MIT develops 'tractor beam' for c...(m a laser to push around and control tin...)

But silicon chips are opaque to light, so applying this technique to them not an obvious move, the researchers say, since the optical tweezers use light beams that have to travel through the material to reach the working surface. The key to making it work in a chip is that silicon is transparent to infrared wavelengths of light - which can be easily produced by lasers, and used instead of the visible light beams.

To develop the system, Lang and Appleyard weren't sure what thickness and surface texture of wafers, the thin silicon slices used to manufacture microchips, would work best, and the devices are expensive and usually available only in quantity. Being at MIT, where there is such a strength in microfabrication, I was able to get wafers that had been thrown out, Appleyard says. I posted signs saying, 'I'm looking for your broken wafers'.

After testing different samples to determine which worked best, they were able to order a set that were just right for the work. They then tested the system with a variety of cells and tiny beads, including some that were large by the standards of optical tweezer work. They were able to manipulate a square with a hollow center that was 20 micrometers, or millionths of a meter, across - allowing them to demonstrate that even larger objects could be moved and rotated. Other test objects had dimensions of only a few nanometers, or billionths of a meter. Virtually all living cells come in sizes that fall within that nanometer-to-micrometers range and are thus subject to being manipulated by the system.

As a demonstration of the system's versatility, Appleyard says, they set it up to collect and hold 16 tiny living E. coli cells at once on a microchip, forming them into the letters MIT.

'/>"/>