1 Introduction

Since the flash temperature was observed at the rubbing contact [1], most of the triboproblems have been analyzed and solved by the concept of frictional temperature rise. However, some of the triboproblems have been remained unsolved by the concept of the energy of frictional temperature rise, in spite of that most of the frictional energy is consumed by the temperature rise. This fact has been suggested that unknown energy source other than the frictional temperature rise is generated at and the vicinity of the sliding contact.

On the other hand, the triboemission phenomena of the electrons, negatively and positively charged particles, and photons from the sliding contact have been systematically studied to give a conclusion that plasma, i.e., “triboplasma” is generated in the gap of the sliding contact, since the same numbers of the negatively and positively charged particles are emitted from the sliding contact and the vicinity [2]. This is because plasma is the state where the same numbers of the negative and positive particles coexist to be neutral [3]. Then the triboplasma was discovered by the success of taking the image using an ICCD camera in the rear gap in 2001 [4] and further in the inside [5,6,7] and the front outside [7, 8] of the sliding contact in dry sliding, and even in oil lubricated sliding contact in 2006 [9, 10]. The triboplasma has been concluded to be generated by discharging of the ambient gas through the electron avalanche process due to the intense electric field caused by the tribocharge-induced electric fields in the gap of the sliding contact. For better understanding the present article, the previously proposed model of the triboelectromagnetic phenomena is shown in Fig. 1 [7].

Fig. 1
figure 1

Model of the triboelectromagnetic phenomena, which embrace the triboplasma generation, triboemission, tribocharging and post-emission; the triboplasma is generated in the front outside, the inside and the rear outside of the sliding contact through the electron avalanche processes due to the intense electric fields caused by tribocharging [4]

Full size image

On the other hand, the friction and wear reducing high molecular weight products formed in a hydrocarbon gas atmosphere was attributed to the action of triboplasma in dry sliding [11, 12] and also the decomposition of stable lubricating oil used for HDI was concluded to be due to the triboplasma action under oil lubrication [13]. The wear of a diamond tool during cutting process of the semiconductors of Si and Ge has also been attributed to the triboplasma action [14].

For developing triboplasma technologies, it is essential to know the role of triboplasma temperatures in the above tribophysical and tribochemical process in addition to the actions caused by the active particles such as electrons, ions, radicals and photons in the plasma. However, the temperature of the triboplasma has been remained unknown.

From the point of the temperature, the plasma is classified into two; (1) the high-temperature plasma, where the temperatures of the ions and neutral particles are almost equal to those of the high energy electrons [15, 16], and (2) the non-equilibrium low-temperature plasma, where the temperatures of the electrons are high, but those of ions and neutral particles are low having almost the same temperature as that of the ambient circumstance [17,18,19]. However, it is also unknown if the triboplasma is the high or the non-equilibrium low-temperature plasma.

In this article, the temperature of the triboplasma has been successfully measured using a highly accurate microscopic infrared camera and drew a conclusion that the triboplasma is the non-equilibrium low-temperature plasma with the temperature rise of 0.69 °C at the highest, and it was found that the measured temperature distribution of the triboplasma changes along the wear track and changes also periodically with the rotation of the disk specimen. But the maximum temperature rise was found to be independent of the repeated sliding and further the sliding velocity. The periodic temperature distribution change is attributed to the tribocharge distribution along the wear track.

2 Experimental Procedure

Figure 2 shows a schematic representation of the apparatus to measure the distribution of the temperature rise above the ambient one, ΔT, using the microscopic high accuracy infrared camera (FLIR Systems, Inc., SC7600-BB). The sensitive wavelength of the camera was 2.5–5.0 μm. The temperature rise of the triboplasma generated in the vicinity of the sliding contact was measured while sliding a diamond pin with a tip radius of R = 4 mm on a sapphire (single crystal Al2O3) disk with a diameter of 50 mm and the thickness of 1 mm under dry sliding in a room temperature of T R = 26 °C, in the ambient room air with the relative humidity of H R = 20%. The diamond pin was embedded into the Ti bar having a diameter of 2 mm and an included angle of 120° to avoid magnetization of the pin observed as in Fe. The surfaces of both specimens were finished with polishing.

Fig. 2
figure 2

Pin-on-disk tribometer installed with a high sensitive infrared camera (IRC: FLIR SC7600-BB) to measure the temperature distribution of the triboplasma

Full size image

The side image of the temperature rise distribution was measured from the horizontal direction with an angle of several degrees from the sapphire disk surface as seen in Fig. 2a, whereas the plane image was measured vertically through the sapphire disk as shown in Fig. 2b. The resolution of the IR camera was 15 μm/pixel, and the integration time (exposure time) was 2.35 ms. The camera detects the combined energy radiated from the plasma itself and the background in the square area of one pixel, and gives the surface temperature calibrated by the blackbody radiation.

Friction experiments were performed under the normal forces of F N = 1 N for measuring the side image and F N = 2 N for the plane image, respectively, under the rotational velocity of ω = 10, 20, 50, 100 and 155 rpm. The wear track diameter was 40 mm. The maximum Hertzian contact pressures under F N = 1 and 2 N were P max·1N = 952 MPa and P max·2N = 1200 MPa, respectively, whereas the mean Hertzian contact pressures under F N = 1 and 2 N were P mean·1N = 634 MPa and P mean·2N = 800 MPa, respectively. They were calculated using the material parameters of diamond and sapphire; Young’s modulus of diamond and sapphire is E dia = 1050 GPa and E sap = 335 GPa, respectively, whereas Poisson ratio of diamond and sapphire is ν dia = 0.1 and ν sap = 0.25, respectively. The test specimens were cleaned with the solvents of n-hexane, ethanol and then acetone in an ultrasonic bath. The sliding experiments were conducted after removing the adsorbed solvents molecules from the specimen surfaces by heating with hot air stream. The friction experiment to measure the effect of sliding velocity on the triboplasma temperature rise was performed with step rise of the sliding velocity on the same wear track in the order of ω = 10 → 20 → 50 → 100 → 155 rpm. Voltages of 330 V would be generated on the diamond pin surface against the sapphire disk at the center of the rear triboplasma during sliding where the rear gap distance is about 7.5 μm [4].

For comparison, temperature distribution was also measured in the tribosystem where no tribocharging occurs to generate triboplasma, i.e., in the tribosystem of a metal of Fe pin with a tip radius R = 2 mm sliding on a metal of Cu disk with a diameter of 25 mm and the thickness of 12 mm under F N = 0.5 N and ω = 80 rpm with the wear track diameter of 20 mm in the ambient room air with the relative humidity of H R = 54%.

3 Results and Discussion

3.1 Temperature Distribution of the Triboplasma

Figure 3a, b shows the side and plane images of the static contact of the diamond pin against the sapphire disk without sliding under F N = 1 and 2 N, respectively, whereas Fig. 3c, d shows the side and plane images of the temperature rise distribution above the ambient one, ΔT, measured by the infrared camera while sliding the diamond pin against the sapphire disk at ω = 10 rpm (V = 2.1 cm/s) under F N = 1 and 2 N, respectively. As discussed below, the temperature distribution of the triboplasma changes with sliding time, t. The temperature distributions in Fig. 3c, d are the ones at the beginning of the sliding. It can clearly be seen in Fig. 3c that the temperature rise is taking place at the two regions of the front and rear gaps of the sliding contact. These temperature rise regions correspond to the regions of “the front outside plasma” and “the rear outside plasma” as shown in the electron avalanche model in Fig. 1 [4]. This means that the measured temperature rise distribution in Fig. 3c is the temperature rise distribution of the triboplasma itself. The plane image in Fig. 3d also shows the plain distribution of the triboplasma temperature rise distribution inside the rear gap. The temperature measurement scale suggests that the maximum temperature rise of the triboplasma is around 0.15 °C at the most. The precise values of the triboplasma temperature rise are shown below.

Fig. 3
figure 3

Side (a) and plane (b) images at the static contact before sliding, and typical side (c) and plane (d) images of the temperature rise distribution during sliding at the time of t = 0.1 s under F N = 1 N and F N = 2 N, respectively, at ω = 10 rpm

Full size image

Figure 4 shows the typical side images of the temperature rise distribution of the triboplasma fully generated in the rear gap of the sliding contact at ω = 20, 50, 100 and 155 rpm under F N = 1 N. In all the rotational velocities, the plasma is generated in the whole rear gap, but the front outside plasma is generated only weakly or scarcely generated, since the front outside plasma is generated only in the case where the rest tribocharge remained after consuming by the rear triboplasma generation has enough intensity to cause the front plasma generation. The temperature rise in these rotational velocities seems to have almost the same maximum values of 0.15 °C at the most as in ω = 10 rpm, showing that the plasma temperature rise does not depend on the sliding velocity. The precise temperature rise distribution of the triboplasma is shown below with the line analysis of the temperature rise distribution.

Fig. 4
figure 4

Typical side images of the temperature rise distribution of the triboplasma which is fully generated in the rear gap under F N = 1 N at ω = 20 (a), 50 (b), 100 (c) and 155 (d) rpm

Full size image

It is probably seen that the size of the rear triboplasma is slightly greater at the lowest sliding velocity of 20 rpm than those at the other velocities. The reason is as follows. As described above, the sliding experiments were continuously performed with the step rise of the sliding velocity on the same wear track. Then during the step velocity increase, such films as graphitic carbon are probably formed. These films must be smeared or transferred to the opposing surfaces. Graphitic carbon films may have higher electrical conductivity than the sliding materials of diamond and sapphire and thus disturb electrical charge accumulation reducing the air discharge to suppress the triboplasma temperature rise. Namely, with the step increase in the sliding velocity on the same wear track, these films with higher electric conductivity may be accumulated and then triboplasma temperature rise decreased. To prove it, we need further work for analyzing the surface films formed.

3.2 Line Analysis of the Temperature Distribution

Figure 5 shows a part of the side image of the temperature rise distribution shown in Fig. 3c and the temperature rise profiles measured along the a–b, c–d, e–f and g–h lines at ω = 10 rpm under F N = 1 N. As seen in the profiles measured along the a–b line, the maximum and average temperature rise of the rear triboplasma at ω = 10 rpm is ΔT max = 0.18 °C and ΔT mean = 0.15 °C, respectively. It can also be seen that the plasma temperature rise is almost constant in the center region of the rear plasma and that it decreases steeply toward the outside from the gap. On the other hand, the front plasma is generated along the arc line from the disk surface toward the edge of the pin as shown by an arrow. This temperature rise pattern corresponds to the electron avalanche model of the front plasma generation depicted in Fig. 1. The temperature rise of the front plasma is most intense at the immediate vicinity of the disk surface having the value of ΔT max = 0.18 °C. It steeply decreases first and then gradually with the distance toward the edge of the pin. The reason why the front plasma is most intense at the immediate vicinity of the disk surface is not clear. We need further investigation for clarifying the mechanism. The average temperature rise of the front plasma measured along the c–d line is ΔT mean = 0.075 °C as seen in the c–d profile, which is the half of that in the rear gap. It shows that the plasma intensity in the center of the front of the sliding contact is much weaker than that of the rear gap plasma.

Fig. 5
figure 5

Side image of the temperature rise distribution and the temperature rise profiles measured along the a–b, c–d, e–f and g–h at t = 0.1 s under ω = 10 rpm

Full size image

Figure 6 shows a part of the plane image of Fig. 3d and the temperature rise profiles measured along the a–b and c–d lines. It is clearly seen in the a–b line profile that the temperature rises in the asperity contact region are negligibly small with only the values of ΔT max = 0.05–0.070 °C, whereas those in the triboplasma regions are much higher having the maximum value of ΔT max = 0.69 °C at the center of the plasma. This great temperature difference gives a clear boundary between the two regions to divide them. This smaller temperature rise in the asperity contact region than those in the plasma region demonstrates that the triboplasma temperature rise is not caused by the frictional temperature rise at the asperity contact but by the tribocharge-induced electric field as described above. The value of ΔT max = 0.69 °C measured in the plane image in Fig. 6 is much higher than that of ΔT max = 0.18 °C measured in the side image in Fig. 5. This means that we have to measure the maximum temperature rise of the triboplasma at the center of the plasma in the plane image.

Fig. 6
figure 6

Plane image of the temperature rise distribution and the temperature rise profiles measured along the a–b and c–d lines at t = 0.2 s under ω = 10 rpm

Full size image

As described above, the experimentally measured sliding asperity contact temperatures are ΔT Exp-2N = 0.05–0.070 °C. Here we estimate theoretically the frictional temperature rise at the asperity contacts using the flash temperature theory, where all the frictional energy is postulated to be converted to the temperature rise. Several temperature rise calculation theories have been proposed [20,21,22,23,24,25]. Of all the theories, we calculated the surface temperature rise above the ambient temperature, ΔT, by the modified Archard’s theory proposed by Dayson using the constant α = 25 to calculate the real area of contact by the frictional tangential stress [25]. The measured friction coefficient at the asperity contact in this experiment was so low of f = 0.08. This low friction coefficient suggests again that some unknown low friction surface films such as graphitic carbon films are formed by the action of the triboplasma to the diamond specimen surfaces. The temperature rise at the sliding asperity contacts was calculated at ω = 10 rpm (V = 2.1 cm/s) under F N = 1 and 2 N, respectively, using the friction coefficient of f = 0.08, the constant of α = 25 [26], thermal conductivity of the diamond, K (dia) = 4.77 cal/(cm s °C, thermal conductivity of the sapphire K (sap) = 0.065 Cal/(cm s °C), yield pressure (Vickers hardness) of sapphire P m(sap) = 23.0 GPa, specific gravity of sapphire ρ (sap) = 3.97, specific heat of sapphire c (sap) = 0.18 Cal/(g °C). Then the theoretically calculated surface temperature rise at the asperity contact was determined to be ΔT Theo-1N = 0.014 °C under F N = 1 N and ΔT Theo-2N = 0.075 °C under F N = 2 N, respectively. The so small frictional temperature rise at the asperity contact is due to the high thermal conductivity of the diamond pin. The theoretically calculated temperature rise, ΔT Theo-2N = 0.075 °C, under F N = 2 N coincides well to the experimentally measured temperature rise of ΔT Exp-2N = 0.05–0.07 °C. This good coincidence shows that both the temperature rise measurement and the theoretical calculation [25] are correct and precise. As the measured maximum triboplasma temperature rise of ΔT max = 0.69 °C in the rear gap under F N = 2 N is much higher than the theoretically calculated and experimentally observed temperature rise at the asperity contact, it is clear that the measured temperature rise in the gap of the sliding contact is not caused by the frictional temperature rise at the asperity contact, but by the triboplasma itself.

Figure 7 shows the rear gap temperature rise distributions and the temperature rise profiles measured along the a–b and c–d lines at ω = 20, 50, 100 and 155 rpm under F N = 1 N, respectively. The temperature rise distributions and the shape of the line profiles in the four kinds of rotational velocities are all similar to those in ω = 10 rpm. Even the values of the maximum and average temperature rises above the ambient one at these velocities are almost the same as those at ω = 10 rpm, i.e., 0.17–0.18 °C and 0.14–0.15 °C, respectively. This leads to a conclusion that triboplasma temperature rise is independent of the sliding velocity.

Fig. 7
figure 7

Side image of the temperature rise distribution and the temperature rise profiles measured along the a–b and c–d lines; a at t = 1.9 s under ω = 20 rpm, b at t = 2.0 s under ω = 50 rpm, c at t = 0.6 s under ω = 100 rpm and d at t = 0.7 s under ω = 155 rpm

Full size image

Figure 8 shows the side image of the temperature rise distribution while sliding the Fe pin with the tip radius of R = 2 mm on the rotating Cu disk under F N = 0.5 N at ω = 80 rpm (V = 6.3 cm/s). It can be seen in Fig. 8 that no temperature rise takes place in the rear sliding gap. This demonstrates that no triboplasma can be generated in the tribosystem of metal sliding on metal where no tribocharging occurs. In this sliding combination, the friction coefficient was f = 0.4. The surface temperature rise was calculated with the modified Archard theory using the f = 0.4, the constant α = 25, the thermal conductivity of iron K (Fe) = 0.15 cal/(cm s °C), the thermal conductivity of copper K (Cu) = 0.923 Cal/(cm s °C), the yield pressure (Vickers hardness) of copper P m(Cu) = 450 MPa, the specific gravity of copper ρ (Cu) = 8.93 and the specific heat of copper c (Cu) = 0.0919 Cal/(g °C). The calculation of the surface temperature rise gave the value of ΔT = 0.20 °C under F N = 0.5 N. This value of ΔT = 0.20 °C is much greater than the calculated frictional temperature rise of ΔT = 0.0136 °C in the diamond pin sliding on the sapphire disk under F N = 1 N. However, no temperature rise distribution occurs in the rear gap region as seen in Fig. 8. This demonstrates again that the temperature rise distribution observed above in the rear gap of the sliding contact in the tribosystem of the diamond pin sliding on the sapphire disk is caused by the triboplasma.

Fig. 8
figure 8

Typical side image of the temperature rise distribution above the ambient one while sliding a Fe pin on a Cu disk under F N = 0.5 N at ω = 80 rpm

Full size image

3.3 Time Dependence of the Plasma Distribution

Figure 9a, b shows the time dependence of the temperature rise distribution in the first revolution cycle from t = 0.1 to 6.0 s and the second one from t = 7.0–12.0 s at ω = 10 rpm, respectively, where the time of one revolution, t 1rev, is 6.0 s. As seen in Fig. 9, the temperature rise distribution changes with time, i.e., with sliding distance. In the beginning at t = 0.1–1.0 s in the first cycle, the plasma is generated in both regions of the front outside and the rear gap of the sliding contact. However, the front plasma disappears at t = 2.0 s, again appears slightly at t = 5.1 s and finally disappears at the end of the first cycle of t = 6.0 s. On the other hand, the rear plasma is initially great at t = 0.1–1.0 s, then it decreases steeply with time and almost ceases at t = 4.0 s, and it again appears slightly at the end of the first cycle of t = 6.0 s. In the second cycle from t = 7.0 to 12.0 s, no front plasma appeared, but the feature of the time dependence of the rear plasma observed in the first cycle was repeated, though the intensity was decreased.

Fig. 9
figure 9

Time dependence of the temperature rise distribution above the room temperature in ω = 10 rpm; time of one revolution, t 1rev = 6.0 s

Full size image

Figure 10a shows the time dependence of the temperature rise distribution in the fourth and fifth cycles from t = 2.8 to 3.8 s and from t = 4.0 to t = 5.0 s, respectively, under ω = 50 rpm where t 1rev = 1.2 s, whereas Fig. 10b shows the ones in the first and second cycles from t = 0.1 to 0.6 s and from t = 0.7 to 1.2 s, respectively, under ω = 100 rpm, where t 1rev = 0.6 s. You can see in comparison with the temperature rise distributions between the fourth and fifth cycles in Figs. 10a in 50 rpm, the time-dependent change in the fourth cycle is exactly repeated in the fifth cycle. You can also see in Fig. 10b in 100 rpm, the time-dependent change in the first cycle is exactly repeated in the second cycle. The origin of the repetition of the time-dependent change of the triboplasma generation is discussed below.

Fig. 10
figure 10

Time dependence of the temperature rise distribution above the room temperature in a ω = 50 rpm, t 1rev = 1.2 s and b ω = 100 rpm, t 1rev = 0.6 s

Full size image

3.4 Area Analysis of the Temperature Distribution

Figure 11 shows the time dependence of the maximum and average temperature rises, ΔT max and ΔT mean, respectively, measured by the area analysis method for the plasma region in the side image under ω = 50 rpm and F N = 1 N. The ΔT max and ΔT mean are defined in Figs. 5, 6 and 7. In Figs. 9 and 10, we have seen that the intensity of the plasma temperature rise distribution changes periodically with time, i.e., with the sliding distance. Corresponding to the periodical change of the temperature rise distribution, both the ΔT max and ΔT mean change periodically with time, where one revolution period time is t rev = 1.2 s. It is also seen that the peak position of ΔT max corresponds to that of the ΔT mean, and that their peak heights, ΔT max-peak and ΔT mean-peak are almost independent of the sliding time, i.e., independent of the sliding distance as seen in Fig. 11. It is noticed that the values of the temperature rise change drastically with time in every revolution, but that the peak values of ΔT max-peak and ΔT mean-peak are almost independent of the repeated sliding.

Fig. 11
figure 11

Time dependence of the maximum and average temperature rises, ΔT max and ΔT mean, respectively, above the room temperature at ω = 50 rpm

Full size image

Table 1 summarizes the values of Δt max-peak-ave and Δt mean-peak-ave and their standard deviations, s max and s mean, for the side and plain images under various sliding velocities, ω. Values of the maximum temperature rise Δt max-peak-max under various ω are also given. The three kinds of values Δt max-peak-max, Δt max-peak-ave and Δt mean-peak-ave measured under the five kinds of rotational velocities have almost the same values of 0.19–0.22 °C, 0.17–0.20 °C and 0.11–0.14 °C, respectively. This confirms the conclusion drawn in the line analysis that triboplasma temperature rise is independent of the sliding velocity. The values of Δt max-peak-max and Δt max-peak-ave are 0.69 and 0.43 °C, respectively, in the plane image column and are much higher than those of the side images under various velocities. As the triboplasma temperature rise is independent of the sliding velocity as described above, the values of Δt max-peak-max and Δt max-peak-ave measured in the plane images in ω = 10 rpm must be almost the same value as those in other rotational velocities. Namely, the maximum and average maximum triboplasma temperature rises are around ΔT max-peak-max = 0.69 °C and ΔT mean-peak-ave = 0.25 °C, respectively, for all the sliding velocities in the present study. This low-temperature rise below 1.0 °C gives the conclusion that the triboplasma is not the high-temperature plasma [15, 16], but the non-equilibrium low-temperature plasma [15,16,17,18,19] since the energy of electron is huge [27], but the mass of the electrons is too small to heat the air molecules by collision [3].

Table 1 Values of Δt max-peak-ave and Δt mean-peak-ave and their standard deviations, s max and s mean, for the side and plain images under various sliding velocities, ω
Full size table

3.5 Mechanism of the Time Dependence Nature

Figure 12a, b and c shows the time dependence of the temperature rise distribution at ω = 10 rpm, the corresponding electron avalanche model and the tribocharge distribution model along the wear track, respectively. This figure explains how the time-dependent changes, i.e., the sliding distance-dependent change, of the rear gap temperature rise distribution of the triboplasma is originated through the electron avalanche process from the non-uniformly distributed tribocharge along the wear track. Previously, it has been reported that the distribution of the surface potential along the wear track does not change with the repeated sliding [28]. Namely, it means that tribocharge distribution along the wear track does not substantially change with the repeated sliding in this kind of number of repeated sliding.

Fig. 12
figure 12

Correlations among the triboplasma temperature distribution at ω = 10 rpm (a), the corresponding electron avalanche (b) and the tribocharge distribution (c) models

Full size image

When the contact point comes into the highly negatively tribocharged region at t = 0.1 s, the rear gap triboplasma is generated intensely. The plasma generation becomes stronger at the higher tribocharge region at t = 1.0 s, and then it becomes weaker, weaker and weakest at t = 2.0–6.0 s with the decrease in the tribocharge intensity on the wear track, since the triboplasma is generated by the electric fields caused by the non-uniformly distributed surface charge, which is independent of the repeated sliding. The front plasma is not always generated since it is generated only when the rest tribocharge remained after consuming by the rear plasma generation process is high enough.

In this article, the triboplasma has been decided to be non-equilibrium low-temperature plasma having the temperature rise of 0.69 °C at the highest as described above. Triboplasma temperature rise is caused by the collision of air molecules with the high velocity electrons accelerated by the tribocharge-induced electric fields. Namely, triboplasma temperature rise depends on the tribocharge density on the wear track surface. However, the tribocharge, which is generated at the sliding asperity contacts on the frictional track, is always consumed during sliding by discharging in the rear gap and then in the front gap. Then the tribocharge is difficult to be accumulated on the wear track surface. This is the reason that the temperature rise is independent of the repeated sliding. This charge consumption process takes place also at all the sliding velocities. Thus, triboplasma temperature rise becomes also independent of the sliding velocity.

In this work, the distribution of the very small increase in the triboplasma temperature rise was measured with a negligibly small asperity contact temperature rise by using the tribosystem where high thermally conductive diamond was used as one of the sliding pair. However, in the case where low thermally conductive solids are used as the sliding pair, the asperity temperature must be elevated significantly and thus will give the heat to the plasma, so that the plasma temperature rise will possibly be elevated with the sliding velocity. In a separate experiment, we have observed that high bulk temperature of the disk specimen contributes to the triboplasma temperature rise. This result will soon be reported. In the case of the environments other than air such as N2 or Ar gases, we need further work.

Based on the present study, new advanced in situ triboplasma application technologies are anticipated to be developed without temperature rise, combined with the plasma application technologies [29,30,31] and the triboplasma control technologies [32,33,34].

4 Conclusions

The temperature of the triboplasma was measured using a highly accurate microscopic infrared camera while a diamond pin sliding on a sapphire disk under dry sliding in the ambient air, and the following conclusions have been drawn.

The triboplasma is the non-equilibrium low-temperature plasma. The temperature rise of the triboplasma above the ambient one is 0.69 °C at the highest. The temperature rise distribution depends on the tribocharge distribution on the wear track surface and changes periodically with the rotation of the disk specimen. This is because the tribocharge-induced surface potential distribution is nearly unchanged with the repeated sliding due to that the tribocharge is consumed continuously by discharging in the rear gap and further in the front gap during rotation. Then the triboplasma temperature rise is independent of the repeated sliding and further independent of the sliding velocity. The temperature measurement is a useful technique to investigate the triboplasma behavior.