Abstract

The first GaN was grown by hydride vapor phase epitaxy (HVPE) more than 30 years ago, and now thick, large-area HVPE layers are being proposed as substrates for GaN devices. Temperature-dependent Hall-effect, photoluminescence (PL), secondary ion mass spectroscopy, and positron annihilation measurements by different groups over the last few years have identified the dominant donors in HVPE GaN as Si and O, and the dominant acceptor as the Ga vacancy. Another potential donor, the N vacancy V N, has been shown to have a larger activation energy (∼70 meV) than that of Si and O (∼30 meV), and thus to contribute little to the conductivity of almost any as-grown sample. However, V N is also an electron trap, with an effective activation energy of about 160 meV. Several other traps appear regularly in HVPE GaN, and one such trap, at 1.0 eV, may be associated with the N interstitial. However, a trap at 0.60 eV is overall the most common trap in GaN grown by HVPE or any other method. We have recently found a correlation between the 0.60-eV trap and the well-known blue band, at 2.8 eV, but unfortunately the blue band itself has not yet been positively identified. Usually, the 0.60-eV trap is isolated, whereas the 1.0-eV trap is often associated with dislocations.

Full Text
Published version (Free)

Talk to us

Join us for a 30 min session where you can share your feedback and ask us any queries you have

Schedule a call