A Visualization Analysis of Machine Learning Applications in Gas Adsorption Using Nanoporous Materials

Area of Science:

• Nanoporous Materials Science
• Computational Chemistry
• Materials Informatics

Background:

• Machine learning (ML) offers novel approaches for gas adsorption research.
• Quantifying the evolution of ML in nanoporous materials research is crucial.
• Bibliometric analysis provides insights into scientific field development.

Purpose of the Study:

• To analyze the publication growth, collaboration networks, journal sources, and thematic evolution of machine learning in gas adsorption research using nanoporous materials.
• To identify key research trends, leading contributors, and collaboration patterns.
• To provide guidance for future data-driven adsorbent discovery.

Main Methods:

• Bibliometric analysis of 730 records from the Web of Science Core Collection (2010-2026).
• Utilized VOSviewer, CiteSpace, and R for analyzing publication trends, collaboration networks, and keyword evolution.
• Screened 1581 initial publications to retain valid records.

Main Results:

• Rapid publication growth observed post-2019, reaching ~280 publications in 2025, indicating accelerated ML integration.
• Broadened journal distribution with high publication weights in leading journals like Chemical Engineering Journal and ACS Applied Materials & Interfaces.
• Identified 10 author clusters with limited cross-community collaboration; China and the US are leading contributors.
• Keyword evolution shows a shift towards data-guided material selection, high-throughput screening, and deep learning.

Conclusions:

• The integration of machine learning in gas adsorption research using nanoporous materials has experienced exponential growth.
• Current research landscape shows strong national contributions but limited inter-community collaboration.
• Future research should focus on data-driven approaches for efficient adsorbent discovery and performance optimization.