3D pattern formation in microchannels

Another direction in our work is to study how to make 3D structures of drops in microchannels. Monodisperse drops are ideal as their size and volume fraction can be easily tuned by changing the water to oil flow rates so different packings can be obtained. We think 3D assemblies made of monodisperse submicron drops are ideal as for example templates for photonic crystals.
Monodisperse drops arrive with a given velocity to a "dimond" shape 2D reservoir (width of the channel increases linearly), so the drop velocity decreases as it enters the reservoir (flow rate is equal to the product of the velocity times the section of the channel, so in principle if the channel becomes wider linearly the velocity should decreases linearly in order to keep the flow rate constant). We should point out that the height of the channels sometime varies as PDMS is a deformable rubber. Pressures inside the channels are generally very high and the larger the channel width the easier it is for the upper wall to deform. In this particular case the reservoir widdens linearly from 100 microns to 500 microns and we have measured heights of 200 microns at the widest section.
We have studied 3D droplet packing at different droplet volume fractions. For a given water flow rate (dispersed phase) the drop volume fraction is tuned by changing the oil flow rate. The highest the oil flow rate the lowest the drop volume fraction. Unfortunatelly by variying the total oil and water flow rate the wall deformation also vary and any packing prediction become more difficult. At this point we are changing the channel design in order to work at fixed pressures (channel heights) and vary volume fraction.
1D-3D-2D transition is observed as the oil flow rate is decreased for a constant water flow rate of 100 microleters per hour. Drops are spherical with a 60 micrometer diameter and are stabilized by 1 micron size colloids. In the central 3D picture is possible to see three layers of spherical drops arranging in an ABC lattice.  

This web page is maintained by:

Galder Cristobal-Azkarate
9 & 15 Oxford Street
Gordon McKay Laboratory, Room #519
Cambridge, MA 02138
617 496 9562

Last update 07/05/03

This work is in collaboration with