Fabrication of direct Z-scheme FeIn2S4/Bi2WO6 hierarchical heterostructures with enhanced photocatalytic activity for tetracycline hydrochloride photodagradation

A series of direct Z-scheme FeIn₂S₄/Bi₂WO₆ hierarchical heterostructures with intimate interface contacts were synthesized by in-situ growth route and characterized by systematical analyses. All as-prepared FeIn₂S₄/Bi₂WO₆ nanocomposites showed significantly enhanced photocatalytic activity towards photodegradation for the removal of tetracycline hydrochloride (TCH) in comparison with individual FeIn₂S₄ and Bi₂WO₆. Meanwhile, the highest photocatalytic degradation activity can be achieved by modulating adding amount of FeIn₂S₄ in FeIn₂S₄/Bi₂WO₆ nanocomposites and the optimized component ratio of FeIn₂S₄ to Bi₂WO₆ is determined to be 10 wt%. The enhanced photocatalytic activity could be ascribed to efficient separation between photogenerated holes and electrons based on the construction of direct Z-scheme system. The high photocatalytic stability of resultant 10 wt% FeIn₂S₄/Bi₂WO₆ nanocomposites was revealed through six successive recycling reactions. The main intermediate generated during TCH photodegradation was explored by HPLC-MS. Besides, the direct Z-scheme photocatalytic mechanism was confirmed by band position analysis, electron spin resonance (ESR) and active species capture experiment.     

Unbalanced probabilistic linguistic decision-making method for multi-attribute group decision-making problems with heterogeneous relationships and incomplete information

Artificial Intelligence Review - In group decision-making problems, decision makers prefer to use several linguistic terms to describe their own perception and knowledge, and give their preference...     

数据中心内服务器能耗数据采集及特征分析

数据中心的高能耗低能效问题正受到广泛关注与研究，但目前没有公开的数据中心内服务器能耗数据集供研究人员使用，且现有过滤式特征选择并不能满足运维人员的需求。为此，提出了一套模拟数据中心内服务器运行状态的仿真环境架构，基于该架构采集了服务器运行各类任务时的多项性能指标和能耗数据。然后将基于因果关系的特征选择应用于能耗数据集的特征分析中，构建出可解释的特征子集和能耗预测结果。实验结果表明，因果特征子集大小约为过滤式特征子集大小的1/3到1/6，并且使用因果特征子集训练的模型在75%的情况下都取得了最优预测精度。     

基于语言模型的预训练技术研究综述

预训练技术当前在自然语言处理领域占有举足轻重的位置。尤其近两年提出的ELMo、GTP、BERT、XLNet、T5、GTP-3等预训练模型的成功,进一步将预训练技术推向了研究高潮。该文从语言模型、特征抽取器、上下文表征、词表征四个方面对现存的主要预训练技术进行了分析和分类,并分析了当前自然语言处理中的预训练技术面临的主要问题和发展趋势。     

偏铁酸钠催化合成共轭亚油酸

以红花菜籽油中亚油酸为原料,偏铁酸钠为催化剂,在单因素试验的基础上,选取反应温度、反应时间、催化剂用量为考察因素,以异构化产物共轭亚油酸(CLA)转化率为响应值,通过响应面分析法对异构化反应进行优化,并建立相应的回归模型,试验结果表明,红花菜籽油中亚油酸异构化为CLA的最佳工艺为:反应温度195℃、反应时间4.7 h、催化剂用量0.6 g,在此条件下得到CLA转化率为92.90％,与预测值93.92％相比,误差为1.02％.然后再通过脲素包合纯化、经2％的H2SO4-CH3OH溶液甲酯化后,用气相色谱-质谱联用仪检测CLA含量为64.27％.     

Efficient photocatalytic H2 evolution over Cu and Cu3P co-modified TiO2 nanosheet

Schottky junction and p-n heterojunction are widely employed to enhance the charge transfer during the photocatalysis process. Herein, Cu and Cu₃P co-modified TiO₂ nanosheet hybrid (Cu–Cu₃P/TiO₂) was fabricated using an in situ hydrothermal method. The ternary composite achieved the superior H₂ evolution rate of 6915.7 μmol g^?1 h^?1 under simulated sunlight, which was higher than that of Cu/TiO₂ (4643.4 μmol g^?1 h^?1) and Cu₃P/TiO₂ (6315.8 μmol g^?1 h^?1) and pure TiO₂ (415.7 μmol g^?1 h^?1). The enhanced activity can be attributed to the collaboration effect of Schottky junction and p-n heterojunction among Cu/TiO₂ and Cu₃P/TiO₂, which can harvest the visible light, reduce the recombination of charge carriers and lower the overpotential of H₂ evolution, leading to a fast H₂ evolution kinetics. This work develops a feasible method for the exploration of H₂ evolution photocatalyst with outstanding charge separation properties.     

Discrimination the geographical origin of Yanchi Tan Lamb with different muscle sections by stable isotopic ratios and elemental profiles

In the study, lamb, feed and soil samples were collected from three regions of China in the year 2017 and 2018. The δ¹³C and δ¹⁵N values in lamb (defatted dry mass and lipids) and grass and the contents of nine elements in lamb, grass and soil were determined. The results indicated that except for Fe, Zn and Rb, most variables of lamb showed significant differences among two or three regions, and the contents of Fe, Sr and Cs showed significant differences between two years, and the contents of Cu and Zn showed significant differences among different muscle sections. Besides, significant correlations were found for δ¹³C between defatted lamb and lipid (r² = 0.973, P < 0.05) and for the element contents between lamb and soil. Finally, four key variables δ¹³C, δ¹⁵N, Sr and Mo were selected through stepwise canonical discriminant analysis with the cross-validation correct classification rate of 99.1%.     

城市轨道交通换乘序列空间适老对策探析

老年人作为生理弱势人群的代表，具有平等享受公共物品的权利。以老年人群搭乘轨道交通中的换乘行为为研究对象，通过路径跟随、访谈、心理量研究等方法，结合国内外文献研究，剖析轨道交通换乘序列空间与老年人群生理条件和行为特点的矛盾关键，进一步提出轨道交通换乘空间设计的适老策略。     

A triple (e−/O2−/H+) conducting perovskite BaCo0.4Fe0.4Zr0.1Y0.1O3-δ for low temperature solid oxide fuel cell

Low-temperature solid oxide fuel cells (LT-SOFCs) have recently gained enormous attention worldwide with a new research trend focusing on single layer fuel cells (SLFC), which have better cell performance than traditional SOFCs at low operating temperatures. In this study, a triple (e^?/O^2?/H⁺) conducting perovskite BaCo_0.4Fe_0.4Zr_0.1Y_0.1O_3-δ (BCFZY) is used as the intermediate layer material for SLFC. A high current density of 994 mA/cm² at 0.6 V and a peak power density of 610 mW/cm² with an OCV of 1.01 V has been achieved with a cell operating temperature of 550 °C, confirming the application feasibility of BCFZY in SLFCs. Furthermore, a typical proton conductor BaZr_0.8Y_0.2O_3-δ (BZY) is introduced into BCFZY to enhance the cell performance. By adjusting the mass ratio of the BCFZY-BZY layer, an optimal power density is obtained, achieving 703 mW/cm² with an OCV of 1.03 V at 550 °C with an 8BCFZY:2BZY (wt%) ratio. These findings prove that the proposed BCFZY-BZY holds great promise for developing SLFCs to realize low-temperature operation.     

Bacterial community succession and biogenic amine changes during fermentation of fish-chili paste inoculated with different commercial starter cultures

The microbial dynamics, diversity and succession characterisation during the fermentation of fish-chili paste, with and without starter cultures by high-throughput sequencing, were investigated to identify the relationship between microbial composition and the accumulation of biogenic amines. Results showed that Firmicutes was the predominant phylum, and Lactobacillus and Weissella were the main genera during fermentation process, regardless of the inclusion of starter cultures. Compared with naturally fermented samples (NS; no starter cultures), Pediococcus and Staphylococcus flourished in samples inoculated with SBM-52 starter culture (Staphylococcus xylosus + Staphylococcus carnosus + Pediococcus pentosaceus + Pediococcus acidilactici). Staphylococcus also accounted for a larger proportion in samples inoculated with WBX-43 starter culture (Staphylococcus xylosus + Staphylococcus carnosus). Furthermore, Enterobacter was inhibited in inoculated samples. Unweighted pair-group method with arithmetic means and principal component analyses revealed that microbiota structures were different among NS, SBM and WBX samples. In case of biogenic amines, the concentration of putrescine, cadaverine and tyramine increased to higher levels in NS samples compared with inoculated samples. Our work also investigated the relationship between core bacterial communities and biogenic amines. This study provided an improved understanding for the effects of starter cultures on the microbial community and the quality of fish-chili paste.