State of Utah Center of Excellence for Biomedical Microfluidics

Home Research Publications Personnel Education

Gas Permeation Pumps
Gas Permeation Pumps Rotary Pumps Osmotic Pumps

 

Gas Permeation Microump

Students: Mark Eddings

News/Media Clips on this Technology:

KSL TV – ‘Researchers Developing “Laboratory on a Chip”’
http://www.ksl.com/?nid=148&sid=702267

Medical Laboratory World

Medical News Today

Abstract
A PDMS diffusion-based membrane pump has been developed for on-chip fluid handling within microfluidic devices. Vacuum and air pressure applied across a thin PDMS membrane were used to bi-directionally manipulate fluid at flow rates of 10-200 nl·min-1. Flow rates can be altered by changing diffusion area, membrane thickness, and applied pressure or vacuum. The pump has been used to rapidly fill and evacuate dead-end channels and chambers. Desired volume plugs are easily isolated and manipulated. The pump can be integrated within densely arrayed microfluidic channels and chambers for mixing, separation, and analysis.

Experimental
In order to fabricate the diffusion membrane pump, previously reported multi-layer soft lithography methods were utilized [1-2]. Devices consisted of a microfluidic channel layer, a thin membrane layer, and the pressure/vacuum control layer (See Figure 1a). Tests were run varying parameters such as diffusion area, membrane thickness, and applied pressure/vacuum to determine the change in flow rate. Separate devices were fabricated to demonstrate fluid handling and dead-end channel filling (See Figure 1b).  A description of the permeation mechanism is shown in Figure 2.

Figure 1. a) Drawing of a microfluidic device used for measuring flow rates. The three-layer PDMS device consists of a fluid channel layer, diffusion membrane, and vacuum source layer. b) Drawing of a microfluidic device used for demonstrating dead-end chamber filling.

Results and Discussion
The flow rate varied linearly with diffusion area and pressure/vacuum (Figure 2). Flow rates from 10-200 nl·min-1 were achieved with the fabricated device. Higher flow rates are theoretically possible assuming a 2-D Steady-State diffusion model, neglecting flow into the bulk regions of the PDMS. Membrane deflection was also modeled and was found to increase the diffusion area by expanding the membrane and decrease the diffusion area by obstruction on the channel ceiling.

Fluid was easily manipulated through turns in cross intersections and to fill dead-end channels and chambers (Figure 3). Individual chambers were filled with different colored fluids without mixing. Rapid fluid plug dispensing and fluid mixing was possible by toggling chambers between pressure and vacuum.

Figure 2.  A comparison of theoretical and experimental flow rate results as for different pressures and diffusion areas.  The deflection membrane had a thickness of 0.025 mm.  Data points represent experimental data while theoretical results are denoted by lines.

Conclusion
The PDMS diffusion pump is a feasible option for complex lab-on-a-chip applications. The micropump is easily fabricated using existing methods and can be highly arrayed for multi-channel configurations. Rapid dead-end channel filling and flow rates in the 200 nl·min-1 range have been demonstrated.  It is also possible to generate slow flow rates below 10 nl·min-1.  Future work will focus on prolonged, slow flow rates for drug delivery applications.

References
[1] D. Duffy, J. McDonald, O. Schueller, G. Whitesides, Rapid Prototyping of Microfluidic Systems in Polydimethylsiloxane, Anal. Chem. 70, pp. 4974-4984 (1998)
[2] M. Unger, H. Chou, T. Thorsen, A. Scherer, S. Quake, Monolithic Microfabricated Valves and Pumps by Multilayer Soft Lithography, Science, Vol. 288, pp. 113-116 (2000)
[3] M. A. Eddings, B. K. Gale, A PDMS-based Gas Permeation Pump for On-chip Fluid Handling in Microfluidic Devices, J. Micromech. Microeng. 16 (2006)

 

A

B

C

D

E

F

 Figure 3.  Pictures of three different fluids, red, green, and blue, filling dead-end chambers (A-F).  Pressure differentials across the membrane below the chambers enabled the flow to make turns at the intersection.  Each fluid was placed in the open well and pumped to each individual chamber.  The fluid in each chamber can be individually returned to the well or can mix with the other wells at the intersection by toggling between pressure and vacuum below the membrane at each chamber.  This technique also enables the isolation of fluid plugs for dispensing and metering similar to an electroosmotic flow setup.

Text Box:  Figure 3.  Pictures of three different fluids, red, green, and blue, filling dead-end chambers (A-F).  Pressure differentials across the membrane below the chambers enabled the flow to make turns at the intersection.  Each fluid was placed in the open well and pumped to each individual chamber.  The fluid in each chamber can be individually returned to the well or can mix with the other wells at the intersection by toggling between pressure and vacuum below the membrane at each chamber.  This technique also enables the isolation of fluid plugs for dispensing and metering similar to an electroosmotic flow setup.

 
Send mail to bruce.gale@utah.edu with questions or comments about this web site.
Copyright © 2006 State of Utah Center of Excellence for Biomedical Microfluidics
Last modified: March 09, 2007
Disclaimer