Estimation of real contact area during sliding friction from interface temperature Estimation of real contact area during sliding friction from interface temperature Sung Keun Chey, Pengyi Tian, and Y[.]
Estimation of real contact area during sliding friction from interface temperature Sung Keun Chey, Pengyi Tian, and Yu Tian Citation: AIP Advances 6, 065227 (2016); doi: 10.1063/1.4955183 View online: http://dx.doi.org/10.1063/1.4955183 View Table of Contents: http://aip.scitation.org/toc/adv/6/6 Published by the American Institute of Physics AIP ADVANCES 6, 065227 (2016) Estimation of real contact area during sliding friction from interface temperature Sung Keun Chey, Pengyi Tian, and Yu Tiana State Key Laboratory of Tribology, Tsinghua University, Beijing 100084, China (Received 14 April 2016; accepted 22 June 2016; published online 29 June 2016) Frictional heat is one of the most important topics in tribological research The real contact area of the frictional pair plays a significant role in accurately estimating the interface temperature, which is closely related to the frictional heat However, conventional methods for measuring the contact area, such as constriction resistance measurements, are not suitable for dynamic frictional motion because of the electrical and thermal interferences at the contact region In this study, a novel method is presented for estimating the real contact area during sliding friction First, the average interface temperature was experimentally measured by the dynamic thermocouple method Then assuming that the total frictional heat power is constant, the measured temperature was used as a constraint to determine the contact area in a finite element model, giving an estimation for the real contact area The calculation results show that the real contact area increases with increasing normal load as predicted by contact theories, and decreases with increasing sliding speed, which could be attributable to the contact dynamics of asperities at the interface Additionally, the limits of the proposed method is discussed C 2016 Author(s) All article content, except where otherwise noted, is licensed under a Creative Commons Attribution (CC BY) license (http://creativecommons.org/licenses/by/4.0/) [http://dx.doi.org/10.1063/1.4955183] When two solids come into contact, the real contact area composed of interacting asperities is usually a very small fraction of the apparent contact area To describe this, Bowden and Tabor,1 Archard,2 and Greenwood and Williamson3 have established the theories of elastic and plastic contacts As a complement to these theories, various experimental methods have been developed to measure the real contact area One of the most common techniques is the electrical contact resistance method, which was first development by Greenwood4 and Holm.5 When an electrical current flows across a pair of contacting bodies, the current drop across the bodies is measured to determine the constriction resistance This resistance can be related to the real contact area, as the resistance depends on the inverse of the circular contact radius.6,7 When the contacting bodies are in dynamic frictional motion, however, the frictional heat at the interface can generate a significant potential gradient via thermoelectric effect This potential gradient can interfere with the constriction resistance measurements Another technique is the thermal contact resistance (or conductance) method In this method, an external heat source is applied to one of the contacting bodies and the heat flow or the temperature distribution across the contact is measured using infrared detectors or thermocouples.8,9 Similar to the electrical contact resistance, heat flow through the interface depends on the contact radius of the separate asperities, so the thermal contact resistance can be related to the real contact area.10–13 During sliding friction, however, frictional heat will interfere with the external heat flow, and the resulting temperature measurement is a mix of external and frictional heat In this case, the contact resistance cannot be measured accurately Additionally, optical methods are widely used to directly observe the real contact area of transparent frictional pairs.14–17 In optical methods, only the transparent materials can be used, and the a Corresponding: tianyu@tsinghua.edu.cn 2158-3226/2016/6(6)/065227/5 6, 065227-1 © Author(s) 2016 065227-2 Chey, Tian, and Tian AIP Advances 6, 065227 (2016) question persists whether the optically measured areas are equivalent to the mechanical contact area due to the limited resolution of the optical devices Based on the above statements, conventional methods have limits in the application for dynamic frictional motion Considering this, a novel method is presented for estimating the real contact area from interface temperature rise during dry sliding friction in this study The proposed method uses the frictional heat generated during the dynamic frictional motion itself Therefore, this method avoids the problems that conventional methods face, which are caused by frictional heat It is well-known that most of the mechanical energy of motion is converted into heat during friction.18–20 Here, all of the frictional work is assumed to be transformed into heat for simplicity: Qtotal = µFz ν (1) where Qtotal is total frictional heat power, µ is the average friction coefficient, Fz is the vertical force, ν is the average sliding speed The frictional heat is dissipated through the real contact area, which is composed of many small asperity contacts According to Archard,21 when the individual asperity contacts are closely packed together, it can be considered that the frictional heat dissipates through an effective area whose size is approximately equal to the sum of the individual asperity contact areas In other words, if the asperities are clustered into a single region, the frictional heat dissipation through individual asperities is equivalent to the same heat being dissipated through a general region consisting of these asperities Using an electrical analogy, Cocks22 gave theoretical and experimental evidences to support this approximation Thus, the effective area Aeff , through which the frictional heat dissipates, can be used to estimate the real contact area Then the total frictional heat power per unit area q is: q = µFz ν/Aeff (2) In the past, Eq (2) was used as an input for finite element calculations for the interface temperature rise during sliding friction.23 In this study, Aeff is assumed to be a constant fraction k, of the wear area, Awear: Aeff (t) = k · Awear(t) (3) In a finite element model, Eq (2) and Eq (3) can be used to model the frictional heat dissipation And by assuming that total frictional heat power given in Eq (1) is constant, the experimentally measured interface temperature can be used as a constraint to calculate the effective area The schematic of the friction experiment setup is shown in Fig The dynamic thermocouple method is used to measure the average interface temperature As developed by Shore to measure temperature rise for cutting tools,24 the dynamic thermocouple method uses two dissimilar metals as a friction pair to measure the thermoelectric potential (emf ) created at the sliding interface as a result of frictional heating In this case, the friction interface between the two metals itself becomes the hot junction of a thermocouple Using the wires made of the same metals as the friction body, a reference junction can be formed An ice bath is typically used for the reference junction By adding a potentiometer to the circuit, the thermoelectric potential between the friction interface and the ice FIG Schematic of the friction experiment 065227-3 Chey, Tian, and Tian AIP Advances 6, 065227 (2016) bath is measured The emf measured using the dynamic thermocouple method corresponds to the average temperature of the friction interface.20,25 A pin-on-plate friction experiment was conducted on a universal micro-tribotester (UMT-2, Center for Tribology, Inc.) using the reciprocating sliding mode Applied load was between 2.5 and 12.5 N, and average sliding speed was between 0.1 and 0.31 m/s The experiment was conducted in a room temperature of 20 ◦C and a relative humidity of 10%, and each run lasted 180 seconds For the dynamic thermocouple setup, the emf values were measured using the UJ33D-1 potentiometer (Shanghai Zhengyang Instrument Works), and an ice bath (0◦C) was used as the reference junction For the specimen, 304 stainless steel was used for the stationary pin and N4 nickel was used for the moving plate The stainless steel specimen was shaped into a cylinder-type pin with a spherical tip to ensure a point contact with the lower body The radius of the steel pin’s hemispherical section was mm and the surface roughness (Ra) was 2.5 µm The nickel plate had a dimension of 40×40×5 mm, and a surface roughness of µm To convert the measured emf values into temperature values, calibration was carried out by using an electric furnace By placing the hot junction of the stainless steel and nickel wires inside the furnace and using an ice bath for the reference junction, the emf values were matched to the corresponding temperature The result is shown in Fig 2(a) The interface temperature measurement results are shown in Fig 2(b) and 2(c) The measured average coefficient of friction26 ranged from 0.247 to 0.569 As shown in Fig 2(b), the interface temperatures quickly reached a stable value after the beginning of friction Then taking the average values of the stable interface temperature, the results with increasing load and sliding speed are shown in Fig 2(c) Having measured the average interface temperature, the next task is to determine Awear as a function of time This was achieved in the following way: first, the wear area of the steel pin was measured at different time intervals using a microscope in a separate friction experiment (load = N, speed = 0.155 m/s) Then Awear(t) was obtained by curve fitting Meanwhile, the final wear area of each steel pin used in the friction experiment was measured after sliding for 180 seconds Finally, Awear(t) was modified to fit the measured wear areas, Awear(t = 180 s) for each experiment parameters The results are shown in Fig The wear area of the steel pin increased steadily with increasing load and sliding speed The finite element analysis was conducted using COMSOL Multiphysics 5.2 A time-dependent 3D model of the friction experiment was constructed using the geometric conditions of the experiment The frictional heat power was modelled by a boundary heat source prescribed at the interface, FIG Experiment results (a) Calibration result using an electric furnace (b) Typical dynamic thermocouple measurements (c) Average interface temperature rise at constant sliding speed and constant load conditions 065227-4 Chey, Tian, and Tian AIP Advances 6, 065227 (2016) FIG Wear area curve-fit result (a) Load effect on wear area at a constant sliding speed of 0.207m/s; (b) Sliding speed effect on wear area at a constant load of 10 N and Eq (2) was used as its magnitude Simultaneously, the size of the boundary heat source was set to increase with time, as given in Eq (3) This guaranteed that the total frictional heat power of the boundary heat source, as given in Eq (1), remains constant at all times At the interface, the pin and the plate mesh were set to share the nodes, which guaranteed that the temperature across the two bodies is continuous For details of the model setup, please refer to the supplementary material.26 To calculate the effective area, the constant of proportionality k in Eq (3) was varied until the calculated average interface temperature of the model matched the experimentally measured temperatures The results are shown in Fig The calculated k ranged between 0.0099 and 0.0495, and the effective area was between 0.0177 and 0.0959 mm2 Aeff increased with increasing load, which is expected as many contact theories predict it.2,6 On the other hand, the effective area decreased with increasing sliding speed, which can be attributed to the faster separation of the contact asperities with higher sliding speed In summary, a novel method for estimating the real contact area during sliding friction was presented using the interface temperature Assuming that the real contact area is a constant fraction of the apparent wear area, the measured temperatures by dynamic thermocouple method was used as constraints in the finite element model to calculate the effective area through which frictional heat dissipates The calculated constant k ranged from 0.0099 to 0.0495, while the real contact area increased with increasing load as predicted by contact theories, and decreased with increasing sliding speed, which could be explained by the faster separation of the asperity contacts The proposed method for estimating the real contact area uses the frictional heat generated during the dynamic frictional motion itself So it effectively overcomes the limits of conventional methods in dynamic motion test which are caused by the inaccuracy due to the interfacial temperature increase The real contact area estimated using the proposed method can be used to study the fundamental mechanism of friction, and to accurately determine physical properties such as the interface temperature FIG Effective contact area (a) constant sliding speed (0.207m/s) (b) constant load (10N) 065227-5 Chey, Tian, and Tian AIP Advances 6, 065227 (2016) Admittedly, this method also has some limitations that it requires conductive and dissimilar materials for the sliding bodies, and that it is only suitable for dry friction and boundary lubrication where solid-solid contact exists When the shear velocity is high enough that mixed or hydrodynamic lubrication exists, the solid contact area is almost zero that the thermoelectric potential cannot be measured any more Also, in case of boundary lubrication, the additional temperature rise contributed by viscous dissipation must be separated from the temperature rise from the contact shearing to improve the accuracy Thus, an optimized model which accounts for these factors is to be adopted to overcome such difficulties This work is supported by the Natural Science Foundation of China (Grant No 51323006 and 51425502) F P Bowden and D Tabor, Proc R Soc Lon A Mat 391 (1939) J F Archard, Proc R Soc Lon A Mat 243, 190 (1957) J A Greenwood and J B P Williamson, Proc R Soc Lon A Mat 295, 300 (1966) J A Greenwood, Brit J Appl Phys 17, 1621 (1966) R Holm, Electrical contacts handbook (Splinger, 1958) V Popov, Contact mechanics and friction: physical principles and applications (Springer Science & Business Media, 2010) L Kogut and K Komvopoulos, J Appl Phys 94, 3153 (2003) C Fieberg and R Kneer, Int J Heat Mass 51, 1017 (2008) M Rosochowska, R Balendra, and K Chodnikiewicz, J Mater Process Tech 135, 204 (2003) 10 M A Lambert and L S Fletcher, J Thermophys Heat Tr 11, 129 (1997) 11 M A Lambert and L S Fletcher, J Heat Transf 124, 405 (2002) 12 N Laraqi and A Bairi, Int J Heat Mass 45, 4175 (2002) 13 P Sadowskiand and S Stupkiewicz, Wear 268, 77 (2010) 14 W G Sawyer and K J Wahl, MRS Bull 33, 1145 (2008) 15 K J Wahl and W G Sawyer, MRS Bull 33, 1159 (2008) 16 A Ovcharenko, G Halperin, I Etsion, and M Varenberg, Trib Lett 23, 55 (2006) 17 B A Krick, J R Vail, J R Persson, and W G Sawyer, Trib Lett 45, 185 (2012) 18 H Uetz and J Föhl, Wear 49, 243 (1978) 19 D A Rigney and J P Hirth, Wear 53, 345 (1979) 20 B Bhushan, Modern Tribology Handbook (CRC press, 2000), Vol 21 J F Archard, Wear 2, 438 (1959) 22 M H Cocks, Ph.D thesis, University of Reading, Reading, 1953 23 F E Kennedy, Y Lu, and I Baker, Tribol Int 82, 534 (2015) 24 H Shore, J Wash Acad Sci 15, 85 (1925) 25 M J Furey, ASLE Trans 7, 133 (1964) 26 See supplementary material at http://dx.doi.org/10.1063/1.4955183 for experiment results and FEM model setup ... ADVANCES 6, 065227 (2016) Estimation of real contact area during sliding friction from interface temperature Sung Keun Chey, Pengyi Tian, and Yu Tiana State Key Laboratory of Tribology, Tsinghua... higher sliding speed In summary, a novel method for estimating the real contact area during sliding friction was presented using the interface temperature Assuming that the real contact area is... application for dynamic frictional motion Considering this, a novel method is presented for estimating the real contact area from interface temperature rise during dry sliding friction in this study