Although the coronavirus disease (COVID-19) emergency status is easing, the COVID-19 pandemic continues to affect healthcare systems globally. It is crucial to have a reliable and population-wide prediction tool for estimating COVID-19-induced hospital admissions. We evaluated the feasibility of using wastewater-based epidemiology (WBE) to predict COVID-19-induced weekly new hospitalizations in 159 counties across 45 states in the United States of America (USA), covering a population of nearly 100 million. Using county-level weekly wastewater surveillance data (over 20 months), WBE-based models were established through the random forest algorithm. WBE-based models accurately predicted the county-level weekly new admissions, allowing a preparation window of 1-4 weeks. In real applications, periodically updated WBE-based models showed good accuracy and transferability, with mean absolute error within 4-6 patients/100k population for upcoming weekly new hospitalization numbers. Our study demonstrated the potential of using WBE as an effective method to provide early warnings for healthcare systems.

Vehicle-mounted ground-penetrating radar (GPR) has been used to non-destructively inspect and evaluate railway subgrade conditions. However, existing GPR data processing and interpretation methods mostly rely on time-consuming manual interpretation, and limited studies have applied machine learning methods. GPR data are complex, high-dimensional, and redundant, in particular with non-negligible noises, for which traditional machine learning methods are not effective when applied to GPR data processing and interpretation. To solve this problem, deep learning is more suitable to process large amounts of training data, as well as to perform better data interpretation. In this study, we proposed a novel deep learning method to process GPR data, the CRNN network, which combines convolutional neural networks (CNN) and recurrent neural networks (RNN). The CNN processes raw GPR waveform data from signal channels, and the RNN processes features from multiple channels. The results show that the CRNN network achieves a higher precision at 83.4%, with a recall of 77.3%. Compared to the traditional machine learning method, the CRNN is 5.2 times faster and has a smaller size of 2.6 MB (traditional machine learning method: 104.0 MB). Our research output has demonstrated that the developed deep learning method improves the efficiency and accuracy of railway subgrade condition evaluation.

Exploring more efficient and low-cost electrocatalysts to replace platinum (Pt) is highly desired to promote the practical hydrogen production through water splitting. Herein, a facile and effective strategy is proposed to fabricate self-standing Al₃Ni₂/Ni electrode with controlled phase composition and surface morphology, which is obtained by one-step electrochemical reduction of Al³⁺ on commercially available nickel in eutectic NaCl-KCl melt. Different from previously reported approaches, uniform Al₃Ni₂ monolith catalyst can directly grow onto Ni substrate. The deposit possesses unique three-dimensional (3D) cauliflower-like morphology comprising of nano- and microparticles due to the rapid nucleation rate during molten salt electrolysis. The as-fabricated Al₃Ni₂/Ni electrode can be directly used as the cathode to catalyze Hydrogen evolution reaction (HER). Impressively, it exhibits remarkable HER activity comparable to commercial Pt, including a low overpotential of 83.4 mV for a current density of 10 mA cm⁻², a small Tafel slope of 40.7 mV dec^-1, and excellent long-term stability over 36 h of continuous HER operation in 0.5 M H₂SO₄ solution. The intrinsic catalytic ability of Al₃Ni₂ with the unique hierarchical structure of nano/microsized grains can offer multiple effects, including massive exposed active sites, enhanced charge transfer and mass transport, and fast gas releasing that synergistically contribute to improving the electrocatalytic performance of HER. This work represents a highly promising approach to the design and one-step controllable fabrication of efficient and self-standing base metal electrode for electrocatalytic hydrogen production.

Waste aluminate phosphor is a valuable secondary resource of rare earth elements (REEs). However, Ce and Tb in aluminate green phosphor can hardly be extracted by direct leaching in an inorganic acid. Therefore, Na₂CO₃ assisted roasting is adopted to decompose the stable spinel structure of Ce_0.67Tb_0.33MgAl₁₁O₁₉ in the present work and to achieve the transformation of REEs to simple oxides. Based on the thermodynamic calculations, systematic experiments of thermal decomposition have been conducted. The thermal decomposition behavior, phase evolution, valence state change, variations in micro and macro morphology of the green phosphor during Na₂CO₃ assisted roasting were examined by using TG-DSC/MS, XRD, XPS, SEM/EDS analyses. The results indicated that the green phosphor began to react with solid Na₂CO₃ at 800 °C, and the reaction was dramatically accelerated with temperature rising above 851 °C. At about 1000 °C, Ce_0.67Tb_0.33MgAl₁₁O₁₉ could completely decomposed into CeO₂, Tb₂O₃ and MgO by roasting in an equivalent mass of Na₂CO₃ for 2 h, while α-Al₂O₃ was hardly attacked in roasting. The decomposition mechanism of Ce_0.67Tb_0.33MgAl₁₁O₁₉ in molten Na₂CO₃ could be depicted by the unreacted shrinking core model, and the reaction rate constant was estimated at approximately nanometers per second. The synergistic effect of cation-oxoanion ensures the successful extraction of CeO₂ and Tb₂O₃ from the green phosphor via Na₂CO₃ assisted roasting method. The converted CeO₂ and Tb₂O₃ can be extracted by using chlorination roasting and separated from non-REE residues. According to these investigations, a new efficient process technology is proposed for sustainable recycling of waste phosphor.

The electrochemical behavior of yttrium and its co-deposition with aluminum were investigated by several transient electrochemical techniques on a tungsten electrode at 973K in NaCl-KCl eutectic melts. The results reveal that the reduction of Y(III) in NaCl-KCl-YCl₃ melts is a one-step process with three-electron exchanged and the reaction is a quasi-reversible diffusion-controlled process at low scan rates (0.05∼0.5 V/s). The calculated diffusion coefficient is approximately 2.8 × 10−⁵cm²/s. After AlCl₃ was introduced into the melts, cyclic voltammetry and open circuit chronopotentiometry showed the formation of two Y-Al intermetallic compounds, indicating that under-potential deposition of yttrium occurred on tungsten electrode covered with liquid Al. The electromotive force was measured at 973K to determine the thermodynamic properties of Y-Al intermetallic compounds, such as the activity of Y in the two-phase coexistence state, relative partial molar Gibbs energies, as well as the standard Gibbs energies of Y-Al intermetallic compounds. Finally, potentiostatic electrolysis was conducted to prepare Y-Al alloys from molten NaCl-KCl-YCl₃ (1.5 mol%)-AlCl3 (1.5 mol%) by the co-reduction method. The cathodic alloys were characterized using X-ray diffraction (XRD) and scan electron micrograph (SEM)-energy dispersive spectrometry (EDS) and the results indicated that the obtained alloys were mainly composed of YAl₂, as well as YAl₃ and YAl phases. The Y-rich phase intermetallic compound YAl, formed in the later period of electrolysis just when the concentration of AlCl₃ is fairly low.

The electrochemical behavior of Nd(III) and its coreduction with Mg(II) were investigated on a molybdenum electrode at 1023 K in eutectic NaCl-KCl melts. The results indicate that the reduction of NdCl₃ in NaCl-KCl melts is a one-step process with three electrons exchanged, and the reaction is an irreversible diffusion-controlled process at low scan rate with the calculated diffusion coefficient of about 6.8 × 10^-5 cm² s^-1. After MgCl₂ is introduced to the melts, the reduction of Nd(III) takes place at a more positive potential value due to the formation of Mg-Nd intermetallic compounds through electrochemical co-deposition and chemical reduction of Nd(III) ions by preferentially deposited Mg. The solid intermetallic compounds of Mg₃Nd, Mg₂Nd, and MgNd observed in open circuit chronopotentiometry curves are transformed to thermodynamically more stable Mg-rich phases of Mg₁₂Nd and Mg₄₁Nd₅ when potentiostatic electrolysis at -2.10 V. It has been confirmed by X-ray diffraction and SEM-EDS microscopy that the cathodic deposits are composed of Mg, Mg₁₂Nd, Mg₄₁Nd₅, and Nd. The present results confirm that it is an effective method for recycling of neodymium from waste NdFeB magnets by means of electrochemical formation of Mg-Nd alloys from NaCl-KCl-MgCl₂-NdCl₃ melts.

Using eight years (2006-2014) of passive (MODIS/Aqua and OMI/Aura) and active (CALIOP/CALIPSO) satellite measurements of aerosols, we yield a three-dimensional (3D) distribution of the frequency of occurrence (FoO) of aerosols over China. As an indicator of the vertical heterogeneity of aerosol layers detected by CALIOP, two types of Most Probable Height (MPH), including MPH_FoO and MPH_AOD, are deduced. The FoO of "Total Aerosol" reveals significant geographical dependence. Eastern China showed much stronger aerosol FoD than northwestern China. The FoO vertical structures of aerosol layer are strongly dependent on altitudes. Among the eight typical ROIs analyzed, aerosol layers over the Gobi Desert have the largest occurrence probability located at an altitude as high as 2.83 km, as compared to 1.26 km over Beijing-Tianjin-Hebei. The diurnal variation (nighttime-daytime) in MPH_AOD varies from an altitude as low as 0.07 km over the Sichuan basin to 0.27 km over the Gobi Desert, whereas the magnitude of the diurnal variation in terms of MPH_AOD is six times as large as the MPH_FoO, mostly attributable to the day/night lidar SNR difference. Also, the 3D distribution of dust and smoke aerosols was presented. The multi-sensor synergized 3D observations of dust aerosols, frequently observed in the zonal belt of 38°N-45°N, is markedly different from that of smoke aerosols that are predominantly located in the eastern and southern parts. The 3D FoO distribution of dust indicates a west-to-east passageway of dust originating from the westernmost Taklimakan Desert all the way to North China Plain (NCP). The findings from the multi-sensor synergetic observations greatly improved our understanding on the long-range aerosol dispersion, transport and passageway over China.