Characterization of Zirconium - Diamond Interfaces
- PDF / 465,430 Bytes
- 6 Pages / 414.72 x 648 pts Page_size
- 107 Downloads / 212 Views
EXPERIMENTAL DETAILS Several natural p-type single crystal semiconducting diamond (100) substrates were used in this study. To remove nondiamond carbon and metal contaminants an electrochemical etch has been employed. Details of this technique have been described earlier [22, 3]. The samples were then loaded into a UHV system consisting of several interconnected chambers featuring capabilities for annealing, metal deposition, ARUPS and AES. Two different in vacuo cleaning processes were used to study the effect of surface treatment on the characteristics of the zirconium - diamond interface. One procedure involved annealing the wafers to 500°C for 10 minutes. And the other involved a 1150'C anneal for 10 minutes. The base pressure in the annealing chamber was - 1 x 10-10 Torr and rose to - 8 x 10`o Torr and -7 x 10- 9 Torr during the anneals, respectively. Subsequent to the anneal a Zr e-beam evaporator was employed to deposit 2 A thick films. A quartz crystal monitor was used to measure the thickness. During deposition the pressure was - 2 x 10-9 Torr. Following each annealing - and deposition step, UPS and AES were used to characterize the samples. The presence of a Zr film was confirmed by means of AES. AFM images of the diamond wafers clearly display linear groves with a depth of - 20 A. This surface structure is due to polishing the samples with diamond grit. No island structures were observed in AFM measurements after 2 A of deposition, indicating a uniform 2D layer. A discharge lamp was employed to excite Hel (21.21 eV) radiation to facilitate the photoemission and a 50 mm hemispherical analyzer with an energy resolution of 0.15 eV was used to detect the emitted electrons. To overcome the workfunction of the analyzer a bias of 2 V was applied to the sample. It was therefore possible to detect the low energy electrons emitted from the NEA surface as a sharp peak at the low energy end of UPS spectra. The position of this feature corresponds to the energy position of the conduction band minimum, Ec. Electrons emitted from Ec appear at Ev + E. in the spectra, where Ev is the energy of the valence band maximum and EG the bandgap energy. Furthermore, electrons from Ev get photoexcited to an energy level at Ev + hv in the conduction band and are obviously detected at Ev + hv in UPS spectra. This corresponds to the high kinetic energy end of the spectra. Therefore the spectral width for a NEA surface is hv - EC. Using the value of hv = 21.21 eV for Hel radiation and EG = 5.47 eV for the bandgap of diamond, a spectral width of - 15.7 eV is obtained. For a surface with a positive electron affinity the low energy cutoff is determined by the vacuum level. This results in a smaller value for the spectral width as compared to the case of a NEA. Photoemission spectra that exhibit features from both the semiconductor and the metal can be used to determine the Schottky barrier height DB (Fig. 1). Electron energy E
EC
Ev + hv
El! (D (-
m 4hma nd
hv - Ea h
Ef
-v/• 8.3 eV
Emission from the diamond
1
Emission from
__
F
Data Loading...
