RESUMO
Beryllium has been considered a potential alternative to magnesium as a p-type dopant in GaN, but attempts to produce conductive p-GaN:Be have not been successful. Photoluminescence studies have repeatedly shown Be to have an acceptor level shallower than that of Mg, but deep Be defects and other compensating defects render most GaN:Be materials n-type or semi-insulating at best. Previous reports use molecular beam epitaxy or ion implantation to dope GaN with Be, almost exclusively. Due to the high toxicity of Be organometallics, reports of GaN:Be by metal-organic chemical vapor deposition (MOCVD) have been largely absent. Here, we report a systematic study of growth of GaN:Be by MOCVD. All doped samples show the established UV band and yellow luminescence signature of GaN:Be, and growth conditions resulting in high-quality GaN with stable Be incorporation were established. Our results show that the MOCVD growth technique allows for Be incorporation pathways that have not been possible with previous growth methodologies and is highly promising in achieving p-type conductivity in GaN:Be.
RESUMO
A combined GaN 3D core-shell and planar pin structure is being developed and demonstrated to achieve the highest potential to increase energy transfer efficiency from the source (ηsrc) and power generated per cm2 (PGaN/cm2) in a betavoltaic (BV) device configuration. Physics-based Sentaurus TCAD and Monte Carlo N-Particle extended (MCNPX) software are employed to obtain the maximum ηsrc and PGaN/cm2 by a parametric study of device dimensions coupled with a 63NiCl2 source. Idealized structure dimensions are determined to be 2⯵m wide, 4⯵m tall GaN pin core-shell mesas, with 63Ni source conformally surrounding the structure with a 2⯵m gap for maximum efficiency of energy transfer. For maximizing power deposited (10⯵m mesa separation) a 3.75x increase in PGaN/cm2 at approximately half the activity density compared to a planar device is achieved for 4⯵m mesa height, with 5.82x improvement in ηsrc.
RESUMO
In this work we propose a novel method of immobilizing nucleic acids for field effect or high electron mobility transistor-based biosensors. The naturally occurring 5' terminal phosphate group on nucleic acids was used to coordinate with semiconductor and metal oxide surfaces. We demonstrate that DNA can be directly immobilized onto ZrO(2), AlGaN, GaN, and HfO(2) while retaining its ability to hybridize to target sequences with high specificity. By directly immobilizing the probe molecule to the sensor surface, as opposed to conventional crosslinking strategies, the number of steps in device fabrication is reduced. Furthermore, hybridization to target strands occurs closer to the sensor surface, which has the potential to increase device sensitivity by reducing the impact of the Debye screening length.