Search

Abstract: Layer‐structured (MoO2)2P2O7 was prepared through the decomposition of MoO2HPO4 and further used as a cathode material for sodium‐ion batteries. Results reveal that the capacity and cycling stability of the (MoO2)2P2O7/Na battery largely depends on the lower cutoff voltage. When cycling between 4.0 ‐1.5 V, (MoO2)2P2O7 delivers a capacity of 180 mAh g−1 in the first cycle, but it drops very fast owing to a phase decomposition of the crystal structure to an amorphous one. The cycling stability is still not good when cycling between 4.0 ‐1.7 V. However, when cycling between 4.0 ‐ 2.0 V, the performance can be significantly improved, which delivers 93 mAh g−1 in the first cycle and, after 50 cycles, the capacity can maintain 55 mAh g−1. This is mainly attributed to the stable crystal structure of the Na‐intercalated (MoO2)2P2O7 electrode. Ex situ X‐ray diffraction examination demonstrates that the volume change of the (MoO2)2P2O7 electrode during the first charging process is only 1 %. [ABSTRACT FROM AUTHOR]

In this article, polyaniline (PANI) nanofibers were prepared and added to low-density polyethylene (LDPE) to produce PANI nanofibers/LDPE composites. LDPE and the composites were tested for direct current (DC) conductivity, breakdown strength, and space charge characteristics. The results suggested that DC breakdown strength of PANI nanofibers/LDPE composites significantly declined once PANI was added, and the decline was more evident with the increase of PANI nanofibers. Meanwhile, the addition of PANI nanofibers contributed to a decrease in the conductivity of LDPE. As the content of PANI nanofibers increased, the conductivity of the composites declined first and then raised. DC conductivity properties of LDPE could be improved by adding an appropriate amount of PANI nanofibers. Compared with LDPE, the space charge distribution was changed in LDPE due to the addition of PANI nanofibers. With the increase of content of PANI nanofibers, the amount of space charges close to the electrodes decreased gradually. [ABSTRACT FROM AUTHOR]

• Comprehensive evaluations of gridded transpiration (T) products, including ERA5L, GLDAS-2.2, and GLEAM-3.6, are conducted using collocation analysis at the global scale for the first time. • Among all products, ERA5L and GLEAM-3.6 outperform GLDAS-2.2 with relatively lower uncertainty regarding signal-to-noise ratio and absolute error variance. • Our results illustrate the importance of considering non-zero error cross-correlation by comparing the evaluation results obtained by triple collocation (TC) and the Extended instrumental variable technique (EIVD). Transpiration (T) constitutes a significant portion of total terrestrial evapotranspiration in ecosystems and functions as the connection between the water and energy cycle. Despite its importance, large-scale quantification of T remains a formidable challenge. Global gridded T datasets are essential for investigating and predicting the regulatory mechanisms of transpiration. However, the scarcity of reliable field references data limits direct validation efforts. This study uses collocation analysis methods, including triple collocation (TC) and extended double instrumental variable algorithm (EIVD), to assess three global T products at 0.25°-daily resolutions. Signal-to-noise ratios (SNR) and absolute error variances (σ ε 2) obtained through EIVD serve as primary indicators. Our results show that both TC and EIVD approaches yield valuable insights into the accuracy and performance of the T products. Moreover, the analysis emphasizes the necessity of considering non-zero error cross-correlation when utilizing collocation analysis for error characterization. Overall, this study presents a comprehensive examination and comparison of collocation analysis for error characterization of three T products. The findings suggested that collocation analysis methods may function as reliable alternatives to tower observations at a global scale, offering potential benefits for data assimilation and merging. [ABSTRACT FROM AUTHOR]

Conventionally, the flow field over the hydrofoil is solved by computational fluid dynamics (CFD), which is a computationally expensive task. As an alternative, deep learning (DL) approximates the flow field without solving the complicated Navier-Stokes (NS) equations. Consequently, this paper proposes a data-driven methodology using Convolutional Neural Networks (CNN) for hydrofoils performance prediction, which automatically detects fundamental features and accurately predicts the flow field of the hydrofoil. The CFD simulation is used to generate the dataset for neural network training and testing. The dataset containing hydrofoil geometry and operating conditions is fed into the CNN to approximate the model for predicting the flow field. Furthermore, we use novel attention gates (AGs) that concentrate on hydrofoils of different shapes and sizes. Models trained with AGs not only learn to restrain unrelated features of the input information, but also emphasize significant features useful for flow field prediction. In addition, the effects of various hyperparameters of models on convergence have been extensively investigated. The experimental results demonstrate that the flow field over a hydrofoil can be obtained within 0.5 s, with a mean relative error of less than 1.5%. • The accuracy of deep learning models for the inference on Reynolds-averaged Navier-Stokes solutions is investigated. • The dataset containing hydrofoil geometry, Reynolds number, and angle of attack is constructed. • The effect of attention gates (AGs) for flow field prediction is investigated. • The influence of different hyperparameters on the convergence and accuracy of deep learning models is studied. • The flow field over a hydrofoil can be obtained within 0.5 s, with a mean relative error of less than 1.5%. [ABSTRACT FROM AUTHOR]