Synthesis of EU-1/ZSM-48 Co-Crystalline Zeolites from High-Silica EU-1 Seeds: Tailoring Phase Proportions and Promoting Long Crystalline-Phase Stability.

A facile, specific, seed-assisted strategy for the synthesis of EU-1/ZSM-48 co-crystalline zeolites in the presence of hexamethonium ions (HM2+ ) has been developed. EU-1/ZSM-48 co-crystalline zeolites with various phase proportions, with EU-1 in the range of 25 wt %-86 wt %, were obtained by adding high-silica EU-1 seeds (SiO2 /Al2 O3 ratio of 300) and adjusting the synthesis parameters. Not only can the phase proportions of EU-1/ZSM-48 co-crystalline zeolites be controlled, but also the stability period for co-crystallization of the two phases can be extended through varying the amount of EU-1 seeds and the HM2+ template. Moreover, with the increase of the EU-1 proportion in the EU-1/ZSM-48 co-crystalline, the framework SiO2 /Al2 O3 ratios of EU-1 phase promotes steadily. Major differences in acidity and textural properties of the EU-1/ZSM-48 co-crystalline zeolites (Coz) were found with varying phase proportions, due to their distinct topological structures, crystal morphology and asymmetry between the EU-1 and ZSM-48 phases. For instance, the EU-1/ZSM-48 zeolite containing 75 wt % of EU-1 (Coz-75) possesses specific acidity and mesoporous characteristics, showing an excellent catalytic activity and stability in n-hexane cracking reaction. Compared to EU-1, ZSM-48, and a mechanical mix of the two zeolites (Mix-75), Coz-75 resulted in the highest hexane conversion and yields of light olefins, with a propylene yield, in particular, up to 38.3 wt %, which is 6.3 % more than that of the Mix-75 sample.

Study on the Nodal Composite Bearing Performance of Nontruncated PHC Pipe Pile and Bearing Platform.

In this paper, low circumferential reciprocating load foot-scale tests were performed on two nontruncated PHC B 600 130 tubular piles with bearing nodes to characterize the damage process and morphology of the specimens and to investigate the load-carrying performance of the members. The test results reveal that under the action of tensile-bending-shear loading, the bearing concrete in the node area buckles and is damaged, the anchored reinforcement in the node area yields, the constraint is weakened, an articulation point is formed, and the node rotational capacity increases. When the embedment depth increases from 200 mm to 300 mm, the ultimate bearing capacities of the positive and negative nodes increase by 31.04% and 36.16%, respectively. A numerical simulation is used to verify the test results. Considering the four types of piles without truncated nodes, the numerical simulation is used to analyze the node-bearing capacity at different embedment depths. Finally, a preferred node type is proposed as follows: a terminal plate welded anchor bar and pipe pile core-filled longitudinal reinforcement anchored into the bearing node, with a preferred embedment depth of 250 mm.

Efficient EEG Feature Learning Model Combining Random Convolutional Kernel with Wavelet Scattering for Seizure Detection.

Automatic seizure detection has significant value in epilepsy diagnosis and treatment. Although a variety of deep learning models have been proposed to automatically learn electroencephalography (EEG) features for seizure detection, the generalization performance and computational burden of such deep models remain the bottleneck of practical application. In this study, a novel lightweight model based on random convolutional kernel transform (ROCKET) is developed for EEG feature learning for seizure detection. Specifically, random convolutional kernels are embedded into the structure of a wavelet scattering network instead of original wavelet transform convolutions. Then the significant EEG features are selected from the scattering coefficients and convolutional outputs by analysis of variance (ANOVA) and minimum redundancy-maximum relevance (MRMR) methods. This model not only preserves the merits of the fast-training process from ROCKET, but also provides insight into seizure detection by retaining only the helpful channels. The extreme gradient boosting (XGboost) classifier was combined with this EEG feature learning model to build a comprehensive seizure detection system that achieved promising epoch-based results, with over 90% of both sensitivity and specificity on the scalp and intracranial EEG databases. The experimental comparisons showed that the proposed method outperformed other state-of-the-art methods for cross-patient and patient-specific seizure detection.