Laser wafer dicing process optimization using the Taguchi approach

Abstract

This study uses an infrared cutting laser system to dice GaAs wafers. The Taguchi method determines the laser wafer dicing parameters [laser power (W), depth of focus (μm), and platform speed (m/s)] in relation to the dicing yield and single-grain compressive strength. An orthogonal array (L9 33), a signal-to-noise ratio, and an analysis of variance are employed to study the effects of these parameters. The result shows that the laser wafer dicing yield ranges from 85.53% to 95.20%, with laser power accounting for 86.08% of the total variation, making it the most significant factor. Additionally, the wafer dicing yield is positively correlated with grain compressive strength. Grain defects and corner chipping produced during the cutting process will reduce the compressive strength of the grains. Under the optimized laser wafer cutting conditions, the dicing yield and grain compressive strength reached 95.87% and 331.93 N/mm2, respectively.