Anode-free lithium metal batteries (AFLMBs), which are manufactured without anode active material, offer great potential for high-energy-density, low-cost energy storage. However, AFLMBs face a long-standing challenge of short lifespan due to the harsh conditions of lacking excess Li-resource and an anode host1-8. This issue is associated with uneven Li deposition/dissolution, rooted in the micro-heterogeneity and mechanical fragility of solid electrolyte interphase (SEI)9. Here we report a practical 500 Wh kg-1-level AFLMB with enhanced lifespan, achieved using a crossover-coupled electrolyte. The electrolyte triggers crossover-coupled interfacial reactions that generate a B-F-based polymer-rich SEI at the anode while suppressing gas evolution at the cathode. The resulting SEI exhibits sub-nanometer homogeneity, high flexibility, and rapid Li-ion transport, and it spontaneously develops a self-adaptive mesh-film structure that ensures uniform ion flux and large-volume-change accommodation, thereby realizing reversible planar Li deposition/dissolution of 5.6 mAh cm-2. Consequently, a 2.7 Ah AFLMB (508 Wh kg-1, 1668 Wh L-1) without any host-material coating demonstrates stable cycling for 100 cycles at 100% depth of discharge (DoD) and 250 cycles at 80% DoD, with 80% capacity retention and a high-power output of 2650 W kg-1 at 96 Wh kg-1. These findings establish crossover-coupled interphase chemistry and address the inherent structural instability of host-free electrodes, advancing the practical implementation of AFLMBs.

Mammalian oocytes are filled by fibric structure called cytoplasmic lattice (CPL), essential for oocyte maturation and early embryonic development1-3. CPL comprises subcortical maternal complex (SCMC) and multiple components, including PADI62,4,5. Despite its discovery in the 1960s, the molecular architecture and assembly mechanisms of CPL have remained poorly understood. Here we present the cryo-electron microscopy (cryo-EM) structure of the CPL isolated from mouse oocytes. Our analysis identified 14 constitutive protein subunits and revealed that CPL is composed of repeating "U-shaped basket" (UB) and "adapter ring" (AR)- featured units, forming a filamentous architecture. AR adopts a two-fold symmetric conformation, containing two NLRP4f, four SCMC and two ZBED3 subunits circularized via two distinct interaction clusters. The UB is anchored by PADI6, a didecamer composed of ten homodimers assembled by two back-to-back pentamers, each forming the lateral side of UB. The underfoot base and up-down sides of the UB are formed by multiple central-symmetric assemblies (UBE2D3-UHRF1-NLRP14) and (TUBB2B-TUBB2A-FBXW24-SKP1) respectively, associating with the PADI6 pentamers to construct the intact UB structure. Two SCMC dimer within each AR connect the up and down sides of two adjacent UBs with an extensive protein-protein interaction network and thus maintain the repetitive connection between the neighboring CPL units. Our work unveils the architectural principles underlying the assembly of this large, periodic CPL filament, offering a molecular basis for understanding CPL's functions in early mammalian embryogenesis and female reproductive disorders.

Perovskite-silicon triple-junction photovoltaics offer efficiency gains beyond dual-junction devices, but at the expense of added complexity (1). Here, we address two key bottlenecks in perovskite-silicon-based triple-junction solar cells: reduced open-circuit voltage in the wide-bandgap top-cell and limited photocurrent generation in the middle-cell (1, 2). A non-volatile additive, 4-hydroxybenzylamine, regulates wide-bandgap perovskite crystallization and passivates defects, promoting oriented growth and suppressing non-radiative recombination. Together with improved energy-level alignment, this yields open-circuit voltages of up to 1.405 V and enhanced stability. To overcome the current limitations in the middle-cell, a three-step deposition strategy enables the formation of thick, low-bandgap perovskite absorbers while preserving microstructural integrity and enhancing electron extraction. In addition, low-refractive-index SiOx nanoparticles that accumulate in the front valleys of the textured silicon bottom-cell act as an optical middle-reflector, enhancing light absorption in the middle-cell. These advances are then combined in 1 cm² perovskite-perovskite-silicon devices, achieving a certified efficiency of 30.02%.

Alkynes are widely used as feedstock chemicals and functional groups in organic chemistry1,2. However, while the hydrogenation from an alkyne to an alkene is well established, typical methods for the reverse reaction - conversion of an alkene to an alkyne, are based on elimination chemistry reported in the 1860s3 and use forcing conditions (strong base or high temperatures)4-6. This precludes more general application on functional molecules. Here we report a recyclable selenanthrene reagent that mediates alkenes desaturation to alkynes under mild conditions. This method shows broad compatibility with both classical leaving groups and sensitive functional groups, enabling application late-stage in the efficient synthesis of complex alkynes. Moreover, this platform enables Z/E alkenes configuration inversion or sorting that are inaccessible with existing methods, highlighting its potential for diverse downstream derivatizations.

Insulin resistance (IR), a primary precursor to type 2 diabetes, is characterized by impaired insulin action in tissues1. However, diagnostic methods remain expensive and inaccessible, which hinders early intervention2,3. Here we present the WEAR-ME study, a large, remotely conducted study of IR (n = 1,165 participants; median body mass index (BMI) = 28 kg m-2, median age = 45 years, median haemoglobin A1c (HbA1c) = 5.4%) that uses time-series data from wearable devices and routine blood biomarkers to train deep neural networks against a ground-truth measure of IR (homeostatic model assessment of IR; HOMA-IR). Using a HOMA-IR cut-off of 2.9, our multimodal model achieved robust performance (area under the receiver operating characteristic curve (AUROC) = 0.80, sensitivity = 76%, specificity = 84%) with data from wearable devices, together with demographic and routine blood biomarker data. To enhance the use of time-series data from wearables, we fine-tuned a wearable foundation model (WFM) pretrained on 40 million hours of sensor data. In an independent validation cohort (n = 72), a model integrating WFM-derived representations with demographic data surpassed a demographics-only baseline (AUROC = 0.75 versus 0.66). Moreover, adding WFM-derived representations to a model with demographics, fasting glucose and a lipid panel substantially improved performance, compared with an identical model without data from wearables (AUROC = 0.88 versus 0.76). We integrate IR prediction into a large language model to contextualize the results and facilitate personalized recommendations. This work establishes a scalable, accessible framework for the early detection of metabolic risk, which could enable timely lifestyle interventions to prevent progression to type 2 diabetes.