Lattice-graded buffer layer for epitaxial growth of lattice-mismatched solar cells
By introducing a lattice gradient buffer layer and a composite DBR into a lattice-mismatched solar cell, the dislocation problem caused by lattice mismatch was solved, resulting in higher photon reflection efficiency and radiation resistance, while reducing production costs.
Patent Information
- Application Number
- CN201910677591.9
- Authority / Receiving Office
- CN · China
- Patent Type
- Patents(China)
- Current Assignee / Owner
- Filing Date
- 2019-07-25
- Publication Date
- 2024-11-29
- Estimated Expiration
- 2039-07-25
AI Technical Summary
In existing lattice-mismatched solar cells, when the thickness of the lattice-mismatched epitaxial layer is less than the critical thickness, the lattice constant returns to its intrinsic value, generating a large number of dislocations, which leads to a decrease in material quality and a narrow DBR reflection spectrum range, limiting the improvement of cell performance.
By combining a lattice-gradient buffer layer with a composite DBR, and by stacking multiple DBRs, a gradient-changing lattice constant and refractive index are designed to form a broadband reflector, reducing dislocation density and widening the reflection range. Combined with a strain compensation structure, stress is released.
It effectively reduces the dislocation density of the lattice-mismatched epitaxial layer, broadens the reflection spectrum range, improves the stability and radiation resistance of the battery, shortens the growth time, and reduces costs.
Smart Images

Figure CN110491957B_ABST
Abstract
Description
Technical Field
[0001] The invention relates to the technical field of solar photovoltaic power generation, and in particular to a lattice gradient buffer layer used for epitaxial growth of lattice mismatch solar cells. Background Art
[0002] With the development of modern industrial technology, energy has become the most important material basis for determining social progress in the survival and development of human society. With the development of society, coal, oil, natural gas and other sources of environmental pollution are serious and non-renewable energy is gradually decreasing. Therefore, the development of green new energy is imminent. Solar energy is an inexhaustible clean energy, so researching and developing solar cell technology will alleviate the shortage of energy in the future to a certain extent. From the development history of photovoltaic power generation technology, solar cells can be roughly divided into three categories: the first generation of crystalline silicon solar cells, the second generation of thin-film solar cells and the third generation of gallium arsenide multi-junction solar cells. Among them, the GaInP / GaInAs / Ge triple-junction solar cell composed of GaInP, GaInAs and Ge sub-cells is the mainstream structure of traditional gallium arsenide multi-junction cells. The conversion efficiency under 500 times concentration exceeds 41%, which is much higher than that of crystalline silicon cells, and there is room for further improvement.
[0003] The traditional triple junction cell structure maintains lattice matching as a whole, with a band gap combination of 1.85 / 1.40 / 0.67eV. However, for the solar spectrum, the band gap of this cell is not the best combination. Due to the large band gap between the GaInAs sub-cell and the Ge sub-cell, the current of the bottom cell in this structure is much larger than that of the middle cell and the top cell. Since the bottom, middle and top triple junction sub-cells are connected in series, according to the current mechanism of the series structure, the current is determined by the smallest current among the three sub-cells. This structure causes a large part of the solar energy loss, limiting the improvement of cell performance.
[0004] Theoretical analysis shows that in order to improve the photoelectric conversion efficiency of triple junction cells, it is necessary to reduce the band gap of the absorption area of the middle cell and the top cell, so that the middle cell and the top cell absorb more light, thereby increasing the current of the middle and top cells and reducing the current of the bottom cell, and finally realizing a current-matched triple junction solar cell. Based on this analysis, the MM (Metamorphic) structure solar cell with lattice mismatch structure was proposed. The MM structure triple junction cell was first applied to the CPV market, and its conversion efficiency can reach more than 42%; in recent years, the MM structure triple junction cell products have been applied to space cells, and the conversion efficiency can reach more than 32%, and its post-irradiation attenuation can reach the level of conventional triple junction cells, that is, the efficiency attenuation is less than 18%, which is much better than other types of photovoltaic cell products. However, when epitaxial lattice mismatch materials are used, if the thickness of the mismatched epitaxial layer is less than the critical thickness, the lattice constant will be consistent with the substrate under the action of deformation energy. Once the critical thickness is exceeded, its lattice constant will return to the inherent value, thereby generating a large number of mismatch dislocations and reducing the material quality. Generally, the size of the critical thickness is related to the mismatch degree. The larger the lattice mismatch degree, the smaller the critical thickness. For heteroepitaxial materials with large lattice mismatch, improving the epitaxial quality of the materials becomes a bottleneck for their application in a wider range of fields and for further improving device performance.
[0005] To solve this problem, the commonly used method is to use a stress-releasing gradient buffer layer (usually InGaAs or GaInP material) to connect the lattice-mismatched Ge substrate and InGaAs material. 0.9 In 0.1 As material can be used on Ge substrate and Ga 0.9 In 0.1 A series of GaAs with varying composition are grown between the As materials. 1-x In x As buffer layer, the composition x changes continuously from 0.01 to 0.1, which matches the lattice of Ge substrate. 1-x In x As buffer layer can release the stress caused by lattice mismatch and reduce Ga 0.9 In 0.1 Defects such as dislocations generated in As materials.
[0006] In order to ensure that the lattice of the mismatched material is completely relaxed and to avoid the increase of lattice dislocations, the composition change rate of the composition gradient buffer layer cannot be too fast, so the thickness of the buffer layer is relatively thick. However, the buffer layer has no special effect in optoelectronics, so people have done relevant research on thinning the buffer layer on the basis of ensuring the effect of the buffer layer. Different from this, the present invention switches the idea and combines the lattice gradient buffer layer with the distributed Bragg reflector (DBR), aiming to develop the potential advantages of the lattice gradient buffer layer so that it also has optical effects.
[0007] The application of DBR structure in semiconductor devices (including light-emitting diodes LED and solar cells) has been relatively mature, and its unique photon reflection ability has greatly improved the performance of semiconductor devices. For example, in a three-junction GaAs solar cell, on the one hand, DBR can significantly reduce the absorption coefficient. Reflecting a part of the excessive photons reaching the bottom battery to prevent them from being converted into heat and released into the battery system is very important for improving battery stability and extending battery life. In order to further reduce the absorption coefficient, the present invention uses multiple sets of DBR to broaden the photon reflection range and increase the number of reflected photons. On the other hand, DBR has a significant effect on improving radiation resistance. Since a large number of irradiation experimental results show that the radiation resistance of GaInAs sub-cells is much worse than that of GaInP sub-cells, it is analyzed that the reason is that the As atomic radius is larger, and its position is not easy to recover after high-energy particle irradiation. DBR is set under the neutron battery, and the DBR structure is adjusted to reflect the sunlight of the corresponding band, so that the photons that are not absorbed by the GaInAs material for the first time are reflected back and absorbed twice, which is equivalent to increasing the effective absorption thickness of GaInAs in disguise, which can effectively reduce the design thickness of GaInAs sub-cells, which is conducive to improving radiation resistance. In the field of LEDs, enhancing light output efficiency through DBR is very important for improving brightness.
[0008] In addition, if other optimizations are applied to the present invention, the advantages of the present invention in improving device performance can be further enhanced. For example, in terms of material selection, a rare nitrogen material with good plasticity and film hardening effect is used, so that defects such as threading dislocations are changed from longitudinal to lateral propagation and stress is released, which has an excellent effect on filtering dislocations. In addition, the DBR is designed as a periodic structure of strain compensation, which improves the collection of minority carriers and can also play a good role as a dislocation barrier.
[0009] In summary, the introduction of this lattice gradient buffer layer with wide-spectrum reflection function into the lattice mismatch structure multi-junction solar cell can not only meet the requirements of reducing the absorption coefficient, but also improve the radiation resistance of the cell, and improve the problem of epitaxial crystal quality deterioration caused by threading dislocations, while reducing the process steps, growth time and raw material loss, which is conducive to reducing costs. In short, the advantages of MM structure multi-junction cells can be maximized to improve cell efficiency. Summary of the invention
[0010] The purpose of the present invention is to overcome the shortcomings and deficiencies of the prior art, and proposes a lattice gradient buffer layer (also called a lattice gradient broadband reflector) applied to the epitaxial growth of a lattice mismatched solar cell, which has a broadband reflection effect. By combining the functions of the gradient buffer layer and the composite DBR, it can reduce the defect density in the epitaxial layer introduced by the mismatch, improve the material quality, and enhance the radiation resistance of the GaInAs sub-cell. It can also broaden the wavelength range of the reflection spectrum, reduce the absorption coefficient, and thus improve the battery stability, which is particularly beneficial for the application in aviation power supplies. At the same time, it reduces the process steps, growth time and raw material loss, which is beneficial to reducing costs.
[0011] To achieve the above-mentioned purpose, the technical solution provided by the present invention is: a lattice gradient buffer layer applied to epitaxial growth of a lattice mismatched solar cell, wherein the lattice gradient buffer layer is located between the lattice mismatched epitaxial material and the substrate of the lattice mismatched solar cell, or between the lattice mismatched epitaxial material and the epitaxial layer with the same lattice constant of the substrate, and is formed by stacking a plurality of sets of composite DBRs with different reflection bands, wherein each set of composite DBRs is composed of a plurality of pairs of DBRs stacked together, and the lattice constants between adjacent DBR pairs vary in a gradient, and each pair of DBRs comprises two layers of semiconductor material, wherein the two layers of semiconductor material have different refractive indices but the same lattice constants or there is a mismatch that can achieve strain compensation.
[0012] Furthermore, the lattice constant of the first pair of DBR pairs adjacent to the substrate matches the substrate, the last pair of DBR pairs matches the lattice mismatched epitaxial material, and the lattice constant of the DBR pairs between them increases gradually from the first pair of DBR pairs to the last pair of DBR pairs.
[0013] Furthermore, the composite DBR has at least two sets, and each set of DBR has a reflection band of (λ1-Δλ) to (λ1+Δλ) to (λ n -Δλ)~(λ n +Δλ), where the comprehensive range is (λ1-Δλ)~(λ n +Δλ) is determined by the band gap of the adjacent sub-cells, λ1 is the central reflection wavelength of the first set of composite DBR, λ n is the central reflection wavelength of the nth set of composite DBR, n is an integer, Δλ is 1 / 2 of the reflection range of a certain set of composite DBR, the central reflection wavelength λ is spaced in the range of 10 to 80 nm, and each set has a period of 3 to 20 pairs.
[0014] Furthermore, the two semiconductor material layers of each pair of DBR pairs need to have different refractive indices, and As-based materials or P-based materials are selected, such as GaInNAs, AlGaInAs, GaInP or AlGaInP, and their thickness design follows the formula: Where d is the thickness, λ is the central reflection wavelength of the expected reflection band, and n is the refractive index of the corresponding semiconductor material. The refractive index of As series materials is greater than that of P series materials, and the higher the aluminum component in the same series of materials, the lower the refractive index.
[0015] Furthermore, the lattice constants of the two semiconductor material layers of each pair of DBRs can be the same, and mismatch that can achieve strain compensation is also allowed, with the mismatch degree being in the range of 0.01% to 5%.
[0016] Compared with the prior art, the present invention has the following advantages and beneficial effects:
[0017] 1. The present invention can connect two structural units with large lattice mismatch, achieve lattice transition and reduce the density of defects such as dislocations.
[0018] 2. By introducing multiple sets of composite DBRs with different reflection bands, the comprehensive reflection half-width can be increased from (100±30)nm to (150±30)nm, which can effectively broaden the DBR reflection range, reduce the absorption coefficient, and help improve the stability of the battery.
[0019] 3. The lattice gradient buffer layer (wide-spectrum reflector) of the present invention improves the problem of narrow photon reflection range of a single set of DBR reflection spectrum. The increase in the number of reflected photons can further reduce the effective thickness of the sub-cell and improve the radiation resistance.
[0020] 4. The composite DBR with wide-spectrum reflection is organically integrated into the lattice gradient buffer layer to improve the performance and stability of the battery while reducing the process steps, growth time and raw material loss, which is conducive to reducing the cost of solar cells.
[0021] 5. Existing semiconductor materials with mature epitaxial processes, such as GaInNAs, AlGaInAs, GaInP, and AlGaInP, are available for selection. Due to the limitations of the specific application location and the band gap of adjacent cells, if GaInNAs with rigid characteristics or AlGaInP with lattice strengthening effects can be used to combine them into DBRs and apply them to MM structure multi-junction solar cells, it can improve photon absorption while filtering defects and better play the role of dislocation barrier.
[0022] 6. The high refractive index (n H ) and low refractive index (n L ) layer is designed as a strain compensation structure, see Figure 1 As shown, the stress introduced by the lattice mismatch can be released, and the interface subjected to tensile and compressive strains is conducive to the lateral slip of penetrating dislocations, further reducing the dislocation density.
[0023] Through experiments, the three-junction MM structure solar cell designed by the present invention was prepared by Vecco's MOCVD, and the growth time was shortened by about 46 minutes compared with the traditional process, which greatly improved the production efficiency and saved a lot of raw materials. The X-ray diffractometer (XRD) test analysis showed that the half-width of the top cell in the three-junction MM structure solar cell epitaxial wafer made by this method had no significant difference from the traditional structure, indicating that the defect density caused by lattice mismatch was reduced to a lower level. The white light reflection spectrum test showed that the reflection wave range of the lattice gradient buffer layer was widened from (840~940)nm to (840~1020)nm, and the overall reflection intensity did not decrease significantly. Through the reflectivity test analysis, compared with the three-junction MM structure battery with a single set of DBR structure, the absorption coefficient of the MM three-junction solar cell made by the present invention was reduced from 0.927 to 0.91, the thermal performance was improved, and the stability of the battery was significantly enhanced.
[0024] After analysis and comparison, the triple-junction MM solar cell made by the present invention has significantly improved radiation resistance compared with the traditional DBR, and the conversion efficiency under the AM0 spectrum can be improved by 2.1% (as shown in Table 1 below), which can greatly improve the output power of the space power supply.
[0025] Table 1 Analysis of radiation resistance performance of MM structure triple junction cell using traditional DBR and lattice gradient buffer layer (broad spectrum reflector) under AM0 spatial spectrum
[0026] BRIEF DESCRIPTION OF THE DRAWINGS
[0027] Figure 1 Schematic diagram of the structure of the lattice gradient buffer layer (also called lattice gradient broadband reflector) of the present invention.
[0028] Figure 2 The schematic diagram is a structural diagram of a lattice-mismatched triple-junction solar cell using the lattice gradient buffer layer (also referred to as a lattice gradient broadband reflector) of the present invention. DETAILED DESCRIPTION
[0029] To further illustrate the content of the present invention, the present invention is described in detail below in conjunction with specific embodiments and drawings.
[0030] like Figure 1 and Figure 2As shown, the lattice mismatched triple junction solar cell provided in this embodiment includes a Ge substrate 1, wherein the Ge substrate 1 is a p-type Ge single crystal wafer with a single-side polishing and a 004 orientation; a GaInP nucleation layer 2, a GaInAs buffer layer 3, a first tunnel junction 4, a lattice gradient buffer layer 5 (also referred to as a lattice gradient wide-spectrum reflector), a GaInAs subcell 6, a second tunnel junction 7 and a GaInP subcell 8 are sequentially grown on the polished surface of the Ge substrate 1 in a layered stacking structure from bottom to top; wherein the lattice gradient buffer layer 5 is formed by stacking two sets of composite DBRs 51 and 52 with different reflection bands, and the two sets of composite DBRs 51 and 52 are composed of multiple pairs of DBR pairs stacked together, and the lattice constants between adjacent DBR pairs vary in a gradient. Each pair of DBR pairs contains two layers of semiconductor material. The refractive index of the two layers of semiconductor material is different, but the lattice constant is the same or a mismatch that can achieve strain compensation is allowed to exist. The mismatch is in the range of 0.01% to 5%. Such a design can not only greatly reduce the density of defects such as a large number of dislocations introduced by the growth of lattice-mismatched epitaxial materials, but also give full play to the role of reflectors. At the same time, it can reduce the process steps, growth time and raw material loss, which is conducive to reducing costs. In addition, the lattice constant of the first pair of DBR pairs adjacent to the substrate matches the substrate, and the last pair of DBR pairs matches the lattice of the lattice-mismatched epitaxial material, while the lattice constant of the DBR pairs between them increases in a gradient direction from the first pair of DBR pairs to the last pair of DBR pairs; the reflection band of each set of DBRs is (λ1-Δλ)~(λ1+Δλ) to (λ n -Δλ)~(λ n +Δλ), where the comprehensive range is (λ1-Δλ)~(λ n +Δλ) is determined by the band gap of the adjacent sub-cells, λ1 is the central reflection wavelength of the first set of composite DBR, λ n is the central reflection wavelength of the nth set of composite DBR, n is an integer, Δλ is 1 / 2 of the reflection range of a certain set of composite DBR, the central reflection wavelength λ is spaced in the range of 10 to 80 nm, and each set has 3 to 20 pairs of cycles; the two layers of semiconductor material of each pair of DBR pairs need to have different refractive indices, and As or P materials can be selected, including GaInNAs, AlGaInAs, GaInP, AlGaInP and other alloy materials, and their thickness design follows the formula: Where d is the thickness, λ is the central reflection wavelength of the expected reflection band, and n is the refractive index of the corresponding semiconductor material. The refractive index of As series materials is greater than that of P series materials, and the higher the aluminum component in the same series of materials, the lower the refractive index.
[0031] In this embodiment, the lattice gradient buffer layer 5 is composed of GaInAs / AlGaInAsDBR and GaInNAs / AlGaInAs DBR with gradually increasing In content. The lattice constant of the first pair of GaInAs / AlGaInAs DBR is the same as that of the substrate, and the last pair of GaInNAs / AlGaInAsDBR is the same as that of the GaInAs subcell. Both are p-type doped layers with a hole concentration of 1×10 18 / cm 3 The reflection wavelength range of the GaInAs / AlGaInAs DBR is 840-940nm, wherein the number of pairs of GaInAs / AlGaInAs combination layers is 12 pairs; the GaInAs material and the AlGaInAs material are designed to be strain-compensated structures, and the lattice mismatch is 0.3%. The reflection wavelength range of the GaInNAs / AlGaInAs DBR is 930-1020nm, wherein the number of pairs of GaInNAs / AlGaInAs combination layers is 4 pairs; the GaInNAs material and the AlGaInAs material are designed to be strain-compensated structures, and the lattice mismatch is 0.3%.
[0032] The GaInP nucleation layer 2 , the GaInAs buffer layer 3 , and the first tunnel junction 4 all maintain lattice matching with the Ge substrate 1 .
[0033] The GaInP nucleation layer 2 is an n-type doped layer with an electron concentration of 2×10 18 / cm 3 , with a thickness of 5nm.
[0034] The GaInAs buffer layer 3 is an n-type doped layer with an electron concentration of 4×10 18 / cm 3 , with a thickness of 500nm.
[0035] The first tunnel junction 4 is a p-AlGaInAs / n-GaInP structure, wherein the thickness of the p-AlGaInAs / n-GaInP is 12 nm.
[0036] The total thickness of the GaInAs sub-cell 6 is 1800 nm, and the optical band gap of the GaInAs material is 1.3 eV.
[0037] The second tunnel junction 7 is a p-AlGaInAs / n-GaInP structure, wherein the thickness of the p-AlGaInAs / n-GaInP is 12 nm.
[0038] The total thickness of the GaInP sub-cell 8 is 600 nm, and the optical band gap of the GaInP sub-cell material is 1.85 eV.
[0039] In summary, the present invention combines the characteristics of the lattice gradient buffer layer and DBR (Distributed Brag Reflector), and introduces the functions of the gradient buffer layer and DBR into one into the MM structure multi-junction solar cell, which can play the role of the lattice gradient buffer layer, reduce the defect density in the epitaxial layer introduced by the mismatch, improve the material quality, and broaden the reflection spectrum wavelength range, improve the radiation resistance of the GaInAs sub-cell, reduce the absorption coefficient, solve the problem that too many photons entering the bottom cell are converted into heat energy to affect the stability of the cell, and at the same time, reduce the process steps, growth time and raw material loss, which is conducive to reducing costs. In short, the present invention can give full play to the photoelectric conversion efficiency of the lattice mismatch solar cell, which is worth promoting.
[0040] The above-described embodiments are only preferred embodiments of the present invention and are not intended to limit the scope of implementation of the present invention. Therefore, all changes made according to the shape and principle of the present invention should be included in the protection scope of the present invention.
Claims
1. Lattice gradient buffer layer used in epitaxial growth of lattice mismatched solar cells, Features: The lattice gradient buffer layer is located between the lattice mismatched epitaxial material and the substrate of the lattice mismatched solar cell, or between the lattice mismatched epitaxial material and the epitaxial layer with the same lattice constant of the substrate, and is formed by stacking multiple sets of composite DBRs with different reflection bands, each set of composite DBRs is composed of multiple pairs of DBR pairs stacked together, and the lattice constants between adjacent DBR pairs change in a gradient, each pair of DBR pairs includes two semiconductor material layers, the two semiconductor material layers have different refractive indices but the same lattice constants or there is a mismatch that can achieve strain compensation, the lattice constants of the two semiconductor material layers of each pair of DBR pairs can be the same, and a mismatch that can achieve strain compensation is also allowed, and the mismatch degree is in the range of 3% to 5%.
2. The lattice gradient buffer layer for epitaxial growth of lattice mismatched solar cells according to claim 1, Features: The lattice constant of the first pair of DBR pairs adjacent to the substrate matches the substrate, the last pair of DBR pairs matches the lattice mismatched epitaxial material, and the lattice constant of the DBR pairs between them increases gradually from the first pair of DBR pairs to the last pair of DBR pairs.
3. The lattice gradient buffer layer for epitaxial growth of lattice mismatched solar cells according to claim 1, Features: The composite DBR has at least two sets, and each set of DBR reflection band (λ 1 -Δλ)~(λ 1 +Δλ) to (λ n -Δλ)~(λ n +Δλ), where the comprehensive range (λ 1 -Δλ)~(λ n +Δλ) is determined by the band gap of the adjacent sub-cell, λ 1 is the central reflection wavelength of the first set of composite DBR, λ n is the central reflection wavelength of the nth set of composite DBR, n is an integer, Δλ is 1 / 2 of the reflection range of a certain set of composite DBR, the central reflection wavelength λ is spaced in the range of 10 to 80 nm, and each set has a period of 3 to 20 pairs.
4. The lattice gradient buffer layer for epitaxial growth of lattice mismatched solar cells according to claim 1, Features: The two semiconductor material layers of each pair of DBR pairs need to have different refractive indices. As-based materials or P-based materials are selected, including GaInNAs, AlGaInAs, GaInP or AlGaInP. The thickness design follows the formula: Where d is the thickness, λ is the central reflection wavelength of the expected reflection band, and n is the refractive index of the corresponding semiconductor material; Among them, the refractive index of As series materials is greater than that of P series materials. The higher the aluminum component in the same series of materials, the lower the refractive index.
Citation Information
Patent Citations
Novel DBR structure for lattice-mismatching multi-junction solar cell
CN109301006A
Forward lattice mismatch triple-junction solar cell
CN109755340A
Lattice gradient buffer layer applied to lattice mismatch solar cell epitaxial growth
CN210272385U
