A combined multi-quantum well active region, light emitting device and method of fabricating the same

Abstract

The application discloses a combined multi-quantum well active region, a light-emitting device and a preparation method thereof, and relates to the field of semiconductor light-emitting devices. The combined multi-quantum well active region is composed of M periods of repeated units of a barrier layer, a quantum well layer and a barrier layer, and two adjacent periods share the same barrier layer, wherein the quantum well layer comprises N well sub-layers and N-1 spacer layers, and the spacer layers are arranged between adjacent well sub-layers, wherein M and N are positive integers not less than 2; the energy band gap of the barrier layer is greater than the energy band gap of the well sub-layer, and the difference between the two is not less than 0.15 eV; and the energy band gap of the spacer layer is greater than the energy band gap of the well sub-layer, and the difference between the two is not less than 0.1 eV. The combined multi-quantum well active region can effectively broaden the gain spectrum.

Description

TECHNICAL FIELD

[0001] The present application relates to the field of semiconductor light-emitting devices, in particular to a combination of a multi-quantum well active region, a light-emitting device and a preparation method thereof. BACKGROUND

[0002] In a semiconductor light-emitting device (such as a semiconductor laser, a light-emitting diode, etc.), an active region is a core active region for realizing carrier injection and recombination, and the structure design of the active region directly determines the photoelectric performance of the device. Among them, the quantum well active region is based on the quantum confinement effect, and a multi-quantum well structure is formed by periodically growing a narrow-bandgap well layer and a wide-bandgap barrier layer, which produces a strong restriction on electrons and holes in the direction perpendicular to the well layer, significantly improving the carrier concentration and the radiation recombination probability in the active region.

[0003] The conventional design of strained quantum well active region generally forms a single split-off level in the conduction band side of the well, and forms multiple split-off levels in the valence band side, such as heavy hole and light hole. In the case of a certain level of biaxial tensile strain introduced in the well layer material, the light hole level will be above the heavy hole level, while in the rest of the cases (including smaller tensile strain, no strain, and compressive strain), the heavy hole level will be above the light hole level. In the former case, the transition mainly occurs between the split-off level in the conduction band and the light hole level in the valence band, and the gain contributes 75% to the TM mode of the polarization light field perpendicular to the well layer, and contributes 25% to the TE mode of the polarization light field parallel to the well layer. In the latter case, the transition mainly occurs between the split-off level in the conduction band and the heavy hole level in the valence band, and the gain contributes 100% to the TE mode of the polarization light field parallel to the well layer, and contributes 0% to the TM mode of the polarization light field perpendicular to the well layer. In such a case, the coverage range of the gain spectrum is determined by the transition level interval (corresponding to the longest wavelength) to the quasi-Fermi level difference of the carrier pair energy band filling (corresponding to the shortest wavelength). Due to the existence of exciton effect and collision effect between carriers, and the collision effect between carriers and lattice, they will further broaden the coverage range of the gain spectrum, but considering all the above factors, for typical InP-based or GaAs-based III-V compound semiconductor materials, the full spectrum FWHM of the gain spectrum is generally about 60 nm. Although more split-off levels can be introduced in the conduction band side of the well by increasing the energy level difference between the barrier and the conduction band edge of the quantum well material (i.e. increasing the barrier height), thereby widening the gain spectrum through multiple transitions between each conduction band level and valence band level. But this way is difficult to accurately control the interval between each transition level and the contribution of each transition level to the gain size due to too few material and structure parameters to control and adjust, so the gain spectrum obtained by this way generally cannot meet the required shape. In addition, the transition level interval in each quantum well can be adjusted by changing the thickness of each quantum well in the multi-quantum well structure of the active region, thereby expanding the gain spectrum, but the main problem of this way is that the transition levels in each well are different, the filling of the transition levels in each well is different when the carriers are injected, and the carrier wave functions at different levels in each well are different, resulting in different overlap integrals, so the electron-hole pair dipole radiation transition matrix elements between the levels in different wells are different. These factors cause the transition of each well to contribute unevenly to the size of the gain, and in severe cases, individual wells may be underfilled, causing the quasi-Fermi level difference to be even lower than the transition level interval, resulting in absorption. In addition, this quantum well structure lacks overall periodicity, which brings certain difficulties to the precision control and calibration of the growth process (such as when X-ray diffraction technology is needed to characterize the grown material and structure).

[0004] Therefore, there is a lack of a quantum well active region capable of effectively broadening the gain spectrum in the prior art. SUMMARY

[0005] The present application aims to provide a combined multi-quantum well active region capable of effectively broadening the gain spectrum.

[0006] To achieve the above-mentioned purpose, the present application provides a combined multi-quantum well active region, which is composed of M periods of barrier layer, quantum well layer and barrier layer as a repeating unit, and adjacent two periods share the same barrier layer, wherein the quantum well layer comprises N well sub-layers and N-1 spacer layers, and the spacer layers are arranged between adjacent well sub-layers, wherein M and N are positive integers not less than 2; the energy band gap of the barrier layer is greater than that of the well sub-layer, and the difference between them is not less than 0.15 eV; the energy band gap of the spacer layer is greater than that of the well sub-layer, and the difference between them is not less than 0.1 eV.

[0007] Optionally, the thickness of the well sub-layer is 2-10 nm, and the thickness of the barrier layer is not less than 6 nm.

[0008] Optionally, the thickness of the spacer layer is 1-5 nm.

[0009] Optionally, the material of the combined multi-quantum well active region is AlGaInAs or InGaAsP.

[0010] The present application also provides a light emitting device comprising the combined multi-quantum well active region as described above.

[0011] Optionally, the material of the spacer layer is matched with the lattice of the substrate of the light emitting device.

[0012] Optionally, the light emitting device is one of a semiconductor laser, a light emitting diode, a super radiation diode and a semiconductor optical amplifier.

[0013] Optionally, the light emitting device is a semiconductor laser, which comprises, from bottom to top, a bottom N-side electrode, an N-type substrate, an N-type buffer layer, an N-side confinement layer, the combined multi-quantum well active region, a P-side confinement layer, a P-type isolation layer, a grating layer, a wetting layer, a P-type cladding layer, a P-side ohmic contact layer and a top P-side electrode.

[0014] The present application also provides a preparation method of the combined multi-quantum well active region as described above, which comprises: epitaxially growing M periods as a repeating unit of barrier layer, quantum well layer and barrier layer, and adjacent two periods share the same barrier layer, wherein the quantum well layer comprises N well sub-layers and N-1 spacer layers, and the spacer layers are arranged between adjacent well sub-layers, wherein M and N are positive integers not less than 2.

[0015] The combination multi-quantum well active region according to the present application can be used to form an effectively broadened gain spectrum, for example, the full width at half maximum of the gain spectrum can be broadened to more than 100 nm, thus being suitable for use in a tunable or array light emitting device, a dense wavelength division multiplexing light emitting device, and a superluminescent diode and a semiconductor optical amplifier with a wideband requirement, etc. In addition, the combination multi-quantum well active region according to the present application can also be used to form a continuously and smoothly broadened gain spectrum, for example, in the range of the interval between two adjacent gain peaks, the gain fluctuation, i.e. the variation of each gain relative to the average value of the gain in the range, is within ±1 dB. BRIEF DESCRIPTION OF DRAWINGS

[0016] Figure 1 is a sectional view of a semiconductor laser based on the combination multi-quantum well active region according to the present application;

[0017] Figure 2 is a sectional view of a combination multi-quantum well active region;

[0018] Figure 3 is a strength normalized gain graph of Example 1 and Comparative Example 1;

[0019] Figure 4 is a strength normalized gain graph of Example 2 and Comparative Example 2. DETAILED DESCRIPTION

[0020] The technical solutions of the present application will be further described below in conjunction with the accompanying drawings and through specific embodiments.

[0021] The present application utilizes two separate well sub-layers which will cause splitting of their original energy levels due to the coupling of their wave functions when they are sufficiently close, thus creating a pair of separated energy levels in both well sub-layers. At this time, the energy levels in these two well sub-layers are completely identical, so the contribution of the same energy level in each well sub-layer to the gain will not be different. The lower energy level in the conduction band side of this pair of well sub-layers has a symmetric wave function, while the higher energy level has an anti-symmetric wave function; the higher energy level in the valence band side has a symmetric wave function, while the lower energy level has an anti-symmetric wave function. Since the dipole radiation recombination transition matrix element of a conduction band electron to a valence band hole is proportional to the overlap integral of their wave functions, in the above structure, the radiation transition can only occur between the lower energy level electron in the conduction band side and the higher energy level hole in the valence band side which both have a symmetric wave function, or between the higher energy level electron in the conduction band side and the lower energy level hole in the valence band side which both have an anti-symmetric wave function. The former has a relatively longer wavelength, while the latter has a relatively shorter wavelength, the wavelength interval between them, and their respective contribution to the gain size can be precisely controlled by the choice of the spacer material composition and thickness between the two well sub-layers. This design principle can be further extended to further expand or modify the gain spectrum by utilizing more separated energy levels created by the coupling of more well sub-layers, to meet the different practical application requirements. In general, there can be N separated energy levels in N coupled well sub-layers separated by N-1 spacers, and the N energy levels in each well sub-layer are the same, so the contribution of each well sub-layer to the gain size in terms of the same energy level is consistent, i.e. in terms of a single electron-hole pair, there is no non-uniformity between the well sub-layers. Restricted by the orthogonality of the carrier wave functions on each energy level and the nature of the dipole radiation recombination transition between energy levels, all radiation transitions can only occur between the N separated conduction band to valence band energy level pairs, and there will be N independent gain peaks on the corresponding gain spectrum, and their interval wavelengths correspond to the wavelengths of the respective transition energy level pairs. By adjusting the material composition and thickness of the N-1 spacers between the well sub-layers, the interval and size of these independent gain peaks can be adjusted, so that the coverage range and shape of the entire gain spectrum can reach certain predetermined targets.

[0022] Based on the above mechanism, the present application provides a combined multi-quantum well active region and a light emitting device based on the combined multi-quantum well active region. In one embodiment, the light emitting device is a semiconductor laser, but it should be understood that the combined multi-quantum well active region proposed by the present application is not limited to be used in semiconductor lasers, but can also be used in light emitting diodes, superluminescent diodes, semiconductor optical amplifiers, etc. Hereinafter, the content of the present application will be introduced by taking a semiconductor laser based on the combined multi-quantum well active region proposed by the present application as an example.

[0023] Figure 1 A two-dimensional structure defined along the semiconductor laser growth direction X and the light propagation direction Z. As shown in FIG. 1, the combined multi-quantum well active region includes a plurality of well sub-layers 10 separated by a plurality of spacer layers 20.Figure 1 As shown, the semiconductor laser comprises, from bottom to top, a bottom N-side electrode 101, an N-type substrate 102, an N-type buffer layer 103, an N-side confinement layer 104, a combined multi-quantum well active region 105, a P-side confinement layer 106, a P-type isolation layer 107, a grating layer 108, a wetting layer 109, a P-type cladding layer 110, a P-side ohmic contact layer 111, and a top P-side electrode 112.

[0024] The bottom N-side electrode 101 and the top P-side electrode 112 are used to connect an external power supply to provide a bias voltage for the semiconductor laser. The N-type substrate 102 is used to provide support for the semiconductor laser. The N-type buffer layer 103 and the N-side confinement layer 104 provide both a vertical confinement for the optical field and a channel for injecting electrons from the conduction band into the active region. The P-side confinement layer 106 and the P-side cladding layer 110 provide both a vertical confinement for the optical field and a channel for injecting holes from the valence band into the active region. The combined multi-quantum well active region 105 is used to provide a wide-spectrum gain for the entire active region. The grating layer 108 is used for longitudinal mode selection of the light. The P-side ohmic contact layer 111 is used to form an ohmic contact to eliminate the Schottky barrier between the electrode metal and the P-type semiconductor. It is worth mentioning that, Figure 1 The semiconductor laser structure shown in the above is only an example, and the combined multi-quantum well active region structure provided by the present application can be applied to any other semiconductor laser, which is not limited by the present application.

[0025] The combined multi-quantum well active region 105 is composed of M periods of barrier layers, quantum well layers, and barrier layers as repeating units, and adjacent two periods share the same barrier layer. The quantum well layer includes N well sub-layers and N-1 spacer layers arranged between adjacent well sub-layers. M and N are positive integers not less than 2. The energy band gap (Eg_B) of the barrier layer is greater than the energy band gap (Eg_W) of the well sub-layer, and satisfies Eg_B - Eg_W ≥ 0.15eV. As an example, the thickness of the well sub-layer is 2nm-10nm, and the thickness of the barrier layer is not less than 6nm. The spacer layer is made of a material matched with the substrate lattice, for example, when the substrate is InP, the spacer layer can be made of InGaAsP or AlGaInAs, and the energy band gap (Eg_S) of the spacer layer is wider than the energy band gap of the well sub-layer, and satisfies Eg_S - Eg_W ≥ 0.1eV. As an example, the thickness of the spacer layer is 1nm-5nm. The thickness and energy band gap of the spacer layer are the main parameters for controlling the energy level splitting, i.e., the separation of the wavelength distance, and a thinner spacer layer and a proper energy band gap are beneficial to increasing the coupling between the well sub-layers on both sides of the spacer layer.

[0026] This active region structure will generate N pairs of coupled separate transition energy levels in the well sub-layers on the conduction band side and the valence band side, respectively, so that N pairs of separate transitions (orthogonality of the wave functions between each energy level excludes more possible transitions) can be generated, which will generate N corresponding separate peaks on the gain spectrum, each at the wavelength corresponding to the transition energy level. By adjusting the parameters (composition and thickness) of the barrier, well sub-layer, and spacer layer, the peaks can be connected smoothly to greatly expand the continuous gain spectrum width, which is conducive to the gain requirements of a wide-spectrum multi-wavelength array laser or other light-emitting devices.

[0027] As shown in Figure 2 In one embodiment, the combined multi-quantum well active region 105 is composed of 2 periods of repeating units of barrier layer 201, quantum well layer, and barrier layer 201, with the adjacent two periods sharing the same barrier layer 201, and the quantum well layer is composed of a stack of well sub-layer 202, spacer layer 203, and well sub-layer 202. In one period, by controlling the composition and thickness of the spacer layer between a pair of well sub-layers, a pair of coupled separate transition energy levels is generated in the two well sub-layers, the electrons on the lower energy level on the conduction band side and the holes on the higher energy level on the valence band side have symmetric wave functions, while the electrons on the higher energy level on the conduction band side and the holes on the lower energy level on the valence band side have anti-symmetric wave functions, so that the transition can and only can occur between the two pairs of energy levels with symmetric wave functions and anti-symmetric wave functions, so that such a gain spectrum will generate two peaks, corresponding to the wavelengths between the two transition energy levels, respectively. By adjusting the composition and thickness of the barrier, well sub-layer, and spacer layer, the wavelength interval and the respective intensity of the two gain peaks can be controlled as required, so as to expand the gain peaks.

[0028] The semiconductor laser in the present application can be a ridge waveguide structure or a buried heterojunction structure, and the present application does not limit this. The semiconductor laser in the present application does not have a particular limitation on the growth material system.

[0029] 1. Embodiment 1

[0030] The semiconductor laser in this embodiment has a ridge waveguide structure and is based on an InP-AlGaInAs-InGaAsP material system, and its two-dimensional structure defined along its growth direction X and light propagation direction Z is as shown in Figure 1 The combined multi-quantum well active region includes 2 periods of repeating units, each unit including 2 well sub-layers and 1 spacer layer, i.e., the structure of the combined multi-quantum well active region is barrier layer-well sub-layer-spacer layer-well sub-layer-barrier layer-well sub-layer-spacer layer-well sub-layer-barrier layer. The specific materials and structural parameters of the combined multi-quantum well active region are shown in Table 1.

[0031] Table 1: Specific materials and structural parameters of the combined multi-quantum well active region

[0032]

[0033] As Figure 3 shown in the comparative example 1 which does not adopt the combined multi-quantum well active region of the present application but adopts the multi-quantum well active region formed by the alternately grown narrow band gap well layer and wide band gap barrier layer commonly used in the prior art, the gain spectrum full width at half maximum FWHM1 is 75 nm; while in the present embodiment, the gain spectrum full width at half maximum FWHM2 is 130 nm, i.e. the gain spectrum is widened by 73.3%. Meanwhile, as can be seen from Figure 3 , the gain spectrum in the present embodiment is continuously and smoothly widened, and the gain fluctuation within the interval of the two gain peaks formed by the widening is within ±0.4 dB.

[0034] 2, Example 2

[0035] The difference between this embodiment and Example 1 is that the semiconductor laser in this embodiment is based on GaAs-InGaAsP-AlGaInP material system. The specific materials and structure parameters of the combined multi-quantum well active region in this embodiment are shown in Table 2.

[0036] Table 2 for specific materials and structure parameters of the combined multi-quantum well active region

[0037]

[0038] As Figure 4 shown in the comparative example 2 which does not adopt the combined multi-quantum well active region of the present application but adopts the multi-quantum well active region formed by the alternately grown narrow band gap well layer and wide band gap barrier layer commonly used in the prior art, the gain spectrum full width at half maximum FWHM3 is 70 nm; while in the present embodiment, the gain spectrum full width at half maximum FWHM4 is 115 nm, i.e. the gain spectrum is widened by 64.3%. Meanwhile, as can be seen from Figure 4 , the gain spectrum in the present embodiment is continuously and smoothly widened, and the gain fluctuation within the interval of the two gain peaks formed by the widening is within ±0.4 dB.

[0039] The present application also provides a preparation method of a semiconductor laser comprising a combined multi-quantum well active region, which comprises:

[0040] S1, epitaxially growing N-type buffer layer, N-side confinement layer, combined multi-quantum well active region, P-side confinement layer, P-type isolation layer, grating layer and wetting layer on the substrate in sequence. The material of the substrate is generally InP or GaAs. When GaAs substrate is used, low refractive index buffer layer needs to be grown to isolate GaAs with high refractive index.

[0041] The preparation method of the combined multi-quantum well active region comprises:

[0042] S11, epitaxially growing M cycles repeatedly with barrier layer, quantum well layer and barrier layer as a repeating unit, and the adjacent two cycles share the same barrier layer, wherein the quantum well layer comprises N well sub-layers and N-1 spacer layers, and the spacer layers are arranged between the adjacent well sub-layers, wherein M and N are positive integers not less than 2.

[0043] S2, fabricating a grating structure on the grating layer. When the grating structure is fabricated, the grating can be written by using the electron beam exposure, or the grating structure can be fabricated by using any other technology in the prior art.

[0044] S3, epitaxially growing a P-type cladding layer and a P-side ohmic contact layer.

[0045] S4, preparing a ridge waveguide, electric contact windowing, fabricating a P-side patterned electrode, thinning the back surface and fabricating an N-side electrode.

[0046] The existing crystal growth technology based on MOCVD or MBE can accurately control the growth of a large-area III-V semiconductor material layer as thin as 2nm, so the structure involved in the present application can be conveniently realized according to the current process manufacturing level.

[0047] In summary, the combined multi-quantum well active region proposed in the present application can be used to form an effectively broadened gain spectrum, for example, the full width at half maximum of the gain spectrum can be broadened to more than 100nm. In addition, the combined multi-quantum well active region can also be used to form a gain spectrum which is continuously and smoothly broadened, for example, in the range between the adjacent two gain peak intervals, the gain fluctuation, that is, the change amount of each gain relative to the average value of the gain in the range, is within ±1dB; in some embodiments, the gain fluctuation can be controlled within ±0.4dB. In the present application, the range between the adjacent two gain peak intervals refers to the range between the top points of the two gain peaks.

[0048] It is worth mentioning that the combined multi-quantum well active region proposed in the present application can be applied to various semiconductor lasers, as well as various semiconductor light-emitting diodes and superluminescent diodes, and various semiconductor optical amplifiers, etc., so that the coverage range of the light-emitting wavelength can be greatly broadened by means of arraying and tuning.

[0049] The present application is described above in combination with specific embodiments, and these descriptions are only for the purpose of explaining the present application, and cannot be explained as limiting the protection scope of the present application in any way. Based on the explanations herein, other specific embodiments of the present application can be conceived by those skilled in the art without creative labor, and these embodiments will fall within the protection scope of the present application.

Claims

1. A combined multi-quantum-well active region, characterized in that, The combined multi-quantum-well active region is composed of M cycles of repeating barrier layers, quantum well layers, and barrier layers as repeating units. Two adjacent cycles share the same barrier layer. The quantum well layer includes N sub-well layers and N-1 interlayers. The interlayers are disposed between adjacent sub-well layers, where M and N are positive integers not less than 2. The energy band gap of the barrier layer is greater than the energy band gap of the sub-well layer, and the difference between the two is not less than 0.15 eV. The energy band gap of the interlayer is greater than the energy band gap of the sub-well layer, and the difference between the two is not less than 0.1 eV.

2. The combined multi-quantum-well active region as described in claim 1, characterized in that, The thickness of the well layer is 2nm to 10nm, and the thickness of the barrier layer is not less than 6nm.

3. The combined multi-quantum-well active region as described in claim 1, characterized in that, The thickness of the partition layer is 1nm to 5nm.

4. The combined multi-quantum-well active region as described in claim 1, characterized in that, The material of the combined multi-quantum well active region is AlGaInAs or InGaAsP.

5. A method for fabricating a combined multi-quantum-well active region as described in any one of claims 1-4, characterized in that, The preparation method includes: repeatedly growing M cycles using a barrier layer, a quantum well layer, and a barrier layer as repeating units, with two adjacent cycles sharing the same barrier layer, wherein the quantum well layer includes N sub-well layers and N-1 interlayers, and the interlayers are disposed between adjacent sub-well layers, wherein M and N are positive integers not less than 2.

6. A light-emitting device, characterized in that, The light-emitting device includes the combined multiple quantum well active region as described in any one of claims 1-4.

7. The light-emitting device as described in claim 6, characterized in that, The material of the separator is matched with the lattice of the substrate of the light-emitting device.

8. The light-emitting device as described in claim 6, characterized in that, The light-emitting device is one of a semiconductor laser, a light-emitting diode, a superluminescent diode, and a semiconductor optical amplifier.

9. The light-emitting device as described in claim 6, characterized in that, The light-emitting device is a semiconductor laser, which, from bottom to top, includes a bottom N-side electrode, an N-type substrate, an N-type buffer layer, an N-side confinement layer, the combined multi-quantum-well active region, a P-side confinement layer, a P-type isolation layer, a grating layer, a wetting layer, a P-type cladding layer, a P-side ohmic contact layer, and a top P-side electrode.

Images