Discipline: Physics
Subcategory: Nanoscience
Session: 1
Room: Capitol
Nixon Ogoi - North Carolina Central University
Co-Author(s): Brandon Ma, North Carolina Central Univerity, Durham, NC Oluwayotin Atikekeresola, North Carolina Central Univerity, Durham, NC Taiwo Ayobami, North Carolina Central Univerity, Durham, NC Fouzia Sahtout, North Carolina Central Univerity, Durham, NC Gaolin Milledge, North Carolina Central Univerity, Durham, NC Abdennaceur Karoui,North Carolina Central Univerity, Durham, NC
Carbon nanotubes (CNTs) are unique tubular structures of nanometer diameter with nanoscale properties that make them suitable for a wide range of applications. Published research about the mechanical properties of CNTs currently exist. However, there is no complete CNT database usable by scientists and engineers. This limits the extent of subsequent research on CNTs, slows the engineering of new and composite materials, and often leads to resource waste. We aim to develop a large and complete dataset that enables materials engineering using Machine Learning (ML) methods. The primary goal is to first generate CNT structural, mechanical, and thermal properties for a large set of chirality values.This presentation outlines the tools, processes that have been used to calculate SWCNT mechanical properties and discusses the initial results. We conducted tensile test on Single Walled Carbon Nanotubes (SWCNTs) with different chirality and lengths on Bridges-2 supercomputer (NSF XSEDE network), using molecular dynamics (MD) implemented in Large-scale Atomic/Molecular Massively Parallel Simulator (LAMMPS) software package. LAMMPS output files of the tensile tests were analyzed using computer tools developed in-house as well as Ovito software. Part of the generated data is validated using experimental data from the literature; this served the calibration of our calculation tools and methods. Our results showed the independence of the stress-strain curves on the SWCNT length. We proposed the dependence is due to local character of CNT mechanical properties, thus, they are mainly influenced by the short-range character of the bonding. On the contrary, the ultimate strength before fracture of the SWCNT depends on chirality. Also, in the case of zigzag SWCNT, chirality (n,0), the ultimate strength increases with the diameter of the nanotube, from 80 GPa up to 137 GPa, while for the armchair SWCNT, chirality (n,n), it is independent of the diameter. The salient feature of the database being developed is the inclusion with the CNT elastic properties, the behavior and parameters that describe the CNT plastic regime. The CNT database is being extended using ML models. Ultimately, materials engineers will use this comprehensive CNT database to design a wide range of CNTs based composite materials, beyond the conventional design, such as extreme operation conditions.
Funder Acknowledgement(s): This work is supported by the National Nuclear Security Administration award NA0003979, the NSAM-ML project, and the NSF XSEDE supercomputer network (now ACCESS) and the Pittsburg Supercomputer Center.
Faculty Advisor: Dr. Abdennaceur Karoui, akaroui@NCCU.EDU
Role: processes that have been used to calculate SWCNT mechanical properties and discusses the initial results. We conducted tensile test on Single Walled Carbon Nanotubes (SWCNTs) with different chirality and lengths on Bridges-2 supercomputer (NSF XSEDE network), using molecular dynamics (MD) implemented in Large-scale Atomic/Molecular Massively Parallel Simulator (LAMMPS) software package.