The development of highly sensitive and rapid response/recovery NO2 sensors that can operate at low temperatures (below 100 °C) and room temperature (RT) holds significant scientific value, yet it remains challenging. Inspired by the macroporous structure of butterfly wings in nature, this work significantly enhances the gas-sensing performance of NO2 sensors at low temperatures through the synergy of vortex effects in the ordered macroporous structure of three-dimensional inverse opal (3DIO) and the engineering of organic-inorganic heterojunctions. 3DIO ZnFe2O4/PANI (ZFOP) composites were synthesized via the sacrificial template method and in situ polymerization. Computational fluid dynamics (CFD) simulations revealed that the ordered macropores can induce vortex effects, which effectively reduce gas flow velocity and prolong the duration of surface interactions, thereby enhancing sensitivity. The organic-inorganic heterointerfaces formed between ZnFe2O4 and PANI can effectively stabilize the polymer chains, improve mechanical robustness, and profoundly modulate the electronic properties of the composite. Experimental results demonstrated that the 3DIO ZFOP sensor exhibits ultrahigh response values toward 100 ppm NO2 (SR1 = 242.6 at 98 °C, SR2 = 4.51 at RT) with an ultrafast response speed of 2 s. Furthermore, the 3DIO ZFOP sensor delivered outstanding selectivity, superior repeatability, and long-term stability. Density functional theory (DFT) calculations verified that the substantial charge transfer during the adsorption process strengthens the adsorption interactions toward NO2. This work provides a highly promising strategy for the facile fabrication of low-power NO2 sensors with high sensitivity and rapid response/recovery rates.
Precise discrimination of structural homologues, namely molecules with highly similar physicochemical properties, remains a critical bottleneck in artificial olfaction and limits its application in precision medicine and environmental monitoring. While deep learning has enhanced pattern recognition, current purely data-driven models often lack physical interpretability and may show limited transferability to homologous analytes not included during training. Here, we present a physics-data dual-driven deep learning framework that integrates time-resolved sensor responses with DFT-derived molecular descriptors, thereby bridging atomic-level electronic information and macroscopic sensing responses for VOC classification. The model achieved direction-averaged accuracy and macro-F1 scores of 83.88 and 83.53%, respectively, in bidirectional validation involving unseen homologues from five chemical categories. Interpretability and ablation analyses further confirmed that the complete DFT-assisted framework contributed substantially to classification. These results demonstrate the potential of integrating microscopic physicochemical descriptors with dynamic sensing signals to recognize unseen VOCs according to their chemical categories, providing a promising route toward interpretable VOC sensing within defined chemical categories.
Flexible capacitive tactile sensors for human biophysical signal monitoring require the simultaneous realization of high sensitivity and a wide detection range, which remains challenging due to the intrinsic structural characteristics of the dielectric layer. Addressing this challenge requires new structural strategies that integrate compliant deformation with mechanical robustness. Here, we present a bioinspired flexible tactile sensor featuring a hierarchical concave microstructure derived from octopus suckers, fabricated via a facile and scalable template-transfer approach that enables precise structural replication and tunable sensing characteristics. Finite element simulations and mechanical characterization were employed to elucidate the pressure-dependent deformation behavior. The results reveal a graded deformation mechanism, where the peripheral rim and central protrusion engage sequentially to delay structural saturation and redistribute stress. As a result, the sensor exhibits segmental sensitivity and high linearity across a wide range of pressures, achieving 8.78 kPa-1 (0-5 kPa), 1.84 kPa-1 (5-230 kPa), and 0.91 kPa-1 (230-500 kPa), with corresponding linearities of 99.34, 97.83, and 98.98%, respectively. Furthermore, the device enables multimodal detection of biophysical signals and spatiotemporal reconstruction of pulse waves, advancing their application in noninvasive medical systems.
Real-time and highly selective monitoring of acetone (C3H6O) is crucial for industrial safety and occupational health. Herein, we report a Ce-doped BiFeO3 (BFO) sensing material synthesized via a sol-gel method, which enables simultaneous regulation of grain size, polarization strength, crystal phase composition, band structure, and defect. The optimized sensor (denoted as BC2FO, 2 mol% Ce-doped) exhibits excellent C3H6O sensing performance over a concentration range of 0.2-10 ppm at 220 °C, and its response value to 2 ppm C3H6O is twice that of the pure BFO sensor. Ex situ XPS and in situ Raman further revealed the sensitive reaction mechanism between C3H6O gas molecules and the BC2FO composite material. Furthermore, the machine learning classification algorithm was introduced for highly accurate identification of C3H6O against isopropanol (C3H8O). An intelligent warning system based on BC2FO micro-electro-mechanical system (MEMS) sensors has been developed, which successfully achieved continuous monitoring and automatic on-site alarm in simulated industrial leakage scenarios. This study presents a comprehensive and robust paradigm combining material engineering, machine learning-assisted recognition, and hardware integration for the detection of C3H6O.
How cell physical state relates to function and stimulus response remains difficult to resolve because most methods measure only one biophysical property at a time. Yet cellular behavior emerges from the interplay of features, including adhesion, morphology, and mechanical dynamics. Building on previous surface plasmon resonance microscopy (SPRM) and related plasmonic microscopy approaches, we developed an SPRM platform for label-free, real-time multiparametric phenotyping of single live cells. By probing the cell-substrate interface with nanometer-scale sensitivity, the platform jointly quantifies three complementary descriptors from the same time-resolved image sequence: adhesion-associated SPR intensity (I), contact area (A), and effective spring constant (k). Applied to Entamoeba histolytica, a highly deformable protozoan parasite, this approach resolved baseline physical states, drug-induced changes, and interaction-dependent responses to bacteria and other cells. Machine learning further classified early single-cell phenotypes relative to treatment-defined viable-reference and non-viable-reference groups. These results establish a label-free SPRM workflow for resolving early interface-associated phenotypic changes in cells during drug exposure and cell-interaction assays.
Timely and reliable identification of bacterial infections in emergency settings remains a major challenge. To address this challenge, here we utilized CRISPR/dCas9-mediated self-assembly to develop a colorimetric and Raman dual-mode biosensor (referred to as dCARD) and applied it to point-of-care bacterial detection. Briefly, dCARD utilized the CRISPR/dCas9-mediated precise recognition of target amplicons to build bridges, attaching gold nanoparticle (GNP)-probe loaded with 4-aminothiophenol (4-ATP) onto the surface of GNP-streptavidin (GNP-SA). The dispersion-to-aggregation change of particles triggered a significant color change from red to purple, along with the formation of Raman hotspots. As a result, the presence of target bacteria was translated into cross-validated color and Raman signals, showcasing high reliability. This biosensor not only maintained the simplicity of visual readout but also integrated the quantitative function of a portable Raman spectrometer, allowing operators to flexibly switch signal modes in different scenarios. The specific primer-based isothermal amplification and CRISPR/dCas9-mediated precise recognition of amplicons together improved the specificity of dCARD, enabling accurate detection of the target bacteria down to about 10 CFU/mL without interference from non-target bacteria. By using a simplified preprocessing strategy, the assay time could be further reduced to less than 30 min. The robustness and programmability of dCARD were also demonstrated by detecting three common bacteria in real urine specimens. We expect this biosensor to evolve into a universal and field-deployable screening tool for bacterial infections.
Biomolecular condensates, formed by proteins, nucleic acids, and other macromolecules through liquid-liquid phase separation, enable molecular enrichment and reaction regulation within membraneless microenvironments. As their physical principles have become better understood, condensates are increasingly being incorporated into in vitro analytical systems. Compared with conventional homogeneous assays, condensate-based systems can enhance effective reaction concentrations, regulate molecular partitioning, accelerate reaction kinetics, and facilitate signal generation through enrichment, selective compartmentalization, and spatial confinement. This review discusses the mechanisms by which biomolecular condensates contribute to in vitro analysis, emphasizing component enrichment, selective partitioning, kinetic enhancement, and programmable phase behavior. We then discuss programmable condensate biosensing systems as emerging platform-level frameworks for molecular recognition, signal processing, and multimodal readout. Representative applications are summarized in nucleic acid and protein analysis, enzyme activity regulation, ion and small-molecule detection, and complex sample analysis. Finally, we highlight key challenges for practical implementation, including robust phase-behavior control in complex biological samples, predictive design, reproducibility, standardization, and biological benchmarking.
Live-cell monitoring of sequence-specific nucleic acids is essential to understanding genome organization, RNA regulation, and disease progression. Clustered regularly interspaced short palindromic repeat (CRISPR)/CRISPR-associated protein (Cas) and Argonaute (Ago) systems provide programmable, guide-directed recognition of DNA or RNA and are increasingly used as platforms for in vivo bioimaging. This review summarizes the structural and mechanistic features of representative CRISPR and Ago effectors and discusses design strategies for sensitive, specific, and multiplexed imaging of genomic loci, extrachromosomal DNA, and endogenous RNA in living cells. We compare the analytical performance and limitations of CRISPR- and Ago-based imaging, with particular emphasis on the major technical and biological challenges affecting their accuracy, applicability, and reliability. Finally, this review offers insights into developing high-resolution and user-friendly bioimaging platforms for fundamental biology and future translational applications.
Acetaminophen is the most commonly used over-the-counter pain relief medication. However, despite its widespread use, high doses of acetaminophen can be toxic and are the leading cause of acute liver failure in children, as well as the most common reason for liver transplants in the United States. Saliva collection offers a more convenient alternative to traditional blood sampling methods, with acetaminophen levels in saliva closely reflecting those in the blood, making it a reliable substitute for measuring acetaminophen levels. In this study, we introduce Lollylab, a lollipop-inspired device designed to collect saliva and transfer it to a laser-induced graphene electrode. The acetaminophen levels in the saliva are then detected using electrochemical techniques by measuring the oxidation of acetaminophen on the electrode surface. Lollylab is particularly user-friendly for children, allowing them to collect saliva simply by licking the candy.
Extracellular vesicles (EVs) are promising liquid-biopsy analytes for hepatocellular carcinoma (HCC) because they carry complementary biomarker information in both surface and luminal compartments and are accessible from multiple clinically relevant biofluids. In HCC, however, the diagnostic performance of EV assays is constrained by low tumor-derived EV fraction, chronic liver disease confounding, matrix-dependent background, enrichment bias, and limited cross-platform comparability. In this Review, we examine EV-based HCC diagnostics through the lens of sensing technology and analytical workflow design. We organize the field around biofluid-specific design constraints, isolation strategies, recognition interfaces, transduction architectures, and benchmarking metrics that determine assay performance in complex liver-disease matrices. We compare established and emerging enrichment methods together with major sensing modalities for EV protein and nucleic-acid analysis, including fluorescence, electrochemical, electrochemiluminescent, field-effect transistor, plasmonic, mass spectrometric, and single-EV platforms. We further discuss how enrichment modules reshape the biological EV fraction being measured and why assay claims should be judged using matrix-aware limits of detection, recovery, throughput, and cost per test. By linking HCC-relevant EV biology to platform selection and validation strategy, this review summarizes design considerations for HCC-EV sensing workflows across early screening, differential diagnosis from benign liver disease, prognosis assessment, recurrence surveillance, and treatment monitoring.
Owing to their porous structures and highly tunable properties, metal-organic frameworks (MOFs) have emerged as a current research hotspot in the field of electrochemiluminescence (ECL). Nevertheless, research on main-group-metal-based MOFs in the ECL field remains limited and mainly focuses on aluminum-based MOFs and indium-based MOFs. This work developed bismuth-based MOFs (Bi-MOFs) as ECL luminophores and explored defect engineering to enhance their ECL efficiency. Using Bi3+ as the metal node and 4,4',4″-((1,3,5-triazine-2,4,6-triyl)tris(azanediyl))tribenzoic acid (H3TATAB) as the organic ligand, a series of Bi-MOFs with tunable defect states, namely, Bi-TATAB-X MOFs, were synthesized. Here, X represented the synthesis temperature. Research found that the synthesis temperature and synthesis time affected the defect level, and there was a clear correlation between the defect level and the ECL efficiency, which was similar to a volcanic eruption. In the Bi-TATAB-X MOF series, the Bi-TATAB-130 MOF, which was synthesized at 130 °C for 2 days, stood out for its excellent cathodic ECL performance due to the accelerated electron injection rate and improved electron-hole recombination efficiency by the defects. The Bi-TATAB-130 MOF integrated dual-locked Y-type DNAzyme-mediated catalytic hairpin self-assembly (CHA) to construct the ECL aptasensor for detecting dibutyl phthalate (DBP), which is a representative of plasticizers, with a limit of detection (LOD) of 3.31 fM. Bi-TATAB-130 not only expanded the ECL emitters of main-group-metal-based MOFs but also established a highly sensitive ECL platform for plasticizer detection.
Since the onset of the COVID-19 pandemic in 2020, the importance of virus detection and development of new diagnostics for emerging or evolving viruses has been a major focus in the field of bioanalytical sensing. Surface-enhanced Raman spectroscopy (SERS) is one technique for virus detection that has taken off in academic research but has not yet transitioned into widespread commercial use despite its enormous potential as a sensitive and specific sensing technology. Herein, we briefly summarize advances in SERS as applied to virus detection over the past 5 years, as there has been rapid development and growth resulting from the need to develop COVID-19 sensors, including critical analysis of the directions of the field. We compare label-free and labeled detection schemes and highlight machine-learning-enabled sensing, with a focus on viruses relevant to human health. Opportunities are proposed and understudied areas are identified in virus SERS sensing alongside outlooks for the field, including the use of moderate affinity agents to facilitate multiplex sensing and adaptation of machine learning algorithms from fields beyond sensing.
Flexible electrochemical gas sensors are promising for wearable electronics, yet conventional hydrogel electrolytes suffer from water evaporation, interfacial instability, and performance degradation under long-term operation. Here, we report a self-powered flexible NO2 electrochemical sensor based on a zinc-air battery (ZAB) structure using a poly(4-acryloylmorpholine)-sodium bis(trifluoromethanesulfonyl)imide-propylene carbonate (PACMO-NaTFSI-PC) organogel electrolyte. The nonvolatile organogel effectively suppresses solvent evaporation and eliminates dehydration-induced failure associated with conventional hydrogels. Meanwhile, the high dielectric constant of propylene carbonate promotes ionic dissociation, while weakly coordinated TFSI- anions optimize Zn2+ solvation and accelerate ion transport within the organogel network. The optimized sensor exhibits excellent environmental stability together with high NO2 sensing performance, including a sensitivity of 1158.9 nA/ppm, rapid response/recovery times of 25.6/19.5 s, and a low detection limit of 3.6 ppb. This work provides a viable strategy for developing highly stable self-powered flexible gas sensors for wearable sensing applications under harsh operating conditions.
Efficient interfacial activation of co-reactants to generate sufficient radical intermediates is fundamental to constructing high-performance co-reactant-type electrochemiluminescence (ECL) biosensing platforms. This work reported a microenvironment-engineered cathodic ECL strategy by integrating PSA-HOF as the luminophore with defect-engineered MIL-88B(Fe) as the co-reactant activation modulator for the sensitive detection of perfluorooctanoic acid (PFOA). Specifically, pyrrole-2-carboxylic acid-regulated Pca-MIL-88B(Fe), as a defect-engineered material, generated coordinatively unsaturated, electron-deficient Fe centers with exceptional Lewis acidity. These defect-derived sites established an optimal interfacial microenvironment for persulfate activation by critically enhancing K2S2O8 enrichment, optimizing adsorption configuration, facilitating electron redistribution, and promoting O-O bond polarization. This collective action significantly lowered the cathodic driving force required for K2S2O8 reduction and accelerated SO4•- generation through a Lewis acid-mediated pathway, ultimately leading to a marked enhancement in cathodic ECL efficiency. As a result, the PSA-HOF/Pca-MIL-88B(Fe) composite exhibited an approximately threefold increase in ECL intensity compared with pristine PSA-HOF. Leveraging this efficient signal amplification mechanism, the developed aptasensor exhibited superior analytical performance with the assistance of PFOA aptamer, featuring a wide linear response range spanning from 10 fM to 1 μM and an ultralow detection limit of 9.8 fM, while also exhibiting outstanding stability, selectivity, and reproducibility. This work provided a robust physicochemical insight into defect-mediated co-reactant activation and paved the way for designing high-efficiency ECL emitters for environmental monitoring.
In light of the escalating demand for highly sensitive and reliable detection methods for ractopamine (RAC) residues in food safety, this study introduces an innovative dual-mode immunosensor. This sensor is constructed from a PZT microcantilever integrated with CTFx@FTP composite nanoparticles and a PANI@rGO-COOH nanofiber membrane. It is capable of simultaneously generating electrical signals and cantilever vibration-amplitude signals, thus facilitating quantitative detection of RAC. B-site Fe incorporation was employed to regulate mixed-valence states and oxygen-vacancy-related defects in CaTi1-xFexO3-σ, while FeTCPP-mediated π-d electronic coupling and the π-conjugated PANI@rGO-COOH network were integrated to construct a composite sensing interface that facilitates charge transfer and electromechanical transduction, enabling specific immunorecognition events to be converted into measurable dual-mode signals. At an Fe doping level of x = 0.20, the sensor demonstrates optimal interfacial transport characteristics, achieving a remarkably low limit of detection of 0.17 ng mL-1, alongside a wide detection range of 0.17 to 50 ng mL-1, and good linearity. The dual-mode signals exhibit good selectivity, storage stability, reproducibility, and repeatability. In spiked pig and bovine serum samples, recovery rates of 97-104% and consistency with ELISA results confirmed the applicability of the proposed sensor in complex biological matrices. The defect-engineering and interface-coupling-based electromechanical synergistic amplification strategy proposed in this study offers a promising pathway for the development of dual-mode electrical and mechanical immunosensors for small molecules. This approach holds significant potential in food safety monitoring.
Electrochemical biosensors integrating the programmable nucleic acid recognition of CRISPR systems with the low-cost and portable electrochemical transduction have emerged as powerful sensing platforms for molecular diagnostics. Early designs mainly relied on turn-off signal transduction, where target-activated CRISPR enzymes cleave probes conjugated to the electrode, resulting in the release of redox reporters from the electrode surface and the consequent signal decrease. Although conceptually straightforward, these turn-off sensors are intrinsically limited by high background, low sensitivity, and large signal variations. To address these limitations, recent efforts have increasingly shifted toward turn-on strategies, in which electrochemical signals are generated in response to target binding. This review highlights recent advances in applying CRISPR technology to electrochemical biosensing, with a focus on the design principles of CRISPR systems and molecular assembly to achieve turn-off and turn-on signal transduction. Nanomaterials, DNA nanotechnology, and amplification strategies facilitate emerging turn-on approaches for sensitive electrochemical sensing. Key challenges and research needs include improving the limit of detection, robustness, and applicability to point-of-care and on-site testing. This review emphasizes the importance of signal-transduction designs and provides perspectives for developing sensitive, specific, and practical CRISPR-based electrochemical biosensors.
Urinary cystatin C (uCysC) has emerged as an endogenous biomarker for renal impairment. In managing renal pathologies, early screening and monitoring of disease progression are crucial for both at-risk individuals and diagnosed patients. The ultimate goal is to prevent and mitigate the associated risks of morbidity and mortality, thereby promoting longevity. However, conventional methods for quantifying uCysC have limitations that hinder their applications in point-of-care for resource-limited settings. In this study, we provide a sample-to-answer quantitative rapid test to monitor uCysC levels. The device uses a vent for urine sampling with a fixed volume, gravitational sedimentation for effective on-device reaction, and a dual-microbead immunoassay for the length-based visual quantification of uCysC. The achieved limit of detection (LOD) is 33.33 ng/mL, with excellent selectivity, pH level tolerance, and high accuracy. This all-in-one device bypasses peripheral lab settings to enable low-cost testing of renal impairment, which will benefit both general users and medical practitioners for telemedicine applications and large-scale disease self-management.
Rheumatic fever is an immunologically mediated inflammatory disease that develops following group A streptococcal infection and triggers valvular inflammation that can progress to chronic rheumatic heart disease (RHD), a major cause of cardiovascular morbidity and mortality in low-resource countries. Early diagnosis remains critical for preventing irreversible valvular damage. Elevated levels of circulating endothelial cell adhesion molecules (CAM), particularly E-selectin; ICAM-1; and VCAM-1 have been associated with disease severity, highlighting their potential as diagnostic biomarkers. Here, we report on the development of a polymeric fluorescent probe based on N-(2-hydroxypropyl)methacrylamide (HPMA) for the sensitive detection of serum E-selectin and VCAM-1. The probe incorporates a high-affinity CAM-binding peptide (Esbp or Vbp), fluorescent reporters (FITC or IR783), and a hexa-histidine (His6)-tag to enable plate surface immobilization. Polymers were synthesized with increasing IR783 molar content to induce fluorescence quenching at high dye loadings while preserving activation upon interaction with serum biomarkers. Their sensing performance was evaluated in both buffer and serum using fluorescence-based assays. Surface-immobilized probes demonstrated dose-dependent detection of labeled E-selectin and VCAM-1 in the microgram scale (1-10 µg/ml) in PBS. In serum, the probe exhibited specific and quantitative binding to E-selectin, while VCAM-1 binding was reduced, likely due to interference from abundant serum proteins. Optimization of IR783 content in the polymer revealed that 2.5 mol % dye loading enabled effective fluorescence activation upon interaction with E-selectin, whereas higher dye loadings led to self-quenching and reduced responsiveness, due to IR783-driven self-assembly that hindered signal activation at the tested biomarker concentrations. The optimized probe successfully detected E-selectin in spiked serum samples. This study demonstrates a simple, modular HPMA-based sensing platform for detecting circulating CAM that can be readily adapted to a broad range of serum biomarkers. By utilizing polymer-based probes that are inherently more stable and lower in cost than antibody-based reagents, the platform addresses key limitations of current diagnostics and offers a promising route toward affordable, rapid assays for RHD in low-resource settings. Further optimization may enhance sensitivity and expand biomarker coverage to improve clinical applicability.
Hydrogen-substituted graphdiyne (HsGDY) is a conjugated carbon material containing sp- and sp2-hybridized carbon atoms that has recently been shown to mechanically actuate upon exposure to acetone, suggesting a strong molecular interaction. Here, we report the first acetone sensor based on HsGDY films synthesized directly on copper foil. Electrochemical impedance spectroscopy reveals pronounced, concentration-dependent impedance changes upon acetone exposure that are absent in bare copper and carbon paper electrodes and strongly suppressed after thermal treatment that reduces alkyne retention. Distribution-of-relaxation-times analysis indicates that acetone primarily alters mass-transport processes within the HsGDY network. Raman spectroscopy shows no detectable change in the alkyne vibrational signature, whereas solid-state 13C NMR reveals clear structural evolution, highlighting the importance of complementary characterization. Structurally related solvents do not produce comparable responses, demonstrating selectivity toward acetone. These results establish HsGDY as a room-temperature acetone sensor and clarify the mass-transfer-dominated mechanism underlying its response.
