VibeTube: a code for accurate extraction of CNTs' chirality from TEM images

Accurate determination of the chiral indices of carbon nanotubes (CNTs) is crucial for their controllable synthesis and practical applications. Transmission electron microscopy (TEM) combined with electron diffraction has been one of the most reliable methods for characterizing the chiral indices of CNTs. However, it is still a challenge to analyze TEM images to extract chirality with high efficiency and accuracy, especially for nanotubes with a diameter larger than 2 nanometers. In this work, a Python code is developed with assistance from an artificial intelligence model (Claude Code) for the auto-extraction of CNTs' chirality from TEM images. Precise and reliable measurement of the diameter was realized by using a new method to determine the average distance between the bright and dark fringes. Symmetry and geometry-based rules enabled measurements of layer lines in the Fourier transformation of TEM images with a sub-pixel resolution. As a result, an accuracy higher than 90% for CNTs with diameters ranging from 0.5 nm to 3.0 nm was achieved. High robustness has been validated by experimentally analyzing the chiralities of CNTs grown by a floating catalyst chemical vapor deposition method. The open-source codes and tools will be valuable for the research community to quantitatively identify the chirality distribution of CNTs with high efficiency.