- Biomedical Optics & Medical Imaging
- Defense & Security
- Electronic Imaging & Signal Processing
- Illumination & Displays
- Lasers & Sources
- Micro/Nano Lithography
- Optical Design & Engineering
- Optoelectronics & Communications
- Remote Sensing
- Sensing & Measurement
- Solar & Alternative Energy
- Sign up for Newsroom E-Alerts
- Information for:
Defense & Security
Mid-IR LEDs project dynamic scenes
While thermal-emitter technology seems to have reached a plateau, infrared light-emitting diodes are poised for potential rebirth.
14 December 2006, SPIE Newsroom. DOI: 10.1117/2.1200611.0520
Dynamic infrared scene projection (DIRSP) is one of the most important—but rarely mentioned—applications of IR-emitting-device technology. DIRSP is a tool for indoor testing of IR seekers and thermal imaging cameras that simulates ‘synthetic’ IR scenarios. An advanced DIRSP system makes use of a Honeywell resistive thermal-emitter array and meets the requirement of simulating IR radiation at apparent temperatures of 300K to 740K in the midwave IR (3 – 5 μm, MWIR) and longwave IR (8 – 12 μm) by heating pixel elements to the required temperature.1 However, a long time constant (∼5ms) limits application of the device to the 200Hz frame rate. This is not acceptable for testing the high-speed IR cameras used in the detection of terrorist activities such as attacks using missiles or explosives. Since the rise-fall time of an image depends on heat dissipation, there is no room for further decreasing this key parameter. The major disadvantage of this heat transfer approach lies in its inability to simulate cold scenarios (such as winter or arctic scenes or a space background) and in its low observability. The performance of this device is therefore impossible to improve.
On the other hand, photonic devices such as MWIR LEDs are poised for a potential rebirth. Recent improvements in device technology and design has enabled power outputs >1mW at room temperature.2,3
To increase the scope of operational scenarios that may be supported by DIRSP, we recently examined4 whether commercially available LEDs could form a platform for photonic MWIR DIRSP devices capable of competing with advanced thermal microemitter technology in testing IR sensors, including forward-looking IR missile warning systems, IR search-and-track devices, and missile seekers.
The active regions of LEDs are grown on InAs substrates by liquid phase epitaxy as InAsSbP/InAsSb double heterostructures. The advantage of these structures is that the peak-emitting wavelength (λp) can be tailored over the MWIR region by merely changing the Sb content in the InAsSb active layer. The chips are packaged as substrate-down planar (450 × 450μm2) or substrate-up circular mesa structures with diameters of 300μm (see Figure 1). We measured the apparent temperature of radiation, Ta, emitted by LEDs tuned at different λp (Ta, the thermographic equivalent of light power density, is determined as the temperature of a blackbody emitting equal power in the spectral range of interest). However, our major goal was to dynamically simulate cold objects and low observables in the MWIR.
Figure 1. (a) The measured apparent temperature changes with current for mesa LEDs tuned at different peak emitting wavelengths (λp). (b) The same graph is plotted for planar LEDs. In both, dashed lines show narrow-band tests with a filter passband Δλ/λp = 8.0% and 10.5% for λp=3.4 and 3.8μm, respectively. Solid lines relate to MWIR-band tests. The insets are diagrams of mesa and planar structures with the active region (a. r.) shown in red.
Figure 1 shows apparent temperature versus current characteristics for mesa and planar LEDs tuned at different λp and recorded at a current pulse length of 50μs and a 25Hz repetition rate. The data show that coming from the short wavelength MWIR side, the apparent radiation temperature gradually decreases as λp increases (mostly due to the Auger recombination impact). Narrow-band Ta values were compared to those of thermal emitters (∼740°K). Only slightly-cooled LEDs (T∼200K) could easily step over this limit in the continuous-wave mode at current I ≥200mA.
We were able to show that combining forward and reverse bias revealed a unique IR LED property: the ability to simulate a hot or cold target with low observability with respect to a given background. Figure 2 illustrates the first demonstration of H. G. Wells' paradox5 in the MWIR. In the initial state (a), there is no current (I = 0) and the thermal imaging camera discriminates between the circular active region (1) of a mesa and the rest of a structure (2) due to the emissivity difference between these regions (Ta = 35°C) and the holder (3, Ta = 33°C) on which the structure sits. This positive thermal contrast becomes higher with a forward bias (b) when I = 50mA; the active region looks ‘warmer’ (Ta = 41°C, ΔT= 8°C). In the negative luminescence6 mode, the camera detects the active region as a low observable (c, I = -20mA, ΔT∼0) or cold object (d, I = -50mA, ΔTa = -5°C).
Figure 2. (a)In the initial state, the MWIR camera maps a mesa-like object (1, 2) and background (3) with positive contrast. (b) Positive bias increases the contrast. (c) Negative bias makes the active region difficult to resolve (zero contrast, low observable). (d) Negative bias can also transform it into an apparently cold object.
To summarise, the 3–5μm LED represents an ideal platform for photonic DIRSP devices operating at room temperature with its relatively high spectral output density and capacity to simulate not only hot targets (Ta∼750°K) but cold objects and low observables as well. Although these power-hungry off-the-shelf devices exhibit a very low external efficiency and suffer from excessive self-heating,4 they have become very strong contenders in the DIRSP market. The long-range future of photonic DIRSP technology is bright and our results provide a starting point for further improvements.
Plasma Phenomena, Institute of Semiconductor Physics
Volodymyr Malyutenko is the head of the department at the Institute of Semiconductor Physics of Kiev. His current research interests include high-level electron processes in IR emitting structures, 2D and 3D computer modeling of heat and light transfer in LEDs, light conversion processes, high-resolution IR scanning microscopy, and IR dynamic scene simulation devices.