Cu/CuxO/Graphdiyne Tandem Catalyst for Efficient Electrocatalytic Nitrate Reduction to Ammonia.

The electrocatalytic reduction reaction of nitrate (NO3 -) to ammonia (NH3) is a feasible way to achieve artificial nitrogen cycle. However, the low yield rate and poor selectivity towards NH3 product is a technical challenge. Here we present a graphdiyne (GDY)-based tandem catalyst featuring Cu/CuxO nanoparticles anchored to GDY support (termed Cu/CuxO/GDY) for efficient electrocatalytic NO3 - reduction. We achieve a high NH3 yield rate of 25.4 mg h-1 mgcat. -1 (25.4 mg h-1 cm-2) with a Faradaic efficiency of 99.8% at an applied potential of -0.8 V versus RHE using the designed catalyst. These performance metrics outperform most reported NO3 - to NH3 catalysts in the alkaline media. Electrochemical measurements and density functional theory reveal that the NO3 - preferentially attacks Cu/CuxO, and the GDY can effectively catalyze the reduction of NO2 - to NH3. This work highlights the efficacy of GDY as a new class of tandem catalysts for the artificial nitrogen cycle and provides powerful guidelines for the design of tandem electrocatalysts. This article is protected by copyright. All rights reserved.

General Synthesis of Metal Indium Sulfide Atomic Layers for Photocatalysis.

The metal indium sulfides have attracted extensive research interest in photocatalysis due to regulable atomic configuration and excellent optoelectronic properties. However, the synthesis of metal indium sulfide atomic layers is still challenging since intrinsic non-van-der-Waals layered structures of some components. Here, a surfactant self-assembly growth mechanism is proposed to controllably synthesize metal indium sulfide atomic layers. Eleven types of atomic layers with tunable compositions, thickness, and defect concentrations are successfully achieved namely In2S3, MgIn2S4, CaIn2S4, MnIn2S4, FeIn2S4, ZnIn2S4, Zn2In2S5, Zn4In16S33, CuInS2, CuIn5S8, and CdIn2S4. The typical CaIn2S4 shows a defect-dependence activity for CO2 photoreduction. The designed S vacancies in CaIn2S4 can serve as catalytic centers to activate CO2 molecules via localized electrons for π-back-donation. The engineered S vacancies tune the non-covalent interaction with CO2 and intermediates, manages to tune the free energy, and lower the reaction energy barrier. As a result, the defect-rich CaIn2S4 displays 2.82× improved reduction rate than defect-poor CaIn2S4. Meantime, other components also display promising photocatalytic performance, such as Zn2In2S5 with a H2O2 photosynthesis rate of 292 µmol g-1 h-1 and CuInS2 with N2-NH4 + conversion rate of 54 µmol g-1 h-1. This work paves the way for the multidisciplinary exploration of metal indium sulfide atomic layers with unique photocatalysis properties.

Asymmetric Electron Redistribution in Niobic-Oxygen Vacancy Associates to Tune Noncovalent Interaction in CO2 Photoreduction.

The role of vacancy associates in photocatalytic CO2 reduction is an open question. Herein, the Nb─O vacancy associates (VNb─O) are engineered into niobic acid (NA) atomic layers to tailor the CO2 photoreduction performance. The intrinsic charge compensation from O to Nb around Nb─O vacancy associates can manipulate the active electronic states, leading to the asymmetric electron redistribution. These local symmetry breaking sites show a charge density gradient, forming a localized polarization field to polarize nonpolar CO2 molecules and tune the noncovalent interaction of reaction intermediates. This unique configuration contributes to the 9.3 times increased activity for photocatalytic CO2 reduction. Meantime, this VNb─O NA also shows excellent photocatalytic activity for NO3 --NH4 + synthesis, with NH4 + formation rate up to 3442 µmol g-1 h-1. This work supplies fresh insights into the vacancy associate design for electron redistribution and noncovalent interaction tuning in photocatalysis.

Deshielding Anions Enable Solvation Chemistry Control of LiPF6-Based Electrolyte toward Low-Temperature Lithium-Ion Batteries.

Severe capacity decay under subzero temperatures remains a significant challenge for lithium-ion batteries (LIBs) due to the sluggish interfacial kinetics. Current efforts to mitigate this deteriorating interfacial behavior rely on high-solubility lithium salts (e.g., Lithium bis(trifluoromethanesulfonyl)imide (LiTFSI), Lithium bis(fluorosulfonyl)imide (LiFSI))-based electrolytes to construct anion participated solvation structures. However, such electrolytes bring issues of corrosion on the current collector and increased costs. Herein, the most commonly used Lithium hexafluorophosphate (LiPF6) instead, to establish a peculiar solvation structure with a high ratio of ion pairs and aggregates by introducing a deshielding NO3 - additive for low-temperature LIBs is utilized. The deshielding anion significantly reduces the energy barrier for interfacial behavior at low temperatures. Benefiting from this, the graphite (Gr) anode retains a high capacity of ≈72.3% at -20 °C, which is far superior to the 32.3% and 19.4% capacity retention of counterpart electrolytes. Moreover, the LiCoO2/Gr full cell exhibits a stable cycling performance of 100 cycles at -20 °C due to the inhibited lithium plating. This work heralds a new paradigm in LiPF6-based electrolyte design for LIBs operating at subzero temperatures.

Weakly supervised learning-based 3D bladder reconstruction from 2D ultrasound images for bladder volume measurement.

BACKGROUND: Accurate measurement of bladder volume is necessary to maintain the consistency of the patient's anatomy in radiation therapy for pelvic tumors. As the diversity of the bladder shape, traditional methods for bladder volume measurement from 2D ultrasound have been found to produce inaccurate results. PURPOSE: To improve the accuracy of bladder volume measurement from 2D ultrasound images for patients with pelvic tumors. METHODS: The bladder ultrasound images from 130 patients with pelvic cancer were collected retrospectively. All data were split into a training set (80 patients), a validation set (20 patients), and a test set (30 patients). A total of 12 transabdominal ultrasound images for one patient were captured by automatically rotating the ultrasonic probe with an angle step of 15°. An incomplete 3D ultrasound volume was synthesized by arranging these 2D ultrasound images in 3D space according to the acquisition angles. With this as input, a weakly supervised learning-based 3D bladder reconstruction neural network model was built to predict the complete 3D bladder. The key point is that we designed a novel loss function, including the supervised loss of bladder segmentation in the ultrasound images at known angles and the compactness loss of the 3D bladder. Bladder volume was calculated by counting the number of voxels belonging to the 3D bladder. The dice similarity coefficient (DSC) was used to evaluate the accuracy of bladder segmentation, and the relative standard deviation (RSD) was used to evaluate the calculation accuracy of bladder volume with that of computed tomography (CT) images as the gold standard. RESULTS: The results showed that the mean DSC was up to 0.94 and the mean absolute RSD can be reduced to 6.3% when using 12 ultrasound images of one patient. Further, the mean DSC also was up to 0.90 and the mean absolute RSD can be reduced to 9.0% even if only two ultrasound images were used (i.e., the angle step is 90°). Compared with the commercial algorithm in bladder scanners, which has a mean absolute RSD of 13.6%, our proposed method showed a considerably huge improvement. CONCLUSIONS: The proposed weakly supervised learning-based 3D bladder reconstruction method can greatly improve the accuracy of bladder volume measurement. It has great potential to be used in bladder volume measurement devices in the future.

Ultrahigh Passive Cooling Power in Hydrogel with Rationally Designed Optofluidic Properties.

The cooling power of a radiative cooler is more than halved in the tropics, e.g., Singapore, because of its harsh weather conditions including high humidity (84% on average), strong downward atmospheric radiation (∼40% higher than elsewhere), abundant rainfall, and intense solar radiation (up to 1200 W/m2 with ∼58% higher UV irradiation). So far, there has been no report of daytime radiative cooling that well achieves effective subambient cooling. Herein, through integrated passive cooling strategies in a hydrogel with desirable optofluidic properties, we demonstrate stable subambient (4-8 °C) cooling even under the strongest solar radiation in Singapore. The integrated passive cooler achieves an ultrahigh cooling power of ∼350 W/m2, 6-10 times higher than a radiative cooler in a tropical climate. An in situ study of radiative cooling with various hydration levels and ambient humidity is conducted to understand the interaction between radiation and evaporative cooling. This work provides insights for the design of an integrated cooler for various climates.

Next-Generation Vitrimers Design through Theoretical Understanding and Computational Simulations.

Vitrimers are an innovative class of polymers that boast a remarkable fusion of mechanical and dynamic features, complemented by the added benefit of end-of-life recyclability. This extraordinary blend of properties makes them highly attractive for a variety of applications, such as the automotive sector, soft robotics, and the aerospace industry. At their core, vitrimer materials consist of crosslinked covalent networks that have the ability to dynamically reorganize in response to external factors, including temperature changes, pressure variations, or shifts in pH levels. In this review, the aim is to delve into the latest advancements in the theoretical understanding and computational design of vitrimers. The review begins by offering an overview of the fundamental principles that underlie the behavior of these materials, encompassing their structures, dynamic behavior, and reaction mechanisms. Subsequently, recent progress in the computational design of vitrimers is explored, with a focus on the employment of molecular dynamics (MD)/Monte Carlo (MC) simulations and density functional theory (DFT) calculations. Last, the existing challenges and prospective directions for this field are critically analyzed, emphasizing the necessity for additional theoretical and computational advancements, coupled with experimental validation.

Deep learning based automatic internal gross target volume delineation from 4D-CT of hepatocellular carcinoma patients.

BACKGROUND: The location and morphology of the liver are significantly affected by respiratory motion. Therefore, delineating the gross target volume (GTV) based on 4D medical images is more accurate than regular 3D-CT with contrast. However, the 4D method is also more time-consuming and laborious. This study proposes a deep learning (DL) framework based on 4D-CT that can achieve automatic delineation of internal GTV. METHODS: The proposed network consists of two encoding paths, one for feature extraction of adjacent slices (spatial slices) in a specific 3D-CT sequence, and one for feature extraction of slices at the same location in three adjacent phase 3D-CT sequences (temporal slices), a feature fusion module based on an attention mechanism was proposed for fusing the temporal and spatial features. Twenty-six patients' 4D-CT, each consisting of 10 respiratory phases, were used as the dataset. The Hausdorff distance (HD95), Dice similarity coefficient (DSC), and volume difference (VD) between the manual and predicted tumor contour were computed to evaluate the model's segmentation accuracy. RESULTS: The predicted GTVs and IGTVs were compared quantitatively and visually with the ground truth. For the test dataset, the proposed method achieved a mean DSC of 0.869 ± 0.089 and an HD95 of 5.14 ± 3.34 mm for all GTVs, with under-segmented GTVs on some CT slices being compensated by GTVs on other slices, resulting in better agreement between the predicted IGTVs and the ground truth, with a mean DSC of 0.882 ± 0.085 and an HD95 of 4.88 ± 2.84 mm. The best GTV results were generally observed at the end-inspiration stage. CONCLUSIONS: Our proposed DL framework for tumor segmentation on 4D-CT datasets shows promise for fully automated delineation in the future. The promising results of this work provide impetus for its integration into the 4DCT treatment planning workflow to improve hepatocellular carcinoma radiotherapy.

Structure-Based Multilevel Descriptors for High-throughput Screening of Elastomers.

To discover new materials, high-throughput screening (HTS) with machine learning (ML) requires universally available descriptors that can accurately predict the desired properties. For elastomers, experimental and simulation data in current descriptors may not be available for all candidates of interest, hindering elastomer discovery through HTS. To address this challenge, we introduce structure-based multilevel (SM) descriptors of elastomers derived solely from molecular structure that is universally available. Our SM descriptors are hierarchically organized to capture both local soft and hard segment structures as well as the global structures of elastomers. With the SM-Morgan Fingerprint (SM-MF) descriptor, one of our SM descriptors, a machine learning model accurately predicts elastomer toughness with a remarkable accuracy of 0.91. Furthermore, an HTS pipeline is established to swiftly screen elastomers with targeted toughness. We also demonstrate the generality and applicability of SM descriptors by using them to construct HTS pipelines for screening elastomers with a targeted critical strain or Young's modulus. The user-friendliness and low computational cost of SM descriptors make them a promising tool to significantly enhance HTS in the search for novel materials.

Machine-Learning-Assisted Optimization of a Single-Atom Coordination Environment for Accelerated Fenton Catalysis.

Machine learning (ML) algorithms will be enablers in revolutionizing traditional methods of materials optimization. Here, we broaden the use of ML to assist the construction of Fenton-like single-atom catalysts (SACs) by developing a methodology including model building, training, and prediction. Our approach can efficiently extract synthesis parameters that exert a substantial influence on Fenton activity and accurately predict the phenol degradation rate k of SACs with a mean error of ±0.018 min-1. The extended synthesis window with accelerated learning enables the realization that the heating temperatures during SAC synthesis significantly influence the Fe-N coordination number, which ultimately dictates their performance. Through ML-guided optimization, a highly efficient SAC dominated by Fe-N5 sites with exceptional Fenton activity (k = 0.158 min-1) is identified. Our work provides an example for ML-assisted optimization of single-atom coordination environments and illuminates the feasibility of ML in accelerating the development of high-performance catalysts.

Cellulose Nanofiber Platform for Pesticide Sequestration in the Gastrointestinal Tract.

Exploitation of nature-derived materials is an important approach to promote environmental sustainability. Among these materials, cellulose is of particular interest due to its abundance and relative ease of access. As a food ingredient, cellulose nanofibers (CNFs) have found interesting applications as emulsifiers and modulators of lipid digestion and absorption. In this report, we show that CNFs can also be modified to modulate the bioavailability of toxins, such as pesticides, in the gastrointestinal tract (GIT) by forming inclusion complexes and promoting interaction with surface hydroxyl groups. CNFs were successfully functionalized with (2-hydroxypropyl)-ß-cyclodextrin (HPBCD) using citric acid as a crosslinker via esterification. Functionally, the potential for pristine and functionalized CNFs (FCNFs) to interact with a model pesticide, boscalid, was tested. Based on direct interaction studies, adsorption of boscalid saturated at around 3.09% on CNFs and at 12.62% on FCNFs. Using an in vitro GIT simulation platform, the adsorption of boscalid on CNFs/FCNFs was also studied. The presence of a high-fat food model was found to have a positive effect in binding boscalid in a simulated intestinal fluid environment. In addition, FCNFs were found to have a greater effect in retarding triglyceride digestion than CNFs (61% vs 30.6%). Overall, FCNFs were demonstrated to evoke synergistic effects of reducing fat absorption and pesticide bioavailability through inclusion complex formation and the additional binding of the pesticide onto surface hydroxyl groups on HPBCD. By adopting food-compatible materials and processes for production, FCNFs have the potential to be developed into a functional food ingredient for modulating food digestion and the uptake of toxins.

Navigating surface reconstruction of spinel oxides for electrochemical water oxidation.

Understanding and mastering the structural evolution of water oxidation electrocatalysts lays the foundation to finetune their catalytic activity. Herein, we demonstrate that surface reconstruction of spinel oxides originates from the metal-oxygen covalency polarity in the MT-O-MO backbone. A stronger MO-O covalency relative to MT-O covalency is found beneficial for a more thorough reconstruction towards oxyhydroxides. The structure-reconstruction relationship allows precise prediction of the reconstruction ability of spinel pre-catalysts, based on which the reconstruction degree towards the in situ generated oxyhydroxides can be controlled. The investigations of oxyhydroxides generated from spinel pre-catalysts with the same reconstruction ability provide guidelines to navigate the cation selection in spinel pre-catalysts design. This work reveals the fundamentals for manipulating the surface reconstruction of spinel pre-catalysts for water oxidation.

Hydrogen Radical-Induced Electrocatalytic N2 Reduction at a Low Potential.

Realizing efficient hydrogenation of N2 molecules in the electrocatalytic nitrogen reduction reaction (NRR) is crucial in achieving high activity at a low potential because it theoretically requires a higher equilibrium potential than other steps. Analogous to metal hydride complexes for N2 reduction, achieving this step by chemical hydrogenation can weaken the potential dependence of the initial hydrogenation process. However, this strategy is rarely reported in the electrocatalytic NRR, and the catalytic mechanism remains ambiguous and lacks experimental evidence. Here, we show a highly efficient electrocatalyst (ruthenium single atoms anchored on graphdiyne/graphene sandwich structures) with a hydrogen radical-transferring mechanism, in which graphdiyne (GDY) generates hydrogen radicals (H•), which can effectively activate N2 to generate NNH radicals (•NNH). A dual-active site is constructed to suppress competing hydrogen evolution, where hydrogen preferentially adsorbs on GDY and Ru single atoms serve as the adsorption site of •NNH to promote further hydrogenation of NH3 synthesis. As a result, high activity and selectivity are obtained simultaneously at -0.1 V versus a reversible hydrogen electrode. Our findings illustrate a novel hydrogen transfer mechanism that can greatly reduce the potential and maintain the high activity and selectivity in NRR and provide powerful guidelines for the design concept of electrocatalysts.

Polyfluorinated crosslinker-based solid polymer electrolytes for long-cycling 4.5 V lithium metal batteries.

Solid polymer electrolytes (SPEs), which are favorable to form intimate interfacial contacts with electrodes, are promising electrolyte of choice for long-cycling lithium metal batteries (LMBs). However, typical SPEs with easily oxidized oxygen-bearing polar groups exhibit narrow electrochemical stability window (ESW), making it impractical to increase specific capacity and energy density of SPE based LMBs with charging cut-off voltage of 4.5 V or higher. Here, we apply a polyfluorinated crosslinker to enhance oxidation resistance of SPEs. The crosslinked network facilitates transmission of the inductive electron-withdrawing effect of polyfluorinated segments. As a result, polyfluorinated crosslinked SPE exhibits a wide ESW, and the Li|SPE|LiNi0.5Co0.2Mn0.3O2 cell with a cutoff voltage of 4.5 V delivers a high discharge specific capacity of ~164.19 mAh g-1 at 0.5 C and capacity retention of ~90% after 200 cycles. This work opens a direction in developing SPEs for long-cycling high-voltage LMBs by using polyfluorinated crosslinking strategy.

Peptide Amphiphile Mediated Co-assembly for Nanoplasmonic Sensing.

Aromatic interactions are commonly involved in the assembly of naturally occurring building blocks, and these interactions can be replicated in an artificial setting to produce functional materials. Here we describe a colorimetric biosensor using co-assembly experiments with plasmonic gold and surfactant-like peptides (SLPs) spanning a wide range of aromatic residues, polar stretches, and interfacial affinities. The SLPs programmed in DDD-(ZZ)x -FFPC self-assemble into higher-order structures in response to a protease and subsequently modulate the colloidal dispersity of gold leading to a colorimetric readout. Results show the strong aggregation propensity of the FFPC tail without polar DDD head. The SLPs were specific to the target protease, i.e., Mpro , a biomarker for SARS-CoV-2. This system is a simple and visual tool that senses Mpro in phosphate buffer, exhaled breath condensate, and saliva with detection limits of 15.7, 20.8, and 26.1 nM, respectively. These results may have value in designing other protease testing methods.

Dual-path Network for Liver and Tumor Segmentation in CT Images Using Swin Transformer Encoding Approach.

BACKGROUND: Liver and tumor segmentation from CT images is a complex and crucial step in achieving full-course adaptive radiotherapy and also plays an essential role in computer-aided clinical diagnosis systems. Deep learning-based methods play an important role in achieving automatic segmentation. OBJECTIVE: This research aims to improve liver tumor detection performance by proposing a dual path feature extracting strategy and employing Swin-Transformer. METHODS: The hierarchical Swin-Transformer is embedded into the encoder and decoder and combined with CNN to form a dual coding path structure incorporating long-range dependencies and multi-scale contextual connections to capture coarse-tuned features at different semantic scales. The features of the two encoding paths and the upsampling path are fused, tested and validated with LITS and in-house datasets. RESULTS: The proposed method has a DG of 97.95% and a DC of 96.2% for liver segmentation; a DG of 80.6% and a DC of 68.1% for tumor segmentation; and a classification study of the tumor dataset shows a DG of 91.1% and a DC of 87.2% for large and continuous tumors and a DG of 71.7% and a DC of 66.4% for small and scattered tumors. CONCLUSION: Swin-Transformer can be used as a robust encoder for medical image segmentation networks and, combined with CNN networks, can better recover local spatial information and enhance feature representation. Accurate localization before segmentation can achieve better results for small and scattered tumors.

A Two-Dimensional van der Waals Heterostructure with Isolated Electron-Deficient Cobalt Sites toward High-Efficiency CO2 Electroreduction.

Electrochemical CO2 conversion is a promising way for sustainable chemical fuel production, yet the conversion efficiency is strongly limited by the sluggish kinetics and complex reaction pathways. Here we report the ultrathin conjugated metalloporphyrin covalent organic framework epitaxially grown on graphene as a two-dimensional van der Waals heterostructure to catalyze CO2 reduction. Operando X-ray absorption and density functional theory calculations reveal the strong interlayer coupling leads to electron-deficient metal centers and speeds up electrocatalysis. The Co(III)-N4 centers exhibit a CO Faradaic efficiency of 97% at a partial current density of 8.2 mA cm-2 in an H-cell, along with a stable running over 30 h. The selectivity of CO approached 99% with a partial current density of 191 mA cm-2 in a liquid flow cell, and the turnover frequency achieved 50 400 h-1 at -1.15 V vs RHE, outperforming most reported organometallic frameworks. This work highlights the key role of strong electronic coupling between van der Waals layers for accelerating the dynamics of CO2 conversion.

Polarized Cu-Bi Site Pairs for Non-Covalent to Covalent Interaction Tuning toward N2 Photoreduction.

A universal atomic layer confined doping strategy is developed to prepare Bi24 O31 Br10 materials incorporating isolated Cu atoms. The local polarization can be created along the CuOBi atomic interface, which enables better electron delocalization for effective N2 activation. The optimized Cu-Bi24 O31 Br10 atomic layers show 5.3× and 88.2× improved photocatalytic nitrogen fixation activity than Bi24 O31 Br10 atomic layer and bulk Bi24 O31 Br10 , respectively, with the NH3 generation rate reaching 291.1 µmol g-1 h-1 in pure water. The polarized Cu-Bi site pairs can increase the non-covalent interaction between the catalyst's surface and N2 molecules, then further weaken the covalent bond order in NN. As a result, the hydrogenation pathways can be altered from the associative distal pathway for Bi24 O31 Br10 to the alternating pathway for Cu-Bi24 O31 Br10 . This strategy provides an accessible pathway for designing polarized metal site pairs or tuning the non-covalent interaction and covalent bond order.

Directing the Architecture of Surface-Clean Cu2O for CO Electroreduction.

Tailoring the morphology of nanocrystals is a promising way to enhance their catalytic performance. In most previous shape-controlled synthesis strategies, surfactants are inevitable due to their capability to stabilize different facets. However, the adsorbed surfactants block the intrinsic active sites of the nanocrystals, reducing their catalytic performance. For now, strategies to control the morphology without surfactants are still limited but necessary. Herein, a facile surfactant-free synthesis method is developed to regulate the morphology of Cu2O nanocrystals (e.g., solid nanocube, concave nanocube, cubic framework, branching nanocube, branching concave nanocube, and branching cubic framework) to enhance the electrocatalytic performance for the conversion of CO to n-propanol. Specifically, the Cu2O branching cubic framework (BCF-Cu2O), which is difficult to fabricate using previous surfactant-free methods, is fabricated by combining the concentration depletion effect and the oxidation etching process. More significantly, the BCF-Cu2O-derived catalyst (BCF) presents the highest n-propanol current density (-0.85 mA cm-2) at -0.45 V versus the reversible hydrogen electrode (VRHE), which is fivefold higher than that of the surfactant-coated Cu2O nanocube-derived catalyst (SFC, -0.17 mA cm-2). In terms of the n-propanol Faradaic efficiency in CO electroreduction, that of the BCF exhibits a 41% increase at -0.45 VRHE as compared with SFC. The high catalytic activity of the BCF that results from the clean surface and the coexistence of Cu(100) and Cu(110) in the lattice is well-supported by density functional theory calculations. Thus, this work presents an important paradigm for the facile fabrication of surface-clean nanocrystals with an enhanced application performance.

Reversible Al Metal Anodes Enabled by Amorphization for Aqueous Aluminum Batteries.

Aqueous aluminum metal batteries (AMBs) are regarded as one of the most sustainable energy storage systems among post-lithium-ion candidates, which is attributable to their highest theoretical volumetric capacity, inherent safe operation, and low cost. Yet, the development of aqueous AMBs is plagued by the incapable aluminum plating in an aqueous solution and severe parasitic reactions, which results in the limited discharge voltage, thus making the development of aqueous AMBs unsuccessful so far. Here, we demonstrate that amorphization is an effective strategy to tackle these critical issues of a metallic Al anode by shifting the reduction potential for Al deposition. The amorphous aluminum (a-Al) interfacial layer is triggered by an in situ lithium-ion alloying/dealloying process on a metallic Al substrate with low strength. Unveiled by experimental and theoretical investigations, the amorphous structure greatly lowers the Al nucleation energy barrier, which forces the Al deposition competitive to the electron-stealing hydrogen evolution reaction (HER). Simultaneously, the inhibited HER mitigates the passivation, promoting interfacial ion transfer kinetics and enabling steady aluminum plating/stripping for 800 h in the symmetric cell. The resultant multiple full cells using Al@a-Al anodes deliver approximately a 0.6 V increase in the discharge voltage plateau compared to that of bare Al-based cells, which far outperform all reported aqueous AMBs. In both symmetric cells and full cells, the excellent electrochemical performances are achieved in a noncorrosive, low-cost, and fluorine-free Al2(SO4)3 electrolyte, which is ecofriendly and can be easily adapted for sustainable large-scale applications. This work brings an intriguing picture of the design of metallic anodes for reversible and high-voltage AMBs.