2.1. The powder-pack boriding process

Samples with a diameter of 20 mm and 4 mm thick were used in this work. The nominal chemical composition for AISI 316L alloy in (wt%) is 0.03 C (max.), 2 Mn (max.), 2-3 Mo (max.), 10-15 Ni (max.), 16-18 Cr (max.), 1 Si (max.), 0.03 S (max), 0.04 P (max), and Fe as balance. [¹⁵]. The samples were embedded in a closed cylindrical case, which contained a GIS® mixture composed of 70% SiC, 20% B₄C and 10% KBF₄, and particle powder size of the mixture around 50 μm. The boriding process was performed at 1273 K for 4 h of exposure in an oven without the use of an inert atmosphere. Once the treatment was completed, the container was removed from the oven and cooled to room temperature. Then, a borided sample was cut in cross-sectioned for the metallographic preparation, and subsequently, a measurement of the layer thickness of the surface to the substrate was observed in a light field by optical microscopy using the Olympus GX51 equipment. In this process, approximately one hundred measurements were taken in different sections of the sample to estimate the thicknesses of FeB and Fe₂B the layers. The thicknesses of the total layer (FeB+Fe₂B) was around of 56.82±0.2 μm.

Fig. 1 show the cross-section view on borided AISI 316L alloy, where the flat layer morphology can be observed due to the influence of alloying elements [¹¹].

Source: The Authors.

Figure 1 The cross-sectional view of borided AISI 316L alloy exposed to the PPBP at 1273 K for 4 h. 

The morphology, growth, and composition of the phases generate by PPBP, are influenced by the alloying elements of the reference material. In pure irons, low and medium carbon steels, the morphology of the layers is intrinsically sawn; however, when the carbon content or other alloying elements increase, the sawn and thickness of the borides layer decrease. Additionally, high concentrations of nickel and chromium, greater than 9 and 6 wt%, respectively, promote the formation of borided with flat layers [¹⁶]. The preferential growth of both intermetallic compounds (Boron-Substrate) is given in the crystallographic direction [001] because the density of the boron atoms is preferential in this direction [¹¹]. The layer consist of the FeB phase with approximately 16.2 wt% boron, with an orthorhombic crystal structure (Fig. 2a) and the phase Fe₂B with an approximate content of 8.83 wt% of boron and tetragonal structure (Fig. 2b) [¹⁷,²⁰]. Bernabé-Molina [¹⁶], mentions the main elements that constitute the stainless steels and their influence during the thermochemical treatment of boriding. The Cr, Mo, Mn elements modify the structure and properties of iron and boron compounds; higher chromium content results in a decrease in the thickness of the layer, as well as the sawn structure. Also, nickel dissolves in Fe₂B, and in some cases, it causes the precipitation of Ni₃B, aiding the layer thickness homogenization. Higher nickel concentration, decrease the layer thickness, and flat the layer-substrate growth interface.

Source: [⁹].

Figure 2 a) Orthorhombic unit structure of the FeB phase. b) Unit structure of tetragonal shape centered on the body of the Fe₂B phase. 

According to the bibliography, the phases FeB and Fe₂B have very similar tribological properties. Table 1 show some mechanical and chemical properties.

Table 1 Some mechanical and chemical properties of FeB and Fe₂B. 

Source: The Authors.

2.2. The sensed Vickers microindentation test

The cross-section of the borided AISI 316L alloy was indented with a commercial indenter (Bruker UMT-2 equipment), which has a Vickers diamond tip. For the indentation load used, load-displacement curves were recorded automatically with the help of the CETR software. In this test, a load of 500 mN was applied for the overall indentations made from the surface to substrate every ~9 μm, and at least three times, the tests were repeated. Fig. 3a and Fig. 3b showing the curves and points of H and E obtained for the borided sample.

Source: The Authors.

Figure 3 Vickers test results a) Curves of load Vs. Displacement, b) Points of hardness and Young´s modulus. 

Finally, X-ray diffraction analysis (XRD) was performed on the surface of borided samples to verify the phases formed. The commercial Bruker D8 equipment with radiation CuKα to λ=0.15418 nm, a voltage of 40 Kv, amperage of 25 mA, a scan between 20 and 80° every 5°, a speed of 0.6 θ/s, and the regulate procedure of scanning 2θ - θ.

2.3. Residual stress calculation

The residual stresses (σ_r) in the iron boride layers were estimated using the Eq. 1 proposed by Totten et al. [²³], taking into account the thermal expansion coefficients of each phase.

Where α_c is the thermal expansion coefficient of each phase equal to FeB=11.5×10^-6 K^-1 and Fe₂B=8.75×10^-6 K^-1, α_s is the thermal expansion coefficient of the AISI 316L=13×10^-6 K^-1, T₁.is the cooling temperature (ambient temperature equal to 303 K), T₂ is the treatment temperature (1273 K), υ_c is the Poisson ratio of the coating, and E is Young’s modulus experimental of the coating.

The value of the Poisson ratio (υ_c) was calculated considering the Eq. 2 proposed by Bhushan [²⁴]:

Where υ_i is the indenter Poisson′s ratio equal to 0.07, E_i is Young’s modulus of the indenter equal to 1141 GPa, and E_r is the reduced Young’s module experimental.

2.4. Sliding reciprocating test

The dry sliding wear tests on the borided AISI 316L alloy were performed at room temperature on a UMT-2 universal tester (Bruker) using a ball-on-flat configuration (Fig. 4), according to parameters of ASTM G113-05 standard procedure with the following conditions: sliding distances of 50 and 150 m, a constant applied loads of 5 and 20 N, and a sliding speed of 30 mm/s. Before the tribological tests, all samples were slightly grounded with abrasive paper and then polished with 0.05 μm Al₂O₃ powder. Alumina balls with a diameter of 6 mm (E=350 GPa, Poisson's ratio υ=0.3) and highly polished surface finish were used as a counterpart material. The mechanic parameters for the experimental conditions of the wear tests are summarized in Table 2. For the overall experimental set, the friction coefficient (CoF) was continually recorded with the aid of the CETR Bruker software; each test was replicated at least three times for all the experimental set.

Source: The Authors.

Figure 4 Schematic diagram of the linear reciprocating wear test using the UMT-2 tribometer. 

Table 2 Contact mechanics parameters obtained for the experimental conditions on the borided AISI 316L alloy. 

Source: The Authors.

After the wear test, wear rate (k) was estimated by Eq. 3 [²⁵]:

Where P is the applied load, S is the total relative distance, and the V is the wear volume emanated over the surface of the worn tracks, was estimated using a non-contact surface profilometry (Contour GT-K 3D, Bruker instrument).

3.1. The microstructure of the Iron boride layers

The formation of iron boride FeB/Fe₂B is a consequence of the high boron potential at high boriding temperatures and the influence of the alloying elements; when an adequate concentration of boron is reached at specific points on the surface of the iron alloy, the Fe₂B crystals begin to nucleate and create a surface layer composed of Fe₂B and FeB, then a diffusion zone appears in the form of peaks with the presence of chromium and molybdenum, these elements act as diffusion barriers at the interface of the borided substrate layer, alongside a reduced amount of carbon, silicon, and boron [²²,²⁴], creating a thin flat layer. Also, the development of the diffusion zone is due to the massive precipitation of Fe outbreaks without coalescence and the presence of products rich in Cr and Si-Mo compounds that grow along the grain boundaries in a preferred crystallographic orientation [001] of the anisotropic crystals in the boride layers (Fig. 5). The preferential growth of FeB/Fe₂B is in the direction of the substrate. High score plus software was used to found the present phases [²⁵-²⁸].

Source: The Authors.

Figure 5 XRD pattern of the iron boride layer. 

3.2. Instrumented indentation Vickers test results

The Young's modulus and the hardness of the boride layer were obtained along the cross-section of the borided AISI 316L obtained by sensed microindentation Vickers test. Fig. 6a shown hardness values in the range of 21 to 19 GPa (for FeB) and 18 to 16 GPa (for Fe₂B), till obtaining a value of 4.5 GPa in the substrate.

Source: The Authors.

Figure 6 The behavior of the a) hardness, b) Young’s modulus, c) H/E ratio, and d) H³/E² ratio along the cross-section of the iron boride layer. 

The Young's modulus is related to the trace size effect obtained during the sensed microindentation test [²⁹, ³⁰]. Also, this value was obtained from the Vickers sensed micro-hardness test. Table 3 show an E value for the FeB phase from 381 to 366 GPa, and in the phase Fe₂B an E value of 365 to 334 GPa is recorded until obtaining a value of 280 GPa in the substrate as shown in the Fig. 6b.

The wear resistance of a material can be evaluated through the plasticity index (H/E ratio). The plasticity of a material is the property to deform permanently and irreversibly when subjected to stresses above its elastic range. According to Fox [³¹], higher values of the plasticity radio, the wear resistance of a material decreases considerably. Fig. 6c shown the behavior of the plasticity index (H/E ratio), along the cross-section profile of the borided AISI 316L alloy, which was obtained from the results of the sensed Vickers microindentation test value of 0.056 and 0.049 for the layer. Also, when the value of H/E is closer to the value of 0.1, the material has excellent wear properties. The contact pressures must be less than the yield stress (σ_f=H/3) of the material; generally, for this type of layer, the contact pressures (Po) do not reach a value higher than 2.5 GPa (σ_f>Po). In this case, yield stresses in the cross-section were between 7.01 to 1.50 GPa for the FeB layer to the substrate, respectively.

Fig. 6d shown the behavior of the plastic deformation resistance (H³/E²), in a range of values from 0.065 to 0.058 GPa for the FeB phase. Subsequently, these values descend to a range of 0.052 to 0.040 GPa for the Fe₂B phase, and finally, in the diffusion zone, the substrate value decreases significantly until reaching the value of 0.001 GPa. In general, these values indicate that PPBP samples have good resistance to plastic deformation in the layer associated with behavior to avoid adhesion failures of the ceramic layer and high wear resistance.

3.4. Sliding reciprocating test results

The measurements of the wear depths, width, and the wear volume were determined for the overall set of experimental conditions using optical profilometry on the surface of the borided AISI 316L with the Vision 64 software installed in the Bruker Contour GT-K 3D Bruker equipment, and the results are summarized in Table 3.

Taking in to account the literature, the untreated material under similar conditions of test a value around 350×10^-6 mm³/Nm of wear loss rate is adequate according to results found by Peruzzo et al.[³²] Also, for the borided AISI 316L, this behavior has not been studied. Thus, the borided layer increased the wear rate around 41 times in comparison with the untreated material.

Table 3 Experimental results. 

Source: The Authors.

3.4.1. Friction coefficient on dry sliding test

The behavior of the friction coefficient (CoF) against different loads and distances for the borided AISI 316L alloy, the values were obtained with the CETR Bruker software presented in Fig. 8.

Source: The Authors.

Figure 8 The behavior of the friction coefficient. a) 5 N, and b) 20 N. 

For the iron boride layer, a short period of execution is observed in Fig. 8a, where the alumina counterpart is coupled with the sample surface during the first 0-10 m of the test due to the high surface hardness, then a transition period where material begins to be removed from the surface of the sample which is evidenced in a better behavior for the distance of 150 m where the CoF is stabilized, showing small variations due to the stick-slip phenomenon, generated by the entrapment and expulsion of wear particles (3 body-mechanism), these wear particles or debris are considered as solid lubricant due to the interaction of the alloy elements with the boron, also presenting additional transition periods due to debris and small porosities on the surface, which interact simultaneously with the formation of grooving, plastic deformation, and oxidation that contains element of the reference and alumina material of the counterpart. On the other hand, in Fig. 8b, the behavior of CoF is similar to that of the 5 N for the distances used.

3.4.2. Wear rate and depth

Fig. 9 shows the profilometry plots obtained on the cross-section views of the wear tracks. From the measurements of the wear depths, the wear volume was determined for the overall set of experimental conditions using the Vision 64 software installed in the Bruker Contour GT-K 3D Bruker equipment. Moreover, the wear rates were determined using Eq. 3.

Source: The Authors.

Figure 9 The behavior of the depth. a) 5 N, and b) 20 N. 

Fig. 9a and Fig. 9b shows that the behavior of depth is more several for a high load to 20 N with a value around ~10 µm, obtaining the highest wear rate around ~9 mm³/Nm of the overall experimental set, attributed to better compliance with the counterpart of Al₂O₃. This behavior decreased in the function of less load and distance with a value of around ~3 µm with a wear rate of around ~5 mm³/Nm. The highest depth on the wear tracks occurs due to the interaction of the different failure mechanisms also associated with the high loads used in the test. In the case of boride layers, the H³/E² ratio has a value of around ~0.065; this value indicates the resistance of the material to wear; when this value is closest to 0.1, the material is difficult to wear. The presence of tensile residual stress around ~1.57 GPa increases the maximum shear stress on the contact zone and promoted wear material loss attributed to the apparition of wear particles due to plastic deformations and material shedding (debris) for the different parameters of the test. For the entire experimental set, no plastic deformations appear at the edges of the tracks, according to the H/E ratio of around ~0.056 GPa near to 0.1. This behavior can be seen in Fig. 9a, and Fig. 9b. Finally, Ashby [³³] states that at sliding speeds below 0.01 m/s for ceramics, there is no significant heating. Likewise, wear is generated by cold mechanisms, predominantly plasticity, cracking, and fracture. For the overall experimental set, the failure mechanisms associated with wear tracks are abrasion, grooving, intergranular cracks (pitting), surface damage (cracking), material agglomeration, and tribo-chemical effect (oxidation) due to the phenomenon of three bodies.