Influence of coating thickness on the morphology of TiC particles

Materials and fabrication of laser cladding coating

The chemical composition of cladding layer powders is 10 wt-% TiC, Al balance, the purity level was 99–99.5%. The mixture powders were prepared by mechanically mixing, as shown in Figure 1. The AZ31B magnesium alloy with dimensions of 80 mm × 60 mm × 7 mm was used as the substrate in the study. The chemical composition of AZ31B magnesium alloy is 2.9 wt-% Al, 0.08 wt-% Si, 0.04 wt-% Ca, 1.1 wt-% Zn, 0.6 wt-% Mn, 0.003 wt-% Fe, 0.01 wt-% Cu, 0.001 wt-% Ni, and Mg balance.

OPEN IN VIEWER

Figure 1

Particle shapes and sizes of powders.

Before the laser cladding, the abrasive paper and alcohol were used to ground and clean the oxidation film and any organic elements on the substrate surface respectively. The composite powders were mixed with alcohol and then coated on the substrate surface. The thickness of the coating powders was 0.8, 1.2, 1.6 and 2.0 mm, respectively. The coating powders were welded by the JK2003SM type Nd: YAG laser equipment. The laser power is 1200 W, the scanning velocity is 8 mm s⁻¹, the spot diameter is 3 mm, the number of superimposed tracks is 3, the overlap ratio is 30%, and the argon flow rate is 25 mL min⁻¹.

Microstructure characterisation

The S4800 scanning electron microscopy (SEM) was used to study the cross-section of the cladding layer. Metallographic samples were mounted, polished and etched by the Keller's reagent (i.e. hydrofluoric acid 2 mL, hydrochloric acid 3 mL, nitric acid 5 mL, and water 100 mL) in line with standard procedures. The morphology and distribution of the ceramic particles were characterised by the S4800 SEM. The phase of the polished coating was identified by a D8 Advance X-ray diffractometer (XRD, 40 kV, 40 mA, Cu Kα radiation, scanning speed was 4° min⁻¹), and the jade 6 software was used to analyse the data.

Mechanical properties

The MHV2000 type digital micro-hardness tester with a 200 N load and 15 s dwell time was used to measure the micro-hardness from the alloy coating surface to the substrate.

An MM-200 dry sliding wear tester was used to test the tribological behaviour of the coatings. The material of the frication pair was GCr15 with a size of Φ50 mm (i.e. outer radius) × 10 mm (i.e. width). Wear conditions were a normal load of 49 N, a sliding speed of 200 r min⁻¹ and a sliding time of 20 min. The wear volume could be calculated by Equation (1) after measured the depth and width of wear scars [10]: (1) where B is the width of wear scar (mm), L is the length of wear scar (mm), and R is the outer radius of wear ring (mm). The worn surface of the coatings was also investigated by the S4800 SEM.

Microstructure of composite coatings

The surface shapes of cladding layers are the about convex curve, as shown in Figure 2(a)–(d). Because the energy distribution of the laser spot is in line with Gaussian distribution, so the laser energy absorbed by the cladding layer powder is also close to the Gaussian distribution, as shown in Figure 2(e). Some scholars [11-13] found that the surface tension and temperature had an inverse relationship, namely, the higher the temperature, the smaller the surface tension of the material. Therefore, the edge of fused cladding layer metal has the higher surface tension than that of the centre area. Under the action of surface tension, the molten metal will gather from the border to the centre, and then the cladding layer convex curve shape was formed.

OPEN IN VIEWER

Figure 2

Cross-section of the composite coatings: (a) 0.8 mm; (b) 1.2 mm; (c) 1.6 mm; (d) 2.0 mm; (e) Formation mechanism of cladding layer.

After calculation, the dilution rates of cladding layers are about 43.3% (0.8 mm), 20.4% (1.2 mm), 15.7% (1.6 mm), and 13.2% (2.0 mm). Because, under the condition of the same laser processing parameters, the absorbed laser energy by the cladding layer is increasing with the increase of thickness so that the magnesium substrate will absorb the lesser energy.

The distribution of TiC ceramic particles in the four cladding layers are not uniform (i.e. reunion phenomenon). The reunion of TiC ceramic particles in the cladding layers is becoming more obvious with the increase of cladding layer thickness. One possibility is that the thermal expansion coefficients of Al and Mg are 23.2 × 10⁻⁶ K and 26.0 × 10⁻⁶K respectively, but the thermal expansion coefficient of TiC is 7.4 × 10⁻⁶K. The difference in thermal expansion coefficient between TiC ceramic particles and cladding metal is too large to produce a good intermiscibility. Another possibility is that the size of TiC particles is so small (about 10 μm) that make it easy to reunite in the cladding layer [14]. However, in the previous study, the same granularity of TiC ceramic particles was added in the nickel base alloy powder, the reunion phenomenon was not founded; the ceramic particles had a good metallurgical bonding with the cladding metal [15]. The thermal expansion coefficients of Ni element, Cr element, and Fe element, the three main elements in nickel base alloy powder, are 13.0 × 10⁻⁶ K, 6.2 × 10⁻⁶ K, and 12.2 × 10⁻⁶ K respectively. Ceramic particles are mainly distributed on the solid solution skeleton of the cladding layer; the Cr element enriches in the solid solution skeleton. The thermal expansion coefficients between Cr and TiC ceramic particle are extremely close. As a result, the reunion phenomena of TiC ceramic particles in cladding layers are due to the higher thermal expansion coefficient difference with the cladding metal. Besides, the ceramic particles can absorb more laser energy than that of cladding powder [16]. With the increase of the cladding layer thickness, the fusion quality of TiC ceramic particles is gradually reduced, and the total amount of ceramic particles is increased. At last, the reunion phenomenon of TiC ceramic particles will become more and more serious.

The phases of cladding metal mainly consist of Al, γ-Al₁₂Mg₁₇, β-Al₃Mg₂, and TiC, as shown in Figure 3. The emergence of the two phases (i.e. γ-Al₁₂Mg₁₇, β-Al₃Mg₂) illustrates that AZ31B magnesium alloy substrate has a dilution effect on the cladding layer. The distribution of TiC phase in the four cladding metals has no rule, especially when the cladding thickness is 1.2 mm, the content of TiC phase significantly increased. Because the scope of the XRD experiment is about 2 mm × 2 mm, when the detection area just located on the conglobate TiC ceramic particles, the peak strength of the TiC phase will increase significantly.

OPEN IN VIEWER

Figure 3

XRD analysis of the composite coatings.

According to the phase diagram of Al–Mg [17], when the content of Al is lower, the intermetallic compound is γ-Al₁₂Mg₁₇, and mainly distributes in the bottom of cladding layers; when the Al content is higher, β-Al₃Mg₂ is the main intermetallic compound distributed in the middle of the cladding layer, as shown in Figure 4. In addition, the content of Mg element decreases on the left side of the fusion line and then increases on the right side. It is due to the hardness of the γ-Al₁₂Mg₁₇ intermetallic compound is higher than that of the substrate. In the preparation process of the metallographic specimen, there will produce height difference between the cladding layer and the substrate, and the height difference is easy to make the electronic information loss on the left position of the fusion line. Finally, the element distribution information has a certain degree of distortion.

OPEN IN VIEWER

Figure 4

Distribution of Al element, Mg element in the boundary of the substrate and composite coatings.

According to the morphology, the TiC ceramic particles in the cladding layers can be divided into three types: the dispersive small particles, as shown in Figure 5(a); the larger size particles gathered by some melted particles, as shown in Figure 5(b); a large area consists a large number of fused and unfused particles, as shown in Figure 5(c). The former two forms of ceramic particles have the good metallurgical combination with cladding layer metal; the third form of ceramic particles have many cracks between particles, but the around cladding layer metal does not have cracks. On the basis of the location in the cladding layer, ceramic particles can be divided into two kinds: the middle of the cladding layer, as shown in Figure 5(d); the bottom of the cladding layer, as shown in Figure 5(e). After the EDS analysis of the TiC ceramic particles and the surrounding cladding layer metal, the TiC ceramic particle in the middle of the cladding layer has a pore structure, and Al and β-Al₃Mg₂ intermetallic compound exist in the pore and around of particles. At the bottom of cladding layers, some cracks generate from the ceramic particles and grow to the cladding layer metal. Because, after the laser cladding, the unfused ceramic particles with irregular shape will become the source of cracks, and there exist large residual stress between the cladding layer metal and the substrate.

OPEN IN VIEWER

Figure 5

Morphology and distribution of TiC in composite coatings.

Mechanical properties

Figure 6 shows the point location of micro-hardness in cladding layers and the micro-hardness distribution. The average value of surface hardness (i.e. 163.5 HV, 129.2 HV, 97.2 HV and 89.0 HV) is lower than that of the middle zone (i.e. 184.5 HV, 142.3 HV, 119.1 HV and 114.0 HV). The main reason is that, on the one hand, the greater the thickness of cladding layer, the less intermetallic compounds content on the surface; on the other hand, in laser cladding process, some impurity float and gather to the cladding layer surface to reduce the surface micro-hardness. The combining zone hardness (i.e. 203.3 HV, 164.0 HV, 134.7 HV and 127.0 HV) is higher than that of the middle zone. Because the combination zone has high magnesium metal compounds content that will enhance the hardness.

OPEN IN VIEWER

Figure 6

Micro-hardness distribution of the composite coating: (a) 0.8 mm; (b) 1.2 mm; (c) 1.6 mm; (d) 2.0 mm.

The average micro-hardness in the middle of cladding layers decreased with the increase of the thickness. The smaller the thickness, the greater the dilution rate of the cladding layer, the content of metal compounds increase gradually. Also, the fluctuations of micro-hardness increase with the increase of cladding layer thickness. It is attributed to the increasement of TiC ceramic particles content, more and more particles inhomogeneously distribute in the cladding layers.

Figure 7 is the relationship between wear volume losses and micro-hardness average values. The wear volume reduces with the increase of the average hardness, namely, and the wear resistance of the cladding layer increases with the increase of hardness. After wear test, four groups of the wear morphology have the exfoliation and ploughing grooves phenomenon. The thicker the cladding layer, the deeper and wider the ploughing grooves. The element distribution on the wear surface of the 1.2 mm thick cladding layer was investigated, as shown in Figure 8. The distributions of Al element and Mg element are uniform; the O element is due to the oxidation of wear surface in the process of wear test; the Fe element is obtained from the frication pair; the Ti element and C element mainly distribute in the exfoliative zone. As a result, the conglobate TiC ceramic particles have a weak binding force with the matrix to peeling off in the process of wear test.

OPEN IN VIEWER

Figure 7

Wear resistance as a function of hardness for the composite coating: (a) 2.0 mm; (b) 1.6 mm; (c) 1.2 mm; (d) 0.8 mm.

OPEN IN VIEWER

Figure 8

Elements distribution of in the worn surface of the composite coating with 1.2 mm thickness.

Appropriate cladding layer thickness reduced the reunion of TiC ceramic particles degree, and the increased intermetallic compounds enhanced the hardness of the cladding layer. The wear resistance of AZ31B magnesium alloy was improved effectively.