Lab-on-a-Chip 시스템의 혈류 역학에 대한 검토: 엔지니어링 관점

Review on Blood Flow Dynamics in Lab-on-a-Chip Systems: An Engineering Perspective

  • Bin-Jie Lai
  • Li-Tao Zhu
  • Zhe Chen*
  • Bo Ouyang*
  • , and 
  • Zheng-Hong Luo*

Abstract

다양한 수송 메커니즘 하에서, “LOC(lab-on-a-chip)” 시스템에서 유동 전단 속도 조건과 밀접한 관련이 있는 혈류 역학은 다양한 수송 현상을 초래하는 것으로 밝혀졌습니다.

본 연구는 적혈구의 동적 혈액 점도 및 탄성 거동과 같은 점탄성 특성의 역할을 통해 LOC 시스템의 혈류 패턴을 조사합니다. 모세관 및 전기삼투압의 주요 매개변수를 통해 LOC 시스템의 혈액 수송 현상에 대한 연구는 실험적, 이론적 및 수많은 수치적 접근 방식을 통해 제공됩니다.

전기 삼투압 점탄성 흐름에 의해 유발되는 교란은 특히 향후 연구 기회를 위해 혈액 및 기타 점탄성 유체를 취급하는 LOC 장치의 혼합 및 분리 기능 향상에 논의되고 적용됩니다. 또한, 본 연구는 보다 정확하고 단순화된 혈류 모델에 대한 요구와 전기역학 효과 하에서 점탄성 유체 흐름에 대한 수치 연구에 대한 강조와 같은 LOC 시스템 하에서 혈류 역학의 수치 모델링의 문제를 식별합니다.

전기역학 현상을 연구하는 동안 제타 전위 조건에 대한 보다 실용적인 가정도 강조됩니다. 본 연구는 모세관 및 전기삼투압에 의해 구동되는 미세유체 시스템의 혈류 역학에 대한 포괄적이고 학제적인 관점을 제공하는 것을 목표로 한다.

KEYWORDS: 

1. Introduction

1.1. Microfluidic Flow in Lab-on-a-Chip (LOC) Systems

Over the past several decades, the ability to control and utilize fluid flow patterns at microscales has gained considerable interest across a myriad of scientific and engineering disciplines, leading to growing interest in scientific research of microfluidics. 

(1) Microfluidics, an interdisciplinary field that straddles physics, engineering, and biotechnology, is dedicated to the behavior, precise control, and manipulation of fluids geometrically constrained to a small, typically submillimeter, scale. 

(2) The engineering community has increasingly focused on microfluidics, exploring different driving forces to enhance working fluid transport, with the aim of accurately and efficiently describing, controlling, designing, and applying microfluidic flow principles and transport phenomena, particularly for miniaturized applications. 

(3) This attention has chiefly been fueled by the potential to revolutionize diagnostic and therapeutic techniques in the biomedical and pharmaceutical sectorsUnder various driving forces in microfluidic flows, intriguing transport phenomena have bolstered confidence in sustainable and efficient applications in fields such as pharmaceutical, biochemical, and environmental science. The “lab-on-a-chip” (LOC) system harnesses microfluidic flow to enable fluid processing and the execution of laboratory tasks on a chip-sized scale. LOC systems have played a vital role in the miniaturization of laboratory operations such as mixing, chemical reaction, separation, flow control, and detection on small devices, where a wide variety of fluids is adapted. Biological fluid flow like blood and other viscoelastic fluids are notably studied among the many working fluids commonly utilized by LOC systems, owing to the optimization in small fluid sample volumed, rapid response times, precise control, and easy manipulation of flow patterns offered by the system under various driving forces. 

(4)The driving forces in blood flow can be categorized as passive or active transport mechanisms and, in some cases, both. Under various transport mechanisms, the unique design of microchannels enables different functionalities in driving, mixing, separating, and diagnosing blood and drug delivery in the blood. 

(5) Understanding and manipulating these driving forces are crucial for optimizing the performance of a LOC system. Such knowledge presents the opportunity to achieve higher efficiency and reliability in addressing cellular level challenges in medical diagnostics, forensic studies, cancer detection, and other fundamental research areas, for applications of point-of-care (POC) devices. 

(6)

1.2. Engineering Approach of Microfluidic Transport Phenomena in LOC Systems

Different transport mechanisms exhibit unique properties at submillimeter length scales in microfluidic devices, leading to significant transport phenomena that differ from those of macroscale flows. An in-depth understanding of these unique transport phenomena under microfluidic systems is often required in fluidic mechanics to fully harness the potential functionality of a LOC system to obtain systematically designed and precisely controlled transport of microfluids under their respective driving force. Fluid mechanics is considered a vital component in chemical engineering, enabling the analysis of fluid behaviors in various unit designs, ranging from large-scale reactors to separation units. Transport phenomena in fluid mechanics provide a conceptual framework for analytically and descriptively explaining why and how experimental results and physiological phenomena occur. The Navier–Stokes (N–S) equation, along with other governing equations, is often adapted to accurately describe fluid dynamics by accounting for pressure, surface properties, velocity, and temperature variations over space and time. In addition, limiting factors and nonidealities for these governing equations should be considered to impose corrections for empirical consistency before physical models are assembled for more accurate controls and efficiency. Microfluidic flow systems often deviate from ideal conditions, requiring adjustments to the standard governing equations. These deviations could arise from factors such as viscous effects, surface interactions, and non-Newtonian fluid properties from different microfluid types and geometrical layouts of microchannels. Addressing these nonidealities supports the refining of theoretical models and prediction accuracy for microfluidic flow behaviors.

The analytical calculation of coupled nonlinear governing equations, which describes the material and energy balances of systems under ideal conditions, often requires considerable computational efforts. However, advancements in computation capabilities, cost reduction, and improved accuracy have made numerical simulations using different numerical and modeling methods a powerful tool for effectively solving these complex coupled equations and modeling various transport phenomena. Computational fluid dynamics (CFD) is a numerical technique used to investigate the spatial and temporal distribution of various flow parameters. It serves as a critical approach to provide insights and reasoning for decision-making regarding the optimal designs involving fluid dynamics, even prior to complex physical model prototyping and experimental procedures. The integration of experimental data, theoretical analysis, and reliable numerical simulations from CFD enables systematic variation of analytical parameters through quantitative analysis, where adjustment to delivery of blood flow and other working fluids in LOC systems can be achieved.

Numerical methods such as the Finite-Difference Method (FDM), Finite-Element-Method (FEM), and Finite-Volume Method (FVM) are heavily employed in CFD and offer diverse approaches to achieve discretization of Eulerian flow equations through filling a mesh of the flow domain. A more in-depth review of numerical methods in CFD and its application for blood flow simulation is provided in Section 2.2.2.

1.3. Scope of the Review

In this Review, we explore and characterize the blood flow phenomena within the LOC systems, utilizing both physiological and engineering modeling approaches. Similar approaches will be taken to discuss capillary-driven flow and electric-osmotic flow (EOF) under electrokinetic phenomena as a passive and active transport scheme, respectively, for blood transport in LOC systems. Such an analysis aims to bridge the gap between physical (experimental) and engineering (analytical) perspectives in studying and manipulating blood flow delivery by different driving forces in LOC systems. Moreover, the Review hopes to benefit the interests of not only blood flow control in LOC devices but also the transport of viscoelastic fluids, which are less studied in the literature compared to that of Newtonian fluids, in LOC systems.

Section 2 examines the complex interplay between viscoelastic properties of blood and blood flow patterns under shear flow in LOC systems, while engineering numerical modeling approaches for blood flow are presented for assistance. Sections 3 and 4 look into the theoretical principles, numerical governing equations, and modeling methodologies for capillary driven flow and EOF in LOC systems as well as their impact on blood flow dynamics through the quantification of key parameters of the two driving forces. Section 5 concludes the characterized blood flow transport processes in LOC systems under these two forces. Additionally, prospective areas of research in improving the functionality of LOC devices employing blood and other viscoelastic fluids and potentially justifying mechanisms underlying microfluidic flow patterns outside of LOC systems are presented. Finally, the challenges encountered in the numerical studies of blood flow under LOC systems are acknowledged, paving the way for further research.

2. Blood Flow Phenomena

ARTICLE SECTIONS

Jump To


2.1. Physiological Blood Flow Behavior

Blood, an essential physiological fluid in the human body, serves the vital role of transporting oxygen and nutrients throughout the body. Additionally, blood is responsible for suspending various blood cells including erythrocytes (red blood cells or RBCs), leukocytes (white blood cells), and thrombocytes (blood platelets) in a plasma medium.Among the cells mentioned above, red blood cells (RBCs) comprise approximately 40–45% of the volume of healthy blood. 

(7) An RBC possesses an inherent elastic property with a biconcave shape of an average diameter of 8 μm and a thickness of 2 μm. This biconcave shape maximizes the surface-to-volume ratio, allowing RBCs to endure significant distortion while maintaining their functionality. 

(8,9) Additionally, the biconcave shape optimizes gas exchange, facilitating efficient uptake of oxygen due to the increased surface area. The inherent elasticity of RBCs allows them to undergo substantial distortion from their original biconcave shape and exhibits high flexibility, particularly in narrow channels.RBC deformability enables the cell to deform from a biconcave shape to a parachute-like configuration, despite minor differences in RBC shape dynamics under shear flow between initial cell locations. As shown in Figure 1(a), RBCs initiating with different resting shapes and orientations displaying display a similar deformation pattern 

(10) in terms of its shape. Shear flow induces an inward bending of the cell at the rear position of the rim to the final bending position, 

(11) resulting in an alignment toward the same position of the flow direction.

Figure 1. Images of varying deformation of RBCs and different dynamic blood flow behaviors. (a) The deforming shape behavior of RBCs at four different initiating positions under the same experimental conditions of a flow from left to right, (10) (b) RBC aggregation, (13) (c) CFL region. (18) Reproduced with permission from ref (10). Copyright 2011 Elsevier. Reproduced with permission from ref (13). Copyright 2022 The Authors, under the terms of the Creative Commons (CC BY 4.0) License https://creativecommons.org/licenses/by/4.0/. Reproduced with permission from ref (18). Copyright 2019 Elsevier.

The flexible property of RBCs enables them to navigate through narrow capillaries and traverse a complex network of blood vessels. The deformability of RBCs depends on various factors, including the channel geometry, RBC concentration, and the elastic properties of the RBC membrane. 

(12) Both flexibility and deformability are vital in the process of oxygen exchange among blood and tissues throughout the body, allowing cells to flow in vessels even smaller than the original cell size prior to deforming.As RBCs serve as major components in blood, their collective dynamics also hugely affect blood rheology. RBCs exhibit an aggregation phenomenon due to cell to cell interactions, such as adhesion forces, among populated cells, inducing unique blood flow patterns and rheological behaviors in microfluidic systems. For blood flow in large vessels between a diameter of 1 and 3 cm, where shear rates are not high, a constant viscosity and Newtonian behavior for blood can be assumed. However, under low shear rate conditions (0.1 s

–1) in smaller vessels such as the arteries and venules, which are within a diameter of 0.2 mm to 1 cm, blood exhibits non-Newtonian properties, such as shear-thinning viscosity and viscoelasticity due to RBC aggregation and deformability. The nonlinear viscoelastic property of blood gives rise to a complex relationship between viscosity and shear rate, primarily influenced by the highly elastic behavior of RBCs. A wide range of research on the transient behavior of the RBC shape and aggregation characteristics under varied flow circumstances has been conducted, aiming to obtain a better understanding of the interaction between blood flow shear forces from confined flows.

For a better understanding of the unique blood flow structures and rheological behaviors in microfluidic systems, some blood flow patterns are introduced in the following section.

2.1.1. RBC Aggregation

RBC aggregation is a vital phenomenon to be considered when designing LOC devices due to its impact on the viscosity of the bulk flow. Under conditions of low shear rate, such as in stagnant or low flow rate regions, RBCs tend to aggregate, forming structures known as rouleaux, resembling stacks of coins as shown in Figure 1(b). 

(13) The aggregation of RBCs increases the viscosity at the aggregated region, 

(14) hence slowing down the overall blood flow. However, when exposed to high shear rates, RBC aggregates disaggregate. As shear rates continue to increase, RBCs tend to deform, elongating and aligning themselves with the direction of the flow. 

(15) Such a dynamic shift in behavior from the cells in response to the shear rate forms the basis of the viscoelastic properties observed in whole blood. In essence, the viscosity of the blood varies according to the shear rate conditions, which are related to the velocity gradient of the system. It is significant to take the intricate relationship between shear rate conditions and the change of blood viscosity due to RBC aggregation into account since various flow driving conditions may induce varied effects on the degree of aggregation.

2.1.2. Fåhræus-Lindqvist Effect

The Fåhræus–Lindqvist (FL) effect describes the gradual decrease in the apparent viscosity of blood as the channel diameter decreases. 

(16) This effect is attributed to the migration of RBCs toward the central region in the microchannel, where the flow rate is higher, due to the presence of higher pressure and asymmetric distribution of shear forces. This migration of RBCs, typically observed at blood vessels less than 0.3 mm, toward the higher flow rate region contributes to the change in blood viscosity, which becomes dependent on the channel size. Simultaneously, the increase of the RBC concentration in the central region of the microchannel results in the formation of a less viscous region close to the microchannel wall. This region called the Cell-Free Layer (CFL), is primarily composed of plasma. 

(17) The combination of the FL effect and the following CFL formation provides a unique phenomenon that is often utilized in passive and active plasma separation mechanisms, involving branched and constriction channels for various applications in plasma separation using microfluidic systems.

2.1.3. Cell-Free Layer Formation

In microfluidic blood flow, RBCs form aggregates at the microchannel core and result in a region that is mostly devoid of RBCs near the microchannel walls, as shown in Figure 1(c). 

(18) The region is known as the cell-free layer (CFL). The CFL region is often known to possess a lower viscosity compared to other regions within the blood flow due to the lower viscosity value of plasma when compared to that of the aggregated RBCs. Therefore, a thicker CFL region composed of plasma correlates to a reduced apparent whole blood viscosity. 

(19) A thicker CFL region is often established following the RBC aggregation at the microchannel core under conditions of decreasing the tube diameter. Apart from the dependence on the RBC concentration in the microchannel core, the CFL thickness is also affected by the volume concentration of RBCs, or hematocrit, in whole blood, as well as the deformability of RBCs. Given the influence CFL thickness has on blood flow rheological parameters such as blood flow rate, which is strongly dependent on whole blood viscosity, investigating CFL thickness under shear flow is crucial for LOC systems accounting for blood flow.

2.1.4. Plasma Skimming in Bifurcation Networks

The uneven arrangement of RBCs in bifurcating microchannels, commonly termed skimming bifurcation, arises from the axial migration of RBCs within flowing streams. This uneven distribution contributes to variations in viscosity across differing sizes of bifurcating channels but offers a stabilizing effect. Notably, higher flow rates in microchannels are associated with increased hematocrit levels, resulting in higher viscosity compared with those with lower flow rates. Parametric investigations on bifurcation angle, 

(20) thickness of the CFL, 

(21) and RBC dynamics, including aggregation and deformation, 

(22) may alter the varying viscosity of blood and its flow behavior within microchannels.

2.2. Modeling on Blood Flow Dynamics

2.2.1. Blood Properties and Mathematical Models of Blood Rheology

Under different shear rate conditions in blood flow, the elastic characteristics and dynamic changes of the RBC induce a complex velocity and stress relationship, resulting in the incompatibility of blood flow characterization through standard presumptions of constant viscosity used for Newtonian fluid flow. Blood flow is categorized as a viscoelastic non-Newtonian fluid flow where constitutive equations governing this type of flow take into consideration the nonlinear viscometric properties of blood. To mathematically characterize the evolving blood viscosity and the relationship between the elasticity of RBC and the shear blood flow, respectively, across space and time of the system, a stress tensor (τ) defined by constitutive models is often coupled in the Navier–Stokes equation to account for the collective impact of the constant dynamic viscosity (η) and the elasticity from RBCs on blood flow.The dynamic viscosity of blood is heavily dependent on the shear stress applied to the cell and various parameters from the blood such as hematocrit value, plasma viscosity, mechanical properties of the RBC membrane, and red blood cell aggregation rate. The apparent blood viscosity is considered convenient for the characterization of the relationship between the evolving blood viscosity and shear rate, which can be defined by Casson’s law, as shown in eq 1.

𝜇=𝜏0𝛾˙+2𝜂𝜏0𝛾˙⎯⎯⎯⎯⎯⎯⎯√+𝜂�=�0�˙+2��0�˙+�

(1)where τ

0 is the yield stress–stress required to initiate blood flow motion, η is the Casson rheological constant, and γ̇ is the shear rate. The value of Casson’s law parameters under blood with normal hematocrit level can be defined as τ

0 = 0.0056 Pa and η = 0.0035 Pa·s. 

(23) With the known property of blood and Casson’s law parameters, an approximation can be made to the dynamic viscosity under various flow condition domains. The Power Law model is often employed to characterize the dynamic viscosity in relation to the shear rate, since precise solutions exist for specific geometries and flow circumstances, acting as a fundamental standard for definition. The Carreau and Carreau–Yasuda models can be advantageous over the Power Law model due to their ability to evaluate the dynamic viscosity at low to zero shear rate conditions. However, none of the above-mentioned models consider the memory or other elastic behavior of blood and its RBCs. Some other commonly used mathematical models and their constants for the non-Newtonian viscosity property characterization of blood are listed in Table 1 below. 

(24−26)Table 1. Comparison of Various Non-Newtonian Models for Blood Viscosity 

(24−26)

ModelNon-Newtonian ViscosityParameters
Power Law(2)n = 0.61, k = 0.42
Carreau(3)μ0 = 0.056 Pa·s, μ = 0.00345 Pa·s, λ = 3.1736 s, m = 2.406, a = 0.254
Walburn–Schneck(4)C1 = 0.000797 Pa·s, C2 = 0.0608 Pa·s, C3 = 0.00499, C4 = 14.585 g–1, TPMA = 25 g/L
Carreau–Yasuda(5)μ0 = 0.056 Pa·s, μ = 0.00345 Pa·s, λ = 1.902 s, n = 0.22, a = 1.25
Quemada(6)μp = 0.0012 Pa·s, k = 2.07, k0 = 4.33, γ̇c = 1.88 s–1

The blood rheology is commonly known to be influenced by two key physiological factors, namely, the hematocrit value (H

t) and the fibrinogen concentration (c

f), with an average value of 42% and 0.252 gd·L

–1, respectively. Particularly in low shear conditions, the presence of varying fibrinogen concentrations affects the tendency for aggregation and rouleaux formation, while the occurrence of aggregation is contingent upon specific levels of hematocrit. 

(27) The study from Apostolidis et al. 

(28) modifies the Casson model through emphasizing its reliance on hematocrit and fibrinogen concentration parameter values, owing to the extensive knowledge of the two physiological blood parameters.The viscoelastic response of blood is heavily dependent on the elasticity of the RBC, which is defined by the relationship between the deformation and stress relaxation from RBCs under a specific location of shear flow as a function of the velocity field. The stress tensor is usually characterized by constitutive equations such as the Upper-Convected Maxwell Model 

(29) and the Oldroyd-B model 

(30) to track the molecule effects under shear from different driving forces. The prominent non-Newtonian features, such as shear thinning and yield stress, have played a vital role in the characterization of blood rheology, particularly with respect to the evaluation of yield stress under low shear conditions. The nature of stress measurement in blood, typically on the order of 1 mPa, is challenging due to its low magnitude. The occurrence of the CFL complicates the measurement further due to the significant decrease in apparent viscosity near the wall over time and a consequential disparity in viscosity compared to the bulk region.In addition to shear thinning viscosity and yield stress, the formation of aggregation (rouleaux) from RBCs under low shear rates also contributes to the viscoelasticity under transient flow 

(31) and thixotropy 

(32) of whole blood. Given the difficulty in evaluating viscoelastic behavior of blood under low strain magnitudes and limitations in generalized Newtonian models, the utilization of viscoelastic models is advocated to encompass elasticity and delineate non-shear components within the stress tensor. Extending from the Oldroyd-B model, Anand et al. 

(33) developed a viscoelastic model framework for adapting elasticity within blood samples and predicting non-shear stress components. However, to also address the thixotropic effects, the model developed by Horner et al. 

(34) serves as a more comprehensive approach than the viscoelastic model from Anand et al. Thixotropy 

(32) typically occurs from the structural change of the rouleaux, where low shear rate conditions induce rouleaux formation. Correspondingly, elasticity increases, while elasticity is more representative of the isolated RBCs, under high shear rate conditions. The model of Horner et al. 

(34) considers the contribution of rouleaux to shear stress, taking into account factors such as the characteristic time for Brownian aggregation, shear-induced aggregation, and shear-induced breakage. Subsequent advancements in the model from Horner et al. often revolve around refining the three aforementioned key terms for a more substantial characterization of rouleaux dynamics. Notably, this has led to the recently developed mHAWB model 

(35) and other model iterations to enhance the accuracy of elastic and viscoelastic contributions to blood rheology, including the recently improved model suggested by Armstrong et al. 

(36)

2.2.2. Numerical Methods (FDM, FEM, FVM)

Numerical simulation has become increasingly more significant in analyzing the geometry, boundary layers of flow, and nonlinearity of hyperbolic viscoelastic flow constitutive equations. CFD is a powerful and efficient tool utilizing numerical methods to solve the governing hydrodynamic equations, such as the Navier–Stokes (N–S) equation, continuity equation, and energy conservation equation, for qualitative evaluation of fluid motion dynamics under different parameters. CFD overcomes the challenge of analytically solving nonlinear forms of differential equations by employing numerical methods such as the Finite-Difference Method (FDM), Finite-Element Method (FEM), and Finite-Volume Method (FVM) to discretize and solve the partial differential equations (PDEs), allowing for qualitative reproduction of transport phenomena and experimental observations. Different numerical methods are chosen to cope with various transport systems for optimization of the accuracy of the result and control of error during the discretization process.FDM is a straightforward approach to discretizing PDEs, replacing the continuum representation of equations with a set of finite-difference equations, which is typically applied to structured grids for efficient implementation in CFD programs. 

(37) However, FDM is often limited to simple geometries such as rectangular or block-shaped geometries and struggles with curved boundaries. In contrast, FEM divides the fluid domain into small finite grids or elements, approximating PDEs through a local description of physics. 

(38) All elements contribute to a large, sparse matrix solver. However, FEM may not always provide accurate results for systems involving significant deformation and aggregation of particles like RBCs due to large distortion of grids. 

(39) FVM evaluates PDEs following the conservation laws and discretizes the selected flow domain into small but finite size control volumes, with each grid at the center of a finite volume. 

(40) The divergence theorem allows the conversion of volume integrals of PDEs with divergence terms into surface integrals of surface fluxes across cell boundaries. Due to its conservation property, FVM offers efficient outcomes when dealing with PDEs that embody mass, momentum, and energy conservation principles. Furthermore, widely accessible software packages like the OpenFOAM toolbox 

(41) include a viscoelastic solver, making it an attractive option for viscoelastic fluid flow modeling. 

(42)

2.2.3. Modeling Methods of Blood Flow Dynamics

The complexity in the blood flow simulation arises from deformability and aggregation that RBCs exhibit during their interaction with neighboring cells under different shear rate conditions induced by blood flow. Numerical models coupled with simulation programs have been applied as a groundbreaking method to predict such unique rheological behavior exhibited by RBCs and whole blood. The conventional approach of a single-phase flow simulation is often applied to blood flow simulations within large vessels possessing a moderate shear rate. However, such a method assumes the properties of plasma, RBCs and other cellular components to be evenly distributed as average density and viscosity in blood, resulting in the inability to simulate the mechanical dynamics, such as RBC aggregation under high-shear flow field, inherent in RBCs. To accurately describe the asymmetric distribution of RBC and blood flow, multiphase flow simulation, where numerical simulations of blood flows are often modeled as two immiscible phases, RBCs and blood plasma, is proposed. A common assumption is that RBCs exhibit non-Newtonian behavior while the plasma is treated as a continuous Newtonian phase.Numerous multiphase numerical models have been proposed to simulate the influence of RBCs on blood flow dynamics by different assumptions. In large-scale simulations (above the millimeter range), continuum-based methods are wildly used due to their lower computational demands. 

(43) Eulerian multiphase flow simulations offer the solution of a set of conservation equations for each separate phase and couple the phases through common pressure and interphase exchange coefficients. Xu et al. 

(44) utilized the combined finite-discrete element method (FDEM) to replicate the dynamic behavior and distortion of RBCs subjected to fluidic forces, utilizing the Johnson–Kendall–Roberts model 

(45) to define the adhesive forces of cell-to-cell interactions. The iterative direct-forcing immersed boundary method (IBM) is commonly employed in simulations of the fluid–cell interface of blood. This method effectively captures the intricacies of the thin and flexible RBC membranes within various external flow fields. 

(46) The study by Xu et al. 

(44) also adopts this approach to bridge the fluid dynamics and RBC deformation through IBM. Yoon and You utilized the Maxwell model to define the viscosity of the RBC membrane. 

(47) It was discovered that the Maxwell model could represent the stress relaxation and unloading processes of the cell. Furthermore, the reduced flexibility of an RBC under particular situations such as infection is specified, which was unattainable by the Kelvin–Voigt model 

(48) when compared to the Maxwell model in the literature. The Yeoh hyperplastic material model was also adapted to predict the nonlinear elasticity property of RBCs with FEM employed to discretize the RBC membrane using shell-type elements. Gracka et al. 

(49) developed a numerical CFD model with a finite-volume parallel solver for multiphase blood flow simulation, where an updated Maxwell viscoelasticity model and a Discrete Phase Model are adopted. In the study, the adapted IBM, based on unstructured grids, simulates the flow behavior and shape change of the RBCs through fluid-structure coupling. It was found that the hybrid Euler–Lagrange (E–L) approach 

(50) for the development of the multiphase model offered better results in the simulated CFL region in the microchannels.To study the dynamics of individual behaviors of RBCs and the consequent non-Newtonian blood flow, cell-shape-resolved computational models are often adapted. The use of the boundary integral method has become prevalent in minimizing computational expenses, particularly in the exclusive determination of fluid velocity on the surfaces of RBCs, incorporating the option of employing IBM or particle-based techniques. The cell-shaped-resolved method has enabled an examination of cell to cell interactions within complex ambient or pulsatile flow conditions 

(51) surrounding RBC membranes. Recently, Rydquist et al. 

(52) have looked to integrate statistical information from macroscale simulations to obtain a comprehensive overview of RBC behavior within the immediate proximity of the flow through introduction of respective models characterizing membrane shape definition, tension, bending stresses of RBC membranes.At a macroscopic scale, continuum models have conventionally been adapted for assessing blood flow dynamics through the application of elasticity theory and fluid dynamics. However, particle-based methods are known for their simplicity and adaptability in modeling complex multiscale fluid structures. Meshless methods, such as the boundary element method (BEM), smoothed particle hydrodynamics (SPH), and dissipative particle dynamics (DPD), are often used in particle-based characterization of RBCs and the surrounding fluid. By representing the fluid as discrete particles, meshless methods provide insights into the status and movement of the multiphase fluid. These methods allow for the investigation of cellular structures and microscopic interactions that affect blood rheology. Non-confronting mesh methods like IBM can also be used to couple a fluid solver such as FEM, FVM, or the Lattice Boltzmann Method (LBM) through membrane representation of RBCs. In comparison to conventional CFD methods, LBM has been viewed as a favorable numerical approach for solving the N–S equations and the simulation of multiphase flows. LBM exhibits the notable advantage of being amenable to high-performance parallel computing environments due to its inherently local dynamics. In contrast to DPD and SPH where RBC membranes are modeled as physically interconnected particles, LBM employs the IBM to account for the deformation dynamics of RBCs 

(53,54) under shear flows in complex channel geometries. 

(54,55) However, it is essential to acknowledge that the utilization of LBM in simulating RBC flows often entails a significant computational overhead, being a primary challenge in this context. Krüger et al. 

(56) proposed utilizing LBM as a fluid solver, IBM to couple the fluid and FEM to compute the response of membranes to deformation under immersed fluids. This approach decouples the fluid and membranes but necessitates significant computational effort due to the requirements of both meshes and particles.Despite the accuracy of current blood flow models, simulating complex conditions remains challenging because of the high computational load and cost. Balachandran Nair et al. 

(57) suggested a reduced order model of RBC under the framework of DEM, where the RBC is represented by overlapping constituent rigid spheres. The Morse potential force is adapted to account for the RBC aggregation exhibited by cell to cell interactions among RBCs at different distances. Based upon the IBM, the reduced-order RBC model is adapted to simulate blood flow transport for validation under both single and multiple RBCs with a resolved CFD-DEM solver. 

(58) In the resolved CFD-DEM model, particle sizes are larger than the grid size for a more accurate computation of the surrounding flow field. A continuous forcing approach is taken to describe the momentum source of the governing equation prior to discretization, which is different from a Direct Forcing Method (DFM). 

(59) As no body-conforming moving mesh is required, the continuous forcing approach offers lower complexity and reduced cost when compared to the DFM. Piquet et al. 

(60) highlighted the high complexity of the DFM due to its reliance on calculating an additional immersed boundary flux for the velocity field to ensure its divergence-free condition.The fluid–structure interaction (FSI) method has been advocated to connect the dynamic interplay of RBC membranes and fluid plasma within blood flow such as the coupling of continuum–particle interactions. However, such methodology is generally adapted for anatomical configurations such as arteries 

(61,62) and capillaries, 

(63) where both the structural components and the fluid domain undergo substantial deformation due to the moving boundaries. Due to the scope of the Review being blood flow simulation within microchannels of LOC devices without deformable boundaries, the Review of the FSI method will not be further carried out.In general, three numerical methods are broadly used: mesh-based, particle-based, and hybrid mesh–particle techniques, based on the spatial scale and the fundamental numerical approach, mesh-based methods tend to neglect the effects of individual particles, assuming a continuum and being efficient in terms of time and cost. However, the particle-based approach highlights more of the microscopic and mesoscopic level, where the influence of individual RBCs is considered. A review from Freund et al. 

(64) addressed the three numerical methodologies and their respective modeling approaches of RBC dynamics. Given the complex mechanics and the diverse levels of study concerning numerical simulations of blood and cellular flow, a broad spectrum of numerical methods for blood has been subjected to extensive review. 

(64−70) Ye at al. 

(65) offered an extensive review of the application of the DPD, SPH, and LBM for numerical simulations of RBC, while Rathnayaka et al. 

(67) conducted a review of the particle-based numerical modeling for liquid marbles through drawing parallels to the transport of RBCs in microchannels. A comparative analysis between conventional CFD methods and particle-based approaches for cellular and blood flow dynamic simulation can be found under the review by Arabghahestani et al. 

(66) Literature by Li et al. 

(68) and Beris et al. 

(69) offer an overview of both continuum-based models at micro/macroscales and multiscale particle-based models encompassing various length and temporal dimensions. Furthermore, these reviews deliberate upon the potential of coupling continuum-particle methods for blood plasma and RBC modeling. Arciero et al. 

(70) investigated various modeling approaches encompassing cellular interactions, such as cell to cell or plasma interactions and the individual cellular phases. A concise overview of the reviews is provided in Table 2 for reference.

Table 2. List of Reviews for Numerical Approaches Employed in Blood Flow Simulation

ReferenceNumerical methods
Li et al. (2013) (68)Continuum-based modeling (BIM), particle-based modeling (LBM, LB-FE, SPH, DPD)
Freund (2014) (64)RBC dynamic modeling (continuum-based modeling, complementary discrete microstructure modeling), blood flow dynamic modeling (FDM, IBM, LBM, particle-mesh methods, coupled boundary integral and mesh-based methods, DPD)
Ye et al. (2016) (65)DPD, SPH, LBM, coupled IBM-Smoothed DPD
Arciero et al. (2017) (70)LBM, IBM, DPD, conventional CFD Methods (FDM, FVM, FEM)
Arabghahestani et al. (2019) (66)Particle-based methods (LBM, DPD, direct simulation Monte Carlo, molecular dynamics), SPH, conventional CFD methods (FDM, FVM, FEM)
Beris et al. (2021) (69)DPD, smoothed DPD, IBM, LBM, BIM
Rathnayaka (2022) (67)SPH, CG, LBM

3. Capillary Driven Blood Flow in LOC Systems

ARTICLE SECTIONS

Jump To


3.1. Capillary Driven Flow Phenomena

Capillary driven (CD) flow is a pivotal mechanism in passive microfluidic flow systems 

(9) such as the blood circulation system and LOC systems. 

(71) CD flow is essentially the movement of a liquid to flow against drag forces, where the capillary effect exerts a force on the liquid at the borders, causing a liquid–air meniscus to flow despite gravity or other drag forces. A capillary pressure drops across the liquid–air interface with surface tension in the capillary radius and contact angle. The capillary effect depends heavily on the interaction between the different properties of surface materials. Different values of contact angles can be manipulated and obtained under varying levels of surface wettability treatments to manipulate the surface properties, resulting in different CD blood delivery rates for medical diagnostic device microchannels. CD flow techniques are appealing for many LOC devices, because they require no external energy. However, due to the passive property of liquid propulsion by capillary forces and the long-term instability of surface treatments on channel walls, the adaptability of CD flow in geometrically complex LOC devices may be limited.

3.2. Theoretical and Numerical Modeling of Capillary Driven Blood Flow

3.2.1. Theoretical Basis and Assumptions of Microfluidic Flow

The study of transport phenomena regarding either blood flow driven by capillary forces or externally applied forces under microfluid systems all demands a comprehensive recognition of the significant differences in flow dynamics between microscale and macroscale. The fundamental assumptions and principles behind fluid transport at the microscale are discussed in this section. Such a comprehension will lay the groundwork for the following analysis of the theoretical basis of capillary forces and their role in blood transport in LOC systems.

At the macroscale, fluid dynamics are often strongly influenced by gravity due to considerable fluid mass. However, the high surface to volume ratio at the microscale shifts the balance toward surface forces (e.g., surface tension and viscous forces), much larger than the inertial force. This difference gives rise to transport phenomena unique to microscale fluid transport, such as the prevalence of laminar flow due to a very low Reynolds number (generally lower than 1). Moreover, the fluid in a microfluidic system is often assumed to be incompressible due to the small flow velocity, indicating constant fluid density in both space and time.Microfluidic flow behaviors are governed by the fundamental principles of mass and momentum conservation, which are encapsulated in the continuity equation and the Navier–Stokes (N–S) equation. The continuity equation describes the conservation of mass, while the N–S equation captures the spatial and temporal variations in velocity, pressure, and other physical parameters. Under the assumption of the negligible influence of gravity in microfluidic systems, the continuity equation and the Eulerian representation of the incompressible N–S equation can be expressed as follows:

∇·𝐮⇀=0∇·�⇀=0

(7)

−∇𝑝+𝜇∇2𝐮⇀+∇·𝝉⇀−𝐅⇀=0−∇�+�∇2�⇀+∇·�⇀−�⇀=0

(8)Here, p is the pressure, u is the fluid viscosity, 

𝝉⇀�⇀ represents the stress tensor, and F is the body force exerted by external forces if present.

3.2.2. Theoretical Basis and Modeling of Capillary Force in LOC Systems

The capillary force is often the major driving force to manipulate and transport blood without an externally applied force in LOC systems. Forces induced by the capillary effect impact the free surface of fluids and are represented not directly in the Navier–Stokes equations but through the pressure boundary conditions of the pressure term p. For hydrophilic surfaces, the liquid generally induces a contact angle between 0° and 30°, encouraging the spread and attraction of fluid under a positive cos θ condition. For this condition, the pressure drop becomes positive and generates a spontaneous flow forward. A hydrophobic solid surface repels the fluid, inducing minimal contact. Generally, hydrophobic solids exhibit a contact angle larger than 90°, inducing a negative value of cos θ. Such a value will result in a negative pressure drop and a flow in the opposite direction. The induced contact angle is often utilized to measure the wall exposure of various surface treatments on channel walls where different wettability gradients and surface tension effects for CD flows are established. Contact angles between different interfaces are obtainable through standard values or experimental methods for reference. 

(72)For the characterization of the induced force by the capillary effect, the Young–Laplace (Y–L) equation 

(73) is widely employed. In the equation, the capillary is considered a pressure boundary condition between the two interphases. Through the Y–L equation, the capillary pressure force can be determined, and subsequently, the continuity and momentum balance equations can be solved to obtain the blood filling rate. Kim et al. 

(74) studied the effects of concentration and exposure time of a nonionic surfactant, Silwet L-77, on the performance of a polydimethylsiloxane (PDMS) microchannel in terms of plasma and blood self-separation. The study characterized the capillary pressure force by incorporating the Y–L equation and further evaluated the effects of the changing contact angle due to different levels of applied channel wall surface treatments. The expression of the Y–L equation utilized by Kim et al. 

(74) is as follows:

𝑃=−𝜎(cos𝜃b+cos𝜃tℎ+cos𝜃l+cos𝜃r𝑤)�=−�(cos⁡�b+cos⁡�tℎ+cos⁡�l+cos⁡�r�)

(9)where σ is the surface tension of the liquid and θ

bθ

tθ

l, and θ

r are the contact angle values between the liquid and the bottom, top, left, and right walls, respectively. A numerical simulation through Coventor software is performed to evaluate the dynamic changes in the filling rate within the microchannel. The simulation results for the blood filling rate in the microchannel are expressed at a specific time stamp, shown in Figure 2. The results portray an increasing instantaneous filling rate of blood in the microchannel following the decrease in contact angle induced by a higher concentration of the nonionic surfactant treated to the microchannel wall.

Figure 2. Numerical simulation of filling rate of capillary driven blood flow under various contact angle conditions at a specific timestamp. (74) Reproduced with permission from ref (74). Copyright 2010 Elsevier.

When in contact with hydrophilic or hydrophobic surfaces, blood forms a meniscus with a contact angle due to surface tension. The Lucas–Washburn (L–W) equation 

(75) is one of the pioneering theoretical definitions for the position of the meniscus over time. In addition, the L–W equation provides the possibility for research to obtain the velocity of the blood formed meniscus through the derivation of the meniscus position. The L–W equation 

(75) can be shown below:

𝐿(𝑡)=𝑅𝜎cos(𝜃)𝑡2𝜇⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯√�(�)=��⁡cos(�)�2�

(10)Here L(t) represents the distance of the liquid driven by the capillary forces. However, the generalized L–W equation solely assumes the constant physical properties from a Newtonian fluid rather than considering the non-Newtonian fluid behavior of blood. Cito et al. 

(76) constructed an enhanced version of the L–W equation incorporating the power law to consider the RBC aggregation and the FL effect. The non-Newtonian fluid apparent viscosity under the Power Law model is defined as

𝜇=𝑘·(𝛾˙)𝑛−1�=�·(�˙)�−1

(11)where γ̇ is the strain rate tensor defined as 

𝛾˙=12𝛾˙𝑖𝑗𝛾˙𝑗𝑖⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯√�˙=12�˙���˙��. The stress tensor term τ is computed as τ = μγ̇

ij. The updated L–W equation by Cito 

(76) is expressed as

𝐿(𝑡)=𝑅[(𝑛+13𝑛+1)(𝜎cos(𝜃)𝑅𝑘)1/𝑛𝑡]𝑛/𝑛+1�(�)=�[(�+13�+1)(�⁡cos(�)��)1/��]�/�+1

(12)where k is the flow consistency index and n is the power law index, respectively. The power law index, from the Power Law model, characterizes the extent of the non-Newtonian behavior of blood. Both the consistency and power law index rely on blood properties such as hematocrit, the appearance of the FL effect, the formation of RBC aggregates, etc. The updated L–W equation computes the location and velocity of blood flow caused by capillary forces at specified time points within the LOC devices, taking into account the effects of blood flow characteristics such as RBC aggregation and the FL effect on dynamic blood viscosity.Apart from the blood flow behaviors triggered by inherent blood properties, unique flow conditions driven by capillary forces that are portrayed under different microchannel geometries also hold crucial implications for CD blood delivery. Berthier et al. 

(77) studied the spontaneous Concus–Finn condition, the condition to initiate the spontaneous capillary flow within a V-groove microchannel, as shown in Figure 3(a) both experimentally and numerically. Through experimental studies, the spontaneous Concus–Finn filament development of capillary driven blood flow is observed, as shown in Figure 3(b), while the dynamic development of blood flow is numerically simulated through CFD simulation.

Figure 3. (a) Sketch of the cross-section of Berthier’s V-groove microchannel, (b) experimental view of blood in the V-groove microchannel, (78) (c) illustration of the dynamic change of the extension of filament from FLOW 3D under capillary flow at three increasing time intervals. (78) Reproduced with permission from ref (78). Copyright 2014 Elsevier.

Berthier et al. 

(77) characterized the contact angle needed for the initiation of the capillary driving force at a zero-inlet pressure, through the half-angle (α) of the V-groove geometry layout, and its relation to the Concus–Finn filament as shown below:

𝜃<𝜋2−𝛼sin𝛼1+2(ℎ2/𝑤)sin𝛼<cos𝜃{�<�2−�sin⁡�1+2(ℎ2/�)⁡sin⁡�<cos⁡�

(13)Three possible regimes were concluded based on the contact angle value for the initiation of flow and development of Concus–Finn filament:

𝜃>𝜃1𝜃1>𝜃>𝜃0𝜃0no SCFSCF without a Concus−Finn filamentSCF without a Concus−Finn filament{�>�1no SCF�1>�>�0SCF without a Concus−Finn filament�0SCF without a Concus−Finn filament

(14)Under Newton’s Law, the force balance with low Reynolds and Capillary numbers results in the neglect of inertial terms. The force balance between the capillary forces and the viscous force induced by the channel wall is proposed to derive the analytical fluid velocity. This relation between the two forces offers insights into the average flow velocity and the penetration distance function dependent on time. The apparent blood viscosity is defined by Berthier et al. 

(78) through Casson’s law, 

(23) given in eq 1. The research used the FLOW-3D program from Flow Science Inc. software, which solves transient, free-surface problems using the FDM in multiple dimensions. The Volume of Fluid (VOF) method 

(79) is utilized to locate and track the dynamic extension of filament throughout the advancing interface within the channel ahead of the main flow at three progressing time stamps, as depicted in Figure 3(c).

4. Electro-osmotic Flow (EOF) in LOC Systems

ARTICLE SECTIONS

Jump To


The utilization of external forces, such as electric fields, has significantly broadened the possibility of manipulating microfluidic flow in LOC systems. 

(80) Externally applied electric field forces induce a fluid flow from the movement of ions in fluid terms as the “electro-osmotic flow” (EOF).Unique transport phenomena, such as enhanced flow velocity and flow instability, induced by non-Newtonian fluids, particularly viscoelastic fluids, under EOF, have sparked considerable interest in microfluidic devices with simple or complicated geometries within channels. 

(81) However, compared to the study of Newtonian fluids and even other electro-osmotic viscoelastic fluid flows, the literature focusing on the theoretical and numerical modeling of electro-osmotic blood flow is limited due to the complexity of blood properties. Consequently, to obtain a more comprehensive understanding of the complex blood flow behavior under EOF, theoretical and numerical studies of the transport phenomena in the EOF section will be based on the studies of different viscoelastic fluids under EOF rather than that of blood specifically. Despite this limitation, we believe these studies offer valuable insights that can help understand the complex behavior of blood flow under EOF.

4.1. EOF Phenomena

Electro-osmotic flow occurs at the interface between the microchannel wall and bulk phase solution. When in contact with the bulk phase, solution ions are absorbed or dissociated at the solid–liquid interface, resulting in the formation of a charge layer, as shown in Figure 4. This charged channel surface wall interacts with both negative and positive ions in the bulk sample, causing repulsion and attraction forces to create a thin layer of immobilized counterions, known as the Stern layer. The induced electric potential from the wall gradually decreases with an increase in the distance from the wall. The Stern layer potential, commonly termed the zeta potential, controls the intensity of the electrostatic interactions between mobile counterions and, consequently, the drag force from the applied electric field. Next to the Stern layer is the diffuse mobile layer, mainly composed of a mobile counterion. These two layers constitute the “electrical double layer” (EDL), the thickness of which is directly proportional to the ionic strength (concentration) of the bulk fluid. The relationship between the two parameters is characterized by a Debye length (λ

D), expressed as

𝜆𝐷=𝜖𝑘B𝑇2(𝑍𝑒)2𝑐0⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯√��=��B�2(��)2�0

(15)where ϵ is the permittivity of the electrolyte solution, k

B is the Boltzmann constant, T is the electron temperature, Z is the integer valence number, e is the elementary charge, and c

0 is the ionic density.

Figure 4. Schematic diagram of an electro-osmotic flow in a microchannel with negative surface charge. (82) Reproduced with permission from ref (82). Copyright 2012 Woodhead Publishing.

When an electric field is applied perpendicular to the EDL, viscous drag is generated due to the movement of excess ions in the EDL. Electro-osmotic forces can be attributed to the externally applied electric potential (ϕ) and the zeta potential, the system wall induced potential by charged walls (ψ). As illustrated in Figure 4, the majority of ions in the bulk phase have a uniform velocity profile, except for a shear rate condition confined within an extremely thin Stern layer. Therefore, EOF displays a unique characteristic of a “near flat” or plug flow velocity profile, different from the parabolic flow typically induced by pressure-driven microfluidic flow (Hagen–Poiseuille flow). The plug-shaped velocity profile of the EOF possesses a high shear rate above the Stern layer.Overall, the EOF velocity magnitude is typically proportional to the Debye Length (λ

D), zeta potential, and magnitude of the externally applied electric field, while a more viscous liquid reduces the EOF velocity.

4.2. Modeling on Electro-osmotic Viscoelastic Fluid Flow

4.2.1. Theoretical Basis of EOF Mechanisms

The EOF of an incompressible viscoelastic fluid is commonly governed by the continuity and incompressible N–S equations, as shown in eqs 7 and 8, where the stress tensor and the electrostatic force term are coupled. The electro-osmotic body force term F, representing the body force exerted by the externally applied electric force, is defined as 

𝐹⇀=𝑝𝐸𝐸⇀�⇀=���⇀, where ρ

E and 

𝐸⇀�⇀ are the net electric charge density and the applied external electric field, respectively.Numerous models are established to theoretically study the externally applied electric potential and the system wall induced potential by charged walls. The following Laplace equation, expressed as eq 16, is generally adapted and solved to calculate the externally applied potential (ϕ).

∇2𝜙=0∇2�=0

(16)Ion diffusion under applied electric fields, together with mass transport resulting from convection and diffusion, transports ionic solutions in bulk flow under electrokinetic processes. The Nernst–Planck equation can describe these transport methods, including convection, diffusion, and electro-diffusion. Therefore, the Nernst–Planck equation is used to determine the distribution of the ions within the electrolyte. The electric potential induced by the charged channel walls follows the Poisson–Nernst–Plank (PNP) equation, which can be written as eq 17.

∇·[𝐷𝑖∇𝑛𝑖−𝑢⇀𝑛𝑖+𝑛𝑖𝐷𝑖𝑧𝑖𝑒𝑘𝑏𝑇∇(𝜙+𝜓)]=0∇·[��∇��−�⇀��+����������∇(�+�)]=0

(17)where D

in

i, and z

i are the diffusion coefficient, ionic concentration, and ionic valence of the ionic species I, respectively. However, due to the high nonlinearity and numerical stiffness introduced by different lengths and time scales from the PNP equations, the Poisson–Boltzmann (PB) model is often considered the major simplified method of the PNP equation to characterize the potential distribution of the EDL region in microchannels. In the PB model, it is assumed that the ionic species in the fluid follow the Boltzmann distribution. This model is typically valid for steady-state problems where charge transport can be considered negligible, the EDLs do not overlap with each other, and the intrinsic potentials are low. It provides a simplified representation of the potential distribution in the EDL region. The PB equation governing the EDL electric potential distribution is described as

∇2𝜓=(2𝑒𝑧𝑛0𝜀𝜀0)sinh(𝑧𝑒𝜓𝑘b𝑇)∇2�=(2���0��0)⁡sinh(����b�)

(18)where n

0 is the ion bulk concentration, z is the ionic valence, and ε

0 is the electric permittivity in the vacuum. Under low electric potential conditions, an even further simplified model to illustrate the EOF phenomena is the Debye–Hückel (DH) model. The DH model is derived by obtaining a charge density term by expanding the exponential term of the Boltzmann equation in a Taylor series.

4.2.2. EOF Modeling for Viscoelastic Fluids

Many studies through numerical modeling were performed to obtain a deeper understanding of the effect exhibited by externally applied electric fields on viscoelastic flow in microchannels under various geometrical designs. Bello et al. 

(83) found that methylcellulose solution, a non-Newtonian polymer solution, resulted in stronger electro-osmotic mobility in experiments when compared to the predictions by the Helmholtz–Smoluchowski equation, which is commonly used to define the velocity of EOF of a Newtonian fluid. Being one of the pioneers to identify the discrepancies between the EOF of Newtonian and non-Newtonian fluids, Bello et al. attributed such discrepancies to the presence of a very high shear rate in the EDL, resulting in a change in the orientation of the polymer molecules. Park and Lee 

(84) utilized the FVM to solve the PB equation for the characterization of the electric field induced force. In the study, the concept of fractional calculus for the Oldroyd-B model was adapted to illustrate the elastic and memory effects of viscoelastic fluids in a straight microchannel They observed that fluid elasticity and increased ratio of viscoelastic fluid contribution to overall fluid viscosity had a significant impact on the volumetric flow rate and sensitivity of velocity to electric field strength compared to Newtonian fluids. Afonso et al. 

(85) derived an analytical expression for EOF of viscoelastic fluid between parallel plates using the DH model to account for a zeta potential condition below 25 mV. The study established the understanding of the electro-osmotic viscoelastic fluid flow under low zeta potential conditions. Apart from the electrokinetic forces, pressure forces can also be coupled with EOF to generate a unique fluid flow behavior within the microchannel. Sousa et al. 

(86) analytically studied the flow of a standard viscoelastic solution by combining the pressure gradient force with an externally applied electric force. It was found that, at a near wall skimming layer and the outer layer away from the wall, macromolecules migrating away from surface walls in viscoelastic fluids are observed. In the study, the Phan-Thien Tanner (PTT) constitutive model is utilized to characterize the viscoelastic properties of the solution. The approach is found to be valid when the EDL is much thinner than the skimming layer under an enhanced flow rate. Zhao and Yang 

(87) solved the PB equation and Carreau model for the characterization of the EOF mechanism and non-Newtonian fluid respectively through the FEM. The numerical results depict that, different from the EOF of Newtonian fluids, non-Newtonian fluids led to an increase of electro-osmotic mobility for shear thinning fluids but the opposite for shear thickening fluids.Like other fluid transport driving forces, EOF within unique geometrical layouts also portrays unique transport phenomena. Pimenta and Alves 

(88) utilized the FVM to perform numerical simulations of the EOF of viscoelastic fluids considering the PB equation and the Oldroyd-B model, in a cross-slot and flow-focusing microdevices. It was found that electroelastic instabilities are formed due to the development of large stresses inside the EDL with streamlined curvature at geometry corners. Bezerra et al. 

(89) used the FDM to numerically analyze the vortex formation and flow instability from an electro-osmotic non-Newtonian fluid flow in a microchannel with a nozzle geometry and parallel wall geometry setting. The PNP equation is utilized to characterize the charge motion in the EOF and the PTT model for non-Newtonian flow characterization. A constriction geometry is commonly utilized in blood flow adapted in LOC systems due to the change in blood flow behavior under narrow dimensions in a microchannel. Ji et al. 

(90) recently studied the EOF of viscoelastic fluid in a constriction microchannel connected by two relatively big reservoirs on both ends (as seen in Figure 5) filled with the polyacrylamide polymer solution, a viscoelastic fluid, and an incompressible monovalent binary electrolyte solution KCl.

Figure 5. Schematic diagram of a negatively charged constriction microchannel connected to two reservoirs at both ends. An electro-osmotic flow is induced in the system by the induced potential difference between the anode and cathode. (90) Reproduced with permission from ref (90). Copyright 2021 The Authors, under the terms of the Creative Commons (CC BY 4.0) License https://creativecommons.org/licenses/by/4.0/.

In studying the EOF of viscoelastic fluids, the Oldroyd-B model is often utilized to characterize the polymeric stress tensor and the deformation rate of the fluid. The Oldroyd-B model is expressed as follows:

𝜏=𝜂p𝜆(𝐜−𝐈)�=�p�(�−�)

(19)where η

p, λ, c, and I represent the polymer dynamic viscosity, polymer relaxation time, symmetric conformation tensor of the polymer molecules, and the identity matrix, respectively.A log-conformation tensor approach is taken to prevent convergence difficulty induced by the viscoelastic properties. The conformation tensor (c) in the polymeric stress tensor term is redefined by a new tensor (Θ) based on the natural logarithm of the c. The new tensor is defined as

Θ=ln(𝐜)=𝐑ln(𝚲)𝐑Θ=ln(�)=�⁡ln(�)�

(20)in which Λ is the diagonal matrix and R is the orthogonal matrix.Under the new conformation tensor, the induced EOF of a viscoelastic fluid is governed by the continuity and N–S equations adapting the Oldroyd-B model, which is expressed as

∂𝚯∂𝑡+𝐮·∇𝚯=𝛀Θ−ΘΩ+2𝐁+1𝜆(eΘ−𝐈)∂�∂�+�·∇�=�Θ−ΘΩ+2�+1�(eΘ−�)

(21)where Ω and B represent the anti-symmetric matrix and the symmetric traceless matrix of the decomposition of the velocity gradient tensor ∇u, respectively. The conformation tensor can be recovered by c = exp(Θ). The PB model and Laplace equation are utilized to characterize the charged channel wall induced potential and the externally applied potential.The governing equations are numerically solved through the FVM by RheoTool, 

(42) an open-source viscoelastic EOF solver on the OpenFOAM platform. A SIMPLEC (Semi-Implicit Method for Pressure Linked Equations-Consistent) algorithm was applied to solve the velocity-pressure coupling. The pressure field and velocity field were computed by the PCG (Preconditioned Conjugate Gradient) solver and the PBiCG (Preconditioned Biconjugate Gradient) solver, respectively.Ranging magnitudes of an applied electric field or fluid concentration induce both different streamlines and velocity magnitudes at various locations and times of the microchannel. In the study performed by Ji et al., 

(90) notable fluctuation of streamlines and vortex formation is formed at the upper stream entrance of the constriction as shown in Figure 6(a) and (b), respectively, due to the increase of electrokinetic effect, which is seen as a result of the increase in polymeric stress (τ

xx). 

(90) The contraction geometry enhances the EOF velocity within the constriction channel under high E

app condition (600 V/cm). Such phenomena can be attributed to the dependence of electro-osmotic viscoelastic fluid flow on the system wall surface and bulk fluid properties. 

(91)

Figure 6. Schematic diagram of vortex formation and streamlines of EOF depicting flow instability at (a) 1.71 s and (b) 1.75 s. Spatial distribution of the elastic normal stress at (c) high Eapp condition. Streamline of an electro-osmotic flow under Eapp of 600 V/cm (90) for (d) non-Newtonian and (e) Newtonian fluid through a constriction geometry. Reproduced with permission from ref (90). Copyright 2021 The Authors, under the terms of the Creative Commons (CC BY 4.0) License https://creativecommons.org/licenses/by/4.0/.

As elastic normal stress exceeds the local shear stress, flow instability and vortex formation occur. The induced elastic stress under EOF not only enhances the instability of the flow but often generates an irregular secondary flow leading to strong disturbance. 

(92) It is also vital to consider the effect of the constriction layout of microchannels on the alteration of the field strength within the system. The contraction geometry enhances a larger electric field strength compared with other locations of the channel outside the constriction region, resulting in a higher velocity gradient and stronger extension on the polymer within the viscoelastic solution. Following the high shear flow condition, a higher magnitude of stretch for polymer molecules in viscoelastic fluids exhibits larger elastic stresses and enhancement of vortex formation at the region. 

(93)As shown in Figure 6(c), significant elastic normal stress occurs at the inlet of the constriction microchannel. Such occurrence of a polymeric flow can be attributed to the dominating elongational flow, giving rise to high deformation of the polymers within the viscoelastic fluid flow, resulting in higher elastic stress from the polymers. Such phenomena at the entrance result in the difference in velocity streamline as circled in Figure 6(d) compared to that of the Newtonian fluid at the constriction entrance in Figure 6(e). 

(90) The difference between the Newtonian and polymer solution at the exit, as circled in Figure 6(d) and (e), can be attributed to the extrudate swell effect of polymers 

(94) within the viscoelastic fluid flow. The extrudate swell effect illustrates that, as polymers emerge from the constriction exit, they tend to contract in the flow direction and grow in the normal direction, resulting in an extrudate diameter greater than the channel size. The deformation of polymers within the polymeric flow at both the entrance and exit of the contraction channel facilitates the change in shear stress conditions of the flow, leading to the alteration in streamlines of flows for each region.

4.3. EOF Applications in LOC Systems

4.3.1. Mixing in LOC Systems

Rather than relying on the micromixing controlled by molecular diffusion under low Reynolds number conditions, active mixers actively leverage convective instability and vortex formation induced by electro-osmotic flows from alternating current (AC) or direct current (DC) electric fields. Such adaptation is recognized as significant breakthroughs for promotion of fluid mixing in chemical and biological applications such as drug delivery, medical diagnostics, chemical synthesis, and so on. 

(95)Many researchers proposed novel designs of electro-osmosis micromixers coupled with numerical simulations in conjunction with experimental findings to increase their understanding of the role of flow instability and vortex formation in the mixing process under electrokinetic phenomena. Matsubara and Narumi 

(96) numerically modeled the mixing process in a microchannel with four electrodes on each side of the microchannel wall, which generated a disruption through unstable electro-osmotic vortices. It was found that particle mixing was sensitive to both the convection effect induced by the main and secondary vortex within the micromixer and the change in oscillation frequency caused by the supplied AC voltage when the Reynolds number was varied. Qaderi et al. 

(97) adapted the PNP equation to numerically study the effect of the geometry and zeta potential configuration of the microchannel on the mixing process with a combined electro-osmotic pressure driven flow. It was reported that the application of heterogeneous zeta potential configuration enhances the mixing efficiency by around 23% while the height of the hurdles increases the mixing efficiency at most 48.1%. Cho et al. 

(98) utilized the PB model and Laplace equation to numerically simulate the electro-osmotic non-Newtonian fluid mixing process within a wavy and block layout of microchannel walls. The Power Law model is adapted to describe the fluid rheological characteristic. It was found that shear-thinning fluids possess a higher volumetric flow rate, which could result in poorer mixing efficiency compared to that of Newtonian fluids. Numerous studies have revealed that flow instability and vortex generation, in particular secondary vortices produced by barriers or greater magnitudes of heterogeneous zeta potential distribution, enhance mixing by increasing bulk flow velocity and reducing flow distance.To better understand the mechanism of disturbance formed in the system due to externally applied forces, known as electrokinetic instability, literature often utilize the Rayleigh (Ra) number, 

(1) as described below:

𝑅𝑎𝑣=𝑢ev𝑢eo=(𝛾−1𝛾+1)2𝑊𝛿2𝐸el2𝐻2𝜁𝛿Ra�=�ev�eo=(�−1�+1)2��2�el2�2��

(22)where γ is the conductivity ratio of the two streams and can be written as 

𝛾=𝜎el,H𝜎el,L�=�el,H�el,L. The Ra number characterizes the ratio between electroviscous and electro-osmotic flow. A high Ra

v value often results in good mixing. It is evident that fluid properties such as the conductivity (σ) of the two streams play a key role in the formation of disturbances to enhance mixing in microsystems. At the same time, electrokinetic parameters like the zeta potential (ζ) in the Ra number is critical in the characterization of electro-osmotic velocity and a slip boundary condition at the microchannel wall.To understand the mixing result along the channel, the concentration field can be defined and simulated under the assumption of steady state conditions and constant diffusion coefficient for each of the working fluid within the system through the convection–diffusion equation as below:

∂𝑐𝒊∂𝑡+∇⇀(𝑐𝑖𝑢⇀−𝐷𝑖∇⇀𝑐𝒊)=0∂��∂�+∇⇀(���⇀−��∇⇀��)=0

(23)where c

i is the species concentration of species i and D

i is the diffusion coefficient of the corresponding species.The standard deviation of concentration (σ

sd) can be adapted to evaluate the mixing quality of the system. 

(97) The standard deviation for concentration at a specific portion of the channel may be calculated using the equation below:

𝜎sd=∫10(𝐶∗(𝑦∗)−𝐶m)2d𝑦∗∫10d𝑦∗⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯�sd=∫01(�*(�*)−�m)2d�*∫01d�*

(24)where C*(y*) and C

m are the non-dimensional concentration profile and the mean concentration at the portion, respectively. C* is the non-dimensional concentration and can be calculated as 

𝐶∗=𝐶𝐶ref�*=��ref, where C

ref is the reference concentration defined as the bulk solution concentration. The mean concentration profile can be calculated as 

𝐶m=∫10(𝐶∗(𝑦∗)d𝑦∗∫10d𝑦∗�m=∫01(�*(�*)d�*∫01d�*. With the standard deviation of concentration, the mixing efficiency 

(97) can then be calculated as below:

𝜀𝑥=1−𝜎sd𝜎sd,0��=1−�sd�sd,0

(25)where σ

sd,0 is the standard derivation of the case of no mixing. The value of the mixing efficiency is typically utilized in conjunction with the simulated flow field and concentration field to explore the effect of geometrical and electrokinetic parameters on the optimization of the mixing results.

5. Summary

ARTICLE SECTIONS

Jump To


5.1. Conclusion

Viscoelastic fluids such as blood flow in LOC systems are an essential topic to proceed with diagnostic analysis and research through microdevices in the biomedical and pharmaceutical industries. The complex blood flow behavior is tightly controlled by the viscoelastic characteristics of blood such as the dynamic viscosity and the elastic property of RBCs under various shear rate conditions. Furthermore, the flow behaviors under varied driving forces promote an array of microfluidic transport phenomena that are critical to the management of blood flow and other adapted viscoelastic fluids in LOC systems. This review addressed the blood flow phenomena, the complicated interplay between shear rate and blood flow behaviors, and their numerical modeling under LOC systems through the lens of the viscoelasticity characteristic. Furthermore, a theoretical understanding of capillary forces and externally applied electric forces leads to an in-depth investigation of the relationship between blood flow patterns and the key parameters of the two driving forces, the latter of which is introduced through the lens of viscoelastic fluids, coupling numerical modeling to improve the knowledge of blood flow manipulation in LOC systems. The flow disturbances triggered by the EOF of viscoelastic fluids and their impact on blood flow patterns have been deeply investigated due to their important role and applications in LOC devices. Continuous advancements of various numerical modeling methods with experimental findings through more efficient and less computationally heavy methods have served as an encouraging sign of establishing more accurate illustrations of the mechanisms for multiphase blood and other viscoelastic fluid flow transport phenomena driven by various forces. Such progress is fundamental for the manipulation of unique transport phenomena, such as the generated disturbances, to optimize functionalities offered by microdevices in LOC systems.

The following section will provide further insights into the employment of studied blood transport phenomena to improve the functionality of micro devices adapting LOC technology. A discussion of the novel roles that external driving forces play in microfluidic flow behaviors is also provided. Limitations in the computational modeling of blood flow and electrokinetic phenomena in LOC systems will also be emphasized, which may provide valuable insights for future research endeavors. These discussions aim to provide guidance and opportunities for new paths in the ongoing development of LOC devices that adapt blood flow.

5.2. Future Directions

5.2.1. Electro-osmosis Mixing in LOC Systems

Despite substantial research, mixing results through flow instability and vortex formation phenomena induced by electro-osmotic mixing still deviate from the effective mixing results offered by chaotic mixing results such as those seen in turbulent flows. However, recent discoveries of a mixing phenomenon that is generally observed under turbulent flows are found within electro-osmosis micromixers under low Reynolds number conditions. Zhao 

(99) experimentally discovered a rapid mixing process in an AC applied micromixer, where the power spectrum of concentration under an applied voltage of 20 V

p-p induces a −5/3 slope within a frequency range. This value of the slope is considered as the O–C spectrum in macroflows, which is often visible under relatively high Re conditions, such as the Taylor microscale Reynolds number Re > 500 in turbulent flows. 

(100) However, the Re value in the studied system is less than 1 at the specific location and applied voltage. A secondary flow is also suggested to occur close to microchannel walls, being attributed to the increase of convective instability within the system.Despite the experimental phenomenon proposed by Zhao et al., 

(99) the range of effects induced by vital parameters of an EOF mixing system on the enhanced mixing results and mechanisms of disturbance generated by the turbulent-like flow instability is not further characterized. Such a gap in knowledge may hinder the adaptability and commercialization of the discovery of micromixers. One of the parameters for further evaluation is the conductivity gradient of the fluid flow. A relatively strong conductivity gradient (5000:1) was adopted in the system due to the conductive properties of the two fluids. The high conductivity gradients may contribute to the relatively large Rayleigh number and differences in EDL layer thickness, resulting in an unusual disturbance in laminar flow conditions and enhanced mixing results. However, high conductivity gradients are not always achievable by the working fluids due to diverse fluid properties. The reliance on turbulent-like phenomena and rapid mixing results in a large conductivity gradient should be established to prevent the limited application of fluids for the mixing system. In addition, the proposed system utilizes distinct zeta potential distributions at the top and bottom walls due to their difference in material choices, which may be attributed to the flow instability phenomena. Further studies should be made on varying zeta potential magnitude and distribution to evaluate their effect on the slip boundary conditions of the flow and the large shear rate condition close to the channel wall of EOF. Such a study can potentially offer an optimized condition in zeta potential magnitude through material choices and geometrical layout of the zeta potential for better mixing results and manipulation of mixing fluid dynamics. The two vital parameters mentioned above can be varied with the aid of numerical simulation to understand the effect of parameters on the interaction between electro-osmotic forces and electroviscous forces. At the same time, the relationship of developed streamlines of the simulated velocity and concentration field, following their relationship with the mixing results, under the impact of these key parameters can foster more insight into the range of impact that the two parameters have on the proposed phenomena and the microfluidic dynamic principles of disturbances.

In addition, many of the current investigations of electrokinetic mixers commonly emphasize the fluid dynamics of mixing for Newtonian fluids, while the utilization of biofluids, primarily viscoelastic fluids such as blood, and their distinctive response under shear forces in these novel mixing processes of LOC systems are significantly less studied. To develop more compatible microdevice designs and efficient mixing outcomes for the biomedical industry, it is necessary to fill the knowledge gaps in the literature on electro-osmotic mixing for biofluids, where properties of elasticity, dynamic viscosity, and intricate relationship with shear flow from the fluid are further considered.

5.2.2. Electro-osmosis Separation in LOC Systems

Particle separation in LOC devices, particularly in biological research and diagnostics, is another area where disturbances may play a significant role in optimization. 

(101) Plasma analysis in LOC systems under precise control of blood flow phenomena and blood/plasma separation procedures can detect vital information about infectious diseases from particular antibodies and foreign nucleic acids for medical treatments, diagnostics, and research, 

(102) offering more efficient results and simple operating procedures compared to that of the traditional centrifugation method for blood and plasma separation. However, the adaptability of LOC devices for blood and plasma separation is often hindered by microchannel clogging, where flow velocity and plasma yield from LOC devices is reduced due to occasional RBC migration and aggregation at the filtration entrance of microdevices. 

(103)It is important to note that the EOF induces flow instability close to microchannel walls, which may provide further solutions to clogging for the separation process of the LOC systems. Mohammadi et al. 

(104) offered an anti-clogging effect of RBCs at the blood and plasma separating device filtration entry, adjacent to the surface wall, through RBC disaggregation under high shear rate conditions generated by a forward and reverse EOF direction.

Further theoretical and numerical research can be conducted to characterize the effect of high shear rate conditions near microchannel walls toward the detachment of binding blood cells on surfaces and the reversibility of aggregation. Through numerical modeling with varying electrokinetic parameters to induce different degrees of disturbances or shear conditions at channel walls, it may be possible to optimize and better understand the process of disrupting the forces that bind cells to surface walls and aggregated cells at filtration pores. RBCs that migrate close to microchannel walls are often attracted by the adhesion force between the RBC and the solid surface originating from the van der Waals forces. Following RBC migration and attachment by adhesive forces adjacent to the microchannel walls as shown in Figure 7, the increase in viscosity at the region causes a lower shear condition and encourages RBC aggregation (cell–cell interaction), which clogs filtering pores or microchannels and reduces flow velocity at filtration region. Both the impact that shear forces and disturbances may induce on cell binding forces with surface walls and other cells leading to aggregation may suggest further characterization. Kinetic parameters such as activation energy and the rate-determining step for cell binding composition attachment and detachment should be considered for modeling the dynamics of RBCs and blood flows under external forces in LOC separation devices.

Figure 7. Schematic representations of clogging at a microchannel pore following the sequence of RBC migration, cell attachment to channel walls, and aggregation. (105) Reproduced with permission from ref (105). Copyright 2018 The Authors under the terms of the Creative Commons (CC BY 4.0) License https://creativecommons.org/licenses/by/4.0/.

5.2.3. Relationship between External Forces and Microfluidic Systems

In blood flow, a thicker CFL suggests a lower blood viscosity, suggesting a complex relationship between shear stress and shear rate, affecting the blood viscosity and blood flow. Despite some experimental and numerical studies on electro-osmotic non-Newtonian fluid flow, limited literature has performed an in-depth investigation of the role that applied electric forces and other external forces could play in the process of CFL formation. Additional studies on how shear rates from external forces affect CFL formation and microfluidic flow dynamics can shed light on the mechanism of the contribution induced by external driving forces to the development of a separate phase of layer, similar to CFL, close to the microchannel walls and distinct from the surrounding fluid within the system, then influencing microfluidic flow dynamics.One of the mechanisms of phenomena to be explored is the formation of the Exclusion Zone (EZ) region following a “Self-Induced Flow” (SIF) phenomenon discovered by Li and Pollack, 

(106) as shown in Figure 8(a) and (b), respectively. A spontaneous sustained axial flow is observed when hydrophilic materials are immersed in water, resulting in the buildup of a negative layer of charges, defined as the EZ, after water molecules absorb infrared radiation (IR) energy and break down into H and OH

+.

Figure 8. Schematic representations of (a) the Exclusion Zone region and (b) the Self Induced Flow through visualization of microsphere movement within a microchannel. (106) Reproduced with permission from ref (106). Copyright 2020 The Authors under the terms of the Creative Commons (CC BY 4.0) License https://creativecommons.org/licenses/by/4.0/.

Despite the finding of such a phenomenon, the specific mechanism and role of IR energy have yet to be defined for the process of EZ development. To further develop an understanding of the role of IR energy in such phenomena, a feasible study may be seen through the lens of the relationships between external forces and microfluidic flow. In the phenomena, the increase of SIF velocity under a rise of IR radiation resonant characteristics is shown in the participation of the external electric field near the microchannel walls under electro-osmotic viscoelastic fluid flow systems. The buildup of negative charges at the hydrophilic surfaces in EZ is analogous to the mechanism of electrical double layer formation. Indeed, research has initiated the exploration of the core mechanisms for EZ formation through the lens of the electrokinetic phenomena. 

(107) Such a similarity of the role of IR energy and the transport phenomena of SIF with electrokinetic phenomena paves the way for the definition of the unknown SIF phenomena and EZ formation. Furthermore, Li and Pollack 

(106) suggest whether CFL formation might contribute to a SIF of blood using solely IR radiation, a commonly available source of energy in nature, as an external driving force. The proposition may be proven feasible with the presence of the CFL region next to the negatively charged hydrophilic endothelial glycocalyx layer, coating the luminal side of blood vessels. 

(108) Further research can dive into the resonating characteristics between the formation of the CFL region next to the hydrophilic endothelial glycocalyx layer and that of the EZ formation close to hydrophilic microchannel walls. Indeed, an increase in IR energy is known to rapidly accelerate EZ formation and SIF velocity, depicting similarity to the increase in the magnitude of electric field forces and greater shear rates at microchannel walls affecting CFL formation and EOF velocity. Such correlation depicts a future direction in whether SIF blood flow can be observed and characterized theoretically further through the lens of the relationship between blood flow and shear forces exhibited by external energy.

The intricate link between the CFL and external forces, more specifically the externally applied electric field, can receive further attention to provide a more complete framework for the mechanisms between IR radiation and EZ formation. Such characterization may also contribute to a greater comprehension of the role IR can play in CFL formation next to the endothelial glycocalyx layer as well as its role as a driving force to propel blood flow, similar to the SIF, but without the commonly assumed pressure force from heart contraction as a source of driving force.

5.3. Challenges

Although there have been significant improvements in blood flow modeling under LOC systems over the past decade, there are still notable constraints that may require special attention for numerical simulation applications to benefit the adaptability of the designs and functionalities of LOC devices. Several points that require special attention are mentioned below:

1.The majority of CFD models operate under the relationship between the viscoelasticity of blood and the shear rate conditions of flow. The relative effect exhibited by the presence of highly populated RBCs in whole blood and their forces amongst the cells themselves under complex flows often remains unclearly defined. Furthermore, the full range of cell populations in whole blood requires a much more computational load for numerical modeling. Therefore, a vital goal for future research is to evaluate a reduced modeling method where the impact of cell–cell interaction on the viscoelastic property of blood is considered.
2.Current computational methods on hemodynamics rely on continuum models based upon non-Newtonian rheology at the macroscale rather than at molecular and cellular levels. Careful considerations should be made for the development of a constructive framework for the physical and temporal scales of micro/nanoscale systems to evaluate the intricate relationship between fluid driving forces, dynamic viscosity, and elasticity.
3.Viscoelastic fluids under the impact of externally applied electric forces often deviate from the assumptions of no-slip boundary conditions due to the unique flow conditions induced by externally applied forces. Furthermore, the mechanism of vortex formation and viscoelastic flow instability at laminar flow conditions should be better defined through the lens of the microfluidic flow phenomenon to optimize the prediction of viscoelastic flow across different geometrical layouts. Mathematical models and numerical methods are needed to better predict such disturbance caused by external forces and the viscoelasticity of fluids at such a small scale.
4.Under practical situations, zeta potential distribution at channel walls frequently deviates from the common assumption of a constant distribution because of manufacturing faults or inherent surface charges prior to the introduction of electrokinetic influence. These discrepancies frequently lead to inconsistent surface potential distribution, such as excess positive ions at relatively more negatively charged walls. Accordingly, unpredicted vortex formation and flow instability may occur. Therefore, careful consideration should be given to these discrepancies and how they could trigger the transport process and unexpected results of a microdevice.

Author Information

ARTICLE SECTIONS

Jump To


  • Corresponding Authors
    • Zhe Chen – Department of Chemical Engineering, School of Chemistry and Chemical Engineering, State Key Laboratory of Metal Matrix Composites, Shanghai Jiao Tong University, Shanghai 200240, P. R. China;  Email: zaccooky@sjtu.edu.cn
    • Bo Ouyang – Department of Chemical Engineering, School of Chemistry and Chemical Engineering, State Key Laboratory of Metal Matrix Composites, Shanghai Jiao Tong University, Shanghai 200240, P. R. China;  Email: bouy93@sjtu.edu.cn
    • Zheng-Hong Luo – Department of Chemical Engineering, School of Chemistry and Chemical Engineering, State Key Laboratory of Metal Matrix Composites, Shanghai Jiao Tong University, Shanghai 200240, P. R. China;  Orcidhttps://orcid.org/0000-0001-9011-6020; Email: luozh@sjtu.edu.cn
  • Authors
    • Bin-Jie Lai – Department of Chemical Engineering, School of Chemistry and Chemical Engineering, State Key Laboratory of Metal Matrix Composites, Shanghai Jiao Tong University, Shanghai 200240, P. R. China;  Orcidhttps://orcid.org/0009-0002-8133-5381
    • Li-Tao Zhu – Department of Chemical Engineering, School of Chemistry and Chemical Engineering, State Key Laboratory of Metal Matrix Composites, Shanghai Jiao Tong University, Shanghai 200240, P. R. China;  Orcidhttps://orcid.org/0000-0001-6514-8864
  • NotesThe authors declare no competing financial interest.

Acknowledgments

ARTICLE SECTIONS

Jump To


This work was supported by the National Natural Science Foundation of China (No. 22238005) and the Postdoctoral Research Foundation of China (No. GZC20231576).

Vocabulary

ARTICLE SECTIONS

Jump To


Microfluidicsthe field of technological and scientific study that investigates fluid flow in channels with dimensions between 1 and 1000 μm
Lab-on-a-Chip Technologythe field of research and technological development aimed at integrating the micro/nanofluidic characteristics to conduct laboratory processes on handheld devices
Computational Fluid Dynamics (CFD)the method utilizing computational abilities to predict physical fluid flow behaviors mathematically through solving the governing equations of corresponding fluid flows
Shear Ratethe rate of change in velocity where one layer of fluid moves past the adjacent layer
Viscoelasticitythe property holding both elasticity and viscosity characteristics relying on the magnitude of applied shear stress and time-dependent strain
Electro-osmosisthe flow of fluid under an applied electric field when charged solid surface is in contact with the bulk fluid
Vortexthe rotating motion of a fluid revolving an axis line

References

ARTICLE SECTIONS

Jump To


This article references 108 other publications.

  1. 1Neethirajan, S.; Kobayashi, I.; Nakajima, M.; Wu, D.; Nandagopal, S.; Lin, F. Microfluidics for food, agriculture and biosystems industries. Lab Chip 201111 (9), 1574– 1586,  DOI: 10.1039/c0lc00230eViewGoogle Scholar
  2. 2Whitesides, G. M. The origins and the future of microfluidics. Nature 2006442 (7101), 368– 373,  DOI: 10.1038/nature05058ViewGoogle Scholar
  3. 3Burklund, A.; Tadimety, A.; Nie, Y.; Hao, N.; Zhang, J. X. J. Chapter One – Advances in diagnostic microfluidics; Elsevier, 2020; DOI:  DOI: 10.1016/bs.acc.2019.08.001 .ViewGoogle Scholar
  4. 4Abdulbari, H. A. Chapter 12 – Lab-on-a-chip for analysis of blood. In Nanotechnology for Hematology, Blood Transfusion, and Artificial Blood; Denizli, A., Nguyen, T. A., Rajan, M., Alam, M. F., Rahman, K., Eds.; Elsevier, 2022; pp 265– 283.ViewGoogle Scholar
  5. 5Vladisavljević, G. T.; Khalid, N.; Neves, M. A.; Kuroiwa, T.; Nakajima, M.; Uemura, K.; Ichikawa, S.; Kobayashi, I. Industrial lab-on-a-chip: Design, applications and scale-up for drug discovery and delivery. Advanced Drug Delivery Reviews 201365 (11), 1626– 1663,  DOI: 10.1016/j.addr.2013.07.017ViewGoogle Scholar
  6. 6Kersaudy-Kerhoas, M.; Dhariwal, R.; Desmulliez, M. P. Y.; Jouvet, L. Hydrodynamic blood plasma separation in microfluidic channels. Microfluid. Nanofluid. 20108 (1), 105– 114,  DOI: 10.1007/s10404-009-0450-5ViewGoogle Scholar
  7. 7Popel, A. S.; Johnson, P. C. Microcirculation and Hemorheology. Annu. Rev. Fluid Mech. 200537 (1), 43– 69,  DOI: 10.1146/annurev.fluid.37.042604.133933ViewGoogle Scholar
  8. 8Fedosov, D. A.; Peltomäki, M.; Gompper, G. Deformation and dynamics of red blood cells in flow through cylindrical microchannels. Soft Matter 201410 (24), 4258– 4267,  DOI: 10.1039/C4SM00248BViewGoogle Scholar
  9. 9Chakraborty, S. Dynamics of capillary flow of blood into a microfluidic channel. Lab Chip 20055 (4), 421– 430,  DOI: 10.1039/b414566fViewGoogle Scholar
  10. 10Tomaiuolo, G.; Guido, S. Start-up shape dynamics of red blood cells in microcapillary flow. Microvascular Research 201182 (1), 35– 41,  DOI: 10.1016/j.mvr.2011.03.004ViewGoogle Scholar
  11. 11Sherwood, J. M.; Dusting, J.; Kaliviotis, E.; Balabani, S. The effect of red blood cell aggregation on velocity and cell-depleted layer characteristics of blood in a bifurcating microchannel. Biomicrofluidics 20126 (2), 24119,  DOI: 10.1063/1.4717755ViewGoogle Scholar
  12. 12Nader, E.; Skinner, S.; Romana, M.; Fort, R.; Lemonne, N.; Guillot, N.; Gauthier, A.; Antoine-Jonville, S.; Renoux, C.; Hardy-Dessources, M.-D. Blood Rheology: Key Parameters, Impact on Blood Flow, Role in Sickle Cell Disease and Effects of Exercise. Frontiers in Physiology 201910, 01329,  DOI: 10.3389/fphys.2019.01329ViewGoogle Scholar
  13. 13Trejo-Soto, C.; Lázaro, G. R.; Pagonabarraga, I.; Hernández-Machado, A. Microfluidics Approach to the Mechanical Properties of Red Blood Cell Membrane and Their Effect on Blood Rheology. Membranes 202212 (2), 217,  DOI: 10.3390/membranes12020217ViewGoogle Scholar
  14. 14Wagner, C.; Steffen, P.; Svetina, S. Aggregation of red blood cells: From rouleaux to clot formation. Comptes Rendus Physique 201314 (6), 459– 469,  DOI: 10.1016/j.crhy.2013.04.004ViewGoogle Scholar
  15. 15Kim, H.; Zhbanov, A.; Yang, S. Microfluidic Systems for Blood and Blood Cell Characterization. Biosensors 202313 (1), 13,  DOI: 10.3390/bios13010013ViewGoogle Scholar
  16. 16Fåhræus, R.; Lindqvist, T. THE VISCOSITY OF THE BLOOD IN NARROW CAPILLARY TUBES. American Journal of Physiology-Legacy Content 193196 (3), 562– 568,  DOI: 10.1152/ajplegacy.1931.96.3.562ViewGoogle Scholar
  17. 17Ascolese, M.; Farina, A.; Fasano, A. The Fåhræus-Lindqvist effect in small blood vessels: how does it help the heart?. J. Biol. Phys. 201945 (4), 379– 394,  DOI: 10.1007/s10867-019-09534-4ViewGoogle Scholar
  18. 18Bento, D.; Fernandes, C. S.; Miranda, J. M.; Lima, R. In vitro blood flow visualizations and cell-free layer (CFL) measurements in a microchannel network. Experimental Thermal and Fluid Science 2019109, 109847,  DOI: 10.1016/j.expthermflusci.2019.109847ViewGoogle Scholar
  19. 19Namgung, B.; Ong, P. K.; Wong, Y. H.; Lim, D.; Chun, K. J.; Kim, S. A comparative study of histogram-based thresholding methods for the determination of cell-free layer width in small blood vessels. Physiological Measurement 201031 (9), N61,  DOI: 10.1088/0967-3334/31/9/N01ViewGoogle Scholar
  20. 20Hymel, S. J.; Lan, H.; Fujioka, H.; Khismatullin, D. B. Cell trapping in Y-junction microchannels: A numerical study of the bifurcation angle effect in inertial microfluidics. Phys. Fluids (1994) 201931 (8), 082003,  DOI: 10.1063/1.5113516ViewGoogle Scholar
  21. 21Li, X.; Popel, A. S.; Karniadakis, G. E. Blood-plasma separation in Y-shaped bifurcating microfluidic channels: a dissipative particle dynamics simulation study. Phys. Biol. 20129 (2), 026010,  DOI: 10.1088/1478-3975/9/2/026010ViewGoogle Scholar
  22. 22Yin, X.; Thomas, T.; Zhang, J. Multiple red blood cell flows through microvascular bifurcations: Cell free layer, cell trajectory, and hematocrit separation. Microvascular Research 201389, 47– 56,  DOI: 10.1016/j.mvr.2013.05.002ViewGoogle Scholar
  23. 23Shibeshi, S. S.; Collins, W. E. The Rheology of Blood Flow in a Branched Arterial System. Appl. Rheol 200515 (6), 398– 405,  DOI: 10.1515/arh-2005-0020ViewGoogle Scholar
  24. 24Sequeira, A.; Janela, J. An Overview of Some Mathematical Models of Blood Rheology. In A Portrait of State-of-the-Art Research at the Technical University of Lisbon; Pereira, M. S., Ed.; Springer Netherlands: Dordrecht, 2007; pp 65– 87.ViewGoogle Scholar
  25. 25Walburn, F. J.; Schneck, D. J. A constitutive equation for whole human blood. Biorheology 197613, 201– 210,  DOI: 10.3233/BIR-1976-13307ViewGoogle Scholar
  26. 26Quemada, D. A rheological model for studying the hematocrit dependence of red cell-red cell and red cell-protein interactions in blood. Biorheology 198118, 501– 516,  DOI: 10.3233/BIR-1981-183-615ViewGoogle Scholar
  27. 27Varchanis, S.; Dimakopoulos, Y.; Wagner, C.; Tsamopoulos, J. How viscoelastic is human blood plasma?. Soft Matter 201814 (21), 4238– 4251,  DOI: 10.1039/C8SM00061AViewGoogle Scholar
  28. 28Apostolidis, A. J.; Moyer, A. P.; Beris, A. N. Non-Newtonian effects in simulations of coronary arterial blood flow. J. Non-Newtonian Fluid Mech. 2016233, 155– 165,  DOI: 10.1016/j.jnnfm.2016.03.008ViewGoogle Scholar
  29. 29Luo, X. Y.; Kuang, Z. B. A study on the constitutive equation of blood. J. Biomech. 199225 (8), 929– 934,  DOI: 10.1016/0021-9290(92)90233-QViewGoogle Scholar
  30. 30Oldroyd, J. G.; Wilson, A. H. On the formulation of rheological equations of state. Proceedings of the Royal Society of London. Series A. Mathematical and Physical Sciences 1950200 (1063), 523– 541,  DOI: 10.1098/rspa.1950.0035ViewGoogle Scholar
  31. 31Prado, G.; Farutin, A.; Misbah, C.; Bureau, L. Viscoelastic transient of confined red blood cells. Biophys J. 2015108 (9), 2126– 2136,  DOI: 10.1016/j.bpj.2015.03.046ViewGoogle Scholar
  32. 32Huang, C. R.; Pan, W. D.; Chen, H. Q.; Copley, A. L. Thixotropic properties of whole blood from healthy human subjects. Biorheology 198724 (6), 795– 801,  DOI: 10.3233/BIR-1987-24630ViewGoogle Scholar
  33. 33Anand, M.; Kwack, J.; Masud, A. A new generalized Oldroyd-B model for blood flow in complex geometries. International Journal of Engineering Science 201372, 78– 88,  DOI: 10.1016/j.ijengsci.2013.06.009ViewGoogle Scholar
  34. 34Horner, J. S.; Armstrong, M. J.; Wagner, N. J.; Beris, A. N. Investigation of blood rheology under steady and unidirectional large amplitude oscillatory shear. J. Rheol. 201862 (2), 577– 591,  DOI: 10.1122/1.5017623ViewGoogle Scholar
  35. 35Horner, J. S.; Armstrong, M. J.; Wagner, N. J.; Beris, A. N. Measurements of human blood viscoelasticity and thixotropy under steady and transient shear and constitutive modeling thereof. J. Rheol. 201963 (5), 799– 813,  DOI: 10.1122/1.5108737ViewGoogle Scholar
  36. 36Armstrong, M.; Tussing, J. A methodology for adding thixotropy to Oldroyd-8 family of viscoelastic models for characterization of human blood. Phys. Fluids 202032 (9), 094111,  DOI: 10.1063/5.0022501ViewGoogle Scholar
  37. 37Crank, J.; Nicolson, P. A practical method for numerical evaluation of solutions of partial differential equations of the heat-conduction type. Mathematical Proceedings of the Cambridge Philosophical Society 194743 (1), 50– 67,  DOI: 10.1017/S0305004100023197ViewGoogle Scholar
  38. 38Clough, R. W. Original formulation of the finite element method. Finite Elements in Analysis and Design 19907 (2), 89– 101,  DOI: 10.1016/0168-874X(90)90001-UViewGoogle Scholar
  39. 39Liu, W. K.; Liu, Y.; Farrell, D.; Zhang, L.; Wang, X. S.; Fukui, Y.; Patankar, N.; Zhang, Y.; Bajaj, C.; Lee, J.Immersed finite element method and its applications to biological systems. Computer Methods in Applied Mechanics and Engineering 2006195 (13), 1722– 1749,  DOI: 10.1016/j.cma.2005.05.049ViewGoogle Scholar
  40. 40Lopes, D.; Agujetas, R.; Puga, H.; Teixeira, J.; Lima, R.; Alejo, J. P.; Ferrera, C. Analysis of finite element and finite volume methods for fluid-structure interaction simulation of blood flow in a real stenosed artery. International Journal of Mechanical Sciences 2021207, 106650,  DOI: 10.1016/j.ijmecsci.2021.106650ViewGoogle Scholar
  41. 41Favero, J. L.; Secchi, A. R.; Cardozo, N. S. M.; Jasak, H. Viscoelastic flow analysis using the software OpenFOAM and differential constitutive equations. J. Non-Newtonian Fluid Mech. 2010165 (23), 1625– 1636,  DOI: 10.1016/j.jnnfm.2010.08.010ViewGoogle Scholar
  42. 42Pimenta, F.; Alves, M. A. Stabilization of an open-source finite-volume solver for viscoelastic fluid flows. J. Non-Newtonian Fluid Mech. 2017239, 85– 104,  DOI: 10.1016/j.jnnfm.2016.12.002ViewGoogle Scholar
  43. 43Chee, C. Y.; Lee, H. P.; Lu, C. Using 3D fluid-structure interaction model to analyse the biomechanical properties of erythrocyte. Phys. Lett. A 2008372 (9), 1357– 1362,  DOI: 10.1016/j.physleta.2007.09.067ViewGoogle Scholar
  44. 44Xu, D.; Kaliviotis, E.; Munjiza, A.; Avital, E.; Ji, C.; Williams, J. Large scale simulation of red blood cell aggregation in shear flows. J. Biomech. 201346 (11), 1810– 1817,  DOI: 10.1016/j.jbiomech.2013.05.010ViewGoogle Scholar
  45. 45Johnson, K. L.; Kendall, K.; Roberts, A. Surface energy and the contact of elastic solids. Proceedings of the royal society of London. A. mathematical and physical sciences 1971324 (1558), 301– 313,  DOI: 10.1098/rspa.1971.0141ViewGoogle Scholar
  46. 46Shi, L.; Pan, T.-W.; Glowinski, R. Deformation of a single red blood cell in bounded Poiseuille flows. Phys. Rev. E 201285 (1), 016307,  DOI: 10.1103/PhysRevE.85.016307ViewGoogle Scholar
  47. 47Yoon, D.; You, D. Continuum modeling of deformation and aggregation of red blood cells. J. Biomech. 201649 (11), 2267– 2279,  DOI: 10.1016/j.jbiomech.2015.11.027ViewGoogle Scholar
  48. 48Mainardi, F.; Spada, G. Creep, relaxation and viscosity properties for basic fractional models in rheology. European Physical Journal Special Topics 2011193 (1), 133– 160,  DOI: 10.1140/epjst/e2011-01387-1ViewGoogle Scholar
  49. 49Gracka, M.; Lima, R.; Miranda, J. M.; Student, S.; Melka, B.; Ostrowski, Z. Red blood cells tracking and cell-free layer formation in a microchannel with hyperbolic contraction: A CFD model validation. Computer Methods and Programs in Biomedicine 2022226, 107117,  DOI: 10.1016/j.cmpb.2022.107117ViewGoogle Scholar
  50. 50Aryan, H.; Beigzadeh, B.; Siavashi, M. Euler-Lagrange numerical simulation of improved magnetic drug delivery in a three-dimensional CT-based carotid artery bifurcation. Computer Methods and Programs in Biomedicine 2022219, 106778,  DOI: 10.1016/j.cmpb.2022.106778ViewGoogle Scholar
  51. 51Czaja, B.; Závodszky, G.; Azizi Tarksalooyeh, V.; Hoekstra, A. G. Cell-resolved blood flow simulations of saccular aneurysms: effects of pulsatility and aspect ratio. J. R Soc. Interface 201815 (146), 20180485,  DOI: 10.1098/rsif.2018.0485ViewGoogle Scholar
  52. 52Rydquist, G.; Esmaily, M. A cell-resolved, Lagrangian solver for modeling red blood cell dynamics in macroscale flows. J. Comput. Phys. 2022461, 111204,  DOI: 10.1016/j.jcp.2022.111204ViewGoogle Scholar
  53. 53Dadvand, A.; Baghalnezhad, M.; Mirzaee, I.; Khoo, B. C.; Ghoreishi, S. An immersed boundary-lattice Boltzmann approach to study the dynamics of elastic membranes in viscous shear flows. Journal of Computational Science 20145 (5), 709– 718,  DOI: 10.1016/j.jocs.2014.06.006ViewGoogle Scholar
  54. 54Krüger, T.; Holmes, D.; Coveney, P. V. Deformability-based red blood cell separation in deterministic lateral displacement devices─A simulation study. Biomicrofluidics 20148 (5), 054114,  DOI: 10.1063/1.4897913ViewGoogle Scholar
  55. 55Takeishi, N.; Ito, H.; Kaneko, M.; Wada, S. Deformation of a Red Blood Cell in a Narrow Rectangular Microchannel. Micromachines 201910 (3), 199,  DOI: 10.3390/mi10030199ViewGoogle Scholar
  56. 56Krüger, T.; Varnik, F.; Raabe, D. Efficient and accurate simulations of deformable particles immersed in a fluid using a combined immersed boundary lattice Boltzmann finite element method. Computers & Mathematics with Applications 201161 (12), 3485– 3505,  DOI: 10.1016/j.camwa.2010.03.057ViewGoogle Scholar
  57. 57Balachandran Nair, A. N.; Pirker, S.; Umundum, T.; Saeedipour, M. A reduced-order model for deformable particles with application in bio-microfluidics. Computational Particle Mechanics 20207 (3), 593– 601,  DOI: 10.1007/s40571-019-00283-8ViewGoogle Scholar
  58. 58Balachandran Nair, A. N.; Pirker, S.; Saeedipour, M. Resolved CFD-DEM simulation of blood flow with a reduced-order RBC model. Computational Particle Mechanics 20229 (4), 759– 774,  DOI: 10.1007/s40571-021-00441-xViewGoogle Scholar
  59. 59Mittal, R.; Iaccarino, G. IMMERSED BOUNDARY METHODS. Annu. Rev. Fluid Mech. 200537 (1), 239– 261,  DOI: 10.1146/annurev.fluid.37.061903.175743ViewGoogle Scholar
  60. 60Piquet, A.; Roussel, O.; Hadjadj, A. A comparative study of Brinkman penalization and direct-forcing immersed boundary methods for compressible viscous flows. Computers & Fluids 2016136, 272– 284,  DOI: 10.1016/j.compfluid.2016.06.001ViewGoogle Scholar
  61. 61Akerkouch, L.; Le, T. B. A Hybrid Continuum-Particle Approach for Fluid-Structure Interaction Simulation of Red Blood Cells in Fluid Flows. Fluids 20216 (4), 139,  DOI: 10.3390/fluids6040139ViewGoogle Scholar
  62. 62Barker, A. T.; Cai, X.-C. Scalable parallel methods for monolithic coupling in fluid-structure interaction with application to blood flow modeling. J. Comput. Phys. 2010229 (3), 642– 659,  DOI: 10.1016/j.jcp.2009.10.001ViewGoogle Scholar
  63. 63Cetin, A.; Sahin, M. A monolithic fluid-structure interaction framework applied to red blood cells. International Journal for Numerical Methods in Biomedical Engineering 201935 (2), e3171  DOI: 10.1002/cnm.3171ViewGoogle Scholar
  64. 64Freund, J. B. Numerical Simulation of Flowing Blood Cells. Annu. Rev. Fluid Mech. 201446 (1), 67– 95,  DOI: 10.1146/annurev-fluid-010313-141349ViewGoogle Scholar
  65. 65Ye, T.; Phan-Thien, N.; Lim, C. T. Particle-based simulations of red blood cells─A review. J. Biomech. 201649 (11), 2255– 2266,  DOI: 10.1016/j.jbiomech.2015.11.050ViewGoogle Scholar
  66. 66Arabghahestani, M.; Poozesh, S.; Akafuah, N. K. Advances in Computational Fluid Mechanics in Cellular Flow Manipulation: A Review. Applied Sciences 20199 (19), 4041,  DOI: 10.3390/app9194041ViewGoogle Scholar
  67. 67Rathnayaka, C. M.; From, C. S.; Geekiyanage, N. M.; Gu, Y. T.; Nguyen, N. T.; Sauret, E. Particle-Based Numerical Modelling of Liquid Marbles: Recent Advances and Future Perspectives. Archives of Computational Methods in Engineering 202229 (5), 3021– 3039,  DOI: 10.1007/s11831-021-09683-7ViewGoogle Scholar
  68. 68Li, X.; Vlahovska, P. M.; Karniadakis, G. E. Continuum- and particle-based modeling of shapes and dynamics of red blood cells in health and disease. Soft Matter 20139 (1), 28– 37,  DOI: 10.1039/C2SM26891DViewGoogle Scholar
  69. 69Beris, A. N.; Horner, J. S.; Jariwala, S.; Armstrong, M. J.; Wagner, N. J. Recent advances in blood rheology: a review. Soft Matter 202117 (47), 10591– 10613,  DOI: 10.1039/D1SM01212FViewGoogle Scholar
  70. 70Arciero, J.; Causin, P.; Malgaroli, F. Mathematical methods for modeling the microcirculation. AIMS Biophysics 20174 (3), 362– 399,  DOI: 10.3934/biophy.2017.3.362ViewGoogle Scholar
  71. 71Maria, M. S.; Chandra, T. S.; Sen, A. K. Capillary flow-driven blood plasma separation and on-chip analyte detection in microfluidic devices. Microfluid. Nanofluid. 201721 (4), 72,  DOI: 10.1007/s10404-017-1907-6ViewGoogle Scholar
  72. 72Huhtamäki, T.; Tian, X.; Korhonen, J. T.; Ras, R. H. A. Surface-wetting characterization using contact-angle measurements. Nat. Protoc. 201813 (7), 1521– 1538,  DOI: 10.1038/s41596-018-0003-zViewGoogle Scholar
  73. 73Young, T., III. An essay on the cohesion of fluids. Philosophical Transactions of the Royal Society of London 180595, 65– 87,  DOI: 10.1098/rstl.1805.0005ViewGoogle Scholar
  74. 74Kim, Y. C.; Kim, S.-H.; Kim, D.; Park, S.-J.; Park, J.-K. Plasma extraction in a capillary-driven microfluidic device using surfactant-added poly(dimethylsiloxane). Sens. Actuators, B 2010145 (2), 861– 868,  DOI: 10.1016/j.snb.2010.01.017ViewGoogle Scholar
  75. 75Washburn, E. W. The Dynamics of Capillary Flow. Physical Review 192117 (3), 273– 283,  DOI: 10.1103/PhysRev.17.273ViewGoogle Scholar
  76. 76Cito, S.; Ahn, Y. C.; Pallares, J.; Duarte, R. M.; Chen, Z.; Madou, M.; Katakis, I. Visualization and measurement of capillary-driven blood flow using spectral domain optical coherence tomography. Microfluid Nanofluidics 201213 (2), 227– 237,  DOI: 10.1007/s10404-012-0950-6ViewGoogle Scholar
  77. 77Berthier, E.; Dostie, A. M.; Lee, U. N.; Berthier, J.; Theberge, A. B. Open Microfluidic Capillary Systems. Anal Chem. 201991 (14), 8739– 8750,  DOI: 10.1021/acs.analchem.9b01429ViewGoogle Scholar
  78. 78Berthier, J.; Brakke, K. A.; Furlani, E. P.; Karampelas, I. H.; Poher, V.; Gosselin, D.; Cubizolles, M.; Pouteau, P. Whole blood spontaneous capillary flow in narrow V-groove microchannels. Sens. Actuators, B 2015206, 258– 267,  DOI: 10.1016/j.snb.2014.09.040ViewGoogle Scholar
  79. 79Hirt, C. W.; Nichols, B. D. Volume of fluid (VOF) method for the dynamics of free boundaries. J. Comput. Phys. 198139 (1), 201– 225,  DOI: 10.1016/0021-9991(81)90145-5ViewGoogle Scholar
  80. 80Chen, J.-L.; Shih, W.-H.; Hsieh, W.-H. AC electro-osmotic micromixer using a face-to-face, asymmetric pair of planar electrodes. Sens. Actuators, B 2013188, 11– 21,  DOI: 10.1016/j.snb.2013.07.012ViewGoogle Scholar
  81. 81Zhao, C.; Yang, C. Electrokinetics of non-Newtonian fluids: A review. Advances in Colloid and Interface Science 2013201-202, 94– 108,  DOI: 10.1016/j.cis.2013.09.001ViewGoogle Scholar
  82. 82Oh, K. W. 6 – Lab-on-chip (LOC) devices and microfluidics for biomedical applications. In MEMS for Biomedical Applications; Bhansali, S., Vasudev, A., Eds.; Woodhead Publishing, 2012; pp 150– 171.ViewGoogle Scholar
  83. 83Bello, M. S.; De Besi, P.; Rezzonico, R.; Righetti, P. G.; Casiraghi, E. Electroosmosis of polymer solutions in fused silica capillaries. ELECTROPHORESIS 199415 (1), 623– 626,  DOI: 10.1002/elps.1150150186ViewGoogle Scholar
  84. 84Park, H. M.; Lee, W. M. Effect of viscoelasticity on the flow pattern and the volumetric flow rate in electroosmotic flows through a microchannel. Lab Chip 20088 (7), 1163– 1170,  DOI: 10.1039/b800185eViewGoogle Scholar
  85. 85Afonso, A. M.; Alves, M. A.; Pinho, F. T. Analytical solution of mixed electro-osmotic/pressure driven flows of viscoelastic fluids in microchannels. J. Non-Newtonian Fluid Mech. 2009159 (1), 50– 63,  DOI: 10.1016/j.jnnfm.2009.01.006ViewGoogle Scholar
  86. 86Sousa, J. J.; Afonso, A. M.; Pinho, F. T.; Alves, M. A. Effect of the skimming layer on electro-osmotic─Poiseuille flows of viscoelastic fluids. Microfluid. Nanofluid. 201110 (1), 107– 122,  DOI: 10.1007/s10404-010-0651-yViewGoogle Scholar
  87. 87Zhao, C.; Yang, C. Electro-osmotic mobility of non-Newtonian fluids. Biomicrofluidics 20115 (1), 014110,  DOI: 10.1063/1.3571278ViewGoogle Scholar
  88. 88Pimenta, F.; Alves, M. A. Electro-elastic instabilities in cross-shaped microchannels. J. Non-Newtonian Fluid Mech. 2018259, 61– 77,  DOI: 10.1016/j.jnnfm.2018.04.004ViewGoogle Scholar
  89. 89Bezerra, W. S.; Castelo, A.; Afonso, A. M. Numerical Study of Electro-Osmotic Fluid Flow and Vortex Formation. Micromachines (Basel) 201910 (12), 796,  DOI: 10.3390/mi10120796ViewGoogle Scholar
  90. 90Ji, J.; Qian, S.; Liu, Z. Electroosmotic Flow of Viscoelastic Fluid through a Constriction Microchannel. Micromachines (Basel) 202112 (4), 417,  DOI: 10.3390/mi12040417ViewGoogle Scholar
  91. 91Zhao, C.; Yang, C. Exact solutions for electro-osmotic flow of viscoelastic fluids in rectangular micro-channels. Applied Mathematics and Computation 2009211 (2), 502– 509,  DOI: 10.1016/j.amc.2009.01.068ViewGoogle Scholar
  92. 92Gerum, R.; Mirzahossein, E.; Eroles, M.; Elsterer, J.; Mainka, A.; Bauer, A.; Sonntag, S.; Winterl, A.; Bartl, J.; Fischer, L. Viscoelastic properties of suspended cells measured with shear flow deformation cytometry. Elife 202211, e78823,  DOI: 10.7554/eLife.78823ViewGoogle Scholar
  93. 93Sadek, S. H.; Pinho, F. T.; Alves, M. A. Electro-elastic flow instabilities of viscoelastic fluids in contraction/expansion micro-geometries. J. Non-Newtonian Fluid Mech. 2020283, 104293,  DOI: 10.1016/j.jnnfm.2020.104293ViewGoogle Scholar
  94. 94Spanjaards, M.; Peters, G.; Hulsen, M.; Anderson, P. Numerical Study of the Effect of Thixotropy on Extrudate Swell. Polymers 202113 (24), 4383,  DOI: 10.3390/polym13244383ViewGoogle Scholar
  95. 95Rashidi, S.; Bafekr, H.; Valipour, M. S.; Esfahani, J. A. A review on the application, simulation, and experiment of the electrokinetic mixers. Chemical Engineering and Processing – Process Intensification 2018126, 108– 122,  DOI: 10.1016/j.cep.2018.02.021ViewGoogle Scholar
  96. 96Matsubara, K.; Narumi, T. Microfluidic mixing using unsteady electroosmotic vortices produced by a staggered array of electrodes. Chemical Engineering Journal 2016288, 638– 647,  DOI: 10.1016/j.cej.2015.12.013ViewGoogle Scholar
  97. 97Qaderi, A.; Jamaati, J.; Bahiraei, M. CFD simulation of combined electroosmotic-pressure driven micro-mixing in a microchannel equipped with triangular hurdle and zeta-potential heterogeneity. Chemical Engineering Science 2019199, 463– 477,  DOI: 10.1016/j.ces.2019.01.034ViewGoogle Scholar
  98. 98Cho, C.-C.; Chen, C.-L.; Chen, C. o.-K. Mixing enhancement in crisscross micromixer using aperiodic electrokinetic perturbing flows. International Journal of Heat and Mass Transfer 201255 (11), 2926– 2933,  DOI: 10.1016/j.ijheatmasstransfer.2012.02.006ViewGoogle Scholar
  99. 99Zhao, W.; Yang, F.; Wang, K.; Bai, J.; Wang, G. Rapid mixing by turbulent-like electrokinetic microflow. Chemical Engineering Science 2017165, 113– 121,  DOI: 10.1016/j.ces.2017.02.027ViewGoogle Scholar
  100. 100Tran, T.; Chakraborty, P.; Guttenberg, N.; Prescott, A.; Kellay, H.; Goldburg, W.; Goldenfeld, N.; Gioia, G. Macroscopic effects of the spectral structure in turbulent flows. Nat. Phys. 20106 (6), 438– 441,  DOI: 10.1038/nphys1674ViewGoogle Scholar
  101. 101Toner, M.; Irimia, D. Blood-on-a-chip. Annu. Rev. Biomed Eng. 20057, 77– 103,  DOI: 10.1146/annurev.bioeng.7.011205.135108ViewGoogle Scholar
  102. 102Maria, M. S.; Rakesh, P. E.; Chandra, T. S.; Sen, A. K. Capillary flow of blood in a microchannel with differential wetting for blood plasma separation and on-chip glucose detection. Biomicrofluidics 201610 (5), 054108,  DOI: 10.1063/1.4962874ViewGoogle Scholar
  103. 103Tripathi, S.; Varun Kumar, Y. V. B.; Prabhakar, A.; Joshi, S. S.; Agrawal, A. Passive blood plasma separation at the microscale: a review of design principles and microdevices. Journal of Micromechanics and Microengineering 201525 (8), 083001,  DOI: 10.1088/0960-1317/25/8/083001ViewGoogle Scholar
  104. 104Mohammadi, M.; Madadi, H.; Casals-Terré, J. Microfluidic point-of-care blood panel based on a novel technique: Reversible electroosmotic flow. Biomicrofluidics 20159 (5), 054106,  DOI: 10.1063/1.4930865ViewGoogle Scholar
  105. 105Kang, D. H.; Kim, K.; Kim, Y. J. An anti-clogging method for improving the performance and lifespan of blood plasma separation devices in real-time and continuous microfluidic systems. Sci. Rep 20188 (1), 17015,  DOI: 10.1038/s41598-018-35235-4ViewGoogle Scholar
  106. 106Li, Z.; Pollack, G. H. Surface-induced flow: A natural microscopic engine using infrared energy as fuel. Science Advances 20206 (19), eaba0941  DOI: 10.1126/sciadv.aba0941ViewGoogle Scholar
  107. 107Mercado-Uribe, H.; Guevara-Pantoja, F. J.; García-Muñoz, W.; García-Maldonado, J. S.; Méndez-Alcaraz, J. M.; Ruiz-Suárez, J. C. On the evolution of the exclusion zone produced by hydrophilic surfaces: A contracted description. J. Chem. Phys. 2021154 (19), 194902,  DOI: 10.1063/5.0043084ViewGoogle Scholar
  108. 108Yalcin, O.; Jani, V. P.; Johnson, P. C.; Cabrales, P. Implications Enzymatic Degradation of the Endothelial Glycocalyx on the Microvascular Hemodynamics and the Arteriolar Red Cell Free Layer of the Rat Cremaster Muscle. Front Physiol 20189, 168,  DOI: 10.3389/fphys.2018.00168ViewGoogle Scholar
Figure 3. FLOW-3D results for Strathcona Dam spillway with all gates fully open at an elevated reservoir level during passage of a large flood. Note the effects of poor approach conditions and pier overtopping at the leftmost bay.

BC Hydro Assesses Spillway Hydraulics with FLOW-3D

by Faizal Yusuf, M.A.Sc., P.Eng.
Specialist Engineer in the Hydrotechnical Department at BC Hydro

BC Hydro, a public electric utility in British Columbia, uses FLOW-3D to investigate complex hydraulics issues at several existing dams and to assist in the design and optimization of proposed facilities.

Faizal Yusuf, M.A.Sc., P.Eng., Specialist Engineer in the Hydrotechnical department at BC Hydro, presents three case studies that highlight the application of FLOW-3D to different types of spillways and the importance of reliable prototype or physical hydraulic model data for numerical model calibration.

W.A.C. Bennett Dam
At W.A.C. Bennett Dam, differences in the spillway geometry between the physical hydraulic model from the 1960s and the prototype make it difficult to draw reliable conclusions on shock wave formation and chute capacity from physical model test results. The magnitude of shock waves in the concrete-lined spillway chute are strongly influenced by a 44% reduction in the chute width downstream of the three radial gates at the headworks, as well as the relative openings of the radial gates. The shock waves lead to locally higher water levels that have caused overtopping of the chute walls under certain historical operations.Prototype spill tests for discharges up to 2,865 m3/s were performed in 2012 to provide surveyed water surface profiles along chute walls, 3D laser scans of the water surface in the chute and video of flow patterns for FLOW-3D model calibration. Excellent agreement was obtained between the numerical model and field observations, particularly for the location and height of the first shock wave at the chute walls (Figure 1).

W.A.C에서 Bennett Dam, 1960년대의 물리적 수력학 모델과 프로토타입 사이의 여수로 형상의 차이로 인해 물리적 모델 테스트 결과에서 충격파 형성 및 슈트 용량에 대한 신뢰할 수 있는 결론을 도출하기 어렵습니다. 콘크리트 라이닝 방수로 낙하산의 충격파 크기는 방사형 게이트의 상대적인 개구부뿐만 아니라 헤드워크에 있는 3개의 방사형 게이트 하류의 슈트 폭이 44% 감소함에 따라 크게 영향을 받습니다. 충격파는 특정 역사적 작업에서 슈트 벽의 범람을 야기한 국부적으로 더 높은 수위로 이어집니다. 최대 2,865m3/s의 배출에 대한 프로토타입 유출 테스트가 2012년에 수행되어 슈트 벽을 따라 조사된 수면 프로필, 3D 레이저 스캔을 제공했습니다. FLOW-3D 모델 보정을 위한 슈트의 수면 및 흐름 패턴 비디오. 특히 슈트 벽에서 첫 번째 충격파의 위치와 높이에 대해 수치 모델과 현장 관찰 간에 탁월한 일치가 이루어졌습니다(그림 1).
Figure 1. Comparison between prototype observations and FLOW-3D for a spill discharge of 2,865 m^3/s at Bennett Dam spillway.
Figure 1. Comparison between prototype observations and FLOW-3D for a spill discharge of 2,865 m^3/s at Bennett Dam spillway.

The calibrated FLOW-3D model confirmed that the design flood could be safely passed without overtopping the spillway chute walls as long as all three radial gates are opened as prescribed in existing operating orders with the outer gates open more than the inner gate.

The CFD model also provided insight into the concrete damage in the spillway chute. Cavitation indices computed from FLOW-3D simulation results were compared with empirical data from the USBR and found to be consistent with the historical performance of the spillway. The numerical analysis supported field inspections, which concluded that deterioration of the concrete conditions in the chute is likely not due to cavitation.

Strathcona Dam
FLOW-3D was used to investigate poor approach conditions and uncertainties with the rating curves for Strathcona Dam spillway, which includes three vertical lift gates on the right abutment of the dam. The rating curves for Strathcona spillway were developed from a combination of empirical adjustments and limited physical hydraulic model testing in a flume that did not include geometry of the piers and abutments.

Numerical model testing and calibration was based on comparisons with prototype spill observations from 1982 when all three gates were fully open, resulting in a large depression in the water surface upstream of the leftmost bay (Figure 2). The approach flow to the leftmost bay is distorted by water flowing parallel to the dam axis and plunging over the concrete retaining wall adjacent to the upstream slope of the earthfill dam. The flow enters the other two bays much more smoothly. In addition to very similar flow patterns produced in the numerical model compared to the prototype, simulated water levels at the gate section matched 1982 field measurements to within 0.1 m.

보정된 FLOW-3D 모델은 외부 게이트가 내부 게이트보다 더 많이 열려 있는 기존 운영 명령에 규정된 대로 3개의 방사형 게이트가 모두 열리는 한 여수로 낙하산 벽을 넘지 않고 설계 홍수를 안전하게 통과할 수 있음을 확인했습니다.

CFD 모델은 방수로 낙하산의 콘크리트 손상에 대한 통찰력도 제공했습니다. FLOW-3D 시뮬레이션 결과에서 계산된 캐비테이션 지수는 USBR의 경험적 데이터와 비교되었으며 여수로의 역사적 성능과 일치하는 것으로 나타났습니다. 수치 분석은 현장 검사를 지원했으며, 슈트의 콘크리트 상태 악화는 캐비테이션 때문이 아닐 가능성이 높다고 결론지었습니다.

Strathcona 댐
FLOW-3D는 Strathcona Dam 여수로에 대한 등급 곡선을 사용하여 열악한 접근 조건과 불확실성을 조사하는 데 사용되었습니다. 여기에는 댐의 오른쪽 접합부에 3개의 수직 리프트 게이트가 포함되어 있습니다. Strathcona 여수로에 대한 등급 곡선은 경험적 조정과 교각 및 교대의 형상을 포함하지 않는 수로에서 제한된 물리적 수리 모델 테스트의 조합으로 개발되었습니다.

수치 모델 테스트 및 보정은 세 개의 수문이 모두 완전히 개방된 1982년의 프로토타입 유출 관측과의 비교를 기반으로 했으며, 그 결과 가장 왼쪽 만의 상류 수면에 큰 함몰이 발생했습니다(그림 2). 최좌단 만으로의 접근 흐름은 댐 축과 평행하게 흐르는 물과 흙채움댐의 상류 경사면에 인접한 콘크리트 옹벽 위로 떨어지는 물에 의해 왜곡됩니다. 흐름은 훨씬 더 원활하게 다른 두 베이로 들어갑니다. 프로토타입과 비교하여 수치 모델에서 생성된 매우 유사한 흐름 패턴 외에도 게이트 섹션에서 시뮬레이션된 수위는 1982년 현장 측정과 0.1m 이내로 일치했습니다.

Figure 2. Prototype observations and FLOW-3D results for a Strathcona Dam spill in 1982 with all three gates fully open.
Figure 2. Prototype observations and FLOW-3D results for a Strathcona Dam spill in 1982 with all three gates fully open.

The calibrated CFD model produces discharges within 5% of the spillway rating curve for the reservoir’s normal operating range with all gates fully open. However, at higher reservoir levels, which may occur during passage of large floods (as shown in Figure 3), the difference between simulated discharges and the rating curves are greater than 10% as the physical model testing with simplified geometry and empirical corrections did not adequately represent the complex approach flow patterns. The FLOW-3D model provided further insight into the accuracy of rating curves for individual bays, gated conditions and the transition between orifice and free surface flow.

보정된 CFD 모델은 모든 게이트가 완전히 열린 상태에서 저수지의 정상 작동 범위에 대한 여수로 등급 곡선의 5% 이내에서 배출을 생성합니다. 그러나 대규모 홍수가 통과하는 동안 발생할 수 있는 더 높은 저수지 수위에서는(그림 3 참조) 단순화된 기하학과 경험적 수정을 사용한 물리적 모델 테스트가 그렇지 않았기 때문에 모의 배출과 등급 곡선 간의 차이는 10% 이상입니다. 복잡한 접근 흐름 패턴을 적절하게 표현합니다. FLOW-3D 모델은 개별 베이, 게이트 조건 및 오리피스와 자유 표면 흐름 사이의 전환에 대한 등급 곡선의 정확도에 대한 추가 통찰력을 제공했습니다.

Figure 3. FLOW-3D results for Strathcona Dam spillway with all gates fully open at an elevated reservoir level during passage of a large flood. Note the effects of poor approach conditions and pier overtopping at the leftmost bay.
Figure 3. FLOW-3D results for Strathcona Dam spillway with all gates fully open at an elevated reservoir level during passage of a large flood. Note the effects of poor approach conditions and pier overtopping at the leftmost bay.

John Hart Dam
The John Hart concrete dam will be modified to include a new free crest spillway to be situated between an existing gated spillway and a low level outlet structure that is currently under construction. Significant improvements in the design of the proposed spillway were made through a systematic optimization process using FLOW-3D.

The preliminary design of the free crest spillway was based on engineering hydraulic design guides. Concrete apron blocks are intended to protect the rock at the toe of the dam. A new right training wall will guide the flow from the new spillway towards the tailrace pool and protect the low level outlet structure from spillway discharges.

FLOW-3D model results for the initial and optimized design of the new spillway are shown in Figure 4. CFD analysis led to a 10% increase in discharge capacity, significant decrease in roadway impingement above the spillway crest and improved flow patterns including up to a 5 m reduction in water levels along the proposed right wall. Physical hydraulic model testing will be used to confirm the proposed design.

존 하트 댐
John Hart 콘크리트 댐은 현재 건설 중인 기존 배수로와 저층 배수로 사이에 위치할 새로운 자유 마루 배수로를 포함하도록 수정될 것입니다. FLOW-3D를 사용한 체계적인 최적화 프로세스를 통해 제안된 여수로 설계의 상당한 개선이 이루어졌습니다.

자유 마루 여수로의 예비 설계는 엔지니어링 수력학 설계 가이드를 기반으로 했습니다. 콘크리트 앞치마 블록은 댐 선단부의 암석을 보호하기 위한 것입니다. 새로운 오른쪽 훈련 벽은 새 여수로에서 테일레이스 풀로 흐름을 안내하고 여수로 배출로부터 낮은 수준의 배출구 구조를 보호합니다.

새 여수로의 초기 및 최적화된 설계에 대한 FLOW-3D 모델 결과는 그림 4에 나와 있습니다. CFD 분석을 통해 방류 용량이 10% 증가하고 여수로 마루 위의 도로 충돌이 크게 감소했으며 최대 제안된 오른쪽 벽을 따라 수위가 5m 감소합니다. 제안된 설계를 확인하기 위해 물리적 수압 모델 테스트가 사용됩니다.

Figure 4. FLOW-3D model results for the preliminary and optimized layout of the proposed spillway at John Hart Dam.
Figure 4. FLOW-3D model results for the preliminary and optimized layout of the proposed spillway at John Hart Dam.

Conclusion

BC Hydro has been using FLOW-3D to investigate a wide range of challenging hydraulics problems for different types of spillways and water conveyance structures leading to a greatly improved understanding of flow patterns and performance. Prototype data and reliable physical hydraulic model testing are used whenever possible to improve confidence in the numerical model results.

다양한 유형의 여수로 및 물 수송 구조로 인해 흐름 패턴 및 성능에 대한 이해가 크게 향상되었습니다. 프로토타입 데이터와 신뢰할 수 있는 물리적 유압 모델 테스트는 수치 모델 결과의 신뢰도를 향상시키기 위해 가능할 때마다 사용됩니다.

About Flow Science, Inc.
Based in Santa Fe, New Mexico USA, Flow Science was founded in 1980 by Dr. C. W. (Tony) Hirt, who was one of the principals in pioneering the “Volume-of-Fluid” or VOF method while working at the Los Alamos National Lab. FLOW-3D is a direct descendant of this work, and in the subsequent years, we have increased its sophistication with TruVOF, boasting pioneering improvements in the speed and accuracy of tracking distinct liquid/gas interfaces. Today, Flow Science products offer complete multiphysics simulation with diverse modeling capabilities including fluid-structure interaction, 6-DoF moving objects, and multiphase flows. From inception, our vision has been to provide our customers with excellence in flow modeling software and services.

Fig. 4. Meshed quarter aluminum model with HAZ regions and support steel plates.

Benchmark study on slamming response of flat-stiffened plates considering fluid-structure interaction

유체-구조 상호작용을 고려한 평판 보강판의 슬래밍 응답에 대한 벤치마크 연구

Dac DungTruongabBeom-SeonJangaCarl-ErikJansoncJonas W.RingsbergcYasuhiraYamadadKotaTakamotofYasumiKawamuraeHan-BaekJua
aResearch Institute of Marine Systems Engineering, Department of Naval Architecture and Ocean Engineering, Seoul National University, Seoul, South Korea
bDepartment of Engineering Mechanics, Nha Trang University, Nha Trang, Viet Nam
cDivision of Marine Technology, Department of Mechanics and Maritime Sciences, Chalmers University of Technology, Gothenburg, Sweden
dNational Maritime Research Institute, National Institute of Maritime, Port and Aviation Technology, Tokyo, Japan
eDepartment of Systems Design for Ocean-Space, Yokohama National University, Kanagawa, Japan
fDepartment of Mechanical Systems Engineering, Tokyo University of Agriculture and Technology, Tokyo, Japan

ABSTRACT

이 논문은 해양구조물의 평보강판의 슬래밍 반응에 대한 벤치마크 연구를 제시합니다. 목표는 유체-구조 상호작용(FSI) 시뮬레이션 방법론, 모델링 기술 및 슬래밍 압력 예측에 대한 기존 연구원의 경험을 비교하는 것이었습니다.

수치 FSI 시뮬레이션을 위해 가장 일반적인 상용 소프트웨어 패키지를 사용하는 3개의 연구 그룹(예: LS-Dyna ALE, LS-Dyna ICFD, ANSYS CFX 및 Star-CCM+/ABAQUS)이 이 연구에 참여했습니다.

공개 문헌에서 입수할 수 있는 경량 선박과 같은 바닥 구조의 평평한 강화 알루미늄 판에 대한 습식 낙하 시험 데이터는 FSI 모델링의 검증에 활용되었습니다. 형상 모델 및 재료 속성을 포함한 실험 조건의 요약은 시뮬레이션 전에 참가자에게 배포되었습니다.

충돌 속도와 강판의 강성이 슬래밍 응답에 미치는 영향을 조사하기 위해 해양 설비에 사용되는 실제 치수를 갖는 평판 보강 강판에 대한 매개변수 연구를 수행했습니다. 보강판에 작용하는 전체 수직력에 대한 FE 시뮬레이션 결과와 이러한 힘에 대한 구조적 반응을 참가자로부터 획득하여 분석 및 비교하였다.

앞서 언급한 상용 FSI 소프트웨어 패키지를 사용하여 슬래밍 부하에 대한 신뢰할 수 있고 정확한 예측을 평가했습니다. 또한 FSI 시뮬레이션에서 관찰된 동일한 영구 처짐을 초래하는 등가 정적 슬래밍 압력을 보고하고 분류 표준 DNV에서 제안한 해석 모델 및 슬래밍 압력 계산을 위한 기존 실험 데이터와 비교했습니다.

연구 결과는 등가 하중 모델이 물 충돌 속도와 플레이트 강성에 의존한다는 것을 보여주었습니다. 즉, 등가정압계수는 충돌속도가 증가함에 따라 감소하고 충돌구조가 더 단단해지면 증가한다.

This paper presents a benchmark study on the slamming responses of offshore structures’ flat-stiffened plates. The objective was to compare the fluid-structure interaction (FSI) simulation methodologies, modeling techniques, and established researchers’ experiences in predicting slamming pressure. Three research groups employing the most common commercial software packages for numerical FSI simulations (i.e. LS-Dyna ALE, LS-Dyna ICFD, ANSYS CFX, and Star-CCM+/ABAQUS) participated in this study. Wet drop test data on flat-stiffened aluminum plates of light-ship-like bottom structures available in the open literature was utilized for validation of the FSI modeling. A summary of the experimental conditions including the geometry model and material properties, was distributed to the participants prior to their simulations. A parametric study on flat-stiffened steel plates having actual scantlings used in marine installations was performed to investigate the effect of impact velocity and plate rigidity on slamming response. The FE simulation results for the total vertical forces acting on the stiffened plates and their structural responses to those forces, as obtained from the participants, were analyzed and compared. The reliable and accurate predictions of slamming loads using the aforementioned commercial FSI software packages were evaluated. Additionally, equivalent static slamming pressures resulting in the same permanent deflections, as observed from the FSI simulations, were reported and compared with analytical models proposed by the Classification Standards DNV and existing experimental data for calculation of the slamming pressure. The study results showed that the equivalent load model depends on the water impact velocity and plate rigidity; that is, the equivalent static pressure coefficient decreases with an increase in impact velocity, and increases when impacting structures become stiffer.

Fig. 4. Meshed quarter aluminum model with HAZ regions and support steel plates.
Fig. 4. Meshed quarter aluminum model with HAZ regions and support steel plates.
Fig. 6. (a) Boundary conditions of water hitting case and (b) water jets at end of the simulation.
Fig. 6. (a) Boundary conditions of water hitting case and (b) water jets at end of the simulation.
Fig. 7. Comparison of prediction and test results for deflection time history of (a) D1 and (b) D2 for Vi = 2.3 m/s.
Fig. 7. Comparison of prediction and test results for deflection time history of (a) D1 and (b) D2 for Vi = 2.3 m/s.
Fig. 8. Comparison of prediction and test results for maximum deflection with different impact velocities.
Fig. 8. Comparison of prediction and test results for maximum deflection with different impact velocities.
Fig. 16. Boundary conditions applied to present FSI simulations (Sym. denotes symmetric, and Cons. denotes constrained)
Fig. 16. Boundary conditions applied to present FSI simulations (Sym. denotes symmetric, and Cons. denotes constrained)
Fig. 24. Distribution of deflections at moment of maximum deflection in: (a) LS-Dyna ALE, (b) Star-CCM+/ABAQUS, (c) ANSYS CFD, and (d) LSDyna ICFD (unit: m).

Keywords

Benchmark studyEquivalent static pressureFlat-stiffened plateFluid-structure interactionPermanent deflectionSlamming pressure coefficient

References

[1] Von Karman TH. The impact on seaplane floats during landing. Washington, DC: National Advisory Committee for Aeronautics; 1929. Technical note No.: 321.
[2] Wagner VH. Über Stoß- und Gleitvorgange ¨ an der Oberflache ¨ von Flüssigkeiten. Z Angew Math Mech 1932;12(4):193–215.
[3] Chuang SL. Experiments on flat-bottom slamming. J Ship Res 1966;10:10–7.
[4] Chuang SL. Investigation of impact of rigid and elastic bodies with water. Report for Department of the Navy. Washington, DC: United States Department of the
Navy; 1970. Report No.: 3248.
[5] Mori K. Response of the bottom plate of high-speed crafts under impulsive water pressure. J Soc Nav Archit Jpn 1977;142:297–305 [Japanese].
[6] Cheon JS, Jang BS, Yim KH, Lee HSD, Koo BY, Ju HB. A study on slamming pressure on a flat stiffened plate considering fluid–structure interaction. J Mar Sci
Technol 2016;21:309–24.
[7] Truong DD, Jang BS, Ju HB, Han SW. Prediction of slamming pressure considering fluid-structure interaction. Part I: Numerical simulations. Ships Offshore
Struct. https://doi.org/10.1080/17445302.2020.1816732.
[8] Truong DD, Jang BS, Ju HB, Han SW. Prediction of slamming pressure considering fluid-structure interaction. Part II: Derivation of empirical formulations. Mar
Struct. https://doi.org/10.1016/j.marstruc.2019.102700.
[9] Greenhow M, Lin W. Numerical simulation of nonlinear free surface flows generated by wedge entry and wave maker motions. In: Proceedings of the 4th
international conference on numerical ship hydrodynamics, Washington, DC; 1985.
[10] Sun H, Faltinsen OM. Water impact of horizontal circular cylinders and cylindrical shells. Appl Ocean Res 2006;28(5):299–311.
[11] Gingold RA, Monaghan JJ. Smoothed particle hydrodynamics: theory and application to non-spherical stars. Royal Astronomical Society 1977;181:375–89.
[12] Shao S. Incompressible SPH simulation of water entry of a free-falling object. Int J Numer Methods Fluid 2009;59(1):91–115.
[13] Souli M, Ouahsine A, Lewin L. ALE formulation for fluid-structure interaction problems. Comput Methods Appl Mech Eng 2000;190(5):659–75.
[14] Livermore Software Technology Corporation (LSTC). ICFD theory manual incompressible fluid solver in LS-DYNA. Livermore Software Technology Corporation;

[15] Livermore Software Technology Corporation (LSTC). LS-DYNA theoretical manual. Livermore Software Technology Corporation; 2006.
[16] FLOW-3D user’s manual. 2018., version 12.0.
[17] Cd-adapco. STAR-CCM+ User’s manual. 2012., version 7.06.
[18] ANSYS fluent user’s guide. 2015.
[19] ANSYS CFX user’s guide. 2014.
[20] Abaqus user’s manual, version 6.13. SIMULIA; 2013.
[21] Luo HB, Hu J, Guedes Soares C. Numerical simulation of hydroelastic responses of flat stiffened panels under slamming loads. In: Proceedings of the 29th
international conference on ocean, offshore and arctic engineering (OMAE2010); 2010 [Shanghai, China].[22] Yamada Y, Takami T, Oka M. Numerical study on the slamming impact of wedge shaped obstacles considering fluid-structure interaction (FSI). In: Proceedings
of the 22nd international offshore and polar engineering conference (ISOPE2012); 2012 [Rhodes, Greece].
[23] Luo HB, Wang H, Guedes Soares C. Numerical and experimental study of hydrodynamic impact and elastic response of one free-drop wedge with stiffened
panels. Ocean Eng 2012;40:1–14.
[24] Sun H, Wang DY. Experimental and numerical analysis of hydrodynamic impact on stiffened side of three dimensional elastic stiffened plates. Adv Mech Eng
2018;10(4):1–23.
[25] Ma S, Mahfuz H. Finite element simulation of composite ship structures with fluid structure interaction. Ocean Eng 2012;52:52–9.
[26] LSTC. Turek & hron’s FSI benchmark problem. 2012.
[27] Califano A, Brinchmann K. Evaluation of loads during a free-fall lifeboat drop. In: Proceedings of the ASME 32nd international conference on ocean, offshore
and arctic engineering (OMAE2013); 2013 [Nantes, France].
[28] LSTC. 3D fluid elastic body interaction problem. 2014.
[29] Yamada Y, Takamoto K, Nakanishi T, Ma C, Komoriyama Y. Numerical study on the slamming impact of stiffened flat panel using ICFD method – effect of
structural rigidity on the slamming impact. In: Proceedings of the ASME 39th international conference on ocean, offshore and arctic engineering (OMAE2020);
2020 [Florida, USA].
[30] Nicolici S, Bilegan RM. Fluid structure interaction modeling of liquid sloshing phenomena in flexible tanks in flexible tanks. Nucl Eng Des 2013;258:51–6.
[31] DNV. DNV-RP-C205 environmental conditions and environmental loads. Det Norske Veritas; October 2010.
[32] Ahmed YM. Numerical simulation for the free surface flow around a complex ship hull form at different froude numbers. Alex Eng J 2011;50(3):229–35.
[33] Ghadimi P, Feizi Chekab MA, Dashtimanesh A. Numerical simulation of water entry of different arbitrary bow sections. J Nav Architect Mar Eng 2014;11:
117–29.
[34] Park BW, Cho S-R. Simple design formulae for predicting the residual damage of unstiffened and stiffened plates under explosive loadings. Int J Impact Eng
2006;32:1721–36.
[35] Truong DD, Shin HK, Cho S-R. Permanent set evolution of aluminium-alloy plates due to repeated impulsive pressure loadings induced by slamming. J Mar Sci
Technol 2018;23:580–95.
[36] Jones N. Structural impact. first ed. Cambridge, UK: Cambridge University Press; 1989.
[37] Zha Y, Moan T. Ultimate strength of stiffened aluminium panels with predominantly torsional failure modes. Thin-Walled Struct 2001;39:631–48.
[38] Sensharma P, Collette M, Harrington J. Effect of welded properties on aluminum structures. Ship Structure Committee SSC-4 2010.
[39] ABS. Guide for slamming loads and strength assessment for vessels. 2011.
[40] Villavicencio R, Sutherland L, Guedes Soares C. Numerical simulation of transversely impacted, clamped circular aluminium plates. Ships Offshore Struct 2012;7(1):31–45.
[41] Material properties database. https://www.varmintal.com/aengr.htm, Assessed date: 16 May 2020.
[42] Ringsberg JW, Andri´c J, Heggelund SE, Homma N, Huang YT, Jang BS, et al. Report of the ISSC technical committee II.1 on quasi-static response. In:
Kaminski ML, Rigo P, editors. Proceedings of the 20th international ship and offshore structures congress (ISSC 2018), vol. 1. IOS Press BV; 2018. p. 226–31.
[43] Shin HK, Kim S-C, Cho S-R. Experimental investigations on slamming impacts by drop tests. J Soc Nav Archit Korea 2010;47(3):410–20 [Korean].
[44] Huera-Huarte FJ, Jeon D, Gharib M. Experimental investigation of water slamming loads on panels. Ocean Eng 2011;38:1347–55.

여수로 방류에 따른 여수로 바닥 슬래브의 손상 메커니즘 검토

여수로 방류에 따른 여수로 바닥 슬래브의 손상 메커니즘 검토

Examinations of Damage Mechanism on the Chuteway Slabs of Spillway under Various Flow Conditions

  • Yoo, Hyung Ju ;
  • Shin, Dong-Hoon ;
  • Lee, Seung Oh
  • 유형주 (홍익대학교 공과대학 건설환경공학과) ;
  • 신동훈 (K-water연구원 물인프라안전연구소) ;
  • 이승오 (홍익대학교 공과대학 건설환경공학과)
  • Published : 2021.06.03

Abstract

최근 기후변화로 인한 집중호우의 영향으로 홍수 시 댐으로의 유입량이 설계 당시보다 증가하여 댐의 안전성 확보가 필요하다(감사원, 2003). 이에 건설교통부(2003)는 기후변화와 댐 노후화에 대비하여 치수능력증대사업을 추진하여 댐의 홍수배제능력을 확보하였고, 환경부(2020)에서는 40년 이상 경과된 댐을 대상으로 스마트 안전관리체계 구축을 통한 선제적 보수보강, 성능개선 및 자산관리로 댐의 장수명화를 목적으로 댐의 국가안전대진단을 추진하고 있다. 이에 본 연구에서는 댐 시설(여수로)의 노후도 평가 시 활용 될 수 있는 여수로 표면손상 원인규명에 대하여 3차원 수치모형(FLOW-3D 및 COMSOL Multiphysics)을 통해 검토하고자 한다. 연구대상 댐은 𐩒𐩒댐으로 지형 및 여수로를 구축하였으며, 계획방류량(200년 빈도) 및 최대방류량(PMF) 조건에서 모의를 수행하였다. 수치모의 계산의 정확도 검토를 위하여 Baffle의 설치를 통하여 시간에 따른 유량의 변화를 설계 값과 비교하였고 오차가 1.0% 이내를 만족하는 것을 확인하였다. 여수로 표면손상의 다양한 원인 중 기존연구(USBR, 2019)를 통하여 공동침식(Cavitation Erosion) 및 수력잭킹(Hydraulic Jacking)에 초점을 두었으며 방류조건 별 공동지수(Cavitation Index)산정을 통하여 공동침식 위험 구간을 확인하였다. 이음부의 균열 및 공동으로 인한 표층부 콘크리트의 탈락현상을 가속화시키는 수력잭킹 검토를 위하여 국부모형을 구축하였고 음압력(Negative Pressure), 정체압력(Stagnation Pressure), 양압력(Uplift Pressure)의 분포를 확인하였다. 최종적으로 COMSOL Multiphysics를 통하여 압력분포에 따른 구조해석을 수행하여 폰 미세스(Von Mises) 등가응력 및 변위를 검토하여 콘크리트의 탈락가능성을 확인하였다. 본 연구는 여수로 공동부 및 균열부에서의 손상메커니즘을 확인할 수 있는 기초적인 연구이지만 향후에는 다양한 지형조건 및 흐름조건에서의 압력분포 분석 및 유체-구조물 상호작용(Fluid-Structure Interaction, FSI)모의를 수행한다면 구조물 노후도 및 잔존수명 평가에 필요한 손상한계함수 도출이 가능할 것으로 기대된다.

Keywords

Investigation of Mould Leakages in a Gravity Casting

Investigation of Mould Leakages in a Gravity Casting

 

This article was contributed by Gabriele Taricco of CM Taricco and Stefano Mascetti of XC Engineering.

Metal leakages in the original gravity casting mould

몰드 설계는 유체 역학과 금속 응고 패턴뿐만 아니라 주형 자체에서 발생할 수 있는 문제와 응력에 대한 반응을 고려해야 하는 매우 복잡한 작업입니다. 이탈리아에 본사를 둔 주형 제작 업체인 CMTaricco 사는 최근에 새로운 주형 중 하나의 하부에서 금속 누출 문제에 직면했습니다. 주형 누출의 원인은 처음에는 분명하지 않았으며 몇 번의 공정 주기 후에만 나타났습니다. 제작 일정에 차질이 생기고 부품 주조 비용이 급격히 증가하기 때문에 문제가 중요한 것은 분명했습니다.

Investigation of an idea

공정 자체는 주입과 오버플로우 설계인 중력 주조 방식이었기 때문에 유체 역학 부분에서는 문제가 발생할 수 없었습니다. GabrieleTaricco (CMTaricco의 소유주)의 가설은 금속 누출이 주형의 열 손실의 설계 불량에서 기인하여 균일하지 않은 분포를 초래한다는 것이었습니다. 변형률과 그에 따라 주형 바닥에서 크고 원하지 않는 변형이 순환하면서 금속이 유출될 수 있는 중요한 영역의 개방까지 주기적으로 시행되었습니다. 이를 확인하고 문제에 대한 신속한 해결책을 찾기 위해 FLOW-3D시뮬레이션을 실행하여 주형이 가열될 때 발생하는 현상을 정확하게 파악했습니다.

Schematic of a critical area where metal was flowing out of the mould

 

A careful setup, to achieve a fast resolution of the issue

문제의 원인은 신속하게 파악할 수 있어야 했기 때문에 최신 Flow-3D기능을 모두 활용하여 정확한 설정이 필요했습니다. 특히, 채택된 meshing기법은 전통적인 설정과 거의 동일한 정확도를 유지하면서 전산 셀의 수를 크게 줄이는데 매우 도움이 되었습니다. 빠른 시뮬레이션으로 주형 세척에 사용된 첫 번째 방법은 주형 내부의 얇은 캐비티를 직교 축과 정렬하기 위해 주형을 수직 축 주위로 회전시키는 것이었습니다.

Rotating the mould around the vertical axis in order to align the inner thin cavity of the mould

 

두 번째 트릭은 내부 공동 (얇은 벽)에 new conformal mesh기능을 사용하는 한편 전체 도메인에 대해 기존의 더 큰 메쉬 블록을 유지하는 것 이었습니다. The conformal mesh는 open volume과 일치하고, 작은 간극을 갖는 cavity로 제한됩니다.

A global view of the mould with cores and its alignment with the mesh blocks

 

마지막으로, 외부 공간을 주형에 제한하기 위해(현재 구두 상자 모양이 되고, 20도 회전하며, 모델 축과 정렬상태) 일부’ 도메인제거’ 요소가 사용되었습니다.즉, FLOW-3D의 내부 솔리드 모델을 통해 직접 연결됩니다

Domain removing components (yellow) were used to limit the space externally to the mould.

 

나머지 설정은 소프트웨어의 권장 기본값 대부분을 이용하여 기존 체계를 따랐습니다. 이러한 기능과 FLOW-3D의 새로운 하위 도메인 분해 기능 덕분에 설계된 9 000 000 셀을 유체 하위 도메인의 경우에만 1 840 000 셀로, 고체 서브 도메인의 경우 2 430 000 셀로 줄이는 것이 가능했습니다.

 

The analysis

주입 시뮬레이션 후, 양호한 주입 패턴을 보장하기 위해 시뮬레이션의 초점이 열 다이 사이클링 분석으로 리디렉션 되었습니다. 이 경우 설정은 일반 데스크 톱 컴퓨터에서 10개의 생산 사이클을 재현하는 데 1시간이면 간단하고 빠릅니다(i7930 K, 상업적 가치 1500달러). 그 결과 CM의 초기 가설이 확인되었습니다. FlowSight를 사용하여 단일 이미지에서 여러 시점과 횡단면에서 온도 필드를 관찰한 결과 온도가 d라는 것이 분명했습니다. 주형의 침입은 예상되는 변형과 금속 누출을 쉽게 유발할 수 있습니다.

Simulation of the mould’s temperature during the die cyclings

 

Further analysis with the Fluid-Structure Interaction module

 

일단 문제가 확인되고 기술 요원이 향상된 금형 설계를 시작하면 CM Taricco는 다이 상의 응력 및 변형에 대한 FEM 해석을 실행하는 최종 확인을 원했습니다. 이 분석을 수행하기 위해 XC Engineering Srl은 CM이 계산을 설정하고 수행하는 것을 도왔습니다. 분석의 결과는 정확히 CM이 생각하고 있는 것을 보여주었습니다. FLOW-3D는 붓기가 거의 걸리지 않은 금형에서 발견 된 실제 변형과 동일한 위치와 크기를 극도의 정확도로 재현 할 수 있었습니다. 이것은 CM에 대한 좋은 소식이었으며, 실제 주조 조건을 기반으로 실제 금형 변형을 예측하기 위해 설계 단계에서 FSI 모듈을 사용하는 추가 권장 사항을 시행했습니다.

Deformation of the mould during the die cyclings, simulated using the Fluid Structure Interaction model. Deformations are amplified x20.

 

Conclusion

해석결과, CM직원은 CFD솔루션의 온도영역에 대한 모든정보를 사용하여 최적화된 새로운 주형을 설계할 수 있었습니다. 새로운 주형은 열 에너지를 보다 효율적인 방법으로 방출할 수 있었으며 주조물은 수 십번의 공정 주기 후에도 금속 누출의 영향을 받지 않았습니다.

The cast part after mould optimization. No critical leak defects are present.

 

Thermal Stress Evolution

Thermal Stress Evolution

FLOW-3D의 열 응력 진화 (TSE) 모델은 모델링 할 수있는 주조 공정의 범위를 확장합니다. FSI / TSE 모델은 주변 유체의 압력 력, 온도 구배 및 지정된 구속 조건에 대한 응답으로 솔리드 및 응고 부품의 모델 응력 및 변형에 대한 유한 요소 접근법을 사용하여 유체와 솔리드 간의 완전 결합 상호 작용을 설명합니다.

불균일 냉각으로 인해 응고 과정에서 열 응력이 발생합니다. 이러한 응력은 주형 벽의 수축과 주조 모양의 불규칙성에 영향을받습니다.

위의 시뮬레이션은 고형 알루미늄 V6 엔진 블록의 Von Mises 응력을 보여줍니다. 이 블록은 강철 다이 내에서 주조 된 알루미늄 A380 합금으로 구성됩니다. 알루미늄의 주입 온도는 527 ° C 였고 초기 다이 온도는 125 ° C였다. 부품을 다이에서 60 초 동안 냉각시킨 후 다이를 열고 주변 조건 (125 ° C)에서 부품을 9 분 동안 계속 냉각시켜 총 10 분의 시뮬레이션 시간을 가졌다. 보여진 폰 미제스 응력은 부품 내부의 전단 응력의 크기를 측정 한 것으로, 파열이 가장 많이 발생하는 부위를 나타냅니다. 응력은 금형과 응고 금속에서 동시에 계산 될 수 있습니다. 메싱은 FLOW-3D의 구조화 된 메쉬를 초기 템플릿으로 사용하여 자동으로 수행 할 수 있습니다. 사용자는 중첩 또는 링크 된 메쉬 블록을 생성하고 V11.0의 새로운 준수 메쉬 기능을 사용하여 메쉬의 로컬 해상도를 제어 할 수 있습니다. 또는 Exodus-II 형식의 타사 메쉬 생성 소프트웨어에서 Finite Element 메쉬를 가져 오는 옵션이 있습니다.

Simulating Thermal Stress

아래 그림은 강철 다이 내에 알루미늄 A380 합금 주물로 구성된 알루미늄 커버입니다. 주입 온도는 654 ℃이고 초기 다이 온도는 240 ℃이다. 부품은 6 초 동안 다이 내에서 냉각되어 부품이 완전히 고화되었다 (러너 시스템 제외). 그런 다음 다이를 열고 부품을 주변 조건 (25 ° C)에서 10 초 더 냉각시켰다. 러너 시스템을 제거한 후 주위 조건에서 10 초간 더 냉각시켰다. 여기에 표시된 일반 변위는 가장 큰 변형 영역을 강조하기 위해 30 번 확대 된 부품 표면의 동작을 나타냅니다.

Component Coupling within the Fluid-Structure Interaction and Thermal Stress Evolution Models

FLOW-3D v11의 새로운 기능은 인접한 유체 구조 상호 작용 (FSI) 구성 요소 및 / 또는 열 응력 진화 (TSE) 응고 유체 영역 사이의 탄성 응력을 허용하는 기존의 유한 요소 역학 해석법으로의 업그레이드입니다. 결합. 이 새로운 기능은 복잡하고 변형이 심한 다중 재료 부품 (예 : 몰드에서 금속 주 조용 응고 또는 바이메탈 게이지)의 열 응력과 변형을 시뮬레이션하고 연결된 유압에서 힘을 시뮬레이션하는 것을 포함하여 풍부한 모델링 가능성을 열어줍니다. 레이디 얼 게이트 및 파이프 라인 지원 시스템과 같은

모델에는 복잡한 프로세스를 효율적으로 계산할 수있는 몇 가지 옵션이 있습니다.

No coupling

이 옵션은 인접한 FSI 구성 요소가 스트레스를 교환하지 않는 단순화 된 사례를 나타냅니다. 이것은 계산 상 효율적이며 구성 요소 간의 응력 상호 작용이 중요하지 않은 시나리오에 적합합니다.

Full coupling

전체 커플 링 옵션은 함께 융합되었지만 재료 특성이 다른 이웃 FSI 구성 요소를 모델링하기위한 것입니다. 두 구성 요소는 서로 떨어져서 당기거나 서로 밀어 낼 수 없지만 인터페이스의 응력은 구성 요소간에 전송됩니다. 이는 바이메탈 스트립과 같은 접합 구조를 모델링하는 데 이상적입니다.

Partial coupling

부분 커플 링 옵션은 인접한 FSI 구성 요소가 마찰 및 수직력을 통해 상호 작용하지만 분리 될 수있는 일반적인 문제를 모델링하기위한 것입니다. 이 옵션은 FSI 구성 요소와 TSE 응고 유체 영역을 결합하는 데 사용할 수 있으므로 다이에서 냉각되는 부품과 주조 부품에 대한 열 응력의 영향을 조사하는 데 이상적입니다.

모델의 새로운 기능을보다 자세히 보여주기 위해 두 가지 시뮬레이션이 제공됩니다. 첫 번째 상황은 전체 커플 링 옵션을 사용하여 시간에 따라 변화하는 온도에 따라 바이메탈 스트립 벤딩을 모델링하는 반면 두 번째 예는 다이 커플 링에서 V6 엔진 블록의 응고 중 열 응력을 보는 부분 커플 링 모델의 사용을 보여줍니다 .

Full Coupling Example: Bimetallic Strip

전체 커플 링 옵션의 가장 단순한 예 중 하나는 온도 구배에 따른 바이메탈 스트립의 움직임입니다. 이러한 스트립은 두 개의 금속이 온도 변화에 반응하여 동일한 속도로 팽창하지 않기 때문에 열 스위치 및 굴곡에서 일반적으로 사용됩니다. 시뮬레이션에서 모델링 된 바이메탈 스트립은 그림 1에서와 같이 동일 치수의 구리 스트립에 접합 된 길이 15cm, 두께 0.5cm의 강철 스트립으로 구성된 캔틸레버 빔입니다.

Schematic of bimetallic strip

그림 1 : 예제 시뮬레이션에 사용 된 바이메탈 스트립의 개략도. 검은 색 화살표는 처짐이 탐지 된 곳을 나타냅니다. 긍정적 인 처짐은 상향이다.
이어서, 스트립을 온도가 70 초 이상 균일하게 변화하는 환경에 두었다. 그림 2는 시뮬레이션을위한 스트립 팁의 편향과 시간 경과에 따른 다양한 온도에서의 분석 솔루션을 보여줍니다. 결과는 온도가 변했을 때와 스트립의 열 관성으로 인한 스트립의 응답 사이의 약간의 지연을 포함하여 몇 가지 흥미로운 특징을 보여줍니다. 이 지연은 해석 솔루션이 온도의 순간 변화를 가정하기 때문에 계산 된 해석 편차와 해석 편향 사이의 타이밍 차이에 영향을 미칩니다. 변위의 진폭 차이는 분석 결과에서 무한히 얇은 스트립의 가정에 기인 할 수 있습니다. 계산 모델의 두께는 장착 지점에서 추가 응력을 추가하여 처짐이 증가합니다.

Bimetallic deflection plot FLOW-3D

그림 2 : 시뮬레이션 시간 동안 스트립의 끝에서의 처짐. 플롯에는 해석 적 (밝은 파란색) 및 계산 된 (빨간색) 편향과 스트립의 평균 온도 (진한 파란색)가 표시됩니다.

Partial Coupling Example: Metal Casting within a Deformable Die

Temperature profile of a v6 engine block

그림 3 : V6 엔진 블록의 온도 프로파일 단면도. 시뮬레이션 시작 7 초.

두 번째 예제 시뮬레이션은 부분 결합 모델을 사용하여 변형 가능한 스틸 다이 내의 금속 주조물에 응력이 발생하는 것을 보여줍니다. 다이의 두 반쪽과 응고 된 유체는 서로 부분적으로 결합되어있어 정상 응력과 마찰을 통해 상호 작용합니다. 이 시뮬레이션은 금형과 주조 부품의 열 응력 변화가 770K의 고 상선 온도 바로 아래에서 293K의 주변 온도까지 냉각되는 것을 보여줍니다. 주조 부품은 A380 알루미늄 합금으로 이루어져 있으며 금형 반은 H-13 강으로 구성됩니다.

캐스트 부품과 주변 다이의 유한 요소 메쉬는 그림 3과 같이 3,665,533 개의 요소와 3,862,378 개의 노드로 구성됩니다. 또한 다이 반쪽과 TSE 응고 된 유체 영역 각각에 대해 서로 다른 메쉬가 표시됩니다. 앞면에있는 빨간색 원은지지 피스톤 (그림에서는 보이지 않음)으로 인한 것입니다.


그림 4는 충진 후 고압 다이 캐스팅 부품 300s의 주조물 온도와 변위 크기로 채색 된 강철 다이 조각을 결합한 이미지를 보여줍니다. 이 시뮬레이션에서, 다이는 응고 알루미늄에 결합되어 응력이 그들 사이에 전달됩니다. 변위 크기는 다이의 에지에서 0에서부터 주조에 인접한 0.1mm 이상까지 다양합니다.

몰드와 응고 된 유체 표면 사이의 계면에서의 응력은 부분적으로 결합되고, 구속 된 수축이 보일 수있다. 그림 4는 시뮬레이션을 통해 주조 부품과 다이 반제품의 절반에 발생하는 변형을 보여줍니다. 다이 반쪽과 주물은 온도가 감소함에 따라 다른 속도로 줄어들므로 간섭 영역에 큰 응력이 발생하고 잠재적 문제 영역이 있음을 나타냅니다. 금형과 부품의 결합 응력을 계산하면 각 부품 내에서 발생하는 응력을 더 잘 예측하고 부품 품질을 개선하고 공구 수명을 연장하는 방법에 대한 통찰력을 얻을 수 있습니다.

Conclusion

서로 다른 솔리드 오브젝트의 상호 작용은 현대의 설계 및 엔지니어링에서 중요한 부분입니다. FLOW-3D에 대한 FSI 구성 요소와 TSE 응고 유체 영역 간의 새로운 커플 링 옵션을 추가하면 오늘날의 엔지니어가 정기적으로 겪게되는 복잡한 형상을 평가할 수있는 유용한 도구를 제공합니다.

열응력 개선 / Thermal Stress Evolution

열응력 개선 / Thermal Stress Evolution

FLOW-3D의 TSE(Thermalstressdiversion)모델은 모델링 가능한 주조 프로세스의 범위를 확장합니다. FSI/SETSE모델은 주변 유체, 열 구배 및 지정된 구속 조건의 압력에 대응하여 솔리드 및 단단한 구성 요소의 응력 및 변형을 모델링 하는 유한 요소 접근 방식을 사용하여 유체와 솔리드 사이의 완전 결합 상호 작용을 설명합니다.

균일하지 않은 냉각에 의해 발생하는 응고 과정 동안 열 스트레스가 발생합니다. 이러한 응력은 주형 벽의 수축 및 주물 형상의 불규칙에 의해 영향을 받습니다.Thermal stress evolution simulation
Von Mises stresses in a solidified aluminum V6 engine block

위의 시뮬레이션은 VonMises가 단단한 알루미늄 V6엔진 블록에서 응력을 나타냅니다. 이 블록은 강철 다이 내에서 주조된 알루미늄 A380합금으로 구성되어 있습니다.

알루미늄의 주입 온도는 527°C였으며 초기 다이 온도는 125°C였습니다. 부품을 60초 동안 다이 내에서 냉각한 후 주변 조건(125°C)에서 9분 동안 부품을 계속 냉각시켜 총 10분의 시뮬레이션 시간을 제공했습니다. 표시된 VonMises 응력은 부품 내 전단 응력의 크기를 측정한 것이며, 따라서 찢어지기 쉬운 부위를 보여 줍니다.

응력은 금형과 응고 금속에서 동시에 계산할 수 있습니다. FLOW-3D의 구조화된 메쉬를 초기 템플릿으로 사용하여 자동으로 메쉬 작업을 수행할 수 있습니다. 사용자는 중첩 또는 링크된 메쉬 블록을 만들고 V1.1.0의 새로운 적합한 메쉬 기능을 사용하여 메쉬의 로컬 해상도를 제어할 수 있습니다. 또는, Exodus-II형식의 타사 메쉬 생성 소프트웨어에서 유한 요소 메쉬를 가져올 수 있습니다.

Simulating Thermal Stress

아래에 표시된 알루미늄 커버는 강철 다이 내 알루미늄 A380합금으로 구성되어 있습니다. 주입 온도는 654°C였으며 초기 다이 온도는 240°C였습니다. 부품이 다이 내에서 6s동안 냉각되었으며 이때 부품이 완전히 경화되었습니다(러너 시스템 제외). 그런 다음 다이를 열고 부품이 주변 조건(25°C)에서 10초 이상 냉각되도록 했습니다. 그런 다음 탕도(runner)시스템을 제거했고, 이후 주변 조건에서 10초간 더 냉각했습니다. 여기에 표시된 정상 변위는 부품 표면의 움직임을 나타내며, 최대 변형 영역을 강조하기 위해 30회 증폭됩니다.

Displacements in a die cast part, die closed
Displacements in a die cast part, die closed.
Displacements in the part and runners, die open
Displacements in the part and runners, die open.
Displacements in the part with runner system removed
Displacements in the part with runner system removed.

Component Coupling within the Fluid-Structure Interaction and Thermal Stress Evolution Models

FLOW-3Dv11의 새로운 기능은 인접한 FSI(유체-구조물 상호 작용)구성 요소 및/또는 TSE(열 스트레스 진화)고체화된 유체 영역 간의 탄성 응력을 결합할 수 있는 기존의 유한 요소 고체 역학 용제의 업그레이드입니다. 이 새로운 기능은 복합 재료 부품(예:주형에서 응고되는 금속 주물 응고제 또는 바이메탈 게이지)의 열 응력과 변형을 시뮬레이션하고 반경 게이트 및 파이프 라인 지지 시스템과 같은 연결된 유압 구조에 가해지는 힘을 시뮬레이션하는 등 다양한 모델링 가능성을 열어 줍니다.

모델에는 복잡한 프로세스를 효율적으로 계산할 수 있는 여러가지 옵션이 있습니다.

No coupling

이 옵션은 인접 FSI구성 요소가 응력을 교환하지 않는 단순화된 경우를 나타냅니다. 그것은 계산적으로 효율적이며 요소들 간의 스트레스 상호 작용이 중요하지 않은 시나리오에 적합하다.

Full coupling

전체 커플링 옵션은 서로 다른 재료 특성을 가진 인접 FSI구성 요소를 모델링 하기 위한 것입니다. 두 구성 요소는 서로 당기거나 미끄러질 수 없지만 인터페이스의 응력은 구성 요소 간에 전달됩니다. 이는 바이메탈과 같이 접합된 구조물을 모델링 하는 데 이상적입니다.

Partial coupling

부분 커플링 옵션은 인접 FSI구성 요소가 마찰력과 정상적인 힘을 통해 상호 작용하지만 분리될 수 있는 일반적인 문제를 모델링 하기 위한 것. 이 옵션은 FSI구성 요소와 TSE의 고체화된 유체 영역을 결합하는 데 사용될 수 있으므로 부품이 다이에서 냉각될 때 주조 부품 및 다이에 대한 열 응력의 영향을 조사하는 데 이상적입니다.

두가지 시뮬레이션이 제시되어 모델의 새로운 특징을 보다 자세히 보여 줍니다. 첫번째 상황에서는 완전한 커플링 옵션을 사용하여 시간이 변화하는 온도에 대응하여 바이메탈 벤딩을 모델링 하는 반면, 두번째 예에서는 다이에서 V6엔진 블록을 응고하는 동안 부분 커플링 모델을 사용하여 열 응력을 확인하는 것을 보여 줍니다.

Full Coupling Example: Bimetallic Strip

전체 커플링 옵션의 가장 간단한 예 중 하나는 온도 구배에 대한 반응으로 바이메탈이 움직이는 것입니다. 이러한 스트립은 온도 변화에 대응하여 두 금속이 동일한 속도로 팽창하지 않기 때문에 열 스위치 및 벤딩에 일반적으로 사용됩니다. 시뮬레이션에서 모델링 된 바이메탈은 그림 1과 같이 길이 15cm, 두께 0.5cm의 강철 스트립으로 구성된 캔틸레버 빔입니다.

Schematic of bimetallic strip
그림 1:예제 시뮬레이션에 사용된 바이메탈의 개략도; 검은 색 화살표는 편향이 프로브 되는 위치를 나타내고, 양의 편향은 상향이다.

그리고 나서 스트립은 온도가 70초에 걸쳐 균일하게 변화하는 환경에 배치되었다. 그림 2는 시간 경과에 따른 다양한 온도에서 시뮬레이션 및 분석 용액을 위한 스트립 팁의 편향을 보여 준다. 결과는 온도가 변한 시기와 스트립의 열적 관성으로 인한 스트립의 반응 사이의 약간의 지연을 포함하여 몇가지 흥미로운 특징을 보여 준다. 이러한 지연은 분석 솔루션이 온도의 즉각적인 변화를 가정하기 때문에 계산된 편향과 분석적 편향 사이의 타이밍 차이에도 영향을 미친다. 변위의 진폭 차이는 분석 결과에서 무한대의 얇은 스트립의 가정에 기인할 수 있다. 계산 모델의 두께는 장착 지점에 응력을 추가하여 편향을 증가시킵니다.

Bimetallic deflection plot FLOW-3D
그림 2:스트립의 끝에서 시뮬레이션 시간에 걸쳐 처짐. 그림에 표시된 것은 스트립의 평균 온도( 진한 파란 색)뿐만 아니라 분석적( 연한 파란 색)및 계산( 빨간 색)편향입니다.

Partial Coupling Example: Metal Casting within a Deformable Die

Temperature profile of a v6 engine block
Figure 3: V6 엔진 블록의 온도 프로파일 단면도. 시뮬레이션 시작 7 초.

두번째 예제 시뮬레이션에서는 부분 커플링 모델을 사용하여 변형 가능한 강철 다이 내 금속 주물의 응력 개발을 보여 줍니다. 다이의 두 절반과 응고된 유체는 부분적으로 서로 결합되어 정상적인 응력과 마찰을 통해 상호 작용합니다. 시뮬레이션은 다이와 주물 부품의 열 응력 변화를 770,000 K의 solidus온도 바로 아래에서 298K의 주변 온도로 냉각하는 모습을 보여 줍니다. 주물 부분은 A380알루미늄 합금으로 구성되어 있고 다이 반쪽은 H-13강철로 구성되어 있습니다.

주조 부품과 주변 다이의 유한 요소 메시는 그림 3과 같이 3,665,533 요소와 3,862,378개 노드로 구성됩니다. 또한 각 다이의 절반에 대해 분리된 메쉬와 TSE고형화된 유체 영역도 나와 있습니다. 전면의 빨간 색 원은 서포트 피스톤 때문입니다(그림과 같이 표시되지 않음).

Thermal stress model
Figure 4 는 채워진 후 고압 다이 캐스팅 부품 300s의 주조물 온도와 변위 크기로 채색 된 강철 다이 조각을 결합한 이미지를 보여줍니다. 이 시뮬레이션에서, 다이는 응고하는 알루미늄에 연결되어 응력이 그들 사이에 전달됩니다. 변위 크기는 다이의 에지에서 0에서부터 주조에 인접한 0.1mm 이상까지 다양합니다.

금형과 응고된 유체 표면 사이의 경계면에서 발생하는 응력이 부분적으로 결합되어 제한된 수축을 확인할 수 있습니다. 그림 4는 시뮬레이션을 통해 주형 부분의 변형과 다이 부분의 절반의 변형을 보여 줍니다. 온도가 감소함에 따라 다이 캐스트와 주물이 서로 다른 속도로 수축하여 간섭 영역에 큰 응력이 발생하고 잠재적인 문제 영역이 나타납니다. 다이와 부품에서 결합된 응력을 계산하면 사용자가 각 구성 요소 내에서 발생하는 응력을 더 잘 예측하고 부품 품질을 개선하고 도구 수명을 연장하는 방법에 대한 통찰력을 제공할 수 있습니다.

Conclusion

다른 단단한 물체들의 상호 작용은 현대 디자인과 공학의 중요한 부분입니다. FSI구성 요소와 TSE고정 유체 영역 간의 새로운 결합 옵션이 FLOW-3D에 추가되어 오늘날의 엔지니어들이 정기적으로 접하는 복잡한 기하학적 구조를 평가하는 데 유용한 도구가 되었습니다.

THE ELASTIC MEMBRANE AND WALL MODEL IN FLOW-3D [FLOW-3D의 탄성 멤브레인과 벽 모델]

1. Introduction
An elastic membrane and wall model has been developed to provide a limited Fluid-Structure Interaction (FSI) capability in FLOW-3D. In the model, deformation of an elastic membrane or an elastic wall impacts the adjacent fluid flow, while fluid pressure, in turn, affects the deformation. These interactions are described in the code in a fully coupled fashion.
The main assumption of the model is that the deformations are small, i.e., the deflections are much smaller than the size of the deforming object (for elastic membranes) or the characteristic lengths of fluid flow and wall thickness (for elastic walls), allowing for a few useful simplifications. The geometries of membranes and elastic walls are assumed to be time-invariant, while the effects of their deformation on fluid flow are described with volume sources and sinks distributed along the fixed fluid-structure interface. With the further assumption that the pressure force is uniformly distributed on the membrane surface, analytical solutions rather than structural analysis algorithms are used to determine the membrane deformation.
There are many potential applications for the model in microfluidic systems, e.g., chemical analysis systems, medical microdosage systems and inkjet devices. The model can be used to simulate flow in piezoelectric valveless pumps which convert membrane vibrations into a pumping action. The model can also be used to simulate droplet formation for piezoelectric inkjet printheads where a membrane or an elastic tube deforms under the force of a piezoelectric actuator to produce a droplet of ink.

Why FLOW-3D?

Why FLOW-3D ?

FLOW-3D는 엔지니어들에게 다양한 분야의 유동해석에 대해 귀중한 통찰력을 제공하는 강력한 모델링 도구입니다. 정확하게 자유 표면 흐름을 예측하는 특별한 기능을 통해 FLOW-3D는 설계 단계에서뿐만 아니라, 생산 공정 개선에도 사용할 수 있는 이상적인 전산 유체 역학 (CFD) 소프트웨어입니다.


TruVOF and Free Surface Modeling

TruVOF - Volume of Fluid - VOF Method

FLOW-3D 는 다른 유동해석 프로그램과는 유체 계면을 다루는 기법이 확연하게 다릅니다. FLOW-3D 는 자유표면의 위치를 추적하고, 그 자유표면에 적절한 동적 경계 조건을 적용하는 특수한 수치기법(numerical method)을 사용합니다. FLOW-3D 에서 모델링된 자유표면은 로스알라모스 국립 연구소에서 Flow Science의 설립자인 C. W. Hirt 박사와 함께 소속한 과학자 그룹에 의해서 개발된 VOF(Volume of Fluid) 기법으로 모델링됩니다. CFD 프로그램의 대부분은 세 가지 기본 VOF 성분 중 실제로는 단지 하나 또는 두 가지만 구현하여 VOF 기능을 통합 구현한 것으로 주장하고 있습니다. CFD 사용자는 이러한 유사 VOF 기법으론 종종 잘못된 결과를 얻을 수 있음을 알고 있어야 합니다. FLOW-3D는 자유표면의 성공적인 해석에 필요한 모든 요소를 포함하고 있습니다. 또한, FLOW-3D는 경계 조건과 계면 추적의 정확도를 증가시키기 위해 원래의 VOF 방법을 크게 향상 시키고 있습니다. 좀더 자세한 사항은 TruVOF 방법을 참고하시기 바랍니다.

Free Gridding Separates Meshing from Geometry Construction

Free gridding capabliity in FLOW-3DGridding에 대한 FLOW-3D’의 기본 접근 방식은 deformed, body-fitted grids의 유연성과 단순한 직사각형 그리드의 장점을 결합합니다. 직사각형 컨트롤 요소의 고정 그리드는 간단하고 매우 바람직한 특성을 갖도록(예를 들어, 향상된 정확도, 작은 메모리 크기 요구, 간단한 수치 근사치) 생성 처리됩니다. 이 접근법은 그리드나 형상을 각각 서로 독립적으로 자유롭게 변경 할 수 있기 때문에 “free-gridding”으로 지칭합니다.

이 기능은 body-fitted 또는 유한 요소 격자를 생성하는 지루한 작업을 하지 않도록 해 줍니다.
Rectangular gridding 의 유연성과 효율은 multi-block 과 conforming meshing 같은 고급 기능에 의해 강화됩니다. 연결되어 있고 부분적으로 중첩된 메쉬 블록은 복잡한 멀티 스케일 유동 도메인과 관심 영역에서 효과적으로 높은 해상도를 가질 수 있는 수단을 제공합니다. Conforming mesh는 직사각형 gridding 구조와 관계 없이 특별한 기하학적 형상, cavities, 얇은 구조물, 경계층 등에 적합한 고품질의 grids를 생성할 수 있습니다.
Read more about FLOW-3D‘s free-gridding approach in CFD-101 >

Modeling Fluid Flow in Complex Geometry is Easy with FAVOR™

FAVOR technique in FLOW-3DFLOW-3D 는 직사각형 격자 내에서 일반 기하학적 영역을 정의하는데 사용되는 FAVOR™ (Fractional Area Volume Obstacle Representation) 방법으로 알려진 특별한 기술을 사용합니다. FAVOR기본 철학은 수치 알고리즘이 각 제어 체적(control volume)에 대해 하나의 압력, 하나의 속도, 하나의 온도 등으로 이루어지는 정보를 기반으로 하므로 형상을 정의하는 데 많은 정보를 사용하는 것이 적합하지 않는다는 것입니다. 따라서, FAVOR ™ 기술은 직사각형 격자의 단순성을 유지하면서 각 제어 체적 내의 유동 값의 정밀도와 일치하는 수준으로 복잡한 기하학적 형상을 나타냅니다.

Read more about FLOW-3D‘s FAVOR™ technique in CFD-101 >

Meshing Capabilities

FLOW-3D 는 복잡한 유체 해석 모델링시 간단하면서도 효율적이고 견고한 대규모 격자 생성 능력을 제공합니다. 특히, FLOW-3D의 효율좋은 단순한 메쉬 구조와 다중 블록 메쉬의 다양한 특징에 의해 최고의 메싱 효율과 견고성을 자랑합니다. 여러개의 메쉬 블록 사용은 관심 영역의 최적화를 허용하고 주어진 시뮬레이션에 필요한 연산 리소스를 훨씬 감소시킵니다. 작은 장애물(Small obstacles), 복잡한 형상 그리고 전체 영역 크기에 비해 좁은 통로(channel)는 다음과 같은 linked, nested, conforming, 그리고/또는 부분 중첩 mesh blocks등의 여러 다중 블록 메싱능력 중 하나를 사용하여 정확히 해결 될 수있습니다. Linked mesh blocks은 관심영역의 격자 생성과 계산 격자의 총 수를 제한하는 데에만 사용될 수 있습니다. Nested mesh blocks은 관심영역 주위의 해상도를 향상시키기 위해 사용될 수 있습니다. Conforming 과 partially overlapping mesh blocks은 단순히 중첩된 급격한 변화와 불규칙한 형상을 해결하기 위해사용될 수 있습니다. FLOW-3D 는 또한 Fluid-Structure Interaction에 필요한 body-fitted 유한요소 격자를 허용합니다. 이 body-fitted 격자는 FLOW-3D 또는 외부 CAD 패키지 S/W로 부터 가져와 자동으로 생성할 수 있습니다. 이 모든 메쉬 기술은 사용자에게 간단하면서 효율적으로 메쉬를 생성하게 하여 솔버 성능 향상 및 해석시간을 줄이는 유연성을 제공합니다.

바이오 분야

Biotechnology

생명 공학 분야에 전산 유체 역학을 적용하는 것은 비교적 새로운 방법으로, 다양한 의료 기기를 효과적으로 사용하거나, 분석 구현하는 방법을 개선하는데 큰 도움이 될 수 있습니다.
FLOW-3D는 하나의 패키지로 구성되어 있으며, 광범위한 범위를 갖는 강력한 시뮬레이션 해석 프로그램 입니다.
FLOW-3D가 가지고 있는 기능으로 자유 표면과 제한된 갇혀 있는 유체의 흐름, 가변 밀도, 상 변화, 움직이는 물체, 기계 및 열 응력 해석이 가능합니다.
 
자세한 내용은 FLOW-3D의 모델링 기능의 전체 목록을 살펴보십시오.
Von Mises stress 분포.
FLOW-3D‘s fluid-structure interaction model 을 이용한 안압 분석 결과.
Courtesy University at Buffalo.

바이오 분야의 다양한 해석 사례


관련 기술자료

Schematic view of the experimental set-up

Short-time numerical simulation of ultrasonically assisted electrochemical removal of strontium from water

September 2023 DOI:10.30955/gnc2023.00436 Conference: 18th International Conference on Environmental Science and Technology CEST2023, 30 August to 2 September 2023, Athens, ...
Fig. 8. Pressure distribution during the infiltration of preform with the 50 ¯m particles and 20 % starches: (a) 25 % filled, (b) 57 % filled, and (c) 99 % filled.

Experimental study and numerical simulation of infiltration of AlSi12 alloys into Si porous preforms with micro-computed tomography inspection characteristics

마이크로 컴퓨터 단층 촬영 검사 특성을 가진 Si 다공성 프리폼에 AlSi12 합금의 침투에 대한 실험적 연구 및 수치 시뮬레이션 Ruizhe ...
Liquid-solid co-printing of multi-material 3D fluidic devices via material jetting

재료 분사를 통한 다중 재료 3D 유체 장치의 액체-고체 공동 인쇄

Liquid-solid co-printing of multi-material 3D fluidic devices via material jetting BrandonHayes,Travis Hainsworth, Robert MacCurdyUniversity of Colorado Boulder, Department of Mechanical ...
Fig. 2: Scheme of the LED photo-crosslinking and 3D-printing section of the microfluidic/3D-printing device. The droplet train is transferred from the chip microchannel into a microtubing in a straight section with nearly identical inner channel and inner microtubing diameter. Further downstream, the microtubing passes an LED-section for fast photo cross-linking to generate the microgels. This section is contained in an aluminum encasing to avoid premature crosslinking of polymer precursor in upstream channel sections by stray light. Subsequently, the microtubing is integrated into a 3D-printhead, where the microgels are jammed into a filament that is directly 3D-printed into the scaffold.

On-Chip Fabrication and In-Flow 3D-Printing of Cell-Laden Microgel Constructs: From Chip to Scaffold Materials in One Integral Process

세포가 함유된 마이크로겔의 온칩 제작 및 인-플로우 3D 프린팅구성:하나의 통합 프로세스에서 칩에서 스캐폴드 재료까지 Vollmer, Gültekin Tamgüney, Aldo BoccaciniSubmitted date: ...
Figure 3 Simulation PTC pipes enhanced with copper foam and nanoparticles in FLOW-3D software.

다공성 미디어 및 나노유체에 의해 강화된 수집기로 태양광 CCHP 시스템의 최적화

Optimization of Solar CCHP Systems with Collector Enhanced by Porous Media and Nanofluid Navid Tonekaboni,1Mahdi Feizbahr,2 Nima Tonekaboni,1Guang-Jun Jiang,3,4 and ...
Energy and exergy analysis of an enhanced solar CCHP system with a collector embedded by porous media and nano fluid

Energy and exergy analysis of an enhanced solar CCHP system with a collector embedded by porous media and nano fluid

Year 2021, Volume 7, Issue 6, 1489 - 1505, 02.09.2021 N. TONEKABONI H. SALARIAN M. Eshagh NIMVARI J. KHALEGHINIA https://doi.org/10.18186/thermal.990897 ...
Fig.1 Schematic diagram of the novel cytometric device

Fabrication and Experimental Investigation of a Novel 3D Hydrodynamic Focusing Micro Cytometric Device

Yongquan Wang*a , Jingyuan Wangb, Hualing Chenc School of Mechanical Engineering, Xi’an Jiaotong University, Xi’an, Shaanxi, 710049, P. R. Chinaa ...
Figure 1. (a) Top view of the microfluidic-magnetophoretic device, (b) Schematic representation of the channel cross-sections studied in this work, and (c) the magnet position relative to the channel location (Sepy and Sepz are the magnet separation distances in y and z, respectively).

Continuous-Flow Separation of Magnetic Particles from Biofluids: How Does the Microdevice Geometry Determine the Separation Performance?

by  Cristina González Fernández1, Jenifer Gómez Pastora2, Arantza Basauri1, Marcos Fallanza1, Eugenio Bringas1, Jeffrey J. Chalmers2 and Inmaculada Ortiz1,* 1Department of Chemical and Biomolecular Engineering, ETSIIT, ...
Fluid velocity magnitude including velocity vectors and blood volumetric fraction contours for scenario 3: (a,b) Magnet distance d = 0; (c,d) Magnet distance d = 1 mm.

Numerical Analysis of Bead Magnetophoresis from Flowing Blood in a Continuous-Flow Microchannel: Implications to the Bead-Fluid Interactions

Jenifer Gómez-Pastora,  Ioannis H. Karampelas,  Eugenio Bringas,  Edward P. Furlani &  Inmaculada Ortiz  Scientific Reports volume 9, Article number: 7265 (2019) Cite this article Abstract 이 연구에서는 ...
Fig. 12. Comparison of simulation results with experimental data for a flow rate of water = Ql=15 ml/hr and a flow rate of air = Qg =3 ml/hr.

Simulation of Droplet Dynamics and Mixing in Microfluidic Devices using a VOF-Based Method

A. Chandorkar Published 2009 Abstract This paper demonstrates that the Volume of Fluid (TruVOF) method in FLOW-3D (a general purpose CFD ...

FlowSight

FlowSight

FlowSight는 FLOW-3DFLOW-3D CAST결과의 정교한 시각화를 제공하도록 설계된 고급 후 처리 도구입니다. FlowSight는 직관적인 후처리 인터페이스 내에서 우수한 결과 분석 기능을 갖춘 모델을 제공합니다. 스플 라인 경로를 따라 임의의 2D클립, 3D클립 및 투명도, 볼륨 렌더링, 고급 데이터 타임 시리즈 플로팅, 간소화 및 벡터 플롯은 사용 가능한 놀라운 도구의 일부에 불과합니다. FlowSight를 사용하면 여러 뷰 포트와 동적 객체 시각화 도구로 구성된 풍부한 기능 세트와 결합되어 있으므로 엔지니어는 분석 및 프레젠테이션 요구 사항에 맞게 CFD결과를 최대한 활용할 수 있습니다.

FlowSight는 모든 FLOW-3DFLOW-3D CAST라이센스에 포함되어 추가비용 없이 사용할 수 있습니다.

새로운–스플 라인 클립!

FlowSight의 스플라인 클립 기능을 사용하면 복잡한 곡면을 따라 클립을 생성할 수 있습니다. ogee weir 위로 물이 흐르는 시뮬레이션에서, 스플 라인은 ogee weir의 표면을 따라 형성됩니다. 그런 다음 스플 라인이 돌출되어 웨어 표면을 따라 물의 자유 표면 높이에 의해 색상이 지정된 클립을 생성합니다.

키 프레임 기능

크고 복잡한 시뮬레이션을 분석 할 때 매우 일반적인 문제는 관심 영역이 형상에 의해 가려지거나 시뮬레이션이 시간이 지남에 따라 변경됨에 따라 관심 영역이 변경 될 수 있다는 것입니다. 키 프레임은 분석 중에 형상을 “분리되도록”허용하고 시점이 시간과 공간을 통해 이동할 수 있도록 하여 이 문제를 해결합니다.

이 애니메이션은 FlowSight의 키 프레임 기능을 사용하여 충전하는 동안 다이 반쪽을 “시각적으로”열고 다이를 채우는 금속을 표시하면서 다이 표면에 고체 온도를 표시하는 방법을 보여줍니다.

Particle Visualization

FlowSight는 파티클(입자) 시각화 기능을 완벽하게 갖추고 있습니다. 입자는 입자 직경, 입자 밀도, 입자 수명, 속도 및 관련성이 있는 기타 변수에 의해 색상이 지정될 수 있습니다. 이 경우, 입자는 각각의 직경의 크기에 의해 착색됩니다.

속도 벡터 필드

FlowSight는 사용자에게 평면 또는 도메인 전체에 걸친 전체 볼륨 속도 및 방향 분석에 속도 벡터 필드를 시각화하는 옵션을 제공합니다. 사용자 지정 가능한 벡터 필드를 사용하면 다양한 색상 지정 및 밀도 조정이 가능하여 선명도를 높일 수 있습니다.

Streamlines & Pathlines

FlowSight의 유선(Streamlines) 기능은 복잡한 동적 패턴을 완전한 충실도로 시각화하여 유동장 속도 방향에 대해 실시간 스냅 샷을 제공합니다. 경로 선(Pathlines)은 시간을 따른 유체 입자의 궤적을 시뮬레이션하는 동안, 히스토리 라인은 유동장에서 유체 입자를 애니메이션 합니다.

Iso-surfaces

Iso-surfaces 은 유체 및 고체 표면을 시각화하는 강력하고 빠른 방법으로, 일정한 난류 에너지 영역을 표시하는 데 적합합니다.

Volume Render

iso-surface에서만 변수를 표시하는 대신 사용자 지정 가능한 볼륨 맵을 사용하여 볼륨 전체에 걸쳐 변수를 표시합니다. 그림에 표시된 바와 같이 각 기포와 주변 액체의 변형률 크기는 볼륨 렌더링과 함께 표시됩니다.

 

Multiple Data Views

숫자 및 다양한 그래프 등의 시각적 형식으로 분석하기

Visualizing Non-inertial Reference Frame Motion

Non-inertial reference frame visualization는 편리한 시뮬레이션 설정을 제공하고 계산 시간을 단축하며 사용자가 사실적인 방식으로 모델을 시각화 할 수 있게합니다.

2D Clips

2D 클립은 모든 단면 평면에서 유체 매개 변수를 시각화하는 데 사용됩니다.

3D Clipping

3D 클리핑 도구를 사용하면 사용자가 6 개 방향 모두에서 등면을 동시에 슬라이스 할 수 있으며, 높은 결함 영역을 감지하고 유체 및 고체 영역 내부의 온도, 압력, 속도 프로파일을 시각화하는 데 유용합니다.

  • 특정 방향의 범위 사이에 애니메이션 제공
  • 한 번에 한 방향으로 스왑
  • 양방향 애니메이션 : 앞으로 및 뒤로

Arbitrary Clips

평면, 원통형, 상자, 원뿔형, 구형 및 간소화된 표면에 대한 시각화를 포함하여 광범위한 유연성으로 표면 뷰를 분석할 수 있습니다. 유체 흐름이 평면이 아닌 표면에 대한 시각화가 필요한 경우 유용합니다. 임의 클립을 사용하면 연속적으로 여러 클립을 만들 수도 있습니다.

Probe Data

포인트 프로브는 시간에 따른 변수의 진화를 보여주고, 라인 프로브는 거리에 따른 변수 값의 변화를 반환합니다. 오른쪽, 프로브는 유체의 응고 비율을 보여줍니다.

Vortex Cores

와류 코어 식별에 사용할 수있는 두 가지 옵션인 와류 및 고유 분석을 통해 코어 강도에 따라 필터링 가능한 결과 생성이 가능합니다.

엔지니어들은 연구를 위해 다양한 시각화 방법을 사용합니다. 유체 흐름에서 와류 코어의 분석은 중요한 문제로, 와류 코어는 속도 필드 내에 와류 구조 (중앙 트레이스)를 나타내는 선 입니다. 기술적으로, FlowSight는 와류 방법 및 고유치 분석에서 속도 벡터와 소용돌이 벡터의 속도장에서의 식별위치는 평행합니다. FlowSight는 사용자에게 와류 코어 식별을 위한 두 가지 옵션을 제공합니다. 코어는 특정 강도 이상 또는 이하로 FlowSight에서 필터링 될 수 있습니다. 코어는 일반적으로 코어 주위에 회전 또는 단순히 순환 강도의 비율에 의해 채색됩니다. 아래의 예에서는, 와류 코어 고유치 값 분석을 이용하여 생성됩니다. 강한 코어는 소용돌이의 중심에 형성되어있는 것을 알 수 있습니다. 이를 통해 사용자는 펌프로 공기 흡입의 가능성을 연구 할 수 있습니다. 코어가 너무 강한 경우, 공기는 강한 와류로 인해 야기되는 열린 통로로부터 흡입될 수 있습니다.

History Data

그래프 도구는 일반적인 히스토리, 진단 및 메시 종속 데이터에 강력한 수준의 분석을 제공하여 서로 다른 시뮬레이션 데이터를 상대적으로 보여줍니다.