Uncooled IR Photodetectors Get a Performance Boost

Józef Piotrowski

End users can look forward to a variety of benefits from current research, including high-performance and fast-response single-element devices.

Within the next year or so, end users can expect significant performance improvements in medium-wavelength IR (3 to 5 μm) and long-wavelength IR (8 to14 μm) photodetectors operating without cryogenic cooling. These high-temperature devices will gain from the use of more advanced architectures, the integration of micro-optics and active elements, and the use of photonic crystals. The HgCdTe and related Hg-based alloys will remain the design materials of choice.

An advanced photodetector based on HgCdTe heterostructure monolithically integrates functions of optical concentration of IR radiation and detection in one element.

Devices anticipated this year include HgCdTe high-temperature photodetectors with response speeds of less than 100 ps. Background-limited operation of medium-wavelength IR devices cooled with thermoelectric coolers (simple two-stage or even one-stage devices), as well as improved performance of uncooled and Peltier-cooled long-wavelength IR detectors are in the offing. In commercially available photodiodes cooled with two-stage thermoelectric coolers, detection capabilities should exceed 1010 cm^1/2Hz^1/2/W at 10.6 μm.

Shown are spectral detectivities of advanced optically immersed HgCdTe devices manufactured by Vigo at the end of 2002: Line 1 shows the background limit D_BLIP*(300 K, 180º); lines 2, 3, 4 and 5, the measured performance at 225 and 300 K, respectively. Maximum detectivity of 4 x 10⁹ cm^1/2Hz^1/2W has now been obtained at 10.6 μm with two-stage thermoelectric cooling.

Uncooled and Peltier-cooled photodetectors also will replace cryogenically cooled devices in applications such as IR gas analyzers, laser rangefinders, warning devices, thermal imagers, fast pyrometers, infrared spectroscopy, laser technology and metrology, lidar, medicine, fiber and free-space high-transfer-rate optical communications, and environment protection. Detector costs will continue to decrease, mostly because of refinement of the HgCdTe epitaxial growth process. Designers will replace expensive CdZnTe substrates with low-cost and large-area hybrid substrates such as GaAs, sapphire and silicon, covered with suitable buffer layers. Device processing will see further improvements.

Engineers can expect commercial availability of two-dimensional arrays of Auger-suppressed medium-wavelength IR photodetectors (British) that can be successfully used for imagers with thermal resolution of ~10 mK.

On the research front, goals for the next two to three years will include design optimization through the use of more complex heterojunction architecture. End users can expect improved quality of HgCdTe-based heterostructures grown by a low-temperature epitaxial process such as molecular beam epitaxy or metallorganic chemical vapor deposition. The main issues here will be the development of controllable doping at both low and high levels, low-temperature activation of acceptor dopants, reduced concentration of the Shockley-Read centers, and sharper, better-defined interfaces. Improved surface techniques will include heterostructural passivation obtained during epitaxial growth.

Another goal is the development of advanced devices that integrate the optical, detection and signal processing functions (concentration of incident radiation, effective use of infrared radiation for generation of charge carriers and internal photoelectric gain) in detector chips. Scientists are studying concepts such as two-color high-temperature photodetectors and low-cost, reproducible arrays of microlenses integrated with detector arrays.

Other research developments could include:

• Significant reduction of the low-frequency noise in uncooled Auger-suppressed, extracted long-wavelength IR photodiodes.

• Use of new narrow-gap semiconductor materials such as III-V compounds, including nitrides.

• Use of type II and III quantum-well devices offering Auger suppression at the equilibrium mode of operation caused by specific band structure.

Computer simulation will be used more in photodetector development. The simulators will account for optical, photoelectrical and thermal phenomena in detector structures.

Meet the author

Józef Piotrowski is the development manager of Vigo System SA in Warsaw, Poland.