Small Science最新文献

2D material (2DM)-based reservoir computing (RC) systems combine the advantages of low-power hardware implementation with lightweight neural network architectures capable of processing complex temporal patterns through minimal training overhead, positioning them as ideal platforms for edge artificial intelligence (AI) applications. Here, a homogeneous RC system via defect engineering in PdSe₂ charge-trap memory (CTM) by ultrafast photoexcitation is demonstrated, which directly generates PdSe_2-xO_x nanodefects, converting volatile states (≈0% retention) into nonvolatile states (≈80% retention) by introducing electron-depleting defects and scattering centers in PdSe₂ channel. This engineering extends relaxation time constants from 15.6 s to 99.4 s and enables multilevel memory (>2⁶ levels) with prolonged retention (>2000 s). Leveraging dual nonlinear/stable operational modes, the physically integrated RC system achieves 91.7% (MNIST) and 93.3% (spoken digits) classification accuracy. Notably, it pioneers electrocardiogram arrhythmia detection (N, L, R, A, and V classes) with 92.3% accuracy, surpassing existing in-memory computing approaches. By establishing a defect engineering paradigm for material-intrinsic neuromorphic devices, this work advances energy-efficient AI hardware for biomedical diagnostics and edge computing applications.

Strain-dependent electronic and optical properties are one of the most appealing features of 2D semiconductors, like monolayer (1L) MoS₂. However, measuring and controlling the homogeneity of strain within the channel is crucial for next-generation MoS₂ field-effect transistors (FETs). This article reports a multiscale investigation of backgated FETs fabricated using large-area 1L MoS₂ flakes grown by liquid-precursor-intermediated chemical vapor deposition on SiO₂/Si substrates. The devices exhibit very attractive properties for ultra-low power applications, such as an I _on/I _off > 10⁶ and a normally off electrical behavior. The combination of temperature-dependent analyses of the FET transfer characteristics and nanoscale resolution potential mapping by Kelvin probe force microscopy shows a fully depleted MoS₂ channel at V _G = 0 and an effective Schottky barrier Φ_B,FB = 0.21 eV at flatband voltage V _FB = 17.9 V. An inhomogeneous tensile strain (ε) distribution along the channel length is revealed by micro-Raman and photoluminescence (PL) mapping, with a reduced ε and blue-shifted PL energy close to the Ni/Au source/drain contacts, suggesting a biaxial compression of 1L MoS₂ induced by metal deposition. The implications of these observations on the effective mass m_eff variation along the channel and the current injection from source/drain contacts have been discussed in the perspective of future ultra-scaled-devices applications.

Organism design incorporates diverse materials with varying properties, such as hard skeletons of biogenic minerals and soft organic skins. However, achieving a balance of flexibility, resilience, and hardness remains a challenge even for organisms. Door snails have a calcareous door (clausilium) that covers the aperture. The clausilium combines hardness for defense and flexibility for opening and closing. Here, this work focuses on the biogenic design of a clausilium stalk as a unique architecture balancing several properties. This study investigates the stalk, a twisted ribbon with high flexibility and resilience, which is identified as a synapomorphic structure in 22 Clausiliidae species across seven subfamilies and 17 tribes. Internal observations reveal a double-layered structure: a hard, dense envelope with aragonite rods arranged in the b axis and a flexible, low-density core with randomly packed aragonite nanoparticles and organic matter. The anisotropic hierarchical design seen in nature is surely useful in the development of artificial materials that combine flexibility, resilience, and hardness.

Developing high-efficiency perovskite/silicon tandem solar cells (PSTs) using scalable deposition methods is crucial for the industrialization of next-generation photovoltaics. However, developing industrially viable deposition techniques to ensure high performance, uniformity, and compatibility with existing silicon manufacturing remains a key challenge. A scalable hybrid two-step deposition process, combining evaporation and inkjet printing, is presented for fabrication of high-performance PSTs. Wide bandgap perovskite solar cells are achieved with power conversion efficiencies (PCEs) of up to 19.8%. Applying this approach to textured silicon bottom cells, the process ensures conformal perovskite growth, critical for industry-relevant tandem integration. Using this technique, highly efficient, fully textured PSTs with a PCE of 27.4% are fabricated. Homogeneous perovskite thin films are formed up to the substrate's very edge, enabling industry standards for silicon edge isolation. These results highlight the potential of hybrid two-step inkjet printing for scalable, high-efficiency PST fabrication, paving the way for industrial adoption.

Characterizing the 3D complex energy materials interface is critical to understand the correlative relationship between performance, degradation, and structures. Unfortunately, the resolution of microscopy and image acquisition speed are limited by the nature of the hardware, causing high-throughput characterization of energy materials to be prohibitive. Herein, REMind, a generative diffusion artificial intelligence model for fast and accurate reconstruction of electrode microstructures via focused ion beam-scanning electron microscopy, is presented. REMind can generate high-resolution internal microstructures between two low-resolution surfaces after training on sufficient high-resolution microstructures, enabling larger milling thickness between slices while keeping high-fidelity imaging. REMind is first demonstrated for reconstructing solid oxide fuel cell (SOFC) anode microstructures. REMind resolves relevant multi-scale structures with low pixel-wise reconstruction error (<10%) and quantifies the generated uncertainty by calculating the generated entropy. Additionally, a multi-scale multi-physics SOFC model is employed to further quantify the reconstructed error regarding the electrochemical performance, i.e., operating current density versus overpotential. REMind shows good transferability, as proven by its ability to reconstruct other energy materials, including catalyst layers of proton exchange membrane fuel cells and solid-state battery composite electrodes, demonstrating the potential for REMind to be used as a general-purpose platform for broad development of energy technology.

Recent advances in liquid-phase transmission electron microscopy (TEM) have enabled the direct visualization of reaction pathways of nanomaterials, providing critical insights into diverse nanoscale processes such as crystallization, phase transition, shape transformation, etching, and nanoparticle motions. Among various liquid cells, graphene liquid cells (GLCs) are particularly advantageous due to the intrinsic properties of graphene-high electrical and thermal conductivity, exceptional mechanical flexibility, and radical scavenging effects-which allow atomic-scale spatial resolution and enhanced imaging stability. This review article highlights the recent progress in GLC-based liquid-phase TEM, focusing on the evolution of structural designs, including veil-type, well-type, liquid-flowing-type, and mixing-type GLCs. Each configuration offers unique advantages tailored to observing distinct types of nanoscale dynamic processes. These studies have elucidated both classical reaction pathways and complex, nonclassical mechanisms involving transient intermediates. Overall, this review highlights how developments in GLC designs have significantly advanced the capabilities of in situ liquid-phase TEM, providing unprecedented opportunities to study nanoscale processes at atomic resolution.

Sustained mechanical stimulation represents a powerful strategy for directing stem cell fate, yet its application within microscale injectable carriers remains limited. This study presents a dynamic microgel platform enabling osteogenic differentiation of single mesenchymal stem cells (MSCs) solely through hydrostatic pressure, without biochemical induction. Individual MSCs are encapsulated in ionically crosslinked, cell-adhesive alginate microgels and stabilized using an alginate-poly-l-lysine-alginate and calcium coating. Application of cyclic hydrostatic pressure at 200 kPa and 0.5 Hz frequency for 30 min per day leads to upregulation of early osteogenic markers RUNX2 and alkaline phosphatase, enhanced collagen I synthesis, and mineralization over 21 days. Results demonstrate that mechanical cues alone are sufficient to orchestrate osteogenic commitment in soft, confined microenvironments, offering a scalable approach to stem cell programming. This work establishes a versatile, high-resolution platform for engineering lineage specification at the single-cell level and highlights the potential of force-driven strategies for scalable production of therapeutic stem cells.

Carbon nanotube (CNT)-based hydrogels have the potential to serve as 3D platforms for nerve regeneration. However, the interplay between different cues governing the formation of a complex tissue-like cellular structure is still ambiguous. Herein, two approaches are adopted to develop PVA/CNT hydrogels using phase inversion method and low kinetic gelation, enabling unprecedented CNT loading capacity (75% w/w) without compromising their elasticity. By controlling key factors affecting cell coverage and maturation, including Young's modulus (YM), CNT concentration, and pore size, distinct thresholds are identified where these factors dominate cell coverage. Results demonstrated that when CNT exceeds 60% w/w or a coating is applied to enhance CNT-cell interaction, CNT effect dominates, increasing cell coverage with increasing CNT concentration. However, below a specific YM threshold, YM dominates cell growth, covering up to 50% of the scaffold surface regardless of CNT concentration or exposure. Lastly, controlling the pore size to 100-250 μm further increased cell coverage to >70%, breaking through previous plateau and upregulating TUBB3 maturity marker. Additionally, certain key factors are seen to synergistically codominate in determining cell growth.

Charge transfer (CT) interactions have rarely been used to organize supramolecules and cross-linked objects. However, the ever-increasing understanding of CT interactions opens up new avenues for the design of innovative materials with tailored electronic properties. Herein, several molar ratios of highly crystalline π-conjugated n%TCNQ@Sq-1,6Py oligomers (n equal molar ratio of tetracyanoquinodimethane (TCNQ) to 1,6-diaminopyrene (1,6Py) moiety) with simultaneously stable intra- and intermolecular CT mechanisms are prepared. As a result, the π-conjugated 200%TCNQ@Sq-1,6Py CT complex indicates stable intra- and intermolecular CT interactions resulting in extremely high electrical conductivity of 8.7 × 10^-2 S cm^-1 at room temperature, a charge-distance capacitance of 70.62 F g^-1 at the current density of 0.625 A g^-1 which significantly increases to 968.7 F g^-1 by doping of polyaniline (PANI) at a current density of 0.312 A g^-1. Finally, it exhibits a capacitance retention of 70% of the initial specific capacitance after 1000 cycles at room temperature. This type of π-conjugated oligomer CT complex can be used to improve existing CT-based energy storage devices, such as capacitors.

2D Ti₃C₂T_x MXene offers high electrical conductivity and a large surface area, making it attractive for electrocatalysis. However, its intrinsic hydrogen evolution reaction (HER) activity remains poor due to the lack of active catalytic sites. To activate the otherwise inert surface, platinum monosulfide (PtS) nanoparticles are synthesized directly on Ti₃C₂T_x nanosheets via thermal decomposition of a single-source precursor, Pt(dmampS)₂, in a solution-based process. This direct growth strategy enables uniform dispersion of PtS nanoparticles and intimate interfacial contact with the MXene surface, without the need for binders or surfactants. The resulting PtS/Ti₃C₂T_x heterostructure exhibits significantly enhanced HER performance, achieving a low overpotential of -104 mV at a current density of -10 mA cm^-2 and a Tafel slope of 48.3 mV dec^-1.