Search

This study focuses on an improved leaching process through the combination of pressurized conditions and direct filtration of acid leaching slurry, which is conductive to improving the filterability of acid leaching systems and the extraction rates of Ti, Al, and Mg components. The effects of sulfuric acid concentration, reaction temperature, particle size of materials, acid–slag ratio, and reaction time on the leaching efficiency were systematically investigated. The results showed that pressurization significantly enhances the filtration efficiency of the reaction slurry. Under the same filtration time, the filtration efficiency increased from 46% under ordinary pressure to 78% under pressurized conditions. Moreover, under the optimal reaction conditions, the extraction rates of Ti, Al, and Mg components were more than 88.21%, 97.8%, and 96.31%, respectively. Additionally, XRD and FTIR showed that titanium oxide sulfate hydrate crystals were produced in the acid-leached residues when the reaction temperature exceeded 190 °C, thereby reducing the extraction rate of Ti component. And the XRD pattern shows that when the reaction temperature is maintained at 190 °C and the reaction time is extended to 150 min, titanium oxide sulfate hydrate crystals will be formed to reduce the extraction rate of the Ti component. In summary, this study not only provides important theoretical support for the resource utilization of high-titanium blast furnace slag but also offers a feasible solution for efficient extraction and convenient filtration, thus holding significant academic and practical implications. [ABSTRACT FROM AUTHOR]

Federated Learning allows edge devices to collaboratively train a shared model on their local data without leaking user privacy. The non-independent-and-identically-distributed (Non-IID) property of data distribution, which leads to severe accuracy degradation, and enormous communication overhead for aggregating parameters should be tackled in federated learning. In this article, we conduct a detailed analysis of parameter updates on the Non-IID datasets and compare the difference with the IID setting. Experimental results exhibit that parameter update matrices are structure-sparse and show that more gradients could be identified as negligible updates on the Non-IID data. As a result, we propose a structure-based communication reduction algorithm, called FedSCR, that reduces the number of parameters transported through the network while maintaining the model accuracy. FedSCR aggregates the parameter updates over channels and filters, identifies and removes the redundant updates by comparing the aggregated values with a threshold. Unlike the traditional structured pruning methods, FedSCR retains the complete model that does not require to be retrained and fine-tuned. The local loss and weight divergence on each device vary a lot because of the unbalanced data distribution. We further propose an adaptive FedSCR, that dynamically changes the bounded threshold, to enhance the model robustness on the Non-IID data. Evaluation results show that our proposed strategies achieve almost 50 percent upstream communication reduction without loss of accuracy. FedSCR can be integrated into state-of-the-art federated learning algorithms to dramatically reduce the number of parameters pushed to the global server with a tolerable accuracy reduction. [ABSTRACT FROM AUTHOR]

The anticancer effects of immune checkpoint inhibitors (ICIs) have been widely examined recently. Although ICIs have been progressively improved for successful gastric cancer treatment, different trials of ICIs such as pembrolizumab and nivolumab have yielded widely variable response rates. Strategies to further improve the efficacy of ICIs are still needed. Previous studies have shown that de novo DNA methylation is acquired by PD1+CD8+ tumor-infiltrating T cells (TILs), which cause a hierarchical downregulation of cytokines such as interferon-γ (IFN-γ). The epigenetic agent 5-Aza-2'-deoxycytidine (DAC) blocks de novo DNA methylation in activated PD1+CD8+ TILs. Such a feature might help enhance the anti-tumor effect of immune checkpoint blockade (ICB) treatment. In this study, polyethylene glycol-poly(ε-caprolactone) (PEG-PCL) nanoparticles (NPs) were linked to the anti-programmed death-1 monoclonal antibody nivolumab to yield αPD1-NPs for targeting TILs with PD1 overexpression using DAC. In addition, the NPs increased DAC stability and improved IFN-γ secretion and the anti-tumor effect of ICB in vitro . Therefore, targeted delivery of DAC reverses the exhaustion of PD1+CD8+ TILs and improves T cell responses and the treatment effect of ICB. These findings suggest that nivolumab-NPs are a potential tool for the delivery of epigenetic drugs, which could enhance the anti-tumor effect of ICB in gastric cancer., Competing Interests: The authors declare no competing interests., (© 2021 The Author(s).)

Existing smart radiation devices suffer from numerous disadvantages such as large thicknesses, limited dimensions, or requirements for sustained electrical power. The present study addresses these issues by proposing a smart thermal control coating based on CaF 2 /VO 2 core-shell (CaF 2 @VO 2 ) structured microspheres prepared by a solvent/hydrothermal-calcination method and distributed within an easily applied polymer matrix. Here, the dielectric-to-metallic transition property of the VO 2 shell material with increasing temperature is used to regulate the optical scattering and absorption characteristics of the CaF 2 @VO 2 core-shell microspheres to realize a positive and reversible increase in the emissivity of the coating from 0.47 at 30 °C to 0.83 at 90 °C. The mechanisms behind this effect are investigated by theoretical analyses and numerical simulations. The present work can expect to promote the further research and development of new coating materials for smart thermal control applications.