Search

Laser powder bed fusion (LPBF) additive manufacturing, as a novel technique provides broad benefits in thermoelectric materials synthesis, such as enhanced printing speed, reduced waste materials, and customized dimension design. The high energy density of laser could potentially balance the synthesis rate and thermoelectric figure-of-merit zT , but the correlations between laser energy and material performance are still vague. Herein, Bi 0.5 Sb 1.5 Te 3 bulks are directly synthesized by LPBF with a recorded synthesis rate of 254 g h−1. The maximum zT , without any post-processing, reaches 1.1 at 75 °C. Induced nanoscale pores by high laser energy printing, obtain a comparable diameter to phonon mean free path, leading to reduced lattice thermal conductivity and enhanced TE performance. We found that input laser energy critically affected thermoelectric performance and concluded the correlation between volumetric energy density E V and power factor PF , PF = 9.05 E V 2 – 30.01 E V + 48.84. It potentially predicts thermoelectric performance with inputted laser energy density. [ABSTRACT FROM AUTHOR]

Highlights • Hybrid pooling is proposed for improving the generalization ability of convolutional neural networks. • The performance of the hybrid pooling is presented in image classification tasks. • The hybrid pooling combined with the dropout technique is competitive to other methods. • Layer-wise control of feature extraction can be realized by the hybrid pooling. Abstract Convolutional neural networks (CNNs) have attracted considerable attention in many application fields for their great ability to deal with image recognition and object detection tasks. A pooling process is an important process in CNNs, which serves to decrease the dimensionality of processed data for reducing computational cost as well as for enhancing tolerance to translation and noise. Although standard pooling methods, such as the max pooling and the average pooling, are typically adopted in many studies, a newly devised pooling method could improve the generalization ability of CNNs. In this study, we propose a hybrid pooling method which stochastically chooses the max pooling or the average pooling in each pooling layer. A characteristic of the hybrid pooling is that the probability for choosing one of the two pooling methods can be controlled for each convolutional layer. In image classification tasks with benchmark datasets, we show that the hybrid pooling is effective for increasing the generalization ability of CNNs. Moreover, we demonstrate that the hybrid pooling combined with the dropout is competitive with other existing methods in classification performance. [ABSTRACT FROM AUTHOR]

As the cathode ternary activated materials with high content of nickel have been widely used in lithium-ion batteries, the recycling of the spent cathode materials has been paid worthy attention. In this work, the sulfate solution with Li, Ni, Co and Mn elements was formed by dissolving the spent LiNi 0.5 Co 0.2 Mn 0.3 O 2 cathode materials in 20% sulfuric acid and 30% H 2 O 2 , assisted with reflux at 90 °C for 1 h. Based on the above solution, the precursor (Ni 0.5 Co 0.2 Mn 0.3 (OH) 2), along with the lithium and sodium sulfates mixture solutions were prepared by co-precipitation at 55 °C. Then Li 2 CO 3 powders and sodium sulfate solution were obtained through evaporating filtered solution separated from the preparation of the precursor and added an amount of Na 2 CO 3 , with the removal of 30% water and separation at about 70 °C. The purity of recovered Li 2 CO 3 is about 99.5%. The residual concentrations of Ni, Co and Mn in the filtered solution separated from the preparation of precursor are only 0.041, 0.025, 0.038 ppm left, respectively. The prepared cathode material LiNi 0.5 Co 0.2 Mn 0.3 O 2 from the precursor exhibits good cycle performance and shows a stable discharge capacity of 140 mAh g−1 at a current rate of 100 mA g−1 between 2.8 and 4.2 V. Image 1 • The facile strategy of entire recovery is proposed from spent LiNi 0.5 Co 0.2 Mn 0.3 O 2 cathode material. • The products of new LiNi 0.5 Co 0.2 Mn 0.3 O 2 , Li 2 CO 3 and Na 2 SO 4 were obtained. • No ammonia is introduced throughout the recovery process. • The filtered solution contains ultra-trace residual concentration of Ni, Co and Mn elements. [ABSTRACT FROM AUTHOR]