Image processing for in-situ feedback control of fiber spinning of carbon nanotubes

Abstract

Carbon nanotubes (CNTs) have gained momentum in industrial applications over the last few decades due to their versatile mechanical, thermal, and electrical properties. Demand for manufactured CNT structures to have precise properties has increased, and with it emerge new manufacturing techniques. This project focuses on an emerging technology that uses an extensional flow apparatus to produce continuous strands of CNTs from a suspension of CNT nanoparticles. The properties of the CNTs produced depend heavily on the configuration of flow rates of the CNT solution and the surrounding solvent. To examine the effects of various flow configurations on a specific property, CNT strand diameter, and provide the basis for a control loop, a LabVIEW virtual instrument (VI) was designed to process experimental images and measure the diameters of the CNT strands produced through an edge detection module. A proof-of-concept experiment was run to do a brief survey of various flow configurations and to test the performance of the LabVIEW system. Several key setup considerations were identified, including lighting of the apparatus to facilitate accurate image processing, and suggestions for improvement of the design of the physical apparatus were identified. The experiment investigated several flow configurations and measured both the average diameter of the CNT strand and the change in diameter as the CNT flows downstream, both of which are important to the control of the flow, where the first one determines the absolute size of the fiber, and the second one relates to the elongational force that the fiber feels. [2] The LabVIEW VI successfully identified and measured the CNT strand 80% of the time, and the unsuccessful cases were examined to determine solutions for improving the VI.