Surface roughness is one of the most important factors to determine the flow and heat transfer characteristics of the microchannel. This paper experimentally and theoretically investigated the effects of surface roughness for the flow and heat transfer behavior within the circular microchannel. The stainless steel circular microchannels were fabricated by electrical spark-erosion perforating and drilling separately to control the relative roughness of the surface which is 1% for drilling method and 1.5% for electrical spark-erosion perforating method. Each test piece includes 44 identical circular microchannels in parallel with diameter of 0.4 mm. In the experiments, the air flowed through the circular microchannels with Reynolds number changing from 200 to 2600. The results showed that the surface roughness in microchannels has a remarkable effect on the performance of flow behavior and heat transfer within the circular microchannel. The values of Poiseuille number and Nusselt number are higher when the surface relative roughness is larger. At the same time, the flow behavior is inconsistent with the behavior within the macrochannel. For the flow behavior, Poiseuille number increases monotonously with the increase of Reynolds number, and is higher than the constant theoretical value. The Reynolds number for the transition from laminar to turbulent flow is between 1400 and 1600. For the heat transfer property, Nusselt number also increases as the increase of the Reynolds number.

Inspired from the idea of developing lattice Boltzmann method (LBM), a discrete ordinate method (DOM) with streaming and collision processes is presented for simulation of microflows in this work. The current method is quite different from the conventional discrete ordinate method (DOM), unified gas kinetic scheme (UGKS) and discrete unified gas kinetic scheme (DUGKS), in which the finite volume method (FVM) or the finite difference method (FDM) is usually utilized to discretize the discrete velocity Boltzmann equation (DVBE). Due to the application of FVM or FDM, the evaluation of the flux of distribution function at the cell interface becomes an essential step for these approaches. Besides that, for the UGKS and DUGKS, not only the flux of distribution functions but also the conservative variables at the cell interface are needed to be computed. These processes require a lot of computational efforts. In contrast, for the developed method, it only needs interpolations within the cell to perform the streaming process. Thus, the computational efficiency can be improved accordingly. To compare the accuracy and efficiency of present scheme with those of DSMC and/or UGKS, several numerical examples including the Couette flow, pressure driven Poiseuille flow and thermal transpiration flow are simulated. Numerical results showed that the solution accuracy of current scheme is comparable to that of DSMC and UGKS. However, as far as the computational efficiency is concerned, the present scheme is more efficient than UGKS.

This paper attempts to experimentally investigate the influence of channel lengths to the flow behavior and heat transfer characteristics in circular microchannels. The diameter of circular microchannels are 0.4mm and the lengths of them are 10mm and 20mm, respectively. All tests were performed with air. The experiments were completed with Reynolds number in the range of 300∼2700. Results of experiments show that the length of microchannels has remarkable effects on the performance of flow behavior and heat transfer characteristics. For the flow behavior, both the friction factor and Poiseuille number drop as the channel length increases, and all the experimental value are higher than the theoretical one. Moreover, the channel length doesn’t influent the value of critical Reynolds number. For the heat transfer characteristics, Nusselt number decreases with the increase of the channel length. The channel length also has a huge influence on the thermal performance. A better thermal performance is obtained in a shorter channel. The results also indicated that in both cases, the friction factor decreases with the increase of the Reynolds number. At the same time, the Poiseuille number increases when the Reynolds number keeps rising. This phenomenon is different from traditional theory that Poiseuille number is widely considered as a constant in laminar regime. It is also observed that the value of critical Reynolds number is between 1500 and 1700 in this paper, this value is lower than the value of theoretical critical Reynolds number of 2300. Nusselt number increases as the increase of the Reynolds numbers, however, the traditional theory considered that it is a constant in laminar regime.

Microchannel heat exchangers have become widely employed in modern systems, found within aerospace applications, waste heat recovery, water treatment processes, air conditioning, biomedical treatments and various industrial process applications. The microchannels increase the ratio of heat transfer surface to volume, thus improving the heat transfer performance significantly whilst reducing the overall weight and size. Moreover, by utilizing secondary flow from Dean Vortices induced by curved microfluidic channels, the fluid flow and heat transfer performance can be enhanced even further beyond conventional straight channels. However, since pressure drops found in microchannels are often quite high, channel lengths must be kept relatively short to balance the friction loss and energy consumption. Due to this, the developing region length at the microchannel entrance area has a greater impact than for macroscale channels, in terms of hydrodynamic and thermal performance over the remaining full developed region. The thermo-hydraulic design for heat transfer microchannel surfaces is strongly dependent on several dimensionless performance indicators, namely Nusselt number ‘Nu’ for heat transfer, and Poiseuille number ‘Po’, which is the product of Fanning friction factor ‘f’ and Reynolds number ‘Re’. These parameters are used to characterize and optimize the performance of microchannel surfaces and heat exchangers in general, also can be used to determine both the thermal and hydraulic developing region lengths at the channel entrance area. Whilst many such studies exist for theoretical analysis and experimental verifications, currently there is little literature on the developing region lengths and impacts researched through the method of Computational Fluid Dynamics (CFD). As such, this paper identifies and explores via quantitative analysis the hydraulic and thermal performance changes created by the relevant developing region lengths at the entrance area of spiral microchannels, as well as determinations and comparisons of these effects over straight channels. The numerical results, generated via COMSOL Multiphysics and contrasted with previous literature on the subject, also compared with the effect of the developing region on the effectiveness and efficiency of both spiral and straight microchannels, finding an improved heat transfer performance but an increased impact of hydraulic friction as well for spiral channels against straight counterpart. Furthermore, significant differences between thermal developing region length and hydraulic developing region length can be observed throughout, which illustrates high challenge and the need for compromise in microchannel design. In this way, implications for the configuration and design of industrial microchannels and micro heat exchangers are self-evident. All the key factors given in this paper are dimensionless, and thus the generated results can be utilized for a variety of flow conditions. Hence, this work should permit an increased understanding for and boost the curved microchannel and micro heat exchanger designs subsequently, through reducing the required numbers of tests and experiments and expediting the development for similar applications followed.

Amongst various porous media, open-cell metallic foams exhibit distinctive properties: relatively low manufacturing cost, ultra-low density, moderate stiffness and strength, and high surface area-to-volume ratio. They have been, therefore, utilized in a variety of applications such as microelectronics cooling, fuel cells, and compact heat exchangers. For such applications, the knowledge of pressure drop of fluid flowing across the foam is often a key issue, enabling control of fluid flow, heat transfer enhancement, planning and designing chemical engineering processes, optimal flow analysis as well as practical designs. We present in this paper an analytical model capable of predicting the pressure drop of a Newtonian incompressible fluid flowing unidirectionally across isotropic and fully-saturated micro open-cell cellular foams within the Darcy and Forchheimer flow regimes. Analytical exploitations are conducted to determine the foam permeability and inertial coefficient. The analytical model is based on the basis of volume-averaging approach and the assumption of piece-wise plane Poiseuille flow with the modified cubic lattice with spherical node at the junction of struts. To better mimic the foam struts shape, a concave-triangular-shaped strut consisting of two nose-to-nose cones is considered and particular attentions have been paid to both analytically and numerically examine the node shape as well as struts shape effect. Built upon a generalized tortuosity model derived from the modified cubic unit cell, an analytical model of permeability on the basis of a cubic unit cell is developed, valid within a typical engineering range of porosity (ε = 0.86 ∼ 0.98) and pore size (0.254 mm ∼ 5.08 mm). With the effect of Reynolds number considered, the pore-scaled Reynolds number dependent drag coefficient expression is introduced and through this the inertial coefficient is analytically modeled on the basis of flow over bluff bodies, which is found to agree well with experimental data from various sources. The modeling procedure for pressure drop (permeability and inertial coefficient) is based on physical principles and geometrical considerations, and the model predictions agree satisfactorily with existing experimental data. Results show that by building the analytical model on the basis of a cubic unit cell to represent the topology of metallic foams, pressure drops as well as hydrodynamic conditions within both the Darcy and Forchheimer regimes in a Newtonian fluid can be analytically predicted.

Non-equilibrium molecular dynamics method is applied to determine the slip length of fluid in the vicinity of the boundaries in Poiseuille flow in a nano-channel with and without nanoscale roughness, in which pressure is imposed to the fluid. Our simulations reveal the boundary roughness effect on the interfacial interaction between the fluid and boundary, which will result in the change of fluid velocity, and this phenomenon is more distinct with boundary roughness height. Bonding energy between solid boundary and fluid atoms is also calculated to make a comparison with that of pure fluid. Roughness changes with highness and its shape, so the key factors controlling fluid flow are presented. The simulation results show that there’s a relationship between the bonding energy at fluid–solid boundary interface and the fluid kinetic properties. In the case of rough surfaces, the bonding strength between liquid and solid atoms has a strong signature in the fluid velocity close to the boundary. In addition, temperature effect is also considered in our simulations, and we find that temperature also affect the hydrodynamic properties of fluid flowing in nano-channel significantly.

This paper studies the roughness effect of the two-dimensional micro Poiseuille gas flows in a curved channel by a modified lattice Boltzmann model. A method relating to the Knudsen number ( Kn ) with the relaxation time is discussed. In addition, to capture the slip velocity on a solid boundary more accurately, a combined boundary scheme (CBC), which combines the no-slip bounce-back and the free-slip specular reflection schemes, is applied to boundary condition treatment. The rough wall of the micro-channel is described by uniformly distributed rectangular or triangular rough elements. The simulation results show that the roughness and the geometries of the channel have great effects not only on velocity distribution but also on pressure distribution.

Different opinions still exist on some basic principles of DSMC method, such as the proper grid dimension and the proper number of particles in a cell. In this paper DSMC simulation of Poiseuille flow is made to evaluate the dependence of simulation results on cell dimension and number of particles per cell. In the simulation process a self adapting block structured grid system is employed to make sure that the number of particles per cell is constant. The simulation covers both slip flow regime and transition flow regime and each regime covers both high pressure and low pressure. Our simulation results indicate that the number of particles per cell and scaling factor exert little influence on simulation result for both slip flow and transition flow when the number of particles per cell surpasses 5, but the dimension of cell influences the accuracy of result obviously. The error caused by cell dimension decreases as the diminish of cell dimension. It is concluded that in the DSMC method it is necessary to make sure that the cell is less than 1/2 of molecular mean free path.

A flow wall collision model between two parallel plane walls was established for argon fluid through a microchannel under the condition of action of periodical external force. Based on this model, two kinds of wall (hydrophilic and hydrophobic) were applied on the flow simulations from non-equilibrium molecular dynamics (NEMD). There are 864 fluid particles and wall particles separately in the simulated system. The non-dimension height of microchannel is 9.667. The velocity profile and temperature profile of argon fluid in hydrophilic microchannel predicted by molecular dynamics simulation are in good agreement with the analytical solution based on the Navier–Stokes and energy equations. The velocity profile and the temperature profile experience a large jump in the layers close to the hydrophobic wall.

In this paper various extended macroscopic models are described and applied to force-driven Poiseuille flow. In particular, details are given for the regularized Grad 13- and 20-moment equations. Extended macroscopic models have, until recently, been limited by the uncertainity surrounding the prescription of boundary conditions on solid-walls. The gas-solid wall interaction plays an important role in describing the dynamics of confined gaseous flows. This problem is tackled in the context of the moment equations whereby the simplified Maxwell microscopic formalism is used to derive boundary conditions for a given moment equation set. The proposed governing equations and boundary conditions are applied to force-driven Poiseuille flow where anomalous thermal behavior is observed as the Knudsen number increases. Results are compared to DSMC data and it is established that the proposed extended macroscopic models can capture this non-intuitive behavior. However, the models show some quantitative disparity in representing this behavior. It is proposed that this is addressed by development of a consistent theory of molecular collision geometries in the extended hydro-dynamic model or by the utilization of more extended moment sets.