Numerical Investigation of Thermofluid Flow in a Chemical Vapor Deposition Furnace Utilized to Manufacture Template-Synthesized Carbon Nanotubes

Abstract

Template-based chemical vapor deposition (TB-CVD) is a versatile technique for manufacturing carbon nanotubes (CNTs) or CNT-based devices for various applications. In this process, carbon is deposited by thermal decomposition of a carbon-based precursor gas inside the nanoscopic cylindrical pores of anodized aluminum oxide (AAO) templates to form CNTs. Experimental results show that CNT formation in templates is controlled by TB-CVD process parameters, such as time, temperature, and flow rate. However, optimization of this process is done empirically, requiring tremendous time and effort. Moreover, there is a need for a more comprehensive and low cost way to characterize the flow in the furnace in order to understand how process parameters may affect CNT formation. In this report, we describe the development of four, 3D numerical models (73 < Re < 1100), each varying in complexity, to elucidate the thermofluid behavior in the TB-CVD process. Using computational fluid dynamic (CFD) commercial codes, the four models are compared to determine how the presence of the template and boat, composition of the precursor gas, and consumption of species at the template surface affect the temperature profiles, velocity fields, mixed convection, and strength of circulations in the system. The benefits and shortcomings of each model, as well as a comparison of model accuracy and computational time, are presented. Due to limited data, simulation results are validated by experiments and visual observations of the flow structure whenever possible. Decent agreement between experimental data and simulation supports the reliability of the simulation.