3.1. Microstructure and Chemical Composition of Layers
The surface morphology and microstructures of the boride layers produced by the hybrid method using boriding under glow discharge conditions carried out at 850 °C for 2 h for three examined substrate variants are shown in
Figure 1. As can be seen, the surface morphology, especially its development, depends on the chemical composition of the substrate subjected to the boriding process (C45 steel or C45 steel covered with a nickel coating using a chemical, electroless method, or C45 steel covered with a nickel coating using a galvanic method). The development of the layer surface seems to be the smallest in the case of the Fe-B-type layer produced directly on C45 steel (
Figure 1a), and the largest in the case of the Fe-Ni-B-type layer produced on steel previously covered with a galvanic nickel coating (
Figure 1e). Surface morphology may significantly impact the course of frictional wear, especially in the running-in period and the initial stages of linear wear. The greater surface development that occurs with a layer of the Fe-Ni-B-P type can promote good lubrication and, therefore, slower wear. It should also be noted that the surface morphology of the boride layer in the Fe-B variant shows the most significant heterogeneity, manifested by the occurrence of dark depression areas, most likely identified with the porosity of the layers. The surface of the layers produced on both previously nickel-plated substrates has a relatively uniform and compact structure. Regardless of the variant, the produced boride layers have a typical needle structure (
Figure 1b,d,f). However, they differ clearly, depending on the chemical composition of the borided substrate, both in thickness and structure.
Metallographic observations of layers’ cross-sections prove that the presence of nickel significantly impacts the kinetics of the layers’ growth and structure (
Figure 1d,f). They indicate that diffusion processes, and in particular boron diffusion in the presence of nickel, run much faster than in iron, which is supported by the fact that boride layers produced on previously nickel-plated substrates, both chemically (Fe-Ni-B-P type) and galvanically (Fe-Ni-B type), are clearly thicker than iron boride layers, with the thickest layers being formed on a nickel galvanic-plated steel substrate. The estimated thicknesses of the layers produced in the boriding process carried out at a temperature of 850 °C for 2 h, measured by the maximum range of the needles growing into the steel substrate, shown in
Table 3, are 53 ± 6 µm, 80 ± 9 µm and 103 ± 5 µm, respectively, for the layers Fe-B, Fe-Ni-B-P and Fe-Ni-B.
It should be noted that in the iron boride layer (the Fe-B type,
Figure 1b), the needles growing into the substrate are the finest. For the mixed iron–nickel boride layer formed on previously nickel-plated steel (the Fe-Ni-B type,
Figure 1f), the needles are slightly thicker, and they are thickest in a multicomponent layer on steel previously chemically nickel-plated (Ni-Fe-B-P type,
Figure 1d). For all tested substrate variants, both precoated with nickel and uncoated, a bright-etching zone, most likely silicon ferrite, can be observed between boride needles and substrate and under the tips of the needles. The darker zones in the steel substrate are composed of a pearlitic and pearlite-ferritic structure. The cause of structural changes in the substrate in the vicinity of the boride layer is most likely local increases in the concentration of silicon and carbon rejected down into the substrate by the growing boride. Such a phenomenon was observed in the borided high alloy nanobainitic steels we recently investigated [
28,
29]. The formation of the pearlitic zone results from the substrate’s enrichment with carbon due to pushing by the forming layer of borides, which do not dissolve carbon. Similarly, silicon, which is present in the C45 steel and does not dissolve in the layer, accumulates in the vicinity of the borides, forming silicon ferrite. The thickness of the carbon enrichment zone, although uneven, ranges between twice and four times the thickness of the boride layer. Pores are an important characteristic of the microstructure of all tested boride layer variants (
Figure 1b,d,f). In addition to the relatively small oval pores located mainly near the surface, large voids are observed in the areas between the boride needles. These zones are located mainly above the transition region between the layer and the substrate. Particularly large voids between the needles, which have a significant share in the volume of the layer, occur in the Fe-B-type layer (
Figure 1b).
Iron–nickel boride layers are less porous and more compact in structure. However, in the Fe-Ni-B-P-type layer produced on a chemically nickel-plated substrate, the porosity of a similar nature as in the Fe-B layer is maintained, although reduced (
Figure 1b,d). The least porous is the Fe-Ni-B boride layer (
Figure 1e). Moreover, the zone of large voids between the needles does not occur in this layer. The details of the microstructure and chemical composition of the tested boride layers are revealed by scanning microscope SEM observations and EDS chemical composition measurements, as shown in
Figure 2,
Figure 3 and
Figure 4.
The differences in contrast visible in the BSE mode show that, in accordance with the theory, the boride layers have a zone structure. For the Fe-B-type layer, two zones can be distinguished. For the Fe-Ni-B and Fe-Ni-B-P types, there are three zones differing in contrast. The internal zones, which, unlike the relatively thin surface zone, constitute the needles, partially penetrate each other in the transition area between the zones. The chemical composition analyses of the elements’ distribution (EDS) in the layers show that in the zones occurring in the boride layers formed on previously nickel-precoated substrates (
Figure 3 and
Figure 4), there is a chemical gradient resulting from the diffusion of iron from the substrate to the forming layer. Consequently, some of the phases forming these zones are complex nickel–iron boride phases of the general (Fe, Ni)
xB
y type. The X-ray diffraction patterns of the investigated model layers are shown in
Figure 5,
Figure 6 and
Figure 7. The combined analysis of the XRD (
Figure 5,
Figure 6 and
Figure 7) and EDS (
Figure 2,
Figure 3 and
Figure 4) and WDS [
36,
37] results of the tested model boride layers shows that hybrid processing combining boriding under glow discharge conditions at 850 °C with prior modification of the substrate by nickel coating deposition results in the formation of the phases indicated in
Table 3.
In light of the presented results of the analyses of the chemical and phase composition of the layers (
Table 3,
Figure 2,
Figure 3 and
Figure 4), it can be concluded that the thin outer zone of the Fe-B-type layer (
Figure 2) is formed by FeB boride, while the needle inner zone is formed by Fe
2B boride. In the multicomponent layer of the Fe-Ni-B-P type produced on the electroless, chemically nickel-precoated substrate, the darker outer zone (
Figure 3b) is built of complex iron–nickel boride (Fe, Ni)B and, most probably, of Ni
4B
3 boride (
Figure 3b, white precipitates); the intermediate inner zone is formed by needles of iron–nickel boride (Fe, Ni)
2B, and the slightly darker needles of iron boride Fe
2B penetrating into the substrate. It should be noted that in the multicomponent iron–nickel boride layers produced on previously chemically nickel-coated steel, nickel-based phosphides occur in deeper areas, with a dominant (Ni, Fe)
3P type phosphide [
36,
37]. It is the residue of the prior chemical nickel coating. This type of coating in the as-deposited state has an amorphous structure which crystallises when heated to the boriding process temperature of 850 °C, transforming into a diffusion layer based on nickel phosphide with Ni
3P stoichiometry. Due to the simultaneous diffusion of iron from the substrate to the coating, this phosphide is transformed into a mixed nickel–iron phosphide of the type (Ni, Fe)
3P. In the structure of the Fe-Ni-B-P layer, this phosphide is located in the upper half of the thickness of the layer below the zone of large voids between boride needles (
Figure 3, cyan areas). In the Fe-Ni-B-type layer produced on galvanically nickel precoated substrate (
Figure 4), the external zone is formed by nickel base, with nickel–iron boride of the type (Ni, Fe)
2B most probably being an almost pure non-alloyed nickel boride near the surface. The inner darker and lighter interpenetrating needle zones are built along the needles of low or iron borides (containing no nickel). The borides are of the (Fe, Ni)
2B and Fe
2B types, respectively, in the needle bottom and the tip. The lighter zone of the needles directly penetrating into the steel substrate is built of practically iron borides of the Fe
xB (containing no nickel). It is worth emphasising here that the monophase (Ni, Fe)
2B type structure of the outer zone of the investigated Fe-Ni-B boride layers free from the brittle FeB boride type obtained in the hybrid process is of crucial importance from the point of view of the layers’ functional properties, in particular, wear resistance improvement and the resulting increase in the layers’ durability.
3.2. Hardness of Layers
The hardness distributions in the near-surface zone of the tested variants of borided C45 steel with different layers are shown in
Figure 8. As can be seen, the highest hardness levels, approximately 2200 HV0.05, are achieved for iron–nickel boride layers of the Fe-Ni-B type. The zone with a hardness above 2000 HV0.05 ranges from approximately 25 µm up to 55 µm from the surface. This zone corresponds to the inner needles zone built of nickel–iron and pure iron (Ni, Fe)
2B- and FeB-type boride, respectively. The hardness of Fe
2B boride may reach up to 2000 HV. Moreover, the iron may bring a hardening effect on the hardness of the Ni
2B phase, as reported by other authors [
1]. Much lower hardness values are observed in the case of the other two variants of boride layers with a maximum in the surface vicinity zone. For the multicomponent iron–nickel boride layers of the Fe-Ni-B-P type, the hardness reaches approximately 1300 HV0.05. For the iron boride layer Fe-B type, it reaches only near 1000 HV0.05, while the (Fe, Ni)B and FeB borides forming the outside zone should ensure a 2000 HV hardness level. As one can assume, the effect may be related to the presence of small pores in the outer zone of the layers (
Figure 1,
Figure 2,
Figure 3 and
Figure 4), which most likely lower the measured hardness values. The low hardness also observed in the outer zone of iron–nickel Fe-Ni-B-type layers next to its porosity is certainly also related to the contribution of the relatively less hard Ni
2B-type boride that forms the zone (
Table 3). The nature of the hardness distribution in the layers of iron borides Fe-B and iron–nickel borides of the Fe-Ni-B-P type is similar in that it shows an extensive reduced hardness region (
Figure 8) situated in the core region of the layer, which can be attributed to the occurrence of an internal zone of larger voids (
Figure 2 and
Figure 3), especially large in the case of Fe-B-type layer, which certainly affects the hardness measurements’ unambiguity. In the Fe-Ni-B-type boride layer, a similar zone of large voids does not occur, which explains the different hardness distribution. The local hardness increase region is observed deep in the Fe-B and Fe-Ni-B-P layers, with the maximum approximately at a depth of 30 µm (c.a. 800 HV) and at 60 µm (c.a. 1300 HV), respectively. The hardness in these regions is related to the iron Fe
2B-type boride occurrence (
Figure 2 and
Figure 3). However, next to the effect of the voids mentioned above, structural factors cause the hardness of the Fe-B and Fe-Ni-B-P layer variants not to reach as high a hardness level as that of Fe-Ni-B. One of those factors is the presence of silicon ferrite areas separating the boride needles growing into the substrate, as well as the much less hard areas present in the multicomponent layers compared to Fe
2B boride (Ni, Fe)
3P phosphide (
Figure 3).
3.4. Wear Resistance
The results of the wear test of the three types of model boride layers: iron borides of the Fe-B type, nickel–iron borides of the Fe-Ni-B type and multicomponent layers based on nickel–iron borides of the Fe-Ni-B-P type produced on C45 steel using the hybrid method are shown in
Figure 10. A clear effect of the type of the layer and, thus, of its chemical composition on the tribological properties has been observed. The nickel precoating of the C45 steel substrate, previous to the boriding, for both chemical, electroless and galvanic method variants, improves the performance of the boride layers. The linear wear rate for a 200 MPa load (
Figure 10a) significantly decreases compared to uncoated borided C45 steel. This means that the nickel–iron borides exhibit better resistance to wear than simple iron borides. As one can notice, the wear rate is lower for the Fe-Ni-B-type layers compared to the Fe-Ni-B-P layers. This may be related to the much higher hardness level and compressive stress observed in the area just below the surface (
Figure 8 and
Figure 9). For the Fe-Ni-B-P-type boride layers tested at the load of 400 MPa (
Figure 10b), the wear rate increased and became higher than for the Fe-Ni-B-P-type boride layer. In the case of those layers, the nickel electroless precoating gives the essential change in the nature of the wear for 400 MPa load. Observed behaviour suggests the controlling effect related to phosphorus. This effect may be attributed to the phosphide (Ni, Fe)
3P type in the multicomponent Fe-Ni-B-P-type boride layers (
Figure 3). This phosphide exhibits high wear resistance (
Figure 10) and may thus slow down the wear rate of the boride layer. Wear trace SEM analysis (
Figure 11) reveals different wear characteristics for the three types of the examined boride layers. The iron boride Fe-B-type layers tend towards decohesion of the superficial needle FeB zone of the layer (
Figure 11a). The layer’s outer zone residues are visible at the top right corner of the figure. The relatively smooth area at the bottom left corner seems to be the effect of seizing. The spalling effect of the FeB boride outer zone in the Fe-B-type layer could explain the lack of the linear wear stage and practically immediate transition to accelerated wear and seizing. In the case of Fe-Ni-B layers, on the relatively smooth wear track, some cavities, most probably related to the layers’ porosity, have been observed (
Figure 11b). The Fe-Ni-B-P (
Figure 11c,d) layers show mixed behaviour—smooth areas situated in the bottom of the ellipsoidal wear test trace seem to suggest good tribological properties in the deeper zone of the layer, which may be related to the highly wear-resistant (Ni, Fe)
3P phosphide contribution.
3.5. Layers’ Cracking Susceptibility
The scratch tests were used to examine the resistance to crack initiation of the three investigated Fe-B-, Fe-Ni-B- and Fe-Ni-B-P-type boride layers. The data analysis was based on the visual identification of the first cracks or other damage effects on the scratch test trace combined with corresponding values of scratch distance and force determination (
Table 4). As can be seen for the multicomponent Fe-Ni-B-P-type iron–nickel boride layers, the first cracks were observed at the lowest scratch distance and force values. This behaviour may be attributed to the (Ni, Fe)
3P phosphide occurrence that is supposed to weaken the boride layer structure. In the case of the Fe-Ni-B-type iron–nickel boride layers, the critical parameters of the cracking initiation are almost twice as high. One should notice that what may seem surprising is that the highest values are registered for the iron boride Fe-B-type layers.
In addition, to investigate the susceptibility to cracking of the examined model boride layers, the bending tests combined with the acoustic measurement of cracking effects and corresponding critical forces and deflection values were carried out. The bending test results confirm the observation obtained through the scratch test, which shows that the Fe-Ni-B-P type layers are most susceptible to crack formation, occurring at relatively low deformation of samples, as shown in
Table 4.
The tested samples were also subject to SEM observations (
Figure 11). The observations show that the iron boride Fe-B-type layers and the multicomponent Fe-Ni-B-P-type nickel boride layers exhibit a visible tendency for spalling (
Figure 11a,b). The spalling effect is particularly extensive in the iron boride layers’ case (
Figure 11a). In the case of the multicomponent boride layer, a slightly lower surface area seems to be damaged (
Figure 11c). The observed spalling effect in both cases is probably related to the low cohesion in the outside boride zone of the layer, which is composed of the most harmful brittle FeB or (Fe, Ni)B-type phase separated from the substrate by the large voids zone between boride needles. The decohesion of the superficial, wear-resistant layer of the boride layer exposes the deeper zones of the layer susceptible to seizing observed in the wear test (
Figure 11). The Fe-Ni boride layers also crack during the bending test (
Figure 12c); however, this process does not lead to the layers’ fragments’ decohesion (
Figure 12c).
The analysis of the layers’ microstructure, chemical composition, and properties described above leads to the following comprehensive summary. The test results of wear resistance tests of the three types of model boride layers are as follows: iron borides of the Fe-B type, iron–nickel borides of the Fe-Ni-B type, and multicomponent layers based on iron–nickel borides of the Fe-Ni-B-P type produced on C45 steel using the hybrid method combining boriding processes under glow discharge conditions with prior nickel coating of steel, respectively, by galvanic or chemical methods, show that regardless of the nickel coating variant, the modification of the steel substrate with a nickel coating prior to boriding processes, leading to the formation of iron–nickel complex boride layers on the steel, results in a radical improvement in wear resistance. The modification of the borided substrate with nickel in a hybrid process using boriding under glow discharge conditions in accordance with the literature [
1] seems to be a potentially advantageous application solution enabling the elimination of known operational problems resulting from the brittleness of iron boride layers, especially those with FeB stoichiometry. However, the comparative analysis of the investigation results of both iron–nickel boride layers produced using the hybrid method proves that the microstructure and properties of Fe-Ni-B and Fe-Ni-B-P layers show significant differences. A critical evaluation seems essential from the point of view of the application suitability of the tested hybrid solutions. This analysis, in a comprehensive assessment, leads to the conclusion that, despite certain limitations, the optimal solution from the point of view of practical use seems to be a solution based on the initial modification of the substrate with nickel using the galvanic coating method. This solution, although it improves the resistance to wear of boride layers to a slightly lower extent than the perceived until now as the most prospective solution using chemical modification [
36,
37], allows for the formation of iron–nickel boride layers of the Fe-Ni-B type, which exhibit several key advantageous functional features. These layers are thicker and more compact than multicomponent layers of the Fe-Ni-B-P type. Their porosity is significantly lower, and most importantly, these layers are free of large voids between the boride needles, which are a source of susceptibility to spalling of the outer zone of the boride layer. What is extremely important is that Fe-Ni-B-type layers, unlike Fe-Ni-B-P-type layers, are only composed of Ni
2B- and Fe
2B-type borides that are less susceptible to cracking; therefore, they have the expected monophase structure, free of unfavourable FeB-type borides. The outer zone of the layers is a favourable, relatively plastic nickel boride of the Ni
2B type [
1] with a relatively small iron content, in contrast to the Fe-Ni-B-P-type layers, in which in the outer zone is formed by an undesirable iron–nickel boride with a stoichiometry of the type (Fe, Ni)B. In terms of performance properties, Fe-Ni-B layers exhibit the highest hardness, up to almost 1000 HV0.05 higher than the hardness recorded in the cross-section of Fe-Ni-B-P layers. These layers are also much more resistant to cracking and do not show the tendency to spalling of the outer zone during the wear resistance tests. Moreover, among the tested layer variants in the case of the Fe-Ni-B-type layers, the highest level of beneficial compressive stresses in the outer zone is recorded, while in multicomponent Fe-Ni-B-P layers, it is the lowest.
One should also notice that the newly developed iron–nickel Fe-Ni-B-type layers are free from an additional drawback manifested by multicomponent Fe-Ni-B-P-type boride layers, which is their relatively low-temperature stability (970 °C) related to the phosphides present in the boride layers structure. This limitation excludes any steel’s core hardening heat treatment subsequent to boriding for those steel grades which are austenitised over 1000 °C [
29]. On the contrary, the iron–nickel Fe-Ni-B-type boride layers produced in the developed hybrid process variant are composed exclusively of boride phases (
Table 3), which are stable up to at least c.a. 1100 °C [
38].
In the conclusion of the above analysis, it seems justified to consider optimal from the application point of view a hybrid solution that allows for the production of iron–nickel borides of the Fe-Ni-B type, combining boriding under glow discharge conditions with a prior modification of the substrate with a nickel coating produced by the galvanic method. In the perspective of the development and potential application of this solution in the surface processing of advanced high-strength nanobainitic steels, further investigations of the novel hybrid method may strongly strengthen their structures.
A prospective direction of further development of the new hybrid method is plasma and gas boriding technologies using BCl
3-based reactive atmospheres, which ensure the precise control of boriding parameters during processing and thus of the forming boride layers’ phase composition and structure. However, in view of the environmental and safety aspects, the most interesting issue to be studied would be the modified hybrid treatment in which the plasma boriding process would use metalorganic-compound-based reactive media such as, e.g., trimethyl borate, B(OCH
3)
3, as investigated by other authors [
40].