&&ReWrAp:HEADERFOOTER:0:ReWrAp&&

Author

Ruchi Gupta

Date of Award

2007

Degree Type

Thesis

Degree Name

Master of Applied Science (MASc)

Department

Engineering Physics

Supervisor

Rafael N. Kleiman

Language

English

Abstract

The motivation for this project is to mimic an electronic microprocessor in the fluidic domain to enable on-chip decision making for lab-on-a-chip applications. Microfabrication can increase the economic feasibility of microfluidic computing by lowering the cost per chip. Monolithic integration also has the potential to reduce the packaging costs and reduce the interconnect delays.

In this thesis a high reliability no-moving part microfluidic NOR gate was modeled and optimized using modern and widely-accepted computational fluid dynamics tools. Our optimized microfluidic NOR satisfies all the stringent device requirements needed to make a computing system. The transfer curves are non-linear, the fan-out is greater than 12, the input and output signals are in the same domain and the flow of information is only in one direction. Fabricating the microfluidic circuits with components of a single fixed geometry is very attractive. The NOR is a universal gate and thus all logic functions can be realized by a combination of NORs.

Three important components of computing: a half adder, a static memory/latch and a clock were also modeled. We have also studied the dynamics of the device, and have understood the origin of the switching time. The performance of the device is only dependent on the Reynolds number of the supply jet. Thus the performance of the device for any fluid: air, water, oil etc, will remain unaltered at the same Reynolds number. Based on our understanding of the power vs. geometry we can predict the critical dimension of the device that can operate at the desirable power consumption.

Finally, we enVISIon applying the designed microfluidic NOR to a newly emergmg technology - i.e. Lab-on-a-chip. We envision fabricating an integrated assembly of microfluidic devices to make a microfluidic processor for lab-on-a-chip systems. This work could have a major impact for biomedical applications.

McMaster University Library

Files over 3MB may be slow to open. For best results, right-click and select "save as..."

Share

COinS