Effect of Si addition on phase structure and wear resistance of CoCrFeMoNi alloy coatings

3.1. Phase analysis

Fig.1 shows the XRD patterns of CoCrFeMoNi HEACs before and after Si addition. It can be seen from Fig.1 that CoCrFeMoNi alloy is composed of a single FCC phase. After the addition of Si element, the alloy phase is transformed into a dual-phase structure of FCC main phase and HCP phase. The diffraction peaks of these two phases are (111), (211), (220), (311), and (103), (200), (112), (201), (300), (213), (302), (106), (220), respectively. Their space group belongs to Fm-3m (225) and P63/mmc (194), respectively. The diffraction peaks of FCC and HCP phase are basically consistent with those of Ni (No. 04-0850) with FCC structure and Fe₂Mo (No. 06-0622) with HCP structure from the standard card published by Joint Committee on Powder Diffraction Standards (JCPDS), respectively, but the diffraction peaks of the two phases are obviously shifted.

Fig. 1XRD patterns of CoCrFeMoNi and CoCrFeMoNiSi alloys

Compared with the above two standard card data, the diffraction peak of CoCrFeMoNi shifts by 1.249° to the small angle direction, and the FCC and HCP phases in the Si-containing alloy shifts by 1.447° and 0.822° to the small angle direction (based on the strongest peak angle of the standard card data corresponding to each phase), respectively. From the Bragg Eq. (1) [32]:

where, $d$ is the crystal plane spacing, $\theta$ is the diffraction angle, and $\lambda$ is the wavelength. Table 1 shows the characteristic parameters of CoCrFeMoNi alloying elements before and after Si element addition. The diffraction peak shift of CoCrFeMoNi alloy is mainly because the atomic radius of Ni is the smallest, and the substitution of other large atomic radii occupying the corresponding lattice sites will cause the expansion of the crystal cells and the increase of the d value of the crystal plane spacing, which will inevitably lead to the reduction of the diffraction angle, that is, the shift to the small angle direction. After the addition of Si element, the angle of the strongest diffraction peak of FCC phase in this alloy is further shifted from 1.249° to 1.447° in the small angle direction. The reason is that the atomic radius of Si (0.132 nm) is only smaller than that of Mo (0.140 nm), which is much larger than the other four alloying elements. After Si enters and occupies the lattice site of other alloying elements, it is easy to cause cell expansion and increase the crystal plane spacing. Table 2 shows the results of cell refinement of XRD data of CoCrFeMoNi alloy before and after Si addition. As can be seen from Table 2, the lattice constant and unit cell volume of FCC phase increase after the addition of Si elements, which is consistent with the previous analysis that the diffraction peak shifts to the small angle direction.

In addition, it can be seen from Fig. 1 that the phase structure of CoCrFeMoNi alloy before and after Si addition has changed from a single FCC phase to a dual-phase structure of FCC main phase and HCP phase. No complex intermetallic compounds are generated, which is mainly related to the high mixing entropy of this alloy. According to Gibbs' free energy law $\mathrm{\Delta }{G}_{mix}$ (Eqs. (2-6)) [33]:

where $R$ is the gas constant, $c_i$ and $c_j$ are the molar percentages of the components $i$ and $j$, respectively, and $\sum _{i=1}^n{c}_i=1$. When $c_1=c_2=\dots =c_i$, the mixed entropy of the system reaches its maximum. $n$ is the number of alloy components, ${\mathrm{\Omega }}_{ij}$ is the interaction parameter between components $i$ and $j$. $\mathrm{\Delta }{H}_{AB}^{mix}$ is the A-B binary mixing enthalpy calculated by the Miedema model [34](see Table 3). $T$ is the thermodynamic temperature. From Eq. (6), the larger the mixing entropy, the more beneficial it is to reduce the Gibbs free energy of the alloy, which is conducive to inhibiting the trend of ordering and segregation of the alloy [33]. The mixing entropy, mixing enthalpy and Gibbs free energy of CoCrFeMoNi alloy before and after the addition of Si elements are calculated from Eqs. (2)-(6), as shown in Table 4. According to the Hume-Ruthery rule, the atomic radius difference $\delta$ can be expressed as:

where $r_i$ is the atomic radius of component $i$, and $\stackrel{-}{r}$ is the average atomic radius of each component of the alloy. The atomic radius difference between alloying elements (see Table 3) and the atomic radius difference of CoCrFeMoNi alloy before and after the addition of Si element (see Table 4) are calculated by the Eqs. (7) and (8). According to the $\delta$-$\mathrm{\Delta }{H}_{mix}$ phase formation rule[35],when $\delta \le$ 6.6 % and –25 kJ/mol $<\mathrm{\Delta }{H}_{mix}<$ 5kJ/mol, the multi-principal alloy is easy to form the solid solution phase structure of single-phase FCC, FCC+BCC dual-phase and single-phase BCC. As can be seen from Table 4, δ and $\mathrm{\Delta }{H}_{mix}$ of the two alloys are 4.36 % and –8.628 kJ/mol, and 4.06 % and –9.080 kJ/mol, respectively, which are in accordance with the formation rules of $\delta$-$\mathrm{\Delta }{H}_{mix}$ phase. In addition, HCP phase is formed in the alloy after the addition of Si element, which is attributed to the change in the electronic structure and atomic interaction of the alloy after adding Si element, and the decrease in its stacking fault energy (SFE) [36]. In the early stage of deformation, dislocations move and interact with each other in the lattice to form stacking faults. With the increase of Si content, the probability of stacking faults increases significantly, and the stacking faults and dislocation density accumulated in the alloy also increase. The atomic mismatch degree in the alloy increases with increasing Si content, which leads to an increase in the friction force between Shockley partial dislocations, and a corresponding increase in the difficulty of cross-slip of spiral dislocations. This makes the dislocation more inclined to plane slip, and provides favorable conditions for the formation of twins. When the deformation reaches the critical strain, mechanical twins will be formed in the alloy. The critical stress that activates the twin is proportional to the magnitude of the stacking fault energy. Because the stacking fault energy decreases with the increase in Si content, the critical stress value of twinning also decreases. When the critical stress is overcome, the twins begin to form and grow, eventually leading to the phase transformation from FCC to HCP. When the stacking fault density is high enough, that is, when the interlayer spacing on the adjacent {1 1 1} planes is reduced to the atomic scale, the FCC structure in the local region may transform into the HCP structure through the periodic arrangement of stacking faults (e.g., ABCABCA...→ABABAB...) [36, 37]. This phase transformation process is closely related to the movement of dislocations, the formation of twins, and the expansion of stacking faults, which has been confirmed in many studies [31, 36, 37]. The formation of HCP phase increases the dislocation density and phase interface of the alloy. Due to the multiple obstacles such as phase boundaries, stacking faults and twin boundaries, the movement of dislocations will be significantly hindered, thereby improving the strength of the alloy [37, 38].

3.2. Microstructure analysis

Fig. 2 is the metallographic photos of the cross section of CoCrFeMoNi alloy coatings before and after Si addition. From Fig. 2, there is a layer of isotropic fine grain zone near the fusion line of both alloys, and the area away from the fusion line is a columnar grain zone. The formation of fine-grained zone is mainly due to their close connection with the unmelted substrate, which can take away a large amount of heat through thermal conduction and generate a large degree of undercooling, resulting in rapid nucleation of the melt in this area and the formation of a fine equiaxed grain zone. The columnar crystal zone is located far away from the fusion line and flows through the heated area when dissipating heat through thermal conduction towards the substrate, resulting in a relatively gentle temperature gradient. Due to the maximum temperature gradient obtained in the opposite direction of heat flow, the growth of columnar crystal zone exhibits significant directionality. In addition, there are a few holes in the cladding layer area of CoCrFeMoNi alloy. No obvious hole defects are observed in the alloy with Si addition. This is mainly because the melting point and density of Si are the smallest among all alloying elements, which can provide a liquid environment with longer solidification time, lower viscosity, and greater fluidity for the melt solidification process. It helps gas molecules or ions in the melt to quickly escape and be filled and welded by the unmelted liquid alloy, thereby reducing the probability of defect formation, which is consistent with the situation reported in the literature [39-41].

Fig. 2Metallographic photos of the cross section of CoCrFeMoNi: a) and CoCrFeMoNiSi; b) alloy coatings

Fig. 3SEM images of CoCrFeMoNi (a) and CoCrFeMoNiSi (b) alloys

Fig. 3 shows SEM images of CoCrFeMoNi alloy before and after the addition of Si element. From Fig. 2 and Fig. 3, it can be seen that both alloys have typical dendritic structures. Table 5 is the EDS analysis results of CoCrFeMoNi alloy before and after the addition of Si element. In order to better understand the segregation of the same element in the dendrite (DR) and interdendrite (ID) regions, the segregation coefficient $K$ is introduced [32]:

where $C_{DR}$ is the dendrite region concentration of an element, and $C_{ID}$ is the interdendrite region concentration of the element. When $C_{DR}=C_{ID}$, $K=$0 indicates that there is no concentration difference between the dendrite and the interdendrite regions, that is, the segregation coefficient is 0. When the concentration difference between $C_{DR}$ and $C_{ID}$ is larger, the value of $C_{DR}$/$C_{ID}$ or $C_{ID}$/$C_{DR}$ is smaller, and the value of $1-C_{DR}$/$C_{ID}$ or $1-C_{ID}$/$C_{DR}$ is larger, that is, the absolute value of $K$ is larger, indicating that the segregation of dendrite and interdendrite region is larger. When the $K$ value is positive, it means that the concentration of the same element in the dendrite region is higher than that in the interdendrite region, that is, it is enriched in the dendrite region. When the $K$ value is negative, it means that the concentration of the same element in the dendrite region is lower than that in the interdendrite region, that is, it is enriched in the interdendrite region. According to Table 5, Mo element in CoCrFeMoNi alloy is mainly enriched in the interdendrite region, while Co element is mainly enriched in the dendrite region. The distribution of Cr, Fe, and Ni is relatively uniform. After adding Si element, Si element is enriched in the interdendrite region, while the enrichment degree of Mo element in the interdendrite region and Co element in the dendrite region decreased significantly. The enrichment of Cr element changed from the original enrichment in the interdendrite region to the enrichment in the dendrite region. The enrichment of Si in the interdendrite region is mainly due to the lowest melting point of Si among all alloying elements (see Table 1), and its precipitation mainly occurs in the interdendrite region in the later stage of solidification. The decrease of Mo and Co enrichment is due to the high fluidity of Si-containing alloy melt, which can improve the diffusion rate and reduce diffusion resistance of each alloying elements. This is conducive to the full diffusion of each alloying element, so the enrichment degree of Mo and Co is reduced. The change of the Cr enrichment region is due to the fact that the melting point of Cr in the alloying element is second only to Mo. In order to reduce the nucleation resistance of Cr during precipitation, it is easy to use the surface of the precipitated Mo solid particles as the substrate for heterogeneous nucleation, so the enrichment region is similar to Mo. Cr only has a non-negative mixing enthalpy with Mo (see Table 3), and has smaller differences in atomic radius, density, and electronegativity with Fe, Co, and Ni (see Table 1), which helps the full diffusion of Cr element in the Si-containing alloy melt with higher fluidity and makes it easier to enrich in the dendrite regions dominated by Fe, Co, and Ni.

3.3. Hardness and wear resistance

Fig. 4 shows the curve of friction coefficient with time for CoCrFeMoNi HEACs before and after the addition of Si element. After adding Si element, the friction coefficient of the coating decreased from 0.6094 to 0.4455, which decreased by 26.90 %.

Fig. 4Relationship between friction coefficient and time of CoCrFeMoNi and CoCrFeMoNiSi coatings

As shown in Fig. 4, the friction coefficients of the two alloys increase rapidly and then gradually stabilize with the increase of time. The friction coefficients of the alloys before and after the addition of Si element are 0.6094 and 0.4455, respectively. This is mainly because in the initial stage of friction, it is in the running-in stage. With the extension of friction time, the initial line contact changes to surface contact, and the surface wear marks of the sample gradually deepen. The hindering effect of the friction pair is strengthened, resulting in an increase of friction resistance and friction coefficient. As the friction continues, the exposed fresh surface will be rapidly oxidized due to the heat generated by high-speed friction. The generated oxide particles adhere to the friction surface to play a lubricating role, and the amplitude of the friction coefficient gradually decreases and tends to be stable. In addition, the hardness of CoCrFeMoNi and CoCrFeMoNiSi coatings are 308.17 HV and 398.42 HV, respectively. After adding Si element, the hardness of the coating increased by 29.29 %. Generally, the hardness is negatively correlated with the friction coefficient [42], that is, the greater the hardness of the sample, the smaller the friction coefficient. After adding Si element, the increase in hardness and the decrease in friction coefficient are mainly related to the increase of the lattice distortion of FCC phase (such as the increase of diffraction peak shift angle) and the formation of high-strength HCP phase [32, 43] in Si containing alloys (see Fig. 1), as well as the reduction of internal defects (such as holes) (see Figs. 2 and 3). The main reason why the phase transformation from FCC to HCP can improve the wear resistance of CoCrFeMoNiSi alloy is that the HCP structure has a higher hardness and a lower number of slip systems than the FCC structure, thereby reducing the plastic deformation and wear during the friction process. The FCC structure has higher symmetry and more slip systems (12), which means that the dislocations in the FCC crystals can move in multiple directions when the force is applied, which can easily lead to plastic deformation and wear. In contrast, the HCP structure has lower symmetry and fewer slip systems Eq. (3), and the direction of dislocation movement is restricted. Therefore, HCP crystals are more difficult to undergo plastic deformation than FCC crystals. This restriction allows the HCP structure to better resist deformation when subjected to wear, thereby improving wear resistance [44]. Montero-Ocampo [45] found that the wear resistance of Co-27Cr-5Mo-0.05C alloy was significantly improved after the transformation of FCC phase to HCP phase, and pointed out that this was mainly due to the reduction of lattice defects by the formation of the HCP phase, which reduced the dissolution rate of the material and the possibility of crack extension.