JPH10229252A - Light pulse generator element - Google Patents

Abstract

PROBLEM TO BE SOLVED: To provide a light pulse generator element, capable of generating an ultrashort and high-peak power light pulses. SOLUTION: This element has a double-hetero structure having an nand p-type clad layers 37, 40 with non-doped active layer 38 sandwiched therebetween. A non-doped clad layer 39 is sandwiched between these layers 38, 40. A laser resonator is divided by two p-type electrodes 42, 43 in the length direction into three regions, i.e., two gain regions 31 and saturable absorptive region 32 sandwiched therebetween. An r-f modulating signal is applied to the gain regions 31 with a d-c reverse bias which is applied to an absortive region 32 for improving the response speed of the region 32, thereby generating a short-pulse width high-power ultrashort pulse.

Description

DETAILED DESCRIPTION OF THE INVENTION

[0001]

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to an optical pulse generating element for generating an ultrashort pulse using a semiconductor laser.

[0002]

2. Description of the Related Art Semiconductor lasers which are small, lightweight and capable of direct modulation are widely used for generating ultrashort pulses.

[0003] PPVasil'ev et al. Provide a saturable absorption region in the cavity of a semiconductor laser and apply a reverse bias voltage to thereby provide a pulse width of 1.65 picoseconds and a peak output of 10.
W light pulses were obtained (PPVasil'ev, IHWhite, D. Bur
ns and W. Sibbett, CLEO'93, Baltimore, 1993, Paper C
ThC8). The passive Q-switching method using the saturable absorption and the gain switching method is characterized in that a peak output several tens of times higher than that of other mode-locking methods or the like can be obtained.

FIG. 7 is a perspective view showing the structure of a semiconductor laser using the passive Q switching method, and FIG.
It is A sectional drawing. First, the laminated structure of the semiconductor laser will be described with reference to FIG. N-type substrate 20 made of GaAs
An n-electrode 28 is deposited on the bottom surface, and an n-type buffer layer 21 made of GaAs, an n-type cladding layer 22 made of Al _0.4 Ga _0.6 As, and Al _x Ga _{1 -x} As (x <0.4) are formed on the top surface. An active layer 23, a p-type cladding layer 24 made of Al _0.4 Ga _0.6 As and a p-type cap layer 25 made of GaAs are sequentially laminated. Such a structure in which the p-type, n-type or neutral active layer is sandwiched between the p-type and n-type cladding layers is called a double hetero (DH) structure.

As shown in FIG. 7, the p-type cap layer 25 has a stripe-shaped protrusion 10 along the longitudinal direction of the resonator, and the protrusion 10 is divided into three in the longitudinal direction. On the p-type cap layer 25, two convex and concave p-electrodes 26 and 27 are laminated, and the two p-electrodes 26 and 27 have a convex portion and a concave portion facing each other. Are arranged in such a manner as to be combined with each other at the projection 10. Therefore, a convex p-electrode 26 is provided on the central protruding portion 10 of the three protruding portions 10, and the protruding portions 1 on both ends are provided.
At 0, a concave p-electrode 27 is laminated. Since the two p electrodes 26 and 27 are not in contact, both electrodes 2
Different electric fields can be applied between 6, 27 and n-electrode 28. As a result, the resonator portion shown in FIG. 8 is electrically divided into three regions, with a saturable absorption region 11 at the center and gain regions 12 on both sides.

When a modulation current is applied to the gain region 12, an optical pulse is modulated. When a DC reverse bias is applied to the absorption region 11, the saturable absorption phenomenon is emphasized. Therefore, high output and short pulse of the optical pulse are realized by using both of them.

[0007]

However, it has been difficult for the conventional DH structure semiconductor laser to reduce the light pulse width to 1 picosecond or less. This is for the following reasons. The response speed of the absorption region depends on the thickness of the depletion layer and the residual carrier concentration. The thicker the depletion layer and the lower the residual carrier concentration, the shorter the response speed and the shorter the light pulse width. On the other hand, in a conventional semiconductor laser having a DH structure, the thickness of a depletion layer formed by applying a reverse bias voltage to a saturable absorber is limited by the thickness of an active layer. Thickening of the active layer causes an increase in threshold value and a decrease in efficiency. Therefore, the thickness of the active layer is limited to about 0.3 μm. Therefore, the response speed of the absorption region cannot be increased, and the light pulse width cannot be sufficiently reduced.

It is an object of the present invention to provide an optical pulse generating element capable of generating an optical pulse having an ultrashort pulse and a high peak output.

[0009]

According to the present invention, an active layer is formed in a first layer.
In an optical pulse generating element having a DH structure laminated so as to be sandwiched between a conductive type clad layer and a second conductive type clad layer, between the active layer and the first conductive type clad layer or between the active layer and the second conductive type clad layer A non-doped cladding layer is interposed in at least one of the layers, a laser optical resonator is formed in a direction perpendicular to the laminating direction, and an electrode layer for supplying current to the active layer via the first cladding layer is formed as a laser optical resonator. , And a reverse DC bias is applied to at least one of the divided electrode layers, and a forward RF bias is applied to at least one other electrode layer. It is characterized by the following.

According to this, since the thickness of the depletion layer can be increased without depending on the thickness of the active layer, the response speed of the absorption region can be improved. In addition, by applying a DC bias in the reverse direction to the absorption region, the response speed of the absorption region can be improved, and by applying an RF bias to the gain region, pulse light with a rapid rise can be obtained.

[0011]

DETAILED DESCRIPTION OF THE PREFERRED EMBODIMENTS An embodiment of the present invention will be described below with reference to the drawings. FIG. 1 is a perspective view showing the structure of an optical pulse generating element according to a first embodiment of the present invention, and FIGS. 2 and 3 are sectional views taken along lines AA and BB, respectively.

First, a laminated structure according to the present embodiment will be described with reference to FIGS. The present embodiment has a DH structure, and an n-electrode 4 is formed on the bottom of an n-type substrate 35 made of GaAs.
4 is deposited, and an n-type buffer layer 36 of GaAs and an n-type cladding layer 3 of Al _0.4 Ga _0.6 As are formed on the upper surface.
7, a non-doped active layer 38 of Al _x Ga _1-x As;
A non-doped cladding layer 39 made of Al _y Ga _1-y As,
P-type cladding layer 40 of Al _0.4 Ga _0.6 As and G
The p-type cap layer 41 made of aAs is sequentially formed by MO-CV
The layers are stacked by the D method. Here, there is a relationship of x <y <0.4.

Here, the thickness of both the active layer 38 and the non-doped cladding layer 39 can be shortened by increasing the thickness, but the threshold value of laser oscillation increases at the same time. Is not preferred. The thinner the active layer 38 is, the more effectively the light is confined, and a higher output can be achieved. Therefore, the thickness of the active layer 38 is about 0.01.
0.3 μm, the thickness of the non-doped cladding layer 39 is about 0.1
It is preferable to set in the range of 0.6 μm. Also,
The thickness of the cladding layers 37 and 40 is sufficient to confine light in the active layer 39, and is preferably about 2 μm. The n-type buffer layer 36 and the p-type cap layer 41
A thickness of about m is preferred.

Further, on the p-type cap layer 41, as shown in FIG. 1, a stripe-shaped protrusion 30 is formed along the entire length of the resonator. The projection 30 is divided into three parts by removing the upper half of the p-type cap layer 41 at two places in the longitudinal direction. As shown in FIG. 3, the surface of the p-type cap layer 41 except for the top surface of the protrusion 30 is
An insulating SiN film 45 is deposited. SiN film 45
Above, a concave p-electrode 42 and a convex p-electrode 43
Are stacked in such a manner as to combine the concave portion and the convex portion at the projection portion 30. Therefore, of the three projections 30, a p-type electrode 43 having a convex shape is laminated on the central projection 30 and a p-type electrode 42 having a concave shape is laminated on the projections 30 at both ends.

The p-electrodes 42 and 43 are in direct contact with the p-type cap layer 41 only at the top surface of the projection 30. In other portions, since the insulating SiN film 45 is interposed between the p-type cap layer 41 and the p-type cap layer 41, the region into which current is injected is limited to only the portion directly below the stripe-shaped protrusions 30, and the threshold voltage is reduced. At the same time as reducing the current, the active layer 38
In the above, a stripe-shaped region is provided with a waveguide property for confining the laser beam, and the oscillating transverse mode is stabilized. p
Since the electrodes 42 and 43 are not in contact with each other, the semiconductor portion 34 can be electrically divided by applying different electric fields to the n-electrode. As a result, in the resonator portion whose cross section is shown in FIG. 2, the gain region 31 immediately below the p-electrode 42 on both sides is separated by the saturable absorption region immediately below the middle p-electrode 43 via the separation portion 33 having no p-electrode on the surface. 32 are sandwiched therebetween.

Next, the operation of the semiconductor device will be described. RF between the p electrode 42 and the n electrode 44 in the gain region 31
A modulation signal is applied, and a DC reverse bias is applied between the p electrode 43 and the n electrode 44 in the saturable absorption region 32.

First, in the gain region 32 to which the RF modulation signal has been applied, the input current is a pulse current. As the input current increases, electrons are gradually accumulated in the active layer 38, and when the electron density exceeds the oscillation threshold, the photon density rapidly increases and laser oscillation occurs. As a result, the accumulated electron-hole pairs are rapidly consumed, and the photon density rapidly decreases from the point where the density falls below the oscillation threshold density, and the laser oscillation stops. Thereby, a laser light pulse having a pulse width shorter than the pulse width of the RF modulation signal itself is obtained.

When the light pulse passes through the absorption region 32, the light at the leading edge of the pulse is absorbed by the absorption region 32. However, when the light output exceeds a certain threshold, the absorption sharply decreases. Is transparent and the light pulse is transmitted. This is because the photoexcited electrons fill up the conductor in the absorption region 32 and there is no transition destination. Thereafter, the photoexcited electrons in the absorption region 32 transition to another region or the like, the absorptance in the absorption region 32 increases again, and the light pulse is absorbed. As a result, the light pulse is compressed.

Here, when no bias voltage is applied to the absorption region 32, a time of about 100 nanoseconds is required from when the light absorption is lost to when the light is restored. However, when a reverse bias is applied, minority carriers generated by light absorption can be swept out, and light absorption can be recovered in a minimum of about 0.1 picosecond. In the absorption region 32, the transition of the photoexcited electrons becomes faster and the absorption recovery becomes faster as the depletion layer is thicker and the residual carrier concentration is lower. In the present embodiment, since the thickness of the depletion layer is increased by forming the non-doped cladding layer 39 as compared with the case where only the active layer 38 is provided, the response speed of the saturable absorption region 32 is increased, and the light pulse The width can be shortened. Further, when viewed along the entire length direction of the resonator, since the saturable absorption region 32 is sandwiched between the two gain regions 31,
There is a feature that the laser pulse collides head-on in the absorption region 31 and the saturation effect is enhanced.

Next, the characteristics of the light pulse generated according to the first embodiment will be described with reference to FIGS. FIG. 4 is a spectrum characteristic diagram of the generated light pulse, and FIG. 5 is a diagram of a measurement result of a harmonic autocorrelation of the light pulse. From FIG. 4, the full width at half maximum of the wavelength is 2.6 nm, and the pulse waveform is Se.
full width at half maximum of the assumed and the pulse width to be ch ² corresponds to 0.4 picoseconds. From FIG. 5, the full width at half maximum of the pulse width is 0.5.
It turns out to be picoseconds. The peak output of the pulse is 30
About W was obtained. The pulse width is about 1 /
3, the peak power has increased three times,
The effect of enhancing the saturable absorption by forming the non-doped layer was confirmed.

Incidentally, as an invention in which a non-doped layer is provided between a p-type or n-type cladding layer and an active layer as described above, there is an invention described in JP-A-63-140590. However, the invention described in this publication aims to prevent diffusion of doped impurities from the doped clad layer to the active layer by providing a non-doped layer having a thickness of about 0.1 to 0.2 μm. is there. The present invention includes a region in which the thickness of the non-doped layer is as thick as about 0.1 to 0.6 μm as compared with the above-described invention. By providing the non-doped layer, the depletion layer is thickened, and the saturable absorption characteristic is mainly increased. The aim is to improve the effect. Preventing diffusion of doped impurities into the active layer is a secondary effect, and differs from the above-described invention in configuration, purpose, and effect.

Next, a second embodiment of the present invention will be described. FIG. 6 is a perspective view showing the structure of the semiconductor device of the present embodiment. Since the cross-sectional structure is basically the same as the cross-sectional structure of the first embodiment shown in FIGS. 2 and 3, the cross-sectional view is omitted here.

As shown in FIG. 6, in the second embodiment, p
The feature is that the structure on the electrode side is different from that of the first embodiment shown in FIG. Specifically, first, of the three protrusions 30 on the p-type cap layer 41, one end and the protrusion 30 at the center have a stripe-like structure with a constant width, but the other end. Has a flare structure in which the width increases linearly toward the end.

JMV Verdiell et al. Described the MOPA (Master Oscill
ator Power Amplifier) (JMVerdiell,
K. Dzurko, DFWelch and DRScifres, "1-W diffract
ion-limited semiconductor MOPA with a monolithicall
y integrated 5-GHz electroabsorption modurator ", CL
EO'95, Baltimore, 1995, Paper CThP4), and reported that high output was achieved by using a flare structure for the projection.

Further, on the surface of the p-type cap layer 41, bracket-shaped p-electrodes 57 and 58 and a rectangular p-electrode 59 are formed.
Are laminated. Bracket-shaped p-electrodes 57 and 58 are arranged in a form in which two corresponding brackets face each other ("") so as to individually cover the two stripe-shaped projections 30, and have a rectangular shape. The p-electrode 59 is arranged so as to cover the protrusion of the flare structure. Since the three p electrodes 57, 58, 59 are not in contact, different electric fields can be applied between the three p electrodes 57, 58, 59 and the n electrode 44. Thus, the resonator portion below the protrusion is electrically divided into a plurality of regions along the longitudinal direction. In this case, the structure is such that the first gain region 51, the separation unit 54, the saturable absorption region 52, the separation unit 54, and the second gain region 53 are arranged in this order from the narrow end to the wide end.

Next, the operation of the second embodiment will be described. The RF modulation signal is applied to the first gain region 51, and the second
A DC bias signal is applied to the gain region 53. Further, a DC reverse bias signal is applied to the saturable absorption region 52.
As in the first embodiment, the saturable absorption phenomenon in the saturable absorption region 52 is enhanced, and the light pulses generated in the first gain region 51 and the second gain region 53
2, it is possible to obtain an optical pulse with a high output and a short optical pulse width.

Further, a high output can be obtained because the gain region portion of the resonance portion has a structure in which the linear portion (first gain region 51) and the flare portion (second gain region 53) are combined. In general, when the gain region is a flare portion, the gain can be obtained in a wide range. However, in the case of the present embodiment, laser light having gained reciprocates between both end faces of the resonator, that is, between the entire lengths of the first gain region 51 and the second gain region 53, and oscillates laser. Therefore, the first gain region 51, which is a linear portion, functions as a diffraction slit and serves as a filter for oscillating only single mode light, and other mode light is removed. As a result, the second gain region 5 which is a flare portion
Also in No. 3, only the light contributing to the single mode oscillation obtains the gain, and the generation of the light of other modes is suppressed, and does not contribute to the oscillation. On the other hand, in the second gain region 53, since the width of the single mode light also spreads to the diffraction limit, the gain is larger than that in the case where the entire width is kept constant without providing a flare portion by the area increase due to this spread. And a high-power single-mode light can be obtained.

The first gain region 51 and the second gain region 5
3, the same RF modulation signal can be applied to generate an optical pulse by gain switching as in the first embodiment.

Here, the case where the non-doped cladding layer is provided between the p-type cladding layer and the non-doped active layer has been described as an embodiment.
It may be provided between the mold cladding layer and the non-doped active layer.
Further, it may be provided at two places between both the clad layers and the active layer.

Although the case where the p-type electrode is divided to provide a single or a plurality of gain regions and a saturable absorption region is described here, the divided electrode may be an n-type electrode.

Further, in the present invention, the case of the ridge stripe structure provided with the stripe-shaped projections has been described, but the stripe structure is not limited to this.
The present invention can be applied to other stripe structures.

[0032]

As described above, according to the present invention,
In the DH structure optical pulse generating element, a non-doped cladding layer is provided on at least one of the p-type or n-type cladding layer and the non-doped active layer, and one electrode is provided along the longitudinal direction of the laser optical resonator. The RF modulation signal is applied to at least one electrode of a gain region, and a DC reverse bias signal is applied to a saturable absorption region.

As a result, first, in the gain region, an optical pulse having a rapid rise can be obtained. In addition, in the absorption region, the synergistic effect of increasing the depletion layer thickness and the application of a DC reverse bias improves the high-speed response of absorption and transmission switching, thereby obtaining a high-power optical pulse with a short pulse width. it can. The non-doped cladding layer also has a secondary effect of preventing impurities from diffusing from the p-type or n-type cladding layer into the active layer.

[Brief description of the drawings]

FIG. 1 is a perspective view showing a structure according to a first embodiment of the present invention.

FIG. 2 is a sectional view taken along line AA in FIG.

FIG. 3 is a sectional view taken along line BB in FIG. 1;

FIG. 4 is a spectrum characteristic diagram of an optical pulse generated by the optical pulse generating element of FIG. 1;

FIG. 5 is a diagram showing a measurement result of a harmonic autocorrelation of the optical pulse shown in FIG. 4;

FIG. 6 is a structural perspective view according to a second embodiment of the present invention.

FIG. 7 is a structural perspective view according to a conventional example.

FIG. 8 is a sectional view taken along line AA of FIG. 7;

[Explanation of symbols]

DESCRIPTION OF SYMBOLS 10 ... Protrusion part, 11 ... Gain area, 12 ... Saturable absorption area, 20 ... N-type substrate, 21 ... N-type buffer layer, 22 ... n
Type clad layer, 23 non-doped active layer, 24 p-type clad layer, 25 p-type cap layer, 26, 27 p-electrode, 28 n-electrode, 30 protrusion, 31 gain region, 3
2 saturable absorption region, 33 separation part, 34 semiconductor part, 35 n-type substrate, 36 n-type buffer layer, 37 n
Type cladding layer, 38 non-doped active layer, 39 non-doped cladding layer, 40 p-type cladding layer, 41 p-type cap layer, 42, 43 p-electrode, 44 n-electrode, 45
SiN film, 51: first gain region, 52: saturable absorption region, 53: second gain region, 54: separating portion, 57, 58,
59 ... p electrode.

Claims (1)

[Claims] 1. An optical pulse generating element having a double hetero structure in which an active layer is stacked between a first conductive type clad layer and a second conductive type clad layer, wherein the active layer and the first conductive type clad layer are provided. A non-doped cladding layer is interposed between at least one of the active layer and the second conductivity type cladding layer between the active layer and the second conductivity type cladding layer, and a laser optical resonator is formed in a direction orthogonal to a stacking direction of the double hetero structure. An electrode layer for supplying a current to the active layer through the first cladding layer is divided into at least two in the longitudinal direction of the laser optical resonator, and the at least one of the divided electrode layers is inverted to at least one of the electrode layers. An optical pulse generation element, wherein a DC bias in a direction is applied and a RF bias in a forward direction is applied to at least one other electrode layer.