Analyzing and Predicting the Viscosity of Polymer Nanocomposites in the Conditions of Temperature, Shear Rate, and Nanoparticle Loading with Molecular Dynamics Simulations and Machine Learning

Predicting the viscosity (?I) of polymer nano-composites (PNCs) is of critical importance as it governs a dominant role in PNCs' processing and application. Machine-learning (ML) algorithms, enabled by pre-existing experimental and computational data, have emerged as robust tools for the prediction of quantitative relationships between feature parameters and various physical properties of materials. In this work, we employed nonequilibrium molecular dynamics (NEMD) simu-lation with ML models to systematically investigate the ?I of PNCs over a wide range of nanoparticle (NP) loadings (e), shear rates (gamma?), and temperatures (T). With the increase in gamma?, shear thinning takes place as the value of ?I decreases on the orders of magnitude. In addition, the e dependence and T dependence reduce to the extent that it is not visible at high gamma?. The value of ?I for PNCs is proportional to e and inversely proportional to T below the intermediate gamma?. Using the obtained NEMD results, four machine-learning models were trained to provide effective predictions for the ?I. The extreme gradient boosting (XGBoost) model yields the best accuracy in ?I prediction under complex conditions and is further used to evaluate feature importance. This quantitative structure-property relationship (QSPR) model used physical views to investigate the effect of process parameters, such as T, e, and gamma?, on the ?I of PNCs and paves the path for theoretically proposing reasonable parameters for successful processing.