LUBRICITY AND TRIBOCHEMICAL REACTIVITY OF ADVANCED MATERIALS UNDER HIGH VACUUM

A. Igartua (1), E. Berriozabal (1), B. Zabala (1), F. Pagano (1), I. Minami(2), N. Doerr(3), C. Gabler(3)

, R. Nevshupa (4), E. Roman (4)

, L. Pleth Nielsen(5), S. Louring(5), L. Muntada(6)

(1) Fundación TEKNIKER, c/ Iñaki Goenaga, 5, 20600 Eibar, Spain. E-mail: amaya.igartua@tekniker.es (2) Luleå University of Technology, Division of Machine Elements, SE97187, Luleå, Sweden  

(3) AC2T research GmbH, Wiener Neustadt, Viktor-Kaplan-Strasse 2C, 2700 Wiener Neustadt, Austria (4) IETCC, CSIC, Serrano Galvache 4, 28033 Madrid, Spain

(5) Danish Technological Institute, KongsvangsAlle 29, 8000 Aarhus, Denmark (6) Brugarolas SA, Camino de la Riera, 36-44, Polígono Industrial Cova Solera, 08191 Rubí, Spain

ABSTRACT

The aim of this study was to develop and characterize advanced tribological materials for space applications. For this purpose a newly developed Ultra High Vacuum (UHV) tribometer CA3UHV was used to determine friction, wear and triboemission of gases and volatiles from atmospheric pressure down to high or ultrahigh vacuum. The study was focused on two classes of materials: a) advanced lubricants based on ionic liquids (ILs), b) engineered diamond-like carbon films as anti-wear and low-friction coatings. Tribochemical mechanisms were identified by analyzing gas triboemission involved in the generation of protective tribofilm layer and/or transfer film and surface chemical characterization. 1. INTRODUCTION

Spacecraft reliability and performance are still, in part, conditioned by tribological systems [1, 2], while solid and liquid lubricants have not experienced radical improvements during the last three decades. Furthermore, advanced tribological materials should perform well both under severe conditions in space and at more normal conditions at atmospheric pressure during testing on earth. Also they should comply with very strict requirements on the outgassing of volatiles both from free surfaces [3] and during friction and mechanical deformation [4, 5] in order to avoid contamination of lenses, sensors and other devices of space systems. This work is focused on development and characterization of two classes of material systems: ionic liquids (ILs) and amorphous carbon (a-C) coatings that can potentially be used in space applications lowering the friction. Ionic liquids are promising liquid lubricants and/or lubricant additives due to their high chemical stability, low volatility, good lubricity and miscibility with organic and synthetic lubricant base stocks [6-8]. In addition, ILs can be further tailored for each specific application by careful selection of cationic and anionic moieties from more than one million of

possible combinations. Previous studies have revealed the importance of tribochemical process between the ILs and the substrate in formation of protective tribofilm on frictional surfaces. ILs with fluorinated anions rendered tribofilms containing non-organic fluorides that afforded good tribological properties [9-11]. However, the knowledge on the mechanisms of tribochemical reaction of ILs is still vague because post mortem approach has been mainly used in previous studies. This approach is based on a comparative study of the surface chemical structure and composition before and after a tribological test. A different approach is based on in situ and real-time characterization of tribochemical processes through triboRaman spectroscopy, ATR FTIR tribometry [12], surface plasmon resonance [13], triboplasma [14, 15] and Mechanically Stimulated Gas Emission (MSGE) spectroscopy [11, 16-18]. The latter technique is a further development towards accurate quantification and behavioural analysis of a conventional mass-spectrometry applied for the analysis of gas triboemission [19-21]. In this work, we applied MSGE spectroscopy for study of tribochemistry of three ILs with phosphonium and ammonium cations in combination with fluorinated and non-fluorinated anions. Among advanced solid lubricants, a-C is considered as a very promising candidate for space applications due to its thermal stability, chemical inertness and superlubricity under certain conditions. However, the challenging drawback of a-C is the lack of coatings that can satisfactorily perform both under vacuum conditions and under atmosphere pressure. In literature, there are a number of studies aimed at improving tribological properties of a-C through surface modification [22], controlling hydrogen content [23, 24], gas-phase lubrication [25], doping with metals (Ag, Ni, Ti, W, etc.) [24, 26, 27] and non-metal elements (Si, F, S, N, etc.) [28], etc. In the present study a novel experimental tribometer equipped with MSGE spectroscopy was used to study tribological behaviour and tribochemical process of various a-C coatings as function of doping

_____________________________________ Proc. ‘16th European Space Mechanisms and Tribology Symposium 2015’, Bilbao, Spain, 23–25 September 2015 (ESA SP-737, September 2015)

substance in a wide range of gas pressures from the atmospheric one to 10-8 hPa. 2. EXPERIMENTAL

2.1. Procedure Development of advanced lubricants requires new specific experimental tools that allow simultaneous characterization of friction, wear, gas triboemission and other tribochemical and tribophysical processes under relevant conditions including the gas pressure ranged from 103 to 10-8 hPa, temperatures from cryogenic one to > 200 ºC, radiation, etc. [29]. Recently such system named Ca3UHV has been developed and constructed by IK4-TEKNIKER (see Figure 1). A patented tribological cell of a ball-on-flat configuration and reciprocating motion was specially designed to avoid any unwanted gas emission from guides, drives, bearings, cooling and heating elements. MSGE was characterized using a fast response Quadrupole Mass Spectrometer (QMS) working in a Multi Ion Detection (MID) mode in which ion currents – time series were recorded for ions with various mass-to-charge ratios (m/z). Features and procedure of MSGE spectrometry were described in detail elsewhere [11, 17, 18]. Coefficient of Friction (CoF) and MSGE time series were recorded simultaneously. Each experiment was performed at least twice in order to confirm repeatability.

Figure 1: General view of Ca3UHV (©IK4-Tekniker) experimental system.

X-ray Photoelectron Spectroscopy (XPS) with monochromatic Al Kα source (1486.6 eV, 100 µm in diameter) and a hemispherical analyzer was employed for characterization of chemical composition and bonding structure of the tribofilms formed on substrates under ILs lubrication. The measurements were done at a base pressure of 2×10-9 Pa. A pass energy was 200 eV for the survey spectra and 50 eV for the high resolution spectra. The spectra were fitted using Gaussian-Lorentzian function. 2.2. Ionic liquids In this study, Si3N4balls of 3 mm in diameter were tested against Ti-6Al-4V (grade 5) discs, 14 mm in diameter and 4 mm height. Normal load was 10 N yielding maximum Hertzian stress 1.9 GPa, maximum sliding velocity was 3mm/s, the stroke was 3 mm during 1000 cycles at a pressure of 1x10-7 hPa at room

temperature. In the experiments with ILs, the specimens were ultrasonically degreased and cleaned subsequently in toluene, isopropanol and petroleum ether baths, 10 min each, both before and after the experiments. The experiments were done always on fresh metal surface. ILs used in this work have been: 1) tributylmethylphosphonium bis(trifluoromethane

sulfonyl)amide([Bu3MeP][Tf2N]), 2) tributylmethylamonium bis(trifluoromethane

sulfonyl)amide ([Bu3MeN][Tf2N]) and 3) tributylmethylphosphonium dimethylphosphate

[Bu3MeP][(MeO)2PO2]. 2.3. a-C coatings In case of a-C coatings, the balls were Al2O3, 3mm in diameter and the disc was coated stainless steel AISI 440C. The normal load was in the range from 2 to 12 N, applying a sliding velocity of 3 mm/s and 3 mm stroke, during 500 cycles, at atmospheric pressure and at 1x10-6 hPa under room temperature. Three a-C coatings were developed and characterized in this work: 1) IBAD-DLC doped during deposition with a Si-O-C-H based oil while cracking the condensing oil on the surface by energetic Cr ions implanted by an industrial scale ion implanter from Danfysik; 2) Si-DLC obtained by physical vapor deposition (PVD) and doped with Si and H on the top of a CrN adhesion layer, and 3) DLC-TR obtained by PVD on the top of a CrN adhesion layer. Carbon coatings were tested as deposited. For IBAD-DLC, Cr+ ions were implanted with the energy of 70 keV. During the film deposition, Si-O-C-H oil was evaporated in the deposition vacuum chamber. Oil vapors were condensing on the surface and incorporated into the coating/substrate by the impinging energetic Cr ions. The coating was dense and quite polished. The PVD coatings were produced in a CemeconCC 800/9 unit applying a DC configuration with graphite and Cr targets. For Si-DLC, additional Si target was also used. The working gas was a mixture of Ar and acetylene at a pressure of 450 mPa. DC substrate bias voltage was 65-165 V. The substrate temperature during the deposition was around 200 ºC. Thickness of the coatings was about 1.5 µm (~ 3 µm including the CrN adhesion layer and the transition layer form the CrN phase to the pure DLC top layer). 3. EXPERIMENTS WITH ILs: RESULTS AND DISCUSSION

The ILs with [Tf2N] anion showed lower coefficient of friction compared to the IL with [(MeO)2PO2] anion (Figure 2). For the ILs with [Tf2N] anion, the surface of the wear tracks was quite smooth, although with the signs of chemical attack. In turn, for

Mass-spectrometer Load-lock chamber

Main chamber

[Bu3MeP][(MeO)2PO2] worn surfaces presented aspects characteristics for high abrasion and adhesion. This could imply that in our case the reduction of friction and wear is controlled mainly by the anionic moiety. The ILs having the same anion [Tf2N] showed similar friction and wear performance, although the IL with phosphonium cation exhibited a slightly lower coefficient of friction and smaller wear scar than the IL with ammonium cation. The effect of anion on friction and wear can be seen from comparison of the ILs having the same cation.

Figure 2. Tribological behaviour of the ionic liquids a) friction coefficient and b) wear volume.

The MSGE spectrometry revealed the existence of a lapse of time between the beginning of the tribotest and the moment when the organic molecules released from the ionic liquids are detected (Figure 3). These molecules are due to the rupture of the cationic organic chains. For ILs with [Tf2N], anion emission of CF3 could be observed immediately after the beginning of the experiment. Initially, the coefficient of friction is relatively high but after a short induction period, there is the formation of a protective tribolayer and consequent reduction of friction coefficient. These findings suggest that emission of CF3 seems to act like a catalyst of the tribochemical processes associated with detachment of trifluoromethyl from [Tf2N] anion that can further react with titanium substrate or recombine and pass to the gas phase. Titanium fluoride tribolayer could promote the tribodesorption of ionic liquid cation fragments (CH3

+, C2H5+, C3H7

+, C4H9+). XPS analysis

carried out on specimens lubricated with [Bu3MeP][Tf2N] and [Bu3MeN][Tf2N] confirmed the formation of a boundary film containing sulfide and

inorganic fluoride both inside and outside the wear tracks. This result has demonstrated high chemical reactivity of the ILs with the substrate even without rubbing. The observed tribolayer, possibly, was responsible for decreasing the friction coefficient. However, [Bu3MeP][(MeO)2PO2] did not yield a phosphorous-rich boundary film as in case of steel [6, 30]. In terms of MSGE, [Bu3MeP][(MeO)2PO2] had longer lapse of time for the rupture of the organic cationic chains than other ILs. In this case, friction coefficient did not diminish showing that the tribolayer was not formed.

Figure 3. Evolution of various ionized gas fragments measured simultaneously with friction force before, during and after tribotest: a) [Bu3MeP][Tf2N], b) [Bu3MeN][Tf2N] and c) [Bu3MeP][(MeO)2PO2].

CONCLUSIONS on the ILs Thus, it can be concluded that [Tf2N] anion was more effective than [(MeO)2PO2] in generating a surface protective film on titanium sliding surfaces under the given tribometrical conditions.

c)

a)

b)

a)

b)

b)

Simultaneous characterization of MSGE and friction has demonstrated its high potential for deep understanding of the mechanisms and pathways of tribochemical reactions occurring in a difficult-to-access interface of a tribological contact in high vacuum. 4. EXPERIMENTS WITH a-C: RESULTS AND DISCUSSION 4.1. Structural characterization and mechanical properties Figure 4 shows cross-sections of two PVD coatings. After deposition, the PVD coatings were annealed in order to understand the temperature stability of the developed coatings. Raman spectrometry was used for structural characterization of the PVD coatings before and after annealing (Figure 5). For Si-DLC, no significant variation in Raman spectra could be observed after annealing to 300 oC for 4 hours. Neither hardness, H, nor elastic modulus, E, experienced any change after annealing maintaining the same values: H = 20 GPa and E = 150 GPa.

Figure 4. Microstructure of PVD coatings a) Si – DLC (scale 5 µm) b) DLC-TR (scale 10 µm).

Figure 5. Raman spectra of the PVD coatings: a) and c) before annealing; b) and d) after annealing to 300 oC in air. a) and b) Si-DLC; c) and d) DLC-TR. In the case of DLC-TR, after annealing G peak considerably decreased, while the ratio D/G increased. It seems that a graphitization of the coating with

temperature has been produced, changing mechanical properties. After annealing, the coating hardness slightly increased, while elastic modulus steeply increased from 104 to 174 GPa. These findings pointed out that Si doping allowed to stabilize a-C structure. 4.2. Tribological characterization under atmosphere Figure 6 shows CoF time series recorded during rubbing of three coatings under atmosphere for two normal loads: 2 and 12 N. All CoF time series were quite stable. Running-in was notable for Si-DLC, while for other two coatings it was almost negligible. The lowest steady state CoF (0.1) was found for the Si-DLC. The effect of the normal load was negligible for Si-DLC, but it was considerable for other two coatings. For DLC-TR, CoF increased from 0.12 to 0.20 when increasing the normal load; however, for IBAD-DLC CoF decreased from 0.15 to 0.10. A hypothesis that could explain this effect is that under higher load, the oil incorporated into IBAD-DLC during deposition process could release onto the surface acting as a lubricant. The increase in CoF for DLC-TR can be conceivably associated with its higher susceptibility to structural destabilization that can occur both under shear deformation and heating. Wear behavior for all three coatings was, in general, well aligned with the friction behavior (Figure 7). IBAD-DLC exhibited the lowest wear under low load. Among the PVD coatings, Si-DLC had less wear than DLC-TR.

Figure 6. CoF time series for three coatings under atmosphere at different normal loads: a) 2 N; b) 12 N. 4.3. Tribological characterization in vacuum (10-6hPa range)

500 1000 1500 2000 25000

1000

2000

3000

4000

5000

Intens

ity  (co

unts)

R aman  s hift  (cm -­‐1)

500 1000 1500 2000 2500

500

1000

1500

2000

2500

3000

3500

4000

Intens

ity  (co

unts)

R aman  s hift  (cm -­‐1)

500 1000 1500 2000 25000

1000

2000

3000

4000

5000

6000

7000

8000

Intens

ity  (co

unts)

R aman  s hift  (cm -­‐1)

500 1000 1500 2000 25000

200

400

600

800

1000

1200

Intens

ity  (co

unts)

R aman  s hift  (cm -­‐1)

a) b)

c) d)

a) a)

b)

When testing under high vacuum, lack of repeatability of the CoF behavior of the IBAD-DLC coating was observed probably due to gas-desorption. CoF of IBAD-DLC was the highest among three coatings at any normal load (Figure 8). Under higher load, IBAD-DLC experiments had to be interrupted due to excessive increase of CoF. Cracks were found on the wear track, especially under higher load, when the coating was severely worn out (Figure 9).

2 N 12 N

IBA

D-D

LC

Si-D

LC

DLC

-TR

Figure 7. Optical images of the wear scars after sliding experiments under atmosphere.

Figure 8. CoF time series for three studied coatings under high vacuum (10-6 hPa range) at different normal loads: a) 2 N; b) 12 N.

Concerning the PVD coatings, different behavior was observed under lower and higher normal loads. Under lower normal load, Si-DLC coatings showed the lowest friction coefficient (0.02) that was almost one order of magnitude lower than in the experiments under atmosphere. For DLC-TR coating was slightly higher but remains in low values. Under 12 N normal load CoF of Si-DLC reached quite low value (<0.05) after a short induction period, but this low value did not stabilize. After short period of time CoF started to grow and reached 0.15 at the end of the test. On the contrary, for DLC-TR CoF was quite steady and below 0.05 during the whole test. In order to study the durability of DLC-TR under given experimental conditions, longer tests were carried out (Figure 10).

2 N 12N

IBA

D-D

LC

Si-D

LC

DLC

-TR

Figure 9. Optical images of the wear scars after sliding

tests under high vacuum (10-6 hPa range).

Figure 10. CoF time series for DLC-TR in long-term experiments under high vacuum (10-6 hPa range) and at normal load of 12 N.

It was found that CoF was low during the whole tests, even after 1200s. After short running-in, CoF stabilized at values below 0.05. After approximately 700s CoF started to increase slowly, but not exceeded 0.07. The fact that CoF for DLC-TR was lower

a)

b)

under vacuum, than in atmosphere can be related with the hydrogen content in this coating. H-termination of cross-linked carbon network is associated with electrostatic repulsion of mating surfaces that decreases energy dissipation under mutual sliding [31]. The tribochemical generation of molecular hydrogen at the interface contributes to separation of the surfaces due to steric effect that also reduces the rate of energy dissipation and CoF [18, 32]. Under atmospheric pressure adsorption of water and oxygen at the sliding interface reduces the effect of electrostatic repulsion, therefore CoF is higher. Previous studies have demonstrated that under mild sliding conditions, such as in this work, tribochemical processes are driven by shear and fracture rather than by frictional heating [5, 17, 18, 33]. Slow but continuous hydrogen release from DLC TR during the tests can be the reason for slow CoF increase after 700 s of sliding. The inflection point can conceivably be corresponded to the minimal H concentration at the subsurface layer, at which low friction is still stable. Wear of Si-DLC was very low under low loads, but increased at higher loads, that was in line with the CoF behavior. For DLC-TR, wear was slightly higher than for Si-DLC under both lower and higher loads. In summary, Si-DLC coating had the lowest friction coefficient under atmospheric pressure and excellent lubricity under high vacuum at low normal load. This could be due to the combination of H and Si doping elements. Hydrogen provides good lubricity in vacuum, whereas Si is good for atmosphere through tribochemical reactions with water and oxygen and formation of amorphous silica at the interface [34, 35]. However, at higher loads in vacuum the coating failed quickly. DLC-TR had much higher load bearing capacity than Si-DLC under high vacuum conditions yielding CoF below 0.05. However, under atmosphere CoF for DLC-TR was quite high, especially at higher loads. Both under atmosphere and vacuum conditions, wear was moderate for DLC-TR and low for Si-DLC. Addition of Si was beneficial for friction under atmosphere and for wear reduction in all cases. 4.4. Tribological behaviour of DLC-TR under varying gas pressure Several sliding experiments have been carried out with DLC-TR coatings under varying gas pressures in the range from 103 hPa to 10-8 hPa in order to determine the critical gas pressure at which a high-friction mode switches to a low-friction one. Results of these tests are shown in Figure 11.

Figure 11. CoF time series for DLC-TR as function of the gas pressure. When the gas pressure decreased from the atmospheric one to 1 hPa (1 mbar), CoF decreased approximately two-fold, but the behaviour did not change significantly, i.e. CoF steadily increased during sliding, although with different friction levels. At gas pressures equal or lower than 10-4 hPa, the CoF behaviour notably changed. Running-in period with duration about 40 s can be clearly distinguished in all graphs. After running-in, CoF slightly decreased and stabilized. The difference in CoF between the running-in and steady state regimes decreased with decreasing gas pressure. In ultrahigh vacuum, the steady state regime changed to slowly decreasing trend at values below 0.05. 4.5. MSGE from PVD a-C coatings MSGE was characterized for the PVD coatings under varying load and sliding velocity in high vacuum at 10-8 hPa gas pressure range. Figure 12 shows ion current time series for various ion species acquired simultaneously with CoF (black) for DLC-TR. Experimental conditions for seven tests denoted in Figure 12 with numbers were the following: 1, 2 – F = 2 N, V = 3 mm/s; 3, 4 – F = 12 N, V = 3 mm/s; 5, 6, 7 – F = 12 N, V = 10 mm/s.

Figure 12. Time series of ion currents for various ionic species and friction force for seven sliding tests carried out for DLC-TR in high vacuum. (a) Complete image with friction force and (b) detail of the ion currents at lower intensity. For H2 the rate of MSGE increased with the increase of both the load and the speed. In a similar way behaved CoF. It appears that the rate of MSGE for alkanes and/or radicals depended more significantly on the load than on the sliding speed. This behavior can be explained considering different tribochemical processes involved in H2 and hydrocarbon emission as suggested earlier in [18, 36]. In fact, H2 emission can be associated mainly with the liberation of molecules trapped into the coating during deposition, and, to a smaller extent, to recombinative tribochemical reactions under intensive shear deformation [32]. In turn, hydrocarbons should be the products of tribochemical reactions similar to those proposed in [36] that involve detachment of methyl radical followed by its recombination with H or another methyl group. Therefore, the rate of tribochemical reaction should depend mainly on the contact area and, therefore, on the normal load. MSGE of H2 should depend on the volume of deformed material and, thus, on the contact area. Figure 13 shows the results of MSGE spectrometry for Si-DLC. The main emitted gas species were hydrogen and methane. MSGE rate was the highest just at the beginning of sliding, reaching the minimum friction. Then, MSGE rate decreased following nearly exponential law. When friction switched from the low level to the high one, the MSGE rate of methane further decreased, while the rate of H2 emission was not affected by changing of the friction regime.

Figure 13. Time series of ion currents for various ionic species and CoF for seven sliding tests carried out for Si-DLC in high vacuum. CONCLUSIONS on the a-C coatings

Tribological properties and MSGE of three different types of a-C coatings have been studied as function of the gas pressure and normal load. IBAD-DLC failed at atmospheric and vacuum conditions in low-cycle tests. Although under atmosphere and at higher loads CoF of IBAD-DLC was quite low, but the coating suffered from intensive wear. Si-DLC showed quite good tribological properties under atmosphere in low-cycle tests with the CoF as low as 0.05. Low friction was also observed for this coating under vacuum at lower normal load. However, the load-bearing capacity of Si-DLC in vacuum was poor and the CoF rose up to 0.15 at higher load, still presenting low wear. For DLC-TR, CoF under vacuum was lower than under atmosphere. At higher load the CoF for DLC-TR reached also excellent lubricity. With decreasing gas pressure, DLC-TR coating exhibited decreasing tendency of CoF that almost stabilized at the pressures below 10-4 hPa. Wear was moderate for DLC-TR and low for Si-DLC. For DLC-TR coating, hydrogen and C1/C2 hydrocarbons were identified among the emitted gases. For H2, the rate of MSGE increased with increasing load and speed. For hydrocarbons, the rate of MSGE depended stronger on the load than on the speed. For Si-DLC, MSGE of hydrogen and hydrocarbons release was slightly higher at the beginning of the experiment when friction was lower. Then MSGE rate of hydrocarbons exponentially decreased and nearly varnished after transition from low-level to high-level of friction. These findings suggested that MSGE is not a temperature-driven process, but can be associated with stress and strain generated in the mechanically affected zone [16, 33]. ACKNOWLEDGEMENTS The authors acknowledge the financing of the EMAITEK Program by the Basque Country. Parts of this work were also funded by the "Austrian COMET-Program" in the frame of K2 XTribology. Surface analysis by XPS was carried out within the "Excellence Centre of Tribology" (AC²T research GmbH). The partners also acknowledge the EUREKA Program Eurostars for financing through “Vacuum DLC” project E� 4653, the Spanish National Agency

CDTI for financing to BRUGAROLAS the Vacuum LUB Project, the project ADEC, a Research project for the benefit of SME’s with reference 3152239. This work was financially supported by the Spanish Ministry of Economy and Competitiveness with the participation of the European Regional Development Fund (FEDER) of the European Union under the projects CIT-420000-2009-53, RYC-2009-0412, BIA2011-25653 and IPT-2012-1167-120000. REFERENCES

1. Jones, W.R. & Jansen, M.J., Space tribology. 2000, NASA Center for Aerospace Information: Hanover, MD. p. 33.

2. Miyoshi, K., Solid Lubricants and Coatings for Extreme Environments: State-of-the-Art Survey, in TECHNICAL MEMORANDUM. 2007, National Aeronautics and Space Administration, Glenn Research Center: Cleveland, Ohio. p. 16.

3. ESA, E., Space product assurance. Thermal vacuum outgassing test for the screening of space materials. 2008, ECSS Secretariat, ESA-ESTEC: Noordwijk. p. 45.

4. Fischer, D., Untersuchung von Reibungsvorschungen mit Hlfe der Exoelektronenemission. 1970, Berlin Technical University: Berlin. p. 65.

5. Nevshupa, R.A., Roman, E., & de Segovia, J.L. (2013) Contamination of vacuum environment due to gas emission stimulated by friction. Tribology International, 59, 23-29.

6. Minami, I. (2009) Ionic Liquids in Tribology. Molecules, 14(6), 2286-2305.

7. Pagano, F., et al. (2012) Dicationic ionic liquids as lubricants. Proceedings of the Institution of Mechanical Engineers, Part J: Journal of Engineering Tribology, 226(11), 952-964.

8. Somers, A., et al. (2013) A Review of Ionic Liquid Lubricants. Lubricants, 1(1), 3-21.

9. Bermúdez, M. & Jiménez, A. (2012) Surface interactions in lubrication of titanium, aluminium, and titanium–aluminium alloys with the ionic liquid [C2mim]Tf2N under increasing temperature. Proceedings of the Institution of Mechanical Engineers, Part J: Journal of Engineering Tribology, 226(11), 977-990.

10. García, A., et al. (2014) Ionic liquids as a neat lubricant applied to steel–steel contacts. Tribology International, 72, 42-50.

11. Totolin, V., et al. (2014) Tribological investigations of ionic liquids in ultra-high vacuum environment. Lubrication Science, 26(7-8), 514-524.

12. Piras, F.M., Rossi, A., & Spencer, N.D. (2002)

Growth of Tribological Films:� In Situ Characterization Based on Attenuated Total Reflection Infrared Spectroscopy. Langmuir, 18(17), 6606-6613.

13. Krick, B., et al. (2012) Optical In Situ Micro Tribometer for Analysis of Real Contact Area for Contact Mechanics, Adhesion, and Sliding Experiments. Tribology Letters, 45(1), 185-194.

14. Nevshupa, R.A. & Nakayama, K. (2003) Triboemission behavior of photons at dielectric/dielectric sliding: Time dependence nature at 10-4-104 s. Journal of Applied Physics, 93(11), 9321-9328.

15. Nakayama, K. & Nevshupa, R.A. (2004) Effect of dry air pressure on characteristics and patterns of tribomicroplasma. Vacuum, 74(1), 11-17.

16. Nevshupa, R., et al. (2015) Tribochemical Decomposition of Light Ionic Hydrides at Room Temperature. The Journal of Physical Chemistry Letters, 2780-2785.

17. Mahrova, M., et al. (2014) Critical Insight into Mechanochemical and Thermal Degradation of Imidazolium-based Ionic Liquids with Alkyl and mPEG Side Chains. Journal of physical chemistry C, 118(39), 22544–22552.

18. Rusanov, A., et al. (2015) Probing the tribochemical degradation of hydrogenated amorphous carbon using mechanically stimulated gas emission spectroscopy. Carbon, 81, 788-799.

19. Mori, S., Kawada, T., & Xu, W.C. (1997) Tribochemical decomposition of formic acid on the nascent surfaces of steel formed by scratching. Applied Surface Science, 108(3), 391-397.

20. Lu, R., et al. (2009) Study of tribochemical decomposition of ionic liquids on a nascent steel surface. Applied Surface Science, 255(22), 8965-8971.

21. Kawada, S., et al. (2014) Tribochemical Reactions of Ionic Liquids Under Vacuum Conditions. Tribology Letters, 54(3), 309-315.

22. Eryilmaz, O.L. & Erdemir, A. (2007) Surface analytical investigation of nearly-frictionless carbon films after tests in dry and humid nitrogen. Surface and Coatings Technology, 201(16–17), 7401-7407.

23. Fontaine, J., et al. (2005) Achieving superlow friction with hydrogenated amorphous carbon: some key requirements. Thin Solid Films, 482(1-2), 99-108.

24. Koskinen, J., et al. (2001) Low friction ta-C films with hydrogen reservoirs. Diamond and Related Materials, 10(3-7), 1030-1035.

25. Martin, J.-M., et al. (2010) Gas-Phase Lubrication of ta-C by Glycerol and Hydrogen Peroxide. Experimental and Computer Modeling.

The Journal of Physical Chemistry C, 114(11), 5003-5011.

26. Kukiełka, S., et al. (2006) Composition, mechanical properties and friction behavior of nickel/hydrogenated amorphous carbon composite films. Surface and Coatings Technology, 200(22–23), 6258-6262.

27. Gayathri, S., et al. (2012) Tribological properties of pulsed laser deposited DLC/TM (TM=Cr, Ag, Ti and Ni) multilayers. Tribology International, 53(0), 87-97.

28. Lan, H., Kato, T., & Liu, C. (2011) Molecular dynamics simulations of atomic-scale tribology between amorphous DLC and Si-DLC films. Tribology International, 44(11), 1329-1332.

29. Nevshupa, R.A., et al. (2015) Ultrahigh vacuum system for advanced tribology studies: Design principles and applications. Tribology International, 86, 28-35.

30. Bermúdez, M.-D., et al. (2009) Ionic Liquids as Advanced Lubricant Fluids. Molecules, 14(8), 2888-2908.

31. Erdemir, A. & Donnet, C. (2006) Tribology of diamond-like carbon films: recent progress and future prospects. Journal of Physics D: Applied Physics, 39(18), R311.

32. Bai, S., et al. (2012) Friction Reduction Mechanism of Hydrogen- and Fluorine-Terminated Diamond-Like Carbon Films Investigated by Molecular Dynamics and Quantum Chemical Calculation. The Journal of Physical Chemistry C, 116(23), 12559-12565.

33. Nevshupa, R. (2009) The role of athermal mechanisms in the activation of tribodesorption and triboluminisence in miniature and lightly loaded friction units. Journal of Friction and Wear, 30(2), 118-126.

34. Knapik, M., et al. (2008) Chemical analysis of interface area in DLC/Si systems. Optical Materials, 30(5), 767-769.

35. Wu, X., et al. (2008) Tribochemical reaction of Si-DLC coating in water studied by stable isotopic tracer. Diamond and Related Materials, 17(2), 147-153.

36. Rusanov, A., et al. (2015) Tribochemistry of hydrogenated amorphous carbon through analysis of Mechanically Stimulated Gas Emission. Diamond and Related Materials, 55(0), 32-40.