1 Introduction Tribosystems with elastomeric sliding partners have characteristic properties in terms of friction and wear behaviours. Elastomer tribology is essentially based on four mutually influencing frictional force components, as described in the rubber friction theory of Kummer and Mayer [1] and Geyer [2]: F total = F adhesion + F hysteresis + F viscosity + F cohesion Aus Wissenschaft und Forschung 14 Tribologie + Schmierungstechnik · 65. Jahrgang · 6/ 2018 Influence from laser surface structuring of elastomers on their friction behaviours with respect to the load dependency J. Voyer, S. Klien, F. Ausserer, I. Velkavrh, P. O. Velazquez, G. Vorlaufer, A. Diem* In der Regel zeigen reine Elastomere aufgrund der hohen Adhäsionsneigung schlechte Gleiteigenschaften. Um diese zu verbessern, werden Zwischenstoffe (Öle, Fette, Lacke usw.) zum Trennen bzw. Funktionalisieren der Kontaktflächen eingesetzt. Eine andere Möglichkeit besteht in der Reduktion der Adhäsionsneigung selbst, z. B. durch Veränderung des Elastomerwerkstoffs selbst oder den hier vorgestellten Ansatz zur Reduktion der effektiven Kontaktfläche des Elastomers durch gezielte Oberflächenstrukturierung. In dieser Studie wurden unterschiedliche rillenförmige Strukturen mittels Ultrakurzpuls-Laser auf Elastomeroberflächen hergestellt, um den Effekt einer Reduktion der gesamten Reibkraft von Elastomeren auf Aluminium- und PA6.6-GF30-Platten im trockenen Tribo-Kontakt zu ermitteln. Dafür wurden das Deformationsverhalten mittels der Methode der finiten Elemente (FEM) simuliert bzw. die Flächenpressungsabhängigkeit des Reibwerts durch experimentelle Laststeigerungsversuche mittels eines RVM-Tribometers untersucht und Vergleiche zu unstrukturierten Elastomeroberflächen als Benchmark hergestellt. Die Ergebnisse dieser Untersuchungen haben gezeigt, dass durch die Veränderung der Elastomeroberfläche durch rillenförmige Strukturen eine Reibkraftreduktion erzielt werden kann. Schlüsselwörter Laserstrukturierung, Reduktion der effektiven Kontaktfläche, Adhäsionsreibung, Oberflächeneigenschaften, Elastomere Pure elastomers usually show poor sliding properties due to their high adhesion tendency. In order to improve these sliding properties, intermediate materials (oils, greases, lacquers, etc.) are used to separate or functionalise the contact surfaces. Another possibility for an improvement of sliding properties is to reduce the adhesion tendency, for example by modifying the elastomeric material itself or through the approach presented in the actual study: to reduce the effective contact area of the elastomer by means of targeted surface structuring. In the present study, different grooved structures were produced on elastomer surfaces using an ultra-short pulsed laser in order to determine their influence on a possible reduction of the total friction force of elastomers against aluminium and PA6.6-GF30 plates in dry tribo-contacts. For this purpose, the deformation behaviour of elastomers was simulated using the finite element method (FEM) along with experimental investigations, through load increase friction tests using a RVM tribometer, on the surface pressure dependency of the friction value and results were compared to those from unstructured elastomer surfaces, acting as benchmark. The results of these investigations have shown that a reduction of friction forces can be achieved through grooved structures on elastomer surfaces. Keywords Laser structuring, reduction of effective contact area, adhesive friction, surface properties, elastomers Kurzfassung Abstract * Dr. Joel Voyer 1 DI (FH) Stefan Klien 1 DI (FH) Florian Ausserer M.Sc. 1 Dr. DI Igor Velkavrh 1 Pedro Osvaldo Velazquez 2 Dr. DI Georg Vorlaufer 2 DI Alexander Diem 1 1 V-Research GmbH, 6850 Dornbirn, Austria 2 AC2T research GmbH, 2700 Wiener Neustadt, Austria T+S_6_18.qxp_T+S_2018 29.10.18 17: 05 Seite 14 The individual force components of the total friction force can be characterized as follows: F adhesion : product of the effective acting shear stresses and the effective contact area F hysteresis : damping losses in the rubber material which occurs during sliding over rough surfaces due to constantly changing deformations of elastomer F viscosity : shear forces of a viscous liquid layer enclosed between elastomer and surface F cohesion : intramolecular cohesive forces; forces occurring during crack formation In systems without any lubrication (dry systems), the adhesive friction component usually dominates and decreases when the surface roughness increases (lower effective contact area). The extent of the adhesive component is also significantly dependent on the materials used. The coefficient of friction of an elastomer tribopairing is also dependent on the normal load or contact pressure and the sliding speed. Due to mutual interactions, it is difficult to assign individual effects to a definite dependency on normal load or speed. Accordingly, a high diversity of load dependencies of the friction coefficient has been described in the past [3]. Tribological optimizations of tribological elastomeric systems are often performed by separating the contact surfaces with interfacial materials such as oils, greases, varnishes, or solid lubricants (graphite, MoS 2 , PTFE, nanotubes). Another possibility is to influence the friction forces by structuring the surfaces (topography), in particular if the adhesive frictional force component represents a significant proportion of the total frictional force. The focus of the current study is based on this last mentioned approach: investigation of the influence of a reduction of the effective contact area by a superficial laser structuring on the entire friction behaviour of selected elastomers. 2 Experimental Procedure For the present study, injection moulded rectangular EPDM (Ethylene-Propylene-Diene-M-Group) and LSR (Liquid Silicone Rubber) pads, having dimensions and surface roughness values listed in Table 1, were produced. Plates of Aluminium and PA6.6-GF30 were chosen and used as counter-bodies for the determination of the dry friction behaviour of the elastomer pads, and their dimensions and surface roughness values are also listed in Table 1. Plain non-structured elastomer pads (pad dimensions and surface roughness values also listed in Table 1) were used as benchmarks for the experimental investigations. In order to determine the effect of a reduction in the effective contact area, some of these benchmark pads were surface processed by means of an ultrashort pulsed laser for producing previously defined surface structures. Some properties of the laser used for surface patterning are listed in Table 2 (additional laser properties were previously published elsewhere [4, 5]). The geometry and dimensions of the desired surface structures are shown and listed in Table 3. For all surface structures, the period (p = 250 µm) and the width of the ridges (b = 50 µm) Aus Wissenschaft und Forschung 15 Tribologie + Schmierungstechnik · 65. Jahrgang · 6/ 2018 Table 1: Geometrical properties and surface roughness values of the investigated elastomer pads and plates Length Width Thickness R a R z Sample Material (mm) (mm) (mm) (µm) (µm) EPDM 3,8 Pad LSR 14 10 5 0,8 5,0 AlSiMgMn 0,9 5,8 Plate PA6.6-GF30 25 20 3 0,8 6,8 Table 2: Properties of the ultrashort pulsed laser used for surface structuring Property Value Wavelength (nm) 1035 Pulse duration (fs) 300 Pulse frequency (kHz) 100 Table 3: Properties of the desired surface structures to be produced on elastomer pads Ridge Dimensions Ratio Contact (µm) Surfaces Elastomer Structure Geometry Width Height Period A structure / (b) (h) (p) A benchmark 1 50 100 EPDM 2 50 60 1 50 60 250 20 % LSR 2 50 30 T+S_6_18.qxp_T+S_2018 29.10.18 17: 05 Seite 15 of a single structural element was firstly created. The structure shown as example in Figure 2 corresponds to the structure 1 of the EPDM elastomer with b = 50 µm and h = 100 µm. By assuming periodic boundary conditions (see Figure 2a) and a plane strain state, this model is representative of an infinitely extended periodic structure. The counter-body (Al or PA plate) was assumed to be fully rigid. The stress-strain behaviour of the elastomer was then described by means of a hyper-elastic material model considering large deformations according to Yeoh [6]. The material parameters were derived from previously measured tensile test data. Computer simulations of the time-dependent deformation behaviour of elastomers were performed using the predefined vertical displacement and sliding speed (50 mm/ min) (see Figures 2b and 2c). The vertical displacement was varied and the resulting nominal contact pressure was evaluated. For initial calculations, a coefficient of friction between the plate and the elastomer of μ = 0.4 was assumed. Aus Wissenschaft und Forschung 16 Tribologie + Schmierungstechnik · 65. Jahrgang · 6/ 2018 were kept constant, while the height of the ridges was varied between 30 µm, 60 µm and 100 µm. The reduction degree of the effective contact areas can be then determined by the ratio between the contact area of the structured pads (A structure ) and the benchmark pads (A benchmark ) and is influenced mainly by the ridge width and pad dimensions. The effective contact area was reduced by a factor of 5 for the structured surfaces with a ridge width of 50 µm compared to the benchmark. The influence of a variation of the ridge angle α (~ 90°) shown in Table 3 on the tribological properties of the elastomer pads was not taken into account in the present study. The real dimensions of the laser-generated surface structures were measured using a confocal white-light microscope (μSurf, NanoFocus AG) and compared with the desired theoretical values. In order to be able to estimate the influence of the laserinduced interactions on the elastomer material, additional samples with identical structures were also produced by means of a structured injection mould and tribologically tested with the same experimental parameters. There were no significant differences in the friction behaviour, showing that the laser-induced interactions during a direct laser-processing of elastomers possess no or limited effects on the tribological behaviours of structured elastomers. The influence of a reduction of the effective contact surface area by laser-produced surface grooves on elastomers was determined using stepwise load-increasing tribological tests with a RVM tribometer (RVM1000, Werner Stehr Tribologie). For this purpose, the unstructured (benchmark) or the groove-structured elastomer pads were tested against Al or PA6.6-GF30 plates under dry tribo-contact using predefined test parameters, which are listed in Table 4. A detailed view of the test setup used is shown in Figure 1. During these tribological experiments, the elastomer pad was moved in a linear oscillating displacement against a plate (Al or PA6.6-GF30) and iteratively loaded, starting with a normal force of 10 N up to 400 N. For the analysis of the results, normal force, frictional force, speed and sample temperature were recorded and analysed over time. 3 Computer Simulations In order to gain a better understanding of the deformation behaviour of the surface structures when subjected to normal loads and/ or a relative motion, in addition to the tribological experiments, computer simulations were also carried out using the finite element method (FEM) and COMSOL-Multiphysics as software. For these simulations, a 2D model which simulates a cross-section Table 4: Test parameters used for the determination of the contact pressure dependency of the friction coefficient of non-structured (benchmark) and structured elastomer pads Parameter Value Elastomer Pads EPDM / LSR Counter Body (Plate) AlSiMgMn / PA6.6-GF30 10, 15, 20, 25, 30, 40, 50, Normal Load (N) 80, 100, 200, 400 Speed (mm/ min) 48 Stroke (mm) 11 Test Temperature (°C) 22 Lubrication None (dry) Figure 1: Experimental setup used for the determination of the contact pressure dependency of the friction coefficient of non-structured (benchmark) and structured elastomer pads through stepwise load increasing tests T+S_6_18.qxp_T+S_2018 29.10.18 17: 05 Seite 16 4 Results and Discussion Typical measured topographies of the individual laserstructured elastomeric pads (EPDM or LSR) are shown in Figure 3 and a comparison of their real dimensions to the desired target values is listed in Table 5. It can be seen that the dimensions of the produced structures are in good agreement with the theoretical target dimensions. The results of the load increasing tests are shown in Figure 4 for EPDM against Al or PA6.6-GF30 and in Figure 5 for LSR against Al or PA6.6-GF30. In the following section, results for EPDM are firstly presented and analysed followed by the results for LSR. The measured friction coefficients for EPDM are shown in Figures 4a and 4b (4a: against Al plate, 4b: against PA plate) as a function of the applied normal load. Figures 4c and 4d show detailed representations for a normal load range up to 100 N, in which the adhesive component of the total frictional force has a high influence. In all diagrams shown, trend curves serving as visual aid for the comparison of the results were added. These trend curves were selected without any physical background and are not intended to be used as a phenomenological description of the load dependency of the friction coefficient. For structure 1 on EPDM against Al (Figures 4a and 4c), no trend curve was added as two opposite tendencies can be observed (F N < 30 N, F N > 30 N). Aus Wissenschaft und Forschung 17 Tribologie + Schmierungstechnik · 65. Jahrgang · 6/ 2018 Figure 2: Graphical representation of the simulation modelling: a) representative single structural element with periodic conditions, b) simulation of the normal load application through a vertical plate displacement, and c) simulation of the relative displacement between elastomer pad and plate a) b) c) Figure 3: Typical topography of laser-structured elastomer pads: a) EPDM structure 1 (remark: different height scale), b) EPDM structure 2, c) LSR structure 1 and d) LSR structure 2 a) c) b) d) T+S_6_18.qxp_T+S_2018 29.10.18 17: 05 Seite 17 Aus Wissenschaft und Forschung 18 Tribologie + Schmierungstechnik · 65. Jahrgang · 6/ 2018 Table 5: Properties of laser-produced surface structures on EPDM and LSR pads Ridges Set-Value Ridges Is-Value A structure / (µm) (µm) A benchmark Elastomer Structure Width Height Period Width Height Period Set- Is- (b) (h) (p) (b) (h) (p) Value Value 1 50 100 250 54 115 260 20 % 20,9 % EPDM 2 50 60 250 49 59 245 20 % 20,0 % 1 50 60 250 46 65 248 20 % 18,5 % LSR 2 50 30 250 49 36 246 20 % 19,9 % Figure 4: Friction coefficients of non-structured (benchmark) and structured EPDM pads as function of the normal load; total normal load range: a) against Al, b) against PA; and detailed representations for F N < 100 N: c) against Al, d) against PA a) c) b) d) Al or PA under dry conditions can be observed for an optimal surface structuring with groove dimensions of b = 50 µm and h = 60 µm. The friction coefficients for LSR are shown in Figures 5a and 5b (5a: against Al plate, 5b: against PA plate) over the full normal load range under study. Detailed representations for normal loads less than 100 N (where the adhesive component of the friction force is dominant) are shown in Figures 5c and 5d, respectively. In all diagrams, trend lines were again added as a visual comparison aid. Both surface structures enable a reduction of friction coefficients for normal loads smaller than In comparison to the benchmark, surface structure 2 allows a reduction of the friction coefficient for normal loads lower than approximately 200-300 N for EPDM pads independently of the counter-body plate (Al or PA), while surface structure 1 allows a reduction of the friction coefficient again for normal loads lower than approximately 200-300 N for EPDM pads against PA only. For normal loads greater than about 300-400 N for EPDM against Al or PA, both surface structures show friction coefficients similar to those of the benchmark. Nevertheless, for low normal loads (F N < 100 N), a halving of the friction coefficient of the EPDM pads against T+S_6_18.qxp_T+S_2018 29.10.18 17: 05 Seite 18 For these LSR pads, the normal load range for a possible halving of the friction coefficient is F N < 50 N against Al plates, while it is F N < 30-40 N against PA plates. Typical results of computer simulations are shown in Figure 6 for EPDM elastomer with surface structure 1. Simulations of the other studied structure and material combinations provided comparable results as those presented in Figure 6. Figure 6a represents the initial state of the surface structure before any normal load application or any relative movement, i.e. in the unloaded condition. At a nominal surface contact pressure of approximately 0.8 MPa, a slight deformation of the structure occurs, as shown in Figure 6b. It can be seen that the structure essentially retains its original shape, so that the real contact surface area is still significantly reduced when compared to the unstructured reference (benchmark). As the nominal surface contact pressure increases to 2.7 MPa, as shown in Figure 6c, the structure collapses and the real contact surface area approaches the value of the unstructured reference surface (benchmark). Thus, these simulation results confirm the observations from experiments: at high normal loads, friction coefficients of structured elastomer pads tend to reach similar values as the benchmark. Aus Wissenschaft und Forschung 19 Tribologie + Schmierungstechnik · 65. Jahrgang · 6/ 2018 approximately 200 N when compared to the benchmark for LSR pads against Al or for normal loads smaller than approximately 100 N for LSR pads against PA. For normal loads greater than 200 N for LSR against Al or for normal loads greater than 100 N for LSR against PA, both surface structures show friction coefficients similar to the benchmark. As previously observed for EPDM, friction coefficients of the structured LSR pads tend to reach similar values as those for the unstructured LSR pads at high normal loads. The reason for this behaviour is believed to be due to the limited load bearing capacity of the structures. The structure ridges are compressed under high normal loads, leading to a deformation-dependent enlargement of the contact surface area and subsequently resulting in friction coefficients similar to those of unstructured pads (benchmark). Because EPDM is harder than LSR, the critical normal load for which this deformation of the ridges occurs is slightly lower for LSR than for EPDM. Nonetheless, a halving of the coefficient of friction of the LSR pads under dry friction conditions can be observed for an optimal surface structuring with groove dimensions of b = 50 µm and h = 60 µm or 30 µm. Figure 5: Friction coefficients of non-structured (benchmark) and structured LSR pads as function of the normal load; total normal load range: a) against Al, b) against PA; and detailed representations for F N < 100 N: c) against Al, d) against PA a) c) b) d) T+S_6_18.qxp_T+S_2018 29.10.18 17: 05 Seite 19 References [1] Kummer, H.W., Unified Theory of Rubber and Tire Friction, Engineering Research Bulletin B-94, Pennsylvania State University, 1966. [2] Geyer, W., Beitrag zur Gummireibung auf trockenen und insbesondere nassen Oberflächen, München, Technische Universität, Dissertation, 1971. [3] Sinha, S.K., Briscoe, B.J., Polymer Tribology, Imperial College Press, ISBN-13 978-1-84816-202-0, 2009. [4] Voyer, J., Klien, S., Ausserer, F., Velkavrh, I., Ristow, A., Diem, A., Friction Reduction Through Sub-Micro Laser Surface Modifications, Tribologie & Schmierungstechnik, 62. Jahrgang, no. 5, pp.13-18 (2015). [5] Voyer, J., Ausserer, F., Klien, S., Ristow, A., Velkavrh, I., Diem, A., Zehetner, J., Stroj, S., Edlinger, J., Heidegger, S., Bertschler, C., Sub-Micro Laser Modifications of Tribological Surfaces, Materials Performance and Characterization: Special Issue on Surface Texturing, vol. 6, no. 2, pp.42-67 (2017). [6] Selvadurai, A.P.S., Deflections of a rubber membrane, Journal of the Mechanics and Physics of Solids, vol. 54, pp.1093-1119 (2006). Aus Wissenschaft und Forschung 20 Tribologie + Schmierungstechnik · 65. Jahrgang · 6/ 2018 5 Summary The present study has shown that a reduction of up to a factor of 2 of the friction coefficient of elastomers through a targeted surface structuring under low surface pressures and dry tribo-contact is possible. The production of the desired surface structures can be realised by a direct processing of the elastomer using an ultrashort pulsed laser. FEM simulations have shown to be beneficial in order to estimate the suitable structural dimensions with regard to their deformation behaviours. Acknowledgments The work presented was funded by the Austrian COMET Programme (Project XTribology, no. 849109) and carried out at V-Research GmbH and AC2T research GmbH, the Excellence Centre of Tribology. Figure 6: Computer simulation of the behaviour of structured EPDM elastomer (structure 1: b = 50 µm, h = 100 µm): a) in non-loaded original state, b) light deformation through sliding and low normal load (nominal contact pressure approx. 0.8 MPa) and c) high deformation or complete flattening of the surface structure through sliding and high normal load (nominal contact pressure approx. 2.7 MPa) a) c) b) T+S_6_18.qxp_T+S_2018 29.10.18 17: 05 Seite 20