출처:http://www.rle.mit.edu/micronano/research_microdrop.htm

Electrohydrodynamic Microdroplet Generation
Sung Jae Kim, Yong-Ak Song

Abstract

The utility of electrohydrodynamics (EHD) method as an effective mean for the microdrop generation is experimentally studied. The electric field developed between the charged liquid sample and the ground electrode cause the electrical body force at the air/liquid interface. Although the Taylor cone which may hinder the accurate control of the drop is formed usual EHD method, the microdrop is generated without the Taylor cone using Teflon AF 1600 coating. The microdrop size can be controlled either by the gap between the electrode and the microchannel or the strength of the electric field. The microdrop can be produced as the same size as the size of microchannel, 20 mm without the Taylor cone.

1. Introduction

The dispensing of microdrop from a capillary with the electric field has been investigated in the field of analytical chemistry and inkjet printing technology. Recently, the EHD method are especially applied to small scale diagnostic devices such as massive parallel drug discovery and DNA microarray. In such devices, the monodisperse size, repeatability and fast generation rate are intensively required. Moreover, the injector should be miniaturized to meet the concept of MEMS and microfluidics.

The method has been studied for over a century since the time of Lord Rayleigh. When the electric field is applied between two electrode, the charged ion at the liquid/air interface experience the electrical forces. Once the electric field strength is high enough to overcome the interface surface tension, the drop is elongated in the parallel direction of the electric field. Depending on the the biasing field and the properties of fluids, the EHD method can be used to produced a spray, a continuous stream and discrete monodisperse microdrops. Since the electric field pull the interface out into a sharp cone called the Taylor cone, this type dispensing is capable of producing much smaller than the size of capillary nozzle. However, the formation of the Taylor cone may hinder the accurate control of the size of the drop because the ejecting probability varies every times even in the same conditions.

 The main focus of the project is producing a mono disperse microdrop under 50 m m diameter, corresponding to 65pl, without the Taylor cone. Moreover, an atmospheric pumping without a mechanical pump should be achieved for simplifying the connection to analytical devices such as AMS and MALDI-TOFMS.

2. Experimental Method

The fabrication of microchannel follows general PDMS chip fabrication processes. The width and height of microchannel as the generator are 15 mm and 20 mm, respectively. To minimize the surface wetting, a semi wedge type chip is made to test. Furthermore Teflon AF 1600 is coated on at the exit of the microchannel to suppress the formation of the Taylor cone. The Teflon provides a ultra hydrophobic surface and the contact angle on the Teflon surface is over 100 degree. Figure 1 shows the microdrop generator system.

Figure 1 (a) The schematic diagram of microdrop generation system and
        (b) the overview of components of the microdrop generator.

It consists of the chip, the carbon tube as a ground electrode and the high voltage power supplier. The motion of the drop is captured by high speed camera up to 8,000 fps. High speed camera needed huge amount of light, because the shutter speed is too fast. Even direct expose to the lens is needed when high frame rate.

 DI water and 0.1% v/v acetic acid are used to test the drop generation. The important differences between two liquid are viscosity and surface tension. Acetic acid gives a little more sphere-shaped drop because it has slightly higher viscosity than DI. But the two liquids have large difference in the surface tension. DI's is three times greater than acetic acid's. Since high surface tension has a large probability to make the Taylor cone, it is better to use the lower surface tension of liquid. However, with low surface tension liquid, the pressure is needed to pump the liquid from the channel because the absence of mother drop won't automatically pull up the liquid from the channel by the surface force. Actually, DI water drop was continuously generated over 3 hours without any pumping. But the generation of acetic acid drop is stopped immediately after stop pumping.

3. Results

When a sample liquid is filled the microchannel, the mother drop is form at the exit due to high wettability of PDMS surface. But an almost sphere shaped drop is formed on the Teflon surface due to the low wettability. Even the Taylor cone is disappeared by using Teflon as shown in Figure 2. The frame rate is 8,000 so that the real time of the first video is less than 30 msec.

(a)   (b)

Figure 2 The microdrop generation on (a) the PDMS surface and (b) the Teflon
coated surface. The real time of the videos is 30 msec.

The effect of the gap between the carbon electrode and the exit of microchannel is tested as shown in Figure 3. The strength of applied electric field is 2,500 V. It shows that shorter gap gives much smaller drop because the shorter gap means the higher electric field. High electric field strongly pulls up the drop so that the drop is detached quickly even if it is extremely small. The black bar is the shadow of the carbon electrode. The Taylor cone still remain when the drop is small in case of DI water due to high surface tension. However, the pumpless actuation is possible. The drops in these movies are generated 3 hours without any pumping. Also the generation rate increase with the smaller drop.

		gap	frequency	D
		(um)	(sec)	(um)
(a)		730	0.566125	108
(b)		600	0.483000	86
(c)		485	0.216500	70
(d)		420	0.117000	52
(e)		340	0.017250	30

Figure 3 The sizes of DI drop and generation frequencies as
                       a function of the gap between the chip and the electrode.

Figure 4 shows the case of acetic acid. Due to lower surface tension than DI water, the Taylor cone is removed even in case of very small drop. The size of the drop is also linearly proportional to the gap.   In this case, the channel gap is fixed at 1 mm and the strength of the electric field is changed. As you can see, the drop is magnificently generated without the Taylor cone and mother drop. The generations are repeatable and reliable. Tests with lots of microchannel are conducted several times. Once the strength of electric field and the gap are set, almost the same size drops are generated every time. The drop in Figure 4(f) has the size of only 21 microns. However, the generation needs pressure due to the low surface tension.

		|E|	D
		(v)	(um)
(a)		1200	152
(b)		1400	120
(c)		1600	106
(d)		1800	72
(e)		2000	60
(f)		2500	21

Figure 4 The sizes of acetic acid drop as a function

of an applied electric field.

The most important future work is the pumpless generation and the connecting to the devices. The modified microchannel design will use the surface tension to pull up the liquid from the inside of the channel. For the connection, the drop which is started from the exit, it should fly 2 cm long so that it enters the devices. Figuring out these problems is now under investigated.

References

1. Lee, E. R. Microdrop Generation, CRC press.
   
2. Basaran, O. A. AIChE J. 48, 1842-1848 (2002).
   
3. Saville, D. A. Annu. Rev. Fluid Mech. 29, 27-64 (1997).