• Fukushima, Sei
• Yamashita, Shunsuke
• Kudo, Yoshihiro
• Arai, Ryoji

Abstract

<jats:p>Defects in semiconductor materials significantly impact their inherent properties, making the evaluation of local defects and their energy levels crucial for controlling device performance. With advancements in monochromators, electron energy loss spectroscopy (EELS) combined with scanning transmission electron microscopy (STEM) has emerged as a promising approach for assessing crystal and band structures of semiconductor materials at the nanoscale. However, there is limited experimental research investigating the relationship between actual defective structures in local regions and mid-gap states. In this study, we conducted high-energy resolution EELS (HR-EELS) measurements with an energy resolution of less than 100 meV to detect the mid-gap states of GaN. Various defects were intentionally induced through Ga-ion implantation, and the defect concentration varied in the depth direction. To understand the origin of the mid-gap states, we performed 4D-STEM analysis and atomic-resolution STEM observations. The HR-EELS measurements provided insights into the depth-dependent valence-loss spectra, revealing that the intensities corresponding to mid-gap states gradually increased toward the surface, whereas the slope at the onsets corresponding to interband transition decreased. Furthermore, local structural analysis unveiled the presence of structural disorder and defective structures, indicating the existence of extended defects such as stacking faults and domain boundaries. Observably, these defective structures were abundant near the surface and less pronounced in deeper regions. Based on these experimental findings, we concluded that the variations in valence-loss spectra can be utilized to qualitatively evaluate the crystal imperfections at the nanoscale.</jats:p>

Topics

• impedance spectroscopy
• surface
• semiconductor
• laser emission spectroscopy
• transmission electron microscopy
• defect
• band structure
• electron energy loss spectroscopy
• stacking fault