Parametric analysis and a predictive model for color difference during laser-induced coloration on titanium

Abstract

Laser-induced coloration on a metallic surface has been of interest to many application arweas. However laser machining of metals involves many complex problems including nonlinear unstable coupled with multiple factors. Therefore there are still some significant challenges in the precise control of color creation. Here we explored the process of the laser-induced coloration and find the connection between surface colors and processing parameters. The Response Surface Methodology (RSM) based experimental design was adopted to explore the influence of the single processing parameter and the interaction between parameters on color changes of titanium. The results showed that the scanning speed laser power repetition rate and hatch distance had significant effects on color changes of titanium. Then we demonstrated that using artificial neural network (ANN) is an effective solution of nonlinear problems in laser-induced coloration which can match the processing parameters and the L*a*b* color values on titanium surface precisely with limited experiments. Finally we successfully used the processing parameters estimated by ANN model to create unique art painting on titanium with nanosecond pulsed laser. This work can provide a potential method to solve the problem in the color consistency and open a new perspective in industrial application of laser-induced coloration technology.