Zhang, et al (2024) Prediction of CO₂adsorption of biocharBiochar is a carbon-rich material created from biomass decomposition in low-oxygen conditions. It has important applications in environmental remediation, soil improvement, agriculture, carbon sequestration, energy storage, and sustainable materials, promoting efficiency and reducing waste in various contexts while addressing climate change challenges. More under KOH activation via machine learning. Carbon Capture Science & Technology. https://doi.org/10.1016/j.ccst.2024.100309

Carbon dioxide (CO2) capture is essential for combating climate change. Biochar, derived from organic materials, has shown potential in CO2 adsorption due to its porous structure and modifiable surface chemistry. To maximize biochar’s CO2 adsorption capabilities, researchers often enhance its physical properties through chemical activation, especially with potassium hydroxide (KOH). A recent study delves into this enhancement process and employs machine learning (ML) to predict the adsorption performance of KOH-activated biochar. This approach promises a more efficient way to optimize biochar production for carbon capture applications.

The Role of KOH Activation in Biochar

Biochar’s natural structure often lacks sufficient pore development for optimal CO2 capture. Chemical activation, particularly with KOH, is widely used to improve this structure. KOH facilitates the formation of micropores and small mesopores within biochar, increasing the surface area available for CO2 adsorption. Additionally, nitrogen-containing groups introduced during the activation process further enhance CO2 adsorption by improving the material’s chemical affinity for CO2 molecules.

This study specifically focuses on biochar activated through a “two-step” KOH treatment. The researchers collected data from 329 samples of biochar subjected to this method, including information on pore structure, elemental composition, and activation conditions.

Machine Learning Models for Prediction

The complex relationship between biochar’s structural properties and its CO2 adsorption capacity poses a challenge for optimization. Machine learning provides a solution by analyzing large datasets and predicting outcomes based on multiple interacting variables. The study tested three ML models: ridge regression (RR), random forest (RF), and multi-layer perceptron (MLP). These models analyzed the effects of various factors, such as activation temperature, pore size, and nitrogen content, on CO2 adsorption.

Among these models, the MLP model performed best, with an accuracy of 96.1% in predicting CO2 adsorption capacity. This outperforms other models used in previous studies, indicating the effectiveness of MLP in handling the complexities of biochar’s CO2 capture potential.

Key Findings

The study’s results highlight several important factors influencing the CO2 adsorption capacity of KOH-activated biochar:

1. Pore Structure: Micropores (pores smaller than 2 nm) are particularly crucial for CO2 adsorption. Increasing micropore volume enhances adsorption, while larger pores (mesopores and macropores) are less effective.
2. Nitrogen Content: The introduction of nitrogen into biochar, particularly in the form of nitrogen-containing groups like pyrrole and graphite nitrogen, significantly improves CO2 capture, especially at low temperatures.
3. Activation Conditions: The optimal KOH-to-biochar ratio for maximizing CO2 adsorption is between 1:1 and 2:1. Additionally, the best activation temperature is around 700°C. Exceeding this temperature can degrade the biochar’s structure, reducing its adsorption capacity.
4. Adsorption Conditions: Lower adsorption temperatures (<20°C) enhance CO2 capture, as CO2 molecules are more easily retained in micropores. Higher pressures also increase adsorption efficiency, though there are diminishing returns at extremely high pressures.

By applying machine learning to predict CO2 adsorption performance, this study provides valuable insights for improving biochar’s effectiveness in carbon capture. The findings suggest that carefully controlling activation conditions, such as KOH ratio and temperature, alongside optimizing pore structure and nitrogen content, can significantly enhance biochar’s ability to adsorb CO2. This research paves the way for more efficient and cost-effective biochar production for environmental applications, including large-scale carbon capture and sequestration efforts.

In the future, further exploration of biochar’s long-term stability and cost-effectiveness in industrial applications will be critical to fully realizing its potential as a sustainable solution for CO2 reduction.