Dissipative particle dynamics simulation of shear flow in a microchannel with a deformable membrane

Thin deformable membranes are encountered in a number of microfluidics-based applications. These are often employed for enhancing sorting, mixing, cross-diffusion transport, etc. Microfluidic systems with deformable membranes can be better understood by employing simple models and efficient computational procedures. In this paper, we present a dissipative particle dynamics model to simulate the interaction between a deformable membrane and fluid flow in a two-dimensional microchannel. The membrane is modeled as a bead-spring system with both extensional and torsional springs to simulate extensional stiffness and bending rigidity, respectively. By performing detailed simulations on a membrane pinned at both ends and oriented parallel to the flow, we observe different steady state conformations. These membrane deflections are found to be relatively large for low bending stiffnesses and small for high stiffnesses. The membrane was found to exhibit a simple bowing out mode for high stiffness values and more complex conformations at lower stiffnesses.     

Dissipative particle dynamics simulation of field-dependent DNA mobility in nanoslits

The dynamics of DNA molecules in highly confined nanoslits under varying electric fields are studied using dissipative particle dynamics method, and our results show that manipulation of the electrical field can strongly influence DNA mobility. The mobility of DNA μ scales with electric field E as $ \mu = \mu^{\text{H}} - k_{1} e^{{ - E/E_{c} }} . $ And the data points for different DNA lengths finally approach each other in strong fields, which suggest that the sensitivity to chain length is almost lost. To explain the unusual field-dependent phenomena, we analyze the time evolution of DNA configurations under different fields. For strong driving potentials when the system is dominated by the electric driving force, the DNA chains are more likely to hold coiled configurations. For weak driving potential when the random diffusion forces dominate, we see frequent dynamic transitions between stretched and coiled configuration, which may increase the drag resistance, therefore reduce the mobility.     

Dissipative particle dynamics simulations of electroosmotic flow in nano-fluidic devices

When modeling the hydrodynamics of nanofluidic systems, it is often essential to include molecular-level information such as molecular fluctuations. To this effect, we present a mesoscopic approach which combines a fluctuating hydrodynamics formulation with an efficient implementation of Electroosmotic flow (EOF) in the small Debye length limit. The resulting approach, whose major ingredient is Dissipative Particle Dynamics, is sufficiently coarse-grained to allow efficient simulation of the hydrodynamics of micro/nanofluidic devices of sizes that are too large to be simulated by ab initio methods such as Molecular Dynamics. Within our formulation, EOF is efficiently generated using the recently proven similitude between velocity and electric field under appropriate conditions. More specifically, EOF is generated using an effective boundary condition, akin to a moving wall, thus avoiding evaluation of the computationally expensive electrostatic forces. Our method is used for simulating EOFs and DNA molecular sieving in simple and complex two-dimensional (2D) and 3D geometries frequently used in nano-fluidic devices. The numerical data obtained from our model are in very good agreement with theoretical results.     

圆柱绕流的二维数值模拟和尾迹分析

为指导机械设计中参数和布局的选择,研究固定在水流中的圆柱结构件的受力情况和流场分布．利用FLUENT中的三种湍流模型对雷诺数为3900的圆柱绕流进行二维数值模拟并进行对比,得到升力因数、阻力因数、分离角、斯特劳哈尔数和涡街尺寸等参数的模拟结果,与参考文献中的实验结果对比验证二维模拟的预测精度．RKE（Realizable k-ε）和雷诺应力模型（Reynolds Stress Model,RSM）均能在此雷诺数下得出接近实验结果的流场,RSM模型使用POWER LAW离散格式的结果优于QUICK格式．与三维模拟的对比表明二维模拟适合在设计初期的快速估算,能够快速得到合适精度的模拟结果．     

Dissipative particle dynamics simulations of liquid nanojet breakup

In this work, we use a dissipative-particle-dynamics (DPD)-based two-phase model to study the breakup of liquid nanojets. We show that the breakup of nanojets is accelerated by the presence of thermal fluctuations. Satellite drops, which are undesirable from an application viewpoint, are not observed in our simulations. We find that the presence of a repulsive wall enclosing the DPD liquid is necessary to prevent clogging of nanojets. We are able to recover the time evolution of minimum jet radius as given by prior theoretical analysis. This study shows that DPD is able to capture the thermal induced breakup phenomena at the sub-micron level. The coarse-grained nature of DPD along with its flexibility to allow for the modeling of complex fluids in combination with the results from this study show that DPD is a useful tool for sub-micron fluid flow simulations.     

Fully explicit dissipative particle dynamics simulation of electroosmotic flow in nanochannels

A fully explicit mesoscale simulation of electroosmotic flow (EOF) in nanochannels is presented by an extended dissipative particle dynamics (DPD) method. To avoid formation of ionic pairs through interacting soft-core charges, a Slater-type smearing distribution borrowed from quantum mechanics is utilized to surround each soft DPD ion with a charge cloud. To account for reduced periodicity normal to the walls direction, a corrected version of 3D Ewald sum is implemented in which a dipole moment term is deducted from energy and force terms of non-frozen charges. Simulation box is then elongated normal to walls to dampen spurious interslab interactions by adding vacuum gaps between periodic images. These measures together with the established unit conversions guarantee perfect match to molecular dynamics results. The transition of EOF velocity profile from parabolic (equivalent to overlap of electric double layers) to plug-like shapes is studied across the changing electric field between 0.06 and 0.41 [V/nm], and varying salt concentration from 0.26 to 2.0 [M]. It is found that 1.25 [V] increase in the driving voltage can potentially enhance the electroosmotic flow rate by 8–11 times in the range of ionic concentrations studied. The range of surface zeta potential calculated as \( 27 < \zeta < 52 \) [mV] in the linear response regime, as identified to occur for 0.24 ≤ E [V/nm], agrees reasonably with numerical and experimental studies.     

Steady laminar flow through a twisted pipe of elliptical cross-section

Previous work by the author on a formal series method of solution for flows in pipes with slowly varying geometries which provides a basis for approximate perturbation solutions is applied to the specific example of flow in a slowly twisted pipe of elliptical cross-section. With the aid of the algebraic and symbolic manipulation programming language REDUCE the calculations to low order have been carried out to reveal some purely three-dimensional effects.     

Numerical investigation of dynamics of elliptical magnetic microparticles in shear flows

We study the rotational dynamics of magnetic prolate elliptical particles in a simple shear flow subjected to a uniform magnetic field, using direct numerical simulations based on the finite element method. Focusing on paramagnetic and ferromagnetic particles, we investigate the effects of the magnetic field strength and direction on their rotational dynamics. In the weak field regime (below a critical field strength), the particles are able to perform complete rotations, and the symmetry property of particle rotational speed is influenced by the direction and strength of the magnetic field. In the strong field regime (above a critical strength), the particles are pinned at steady angles. The steady angle depends on both the direction and strength of the magnetic field. Our results show that paramagnetic and ferromagnetic particles exhibit markedly different rotational dynamics in a uniform magnetic field. The numerical findings are in good agreement with theoretical prediction. Our numerical investigation further reveals drastically different lateral migration behaviors of paramagnetic and ferromagnetic particles in a wall-bounded simple shear flow under a uniform magnetic field. These two kinds of particles can thus be separated by combining a shear flow and a uniform magnetic field. We also study the lateral migration of paramagnetic and ferromagnetic particles in a pressure-driven flow (a more practical flow configuration in microfluidics), and observe similar lateral migration behaviors. These findings demonstrate a simple but useful way to manipulate non-spherical microparticles in microfluidic devices.     

Molecular dynamics simulation on flow behavior of nanofluids between flat plates under shear flow condition

Molecular dynamic model of nanofluid between flat plates under shear flow conditions was built. The nanofluid model consisted of 12 spherical copper nanoparticles with each particle diameter of 4 nm and argon atoms as base liquid. The Lennard–Jones (LJ) potential function was adopted to deal with the interactions between atoms. Thus, the motion states of nanoparticles during the process of flowing were obtained and the flow behaviors of nanofluid between flat plates at different moments could be analyzed. The simulation results showed that an absorption layer of argon atoms existed surrounding each nanoparticle and would accompany with the particle to move. The absorption layer contributed little to the flow of nanoparticles but much to the heat transferring in nanofluids. Another phenomenon observed during shear flowing process was that the nanoparticles would vibrate and rotate besides main flowing with liquid argon and these micro-motions could strengthen partial flowing in nanofluids.     

Dissipative particle dynamics simulations for fibre suspensions in newtonian and viscoelastic fluids

A versatile model of fibre suspensions in Newtonian and viscoelastic fluids has been developed using dissipative particle dynamics method. The viscoelastic fluid is modelled by linear chains with linear connector spring force (the Oldroyd-B model), which is known to be a reasonable model for the so-called Boger fluid (a dilute suspension of polymer in a highly viscous solvent). The numerical results are in excellent agreement with the analytical results of the Oldroyd-B model in simple shear flow. An effective meso-scale model of fibre in DPD is proposed and then incorporated with simple Newtonian fluid and our Boger fluid to enable entirely study rheological properties of fibre suspensions in both Newtonian and viscoelastic solvents. The numerical results are well compared with available experimental data and other numerical models.     

两亲性肽纳米纤维聚集的耗散粒子动力学模拟

根据PA分子基团的化学性质和粒子大小,将PA分子分成DPD粒子,将适当大小的水簇视为溶剂粒子,克服了模拟技巧的困难,首次用耗散粒子动力学(dissipative particle dynamics,DPD)方法模拟了两亲性肽分子PA的自组装相行为.模拟结果表明:在水溶剂的辅助下,在三维纤维状的圆柱聚合体中,这些分子能进行自组装.DPD的模拟结果受溶剂珠粒的大小、温度及PA与溶剂的比例等多种因素的影响.研究同时发现,在圆筒状纤维中,使用5～9个水簇的溶剂粒子可得到PA分子的聚合体.形成纳米纤维聚合体的条件是PA与水珠粒的比值大于1∶6,温度高于340 K.筒状聚合体直径的估计值符合实验测量值.     

The advance from 2D electrostatic to 3D electromagnetic particle simulation

The evolution of a 3D, E-M code from earlier less ambitious codes (and their critical analyses) is presented. The code simulates the interaction of some 10⁶ superparticles with over 10⁵ “superphotons” (field modes in k-space). Particle shaping is done in k-space, isotropically. Interpolation of the fields and their excitation is done with optimally fitted quadratic splines, using a 64³ grid. The field modes are updated with no dispersion error, the particles by a time-centered (optionally relativistic) algorithm. Data management conflicts, arising from the global nature of the field information versus the localization of particle processing, are resolved by means of the partial random accessibility of external storage. Results of trial runs with a uniform magnetised plasma show the simultaneous presence of such widely differing phenomena as Alfvén waves, Whistlers, Landau damping, trapping by electrostatic waves, etc., in the computer model.     

Non-newtonian laminar 2D swirl flow design by the topology optimization method

Structural and Multidisciplinary Optimization - The performance of fluid devices, such as channels, valves, nozzles, and pumps, may be improved by designing them through the topology optimization...     

Numerical simulation of capsule deformation in simple shear flow

The transient deformation of two-dimensional non-circular and three-dimensional non-spherical capsules in simple shear flow is studied numerically, using the hybrid immersed boundary and multi-block lattice Boltzmann method recently proposed by the present authors. The capsules are modeled as Newtonian liquid drops enclosed by elastic membranes; the liquids inside and outside the capsule have the same physical properties. The present results show important different behaviors between two-dimensional and three-dimensional capsules in shear flow. For two-dimensional non-circular capsules without considering the membrane bending rigidity, or considering the bending with the minimum bending-energy configurations (shapes at which the bending-energy has a global minimum) being uniform-curvature shapes, the capsules will always achieve the steady tank treading motion (a capsule deforms to a steady shape with a steady inclination and the membrane rotates around the liquid inside). However, for three-dimensional non-spherical capsules without membrane bending rigidity, such a steady mode does not exist; with the shear rate decreasing, the three-dimensional capsules’ motion changes from swinging mode (a capsule undergoes periodic shape deformation and inclination oscillation while its membrane is rotating around the liquid inside) to flipping mode. The deformation of two-dimensional capsules, with their initial non-circular shapes taken as the minimum bending-energy configurations, is also considered. It is quite interesting to find such two-dimensional capsules behave qualitatively similar to three-dimensional capsules: there is a swinging-to-flipping transition induced by lowering the shear rate.     

Analysis of mixed motion in deterministic ratchets via experiment and particle simulation

Deterministic lateral displacement (DLD) ratchets are microfluidic devices, which are used for size-based sorting of cells or DNA. Based on their size, particles are showing different kinds of motion, leading to their fractionation. In earlier studies, so-called zigzag and displacement motions are observed, and in recent study by our group (Kulrattanarak et al., Meas Sci Technol, 2010a; J Colloid Interface Sci, 2010b), we have shown that also mixed motion occurs, which is an irregular alternation of zigzag and displacement motion. We have shown that the mixed motion is due to asymmetry of the flow lane distribution, induced by the symmetry breaking of the oblique primitive lattice cell (Kulrattanarak et al. 2010b). In this study, we investigate mixed motion in depth by numerical and experimental analysis. Via 3D simulations, we have computed explicit particle trajectories in DLD, and are able to show that there are two critical length scales determining the type of motion. The first length scale d _f,1 is the first flow lane width, which determines the transition between zigzag motion and mixed motion. The other length scale, d _f,c , determines the transition between mixed motion and displacement motion. Based on our experimental and numerical results we have been able to correlate the migration angle of particles showing mixed motion to the particle size, relative to the two critical length scales d _f,1 and d _f,c .     

Sudden expansions in circular microchannels: flow dynamics and pressure drop

Micro particle shadow velocimetry is used to study the flow of water through microcircular sudden expansions of ratios e = 1.51 and e = 1.96 for inlet Reynolds numbers Re _d < 120. Such flows give rise to annular vortices, trapped downstream of the expansions. The dependency of the vortex length on the Reynolds number Re _d and the expansion ratio e is experimentally investigated in this study. Additionally, the shape of the axisymmetric annular vortex is quantified based on the visualization results. These measurements favorably follow the trends reported for larger scales in the literature. Redevelopment of the confined jet to the fully developed Poiseuille flow downstream of the expansion is also studied quantitatively. Furthermore, the experimentally resolved velocities are used to calculate high resolution static pressure gradient distributions along the channel walls. These measurements are then integrated into the axisymmetric momentum and energy balance equations, for the flow downstream of the expansion, to obtain the irreversible pressure drop in this geometry. As expected, the measured pressure drop coefficients for the range of Reynolds numbers studied here do not match the predictions of the available empirical correlations, which are commonly based turbulent flow studies. However, these results are in excellent agreement with previous numerical calculations. The pressure drop coefficient is found to strongly depend on the inlet Reynolds number for Re _d < 50. Although no length-scale effect is observed for the range of channel diameters studied here, for Reynolds numbers Re _d < 50, which are typical in microchannel applications, complex nonlinear trends in the flow dynamics and pressure drop measurements are discovered and discussed in this work.     

Rotational motion and lateral migration of an elliptical magnetic particle in a microchannel under a uniform magnetic field

We have numerically investigated the motion of an elliptical magnetic particle in a microfluidic channel subjected to an external uniform magnetic field. By using the direct numerical simulation method and an arbitrary Lagrangian–Eulerian technique, the involved particle–fluid-magnetic field problem can be solved in a fully coupled manner. The numerical predictions of the particle trajectory and orientation with and without a uniform magnetic field are in qualitative agreement with the existing experimental results, and numerical results have revealed the impacts of key parameters such as inlet flow velocity, magnetic field direction, and particle shape on the rotational motion and lateral migration of the elliptical particle. Meanwhile, the shape-based particle separation in a low Reynolds number flow with the aid of an applied uniform magnetic field has also been numerically demonstrated.     

End-wall effects on vortex shedding in planar shear flow over a circular cylinder

A prevailing controversy regarding the suppression of periodic vortex shedding from a circular cylinder embedded in a planar shear flow has been addressed. Three-dimensional computer simulations utilizing the advanced MGLET software [11] demonstrated the importance of the end-wall conditions. Earlier results from two-dimensional simulations at Re = 100 were reproduced only with free-slip conditions. With no-slip conditions imposed at one or both end-walls, the vortex shedding was suppressed near the no-slip boundary and the shedding pattern was substantially affected even at mid-span. The Strouhal number decreased when the shear-rate parameter was increased from 0.1 to 0.2, irrespective of the choice of boundary conditions.     

Numerical simulations of particle migration in a viscoelastic fluid subjected to shear flow

Particle migration is a relevant transport mechanism whenever suspensions flow in channels with gap size comparable to particle dimensions (e.g. microfluidic devices). Several theoretical as well as experimental studies have been performed on this topic, showing that the occurring of this phenomenon and the migration direction are related to particle size, flow rate, and the nature of the suspending liquid.In this work we perform a systematic analysis on the migration of a single particle in a sheared viscoelastic fluid through 2D finite element simulations in a Couette planar geometry. To focus on the effects of viscoelasticity alone, inertia is neglected. The suspending medium is modeled as a Giesekus fluid.An ALE particle mover is combined with a DEVSS/SUPG formulation with a log-representation of the conformation tensor giving stable and convergent results up to high flow rates. To optimize the computational effort and reduce the remeshing and projection steps, needed as soon as the mesh becomes too distorted, a ‘backprojection’ of the flow fields is performed, through which the particle in fact moves along the cross-streamline direction only, and the mesh distortion is hence drastically reduced.Our results, in agreement with recent experimental data, show a migration towards the closest walls, regardless of the fluid and geometrical parameters. The phenomenon is enhanced by the fluid elasticity, the shear thinning and strong confinements. The migration velocity trends show the existence of a master curve governing the particle dynamics in the whole channel. Three different regimes experienced by the particle are recognized, related to the particle-wall distance. The existence of a unique migration behavior and its qualitative aspects do not change by varying the fluid parameters or the particle size.