Conference Summary International Symposium on Blue
Laser and Blue LEDs
The International symposium on Blue Laser and Blue LEDs
was held at the Chiba University Convention Center on
March 5-7, 1996 and was chaired by Professor A.
Yoshikawa of Chiba University. The conference amounted
to a showdown between Gallium Nitride versus Zinc
Selenide based optoelectronic materials and devices.
The conference was well attended with over 130 papers
(62 papers containing Nitride studies) and more than 400
attendees. A large sake drinking party insured
competitive discussions amongst the attendees. No clear
winner for the blue laser diode emerged yet with both
materials showing progress. Sony Corporation annouced
improved reliability with ZnSe laser diodes displaying
101.5 hour CW lifetimes (A. Ishibashi) and two GaN
research groups, S. Nakamura (Nichia Chemical
Industries) and I. Akasaki and H. Amano et al. (Meijo
University), now reporting GaN lasers operating pulsed at
room temperature. However, based on superior
reliability and efficiency the GaN LEDs appear to be
quickly gaining market acceptance for full-color outdoor
displays. The major problems remaining for CW
operation in the Nitrides appears to be related to poor p-
contacts (heating), high threshold currents, and high
quality mirror formation. The role of dislocations in GaN
is still unclear at present.
With this review intended for the MRS Nitride Internet
Journal I will highlight some of the interesting GaN
papers presented. My apologies for not summarizing
more presentations which were equally important.
Lasers
S. Nakamura reported on improvements made in blue
lasers at Nichia Chemicals Industries. Instead of 26
quantum wells the newest lasers are using 7 InGaN
quantum wells. Threshold current density was 4kA/cm2
and Vthreshold has been reduced to 20V. The key layers
from a MOCVD growth perspective are the
implementation of a 100nm In0.1Ga0.9N layer below the
Al0.12Ga0.88N lower cladding layer which is reported to
prevent cracking in the AlGaN. Also, a 20nm Al0.2Ga0.8N
layer is employed directly above the MQW active region
to suppress In desorption. GaN lasers were grown on
both a-face and c-face sapphire(0001) Nakamura's most
surprising result was that he could obtain cleaved facets
when he grew on the A-face of sapphire. Apparently two
of the R-faces are parallel when using A-face sapphire.
Mirrors are then subsequently HR coated (60-70%) to
reduce threshold current density. S. Nakamura reported
that he believes that heat generation is limiting CW
operation. His talk concluded with a demonstration of
the MQW laser, which was dim but still displayed a
coherent blue-violet beam. A second group headed by I.
Akasaki and H. Amano at Meijo University also has
reported lasing at room temperature under pulsed
conditions. The shortest wavelength lasing at 376nm in
an InGaN/GaN DH structure was reported with a threshold
of 16V. A narrow p-contact layer was created by low
energy electron beam irradiation (LEEBI) and is thought
to result in the lower threshold voltage. Details of the
lasing structure will be published shortly.
Optical Pumping
Optical pumping of GaN was reported on by at least three
groups. I. Akasaki and H. Amano of Meijo University grew
a AlGaN/GaN/InGaN QW structure on a SiC substrate
which exhibited the lowest reported threshold value of
27kW/cm2. Their dislocation density on SiC was
reported to be 2E+8cm-2. An interesting paper on
vertical cavity stimulated emission from optically
pumped GaN was given by K. Domen et al. of Fujitsu
Laboratories (Atsugi). This paper was notable because
they reported an extremely large threshold optical gain
of 15,000cm-1 from a gain analysis of their structure.
Their structure consisted of a GaN epilayer grown on a
spinel substrate (MgAl2O4). The mirrors in their cavity
were formed at the GaN/air interface (22%) and
GaN/substrate interface (4.7%). An optically pumped
vertical cavity surface emitting laser was also reported
on by J. Redwing of ATMI (Danbury, MA). Stimulated
emission was observed from AlGaN/GaN double
heterostructure on sapphire. Their cavity also did not
employ high reflectivity mirrors.
Substrates and Dislocations
Growth on small bulk substrates of GaN up to 5mm wide
was reported by S. Porowski et al of the Polish Academy
of Sciences (Warsaw, Poland). The small bulk crystals
were grown from a liquid Ga solution under 20kbar of N2
overpressure. Double crystal X-Ray diffraction exhibited
20-40 arcsec and homoepitaxial films deposited on these
substrates showed reduced dislocation densities of 1E+6
to 1E+7cm-2. GaN growth on a variety of alternative
substrates was also reported. A. Kuramata (Fujitsu Labs)
obtained good single crystal GaN and cleaved facets when
growing on Spinel (MgAl2O4). J.F. Schetzina of North
Carolina State University reported on MBE of GaN on
GaN/SiC substrates which were supplied by CREE
Research Inc. The dislocation density of MBE is good and
essentially replicated the density observed for the
underlying GaN buffer layers which was approximately
1E+8cm-2. A very interesting report of the GaN
microstructure was reported on by F.A. Ponce of Xerox
(Palo Alto, CA). F. A. Ponce reported that the
microstructure of GaN is columnar in nature and is
bounded by low angle grain boundaries with both edge and
screw dislocations. The individual grains are slightly
twisted with respect to one another. The dislocation
density and microstructure seem to originate from the
low temperature GaN buffer layer and nitridation
procedure. K. Uchida et al. of Hitachi (Tokyo, Japan)
reported that the nitridation which takes place at 1050 C
creates an AlO1-xNx amorphous layer upon which the fine
grain GaN buffer layer is subsequently deposited at 550C.
Steven DenBaars
