Research Highlights


Ultra-fast separations of large DNA molecules

Conventional methods for separating DNA molecules of different sizes are based on agarose gel electrophoresis. Unfortunately, long DNA molecules (>20 kbp) exhibit complex behaviour when moving through gel or polymer matrices, resulting in reduced separation efficiencies. Consequently, pulsed field gel electrophoresis (PFGE) is normally used to efficiently separate long chain DNA molecules. Although successful, PFGE methods suffer from excessively long run times and are normally limited to fragment lengths below 10 Mbp. Robert Austin and co-workers at Princeton and Cambridge Universities have addressed this problem by implementing PFGE within microfabricated hexagonal arrays. Devices fabricated from quartz using standard photolithographic and reactive ion etching techniques result in the formation of arrays of 2 μm pillars with 2 μm spacings arranged in a hexagonal lattice. By applying a pulsed field along alternating axes of the array (separated by 120°) DNA molecules exhibit a net motion along the bisector of the axes, with migration speeds that depend on their length (see Fig. 1). In addition, an entropic barrier is used to pre-concentrate DNA molecules at the array entrance and therefore reduce any band broadening. Initial experiments demonstrate efficient PFGE separations of T4 (168.9 kbp) and λ (48.5 kbp) DNA in times as low as 10 s. Furthermore, the authors note that even faster run times should be achievable by the application of higher electric fields for shorter periods, and that the technology can be extended to the separation of polymers of all lengths.
Microscopic view of the hexagonal array. Cartoon DNA molecules illustrate the motion of individual molecules of different lengths. Shorter molecules move farther in the array because subsequent to reorientation along the axis of the field they move in an unhindered straight line for the duration
of the pulse. Longer molecules, on the other hand, spend most of the pulse period retracing their paths. (Adapted with permission. Copyright 2001, The American Chemical Society.)
Fig. 1 Microscopic view of the hexagonal array. Cartoon DNA molecules illustrate the motion of individual molecules of different lengths. Shorter molecules move farther in the array because subsequent to reorientation along the axis of the field they move in an unhindered straight line for the duration of the pulse. Longer molecules, on the other hand, spend most of the pulse period retracing their paths. (Adapted with permission. Copyright 2001, The American Chemical Society.)
Analytical Chemistry, 2001, 73, 6053

Ultrasonic mixer

The efficient mixing together of two or more reagent streams is a primary issue of concern in the development of highly integrated chip-based analysis systems. Since most microfluidic systems have micron-sized features and incorporate low volumetric flow rates, associated Reynolds numbers are extremely small. Very simply, this means that flow is normally laminar, turbulence is rarely observed and mixing is dominated by diffusion. To date, the vast majority of microfabricated mixing devices act to enhance interdiffusion between adjacent fluid streams in a passive manner. An alternative approach is through the use of an active micromixing device (where external energy is transferred into the device to aid mixing). Ryutaro Maeda and colleagues at the National Institute of Advanced Industrial Science and Technology in Japan have recently evaluated the design and operation of an active ultrasonic micromixer for use in microfluidic systems. The mixer is fabricated from glass and silicon and ultrasonic vibration is induced by a piezoelectric lead–zirconate–titanate ceramic adhered to a thin-film silicon diaphragm. The authors demonstrate that turbulent mixing can be generated in a 20 μL mixing chamber by ultrasonic vibration. Furthermore, efficient mixing can be achieved at a variety of flow rates and is unaffected by the presence of gas-bubbles within the mixing chamber. Nevertheless, mixer performance is compromised by excessive heat generation during energy input and the fact that ultrasound at kHz frequencies can be extremely harmful to many biological samples. These issues limit potential applications for the technology, although the authors are currently addressing improvements in both areas.Sensors and Actuators A, 2001, 93, 266

Chemiluminescence detection in 3-D microarrays

Fluorescence detection schemes are almost exclusively used to image binding events within two-dimensional solid support microarray platforms. They afford both exquisite sensitivity and high spatial resolution but require the use of an external excitation source. Chemiluminescence (CL) imaging techniques provide an alternative approach to analyte detection in array based systems. Attractive features of CL methods include inherently low background noise, high sensitivity, high specificity and reduced instrumental complexity. Nevertheless, CL methods have rarely been used to image arrays. This is primarily due to reduced sensitivity and spatial resolution when compared to fluorescence techniques. Since the light emitting species are normally not surface-associated, they may diffuse across the array surface and cause emission in a different part of the array. Brady Cheek and co-workers at MetriGenix Inc., Gaithersburg, have addressed this problem by modifying the structure of the array format. The authors report the use of a microarray platform in which molecular interactions occur within a three-dimensional volume of ordered microchannels rather than on two-dimensional surfaces. By localizing light-emitting species within the microchannels (that connect top and bottom surfaces of the array) high-sensitivity CL detection can be achieved in a microarray format. Due to the additional surface area provided by the microchannels signal improvements of two orders of magnitude are obtained compared to flat glass arrays. In addition, the authors demonstrate two-channel CL measurements that incorporate multiple enzyme-substrate combinations.

Analytical Chemistry, 2001, 73, 5777


‘Smart’ gas sensing chip

Much current work in the area of chemical gas sensing is focussed on either the design of highly selective biological recognition elements or the fabrication of arrays of partially selective sensors for multi-component analysis. Developments in micromachining methods have also facilitated the development of both planar and MEMS based sensors. Recently, Andreas Hierlemann and colleagues at ETH, Zurich, combined a number of these developments in fabricating an integrated chemical microsensor incorporating three distinct transducers (mass sensitive, calorimetric and capacitive) to detect volatile organic compounds. Integration of both electronic and micromechanical components on one device affords both control and monitoring of sensor functions and conditioning of transduced signals. Specifically, each transducer relies on sensitive poly(etherurethane) layers for analyte detection. The capacitive sensor monitors changes in the dielectric constant of the polymer upon exposure to analyte species; changes in the oscillation frequency of a micromachined cantilever coated in polymer upon analyte adsorption provide a measurement of mass; and the thermoelectric calorimeter detects enthalpy changes on adsorption or desorption of analyte molecules in the polymer film. Efficient functioning of all transducers is achieved when the sensor is exposed to ethanol and toluene loaded gas samples, and the authors note that analyte identification can be easily improved by creating arrays of sensor chips that incorporate different polymer layers.

Nature, 2001, 414, 293


Microbubbles

Although gas bubbles are the scourge of many microfluidic systems, they can often be put to good use. For example single-bubble micro-pumps are based on the principle that under certain conditions, thermally generated bubbles can rapidly and efficiently move fluids. Furthermore, micron sized gas bubbles are used in a diversity of medical applications, including ultrasound, tumor destruction and targeted drug delivery. Alfonso Gañán-Calvo and José Gordillo from the University of Seville report, in the December 31st issue of Physical Review Letters, a simple microfluidic phenomenon which allows the mass production of micron sized gas bubbles with highly monodisperse and controllable diameters. By continuously supplying a gas stream from a capillary positioned close to and slightly upstream of an orifice (30–500 μm diameter) through which liquid is forced, cusp-like attached gas bubbles form. The gas jet is then focused through the orifice by the surrounding liquid stream and breaks up into a stream of microbubbles. These microbubbles are produced at constant frequency and with uniform size. Depending on the density, viscosity and surface tension of the surrounding liquid medium bubbles with diameters ranging from 5 to 120 μm can be produced.
Micrograph of the gas microsensor system chip (7 × 7 mm). The different components include: 1, flip-chip frame; 2, reference capacitor; 3, sensing capacitor; 4, calorimetric sensor and reference; 5, temperature sensor; 6, mass-sensitive resonant cantilever; and 7, digital interface. (Adapted with permission. Copyright 2001, Nature Publishing Group.)
Fig. 2 Micrograph of the gas microsensor system chip (7 × 7 mm). The different components include: 1, flip-chip frame; 2, reference capacitor; 3, sensing capacitor; 4, calorimetric sensor and reference; 5, temperature sensor; 6, mass-sensitive resonant cantilever; and 7, digital interface. (Adapted with permission. Copyright 2001, Nature Publishing Group.)

Physical Review Letters, 2001, 87, 274501


Design of multiple reaction systems

Over the past decade there has been significant interest in the development of microfabricated DNA analysis devices. Most of this research effort has been motivated by enhanced analytical performance, superior component integration, increased throughput and improved automation. Recent studies in the area have demonstrated a clear focus on functional integration of analytical components within monolithic devices. For many genotyping applications an assay may often involve a number of reactions requiring different thermal control. Consequently, the ability to perform multiple thermal reactions within a monolithic structure is highly desirable. To allow for efficient integration and operation of chip function within monolithic structures heating zones should be highly localized (or isolated) so as to minimize heat gradients across the substrate. Mark Burns and co-workers at the University of Michigan, Ann Arbor, have presented a new methodology for the design and manufacture of low-power, small-size microfabricated multiple reaction systems using the concept of heat integration. The approach is used to minimize the overall power consumption within a microdevice by optimizing component placement (based on factors such as substrate material, operating temperatures and fabrication complexity). Initial experiments were used to design and test reaction systems for DNA analysis. Resulting devices incorporated heaters positioned adjacent to, rather than directly under, reaction chambers (greatly simplifying the fabrication process) and were successful in performing DNA restriction digest reactions. The general methods outlined by the authors should prove extremely useful when fabricating highly integrated analysis systems for genetic analysis.
Conceptual schematic of a microchannel glass array. The chip is composed of an ordered array of microchannels that connect the planar surfaces. Analyte-specific reagents,
or probes, are deposited on the chip in spots. Each spot incorporates several individual channels. (Adapted with permission. Copyright 2001, The American Chemical Society.)
Fig. 3 Conceptual schematic of a microchannel glass array. The chip is composed of an ordered array of microchannels that connect the planar surfaces. Analyte-specific reagents, or probes, are deposited on the chip in spots. Each spot incorporates several individual channels. (Adapted with permission. Copyright 2001, The American Chemical Society.)
Sensors & Actuators A, 2002, 95, 250

Hydrodynamic chromatography on-chip

Liquid chromatography (LC) is still the most widely used of all analytical separation techniques, with annual sales of HPLC equipment around the billion-dollar mark. In conventional LC the stationary phase is normally a porous gel packed into the column or attached directly to the walls of an open channel. The difficulties in transferring LC methods to chip-based formats have traditionally been associated with the introduction of packing materials into the microfluidic network. To allow efficient open-channel LC separations a high area-to-volume ratio channel should ideally be combined with a small injection volume. Albert van den Berg and colleagues at the MESE Research Institute, University of Twente, have recently reported a method for performing hydrodynamic chromatography on chip devices where separation is based purely on geometrical characteristics. In hydrodynamic chromatography size discrimination is based on the fact that for parabolic flow in a channel, larger molecules are unable to approach the channel walls as closely as smaller molecules, and thus have large velocities. In their first generation device, three parallel, high-aspect ratio injection channels are used to deliver sample into a low-aspect ratio separation channel. Chromatographic performance was then assessed through the separation of fluorescein and fluorescent polymer beads (26 nm diameter). Future applications of the technology will focus on the analysis of large molecules in the 10 kDa–10 MDa range.

Sensors & Actuators B, 2002, 82, 111

Andrew J. de Mello

This journal is © The Royal Society of Chemistry 2002
Click here to see how this site uses Cookies. View our privacy policy here.