Liquid-nitrogen (LN2) based cryogenic cooling is widely employed in bone cryosurgery. However, conventional static immersion is limited by vapor-film formation and conduction-dominated heat transfer within the bone, leading to slow and non-uniform cooling. In this study, a convection-assisted LN2 cryogenic system is investigated to enhance heat-transfer performance during bone freezing. A coupled multiphysics framework integrating phase-field modeling for LN2-air displacement, incompressible Navier-Stokes equations, and transient heat conduction is developed to simulate static, continuous-flow, and intermittent-flow cooling modes. Numerical predictions are validated against ex-vivo cattle bone experiments, demonstrating good agreement in cooling trends and thermal penetration depth. Results show that forced LN2 convection significantly accelerates early-time cooling, producing an approximately two-fold enhancement in the initial cooling rate compared to static immersion, while higher flow rates exhibit diminishing thermal returns due to internal conduction resistance within the bone. A dimensionless analysis based on the Peclet number and a normalized Biot number reveals a transition from convection-assisted surface cooling to conduction-limited internal heat transfer, which governs the performance of both continuous and intermittent cooling modes. Intermittent LN2 delivery strategies are shown to exploit this regime transition effectively, achieving up to similar to 70% reduction in cryogen consumption while maintaining comparable minimum temperatures. Patient-specific simulations further confirm that flow-assisted cooling enhances thermal uniformity and penetration in anatomically realistic geometries. The results establish a generalized heat-transfer framework for cryogenic cooling of low-diffusivity solids and provide quantitative guidance for optimizing convection-assisted LN2 cooling systems.

This study presents a combined numerical and experimental investigation of I-typeparallel microchannel heat sinks to evaluate the influence of hydraulic diameter, aspectratio, and fin width on thermo-hydraulic performance under constant mass flow rateand heat flux. A surrogate-based optimisation framework integrating Artificial NeuralNetworks (ANN) and Genetic Algorithms (GA) was employed to identify the optimumgeometry. The optimised configuration (Dh = 0.579 mm, α = 1.79, and Wfin =0.601 mm) achieved a 52.3% reduction in pressure drop, reduced the wall superheat(Tmax-Tin) from 32.2 K to 28.4 K (12% reduction), and improved the temperature nonuniformityindex ψ by 7.4% relative to the baseline case. Experimental validation usingIR thermography and thermocouples with water and acetone showed good agreementwith CFD predictions. In addition, system-level friction factor and Nusselt numbercorrelations accounting for manifold effects were developed. The proposed frameworkprovides an effective approach for designing high-performance microchannel heat sinksfor advanced electronics cooling applications.

Microfluidics has emerged as an innovative approach for tuneable and controlled synthesis of nanoparticles, offering distinct advantages over conventional methods. Precise control of the physicochemical properties of nanoparticles is essential for optimising their therapeutic performance. In this study, we address the analytical challenge of predicting Poly (lactic-co-glycolic acid) (PLGA) nanoparticle size during synthesis by investigating whether mixing time (τmix), measured in-line and in real time within a hydrodynamic flow focusing (HFF) microfluidic platform, can serve as a predictor of nanoparticle size. Because nanoparticle size is conventionally assessed only offline after synthesis is complete, there is currently no methodology for real-time size prediction during microfluidic manufacturing. To evaluate mixing time as an analytical predictor, flow rate ratio (FRR) and total flow rate (TFR) were systematically varied to generate a broad range of mixing conditions. While FRR is widely regarded as the dominant parameter controlling nanoprecipitation, the influence of TFR on mixing dynamics and nanoparticle formation remains less understood. Mixing time was quantified in-line using a novel flow-visualisation method, and the observed trends were subsequently validated using computational fluid dynamics (CFD). Results from flow visualisation and CFD showed that τmix is strongly dependent on FRR but that, despite being neglected from most theoretical models, TFR also has a measurable effect on τmix. This effect was further demonstrated through PLGA NPs synthesis under 16 different flow conditions. As expected, decreasing FRR reduced mixing time and led to the formation of smaller nanoparticles. However, increasing TFR also reduced mixing time and consequently decreased nanoparticle size. For example, at a constant FRR of 0.05, increasing TFR from 250 μL/min to 1000 μL/min reduced PLGA NP size from 156.6 ± 2.56 nm to 128.9 ± 2.53 nm. Together, these results establish τmix as a real-time, in-line measurable parameter that correlates strongly with offline-determined PLGA nanoparticle size. In this analytical framework, nanoparticle size is treated as the target quality attribute, while mixing time, derived from in-line flow visualisation, is evaluated as a measurable surrogate parameter for predicting that attribute during microfluidic nanoprecipitation.

Microfluidics has emerged as an innovative approach for tuneable and controlled formulation of nanoparticles, offering distinct advantages over conventional methods. Precise control of the physicochemical properties of nanoparticles is essential for optimising their therapeutic performance. In this study, we systematically investigate the influence of flow rate ratio (FRR) and total flow rate (TFR) on the formulation of Poly(lactic-co-glycolic acid) (PLGA) nanoparticles using a continuous hydrodynamic flow focusing (HFF) microfluidic platform and benchmark the outcomes against a conventional batch process. Mixing times under different flow conditions were quantified by computational fluid dynamics (CFD) analysis and validated experimentally to establish mechanistic links between microchannel mixing and the resulting nanoparticle properties. Our results show that increasing TFR and decreasing FRR shorten the mixing time, leading to the formation of smaller nanoparticles. PLGA NPs size decreased from 156.6 ± 2.56 nm to 128.9 ± 2.53 nm with an increase in TFR from 250 mL/min to 1000 mL/min while keeping the FRR constant at 0.05. While by increasing the FRR from 0.05 to 0.075 at a constant TFR of 250 uL/min PLGA NPs size increases from 156.6 ± 2.56 nm to 168.4 ± 2.46nm. This work highlights the critical role of flow parameters in governing nanoprecipitation kinetics and establishes a tuneable platform for the scalable production of PLGA nanoparticles with well-defined physicochemical properties for biomedical applications.

Efficient plasma separation from whole blood is an essential preprocessing step in clinical diagnostics, as plasma is preferred for most colorimetric and biochemical tests over whole blood. Traditional separation methods, such as centrifugation, need laboratory facilities and trained personnel, making them unsuitable for point-of-care (POC) diagnostics. Although paper-based microfluidic devices emerged as a promising solution to the above limitations, most existing paper-based devices rely on filtration membranes, chemical functionalization, or both, increasing device complexity and cost. In the present study, membrane- and reagent-free paper-based microfluidic devices have been designed and developed that achieve blood-plasma separation solely through geometric control of the wicking pathway, eliminating the requirement of any external separation aids. Two devices were designed and fabricated embedding constriction-expansion flow path, with and without a localized hydrophobic barrier, to selectively trap red blood cells while allowing plasma to wick through the porous cellulose network. Experimental results demonstrate consistent plasma separation, with a maximum plasma separation efficiency of ∼64% for the optimized design. Further, to elucidate the underlying mechanism, an analytical formulation for capillary transport considering non-Newtonian blood rheology coupled with two-phase numerical simulations was developed. Finally, the reliability of the proposed devices was further validated through protein (albumin) analysis of the separated plasma, which closely matched the clinical laboratory measurements. The present work establishes a simple, low-cost, equipment- and chemical-free strategy for effective blood-plasma separation achieved solely through structural design, while preserving analyte integrity, offering a viable solution for POC diagnostic applications.

Electrolysis-generated hydrogen bubble plumes are influenced by bubble departure, coalescence, diffusive growth, and plume-induced hydrodynamics. Imposing an external acoustic field can alter the evolution and transport of bubbles. The interaction between bubbles and a 200 kHz ultrasonic field was experimentally studied for a power range of 0-9 W. High-speed imaging was used to determine bubble diameter and translational velocity at multiple vertical locations above the cathode. In the absence of ultrasound, the bubble diameters (25-120 mu m) follow a log-normal distribution near the electrode surface. Under ultrasonic excitation, the maximum bubble size reaches 400 mu m at 9 W, and the distribution becomes near-normal at 20 mm from the cathode. The cumulative size distributions show an increase in the lower decile up to the mid-plane of the ultrasonic transducer, followed by a decrease at higher elevations. These trends are interpreted in terms of bubble coalescence, rectified diffusion, and Bjerknes forces due to the acoustic pressure field. A power-law relationship is observed between the bubble velocity and the bubble diameter. The results indicate that ultrasound modifies the post-detachment bubble dynamics and not nucleation at the electrode. The findings are expected to help develop multiphase flow models of bubble plumes under acoustic forcing.

Accurate reconstruction of local surface temperature and heat flux fields is essential for characterizing thermal non-uniformity in microchannel heat sinks. Infrared thermography provides full-field foil-side temperature measurements, but the channel-side surface temperature and local heat flux are not directly accessible. Conventional reconstruction requires pixel-wise energy-balance calculations and three-dimensional conduction simulations, which become computationally expensive for large parametric datasets. In this work, a residual-context RED-Net framework is developed to reconstruct these hidden thermal fields directly from measured foil-temperature maps. The input is a seven-channel representation comprising normalized foil temperature, spatial-coordinate maps, and global foil-temperature statistics. Two task-specific models are trained independently using the same residual encoder–decoder architecture: one for channel-side surface temperature and another for local heat flux. For the retained evaluation cases, the surface-temperature model achieves mean absolute errors of 0.060°C and 0.063°C for the flow-rate and heat-flux-variation groups, respectively. For the three retained geometry-variation test cases, the corresponding mean absolute error is 0.118°C. A retained additional-geometry test case gives a mean absolute error of 0.992°C. The heat-flux model achieves mean absolute errors of 0.032W/cm2 and 0.0098W/cm2 for flow-rate and heat-flux-variation groups, respectively. RED-Net reduces repeated reconstruction from several hours to less than one second per case and provides a low-error surrogate for operating-condition reconstruction. Selected geometry-transfer results are promising, whereas broader extrapolation requires explicit geometry-aware descriptors.

This experimental study investigates the hydrodynamic and thermal phenomena in the concurrently hydrodynamically and thermally developing phase of laminar, transitional, and turbulent mixed convection flows of water in a vertical tube with buoyancy effects. A mixed convection experimental test-rig was designed, built, and validated for the experimentation. Experiments were conducted for Reynolds number (Re) varying from 494 to 15053, Grashof number (Gr) from 1.44 x 104 to 1.35 x 107,Prandtl number (Pr) from 3.1 to 7.1, and Richardson number (Ri) from 0 to 1.5 with a tube length-to-diameter ratio of 130, undergoing constant wall heat flux. A facility for flow visualization is also made following the test section for identifying the onset of transition. The flow visualization indicates an earlier transition in buoyancy-opposed flow in comparison to the buoyancy-aided flow. The effects of varying the Gr , Re , at fixed Ri on friction factor (f) and Nusselt number (Nu) were investigated for both the flows. Experimental results showed that at fixed Ri with the increment in Re , the average f gradually reduces in the laminar region, starts increasing in the transitional phase, and further reduces in the turbulent regime. In contrast, the average Nu increases in the laminar region, remains unchanged in the transitional phase, and then increases with the increase in Re in the turbulent region. Further, in the transitional phase, with the increase in Ri , the average f lowers, and the average Nu enhances in buoyancy-aided and opposed flows, respectively. It has also been observed that the transition is delayed with the increase in Ri in both flows and shortens the span of the transitional regime. The transition onset occurs earlier in the opposing flow contrary to the aiding flow for the same Re and Gr .

Flow-electrode capacitive deionization (FCDI) is an innovative approach for removing charged ions from untreated water, utilizing the interaction between ions and flow carbon electrodes. A review of recent publications on FCDI reveals a predominant focus on salt removal from water (desalination) and electro-sorption processes. Though desalination is just one step in improving the water quality, it is worthwhile looking at the research in the context of FCDI techniques that involve other water treatment methods. This paper offers a detailed review of recent literature on FCDI applications in wastewater treatment. Given the broad scope of wastewater treatment, the specific areas where FCDI shows promise, including removal of heavy metal and radioactive elements, organic micropollutant elimination, halogen removal, and resource recovery, are addressed. Additionally, we assess the current research landscape and propose potential future directions in this evolving field.

In practical heat exchangers, where tube lengths are typically shorter, the flow is largely developed. In this numerical investigation, we examined the thermal and hydrodynamic characteristics within the transitional developing regime of mixed convection for buoyancy-aiding and opposing flows through a vertical tube. Employing a two-dimensional axially symmetric approach with a steady state, simulations were conducted for a range of Reynolds number (2000 <= Re <= 5000 ), Grashof number ( 4 x 10(5) <= Gr <= 2.5 x 10(6) ), and Richardson number (Ri) of 0.1 for a nondimensional length (L/D) of 150 subjected to uniform wall heat flux condition. The study involved a comparative analysis of two transition models: transition kappa - kappa(l) - omega and transition shear stress transport (SST) and the model with better accuracy was selected. The findings reveal the crucial role of buoyancy in the laminar-turbulent transition for both assisting and opposing flows. In both flow scenarios, the coefficient of skin friction (C-f) and Nusselt number (Nu) exhibit an increase with increasing Re at a constant Ri. Notably, under the same Ri conditions, the opposing flow exhibits higher pressure drop (assessed by C-f) and heat transfer (assessed by Nu). For buoyancy-aiding and opposing flows, the hydrodynamic entry length reduces from L/D similar to 70 to 60, while the thermal entry length extends with increasing Gr and Re, when Ri is fixed. Furthermore, the onset of transition in mixed convection is dependent on the wall heat flux and is earlier in opposed flow than assisted flow. Based on the intermittent behavior, the lower transitional zone (2000 < Re < 3000) and the higher transition zone (3000 < Re < 5000) have been demonstrated.

In this study, saturated pool boiling experiments were conducted on copper minichannel and flat surfaces at atmospheric pressure using water as the working fluid. The heat transfer performance was assessed through point measurements with a heater block, cartridge heaters, and thermocouples, as well as field measurements using a thin-foil heater and infrared thermography. Two copper minichannel surfaces with square cross sections of 1 mm (minichannel-1) and 2 mm (minichannel-2) side lengths were tested and compared to a flat surface. Minichannel-1 and minichannel-2 enhanced the critical heat flux (CHF) by 17% and 45%, respectively, and improved the heat transfer coefficient by 24-40% and 51-75%, respectively, compared to the flat surface. Minichannel-2 exhibited the lowest and most uniform boiling surface temperature, making it the best performer among the three. There was no significant change in departure frequency among the surfaces, and no significant change in departure diameter for the flat surface and minichannel-1. However, minichannel-2 had lower departure diameters due to its deeper channels, which prevented bubble coalescence and maintained low departure diameters. Additionally, minichannel-2 delayed vapor film formation by breaking it with its deeper fins, thereby improving CHF and slightly enhancing bubble dynamics. The enhancement in boiling heat transfer is primarily attributed to the increased surface area provided by the minichannels, with a minor contribution from improved bubble dynamics. However, the dominant factor in enhancing pool boiling heat transfer on minichannel surfaces is the increase in surface area.

This study explores the potential of hexagonal boron nitride (h-BN) h-BN) nanoparticles suspended in Deep Eutectic Solvent (DES) as a thermal medium or coolant. The eutectic point of the DES, which comprises dibenzyl ether and diphenyl ether as HBA and HBD, respectively, is predicted using the COnductor-like S creening MO del- S egment A ctivity C oefficient (COSMO-SAC) thermodynamic model. The Nuclear Magnetic Resonance (NMR) spectroscopy technique is used to evaluate the hydrogen interaction of the DES. The nanofluids are synthesized at five different concentrations of h-BN nanoparticles: 0.01-0.12 wt%. The uniform dispersion of the nanoparticles in the bulk media is investigated by zeta potential stability analysis. The basic physical-chemical properties, namely thermal stability and freezing point, of the DES are further measured. The experimental evaluation of nanofluids comprises studying the effective thermophysical characteristics. The results demonstrate that the average thermal conductivity increase for Nanoparticles-Enhanced DES (NEDES2) is 7.4%. The specific heat capacity (average) of NEDES5 nanofluid increased by 33%. For a small quantity of nanoparticle dispersion, the viscosity of nanofluids increases somewhat relative to the base fluid and is temperature-dependent. While the density decreases with the addition of nanoparticles. Precise predictions of thermophysical properties are essential for addressing and improving a wide range of heat and mass transfer problems, especially in the field of engineering and associated domains. This is especially important for complex fluids, including nanofluids, which have a crucial impact on advancing new technologies. The importance of thermal performance parameters in heat transfer research is apparent. However, there is a noticeable lack of dependable thermophysical properties for nanofluids that can be consistently used in various issues and sophisticated industrial systems. A machine learning technique is used to forecast the thermophysical characteristics of nanofluids. For several models, the deviation between nanofluid's predicted and experimental thermophysical characteristics is less than 6%. Apart from thermal conductivity, all other nanofluid parameters vary linearly with temperature, with R2 2 values above 0.99 and RMSE less than 0.0332.

Polymer electrolyte membrane (PEM) fuel cell directly converts the chemical energy of the fuel into electrical energy thus providing higher energy conversion efficiency. Liquid water causes flooding in the Gas Diffusion Layer, Catalyst Layer as well as in channels due to which the active catalyst surface area in the CL reduces and hence the reaction rate reduces. Phase transfer between liquid water and water vapour along with electro-osmotic drag and reverse diffusion through the membrane, are the factors influencing liquid water buildup in the domains. In the present work a three dimensional, non-isothermal, two phase numerical model incorporating cathode and anode domains for studying water distribution is developed. Water formation and distribution across the domain are studied using the mixed flow distributor. The developed numerical model is validated by comparing the polarization curve with the experimental results. The study shows that the ohmic region in the polarization curve increases and the concentration losses decrease due to the use of pure oxygen at inlet which can be important for applications such as in spacecrafts. The effects of liquid water accumulation in the domain on the reactant transport, current density and temperature distribution are also investigated.

Artificial superhydrophobic and self-cleaning surfaces are desirable in many engineering applications. Lotus leaves have long been the benchmark for the design and fabrication of artificial non-wetting surfaces. Here, we report the design and fabrication of superhydrophobic surfaces that mimic the behavior of lotus leaves. Akin to the microstructure of lotus leaves, an intrinsically hierarchical microstructure is created using carbon nanotubes (CNTs). The conventional CNT growth protocol is modified to induce multiscale features with lower diameter CNTs on the top of thicker CNTs. Together they form a dandelion seed head type structure with thicker CNTs such as “beak” and thinner CNTs such as “pappus.” The wetting and self-cleaning behavior of the CNT coated surface is compared with the lotus leaves. The wetting behavior of never-wet commercial spray and Cu and Ni foils are also recorded for comparison. The contact angle, contact angle hysteresis, and sliding angle of water drops on CNT coated surface are comparable with the lotus leaves. The wobbling motion of water drops on the CNT coated surface is similar to that on the lotus leaves and spray coated surface with varying contact line length with time. It also induces the self-cleaning characteristics of CNT coated surfaces similar to lotus leaves. Finally, the present study demonstrates a feasible strategy to design and fabricate lotus leaves like artificial superhydrophobic surfaces with hierarchical CNT structures.

The performance of a PEM fuel cell that uses hydrogen as the fuel and pure oxygen as the oxidant strongly depends on water management, which has been primarily studied in a single-channel domain. Therefore, there is a need to examine water distribution throughout the entire fuel cell domain, including both the anode and cathode sides. Liquid water can cause flooding in the gas diffusion layer, catalyst layer, and channels, reducing the active surface area of the catalyst and, consequently, the reaction rate. Phase transfer between liquid water and water vapor influences the buildup of liquid water in these domains. In the present work, a three-dimensional, non-isothermal, two-phase numerical model incorporating both the cathode and anode domains has been developed to study water distribution. This model includes water phase transition in the gas diffusion layer, catalyst layer, and channels. The mixed flow distributor is used to analyze water formation and distribution throughout the domain. The study shows that using pure oxygen at the inlet increases the ohmic region in the polarization curve and decreases concentration losses, which could be important for applications such as spacecraft. Additionally, the effects of liquid water accumulation in the porous layers on reactant transport and cell performance are investigated.

The study of flow behavior in the simultaneously developing transitional regime of mixed convection flows is rare. It has been believed that the transitional regime will give a good compromise between pressure drop and heat transfer compared to laminar and turbulent flow regime. In this experimental study, the friction factor (f) and Nusselt number (Nu) characteristics for buoyancy-assisted and opposed flows of water in concurrently developing transitional regime of mixed convection through a vertical tube are studied. Experiments were done for Reynolds numbers (Re) varying from 500 to 15,000, Grashof numbers (Gr) from 1.25 x 10(4) to 5 x 10(6), Prandtl numbers (Pr) from 3 to 7, and Richardson numbers (Ri) from 0 to 0.1 subjected to uniform heat flux boundary conditions. A flow visualization provision after the test section which confirms an early transition in buoyancy-opposing flow (Rec = 2264) compared to buoyancy-aiding flow (Rec = 2468) at a fixed Ri of 0.1. Further, with the increase in Ri from 0 to 0.1, the average f decreases, and the average Nu increases in both aiding and opposing flows. It confirms that the onset of transition gets delayed with the increase of heat flux supplied in both the flows. Based on the present outcomes, an efficient heat exchanging device can be operated either to delay or advance the transition in a vertical pipe flow for optimum heat transfer.

Supplementary Figures 1-12, Table 1, Methods from Computationally Guided Photothermal Tumor Therapy Using Long-Circulating Gold Nanorod Antennas