Black Silicon Offers Enhanced Responsivity to Overcome Near-Infrared Photodetection Challenges
Nanostructured black silicon represents a straightforward approach to avoid bottlenecks in aerospace and defense sensing.
TONI PASANEN, ELFYS INC.
Detecting and measuring near-infrared (NIR) radiation with high sensitivity is essential for many
modern defense and security technologies. Applications deploying laser-guided systems, as well as range-finding and night-vision devices, require efficient photodetection to ensure precise operation under varying conditions. Another requirement for these applications is the high speed of photodetection. Speed is vital to enable instantaneous reaction to rapid changes in the light signal.
Courtesy of ElFys.
Achieving both high sensitivity and fast photodetection speed within the NIR band presents a significant technical challenge. At the component level, conventional photodiodes fabricated from silicon face several performance limitations for NIR applications. Given these drawbacks, opportunity exists for innovative alternative solutions to overcome these hurdles. The use of nanostructured black silicon for the photodiode is one such approach.
Responsivity versus speed
The critical performance metrics describing the sensitivity and speed of a photodetector are responsivity and rise time, respectively. Responsivity refers to a photodiode’s conversion of incident light into an electrical signal. High responsivity is essential for detecting weak or distant light signals, which is important for applications such as laser guidance and surveillance.
Rise time describes the speed at which a photodetector responds to changes in light intensity. It is defined more precisely as the time it takes for the detector electrical signal to rise from 10% to 90% of its final value after the device is subjected to light.
In photodetection, silicon photodiodes are commonly used due to their compatibility with existing semiconductor fabrication processes, plus their wide availability and overall cost-effectiveness. Although silicon is an excellent material for visible light detection, the absorption of NIR radiation in silicon is weak due to the decreasing absorption coefficient with increasing wavelength, which causes the photodetector response to drop. Silicon’s limitations for the NIR wavelength range — which become especially apparent beyond 950 nm — pose a significant challenge to the defense and security industries, in which the 1064-nm wavelength is of considerable importance. Several applications in the field, such as missile target designation, laser warning systems, and laser rangefinders, commonly use the neodymium-doped yttrium aluminum garnet (Nd:YAG) laser at this wavelength.
A straightforward way to improve the responsivity in the NIR region is to increase the effective optical path length that light travels in the silicon material; and an obvious way to achieve this is
to increase the thickness of the photodetector. When light needs to pass through
a thicker piece of material, it naturally
has a higher likelihood of being absorbed.
However, increasing the detector chip thickness simultaneously increases the rise time and makes the device slower. The reason for this is that in addition to the resistor-capacitor (RC) time constant of the detector circuit, the rise time also depends on the time that it takes to collect the light-generated charge carriers within the silicon material to the electrodes. NIR radiation penetrates deep in silicon, far away from the charge carrier-collecting p-n junction, which is typically located at the front of the device. A thicker chip increases the distance that carriers must travel in the depletion region to reach the electrodes. Simply, the transit time becomes longer.
Moreover, increased thickness necessitates the use of higher bias voltages to avoid part of the absorption occurring outside the depletion region (where no electric field exists). Without sufficient bias, the extremely slow diffusion-type movement of the charge carriers completely dominates the rise time.
This points to the inherent trade-off that exists between the two fundamental performance metrics of the photodetector.
The chip thickness must be carefully optimized for the specific application to maximize the detector performance.
Figure 1 highlights this by showing a simulation on photodetector rise time and responsivity at the 1064-nm wavelength as a function of detector chip thickness. While a minimum rise time, indicating fastest operation, is achieved with a chip thickness of ~150 µm, the responsivity is heavily reduced due to weaker NIR absorption. In contrast, highest absorption and responsivity are obtained with a thick photodetector chip, though the device
operates slowly. The optimal combination is somewhere between these two extremes and depends on the application.
Figure 1. Simulated rise time and responsivity of a black silicon four-quadrant photodiode at 1064 nm as a function of chip thickness. The rise time calculation assumes a 40-sq-mm active area and full depletion at 150-V bias. Courtesy of ElFys.
Photodiodes based on III-V semiconductor compounds offer an alternative solution for NIR detection. These materials absorb NIR radiation efficiently even if the layer is very thin, and they provide for high operation speed. However, processing of the III-V materials is more expensive than with silicon due to the complexity of the epitaxial processes used to grow the thin films, the higher cost of the needed substrates, and the lower scalability of the manufacturing processes. The material quality with III-V semiconductors is also typically lower than in silicon, which limits the available photodiode size. III-V compounds are also not naturally compatible with silicon-based electronics. This makes integration with CMOS circuits more challenging.
Better antireflection for responsivity gains
When light hits the photodetector surface, a portion of it is reflected away due to the difference in the refractive index between air and the photodetector material. This reflection loss can significantly reduce the amount of light entering the device: For a bare silicon surface, reflectivity is
typically as high as 30% for normal incidence at NIR. This means that 30% of the light is lost before it reaches the active region, resulting in an equally large drop in the responsivity.
A commonly used technique to reduce the optical losses is to introduce an anti-reflection (AR) coating on the detector surface. AR coatings have an intermediate refractive index between air and the photodetector material, and use of this coating on the surface minimizes reflections by creating destructive interference for reflected light waves. More light is enabled to enter the silicon material, and more photons are absorbed as a result. The material and thickness of the AR coating are optimized so that the reflection minimum occurs at the wavelength range of interest.
Figure 2. A black silicon four-quadrant detector chip (top) and a packaged component optimized for 1064 nm. In addition to defense applications, black silicon-based four-quadrant detectors are beneficial for other applications for which precise laser beam tracking or alignment is needed. The technology holds promise for medical imaging technologies and industrial process control. Black silicon nanostructures could also enable new capabilities in scientific instrumentation and analytical devices operating in the NIR. Courtesy of ElFys.
As an alternative to the AR coating, the approach of texturizing the photodetector surface opens the possibility to leverage additional (and often very distinctive) optical properties. An advanced version of the technique of texturization of the photodetector surface uses nanostructured black silicon (Figure 2). Black silicon consists of nanostructures smaller than the wavelength of light and acts as an
effective medium with a gradual refractive index transition from air to silicon. Since the nanostructures eliminate the optical interface between these two materials, they efficiently suppress reflections across a broad range of wavelengths and incident angles from the ultraviolet to
the visible and extending to the NIR (Figure 3).
Black silicon has other interesting optical properties. Notably, the irregular surface of black silicon scatters incoming light, effectively increasing the optical path length within the photodetector
material. This in turn increases the likelihood of photon absorption, particularly for longer wavelengths in the NIR region (Figure 3). Therefore, black silicon enables thinner photodetector chips to achieve the same (or better) light absorption compared with thicker counterparts. This relaxes the trade-off between the responsivity and speed of the detector and gives more freedom to optimize the performance of the device for a specific target application.
Figure 3. The spectral absorbance of silicon with a black silicon or planar uncoated surface (above). Courtesy of ElFys.
Black silicon four-quadrant detectors
Though not all precision munitions harness laser technology, laser-guided missiles represent one effective use of 1064-nm sources in defense and military applications. With laser-guided missiles, four-quadrant detectors are often used. These sensors divide the detection area into four equal segments. As a laser beam shines onto the surface of the photodetector, the position of the beam on the surface determines the amount of light received by each of the four quadrants.
One of the main challenges facing conventional four-quadrant detectors is the limited responsivity that can be provided at 1064 nm while still maintaining fast speed of operation. Conventional silicon photodiodes even with an optimized AR coating typically exhibit responsivity of <0.5 A/W at this wavelength at room temperature.
By enhancing light absorption and minimizing losses due to reflection, black
silicon photodiodes exhibit higher quantum efficiency, translating to improved responsivity. Indeed, black silicon four-
quadrant detectors can achieve responsivities >0.6 A/W, representing an improvement of up to 30% compared with conventional technologies. Higher responsivity combined with low dark current offers the capability to lock on to the target at longer distances, improving hit accuracy and circular error probable in application. It also makes the system more robust against any disturbances in the air that weaken the signal, such as smoke or dust. Furthermore, as a more sensitive
detector can detect weaker signals at greater distances, it allows an operator to designate targets from farther away. This increases the standoff distance, reducing the risk of the operator being within range of enemy fire or counterattacks.
In application: YAG-laser guidance
YAG-laser missile guidance refers to
the use of a YAG laser to designate (and guide) missiles to their targets. This technology is widely used in precision, laser-guided munitions to strike targets with high accuracy.
In the YAG-laser guidance system, a laser designator is used to mark, or “paint” the target with a beam at 1064 nm (Figure 4). The operator uses the YAG laser designator to illuminate the target. The device can be operated manually by ground-based infantry or mounted, such as on an aircraft. The laser beam reflects off the target’s surface, creating a “spot” onto which the missile can lock. The designator often operates in a coded pulse mode, sending a specific laser pulse sequence to ensure that the target is uniquely identifiable and avoid interference with the environment and other systems that may be present.
Figure 4. The operating principle of yttrium aluminum garnet (YAG)-laser missile guidance. Courtesy of ElFys.
As the missile is launched, it enters its initial trajectory. The missile may use inertial navigation during the early phase of the flight, to travel toward the general area of the target. The missile is equipped with a laser seeker, which activates as
the missile approaches the target area,
and the seeker detects the reflected laser beam by a location-sensitive NIR photodetector and determines the location of the laser spot. Data from the laser seeker is translated into real-time adjustments to the missile’s trajectory via the onboard guidance control system. This system
adjusts the missile’s guidance fins or thrust vectoring systems to keep the flight path aligned with the reflected laser
signal and to maintain focus on the target.
Laser beam tracking systems that are used for this application rely on rapid and accurate photodetection to monitor the position of a laser beam. As a result, these systems commonly use four-quadrant photodetectors.
After the laser beam hits the surface of the photodetector, and the beam position determines the amount of light at each quadrant, each quadrant then generates a photocurrent proportional to the light intensity that it receives. A comparison of the differences in signals between the quadrants enables the system to calculate the beam’s position relative to the center of the detector (Figure 5). This information is similarly forwarded to the control system, which facilitates any necessary adjustment to the position of the tracking
mechanism and the trajectory of the mis-
sile to re-center the laser beam on the photodetector.
Figure 5. A four-quadrant detector can determine the position of a laser beam on the detector surface by comparing the output
currents between the quadrants. Courtesy of ElFys.
Such systems obviously set demanding
requirements for photodetector performance. The speed of the missile demands that the photodetector respond promptly to changes in light intensity to avoid missing the target.
At the same time, the responsivity must be assured, because the light signal may often be weak due to the long distance to the target and disruptions in the environment, including smoke, rain, and/or fog.
Meet the author
Toni Pasanen is a senior project engineer at ElFys Inc., and a cofounder of the company. He has a background in applied research
on semiconductor-based optoelectronic devices and expertise on nanostructured black silicon surfaces and thin films; email: [email protected].
/Buyers_Guide/ElFys_Inc/c32279