Parametric study of the Crossing elongation effect on the mixing performances using short Two-Layer Crossing Channels Micromixer (TLCCM) geometry

https://doi.org/10.1016/j.cherd.2020.03.010Get rights and content

Highlights

  • Numerical investigations of a short TLCCM micromixer with an elongation of the crossing zone were performed.

  • The chosen micromixer showed an excellent mixing index which exceeds 85.67% for Re = 0.2 and reaches 99.22% from Re = 50.

  • Our micromixer has a lower pressure drop compared to other geometries studied recently.

Abstract

In this study we investigate the mixing performances of a modified micromixer which achieves a very good mixing quality that can be compared to other micromixers proposed recently, our idea proposes a modification of the crossing zone to reduce the unit number. The numerical simulations have been carried out at low Reynolds numbers using the CFD Fluent code to solve the 3D momentum equations, continuity equation, and the species transport equations. The elongation of the crossing zone is defined by a parameter called aspect ratio (l/W). A parametric study was realized using five values of aspect ratio (l/W) from 0 to 1 in a wide range of Reynolds numbers: from 0.2 to 80. To analyze the obtained results through the numerical simulations, the mass fraction contours, velocity vectors, velocity profiles, and pressure losses were presented in different cross-sectional planes and positions. The selected geometry (with l/W = 1) has excellent mixing performances where the obtained mixing index exceeds 85.67% for Re = 0.2 and reaches 99.22% for Re = 50, it also has a lower pressure drop compared to other geometries studied recently. Therefore, the selected micromixer shows high mixing performances at low Reynolds numbers, so it can be employed to improve fluid mixing in various microfluidic systems.

Introduction

Micromixers are widely used in various industrial applications, they represent the important components in microfluidic systems and have many uses in the bioengineering fields, biomedical applications and chemical engineering (Lee and Fu, 2018). Mixing in laminar flow and at low Reynolds numbers is commonly used in different fields (Lee et al., 2016), for example chemical synthesis, chemical reactions, extraction and purification, emulsion processes, polymerization, DNA analysis, detection and analysis of chemical or biochemical content.

Micromixers are generally classified into two categories: active and passive micromixers (Lee and Fu, 2018; Lee et al., 2016; Ghanem et al., 2014). Active micromixers require external energy for the mixing process to enhance the mixing efficiency (dielectrophoretic, electrowetting shaking, magneto-hydrodynamic, and ultrasound disturbance) (Lee et al., 2011). These types of mixers are more efficient. However, they are more difficult to integrate and expensive than passive types and require more detailed fabrication while mixing in passive micromixers is a consequence of the interaction between flow and channel geometry. Passive micromixers have particular importance according to their simple structures and easy manufacturing (Hossain et al., 2017).

The use of chaotic advection which generates secondary flows is an appropriate way to increase mixing efficiency, in the case of Newtonian fluids with stationary flows in laminar regime (Aref, 1984), the planar micromixer geometries do not lead to obtaining a homogeneous mixing and consequently, the mass transfer will be ineffective (Hossain et al., 2009; Ansari and Kim, 2010; Solehati et al., 2014), while the use of the chaotic micromixer geometries promote the mass transfer which allows improving the mixing performance in a significant way (Ansari and Kim, 2009; Lin, 2015; Hossain and Kim, 2015).

Many studies have been carried out on the mixing of Newtonian fluids in various types of micromixers. In passive micromixers where the walls are immobile, the mixing of the flowing fluids can be ensured by the key phenomenon which is the chaotic advection. The channels geometry makes it possible to disturb the laminar velocity field which generates the compression, stretching and folding of the fluid layers, this deformation leads to a more homogeneous and rapid mixing (Aref, 1984).

Several models of planar micromixers have been designed and studied previously and recently through numerical and experimental works, Soleymani et al. (2008) studied the development of vortices in a T-type micromixers where the Reynolds number values are taken between 12 and 240, they examined the effect of the angle junction, the aspect ratio, and the throttling on the mixing efficiency, their results reveal that the mixing quality is improved by the appearance of the vortices which are generated by the increase of the flow rate and also by the geometrical parameters. Different micromixer geometry: curved channel; square-wave channel; and zig-zag channel, were studied numerically using a CFD code by Hossain et al. (2009), they evaluated the mixing efficiency in the Reynolds number range (from 0.267 to 267) and found that the square-wave micromixer gives the best mixing quality while the curved channel shows a low-pressure drop. The mixing efficiency has been analyzed by Ansari and Kim (2010) for planar split and recombines micromixers with rhombic and circular sub-channels where the Reynolds numbers are ranging from 1 to 80. They found that the unbalanced geometries have higher mixing quality compared to the balanced geometries, in particular, the geometries of circular sub-channels where the pressure-drop is also low. A planar micromixer with curved C-shaped channel and radial baffles which generate multidirectional vortices has been examined by Tsai and Wu (2011), the range of Reynolds number is from 0.054 to 81. Their results indicate that the mixing performance of the micromixer with the first baffle related to the internal cylinder and the second related to the external cylinder is better than that of the micromixer with the reverse arrangement of baffles, while the pressure drops are quite strong. A numerical and experimental study on eight planar micromixers have been carried out by Cheri et al. (2013) at a range of Reynolds number between 0.1 and 40. The results show that the best mixing index is obtained with the micromixer characterized by a round corner rectangular chamber and a straight shape obstacle where the ratio between the mixing index and the pressure drop is maximal. Solehati et al. (2014) have realized a numerical study to analyze the mixing performance of microchannel with wavy structure compared to the basic straight microchannel using various Reynolds numbers ranging from 1 to 200. They obtained that microchannel with wavy structure provides higher mixing performance than that of basic microchannel due to the occurrence of chaotic flow created by wavy curves structure. A planar modified zigzag micromixer has been numerically studied by Chen and Li (2017) at a wide range of Reynolds numbers (0.1–100), they found that the inverse flow generated by the zigzag modification allows to enhance significantly the mixing index compared to the zigzag micromixer, however the pressure drop in the modified zigzag micromixer is higher than the zigzag micromixer. A planar micromixer with spiral microchannel has been proposed by Vatankhah and Shamloo (2018), who conducted a numerical parametric study which investigates the distance center to center of successive channels. They found that decreasing this parameter allows to enhance the mixing index. Borgohain et al. (2018) have studied numerically using CFD code the effect of the placement of obstacles designed as thin curved ribs on the mixing performance of a cross T-shaped micromixer. Their results show that the inclusion of curved obstacles along the channel provides better geometry that improves the mixing efficiency in the microchannel.

The need of generating chaotic advection with average flow regimes has been attracted the interest of several researchers with numerical and experimental works to design adequate geometries that can generate three-directional flows. Among the first works, we indicate the work of Liu et al. (2000), they have conducted a numerical and experimental study by evaluating the mixing efficiency in the three-dimensional micromixer, the C-shaped micromixer and the straight channel, for a Reynolds range between 6 and 70, they found that the three-dimensional micromixer has a mixing index clearly superior to that of the two other micromixers. Beebe et al. (2001) carried out a qualitative analysis of the mixing performance of a three-dimensional micromixer by both experimental and numerical study at a Reynolds number range of 1–20. The designed geometry is of repetitive “L” shaped units. Their results show that the mixing efficiency of the three-dimensional micromixer is clearly superior to that of the square-wave micromixer. A parametric study of the mixing in a chaotic micromixer consisting of repeating “L” shaped patterns was performed numerically by Ansari and Kim (2009). Reynolds numbers rise from 1 to 70. Both the mixing index and the pressure loss values have been limited in optimal intervals of Reynolds numbers where the mixing performance is significantly better with a low pressure loss. Nimafar et al. (2012) carried out a comparative experimental study of three micromixers H, O and T-micromixer at low Reynolds number (0.08–4.16), they found that the H-micromixer has the best mixing performance compared to the two others. Another comparative numerical investigation has been performed by Alam and Kim (2013) to evaluate the mixing performance of four micromixers characterized by circular mixing chambers interconnected by various constriction channels at Reynolds numbers change from 0.1 to 100. They obtained that their proposed micromixer with crossing constriction channels reveals excellent mixing performance at low Reynolds numbers less than 50 with lower pressure drop than the three others. An experimental and numerical investigation has been carried out by The et al. (2015) to evaluate the mixing efficiency of 3D micromixer at various Reynolds number from 0.5 to 100, the highest mixing index value obtained in this range was 95% with slightly high pressure drop. Five 3D serpentine micromixers were numerically examined by Lin (2015) for a large interval of Reynolds numbers (8–160), the chaotic advection mechanism as rotation, continuous rotation and 3D stretching was identified to know their effect on the mixing performance as well the effect of the addition of grooves and their positions, however, the pressure drop of all micromixers was similar and significantly high. The effects of splitting, recombination and chaotic advection mechanisms were performed by Chen and Shen (2017) using two types of micromixers with repeating E-shaped units, their results show that the mixing index has achieved 96% but the pressure drop has quickly increased. Numerical simulations were realized by Ruijin et al. (2017) to compare the mixing efficiency of three micromixers: Baker and F-shaped micromixers, they found that at low Reynolds numbers the Baker micromixer has highest mixing index.

In order to enhance the chaotic advection that improves the mixing quality, another type of micromixers that is characterized by two fluid layers has been examined in very recent works. We note that the first work in this field has been realized by Xia et al. (2005), who have studied numerically and experimentally an excellent micromixer composed of two-layer crossing channels, the chaotic advection created in this micromixer allows to have a very high mixing index which reaches 96% at low Reynolds numbers. The dominance of this micromixer has been proven compared with other types. Numerical analyses were performed by Hossain and Kim (2015) to evaluate the mixing performances of a micromixer composed with repeating OH-shaped units in a range of Reynolds number from 0.1 to 120. The studied micromixer reached 88.4% at Re = 30. After that, they carried out a parametric numerical study (Hossain and Kim, 2016) using the geometry proposed by Xia et al. (2005) in a Reynolds number interval change from 0.2 to 40. The results revealed that the mixing index and the pressure drop were significantly influenced by the geometric parameters. Experimental and numerical investigations have been carried out by Hossain et al. (2017) where the Reynolds number range was (0.2–120), high mixing index values were obtained: 96%–99% with lower pressure drop than TLCCM of Xia et al. (2005). Recently, Raza et al. (2018) have performed a numerical study to investigate the mixing performance of a short micromixer with repeating OX-shaped unites in the Reynolds number range: 0.1–200, they also studied the effect of the geometric parameters on the mixing performances. Their results revealed that at least 87% of mixing has obtained, while, the pressure drop increased rapidly.

Motivated by the work of Hossain et al. (2017), our goal is to design a geometry that offers excellent mixing performances with low pressure losses using a short length (the half of that of Hossain et al. (2017)). The mixing performances were examined using three dimensional Navier-Stokes equations over Reynolds number interval of 0.2–80. The effect of the geometrical parameter which is the aspect ratio on the mixing efficiency has been investigated, than the obtained results of the pressure losses were compared with that of recent works.

Section snippets

Micromixer designs

The basic geometric model is that of Hossain et al. (2017). Our idea is to exploit the elongation of the crossing zone which is defined by an aspect ratio (l/W), where l/W = 0 represents the reference case (Hossain et al., 2017). The current proposed geometry is composed of two superimposed layers, with two shifted inlets, the inlet 1 is linked to the top layer while the inlet 2 is linked to the bottom layer.

Fig. 1 shows a detailed description of the studied micromixer geometry, such as: Fig. 1

Governing equations

In this study, the flow and mixing simulations were performed using Computational Fluids Dynamics (CFD) code ANSYS Fluent 16.0. The governing equations of 3D steady and incompressible flows are the continuity equation (Eq. (1)) and the momentum equation (Eq. (2)) which can be expressed in the following form:V=0VV=-1ρP+ν2Vwhere V, ρ and P represent the velocity, fluid density and static pressure respectively.

The species transport equation (Eq. (3)) for a fluid with constant density is a

CFD code validation

In order to validate our obtained results by the CFD code, a quantitative comparison has been realized with a recent work of Raza et al. (2018), where the mixing index values at the exit plane of the micromixer at different Reynolds number were compared as shown in Fig. 2, it is clear from this figure that our results are very comparative and coincide with those of Raza et al. (2018), where the values of the mixing index have the same evolution over a wide range of Reynolds number.

To clarify

Conclusions

Numerical simulations of the mixing of Newtonian fluids were performed in a chaotic modified two-layer crossing channels micromixer. The numerical results were in good agreement with the recent numerical data. The proposed micromixer with four mixing units had at least 85% of mixing index with Re = 0.2 and exceeds 90% for Re = 20 and reaches more than 99.22 % from Re = 50. A parametric numerical study was carried out for four aspect ratios (l/W = 0.25–1), the quantitative and qualitative

Declaration of interests

The authors declare that they have no known competing financial interests or personal relationships that could have appeared to influence the work reported in this paper.

Acknowledgment

This work was supported by the Directorate General for Scientific Research and Technological Development DGRSDT.

References (29)

Cited by (12)

  • Vortex-induced vibration effects on mixing performance

    2022, Chemical Engineering and Processing - Process Intensification
    Citation Excerpt :

    In this type of micromixers, the mixing efficiency is improved without the need for any external force only by modifying the mixer structure to induce perturbations in the flow. Among the proposed passive methods in practice, some methods such as parallel [22] and serial [23] lamination, increasing number of working inlets [24], changing inlet flow direction [6], and using channels with modified geometry such as zigzag [25], spiral [26,27], intersecting [28,29], grooved [30,31] and twisted [32–34] channels are more fascinating for engineers. Although among all passive methods, the geometric modification of micromixers has received more attention, due to manufacturing considerations, their complex structures often limit their application.

  • A structural optimization model of a biochemical detection micromixer based on RSM and MOEA/D

    2022, Chemical Engineering and Processing - Process Intensification
    Citation Excerpt :

    The experimental results show that, compared with a simple curved mixer, the mixing percentage is increased by 0 to 2.04 times, and the pressure drop is only increased by 0 to 1.23 times, and the performance of the micromixer is greatly improved. Kouadri et al. [16] adopted spatial cross structure, C-shaped channel and serpentine channel in the micromixer structure. Through optimization, the thermal mixing performance and thermodynamics of the micromixer were effectively improved.

  • Comparative study of mixing behaviors using non-Newtonian fluid flows in passive micromixers

    2021, International Journal of Mechanical Sciences
    Citation Excerpt :

    Their results confirm the better mixing index of Model B compared to that of Model A which related to the geometry structure. A numerical parametric study was performed by Kouadri et al. [47] using optimized TLCCM micromixer of 3.75 mm with five values of aspect ratio (l/W) from 0 to 1 in a wide range of Reynolds numbers: from 0.2 to 80. their obtained results show that the ratio of (l/W = 1) give the excellent mixing performances where their obtained mixing index overrides 85.67% for Re = 0.2 and amounts to 99.22% for Re = 50, which has the lowest pressure drop compared to other micromixers studied recently.

View all citing articles on Scopus
View full text