Synergistic Lubrication for Textured Surfaces Using Polar and Nonpolar Lubricants

In the last few years, the excessive use of fossil fuels and air pollution have played a significant role in global warming, posing a serious threat to mankind [1,2]. Vehicles use about 30% of the energy consumed by the public transportation industry, with around one-third of that consumption induced by friction and wear of the moving mechanical components [3,4]. Friction and wear not only increase the energy loss but also shorten the lifetime of the components. As a result, it is critical to reduce friction and wear in mechanical systems. Traditional modern tribological improvement approaches, including coatings, lubricant additives, and solid lubricants, have emerged and demonstrated their promising potential in engineering fields [3–10]. However, these methods are subjected to a much complicated and costly procedure. Using a simple lubricant is the easiest way to improve the tribological behavior at the interface, since a lubricant can protect the friction surface and optimize the friction coefficient owing to its viscosity and chemical properties. Therefore, many studies had investigated the influence of lubes properties on the friction performances [11–15]. The chemical properties of lubricants have a crucial role in the tribological performance especially in the boundary lubrication state. The lubricant with polar structure can improve the state of boundary lubrication by forming a stable physical structure with the metal surface even at high load [13,14]. Gryglewicza et al. [15] improved the tribological properties by adding polar additives to form hydrodynamic films at high temperatures. The study by Hwang et al. [14] showed that the polarity of the oxygen atom-electron pair in the lubricant exhibited better lubricating performance in the boundary lubrication. Furthermore, lubricants with different chemical properties exhibit different tribological properties [16]. Zolper et al. [17] analyzed the relationship between lubricant properties and the molecular structure of poly-α-olefin and polydimethylsiloxane. The friction and wear properties of the ester oil and mineral oil were also analyzed [18,19]. Crook [20] investigated the relationship between the effective viscosity and frictional force. Zolper et al. [21] established a molecular rheology model to look at the relationship between the lubricant molecular structure and rheological behavior.

Furthermore, the surface morphology and roughness of the friction interface also significantly affect the tribological behavior. In fact, texturing that is fabricating appropriate pits, grooves, or protrusions on the surfaces can reduce friction and provide an anti-wear effect [22,23]. Laser treatment on the surface of piston rings has been shown to effectively reduce engine fuel consumption and friction by up to 4% and 50%, respectively [24]. Depending on the lubrication states, the textured surface plays distinct and vital functions including: the storage of wear debris [25–27], the hydrodynamic lift effect, the surface roughness peak contact reduction [28–31], and the lubricant storage [32,33]. Tribological behavior under an oil lubricant is affected by the lubrication regime. The film thickness ratio (λ), which is the proportion of minimum oil film thickness to surface roughness, can be used to determine the lubrication regime. The lubrication regime of oil lubrication can be classified as boundary (λ < 1), mixed (1 < λ < 3), and hydrodynamic (λ > 3) lubrication condition. The lubrication condition of a mechanical system is generally assumed to be under mixed lubrication between running and stop periods, implying that the contact of asperities and hydrodynamic pressure exist simultaneously [34]. The degree of hydrodynamic pressure generation can be improved by using surface texture [35]. Meanwhile, the lubricant chemical of the lubricants can have a significantly influence on the formation of the boundary film [36,37].

Although some previous research work has been reported on the effect of texture characteristics on tribological performance, the majority of these studies have focused on the relationship between surface texturing and operating conditions. Furthermore, numerous studies separately concentrate on improving the tribology performance by optimizing the lubricant with additives. In this paper, the synergistic effect of surface texturing and various lubricants on friction and wear of the tribopair was investigated. Three different oils were used including, a low-viscosity and nonpolar white oil, a high viscosity and nonpolar silicone oil, and a highly viscous and polar castor oil. The theoretical oil film thickness was simulated. The contact angle and tribological test were performed to study the influence of viscosity and polarity of oil on the tribological performance of surface texturing under various operating conditions. Moreover, the corresponding mechanisms are discussed based on surface analysis technologies.

Three types of lubricants were selected including: white oil (Sinopharm Chemical Reagent Co., Ltd, China), silicone oil (Dow Corning, America), and castor oil (Sinopharm Chemical Reagent Co., Ltd, China). Table 1 summarizes the characteristics of each lubricant, and Fig. 1 shows the molecular structures of silicone oil and castor oil.

The pin was made of T2 copper, while the disks used in the experiments were made of SUS304 stainless steel. The relevant parameters are shown in Table 2.

In order to evaluate the chemical structure of the lubricant, an attenuated total reflectance Fourier transform infrared (ATR-FTIR) spectroscopy analyses were used. The Fourier infrared spectrometer (Nicobet iS50) recorded a spectra in a range of 600–4000 cm⁻¹ under an ATR module. During the test, a few drops of each lubricant were placed on a cleaned ATR plate. The spectra were obtained after three scans at a resolution of 4.0 cm⁻¹.

The laser processing method was used to fabricate the textured disc. The laser machine was a BasiCube 10 (SCANLAB Company, Germany). The following laser parameters are used during the surface processing: (i) the laser spot diameter of 20 µm; (ii) the wavelength of 1064 nm; (iii) the output power of 12 W; (iv) the focal length of 78 mm; (v) the frequency adjusted to 20 kHz; (vi) the scanning speeding at 200 mm/s; and (vii) the laser fluence of 4.4 × 10⁵ J/cm². In addition, the laser filling method was a combination of straight lines and circles.

Previous studies have found that circular concave surface texture with geometrical parameters ranging from 100 µm to 1000 µm in diameter and 5% to 15% in area density exhibits a favorable tribological performance [38–41]. Based on the previous investigations and machining accuracy, the parameters selected for processing were circular texture pits grouped in a matrix. The area ratio of circular texture was 10%, with a diameter D of 100 µm, a depth of 20 µm, and the center-to-center distance L of 280 µm as illustrated in Fig. 2(c). Furthermore, the three-dimensional topography of one pit was displayed using a confocal laser as shown in Fig. 2(b). The contour at the line in Fig. 2(b) is depicted in Fig. 2(d), at a depth of 30 µm. Following the laser texturing process, the surface was meticulously polished using 400#, 800#, and 1200# grain-sized sandpapers to keep the surface roughness (Ra) under 0.08 µm. Afterwards, the polished discs were immersed in ethyl alcohol solution and cleaned for 10 min using an ultrasonic wave concussion to remove any contaminants. The texture pits had a circular shape and arranged in an array as shown by scanning electron microscopy (SEM) (Japanese-made SU-8020, Hitachi, Tokyo, Japan) in Fig. 2(a).

A tribological tester (UMT-2, CETR, USA) was used to examine the tribological properties of the surface. The length of the pin is 6 mm, and the diameter is 2 mm with the surface roughness (Ra) under 0.23 µm. The relative motion mode was reciprocating, and the relative sliding distance was 6 mm shown in Fig. 3. The temperature during the experiment was 25 °C. The sliding speed frequency was 2 Hz, and varied loads of 5 N, 10 N, and 15 N were used in this research, with the contact pressures summarized in Table 3. Before the experiments, 0.3 ml of lubricants were used to wet the friction interface [36]. The worn surfaces were examined by a Hitachi SU-8020 SEM and the FTIR data of lubricants after used were collected by the Nicobet iS50 ATR-FTIR system.

The ATR-FTIR spectra of the three lubricants used are shown in Fig. 4. The stretching vibrations of CH₂ and CH₃ caused the powerful signals revealed in the spectra of white oil. For silicone oil, the spectrum depicted the signals of polydimethylsiloxane. Each signal on the spectrum was marked with the corresponding functional group. There was a wide signal for castor oil at about 3400 cm⁻¹ due to the O–H vibrations related to the nature of castor oil. In addition, the relationship between functional groups and signals has been noted. It can be seen that white oil and silicone oil are nonpolar lubricants, while castor oil has a strong polarity.

Table 4 shows the lubrication regimes. White oil lubrication resulted in the thinnest film thickness, indicating that the system is in a boundary lubricating regime. The mixed lubricated state with a comparable oil film thickness was determined for the systems lubricated with silicone oil and castor oil.

Figure 5 shows the contact angles of the three different lubricants on bare and textured surfaces. Castor oil had a larger contact angle compared with silicone oil and white oil on patterned and non-pattern surfaces, and the silicone oil has the smallest contact angle among the three lubricants. According to Ref. [44], the surface energy of silicone oil, white oil, and castor oil gradually increases. At the same time, Fig. 5 also shows that castor oil had the largest intermolecular forces, followed by white oil, and the silicone oil displayed the smallest value. The contact angles of three lubricants on the textured surface were larger than those of the smooth surface. This phenomenon demonstrates how tailored texture manufactured by laser processing can reduce the surface free energy [45]. The higher the contact angle, the stronger the cohesion between the lubricant molecules. The friction surface has a few molecular layers of lubricants under the boundary lubrication. Moreover, because of the large molecular cohesion, there is a greater possibility of separating the mating surfaces and improving the tribology performance.

The curve of coefficient of friction (COF) versus time on bare and textured surfaces using different lubricants under 5 N load and 2 Hz sliding frequency was displayed in Fig. 6. The COF of the low-viscosity white oil on the smooth surface increased continuously, peaking at 1500 s, indicating that the lubricating oil film is broken. On the textured surface, the friction coefficient curve was not only low but also stable. This difference indicates that texture can store lubricant to a certain extent. For the silicone oil with high viscosity and less polarity, the friction coefficient on the smooth surface was much higher than that on the textured surface. The friction coefficients on both surfaces had multiple fluctuations over time. This trend, to some degree, also indicates that the lubricating film has a stable rupture condition during the friction process. This phenomenon might be related to the weak polarity of silicone oil and the precarious existence of the formed lubricating film.

For castor oil, which have a high viscosity and polarity, the COF on the smooth surface was very similar to that of the textured surface close to 0.06. Compared with the bare surface, the COF of the textured surface took less time to stabilize, and the friction coefficient remained constant without increasing. The textured surface processed by the selected parameters has the advantage of improving the friction performance.

The wear tracks under the load of 5 N are presented in Fig. 7. The wear track was severe when using white oil. However, it was greatly reduced when using silicone oil. Furthermore, the track became unnoticeable using castor oil compared to silicone oil. The white oil had the deepest wear depth, while the castor oil had the least. Furthermore, as shown in Fig. 7(a), there were some furrows on the wear track when using white oil, whereas there were few furrows on the wear tracks for silicone oil and castor oil as illustrated in Figs. 7(b) and 7(c).

Because the white oil had the lowest viscosity, the lubricant film could not be generated sufficiently to separate the mating surface, leading to a relatively serious asperity interaction, increasing substantially the COF. On the other hand, silicone oil and castor oil having a higher viscosity could more efficiently separate the mating surface to generate a thick lubricant film. As a result, lowered COFs were obtained. Silicone oil had a higher COF compared to castor oil, indicating that the chemical features of lubricating oils may potentially impact the resulted tribological behaviors. As seen in Fig. 4, castor oil molecules exhibited polar hydroxyl groups, which approached the surface and interacted to produce adsorption layers, contributing to COF reduction.

As previously stated, the lubricating performance of lubricant molecules is related to their ability to produce a lubrication film and interact with mating surfaces. The tribological performance of the three oils were further investigated under higher loads. Figure 8(a) shows a comparison of the COF for white oil, silicone oil, and castor oil under the load of 10 N. The lowest friction coefficient of around 0.09 was obtained with castor oil. The friction coefficient of silicone oil was about 0.2. The friction coefficient with white oil was still the highest approximately 0.22. The COF under white oil and castor oil as lubricants reached a stable stage after approximately 10 s, while it took nearly 700 s for silicone oil to stabilize. When compared to white oil and castor oil, the friction coefficient curve of silicone oil fluctuates in the final stabilization stage.

Figure 8(b) shows the time-dependent variation in friction coefficient curves for white oil, silicone oil, and castor oil under the load of 15 N. When the lubricant was castor oil, the friction coefficient was 0.086, and the running-in period was about 10 s. In the stable zone, the friction coefficient curve was very smooth. For white oil and silicone oil, the friction coefficients were similar in the stable zone, whereas the friction coefficient of white oil was slightly lower than that of silicone oil. The running-in period of white oil was about 100 s, which was much lower than that of approximately 700 s for silicone oil. In addition, the friction curve with silicone oil as a lubricant still fluctuated slightly in the stable zone.

Figure 9 illustrates the energy dispersive spectroscopy (EDS) spectra of the worn surfaces with three lubricants under the load of 10 N. After removing the residual lubricants, the chemical elements of the surfaces were detected. For white oil, there were many evident furrows on the surface (Fig. 9(a)). The surface elements, in addition to the elements carried by the metal disc, contained carbon and oxygen elements, and carbon elements were distributed on the entire surface as seen in Fig. 9(b). There is silicon adhesion on the surface (Fig. 9(d)) when using silicone oil, and the silicon elements were mainly concentrated in the furrows as seen in Fig. 9(c), and the copper element on the surface had a large amount of residues [46]. There were only few faint scratches on the worn surface using castor oil as the lubricant (Fig. 9(e)). Furthermore, there was a significant oil residue on the surface, composed of carbon (Fig. 9(f)), demonstrating that castor oil has a strong adhesion on the surface.

The wear-rate of the pin using white oil, silicone oil, and castor oil at a sliding speed of 2.0 Hz is shown in Fig. 10(a). White oil has the largest wear-rate, followed by silicone oil, and castor oil as the lowest wear-rate. When using white oil, the wear-rate obviously increased as the load increased. For silicone oil, the wear-rate further increased as compared to white oil. For castor oil, the wear-rate decreased gradually as the loads increased. The slightest differences in the wear-rates of silicone oil and castor oil are mostly due to the high viscosity, leading to the development of the oil film bearing capacity. Simultaneously, the lubricant containing polar molecules can be adsorbed on the surface of the friction pair to reduce the wear. According to the friction state estimation model described in the literature, the experiments are under the mixed lubrication state.

Figure 10(b) shows the average COFs of the tribopair using various lubricants. For castor oil, the average friction coefficient was stable by approximately 0.085, 0.090, and 0.086 under 5 N, 10 N, and 15 N, respectively. The friction coefficient had minimal change as the load increased, and the running-in duration was less than 10 s. For white oil, the average friction coefficient was at around 0.211, 0.209, and 0.203 at 5 N, 10 N, and 15 N, respectively. The friction coefficient changed slightly with the increased load, but the running-in period became shorter. For silicone oil, the average friction coefficient was stable at 0.181, 0.193, and 0.208 for 5 N, 10 N, and 15 N, respectively. The friction coefficient increased significantly as the testing load increased, and the running-in period duration became longer.

Figure 10(c) shows the Stribeck curves of silicone oil and castor oil versus the Stribeck–Hersey number, which considers the effects of pressure, velocity, and viscosity. It is meaningful to look at the Stribeck curves of the two lubes with their molecular structures (Fig. 1). Since the viscosity of silicone oil is 510 mm²/s and that of castor oil is 610 mm²/s, the castor oil lowest COF trough was lower than that of silicone oil. For the castor oil with polar hydroxyl group, the COF was lower than that of silicone oil with no polar group, especially at the front portion of the curve, which revealed that chemical properties with polar groups in boundary and mixed lubrication regime contributed to the castor oil low COF [14,46].

Figure 11 reveals the comparison of the infrared spectrum of the lubricants before and after the tribological testing with a load of 15 N. There was no change in the chemical functional groups before and after friction for silicone oil, reflecting that silicone oil is a kind of high temperature resistant and chemical stable polymer lubricating oil. After the tribological reaction of white oil, the type and content of peaks around 800–660 changed, indicating the formation and change of the C–H bond. These changes also demonstrate that white oil has experienced some chemical decomposition or oxidation. For castor oil, the fluctuation was around 3000, indicating that the C=C had changed, and the fluctuation around 1300–800 manifested that the ether (C–O), hydroxyl (C–OH), carbonyl (C=O), and acid C(O)OH had varied. It was demonstrated that castor oil experienced an oxidative hydrolysis reaction, which is the same as the tribochemical reaction of castor oil mentioned in the previous literature [47–52]. The hydrolysis of castor oil produces more polar molecules, which are more favorable for forming boundary films.

For the boundary and mixed lubrication, the friction force depends entirely on the residual roughness peak, contact degree, and oil shear force. The residual roughness peak contact degree is small when the oil film thickness is large. The thickness of the oil film is proportional to the viscosity of the lubricating fluid. According to the Renold equation, the greater the viscosity of the lubricating fluid, the greater the thickness of the formed oil film. The dynamic viscosity of white oil is 27.5 mPa s, and the dynamic viscosities of castor oil and silicone oil are as high as 583.2 mPa s and 515 mPa s separately. When white oil was used as a lubricant, the oil film thickness was thin, and the residual roughness peak contact degree was high, as seen in Fig. 12(a). The oil film formed during the friction process was thin, and the contact area of the rough peak of the contact surface was large. Thus, the coefficient of friction was relatively high. When silicone oil and castor oil were used as lubricants, the film thicknesses are similar as shown in Table 4. The boundary lubrication contacts at the interface were improved, and the residual roughness peak contact degree became lower. At the same time, the friction force became smaller, as seen in Figs. 12(b) and 12(c), respectively. For silicone and castor oil, it has been evaluated that the friction coefficient of castor oil was much lower than that of silicone oil, and the running-in period was also shorter. The superior lubricating performance of castor oil is due to the polar molecules of castor oil that can adhere to the friction surface to form an orderly and closely packed boundary adsorption film, reducing the severe interaction between asperity contacts [51,52]. In comparison, silicone oil is nonpolar and could hardly adhere to the contact surface, providing little protection to asperity contacts. Thus, among the three investigated oils, castor oil displayed the lowest COF and the shorter running-in time period.

This study reported a novel strategy to achieve enhanced lubrication for steel tribopairs by combing surface texturing and a desirable oil with appreciable viscosity and polarity properties as lubricant. The obtained results are listed below:

1. The designed and fabricated texture surface can improve the friction state using different lubricants. Under boundary lubricant, texture mainly plays the role of oil storage. For lubricants with higher viscosity, texture can shorten the running-in period, perform secondary lubrication, and improve the load bearing capacity of the lubricating film. Laser surface processing reduces the surface energy, which to some extent can improve the tribological performance with boundary and mixed lubrication.

2. Among the three oils, castor oil displays the lowest coefficient of friction and wear, while white oil has the highest coefficient of friction and wear; the friction coefficient of silicone oil presents a rising trend as the load increases.

3. Lubricants with higher viscosity could alleviate the boundary contacts by generating a thicker film to separate the tribological surfaces, especially under the textured surface. When the lubricant's viscosity is larger, the effect of the textured surfaces to increase the thickness of the oil film is more pronounced.

4. Compared with white oil and silicone oil, the chemical properties of castor oil have a significant effect on the tribological performance when the load increases. Castor oil presents the best tribological performance under mixed lubrication, which is dominant because it displays the highest viscosity and exhibits polar functional groups. During friction of mating surfaces, castor oil hydrolyzes to produce more polar molecules. Polar castor oil molecules could be adsorbed on the lubricating surfaces for the existing asperity contacts to improve the boundary lubrication state.

The authors would like to thank the financial support from the National Natural Science Foundation of China (Nos. 51975454 and 52075418).

There are no conflicts of interest.

The authors attest that all data for this study are included in the paper.