Carbon nanotubes can be manufactured using neural networks
Carbon nanotube thin films hold great promise in advanced optoelectronics and energy, as well as medicine. However, their production process is subject to strict standardization requirements and close supervision, so they are unlikely become widespread anytime soon.
The multiphase manufacturing process of nanotubes, which is difficult to manage, is a major obstacle to unlocking their vast potential. Dmitry Krasnikov, a Skoltech researcher and one of the authors in the study, explains that artificial neural networks can be used to analyze experimental data.
The authors of a paper published in Carbon, a prestigious journal, show how machine learning, and in particular, ANNs trained on experimental parameters such as temperature and gas pressure, can be used to monitor the properties and performance of carbon nanotube films.