Sadanori Yamanaka

Device grade B-doped homoepitaxial diamond thin films

We review our recent results on properties of high quality B-doped homoepitaxial diamond thin films grown by chemical vapor deposition (CVD) by using trimethylboron (TMB) as a B source gas. The conventional (one-step) and the two-step growth methods were used for film preparation. The latter realized smooth surface without any non-epitaxial crystallites (UCs). The films showed a sharp Raman shift peak at 1332 cm -1 , a strong free-excitonic emission at room temperature, and high Hall mobility as high as 1000 cm 2 /Vs or more, indicating high-quality diamond. We also successfully realized resistivity control of the films in a wide range from 10 0 to 10 5 Ω cm due to low compensation ratio. Using the two-step growth method, Schottky junctions with the ideality factor n of 1.1 or less and undetectable leakage current could be prepared between various metals such as Al, Zn, Cr, Ni, Au or Pt and oxidized B-doped CVD diamond films. In particular, we successfully made nearly ideal Schottky junctions using highly B-doped films in the order of 10 17 cm -3 , indicating that the quality of the present B-doped films is comparable with or higher than those of conventional semiconductors such as Si and GaAs.

n-Type Control by Sulfur Ion Implantation in Homoepitaxial Diamond Films Grown by Chemical Vapor Deposition

n-type control was achieved by sulfur-ion-implantation in homoepitaxial diamond (100) films grown by chemical vapor deposition (CVD) for the first time. Sulfur-implantation was carried out with energies of up to 400 keV at 400°C. The activation energy of the conductivity was 0.19–0.33 eV depending on the conditions of ion implantation. A junction between this layer and a boron-doped p-type layer was fabricated by combining sulfur-implantation with gas-phase boron doping during CVD. The junction exhibited clear pn junction properties. The capacitance of the junction decreased with reverse bias voltage, which confirms that the depletion region of the junction was actually extended with the reverse bias voltage.

Hydrogen analysis of CVD homoepitaxial diamond films by high-resolution elastic recoil detection

We have measured hydrogen depth profiles in chemical vapor deposition (CVD) diamond films by high-resolution elastic recoil detection. The depth resolution of nearly 0.2 nm is achieved by means of a high-resolution magnetic spectrometer. The hydrogen profile in the as-grown sample shows a sharp peak at the surface. The peak has a small tail toward larger depth, showing that some hydrogen atoms are incorporated in the subsurface region. There is no difference in the hydrogen depth profiles between the undoped and B-doped diamond films. After surface hydrogen is removed by an acid solution, the CVD diamond is rehydrogenated using a hydrogen plasma at 800 °C. The rehydrogenated sample shows almost the same hydrogen profile as the as-grown sample. The hydrogen profile hardly changes by annealing at ∼400 °C, though a small change in the subsurface region cannot be excluded.

Structural properties of sulfur-implanted diamond single crystals

The lattice location of sulfur implanted into diamond single crystals has been investigated using particle induced X-ray emission and ion channeling. Sulfur atoms were implanted into high-quality undoped homoepitaxial diamond [100] film grown by microwave plasma assisted chemical vapor deposition onto high-temperature and high-pressure synthetic Ib diamond [100] substrates, as well as into Ib diamond substrates directly, at 400/spl deg/C up to the concentration of 1/spl times/10/sup 20//cm/sup 3/. They were annealed at 800/spl deg/C in vacuum for 100 min after the implantation. Sulfur dopant was found to occupy preferentially substitutional sites in the host lattice. The possible maximum displacement of sulfur dopant was 0.14 A from axis, and 0.07 A from axis. The substitutional fraction of sulfur was 0.5 and 0.7 along and along direction, respectively. The depth profile of sulfur distribution measured by SIMS coincides with that of simulated vacancy depth profile associated with the sulfur implantation, rather than expected dopant distribution. These results suggest the redistribution of sulfur, and possible sulfur-residual damage (vacancy) coupling in the diamond crystal after the implantation.