Lava tubes play an important role in base construction and understanding planetary geological development on extraterrestrial bodies. However, our understanding of lava tube morphology on Earth as a reference for extraterrestrial bodies remains limited. Existing research mainly focuses on their formation mechanisms and uses a single approach to study their morphology. To fill this gap, we conducted a multi-platform investigation of a lava tube near Jingpo Lake, Heilongjiang Province, China, employing space-, air-, and ground-based techniques. The method presented in this paper can assist in the comprehensive exploration of lunar lava tubes in the future. First, data covering this large lava tube area were collected using synthetic aperture radar (SAR), due to its all-weather capability, from which distribution information on lava tube skylights was retrieved. Furthermore, close-range point cloud data of the lava tube skylight areas were collected from drones equipped with the DJI Zenmuse L1 LiDAR sensor, which allows for high-resolution data acquisition and the creation of detailed models, providing local high-precision point clouds. Additionally, Light Detection and Ranging (LiDAR) was used for detailed point cloud collection inside the lava tubes, and through the registration of high-precision point clouds of the skylights both inside and outside the tube, comprehensive point cloud data of the entire lava tube were obtained. Hyperspectral LiDAR was also used to collect data of the tube walls under dark conditions, addressing the issue of consistency matching between hyperspectral imaging and laser far-field spots, enabling integrated spatial-spectral data acquisition. Finally, ground-penetrating radar was also employed for subsurface structural analysis, benefiting from its high penetration capabilities to acquire more detailed information on the underground morphology of the lava tubes. Our experiments demonstrate that this multi-sensor approach provides comprehensive information on terrain, lava tube orientation, cave distribution, and the internal environment of the lava tube system.

The lack of stable orange-red converters hinders the development of blue laser-driven full-color lighting sources. Here, an orange-emitting Lu2CaMg2Si3O12: Ce3+ (LCMSO: Ce) garnet phosphor was prepared and fabricated into a phosphor-in-glass film (PiG film) via high-temperature co-sintering (640 degrees C optimized) on a sapphire substrate. Benefiting from exceptional chemical stability against glass erosion and oxidative decomposition during cosintering (600-680 degrees C), the LCMSO: Ce PiG film achieves enhanced optical conversion efficiency (from 33% to 40%) via thermal defect annihilation. This stands in sharp contrast to nitride-based (CASN: Eu) PiG films, which suffer a drastic efficiency drop (from 35% to 14%) due to thermal degradation. Furthermore, a double-layer composite PiG film was constructed by combining the orange LCMSO: Ce PiG film with an in-house synthesized cyan Ca3Sc2Si3O12: Ce3+ PiG film. By modulating the thickness ratio of the two layers, continuous emission tuning from cyan to orange-red was realized, yielding high-quality white light (Ra = 80) at an optimized 1:1 ratio. Under 8.33 W laser irradiation (16.66 W/mm2), the composite PiG film delivers a luminous flux of 610.1 lm and a luminous efficacy of 73.2 lm/W without emission saturation. In addition, the composite PiG film exhibits robust long-term stability (under continuous excitation at 10 W/mm2 for 1800 s) and excellent thermal stability (85% of emission intensity retained at 420 K). This study demonstrates the great potential of the LCMSO: Ce PiG film as a high-performance color converter for advanced high-power laser-driven lighting applications.

Ultrasmall‐sized cesium lead iodide (CsPbI 3 ) quantum dots (QDs) are promising candidates for achieving spectrally stable pure‐red perovskite light‐emitting diodes (PeLEDs) meeting Rec. 2020 standards. However, the corresponding devices hardly achieve satisfactory external quantum efficiency (EQE), current efficiency (CE), and luminance simultaneously because of the use of largely excessive insulating long‐chain ligands and additional difficulties in the defect control of ultrasmall CsPbI 3 QDs. Herein, we develop an alkyl iodide‐assisted ligand modulation strategy for CsPbI 3 QDs toward high‐efficiency and bright pure‐red PeLEDs. We elucidate an in‐situ nucleophilic bimolecular (S N 2) substitution reaction between the oleylamine and additionally incorporated short‐chain 1‐iodooctane (IO) molecules during the materials synthesis. The reaction‐generated hydriodic acid (HI) induces non‐destructive surface etching of QDs, enabling exceptional luminescent properties of the strongly confined products. In addition, the S N 2 reaction‐derived secondary amine strongly adsorbs at the surface of QDs, which stabilizes the products with a reduced ligand density, simultaneously enhancing photoluminescence stability and electrical properties of the assembled emissive layers. The resultant devices emitting at 632 nm demonstrate a peak EQE of 21.56%, an impressive luminance of 13,132 cd m −2 , and an exceptional CE of 20.73 cd A −1 , which outperforms state‐of‐the‐art Rec. 2020 pure‐red PeLEDs utilizing ultrasmall‐sized colloidal CsPbI 3 QDs.

Bistable structures have broad applications in intelligent mechanical design due to their simplicity, low energy consumption, and high reliability. Self-propagating mechanisms, composed of bistable units connected by elastic components, enable tunable energy transfer and state transitions. When triggered, the initial unit overcomes its energy barrier, releasing energy to sequentially activate adjacent units in a domino-like chain reaction. This study introduces a fan-shaped bistable unit arranged in a one-dimensional chain to create a self-propagating mechanism. A discrete numerical model based on the Lagrangian equation is employed to investigate the influence of structural parameters, such as spring stiffness, friction, and damping on deployment dynamics. Results show that these parameters can precisely control deployment speed and energy transmission modes. The mechanism demonstrates robustness, unidirectionality, uniformity, tunability, and smart structure capabilities, enabling continuous energy transfer and automatic deployment based on single-point triggering. Additionally, a continuous model is developed to predict global dynamic behavior, validated through comparison with the discrete model. These findings highlight the potential of bistable self-propagating structures for intelligent design applications in aerospace, civil engineering, and mechanical engineering, offering novel solutions for deployable systems.

ABSTRACT Perovskite‐silicon tandem solar cells hold great promise for the development of next‐generation cost‐effective photovoltaic technologies, and their performance gains largely rely on advanced materials and interface engineering of wide‐bandgap perovskite sub‐cells. Herein, we develop a cooperative self‐assembled monolayer (co‐SAM) strategy by incorporating a rationally designed alkyl‐phosphonic acid, 12‐methoxydodecylphosphonic acid (MeODPA), into commonly used carbazole‐based SAMs, leading to an obvious enhancement in the performance of single‐junction wide‐bandgap (WBG) (1.68 eV) and tandem PSCs. We demonstrate that the methoxy terminal group and flexible alkyl chain of MeODPA synergistically improve substrate wettability and molecular coverage of SAMs on the buried substrate, promoting perovskite crystallization and stabilizing the buried interfacial contacts. Moreover, we uncover a previously overlooked function of the co‐SAMs in regulating the PbI 2 distribution at the top surface, which facilitates more effective surface passivation, leading to improved light stability of resultant perovskite films. Leveraging these advancements, we fabricate single‐junction WBG and perovskite‐silicon tandem devices with exceptional power conversion efficiencies (PCEs) of 24.11% and 32.16%, respectively. This work provides a comprehensive understanding of the coordinated effects of co‐SAM on bulk and dual‐interface properties of perovskite films, offering valuable insights for further performance improvements of both single‐junction and tandem perovskite photovoltaics.

Abstract Tandem light-emitting diodes (LEDs) offer advantages in efficiency, brightness and lower the current density for a target brightness, but their implementation in solution-processed perovskite LEDs is limited by interconnection strategies that introduce high voltage penalties and poor robustness. Here we demonstrate a hierarchically structured interconnection layer (ICL) integrating a near-degenerate charge-generation junction with a crystalline-amorphous oxide composite, enabling low-barrier transport and solvent-tolerant stacking with reduced loss. Using this ICL, we realize double-junction perovskite LEDs in which two subunits operate efficiently in series, achieving a peak external quantum efficiency (EQE) of 42.6%, a maximum radiance of 690 W sr −1 m −2 , and low driving voltage. The ICL is compatible with multiple perovskite emitters and supports scalable multijunction integration, yielding multijunction devices with sub-bandgap turn-on voltages and high EQEs exceeding 60%. Our results establish a low-loss interconnection strategy for series-stacked perovskite emitters, providing a practical route toward high-brightness and multi-wavelength perovskite light sources.

All-inorganic CsPbI3 perovskite holds great promise for high-performance deep-red light-emitting diodes (LEDs), yet its hardly controllable phase transition and crystallization readily induce abundant defects in the resultant thin-film emitters. The development of alkaline substrate-assisted CsPbI3 modulation has enabled significant performance improvements in associated perovskite LEDs (PeLEDs). However, a thorough understanding and effective management of the buried interfacial reactions remains elusive. Here, we develop a magnesium (Mg)-doped alkaline zinc hydroxide (Zn(OH)2) substrate with delicately regulated surface properties and systematically investigate the crystallization and degradation of CsPbI3 emissive layers deposited on top. We reveal that the Mg doping is effective in reducing oxygen vacancies and surface hydroxyls, which directly attenuates the substrate basicity. The Mg-doped Zn(OH)2 substrate suppresses the alkaline-interface-induced organic deprotonation and alleviates the rapid transition process from intermediate phases to CsPbI3 perovskite, enabling high-quality emitters with reduced defects. More importantly, this facile interface engineering substantially mitigates the detrimental interfacial degradation and structural collapse of perovskite emitters under continuous electrical and thermal stresses. Consequently, we obtain deep-red PeLEDs with a peak external quantum efficiency of 23.62% and an exceptional operational half-lifetime of 376 h at 20 mA cm-2, representing one of the best-performing devices utilizing bulk CsPbI3 emitters.

Ultrasmall-sized cesium lead iodide (CsPbI3) quantum dots (QDs) are promising candidates for achieving spectrally stable pure-red perovskite light-emitting diodes (PeLEDs) meeting Rec. 2020 standards. However, the corresponding devices hardly achieve satisfactory external quantum efficiency (EQE), current efficiency (CE), and luminance simultaneously because of the use of largely excessive insulating long-chain ligands and additional difficulties in the defect control of ultrasmall CsPbI3 QDs. Herein, we develop an alkyl iodide-assisted ligand modulation strategy for CsPbI3 QDs toward high-efficiency and bright pure-red PeLEDs. We elucidate an in-situ nucleophilic bimolecular (SN2) substitution reaction between the oleylamine and additionally incorporated short-chain 1-iodooctane (IO) molecules during the materials synthesis. The reaction-generated hydriodic acid (HI) induces non-destructive surface etching of QDs, enabling exceptional luminescent properties of the strongly confined products. In addition, the SN2 reaction-derived secondary amine strongly adsorbs at the surface of QDs, which stabilizes the products with a reduced ligand density, simultaneously enhancing photoluminescence stability and electrical properties of the assembled emissive layers. The resultant devices emitting at 632 nm demonstrate a peak EQE of 21.56%, an impressive luminance of 13,132 cd m-2, and an exceptional CE of 20.73 cd A-1, which outperforms state-of-the-art Rec. 2020 pure-red PeLEDs utilizing ultrasmall-sized colloidal CsPbI3 QDs.

All-inorganic CsPbI3 perovskite holds great promise for deep-red perovskite light-emitting diodes (PeLEDs) owing to its superior color purity and thermal stability. However, its performance is severely limited by deep-level defects such as uncoordinated Pb2+ sites and A-site vacancies, which originate from the uncontrollable crystallization process and are inadequately addressed by conventional passivation strategies that difficult to reach defects buried inside grains. These defects act as non-radiative recombination centers, reducing photoluminescence quantum yield, accelerating efficiency roll-off, and compromising device operational stability. To address this issue, we introduce free Guanidine (Gua) molecules into the perovskite precursor to achieve bulk incorporation of Gua throughout the perovskite lattice and grain boundaries. The electron-rich nitrogen atoms of Gua strongly coordinate with undercoordinated Pb2+ sites via Lewis acid–base coordination and the excess Gua inserts in the perovskite crystal unit, enabling passivation of bulk defects. Meanwhile, the strong Gua-Pb coordination introduces an additional dissociation step that retards perovskite crystallization. Eventually, we obtain deep-red PeLEDs with a peak external quantum efficiency of 22.63% and an extended operational half-lifetime of 247 h at 20 mA cm−2, representing a viable pathway toward bulk passivation for high-performance perovskite optoelectronics.

Near-infrared (NIR) light-emitting diodes (LEDs) based on perovskite quantum dots (QDs) hold broad application prospects in night vision, biomedical treatment, and optical communications. NIR perovskite QD-based LEDs (QLEDs) have made significant progress, with external quantum efficiencies (EQEs) exceeding 20% over the past two years, but they still suffer from low radiance and poor operational stability. Here, we used ammonium sulfamate to construct the surface hydrogen-bond network, thereby enabling QLEDs with high radiance and long lifetime. The ammonium sulfamate forms N & horbar;H & centerdot;& centerdot;& centerdot;I and N & horbar;H & centerdot;& centerdot;& centerdot;O bonds with FAPbI(3) QDs to inhibit the desorption of FA(+) and I-, thus stabilizing the QD surface structure. The as-fabricated QLED showed a high radiance of 109 167 mW, a maximum EQE of 20.4% at 785 nm sr(-1) m(-2), and an excellent operational stability of 1286 min at an initial radiance of 1000 mW sr(-1) m(-2), representing a significant improvement over the previous reported record radiance (<40 000 mW sr(-1) m(-2)) and record lifetime (160 min). This work offers a practical solution to enhance the radiance and operational stability of perovskite NIR QLEDs and also promotes their industrialization.

Understanding and accurately predicting heat conduction in composite materials is crucial for various engineering applications. In this paper, a physics-informed finite difference U-Net (FD-U-Net) is proposed to solve two dimensional steady-state heat conduction problem in composites by directly computing from the images of heterogeneous microstructures without labeled data. Using the encoder-decoder structure, FD-U-Net can establish the mapping from microstructure image to steady-state temperature field. The loss function of FD-U-Net integrates the principles of finite difference, ensuring physical consistency of the prediction. It also enables model training without labeled data. To optimize model performance, the critical training hyperparameters are investigated, including the number of finite difference kernel iterations, batch size, and choice of loss function. The prediction by FD-U-Net is compared to finite difference and data-driven U-net, demonstrating superior accuracy and computational efficiency. Moreover, the architecture of FD-U-Net also effectively handles heat conduction in composites of multi-phase or with internal heat source. By integrating a post-processing module, the FD-U-Net model enables direct derivation of homogenized material properties from microstructure images, highlighting its practical utility across applications. This work introduces a novel AI-physics approach for extracting, analyzing, and interpreting high-dimensional physical quantities and properties from digital images, incorporating properly defined boundary conditions.

Quantum dot light-emitting diodes (QLEDs) are promising for cost-effective and color-saturated lighting and display applications. As a key to realizing full-color emission of the emerging technology, blue QLEDs usually exhibit inferior performance compared with their red and green counterparts, stemming from additional challenges in achieving efficient and balanced charge injection in state-of-the-art devices. Herein, we develop a lithium chloride (LiCl)-assisted approach to simultaneously modulate defect and electrical properties of commonly used magnesium-doped zinc oxide (ZnMgO) nanocrystals, enabling blue QLEDs with obviously enhanced performance. We demonstrate that the chlorine ions of LiCl induce effective defect passivation of resultant nanocrystals, which effectively alleviates problematic interfacial exciton quenching between ZnMgO and the emissive layer. Meanwhile, the LiCl-induced modulations on the electrical property and energy level structure of ZnMgO ensure enhanced charge-injection balance in resultant devices. The champion QLED with emission peaking at 471 nm demonstrates a high external quantum efficiency (EQE) of 17.0% and a maximum luminance of 145,153 cd/m2 at a low driving voltage of 6 V, representing one of the best-performing blue QLEDs reported to date. Moreover, the device shows a measured T95 lifetime of 12.5 h at 1,092 cd/m2, which is over five-fold improvement than devices using conventional ZnMgO nanocrystals.

Tin (Sn)-halide perovskites are emerging environmentally-friendly candidates for achieving near-infrared light-emitting diodes with extended emission beyond their toxic lead counterparts. Despite the rapid development of Sn-based perovskite LEDs (PeLEDs), a thorough understanding and advanced modulations of the crystallization engineering of Sn-perovskites remain necessary to further boost the performance characteristics of state-of-the-art devices. Herein, the additive engineering of typical amino acids of L-phenylalanine (Phe) and L-tyrosine (Tyr) is comprehensively investigated in manipulating the efficient heterostructured Sn-perovskite emissive layers and ensuing PeLEDs. It is demonstrated that the Tyr possessing an additional reductive phenol moiety than that of Phe functions more effectively in retarding the fast crystallization and inhibiting the Sn2+ oxidation, ensuring the fabrication of Sn-perovskite emissive layers with high uniformity and enhanced luminescence properties. More importantly, the incorporated Tyr molecules induce increased electron density of Sn2+ nuclei, which promotes the formation of low-dimensional perovskite components at the bottom of the heterostructured emissive layers and enables more efficient charge injection in the resultant PeLEDs. The optimized Tyr-derived device demonstrates a high peak external quantum efficiency (EQE) of 13.2% along with a maximum radiance of 165 W sr-1 m-2, which sets a new benchmark for Sn-based near-infrared PeLEDs.

This study proposes an active-passive hybrid vibration isolator using piezoelectric (PZT)-enhanced quasi-zerostiffness (QZS) units. The design features bi-cosine curved beams with PZT actuators replacing central constraints, enabling precise deflection control and leveraging inherent QZS properties for lowfrequency suppression. Theoretical modeling confirms the nonlinear dynamics and QZS behavior. A PID control converts displacement errors into PZT actuation. Simulations demonstrate that PZT intervention significantly reduces vibration transmissibility near resonance across varied conditions. Experiments validate the design: under low-frequency excitation, active control effectively suppresses displacement response and maintains it below critical levels within the resonance band, while passive QZS provides isolation elsewhere. The synergy of passive QZS and active PZT control achieves enhanced low-frequency vibration isolation, validated numerically and experimentally.

X-ray absorption spectroscopy (XAS) is a critical analytical technique for comprehensively characterizing the electronic configurations and atomic structures of materials. The rapid growth in both data volume and complexity, driven by modern synchrotron radiation facilities, necessitates computational frameworks capable of efficiently processing large-scale XAS datasets. To address this need, we introduce XASDAML, a machine-learning-based platform that integrates the entire data processing workflow. The framework coordinates key operational processes, including spectral-structural descriptor generation, predictive modeling and performance validation, while facilitating statistical analyses through principal component decomposition and clustering algorithms to uncover latent patterns within datasets. Designed with modular architecture, the system enables independent modification or enhancement of individual components, ensuring flexibility to meet evolving analytical demands. Implemented through a Jupyter Notebook-based interface, the platform ensures accessibility for researchers. The framework is validated with two case studies: (i) copper-foil EXAFS data show that it can predict coordination numbers and radial distribution functions; and (ii) XANES spectra of the spin-crossover complex Fe(phen)3 uncover bond-length changes between the low-spin and high-spin states. Comprehensive validation highlights robust toolkit functionalities, including statistical descriptor analyses, spectral visualization, and prediction of widely employed structural descriptors closely reflecting local atomic environments. By establishing standardized and extensible procedures for integrating machine learning into XAS analysis, XASDAML enhances research efficiency, promotes richer data insights, and provides a versatile computational resource tailored to the expanding needs of XAS research.

Origami structures hold promising potential in space applications, such as ultra-large-area solar arrays, deployable space stations, and extra-terrestrial modular foldable buildings. However, the development of thick-panel origami structures has been limited, relying on a few typical origami patterns without a comprehensive design theory for multi-crease, multi-vertex thick-panel configurations. Additionally, realizing closed Polyhedra in thick-panel origami presents substantial challenges. Here, we introduce a design methodology inspired by origami and kirigami principles for one-degree-of-freedom (one-DOF) flat-foldable thick-panel origami-kirigami structures, including modular scalable arrays and closed polyhedral structures. The thick-panel origami-kirigami modular scalable arrays incorporate mixed four-crease vertices and (2n + 4)-crease vertices, enabling one-DOF flat-foldability and modular expansion of thick-panel units. The thick-panel origami-kirigami closed polyhedral structures, including tetrahedrons, square pyramids and triangular prisms, possess one-DOF inward-flat-foldability and structural closure after unfolding. This novel design framework for thick-panel origami-kirigami structures is capable of structural design from centimeter to meter scale, validated by kinematic analysis and prototype experiments. Chong Zhao and colleagues present a novel design methodology for thick-panel origami-kirigami structures, enabling modular scalable arrays and closed polyhedral structures. This approach combines origami and kirigami principles to facilitate structural design across varying scales, supported by kinematic analysis and prototype validation.

We present a dedicated and comprehensive database for experimental X-ray absorption spectroscopy (XAS) data, an integrated and user-friendly platform that combines spectrum visualization, raw data processing, spectrum matching and downloading, thereby offering a holistic solution for understanding XAS data. Leveraging the unique nature of XAS data, we have designed the MySQL table structure and developed professional plotting tools to present the data. Furthermore, the database - named XASDB - incorporates a variety of data processing tools, including data normalization as well as flexible searching and downloading options. A toolkit called XASMatch is integrated in the platform, enabling users to identify the most similar spectra within the database to their input spectra and rank with scores. An API interface is also provided to facilitate sharing data with other XAS databases and users. A total of 152 spectra measured at Beijing Synchrotron Radiation Facility for metals, oxides and minerals standards are included in the database. Applications integrated with artificial intelligence are advancing, since more high solution spectra with new standards will be available after operation of a high energy photon source to be commissioned at the end of the year.

Detecting the thermal environment of extraterrestrial planets in situ is crucial for understanding Earth and terrestrial planets' formation and evolution. Beyond measuring the thermal properties of planetary regolith, it's essential to measure thermophysical parameters within complex stratigraphic mixtures, including rocks and regolith, to study internal heat flow distribution effectively.This paper introduces an improved accuracy measurement system for thermophysical parameters, utilizing a small probe designed specifically for lunar rocks. The system employs the transient plane source method, capable of measuring temperature changes and heat flux during rock heating on the Moon. Furthermore, it presents an inversion model that combines artificial neural networks and genetic algorithms to accurately determine the thermophysical parameters of the rocks being tested. The accuracy of both the system and inversion model is verified through atmospheric and vacuum experimental setups. The inversion error for thermal conductivity is consistently under 5%, while for density and specific heat capacity, it ranges around 10%. This paper serves as a valuable reference for future in-situ thermal detection on the Moon, offering a reliable detection method and inversion approach.

Lead-halide perovskite light-emitting diodes (PeLEDs) are intrinsically capable of delivering high efficiency at high current densities compared to conventional solution-processed light-emitting diodes. While such performance and relevant high radiance have been well demonstrated in green and near-infrared ones, blue PeLEDs have lagged far behind due to extremely severe luminance-efficiency roll-off, especially in the pure-blue region (<480 nm, a CIEy coordinate below 0.15). Here, by tackling the critical limitations of phosphonic acid functional carbazoles (PACs) as hole injection layers and simultaneously leveraging their advantages on hole injection, we achieved ultrabright pure blue PeLEDs with minimized efficiency roll-off at high brightness with a CIEy coordinate below 0.15. We show that devices based on prevailing small-molecule PACs generally exhibit significant leakage currents. This is due to a synergistic effect of uneven surface coverage from reverse micelle formation and the nanoisland structure of thin-film lead-halide perovskite emitters. By using polymeric PACs instead, we demonstrate bright blue PeLEDs showing a peak luminance of ∼29 800 cd m-2 (478 nm, at a CIEy coordinate below 0.15). We also achieve a high brightness reaching ∼140 000 cd m-2 under pulsed driven. Our study not only provides a useful guidance for developing bright blue PeLEDs but also resolves a long-standing puzzle regarding the interfacial properties of PACs and their impact on hole transport, and it helps with the further design of these materials for lead-halide perovskite applications.