Kinetics and Tribological Characterization of Pack-Borided AISI 1025 Steel

The mass balance equation

The model is concerned with the growth of Fe₂B layer on a saturated substrate with boron atoms as shown in Figure 1. The f(x,t) function describes the distribution of boron concentration in the ferritic matrix before the nucleation of Fe₂B phase. t0Fe2B represents the incubation period needed to form the Fe₂B phase when the matrix reaches a saturation state with boron atoms. CupFe2B is the upper limit of boron content in Fe₂B (=60×103mol m−3) while ClowFe2BT represents the lower limit of boron content in Fe₂B =(−0.0004T+60.373)×103 mol/m³ [25] and the point x(t=t)=v is the Fe₂B layer thickness [28–30]. The term CadsB is the effective adsorbed boron concentration during the boriding process [31]. From Figure 1, a1=CupFe2B−ClowFe2BT defines the homogeneity range of the Fe₂B layer, a2=ClowFe2BT−C0 is the miscibility gap and C0 is the boron solubility in the matrix. The diffusion zone in the substrate underneath the compound layer can be ignored (C0≈0 mol/m³) [32, 33]. The following assumptions are considered for the diffusion model:

1. The growth kinetics is controlled by the boron diffusion in the Fe₂B layer.

2. The Fe₂B phase nucleates after a specific incubation time.

3. The boride layer grows because of the boron diffusion perpendicular to the specimen surface.

4. Boron concentrations remain constant in the boride layer during the treatment.

5. The boride layer is thin compared to the sample thickness.

6. A uniform temperature is assumed throughout the sample.

7. Planar morphology is assumed for the phase interface.

Boundary conditions:

where A(=1⋅1) is defined as the unit area and C0 represents the boron concentration in the matrix. The flux JFe2B and JFe are obtained from the Fick’s first law as

and

The term JFe is null since the boron solubility in the matrix is very low (≈0 mol/m³) [32, 33].

Thus, eq. (4) can be rewritten as

If the boron concentration profile in Fe₂B is constant for the treatment time, Fick’s second law is reduced to an ordinary second-order differential equation as follows:

By solving eq. (8) and applying the boundary conditions proposed in eqs (2) and (3), the boron concentration profile in Fe₂B is expressed by eq. (9) if the boron diffusion coefficient in Fe₂B is constant for a particular temperature:

By substituting the derivative of eq. (9) with respect to the distance x(t) into eq. (7), eq. (10) is obtained:

for 0≤x≤v.

By substituting the derivative of the parabolic growth law v=2εDFe2B1/2t1/2 with respect to the time t into eq. (10), eq. (11) is obtained:

The normalized growth parameter (ε) for the (Fe₂B/substrate) interface can be estimated numerically by the Newton–Raphson method. It is assumed that expressions CupFe2B, ClowFe2BT and C0 do not depend significantly on temperature (in the considered temperature range) [30].

A schematic representation of the square of the layer thickness against linear time v2=4ε2DFe2Bt=4ε2DFe2Btv+t0Fe2B is displayed in Figure 2. tv=t−t0Fe2B is the effective growth time of the Fe₂B layer and t is the treatment time.

Figure 1:

A schematic boron concentration profile through the Fe₂B layer.

Figure 2:

A schematic representation of the square of the layer thickness against treatment time.

The boriding process

AISI 1025 steel was used as the substrate in this study. It had a nominal chemical composition of 0.22–0.28 % C, 0.07–0.60 % Si, 0.30–0.60 % Mn, 0.030 % P and 0.050 % S.

The samples have a cubic shape with dimensions of 10 mm × 10 mm × 10 mm. Prior to the boriding process, the samples were polished, ultrasonically cleaned in an alcohol solution and deionized water for 15 min at room temperature. Afterward the samples were dried and stored under clean room conditions. The samples were embedded in a closed cylindrical case (AISI 304L), containing a fresh Durborid powder mixture. The powder boriding medium, with an average particle size of 30 μm, is composed of an active source of boron (B₄C), an inert filler (SiC) and an activator (KBF₄). The active boron is then supplied by the powder quantity placed over and around the material surface. The pack-boriding process was performed in a conventional furnace under a pure argon atmosphere. This thermochemical treatment was carried out in the temperature range of 1,123–1,273 K for a variable time (2, 4, 6 and 8 h). The boriding temperatures were selected in accordance with the position of the solidus line in the Fe–B phase diagram. Once the treatment was complete, the container was removed from the furnace and slowly cooled to room temperature.

Experimental techniques

The borided and etched samples were cross-sectioned and examined by SEM (JEOL JSM 6300). For a kinetic study, the boride layer thickness was automatically measured with the aid of MSQ PLUS software. To ensure the reproducibility of the measured thicknesses of layers, 50 measurements were taken from different sections of the borided samples to estimate the Fe₂B layer thickness, defined as an average value of the long boride teeth [35, 36]. The presence of different borides formed at the surface of AISI 1025 steel was determined by means of XRD equipment (Equinox 2000) using CoKα radiation at λ = 0.179 nm. The Daimler–Benz Rockwell-C technique, using an indenter hardness tester, was performed to get a qualitative information on the cohesive strength of the boride layers to the substrate. The well-known Rockwell-C indentation test is prescribed by the VDI 3198 norm, as a destructive quality test of coated compounds [37–39].

The principle of this method was presented in Figure 3. A load of 1,471 N was applied to cause coating damage adjacent to the boundary of the indentation. Three indentations were made for each borided sample to assess the cohesion test. The indentation craters on the surfaces of samples were observed by SEM (JEOL JSM 6300).

Figure 3:

Principle of the VDI 3198 indentation test [37].

In this technique, a conical diamond indenter penetrated into the surface of an investigated layer, thus inducing massive plastic deformation to the substrate and fracture of the boride layer.

The damage of the boride layer was compared with the adhesion strength quality maps HF1–HF6 (see Figure 3). In general, the adhesion strengths HF1–HF4 are defined as sufficient adhesion, whereas HF5 and HF6 represent insufficient adhesion (HF is the German short form of adhesion strength) [37–39].

The pin-on-disc wear tests were achieved at ambient conditions without lubrication. Before the test, the samples were cleaned with acetone in order to remove contaminants from the surface. The tested samples had a disc shape with a diameter of 25.4 mm and a thickness of 10 mm. Tribological tests were performed with a diamond-made indenter with a 10-mm-diameter hemispheric. The pin-on-disc tests were carried out in dry sliding conditions using a CSM tribometer (see Figure 4) at room temperature with a relative humidity of 40 %. All tests were conducted for a total sliding distance of 500 m with a sliding speed of 0.08 m/s and the covered radial distance was of 14 mm under a normal load of 5 N. The machine, designed for the pin-on-disc tests, is used to determine the magnitude of friction coefficient and wear as two surfaces rub together. In one measurement method a pin or a sphere is loaded onto the test sample with a precisely known force. The pin is mounted on a stiff lever, designed as a frictionless force transducer. The friction coefficient is determined during the test by measuring the deflection of the elastic arm [21]. Diamond-made indenter with a 10-mm-diameter hemispheric, commonly employed, was used to slide against the surface of borided AISI 1025 steel.

Figure 4:

Schematic diagram of typical pin-on-disc test device (1: elastic arm; 2: weight (1, 2, 5 and 10 N); 3: friction force sensor; 4: pin, ball holders; 5: wear track; 6: rotating disc or cap for liquid testing).

Before the scratch wear tests, the samples with a rectangular shape of dimensions 12 mm × 7 mm × 7 mm were cleaned with acetone in order to remove the contaminants from the surface. The test consists in scratching the sample surface by using an LG Motion Ltd (scratch) with a single-pass under increasing normal load at a rate of 10 N/mm of covered distance. Applied loads were between 0 and 90 N. This permitted the determination of the critical load (L_c) corresponding to the apparition of the layer damage. The scratch wear tests were carried out in dry sliding conditions (at ambient conditions without lubrication) using an LG Motion Ltd (see Figure 5).

Figure 5:

Schematic diagram of typical scratch test device (1: Rockwell-C indenter; 2: weight (1, 2, 5, 10,…, 90 N); 3: trail obtained; 4: tangent force; 5: horizontal displacement of the borided sample (x(t)); 6: substrate; 7: borided surface).

This technique of characterization involves generating a controlled scratch with a sharp tip on a selected area. The tip material (commonly diamond or hard metal (WC)) is drawn across the borided surface under constant, incremental or progressive load. At a certain critical load, the boride layer will start to fail. The critical loads are very precisely detected by means of an acoustic sensor attached to the load arm but can also be confirmed and collated with observations from a built-in optical microscope. The critical load data is used to quantify the adhesive properties of different boride layer–substrate combinations.

SEM observations of boride layers

Figure 6 shows the SEM pictures of the cross sections of boride layers grown on AISI 1025 steel treated at different temperatures for 6 h of treatment. The formed Fe₂B layers exhibited a sawtooth morphology where the boride needles with different lengths penetrate into the substrate. This peculiar morphology was observed in Armco Fe and low carbon steels [13, 18, 27] with a textured growth. For the pack-borided AISI 1025 steel, the preferential growth direction in Fe₂B is along the [001] crystallographic direction because the atom density of boron is maximum along this direction. Therefore, the boride grains with the [001] direction perpendicular to the sample surface grow faster in the same direction of the boron concentration gradient [40]. Since the growth of the boride layer is a controlled diffusion process with a highly anisotropic nature, higher temperatures and/or longer times encouraged the Fe₂B crystals to make contact with adjacent crystals and forced them to retain an acicular shape [40].

Figure 6:

SEM micrographs of the cross sections of Fe₂B layers formed on AISI 1025 steel during 6 h for different boriding temperatures: (a) 1,123 K, (b) 1,173 K, (c) 1,223 K and (d) 1,273 K.

The Fe₂B layer thickness is increased with increasing temperature because the phenomenon of boron diffusion at solid state is activated by increasing the process temperature.

XRD analysis

Figure 7 gives the XRD spectra obtained at the surface of borided AISI 1025 steel at 1,273 K for a treatment time of 8 h. It is seen that the intensities of diffraction peaks are different and depend on the crystallographic orientations of Fe₂B crystals. In addition, the presence of Fe₂B phase was observed in the SEM micrographs of borided AISI 1025 steel (see Figure 6).

Figure 7:

XRD spectra obtained at the surface of the borided AISI 1025 steel at 1,273 K for 8 h of treatment.

The diffraction peaks relative to the Fe₂B phase are readily identified. The same situation was also observed for all the boriding conditions. The growth of Fe₂B layer has a highly anisotropic nature. During the boriding process, boride nuclei first formed on the sample surface then grew toward the substrate. According to Palombarini and Carbucicchio [40], the orientations of the boride nuclei were first random. The [001] direction, corresponding to a maximum of atomic density of boron, is the easiest path for the boron diffusion in the Fe₂B phase, because of the tendency of boride crystals to grow faster along a direction of minimum resistance, perpendicular to the external surface. Afterward the growth of the boride grains along other orientations apart from the [100] direction is slower and is soon suppressed because their growth encounters other grains. As the metal surface is covered, an increasing number of Fe₂B crystals come in contact with adjacent crystals and they are forced to grow in the direction of material substrate, retaining an acicular shape. Furthermore, the diffusion of boron with the formation of Fe₂B is accompanied by the grains coalescence, which fills the structural pores and defects in the substrate material. This leads to consolidation and formation of the rather dense Fe₂B layers [41].

Rockwell-C adhesion test

Figure 8 shows the SEM image of the indentation crater by VDI adhesion test on the surface of AISI 1025 steel borided at 1,273 K for 8 h. The indentation crater obtained on the surface of the pack-borided AISI 1025 steel revealed that there were radial cracks at the perimeter of indentation craters. However, a small quantity of spots with flaking resulting from delamination was visible and the adhesion strength quality of this boride layer is related to the HF3 standard.

Figure 8:

SEM image of the indentation crater by VDI adhesion test on the surface of AISI 1025 steel borided at 1,273 K for 8 h.

It is reported that the adhesion strength quality of the boride layers formed on AISI W2 steel depends on the boriding parameters (time and temperature) [42]. In a recent study, Flores-Rentería et al. [27] have used the Daimler–Benz Rockwell-C indentation technique to examine the cohesive property of the Fe₂B layer formed on AISI 1026 steel at 1,273 K for 4 h of treatment. They found that the cohesion strength quality of the boride layer was related to HF4 category. Ortiz-Dominguez et al. [21] have also employed the same characterization technique to study the adhesion of Fe₂B layer on AISI D2 steel obtained at 1,273 K for 4 h of treatment. They showed that the cohesion strength quality of the boride layer on AISI D2 steel was related to HF5 category.

Tribological characterization

Figure 9 shows the variation of friction coefficient of diamond indenter during sliding against borided surface at 1,273 K with exposure time of 8 h and unborided substrate. It is seen that the borided sample exhibited a friction coefficient lower than that of the unborided substrate. The average friction coefficient for the borided sample at 1,273 K for 8 h of treatment, ranged from 0.015 to 0.030, while the average friction coefficient was between 0.105 and 0.107 for the unborided substrate. It is concluded that the presence of Fe₂B layer improves the tribological behavior of AISI 1025 steel because of the high hardness of Fe₂B phase. In addition, it is well known that the hardness of boride layer plays an important role in improving the wear resistance.

Figure 9:

Variation of friction coefficient of diamond indenter during sliding against borided surface at 1,273 K with exposure time of 8 h and unborided substrate.

Figure 10 gives the SEM images of the unborided and borided surfaces obtained at 1,123 K with exposure time of 8 h, respectively. Figure 10(a) and (b) points out the presence of wear debris, cracks and scratching lines on the unborided surfaces. Figure 10(c) and (d) show the wear scars on the borided surfaces, where some pits and scratching lines are visible.

Figure 10:

SEM micrographs of wear scar on surfaces of the AISI 1025 steel: (a) and (b) unborided surface and (c) and (d) borided surface at 1,123 K for 8 h.

Figure 11 displays the SEM images of the borided surfaces of AISI 1025 steel at 1,123 and 1,273 K with exposure time of 8 h, respectively. It is noticed that the cracks propagate in depth along the scratch trails. They have either a curvilinear form (see Figure 11(a)) or a mosaic (see Figure 11(b)). According to the literature, these types of cracks are characteristic of a Hertzian fracture on brittle solids when a blunted indenter is used. These cracks propagate in depth in a semi-conical shape and start at flaws near the contact surface where high-tension stresses develop [43, 44]. There was not a case of adhesive scaling at the Fe₂B/substrate interface was observed. This was expected since it is well established that the coatings achieved at high temperature present a good adhesion due to the interdiffusion phenomenon ensuring the continuity of metallic interface.

Figure 11:

SEM micrographs of wear scar on surfaces of AISI 1025 steel for two boriding conditions: (a) 1,123 K for 8 h and (b) 1,273 K for 8 h.

Estimation of boron activation energy

The kinetic data regarding the formation of Fe₂B layers on the AISI 1025 steel was used to estimate the boron diffusion coefficient in the Fe₂B layers by applying the proposed diffusion model. The determination of ε parameter, using the mass balance equation (eq. (11)), is necessary to deduce the value of boron diffusion coefficient in Fe₂B at each boriding temperature. Figure 12 gives the time dependence of the square of Fe₂B layer thickness at different temperatures, where the slopes of the straight lines provide the values of growth constants (=4ε2DFe2B). These values can be obtained by a linear fitting through origin at each boriding temperature. Table 1 lists the estimated value of boron diffusion coefficient in Fe₂B at each temperature along with the squared normalized value of ε determined from eq. (11). On the basis of Arrhenius relationship, the activation energy for the boron diffusion in the Fe₂B layer was determined by the slope obtained in the plot of lnDFe2B as a function of reciprocal boriding temperature shown in Figure 13. The temperature dependence of boron diffusion coefficient in Fe₂B was deduced as

Figure 12:

The time dependence of the square of Fe₂B layer thickness at different temperatures.

Figure 13:

Arrhenius relationship for the boron diffusion coefficient in the Fe₂B layer.

where R = 8.314 J/mol/K and T the absolute temperature in Kelvin. The obtained value of boron activation energy (=174.36 kJ/mol) of AISI 1025 steel was interpreted as the amount of energy for the movement of boron atoms in the easiest path direction [100] along the boride layer that minimizes the growth stresses. The diffusion phenomenon of boron atoms can occur along the grain boundaries and also in volume to form the Fe₂B layer on AISI 1025 steel.

Table 2 compares the value of boron activation energy for AISI 1025 steel with the values found in the literature for some borided steels [15, 27, 45–47]. It is concluded that the reported values of boron activation energy depended on various factors such as the boriding method, the boriding agent and the chemical composition of base steel. The obtained value of boron activation energy for AISI 1025 steel is very comparable to that estimated (=178.4 kJ/mol) for AISI 1026 steel [27] by using the same boriding agent. It is noticed that the estimated values of boron activation energy for AISI 1020 and 1040 steels by Celik et al. [45], determined from an empirical approach, are inconsistent with the literature results.

Experimental validation of the kinetic model

To extend the validity of the present model, additional boriding parameters were considered. Figures 14 gives the SEM micrographs of the cross sections of Fe₂B layers formed on AISI 1025 steel at 1,188 K with exposure times of 3 and 5 h; and 1,248 K for 4 and 6 h, respectively.

Figure 14:

SEM images of the Fe₂B layers formed on the AISI 1025 steel for the boriding conditions: (a) 1,188 K for 3 h, (b) 1,188 K for 5 h; (c) 1,248 K for 4 h, and (d) 1,248 K for 6 h.

Equation (16) predicts the experimental Fe₂B layer thickness for the given boriding conditions:

The results obtained from eq. (16) are in good agreement with the experimental data and the theoretical results displayed in Table 3.

Hence, eq. (16) can be used to determine the optimum values of Fe₂B layers’ thicknesses for AISI 1025 steel, depending on the boriding parameters (time and temperature).

It is standard practice to match the case depth with the intended industrial application and base material. As a rule, thin layers (e. g. 15–20 µm) are used to protect against adhesive wear (such as chipless shaping and metal stamping dies and tools), whereas thick layers are recommended to combat abrasive wear (extrusion tooling for plastics with abrasive fillers and pressing tools for the ceramic industry). In the case of low carbon steels and low alloy steels, the optimum boride layer thicknesses range from 50 to 250 µm, and for high alloy steels, the optimum boride layer thicknesses are between 25 and 76 µm.

In addition, this model can be extended to predict the growth kinetics of a double layer (FeB + Fe₂B) at the surface of different ferrous alloys.