Optimizing the growth of vertically aligned carbon nanotubes by literature mining and high-throughput experiments

Vertically aligned carbon nanotube (VACNT) arrays with good mechanical properties and high thermal conductivity can be used as effective thermal interface materials in thermal management. In order to take advantage of the high thermal conductivity along the axis of nanotubes, the quality and height of VACNT arrays need to be optimized. However, the immense size of the synthesis parameter space for VACNT arrays and the interdependence of structural features make it challenging to improve both the height and quality of VACNT arrays. Here, we developed a literature mining approach combined with machine learning and high-throughput design to efficiently optimize the height and quality of VACNT arrays. To reveal the underlying relationship between VACNT structures and key growth parameters, we employed a random forest regression (RFR) to model a set of published sample data (864 samples) and adopted the Shapley Additive exPlanations (SHAP) method. High-throughput experiments were designed to modulate four key parameters: growth temperature, growth time, catalyst compositions, and carbon source concentration. It was found that the screened Fe/Gd/Al2O3 catalyst was able to grow VACNT arrays with millimeter-scale height and improved quality. Our results demonstrate that text mining, high-throughput optimization, and data-based machine learning can effectively deal with multi-parameter processes such as nanotube growth and improve the control over the structures.