Small Sensors Show Big Potential

December 2006
By Henry S. Kenyon

Researchers are developing nanoscale sensors to detect chemical and biological agents. Embedded on a silicon microchip, this prototype carbon nanotube sensor (inset) can be set to detect specific types of molecules. Microchips can be designed to hold hundreds of nanotubes, creating highly sensitive miniature detectors.
Arrays of sensitive microscopic detectors offer new ways to identify harmful gases and liquids quickly.

Researchers have developed nanoscale sensors capable of detecting trace amounts of chemical and biological agents. The tiny devices can be placed on microchips, creating the potential for highly accurate networked sensors embedded in a variety of equipment and systems.

Conducted by Arizona State University (ASU), Tempe, Arizona, and Motorola’s Embedded Systems Research Laboratories, Tempe, the work focuses on using single-walled carbon nanotubes (SWNTs) to detect chemical agents. According to Dr. Nongjian Tao, a professor in ASU’s department of electrical engineering who is heading the research, SWNTs are known for their electronic properties, which makes them good material for microchips and transistors.

Tao explains that chemical and biological sensors typically operate by converting chemical reactions into electrical signals. These reactions occur when molecules bind to the surface of the sensor. The engineering challenge lies in converting these molecular-level events into a useful electronic signal.

An SWNT consists of a sheet of carbon atoms rolled into a tube. Every atom in this structure is exposed to the environment, which allows the tube to detect any changes such as chemicals binding to its surface. This high sensitivity makes carbon nanotubes a popular choice for sensors, Tao states. When used as a sensor, the inside of the SWNT is not important because the chemical interactions take place on the outer surface of the tube. The tube’s small size is useful for embedding high-density arrays on small platforms such as microchips. SWNT sensitivity provides an advantage because it allows for rapid signal response.

Developing the small SWNT structures into sensors is difficult. Modifying SWNTs to attract specific chemicals strongly will change the tubes’ electronic properties, ruining their ability to work as sensors. Tao’s solution attaches peptides to the surface of the tubes because peptides do not significantly alter the tubes’ electronic properties and are chemically stable.

Peptides are molecules consisting of roughly 20 amino acids that form the building blocks of proteins, natural biological systems that can recognize a variety of substances. By changing the sequencing of the amino acids, scientists can make different types of peptides. “That’s the reason antibodies can recognize antigens. The human body can make antibodies by tuning the sequence of the amino acids,” Tao says.

Researchers use this building block approach to create different peptides that are attached to an array of SWNTs embedded on a microchip. The various peptides are modified to detect specific chemicals and biological agents. Tao explains that this process allows each nanotube on the chip to distinguish a particular substance. “You can tune the functionality of the device by tuning the sequence and the length of the peptides,” he observes.

ASU researchers demonstrated the proof of concept by setting the tubes to detect several simple chemicals and heavy metal ions. Tao notes that the heavy metal ions were chosen because the U.S. Food and Drug Administration and the U.S. Environmental Protection Agency had provided funding to develop methods of detecting these ions in drinking water. He adds that SWNTs can be modulated to detect a range of substances from toxic gases in the air to chemicals in water.  

In addition to carbon nanotubes, Tao’s team also studies the use of conducting polymers as detectors. He explains that the concept can be used in a manner similar to SWNTs. Because these polymers are naturally conductive, the electric current flowing through a polymer-equipped device can be measured to detect any binding events taking place on the device’s surface.  Polymers are flexible and can be synthesized for specific purposes; however, conductive polymers do not have the electronic properties of carbon nanotubes.

Polymers used as chemical detectors are referred to as conducting polymer nanojunctions, which consist of two semiconducting metal electrodes with a small gap in between. This space is bridged with the conducting polymer to create a nanoscale junction. Nanoscale sensors have several advantages. Tao explains that the small size reduces the surface-to-volume ratio of a device, making it more sensitive. The smaller volume also decreases response time to a reaction.

Tao notes that similar methods are used to convert conducting polymers and SWNTs into sensors. The polymers, when activated with probe molecules, can recognize different chemicals and biological agents. Peptides also can be attached to polymer nanojunctions. He shares that this type of chemistry is somewhat easier to accomplish with polymers than with nanotubes. “When you’re trying to modify or attach something to carbon nanotubes [with peptides], sometimes you can actually change the properties of the carbon nanotubes. Conductive polymers give you more freedom to choose the right way to do it,” he says.

But these nanotechnology applications have some challenges to overcome before they can fully mature. An important part of the development process will be integrating the various sensors and pieces into a single device. Tao believes that purely nanoscale applications do not solve all of the challenges inherent in detecting individual molecules. He adds that by combining the unique features of nanotechnology and conventional technology, scientists may be able to create structures that cannot be made with nanomaterials or bulk materials alone.

For future chemical and biological agent detectors to be effective, the sensors will have to sample gases and liquids. Tao explains that nanoscale sensors by themselves are not good enough for this testing. He notes that liquids and gases must be introduced to the sensors under controlled circumstances rather than by exposing nanoscale devices directly to the environment. One method for transferring substances to nanosensors uses microchip-based laboratories and microscopic channels to introduce liquids or gas to the detectors in a controlled fashion. Tao observes that most laboratory-on-a-chip systems now use chemical or electro-optical detectors. He adds that introducing carbon nanotube and polymer sensors into the chip’s fluid channels creates new possibilities for detecting different types of chemical and biological agents.

Amplifying nanoscale signals to where they can produce data remains a challenge. Interfaces between the nanoscale and microscale must be created when a chip is manufactured. Tao notes that SWNTs are connected to electrodes that are quite large in comparison to the tubes. He also points out that SWNTs compose only a small percentage of a microchip’s total area.

Scale is less of a problem for polymer nanojunctions because the electrodes are placed close together, and the polymer is grown to bridge the junctions between the electrodes. But packaging and interconnection still present difficulties. “There is an interface issue here,” Tao explains. “Individually, you can make a tiny nanoscale device, but this device has to be hooked up to external circuits,” he says.

For microscale and nanoscale sensors to operate, the equipment must be part of a larger apparatus that sucks in gas or fluid to introduce to the sensors. This delivery system and its interface with the sensors have not yet been developed. “What we have shown, basically, are the small tools that are the detection part. That’s just a portion of the whole device. Sensitivity can be very good if you can deliver the target molecules onto the sensors,” he observes.

But much work remains before the sensors can be developed into commercial products such as handheld chemical and biological agent detectors and analyzers. Tao explains that many practical issues need to be considered. Although researchers have used polymer nanojunctions to detect heavy metal ions in water, this was accomplished under laboratory conditions.

Tao believes that the technology is still several years away from maturity, and he cautions that developing a specific product is not necessarily a key direction for the research. But while the focus is still on basics, he adds that ASU is working with companies such as Motorola to determine the product potential of specific applications.

The ASU team has worked on nanoscale sensors for almost five years, and Motorola has been researching carbon nanotubes for more than a decade. In the past few years, Motorola has partnered with ASU on biological and chemical agent sensor application for SWNTs, explains Dr. Vida Ilderem, vice president of Motorola’s Embedded Systems Research Laboratories.

Ilderem shares that the technology is not mature enough to be applied directly to a sensor system. She notes that one goal of the research is to have many analytical systems on a single microchip so that the sensor can detect multiple agents. But none of the industrial aspects of developing microchip-based sensors has been considered. “For this to become a reality, you have to set up manufacturing, a supply chain and the whole infrastructure to use it. That doesn’t exist today,” she says.

The current state of the technology demonstrates that nanoscale sensors can distinguish chemical and biological agents with a high degree of sensitivity. Laboratory tests confirm that the sensors can detect molecules in the parts-per-billion range. Ilderem notes that successfully identifying a particular agent depends on its properties and the signal-to-noise ratio.

Developers also will need to amplify the signal levels from the nanoscale up through the microscopic levels to provide alerts and data. Ilderem explains that there are ways to amplify these signals to provide actionable information. “We have designed interface circuitry, and we’re looking at different ways to amplify these signals to be able to network them and take action,” she relates.

Motorola is looking at the technology for its potential to develop chemical and biological agent sensors for homeland security and public safety applications. Ilderem believes that the detectors will have to be networked together to provide more accurate alerts. The sensors would be one part of a larger system capable of detecting the overall levels of specific agents and providing varied notification levels. She adds that Motorola is seeking ways to leverage its experience with networked systems to digitize and distribute chemical and biological alert data.


Web Resources
Arizona State University Ira A. Fulton School of Engineering:


Enjoyed this article? SUBSCRIBE NOW to keep the content flowing.