ioChem-BD Browse module - collection page

We conducted Density Functional Theory (DFT) simulations on pristine carbon cavities decorated with heteroatoms in various carbon/heteroatom ratios. They are located in N_carbons, S_carbons, and P_carbons repositories. Subsequently, we synthesized Single Atom Catalysts in doped-carbons that are contained in M_N_SACs, M_S_SACs, and M_P_SACs repositories.

To evaluate the performance of machine learning algorithms in predicting the stability descriptor E ^DFT _ads, we generated interactive parity plots for the two high-performing algorithms: Random Forest Regression (RFR) and regression with Bayesian Machine Scientist (BMS). These plots, parity_plot_rfr_m_logocv.html and parity_plot_bms_m_logocv.html, respectively, were derived from the Metal-Leave-One-Out Cross Validation (M-LOGOCV) technique.

The interactive parity plots offer valuable visualizations, allowing a direct comparison between the machine learning predictions and the DFT data. They are especially beneficial for bimetallic systems, providing essential insights for the design and development of efficient catalysts across various applications.

Share - copy and redistribute the material in any medium or format. Adapt - remix, transform, and build upon the material for any purpose, even commercially. The licensor cannot revoke these freedoms as long as you follow the license terms.

Please use this identifier to cite or link to this collection: DOI: 10.19061/iochem-bd-1-329

This dataset derived results are published in:

Manuscript title: A generalized model for estimating adsorption energies of single atoms on doped carbon materials

Journal: J. Mater. Chem. A

DOI: 10.1039/D3TA05898K