ABSTRACT
Manufacturing polypropylene (PP) composites to meet customers' needs is difficult, time-consuming, and costly, owing to the ever-increasing diversity and complexity of the corresponding specifications and the trial-and-error method currently used to satisfy the required physical properties. To address this issue, we developed three models for predicting the physical properties of PP composites using three machine learning (ML) methods: multiple linear regression (MLR), deep neural network (DNN), and random forest (RF). Further, the industrial data of 811 recipes were acquired to verify the developed models. Data categorization was performed to account for the differences between data and the fact that different recipes require different materials. The three models were then deployed to predict the flexural strength (FS), melting index (MI), and tensile strength (TS) of the PP composites in nine case studies. The predictive performance results differed according to the physical properties of the composites. The FS and MI prediction models with MLR exhibited the highest R2 values of 0.9291 and 0.9406. The TS model with DNN exhibited the highest R2 value of 0.9587. The proposed models and study findings are useful for predicting the physical properties of PP composites for recipes and the development of new recipes with specific physical properties.
ABSTRACT
Commercial humic acids mainly obtained from leonardite are in increasing demand in agronomy, and their market size is growing rapidly because these materials act as soil conditioners and direct stimulators of plant growth and development. In nature, fungus-driven nonspecific oxidations are believed to be a key to catabolizing recalcitrant plant lignins, resulting in lignin humification. Here we demonstrated the effective transformation of technical lignins derived from the Kraft processing of woody biomass into humic-like plant fertilizers through one-pot Fenton oxidations (i.e., artificially accelerated fungus reactions). The lignin variants resulting from the Fenton reaction, and manufactured using a few different ratios of FeSO4 to H2O2, successfully accelerated the germination of Arabidopsis thaliana seeds and increased the tolerance of this plant to NaCl-induced abiotic stress; moreover, the extent of the stimulation of the growth of this plant by these manufactured lignin variants was comparable or superior to that induced by commercial humic acids. The results of high-resolution (15 T) Fourier transform-ion cyclotron resonance mass spectrometry, electrostatic force microscopy, Fourier transform-infrared spectroscopy, and elemental analyses strongly indicated that oxygen-based functional groups were incorporated into the lignins. Moreover, analyses of the total phenolic contents of the lignins and their sedimentation kinetics in water media together with scanning electron microscopy- and Brunauer-Emmett-Teller-based surface characterizations further suggested that polymer fragmentation followed by modification of the phenolic groups on the lignin surfaces was crucial for the humic-like activity of the lignins. A high similarity between the lignin variants and commercial humic acids also resulted from autonomous deposition of iron species into lignin particles during the Fenton oxidation, although their short-term effects of plant stimulations were maintained whether the iron species were present or absent. Finally, we showed that lignins produced from an industrial-scale acid-induced hydrolysis of wood chips were transformed with the similar enhancements of the plant effects, indicating that our fungus-mimicking processes could be a universal way for achieving effective lignin humification.
