First demonstration of 10 microns FPAs in InAs/GaSb superlattices (invited)

As an emerging infrared material, the InAs/GaSb superlattices have received more and more interest nearly 30 years after its initial conception in the 1970s. Experimental work on photon detectors using this material has a history of less than 10 years. After intensive research and with cutting edge molecular beam epitaxy (MBE) cell designs, high quality superlattice material growth becomes a reality. The most important figure of merit for developing the focal plane arrays (FPAs), the RoA value, has been continuously increased in the past few years. At state of the art, the material quality has been improved to a level that imaging FPAs become a possibility. Due to high flexibility in the material design, the room left for improvement is unknown. The original FPA using this superlattice material was reported from Center for Quantum Devices back in the year of 2003 with a cutoff wavelength of 8 μm. High performance FPA with cutoff wavelength of 5 μm was reported from Germany one year afterwards. Due to the exponential decrease of R ₀A with cutoff wavelength, combined with the passivation issues unresolved, the FPA imaging in the long wavelength infrared (LWIR) range is significantly more challenging than in the mid-wavelength infrared (MWIR) range. With our material optimization in the LWIR range, a R ₀A value beyond 100 UΩ·cm ² can be routinely achieved at liquid nitrogen temperature for detectors with cutoff wavelength around 10 μm. The quantum efficiency ranges from 30-50% depends on different material designs. The background doping level was measured to be in the mid 10 ¹⁴ cm ^-3 using capacitance-voltage (CV) characterization technique, as shown in Figure 1. We have made our initial demonstration using these materials with a format of 256×256, as shown in Figure 2. The mesa size is 25 μm×25 μm with a pitch of 30 μm. The FPA mesas were formed using both dry and wet etching techniques. SiO ₂ was used as a capping passivation layer on top of the mesas for the work we report here. At the current stage of development, imaging quality from dry etched samples is better than that of the wet etched ones. The FPAs we made begin to detect hot soldering iron around 180 K and start to see humans around 100 K. Although the quality needs to be improved, the concept has been proved. With developing processing techniques, we expect to have better imaging quality in the near future.