Combined Experimental and Machine Learning Study on the Interplay between Delignification and Mechanical Properties...

Area of Science:

• Materials Science
• Biotechnology
• Sustainable Engineering

Background:

• Structure-retaining delignification of wood is crucial for creating biobased composite scaffolds.
• Current methods often compromise sustainability and scalability.

Purpose of the Study:

• To develop a sustainable and scalable method for wood delignification.
• To improve the mechanical properties of wood-based scaffolds.
• To assess the environmental impact and predict material performance.

Main Methods:

• Reconstruction of poplar wood using room-temperature partial delignification, followed by densification.
• Mechanical property testing at various fiber directions.
• Life Cycle Assessment (LCA) to evaluate environmental impacts.
• Development of an 'unsupervised, supervised classification, supervised regression' (USS) learning framework for property prediction.

Main Results:

• Partial delignification at room temperature yielded superior mechanical properties at 45° and 90° fiber directions compared to full delignification.
• The process facilitated sample up-scaling and allowed multiple reuses of the delignification solution without quality loss.
• Repeated reuse of the delignification solution significantly reduced global warming potential (GWP) and ecosystem quality (EQ) impacts, as per LCA.
• The USS learning framework accurately predicted mechanical properties and identified key influencing parameters.

Conclusions:

• Room-temperature partial delignification offers a sustainable and scalable route to high-performance wood-based scaffolds.
• This method enhances mechanical properties and minimizes environmental footprint.
• Predictive modeling provides valuable insights into material performance and fabrication parameters.