In Situ Observation of Diffusion Mixing in a Micro-fluidic Mixer

Yuta Morizane and Toshiyuki Horiuchi

Tokyo Denki University, 5 Senju-Asahi-cho, Adachi-ku, Tokyo 120-8551, Japan

Keywords: Micro-Fluidic Device, Diffusion Mixing, Projection Exposure, Micro-Mixer, Phenolphthalein.

Abstract: Mixing of laminar two-liquid flows in micro-mixers was visualized and analyzed by mixing alkaline solution

and phenolphthalein. Micro-mixers with flow paths fabricated by optical projection lithography were used.

The two liquids were injected using a syringe pump from Y-shape inlets, and states of mixing were observed

using an optical microscope with a high-resolution digital camera. By using above mentioned two liquids,

transparent liquids were colored in red when they were mixed. For this season, the mixing was clearly

visualized. Because Reynolds number of the flow was so small as 0.27-17.7, the flow became the laminar

one. Accordingly, two liquids were not mixed near the junction where they were joined together. However,

they were gradually mixed by diffusion during they flowed in the mixer paths. It was clarified that the mixing

ratio varied depending on the flow-path shape, flow rate, and flow-path width.

1 INTRODUCTION

Recently, various analyses using micro-fluidic

devices are proposed in chemical, bio, and medical

fields (Akhtar, et al., 2015) (Inami, et al., 2009) (Pu

and Liu, 2004) (Stone, et al., 2004) (Somaiyeh, et al.,

2015) (Watts and Wiles, 2007). There are many

advantages for using micro-fluidic devices. For

example, fluids can be mixed in a narrow area, and

even dangerous chemicals can be safely mixed. In

addition, because the contact area per unit volume of

liquids is increased, high mixing efficiency is

obtained, and mixing speed is improved. For this

reason, it is expected that such micro-fluidic devices

are applicable to analyses of DNAs, genomes, and

proteins. They will be especially useful for low-

volume wide-variety manufacturing of

pharmaceutical devices in the future.

In the field of bio-technology, among various

micro-fluidic devices, micro-mixers are conveniently

used for various analyses and diagnoses. There are

two types of micro-mixers. One is a type in which

external mixing energies are supplied. For example,

electric fields (Harnett, et al., 2008) (Oddy, et al.,

2001) or vibrations (Yang, et al., 2001) are given.

Second-type devices are the ones using no external

energies (Liu, et al., 2000). It is reported that mixing

efficiencies are improved by contriving channel

shapes (Wang and Hu, 2010) (Wang, et al., 2002)

(Wang, et al., 2012).

Using external energy, efficient mixing is

realized. However, the structure of mixer becomes

complicated. In addition, in the case of bio-liquids,

external mixing energy sometimes gives harmful

influence to the liquids. On the other hand, micro-

mixers without using external energies are easily used

under any circumstances. In addition, because a large

number of micro-fluidic devices are often used

simultaneously, they had to be easily made with low

costs. For this reason, simply manufacturable mixers

without using external energies are preferred.

Therefore, such type simple micro-mixers are

investigated here.

In this research, micro-mixers were fabricated

using a newly developed low-cost 1:1 projection

exposure system (Morizane et al., 2014)

(Morizane

and Horiuchi, 2015). The system has a 20-mm square

exposure field, and straight-type, snail-type, and

meander-type mixers with sizes of 10×15 mm

10×15 mm

and 15×15 mm

were fabricated. The

flow paths with a width of 100 μm were directly

fabricated using resist patterns with vertical side

walls. The mixers were assembled by binding acrylic

lid plate and vessel plate sandwiching the flow path

substrates by four screws.

To evaluate the states of mixing clearly, alkaline

solution and phenolphthalein are mixed, reffering to

the past researches. In the first past research, a snail-

type micro-mixer was fabricated in 2-mm square area

using 1/19 projection lithography (Horiuchi, et al.,

Morizane, Y. and Horiuchi, T.

In Situ Observation of Diffusion Mixing in a Micro-ﬂuidic Mixer.

DOI: 10.5220/0005757902090215

In Proceedings of the 9th International Joint Conference on Biomedical Engineering Systems and Technologies (BIOSTEC 2016) - Volume 1: BIODEVICES, pages 209-215

ISBN: 978-989-758-170-0

209

② Exposure

① Resist coating

③ Development

④ Assemble

2000), and in the second past research, meander-type

micro-mixer was fabricated in 6×15 mm

area using

contact lithography (Horiuchi and Yoshino, 2014). In

these past researches, mixing performances were

observed and evaluated using colored waters.

However, it was difficult to quantitatively evaluate

the mixing, and clarify what parameters mainly

influence the mixing. For this reason, above

mentioned liquids were selected. It is considered that

because the color of mixed parts are changed from no

color to vivid red, they are distinguished clearly.

Two liquids are injected using a syring pump, and

the mixing states are in situ observed by an optical

microscope with a digital camera. As a results,

mixing states are clearly visualized. Mixing ratio and

main parameters influencing the mixing are

investigated and discussed in detail.

2 FABRICATION OF

MICRO-MIXERS

Micro-mixers were fabricated, by the procedure

shown in Figure 1. At first, flow path patterns were

fabricated on a synthetic quartz substrate using

projection lithography. Negative resist SU-8 (Micro

Chem) with a thickness of 100 µm was used. Next,

the substrate was softly baked at 65 °C for 20 min and

at 95 °C for 50 min continuously. Then, the resist was

exposed to the shape of flow paths using the 1:1

projection exposure system shown in Figure 2. The

sizes of the system were 300 mm wide, 600 mm deep,

and 1,100 mm high. As a light source, an ultra-violet

lamp with a main wavelength of 365 nm (Inflidge,

UV-CURE120) was used. A macro lens for a camera

(Sigma, 50 mm, F2.8 EX DG MACRO) was used as

the projection lens. Numerical aperture was set to

0.089. The exposed substrate was rebaked. The post-

exposure bake was done at 65 °C for 10 min and 95

°C for 30 min. After cooling down the substrate to

room temperature, the resist on the substrate was

developed in SU-8 developer for 10 min. By exposing

the resist under appropriately defocused conditions,

flow path patterns with almost vertical side walls

were obtained, as shown in Figure 3.

Three types of mixers with different flow-path

patterns were fabricated, as shown in Figure 4. One

was a straight-type mixer with Y-shaped inlets.

Second one was a snail-type mixer. Third one was a

meander-type mixer. The quartz substrate with the

flow path pattern was sandwiched by acrylic vessel

and lid plates, and assembled to a micro-mixer by

binding the four corners using M3 screws. Teflon

tubes with an outer diameter of 1.6 mm and inner

diameter of 0.7 mm were attached to two inlets and

one outlet using an adhesive. To prevent the

distortions of lid and vessel plates, and improve the

sealing, a rubber sheet was inserted between the

vessel plate and the substrate.

Figure 1: Fabrication method of micro-mixer.

Figure 2: Exposure system used for the research.

Figure 3: Profiles of flow-path patterns.

(b) Bird’s-eye view. (a) Cross section.

100 μm

Reticle

Quartz substrate

placed on Z stage

Illumination optics

Collective lens

Projection lens

XY stage

UV light source

100 m

Resist

Quartz substrate

Light rays

Reticle

Projection lens

Sensitized parts

Lid plate

Rubber sheet

Vessel plate

BIODEVICES 2016 - 9th International Conference on Biomedical Electronics and Devices

210

(a) Straight-type. (b) Snail-type.

Figure 4: Shapes and sizes of fabricated micro-mixers.

Assembled micro-mixer is shown in Figure 5. By

grace of elasticity of the rubber sheet and SU-8, the

flow paths of mixers were successfully sealed and no

leaks were observed.

Figure 5: Assembled micro-mixer.

3 TWO LIQUID MIXING

3.1 Evaluation Method

Mixing of two liquids injected in micro-fluidic paths

were in situ observed. Micro-mixers explained in

section 2 were used for the experiments. The in situ

observation method is shown in Figure 6. The mixers

were placed one by one on the stage of optical

microscope (Arms system, IMZ-20CU) with a high-

resolution digital camera (Arms system, HV-20CU).

The two liquids were simultaneously injected using a

syringe pump (AS ONE, MSP-3D). Because all of the

acrylic lid, resist flow path, and synthetic quartz

substrate were transparent, and a white paper sheet

was laid under the quartz substrate, liquids were

observed as they were in natural colors, if they were

colored. In the past researches, waters colored in

advance were mixed. However, it was difficult to

observe the mixing state clearly. For this reason, as a

method for evaluating liquid mixing more clearly,

two transparent liquids colored only when they were

mixed were chosen. In concrete, strong alkali and

phenolphthalein were mixed. As the alkali, 3%

TMAH (Tetra methyl ammonium hydroxide) with pH

of 13 (Tokyo Ohka Kogyo, PMER P-7G) was used.

The phenolphthalein was dissolved in ethanol, and

measured pH was 7. When they were mixed, they

were colored in vivid red. According, weather they

were mixed or not were clearly discriminated. Red

and colorless parts were very easily distinguished.

Thus, mixing of two liquids was visualized on a

monitor display as the in situ image of the digital

camera attached to the optical microscope.

To evaluate the mixing ratio quantitatively, the red

and colorless areas were binarized to black and white.

The threshold was decided by the red color of the

fully mixed area. Next, the black and white ratio in

the width direction of the flow paths was calculated

as the mixing ratio R

. The mixing ratio R

was

defined by eq. (1), as shown in Figure 7, and

evaluated at the points shown in Figure 8.

R =

(1)

Figure 6: Set up for injecting liquids in the micro-mixer and

in situ observation system.

Vessel plate

Tube

Bolt and nut

Quartz substrate

with flow paths

Lid plate

Rubber sheet

10 m

In Situ Observation of Diffusion Mixing in a Micro-ﬂuidic Mixer

211

Figure 7: Figure for explaining mixing ratio R

(a) Straight-type. (b) Snail-type.

Figure 8: Points where the mixing ratio R

was evaluated.

3.2 Mixing Ratio Dependence on Flow

Rate and Flow-Path Shape

Mixing ratios were measured along each flow path by

1 or 2 mm intervals, and the results are shown in

Figures 9-11. It was known that the red part widths

gradually increased along the flow paths after the two

liquids were joined using the Y-shape inlets. When

the flow rate was changed from 0.1 (Re=0.27) to 6.4

(Re=17.7) ml/h, the mixing ratio changed almost

regularly. Here, Re is the Reynolds number. It was

clarified that the mixing ratio strongly depended on

the flow rate. Because Reynolds numbers were very

small, and the liquids were mixed by diffusion, the

mixing ratio became large for small flow rate.

Next, by comparing the mixing ratios between the

snail-type and the meander-type channels, difference

of the mixing ratio was investigated. Under all the

flow rate conditions of 0.1-6.4 ml/h, mixing ratios of

the snail-type mixer were a little higher than that of

the meander-type mixer. It was observed that the

liquids in the snail-type channel were rapidly mixed

after passing corners. Accordingly, it was considered

that the right-angled corners like those used in the

snail-type channel were effective for mixing.

A panoramic photograph of the visualized

meander-type channel (flow rate=0.4) is shown in

Figure 12.

Figure 9: Mixing ratio of straight-type micro-mixer with a

flow-path width of 100 μm.

Figure 10: Mixing ratio of snail-type micro-mixer with a

flow-path width of 100 μm.

Figure 11: Mixing ratio of meander-type micro-mixer with

a flow-path width of 100 μm.

Visual difference of mixed flow are shown in

Figure 13. These photographs were taken at the point

3 of the snail-type mixer. It is clearly known that the

mixed area or mixing ratio decrease depending on the

flow rate.

BIODEVICES 2016 - 9th International Conference on Biomedical Electronics and Devices

212

Figure 12: Panoramic photograph of meander-type mixer.

It is clearly observed that the flow width of mixed liquids

increases gradually from the junction point to the outlet.

Figure 13: Visual differences of mixed flows observed at

point 3 of the snail-type mixer.

3.3 Influence of Flow-Path Width

Mixers with a width and depth of 100 μm were used for the

experiments shown in section 3.2. In this section, a straight-

type mixer with a width of 50 μm and a depth of 100 μm

was used, and the mixing performances were compared.

Figure 14 shows the results. Comparing Figure 14 with

Figure 9, it was found that the mixing ratios of the channel

with a narrow flow path were higher than those with a wide

flow path.

Figure 14: Mixing ratio of straight-type micro-mixer with a

flow-path width of 50 μm.

4 DISCUSSION

Changes of mixing ratios caused by differences of

flow rate and flow path were discussed. In the case of

micro-fluidic devices, the width and depth of flow

path are very small. For this reason, Reynolds number

becomes very small, and the flow becomes laminar.

In a laminar flow, mixing by turbulences of flow is

not expected, and giving external forces for stirring

the flow in the micro-flow path is also difficult.

Therefore, molecular diffusion becomes the most

probable mixing phenomenon.

In diffusion mixing, diffusion time t is expressed by

t ≈ L

/ D (yamamoto, et al., 2013).

(2)

Here, D and L are diffusion coefficient and length,

respectively. Accordingly, t is proportional to L

, if

the diffusion coefficient D is constant. In the case of

micro-mixer investigated here, diffusion length L

corresponds to (w

/2) because the mixing of liquids

expand from the flow-path center to the outside. On

the other hand, diffusion time t corresponds to the

distance d from the junction point. If the flow rate Q

and the cross section A of flow path are constant, t is

calculated by the following equation because the flow

velocity v is obtained by v = Q/A.

t ===

(3)

Referring to the experimental results shown in

Figures 9-14, the mixing ratio gradually saturates and

approaches to 100 %, and roughly speaking, the

curves seem almost parabolic, as shown in Figure 15.

If the curve is parabolic, d is proportional to 





Because R

equals to w

/ w

= 2L/ w

and d = (Q/A)

In Situ Observation of Diffusion Mixing in a Micro-ﬂuidic Mixer

213

t, t is proportional to L

. Therefore, it is thought that

the experimental results have a tendency

approximately agreed with eq. (2).

Figure 15: Almost parabolic relationship between the

mixing ratio R

and the distance d from the junction point.

If the curve is parabolic, t is proportional to L

5 CONCLUSIONS

Micro-fluidic mixer patterns were formed on quartz

substrates using 1:1 optical projection lithography

system, and three types of micro-mixers were

fabricated sandwiching the quartz substrates by

acrylic lid and vessel plates. Mixing of two liquids

were investigated by injecting strong alkali and

phenolphthalein solutions simultaneously from Y-

shape inlets. The state of mixing was in situ observed

on a monitor display using an optical microscope with

a high-resolution digital camera. The mixing was

clearly visualized caused by the chemical color

change from no color to red. Because the flow width

colored in red gradually increased along the flow

paths, mixing ratio was measured as the colored width

devided by the full width of the flow path.

Then, the mixing ratio dependence on flow rate

and flow path was clarified. The mixing ratio

increased when flow rate was decreased, and narrow

flow-path was used. It was demonstrated that the new

in situ observation method was effective to clarify the

diffusion mixing in micro-fluidic mixers. It was also

found that right-angled corners of flow paths were

effective for advancing the mixing of liquids.

In this research, flow path depth was fixed to 100

μm. It is considered that the mixing is also influenced

by the flow path depth or the aspect ratio of the flow-

path cross section. It is necessary to investigate

hereafter.

ACKNOWLEDGEMENTS

This work was partially supported by Research

Institute for Science and Technology of Tokyo Denki

University, Grant Number Q15T-03.

REFERENCES

Akhtar, M., Brandhoff, L., Driesche, S., Vellekoop, J, M.,

2015. Air-droplets as Gas Reservoir to Provide O2 to

the Stored-Aqueous Droplets in Micro-channels.

Procedia Engineering, 120, pp. 92-95.

Harnett, C. K., Templeton, J., Dunphy-Guzman, K. A.,

Senousy, Y. M., Kanouff, M. P., 2008. Model based

design of a microfluidic mixer driven by induced charge

electroosmosis. Lab Chip, 8, pp. 565-572.

Horiuchi, T., Yoshino, S., 2014. Fabrication of Precise

Micro-fluidic Devices Using a Low-cost and Simple

Contact-exposure Tool for Lithography. Proceedings of

The third International Conference on Biomedical

Electronics and Devices, pp. 5-11.

Horiuchi, T., Watanabe, H., Hayashi, N., Kitamura, T.,

2010. Simply Fabricated Precise Microfluidic Mixer

with Resist Flow Paths Sealed by an Acrylic Lid.

Proceedings of The third International Conference on

Biomedical Electronics and Devices, pp. 82-87.

Inami, H., Tsuge, K., Matsuzawa, M., Sasaki, Y., Togashi,

S., Komano, A., Seto, Y., 2009. Semi-automated

bacterial spore detection system with micro-fluidic

chips for aerosol collection, spore treatment and ICAN

DNA detection. Biosensors and Bioelectronics, 24, pp.

3299-3305.

Liu, R. H., Stremler, M. A., Sharp, K. V., Olsen, M. G.,

Santiago, J. G., Adrian, R. J., 2000. Passive Mixing in

a Three-Dimensional Serpentine Microchannel.

MICROELECTROMECHANICAL SYSTEMS, 9, pp. 2.

Morizane, Y., Uchino, S., Watanabe, Y., Horiuchi, T., 2014.

Development of a Simple and Low-Cost Projection

Exposure System for Fabricating Micro-Components

with Large-Size Patterns. Dig. Papers, Photomask

Japan 2014, pp. 42.

Morizane, Y., Horiuchi, T., 2015. Fabrication of

Microfluidic Paths Using Thick and Vertical Resist

Patterns Printed by Exposure with Intentional Large

Defocus. Dig. Papers, Photomask Japan 2015, pp. 43.

Oddy, M. H., Santiago, J. G., Mikkelsen, J. C., 2001.

Electrokinetic Instability Micromixing. Analytical

Chemistry, 73, pp. 5822-5832.

Somaiyeh, A., Haghighi, M., Ebadi, A., 2015. Direct

synthesis of DME over nanostructured CuO–ZnO–

Al2O3/HZSM-5 catalyst washcoated on high pressure

microreactor. Chemical Engineering Journal 262, pp.

1175-1186.

Stone, H. A., Stroock, A. D., Ajdari, A., 2004. Engineering

flows in small devices: Microfluidics toward a lab-on-

a-chip. Annual Review and Fluid Mechanics, 36, pp.

381-411.

BIODEVICES 2016 - 9th International Conference on Biomedical Electronics and Devices

214

Pu, Q. S., Liu, S. R., 2004. Microfabricated electroosmotic

pump for capillary-based sequential injection analysis.

Analytica Chimica Acta, 511, pp. 105-112.

Yang, Z., Matsumoto, S., Goto, H., Matumoto, M., Maeda,

R., 2001 Ultrasonic micromixer for microfluidic

systems. Sensors and Actuators, 93, pp. 266-272.

Yamamoto, D., Maki, T., Watanabe, S., Tanaka, H.,

Miyahara, T. M., Mae, K., 2013. Synthesis and

adsorption properties of ZIF-8 nanoparticles using a

micromixer. Chemical Engineering Journal, 227, pp.

145-150.

Wang, C. T., Hu, Y. C., 2010. Mixing of liquids using

obstacles in Y-type microchannels. Science and

Engineering, 13, pp. 385-394.

Wang, H. Z., Iovenitti, P., Harvey, E., Masood, S., 2002.

Optimizing layout of obstacles for enhanced mixing m

microchannels. Smart Mater and Structres, 11, pp. 662-

667.

Wang, L., Liu, D., Wang, X., Han, X., 2012. Mixing

enhancement of novel passive microfluidic mixers with

cylindrical grooves. Chemical Engineering Science, 81,

pp. 157-163.

Watts, P., Wiles, C., 2007. Micro reactors: a new tool for

the synthetic chemist. Organic & Biomolecular

Chemistry, 5, pp. 48.

In Situ Observation of Diffusion Mixing in a Micro-ﬂuidic Mixer

215