Experimental Study On SiC Single Crystal Machining With Reciprocating And Rotating Electroplated Diamond Wire Saw

With the rapid development of the large-scale integrated circuits and micro-electronics, photonics technology, applications of semiconductor materials such as SiC single-crystal become very popular.The rapid development of technology not only requires the diameter of SiC single-crystal substrates is increasing, but the requirement of the wafer surface machining quality is getting higher and higher. SiC wafers cutting process is the first and most important step in the whole manufacturing process,cutting quality of its directly affects the efficiency of wafer machining of the follow-up processes and the final quality. Fixed abrasive diamond wire saw machining technology is becoming more and more popular, due to features such as narrow kerf, high removal rate, decent workpiece quality and clean machining environment, etc. Meanwhile it is expected to be an essential future topic of hard-brittle materials.Based on the cutting and grinding removal mechanism of hard-brittle materials and the basic principle of indentation fracture mechanics, the macro-and micro-removal mechanism of the hard-brittle materials was studied and analyzed in this paper. According to the relative motion relationship between wire saw and workpiece in cutting process and the distribution of diamond grits on the wire saw surface, cutting model of the diamond wire saw and force model of a single diamond grain were established, and cutting depth and cutting force formula of a single diamond abrasive grit were deduced. Based on this, cutting force formula of the wire saw was obtained.The experiment that reciprocating and rotating diamond wire saw cutting machine cuts SiC wafer was researched in orthogonal test method in this paper. The influence of saw wire speed, feed rate and rotating speed of workpiece in cutting process were analyzed. The optimized cutting parameters which can improve surface quality of slices and reduce tangential cutting force were obtained. The results showed that:the surface quality of slices which was cloud-shaped can be improved significantly due to the adding of the work-piece rotation; The workpiece feed rate had greater impact on the tangential cutting force and surface roughness than the saw wire speed and the workpiece rotation speed. An important way to obtain the smaller tangential sawing force and better surface quality of slices is considering of a reasonable match among saw wire speed, feed rate and rotating speed of workpiece comprehensively.Take use of BP network of the artificial neural network (ANN) technology in MATLAB to process the test data, a predicting model based on the influence of wire saw speed, feed rate and rotating speed of workpiece on wafer surface roughness was established. The results showed that:the model can predict the surface roughness better, the maximum simulate relative error and the largest predict error of it are 7.05 and 13.4 percent, it provides a effective method for surface quality prediction based on cutting parameters.