|M.Sc Student||Yom-Tov Nir|
|Subject||Accurate Electrical Measurements of Doped Diamond|
|Department||Department of Physics||Supervisors||Professor Emeritus Rafael Kalish|
|Dr. Yuval Yaish|
|Full Thesis text - in Hebrew|
Diamond is a semiconductor of extraordinary electrical and physical properties, which make it a promising material for certain uses in the electronics industry. Among its unique properties are wide band gap, a very high chemical inertness and high heat conductance.
In order to take advantage of these unique properties, dopants with reasonable activation energies are needed. Boron is a possible p-type dopant, while no known n-type dopant with a reasonable activation energy exists.
Partial success in converting a Boron doped diamond to an n-type diamond by hydrogenation was reported. According to the reports, the p-type to n-type conversion depends on the local crystal quality, and might change from one region to the other on the same diamond layer. Measurements for diamonds showing this local p to n conversion were done without electrically isolating the measured region, thus it is hard to estimate the error caused by the lack of proper isolation.
The present work deals with the creation and characterization of mesa structures in diamond using RIE, by adapting a common technology used in Silicon microelectronics and checking the effect of an Oxygen RIE on the electrical properties of a Boron doped diamond.
A set of experiments was done in order to characterize the carrier concentration of a diamond layer before and after mesa creation using Hall effect measurements at non ideal Van Der Pauw geometry. The results of these experiments show that due to the inclusion of conductive surroundings, it is possible to achieve an artificial Hall sign reversal. When repeating the experiments with a better contacts configuration, the artificial Hall sign reversal disappears.
A finite element simulation (Comsol) was used to simulate the experimental results. It was shown that non uniformity of the sample, in addition to a non ideal Van Der Pauw configuration of the contacts, can indeed lead to an artificial Hall sign reversal. However, it was impossible to reproduce the sign reversal when using the typical carrier concentrations and mobilities of the diamond layers that were used in the experiments.
This work shows that the existence of conductive surroundings can have a tremendous effect on the electrical measurements of the carriers at the diamond layer. This effect is apparently enhanced by RIE. A drop in the carrier mobility, and a rise in the carrier concentration were seen in the simulation, while the Hall voltage sign reversal was not exactly reproduced in the simulation.