Uncooled Infrared Detector Array Research at Koç University

In this article we focus on a new thermomechanical infrared (IR) detect

Issue 21

What’s the state of the art in Thermomechanical IR Detector array research at Koç University?
Aselsan (Turkey) has been the main funding resource behind this development project since 2006 and licensed the original patent by Koç University researchers. The goals of the initial project were to develop a 160x120 array with 50m pixels. The research resulted in a number of scientific publications and patentable ideas. The first phase of the project was completed recently and the major goals were met. Currently, still images and real-time video can be acquired in the 8 -14um IR band (Figure 1) at 30 fps. Higher resolution arrays with 35 um pitch are being developed with a target NETD level of 100 mK. Future efforts will be focused on miniaturization, increasing the resolution while reducing the pixel size further, and integration of the optical readout within the detector package. Our group collaborates with a number of research laboratories in Turkey, USA, and Europe for microfabrication of the detector arrays, packaging, and integration.

How does the technology work?
Operation Principle of the Uncooled Thermomechanical Detector Arrays with Optical Readout is described in this section. Among the various methods for sensitive displacement measurement, diffraction gratings embedded with MEMS sensors has been one of the leading methods. Atomic Force Microscopy, Grating Light Valve (GLV) display systems, MEMS microphones, bio/chemical sensors are among the applications that benefit the sensitivity advantage of diffraction gratings. The optical detection of the uncooled thermomechanical imagers, developed at OML, is also carried out using embedded diffraction gratings
As shown in Figure 2b, the detectors are composed of IR absorbing membranes that are connected to the substrate via bimaterial legs. Once IR energy is absorbed, the bending of the bimaterial legs create a displacement on the membrane. The movement of the membrane, and the reflector that is located on the membrane, causes the optical path difference within the gap between the sensor and the grating. This displacement can be captured by illuminating the FPA with a laser from the backside of the transparent substrate and the diffracted light can be observed using a photodetector. Sub-nanometer interferometric sensitivity can be easily achieved using diffraction grating sensors. Thanks to the sinusoidal intensity of diffracted light with respect to detector movement, the FPA is immune to saturation or “sun blindness” and can detect a very broad range of temperatures with high sensitivity. The acquired images in Figure 1 were captured by observing the diffracted light from the FPA by using a CCD camera. Figure 3 shows the vacuum-sealed package developed in collaboration with VTT (Finland). It houses the detectors and includes a TEC, a heat sink, and a pressure sensor. An interesting feature of the package is that, one side is an IR window for the detectors and the other side is a visible window for optical sensing.
What are the most important advantages of this technology?
There exist a number of advantages that differentiate uncooled thermomechanical detector arrays with optical readout from other uncooled technologies;

• The MEMS sensor array is decoupled from the readout IC (ROIC). Therefore, each part can be designed and optimized separately allowing for different wafer material and size selection for the IR sensor and ROIC.
• The IR detector array is completely passive, i.e., does not require any electrical connections to the substrate and very good thermal isolation can be achieved.
• Designs are easily scalable to small pitch and higher resolutions as it doesn’t require electrical interconnections on the detector chip.
• Microfabrication is simple, uses standard MEMS processes and requires only a 4-mask process.
• The array is immune to saturation (or sun-blindness effect while observing high temperature targets) due to cyclic readout signal of the interferometer.
• Large dynamic range can be achieved. This is possible by incorporating readout algorithms to extend the range of the optical interferometry beyond /4 offered by the standard readout using one laser.
• Diffraction grating based readout is proven to achieve very high precision for detecting small mechanical deflections therefore theoretical NETD levels as low as ~10mK seem possible.

What are the other major projects at Koç University, Optical Microsystems Laboratory (OML)?
OML currently has about 20 members working under the direction of Prof. Hakan Urey and has a large number of national and international collaborators. OML’s research is focused on optical MEMS, which are systems that combine micro-optics, micro-mechanics, and micro-electronics with applications in displays, imaging, biosensors, and spectrometry. In addition to the Aselsan sponsored Thermomechanical IR detector array project, the following are the current major projects in the group: (i) MEMS scanners and micro-optics for pico-projectors and 3D displays (sponsor Microvision Inc., USA); (ii) MEMS Fourier transform spectrometer development (project MEMFIS funded by EC 7th Framework Program-FP7); (iii) 3D laser TV development (project HELIUM3D funded by EC 7th Framework Program-FP7); (iv) Multi-analyte screening biosensor development (project MIMOSA, funded by TÜB?TAK).