The grinding surface characteristics and evaluation of particle-reinforced aluminum silicon carbide

2.1 Sample preparation

The test materials used for the experiments are made of cast aluminum alloy through the pressureless infiltration process. The basic materials, cast aluminum, are therefore reinforced by SiC ceramic particles. Owing to its high plasticity, the cast aluminum alloy is characterized with continuous plastic deformation. The elasticity modulus of SiC particles is 420 GPa, along with the Poisson’s ratio of 0.15, particle volume fraction of 62%, and average particle size of 40 μm. The detailed information about the materials is listed in Table 1. The surface for grinding was selected to be the upper sides of the SiCp/Al. Samples were divided into small pieces of 15 mm×35 mm×10 mm by using a wire-cutting device. Then, they were polished to have a surface roughness (R_a) of 100 nm.

2.2 Experimental setup and conditions

The grinding equipment was the advanced Ultrasonic 50 (DMG, Bielefeld, Germany) for the tests, the maximum ultrasonic power of which was 300 W and the feeding precision was 1 μm. The high-frequency axial vibration signals of 16.5–30 kHz released by the ultrasonic generator were then converted into mechanical vibrations by the transducer. The amplitude of mechanical vibration was increased by 1–20 μm by the transformer, which was then passed onto a diamond-grinding wheel. The wheel then constantly worked on the surfaces of the workpieces at quite high frequency, which was ultrasonic grinding, as shown in Table 1. Before grinding, the laser interferometer (precision, 1 nm) was used to check the amplitude of various cutting devices at different frequencies to determine the ultrasonic setup of each cutter. For the purpose of this paper, the ultrasonic frequency was set to be 20 kHz and the amplitude was 4 μm. It was traditional grinding when the ultrasonic assistance was shut off. The diamond-cutting tool was made by DMG SAUER, and it was a nickel-eletroplating diamond grinding wheel with a diameter of 5 mm and granularity of D126 (120/140). The workpiece was made of SiCp/Al with a dimension of 140 mm×24 mm×15 mm. The cooling liquid for inner cooling and outer cooling of the testing device was the specialized grinding liquid made by Blaster, as is shown in Figure 1. The principal axis of the device run at 5000 rpm, and the feeding rate was 40 mm/min. The grinding depths, α_p, respectively, were 10, 20, 30, and 40 μm.

Figure 1:

Experimental setup.

2.3 Experimental measurement

Kistler 9257B, a high-precision microdynamometer (Kistler 9257B, Winterthur, Switzerland), was used to record the data of the grinding tests on SiCp/Al. The data were collected at a frequency of 9 kHz. The surface morphology of ground workpieces was studied under a scanning electron microscope (SEM) (FEI, Eindhoven, The Netherlands) by using different cutting tools. The SEM was a kind of environmental SEM made by the Dutch company FEI, and its model was FEI Quanta200 FEG. Since silicon carbide was not conductive, the low-vacuum mode was chosen for observation and measurement. The Olympus optical microscope was used for examining the wear of cutting tools. For the purpose of enhancing the reliability of the test data, three tests were performed on each set of process parameters to obtain the average value. When the ultrasonic generator was turned off, traditional grinding tests were carried out.

3.1 The surface characteristics and formation mechanism analysis of particle-reinforced SiCp/Al

The surface morphologies of SiCp/Al after grinding are shown in Figures 2–7. Considering the structural characteristics of composite materials, the damages could be divided into the following modes: particle damages, matrix destructions, and interface failures. Particle damages include grain crush, surface scratch, indention, fallout, crack, expansion, and attachment of particles; matrix destructions cover matrix cracking, coating and substrate warping, and surface grain; and interface failures include interface cracking and debonding.

Figure 2:

The destruction of the particles under the extrusion.

Figure 3:

Surface grains and texture features.

Figure 4:

Aluminum matrix cracking and Silicon carbide particles pulled up.

Figure 5:

Particle’s cracks and propagation.

Figure 6:

Aluminum spurs and particle attachment.

Figure 7:

The ideal surface morphology.

In the process of grinding, SiCp/Al was broken and scratched under the squeezing and engraving of the diamond-grinding wheel, as is shown in Figure 2. Under the effect of repeated pressing and scratching, the surface of SiCp/Al formed grains, as is shown in Figure 3. The aluminum substrate had shear slip and deformation, and the silicon carbide particles were pushed inside, as is shown in Figure 3. When the grinding force was larger than the interface bonding strength of the composite, the substrate would crack and even the silicon carbide particles would fall off, as is shown in Figure 4.

When the stress on the silicon carbide particles was beyond its elastic limit, there were cracks and crack propagation, as is shown in Figure 5. The aluminum substrate was softened under high temperature and high pressure, and coated on the surface. Meanwhile, the workpieces had many “aluminum spurs” on their surfaces at the repeated scratches of the diamond-grinding wheel. Sometimes, the broken silicon carbide particles would also attach on the surfaces, as is shown in Figure 6.

Because SiCp/Al is going to be used for optical structural parts and electric packaging material, the requirements on the surface morphology is relatively high and the above-mentioned phenomenon should be avoided. The ideal surface morphology of the workpiece after grinding should be smooth, and the particles are removed plastically as is shown in Figure 7.

3.2 Evaluation of surface morphology after grinding

The optical structural parts should have high reflectivity to a certain waveband of electromagnetic wave, and the consistency in shape and size of the mirror should be ensured. The electronic packaging material should have a smooth surface, and the specific requirements include three-dimensional roughness and as small as possible broken area. Therefore, S_q, S_dr, and S_fd are three parameters used to evaluate the surface morphology after grinding. S_q is the root mean square deviation of the three-dimensional profiles to indicate the degree of deviation from the base plate. The deviation equation is Sq=1MN∑j=1N∑i=1Mz2(xi, yi), where z(x_i, y_i) is the residual surface and M and N are, respectively, the discrete sampling numbers in x and y direction. S_dr is the ratio of broken area, i.e., the increment of surface area in the sampling area. Assume the expanded area is A, then Sdr=A-(M-1)(N-1)ΔxΔy(M-1)(N-1)ΔxΔy, where M and N are, respectively, the discrete sampling numbers in x and y direction. ΔxΔy is the area element of the sampling area. S_fd is the fractal dimension, comprehensively reflecting the irregular shape of surfaces and the ability to fill in the gaps. Studies [18], [19], [20], [21], [22] show that fractal dimension is suitable for evaluating the surfaces of composites. The larger the fractal dimension is, the more details the curved lines have and the richer the characteristics are.

3.3 Influences of the grinding parameters on surface morphology

To analyze the influences of different grinding parameters on the surface morphology of SiCp/Al, three tests were carried out for each set of parameters to compare the root mean square and the average value of the broken area.

As is shown in Figure 8, the root mean square increases as the grinding penetrates deeper, which indicates that the surface becomes rougher. At the same time, the root mean square of surfaces after traditional grinding is larger than that after ultrasonic grinding. The surface is relatively rough after traditional grinding and it is smooth after ultrasonic grinding, as is shown in Figure 9.

Figure 8:

Grinding S_q: traditional vs. ultrasonic.

Figure 9:

Surfaces after traditional and ultrasonic grinding.

Judging from Figure 10, the broken area becomes larger as the grinding depth increases. It is smaller after ultrasonic grinding than after traditional grinding. It also explains the superiority of ultrasonic assistance during grinding, as is shown in Figure 11.

Figure 10:

Broken areas: traditional vs. ultrasonic.

Figure 11:

Broken area comparison after traditional and ultrasonic grinding.

As is shown in Figure 12, the fractal dimension grows larger as the grinding depth increases during traditional grinding. That is because there are more broken particles. It is the opposite in ultrasonic grinding. The tiny broken particles are caused by ultrasonic vibration. As the grinding depth increases, it will gradually offset the influences of ultrasonic vibration and weaken the ultrasonic effects. Therefore, the fractal dimension is smaller. The broken particles are relatively bigger in size in traditional grinding, and the fractal dimension of the surfaces is greater than that in ultrasonic grinding.

Figure 12:

Surface S_fd: traditional vs. ultrasonic.