Tribological properties and microstructures of Al₂O₃-TiC-TiB₂ reinforced composites

3.1 XRD and EDS analysis

As was shown in Figure 1, the surface of the Fe₃Al-B₄C-YPSZ-Al₂O₃ laser clad composites mainly consisted of Fe₃Al, Ti₃Al, TiB₂, TiC, Al₂O₃, TiO₂, t-ZrO₂ and ZrSiO₄. It should be mentioned that the Y₂O₃ diffraction was not present in the XRD diagram due to its too low content. As mentioned previously, there was SiO₂ in the water glass. According to the research of Xue et al. [10], it was known that the addition of YPSZ may decrease the intergranular glassy SiO₂ phases by forming ZrSiO₄ and hence improve various mechanical properties of such composites [11]. As the stabilizer of ZrO₂, Y³⁺ in Y₂O₃ can suppress c→t→m transformation, and the content of Y³⁺ has a different suppression effect on ZrO₂ transformation, which enables metastable phase tetragonal ZrO₂ (t-ZrO₂) to exist, while t-ZrO₂ phase structure has stress-induced transformation character, which has transformation toughness action [12].

Figure 1:

X-ray diffraction diagram of the Fe₃Al-B₄C-YPSZ-Al₂O₃ laser clad composites.

In fact, during the laser-cladding process, Ti were able to enter into the molten pool due to the dilution effect of the substrate to laser molten pool; moreover, Ti and Al were all the activity elements, which were able to react with oxygen in the air leading to the formation of TiO₂ and Al₂O₃, respectively. In addition, a large amount of Al in the molten pool came from the substrate. Thermodynamically, a portion of Al had high affinity with Ti, favoring the formation of Ti₃Al.

It was known that Al₂O₃ was also able to react with a portion of TiB₂ in the molten pool, leading to the formations of Ti₃Al and B, thus resulting in the production of the high Ti₃Al diffraction peak. In addition, with decrease of TiB₂, the dilution rate of the substrate to the coating increased, leading to the complete melting of retained TiB₂. As a result, the fine microstructure was obtained. The XRD diagram also indicated that TiB₂-TiC can be produced through in situ metallurgical reactions during the laser-cladding process, the chemical reaction which accorded with thermodynamic laws can take place as follows [13]:

The EDS image indicated that there were B, C, O, Al, Ti and Zr in point 1 of the upper layer (see Figure 2A and B). Combining the XRD image indicated that the precipitates of the upper region consisted mainly of t-ZrO₂, TiO₂, TiB₂ and TiC. It was known that the melting points of TiC (3420°C) and TiB₂ (2980°C) were higher than those of TiO₂ (2050°C)-ZrO₂ (2677°C), so it can be deduced that t-ZrO₂/TiO₂ precipitated around TiB₂/TiC. Vasiliev and Padture [14] reported that Al³⁺ was otherwise insoluble in t-ZrO₂ at room temperature under equilibrium conditions and was in solid solution with ZrO₂, leading to the stabilization of t-ZrO₂.

Figure 2:

SEM image of the upper layer in the Fe₃Al-B₄C-YPSZ-Al₂O₃ laser clad composites and the EDS spectrum analysis, (A) microstructures and (B) points 1 and 2.

The EDS image of point 2 indicated that there were C, O, Al, Si, Ti, Fe and V in the matrix of the upper layer, and the EDS images of these two different points are shown in Table 1. According to the XRD test result, it can be deduced that there were Fe₃Al, Ti₃Al, Al₂O₃, ZrSiO₄ and TiC in the matrix of the upper layer. It was considered that there was a minor amount of inter-dendritic TiC-Fe₃Al eutectic in the matrix of the upper layer. According to Jackson’s theory of interface structure, the solid/liquid interface of the TiC carbide is atomically smooth, and its growth mechanisms are predominantly by lateral growth under equilibrium and near-equilibrium solidification conditions [15]. Thus, it was reasonable that the t-ZrO₂-TiB₂-TiC precipitates embedded in the composite matrix; moreover, it should be mentioned that ZrO₂ has an action of transformation to toughening on the Al₂O₃ matrix [16].

3.2 Microstructures analysis

As is shown in Figure 3A, a stratification was observed in such laser clad composites. It was considered that the completed melting of both ceramics and alloy had occurred during the laser radiation, and there was some redistribution of components because of the difference of the mass density and interfacial tension between two liquid phases, boosted by the drastic convection in the molten pool. The clad zone can be divided into three distinct regions: the upper layer, the bottom layer and the transition layer. The upper layer was approximately 250 μm thick, and there was t-ZrO₂ in this region. It was known that the density of YPSZ was 4.15 g/cm³, which was lower than that of the density of TiB₂ (4.52 g/cm³) or TiC (4.93 g/cm³). Thus, it was reasonable that during the laser-cladding process, YPSZ had a very marked trend to float on the top of the molten pool; moreover, the production of TiO₂ reduced the thermal stress and also the solidification shrinkage of such composites, which was beneficial to the controllable cracking in the clad layer. The SEM micrographs of the upper layer were shown in Figure 3B for sections perpendicular to the growth direction. Combining the XRD image, it can be deduced that TiB₂ was produced in this region. The investigation [17] indicated that during the laser-cladding process, TiB₂ precipitated earlier due to the high melting point, which gave nucleation sites for the future grains, favoring the formation of a fine microstructure.

Figure 3:

SEM micrographs of the Fe₃Al-B₄C-YPSZ-Al₂O₃ laser clad composites, (A) upper layer and (B) TiB₂ and t-ZrO₂.

The SEM image indicated that the TiB₂-TiC precipitates were produced in the bottom layer, and Fe₃Al-Ti₃Al was the matrix of this region (see Figure 4A). The SEM image also indicated that some of the TiB₂ precipitations were located at the grain boundaries and appeared as a small cluster (see Figure 4B). A large amount of the borons were present in the molten pool; near the interface there also existed a large amount of liquid titanium due to the dilution action, so the liquid titanium can react with boron leading to the formation of borides in the interface zone (see Figure 4C). It was also noted that the α-Ti acicular martensite was produced in the heat-affected zone, which was beneficial in improving the mechanical properties.

Figure 4:

SEM micrographs of the Fe₃Al-B₄C-YPSZ-Al₂O₃ laser clad composites, (A) bottom layer, (B) TiC and the TiB₂ cluster, and (C) the interface zone.

3.3 Micro-hardness and wear resistance

The micro-hardness as a function of depth from the top of the coating to substrate is shown in Figure 5. The micro-hardness distributions of the Fe₃Al-B₄C-YPSZ-Al₂O₃ laser clad composites exhibited clearly three steps corresponding to the upper layer, the bottom layer and the transition layer. It was noted that the micro-hardness of the upper layer was in a range of 1650~1750 HV_0.2, which was approximately 4~5 times higher compared with substrate (about 360 HV_0.2). It was considered that an enhancement of the micro-hardness of this upper layer was mainly attributed to the phase constituent. In addition, the micro-hardness distributions decreased to 1200~1350 HV_0.2 in the bottom layer. An enhancement of the micro-hardness in this region was mainly ascribed to the production of the Fe₃Al-Ti₃Al-TiB₂-TiC hard phases/the grain refinement strengthening.

Figure 5:

Micro-hardness distributions of the Fe₃Al-B₄C-YPSZ-Al₂O₃ laser clad composites.

The friction coefficient of such composites and Ti-3Al-2V substrate as a function of the normal load sliding against the quenching and tempering steels is shown in Figure 6. It was noticed that the friction coefficient of Ti-3Al-2V was insensitive to the normal load. Nevertheless, such composite was very sensitive to the normal load. It was clear that the friction coefficient of such composites was always lower than that of the substrate. The coefficient of such composites decreased from 0.46 to 0.37 with increase of the contact load ranging from 15 to 55 N. The friction coefficient test results indicated that such composites exhibited better wear resistance than that of the substrate under drying sliding condition [18].

Figure 6:

Variations of friction coefficients of the Fe₃Al-B₄C-YPSZ-Al₂O₃ laser clad composites and the Ti-3Al-2V substrate.

When the load was 5 kg, the wear test result revealed that the wear volume loss of the Ti-3Al-2V alloy substrate was approximately 4~6 times greater than that of the composites (see Figure 7), which was mainly ascribed to the lower micro-hardness of the substrate; moreover, the wear test results also demonstrated that this coating was tightly bonded to the substrate and exhibited an effective property of the wear resistance. The wear resistance of the bottom layer is worse compared with the upper layer. Nevertheless, the bottom layer also showed better wear resistance relative to the substrate due to the actions of the hard phases and the grain refinement strengthening. In addition, the investigation [19] revealed that Al₂O₃ can further improve the hardness, elastic modulus and high temperature mechanical properties of the Al₂O₃-ZrO₂-Y₂O₃ coating. Moreover, TiO₂ is the most effective additive for improving sintering and mechanical properties [20]. Thus, it was considered that the production of Al₂O₃-TiO₂ was beneficial in improving the wear resistance of the upper layer. Furthermore, the production of Ti₃Al, TiC-Fe₃Al eutectic and TiC-Al₂O₃ also improved the wear resistance of this region. Therefore, it was reasonable that the upper layer of the composites showed an excellent wear resistance.

Figure 7:

Wear volume losses of the Fe₃Al-B₄C-YPSZ-Al₂O₃ laser clad and the Ti-3Al-2V alloy.

The SEM results revealed that the upper layer surface was smoother than that of the substrate, and the upper layer surface was characterized by the presence of few grooves and little adhesive features (see Figure 8A and B); this was proved that such composites showed better wear resistance than that of the substrate. As is shown in Figure 8C, the hard phases were present in the bottom layer, which increased the wear resistance. Nevertheless, due to the lower micro-hardness of the bottom layer, the obvious grooves were produced in the bottom layer surface after the wear test.

Figure 8:

Worn morphologies of (A) the Ti-3Al-2V alloy, (B) the upper layer, and (C) the bottom layer of the Fe₃Al-B₄C-YPSZ-Al₂O₃ laser clad composites.