The Effects of the Finishing Polish Process on the Tribological Properties of Boride Surfaces of AISI 4140 Steel

Abstract

In sealing, sliding, and power transmission operations, surface quality and contact tolerances have high impacts on material system efficiency. Although the boriding process improves the wear resistance of metallic surfaces, it increases surface roughness, affecting the tribological efficiency of material systems. This study presents the tribological results of AISI 4140 boriding surfaces tested using a dehydrated paste pack boriding method with and without a finishing polish process to reduce the roughness. The duration of the boriding process was 1 h at 1123, 1173, 1223, and 1273 K using boron paste obtained from a commercial source and using a pot-polishing process with Al₂O₃ with a particle size of 0.5 μm for 25 min. The samples with and without the finishing polish process were structurally characterized using X-ray diffraction, and the boride coating adhesion was determined using Rockwell C indentation. The tribological properties of the boride surface with and without the finishing polish process were determined using a reciprocating sliding test, with a ZrO₂ ball as a counter body. The boride surfaces’ crystalline structure changed with polishing, which revealed the FeB phase and reduced the roughness value. These modifications in the surface characteristics altered the adhesion and tribological performance of the coating, resulting in a more stable tribological performance on the polished boride surfaces, with a reduction in the coefficient of friction (Cof) value from 0.75 ± 0.02 for the tribological test on the 1123 K-P sample to 0.59 ± 0.002 for the 1273 K-P sample surface at 20 N of applied load.

1. Introduction

In various industries, several applications include high-demand tribological systems where the standard materials have demonstrated a lack of performance. The high-demand tribological system can consist of high load, high or low temperature, high humidity, a corrosive environment, and other parameters that can significantly contribute to wear and friction or reduce the efficiency of lubricant. AISI 4140 steels are used in bolts, gears, couplings, pump shafts, and other applications in the aerospace, automotive, gas, and oil industries in low-, medium-, and high-demand tribological systems [1,2,3,4]. Several studies have investigated the modification of the surface characteristics of AISI 4140 steel to improve its wear resistance, with the boriding process being one of the studied alternatives [5,6,7,8,9]. The characteristics of the boride layer in AISI 4140 steel depend on the thermochemical process parameter. These layers can present the FeB and/or Fe₂B phases, sawtooth-like morphology, porosity, mono- or multilayer morphology, and variation in the thickness value. Additionally, the hardness and elastic modulus values are around 17 GPa and 286 GPa, respectively [6,7,8,9]. The wear resistance of AISI 4140 steel improves with the boriding treatment, which significantly reduces the wear rate, with a similar coefficient of friction (Cof) to the AISI 4140 substrate in the dry-sliding operation [8,9,10,11].

The boride treatment can be applied using pack, salt bath immersion, paste salt bath electrolysis, ion boride, double-layer glow discharge plasma, and liquid phase plasma boriding techniques [12]. Some of these techniques have been used to improve the wear resistance of metallic materials such as Ti alloys, e.g., TiAlV, TiAlSn, TiA₂, and TiB₂, among others [12,13,14,15,16,17,18]. Pack boriding applied to cemented carbides, particularly WC-Co systems, significantly improves surface properties. Studies have shown an increase in surface hardness from around 1492 HV to 2000 HV with boride layers of 28 μm. The pack boriding process forms complex boride phases, such as Co₂B and W₂CoB₂, which enhance wear resistance [19]. For Ti-6Al-4V, boriding at 900 °C results in a significant increase in surface hardness to 2100 HV, compared with 300 HV in the untreated alloy, marking a 600% improvement. Additionally, after cyclic oxidation at 900 °C for 100 h, the borided alloy showed only one-fortieth of the weight gain compared with the untreated material, demonstrating excellent oxidation resistance [13]. Similarly, the borided Ti-6Al-4V alloy formed a dense TiB₂ layer (15 μm), which enhanced both hardness and oxidation resistance [14]. The pack boriding process typically forms a dual-layer structure consisting of TiB2 and TiB phases, which significantly enhances the surface properties of the substrate [15,16,20]. In cobalt alloys, such as Haynes 25, boriding treatment has demonstrated up to 18 times higher wear resistance at room temperature and four times higher at elevated temperatures of 500 °C, with boride phases such as CoB and Co₂B contributing to enhanced performance [21]. These advancements in pack boriding technology continue to expand its applications across different material systems, offering improved surface properties and enhanced performance in demanding environments. Pack and paste boriding processes have been widely implemented among the various boriding methods due to their cost-effectiveness and relative simplicity. In recent years, a novel approach using dehydrated paste pack boriding (DPPB) has gained significant attention due to its ability to produce uniform and well-adhered boride layers under controlled temperature conditions. The significance of DPPB lies in its ability to generate uniform boride layers, with improved control over layer thickness and morphology. This process has shown particular promise in treating various steel grades, including high-speed steels [22]. Furthermore, recent studies have indicated that DPPB can be effectively combined with other surface treatments to create duplex coatings, expanding its potential applications in industrial settings where enhanced surface properties are required [23].

Nevertheless, roughness is another essential characteristic that controls the performance of a tribological system, and the acceptable initial roughness value depends on the application tolerances (surface texture, ISO-1302) of each sliding system. The initial roughness value and its evolution are related to the initial contact stress distribution and its evolution; the stress distribution controls the friction force that impacts the wear of the contact surfaces, which can increase or decrease the efficiency of the tribo-system [24,25]. However, the boriding process increases the surface roughness, and depending on the thermochemical process, the roughness value could overcome the standard value for some applications [26,27,28]. To address these issues, boride surfaces can be subjected to a finishing polish process in order to achieve the tolerances and surface quality required for some high-demand applications in the aerospace, automotive, oil, and gas industries. These industries require a high control of the surface texture and tolerance, which the finishing polish process treatment achieves by modifying the characteristics and tribological properties of boride surfaces.

Researchers who studied the tribological performance of different boride metallic surfaces, such as Zagkliveris et al. [9] and Márquez-Cortés et al. [8], among others, carried out finishing polish processes before conducting tribological tests. To the best of our knowledge, no studies have reported the tribological performance of surfaces with and without a finishing polish process [28,29,30,31]. This work presents the initial results of the modification of the characteristics and properties of AISI 4140 surfaces borided at treatment temperatures of 1123, 1173, 1223, and 1273 K with and without a finishing polish process. The results of this study could be employed in order to improve the operation time of several tribo-systems in the aerospace, automotive, oil, and gas industries, increasing sliding-systems efficiency and reducing pollution, and as a resource for users in different applications.

2. Materials and Methods

2.1. Surface Preparation Process

An AISI 4140 bar 50.8 mm (2 inches) in diameter, obtained from a commercial source, was cut into cylindrical samples 8 mm in thickness. Before the thermochemical process, the samples were polished with SiC sandpaper from 180 to 1500, followed by a diamond paste of 3 μm to obtain an N4 surface quality (ISO 1302, in the range from 0.125 to 0.32 μm). The samples were cleaned in bi-distilled water and an industrial soap solution, consisting of bi-distilled water, acetone, and ethylic alcohol, in an ultrasonic bath, with each treatment carried out for 10 min. The samples were borided using a dehydrated paste pack boriding (DPPB) process at 1123, 1173, 1223, and 1273 K for 1 h using Durborid G boron paste obtained from a commercial source (Hef-Durferrit-Mexico). The boron paste was composed of boron carbide (B₄C) and cryolite (Na₃AlF₆) containing up to 10–15% water content. To prepare the dehydrated paste, it was dried at 393 K (120 °C) for 20 min to eliminate residual moisture, as described by N. López-Perrusquia et al. [32], M. Y. García-Santibañez et al. [33], and M.A. Doñu-Ruiz et al. [22] (see Figure 1). This drying step was critical to ensure uniform boron diffusion and the formation of a consistent surface layer during the boronizing process. After the thermochemical treatment, the samples were cleaned in ethylic alcohol in an ultrasonic bath for 10 min. Some boride samples at 1123, 1173, 1223, and 1273 K thermochemical treatment temperatures were submitted to the finishing polish process (FPP) for 25 min using a slurry of Al₂O₃ of 0.5 μm at 10% w/w and cleaned with the cleaning process previously mentioned. The boride samples that did not receive the finishing polish process were identified as 1123 K, 1173 K, 1223 K, and 1273 K, and the samples that received the finishing polish process were identified as 1123 K-P, 1173 K-P, 1223 K-P, and 1273 K-P.

2.2. Surface Characterization

The morphology of the boride samples was characterized using a metallographic process with optical microscopy and a Scanning Electron Microscope (CUBE, Emcrafts—EE.UU.) to obtain images of boride layers before and after the finishing polish process (FPP). To study the boride layer morphology, the samples were cut using a diamond cutting disc and then polished with sandpaper and a 5 μm Alumina solution to achieve a mirror-like surface finish. After polishing, the boride samples were treated with a solution containing 2% nitric acid in ethanol for 3 s. The thickness of the layers was measured using the images obtained with optical microscopy and ImageJ 1.53k software, USA. The roughness (Rq) values of the boride samples with and without the finishing polish process were measured using an optical profilometer and calculated using Vision 4.2 software, Veeco instruments, USA. The crystalline structure of the samples was obtained with the XRD patterns of the samples in the Bragg–Brentano configuration from 20 to 90° in 2θ, using a Copper (Cu-Kα = 1.5406 Å) source. The adhesion of the boride layers to the AISI 4140 steel substrate was determined using the Daimler–Benz Rockwell C adhesion test, applying 150 KgF of load. The adhesion marks were studied using SEM images and classified as acceptable or unacceptable failures, following the VDI 3198 standard.

2.3. Tribological Characterization

The tribological characterization of the boride surfaces (E = 283 GPa and υ = 0.3 [8]) of AISI 4140 steel (E = 206 GPa and υ = 0.3) was carried out in a sliding configuration system using a ball-on-flat contact with reciprocating movement of 10 mm of the racetrack at room temperature (around 273 K) and around 35% humidity. Each test was 1800 s (30 min) in duration with a tangential speed of 20 mm/s, using a high-quality surface ball of ZrO₂ (H = 15.7 GPa, E = 250 GPa, and υ = 0.32) 10 mm in diameter as a counter body (according to ASTM G133 standard) [8]. The applied loads were 10, 15, and 20 N, producing a contact pressure on boride surfaces of 1.16, 1.33, and 1.47 GPa, respectively. The friction force produced during the sliding operation was registered in real time, and the wear track produced was characterized using optical microscopy (Zeiss-Mexico), optical profilometry (Zygo-3D profilometer-Mexico), and SEM (JEOL-JSM-7800F-Mexico).

3. Results

3.1. Surface Characterization

3.1.1. Morphology

The 1123 K, 1173 K, 1223 K, and 1273 K samples presented a sawtooth-like morphology with porosity, which is generally obtained with the boriding process in a carbon steel matrix (see Figure 2) [34,35,36]. This kind of morphology combines peak (P) and valley (V) values of diffusion that increase with an increment in the treatment temperature, showing values of P_1123K = 40.6 ± 2.4 μm and V_1123K = 12.8 ± 3.2 μm, P_1173K = 48.2 ± 3 μm and V_1173K = 23.18 ± 4.2 μm, P_1223K = 115.4 ± 3.5 μm and V_1223K = 87.9 ± 5 μm, and P_1273K = 127.2 ± 7 μm and V_1273K = 80.2 ± 9 μm (see Figure 1a and

Table 1. Boride surfaces of AISI 4140: variation in the peak and valley values of sawtooth-like morphology; roughness before and after the finishing polish process (FPP).

Treatment Temperature (K)	Morphology		Roughness (Rq)
	Peak (μm)	Valley (μm)	Before FPP (μm)	After FPP (μm)
1123	40.6 ± 2.4	12.8 ± 3.2	0.6 ± 0.1	0.4 ± 0.04
1173	48.2 ± 3	23.18 ± 4.2	0.9 ± 0.3	0.7 ± 0.08
1223	115.4 ± 3.5	87.9 ± 5	1.1 ± 0.5	0.97 ± 0.03
1273	127.2 ± 7	80.2 ± 9	2.6 ± 1	1 ± 0.07

). Figure 2b shows the SEM images of the AISI 4140 boride surfaces at 1223 K and 1273 K before and after the finishing polish process, revealing the surface characteristics of the 1223 K-P and 1273 K-P samples. The wear on the 1223 K sample produced by the finishing polish process caused a reduction in the peak distance of the sawtooth-like morphology (P at 1223) of the boride layer of 26.8 ± 8 μm, whilst on the 1273 K sample, the finishing polish process reduced the P to 10.2 ± 7 μm at a 1273 distance, resulting in a sawtooth-like morphology of 89.1 ± 6 μm and 107.7 ± 5 μm for the 1223 K-P and 1273 K-P samples, respectively. Lower wear was produced by the finishing polish process of the 1273 K sample compared with the 1223 K sample, which reduced the peak dimension of the sawtooth-like morphology. This result was due to the higher treatment temperature of the 1273 K sample, which improved the diffusion of boron atoms in the metallic matrix, producing a higher hardness of the boride layer than the samples at 1223 K and reducing the wear produced by the finishing polish process on the surfaces [8,10,37,38,39].

The roughness (Rq) value increased with increments in the treatment temperature from 0.6 ± 0.1 to 0.9 ± 0.3, 1.1 ± 0.5, and 2.6 ± 1 μm for the 1123 K, 1173 K, 1223 K, and 1273 K samples, respectively (see Figure 2c). The increments in the roughness changed the surface finish classification from N4 (0.25 ± 0.02 μm) for the substrate to N6 for the 1123 K and 1173 K samples, N7 for the 1223 K samples, and N8 for the 1273 K samples. The roughness value of the sample with the finishing polish process decreased for the samples without the finishing polish process, showing values of 0.4 ± 0.04, 0.7 ± 0.08, 0.97 ± 0.03, and 1 ± 0.07 μm, for the 1123 K-P, 1173 K-P, 1223 K-P, and 1273 K-P samples, respectively. This variation in the roughness changed the surface finish classification to N5 for the 1123 K-P sample, N6 for the 1173 K-P sample, and N7 for the 1223 K-P and 1273 K-P samples (see Figure 2d and Table 1).

3.1.2. Crystalline Structure

Figure 3 shows the XRD patterns obtained from the 1123 K, 1173 K, 1223 K, and 1273 K samples with and without the finishing polish process. The 1123 K and 1123 K-P samples had similar XRD patterns, with peaks at 42.6° and 45.07° that correspond to the (002) plane of the Fe₂B (PDF 00-036-1332) phase and the (021)/(121) planes of the FeB (PDF 00-032-0463) phase. On the other hand, the patterns of the 1173 K, 1223 K, and 1273 K samples changed after the finishing polish process. The XRD pattern of the 1173 K sample had a similar crystalline structure to the 1123 K sample, combining the FeB and Fe₂B phases. Still, for the pattern of the 1173 K-P sample, the peaks at 42.6° that correspond to the (002) plane of the Fe₂B phase disappear, and the intensity of the (020) plane increases at 32.5°, the (101) plane at 37.7°, and the (210) plane at 47.7° of the FeB phase. The increment in the intensity of the FeB peaks indicates that, after the finishing polish process, the surfaces exhibited a higher presence of the FeB phase, reducing the Fe₂B phase for the abrasive operation. The XRD pattern of the 1223 K sample exhibited a peak at 28.3°; this peak does not correspond to the FeB or Fe₂B phases, and thus this peak was considered as a peak of the residual material from the precursor that adhered to the surface. This peak could be a combined peak of the boron carbide (B₄C) [40,41], residual carbon in the graphitic phase [42,43,44], the boron oxide in the B₂O₃ [45,46,47,48], and boric acid (H₃BO₃) [49,50,51]. The 1223 K-P sample shows the (020), (021), (210), and (220) planes of the FeB phase, with the (020) plane being preferential. This result indicates that the crystalline structure changed its preferential plane in the FeB phase from (021) for the layers produced at 1123 K and 1173 K to (020) at the 1223 K treatment temperature. The 1273 K sample presented the (020), (002), (121), and (330) planes of the Fe₂B phase and the (221) plane of the FeB phase, with the same peak at 28.3° as the pattern of the 1223 K sample. Similar to the 1223 K sample, the XRD pattern of the 1273 K-P sample shows peaks that correspond to the (020), (101), (021), (210), (221), and (002) planes of the FeB phase, indicating that the phase that dominates the surface of the boride polishing surfaces was the FeB phase with the (021) plane as the preferential plane. Although the XRD patterns of the boride surfaces before the finishing polish process were observed to exhibit the Fe₂B phase, the XRD pattern of the boride surfaces showed the presence of the FeB phase, indicating that the Fe₂B phase was on the top of the surfaces of the samples and was removed with the finishing polish process.

3.2. Adhesion Test

Figure 4 shows the SEM images of the marks produced by the Rockwell C adhesion tests on the 1123 K, 1173 K, 1223 K, and 1273 K boride samples before and after polishing. According to the VDI3198 standard, the 1123 K and 1173 K samples presented an HF1 classification due to the adhesion test marks of both samples, which showed radial cracks in a semi-linear direction without spallation of the layer. The 1223 K and 1273 K samples had HF2 classification of acceptable failure due to adhesion mark cracks with a radial form and random direction, and small spallation of the layer on fractures of the surfaces of the 1273 K sample related to the HF2 classification. However, the adhesion test marks of the samples changed with the finishing polish process, showing that the adhesion marks of the 1123 K-P and 1173 K-P samples had a higher number of radial cracks, with these adhesion marks classified as HF2. The 1223 K-P and 1273 K-P samples had an HF3 classification due to the adhesion marks produced on these samples, exhibiting small spallation on the fractures on the borders of the marks, with the most significant marks observed on the 1273 K-P surface. The deformation zone size (DZ) and the number, direction, and size of radial cracks indicate the elastoplastic process’s energy liberation resistance for the layer–substrate interfaces. The reduction in the deformation zone (DZ) value from 731 ± 11 to 686 ± 9 and 692 ± 12 to 628 ± 20 μm for the 1123 K, 1173 K, 1223 K, and 1273 K samples, respectively, shows that the 1123 K sample is more brittle than the other samples. Nevertheless, the 1273 K-P sample presented a higher DZ than the 1273 K sample, with a similar number of radial cracks, indicating that the finishing polish process exposed the more brittle boride layer. The increment in the DZ value after the finishing polish process was observed in the adhesion marks on the 1123 K, 1173 K, and 1273 K samples, indicating that the roughness and the Fe₂B phase on the top of the surface could reduce the stress values in the adhesion tests.

Conversely, the number of radial cracks decreased with the increase in treatment temperature. The reduction in radial cracks was achieved due to the increment in temperature, which improved the formation of granular morphology in the boride layer, reducing the cohesion of the layer and improving stress liberation. The increment in the granular morphology can be observed in the SEM images (see Figure 4), where the 1123 K, 1123 K-P, 1173 K, and 1173 K-P samples show the initial formation of the granular morphology, with layer fractures in a semi-linear direction, while the 1223 K, 1223 K-P, 1273 K, and 1273 K-P samples exhibit a granular morphology with radial cracks in random directions. This type of granular morphology was reported by Martini et al. [5], where the Fe₂B crystals grow on the iron surfaces at 1123 K; moreover, the work of García-Santibañes et al. [33] shows a granular morphology of the boride layer [6,7].

3.3. Tribological Results

3.3.1. Wear

The wear tracks produced on the surfaces of the 1123 K sample during the sliding test at 10, 15, and 20 N of applied load have plastic deformation (PD) on the borders and some abrasion (Ab) marks on the center of the wear track. The plastic deformation and the abrasion marks on a tribolayer (Tl) were formed due to the debris accumulation on the worn zones (see Figure 5). A similar wear mechanism was observed on the surface of the 1173 K sample during the sliding tests at 10, 15, and 20 N of applied load, with the difference that these wear tracks did not overcome the deeper valley of the roughness. The worn zones produced for the sliding tests at 10, 15, and 20 N of applied load on the 1123 K and 1173 K samples were covered by tribolayers that had zones with flake deformation (fl), fragile fracture (ff), and detachment (d) that were caused by the fatigue process and exposed the granular microstructure of the boride layer (see Figure 5). However, the tribolayer formed on the worn zones reduced the wear, protecting the contact surfaces. The counter body used during the sliding tests shows a clean wear track with abrasion and small zones covered with a layer formed by material transferred from the boride surfaces.

Figure 6 exhibits the profiles of the wear tracks produced during the sliding tests on the boride AISI 4140 steel surfaces at 1123 K, 1173 K, 1223 K, and 1273 K with and without the finishing polish process. The wear track profiles show a variation in the Y (μm) axis due to the increment in roughness value with the increment in the temperature treatment and the material accumulation formed by the plastic deformation of the asperities and deposition of the debris of the fractured aspirates. This material accumulation produced a tribolayer that protects the boride surfaces during the sliding tests, avoiding the lowest valleys of roughness during the sliding tests.

The width size of the wear track produced by the sliding test at 10 and 15 N increases with the increment in the temperature treatment on the 1123 K, 1173 K, 1223 K, and 1273 K samples. The width size of the worn zone increases with the increment in the roughness value because the wear track is mainly formed by the plastic deformation and fracture of the roughness that forms a tribolayer. Similarly, except for the surface of 1123 K samples, the wear tracks did not overcome the deepest valley of the surface’s roughness, indicating that the tribolayer protected the surfaces (see Figure 6). The width value of the wear tracks produced during the sliding tests at 20 N of applied load on the boriding surfaces without the finishing polish process decreased with the increment in the treatment temperature due to the increment in the surface’s roughness and thickness of the boride layer, thus improving the tribolayer formation and decreasing the plastic deformation of the sample surfaces, respectively. As observed in the wear track profiles, the depth of the worn zone produced on polished surfaces is greater than the valley of the roughness, thus causing wear on the boride layer (see Figure 6). The samples with the finishing polish process developed a similar worn zone width size during the sliding tests at 10 and 15 N of applied load. Meanwhile, the wear track produced for the sliding test at 20 N on the surfaces of the 1173 K-P sample presented a higher width size, with debris accumulation in the center of the wear track.

The width size of the wear tracks produced during the test at 20 N of applied load was smaller on the polished sample surfaces than on the non-polished sample surfaces because the wear track produced on the boride layer of the non-polished samples was mainly the plastic deformation and fractures of the asperities that produced a tribolayer on the worn zone. The asperities on the polished surfaces were removed, reducing the materials that were plastically deformed.

3.3.2. Coefficient of Friction (Cof)

The friction force produced during sliding tests on AISI 4140 steel surfaces at 10, 15, and 20 N of applied load increased with the increment in applied load, showing a coefficient of friction (Cof) value ranging from 0.35 to 0.45 that corresponds to a metallic–ceramic contact [8,25]. Figure 7a–c show the Cof performance produced during the sliding tests at 10, 15, and 20 N of applied load on boride surfaces with and without the finishing polish process. All the tests began with a low Cof value that increased with the sliding distance to reach a steady-state period. However, the 1123 K surfaces exhibited a more unstable initial period, having a performance of the Cof value that began with a low value, increasing to a maximum value and decreasing with variation to reach a stable Ff value period. This variation in the Cof value was observed for a higher elastoplastic deformation of these surfaces during the formation and adaptation of the tribolayer on the worn surfaces than the other samples.

Figure 7d shows the variation in the average coefficient of friction (Cof) value when applying a load of 10, 15, or 20 N, and boride surfaces with and without the finishing polish process. Although the Cof value of the boride surfaces without the finishing polish process shows a decrease with the increment in the treatment temperature, especially for the tests at 15 and 20 N of applied load, from 0.72 ± 0.01 for the 1123 K sample at 20 N of applied load to 0.57 ± 0.007 for the 1273 K sample at 20 N, this reduction shows a high variation, especially for the Cof of the 1173 K surfaces, which produced the highest Cof value of 0.78 ± 0.003 for the tests at 10 N of applied load and decreased to 0.62 ± 0.02 for the tests at 15 and 20 N of applied load. This variation was produced by the initial roughness value, and the elastoplastic performance of the tribolayer was produced by the debris accumulation of the fracture and plastic deformation of the asperities [52,53]. A similar Cof value was reported by Márquez-Corteet al. [8] for the sliding test conducted in the steady-state period. The Cof values of the tribo-systems formed by the boride polishing surfaces had a reduction from 0.75 ± 0.02 for the 1123 K-P surfaces at 20 N of applied load to 0.59 ± 0.002 for the 1273 K-P surfaces at 20 N of applied load, with more stable performance. K.J. Kubiak et al. [27] reported that the initial roughness has a high influence in a tribo-system operation, and M. Sedlacek et al. [52] reported that the Ff is lower with a high roughness value in dry conditions. As was observed, the increment in the temperature treatment reduces the Cof value due to the reduction in the elastoplastic deformation of the surfaces, this due to the increment in the thickness of the boride layer and a higher presence of FeB that has a higher hardness than the Fe₂B phase, as well as the increment in the roughness that reduces the friction force during the sliding test [52,53]. The Cof value decreases with the increment in roughness value due to the tangential shear stress and fatigue process that produces plastic deformation and fractures the large-size asperities in a sliding operation, which are lower than on smoother and denser surfaces, producing a lower friction force and higher wear width on high rough surfaces than on the polished surface. The surface deformation was reduced with the increment in layer thickness and increased the FeB precedence in the boride surfaces due to the FeB phase having a higher hardness than the Fe₂B phase, which increased the load capacity and reduced the surface deformation, thus reducing the Ff during the sliding operation [37,38,39].

Temperature significantly affects the tribological performance of sliding surfaces due to an increase or decrease in heat energy, which can alter the characteristics of the contact surfaces. Sliding operations generate heat energy that is influenced by factors such as pressure, motion, and friction force. Maalekian et al. [54] and Chey et al. [55] propose the following formula to approximate the heat energy produced during the sliding operation:

$$q={\tau }_{fric}v; {\tau }_{fric}=\mu P$$

where q is the heat energy generated during the rubbing operation for friction stress (τ_fric) at the average sliding speed (v). τ_fric is determined using the average friction coefficient (μ) and the normal force P applied during the sliding operation. However, the variation in the contact temperature depends on the material characteristics of the contact surfaces.

Figure 8 illustrates a theoretical approximation of the heat energy produced during sliding tests involving AISI 4140 steel, AISI 4140 boride surfaces, and AISI 4140 boride surfaces that underwent a finishing polish process (FPP). Figure 8 shows a general trend of increasing heat energy with the increase in normal force for all sliding tests, with higher values observed at a 20 N applied load [55,56,57,58]. The AISI 4140 surfaces exhibited lower heat energy production due to the non-boride surface having lower μ values at 10, 15, and 20 N. Nonetheless, the heat energy produced during the sliding tests and the surrounding atmospheric conditions altered the characteristics of the AISI 4140 surface, leading to oxidation in the worn zone [59,60,61]. The heat energy q generated during the sliding tests was greater on boride surfaces compared with non-boride surfaces; this result was especially noted on the 1123 K sample of the boride surfaces without the FPP. The 1123 K and 1123 K-P samples exhibited the highest q values at 10 and 20 N of applied load due to their higher μ values, for both the boride samples with and without the FPP. Additionally, the q value decreased with an increase in treatment temperature on the boride surfaces, indicating that greater boride layer thickness and roughness reduced frictional stress during the sliding operations. Furthermore, the lower q value for boride surfaces that underwent FPP demonstrated improved performance, suggesting that the FPP positively altered the frictional behavior.

4. Discussion

In a tribo-system, the characteristics of the surfaces in contact and relative motion control the wear and friction performance during the system operation. Consequently, when a system application requires a specific surface tolerance, a finishing polish process is necessary to satisfy the surface’s requirements. The boriding process enhances the wear resistance of the AISI 4140 steel surface and modifies the surface characteristics by the diffusion of boron atoms into the metallic matrix [8,9,10,28]. However, the boriding process increases the roughness, and for the AISI 4140 surface, the roughness values increased from 0.25 ± 0.02 before the boriding process to 0.6 ± 0.1, 0.9 ± 2, 1.1 ± 1, and 2.6 ± 0.3 μm for the 1123 K, 1173 K, 1223 K, and 1273 K samples, respectively [26,27,28].

This increase in roughness requires a finishing polish process using an abrasive wear operation to achieve a specific roughness tolerance value for some sliding operations in the aerospace, automotive, gas, and oil industries. The wear of the boride surface by the polishing process reduces the size of asperities; however, at the same time, it modifies the surface characteristics, such as the crystalline structure. The boride surfaces without the polishing process studied in this work exhibited the principal planes of the Fe₂B (PDF 00-036-1332) phase with some planes of the FeB (PDF 00-032-0463) phase. This result is in accordance with studies reported by several authors that work with AISI 4140 boride steel with a monophasic structure and is due to the layer exhibiting a sawtooth-like morphology with a homogenous appearance [8,11]. Nevertheless, the XRD patterns of the 1173 K-P, 1223 K-P, and 1273 K-P boride samples studied in this work exhibited the FeB phase, highlighting the presence of the (020), (101), and (210) planes at 32.5°, 37.7°, and 47.7°, respectively. The polishing process does not have sufficient energy to change the crystalline structure of the surfaces; therefore, the polishing process just removes the surface material that contains the Fe₂B phase, revealing the boride layer with the FeB phase. The revelation of the FeB phase indicates that AISI 4140 boride surfaces have a multilayer coating that combines the Fe₂B and FeB phases in the top layer, followed by an FeB phase layer and Fe₂B phase layer. These three layers are described in further detail in other studies, such as those by Doñu-Ruiz et al. [62] and Lopez-Perrusquia et al. [6], where cracks and spallation of a top layer were observed. The boride treatment temperature produces the multilayer effect because the boride surfaces at 1123 K with and without the boride process showed the same crystalline structure of the boride coating with the Fe₂B phase.

The samples with and without the polishing process presented the “Acceptable Failure” category of the VDI 3198 standard. However, the number of cracks out of the Rockwell C indenter contact area was reduced with the increment in the treatment temperature, with a lower number on the polished surfaces than on the non-polished surfaces. The variation in the number of fractures indicates that the boride coating increases its capacity to absorb energy before being fractured with the increment in the thickness of the coating. Similarly, the finishing process could eliminate a top layer, reducing the number of fractures produced in the boride layer.

Although the wear track of the boride samples with and without the finishing polish process presented a similar wear mechanism that includes the protection of the surfaces in contact with a tribolayer, the finishing polish process modified the tribological performance of the boride surfaces. The width values of the worn zone produced during the sliding tests at 10, 15, and 20 N of applied load were similar. In contrast, the width size on the non-polished surfaces increased during the test at 10 and 15 N of applied load on non-polished surfaces and decreased during the tests at 20 N of applied load with the increment in the treatment temperature. The variation in the width size of the wear track on the non-polished boride samples was caused by the following:

(1)

The increment in the roughness value with the increment in the temperature treatment.

(2)

The top layer presented a lower stress resistance due to its Fe₂B phase content.

(3)

The increment in the coating thickness reduced the surface deformation during the sliding tests (see Figure 9).

(4)

The modification of the surface characteristics of the boride treatment with the increment in the temperature of the boride thermal treatment reduced the frictional stress (τ_fric), reducing the heat energy generated by the rubbing operations.

Figure 9. Sketch of the modification and contact of the boride surfaces before and after the finishing polish process.

Figure 9. Sketch of the modification and contact of the boride surfaces before and after the finishing polish process.

These factors affected the friction force produced during the sliding tests, producing more performance instability of the Cof values and variation in the width size of the wear track than the finishing polish process. These tribological results show that polished surfaces have a better wear and friction performance, and according to Kubiak et al. [53,63], contact surfaces with high roughness value have a shorter lifetime. Although the variation in some parameters of the tribological system can modify the wear and friction performance, the tribological studies of boride surfaces presented during the sliding test at an applied load of around 20 N resulted in a similar Cof value (around 0.6) and the formation of a tribolayer on the worn surfaces. This result indicates that the boride surfaces without the FPP operation present a high plastic deformation and a wear mechanism of adhesion, abrasion, and fragile fractures of the tribolayer [8,9,10,11,64,65]. Additionally, the FPP operation alters the heat energy generated on the boride surfaces, resulting in a more stable variation in the q value as the treatment temperature increases. This behavior may be attributed to the more stable structural properties and roughness characteristics observed in the samples treated with the FPP operation. These samples exhibit less than 11% variation in roughness and show a greater presence of the FeB phase.

5. Conclusions

• The boride AISI 4140 surfaces at treatment temperatures of 1123, 1173, 1223, and 1273 K presented the FeB and Fe₂B boride phases, with an incremental boride layer thickness and roughness with a sawtooth-like morphology.
• The finishing polish process reduced the roughness of the boride surfaces, obtaining an N7 surface quality classification for the 1273 K sample and removing a top layer formed by the Fe₂B phase and residual material of the thermochemical treatment, revealing the FeB phase of the boride layers.
• Although the boride layer with and without the finishing process presented an acceptable adhesion characteristic, the number of fractures around the Rockwell C mark decreased with increments in the temperature of the thermochemical treatment, showing an increment in the absorption of energy before the layer fracture with increments in the thickness of the layer.
• The boride surfaces with and without the finishing polish process showed a similar wear track, with the formation of a tribolayer that protected the surfaces. However, the finishing polish process improves the stability of the tribological performance, showing a more constant reduction in the Cof value due to the revelation of the FeB phase and increments in the thickness of the boride layer by increments in the treatment temperature that reduce the boride surface deformation, production of heat energy, and adhesion forces between the Pin and the boride surfaces.