Computational Investigation of Chirality-Based Separation of Carbon Nanotubes Using Tripeptide Library.

Carbon nanotubes (CNT) have fascinating applications in flexible electronics, biosensors, and energy storage devices, and are classified as metallic or semiconducting based on their chirality. Semiconducting CNTs have been teased as a new material for building blocks in electronic devices, owing to their band gap resembling silicon. However, CNTs must be sorted into metallic and semiconducting for such applications. Formerly, gel chromatography, ultracentrifugation, size exclusion chromatography, and phage display libraries were utilized for sorting CNTs. Nevertheless, these techniques are either expensive or have poor efficiency. In this study, we utilize a novel technique of using a library of nine tripeptides with glycine as a central residue to study the effect of flanking residues for large-scale separation of CNTs. Through molecular dynamics, we found that the tripeptide combinations with threonine as one of the flanking residues have a high affinity for metallic CNTs, whereas those with flanking residues having uncharged and negatively charged polar groups show selectivity towards semiconducting CNTs. Furthermore, the role of interfacial water molecules and the ability of the tripeptides to form hydrogen bonds play a crucial role in sorting the CNTs. It is envisaged that CNTs can be sorted based on their chirality-selective interaction affinity to tripeptides.