Transverse cavity vertical cavity surface emitting laser and method for manufacturing the same

Abstract

The application provides a transverse cavity type vertical cavity surface emitting laser and a preparation method thereof, and relates to the technical field of semiconductor lasers. The vertical cavity surface emitting laser comprises a substrate, an n-type distributed Bragg reflection layer, an active region, an oxidation limiting layer, a p-type distributed Bragg reflection layer and an electrode structure. The p-type distributed Bragg reflection layer comprises a first reflection part close to the electrode structure and a second reflection part between the first reflection part and the oxidation limiting layer, the second reflection part is provided with a transverse cavity extending from the outside to the inside of the vertical cavity surface emitting laser, and the effective transverse dimension of the second reflection part is smaller than that of the first reflection part. In this way, a relatively reduced-diameter reflection area can be formed above the oxidation limiting layer, the distribution of the injected current to the central light-emitting area is improved, the equivalent refractive index and loss distribution of the peripheral area of the second reflection part are adjusted, and the coordination between the injected current distribution and the transverse mode control is improved.

Description

Technical Field

[0001] This application relates to the field of semiconductor laser technology, and more specifically, to a transverse cavity type vertical cavity surface-emitting laser and its fabrication method. Background Technology

[0002] Vertical-cavity surface-emitting lasers (VCSELs) are characterized by low threshold current, emission direction perpendicular to the substrate, ease of arraying, and convenient coupling with fiber optic or optical systems. They have been widely used in short-distance optical communication, optical interconnects, 3D sensing, and optoelectronic integration. A typical VCSEL usually consists of a substrate, an n-type distributed Bragg reflector, an active region, an oxide confinement layer, a p-type distributed Bragg reflector, and an electrode structure. The oxide confinement layer forms the lateral confinement region for current and optical field.

[0003] As vertical-cavity surface-emitting lasers (VCSELs) evolve towards high-speed modulation and highly stable output, the injection current distribution and lateral mode control within the device require greater coordination. If current limiting is primarily achieved using an oxide confinement layer, the injected current tends to experience concentrated lateral contraction near the oxide confinement region, leading to mutual interference between carrier injection distribution in the central emitting region, ineffective injection in the peripheral region, and higher-order lateral modes. Simultaneously, simply adjusting the oxide confinement aperture size may result in trade-offs between current limiting, heat accumulation, series resistance, and lateral mode selection.

[0004] Therefore, improving the coordination between injection current distribution and transverse mode control in vertical cavity surface-emitting lasers has become a technical problem that needs to be solved. Summary of the Invention

[0005] To address the shortcomings of existing technologies, this application provides a transverse cavity type vertical cavity surface-emitting laser and its fabrication method.

[0006] In a first aspect, this application discloses a transverse cavity type vertical cavity surface-emitting laser, comprising: Substrate; An n-type distributed Bragg reflector layer is disposed above the substrate; The active region is positioned above the n-type distributed Bragg reflector layer. An oxidation confinement layer is disposed above the active region; A p-type distributed Bragg reflector layer, disposed above the oxide confinement layer; and An electrode structure electrically connected to the p-type distributed Bragg reflector layer; The p-type distributed Bragg reflector layer includes a first reflective portion close to the electrode structure and a second reflective portion located between the first reflective portion and the oxide confinement layer. The second reflective portion has a transverse cavity extending inward from the outer side of the vertical cavity surface-emitting laser, and the effective transverse dimension of the second reflective portion is smaller than the effective transverse dimension of the first reflective portion.

[0007] Optionally, the lateral cavity is located in the peripheral region of the second reflective portion, and the second reflective portion includes a central semiconductor region defined by the lateral cavity; The oxidation confinement layer includes a central oxide hole, and the central semiconductor region overlaps with the central oxide hole in a direction perpendicular to the substrate; The effective lateral dimension of the central semiconductor region is greater than or equal to the effective lateral dimension of the central oxide hole.

[0008] Optionally, the transverse cavity may be a continuous annular cavity or multiple cavities spaced apart along the circumference.

[0009] Optionally, the p-type distributed Bragg reflector layer includes multiple semiconductor layer pairs, and the lateral cavity extends through at least two of the multiple semiconductor layer pairs in a direction perpendicular to the substrate.

[0010] Optionally, the lateral cavity is located on the side of the second reflective portion near the oxide confinement layer and does not extend to the oxide confinement layer.

[0011] Optionally, the effective lateral dimension of the first reflective portion is a first dimension, the effective lateral dimension of the central semiconductor region is a second dimension, and the effective lateral dimension of the central oxide hole is a third dimension, wherein the first dimension is greater than the second dimension, and the second dimension is greater than or equal to the third dimension.

[0012] Optionally, the radial extension depth of the lateral cavity varies along a direction perpendicular to the substrate, so that the second reflective portion has a stepped or gradient effective lateral dimension.

[0013] Optionally, the cavity wall of the transverse cavity is covered with a passivation layer.

[0014] Secondly, this application discloses a preparation method, including: An n-type distributed Bragg reflector layer, an active region, an oxide confinement layer, and a p-type distributed Bragg reflector layer are sequentially formed on a substrate. The p-type distributed Bragg reflector layer comprises multiple pairs of semiconductor layers stacked along a direction perpendicular to the substrate. The p-type distributed Bragg reflector layer is etched to form a device mesa; The oxide confinement layer is selectively oxidized to form a central oxide pore; Laterally etch the reflective portion of the p-type distributed Bragg reflector layer near the active region from the outside of the device mesa to form a transverse cavity extending inward from the outside of the vertical cavity surface emitter laser. During the lateral etching process, a probe light is applied to the p-type distributed Bragg reflector layer in a direction perpendicular to the substrate. The reflected signal formed by the probe light after reflection from the p-type distributed Bragg reflector layer is collected, and the lateral etching is controlled based on the reflected signal. An electrode structure is formed that is electrically connected to the p-type distributed Bragg reflector layer.

[0015] Optionally, the reflected signal changes as the lateral etching progresses, and the change is caused by the gradual formation of the lateral cavity in the p-type distributed Bragg reflector layer and the optical coupling formed between the lateral cavity and the plurality of semiconductor layer pairs.

[0016] Optionally, the reflected signal includes a reflectance spectrum, and the method further includes: Extract reflectance spectral features from the reflectance spectrum; The current radial extension depth of the transverse cavity is determined based on the predetermined mapping relationship between the reflection spectrum feature quantity and the radial extension depth of the transverse cavity. The predetermined mapping relationship is established based on the periodic optical reflection characteristics of the plurality of semiconductor layer pairs and the change in equivalent refractive index distribution caused by the transverse cavity.

[0017] Optionally, the reflection spectrum characteristics include at least one of the following: the displacement of the reflection peak position, the change in the reflection peak intensity, the change in the slope of the reflection band edge, the spectral oscillation period, or the characteristic parameters of the resonance sideband introduced by the transverse cavity.

[0018] Optionally, the lateral etching is terminated when the current radial extension depth reaches the target depth; or, Based on the difference between the current radial extension depth and the target depth, at least one of the lateral etching rate, etchant concentration, or etching ambient temperature is adjusted.

[0019] Optionally, the area of ​​illumination of the probe light on the p-type distributed Bragg reflector layer at least covers the peripheral area of ​​the second reflective portion.

[0020] Compared to existing technologies, this application forms a lateral cavity on the side of the p-type distributed Bragg reflector layer near the oxide confinement layer, creating a locally narrowed structure with a smaller effective lateral dimension compared to the side near the electrode structure. Based on this structure, the injected current, during its propagation from the electrode structure to the active region, is guided and constrained by the locally narrowed region before reaching the oxide confinement layer. This improves the current distribution towards the central emitting region and reduces abrupt changes in local current distribution that might occur when current contraction relies solely on the oxide confinement layer. Simultaneously, the lateral cavity alters the equivalent refractive index and loss distribution in the peripheral region of the p-type distributed Bragg reflector layer, making higher-order lateral modes with more pronounced lateral extension more susceptible to disturbances or losses, thereby enhancing lateral mode selectivity.

[0021] Therefore, this application can improve the coordination between injection current distribution and transverse mode control in a vertical cavity surface-emitting laser without simply relying on reducing the size of the oxide confinement aperture. Attached Figure Description

[0022] Figure 1 A schematic diagram of a longitudinal cross-sectional structure of a transversely perforated PDBR VCSEL provided in this application embodiment; Figure 2 A schematic diagram of a three-dimensional structure of a horizontally perforated PDBR VCSEL provided in this application embodiment; Figure 3 This is a flowchart illustrating a method for fabricating a vertical-cavity surface-emitting laser, as provided in an embodiment of this application.

[0023] Reference numerals: 1. Substrate; 2. n-type distributed Bragg reflector layer; 3. Active region; 4. Oxide confinement layer; 5. First reflective portion; 6. Second reflective portion; 7. Lateral cavity; 8. Electrode structure; 9. Injected current transport path. Detailed Implementation

[0024] The technical solutions in the embodiments of this application will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of this application, and not all embodiments.

[0025] like Figure 1 and Figure 2 As shown, the lateral cavity type vertical cavity surface-emitting laser includes a substrate 1, an n-type distributed Bragg reflector 2 disposed above the substrate 1, an active region 3 disposed above the n-type distributed Bragg reflector 2, an oxide confinement layer 4 disposed above the active region 3, a p-type distributed Bragg reflector 4 disposed above the oxide confinement layer 4, and an electrode structure 8. The oxide confinement layer 4 may include a central oxide hole, which is used to define the lateral passage region of current and optical field.

[0026] The p-type distributed Bragg reflector includes a first reflective portion 5 near the electrode structure 8 and a second reflective portion 6 located between the first reflective portion 5 and the oxide confinement layer 4. The second reflective portion 6 has a lateral cavity 7 extending inward from the outside of the vertical-cavity surface-emitting laser, such that the effective lateral dimension of the second reflective portion 6 is smaller than the effective lateral dimension of the first reflective portion 5. The injected current transport path 9 schematically represents the path of current transport from the electrode structure 8 through the p-type distributed Bragg reflector to the active region 3.

[0027] In this application, the direction perpendicular to substrate 1 can be understood as the stacking direction of each semiconductor epitaxial layer, and the lateral direction can be understood as the direction generally parallel to the main surface of substrate 1. The outer side of the vertical cavity surface-emitting laser can be understood as the region on the side where the device mesa or the sidewall of the p-type distributed Bragg reflector layer is located. The lateral cavity 7 extends from the outside to the inside, meaning that the lateral cavity 7 extends a certain distance from the device sidewall towards the device center region, but it is not required that the lateral cavity 7 penetrates the entire p-type distributed Bragg reflector layer. The lateral cavity 7 can be understood as the lateral extension space in the second reflective portion 6 that is not occupied by semiconductor material. This lateral extension space can change the material continuity, equivalent refractive index distribution, and effective conductivity range of the peripheral region of the second reflective portion 6.

[0028] In some embodiments, the substrate 1 can be a GaAs substrate, the n-type distributed Bragg reflector layer 2 can include 25 to 40 pairs of n-type semiconductor layer pairs, each n-type semiconductor layer pair can include two AlGaAs-based semiconductor layers with different refractive indices; the active region 3 can include 1 to 7 quantum wells, and the quantum well material can be selected as GaAs, InGaAs, or other III-V group semiconductor materials according to the target emission wavelength; the oxide confinement layer 4 can include a high-aluminum AlGaAs layer, and the thickness of the high-aluminum AlGaAs layer can be 20 nm to 100 nm. The p-type distributed Bragg reflector layer can include 15 to 35 pairs of p-type semiconductor layer pairs, each p-type semiconductor layer pair can include two AlGaAs-based semiconductor layers with different refractive indices.

[0029] The number of layers and materials mentioned above are merely feasible examples; the actual number of layers can be adjusted based on the target wavelength, reflectivity, series resistance, and epitaxial process capabilities.

[0030] In some embodiments, the first reflective portion 5 can be a reflective region near the electrode structure 8 in the p-type distributed Bragg reflector layer, and the second reflective portion 6 can be a reflective region near the oxide confinement layer 4 in the p-type distributed Bragg reflector layer. The first reflective portion 5 and the second reflective portion 6 can belong to different height segments of the same p-type distributed Bragg reflector layer, and no additional material interface or independent functional layer is required between them.

[0031] In other words, the first reflective portion 5 and the second reflective portion 6 can be formed by the same epitaxial growth process. The main difference between the two lies in their height position and whether a transverse cavity 7 is provided.

[0032] In some embodiments, the effective lateral dimension of the device mesa can be from 10 μm to 50 μm. The effective lateral dimension of the first reflective portion 5 can be from 8 μm to 40 μm, and the effective lateral dimension of the second reflective portion 6 after the lateral cavity 7 is provided can be from 3 μm to 25 μm, which is smaller than the effective lateral dimension of the first reflective portion 5. For example, in a high-speed vertical-cavity surface-emitting laser in the 850 nm or 940 nm band, the effective lateral dimension of the first reflective portion 5 can be from 18 μm to 24 μm, and the effective lateral dimension of the second reflective portion 6 can be from 6 μm to 14 μm.

[0033] Accordingly, the radial depth of the lateral cavity 7 extending inward from the outside of the device can be from 2 μm to 9 μm. These dimensions can be adjusted according to the target light-emitting aperture, current density, heat dissipation requirements, and mesa size.

[0034] In this application, the effective lateral dimension can be understood as the dimension of the corresponding structure in a plane parallel to the main surface of substrate 1, used to characterize the effective material distribution range. For structures with an approximately circular cross-section, the effective lateral dimension can be the diameter or an equivalent diameter; for structures with a non-circular, irregular, or locally open cross-section, the effective lateral dimension can be the equivalent width, the equivalent diameter, or a feature dimension determined by the outer contour of the effective semiconductor region.

[0035] For example, the area in a cross-section that still retains semiconductor material and can participate in current transmission can be converted into a circular area of ​​the same area, and the diameter of this circular area can be used as the equivalent lateral dimension. As long as the effective material distribution range of the second reflective portion 6 is smaller than that of the first reflective portion 5 due to the arrangement of the lateral cavity 7, the effective lateral dimension of the second reflective portion 6 can be considered smaller than that of the first reflective portion 5.

[0036] In some embodiments, the axial height of the second reflective portion 6 can be from 0.2 μm to 3 μm, corresponding to 2 to 20 pairs of semiconductor layer pairs in the p-type distributed Bragg reflector layer. A lateral cavity 7 can be formed within at least a portion of the height of the second reflective portion 6, causing the second reflective portion 6 to exhibit a locally reduced diameter structure on the side near the oxide confinement layer 4. If the axial height of the second reflective portion 6 is too small, its effect on regulating the injected current transmission path 9 and the peripheral mode environment may be insufficient; if the axial height of the second reflective portion 6 is too large, it may increase the processing difficulty and adversely affect the overall reflective performance, thermal diffusion path, or mechanical support of the p-type distributed Bragg reflector layer. Therefore, the axial height of the second reflective portion 6 can be selected based on the layer pair thickness, target current density, and target mode loss of the p-type distributed Bragg reflector layer.

[0037] In some embodiments, the lateral cavity 7 can be formed by lateral etching of the peripheral region of the second reflective portion 6 from the mesa sidewall of the device. Lateral etching can be achieved using wet selective etching, dry etching, reactive ion etching, plasma-assisted etching, or a combination thereof. For example, the sidewall of the p-type distributed Bragg reflector layer can be exposed first by mesa etching, and then lateral material removal can be performed within the height range corresponding to the second reflective portion 6 to form the lateral cavity 7. After the lateral cavity 7 is formed, the semiconductor conductive path in the peripheral region of the second reflective portion 6 is weakened, and the effective lateral dimension of the second reflective portion 6 is reduced relative to the first reflective portion 5.

[0038] Based on the above structure, when the injected current from electrode structure 8 is transmitted to the active region 3 through the p-type distributed Bragg reflector layer, it is affected by the local diameter reduction structure of the second reflector portion 6, causing the injected current transmission path 9 to be distributed towards the central region of the device before reaching the oxide confinement layer 4. Compared with the structure in which the current lateral contraction only occurs near the oxide confinement layer 4, this structure allows the injected current transmission path 9 to undergo spatial adjustment earlier within the p-type distributed Bragg reflector layer, thereby improving the coordination between the injected current distribution and the lateral mode control.

[0039] Meanwhile, the transverse cavity 7 changes the equivalent refractive index and loss distribution of the outer region of the second reflection part 6, making the mode components with more obvious transverse extension more susceptible to disturbance in the outer region, which is beneficial to improving the transverse mode selectivity of the vertical cavity surface-emitting laser.

[0040] It should be noted that the above size range is for illustrating some implementable examples of the structure of this application and does not mean that this application can only be implemented within these specific size ranges. For vertical cavity surface-emitting lasers with different emission wavelengths, different epitaxial material systems, or different target output powers, the height of the first reflective part 5 and the second reflective part 6, the effective lateral dimension, and the radial extension depth of the lateral cavity 7 can be adjusted according to optical simulation, current spread simulation, thermal simulation, or sample calibration results.

[0041] In some embodiments, the lateral cavity 7 is located in the peripheral region of the second reflective portion 6. This peripheral region can be understood as the area in the second reflective portion 6 closer to the mesa sidewall relative to the device's central axis. The central region in the second reflective portion 6 not occupied by the lateral cavity 7 can form a central semiconductor region. This central semiconductor region can be understood as the region in the second reflective portion 6 that maintains the continuity of the semiconductor material, providing a current path for transmission from the electrode structure 8 toward the active region 3 within the p-type distributed Bragg reflector layer.

[0042] By placing the lateral cavity 7 in the peripheral region instead of in the center of the second reflective portion 6, the semiconductor conductive path on the central side can be preserved while weakening the peripheral conductive path, making it easier for the injected current to travel downwards along the central region of the device.

[0043] In some embodiments, the oxide confinement layer 4 includes a central oxide hole. The central oxide hole can be understood as an unoxidized region or a region with relatively high conductivity and optical transmittance in the oxide confinement layer 4 through which injected current and light fields primarily pass. The central semiconductor region overlaps with the central oxide hole in a direction perpendicular to the substrate 1, allowing the current channel defined by the second reflective portion 6 to correspond to the lateral passage region defined by the oxide confinement layer 4.

[0044] In this way, the injected current is already distributed toward the central region when it passes through the second reflection part 6, and can continue to enter the central oxide hole of the oxide confinement layer 4 without needing to undergo a significant lateral turn near the oxide confinement layer 4.

[0045] For example, the effective lateral dimension of the central oxide aperture can be 2 μm to 12 μm, such as 3 μm to 8 μm; the effective lateral dimension of the central semiconductor region can be 4 μm to 18 μm, such as 6 μm to 14 μm, and is greater than or equal to the effective lateral dimension of the central oxide aperture. In this way, the second reflective portion 6 can pre-guide the injected current above the oxide confinement layer 4, but the retained central semiconductor region will not be smaller than the passage area corresponding to the central oxide aperture, thereby preventing the second reflective portion 6 itself from becoming a current bottleneck narrower than the central oxide aperture. In one example, the effective lateral dimension of the central oxide aperture is 6 μm, and the effective lateral dimension of the central semiconductor region is 10 μm. This allows for a more pronounced centralized current distribution above the oxide confinement layer 4 while reducing the risk of increased series resistance and localized heating caused by an excessively narrow central semiconductor region.

[0046] In some embodiments, the transverse cavity 7 can be a continuous annular cavity. The continuous annular cavity can be continuously distributed along the circumference of the second reflective portion 6 and positioned around the central region of the second reflective portion 6. The radial extension depth of the continuous annular cavity can be from 2 μm to 9 μm, and the axial height can be from 0.2 μm to 3 μm. With this structure, the second reflective portion 6 has good symmetry in the circumferential direction, which is beneficial for obtaining a more uniform current contraction effect and a more stable transverse optical field distribution.

[0047] In other embodiments, the lateral cavities 7 can also be multiple cavities spaced apart circumferentially. These cavities can be arranged around the central region of the second reflective portion 6, and the number of cavities can be from 4 to 32, for example, 6, 8, 12, or 16. The circumferential duty cycle of the multiple cavities can be from 30% to 85%, meaning that within the same lateral cross-section, the total angle occupied by the lateral cavities 7 in the circumferential direction can be from 30% to 85% of the full circumferential angle. Unremoved semiconductor material regions can be retained between any adjacent cavities to form mechanical support and heat conduction paths. This structure is advantageous in balancing mechanical strength, heat dissipation capacity, and process window while forming the lateral cavities 7.

[0048] In some implementations, the p-type distributed Bragg reflector layer comprises multiple pairs of semiconductor layers. Each pair of semiconductor layers may include two types of semiconductor layers with different refractive indices, such as a high-refractive-index AlGaAs-based semiconductor layer and a low-refractive-index AlGaAs-based semiconductor layer.

[0049] The physical thickness of a single semiconductor layer pair can be determined based on the target emission wavelength and the refractive index of the material; taking a GaAs-based VCSEL in the 850nm to 940nm band as an example, the physical thickness of each semiconductor layer pair can be in the range of approximately 100nm to 180nm. The p-type distributed Bragg reflector layer can include 15 to 35 pairs of semiconductor layer pairs, and the second reflective portion 6 can correspond to 2 to 20 pairs of semiconductor layer pairs near the oxide confinement layer 4.

[0050] In some embodiments, the lateral cavity 7 extends along a direction perpendicular to the substrate 1 through at least two of the plurality of semiconductor layer pairs. That is, the lateral cavity 7 does not correspond to a single thin-layer region, but forms a cavity region with a certain longitudinal coverage range within the height range of the second reflective portion 6. For example, the lateral cavity 7 may correspond to 2 to 12 pairs of semiconductor layer pairs, or it may correspond to 3 to 8 pairs of semiconductor layer pairs.

[0051] In this way, the peripheral region of the second reflective portion 6 is affected by the lateral cavity 7 within the height range of multiple semiconductor layer pairs, thereby enabling the second reflective portion 6 to form an effective lateral dimension that is smaller than that of the first reflective portion 5 more stably.

[0052] In some embodiments, to establish a clearly defined structural relationship between the lateral cavity 7 and the oxide confinement layer 4, the lateral cavity 7 may be located on the side of the second reflective portion 6 closest to the oxide confinement layer 4, and may not extend into the oxide confinement layer 4. At least one layer of semiconductor material may be retained between the lower boundary of the lateral cavity 7 and the oxide confinement layer 4, or a spacing distance of 0.05 μm to 1.0 μm may be maintained. For example, the lower boundary of the lateral cavity 7 may be 0.1 μm to 0.5 μm from the oxide confinement layer 4.

[0053] In this way, the transverse cavity 7 is mainly used to adjust the injection current distribution and transverse mode environment above the oxide confinement layer 4, while the oxide confinement layer 4 still maintains its own current and optical field confinement function, so as to prevent the transverse cavity 7 from destroying the formation and stability of the central oxide hole.

[0054] In some embodiments, the effective lateral dimension of the first reflective portion 5 can be denoted as the first dimension, the effective lateral dimension of the central semiconductor region can be denoted as the second dimension, and the effective lateral dimension of the central oxide hole can be denoted as the third dimension. The first dimension is larger than the second dimension, and the second dimension is greater than or equal to the third dimension. Further, the first dimension can be 1.2 to 4 times the second dimension, and the second dimension can be 1.0 to 2.5 times the third dimension.

[0055] For example, the first dimension can be 20μm, the second dimension can be 10μm, and the third dimension can be 6μm; or the first dimension can be 24μm, the second dimension can be 12μm, and the third dimension can be 8μm.

[0056] The aforementioned dimensional relationship allows for a larger lateral range on the side closer to electrode structure 8 to meet the needs of electrode contact, current expansion, and reflection structure arrangement; at the same time, it creates a relatively smaller but not too narrow current transmission region on the side closer to oxide confinement layer 4, thereby achieving a balance between current pre-distribution, series resistance control, and lateral mode adjustment.

[0057] It should be noted that the above size and quantity ranges are merely feasible examples for ease of explanation. For vertical cavity surface-emitting lasers with different target wavelengths, different material systems, different mesa sizes, or different output power levels, the dimensions of the central oxide hole, the central semiconductor region, the transverse cavity 7, and the first reflective portion 5 can be adjusted through current spread simulation, optical field mode simulation, thermal simulation, or sample calibration.

[0058] In some embodiments, the radial extension depth of the lateral cavity 7 can vary along a direction perpendicular to the substrate 1. Here, the radial extension depth can be understood as the distance the lateral cavity 7 extends from the outer side of the device towards the center of the device within a cross-section parallel to the main surface of the substrate 1. For an approximately circular mesa structure, the radial extension depth can correspond to the radial distance between the outer contour of the mesa and the inner boundary of the lateral cavity 7; for a non-circular mesa structure, the radial extension depth can correspond to the distance the lateral cavity 7 extends into the second reflective portion 6 along the corresponding lateral direction.

[0059] For example, the second reflective portion 6 can be divided along a direction perpendicular to the substrate 1 into an upper region near the first reflective portion 5, a transition region in the middle, and a lower region near the oxide confinement layer 4. The radial extension depth of the lateral cavity 7 in the upper region can be 2 μm to 4 μm, the radial extension depth in the transition region can be 4 μm to 7 μm, and the radial extension depth in the lower region can be 6 μm to 9 μm.

[0060] Therefore, the effective lateral dimension of the second reflective portion 6 can be gradually reduced from the side closer to the first reflective portion 5 to the side closer to the oxide confinement layer 4, thus forming a stepped effective lateral dimension. This structure enables the injected current transmission path 9 to be distributed gradually towards the central region within the second reflective portion 6, reducing the degree of abrupt contraction of the injected current transmission path 9 at a single height position.

[0061] For example, the radial extension depth of the lateral cavity 7 can also vary continuously along a direction perpendicular to the substrate 1, so that the second reflective portion 6 has a gradually changing effective lateral dimension. For example, the radial extension depth of the lateral cavity 7 can gradually increase from about 2 μm near the side of the first reflective portion 5 to about 8 μm near the side of the oxide confinement layer 4; correspondingly, the effective lateral dimension of the second reflective portion 6 can gradually decrease from about 16 μm to about 8 μm.

[0062] The aforementioned gradient structure can be achieved by controlling the lateral etching time, etching window height, etchant concentration distribution, etching selectivity, or multiple step-by-step etching. Compared to abrupt diameter reduction structures, gradient structures are better at reducing the degree of abrupt changes in current distribution and optical boundary conditions, thereby achieving a more balanced effect between current pre-distribution, mode perturbation, and process tolerance.

[0063] In some embodiments, the radial extension depth of the lateral cavity 7 can also be varied in the opposite way, for example, gradually decreasing from the side near the first reflective portion 5 to the side near the oxide confinement layer 4, or forming the maximum radial extension depth in the middle region of the second reflective portion 6. These different variations can be selected according to the target current distribution, target lateral mode loss, and thermal diffusion requirements. For devices that wish to enhance the current pre-convergence effect near the oxide confinement layer 4, the lateral cavity 7 can have a larger radial extension depth near the oxide confinement layer 4; for devices that wish to retain more heat conduction paths near the active region 3, the radial extension depth of the lateral cavity 7 near the oxide confinement layer 4 can be appropriately reduced.

[0064] In some embodiments, the cavity wall of the lateral cavity 7 may be covered with a passivation layer. The cavity wall can be understood as the semiconductor sidewall surface exposed within the cavity after the lateral cavity 7 is formed. Since the lateral cavity 7 can be formed by lateral etching, wet etching, or dry etching, surface states, etching damage, native oxide layers, or local defects may exist at the cavity wall. By forming a passivation layer on the cavity wall, the impact of cavity wall surface defects on carrier recombination, sidewall leakage current, and long-term device stability can be reduced.

[0065] For example, the passivation layer may include silicon oxide, silicon nitride, aluminum oxide, silicon oxynitride, or combinations thereof. The thickness of the passivation layer can be from 5 nm to 200 nm, for example, from 10 nm to 80 nm. If the passivation layer is too thin, its suppression of cavity wall defects and leakage paths may be insufficient; if the passivation layer is too thick, it may excessively alter the effective spatial size of the lateral cavity 7 or affect the equivalent refractive index distribution of the peripheral region of the second reflective portion 6. Therefore, the passivation layer thickness can be selected according to the size of the lateral cavity 7, the target mode loss, and reliability requirements.

[0066] In some embodiments, the passivation layer can be formed by plasma-enhanced chemical vapor deposition (PECVD), atomic layer deposition (ALD), low-pressure chemical vapor deposition (LPCVD), sputtering deposition, or thermal oxidation. For structures where the lateral cavity 7 has a large aspect ratio or where the cavity walls are difficult to cover, ALD can be used to form the passivation layer to improve the uniformity of cavity wall coverage. For structures where the lateral cavity 7 has a large opening and the cavity walls are easy to cover, plasma-enhanced chemical vapor deposition can be used to form a silicon oxide layer or a silicon nitride layer. After the passivation layer is formed, the lateral cavity 7 can still remain a cavity structure, and the passivation layer mainly covers the cavity wall surface without completely filling the lateral cavity 7.

[0067] In some embodiments, the cavity wall may be surface-cleaned or lightly wet-treated before the passivation layer is formed to remove etching residues or native oxides. For example, the cavity wall surface may be treated with at least one of the following methods: deionized water cleaning, dilute acid cleaning, sulfur-containing solution treatment, or plasma pretreatment, before the passivation layer is deposited.

[0068] The above processing method can be selected based on the material system of the p-type distributed Bragg reflector and the compatibility with subsequent electrode processes. After the passivation layer is formed, the insulating layer deposition, windowing, and electrode structure 8 can continue, enabling the electrode structure 8 to achieve electrical connection with the p-type distributed Bragg reflector.

[0069] See Figure 3 The flowchart below illustrates a method for fabricating a vertical-cavity surface-emitting laser, as provided in this application embodiment, including steps S101 to S106, wherein: S101: An n-type distributed Bragg reflector layer, an active region, an oxide confinement layer, and a p-type distributed Bragg reflector layer are sequentially formed on a substrate, wherein the p-type distributed Bragg reflector layer comprises a plurality of semiconductor layer pairs stacked along a direction perpendicular to the substrate; S102: Etch the p-type distributed Bragg reflector layer to form a device mesa; S103: Selectively oxidize the oxide confinement layer to form a central oxide pore; S104: Laterally etch the reflective portion of the p-type distributed Bragg reflector layer near the active region from the outside of the device mesa to form a transverse cavity extending inward from the outside of the vertical cavity surface emitter laser. S105: During the lateral etching process, a probe light is applied to the p-type distributed Bragg reflector layer in a direction perpendicular to the substrate, and a reflection signal formed by the probe light after reflection by the p-type distributed Bragg reflector layer is collected. The lateral etching is then controlled based on the reflection signal. S106: Form an electrode structure electrically connected to the p-type distributed Bragg reflector layer.

[0070] See Figure 3 This method can be used to prepare, for example Figure 1 and Figure 2 The diagram shows a transverse cavity type vertical cavity surface-emitting laser. The method includes steps S101 to S106. It should be noted that... Figure 3 The sequence of steps in the document is used to illustrate an feasible preparation process. Without affecting the formation of the transverse cavity 7 and the electrical connection of the electrode structure 8, some cleaning, mask formation, insulating layer formation, annealing, or testing steps can be inserted between the corresponding steps according to the process conditions.

[0071] In step S101, an n-type distributed Bragg reflector layer 2, an active region 3, an oxide confinement layer 4, and a p-type distributed Bragg reflector layer are sequentially formed on substrate 1. Specifically, the epitaxial structure can be formed on the GaAs substrate using metal-organic chemical vapor deposition or molecular beam epitaxy. The n-type distributed Bragg reflector layer 2 may include 25 to 40 pairs of n-type semiconductor layer pairs, and the p-type distributed Bragg reflector layer may include 15 to 35 pairs of p-type semiconductor layer pairs. Each semiconductor layer pair may include two types of semiconductor layers with different refractive indices, such as AlGaAs-based semiconductor layers with different Al compositions. For GaAs-based vertical-cavity surface-emitting lasers in the 850nm to 940nm wavelength range, the thickness of each layer in a single semiconductor layer pair can be set according to the quarter-wavelength condition, and the actual physical thickness can be determined by combining the material refractive index and the target reflection wavelength.

[0072] The active region 3 may include one or more quantum wells and barrier layers located on both sides of the quantum wells. The oxide confinement layer 4 may be a high-aluminum AlGaAs layer, such as an AlGaAs layer with an Al content greater than 0.90, and its thickness may be 20 nm to 100 nm. This oxide confinement layer 4 forms a central oxide hole during subsequent selective oxidation to define the lateral passage region of current and light field. The side of the p-type distributed Bragg reflector layer closest to the electrode structure 8 may subsequently serve as the first reflective portion 5, and the side closest to the oxide confinement layer 4 may subsequently serve as the second reflective portion 6.

[0073] In step S102, the p-type distributed Bragg reflector layer is etched to form a device mesa. Specifically, a photoresist mask or hard mask can be formed on the p-type distributed Bragg reflector layer, and then the device mesa is formed using inductively coupled plasma etching, reactive ion etching, or wet etching. The effective lateral dimension of the device mesa can be from 10 μm to 50 μm, for example, from 18 μm to 30 μm. The mesa etching can at least expose the sidewalls of the oxide confinement layer 4 for subsequent selective oxidation; it can also further expose the sidewall region in the p-type distributed Bragg reflector layer where the lateral cavity 7 is to be formed, for subsequent lateral etching from the outside of the device mesa.

[0074] In step S103, the oxide confinement layer 4 is selectively oxidized to form a central oxide hole. Specifically, the epitaxial wafer forming the device mesa can be placed in a moisture-containing oxidation environment, causing the oxide confinement layer 4 to undergo lateral oxidation from the sidewall of the device mesa towards the center, thereby forming a peripheral oxidized region and a central unoxidized region in the oxide confinement layer 4. This central unoxidized region can serve as the central oxide hole. The selective oxidation temperature can be from 360°C to 450°C, and the oxidation time can be determined based on the Al composition, thickness, mesa size, and target central oxide hole size of the oxide confinement layer 4.

[0075] For example, in one embodiment, the effective lateral dimension of the central oxide pore can be controlled between 2 μm and 12 μm, and further between 3 μm and 8 μm.

[0076] In S104, the reflective portion of the p-type distributed Bragg reflector layer near the active region 3 is etched laterally from the outside of the device mesa to form a transverse cavity 7 extending inward from the outside of the vertical cavity surface-emitting laser.

[0077] Specifically, a mask structure can be formed on the outer side of the mesa to define the range of lateral etching height, exposing a region of the p-type distributed Bragg reflector layer near the oxide confinement layer 4 to the etching environment. Subsequently, this region can be laterally material-removed using wet selective etching, dry etching, plasma-assisted etching, or a combination of the above methods, thereby forming a lateral cavity 7. The radial extension depth of the lateral cavity 7 can be 2 μm to 9 μm, and the axial height can be 0.2 μm to 3 μm, or correspond to 2 to 20 pairs of semiconductor layers in the p-type distributed Bragg reflector layer.

[0078] In some embodiments, the lateral etching in S104 can be performed after the formation of the central oxide hole, such that the lateral cavity 7 is located above and spaced apart from the oxide confinement layer 4.

[0079] Thus, the lateral cavity 7 is mainly used to regulate the current pre-distribution and peripheral mode environment within the p-type distributed Bragg reflector layer, while the oxide confinement layer 4 is still used to define the central through region near the active region 3. In other embodiments, the order of lateral etching and selective oxidation can also be adjusted according to the epitaxial material selectivity and process window, but the lateral cavity 7 in the final structure should be formed in the second reflective portion 6 of the p-type distributed Bragg reflector layer.

[0080] For example, in the case of wet selective etching, a mixed solution of dilute hydrochloric acid, citric acid and hydrogen peroxide, a mixed solution of phosphoric acid and hydrogen peroxide, or a buffered hydrofluoric acid solution can be used as the etchant, and the rate and depth of lateral etching can be controlled by adjusting the etchant concentration, temperature, and etching time. In the case of dry etching, a plasma etching process using boron trichloride, chlorine, silicon tetrachloride, or a mixture thereof as the reactant gas can be used, and the rate and directionality of lateral etching can be controlled by adjusting the RF power, bias power, reactant gas flow rate, and chamber pressure. When using a combination of the above methods, sidewall openings can be formed first by dry etching, and then further lateral material removal can be performed along specific aluminum composition layers by wet selective etching.

[0081] In step S105, during the lateral etching process, a probe light is applied to the p-type distributed Bragg reflector layer in a direction perpendicular to the substrate 1, and the reflection signal formed by the reflection of the probe light through the p-type distributed Bragg reflector layer is collected. The lateral etching is then controlled based on the reflection signal.

[0082] It should be noted that S105 can be executed synchronously with S104, that is, the probe light is continuously or intermittently applied and the reflected signal is collected during the lateral etching process, but it does not mean that S105 must be executed after S104 is completed.

[0083] For example, the probe light can be provided by a broadband light source, a tunable laser, or a narrowband laser, and illuminated above the device stage via an objective lens, a fiber collimator, or a microscopic reflectance measurement optical path. The wavelength range of the probe light can cover the reflection bands or edges of the p-type distributed Bragg reflector.

[0084] For example, for 850nm band devices, the probe light can cover the range of 800nm ​​to 900nm; for 940nm band devices, the probe light can cover the range of 880nm to 1000nm. The irradiation area of ​​the probe light on the p-type distributed Bragg reflector layer can cover at least a portion of the peripheral region of the second reflective portion 6, so that the reflected signal can reflect the formation process of the transverse cavity 7.

[0085] During lateral etching, as the lateral cavity 7 gradually forms from the outside of the device inward, the equivalent refractive index distribution, semiconductor material continuity, and local reflection boundaries of the peripheral region of the second reflective portion 6 change. Since the p-type distributed Bragg reflector layer comprises multiple semiconductor layer pairs stacked along a direction perpendicular to the substrate 1, optical coupling occurs between the lateral cavity 7 and these semiconductor layer pairs, causing the reflected signal to change with the etching process. Based on this change, the progress of lateral etching can be assessed, for example, to determine whether the lateral cavity 7 is approaching the target radial extension depth.

[0086] For example, the intensity of the reflection peak, the position of the reflection peak, the slope of the reflection band edge, or the spectral oscillation characteristics can be pre-selected as monitoring indicators. When the monitoring indicators reach a predetermined range corresponding to the target lateral cavity depth, lateral etching is stopped; when the monitoring indicators show that the lateral cavity 7 is not formed sufficiently, etching can continue; when the monitoring indicators deviate from the target range, the etching rate, etchant concentration, or etching environment temperature can be adjusted.

[0087] In this way, the formation depth of the lateral cavity 7 can be controlled in situ during the lateral etching process, reducing the process deviation caused by relying solely on a fixed etching time.

[0088] In step S106, an electrode structure 8 electrically connected to the p-type distributed Bragg reflector layer is formed. Specifically, after forming the lateral cavity 7, the device surface can be cleaned, and an insulating layer, an opening region, and a metal layer can be formed as needed. The electrode structure 8 may include a p-type electrode portion located above the p-type distributed Bragg reflector layer, such as a p-type ring electrode; it may also include an n-type electrode portion electrically connected to the n-type distributed Bragg reflector layer 2 or the substrate 1. The metal layer can be formed by evaporation, sputtering, electroplating, or stripping processes, and ohmic contact performance can be improved by rapid thermal annealing.

[0089] Through the above steps S101 to S106, a lateral cavity 7 can be formed on the side of the p-type distributed Bragg reflector layer near the oxide confinement layer 4 while maintaining the basic epitaxial hierarchical structure of the vertical-cavity surface-emitting laser (VCSEL). The lateral cavity 7 creates a region with a smaller effective lateral dimension for the second reflective portion 6 relative to the first reflective portion 5, thereby pre-guiding the injected current before it enters the oxide confinement layer 4, and simultaneously altering the equivalent refractive index and loss distribution of the peripheral region of the second reflective portion 6. This improves the coordination between the injected current distribution and lateral mode control in the VCSEL.

[0090] Optionally, during the formation of the lateral cavity 7, the radial extension depth of the lateral cavity 7 directly affects the effective lateral dimension of the second reflective portion 6. If the radial extension depth of the lateral cavity 7 is insufficient, the peripheral region of the second reflective portion 6 will still retain a large number of semiconductor conductive paths, and the effect of current distribution to the central region and the effect of lateral mode perturbation may be insufficient. If the radial extension depth of the lateral cavity 7 is too large, the semiconductor region retained in the second reflective portion 6 may be too narrow, leading to increased risks of series resistance, local thermal resistance, or structural reliability.

[0091] Therefore, in addition to setting the etching time in advance, the formation state of the transverse cavity 7 can be monitored in situ during the lateral etching process, so as to more stably control the radial extension depth of the transverse cavity 7.

[0092] In some embodiments, the optical reflection properties of the p-type distributed Bragg reflector layer itself can be used to monitor the lateral etching process. The p-type distributed Bragg reflector layer comprises multiple pairs of semiconductor layers stacked along a direction perpendicular to the substrate 1. Each semiconductor layer pair may include semiconductor layers with different refractive indices, thus this layer structure exhibits a strong reflection response to probe light in a specific wavelength band. During the lateral etching process, as the lateral cavity 7 is gradually formed from the outside to the inside of the device mesa, the semiconductor material in the peripheral region of the second reflective portion 6 is gradually removed, and the radial extension depth of the lateral cavity 7 gradually increases, causing changes in the equivalent refractive index distribution, local reflection boundary, and optical path distribution of this region. These changes cause the reflected signal formed by the probe light reflected by the p-type distributed Bragg reflector layer to change with the lateral etching process.

[0093] In this application, the optical coupling formed between the lateral cavity 7 and the multiple semiconductor layer pairs can be understood as follows: after the lateral cavity 7 is formed, the cavity region and the regions where the multiple semiconductor layer pairs are located are spatially adjacent, corresponding, or partially overlapping, causing the periodic optical reflection characteristics of the multiple semiconductor layer pairs to be affected by the lateral cavity 7. This effect can manifest as changes in the intensity, peak position, band edge variation trend, or local spectral shape of the reflected signal with the radial extension depth of the lateral cavity 7.

[0094] The optical coupling here does not require that the transverse cavity 7 and each semiconductor layer pair have the same interaction strength, nor is it limited to forming a specific resonant mode; as long as the formation of the transverse cavity 7 can make the reflection response of the probe light produce a identifiable change with the lateral etching process, it can be used to characterize the formation state of the transverse cavity 7.

[0095] In some implementations, the reflected signal may include a reflection spectrum. The reflection spectrum can be acquired using a broadband light source and a spectrometer, or using a tunable light source and a photodetector. For example, a broadband probe light can be directed perpendicularly to the substrate 1 onto the device mesa, and the reflected light returned from the p-type distributed Bragg reflector layer can be acquired to obtain a spectral curve showing the reflection intensity as a function of wavelength. For 850nm devices, the probe light can cover the 800nm ​​to 900nm range; for 940nm devices, the probe light can cover the 880nm to 1000nm range. The sampling interval for the reflection spectrum can be 0.1 seconds to 10 seconds, and the spectral resolution can be 0.1nm to 5nm. The above parameters are feasible examples and can be adjusted according to the target wavelength, etching rate, light source bandwidth, and spectrometer performance.

[0096] In some implementations, reflectance spectral features can be extracted from the reflectance spectrum. Reflectance spectral features can be understood as quantitative data that characterizes how the reflectance spectrum changes with the lateral etching process.

[0097] In practice, a characteristic band can be selected in the reflection spectrum first. This characteristic band can be located within the high-reflection band of the p-type distributed Bragg reflector layer, near the edge of the reflection band, or in a band that is sensitive to the formation process of the lateral cavity 7. Subsequently, the reflection spectrum within this characteristic band can be normalized, and the average reflection intensity, integrated reflection intensity, peak position, band edge slope, or curve shape change within this characteristic band can be extracted as characteristic quantities of the reflection spectrum. The normalization process can use the initial reflection spectrum before the start of lateral etching as a reference, or it can use the reflection spectrum of a reference sample as a reference.

[0098] For example, an initial reflection spectrum, denoted as initial spectrum A, can be acquired before lateral etching begins. During lateral etching, the current reflection spectrum, denoted as current spectrum B, is acquired at predetermined time intervals. Subsequently, the degree of change of current spectrum B relative to initial spectrum A is compared within a preset wavelength window. For example, the overall change in reflection intensity, the shift of reflection peaks, or the deviation of curve shape within this wavelength window is calculated, and the calculation results are used as reflection spectrum characteristic quantities. If the actual reflection spectrum contains noise, moving average, low-pass filtering, or multiple sampling averaging can be applied to the reflection spectrum to improve the stability of the characteristic quantities.

[0099] In some embodiments, the current radial extension depth of the transverse cavity 7 can be determined based on a predetermined mapping relationship between the reflection spectrum characteristics and the radial extension depth of the transverse cavity 7. This predetermined mapping relationship is established based on the periodic optical reflection characteristics of multiple semiconductor layer pairs and the change in equivalent refractive index distribution caused by the transverse cavity 7.

[0100] Specifically, multiple semiconductor layers give the p-type distributed Bragg reflector layer reflection spectrum characteristics related to layer thickness, refractive index, and target reflection wavelength; the formation of the transverse cavity 7 changes the local material distribution and equivalent refractive index of the peripheral region of the second reflective portion 6, causing the reflection spectrum characteristic quantities to change with the radial extension depth. Based on this correspondence, the reflection spectrum characteristic quantities can be converted into the current radial extension depth of the transverse cavity 7.

[0101] For example, multiple calibration samples can be prepared, each with a different known radial extension depth of the lateral cavity, such as 2 μm, 4 μm, 6 μm, 8 μm, and 10 μm. The same probe optical path is used to collect the reflectance spectra of each calibration sample, and the reflectance spectral features are extracted according to the same rules.

[0102] Subsequently, a lookup table, piecewise linear relationship, or fitting curve can be established between the reflection spectrum features and the radial extension depth. During the actual lateral etching process, the currently acquired reflection spectrum features are substituted into this predetermined mapping relationship to determine the current radial extension depth of the lateral cavity 7.

[0103] In another example, the predetermined mapping relationship can also be established through a combination of optical simulation and a small number of calibration samples. An optical model can be built based on the material refractive index, layer thickness, target reflection wavelength, and radial extension depth of multiple semiconductor layer pairs in a p-type distributed Bragg reflector layer, calculating the reflection spectral response at different radial extension depths. The simulation results are then corrected using a small number of samples at known depths. The optical model can employ a transfer matrix model, a finite difference time-domain model, or a finite element model. This predetermined mapping relationship can simultaneously reflect the periodic optical reflection characteristics of multiple semiconductor layer pairs and the equivalent refractive index distribution changes caused by the transverse cavity, thereby improving the reliability of the determination of the current radial extension depth.

[0104] For example, with the target radial extension depth set to 6 μm, the reflection spectrum can be acquired every second during lateral etching, and the reflection spectrum characteristics can be extracted from the selected wavelength window. If the current radial extension depth is determined to be 3 μm according to the predetermined mapping relationship, lateral etching continues; if the current radial extension depth is determined to be in the range of 5.5 μm to 6.5 μm, it is considered that the lateral cavity 7 has approached the target depth, and lateral etching can be terminated or switched to a low-speed etching state; if the reflection spectrum characteristics do not significantly match the predetermined trend, etching can be paused and the probe optical path, etchant concentration, etching temperature, or sample position can be checked. In this way, batch fluctuations caused by relying solely on a fixed etching time to control the depth of the lateral cavity 7 can be reduced, making the effective lateral dimension of the second reflective part 6 closer to the design value.

[0105] Optionally, to improve the stability of determining the radial extension depth of the transverse cavity 7 based on the reflection spectrum, one or more reflection spectral features can be extracted from the reflection spectrum. These features are not limited to a single spectral parameter but can be selected based on the reflection band position of the p-type distributed Bragg reflector layer, the probe light band, the etching rate, and the spectral noise. By using multiple reflection spectral features simultaneously or selectively, the risk of misjudgment due to fluctuations in light source intensity, sample position shifts, or local surface scattering caused by a single feature can be reduced.

[0106] For example, the reflectance spectrum characteristics may include the displacement of the reflectance peak position. Specifically, before the lateral etching begins, an initial peak wavelength of a reflectance peak can be determined from the initial reflectance spectrum; during the lateral etching process, the current peak wavelength of the same reflectance peak can be determined from the current reflectance spectrum; and the offset of the current peak wavelength relative to the initial peak wavelength can be used as the displacement of the reflectance peak position.

[0107] The displacement can be expressed in nanometers; for example, peak position changes on the order of 0.2 nm, 0.5 nm, or 1 nm can all be considered identifiable spectral changes. Since the gradual formation of the transverse cavity 7 alters the equivalent refractive index distribution of the peripheral region of the second reflective portion 6, the position of the reflection peak may shift with radial extension depth.

[0108] In another example, the reflectance spectrum characteristic can include the change in the intensity of the reflection peak. Specifically, the peak intensity of a specific reflection peak in the initial reflectance spectrum can be used as the reference intensity, the peak intensity of the corresponding reflection peak in the current reflectance spectrum can be used as the current intensity, and the change in the reflectance peak intensity can be determined based on the difference or ratio between the two. To reduce the impact of light source output fluctuations, the reflectance spectrum can be normalized first, for example, using the average reflectance intensity of a reference band, the reflectance intensity of an un-etched region, or the reflectance intensity of a standard reflector as the normalization reference.

[0109] For example, the reflectance spectrum characteristic quantity can include the change in the slope of the reflection band edge. The reflection band edge can be understood as the band in the reflectance spectrum of a p-type distributed Bragg reflector that transitions from a high-reflectance region to a low-reflectance region. A wavelength window, for example, with a width of 2 nm to 20 nm, can be selected near the reflection band edge, and the slope or average slope of the reflection intensity as a function of wavelength within this window can be calculated. As the transverse cavity 7 is formed, the equivalent refractive index and local reflection conditions of the peripheral region of the second reflective portion 6 change, and the slope of the reflection band edge may change. Therefore, this slope change can be used to help determine the degree of formation of the transverse cavity 7.

[0110] For example, a reflectance spectrum characteristic may include the spectral oscillation period. Here, the spectral oscillation period can be understood as the wavelength interval between adjacent local peaks, adjacent local valleys, or adjacent peaks and valleys in the reflectance spectrum. In other embodiments, when analyzing the evolution sequence of reflectance intensity within a selected wavelength window as a function of lateral etching time, the time interval between adjacent local peaks or adjacent local valleys can also be used as an oscillation characteristic of the reflectance spectrum evolution. For broadband reflectance spectra, multiple local peaks or valleys can be identified within a selected wavelength window, and the spectral oscillation period can be obtained based on the spacing between adjacent peaks or valleys. Changes in the spectral oscillation period can reflect changes in the optical coupling state between the lateral cavity 7 and multiple semiconductor layer pairs.

[0111] For example, the reflectance spectrum characteristics may include characteristic parameters of the resonant sidebands introduced by the transverse cavity 7. A resonant sideband can be understood as a local additional peak, shoulder, sideband undulation, or local depression appearing near the main reflectance peak or reflectance band in the reflectance spectrum. Characteristic parameters of the resonant sideband may include sideband position, sideband intensity, sideband width, wavelength interval between the sideband and the main peak, or the ratio of sideband intensity to the main peak intensity.

[0112] It should be noted that the resonance sideband is not required to appear clearly in all embodiments. When the resonance sideband can be stably identified in the actual spectrum, it can be used as one of the characteristic quantities of the reflection spectrum. When the resonance sideband is not obvious, the reflection peak position, reflection peak intensity, reflection band slope, or spectral oscillation period can also be used as the characteristic quantities of the reflection spectrum.

[0113] In some implementations, the aforementioned reflection spectrum characteristics can be used individually or in combination. For example, the displacement of the reflection peak position can be used as the primary determination factor, and the change in reflection peak intensity can be used as a verification factor; alternatively, the change in the slope of the reflection band edge and the spectral oscillation period can be used together to determine whether the transverse cavity 7 has entered the vicinity of the target depth. In cases of combined use, multiple determination conditions can be preset. For instance, when the displacement of the reflection peak position enters the first target range and the change in reflection peak intensity enters the second target range, it is determined that the current radial extension depth of the transverse cavity 7 has reached the vicinity of the target depth. This approach improves tolerance to noise and localized process disturbances.

[0114] In some implementations, lateral etching can be terminated when the determined current radial extension depth reaches the target depth. The target depth can be determined based on the target effective lateral dimension of the second reflective portion 6, the target dimension of the central semiconductor region, and the dimension of the central oxide hole.

[0115] For example, with an effective lateral dimension of 20 μm for the device mesa and an effective lateral dimension of 10 μm for the target central semiconductor region, the target radial extension depth of the lateral cavity 7 from one side inward can be set to approximately 5 μm; considering process deviations, the target depth range can be set to 4.5 μm to 5.5 μm. When the current radial extension depth is determined to be within this range based on the reflectance spectrum characteristics, the etchant supply can be stopped, plasma etching can be stopped, or the sample can be removed from the etching environment to terminate the lateral etching.

[0116] In other implementations, lateral etching can be adjusted based on the difference between the current radial extension depth and the target depth. For example, when the current radial extension depth is significantly less than the target depth, the current etching conditions can be maintained or the etching rate can be appropriately increased; when the current radial extension depth is close to the target depth, the etching rate can be reduced to avoid over-etching; when the current radial extension depth exceeds the target depth or the trend is abnormal, lateral etching can be immediately terminated and the abnormal batch recorded. The etching rate can be changed by adjusting the plasma power, reactive gas flow rate, etchant concentration, etching environment temperature, or sample exposure conditions. The specific control method can be determined based on the dry or wet etching process used.

[0117] In the wet lateral etching example, the etching rate can be adjusted by the etchant concentration and the etching environment temperature. If the difference between the current radial extension depth and the target depth is large, a higher etching rate can be used; if the current radial extension depth is close to the target depth, for example, the difference is less than 0.5 μm, the etchant concentration or the etching environment temperature can be reduced, so that the lateral etching enters a slower finishing stage.

[0118] In dry or plasma-assisted etching examples, the etching rate can be varied by adjusting the RF power, bias power, reactive gas flow rate, or chamber pressure. The above examples illustrate feedback control methods; actual process parameters can be determined based on the material system and etching selectivity of the p-type distributed Bragg reflector layer.

[0119] In some embodiments, the area of ​​the probe light irradiated on the p-type distributed Bragg reflector layer at least covers the peripheral region of the second reflective portion 6. Here, "coverage" can be understood as: the projection range of the probe light spot on the main surface of the substrate 1 at least partially overlaps with the peripheral region of the second reflective portion 6, enabling the reflected signal to include the optical response of the region where the transverse cavity 7 is formed. If the probe light only irradiates the central region of the second reflective portion 6, and does not irradiate or substantially does not irradiate the region where the transverse cavity 7 is formed, the acquired reflected signal may primarily reflect the optical response of the unetched central region, making it difficult to accurately reflect the radial extension depth of the transverse cavity 7.

[0120] For example, the probe light spot can be set to be circular or approximately circular, and the spot diameter can be 50% to 120% of the effective lateral dimension of the device mesa. When the effective lateral dimension of the device mesa is 20 μm, the probe light spot diameter can be set to 10 μm to 24 μm, and the center of the spot is substantially aligned with the center of the device mesa; in this way, the spot can simultaneously cover the central region and the peripheral lateral cavity formation region. In another example, ring illumination, off-center illumination, or scanning illumination can also be used to make the probe light focus on covering the peripheral region of the second reflective portion 6.

[0121] For example, the probe light can be focused near the edge of the stage using a microscope objective and scanned circumferentially to obtain the formation state of the transverse cavity 7 at different circumferential positions.

[0122] In some implementations, to improve the stability of the reflected signal, the position of the probe light-illuminated area can be recorded when acquiring the reflected spectrum, and the relative position of the light spot and the device mesa can be kept constant during lateral etching. For arrayed devices, one or more representative light-emitting units in the array can also be selected as monitoring units, so that the probe light-illuminated area covers the peripheral area of ​​the second reflective portion 6 in the corresponding light-emitting unit.

[0123] This allows the reflection spectrum characteristics to more directly reflect the changes in the region where the transverse cavity 7 is formed, thereby improving the reliability of radial extension depth determination and lateral etching feedback control.

[0124] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of this application, and are not intended to limit them. Although this application has been described in detail with reference to the foregoing embodiments, those skilled in the art should understand that modifications can still be made to the technical solutions described in the foregoing embodiments, or equivalent substitutions can be made to some or all of the technical features therein. Such modifications or substitutions do not cause the essence of the corresponding technical solutions to deviate from the spirit and scope of the technical solutions of the embodiments of this application.

Claims

1. A transverse cavity type vertical cavity surface-emitting laser, characterized in that, include: Substrate; An n-type distributed Bragg reflector layer is disposed above the substrate; The active region is positioned above the n-type distributed Bragg reflector layer. An oxidation confinement layer is disposed above the active region; A p-type distributed Bragg reflector layer is disposed above the oxide confinement layer; as well as An electrode structure electrically connected to the p-type distributed Bragg reflector layer; The p-type distributed Bragg reflector layer includes a first reflective portion close to the electrode structure and a second reflective portion located between the first reflective portion and the oxide confinement layer. The second reflective portion has a transverse cavity extending inward from the outer side of the vertical cavity surface-emitting laser, and the effective transverse dimension of the second reflective portion is smaller than the effective transverse dimension of the first reflective portion; The lateral cavity is located in the peripheral region of the second reflective portion, and the second reflective portion includes a central semiconductor region defined by the lateral cavity; The oxidation confinement layer includes a central oxide hole, and the central semiconductor region overlaps with the central oxide hole in a direction perpendicular to the substrate; The effective lateral dimension of the central semiconductor region is greater than or equal to the effective lateral dimension of the central oxide hole; The lateral cavity is located on the side of the second reflective portion near the oxide confinement layer and does not extend into the oxide confinement layer.

2. The vertical-cavity surface-emitting laser according to claim 1, characterized in that, The transverse cavity is a continuous annular cavity or multiple cavities spaced apart along the circumference.

3. The vertical-cavity surface-emitting laser according to claim 1, characterized in that, The p-type distributed Bragg reflector layer includes multiple semiconductor layer pairs, and the lateral cavity extends through at least two of the multiple semiconductor layer pairs in a direction perpendicular to the substrate.

4. The vertical-cavity surface-emitting laser according to claim 1, characterized in that, The effective lateral dimension of the first reflective portion is a first dimension, the effective lateral dimension of the central semiconductor region is a second dimension, and the effective lateral dimension of the central oxide hole is a third dimension. The first dimension is greater than the second dimension, and the second dimension is greater than or equal to the third dimension.

5. The vertical-cavity surface-emitting laser according to claim 1, characterized in that, The radial extension depth of the lateral cavity varies along a direction perpendicular to the substrate, giving the second reflective portion a stepped or gradient effective lateral dimension.

6. The vertical-cavity surface-emitting laser according to claim 1, characterized in that, The walls of the transverse cavity are covered with a passivation layer.

7. A method for fabricating a vertical-cavity surface-emitting laser according to any one of claims 1 to 6, characterized in that, include: An n-type distributed Bragg reflector layer, an active region, an oxide confinement layer, and a p-type distributed Bragg reflector layer are sequentially formed on a substrate. The p-type distributed Bragg reflector layer comprises multiple pairs of semiconductor layers stacked along a direction perpendicular to the substrate. The p-type distributed Bragg reflector layer is etched to form a device mesa; The oxide confinement layer is selectively oxidized to form a central oxide pore; Laterally etch the reflective portion of the p-type distributed Bragg reflector layer near the active region from the outside of the device mesa to form a transverse cavity extending inward from the outside of the vertical cavity surface emitter laser. During the lateral etching process, a probe light is applied to the p-type distributed Bragg reflector layer in a direction perpendicular to the substrate, and the reflected signal formed by the probe light after being reflected by the p-type distributed Bragg reflector layer is collected. The lateral etching is then controlled based on the reflected signal. as well as An electrode structure is formed that is electrically connected to the p-type distributed Bragg reflector layer.

8. The method according to claim 7, characterized in that, The reflected signal changes as the lateral etching progresses, and the change is caused by the gradual formation of the lateral cavity in the p-type distributed Bragg reflector layer and the optical coupling formed between the lateral cavity and the plurality of semiconductor layer pairs.

9. The method according to claim 8, characterized in that, The reflected signal includes a reflection spectrum, and the method further includes: Extract reflectance spectral features from the reflectance spectrum; The current radial extension depth of the transverse cavity is determined based on the predetermined mapping relationship between the reflection spectrum feature quantity and the radial extension depth of the transverse cavity. The predetermined mapping relationship is established based on the periodic optical reflection characteristics of the plurality of semiconductor layer pairs and the change in equivalent refractive index distribution caused by the transverse cavity.

10. The method according to claim 9, characterized in that, The reflection spectrum characteristics include at least one of the following: the displacement of the reflection peak position, the change in the reflection peak intensity, the change in the slope of the reflection band edge, the spectral oscillation period, or the characteristic parameters of the resonance sideband introduced by the transverse cavity.

11. The method according to claim 9, characterized in that, The lateral etching is terminated when the current radial extension depth reaches the target depth; or... Based on the difference between the current radial extension depth and the target depth, at least one of the lateral etching rate, etchant concentration, or etching ambient temperature is adjusted.

12. The method according to claim 8, characterized in that, The area of ​​illumination of the probe light on the p-type distributed Bragg reflector layer at least covers the peripheral area of ​​the second reflective portion.

Images