Machine learning-driven optimization for sustainable CO2-to-methanol conversion through catalytic hydrogenation
Bibliographic record
Abstract
Growing concerns about greenhouse gas emissions have accelerated research into converting CO 2 into valuable products like methanol. Catalytic hydrogenation, utilizing a catalyst in a thermochemical process, offers a promising solution for reducing atmospheric CO 2 and combating climate change. However, optimizing operating conditions and selecting suitable catalysts for CO 2 to methanol conversion remains challenging due to the complex interplay between catalyst properties and reaction performance. This research leveraged machine learning (ML) to model CO 2 to methanol conversion using a comprehensive experimental database. ML models were developed to predict CO 2 conversion efficiency, methanol selectivity, and CO selectivity, facilitating process optimization, techno-economic analysis, and life cycle assessment (LCA). The gradient boosting regression model emerged as the most accurate, with coefficients of determination (R 2 > 0.86) and low error metrics (RMSE < 9.99, MAE < 5.99). De novo predictions demonstrated an acceptable linear relationship with the completely unseen dataset. Feature importance analysis identified temperature and gas hourly space velocity (GHSV) as the most significant descriptors. The optimal conditions for maximum CO 2 conversion efficiency and methanol selectivity were identified as temperatures between 330 and 370 °C, a pressure of 50 bar, and a GHSV of 6,500–14,000 mL/g.h. The techno-economic analysis highlighted H 2 purchase price, methanol selling price, and CO 2 feedstock costs as critical economic factors, with a payback period of 4.6 years. The LCA demonstrated a 270 % reduction in carbon emissions through catalytic hydrogenation of CO 2 to methanol. This study underscored the importance of using sustainable H 2 and electricity sources to enhance the economic and environmental benefits of the process.
Fetched live from OpenAlex and de-inverted. Abstracts are not stored in this database: the inverted indexes are 8.6 GB of the frame’s 9.3 GB of text, and the host has 13 GB free.
How this classification was reachedexpand
Full frame machine prediction
Teacher imitationNot calibrated prevalence, not ground truth. Human validation pending. The Gemma side is a direct model label for every work in the frame, read from the title-only record. The Codex side is a classifier learned from the 10,348 direct Codex labels and calibrated to design-weighted sample rates; fields without enough sample support carry no Codex call. Candidate is the union of the two sides; consensus is their intersection. These outputs are machine_predicted_unvalidated and are not human labels.
Distilled classifier scores by category (both heads)
| Category | Codex | Gemma |
|---|---|---|
| Metaresearch | 0.001 | 0.002 |
| Meta-epidemiology (narrow) | 0.001 | 0.001 |
| Meta-epidemiology (broad) | 0.001 | 0.001 |
| Bibliometrics | 0.000 | 0.001 |
| Science and technology studies | 0.000 | 0.000 |
| Scholarly communication | 0.001 | 0.001 |
| Open science | 0.001 | 0.001 |
| Research integrity | 0.001 | 0.001 |
| Insufficient payload (model declined to judge) | 0.001 | 0.000 |
Machine scores (provisional)
The two teacher heads of the student model, read on this work. A score orders the frame for review; it never asserts a category, and the validation status ships verbatim with every row.
Baseline scores from an immature model (maturity gate not passed, 7 training rounds). Scores rank; they never assert a category.
score_only:v0-immature-baseline · verbatim from the scoring run: score_only means the number may rank works, and no category label ships from itClassification
machine, unvalidatedMachine predicted; a candidate call from one source (direct Gemma or distilled Codex), not a consensus.
How this classification was reached, model by model and score by score, is at the end of the page under "How this classification was reached".