Aus der Praxis für die Praxis Tribologie + Schmierungstechnik 64. Jahrgang 4/ 2017 55 Friction reduction due to plasmapolymeric coating in spite of lubrication deficiency S. Schmidt, D. Paulkowski, K. Vissing* Reibungsreduzierung und dadurch Energieeinsparung sind von besonderem Interesse für viele Industrien wie beispielsweise den Automobilbereich, Windenergie und alle anderen Bereiche in denen elastomere Komponenten wie Dichtringe verwendet werden. Mangelschmierung kann unvorhersehbar auftreten und die Dichtung schädigen. Durch das Aufbringen einer plasmapolymeren Beschichtung auf Elastomere kann die Reibung reduziert werden. Vier Elastomertypen (Acrylatkautschuk, Fluorkautschuk sowie zwei Nitril-Butadien- Kautschuke) wurden jeweils mit und ohne plasmapolymere Beschichtung untersucht. Die Experimente wurden mit flachen beschichteten und unbeschichteten Elastomerplatten durchgeführt und die Reibung mit Hilfe eines Universal Material Tester (UMT3) Systems in oszillierender Pin-on-Plate Kontaktgeometrie gemessen. Die Stirnseite eines zylindrischen Radlagers mit einem Durchmesser von 6 mm wurde als Gegenkörper verwendet. Diese Kontaktgeometrie wurde sowohl wegen der Differenzierbarkeit von Schmierstoffen im oszillierenden Test, als auch wegen der 10 4 mal größeren Kontaktfläche im Vergleich zu einer Dichtlippe eines Radialwellendichtrings ausgewählt. Die tribologischen Untersuchungen wurden unter Umgebungsbedingungen mit einer Geschwindigkeit von 200 mm/ s und einem Hub von 11 mm durchgeführt. Die verwendete Normalkraft von 10,6 N repräsentiert eine initiale Flächenpressung von 0,5 MPa. Um Mangelschmierung zu erreichen, wurden die Tests mit verschiedenen kleinen Mengen eines voll additivierten Getriebeöles bis hin zur kompletten Vermeidung von Schmierstoff durchgeführt. Die beschichteten Elastomere zeigten eine reduzierte Reibung im Vergleich zu den unbeschichteten Proben. Bei den beschichteten Proben zeigten die Messungen ohne Schmierung die geringste Reibung. Reibungskoeffizienten bis hinunter zu 0,17 wurden erreicht. Im Vergleich mit 0,48 bei trockenem Lauf unbeschichteter Proben ergibt sich damit eine Reibungsreduzierung um 65 %. Dies demonstriert die Notlaufeigenschaft für plasmapolymer beschichtete Komponenten. Für die geschmierten Messungen ergaben sich Unterschiede für die verschiedenen Schmierstoffmengen. Die plasmapolymere Beschichtung und damit die Reibungsreduzierung sind durch Anwendung der plasmaunterstützten chemischen Gasphasenabscheidung (PECVD) zu geringen Kosten verfügbar. Mittels PECVD ist es problemlos möglich komplexe 3D-Geometrien, wie beispielsweise Dichtringe, zu beschichten. Der Prozess wurde bereits für Radialwellendichtringe aufskaliert um eine Serienproduktion zu ermöglichen. Schlüsselwörter Plasmapolymere Beschichtung, Mangelschmierung, Elastomer, Gummi, Reibungsreduzierung, Energieeinsparung Reducing friction and thereby saving energy is of special interest for several industries like automotive, wind energy and any other section using elastomeric components, for example sealing rings. Unpredictable, deficient lubrication can occur and damage the sealing. Applying a plasmapolymeric coating on elastomers the friction can be reduced. Four elastomer types (acrylic rubber, fluoric rubber and two different nitrile butadiene rubbers) were investigated each with and without plasmapolymeric coating. The experiments were performed on coated and uncoated flat elastomer plates and the friction reduction was measured using a Uni- Kurzfassung Abstract versal Material Tester (UMT3) system with oscillating Pin-on-plate contact geometry. The face of a cylindric wheel bearing with 6 mm in diameter was used as counterpart. The contact geometry was chosen due to the differentiability of lubricants in oscillating tests as well as the 10 4 times bigger contact area compared with the sealing lip of a radial shaft sealing. The tribological tests were done under ambient conditions with a velocity of 200 mm/ s and a stroke length of 11 mm. The used normal force of 10.6 N represents an initial contact pressure of 0.5 MPa. The tests were performed using different small amounts of fully formulated gear oil * Stefanie Schmidt (M. Sc.), Dr. Dominik Paulkowski, Dr. Klaus Vissing, Fraunhofer IFAM, 28359 Bremen T+S_4_17 07.06.17 17: 27 Seite 55 Aus der Praxis für die Praxis 1 Introduction Elastomers are used in several industries and their tribological properties are of crucial interest, for example for sealing rings or other elastomeric devices. A reduced friction of sealing rings in an automotive application realizes the reduction of CO 2 emissions due to energy saving. Friction reduction on elastomers can be achieved by applying a plasmapolymeric coating on the elastomers. Investigations regarding reduced dry and lubricated friction of flat elastomer plates using plasmapolymeric coatings were published by the authors in the past [1]- [4]. Moreover the reduction of friction on coated radial shaft sealings was demonstrated [5]. Consequently, investigations at lubrication deficiency are presented in this article to demonstrate emergency running properties of plasmapolymeric coated elastomers. For the developed chemical composition of the plasmapolymeric coatings to reduce friction of plastic and elastomeric materials a patent was issued [6]. 2 Experimental details 2.1 Substrate preparation As substrates four different elastomer types were used: acrylic rubber (ACM), fluoric rubber (FKM) and two different nitrile rubbers (NBR). One nitrile rubber type was sooty (NBR(s)) and the other was minerally filled (NBR(m)). The Shore-A hardness of ACM, FKM, and NBR(s) was 75, NBR(m) had a Shore-A hardness of 72. The samples had a size of 1 cm x 2 cm or 2 cm x 2 cm and were cut out of bigger plates. The ACM, FKM, and NBR(s) plates had a thickness of 2 mm. The NBR(m) was 3 mm thick. Prior to the film deposition the samples were cleaned in a wet chemical solution in an ultrasonic bath at 60 °C. 2.2 Film preparation The elastomer substrates were coated with a plasmapolymeric coating (SiO x C y H z ). The films were prepared using a plasma enhanced chemical vapor deposition (PECVD) process with oxygen (O 2 ) and Hexamethyldisiloxan (HMDSO) as process gases. The Young’s modulus of the plasmapolymeric coating determined by nanoindentation with a Berkovich indenter is 38.4 GPa ± 2.0 GPa [4]. Before starting the deposition process, the chamber was evacuated to a base pressure below 1 x 10 -2 mbar. The working pressure during the deposition process was around 2 x 10 -2 mbar. For improved adhesion a thin interlayer between substrate and topcoat was applied using oxygen and Hexamethyldisiloxan as well. Prior to the deposition process the samples were activated in a hydrogen-oxygen plasma. Due to the PECVD process the substrate temperature during all process steps was below 37 °C. 2.3 Experimental setup The tribological tests were carried out using an Universal Material Tester (UMT3) system [7] with oscillating Pinon-plate contact geometry (see Figure 1). The face of a cylindrical wheel bearing with 6 mm in diameter was used as counterpart (Schaeffler ZRB6X6 attended to DIN5402-1, log. profile). This contact geometry was chosen due to two different reasons. First of all there is the need for distinguishing sorts of different oils [1], [3]. In the further contemplation 56 Tribologie + Schmierungstechnik 64. Jahrgang 4/ 2017 down to avoiding lubricants to demonstrate the effect of lubrication deficiency. The coated elastomers showed a reduced friction compared to the uncoated ones. If a coating was applied to the elastomers the measurements without any lubrication showed the lowest friction. Friction coefficients down to 0.17 were achieved. Compared to 0.48 for the dry measurement with uncoated substrate, a friction reduction of 65 % was reached. This demonstrates emergency running properties for plasmapolymeric coated components. For the lubricated measurements differences between the different lubrications appeared. The plasmapolymeric coating and therewith the friction reduction is available at low costs using a plasma enhanced chemical vapor deposition technique (PECVD). With PECVD it is easily possible to apply coatings on complex 3D-geometries like sealing rings. The process was already scaled up for radial shaft sealings to enable series production. Keywords plasmapolymeric coating, lubrication deficiency, elastomer, rubber, friction reduction, energy saving Figure 1: Schematic of the Pin-on-plate contact geometry. Uncoated cylindrical wheel bearing with 6 mm diameter in contact with coated or uncoated elastomer plate. The flat elastomer plate is glued to a steel plate and is oscillating. T+S_4_17 07.06.17 17: 27 Seite 56 Aus der Praxis für die Praxis there is the 10 4 times bigger contact area compared with the sealing lip of a radial shaft sealing (see Figure 2). The contact situation enables that the tests can be performed under deficiency lubrication. The tribological tests were done under ambient conditions with a velocity of 200 mm/ s and a stroke length of 11 mm. The used normal force of 10.6 N represents an initial contact pressure of 0.5 MPa. The elastomer samples were glued on a steel plate to be fixed in the UMT3. Thereby the tool traces which arise during creation of the elastomer plates were arranged in the oscillating direction. The tests were performed using fully formulated gear oil ref. 2 of [3] (kinematic viscosity @40 °C 86.2 mm 2 / s and @100 °C 12.3 mm 2 / s). This commercial gear oil is based on PAO. Different small amounts of the gear oil and avoiding lubricants demonstrate the effect of lubrication deficiency. As first step undiluted lubricants were used in the form of 0.5 µl droplets. To achieve very small lubrication amounts the oil was solved in solvent (isopropanol or hexane) and was applied to the sample. Before starting the tribological measurement it was waited for a drying off of the solvent to achieve very small amounts of lubrication. Dilutions with 0.5 %, 5.0 % and 20 % oil were tested. Every time a drop of 10 µl was deposited on the sample. The outcomes of these were lubrication amounts of 0.05 µl oil, 0.5 µl oil and 2.0 µl oil, respectively. As reference measurement the pure solvent was performed as well. 3 Theory Rubber friction shows a strong dependency to the sliding velocity [8]-[13]. Due to the surface roughness of the contacting mating partners there is only local molecular adhesion. Depending on the velocity shearing, deforming, breaking and re-making of each bonds has to be expected. In the case of the viscoelastic elastomers the material creeps into the cavities of the surface roughness. Subjecting to the time of contact at sliding motion both effects of reassembling bonds, filling cavities as well as changing viscoelasticity result in a maximum of friction in sliding friction. In passenger cars the variation of velocity represents driving at different speeds. To reduce the friction a plasmapolymeric coating with separating behavior was applied [2]-[5]. Regarding the ravel structure of the molecule chains of elastomers a plasmapolymeric coating provides a very similar configuration. One fulfilled demand is keeping the elasticity of the elastomers. Another fact is the strong adhesion of the coating to the elastomeric substrate due to the similar configuration. In combination with a slightly harder version of plasmapolymeric topcoat in relation to the hardness of the elastomer the wear resistance is increasing [14], [15]. Plasmapolymeric coatings reveal chemical resistance and can be used at low and high temperature. The used deposition technology PECVD allows series-production at low costs. For radial shaft sealings the process was already scaled up to enable series production [16]. The wear resistance of the plasmapolymeric coating protects the elastomer even at lubrication deficiency. The low adhesion on the surface of the coating leads to reduced friction at the same level as at the lubricated samples. 4 Results and discussion As described in section 2.3 different ways to achieve lubrication deficiency were tested. Figure 3 shows the first results using a 0.5 µl drop of pure oil and tests with 0.05 µl oil resulting from a dilution of oil in isopropanol or hexane. As expected, the plasmapolymeric coated samples showed lower coefficients of friction in comparison with the uncoated ones. In the case of undiluted oil the only exceptions are ACM and the NBR(s) due to the known behavior using gear oil ref. 2 [3]. Reduced viscosity was able to close the gap [3]. For the present investigations of lubrication deficiency it means unfavorable conditions for the plasmapolymeric coating. In spite of that fact the plasmapolymeric coating revealed positive effects. For the uncoated samples significant differences between the three lubrication types occurred, whereas in the most cases the lubrication with a diluted drop in isopropanol led to the highest coefficients of friction, for example values around 1.3 for uncoated ACM and NBR(m). Tribologie + Schmierungstechnik 64. Jahrgang 4/ 2017 57 a) b) Figure 2: a) Sketch of the contact situation of a radial shaft sealing in contact with a shaft and b) sketch of the contact situation in the Pin-on-plate contact geometry. The contact area of the Pin-on-plate geometry is 10 4 times bigger compared with the sealing lip of a radial shaft sealing, ensuring lubrication deficiency. T+S_4_17 07.06.17 17: 27 Seite 57 Aus der Praxis für die Praxis Using plasmapolymeric coated samples and the same measurement conditions, the coefficient of friction is only in the range of 0.2, which means a reduction by 84 %. Regarding the measurements with a diluted drop using isopropanol, all uncoated samples showed higher coefficients of friction than the measurements done with 0.5 µl of pure oil. There are two possible reasons for this behavior. On the one hand the diluted drop led to such a small amount of oil that lubrication deficiency with higher friction coefficients occurred. On the other hand, isopropanol is a polar solvent which could lead to a superficial swelling of the elastomer and thereby a change of the surface properties. In spite of short immersion time absorption of oil and swelling are accelerated or even only possible thereby. The durability is different for the tested elastomer types. Whereas ACM is more or less not resistant, FKM is resistant and NBR somewhat. This corresponds to the received friction coefficients where ACM showed a very high value and FKM a much lower one. In contrast, the plasmapolymeric coated samples showed no effect concerning isopropanol. The coefficients of friction were much lower than for the uncoated samples and the values were equivalent to the measurements done with the 0.5 µl drop of pure oil or even lower. The 0.5 µl drop of pure oil led to sufficient lubrication in the tribological system in spite of its small amount, based on the low friction coefficients for uncoated and plasmapolymeric coated samples. The fact, that the coated samples had the same values for the diluted drop in isopropanol than for the pure oil drop demonstrates that the plasmapolymeric coating protects the elastomer surface and substitutes the low lubrication. Further investigations were performed with the nonpolar solvent hexane to analyze whether the high friction co- 58 Tribologie + Schmierungstechnik 64. Jahrgang 4/ 2017 Figure 3: Coefficient of friction for different elastomers with different types of lubrication: a small drop (0.5 µl) of pure oil and a dilution of oil in either isopropanol or hexane. Measuring time five minutes, velocity 200 mm/ s, initial contact pressure 0.5 MPa. Figure 4: Variation of dilution of oil in hexane, as references dry and pure hexane measurements. Coefficient of friction of different uncoated and plasmapolymeric coated elastomers. Measuring time five minutes, velocity 200 mm/ s, initial contact pressure 0.5 MPa. T+S_4_17 07.06.17 17: 27 Seite 58 Aus der Praxis für die Praxis efficients of the uncoated samples were caused by an influence of the isopropanol or not. The nonpolar solvent hexane does not affect the elastomers. These experiments are shown in Figure 3 as well. The coefficients of friction of the uncoated samples are lower compared with the measurements using isopropanol, except for NBR(s). The coating revealed a friction reduction of diluted oil in hexane at NBR(s) from 0.76 in the uncoated case to 0.19 in the plasmapolymeric coated case. That means a reduction of 75 % due to the plasmapolymeric coating. In the case of FKM the value reaches the level of the pure oil drop. Everything taken into account the plasmapolymeric coated samples achieved lower friction coefficients, which mean that the plasmapolymeric coating improves the surface properties. The experimental series was continued with experiments using different dilutions of oil in hexane, which does not affect the elastomer. As references pure hexane as lubricant and dry measurements were done as well. The coefficients of friction of these measurements are depicted in Figure 4. On uncoated ACM and NBR(m) pure hexane led to an increased friction compared with the dry measurements. Using small amounts of diluted oil a decreasing friction could be observed with increasing oil amount for ACM and the two NBRs. At FKM a minimal increase within the error bars occurred. The uncoated FKM showed no differences between the different lubrications. Regarding the plasmapolymeric coated samples it revealed again, that the coating decreases the friction coefficients in general. Depending on elastomer type and lubrication, on the one hand the friction reduction is significant. On NBR(m) the dry measurement showed a friction reduction of 65 % from 0.48 down to 0.17. On NBR(s) the friction reduction is biggest with 70 %, from 0.64 down to 0.19. In the worst case the coated samples achieved the same values as the uncoated samples, for example at the most FKM measurements. In general, the plasmapolymeric coating achieved friction reduction even and also particularly at lubrication deficiency. There are two possible reasons for the positive effect on lubrication deficiency. On the one hand, the amount of additives might reduce the bonding to the surface. On the other hand, the friction reduction of the plasmapolymeric coated samples may be partially achieved due to a better wetting and therefore less oil on the same area. The surface energies for the uncoated elastomers and the plasmapolymeric coating are listed in Table 1. The surface energy of the plasmapolymeric coating is in the range of commonly used oils. But it must be assumed that the lubrication in these experiments was so small, that there was only a very limited amount of oil. The plasmapolymeric coating shows nevertheless emergency running properties at missing or lubrication deficiency (Figure 4). FKM showed in the uncoated but lubricated case lower friction coefficients than ACM and NBR. This could be attributable to the lower surface energy and thereby a worse wetting behavior. The oil drop spread over a smaller area so that there is more oil in the middle of the sample where the tribological test is performed. The photographs of uncoated ACM and FKM with a diluted and evaporated drop of oil in hexane demonstrate this effect (Figure 5). The theory of a better wettability of the plasmapolymeric coated samples can be verified with the photographs presented in Figure 6. These pictures show an uncoated and a plasmapolymeric coated FKM sample after the tribological experiment. In the case of the uncoated sample the oil is only in the area where the oil was Tribologie + Schmierungstechnik 64. Jahrgang 4/ 2017 59 Table 1: Surface energy of uncoated elastomers and plasmapolymeric coating. The surface energy for a nonpolar base oil is approx. 29 mN/ m and for a polar base oil approx. 36 mN/ m. Material Surface energy [mN/ m] ACM, uncoated 25.0 NBR(m), uncoated 22.4 NBR(s), uncoated 23.9 FKM, uncoated 21.2 Plasmapolymeric coating 31.2 b) Figure 5: Comparison of drop sizes of uncoated a) ACM and b) FKM with a 10 µl diluted drop of oil in hexane (0.5 % oil: hexane, resulting 0.05 µl oil after drying off of hexane). a) T+S_4_17 07.06.17 17: 27 Seite 59 Aus der Praxis für die Praxis dropped down and where the counterpart was moving and thereby spreading the oil. Regarding the plasmapolymeric coated sample the oil is spread over the whole sample. This means less oil in the area of tribological testing. Despite this further reduced lubrication in the friction test the coated sample persisted wear and showed low friction. The plasmapolymeric coating demonstrates therewith good emergency running properties. 5 Conclusion The tribological measurements were performed using uncoated and plasmapolymeric coated elastomer samples at lubrication deficiency. Depending on the elastomer type the plasmapolymeric coated samples showed a significant friction reduction at ACM and NBR or at least values in the same range as uncoated for FKM. Regarding the different lubrication amounts at ACM and NBR the dry measurements with coated substrates showed the lowest friction coefficients down to 0.17. It was shown that the plasmapolymeric coating could protect the elastomer surface against swelling due to solvent influence and oil absorption. The plasmapolymeric coating improves the wettability of oil on the sample surface. Consequently, the plasmapolymeric coating shows good emergency running properties under lubrication deficiency. Acknowledgement The authors acknowledge the financial support of the Bundesministerium für Wirtschaft und Energie (BMWi) and the participating companies. The support code was 03ET1187 B. References [1] D. Paulkowski, Rapid wear testing on coated elastomers using model test rig, Fall Rubber Colloquium KHK, Hannover 2016; D. Paulkowski, Verschleißmessung an beschichteten Elastomeren durch Schnelltests am Modellprüfstand, Forschung meets KMU - Analytik und Tribologische Kontakt-Systeme, Dortmund, 24.06.2015. [2] D. Paulkowski, K. Vissing, Tribological improvement of elastomers using plasmapolymeric coatings, Proceedings of Tribologie Fachtagung 2011; 15/ 1-15/ 14. [3] D. Paulkowski, K. Vissing, Reduction of elastomeric friction in lubricated contact using plasmapolymeric coatings, Proceedings of Tribologie Fachtagung 2012; 52/ 1-52/ 9. [4] D. Paulkowski, K. Vissing, R. Wilken, Composition of plasmapolymeric coatings using O2/ HMDSO gas mixtures and application on elastomers for tribological improvement, OR1007 (extended abstract), Proceedings of 13 th International Conference on Plasma Surface Engineering (PSE), Garmisch-Partenkirchen, Germany, 2012. [5] D. Paulkowski, K. Vissing, Plasmapolymeric coatings improve radial shaft sealing on application, Proceedings of Tribologie Fachtagung 2013; 87/ 1-87/ 11. [6] Patent „Dichtungsartikel“: DE 10 2008 002 515 A1. [7] Universal Material Tester (UMT3), Bruker formerly Center for Tribology Inc., 1717 Dell Ave., Campbell, CA 95008. [8] M. L. Williams, R. F. Landel, J. D. Ferry, The temperature dependence of relaxation mechanisms in amorphous polymers and other glass-forming liquids, J. Am. Chem. Soc. (1955), 77 (14), 3701-3707. [9] K. A. Grosch, The relation between the friction and viscoelastic properties of rubber, Proc. Roy. Soc., S. A (1963), 274, 21-39. [10] A. Schallamach, A theory of dynamic rubber friction, Wear, Volume 6, Issue 5, September-October 1963, 375-382. [11] M. Barquins, Sliding friction of rubber and Schallamach waves - A review, Material Science and Engineering (1985), 73, 45-63. [12] B. N. J. Persson, A. I. Volokitin, Theory of rubber friction: Nonstationary sliding, Physical Review B (2002), Vol. 65, 134106. [13] S. Sills, K. Vorvolakos, M. K. Chaudhury, R. M. Overney, Molecular origins of elastomeric friction, Fundamentals of Friction and Wear, NanoScience and Technology (2007), Part 7, 659-676. [14] D. Paulkowski, S. Karpinski, K. Vissing, Friction and wear resistance of plasmapolymeric coatings applied on elastomers, NordTrib 2014, Denmark; Finnish Journal of Tribology, Vol. 32, No. 2, 12-19 (2014). [15] D. Paulkowski, S. Karpinski, K. Vissing, R. Wilken, Wear protection of elastomers using plasmapolymeric coatings, poster PO4100 at 14th International Conference on Plasma Surface Engineering (PSE), Garmisch-Partenkirchen, Germany, 2014; Plasmapolymeric coating protects elastomers against wear, RFP 2/ 2017-Volume 12 [16] D. Paulkowski, K. Vissing, M. Santos, Cost-efficient production of plasmapolymeric coatings on rotary shaft sealings, Fluid Sealings 2016, Manchester (UK). 60 Tribologie + Schmierungstechnik 64. Jahrgang 4/ 2017 b) Figure 6: Comparison of oil spreading after tribological measurement of a) uncoated FKM and b) plasmapolymeric coated FKM. 0.5 µl pure oil drop. a) T+S_4_17 07.06.17 17: 27 Seite 60