Microscale studying and machine learning of polymer wetting behavior in direct ink writing under different gravitational intensity

Understanding polymer wetting under varying gravity is crucial for advancing in-space additive manufacturing, particularly in extrusion-based methods such as direct ink writing (DIW). In microgravity, surface tension dominates fluid behavior, making wetting dynamics highly sensitive to polymer properties such as chain length, concentration, and viscosity. This study employs a multiscale framework using coarse-grained molecular dynamics (CGMD) to explore how these factors, along with the bond number (representing gravity), affect the wetting of polyethylene glycol solutions on iron substrates. Simulations show that higher bond numbers improve lateral chain alignment and conformational relaxation, enhancing wetting. To generalize predictions, a dataset of 200 simulations were used to train machine learning models, with XGBoost achieving the highest accuracy. A data augmentation strategy mitigated the small sample size issue. Shapley additive explanation analysis identified bond number as the most influential factor. This integrated CGMD–machine learning approach provides both physical insight and predictive capability, guiding the formulation of polymer inks and printing parameters for adaptive DIW under variable gravity conditions.