High-throughput design method for low-dimensional materials

Abstract

The application discloses a high-throughput design method for low-dimensional materials, which utilizes existing crystal structures and performs element recombination to construct a structure library for high-throughput calculation, and optimizes a target material according to a calculation result, wherein, the structure is subjected to dimension reduction processing and / or dimension reduction calculation. The application solves the problem that the existing method cannot quickly and efficiently design low-dimensional materials by performing dimension reduction processing and / or dimension reduction calculation on the structure, improves the screening and prediction efficiency, and improves the design accuracy, and can be used to further design high-performance optoelectronic materials and devices.

Description

Technical Field

[0001] This invention belongs to the field of semiconductors, and specifically relates to a low-dimensional material design method. Background Technology

[0002] With the development of the semiconductor industry, the requirements for material parameters in micro- and nano-optoelectronic devices are becoming increasingly stringent, and many existing materials are no longer sufficient to meet these increasingly demanding application requirements. To achieve the desired application results, it is necessary to optimize the structure of existing materials to improve their performance. Furthermore, entirely new materials may need to be constructed based on specific needs, leading to a significant increase in time and costs. Therefore, how to significantly shorten the research and development time of new optoelectronic materials and improve research and development efficiency has become an urgent problem to be solved.

[0003] It is well known that using experimental trial and error to screen materials is time-consuming and labor-intensive. Modeling-based screening, however, can help select materials that meet performance requirements from thousands of candidate materials, significantly narrowing the scope of experimental verification. Combining material screening, multi-scale modeling, and experimental verification can effectively and rapidly screen the desired materials. However, currently, there is a lack of rapid and efficient design methods for low-dimensional materials.

[0004] Therefore, researchers are dedicated to discovering a method that can rapidly, efficiently, and accurately design low-dimensional semiconductor materials. Summary of the Invention

[0005] In view of this, the present invention provides a high-throughput design method for low-dimensional materials. This design method helps to solve the problem that existing methods cannot perform rapid, efficient, and accurate design of low-dimensional materials.

[0006] To achieve the above objectives, according to embodiments of the present invention, the present invention provides a high-throughput design method for low-dimensional materials, which utilizes existing crystal structures and performs element recombination to construct a structure library for high-throughput calculations, and selects and optimizes target materials based on the calculation results, wherein the method includes dimensionality reduction processing and / or dimensionality reduction calculations on the structure.

[0007] Optionally, the method includes the following steps:

[0008] Set the general structural formula for the target materials to be selected;

[0009] Based on the general structural formula of the target material, obtain an existing crystal structure library that conforms to the general formula;

[0010] The existing crystal structure library is reconstructed to obtain a recombinant crystal structure library;

[0011] The recombined crystal structure is reduced in dimension and optimized, and parameters including band structure, absorption spectrum and carrier mobility are calculated. Target materials are then selected based on these parameters.

[0012] Optionally, the existing crystal structure library is reconstructed to obtain a recombined structure library, wherein the reconstruction includes one or more combinations of element substitution and structural relationship adjustment.

[0013] Optionally, the recombinant crystal structure library is subjected to dimensionality reduction and optimization, and parameters including band structure, absorption spectrum, and carrier mobility are calculated. Target materials are then selected based on these parameters. The calculation of the band structure includes calculating the valence band electronic structure. The steps are as follows:

[0014] The three parameters, truncated wave vector k_control, number of bands at the valence band apex Num, and reduced radius R0_red, are obtained by recombining the structure library.

[0015] Construct an effective mass Hamiltonian matrix, and ensure that the Hamiltonian matrix is ​​a Hermitian matrix;

[0016] The Hamiltonian matrix is ​​diagonalized by calling the diagonalization function eig to obtain the band eigenvalues ​​and eigenvectors.

[0017] Use the diag function to extract the diagonal eigenvalues ​​of the diagonalized Hamiltonian;

[0018] Sort the extracted eigenvalues;

[0019] Enter the next wave vector k cycle until the wave vector k_control is truncated, and output the sorted eigenenergys corresponding to all calculated wave vectors to obtain the valence band electronic structure.

[0020] Optionally, an effective mass Hamiltonian matrix is ​​constructed, and it is guaranteed that the Hamiltonian matrix is ​​a Hermitian matrix.

[0021] Optionally, the recombinant crystal structure library is reduced in dimension and optimized, and screening parameters including band structure, absorption spectrum and carrier mobility are calculated. Target materials are obtained by screening according to the screening parameters. The structure optimization includes: step-by-step optimization of the macrostructure of the recombinant material.

[0022] Optionally, the dimensionality reduction includes cutting along the crystal orientation and adding a vacuum layer.

[0023] Optionally, cutting along the crystal direction and adding a vacuum layer includes: cutting the three-dimensional material along the crystal direction to obtain a two-dimensional crystal plane, and then adding a vacuum layer along the cut crystal direction. The vacuum layer.

[0024] Optionally, the high-throughput computation is performed in VASP and Matcloud, and the parameters set in the INCAR file are as follows:

[0025] PREC = ACCURATE

[0026] ENCUT = 600eV

[0027] EDIFF = 1 × 10 -6 eV

[0028]

[0029] LREAL = FALSE;

[0030] When optimizing the structure, set the parameters as follows: ISIF=3, IBRION=2; when calculating self-consistency, band structure, and optical absorption coefficient, set the parameters as follows: ISIF=2, IBRION=-1.

[0031] To achieve the above objectives, according to embodiments of the present invention, the present invention also provides a low-dimensional material, obtained by the low-dimensional material high-throughput design method described above.

[0032] The beneficial effects of this invention are:

[0033] To design low-dimensional materials quickly and efficiently, existing materials can be recombinated to infer potentially undiscovered semiconductor optoelectronic materials. Therefore, existing crystal structures are utilized and elements are recombined to construct a structure library for high-throughput computation. This involves large-scale calculations, followed by selection and optimization based on the results to obtain target materials. Before high-throughput or optimization calculations, appropriate dimensionality reduction is performed, and calculations are then performed on the low-dimensional materials to more accurately determine their physical and optoelectronic properties, thus facilitating selection. Attached Figure Description

[0034] Figure 1 This is a flowchart of a high-throughput design method for low-dimensional materials according to Embodiment 1 of the present invention;

[0035] Figure 2 : Adding to the two-dimensional CsPbCl3 perovskite material in Example 1 of this invention Crystal structure diagram after optimization of vacuum layer structure;

[0036] Figure 3 : This is the band structure diagram of the two-dimensional CsPbCl3 perovskite material in Example 1 of this invention;

[0037] Figure 4 : This is a graph showing the light absorption coefficient of the two-dimensional CsPbCl3 perovskite material in Example 1 of this invention;

[0038] Figure 5 Adding to the two-dimensional Cs2AgBiCl6 perovskite material in Example 1 of this invention Crystal structure diagram after optimization of vacuum layer structure;

[0039] Figure 6 : This is the band structure diagram of the two-dimensional Cs2AgBiCl6 perovskite material in Example 1 of this invention;

[0040] Figure 7 : This is a graph showing the light absorption coefficient of the two-dimensional Cs2AgBiCl6 perovskite material in Example 1 of this invention;

[0041] Figure 8 Flowchart of a method for obtaining the valence band structure of wurtzite semiconductor nanowires;

[0042] Figure 9 Valence band electronic structure of wurtzite CdSe nanowires with a radius of 6 nanometers;

[0043] Figure 10 Valence band electronic structure of wurtzite ZnO nanowires with a radius of 6 nanometers. Detailed Implementation

[0044] The embodiments described are provided to better illustrate the present invention, but are not intended to limit the scope of the invention to the embodiments described. Therefore, non-essential improvements and adjustments made to the embodiments by those skilled in the art based on the above description are still within the scope of protection of the present invention.

[0045] The present application will now be described in detail with reference to the accompanying drawings, which will enable those skilled in the art to readily implement the invention. However, the invention can be implemented in many different ways and is not limited to the exemplary embodiments described herein. The endpoints and any values ​​of the ranges disclosed herein are not limited to the precise ranges or values, and these ranges or values ​​should be understood to include values ​​close to these ranges or values. For numerical ranges, the endpoint values ​​of the various ranges, the endpoint values ​​of the various ranges and individual point values, and individual point values ​​can be combined with each other to obtain one or more new numerical ranges, which should be considered as specifically disclosed herein.

[0046] The objectives and advantages of the exemplary embodiments of the present invention will become clearer and more apparent from the description below, but the objectives and advantages of the exemplary embodiments of the present invention are not limited to the following description. In the description of the exemplary embodiments, detailed explanations of the prior art will be omitted where it is believed that such detailed explanations might unnecessarily obscure the focus of the present invention.

[0047] Basic principle explanation:

[0048] Currently, semiconductor materials, depending on their crystal symmetry, have general structural formulas under different space groups. These formulas include some existing structures that have already been discovered. Structures with the same formula within the same space group exhibit certain similar properties and possess more constants, thus simplifying calculations. Therefore, this invention can first define the general structural formulas of the target material under different space groups. For example, Example 1 corresponds to the perovskite material with the CsPbCl3 structural formula in space group 221 (Pm-3m) and the perovskite material with the A2BCX6 structural formula in space group 225 (Fm-3m). Example 2 corresponds to the wurtzite semiconductor with the AB binary compound structural formula in space group 186 (P63mc). Based on these structures, element substitutions and structural adjustments can be made according to the properties of their respective space groups and formulas, thereby constructing novel materials in the model. The existing materials' various structures are extracted from the modeling data to form an existing crystal structure library. Because the structure of new materials is similar to the existing structure, adjusting the existing structure and starting the reconstruction will more quickly obtain materials with reasonable structures. Reconstructing the existing crystal structure library yields a recombined structure library composed of different structural data after element substitution and structural adjustment. After the above reconstruction steps, because existing materials often have three-dimensional structures, dimensionality reduction processing is needed for different dimensions of low-dimensional materials, such as zero-dimensional, one-dimensional, and two-dimensional (taking Example 1 as an example). First, the recombined structure library is dimensionality reduced, and then targeted high-throughput calculations are performed. High-throughput calculations refer to calculations including a series of parameters such as band structure, absorption spectrum, and carrier mobility. In addition to dimensionality reduction in modeling, it is often necessary to adjust the calculation method for these parameters (such as the calculation adjustment for low-dimensional materials in Example 2). The preliminary screening rules are: response wavelength covers the ultraviolet to visible light band (or according to the required band of different optoelectronic devices); the band gap width of the material is 1.5-3.2 eV; the light absorption is relatively strong; and the carrier mobility is as high as possible.

[0049] Dimensionality reduction was performed using Material Studio software, and calculations were performed using a self-written Linux script and the first-principles calculation software VASP. VASP, short for Vienna Ab-initio Simulation Package, is one of the most popular software programs for materials simulation and computational materials science research. It can calculate the electronic structure of materials (energy levels, charge density distribution, band structures, and electronic density of states), as well as the optical, magnetic, and lattice dynamic properties of materials.

[0050] Example 1

[0051] Specifically, such as Figure 1-7As shown, this paper takes the design of perovskite materials CsPbCl3 (space group 221) and Cs2AgBiCl6 (space group 225) as examples to illustrate the specific design method in detail:

[0052] Sample generation includes setting a general formula for the structure of candidate target materials; obtaining an existing crystal structure library that conforms to the general formula; reconstructing the existing crystal structure library to obtain a recombined structure library; performing dimensionality reduction and optimization on the recombined crystal structure library, and calculating parameters including band structure, absorption spectrum and carrier mobility, and selecting target materials based on the parameters; and performing dimensionality reduction on the target material library.

[0053] S1 selects the general structural formula and space group from databases (such as Materials Project) according to data requirements. The perovskite materials are selected from the original VASP calculation results of the ABX3 structure (general structural formula) in space group 221 and the A2BCX6 structure in space group 225 of Materials Project. More than a dozen materials such as CsCaCl3, CsCaF3, CsPbCl3, and Cs2AgBiCl6 are relatively suitable.

[0054] S2 then uses scripts to create a large number of random combinations of structural changes based on the existing material structure (e.g., element substitution and atomic position adjustment), thereby forming more than 2,000 perovskite materials.

[0055] S3 optimized the structure of the selected perovskite material samples. First, the samples were modeled in Material Studio. Then, the (001) crystal orientation was selected to cut and reduce the dimensionality to two-dimensional material. A vacuum layer was then added in Material Studio to shield the interactions between different layers of the two-dimensional material. Finally, VASP software was used for structural optimization, self-consistent calculations, band structure calculations, and optical absorption coefficient calculations. The more efficient PBE exchange-correlated functional was used in the calculations, with energy and force convergence criteria of 1×10⁻⁶. -6 eV and The cutoff energy is set to 600 eV. When performing structural optimization, a method with a fixed lattice constant in the vacuum layer direction is required; therefore, the OPTCELL file needs to be configured.

[0056] The specific methods for setting other relevant parameters in the INCAR file in the above steps are as follows:

[0057] PREC = ACCURATE

[0058] LREAL = FALSE

[0059] When optimizing the structure, set the parameters as follows: ISIF=3, IBRION=2; when calculating self-consistency, band structure, and optical absorption coefficient, set the parameters as follows: ISIF=2, IBRION=-1.

[0060] For the ABX3 structure in space group 221, due to its smaller number of unit cell atoms, the KPOINTS file was set to 9×9×1 using the Gamma method during structure optimization and self-consistent calculations. For optical absorption coefficient calculations, the KPOINTS file, with more densely packed points in the reciprocal lattice space, was set to 19×19×1 using the Gamma method. For the A2BCX6 structure in space group 225, with its larger number of unit cell atoms, the KPOINTS file was set to 5×5×1 using the Gamma method during structure optimization, 7×7×1 during self-consistent calculations, and 19×19×1 during optical absorption coefficient calculations.

[0061] S4 then uses a self-written Linux calculation script to perform high-throughput calculations on parameters such as the band structure, absorption coefficient, and mobility of the samples. During the calculation, density functional theory is used to calculate X material samples that meet the specified requirements, where X is required to be between 1000 and 10000. Excellent perovskite material samples are then selected from the results. These materials must meet the following conditions: response wavelength covers the ultraviolet to visible light band, and the material's band gap width is between 1.5 and 3.2 eV. Figure 3 , Figure 6 The band structures of the two-dimensional CsPbCl3 and Cs2AgBiCl6 materials shown are illustrated, and it can be seen that the calculated band gaps are all in the range of 1.5-3.2 eV. Furthermore, as... Figure 4 , Figure 7 As shown, the calculated light absorption coefficients of the two-dimensional CsPbCl3 and Cs2AgBiCl6 materials are relatively large.

[0062] Through the above process, the present invention can:

[0063] (1) The design of a large number of samples can be completed quickly, which greatly shortens the material research and development cycle;

[0064] (2) Simultaneously search existing materials in the structure library and design new materials to increase the number of sample types that meet performance requirements.

[0065] (3) The design sample is lossless and will not generate additional material costs.

[0066] Example 2

[0067] Specifically, such as Figure 8-10As shown, ZnO and CdSe of the P63mc space group are used as examples to illustrate how to perform dimensionality reduction. Using a self-developed software program (Low-dimensional semiconductor material electronic structure calculation software V1.0, application number: 2022R11S1344426), the valence band electronic structure of low-dimensional semiconductor nanowires can be obtained more quickly and accurately. The valence band electronic structure can then be used as a screening parameter, and the material can be screened using a method similar to that in Example 1.

[0068] SS1 obtains three parameters for wide-bandgap wurtzite semiconductor nanowires: the truncated wave vector k_control, the number of bands at the valence band apex Num, and the reduced radius R0_red. k_control is the truncated wave vector selected during the calculation, with a step size of 0.005 × 10⁻⁶. 10 m -1 For wide-bandgap semiconductors, Num is set to 6, R0_red is the reduced radius, and the actual radius is R0 = R0_red × 1 × 10. -10 m. Input the three initialization parameters mentioned above, and loop through the wave vector k.

[0069] SS2 constructs the effective mass Hamiltonian matrix of the nanowire system, which should be a Hermitian matrix; the Hamiltonian matrix is ​​diagonalized using the diagonalization function eig to obtain the band eigenvalues ​​and eigenvectors; the diagonal eigenvalues ​​of the diagonalized Hamiltonian are extracted using the diag function; and the extracted eigenvalues ​​are sorted.

[0070] For nanowire systems, to achieve more accurate calculations, the quantum confinement effect needs to be considered. Therefore, the six-band effective mass Hamiltonian of the bulk material is converted into the effective mass Hamiltonian of the low-dimensional nanowire system, specifically:

[0071] The effective mass Hamiltonian of the six bands of a bulk material is represented by the following 6×6 matrix:

[0072]

[0073] Where H U and H L The following is a 3×3 matrix:

[0074]

[0075] Matrix elements P1, P3, F, G, F*, and G* are the effective mass matrix elements of wurtzite material, containing the effective mass parameters and wave vector k. x ,k y and k z m0 is the mass of the free electron, H so It is the spin-orbit coupling Hamiltonian of semiconductor materials;

[0076] For nanowire systems, the motion of charge carriers in the (r,θ) plane perpendicular to the nanowire axis is restricted, while their motion along the nanowire axis is free. Therefore, k... z It is a good quantum number, and k x and k y It is no longer a good quantum number and needs to be rewritten;

[0077] The operator k is not a good quantum number. x and k y Rewritten as a differential operator, i.e.: and Where i is the imaginary unit.

[0078] Setting the zeros of the cylindrical Bessel function: because the cylindrical Bessel function jLh(r h ) is a continuously oscillating function. In the calculation, it is necessary to set the zeros of different orders of Béssel functions, where L... h This refers to the order of the cylindrical Bessel function. For example, the zeroth-order cylindrical Bessel function j0(r) h The first zero of ) is 2.4048, and the second zero is 5.5201; the first-order cylindrical Bessel function j1(r h The first zero point is 3.8317, and the second zero point is 7.0156.

[0079] After transforming the effective mass Hamiltonian of the bulk material into the effective mass Hamiltonian of the nanowire system, the Schrödinger equation for the effective mass of the nanowire can be solved using the special function expansion method to obtain its valence band electronic structure, as follows:

[0080] Since the nanowire system is cylindrical, assuming that the charge carriers are strictly confined within the nanowire system, the wave function of the charge carriers at the nanowire boundary is zero. Therefore, the wave function of the hole can be expanded using the Bessel function into the following six-component form:

[0081]

[0082] in It is L h Cylindrical Bessel function, and It is L h The nth order of columnar Bessel functions h There are zero points, and R0 is the radius of the nanowire. It is L h The normalization constant of the columnar Bessel function; the effective mass Hamiltonian of the nanowire system is written in terms of the following matrix elements under the wave function:

[0083] <ψ h (ρ h,θ h ,z h )|H h |ψ h (ρ h ,θ h ,z h )>

[0084] SS3 enters the next wave vector k-cycle until the wave vector k_control is truncated, and outputs the ordered eigenenergys corresponding to all calculated wave vectors. For example... Figure 9 , 10 The figures show the valence band electronic structures of wurtzite CdSe nanowires with a radius of 6 nm and wurtzite ZnO nanowires with a radius of 6 nm, respectively.

[0085] The above does not list all possible applications.

[0086] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention and are not intended to limit it. Although the present invention has been described in detail with reference to preferred embodiments, those skilled in the art should understand that modifications or equivalent substitutions can be made to the technical solutions of the present invention without departing from the spirit and scope of the technical solutions of the present invention, and all such modifications or substitutions should be covered within the scope of the claims of the present invention.

Claims

1. A high-throughput design method for low-dimensional materials, which utilizes existing crystal structures and performs element recombination to construct a structure library for high-throughput calculations, and selects and optimizes target materials based on the calculation results, characterized in that... This includes dimensionality reduction processing and / or dimensionality reduction calculation of the structure, including the following steps: Set the general structural formula for the target materials to be selected; Based on the general structural formula of the target material, obtain an existing crystal structure library that conforms to the general formula; The existing crystal structure library is reconstructed to obtain a recombinant crystal structure library; The recombined crystal structure library is dimensionality reduced and optimized, and parameters including band structure, absorption spectrum, and carrier mobility are calculated. Target materials are then selected based on these parameters. The calculation of the band structure includes calculating the valence band electronic structure. The steps are as follows: Obtain the three parameters: truncated wave vector k_control, number of bands at the valence band apex Num, and reduced radius R0_red; Construct an effective mass Hamiltonian matrix, and ensure that the Hamiltonian matrix is ​​a Hermitian matrix; The Hamiltonian matrix is ​​diagonalized by calling the diagonalization function eig to obtain the band eigenvalues ​​and eigenvectors. Use the diag function to extract the diagonal eigenvalues ​​of the diagonalized Hamiltonian; Sort the extracted eigenvalues; Enter the next wave vector cycle until the wave vector reaches the truncated wave vector k_control, and output the sorted intrinsic energies corresponding to all calculated wave vectors to obtain the valence band electronic structure. The dimensionality reduction includes: cutting along the crystal orientation and adding a vacuum layer; Cutting along the crystal direction and adding a vacuum layer includes: cutting the three-dimensional material along the crystal direction to obtain a two-dimensional crystal plane, and then adding a vacuum layer of 10 Å-20 Å along the cut crystal direction.

2. The high-throughput design method for low-dimensional materials as described in claim 1, characterized in that, The existing crystal structure library is reconstructed to obtain a recombined crystal structure library, wherein the reconstruction includes one or more combinations of element substitution and structural relationship adjustment.

3. The high-throughput design method for low-dimensional materials as described in claim 1, characterized in that, The recombined crystal structure library is reduced in dimension and optimized, and parameters including band structure, absorption spectrum and carrier mobility are calculated. Target materials are then selected based on these parameters. The structure optimization includes stepwise optimization of the overall structure of the recombined materials.

4. The high-throughput design method for low-dimensional materials as described in claim 1, characterized in that, The high-throughput computation is performed in VASP and Matcloud, and the parameters set in the INCAR file are as follows: ； When optimizing the structure, set the parameters as follows: ISIF=3, IBRION=2; when calculating self-consistency, band structure, and optical absorption coefficient, set the parameters as follows: ISIF=2, IBRION=-1.

5. A low-dimensional material, characterized in that, The design was obtained using the low-dimensional material high-throughput design method as described in any one of claims 1-4.

Images