JP2005099387A - Semiconductor optoelectronic waveguide - Google Patents

Abstract

<P>PROBLEM TO BE SOLVED: To provide a semiconductor optoelectronic waveguide with an nin type heterostructure performing stable operation of an optical modulator. <P>SOLUTION: On the top and bottom surfaces of a core layer 11 whose structure is so determined that electrooptic effect effectively operates at an operating light wavelength and light absorption causes no problem, intermediate clad layers (12-1 and 12-2) having band gaps larger than the band gap of the core layer 11 are provided so that carriers generated by the light absorption are not trapped by a heterointerfacce, and clad layers 13-1 and 13-2 having band gaps larger than those intermediate clad layers are provided on the top surface of the intermediate clad layer 12-1 and the bottom surface of the intermediate clad layer 12-2 respectively. On the top surface of the clad layer 13-1, a p-type layer 15 and an n-type layer 16 are laminated in order, and in the range of an applied voltage used in an operation state, the entire area of the p-type layer 15 and a partial or the entire area of the n-type layer 16 are depleted. <P>COPYRIGHT: (C)2005,JPO&NCIPI

Description

  The present invention relates to a semiconductor optoelectronic waveguide, and more particularly to a semiconductor optoelectronic waveguide having a nin type heterostructure that enables stable operation of an optical modulator.

  In recent large-capacity optical communication systems, optical signals modulated at a high speed of Gbit / s or higher are transmitted. However, the longer the transmission distance, the more easily affected by the dispersion effect of the fiber, and the pulse waveform is distorted. It is necessary to use an optical signal with little ping. For this reason, the generation of a normal optical signal is not performed by direct modulation of a laser diode (LD) having extremely large chirping, but by combining a DC-operated LD and an external modulator.

Conventional typical external modulator used in long-distance transmission of the optical signal is a LN modulator configured in accordance with LiNbO ₃ (LN) waveguides. The principle of operation of the LN modulator is that a refractive index change based on the electro-optic effect is generated in an optoelectronic waveguide in which an optical waveguide and an electric waveguide are coupled, and the phase change is given to the light by this refractive index change. is there. Such an LN modulator can function as an optical intensity modulator incorporating an optical phase modulator or a Mach-Zehnder (MZ) interferometer, or as a highly functional optical switch configured by combining multiple waveguides. It is.

However, LiNbO ₃ is a main material of the LN modulator to a dielectric, passivation and waveguide advanced technology for processing is required. In addition, since the waveguide length is relatively long, it is necessary to use special photolithography different from a normal semiconductor process, and the size of the package on which the LN modulator is mounted must be increased. For these reasons, the LN modulator module has a problem that the manufacturing cost is high and the size of the optical transmitter module becomes relatively large.

  Semiconductor optical modulators based on the same operating principle as LN modulators are also known. For example, GaAs optical modulators using Schottky electrodes on semi-insulating GaAs and using them as optoelectronic waveguides, or hetero pn junctions There are known InP / InGaAsP optical modulators in which a voltage is efficiently applied to a waveguide core portion in addition to optical confinement by utilizing the above.

  FIG. 3 is a band diagram of a waveguide constituting a typical conventional InP / InGaAsP optical modulator, in which 31 is a core layer of the waveguide, and 32-1 and 32-2 are first cladding layers. , 33-1 and 33-2 are p-type and n-type second cladding layers, respectively. 30-1 and 30-2 are electrons and holes, respectively, and a voltage is applied to the p-type second cladding layer 33-1 and the n-type second cladding layer 33-2, A desired electro-optic effect is induced in the core layer 31 to realize light modulation. In such a conventional waveguide, since voltage application to the core layer 31 is performed by a pn junction, there is little leakage current, and carriers generated by light absorption can be easily flowed to the outside. Stable operation is realized.

  However, the GaAs optical modulator provided with the Schottky electrode has a problem that the operating voltage becomes high. InP / InGaAsP optical modulators have a narrow operating band due to the propagation loss of electrical signals due to the high resistance of the p-clad layer, and the waveguide length due to the large optical absorption of the p-type cladding layer. There is a problem that it cannot be made long and it is difficult to reduce the operating voltage. The propagation loss of the electric signal in the InP / InGaAsP optical modulator occurs in the process of charging / discharging the pn junction through the resistance of the signal line and the resistance of the p-type second cladding layer 33-1. In particular, the resistance of the p-type second cladding layer 33-1 is a problem that cannot be avoided because it is caused by the material physical properties of a low hole mobility and a high resistance value. In view of such a problem, a waveguide having a nin type structure has recently been proposed.

  FIG. 4 is a band diagram of a waveguide having a nin type structure in which both clad layers (33-1 and 33-2) on both sides of the waveguide of the InP / InGaAsP optical modulator shown in FIG. A voltage is applied between these two n-type electrode layers to operate the device. In this figure, 41 is a core layer of the waveguide, and 42-1 and 42-2 are first cladding layers. The difference from the configuration shown in FIG. 3 is that both electrode layers (44-1 and 44-2) are n-type, and the p-type second cladding layer 33-1 in FIG. However, it is replaced by an Fe-doped semi-insulating layer 45 having a deep Fe level 46 and an n-type electrode layer 44-1 (see Patent Document 1). The n-type electrode layer 44-2 corresponds to the n-type second clad layer 33-2 in FIG. 3, and 40-1 and 40-2 are electrons and holes, respectively. .

  In such a configuration, since the deep Fe level 46 of the semi-insulating layer 45 acts as an ionized acceptor, the band is bent by the charge to form a potential barrier against electrons, and as shown by the arrows in the figure, the band The electrons 44-1 and the holes 40-2 in the vicinity of the curved portion are recombined through the deep Fe level 46 in the semi-insulating layer 45. Therefore, the leakage current of electrons is suppressed by this potential barrier, and an electric field can be applied to the core layer 41.

  However, since the density of the deep Fe level 46 is not sufficiently high in the waveguide having this structure, the ionization state of the level changes depending on the bias. Such bias dependence of the ionized state causes a change in the depletion layer thickness due to a voltage change, resulting in a failure to maintain a proportional relationship between the applied voltage and the electric field applied to the core layer 41. Furthermore, since the interval between trapping and releasing carriers by the deep Fe level 46 is relatively long, it becomes difficult to respond to high-speed modulation signal processing, resulting in a problem that the modulation intensity has frequency dispersion.

  The basic concept of “operating a device by applying a voltage between two n-type electrode layers” has been conventionally known in the field of electronic devices as a so-called bulk barrier diode. As an applied example, there is a report of “a modulator in which a core layer for inducing a carrier band filling effect of a quantum well is introduced” (see Patent Document 2). Since this optical modulator uses the insertion / extraction of electrons to / from the quantum well, it is impossible in principle to increase the operation speed as compared with the optical modulator using the electro-optic effect.

JP 2003-177368 A U.S. Pat.No. 5,647,029

  The present invention has been made in view of such problems, and an object of the present invention is to provide a semiconductor optoelectronic waveguide having a nin type heterostructure that enables stable operation of an optical modulator. It is in.

  In order to achieve such an object, the present invention provides a first semiconductor clad layer and a second semiconductor clad on each of one main surface and the other main surface of a semiconductor core layer exhibiting an electro-optic effect. And a third semiconductor clad layer functioning as an n-type electrode layer are sequentially stacked, and a band gap of the first semiconductor clad layer is larger than a band gap of the semiconductor core layer, Each of the band gaps of the semiconductor clad layer and the third semiconductor clad layer is larger than the band gap of the first semiconductor clad layer, and is laminated on one main surface or the other main surface of the semiconductor core layer. Between the second semiconductor cladding layer and the third semiconductor cladding layer, the second semiconductor cladding layer side is p-type, and the third semiconductor cladding layer side is n-type. A semiconductor optoelectronic waveguide, wherein a pn junction layer is provided.

  Preferably, each layer thickness and impurity concentration of the pn junction layer is such that, in the operating state of the semiconductor optoelectronic waveguide, the entire p layer is depleted, while the n layer is at least partially depleted. Is set.

Preferably, the impurity concentration of the p layer of the pn junction layer is 1 × 10 ¹⁷ cm ⁻³ or more, and the impurity concentration of the n layer is 5 × 10 ¹⁷ cm ⁻³ or more.

  Preferably, the n layer of the pn junction layer is doped with an impurity that forms a deep level in addition to the n-type impurity.

  Preferably, the band gap energy of the n layer of the pn junction layer is smaller than the band gap energy of the p layer of the pn junction layer.

  More preferably, the deep level impurity doped in the n layer of the pn junction layer is Fe.

  According to the present invention, it is possible to facilitate the band profile control of the nin-type heterostructure included in the optoelectronic waveguide, and therefore it is possible to provide a semiconductor optoelectronic waveguide that enables stable operation of the optical modulator. It becomes. As a result, a more stable optical modulation operation can be realized without impairing the features of the nin-type heterostructure semiconductor optoelectronic waveguide having a low drive voltage, which contributes to lower power consumption and lower cost of the module.

  Embodiments of the present invention will be described below with reference to the drawings.

  FIG. 1 is a diagram for explaining a configuration example of a semiconductor optoelectronic waveguide according to the present invention. FIG. 1A is a perspective view of the optoelectronic waveguide, and FIG. 1B is a band diagram thereof. 11 is a core layer, and its structure is determined so that the electro-optic effect works effectively at the operating light wavelength and the light absorption does not become a problem. For example, in the case of a 1.5 μm band device, the quantum well layer and the barrier layer are formed of InGaAlAs compounds, and the core layer 11 having a multiple quantum well structure in which the Ga / Al composition of these layers is made different.

  On the upper and lower surfaces of the core layer 11, an intermediate cladding having a composition such as InGaAlAs having a band gap larger than the band gap of the core layer 11 in order to prevent carriers generated by light absorption from being trapped at the heterointerface. Layers (12-1 and 12-2) are provided.

  On each of the upper surface of the intermediate cladding layer 12-1 and the lower surface of the intermediate cladding layer 12-2, cladding layers 13-1 and 13-2 having a larger band gap than these intermediate cladding layers and having a composition such as InGaAlAs are provided. Is provided.

On the upper surface of the cladding layer 13-1, for example, an InGaAlAs p-type layer 15 and an InGaAlAs n-type layer 16, for example, are sequentially laminated. The region and the partial region or the entire region of the n-type InGaAlAs layer 16 are depleted. The doping concentration profiles of these layers are determined so that the potential change of the band in such a depletion region is sufficiently large, that is, a sufficient potential barrier for electrons is induced. The doping concentration of these layers is preferably 1 × 10 ¹⁷ cm ⁻³ or more for the p-type layer 15 and 5 × 10 ¹⁷ cm ⁻³ or more for the n-type layer 16. For example, the p-type layer 15 has a doping concentration of 2 × 10 ¹⁷ cm ⁻³ and the n-type layer 16 has a doping concentration of 1 × 10 ¹⁸ cm ⁻³ .

  Each of the upper surface of the n-type InGaAlAs layer 16 and the lower surface of the cladding layer 13-2 is provided with n-type layers 14-1 and 14-2 having a composition such as InGaAlAs that function as the cladding layer. 1 is provided with an electrode 18-1. The band gaps of these n-type layers 14-1 and 14-2 are set larger than the band gaps of the intermediate cladding layers 12-1 and 12-2. And n-type layer 14-2 which is the lowest layer of these laminated structures is provided on a partial region of the main surface of n-type electrode layer 17 having electrode 18-2.

  In order to function as an optoelectronic waveguide, a waveguide structure including a mesa structure having a cross section as illustrated in FIG. 1A is used, and the electrodes 18-1 and 18-2 are in a state where light is propagated through the waveguide. The electric signal is input from the above, and a voltage is applied between the n-type layer 14-1 and the n-type layer 14-2.

  As can be understood from FIG. 1B showing the band diagram in the voltage application state, the n-type layer 14-1 is formed by the potential barrier formed by the presence of the p-type InGaAlAs layer 15 and the n-type InGaAlAs layer 16. On the other hand, the leak current due to the electron injection from the hole 10-2 is suppressed, while the hole 10-2 generated by light absorption (although slightly) is a shallow level acceptor in the p-type InGaAlAs layer 15 and the n-type InGaAlAs layer 16. Then, recombination is performed via a donor, whereby a voltage can be applied to the core layer 11.

  When the band diagram of FIG. 1B is compared with the band diagram shown in FIG. 4, the waveguide of the conventional configuration induces a potential change by ionizing deep levels, whereas in the structure of the present invention, The potential shape is surely controlled by determining the concentrations of the acceptor and the donor in the shallow level so that a desired electric field strength is applied to the core layer 11.

  In FIG. 1, a pn junction layer composed of the p-type InGaAlAs layer 15 and the n-type InGaAlAs layer 16 is provided between the cladding layer 13-1 and the n-type layer 14-1, but this configuration is changed. Thus, it may be provided between the cladding layer 13-2 and the n-type layer 14-2.

  During operation, electrons 10-1 and holes 10-2 are generated by light absorption in the core layer 11, although only slightly. Among these, the electrons 10-1 easily reach the n-type layer 14-2, but the holes 10-2 may accumulate near the n-type InGaAlAs layer 16 with sharp band curvature. The accumulated hole 10-3 becomes a forward bias factor in the pn junction between the p-type InGaAlAs layer 15 and the n-type InGaAlAs layer 16, and therefore pushes down the potential barrier in this region and causes the core layer 11 to have a voltage. May be difficult to be applied and cause electron injection from the n-type layer 14-1 side.

  In this embodiment, in order to quickly recombine such accumulated holes 10-3, the p-type InGaAlAs layer 15 and the n-type InGaAlAs layer 16 are made highly doped layers to reduce the thickness of the pn junction. As a result, the electrons and the accumulated holes are spatially approximated to increase the probability of interband recombination indicated by the arrows in FIG. As a result, the holes 10-3 generated in the core layer 11 and accumulated near the n-type InGaAlAs layer 16 are quickly removed, and the potential barrier formed by the p-type InGaAlAs layer 15 and the n-type InGaAlAs layer 16 is removed. Can be suppressed.

  In the waveguide of this embodiment, a layer corresponding to the n-type InGaAlAs layer 16 in FIG. 1 is doped with an impurity that forms a deep level such as Fe together with a donor impurity. Note that the doping amount of impurities forming deep levels is set sufficiently lower than the doping amount of donor impurities. According to such doping, an impurity forming a deep level does not greatly affect the band profile, but the recombination probability through the deep level is increased, and light absorption absorbs in the core layer 11. It is possible to quickly remove the generated holes.

FIG. 2 is a band diagram of the waveguide of this embodiment. The layer corresponding to the n-type InGaAlAs layer 16 in FIG. 1 is an n-type layer 19 having a smaller band gap energy such as InGaAsP. Holes generated in the core layer 11 by light absorption by making the band gap difference (ΔE _G ) and the doping profile between the p-type layer 15 such as InGaAlAs and the n-type layer 19 such as InGaAsP into a desired shape. A part of 10-2 reaches the n-type InGaAsP layer 19 (10-3), thereby enabling faster recombination. Here, in controlling the potential shape, it is preferable that the valence band discontinuity between the p-type InGaAlAs layer 15 and the n-type InGaAsP layer 19 is smaller than the conduction band discontinuity. This is because the smaller the valence band discontinuity, the easier the holes pass through the interface between the p-type InGaAlAs layer 15 and the n-type InGaAsP layer 19.

  In the above description, InGaAlAs and InGaAsP have been exemplified as the constituent materials of the waveguide in explaining the present invention. However, the present invention is not limited to these materials, and the present invention is not limited to III-V group compound semiconductors including AlGaAs. The waveguide of the invention may be configured.

It is a figure for demonstrating the structural example of the semiconductor optoelectronic waveguide of this invention, (a) is a perspective view of this optoelectronic waveguide, (b) is the band diagram. 6 is a band diagram of a waveguide of Example 4. It is the band diagram of the waveguide which comprises the conventional typical InP / InGaAsP optical modulator. 4 is a band diagram of a waveguide having a nin type structure in which both clad layers on both sides of the waveguide of the InP / InGaAsP optical modulator shown in FIG. 3 are n type.

Explanation of symbols

11 Core layer 12-1, 12-2 Intermediate cladding layer 13-1 13-2 Cladding layer 14-1, 14-2 n-type layer 15 p-type InGaAlAs layer 16 n-type InGaAlAs layer 17 n-type electrode layer 18-1, 18-2 Electrode 19 n-type InGaAsP layer 30-1, 40-1 Electron 30-2, 40-2 Hole 31, 41 Core layer 32-1, 32-2, 42-1, 42-2 of waveguide 1 cladding layer 33-1 p-type second cladding layer 33-2 n-type second cladding layer 44-1 and 44-2 n-type electrode layer 45 Fe-doped semi-insulating layer 46 deep Fe level

Claims (6)

A third semiconductor functioning as a first semiconductor clad layer, a second semiconductor clad layer, and an n-type electrode layer on each of one main surface and the other main surface of the semiconductor core layer exhibiting an electro-optic effect And a clad layer are sequentially laminated,
The band gap of the first semiconductor cladding layer is larger than the band gap of the semiconductor core layer,
Each of the band gaps of the second semiconductor cladding layer and the third semiconductor cladding layer is larger than the band gap of the first semiconductor cladding layer,
Between the second semiconductor clad layer and the third semiconductor clad layer stacked on one main surface or the other main surface of the semiconductor core layer, the second semiconductor clad layer side is p-type. A semiconductor optoelectronic waveguide characterized in that a pn junction layer having an n-type on the third semiconductor cladding layer side is provided.   Each layer thickness and impurity concentration of the pn junction layer are set so that the entire region of the p layer is depleted while the n layer is at least partially depleted in the operating state of the semiconductor optoelectronic waveguide. The semiconductor optoelectronic waveguide according to claim 1, wherein: The impurity concentration of the p layer of the pn junction layer is not less than 1 × ¹⁰ 17 ^{cm -3,} according to claim 1 or 2, wherein the impurity concentration of the n-layer is 5 × ¹⁰ 17 ^{cm -3} or more Semiconductor optoelectronic waveguide.   4. The semiconductor optoelectronic waveguide according to claim 1, wherein the n layer of the pn junction layer is doped with an impurity that forms a deep level in addition to an n-type impurity.   5. The semiconductor optoelectronic waveguide according to claim 1, wherein the band gap energy of the n layer of the pn junction layer is smaller than the band gap energy of the p layer of the pn junction layer. 6. The semiconductor optoelectronic waveguide according to claim 4, wherein the deep level impurity doped in the n layer of the pn junction layer is Fe.

Images