The rapid development of flexible and wearable electronics creates a pressing demand for high-precision, efficient, and environmentally friendly microfabrication techniques for polymer substrates. Polyimide (PI) film is an ideal base material for flexible printed circuit boards (FPCBs) due to its exceptional thermal stability, mechanical strength, and dielectric properties. Laser direct writing, particularly using CO₂ lasers, has emerged as a promising non-contact method. However, controlling the ablation quality, especially minimizing the heat-affected zone (HAZ) and achieving desirable groove morphology for subsequent circuit embedding, remains a significant challenge. This study aims to systematically optimize the CO₂ continuous laser ablation process for PI films. The core objective is to establish a quantitative relationship between key laser parameters and processing quality, thereby identifying an optimal parameter set to achieve clean, dimensionally controlled grooves suitable for flexible circuit fabrication.

A commercial CO₂ continuous laser engraving system with a maximum power of 80 W and a wavelength of 10.6 μm was employed. The material was 0.25 mm thick bare PI film. The research methodology comprised several key steps. First, the optimal focal position (stand-off distance) was determined by conducting single-point ablation tests. The laser nozzle-to-film distance varied from 2 mm to 10 mm. The diameters of the ablated pits were measured using an ultra-depth-of-field microscope, and the corresponding laser energy density was calculated (Fig.2, Fig.3, Table 3). This data, combined with subsequent line width measurements at different distances, pinpointed the focal point for minimal spot size and maximum energy density (Fig.5, Fig.8). Subsequently, a single-factor experimental approach was adopted to investigate the effects of laser power (6.4 W, 8.0 W, 9.6 W, 11.2 W, 12.8 W), scanning speed (75 mm/s to 175 mm/s), and number of scans (1 and 2) on the ablation quality. For each parameter combination, straight lines were ablated on the PI film. The primary response variable, the line width, was obtained by taking the average of three measurements per line using the ultra-depth-of-field microscope (Fig.6, Fig.7). The surface morphology, including the presence of carbonized graphene-like residues and HAZ, was simultaneously observed and analyzed to comprehensively assess processing quality (Fig.11, Fig. 12, Fig.14).

The point ablation tests revealed that the calculated energy density first increased and then decreased with increasing nozzle distance, peaking at 334.46 J/mm² at a distance of 6 mm (Fig.5). Correspondingly, the ablated line width reached a minimum at this same distance (Fig.8), confirming it as the optimal focal position (0 mm defocus) for achieving the finest feature size and highest energy utilization. The single-factor experiments on power and speed demonstrated their dominant influence on line width. Line width exhibited a general increasing trend with higher laser power and a decreasing trend with faster scanning speeds (Fig.9, Fig.10). Notably, within the power range of 8.0 W to 9.6 W, the increase in width was less pronounced. Microscopic observation attributed this to the formation and strong adhesion of laser-induced carbonaceous material (graphene) within the groove, which partially hindered direct ablation of the underlying PI. At lower speeds (e.g., 75 mm/s) or higher powers, this material was more easily removed by the auxiliary gas, resulting in wider and cleaner grooves. However, excessive power (11.2 W) led to a large and rough HAZ, and a power of 12.8 W caused complete penetration. At very high speeds (e.g., 175 mm/s), energy input was insufficient for consistent carbonization. The optimal surface quality, characterized by a clean, yellow-base groove free of debris, was achieved at 9.6 W and 75 mm/s, producing a line width of 338.14 μm (Fig.12). In contrast, other parameter sets resulted in either residual carbon deposits or excessive thermal damage. Experiments involving two scan passes showed only a marginal increase in line width compared to the change induced by varying power or speed (Fig.13). The primary effect of a second scan was to deepen the groove or alter the state of the carbonized layer, sometimes leading to poorer surface morphology or complete penetration if the energy was already high (Fig.14). This confirmed that laser power and scanning speed were the principal factors governing line width. A direct comparison between the optimized parameter set (9.6 W, 75 mm/s, 1 scan, 6 mm focal distance) and a non-optimized set validated the findings. The groove produced with optimized parameters was clean and well-defined, whereas the other showed debris or over-ablation (Fig.15).

This study successfully establishes and validates a process optimization framework for CO₂ continuous laser ablation of PI films. The main conclusions are as follows: the optimal focal distance for the specific experimental setup is identified as 6 mm between the laser nozzle and the PI film, which delivers the highest energy density and enables the smallest achievable line width. A quantitative model linking laser parameters to processing outcomes is developed. Laser power and scanning speed are the dominant factors controlling ablation line width and surface quality, primarily through their effect on the energy density delivered per unit area and the behavior of the laser-induced carbonization layer. The number of scans has a negligible effect on line width but can modify groove depth and surface morphology; it is not a primary controlling factor for width. The identified optimal parameter combination (9.6 W laser power, 75 mm/s scanning speed, single scan) produces a clean ablation groove with a width of 338.14 μm, which is suitable for subsequent processes such as conductive material embedding in flexible circuit board fabrication. This work provides a practical guideline for achieving high-quality, efficient laser processing of PI films for flexible electronics.