Modeling and experimental investigation of parameter regularity and morphological control in CO ₂ laser machining PMMA microchannels

Abstract This study presents a high-efficiency CO 2 laser micro-machining strategy for fabricating precision microfluidic architectures in polymethyl-methacrylate (PMMA) substrates, with a focus on establishing a predictive model for microchannel geometry. Through the systematic integration of finite element analysis (FEA) and experimental validation, the work investigates the thermomechanical interactions and geometric characteristics during CO 2 laser machining of PMMA plates. The computational model is developed to simulate the evolution of the transient temperature field and predict distinct temperature regimes for single-pass and double-pass moving processes, reaching approximately 300 °C and 350 °C, respectively. Furthermore, the double-pass moving process exhibits a noticeable temperature superposition effect due to residual heat accumulation. Experimentally, the influence of laser power ranging from 8 to 13 W and moving speed varying from 5 to 30 mm/s on microchannel dimensions is systematically examined across 36 parameter sets. Each processing condition is replicated three times, with dimensional measurements taken at three distinct locations per microchannel, and the resulting standard deviations remain below 5 % of the mean values, confirming good reproducibility under consistent processing conditions. The results show microchannel width ranging from 150 ± 5 µm to 477 ± 15 µm and depth from 102 ± 5 µm to 3,266 ± 20 µm. Systematic analysis of the experimental cases reveals consistent trends whereby microchannel depth and width exhibit positive correlations with increasing laser power under constant moving speeds, while demonstrating inverse relationships with elevated moving velocities under fixed laser power conditions. The FEA simulations show strong agreement with experimental measurements, with microchannel width and depth prediction errors of less than 18 %. The key novelty of this work lies in the development of a high-fidelity, computationally efficient model that accurately predicts microchannel morphology by incorporating Gaussian energy distribution and thermal accumulation effects, validated across a wide range of practical processing parameters. This synergistic experimental-numerical approach provides a robust parametric optimization platform for laser-based microfabrication, significantly advancing the development of microfluidic systems through intelligent process design.