Cantilever Thermal Detector to Challenge Microbolometer

Kevin Robinson

Researchers at Oak Ridge National Laboratory have developed an uncooled thermal infrared detector that they hope will cut the costs of IR cameras by one-third.
Researchers Panos Datskos and Slobodan Rajic have developed a microcantilever, constructed with a proprietary microfabrication technique, that bends as it absorbs thermal radiation.
Researchers make the cantilevers out of semiconductor materials such as silicon or GaAs. The devices are 50 to 200 µm long, 10 to 30 µm wide and 0.4 to 4 µm thick, and can detect temperature differences as small as 10²⁶K.
Coatings are applied to enhance the cantilevers' absorption properties. Datskos said the coatings not only tailor the device to the most useful parts of the infrared spectrum, but also can enhance the cantilever's bending, thereby improving its signal. With the proper coating, it detects a specific wavelength range; the degree of bending determines the emission intensity.

How much bend?
There are three methods for detecting the degree of bending: piezoelectric resistance, a laser and position-sensitive photodetector, and capacitance. The researchers have performed several tests using piezoelectric resistance, but Rajic said that it is not the most sensitive method because the electric current running through the cantilever generates heat, making the device less sensitive. The laser method, similar to the method used to detect cantilever deflection in atomic force microscopy, is extremely accurate, but applying the technique to more than one cantilever at a time is troublesome.
The capacitance measurement detects changes in capacitance between the cantilever and the substrate. It occupies the middle ground in the trade-off between sensitivity and simplicity. Rajic said it would be useful in measuring cantilever arrays.
The US Department of Energy has funded the research, but Rajic said he hopes commercial funding will move up development. At the SPIE Photonics West/Biomedical Optics conference, he presented the device as a reliable way to perform thermal sensing in vivo. A cantilever or array of cantilevers can fit in an endoscope, delivering the sensitivity and response time necessary for measurements inside the body.
Rajic said the device is two or three years from widespread commercial development.