Method of obtaining high-power single-mode laser

Abstract

The present invention relates to a method (100) for obtaining high-power single-mode edge-emitting semiconductor laser designs.

Description

[0001] METHOD OF OBTAINING HIGH-POWER SINGLE-MODE LASER

[0002] Technical Field

[0003] The present invention relates to a method for obtaining high-power single-mode edge-emitting semiconductor laser designs.

[0004] Background of the Invention

[0005] Semiconductor lasers have wide range of application areas such as telecommunications, biology and optoelectronic devices and have completely changed the field of optical technology. Vertical-cavity surface-emitting lasers (VCSELs) and edge-emitting lasers (EELs) constitute two main classes of diode lasers. VCSELs have the capacity to emit high-quality spectral and spatial Gaussian beams. On the other hand, EELs are used when higher powers are required.

[0006] Laser beam quality is a critical parameter for laser devices. This parameter, measured by the M2factor, is a dimensionless value used to quantitatively identify beam quality. It is calculated by using the waist width and divergence of a laser beam. The higher the M2value, the worse the beam quality. The ideal M2value is 1, which is the value of a Gaussian beam with a bell-shaped peak along its central optical axis. Single-mode laser light (with an M2value of 1) provides optimal focusing, minimal divergence, predictable propagation and efficient coupling, making it highly desirable in a broad spectrum of applications. A Gaussian beam with an M2value of 1 can be focused to the smallest spot size, maximizing the intensity at the focal point. This is of vital importance for applications such as laser cutting, laser processing and laser welding, where precision and high-power density are required. It is therefore crucial to combine single-mode beam profile with high-power EELs.

[0007] The main approach for improving beam quality while maintaining high powers is to optimize the epitaxial design. In an optimal epitaxial design, the output power can be further increased by increasing the width and length of the laser waveguide (cavity). Wider waveguides increase the number of modes. There is a width limit for the emergence of higher-order modes, which restricts applications that require higher power. Reducing cross-sectional areas to maintain beam quality and achieve higher power at high currents brings nonlinear losses, filamentation and catastrophic optical damage (COD) and affects the reliability of the laser, further degrading beam quality.

[0008] The use of tapered design, lateral heterogeneities, and modulations in the refractive index for filtering out higher-order side modes are among the current strategies to address this problem. The tapered design enables adiabatic coupling between single-mode laser and multimode waveguide sections without exciting higher-order modes. Despite their ability to limit higher-order mode initiation, there are significant problems with these methods. The expansion of the waveguide (expansion angle a) may generally trigger undesirable effects on the beam shape. A further limitation for most of these designs is that they also generate losses on the fundamental mode and require significant steps to be added to the fabrication procedure. Having said all this, it is of crucial importance to develop a repeatable method in order to produce industrial-quality single-mode EELs.

[0009] Recent discoveries in the field of photonics promise a flexible alternative for improving the properties of laser systems. Designing optical systems in a specific way allows the use of parity-time symmetry (PTS), which in turn provides the possibility to selectively apply losses to higher-order modes. The general schematic of the PTS consists of a pumped main waveguide and a lossy partner waveguide. The PTS condition is achieved by limiting the gain / loss contrast by maintaining the coupling between the main and matching (lossy) waveguides. This concept can be used in both VCSELs and EELs. Most optically pumped PTS lasers operate at laboratory scale. However, electrically pumped ones have very low power levels for enabling single-mode propagation. This is related to the so- called exceptional point, the coupling constant between the main and matching waveguides, and the loss in the matching (lossy) waveguides. In typical PTS designs, fine-tuning the applied gain / loss contrast is essential for achieving the PTS condition. Unoptimized gain / loss can lead to undesirable losses in fundamental mode and power. In order to overcome this, a PTS modification, the quasi-PTS design, is proposed. According to this concept, the passive waveguide is specifically designed to support only the higher-order mode in the main waveguide. However, an exceptional point still exists, above this point, increasing the gain causes the PTS to degrade and the design to fail. For this reason, in order to increase this exceptional point, there is a need for a new method of obtaining regulated passive waveguides.

[0010] The Chinese patent document no. CN109244828A, an application included in the state of the art, discloses a Bragg reflection waveguide laser (PT-BRW) using a parity-time (PT symmetry), i.e. a high-power laser based on the PT Bragg reflection waveguide, and the preparation method thereof. The said laser has the advantages of high power conversion efficiency (PCE), high catastrophic optical damage (COD) threshold, low laser threshold and easy heat dissipation. The said invention provides a high-power semiconductor laser based on a PT Bragg reflective waveguide, comprising an InP substrate and an InP cladding layer. The InP substrate and InP cladding layer are provided with a central cavity with a low refractive index and a gap with a low refractive index. The PT Bragg reflection grating regions on both sides of the central cavity, the central cavity with low refractive index and the PT Bragg reflection grating region extend along the entire z-axis, and the PT Bragg reflection grating regions on both sides are injection regions. The low refractive index central cavity is composed of a low refractive index material (such as SiC or the like) wrapped with a high refractive index strip waveguide, and the PT Bragg reflection grating region comprises quantum wells (QWs). Preferably, the InP substrate is a lower end surface and an upper end surface of the InP cladding layer is provided with a contact electrode. Preferably, the low refractive index central cavity is composed of a SiOi material enveloping a narrow InAlGaAs strip waveguide. The effective refractive index of the low refractive index central cavity is lower than the effective refractive index of the PT Bragg reflection grating regions on both sides. The refractive index of the central cavity is lower than that of the Bragg reflection regions on both sides, so that the scale of the central cavity can be larger, and at the same time, the energy of the laser mode can be more distributed in the PT Bragg grating regions on both sides. Preferably, the PT Bragg reflection grating regions on both sides are symmetrically distributed with respect to the low refractive index central cavity, the laser mode is stable, and the preparation process is simple. Preferably, the PT Bragg reflection grating region includes an upper layer, an intermediate layer and a lower layer from top to bottom. The upper layer is an InGaAsP / InP grating layer, the middle layer is an InAlGaAs quantum well active layer, and the lower layer is a doped InAlGaAs / InP grating. Light absorption is provided by the underlying doped InAlGaAs / InP, while the filter mode is obtained by the joint action of the upper layer, intermediate layer and lower layer. The conventional Bragg grating itself has a mode selection function. The PT-based Bragg grating of the said invention has a PT structure that comprises gain, loss and fluctuations in the refractive index. In the said invention, the upper layer provides fluctuations in the refractive index, the intermediate layer provides gain, and the lower layer provides loss.

[0011] Summary of the Invention

[0012] An object of the present invention is to realize a method for obtaining high-power single-mode edge-emitting semiconductor laser designs. Detailed Description of the Invention

[0013] “Method of Obtaining High-Power Single-Mode Laser” realized to fulfd the objectives of the present invention is shown in the figures attached, in which:

[0014] Figure 1 is a flowchart of the inventive method.

[0015] Figure 2 is a 3-dimensional view of a) PTS, b) q-PTS and c) h-PTS laser structures.

[0016] Figure 3 is a cross-sectional view of a) PTS, b) q-PTS and c) h-PTS laser structures.

[0017] Figure 4 is a top view of the bent a) PTS, b) q-PTS and c) h-PTS laser structures.

[0018] Figure 5 is a top view of the recessed a) PTS, b) q-PTS and c) h-PTS laser structures.

[0019] Figure 6 is a) 3-dimensional b) cross-sectional view of the asymmetric a-PTS laser structure.

[0020] The components illustrated in the figures are individually numbered, where the numbers refer to the following:

[0021] 100. Method

[0022] The inventive method (100) for obtaining high-power single-mode edge-emitting semiconductor laser designs comprises the steps of obtaining a laser chip by forming an active region between the optical cavity and the cavity by growing laser epitaxy on a semiconductor substrate (101); preparing waveguides of different widths through etching on the laser chip (102); forming a dielectric layer on waveguides (103); opening an injection window on some waveguides (104); and coating the entire laser chip first with contact metals and then with gold electroplating (105).

[0023] At the step of obtaining a laser chip by forming an active region between the optical cavity and the cavity by performing laser epitaxy on a semiconductor substrate (101) of the inventive method (100), laser epitaxy is grown on a semiconductor substrate (A) in the form of GaAs, InP, GaN, Si by using a method in the form of MOCVD or MBE, and in the epitaxial design, it is prepared in such a way as to form vertical optics for optimal confinement and carrier transfer by being interposed between an active quantum well (QW) in the form of InGaAs, InAlGaAs, InGaAsP, GaAsP.

[0024] At the step of preparing waveguides of different widths through etching on the laser chip (102) of the inventive method (100), waveguides with etching height (R) of 0.3-5 pm, width of Wm: 3-20 pm or Wp: 1-5 pm; width of 5-10 pm for dual-mode operation; distance between them (S) of 1-7 pm; length (L) of 1-10 mm are formed on the chip (A) by wet or dry etching. In one embodiment of the invention, 3 to 5 waveguides are formed, although any number of waveguides can be formed on the substrate (A).

[0025] At the step of forming a dielectric layer on waveguides (103) of the inventive method (100), the chip on which waveguides are formed is electrically insulated with a dielectric layer (K) in the form of SiOi or SisN4 in the thickness range of 5- 1000 nm.

[0026] At the step of opening an injection window on some waveguides (104) of the inventive method (100), an injection window (E) is opened on the waveguides (I) that are intended to be pumped, i.e. active, while no action is taken on the waveguides (P) that are intended to be kept unpumped, i.e. passive, of those electrically pumped waveguides. The waveguides on the electrically isolated substrate (A) on which the waveguide is formed are prepared to be available in variable order and width as active (I) and passive (P). In one embodiment of the invention, the passive waveguide (P), active waveguide (I) and passive waveguide (P), all of which have a length of L and a width of Wm, form the PTS laser structure (Figures 2a, 3 a) by being arranged side by side. In another embodiment of the invention, the q-PTS (quasi-PTS) laser structure (Figures 2b, 3b) is obtained by forming, side by side, two passive waveguides (P) with a width of Wp and a length of L, an active waveguide (I) with a width of Wm and a length of L, and two passive waveguides (P) with a width of Wp and a length of L, respectively. In a further embodiment of the invention, the h-PTS (hybrid PTS) laser structure (Figures 2c, 3c) is obtained by forming, side by side, a passive waveguide (P) with a width of Wm and a length of L, a passive waveguide (P) with a width of Wp and a length of L, an active waveguide (I) with a width of Wm and a length of L, a passive waveguide (P) with a width of Wp and a length of L, and a passive waveguide (P) with a width of Wm and a length of L, respectively. In a further embodiment of the invention, the PTS, q-PTS and h-PTS laser structures (Figure 4a, b, c) are obtained by shaping the output surfaces of the passive waveguides so as to have a length of AL (up to L / 4) and to be bent with a displacement of AS (greater than 1 micron), while the length of all waveguides is L. In a further embodiment of the invention, the recessed PTS, q-PTS and h-PTS laser structures (Figure 5a, b, c), prepared in such a way that the length of the active waveguide (I) is L while the length of the passive waveguides (P) is AL shorter than L at one end, are obtained. In a further embodiment of the invention, the asymmetric a-PTS laser structure (Figure 6) is obtained by forming, side by side, three passive waveguides (P) with a width of Wp, an active waveguide (I) with a width of Wm, and two passive waveguides (P) with a width of Wm, all of which have a length of L, respectively.

[0027] At the step of coating the entire laser chip first with contact metals and then with gold electroplating (105) of the inventive method (100), the entire chip (A), on which active and passive waveguides are formed, is coated with contact metals (M) prepared with a mixture of Ti, Pt, Ni and Au in order to ensure proper adhesion and to adjust the metallic work functions for better carrier injection. The metal coating is then thickened with gold electroplating in order to facilitate die / wire bonding.

[0028] The lasers obtained by the inventive method (100) are shown in Figure 2. Here, the improved PTS (Figure 2a) utilizes additional passive waveguides of the same width as the main waveguide. Figure 2b shows the improved q-PTS design, which is specially designed not to affect the fundamental mode. In this design, there is a main waveguide with a width of Wm, which is below the cutoff frequency of mode 3 for the laser, and passive waveguides with widths of Wpl, Wp2, Wp3 and Wp4. Another variation of the proposed design is a hybrid PTS (Figure 2c). In this design, the active waveguide is placed side by side with two narrow passive waveguides with widths of Wp. This design has wide waveguides with widths of Wm and a second pair of passive waveguides. Details about this design are presented in Figure 2c and Figure 3c.

[0029] Cross-sectional images for the designs are illustrated in Figure 3. The width of the passive waveguides has been calculated through simulations. In these passive waveguides, the fundamental mode has the same effective index as the higher- order mode in the main waveguide. The etching height (R) is obtained by etching just above the vertical cavity in the EELs. The gap between two adjacent coasts (S) is set as approximately 3 microns. When a typical InGaAs QW is placed between AlGaAs reflective vertical layers, the optimum value of R is about 1200nm. To allow only modes 1 and 2 and to provide higher power, the width of the main waveguide should remain below 9 microns and be set as approximately 8 microns. The width of the passive waveguides is set as approximately 2.35 microns in order to optimally match the mode 2 of the main waveguide. These widths may have equal or different widths in order to compensate for production errors. The improved PTS design (Figure 2a and Figure 3a) has two passive waveguides that can be extended up to four. The improved PTS design (Figure 2b and Figure 3b) may have more than four passive waveguides. For example, it can be designed with three or four pairs of passive waveguides. However, as supported by our theoretical studies, a design with more than four waveguides would have less effect. Finally, the improved hybrid PTS design (Figure 2c and Figure 3c) has a fixed design where two narrow passive waveguides are adjacent to two wider waveguides.

[0030] Furthermore, two additional variations are provided for the improved designs presented in Figure 2 and Figure 3. One of these variations is to further degrade the feedback and increase the loss by bending the passively unpumped waveguides. Top views of these designs are shown in Figure 4. All lasers are of length L, while the output surfaces of the passive waveguides are refracted with a length of AL and a displacement of AS. The other parameters are the same as in Figure 3. Other variations are obtained by making recessed passive partners as shown in Figure 5. In these designs, only the length of the main injected waveguide is L. The other passive waveguide coasts are shorter with AL.

[0031] Within these basic concepts; it is possible to develop various embodiments of the inventive “Method (100) of Obtaining High-Power Single-Mode Laser”; the invention cannot be limited to examples disclosed herein and it is essentially according to claims.

Claims

CLAIMS1. A method (100) for obtaining high-power single-mode edge-emitting semiconductor laser designs; characterized by the steps of; obtaining a laser chip by forming an active region between the optical cavity and the cavity by growing laser epitaxy on a semiconductor substrate (101); preparing waveguides of different widths through etching on the laser chip (102); forming a dielectric layer on waveguides (103); opening an injection window on some waveguides (104); and coating the entire laser chip first with contact metals and then with gold electroplating (105).

2. A method (100) according to Claim 1; characterized in that at the step of obtaining a laser chip by forming an active region between the optical cavity and the cavity by growing laser epitaxy on a semiconductor substrate (101), laser epitaxy is grown on a semiconductor substrate (A) in the form of GaAs, InP, GaN, Si by using a method in the form of MOCVD or MBE, and in the epitaxial design, it is prepared in such a way as to form vertical optics for optimal confinement and carrier transfer by being interposed between an active quantum well (QW) in the form of InGaAs, InAlGaAs, InGaAsP, GaAsP.

3. A method (100) according to Claim 1 or 2; characterized in that at the step of preparing waveguides of different widths through etching on the laser chip (102), waveguides with etching height (R) of 0.3-5 pm, width of Wm: 3-20 pm or Wp: 1-5 pm; width of 5-10 pm for dual-mode operation; distance between them (S) of 1-7 pm; length (L) of 1-10 mm are formed on the chip (A) by wet or dry etching.

4. A method (100) according to Claim 3; characterized in that 3 to 5 waveguides are formed, although any number of waveguides can be formed on the substrate (A).

5. A method (100) according to any one of the preceding claims; characterized in that at the step of forming a dielectric layer on waveguides (103), the chip on which waveguides are formed is electrically insulated with a dielectric layer (K) in the form of SiOi or SisN4 in the thickness range of 5-1000 nm.

6. A method (100) according to any one of the preceding claims; characterized in that at the step of opening an injection window on some waveguides (104), an injection window (E) is opened on the waveguides (I) that are intended to be pumped, i.e. active, while no action is taken on the waveguides (P) that are intended to be kept unpumped, i.e. passive, of those electrically pumped waveguides.

7. A method (100) according to Claim 6; characterized in that the waveguides on the electrically isolated substrate (A) on which the waveguide is formed are prepared to be available in variable order and width as active (I) and passive (P).

8. A method (100) according to any one of the preceding claims; characterized in that the passive waveguide (P), active waveguide (I) and passive waveguide (P), all of which have a length of L and a width of Wm, form the PTS laser structure by being arranged side by side.

9. A method (100) according to Claim 6 or 7; characterized in that the q- PTS (quasi-PTS) laser structure is obtained by forming, side by side, two passive waveguides (P) with a width of Wp and a length of L, an activewaveguide (I) with a width of Wm and a length of L, and two passive waveguides (P) with a width of Wp and a length of L, respectively.

10. A method (100) according to Claim 6 or 7; characterized in that the h- PTS (hybrid PTS) laser structure is obtained by forming, side by side, a passive waveguide (P) with a width of Wm and a length of L, a passive waveguide (P) with a width of Wp and a length of L, an active waveguide (I) with a width of Wm and a length of L, a passive waveguide (P) with a width of Wp and a length of L, and a passive waveguide (P) with a width of Wm and a length of L, respectively.

11. A method (100) according to Claim 6 or 7; characterized in that the PTS, q-PTS and h-PTS laser structures are obtained by shaping the output surfaces of the passive waveguides so as to have a length of AL (up to L / 4) and to be bent with a displacement of AS (greater than 1 micron), while the length of all waveguides is L.

12. A method (100) according to Claim 6 or 7; characterized in that the recessed PTS, q-PTS and h-PTS laser structures, prepared in such a way that the length of the active waveguide (I) is L while the length of the passive waveguides (P) is AL shorter than L at one end, are obtained.

13. A method (100) according to Claim 6 or 7; characterized in that the asymmetric a-PTS laser structure is obtained by forming, side by side, three passive waveguides (P) with a width of Wp, an active waveguide (I) with a width of Wm, and two passive waveguides (P) with a width of Wm, all of which have a length of L, respectively.

14. A method (100) according to any one of the preceding claims; characterized in that at the step of coating the entire laser chip first with contact metals and then with gold electroplating (105), the entire chip (A),on which active and passive waveguides are formed, is coated with contact metals (M) prepared with a mixture of Ti, Pt, Ni and Au in order to ensure proper adhesion and to adjust the metallic work functions for better carrier injection.

15. A method (100) according to Claim 14; characterized in that the metal coating is then thickened with gold electroplating in order to facilitate die / wire bonding.

Images