Investigation of Metal Contacts to Group III-nitride Semiconductors

Open Access
- Author:
- Glickstein, James Jay
- Area of Honors:
- Materials Science and Engineering
- Degree:
- Bachelor of Science
- Document Type:
- Thesis
- Thesis Supervisors:
- Suzanne E Mohney, Thesis Supervisor
Ivica Smid, Faculty Reader
Judith A Todd Copley, Faculty Reader
Suzanne E Mohney, Thesis Honors Advisor - Keywords:
- semiconductor
III-V
III-Nitride
GaN
InN
wurtzite
crystal polarity
metal contact - Abstract:
- Due to recent advancements in growth techniques and the understanding of its electronic band structure, the group III-N semiconductor InN is poised to become a material of choice for many future electronic device applications. Before this is possible, the interaction of InN with thin metal films must be better understood in order to design electrical contacts with desirable behaviors. In this study, the thermal stability of various metal contacts on InN has been predicted using calculated ternary phase diagrams. Thermodynamic data from the literature was used to make phase-stability calculations between InN and the following metals: Ta, Mo, Re, Ni, Pd, Pt, and Au. Periodic trends were observed in the predicted contact behavior based on the phase diagrams. Late transition metals were predicted to react with InN at room temperature or after annealing at 400°C to form stable metal-In phases. Middle transition metals were predicted to be stable with InN even after high-temperature annealing. Early transition metals were predicted to react at room temperature under N2 gas to form stable metal-N phases. While these thermodynamic predictions will allow future researchers to know what contact behavior to expect for certain metals, experiments must be performed to observe the effects of reaction kinetics and other variable factors that may greatly affect the expected behavior in practice. One of the most important of these factors that must be considered for designing contacts to InN is the effect of III-N wurtzite polarity on contact phase formation. Due to the lack of availability of InN samples and the identical wurtzite crystal structure of GaN, samples of N- and Ga-polar GaN were used to experimentally determine the effect of different polarity on Ni contact formation at different annealing temperatures. The (0001) Ga-polar samples were shown by X-ray diffraction to react with an 80 nm sputtered Ni film at annealing temperatures above 800°C to produce Ni3Ga, consuming the Ni layer in the process. The (000-1) N-polar samples also reacted with an identically deposited Ni film at annealing temperatures above 800°C, this time forming Ni5Ga3. An oxide, Ga2O3, is also thought to have been produced in the reaction on the N-polar sample. These results indicate that a reaction occurs in both cases above 800°C and is more extensive on the N-polar face. The calculated results obtained in this thesis outline the expected reaction of various contact metals on InN. These guidelines will allow researchers to minimize the experiments necessary to find a suitable contact metal for a given application. The experimental results with GaN indicate that reactions will likely proceed faster on the N-polar surface of InN compared to the In-polar surface. This preliminary study may act as a guideline for future work with group III-N contact metallurgy focused on the importance of crystal polarity in predicting phase formation.