Objective
- This action aims to grow by metalorganic vapour phase epitaxy device quality (A1, Ga)N alloy semiconductors using complementary approaches from the two partner laboratories. The final goal is to fabricate a nitride based UV detector, basically a p-n junction.
- At the end of the first year, state of the art undoped GaN has been grown by MOVPE on sapphire, i.e. design of a proper buffer for GaN on sapphire, appropriate growth conditions. In depth characterisation including low temperature photoluminescence, double X-ray diffraction, SEM and TEM has been achieved. Even though GaN based devices have been fabricated by epitaxial growth on sapphire, GaN is highly dislocated. High Resolution Transmission Electron Microscopy (HRTEM) was used to investigate the nanostructure of GaN layers. An in depth evaluation of defects was achieved, it appears that GaN layers on sapphire exhibit about 1010 cm-2 dislocations and 108 cm-2 nanopipes.
- During the second year, improvement of the electronic quality of undoped GaN has been achieved using a better buffer structure. The effort was concentrated on in depth structural characterisation and p-doping of GaN. Also n-doping was achieved, thus making possible the first p-n structures, thereby paving the way towards a U.V. detector. The implementation at CRHEA of a Reactive Ion Etching (RIE) facility allowed to fabricate GaN based devices. The first p-n junctions were made and tested.
- The University of Monastir implemented of a hot filament for the thermal cracking of NH3. In the meantime, t-butylamine (t-NH2) as a new nitrogen precursor has been tested.
- From our experiences compared to published data from the scientific community, alternative nitrogen source and hot wire activation of NH3 remain open options. The University of Monastir carried out experiments using t-NH2. When t-NH2 was replacing NH3 as nitrogen precursor, in basically the same experimental growth conditions, the layers contain an huge amount of carbon, consequently, t-NH2 is not suitable for the epitaxial growth of GaN.
- Presently, the growth of good quality GaN epitaxial layers on sapphire is achieved. The development of an alternative nitrogen source has been reached, it seems that NH3 remains the best nitrogen precursor for the growth of GaN.
- For the design of any device based upon GaN/(A1, Ga)N heterostructure, the knowledge of band offsets is of critical importance. Therefore, calculations have been carried out to get good estimate of the band edge discontinuities that occur within heterodevices based on III-V nitride alloys. A semi-empirical approach, based on er ab initio computations or experimental results, was used. Henceforth, the band edge discontinuities lie in the order of 1 eV for the valence bands of AIN and GaN. However, it is still necessary to refine the computation to get more acceptable figures.
- Several nitrogen sources have been tested to grow GaN by Metal Organics Vapour Phase Epitaxy (MOVPE). Among different nitrogen sources (plasma NH3 or N2); NH3 triethylamine, NH3 provided the best results;
- The research efforts were put in four directions :
- proper design of a buffer layer (CRHEA);
- Mg p-doping (CHREA);
- design of a nitrogen source using a W hot filament (Monastir);
- use of t-butylamine (Monastir).
- Before fabricating any minority carrier device, the residual n-doping of as grown GaN should be carefully controlled. The origin of the residual n-conduction in GaN remains not clear, it could be either nitrogen vacancies or some impurities or even Ga interstitial. To avoid compensation in p-type sample, the residual n-conduction should be reduced down to the low 1016 cm-3 . This is related to the deep nature of p-dopant (~200meV) leading to very low ionisation efficiency.
Topic(s)
Data not availableCall for proposal
Data not availableFunding Scheme
CSC - Cost-sharing contractsCoordinator
06560 Valbonne
France