A new wave of microfluidic devices - The Industrial Physicist

Microfluidics refers to a set of technologies that control the flow of minute amounts of liquids or gases—typically measured in nano- and picoliters — in a miniaturized system. “Unlike microelectronics, in which the current emphasis is on reducing the size of transistors, microfluidics is focusing on making more complex systems of channels with more sophisticated fluidhandling capabilities,” says George Whitesides, Mallinckrodt Professor of Chemistry and Chemical Biology at Harvard University (Figures 4, 5, and 7). Although microand macrofluidics systems require similar components— including pumps, valves, mixers, filters, and separators — the small size of microchannels causes their flow to behave differently (see “Micro versus macro”, table below). Hence, microscale components require new fabrication methods.

Microfluidics devices, first developed in the early 1990s, were initially fabricated in silicon and glass using photolithography and etching techniques adapted from the microelectronics industry, which are precise but expensive and inflexible. The trend recently has moved toward the application of soft lithography—fabrication methods based on printing and molding organic materials (see The Industrial Physicist, August/September 2002)—to build microdevices.

Microfluidics’ appeal lies in the fact that the microchips require only a small amount of sample and reagent for each process—only a few tens or hundreds of nanoliters compared with the 100 ml required by existing plate assays. Microscale reactions also occur much faster because of the unique physics of small fluid volumes, and microfluidics technologies are easily automated to do routine assay and sample preparation on standardized chips with little human intervention. Such chips hold the promise of combining multiple functions on a single chip, including purification, labeling, reaction, separation, and detection. Microfluidics would guide the sample automatically from one station to another on the chip.

Applications
The most mature application of microfluidics technology is ink-jet printing, which uses orifices less than 100 µm in diameter to generate drops of ink. Today, inkjet printing is moving out of the office and into biotechnology to deliver reagents to microscopic reactors and deposit DNA into arrays on the surface of biochips.

Most microfluidics devices today use electrokinetic and pressure methods to move small amounts of fluid around a microchip. However, although such techniques are useful for certain niche markets, they lack the flexibility required for universal application—a key selling point to scientists seeking to reduce capital investment. Fluidigm Corp. (San Francisco, CA) believes that multilayer soft lithography (MSL) provides a solution. Developed in the late 1990s by Caltech biophysicist Stephen Quake, MSL enables the company to fabricate three-dimensional structures from multiple layers of soft elastomer by imprinting each layer and then binding them together to form the pumps, valves, and channels integral to the chip. This enables a single chip to serve many functions, including sample preparation, manipulation of live cells, perfusion of reagents, and analyte detection. That kind of flexibility gives any microfluidics technology a competitive edge in emerging markets.

The Bioanalyzer 2100, a collaboration between Caliper and Agilent Technologies (Palo Alto, CA), targets individual researchers working with small samples of DNA, RNA, proteins, and cells. Caliper’s AMS-90 line is geared toward higher-throughput applications, such as the quality control of DNA sorting, during which hundreds of samples are run at a time.

Scientists at Sandia National Laboratories (Livermore, CA) are developing ChemLab, a portable, handheld chemical-analysis system for homeland security, defense, and environmental and medical applications. Currently a prototype, ChemLab can detect chemicalwarfare agents and proteins, as well as biotoxins such as ricin, staphylococcal enterotoxin B, and botulinum toxin. It can also identify viruses and bacteria using protein fingerprinting. Sandia expects to commercialize the system within the next two years.

Physicist David Grier of the University of Chicago has developed a new technique known as holographic optical tweezers. In this approach, a laser beam is sent into a hologram and divided into myriad subbeams, which can independently suspend and manipulate many tiny particles for transportation, mixing, or reacting. Using this technique, ensembles of microspheres can be moved into patterns and set to spinning by the holographically sculpted light fields (Figure 1).

When applied to fluid samples of biomolecules, this holographic multiplexing produces what Grier has dubbed “optical fractionation,” an optical equivalent of gel electrophoresis, the workhorse of the biotechnology industry. It enables electric fields to differentially drive and separate macromolecules. However, Grier’s approach is more flexible than gel electrophoresis because it does not require a viscous gel.

By merely tweaking the computer- generated hologram or the laser wavelength, the user can sort objects ranging from the 10-nm (important for viruses) up to the 100-µm scale.

Other advantages of optical fractionation include continuous rather than batchmode operation, and optimal performance simply requires varying the laser power, laser wavelength, and/or the geometry of the trap arrays. Thus, the same apparatus can sort samples on the basis of size, surface charge, dielectric constant, magnetic permeability, and shape. The technique “ offers exponential sensitivity for fractionation by size,” says Grier. “This is unprecedented. Other techniques offer linear, quadratic, or slightly better selectivity, whereas optical fractionation is qualitatively more selective.”

In April, Micronics Inc., (Redmond, WA), patented a credit-card-sized microfluidics device to perform immunoassays, which it achieves by exploiting molecular binding reactions and differential diffusion rates in microchannels (Figure 6). The microassay is applicable to a wide range of analytes, including therapeutic drugs, molecular biological markers, and environmental contaminants. And University of Michigan researchers, led by Gary Smith of the department of obstetrics and gynecology, have developed a prototype microfluidics device that rapidly and automatically sorts sperm and isolates the most viable swimmers for injection into an egg. Their ultimate goal is to create a self-contained, at-home test for men to screen for infertility or to judge the success of vasectomy or vasectomy-reversal procedures. The device sorts sperm on the basis of their speed.

Surface Logix (Cambridge, MA) targets its microfluidics systems to advanced cell-based assays for drug research. It is developing multiplexed cell-based assays that provide precise control over cell location, culture condition, and reagent delivery. Benefits include the ability to isolate, manipulate, and monitor individual cells; real-time analysis of physiological changes; and compatibility with standard assay formats and detection devices.

With Fluidigm up and running, Quake has turned to developing another microfluidics device. This one allows the careful metering of reagents to facilitate protein crystallization under a variety of conditions, including pH, viscosity, surface tension, or various solvents.

Micro versus Macro

Microfluidics hardware requires different methods of construction and design. Conventional devices cannot simply be scaled down because the basic physics changes at the microscale. When the dimensions of a device or system become small enough, particles suspended in a fluid become comparable in size to the device itself, which dramatically alters system behavior.

Although the fluid properties remain the same at the microscale, surface tension, viscosity, and electrical charges can become dominant forces on a fluid because the surface-to-volume ratio is much greater than for macroscale systems. Also, no one fully understands how heat transfer and mass transfer function at the microscale, and what effect they might have on the device.

“At large scales, the inertia of the fluid is important, whereas at smaller scales, it is not important at all,” explains Brian Kirby, a staff scientist at Sandia National Laboratories. “ At the macroscale, the notion of applying a voltage to a fluid and expecting it to have an impact is ridiculous; we use mechanical pumps at the macroscale that exploit inertia. When you get down to length scales on the order of 10 µm, applying voltages to fluids encased in a channel can be used to manipulate them.”

Another area where microfluidics could prove valuable is structural genomics (Figure 3). This market sector will grow at a compound annual rate of 32% to an estimated $1.4 billion over the next five years, according to a 2002 study by Front Line Strategic Consulting, Inc. (San Mateo, CA). Chow reports that Caliper is currently developing a prototype microfluidics system for genomics analysis, with the initial focus on single-nucleotide polymorphism analysis, which can detect the substitution of one amino acid for another in a protein. “By adding the correct reagent to a channel containing a genomic DNA sample, then selectively amplifying a certain gene segment, we can study the characteristics of that particular segment of interest.” Other lab-on-a-chip companies exploring this area include Nanogen (San Diego, CA) and Orchid Biosciences (Princeton, NJ).

Chemistry
Kevin Killeen, project manager of microfluidics and sensors at Agilent, contends that the field has barely tapped the possibilities of microfluidics flow-through processes in chemistry. “Thus far, we have only been scratching the surface by emulating what is currently done in batch mode,” he says. “But microfluidics means we can prepare samples for chemical separation and detection in a different way than a lab bench chemist would have (see 2003 Industrial Physics Forum, page 30, “Polymer microfluidics for chemical analysis,” Kevin Killeen).

The advent of the nanotechnology revolution opens even more opportunities for innovation. For example, Caliper’s Chow foresees combining microfluidics processing with emerging devices such as nanotechnologybased sensors to create new types of assays with new functions that have no analogy in the macroscopic world. “ There are two ways to think about miniaturization,” she says. “First, you can take existing technology and simply make it smaller. But it is much more powerful to create something that can only be done in the miniaturized world—this will be a truly enabling technology.”

“It might not be the exact equivalent of what happened in the integrated circuit (IC) industry,” says Killeen of microfluidics’ future. “But I do think it will, in a similar way, do for chemistry what the IC industry did for electronics. Microfluidics is going to revolutionize the way things are done. It is going to make processes more efficient. It’s going to miniaturize chemistry and make possible chemical reactions that can’t be done in batch mode today because they are either too unstable or the chemicals are too precious. It is hard to say exactly what will happen, but I know it will touch virtually every aspect of our lives.”