Position of fermi level on Al0.2Ga0.8N surface and distribution of electric field in Al0.2Ga0.8N/GaN heterostructures without and with AlN layer

Abstract

Position of Fermi level on Al0.2Ga0.8N surface and distribution of electric field in Al0.2Ga0.8N/GaN transistor heterostructures without and with AlN layer were studied experimentally using contactless electroreflectance and theoretically solving Schrodinger-Poisson equation with various surface boundary conditions. It has been observed that the thin AlN layer changes very strongly the distribution of electric field in this heterostructure but the Fermi level position on Al0.2Ga0.8N surface does not change significantly. Its position is the same within experimental uncertainly (i.e., ∼0.5 eV below conduction band) for both bulk Al0.2Ga0.8N and Al0.2Ga0.8N/GaN heterostructures.

Contactless electroreflectance studies of surface potential barrier in AlGaN/n‐AlGaN structures with various Al concentrations

“…This value is about 2.5 times larger than the value of the surface potential barrier of GaN/n‐GaN structures studied in this paper, as well as in Al 0.2 Ga 0.8 N/n‐Al 0.2 Ga 0.8 N structures and Al 0.2 Ga 0.8 N/GaN heterostructures studied in Ref. , where the surface potential barrier of 0.5 eV was observed for Al 0.2 Ga 0.8 N.…”

“…For AlGaN the Fermi level at the surface has not been studied extensively. Recent CER studies of Al 0.2 Ga 0.8 N/n‐Al 0.2 Ga 0.8 N structures and Al 0.2 Ga 0.8 N/GaN heterostructures show that the Fermi level at the Al 0.2 Ga 0.8 N surface is located about 0.5 eV below the CB which is very similar to results obtained for GaN . In general, no sudden change is expected with the increase of Al concentration but it is possible that the surface potential barrier can increase with increasing in the band gap of AlGaN alloy, i.e ., the Fermi level can move slightly further away from the CB.…”

“…Both studies described here and in Ref. were performed at room temperature in ambient (atmospheric) air so measurement conditions cannot be responsible for the change in Fermi level position from 0.5 to 1.4 eV. Moreover the intensity of sample illumination was kept as low as possible and no change could be seen in CER spectra when measured several times so all effects related to a change of the Fermi level position due to photoabsorption/photodesorption of oxygen can be neglected.…”

“…Distribution of built‐in electric field in AlGaN/GaN‐based transistor structures is affected by many parameters such as polarization effects , an intentional (or non‐intentional) doping as well as the Fermi level position at semiconductor surface . It has been shown that the position of Fermi level at the surface of (GaN/)AlGaN/GaN heterostructure can be treated as the surface boundary condition, which strongly influences both the distribution of electric field inside the (GaN/)AlGaN/GaN heterostructure and the two‐dimensional electron gas concentration at the AlGaN/GaN interface . In this context it is very interesting to know the Fermi level position at the surface of AlGaN with various Al concentrations, i.e ., the surface potential barrier in AlGaN/n‐AlGaN structures.…”

“…. The value of this field can be estimated with the knowledge of thickness of undoped layer and the Fermi level position at the surface and at the (Al)GaN/n‐(Al)GaN interface . The built‐in electric field in undoped (Al)GaN layer can be measured by CER spectroscopy since CER spectra exhibit Franz–Keldysh oscillations (FKOs) for bulk‐like transitions in layers with medium (or high) built‐in electric fields.…”