Something the size of a stamp can monitor the oceans, Giselle Weiss reports.
David Barrow fishes in his shirt pocket for a sliver of clear plastic the size of a postage stamp. What look like rows of fine threads in the centre of the plastic turn out to be a single meandering channel. The channel is designed to separate mixtures that usually requires a bulky piece of equipment in a laboratory. The miniature version can do the same thing better, says Barrow, professor of engineering at Cardiff University, and in a fraction of the time. It is an example of a technology called microfluidic microsystems, or more popularly, labs-on-a-chip.
The idea of moving, controlling and measuring fluids and gases in miniature microfluidics was pioneered by a handful of people, among them Andreas Manz, professor of chemistry at Imperial College, London. Manz says he first seized on the idea after an accident in 1986 at the Sandoz chemical plant in Basel, Switzerland, released a large amount of toxic waste into the Rhine. He wanted a system that was flexible, but small and "reasonably cheap".
In 1991, Manz proved his concept by building a miniature device that separated two dyes more quickly than the classical lab apparatus. Since then, research and commercial interest in microfluidic systems has increased dramatically. The flood of biochemical information from new drug-discovery techniques and the human genome project is just one obvious target for the fast and accurate processing that labs-on-a-chip can provide.
The United States and Japan were first to develop the technology. But in the past year the UK government's Foresight Link scheme, together with British industry, provided roughly Pounds 3 million for a 30-month programme in which a consortium of seven universities and 12 companies will develop and commercialise microdevices. Whereas the push in other countries has been on devices for use in medical diagnostics and genetic testing, the UK effort emphasises the making and analysing of chemicals.
"Microfluidics is just microplumbing, really," says Barrow. The same plumbing you find in a process plant, the gills of a fish or rocket engines, only smaller. Much smaller. The systems are made from a range of materials, including silicon and plastic, and borrow techniques from microelectronics to pattern and shape channels finer than spider silk and reservoirs for moving and processing samples.
Although most devices being developed consist of a simple component - like a single transistor, says Manz - researchers envisage a future in which perhaps ten different components, such as lasers, electronics and mechanical parts, sit alongside fluidic components on the same chip.
Microfluidic systems offer more precise results than traditional instruments in less time. They also consume less power and can work with a smaller sample. That matters, for example, in the pharmaceutical industry, where researchers may have only a tiny quantity of a potential drug they wish to analyse. Disposability means you can throw the devices away. Portability means systems can be distributed widely, in places inaccessible to larger machines.
Barrow, an ecologist by training, believes microfluidic devices are key to a sustainable future. Microanalysers to test for dissolved carbon gases in the ocean, related to greenhouse gases, could be mounted on buoys at sea, where they could monitor levels continuously, removing the need to collect and transport sea water that can invalidate measurements. Devices to test for nitrates or other nutrients in soil water could be distributed around a farmer's land, helping to control pollution and perhaps increase crop yields.
Confining molecules in liquid or gas in smaller spaces allows them to react very quickly. That makes reactions safer. "When you do explosive reactions in microfluidic tubes they aren't explosive. You can contain them a lot more," says Barrow. It also becomes possible to experiment with new chemistries.
"But of course," he acknowledges, "you don't get much." The problem of how to get more is one that Colin Ramshaw is tackling at the University of Newcastle. Ramshaw is applying the principles of microfluidic technology to the concept of a benchtop factory. Lab-on-a-chip systems for making fine chemicals, for example, could be stamped onto a sheet of plastic, "like wallpaper", says Ramshaw, cut into leaves and bonded into a block the size of a shoebox. Although each individual device might produce only a few drops per minute, if you multiply that by a thousand, he says, you could produce a few hundred tonnes per year - a significant amount for a pharmaceutical product.
Not only can microfluidic technology be applied to analytical systems and to small process plants, says Barrow, but microfluidic engines are being designed to steer satellites the size of pumpkins in space. And tiny fluidic manifolds could be used instead of fans to keep computer chips from overheating.
The challenges in constructing a working lab-on-a-chip are many. Samples need to be pretreated to filter out particles, for example in blood or pond water, or to concentrate the sample sufficiently to get a useful result. Getting fluids down little tubes is hard, but valves and pumps are unreliable. Electrical forces offer an alternative for shunting fluids about. Because chips consist of two and sometimes more layers, a good seal is essential, but inert materials such as Teflon will not stick. "The world is designed to give us problems," says Barrow.
It is still early days. Although microfluidic systems for analysing DNA are available commercially, other commercial applications are still several years down the road. At the moment, the barriers are technical. But even when the kinks are ironed out, Manz muses, in the end it is not the technology itself but its capacity to solve problems that will change the way people think. "We have to be better than 150 years of chemistry and engineering," he says.