Comparing picosecond to femtosecond laser pulse duration in laser cleaning processes with a focus on diamond-like carbon (DLC) coating removal

Laser de-coating is an increasingly promising technology for selective coating removal that eliminates the need for chemical cleaning. Understanding the removal mechanism and optimising laser parameters is critical to achieving maximum performance and substrate integrity. This paper presents a comparative study of the mechanism of diamond-like carbon (DLC) coating removal when using a femtosecond laser compared to a picosecond laser. Two different lasers of pulse durations of 86 fs and 150 ps and 1030 nm and 1060 nm wavelength respectively were compared in the study. The laser de-coating process of a DLC coating focused on the optimisation of key parameters and their effects on surface integrity. Fractional factorial experiments were conducted using a Taguchi L9 array and analysis of variance (ANOVA) was used to investigate systematically the effects of pulse energy, repetition rate, scanning speed and beam overlap on cleaning efficiency (quantified by residual coating elemental composition), surface roughness, residual stress, atomic elemental composition and nano-hardness. The findings indicate that femtosecond lasers reduced thermal effects. Their use resulted in reduced surface roughness, lower compressive residual stress relief, minimal atomic residual coating and nano-hardness retention. Quantitatively, femtosecond processing achieved up to 70% carbon reduction (from 25% to 7.45%), 74% chromium reduction (from 65% to 17.1%) and 95% nitrogen reduction (from 40% to 1.89%) by the final pass, demonstrating high coating-removal completeness. In contrast, picosecond lasers, while effectively removing material, demonstrated moderate thermal diffusion and occasional material redeposition. The surface cleaned by femtosecond laser exhibited a surface morphology characterised by a powder-like or recrystallised appearance, whereas the picosecond laser treatment produced a densely packed, low aspect ratio and worm-like structure. Beyond a direct performance comparison, this study introduces a normalised optimisation framework (fluence relative to ablation threshold, overlap, scan strategy and pass sequencing) to determine when a cost-preferred picosecond process can be driven result in femtosecond-like levels of surface integrity. Using the optimised parameters, picosecond de-coating approaches femtosecond outcomes in key integrity measures, including final-pass surface roughness and residual stress convergence, while highlighting where further optimisation is needed to mitigate thermal softening and material redeposition. These results provide transferable design rules for selecting and tuning ultrashort-pulse parameters for industrial DLC removal. This study enhances the understanding of laser-material interactions and contributes to the development of sustainable, high-precision laser cleaning technologies, ultimately enhancing process efficiency and component longevity.