Broadband high-power superluminescent diodes based on tunnel-connected cascades

Abstract

This invention provides a broadband high-power superluminescent diode based on tunnel junction cascade, comprising an epitaxial structure and matched n-electrodes and p-electrodes. The epitaxial structure includes an n-type confinement layer, a p-type confinement layer, x light-emitting structures, and (x-1) tunnel junctions. The x light-emitting structures and (x-1) tunnel junctions are disposed between the n-type and p-type confinement layers along the epitaxial growth direction. Each tunnel junction is disposed between two adjacent light-emitting structures, and the two adjacent light-emitting structures are electrically cascaded via the tunnel junctions. The n-type and p-type confinement layers form a unique pair of confinement layer structures, and the x light-emitting structures operate together within the same pair of confinement layer structures. By setting a unique pair of confinement layer structures, this invention eliminates the multi-waveguide characteristics of the longitudinal refractive index distribution, thereby suppressing longitudinal multimode and near-field multi-peak phenomena, significantly improving the coupling efficiency and beam quality of the device, and simultaneously achieving high power and broadband output.

Description

Technical Field

[0001] This invention specifically relates to a broadband high-power superluminescent light-emitting diode based on tunnel junction cascade, belonging to the field of semiconductor technology. Background Technology

[0002] Superluminescent diodes (SLDs) combine the broadband emission characteristics of LEDs with the high brightness output capability of semiconductor optical amplifiers, offering advantages such as low coherence, broad spectrum, and high stability. They are widely used in fiber optic sensing, fiber optic gyroscopes, and optical coherence tomography (OCT). For OCT systems, the center wavelength, spectral bandwidth, and output power of the light source are key parameters determining imaging depth, axial resolution, and system signal-to-noise ratio (SNR). Higher output power improves SNR and imaging depth, while a wider spectral bandwidth enhances axial resolution and suppresses coherence artifacts. Especially in the approximately 1.0 μm–1.06 μm band, the demand for high-power broadband SLDs continues to grow due to their combination of strong tissue penetration and low water absorption.

[0003] Currently, the mainstream implementation path of SLD devices mostly adopts a single active region structure with single or multiple quantum wells, and obtains higher output power by increasing the injection current, extending the gain length, or increasing the optical confinement factor. Simultaneously, techniques such as end-face coating and tilted waveguides are used to suppress resonance and maintain superradiative operation. This approach has advantages such as relatively simple epitaxy, easy control of the longitudinal guided mode, and high process consistency. However, its performance improvement is constrained by an inherent mechanism: with enhanced injection, the intensified stimulated emission process exacerbates intraspectral gain competition, causing the output spectrum to narrow. Furthermore, the presence of residual reflection at the end face and waveguide scattering increases the risk of laser-like oscillations, leading to spectral ripple or even lasing components. In addition, the temperature rise and carrier overflow caused by high current density operation result in decreased efficiency and deteriorated reliability, making it difficult to sustain the synergistic optimization of high power, wide spectrum, and low coherence at higher power levels.

[0004] There are two main paths for the gain structure of existing broadband high-power SLDs: The first is to use multiple quantum wells within a single active region, achieving gain spectrum superposition through the bandgap difference between different quantum wells to broaden the bandwidth. This type of structure has relatively simple epitaxy and easy control of the longitudinal mode field, but its power increase usually still depends on increasing the injection current density and gain intensity. At high power, intraspectral gain competition is more likely to occur, leading to bandwidth compression, accompanied by thermal effects, carrier overflow, and decreased reliability. Simultaneously, when strained quantum wells are densely stacked, local strain concentration increases the risk of epitaxial defects. The second approach uses multiple active regions and achieves electrical cascading through tunnel junctions to obtain higher power and achieve broadband output through the center wavelength difference between different active regions. However, existing schemes often repeatedly set thick confinement and waveguide layers on both sides of each active region, causing the highly doped tunnel junction to be located at the optical field node, resulting in a multi-waveguide characteristic longitudinal refractive index distribution. This easily supports multiple longitudinal guided modes, leading to near-field multi-peaking at the end face (spot splitting) and reduced coupling efficiency. Summary of the Invention

[0005] The main objective of this invention is to provide a broadband high-power superluminescent diode based on tunnel junction cascade, which can improve output power while maintaining wide bandwidth and low coherence characteristics, and achieve end-face fundamental mode spot output, thereby overcoming the shortcomings of the prior art.

[0006] To achieve the aforementioned objectives, the technical solution adopted by this invention includes: This invention provides a broadband high-power superluminescent diode based on tunnel junction cascade, comprising an epitaxial structure and n-electrodes and p-electrodes matched with the epitaxial structure. The epitaxial structure comprises an n-type confinement layer, a p-type confinement layer, x light-emitting structures, and (x-1) tunnel junctions. The x light-emitting structures and (x-1) tunnel junctions are disposed between the n-type confinement layer and the p-type confinement layer along the epitaxial growth direction. Each tunnel junction is disposed between two adjacent light-emitting structures, and the two adjacent light-emitting structures are electrically cascaded via a tunnel junction. The n-type confinement layer and the p-type confinement layer form a unique pair of confinement structures, and the x light-emitting structures operate within the same pair of confinement structures, where x ≥ 2.

[0007] This invention avoids the need for repeated thick confinement layers and thick waveguide layers around each sub-active region by setting a unique pair of confinement structures, thus eliminating the multi-waveguide characteristics of the longitudinal refractive index distribution. This suppresses longitudinal multimode and near-field multi-peak phenomena at the structural level, significantly improving the coupling efficiency and beam quality of the device, and enabling the output of a wide spectrum, low coherence, and fundamental mode beam under high power conditions.

[0008] Furthermore, the thickness of the n-type confinement layer is 1.0 μm to 2.5 μm, and the n-type doping concentration is 5. 10 17cm -3 ~5 10 18 cm -3 The thickness of the p-type confinement layer is 1.0 μm to 2.5 μm, and the p-type doping concentration is 5%. 10 17 cm -3 ~5 10 18 cm -3 .

[0009] This invention optimizes the thickness and doping concentration of the confinement layer to ensure effective confinement of the light field, while reducing the series resistance, which is beneficial to improving the power conversion efficiency of the device.

[0010] Furthermore, the thickness of the tunnel junction is set to reduce free carrier absorption. By thinning the tunnel junction, the absorption loss of free carriers of photons in the heavily doped region is effectively reduced, thereby improving the overall optical output power.

[0011] Furthermore, the tunnel junction includes a p-type heavily doped layer and an n-type heavily doped layer in a stacked contact.

[0012] Furthermore, the thickness of the heavily p-type doped layer is 5 nm to 15 nm, and the p-type doping concentration is 1. 10 19 cm -3 ~1 10 20 cm -3 The thickness of the n-type heavily doped layer is 5 nm to 15 nm, and the n-type doping concentration is 1. 10 19 cm -3 ~1 10 20 cm -3 .

[0013] Furthermore, the light-emitting structure includes an n-type waveguide layer, a light-emitting region, and a p-type waveguide layer stacked sequentially. The n-type waveguide layer of the first light-emitting structure at the bottom layer is stacked and in contact with the n-type confinement layer, and the p-type waveguide layer of the x-th light-emitting structure at the top layer is stacked and in contact with the p-type confinement layer.

[0014] Furthermore, the thickness of the n-type waveguide layer is 20 nm to 300 nm, and the n-type doping concentration is 5. 10 17 cm -3 ~5 10 18 cm -3 The thickness of the p-type waveguide layer is 20 nm to 300 nm, and the p-type doping concentration is 5%. 10 17 cm -3 ~5 10 18 cm -3 By adjusting the thickness and doping concentration of the waveguide layer, the optical field confinement factor and carrier injection efficiency were optimized, ensuring the gain performance of each emitting region.

[0015] Furthermore, the light-emitting region includes two quantum barrier layers and a quantum well layer stacked between the two quantum barrier layers. The quantum well layers with different light-emitting structures have different thicknesses or compositions, and the different light-emitting structures have different center wavelengths. By designing different light-emitting structures with different center wavelengths, and utilizing the principle of multi-center wavelength gain superposition, the output spectral bandwidth of the device is effectively broadened, achieving wide-spectrum output.

[0016] Furthermore, the thickness of the quantum barrier layer is 3 nm to 15 nm, and the thickness of the quantum well layer is 2 nm to 10 nm.

[0017] In a typical implementation, the epitaxial structure further includes an ohmic contact layer, which is stacked on the p-type confinement layer, and the p-electrode is electrically connected to the ohmic contact layer.

[0018] In a typical implementation, the epitaxial structure has a tilted ridge waveguide structure, the bottom of which is located within the p-type confinement layer, and the p-electrode is disposed on the tilted ridge waveguide structure.

[0019] Furthermore, the width of the ridge waveguide structure is 2 μm to 10 μm, and the height is 0.5 μm to 2 μm.

[0020] This invention can suppress higher-order modes by adjusting the height and width of the ridge waveguide structure, thereby achieving both wide spectrum, low coherence and fundamental mode beam output at high power.

[0021] In a typical implementation, the n-type confinement layer is stacked on the n-type buffer layer, the n-type buffer layer is stacked on the n-type substrate, and the n-electrode is electrically connected to the n-type substrate.

[0022] Furthermore, antireflective coatings are provided on both end faces of the device.

[0023] Compared with the prior art, the advantages of the present invention include: This invention provides a specific broadband high-power superluminescent light-emitting diode based on tunnel junction cascading, exhibiting high output power and superior stability at high power. By electrically cascading multiple light-emitting structures through tunnel junctions, this invention achieves carrier reuse and increases the effective gain segment length, allowing for longitudinal segmented distribution of gain and thermal load. This improves the overall output power without significantly increasing the current density of individual segments, mitigating temperature rise, gain saturation, and efficiency roll-off at high power, and enhancing continuous operation stability.

[0024] The present invention provides a specific broadband high-power superluminescent diode based on tunnel junction cascade that achieves broadband output. By utilizing the structural degrees of freedom of multiple light-emitting structures, the composition or well width of the quantum well of each light-emitting structure is controlled to form a multi-center wavelength gain superposition, thereby obtaining a wider gain spectrum.

[0025] The present invention provides a specific broadband high-power superluminescent diode based on tunnel junction cascade with good epitaxial manufacturability. For strained quantum well systems, dispersing the quantum wells can reduce the intensity of local strain superposition and reduce the risk of defect propagation, thereby improving device consistency and reliability.

[0026] The present invention provides a specific broadband high-power superluminescent diode based on tunnel junction cascade, which improves the stability of the fundamental mode light spot output. By enabling multiple light-emitting structures to share a single confinement structure, the longitudinal multimode caused by repeated thick confinement layers and thick waveguide layers on both sides of each light-emitting structure is avoided. Structurally, the near-field multi-peaking at the end face is suppressed, thereby improving the stability of the fundamental mode light spot output and the system coupling efficiency. Attached Figure Description

[0027] Figure 1 This is a schematic diagram of the epitaxial structure of a broadband high-power superluminescent diode based on tunnel junction cascade, provided in a typical embodiment of the present invention. Figure 2 This is a flowchart illustrating the fabrication process of a broadband high-power superluminescent diode based on tunnel junction cascade, provided in a typical embodiment of the present invention. Figure 3 This is a schematic diagram of a broadband high-power superluminescent diode based on tunnel junction cascade, provided in a typical embodiment of the present invention. Figure 4a This is a two-dimensional near-field distribution of a broadband high-power superluminescent diode in Embodiment 1 of the present invention; Figure 4b This is a one-dimensional near-field distribution of a broadband high-power superluminescent diode in Embodiment 1 of the present invention; Figure 5aThis is a two-dimensional far-field distribution of a broadband high-power superluminescent diode in Embodiment 1 of the present invention; Figure 5b This refers to the two-dimensional far-field divergence angle of a broadband high-power superluminescent diode in Embodiment 1 of the present invention. Detailed Implementation

[0028] In view of the shortcomings of the prior art, the inventors of this invention, through long-term research and extensive practice, have proposed the technical solution of this invention. The following will further explain and illustrate the technical solution, its implementation process, and its principles in conjunction with the accompanying drawings and specific embodiments.

[0029] In a typical implementation, a broadband high-power superluminescent light-emitting diode based on tunnel junction cascade includes an epitaxial structure and n-electrodes and p-electrodes matched with the epitaxial structure. The epitaxial structure is the core part of the device to realize optical gain and emission. It includes an n-type confinement layer, a p-type confinement layer, x light-emitting structures (also called sub-active regions) and (x-1) tunnel junctions, where x is an integer greater than or equal to 2.

[0030] Specifically, x light-emitting structures and (x-1) tunnel junctions are positioned between the n-type confinement layer and the p-type confinement layer along the epitaxial growth direction. More specifically, the n-type and p-type confinement layers form a unique pair of confinement structures, and all x light-emitting structures operate within this pair. This "unique pair of confinement structures" design is a key feature that distinguishes this invention from existing technologies. In existing multi-sub-active region cascade technology schemes, to confine the optical field of each sub-active region, thick confinement layers and waveguide layers are often repeatedly set on both sides of each sub-active region (i.e., the light-emitting structure in this embodiment). This repeated setting leads to a significant multi-waveguide characteristic in the longitudinal refractive index distribution of the entire epitaxial structure, i.e., the refractive index fluctuates multiple times in the longitudinal direction. During transmission, the optical field is segmented by this multi-waveguide structure, resulting in multiple peaks in the near-field distribution at the device end face, i.e., a beam splitting phenomenon. This severely reduces the coupling efficiency between the device and subsequent optical systems such as optical fibers.

[0031] In contrast, this embodiment uses a single pair of confinement structures to uniformly "wrap" all the light-emitting structures within the same optical resonant cavity. This design eliminates the multi-waveguide characteristics in the longitudinal refractive index distribution, allowing the entire active region to form a unified waveguide structure in the longitudinal direction. As a result, the transmission mode of the optical field in the longitudinal direction is effectively controlled, suppressing the generation of longitudinal multimodes, thus ensuring that the device can output an ideal fundamental mode spot. This significantly solves the problem of near-field multi-peak formation at the end face and improves coupling efficiency.

[0032] Specifically, the thickness of the n-type confinement layer is 1.0 μm to 2.5 μm, and the n-type doping concentration is 5. 1017 cm -3 ~5 10 18 cm -3 The thickness of the p-type confinement layer is 1.0 μm to 2.5 μm, and the p-type doping concentration is 5%. 10 17 cm -3 ~5 10 18 cm -3 The thicknesses of the n-type confinement layer and the p-type confinement layer can be the same or different.

[0033] Specifically, each tunnel junction is positioned between two adjacent light-emitting structures, and the two adjacent light-emitting structures are electrically cascaded via a tunnel junction. The tunnel junction configuration enables carrier reuse, allowing carriers to flow sequentially through multiple light-emitting structures, thereby increasing the total output power of the device without significantly increasing the injected current density. For example, when x is 3, the epitaxial structure includes three light-emitting structures and two tunnel junctions. The three light-emitting structures are arranged sequentially along the epitaxial growth direction, and the two tunnel junctions are located between the first and second light-emitting structures, and between the second and third light-emitting structures, respectively. It should be understood that the specific value of x can be selected according to actual power requirements and process conditions. For example, x can also be 2, 4, or larger integers, as long as the condition x ≥ 2 is met, it falls within the scope of protection of this invention.

[0034] Specifically, the thickness of the tunnel junction is set to reduce free carrier absorption. More specifically, the tunnel junction includes a stacked p-type heavily doped layer and an n-type heavily doped layer, with the p-type heavily doped layer having a thickness of 5 nm to 15 nm and a p-type doping concentration of 1. 10 19 cm -3 ~1 10 20 cm -3 The thickness of the heavily doped n-type layer is 5 nm to 15 nm, and the n-type doping concentration is 1. 10 19 cm -3 ~1 10 20 cm -3 .

[0035] Specifically, regarding the internal structure of the light-emitting structure, it comprises sequentially stacked n-type waveguide layers, a light-emitting region, and a p-type waveguide layer. The n-type waveguide layer of the bottommost light-emitting structure is stacked and in contact with the n-type confinement layer, and the p-type waveguide layer of the topmost x-th light-emitting structure is stacked and in contact with the p-type confinement layer. This arrangement allows the x light-emitting structures to be stacked in series along the epitaxial growth direction, and the entire structure is enclosed by the confinement structure, forming a unified waveguide structure. This avoids the multi-waveguide effect caused by the repeated placement of confinement layers in existing technologies.

[0036] Specifically, the thickness of the n-type waveguide layer is 20 nm to 300 nm, and the n-type doping concentration is 5. 10 17 cm -3 ~5 10 18 cm -3 The thickness of the p-type waveguide layer is 20 nm to 300 nm, and the p-type doping concentration is 5%. 10 17 cm -3 ~5 10 18 cm -3 The refractive index of the waveguide layer is between that of the confinement layer and the emitting region, thus playing the role of guiding the light field and confining the mode.

[0037] Specifically, the luminescent region comprises two quantum barrier layers and a quantum well layer stacked between the two quantum barrier layers. The quantum well layers with different luminescent structures have different thicknesses or compositions, and different luminescent structures have different center wavelengths. This "barrier-well-barrier" sandwich structure constitutes the basic unit of carrier recombination luminescence. The quantum well layer, acting as a potential well, effectively confines electrons and holes within a narrow region, increasing the recombination probability; while the quantum barrier layers on both sides act as potential barriers, preventing carrier escape and further enhancing luminescence efficiency. Through this design, the multiple luminescent structures in the epitaxial structure no longer produce a single wavelength output, but rather form a wavelength gradient distribution. Specifically, the material composition (e.g., the proportion of In in InGaAs) or geometric dimensions (thickness) of the quantum well layer directly determines its band structure, and thus the center wavelength of the emission. When the thickness of the quantum well layer changes, the strength of the quantum confinement effect changes accordingly; a thicker quantum well typically corresponds to a narrower band gap and a longer emission wavelength. Similarly, adjusting the material composition can also change the band gap width, thereby adjusting the wavelength.

[0038] Typically, the thickness of the quantum barrier layer is 3 nm to 15 nm, and the thickness of the quantum well layer is 2 nm to 10 nm. This parameter range is a preferred design for GaAs-based materials in the 1050 nm wavelength band. For example, in a specific implementation, the thicknesses of the quantum well layers of the three luminescent structures can be set to 3.9 nm, 4.0 nm, and 4.1 nm, respectively, achieving a staggered distribution of the center wavelength through slight differences in thickness. It should be understood that the specific values ​​of thickness and composition mentioned above are merely illustrative. In practical applications, those skilled in the art can adaptively adjust the thickness and composition of the quantum well according to the target output wavelength band (such as the 1300 nm or 1550 nm band) and the selected material system (such as InP-based). As long as the condition that different luminescent structures have different center wavelengths is met, it falls within the protection scope of this invention.

[0039] Specifically, the epitaxial structure has a tilted ridge waveguide structure, such as... Figure 3 As shown, the tilted ridge waveguide structure is formed on the surface of the epitaxial structure. Positionally, the bottom of the ridge waveguide structure is located within the p-type confinement layer. This positional setting is crucial. If the etching (the tilted ridge waveguide structure is formed in conjunction with the etching process) is too deep, penetrating the p-type confinement layer and entering the light-emitting region, it will introduce severe surface recombination loss, reducing luminous efficiency. Conversely, if the etching is too shallow, an effective refractive index difference cannot be formed, making it difficult to achieve lateral confinement of the optical field. Therefore, controlling the bottom of the ridge waveguide within the p-type confinement layer ensures both a good current injection channel and sufficient optical field confinement capability. Simultaneously, the etching depth also affects the divergence angle of the structure. A deeper etching depth results in stronger lateral confinement of the optical field, leading to increased gain and output power, but also a larger far-field divergence angle. A shallower etching depth results in weaker lateral confinement of the optical field, decreased gain and output power, and a smaller divergence angle.

[0040] Furthermore, the p-electrode is disposed on the tilted ridge waveguide structure. By covering the surface of the tilted ridge waveguide structure, the p-electrode guides the injected current to the light-emitting region beneath the structure, achieving targeted carrier injection and recombination emission. In a preferred embodiment, the tilted ridge waveguide structure has a width of 5 μm and a height of 1.2 μm. It should be understood that the above parameters are merely illustrative; in practical applications, the width of the tilted ridge waveguide structure can be adjusted according to the target power and spot size, and the height can be selected based on the specific thickness of the epitaxial layer, as long as the requirements for fundamental mode output and current limitation are met.

[0041] Specifically, regarding the electrode contact portion, the epitaxial structure also includes an ohmic contact layer. This ohmic structure layer is stacked on top of the p-type confinement layer, and the p-electrode is electrically connected to the ohmic contact layer. Specifically, the ohmic contact layer typically uses a heavily doped p-type material, with a thickness of approximately 200 nm and a doping concentration of up to 2. 10 19 cm -3 .

[0042] Regarding the device's supporting substrate, an n-type confinement layer is stacked on an n-type buffer layer, which in turn is stacked on an n-type substrate. The n-electrode is electrically connected to the n-type substrate. The n-type substrate, serving as the growth substrate for the entire epitaxial structure, provides mechanical support. In practical applications, the n-type substrate is typically made of n-GaAs material with good conductivity, with a thickness of approximately 350 μm, to balance mechanical strength and heat dissipation performance. The n-type buffer layer can be made of GaAs material homogeneous with the substrate, with a thickness of approximately 300 nm.

[0043] In a more typical implementation scheme, please refer to Figure 2 A fabrication process for a broadband high-power superluminescent diode based on tunnel junction cascades includes: Step 1: Grow the aforementioned epitaxial structure on an n-type substrate.

[0044] Step 2: Define the tilted waveguide pattern using photolithography, and fabricate the tilted waveguide structure using dry etching or wet etching.

[0045] Step 3: Deposit a passivation layer to cover the sidewalls to prevent surface leakage and improve device reliability. SiO2 is usually selected as the passivation film.

[0046] Step 4: Create a window on the passivation film using photolithography / etching to expose the ohmic contact layer, and deposit p-side metal as the p-electrode.

[0047] Step 5: Annealing causes the p-electrode to form an ohmic contact with the p-type ohmic contact layer of the epitaxial structure.

[0048] Step 6: Thicken the p-electrode to reduce series resistance.

[0049] Step 7: Thin the n-type substrate to reduce the bulk resistance.

[0050] Step 8: Deposit metal on the n-face as the n-electrode.

[0051] Step 9: Annealing allows the n-electrode to form an ohmic contact with the n-type substrate.

[0052] A schematic diagram of the three-dimensional structure of the tilted waveguide SLD device prepared by the above process is shown below. Figure 3 As shown.

[0053] This invention provides a broadband high-power superluminescent light-emitting diode (SLED) based on cascaded tunnel junctions. By adjusting the thickness and composition of each quantum well, multi-center wavelength gain can be superimposed, resulting in a wider output spectrum. Free carrier absorption can be reduced by decreasing the thickness of the heavily doped tunnel junctions. Parasitic feedback can be suppressed by depositing antireflection coatings on both end faces and changing the tilt angle of the tilted waveguide to maintain superluminescent operation. Higher-order modes can be suppressed by adjusting the height and width of the tilted ridge waveguide structure, thus achieving a balance between broadband, low coherence, and fundamental mode output at high power. The thickness of the waveguide layer and confinement layer can be adjusted within a certain range to ensure that only the fundamental mode output is supported in the longitudinal direction.

[0054] The technical advantages of the broadband high-power superluminescent diode based on tunnel junction cascade provided by this invention are mainly reflected in: (1) High power stability: By adopting multiple light-emitting structures and realizing electrical cascade with tunnel junctions, the gain and heat generation are distributed in the longitudinal segment, and the carrier reuse and effective gain segment length are increased at the same time. Thus, the total light output is improved without significantly increasing the single-segment current density, which is beneficial to alleviate the thermal effect and efficiency roll-off under high power and improve the continuous working stability; (2) Broad spectrum output capability: By utilizing the structural degrees of freedom of multiple light-emitting structures, the composition and well width of the quantum well of each light-emitting structure are segmented and controlled to realize the superposition of multi-center wavelength gain and spectral flattening, thereby obtaining Wider output spectrum; (3) Epitaxial manufacturability: For strain quantum well systems such as InGaAs, dispersing the quantum wells among multiple light-emitting structures and separating them with spacer layers can reduce the concentration and superposition intensity of local strain energy and reduce the risk of defect propagation; (4) Fundamental mode spot output: In response to the longitudinal multimode problem that may be introduced by the multi-active region structure, this invention unifies and simplifies the design of the longitudinal waveguide and confinement layer, so that multiple light-emitting structures work together in the same confinement structure, avoids the repeated setting of thick confinement layers and thick waveguide layers around each light-emitting structure, reduces the risk of near-field multi-peak at the end face, and thus takes into account the wide spectrum, low coherence and fundamental mode spot output under high power conditions, and improves the coupling efficiency of the device.

[0055] Example 1 A broadband high-power superluminescent light-emitting diode based on tunnel junction cascade is disclosed. This device is based on GaAs material system, emits light in the 1050 nm band, and is used in optical coherence tomography (OCT) systems.

[0056] Specifically, combined Figure 1 As shown, in this embodiment of the invention, the device has a thickness of 350 μm and a Si doping concentration of 2 × 10⁻⁶. 18 cm -3 The n-type GaAs substrate 1 has the following epitaxial structure from bottom to top along the epitaxial growth direction: n-type GaAs buffer layer 2, with a thickness of 300 nm and a Si doping concentration of 2 × 10⁻⁶.18 cm -3 ; n-type Al 0.3 Ga 0.7 As confinement layer 3 has a thickness of 1.5 μm and a Si doping concentration of 2 × 10⁻⁶. 18 cm -3 ; The first light-emitting structure comprises n-type Al atoms stacked sequentially. 0.15 Ga 0.85 As waveguide layer 4, first luminescent region 8, p-type Al 0.15 Ga 0.85 As waveguide layer 9, n-type Al 0.15 Ga 0.85 The thickness of As waveguide layer 4 is 125 nm, and the Si doping concentration is 5 × 10⁻⁶. 17 cm -3 p-type Al 0.15 Ga 0.85 The thickness of As waveguide layer 9 is 20 nm, and the C doping concentration is 1×10⁻⁶. 18 cm -3 The first light-emitting region 8 includes an undoped GaAs quantum barrier layer 5 and an undoped In layer stacked sequentially. 0.3 Ga 0.7 As quantum well layer 6, undoped GaAs quantum barrier layer 7, GaAs quantum barrier layer 5 has a thickness of 5 nm, In 0.3 Ga 0.7 The thickness of As quantum well layer 6 is 3.9 nm and the thickness of GaAs quantum barrier layer 7 is 5 nm. The first tunnel junction 12 includes a p-type heavily doped GaAs layer 10 and an n-type heavily doped GaAs layer 11 stacked sequentially. The p-type heavily doped GaAs layer 10 has a thickness of 9 nm and a C doping concentration of 5 × 10⁻⁶. 19 cm -3 The thickness of the n-type heavily doped GaAs layer 11 is 9 nm, and the Si doping concentration is 5 × 10⁻⁶. 19 cm -3 ; The second light-emitting structure includes n-type Al atoms stacked sequentially. 0.15 Ga 0.85 As waveguide layer 13, second luminescent region 17, p-type Al 0.15 Ga 0.85 As waveguide layer 18, n-type Al 0.15 Ga 0.85 The thickness of the As waveguide layer 13 is 20 nm, and the Si doping concentration is 1×10⁻⁶. 18 cm -3 p-type Al 0.15 Ga0.85 The thickness of the As waveguide layer 18 is 20 nm, and the C doping concentration is 1 × 10⁻⁶. 18 cm -3 The second light-emitting region 17 includes an undoped GaAs quantum barrier layer 14 and an undoped In layer 14, which are stacked sequentially. 0.3 Ga 0.7 As quantum well layer 15, undoped GaAs quantum barrier layer 16, GaAs quantum barrier layer 14 with a thickness of 5 nm, In 0.3 Ga 0.7 The thickness of the As quantum well layer 15 is 4 nm and the thickness of the GaAs quantum barrier layer 16 is 5 nm. The second tunnel junction 21 includes a p-type heavily doped GaAs layer 19 and an n-type heavily doped GaAs layer 20 stacked sequentially. The p-type heavily doped GaAs layer 19 has a thickness of 9 nm and a C doping concentration of 5 × 10⁻⁶. 19 cm -3 The thickness of the n-type heavily doped GaAs layer 20 is 9 nm, and the Si doping concentration is 5 × 10⁻⁶. 19 cm -3 ; The third light-emitting structure includes n-type Al atoms stacked sequentially. 0.15 Ga 0.85 As waveguide layer 22, third luminescent region 26, p-type Al 0.15 Ga 0.85 As waveguide layer 27, n-type Al 0.15 Ga 0.85 The thickness of the As waveguide layer 22 is 20 nm, and the Si doping concentration is 1×10⁻⁶. 18 cm -3 p-type Al 0.15 Ga 0.85 The thickness of the As waveguide layer 27 is 125 nm, and the C doping concentration is 5 × 10⁻⁶. 17 cm -3 The third luminescent region 26 includes an undoped GaAs quantum barrier layer 23 and an undoped In layer 26 stacked sequentially. 0.3 Ga 0.7 As quantum well layer 24, undoped GaAs quantum barrier layer 25, GaAs quantum barrier layer 23 with a thickness of 5 nm, In 0.3 Ga 0.7 The thickness of the As quantum well layer 24 is 4.1 nm and the thickness of the GaAs quantum barrier layer 25 is 5 nm. p-type Al 0.3 Ga 0.7 As confinement layer 28 has a thickness of 1.5 μm and a C doping concentration of 2 × 10⁻⁶. 18 cm -3 ; p-type heavily doped GaAs ohmic contact layer 29, 200 nm thick, with a C doping concentration of 2 × 10⁻⁶. 19 cm -3 .

[0057] Combination Figure 3 As shown, the surface of the epitaxial structure of this broadband high-power superluminescent diode also has a tilted ridge waveguide structure. The width of the tilted ridge waveguide structure is 5 μm, and the height, i.e., the etching depth, is 1.2 μm. The near-field and far-field distributions of this structure are both ideal fundamental mode light spots. The broadband high-power superluminescent diode based on tunnel junction cascade in Example 1 abandons multiple single waveguide superposition structures and only contains a pair of confinement structures, corresponding to only a single light spot, thus avoiding multi-peak situations.

[0058] Combination Figure 4a , Figure 4b and Figure 5a , Figure 5b The effect of the broadband high-power superluminescent diode of Embodiment 1 of the present invention was verified. Figure 4a , Figure 4b These are the two-dimensional and one-dimensional near-field distributions of this broadband high-power superluminescent diode, respectively. Figure 5a , Figure 5b These represent the two-dimensional far-field distribution and far-field divergence angle of this broadband high-power superluminescent diode. Figure 4a , Figure 4b and Figure 5a , Figure 5b It can be seen that both the near field and the far field are single light spots, which greatly improves the fiber coupling efficiency.

[0059] This invention provides a specific broadband high-power superluminescent light-emitting diode based on tunnel junction cascading, exhibiting high output power and superior stability at high power. By electrically cascading multiple light-emitting structures through tunnel junctions, this invention achieves carrier reuse and increases the effective gain segment length, allowing for longitudinal segmented distribution of gain and thermal load. This improves the overall output power without significantly increasing the current density of individual segments, mitigating temperature rise, gain saturation, and efficiency roll-off at high power, and enhancing continuous operation stability.

[0060] The present invention provides a specific broadband high-power superluminescent diode based on tunnel junction cascade that achieves broadband output. By utilizing the structural degrees of freedom of multiple light-emitting structures, the composition or well width of the quantum well of each light-emitting structure is controlled to form a multi-center wavelength gain superposition, thereby obtaining a wider gain spectrum.

[0061] The present invention provides a specific broadband high-power superluminescent diode based on tunnel junction cascade with good epitaxial manufacturability. For strained quantum well systems, dispersing the quantum wells can reduce the intensity of local strain superposition and reduce the risk of defect propagation, thereby improving device consistency and reliability.

[0062] The present invention provides a specific broadband high-power superluminescent diode based on tunnel junction cascade, which improves the stability of the fundamental mode light spot output. By enabling multiple light-emitting structures to share a single confinement structure, the longitudinal multimode caused by repeated thick confinement layers and thick waveguide layers on both sides of each light-emitting structure is avoided. Structurally, the near-field multi-peaking at the end face is suppressed, thereby improving the stability of the fundamental mode light spot output and the system coupling efficiency.

[0063] It should be understood that the above embodiments are merely illustrative of the technical concept and features of the present invention, and are intended to enable those skilled in the art to understand the content of the present invention and implement it accordingly. They should not be construed as limiting the scope of protection of the present invention. All equivalent changes or modifications made in accordance with the spirit and essence of the present invention should be covered within the scope of protection of the present invention.

Claims

1. A broadband high-power superluminescent light-emitting diode based on tunnel junction cascade, comprising an epitaxial structure and n-electrodes and p-electrodes matched with the epitaxial structure, characterized in that, The epitaxial structure includes: an n-type confinement layer, a p-type confinement layer, x light-emitting structures, and (x-1) tunnel junctions. The x light-emitting structures and (x-1) tunnel junctions are disposed between the n-type confinement layer and the p-type confinement layer along the epitaxial growth direction. Each tunnel junction is disposed between two adjacent light-emitting structures. The two adjacent light-emitting structures are electrically cascaded through a tunnel junction. The n-type confinement layer and the p-type confinement layer form a unique pair of confinement structures. The x light-emitting structures work together in the same pair of confinement structures, where x ≥ 2.

2. The broadband high-power superluminescent diode based on tunnel junction cascade as described in claim 1, characterized in that: The thickness of the n-type confinement layer is 1.0 μm to 2.5 μm, and the n-type doping concentration is 5. 10 17 cm -3 ~5 10 18 cm -3 The thickness of the p-type confinement layer is 1.0 μm to 2.5 μm, and the p-type doping concentration is 5%. 10 17 cm -3 ~5 10 18 cm -3 .

3. The broadband high-power superluminescent diode based on tunnel junction cascade as described in claim 1, characterized in that: The tunnel junction includes a stacked p-type heavily doped layer and an n-type heavily doped layer; Preferably, the thickness of the heavily p-type doped layer is 5 nm to 15 nm, and the p-type doping concentration is 1. 10 19 cm -3 ~1 10 20 cm -3 The thickness of the n-type heavily doped layer is 5 nm to 15 nm, and the n-type doping concentration is 1. 10 19 cm -3 ~1 10 20 cm -3 .

4. The broadband high-power superluminescent diode based on tunnel junction cascade as described in claim 1, characterized in that: The light-emitting structure includes an n-type waveguide layer, a light-emitting region, and a p-type waveguide layer stacked sequentially. The n-type waveguide layer of the first light-emitting structure at the bottom layer is stacked and in contact with the n-type confinement layer, and the p-type waveguide layer of the x-th light-emitting structure at the top layer is stacked and in contact with the p-type confinement layer.

5. The broadband high-power superluminescent diode based on tunnel junction cascade as described in claim 4, characterized in that: The thickness of the n-type waveguide layer is 20 nm to 300 nm, and the n-type doping concentration is 5%. 10 17 cm -3 ~5 10 18 cm -3 The thickness of the p-type waveguide layer is 20 nm to 300 nm, and the p-type doping concentration is 5%. 10 17 cm -3 ~5 10 18 cm -3 .

6. The broadband high-power superluminescent diode based on tunnel junction cascade as described in claim 4 or 5, characterized in that: The light-emitting region includes two quantum barrier layers and a quantum well layer stacked between the two quantum barrier layers. The quantum well layers with different light-emitting structures have different thicknesses or compositions, and the different light-emitting structures have different center wavelengths. Preferably, the thickness of the quantum barrier layer is 3 nm to 15 nm, and the thickness of the quantum well layer is 2 nm to 10 nm.

7. The broadband high-power superluminescent diode based on tunnel junction cascade as described in claim 1, characterized in that, The epitaxial structure further includes an ohmic contact layer, which is stacked on the p-type confinement layer, and the p-electrode is electrically connected to the ohmic contact layer.

8. The broadband high-power superluminescent diode based on tunnel junction cascade as described in claim 1 or 7, characterized in that: The epitaxial structure has a tilted ridge waveguide structure, the bottom of which is located within the p-type confinement layer, and the p-electrode is disposed on the tilted ridge waveguide structure. Preferably, the width of the ridge waveguide structure is 2 μm to 10 μm and the height is 0.5 μm to 2 μm.

9. The broadband high-power superluminescent diode based on tunnel junction cascade as described in claim 1 or 7, characterized in that: The n-type confinement layer is stacked on the n-type buffer layer, the n-type buffer layer is stacked on the n-type substrate, and the n-electrode is electrically connected to the n-type substrate.

10. The broadband high-power superluminescent diode based on tunnel junction cascade as described in claim 1, characterized in that: Anti-reflection coatings are provided on both end faces of the device.

Images