Method and device for evaluating the effect of doping on the performance of lithium iron phosphate cathode material

Abstract

The application discloses a kind of method and device for evaluating the influence of doping on the performance of lithium iron phosphate positive electrode material, belong to the technical field of lithium iron phosphate positive electrode material.The method comprises: geometric optimization is carried out on lithium iron phosphate (LFP) supercell model, based on the S-LFP supercell model obtained after optimization, the doping model of the optimal doping site is constructed, and geometric optimization is carried out to obtain X-LFP supercell model;Based on X-LFP supercell model, evaluate the influence of doping atoms on the overall performance and / or lithium ion diffusion performance of LFP, wherein the geometric optimization is realized by spin polarization density functional calculation based on GGA-PBE functional and projected augmented wave (PAW) method.The quantitative relationship between doping parameters and material performance can be obtained by the application, the influence of doping on the performance of LFP positive electrode material is accurately evaluated, and the research and development cycle is significantly shortened, and the research and development cost is reduced.

Description

Technical Field

[0001] This invention relates to the technical field of lithium iron phosphate cathode materials, and in particular to a method for evaluating the effect of doping on the performance of lithium iron phosphate cathode materials. Background Technology

[0002] Lithium iron phosphate (LiFePO4, LFP) has become a core choice for cathode materials in lithium-ion batteries due to its advantages such as high theoretical specific capacity, excellent safety, good cycle stability, and low raw material cost. It is widely used in new energy vehicles, energy storage power stations, and other fields. However, LFP also has the following inherent defects: low electronic conductivity and small lithium-ion diffusion coefficient, which limits its performance in high-rate charge and discharge scenarios, making it difficult to meet the ever-increasing demand for battery power density in the new energy industry.

[0003] To address the aforementioned issues, existing technologies generally employ doping modification to optimize the performance of lithium iron phosphate (LFP). This involves introducing heteroatoms into the LFP crystal structure to modulate its electronic and crystal structures, thereby improving electronic conductivity and lithium-ion migration capabilities. Carbon (C) and boron (B) are two of the most common doping elements. C possesses excellent electrical conductivity, which can improve the material's electronic conductivity and thus its rate performance. B has a small atomic radius and high electronegativity; doping with boron can influence the crystal structure and electron cloud distribution of lithium iron phosphate, optimizing the lithium-ion diffusion path and improving ionic conductivity.

[0004] Some studies have further attempted to co-dope CB, hoping to achieve synergistic effects by utilizing the electronic conduction advantage of C and the lattice regulation advantage of B. However, existing co-doping studies have the following obvious limitations: (1) The studies are mainly based on experimental trial and error. By preparing CB co-doped LFP samples with different doping ratios and different doping sites, the performance is analyzed by electrochemical tests (such as cyclic voltammetry and charge-discharge tests) and physical characterization (such as XRD, XPS, and SEM). There is a lack of theoretical modeling and simulation prediction methods that can be used as guidance or reference. (2) The studies cannot reveal the intrinsic influence mechanism of CB co-doping on the electronic structure (such as band gap and density of states) and lithium-ion diffusion path of LFP at the atomic level. It is difficult to obtain the quantitative relationship between doping parameters (such as doping ratio and site) and material properties. (3) The experimental cycle is long (usually requiring several weeks or even months), the cost is high, and the efficiency of doping parameter optimization is low.

[0005] Therefore, there is an urgent need in this field to develop a method that can systematically evaluate the impact of CB co-doping on the performance of lithium iron phosphate cathode materials at the atomic level, in order to replace the traditional experimental trial-and-error mode, shorten the R&D cycle, reduce R&D costs, and provide technical guidance for the optimization of doping schemes. Summary of the Invention

[0006] To address the shortcomings of existing technologies, the present invention aims to provide a method and apparatus for evaluating the impact of doping on the performance of lithium iron phosphate cathode materials. This method can reveal the doping mechanism at the atomic level, obtain the quantitative relationship between doping parameters and material performance, and systematically and accurately evaluate the impact of doping on the performance of lithium iron phosphate cathode materials. It overcomes the shortcomings of existing technologies that are mainly based on experimental trial and error, shortens the research and development cycle, reduces research and development costs, and provides technical guidance for obtaining the optimal doping scheme.

[0007] The technical solution of the present invention is as follows: A method for evaluating the effect of doping on the performance of lithium iron phosphate cathode materials, comprising: S1 extends the original cell model of lithium iron phosphate into a supercell model, and performs the first geometric optimization on it to obtain a stable lithium iron phosphate supercell model, namely the S-LFP supercell model. S2 Based on the S-LFP supercell model, a doped LFP supercell model is constructed, consisting of doped atoms, including: S21 determines the optimal doping site of the doped atom in the S-LFP supercell model by comparing the changes in system energy; S22 constructs a doped LFP supercell model by replacing one original atom at the optimal doping site in the S-LFP supercell model with one doped atom. S3 performs a second geometric optimization on the doped LFP supercell model to obtain a stable doped LFP supercell model, namely the X-LFP supercell model. S4. Based on the X-LFP supercell model, evaluate the impact of doping atoms on the overall performance and / or lithium-ion diffusion performance of lithium iron phosphate. The first and second geometric optimizations are based on first-principles density functional theory and are achieved by spin-polarized density functional calculations using the PBE functional and projected fused plane wave (PAW) method in the generalized gradient approximation. During the calculation, lattice parameters and atomic coordinates relax simultaneously until the energy change and interatomic interaction forces meet the convergence criteria.

[0008] In the above scheme of this invention, supercell expansion can effectively avoid the influence of boundary effects on the calculation results, while balancing computational accuracy and efficiency. The geometric optimization method adopted uses a full-structure optimization approach with synchronous relaxation lattice parameters and atomic coordinates, combined with the PBE functional under the generalized gradient approximation (GGA) framework to accurately describe electron exchange correlations, the projected plane wave (PAW) method to efficiently handle the interaction between ionic cores and valence electrons, and spin polarization density functional calculations to fully restore the lattice distortion, electron spin characteristics, and atomic coordination environment changes of the doped system, thereby accurately obtaining the lowest energy and structurally stable ground state configuration of doped lithium iron phosphate. At the same time, the conjugate gradient method is used for geometric optimization, which has high convergence efficiency and small memory footprint, and combines computational stability and efficiency in atomic-scale supercell structure optimization.

[0009] According to some preferred embodiments of the present invention, S4 includes: S41. Based on the X-LFP supercell model, the doping formation energy is calculated, and the influence of dopant atoms on the stability of lithium iron phosphate is evaluated using the doping formation energy; wherein, the doping formation energy is calculated using the following calculation model: in, Indicates the doping formation energy. This represents the total energy of the doped system obtained from the X-LFP supercell model. E represents the total energy of lithium iron phosphate obtained from the S-LFP supercell model. y E z Let E represent the reference state energies of the dopant atom and the doped atom, respectively, where E y The values ​​are derived from the ground-state energy of a single-element crystal doped with y atoms, E. z The numerical values ​​are taken from the average chemical potential of the doped atom z in the S-LFP supercell model, n y n z These represent the number of doping atoms and the number of atoms that are doped, respectively.

[0010] More preferably, the doping formation energy can be calculated using the following computational model: in, This represents the total energy of the doped system. When calculating C doping, it corresponds to the total energy of the C-LFP supercell model; when calculating B doping, it corresponds to the total energy of the B-LFP supercell model; and when co-doping CB, it corresponds to the total energy of the CB-LFP supercell model. E P This represents the reference state energy of the doped P atom, and the value is taken from the average chemical potential of the P atom in the S-LFP supercell model. E C ,E B These represent the reference state energies of C and B atoms, respectively, with values ​​taken from the ground state energies of C and B elemental crystals.

[0011] According to some preferred embodiments of the present invention, S4 further includes: S42 Based on the X-LFP supercell model, calculate one or more of the band gap, density of states, and differential charge density of the doped system, and evaluate the influence of doped atoms on the electronic conductivity of lithium iron phosphate based on the calculation results. The calculation of the bandgap value includes: Static self-consistent calculations are performed on the X-LFP supercell model to obtain the ground-state self-consistent charge density of the doped system. Based on this ground-state self-consistent charge density, a high-symmetry point path within the Brillouin zone is selected. Along the selected high-symmetry point path, the electronic eigenvalues ​​corresponding to each sampling point are calculated to obtain the energy-wave vector relationship curve, i.e., the band structure diagram. Through the band structure diagram, the energy difference between the valence band top and the conduction band bottom is obtained, i.e., the band gap value. The calculation of the density of states includes: The Brillouin zone is uniformly sampled using the ground-state self-consistent charge density, and the number of electronic states in each sampling interval is counted to obtain the total density of states of the doped system and the partial density of states of each atom. The calculation of the differential charge density includes: The total charge density of the doped system is extracted from the calculation results of the X-LFP supercell model obtained through the second geometric optimization; keeping the lattice parameters and atomic coordinates of the X-LFP supercell model unchanged, all doped atoms are deleted to construct a lithium iron phosphate framework model, and the total charge density of the model is calculated to obtain the framework charge density; keeping the lattice parameters and atomic coordinates of the X-LFP supercell model unchanged, only one type of doped atom is retained to construct an isolated doped atom model, and its total charge density is calculated to obtain the isolated doped atom charge density; the total charge density of the doped system is subtracted from the framework charge density, and then the isolated doped atom charge density of all doped atoms is subtracted to obtain the differential charge density.

[0012] In specific implementations, the differential charge density of C-doped or B-doped is obtained by subtracting the framework charge density from the total charge density of the doped system, and then subtracting the isolated doped atom charge density of C atoms or the isolated doped atom charge density of B atoms. The differential charge density of CB co-doped is obtained by subtracting the framework charge density from the total charge density of the doped system, and then subtracting the isolated doped atom charge density of C atoms and the isolated doped atom charge density of B atoms.

[0013] According to some preferred embodiments of the present invention, the evaluation of the effect of dopant atoms on lithium-ion diffusion performance includes: S51 Based on the crystallographic features of one-dimensional diffusion channels and the CI-NEB method, the diffusion path of lithium ions in the X-LFP supercell model is constructed to obtain the diffusion model; S52 performs a third geometric optimization on the diffusion model to obtain a stable diffusion model, namely the ST-LFP model; S53 Based on the ST-LFP model, calculate the lithium-ion diffusion barrier and / or diffusion coefficient of the doped system, and evaluate the effect of doping on lithium-ion migration ability. The third geometric optimization is achieved by performing spin polarization density functional calculations using the PBE functional and projected fused plane wave (PAW) method in the generalized gradient approximation. During the calculation, the lattice parameters and atomic coordinates relax simultaneously until the energy change and interatomic interaction forces meet the convergence criteria. This is completely consistent with the first and second geometric optimizations, except for the convergence criteria.

[0014] According to some preferred embodiments of the present invention, the convergence criterion for the third geometric optimization is: energy change less than or equal to 1 × 10⁻⁶. -4 eV / atom, interatomic interaction force is less than or equal to 0.05 eV / Å.

[0015] According to some preferred embodiments of the present invention, S51 includes: selecting a lithium ion in the supercell of the X-LFP supercell model as the migration target, its state at the original lattice site is the initial state, setting an adjacent lithium ion lattice site as a vacancy, and the state of its migration to the vacancy site is the final state. Using the CI-NEB method, four intermediate states are inserted between the initial state and the final state to construct a diffusion path and obtain a diffusion model; S52 includes: performing the third geometric optimization on the initial state and the final state of the diffusion model respectively to ensure that the initial state and the final state are both local minimum energy configurations to obtain the ST-LFP model.

[0016] According to some preferred embodiments of the present invention, the calculation of the lithium-ion diffusion barrier and / or diffusion coefficient includes: reading the energy value of the transition state (the highest energy point on the diffusion path) and the energy value of the initial state from the ST-LFP model, subtracting the two to obtain the diffusion barrier of lithium ions along the diffusion path; and calculating the diffusion coefficient of lithium ions by combining the Arrhenius equation.

[0017] According to some preferred embodiments of the present invention, the convergence criterion for the first geometry optimization is: energy change less than or equal to 1 × 10⁻⁶. -5 eV / atom, interatomic interaction force is less than or equal to 0.05 eV / Å.

[0018] According to some preferred embodiments of the present invention, the convergence criterion for the second geometry optimization is: energy change less than or equal to 1 × 10⁻⁶. -5 eV / atom, interatomic interaction force is less than or equal to 0.05 eV / Å.

[0019] According to some preferred embodiments of the present invention, the supercell expansion uses a supercell expansion algorithm to expand by 1 times along the lattice a-axis direction of the original unit cell model, by 2 times along the lattice b-axis direction, and by 1 time along the lattice c-axis direction.

[0020] According to some preferred embodiments of the present invention, in the spin polarization density functional calculation, the Monkhorst-Pack scheme is used for k-point sampling, the sampling interval along the reciprocal space is set to 3×3×7, and the plane wave cutoff energy is set to 650eV.

[0021] According to some preferred embodiments of the present invention, the S-LFP supercell model and / or X-LFP supercell model are output in the form of a standard crystal structure file.

[0022] The standard crystal structure file, such as POSCAR format and / or CIF format, contains complete information such as the supercell's lattice constant, atomic coordinates, and the types and quantities of atoms.

[0023] According to some preferred embodiments of the present invention, the doping atom is selected from C atoms and / or B atoms.

[0024] Based on the above methods, the present invention can further provide an apparatus for evaluating the influence of doping on the performance of lithium iron phosphate cathode materials, which includes a supercell expansion module for constructing the lithium iron phosphate supercell model, a doping construction module for constructing the doped LFP supercell model, an optimization calculation module for constructing the S-LFP supercell model and the X-LFP supercell model, and an evaluation and analysis module for evaluating the influence of doped atoms on the overall performance and / or lithium-ion diffusion performance of lithium iron phosphate through the X-LFP supercell model.

[0025] The present invention has the following beneficial effects: (1) This invention solves the technical limitations of traditional material preparation which is dominated by experimental trial and error. Based on first-principles simulation of density functional theory, it can evaluate the effect of doping on LFP performance without the need to prepare physical samples, which significantly shortens the research and development cycle, reduces the research and development cost, and greatly improves the efficiency of scheme optimization. (2) The entire technical process of the present invention does not involve the preparation, processing and testing of physical samples, and there is no consumption of chemical reagents or emission of chemical waste, thus avoiding environmental pollution from the source and conforming to the concept of green research and development; (3) This invention overcomes the one-sidedness of the single performance evaluation adopted by the prior art, and can realize the systematic calculation of key parameters such as doping formation energy, band gap, differential charge density, lithium ion diffusion barrier and diffusion coefficient, and simultaneously cover lattice stability, electronic conduction performance and lithium ion diffusion performance, and construct a multi-dimensional and all-round comprehensive evaluation system. (4) The first-principles calculation method adopted in this invention is compatible with any first-principles calculation platform that conforms to academic norms. It can be flexibly extended to the performance evaluation of multiple single-element or multi-element co-doped LFPs without reconstructing the core process, providing a standardized and reusable technical framework for the doping modification research and development of lithium iron phosphate cathode materials. (5) In some preferred embodiments, the present invention further defines all core calculation parameters (such as k-point grid, plane wave cutoff energy, energy / force convergence criterion, etc.), and combined with its standardized calculation steps, it can ensure that different R&D personnel obtain consistent evaluation results on different computing platforms, which meets the requirements of reproducibility in scientific research. (6) By testing the convergence of the core calculation parameters, this invention can significantly shorten the calculation time while ensuring the calculation accuracy, effectively improve the utilization rate of computing resources, and avoid the ineffective waste of computing time. (7) This invention can reveal the doping mechanism at the atomic level, obtain the quantitative relationship between doping parameters and material properties, and systematically and accurately evaluate the impact of doping on the performance of lithium iron phosphate cathode materials. Attached Figure Description

[0026] Figure 1 Comparison of crystal structures of the S-LFP supercell model (a), C-LFP supercell model (b), B-LFP supercell model (c), and CB-LFP supercell model (d) constructed for Example 1; Figure 2 Comparison of band structures of the S-LFP supercell model (a), C-LFP supercell model (b), B-LFP supercell model (c), and CB-LFP supercell model (d) constructed for Example 1; Figure 3 A comparison diagram of the density of states of the S-LFP supercell model (a), C-LFP supercell model (b), B-LFP supercell model (c), and CB-LFP supercell model (d) constructed for Example 1; Figure 4 A comparison of differential charge densities for the S-LFP supercell model (a), C-LFP supercell model (b), B-LFP supercell model (c), and CB-LFP supercell model (d) constructed for Example 1; Figure 5A comparative diagram of the lithium-ion diffusion path configurations of the S-LFP supercell model (a), C-LFP supercell model (b), B-LFP supercell model (c), and CB-LFP supercell model (d) constructed for Example 1; Figure 6 Comparison of diffusion barriers for the S-LFP supercell model (a), C-LFP supercell model (b), B-LFP supercell model (c), and CB-LFP supercell model (d) constructed for Example 1. Detailed Implementation

[0027] The technical solutions of the present invention will be further described below with reference to specific embodiments and examples. The embodiments described below are only some, not all, of the embodiments of the present invention. All other embodiments obtained by those skilled in the art based on the embodiments of the present invention without creative effort should fall within the scope of protection of the present invention.

[0028] All calculations in the following examples are based on first-principles density functional theory (DFT) (which obtains the ground state energy and electronic structure of multi-electron systems by solving the Kohn-Sham equations, predicting material properties at the atomic level without relying on empirical parameters).

[0029] The following embodiments use the Vienna First Principles Simulation Software (VASP) as the computing platform and follow a standardized parameter system, including: The PBE functional under the generalized gradient approximation (GGA) is used in the electron exchange correlation. The interaction between the ionic core and valence electrons is described using the projected added plane wave (PAW) method; The self-consistent iterative energy convergence accuracy is set to 1×10. -5 eV / atom; The atomic force convergence criterion is set to ≤0.05 eV / Å; All model-relaxed structure files are output in the VASP standard format, compatible with POSCAR / CONTCAR file specifications.

[0030] Example 1 The following steps were used to evaluate the performance effects of C doping, B doping, or C / B co-doping on lithium iron phosphate cathode materials: S1 constructs a stable supercell model for lithium iron phosphate (LFP), specifically including: S11 yields the original unit cell model of LFP, as follows: The LFP (Lithium Iron Phosphate) CIF file was retrieved and exported from the Inorganic Crystal Structure Database (COD). The CIF file was opened using crystallography software. Based on existing theoretical and experimental data, the completeness and rationality of the provided crystal structure parameters were determined. If the crystal structure parameters were complete and rational, the provided unit cell model was confirmed as the original unit cell model of LFP. This original unit cell model contains 4 Li atoms, 4 Fe atoms, 4 P atoms, and 16 O atoms, belongs to the orthorhombic crystal system, has the space group Pnma, and the lattice parameters are: a = 10.33 Å, b = 6.01 Å, c = 4.69 Å, α = β = γ = 90°. Here, a, b, and c are the lengths of the lattice a-axis, b-axis, and c-axis, respectively. The b-axis represents the direction of the one-dimensional diffusion channel of lithium ions in lithium iron phosphate. α, β, and γ represent the angles between the lattice b-axis and c-axis, the lattice a-axis and c-axis, and the lattice a-axis and b-axis, respectively. S12 performs supercell expansion on the original unit cell model to obtain the LFP supercell model, as follows: Based on the original unit cell model, a 1×2×1 (i.e., expanding by 1 time along the a-axis, 2 times along the b-axis, and 1 time along the c-axis) supercell model was constructed using the supercell expansion algorithm, resulting in the LFP supercell model. The model contains 8 Li atoms, 8 Fe atoms, 8 P atoms, and 32 O atoms, with lattice parameters of a=10.33Å, b=12.02Å, c=4.69Å, α=β=γ=90°. S13 performs a first geometric optimization on the LFP supercell model to obtain a structurally relaxed LFP supercell model, namely the stable LFP supercell model S-LFP, as follows: The LFP supercell model was imported into the Vienna First Principles Simulation Software (VASP), and geometric optimization was performed using the structure optimization module based on density functional theory. In the optimization calculation, the plane wave cutoff energy was set to 650 eV, the k-point sampling scheme was set to the Monkhorst-Pack scheme, and the sampling partitioning parameters along the reciprocal space were set to 3×3×7 (i.e., 3 sampling points along the a-axis, 3 sampling points along the b-axis, and 7 sampling points along the c-axis in the reciprocal space). Simultaneously, spin polarization calculations were enabled to perform geometric optimization of the LFP supercell model with synchronous relaxation of lattice parameters and atomic coordinates until the total system energy and interatomic forces met the convergence criteria, resulting in the optimized LFP supercell model, i.e., the stable LFP supercell model, as shown in the appendix. Figure 1 As shown in (a); Output a POSCAR format file containing complete lattice constants, atomic coordinates, and type information of the stable LFP supercell model; The stable LFP supercell model lattice parameters obtained in this embodiment are: a=10.39Å, b=12.12Å, c=4.74Å.

[0031] Based on the stable LFP supercell model S-LFP obtained in S1, S2 constructs C-doped, B-doped, and CB-co-doped LFP supercell models, specifically including: S21 determines the optimal doping sites for C and B in the LFP crystal structure by comparing the changes in the total system energy when C and B atoms replace different atomic sites, as follows: Based on the atomic occupancy characteristics of the LFP crystal structure, two potential doping methods were initially identified, including: C atoms and B atoms replacing P atoms in the crystal lattice, i.e., the P replacement method; and C atoms and B atoms replacing O atoms in the crystal lattice, i.e., the O replacement method. On the 1×2×1 LFP supercell model obtained in step S12, substitution models with P substitution (coordination number 4) and O substitution (coordination number 6) are constructed. The substitution models are geometrically optimized using the density functional theory-based method consistent with S13. After convergence, the total system energy of each substitution model is extracted from the calculation result file. The site corresponding to the substitution method with the lower total system energy is taken as the optimal doping site. In this embodiment, the test found that the energy of replacing an O atom with a C or B atom is higher than the energy of replacing a P atom, so the optimal doping site was determined to be the P atom site. S22 Based on the optimal doping sites, construct LFP supercell models with C-doped, B-doped, and CB-co-doped sites, as follows: Based on the optimal doping site determined in S21 as the P atom site, the stable LFP supercell model output in S13 (the S-LFP model below, as shown in the appendix) is obtained. Figure 1 (a) Supercell models with C-doped, B-doped, and CB-co-doped structures were constructed on the supercell, and the corresponding POSCAR files were output. The C-doped model was obtained by replacing one P atom in the S-LFP model with a C atom, as shown in the attached figure. Figure 1 As shown in (b); the B-doping model is obtained by replacing one P atom in the S-LFP model with a B atom, as shown in the appendix. Figure 1 (c) shows the CB co-doping model, which is obtained by replacing one P atom in the S-LFP model with C and B atoms respectively, as shown in the appendix. Figure 1 As shown in (d).

[0032] S3 performs a second geometric optimization on the C-doped, B-doped, and CB-co-doped LFP supercell models respectively to obtain structurally relaxed C-doped, B-doped, and CB-co-doped LFP supercell models, i.e., stable doping models, specifically including: The C-doped, B-doped, and CB-co-doped LFP supercell models obtained in S22 were imported into the Vienna First Principles Simulation Software (VASP) package, and geometric optimization was performed using the same calculation method as in S13. Specifically, the plane wave cutoff energy was set to 650 eV, the k-point sampling scheme was set to the Monkhorst-Pack scheme, the partitioning parameter along the reciprocal space was set to 3×3×7, and the energy convergence accuracy was ≤1×10⁻⁶. -5 eV / atom, force convergence criterion ≤0.05eV / Å, lattice parameters and atomic coordinates are simultaneously relaxed to obtain optimized LFP supercell models for C-doped, B-doped and CB-co-doped, i.e. stable doping models (denoted as C-LFP, B-LFP and CB-LFP supercell models, respectively). Output a POSCAR format file containing the complete lattice constants, total energy, and atomic coordinates of the C-LFP, B-LFP, and CB-LFP supercell models.

[0033] S4. Based on the aforementioned stable doping model, evaluate the impact of C doping, B doping, and CB co-doping on the overall performance of LFP, specifically including: S41 Based on the aforementioned stable doping model, the doping formation energy is calculated, and the effects of C doping, B doping, and CB co-doping on the stability of LFP are evaluated as follows: The doping formation energies of C-doped, B-doped, and CB-co-doped cells were calculated using the following computational model: in, E f It represents the doping formation energy, which is used to quantify the thermodynamic stability of a doped system. The smaller the doping formation energy, the higher the thermodynamic stability of the doped system and the easier it is to spontaneously form. The total energy of the doped system is represented by the value extracted from the calculation results of the stable doping model obtained through the second geometric optimization in step S3; where C doping corresponds to the C-LFP supercell model, B doping corresponds to the B-LFP supercell model, and CB co-doping corresponds to the CB-LFP supercell model. This represents the total energy of the pure LFP, and its value is extracted from the calculation results of the S-LFP supercell model obtained through the first geometric optimization in step S13. E P The reference state energy represents the doped atom (P atom in this embodiment), and the value is taken from the average chemical potential of P atom in the S-LFP supercell model. E C , EB These represent the reference state energies of C and B atoms, respectively, with values ​​taken from the ground state energies of C and B elemental crystals. n y Indicates the number of doped atoms in the system; n C , n B : These represent the doping amounts of C atoms and B atoms in the system, respectively; In this embodiment, for different doping systems, the parameters are... n y , n C , n B The possible values ​​are as follows: C-doped systems: n y =1 (replaces 1 P atom). n C =1, n B =0; B-doped system: n y =1 (replaces 1 P atom). n B =1, n C =0; CB co-doped system: n y =2 (replacing 2 P atoms) n C =1, n B =1; All the above energy parameters were obtained through first-principles density functional calculations and can be directly extracted from the geometric optimization result files of the corresponding models. The calculation parameters are consistent throughout the process, ensuring the reliability and comparability of the results.

[0034] The effects of C doping, B doping, and CB co-doping on the stability of LFP were evaluated based on the doping formation energy. The smaller the doping formation energy, the more stable and easier the corresponding doped system is to form. In this embodiment, the formation energy of C doping is calculated to be -1.20 eV, the formation energy of B doping is -1.65 eV, and the formation energy of CB co-doping is -3.02 eV, indicating that the CB co-doped system is more stable and easier to form. S42 Based on the aforementioned stable doping model, the band structure, density of states, and differential charge density of the doped systems obtained by C doping, B doping, and CB co-doping are calculated, and the effects of C doping, B doping, and CB co-doping on the electronic conductivity of LFP are evaluated as follows: S421 performs static self-consistent calculations on the stable doping model to obtain the ground-state self-consistent charge density of the doped system. Based on this ground-state self-consistent charge density, a high-symmetry point path within the Brillouin zone is selected (e.g., Γ→Z→T→Y→S→X→U→R→Γ, etc., the specific path is determined according to the symmetry of the Pnma space group of the LFP orthogonal crystal system). Along the selected high-symmetry point path, the electronic eigenvalues ​​corresponding to each k-point are calculated to obtain the energy-wave vector relationship curve, i.e., the band structure diagram. From the band structure diagram, the band gap value of the doped system is obtained by reading the energy difference between the valence band top and the conduction band bottom. Based on the size of the band gap value, the influence of C doping, B doping, and CB co-doping on the electronic conductivity of LFP is evaluated. Among them, the smaller the band gap value, the more significant the improvement in the electronic conductivity of the corresponding doped system. In this embodiment, the calculation results show that the band gap of the pure LFP based on the S-LFP supercell model is 3.67 eV, the band gap of the C-doped crystal based on the C-LFP supercell model is 1.68 eV, the band gap of the B-doped crystal based on the B-LFP supercell model is 1.46 eV, and the band gap of the CB co-doped crystal based on the CB-LFP supercell model is 1.42 eV, as shown in the attached figures. Figure 2 As shown in (a), 2(b), 2(c), and 2(d), the co-doped crystal has the smallest band gap, indicating that the energy required for electrons to transition from the valence band to the conduction band is the lowest after CB co-doping, the resistance to electron transition is the smallest, and the electronic conduction performance is more significantly improved. S422 uses the same self-consistent charge density as in step S421 (i.e., the obtained ground-state self-consistent charge density) to uniformly sample the Brillouin zone, count the number of electronic states in each energy range, and obtain the total density of states of the doped system and the partial density of states of each atom (Li, Fe, P, O, C, B). Through the partial density of states, the contribution of each atomic orbital to the total density of states is determined, and the orbital hybridization interaction between the doped atom and the matrix atom is analyzed. Based on the strength of the interaction, the effects of C doping, B doping, and CB co-doping on the electronic conduction performance of LFP are evaluated. Among them, the stronger the orbital hybridization interaction between the doped atom and the matrix atom, the greater the overlap of their electron clouds, which is more conducive to improving the electronic transport characteristics of the LFP system and thus improving the electronic conduction performance. In this embodiment, the density of states diagrams for undoped, C-doped, B-doped, and CB-co-doped states are shown in the attached figure. Figure 3As shown in (a), 3(b), 3(c), and 3(d), in the CB co-doped (3(d)), the 2p orbital state densities of C and B have obvious peaks in the energy range of -10 eV to -5 eV, which overlap to some extent with the 2p orbital state densities of O and 3p orbital state densities of P. This indicates that there is a strong orbital hybridization interaction between C and B atoms and O and P atoms, which is conducive to the formation of electronic conduction channels and significantly improves the electronic conduction performance of LFP. S423 Extract the total charge density of the doped system from the calculation results of the stable doping model (X-LFP supercell model) obtained through the second geometric optimization in step S3; keep the lattice parameters and atomic coordinates of the X-LFP supercell model unchanged, delete all doped atoms, construct a lithium iron phosphate framework model, calculate the total charge density of the model, and obtain the framework charge density; keep the lattice parameters and atomic coordinates of the X-LFP supercell model unchanged, retain only any one type of doped atom, construct an isolated doped atom model, calculate the total charge density of the model, and obtain the isolated doped atom charge density; subtract the framework charge density from the total charge density of the doped system, and then subtract the isolated doped atom charge density of all doped atoms to obtain the differential charge density; Based on the above implementation process, in this embodiment, the differential charge density of C doping or B doping is obtained by subtracting the framework charge density from the total charge density of the doping system, and then subtracting the isolated doped atom charge density of C atoms or the isolated doped atom charge density of B atoms. The differential charge density of CB co-doping is obtained by subtracting the framework charge density from the total charge density of the doping system, and then subtracting the isolated doped atom charge density of C atoms and the isolated doped atom charge density of B atoms. Based on the obtained differential charge density, a differential charge density distribution map can be further drawn to intuitively determine the direction and degree of charge transfer between the dopant atom and the matrix atom. The more complete the charge transfer, the stronger the electronic interaction between the dopant atom and the matrix atom, which is more conducive to carrier transport and the more significant the improvement in the electronic conduction performance of LFP. In this embodiment, the differential charge density distribution diagrams corresponding to no doping, C doping, B doping, and CB co-doping are attached. Figure 4 As shown in (a), 4(b), 4(c), and 4(d) (yellow electron clouds represent electron gain, and blue electron clouds represent electron loss), in CB co-doping (4(d)), the electron cloud around the C atom is blue, and the electron cloud on the C-O bond is yellow, indicating that the C atom lost electrons and transferred them to O to form C-O bonds. The electron cloud around the B atom is yellow, and the electron cloud of the neighboring Fe atom is blue, indicating that B gained electrons while Fe lost electrons, and the charge transfer was sufficient, further verifying that it significantly improved the electronic conduction performance of LFP.

[0035] S5. Based on the aforementioned stable doping model, the X-LFP supercell model, the effects of C doping, B doping, and CB co-doping on the lithium-ion diffusion performance of LFP are evaluated, specifically including: S51 Based on the crystallographic features of one-dimensional diffusion channels and the CI-NEB method, the diffusion path of lithium ions in a stable doping model is constructed, and the diffusion model is obtained as follows: Based on the crystallographic characteristics of the one-dimensional diffusion channel of lithium ions along the

[010] direction in the LFP crystal structure, specifically, in the stable doping model, a lithium ion within the supercell is selected as the migrating ion, and its adjacent lithium ion lattice site is set as a vacancy. This determines the initial state (IS, lithium ion at the original lattice site) and the final state (FS, lithium ion migrating to the adjacent vacancy site) of the lithium ion diffusion path. Using the climbing-type micro-motion elastic band (CI-NEB) method, four intermediate images are inserted between the initial and final states to construct the complete lithium ion diffusion path, obtaining the diffusion model of the corresponding system, as shown in the attached figure. Figure 5 As shown (where (a) is the pure LFP diffusion model, (b) is the C-LFP diffusion model, (c) is the B-LFP diffusion model, and (d) is the CB-LFP diffusion model); S52 performs a third geometric optimization on the diffusion model to obtain a diffusion model with structural relaxation and a minimum energy path, i.e., a stable diffusion model, as follows: To accurately simulate the diffusion process of lithium ions, a thorough structural optimization, namely the third geometric optimization, was performed on the diffusion model. The optimization employed the GGA-PBE functional and projected augmented plane wave (PAW) method used in S13 and S22. During optimization, the lattice constant was fixed, and only atomic coordinates were allowed to relax freely to ensure the stability and reliability of the diffusion path calculation. Periodic boundary conditions were used to simulate the true state of the material in an infinite crystal space. The optimization process was as follows: first, geometric optimization was performed on the initial and final states of the diffusion model to ensure that both the initial and final states were local minimum energy configurations; then, structural relaxation was performed on the complete diffusion path including the initial state, intermediate images, and final state to obtain the minimum energy path for lithium ion migration, ultimately resulting in a structurally fully relaxed stable diffusion model (ST-LFP). The convergence criterion for geometric optimization was set to an energy change less than or equal to 1 × 10⁻⁶. -4 eV / atom, interatomic interaction force less than or equal to 0.05 eV / Å; S53 Based on the ST-LFP model, the lithium-ion diffusion barrier and diffusion coefficient of the doped system are calculated, and the influence of doping on lithium-ion migration ability is evaluated as follows: From the ST-LFP model, the energy value of the highest energy point (i.e., the transition state) on the diffusion path is read and the energy value of the initial state is calculated. The difference between the two is used to obtain the diffusion barrier of lithium ions along the diffusion path. Using the Arrhenius equation, the diffusion coefficient of lithium ions at 300 K (room temperature) was calculated. The effects of C doping, B doping, and CB co-doping on lithium-ion migration ability are evaluated based on the diffusion coefficient. The larger the diffusion coefficient, the stronger the lithium-ion migration ability. In this embodiment, the calculation results show that the diffusion barrier of pure LFP is 0.51 eV, the diffusion barrier of C-doped LFP is 0.36 eV, the diffusion barrier of B-doped LFP is 0.63 eV, and the diffusion barrier of CB co-doped LFP is 0.41 eV, as shown in the attached figures. Figure 6 As shown in (a)-(d); the diffusion coefficient of pure LFP is D=4.3×10 -12 cm² / s, C-doped diffusion coefficient D = 1.55 × 10⁻⁶ cm² / s -9 cm² / s, B-doped diffusion coefficient D = 4.85 × 10⁻⁶ cm² / s -14 cm² / s, CB co-doped diffusion coefficient D = 2.3 × 10⁻⁶ cm² / s -10 cm² / s. Therefore, the lithium-ion migration ability is: C doping > CB co-doping > pure LFP > B doping.

[0036] Based on the multi-dimensional calculation results of the above band structure, density of states, differential charge density, lithium-ion diffusion barrier, and diffusion coefficient, this embodiment can achieve targeted screening of doped systems according to actual performance requirements, such as: When the doping system focuses more on improving electronic conductivity, the CB co-doping system is selected. When the doping system focuses more on enhancing lithium-ion diffusion kinetics, the C doping system is selected to achieve a precise match between performance requirements and modification schemes. For general scenarios requiring synergistic optimization of electronic conduction and ion diffusion performance, the CB co-doped system is selected to achieve a comprehensive balance between the two types of transport performance (although its lithium-ion diffusion performance is slightly lower than that of the C-doped system, its electronic conduction advantage is outstanding, which can effectively avoid the shortcomings of single transport performance).

[0037] Meanwhile, the orbital hybridization characteristics and lithium-ion diffusion barrier rules obtained in this embodiment can be used to specifically control the doping ratio and process parameters of carbon and boron sources, so as to enhance the electron transport capability of the CB co-doped system and the ion diffusion capability of the C doped system, and give full play to the performance advantages of each doping system. Furthermore, based on the performance differences of different doping systems obtained in this embodiment, it is possible to customize the selection of materials according to specific scenarios. For example, the C-doped system is suitable for high-power fast-charging power batteries, while the CB co-doped system is more suitable for high-capacity long-cycle energy storage batteries.

[0038] It should be noted that the above descriptions are merely preferred embodiments of the present invention and should not limit the scope of protection of the technical solutions of the present invention. Any modifications made to the technical solutions described in the foregoing embodiments, or equivalent substitutions of technical features, by those skilled in the art within the spirit and principles of the present invention, should be included within the scope of protection of the present invention.

Claims

1. A method for evaluating the effect of doping on the performance of lithium iron phosphate cathode materials, characterized in that, It includes: S1 extends the original cell model of lithium iron phosphate into a supercell model, and performs the first geometric optimization on it to obtain a stable lithium iron phosphate supercell model, namely the S-LFP supercell model. S2 Based on the S-LFP supercell model, a doped LFP supercell model is constructed, consisting of doped atoms, including: S21 determines the optimal doping site of the doped atom in the S-LFP supercell model by comparing the changes in system energy; S22 constructs a doped LFP supercell model by replacing one original atom at the optimal doping site in the S-LFP supercell model with one doped atom. S3 performs a second geometric optimization on the doped LFP supercell model to obtain a stable doped LFP supercell model, namely the X-LFP supercell model. S4. Based on the X-LFP supercell model, evaluate the impact of doping atoms on the overall performance and / or lithium-ion diffusion performance of lithium iron phosphate. The first and second geometric optimizations are based on first-principles density functional theory and are achieved by spin-polarized density functional calculations using the PBE functional in the generalized gradient approximation and the projected plane wave method. During the calculation, the lattice parameters and atomic coordinates relax simultaneously until the energy change and interatomic interaction forces meet the convergence criteria.

2. The method according to claim 1, characterized in that, S4 includes: S41. Based on the X-LFP supercell model, the doping formation energy is calculated, and the influence of dopant atoms on the stability of lithium iron phosphate is evaluated using the doping formation energy; wherein, the doping formation energy is calculated using the following calculation model: in, Indicates the doping formation energy. This represents the total energy of the doped system obtained from the X-LFP supercell model. This represents the total energy of lithium iron phosphate obtained according to the S-LFP supercell model. E y , E z Let represent the reference state energies of the dopant atom and the doped atom, respectively. E y The values ​​are derived from the ground-state energy of a single-element crystal doped with y atoms. E z The numerical values ​​are derived from the average chemical potential of the doped atom z in the S-LFP supercell model. n y , n z These represent the number of doping atoms and the number of atoms that were doped, respectively. And / or, S42 Based on the X-LFP supercell model, calculate one or more of the band gap, density of states, and differential charge density of the doped system, and evaluate the influence of doped atoms on the electronic conductivity of lithium iron phosphate based on the calculation results. The calculation of the bandgap value includes: Static self-consistent calculations are performed on the X-LFP supercell model to obtain the ground-state self-consistent charge density of the doped system. Based on this ground-state self-consistent charge density, a high-symmetry point path within the Brillouin zone is selected. Along the selected high-symmetry point path, the electronic eigenvalues ​​corresponding to each sampling point are calculated to obtain the energy-wave vector relationship curve, i.e., the band structure diagram. Through the band structure diagram, the energy difference between the valence band top and the conduction band bottom is obtained, i.e., the band gap value. The calculation of the density of states includes: The Brillouin zone is uniformly sampled using the ground-state self-consistent charge density, and the number of electronic states in each sampling interval is counted to obtain the total density of states of the doped system and the partial density of states of each atom. The calculation of the differential charge density includes: The total charge density of the doped system is extracted from the calculation results of the X-LFP supercell model obtained through the second geometric optimization; keeping the lattice parameters and atomic coordinates of the X-LFP supercell model unchanged, all doped atoms are deleted to construct a lithium iron phosphate framework model, and the total charge density of the model is calculated to obtain the framework charge density; keeping the lattice parameters and atomic coordinates of the X-LFP supercell model unchanged, only one type of doped atom is retained to construct an isolated doped atom model, and its total charge density is calculated to obtain the isolated doped atom charge density; the total charge density of the doped system is subtracted from the framework charge density, and then the isolated doped atom charge density of all doped atoms is subtracted to obtain the differential charge density.

3. The method according to claim 1, characterized in that, in, The evaluation of the influence of doped atoms on lithium-ion diffusion performance includes: S51 Based on the crystallographic features of one-dimensional diffusion channels and the CI-NEB method, the diffusion path of lithium ions in the X-LFP supercell model is constructed to obtain the diffusion model; S52 performs a third geometric optimization on the diffusion model to obtain a stable diffusion model, namely the ST-LFP model; S53 Based on the ST-LFP model, calculate the lithium-ion diffusion barrier and / or diffusion coefficient of the doped system, and evaluate the effect of doping on lithium-ion migration ability. The third geometric optimization is achieved by performing spin polarization density functional calculations using the PBE functional and projected plane wave method in the generalized gradient approximation. During the calculation, lattice parameters and atomic coordinates relax simultaneously until the energy change and interatomic interaction forces meet the convergence criteria.

4. The method according to claim 3, characterized in that, The convergence criterion for the third geometric optimization is: energy change less than or equal to 1 × 10⁻⁶. -4 eV / atom, interatomic interaction force is less than or equal to 0.05 eV / Å.

5. The method according to claim 3, characterized in that, S51 includes: selecting a lithium ion in the X-LFP supercell model as the migration target, its state at the original lattice site is the initial state, setting an adjacent lithium ion lattice site as a vacancy, and the state of the lithium ion migrating to the vacancy site is the final state. Using the CI-NEB method, four intermediate states are inserted between the initial state and the final state to construct a diffusion path and obtain a diffusion model. S52 includes: performing the third geometric optimization on the initial state and the final state of the diffusion model to ensure that the initial state and the final state are both local minimum energy configurations to obtain the ST-LFP model.

6. The method according to claim 5, characterized in that, The calculation of the lithium-ion diffusion barrier and / or diffusion coefficient includes: reading the energy value of the transition state (the highest energy point on the diffusion path) and the energy value of the initial state from the ST-LFP model, subtracting the two to obtain the diffusion barrier of lithium ions along the diffusion path; and calculating the diffusion coefficient of lithium ions by combining the Arrhenius equation.

7. The method according to claim 1, characterized in that, The convergence criterion for the first geometric optimization is: energy change less than or equal to 1 × 10⁻⁶. -5 eV / atom, interatomic interaction force less than or equal to 0.05 eV / Å; and / or, the convergence criterion for the second geometry optimization is: energy change less than or equal to 1×10 -5 eV / atom, interatomic interaction force is less than or equal to 0.05 eV / Å.

8. The method according to claim 1, characterized in that, in, The supercell expansion uses a supercell expansion algorithm, expanding by a factor of 1 along the lattice a-axis of the original unit cell model, by a factor of 2 along the lattice b-axis, and by a factor of 1 along the lattice c-axis; and / or, in the spin polarization density functional calculation, the Monkhorst-Pack scheme is used for k-point sampling, the sampling segmentation parameter along the reciprocal space is set to 3×3×7, and the plane wave cutoff energy is set to 650eV; and / or, the S-LFP supercell model and / or X-LFP supercell model are output in the form of a standard crystal structure file.

9. The method according to any one of claims 1-8, characterized in that, The doping atoms are selected from C atoms and / or B atoms.

10. An apparatus for evaluating the effect of doping on the performance of lithium iron phosphate cathode materials according to any one of claims 1-9, comprising a supercell expansion module for constructing the lithium iron phosphate supercell model, a doping construction module for constructing the doped LFP supercell model, an optimization calculation module for constructing the S-LFP supercell model and the X-LFP supercell model, and an evaluation and analysis module for evaluating the effect of doped atoms on the overall performance and / or lithium-ion diffusion performance of lithium iron phosphate through the X-LFP supercell model.

Images