The progress made by microfabrication technologies has escalated the interest in particle manipulation on micron scale. Particles are not only limited to solid particles but also include cells, droplets and bubbles. Activities involved in manipulation are trapping, sorting, separation and so on. The devices designed to execute such activities are aimed for a particular application. However, there lies a gap in literature to design an all-in-one device. Microfluidics research carried out at the Laboratory for Aero and Hydrodynamics is currently working on designing such devices which can carry out different manipulation tasks.
This thesis is aimed at designing and fabricating such a device where two particles can be manipulated simultaneously. The application of designing such a device would be to study cell-cell interaction, droplet coalescence and so on. To achieve this, a microfluidic device is designed as a Hele-Shaw flow cell. Since the height-averaged velocity in a Hele-Shaw flow cell is irrotational, analytical solution is possible for such a design. Although a Hele-Shaw flow is viscous flow between two closely spaced plates, it produces same streamline patterns as a 2D ideal flow. This enables the use of basic building blocks of an ideal flow field such as sources, sinks, uniform flow and so on. In this case, sources and sinks are superimposed to get the flow field inside the microfluidic device. The microfluidic device is fabricated using Polydimethylsiloxane (PDMS), a commonly used polymer to fabricate microfluidic devices. The design of the device is validated by comparing experimental flow fields, streamlines and stagnation points to the corresponding analytical solution. Single particle manipulation activities are carried out in the device. This is executed by manually controlling the flow rates or the strength of sources and sinks. Firstly, the particle is trapped at the stagnation point for a certain timespan. Finally, some manipulation activities are carried out using a single particle with the aid of streamlines produced by the sources and sinks.

The interest in manipulating particles, droplets and bubbles have garnered significant attention in recent years, owing to the advantages offered by micro-fluidics and the advancement in micro-fabrication technologies. These manipulation activities have found its applications in myriad fields of engineering, ranging from medical diagnostics to chemical industry to drug discovery. This has increased demand for the development of devices such as 'Lab on a chip', which performs laboratory-sized experiments and analysis on a single small chip, with the same speed and accuracy as its room-sized counterpart. However, manipulation activities carried out in these devices has fixed channels, designed to serve purpose for specific manipulation tasks. This makes the device suitable for a specific application. Addressing this aspect, a device designed without having any real channels would give an opportunity to integrate multiple functionalities onto a single-chip in the long run. As a first step towards reaching this 'bigger picture', it is necessary to explore the feasibility of manipulating particles, droplets and bubbles by generating so called 'virtual channels'.
The present thesis focuses on an attempt to manipulate particles without the use of any real channels or external field. Although such manipulation is desired in the micro-scale, a top down approach is preferred and hence, the manipulation is carried out in a scaled up model. First, a Hele-Shaw flow cell is designed with sources and sinks in the millimeter scale to deviate streamlines in the same range. Thereafter, four different velocity fields are studied under different combination of sources and sinks, which are then compared to the computational ones. The property of a Hele-Shaw cell that the averaged velocity over the height of the channel is irrotational, makes it possible to compute velocity fields by the use of potential flow theory. The same velocity fields are hence, computed using a discrete source based Panel Method. A good agreement is found between computation and experiment, making PIV measurements not a necessary option for evaluation of velocity fields under these sources and sinks. Finally, individual particle is inserted into the Hele-Shaw cell and manipulated using unsteady fields. The manipulation includes tasks such as diverting particles having same initial position to different end locations; trapping particle for different instances of time and then releasing them into different directions; flipping positions of two particles; and deflecting a particle by ninety degrees. The individual particle trajectory for the above manipulation activities are tracked down using a particle tracking code and then compared with the ones generated using the Panel Method. This, however, excludes the activities where particles need to be trapped because of particle fluctuation near stagnation point. Overall, the Panel Method serves well in predicting particle path-lines and can be used as a tool for manipulating particles.