Cancer remains one of the leading causes of mortality worldwide, early detection is essential for improving cancer patient survival. Although advanced imaging techniques and point-of-care (POC) devices have emerged, they are often invasive, costly and insufficiently sensitive for detecting cancers at their earliest stages. Exosomes are nanoscale extracellular vesicles (EVs) (30–150 nm) released by tumor cells that contain proteins, lipids, RNA, and DNA. Because they represent the molecular status of their parent cells, exosomes can be used as biomarkers. However, the practical translation of traditional isolation techniques, such as ultracentrifugation, precipitation, and immunoaffinity capture, is limited due to their time-consuming nature, high sample volume requirements, and tendency to produce heterogeneous preparations. While nanomaterials excel in tunability and controlled engineering, exosomes provide a biologically informed and physiologically relevant alternative. Microfluidic platforms coupled with biosensors, such as lab-on-a-chip (LOC) devices, paper-based platforms, and other tiny systems, provide for rapid, sensitive, and portable exosome analysis in POC testing. The use of aptamers and peptides as biorecognition elements improves these systems by providing greater reproducibility, lower cost, sensitivity and specificity than antibodies, increasing clinical feasibility. This review discusses current breakthroughs in aptamer and peptide functionalized microfluidic biosensors for exosome-based cancer diagnostics, focusing on their translation into POC applications. Finally, we highlight present challenges, such as repeatability, clinical validation, and scalable production, as well as potential future prospects, such as smartphone-based readouts and artificial intelligence (AI) integration for next-generation cancer diagnostics.
Continuous-flow microfluidics presents a scalable, consistent, and eco-friendly alternative to conventional batch nanomanufacturing for applications in energy, medicine, and the environment. This review examines how channel designs, laminar flow, and droplet formation allow for precise control over the dimensions and shapes of metallic, polymeric, and hybrid nanostructures. It emphasizes the “green” benefits, such as decreased reagent use and the employment of bio-based solvents, and evaluates scale-up methods like numbering-up and parallelization for industrial feasibility. Additionally, emerging applications, including image-guided drug delivery, diagnostics, photocatalysis, energy storage, and antimicrobial systems, are explored to demonstrate the translational and interdisciplinary potential of this technology. The review also identifies key challenges, such as clogging, fouling, a lack of standardization, and the necessity for strong industrial adoption, while suggesting future directions in AI-optimized process development and automated nanomanufacturing. This review positions continuous-flow microfluidic synthesis as a scalable and sustainable method for the industrial production of advanced nanomaterials.
Building upon an established single-cell hyperspectral drug stimulation detection method, this study systematically developed a dual-mode drug stimulation analysis-genotyping platform by utilizing single-cell nucleic acid amplification technology. On one hand, hyperspectral imaging was employed to analyze drug stimulation responses at the single-cell level. On the other hand, single-cell whole-genome amplification combined with sequencing was employed for genotyping of the targeted cells. To support the complicated on-chip biochemical processing and detection workflows, we designed an integrated microfluidic chip incorporating on-chip pneumatic membrane microvalves. This platform facilitates precise fluid handling, integrates multi-step reactions for single-cell nucleic acid analysis, and allows hyperspectral assessment of single-cell drug stimulation responses. As a result, by correlating the spectral response of single cells under drug stimulation with their genotypes, this study successfully establishes a comprehensive analysis that bridges the gap from phenotype to genotype.
To prevent arteriovenous fistula (AVF) stenosis, elucidating the functional regulation mechanism of endothelial cells by wall shear stress (WSS) at the AVF site was crucial. In this study, we proposed a novel microfluidic model to reproduce the WSS within the AVF region and revealed the associated mechanobiology mechanism. The computed tomography angiography (CTA) and computational fluid dynamics (CFD) simulation techniques were used to obtain the WSS with different characteristics on endothelial cell surface at AVF site. The constructed microfluidic model was employed to perform hemodynamics and cell experiments. The hemodynamic experiment results demonstrated that the AVF-specific WSS were successfully reproduced by the proposed model. Our cell-based experiments revealed that the pulsatile WSS promoted cell migration. Conversely, low intensity WSS and oscillatory pathological WSS were observed to significantly reduce endothelial cell migration, with oscillatory WSS resulting in the most severe damage to cell function. The neointimal hyperplasia (NIH) progression of AVF was accelerated by WSS-mediated endothelial injury. The proposed microfluidic model provided an in vitro research platform for clarifying AVF stenosis mechanisms, and the findings indicated that optimizing hemodynamic conditions at the AVF site was a potential strategy for maintaining patency of AVF.
Visual experiments on a microfluidic chip provide a powerful platform for mechanism studies on multiphase flow and displacements in subsurface porous media. Generally, the inlet-outlet conditions and injection confinement naturally inherit those from microfluidics for microsystems, which can influence the flow behavior and displacement stages strongly, yet may not be appropriate for subsurface scenarios. In this work, inspired by field-scale injection processes, a capillary-number-mapped inlet-boundary design is developed and tested for heterogeneous micromodels, in which point injection is combined with variable boundary confinement to introduce a progressive along-path capillary number evolution and reduce inlet-induced discontinuities. Compared with two common configurations: a uniform injection with constant confinement and a point injection with constant confinement, experiments spanning multiple orders of magnitude in the capillary number identify three representative stages with the corresponding capillary number ranges highly sensitive to inlet–boundary designs. The proposed geometry design suppresses rapid early-stage transverse invasion at the inlet and enables a progressive expansion of matrix mobilization as capillary number increases, thereby alleviating the abrupt mobilization transition observed in conventional point-injection layouts while reducing the experimental fluctuation caused by stochastic pathway selection under uniform injection. Interface-evolution metrics, pressure-drop responses, and skeleton-based topology analyses support these findings consistently, suggesting that the proposed design helps improve the physical consistency of quasi-two-dimensional micromodel displacement experiments for interpreting in-plane preferential flow and matrix mobilization.
Viscoelastic microfluidics has emerged as a powerful, label-free platform for submicron particle manipulation by exploiting the intrinsic rheological properties of polymeric fluids without the need for external fields. However, achieving precise focusing at the submicron scale remains challenging due to the significantly weakened hydrodynamic forces acting on small particles. In this study, we present a high-performance microfluidic system that integrates high-aspect-ratio (HAR) straight channels with hydrodynamic pinching to enhance elasto-inertial focusing of micro- (2.1 μm) and submicron-sized (0.87 μm and 0.5 μm) particles. Using devices with physical aspect ratios (ARphys = H/W) of 3.3, 5, and 10, we systematically investigated the transition of particle migration behavior across varying PEO concentrations and flow conditions. To unify the combined effects of static channel geometry and dynamic flow-induced confinement, we introduce the concept of the effective aspect ratio (AReff). Our results show that while an ARphys of 10 enables stable sheathless focusing, devices with lower ARphys can achieve comparable performance by increasing the flow rate ratio (FRR), thereby elevating AReff. This synergistic coupling amplifies the lateral elastic stress gradient by confining particles within an intensively compressed sample stream, effectively overcoming the volume-dependent force scaling limitations (Fel ∝ d3). By optimizing the flow configuration within the ARphys = 10 device (FRR = 4, Qtotal = 40 µL/min, AReff ≈ 50), we successfully suppressed the Brownian diffusion, achieving the highest focusing quality (FWHM ≈ 10 μm for 0.5 μm particles). By quantitatively evaluating focusing performance through the AReff, this work provides a robust physical framework and practical design guidelines for the manipulation of sub-micrometer targets in relatively short channels (L/Dh ≈ 255). The proposed device offers a high-resolution, high-throughput solution for future analytical and diagnostic applications.
Biofilm associated infections represent a significant clinical challenge due to their inherent tolerance to antibiotics, ability to evade host immune responses, and persistence in chronic disease settings. Conventional therapies often fail to eradicate mature biofilms, particularly in polymicrobial infections or on medical devices, highlighting the urgent need for innovative interventions. Nanomedicine has emerged as a promising strategy, offering platforms such as polymeric nanoparticles, liposomes, metallic, and hybrid systems capable of enhancing drug penetration, targeting microbial communities, and enabling controlled release of antimicrobials. Comparative studies suggest that combination approaches, including enzyme functionalized or stimuli responsive nanoparticles paired with antibiotics, may outperform single modality treatments by simultaneously disrupting the biofilm matrix and delivering active agents. Despite encouraging preclinical findings, several barriers hinder clinical translation. These include challenges in large scale synthesis, reproducibility, stability, regulatory approval, and comprehensive safety assessment, as well as the complexity of accurately modeling biofilm heterogeneity and host interactions. Additionally, nanoparticle biodistribution, protein corona formation, and potential microbial adaptation present further translational hurdles. Future research should focus on optimizing multifunctional nanoplatforms, refining preclinical biofilm models, assessing long term safety, and integrating strategies to anticipate resistance. Addressing these interconnected scientific, technical, and regulatory challenges will likely be critical for harnessing nanomedicine as a transformative tool in the management of persistent biofilm related infections.
Label-free cell sorting technology holds significant value in fundamental biomedical research and clinical precision diagnostics. However, existing dielectrophoresis sorting chips generally face the challenge of balancing high throughput with high resolution, while traditional metal electrodes are prone to bubble formation and sample contamination. To address these issues, this study proposes a ring-array dielectrophoresis microfluidic chip based on liquid electrodes. Through the synergistic design of semicircular flow channels and 36 circumferentially arranged liquid electrodes, the effective selection region of the electric field is significantly extended within the limited chip area, while a gradient electric field is formed along the flow direction, enabling continuous and dynamic deflection separation of particles of similar size. Key parameters such as voltage, frequency, and flow rate were optimized using a multiphysics simulation system, and the chip was fabricated via 3D printing. Fluorescent polystyrene microspheres of 5 μm and 15 μm were used to simulate non-target and target cells, respectively, and sorting experiments were conducted under various optimized parameter conditions. Compared to existing dielectrophoresis chips, this design maintains excellent resolution during high-throughput operations while fundamentally eliminating bubble formation and sample contamination caused by liquid electrodes, featuring a simple structure and convenient fabrication. This study provides a novel solution that balances performance and practicality for label-free, high-activity cell sorting, with subsequent efforts focused on validation with real-cell samples and exploration of integration with point-of-care diagnostic systems.
Capillaric microfluidics offer miniaturization and simplified liquid handling without the need for peripheral equipment. Paper materials like cellulose and nitrocellulose are commonly used as passive pumps in capillaric microfluidics due to their porous structure capable of wicking relatively large volumes of liquid in a compact device footprint. Growing interest in capillaric microfluidics over the past four decades has provided a wide range of self-powered fluidic control elements and expanded the range of available applications. However, the integration of paper pumps with microchannels is an often overlooked and delicate process with tips and tricks – often spread via word of mouth – that can significantly impact the performance and reproducibility of capillaric microfluidics. In this work, we compared fabrication and connection strategies between a 3D printed microfluidic device, nitrocellulose substrate, and cellulose wicking pump to determine the optimal design for reliable and user-friendly handling. Leveraging prior work in the field, we found that a thick cellulose wicking pump, secured to the device and connected to the nitrocellulose substrate using adhesive tape to apply pressure resulted in low variation of the fluid flow within the system. We also showed that overlapping the nitrocellulose with the outlet microchannel produced repeatable results for users of various training levels. These results demonstrate an effective, reliable, and user-friendly microchannel-paper interface that could prove suitable for a broad range of applications, including diagnostic immunoassays. Our characterization and conclusions also offer transparency into a crucial yet under-discussed aspect of capillaric microfluidics design and fabrication.
We numerically investigate wettability-gradient-driven propulsion of a shear-thinning droplet on micro-textured and confined substrates. The droplet rheology is modelled using the Carreau–Yasuda constitutive relation, and interfacial dynamics are resolved using a finite-element level-set framework incorporating consistent slip regularization and mesh-independent resolution. We develop a scaling analysis from the numerical framework, establishing the coupled dependence of droplet propulsion on wettability gradient, confinement, roughness, and contact-line dynamics. This is achieved through non-dimensional parameters including capillary number (Ca), roughness height ( ), and rescaled slip length ( ε ), providing mechanistic insights into wettability-driven transport of non-Newtonian fluids. A theoretical scaling relation describing roughness-dependent propulsion is further developed and validated against numerical results. Droplet propulsion exhibits a non-monotonic dependence on the flow behaviour index, with optimal transport arising in shear-thinning regimes that enhance internal circulation while balancing kinetic energy generation and viscous dissipation. Surface texture strongly modulates contact-line stresses; triangular asperities generate the highest propulsion by amplifying local shear and sustaining pressure imbalance. Roughness height and confinement ratio both display optimal intermediate ranges, beyond which contact-line distortion, curvature-induced resistance, and increased hydrodynamic resistance suppress motion. The results identify the coupled roles of rheology, texture, confinement, and energetics in governing passive droplet transport on wettability-graded surfaces.
Continuously rotating a single microdroplet within a circular loop channel keeps the droplet in motion while holding it within a fixed field of view. Existing approaches to droplet rotation, however, typically rely on multi-layer devices equipped with pneumatic control valves, which complicate both fabrication and operation. In this study, we present a method, free of embedded pneumatic valves, in which a single droplet is rotated in a circular loop microchannel by applying three pressure waveforms, analogous to the driving currents of a three-phase synchronous motor, to three radially arranged control channels. The operating principle is formulated using this motor analogy, and an equation of motion is derived that explicitly accounts for the time delays arising from the pressure pumps and the feedback loop. Computational fluid dynamics simulations and experiments consistently show that the measured optimal lead angle closely matches the theoretical prediction of 90° across a range of system parameters, including tubing length. The proposed framework simplifies the integration of loop-based droplet manipulation into existing lab-on-a-chip devices and provides a practical design guide for controlling such systems.
Microflows like Knudsen pumps often include rarefied gases featuring different degrees of rarefaction. This different modeling complexity requires space- and time-adaptive rarefied gas models that resolve the physical effects efficiently in each subdomain of the microflow. Different-order moment models are effective at describing rarefied gases in microflows with respective levels of complexity in each subdomain. In this work, analytical model-error estimators for the Hyperbolic Moment Equations (HME) model are derived that are then used to define a space- and time-adaptive moment model for microflows of rarefied gases. First, a domain decomposition strategy of the microflow into subdomains, each modelled by an HME model of an appropriate order, is presented, using domain decomposition criteria that are based on the exact model difference between a higher-order HME model and a lower-order HME model and on chosen error thresholds. Secondly, a non-linear adaptation of a recently developed padded buffer cell approach is presented to couple these varying-order HME models using a single finite volume scheme. Finally, a smoothing of the domain decomposition is proposed to limit oscillations generated by the coupling. While the performance depends on the thresholds and the smoothing parameter, the proposed adaptive moment model is able to capture the different degrees of rarefaction of the rarefied gas in the microflow and yields accurate numerical results compared to experimentally validated DVM benchmark data while obtaining computational speedups of up to 40 percent compared to using a high-order HME model in the entire domain.
Precise control of shear stress is essential for maintaining β -cell physiology in microfluidic perifusion assays, where transient pressure fluctuations can alter glucose-stimulated responses. In this work, we present a boundary-layer-driven microTesla ( μ Tesla) pump engineered to deliver stable, non-pulsatile flow for shear-sensitive biological studies. Using combined computational fluid dynamics and experimental validation, we examined how surface texturization, slip length, and fluid viscosity collectively influence boundary-layer development, shear distribution, and pump output. The μ Tesla II pump produced laminar, viscosity-adaptive flow across Newtonian and non-Newtonian regimes, maintaining consistent outlet velocity even under high-viscosity conditions. Surface textures modulated slip length to enhance viscous coupling, yielding smooth acceleration and reduced startup surges relative to conventional syringe and peristaltic pumps. When applied to β -cell perifusion, μ Tesla II minimized flow-induced artifacts, preserving physiological calcium and ATP dynamics during glucose stimulation. Overall, these findings establish the μ Tesla II as a low-shear perifusion platform that reduces flow-induced artifacts, improves the reproducibility of short-term flow-based cellular assays, and supports the advancement of organ-on-chip and diabetes research models.
Chronic kidney disease (CKD) is a progressive disorder that requires early diagnosis and continuous monitoring to prevent irreversible renal damage. Clinically significant indicators of kidney function and disease progression include urinary biomarkers, especially albumin and creatinine. Nevertheless, traditional analytical techniques for their determination are frequently laborious, necessitate centralized laboratory infrastructure, and are not ideal for quick or point-of-care (POC) examination. To address these limitations, this study presents the design and fabrication of a novel three-dimensional (3D) microfluidic device for the simultaneous detection of albumin protein and creatinine in human urine samples from CKD patients. In order to improve fluid handling, mass transfer, and analytical efficiency while reducing sample and reagent usage, the suggested 3D microfluidic design incorporates intricate channel designs. The device reduces operating complexity and analysis time by enabling multiplexed biochemical analysis inside a single compact system. The synthesis of silver nanoparticles AgNPs@BS and copper nanoparticles CuNPs@BS was achieved using Boswellia sacra (BS) as a dual-function phytochemical platform, acting concurrently as both a reducing and stabilizing (capping) agent. The detection process is then based on the aggregation of AgNPs@BS at pH = 11 induced by creatinine and the aggregation of CuNPs@BS induced by albumin, resulting in a distinct and visually observable colour shift. With a regression value (R2) of 0.9837 and a limit of detection (LOD) of 4.52 µg/mL, the AgNPs@BS sensor demonstrated a linear relationship between the A670/A403 extinction ratio and creatinine concentration range of 20 to 150 µg/mL. Additionally, the CuNPs@BS-based colorimetric assay exhibited a linear relationship between color intensity and albumin concentration over the range of 5–150 µg/mL, with LOD of 3.87 µg/mL. Overall, the developed 3D microfluidic device offers a promising approach for integrated urinary biomarker analysis and represents a step toward portable and POC diagnostic systems for chronic kidney disease monitoring.
The dynamic evolution of gas–solid–liquid interfaces exert a critical influence on reactive transport in subsurface engineering, including geological CO2 sequestration and unconventional energy development. However, the microscale interfacial processes by which reaction-induced bubbles regulate mineral dissolution remain poorly understood, owing to the challenge on visualizing in the opaque natural rock. In this work, emerging mineral microfluidic platform was employed to directly visualize and quantify the coupling between CO2 bubble generation and single calcite dissolution. Three distinct dissolution regimes were identified depending on flow rate and acid concentration, namely bubble-free, bubble-shielding, and bubble-sealing dissolution. At low acid concentration, dissolution proceeds smoothly without bubble formation, and the effective reaction rate increases with flow rate following a power-law trend. Once a critical acid concentration is exceeded, CO2 bubbles nucleate and grow on the reactive surface, sharply reducing the effective reactive area. Acid transport becomes confined to narrow channels surrounding the bubbles, suggesting a progressive shift from convection-dominated to diffusion-dominated mass transfer. Bubble shielding significantly decreases the effective reaction rate by 81.5
Non-poly(A) messenger ribonucleic acid (mRNA) purification remains a critical bottleneck for cell-free translation platforms like transcription-translation coupled with the association of puromycin linker (TRAP) display for peptide screening, where sub-optimal thermal release can compromise results. Herein, we demonstrate a “complex system response” approach for optimizing thermal elution from probe-coated magnetic beads. Using an orthogonal array composite design with two optimization rounds (17 three-factor + 9 two-factor experiments), we systematically explored temperature (60–95 °C), time (5–30 min), pH (7–9), and their interactions, revealing an unanticipated optimum of 60 °C for 23.5 min at pH 9 (vs. conventional protocols of 90 °C for 10 min). These conditions were associated with a maximum release efficiency of 71 ± 7
The influence of the flow rate ratio on plug hydrodynamics is investigated experimentally in a rectangular T-junction microchannel under different conditions, where water plugs are dispersed in a silicone oil carrier phase. The study examines plug length, generation frequency, velocity, and shape. The results show that the resultant forces at the junction, characterised by the two-phase capillary number CaTP, predominantly govern all plug characteristics. Under constant CaTP, the plug length increases linearly with the flow rate ratio, while the generation frequency remains within the range of 41–55 Hz. Plug velocity is primarily controlled by the liquid film thickness, which varies with the flow rate ratio, whereas plug shape is only weakly influenced by this parameter. These findings enhance understanding of two-phase plug flow in microchannels and provide valuable insights for controlling plug flow regimes in microelectronic cooling applications.
This paper examines deformation and transport behavior of compound droplets when transported through microchannels with different geometries, including sinusoidal, sawtooth, diffuser, nozzle obstructions. The effect of capillary number (varying between 0.025 and 1) and obstruction size (0.4 Ro -0.8 Ro) on transition behaviour of compound droplets is studied systematically. The results indicate that capillary number is significant factor in deformation and stability of droplets. One of the values tested gave reliable outcomes with capillary number of 0.1 and even more accurate with 0.025. This implies that lower capillary numbers result in more accurate simulations, as they are precise in capturing interface dynamics and droplet-obstruction interactions. Transit time of outer and inner droplets decreases with obstruction ratio of 0.4–0.8, which means that obstructions of higher size facilitate faster droplet passage. This suggests that a lower capillary number enhances the precision of simulations by better capturing interface dynamics and droplet–obstruction interactions. The transit time for both outer and inner droplets decreases with increasing obstruction ratio from 0.4 to 0.8, indicating that larger obstructions facilitate faster droplet passage.
The equivalent circuit design method has evolved into a relatively mature technique for designing microfluidic devices. This approach simplifies microfluidic device design by approximating fluid behavior within the devices as analogous to electron behavior in electrical circuits, thereby leveraging the analogy between microfluidic hydraulics and electrical circuits. Based on this technique, scientists have developed microfluidic devices with various functionalities, such as multi-concentration mixing, cell capture, flow distribution, concentration gradient generation, and droplet generation. However, due to the disparity in physical properties between microfluidic systems and electrical circuits, the impedance effect experienced by fluids during the analogical design process is not pronounced—specifically, when mapping fluidic characteristics like high and low Reynolds numbers to circuit properties such as high and low current frequencies. This gap has resulted in a lack of microchannel design methods analogous to inductive components in electrical circuits when applying the equivalent circuit approach to microfluidic chip design. Here, we propose a design strategy for microfluidic valves with inductive properties and their topological optimization. The valve structures designed by this method exhibit high flow resistance under high Reynolds number conditions and low resistance under low Reynolds number conditions. Consequently, this work enriches the equivalent circuit design methodology by introducing the equivalent inductive microfluidic module—Microfluidic inductor as a new available module, along with its corresponding design method.
This work presents a physics-guided design, optimization, fabrication, and experimental validation of Fresnel-ring-based single-transducer acoustic tweezers capable of precise three-dimensional manipulation of microparticles in liquid without reliance on (or being hampered by) standing waves. A planar piezoelectric transducer patterned with concentric half-wavelength annular electrodes is engineered to generate three axially separated focal points whose constructive interference forms a spatially enclosed radiation-force potential well. By systematically optimizing the inter-focal spacing using Finite Element Modeling (FEM), we establish quantitative design rules that link the primary focal length, inter-focal distance, trapping-zone geometry, and particle-size selectivity. Hydrophone-based three-dimensional pressure mapping validates the simulated acoustic field and confirms the formation of a well-defined low-pressure region bounded by surrounding high-pressure lobes. Experimental results demonstrate robust, size-selective confinement of microspheres and stable trapping of large biological specimens, including a 700 μm zebrafish embryo, at user-defined axial distances from the transducer surface. The observed trapping behavior agrees with theoretical predictions based on acoustic radiation force scaling and streaming-induced drag. By explicitly correlating focal configuration, radiation potential landscape, and particle-size-dependent stability, this study establishes a scalable, microfabrication-compatible framework for single-transducer acoustic tweezers. The technology based on multi-focal acoustic tweezers enables programmable trapping size, trapping position, selective manipulation, and controlled particle lifting and ejection, providing a versatile platform for advanced BioMEMS applications and next-generation contactless micromanipulation systems.
