Machine Learning—Guided Design of Biomass-Based Porous Carbon for Aqueous Symmetric Supercapacitors

Manickam Minakshi; Apsana Sharma; Ferdous Sohel; Almantas Pivrikas; Pragati A. Shinde; Katsuhiko Ariga; Lok Kumar Shrestha

doi:10.1002/cplu.202500342

Back

Machine Learning—Guided Design of Biomass-Based Porous Carbon for Aqueous Symmetric Supercapacitors

Journal article

Open access

Peer reviewed

Machine Learning—Guided Design of Biomass-Based Porous Carbon for Aqueous Symmetric Supercapacitors

Manickam Minakshi, Apsana Sharma, Ferdous Sohel, Almantas Pivrikas, Pragati A. Shinde, Katsuhiko Ariga and Lok Kumar Shrestha

ChemPlusChem (Weinheim, Germany), Vol.90(10), e202500342

2025

DOI: https://doi.org/10.1002/cplu.202500342

PMID: 40878512

Appears in Open Access via Read & Publish Agreements

Files and links (1)

pdf

Published3.53 MBDownload View

CC BY-NC-ND V4.0, Open Access

Abstract

biomass

carbon

dopant

energy

machine learning

storage

[Display Omitted] Biomass-derived porous carbon electrodes have attracted significant attention for high-performance supercapacitor applications due to their sustainability, cost-effectiveness, and tunable porosity. To accelerate the design and evaluation of these materials, it is essential to develop accurate and efficient strategies for optimizing their physicochemical and electrochemical properties. Herein, a machine learning (ML) approach is employed to analyze experimental data from previously reported sources, enabling the prediction of specific capacitance (F g−1) based on various material characteristics and processing conditions. The trained ML model evaluates the influence of factors such as biomass type, electrolyte, activating agent, and key synthesis parameters, including activation and carbonization temperatures and durations, on supercapacitor performance. Despite growing interest, comprehensive studies that correlate these variables with performance metrics remain limited. This work addresses this gap by using ML algorithms to uncover the interrelationships between biomass-derived carbon properties, synthesis conditions, and specific capacitance. Herein, it is demonstrated that an optimal combination of a carbonized honeydew peel to H3PO4 ratio of 1:4 and an activation temperature of 500 °C yields a highly porous carbon material. When used in a symmetric device with 1 M H2SO4 electrolyte, this material, rich in oxygen and phosphorus species, achieves a high specific capacitance of 611 F g−1 at a current density of 1.3 A g−1. Correlation analysis reveals a strong synergy between surface area and pore volume (correlation coefficient = 0.8473), and the ML-predicted capacitance closely aligns with experimental results. This ML-assisted framework offers valuable insights into the critical physicochemical and electrochemical parameters that govern supercapacitor performance, providing a powerful tool for the rational design of next-generation energy storage materials.

Details

Title: Machine Learning—Guided Design of Biomass-Based Porous Carbon for Aqueous Symmetric Supercapacitors
Authors/Creators: Manickam Minakshi - Murdoch University, Centre for Water, Energy and Waste
Apsana Sharma - College of Science, Technology, Engineering & Mathematics, Murdoch University, Murdoch, WA, Australia
Ferdous Sohel - Murdoch University, Centre for Crop and Food Innovation
Almantas Pivrikas - Murdoch University, School of Mathematics, Statistics, Chemistry and Physics
Pragati A. Shinde - National Institute for Materials Science
Katsuhiko Ariga - National Institute for Materials Science
Lok Kumar Shrestha - National Institute for Materials Science
Publication Details: ChemPlusChem (Weinheim, Germany), Vol.90(10), e202500342
Publisher: Wiley-VCH GmbH
Number of pages: 13
Grant note: JP20H00392; JP23H05459 / Japan Society for the Promotion of Science (http://data.elsevier.com/vocabulary/SciValFunders/501100001691) Japan Society for the Promotion of Science (http://data.elsevier.com/vocabulary/SciValFunders/501100001691)
Identifiers: 991005810349707891
Murdoch Affiliation: Centre for Crop and Food Innovation; School of Mathematics, Statistics, Chemistry and Physics; School of Information Technology; Centre for Water, Energy and Waste; College of Science, Technology, Engineering and Mathematics
Language: English
Resource Type: Journal article

Metrics

2 File views/ downloads

18 Record Views

5 Times Cited - Web of Science

InCites Highlights

These are selected metrics from InCites Benchmarking & Analytics tool, related to this output

Collaboration types: Domestic collaboration; International collaboration
Citation topics: 2 Chemistry; 2.62 Electrochemistry; 2.62.52 Electrode Materials
Web Of Science research areas: Chemistry, Multidisciplinary
ESI research areas: Chemistry