Pečar et al; Informacije Midem, Vol. 47, No. 3(2017), 147 – 154 152 face in addition to the weakening of the silane network [17]. For the silane network itself, high crosslink density can provide a major increase in resistance. However, networks formed by typical silanes, containing three silanol groups, tend to be cyclic, decreasing their re- sistance to dissolution [18]. Addressing failure mecha- nisms in all three locations is necessary to ensure hy- drolytic stability [6,18]. In further investigation, burst pressure tests were per- formed on amine-PDMS linker test devices fabricated on PC, ABS and acrylic glass using pressurized water. Test devices withstood 2 bar water pressure for 3 hours without any delamination observed. Moreover, amine- PDMS linker devices were soaked in water for one week and successfully withstood all additional burst pressure tests. All tests confirmed hydrolytic stability of TP-PD- MS bonds established through amine-PDMS linker. It is speculated that the water-repelling nature of the PDMS component in amine-PDMS linker prevented penetra- tion of the aqueous solutions at the interface improv- ing bond hydrolytic resistance [2, 3, 6]. Finally, developed bonding processes employing AP- TES and amine-PDMS linker were applied in modified micropump fabrication process. Based on the poor re- sults of water tests on burst pressure devices, we fur- ther modified the APTES application by using multiple deposition steps (3 to 5 deposited layers), thus expect- ing the improvements of bonds hydrolytic resistance. Our previously developed piezoelectric microcylinder pump prototypes [19] comprise activated PDMS elasto- mer layer bonded on its bottom side to the supporting bottom glass. Supporting bottom glass includes im- provised fluidic connections and serves as a functional part of the micropump affecting micropump perfor- mance characteristics. Replacing supporting bottom glass with TP could pave the path toward micropump mass production in terms of enclosing the micropump in professional TP housing comprising professional flu- idic and electric connections. In the initial stage, supporting bottom glass was re- placed with flat TP substrate. Exploded view of a typical TP microcylinder pump structure is shown in Fig. 7. The TP microcylinder pump comprises PDMS elasto- mer layer with molded micropump chamber, fluidic mi- crochannel and rectifying elements (Fig. 7 c). Addition- ally, two through-holes are punched into an elastomer, one into the center of the micropump chamber and the other one at the end of the channel. PDMS elastomer layer (Fig. 7 c) and PDMS fluidic connections (Fig. 7 e) are covalently bonded to the supporting TP substrate (Fig. 7 d) by employing developed multiple deposition APTES or single deposition amine-PDMS linker bond- ing process. One inlet and one outlet fluid port is drilled through a supporting TP substrate that supply and drain the fluid into and out of the pump. The micropump cham- ber and the microchannel are sealed with a thin glass membrane (Fig. 7 b) by employing oxygen plasma PDMS-glass covalent bonding process. Piezoelectric actuator (Fig 7 a) is positioned in the axis of a micro- pump chamber, coupled rigidly to the micropump membrane through silver filled epoxy adhesive (EPO- TEK EE129-4). During excitation loosely attached glass membrane and PDMS elastomer layer deform in a controlled man- ner, which enables compression and expansion of the centrally placed inlet cylindrical port, micropump chamber and outlet throttle shaped port with a specific phase lag, contributing to efficient micropump opera- tion. Pumping test were performed by pumping air and DI water media. Both approaches passed air pumping tests, reaching maximum flow rate performance of 8 ml min-1 and maximum backpressure performance of 100 mbar at applied square excitation waveform Figure 7: Exploded view of a typical TP microcylinder pump structure (dimensions are not to scale). B. Pečar et al; Informacije Midem, Vol. 47, No. 3(2017), 147 – 154 153 with an amplitude of 140 V and a frequency of 300Hz. However, during DI water pumping tests and despite improved multiple layer deposition process of APTES, micropumps degraded after several minutes of opera- tion. This is depicted in Fig. 8 as APTES MP3 character- istics and further expanded on time scale in an in-set (see Fig. 8 central plot on the figure plane). Figure 8: Long term flowrate stability characteristics of three representative micropump devices and ambient temperature. In contrast, amine-PDMS linker composed of a PDMS backbone incorporating an amine side group estab- lished hydrolytically stable covalent bonds. This was confirmed by long-term flowrate stability measure- ments on two micropumps MP1 and MP2 with typical performances yielding initial maximum DI water flow- rate performance of 1.2 ml min-1 and 1.4 ml min-1 at ap- plied RC excitation waveform [20] with an amplitude of 140 V and a frequency of 300 Hz. Figure 8 also includes measurement setup ambient temperature (dotted line closest to the upper edge of the diagram). Due to par- tial correlation between flowrate performance char- acteristics of MP1 and MP2 and measurement setup ambient temperature it was concluded that transient deviations in flowrate characteristics are to be attrib- uted to temperature changes of the medium viscosity, the micropump and driving electronics throughout the measurement. Therefore, long term stability measure- ments should be improved by setting measuring setup in a temperature stabilized chamber. Micropumps em- ploying amine-PDMS linker bonding process exhibit no deterioration in their performance after eight weeks of continuous operation. 4 Conclusions Low temperature process for TP-PDMS irreversible covalent bonding was presented. Process is based on silylation of the TP surfaces through the use of orga- nofunctional silanes. As the silanol groups are polar in nature, they make the exposed surface highly hy- drophilic and this was observed by measuring WCAs. However in this study, no direct correlation between wettability of TP surfaces deposited with APTES and fi- nal bond strength was found. In further investigation, burst pressure tests were performed on designed and fabricated PDMS test devices employing TPs function- alized with APTES or amine-PDMS linker. All devices sustained air pressures of 5 bars for at least 3 hours, but only amine-PDMS linker test devices sustained continu- ous water pressure as high as 2 bars without delamina- tion. Bonds established via APTES and subjected to wa- ter decayed in spite of additional post-deposition heat treatment and modification of post-deposition plasma treatment. In further application oriented study, both bonding processes were applied on piezoelectric mi- cropumps where glass substrate was replaced by ther- moplastic substrate. Even implementation of multiple deposition steps of APTES was insufficient in prevent- ing hydrolysis of covalent bonds, resulting in micro- pumps performance deterioration. On the other hand, micropumps employing amine-PDMS linker exhibit no deterioration in their performance even after eight weeks of continuous operation. 5 Acknowledgments Authors would like to thank the Slovenian Research Agency/ARRS (P2-0105, P2-0244), Ministry of Educa- tion, Science and Sport and Kolektor Group d.d. for their support of this work. 6 References 1. Yetisen, A. K., Akram, M. S., & Lowe, C. R. (2013). Based microfluidic point-of-care diagnostic de- vices. Lab on a Chip, 13(12), 2210-2251. 2. Lee, S. K., Lee, H., & Ram, J. R. (2016). U.S. Patent No. 9,422,409. Washington, DC: U.S. Patent and Trademark Office. 3. Wu, J., & Lee, N. Y. (2014). One-step surface modifi- cation for irreversible bonding of various plastics with a poly (dimethylsiloxane) elastomer at room temperature. Lab on a Chip, 14(9), 1564-1571. 4. Suzuki, Y., Yamada, M., & Seki, M. (2010). Sol–gel based fabrication of hybrid microfluidic devices composed of PDMS and thermoplastic substrates. Sensors and Actuators B: Chemical, 148(1), 323- 329. 5. Tsao, C. W., & DeVoe, D. L. (2009). Bonding of ther- moplastic polymer microfluidics. Microfluidics and Nanofluidics, 6(1), 1-16. B. Pečar et al; Informacije Midem, Vol. 47, No. 3(2017), 147 – 154 154 6. Lee, K. S., & Ram, R. J. (2009). Plastic–PDMS bond- ing for high pressure hydrolytically stable active microfluidics. Lab on a Chip, 9(11), 1618-1624. 7. Tang, L., & Lee, N. Y. (2010). A facile route for ir- reversible bonding of plastic-PDMS hybrid mi- crodevices at room temperature. Lab on a Chip, 10(10), 1274-1280. 8. Karakoy, M., Gultepe, E., Pandey, S., Khashab, M. A., & Gracias, D. H. (2014). Silane surface modifi- cation for improved bioadhesion of esophageal stents. Applied surface science, 311, 684-689. 9. Garbassi, F., Morra, M., Occhiello, E., & Garbassi, F. (1998). Polymer surfaces: from physics to technol- ogy (pp. 169-200). Chichester: Wiley. 10. Hillborg, H., & Gedde, U. W. (1999). Hydrophobicity changes in silicone rubbers. IEEE Transactions on Dielectrics and Electrical insulation, 6(5), 703-717. 11. Bhattacharya, S., Datta, A., Berg, J. M., & Gango- padhyay, S. (2005). Studies on surface wettabil- ity of poly (dimethyl) siloxane (PDMS) and glass under oxygen-plasma treatment and correlation with bond strength. Journal of microelectrome- chanical systems, 14(3), 590-597. 12. Aran, K., Sasso, L. A., Kamdar, N., & Zahn, J. D. (2010). Irreversible, direct bonding of nanoporous polymer membranes to PDMS or glass microde- vices. Lab on a Chip, 10(5), 548-552. 13. Goebel, R. (2003). U.S. Patent No. 6,613,439. Wash- ington, DC: U.S. Patent and Trademark Office. 14. Pape, P. G., & Plueddemann, E. P. (1991). Methods for improving the performance of silane coupling agents. Journal of adhesion science and technol- ogy, 5(10), 831-842. 15. Plueddemann, E. P. (1991). Reminiscing on silane coupling agents. Journal of Adhesion Science and Technology, 5(4), 261-277. 16. Tesoro, G., & Wu, Y. (1991). Silane coupling agents: the role of the organofunctional group. Journal of adhesion science and technology, 5(10), 771-784. 17. Battjes, K. P., Barolo, A. M., & Dreyfuss, P. (1991). New evidence related to reactions of aminated silane coupling agents with carbon dioxide. Journal of Adhesion Science and Technology, 5(10), 785-799. 18. Plueddemann, E. P. (1991). Adhesion through si- lane coupling agents. In Fundamentals of adhe- sion (pp. 279-290). Springer US. 19. Dolžan, T., Pečar, B., Možek, M., Resnik, D., & Vrtačnik, D. (2015). Self-priming bubble tolerant microcylinder pump. Sensors and Actuators A: Physical, 233, 548-556. 20. http://www.electronics-tutorials.ws/rc/rc_3.html (17.10.2017) Arrived: 31. 08. 2017 Accepted: 26. 10. 2017 B. Pečar et al; Informacije Midem, Vol. 47, No. 3(2017), 147 – 154