Towards control of carbon nanotube synthesis process using prediction-based fast Monte Carlo simulations

Precise control of the lengths of carbon nanotube (CNT) and other nanostructures is important for various industrial applications. However, time-resolution (∼1 min) of current in situ measurements does not allow control of lengths to within 20 nm. We present an approach to combine intermittent in situ measurements with length estimates from a fast atomistic Monte Carlo (MC) simulation of CNT synthesis. The MC simulation time was reduced by >70% through prediction of the nonlinear and nonstationary growth increments, and initialization of relaxation process (the most computationally intensive step in MC simulations) with the near-optimum predicted positions, leading to one of the longest (∼194 nm) CNTs from atomistic simulations. A utility function of growth predictions was defined so that its maximization specified the end-point of the synthesis process. Extensive simulation studies indicate that the approach can be used to control CNT lengths to within 1 nm of specifications.