A recent study published in Energy demonstrates the use of machine learning (ML) to optimize the design and prediction of porous properties in biomass-derived biochar. Biochar, a carbon-rich material derived from biomassBiomass is a complex biological organic or non-organic solid product derived from living or recently living organism and available naturally. Various types of wastes such as animal manure, waste paper, sludge and many industrial wastes are also treated as biomass because like natural biomass these More, is widely used in energy, agriculture, and environmental applications due to its customizable porosityPorosity of biochar is a key factor in its effectiveness as a soil amendment and its ability to retain water and nutrients. Biochar’s porosity is influenced by feedstock type and pyrolysis temperature, and it plays a crucial role in microbial activity and overall soil health. Biochar More and surface properties.

The researchers developed ML models to predict key biochar porosity properties, including specific surface area (SSA), total pore volume (Total_PV), micropore volume (Micro_PV), mesopore volume (Meso_PV), and average pore size (Average_PS). These models achieved high accuracy, with R² values ranging from 0.76 to 0.91 for individual properties and an average of 0.87 for a multi-target model. The analysis highlighted that the type and ratio of activation agents were critical factors influencing porosity.

Using ML, the team optimized pyrolysisPyrolysis is a thermochemical process that converts waste biomass into bio-char, bio-oil, and pyro-gas. It offers significant advantages in waste valorization, turning low-value materials into economically valuable resources. Its versatility allows for tailored products based on operational conditions, presenting itself as a cost-effective and efficient More and activation processes, enabling the production of high-porosity biochar. For example, peanut shell biochar achieved an SSA of approximately 1800 m²/g and a pore volume of ~1 cm³/g, validated with experimental R² of 0.98. This represents a significant improvement over traditional methods that often yield lower porosity.

The study underscores the potential of ML in engineering biochar with tailored properties for specific applications, advancing its performance in areas like gas adsorption, water purification, and catalysis. By integrating diverse activation agents and treatments, this approach paves the way for smarter biochar production strategies.