Numerical Investigation of a Periodic Heating Thermal-Bubble Actuated Diffuser–Nozzle Valveless Pump

A valveless micropump actuated by thermal bubbles which generated by an electrode heater mounted with a pair of diffuser nozzles has been numerically studied by commercial CFD software FLUENT. The relationship between the net flow rate and the superheating and heat supplying frequency has been investigated. The depth of the diffuser–nozzle micropump used in current numerical simulation model is 200 μm, and the diameter of micropump chamber is 1 mm. The pair of diffuser–nozzles are with gaps expanding from 30 to 274 μm and open angles of 7°. The working fluid is methanol in present study. The results show that the pump has different optimal driving frequency with different superheating. The cycle composed of bubble growth and shrinking costs more time at higher superheating. The maximum volume flow rate and the maximum pump pressure will increase with increasing superheating, simultaneously; and the optimal pulse duty, the maximum volume flow rate and pump pressure decrease with increasing superheating. The maximum volume flow rate and the maximum pump pressure are 29.6 μL/min and 680 Pa at ΔT = 15°C, respectively.     

Flow Boiling Heat Transfer in Two-Phase Micro Channel Heat Sink at Low Water Mass Flux

Boiling heat transfer at water flow with low mass flux in heat sink which contained rectangular microchannels was studied. The stainless steel heat sink contained ten parallel microchannels with a size of 640 × 2050 μm in cross-section with typical wall roughness of 10–15 μm. The local flow boiling heat transfer coefficients were measured at mass velocity of 17 and 51 kg/m²s, heat flux on 30 to 150 kW/m² and vapor quality of up to 0.8 at pressure in the channels closed to atmospheric one. It was observed that Kandlikar nucleate boiling correlation is in good agreement with the experimental data at mass flow velocity of 85 kg/m²s. At smaller mass flux the Kandlikar model and Zhang, Hibiki and Mishima model demonstrate incorrect trend of heat transfer coefficients variation with vapor quality.     

Solvent control of cellulose acetate nanofibre felt structure produced by electrospinning

Non-woven structures of cellulose acetate (CA) fibres of 90 nm–5 μm in diameter (spinning parameters 90 nm beaded fibres: 12% CA in EtOH-DMSO 1/1, 22 kV, 30 cm, 0.5 mL/h; maximum 5 μm diameter fused fibres spun with 14% CA in Ac-BenzOH 2/1, 22 kV, 24 cm, 13 mL/h) were produced by electrospinning. On the basis of Hansen solubility theory, composition of binary solvent mixtures (ketones—acetone, methyl ethyl ketone (MEK), and alcohols—benzyl alcohol, propylene glycol and dimethylsulphoxide) was optimized with respect to control of fibre felt morphology. Fibre networks of high packing density were obtained with binary low-volatile alcohols/MEK solvent mixtures, a decreased spinning distance and an increased feed rate. Substituting MEK by acetone in the solvent mixture resulted in the formation of nanofibre felt with a low degree of fibre cross-links. Thus, solvent control is a key aspect for control of electrospun fibre felt structures, which may serve as scaffolds for tissue engineering.     

Numerical simulation of gas-particle flows behind a backward-facing step using an improved stochastic separated flow model

An improved stochastic separated flow (ISSF) model developed by the present authors is tested in gas-particle flows behind a backward-facing step, in this paper. The gas phase of air and the particle phase of 150 μm glass and 70 μm copper spheres are numerically simulated using the k–ɛ model and the ISSF model, respectively. The predicted mean streamwise velocities as well as streamwise and transverse fluctuating velocities of both phases agree well with experimental data reported by Fessler. The reattachment length of 7.6H matches well with the experimental value of 7.4H. Distributions of particle number density are also given and found to be in good agreement with the experiment. The sensitivity of the predicted results to the number of calculation particles is studied and the improved model is shown to require much less calculation particles and less computing time for obtaining reasonable results as compared with the traditional stochastic separated flow model. It is concluded that the ISSF model can be used successfully in the prediction of backward-facing step gas-particle flows, which is characterised by having recirculating regions and anisotropic fluctuating velocities. Received 20 June 2000     

Regimes of Two-Phase Flow in Short Rectangular Channel

Experimental study of two-phase flow in the short rectangular horizontal channel with height 440 μm has been performed. Characteristics of liquid motion inside the channel have been registered and measured by the Laser Induced Fluorescence technique. New information has allowed determining more precisely the characteristics of churn regime and boundaries between different regimes of two-phase flow. It was shown that formation of some two-phase flow regimes and transitions between them are determined by instability of the flow in the lateral parts of the channel.     

Effect of a moving boundary on pulsatile flow of incompressible fluid in a tube

The pulsatile flow in a pipe with a moving boundary has been studied for a viscous, incompressible fluid by solving the Navier-Stokes equations numerically. The governing equations were formulated in boundary fitted curvilinear coordinates and a finite volume discretization procedure was used to solve the problem. This analysis is based on the assumption that the flow has a simple periodic pulsation and the shape of the wall changes according to the frequency of pulsation. The presence of the moving boundary causes unsteadiness in the flow behaviour as the vibrating wall has a nonlinear interaction with the flow. A detailed analysis of the flow field is presented here for a range of frequencies (5≤α≤10) where α is the reduced frequency parameter and a Reynolds number of 100.     

Fabrication of silicon based glass fibres for optical communication

Silicon based glass fibres are fabricated by conventional fibre drawing process. First, preform fabrication is carried out by means of conventional MCVD technique by using various dopants such as SiCl₄, GeCl₄, POCl₃, and FeCl₃. The chemicals are used in such a way that step index single mode fibre can be drawn. The fibre drawing process consists of various steps such as heating the preform at elevated temperature, diameter monitor, primary and secondary coating, and ultra violet radiation curing. The fibres are then characterized for their geometrical and optical properties. The drawn fibre has diameter of core and cladding to be 8.3 μm and 124.31 μm, respectively whereas non-circularity is found to be 4.17% for core and 0.26% for cladding as seen from phase plot. Mode field diameter is found to be 8.9 μm and 9.2 μm using Peterman II and Gaussian method, respectively. The fabricated fibres showed the signal attenuation of 0.35 dB/km and 0.20 dB/km for 1310 nm and 1550 nm, respectively as measured by the optical time domain reflectometer (OTDR).     

Dependence of flow characteristics of rectangular cylinders near a wall on the incident velocity

Numerically simulated results are presented for a family of rectangular cylinders with aspect ratios r ₁ (=b/a with height a and width b) ranging from 0.1 to 1.0 (square cylinder) to gain a better insight into the dependency of the aerodynamic characteristics on the operational dimensionless parameters, namely Reynolds number Re and aspect ratio r ₁. This work describes the flow from a long cylinder of rectangular cross-section placed parallel to a wall and subjected to a uniform shear flow. The flow is investigated in the laminar Reynolds number range (based on the incident stream at the cylinder upstream face and the height of the cylinder) at cylinder to wall gap height 0.5 times the cylinder height. The governing unsteady Navier?CStokes equations are solved numerically through a finite volume method on a staggered grid system using QUICK scheme for convective terms. The resulting equations are then solved by an implicit, time-marching, pressure correction-based SIMPLE algorithm for Reynolds number up to 1,000. The critical Reynolds numbers at which vortex shedding from the cylinder is started are specified for both the cases: far from the wall and near to the wall. It is reported that the vortex shedding from the rectangular cylinder of lower aspect ratio r ₁ (???0.25) becomes regular and insensitive to the Reynolds number, while the aerodynamic characteristics of the rectangular cylinders with higher aspect ratio r ₁ (???0.5) are strongly dependent on the Reynolds number.     

On the motion of superparamagnetic particles in magnetic drug targeting

A stochastic ODE model is developed for the motion of a superparamagnetic cluster suspended in a Hagen-Poiseuille flow and guided by an external magnet to travel to a target. The specific application is magnetic drug targeting, with clusters in the range of 10–200 nm radii. As a first approximation, we use a magnetic dipole model for the external magnet and focus on a venule of 10⁻⁴ m radius close to the surface of the skin as the pathway for the clusters. The time of arrival at the target is calculated numerically. Variations in release position, background flow, magnetic field strength, number of clusters, and stochastic effects are assessed. The capture rate is found to depend weakly on variations in the velocity profile, and strongly on the cluster size, the magnetic moment, and the distance between the magnet and the blood vessel wall. A useful condition is derived for the optimal capture rate. The case of simultaneous release of many clusters is investigated. Their accumulation in a neighborhood of the target at the venule wall follows a normal distribution with the standard deviation roughly half of the distance between the magnet and the target. Ideally, this deviation should equal the tumor radius, and the magnet should be beneath the center of the tumor. The optimal injection site for a cluster is found to be just prior to arrival at the target. Two separate mechanisms for capturing a cluster are the magnetic force and, for radii smaller than 20 nm, Brownian motion. For the latter case, the capture rate is enhanced by Brownian motion when the cluster is released near the wall.     

Numerical simulation of microparticles transport in a concentric annulus by Lattice Boltzmann Method

Dispersion and removal of microaerosol particles are investigated numerically in a horizontal concentric annulus by Lattice Boltzmann Method and Lagrangian Runge–Kutta procedure with the assumption of one-way coupling. Drag, buoyancy, gravity, shear lift, Brownian motion and thermophoretic are forces that are included in particle equation of motion. All simulations were performed at Rayleigh number of 10⁴ and particles specific density of 1000. The effect of aspect ratio and particles diameter were determined on particles behavior such as removal and dispersion. Results show that recirculation power increases by decreasing of cylinders gap. Particles move in a thinner quasi-equilibrium region by increasing of their diameter and decreasing of cylinders gap. Brownian motion is dominant removal mechanism in particle with diameter of 1 μm.     

EULERIAN AND LAGRANGIAN SIMULATIONS OF PARTICLE DEPOSITION FROM A POINT SOURCE IN A TURBULENT CHANNEL FLOW

Results comparing Eulerian and Lagrangian simulations of particle deposition from a point source in a channel are presented. The mean turbulent flow field is simulated using a two-equation k-ε turbulence model. In the first, approach, diffusion of aerosol particles is studied by solving the corresponding advection-diffusion equation. Deposition of particles in the intermediate size range are analyzed by considering both the turbulent eddy diffusion and the eddy impaction processes, as well as the Brownian diffusion effects. In the second approach, the turbulence fluctuating velocity field are numerically simulated as a Gaussian random process. The Lagrangian trajectories of aerosol particles in the channel are then evaluated by solving the corresponding particle equation of motion. Effects of Brownian diffusion on particle motions are also included. A series of digital simulations for particles of various sizes which are released at different locations across the channel are carried out. Depositions of different size particles on the wall under a variety of conditions are analyzed. The relative significance of turbulence and Brownian effects are also discussed.     

EULERIAN AND LAGRANGIAN SIMULATIONS OF PARTICLE DEPOSITION FROM A POINT SOURCE IN A TURBULENT CHANNEL FLOW

ABSTRACT

Results comparing Eulerian and Lagrangian simulations of particle deposition from a point source in a channel are presented. The mean turbulent flow field is simulated using a two-equation k-? turbulence model. In the first, approach, diffusion of aerosol particles is studied by solving the corresponding advection-diffusion equation. Deposition of particles in the intermediate size range are analyzed by considering both the turbulent eddy diffusion and the eddy impaction processes, as well as the Brownian diffusion effects. In the second approach, the turbulence fluctuating velocity field are numerically simulated as a Gaussian random process. The Lagrangian trajectories of aerosol particles in the channel are then evaluated by solving the corresponding particle equation of motion. Effects of Brownian diffusion on particle motions are also included. A series of digital simulations for particles of various sizes which are released at different locations across the channel are carried out. Depositions of different size particles on the wall under a variety of conditions are analyzed. The relative significance of turbulence and Brownian effects are also discussed.     

Investigation of the process of diamagnetic particle separation in a high-gradient ordered-structure magnetic field

On the basis of the model of a flow-type magnetic filter with a transversely magnetized ordered system of long ferromagnetic rods of rectangular cross section, the process of high-gradient magnetic separation of microscopic diamagnetic particles (potato starch granules of sizes 8–30 μm) from a liquid suspension has been investigated. The registered laws of change in the concentration and size distribution of particles at the suspension outlet from the filter agree with the theoretical conclusions obtained from the analysis of the magnetic field structure and thecharacter of the particle motion in the filter volume.     

The role of friction in mixing and segregation of granular material

We use discrete element modelling to investigate the processes of mixing and size segregation in a polydisperse mixture of spherical particles in a three-dimensional rectangular box and analyze the influence of friction between the particles on segregation. The packed bed is stirred by a rectangular bar moving periodically in the horizontal direction. The parameters were introduced to characterise the segregation and mixing intensities, and a differential equation was proposed to describe the evolution of segregation intensity approaching exponentially a certain steady state value. It was found that the dynamic friction coefficient has a non-monotonous influence on the processes of mixing and size segregation in poly-disperse granular systems. Critical value of the dynamic friction coefficient μ_crit was identified. For the values of friction μ > μ_crit, behaviour of granular material can be characterised as a “laminar” flow with dominating convective motion of packed bed. For values of friction μ < μ_crit, behaviour of granular mattter can be characterised as “turbulent” flow with dominating “local” mixing inside the packed bed.     

Transport study of a novel polyfluorene/poly(p-phenylenevinylene) copolymer by various mobility models

The polyfluorene/poly(p-phenylenevinylene) copolymer based hole-only devices are fabricated and the current–voltage characteristics are measured as a function of temperature. The hole current is fitted well with space-charge limited and field-dependent mobility model, which provides a direct measurement of the hole mobility μ as a function of electric field E and temperature. The mobility is fitted with existing Gill’s model, Gaussian disorder model, correlated Gaussian disorder model and Brownian motion model. Energy hopping time and activation energy are obtained from Brownian motion model. Microscopic transport parameters are derived and a consistent picture of the influence of the molecular structure of the polymer on the charge transport is depicted. For the polyfluorene/poly(p-phenylenevinylene) copolymer, although with a high degree of irregularity in structure and larger energetic disorder, the two bulky structure favors charge delocalization and remove defect sites, results in a higher mobility. The results suggest space-charge limited and field-dependent mobility model combine with various mobility model, include Brownian motion model, is a useful technique to study charge transport in thin films with thicknesses close to those used in real devices.     

Slip flow and heat transfer in rectangular and circular microchannels using hybrid FE/FV method

Rarefied gas flows typically encountered in MEMS systems are numerically investigated in this study. Fluid flow and heat transfer in rectangular and circular microchannels within the slip flow regime are studied in detail by our recently developed implicit, incompressible, hybrid (finite element/finite volume) flow solver. The hybrid flow solver methodology is based on the pressure correction or projection method, which involves a fractional step approach to obtain an intermediate velocity field by solving the original momentum equations with the matrix‐free, implicit, cell‐centered finite volume method. The Poisson equation resulting from the fractional step approach is then solved by node based Galerkin finite element method for an auxiliary variable, which is closely related to pressure and is used to update the velocity field and pressure field. The hybrid flow solver has been extended for applications in MEMS by incorporating first order slip flow boundary conditions. Extended inlet boundary conditions are used for rectangular microchannels, whereas classical inlet boundary conditions are used for circular microchannels to emphasize on the entrance region singularity. In this study, rarefaction effects characterized by Knudsen number (Kn) in the range of 0 ⩽ Kn ⩽ 0.1 are numerically investigated for rectangular and circular microchannels with constant wall temperature. Extensive validations of our hybrid code are performed with available analytical solutions and experimental data for fully developed velocity profiles, friction factors, and Nusselt numbers. The influence of rarefaction on rectangular microchannels with aspect ratios between 0 and 1 is thoroughly investigated. Friction coefficients are found to be decreasing with increasing Knudsen number for both rectangular and circular microchannels. The reduction in the friction coefficients is more pronounced for rectangular microchannels with smaller aspect ratios. Effects of rarefaction and gas‐wall surface interaction parameter on heat transfer are analyzed for rectangular and circular microchannels. For most engineering applications, heat transfer is decreased with rarefaction. However, for fluids with very large Prandtl numbers, velocity slip dominates the temperature jump resulting in an increase in heat transfer with rarefaction. Depending on the gas‐wall surface interaction properties, extreme reductions in the Nusselt number can occur. Present results confirm the existence of a transition point below and above wherein heat transfer enhancement and reduction can occur. Copyright © 2011 John Wiley & Sons, Ltd.     

Aerodynamic properties of a wing performing unsteady rotational motions at low Reynolds number

Summary The aerodynamic forces and flow structures of a wing of relatively small aspect ratio in some unsteady rotational motions at low Reynolds number (Re=100) are studied by numerically solving the Navier-Stokes equations. These motions include a wing in constant-speed rotation after a fast start, wing accelerating and decelerating from one rotational speed to another, and wing rapidly pitching-up in constant speed rotation. When a wing performs a constant-speed rotation at small Reynolds number after started from rest at large angle of attack (=35°), a large lift coefficient can be maintained. The mechanism for the large lift coefficient is that for a rotating wing: the variation of the relative velocity along the wing-span causes a pressure gradient and hence a spanwise flow which can prevent the dynamic stall vortex from shedding. When a wing is rapidly accelerating or decelerating from one rotational speed to another, or rapidly pitching-up during constant speed rotation, even if the aspect ratio of the wing is small and the flow Reynolds number is low, a large aerodynamic force can be obtained. During these rapid unsteady motions, new layers of strong vorticity are formed near the wing surfaces in very short time, resulting in a large time rate of change of the fluid impulse which is responsible for the generation of the large aerodynamic force.     

Instabilities of two-phase flows in short rectangular microchannels

Two-phase (liquid-gas) flows in a short horizontal slit channel of rectangular cross section with heights (thicknesses) from 100 to 500 μm have been experimentally studied using the laser-induced fluorescence and schlieren photography methods. It is established that the formation of various two-phase flow regimes and the transitions between different regimes are determined by instabilities of the liquid-gas flow in the side parts of a channel. In a 100-μm-thick channel, a frontal instability has been observed during the liquid-gas interaction in the region of liquid output from the nozzle.     

Self-assembly of polystyrene microspheres within spatially confined rectangular microgrooves

A convective self-assembly of mono-sized polystyrene spheres with diameters ranging from 262 to 1000 nm was conducted on patterned silicon wafers with one-dimensional, periodic rectangular microgrooves of different widths (0.65–6 μm). The latex beads were driven into the spatially confined microgrooves by the capillary interactions and the confined wall during solvent evaporation, resulting in a range of packing structures. Processing variables including evaporation temperature, particle size (D), groove width (W), and groove height (H) were examined experimentally, and geometrical models were proposed to explain the various packing structures obtained. The degree of spatial freedom for the particles to rearrange themselves in the confined channels is found critical to the assembled particle-packing structure.