Thermal Deformation Of Micro Uncooled Infrared Sensor With Capacitive Readout

Infrared (IR) imaging is the key technology for night vision and remote sensing for both civilian and military applications. The thermal infrared image sensor is a device that forms an image using infrared radiation even in night and bad weather conditions. With the development of micro fabrication, the micro infrared sensor can be formed a Focal Plane Array (FPA) which consist of millions of image pixels. Comparing with cooled IR sensor, micro uncooled infrared sensor has certain advantage such as low cost, leas weight, low power consumption. The image pixel is the object of this study.Literature review has been conducted for the development of infrared sensor and some useful concepts of infrared radiation has been listed. Based on the past development, we propose a novel bi-materials cantilever micro uncooled infrared sensor pixel, which consists of bi-materials cantilever, thermal isolation, electrical pad and base. The pixel structure is a key issue of the study, its design must take into account several factors, such as its compatibility, absorb material for specific radiation wave length, thermal property, thermal conductivity, self-radiation and size. The pixel working principal can be described as:the absorbing layer will be heated up after receiving radiation, the bi-materials cantilever beam will be deformed due to the different of the bi-material's thermal properties, the deformed cantilever will cause capacitive change, which can be detected by a specific IC circuits, then the temperature distribution of the object can be measured indirectly. For quality insurance, we also studied the motion of object and infrared sensor influence on the cantilever beam.Theoretical and finite element (FEM) have been formulated for the problem, the thermal deformation has been carried out, the FEM results are very good agree with our analytical solution, which proved our formulation is correct. The change of capacitance due to the deformation can be used to detect the temperature distribution, this make it possible to convert the object into a thermal image, due to resource constraints, we limit our study only on the theoretical aspects of the devices, and no discussion on fabrication.