Jing Zhang

Nature Photonics, Published online: 19 December 2025; doi:10.1038/s41566-025-01818-7

A nanostencil lithography technique enables fabricating arrays of green-emitting OLEDs with pixels as small as 100 nm and an external quantum efficiency of 13.1%.

A previously unreported selective diffusion behavior governed by chemical heterogeneity is described, along with the demonstration of a blocking effect of the ZrB₂ intergranular phase on Tb diffusion. These novel findings establish fundamental microstructure design principles that facilitate the high-efficiency diffusion of heavy rare earth elements, offering a practical pathway to overcome the coercivity bottleneck in grain boundary diffusion magnets with high Ce contents.

ABSTRACT

Unraveling the effects of chemical heterogeneity and intergranular phases on heavy rare earth (Dy/Tb) diffusion is crucial for developing high-performance, cost-effective grain boundary diffusion (GBD) magnets with high Ce substitution. Herein, a previously unreported selective diffusion behavior governed by chemical heterogeneity is uncovered, along with the demonstration of a blocking effect of the ZrB₂ phase on Tb diffusion. Comprehensive experimental and theoretical analysis reveals that Tb preferentially diffuses toward Nd-rich grains rather than LaCe-rich grains, thereby hindering Tb diffusion and weakening the magnetic hardening effect in the LaCe-rich grains. This accounts for the lower coercivity increment in multi-main-phase (MMP) GBD magnets (5.15 kOe) relative to that in single-main-phase (SMP) GBD magnets (5.52 kOe). Despite achieving a higher room-temperature coercivity (12.43 kOe), the MMP GBD magnets exhibit inferior thermal stability due to the weaker demagnetization resistance of LaCe-rich grains. Furthermore, the ZrB₂ phase exhibits significantly stronger blocking of Tb diffusion than the REFe₂ phase, which results in a non-uniform Tb distribution on the grain surfaces. These novel findings establish fundamental microstructure design principles that facilitate the high-efficiency diffusion of heavy rare earth elements, providing a viable pathway to overcome the coercivity bottleneck in GBD magnets with high Ce contents.

A unique saddle-shaped hierarchical lamellar phase (saddle-SML) featuring with negative Gaussian curvature—a defining feature of bicontinuous phases—while maintaining the topology of lamellar phase has been first obtained within a binary self-assembly system containing block copolymer and surfactant. This phase confirms the thermostability of the interfacial curvature intermediate, implying the existence of a smooth lamellar-bicontinuous transition pathway.

Abstract

Curvatures are fundamental parameters governing the topological changes and phase separations in soft matter systems. Explorations of Gaussian curvature, mean curvature, and their combination (as interfacial curvature linked to packing parameter) are essential for understanding the dynamic self-assembly processes, especially for the critical and classic lamellar-to-bicontinuous transition. However, whether intermediate states arising from the changes in these curvatures exist remains a long-standing controversy. Herein, this issue is addressed by identifying a unique saddle-shaped hierarchical lamellar phase in a binary self-assembly system containing polystyrene-b-poly(acrylic acid) (PS-b-PAA) and stearyltrimethylammonium bromide (STAB). This intermediate phase exhibits characteristics with negative Gaussian curvature similar to bicontinuous phase while maintaining lamellar topologies. Its existence attributes to the competing effects of the inserted STAB micelles imposed on PAA segments: electrostatic screening reduces PAA effective volume, while volume expansion increases it. This competition balance promotes the lamellae toward saddle-shapes with enhanced molecular capacity and modulates their interfacial curvature to a critical intermediate value that is rarely stable in block copolymer systems. The thermostability of this intermediate phase serves as compelling evidence for the smooth interfacial curvature evolution pathway. This discovery provides insights into phase transformations in biological membranes, marking a significant advance in understanding complex soft matter systems.

Nature Materials, Published online: 16 December 2025; doi:10.1038/s41563-025-02442-0

The small-bandgap semiconductors LnCd3P3 (Ln = La, Ce, Pr and Nd) are promising materials to study emergent phenomena from geometric frustration across bond, spin and charge degrees of freedom.

Rare-Earth luminescent (REL) materials are valued for their sharp multiwavelength emissions, photon up- and down-conversion, and strong x-ray absorption. MXenes, in contrast, are highly versatile two-dimensional (2D) materials with exceptional electronic and structural properties, yet their use as REL hosts remains largely unexplored. Here, we demonstrate, for the first time, rare-Earth-modified 2D MXene sheets as luminescent materials, establishing a new pathway for their integration into flexible, multifunctional, and miniaturized photonic systems. Europium (Eu3+) ions were incorporated into Mo2CTx nanosheets via hydrothermal treatment. At moderate temperature (230 °C), the layered structure was preserved with an expanded interlayer spacing, enabling ∼1-0 at.% Eu3+ incorporation into oxygen-rich Mo–O–C domains. At higher temperature (360 °C), the MXene layers collapsed, yielding highly crystalline MoOx and Mo–O–C phases with increased Eu content (∼15 at.%). Photoluminescence spectra exhibited sharp Eu3+⁵D0 → ⁷F transitions, dominated by a strong emission at 615 nm. The emission intensity was markedly enhanced at higher temperature due to improved Eu incorporation and oxide crystallinity. Importantly, this work establishes a luminescence mechanism in MXenes beyond previously reported quantum dots, demonstrating that layered MXene frameworks themselves can host rare-Earth optical centers. This unique non-quantum dot pathway highlights the potential of REL-functionalized MXenes for space-relevant applications requiring multifunctionality, such as multiwavelength lidar surveillance and communication, photon conversion from infrared and x-rays to visible light, electromagnetic shielding, and radiation protection, while retaining lightweight and mechanically flexible form factors.

This study reports a tenfold enhancement of mechanoluminescence in SrAl₂O₄:Dy³⁺, Eu²⁺ phosphors by Nd³⁺ doping. The material exhibits strong green light emission under both friction and ultrasound stimulation, alongside near-infrared persistent luminescence. Its multifunctionality enables anti-counterfeiting, ultrasound-triggered labeling, and dual-mode optical thermometry, highlighting its potential for advanced sensing and security technologies.

Abstract

Mechanoluminescence (ML) is an appealing phenomenon in which a material emits light in response to mechanical or acoustic stimulation. The article shows a novel strategy to enhance the persistent mechanoluminescence signal in SrAl₂O₄:Dy³⁺, Eu²⁺ phosphors by co-doping them with Nd³⁺ ions. The addition of Nd³⁺ significantly improves both friction-induced and ultrasound-induced persistent ML. An efficient energy transfer from Eu²⁺ to Nd³⁺ also introduces strong, long-lasting near-infrared (NIR) emission. The NIR persistent luminescence extends the material's functionality to deep-tissue bioimaging and optical thermometry based on lifetime and intensity ratio-based readouts. The SrAl₂O₄:Dy³⁺-Eu²⁺-Nd³⁺ phosphors, synthesized via a solid-state method, demonstrate robust and reproducible ML and photoluminescence responses, establishing their potential for real-world applications. As a proof of concept, the material in anti-counterfeiting labeling is successfully applied, where invisible handwriting became visible upon mechanical stimulation. The findings establish Nd³⁺ doping as an effective approach to enhance ML performance in the studied materials and open pathways for their integration into next-generation optoelectronics, sensing, and security technologies.

Nature Nanotechnology, Published online: 15 December 2025; doi:10.1038/s41565-025-02086-w

A tension-modulated DNA origami-based nanosensor, compatible with bulk biochemical analysis in cell lysates, is used to assess the force-induced interaction of proteins involved in cellular mechanotransduction.

The tear-and-stack method enables the creation of twisted SrTiO₃ bilayers with accurate twist-angle control, which yield atomically sharp oxide moiré superlattices with emergent exotic topological polar vortices, thereby opening a new pathway for twistronics based on 2D-like non-van der Waals oxides.

Abstract

Oxide-based moiré superlattice is an emerging field for its exotic properties and abundant design freedom. However, due to the dangling bond and much stronger interlayer adhesion with the supporting substrate, fabricating twisted complex oxide is challenging, and an oxide moiré lattice with a clean interface remains elusive. Here, square moiré superlattices in twisted SrTiO₃ (STO) bilayer are constructed using a 2D like “tear-and-stack” method, achieving unprecedent control resolution and superior interface quality. Through depth-dependent atomic-scale analysis and electronic reconstruction, the upper and lower STO layers are found near the twisted interface exhibit opposite shear strain, evidencing a strong coupling confined within 2 unit cells (±0.8 nm) from the interface. The strain gradient of twisted bilayer STO gives rise to alternating clockwise and counter-clockwise polarization originating from the flexoelectric effect, leading to a large-scale array of polar vortex. This motivates to fabricate twisted STO bilayers with a freestanding parent layer as thin as 0.8 nm, in which a polar vortex is also confirmed. The “tear-and-stack” method is generic to create twisted moiré superlattices in a wide range of oxide material systems, and it demonstrates the feasibility of “2D-like” oxide twistronics.

A water-assisted rapid evaporation strategy enables melt-free synthesis of transparent, multicolor luminescent hybrid lanthanide-based metal halide glasses. Triplet exciton-mediated energy transfer from phosphorescent organic cations to lanthanide ions significantly enhances the scintillation efficiency. The resulting (BuTPP)3TbCl6 glass demonstrates a high light yield and high transmittance, making it a promising candidate for high-resolution X-ray imaging.

Abstract

Organic–inorganic hybrid metal halide glasses are an emerging class of scintillators for high-resolution X-ray imaging, but their synthesis via conventional melt-quenching method is hindered by thermal decomposition of organic components and strong recrystallization tendency. Moreover, the energy dissipation during relaxation of secondary electrons further limits their scintillation performance. To address these issues, a water-assisted rapid evaporation method is developed that bypasses melting, enabling the synthesis of transparent, multicolor luminescent lanthanide-based metal halide glasses, (BuTPP)₃LnCl₆ (BuTPP⁺ = butyltriphenylphosphonium). A triplet exciton-mediated energy transfer mechanism is further introduced to enhance the utilization of secondary electrons. The incorporation of heavy lanthanides (such as Tb³⁺) enhances spin-orbit coupling in phosphorescent organic cations BuTPP⁺, promoting intersystem crossing and generating abundant triplet excitons during the relaxation of secondary electrons. These excitons undergo near-unity energy transfer to Tb³⁺ ions, facilitated by optimal energy level alignment (BuTPP⁺ triplet: 21739 cm⁻¹; Tb^{3+ 5}D₄: 20492 cm⁻¹), thereby minimizing energy dissipation and yielding intense radioluminescence. The representative (BuTPP)₃TbCl₆ glass exhibits a high light yield comparable to commercial Bi₄Ge₃O₁₂ scintillator, outstanding spatial resolution (26.8 lp mm⁻¹), and robust radiation stability. This work not only provides a scalable synthesis strategy for hybrid glasses but also establishes an effective exciton management strategy for high-performance scintillators.

Nature, Published online: 02 December 2025; doi:10.1038/s41586-025-09954-4

Bulk superconductivity up to 96 K in pressurized nickelate single crystals

Nature, Published online: 02 December 2025; doi:10.1038/d41586-025-03933-5

The single-celled organism can grow at 63 °C, a record for eukaryotic life.

Nature Photonics, Published online: 02 December 2025; doi:10.1038/s41566-025-01811-0

Vitrification of polymer solutions yields ultrasmall fluorescent polymer dots that combine dye-like size with nanoparticle brightness, enabling nanometre-precision live-cell tracking on standard microscopes.

Ultra-durable cost-effective information-encoded anti-counterfeiting labels are fabricated to secure semiconductor chips. A novel method is used to self-assemble colloidal quantum wells (CQWs) into color bars. Information can be encoded spatially, spectrally, and opto-spatially. The randomness in the boundaries of the stripes makes the labels unique and hard to replicate. These labels have high stability against high temperature, humidity, and  ultraviolet (UV) exposure.

Abstract

Forgery, a serious universal problem, is causing huge economic losses every year. Against forgery, information-encoded labelling systems have attracted significant attention for a diverse range of anti-counterfeiting applications. Here, cost-effective and ultra-durable nanocrystal-based labels are proposed and demonstrated in which information can be encoded as physically unclonable functions (PUFs) of hardware-oriented security systems. The fabrication method of the PUFs is based on the self-assembly of colloidal quantum wells (CQWs) and generation of unclonable features within their pattern at a liquid–liquid interface. These CQW PUFs are analyzed with well-known statistical tests, which show a uniqueness level of 0.5060 ± 0.0323 and prove their randomness. In addition, a feature-matching algorithm is used to authenticate these information-encoded CQW PUFs. For the safety of the semiconductor chips, a CQW PUF is attached to the surface of the chip to protect against hardware cyber-attacks. Eventually, fabricated labels are examined against high temperatures and moisture environments. The fabricated CQW label is durable for a period of 150 days it is tested, demonstrating ultra-high stability of the label. High stability and durability, cost-effectiveness, and high encoding capacity make these proposed nanocrystal labels extremely attractive for large-scale commercialization.

A stretchable silver nanosheet nanomembrane (AgNS NM) is developed through partially-overlapped face-to-face contacts assembly of AgNSs, enhancing isotropic inter-sheet connectivity and maximizes material performance. Simulation and experiments reveal an AgNS-geometry-dependent design principle of the NM. The ultrathin AgNS NM features high conductivity and omnidirectional stretchability, enabling the fabrication of an impedance-tomography-based electronic skin device for spatially-resolved, real-time tactile mapping.

Abstract

Achieving isotropic electrical and mechanical properties is essential for skin-integrated electronics to operate reliably under complex, multidirectional skin deformations. However, nanomaterial-based composites in skin electronics often rely on anisotropic filler configurations to meet demanding requirements for high-quality bio-interfacing materials, such as ultrathin thickness, high conductivity, and stretchability. While directional alignment of high-aspect-ratio nanofillers facilitates dense percolation, it compromises isotropic material uniformity. To overcome the trade-off between high performance and omnidirectional material properties in the nanocomposites, a controlled assembly strategy is proposed for silver nanosheets (AgNSs) that forms face-to-face contacts with partial overlaps, enhancing inter-sheet contact area and reducing contact resistance. Implementing this assembly configuration in an ultrathin elastomeric membrane yields a silver nanosheet nanomembrane (AgNS NM) with both isotropic material properties and high performance, featuring a high conductivity of ≈115 000 S cm⁻¹, a stretchability of ≈50%, and a total thickness of ≈235 nm. Coarse-grained molecular dynamics simulations (CGMD) reveal that the degree of overlap correlates with nanosheet geometry, providing design insights for controlling interfacial contact configurations in nanomaterials. Finally, the potential of the AgNS NM for bio-interfacing applications is demonstrated through an electrical impedance tomography-based tactile electronic skin, enabling reliable multi-point pressure mapping and real-time tracking.

This study proposes a bilayer strain sensor fabrication strategy that enables large-scale patterning via laser direct writing on 3D stretchable substrates. The resulting sensors exhibit tunable performance through laser-induced microstructural engineering. By constructing a sensor array on a hemispherical surface, the authors further demonstrate the potential of this approach for curved-surface strain mapping and deformation reconstruction.

Abstract

Stretchable strain sensors are capable of accurately mapping deformation across object surfaces, serving as critical components in structural health monitoring and failure mitigation across diverse systems. However, the inherent geometric complexity and non-uniformity of real-world surfaces pose significant challenges to conformal sensor integration. Moreover, conventional strain sensors face inherent trade-offs among sensitivity, detection range, and tunability, limiting their adaptability in practical applications. This study introduces a laser direct writing strategy that combines material and process innovations to enable scalable fabrication of strain sensors on preformed stretchable curved surfaces. Precise control of laser-induced microstructures allows programmable tuning of electromechanical properties, enabling selective behaviors such as high linearity, strain insensitivity, or pronounced resistance changes at small strains. The resulting devices exhibit a high gauge factor of up to 10⁶, a strain detection range exceeding 100%, a minimum detectable strain of 0.1%, and excellent linearity (correlation coefficient > 0.98) within defined operational ranges. As a proof of concept, a sensor array is implemented for strain mapping and deformation reconstruction on a hemispherical 3D stretchable substrate, demonstrating the capability of this approach for high-resolution strain monitoring on complex, non-planar geometries.

MicroSphere Autolithography (µSAL) enables scalable fabrication of patchy particles with customizable surface motifs. Focusing light through dielectric microspheres creates well defined, tunable patches via a conformal poly(dopamine) photoresist. Nearly arbitrary surface patterns can be achieved, with the resolution set by the index contrast between the sphere and the surrounding medium.

Abstract

Patchy particles, i.e., colloidal particles whose surface properties have been modified in predetermined patterns, can serve as building blocks for efficient self-assembly of well-defined, ordered structures. This paper introduces MicroSphere AutoLithography (µSAL), a scalable lithographic method for production of patchy particles with arbitrary patch motifs. This technique leverages dielectric microspheres as both a lithographic substrate and the illuminating optic, using the fact that when a plane wave of light is refracted through a sphere, it produces a circular patch of high-intensity illumination on the back hemisphere. Exposing a collection of microspheres to multiple plane waves, every sphere simultaneously projects identical patterns of illuminated patches onto its own surface. Here, µSAL is demonstrated in barium titanate glass (BTG) microspheres that are coated with a thin (≈40 nm) conformal film of poly(dopamine) acting as the photoresist, fixating the optical pattern into a permanent metal structure through light-induced reduction of silver ions from the liquid suspension. Varying the index of refraction of the BTG spheres and the suspension produces a range of patch sizes and geometries in good agreement with theoretical modeling.

A robust MXene gelation strategy induced by aniline and hydrochloric acid is proposed, producing a skeleton-reinforced hydrogel that enables conformal densification via capillary shrinkage with minimal active site loss. The resultant 225 µm-thick electrode delivers record gravimetric (395 F g⁻¹) and high areal (16.1 F cm⁻²) capacitances, offering new insights for constructing robust MXene architectures in practical energy devices.

Abstract

Transition metal carbides/nitrides (MXenes), with intrinsic high density and pseudo-capacitance, along with the capability for liquid-phase assembly mediated by highly tunable colloidal chemistries, are promising candidates for developing thick electrodes toward high-energy devices. However, the manufacture of high-performance thick MXene electrodes faces fundamental challenges, including nanosheet restacking, 3D structural collapse, and surface oxidation. Here, a robust MXene gelation strategy induced by aniline (ANI) and hydrochloric acid is proposed, producing a skeleton-reinforced hydrogel that enables conformal densification via capillary shrinkage with minimal active site loss. During gelation, ANI absorbs onto MXene surfaces and polymerizes, simultaneously reinforcing the 3D network through covalent bonding while forming temporary hydrophobic layers to protect active sites. Subsequent thermal treatment effectively removes the surface-bound ANI and its oligomers, restoring the active sites for capacitive energy storage. At a thickness of 225 µm, the resulting electrode achieves a record gravimetric capacitance (395 F g⁻¹) among reported MXene electrodes over 40 µm, even surpassing that of a 7 µm MXene film, and delivers a high areal capacitance of 16.1 F cm⁻². This work provides a new insight for assembling robust MXene architectures toward practical MXene-based devices.

This study presents a self-powered, flexible, multi-sensory artificial synaptic device based on a heterojunction structure of reduced graphene oxide and sulfur vacancy-rich ZnIn₂S₄, which ingeniously integrates optical perception and tactile sensing capabilities within a monolithic platform.

Abstract

The burgeoning fields of artificial intelligence and bio-inspired robotics necessitate the development of advanced artificial synapses that transcend the limitations of conventional unitary and rigid systems. This study presents a self-powered, flexible, multi-sensory artificial synaptic device based on a heterojunction structure of reduced graphene oxide (rGO) and sulfur vacancy-rich ZnIn₂S₄ (Sv-ZIS), which ingeniously integrates optical perception and tactile sensing capabilities within a single platform. By engineering interfacial voids and defect states, the dynamic photo-carrier trapping and releasing enable the emulation of fundamental synaptic behaviors under optical stimulation, with relaxation time reaching 200 s. Concurrently, the device exhibits a quantifiable mechanical response to strain and external force, originating from the contact potential difference between the Sv-ZIS and rGO layers. Remarkably, all functionalities are achieved at zero bias, ensuring minimal energy consumption. Leveraging the multi-sensory ability, an artificial neural network is constructed for intelligent object recognition based on their optical transmittance and weight characteristics, achieving a high accuracy of 97.67%. This work establishes a novel paradigm for the development of energy-efficient, flexible, and multi-modal neuromorphic devices, paving the way for advanced applications in soft robotics, wearable electronics, and human-machine interfaces.

Scalable CVD-grown lateral heterostructures stacked with controllable twist angles form quadruple moiré pockets, enabling programmable tuning of phonon, exciton, and SHG. Material-selective lattice relaxation modulates strain, interlayer coupling, Davydov splitting, valley polarization, and anomalous SHG enhancement. Such multi-moiré networks establish a scalable pathway for multifunctional moiré engineering, opening opportunities in opto-straintronics, nonlinear optoelectronics, and quantum photonics.

Abstract

Moiré-engineering in 2D transition-metal dichalcogenides offers access to correlated quantum phenomena. However, simultaneous control over twist-angle (θ) and material combinations to tune phonons, excitons, and their collective interactions remains limited. This study presents scalable, quadruple moiré-pockets formed by vertically stacking chemical vapor deposition-grown monolayer MoS₂–WS₂ and MoSe₂–WSe₂ lateral heterostructures with controlled θ (0⁰–60°), within a single flatland. Moiré non-rigidity induces lattice-relaxation via rotational reconstruction (θ<8°) and volumetric dilation (θ>8°), resulting in strain-mediated phonon frequency-softening and linewidth-broadening, respectively. Strain localizes selectively in mechanically softer crystal for θ<8°, while an epitaxial-pseudomorphic pattern dominates for θ>8°. Degree of phonon-reconfiguration and angle-resolved photoemission spectroscopy uncover the role of interfacial orbitals in modulating interlayer coupling. At aligned angles (θ–0° and 60°), specifically, MoS₂ exhibits Davydov splitting and reduced valley polarization, reflecting symmetry breaking and chiral phonon effects. At θ–3°, WS₂/WSe₂ shows up to 480% enhancement in second-harmonic generation (SHG), while WS₂/MoSe₂ records the lowest due to variations in interlayer-coherence and band-offset-driven phase delay. Notably, at θ–60°, only WS₂/MoSe₂ exhibits an anomalous 300% SHG enhancement, attributed to large phase delay and reconstruction-induced strain. Electronic bandstructure calculations support these observations. These findings offer programmable multi-moiré platforms for opto-straintronics, sensing, and on-chip quantum photonics applications.

A 3D electrochemo-mechanical modeling and simulation framework based on high-resolution FIB-SEM tomography is developed to unravel the interplay between charge conditions and mechanical degradation in a high-SiO_x-content electrode. It enables direct visualization and quantification of microstructure evolution and stress localization, highlighting the critical role of charge condition in balancing capacity and structural integrity in silicon-rich anodes.

Abstract

Silicon is a promising anode material due to its high theoretical capacity, but its extreme volume change (>300%) during cycling leads to contact loss, electrode delamination, and crack propagation, ultimately compromising mechanical integrity. While operando imaging captures morphological evolution, it remains insufficient to resolve the coupled electrochemical, mechanical, and microstructural dynamics that govern degradation. Here, a microstructure-resolved digital twin model of SiO_x/graphite composite electrodes is presented to diagnose electrochemo-mechanical behavior. A 3D structure reconstructed from high-resolution FIB-SEM tomography is integrated into a coupled simulation framework that captures Li⁺ diffusion, interfacial electrochemical reactions, and concentration-dependent mechanical strain. Simulations reveal that volumetric expansion distorts internal conduction pathways—enhancing electronic conduction via broadened solid–solid interfaces while impeding ion transport through increased tortuosity. Moreover, charge-rate-dependent analysis shows that the charging rate governs the balance between the state of charge (SoC) and local stress. Increasing the rate from 0.5C to 4C reduces stress by limiting the SoC level, thereby mitigating mechanical degradation and enhancing cycling stability. This digital twin framework enables quantitative diagnostics of stress-driven failure and offers design guidelines for the development of mechanically robust, high-performance silicon-based anodes.

Nature Communications, Published online: 23 November 2025; doi:10.1038/s41467-025-66697-6

A directional electrodynamic decomposition method using a LaNiO3 sacrificial layer is proposed for freestanding films. It boosts oxide film release rate to 600 mm2 /min via electric field-enhanced adsorption energy and electron transfer.

Digital Twins of random porous tape-cast solid-state battery architectures across µm to mm feature sizes from FIB-SEM to X-Ray µCT, respectively.

Abstract

Solid-state lithium batteries (SSBs) have the potential to overcome conventional Li-ion batteries in performance and safety. SSBs are typically manufactured through casting techniques to produce 3D electrode architectures that increase interfacial surface area and improve cell kinetics. However, these microstructures are often very complex and the exact influence of their geometries on their electrochemical performance can best be determined through numerical simulations. The main obstacle to numerically studying porous electrodes is the difficulty of obtaining high-quality geometric data. In place of characterizing real samples, a costly and time intensive process, researchers seeking to intelligently design SSB microstructures may choose to run their calculations on stochastically generated digital twins (DTs). Naturally, the quality of these results rests on the fidelity of the DTs with real materials. This work develops a stochastic model and validation suite to produce the highest fidelity computer-generated microstructures. The DTs generated with this model are compared to several solid-state electrolyte samples characterized with Focused-Ion Beam, Scanning Electron Microscopy (FIB-SEM), and X-Ray Micro-Computed Tomography (µCT). An analysis of these microstructures across a series of tape-casting parameters (including particle size, ceramic volume fraction, and particle roughness) reveals the existence of several geometric relationships that are characteristic of the tape-casting process.

Herein, a systematic digital twin workflow tailored for generating high-fidelity virtual representations of anisotropic composite microstructures and giga-voxel meso-structural models is presented, leveraging a harmonious integration of top–down image-based modeling and bottom–up data-driven voxel generation.

Abstract

Giga-voxel digital models offer abundant geometric detail; however, no mainstream method currently exists to efficiently distribute individual voxels across massive image volumes, and designing complex anisotropic composite materials remains infeasible due to the absence of promising methods. Herein, we propose a systematic digital twin workflow tailored for generating high-fidelity virtual representations of anisotropic composite microstructures and giga-voxel meso-structural models, leveraging a harmonious integration of top-down image-based modeling and bottom-up data-driven generation. Our study demonstrates the efficacy of micro-digital representations as foundational building blocks within a continuum of digital assembly processes tailored for mesostructural models. Utilizing 3D image data, specifically X-ray tomography, our data-driven modeling meticulously characterizes the geometric attributes of the experimentally observed objects, thereby facilitating the creation of digital unit twins, each endowed with distinct identities assigned through a random seed generation. The closed-loop system provides feedback mechanism between data and model to ensure the 3D quality of the generated models. For hierarchical organization at the giga-voxel level, the digital unit twins are methodically expanded into cohesive 3D architectures based on assembly relationship at length scales of more than four orders of magnitude. Remarkably, this hierarchical model provides intricate insight into micro-to-macro geometrics while preserving the intrinsic microstructure.