Unfortunately, due to their extreme size, the forces at work in these chips are difficult to measure and sometimes understand. A few interested assistant professors at Johns Hopkins Whiting School of Engineering and members of the school's Institute for NanoBioTechnology, have undertaken what is called dimensional analysis to try to understand the physics involved in these microscopic devices. Dimensional analysis is simply the scaling up of a system so that the relative mechanics remain unchanged, but allowing the forces at work to be measured with greater ease.

Joelle Frechette and German Drazer, the aforementioned assistant professors specializing in chemical and biomolecular engineering, turned to a popular children's toy to mimic the internal physics of microfluidic arrays. Looking like a high school science fair project, their test bed consists of little more than an aquarium, a few gallons of glycerol, some steel and plastic beads and LEGOs.

Constructed primarily of the common three-high single block tube, stacked two high for each peg, on a LEGO board and secured to a piece of plexiglass for added rigidity, the group formed a lattice not unlike the well-known Price is Right Plinko board. And their tests operated in much the same manner, sans whammies.

Department of Chemical and Biomolecular Engineering graduate students Manuel Balvin and Tara Iracki, along with undergraduate Eunkyung Sohn, performed several trials for each of size of the various stainless steel and plastic balls. Dropping the test beads into the glycerol and tracking their movement to the bottom, the researchers were able to see how their size, mass and interaction with the pegs led each type to a certain path.

In order to vary the results and get an idea of how changes in the array would affect particles, the students rotated the LEGO peg board a small amount and performed the same tests each time. Somewhat unsurprisingly, they found the larger, heavier particles would always take a deterministic path to the bottom, which is to say they could predict the outcome correctly for each drop. The smaller particles took on a more random pattern, sometimes not falling straight at all, but shifting over several rows before continuing to fall.

This gave them some insight on the mechanics of the lattice and particles in relation to the force angle, or the angle of the rows of pegs in comparison with the force of gravity. "Our experiment shows that if you know one single parameter—a measure of the asymmetry in the motion of a particle around a single obstacle—you can predict the path that particles will follow in a microfluidic array at any forcing angle, simply by doing geometry," explained Drazer.

Frachette also commented on the outcome of the tests; "There are forces present between a particle and an obstacle when they get really close to each other, which are present whether the system is at the micro- or nanoscale or as large as the LEGO board. In this separation method, the periodic arrangement of the obstacles allows the small effect of these forces to accumulate, and amplify, which we suspect is the mechanism for particle separation."

The researches expect that the effect they observed for some of the test balls, known as phase locking, where the balls would follow the same trajectory even at different forcing angles, would scale down to micro- and nanoscale assemblies. The results could help improve the effectiveness of future microfluidic arrays. Different particles could be deterministicly separated and sorted by using phase locking and some quick geometry.

However, Dazer cautioned that their test is only valid at certain solution densities. Increasing the amount of the to-be-sorted constituent in the solution will inevitably lead to more particle-particle interaction which would likely render the technique inaccurate and undependable.