In this paper, we present a theoretical study aimed at investigating the rate-limiting factors in thin-film evaporative heat transfer processes, considering the finite-rate evaporation kinetics. The problems of evaporation of a flat thin-film in either pure vapours or vapour-inert-gas mixtures are analysed based on the non-dimensionalised macroscopic transport equations for continuum fluids, coupled with out-of-equilibrium kinetic boundary conditions. Both the full numerical solutions and asymptotic analytical solutions at slow evaporation limit are provided and applied to analyse thin water film evaporation. Existing solutions, assuming negligible heat transfer in the gas domain, or negligible temperature jump across the non-equilibrium kinetic layer, or more boldly a thermodynamically equilibrial interface (i.e. its temperature is at the saturation temperature), can be fully recovered from the more general solutions presented here. Our results show that while these assumptions hold in special cases, they can lead to significant errors in many conditions, especially when the film thickness δ is reduced to a few micrometers or thinner. We show that the conventional views that the rate-limiting factors in thin-film evaporative heat transfer is either the heat diffusion through the liquid film or the mass transfer in the gas domain only apply to thick film (i.e. δ≫λ where λ is the mean free path in the vapour phase). As δ decreases to a few micrometer or smaller (more precisely when the Knudsen number Kn increases beyond O(1) in an pure vapour environment or when the kinetic Peclet number Pe is reduced below O(1) in inert gases), the interfacial thermal resistance due to the evaporation kinetics can be on the same orders of magnitude as the thermal resistance in the liquid film. The analysis also allow us to compare the heat transfer processes during the evaporation of a thin-film in pure vapours to those in inert gases, providing deeper insight into the effectiveness of various strategies for exploring the evaporation process in practical thermal management.

A general mixed kinetic-diffusion boundary condition is formulated to account for the out-of-equilibrium kinetics in the Knudsen layer. The mixed boundary condition is used to investigate the problem of quasi-steady evaporation of a droplet in an infinite domain containing inert gases. The widely adopted local thermodynamic equilibrium assumption is found to be the limiting case of infinitely large kinetic Péclet number ${{Pe}_k}$ , and it introduces significant error for ${{Pe}_k} \leqslant O(10)$ , which corresponds to a typical droplet radius $a$ of a few micrometres or smaller. When compared with experimental data, solutions based on the mixed boundary condition, which take into account the temperature jump across the Knudsen layer, better predict the time evolution of $a$ than the classical $D^2$ -law (i.e. $a^2 \propto t$ , where $t$ denotes time). In the slow evaporation limit, an analytical solution is obtained by linearising the full formulation about the equilibrium condition, which shows that the $D^2$ -law can be recovered only in the large ${{Pe}_k}$ limit. For small ${{Pe}_k}$ , where the process is dominated by kinetics, a linear relation, i.e. $a \propto t$ , emerges. When the gas phase density approaches the liquid density (e.g. at high-pressure or low-temperature conditions), the increase in the chemical potential of the liquid phase due to the presence of inert gases needs to be accounted for when formulating the mixed boundary condition, an effect largely ignored in the literature so far.

An integrated batteries and power electronics system has great potential in improving the compactness, flexibility and multifunctionality in electrical vehicles. However, it needs to overcome thermal management challenges due to different operation temperatures of batteries and power electronics. To tackle this issue, this paper presents the first systematic study on the heat transfer characteristics in phase change materials (PCMs) based thermal management system sandwiched between two significantly different constant heat sources, through an experimentally validated transient three-dimensional heat transfer model. All key parameters of PCMs affecting the system are identified through a new dimensionless formulation, including the ratio of horizontal and vertical thermal conductivity kxy and kz, the aspect ratio (ratio of thickness and length), and Jakob number Ja (ratio of sensible heat and latent heat). Modelling results show that the system operation duration & tau;d increases monotonically with the increase of kxy. However, increasing kz is not always benificial for & tau;d and there is an optimal value for kz when keeping kxy unchanged. In addition, the increase of PCMs thickness can prolong & tau;d monotonically. The reduction of Ja, e.g. increasing latent heat, is always beneficial for & tau;d.

Evaporation of metastable droplets in an immiscible liquid can occur in a wide range of practical applica-tions and in particular in cryogenic systems due to the low saturation temperature of the cryogenic liquid. In this article, we show experimentally that the incomplete explosion of a metastable droplet can lead to a 'film evaporation' configuration, where a remaining droplet is eventually separated from the host immis-cible liquid by an established vapour layer. Based on a transient heat diffusion model, we identified that the necessary but not sufficient condition for the complete explosion of a metastable droplet is that the droplet must absorb, before explosion, an amount of heat larger than the latent heat required for a full evaporation, i.e. the global Jacob number Ja >= 1 . In addition, the measured superheat limits of pentane and isohexane are 420.65 K and 4 4 4.95 K respectively, which are very close to the theoretical Spinodal points (<= 1.5 K) and other reported values in existing literature. Therefore, the experimental setup re-ported here can be used as an alternative way to quantify the superheat limit. The Droplet relaxation time was also found scaled to the square root of the diffusion characteristic time. (c) 2021 Elsevier Ltd. All rights reserved.

In this article, we examine the hybridisation of refrigerated commercial vehicles through replacing the traditionally used auxiliary diesel engine with a non-polluting, non-electric unit as an effective emissions reduction alternative. The zero-emission hybrid solution presented in this article is a Liquid Nitrogen (LN2) engine system, featured with a novel integrated Heat Exchange Fluid (HEF) subsystem, that can provide simultaneous cooling and auxiliary power in, for example, refrigerated trucks. Evaporation of LN2 provides the cooling/refrigeration power. The resulting high pressure gaseous N2 then expands in the engine, producing shaft power. A major contribution of this research is the use of a novel direct in-cylinder HEF supply technology which we show experimentally that it leads to reliable and significantly enhanced engine performance. Specifically, a detailed experimental investigation into the effects of HEF temperature and flow rate at different inlet N2 conditions and engine speeds on engine performance is presented. Results from a thermodynamic analysis, based on an idealised cycle, are also presented to better understand the engine performance and assess the potential of the proposed engine architecture. The results show up to 41% brake thermal efficiency and up to 172 kJ/kg-LN2 specific work from the engine system, which are significantly higher figures when compared to previously reported maximum values in the literature (i.e. 9.2% and 40 kJ/kg-LN2, respectively). It is also shown that the thermodynamic model can predict with good accuracy the upper and lower limits of the measured indicated power and efficiency.

It has been proven that performance gains in liquid nitrogen (LN2) engine systems, generating simultaneous cooling and auxiliary power, can be achieved through integration of a dedicated heat exchange fluid (HEF) circuit. The novel, HEF enhanced LN2 engine system can be utilised as an optimised hybrid solution for commercial refrigeration trucks. Although the benefits arising from HEF addition have been researched, there are no articles investigating the effect of changing the HEF composition on engine performance. This article reports a detailed experimental investigation on the performance of a novel, HEF enhanced LN2 engine system. The key contribution of the current study is the knowledge generated from investigating the impact of different HEF compositions on the engine performance under different HEF temperatures, N2 inlet conditions and engine speeds. The HEF composition was varied through changing the water content in the mixture. A thermodynamic model based on an idealised cycle was used to assist interpretation of the experimental results and assess the potential of the proposed engine architecture. The experimental study demonstrated up to 42.5% brake thermal efficiency, up to 2.67 kW of brake power and up to 174 kJ/kg specific energy, which were higher than previously published figures for LN2 engine systems. A reduction in the HEF water content was found to generally increase the engine power output at a HEF temperature of 30 °C. However, at a HEF temperature of 60 °C, the impact of HEF composition was found to be minor and nonmonotonic. The thermodynamic model predicted the upper and lower limits of the measured indicated power and indicated thermal efficiency with acceptable accuracy.

Cryogenic fluids such as liquid nitrogen can be used to produce power with negligible tailpipe emissions. In this work, quantitative evaluations of evaporation rates of a nitrogen droplet immersed into different immiscible liquids maintained at different bulk temperatures were carried out by accurately tracking the three-dimensional evaporation process using a high-speed backlight imaging system. A comparison of the evaporation rates measured by the surrounding bubble growth and heat transfer rates at different bulk temperatures has been made for a droplet evaporating in 2-propanol, methanol, n-pentane and n-hexane.

Liquid nitrogen or other cryogenic liquids have the potential to replace or augment current energy sources in cooling and power applications. This can be done by the rapid evaporation and expansion processes that occur when liquid nitrogen is injected into hotter fluids in mechanical expander systems. In this study, the evaporation process of single liquid nitrogen droplets when submerged into n-propanol, methanol, n-hexane, and n-pentane maintained at 294 K has been investigated experimentally and numerically. The evaporation process is quantified by tracking the growth rate of the resulting nitrogen vapour bubble that has an interface with the bulk liquid. The experimental data suggest that the bubble volume growth is proportional to the time and the bubble growth rate is mainly determined by the initial droplet size. A comparison between the four different bulk liquids indicates that the evaporation rate in n-pentane is the highest, possibly due to its low surface tension. A scaling law based on the pure diffusion-controlled evaporation of droplet in open air environment has been successfully implemented to scale the experimental data. The deviation between the scaling law predictions and the experimental data for 2-propanol, methanol and n-hexane vary between 4 and 30% and the deviation for n-pentane was between 24 and 65%. The more detailed bubble growth rates have been modelled by a heuristic one-dimensional, spherically symmetric quasi-steady-state confined model, which can predict the growth trend well but consistently underestimate the growth rate. A fixed effective thermal conductivity is then introduced to account for the complex dynamics of the droplet inside the bubble and the subsequent convective processes in the surrounding vapour, which leads to a satisfactory quantitative prediction of the growth rate. (C) 2019 Elsevier Ltd. All rights reserved.

This paper presents corrections for existing hydrodynamic instability-based critical heat flux (CHF) models in pool boiling by taking into account the effect of the viscosity, geometry and size of the liquid–vapor interface. Based on the existing literature, the Kelvin–Helmholtz (KH) theory, used by the most commonly adopted CHF models, can lead to noticeable errors when predicting the instability conditions. The errors are mainly due to the inaccuracy of the inviscid flow assumptions and the oversimplification of the interface geometry. In addition, the literature suggests the most unstable condition predicted by the viscous correction for viscous potential flow (VCVPF) theory for the cylindrical interfaces best match the observed air column breakup conditions in water. In this paper, the most unstable instability conditions predicted by the VCVPF theory are used to correct the existing CHF models. The comparison between the existing and corrected CHF models suggests that the corrected models always predict a higher CHF value. In addition, the corrected Zuber model predicts similar CHF value to the Lienhard and Dhir model. The comparison with experimental data suggests that the correction to the Zuber model can increase its prediction accuracy in most cases, but not necessary for the Lienhard and Dhir model. When compared to experimental CHF data for boiling cryogens at different pressures, the corrected CHF models are consistently more accurate than the original CHF models.

•A new model has been developed to predict the Critical Heat Flux (CHF) in pool boiling based on hydrodynamic instability and bubble growth theory.•The new model can predict the critical heat flux when boiling methanol on the small flat surface and long horizontal cylinders with different size to within 5% uncertainty.•The new model can predict the critical heat flux on different contact angles to within 10% uncertainty for both hydrophilic and hydrophobic surfaces.•The new model can predict the critical heat flux when boiling water on both flat surface and long horizontal cylinders to within 5% uncertainty up to 10 bar system pressure and the critical heat flux when boiling methanol on a flat surface to within 10% uncertainty.•The fundamental equations and different sub-models can predict the CHF and detailed bubble processes well.

This paper derives analytical solutions for the critical Kelvin-Helmholtz (KH) instability conditions at the interface between a cylindrical gas column and a pool of viscous immiscible fluid confined in a chamber of finite size. The analysis focuses on conditions of negligible heat and mass transfer. The derivations are based on the established approaches reported in the literature with different boundary conditions. The most unstable instability conditions have also been calculated numerically. Experiments designed to measure the actual air column break-up conditions in water have been carried out to validate the analytical models. Comparisons show that the most unstable conditions predicted by the Viscous Corrections of the Viscous Potential Flow KH model are the best match to the experimentally measured break-up conditions. Parametric investigation of the instability theories shows that the vapour column size has a noticeable effect on the critical conditions, but has a negligible effect on the most unstable conditions when the column radius is greater than 1.2 mm. Furthermore, the critical instability conditions are sensitive to the chamber size and the perturbation symmetry, while the most unstable conditions are insensitive to these parameters. (C) 2018 Elsevier Ltd. All rights reserved.

In this paper, droplet impact on a porous surface is experimentally investigated over a wide range of Weber numbers and surface temperatures. Regime transition criteria have been deduced to determine droplet post-impingement behaviour as a function of the Weber number and surface temperature for which a droplet impacting on a porous surface. Based on the energy balance, an analytical model with improved boundary layer description is proposed to predict maximum spreading of droplet following impact on porous surfaces when the effect of heat transfer is negligible. The results of the model indicate that the spreading process after droplet impact on porous surfaces is governed by the viscous dissipation and matric potential. The maximum-spread model predictions agreed well with experimental measurements reported in this paper and the literature over a large range of Weber numbers and different porous surfaces. (C) 2018 Elsevier Ltd. All rights reserved.

Flow coefficients of intake valves and port combinations were determined experimentally for a compressed nitrogen engine under steady-state and dynamic flow conditions for inlet pressures up to 3.2 MPa. Variable valve timing was combined with an indexed parked piston cylinder unit for testing valve flows at different cylinder volumes while maintaining realistic in-cylinder transient pressure profiles by simply using a fixed area outlet orifice. A one-dimensional modeling approach describing three-dimensional valve flow characteristics has been developed by the use of variable flow coefficients that take into account the propagation of flow jets and their boundaries as a function of downstream/upstream pressure ratios. The results obtained for the dynamic flow cases were compared with steady-state results for the cylinder to inlet port pressure ratios ranges from 0.18 to 0.83. The deviation of flow coefficients for both cases is discussed using pulsatile flow theory. The key findings include the followings: (1) for a given valve lift, the steady-state flow coefficients fall by up to 21% with increasing cylinder/manifold pressure ratios within the measured range given above and (2) transient flow coefficients deviated from those measured for the steady-state flow as the valve lift increases beyond a critical value of approximately 0.5 mm. The deviation can be due to the insufficient time of the development of steady-state boundary layers, which can be quantified by the instantaneous Womersley number defined by using the transient hydraulic diameter. We show that it is possible to predict deviations of the transient valve flow from the steady-state measurements alone.

A new optical diagnostic technique has been used to measure the spatially distributed temperatures, soot diameters, and soot volume fractions in several different ethylene laminar diffusion flames to investigate the effect of adding hydrogen and helium on the soot formation. The test results show that adding hydrogen increases the flame temperature in all regions, while adding helium does not significantly affect the flame temperature in the reaction region but does increase the flame temperature elsewhere. The flame heights when adding helium and hydrogen can be calculated using the correlation introduced by Roper if the ethylene diffusion coefficient is used. This indicates that the flame height is determined by the diffusion of ethylene molecules when the hydrogen fraction is below 20%. It was also found that either adding helium or hydrogen does not significantly affect the soot diameter but does reduce the soot volume fraction. A total of 20% of helium addition by volume was measured to reduce the total soot number by 19%, while a total of 20% of hydrogen addition reduced the total soot number by 23%. In comparison, replacing the hydrocarbon with hydrogen is much more effective in reducing soot formation. Replacement of 25% ethylene by hydrogen was measured to reduce the total soot number by 66%. Apart from demonstrating the influence of hydrogen and helium on ethylene diffusion flames, these measurements provide additional data for modelers of diffusion flames, especially those with an interest in the formation of particulate matter.

A new low-cost optical diagnostic technique, called Cone Beam Tomographic Three Colour Spectrometry (CBT-TCS), has been developed to measure the planar distributions of temperature, soot particle size, and soot volume fraction in a co-flow axi-symmetric laminar diffusion flame. The image of a flame is recorded by a colour camera, and then by using colour interpolation and applying a cone beam tomography algorithm, a colour map can be reconstructed that corresponds to a diametral plane. Look-up tables calculated using Planck's law and different scattering models are then employed to deduce the temperature, approximate average soot particle size and soot volume fraction in each voxel (volumetric pixel). A sensitivity analysis of the look-up tables shows that the results have a high temperature resolution but a relatively low soot particle size resolution. The assumptions underlying the technique are discussed in detail. Sample data from an ethylene laminar diffusion flame are compared with data in the literature for similar flames. The comparison shows very consistent temperature and soot volume fraction profiles. Further analysis indicates that the difference seen in comparison with published results are within the measurement uncertainties. This methodology is ready to be applied to measure 3D data by capturing multiple flame images from different angles for non-axisymmetric flame.

Three different fractions (2%, 5%, and 10% of stoichiometric, or 2.38%, 5.92%, and 11.73% by energy fraction) of hydrogen were aspirated into a gasoline direct injection engine under two different load conditions. The base fuel was 65% iso-octane, and 35% toluene by volume fraction. Ignition sweeps were conducted for each operation point. The pressure traces were recorded for further analysis, and the particulate emission size distributions were measured using a Cambustion DMS500. The results indicated a more stable and faster combustion as more hydrogen was blended. Meanwhile, a substantial reduction in particulate emissions was found at the low load condition (more than 95% reduction either in terms of number concentration or mass concentration when blending 10% hydrogen). Some variation in the results occurred at the high load condition, but the particulate emissions were reduced in most cases, especially for nucleation mode particulate matter. Retarding the ignition timing generally reduced the particulate emissions. An engine model was constructed using the Ricardo WAVE package to assist in understanding the data. The simulation reported a higher residual gas fraction at low load, which explained the higher level of cycle-by-cycle variation at the low load.