Research Progress on the Design of Surface Texture in Tribological Applications: A Mini-Review

Abstract

Surface texturing technology, as an advanced method to improve surface tribological properties of friction pairs, has been widely used in many fields. In this work, the influence of surface texture parameters on tribological properties of friction pair surfaces are reviewed. For the currently most developed surface textures with symmetry and simple geometries and distributions, it is found that they could help reduce friction mainly by enhancing their dynamic pressure lubrication capability, storing abrasive debris and lubricants for dynamic lubrication or promoting the formation of friction films on surfaces of friction pairs. The dominant design parameters of surface textures influencing their tribological performance are found to be shape, geometry and density, while working condition, including contact mode and lubrication situation, also has a significant influence on the performance of surface textures with specific features. Asymmetric textures and multi-scale composite textures also show great tribological performance, while the coupling mechanism across different factors is still unclear, which makes it a challenge to maximize the advantage of asymmetric or multi-scale composite textures. The development of machine learning provides promising approaches for the multi-parameter optimization of surface textures, which is expected to promote and accelerate the design of advanced surface textures.

1. Introduction

Friction and wear are common phenomena in production and daily life, and scientists have been making efforts to minimize the cost induced by friction and wear for a long time [1,2,3,4,5]. In modern industry, around 60–80%of the occurrence failure of mechanical components are caused by wear, and the energy consumed by friction globally accounts for nearly 1/4 of the total energy used [6,7]. Apart from developing wear-resistant materials, lubricating media and surface treatment technology [8,9,10], controlling surface structure is also an effective way to control the friction and wear of materials. It has been observed that non-smooth surfaces do not always lead to more friction and wear [11], and this phenomenon has attracted great attention [12,13,14,15,16,17]. Experimental results show that surfaces with specific textures could result in less friction and wear compared to untextured surfaces [18,19,20]. Although the anti-friction mechanisms of surface textures are still lacking a unified theory, the effect of surface textures in reducing friction and wear has been widely accepted, and textures on the surface of friction pairs have been used more and more in tribological applications [20,21,22,23,24].

Surface textures are specific structures on the surface of friction pairs with a specific distribution arrangement [25,26,27]. Surface textures have been applied in many fields such as cutting tools and human joints for the purpose of friction reduction [28,29,30]. Surface textures exhibit varying anti-friction properties under different tribological conditions, so establishing a correlation between surface texture parameters and their performance across different working conditions is crucial for advancing the design of sophisticated surface textures. Generally speaking, factors such as shape, geometric shape, distribution and density are significant in determining the performance of surface textures, and most of the current research on surface texture design focuses on the optimization these features.

Here, we reviewed the research progress on the design of surface texture to look into the influence of surface texture design parameters on the tribological properties of surfaces. We discussed the effects of the design parameters of surface textures on the tribological properties of friction pair surfaces and illustrated friction reduction mechanisms for different types of surface textures under different working conditions. The advantages of machine learning in dealing with complex parameter optimization are also introduced, which provides a new and effective approach for the design of surface textures. This work is expected to provide references for a more effective optimization of design parameters in surface textures.

2. The Effect of Symmetric Texture Parameters on Tribological Properties

The geometric dimensions of surface texture have been studied from micro to macro levels to suit different working conditions [31,32,33]. According to the relative height between surface texture and friction pair surface, surface texture could be classified into two types: the convex texture (Figure 1a) and the pit texture (Figure 1b). Among the two types of textures, the pit texture attracts more attention and is currently more commonly used, which is mainly because of its better manufacturability. The pit-textured structures could be obtained by different mechanical, physical or chemical approaches, which are more mature than the additive manufacturing approach used to fabricate convex-textured structures. The pit texture has been proven to have good performance in anti-friction and anti-wear properties, load-bearing capacity and self-cleaning properties, and it could meet the requests of multiple applications with a relatively simple optimization of parameters [10,11]. Therefore, this work mainly focuses on the design and development of the pit texture.

2.1. Surface Texture Contour Shape

Significant advances in computer graphics, coupled with a deeper understanding of the microstructure of biological body surfaces, have greatly facilitated the development of surface texture design techniques. These technological advances have not only simplified the design process but also improved the accuracy of the design results [34]. The excellent surface properties of biological body surfaces and the working mechanisms behind them have been an important source of inspiration for surface texture design. However, not all of these naturally generated surface structures can be directly applied to engineering practice. Therefore, performing graphical optimization becomes a crucial step in translating these bio-inspired structures into practical designs [35]. Zhang et al. [36] used a genetic algorithm to simulate the biological evolution process in order to investigate the effect of surface texture contour shape on tribological performance for shape contour optimization. After 28 iterations of the genetic algorithm, two types of textures, fish-shaped and bullet-shaped, were finally obtained. In order to verify the tribological properties of optimized texture shapes, three shapes of textures were prepared on GCr15 discs (Figure 2), in which all design parameters were kept the same apart from the shape. Unidirectional sliding experiments were carried out on the UMT-2 friction and wear tester; the average friction coefficient is shown in Figure 2d. It can be observed that the fish-shaped texture has the lowest friction coefficient. However, the longer boundary perimeter was found to result in more severe stress concentration, so the fish-shaped texture would fail earlier during relative friction. The experimental results show that the bullet-shaped textures have lower friction coefficients under all operating conditions compared to the circular textures, indicating that the bullet shape is more suitable for textures under the unidirectional sliding condition than the circular shape. Therefore, the optimal contour shape for all considerations was found to be the bullet shape.

Guo et al. [37] found that the texture shape of the material surface has a significant impact on the lubrication effect. They studied the effects of three types of convex textures on the surface of ultra-high-molecular-weight polyethylene materials under water lubrication. The three textures are cylindrical, cubic and rectangular. Due to the limitations of convex texture processing methods, the texture size is in the millimeter range. Friction and wear experiments were designed based on the service conditions of the materials. The average friction coefficient is shown in Figure 3. The coefficient of friction at a load of 0.3 MPa is obviously larger than the other cases, which is determined by the characteristics of the ultra-high-molecular-weight polyethylene materials. The textured surfaces have fewer scratches and material transfer compared to non-textured surfaces, and the cubic texture shows the best wear performance. Khan et al. [38] prepared three types of textures, namely straight grooves, curved grooves and circular pits, on the surface of tungsten carbide cutting tools using lasers. Compared to non-textured surfaces, the wear of these textured cutting tools is significantly reduced, which is attributed to the presence of textured structures effectively reducing the friction coefficient and controlling the generation of heat. It should be noted that although the wear resistance of the circular pit texture is the worst among the three textures, as is reflected in the service life of the tool, it still has a longer service life than tools without textured surfaces. This indicates that even textures with relatively low wear resistance can improve the durability of cutting tools to a certain extent, providing important references for tool design and material optimization.

The improvement of surface tribological properties by lubrication is significant, and different shapes of textures exhibit different friction reduction mechanisms under lubricated conditions. Morris et al. [39] combined numerical simulations and experimental tests to ascertain the positive effects of pressure perturbations induced by microfluidic mechanisms on the entrapment of lubricant in cavities of textures, and the improvement in micro-wedge flow dynamics was also estimated. Reciprocal friction tests on a V-shaped textured surface showed that the reciprocating sliding lubricant was entrained into the texture pits for storage, leading to varying degrees of effectiveness of the micro-wedge effect. The reduction in friction coefficient was most pronounced when the angle of the V-shaped texture was 80° and the depth was 3 μm. Higher texture density can store more lubricant, but the texture density should be designed carefully considering the air leakage problem when the piston ring is in contact with the cylinder liner. Wang et al. [40] ascribed the friction reduction mechanism of pit textures under oil lubrication to the variable cavitation effects induced by different shapes of pits. The friction coefficients of the three types of textures studied under lubrication conditions are shown in Figure 4. The square texture shows the lowest friction coefficient due to its larger compression area and superior dynamic pressure-bearing capacity. The dynamic lubrication effect provided by the texture has been confirmed in both simulation and experimental analysis and is currently one of the accepted texture friction reduction mechanisms.

In addition to the texture shape, there is a correlation between the dynamic pressure-bearing capacity and the depth and occupancy of pits. In the study by Nie et al. [41], a detailed simulation analysis was conducted on the relationship between surface texture parameters and bearing capacity; the results obtained are shown in Figure 5. H is the ratio of oil film thickness to texture depth, and W is the ratio of texture width to texture unit width. When W < 0.5, the load-bearing capacities of textured surfaces are lower than those of the non-textured surface, while when W > 0.5, the load-bearing capacities of textured surfaces is significantly higher than those of the non-textured surface. In the study by Qiu et al. [42], the geometric shapes of six commonly used textures were optimized, and it was found that contours with curves had a greater advantage in reducing the friction coefficient than straight edge contours, which was almost opposite to the results obtained by Wang et al. [40]. The main reason is that the lubrication condition of Qiu et al. [42] was gas lubrication and that only simulation analysis was performed, while the experimental results of Wang et al. [40] were obtained in oil lubrication. Babu et al. [43] combined simulation and experiment to study the frictional performance of several shape textures under boundary lubrication conditions. From the experimental results in Figure 6, it can be found that among different texture shapes, the elliptical texture shows the lowest friction coefficient at different slip velocities, whereas the circular texture shows the highest friction coefficient.

From the above discussion, one may see that the texture shape has a significant effect on the reduction of surface friction coefficient. Based on the analysis on friction coefficient and wear rate, the performance of different texture shapes under the same working conditions can be effectively compared, and the optimal texture design can be determined. It is worth noting that with changes in lubrication conditions, the same texture shape may exhibit vastly different performance. This is mainly due to differences in friction reduction mechanisms. Under dry friction conditions, surface texture achieves anti-friction ability mainly by reducing contact area and storing debris, while under lubricated conditions, the primary anti-friction mechanism becomes the enhancement of load-bearing capacity. Therefore, different lubrication states may lead to changes in texture properties. Specifically, a circular texture may not be comparable to other textures in terms of friction reduction effect, but its design and processing are relatively simple, and it can still provide a certain degree of friction reduction compared to surfaces without a texture [10]. This allows circular textures to still play a role in practical applications, especially in situations where high processing requirements are needed.

2.2. Geometric Parameters

The optimization of parameters is a crucial step in the design of products. For surface textures with specific shapes, the active optimization of geometric parameters is essential to improving their tribological properties [11,14,25]. Huang et al. [44] were inspired by the surface microstructure of tree frogs and used the response method to optimize the geometric size of the texture. A biomimetic hexagonal texture design was carried out on the surface of AISI 4140 steel. The average friction coefficient was reduced by 20% after the optimization; this is ascribed to the uniform lubricating film generated on the optimized textured surface, which can help decrease the direct contact between friction pairs. Zhang et al. [45] experimentally verified that the friction coefficient of groove-textured surfaces is lower than that of non-textured surfaces. In the state of oil lubrication, the wear resistance of textured surfaces with different parameters also shows differences. The effect of groove width and depth on wear rate is shown in Figure 7. The results show that the wear rate of textured surfaces with larger depth and width values is lower. The SEM characterization indicates that the observed phenomenon is attributed to the presence of abrasives in the operational environment. The grooves with increased width and depth can capture more abrasives, which reduces residual abrasives on the surface and turns the wear type from abrasive wear to mild abrasive wear and scratches. Patel et al. [46] also found that deeper and wider textures could significantly decrease the friction coefficient.

Xu et al. [47] studied the effects of full oil lubrication and oil-deficient lubrication on the formation of lubricating oil film and the wear performance of elliptical textures. The presence of a friction film is key to reducing surface friction wear. Under the same lubrication conditions, textured surfaces have thicker lubrication films. XPS analysis shows that carbon, oxygen and sulfur are the primary elements composing the friction film. With sufficient lubrication, the contents of carbon and oxygen in the film are more than with starved lubrication. If the ratio of the long axis to the short axis of an ellipse is too high, it will increase the contact stress and damage the friction film, forming wear debris. M. Varman et al. [48] studied the effect of different diameters and depths of textures under lubrication conditions on the tribological properties of DLC (diamond-like carbon) coatings. Figure 8 shows the influence of texture diameter and depth on the wear volume. It is found that the pit texture with a depth to diameter ratio of 0.06 has the lowest wear loss. Friction reduction on textured surfaces is not always better than on untextured surfaces, especially if the texture is not well-designed. The depth and diameter of the texture largely determine the lubricant storage capacity. Well-designed texture pits can effectively store lubricant and thus reduce friction. However, if the pit is too deep or too large, it may cause the oil film to rupture and increase friction. However, shallow or small pits may not be able to efficiently store lubricants, leading to an impact on the lubrication effect.

The design process of surface texturing is complex and involves numerous influencing factors, making it extremely challenging to reveal the relationship between texture parameters and surface functionality solely through simple statistical analysis. An adequate understanding and effective control of related influencing factors are the key to achieving optimal textured surfaces. During the design stage, it is essential to take different factors into consideration, including properties of materials and morphological features of textures such as shape, size and distribution density. Zeng et al. [49] proposed a method in the article that utilizes 3D surface texture parameters and statistical functions to establish a connection between surface geometric features and surface functionality, which is more accurate and effective than 2D parameters. The research results also reveal that establishing the relationship between the geometric shape of surface texture and its function cannot rely solely on a single or a few parameters. Given the numerous parameters involved and the complexity of experimental conditions, relying solely on human labor for statistics and analysis is impractical. In the context of the big data era, with the increasing improvement of computer technology and artificial intelligence algorithms, the use of machine learning algorithms to predict the relationship between texture parameters and surface properties has shown promising prospects. Machine learning can effectively establish potential connections in complex data, build multidimensional models and enable researchers to optimize texture design with higher accuracy and efficiency. This method has been applied in material design and surface texture design [50,51,52]. Li et al. [53] proposed a method based on machine learning for designing texture parameters, which trains and evaluates a large number of collected experimental parameters to reverse predict the relationship between texture parameters and performance functions. They found that the coefficient of friction is mainly affected by the texture geometry and the predictions from machine learning algorithms have a reliability of more than 90%, which effectively accelerates the design and optimization of surface texture parameters.

2.3. Texture Density

The surface texture could lower the friction coefficient by reducing the contact area between friction pairs. Both convex and pit textures can effectively reduce the contact area and thus lower the friction coefficient. Therefore, a reasonable design of the proportion of the textured area on the whole surface is essential to achieve the desired tribological performance. Due to the generally small size of surface texture, higher texture density can increase processing difficulty and have adverse effects on surface quality. Therefore, it is necessary to optimize the texture density for different application surfaces.

Li et al. [20] prepared a V-shaped texture on the surface of hard alloys to achieve the purpose of reducing friction. The experimental environment was lubricated with lubricating oil, and the friction direction was consistent with the V-shaped convergence direction. Figure 9 shows the friction coefficients for different texture densities. The results shows that the texture density of 9.5% has the lowest and most stable friction coefficient under different loads. If the density is too high, it will actually increase the friction coefficient, while if the density is too low, it will not provide sufficient bearing capacity. Additionally, for lubrication conditions, Zhang [45] et al. concluded in their study that a larger texture density significantly enhances the wear resistance of the surface, and this conclusion is based on the fact that a larger texture density can capture the abrasive material present between the surfaces to a greater extent and minimize the three-body abrasive wear. Nui et al. [54] found in their study that increasing the density and diameter of pits appropriately is beneficial for reducing friction and wear when the texture depth is small. With the same number of textured units, increasing the diameter of the pits results in an increase in the density of the texture, so the increase in both density and diameter can enhance the ability of textures to capture abrasive particles, thereby reducing the wear rate.

Yuan et al. [55] prepared pit textures on the surfaces of four metal materials and used variance analysis to comprehensively analyze and determine the optimal texture density for each material. However, the optimal texture density for different materials is different. Therefore, optimizing texture density cannot completely rely on literature references or simulations. Instead, for each material, texture parameters must be optimized through a systematic design of experiments and statistical analyses to ensure optimal tribological properties. The surface texture not only stores abrasive debris to reduce three-body abrasive wear but also stores some tiny solid lubricants, which diffuse out from the inside of the texture pits to lubricate the friction partner when it moves relative to each other. The surface texture not only captures debris and prevents abrasive wear but also serves as a reservoir for small solid lubricants. These lubricants gradually diffuse from the texture to the surface, providing lubrication during the relative motion of the friction pair. The study by Zhang et al. [56] confirms this phenomenon, but it requires the surface texture density to be within a reasonable range. From Figure 10, it can be concluded that the wear rate is lowest when the surface texture density is 36% and increases when it is greater than or less than 36%. The highest wear rate occurs on the non-textured surface. Electron probe microanalysis and EDS analysis results indicate that the presence of surface texture results in the diffusion of solid lubricants to the worn surface, thereby reducing direct contact between friction pairs. When the texture density is 36%, the solid lubricant has the best synergistic effect with the surface texture and exhibits the best tribological performance.

The lubricant plays a crucial role in reducing friction and wear between friction pairs. In addition to conventional lubricants and greases, water-based lubricants have been found to be of practical value in specific application scenarios. In particular, there are industrial processes where the operating environment of the friction pair is inherently wet, such as in underwater operations, hydraulic machinery and marine engineering. Zou et al. [57] confirmed under water lubrication conditions that there is a positive correlation between texture-bearing capacity and wear resistance, where texture-bearing capacity is also related to texture density. The best bearing capacity is achieved at a texture density of 2%, and the wear volume is also the smallest, as shown in Figure 11. Li et al. [58] also pointed out the considerable significance of texture density on anti-wear performance. Their results show that modifying the texture density can obviously enhance the tribological properties of the textured surface. The key factors contributing to improved anti-wear performance in friction pairs are the enhancement of dynamic pressure lubrication capabilities and the formation of friction film.

2.4. Other Factors

Apart from shape, dimensions and density, there are many other factors influencing the design of surface textures. For example, the distribution arrangement was observed to largely affect the performance of surface textures. However, there is a lack of systematic research on the arrangement of surface textures, and it is unclear how the texture arrangement influences wear behaviors. Shen et al. [59] studied the differences in the arrangement of V-shaped textures on the improvement of bearing capacity. Compared to V-shaped textures with the parallel arrangement shown in Figure 12a, the V-shaped arrangement shown in Figure 12b shows improved lubrication performance and load-bearing capacity, which is ascribed to the more effective convergence of the lubricating oil film with the V-shaped arrangement [11]. Li et al. [60] observed significant influences of texture shape and lubricant flow direction on the fluid dynamic pressure effects. The reverse-arranged gourd-shaped texture and the circular pit texture are found to have close friction coefficients, while the forward-arranged gourd-shaped texture has a significantly reduced friction coefficient. Therefore, the convergence of the texture in the direction of lubricant flow is considered to be an important factor affecting the hydrodynamic pressure effect of textures. In some studies [10,11,43,54], it is believed that the ratio of texture depth to diameter has an impact on the surface of friction pairs, and there is a certain correlation between the friction coefficient and the ratio of texture depth to diameter. The ratio of depth to diameter affects the secondary lubrication ability of lubricants [61,62].

The wear products would affect the tribological interface and the subsequent frictional processes. In some cases, abrasive debris generated during tribological processes would result in three-body abrasive wear and exacerbate material loss. In contrast, some materials may form friction films during tribological processes, and in some cases the wear products could also act as a lubricant to reduce the coefficient of friction and reduce wear losses. Kong et al. [63] prepared hetero-structured NiCr-WS₂-Ti self-lubricating composites, and the sulfide products after the interfacial friction and oxidation reaction acted as a solid lubricant, which facilitated the formation of a protective layer on the sliding surfaces. At elevated temperatures, titanium would be firstly oxidized due to its high activity, which promotes the spontaneous formation of tiny textures on the surface of material, the textures can store the self-lubricating material debris, and the oxides could serve as a protective glaze layer on the contact surface. Srivyas et al. [64] actively designed the texture parameters and prepared the textured surface using lasers. The synergistic lubrication between the textured surface and lubricant was investigated, and the results showed that the coefficient of friction and wear of the textured surface were lower than those of the normal surface under all the design conditions, which was attributed to the synergistic action between the textured surface and the lubricant in the formation of a self-lubricating layer, which reduces the coefficient of friction and the wear loss.

When the friction pair surface has hydrophobicity, it can effectively reduce the liquid in contact with the surface, reduce the erosion of liquid molecules on the surface and reduce adhesive wear and corrosion wear [65,66]. Shum et al. [67] prepared textures with different geometric parameters on the surface using laser surface texturing technology, and they then deposited DLC coatings on the surface by magnetron sputtering to form a textured DLC surface. The surface contact angle of the textured DLC coating was observed to be greater than 130°. The wear experiment was conducted using a reciprocating ball plane wear tester in a simulated body fluid environment. When the texture density was 10%, the distance between adjacent texture units was most conducive to reducing wear, which was related to the surface hydrophobicity and anti-adhesion caused by the texture units. Wang et al. [68] found a correlation between the change in contact angle caused by surface texture and the friction coefficient. Hydrophobic textures were prepared on the surface of YG8 hard alloy cutting tools using a femtosecond laser. Contact angle measurements showed a 123% increase in specific surface contact angle compared to ordinary surfaces. Friction and wear experiments showed a 61.154% decrease in surface friction coefficient compared to ordinary surfaces. SEM and EDS results showed that the micro-textures can significantly reduce adhesive wear, promote the separation of abrasive particles and prevent the aggravation of wear.

2.5. Summary

Surface texture plays an important role in controlling friction and wear by reducing friction coefficients and wear loss during relative motion, which greatly benefits the surface life and reliability of components. The reduction of the friction coefficient improves the energy utilization efficiency of mechanical systems, reducing energy waste and costs, and the decrease in wear loss increases the surface life and reduces maintenance costs. Although it is still hard to give a quantitative techno-economic evaluation for the surface texture technology at the current stage, the development of manufacturing technology and continuously improving performance makes the application of texture technology promising in terms of both efficiency and economy. The main mechanisms contributing to the anti-friction properties of surface textures are found to be enhancing the dynamic pressure lubrication capability, storing abrasive debris and lubricants for dynamic lubrication and promoting the formation of friction films [45,48,54]. At present, the surface texture is more common in circular, oval, square and other symmetrical texture shapes, and there are also a variety of shape composite surface textures [36,43,47]. Compared with non-textured surfaces, traditional types of textured surfaces typically exhibit superior wear resistance [47]. However, in specific application scenarios, specially designed irregular textures can provide better wear resistance due to their unique geometric features (Table 1). Currently, compared to other preparation techniques, laser processing technology is the most commonly used method for preparing surface textures due to its flexibility and high accuracy. By adjusting the frequency, power and scanning strategy of the laser, the physical features of textures can be precisely controlled, and with the advancement of technology, the cost of laser-textured surfaces is expected to be lowered. However, the geometric dimensions of surface texture are still limited to the micrometer level. The surface occupancy rate of the texture in current research typically ranges from 2% to 20%, and in some special cases the occupancy rate may come up to 30–40% [47]. At present, there is a lot of research on textured surfaces of hard alloy materials, titanium alloys, stainless steel and other materials. This is because these materials are widely used in wear resistance and corrosion resistance, and textured surfaces further improve the frictional properties of wear-resistant material surfaces [20,38,44,55]. The formation of surface lubricating film is related to the surface material and lubricant. Textured surfaces and lubricated surfaces are more likely to form stable lubricating films in the later stage of friction compared to non-textured surfaces. Furthermore, a reasonable texture unit design can dynamically release lubricants, store abrasive particles and ensure the continuous existence of the lubricating film. The main factors influencing lubricating films include the type of lubrication, the self-lubricating properties of the surface material and the dynamic pressure lubrication capability of the texture unit [39,44,47,63,64]. The anti-friction effect of surface texture involves numerous complex influencing factors (Table 1), which require a comprehensive consideration of the interaction of these factors in the design stage. In order to achieve optimal tribological performance, a systematic approach must be adopted to comprehensively analyze and evaluate the effects of various parameters. Machine learning, as an advanced data analysis tool, provides an effective means to handle and optimize multi-parameter problems in surface texture design. By training algorithms to identify complex patterns and trends, machine learning can predict the contributions of different texture parameters to tribological properties and guide rational design optimizations. It is worth noting that when designing the texture shape, the actual working conditions in the application should be taken into account. Radial bearings usually use surface textures with circular pits. However, due to the problem of combustion pressure leakage, circular textures are not suitable for piston rings.

3. The Effect of Asymmetric Texture and Multi-Scale Composite Texture on Tribological Performance

3.1. Asymmetric Texture

Asymmetric design plays an important role in nature, science and engineering. Asymmetry is an important concept in the field of topology in mathematics. In material design, asymmetry can be used to enhance the mechanical properties of materials, such as improving material strength and toughness through asymmetric microstructures [69,70]. Asymmetric design also exists in the design of surface textures to achieve the improvement of surface properties. Shen et al. [71] proposed a numerical method for optimizing surface texture shape based on the SQP algorithm, which eliminates the limitation on the occupancy rate of texture area in the optimization algorithm and can maximize the finding of the optimal texture shape. After optimization, it is found that the optimal texture geometry is similar to the V-shape texture but with a flat front end and an asymmetric structure, as shown in Figure 13. Such an asymmetric texture has much better load-bearing capacity compared to the texture with a symmetric V-shape, which is consistent with the results from the experimental and theoretical work by Li et al. [72]. A similar concept is proposed by Han et al. [73], in which ultimate bearing capacity is used as an effective basis to evaluate the lubrication characteristics of textured surfaces. The research results indicate that asymmetric micro pits can achieve greater load-bearing capacity than symmetric rectangular micro pit textures, which will help improve surface lubrication performance. Asymmetric micro pit structures may become a highly promising candidate shape for future texture design.

The effect of asymmetric texture on surface friction under dry friction conditions was studied by Tewelde et al. [74], in which the groove has an asymmetric cross section as shown in Figure 14. The results show that the sliding direction has a significant influence on the performance of asymmetric textures. The friction coefficient under clockwise rotation is 28.1% higher compared to that under counterclockwise rotation. This is related to the variation in the sharpness of the asymmetric texture edges under different relative motions. Jiang et al. [75] proposed a numerical method based on a genetic algorithm to explore the optimal type of texture under reciprocating sliding conditions. The results showed that the asymmetric texture optimized by numerical methods has the ability to improve overall performance and achieve directional friction control. It also indicates that the design of texture types should be based on actual working conditions, as the optimal texture is not necessarily symmetrical or asymmetric. Schuh et al. [76] successfully achieved the preparation of asymmetric texture on the friction pair surface through end milling and experimentally verified the effect of asymmetric texture surface on reducing the friction coefficient. It shows that there exists an optimal value of β for reducing the friction coefficient of the asymmetric texture, the optimal value of β depends on the height of the operating gap and the optimal value in the study was obtained from the analysis of the friction coefficient, as shown in Figure 15.

The asymmetric textures have shown unique advantages and great potential in directional sliding and enhanced dynamic lubrication compared to symmetric textures in specific cases [71,74]. However, the research on asymmetric surface textures is relatively limited. Through a statistical analysis of the relevant experimental literature, as shown in Table 1, it can be found that there are relatively few studies on asymmetric textures, which may be related to the limitations of their preparation methods. The preparation methods of asymmetric textures in Table 1 mainly rely on mechanical processing, and this method has a limited selection of design parameters. Moreover, the values of surface texture design parameters are at the mesoscale, which increases the difficulty of preparing asymmetric textured surfaces. In contrast, symmetric textures are more easily obtained owing to relatively mature preparation processes and simple geometries. Therefore, in order to promote the development of asymmetric surface texturing technology, future research focusing on innovative preparation methods are desired, which is expected to overcome the challenge in fabrication and minimize the potential benefits of asymmetric textures in tribological applications.

Table 1. Brief summary of experimental data from some references (d (depth), dia (diameter), h (heigh), w (width), CA (Crosshatch angle), β (β is the angle that the material was tilted to make the texture), s (spacing), r (ratio of major axis to minor axis of ellipse)).

	Profile Shape	Geometrical Parameter	Texture Density% or Spacing	Sliding Type	Lubrication Type	Tribological Performance	Friction Reduction Mechanism	Ref.
Laser	Chevron	d = 32 μm	5.05, 9.5, 13.02, 15.2	Unidirectional sliding	Liquid lubricant	The surface with 9.5% texture density shows the lowest friction coefficient.	Increase fluid dynamic pressure	[20]
Laser	Linear groove	d = 20–100 μm w = 20–50 μm	32	Reciprocating sliding	Dry sliding	Compared to non-textured surfaces, the friction coefficient is reduced by 32%, the wear rate is reduced by over 84%.	Form friction films	[35]
Laser	Circular, bullet, fish	d = 15 μm	15.2	Unidirectional sliding	Oil	Compared to circular texture, the friction coefficients of fish shaped and bullet shaped materials are reduced by 21.1% and 30.1%, respectively.	Increase fluid dynamic pressure	[36]
Turning	Cylindrical, cube, cuboid	d = 1.0 mm	/	Unidirectional sliding	Water	Compared to non-textured surfaces, the decrease in friction coefficient is not significant, and the wear loss is reduced by 55%, 63% and 58% for cylindrical, cube and cuboid textures, respectively.	Store debris	[37]
Laser	Chevron	d = 3 μm	9.1 11.5	Reciprocating sliding	Oil	The surface with texture density of 11.5% has the lowest friction coefficient.	Increase the thickness of the lubricating film	[39]
Laser	Square, Circular, triangle	d = 90 μm	8	Unidirectional sliding	Oil	Compared to non-textured surfaces, the friction coefficients are reduced by 17%, 13% and 6% for square, circular and triangle textures, respectively.	Increase fluid dynamic pressure	[40]
Chemical Etching	Square	h = 10, 40, 70, 100 μm	10–50	Unidirectional sliding	Oil	Compared to non-textured surfaces, the lower the height and texture density, the more significant the decrease in friction coefficient: The maximum friction reduction rate of a surface with a height of 10 μm and a texture density of 10% is 80.6%.	Increase fluid dynamic pressure	[43]
Laser	Hexagonal (bionic)	d = 490 μm	s = 360 µm	Reciprocating sliding	Solid lubricant	Compared to non-textured surfaces, the friction coefficient is decreased by 20.82%, and the wear loss is decreased by 65.65%.	Form friction films	[44]
Milling	Linear groove	d = 400 μm w = 1.5 mm	34	Unidirectional sliding	Lubricating oil contains sands	The wear resistance is 4.9 times that of non-textured surfaces.	Store silica sand	[45]
photoetching	Ellipse	r = 1.5 (ST), 2 (MT), 3 (LT)	s = 2 mm	Reciprocating sliding	Starved oil lubrication	Compared to the non-textured surface, the wear loss is decreased by 10.1% and 23.4% for ST and MT surfaces and increased by 35.5 for LT surface.	Form friction films	[47]
Laser	Dimple	d = 6–30 μm dia = 50–300 μm	20	Reciprocating sliding	Oil	The samples with texture diameters of 100 μm and depths of 6 μm show enhanced tribological properties.	Store lubricants and remove wear particles	[48]
Laser	Dimple	d = 10 μm	5, 10, 15	Unidirectional sliding	Starved oil lubrication	Compared to the non-textured surface, the friction coefficient is decreased by 51%, 55% and 48% for texture densities of 5%, 10% and 15%, respectively.	Store debris and form lubricating films	[54]
EDM wire cutting	Rectangular Micro-grooves	d = 500 μm w = 300 μm	24, 30, 36, 42	Reciprocating sliding	Grease	Compared to the non-textured surface, the wear rate is the most reduced by 38.6% with the texture density of 36%.	Store lubricating grease	[56]
Laser	Dimple	dia = 15 μm	0.5, 1, 2, 4, 6	Unidirectional sliding	Deionized water	Compared to the non-textured surface, the wear loss is decreased by 21%, 56%, 73%, 66% and 39%, for texture densities of 0.5%, 1%, 2%, 4% and 6%, respectively.	Increase fluid dynamic pressure	[57]
Laser	Dimple	dia = 90 μm d = 10 μm	2, 4, 6, 8, 10	Unidirectional sliding	Graphene/5CB suspension lubrication	The surface with texture density of 8% shows the lowest friction coefficient.	Store lubricating oil, reduce contact area, secondary lubrication	[58]
photoetching	Micro-grooved crosshatch	CA = 20°–60° W = 40 μm	20	Unidirectional sliding	Paraffin oil	The friction coefficient first increases and then decreases, with the highest value at 30°, which is about 6 times higher than that at 60°.	Promote fluid flow	[62]
Laser	Square	d = 10 μm w = 50 μm	s = 100 µm	Reciprocating sliding	Dry sliding, lubrication (virgin PAO-4, SAE20W50)	Compared to the non-textured surface, the friction coefficient of the textured surface is decreased by 63.88%, 70.53% and 44.99%, and the wear loss is decreased by 32.13%, 60.30% and 86.03% under conditions of dry sliding, virgin PAO-4 lubrication and SAE20W50 lubrication, respectively.	Form transfer layers	[64]
Turning	Asymmetric depth of groove	d = 4 μm	20, 30, 40, 70	Opposite bidirectional rotational sliding	Dry sliding	The sharpness of the groove outlet edge increases when rotated clockwise, resulting in a higher friction coefficient. The increase in friction coefficient is 61.3% and 28.1% for texture densities of 20% and 70%, respectively.	Reduce contact area	[74]
Milling	Asymmetric-depth-profile textures	β = 5.3°, 9.4°, 14°, 21.7°	s = 8.953 mm	Bidirectional relative motion	Newtonian fluids	The asymmetric textures result in lower friction coefficient than symmetric textures. The asymmetric texture with a base angle β = 5.3 ° shows the lowest effective friction coefficient.	Reduce shear stress and generate normal load	[76]

Table 1. Brief summary of experimental data from some references (d (depth), dia (diameter), h (heigh), w (width), CA (Crosshatch angle), β (β is the angle that the material was tilted to make the texture), s (spacing), r (ratio of major axis to minor axis of ellipse)).

	Profile Shape	Geometrical Parameter	Texture Density% or Spacing	Sliding Type	Lubrication Type	Tribological Performance	Friction Reduction Mechanism	Ref.
Laser	Chevron	d = 32 μm	5.05, 9.5, 13.02, 15.2	Unidirectional sliding	Liquid lubricant	The surface with 9.5% texture density shows the lowest friction coefficient.	Increase fluid dynamic pressure	[20]
Laser	Linear groove	d = 20–100 μm w = 20–50 μm	32	Reciprocating sliding	Dry sliding	Compared to non-textured surfaces, the friction coefficient is reduced by 32%, the wear rate is reduced by over 84%.	Form friction films	[35]
Laser	Circular, bullet, fish	d = 15 μm	15.2	Unidirectional sliding	Oil	Compared to circular texture, the friction coefficients of fish shaped and bullet shaped materials are reduced by 21.1% and 30.1%, respectively.	Increase fluid dynamic pressure	[36]
Turning	Cylindrical, cube, cuboid	d = 1.0 mm	/	Unidirectional sliding	Water	Compared to non-textured surfaces, the decrease in friction coefficient is not significant, and the wear loss is reduced by 55%, 63% and 58% for cylindrical, cube and cuboid textures, respectively.	Store debris	[37]
Laser	Chevron	d = 3 μm	9.1 11.5	Reciprocating sliding	Oil	The surface with texture density of 11.5% has the lowest friction coefficient.	Increase the thickness of the lubricating film	[39]
Laser	Square, Circular, triangle	d = 90 μm	8	Unidirectional sliding	Oil	Compared to non-textured surfaces, the friction coefficients are reduced by 17%, 13% and 6% for square, circular and triangle textures, respectively.	Increase fluid dynamic pressure	[40]
Chemical Etching	Square	h = 10, 40, 70, 100 μm	10–50	Unidirectional sliding	Oil	Compared to non-textured surfaces, the lower the height and texture density, the more significant the decrease in friction coefficient: The maximum friction reduction rate of a surface with a height of 10 μm and a texture density of 10% is 80.6%.	Increase fluid dynamic pressure	[43]
Laser	Hexagonal (bionic)	d = 490 μm	s = 360 µm	Reciprocating sliding	Solid lubricant	Compared to non-textured surfaces, the friction coefficient is decreased by 20.82%, and the wear loss is decreased by 65.65%.	Form friction films	[44]
Milling	Linear groove	d = 400 μm w = 1.5 mm	34	Unidirectional sliding	Lubricating oil contains sands	The wear resistance is 4.9 times that of non-textured surfaces.	Store silica sand	[45]
photoetching	Ellipse	r = 1.5 (ST), 2 (MT), 3 (LT)	s = 2 mm	Reciprocating sliding	Starved oil lubrication	Compared to the non-textured surface, the wear loss is decreased by 10.1% and 23.4% for ST and MT surfaces and increased by 35.5 for LT surface.	Form friction films	[47]
Laser	Dimple	d = 6–30 μm dia = 50–300 μm	20	Reciprocating sliding	Oil	The samples with texture diameters of 100 μm and depths of 6 μm show enhanced tribological properties.	Store lubricants and remove wear particles	[48]
Laser	Dimple	d = 10 μm	5, 10, 15	Unidirectional sliding	Starved oil lubrication	Compared to the non-textured surface, the friction coefficient is decreased by 51%, 55% and 48% for texture densities of 5%, 10% and 15%, respectively.	Store debris and form lubricating films	[54]
EDM wire cutting	Rectangular Micro-grooves	d = 500 μm w = 300 μm	24, 30, 36, 42	Reciprocating sliding	Grease	Compared to the non-textured surface, the wear rate is the most reduced by 38.6% with the texture density of 36%.	Store lubricating grease	[56]
Laser	Dimple	dia = 15 μm	0.5, 1, 2, 4, 6	Unidirectional sliding	Deionized water	Compared to the non-textured surface, the wear loss is decreased by 21%, 56%, 73%, 66% and 39%, for texture densities of 0.5%, 1%, 2%, 4% and 6%, respectively.	Increase fluid dynamic pressure	[57]
Laser	Dimple	dia = 90 μm d = 10 μm	2, 4, 6, 8, 10	Unidirectional sliding	Graphene/5CB suspension lubrication	The surface with texture density of 8% shows the lowest friction coefficient.	Store lubricating oil, reduce contact area, secondary lubrication	[58]
photoetching	Micro-grooved crosshatch	CA = 20°–60° W = 40 μm	20	Unidirectional sliding	Paraffin oil	The friction coefficient first increases and then decreases, with the highest value at 30°, which is about 6 times higher than that at 60°.	Promote fluid flow	[62]
Laser	Square	d = 10 μm w = 50 μm	s = 100 µm	Reciprocating sliding	Dry sliding, lubrication (virgin PAO-4, SAE20W50)	Compared to the non-textured surface, the friction coefficient of the textured surface is decreased by 63.88%, 70.53% and 44.99%, and the wear loss is decreased by 32.13%, 60.30% and 86.03% under conditions of dry sliding, virgin PAO-4 lubrication and SAE20W50 lubrication, respectively.	Form transfer layers	[64]
Turning	Asymmetric depth of groove	d = 4 μm	20, 30, 40, 70	Opposite bidirectional rotational sliding	Dry sliding	The sharpness of the groove outlet edge increases when rotated clockwise, resulting in a higher friction coefficient. The increase in friction coefficient is 61.3% and 28.1% for texture densities of 20% and 70%, respectively.	Reduce contact area	[74]
Milling	Asymmetric-depth-profile textures	β = 5.3°, 9.4°, 14°, 21.7°	s = 8.953 mm	Bidirectional relative motion	Newtonian fluids	The asymmetric textures result in lower friction coefficient than symmetric textures. The asymmetric texture with a base angle β = 5.3 ° shows the lowest effective friction coefficient.	Reduce shear stress and generate normal load	[76]

3.2. Multi-Scale Composite Texture

Considering the fact that a single texture has shown the capacity to improve surface tribological properties, it is attractive to explore the design and application of multiscale composite textures. There are generally two types of multiscale composite textures. One is to combine multiple textures of different shapes on the same surface, and the other one is to combine textures of the same shape but different sizes on the same surface [77,78,79]. Segu et al. [80] demonstrated in their study that a combination of circular and elliptical surface textures was used to prepare multi-scale textures with specific formula arrays, as shown in Figure 16. Friction and wear experiments were conducted on surfaces with multi-scale composite textures and non-textured surfaces on a unidirectional friction and wear tester. It was found that the surface with multi-scale texture has a lower and more stable friction coefficient than the non-textured surfaces, which is attributed to the enhanced hydrodynamic lubrication. Meng et al. [81] studied the effect of multi-scale composite pit texture on the frictional performance of sliding bearings, and the results showed that compared to a single texture, a multi-scale composite texture performed excellently in improving load-bearing capacity and reducing the friction coefficient. However, it is not necessarily the case that multi-scale composite textures are superior to single textures or smooth surfaces in improving surface tribological performance. The research of Zou et al. [57] shows that multi-scale composite textures do not significantly improve tribological performance under certain conditions. Therefore, whether it is a single texture or a multi-scale composite texture, it is necessary to fully consider the parameters and experimental conditions of the texture itself when using it.

3.3. Summary

Asymmetric textures and multi-scale composite textures have shown unique advantages in improving tribological properties compared to single-texture and non-textured surfaces. Due to the fact that there is no consensus on the mechanism of the tribological performance of single and symmetric textures, how multi-scale composite or asymmetric textures contribute to the reduction of friction and wear is not clear, and related research deep into the mechanism is limited at the current stage. More influencing factors such as the complex combination of shape, geometry and distribution of textures and the complexity of geometric features of asymmetric textures makes it hard to establish a systemic approach for the design of asymmetric or composite textures. However, the development of processing techniques makes it easier to access asymmetric and composite textures with complex parameters, which promotes the design, discovery and understanding of complex textures in tribological applications.

4. Summary and Prospect

In this work, we review the research progress on the design of surface textures in tribological applications. The influence of surface texture parameters on the tribological properties of friction pairs is discussed from the perspective of texture parameters. It can be concluded that the shape, geometry and density of textures have a significant impact on tribological performance while the distribution of textures has a smaller impact on tribological performance. The main anti-friction mechanism of surface textures includes enhancing the dynamic pressure lubrication capability, storing abrasive debris and lubricants for dynamic lubrication or promoting the formation of friction films on surfaces of friction pairs. A texture with a unidirectional convergence shape, such as a V-shape, in unidirectional lubrication conditions has good tribological properties. The formation of friction film is related to the surface material and lubrication state. Reasonable surface texture design can promote the formation of friction film and reduce the friction coefficient. Under dry friction conditions, a reasonable design of texture depth can store more debris and reduce abrasive wear. Asymmetric textures and multi-scale composite textures also show great anti-friction performance and have potential in specific applications, but they still lack systematic research. The main factor hindering their development is the limited manufacturing approach. Surface texturing technology has been proven to effectively reduce the friction coefficient and energy consumption through the optimization of texture parameters, but considering the significant influence of working conditions on the performance of surface textures, the current design of surface textures still largely relies on case studies. Therefore, some future research directions are proposed based on the review:

• The development of a general mathematical method for designing optimal texture parameters based on the anti-friction mechanism and experimental design parameters;
• The development of standardized testing methods and computational models to better understand and predict the performance of surface textures;
• The development of preparation methods for asymmetric textures;
• The development of a multi-parameter optimization method based on machine learning, which could be used to analyze and predict the tribological properties of textured surfaces, evaluate the weight of different parameters and optimize the design of surface textures.