Related Experiment Video
Updated: Jun 6, 2025
10:18
Extraction of Lignin with High β-O-4 Content by Mild Ethanol Extraction and Its Effect on the Depolymerization Yield
Published on: January 7, 2019
20.4K
Enhancing Lignin-Carbohydrate Complexes Production and Properties With Machine Learning.
Daryna Diment1, Joakim Löfgren2, Marie Alopaeus3
1Department of Bioproducts and Biosystems, School of Chemical Engineering, Aalto University, Vuorimiehentie 1, Espoo, 02150, Finland.
Chemsuschem
|November 25, 2024
Summary
Machine learning optimized biorefining to produce high-yield lignin-carbohydrate complexes (LCCs). This novel approach significantly enhances LCC production, enabling high carbohydrate content for valuable applications.
Area of Science:
- Biorefining and Biomaterials Science
- Sustainable Chemistry and Engineering
Background:
- Lignin-carbohydrate complexes (LCCs) offer synergistic potential for high-value products but face challenges in high-yield production.
- Conventional methods like ball milling and enzymatic hydrolysis have limitations in LCC yield and targeted synthesis.
Purpose of the Study:
- To develop a novel, machine learning-guided approach for the targeted production of LCCs with high carbohydrate content and yield.
- To optimize the AquaSolv Omni (AqSO) biorefinery process for enhanced LCC synthesis.
Main Methods:
- Employed machine learning, specifically Bayesian Optimization, to tune biorefinery processing conditions (temperature, severity, liquid-to-solid ratio).
- Utilized Pareto front analysis to identify optimal trade-offs between LCC yield and carbohydrate content.
- Assessed LCC properties including glass transition temperature (Tg), surface tension, and antioxidant activity.
Main Results:
- Achieved high yields of LCCs (up to 15 wt%) with significant carbohydrate content (up to 60/100 Ar).
- Identified processing conditions yielding LCCs with 8-15 wt% yield and 10-40/100 Ar carbohydrate content.
- Found that high-carbohydrate LCCs generally exhibit low glass transition temperatures and surface tension.
Conclusions:
- The ML-guided biorefinery concept significantly improves LCC production yields compared to conventional methods.
- This approach enables the scalable production of LCCs with tailored properties for high-value applications.
- Optimized LCCs demonstrate potential for diverse applications based on their characterized physical and chemical properties.
