Part of the Oxford Instruments Group
Expand

Understanding SWIR

SWIR imaging unlocks a whole new world beyond the limits of the visible light. Whether in industry or research laboratories, SWIR imaging reveals what the human eye cannot perceive. Thanks to its versatility, SWIR stands out across diverse fields, from industrial inspection or medical research to night-surveillance and telecommunications. In this focus on SWIR, we showcase the technology and explore its multiple applications.

Short Wave Infrared (SWIR) is defined as a region in the infrared spectrum that traditionally spans from 900 to 1700 nm. However, SWIR encompass what spectroscopists call the Near InfraRed (NIR) region or NIR-I, and NIR-II for biologists corresponding to two biological windows.

Figure 1: UV to LWIR spectrum

Similar to the visible spectrum, SWIR imaging relies on the reflectance properties of the materials: photons are either reflected or absorbed by the object, creating the contrast and shadows needed for high resolution imagery. Images are comparable to those in the visible in sensitivity and detail, featuring distinct shadows and high contrast, facilitating interpretation where visible light would be ineffective. However, they are monochromatic.

Unlike longer infrared wavelengths, where Mid-Wave Infrared (MWIR) and Long-Wave Infrared (LWIR) emissions originate from the object itself, SWIR experiences less interference from atmospheric elements enabling clearer imaging in challenging conditions. However, like MWIR and LWIR spectral bands, SWIR enables detection of thermal radiation

In this focus on SWIR, we showcase the technology and explore its multiple applications. Depending on the wavelength of interest, different technologies are available:

  • Germanium (Ge)
  • Lead-Tin-Telluride (PbSnTe)
  • Indium Antimonide (InSb)
  • Mercury Cadmium Telluride (MCT/HgCdTe)
  • Indium Gallium Arsenide (InGaAs)

Sensors like MCT for the most popular, allow to reach longer wavelengths. However, InGaAs sensors offer a costeffective, performant and practical solution. Here, we focus on the InGaAs sensors that were designed for SWIR-imaging.

SWIR imaging (900-1700 nm) produces high-resolution, monochromatic images with distinct shadows and high contrast. InGaAs sensors are a cost-effective, highperformance solution for SWIR imaging. SWIR imaging finds applications in various fields (astronomy, life science, military, beam profiling, …)

Working Principle

An InGaAs sensor converts light in current thanks to the inner photoelectric effect. A sensor is composed of several arrays of photodiodes (pixels). Most of the modern InGaAs sensors have P-I-N (Positive Intrinsic Negative) photodiodes (pixels). There are other photodiodes structures that exist (PN, Schottky, APD, …), but we focus on the PIN structure for simplicity and cause it is widely spread.

The physical principle behind these sensors is a P-I-N junction (1). Compared to a simple PN junction, a P-I-N photodiode is constituted of an “intrinsic” region sandwiched between a P-semiconductor where holes (missing electrons) are the most predominant carriers of charges and a N-semiconductor with electrons as the main carriers of charges

A P-I-N diode, thanks to its intrinsic region offers:

  • Efficient photon absorption because it is undoped (no free charge carriers)
  • Enhanced depletion region quantum efficiency because it can better collect charge carriers
  • Improved response times thanks to its reduced capacitance

P-I-N Photodiode

1. Light is absorbed
2. Electron (e-)-hole (p+) pairs are generated
3. A P(I)N Junction is created
4. An electric field Ê is induced and separates the electron-hole pairs
5. By moving, charge carriers induce a measurable electric current.

Figure 2: Scheme of a P-I-N junction

Among sensor’s characteristics, performances of a SWIR detector are determined by the photodiode quality and CMOS ReadOut Integrated Circuits (ROIC) integrated in the detector:

  • Dark current
  • Readout noise
  • Frame rate
  • Sensitivity

InGaAs Sensors

InGaAs sensors operate in charge integration mode, where the generated charge is accumulated to enhance the output signal, making them ideal for low-light detection. However, since silicon detectors lose sensitivity above 1100 nm, the focal plane array (FPA) must incorporate different materials to effectively cover the short-wave infrared (SWIR) wavelengths.

InGaAs sensors, which integrate hybrid components with CMOS technology, enable efficient signal processing and extend sensitivity into the SWIR range.

Figure 3: Schematic representation of components used in a VIS-SWIR InGaAs detector array

Extended Range InGaAs Sensors

Standard InGaAs sensors have a long wavelength cutoff at 1680 nm. However, many applications require detection at longer wavelengths (2). To address this, some SWIR cameras, often referred to as “eSWIR” (extended SWIR) cameras, use unique optics and electronic components optimized and coated to extend their range up to 2500 nm. Extended InGaAs provides a cost-effective solution for wavelengths between 1680 and 2500 nm compared to other technologies like HgCdTe sensors (3). (To know more about HgCdTe sensors, please have a look at our C-RED One camera)

Technically, InGaAs has a face cubic centered structure (see Fig.) and is usually grown on a InP substrate by epitaxy to keep the same lattice constant between the substrate and InGaAs cells (4). Extending the detection range to 2500 nm is achieved by varying the fraction of indium in the ternary compound, which changes the lattice constant between layers. Doping indium into GaAs decreases the band gap, enabling the detection of longer wavelengths. More details can be found in our focus on “Extended Range SWIR Imaging”.

Figure 4: Crystallographic structure of a InGaAs cristal grown by epitaxy on a InP layer. Both InP and InGaAs lattices are matched. As in red, P in blue and Ga/In in green

The Importance of Cooling to Reduce Noise

Advantages Disadvantages
Low power consumption Less performant than HgCdTe above 1680 nm
Can operate at room temperature Expensive compared to standard CMOS
Sensitive, great quantum efficiency and high frame rate Need cooling for better imaging or eSWIR
Smaller size, easy to integrate  
Cost-effective compared to HgCdTe sensors  

InGaAs sensors are an array of P-I-N photodiodes that convert photons to electrons. The resulting voltage can be measured and provide images. Noise must be taken into account for better image quality. This can be done by choosing a proper cooling

Context

Generally, InGaAs detectors can operate at room temperature, as they show good performance and an acceptable signal-to-noise ratio.

However, a little variation of temperature can dramatically change the noise. Thus, cooling the InGaAs FPA significantly reduces the dark current per pixel, which allows higher quality images.

Infrared detectors are cooled to improve the Signal / Noise ratio and to keep the detector element at a constant temperature. Cooling changes the spectral response but, cooled cameras and sensors are more expensive.

InGaAs sensors are small sized and use little power compared to CCD. They have fast time response, high quantum efficiency and combined with a cooling system, they can enable longer integration times.

Like most electronic devices, SWIR cameras work with inherent noise that can interfere with image quality. Compared to the sensors operating in the visible spectrum, InGaAs sensors have lower band gap and thus, have slightly higher noise.
Three contributions of noises must be taken into account when choosing a camera:

  • Dark noise: arises from thermally generated electrons and it is typically the predominant noise for InGaAs sensors
  • Readout noise: Inherent to the amplifiers and due to the process of converting charges into a voltage that can be quantified. Very often, this term refers to the total noise of the camera that can be misleading for the user
  • Photon shot noise: depends on the arrival rate of photons onto the sensor and follows Poisson statistics

To reduce the total noise, especially at long exposure times and improve camera’s performance, InGaAs sensors need to be cooled. There are several techniques to transfer heat out of the system:

  • Thermoelectric cooling (TEC-cooling): using the Peltier effect, heat is driven out of the sensor by applying a current through a succession of semi-conductors.
  • Water cooling: thanks to its heat capacity, water is a great candidate to store heat and drive it out of the system
  • Stirling cooling: cooling is achieved by compressing and expanding an inert gas (usually Helium) drawing heat out of the system. Vibrations may arise from this cooling method.
  • Liquid Nitrogen: nitrogen can cool down the sensor to very low temperatures due to its low liquefaction point (−196 °C; 77 K). However, special and costly equipment is required and may be complex to use (refill / maintenance every 4 to 8 hours)

InGaAs sensors can operate at room temperature, but proper cooling improves the image quality especially at longer exposure times where photon shot and dark noise become important.

Applications

In life sciences, researchers manipulate small animals for preclinical studies. They image live, using fluorescence techniques. However, most fluorophores lie in the 500-900 nm optical window and show limits when it comes to imaging. With the rise of new fluorophores such as carbon nanotubes, quantum dots or rare-earth down-conversion nanoparticles, a new optical window known as NIR-II/SWIR is available and offers great imaging opportunities (5). In this optical window, wavelengths ranging from 1000 to 1700 nm allow researchers to get rid of high background noise due to light scattering and cells autofluorescence, and thus achieve better imaging depth.

Optimal is a company in Grenoble that specializes in small animal imaging using Indocyanine nanoparticles (Serb Laboratories) that can be injected in the body and eliminated by urinary or hepatic routes after imaging. Increased penetration depth makes it a great candidate for medical imagery deep inside biological tissues. Thanks to the very low noise SWIR camera C-RED 2* and these nanoparticles, the whole circulatory system of a mouse can be imaged with great contrast. To know more, please have a look at our focus on small animal imaging.

Figure 5: Pictures taken with the C-RED 2 camera (*) showing a mouse after injection of “GoldenEyes” gold nanoparticles in ambient light conditions (on the left) and under λ=808 nm laser illumination (on the right). The bright signal on the image depicts the circulatory system of the mouse.

(*) SWIR InGaAs cameras are considered as a dual-use good. Dual-use goods are products or services that can have both a military and civilian application. As such, exportation of these goods is subject to restrictive controls for some countries, requiring an export license.

Astronomy

SWIR is also used in optical astronomy and adaptive optics. The use of infrared radiation for astronomy is crucial because unlike light at visible wavelengths, infrared light is not blocked by interstellar dust. It finds applications in exo-planets research, adaptative optics, laser communications, astronomical observations, ground-to-ground or ground-to-space transmissions,…

Figure 6: Photography of Saturn taken by an amateur Celestron 8 inch telescope with C-RED 2 (Courtesy of Jean-Luc Gach)

Detection and Sorting

Colors that can be confused or look similar in visible, can be better differentiated with SWIR. InGaAs sensors offer high resolution images at high frame rate and find many applications in industry for quality check or sorting, saving time and are highly reliable thanks to the associated computer vision algorithms.

Figure 7: On the left: visible imaging of beans. On the right: SWIR imaging of the same beans: defaults are easily perceptible

Surveillance: Steam, Fog and Smoke Penetration

The principal advantage of SWIR is the atmospheric penetration offered by SWIR wavelengths compared to visible or NIR bands. Indeed, SWIR is very useful in difficult atmospheric conditions, whereas visible cameras are blind. Thus, a lot of applications impossible to achieve with visible light are possible: imaging in fog, steam water, or dense smoke like forest fire. Seeing through some materials like silicon or glass is also possible with SWIR.

Figure 8: Haze across the Potomac River in visible light (left) and in SWIR (right) (“What good is SWIR?” R.G.Driggers et al.) (courtesy of Peter Judd and James Waterman, NRL).

Solar Panels Inspection

SWIR cameras can be embedded on drones to provide fast and high resolution images for the detection of cracks and defects on solar cells. A laser beam is directed towards the solar panel and excites its cells. These cells remit light in the SWIR by photoluminescence. The photoluminescent signal is collected by a C-RED 3 camera with an objective mounted on the drone and the resulting image shows if defects or cracks are present on the solar panel. Check our learning centre to know more about photovoltaic inspection.

Figure 9: Laser-induced Photoluminescence Imaging (LIPI) for fast and accurate inspection of photovoltaic plants. On the left: drone identifying a solar panel. On the right, a solar panel with visible defects on its surface (Courtesy of First Light Imaging, Creative Sight™, Optoprim, and the Technical University of Denmark)

Conclusion

SWIR imaging, driven by InGaAs sensors, has proven to be an essential tool across various industries, offering clarity where visible light fails. Its ability to deliver high-contrast and resolution in difficult conditions makes it ideal for applications in various fields like life science, industrial inspection, security,…

Although InGaAs sensors can operate at room temperature, a cooling system should be considered for better imaging or at longer integration time to minimize dark current. Thanks to recent advancements in the technology, extended SWIR provides cost-effective solutions for professionals needing to reach longer wavelength at a lower cost.

SWIR imaging is a critical tool for professional seeking high-performance for demanding applications,

References

1. InGaAs for infrared photodetectors. Physics and technology. Kaniewski, Jakub et Piotrowski, Jozef. 1, 2004, OPTOELECTRON REV, Vol. 12, pp. 139-148.
2. C-RED 2 ER: an extended range SWIR camera with applications in hyperspectral imaging. De Kernier, Isaure, et al. San Francisco : Proceedings Volume 11997, Optical Components and Materials XIX, 2022. SPIE OPTO. p. 119970V.
3. HgCdTe Infrared Detectors. Norton, Paul. 2002, OPTO-ELECETRON REV, Vol. 10, pp. 159-174.
4. Dopant–dopant interactions in beryllium doped indium gallium arsenide: An ab initio study. Kulish, Vadym, et al. 4, 2018, J MATER RES, Vol. 33, pp. 401-413.
5. Shortwave-infrared (SWIR) emitters for biological imaging: a review of challenges and opportunities. Thimsen, Elijah, Sadtler, Bryce et Berezin, Mikhail Y. 5, 2017, Nanophotonics, Vol. 6, pp. 1043-1054.

For more information, or to discuss your application, feel free to contact us:

fli-contact@oxinst.com

Related assets