A quick walk around any electronics shop will show you that bigger is no longer better. Mobile phones are shrinking alarmingly, and advances in computing power has seen palmtop computers go from glorified battery-powered filofaxes to sophisticated devices capable of guiding you to the nearest bar selling obscure spirits. Stuart Nathan reports.
In the laboratory field, miniaturisation is set to have even more dramatic effects. Advances in microscopic fabrication techniques and new understandings of the behaviour of liquids in confined spaces is leading the development of microchip-sized devices which incorporate the functions of an entire analysis or synthesis laboratory into a space less than an inch square.
Lab-on-a-chip devices do everything a room-sized laboratory can do in a space the size of a fingernail. Tiny amounts of liquids sample sizes can be as small as a picolitre, a millionth of the volume of the average adrop' flow through channels around 50microns across and 10microns deep, thinner than a human hair, etched into a layer of quartz, glass or polymer. On their way around the chip, the molecules in the sample zip past reservoir chambers which add reagents and over detectors that measure the rate of reactions. Chip-mounted devices can diagnose diseases from blood samples; synthesise arrays of novel compounds; and separate mixtures of compounds. And the list of applications seems limited only by the developers' imaginations.
Speed is the driving force for development of lab-on-a-chip devices. As pharmaceutical competition becomes ever more intense, companies are looking for ways to screen compounds faster, to make sure the most effective compounds are developed and reach the market in the fastest possible time. Lab-on-a-chip devices score heavily because they use tiny amounts of substances and provide results fast.
It is the field of microfluidics the study of moving minute amounts of liquids in confined spaces which provided the micro-labs with their equivalent of pumps. The most common way of pumping the liquids is by a technique known as electo-osmosis, which exploits the property of many liquids to move under the influence of an electric field. A technique developed by California-based microfluidics specialist Caliper Technologies the first company to commercialise labs-on-chips generates these fields with electrodes attached to computer-controlled power supplies and placed in the sample reservoirs at either end of the chips' channels. The flow rates, typically of about a millilitre per second, are in direct proportion to the strength of the electrical field, and therefore can be controlled very precisely.
The technique works by generating very slight charges in the liquid. At the channel walls, the charge produced by the electrodes induces a small opposite charge in the adjacent molecules of the liquid, which forces them to aline up'. Applying a voltage than forces the liquid to move along the channel en masse. The leading edge of the liquid is sharp, unlike the ablurring' that occurs when pumping is done by pressure, making the labs-on-chips ideal for applications involving separating mixtures, or processing different compounds in batches (such as synthesising libraries of compounds).
The pumps can be surprisingly powerful. At Sandia National Laboratory in San Francisco, Christopher Bailey and Don Arnold have developed an electokinetic pump that can generate pressures over 9000psi. "We're getting orders and orders of magnitude more pressure than other systems,' comments Arnold. aIt's limited by the connector I can generate pressures on a chip of 10000psi, but blow the connector. We've generated pressures to the point where the glass breaks from mechanical failure.“
This is especially astounding considering the size of the components. In Fig. 1, a surface acoustic wave sensor array, a preconcentrator that collects chemical vapours for gas-phase analysis and a gas chromatograph, all developed at Sandia, sit neatly inside a peapod. The entire device, a liquid sample analyser, includes power supplies (running off batteries designed for compact cameras) and solid-state relays to direct energy flows, and is about the size of a thick paperback book. The developers expect to shrink this further, to around the size of a palmtop computer.
The channels etched into the chips are so small that there is virtually no turbulence in the flow. Moreover, the control exerted by the fields means that there is no need for any other types of pump or valve; in other words, the chips have no moving parts.
The detectors are equally ingenious. For example, the Sandia gas chromatograph uses acoustic wave sensors to detect components of a mixture. These are coated in compounds which adsorb the components coming off the column, leading to a shift in acoustic wave frequency. The different degrees of absorption of each molecule give a characteristic fingerprint which can be detected by an on-board computer. The size of the device works in the detectors' favour here: the acoustic wave sensors become more sensitive as the get smaller. Other detectors developed by the Sandia labs are based on lasers and photodiodes to detect light absorption by the sample components. These are built in to the chip.
At Caliper, researchers are working on a chips which can perform experiments as well as analyses. One chip is capable of a complete biochemical experiment, involving the competitive binding of biotin and the protein streptavidin at various concentrations of biotin. It does this by mixing a buffered solution of biotin with another solution containing biotin labelled with the fluorescent dye fluorescein, pumped from a separate reservoir. This solution is then mixed with streptavidin, pumped from yet another point on the chip. If the labelled biotin binds to the protein, its fluorescence is quenched; if the unlabelled molecule binds, the fluorescence stays the same. As the concentration of unlabelled biotin increases, more labelled molecules remain unbound. A detector measures the fluorescence signal, and gives a measure of how binding is affected by concentration. Systems like this can be adapted for a variety of biochemical systems and experiments; for example, a very similar system could screen drugs designed to block binding of certain molecules onto receptor sites.
But the use of lab-on-a-chip systems is not limited to research. Because the systems are so portable, they are prime material for field devices.
Potential applications include systems to detect the presence of explosives and chemical weapons. Already being developed at Sandia, these devices could also be used to detect landmines. It is a small step from this to packages to detect and analyse chemical and biological hazards, pollutants and emissions. In the medical field, lab-on-a-chip devices could give doctors a powerful tool to diagnose patients at their bedside, and even check the effectiveness of their drug therapies. And at chemical plants, they could be used to optimise industrial processes by checking the quality of products in real time.
However, there are still challenges. Caliper's researchers have found that, at such a small scale, the material of the chip can itself have significant effects on reactions; they are currently designing methods to suppress these effects. Teams are also working on methods of sample preparation and handling, and speed of data handling.
