This study investigates bubble formation, pressure drop (flow rate), and wall temperature variation during fluid heating in a lattice-shaped microchannel under constant heat flux conditions. The microchannel was designed with periodically arranged pillars to trap bubbles in specific regions, enabling simplified and controlled observation of bubble behavior and its influence on neighboring bubble formation. Imaging and infrared thermography were used to visualize bubble dynamics and measure wall temperature distributions. The fluid was supplied under constant driving pressure, and the flow rate during bubble formation was recorded. To complement the experiments, a numerical simulation was performed by coupling two-dimensional flow analysis with a probabilistic element-filling model. Two bubble formation models were considered: a random model, in which bubbles form independently, and a neighbor-driven model, in which the filling probability increases near already filled regions, mimicking bubble propagation. The neighbor-driven model produced clustered bubble patterns and wider flow passages, resulting in lower pressure drops compared to the random case. An analytical model for pressure drop was developed based on the simulation results, incorporating both random and propagative bubble formation mechanisms. This model was applied to the experimental data and showed good agreement with the measured relationship between unfilled ratio and flow rate. Temperature measurement showed that the effect of heat transfer to the bubble propagation was relatively small in this study. The findings underscore the importance of accounting for both stochastic and deterministic effects in bubble formation within microchannels.

Direct numerical simulations (DNS) are conducted to investigate the effects of spanwise domain size on stably stratified turbulent shear layers. The focus is on the formation and spatial organization of elongated large-scale structures (ELSS), which emerge following the transition from Kelvin–Helmholtz instability and characterized by streamwise extents far exceeding the shear layer’s thickness. Simulations are conducted for a temporally developing shear layer under stable density stratification. The spanwise extent is varied, while the streamwise and vertical domain sizes are fixed. Flow visualizations, one-point statistics, energy spectra, and two-point correlation functions are used to assess the influence of spanwise confinement on the transition process and late-time turbulence characteristics. The results show that when the spanwise domain size is very small, the transition process is altered and ELSS fail to develop properly. For intermediate domain sizes, the streamwise elongation of ELSS is captured, but their meandering and spatial repetition are suppressed. Statistical analysis reveals that while the meandering of ELSS contributes to large-scale structure, the presence of multiple alternating ELSS in the spanwise direction is more critical to the overall flow statistics. These findings emphasize the importance of spanwise configurations of ELSS in the dynamics and energetics of stably stratified shear layers.

This study investigates the puffing-induced secondary atomization and evaporation behavior of a bi-component droplet composed of miscible species (n-heptane and n-hexadecane) using three-dimensional numerical simulations. The gas–liquid interface is captured using the coupled level-set and volume of fluid (CLSVOF) method, and evaporation of two components is modeled based on a non-ideal vapor–liquid equilibrium (VLE) formulation. The simulation consists of two stages: a stationary stage to establish initial thermal and concentration fields assuming a quiescent droplet, and a puffing stage where a vapor bubble of n-heptane is initially embedded within the droplet. The results qualitatively reproduce the fundamental dynamics of puffing droplets, including bubble growth, burst, vapor ejection, ligament formation, and its fragmentation as reported in previous experimental and numerical studies. The analysis reveals a strong dependence of the evaporation behavior on the embedded bubble positioning. While n-heptane shows a sharp decline in evaporation rate after bubble bursts, evaporation rate of n-hexadecane exhibits a transient enhancement owing to rapid redistribution over the droplet surface. Furthermore, placing the bubble deeper inside the droplet leads to stronger puffing and sustains fuel, especially n-heptane, evaporation, highlighting the critical role of internal distribution of species and bubble positioning in bi-component droplet evaporation under puffing conditions.

We introduce a deep-learning framework that turns a single moving camera into a reliable 3-D sensor for thermal imaging. Portable or drone-mounted inspections increasingly demand lightweight depth acquisition, and passive light field imaging is attractive because it dispenses with active hardware such as LiDAR. Unfortunately, conventional light field cameras rely on large-scale lens arrays, and simply sweeping a monocular camera in their place introduces subtle inter-view vertical offsets that quickly erode reconstruction accuracy: in our simulation, shifting five of nine views increased the Root-Mean-Square Error (RMSE) by a factor of 2.9. To counter this effect we adopt a two-stage strategy in which a Residual U-Net (ResUNet) first rectifies the displaced images and the resulting aligned sequence is then processed by EPINET to estimate depth. In the initial experiments with a synthetic RGB dataset, the rectifier reduced RMSE by 14.2% on average (up to 33.8% in the best case) and increased Peak Signal-to-Noise Ratio (PSNR) by 0.3 dB. These gains were accompanied by the disappearance of streak-like artifacts and the recovery of clean linear structures in epipolar plane images. Subsequently, in the thermal image simulation, the proposed method consistently improved depth estimation across all relative-depth metrics, while also mitigating systematic errors in isothermal regions and enhancing the clarity of temperature boundaries. Our study is, to the best of our knowledge, the first to combine learned rectification with light field depth inference for calibration-free, single-camera thermal diagnostics. Source code and trained models are available at https://github.com/ryutaroLF/ResUNet_EPI_Stabiliser/ .

In this study, the diffusion coefficients of carbon dioxide (CO₂) in ionic liquids were experimentally measured using the phase-shifting interferometer. The temperature dependence of the diffusion coefficient from 298 K to 323 K and the pressure dependence in the range of 0.2 MPa-CO₂ to 1.0 MPa-CO₂ were evaluated. Ionic liquids are expected to replace conventional amine solutions as absorbents in CO₂ separation and capture processes because of their non-volatility and excellent gas absorption performance. Understanding the diffusivity of CO₂ into ionic liquids at the gas-liquid interface is essential to realize a novel and efficient CO₂ separation and capture process. In this study, the phase-shifting interferometer was applied to precisely observe the CO₂ diffusion process in the vicinity of the gas–liquid interface in a short duration. As absorbents, l-butyl-3-methylimidazolium tetrafluoroborate ([bmim][BF₄]) and 1-butyl-3-methylimidazolium bis(trifluoromethylsulfonyl)imide ([bmim][Tf₂N]) were used to evaluate the effect of anions. The diffusion coefficient of CO₂ into ionic liquids was experimentally evaluated from transient concentration distributions. As results, the measured diffusion coefficients were on the order of 10⁻¹⁰ m²/s for both the temperature and pressure dependency experiments. The diffusion coefficient of CO₂ in both the ionic liquids was largely constant within the pressure range applied in the experiment, and no strong pressure dependence was observed. In contrast, the diffusion coefficient of [bmim][Tf₂N] showed a strong temperature dependence, increasing monotonically with increasing temperature, in the applied temperature range, which is typically employed in absorption towers.

Understanding droplet penetration and evaporation in inkjet printing is essential for evaluating the energy requirements during drying and fixation, which are reflected in the droplet lifetime. This study investigates model ink droplets composed of water, propylene glycol, and glycerol deposited on a porous polyimide substrate with pore diameters of 300 nm and 1000 nm. The wetting conditions are either hydrophilic or hydrophobic. An experimental approach was employed by monitoring the geometric evolution of 100 pL droplets on the substrate. Immediately after the droplet impact, all cases exhibited spreading behavior for 5–20 ms. Moving to the next stage, most droplets showed a quasi-pinned contact line. Droplets containing 50 wt% water on hydrophobic surfaces penetrated significantly faster on the smaller pores (300 nm), resulting in a droplet lifetime approximately 30% shorter than on the 1000 nm pores. In contrast, droplets with higher water contents (75 wt% and 95 wt%) exhibited no significant difference in penetration time across hydrophobic surfaces, regardless of pore size. On hydrophilic substrates, all mixtures consistently demonstrated faster penetration. The simultaneous evaporation along the droplet lifetime is significant for higher water content on hydrophobic media. An analytical model based on the surface energy described by Owen-Wendt-Rabel-Kaelble (OWRK) was applied. Contact angle and penetration rate from the Young and Lucas-Washburn (LW) equation are calculated by using OWRK for the cos θ term and compared with the experiment result. The comparison shows Young-OWRK calculation will have an accurate prediction for droplets having the identical size with surface energy measurement (~2 μL), while a disparity is observed for the smaller droplet size (~100 pL) due to having more dynamics from the nozzle ejection. LW-OWRK equation calculation results on much faster penetration compared to the experiment, implying the need to use a more complex LW equation to represent the porous media characteristics.

Methane-ammonia blended combustion has emerged as a promising strategy to reduce carbon emissions while maximizing the utilization of existing fossil-fuel-based burners, serving as a viable intermediate step towards carbon neutrality. In order to effectively implement methane-ammonia combustion strategies in practical combustors, it is crucial to understand the combustion characteristics of premixed methane-ammonia-air flames, particularly focusing on how turbulence and varying ammonia blending ratios influence flame structure and NO formation. Therefore, in the present study, turbulent premixed methane-ammonia-air jet flames at different ammonia blending ratios were investigated using two-dimensional direct numerical simulation (DNS). Local heat release rates, flame front curvature, tangential strain rates, NO production rates, and their correlations were systematically evaluated. Results indicate that, under turbulent flame interactions, flame front curvature has a dominant influence; specifically, flame elements convex toward the burned side exhibit increased heat release rates. These effects are consistently predicted across different ammonia blending ratios when appropriately normalized with their corresponding laminar flame quantities. However, a comparison between stretched laminar flames and local turbulent flame elements reveals that convex flame elements significantly enhance NO formation, which cannot be fully captured by laminar planar flame analyses alone. Therefore, it is essential to incorporate curvature effects for accurately evaluating NOx emissions in turbulent combustion.

2D numerical simulations of a wrinkled lean hydrogen (H₂)-air premixed flame are performed with and without considering the Soret effect to investigate the Soret effect on flame propagation. The equivalence ratio and temperature of unburnt premixed gas are 0.5 and 300 K, respectively, and ambient pressure is 1 atm. Results show that neglecting the Soret effect underestimates the burning velocity and flame surface area by approximately 4% and 5.5%, respectively. Neglecting the Soret diffusion of H₂ overestimates the Lewis number of H₂. This overestimated Lewis number suppresses the growth of the flame front perturbation with a large wavelength and causes the underestimation of the flame surface area. Moreover, neglecting the Soret effect underestimates the total heat release near the maximum of the progress variable. At the flame bulge, the Soret diffusion enhances the transport of H₂ toward the convex flame front, increasing the local mixture fraction. This increase in the mixture fraction means that the local equivalence ratio approaches stoichiometry under the current fuel-lean conditions. Therefore, neglecting the Soret effect results in a leaner local mixture and an underestimation of the reaction progress in this region. Additionally, neglecting the Soret diffusion of H results in the underestimation of H supply to burnt gas side. Thus, chain-branching reaction of H + O₂ = O + OH and the formation of H₂O from intermediate species (H, O, OH) in the burnt gas region at flame bulge are suppressed. Consequently, the heat release in the flame bulge, where the progress variable is near its peak, is also underestimated. The underestimations of flame surface area and heat release result in the underestimation of the burning velocity.

Head-on quenching (HOQ) is one of the canonical configurations of flame-wall interaction (FWI), where a premixed flame propagates at a normal angle to the wall and quenches. However, few studies have investigated the effect of convective flow on the HOQ process, which is relevant to the turbulent FWI in practical combustors. Therefore, the effect of convective flow on the HOQ of a laminar premixed flame was numerically investigated in an inert, isothermal wall-stagnating flow. The magnitude of convection was controlled by the nozzle velocity. Two kinds of fuels, namely n-heptane and a methane/hydrogen mixture, were considered. It was found that the quenching distance and the maximum wall heat flux were influenced by two kinds of effects induced by the flow field. The first effect is the convective effect, which transports the flame closer to the wall with increasing nozzle velocity. The second effect is the flame stretch effect, which strengthens (weakens) the flame for mixtures with the Lewis number Le less (greater) than unity. For mixtures with Le < 1 (e.g., lean methane/hydrogen/air), the two effects cooperate to decrease the quenching distance monotonically with increasing nozzle velocity. Meanwhile, for mixtures with Le > 1 (e.g., n-heptane/air), the two effects compete, and the quenching distance first increases and then decreases with increasing nozzle velocity. Meanwhile, for n-heptane flames, the HOQ process in a stagnation flow is governed by the Damköhler number and the flame power. For methane/hydrogen flames, the maximum wall heat flux did not scale with the flame power. It was implied that the quenching mode may transition away from thermal quenching for mixtures with high hydrogen addition ratios, which requires further investigation.

Use of alternative fuels, such as e-fuel , will be more common in automobile industries. However, diversity in fuel composition influences combustion characteristics of an engine, requiring highly-robust control for stable and high-efficient operation. For that purpose, real-time monitoring of combustion is necessary. Ion current sensor has attracted attention for its low cost and potential in catching information about combustion. The objective of this study is to extract information about fuel composition from ion current signal. Experiments were conducted on a gasoline engine bench with a spark plug modified to detect ion current. To simulate the diversity in fuel composition, different flow rates of CH₄ and CO₂ gas were introduced into intake manifold while gasoline was directly injected into cylinders. To analyze ion current data, which is characterized by noise and high cycle-to-cycle variation, principal component analysis (PCA) was employed. Principal components were extracted from the condition under the base fuel setting, and ion current history for other fuel settings were reconstructed using the principal components. To quantify the deviation in ion current signal caused by fuel variation, anomaly score was defined as the reconstruction error. Results showed that as gasoline injection amount decreased, CH₄ and CO₂ flow rates increased or equivalence ratio deviated from stoichiometric condition, anomaly scores tended to be higher. This demonstrates that the conditions deviating further from the base setting lead to high anomaly scores. Moreover, anomaly scores were higher under conditions with CO₂ introduction than those with CH₄, which implies that inactive gas has larger impact on ion current behavior than combustible gas. Finally, to evaluate the effect of calculation process, calculation sections of mean anomaly scores and number of principal components used for reconstruction were varied. It was clarified that adjustment of calculation process could have large effect on anomaly scores.

Lean combustion technology has been one of the promising approaches for developing next-generation engines with high thermal efficiency and low environmental impact. However, the practical lean limit of such engines is typically constrained by combustion stability. To address this issue, the present study introduces a newly designed intake adapter that can effectively generate in-cylinder tumble flow during the engine’s intake stroke, which is expected to promote ignition and flame propagation. The effect of this adapter was evaluated using a single-cylinder spark-ignition (SI) engine, with a particular focus on thermal efficiency and cycle-to-cycle variation. The results show that, with the adapter, the engine achieved a significant extension of the lean limit compared to the baseline case without flow control. Specifically, under operating conditions of IMEP=1.15 MPa and a compression ratio (CR) of 17, stable combustion was sustained at an air–fuel equivalence ratio of λ=2.26. Meanwhile, thermal efficiency improved by up to 3% relative to the no-adapter case. Furthermore, under lean conditions, the concentrations of unburned hydrocarbons (HC) and carbon monoxide (CO) in the exhaust were significantly reduced, indicating that the adapter enables stable combustion even under ultra-lean conditions. Finally, it was found that adjustments to the geometric parameters of the intake adapter can further enhance both thermal efficiency and lean limit performance under high-load and high-compression-ratio operations.

This study investigates the frost growth behaviors on a silver iodide (AgI) dot-patterned surface under desublimation conditions. The surface was fabricated using a two-step process combining UV lithography and UV nanoimprinting, resulting in a flat substrate with AgI dots embedded in a UV-curable resin (X433). The main objectives were to capture 3D frost morphologies through replica method and to evaluate whether such localized nucleation sites can induce laterally confined frost growth as observed in previous studies using AgI stripe patterns. Experiments were conducted at ambient temperatures of 2 ℃ and 10 ℃ with 40% relative humidity. Surface temperatures of −15.5 ℃ and −20.5 ℃ were selected to realize desublimation conditions, where the dew point is below freezing point. Frost formations on the AgI-patterned and bare resin surfaces were compared. On the AgI dot-patterned surface, frost nucleated selectively at certain AgI dots and grew horizontally, forming isolated columnar and planar crystals. These structures remained spatially separated for over 3 hours. In contrast, the bare surface exhibited typical frost behavior, i.e. dropwise condensation, droplet freezing, ice bridging, and eventually full coverage. Frost growth rate increased at lower surface temperature on both surfaces. On the AgI-patterned surface, the combination of higher ambient temperature and lower surface temperature further promoted nucleation and growth from the AgI dots, enhancing horizontal crystal development. To reconstruct the 3D frost structures, the replica method was applied. The AgI-patterned surface exhibited clear, laterally extended frost morphologies, while the bare PET surface showed random, interconnected frost growth. These results demonstrate that AgI dot patterns can effectively localize nucleation and preserve isolated and directional frost growth, offering a valuable approach for investigating frost growth mechanisms and developing frost controlling strategy.

Water is fundamental to a range of natural phenomena and is equally crucial in various engineering applications that demand efficient energy utilization, such as energy conversion with chemical reactions. However, the microscopic mechanisms of energy transfer in water remain unclear. This study investigated the instantaneous energy transfer (IET), formulated using forces obtained from ab initio molecular dynamics (AIMD) with a comprehensive analysis of its correlation with intermolecular and intramolecular structures in liquid water. By comparing our findings with those obtained from classical molecular dynamics (CMD), we examined the validity of the AIMD-based method and assessed the performance of the TIP4P/2005f water model. The results of the O–O, O–H, and H–H radial distribution functions (RDFs) indicated that the strongly constrained and appropriately normed (SCAN) functional in AIMD provided a more accurate representation of experimental values compared with the PBE-D3 functional. On the other hand, although the TIP4P/2005f model in CMD accurately reproduced some structural features, the classical force field exhibited some limitations, particularly in reproducing the height and width of the first peak in the O–O RDF. Moreover, we identified correlations between the IET and distance from the target oxygen atom to its nearest oxygen or hydrogen atoms, revealing that the characteristics of IET depend on this distance. Specifically, the mean IET: IET efficiency (IETE) was higher at shorter interatomic distances, indicating that both instantaneous intermolecular and intramolecular structures determine the IETE. It was also shown that variations in OH bond length significantly contributed to IETE. Additionally, our findings revealed that AIMD utilizing the SCAN functional shows higher IETE compared with that of CMD employing the TIP4P/2005f model. Through this study, the proposed method to evaluate IET has been validated, which will give fundamentals to understand transport phenomena in condensed phase in the framework of AIMD.

The relaxation from intra- to intermolecular vibrations of water is compared in the cases of the infrared heating that emits the radiation to the infrared absorption band of water with a wavelength-selective emitter and the conductive heating with a heater. The wavelength-selective emitters have a metamaterial structure of Au/Cr/Al₂O₃/Au/Cr/substrate and continuously emit the radiation in the wavelength range about 3.03 or 6.06 μm, corresponding to the stretching or bending vibrations of water, respectively. The relative reflectance of water is measured by using the attenuated total reflection method and a Fourier transform infrared spectrometer in both cases. The increase in relative reflectance at the wavelength range from 14.5 to 14.9 μm, corresponding to the intermolecular vibrations of water, is obtained by subtracting the relative reflectance in the conductive heating from the relative reflectance in the infrared heating. The t-test determines that the increase in relative reflectance can be considered statistically positive. It is concluded that the infrared heating that emits the radiation to the infrared absorption band of water can promote the intermolecular vibration of water better than the conductive heating with a heater. This conclusion means that the infrared heating with wavelength-selectivity in the infrared absorption band of water may enhance water evaporation more than the conductive heating.

Sweat monitoring is gaining increasing attention as a key component of next-generation health technologies. In this study, we developed a compact and flexible wearable sensor based on microelectromechanical systems (MEMS) technology for real-time sweat rate measurement. The sensor detects thermal dissipation of a heat source due to sweat flow within a microchannel, allowing for non-invasive and continuous monitoring. It consists of a microheater and two temperature sensors placed symmetrically upstream and downstream along the microchannel wall. When sweat enters the channel, the convective flow causes a temperature difference between the two sensors, which varies according to the sweat rate. Numerical simulations using finite element analysis were conducted to optimize the sensor geometry, particularly the spacing between the heater and the temperature sensors. The device was fabricated by patterning Cr-Au electrodes onto a flexible polyimide substrate and forming a PDMS microchannel over the sensing area. During exercise experiments, the device was attached to the upper arm of a participant, and the sensor output was recorded wirelessly using a compact data acquisition system. The results showed a clear increase in sweat rate during exercise and a gradual decay during the post-exercise recovery phase. These results demonstrate that the developed device can reliably capture temporal changes in sweat rate, making it a promising candidate for wearable applications in personalized health monitoring and fitness tracking.