Frequency Band-Controllable Acoustic Topological Insulators Based on Local Dimensional Coordination.

The potential value of acoustic topological insulators is of wide interest because of their special topology-protected edge-state unidirectional transmission properties. In this article, a frequency band-controlled acoustic topological insulator based on local dimensional coordination is designed. It is capable of controlling the unidirectional transmission frequency range by controlling the position of the band gap within a certain range. The energy band structure and eigenmodes are analyzed by using the finite element method. The range of control of the band gap position by the local dimension length is derived. The control of the band gap over the unidirectional propagation frequency is experimentally verified. This work provides a way to precisely control the unidirectional transmission frequency of topology-protected edge states in a specific band, opening up new possibilities for their potential value.

UniParser: Multi-Human Parsing With Unified Correlation Representation Learning.

Multi-human parsing is an image segmentation task necessitating both instance-level and fine-grained category-level information. However, prior research has typically processed these two types of information through distinct branch types and output formats, leading to inefficient and redundant frameworks. This paper introduces UniParser, which integrates instance-level and category-level representations in three key aspects: 1) we propose a unified correlation representation learning approach, allowing our network to learn instance and category features within the cosine space; 2) we unify the form of outputs of each modules as pixel-level results while supervising instance and category features using a homogeneous label accompanied by an auxiliary loss; and 3) we design a joint optimization procedure to fuse instance and category representations. By unifying instance-level and category-level output, UniParser circumvents manually designed post-processing techniques and surpasses state-of-the-art methods, achieving 49.3% AP on MHPv2.0 and 60.4% AP on CIHP. We have released our source code, pretrained models, and demos to facilitate future studies on https://github.com/cjm-sfw/Uniparser.

Oxygen Electrode PrBa0.5Sr0.5Co1.5Fe0.5O5+δ-BaZr0.1Ce0.7Y0.1Yb0.1O3-δ with Different Composite Proportions for Proton-Conducting Solid Oxide Electrolysis Cells.

Proton-conducting solid oxide electrolysis cell (H-SOEC), as a hydrogen production device using proton conductor oxides as an electrolyte, has gained attention due to its various advantages of being more suitable for operating conditions at intermediate and low temperatures. However, its commercialization urgently needs to address the issue of insufficient catalytic activity of the oxygen electrode at lower temperatures. In this work, PrBa0.5Sr0.5Co1.5Fe0.5O5+δ-BaZr0.1Ce0.7Y0.1Yb0.1O3-δ (PBSCF-BZCYYb) series composite materials (denoted as PBSCF-BZCYYb46, PBSCF- BZCYYb55, and PBSCF-BZCYYb64 based on the mass ratios of PBSCF and BZCYYb as 4:6, 5:5, and 6:4, respectively) are prepared and applied as oxygen electrodes for H-SOECs. The H-SOECs with the structure of PBSCF-BZCYYb|BZCYYb|NiO-BZCYYb (active layer)|NiO-BZCYYb (support layer) are prepared and recorded as Cell 1, Cell 2, and Cell 3 with PBSCF-BZCYYb46, PBSCF-BZCYYb55, and PBSCF-BZCYYb64 as oxygen electrodes. The H-SOECs exhibit electrolysis current densities of 669.00, 743.80, and 503.30 mA cm-2 under 1.3 V at 650 °C, respectively. The cells also show considerable stability in the constant voltage electrolysis of 179.5, 152.8, and 83.0 h, respectively. Through the comparison of various electrochemical properties, PBSCF-BZCYYb55 is considered the most promising oxygen electrode material in this work.

Efficient encapsulation of fat-soluble food-derived biofunctional substances (curcumin as an example) in dual-modified starch-based nanoparticles containing large conjugated systems.

Acid-ethanol hydrolysis and subsequent cinnamic acid (CA) esterification were employed to prepare a series of dual-modified starches efficiently loaded with curcumin (Cur) utilizing large conjugation systems provided by CA. Structures of the dual-modified starches were confirmed by IR and NMR, and their physicochemical properties were characterized by SEM, XRD and TGA. The nanoparticles fabricated from the dual-modified starch have perfect spherical shape (250.7-448.5 nm, polydispersity index <0.3), excellent biosafety (no hematotoxicity, no cytotoxicity, no mutagenicity) and high loading of Cur (up to 26.7 % loading). By XPS analysis, this high loading was believed to be supported by the synergistic effect of hydrogen bonding (provided by hydroxyl groups) and π-π interactions (provided by large conjugation system). In addition, the encapsulation of dual-modified starch nanoparticles effectively enhanced the water solubility (18-fold) and physical stability (6-8-fold) of free Cur. In vitro gastrointestinal release showed that Cur-encapsulated dual-modified starch nanoparticles were released more preferably than free Cur and that the Korsmeyer-Peppas model was the most suitable release model. These studies suggest that dual-modified starches containing large conjugation systems would be a better alternative for encapsulating fat-soluble food-derived biofunctional substances in functional food and pharmaceutical applications.