High-specific-surface-area magnetic porous carbon microspheres (MPCMSs) were fabricated by annealing Fe(2+)-treated porous polystyrene (PS) microspheres, which were prepared using a two-step seed emulsion polymerization process. The resulting porous microspheres were then sulfonated, and Fe(2+) was loaded by ion exchange, followed by annealing at 250 °C for 1 h under an ambient atmosphere to obtain the PS-250 composite. The MPCMS-500 and MPCMS-800 composites were obtained by annealing PS-250 at 500 and 800 °C for 1 h, respectively. The iron oxide in MPCMS-500 mainly existed in the form of Fe3O4, which was concluded by characterization. The MPCMS-500 carbon microspheres were used as catalysts in heterogeneous Fenton reactions to remove methylene blue (MB) from wastewater with the help of H2O2 and NH2OH. The results indicated that this catalytic system has a good performance in terms of removal of MB; it could remove 40 mg L(-1) of MB within 40 min. After the reaction, the catalyst was conveniently separated from the media within several seconds using an external magnetic field, and the catalytic activity was still viable even after 10 removal cycles. The good catalytic performance of the composites could be attributed to synergy between the functions of the porous carbon support and the Fe3O4 nanoparticles embedded in the carrier. This work indicates that porous carbon spheres provide good support for the development of a highly efficient heterogeneous Fenton catalyst useful for environmental pollution cleanup.

Image classification is one of the predominant tasks in computer vision. So far, there are many approaches in image classification, and the most typical methods are Convolutional Neural Networks (CNN), BOF-based algorithms, etc. Most of these methods have a good performance, but there are still some limitations. Capsule Network (CapsNet) is the most advanced algorithm, which realizes the operation based on active vector and dynamic routing, and can overcome limitations of the original algorithm. This paper attempts to apply CapsNet into image classification as well as another two efficient classification methods, which are CNN and Fully Convolutional Network (FCN). We use two datasets: MNIST and CIFAR-10 to train our model and tested the networks. Finally, compare and evaluate their performances in aspects of time cost, loss, accuracy and the number of parameters.

No abstract

Here we report the formation of giant {Mo 126 W 30 } polyoxometalate cages templated by π-conjugated planar ligands. Their inorganic shells incorporate arrays of 18 stacked aromatic carboxylates, such as 1,3,5-benzenetricarboxylate or 5-nitroisophthalate, and the assembly is stabilized by multiple π-π interactions. NMR data confirmed that the cages are stable in solution and the inner aromatic templates can be postsynthetically replaced by selected aliphatic dicarboxylates, paving the way for endo-functionalization and exploration of the reactivities within the cage cavities.