Fig. 6: Virtual Screening of High-Performance VII. | npj Computational Materials

Fig. 6: Virtual Screening of High-Performance VII.

From: Exploring high-performance viscosity index improver polymers via high-throughput molecular dynamics and explainable AI

Fig. 6

a Comparison of polymer PVI distributions among different ML models (XGBoost, KRR, MLP, RF) using datasets PoLyInfo, PI1M, Polyimides, and the VIIInfo dataset. Screening criteria: synthetic accessibility score (SA) < 3.2, proportional viscosity index (PVI) > 480, and thickening efficiency (TE) < 30. Comparative analysis of selected polymers for b viscosity at 313 K, c viscosity at 373 K, d PVI, e TE, and f SA. g Typical molecular structures of the selected high-performance polymers. PIB, PMA, OCP, HSD, and MSC are viscosity index improver polymers, and their structural formulas are shown in Fig. S2.

Back to article page