JP2536710B2 - Semiconductor laser - Google Patents

Description

Detailed Description of the Invention

[0001]

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to a semiconductor laser, and more particularly to a semiconductor laser for optical communication.

[0002]

2. Description of the Related Art FIG. 2 shows a sectional view of a conventional DC-PBH structure semiconductor laser. For example, JP-A-61-108183
See the official gazette. The shortest distance between the active layer 23 and the n-type current blocking layer 20 in the conventional structure (d in FIG. 2), in other words, the distance between the upper end of the active layer in the mesa portion and the mesa top is
It is 1.0 μm. The operating current at 85 ° C.-20 mW of this device was 90 mA or more.

[0003]

[Problems to be Solved by the Invention] Optical Ti
me Domain Reflectmeter (OT
In order to suppress heat generation from the active layer in the high power semiconductor laser for DR) or the semiconductor laser for environment resistance,
It is desirable that the drive current is as small as possible. The object of the present invention is to eliminate the drawbacks of the DC-PBH laser having such a conventional structure and to improve the high temperature operation characteristics, for example, 85 ° C.-20 mW.
In order to improve the characteristics of the high output laser or environment resistant laser, the drive current at the time is set to 80 mA or less.

[0004]

The present invention provides an n-type semiconductor substrate
DC-PBH (double-cha) formed on a plate
nnel planar burried hetero
In a semiconductor laser, the shortest distance between the n-type current block layer and the active layer is 0.1 to 0.3 μm.
Is characterized in that. This distance can be said to be the distance between the upper end of the active layer in the mesa portion and the mesa top, or the thickness of the p-type cladding layer on the active layer in the mesa portion. Further, the active layer is 57 A thick and has a 1.40 μm composition of InGaAs.
P-well and 100 A thick 1.13 μm composition InGa
The MQW is composed of an AsP barrier.

[0005]

In order to realize a low oscillation threshold current and a high differential efficiency in a semiconductor laser, it is important to reduce unnecessary leakage current flowing in a portion other than the active layer. The mechanism of the leakage current of the semiconductor laser having the conventional structure will be described with reference to FIG. The leakage current flowing on both sides of the active layer 23 is suppressed to some extent by the pnpn structure of the current block layer. However, from the p-type cladding layer 22 above the active layer 23 to the pnp
When the gate current (Ig in FIG. 2) flowing into the n-type p-type current blocking layer 21 increases, not only the oscillation threshold current and the differential efficiency deteriorate, but also the turn-on voltage decreases due to the thyristor action of the pnpn structure, and once turned on. Large leakage current begins to flow.

In the conventional structure, from the p-type clad layer 22 above the active layer 23 to the p-type current blocking layer 2 of the pnpn structure.
The generation of the gate current Ig flowing into 1 is unavoidable. Particularly at high temperature, the gate current Ig increases, the turn-on voltage decreases, and the oscillation threshold current is high, and the differential efficiency is further reduced.

Next, the result of analysis of the semiconductor laser structure shown in FIG. 3 by a two-dimensional optical device simulator will be shown. FIG. 3 is a diagram schematically showing the DC-PBH structure and shows the structure used for the simulation. 85 in FIG.
The current-light output characteristic in ° C is shown. n-type current blocking layer
The shortest distance from the active layer, d = 0.1 to 1.0 μm (d in FIG. 3), was calculated. FIG. 4B shows 85 ° C.-optical output 5
It is the d efficiency of the differential efficiency and the leakage current at mW. As can be seen from FIG. 4B, the leakage current reaches the minimum value at d = 0.2 μm, and is saturated below that.
Along with this, the differential efficiency is about 0.
The maximum value is reached around 1 to 0.3 μm. Therefore, d
It can be seen that the optimum value is 0.1 to 0.3 μm .

FIGS. 5 (a) and 5 (b) show the band edge energies 36 and 38 at the oscillation threshold current at a position near the upper part of the active layer 23 and parallel to the active layer 23 (AA 'in FIG. 3). This is a plot of Fermi levels 37 and 39. 5 (a) is d = 0.6 μm, and FIG. 5 (b) is d =
This is the case of 0.2 μm. The potential barriers against holes in the p-type cladding layer 22 above the active layer 23 and the p-type current blocking layer 21 beside the active layer 23 are almost zero in (a).
Although it is V, it is 0.15 V in (b). Further, from the difference between the pseudo-Fermi level 39 of holes and the band edge energy 38 of the valence band, in the case of (b), the hole concentration at this position of the p-type current blocking layer 21 is higher than that of (a). You can see that it is decreasing. That is, it is shown that the current from the p-type clad layer 22 above the active layer 23 to the p-type current block layer 21 beside the active layer 23 is less likely to flow in the case of (b) than in the case of (a). There is.

FIGS. 6A and 6B show a position of the current blocking layer near the edge of the active layer 23, which is perpendicular to the active layer 23 (B- in FIG. 3).
Electron 41 and hole concentration 4 at the oscillation threshold voltage in B ')
It is a plot of 0. In FIG. 6A, d = 0.6.
.mu.m, (b) are for d = 0.2 .mu.m. In the case of (b), the width of the region where the hole concentration is reduced near the upper pn junction of the p-type current block layer 21, and d
(0.2 μm) is about the same size,
In the case of (a), d (0.6 μm) is considerably larger. From these, it can be seen that in the case of d = 0.2 μm, the current from the p-type cladding layer 22 to the p-type current blocking layer 21 is less likely to flow electrostatically.

In the DC-PBH structure, a portion of the pnpn current blocking layer corresponding to the gate of the thyristor structure and the active layer portion are electrically connected, and the distance d is small (d = 0. 1 to 0.3 μm), the current flowing into the current blocking layer is spatially limited,
At the np current block layer interface, a depletion layer is formed with a voltage value around the oscillation threshold voltage. If the depletion layer is formed at the end of the p-type clad layer above the active layer, the wetting current is also suppressed electrostatically, and the wetting current is suppressed even if a voltage is applied, so that the thyristor is hard to turn on.

However, as in the conventional structure, d is 1.0 μm.
In the case of a large amount, a large current flows into the current blocking layer, and it is considered that the current is turned on. Further, as described with reference to FIGS. 5 and 6, from the comparison of d = 0.2 μm and d = 0.6 μm, a sufficient effect cannot be obtained at d = 0.6 μm. As shown in FIG. 4 , d = 0.1
It can be seen that the effect is remarkable when the thickness is 0.3 μm .

[0012]

DESCRIPTION OF THE PREFERRED EMBODIMENTS The present invention will be described with reference to the drawings. InGaAsP / I which is one embodiment of the present invention will be described with reference to FIG.
The nP-based MQW DC-PBH will be described. First, the n-InP buffer layer 2 (Si: 1.2 × 1) is formed on the n-InP (001) substrate 1 by using the MOVPE method.
0 ¹⁸ cm ⁻³ doped) 0.5 μm, 1.13 μm composition n-InGaAsP SCH layer 3 (Si: 1.2 × 10
¹⁸ cm ^-3 dope) 600 A (angstrom),
57A thick 1.40 μm InGaAsP well 4
(Non-doped) and 100 A thick 1.13 μm composition In
7-layer MQW composed of GaAsP barrier 4 (non-doped)
Structure 5, p-InP cladding layer 6 (Zn: 3.0 × 10
¹⁷ cm ⁻³ doped) 0.2 μm to grow an MQW wafer. Although the MOVPE method is used in this embodiment,
This is also possible in the LPE method or MBE method.

Next, 5.0 μ for forming the DC-PBH mesa
Double channels 7 are formed on the MQW wafer by a Br-methanol-based etchant using two 3.0 μm-wide stripe masks separated by m. And LPE
Method is used to bury and grow the double channel formed on the MQW wafer.

The final embedded shape is shown in FIG.
As shown in, the shortest distance d between the 1.4 μm wide active layer 8 and the n-type current blocking layer 9 is 0.2 μm. Compared with the conventional DC-PBH structure shown in FIG. 2, the shortest distance between the active layer 8 and the n-current blocking layer 9 (here, p-InP class) is used.
The thickness of the pad layer) is different.

The current-light output characteristics at 85 ° C. of the device manufactured based on the above steps are 75 mA at 20 mW, which is improved by 10 mA or more as compared with the conventional structure (about 90 mA or more). .

The present invention is based on the InGaAsP /
InP-based DC-PBH laser as well as AlGaAs /
It can also be applied to a GaAs DC-PBH laser.

[0017]

According to the present invention, a laser having excellent characteristics during high temperature operation can be obtained. In particular, the differential efficiency is excellent in high temperature operation, and the drive current can be reduced.

[Brief description of drawings]

FIG. 1 is an InGaAsP / InP-based MQWD of the present invention.
The figure which shows a C-PBH semiconductor laser. (A) is a figure which shows a MQW wafer, (b) is a figure which shows an embedding shape.

FIG. 2 is a cross-sectional view of a conventional DC-PBH structure semiconductor laser.

FIG. 3 is a structural schematic diagram for explaining the present invention.

FIG. 4 is a diagram for explaining the present invention, in which (a) is a current −
FIG. 6 is a diagram showing d dependence of light output characteristics, and FIG. 6B is a diagram showing d dependence of differential efficiency and leakage current.

5 is a view for explaining the present invention, which is taken along the line AA ′ in FIG.
It is a figure which shows the distance in a direction, and an energy level.

FIG. 6 is a view for explaining the present invention, and is BB ′ of FIG.
It is a figure which shows the distance and carrier concentration in a direction.

[Explanation of symbols]

1 n-InP (001) substrate 2 n-InP buffer layer 3 1.13 μm composition n-InGaAsP SCH layer 4 1.40 μm composition n-InGaAsP well 5 1.13 μm composition InGaAsP barrier 6 7-layer MQW structure 7 p-InGaAsP clad Layer 8 Double channel 9 Active layer 10 p-InGaAsP cap layer 11 p-InP layer 12 n-InP current block layer 13 p-InP current block layer 14 p-InP clad layer 15 n-InP buffer layer 16 Electrode 17 SiO ₂ film , 18 p-InGaAsP cap layer 19 p-InP layer 20 n-InP current blocking layer 21 p-InP current blocking layer 22 p-InP clad layer 23 active layer 24 n-InP buffer layer 25 n-InP substrate 26 electrode 27 electrode 28 p-In GaAsP cap layer 29 p-InP layer 30 n-InP current blocking layer 31 p-InP current blocking layer 32 P-InP clad layer 33 active layer 34 n-InP substrate 35 electrode 36,38 band edge energy 37,39 pseudo-Fermi level 40 hole concentration 41 electron concentration

Claims (2)

(57) [Claims] 1. A DC-PB formed on an n-type semiconductor substrate.
H (double-channel planar bu
In a semiconductor laser with a ried heterostructure, the shortest distance between the n-type current block layer and the active layer
A semiconductor laser having a separation of 0.1 to 0.3 μm . 2. The active layer is a multi-quantum well (MQW) comprising a 1.40 .mu.m composition InGaAsP well of 57 angstrom (A) thickness and a 1.13 .mu.m composition InGaAsP barrier of 100 A thickness. Semiconductor laser.