A method for screening diatomic catalysts for lithium-sulfur batteries and application thereof
By constructing a new descriptor Ddf and combining first-principles calculations and machine learning methods, high-performance biatomic catalysts for lithium-sulfur batteries were screened. This solved the problem of insufficient accuracy of traditional descriptors, achieved efficient and accurate catalyst screening, and improved the performance of lithium-sulfur batteries.
Patent Information
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- SICHUAN UNIV
- Filing Date
- 2026-03-10
- Publication Date
- 2026-06-12
AI Technical Summary
Existing technologies struggle to efficiently screen high-performance biatomic catalysts for lithium-sulfur batteries. Traditional descriptors, such as d-band centers, exhibit limited linear correlation with adsorption energy, resulting in insufficient accuracy in predicting material properties.
A novel descriptor Ddf is constructed using first-principles calculations and machine learning methods. High-performance diatomic catalysts are screened through density functional theory calculations and genetic algorithm optimization. Furthermore, a high-precision adsorption energy descriptor Ddf is constructed by combining density functional theory and machine learning methods to screen diatomic catalysts with stability and high catalytic activity.
This approach enables efficient and accurate catalyst screening, reduces computational costs and time, improves prediction accuracy, and identifies the Os2N8 catalyst as significantly enhancing the rate performance and cycle stability of lithium-sulfur batteries.
Smart Images
Figure CN122201498A_ABST
Abstract
Description
Technical Field
[0001] This invention belongs to the fields of lithium-sulfur battery technology and materials simulation and calculation technology, specifically relating to a method for screening biatomic catalysts for lithium-sulfur batteries. Background Technology
[0002] Lithium-sulfur batteries are considered one of the most promising next-generation high-energy-density storage systems due to their extremely high theoretical specific capacity (~2600 Wh / kg), low cost, and environmental friendliness. However, in addition to the poor conductivity of the cathode active material, the commercialization of lithium-sulfur batteries has long been limited by two key issues: first, the "shuttle effect" of lithium polysulfides, which leads to the loss of active material and a decrease in coulombic efficiency; and second, the slow kinetics of sulfur redox reactions, resulting in poor rate performance and rapid capacity decay.
[0003] To overcome the aforementioned challenges, the development of multifunctional cathode materials has become a research hotspot. Ideal materials should possess high conductivity, strong lithium polysulfide anchoring ability, and high catalytic activity to effectively adsorb and promote the conversion of polysulfides. As an extension of single-atom catalysts, diatomic catalysts, due to their more complex structure, more flexible active sites, tunable electronic structure, and near-100% atom utilization, show great potential in suppressing the shuttle effect and accelerating polysulfide conversion.
[0004] Traditionally, screening high-performance catalysts through trial and error or density functional theory calculations is inefficient and costly. Therefore, establishing key performance descriptors for materials and rapidly predicting their properties is crucial for high-throughput screening and prediction of new materials. In existing technologies, d-band centers are widely used descriptors, but their linear correlation with adsorption energy is limited. While the energy difference between metal d-band centers and sulfur p-band centers improves prediction accuracy to some extent, as a single-parameter descriptor, it still cannot fully reflect the complexity of electrochemical systems, leading to significant biases in some systems. Therefore, there is an urgent need to develop novel descriptors that can more accurately quantify the impact of d-orbital characteristics on adsorption and catalytic activity. Summary of the Invention
[0005] In view of this, the present invention provides a method for screening biatomic catalysts for lithium-sulfur batteries based on first-principles calculations and machine learning, so as to quickly and accurately screen high-performance biatomic catalysts that can effectively anchor polysulfides and catalyze their conversion.
[0006] To achieve the above objectives, the present invention adopts the following technical solution: A method for screening diatomic catalysts for lithium-sulfur batteries, the method comprising the following steps: S1. Based on density functional theory, two diatomic catalyst models with configurations of TM2N8 and TM2N6 are constructed, where TM represents a transition metal element; S2. Optimize the structure and evaluate the stability of the two-configuration diatomic catalyst models, and screen out the catalyst models with stability; the stability includes thermodynamic stability and kinetic stability. S3. For all the selected stable diatomic catalyst models, density functional theory calculations are used to obtain their electronic structure information and adsorption energy for sulfur-containing molecules. The electronic structure information includes the d-orbital state density; S4. Extract d-orbital related characteristic parameters from the d-orbital density of states obtained in step S3, and perform correlation analysis with the obtained adsorption energy of sulfur-containing molecules to screen out characteristic parameters that are strongly correlated with adsorption energy. The d-orbital related characteristic parameters include the d-orbital density of states integral area, d-orbital energy span, d-band center, energy difference between d-orbital and sulfur p-orbital, the highest intensity of d-orbital and its corresponding energy value, the highest energy level of d-orbital, the lowest energy level of d-orbital, the density of states integral area of occupied d-orbitals, and the ratio of the density of states integral area of occupied to non-occupied d-orbitals. S5. For the selected characteristic parameters that are strongly correlated with adsorption energy, a genetic algorithm is used to optimize the coefficients in a high-dimensional parameter space to construct the adsorption energy descriptor D. df Descriptor D df The details are as follows; D df =exp(λ·A t / W d ) Among them, A t W is the integral area of the density of states of d orbitals. d Let λ be the energy span of the d orbital, and λ be a positive constant optimized by the genetic algorithm. S6. Using density functional theory, calculate the reaction free energy change of sulfur-containing compound reduction reactions on the surface of all stable diatomic catalyst models, and obtain the thermodynamic energy barrier of sulfur reduction reactions. S7. The adsorption energy descriptor D constructed according to step S5. df Based on the thermodynamic energy barrier of the sulfur reduction reaction obtained in step S6, a linear relationship between the thermodynamic energy barrier and the adsorption energy difference is constructed, and the bimetallic atom catalyst with the best comprehensive performance is screened based on the theoretical calculation results.
[0007] In some specific embodiments, preferably, the substrate of the diatomic catalyst model in step S1 is a two-dimensional heterostructure of boron nitride and graphene; wherein, bimetallic atoms are doped in the boron nitride portion of the structure; The transition metal elements include all transition metal elements in the fourth, fifth, and sixth periods, excluding Cd, Hg, and lanthanides.
[0008] In some specific embodiments, preferably, the structural optimization in step S2 involves: performing structural optimization on all catalyst models using first-principles calculations; The thermodynamic stability is indicated by the negative formation energy calculated using density functional theory for all catalyst models, suggesting that the structure is thermodynamically stable. The kinetic stability: According to ab initio molecular dynamics simulations, no significant distortion or bond breakage occurred in the atomic structure after the simulation, indicating that the structure has good kinetic stability.
[0009] Furthermore, the transition metals in the stable catalyst models after screening are V, Cr, Mn, Fe, Co, Ni, Cu, Nb, Mo, Tc, Ru, Rh, Pd, Ta, W, Re, Os, Ir, and Pt.
[0010] Furthermore, the electronic structure information mentioned in step S3 also includes the non-degenerate d-orbit coupled Hamiltonian population of the TM2N6 configuration; Calculations using density functional theory also include obtaining the TM d–S p orbital density of states and orbital coupled Hamiltonian configuration of the adsorption configuration. The sulfur-containing molecules include S8, Li2S4, Li2S2, and Li2S.
[0011] In some specific embodiments, preferably, the correlation analysis in step S4 is Pearson correlation analysis.
[0012] To ensure that the obtained descriptor D df To achieve predictive accuracy, step S5 also includes using the adsorption energies of S8, Li2S4, Li2S2, and Li2S molecules to modify the adsorption energy descriptor D. df Perform linear fitting to verify its accuracy.
[0013] In order to understand A t / W d How adsorption is affected, and verifying its accuracy also includes: constructing a descriptor A for the integral value of the TM d–S p orbital coupled Hamiltonian layout based on the selected feature parameters. t / W d 0.5 ; Among them, A t W is the integral area of the density of states of the d orbitals. d For d orbital energy span; Descriptor A t / W d 0.5Construction method and descriptor D df The construction method is the same.
[0014] Furthermore, regarding the acquired descriptor A t / W d 0.5 It also includes: statistically averaging the Li–S bond length based on the adsorption configuration, and constructing a Li–S bond length-A… t / W d 0.5 The linear relationship is used to predict the ease of decomposition.
[0015] In some specific embodiments, preferably, the thermodynamic energy barrier of the sulfur reduction reaction in step S6 is obtained by calculating the reaction free energy change of Li2S4 on the catalyst surface to Li2S2 and further to Li2S using density functional theory.
[0016] The diatomic catalyst obtained by the above screening method has an Os2N8 configuration, that is, two osmium atoms are anchored on a two-dimensional heterostructure substrate of boron nitride and graphene, and each osmium atom is coordinated with four nitrogen atoms.
[0017] A lithium-sulfur battery cathode material comprising the aforementioned diatomic catalyst.
[0018] Compared with the prior art, the beneficial effects of the present invention are as follows: (1) A highly efficient and accurate catalyst screening method is provided: This invention combines density functional theory calculation and machine learning. By constructing a high-precision descriptor, it can achieve high-throughput screening of diatomic catalysts, significantly reduce computational costs and time, and accelerate the discovery of high-performance materials.
[0019] (2) A novel electronic structure descriptor D was proposed. df This descriptor is constructed based on the integral area of the d-orbital density of states and the d-orbital energy span. It has higher prediction accuracy than traditional d-band center descriptors. The correlation coefficient R² for the adsorption energy of S8 and lithium polysulfides can reach 0.76–0.97, which is 28%–44% higher than that of traditional d-band center descriptors.
[0020] (3) The catalytic mechanism was revealed in depth: Through partial density of states and crystal orbital Hamiltonian population analysis, the different orbital hybridization modes (dz2–pz and dz2–px) between the two configurations TM2N8 and TM2N6 and sulfur atoms were clarified, as well as the regulatory effect of d–d orbital coupling on electronic structure, providing a solid physical basis for the construction of descriptors.
[0021] (4) Screening out high-performance catalyst Os2N8: This catalyst is thermodynamically stable, can effectively adsorb sulfur-containing molecules, and exhibits the lowest energy barrier (only 2.96 eV) for key sulfur reduction steps, which is expected to significantly improve the rate performance and cycle stability of lithium-sulfur batteries. Attached Figure Description
[0022] Figure 1 This is a structural stability analysis diagram of the diatomic catalyst in Example 1 of the present invention. Wherein; (a) is a schematic diagram of the atomic structure of the TM2N8 and TM2N6 configurations; (b) Formation energies of TM2N8 and TM2N6 structures formed by different transition metals; (c) is a graph showing the energy and temperature versus time in a 10 ps ab initio molecular dynamics simulation of Cu2N8 at 300 K; (d) shows the energy and temperature versus time in a 10 ps ab initio molecular dynamics simulation of Pd2N6 at 300 K.
[0023] Figure 2 This is a d-orbital density of states diagram of 19 TM2N8 structures in Embodiment 1 of the present invention.
[0024] Figure 3 This is a d-orbital density of states diagram of 19 TM2N6 structures in Embodiment 1 of the present invention.
[0025] Figure 4 The adsorption energies of the TM2N8 and TM2N6 structures for S8 molecules in Example 1 of this invention are given.
[0026] Figure 5 The adsorption energies of the TM2N8 and TM2N6 structures for Li2S4 molecules in Example 1 of this invention are given.
[0027] Figure 6 The adsorption energies of the TM2N8 and TM2N6 structures for Li2S2 molecules in Example 1 of this invention are given.
[0028] Figure 7 The adsorption energies of the TM2N8 and TM2N6 structures for Li2S molecules in Example 1 of this invention are given.
[0029] Figure 8 In the TM2N8 and S8 adsorption configurations of this invention, the TM d orbital and Sp orbital are coupled to the ICOHP value.
[0030] Figure 9 The ICOHP value represents the coupling of the TM d orbital and the Sp orbital in the TM2N6 and S8 adsorption configurations in Example 1 of this invention.
[0031] Figure 10 This diagram illustrates the relationship between descriptors and performance obtained through machine learning in Embodiment 1 of the present invention, as well as a schematic diagram of the adsorption enhancement mechanism. Wherein: (a) shows the adsorption energy of the TM2N8 structure for S8 and the adsorption energy of D. df Linear relationship diagram; (b) shows the adsorption energy of the TM2N8 structure for Li2S and the relationship between D df Linear relationship diagram; (c) shows the adsorption energy of the TM2N6 structure for S8 and D. df Linear relationship diagram; (d) represents the adsorption energy of the TM2N6 structure for Li2S and the relationship between D and df Linear relationship diagram; (e) shows the ICOH values and At / W values for the coupling of TM d and Sp orbitals in the Li2S adsorption configurations of TM2N8 and TM2N6 structures. d 0.5 Linear relationship curve; (f) Schematic diagram of the effect of metal d orbital shape on adsorption strength.
[0032] Figure 11 This demonstrates the relationship between the Li–S bond length and A in Li2S adsorbed on the surface of the catalyst structure in Example 1 of the present invention. t / W d 0.5 Linear relationship graph and graph showing the relationship between the sulfur reduction thermodynamic energy barrier and the adsorption energy difference between Li₂S₂ and Li₂S. Among them: (a) The Li–S bond length and At / W ratio in Li2S on the TM2N8 structure d 0.5 Linear relationship diagram; (b) The Li–S bond length and At / W ratio in Li2S on the TM2N6 structure d 0.5 Linear relationship diagram; (c) is a graph showing the relationship between the sulfur reduction thermodynamic energy barrier on the TM2N8 structure and the difference in adsorption energy between Li2S2 and Li2S. (d) is a graph showing the relationship between the sulfur reduction thermodynamic energy barrier on the TM2N6 structure and the difference in adsorption energy between Li2S2 and Li2S.
[0033] Figure 12 This is a flowchart of the technical solution of Embodiment 1 of the present invention. Detailed Implementation
[0034] The present invention will be further described in detail below with reference to specific embodiments, so that those skilled in the art can more clearly understand the present invention. Unless otherwise specified, the technical means used in the following embodiments are all conventional means well known to those skilled in the art, and all reagents and consumables are commercially available products.
[0035] Example 1 This embodiment provides a method for screening biatom catalysts for lithium-sulfur batteries, specifically including the following steps: 1. Construction and stability assessment of a library of two-atom catalyst models 1.1 Model Construction Using two-dimensional planar heterostructures of BN and Gr as substrates, two metal diatomic catalyst models with different coordination configurations were constructed (see...). Figure 1 a) refers to the TM2N8 configuration (each transition metal atom is coordinated to four N atoms) and the TM2N6 configuration (each transition metal atom is coordinated to three N atoms, and there is a bonding interaction between the two metal atoms). The transition metals TM include all transition metal elements in the fourth, fifth, and sixth periods, except for the highly biotoxic Cd and Hg elements and the lanthanides.
[0036] 1.2 Structural Optimization All constructed TM2N8 and TM2N6 structures were optimized using first-principles calculations. The software used was the Vienna Ab-initio Simulation Package (VASP). The key optimization parameters were ENCUT=520 (eV), EDIFF=1E-5 (eV), and EDIFFG= 0.02 (eV / Å), IVDW=12. The high-symmetry k-point scattering density in the Brillouin zone is 0.04×2π / Å. The optimized geometry shows that the model containing group IIIB / IVB transition metals (Sc, Ti, Y, Zr, Hf) exhibits significant structural distortion along the z-axis, indicating poor stability, and therefore was excluded in subsequent studies. In the subsequent stability prediction, 22 transition metal elements were considered, including V, Cr, Mn, Fe, Co, Ni, Cu, Zn, Nb, Mo, Tc, Ru, Rh, Pd, Ag, Ta, W, Re, Os, Ir, and Pt.
[0037] 1.3 Stability Assessment Thermodynamic stability: The formation energy of all catalyst models corresponding to two configurations for 22 metallic elements was calculated using density functional theory. The formula for calculating the formation energy is: E f =E (defect) -E pure -ΣΔn x μ x in, E( defect) and E pure These represent the total energy of a substrate with defects and a intact substrate, respectively. Δn x and μ x These represent the number of atoms added or removed and their chemical potentials, respectively. The chemical potential of an atom is considered as the energy of an atom in a vacuum, and is calculated by placing a single atom in an orthorhombic unit cell with x, y, and z dimensions greater than 10 Å.
[0038] The formation energy calculation results show (see) Figure 1 (b) Except for Zn, Ag, and Au-based catalysts, the formation energies of all other TM2N8 and TM2N6 configurations are negative, indicating that these structures are thermodynamically stable. For the same transition metal, the formation energy of the TM2N6 configuration is slightly lower than that of the TM2N8 configuration, indicating that the former is more stable.
[0039] Kinetic stability: The Cu₂N₈ and Pd₂N₆ structures, which have the highest formation energy among thermodynamically stable structures, were selected as representatives for ab initio molecular dynamics simulations (see...). Figure 1 (c, d). Simulation was performed for 10 ps in the NVT ensemble at 300 K. The results show that the energy and temperature of the system quickly reached equilibrium after the simulation started, and fluctuated slightly within the range of 1 eV and 50 °C, respectively. Moreover, no obvious distortion or bond breakage occurred in the atomic structure after the simulation, confirming that this type of catalyst has good kinetic stability.
[0040] 2. Electronic Structure Analysis To investigate the effects of different configurations and different transition metals on the electronic structure, the density of states for all stable structural models was calculated (see...). Figure 2 and Figure 3 The d-orbit coupling Hamiltonian population of the TM2N6 structure and the TM2N6 structure.
[0041] 2.1 TM–TM Interactions The charge density difference plot shows that in the TM2N6 configuration, a TM–TM bond is formed between the two TM atoms; while in the TM2N8 configuration, there is no direct bonding between the TM atoms.
[0042] Further analysis of the partial wave density of states and d-orbital coupled Hamiltonian population of TM atoms in TM2N6 revealed strong hybridization of the d orbitals between TM and TM atoms, primarily manifested as d-orbital hybridization. xy and d x2-y2 The contribution of the orbit also exists in d. xz -d yz d z2 -d z2Multiple coupling mechanisms are involved. This d–d orbital coupling leads to a decrease in d orbital energy level degeneracy, an increase in peak positions, and a reconstruction of the electronic structure, which is the electronic origin of TM2N6's unique catalytic activity.
[0043] 3. Calculation of adsorption energy of sulfur-containing molecules Density functional theory calculations were used to evaluate the adsorption capacity of different configuration diatomic catalysts for sulfur-containing molecules, and the orbital interaction mechanism was revealed at the electronic structure level.
[0044] 3.1 Adsorption Energy Calculation Method To evaluate the anchoring ability of diatomic catalysts for polysulfides, S8 molecules and three representative lithium polysulfides—Li2S, Li2S2, and Li2S4—were selected as adsorbates, and their adsorption energies on the surfaces of TM2N8 and TM2N6 catalysts were calculated. Adsorption energy E ad The calculation formula is as follows: E ad= E substrate···LiPS E substrate E LiPS Among them, E substrate···LiPS E represents the total energy of the catalyst substrate and the sulfur-containing molecule adsorption system. substrate E represents the energy of the catalyst substrate. LiPS E represents the energy required to isolate a sulfur-containing molecule. ad A negative value indicates that the adsorption process is exothermic, and the more negative the value, the greater the adsorption strength.
[0045] 3.2 Adsorption Energy Results The calculation results show (see) Figure 4-7 The adsorption energy range of the TM2N6 configuration for S8 and lithium polysulfides is as follows: 1.16 eV to The adsorption energy range of the TM2N8 configuration is 8.57 eV. 0.75 eV to The value of 4.74 eV indicates that both configurations exhibit good adsorption capacity. For the same transition metal element, the TM₂N₆ configuration generally shows stronger adsorption capacity than the TM₂N₈ configuration. This is due to the stronger synergistic adsorption effect generated by the d–d orbital coupling between the two transition metal atoms in the TM₂N₆ configuration. This enhancement effect is most significant in group VB-VIIIB transition metals, which have high d orbital occupancy and electron delocalization, enabling them to form two chemical bonds with sulfur-containing molecules and maximize adsorption interactions.
[0046] 3.3 Adsorption Energy Variation Pattern Further analysis revealed a periodic pattern in the adsorption energy with respect to the type of transition metal. Within the same period, the adsorption energy gradually decreases as the reducing power of the metal weakens; within the same subgroup, the adsorption capacity generally increases with increasing period number. For example, in the system where S8 is adsorbed by the TM2N8 configuration, the adsorption energy of the 3d transition metal catalyst is... 1.13 eV to 2.27 eV, 4d transition metal reinforcement to 1.12 eV to 3.80 eV, 5d transition metal further enhanced to 1.11 eV to 4.74 eV. This periodic trend reflects the stronger delocalization of the d orbitals in 4d / 5d metals, which is conducive to the formation of stronger TM–S bonds.
[0047] 4. Analysis of the interaction between catalysts and polysulfide molecular orbitals To elucidate the electronic structure origin of the differences in adsorption capacity, the partial density of states and crystal orbital Hamiltonian population of the TM–S bond were analyzed using the Li2S adsorption system as an example.
[0048] 4.1 Orbital Hybridization Characteristics of the TM2N8 Configuration With V2N 8… Taking the Li₂S system as an example, density of states analysis shows that below the Fermi level, approximately At 3.5 eV, the d of the V atom z2 orbital and p of S atom z The orbitals exhibit significant resonance peak overlap, indicating the formation of strong σ-type bonding interactions. Simultaneously, the d-axis of the V atom... yz orbital and p of S atom y orbit, d xz orbital and p of S atom x The orbitals also exhibit varying degrees of overlap, forming π-type bonding interactions. Systematic orbital coupling Hamiltonian population analysis of the TM2N8-adsorbed Li2S configuration shows (see...) Figure 8 ), the formation of TM-S bonds is mainly due to d z2 -p z d yz -p y and d xz -p x The three orbitals contribute, where d z2 -p z σ bonds play a dominant role in adsorption strength, while d bonds... yz -p y and d xz -p x π bonds provide additional stabilization energy.
[0049] 4.2 Orbital Hybridization Characteristics of the TM2N6 Configuration With V2N 6… Taking the Li₂S system as an example, the partial density of states analysis shows that below the Fermi level, approximately... At 4.1 eV, the d of the V atom z2 The orbitals are mainly related to the p orbitals of the S atom. x Orbital coupling occurs, forming half-σ bonds. This is due to the bonding between bimetallic atoms in the TM2N6 configuration and the resulting geometrical differences, leading to changes in adsorption sites and orbital orientations. Orbital coupling Hamiltonian population analysis shows (see...) Figure 9 In the TM2N6 configuration, the TM–S bond formation is based on d… z2 -p x Semi-σ bonds are dominant, while d xz -p z and d z2 -p z Orbital pairs also contribute to this.
[0050] 4.3 Relationship between the number of bonding orbitals and bond strength Further analysis of the relationship between the number of bonding dp orbital pairs in the TM-S bond and the coupled Hamiltonian population integral revealed a positive linear correlation, indicating that bond strength is determined by the number of orbital pairs involved in bonding. Notably, 4d and 5d transition metals, due to their larger atomic radii, exhibit greater spatial extension of their d orbitals, enabling them to generate a larger overlap integral with the 3p orbitals of the S atom, thus forming more bonding orbital pairs. This explains why 4d / 5d transition metals possess stronger bonding ability and catalytic activity compared to 3d transition metals.
[0051] 5. Adsorption energy descriptor D based on machine learning df Construction To rapidly predict the adsorption capacity of diatomic catalysts for polysulfides, we developed a novel descriptor by combining first-principles calculations with machine learning methods.
[0052] 5.1 Feature Extraction and Correlation Analysis Twelve characteristic parameters were extracted from the total d-orbital density of states data of all stable catalyst models. The linear correlation between these characteristics and the adsorption energies of S8, Li2S4, Li2S2, and Li2S was analyzed using Pearson correlation coefficient. The results show that among all characteristics, the integral area of the d-orbital density of states (A0) is the highest. t ) and d orbital energy span (W d The correlation between adsorption energy and the adsorption energy is the highest. However, the correlation coefficients between the traditional d-band center and the d–p-band center difference and the adsorption energy are relatively low.
[0053] 5.2 Descriptor Construction Select A, which has the highest correlation with adsorption energy. t and W d As a fundamental feature, a new adsorption energy descriptor D was constructed by optimizing the coefficients in a high-dimensional parameter space using a genetic algorithm. df =exp(λ·A t / W d ), where λ is a positive constant optimized by a genetic algorithm.
[0054] 5.3 Descriptor Performance Evaluation (a) D df Linear fitting was performed on the adsorption energies of S8, Li2S4, Li2S2, and Li2S adsorbed by the TM2N8 and TM2N6 configurations, respectively (see [link to documentation]). Figure 10 a–d).
[0055] (b) The results show that D df The adsorption energies for all sulfur species showed a very strong linear correlation. For the TM2N8 configuration, the predicted correlation coefficient R0 was [value missing]. 2 Between 0.76 and 0.97; for the TM2N6 configuration, R 2 Between 0.82 and 0.95. This indicates that D df It is an accurate and universal adsorption energy descriptor that is unaffected by the coordination environment and has good mobility.
[0056] (c) In contrast, the same dataset was used to study the traditional d-band center (ε) d ) and d–p band central difference (ε d-p The tests were conducted. The results showed that the linear relationship between these two descriptors and the adsorption energy was poor (R0). 2 (Generally below 0.7). Compared to them, D df The prediction accuracy has been improved by 28%–44%, demonstrating a significant advantage.
[0057] 5.4 Physical meaning of descriptors In order to understand A t / W d To understand how adsorption is affected, we correlated it with the ICOHP value of the d–p orbital coupling of the TM–S bond. We found that A… t / W d 0.5 It shows a strong linear correlation with the ICHP value (R 2 >0.78, see Figure 10 e). Smaller A t / W d 0.5 A larger value corresponds to a wider, flatter d orbital, meaning there are more unoccupied antibonding states near the Fermi level, which favors strong hybridization with s atomic orbitals, resulting in strong adsorption. Conversely, a larger A value...t / W d 0.5 The value corresponds to a localized, sharp d-orbital, indicating weak adsorption capacity (see...). Figure 10 f).
[0058] 6. Prediction of Li2S decomposition behavior The decomposition of Li₂S is the initial step in the charging process and also the kinetically rate-limiting step. Research has found that descriptor A... t / W d 0.5 There is a good linear relationship between the average Li–S bond length and Li2S (TM2N8: R²=0.88, TM2N6: R²=0.75, see...). Figure 11 (a, b). A longer bond length means a weaker Li–S bond, making Li₂S more prone to decomposition, and resulting in a smaller overpotential in the lithium-sulfur battery. Therefore, A t / W d 0.5 It can serve as an effective descriptor to predict the ease of decomposition of Li2S.
[0059] 7. Calculation of the thermodynamic energy barrier for sulfur reduction 7.1 Calculation of Sulfur Reduction Reaction Pathway For all structurally stable catalysts, calculate the Gibbs free energy change during the discharge process, from Li₂S₄ to Li₂S on its surface. The calculation is based on the following reaction equation: Li2S4 +2Li₂S₂+10Li + +2e - →Li2S2 +3Li₂S₂+8Li + (ΔG1); Li2S2 +3Li₂S₂+8Li + +8e - →Li2S +7Li2S(ΔG2).
[0060] 7.2 Results Analysis The calculation results show that the conversion step from Li2S2 to Li2S (ΔG2) requires the highest energy and is the rate-determining step of the entire sulfur reduction reaction.
[0061] Among all TM2N8 configuration catalysts, Os2N8 exhibits the lowest ΔG2 energy barrier, at only 2.96 eV. Ta2N8, on the other hand, has the highest energy barrier (8.71 eV).
[0062] Among all TM2N6 configurations, Ru2N6 exhibits the lowest energy barrier (3.34 eV).
[0063] 7.3 Mechanism Analysis The difference in adsorption energy between Li2S2 and Li2S (E) d Correlation between ΔG2 and the reaction energy barrier revealed a strong linear relationship. This indicates that the magnitude of ΔG2 is primarily determined by the difference in adsorption strength between the intermediate and final products. For example, excessive adsorption of Li2S2 by the Ta-based catalyst inhibits its further desorption and conversion, leading to an increase in the reaction energy barrier.
[0064] The above series of investigations demonstrate that this invention, through a combination of first-principles calculations and machine learning, has successfully constructed a novel descriptor D capable of accurately predicting the adsorption capacity of diatomic catalysts for polysulfides. df The physical implications were revealed. The ICOHP value of the TM d–S p orbital coupling between the catalyst and sulfur-containing molecules was verified, and the energy barrier for Li₂S decomposition was determined by A. t / W d 0.5 Description: The thermodynamic energy barrier for sulfur reduction can be described by the difference in adsorption energies between Li₂S₂ and Li₂S. Through a series of density functional theory calculations, a diatomic catalyst with the configuration of Os₂N₈ was finally selected. It exhibits excellent performance in thermodynamic stability, adsorption capacity, and catalytic activity, and has great potential for application in the cathode of high-performance lithium-sulfur batteries.
[0065] Unless otherwise specified, all raw materials used in this invention are existing substances that can be purchased directly from the market.
[0066] The above are merely preferred embodiments of the present invention and are not intended to limit the scope of protection of the present invention. Any modifications, equivalent substitutions, improvements, etc., made 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 screening diatom catalysts for lithium-sulfur batteries, characterized in that, The method includes the following steps: S1. Based on density functional theory, two diatomic catalyst models with configurations of TM2N8 and TM2N6 are constructed, where TM represents a transition metal element; S2. Optimize the structure and evaluate the stability of the two-configuration diatomic catalyst models, and screen out the catalyst models with stability; the stability includes thermodynamic stability and kinetic stability. S3. For all the selected stable diatomic catalyst models, density functional theory calculations are used to obtain their electronic structure information and adsorption energy for sulfur-containing molecules. The electronic structure information includes the d-orbital state density; S4. Extract d-orbital related characteristic parameters from the d-orbital density of states obtained in step S3, and perform correlation analysis with the obtained adsorption energy of sulfur-containing molecules to screen out characteristic parameters that are strongly correlated with adsorption energy. The d-orbital related characteristic parameters include the d-orbital density of states integral area, d-orbital energy span, d-band center, energy difference between d-orbital and sulfur p-orbital, the highest intensity of d-orbital and its corresponding energy value, the highest energy level of d-orbital, the lowest energy level of d-orbital, the density of states integral area of occupied d-orbitals, and the ratio of the density of states integral area of occupied to non-occupied d-orbitals. S5. For the selected characteristic parameters that are strongly correlated with adsorption energy, a genetic algorithm is used to optimize the coefficients in a high-dimensional parameter space to construct the adsorption energy descriptor D. df Descriptor D df The details are as follows; D df =exp(λ·A t / W d ) Among them, A t W is the integral area of the density of states of the d orbitals. d Let λ be the energy span of the d orbital, and λ be a positive constant optimized by the genetic algorithm. S6. Using density functional theory, calculate the reaction free energy change of sulfur-containing compound reduction reactions on the surface of all stable diatomic catalyst models, and obtain the thermodynamic energy barrier of sulfur reduction reactions. S7. The adsorption energy descriptor D constructed according to step S5. df Based on the thermodynamic energy barrier of the sulfur reduction reaction obtained in step S6, a linear relationship between the thermodynamic energy barrier and the adsorption energy difference is constructed, and the bimetallic atom catalyst with the best comprehensive performance is screened based on the theoretical calculation results.
2. The screening method according to claim 1, characterized in that, The substrate of the biatomic catalyst model in step S1 is a two-dimensional heterostructure of boron nitride and graphene; wherein, bimetallic atoms are doped in the boron nitride part of the structure; The transition metal elements include all transition metal elements in the fourth, fifth, and sixth periods, excluding Cd, Hg, and lanthanides.
3. The screening method according to claim 1, characterized in that, The structural optimization described in step S2 involves performing structural optimization on all catalyst models using first-principles calculations. The thermodynamic stability is indicated by the formation energy of all catalyst models calculated using density functional theory. A negative formation energy indicates that the structure is thermodynamically stable. The kinetic stability: According to ab initio molecular dynamics simulations, no significant distortion or bond breakage occurred in the atomic structure after the simulation, indicating that the structure has good kinetic stability.
4. The screening method according to claim 1, characterized in that, The electronic structure information mentioned in step S3 also includes the non-degenerate d-orbit coupled Hamiltonian population of the TM2N6 configuration; Calculations using density functional theory also include obtaining the TM dS p orbital density of states and orbital coupled Hamiltonian configuration of the adsorption configuration. The sulfur-containing molecules include S8, Li2S4, Li2S2, and Li2S.
5. The screening method according to claim 4, characterized in that, Step S5 also includes using the adsorption energies of S8, Li2S4, Li2S2, and Li2S molecules to describe the adsorption energy descriptor D. df Perform linear fitting to verify its accuracy.
6. The screening method according to claim 5, characterized in that, The process of verifying its accuracy also includes: constructing a descriptor A for the integral value of the TM d–S p orbital coupled Hamiltonian layout based on the selected feature parameters. t / W d 0.5 ; Among them, A t W is the integral area of the density of states of the d orbitals. d For the d orbital energy span.
7. The screening method according to claim 6, characterized in that, For the obtained descriptor A t / W d 0.5 It also includes: statistically averaging the Li–S bond length based on the adsorption configuration, and constructing a Li–S bond length-A… t / W d 0.5 The linear relationship is used to predict the ease of decomposition.
8. The screening method according to claim 1, characterized in that, The thermodynamic energy barrier of the sulfur reduction reaction described in step S6 is obtained by calculating the reaction free energy change of Li2S4 to Li2S2 and further to Li2S on the catalyst surface using density functional theory.
9. A diatomic catalyst obtained by screening using the screening method according to any one of claims 1-8, characterized in that, The catalyst has an Os2N8 configuration, in which two osmium atoms are anchored on a two-dimensional heterostructure substrate of boron nitride and graphene, and each osmium atom is coordinated with four nitrogen atoms.
10. A lithium-sulfur battery cathode material, characterized in that, It includes the diatomic catalyst as described in claim 9.