The convergence of artificial intelligence (AI) and carbon nanotube (CNT) chemistry is accelerating innovations in the synthesis, functionalization, and advanced applications of carbon-based nanomaterials. This review highlights recent AIdriven methodologies, including neural networks, ensemble learning, metaheuristics, and hybrid frameworks that are redefining the design, surface engineering, and structure–property relationships of CNTs. Special attention is given to their roles in clean energy technologies, polymer nanocomposites, environmental systems, and nanoelectronics. Advances such as autonomous synthesis guided by deep learning, high-throughput experimentation, and AI-enabled property prediction are critically reviewed. Challenges including data fragmentation, class imbalance, and lack of benchmarking are addressed, alongside future directions such as physics-informed machine learning, robotics integration, and multi-objective optimization. This review positions AI as a disruptive catalyst in advanced CNT research, offering intelligent automation and predictive insights across diverse carbon-material applications.

AI-driven strategies for the design and functionalization of carbon nanotubes: a critical review
    Abstract
    1  Introduction
    2 Artificial intelligence in carbon nanotube research and modeling
        2.1 Why these AI algorithms for CNT problems
        2.2  Neural networks and their applications in CNT research
        2.3 Metaheuristic and evolutionary optimization algorithms
        2.4  Tree‑based and ensemble learning models
        2.5  Advanced machine‑learning techniques in CNT research
        2.6  Hybrid and integrated Machine-Learning models in CNT research
            2.6.1 Why hybrid models? Rationale and observed effects in CNT research
    3  Optimization of Carbon-Nanotube synthesis and surface modification
        3.1  CVD process control
        3.2 Functionalization of CNT surfaces
    4 Reinforcing polymer composites with carbon nanotubes
        4.1  Mechanical properties of CNT–Polymer composites
            4.1.1 Why CNTs stiffen and strengthen polymers
            4.1.2  State-of-the-Art ML models for mechanical response
            4.1.3  Physics insights extracted by ML
            4.1.4  Guidelines for future work
    4.2 Thermal and electrical properties of CNT–Polymer composites
    5  Environmental applications and wastewater treatment
        5.1 Adsorption of pollutants and heavy metals
        5.2  Integrating nanotube technology with advanced oxidation processes (AOPs)
    6  Sensing and electronic properties of carbon nanotubes
        6.1  Gas sensors and biosensing
            6.1.1  Gas sensors using CNTs
            6.1.2  Biosensors based on CNTs
    6.2  Modifying CNT structure to improve sensing response
        6.2.1  Metal doping and functionalization
        6.2.2  Defect engineering and functional groups
        6.2.3  Deep learning for structural identification
    7 Conclusion, analytical Reflection, and future directions
    References

• Farhang Daneshmand(Department of Mechanical Engineering, Penn State University-Scranton, Dunmore, PA 18512, USA)
• Mohammad Mohammadi(Department of Mechanical Engineering, Shi.C., Islamic Azad University, Shiraz, Iran)
• Yousef Bazargan Lari(Department of Mechanical Engineering, Shi.C., Islamic Azad University, Shiraz, Iran) Corresponding author
• Seyed Mohammad Reza Nazemosadat(Department of Mechanical Engineering, Shi.C., Islamic Azad University, Shiraz, Iran)