Advertisement

Tribology Letters

, 66:142 | Cite as

In Situ Observation of Heat Generation Behaviour on Steel Surface During Scuffing Process

  • Yasuo Matsuzaki
  • Kazuyuki Yagi
  • Joichi Sugimura
Open Access
Original Paper
  • 781 Downloads

Abstract

In the current study, the heat generation behaviour at plastic flow area in the contact area during a scuffing process was observed in situ using a monochrome high-speed camera that detected both visible and near-infrared light. The scuffing test was conducted using a pin-on-disc test rig comprising a rotating sapphire disc and a fixed martensitic steel pin. Engine oil was supplied as a lubricant. The images captured by the high-speed camera clearly showed changes in the contact area and the heat generation behaviour. It was found that the heat generation behaviour could be classified into three stages during scuffing. In the first stage, local heat generation occurred intermittently at local severe contact points, such as the passing of transfer layers on the sapphire disc and the trailing edge of the contact area. In the second stage, heat generation occurred intermittently over larger areas in which heat had been generated until that time. When the entire contact area had previously generated heat, heat generation became continuous throughout the contact area in the third stage. During the third stage, the contact area was increased rapidly, which caused catastrophic failure. These results highlight important questions regarding material phenomena that remain to be investigated.

Keywords

Scuffing In situ observation Heat generation Plastic flow Subsurface 

1 Introduction

Scuffing is a catastrophic surface failure occurring at the contact areas between the sliding surfaces of machine parts operated under high sliding speeds and high loads, such as piston–cylinder pairs, journal bearings in crankshaft–connecting rod pairs, and gears. At present, decreases in oil viscosity and downsizing of machine parts are necessary to improve the fuel efficiency of machinery. Recently, SAE 0W-8 is available in the market [1]. With these changes, lubricated areas are subjected to more severe conditions that induce surface failure, attributed to the direct contact between mating surfaces. As scuffing occurs, the friction and temperature both increase drastically, and the mating surfaces adhere to each other, which can cause machinery shutdown in the worst-case scenario. Therefore, the mechanisms of scuffing initiation and progression must be understood to achieve highly efficient and safe machine operation.

Scuffing seems to be initiated by the breakdown of the protective films intended to prevent direct contact between mating surfaces. Based on the hydrodynamic lubrication theory, it has been proposed that as the friction surface temperature exceeds a critical temperature, the hydrodynamic fluid film breaks, resulting in scuffing [2, 3, 4, 5, 6, 7]. It is also suggested that oil starvation due to the agglomeration of wear particles can cause the breakdown of the fluid film, leading to scuffing [8, 9]. From the perspective of boundary lubrication, it was proposed that scuffing occurs after the thermal and mechanical removal of chemical reaction films originating from the additives [10, 11]. The removal and reduction of oxide films may also lead to seizure [12, 13, 14]. The magnitude and orientation of the surface roughness are key factors influencing for fluid film breakdown [15] and lubricant flow [16, 17]. Although the breakdown of surface protective films demonstrably leads to scuffing, it is only one of several processes in scuffing. The next stage of the scuffing process after protective film breakdown is the material failure process of the mating surfaces.

After initiation, scuffing does not always induce catastrophic failure. When scuffing does not progress to the final stage, the process is suppressed by the stable conditions termed “running-in” [18]. Yoon et al. [19] constructed a transition map characterising the three regions of mild wear; microscuffing, which means friction spikes preceding catastrophic scuffing failure; and catastrophic scuffing failure, depending on the load and sliding speed. Regarding the progression of scuffing, Zhang et al. [20] proposed that scuffing progresses when heat generation by the scuffing of an asperity affects adjacent asperities. Hershberger et al. [21] suggested that scuffing occurs when the thermal softening of the friction surface exceeds the work hardening; at this point, the remaining stable contact area is continuously deformed because the softened area supports lower normal and shear forces. Ajayi et al. [22] suggested that scuffing progresses continuously when the local heat generation induced by scuffing exceeds the heat dissipation. However, Markov et al. [23] suggested that only friction heating and mechanical property changes at the bulk or macroscale level can influence scuffing progression, because microscale plastic deformation of asperities has insufficient time for heating and changing the contact surface properties. In addition to changes in properties, changes in the morphology of the friction surface also affect the scuffing progression. Jiajun et al. [24] proposed that plastic deformation at the asperity level continues and promotes severe scuffing if the surface smoothness cannot be improved after the plastic deformation. In contrast, the rheological properties of lubricant oil may have little effect on the scuffing progression. Piekoszewski et al. [25, 26] indicated that the critical load for scuffing initiation was increased with increases in oil viscosity, but that the oil viscosity did not affect the scuffing load for the final stage of scuffing.

A specific layer has been observed on scuffed surfaces. Rogers [27, 28] characterised the scuffed surface and subsurface based on the microscopic appearance, etching, and hardness testing, and reported that the scuffed surface was covered by a white layer that was harder than the bulk material. Using transmission electron microscopy, Torrance et al. [29] found a very fine martensite structure in the white layer. Leng et al. [30] proposed that the white layer was a form of microscopic friction weld between asperities and that its formation mechanism was similar to that of adiabatic shear bands [31, 32, 33]. Ajayi et al. observed the microstructural changes in the near-surface material before, during, and after scuffing, and indicated that a severe deformation layer was formed during scuffing [34]. Kuznetsov et al. [35] observed a nanocrystalline structure on the friction surface under very high load and proposed that surface structure modification containing nanosized grain–subgrain structures induced instability of the friction surface against shear deformation by grain boundary slippage, which caused intense adhesion and transfer of metal onto the counter surface that could lead to scuffing.

As described above, much work has been conducted to investigate the scuffing mechanism from several perspectives. During the last decade, the authors employed a different approach from that used in most previous studies to investigate the scuffing process more clearly. The authors developed in situ observation systems for the contact area comprising a rotating sapphire disc and a stationary steel pin. Yagi et al. [36] directly observed the contact area, reporting local plastic flow on the steel surface during scuffing. Li et al. [37] indicated that plastic flow firstly occurred at the point where the wear particles accumulated. Yagi et al. [38, 39] conducted simultaneous in situ observation and synchrotron X-ray diffraction analysis of the contact area. The capturing system used in their work comprised a visible-light camera with a Xenon flashing lamp as the light source and a near-infrared-light camera. They reported that the plastic flow area showed a high temperature exceeding 1000 °C and that the martensite–austenite phase transformation occurred repeatedly during scuffing. Izumi et al. [40] used synchrotron X-ray diffraction analysis to investigate the relationship among the frequency of heat spot, austenite fraction, and wear volume during scuffing with four steels. Matsuzaki et al. [41] conducted in situ fast and long-term observation using four cameras with freely adjusted time intervals. They found that plastic flow initially occurred on the passing of a transfer layer and that the duration of plastic flow was a few milliseconds. As stated above, instantaneous plastic flow and heat generation were observed; however, the continuous behaviour of plastic flow could not be observed, because the frame rates of the cameras used in the previous studies [36, 37, 38, 39, 40, 41] were insufficient.

In the current study, in situ observation was performed using a monochrome high-speed camera that captured both visible and near-infrared light. This observation system could capture simultaneously and precisely the change of the contact area and the heat generation behaviour. The scuffing process was investigated using this advanced in situ observation system.

2 Experimental Procedure

Figure 1 shows the in situ observation system for the scuffing test used in the current study. The system contains two parts of the tribometer and the capturing system, located above the tribometer.

Fig. 1

Schematic of test apparatus

The tribometer for the scuffing test is the pin-on-disc type, in which the contact area is created between a rotating sapphire disc and a fixed steel pin with curvature at the top. Sapphire is used for the disc material because it can transmit visible and near-infrared light. The diameter and thickness of the sapphire disc are 50.0 and 10.0 mm, respectively. The material of the steel pin is JIS SUJ2, which is equivalent to AISI52100. The diameter of the steel pin is 4 mm, as shown in Fig. 2, and the radius of curvature of the top surface is 12.7 mm. Figure 3 shows the Vickers hardness of the SUJ2 pin over a broad temperature range. The hot hardness was measured three times at each temperature, and Fig. 3 shows the averaged values. The sapphire disc is rotated by an AC servomotor through a timing belt. The pin is loaded against the rotating sapphire disc using an air cylinder and a load cell is used to measure the applied load. The sliding track radius on the disc is 16 mm. Because the motion range of the steel pin is limited, the steel pin can be worn out by up to 2 mm in height. When friction occurs at the contact area, the upper part of the tribometer moves in the horizontal direction along linear guides and the friction is measured by a load cell for friction. Type K thermocouple is embedded in the steel pin at the depth of 1 mm. The applied load, friction, and temperature of the embedded thermocouple are simultaneously measured. As a lubricant oil, SAE 0W-8 multigrade engine oil was used because scuffing becomes important in such low-viscosity engine oil. Table 1 shows the additives and properties of the engine oil. Table 2 shows the mass concentrations of elements contained in the engine oil. The lubricant oil is supplied toward the leading edge of the contact area by a rotary pump with controlled temperature. Before supplying the lubricant oil, it was heated in an oil bath established outside of the tribometer.

Fig. 2

Schematic of pin specimen

Fig. 3

Vickers hardness of SUJ2 pin at different temperatures

Table 1

Additives and properties of engine oil

SAE viscosity grade

0W-8

Viscosity index improver

Polymethacrylate

Anti-wear additive

Zinc dialkyldithiophosphate (ZDDP, Secondary type)

Metallic detergent

Overbased calcium salicylate

Antioxidant

Amine type, Phenol type

Friction modifier

Molybdenum dithiocarbamate (MoDTC)

Antifoaming agent

Silicone

Kinetic viscosity, mm2/s

20.6 at 40 °C

4.7 at 100 °C

Table 2

Mass concentrations of elements contained in engine oil

Element

Ca

Mo

P

Zn

S

Mass concentration in engine oil, wt%

0.20

0.07

0.08

0.09

0.22

The capturing system consists of a microscope, halogen lamp, and high-speed camera. The high-speed camera is a monochrome complementary metal-oxide semiconductor (CMOS) camera that detects light in the wavelength range from 350 to 1100 nm. The camera captures reflected visible images of the halogen lamp as well as high temperatures exceeding 600 °C. Its pixel size is 20 × 20 µm/pixel with 256 brightness levels. Figure 4 shows a representative image of the contact area during scuffing without a light source for the high-speed camera. Heat generation at the plastic flow area is imaged as the white area, indicating high brightness.

Fig. 4

Captured image during scuffing without light source

The test was conducted at a load of 5000 N and a constant rotational speed of 1790 rpm (= a constant sliding speed of 3.0 m/s). Then, the sapphire disc rotates one time in 33.5 ms. Stepwise [7, 9, 10, 11, 12, 13, 19, 21, 22, 29, 30, 34, 36, 37] or contentious [25, 26] increases in load are generally chosen for scuffing tests. Gradually increasing the load until scuffing occurs is suitable for determining the threshold load for scuffing. By contrast, if an oil supply problem or overload occurs within a short period in the contact area, the lubrication condition suddenly changes from hydrodynamic lubrication to boundary lubrication. In this severe case, it is more important to understand the scuffing phenomenon. Therefore, in the current study, shock loading was chosen as the loading process to simulate drastic variations in the contact area. The supply temperature of the lubricant oil was set to 80 °C. The exposure time of the high-speed camera was set to 100 µs and the frame rate of the high-speed camera was set to 3000 fps. Three tests were conducted under the same test conditions to confirm the repeatability of the results.

3 Results

Figure 5 shows the variations in the friction coefficient and applied load during each test. The result of Test 1 in Fig. 5a is explained below. The applied load reaches a set value of 5000 N in less than 0.2 s. After the test begins, the friction coefficient is approximately 0.025 and low in fluctuation. From approximately t = 6 s, the fluctuation of the friction coefficient becomes larger. From approximately t = 9 s, the friction coefficient is increased dramatically, indicating the occurrence of scuffing. The friction coefficient reaches the maximum value of 0.35 at the test time of t = 9.016 s. Afterward, the friction coefficient is decreased to approximately 0.23 at the test time of approximately t = 9.1 s. During scuffing, the pin is worn rapidly, and the pin holder moves upward. When the motion of the pin reaches the moveable range, the friction coefficient is decreased to zero. In comparison with the results of Tests 1, 2, and 3 in Fig. 5a–c, although the times until scuffing occurs are different, the behaviours of the variations in the friction coefficient are similar.

Fig. 5

Variations in friction coefficient and applied load during a Test 1, b Test 2, and c Test 3

Figure 6 shows images of the contact area captured before the initiation of scuffing in Test 1. The image at the test time of t = 4.1780 s is the first of the captured images. It is observed that a black film is formed in the contact area at this time. At the test time of t = 6.1780 s, the black film remains with no change in the size of the contact area and a low friction coefficient. Moreover, transfer layers on the sapphire disc are seen in some areas. At the test time of t = 6.4077 s, the removal of the black film begins from the centre of the contact area. As the black film is gradually removed, the fluctuation of the friction coefficient is increased, as shown in Fig. 5a. At the test time of t = 7.4723 s, the surface at the trailing side of the contact area flows along the sliding direction. At the test time of t = 8.7993 s, most of the black film is removed and heat generation is first observed in the area over which one transfer layer passes.

Fig. 6

Removal of black film and scuffing initiation in Test 1

Figure 7 shows images of the contact area captured in the early stage of the scuffing process in Test 1. In the monochrome images at the top of each sub-figure, the reflected light of the halogen lamp and the near-infrared light originating from heat generation are combined. To reveal the behaviour of heat generation in the contact area, each monochrome image was binarized using a threshold brightness value. The binarized images are shown below the monochrome images. The threshold brightness value was set to 220 in the current study to distinguish the heat generation area. It is found that heat generation occurs intermittently at local severe contact points with the passing of transfer layers and at the trailing edge during the early stage of the scuffing process. At the test time of t = 8.8097 s, heat generation is observed after the passing of a transfer layer, similar to that at t = 8.7993 s. At the test time of t = 8.8123 s, heat generation is seen at the trailing edge of the contact area, regardless of the passing of transfer layers. This stage shown in Fig. 7 is called the first stage. Figure 8 shows larger images of the contact area at the test time of t = 8.8097 s. Heat generation area was very small compared to the contact surface at the initiation of scuffing.

Fig. 7

Monochrome and binarized images in the first stage of Test 1

Fig. 8

Larger images of the contact area shown in Fig. 7b

Figure 9 shows the expansion of the heating areas by focusing on a certain transfer layer in the first stage with both monochrome and binarized images in Test 1. The time interval between the adjacent images in each row is 0.3 ms, indicating one frame. The interval between the rows comprising paired monochrome and binarized images indicates one rotation of the sapphire disc. At the test time of t = 8.7990 s, it is observed that a transfer layer passes through the contact area and heat generation occurs at the trailing edge. At the test time of t = 8.7993 s, heat generation occurs with the passing of the transfer layer at the centre of the contact area. At the test time of t = 8.7996 s, where the sapphire counter surface has moved by 0.9 mm since t = 8.7993 s, the heat generation area has not expanded in this duration. At the test time of t = 8.9003 s in Fig. 9k, where the sapphire disc has rotated three times since t = 8.7993 s in Fig. 9b, heat generation occurs at the passing of the transfer layer, as in Fig. 10b. The heat generation area shown in Fig. 9k is larger than that in Fig. 9b. During the first stage of the scuffing process, heat generation occurs, and the generation area is expanded as the same transfer layer passes the contact area.

Fig. 9

Expansion of heat generation area during several sequential passing of transfer layer in Test 1

Fig. 10

Monochrome images in the second and third stages in Test 1

Figure 10 shows the variations in the contact area after the first stage of scuffing in Test 1. The three images in each row indicate the change in the contact area during one rotation of the sapphire disc. The interval between the rows indicates one rotation of the sapphire disc. It is observed that the blotchy darker areas on the contact area are synchronised with the rotation of the disc, as seen in Fig. 10a–i. From this result, it can be concluded that the conformity between the surfaces depends on the location of the sapphire disc. Figure 11 shows the binarized images converted from the images in Fig. 10. It is found that the intermittent heat generation is eventually changed into continuous heat generation. At the test time of t = 8.9423 s, local heat generation as shown in Fig. 11a occurs at the passage of a transfer layer, as shown in Fig. 10a, likely in the first stage of scuffing. At the test time of t = 8.9486 s, however, heat generation occurs over larger areas, as shown in Fig. 11b, regardless of the passage of transfer layers as shown in Fig. 10b. At the test time of t = 8.9533 s, heat generation is observed at the edge of the contact area. Between Fig. 11a, i, it is found that heat generation occurs intermittently over a larger area and that the heat generation area increases gradually. This period is called the second stage. From the test time of t = 9.0440 s, heat generation becomes continuous throughout the contact area, as shown in Fig. 11j–l; the contact area expands rapidly, as shown in Fig. 10j–l. This stage is called the third stage, which is the final stage of the scuffing process.

Fig. 11

Binarized images in the second and third stages in Test 1

Figure 12 shows histograms of the area ratio of the brightness value on the contact areas before and during scuffing. Figure 12a was calculated from the contact area shown in Fig. 6d at the test time of t = 7.4723 s, which is before scuffing. Figure 12b was calculated from the contact area shown in Fig. 10l at the test time of t = 9.0553 s, which is during scuffing. There are very few instances of the number of brightness value over 220 before scuffing and the intensity during scuffing is much stronger.

Fig. 12

Area ratio of brightness value on the contact areas a calculated from Fig. 6d, which is before scuffing, and b that calculated from Fig. 10l, which is during scuffing

Figure 13 shows the variations in the contact area during the test in Test 1, as calculated from the captured images. During the removal of the black film as shown in Fig. 6, the contact area is slightly increased. In the scuffing process, the area is increased rapidly. Figure 14 shows the variations in the heating area and temperature during the test of Test 1, Test 2, and Test 3. The heating area was measured by calculating the number of white pixels in the binarized images. In all results of Tests 1, 2, and 3, as shown in Fig. 14a–c, during the removal of the black film, heat generation cannot be observed; however, when scuffing occurs, the heating area increases rapidly. Figure 15 shows enlarged graphs of the heating area and temperature during the scuffing process in Tests 1, 2, and 3. The variations in the heating area in all tests show similar behaviour. In the first stage, heat generation occurs intermittently in small areas. In the second stage, heat generation occurs intermittently as the heating area gradually increases. In the third stage, heat generation becomes continuous and the heating area rapidly increases with increasing contact area.

Fig. 13

Variations in contact area calculated from captured images in Test 1

Fig. 14

Heating area calculated from binarized images and temperature measured by thermocouple during entire test of a Test 1, b Test 2, and c Test 3

Fig. 15

Heating area calculated from binarized images and temperature measured by thermocouple during scuffing process of a Test 1, b Test 2, and c Test 3

4 Discussion

In situ observation of the contact area using a monochrome high-speed camera, which can detect both visible and near-infrared light, was used in the current study. Therefore, the captured images are grey scale, as shown in Fig. 10, which may contain both reflected light due to halogen lamp and near-infrared light due to heat generation. Brightness histograms were calculated as shown in Fig. 12. The plastic flow area shows very high intensity. The reflected light component of the captured images could be removed by binarization of the images with the threshold value. The binarized images emphasise the heat generation behaviour on the contact area. A merit of this method is that, despite a fast phenomenon of scuffing, it could be possible to simultaneously capture the transition of the contact area and heat generation behaviour. In the following discussion, binarized images are suitably processed to understand the heat generation behaviour.

The transition of the contact area during the scuffing process was observed by the capturing system. At the start of the record, a black film was observed in the contact area. After the black film was mostly removed, plastic flow occurred, which is identified as the initiation of scuffing. The imaging system precisely captured the significant heat generation at the plastic flow area, which caused a catastrophic failure with a marked increase in friction. The heat generation behaviour observed in the test could be classified into three stages, as detailed in the following.

The black film could be observed on the contact area at the test time of 4.178 s, which is the start of capturing, as shown in Fig. 6a. While the black film covered the contact area, the value and the fluctuation of the friction coefficient were both low, as shown in Fig. 5. Low wear was maintained, as the contact area did not expand while the black film was present. When the removal of the black film began, wear was promoted as shown in Fig. 6d at the test time of t = 7.4723 s, along with increased fluctuation of the friction coefficient. After most of the black film was removed, heat generation was observed as shown in Fig. 6f. Although the properties of the black film, such as the thickness and composition, were not investigated in the current study, the black film was probably wear particles, an oxide film, and a chemical reaction film originating from the ZDDP [42, 43] and MoDTC [44] additives.

During the scuffing process after the first occurrence of plastic flow, the heat generation behaviour could be divided into three stages. In the first stage, local heat generation occurred repeatedly as the transfer layers passed and at the trailing edge of the contact area, as shown in Fig. 9. Figure 16 shows a colour map of the number of heat generation instances, which was calculated from 431 binarized images captured during the first stage in Test 1. It is found that heat generation occurs frequently at the centre and trailing edge of the contact area, while non-heated areas still remain during the first stage. During the second stage, heat generation occurred regardless of the passage of transfer layers and the heat generation area was larger than it was in the first stage, as shown in Fig. 10b, e, h. Figure 17 shows the heating area at the test time of t = 8.9436 s, which is the beginning of the second stage in Test 1. It is found that, at the beginning of the second stage, heat generation began in the area that experienced first-stage heating. Figure 18 shows the heat generation area and heat-experienced area at each stage of the scuffing process in Tests 1, 2, and 3. The test time of the images of the first and second stages is near the end of the respective stages. The test time of the images of the third stage is near the start of the third stage. In all tests, the heat-experienced area does not cover the contact area in the first and second stages; however, it covers the contact area in the third stage. Figure 19 shows the heat generation area and heat-experienced area with a focus on the transition from the second to the third stage in Test 1. During the second stage, heat generation occurs intermittently and the area experiencing heating expands over the contact area. As the previously heated area gradually encompasses the contact area, as shown in Fig. 19i at the test time of t = 9.0280 s, heat generation becomes continuous, which indicates the beginning of the third stage. During the early part of the third stage, as shown in Fig. 19i–p, the heat generation area also begins to expand throughout the contact area.

Fig. 16

Map of heat generation frequency during the first stage in Test 1

Fig. 17

Heating area at the beginning of the second stage in Test 1

Fig. 18

Heat generation behaviour and heat-experienced area at each stage of scuffing in a Test 1, b Test 2, and c Test 3

Fig. 19

Heat generation behaviour and heat-experienced area during the second and third stages in Test 1

During the first stage of heat generation, heat is locally generated at severe contact spots such as passing of transfer layers and the trailing edge of the contact area. High heat generation at plastic flow locations was also observed in a previous study [39]. Matsuzaki et al. [41] observed that plastic flow occurred upon the passage of the transfer layer during scuffing. Transfer layers originating from the steel pin cause steel–steel contact and produce a strong adhesive force [45]. Additionally, the transfer layer has higher hardness as well as convex shapes because of work hardening [46]. Such characteristics of transfer layers as the same material as the counter surface, higher hardness, and convex shapes appear to cause localised high pressure with strong adhesive force, resulting in high heat generation during impact contacts. Mishina and Sasada [47] and Hase et al. [48] also suggested that the passing of the transfer layer induced strong adhesion between the friction surfaces. By contrast, at the trailing edge of the contact area, heat generation occurred regardless of the passage of the transfer layer. This is probably because the stress at the edge of the contact area was very high.

Heat generation tends to occur more frequently on the area that has experienced heating than on the non-heated area, as observed during the second and third stages. Several mechanisms by which the heated area tends to repeatedly generate heat can be suggested. The first mechanism is the material softening of the top surface layer, arising from the change in microstructure of the heated area. High temperatures of > 1000 °C in the heat generation area, phase transformation during scuffing [38, 39, 40], and grain refinement in this specific layer after scuffing [27] have been reported. Grain refinement occurs after phase transformation because of the high temperature and shear stress at the heated area. The refined grains have many grain boundaries that become nucleation sites, thus facilitating dynamic recrystallisation [49], which causes material softening during plastic deformation. The material softening may promote the occurrence of plastic flow of the contact area. Moreover, the adhesion force increases because the real contact area increases due to material softening, which increases the shear force on the plastic flow area. The second mechanism involves an increase in the adhesion stress at the heat-experienced area by grain refinement. Kato et al. [50] suggested that grain refinement caused an increase in the adhesion stress because the fine grain material, which contains many grain boundaries, is unstable and active. The third mechanism is the removal of the oxide film, which would otherwise prevent strong adhesion forces between the mating surfaces [51]. The fourth mechanism is a reduction of the surface reactivity of the grain refinement layer at the heat-experienced area related to the adsorption of oil, oxide film formation, and reaction of additives. Grew et al. [52] reported that, because the lattice strain of the grains was small, the surface reactivity was low, which slows down the oxide film formation. The lattice strain at the heat-experienced area may be low due to high temperature. Therefore, the performance of the adsorption of oil, oxide film formation, and reaction of additives may decrease at the heat-experienced area. It is controversial whether the four mechanisms affect the scuffing phenomenon synergistically or not. The influence of metallographic changes in the subsurface on material softening and surface reactivity should be investigated in the future.

When non-heated areas remain, as shown in Figs. 16, 18, and 19, the non-heated areas can support the applied load; thus, heat generation occurs intermittently during the second stage, as shown in Fig. 11. The temperatures at the start of the first, second, and third stages in Test 1 were ~ 270 °C, 300 °C, and 370 °C, respectively, as shown in Fig. 15a. The hardness values of the bulk material in these stages were possibly 420 Hv, 370 Hv, and 250 Hv, respectively, as assumed from Fig. 3. The non-heated area during the first and second stages was considered to remain hard. During the third stage, as the area that has been heated encompasses the whole contact area, heat generation becomes continuous because the applied load is always supported by a previously heated area. At the beginning of the third stage, the contact area is considered to have surface waviness; thus, heat is locally generated during the first portion of the third stage and gradually expands to the whole contact area during the middle and final portions of the third stage.

The increase of the friction coefficient during the scuffing process, as shown in Fig. 5, appears to arise from the increases in the area and frequency of heat generation. The friction increase is important in the increase of the temperature, which causes the macroscale softening of the steel pin. The hardness of the steel is decreased with increasing temperature, as shown in Fig. 3. Therefore, the real contact area is increased with the softening of the steel pin surface, thereby increasing friction. The increased friction causes increased heat generation, inducing greater temperature increases. The feedback process based on the relationship between friction and temperature appears to contribute to the progress of scuffing.

Figure 20 shows a schematic summarising the transition of the heat generation behaviour that causes macroscale scuffing, based on the above discussion. It is found that macroscale scuffing occurs after the contact area is covered by the heated area. This finding is revealed here for the first time utilising the in situ observation method. However, the current understanding of the scuffing process raises important questions on the materials science that remain to be investigated. Additionally, tests in which other parameters such as temperature, applied load, sliding speed, and surface roughness are changed should also be conducted. Further research should be continued to attain a better understanding of the scuffing phenomenon.

Fig. 20

Schematic of transition of heat generation behaviour leading to macroscale scuffing

5 Conclusions

In the current study, in situ observation of the scuffing process was conducted using a monochrome high-speed camera. The camera employed in the current study has a detectable range varying from visible to near-infrared wavelength and could capture fast images with several thousand frames per second. The high-speed camera could capture precisely combine images of reflected visible light showing the removal of the protective film and plastic flow and the expansion of the contact area and near-infrared light emitted from heat spots at plastic flow areas. In the current study, the scuffing behaviour was investigated by using the current in situ observation system under shock loading in engine oil lubrication. The current in situ observation system could show the following precise phenomena in the scuffing process.

  1. 1.

    Before scuffing, the black film that covered the contact area was gradually removed. When most of the black film was removed, heat generation started.

     
  2. 2.

    The heat generation behaviour arising from plastic flow in the contact area during the scuffing process could be classified into three stages.

     
  3. 3.

    During the first stage, heat generation occurred intermittently at severe local contact points, such as the passage of transfer layers on the sapphire disc and the trailing edge of the contact area.

     
  4. 4.

    During the second stage, heat generation occurred intermittently over larger areas than those in the first stage. The heat generation during the second stage initiated in the areas that had been heated during the first stage.

     
  5. 5.

    When the entire contact area had been previously heated, heat generation occured continuously throughout the contact area; this was called the third stage of scuffing. During the third stage, the contact area increased rapidly with increasing friction and temperature.

     

Notes

Acknowledgements

This study has been supported by a Strategic Innovation Promotion (SIP) Program of Council for Science, Technology and Innovation of “Innovative Combustion Technology” (Management institution: JST).

Funding

Funding was provided by Japan Science and Technology Agency.

References

  1. 1.
    Galvão, C., Tomanik, E., Fujita, H., Paes, E., Morais, P.: Advantages and challenges for low viscosity oils in emergent countries. SAE Tech. Pap. (2017).  https://doi.org/10.4271/2017-36-0387 CrossRefGoogle Scholar
  2. 2.
    Blok, H.: The flash temperature concept. Wear. 6, 483–494 (1963).  https://doi.org/10.1016/0043-1648(63)90283-7 CrossRefGoogle Scholar
  3. 3.
    Lee, S.C., Chen, H.: Experimental validation of critical temperature-pressure theory of scuffing. Tribol. Trans. 38, 738–742 (1995).  https://doi.org/10.1080/10402009508983467 CrossRefGoogle Scholar
  4. 4.
    Dyson, A.: Scuffing—a review. Tribol. Int. 8, 77–87 (1975).  https://doi.org/10.1016/0301-679X(75)90056-0 CrossRefGoogle Scholar
  5. 5.
    Dyson, A.: Scuffing—a review. Part 2: the mechanism of scuffing. Tribol. Int. 8, 117–122 (1975).  https://doi.org/10.1016/0301-679X(75)90029-8 CrossRefGoogle Scholar
  6. 6.
    Bowman, W.F., Stachowiak, G.W.: A review of scuffing models. Tribol. Lett. 2, 113–131 (1996).  https://doi.org/10.1007/BF00160970 CrossRefGoogle Scholar
  7. 7.
    Li, S., Kahraman, A., Anderson, N., Wedeven, L.D.: A model to predict scuffing failures of a ball-on-disk contact. Tribol. Int. (2013).  https://doi.org/10.1016/j.triboint.2012.11.007 CrossRefGoogle Scholar
  8. 8.
    Enthoven, J., Spikes, H.A.: Infrared and visual study of the mechanisms of scuffing. Tribol. Trans. 39, 441–447 (1996).  https://doi.org/10.1080/10402009608983550 CrossRefGoogle Scholar
  9. 9.
    Enthoven, J.C., Spikes, H.A.: Visual observation of the process of scuffing. Tribol. Ser. 30, 487–494 (1995).  https://doi.org/10.1016/S0167-8922(08)70654-0 CrossRefGoogle Scholar
  10. 10.
    Choa, S.H., Ludema, K.C., Potter, G.E., Dekoven, B.M., Morgan, T.A., Kar, K.K.: A model of the dynamics of boundary film formation. Wear. 177, 33–45 (1994).  https://doi.org/10.1016/0043-1648(94)90115-5 CrossRefGoogle Scholar
  11. 11.
    Palacios, J.M.: Films formed by antiwear additives and their incidence in wear and scuffing. Wear. 114, 41–49 (1987).  https://doi.org/10.1016/0043-1648(87)90014-7 CrossRefGoogle Scholar
  12. 12.
    Bjerk, R.O.: Oxygen—an extreme-pressure agent. ASLE Trans. 16, 97–106 (1973).  https://doi.org/10.1080/05698197308982710 CrossRefGoogle Scholar
  13. 13.
    Cutiongco, E.C., Chung, Y.-W.: Prediction of scuffing failure based on competitive kinetics of oxide formation and removal: application to lubricated sliding of AISI 52100 steel on steel. Tribol. Trans. 37, 622–628 (1994).  https://doi.org/10.1080/10402009408983338 CrossRefGoogle Scholar
  14. 14.
    Saeidi, F., Taylor, A.A., Meylan, B., Hoffmann, P., Wasmer, K.: Origin of scuffing in grey cast iron-steel tribo-system. Mater. Des. 116, 622–630 (2017).  https://doi.org/10.1016/j.matdes.2016.12.044 CrossRefGoogle Scholar
  15. 15.
    Cann, P., Ioannides, E., Jacobson, B., Lubrecht, A.A.: The lambda ratio—a critical re-examination. Wear. 175, 177–188 (1994).  https://doi.org/10.1016/0043-1648(94)90181-3 CrossRefGoogle Scholar
  16. 16.
    Horng, J.H.: True friction power intensity and scuffing in sliding contacts. J. Tribol. 120, 829 (1998).  https://doi.org/10.1115/1.2833786 CrossRefGoogle Scholar
  17. 17.
    Li, S.: Influence of surface roughness lay directionality on scuffing failure of lubricated point contacts. J. Tribol. 135, 041502 (2013).  https://doi.org/10.1115/1.4024783 CrossRefGoogle Scholar
  18. 18.
    Ludema, K.C.: A review of scuffing and running-in of lubricated surfaces, with asperities and oxides in perspective. Wear. 100, 315–331 (1984).  https://doi.org/10.1016/0043-1648(84)90019-X CrossRefGoogle Scholar
  19. 19.
    Yoon, H., Zhang, J., Kelley, F.: Scuffing characteristics of sae 50b38 steel under lubricated conditions. Tribol. Trans. 45, 246–252 (2002).  https://doi.org/10.1080/10402000208982547 CrossRefGoogle Scholar
  20. 20.
    Zhang, C., Cheng, H.S., Wang, Q.J.: Scuffing behavior of piston-pin/bore bearing in mixed lubrication—part II: scuffing mechanism and failure criterion. Tribol. Trans. 47, 149–156 (2004).  https://doi.org/10.1080/05698190490279100 CrossRefGoogle Scholar
  21. 21.
    Hershberger, J., Ajayi, O.O., Zhang, J., Yoon, H., Fenske, G.R.: Evidence of scuffing initiation by adiabatic shear instability. Wear. 258, 1471–1478 (2005).  https://doi.org/10.1016/j.wear.2004.10.010 CrossRefGoogle Scholar
  22. 22.
    Ajayi, O.O., Lorenzo-Martin, C., Erck, R.A., Fenske, G.R.: Scuffing mechanism of near-surface material during lubricated severe sliding contact. Wear. 271, 1750–1753 (2011).  https://doi.org/10.1016/j.wear.2010.12.086 CrossRefGoogle Scholar
  23. 23.
    Markov, D., Kelly, D.: Mechanisms of adhesion-initiated catastrophic wear: pure sliding. Wear. 239, 189–210 (2000).  https://doi.org/10.1016/S0043-1648(99)00373-7 CrossRefGoogle Scholar
  24. 24.
    Jiajun, L., Zhiqiang, L., Yinqian, C.: The study of scuffing and pitting failure of cam-tappet rubbing pair. Wear. 140, 135–147 (1990).  https://doi.org/10.1016/0043-1648(90)90127-V CrossRefGoogle Scholar
  25. 25.
    Piekoszewski, W., Szczerek, M., Tuszynski, W.: The action of lubricants under extreme pressure conditions in a modified. Wear. 249, 188–193 (2001).  https://doi.org/10.1016/S0043-1648(01)00555-5 CrossRefGoogle Scholar
  26. 26.
    Szczerek, M., Tuszynski, W.: A method for testing lubricants under conditions of scuffing. Part I. Presentation of the method. Tribotest. 8, 273–284 (2002).  https://doi.org/10.1002/tt.3020080402 CrossRefGoogle Scholar
  27. 27.
    Rogers, M.D.: Metallographic characterisation of transformation phases on scuffed cast-iron diesel engine components. Tribology. 2, 123–127 (1969).  https://doi.org/10.1016/S0041-2678(69)80249-6 CrossRefGoogle Scholar
  28. 28.
    Rogers, M.D.: The mechanism of scuffing in diesel engines. Wear. 15, 105–116 (1970).  https://doi.org/10.1016/0043-1648(70)90004-9 CrossRefGoogle Scholar
  29. 29.
    Torrance, A.A., Cameron, A.: Surface transformations in scuffing. Wear. 28, 299–311 (1974).  https://doi.org/10.1016/0043-1648(74)90187-2 CrossRefGoogle Scholar
  30. 30.
    Leng, J.A., Davies, J.E.: Examination of wear debris produced using a four-ball machine. Tribol. Int. 22, 137–142 (1989).  https://doi.org/10.1016/0301-679X(89)90174-6 CrossRefGoogle Scholar
  31. 31.
    Derep, J.L.: Microstructure transformation induced by adiabatic shearing in armour steel. Acta Met. 35, 1245–1249 (1987).  https://doi.org/10.1016/0001-6160(87)90005-8 CrossRefGoogle Scholar
  32. 32.
    Rogers, H.C.: Adiabatic plastic deformation. Annu. Rev. Mater. Sci. 9, 283–311 (1979).  https://doi.org/10.1146/annurev.ms.09.080179.001435 CrossRefGoogle Scholar
  33. 33.
    Xu, Y., Zhang, J., Bai, Y., Meyers, M.A.: Shear localization in dynamic deformation: microstructural evolution. Metall. Mater. Trans. A 39, 811–843 (2008).  https://doi.org/10.1007/s11661-007-9431-z CrossRefGoogle Scholar
  34. 34.
    Ajayi, O.O., Hersberger, J.G., Zhang, J., Yoon, H., Fenske, G.R.: Microstructural evolution during scuffing of hardened 4340 steel—implication for scuffing mechanism. Tribol. Int. 38, 277–282 (2005).  https://doi.org/10.1016/j.triboint.2004.08.011 CrossRefGoogle Scholar
  35. 35.
    Kuznetsov, V.P., Tarasov, S.Y., Dmitriev, A.I.: Nanostructuring burnishing and subsurface shear instability. J. Mater. Process. Technol. 217, 327–335 (2015).  https://doi.org/10.1016/j.jmatprotec.2014.11.023 CrossRefGoogle Scholar
  36. 36.
    Yagi, K., Ebisu, Y., Sugimura, J., Kajita, S., Ohmori, T., Suzuki, A.: In situ observation of wear process before and during scuffing in sliding contact. Tribol. Lett. 43, 361–368 (2011).  https://doi.org/10.1007/s11249-011-9817-3 CrossRefGoogle Scholar
  37. 37.
    Li, H., Yagi, K., Sugimura, J., Kajita, S., Shinyoshi, T.: Role of wear particles in scuffing initiation. Tribol. Online. 8, 285–294 (2013).  https://doi.org/10.2474/trol.8.285 CrossRefGoogle Scholar
  38. 38.
    Kajita, S., Yagi, K., Izumi, T., Koyamachi, J., Tohyama, M., Saito, K., Sugimura, J.: In situ X-ray diffraction study of phase transformation of steel in scuffing process. Tribol. Lett. 57, 6 (2015).  https://doi.org/10.1007/s11249-014-0443-8 CrossRefGoogle Scholar
  39. 39.
    Yagi, K., Kajita, S., Izumi, T., Koyamachi, J., Tohyama, M., Saito, K., Sugimura, J.: Simultaneous synchrotron X-ray diffraction, near-infrared, and visible in situ observation of scuffing process of steel in sliding contact. Tribol. Lett. 61, 19 (2016).  https://doi.org/10.1007/s11249-015-0636-9 CrossRefGoogle Scholar
  40. 40.
    Izumi, T., Yagi, K., Koyamachi, J., Saito, K., Sanda, S., Yamaguchi, S., Ikehata, H., Yogo, Y., Sugimura, J.: Surface deteriorations during scuffing process of steel and analysis of their contribution to wear using in situ synchrotron X-ray diffraction and optical observations. Tribol. Lett. 66, 120 (2018).  https://doi.org/10.1007/s11249-018-1062-6 CrossRefGoogle Scholar
  41. 41.
    Matsuzaki, Y., Yagi, K., Sugimura, J.: In-situ fast and long observation system for friction surfaces during scuffing of steel. Wear. 386–387, 165–172 (2017).  https://doi.org/10.1016/j.wear.2017.06.013 CrossRefGoogle Scholar
  42. 42.
    Fujita, H., Spikes, H.A.: The formation of zinc dithiophosphate antiwear films. Proc. Inst. Mech. Eng., Part J 218, 265–277 (2004).  https://doi.org/10.1243/1350650041762677 CrossRefGoogle Scholar
  43. 43.
    Zhang, J., Spikes, H.: On the mechanism of ZDDP antiwear film formation. Tribol. Lett. 63, 24 (2016).  https://doi.org/10.1007/s11249-016-0706-7 CrossRefGoogle Scholar
  44. 44.
    Morina, A., Neville, A., Priest, M., Green, J.H.: ZDDP and MoDTC interactions and their effect on tribological performance—tribofilm characteristics and its evolution. Tribol. Lett. 24, 243–256 (2006).  https://doi.org/10.1007/s11249-006-9123-7 CrossRefGoogle Scholar
  45. 45.
    Rabinowicz, E.: The influence of compatibility on different tribological phenomena. ASLE Trans. 14, 206–212 (1971).  https://doi.org/10.1080/05698197108983244 CrossRefGoogle Scholar
  46. 46.
    Rigney, D.A., Chen, L.H., Naylor, M.G.S., Rosenfield, A.R.: Wear processes in sliding systems. Wear. 100, 195–219 (1984).  https://doi.org/10.1016/0043-1648(84)90013-9 CrossRefGoogle Scholar
  47. 47.
    Mishina, H., Sasada, T.: Observation of micro-structure in seized portion and mechanism of seizure. Trans. ASME J. Tribol. 108, 128 (1986).  https://doi.org/10.1115/1.3261132 CrossRefGoogle Scholar
  48. 48.
    Hase, A., Mishina, H., Wada, M.: Fundamental study on early detection of seizure in journal bearing by using acoustic emission technique. Wear. 346–347, 132–139 (2016).  https://doi.org/10.1016/j.wear.2015.11.012 CrossRefGoogle Scholar
  49. 49.
    Poorganji, B., Miyamoto, G., Maki, T., Furuhara, T.: Formation of ultrafine grained ferrite by warm deformation of lath martensite in low-alloy steels with different carbon content. Scr. Mater. 59, 279–281 (2008).  https://doi.org/10.1016/J.SCRIPTAMAT.2008.03.041 CrossRefGoogle Scholar
  50. 50.
    Kato, H., Yamashita, K., Todaka, Y.: Wear properties of ultra-fine grained metals produced by severe plastic deformation. In: Tribology conference 2018 Spring Tokyo, F31 (2018)Google Scholar
  51. 51.
    Kovalchenko, A.M., Blau, P.J., Qu, J., Danyluk, S.: Scuffing tendencies of different metals against copper under non-lubricated conditions. Wear. 271, 2998–3006 (2011).  https://doi.org/10.1016/J.WEAR.2011.06.009 CrossRefGoogle Scholar
  52. 52.
    Grew, W., Cameron, A.: Role of austenite and mineral oil in lubricant failure. Nature. 217, 481–482 (1968).  https://doi.org/10.1038/217481a0 CrossRefGoogle Scholar

Copyright information

© The Author(s) 2018

Open AccessThis article is distributed under the terms of the Creative Commons Attribution 4.0 International License (http://creativecommons.org/licenses/by/4.0/), which permits unrestricted use, distribution, and reproduction in any medium, provided you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons license, and indicate if changes were made.

Authors and Affiliations

  1. 1.Department of Hydrogen Energy Systems, Graduate School of EngineeringKyushu UniversityFukuokaJapan
  2. 2.Department of Mechanical Engineering, Faculty of EngineeringKyushu UniversityFukuokaJapan
  3. 3.International Institute for Carbon-Neutral Energy Research (I²CNER)Kyushu UniversityFukuokaJapan

Personalised recommendations