No abstract

No abstract

Carbon nanotubes (CNTs) possess remarkable mechanical, electrical, thermal, and optical properties that predestine them for numerous potential applications. The conventional chemical vapor deposition (CVD) route for the production of CNTs, however, suffers from costly and complex issues. Herein, we demonstrate a general and high-yield strategy to grow nitrogen-doped CNTs (NCNTs) on three-dimensional (3D) graphite felt (GF) substrates, through a direct thermal pyrolysis process simply using a common tube furnace, instead of the costly and complex CVD method. Specifically, the NCNTs-decorated GF (NCNT-GF) electrode possesses enhanced electrocatalytic performance towards cerium redox reactions, mainly due to the catalytic effect of N atoms doped into NCNTs, and ingenious and hierarchical 3D architecture of the NCNT-GF. As a result, the cell with the NCNT-GF serving as a positive electrode shows the improved energy efficiency with increases of about 53.4% and 43.8% over the pristine GF and the acidly treated GF at a high charge/discharge rate of 30 mA cm-2, respectively. Moreover, the as-prepared NCNT catalyst-enhanced electrode is found to be highly robust and should enable a long-term cycle without detectable efficiency loss after 500 cycles. The viable synthetic strategy reported in this study will contribute to the further development of more active heteroatom-doped CNTs for redox flow batteries.

No abstract

In order to improve the accuracy of signal recovery after transmitting over atmospheric turbulence channel, a deep-learning-based signal detection method is proposed for a faster-than-Nyquist (FTN) hybrid modulated optical wireless communication (OWC) system. It takes advantage of the long short-term memory (LSTM) network in the recurrent neural network (RNN) to alleviate the interdependence problem of adjacent symbols. Moreover, an LSTM attention decoder is constructed by employing the attention mechanism, which can alleviate the shortcomings in conventional LSTM. The simulation results show that the bit error rate (BER) performance of the proposed LSTM attention neural network is 1 dB better than that of the back propagation (BP) neural network and outperforms by 2.5 dB when compared with the maximum likelihood sequence estimation (MLSE) detection method.