We demonstrate high-precision hyperdimensional computing using an in-memory computing (IMC) architecture based on ferroelectric Hf0.5Zr0.5O2 (HZO) capacitors. By exploiting the high polarization charge density of CMOS back-end-compatible HZO, we achieved 32 well-separated and linearly programmable intermediate states in 10-nm-thick capacitors making them suitable as capacitive IMC elements. In recent times, capacitive IMC emerged as a promising energy- and latency-efficient route for data-intensive computing tasks. However, compute-in-memory elements require non-volatile, reproducible, and multi-bit operation. In this work, we show that through optimized device fabrication without vacuum break between oxide and nitride depositions and tailored thermal engineering, the HZO capacitors can exhibit high remanent polarization (2Pr = 75 µC/cm²). Structural studies highlight a high orthorhombic phase fraction and clean HZO/TiN interface. The intermediate polarization states exhibit controllable, linear, and reproducible capacitance modulation via voltage-driven polarization switching, enabling reliable multi-bit device operation and non-destructive readout. Leveraging these 5-bit ferroelectric capacitors, it is possible to store 15-bit numerical values using only three capacitors to implement high-precision capacitive IMC in a hyperdimensional computing task, achieving improved inference accuracy of 92.3% and 2.3x reduced areal footprint compared to binary encoding. These results highlight the importance of advanced materials engineering to achieve high bit-precision and state linearity in ferroelectric capacitors for scalable capacitive in-memory computing.
Precise regulation of hydraulic resistance is essential for controlling flow distribution, particle transport, and chemical synthesis in microfluidics. Despite frequent comparisons between microfluidic networks and electronic circuits, components analogous to electronic rheostats for high-precision continuous resistance tuning remain absent, with flow regulation typically relying on microvalves with discrete open/closed states and external flow-control hardware. Inspired by geometry-modulated flow resistance in heart valves, we introduce magnetic microfluidic rheostats composed of soft magnetic cantilever arrays, whose field-controlled bending modulates the hydraulic diameter to achieve stable and continuous resistance tuning with a minimum measured increment of ~1.7%. Using external permanent magnets instead of complex actuation systems, the rheostat enables low-cost, energy-efficient operation while maintaining high precision. Incorporating the rheostat into a Wheatstone fluidic bridge allows sub-10 µm/s flow balancing and precise microparticle manipulation. Moreover, rheostats with tailored magnetic responses enable one-to-many control under a single magnetic field, allowing multiple channels to be regulated synchronously or asynchronously. This approach simplifies system architecture compared with conventional one-to-one actuation, while maintaining 1-2% composition accuracy across multi-component streams. These results establish microfluidic rheostats as dedicated resistance regulators that expand the functional toolkit of microfluidics for scalable, high-precision manipulation and synthesis.
Abstract Neuronal magnetic signal recording intrinsically provides vector information and tissue transparency, offering a potential route to overcome the spatial resolution limitations of conventional electrophysiological recordings. However, in situ detection of neuronal magnetic signals at the cellular level remains highly challenging due to the limited sensitivity of microscale magnetic sensors and the presence of background noise. Here, we report a high-sensitivity implantable differential magnetrode based on a dual-pinned magnetic tunnel junction (TMR). By implementing a spatially decoupled differential architecture with a long baseline (5 mm), together with a sensitivity dynamically matched interface circuit, the device effectively suppresses environmental common-mode noise and achieves an ultralow detection limit of 68 pT/√Hz at 1 kHz. Benefiting from an optimized bidirectional SiO₂/Si₃N₄ protective interface, the device exhibits good biocompatibility and stability. Using this magnetrode, we detected action potential-related magnetic signals in the CA1 region of the rat hippocampus. Crucially, we experimentally observed polarity reversals in spike waveforms that correlate with neuronal spatial orientation, demonstrating the potential of magnetic recording to distinguish neural current directionality based on vector information. This work provides a novel sensing tool for microscale neuroscience research and offers a new technical pathway for resolving the spatial topology of complex neural circuits.
Flexible and stretchable structural systems require geometries capable of maintaining mechanical stability under repeated deformation while minimizing localized stress concentration. In this study, a beetle elytra-inspired flexible architecture based on the interdigitated suture geometry of the diabolical ironclad beetle was developed and compared with conventional Peano curve-based structures. The proposed geometry incorporated elongated elliptical motifs designed to enhance tensile deformability and promote stress redistribution during loading. Finite element analysis demonstrated that the beetle-inspired structures exhibited more uniform strain propagation and lower local stress concentration than conventional structures, particularly in the horizontal configuration. Uniaxial tensile testing further confirmed enhanced deformation sustainability and delayed structural failure in the beetle-inspired architectures. In addition, comparative simulations of deformable sensor-interconnect systems revealed reduced peak stress accumulation near rigid island regions. Overall, the present study demonstrates that beetle-inspired interdigitated geometries can effectively enhance flexibility and mechanical stability in stretchable systems, suggesting their strong potential for future flexible and deformable devices requiring high durability under repeated mechanical deformation.
Pancreatic ductal adenocarcinoma (PDAC) is a highly aggressive solid tumor, closely associated with its unique tumor microenvironment (TME), which is characterized by a dense desmoplastic stroma. Abundant stromal cells, primarily fibroblasts, constitute the majority of cells in the tumor mass and exhibit pronounced spatial heterogeneity. Importantly, the spatial distribution of tumors and fibroblasts is vital for shaping the TME and critically influencing therapeutic responses. Here, we present a facile microarray chip for generating architecturally defined 3D PDAC heterospheroids. This platform enables us to mimic the dynamic interactions between tumor and stromal cells and to investigate how spatial organization influences stroma heterogeneity, tumor invasion and chemoresistance. The chip incorporates square concave microstructure array allowing controllable and reproducible production of uniform-sized spheroids. By simply altering the cell seeding sequence, we successfully constructed heterospheroids with distinct spatial distributions of cancer cells and fibroblasts. We further demonstrated that these organizational patterns modulate tumor-stroma crosstalk and ultimately regulate tumor invasive behavior. Furthermore, the heterospheroids with defined patterns exhibited distinct drug responses, and the potential for combination therapy evaluation was also verified. Beyond providing a robust platform for engineering heterospheroids with controllable tumor-stroma architectures, this system offers a robust 3D co-cultured model for advancing cancer research and drug screening.
Simultaneous acquisition of quantitative phase and fluorescence information is essential for comprehensive cellular analysis, as these complementary modalities provide structural and biochemical insights, respectively. However, conventional dual-mode imaging systems require fluorescent labeling, complex optical architectures, and multiple acquisition channels, which limit scalability, increase cost, and hinder deployment in low-resource or high-throughput settings. Here, we present an AI-driven dual-mode phase and label-free fluorescence imaging platform using a single-shot Gabor hologram. The proposed framework enables simultaneous reconstruction of quantitative phase images and virtual fluorescence channels from a single low-cost holographic measurement, eliminating the need for fluorescent staining, multi-shot acquisition, or multimodal optical hardware. To achieve this, we introduce a one-sided unsupervised diffusion model that learns a unidirectional mapping from Gabor holograms to dual-mode outputs without requiring paired training data or cycle-consistency constraints. The model integrates contrastive learning-based hologram synthesis with hologram-conditioned denoising diffusion to ensure high structural fidelity and robust cross-modal reconstruction. Ground-truth phase and fluorescence images acquired from a conventional dual-mode optical system are used only during training, while inference relies exclusively on a single-shot Gabor hologram captured with a minimal optical configuration. Experimental validation across multiple cancer cell lines and organelle-specific fluorescence channels demonstrates that the proposed platform accurately recovers cellular morphology and subcellular distributions, achieving an FID of 57.74, SSIM of 0.76, PSNR of 26.89 dB, and LPIPS of 0.12, and further generalizes to unseen conditions including higher magnification, different cell type, low-illumination, and defocused acquisitions. These results support quantitative analysis, cell-type discrimination, and drug-response assessment, establishing a scalable and cost-effective platform that paves the way toward compact, high-content label-free imaging solutions.
Monitoring of organ health and tissue damage through real-time sensing of chemical parameters such as pH and oxygen concentration could provide helpful information in the field of solid organ transplantation. Several sensors have been developed to detect oxygen concentrations and pH levels in tissues separately; however, there is no report to date on a single device in such context that can detect both pH and oxygen alterations simultaneously and in real time. This paper reports the development of a single, bi-functional optical sensor device that can simultaneously and reversibly respond to changes in both pH and oxygen concentration. The proposed optical sensor integrates both pH- and oxygen-sensitive probes, and is optimized to achieve minimal cross-sensitivity during simultaneous measurements. The sensor is excited with light of 405 nm wavelength to detect pH and oxygen changes respectively at emission wavelengths of 520 nm and 600 nm. The sensor detects pH and oxygen alterations with a response time of about 12.5 s and less than 1.5 s, respectively. The sensor exhibits a sensitivity of 75.75% for pH variations and 20.4% per mg/ml for changes in oxygen concentration. The sensor was tested on a human liver declined for transplantation to analyse the initial sensor behaviour on human tissue. Including the sensor in an organ perfusion setup to monitor pH and oxygen levels could potentially provide insight into the microcirculation of an organ before transplantation.
The evolution of wireless communication imposes stringent requirements on RF front-end modules, specifically demanding switches with ultra-thin profiles and wideband coverage up to Ku band. However, conventional packaged MEMS switches often result in excessive device height, creating a bottleneck for compact system integration. To address these challenges, this study presents a TSV-packaged Single Pole Six Throw (SP6T) RF MEMS switch that simultaneously achieves an ultra-thin profile and superior signal integrity. The device features a total height of only 300 μm, achieved through a substrate grinding process. To compensate for the impedance variations induced by the ultra-thin substrate and air cavity, two types of Heterotypic Microstrip (HMS) transmission lines are proposed and optimized for signal matching from DC to Ku band. The switches are fabricated using a combination of surface and bulk micromachining, including gold-alloy electroplating, Au-Sn bonding, and TSV processes. Measurement results verify the design strategy: the switch with a 300-μm height exhibits significantly improved insertion loss (IL) suppression compared to thicker counterparts. The best IL reaches 2.0 dB at 18 GHz, with return loss better than 14 dB and isolation exceeding 18 dB across the DC ~18 GHz band. Furthermore, mechanical and thermal simulations confirm that the substrate thinning does not compromise contact force or heat dissipation. These results demonstrate that the proposed packaging scheme successfully solves the conflict between miniaturization and high-frequency performance, offering a promising solution for next-generation mobile RF systems.
Ultraviolet (UV) plasmonic materials are promising for biosensing and nanophotonics, but practical deployment has been limited by oxidation and corrosion in UV-active metals. Here we introduce chiral silicon nanohelices as a robust and scalable platform for UV plasmonics enabled by interband-driven negative permittivity of silicon below ~300 nm. Wafer-scale arrays of three-dimensional chiral Si nanohelices are fabricated by glancing angle deposition and exhibit pronounced UV chiroptical responses, including strong circular dichroism near 270 nm. The nanohelices show exceptional stability, retaining their UV chiral plasmonic signatures after >4 years of ambient storage and under strongly acidic conditions (pH 1.9). As a proof of concept, we translate the UV chiroptical resonance shifts into a refractive index sensing scheme, demonstrating competitive detection performance in the UV. These results establish silicon nanohelices as a chemically robust, application-ready platform for UV chiral plasmonics.
Capacitive micromachined ultrasonic transducers (CMUTs) are key components of ultrasonic technology, which have broad applications in medical, industrial, and military fields. The theoretical model enables rapid analysis of their electromechanical performances, which in turn guides the iterative design of CMUTs arrays. The parallel-plate capacitive devices with variable cavity height, such as T-shape cavity CMUTs, utilize the electrostatic softening effect to induce piston-like deflection in the membrane and have been demonstrated to significantly decrease the collapse voltage while improving the acoustic performance. However, the theoretical model is lacking for CMUTs with T-shape cavities, limiting their optimization design. This paper introduces theoretical models for CMUTs with T-shape cavities actuated by electrostatic force. The model integrates the Galerkin method, a partial expansion method of nonlinear electrostatic force, and an energy equivalence method, thus facilitating the derivation of theoretical expressions for key mechanical behaviors, such as static deflection, collapse voltage, and resonant frequency. The finite element model and experimental verification are used to demonstrate the theoretical results, showcasing generality and high analytical accuracy (error less than 5%) over a large range of bias voltages (up to 90% of the collapse voltage), membrane dimensions (diameter-to-thickness ratio of 30 to 110), and cavity heights (cavity height-to-membrane thickness ratio of 0.2 to 1.0). Meanwhile, theoretical models are used in the optimization analysis, which demonstrates that the CMUTs with T-shape cavities can achieve a 12% increase in average membrane displacement and 47% decrease in collapse voltage compared to conventional CMUTs. The theoretical expressions can serve as a basis for the design of a series of parallel-plate capacitive devices with variable cavity height.
Efficient delivery of large, negatively charged self-amplifying RNA (saRNA) into dendritic cells (DCs) is critical for next-generation cancer vaccines. However, this remains challenging due to the high sensitivity of DCs to chemical carriers and high-voltage electroporation. In this study, an integrated nanopore-electroporation (NEP) microdevice was developed by combining 200 nm track-etched polycarbonate membrane, bidirectional PDMS microfluidic channels, and Pt/ITO electrodes to localize the electric field and induce membrane permeabilization at low voltage (≤30 V). Multiphysics simulations revealed that 200 nm nanopores concentrated the electric field at the cell-membrane interface, generating transmembrane potentials exceeding 3 V. Using DC2.4, the NEP system achieved 75% propidium iodide (PI) uptake at 25 V with 90% viability, confirming controllable nanoscale perforation. Direct delivery of GFP-encoding saRNA achieved approximately 50% transfection efficiency with sustained protein expression for more than 96 h, significantly outperforming mRNA at an equal dose. Long-term viability (>85% at 96 h) and negligible cytotoxicity demonstrated the excellent biocompatibility of the device. This reagent-minimal, modular NEP platform thus provided a high-efficiency, low-toxicity route for saRNA delivery into hard-to-transfect immune cells, offering a versatile engineering framework for DC-based cancer immunotherapy, RNA vaccine development, and broader cell gene-modification applications.
Effective cancer immunotherapy is hindered by immunosuppressive crosstalk within the tumor microenvironment. We engineered a tumor immune microenvironment-on-a-chip (TIMoC) that recapitulates the vascularized, hypoxic, and spatially organized niche of human solid liver tumors. We employed TIMoC to dissect the reciprocal interaction between macrophages and natural killer (NK) cells. Macrophages induced NK cell dysfunction, while dysfunctional NK cells promoted M2 macrophage polarization. This bidirectional impairment created a self-perpetuating immunosuppressive loop. TIMoC served as an in vitro screening tool, confirming the limited efficacy of TIGIT blockade in a multicellular context and revealing synergistic anti-tumor activity for combinations of a macrophage-reprogramming agent (resiquimod) with NK cell-targeting antibodies. By incorporating patient-derived organotypic tumor spheroids and autologous immune cells, the personalized TIMoC platform modeled patient-specific responses and evaluated effective drug combinations, demonstrating its potential to guide precision immunotherapy. This work elucidates a key immunosuppressive axis and introduces a versatile platform for rationally designing combination immunotherapies.
Quantitative assessment of chimeric antigen receptor T (CAR-T) activity in solid tumors remains challenging, as immune engagement can induce early mechanical softening of tumor spheroids before overt cell death becomes detectable. To address this limitation, we developed Cytotongue, a three-dimensional (3D)-printed microfluidic aspiration system for real-time, treatment-integrated aspiration-response phenotyping of live tumor spheroids. Using human epidermal growth factor receptor 2 (HER2)-positive BT-474 breast cancer spheroids exposed to CAR-T cells, the Cytotongue system quantified aspiration-induced elongation dynamics and assay-specific deformation indices derived from an empirical biphasic deformation framework. This approach enabled detection of CAR-T-induced mechanical softening at low effector-to-target ratios (1:1-2:1) within 24 h, conditions under which conventional propidium iodide (PI) staining showed minimal response. System-level validation demonstrated robust performance, with high Z' factors (0.74-0.93), large effect sizes, and low coefficients of variation. Moreover, Cytotongue distinguished CAR-T-associated mechanical softening accompanied by apoptosis from doxorubicin-induced deformation responses despite comparable PI readouts. Collectively, this work introduces a treatment-integrated microfluidic aspiration system that enables real-time mechanical phenotyping of tumor spheroids, providing a new physical dimension for evaluating immunotherapeutic and drug responses.
The relentless miniaturization of microelectronics demands energy storage systems with ultrahigh energy and power densities in ultracompact footprints. Microsupercapacitors (MSCs) are promising due to their rapid charge-discharge capabilities, but conventional electrode architectures suffer from a trade-off between energy and power density as thickness increases, exacerbated by electron transport resistance and mechanical instability. Here, we present a paradigm-shifting, truly millimeter-thick (up to 1.2 mm), high-aspect-ratio (7:1), three-dimensional microelectrode architecture that fundamentally decouples electron transport distance from electrode thickness. By roll-to-roll calendering of alternating multilayer current collectors and electrode films, we achieve precisely aligned multilayer structures with total thicknesses exceeding 1 mm, specifically demonstrating a 1.2 mm-thick, 9-layer device in an ultracompact footprint, followed by precision laser engraving to define interdigitated gaps below 180 µm, establishing a parallel electron transport network. Therefore, the architecture enables linear scaling of both energy and power density with thickness, delivering a peak energy density of 1733 µWh cm-2, comparable to that of 3D micro-batteries, and a peak power density of 153 mW cm-2, surpassing that of the state-of-the-art microsupercapacitors, alongside exceptional stability demonstrated by 95% capacitance retention after 10,000 cycles at 2000 mV s-1. Roll-to-roll manufacturing combined with laser engraving enables large-area, industrial-scale production of microsupercapacitors, resolving the persistent trilemma among electrode thickness, microscale resolution, and mechanical stability. This advancement delivers a transformative solution for next-generation micro-energy systems achieving high energy density, high power density, and strong mechanical stability.
Nano-thick epidermal electronics have extraordinary lightweight and conformability characteristics, thus enable high-quality signal acquisition and imperceptible wearing experience. However, reducing the device thickness to nanoscale makes the rigidity extremely low thus the device curls spontaneously, imposing unprecedented challenges to the device fabrication, handling and recycling. To resolve the dilemma of epidermal electronics, we propose an edge-supporting strategy that maintains the benefits of nano thickness, meanwhile obtains practicality and reusability via reinforced edge scaffold. The nano-thick (~500 nm thick) core area of the device provides perfect conformal, firm, and burdenless contact with human skin. On the other hand, the edge scaffold (~5 μm thick) plays an important role in protecting internal fragile film from external forces and providing rigid support during handling and recycling. In addition, the edge scaffold enables spontaneous release of nano-thick electronics from the substrate in water, getting rid of the use of sacrificial layers and etchants. We demonstrate the use of the edge-supporting strategy to develop a physiological electrode with low impedance less than 3.6 kΩ at 1 kHz, a resistance-type temperature sensor with fast response time less than 0.5 ms and a nanoengineering-based pressure sensor with high sensitivity of 62.9 kPa-1. Our work provides an effective method to handle and release nano-thick electronics without using external stamps, thereby enabling truly practical applications of ultrathin flexible electronics in the fields of wearable technology, healthcare and so on.
Film bulk acoustic resonators (FBARs) are widely used in radio frequency (RF) filters for wireless communication because of their high operating frequency and high quality factor. With the increase of high-power applications, ensuring device robustness has become a critical challenge. This study presents an investigation into the high-power failure behaviors and mechanisms of FBARs, specifically examining the role of active area, film thickness, and geometry. Experimental results demonstrate that small-area FBARs exhibit distinct failure characteristics compared to large-area devices. Small-area devices are governed by progressive spallation at electrode edges, which is induced by high-temperature oxidation and stress concentration, whereas large-area FBARs are prone to sudden structural fracture or short-circuiting caused by excessive thermal stress. Crucially, the study reveals a thickness-dependent transition in large-area devices, where short-circuiting and structural fracture correspond to distinct stress-severity regimes. Furthermore, dynamic evaluations demonstrate that these FBARs preserve strict electrical linearity right up to the point of catastrophic collapse. Based on these phenomenological findings, a thermo-mechanical coupling mechanism is proposed that goes beyond the conventional thermal-only model. Finally, we propose new design guidelines to enhance the power handling capability of FBAR devices.
Drug delivery strategies with excellent spatiotemporal controllability and biocompatibility hold exciting prospects in personalized medicine and biopharmaceuticals. Among them, all-optical delivery strategies have attracted significant attention due to their low invasiveness and high positioning accuracy. However, current all-optical schemes predominantly rely on conservative optical gradient force trapping or transient scattering forces propulsion, which cannot sustain stable driving and limit delivery speed. Here, we present an all-optical delivery strategy based on Airy beam, enabling high-speed and subcellular-precision delivery through the synergistic regulation of gradient-scattering forces. The dominant scattering force of Airy beams propels drug carriers at speeds exceeding 400 µm/s, while the gradient force ensures a subcellular delivery precision of ~1 µm during high-speed propulsion. Quantitative measurements using a custom axial imaging module reveal at least a 7-fold speed enhancement over conventional all-optical methods. When applied to anticancer drug delivery, the localized mechanical forces generated by high-speed propulsion enhance stable adhesion of carriers to the cell membrane, thereby promoting drug uptake and accelerating tumor cells apoptosis. This work establishes Airy beams as a powerful tool for drug delivery, opening a new avenue for high-performance all-optical delivery platforms.
This study investigates the miniaturization pathway of cycloidal mass spectrometry, a technique distinguished by its inherent perfect-focusing properties, and introduces a planar, stacked-layer MEMS-based cycloidal mass analyzer. A comprehensive numerical analysis was conducted to elucidate the effects of key design parameters, namely electric sector geometry, electrode geometry, and operating vacuum, on ion focusing and mass separation performance. The results indicate that appropriate optimization of geometric dimensions and electrode arrangement density effectively mitigates electric field distortion, thereby improving ion beam focusing and resolution. Configured with an electric sector size of 32.60 mm × 41.10 mm × 6.85 mm and 83 electrode pairs, and operated under a vacuum of 1 × 10-3 Pa, the device achieves full width at half maximum (FWHM) values of approximately 0.03 Da for light ions (H2+, He+) and about 0.40 Da for heavier ions (Ar+, CO2+). Across the m/z range of 2-50 Da, the analyzer maintains a resolution greater than 99 and an ion detection efficiency exceeding 92%, demonstrating robust separation and transmission performance over a wide mass-to-charge range.
Piezo channels are integral to numerous physiological functions: Piezo 1 modulates blood pressure through shear stress sensing and plays a role in cellular development and epigenetic processes. Conversely, Piezo 2 is implicated in sensory perceptions, including tactile sensation, balance, and nociception. Piezo channels are biophysical mechanosensitive ion channels that convert mechanical forces into electrical signals within cells. In this review, we investigate the recent advancements in the application of noninvasive biophysical techniques to modulate Piezo ion channels, thereby affecting diverse physiological functions. We also explore different electromechanical modulation methods for Piezo channels and discuss their significance. Furthermore, we emphasize various optical and magnetic techniques for modulating Piezo channels. Finally, we examine the potential applications of Piezo channel modulation for the treatment of neurodegenerative diseases, an area with significant potential to impact healthcare. The noninvasive activation of Piezo ion channels through mechanobiology holds significant potential for advancements in healthcare and for understanding and addressing neurodegenerative and degenerative diseases.
Strain-diluted dissipation has emerged as a promising technique to modulate the quality (Q) factor of resonant nanoelectromechanical systems (NEMS). However, comprehensive understanding and precise control of this effect under varying temperatures have remained elusive. Here we investigate the temperature-modulated dissipation dilution mechanisms in two-dimensional (2D) NEMS resonators. We develop an explicit temperature-dependent dissipation dilution model highlighting the roles of thermally-induced strain and temperature fluctuations during vibration, which well captures experimental observations: with temperature increasing from 77 K to 355 K, Q factor in graphene resonators first decreases and then increases, while that in molybdenum disulfide (MoS2) resonators monotonically decreases. Furthermore, based on the model, we design a graphene-MoS2 heterostructure NEMS resonator with near-zero effective thermal expansion, which experimentally exhibits high temperature stability in both frequency and Q factor. Our results advance the understanding of dissipation dilution mechanisms and pave the way for developing thermally-stable resonant transducers and logic components.
