Search

Employing the solid polymer electrolyte (SPE) instead of traditional liquid electrolyte is an effective way to develop high safety and high-energy density solid-state lithium batteries. Herein, for the first time, the Ni3B2O3 (NBO) nanorods are incorporated into polyethylene oxide (PEO)-based SPE. Particularly, the optimized NBO-embedded SPE shows a high ionic conductivity of 8.5 × 10−5 S cm−1 at 30 °C, lowering the operating temperature of PEO-based SPE substantially. The corresponding LiFePO4/Li battery demonstrates a high discharge capacity of 154 mAh g−1 after 80 cycles at 0.2 C under 30 °C, with favorable capacity retention of 97.5%. The remarkable properties are attributed to the high ionic conductivity of modified SPE at ambient temperature, which is resulted from the decreased crystallinity and melting transition point, increased movement of PEO chain, and promotion of lithium salt dissociation, as well as the formation of the lithium ion migrating pathway on the interface between PEO and NBO nanorods.

Omics software tools have reshaped the landscape of modern biology and become an essential component of biomedical research. The increasing dependence of biomedical scientists on these powerful tools creates a need for easier installation and greater usability. Packaging, virtualization, and containerization are different approaches to satisfy this need by wrapping omics tools in additional software that makes the omics tools easier to install and use. Here, we systematically review practices across prominent packaging, virtualization, and containerization platforms. We outline the challenges, advantages, and limitations of each approach and some of the most widely used platforms from the perspectives of users, software developers, and system administrators. We also propose principles to make packaging, virtualization, and containerization of omics software more sustainable and robust to increase the reproducibility of biomedical and life science research., arXiv

Presentation:Addressing bias in big data and AI for healthcare: a call for open science

In this study, a material-oriented metallurgical strategy, which is of such characteristics as short process, energy conservation and cost saving, is proposed to treat high-grade nickel matte and simultaneously produce LiNi0.80Co0.15Al0.05O2 cathode material for lithium ion batteries. This process is mainly composed of four steps, which is oxidation roasting, two-stage leaching, purification and material synthesis. First, high-grade nickel matte is roasted to eliminate sulfur under air atmosphere. Then the roasted product is first leached to selectively remove 78.10% Cu with little Ni and Co extraction at atmospheric temperature under optimum conditions (25 °C, HCl 1.50 mol L−1, L/S 7.0, 30 min). After that, 96.23% Ni, 96.58% Cu and 84.18% Co are extracted in thermal leaching process under optimum conditions (70 °C, HCl 3.0 mol L−1, L/S 10.0, 120 min), leaving Fe, Zn and Ca in residue. The obtained lixivium is purified by solvent extraction method to adjust the mass ratio of Cu and Ni to the range that meets the requirement for preparation of cathode material. Finally, LiNi0.80Co0.15Al0.05O2 material is produced by spray pyrolysis and sintering. The electrochemical performance of the resulting material is comparable with that prepared from analytical reagents, indicating such process can realize the comprehensive utilization of high-grade nickel matte.

The treatment of lepidolite by a multistage leaching process was studied for the extraction of alkali metals as alkali alums and brines. The precipitation characteristics of the aluminum and fluorine in the solution were also investigated. The transformation of the mineralogical species in lepidolite into albite and silicon dioxide was tentatively identified by X-ray diffraction measurements. The crystallization was aimed at the separation of rubidium and cesium from an acidic solution containing potassium as the main contaminant. The alum mixture contained on average 7.5% potassium, 1.3% rubidium and 0.30% cesium. The main crystalline components were aluminum sulfate and AlF(SO4)·5H2O after concentration. The aluminum in the solution was removed by adding potassium sulfate with a K/Al mole ratio of 2 at various pHs to yield an overall aluminum precipitation of 62.1–66.6%. Under weakly acidic and neutral conditions, soluble aluminum may transform into insoluble aluminum, such as Al(OH)3. Additionally, soluble AlFx(3−x)+ species tended to precipitate directly with OH−, thereby achieving the formation of Al-OH-F complexes. The AlFx(3−x)+ species at high pH were partially dissociated into F− and AlOH2+, Al(OH)2+ and Al(OH)4−, and the buffering effect of F− inhibited the transformation of the precipitated Al(OH)3 into soluble aluminates due to the formation of Al-OH-F complexes. The residual aluminum and fluorine contents were on average 0.0065 g/L and 0.042 g/L, respectively, in the mother liquor after neutralization at pH 9. The loss of lithium increased from 1.1% to 5.4% as the pH increased from 5 to 9. These findings imply that aluminum and fluorine can play significant roles in the recovery of alkali metals from lepidolite, and the precipitation of aluminum from leaching solutions as alums can cause a loss of lithium.

Infinite volume expansion and high reactivity severely hinder the commercial applications of lithium metal batteries. Moreover, the short circuit caused by Li dendrites is fatal for the long-term performance of lithium metal batteries and may cause safety problems. Herein, lithiophilic silver/lithium (Ag/Li) composite anodes are synthesized via a spontaneous displacement reaction between silver nitrate and Li foil at room temperature, which can settle the above challenges by regulating Li nucleation and homogenizing Li-ion flux. Benefiting from the lithiophilic design, Li ion realizes zero nucleation overpotential on the as-obtained Ag/Li anodes with abundant spherical silver nanoparticles. Additionally, the by-product, lithium nitrate, acts as an electrolyte additive, which further ensures uniform Li deposition. Consequently, Ag/Li symmetric cells demonstrate ultralong lifespan with repeated plating/stripping for 1000 cycles at the current densities of 1 mA cm−2 and 5 mA cm−2, with negligible voltage fluctuations. Furthermore, the full-cell devices, paired with LiFePO4, also exhibit excellent electrochemical performance. The facile and effective strategy provides a promising prospect for commercial applications of lithium metal batteries.

Tin oxides as some of the potential candidates for commercial graphite anodes suffer from serious drawbacks such as poor electronic conductivity and large volume variation during cycling. Dispersing nanostructured SnOx into a carbon matrix may be an effective strategy to solve these problems. In this work, a novel dried plum-like yolk–shell SnOx/C architecture with high specific surface area and superior structural stability is constructed by an extremely facile and scalable spray pyrolysis method in ten seconds. The SnOx nanodots (

Summary Low-cost, scalable energy storage is the key to continuing growth of renewable energy technologies. Here a battery with sedimentary slurry electrode (SSE) is proposed. Through the conversion of discrete particles between sedimentary and suspending types, it not only inherits the advantages of semi-solid flow cell but also exhibits high energy density and stable conductive network. Given an example, the zinc SSE (ZSSE) delivers a large discharge capacity of 479.2 mAh g−1 at 10 mA cm−2. More importantly, by renewal of the slurry per 20 cycles, it can run for 112 and 75 cycles before falling below 80% of designed capacity under 10 mA cm−2 (20% DODZn) and 25 mA cm−2 (25% DODZn), respectively. The lost capacity after cycles is able to recover after slurry renewal and the end-of-life SSE can be easily reused by re-formation. The concept of SSE brands a new way for electrochemical energy storage., Graphical Abstract, Highlights • A renewable semi-solid sedimentary slurry battery is proposed • Lost capacity after cycles is able to recover after slurry renewal • End-of-life SSE can be easily reused by re-formation, Energy Materials; Energy Storage; Energy Systems; Materials Characterization

The Nb-doped Li[Li0.2Mn0.54-x/3Ni0.13-x/3Co0.13-x/3Nbx]O2 (x = 0, 0.01, 0.02, 0.03) materials are synthesized and characterized by X-ray diffractometry (XRD), transmission electron microscope (TEM), scanning electron microscopy (SEM), charge–discharge measurements, AC impedance spectroscopy as well as cyclic voltammetry. XRD results indicate that Nb-doped Li1.2Ni0.13Co0.13Mn0.54O2 materials exhibit similar XRD patterns to the pristine sample. The results of EDS mapping confirm the homogenous distribution of Nb element in the materials. The Nb substituted for transition metal ions with x = 0.02 is found to be the most effective in improving the cycling properties and in lowering the voltage fade. Particularly, it exhibits remarkably improved cycling performance with the capacity retention ratio of 83% compared with 54% for pristine one after 100 cycles at 1C. Besides, it shows much lower discharge midpoint voltage decay (0.1913 V) than that of pristine sample (0.6951 V) after 100 cycles. The improved electrochemical performance can be attributed to the better structure stability and decreased charge transfer resistance caused by dopant of Nb. These results reveal that Nb-doping can be a promising method for improving the performance of Li-rich material.

Even though Li-rich layered oxide is considered to be a promising cathode material for Li-ion batteries because of the high discharge capacity, it still suffers from many fundamental challenges such as low initial coulombic efficiency, the voltage fading upon cycling and poor rate performance. In this work, lithium-rich layered Li[Li0.2Ni0.13Co0.13Mn0.54]O2 cathode materials are pretreated by NH4F through a facile solid state method to overcome these problems. After pretreating, the spinel phase LiM2O4 coating layer is formed accompanying with the formation of oxygen vacancies caused by NH3 decomposed from NH4F. Compared with the pristine material, materials pretreated by appropriate amount of NH4F present significantly improved electrochemical performances such as high initial coulombic efficiency and improved stability in successive electrochemical cycling, excellent rate capability and suppressed voltage decay. The pretreated material delivers the highest initial coulombic efficiency of 102.2% (NF10) compared to 74.5% of NF0 and the highest voltage retention of 90.6% compared to 86.2% of NF0. The NF0 delivers the best comprehensive performance, with higher initial coulombic efficiency and voltage retention, improved rate performance and capacity stability.

To suppress the capacity fade of Li-rich Li1.2Ni0.13Co0.13Mn0.54O2 material as cathode materials for lithium-ion battery, we introduce a LiF coating layer on the surface to improve the cycling performance of Li1.2Ni0.13Co0.13Mn0.54O2 material. The modified sample shows a capacity of 163.2 mAh g−1 with a capacity retention of 95% after 100 cycles at a current density of 250 mA g−1, while the pristine sample only delivers a capacity of 129.9 mAh g−1 with a capacity retention of 82%. Compared with the pristine material, the LiF-modified sample exhibits an obvious enhancement in the electrochemical performance, which will be very beneficial for this material to be commercialized on the new energy vehicles and other related areas.

In this paper, SiOx@TiO2/C fibers with TiO2/C composite layer were successfully prepared by one-step electrospinning. Carbon coating can enhance conductivity and act as bulk expansion buffer matrix, while TiO2 coating is beneficial to improving the mechanical stability of composite. Thus, a highly effective and stable conductive network is formed. Benefited from the protection of TiO2/C coating layer, SiOx@TiO2/C fibers demonstrates enhanced electrochemical performance. It exhibites an initial charge capacity of 1125.3 mAh g−1 at 0.1 A g−1, and a stable charge capacity of 855.0 mAh g−1 can be retained with a retention of 89.5% after 100 cycles at 0.4 A g−1, which is better than that of M-SiOx (32.3%) and SiOx@C (52.5%) materials. Moreover, SiOx@TiO2/C fibers show a high charge capacity of 640.4 mAh g−1 after 200 cycles cycled at a high current density of 1 A g−1.

In this work, Sn is used to dope the Li-rich cathode material to improve the electrochemical performance of Li ion battery. After Sn treatment, the lattice parameters a , c and lattice volume V become larger. Compared with the pristine sample, the Sn-contained samples show longer plateaux at about 4.5 V in the first charging process, which means that Sn can activate the Li 2 MnO 3 component. Meanwhile, with appropriate content of Sn doping, the sample exhibits enhanced rate capability and cycling stability. Especially, the sample S10 shows the best electrochemical performance, with a capacity retention of 88.66% after 100 cycles at 1 C (1 C=250 mA g −1 ). The mechanisms of Sn doping have also been investigated. The increased activation of Li 2 MnO 3 is due to the improved conductivity of Li 2 MnO 3 phase by Sn doping, and the enhanced electrochemical performance is mainly ascribed to the increased ability of Li ion diffusing into bulk phase and the improved structure stability during the prolonged charge-discharge cycles. It is suggested that Sn doping is an effective way to improve the electrochemical performance of Li-rich cathode material.

Single-crystal nickel-rich cathode materials (SC-NRCMs) are the most promising candidates for next-generation power batteries which enable longer driving range and reliable safety. In this review, the outstanding advantages of SC-NRCMs are discussed systematically in aspects of structural and thermal stabilities. Particularly, the intergranular-crack-free morphology exhibits superior cycling performance and negligible parasitic reactions even under severe conditions. Besides, various synthetic methods are summarized and the relation between precursor, sintering process, and final single-crystal products are revealed, providing a full view of synthetic methods. Then, challenges of SC-NRCMs in fields of kinetics of lithium diffusion and the one particularly occurred at high voltage (intragranular cracks and aggravated parasitic reactions) are discussed. The corresponding mechanism and modifications are also referred. Through this review, it is aimed to highlight the magical morphology of SC-NRCMs for application perspective and provide a reference for following researchers.

The composite of Co 3 O 4 , Co and graphitized carbon is synthesized by carbonization and cobalt-catalyzed-graphitization of carboxymethyl chitosan, followed by low-temperature thermal oxidation. At the high temperature, the Co(+2) is reduced to metallic Co and the produced Co acts as a catalyst for the graphitization of the pyrolytic carbon. The low-temperature oxidation is conducted to selectively oxidize Co to Co 3 O 4 while the carbon remains unreacted. The structure analysis indicates that three crystal phases (graphite, metallic Co, Co 3 O 4 ) accompanied with amorphous carbon are co-existed in the composite. Morphological results demonstrate that a lot of graphite grains around the Co element are distributed in the carbon. The electrochemical testing results indicate that the composite shows good electrochemical performance. It delivers a reversible capacity of 843 mAh g −1 at low current density, and remains 88.9% after 60 cycles at 200 mA g −1 . Even performed at 1 A g −1 , it also exhibits 493 mAh g −1 . The performance improvement is mainly due to the high capacity of Co 3 O 4 , high conductivity of graphitic carbon and metallic cobalt, the porous structure offering enough ion transport pathways and relieving the strain during cycling.

The major obstacle for the application of high-voltage cathode materials is the anodic instability of the electrolyte. On the guidance of density functional theory calculation, we develop a new high-voltage electrolyte comprising 1 mol L−1 LiPF6 dissolved in fluoroethylene carbonate (FEC) and ethyl difluoroacetate (DFEAc) (FEC/DFEAc = 3:7 in wt.% ratio). It shows a high conductivity (9.48 mS cm−1) and high anodic stability compared with the conventional electrolyte of 1 mol L−1 LiPF6 dissolved in ethylene carbonate (EC) and ethyl-methyl carbonate (EMC) (EC/EMC = 3:7 in wt.% ratio). In addition, Li/LiNi0.5Co0.2Mn0.3O2 half-cells with the new electrolyte display an excellent cycling ability with capacity retention of 89.23% after 100 cycles at 4.6 V (vs. Li/Li+). Although the electrolyte still will be oxidized at the cathode surface, the Li2CO3 and carbon-fluoride species originated from DFEAc and FEC are beneficial to building a stable cathode/electrolyte interface as revealed by the TEM and XPS results.

In this work, we propose a novel system for smart copy-paste, enabling the synthesis of high-quality results given a masked source image content and a target image context as input. Our system naturally resolves both shading and geometric inconsistencies between source and target image, resulting in a merged result image that features the content from the pasted source image, seamlessly pasted into the target context. Our framework is based on a novel training image transformation procedure that allows to train a deep convolutional neural network end-to-end to automatically learn a representation that is suitable for copy-pasting. Our training procedure works with any image dataset without additional information such as labels, and we demonstrate the effectiveness of our system on two popular datasets, high-resolution face images and the more complex Cityscapes dataset. Our technique outperforms the current state of the art on face images, and we show promising results on the Cityscapes dataset, demonstrating that our system generalizes to much higher resolution than the training data., 12 pages, 9 figures

In this thesis, our focus is learning a controllable representation and applying the learned controllable feature representation on images synthesis, video generation, and even 3D reconstruction. We propose different methods to disentangle the feature representation in neural network and analyze the challenges in disentanglement such as reference ambiguity and shortcut problem when using the weak label. We use the disentangled feature representation to transfer attributes between images such as exchanging hairstyle between two face images. Furthermore, we study the problem of how another type of feature, sketch, works in a neural network. The sketch can provide shape and contour of an object such as the silhouette of the side-view face. We leverage the silhouette constraint to improve the 3D face reconstruction from 2D images. The sketch can also provide the moving directions of one object, thus we investigate how one can manipulate the object to follow the trajectory provided by a user sketch. We propose a method to automatically generate video clips from a single image input using the sketch as motion and trajectory guidance to animate the object in that image. We demonstrate the efficiency of our approaches on several synthetic and real datasets.

Nanoparticles-assembled Co 3 O 4 microspheres with different primary particle size are synthesized via a facile one-step ultrasonic spray pyrolysis process, using CoCl 2 ·6H 2 O as precursor. The as-synthesized Co 3 O 4 powders are directly used as anode material for Li-ion batteries without further calcinations. The primary particle size is tuned by controlling the spray pyrolysis temperature and its effects on the electrochemical performance of the prepared samples are systemically investigated. The results reveal that primary particles size exerts significant effect on the initial coulombic efficiency, specific capacity and rate performance. In addition, we find that the temperature of powder collector strongly influences the chlorine content in the products, which is resulted from the chloridization of Co 3 O 4 by gaseous HCl from the off-gas. A simple and effective strategy is proposed to prevent the chloridization reaction. The sample with the smallest primary particle size and low Cl content shows the highest initial coulombic efficiency of 73% and a high reversible capacity of 1340 mA h g −1 after 50 cycles at a current density of 200 mA g −1 .

Hierarchical mesoporous hybrid NiO–MnCo2O4 microspheres are synthesized by a robust one-pot spray pyrolysis process. The mesoporous hybrid NiO–MnCo2O4 microspheres are comprised of homogeneously dispersed nanoscaled NiO and MnCo2O4 subunits. Specific surface area of the mesoporous hybrid NiO–MnCo2O4 microspheres is determined to be 10.2 m2 g−1 with dominant pore size of 25.8 nm. The as-obtained material is evaluated as anode material for LIBs. The as-prepared hybrid NiO–MnCo2O4 sample demonstrate better electrochemical performances, including higher reversible capacity and superior rate capability, compared with the single-phase NiO and MnCo2O4 samples. These superior electrochemical performances should be mainly attributed to its porous microstructure characteristics and the synergistic effects between the well-dispersed NiO and MnCo2O4 nanophases. The strategy is simple and flexible, so it could be readily generalized to construct other hybrid metal oxides materials.

Li[Li 0.2 Mn 0.54 Ni 0.13 Co 0.13 ]O 2 materials doped with different Zr contents are studied using X-ray diffractometry (XRD), energy dispersive X-ray analysis (EDX), scanning electron microscopy (SEM), charge–discharge measurements, AC impedance spectroscopy as well as cyclic voltammetry. These results demonstrate that the element Zr is distributed uniformly in these materials. With the increase of Zr content, the lattice parameters a , c and lattice volume V become larger. The discharge cycling curves of cells show that Zr doping enhances cycle life compared with the undoped one. The sample with x =0.01 shows the highest capacity and best cycling performance, especially at high current density. The Li[Li 0.2 Mn 0.54− x /3 Ni 0.13− x /3 Co 0.13− x /3 Zr x ]O 2 ( x =0.01) delivers a capacity of 133.6 mA h g −1 (after 100 cycles) at the current density of 250 mA g −1 , corresponding to the capacity retention of 83.8%. The favorable performance of the doped samples could be attributed to its enhanced structural stability and Li + mobility.

The x LiVPO 4 F– y Li 3 V 2 (PO 4 ) 3 composite cathode materials have been synthesized through a soft chemical route with mechanical activation assistance followed by low-temperature calcination. Structural analysis indicates that two phases, i.e. monoclinic Li 3 V 2 (PO 4 ) 3 and triclinic LiVPO 4 F coexist in the composites. With the addition of LVP, the average particle size becomes smaller. The prepared composite materials show good electrochemical performance, and especially a good cyclic performance, the most suitable x / y ratio being 1:5 where the sample can deliver a reversible capacity of 139 mA h g −1 .

Lithium bis(fluorosulfony)imide (LiFSI) is a promising alternative lithium salt to replace lithium hexafluorophosphate (LiPF6) due to its high conductivity and excellent compatibility with electrode material. On the other hand, the aluminum corrosion caused by LiFSI hinders its application in lithium ion battery. To solve this problem, lithium difluoro(oxalato)borate (LiDFOB) is added to suppress the aluminum corrosion in LiFSI-based nonaqueous carbonate electrolyte. The electrochemical tests in three-electrode cells and graphite/LiCoO2 full cells confirm that the addition of LiDFOB is beneficial to suppress the aluminum corrosion. In addition, the mechanism is proposed that the oxidation products of LiDFOB form a passivating film at the aluminum surface to suppress the corrosion based on the experimental results from SEM and XPS tests.

Li metal anode is one of the most promising anodes for next-generation high-energy-density batteries. However, some lethal challenges, such as Li dendrite, inferior coulombic efficiency, and infinite volume change during repeated Li plating/stripping restrict its practical application. Although carbon-based materials are ideal hosts for Li deposition, unsatisfied lithiophilic property and vulnerable solid electrolyte interphase (SEI) film still remain unsolved. Herein, we report the fluorinated mesoporous carbon (FMC) nanosheets derived from spent cathodic carbon of aluminum electrolysis as a versatile dendrite-free current collector. Two types of C–F bonds are discovered in FMC. One is C(sp2)-F, which serves as lithiophilic site for Li nucleation with reduced barrier. The other is C(sp3)-F, which breaks to produce extra LiF during Li deposition, facilitating the formation of stable LiF-rich SEI film. The synergistically designed Li@FMC|LiFePO4 full cells demonstrate improved cycling performance with high coulombic efficiency. This work provides possibility for direct utilization of waste electro-carbon in energy storage application.

Synthesizing spinel/layer Li-rich materials (SLLRs) is a promising way to improve the electrochemical performance of Li-rich material. In this work, the influence of introducing spinel phase into layered one on the coulombic efficiency is studied by electrochemical method. During CV testing, the composite exhibits no oxidation peak around 2.5–3.0 V when it starts with positive scanning, while demonstrates a reduction peak at ∼2.57 V as it begins with negative scanning. This means the spinel phase formed during the synthetic process is lithium-deficient, consuming lithium from lithium metal anode during discharge in Li-half cell. Charge-discharge results show that this spinel phase contributes to extra capacities when cells are discharged below 3.0 V, leading to a higher initial coulombic efficiency (ICE) of SLLRs in half cells. However, when tested in full cell using graphite as anode, ICE of SLLRs equals to that of Li-rich material while the capacity is lower. This is because that the graphite anode in full cell is unable to provide extra lithium ions and to compensate lithium deficiency of cathode material. Strategy like electrochemical pre-lithiation is suggested to improve the electrochemical performance of SLLRs in full cell.

Nonprecious bifunctional electrocatalysts with satisfactory cost and high catalytic performance toward oxygen reduction reaction (ORR) and oxygen evolution reaction (OER) are essential to promote the commercialization of metal-air batteries. Herein, 3D honeycombed hierarchical porous N, O-doping carbon (HHPC) is fabricated via a facile and scalable strategy of hard template-activator assisted pyrolysis. The obtained HHPC displays unique macro-meso-microporous framework, N, O-codoped effect and order-disorder hybrid characteristic, providing sufficient space for O2 and electrolyte transport, abundant defect sites, enhanced electronic conductivity and stability for oxygen conversion reaction. As such, the HHPC exhibits superior bifunctional electrocatalytic activity with a narrow ΔE value of 0.8 V. And a long-term stability and reversibility with a narrow voltage gap increasing of 0.09 V over 1165 cycles (∼388 h) at 10 mA cm−2 are achieved in the HHPC-based rechargeable Zn-air battery, outperforming most of the state-of-the-art carbon-based Zn-air batteries.

A two stage hydrochloric acid leaching process is proposed for the treatment of lepidolite and recovery of alkali metals. The separation characteristics of aluminum and fluoride in the leachate were also determined. The transformation of lepidolite into silica-like residue was tentatively identified through X-ray diffraction measurements. The leach liquor was heated and concentrated and by evaporation to drive off to remove some of the hydrogen chloride and hydrogen fluoride from the acidic leachate, followed by cooling to deposit crystals. This yielded an overall hydrogen chloride and hydrogen fluoride removal of 25.4% and 2.52%, respectively. The main components of the crystals were aluminum and potassium chlorides, with small amounts of fluoroaluminate, polyaluminum chloride, and aluminum hydroxyfluoride. The crystals were calcined at 623 K, with a further 43.9% of hydrogen chloride and 4.33% of hydrogen fluoride being driven off and captured by adsorption equipment, which was beneficial for the subsequent separation of impurities. Investigation of the major mineral phases revealed that lower temperature calcination of the crystals tended to produce less-soluble polyaluminum chloride, aluminum hydroxyfluoride, fluoaluminate, and alumina. Water leaching of crystals after calcination at 623 K gave a leach liquor with 7.74 g/L of lithium, 0.15 g/L of aluminum and 0.25 g/L of fluoride. The fluoaluminate in the leach liquor led to the production of secondary precipitates after neutralization. The [AlFn]3-n species were partially dissociated to F− and Al(OH)4− at a strongly basic pH. The residual aluminum and fluoride concentrations were 0.01 g/L and 0.05 g/L, respectively, at the pH of 9. Since alkali chlorides such as lithium chloride are very soluble, there was less lithium loss during neutralization than in the corresponding sulfate system.

We present a novel system for sketch-based face image editing, enabling users to edit images intuitively by sketching a few strokes on a region of interest. Our interface features tools to express a desired image manipulation by providing both geometry and color constraints as user-drawn strokes. As an alternative to the direct user input, our proposed system naturally supports a copy-paste mode, which allows users to edit a given image region by using parts of another exemplar image without the need of hand-drawn sketching at all. The proposed interface runs in real-time and facilitates an interactive and iterative workflow to quickly express the intended edits. Our system is based on a novel sketch domain and a convolutional neural network trained end-to-end to automatically learn to render image regions corresponding to the input strokes. To achieve high quality and semantically consistent results we train our neural network on two simultaneous tasks, namely image completion and image translation. To the best of our knowledge, we are the first to combine these two tasks in a unified framework for interactive image editing. Our results show that the proposed sketch domain, network architecture, and training procedure generalize well to real user input and enable high quality synthesis results without additional post-processing.

The cycling performance of graphite/LiNi 0.5 Co 0.2 Mn 0.3 O 2 battery in the electrolyte with different content (0, 0.5, 1.0, 2.0 wt.%) 1-propylphosphonic acid cyclic anhydride (PACA) is studied in this work. When 0.5 wt.% PACA additive is incorporated into the blank electrolyte, the capacity retention of graphite/LiNi 0.5 Co 0.2 Mn 0.3 O 2 battery at high voltage after 100 cycles is increased from 79.12% to 91.84%. Electrochemical impedance spectroscopy (EIS), X-ray diffraction (XRD), and X-ray photoelectron spectroscopy (XPS) results reveal the mechanism that the surface film derived from PACA can decrease the surface film impedance of the cell (from 40.63 to 16.57 Ω), stabilize the structure of cathode material. Moreover, the investigation of PACA on the cycling performance of graphite electrode demonstrates that PACA function on LiNi 0.5 Co 0.2 Mn 0.3 O 2 instead of graphite.

We study the problem of building models that disentangle independent factors of variation. Such models could be used to encode features that can efficiently be used for classification and to transfer attributes between different images in image synthesis. As data we use a weakly labeled training set. Our weak labels indicate what single factor has changed between two data samples, although the relative value of the change is unknown. This labeling is of particular interest as it may be readily available without annotation costs. To make use of weak labels we introduce an autoencoder model and train it through constraints on image pairs and triplets. We formally prove that without additional knowledge there is no guarantee that two images with the same factor of variation will be mapped to the same feature. We call this issue the reference ambiguity. Moreover, we show the role of the feature dimensionality and adversarial training. We demonstrate experimentally that the proposed model can successfully transfer attributes on several datasets, but show also cases when the reference ambiguity occurs., Comment: Submitted to ICLR 2018

In this paper we introduce silhouette constraints to improve the quality of unconstrained 3D face reconstruction. Previously, state of the art unconstrained 3D face reconstruction techniques relied solely on photometric consistency and matching sparse facial landmarks. We show that constraining the silhouettes of the 3D reconstruction to match silhouettes in the input images can further improve reconstruction quality. Our technique automatically detects silhouettes and iteratively matches silhouette points computed from the current 3D reconstruction with silhouettes in the input images. We demonstrate that our results improve on the previous state of the art in unconstrained 3D face reconstruction and that our additional constraints can easily be included in the iterative reconstruction at little additional cost.

Sulfation roasting followed by water leaching process was used to extract lithium from lepidolite. Various parameters including roasting temperature, the amount of additions, and solid/liquid ratio in leaching process were optimized. The lithium extraction efficiency of 91.61% could be reached with a mass ratio of lepidolite/Na 2 SO 4 /K 2 SO 4 /CaO of 1:0.5:0.1:0.1, and roasting at 850 °C for 0.5 h. XRD analysis showed that sulfation roasting caused the decompositions of the original aluminosilicate to NaSi 3 AlO 8 , KAlSi 2 O 6 and CaAl 2 Si 2 O 8 . The phases of CaF 2 and Ca 4 Si 2 O 7 F 2 are observed due to the addition of CaO.

A novel technique was developed to extract lithium from lepidolite. The lepidolite was pre-roasted at a high temperature with water steam atmosphere for defluorination. Then the defluorinated lepidolite was leached in a lime–milk autoclave. Various parameters including the defluorination percentage of lepidolite, milling time, temperature, time, lime-to-defluorinated lepidolite ratio, and liquid-to-solid ratio in the leaching process were optimized. The lithium extraction efficiency can reach 98.9% under the optimal conditions. The purity of the lithium carbonate obtained can be up to 99.9%.

Submicron LiNi0.6Co0.2Mn0.2O2 as a cathode material for lithium-ion batteries was prepared by a spray drying assisted solid-state method. The crystal structure, morphology and electrochemical properties of LiNi0.6Co0.2Mn0.2O2 were characterized by X-ray diffraction, scanning electron microscopy, transmission electron microscope, energy-dispersive spectroscopy and charge–discharge testing. Single-phase layered LiNi0.6Co0.2Mn0.2O2 was obtained. The material delivered a reversible charge/discharge capacity of 179.8 mAh g−1 at C/10, and showed excellent cycling performance and rate capability.

The effect of pre-roasting on leaching of the gamierite laterite ore, obtained from Yunnan province, China, was investigated in this study. The phase transformation of laterite minerals roasted at different temperatures was investigated with X-ray powder diffraction (XRD). The roasting experiment results show that there are two phase transformation processes of dehydroxylation of goethite and lizardite at roasting temperatures of 277 °C and 610 °C, respectively, which accord with the result of DTA–TG analysis. Pre-roasting of the laterite not only alters its mineralogical composition but also increases its porosity and surface area, thus making it more amenable to leaching. Compared to the leaching result of raw ore and ores roasted at different temperatures, it indicates that increasing roasting temperature up to 300 °C appears to provide the optimum nickel recovery and further heating appears to be detrimental to the nickel recovery.

In spent battery material, there are plenty of valuable metals, such as copper, nickel, cobalt, manganese. Recovery of valuable metals from spent battery material not only protects the environment but also improves the utilization of resources and decreases the cost of battery material. In this study, hydrochloric acid is used as lixivant with characteristics of faster leaching rate and being recycled easily. The optimal conditions are that hydrochloric acid concentration is 6 mol/L, reaction temperature is exactly 60 °C, liquid/solid ratio is 8:1, (H2O2)mol/(MeS)mol = 2, and the leaching time is 2 h, the results show that the dissolution yields of Ni, Co and Mn can be 95 wt.% at least. The basic purification concept of the leaching solution includes that copper is removed through replacement by iron powder followed by iron precipitation in goethite method. The results show that Cu and Fe can be removed 99 wt.% at the least. At the same time, the loss of Ni, Co and Mn is not beyond 2 wt.%, 3 wt.％ and 2 wt.%, respectively. This method makes the preparation of pure NixCoyMnz ternary system precursor economical. The process seems to be able to claim base metals from waste in a reliable and feasible way.

The LiMn2O4/graphite battery was fabricated and its 3 C/10 V overcharge performance was studied. Spinel LiMn2O4 was synthesized by solid-state method and 325680-type size full battery was fabricated. The structure and morphology of the powders were characterized by XRD and SEM technique, respectively. The battery explodes after 3 C/10 V overcharged test, and surface temperature of the battery case arrives at 290 °C in 12 s after exploding. Black air is given out with blast. Carbon, MnO, and Li2CO3 are observed in the exploded powders. The cathode electrode remains spinel structure with 5.0 V charged. Cracks in the cathode electrode particles are detected with the increase of voltage by SEM technique. The 5.0 V charged electrode can decompose into Mn3O4 at 400 °C. It is demonstrated that the decomposition of 5.0 V charged electrode can be promoted and Mn4+ can be deoxidized to Mn2+ by carbon and electrolyte through the simulation of blast process.

Spinel LiMn2O4 was synthesized by a solid-state method. A 204468-size battery was fabricated and stored at 55°C. The structure and morphology of the LiMn2O4 cathode were analyzed by X-ray diffraction (XRD) and scanning electron microscopy (SEM) technique. Energy dispersive spectroscopy (EDS) was used to analyze the surface component of the carbon anode. The discharge capacities of LiMn2O4 stored for 0, 24, 48, and 96 h are 106, 98, 96, and 92 mAh·g−1, respectively. The cyclic performance is improved after storage. The capacity retentions of LiMn2O4 stored for 0, 24, 48, and 96 h are 83.8%, 85.8%, 86.9%, and 88.6% after 180 cycles. The intensity of all the LiMn2O4 diffraction peaks is weakened. Mn is detected from the carbon electrode when the battery is stored for 96 h. Cyclic voltammograms and electrochemical impedance spectroscopy (EIS) were used to examine the surface state of the electrode after storage. The results show that the resistance and polarization of LiMn2O4/electrolyte is increased after storage, which is responsible for the fading of capacity.

The monoclinic-type Li 3 V 2 (PO 4 ) 3 cathode material was synthesized via calcining amorphous Li 3 V 2 (PO 4 ) 3 obtained by chemical reduction and lithiation of V 2 O 5 using oxalic acid as reducer and lithium carbonate as lithium source in alcohol solution. The amorphous Li 3 V 2 (PO 4 ) 3 precursor was characterized by using TG–DSC and XPS. The results showed that the V 5+ was reduced to V 3+ by oxalic acid at ambient temperature and pressure. The prepared Li 3 V 2 (PO 4 ) 3 was characterized by XRD and SEM. The results indicated the Li 3 V 2 (PO 4 ) 3 powder had good crystallinity and mesoporous morphology with an average diameter of about 30 nm. The pure Li 3 V 2 (PO 4 ) 3 exhibits a stable discharge capacity of 130.08 mAh g −1 at 0.1 C (14 mA g −1 ).

LiFePO4/C composite cathode material was prepared by carbothermal reduction method, which uses NH4H2PO4, Li2CO3 and cheap Fe2O3 as starting materials, acetylene black and glucose as carbon sources. The precursor of LiFePO4/C was characterized by differential thermal analysis and thermogravimetry. X-ray diffraction (XRD), scanning electron microscopy (SEM) micrographs showed that the LiFePO4/C is olivine-type phase, and the addition of the carbon reduced the LiFePO4 grain size. The carbon is dispersed between the grains, ensuring a good electronic contact. The products sintered at 700 °C for 8 h with glucose as carbon source possessed excellent electrochemical performance. The synthesized LiFePO4 composites showed a high electrochemical capacity of 159.3 mAh g−1 at 0.1 C rate, and the capacity fading is only 2.2% after 30 cycles.

Salt roasting with Na 2 SO 4 + CaCl 2 followed by water leaching was used to extract alkali metals from lepidolite. The experiments indicated that the best mass ratio of lepidolite / Na 2 SO 4 / CaCl 2 during roasting was 1:0.5:0.3. The extraction of Li , Rb and Cs were all > 90% after 0.5 h at 880 °C. The recovery of Li was essentially constant when roasted at 830–930 °C. The flexible roasting condition is easily controlled in industrial application. After leaching in water, the solution was cooled to − 5 °C for 2 h to crystallise 92.1% of the sulphate and 3.9% of the chloride as Na 2 SO 4 ·10H 2 O and NaCl , respectively. Evaporation and precipitation using Na 2 CO 3 produced Li 2 CO 3 crystals with > 99.5% purity and a solution from which Rb and Cs could be recovered.

Artificial intelligence (AI) has an astonishing potential in assisting clinical decision making and revolutionizing the field of health care. A major open challenge that AI will need to address before its integration in the clinical routine is that of algorithmic bias. Most AI algorithms need big datasets to learn from, but several groups of the human population have a long history of being absent or misrepresented in existing biomedical datasets. If the training data is misrepresentative of the population variability, AI is prone to reinforcing bias, which can lead to fatal outcomes, misdiagnoses, and lack of generalization. Here, we describe the challenges in rendering AI algorithms fairer, and we propose concrete steps for addressing bias using tools from the field of open science.