(Photo: courtesy of Hans Stakelbeek / FMAX; originally published in Delta)

A Pinball Machine for molecules 

Dr. ir. Gea Parikesit designed a chip for sorting DNA molecules based on their sizes. This could offer new possibilities for performing genetic tests. Even though some things went wrong during the course of the research, in the end the chip still works as desired.

by: Jos Wassink (originally published in Delta)

"The situation forced me to think more creatively," said Gea Oswah Fatah Parikesit, trying to summarize his research experience. He took a small Tupperware box out from his backpack, and from inside the box he grabbed a small plastic transparent block with the size of an ordinary matchbox. Two copper clamps on the top of the block hold a chip, which is as small as a postage stamp. Small channels were seen inside the block, which can be used for fluidic connections between the chip and the instruments outside the block. At the bottom of the block, a PCB (printed circuit board) is installed for electrical connections. "This is my Lab-on-a-Chip (LOC)," said Parikesit. "You only need to set up the fluidic channels and electrical connections and you can directly put the whole block under a microscope to see what is going on inside the chip." A part of his thesis shows what you would see through the microscope: inside a semi-circular device on the chip, DNA molecules labeled with fluorescent dyes (these are some molecular flags that glow when hit by a laser) travel toward a small corner. They then make a sharp U-turn before going through one of the available outlets and disappear in the other end of the device. The whole device looks just like a Pinball Machine in a molecular scale.

The development of the device was inspired by a Mass Spectrometer, a tool that works in a vacuum and separate molecules based on their masses. The larger the mass of a molecule, the further away this molecule 'flies' to the left. "That was also the initial plan with this fluidic device", said Parikesit. Electrodes fabricated in the left- and right-side of the semi-circular chamber would be able to control how far the passing DNA molecules make a U-turn path. Longer DNA molecules have a larger electrical charge compared to shorter molecules. The longer molecules would then 'feel' a stronger electrostatic force and eventually are deflected further away to the left. That would offer a possibility to separate DNA molecules, while contained in a fluid solution, based on their sizes.

DIMES (the clean-room facility in TU Delft) fabricated the chip, comprising four identical devices, with an area of only around a squared centimeter. Each device have a radius of 1 millimeter, while
the inlet channel is about one-tenth millimeter wide and only less than a micrometer (one-thousands millimeter) deep. However, as the first experimental attempts showed that the devices always have uncontrollable fluidic leakages (whenever the electrodes are fabricated), Parikesit then proposed to just leave out the electrodes and investigate whether he could still use it to achieve the separation of the molecules.

Very strangely, the device still works. An external electric field, applied simply between the inlet and the outlet, pushes the molecules along the device. The PhD candidate then showed in a graph how the molecules with different sizes travelled inside the device. "I wanted (to use this electric field for making) the molecules to be sorted based in their sizes, but nobody has ever done that before." He observed that heavier molecules are deflected further away compared to lighter molecules, just as in a Mass Spectrometer. However, this happens only when the DNA molecules travelled close to the sharp corner in the end of the inlet channel. This way, he could sort two types of DNA molecules from each other, where one type is three times longer than the other.

According to one of his supervisors, Prof. dr. Yuval Garini (Bar-Ilan University, Israel), Gea Parikesit has developed an important and fundamental work. The idea of a "Lab-on-a-Chip" (LOC) is not new, and some devices are even already available commercially. But, according to Garini, the small scale of Parikesit's chips make them unique. "The channel depth is around 150 nanometer, whereas conventional devices are between ten to hundred times deeper than that." The advantage of working with such a small scale is that you could perform a medical diagnosis using only a small number of molecules.

The quality of the molecular sorting is still crude, but the sorting principle has now been shown experimentally. The PhD candidate also offered a few suggestions to improve the sorting quality: you could make the width of the inlet channel smaller, or improve the control of the inlet flow, or serially connect several identical devices. To begin with, there are now already four devices fabricated on a single chip. Garini: "All starting phase is difficult, but once it works, then it would be easy to continue later on and achieve an improvement with a factor of ten or more."

'Nanofluidic electrokinetics', Gea Oswah Fatah Parikesit.
Promoted by Prof. Ted Young, Applied Sciences, on 14 januari 2008.

---

(click the image below to see the full-sized scan of the original article)