Leng, Lei, et al (2024) Machine-learning-aided prediction and engineering of nitrogen-containing functional groups of biochar derived from biomass pyrolysis. Chemical Engineering Journal. https://doi.org/10.1016/j.cej.2024.149862

In this study, the focus is on biochar, a carbonaceous material derived through biomass pyrolysis, with potential applications. The research utilizes machine learning (ML) to predict and engineer nitrogen-containing functional groups in biochar, specifically amine-N (N-A), pyrrolic-N (N-5), and pyridinic-N (N-6). The single-target random forest model demonstrates accuracy, achieving a test R2 of 0.91–0.97 for predicting N recovery and functional group contents. The study reveals that pyrolysis temperature is a dominant factor influencing these characteristics.

A multi-target random forest model with an average test R2 of 0.93 is employed to optimize pyrolysis parameters for designing biochar N-functional groups. Experimental verification follows, demonstrating relative errors within 15%. This integration of ML in biochar engineering underscores its potential for efficient material optimization, overcoming the limitations of traditional trial-and-error methods. The findings emphasize the significance of tailored N-functional groups in biochar for enhanced performance across various applications.



Leave a Reply

Trending

Discover more from Biochar Today

Subscribe now to keep reading and get access to the full archive.

Continue reading