research review
rhombus4
Koreans slam Auger as the
determined the non-radiative carrier lifetime
from time-resolved photoluminescence
measurements. Lead author Han-Youl Ryu
primary cause of LED droop
says that this type of photoluminescence
measurement could not be used on the LED
chips. “Instead, we performed the
A Korean partnership has joined the To slash the number of fit parameters to just measurement on an LED wafer with similar
controversial debate over the origin of LED one, the researchers assumed an injection layer structures, and estimated the non-
droop. It claims that its sophisticated efficiency of 100 percent. radiative carrier lifetime to be about 50 ns.”
approach to curve fitting of experimental
data demonstrates that Auger recombination In addition, they exploited the fact that the The researchers then deduced an Auger
only makes a small contribution to droop, gradient of the peak of the internal quantum recombination rate of 10
-27
cm
6
/s, by
the decline in GaN LED external quantum efficiency as a function of drive current is assuming a maximum internal quantum
efficiency at higher current densities. zero at the maximum value for efficiency. efficiency of 79 percent, and determining the
This allowed the construction of new value for the SRH recombination rate from
This team from Inha University and equations that related the rates for SRH the non-radiative carrier lifetime
Hanyang University arrived at this recombination, bimolecular radiative measurements.
conclusion by manipulating the standard recombination, and Auger recombination to
rate equations, and reducing fit parameters the quantum well thickness and the This value for the Auger recombination rate
from three to just one. Selecting a suitable maximum values for current density and is at least 1000 times higher than that
value for the Shockley-Read-Hall (SRH) internal quantum efficiency. quoted by other groups that claim Auger is
non-radiative recombination coefficient the primary cause of droop, and implies that
generated a value for Auger recombination Experimental results were obtained by taking an alternative non-radiative mechanism is
(the non-radiative interaction of an electron, a 460 nm LED produced by a domestic needed to account for declining LED
a hole, and a third carrier) that is supplier, and measuring its internal quantum efficiencies at high drive currents.
substantially different from that quoted efficiency as a function of drive current. This
by those backing Auger as the cause of supplier also provided a value for the H.-Y Ryu et al. (2009) Appl. Phys. Lett. 95
LED droop. quantum well thickness, and the researchers 081114
Patterned
etching. A 30 nm thick GaN buffer was then performances were inferior to that on c-
deposited onto this substrate by MOCVD at sapphire,” says Narihito Okada.
sapphire creates
460
0
C, followed by the growth of GaN at
temperatures ranging from 900
0
C to They now want to understand why the
1000
0
C. Lower deposition temperatures performances of LEDs grown on the semi-
semi-polar GaN
created an undesirable mixture of (1122) and polar and non-polar planes are inferior to
(112 0) GaN, but growth at 1000
0
C those grown on the c-plane. However, the
produced (1122) only. Cathodoluminscence primary goal of this research team is to grow
Researchers from Yamaguchi University, measurements revealed a threading high-quality GaN layers with defect density
Japan, have produced semi-polar (1122) dislocation of more than 3 x 10
8
cm
-2
in an below 10
8
cm
-2
.
GaN on a maskless r-plane patterned n-doped layer.
sapphire surface. Their advance could help N. Okada et al. (2009) Appl. Phys. Express
to spur the development of green-emitting The researchers have produced LEDs on 2 091001
devices that are grown on the semi-polar this material with the processes that were
planes of GaN. These faces enable the initially developed for device fabrication on
fabrication of epistructures that are not c-plane sapphire. “However, the
hampered by strong internal electric fields,
and they facilitate the growth of InGaN
layers with a high enough indium content for
green emission.
This Japanese team is not the first to grow
semi-polar GaN on patterned, foreign
substrates. Silicon has also been used, but
this has the downside of absorbing some of
the light that is generated by the device.
Fabrication of a series of semi-polar GaN Fig1: Photolithography and etching forms
films began with the creation of micron-sized grooves in a sapphire substrate with a
grooves in sapphire via a combination of depth of up to 1 µm, and groove and Fig2: MOCVD growth of GaN at 1000
0
C
photolithography and fluorine-based terrace widths of 3 µm. Subsequent creates semi-polar material with a high
inductively coupled plasma reactive-ion growth of GaN forms a semi-polar film crystal quality
42
www.compoundsemiconductor.net October 2009
Page 1 |
Page 2 |
Page 3 |
Page 4 |
Page 5 |
Page 6 |
Page 7 |
Page 8 |
Page 9 |
Page 10 |
Page 11 |
Page 12 |
Page 13 |
Page 14 |
Page 15 |
Page 16 |
Page 17 |
Page 18 |
Page 19 |
Page 20 |
Page 21 |
Page 22 |
Page 23 |
Page 24 |
Page 25 |
Page 26 |
Page 27 |
Page 28 |
Page 29 |
Page 30 |
Page 31 |
Page 32 |
Page 33 |
Page 34 |
Page 35 |
Page 36 |
Page 37 |
Page 38 |
Page 39 |
Page 40 |
Page 41 |
Page 42 |
Page 43 |
Page 44