Enhanced phytoremediation of vanadium using coffee grounds and fast-growing plants: Integrating machine learning for...

Area of Science:

• Environmental Science
• Bioremediation
• Plant Science

Background:

• Vanadium (V) contamination presents environmental challenges.
• Phytoremediation offers a sustainable approach but is often limited by plant growth rates.
• Integrating coffee grounds with fast-growing plants can enhance phytoremediation efficiency.

Purpose of the Study:

• To investigate the enhancement of vanadium (V) phytoremediation by incorporating coffee grounds with fast-growing plants.
• To evaluate the impact of coffee grounds on plant stress indicators and microbial communities.
• To develop and validate machine learning models for predicting and optimizing V phytoremediation.

Main Methods:

• Utilized barley grass, wheat grass, and ryegrass for V phytoremediation experiments.
• Incorporated coffee grounds to assess their effect on V removal and plant stress.
• Employed Gradient Boosting and XGBoost models for prediction and optimization, evaluated using MSE and R².
• Analyzed changes in plant root enzyme activities (CAT, POD, SOD) and microbial community abundance.

Main Results:

• Ryegrass achieved 48.7% V⁵⁺ removal in 6 days.
• Wheat grass with coffee grounds showed increased V⁵⁺ removal from 30.51% to 62.91%.
• Machine learning models demonstrated high predictive accuracy (R² = 0.95, MSE = 1.20).
• Coffee grounds reduced oxidative stress markers in ryegrass roots and altered microbial communities.

Conclusions:

• Coffee grounds significantly enhance V phytoremediation efficiency in fast-growing plants.
• Machine learning provides a powerful tool for optimizing phytoremediation processes.
• This integrated approach offers a novel and sustainable solution for vanadium-contaminated environments.