Henna plant biomass enhanced azo dye removal: Operating performance, microbial community and machine learning
Shilin Wen1, Jingang Huang2, Weishuai Li1
1College of Materials and Environmental Engineering, Hangzhou Dianzi University, Hangzhou, 310018, PR China.
Chemosphere
|February 19, 2024
Summary
Henna plant biomass effectively promotes azo dye bio-reduction by providing electron donors and mediators. Machine learning, specifically XGBoost, accurately predicts dye removal efficiency, validating henna as a sustainable solution.
Area of Science:
- Environmental Biotechnology
- Applied Microbiology
- Machine Learning in Environmental Science
Background:
- Azo dye bio-reduction relies on electron donors and redox mediators.
- Natural biomass offers potential for sustainable dye remediation.
- Predictive modeling can optimize bioremediation processes.
Purpose of the Study:
- To investigate henna biomass for azo dye (Acid Orange 7) bio-reduction.
- To apply machine learning for understanding and predicting the removal process.
- To validate the efficacy of henna-assisted dye removal.
Main Methods:
- Supplementation of natural henna plant biomass for azo dye bio-reduction.
- Analysis of henna hydrolysis and fermentation products (VFA, lawsone).
- Application and validation of machine learning algorithms (XGBoost) for prediction.
Main Results:
- Henna biomass provided volatile fatty acids (VFA) and lawsone, facilitating AO7 bio-reduction.
- Specific bacterial enrichment (Firmicutes, Levilinea, Paludibacter) observed.
- XGBoost accurately predicted AO7 removal (91.6% efficiency, 3.95% error) under optimized conditions.
Conclusions:
- Henna biomass is a cost-effective and robust agent for azo dye bio-reduction.
- Machine learning models like XGBoost are effective for predicting bioremediation performance.
- Integrated approach of biomass addition and ML modeling enhances azo dye removal strategies.
