Evaluation of the Effectiveness of Surface Defect Removal by Slide Burnishing

Abstract

This study determines the influence of technological parameters of slide burnishing on the size of surface defects (scratches). The experiment was performed on ring-shaped samples of C45 steel. The samples had scratches made on their surface with a nominal depth from 10 μm to 70 μm. Slide burnishing was carried out using a variable force and feed. It was observed that regardless of the applied force and feed, scratches with a nominal depth of 10 μm and 20 μm were completely removed, and a “crushing” effect occurred. As for other surface defects, they were 2 to 27 times smaller compared to their values before burnishing. The surface roughness parameters Ra, Rt, Rpk, Rk, and Rvk decreased. Their values were 42% to 91% lower than those observed after grinding. The thickness of the strengthened layer ranged from 10 μm to 15 μm, and the degree of strengthening was from 20% to 38% at a depth of 1 μm. Compressive residual stresses occurred in the surface layer. Taking into account the surface layer properties and the effectiveness of surface defect removal, it should be noted that the most beneficial effects were obtained at F = 150 N and f = 0.03 mm/rev.

1. Introduction

Burnishing is one of the methods for shaping the surface texture and physical properties of the surface layer of manufactured objects. In the burnishing process, the tool is pressed against the workpiece with a specified force, and the tool component, in direct contact with the workpiece surface, the so-called burnishing component, moves along this surface in a rolling or sliding manner [1,2]. Depending on the shape of the burnishing element rolled over the workpiece surface, a distinction is made between ball burnishing and rolling burnishing [3,4]. In slide burnishing, the burnishing element usually has a spherical cup with a defined radius on its end [5].

Slide burnishing is used as a finishing operation for workpieces of various shapes. In work by V. Ferencsik et al. [6], it was shown that a reduction in the roughness of the machined surface could be achieved by sliding burnishing of the cylindrical surface of an EN AW-2011 aluminum alloy workpiece. A study by M. Dix and M. Posdzich showed that the combined ball-end milling and burnishing processing of additively manufactured prismatic objects made of AlSi10Mg aluminum alloy, performed on a 5-axis CNC machine tool, made it possible to reduce surface roughness and shape compressive residual stresses [7]. A study by M. Korzynski et al. demonstrated that slide burnishing of valve stems made of 317Ti steel led to improved surface roughness, increased microhardness, and the generation of compressive residual stresses, which resulted in an approximately 4-fold increase in the failure-free operation time of the valves [8]. Research by A. Dzierwa et al. showed that slide burnishing of 42CrMo4 steel discs with a hardness of 40 ± 2 HRC improved the surface layer and tribological properties of the workpieces [9].

The effects of slide burnishing greatly depend on the technological conditions of this process, such as the shape, dimensions, and material of a burnishing element, burnishing parameters, and lubricating liquids. According to J. Maximov et al., diamond elements with a spherical cup are most often used for slide burnishing (84%), while hard alloy (11%) and hardened steel (4%) elements are used to a lesser extent [5]. W. Zielecki et al. used burnishing elements with the radius R = 1 ÷ 2.5 mm and made of synthetic diamond composite for slide burnishing of X19NiCrMo4 steel samples, which resulted in a 23.8% increase in their fatigue strength compared to turned samples [10]. A special tool with a spherical diamond tip for damping vibrations generated by the technological machine bed was used in the cryogenic diamond burnishing process of 17-4 PH steel [11]. H. Kato et al. obtained a significant reduction in surface roughness and increased microhardness of the surface layer of HV250 normalized carbon steel samples by subjecting them to slide burnishing using non-rotating silicon nitride ceramic balls with the diameters of 6.35 mm and 12.7 mm [12]. M. Posdzich et al. investigated the surface texture of 3.1645 aluminum alloy samples after slide burnishing using a burnishing element in the form of a sliding pin made of DLC-coated cemented carbide ended with a 10 mm diameter spherical surface [13].

The results of a study investigating the effects of selective laser melting parameters of Ti6Al4V titanium alloy samples and slide burnishing parameters of these samples on the hardness of their surfaces were reported in [14]. A study by T. Dyl et al. showed that slide burnishing performed using a 2.5 mm radius diamond tool on X2CrNiMo17-12-2 corrosion-resistant steel specimens resulted in reduced roughness and increased surface hardness of these specimens, with the burnishing effects depending on the feed rate and burnishing force [15]. The beneficial effect of diamond slide burnishing on the surface roughness of aluminum matrix composites samples was also found in [16]. K. Konefal et al. analyzed the correlations between surface roughness parameters and residual stresses in X6CrNiMoTi17-12-2 stainless steel samples after slide burnishing with a diamond tool, also focusing on corrosion resistance [17]. Corrosion resistance was also investigated in [18], where the results demonstrated that subjecting samples made of 304 and 316L steel to a treatment combining turning with slide burnishing reduced their susceptibility to stress corrosion cracking. A study A. Skoczylas et al. investigated the effect of feed rate and diamond sliding burnishing force of X6CrNiTi18 stainless steel on its surface topography, residual stresses, microhardness, and positron mean lifetime [19]. The impact of slide burnishing parameters on the surface roughness of 42CrMo4 steel, with its hardness ranging from 21 ÷ 22 HRC, was determined via experiments and analyses using artificial neural networks in one study [20]. FEM simulations and their experimental validation were employed to evaluate the effect of burnishing force on the roughness of a machined surface in slide burnished galvanized steel sheets [21]. The FEM method was also employed by J. Chodór et al. to predict the residual stresses and strains in 41Cr4 steel samples slide burnished using a tool with a diamond tip of 3.5 mm in radius [22].

In the slide burnishing process, an emulsion may be used to reduce the friction coefficient between the tool and the workpiece [16], or the process can be carried out without the use of machining fluids [12]. Studies conducted on C45 steel samples showed that the chemical composition of a liquid affected the roughness of a machined surface, including surface free energy [23]. The chemical composition of the cutting fluid, especially surface-active additives, was also found to affect the surface layer condition and fatigue life of titanium alloy samples [24].

Regular micro-irregularities and surface defects such as inclusions, pitted surfaces, patches, crazing, and scratches can also occur on the surface of a burnished workpiece [25,26]. Surface defects can have different dimensions, so some defects may be difficult to detect. In studies by R. Ameri et al. [27] and D. Bai et al. [28], various methods for detecting surface defects were described. The quality of manufactured items can be improved, among other things, by removing surface defects. This is very important due to the further use of the elements. A work by J. Matuszak et al. presented the results of a study on the effectiveness of brushing treatment in surface defect removal in samples made of aluminum alloy EN AW-2024 and magnesium alloy AZ 91HP [29]. In turn, V. Sandell et al. investigated the effect of chemical treatment on the surface and sub-surface defects of Ti-6Al-4V titanium alloy objects that had been subjected to electron beam melting [30].

The literature review shows that surface defects are removed by brushing [29] or chemical treatment [30]. The positive effects obtained by slide burnishing (low surface roughness, increased microhardness, formation of compressive residual stresses) encourage research to be carried out with the aim of using slide burnishing to reduce surface defects. The aim of this work is to evaluate the effect of slide burnishing on the size of surface defects occurring in machined workpieces.

2. Materials and Methods

The study was performed on thin-walled samples made of C45 steel, with the following dimensions: outer diameter d = 56 mm, inner diameter d_o = 50 mm, and width b = 10 mm. C45 is often used to make medium-duty machinery and plant components, e.g., discs or unhardened gears. Our previous study [23] confirmed the susceptibility of this steel grade to slide burnishing. The pre-treatment of thin-walled samples was grinding. It was carried out on a shaft grinder using a Norton Saint–Gobain electro corundum grinding wheel with the dimensions of 400 × 50 mm and the characteristics of 8A60M7VS3, using a grinding speed of v_s = 35 m/s and a grinding depth of a_p = 0.01 mm.

Surface defects (scratches) were made on the outer surface of the ring samples. To make the scratches, special “scribers” were used, the so-called surface defect generators, which were made from a sintered carbide rod. The working surfaces of the scribers were ground to obtain a conical surface. The double generating angle of cone formers was 90°. The rounding radius of the cone was approximately 40 μm. Figure 1 shows the view of generators and the photograph of a scriber blade.

Surface defects were generated using a Dematec DEM4000 machine (at Lublin University of Technology) by a digital multimeter. To ensure repeatability, a short-circuit detection option was used after the contact between the crack generator and the test sample. One end of the multimeter was connected to the defect generator, and the other to the ring sample. The generator was approached to the sample surface with a step of 10 µm. At each time a short-circuit was obtained, the “scriber” was pressed into the surface by a specified value. Five different nominal depths were used in the study: 10, 20, 30, 50, and 70 µm. The generator’s feed rate was 300 mm/min. Figure 2 shows the stand for making surface defects.

The next step was slide burnishing (SB) to reduce surface defects. The ring-shaped specimens were mounted on a mandrel that was fixed in the chuck of a C11/MB universal lathe (manufactured in Bulgaria). During slide burnishing, the mandrel performed a rotary movement at a speed of n = 160 rev/min. The tool consisted of a spherical cup-shaped tip with a radius of R = 2.0 mm and a spring mechanism (for exerting a pressing force F on the specimen). During burnishing, the tool performed a feed motion f. Figure 3 shows the slide burnishing stand, and

Table 1. Technological parameters of slide burnishing.

No.	Slide Burnishing Force F [N]	Feed f [mm/rev.]
1	100	0.03
2	100	0.08
3	150	0.03
4	150	0.08

lists the technological parameters of the experiment. The slide burnishing conditions were selected based on our previous works [23,24].

The visual evaluation of induced scratches and the effectiveness of their “removal” by slide burnishing was performed using a Keyence VHX5000 digital microscope (Keyence Ltd. HQ & Laboratories, Osaka, Japan). During the analyses, 500× magnification and a technique of combining images were used. The Keyence VHX5000 microscope software allows us to create a 3D model (the assembly of a series of image functions) and then measure the depth in a section perpendicular to the surface of the test object. Figure 4 shows the model of a scratch and its depth.

The work also investigated surface topography and roughness, microhardness, and residual stresses. Topography and surface roughness measurements were performed on a T8000RC 120-400 device from Hommel Etamic (Jenoptik, Villingen-Schwenningen, Germany). The scanned surface area was 1.5 × 1.5 mm, and the analyzed 2D surface roughness parameters were RaRt, Rpk, Rk, and Rvk. In 2D measurements, the measuring length was 0.8 mm, and the measuring section was 4.8 mm. A Leco LM 700at (Leco, St. Joseph, MI, USA) microhardness tester was used to measure microhardness on the surface (penetrator load 500 g) and on angled sections (penetrator load 100 g). Residual stress measurements were performed using a Theta-Theta EDGE portable diffractometer (G.N.R. S.r.l, Agrate Conturbia, Italy). The X-ray source was a chromium lamp. The diffractometer was equipped with a vanadium filter and a 0.5 mm diameter collimator. The X-beam exposure time was 30 s.

3. Results and Discussion

3.1. Visual Defects Analysis Before and After Slide Burnishing

In the first step of the analysis, photographs of the defects before and after slide burnishing are presented in tables, followed by their quantitative description.

Table 2. Example of a comparison of the scratches before and after slide burnishing, for F = 100 N and f = 0.08 mm/rev. (magnification ×500).

Nominal Depth (µm)	Before Slide Burnishing	After Slide Burnishing
10
20
30
50
70

shows an example of a comparison of the scratches before and after burnishing conducted with a force F = 100 N and a feed f = 0.08 mm/rev. Before burnishing, one can observe characteristic defects resulting in the shape of the scratch generator moving along the axis of the workpiece. The tool left a symmetrical groove with characteristic material flashes above the base surface (as shown in Figure 4). The burnishing process removed the scratch, especially for smaller initial depths. Up to a nominal depth of 20 μm, the scratches were completely removed, whereas for a nominal depth ≥30 μm, the traces of surface defects are still visible. For all cases, the flashes above the base surface were removed as a result of the action of the burnishing tool on the machined surface.

Table 3. Example of a comparison of the scratches after burnishing as a function of the applied technological parameters for a scratch with a nominal depth of 70 μm (magnification ×500).

	F = 100 N	F = 150 N
f = 0.03 mm/rev.
f = 0.08 mm/rev.

presents an example of a comparison of the surface defects after burnishing as a function of the parameters used for slide burnishing a scratch with the largest nominal depth of 70 μm. The use of a smaller feed results in smaller contact distances and a longer contact time, which increases the intensity of plastic deformation and, thus, the effectiveness of removing surface defects. Similarly, as the force is increased, the effectiveness of removing surface defects increases.

Figure 5 and Figure 6 show the depths of surface defects obtained during the experiment. At the burnishing forces of F = 100 N and F = 150 N, the surface defects with a nominal depth of 10 and 20 µm were removed. For the depth of 30 µm and above, there was a significant reduction in the depth of surface defects. The maximum degree of defect depth reduction is 97% and was obtained for a defect with a nominal depth of 30 μm and burnished with F = 150 N and f = 0.03 mm/rev. Reducing the feed and increasing the force leads to a reduction in the depth of the surface defects. The depth of the surface defects after slide burnishing is 2 to 27 times smaller than the nominal value. For the entire range of burnishing technological parameters applied in the experiment, the flashes protruding above the base surface of the ring-shaped samples were removed. The factors affecting the effectiveness of surface defect removal are pressure force and burnishing feed. Taking the economic aspect into account, an increase in feed leads to reduced machining time as well as less surface defect removal effectiveness. In turn, an increase in force at the same feed results in better defect removal efficiency. However, the process should be analyzed in conjunction with surface topography characteristics since too high a burnishing force may lead to surface degradation or discontinuities in the surface layer structure.

3.2. Surface Topography

Following grinding, which was used as a pre-treatment before slide burnishing, a unidirectional arrangement of micro-irregularities was obtained on the surface (Figure 7), which results from the action of the grinding wheel. Visible on the surface are the peaks of elevations that dominate the total height of the micro-irregularities. The Sp parameter is approximately 60% of the Sz parameter value. An analysis of the arrangement of the resulting micro-irregularities reveals that groove formation and micro-cutting dominate during grinding. Some discontinuities and pinchers are visible on the surface, which confirms the occurrence of these mechanisms during grinding.

The surface topography of the samples changes after slide burnishing. The peaks of the micro-irregularities become flattened (

Table 4. Surface topography and 3D parameters after slide burnishing conducted with variable technological parameters.

	F = 100 N	F = 150 N
f = 0.03 mm/rev.
	Sa = 0.176 μm; Sz = 2.42 μm Sv = 1.36 μm; Sp = 1.06 μm	Sa = 0.183 μm; Sz = 1.95 μm Sv = 0.955 μm; Sp = 0.994 μm
f = 0.08 mm/rev.
	Sa = 0.284 μm; Sz = 3.32 μm Sv = 2.02 μm; Sp = 1.30 μm	Sa = 0.285 μm; Sz = 3.88 μm; Sv = 2.24 μm; Sp = 1.43 μm

). High-strain areas are visible on the surface. These areas are characterized by large irregularities, which indicates the occurrence of friction and adhesive interaction between the tool tip and the burnished surface, which is particularly noticeable for the feed f = 0.03 mm/rev. The use of a feed of f = 0.03 mm/rev during slide burnishing causes the micro-irregularities to become almost completely “crushed” after grinding, whereas at f = 0.08 mm/rev, they are still visible on the surface. The use of the force F = 150 N during slide burnishing with f = 0.03 mm/rev translates into a reduction in the height parameters compared to the surface burnished with F = 100 N and f = 0.03 mm/rev. Interestingly, after slide burnishing conducted with F = 150 N and f = 0.03 mm/rev, the height of the peaks and the depth of the depressions have similar values. A comparison of the applied feeds during slide burnishing demonstrates that an increase in feed causes an increase in 3D surface roughness parameters, whatever the applied force F.

The greatest changes in relation to the ground surface were obtained for the Sz and Sp parameters. An interesting result is a slight increase in the 3D height parameters for the surface burnished with F = 150 N and f = 0.08 mm/rev compared to the sample that was slide burnished with F = 100 N and f = 0.08 mm/rev. This can be explained by the most likely occurrence of shear damage under the sample surface. The minimum values of 3D height parameters were obtained for the sample slide burnished with F = 150 N and f = 0.03 mm/rev. The Sa parameter for the surface subjected to burnishing with F = 150 N and f = 0.03 mm/rev is about 20% higher than that obtained after slide burnishing of X6CrNiTi18 steel [19]. Compared to the results of previous studies, the values of the 3D surface roughness parameters obtained in this work are similar to those reported after slide burnishing of 317Ti steel [8].

3.3. Surface Roughness

Figure 8, Figure 9 and Figure 10 show the influence of the technological parameters of slide burnishing on the 2D surface roughness parameters. All analyzed 2D surface roughness parameters after slide burnishing are lower than those after grinding. The application of a higher burnishing force (F = 150 N) results in more intensive plastic deformation of the material. The plasticized material fills the depression created by grinding. This causes a reduction in the amplitude parameter Ra (Figure 8) and the height parameter Rt (Figure 9). The Ra and Rt parameters are 70% to 87% lower compared to the values after grinding, which may have a positive effect on tribological and fatigue life. An increase in fatigue life, associated with a decrease in surface roughness parameters, was obtained after slide burnishing of C45 steel workpieces in our previous study [31]. Regardless of the applied force F, an increase in burnishing feed causes the parameters Ra, Rt, Rpk, Rk, and Rvk to increase. This can be explained by an increase in the distance between successive passes of the burnishing tool. A larger distance between the tool paths results in incomplete deformation of micro-irregularities after grinding.

Whatever the technological parameters of slide burnishing, the parameters of the Abbott-Firestone curve change (Figure 10). After slide burnishing, the Rpk parameter is lower from 42% to 70%, Rk from 74% to 85%, and Rvk is lower from 80% to 91% with respect to its value after grinding. This means that the performance properties of the workpieces have improved. The reduced Rpk parameter value means an increase in the service life of mating components after the lapping time. The low Rk value allows us to assume an increase in the load capacity of the surface after slide burnishing.

Compared to the results of previous studies, an analysis of the 2D surface roughness parameters obtained in this work demonstrates that the values of the Ra and Rt parameters are similar to those obtained in [15], but lower than those reported in [12]. All analyzed surface roughness parameters are from 42% to 91% lower than before slide burnishing, which is a better result than in [10]. The minimum value of the Ra parameter was obtained for F = 150 N and f = 0.03 mm/rev (Ra = 0.17 μm), which is more than 3 times higher than the value obtained after burnishing of D16T aircraft aluminum alloy [32]. As for the Abbott-Firestone curve parameters, the reduction in these parameters is similar to that observed in our previous work [23], despite the lower burnishing force, yet greater than in [24].

3.4. Microhardness

Microhardness changes as a result of slide burnishing. Microhardness measurements performed on the surface of the workpiece (Figure 11) confirm that the use of a low feed and a high force causes the largest increase in microhardness (for F = 150 N and f = 0.03 mm/rev ΔHV0.5 ≈ 89). The obtained maximum value of microhardness on the surface is higher (by about 20 HV) than that obtained in our previous work [23], where oil was used as a lubricant and the force was F = 180 N. Regardless of the burnishing feed, the application of a higher force results in a greater increase in microhardness, with these changes being more visible for f = 0.08 mm/rev. The difference is about 27 HV0.5. The use of a greater force during slide burnishing allows for “easier” plastic deformation of the near-surface layers and the disruption of the material’s cohesion. This causes an increase in the dislocations’ density and their displacement, which leads to an increase in microhardness [19]. On the other hand, the use of a higher feed leads to reduced structural homogeneity due to an increase in the distance between successive tool passes, which, consequently, causes lower microhardness.

Figure 12 and Figure 13 show the microhardness distribution as a function of the distance from the surface for samples burnished with variable force F and variable feed f and for ground samples. After pre-treatment (grinding), there is a slight increase in microhardness. The microhardness increase is approximately ΔHV0.1 ≈ 20, and the depth of the changes is 5 μm. A similar microhardness distribution was obtained in our earlier work [23]. In the case of the samples burnished with a force of F = 100 N (Figure 12), the thickness of the strengthened layer is about 10 μm (standard deviation bars do not overlap at this depth). The largest increase in microhardness in relation to the value after grinding occurs just at the surface at a depth of 1 μm for the sample burnished with F = 100 N and f = 0.03 mm/rev and is ΔHV0.1 = 76, while for the sample burnished with F = 100 N and f = 0.08 mm/rev—it occurs at a depth of 3 μm and amounts to ΔHV0.1 = 52. Regarding the samples burnished with a force of F = 150 N (Figure 13), the thickness of the strengthened layer is approximately 15 μm, regardless of the burnishing feed. The largest microhardness changes for the two burnishing feeds occur at a depth of 1 μm where ΔHV0.1 = 92 for f = 0.03 mm/rev and ΔHV0.1 = 82 for f = 0.08 mm/rev. An analysis of the effect of feed depending on the force F shows that the influence of this factor on microhardness is more noticeable during burnishing conducted with F = 100 N, yet it only occurs at a depth of 1 μm.

Relating the obtained results to previous studies, it can be concluded that the thickness of the strengthened layer after slide burnishing is similar to that reported in our previous work [23] and that obtained for X19NiCrMo4 steel [10], and resembles the results obtained after slide burnishing of selective laser melted Ti6Al4V [14], but lower than that obtained for X6CrNiTi18 steel [19]. In this study, the degree of hardening from 20% to 38% was obtained, which is a value similar to that reported in [10] but lower than the value obtained after slide burnishing of X6CrNiTi18 steel [19]. As the burnishing force is increased, more intense plastic deformation occurs due to friction, which translates into higher microhardness and thickness of the strengthened layer. The influence of force F on the obtained effects is similar to the results reported in [12], while the influence of feed is similar to that observed in [14].

3.5. Residual Stress

As a result of pre-treatment (grinding), compressive stresses occur in the top layer, just next to the surface, and their absolute value is approximately σ = 112 MPa (Figure 14). Slide burnishing causes an increase in the absolute value of compressive residual stresses. The greatest increase was obtained after slide burnishing (over 6.5 times) conducted with F = 150 N and f = 0.03 mm/rev, similar to the surface microhardness. The value of residual stresses is σ = −735 MPa. This is more than twice the value obtained in our previous work [31], which may be due to the use of a lower burnishing feed in the experiment. It should be noted that the use of a higher burnishing force and a lower feed makes the absolute values of residual stresses slightly increase. An analysis of the influence of technological parameters on the residual stresses reveals no significant differences in the obtained values.

4. Conclusions

Pre-ground-shaped samples made of C45 steel, with surface defects on the circumference in the form of scratches (having different nominal depths) were subjected to the slide burnishing process. The following conclusions summarize the research results:

• After slide burnishing, regardless of the applied force F and feed f, scratches with nominal depths of 10 μm and 20 μm were removed.
• For the scratches with their nominal depth ranging from 30 μm to 70 μm after SB, only their depth decreased (it became 2 to 27 times smaller than before slide burnishing).
• The use of slide burnishing changed surface topography. The parameters Sa, Sz, Sp, and Sv decreased.
• After slide burnishing, the parameters of the Abbott–Firestone curve, Ra and Rt, were lower than those after grinding regardless of the SB conditions.
• As a result of the SB treatment, the surface layer was strengthened. The microhardness increased by 20–38% at a depth of 1 μm, and the thickness of the strengthened layer was 10 ÷ 15 μm.
• In the surface layer, just below the slide burnished layer, there occurred compressive residual stresses, the value of which oscillated around σ = −707 ± 19 MPa regardless of the slide burnishing parameters.
• Taking into account the obtained surface layer properties and the effectiveness of “crushing” surface defects on C45 steel samples, it can be stated that the most advantageous effects were obtained when the slide burnishing process was conducted with F = 150 N and f = 0.03 mm/rev.
• Based on the obtained results, it can be assumed that the proposed method of reducing the depth of surface defects will work well in the case of small-depth defects. For large-depth defects, slide burnishing is less effective in reducing them.