Multi-parameter synergistic dynamic screening method for adsorbed species on surface of catalyst

Abstract

The invention belongs to the field of computational chemistry, and particularly relates to a multi-parameter collaborative dynamic screening method for adsorbed species on the surface of a catalyst. Comprising the following steps that an initial crystal structure is constructed and optimized through VESTA software, and a stable crystal structure model is obtained through density functional theory optimization; integrating working condition parameters such as a solvent, pH and potential, and drawing a Bayer diagram to label adsorption species and a mapping relation; and calculating a zero charge point and capacitance to obtain electrochemical parameters, and determining stable adsorption species under actual reaction through constant potential molecular dynamics simulation. According to the scheme, a deep structure-activity relationship between an active site structure and catalytic activity under a working condition is established through integration of all working condition parameters, labeling of adsorption species by a Boubye diagram, constant potential molecular dynamics simulation and the like; the problems that a systematic method combined with actual working condition parameters is lacked in prediction of the adsorbed substances on the surface of the catalyst, and real adsorbed species are difficult to accurately screen are solved, and accurate capture and efficient screening are achieved.

Description

TECHNICAL FIELD

[0001] The present scheme belongs to the field of computational chemistry, and specifically relates to a multi-parameter synergistic catalyst surface adsorbed species dynamic screening method. BACKGROUND

[0002] The accurate prediction of catalyst surface adsorbed species is the core basis for the design and performance optimization of catalytic materials. The core goal is to capture the real adsorbed species on the catalyst surface under actual reaction conditions through theoretical simulation, to provide reliable basis for revealing the catalytic mechanism and screening high-efficiency catalysts.

[0003] In the field of catalyst surface adsorbed species prediction, traditional methods mostly rely on static simulation framework in vacuum environment, and only single-state calculation of material crystal structure is performed through density functional theory or molecular mechanics, which can only predict the adsorbed species and related performance parameters under ideal conditions. However, such methods have significant limitations. The simulation process does not include key working condition parameters in actual electrochemical reactions. Neither does it consider the solvation effect of solvent molecules on adsorbed species, nor does it involve the surface protonation state changes caused by pH fluctuations in the reaction system, nor does it include the influence of electrode potential regulation on electron transfer and adsorption behavior. Therefore, the adsorbed species obtained by simulation deviates significantly from the actual adsorption state on the catalyst surface in the real reaction environment, and it is difficult to reflect the regulation law of working condition changes on adsorption behavior.

[0004] At present, some improvements have been made to overcome the shortcomings of traditional methods. For example, a high-efficiency electrocatalyst screening method based on HPC and AI fusion disclosed in Chinese patent CN114141318A collects original crystal structure, analyzes adsorption site characteristics, combines machine learning model to predict adsorption energy, and verifies it by DFT, which to some extent solves the problems of high cost, long cycle and insufficient machine learning samples in traditional experimental screening.

[0005] However, this method still does not break through the limitation of ideal conditions. The prediction of adsorbed species is mainly based on adsorption energy calculation in vacuum environment, without integrating actual working condition parameters such as solvent, pH and potential. There is a lack of systematic method combining the above working condition parameters, and it is impossible to establish the correlation between working condition changes and adsorbed species, which makes it still difficult to accurately screen the real structure of catalyst surface under actual reaction conditions. SUMMARY

[0006] The purpose of the present scheme is to provide a multi-parameter synergistic catalyst surface adsorbed species dynamic screening method to solve the technical problem that in the field of catalyst surface adsorbed species prediction, there is a lack of systematic method combining actual working condition parameters, which makes it difficult to accurately screen the real adsorbed species on the catalyst surface under actual reaction conditions.

[0007] In order to achieve the above object, the scheme provides a multi-parameter coordinated catalyst surface adsorbed species dynamic screening method, comprising the following steps:

[0008] S10: An initial crystal structure model of the target material is constructed by VESTA software, and the initial crystal structure model matching the actual material characteristics is obtained through atomic coordinate optimization and unit cell parameter adjustment;

[0009] S20: The initial crystal structure model is optimized by density functional theory, and the optimization process includes setting calculation accuracy, calculating electronic structure, optimizing geometric structure and considering correction van der Waals force, and finally obtaining a stable crystal structure model;

[0010] S30: The stable crystal structure model is constructed by an implicit solvent model to simulate solvent effect, the H + / OH- ion concentration or surface protonation state is adjusted to cover the pH 0-14 range to simulate pH value, and the constant potential algorithm is used to apply-1.5V to 1.5V target potential to simulate potential conditions, and the crystal structure model integrating actual working parameters is obtained;

[0011] S40: The Buryi diagram of the crystal structure model is drawn with pH as the horizontal axis and potential as the vertical axis, and the real adsorbed species corresponding to each stable region is marked in the Buryi diagram, and the mapping relationship between pH, potential and real adsorbed species is analyzed;

[0012] S50: Based on the real adsorbed species in the mapping relationship, the potential corresponding to the charge zero point is fitted by calculating the surface charge density under different potentials, and the zero charge point is obtained; the correctness of the material constant potential theoretical calculation result is verified by calculating the capacitance, and the electrochemical performance parameters are obtained;

[0013] S60: According to the preset target working condition, the advantage adsorbed species is screened out from the real adsorbed species, the dynamic adsorption process of the advantage adsorbed species on the catalyst surface is simulated by constant potential molecular dynamics method, the energy fluctuation, diffusion energy barrier and bond length change parameters of the adsorption configuration are calculated, the long-term stability evaluation result is obtained combined with the preset stability threshold, and finally the catalyst surface adsorbed species stably existing under the actual reaction condition is determined.

[0014] The principle and technical effect of the scheme are that: the scheme integrates all range of actual working condition parameters systematically, accurately covers pH 0-14 (including acidic, neutral, alkaline environment), potential -1.5V to 1.5V, and includes the specific action of electrolyte ions such as Cl-, SO42-, and the like, and reproduces the real reaction micro-environment of the catalyst through multi-parameter collaborative simulation; on this basis, the atomic arrangement of the material crystal structure, the surface electronic state and the working condition parameters are coupled to construct a dynamic mapping of the combination of working condition parameters, material structure and adsorbed species, and capture the formation and conversion rules of adsorbed species under different conditions. This way can completely reproduce the real scene characteristics of fuel cells, water electrolysis and the like, break through the defects of traditional simulation of incomplete working condition parameters, and accurately capture the real adsorbed species. For example, when simulating the alkaline working condition of industrial electrolytic tank (such as pH = 13, potential 1.2V, high concentration of OH-), the dominant adsorbed species can be accurately identified as O instead of the ideal environment misjudgment of OH, providing micro-reliable basis for catalyst performance evaluation and optimization.

[0015] The scheme takes the Bucky plot as the core tool, draws a chart with pH as the horizontal axis and potential as the vertical axis, and combines the adsorption energy data calculated by DFT with the synergistic effect of working condition parameters, and clearly marks the most stable adsorbed species on the catalyst surface in each pH / potential interval in the chart, forming a direct correspondence between working conditions and stable adsorbed species. In this way, the distribution rule of adsorbed species in the whole working condition range can be intuitively displayed, and the key adsorbed species under the target working condition can be quickly located without complex calculation, significantly improving the screening efficiency. For example, when analyzing the acidic fuel cell working condition, the stable adsorbed species can be directly read from the Bucky plot as *OOH, which improves the simulation efficiency compared with the traditional trial-and-error method.

[0016] The scheme adopts constant potential molecular dynamics method to simulate the long time scale of the dominant adsorbed species screened by the Bucky plot, keeps the potential consistent with the target working condition, and real-time tracks the changes of bond length, energy fluctuation, diffusion energy barrier and other dynamic parameters of the adsorbed species and the catalyst surface, and analyzes the dynamic behavior through atomic coordinate trajectory. The scheme reveals the interaction mechanism between the adsorbed species and the catalyst surface from the atomic level, accurately evaluates the long-term stability, for example, simulating the behavior of *OOH species under 0.6V potential, which can capture the slow growth of bond length, the decay trend of energy fluctuation expansion, and provide micro-evidence for predicting the long-term activity of the catalyst.

[0017] When simulating complex working conditions containing multiple electrolyte ions, the scheme focuses on the competitive adsorption of electrolyte ions and target adsorbed species, calculates the adsorption energy change of the target species under different ion concentrations, and determines the influence mechanism of ion competition on adsorption stability through DFT electronic structure analysis. The scheme can find the micro-rules that cannot be revealed by traditional performance prediction models, and provide targeted guidance for catalyst design under complex working conditions.

[0018] The scheme establishes the deep structure-activity relationship of the working condition parameters, adsorbent species characteristics, and catalyst activity / stability. By comparing the adsorbent species states (such as bond length, adsorption energy, and charge distribution) of different catalysts under the same working conditions, and combining with the electrochemical performance data, the key properties affecting the activity / stability are located, which can accurately analyze the microcosmic root of the performance difference and provide a clear direction for optimization.

[0019] In summary, the scheme integrates full working condition parameters, adsorbent species labeled by Bader diagram, and constant potential molecular dynamics simulation to establish deep structure-activity relationship, solves the problems of lack of systematic method combining actual working condition parameters in predicting catalyst surface adsorbent, and difficulty in accurately screening real adsorbent species, and realizes accurate capture and efficient screening.

[0020] Further, in the optimization process of step S20, the specific parameters and synergistic mechanism of each sub-step are as follows:

[0021] S21: In the initialization stage, set the calculation precision and basic parameters suitable for the characteristics of the material, including turning on the spin polarization according to whether the catalyst has magnetism, and judging whether to continue the calculation by calling the charge density file calculated in the previous calculation to reduce repeated operation;

[0022] S22: In the electronic structure calculation stage, the plane wave cutoff energy matching the element composition of the material is used, the algorithm is used to accelerate the electronic self-consistent field calculation, the electron occupation number calculation method is associated with the force convergence requirement of geometric optimization, and the synergistic optimization of electronic state and atomic position is ensured; In the geometric structure optimization stage, the conjugate gradient algorithm is used to update the atomic position, and the atomic force reaches the stable threshold as the convergence standard. The electronic structure calculation results are synchronized regularly in the iteration process to avoid local energy minimum value;

[0023] S23: In the advanced processing stage, the weak interaction description is optimized by van der Waals force correction, and specific processing is performed for strong correlation electronic system. The solvent environment simulation is not enabled at this stage to avoid parameter conflict with the implicit solvent model;

[0024] S24: Output the optimized structure parameters, energy data, and electronic state information to provide a basis for subsequent analysis.

[0025] The optimization process of step S20 cooperates spin polarization and strong correlation electron processing, which not only accurately adapts to the electronic state characteristics of magnetic catalysts, but also captures the strong correlation effect of transition metal d orbital electrons. This fine optimization of atomic level electronic behavior directly serves the subsequent micro recognition of adsorbent species. For example, in the optimization of spinel structure containing Co, the influence of valence state distribution of Co 3+ / Co 2+ on the surface adsorption site can be accurately analyzed.

[0026] The parameter linkage mechanism of electronic structure and geometry optimization not only accelerates the convergence of structure through the conjugate gradient algorithm, but also avoids the deviation of structure caused by local energy minimum. This efficient and accurate double effect ensures that the optimized crystal structure can truly reflect the surface atomic arrangement of the material under actual working conditions. For example, in the optimization of layered hydroxide catalysts, the key configuration of interlayer hydroxyl groups can be accurately preserved, providing a reliable structural basis for marking the stable region of adsorbate species in the Born chart.

[0027] In addition, the advanced processing stage does not enable solvent simulation to avoid parameter conflict design, forming a step-by-step focused synergy logic, where step S20 focuses on material intrinsic structure optimization, and step S300 integrates working condition parameters separately. The connection of the two ensures both structural accuracy and parameter interference avoidance. This step-by-step optimization strategy can provide clear variable control for the identification of adsorbate species in complex systems.

[0028] Further, in step S30, the integration of actual working condition parameters is achieved through an automated script, which includes the following steps:

[0029] S31: Pre-set target potential range and determine key nodes, script automatically traverses each potential node, creates an independent working directory for each node, and copies the basic crystal structure file;

[0030] S32: Call tools in each directory to extract the total number of valence electrons of the material, and calculate the adjusted valence electron parameters based on the current potential node;

[0031] S33: Generate an adapted calculation configuration file based on the parameters, which integrates electron step convergence accuracy, ion step force convergence standard, van der Waals force correction method, strong correlation electron processing parameters, and parallel computing settings, and reserves solvent environment simulation parameter interface for subsequent working condition expansion;

[0032] S34: Through script linkage, realize the automatic matching of potential parameters, electronic structure calculation parameters, and structure files, complete the batch configuration of calculation environment under different potential working conditions, and ensure the synergy integration of actual working condition parameters and simulation calculation settings.

[0033] The actual working condition parameter integration is realized through an automatic script. The scheme can not only create an independent directory by presetting a potential node and traversing, efficiently complete the batch configuration of multiple potential working conditions, and avoid the repetition and errors of manual operation, but also can extract the valence electron number by calling a tool and adjust the parameters combined with the potential calculation to ensure the accurate matching of the potential simulation and the intrinsic electron characteristics of the material, thereby solving the parameter disconnection problem in traditional manual configuration. The configuration file integrates multiple parameters such as electron step convergence and ion step force convergence, and reserves a solvent environment simulation interface, which not only ensures the synergy of the current working condition simulation, but also provides compatibility for subsequent expansion of complex parameters such as pH and electrolyte ions, and realizes seamless connection of basic potential configuration and multi-working condition expansion.

[0034] The script core of the scheme is the deep synergy of working condition parameters and calculation settings, rather than simply pursuing batch processing speed. For example, in the simulation of hydrogen production from seawater with high concentration of Cl-, the script can reserve an electrolyte ion parameter interface in the configuration file and accurately adjust the valence electron parameters combined with the potential node, so that the simulation environment of the subsequent adsorbed species (such as *H and *Cl) is highly consistent with the real seawater system.

[0035] Further, in the optimization process of step S20, the parameter dynamic adjustment based on material characteristics is used to realize the synergistic optimization of electronic structure and geometric structure, which specifically includes selecting matching electronic structure calculation parameters for material element composition, correlating electronic state convergence criteria and atomic force convergence requirements, making specific corrections for weak interaction and strong correlation electronic systems, and isolating solvent environment simulation parameters in the advanced processing stage to avoid conflicts with subsequent working condition parameter integration.

[0036] The scheme realizes the deep synergistic optimization of electronic structure and geometric structure through parameter dynamic adjustment based on material characteristics, so that the optimized crystal structure is more consistent with the intrinsic characteristics of the actual material, providing a reliable foundation for subsequent actual working condition parameter integration and adsorbed species identification. By selecting matching electronic structure calculation parameters for material element composition, the scheme avoids structural deviation caused by parameter disconnection by correlating the design of electronic state convergence criteria and atomic force convergence requirements, significantly improving the authenticity of structure optimization; the specific correction for weak interaction and strong correlation electronic systems enhances the description accuracy of electronic structure and geometric characteristics of complex material systems, and can accurately capture the micro interaction mechanism of the material surface.

[0037] The high-level processing stage in the scheme isolates the solvent environment simulation parameters, avoiding parameter conflicts with the subsequent working condition parameter integration of the implicit solvent model, forming an ordered connection logic of structure intrinsic optimization and working condition parameter loading, and ensuring the coherence and stability of the whole process. The scheme takes the accurate identification of adsorbate species under actual working conditions as the core goal. For example, when dealing with layered catalysts with weak interactions, through targeted weak interaction correction, the key structural features between layers can be accurately retained, making the subsequent analysis of the stability region of adsorbate species more consistent with actual reaction observations.

[0038] Further, in step S31, the script realizes automatic traversal through a preset list of potential nodes, each node corresponding to a directory named with a potential value. The file integrity is verified synchronously when copying the basic crystal structure file, ensuring the consistency of the initial structure of each directory, and the directory naming rule is mapped to the coordinate of the potential axis of the subsequent Bader diagram.

[0039] Through the mapping design of potential nodes and directory naming, the scheme can realize batch configuration of multiple working conditions, and provide a structured data basis for subsequent visualization analysis of adsorbate species and potential relationship. The naming rule of the scheme is directly related to the working condition parameters. When analyzing the *O adsorbate species under the working condition of alkaline electrolytic water, the corresponding data can be quickly located through the directory name, avoiding parameter and data matching errors. This direct correlation between working conditions, data, and analysis can provide accurate correspondence for the whole chain of working condition parameters, adsorbate species, and visualization analysis, effectively improving the coherence and accuracy of the screening process. The direct correlation between working conditions, data, and analysis in the scheme focuses on the accurate mapping of adsorbate species and working conditions. For example, when drawing the *OH adsorption stability region of a catalyst at 0.8V potential, the calculation data corresponding to this potential can be directly located through the directory name, ensuring that the stability interval of the adsorbate species in the Bader diagram strictly matches the actual potential condition. This way of working condition data and analysis can ensure the accuracy of adsorbate species identification under actual working conditions.

[0040] Further, in step S32, the called tool is a material electronic structure analysis tool, which extracts the total valence electron number through a specified function module, and calculates the adjusted valence electron parameter based on the linear correlation formula of the current potential node. This parameter is directly written into the core parameter item of the calculation configuration file, realizing the dynamic matching of potential and material electronic characteristics.

[0041] The dynamic matching of potential and material electronic properties is realized by the core parameter item of the configuration file, such as NELECT, the adjusted valence electron number, so that the valence electron parameters are adjusted in real time in association with the potential, ensuring that the potential simulation is not only a simple loading of external conditions, but also deeply coupled with the intrinsic electronic structure of the material, such as the valence electron distribution and the surface electron density. This internal coupling can accurately capture the influence of potential changes on the surface electronic environment of the catalyst, for example, in the simulation of H adsorption under a potential of-0.5V in seawater hydrogen production, the dynamic adjustment of valence electron parameters can reflect the changes in surface electron density under this potential, so that the calculation results of H adsorption energy are more consistent with the actual electronic state of the reaction, effectively improving the accuracy of adsorbate species identification, providing a reliable electronic structure basis for subsequent annotation of stable regions of adsorbate species in the construction of the Bader diagram, calculation of electrochemical performance, and accurate characterization of the real adsorption behavior on the catalyst surface under actual working conditions.

[0042] Further, in step S33, the generated calculation configuration file includes multiple sets of sub-configuration files for structure optimization, electronic state analysis, and solvent environment simulation, and each sub-configuration file is associated through a unified parameter interface. The solvent environment simulation sub-configuration file separately enables solvent model parameters, and forms a stage-by-stage enabling mode with other sub-configuration files.

[0043] The stage-by-stage enabling design of multiple sets of sub-configuration files not only meets the specific parameter needs of different calculation stages such as structure optimization, electronic state analysis, and solvent simulation, but also ensures the coherence of parameter transmission through a unified interface. When simulating the working conditions of an alkaline electrolytic tank (such as pH=13 and potential 1.2V), the structure of the catalyst surface can be pre-optimized through the structure optimization sub-configuration file, the surface electron distribution can be analyzed through the electronic state analysis sub-configuration file, and finally the high-concentration OH-parameters can be loaded through the solvent environment sub-configuration file, so that the simulation environment of O adsorbate species is more consistent with the real reaction.

[0044] Further, in step S22, the intermediate trajectory file generated by each ion iteration synchronously records the atomic force direction and the electron state density change rate. The intermediate trajectory file is used not only for local minimum value judgment of geometric optimization, but also as an initial trajectory input for constant potential molecular dynamics simulation in step S60, realizing the trajectory data reuse of structure optimization and stability evaluation combination.

[0045] From the perspective of multi-scale structure optimization in step S20, the continuous recording of the atomic force direction in the intermediate trajectory file can be used to accurately judge whether the geometric optimization falls into a local energy minimum value. When the force direction of adjacent iteration steps changes abruptly, non-global stable structures can be identified in time to avoid passing such deviated structures to subsequent steps. At the same time, the synchronous recording of the electron state density change rate can reflect the dynamic evolution of the surface electronic environment of the material during optimization, providing a basis for understanding the electronic mechanism of structural stability.

[0046] More importantly, the intermediate trajectory file is directly input as the initial trajectory of the constant potential molecular dynamics simulation in step S60, so that the stability evaluation is not started from the isolated optimized structure, but continues the inertial trend of atomic motion in step S20. This continuation makes the energy fluctuations, diffusion energy barriers and other parameters calculated in step S60 more consistent with the dynamic evolution law of the structure in the actual reaction. For example, when simulating the stability of O adsorption on the surface of a catalyst, the intermediate trajectory file can be traced back to the slight deviation of an atom caused by unbalanced force in step S20, which is difficult to detect in the isolated optimized structure, but can be amplified as the instability of the adsorption configuration in the long-term simulation of step S60, ultimately affecting the judgment of the real stable adsorption species.

[0047] This data multiplexing of the scheme breaks through the data fragmentation problem of structure optimization and stability evaluation in traditional simulation, not only reduces the resource consumption of repeated calculation, but also significantly improves the accuracy and reliability of identifying stable adsorption species on the surface of the catalyst under actual working conditions by preserving the dynamic correlation information of the whole process.

[0048] Further, the surface charge density distribution of the adsorption species under different potentials is extracted synchronously when calculating the zero charge point, generating a three-dimensional mapping table of potential, charge polarity and adsorption energy; in the constant potential molecular dynamics simulation, the capacitance value in the three-dimensional mapping table is used as the judgment threshold of the standard deviation of bond length fluctuation, and the energy fluctuation reference of the adsorption configuration is corrected by the potential of the zero charge point, so that the three-dimensional mapping table is used to verify the calculation accuracy of the capacitance and assist in optimizing the dynamic parameters of the stability evaluation, finally outputting the correlation curve of the electrochemical performance and stability, and using the correlation curve for precise labeling of the stable region of the Pourbaix diagram.

[0049] In terms of technical implementation, the multi-scale structure optimization of the scheme outputs stable material structures based on the density functional theory, laying a reliable foundation for subsequent simulation; the integration of actual working parameters upgrades the vacuum static simulation to a dynamic environment close to the real reaction through the implicit solvent model, pH gradient coverage and wide-range potential loading, solving the deviation caused by the neglect of key factors such as solvent, pH and potential in traditional simulation; the dynamic mapping of the adsorbent presents the stable adsorption species under different combinations of pH and potential through the Pourbaix diagram, establishing a clear correlation between working conditions and adsorbents; the calculation of electrochemical performance is verified by the zero charge point and capacitance, ensuring the consistency of theoretical calculation and actual performance; the constant potential molecular dynamics simulation further evaluates the long-term stability of the dominant adsorption species, finally locking the real adsorption species under actual reaction conditions.

[0050] The whole-chain design of structure optimization, working condition integration, adsorption mapping, performance verification and stability evaluation not only significantly improves the characterization accuracy of the adsorbent, but also realizes the deep correlation from working condition parameters to material structure, adsorption behavior and electrochemical activity / stability, completely solves the core problem of the disconnection between traditional vacuum simulation and actual application scene, and provides precise and reliable theoretical guidance for efficient screening and optimization of catalysts. BRIEF DESCRIPTION OF DRAWINGS

[0051] Figure 1 A flowchart of a multi-parameter coordinated catalyst surface adsorbate dynamic screening method in an embodiment of the present application is shown. DETAILED DESCRIPTION

[0052] The concept and technical effects of the present application will be described below in conjunction with the embodiments to clearly and completely understand the purpose, features and effects of the present application. Obviously, the described embodiments are only a part of the embodiments of the present application, not all embodiments, and other embodiments obtained by those skilled in the art without creative labor based on the embodiments of the present application are within the scope of the present application:

[0053] As shown in Figure 1 The present scheme provides a multi-parameter coordinated catalyst surface adsorbate dynamic screening method, comprising the following steps:

[0054] S10: An initial crystal structure model of the target material is constructed by VESTA software, and the initial crystal structure model matching the actual material characteristics is obtained through atomic coordinate optimization and unit cell parameter adjustment;

[0055] S20: The initial crystal structure model is optimized by density functional theory, and the optimization process includes setting calculation accuracy, calculating electronic structure, optimizing geometric structure and considering correction of van der Waals force, and finally obtaining a stable crystal structure model;

[0056] S30: The stable crystal structure model is constructed by an implicit solvent model to simulate the solvent effect, the H + / OH- ion concentration or surface protonation state is adjusted to cover the pH 0-14 range to simulate the pH value, and the constant potential algorithm is used to apply a target potential of-1.5V to 1.5V to simulate the potential condition, and a crystal structure model integrating actual working condition parameters is obtained;

[0057] S40: A Pourbaix diagram is drawn for the crystal structure model with pH as the horizontal axis and potential as the vertical axis, and the real adsorbate corresponding to each stable region is marked in the Pourbaix diagram, and the mapping relationship between pH, potential and real adsorbate is analyzed;

[0058] S50: Based on the real adsorption species in the mapping relationship, the potential corresponding to the charge zero point is fitted by calculating the surface charge density under different potentials, and the zero charge point is obtained; the correctness of the theoretical calculation result of the material constant potential is verified by calculating the capacitance, and the electrochemical performance parameters are obtained;

[0059] S60: According to the preset target working condition, the advantage adsorption species is selected from the real adsorption species, the dynamic adsorption process of the advantage adsorption species on the catalyst surface is simulated by the constant potential molecular dynamics method, the energy fluctuation, diffusion energy barrier and bond length change parameters of the adsorption configuration are calculated, and the long-term stability evaluation result is obtained combined with the preset stability threshold, and finally the catalyst surface adsorption species stably existing under the actual reaction condition is determined.

[0060] In the optimization process of step S20, the specific parameters and synergistic mechanism of each sub-step are as follows:

[0061] S21: In the initialization stage, the calculation precision and basic parameters suitable for the material characteristics are set, including starting the spin polarization according to whether the catalyst has magnetism, and judging whether to continue the calculation by calling the charge density file of the previous calculation to reduce repeated operation;

[0062] S22: In the electronic structure calculation stage, the plane wave cutoff energy matching the material element composition is used, the electronic self-consistent field calculation is accelerated by algorithm, the electron occupation number calculation method is associated with the convergence standard of the geometric optimization force convergence requirement, and the synergistic optimization of the electronic state and the atomic position is ensured; In the geometric structure optimization stage, the conjugate gradient algorithm is used to update the atomic position, and the atomic force reaches the stable threshold as the convergence standard, and the electronic structure calculation result is synchronized regularly in the iteration process to avoid local energy minimum value;

[0063] S23: In the advanced processing stage, the weak interaction description is optimized by van der Waals force correction, and specific processing is performed for strong correlation electron system, and solvent environment simulation is not enabled to avoid parameter conflict with implicit solvent model;

[0064] S24: The optimized structure parameters, energy data and electronic state information are outputted to provide basis for subsequent analysis.

[0065] Specifically, in the optimization process of step S20, the synergistic optimization of the electronic structure and the geometric structure is realized by dynamic adjustment of the parameters adapted to the material characteristics, which specifically includes selecting matching electronic structure calculation parameters according to the material element composition, associating the electronic state convergence standard with the atomic force convergence requirement, specifically modifying the weak interaction and strong correlation electron system, and isolating the solvent environment simulation parameters in the advanced processing stage to avoid parameter conflict with the subsequent working condition parameters.

[0066] More specifically, in step S21, after the initial crystal structure model of the target material is constructed by the VESTA software, it is exported in a format that can be recognized by the density functional theory calculation program. When entering the structure optimization stage, first set the basic calculation parameters, i.e., determine the calculation precision level according to the crystal structure characteristics of the material; at the same time, check whether the current calculation directory contains the electronic structure result file generated by the previous optimization, if it exists, call the result to continue the calculation, otherwise initialize the electronic structure from zero to reduce the amount of repeated calculation.

[0067] In step S22, electronic structure calculation is performed using density functional theory, the plane wave cutoff energy is selected according to the element composition of the material, and the iterative algorithm is used for electronic self-consistent field calculation. Gaussian broadening is used, and the convergence criteria are set to ensure that the electronic structure calculation results match the force convergence requirements of subsequent geometric structure optimization, realizing the coordinated optimization of electronic state and atomic position. Based on the atomic force data obtained from the electronic structure calculation, the conjugate gradient algorithm is used to adjust the atomic coordinates and cell parameters. After completing a round of atomic position update, the electronic structure calculation is performed again, and the iteration is repeated until the atomic force reaches a stable threshold or the maximum iteration step is reached. During the iteration process, the electronic structure results are synchronized regularly to avoid structural optimization deviation caused by local energy minimum, and to ensure that a globally stable crystal structure is obtained.

[0068] In step S23, for the possible interlayer van der Waals forces in the material, DFT-D3 is used for optimization description; for systems containing strong correlated electrons, the accuracy of electronic structure description is improved by introducing Hubbard U correction and other methods. At this stage, any solvent environment simulation parameters are not enabled to avoid conflicts with the parameters of the implicit solvent model in step S30, ensuring the independence of the integration stage of the working condition parameters.

[0069] In step S24, after optimization, the key parameters of the stable crystal structure, the total energy data of the system and the electronic state information are extracted and saved to form a standardized output file, providing basic data support for the actual working condition parameter integration in step S30, the drawing of the Born diagram in step S40 and the subsequent calculation of electrochemical performance.

[0070] In this specific implementation, through the dynamic adjustment of parameters adapted to the characteristics of the material, the deep collaborative optimization of electronic structure and geometric structure is realized, and the optimized crystal structure is more consistent with the intrinsic characteristics of the actual material, providing a reliable foundation for the subsequent actual working condition parameter integration and adsorbate species identification.

[0071] The selection of matching electronic structure calculation parameters for material element composition, combined with the association design of electronic state convergence standard and atomic force convergence requirement, avoids structural deviation caused by parameter disconnection, significantly improves the authenticity of structure optimization; The specific correction of weak interaction and strong correlation electronic system enhances the description accuracy of electronic structure and geometric characteristics of complex material system, which can accurately capture the micro mechanism of material surface.

[0072] The simulation parameters of the solvent environment in the advanced processing stage avoid the parameter conflict of the implicit solvent model in the integration of subsequent working condition parameters, form an ordered connection logic of intrinsic structure optimization and working condition parameter loading, and ensure the coherence and stability of the whole process.

[0073] Specifically, in step S22, the intermediate trajectory file generated by each ion iteration synchronously records the atomic force direction and the rate of change of electronic state density, which is used for local minimum value judgment of geometric optimization and as initial trajectory input for constant potential molecular dynamics simulation in step S60, realizing trajectory data reuse combining structure optimization and stability evaluation.

[0074] In step S30, the integration of actual working condition parameters is realized through an automatic script, and the implementation process specifically includes the following steps:

[0075] S31: preset the target potential range and determine the key nodes, the script automatically traverses each potential node, creates an independent working directory for each node and copies the basic crystal structure file;

[0076] S32: call the tool in each directory to extract the total number of valence electrons of the material, and calculate the adjusted valence electron parameters based on the current potential node;

[0077] S33: generate an adaptive calculation configuration file based on the parameters, which integrates electronic step convergence accuracy, ion step force convergence standard, van der Waals force correction method, strong correlation electron processing parameters and parallel computing settings, and reserves a solvent environment simulation parameter interface for subsequent working condition expansion;

[0078] S34: realize the automatic matching of potential parameters, electronic structure calculation parameters and structure files through script linkage, complete the batch configuration of calculation environment under different potential working conditions, and ensure the cooperative integration of actual working condition parameters and simulation calculation settings.

[0079] Specifically, in step S31, the script realizes automatic traversal through a preset potential node list, each node corresponds to a directory named by the potential value, and the file integrity is verified synchronously when copying the basic crystal structure file, ensuring the consistency of the initial structure of each directory, and the directory naming rule is mapped and associated with the potential axis coordinates of the subsequent phase diagram.

[0080] In step S32, the tool called is the material electronic structure analysis tool, the total valence electron number is extracted by specifying the function module, the adjusted valence electron parameter is calculated according to the linear correlation formula of the current potential node, the parameter is directly written into the core parameter item of the calculation configuration file, and the dynamic matching of the potential and the material electronic characteristics is realized.

[0081] In step S33, the generated calculation configuration file includes multiple sets of sub-configuration files for structure optimization, electronic state analysis and solvent environment simulation, and each sub-configuration file is associated through a unified parameter interface, wherein the solvent environment simulation sub-configuration file separately enables the solvent model parameter, and forms a stage-enabled mode with other sub-configuration files.

[0082] In one specific embodiment of the present scheme, the actual working condition parameters are integrated through the following automatic script process, and the specific steps are as follows:

[0083] S31: The preset target potential range is-1.5V to 1.5V, the key node values are selected as (-1.5, -1, -0.5, 0, 0.5, 1, 1.5), and the list is stored in the script variable values. The script iterates through each potential node, automatically creates an independent working directory named "e+ potential value" for each node (such as the potential-1.5V corresponding directory e-1.5), and copies the basic crystal structure file optimized in step S20 to each directory to ensure the consistency of the initial structure of each working condition.

[0084] S32: In each working directory, the 103 function module of the electronic structure analysis tool vaspkit is called to extract the total valence electron number of the material. According to the current potential node value, the adjusted valence electron parameter new_nelect is calculated through the bc calculation tool (the formula is "total valence electron number+potential value"), and the dynamic binding of the potential and the material electronic characteristics is realized.

[0085] S33: Based on the new_nelect parameter, three sets of INCAR configuration files suitable for different calculation scenarios are automatically generated by the script:

[0086] INCAR-1: For structure optimization scenario, set the identifier as "OPT", adopt Normal precision (PREC=Normal), conjugate gradient algorithm (IBRION=2), close wave function file saving (LWAVE=F), and enable strong correlation electron correction through LDAU=T (the parameter setting "T" in the script represents "true"), and the van der Waals force correction adopts the DFT-D3 scheme (IVDW=12).

[0087] INCAR-2: For the electronic state analysis scenario, identified as "Vancume", with Accurate precision (PREC=Accurate), state density output turned on (LORBIT=11), wave function file preserved (LWAVE=T) to support subsequent electronic structure analysis, and the number of iterations increased to 500 to ensure electronic self-consistent convergence.

[0088] INCAR-3: For the solvent environment simulation scenario, identified as "sol", implicit solvent model enabled (LSOL=T, set as "true" in the script), solvent dielectric constant configured (EB_K=80) and Debye length (LAMBDA_D_K=3) to adapt to the actual electrolyte environment, and the broadening coefficient (SIGMA=0.03) adjusted to improve the accuracy of the electronic state description under solvation conditions.

[0089] At the same time, two sets of KPOINTS files (KPOINTS-1 and KPOINTS-2) are generated, corresponding to different k-point grid densities, to adapt to the structure optimization and electronic state calculation with different precision requirements.

[0090] S34: The script copies the auxiliary script cpvasp.sh to each directory to support file management and batch submission during calculation; after traversal, the NELECT parameter values in all directories are summarized by the grep NELECT* / INCAR* command to form a potential valence electron parameter mapping table, providing structured data support for subsequent band diagram drawing (step S40).

[0091] In this specific embodiment, the script realizes physical isolation of different potential working conditions through independent directories (folder_name), avoiding parameter cross-contamination; the uniform variable values cover a wide range of potentials, ensuring the comprehensiveness of the actual working conditions. This design not only realizes batch configuration efficiency, but also through directory naming and direct mapping of potential, makes subsequent analysis (such as the association of potential axis and data in the band diagram) unnecessary for additional parameter matching, reducing human error.

[0092] At the same time, the new_nelect parameter dynamically associates the potential and the number of valence electrons, so that NELECT adjusts in real time with the potential, ensuring that the electronic structure calculation (such as the electronic self-consistent field iteration) truly reflects the change in surface electron density under a specific potential. This coupling breaks the traditional simulation logic of potential as an external condition, accurately capturing the influence of potential on the electronic properties (such as charge distribution and bonding ability) of the adsorption site. For example, when simulating *H adsorption at -0.5V, the valence electron parameter can be adjusted to accurately reflect the surface electron state, improving the authenticity of the adsorption energy calculation.

[0093] Further, the differentiated design of the three INCAR files (including structure optimization, electronic state analysis, and solvent simulation) enables the phased activation of parameters, avoiding the conflict between solvent parameters and non-solvent calculations, and ensuring the optimal configuration of parameters in different scenarios. This design of one script adapting to multiple scenarios can switch the calculation target without manual modification of the configuration file, solving the consistency problem caused by repeated parameter adjustment in traditional simulation.

[0094] In addition, the fine adjustment of INCAR-3 to SIGMA (0.03) in the script can reduce the artificial broadening error of the electronic state in the solvent environment, making the energy level distribution of the solvated adsorbed species closer to the experimental observation, which is a detail optimization that cannot be achieved by the uniform parameter template in traditional high-throughput screening. The differentiated grid design of the KPOINTS file saves computing resources for low-precision optimization (KPOINTS-1), ensures data quality for high-precision electronic state analysis (KPOINTS-2), and achieves dynamic balance between efficiency and precision, rather than simply pursuing batch processing speed.

[0095] More specifically, the surface charge density distribution of the adsorbed species under different potentials is extracted simultaneously during zero-charge-point calculation to generate a three-dimensional mapping table of potential, charge polarity, and adsorption energy; in constant-potential molecular dynamics simulation, the capacitance value in the three-dimensional mapping table is used as the threshold for the standard deviation of bond length fluctuation, and the energy fluctuation reference of the adsorbed configuration is corrected by the potential of the zero-charge point, so that the three-dimensional mapping table is used not only to verify the accuracy of capacitance calculation but also to assist in optimizing the dynamic parameters for stability evaluation, finally outputting the correlation curve between electrochemical performance and stability, and using the correlation curve to accurately label the stable region of the Pourbaix diagram.

[0096] In one specific embodiment of the present scheme, during zero-charge-point calculation, the surface charge density distribution under different potentials is extracted from the LOCPOT file of the VASP software, and a Python script is called to calculate the charge polarity, which includes positive / negative / neutral. Combined with the adsorption energy data from the preliminary screening results of the Pourbaix diagram in step S40, a three-dimensional mapping table of potential (-1.5V to 1.5V), charge polarity, and adsorption energy is generated, and the three-dimensional mapping table is stored in CSV format, containing 200 groups of data with a potential step of 0.1V.

[0097] In constant-potential molecular dynamics simulation, the script automatically reads the capacitance value (cap) in the three-dimensional mapping table, and sets the threshold for the standard deviation of bond length fluctuation to cap x 5%; at the same time, the potential of the zero-charge point (pzc) in the table is extracted to correct the energy fluctuation reference of the adsorbed configuration, i.e. energy fluctuation value = model energy - reference energy corresponding to pzc.

[0098] In the generation of the correlation curve, the electrochemical performance parameters (including pzc, cap) and the stability parameters (including bond length fluctuation, energy fluctuation) are fitted into smooth curves by Origin software, and the script embeds the curve data into the stability region boundary of the Bode plot (such as marking "cap=0.12 F / g" beside the boundary line, "), to achieve quantitative labeling.

[0099] The above is only an embodiment of the present application, and the well-known specific structures and characteristics in the scheme are not described in detail. It should be pointed out that for those skilled in the art, without departing from the structure of the present application, a number of modifications and improvements can be made, which should also be considered as the protection scope of the present application, which will not affect the effect and practicality of the patent. The protection scope claimed in this application should be subject to the content of its claims, and the specific implementation mode and the like recorded in the specification can be used to explain the content of the claims.

Claims

1. A method for dynamic screening of adsorbed species on the surface of a catalyst using a multi-parameter synergistic approach, characterized in that, Includes the following steps: S10: Construct an initial crystal structure model of the target material using VESTA software, and obtain an initial crystal structure model that matches the characteristics of the actual material through atomic coordinate optimization and unit cell parameter adjustment; S20: The initial crystal structure model is optimized using density functional theory. The optimization process includes setting the calculation precision, calculating the electronic structure, optimizing the geometric structure, and considering the correction of van der Waals forces, ultimately obtaining a stable crystal structure model. S30: The stable crystal structure model constructs a solvent environment through an implicit solvent model to simulate solvent effects, by adjusting H... + / OH- ion concentration or surface protonation state covers the pH range of 0-14 to simulate pH value, and a target potential of -1.5V to 1.5V is applied by constant potential algorithm to simulate potential conditions, so as to obtain a crystal structure model integrating actual working condition parameters; S40: Plot a Boubai diagram with pH as the horizontal axis and potential as the vertical axis for the crystal structure model, and mark the actual adsorbed species corresponding to each stable region in the Boubai diagram. Analyze the mapping relationship between pH, potential and actual adsorbed species. S50: Based on the real adsorbed species in the mapping relationship, the zero charge point is obtained by calculating the surface charge density under different potentials and fitting the potential corresponding to the zero charge point; The correctness of the theoretical calculation results of the material's constant potential was verified by calculating the capacitance, and the electrochemical performance parameters were obtained. S60: Based on the preset target operating conditions, dominant adsorbent species are screened from real adsorbent species. The dynamic adsorption process of dominant adsorbent species on the catalyst surface is simulated by the constant potential molecular dynamics method. The energy fluctuation, diffusion barrier and bond length change parameters of the adsorption configuration are calculated. Combined with the preset stability threshold, the long-term stability evaluation results are obtained, and finally the adsorbent species on the catalyst surface that are stable under the actual reaction conditions are determined.

2. The method for dynamic screening of adsorbed species on the catalyst surface with multi-parameter synergy according to claim 1, characterized in that: During the optimization process in step S20, the specific parameters and coordination mechanisms for each sub-step are as follows: S21: Initialization phase, setting the calculation accuracy and basic parameters to suit the material properties, including whether the catalyst has magnetic properties to enable spin polarization, and determining whether to continue the calculation by calling the charge density file calculated in the previous step, reducing redundant calculations; S22: In the electronic structure calculation stage, a plane wave cutoff energy matching the material element composition is adopted. The calculation of the electronic self-consistent field is accelerated by the algorithm. The electronic occupancy number calculation method is correlated with the force convergence requirement of geometric optimization to ensure the synergistic optimization of electronic states and atomic positions. During the geometric structure optimization stage, the conjugate gradient algorithm is used to update the atomic positions, and the convergence criterion is that the atomic force reaches a stable threshold. During the iteration process, the electronic structure calculation results are periodically synchronized to avoid local energy minima. S23: Advanced processing stage, the description of weak interactions is optimized by van der Waals force correction, and specific processing is performed on strongly correlated electron systems. Solvent environment simulation is temporarily not enabled to avoid parameter conflicts with implicit solvent models. S24: Outputs optimized structural parameters, energy data, and electronic state information, providing a basis for subsequent analysis.

3. The method for dynamic screening of adsorbed species on the catalyst surface with multi-parameter synergy according to claim 2, characterized in that: In step S30, the integration of actual operating parameters is achieved through automated scripts, and the specific implementation process includes the following steps: S31: Preset the target potential range and determine key nodes. The script automatically traverses each potential node, creates an independent working directory for each node and copies the basic crystal structure file. S32: Use tools in each directory to extract the total number of valence electrons of the material, and calculate the adjusted valence electron parameters in combination with the current potential node; S33: Generate a suitable calculation configuration file based on this parameter. The configuration file integrates the electronic step convergence accuracy, ion step force convergence standard, van der Waals force correction method, strongly correlated electron processing parameters and parallel computing settings, and reserves an interface for solvent environment simulation parameters to adapt to subsequent working condition expansion. S34: Automatic matching of potential parameters, electronic structure calculation parameters and structure files is achieved through script linkage, and batch configuration of calculation environment under different potential conditions is completed to ensure the coordinated integration of actual operating condition parameters and simulation calculation settings.

4. The method for dynamic screening of adsorbed species on the catalyst surface with multi-parameter synergy according to claim 2, characterized in that: In the optimization process of step S20, the electronic structure and geometric structure are optimized in a coordinated manner by dynamically adjusting the parameters adapted to the material properties. Specifically, this includes selecting matching electronic structure calculation parameters for the material element composition, correlation electronic state convergence criteria and atomic force convergence requirements, specific corrections for weakly interacting and strongly correlated electronic systems, and isolating the solvent environment simulation parameters in the advanced processing stage to avoid conflicts with the integration of subsequent operating condition parameters.

5. The method for dynamic screening of adsorbed species on the catalyst surface with multi-parameter synergy according to claim 4, characterized in that: In step S31, the script automates traversal through a preset list of potential nodes. The directory corresponding to each node is named with the potential value. When copying the basic crystal structure file, the integrity of the file is checked simultaneously to ensure the consistency of the initial structure of each directory. The directory naming rules are mapped to the potential axis coordinates of the subsequent Boubai diagram.

6. The method for dynamic screening of adsorbed species on the catalyst surface with multi-parameter synergy according to claim 5, characterized in that: In step S32, the tool called is the material electronic structure analysis tool. The total number of valence electrons is extracted by the specified functional module, and the adjusted valence electron parameters are calculated by combining the linear correlation formula of the current potential node. These parameters are directly written into the core parameter item of the calculation configuration file to achieve dynamic matching between potential and material electronic properties.

7. The method for dynamic screening of adsorbed species on the catalyst surface with multi-parameter synergy according to claim 6, characterized in that: In step S33, the generated calculation configuration file includes multiple sub-configuration files for adaptive structure optimization, electronic state analysis, and solvent environment simulation. Each sub-configuration file is associated through a unified parameter interface. The solvent environment simulation sub-configuration file enables solvent model parameters separately, forming a phased activation method with other sub-configuration files.

8. The method for dynamic screening of adsorbed species on the catalyst surface with multi-parameter synergy according to claim 2, characterized in that: In step S22, the intermediate trajectory file generated in each ion iteration synchronously records the direction of atomic force and the rate of change of electronic state density. The intermediate trajectory file is used for both the local minimum value judgment of geometry optimization and the initial trajectory input for the constant potential molecular dynamics simulation in step S60, realizing the reuse of trajectory data that combines structure optimization and stability assessment.

9. The method for dynamic screening of adsorbed species on the catalyst surface with multi-parameter synergy according to claim 5, characterized in that: During zero-charge point calculation, the surface charge density distribution of adsorbed species under different potentials is extracted simultaneously to generate a three-dimensional mapping table of potential, charge polarity, and adsorption energy. In the constant potential molecular dynamics simulation, the capacitance value in the three-dimensional mapping table is used as the threshold for determining the standard deviation of bond length fluctuations. At the same time, the energy fluctuation benchmark of the adsorption configuration is corrected by the zero-charge point potential. This allows the three-dimensional mapping table to be used to verify the accuracy of capacitance calculations and to assist in optimizing the dynamic parameters for stability assessment. Finally, the correlation curve between electrochemical performance and stability is output, and the correlation curve is used for accurate labeling of the stable region of the Boubai diagram.

Images