Aus Wissenschaft und Forschung 22 Tribologie + Schmierungstechnik 63. Jahrgang 1/ 2016 Introduction Gaseous atmospheres have a significant influence on the friction and wear behaviour of metallic materials. In an O 2 -rich atmosphere, the friction and wear behaviour of metallic surfaces will be dominated by the adsorption of O 2 and the resulting formation of oxide coatings. In comparison to pure and non-oxidized metallic surfaces, for which very high adhesion may occur, these oxide coatings possess better sliding properties [1]. In previous studies [2-4], it was already determined that the gases Ar, N 2 and CO 2 may have a significant influence on the friction and wear behaviour, but the results were sometimes contradictory. The focus of the present study is the investigation of the sliding and wear behaviour in various gaseous atmospheres (air, Ar, N 2 and CO 2 ) using a SRV testing apparatus. Experimental Setup A testing apparatus based on the standard DIN 51834-1 [5] was adapted by adding a gas container, with which normal ball-on-disk wear tests could still be performed. This setup adaptation enabled the investigation of the influence of various gaseous atmospheres (air, Ar, N 2 and CO 2 ) on the tribological behaviour of steel-steel pairing. Figure 1 shows the modified test chamber of the testing apparatus. The setup consists of upper and lower sample holders, both joined by a rubber sleeve, which assures that the test chamber is gas tight. The upper sample holder is designed to accommodate standard balls having a diameter of 10 mm, while the lower sample holder possesses a gas inlet and a gas outlet and is designed to accommodate standard discs with a diameter of 24 mm and a height of 7.85 ± 0.05 mm. Furthermore, the standard disc holder in the lower sample holder was countersunk, enabling that experiments using liquid media (lubricants, salt water, etc.) and gaseous atmospheres could be performed. In order to monitor and control the gas flow and the gas pressure of the gaseous atmospheres in the gas container, * DI (FH) Florian Ausserer DI (FH) Stefan Klien, Dr Igor Velkavrh V-Research GmbH, 6850 Dornbirn, Austria Pierre Forêt, Linde AG, Linde Gases Division 85716 Unterschleißheim, Germany, DI Alexander Diem V-Research GmbH, 6850 Dornbirn, Austria Investigations of the Sliding and Wear Behaviour in Various Gaseous Atmospheres Using a SRV Testing Apparatus F. Ausserer, S. Klien, I. Velkavrh, P. Forêt, A. Diem* Eingereicht: 15. 6. 2015 Nach Begutachtung angenommen: 5. 7. 2015 Um den Einfluss von verschiedenen industriell eingesetzten Gasen (Ar, N 2 , CO 2 und Luft) auf das tribologische Verhalten von Stahl-Stahl-Paarungen zu untersuchen, wurde ein Versuchsprüfstand nach DIN 51834-1 um eine Gaskapselung für eine Kugel-Scheibe-Kontaktpaarung erweitert. Alle unter Kalibrierungsbedingungen gemessenen Reibkoeffizienten aus Versuchen mit dem modifizierten Aufbau lagen innerhalb dem vom Optimol Instruments GmbH genannten Toleranzband. Die Ergebnisse der durchgeführten Kugel-Scheibe Versuche (Stahl-Stahl) unter Gasbeladung decken sich mit den in der Literatur angeführten Beobachtungen. Schlüsselwörter Gasatmosphären, Luft, Argon, Stickstoff, Kohlendioxid, SRV, Stahl To investigate the influence of various industrially used gases (Ar, N 2 , CO 2 and air) on the tribological behaviour of steel-steel contacts, a test bench according to DIN 51834-1 was extended by gas containment for a ball-on-disk setup. All measured friction coefficients obtained from experiments under calibration conditions using the modified setup were within the tolerance band specified by Optimol Instruments GmbH. The results of the ball-on-disc tests (steel-steel) under gas atmosphere are consistent with the literature. Keywords Gaseous atmospheres, Air, Argon, Nitrogen, Carbon dioxide, SRV, Steel Kurzfassung Abstract T+S_1_16 21.12.15 10: 54 Seite 22 Aus Wissenschaft und Forschung Tribologie + Schmierungstechnik 63. Jahrgang 1/ 2016 23 supplementary sensors were added to the SRV testing apparatus. Furthermore, an O 2 particle sensor was integrated to the system for the detection of any possible impurities, which could be present due to leakage of the gas container. Figure 2 illustrates a schematic of the setup along with the integrated sensors. Correlation of the modified test setup with the Standard DIN 51834-1 procedure (SRV) The adaptation of the SRV testing apparatus (DIN 51834-1) [5] with a gas container has modified the testing system and may have an influence on the determination of the friction and wear values [6]. In order to quantify the influence of the stiffness of the rubber sleeve and the pressure exerted by the gas on the testing apparatus, several pre-tests without and with the sleeve along with an overpressure of 0.02 bar were performed. Furthermore, the method used and presented in a previous study for the characterization of the adhesion behaviour and the analysis of the friction signal using the SRV testing apparatus was chosen [7]. In this method, adhesion values are determined for a wide range of loads. The results of these measurements are presented in Figure 3 in which the coefficient of friction is shown along with the adhesion values for the different loads tested. With the help of this diagram, one may determine if the gas container having a gas overpressure of 0.02 bar has an influence on the standard SRV setup. In Figure 3, it may be easily seen that for loads higher than 150 N, the use of a gas container having a gas overpressure of 0.02 bar has only a negligible influence on the standardized SRV tests. For loads lower than 150 N, the application of the force (F N ) through the SRV testing apparatus must be corrected. For example, for a required normal force F N = 30 N using a gas overpressure of 0.02 bar, the calculated normal force should be raised by a value of 5 N, corresponding to a total normal force F N = 35 N. This results in the fact that the friction coefficient measured by the SRV testing apparatus (µ srv ) has to be corrected using the following equation: N F r e n i a t n o C s a G r f sc i D sc l l a B t e l t u O s a G t e l n I s a G n o C s a G r e n i a t n N F e r ssu e r P s a G r so n e S 2 O e r ssu e r p r e v o r a b 2 0 . 0 r e l l o r t n o C r i A r A 2 O C 2 N Figure 1: Adaptation of the SRV testing apparatus with a gas container Figure 2: Schematic of the experimental setup with additional gas and particle sensors Figure 3: Analyse of the friction signal for the characterization of the adhesion behaviour T+S_1_16 21.12.15 10: 54 Seite 23 Aus Wissenschaft und Forschung 24 Tribologie + Schmierungstechnik 63. Jahrgang 1/ 2016 (1) (2) (3) where µ SRV friction coefficient measured by the SRV apparatus using F N,SRV and F R F R friction force [N] F N,SRV normal force applied by the SRV apparatus [N] µ real coefficient of friction F N real normal force applied on the sample [N] p gas pressure [N/ mm 2 ] A surface of the gas container along the direction of the normal force [mm 2 ] From equations (1) to (3), one obtains: (4) From equation (4), the coefficient of friction µ SRV must be corrected using the following factor: (5) After consideration of the correction procedure presented above, all the measurements of the coefficient of friction obtained in experiments performed using calibration procedures and calibration oil and using the modified SRV testing apparatus lied within the tolerance zone of the declaration of conformity of the SRV’s manufacturer (Optimol Instruments GmbH). From these results, it may be considered that the modification of the SRV testing apparatus has only a small influence on the test setup. Results of experiments (unlubricated) The influence of the various gaseous atmospheres on the tribological ball-disc system was studied using the test parameters listed in Table 1. Figure 4 shows an overview of all results obtained from the various experiments performed at different tempera- Figure 4: Overview of the coefficient of friction and wear results from experiments (unlubricated) F F SRV N R SRV = µ F F N R = µ A p F F SRV N N ⋅ − = A p F F SRV N R ⋅ − = µ or ⋅ − ⋅ = SRV N SRV F A p 1 1 µ µ ⋅ − S R V N F A p 1 1 Table 1: Test parameters for experiments performed (unlubricated) Normal Load 30 N (real normal force applied on the sample) Frequency 20 Hz Stroke 2 mm Test Duration 60 min Material Pairing 100Cr6 - 100Cr6 (ball disc) Gaseous Atmospheres air, Ar, N 2 , CO 2 Temperatures RT (Room Temperature), 200 °C RT 200 °C (a) (b) (c) (d) T+S_1_16 21.12.15 10: 54 Seite 24 Aus Wissenschaft und Forschung Tribologie + Schmierungstechnik 63. Jahrgang 1/ 2016 25 tures and in various gaseous atmospheres. For experiments performed at room temperature, the coefficient of friction and the wear measured in the various gaseous atmospheres (air, Ar, N 2 , CO 2 ) are shown in Figure 4a and Figure 4b, respectively. From these figures, it may be observed that CO 2 has a positive influence on the coefficient of friction and on the wear, in comparison to air, N 2 and Ar. The friction coefficient in Ar and in N 2 is higher than the one measured in air. However, the wear measured in N 2 is lower than the one measured in air and in Ar. For experiments performed at 200 °C, the coefficient of friction and the wear measured in the various gaseous atmospheres (air, Ar, N 2 , CO 2 ) are shown in Figure 4c and Figure 4d, respectively. From these figures, it may be observed that CO 2 has the lowest coefficient of friction and also the lowest wear value when compared to air, Ar and N 2 . On the other hand, Ar and N 2 have similar coefficient of friction and wear values and both are lying between the values of air and CO 2 . Similarly, the wear measured in Ar and N 2 is lower than the one measured in air, but higher than in CO 2 . Coefficient of friction (steady state) The averaged coefficient of friction values measured in air, Ar, N 2 and CO 2 atmospheres are shown in Figure 5. For the calculation of the average coefficient of friction values, 3 experiments for each atmosphere (air, Ar, N 2 , CO 2 ) and temperature (room temperature / 200 °C) were evaluated for a time frame between 33 and 60 minutes. The choice of the time frame was based on the shortest steady-state friction interval which occurred in experiment CO2-2, shown in Figure 7. Here a stable coefficient of friction was measured after a time interval of 33 minutes, while in other experiments steady-state friction behaviour occurred already after shorter time intervals. Figure 6 shows the wear rate (µm/ min) for the air, Ar, N 2 and CO 2 atmospheres. For the calculation of the wear rate, 3 experiments for each atmosphere and temperature were evaluated. Discussion - Results of experiments (unlubricated) The experiments performed at room temperature have shown that the averaged coefficient of friction is at its minimum for the CO 2 atmosphere, followed by the air, N 2 and Ar atmospheres, respectively. Furthermore, as shown in Figure 7, it could be observed that for the CO 2 atmosphere at room temperature, 2 different levels for the coefficient of friction may be measured. When the system is in the lowest stable friction state (Figure 7, test CO2-3), a stable coefficient of friction of ~ 0.4 and Figure 5: Average Coefficient of friction for various gaseous atmospheres and temperatures from unlubricated experiments Figure 6: Wear rates various gaseous atmospheres and temperatures from unlubricated experiments Figure 7: Coefficient of friction at room temperature from unlubricated experiments in CO 2 atmosphere Figure 8: Wear measurement curves at room temperature from unlubricated experiments in CO 2 atmosphere T+S_1_16 21.12.15 10: 54 Seite 25 Aus Wissenschaft und Forschung 26 Tribologie + Schmierungstechnik 63. Jahrgang 1/ 2016 a wear rate of ~ 5 µm/ h are measured. If this stable state is perturbed (Figure 7, tests CO2-1, CO2-2), the coefficient of friction suddenly increases to a higher level of ~ 1.2. This jumping behaviour of the coefficient of friction from a low to a high level was already previously observed; however in correlation with the pressure of the CO 2 gas [4]. This stable state perturbation may also be observed in the wear signal of Figure 8 (tests CO2-1, CO2-2), which could be identified as a negative wear value. The perturbation of the stable state during the wear tests may be caused by the formation of wear particles, which may unfavourably agglomerate themselves between the ball and disc and prevent from formation and/ or contact between CO 2 -reacted tribolayers, which results in the increase of the coefficient of friction and of the wear rate. However, at high temperature (200 °C), no such sudden increase of the coefficient of friction could be observed in the CO 2 atmosphere. In our previous work [8] it was shown that the low friction CO 2 -reacted tribolayers consist of iron carbonate FeCO 3 and/ or iron bicarbonate Fe2(CO 3 ) 3 . Possibly at high temperatures these tribolayers do not form as efficiently and/ or partially lose their low-friction and low-wear properties. At room temperature, the Ar and N 2 atmospheres showed similar coefficient of friction (Figure 5), but their wear behaviour was significantly different (Figure 6). Figure 9 shows the wear scars of balls and discs tested at room temperature in Ar and N 2 atmospheres, respectively. It may be observed that a strong adhesion mechanism took place in the Ar atmosphere in comparison to the N 2 atmosphere, which resulted in an unsteady high wear rate and a high coefficient of friction. Under air atmosphere and at room temperature, an increased formation of red rust (Fe oxide, most probably Fe 2 O 3 [8]) could be observed. This red rust formation influences positively the coefficient of friction; however, it affects negatively the wear behaviour. For a higher temperature (200 °C), a positive effect on the coefficient of friction and on the wear behaviour could be identified when the tests were performed in the Ar, N 2 and CO 2 atmospheres in comparison to the air atmosphere. Results of lubricated experiments (with lubricant) The influence of the different gas atmospheres on the lubricated tribological ball-disc contact was investigated using the experimental parameters listed in Table 2. These parameters correspond to the ones prescribed by the German standard DIN 51834-2 [[6]. Figure 10 and Figure 11 show the results of experimental wear tests. One may observe that the lowest coefficient of friction was measured in air atmosphere (Figure 10), followed by the N 2 , Ar and CO 2 atmospheres respectively. Figure 11 shows the wear behaviour of lubricated contacts under different gaseous atmospheres. The lowest wear value was measured for the N 2 atmosphere, followed by CO 2 , Ar and air atmosphere respectively. 1 mm 200 µm 1 mm 1 mm Ar (RT) N 2 (RT) Ball Disc Figure 9: Ball and disc wear scars for Ar and N 2 atmospheres at room temperature Table 2: Test parameters for lubricated experiments (performed with lubricants) Normal Load 200 N (real normal force applied on the sample) Frequency 50 Hz Stroke 2 mm Duration 120 min Material Pairing 100Cr6 - 100Cr6 (ball disc) Gaseous Atmospheres Air, Ar, N 2 , CO 2 Temperature 120 °C Lubricant FVA 3 oil 1 mm 1 mm 1 mm 200 µm T+S_1_16 21.12.15 10: 54 Seite 26 Aus Wissenschaft und Forschung Tribologie + Schmierungstechnik 63. Jahrgang 1/ 2016 27 Discussion - Results of lubricated experiments (with lubricant) The FVA3 oil used for the present study is a mineral base oil group I with an aromatic content of 6.5 % and contains a small amount of additives. The different wear behaviours for different gas atmospheres shown in Figure 11 may be due to the oil aging caused by oxidation. The oil chemical modifications due to the SRV wear tests were determined by FTIR spectroscopy (Fourier Transform Infrared Spectroscope, Bruker Tensor 27). Fresh oil was used as a reference sample. The FTIR spectra of the oil analysis are shown in Figure 12. Only marginal spectral differences which are indicative of oxidation of the oil (wavenumbers at 1710 cm -1 ) were observed in the FTIR oil spectra. The extent of oil aging was correlated with oil oxidation (starting with the most pronounced aging): air > Ar ~ N 2 > CO 2 > fresh oil (reference). The oil aging is most pronounced for the experiments in air. Slight oil oxidation was also detected for the other gases. Summary By integrating additional gas containment in the experimental SRV4 test rig according to DIN 51834-1, it is possible to create atmospheres with constant gas overpressure. All measured friction coefficients obtained from experiments under calibration conditions using the modified test setup and taking into account the correction factors were within the tolerance band specified by Optimol Instruments GmbH. The results of the ball-disc tests (steel-steel contacts) are consistent with the literature. It could also be shown that two different friction coefficient levels can form under CO 2 atmosphere. When the tribo-system stabilizes itself to the lower level of the friction coefficient, a stable coefficient of friction is measured along with a low wear rate. On the other hand, if the tribo-system is disturbed, the friction coefficient jumps to an upper level and the wear rate increases. Furthermore, for the lubricated tests, it was shown that the gas atmospheres have an influence on the aging of the oil due to oxidation. A positive influence on the friction and wear behaviours of the tribological ball-disc contacts for the N 2 and CO 2 atmospheres could be observed. Acknowledgments This work was funded by the „Austrian COMET-Program“ under the scope of K2 XTribology and was carried out at V-Research GmbH and AC 2 T research GmbH in cooperation with Linde AG / Linde Gas Division. Figure 10: Coefficient of friction for the lubricated experiments with FVA 3 oil at room temperature Figure 11: Wear for the experiments with FVA 3 oil at room temperature Figure 12: FTIR spectra of the FVA3 oil before and after the tests for air, Ar, N 2 and CO 2 atmospheres T+S_1_16 21.12.15 10: 54 Seite 27 Aus Wissenschaft und Forschung 28 Tribologie + Schmierungstechnik 63. Jahrgang 1/ 2016 References [1] G.W. Stachowiak and A.W. Batchelor, Engineering Tribology, Third Edition, Elsevier Butterworth-Heinemann (2005) Oxford, UK. [2] M. Qiu, L. Chen, Tribological characteristics of chromium steels in three various atmospheres under high-speed conditions, Wear, Vol. 268 (2010), p. 1342-1346. [3] A.F. Smith, Influence of environment on the unlubricated wear of 316 stainless steel at room temperature, Tribology International, Vol. 19 (1986), p. 3-10. [4] X. Wu, P. Cong, H. Nanao, I. Minami, S. Mori, Tribological behaviors of 52100 steel in carbon dioxide atmosphere, Tribology Letters, Vol. 17, No. 4 (2004), p. 925- 930. [5] DIN 51834-1, Tribologische Prüfungen im translatorischen Oszillations-Prüfgerät, Teil1: Allgemeine Arbeitsgrundlagen, Beuth Verlag, Berlin, Germany, 2004. (German Standard DIN 51834-1: Tribological experiments in a translating-oscillating testing apparatus, Part 1: General working principles, in German). [6] DIN 51834-2, Tribologische Prüfungen im translatorischen Oszillations-Prüfgerät, Teil 2: Bestimmung von Reibungs- und Verschleißmessgrößen für Schmieröle, Beuth Verlag, Berlin, Germany, 2004. (German Standard DIN 51834-2: Tribological experiments in a translating-oscillating testing apparatus, Part 2: Determination of friction and wear data for lubricating oils, in German). [7] S. Klien, A. Ristow, J. Rébel, R. Jisa, A. Diem, Charakterisierung des Haftreibverhaltens und Reibsignalanalyse am Schwing-Reib-Verschleiß-Prüfstand, GFT 2012. (Characterization of the static friction behavior and analysis of the friction signal on vibration-friction wear test apparatus, in German) [8] I. Velkavrh, F. Ausserer, S. Klien, J. Brenner, P. Forêt, A. Diem, The effect of gaseous atmospheres on friction and wear of steel-steel contacts, Tribology International, Vol. 79 (2014) 99-110. Firmenportrait V-Research sichert den Spielraum. Für die richtigen Entscheidungen. Je komplexer die Anforderungen, desto wichtiger ist Spielraum für das Auffinden neuer Lösungswege. Forschungs- und Entwicklungspartnerschaften wie sie V- Research seit vielen Jahren erfolgreich betreibt, schaffen neue Bewegungsfreiheit im Denken und Handeln. V-Research steht für industrieorientierte Forschung und Entwicklung in zwei Arbeitsgebieten: • Tribo Design - die Optimierung tribologisch beanspruchter Systeme sowie • Design Automation - die Automatisierung von Konstruktions- und Entwicklungsprozessen In diesen Themenfeldern übernehmen wir Forschungs- und Entwicklungsprojekte, bieten Industriebetrieben Technologieberatung an und stehen ihnen bei der Umsetzung ihrer Innovationsziele mit unserer Methodenkompetenz, unserem technologischen Wissen und unserer Erfahrung in der angewandten Forschung und Entwicklung zur Seite. Unser Ziel ist es, den Innovationsvorhaben unserer Auftraggeber zum wirtschaftlichen Erfolg zu verhelfen. Beispiel einer erfolgreichen Kooperation: ZIMM und V-Research Die ZIMM Maschinenelemente GmbH + Co KG ist ein international tätiges Unternehmen mit mehr als 35 Jahren Erfahrung in der Antriebstechnik. ZIMM vertreibt und entwickelt hochwertige Antriebselemente für präzise elektromechanische Verstellungen von linearen Bewegungen - und das bei Lasten bis zu 1.000 kN pro Getriebe. Diese Hubgetriebe werden in automatisierten Produktionsanlagen auf der ganzen Welt eingesetzt: von der Airbus-Produktion in Hamburg bis zur Herstellung von Flachbildschirmen von Samsung und LG in Korea. Wenn es um Forschung und Entwicklung geht, arbeitet ZIMM seit Jahren mit V-Research zusammen. Schwerpunkt der Arbeit von V-Research ist dabei die Optimierung der bestehenden Werkstoffpaarungen, Oberflächen, Beschichtungen und Schmierstoffen sowie die Unterstützung bei der Auswahl und Evaluation ganz neuer Werkstoffkonzepte. So konnte z. B. beim Herzstück der Spindelhubgetriebe, der Schneckenverzahnung und dem Bewegungsgewinde, durch ein umfassendes Neudesign die Lebensdauer des Hubsystems um 55 % und die Traglasten um 30 % gesteigert werden bei gleichzeitiger Verringerung der Herstellungs- und Materialkosten um 30 %. V-Research GmbH Industrielle Forschung und Entwicklung CAMPUS V Stadtstraße 33 · 6850 Dornbirn · Austria +43 5572 394159 · E office@v-research.at www.v-research.at v research Industrielle Forschung und Entwicklung T+S_1_16 21.12.15 10: 54 Seite 28