For counteracting background current of photoconductive (PC) PbS detector, an example of layout design and analyze 1×128 linear PbS infrared focal plane array (IRFPA) detector using resistance selection of blind sensitive element is given. 1×128 linear PC PbS infrared focal plane array detector is fabricated and characterized by IRFPA test-bench. Results show that average responsivity of detector is 4.19×10<sup>6</sup>V/W; average detectivity of detector is 5.79×10<sup>9</sup>cm•Hz<sup>1/2</sup>•W<sup>-1</sup>.
Material characteristics of InSb wafer and Si wafer are fundamental factors to fabricating qualified infrared focal plane array (IRFPA). Parallelism and flatness of wafer are critical to yield and reliability of large format IRFPA. Influence of materials’ parallelism and flatness on indium bump growing, flip-chip bonding and back thinning in the fabrication of large format IRFPA is analyzed. Parallelism of material brings additional nonuniformity to IRFPA. Parallelism after back thinning is only determined by the parallelism of readout integrated circuit (ROIC) and isn’t affected by that of detector. Influence of material flatness is non-contacting or bad-contacting during flip-chip bonding, which results in bad pixels of IRFPA. According to actual fabrication condition of IRFPA, flatness of one single detector and ROIC chip should be better than 1 µm. Parallelism of ROIC chip should be better than 2 μm. Optical flat is the most convenient approach for InSb material morphology test. Utilizing higher indium bump and press in flip-chip bonding, designing larger contact metal under indium bumps or fabricating indium bumps with smaller diameter in center, selecting distribution of chips on wafer are put forward to reduce influence of morphology. Yield of large format IRFPA is improved.
Performance of IRFPA depends greatly on the amount and distribution of bad pixels. In this paper, general causes of bad
pixel in IRFPA are analyzed. Most bad pixels of IRFPA can be classified into four types for flip-chip bonding structure.
The amount of bad pixels in IRFPA often increases after long-term operation. This strongly affects application of IRFPA.
High temperature storage and temperature shock are effective ways to expose these potential bad pixels in advance. High
temperature storage and temperature shock are carried out on some IRFPA samples. Four kinds of variation for bad
pixels are investigated. They are variations of amount, characteristics, bad pixels on margin and bad pixels in different
IRFPA. Results show potential bad pixels damaged after these tests. New bad pixels are tested, analyzed and classified.
Each type of bad pixel is corresponding to defect of specified manufacture procedure. This indicates the potential
improving directions. Methods that could reduce bad pixels are briefly discussed. Results shown in this paper can help to
improve manufacture technology of IRFPA and then the performance of infrared imaging system.
Infrared Focal Plane Array (IRFPA) of InSb is designed to work under the condition of zero bias voltage or nearly to
zero bias voltage. InSb IRFPA can't work normally if PN junction is in the condition of high reverse bias voltage for
several minutes. In order to analyze the causation of the phenomena some experiments were designed to simulate the
condition of high reverse bias voltage. It is found that when the reverse bias voltage exceeds the limited value, the
responsibility of InSb detector becomes lower. The phenomena will not change unless the operating temperature is raised
to room temperature and kept for a long time. Response characteristic of InSb PN junction under high reverse bias
voltage is briefly described in this paper. The factors affecting response characteristic are discussed. The limited value of
the reverse voltage is given. The result is useful to design the driving circuit of InSb IPFPA. It also plays the guidance
part in application of InSb detector.
I-V temperature characteristic is very important to InSb IRFPA. In order to make further studies on I-V temperature
characteristic, some experiments were done. In the experiments, the operating temperature of the InSb array was
gradually raised from 77k. It is shown that reverse current doesn't simply increase with the increase of the operating
temperature. The reason can be attributed to the composing of reverse currents at different operating temperature. In this
paper, the I-V characteristic of InSb diode at different operating temperature is briefly described. The dominant
components of reverse current and its temperature characteristic are discussed. The change of detector impedance is
analyzed as operating temperature is changed. At the same time, the optimized operating temperature of InSb IRFPA is
presented. The limit of operating temperature at which InSb IRFPA can work normally is also given.
Nonuniformity impacting factors on an IRFPA with direct injection (DI) readout circuit and InSb PV detector are
analyzed in this paper. The nonuniformity of threshold voltage results in deviation of detector bias and current. Then the
detector nonuniformity and injection efficiency nonuniformity occur. Another expression for injection efficiency is
deduced and its relation with detector bias is obtained. Relation between FPA nonuniformity and detector I-V
characteristics is analyzed. The bias range from -0.2V to -0.1V is an ideal operating region for detector array. Small
deviation of detector current, insensitive to nonuniformity of threshold voltage, uniform response of detector array, high
detector impedance and high injection efficiency are all satisfied in this region. The best gate voltage range of injection
MOSFET is from 3.6V to 3.7V. FPA has minimum nonuniformity in this region, which is corresponding to detector bias
from -0.2V to -0.1V. Results shown in this paper optimize the performance of FPA.