ECE: Electrical & Computer Engineering
ECE News

Shrinking Lab Equipment onto a Chip

Print Version

image of article

.pdf Print Version
204 KB

More Information


For more information, visit the The Virginia Tech Microelectromechanical Systems Laboratory (VT MEMS Lab) .

Etching process may impact engineering educatoin


The single-pass, multi-depth etching process developed by Masoud Agah's research team my change how university students in many fields study MEMs.

Read More

GC Matrix System tackles faster analysis and lower cost with micro design.

Masoud Agah

Masoud Agah has received an NSF CAREER Award to develop micro gas chromatographs

GC Matrix Screenshot

Masoud Agah has developed a microchip etching process and a multi-column architecture that he hopes to use to push gas chromatography (GC) out of its traditional laboratory setting and into the field. He wants to shrink GC units from their present table-top size to a credit-card size so they can be carried easily for instant analysis of hundreds of elements.

Gas chromatography is the primary technique used in scientific, medical and industrial laboratories for separating and analyzing volatile compounds. Portable microGC units would have many uses, ranging from doctors using breath analysis for diagnosing disease, to environmental monitoring of air and water contaminants to testing for chemical warfare agents.

Gas chromatography separates chemicals from a compound sample that has been gasified. Using an inert carrier gas, the sample is propelled through a column that has coated interior walls. Molecules from the sample can be absorbed by the coating, which affects the rate at which different chemicals exit the column. Since different types of molecules exit the column at different times, thermal conductivity detectors at the end of the column can identify the chemicals in the sample. In practice, several runs are used to confirm results and the time spent in the column can sometimes run more than an hour.

"It can often take four days to send a sample to the lab and get firm results using gas chromatography or mass spectrometry," Agah says. "There are many applications that can't wait that long. So the separation and detection times become critical."

Micro Gas Chromatograph

A number of research teams worldwide are working to develop fast, accurate microGC units, but Agah's concept focuses on low-cost production methods and a novel, multi-column architecture. His etching process may also help change how students in many fields study microsystems.

His team's preliminary results are encouraging and the National Science Foundation (NSF) has awarded him a $400,000 CAREER Award to support the effort. CAREER grants are NSF's most prestigious awards for junior faculty members who are considered likely to become academic leaders.

Agah's plan involves at least two critical advances &mdash employing an array of microtubes instead of the single tube that is conventionally used in gas chromatography and developing a single-pass etching process for multiple channel depths.

Matrix approach

Most microGC systems use a similar structure to the large, table-top systems, employing a single, long column. While tabletop systems use columns about 10 meters long, Agah describes the current trend in microGC systems is to use 3-meter-long columns. "These systems cannot separate more than 20-30 compounds and have separation times measured in several minutes," he explains. Faster analysis has been achieved by reducing the column length to less than 50cm and using rapid temperature programming. "These systems have separated multi-component mixtures in a few seconds. But there is a tradeoff between speed and resolution. They can only separate fewer than 10 compounds. They can go fast and few, but not complex," he adds.

Agah's goal is to separate complex mixtures &ndash more than 100 compounds &mdash in a few seconds. He is tackling the speed/resolution tradeoff by using a novel architecture he calls GC Matrix.

Agah's goal is to separate complex mixtures &ndash more than 100 compounds &mdash in a few seconds. He is tackling the speed/resolution tradeoff by using a novel architecture he calls GC Matrix. The GC Matrix will consist of short microGC columns less than 30 centimeters long, with on-chip heaters, temperature sensors, micro flow meters, and thermal conductivity detectors. The columns will be integrated into a matrix configuration, with each row responsible for certain boiling points. Within each row, different columns will have different polarities and even different coatings.

"At the end of each row, we'll be spitting out results," he describes. "That's how we can get speed. We have independent rows, but also get the needed complexity because we'll have different polarities in each row." The performance of each mGC column will still depend on a tradeoff between speed and resolution, he says, "but the GC Matrix will enable us to tune these attributes independently on a system level." Agah already has developed columns that can separate a limited number of volatile compounds in less than 10 seconds.

First-ever integration

Conventional Gas ChromatographyGas chromatgraphy using GC Matrix

Masoud Agah has received an NSF CAREER Award to develop micro gas chromatographs

A GC Matrix system would integrate all the sensors on a column for the first time. "We're talking about a very compact microchip with integrated pressure, flow, thermal conductivity, and temperature sensors, with columns and heaters."

"For easy commercialization, we want to develop technology that is simple, reliable, low-cost and robust," he explains. "With this kind of compactness, it's important to minimize the fabrication complexity. As you increase production steps, your yield goes down, because every additional step introduces more opportunity for problems, such as contamination."

Instead of the typical silicon glass, Agah's team is using a silicon base employing a new fabrication technology they call the CMOS-Compatible Predictable 3D Buried-Channel Process. "For our integrated columns to work, we will need to etch to four or five different depths from 10 to 120 microns." He explains that the team wants to fabricate the system using a single pass for etching and a single pass for deposition with no bonding.

The team has eliminated a bonding process for both power consumption and production reasons. "If we use bonding, we will increase the total mass of the chip," Agah says. "When we increase the mass of a chip, it needs more power to heat it up. By getting rid of bonding and its thick pyrex substrate, we reduce mass. Then our columns can be floating structures connected with thin dielectric layers. We can heat it up rapidly, because it is isolated and the power consumption goes down. We cannot do this with silicon glass. We'll be able to keep the processing temperature during fabrication less than 300°C, which makes it compatible with conventional IC manufacturing."

Single-pass etching process

Agah at Work

The team has recently tested a new method of etching silicon to different depths in one pass by maximizing a reactive ion etching (RIE) lag. "We have concluded that with a structure that has wider openings, the etch is deeper. Based on the holes in the array and the dimensions of the holes and spacing between them, we can control the depth and width of these channels."

Since the technology allows different depths and widths, even a single channel can have varying dimensions, according to Agah. "We can even make nozzles with this," he says. "This is just not achievable in pyrex."

The single-pass etching technique is not only useful for Agah's microGC system, but holds promise for other microfluidic uses in life sciences and chemical analysis and in engineering education (see sidebar).

Although he has demonstrated the new, single-pass, multi-depth etching process and modeled the GC Matrix, there are many challenges facing Agah's team. Several hurdles include tuning Virginia Tech's existing equipment to accommodate the required sensitivity or even acquiring new deposition capability.

"The beauty of research," he says, "is that we don't know what will happen until we do it."

Other hurdles are technical. Once the etching technology is perfected, the team will develop the microcolumn with all the microsensors on it. "To do that," Agah says, "we first must fabricate the individual components to make sure we can get the sensitivity we need for each component, then integrate it on a chip. That seems routine, but a least one part might become troublesome &mdash coating the columns. We can use typical coating techniques or use nanotechnology.

"Then, we'll need to make the 30 cm columns coated with different polymers that separate different components. Since all of these sensors are integrated, will the coating affect the performance? Will there be any deficiency in sensitivity and detectability?" he wonders.

"The beauty of research," he says, "is that we don't know what will happen until we do it. What makes it fun is that we see so many challenges in front of us, and we know there will be more as we move forward."

For more information: www.ece.vt.edu/mems.

Etching process may impact engineering educatoin

The single-pass, multi-depth etching process developed by Masoud Agah’s research team may change how students in many fields study microelectromechanical systems (MEMS). As microsystems become more common, students in many disciplines, including ECE, mechanical engineering, engineering science and mechanics, chemical engineering, and biosciences are being introduced to the technology. As part of Agah’s CAREER project, he will be incorporating the process in a laboratory course for students from these disciplines.

The new etching process is expected to make fabrication inexpensive and speedy enough for classroom use in real applications. “We’ll be able to give students a deep, hands-on education in microsystems from design to fabrication to applications. Equally important, we’ll be able to have them work in truly interdisciplinary teams while in school,” he says.

“With this etching technology, we can make many different systems, such as pressure sensors, micro mixers, particle (cell) separators, gas analyzers and more,” he adds. In the course, each multidisciplinary team will design a device based on the technology. “They can design whatever they are interested in. It can be a sensor. It can be a microfluidic device.”

Once they determine the overall design and the depth and width of their channels, they will have access to the software Agah’s team is developing for the gas chromatography project to determine the layout. “We’ll be able to put all the designs onto one photo mask and have them all fabricated simultaneously using the single-pass process. Everybody will get trained on the different tools and fabrication processes and will contribute to the fabrication of class microchips.”

With the chips fabricated, the teams will be able to analyze and test their systems, and determine what design changes would be needed.

Teams of up to three students will work together for the design, with each student taking the responsibility of his or her own discipline. The students will learn to communicate about their discipline at the level of students from other fields. “MEMS is by nature interdisciplinary,” he says. “Learning to communicate to professionals outside their field and learning to work on a multidisciplinary team will make our students more valuable to employers.”Micron Technology is helping develop the course. “Getting the perspective from an IC company is very valuable,” Agah says. “Micron will help ensure that our students learn skills that are state-of-the-art.”