JPH02291188A - Semiconductor laser device - Google Patents

Abstract

PURPOSE:To improve the high-frequency characteristics of a semiconductor laser by a method wherein the part other than the vicinity of the luminous region of the laser is covered with a high-resistance semiconductor layer and at the same time, an active layer is formed into a buried structure. CONSTITUTION:Almost half of an N-type InP semi-insulative substrate 1 with an InP high-resistance layer 9 laminated on it is etched using a silicon nitride film 11 as a mask. Then, an n-type InP buffer layer 2, an InGaAsP active layer 3, a p-type InP clad layer 4 and a p-type InGaAsP contact layer 5 are formed in order on the stepped base of the surface of an inverse mesa using the film 11 as a mask by an MOCVD device. Then, after the film 11 is removed using a buffered hydrofluoric acid, a silicon nitride film 12 is anew formed and a switching is performed until the layer 2 is exposed using this film 12 as a mask. Then, after the layer 3 is selectively etched using a sulfuric acid etchant, an n-type InP mass transport buried layer 20 is formed. After the film 12 is removed using a buffered hydrofluoric acid, a p side electrode 7 is deposited. Moreover, an n side electrode 8 is formed on the side of the substrate 1. Thereby, a signal current is almost all fed to the layer 3 up to its high-frequency band and a semiconductor laser having superior high-frequency characteristics is obtained.

Description

[Detailed description of the invention] [Industrial application field] The present invention relates to a semiconductor laser device.

[Conventional technology]

Optical semiconductor devices such as light emitting diodes and photodiodes using 111-V group compounds are used as key devices in optical fiber communications, optical information processing, and the like. In particular, semiconductor lasers are the most important element in the development and practical implementation of long-distance, high-capacity optical fiber communication systems, and efforts have been made in recent years to improve their speed.

In order to increase the speed of semiconductor lasers, it is important to reduce the extra capacitance (parasitic capacitance) that exists outside the active layer region, which is the light emitting region, in order to reduce leakage of high-frequency signals. This is pointed out by Kobayashi et al. in paper number 918 of the ``Collection of Lectures at the 1980 Spring National Conference of the Institute of Electronics and Communication Engineers.'' Tsumashi, is this parasitic capacitance completely reduced? In order to achieve
Modulation at high speeds on the order of Gb/s becomes possible.

[Problem that the invention seeks to solve]

By the way, in the case of high-performance embedded semiconductor lasers used as light sources for optical communications, a p-n reverse bias junction is used as a current confinement mechanism, so the p-n junction capacitance is large and high-frequency signal current is It leaks to regions other than the active layer via this pn junction and the resistance of the semiconductor layer. Therefore, it is not possible to obtain an ultrahigh-speed semiconductor laser simply by forming an insulating film such as Si02 on the surface of the semiconductor layer. In addition, Si02 itself has a capacitance, and for example, the area is about the size of a normal semiconductor laser element (300 x 250 x 1?
? +2), SiO with a thickness of about 6000. When a film is formed, the capacitance of Si02 white is about 10 pF9.
, it cannot be said that the capacity is small for high frequency modulation of 5 GHz or higher.

? Furthermore, since the thermal expansion coefficients of S10 and the semiconductor differ by about one order of magnitude, strain remains inside the semiconductor after the S10 is formed, which adversely affects the reliability of the semiconductor laser.

The purpose of the present invention is to effectively concentrate high frequency current in the active layer,
An object of the present invention is to provide a highly reliable semiconductor laser device that is capable of ultra-high-speed modulation.

[Means to solve all problems]

In the semiconductor laser device of the present invention, a step is provided on a first conductivity type semiconductor substrate on which a high resistance semiconductor layer is laminated so that a part of the first conductivity type semiconductor substrate is exposed, and the active layer is provided with a step that defines the step. a first conductivity type semiconductor layer in contact with the resistive semiconductor layer;
Semiconductor layers including an active layer, a second conductivity type semiconductor layer, and a contact 1 layer are stacked, and the contact layer, the second conductivity type semiconductor layer, and the active layer are etched in a stripe shape,
A first layer having a refractive index smaller than that of the active layer is provided at the end of the active layer.
Alternatively, it is characterized by embedding a semiconductor layer of the second conductivity type.

The following margins [Function] By configuring the present invention as described above, the active layer, which is the light emitting region, is surrounded by a high resistance semiconductor layer and a small volume of In.
Since it is a P layer (first conductivity type semiconductor layer), there is extremely little so-called parasitic capacitance. Therefore, almost all of the signal current flowing inside the semiconductor up to the high frequency range is supplied to the active layer, resulting in a semiconductor laser device with excellent high frequency characteristics.

〔Example〕

Next, the present invention will be explained using examples.

First, referring to FIG. 1, an InP high resistance layer 9 is laminated on an n-InP substrate 1, and a part of it reaches the n-InP substrate 1 and is removed to form an inverted mesa shape (having an inverted mesa surface). ) A step is formed. The flat surface on the right side of this step (second surface,
nP buffer layer 2, I
nGaAsP active layer 3, p-nP cladding layer 4,
and p-n[}aAsP contact layers 5 are sequentially formed. Here, n(}aAsP active layer 3, p- n
The P cladding layer 4 and the p-1nGaAeP contact layer 5 are etched so that their right side surfaces have an inverted mesa shape. This is because when forming the electrode, it is necessary to electrically separate the bottom surface of the step and the flat surface on the left side of the step (referred to as the first surface, the top surface of the step) (P-side electrode). If the shadow of the surface is used, separation can be formed only by a vapor deposition process, as described later. In this structure, the nGaAsP active layer B is surrounded by a high-resistance semiconductor layer, which is an InP layer with a small volume, so that almost all the signal current injected from the p-side electrode 7 flows to the InGaAsP active layer B, so that high frequency The structure has excellent response characteristics.

FIGS. 2(a) to 2(e) show the manufacturing process of this embodiment.

First, as shown in FIG. 2(a), an InP high resistance layer 9 of approximately 5 μm thickness is laminated using a silicon nitride film 11 as a mask.
-Etch almost half of the nP substrate 1 to a depth of about 6 μm. At this time, the etching is in the <ioo> direction,
Prom methyl heart fluid (brom 0,2c) was used as an etching solution.
A mixed bath solution of c and 100 cc of methyl alcohol is used.

As a result, the stepped surface of the inverted mesa (
(etched surface) can be formed.

Next, as shown in FIG. 2(b), using the silicon nitride film 11 as a mask, an n-InP buffer layer 2 k 2 μmr and nGaAsP active layer 3' is placed on the bottom surface of the step on the reverse mesa surface.
k0.1μ stripe, p 1 piece nP cranoid layer 4 wo 2ttm, p
-InGaAsP contact layer 5 total 1 μm + MO-OV
Sequential formation using D apparatus. Here, since a silicon nitride film is used as a mask, polycrystals are not grown on the mask.Furthermore, since steps are formed on the reverse mesa surface, the growth surface of each layer can be grown relatively flat.

As shown in FIG. 2(C), after removing the silicon nitride film 11 using Vanguard hydrofluoric acid, a new silicon nitride film 12 is formed, and this is used as a mask to form the fcn-nP buffer. Layer 2, nGaAs
P active layer 3, p-InP cladding layer 4, p-InP nGaA
The sP contact layer 5 is etched using a kebrome methyl solution until at least the nInP layer 1 is exposed. The silicon nitride film serving as a mask is formed so that the width of the nGaAsP active layer 6 is approximately 1.5 μm after etching.

As shown in FIG. 2(A), the InGaAsP active layer 6 is selectively etched using a sulfuric acid-based etchant (mixture of 1 part water, 1 part hydrogen peroxide, and 2 parts sulfuric acid), which is a selective etching solution for InGaAsP. etching. At this time, the width of the etched nGaAeP active layer 6 was about 1 μm. Next, regular volume 40, issue 7 (1982), 568-5
(See page 70) to form a pn-InP layer 20'k.

Finally, as shown in FIG. 2(e), after removing the silicon nitride film 12 using buffered hydrofluoric acid, the p-side electrode Z is deposited. Here, Cr and Au f are used as vapor deposited metals.
The layers were sequentially deposited using a resistance heating vacuum evaporation method. In the vacuum evaporation method, the metal to be evaporated advances almost linearly from the evaporation source because the evaporation is performed in a high vacuum. Therefore, the top surface of the step (
The P-side electrode) and the part deposited on the bottom of the step can be separated. After further heat treatment for 5 minutes in a hydrogen atmosphere at 380°C, the n-InP substrate 1 side was mirror polished to a thickness of approximately 150 mm.
After making it μm, the n-side electrode 8 is made of Ti, Au'in-1n.
They are sequentially formed on the P substrate side and heat treated to complete the process.

In the semiconductor laser of this embodiment, both sides of the InGaAsP active layer 6 are covered with high resistance layers or InP layers with a small volume. Therefore, since there is almost no extra junction capacitance due to p-n junctions, etc., the electric signal supplied from the electrode metal is mostly made of InGaAs from direct current to high frequency range.
P is supplied to the active layer 6. Therefore, a semiconductor laser device with excellent high frequency response characteristics is provided. Further, in this structure, since the IHGaASF active layer B is surrounded by an InP layer having a slightly low refractive index, a mode control structure can be easily realized in which the oscillation transverse mode of the semiconductor laser is set to all fundamental modes.

Furthermore, since the mass transport region is small, the deterioration of high frequency response characteristics is also very small. And, engineering nGa
Reliability is improved because the AsP active layer 6 is not exposed to air.

This semiconductor laser wafer has a cavity length of 600 μm.
The semiconductor laser was cleaved so that the strip line was cleaved and fused directly onto the strip line, a semiconductor laser was assembled, and the small signal frequency characteristics were completely measured. As a result, a value of 1QGHz or more was obtained as a 3dB band at a bias current value twice the oscillation threshold. This 3 dB band can be increased to a value higher than 9p by increasing the photon density by shortening the cavity, decreasing the photon lifetime, cooling, etc. In the case of this semiconductor laser device, it oscillates in the fundamental transverse mode and the 3 dB band 13 GHz as band
It became.

The present invention has been explained above using diagrams, but here, InG
Although an aA s P-based semiconductor laser is used, it is also applicable to semiconductor lasers made of other materials, such as [}aAIAs-based semiconductor lasers.

? [Effects of the Invention] As explained above, in the present invention, by covering the portion other than the vicinity of the light emitting region with a high-resistance semiconductor layer, etc., it is possible to eliminate the parasitic capacitance inside the semiconductor laser as much as possible. Furthermore, since the active layer has a so-called buried structure without using a dielectric film such as SiO2, reliability and static characteristics are greatly improved. Therefore, an ultrahigh-speed semiconductor laser device having a modulation band of 1 [] GHz or more and excellent reliability can be easily obtained.

[Brief explanation of drawings]

FIG. 3 is a cross-sectional view for explaining the entire manufacturing process of the example. In the figure, 1 is an n-InP substrate, 9 is an InP high resistance layer,
2 is an n-■nP buffer layer, 3 is an I nC}aAeP active layer, 4 is a p-■nP cladding layer, 5 idp-H
GaAsP contact layer, 20'J-np-q transport buried layer.

Claims (1)

[Claims] 1. A first conductive type semiconductor substrate, a high resistance semiconductor layer formed on the first conductive type semiconductor substrate by exposing a part of the first conductive type semiconductor substrate, and the first conductive type semiconductor substrate. a step portion defined by the high-resistance semiconductor layer; a first conductivity type semiconductor layer laminated on the exposed first conductivity type semiconductor substrate and having one end in contact with the step portion; and the first conductivity type semiconductor layer. An active layer, a second conductivity type semiconductor layer, and a contact layer, which are sequentially laminated on a part of the top and have one end in contact with the stepped portion, and are embedded in the other end of the active layer and have a refractive index smaller than that of the active layer. a conductive type semiconductor layer, wherein the one end of the active layer is in contact with the high-resistance semiconductor layer defining the step portion.