Response to 'Comment on 'Interpretation of Fermi level pinning on 4H-SiC using synchrotron photoemission spectroscopy'' [Appl. Phys. Lett. 85, 2661 (2004)]

摘要

In our previous paper, we observed the Fermi level movements with in situ deposition of Ni contact on 4H-SiC using synchrotron radiation photoemission spectroscopy. It was found that the Fermi level (E_(F)) pinning at the surface of p-type SiC originated from a hole trap (H1), observed by DLTS measurements. Recently, Lin commented on the analysis of experimental results. Here we present corresponding answers. First, Lin claimed that the 4-A-thick Ni layer deposited on SiC was too thin to determine the barrier height because the surface band bending changes with the coverage of metal atoms. In the article referred to by Lin, the barrier heights of Ti and Mg on GaN gradually increased up to the thickness of ～15 A, but the barrier heights of Al, Au, and Pt on GaN were stabilized within a thickness of ～5 A. In other words, the change of barrier height with the metal coverage depends on the kind of metal. No report was found on the variation of barrier height for Ni on SiC. Figure 1 shows the change of Si 2p core level with the thickness of in situ deposited Ni layer on SiC. The binding energy of Si 2p core level changed only ～0.1 eV with depositing 15-A-thick Ni on n-type SiC, and no changes in binding energy were found on p-type SiC. This is in good agreement with the previously reported results that the change of the E_(F) position was stabilized within ～1 monolayer (ML) for Ni on GaAs. From these measurements, it is apparent that the 4-A-thick Ni layer is thick enough to form the stabilized barrier height on SiC surface. Meanwhile, the discrepancy between the barrier heights on p-type SiC determined using different approaches, i.e., φ=～1.15 eV from I-V and φ=0.95 eV from SRPES, could arise from fitting the valence band spectrum in Fig. 2(b) of Ref. 1.