Large-scale multi-particle cold spray simulation framework for deposit morphology and deformation analysis

Cold spray (CS) is a solid-state particle deposition technology extensively utilized for coating and more recently for additive manufacturing applications. While detailed numerical modeling of CS deposition offers substantial benefits for process optimization, existing methods are hindered by significant computational demands. As a result, simulations are typically confined to a small number of powder particles, which does not adequately reflect the actual deposition dynamics and therefore limits the effectiveness of simulations for deposition assessment. In this study, a high-fidelity open-source multi-particle impact framework (Free2Spray) is developed to simulate large-scale (>10,000 particles) CS deposition at a considerably convenient computational cost. The model can accurately track material changes during deposition by splitting the process into sections, each with randomly placed particles. These sections are added step by step, carrying over all data from one to the next. Python scripts are developed to automate this workflow, requiring the user to specify a few key parameters. The framework is also equipped with our recently developed high-precision material model implemented via a user-defined hardening (VUHARD) subroutine to accurately capture the realistic particle deformation. The developed framework is proven to accurately predict the shape profiles and surface roughness of experimental single-layer single-track deposits under various nozzle scanning speeds and reproduces the experimental cross-sectional particle deformation and flattening ratios. Compared with the standard Eulerian schemes, the proposed framework reduces the computational time and memory usage by 34% and 35%, respectively. These results demonstrate that the proposed framework significantly improves computational efficiency while maintaining high accuracy, enabling personal computers to simulate the CS deposition process with ten million elements and providing a reliable tool for predicting deposit morphology and particle deformation behavior. The source codes are provided (https://github.com/xuanyge/Free2Spray.git) to enable interested researchers to utilize and implement them.