Semiconductor device

Details Semiconductor device Semiconductor device

2000-07-05

2003-10-07

Leung, Quyen (Department: 2828)

Coherent light generators

Particular active media

Semiconductor

C372S046012

Reexamination Certificate

active

06631150

ABSTRACT:

BACKGROUND OF THE INVENTION
1. Field of the Invention
This invention relates to a semiconductor device and particularly, but not exclusively, to a semiconductor laser device that emits visible radiation in the wave-length range 630 nm to 680 nm. The laser device may be of the edge-emitting or of the surface-emitting type.
2. Description of the Related Art
Laser devices or laser diodes (LDs) fabricated in the (Al,Ga,In)P material system which emit visible light in the 630 nm-680 nm wavelength range are becoming increasingly important components of professional and consumer products. For example, it is envisaged that the Digital Video Disc (DVD) system will employ a 635 nm-650 nm wavelength LD capable of delivering up to 30 mW output power up to a temperature of 60° C. The next generation of semiconductor lasers will need an even greater maximum power output up to a higher (eg. 70° C.) operating temperature.
By the (Al,Ga,In)P system is meant the family of compounds having the general formula (Al
x
Ga
1−x
)
1−y
In
y
P, where both x and y are between 0 and 1. One particular advantage of this semiconductor system is that it is lattice-matched to a Gabs substrate when the indium mole fraction, y, is equal to 0.48.
A principal limitation of current (Al,Ga,In)P laser diodes is that they are incapable of operating for long periods (or with a sufficiently low threshold current) at the highest specified operating temperature. It is generally believed that this is caused by electron leakage from the active region of the device into the surrounding optical guiding region and subsequently into the p-type cladding region.
The generic structure of a separate confinement laser structure intended to generate light at 630-680 nm will now be described with reference to FIGS.
1
(
a
) and
1
(
b
).
FIG.
1
(
a
) is a schematic band structure of a separate confinement laser structure fabricated in the (Al,Ga, In)P system. It consists of an n-doped (Al
0.7
Ga
0.3
)
0.52
In
0.48
P cladding region
1
, an (Al
0.5
Ga
0.5
)
0.52
In
0.48
P optical guiding region
2
,
4
, a GaInP quantum well active region
3
disposed within the (Al
0.5
Ga
0.5
)
0.52
In
0.48
P optical guiding region, and a p-doped (Al
0.7
Ga
0.3
)
0.52
In
0.48
P cladding regions. A p-type contact layer (not shown in FIG.
1
(
a
)) may be provided on the p-type cladding region
5
, and an n-type contact layer (also not shown in FIG.
1
(
a
)) may be provided on the n-type cladding region
1
. Optical transitions giving rise to laser action in the quantum well active region
3
of the laser diode originate from electrons in the &Ggr;-band in the GaInP quantum well active region.
The terms &Ggr;-band and X-band as used herein refer to symmetry points it the Brillouin zone and are standard terms in solid state physics, see for example R. A. Smith “Semiconductors”, (Cambridge University Press, 1978). The terms &Ggr;-minimum and X-minimum refer to the minimum energy level of the &Ggr;-band and the X-band, respectively.
The minimum energy in the conduction band of (Al,Ga,In)P is a function of the aluminium content. There is a crossover from a &Ggr;-band minimum to an X-band minimum at an aluminium concentration of about 0.55.
FIG.
1
(
b
) illustrates a similar laser structure to that of FIG.
1
(
a
), but the cladding regions
1
,
5
are formed of AlInP rather than (Al
0.7
Ga
0.3
)
0.52
In
0.48
P in order to increase the potential barrier confining &Ggr;-electrons in the optical guiding region
2
.
4
.
In FIGS.
1
(
a
) and
1
(
b
) the active region
3
is shown as containing a single quantum well layer. As is well known, however, the active region
3
may alternatively consist of a plurality of quantum well layers, with each quantum well layer being separated from an adjacent quantum well layer by a barrier layer.
Degradation of semiconductor lasers has been a major problem in developing commercial devices. AlGaAs/GaAs lasers having a lasing wavelength of around 0.85 &mgr;m were initially developed in the 1970s, but early examples of these lasers degraded quickly during use and, as a result, had a low lifetime and were unsuitable for commercial applications. It took a considerable time to overcome the significant degradation problems involved with these lasers. One of the main factors in the degradation of AlGaAs/GaAs lasers was crystal defects introduced during the crystal growth and fabrication processes These defects are known as “dark lines” defects, and give rise to a localised region of high non-radiative recombination centres which have high optical absorption. The problem of these dark line defects was eventually overcome by improving the structural quality of the GaAs, and long lifetime AlGaAs,GaAs lasers are now commercially available. M. Fukuda reports, in “Reliability and Degradation of Semiconductor Lasers and LEDs” ISBN 0-89006-465-2, that AlGaAs/GaAs lasers having a life time greater than 10,000 hours are now commercially available.
In order for (Al,Ga,In)P lasers to be commercially successful, these lasers must have a life time comparable with that of AlGaAs/GaAs lasers.
At present, wide bandgap phosphide lasers operating in the visible spectrum at a wavelength of about 650 nm display a severe degradation problem. Although the lifetime of low power phosphide lasers is approximately 10,000 hours which is satisfactory for commercial purposes, a typical lifetime of a high-power phosphide laser is only about 5,000 hours which is not commercially acceptable. Furthermore, it is necessary to anneal the lasers in order to obtain these lifetimes, and lasers that are not annealed have much shorter lifetimes.
The degradation problem is particularly serious for lasers fabricated using molecular beam epitaxy (MBE). At present, phosphide laser structures that are grown by MBE have to be thermally annealed in order to improve their reliability and to decrease the threshold for laser operation. It is presumed that the annealing process removes (or at least moves) some of the non-radiative recombination centres in the material. It is, however, undesirable to carry out an annealing step. One common p-type dopant for the p-type cladding region is beryllium, and if a beryllium-doped laser device is annealed beryllium can diffuse from the p-type cladding region into the active region. Such diffusion will degrade the performance of the laser device, and may also lower the yield of the manufacturing process.
M. Jalonen et al., report, in “Applied Physics Letters” Vol. 71 No. 4 p.479 (1997), laser oscillation thresholds for lasers grown by the MOCVD process in the range 0.2-0.4 kA/cm
2
compared to thresholds for material grown by MBE of >1 kA/cm
2
, for an emission wavelength of 680 nm. It is possible that the poor degradation characteristics and higher laser thresholds for lasers grown by MBE arise because these materials have significantly more non-radiative recombination centres (or defects) than material grown by MOCVD. It
1
s desirable to remove non-radiative defects in order to improve the reliability of phosphide lasers, and to lower their lasing threshold.
One particular type of defect that could well be responsible for the severe degradation and high laser thresholds in wide bandgap phosphide lasers is the anti-phase domain defect (APD). These defects arise in phosphide materials as a consequence of ordering occuring in the crystal structure. APDs give rise to a region of high non-radiative recombination, and of high optical absorption. Thus, the higher is the density of APDs, the higher will be the optical absorption and the higher will be the threshold for laser oscillation. A high density of domain boundaries also results in fluctuations in the refractive index of the material, and these fluctuations will cause additional photon loss and will hence decrease the quantum efficiency of the laser. The presence of APDs may also lower the yield of the manufacturing process.
C. Geng et al report, in “Journal of Crystal Growth” Vol. 170, page 418 (1997), that the domain structure can be responsible for the worsened laser per

Affiliated with

Also associated with

Rating

Semiconductor device does not yet have a rating. At this time, there are no reviews or comments for this patent.

If you have personal experience with Semiconductor device, we encourage you to share that experience with our LandOfFree.com community. Your opinion is very important and Semiconductor device will most certainly appreciate the feedback.

Rate now

Comments { 0 }

Profile ID: LFUS-PAI-O-3114845