Cardiff University | Prifysgol Caerdydd ORCA
Online Research @ Cardiff 
WelshClear Cookie - decide language by browser settings

Vertical leakage mechanism in GaN on Si high electron mobility transistor buffer layers

Choi, F S, Griffiths, J T, Ren, Chris, Lee, K B, Zaidi, Z H, Houston, PA, Guiney, I, Humphreys, C J, Oliver, R A and Wallis, D J 2018. Vertical leakage mechanism in GaN on Si high electron mobility transistor buffer layers. Journal of Applied Physics 124 (5) , 055702. 10.1063/1.5027680

[img]
Preview
PDF - Accepted Post-Print Version
Download (1MB) | Preview

Abstract

Control of leakage currents in the buffer layers of GaN based transistors on Si substrates is vital for the demonstration of high performance devices. Here we show that the growth conditions during the metal organic chemical vapour deposition (MOCVD) growth of the graded AlGaN strain relief layers (SRL) can significantly influence the vertical leakage. Using scanning capacitance microscopy (SCM), secondary ion mass spectrometry (SIMS) and transmission electron microscopy (TEM) we investigate the origins of leakage paths and show that they result from the preferential incorporation of oxygen impurities on the side wall facets of the inverted hexagonal pyramidal pits which can occur during the growth of the graded AlGaN SRL. We also show that when 2D growth of the AlGaN SRL is maintained a significant increase in the breakdown voltage can be achieved even in much thinner buffer layer structures. These results demonstrate the importance of controlling the morphology of the HEMT buffer layer as even a very low density the leakage paths identified would provide leakage paths in large area devices.

Item Type: Article
Date Type: Publication
Status: Published
Schools: Engineering
Publisher: AIP Publishing
ISSN: 0021-8979
Date of First Compliant Deposit: 24 July 2018
Date of Acceptance: 18 July 2018
Last Modified: 13 Oct 2018 07:28
URI: http://orca.cf.ac.uk/id/eprint/113414

Actions (repository staff only)

Edit Item Edit Item

Downloads

Downloads per month over past year

View more statistics