% IMPORTANT: The following is UTF-8 encoded. This means that in the presence
% of non-ASCII characters, it will not work with BibTeX 0.99 or older.
% Instead, you should use an up-to-date BibTeX implementation like “bibtex8” or
% “biber”.
@INPROCEEDINGS{Pecz:827190,
author = {Pecz, Bela and Kovacs, Andras and Dunin-Borkowski, Rafal
and Yakimova, Rositza and Heuken, Michael},
title = {{N}itride layers grown on patterned graphene/{S}i{C}},
address = {Weinheim, Germany},
publisher = {Wiley-VCH Verlag GmbH $\&$ Co. KGaA},
reportid = {FZJ-2017-01388},
pages = {630 - 631},
year = {2016},
comment = {European Microscopy Congress 2016: Proceedings},
booktitle = {European Microscopy Congress 2016:
Proceedings},
abstract = {Self-heating of high power GaN devices during their
operation is a major drawback that limits the performance.
Integration of sheets with very high thermal conductivity
material could help in this matter. After some unsuccessful
GaN growth experiments carried out directly on graphene, we
succeeded to grow nitride layers on patterned
graphene/6H-SiC by Metalorganic Chemical Vapour Deposition
(MOCVD). The growth is similar to the well-known Epitaxial
Lateral Overgrowth method in which the graphene buried
stripes are overgrown laterally from the window regions,
where AlN could grow on bare SiC with epitaxy. An AlN buffer
layer was first deposited on patterned graphene/6H-SiC
surface followed by a deposition of ~ 300 nm thick
Al0.2Ga0.8N and ~ 1.5 µm thick GaN layer. The AlN buffer
deposited onto the graphene stripe was grown in a 3D way
(Fig.1a). The heterostructure was studied using
aberration-corrected transmission electron microscopy (TEM)
methods in combination of electron energy-loss X-ray
spectroscopy (EDXS) and electron energy-loss spectroscopy
(EELS). TEM specimens were prepared using both conventional
and focused ion beam methods.The most surprising details of
this study is the appearance of the AlN/GaN superlattices,
which were formed in a self-organised way over the buffer
layer. Instead the ternary AlGaN we have superlattice (Fig.
1.b and c) in which the thickness of the AlN/GaN is
determined by the available elements from the Al0.2Ga0.8N
which we wanted to grow. The control sample (without
graphene) showed a much more flat AlN buffer and a ternary
Al0.2Ga0.8N on that without any phase separation. EDXS
mapping and also superlattice reflections show however,
clearly the complete phase separation in the case the
nitride layers are grown on graphene. We suppose, that some
excess carbon induced the phase separation.The detailed TEM
studies revealed the AlN nucleation directly on SiC and
lateral overgrowth of graphene island as shown in Fig.2a.
The high resolution image in Fig.2.b shows three layers of
graphene and the AlN that is in epitaxy with SiC. Both
interfaces are sharp and no interdiffusion of the elements
are observed according to the Si, C (not shown) and Al maps
in Fig. 2c The results show that high quality GaN layer over
graphene/SiC can be grown with MOCVD that can serve as
templates for high power GaN devices.},
month = {Aug},
date = {2016-08-28},
organization = {16th European Microscopy Congress (EMC
2016), Lyon (France), 28 Aug 2016 - 2
Sep 2016},
cin = {PGI-5 / ER-C-1},
cid = {I:(DE-Juel1)PGI-5-20110106 / I:(DE-Juel1)ER-C-1-20170209},
pnm = {143 - Controlling Configuration-Based Phenomena (POF3-143)},
pid = {G:(DE-HGF)POF3-143},
typ = {PUB:(DE-HGF)8 / PUB:(DE-HGF)7},
doi = {10.1002/9783527808465.EMC2016.6338},
url = {https://juser.fz-juelich.de/record/827190},
}