In n-i-p perovskite solar cells, the SnO2/perovskite buried interface plays a decisive role in governing crystallization and charge extraction. Here, trisodium methylglycine diacetate (MGDA) is introduced onto SnO2 to regulate precursor behavior during crystallization. Molecular-dynamics simulations using PbI2∙DMSO as a representative solvent-coordinated lead-iodide species show that MGDA increases interfacial precursor localization while restricting lateral motion and aggregation. This modified precursor state is accompanied by slower but more extended late-stage growth/reorganization during thermal annealing, leading to larger grains, fewer buried-interface voids, improved crystallinity, and lower residual tensile stress. The optimized devices deliver a power conversion efficiency of 25.50%, while 7 × 7 cm[2] rigid and 5 × 5 cm[2] flexible mini-modules reach 22.49% and 17.38%, respectively, with improved ambient-storage and thermal stability. These results reveal the link between interfacial precursor behavior, subsequent crystallization, and device performance, providing guidance for buried-interface design in perovskite solar cells.

Solid-solution catalysts hold great promise as low-cost catalysts for the growth of small-diameter single-walled carbon nanotubes (SWCNTs); however, their catalytic performance is limited by low metal-site utilization and unfavorable interfacial structures in conventional architectures. Here, we report a facile design strategy for FeMgAlOx solid-solution catalysts featuring abundant active sites and enhanced interfacial structures that suppress nanocluster agglomeration. The successful synthesis of this catalyst relies on a high-temperature-induced two-step strategy: high-temperature oxidation enhances crystallinity, while high-temperature air/CO2 treatment facilitates the formation of strong metal-support interactions (SMSI) on inert metal oxide supports. In this architecture, the ultrathin two-dimensional morphology and ordered atomic arrangement promote active-metal spillover, yielding abundant active sites; meanwhile, CO2 induction significantly enhances the SMSI-mediated confinement capability for Fe nanoclusters. Crystalline evolution and SMSI regulation synergistically facilitate the exsolution of active Fe species from the bulk solid solution, thereby forming thermally stable Fe nanoclusters. Furthermore, optimized single-walled carbon nanotube (SWCNT) growth kinetics confirm that the modified solid solution possesses both a higher density of active sites and enhanced intrinsic catalytic activity. Ultimately, our catalyst enables the growth of SWCNTs with diameters ranging from 0.96 to 3.33 nm, offering a simple and low-cost route for SWCNT synthesis with broad application prospects.

Compact all-solid-state lasers require nonlinear optical crystals that combine strong second-harmonic generation (SHG) with large birefringence, yet planar π-conjugated molecules often pack antiparallel, canceling their macroscopic second-order response. Here, we break this trade-off through dimensional competition between halogen bonding and π-π stacking in two nitroazole-based organic frameworks. The parent (IClPzNO2)2·H2O forms a centrosymmetric two-dimensional framework with coplanar chromophores and a birefringence of 0.705 at 546 nm. Combined skeletal and halogen substitutional editing, from pyrazole (Pz) to imidazole (Im) and from an ortho-I/Cl to a meta-I/I substitution pattern, redirects in-plane hydrogen-bonding aggregation via relocating the N─H donor, and then extends C─I···O connectivity from two to three dimensions. The resulting halogen-bonded network governs chromophore orientation and produces a twofold-interpenetrated polar framework, (I2ImNO2)2·H2O (Pna21). The crystal exhibits a phase-matchable powder SHG response 20 times that of KDP (KH2PO4), the largest value among five-membered-ring and nitro/nitrate NLO crystals, as well as a giant birefringence of 0.554 at 546 nm. To the best of our knowledge, this combination occupies a previously unreported performance regime. These results identify interaction dimensionality as a practical design parameter for aligning polar π-conjugated molecules and combining strong SHG with giant optical anisotropy.

The sluggish kinetics of the oxygen evolution reaction (OER) impede efficient water electrolysis, motivating the development of robust, non-precious-metal electrocatalysts. Here, we construct a self-supported bimetallic metal-organic framework precatalyst (PAx-NiCo-BDC) through a two-step strategy using phytic acid (PA) for mild etching and multidentate coordination. This treatment introduces oxygen-vacancy-related coordination defects and regulates the precursor morphology. The optimized PA3-NiCo-BDC requires only 212 mV to reach 10 mA cm[-] [2], exhibits a Tafel slope of 20.2 mV dec[-] [1], and shows excellent durability, outperforming commercial RuO2. An anion-exchange-membrane water electrolyzer employing PA3-NiCo-BDC delivers 1 A cm[-] [2] at 1.72 V and operates stably for 300 h. In situ Raman spectroscopy reveals that PA-induced defects enable the lower-potential formation of Ni[3+]-O-containing NiOOH-like surface species. Differential electrochemical mass spectrometry identifies the adsorbate evolution mechanism as the dominant OER pathway. Density functional theory calculations further show that defect-induced d-band-center modulation optimizes the adsorption and desorption of OER intermediates at Ni sites and lowers the OER energy barrier. This green coordination-engineering strategy provides a versatile route to efficient and durable MOF-based electrodes for practical water electrolysis.

Hydrostatic preload increases response nonlinearity and limits weak-force detection in underwater tactile sensing. Here, we present a nanocomposite-enhanced iontronic tactile sensor (NeITS) that combines high sensitivity with wide-range linearity through a coupled mechano-ionic mechanism. Hydrophobic SiO2 nanoparticles reinforce the ionogel, suppress premature geometric saturation, and sustain contact-area evolution, while pressure-driven interfacial ion enrichment enhances electric double-layer capacitance. The optimized sensor achieves a sensitivity of 55.36 kPa[─] [1] over a 500 kPa linear range, with response and recovery times of 50 ms. It resolves an additional contact pressure of approximately 2 kPa at hydrostatic preloads of 0, 93, and 116 kPa with closely comparable response amplitudes. A 4 × 4 NeITS array enables real-time pressure mapping and shape imaging. Integrated into an underwater robotic gripper, the array supports neural-network-assisted recognition of nine representative benthic biological targets, achieving a mean classification accuracy of 97.8% under temporally separated window-level evaluation. This work establishes a materials-level strategy for hydrostatic-pressure-tolerant iontronic sensing and advances tactile perception for underwater robotic manipulation.

Neuromorphic visual computing aims to fuse sensing, memory, and processing into a single hardware stack, enabling reliable analog weight updates. Defect engineering in two-dimensional (2D) semiconductors is an effective strategy for achieving such synaptic functionality. However, in most 2D materials, the defects that confer plasticity also compromise lattice stability and degrade intrinsic performance. This work demonstrates that Bi2O2Se provides a robust, defect-tolerant platform. The growth process introduces selenium vacancy (VSe) without compromising covalent structure or the approximately 0.8 eV bandgap, yielding stable and programmable trap centers. Combined with a low-trap hexagonal boron nitride (h-BN) gate dielectric that suppresses extra interfacial states, the resulting Bi2O2Se/h-BN transistor exhibits analog conductance modulation, excitatory postsynaptic currents, double-pulse facilitation, and long-term potentiation/inhibition, supporting hybrid electro-optic learning via vacancy-assisted charge trapping and persistent photoconductivity. Mapping measured device characteristics onto designed convolutional neural networks achieves 90.1% accuracy in complex image recognition, establishing an optoelectrical neuromorphic multifunctional platform for edge intelligence systems.

The inevitable recurrence of glioblastoma multiforme post-resection is primarily driven by infiltrative residual cells and a profoundly immunosuppressive tumor microenvironment (TME) that evades innate immune surveillance. In the present study, an injectable, photo-crosslinkable carboxymethyl chitosan (CMCS)-based hydrogel (si-HAX1&Quercetin@Gel) was engineered to achieve localized, synergistic chemo-immunotherapy within the surgical cavity. The findings demonstrate that the localized release of si-HAX1 triggers robust endoplasmic reticulum stress-mediated immunogenic cell death, leading to significant CRT surface translocation. Concurrently, Quercetin acts as a small-molecule antagonist of the CD47-signal regulatory protein α axis, dismantling the "don't eat me" shield that protects glioma cells from macrophage-mediated phagocytosis. In an in vivo postoperative glioblastoma model, the si-HAX1&Quercetin@Gel effectively suppressed tumor regrowth, increased the M1/M2 macrophage ratio and recruited CD8+ T cells, thereby extending median survival time. Moreover, this localized intervention sensitized residual tumors to intratumorally administered anti-programmed cell death protein 1 therapy by upregulating IFN-γ and remodeling the TME. By integrating biomaterial engineering with deep molecular targeting of HAX1 and CD47, we provide a robust strategy to transform the post-surgical cavity into an active immunological zone, offering a potent solution to combat malignant glioma recurrence.

The quest to employ colloidal self-assembly as a scalable means of fabricating photonic crystals with a complete photonic band gap at optical frequencies has proved challenging throughout the first quarter of the 21st century. Meanwhile, the field of topological photonics has emerged, presenting new, and largely unexploited, functionalities for self-assembled colloidal materials. The robustness of topological photonic properties to defects that are practically unavoidable in self-assembled colloidal structures appears to be especially attractive. Here, the self-assembly of a colloidal honeycomb crystal is demonstrated in computer simulations, with an assembly strategy that not only suppresses the formation of real-space topological defects observed in experiments but also results in topological photonic properties in reciprocal space. This assembly strategy promotes the emergence of long-range order and allows for the breaking of spatial inversion symmetry with two distinct sub-lattices. A sufficient dielectric contrast between the colloidal particles sitting on these two sub-lattices is exploited to realize a valley-Hall topological photonic phase, allowing for unidirectional propagation of circularly polarized light through waveguides composed of the designer crystals. The design rules are tested for their robustness to open the door to scalable fabrication of topological photonic devices via colloidal self-assembly.

Polymer-based conductive aerogels are promising for flexible piezoresistive sensors but suffer from skeleton deformation and conductive filler detachment under high-strain and long-term compression. Here, an aminopropyltriethoxysilane/carboxylated carbon nanotubes/waterborne polyurethane (APTS/CNTs/WPU) aerogel featuring a semi-interpenetrating network is developed through a strategy combining low-temperature demulsification-induced gelation with gel-confined freeze-casting and simultaneously possesses a highly elastic skeleton, continuous conductive pathways, and reinforced filler/matrix interfaces. The low-temperature condition retards the electrostatic demulsification rate between anionic WPU and protonated APTS, transforming the otherwise rapidly destabilizing emulsion into a controllable gelation process. Accompanied by the gradual occurrence of electrostatic interactions, hydrogen bonding, and hydrolysis-condensation reactions of silane, a semi-interpenetrating gel network is established within minutes. The formed gel network prevents re-agglomeration of CNTs and ensures their stable immobilization within the skeleton during freeze-casting. The aerogel exhibits 99.1% elastic recovery after 10 cycles at 80% strain and 5.1% plastic deformation after 10 000 cycles. The piezoresistive sensor achieves a detection range of 0-408 kPa, a maximum sensitivity of 12.33 kPa[-1], and stable response over 10 000 cycles. Furthermore, a sensing array integrated with a wireless/wired signal acquisition system is constructed for monitoring gait pressure distribution, fully demonstrating its substantial application potential in human motion perception and wearable electronics.

Enzyme-driven micromotors provide a promising platform for mimicking biological chemotaxis and for developing responsive materials. We recently developed a method to make cell-sized asymmetric urease-functionalized poly(lactic-co-glycolic acid) (PLGA) microcapsules that are highly motile in the presence of urea. Here, we investigate the chemotactic behavior of these microcapsules in urea gradients in a chemotaxis chamber. We demonstrate that urease-functionalized microcapsules exhibit negative chemotaxis, migrating down urea gradients over a wide range of gradient magnitudes. The chemotactic flux increases with increasing urea gradients but saturates at high gradients, whereas increasing the average substrate concentration in a fixed gradient suppresses directional migration. We find that the direction of migration is independent of the orientation of the asymmetric particles. Buffered environments suppress chemotaxis, supporting a mechanism of motion that involves local pH changes or ionic self-diffusiophoresis. A transport model of substrate consumption demonstrates that chemotactic behavior correlates with the relative concentration difference across the particle, whereas particle speed scales with the absolute concentration/corresponding osmotic pressure difference. These results establish negative chemotaxis as a robust feature of urease-functionalized polymeric microcapsules and provide mechanistic insight into how enzymatic activity, substrate gradients, and ionic environments govern directional motion.

The discovery of in-plane ordered MAX (i-MAX) phases containing lanthanides provides a promising route to novel MXenes. Most MXenes have transition elements occupying the M site. However, the distinctive optical and magnetic properties of lanthanides could yield multifunctional MAX phases and MXenes with applications in aeronautics, radiation shielding, and space technology. Previously, lanthanide-containing i-MAX phases were etched to produce ordered vacancy MXenes via lanthanide removal. Here, we report the first synthesis of (Mo2/3Nd1/3)2CTx i-MXene retaining Nd and compare wet chemical etching, fluorine-free molten salt etching, and hydrothermal etching routes. A key challenge is removing Al without also removing Nd. Etching with concentrated hydrofluoric acid (HF) yielded i-MXene along with neodymium fluoride. Etching using lithium fluoride (LiF) and hydrochloric acid (HCl) produced an accordion-like morphology typical of MXenes. Mixed acids (HF and HCl) yielded more uniform etching compared to other etchants. Both molten salt etching and alkali etching proved less effective. We discuss the potential and challenges of each etching method. Based on the similar chemical behavior of lanthanides, we expect results for the Nd-containing i-MAX phase to be representative. This comparative study reveals significant morphological and structural differences, paving the way for the synthesis of a new family of lanthanide-containing i-MXenes.

Rapid advancement of wearable electronics, the Internet of Things, and artificial intelligence has generated a growing demand for flexible sensors capable of reliable and high-precision pressure monitoring under complex conditions. However, traditional piezoresistive sensors often suffer from sensitivity degradation and structural fatigue caused by moisture exposure and long-term cyclic loading. Here, we report a ternary composite aerogel-based piezoresistive sensor composed of tobacco stem-derived TEMPO-oxidized cellulose nanofibers (TTOCNFs), MXene, and polymethylsilsesquioxane (PMSQ) through cross-scale interfacial engineering. In this design, TTOCNFs form a flexible skeleton, MXene establishes an efficient conductive network, and PMSQ reinforces the framework while imparting hydrophobicity, mitigating moisture-induced performance degradation. An ice-templating strategy further constructs a honeycomb-like skeleton, enabling efficient stress distribution under mechanical deformation. Benefiting from these structural and interfacial synergies at different scales, the sensor exhibits high sensitivity (1612.8 kPa[-1]), broad detection range (2.4 Pa-48.8 kPa), rapid response and recovery (117 and 80 ms), and outstanding cycling stability (>1000 cycles), while maintaining reliable performance in humid environments. It enables high-fidelity monitoring of subtle physiological signals and human motions, demonstrating strong potential for health monitoring and human-machine interaction. This work provides a scalable and sustainable design strategy for environmentally robust aerogel-based pressure sensors toward smart wearable electronics.

This work demonstrates a sustainable upcycling strategy that transforms discarded mask straps into flexible direct-current triboelectric nanogenerators (DC-TENGs) through an eco-friendly NaCl-assisted modification. The PEDOT:PSS/PVA/Pullulan composite simultaneously serves as the electrode and triboelectric layer, enabling a simplified and metal-reduced fabrication. The saline treatment introduces Na[+]-PSS[-] coordination, which sustainably enhances molecular mobility, interfacial polarization, and charge transfer without the need for toxic dopants or complex post-processing. The resulting device delivers a peak voltage of ∼1 V, a current of ∼180 µA, and a maximum power density of 0.5 W m[-] [2] under mechanical deformation. By utilizing water-based NaCl as a benign ionic modifier, this approach exemplifies a circular and green pathway to improve energy harvesting performance while mitigating mask waste pollution. The developed NaCl-enabled DC-TENG offers a practical route toward sustainable, self-powered wearable and communication systems aligned with global carbon-neutral and SDG objectives.

Real-time, high-precision underwater monitoring and optical communication are critical for marine resource exploration yet remain hindered by the lack of photodetectors capable of stable, autonomous operation in harsh seawater environments. Herein, we report a robust, self-powered photoelectrochemical (PEC) deep-ultraviolet (DUV) photodetector based on an epitaxial La-doped SrSnO3 (LSSO)/SrRuO3 (SRO) architecture. The single-crystalline epitaxial LSSO film and the high-quality LSSO/SRO interface suppress defect-induced recombination and enable efficient separation and transport of photogenerated carriers. Through systematic optimization of La-doping concentration and operating bias, the device achieves a remarkable peak responsivity of 459.84 mA W[-1] in natural seawater, alongside rapid response speeds (rise/fall times of 2.98/2.22 ms). The device shows excellent long-term stability in natural seawater during PEC operation, with the photoactive surface directly exposed to the electrolyte. Specifically, the photocurrent remains nearly unchanged after 100 days of ambient storage and 30 days of immersion in natural seawater, retaining over 90% of its initial value after 2000 s of continuous cycling. This work demonstrates that epitaxial oxide thin films can effectively bridge the gap between high-performance optoelectronics and harsh-environment adaptability, paving the way for next-generation deep-sea exploration technologies.

Chemically powered microrobots provide a promising route toward untethered operation at small length scales, but their motion near surfaces is often difficult to predict because it depends strongly on both particle design and the local interfacial environment. Here, we investigate the two-dimensional surface locomotion of cell-sized self-propelled microrobots in aqueous hydrogen peroxide. Planar 15 µm microrobots were fabricated by photolithography with different structural materials and electrode configurations, and their motion was examined on substrates with distinct surface conditions. Variations in microrobot structure and substrate environment produced differences in mobility and locomotion mode, including translational, circular, and spiral trajectories. Our force-balance analysis indicates that these behaviors reflect the combined effects of self-phoretic propulsion driven by ionic gradients, viscous drag, and surface interactions. We further find that translational velocity is coupled to trajectory curvature under different interfacial conditions. This relationship is captured by a reduced kinematic model and is consistent with both experiments and simulations. Together, these results provide an experimental and physical framework for understanding and tuning the surface locomotion of chemically powered microrobots.