A numerical model for designing Geiger-mode 4H-silicon carbide avalanche photodiodes (GM-4H-SiC-APDs) in near-ultraviolet (NUV) wavelength ranges, developed by researchers at the DEVCOM U.S. Army Research Laboratory, could resolve several issues that typically arise when APDs are used to detect photons at the longer UV wavelengths. 4H-SiC APDs demonstrate high sensitivity in the deep ultraviolet (DUV) range. However, longer wavelengths require APDs with specially-designed architectures to improve unity-gain quantum efficiency (QE) so that single-photon sensitivity can be maintained. At certain wavelengths, a process known as impact ionization can cause photons to generate electron-hole pairs when they are absorbed by photodiodes. When impact ionization occurs in an electric field, it can cause charge multiplication. An APD is biased above its “breakdown voltage,” at which point impact ionizations reach a self-sustaining rate, resulting in a distinct electrical pulse that is readily detectable. Researchers created a numerical model that helps design Geiger-mode avalanche photodiodes (APDs) for near-ultraviolet (NUV) light detection. Courtesy of Openverse/Radovan Blažek. For an APD to detect single photons in the presence of mechanisms that generate impact ionization, there must be a high probability that the APD will absorb incident photons of the desired wavelength. The ability to support high fields, and the presence of high QE at the desired wavelength, are critical factors in determining the sensitivity of such an APD device. To reliably detect photons at higher wavelengths where absorption is weaker, the baseline photon capture of 4H-SiC APDs needs to be made more efficient, as indicated by the unity-gain QE. Typically, APD developers resolve this issue by increasing the thickness of the APD’s absorber layers. “APDs with much thicker absorber layers (10s of microns) must be utilized to improve the NUV response, which necessitates switching from a conventional PIN architecture (usually less than 3 μm thick) to a separate-absorption charge-multiplication [SACM] architecture,” researcher Jonathan Schuster said. “However, this involves unique challenges, such as deviating from existing front-side absorber SACM architectures to a very thick backside one.” Schuster and his team overcame these challenges by developing a numerical model with a calibrated 4H-SiC material library, and designed SACM APDs that are predicted to exhibit high single-photon detection efficiency in the NUV range. The researchers used the model to design NUV-enhanced SACM structures. They considered two architectural designs for their APDs — non-reach-through (NRT) and reach-through (RT). They designed NRT-SACM APDs, with unity-gain QE of up to 32%, and RT-SACM APDs, with unity-gain QE of up to 71%, for photons with a wavelength of 340 nm. Both designs maintained a large electric field in the multiplication layer for Geiger-mode operation. Schuster sums up observations made by his team, regarding the new APD designs for NUV wavelengths. “For the NRT-SACM case, it was determined that the doping profiles must be engineered such that two competing mechanisms are balanced, maximizing the minority carrier diffusion length in the absorber layer (AL) while minimizing the corresponding potential barrier at the AL/charge layer (CL) interface,” he said. “Conversely, in a RT-SACM architecture, it was determined that a narrow range of total charge in the CL properly modulated the electric field to be non-zero in the AL and sufficiently large in the multiplication layer to operate above avalanche breakdown.” The improvements in QE, incorporated into these designs, could translate to a broader application of APDs for high-wavelength photodetection. The researchers identified several design rules that may be useful for designing Geiger-mode APDs for single-photon counting applications in the NUV range. In addition, they discovered that charge layer designs in APDs are inflexible when it comes to deviations in layer thickness or doping, which could make their fabrication more challenging. Optimizing APD design for NUV photodetection could lead to more sensitive, efficient APDs for various applications that require UV photon detection, such as solar-blind UV detection, combustion monitoring, and environmental UV monitoring. In the future, the numerical model could be instrumental in introducing APDs that are highly responsive detectors in the NUV region for use in a range of applications. The research was published in IEEE Journal of Quantum Electronics (www.doi.org/10.1109/JQE.2025.3591762).
