Hydrogen desorption study of as-synthesized carbon nanotubes using artificial neural network

Abstract

Carbon nanotubes are considered for hydrogen storage due to their low density, high strength, and hydrogen adsorption characteristics. Reports suggest that the total surface area of carbon affects the hydrogen storage capacities in carbon nanotubes. Based on the experimental data of as-synthesized carbon nanotubes, an artificial neural network (ANN) model was developed to study the relationship between the surface area of carbon and the hydrogen desorption. The model was also used to study the effect of carbon and alumina content to the hydrogen desorption. A feedforward ANN was used for the prediction. The ANN was trained using the Levenberg-Marquardt training algorithm.