A method for measuring the absorption coefficient of quantum well material

Abstract

The application discloses a method for measuring the absorption coefficient of quantum well material, which comprises the following steps: (1) growing an epitaxial layer containing a quantum well structure on a substrate; (2) obtaining the total thickness of the quantum well layer; (3) preparing a sample from the epitaxial wafer and obtaining the reflectivity of the upper and lower interfaces of the epitaxial layer in the sample; (4) testing the external quantum efficiency curve when the photo-generated carriers in the quantum well structure are 100% collected, and analyzing the internal quantum efficiency curve of the light response of the quantum well structure according to the external quantum efficiency curve; and (5) solving the relationship between the absorption coefficient and the internal quantum efficiency, the reflectivity and the total thickness of the quantum well layer to obtain the absorption coefficient. The method provides a way for obtaining the absorption coefficient of the quantum well material, has the advantages of simplicity, practicality, no requirement for the substrate material, high accuracy and the like, avoids the shortcomings that the traditional ellipsometry method is not applicable to measuring the quantum well structure and the transmission method requires a transparent substrate, and is of great significance for device design.

Description

Technical Field

[0001] This invention relates to the field of semiconductor device material parameter measurement technology, and in particular to a method for measuring the absorption coefficient of quantum well materials in optoelectronic devices. Background Technology

[0002] InGaN ternary alloy materials not only possess a continuously tunable direct bandgap structure (0.65-3.42 eV), but also have advantages such as high absorption coefficient, high electron mobility, and strong radiation resistance, thus attracting widespread attention in the field of optoelectronic devices. The absorption coefficient of InGaN materials is closely related to the In composition and is an important optical parameter in device design. There are reports on the absorption coefficient of GaN materials both domestically and internationally. Researchers have used the ellipsometry to measure the refractive index of GaN materials, and some researchers have measured the reflectivity and absorption coefficient of GaN through reflection and transmission spectra (see references: Lian Chuanxin, Li Xiangyang, Liu Ji. Ellipsometry of GaN refractive index [J]. Journal of Infrared and Millimeter Waves, 2004, 23(4). and Du Xiaoqing. Measurement of optical parameters of GaN epitaxial layer using reflection and transmission spectra [J]. Optics and Optoelectronics Technology, 2010(1):4); however, there are very few reports on the absorption coefficient of InGaN materials, especially for InGaN materials with high In composition. This is mainly because InGaN materials face numerous challenges in epitaxial growth, such as In segregation, making it difficult to obtain sufficiently thick and relatively uniformly indistinct high-In-content InGaN bulk materials. Furthermore, the absorption coefficient of InGaN materials is also related to the device structure. Currently, most GaN-based devices employ quantum well structures to increase In content while maintaining material quality. The absorption coefficient of such InGaN quantum well materials differs significantly from that of bulk materials. Charge carriers in quantum wells possess only two degrees of freedom, and the relationship between the density of states (DSO) and energy in quantum wells is step-shaped, while the DSO of three-dimensional bulk materials exhibits a parabolic relationship. Since the absorption coefficient is closely related to the DSO, the absorption coefficient of three-dimensional bulk materials exhibits a parabolic shape with energy, while the absorption coefficient of quantum well materials shows a step-shaped relationship. Simultaneously, strong polarization exists in InGaN materials, causing band tilting, which makes the absorption coefficient of InGaN quantum well materials not step-shaped, but closer to an S-shaped curve. In order to better predict device performance and design devices, it is essential to obtain the absorption coefficient of the quantum well InGaN material in actual devices.

[0003] There are generally two methods for measuring the absorption coefficient of InGaN bulk materials: ellipsometry and spectrophotometry. In 2018, researchers used ellipsometry to measure the absorption coefficient of InGaN bulk materials with an In composition of 0.5-0.8 nm, which were epitaxially grown on sapphire substrates using MEPA-MOCVD technology (see: Zhang C, Li Y, Qian Y, et al. Surface and optical properties of indium-rich InGaN layers grown on sapphire by migration-enhanced plasma assisted metal organic chemical vapor deposition[J]. Materials Research Express, 2018, 6). However, ellipsometry is complex for data processing, and establishing a dispersion model becomes even more complicated and difficult when the InGaN material is a multilayer film system with unknown properties. Furthermore, for InGaN materials with a thickness of less than 100 nm, the fitted absorption coefficient has a large error, making it difficult to meet the characterization requirements of the absorption coefficient of quantum well InGaN materials. Spectrophotometry, on the other hand, calculates the absorption coefficient of the material based on the transmission and reflection spectra measured by a spectrophotometer. In 1997, researchers used this method to measure the absorption coefficients of epitaxial GaN and AlGaN bulk materials on sapphire substrates (see Brunner D, Angerer H, Bustarret E, et al. Optical constants of epitaxial AlGaN films and their temperature dependence[J]. Journal of Applied Physics, 1997, 82(10): 5090-5096. and Muth JF, Lee JH, Shmagin I K, et al. Absorption coefficient, energy gap, exciton binding energy, and recombination lifetime of GaN obtained from transmission measurements[J]. Applied Physics Letters, 1997, 71(18): 2572-2574.).In 2002, researchers grew a 200 nm AlN buffer layer on a sapphire substrate using molecular beam epitaxy (MBE). They then epitaxially grew a 240 nm high-In-content (x>0.5) InGaN bulk material on the buffer layer, and finally characterized the absorption coefficient using a spectrophotometer (see Wu J, Walukiewicz W, Yu KM, et al. Small band gap bowing in In1-xGaxN alloys[J]. Applied Physics Letters, 2002, 80(25)). However, spectrophotometry is a transmission-based measurement, which has limitations. It is only suitable for measuring transparent samples and cannot measure materials epitaxially grown on opaque substrates. Furthermore, interference from substrate reflection and absorption is difficult to eliminate during testing, leading to measurement errors. InGaN devices are typically epitaxially grown on substrates such as silicon, silicon carbide, and sapphire. Silicon and silicon carbide are opaque substrates, and their absorption coefficients cannot be measured using this method. Therefore, characterizing the absorption coefficient of InGaN quantum well materials through other methods is particularly important for better device design. Summary of the Invention

[0004] The purpose of this invention is to provide a simple and easy method for measuring the absorption coefficient of quantum well materials, which is unaffected by the substrate.

[0005] The objective of this invention is achieved as follows:

[0006] A method for measuring the absorption coefficient of quantum well materials, characterized by the following specific steps:

[0007] S1: Fabrication of epitaxial wafer: An epitaxial layer containing a quantum well structure is grown on a substrate. The epitaxial layer includes a first conductive layer, a quantum well structure, and a second conductive layer. The quantum well structure is composed of alternating quantum well layers and quantum barrier layers.

[0008] S2: Obtain the total thickness of the quantum well layer;

[0009] S3: Perform chip processing on the epitaxial wafer, deposit p-electrodes and n-electrodes to obtain a sample, and obtain the reflectivity of the upper interface and the lower interface of the epitaxial layer in the sample.

[0010] S4: Apply a suitable bias voltage and test the external quantum efficiency (EQE) curve when the photogenerated carriers generated in the quantum well structure are 100% collected; based on the external quantum efficiency curve, analyze and extract the internal quantum efficiency curve that is only the optical response of the quantum well structure.

[0011] S5: The absorption coefficient is obtained by solving the equations relating the absorption coefficient to the internal quantum efficiency, reflectivity, and the total thickness of the quantum well layer.

[0012] Furthermore, the total thickness of the quantum well layer in step S2 is obtained according to the experimental design or by characterization using X-ray diffraction, scanning electron microscopy, or secondary ion mass spectrometry. The total thickness of the quantum well layer is determined by the thickness of the quantum well layer and the number of periods of the quantum well structure, and the number of periods of the quantum well structure is greater than or equal to 1.

[0013] Furthermore, in step S3, the reflectance of the upper interface of the epitaxial layer in the sample is the reflectance of the light incident surface and the air interface, and the reflectance of the lower interface is the reflectance of the contact surface between the lower surface of the epitaxial layer and the electrode or substrate. The reflectance is obtained by specific measurement with a spectrophotometer, or by calculation according to the law of reflection, or by empirical value.

[0014] The appropriate bias voltage in step S4 includes forward bias, zero bias, and reverse bias.

[0015] Further, the appropriate bias voltage in step S4 is determined by the following process: gradually increasing the reverse bias voltage and measuring the external quantum efficiency curve of the sample. When the external quantum efficiency curve does not change with the increase of the applied reverse bias voltage in the wavelength range that is only the optical response of the quantum well structure, any bias voltage greater than or equal to the reverse bias voltage range is the bias voltage when the photogenerated carriers generated in the quantum well are 100% collected.

[0016] Furthermore, in step S4, within the wavelength range of the quantum well structure's optical response only, the internal quantum efficiency is equal to the measured external quantum efficiency.

[0017] Furthermore, when the quantum well structure contains V-pits, the measured external quantum efficiency in step S4 needs to be corrected: the measured external quantum efficiency is divided by the volume ratio of the quantum well platform region; the internal quantum efficiency is equal to the corrected external quantum efficiency; the volume ratio of the quantum well platform region is the ratio of the volume of the quantum well platform region to the volume of the quantum well assuming the sample has no V-pits. This is because when the quantum well structure contains V-pits, the sidewalls of the V-pits are basically non-absorbent, and the effective absorption region is only the quantum well region of the platform. Therefore, in order to make the measured absorption coefficient more accurate and eliminate the interference of the non-absorbent region, the external quantum efficiency needs to be corrected, that is, the measured external quantum efficiency is divided by the volume ratio of the quantum well platform region. At this time, the internal quantum efficiency is equal to the corrected external quantum efficiency.

[0018] Furthermore, if the chip process in step S3 does not include a surface roughening process, the absorption coefficient obtained in step S5 is the absorption coefficient of the quantum well material.

[0019] Furthermore, when the chip process in step S3 includes a surface roughening process, due to the non-perpendicular incident light beam, the absorption coefficient obtained in step S5 is the device-level equivalent absorption coefficient of the quantum well material. After the chip surface is roughened, the light beam actually enters the sample at an oblique angle, which is equivalent to an increase in the thickness of the quantum well layer. Therefore, the measured absorption coefficient is larger than the absorption coefficient of the quantum well material itself; to distinguish it, this invention refers to it as the device-level equivalent absorption coefficient. To improve device performance, actual optoelectronic devices (such as photodetectors, solar cells, etc.) undergo surface roughening treatment to reduce light loss caused by surface reflection. Therefore, the device-level equivalent absorption coefficient measured by this method is more practical in device simulation design.

[0020] Furthermore, the relationship in step S5 is derived based on the Lambert-Beer law, and is as follows:

[0021] η=(1-R1){[1-exp(-αd)]+[1-exp(-αd)]×R2exp(-αd)} (1)

[0022] Where η is the internal quantum efficiency, R1 is the reflectivity of the upper interface of the epitaxial layer in the sample, R2 is the reflectivity of the lower interface of the epitaxial layer in the sample, d is the total thickness of the quantum well layer, and α is the absorption coefficient of the quantum well material.

[0023] Compared with existing technologies, the present invention has the following advantages:

[0024] 1. This invention proposes a method for measuring the absorption coefficient of quantum well materials. The measurement process is simple and easy to perform, and the obtained absorption coefficient is relatively accurate, which is of great significance for device design.

[0025] 2. Unlike transmission measurement, this invention is applicable to measuring the light absorption coefficient of quantum well InGaN materials grown on opaque substrates (such as Si substrates). Attached Figure Description

[0026] Figure 1 This is a cross-sectional view of the epitaxial wafer prepared in Example 1;

[0027] Figure 2 This is a schematic diagram of the optical model used to measure the absorption coefficient in Example 1;

[0028] Figure 3 This is a schematic diagram of the reflectance curves of the upper and lower interfaces of the epitaxial layer in the sample of Example 1;

[0029] Figure 4 This is a schematic diagram of the external quantum efficiency curves of the sample in Example 1 under bias voltages of 0V, -3V, and -5V;

[0030] Figure 5 This is a half-section view of the epitaxial layer of the sample in Example 1;

[0031] Figure 6 This is a schematic diagram of the absorption coefficient of the quantum well InGaN material measured in Example 1;

[0032] Figure 7 This is a schematic diagram of the reflectance of the interface on the epitaxial layer in the sample of Example 2;

[0033] Figure 8 This is a schematic diagram of the external quantum efficiency curves of the sample in Example 2 under bias voltages of 0V, -2V, -5V, and -8V.

[0034] Figure 9 This is a schematic diagram of the absorption coefficient of the quantum well InGaN material measured in Example 2;

[0035] Figure 10 A summary graph of the absorption coefficients of InGaN materials obtained from calculations in Models 1 and 2 and measurements in Examples 1 and 2 from the literature;

[0036] Figure 11 This is a cross-sectional view of the epitaxial wafer prepared in Example 3;

[0037] Figure 12 This is a cross-sectional view of the epitaxial wafer prepared in Example 4;

[0038] Figure 13 This is a schematic diagram of the reflectance of the interface on the epitaxial layer in the sample of Example 4;

[0039] Figure 14 This is a schematic diagram of the sample in Example 4. Detailed Implementation

[0040] The present invention will now be described in further detail with reference to the embodiments and the accompanying drawings.

[0041] To make the objectives, technical solutions, and beneficial effects of this invention clearer, the technical solutions of this invention will be described in more detail below with reference to the embodiments and accompanying drawings. It should be noted that the accompanying drawings of this invention are all simplified and not precisely to scale, and are only used to facilitate the explanation of this invention.

[0042] The absorption coefficient measured in this invention is the intrinsic absorption coefficient, without considering exciton absorption, impurity absorption, etc. For example, the absorption of the yellow band of GaN is ignored, but this part contributes to EQE, so the final measured absorption coefficient includes this contribution.

[0043] In this invention, interlayer reflection and refraction at the epitaxial layer interface are ignored. It is assumed that the absorption of light follows the Lambert-Beer law. The light beam enters the sample perpendicularly, is absorbed once by the quantum well, reaches the bottom of the sample, is reflected perpendicularly according to the reflectivity of the lower interface of the epitaxial layer, and is reflected once. After being absorbed twice by the quantum well, it exits the surface of the sample perpendicularly and reaches the air.

[0044] The basic principle behind applying a bias voltage to achieve 100% collection efficiency for photogenerated carriers within the quantum well is as follows: It is well known that photogenerated carriers generated in the depletion region of a pn junction can be rapidly separated by the built-in electric field and then collected by the electrodes, achieving a collection efficiency of approximately 100%. For InGaN quantum well optoelectronic devices, applying a reverse bias voltage gradually increases the electric field in the depletion region, widening it and gradually covering the quantum well region. More photogenerated carriers generated within the quantum well are collected, increasing the external quantum efficiency in the corresponding wavelength band. When the reverse bias voltage is sufficiently large, the quantum well will be completely in the depletion region, and 100% of the photogenerated carriers generated within the quantum well will be collected. With further increases in the reverse bias voltage, the external quantum efficiency remains unchanged; that is, the EQE curves within this bias voltage range coincide in the wavelength range representing only the structure-optical response of the quantum well. Neglecting external circuit losses, it is assumed that electrons and holes reaching the pn junction contribute to the external circuitry, and the device's collection efficiency is 100% within the wavelength range representing only the structure-optical response of the quantum well. It is worth noting that when the quantum well structure contains a V-pit, the presence of the V-pit forms a three-dimensional pn junction, introducing a strong transverse electric field. This allows most of the photogenerated carriers generated within the quantum well to escape from the sidewalls of the V-pit under the influence of the transverse electric field, and then be collected by the electrodes. Therefore, there are cases where photogenerated carriers generated within the quantum well can be 100% collected even at 0V or even a positive bias voltage.

[0045] The fundamental principle of deriving internal quantum efficiency from external quantum efficiency:

[0046] The definition of internal quantum efficiency is:

[0047]

[0048] The definition of external quantum efficiency is:

[0049]

[0050] Within the wavelength range of the quantum well structure's optical response, since the collection efficiency of photogenerated carriers within the quantum well is 100%, meaning that an electron-hole pair generated within the quantum well is collected, the internal quantum efficiency equals the external quantum efficiency, according to the definitions of internal and external quantum efficiency.

[0051] Example 1:

[0052] To measure the absorption coefficient of InGaN quantum well material, an epitaxial wafer of the InGaN quantum well device was first fabricated. The cross-sectional view of the epitaxial wafer is shown in the figure below. Figure 1 From bottom to top, they are: substrate 100, buffer layer 200, n-type GaN layer 300, preparation layer 400, quantum well structure 500, p-type AlGaN electron blocking layer 600, p-type GaN layer 700, and V-pit 800.

[0053] The substrate 100 is made of Si.

[0054] The buffer layer 200 is made of AlN and has a thickness of 150 nm.

[0055] The n-type layer 300 is made of GaN, with a thickness of 2.3 μm and a Si doping concentration of 5 × 10⁻⁶. 18 / cm 3 .

[0056] The preparation layer 400 is made of GaN material, with a thickness of 0.22 μm and a Si doping concentration of 3 × 10⁻⁶. 17 / cm 3 .

[0057] The quantum well structure 500 consists of alternating InGaN well layers and GaN barrier layers. The InGaN well layer material is In. 0.3 Ga 0.7 The nitrogen layer is 2.5 nm thick and undoped. The GaN barrier material is GaN, 13 nm thick, with a Si doping concentration of 1 × 10⁻⁶. 17 / cm 3 One InGaN well layer and one GaN barrier layer constitute one cycle, and the 500 quantum well structure contains 9 cycles of InGaN / GaN.

[0058] p-type AlGaN electron blocking layer 600 material is Al 0.2 Ga 0.8 N, with a thickness of 10 nm and a Mg doping concentration of 1 × 10⁻⁶. 19 / cm 3 .

[0059] The p-type GaN layer 700 is made of GaN, with a thickness of 170 nm and a Mg doping concentration of 5 × 10⁻⁶. 19 / cm 3 .

[0060] V-pits 800 are generated in the preparation layer 400 and extend throughout the entire quantum well structure 500. The average V-pit radius r is 78 nm, and the average V-pit density ρ is 1 × 10⁻⁶. 9 / cm 2 The V-crater angle is 56.08°. There are a large number of V-craters of varying sizes and spacing in the epitaxial layer. The radius and density of the V-craters at this point are statistical average results.

[0061] In this embodiment, the quantum well In 0.3 Ga 0.7 The N material is considered to be of uniform In composition. According to the experimental design, the thickness of a single quantum well is 2.5 nm, the thickness of a single quantum barrier is 13 nm, and the total thickness of the quantum wells is 22.5 nm.

[0062] This embodiment measures the absorption coefficient of the quantum well InGaN material through the following process:

[0063] (1) The prepared epitaxial wafer was processed using chip fabrication. A NiAg electrode was deposited on the surface of the p-type GaN material. A co-electrode was placed during the deposition process. The reflectance of the NiAg deposited on the co-electrode was then measured using a spectrophotometer. The results are as follows: Figure 3 The average value in the range of 400nm-600nm is 90%, and the reflectivity of the lower interface of the epitaxial layer (the interface between p-GaN and NiAg) in the sample is 90%.

[0064] (2) Substrate transfer was performed using existing chip technology. A metal bonding layer was fabricated on the p-electrode and bonded to the substrate. The substrate 100 was removed, and AlN 200 was removed using dry etching to expose the n-GaN surface. The n-GaN surface was roughened using wet etching. At this point, the reflectance of the roughened n-GaN surface was measured using a spectrophotometer. See [reference needed]. Figure 3 The average value is <1%, which is ignored in the calculation, that is, the reflectivity of the upper interface of the epitaxial layer (the interface between n-GaN and air) in the sample is 0%; then n electrodes are fabricated on n-GaN, and the device is finally fabricated as a vertical structure chip with the n-face facing up, and has a nickel silver reflector on the p-face.

[0065] (3) The fabricated device was tested using a QE quantum efficiency tester to obtain external quantum efficiency curves in the wavelength range of 300nm-600nm under different reverse bias voltages (0V, -3V, -5V), such as... Figure 4 As shown, the external quantum efficiencies under different bias voltages coincide, indicating that the photogenerated carriers generated in the quantum well are completely collected. This means that the external quantum efficiency is equal to the internal quantum efficiency within the wavelength range of the quantum well structure's optical response.

[0066] (4) According to Figure 4The EQE curve only determines the wavelength range of the quantum well structure's optical response: Since the n-GaN layer of the prepared sample is thick enough, it completely absorbs photons up to 365 nm, and the electron-hole pairs generated within the n-layer cannot be collected, so the external quantum efficiency up to 365 nm is 0; photons in the wavelength range of 365-385 nm are mainly absorbed by GaN impurity energy levels and by the quantum well InGaN material; photons in the wavelength range of 385-550 nm are only absorbed by the quantum well InGaN material, so for photons in the wavelength range of 385-550 nm, the internal quantum efficiency is equal to the external quantum efficiency.

[0067] (5) Since the sidewalls have a low In composition and basically do not absorb light, the electron-hole pairs generated in the quantum well in the plateau region contribute to EQE; if a small unit containing a V-pit is approximated as a cylinder and the V-pit as a cone, then the half-section of the device is as follows: Figure 5 As shown, rotating the half-section 360° around the rotation axis yields a complete small unit containing V-shaped pits; Figure 5 The radius of the first quantum well is r1, θ is the V-pit angle with a value of 56.08°, r is the V-pit radius with a value of 78 nm, and R is... Figure 5 If the radius of the cylinder containing the V-pit is obtained by rotating 360°, then the volume of the quantum well platform region is:

[0068]

[0069] The volume percentage of the quantum well platform region is:

[0070]

[0071] in:

[0072]

[0073] As mentioned above, ρ is the density of the V-shaped pit, with a value of 1 × 10⁻� 9 cm -2 h represents the V-hole density, which is the height from the beginning of the first quantum well to the end of the growth of the last quantum well. For this sample, it is the thickness of 9 wells plus the thickness of 8 barriers, with a value of 126.5 nm. r1 can be obtained from trigonometric relationships:

[0074]

[0075] Substituting all the parameters, we can obtain Vratio, which is 84.7%. Therefore, the corrected external quantum efficiency for calculating the absorption coefficient using the relational formula should be:

[0076]

[0077] (6) Within the wavelength range of the quantum well structure's optical response, GaN material is essentially transparent. An optical model is established, such as... Figure 2 As shown; the incident light is perpendicularly incident from the upper surface n of the sample, and there is a reflection loss at the incident surface. After one absorption in the quantum well, it reaches the bottom p electrode. According to the Lambert-Beer law, the light intensity I1 reaching the bottom of the sample is:

[0078] I1=(1-R1)I0[exp(-αd)] (9)

[0079] I1 is the light intensity reaching the bottom of the sample, I0 is the incident light intensity, R1 is the reflectivity of the interface on the epitaxial layer in the sample, d is the total thickness of the quantum well layer, and α is the absorption coefficient of the quantum well material.

[0080] The intensity of light absorbed in a single pass through the quantum well is:

[0081] I 一次吸收 =(1-R1)I0-I1 (10)

[0082] After the incident light reaches the bottom, it is reflected vertically back to the sample. After secondary absorption by the quantum well, it is reflected off the sample surface. Similarly, according to the Lambert-Beer law, the light intensity I2 emitted from the sample surface is:

[0083] I2=R2I1[exp(-αd)] (11)

[0084] I2 is the light intensity emitted from the sample surface, and R2 is the reflectivity of the lower interface of the epitaxial layer in the sample.

[0085] The intensity of light absorbed in the second phase of the quantum well is:

[0086] I 二次吸收 =R2I1-I2 (12)

[0087] Therefore, the total light intensity absorbed by the quantum well is:

[0088] I 总吸收 =I 一次吸收 +I 二次吸收 =(1-R1)I0-I1+R2I1-I2 (13)

[0089] Substituting equations (9) and (11) into equation (13), we get:

[0090] I 总吸收 =(1-R1)I0{[1-exp(-αd)]+[1-exp(-αd)]×R2 exp(-αd)} (14)

[0091] Considering only intrinsic absorption, the device generates an electron-hole pair for every photon absorbed, and the internal quantum efficiency is:

[0092]

[0093]

[0094] Substituting the total thickness of the obtained quantum well layer, the reflectivity of the upper interface and the lower interface of the epitaxial layer in the sample, and the external quantum efficiency into equation (16), the absorption coefficient in the wavelength range of the quantum well structure's optical response is calculated, as follows: Figure 6 .

[0095] Example 2

[0096] Compared to Example 1, the preparation layer 400 in the epitaxial wafer is In 0.08 Ga 0.92 The N / GaN superlattice structure has a period of 2nm / 5nm, totaling 24 periods; the quantum well structure 500 contains 8 periods of In. 0.3 Ga 0.7 N (2.5nm) / GaN (13nm), see [reference] Figure 1 .

[0097] According to the experimental design, in this embodiment, the quantum well In 0.3 Ga 0.7 The N material is considered to have a uniform In composition. The thickness of a single quantum well is 2.5 nm, the thickness of a single quantum barrier is 13 nm, and the total thickness of the quantum well layer is 20.0 nm. The average V-pit radius r is 76 nm, and the average V-pit density ρ is 1 × 10⁻⁶. 9 / cm 2 The included angle of the V-shaped pit is 56.08°.

[0098] The process for measuring the absorption coefficient in this embodiment is as follows:

[0099] (1) The device is fabricated into a vertical structure chip with the n-side facing up using a chip process similar to that in Example 1, and has a nickel-silver reflector on the p-side. The difference is that the n-GaN surface is not roughened.

[0100] (2) The reflectivity of the lower interface of the epitaxial layer in the sample (the interface between p-GaN and NiAg electrodes) is the same as that in Example 1, with an average value of 90%. The upper interface of the epitaxial layer in the sample (the interface between n-GaN and air) is considered to be a flat surface, and the reflectivity is obtained by the law of reflection. The calculation formula is as follows:

[0101]

[0102] The refractive index of air is 1, and the refractive index of n(GaN) is calculated by equation (18), where λ is the wavelength. A0, A1, and A2 can be obtained by fitting the GaN refractive index measured by an ellipsometer. Here, the values ​​in the literature are used, where A0 = 2.26, A1 = 330.1, and A2 = 265.7. Substituting these values ​​into equation (18) yields the refractive index of the n(GaN) material. Then, the calculated refractive index is substituted into equation (17) to calculate the reflectivity of the interface on the epitaxial layer in the sample, as shown below. Figure 7 ;

[0103]

[0104] (3) Test the EQE curves of the sample from Example 2 under different bias voltages (0V, -2V, -5V, -8V), as follows: Figure 8 Compared to Example 1, the preparation layer is a superlattice structure. In this case, photons in the 365-425nm range are absorbed by the GaN impurity energy level and by the quantum well material and the superlattice. Therefore, photons in the 425nm-550nm wavelength range are only absorbed by the quantum well material. The EQE curves from -2V to -8V coincide in the wavelength range (425nm~550nm) that are only the optical response of the quantum well structure, indicating that the quantum well is completely in the depletion region. The quantum efficiency in the wavelength range that is only the optical response of the quantum well structure is equal to the external quantum efficiency.

[0105] (4) Similar to process (5) in Example 1, the volume ratio Vratio of the quantum well platform region is found to be 85.5%. Therefore, the EQE calculated by substituting into the formula is:

[0106]

[0107] The absorption coefficient is calculated by substituting the reflectivity of the upper interface and the lower interface of the epitaxial layer in the sample, the total thickness of the quantum well layer, and the EQE into formula (16). Figure 9 .

[0108] The literature reports two models for calculating the absorption coefficient of InGaN bulk materials. Model 1 is as follows:

[0109]

[0110] Model 2 is as follows:

[0111]

[0112] Where: E is the incident photon energy, E g Let a and b represent the band gap of the material. These are empirical values; when the In content is 0.3%, a is 1.6222 and b is 0.0375. Models 1 and 2 are used to calculate the In content. 0.3 Ga 0.7The absorption coefficient of the N-body material was compared with the In measured in Examples 1 and 2. 0.3 Ga 0.7 The absorption coefficients of N-quantum well materials were compared, see Figure 10 The absorption coefficients measured in Examples 1 and 2 both exhibit a tailed curve, while the calculated value is parabolic. This is because the measured values ​​are for the InGaN quantum well material, while the model calculations yield the absorption coefficient for the InGaN bulk material. The absorption coefficient of the InGaN quantum well material measured in Example 1 is significantly larger than the theoretically calculated value. This is because the n-GaN surface was roughened, causing light to enter the sample at an oblique angle, resulting in the absorption of more photons. In this case, the measured absorption coefficient is the device-level equivalent absorption coefficient. When the n-GaN surface is not roughened, as in Example 2, the measured absorption coefficient is the absorption coefficient of the InGaN quantum well material itself, which is close to the theoretical calculation. This also demonstrates that the absorption coefficient measured by this method is relatively accurate.

[0113] Example 3:

[0114] Compared to Example 2, the quantum well structure of the sample does not contain V-pits, and the epitaxial wafer cross-sectional view is shown below. Figure 11 As shown. The process for measuring the absorption coefficient is basically the same as in Example 2, except that:

[0115] Without needing to obtain the volume ratio of the quantum well platform region, the absorption coefficient can be calculated by substituting the measured EQE into formula (16).

[0116] Example 4:

[0117] Compared to Example 3, the epitaxial wafer uses a sapphire substrate and has no pre-layer structure on top of the n-type GaN; the chip fabrication process is a same-side structure process. A cross-sectional view of the epitaxial wafer in this example is shown below. Figure 12 From bottom to top, they are: substrate 111, buffer layer 211, n-type GaN layer 311, quantum well structure 411, p-type AlGaN electron blocking layer 511, and p-type GaN layer 611.

[0118] The substrate 111 is made of sapphire.

[0119] The buffer layer 211 is made of GaN, with a thickness of 2 μm, and is undoped.

[0120] The n-type layer 311 is made of GaN, with a thickness of 2 μm and a Si doping concentration of 5 × 10⁻⁶. 18 / cm 3 .

[0121] The quantum well structure 411 consists of alternating InGaN well layers and GaN barrier layers, with the InGaN well layer material being In... 0.3 Ga 0.7The nitrogen layer is 2.5 nm thick and undoped. The GaN barrier material is GaN, 13 nm thick, with a Si doping concentration of 1 × 10⁻⁶. 17 / cm 3 One InGaN well layer and one GaN barrier layer constitute one cycle, and the 500 quantum well structure contains 8 cycles of InGaN / GaN.

[0122] The p-type AlGaN electron blocking layer 511 material is Al 0.2 Ga 0.8 N, with a thickness of 20 nm and a Mg doping concentration of 1 × 10⁻⁶. 19 / cm 3 .

[0123] The p-type GaN layer 611 is made of GaN, with a thickness of 170 nm and a Mg doping concentration of 5 × 10⁻⁶. 19 / cm 3 .

[0124] According to the experimental design, in this embodiment, the quantum well In 0.3 Ga 0.7 The N material is considered to have a uniform In composition, with a single quantum well thickness of 2.5 nm, a single quantum barrier thickness of 13 nm, and a total quantum well layer thickness of 20.0 nm.

[0125] The process for measuring the absorption coefficient in this embodiment is as follows:

[0126] (1) Indium tin oxide (ITO) was grown on an epitaxial wafer using electron beam evaporation as a transparent current spreading layer. The reflectance of the upper interface (ITO-air interface) of the epitaxial layer was then measured using a spectrophotometer. Figure 13 As shown, the reflectivity in the 360-550nm range is approximately 10%, therefore the reflectivity R1 of the upper interface of the epitaxial layer is 10%. The lower interface of the epitaxial layer in the sample is the contact surface between n-GaN and the sapphire substrate, with a reflectivity of <3%, which is ignored here. That is, the reflectivity R2 of the lower interface of the epitaxial layer in the sample is 0.

[0127] (2) The epitaxial wafer is etched to expose part of the n-GaN, and p-electrodes and n-electrodes are deposited by vapor deposition. The final device schematic diagram is shown below. Figure 14 As shown;

[0128] (3) Similar to step (3) in Example 2, the prepared sample is tested using a quantum efficiency tester to obtain the external quantum efficiency curves under different reverse bias voltages. When the reverse bias voltage is large enough, the quantum well region is completely in the depletion region. At this time, the external quantum efficiency value in the wavelength range of the quantum well structure light response does not change with the increase of the bias voltage. The external quantum efficiency in the wavelength range of the quantum well structure light response is equal to the internal quantum efficiency.

[0129] (4) Substitute the reflectivity of the upper interface of the epitaxial layer, the reflectivity of the lower interface, the total thickness of the quantum well layer and EQE in the sample into formula (16) to calculate the absorption coefficient of the quantum well material.

[0130] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention, and not to limit them; although the present invention 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 of the technical features; and these 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 the present invention.

Claims

1. A method of measuring the absorption coefficient of a quantum well material, characterized by: Comprising the following steps: S1: preparing an epitaxial wafer: growing an epitaxial layer containing a quantum well structure on a substrate, the epitaxial layer comprising a first conductive layer, a quantum well structure, a second conductive layer, the quantum well structure being formed by quantum well layers and quantum barrier layers alternating; S2: obtaining the total thickness of the quantum well layer; S3: performing a chip process on the epitaxial wafer, evaporating p electrodes and n electrodes, obtaining a sample, and obtaining the reflectivity of the upper interface and the lower interface of the epitaxial layer in the sample; S4: applying a suitable bias, testing the external quantum efficiency when the photo-generated carriers in the quantum well are 100% collected, and then drawing an external quantum efficiency curve according to the external quantum efficiency; according to the external quantum efficiency curve, analyzing and stripping out the internal quantum efficiency curve of the quantum well structure light response only; S5: according to the relationship between the absorption coefficient and the internal quantum efficiency, the reflectivity, and the thickness of the quantum well layer, the absorption coefficient is solved; Wherein: the relationship between the absorption coefficient and the internal quantum efficiency, the reflectivity, and the thickness of the quantum well layer is: where ƞ is the internal quantum efficiency, R is the reflectivity of the upper interface of the epitaxial layer in the sample, R is the reflectivity of the lower interface of the epitaxial layer in the sample, and d is the total thickness of the quantum well layer, is the absorption coefficient of the quantum well material.

2. The method of measuring the absorption coefficient of quantum well material according to claim 1, wherein: The total thickness of the quantum well layer in step S2 is obtained according to the experimental design or by X-ray diffraction , a scanning electron microscope, a secondary ion mass spectrometer, and the like. The total thickness of the quantum well layer is determined by the thickness of the quantum well and the number of periods of the quantum well, and the number of periods of the quantum well is greater than or equal to 1.

3. The method of measuring the absorption coefficient of quantum well material according to claim 1, wherein: The reflectivity of the upper interface of the epitaxial layer in the sample in step S3 is the reflectivity of the light incident surface and the air interface, and the reflectivity of the lower interface is the reflectivity of the lower surface of the epitaxial layer and the contact surface of the electrode or the substrate. The reflectivity is obtained by a spectrophotometer, or calculated by the reflection law, or obtained by an empirical value.

4. The method of measuring the absorption coefficient of quantum well material according to claim 1, wherein: The suitable bias in step S4 is determined by the following process: gradually increasing the reverse bias, measuring the external quantum efficiency curve of the sample, and when the external quantum efficiency curve does not change with the increase of the applied reverse bias in the wavelength range of the quantum well structure light response only, any bias greater than or equal to the reverse bias range at this time is the bias when the photo-generated carriers in the quantum well are 100% collected.

5. The method of measuring the absorption coefficient of quantum well material according to claim 1 or 4, wherein: In step S4, in the wavelength range of the quantum well structure light response only, the internal quantum efficiency is equal to the measured external quantum efficiency.

6. The method of measuring the absorption coefficient of quantum well material according to claim 1 or 4, wherein, When the quantum well structure contains V-pits, the measured external quantum efficiency in step S4 needs to be corrected: divide the measured external quantum efficiency by the volume ratio of the quantum well platform region; the internal quantum efficiency is equal to the corrected external quantum efficiency.

7. The method of measuring the absorption coefficient of quantum well material according to claim 1, wherein: When the chip process in step S3 does not contain a surface roughening process, the absorption coefficient obtained in step S5 is the absorption coefficient of the quantum well material.

8. The method of measuring the absorption coefficient of quantum well material according to claim 1, wherein: When the chip process in step S3 contains a surface roughening process, due to the non-normal incidence of the light beam, the absorption coefficient obtained in step S5 is the device-level equivalent absorption coefficient of the quantum well material.

Images