(83i) Investigation of Condition for Bubble Detachment from Microchannels

Conditions for Bubble Detachment from Microchannels

1. Introduction   Micromachining technologies have recently been used to make miniature devices for synthetic applications. These devices are called microreactors[1]. Several successful microreactors have been developed[2], but additional work is required before they can be used in industrial processes.   In a microreactor, two liquids make contact inside minute channels in order to mix due to molecular diffusion. By definition, a minute mixing channel is narrow; therefore, the throughput of such a mixing channel is inevitably low. To increase the throughput, two liquids flow alternately in one channel. This flow is called multilayer flow. Each layer of liquid has a width of about 100 µm in a single channel.   The central contraction flow type microreactor, which was developed by Hitachi Ltd., forms 48 pairs of multilayer flows and has a throughput of 20 ml/min. Multilayer flow microreactors use only 1 pair of pumps, which means that if one channel is blocked by a bubble, the subsequent change in flow effects the other channels. This means it is impossible to accurately predict a microreactor's performance.   As bubbles are the root cause of the problems with predicting microreactor performance, we developed a prediction method of removing bubbles. We first developed our method as a model and then demonstrated its effectiveness by experiment.

2. Bubble blockade in microchannels   A photograph of a microreactor is shown in figure 1. The two reactants are pumped into the mirocreactor. In the reactor, the reactants are divided into plural channels in which the two reactants flow in alternate confluent layers, forming multilayer flow. An enlarged internal view of a microreactor is shown in figure 2. One reactant flows from the top of the figure to the bottom; the other liquid flows from the nozzles in the center. These two liquids form a multilayer flow of 6 pairs. In fig. 2, three of the six nozzles are blocked by bubbles. In this state, multilayer flow forms in only 3 pairs; therefore, it is impossible for uniform flow to form, and if uniform flow does not form, the mixture ratio of the two reactants is not correct, meaning that the product yield decreases.   We have developed a method that uses differential pressure to detach bubbles. Differential pressure was chosen because it is dependent on the flow rate, which is relatively easy to measure and control.

3. Bubble detachment model   The surface tension of a liquid determines the conditions under which a bubble detaches itself from a nozzle; therefore, it is possible to calculate the conditions when a bubble will detach from a nozzle condition using surface tension as a parameter. We assumed the conditions detailed below for when a bubble attached to a nozzle will become detached due to a differential pressure.   First, a bubble, which is floating in a reactant container or tube, attaches itself to a nozzle. Next, if the pressure upstream become higher than that downstream, the bubble moves downstream. When the bubble stops moving, it is possible to use equation (1) because the forces on both sides of the bubble are in equilibrium.

(P₁: pressure downstream, P₂: pressure upstream, g: surface tension, ?Æ: contact angle of upstream, ?": contact angle of downstream, r: nozzle radius)

			-------(1)

  In addition, the maximum force acting on the upstream side of the bubble is when the surface tension of the bubble becomes maximum, which is when the radius of the downstream side of the bubble becomes a minimum, which, in turn, is when the bubble radius is equal to the radius of the nozzle. Therefore, as the contact angle of the downstream side of the bubble reaches 0 degrees, it is possible to derive equation (2).

  According to Equation (2), if a larger differential pressure than ?¢P is generated between the two sides, the attached bubble will detach.   We used ethanol solution to demonstrate that equation (2) was valid because it easy to adjust the surface tension of ethanol by varying its concentration. The surface tension for several concentrations was used as a parameter in our calculations.

3. Experimental result and discussion   We measured the differential pressure at which a bubble detached from a nozzle for a given flow rate. Ethanol solution was pumped into the microreactor, and the flow rate was gradually increased. The conditions under which bubbles detached were observed. The flow rate was measured when all the bubbles were had detached.   The differential pressure when the number of attached bubbles became 0 for each value of surface tension of the solution is shown in Fig. 3. The figure also shows the calculated value for the differential pressure. Since the calculated value and experiment value agreed to ±9%, the validity of the bubble detachment model was verified. 4. Conclusion   We constructed a bubble-detachment model in order to investigate the conditions under which a bubble will detach from a nozzle in a microreactor. We used this model to investigate the relationship between the differential pressure across a bubble that results in it detaching and the surface tension of the bubble. Our experimental results and the calculations were in good agreement; therefore, we verified that controlling the flow rate is a practicable method of removing bubbles from microreactors. 

REFERENCES

1. Benson, R. S. and J.W. Ponton, "Process Miniaturisation - A Route to Total Environmental Acceptability?", Trans. IChemE, Vol. 71, Part A, (1993) 160–168.

2. Ehrfeld, W., Golbig, K., Hessel, V., Lowe, H. and Richter, T., "Characterization of Mixing in Micromixers by a Test Reaction: Single Mixing Units and Mixer Arrays", Ind. Eng. Chem. Res. 38, 3 (1999) 1075–1082.