Optical Devices

Abstract

An optical device according to the present invention comprises a first semiconductor layer (104) and a second semiconductor layer (105), between which an active region (131) is sandwiched, and which are formed to be in contact with a lateral surface of a core (102). The first semiconductor layer (104) and the second semiconductor layer (105) are formed to be thinner than the core (102). The first semiconductor layer (104) comprises a first region (104a) and a first expanding region (104b); and the second semiconductor layer (105) comprises a second region (105a) and a second expanding region (105b). The first expanding region (104b) has a trapezoidal shape that becomes wider in the direction from the first region (104a) side toward a third semiconductor layer (106). The second expanding region (105b) has a trapezoidal shape that becomes wider in the direction from the second region (105a) toward a fourth semiconductor layer (107).

Description

[Technical Field]

[0001] The present invention relates to an optical waveguide type optical device. [Background technology]

[0002] A rib-type current injection structure has been proposed for a lateral injection membrane optical device that injects current laterally into an active layer embedded in a core (Patent Document 1). This optical device uses a so-called rib-type optical waveguide structure in the active region where the active layer is embedded. This structure provides advantages such as extremely strong lateral optical confinement, electrode placement near the active layer, and highly efficient mode conversion with a channel-type passive waveguide. In particular, by providing a reflecting section using a one-dimensional photonic crystal (hereinafter referred to as a "nanobeam") in the passive optical waveguides that sandwich the active region in the waveguiding direction, a laser with strong optical confinement and low resistance can be constructed.

[0003] In this laser, the nanobeam mirror and the active region are connected by a core width conversion taper of the channel-type passive optical waveguide, and a mode conversion taper between the active region of the rib-type waveguide structure and the channel-type passive optical waveguide. [Prior art documents] [Patent documents]

[0004] [Patent Document 1] International Publication No. 2021 / 124440 Summary of the Invention [Problem to be solved by the invention]

[0005] However, the above-mentioned technology has the following problem: In the laser with the above-mentioned conventional structure, when the active layer length is made extremely short (for example, 2.5 μm or less), the mode conversion structure between the active region and the passive optical waveguide does not provide sufficient heat insulation, and it is not possible to obtain a sufficiently high resonator Q value (for example, sufficiently larger than 2,000) to obtain good laser characteristics.

[0006] To solve this problem, if the slope of the taper in the mode conversion structure between the active region and the passive optical waveguide is made gentler than the typical value, the length of the connection between the thinned rib portion and the region where the electrode is formed will be significantly shortened. This connection is the region that serves as the current injection path, and shortening it increases the device resistance, degrading the laser characteristics. Thus, with conventional structures, there is a trade-off between a high resonator Q value and low resistance.

[0007] The present invention has been made to solve the above problems, and aims to reduce the element resistance in a lateral injection membrane optical device while maintaining sufficient thermal insulation between the active region and the passive optical waveguide. [Means for solving the problem]

[0008] The optical device according to the present invention comprises a cladding layer, a core made of compound semiconductors formed on the cladding layer, an active layer embedded in an active region of the core, a first semiconductor layer made of an n-type compound semiconductor and a second semiconductor layer made of a p-type compound semiconductor formed on the cladding layer and in contact with the side surfaces of the core, sandwiching the active region therebetween, a third semiconductor layer made of an n-type compound semiconductor formed on the cladding layer and disposed so as to sandwich the first semiconductor layer between the cladding layer and the active region, and connected to the first semiconductor layer, a fourth semiconductor layer made of a p-type compound semiconductor formed on the cladding layer and disposed so as to sandwich the second semiconductor layer between the cladding layer and the active region, and connected to the second semiconductor layer, a first electrode connected to the third semiconductor layer, a second electrode connected to the fourth semiconductor layer, and two regions sandwiching the active region. and a first passive optical waveguide and a second passive optical waveguide connected to the active region, the first semiconductor layer and the second semiconductor layer being formed thinner than the core, the first semiconductor layer comprising a first region having a trapezoidal shape that narrows in plan view from the core side toward the third semiconductor layer side, and a first expanded region having a trapezoidal shape that widens in plan view from the first region side toward the third semiconductor layer side, the second semiconductor layer comprising a second region having a trapezoidal shape that narrows in plan view from the core side toward the fourth semiconductor layer side, and a second expanded region having a trapezoidal shape that widens in plan view from the second region side toward the fourth semiconductor layer side, and the first region and the second region have tapered shapes at their ends in the waveguiding direction that narrow in plan view the further away from the center of the active region. [Effects of the Invention]

[0009] As described above, according to the present invention, a first expansion region having a trapezoidal shape that becomes wider toward the third semiconductor layer is formed in the first semiconductor layer, and a second expansion region having a trapezoidal shape that becomes wider toward the fourth semiconductor layer is formed in the second semiconductor layer.Therefore, in a lateral injection membrane optical device, it is possible to reduce the element resistance while ensuring sufficient thermal insulation between the active region and the passive optical waveguide. [Brief explanation of the drawings]

[0010] [Figure 1A] FIG. 1A is a plan view showing a configuration of an optical device according to an embodiment of the present invention. [Figure 1B] FIG. 1B is a cross-sectional view showing a partial configuration of an optical device according to an embodiment of the present invention. [Figure 2] FIG. 2 is a plan view showing the configuration of an optical device having a comparative structure. [Figure 3A] FIG. 3A is a plan view showing a state of an optical device in the middle of a process, for illustrating a method for manufacturing an optical device according to an embodiment of the present invention. [Figure 3B] FIG. 3B is a cross-sectional view showing a state of the optical device in the middle of a process for explaining a method for manufacturing an optical device according to an embodiment of the present invention. [Figure 3C] FIG. 3C is a plan view showing a state of the optical device in the middle of a process for explaining the method for manufacturing the optical device according to the embodiment of the present invention. [Figure 3D] FIG. 3D is a plan view showing a state of the optical device in the middle of a process for explaining a method for manufacturing an optical device according to an embodiment of the present invention. [Figure 3E] FIG. 3E is a cross-sectional view showing a state of the optical device in the middle of a process for explaining a method for manufacturing an optical device according to an embodiment of the present invention. [Figure 3F] FIG. 3F is a plan view showing a state of the optical device in the middle of a process, for explaining a method for manufacturing an optical device according to an embodiment of the present invention. [Figure 4] FIG. 4 is a plan view showing the configuration of another optical device having a comparative structure. [Figure 5] FIG. 5 is a characteristic diagram showing the results of calculating the resonator Q value by sweeping Wtp2 in the old structure B, where the active layer length LBH is 2.5 μm, the thicknesses tside of the first semiconductor layer 104″ and the second semiconductor layer 105″ are 100 nm, and Ltp2 is 0.4 μm. [Figure 6] FIG. 6 is a plan view showing the configuration of another optical device having a comparative structure. [Figure 7]Figure 7 shows a characteristic diagram (a) in which the normalized resistance value R' is calculated based on equation (1) and plotted as a function of the active layer length LBH, and a characteristic diagram (b) in which the normalized resistance value R'LBH per unit active layer length is plotted by multiplying R' by LBH. DETAILED DESCRIPTION OF THE INVENTION

[0011] An optical device according to an embodiment of the present invention will be described below with reference to Figures 1A and 1B. Figure 1B shows a cross section perpendicular to the waveguide direction. The x-axis direction (x-direction) shown in Figure 1B is the thickness direction, the y-axis direction (y-direction) is the current injection direction, and the z-axis direction (z-direction) is the waveguide direction.

[0012] This optical device comprises a cladding layer 101, a core 102 formed on the cladding layer 101, an active layer 103 embedded in the core 102, and a first semiconductor layer 104 and a second semiconductor layer 105 formed on the cladding layer 101 and in contact with the side surfaces of the core 102, sandwiching an active region 131 in a direction (y direction) parallel to the surface of the cladding layer 101 and perpendicular to the waveguide direction (z direction).

[0013] The cladding layer 101 is made of, for example, silicon oxide. For example, the cladding layer 101 can be a silicon oxide layer formed on a substrate such as Si. The core 102 is made of, for example, a III-V group compound semiconductor such as InP. For example, the core 102 can be formed by depositing InP on the cladding layer 101 by a well-known metalorganic chemical vapor deposition method or the like.

[0014] The active layer 103 is embedded in the active region 131 of the core 102. The active layer 103 has an outer shape of, for example, a rectangular parallelepiped. The first semiconductor layer 104 and the second semiconductor layer 105 are disposed on either side of the active region 131. The first semiconductor layer 104 is made of, for example, an n-type III-V compound semiconductor such as n-type InP. The second semiconductor layer 105 is made of, for example, a p-type III-V compound semiconductor such as p-type InP.

[0015] This optical device also includes a third semiconductor layer 106 formed on the cladding layer 101, arranged to sandwich the first semiconductor layer 104 between itself and the active region 131, and connected to the first semiconductor layer 104. The optical device also includes a fourth semiconductor layer 107 formed on the cladding layer 101, arranged to sandwich the second semiconductor layer 105 between itself and the active region 131, and connected to the second semiconductor layer 105. The third semiconductor layer 106 is made of an n-type III-V compound semiconductor such as n-type InP. The fourth semiconductor layer 107 is made of a p-type III-V compound semiconductor such as p-type InP.

[0016] This optical device also includes a first electrode 108 electrically connected to the third semiconductor layer 106 and a second electrode 109 electrically connected to the fourth semiconductor layer 107. In this example, the cladding layer 101 side is the lower side, and the upper side of the core 102 is clad with air.

[0017] In this optical device, the first semiconductor layer 104 and the second semiconductor layer 105 are formed to be thinner than the core 102. The thinner regions can be referred to as "trench regions." In this example, the core 102, the first semiconductor layer 104, the second semiconductor layer 105, the third semiconductor layer 106, and the fourth semiconductor layer 107 are integrally formed. A first passive optical waveguide 132 and a second passive optical waveguide 133 are arranged on either side of the active region 131 in the waveguiding direction, and are optically connected to the active layer 103 (active region 131).

[0018] The active region 131, in which the first semiconductor layer 104 and the second semiconductor layer 105 are formed thinner than the core 102, has a so-called rib-type optical waveguide structure in which the first semiconductor layer 104 and the second semiconductor layer 105 form a slab. On the other hand, the first passive optical waveguide 132 and the second passive optical waveguide 133 have a so-called channel-type optical waveguide structure.

[0019] In addition to the above-described configuration, in the optical device according to the embodiment, the first semiconductor layer 104 includes a first region 104a and a first expanded region 104b, and the second semiconductor layer 105 includes a second region 105a and a second expanded region 105b.

[0020] The first region 104a has a trapezoidal shape that narrows from the core 102 side toward the third semiconductor layer 106 side in a planar view. The first expanded region 104b has a trapezoidal shape that widens from the first region 104a side toward the third semiconductor layer 106 side in a planar view. The second region 105a has a trapezoidal shape that narrows from the core 102 side toward the fourth semiconductor layer 107 side in a planar view. The second expanded region 105b has a trapezoidal shape that widens from the second region 105a side toward the fourth semiconductor layer 107 side.

[0021] In this example, the first region 104a and the second region 105a have a planar shape of an isosceles trapezoid with its base on the side of the active layer 103. Also, in this example, the first expanded region 104b has a planar shape of an isosceles trapezoid with its base on the side of the third semiconductor layer 106, and the second expanded region 105b has a planar shape of an isosceles trapezoid with its base on the side of the fourth semiconductor layer 107. Also, the first region 104a and the second region 105a have tapered shapes at their ends in the waveguide direction, with the width in planar view narrowing as they move away from the center of the active region 131 (active layer 103).

[0022] In this example, the first semiconductor layer 104 includes a first neck region 104c connecting the first region 104a and the first expanded region 104b, and the second semiconductor layer 105 includes a second neck region 105c connecting the second region 105a and the second expanded region 105b. The first neck region 104c has a uniform width from the core 102 side to the third semiconductor layer 106 side in a planar view. The second neck region 105c has a uniform width from the core 102 side to the fourth semiconductor layer 107 side in a planar view.

[0023] Furthermore, in this example, a first recess 101a and a second recess 101b are formed in the cladding layer 101 in regions sandwiching the first semiconductor layer 104 in the waveguide direction, and a third recess 101c and a fourth recess 101d are formed in the cladding layer 101 in regions sandwiching the second semiconductor layer 105 in the waveguide direction. Each of the first recess 101a, the second recess 101b, the third recess 101c, and the fourth recess 101d is formed so that the distance (distance in the y direction) between the first recess 101a, the second recess 101b, the third recess 101c, and the fourth recess 101d and the linear portions of the cores 102 of the first passive optical waveguide 132 and the second passive optical waveguide 133 increases with increasing distance from the active region 131.

[0024] In this example, the first passive optical waveguide 132 and the second passive optical waveguide 133, which are arranged on either side of the active region 131 in the waveguiding direction, are optically connected to the active layer 103 (active region 131) via tapered regions whose core widths become smaller with increasing distance from the active region 131. The core widths of the first passive optical waveguide 132 and the second passive optical waveguide 133 may also be set to be the same as the core width of the active region 131.

[0025] Furthermore, the optical device according to the embodiment may include a resonator 111 formed on both sides of the active region 131 in the waveguiding direction. However, the resonator 111 is not necessary. In this example, the resonator 111 is configured from a photonic crystal structure formed in the core 102. The photonic crystal structure is configured by arranging a plurality of through holes penetrating the core 102 in the thickness direction in the core 102 of the first passive optical waveguide 132 and the second passive optical waveguide 133 in the waveguiding direction (nanobeam). However, instead of the photonic crystal structure, a diffraction grating may be formed on the core 102 of the first passive optical waveguide 132 and the second passive optical waveguide 133, and the resonator may be configured using these as reflectors.

[0026] As described above, by forming the resonator (reflector) 111 and sandwiching the active region 131 between the resonators 111 to confine light in the active region 131, the optical device can be operated as a current injection laser. As a mechanism for extracting light, for example, the number of periods of the photonic crystal structure constituting the resonator 111 of the first passive optical waveguide 132 can be reduced, and the transmitted component can be used as the output. Also, for example, a Si core can be formed to be placed as close as possible to the core 102 of the first passive optical waveguide 132 as optically coupling is possible, and oscillating light can be extracted through an optical waveguide using this Si core.

[0027] The optical device according to the above-described embodiment includes the first region 104a and the first expanded region 104b, and the second semiconductor layer 105 includes the second region 105a and the second expanded region 105b. This significantly improves the thermal insulation of the tapered structure between the active region 131, which is a rib-type optical waveguide, and the first passive optical waveguide 132 and the second passive optical waveguide 133, which are channel-type optical waveguides, compared to a conventional structure. This allows, for example, when the optical device is used as a laser, to have a higher resonator Q value compared to a conventional structure. Additionally, the optical device according to the embodiment can reduce device resistance compared to a conventional structure.

[0028] Furthermore, the optical device according to the embodiment can be fabricated by a practical and easily realized method by including the first recess 101a, the second recess 101b, the third recess 101c, and the fourth recess 101d, thereby eliminating the inconveniences in the device structure that arise when this fabrication method is adopted and achieving good device characteristics.

[0029] Here, the optical device according to the embodiment will be described in detail by comparing it with the conventional structure shown in Fig. 2. In the conventional structure (hereinafter referred to as "old structure A"), the first semiconductor layer 104' and the second semiconductor layer 105' formed on either side of the core 102, sandwiching the active region 131, each have a trapezoidal shape that narrows as it moves away from the core 102 side in plan view. In this structure, the length in the z direction is L tp2, the length in the x direction is W tre and the tangent is W tre / L tp2 The structure with a steep slope given by tapers the areas between the active region 131 and the first passive optical waveguide 132 and second passive optical waveguide 133. Note that hereinafter, the connection between the active region 131, which is a rib-type optical waveguide, and the first passive optical waveguide 132 and second passive optical waveguide 133, which are channel-type optical waveguides, will be referred to as "rib-channel."

[0030] On the other hand, as shown in FIG. 1A, the optical device according to the embodiment has a length L in the z direction by providing the first region 104a and the second region 105a. tp2 , tangent is W tp2 / L tp2 (W tp2 <W tre ) is used. Furthermore, according to the embodiment, after the modes of the first passive optical waveguide 132 and the second passive optical waveguide 133 are converted into a rib type by the taper of the first region 104a and the second region 105a, neck The trench region that becomes the current injection path through the neck part of ext / L ext The shape of the neck portion 104c is given by the following equation: The neck portion 104c is a region defined by the first and second neck regions 104c and 105c in the z direction.

[0031] The first recess 101a, the second recess 101b, the third recess 101c, and the fourth recess 101d formed in the cladding layer 101 have a tangent W margin / L margin The tapered shape of the cladding layer 101 is tapered away from the resonator 111 (nanobeam mirror) toward the outside. In the old structure A shown in Fig. 2, the first recess 101a, the second recess 101b, the third recess 101c, and the fourth recess 101d, i.e., the etched regions of the cladding layer 101, do not exist because a realistic manufacturing process is not taken into consideration.

[0032] Next, a method for fabricating an optical device according to an embodiment will be described with reference to FIGS. 3A to 3F. First, a thin semiconductor layer made of, for example, InP is formed on cladding layer 101, and then an InP-based semiconductor layer or semiconductor laminate structure that will become active layer 103 is formed thereon. The semiconductor laminate structure is, for example, a multiple quantum well structure. For example, the semiconductor laminate structure can be fabricated by bonding a wafer made of a III-V compound semiconductor on which the above-mentioned semiconductor laminate structure is crystal-grown to a Si wafer with an oxide film that will become cladding layer 101. Thereafter, the InP-based semiconductor layer or semiconductor laminate structure that will become active layer 103 is patterned using known lithography and etching techniques, thereby forming active layer 103.

[0033] Next, by forming the active layer 103, InP is regrown from the thin semiconductor layer made of InP exposed around the active layer 103, and a thick semiconductor layer 120 (thickness t membrane Then, impurities are introduced to form regions of each conductivity type (Reference 1).

[0034] Next, by using known lithography and etching techniques, regions to become the first semiconductor layer 104, the second semiconductor layer 105, and regions to become the third semiconductor layer 106 and the fourth semiconductor layer 107 are formed. For example, first, a silicon oxide film is formed on the semiconductor layer 120, and the formed silicon oxide film is patterned by using known lithography and etching techniques to form a mask pattern 121.

[0035] Next, the semiconductor layer 120 is etched using the mask pattern 121 as a mask to form regions that will become the first semiconductor layer 104 and the second semiconductor layer 105, and regions that will become the third semiconductor layer 106 and the fourth semiconductor layer 107. Also, in this process, the shapes of the photonic crystal structure that will become the core 102 of the first passive optical waveguide 132 and the second passive optical waveguide 133, and the resonator 111 are formed. All of the InP (semiconductor) in regions other than the first passive optical waveguide 132, the second passive optical waveguide 133, and the core 102 is removed to expose the top surface of the cladding layer 101 (FIGS. 3A and 3B).

[0036] At this stage, as shown in FIG. 3B, a mask pattern 121 having the planar shape shown in the plan view of FIG. 3C remains on the semiconductor layer 120 in which the regions to become the first semiconductor layer 104, the second semiconductor layer 105, the third semiconductor layer 106, and the fourth semiconductor layer 107 have been formed.

[0037] Next, as shown in FIGS. 3D and 3E, a resist pattern 122 having openings 122a and 122b is formed on the mask pattern 121. For example, a resist film is formed by applying a positive resist, and then the resist film is exposed and developed using a known photolithography technique, thereby forming the resist pattern 122. In the openings 122a and 122b, the semiconductor layer 120 is patterned using the mask pattern 121, so that the exposed portions of the upper surface of the cladding layer 101 are visible. These portions become the first recess 101a, the second recess 101b, the third recess 101c, and the fourth recess 101d.

[0038] Next, the mask pattern 121 is etched away from the portions exposed in the openings 122a and 122b using the resist pattern 122. At this time, the exposed portion of the upper surface of the cladding layer 101 is also etched at the same time, as described above, to form the first recess 101a, the second recess 101b, the third recess 101c, and the fourth recess 101d.

[0039] As described above, the mask pattern 121 is removed from the regions that will become the first semiconductor layer 104 and the second semiconductor layer 105, the first recess 101a, the second recess 101b, the third recess 101c, and the fourth recess 101d are formed, and the resist pattern 122 is removed to expose the regions that will become the first semiconductor layer 104 and the second semiconductor layer 105. At this point, the regions that will become the first semiconductor layer 104 and the second semiconductor layer 105, as well as the regions that will become the third semiconductor layer 106 and the fourth semiconductor layer 107, have a thickness of t membrane It remains as it is.

[0040] Next, by etching using the mask pattern 121 from which the regions to be the first semiconductor layer 104 and the second semiconductor layer 105 have been removed, the regions to be the first semiconductor layer 104 and the second semiconductor layer 105 of the semiconductor layer 120 are removed to an appropriate thickness (t membrane -t side ) and etch to thickness t side As a result, as shown in FIG. 3F, first semiconductor layer 104 (first region 104a, first enlarged region 104b, first neck region 104c) and second semiconductor layer 105 (second region 105a, second enlarged region 105b, second neck region 105c) that are thinner than other portions are formed.

[0041] Thereafter, electrodes are formed as shown in Reference 1, thereby obtaining the optical device (photonic crystal laser diode) shown in Figures 1A and 1B. Although omitted in Figure 1B, (part of) mask pattern 121 used to form first semiconductor layer 104 and second semiconductor layer 105, which are thinner than other parts, may remain on core 102 on parts of core 102 where no electrodes are formed. Even if part of mask pattern 121 remains on core 102 in this way, there is no problem in terms of characteristics.

[0042] Next, the effects on device characteristics and advantages in the device fabrication process brought about by the present invention will be described. In order to separate the effects of the present invention, a structure as shown in FIG. 4 (hereinafter referred to as "old structure B") will be considered as an intermediate structure between the optical device according to the embodiment described with reference to FIGS. 1A and 1B and the old structure A described with reference to FIG. 2. The first semiconductor layer 104" and the second semiconductor layer 105" of the old structure B have the same tapered structure (rib-channel transition taper) for mode conversion in the rib-channel as the structure of the embodiment, and the other structures are the same as those of the old structure A.

[0043] In the following calculations, the values ​​shown in Table 1 were commonly used as typical device structure parameters.

[0044] [Table 1]

[0045] In addition, when calculating the resonator Q value of an optical device that has a laser structure by incorporating a resonator 111, a three-dimensional finite-difference time-domain method is used, and resonators 111 made of nanobeams with sufficiently low transmittance are provided in both the +z and -z directions, so that an intrinsic Q value can be obtained that does not include a decrease in the Q value due to light extraction.

[0046] In the old structure B, the active layer length L BH = 2.5 μm, the thickness t of the first semiconductor layer 104″ and the second semiconductor layer 105″ side = 100 nm, L tp2 = 0.4 μm and W tp2 The results of calculating the resonator Q value by sweeping are shown in Figure 5. tp2 = 0.2 μm, the resonator Q value is at its maximum value, and the specified L tp2 Optimal W tp2 In addition, in Figure 5, W tp2 =W tre = 0.7 μm is equivalent to the old structure A, and in the old structure B, Wtp2 It can be seen that by optimizing the Q value, a significantly higher value can be obtained than in the previous structure A.

[0047] The resonator Q value W tp2 The dependency can be understood by the following mechanism: First, W tp2 If W is too wide, as was the case with the previous structure A, the slope of the rib-channel transition taper becomes too steep, and the mode conversion between the two cannot be performed with good adiabaticity, resulting in a deterioration of the Q value. tp2 If the length L is too narrow, the thermal insulation in the rib-channel transition taper portion is good, but the taper ends before the mode on the active region 131 side has fully expanded, and the portion parallel to the x-axis (i.e., the angle of 90 degrees) (length L neck This also deteriorates the insulation performance because the part (

[0048] Therefore, as a result of balancing these two, the optimum W that gives the maximum Q value is tp2 The present invention provides a solution to this nontrivial extremum problem, and tp2 The optimum W value for which the highest Q value is obtained tp2 In this example, the active layer length is L BH The very short value of t = 2.5 μm is used as the thickness of the first and second semiconductor layers. side The thickness of L = 100 nm is assumed, which is a severe condition for obtaining a high Q value. Therefore, the Q value of the old structure A is about 2,000, which is insufficient for obtaining good laser characteristics. However, by using the rib-channel transition taper optimized by the present invention, L tp2 While the same, the Q factor has been increased to approximately 7,500, which is a significant and important improvement in obtaining good laser characteristics.

[0049] Next, consider a structure as shown in FIG. 6 (hereinafter referred to as "old structure B'") that is closer to the structure of the optical device according to the embodiment. As is clear from the figure, old structure B' is a structure in which a first recess 101a', a second recess 101b', a third recess 101c', and a fourth recess 101d' are added to old structure B, thereby creating a hollow in the lower part of the cladding layer 101. None of the first recess 101a', the second recess 101b', the third recess 101c', and the fourth recess 101d' extend into the linear region of the core 102 of the second passive optical waveguide 133.

[0050] 3A to 3F, the regions where the recesses are formed are essential in consideration of the alignment accuracy of the actual manufacturing process. Ideally, the mask pattern used for thinning should be patterned only in the regions to be thinned into the fourth and fifth semiconductor layers. To achieve this, the openings in the resist pattern 122 described with reference to FIG. 3D are formed only in the target regions.

[0051] However, in reality, due to limitations imposed by the drawing resolution and alignment accuracy of the lithography process, patterning as described above is extremely difficult, and if exposure is attempted only on the regions to be thinned as the fourth and fifth semiconductor layers, problems such as blunting of the tip shape of the rib-channel conversion taper and generation of semiconductor layer residue due to misalignment can occur. To avoid these problems, a resist pattern 122 having openings 122a and 122b wider than the regions to be the fourth and fifth semiconductor layers is used, as described with reference to FIG. 3D .

[0052] However, as shown in FIG. 6, when the first recess 101a', the second recess 101b', the third recess 101c', and the fourth recess 101d' are formed, the hollows in the cladding layer 101 appear suddenly when viewed along the z-axis direction, which is felt by the mode of the laser resonator, resulting in a significant decrease in the Q value.

[0053] The first recess 101a, the second recess 101b, the third recess 101c, and the fourth recess 101d of the embodiment are intended to solve the above-mentioned problems. tp2 and W tp2 Table 3 shows the results of calculating the Q values ​​for the old structure B, the old structure B', and the optical device according to the embodiment (proposed structure) using the optimum value obtained from Figure 5. Table 3 also shows each parameter of the proposed structure assumed in the calculation. The values ​​of these parameters are typical values ​​for obtaining good device characteristics in the proposed structure.

[0054] [Table 2] JPEG0007747196000003.jpg68121

[0055] First, in the old structure B, which does not take into consideration the hollowing of the cladding layer 101 in the recessed area, 3 This is because the optimized L tp2 and W tp2 In contrast, in the old structure B' in which the first recess 101a', the second recess 101b', the third recess 101c', and the fourth recess 101d' are added, the Q value is 4.89×10 3 (a 42% drop).

[0056] On the other hand, the Q value of the proposed structure is 7.78 × 10 3 The decrease in the Q value is only 8.4% compared to the previous structure B. In other words, the proposed structure effectively suppresses the decrease in the Q value due to the formation of recesses in the cladding layer 101, while at the same time achieving low resistance by expanding the lengths in the z direction of the first and second semiconductor layers for current injection without significantly decreasing the Q value.

[0057] Here, the reason why the Q value is not significantly reduced even though the first expanded region 104b and the second expanded region 105b of the first semiconductor layer 104 and the second semiconductor layer 105, whose lengths in the z direction are expanded, are close to the first passive optical waveguide 132 and the second passive optical waveguide 133 that form the nanobeam, is as follows: Generally, when another optical waveguide or slab containing a mode with a similar effective refractive index exists near an optical waveguide or nanobeam, directional coupling occurs between the two. When the former forms a resonator, the directional coupling that occurs effectively functions as light extraction, reducing the Q value.

[0058] In contrast, in the proposed structure, various transverse modes are established in the third semiconductor layer 106 and the fourth semiconductor layer 107, on which the first electrode 108 and the second electrode 109 are formed, but because the first semiconductor layer 104 and the second semiconductor layer 105 are thinned, the mode distribution is determined by the strong refractive index contrast between the first semiconductor layer 104 and the second semiconductor layer 105 and the thicker third semiconductor layer 106 and the fourth semiconductor layer 107, and there is very little leakage of the modes into the first semiconductor layer 104 and the second semiconductor layer 105. Therefore, even if the first expanded region 104b and the second expanded region 105b have regions close to the first passive optical waveguide 132 and the second passive optical waveguide 133, no significant optical coupling occurs with the third semiconductor layer 106 and the fourth semiconductor layer 107, and a high Q value can be maintained.

[0059] Next, we will compare the device resistance in each structure to see the effect of the present invention on reducing resistance. In the rib-type current injection structure, the series resistance component of the thinned first and second semiconductor layers is dominant, so we will only consider this resistance component below. The resistivity ρ and thickness t of the first and second semiconductor layers are side is common to each structure, and the dimensionless resistance value R' normalized by these common factors is calculated using the following formula.

[0060]

number

[0061] In the above formula, R is the series resistance of the first semiconductor layer and the second semiconductor layer, L tre (x) is the length of the first semiconductor layer and the second semiconductor layer in the optical axis direction (waveguide direction) at the coordinate x. For the three structures, the proposed structure, the old structure A, and the old structure B, the normalized resistance value R' is calculated based on equation (1) to obtain the active layer length L BH The plot of R' as a function of L is shown in Figure 7(a). BH Multiplying by this gives the normalized resistance R'L per unit active layer length. BH The plot of the normalized resistance is shown in Figure 7(b). tre ,L tp2 ,W tp2 ,W neck ,L ext ,W ext ) were the typical values ​​shown in Tables 1, 2, and 3. In Figure 7, the solid line represents the proposed structure, the dashed line represents the old structure A, and the dotted line represents the old structure B.

[0062] As shown in Figure 7, the proposed structure has a significantly lower resistance than both the previous structures A and B. This resistance reduction effect is particularly noticeable in the case of an extremely short active layer, where the increase in conductance due to the first and second expansion regions is relatively large. For example, BH At 2.5 μm, the difference is 28.6% compared to the previous structure A, L BH At 1.3 μm, a 25.9% reduction in resistance was achieved compared to the previous structure A.

[0063] As described above, according to the present invention, a first extended region having a trapezoidal shape that becomes wider toward the third semiconductor layer is formed in the first semiconductor layer, and a second extended region having a trapezoidal shape that becomes wider toward the fourth semiconductor layer is formed in the second semiconductor layer, so that in a lateral injection membrane optical device, it is possible to achieve low element resistance while ensuring sufficient thermal insulation between the active region and the passive optical waveguide. Furthermore, according to the present invention, the above-mentioned structure can be realized by a practical and easy manufacturing process.

[0064] In the present invention, the structural design of the rib-channel transition taper is determined by the parameter W tp2 This is done by optimizing the specified L tp2 (L tp2 If the length is too long, the neck length L neck becomes too short and the resistance increases, tp2 One of the features of this design is that it has an optimally designed tapered structure that can obtain the highest resonator Q value in a wide range of applications (the length cannot be made too long). tp2 This is based on the non-trivial extreme value problem that if the width is too wide or too narrow, the insulation performance will deteriorate.

[0065] The characteristic structure described above ensures sufficient thermal insulation, and the expanded region expands the thinned first and second semiconductor layers in the waveguide direction, thereby achieving low resistance without compromising the high resonator Q value. This makes good use of the unique properties of the rib-type current injection structure, in that the modes of the fourth and fifth semiconductor layers, where the electrodes are formed, are strongly localized within this region and barely seep into the thinned first and second semiconductor layers, so even if there are regions in the first and second semiconductor layers that are close to the first and second passive optical waveguides, no significant optical coupling occurs.

[0066] Furthermore, as a new, realistic and easily realizable manufacturing process has been proposed in the embodiment, a unique innovation in the drawing pattern in the lithography process makes it possible to simultaneously achieve the above-mentioned high Q value and low resistance. The primary purpose of adiabatically escaping to the outside by providing a tapered structure of about several μm in the recesses formed in the cladding layer is to suppress a decrease in the Q value caused by the recesses formed in the cladding layer, but in conjunction with this, it is only by securing a wide manufacturing margin for forming the thinned region and forming openings in the mask pattern over a wide range that it becomes possible to expand the first and second semiconductor layers to be thinned.

[0067] Some or all of the above-described embodiments may also be described as, but are not limited to, the following supplementary notes.

[0068] [Appendix 1] a cladding layer; a core made of a compound semiconductor formed on the cladding layer; an active layer embedded in the active region of the core; a first semiconductor layer made of an n-type compound semiconductor and a second semiconductor layer made of a p-type compound semiconductor formed on the cladding layer, sandwiching the active region and contacting a side surface of the core; a third semiconductor layer formed on the cladding layer, disposed so as to sandwich the first semiconductor layer between the cladding layer and the active region, and made of an n-type compound semiconductor connected to the first semiconductor layer; a fourth semiconductor layer formed on the cladding layer, disposed so as to sandwich the second semiconductor layer between the cladding layer and the active region, and made of a p-type compound semiconductor connected to the second semiconductor layer; a first electrode connected to the third semiconductor layer; a second electrode connected to the fourth semiconductor layer; a first passive optical waveguide and a second passive optical waveguide, each of which is formed by the cores in two regions sandwiching the active region, and which are connected to the active region; Equipped with the first semiconductor layer and the second semiconductor layer are formed thinner than the core, the first semiconductor layer includes a first region having a trapezoidal shape whose width narrows from the core side toward the third semiconductor layer side in a plan view, and a first expanded region having a trapezoidal shape whose width widens from the first region side toward the third semiconductor layer side, the second semiconductor layer includes a second region having a trapezoidal shape whose width narrows from the core side toward the fourth semiconductor layer side in a plan view, and a second expanded region having a trapezoidal shape whose width widens from the second region side toward the fourth semiconductor layer side, The first region and the second region have tapered shapes at their ends in the waveguiding direction, the width of which becomes narrower in a plan view as the distance from the center of the active region increases. An optical device characterized by:

[0069] [Appendix 2] 10. The optical device of claim 1, a first recess and a second recess formed in a cladding layer in a region sandwiching the first semiconductor layer in a waveguiding direction; a third recess and a fourth recess formed in the cladding layer in a region sandwiching the second semiconductor layer in a waveguiding direction; Equipped with Each of the first recess, the second recess, the third recess, and the fourth recess is formed so that the distance from the core increases as the recess moves away from the active region. An optical device characterized by:

[0070] [Appendix 3] 3. The optical device according to claim 1, the first semiconductor layer includes a first neck region that connects the first region and a first expansion region and has a uniform width from the core side to the third semiconductor layer side in a plan view; The second semiconductor layer includes a second neck region that connects the second region and the second expansion region and has a uniform width from the core side to the fourth semiconductor layer side in a plan view. An optical device characterized by:

[0071] [Appendix 4] In the optical device according to any one of Supplementary Notes 1 to 3, An optical device further comprising a resonator formed on both sides of the active region in a waveguiding direction.

[0072] [Appendix 5] 5. The optical device according to claim 4, An optical device, wherein the resonator is configured from a photonic crystal structure formed in the core.

[0073] [Appendix 6] 5. The optical device according to claim 4, 10. An optical device according to claim 9, wherein the resonator is composed of a diffraction grating formed on the core.

[0074] It should be noted that the present invention is not limited to the embodiments described above, and it is clear that many modifications and combinations can be made by a person having ordinary knowledge in the art within the technical concept of the present invention.

[0075] [Reference 1] K. Takeda et al, "Optical links on silicon photonic chips using ultralow-power consumption photonic-crystal lasers", Optics Express 29, Vol. 29, No. 16, pp. 26082-26092, 2021. [Explanation of symbols]

[0076] 101...cladding layer, 101a...first recess, 101b...second recess, 101c...third recess, 101d...fourth recess, 102...core, 103...active layer, 104...first semiconductor layer, 104a...first region, 104b...first expansion region, 104c...first neck region, 105...second semiconductor layer, 105a...second region, 105b...second expansion region, 105c...second neck region, 106...third semiconductor layer, 107...fourth semiconductor layer, 108...first electrode, 109...second electrode, 111...resonator, 131...active region, 132...first passive optical waveguide, 133...second passive optical waveguide.

Claims

1. a cladding layer; a core made of a compound semiconductor formed on the cladding layer; an active layer embedded in the active region of the core; a first semiconductor layer made of an n-type compound semiconductor and a second semiconductor layer made of a p-type compound semiconductor formed on the cladding layer, sandwiching the active region and contacting a side surface of the core; a third semiconductor layer formed on the cladding layer, disposed so as to sandwich the first semiconductor layer between the cladding layer and the active region, and made of an n-type compound semiconductor connected to the first semiconductor layer; a fourth semiconductor layer formed on the cladding layer, disposed so as to sandwich the second semiconductor layer between the cladding layer and the active region, and made of a p-type compound semiconductor connected to the second semiconductor layer; a first electrode connected to the third semiconductor layer; a second electrode connected to the fourth semiconductor layer; a first passive optical waveguide and a second passive optical waveguide, each of which is formed by the cores in two regions sandwiching the active region, and which are connected to the active region; Equipped with the first semiconductor layer and the second semiconductor layer are formed thinner than the core, the first semiconductor layer includes a first region having a trapezoidal shape whose width narrows from the core side toward the third semiconductor layer side in a plan view, and a first expanded region having a trapezoidal shape whose width widens from the first region side toward the third semiconductor layer side, the second semiconductor layer includes a second region having a trapezoidal shape whose width narrows from the core side toward the fourth semiconductor layer side in a plan view, and a second expanded region having a trapezoidal shape whose width widens from the second region side toward the fourth semiconductor layer side, The first region and the second region have tapered shapes at their ends in the waveguiding direction, the width of which becomes narrower in a plan view as the distance from the center of the active region increases. An optical device characterized by:

2. 2. The optical device according to claim 1, a first recess and a second recess formed in the cladding layer in a region sandwiching the first semiconductor layer in a waveguiding direction; a third recess and a fourth recess formed in the cladding layer in a region sandwiching the second semiconductor layer in a waveguiding direction; Equipped with The first recess, the second recess, the third recess, and the fourth recess are each formed so that the distance from the core increases as the recess moves away from the active region. An optical device characterized by:

3. 3. The optical device according to claim 2, the first semiconductor layer includes a first neck region that connects the first region and a first expansion region and has a uniform width from the core side to the third semiconductor layer side in a plan view; The second semiconductor layer includes a second neck region that connects the second region and the second expansion region and has a uniform width from the core side to the fourth semiconductor layer side in a plan view. An optical device characterized by:

4. The optical device according to any one of claims 1 to 3, An optical device further comprising a resonator formed on both sides of the active region in a waveguiding direction.

5. 5. The optical device according to claim 4, An optical device, wherein the resonator is configured from a photonic crystal structure formed in the core.

6. 5. The optical device according to claim 4, 10. An optical device according to claim 9, wherein the resonator is composed of a diffraction grating formed on the core.

Images