Cuprate high-temperature superconductors are prototypical strongly correlated systems. Clarifying the complex orbital interactions within these systems has been one of the greatest challenges in condensed matter physics. Because the coupling between orbital bands cannot be disentangled, conventional density functional theory (DFT) has struggled to provide a microscopic picture of individual orbital bands. To investigate the contribution of core-orbital coupling (COC) to strong correlation—which is neglected by the adiabatic approximation in DFT—this paper extracts the ubiquitous pyramidal layer (based on the CuO2 plane oxygens and neighboring rare-earth ions) and uses inert gas atom substitution (He replacing O 2s, Ne replacing RE p) to construct a minimal model that isolates COC. Using standard DFT to calculate supercells of varying sizes (n×n, n = 1–15) to simulate doping, we reveal that COC can generate stable spin-singlet pairs (Px, Py, Pz) formed through s–p hybridization, with a binding energy of approximately 0.2 eV. Within the CuO2 plane, the Px/Py states form either linear long-range chains (n = 2) or vortex-like patterns (n ≥ 3), depending on the degree of translational symmetry breaking. The COC network is robust against up to 25% vacancies, may corresponding to the quantum critical point (p = 0.25) in the cuprate doping phase diagram. Strong spin-orbit coupling leads to spin-momentum locking, stabilizing the spin-singlet configuration. These results show similarities with the resonating valence bond (RVB) theory, Kosterlitz-Thouless (KT) theory, and valence bond solid (VBS) picture, and also open up a practical route for using standard DFT tools to study orbital interactions in strongly correlated systems.

Intrinsically stretchable electronics can be used to make wearable devices that collect large amounts of multimodal sensory data. This has led to a demand for enhanced near-sensor computing capabilities that can process such data. One potential solution is neuromorphic edge computing implemented using stretchable organic electrochemical transistors, but the lack of a scalable fabrication method for these transistors has limited the size and complexity of the systems. Here we report the fabrication of large-scale, intrinsically stretchable organic electrochemical transistor arrays with a density up to 10,000 transistors per square centimetre. The transistors exhibit good synaptic performance, including linear, precise conductance programming and good retention. The performance uniformity of the array enables the hardware implementation of a stretchable artificial neural network for processing health data, including heart attack risk assessment and kernel convolution for locating propagation wavefronts. We also explore the potential of implementing reinforcement learning algorithms on the neuromorphic circuits for use in soft robotics.

Effective nose-to-brain delivery is limited by the mucus-mucosal barrier, which severely hinders drug transport after nasal administration. Although nanostructured lipid carriers (NLCs) offer promising solutions by improving their retention duration in the mucosal layer or promoting mucosal permeation and intracellular uptake, the nanoparticle-mucus interactions, epithelial uptake, and mechanisms supporting transcellular transport in the nasal cavity remain poorly learned. This study was developed to assess the interactions between mucin and NLCs, the absorption and transport capabilities of Cannabidiol (CBD) nanostructured lipid carriers (CBD-NLCs) in rat nasal mucosal epithelial (RNME) cells. Molecular docking (MD) was used to evaluate the interactions between the human mucin protein MUC5AC (hMUC5AC) and some main components of CBD-NLCs. In vitro experiments were conducted to assess the mucus penetration of CBD-NLCs. Cellular uptake, localization, and transport mechanisms of coumarin-6-labeled NLCs (C6-NLCs) in RNME cells were examined by confocal laser scanning microscopy, endocytosis inhibition assays, and transcellular transport assays. Our results show that there were no interactions between the main components of NLCs and hMUC5AC according to MD simulations. In vitro experiments, NLCs promoted mucus penetration and had little interaction with mucin. Cellular studies confirmed cytoplasmic (non-nuclear) localization of C6-NLCs in RNME cells, and transport assays showed that the clathrin-mediated endocytic route was a predominant pathway for their internalization. When across the RNME cell monolayer, NLCs had a great benefit on CBD transportation across monolayers of RNME cells. Our results clarify the mechanism by which NLCs enhance mucosal penetration and facilitate cellular trafficking in RNME cells. Their little interaction with mucin, combined with efficient cellular uptake and transcellular transport, supports the potential of NLCs as a promising nanoplatform for nose-to-brain drug delivery.

Despite their structural precision, supramolecular coordination complexes (SCCs) often suffer from weak near-infrared (NIR) absorption and inadequate tumor targeting, limiting their theranostic efficacy. To overcome these barriers, we engineered two biomimetic metallacycle nanoparticles, CMa and CMb, via coordination-driven self-assembly of an organoplatinum acceptor with aza-BODIPY ligands bearing varying numbers of electron-donating N,N-dimethylamino groups. The optimized CMb nanoparticles, incorporating four strong electron-donating groups, exhibited a pronounced redshift in absorption into the NIR-I region (832 nm) and an exceptional photothermal conversion efficiency of 55.72%. Subsequent encapsulation with Pluronic F127 and coating with a HeLa cell membrane conferred good colloidal stability, biocompatibility, and specific homologous targeting. In vitro studies confirmed that CMb NPs efficiently generated heat and singlet oxygen under 808 nm laser irradiation, leading to potent synergistic cell killing through combined photothermal therapy (PTT), photodynamic therapy (PDT), and platinum-based chemotherapy. More importantly, in vivo experiments demonstrated that CMb NPs-mediated photo-chemo therapy effectively ablated primary tumors, induced immunogenic cell death, and triggered a systemic antitumor immune response, as evidenced by enhanced dendritic cell maturation and cytotoxic T-cell infiltration. When synergized with an anti-PD-L1 antibody, this combination therapy achieved complete tumor eradication and robust inhibition of distant tumors, effectively reversing the immunosuppressive tumor microenvironment. This work presents a versatile biomimetic nano-platform that integrates supramolecular design with cell membrane camouflage for advanced synergistic cancer therapy.

Stretchable organic light-emitting diodes (OLEDs) hold great promise for wearable displays and optical biointerfaces, yet progress is hindered by the intrinsic trade-off between mechanical stretchability and high emission efficiency. Here, we present a broadly applicable strategy to enhance both stretchability and light-emitting performance in thermally activated delayed fluorescence (TADF) polymers through the incorporation of optoelectronically inert small-molecule plasticizers. Using dioctyl phthalate (DOP) as a model additive, we show that plasticizers function as molecular spacers, expanding free volume to suppress triplet exciton quenching while facilitating stress-dissipative chain mobility. The resulting composites achieve approaching-unity photoluminescence quantum yield (PLQY), stretchability beyond 110% strain, and improved electroluminescent efficiency, with external quantum efficiency (EQE), reaching 12.6% in rigid devices and 3.05% in fully stretchable OLEDs. This strategy is effective across a range of TADF polymers, demonstrating plasticizer engineering as a simple, scalable design principle for intrinsically stretchable optoelectronic materials.

Photodynamic therapy (PDT) utilizing organic photosensitizers like indocyanine green (ICG) faces several intrinsic limitations. These challenges include a propensity to aggregate, insufficient stability, and low intersystem crossing (ISC) efficiency that yields inadequate reactive oxygen species (ROS). Furthermore, the tumor microenvironment imposes additional restrictions on its efficacy. To address these challenges, we implemented a cetyltrimethylammonium bromide (CTAB)-templated metal coordination approach and rationally designed cerium(IV)-coordinated, self-assembled ICG nanoparticles (CINPs). Cooperative coordination between Ce4+ and ICG, combined with hydrophobic interactions, significantly enhances ICG stability. It also optimizes the energy gap between the lowest singlet excited state (S1) and the lowest triplet state (T1), thereby promoting ISC and equipping the system with robust ROS-generating capacity. Moreover, Ce4+ exhibits dual catalytic activity, on the one hand catalyzing the decomposition of endogenous hydrogen peroxide (H2O2) to generate oxygen and alleviate tumor hypoxia, and on the other hand oxidizing and depleting intracellular glutathione (GSH) to weaken antioxidant defenses. In a hepatocellular carcinoma model, CINPs harnessed efficient ROS production to induce mitochondrial dysfunction, lipid peroxidation (LPO), and DNA strand breaks, which collectively activated multiple cell death pathways and significantly suppressed tumor growth. Unlike nanoplatforms that require elaborate designs, complex compositions, and fine chemical synthesis, the nanoparticles developed here are assembled from a small set of reliable, clinically established materials and can be rapidly formed through a simple process in less than 30 min. Despite the streamlined preparation, CINPs effectively remodel the tumor microenvironment, induce vigorous ROS generation during treatment, and achieve potent antitumor activity via oxidative stress-related mechanisms across multiple pathways, highlighting strong translational potential and broad prospects for clinical application.

Immune checkpoint blockade (ICB) has shown clinical promise in cancer immunotherapy, but colorectal cancer (CRC) remains difficult to treat due to a complex immunosuppressive tumor microenvironment (TME). This TME features dysregulated CD47/SIRPα and PD-1/PD-L1 pathways, T cell exhaustion, and infiltration of immunosuppressive cells. Therefore, novel strategies to reshape TME are critically needed. Here we developed bioinspired nanodevices (SPCM@DMgTi-ADM) for CRC immunotherapy. These nanodevices consist of dendritic titanium-magnesium nanoparticles (DMgTi) that serve as both a sonosensitizer and a carrier for aldometanib (ADM). The nanoparticles are coated with genetically engineered membranes displaying SIRPα and PD-1 decoy receptors (SPCM). This design achieves cascade reinforcement between AMP-activated protein kinase (AMPK) activation and cGAS-STING activation. The SPCM coating simultaneously blocks SIRPα/CD47 and PD-1/PD-L1 axes, preliminarily reshaping the TME. Released Mg2+ induces conformational changes in LFA-1 on CD8+ T cells, promoting their tumor infiltration and cytotoxic function. ADM together with sonodynamic therapy (SDT) induces AMPK activation, which drives autophagy-dependent ferroptosis. This synergistic process triggers strong Immunogenic cell death (ICD). The released dsDNA potently activates the cGAS-STING, which in turn inhibits GPX4 and sustains ferroptotic stress. This creates a self-amplifying loop: ferroptosis promotes dsDNA release, which activates cGAS-STING; STING then suppresses GPX4, worsens ferroptosis and further boosting anti-tumor immunity. Both in vitro and in vivo studies confirm that our nanodevice effectively evaluates tumors and activates systemic anti-tumor immunity, offering a clinically translatable strategy for precision CRC therapy.

Achieving efficient and homogeneous mixing in highly concentrated solid-liquid systems remains a major challenge, since existing simulation methods cannot accurately capture the dynamic continuous mixing, thus failing in effective prediction and optimization. In this work, an Euler-Euler model coupled with the dynamic mesh technology and kinetic theory of granular flow was developed to simulate the transient continuous mixing of dense solid-liquid suspension. The liquid shear stress, dispersive mixing, distributive mixing and dispersivedistributive coupling mixing were resorted to elucidate the dense solid-liquid mixing mechanism. On this basis, novel screw elements were designed and associated mixing behaviors were comprehensively investigated. The results demonstrate that the novel simulation method provides high accuracy in predicting the mixing performance. Compared with the conventional K45/5/32 kneading element, the axial kneading slot element had increased the non-uniformity of tensile strength and shear strength by 16.5% and 64.7%, respectively, leading to a 48.7% improvement in the mixing uniformity of the dense solid-liquid system. Moreover, the axial kneading slot element increases the probability density of liquid shear stress in the range of 610-1300 Pa from 0.41 in K45/5/32 element to 0.56, increasing moderate shear stress zones by 15% while maintaining the high liquid shear stress region (>4200 Pa) to disrupt agglomerated particles.

Spinal cord injury (SCI) is a devastating neuropathological condition. Currently, there is an urgent need for highly effective therapies for SCI treatment. Here we developed a multifunctional hydrogel therapy (LPPXN), by rationally integrating pharmacologically active nanomicelles into hydrogels composed of noncovalently cross-linked nanoparticles that are self-assembled by a functionalized amphiphilic triblock polymer. LPPXN exhibits temperature-responsive gelation, high strength, favorable bioadhesive properties, and excellent shear-thinning and self-healing capabilities under pathological conditions. Following local injection, LPPXN can be sustained for over one month. Therapeutically, LPPXN significantly improved the structural integrity of injured spinal cords and promoted function recovery in a mouse model of SCI. Furthermore, LPPXN demonstrated beneficial therapeutic effects in mice with SCI combined with ischemia-reperfusion injury, a model closely replicating real-world scenarios. Mechanistically, LPPXN treatment promoted neuroprotective astrocyte polarization and structured network assembly at the SCI lesion site, while reconstructing a regenerative niche to enhance neural preservation and protection. This multifaceted efficacy was primarily mediated through suppressing oxidative/inflammatory cascades, inducing anti-inflammatory polarization of macrophages and microglia, and modulating the CCL2/CCL5-JAK-STAT signaling pathway. Notably, LPPXN showed excellent tissue biocompatibility in the spinal cord. Accordingly, LPPXN warrants further development as a promising therapeutic option for SCI and other nerve injury-associated diseases.

Stretchable organic light-emitting diodes (OLEDs) are transforming human-machine interfaces and wearable technologies; still, their performance is considerably inferior to commercial, non-stretchable OLEDs, mainly limited by inefficient electron injection. We address this by redesigning both the electron transport layer and the cathode. For the former, we design a copolymer structure with high stretchability and ideal energy levels, achieving performance comparable with standard small-molecule electron transport layers. For the latter, we leverage the liquid metals embrittlement effect to confer stretchability to aluminium thin films, without compromising their electrical and optical characteristics. Combining these designs, we demonstrate fully stretchable OLEDs with a very high external quantum efficiency of 8% and a very low turn-on voltage of 3.5 V, which is on par with the reference rigid OLEDs utilizing the same emitter. This work tackles a crucial bottleneck in stretchable OLED development, bridging the performance gap between stretchable OLEDs and standard rigid OLEDs at the device level, paving the way for high-performance, skin-like displays.

Memristor-based analog in-memory learning (AIML) has emerged as a promising approach to improve energy efficiency in deep neural network training. However, non-idealities in memristive devices, such as nonlinearity, asymmetry, and cycle-to-cycle (C2C) and device-to-device (D2D) variations, pose significant challenges. These issues lead to increased energy consumption, reduced write precision, and compromised in-situ learning performance. To address these problems, we propose a mixed-precision training strategy that combines gradient accumulation with single pulse blind write method. We analyze the failure mechanisms of in-situ learning without these techniques and systematically investigate how various non-idealities affect AIML performance. We demonstrate that, by using our GA-Single Pulse strategy, high accuracy (95.36%) can be achieved even under significant non-idealities, including device conductance states being limited to 10 pulses for potentiation as well as 5 pulses for depression, the asymmetry of conductance state constrained to a factor of 2, the nonlinearity in long-term potentiation/ long-term depression curve reaching up to 5, C2C variation as high as 50%, and D2D variation extending up to 40% for learning handwritten digits in MNIST handwritten digit dataset, outperforms all previous reports. The results suggest that the idealities of memristive devices may not be as critical as previously assumed for AIML's practical deployment.

The presence of first-pass metabolism and the obstacle of the blood-brain barrier may reduce the effectiveness of oral antiepileptic medications. Nasal drug delivery has been considered a promising selective route to the brain for drugs with low aqueous solubility in the treatment of CNS disease. The purpose of our study was to improve the bioavailability as well as brain targetability of cannabidiol (CBD) by encapsulating it in nanostructured lipid carriers (NLCs) and delivering the formulation via the nasal route. CBD-NLCs were effectively prepared with the appropriate particle sizes and polydispersity index for the nasal route (77.71 nm± 0.79 and 0.23 ± 0.00, respectively). These particles demonstrated a high entrapment efficiency and drug loading of 99.24% ± 0.07 and 8.73% ± 0.56 w/w, respectively. According to the FTIR, XRD, and DSC data, CBD was either in the amorphous state or distributed molecularly in the lipid matrix. The absorption of CBD-NLCs in the nasal cavity was significantly superior to pure CBD (the rate constants were 9.02 ± 1.64 and 2.10 ± 0.25 μg/min, respectively). Compared to the intravenous administration of CBD, CBD-NLCs showed a drug targeting efficiency of 277.82% after nasal administration, indicating a more efficient brain targeting. In a rat model where seizure activity was induced by PTZ, intranasal administration of CBD-NLCs significantly prolonged seizure latency and decreased the Racine score. All of the results suggest CBD-NLCs administered intranasally might be a promising alternative to traditional epilepsy treatments.

One of the greatest obstacles to achieving implantable electronics with long-term functionality and minimized inflammatory reactions is the immune-mediated foreign-body response (FBR). Recently, semiconducting polymers with mixed electron-ion conductivity have been demonstrated as promising candidates to achieve direct electrical interfacing on bio-tissues. However, there is limited understanding of their immune compatibility in vivo, and strategies for minimizing the FBR through molecular design remain underexplored. Here we introduce a set of molecular design strategies for enhancing the immune compatibility of semiconducting polymers. Specifically, we show that selenophene, when incorporated in the backbone, can mitigate the FBR by suppressing macrophage activation. In addition, side-chain functionalization with immunomodulatory groups decreases the FBR further by downregulating the expression of inflammatory biomarkers. Together, our synthesized polymers achieve suppression of the FBR by as much as 68% (as indicated by the collagen density). In the meantime, these immune-compatible designs still provide a high charge-carrier mobility of around 1 cm2 V-1 s-1. We anticipate that such immune-compatible design principles can be translated to a variety of conjugated polymers to suppress the FBR for implantable applications.

The high redox levels within tumors position chemodynamic therapy (CDT) as a promising therapeutic approach. However, the CDT efficiency of dihydroartemisinin (DHA) is limited by rapid clearance from bloodstream, along with inadequate endogenous ferrous ions within tumor microenvironment and heightened anti-oxidative defense inside tumor cells. To overcome these limitations, we developed an innovative virus-like hollow mesoporous manganese nanocage, loaded with DHA and subsequently cloaked with red cell membrane, designed to trigger a tumor-microenvironment-responsive free radical generation, synergized with glutathione (GSH) exhaustion for enhanced CDT efficacy. Upon accumulation in tumor tissues via the enhanced penetration and retention (EPR) effect, the high concentration of GSH in cancer cells initiates the degradation of the nanocages. This process specifically and efficiently released both Mn2+ and DHA while simultaneously depleting GSH. The released Mn2+ further catalyzed the conversion of DHA to generate large amounts of highly toxic carbon-center (•C) radicals accompanied by the generation of Mn4+. The •C radical generation led to severe mitochondria dysfunction and DNA damage, potentially causing cancer cell death. The concurrently generated Mn4+ continued to depelete intracellular GSH and induce lipid peroxidation, thereby weakening cancer cells' anti-oxidative defenses and amplifing oxidative stress. The viability of 4T1 cells treated with DHA@vhmMN@RM was significantly lower (about 30 %) than other groups. This work presents a novel nanosystem that specifically enhances the therapeutic effect of CDT by leveraging a tumor microenvironment-responsive free radical generation, coupled with GSH exhaustion, offering a new avenue for targeted drug delivery and synergistic cancer therapy.

The integration of multimodal therapies into a single nanoplatform promises significant advances in precision oncology, yet structural instability, premature drug leakage, and insufficient immune activation remain key challenges. Herein, a supramolecular metallacycle-based nanoplatform ( M2S-AD NPs) is engineered through coordination-driven self-assembly and β -cyclodextrin-mediated host-guest encapsulation. The metallacycle exhibits aggregation-induced emission (AIE) characteristics to amplify NIR-II fluorescence (1084 nm) for real-time tumor imaging and achieves a record-high photothermal conversion efficiency (42.7%) under 808 nm irradiation, enabling deep-tissue photothermal ablation. Leveraging tumor microenvironment (TME)-responsive disulfide bonds and pH-labile Pt-N coordination, M2S-AD NPs selectively release chemotherapeutic agents in glutathione-rich tumors while maintaining stability under physiological conditions. Synergistically, localized hyperthermia induces immunogenic cell death, releasing tumor antigens to prime dendritic cells, while co-delivered anti-PD-L1 antibodies reverse immunosuppression, amplifying cytotoxic T lymphocyte infiltration. In 4T1 breast tumor models, this combinatorial strategy eradicates primary tumors and suppresses distant metastases with >80% tumor growth inhibition without inducing systemic toxicity. By unifying AIE-enhanced imaging, TME-triggered drug release, and immunomodulation, this work establishes a supramolecular engineering paradigm for next-generation cancer theranostics, bridging material innovation with immunooncology to combat metastatic malignancies.

Stretchable light-emitting devices are poised to play a central role in advancing human–technology integration, enabling applications such as on-skin displays, optical sensing, and implantable phototherapy. Among them, stretchable organic light-emitting diodes (OLEDs) are particularly attractive due to their high efficiency and potential biocompatibility. The recent realization of thermally activated delayed fluorescence (TADF) in stretchable emitters, enabling triplet exciton harvesting, has increased external quantum efficiency (EQE) to 10%. However, triplet–triplet annihilation (TTA) remains a key barrier to further improvement toward commercial-grade performance. Here, we introduce a stretchable host–guest emitter design that overcomes this quenching mechanism by uniformly dispersing TADF small-molecule guests within a newly designed stretchable host polymer. This architecture enables efficient exciton transfer while suppressing TTA, yielding an external quantum efficiency (EQE) of 20.3%—doubling the performance of prior state-of-the-art. Notably, the guest molecules also act as plasticizers, enhancing stretchability beyond 150%. With generalizability to different TADF emitters, this work establishes a foundational strategy for mitigating TTA in stretchable emissive layers, advancing soft optoelectronics toward commercial viability with mechanical durability and practicality.

The ionic transportation process is the main driving mechanism in novel microelectronic devices, such as resistive random access memories (RRAMs) and ionic-gated transistors. Understanding the ionic migration under multiple coupled physical fields provides valuable information and guidelines for device design and optimization. The oxygen vacancies transportation processes include drift under an electric field, Fick diffusion under a concentration gradient, and Soret diffusion or thermophoresis under a temperature gradient, where the last effect is often neglected. The working mechanism of RRAM devices depends on the conductive filament formation and rupture inside the resistive switching (RS) layer. The reliability and uniformity issues of RRAM devices come from the complex ionic transportation in amorphous materials under multiple coupled physical fields. In this work, we propose a technology computer-aided design (TCAD) model for RRAM devices that includes the full ionic transport mechanisms as well as takes the ionic generation and recombination processes into consideration. Based on this model, we investigated the effect of different parameters on device behavior.

Flexible wearable sensors that can intimately adhere to the human body for real-time monitoring of human activities and physiological signals have attracted great attention owing to their potential in personalized healthcare and human-machine interfaces. Gelatin-based biogels are promising materials in wearable sensors due to their good biocompatibility, biodegradability, and sustainability. However, conventional gelatin-based biogels are usually weak and brittle (tensile strength < 10 kPa and stretchability < 50%), and thus cannot be applied in flexible wearable devices. Therefore, further efforts are needed to engineer tough gelatin-based biogels that meet the demands of flexible wearable sensors. In this perspective, we summarize recent progress in designing tough gelatin-based biogels and their wide applications in wearable sensing devices, while highlighting potential future directions in this field.