Nanjing University of Science and Technology

About: Nanjing University of Science and Technology is a education organization based out in Nanjing, China. It is known for research contribution in the topics: Catalysis & Computer science. The organization has 31581 authors who have published 36390 publications receiving 525474 citations. The organization is also known as: Nánjīng Lǐgōng Dàxué & Nánlǐgōng.

Three-Dimensional Hierarchical Structure ZnO@C@NiO on Carbon Cloth for Asymmetric Supercapacitor with Enhanced Cycle Stability.

Abstract: In this work, we synthesized the hierarchical ZnO@C@NiO core–shell nanorods arrays (CSNAs) grown on a carbon cloth (CC) conductive substrate by a three-step method involving hydrothermal and chemical bath methods. The morphology and chemical structure of the hybrid nanoarrays were characterized in detail. The combination and formation mechanism was proposed. The conducting carbon layer between ZnO and NiO layers can efficiently enhance the electric conductivity of the integrated electrodes, and also protect the corrosion of ZnO in an alkaline solution. Compared with ZnO@NiO nanorods arrays (NAs), the NiO in CC/ZnO@C@NiO electrodes, which possess a unique multilevel core–shell nanostructure exhibits a higher specific capacity (677 C/g at 1.43 A/g) and an enhanced cycling stability (capacity remain 71% after 5000 cycles), on account of the protection of carbon layer derived from glucose. Additionally, a flexible all-solid-state supercapacitor is readily constructed by coating the PVA/KOH gel electrolyte bet...

Understanding the Disharmony Between Dropout and Batch Normalization by Variance Shift

Abstract: This paper first answers the question ``why do the two most powerful techniques Dropout and Batch Normalization (BN) often lead to a worse performance when they are combined together in many modern neural networks, but cooperate well sometimes as in Wide ResNet (WRN)?'' in both theoretical and empirical aspects. Theoretically, we find that Dropout shifts the variance of a specific neural unit when we transfer the state of that network from training to test. However, BN maintains its statistical variance, which is accumulated from the entire learning procedure, in the test phase. The inconsistency of variances in Dropout and BN (we name this scheme ``variance shift'') causes the unstable numerical behavior in inference that leads to erroneous predictions finally. Meanwhile, the large feature dimension in WRN further reduces the ``variance shift'' to bring benefits to the overall performance. Thorough experiments on representative modern convolutional networks like DenseNet, ResNet, ResNeXt and Wide ResNet confirm our findings. According to the uncovered mechanism, we get better understandings in the combination of these two techniques and summarize guidelines for better practices.

Ultralarge All-Inorganic Perovskite Bulk Single Crystal for High-Performance Visible–Infrared Dual-Modal Photodetectors

Abstract: Visible-infrared dual-modal light harvesting is crucial for various optoelectronic devices, particularly for solar cells and photodetectors. For the first time, this study reports on large 25 cm3-volume all-inorganic perovskite CsPbBr3 single crystal (SC) with an emphasis on the observed visible-infrared dual-modal light harvesting and sensing as demonstrated by the high-performance visible-infrared dual-modal photodetectors. First, ultralarge 25 cm3-volume CsPbBr3 SC ingots with trapping state density as low as of 1 × 109 cm−3 have been achieved by a modified Bridgman growth method. The volume reported here is the largest CsPbX3 (X = Cl, Br, I) all-inorganic perovskite system up to now, and the SC can be facilely cut into SC wafers with a diameter of 25 mm for various optoelectronic devices. Furthermore, these CsPbBr3 SCs exhibit a visible absorbance coefficient, a near-infrared (IR) two-phonon absorption coefficient, a carrier diffusion length, and a mobility as high as of 105 cm−1, 3.7 cm per Goeppert-Mayer (GM), 10 µm and 2000 cm2 V−1 s−1, respectively. These merits match well to the requirements of high-performance Vis-IR dual-modal light harvesting optoelectronic devices, which has been demonstrated by the CsPbBr3 SC photodetectors operated under the irradiation of both visible and IR light sources with light on/off ratio higher than 103. These results demonstrate the CsPbBr3 SCs with high visible-infrared dual-modal light harvesting capability and excellent electrical transporting properties have a huge potential in various optoelectronic devices, such as solar cells, photodetectors, and lasers.

Hierarchically Structured Co3O4@Pt@MnO2 Nanowire Arrays for High-Performance Supercapacitors

Abstract: Here we proposed a novel architectural design of a ternary MnO2-based electrode - a hierarchical Co3O4@Pt@MnO2 core-shell-shell structure, where the complemental features of the three key components (a well-defined Co3O4 nanowire array on the conductive Ti substrate, an ultrathin layer of small Pt nanoparticles, and a thin layer of MnO2 nanoflakes) are strategically combined into a single entity to synergize and construct a high-performance electrode for supercapacitors. Owing to the high conductivity of the well-defined Co3O4 nanowire arrays, in which the conductivity was further enhanced by a thin metal (Pt) coating layer, in combination with the large surface area provided by the small MnO2 nanoflakes, the as-fabricated Co3O4@Pt@MnO2 nanowire arrays have exhibited high specific capacitances, good rate capability, and excellent cycling stability. The architectural design demonstrated in this study provides a new approach to fabricate high-performance MnO2-based nanowire arrays for constructing next-generation supercapacitors.

Two-Dimensional, Porous Nickel–Cobalt Sulfide for High-Performance Asymmetric Supercapacitors

Abstract: High specific surface area, high electrical conductivity, and abundant channels have been recognized to favor pseudocapacitors, but their realization at the same time is still a great challenge. Here, we report on nickel–cobalt sulfide nanosheets (NSs) with both ultrathin thickness and nanoscale pores for supercapacitors. The porous Ni–Co sulfide NSs were facilely synthesized through micelle-confined growth and subsequent sulfuration. The NSs are as thin as several nanometers and have a large number of pores with a mean size of ∼7 nm, resulting in ultrahigh atom ratio at surface with unique chemical and electronic structure. Therefore, fast diffusion of ions, facile transportation of electrons and high activity make great synergistic contributions to the surface-dependent reversible redox reactions. In the resulted supercapacitors, a specific capacitance of 1304 F g–1 is achieved at a current density of 2 A g–1 with excellent rate capability that 85.6% of the original capacitance is remained at 20 A g–1. ...