C, O, Si Impurities in PVT-AlN – Incorporation and measurement by SIMS/Absorption
Electronic Semiconductors, Components and Circuits, Integrated Systems, Sensor Technology, Theoretical Electrical Engineering
Final Report Abstract
Aluminum nitride (AlN) is a wide-bandgap semiconductor crystal that has garnered significant attention for its direct ultraviolet (UV) bandgap of 6.1 eV, making it highly suitable for UV optoelectronic devices. While initial interest in UV devices focused on water sanitation, the COVID-19 pandemic highlighted a broader need for effective disinfection of air and surfaces encountered in daily life. UV light, particularly at wavelengths below 265 nm, is capable of denaturing molecular bonds in viruses and bacteria, providing sterilization without the use of harmful chemicals. Moreover, deep-UV light (<230 nm) has shown promise for safe exposure to human skin, opening the door to more versatile applications. By alloying AlN with gallium nitride (GaN) to form aluminum gallium nitride (AlGaN), these desired emission wavelengths can be achieved. However, a key bottleneck in the performance of AlGaN-based UV devices is the choice of substrate used during epitaxial growth. On one hand, AlGaN devices have been grown on sapphire (Al₂O₃) substrates, which benefit from industrial wafer sizes (>8 inches) and optical transparency in the UV. However, the lattice mismatch between sapphire and AlGaN results in a high density of threading dislocations within the active device layers. These dislocations serve as centers for non-radiative recombination, thereby reducing device efficiency. On the other hand, AlN is a more compatible substrate for AlGaN, sharing the same wurtzite crystal structure and a closely aligned lattice constant. This greatly reduces dislocation density and improves the crystalline quality of the epitaxial layers. However, large-diameter singlecrystal AlN substrates remain difficult to produce with current technologies. Furthermore, AlN wafers grown by physical vapor transport (PVT) often exhibit strong UV absorption due to high concentrations of unintentional impurities. This absorption undermines device efficiency, particularly for flip-chip LED designs that extract light through the substrate. These challenges limit the widespread adoption of PVT-grown AlN as a viable substrate for high-performance UV devices. This project aims to address these limitations by enabling better process control in PVT growth, specifically in impurity content and wafer size/quality. To achieve this, the study calibrates UV absorption spectroscopy—a fast, non-destructive technique— against secondary ion mass spectrometry (SIMS), which provides precise impurity quantification. By analyzing the shape and magnitude of the UV absorption coefficient across wavelengths, the research reveals characteristic signatures associated with impurities such as carbon, oxygen, and silicon. These spectral features reflect both the concentration and relative ratios of impurities present in the material. This correlation allows for rapid assessment of PVT process variations and offers a path toward scalable, high-purity AlN substrates for next-generation UV optoelectronics.
Publications
-
Prediction of impurity concentrations in AlN single crystals by absorption at 230 nm using random forest regression. CrystEngComm, 27(2), 184-190.
Klump, Andrew; Hartmann, Carsten; Bickermann, Matthias & Straubinger, Thomas
