1 Introduction

In sliding contact systems, most of the frictional energy is transformed into heat, which in turn leads to a rise of temperature in the contact interface [1, 2]. The frictional heat between tribo-pairs plays a critical role in tribological processes, and it can result in phase transformations on the tribo-pair surface, such as local melting or gluing, thermal distortions, hot cracking, oil film thinning and even vaporization, and further influence frictional stability [3,4,5,6]. In typical disk brake systems, the temperature rises are conducive to decreasing the friction coefficient; however, they increase wear and energy dissipation and decrease the brake reliability [3], but completely removing the wear of material surface and the generation of frictional heat is challenging.

Nanoparticles as additives in lubricating oil or as nanofillers in composites can effectively reduce friction and wear [7,8,9,10]. Nanoadditives containing Cu have attracted tremendous attention because of their unique characteristics that can improve the tribological properties of tribo-pairs with friction-reducing and anti-wear properties. The function mechanisms of the nanoadditives as lubricants have been attributed to ball-bearing effect, colloidal effect, protective film effect, small-size effect, and third-body effect [7, 8, 11]. Padgurskas et al. [11] reported the tribological properties of various lubricant additives, such as Fe, Cu, and Co nanoparticles. They also declared that Cu nanoparticles as additives in lubricating oil can effectively reduce friction and wear. However, nanoparticles used as additives in the lubricants easily oxidize and agglomerate during preparation and storage. These disadvantages weaken the friction-reducing and anti-wear properties to some extent and even limit their tribological applications.

During the past few years, Cu nanoparticles have been synthesized through different methods, such as vacuum vapor deposition, radiolytic reduction, microemulsion techniques, sonochemistry, thermal decomposition, and chemical reduction method, in both scientific and industrial fields [12, 13]. Thermal decomposition is a novel method that has emerged in recent years for the synthesis of stable monodispersed copper nanomaterials. In our previous work, we developed a facile and controllable synthesis method under low temperature to prepare hybrid particles (Cu and Cu2O) through the thermal decomposition of copper formate–octylamine complexes as the organic precursor [14, 15]. We also found that the reaction temperature, time, and content of the precursor influence the crystallite morphology [15]. These findings suggest that Cu nanoparticles in situ generated in organic precursors at the friction interface may develop friction-reducing and anti-wear abilities.

Sutter et al. [16] reported that the frictional heat is aroused by the instantaneous friction of surface asperities, and the heat is mainly continuously stellate distributed around the asperities. The research also indicated the highest temperature rise of 1100 ℃ on the surface of tribo-pairs, but the local high-temperature region is less than 100 μm in diameter. Kalin et al. [17] studied the flash temperature rises of friction pairs under the fretting condition. They found that the flash temperature could reach 1000 ℃ under dry friction condition and that the lower limit value of flash temperature is approximately 400℃ in the lubricating state.

Rodríguez Ripoll et al. [18] investigated in situ formation of MoS2 and WS2 tribofilms by the synergy between transition metal oxide nanoparticles and conventional sulfur-containing anti-wear and extreme pressure additives. They indicated that the formation of these low friction tribofilms can be obtained under reciprocating sliding contact and under extreme pressure conditions. Even so, limited reports are available in the literature to improve tribological properties through the real-time decomposition of organic precursors under the action of frictional heat on the contact interface. This study aims to explore the effect of frictional heat on the tribological properties and tribological mechanisms during sliding in copper formate–octylamine complexes as the organic precursor. Interestingly, Cu nanoparticles are induced by frictional heat, and the detailed formation mechanism is discussed in this work.

2 Experimental Materials and Methods

2.1 Materials and Specimens

All chemical reagents used in our experiments were analytical reagent grade. Copper formate tetrahydrate was obtained from Guanghua Chemical Co. Ltd. Octylamine was purchased from Aladdin.

Two types of test materials, a 316L stainless steel (SS) flat against a 316L SS ball and a medium carbon steel (the trademark of material is 45 steel in China) flat against AISI 52,100 standard bearing steel (GCr15) ball, were used as tribo-pairs, respectively. The flat specimens with size of 30 mm × 15 mm × 5 mm were polished to an average surface roughness (Ra) of about 0.04 μm. Ball specimens with a diameter of 40 mm and a surface roughness (Ra) of 0.04 μm were used as counterparts. The chemical composition and mechanical properties of the test materials are listed as follows: 316L SS (in wt%: 17.2 Cr, 13.6 Ni, 1.2 Mn, 0.63 Si, and 2.2 Mo) with a hardness of 172 HV, medium carbon steel (in wt%: 0.45 C, 0.26 Si, 0.65 Mn, 0.16 Cr, 0.23 Ni, 0.21 Cu) with a hardness of 197 HB, and GCr15 steel (in wt%: 1.0 C, 1.49 Cr, 0.31 Mn, 0.26 Si, 0.009 P, 0.004 S) with a hardness of 766 HV.

2.2 Experimental Procedure

Copper formate tetrahydrate (Chemical formula Cu (HCOO)2·4H2O) was mixed with octylamine (Chemical formula NH2C8H17) at two molar ratios of 0.25:1 and 0.5:1, transferred into a 250 mL three-necked flask and then stirred for 1 h at 35 ℃, as shown in Fig. 1(I–II). Then, the precursor of copper formate–octylamine complex was formed and was stored in a dark place for 12 h (Fig. 1–III). For comparison, pure octylamine solution (marked as “O”) and saturated copper formate solution (marked as “C”) were also prepared. These four lubricants were denoted as O:C = 1:0 solution (simplex octylamine), O:C = 1:0.25 solution, O:C = 1:0.5 solution, and C solution (simplex copper formate), respectively.

Fig. 1
figure 1

Preparation of the precursor as lubricant

Full size image

The thermal stability of the precursor (Fig. 1VI) was investigated by thermogravimetric analysis (TGA) using a thermal analyzer (Q5000) under the protection of nitrogen atmosphere. To further analyze the characteristic of the pyrolysis product of the lubricant, the precursor was heated to 130 ℃ and started to decompose in nitrogen atmosphere. The color of the precursor gradually changed from light blue to dark brown after the reactions continued for 15 min to 5 h. The obtained pyrolysis products were examined by X-ray diffraction (XRD, X’PERT PRO, PANalytical) with Cu Kα radiation (k = 1.5406 A) within 10°–80°.

Tribology tests were performed using a UMT-3 tribometer in the ball-on-plate reciprocating wear mode (Fig. 1–VII) under lubrication with an applied load of 10 N, a frequency of 4 Hz, a sliding cycles number of 5000, and a slide stroke of 7.5 mm. All tests were performed at room temperature with a relative humidity of 55%, and each test was repeated at least three times to ensure the reliability of the results. During the tests, the ball specimen was attached to the upper holder, which could rubbing on the lower flat specimen. In addition, a container was mounted to the lower holder and filled with lubricants so that the top of the steel ball was completely immersed with the lubricant. Prior to each test, the specimens were ultrasonically cleaned with acetone and benzine mixture for 10 min.

After the tribological tests, the ball and plate samples were ultrasonically cleaned in ethanol. The morphologies of the worn surfaces were observed by scanning electron microscopy (SEM, SU8010, Hitachi, Japan), and 3D surface profilometry (Bruker, Contour GT–K, USA). Furthermore, the chemical compositions and the chemical states of the typical elements on the worn surfaces were detected using X-ray photoelectron spectroscopy (XPS, Kratos AXIS Ultra DLD, Japan) and energy dispersive spectroscopy (EDS, Xflash 6160, Bruker, USA).

3 Results and Discussion

3.1 Friction Behavior Analysis

The coefficient of friction (COF) for the ball-on-plate reciprocating wear of ASIS 316L SS under four lubricants are shown in Fig. 2. The fluctuation of COF is dramatically larger in pure octylamine (O:C = 1:0, Fig. 2a) or copper formate saturated solution (C solution, Fig. 2d) than that in the copper formate–octylamine complexes (Fig. 2b and c). The COF reaches the smallest value as O:C = 1:0.5 (Fig. 2c), which indicates that the complex has a good lubrication effect. The COF gradually decreases with increasing cycle number within the first 300 friction cycles (red box in Fig. 2b and c) under the lubricant composed of copper formate–octylamine complexes, which remains at a low level and relatively stable after 1000 sliding cycles.

Fig. 2
figure 2

Time-varying friction coefficient curves of 316L SS in four lubricants

Full size image

A similar phenomenon can also be observed in the friction pair consisting of the middle carbon steel against GCr15 bearing steel. The average COF of two types of friction pairs in different lubricating media is shown in Fig. 3 to compare the effects of lubricants on the friction-reducing performance of friction pairs. The lowest COF is achieved when the complex ratios were O:C = 1:0.5 for the two friction pairs, indicating that the precursor as a lubricant with the complex ratio of O:C = 1:0.5 has a good friction-reducing efficiency.

Fig. 3
figure 3

Comparison of average friction coefficients of two types of friction pairs in different lubricating media

Full size image

3.2 Wear Surface Analysis

The low COF usually reduces the surface wear. Therefore, the wear of the friction pairs can be dramatically mitigated when the precursor is involved in friction. Figure 4a illustrates the cross-sectional profile of the wear scars under the lubricants with three mixing ratios of copper formate and octylamine. The profile is obtained by measuring the 3D topography of the wear area (Fig. 4c and d). As shown in Fig. 4a, when the ratios of octylamine and copper formate are 1:0, 1:0.25, and 1:0.5, the depths of the cross-sectional wear profile are 4.43, 2.58, and 1.67 μm, respectively. That is, the anti-wear properties of the precursor at O:C = 1:0.5 is about 2.5 times better than that in the pure octylamine solution (O:C = 1:0). Furthermore, the volume loss (Fig. 4b) can be calculated according to the 2D cross-section profile for different lubrication media. When O:C = 1:0.5, the volume loss is less than 1/7 of the volume loss of the pure octylamine solution.

Fig. 4
figure 4

Wear scar profile of 316L SS in different lubricants: a 2D profiles; b volume loss; 3D profiles for: c O:C = 1:0.5; d O:C = 1:0

Full size image

The surface morphologies of the wear scar of 316L SS in different lubricating media are shown in Fig. 5 to analyze the effect of different lubricants on the surface morphology of the wear scar. As shown in Fig. 5a, the worn surface is full of ploughing where the material is loose when the pure octylamine solution is used as the lubricating medium, and the wear mechanism is mainly abrasive wear accompanied by the deep ploughing.

Fig. 5
figure 5

Wear morphology of 316L SS in different lubricating media: a O:C = 1:0; b O:C = 1:0.25; c O:C = 1:0.5

Full size image

However, when the precursor is used as the lubricant (i.e., O:C = 1:0.5 and O:C = 1:0.25), as shown in Fig. 5b and c, such deep ploughing does not appear on the worn surface and is replaced by the black regions that are strips in shape and accords with the sliding direction. Further details will be discussed below. In addition, significant delamination characteristics appear on local worn surface (Fig. 5c). This result indicates that the wear mechanism under precursor lubrication conditions is slight abrasive wear accompanied by significant fatigue wear.

Furthermore, EDS mapping images of the worn scar containing the black stripe regions are shown in Fig. 6. The results show that the black stripe regions in Fig. 6a is rich in elements such as Cu (see Fig. 6b), C (see Fig. 6c) and O (see Fig. 6d). Moreover, EDS was also conducted for the areas marked by yellow rectangle on the 45 steel surface in Fig. 7a, and the EDS mapping images of the worn surface sliding in the copper formate saturated aqueous solution are shown in Fig. 7b. Therefore, analysis results show that the same phenomenon can also be verified on the tribo-pairs for 45 steel against GCr15.

Fig. 6
figure 6

Surface morphology (a) and elemental distribution on the wear surface of 316L under lubrication precursor condition with O:C = 1:0.5 and three corresponding element distribution maps (b–d)

Full size image
Fig. 7
figure 7

Position for EDS mapping on the wear surface of 45 steel under lubrication precursor condition with O:C = 1:0.5 (a, b) and three corresponding element distribution maps (c–e)

Full size image

Figure 8a shows the EDS spectra of black stripe regions on the worn surface of 316L in different media and 316L substrate. Cu peaks are not detected in the EDS spectrum as sliding in the octylamine solution and copper formate saturated aqueous solution. However, the obvious Cu peaks are observed in the EDS spectrum of the worn surface when the precursors with composing ratios of O:C = 1:0.5 and O:C = 1:0.25 are used as lubricants. Figure 8b shows the atomic percentage between the worn surface and the 316L substrate under four lubricants. Compared with that on the 316L substrate, the oxygen content on the worn surface increases with varying degrees because of oxidation. Interestingly, the Cu contents in the black stripe region are 3.50% and 8.34%, when the ratios of the precursors as lubricants are O:C = 1:0.25 and O:C = 1:0.5, respectively, which means that the sliding of the sample in these two solutions results in the formation of Cu-containing substances in the black stripe region. The Cu element on the worn surface of the saturated copper formate solution and pure octylamine solution slightly differs from that on the 316L substrate. This result indicates that Cu-containing substances are not produced during the friction.

Fig. 8
figure 8

Comparison of EDS spectra (a) and atomic percentage of black stripe regions on the worn surface of 316L in different medium environments (b)

Full size image

Further analysis on the residual lubricating medium on the worn surface (Fig. 9a and b) shows that a large number of aggregated spherical nanoparticles with a size range of 100 nm to 500 nm exist in the residual lubricating medium. EDS mapping images (Fig. 9c–d) reveal that these aggregated nanoparticles are rich in Cu and O. This phenomenon is the cause of the high Cu content in the dark gray areas on the worn surface shown in Figs. 7 and 8.

Fig. 9
figure 9

Morphology of nanoparticles produced on the worn surface and their element distribution characteristics: (a, b) SEM photos of nanoparticles; c Cu element distribution; d O element distribution

Full size image

3.3 Mechanism of Friction-Reducing and Anti-wear

Copper is a metal with good ductility, low shear strength, and low COF, its nanoparticles are often used as lubricant additives. Nanoparticles (such as copper and its oxides) can reduce friction through a small sliding resistance and protect the substrate from wear by preventing metal-to-metal contact [11, 19, 20]. As mentioned earlier, the precursor could generate nanocopper crystallites and Cu2O crystallites after thermally decomposition as shown in Fig. 1.

Figure 10a shows the TGA test result of precursor (corresponding Fig. 1VI). The TG curve shows that precursor decomposition began at 70 ℃ and finished at 160 ℃. Its decomposition temperature is significantly lower than that of anhydrous copper formate. The weight loss ratio is about 84.8%, which is consistent with the theoretical value of 85.6%.

Fig. 10
figure 10

a TGA curve of the precursor; b SEM image of the Cu2O microcrystals; c XRD pattern of the pyrolysis product

Full size image

Figure 10c shows the XRD pattern of the pyrolysis products (Fig. 1V), and the diffraction peaks of 36.52°, 42.45°, 61.50°, and 73.76° correspond to the Cu2O (111), (200), (220), and (311) planes (JCPDF 0,770,199). The diffraction peaks are very sharp, which indicated good crystallinity. The diffraction peaks of 43.33° and 50.68° correspond to the (111) and (200) planes of copper (JCPDF040836). The intensity of diffraction peaks of copper is weak, which indicates that the copper content is low. This result should be attributed to the fact that the Cu generated by the thermal decomposition of the precursor is easily oxidized to Cu2O. Furthermore, no pyrolysis product is further oxidized into CuO. Figure 10b shows the micromorphology of Cu2O microcrystals under free thermal pyrolysis. The SEM image indicates that Cu2O particles form in the shape of regular hexahedron, and the average edge length is 1.25 μm.

Generally, the surface of mechanical parts is not smooth, and it has a large number of microconvex. When the friction is conducted on a rough surface, instantaneous temperature points called flash temperature (Tf) can be generated in the area within a few micrometers. The flash temperature can be predicted by using a classical model from Jaeger’s work for a moving heat source. The formula can be expressed as [21, 22]

$$T_{{\text{f}}} = 1.13\sqrt {\frac{{la}}{v}}\;{\frac{{\mu pv}}{k}},$$
(1)

where μ is the dynamic friction factor, p is the contact pressure, v is the sliding velocity, l is the distance paralleling to the sliding direction, k is the thermal conductivity, and a is the thermal diffusivity.

The thermal diffusivity a also can be expressed as

$$a = \frac{k}{c\rho },$$
(2)

where k is the thermal conductivity of counter pair, c is the specific heat capacity, ρ is the density of tribo-pairs.

Therefore,

$$T_{{\text{f}}} = 1.13\mu p\sqrt {\frac{lv}{{kc\rho }}} ,$$
(3)

According to the above formula, the flashing temperatures of the friction surface of 316L SS in the three lubrication conditions are 296 °C, 275 °C, and 257 °C when O:C = 1:0, O:C = 1:0.25 and O:C = 1:0.5, respectively. As shown in Fig. 10a, the frictional flash temperatures are much higher than the thermal decomposition temperature of octylamine and copper formate complex). In other word, the local frictional flash temperature during the frictional process is sufficiently high to drive the precursor to be decomposed and form copper nanoparticles. The nanocopper could be produced by frictional heat and provides lower shear stability (low friction) and fast relaxation of contact stress between the friction pairs to prevent the deformation of the asperity contact and produce anti-wear and friction-reducing properties [2, 23].

As shown in Fig. 11, the properties of 316L worn surface (under precursor and simplex octylamine, respectively) are studied by XPS to explain the existence of copper nanoparticles and their friction-reducing and anti-wear mechanisms. The corresponding positions are marked in Fig. 5. The XPS results in Fig. 11a indicate that the main and satellite peaks of Cu 2p3/2 and Cu 2p1/2 could be detected on the worn surface lubricated by the precursor. However, no characteristic peak is found on the worn surface lubricated by pure octylamine or C solution (see Fig. 11b and its inset). Furthermore, Fig. 11a shows that the Cu 2p3/2 spectrum is located at about 933 eV fitting into two components, corresponding to Cu1+ in Cu2O (Cu2+ at 932.6 eV) and Cu2+ in CuO (Cu2+ at 934.8 eV) [24]. The Cu 2p1/2 spectrum in the binding energy range of 950–958 eV can be deconvolved into two peaks. The peaks centering at 952.6 and 954.8 eV are assigned to Cu2O and CuO [24]. According to our previous work [15] and other reports [25, 26], CuO and Cu2O are not produced by the thermal decomposition of the precursor complexes. However, the addition of oleic acid produces Cu2O under similar reaction conditions because of the strong reducibility of nano-Cu [14]. Conversely, no nanocopper can be detected on the worn surface possibly because the friction process in this investigation is carried out in the air environment (oxygen-rich). The copper nanoparticles generated by the frictional thermal pyrolysis are oxidized into Cu2O and CuO under the effect of tribo-oxidation. Therefore, the synthesis steps and chemical reactions are concluded as follows:

$${\text{Cu}}\left( {{\text{COOH}}} \right)_{{2}} + {\text{ 2NH}}_{{2}} {\text{C}}_{{8}} {\text{H}}_{{{17}}} \to {\text{Cu}}\left( {{\text{COOH}}} \right)_{{2}} \left( {{\text{NH}}_{{2}} {\text{C}}_{{8}} {\text{H}}_{{{17}}} } \right)_{{2}} ,$$
(4)
$${\text{Cu}}\left( {{\text{COOH}}} \right)_{{2}} \left( {{\text{NH}}_{{2}} {\text{C}}_{{8}} {\text{H}}_{{{17}}} } \right)_{{2}} \to {\text{Cu }} + {\text{ H}}_{{2}} + {\text{ 2CO}}_{{2}} + {\text{ 2NH}}_{{2}} {\text{C}}_{{8}} {\text{H}}_{{{17}}} ,$$
(5)
$${\text{Cu }} + {\text{ O}}_{{2}} \to {\text{Cu}}_{{2}} {\text{O }} + {\text{ CuO}},$$
(6)
Fig. 11
figure 11

XPS spectra of different conditions on the worn surface of 316L (corresponding to the mark in Fig. 5): a under precursor condition; b under simplex octylamine condition

Full size image

Different from the Cu2O microcrystals in Fig. 10b, the nanoparticles produced on the worn surface exhibit a microsphere shape (see Fig. 9a and b), with the diameter range of only about 100–500 nm, and are significantly smaller than those in Fig. 10b. This phenomenon may be attributed to the fact that the nanoparticles are produced in the shearing and rolling of the friction process on the friction interface.

Figure 12 shows the reduced the friction and wear mechanism for the tribosystem. This result can be explained as follows: (i) frictional heat induces the thermal decomposition of precursors to generate nanoparticles during rubbing (see Fig. 10b). That is, the nanocopper could be in situ generated at the frictional interface. (ii) The nanoparticle deposition onto the rubbing surface results in effective rolling friction and self-lubricating effect [20]. In the beginning of the friction process, more and more nanoparticles are generated, causing the COF between the friction pairs tends to decrease (see Fig. 2b and c). After that, the COF maintains a low and stable value (see Figs. 2 and 3). (iii) As the friction proceeds continuously, the worn surface becomes rough, but the nanoparticles are filled into the scar and grooves on the worn surface. Then, the protective film forms when nanoparticles are compacted on the contact surface at a high contact pressure and temperature during friction [20, 27]. As a result, the wear resistance is improved significantly. Meanwhile, the deposited nanoparticles transform into a self–healing film in the frictional process, which makes the friction surface flat and smooth. This so-called self-repairing capacity of the nanoparticles also decreases the frictional force.

Fig. 12
figure 12

Friction-reducing and anti-wear mechanism of the precursor between tribo-pairs surface

Full size image

4 Conclusions

A copper formate–octylamine complex was prepared as a precursor, and tribological tests were performed to explore the role of Cu nanoparticles generated by the thermal decomposition of precursor during the friction process to achieve friction-reducing and anti-wear properties. For this reason, the tribological properties of tribo-pairs in different lubricating media were investigated in detail. On basis of the above works, the following conclusions can be drawn:

  1. 1.

    The prepared precursor decomposes at about 70 °C, and the decomposition finishes completely at 160 °C. The temperature mentioned above is significantly lower than the local flash temperature between tribo-pairs during the friction. That is, the frictional heat could cause the precursor to generate copper nanoparticles by thermal decomposition. This method can achieve the real-time generation of nanocopper between the friction pairs and effectively avoid the agglomeration of adding nanoparticles in the similar lubrication. Therefore, this method has a potential application in friction-reducing and anti-wear.

  2. 2.

    The complex composed of copper formate–octylamine is an extremely good lubrication because it can be decomposed to produce copper nanoparticles and can effectively reduce friction and wear. Compared with pure octylamine or copper formate solution, the copper formate–octylamine complex with two ratios can effectively reduce the friction coefficient between the two friction pairs and greatly reduce the wear. The friction coefficient is stable and smallest, and the wear volume is 1/8 less than that of the pure octylamine solution when the compounding ratio is O:C = 1:0.5.

  3. 3.

    When the precursor is used as a lubricant, the SEM photograph shows black stripe regions on the worn surface, and this area is rich in Cu corresponding to Cu2O and CuO. The worn surface is flat and smooth, attributing to the self-lubricating and self-repairing effects of the nanoparticles. The wear mechanism is mainly slight abrasive wear, accompanied with significant fatigue wear and oxidative wear. In pure octylamine or copper formate solution, grooves are obvious but Cu is not detected on the worn surface.