Experimental investigation and optimization of femtosecond laser processing parameters of silicon carbide–based on response surface methodology

Abstract

The femtosecond laser, with its extremely high peak power and ultrashort pulse duration, is widely used in silicon carbide (SiC) processing. Herein, the response surface method was used to develop four quadratic regression models to predict the groove depth, width, heat affected zone (HAZ), and material removal rate (MRR). The effects of each parameter were investigated using variance analysis. In addition, the interaction of processing parameters with groove depth, width, HAZ and MRR was investigated, which provided an important analysis for the precise selection of processing parameters. The laser fluence is shown to significantly impact the groove depth, width, HAZ, and MRR. The scan speed significantly influences the groove depth, width, and HAZ but not the MRR. The repetition rate and multipass scanning significantly influence the groove depth, HAZ, and MRR but only moderately influence the groove width. The models show that changing the energy received by the groove can change the groove depth, width, HAZ, and MRR. The groove depth, width, and HAZ increase as the laser fluence, repetition rate, and multipass scanning increase; and they decrease as the scan speed increases. The MRR increases as laser fluence and multipass scanning increase. Meanwhile, the MRR varies slightly with repetition rate and scan speed. A nonlinear functional relationship exists in the interaction process of processing parameters. The processing parameters were optimized using response surface methodology with the maximum MRR and minimum HAZ as the objective function. Within the range of the processing parameters used, the average relative error between the experimental and predicted values was less than 5%. Consequently, the established regression models can accurately predict the groove depth, width, HAZ, and MRR in SiC femtosecond laser processing.