Calculation method and system for metal atom dissolution at junction of battery positive electrode and electrolyte

Abstract

The invention provides a calculation method and system for metal atom dissolution at the junction of a battery positive electrode and an electrolyte. The method comprises the following steps: acquiring material information of a battery positive electrode material and electrolyte molecules, and controlling calculation parameters for calculating a simulation process; based on the material information and the calculation parameters, constructing a corresponding interface structure model of the positive electrode material and the electrolyte; performing balanced MD simulation on the positive electrode material and electrolyte interface structure model through an isothermal isobaric NPT ensemble algorithm to obtain balanced lattice parameters; carrying out constrained MD simulation on the positive electrode material and electrolyte interface structure model based on equilibrium lattice parameters through an isothermal isovolumetric NVT ensemble algorithm to obtain an average force borne by a reaction coordinate, and determining the reaction coordinate based on the distance between a metal atom dissolved out of the junction of the battery positive electrode material and the electrolyte and a reference atom; and carrying out thermodynamic integral calculation on the average force to obtain a free energy curve of dissolved-out metal atoms at the junction of the battery positive electrode material and the electrolyte.

Description

Technical Field

[0001] This application belongs to the technical field of computational simulation of battery cathode materials. More specifically, it relates to a calculation method and system for the dissolution of metal atoms at the interface between the battery cathode and the electrolyte. Background Art

[0002] Lithium-ion batteries have the advantages of high energy density, high working voltage, long cycle life, and low self-discharge, and are widely used in consumer electronics, electric vehicles, energy storage, and other fields. There are three basic components in a lithium-ion battery: the cathode, the anode, and the electrolyte. Among them, the cathode materials in lithium-ion batteries are mostly transition metal oxides or oxygen-containing salts containing lithium such as nickel, cobalt, manganese, and iron, which are the performance bottleneck and development focus of lithium-ion batteries.

[0003] Currently, a key problem faced in the development of cathode materials in lithium-ion batteries is that with the progress of charge and discharge cycles, there is significant dissolution of active transition metals at the interface between the cathode and the electrolyte. On the one hand, the dissolved transition metal ions can migrate to the surface of the anode, resulting in an increase in battery impedance. On the other hand, they will induce the restructuring of the surface structure of the cathode material and exacerbate the loss of redox active sites, causing attenuation of battery capacity. The above metal dissolution problem is particularly serious in manganese-containing cathode materials with lower costs.

[0004] Although there have been many experimental studies, the industry has not yet formed a unified understanding of the mechanism of metal dissolution at the interface between the cathode and the electrolyte, which restricts the further development of lithium-ion battery cathode materials in the direction of cost reduction and improved cycle performance. In terms of computational research, existing material calculation methods have problems such as large computational amount, too low efficiency, and insufficient generality of reaction coordinates when calculating and simulating the metal dissolution process at the interface between the cathode material and the electrolyte, hindering the development of new cathode materials. Summary of the Invention

[0005] The purpose of the embodiments of this application is to provide a calculation method and system for the dissolution of metal atoms at the interface between the battery cathode and the electrolyte, aiming to solve the technical problems of excessive computational amount, too low efficiency, and insufficient generality of reaction coordinates existing in the existing material calculation methods when calculating and simulating the metal dissolution process at the interface between the cathode material and the electrolyte.

[0006] To achieve the above purpose, according to the first aspect of this application, a calculation method for the dissolution of metal atoms at the interface between the battery cathode and the electrolyte is provided. The method includes:

[0007] Obtain the material information of the battery cathode material and the electrolyte molecules, as well as the calculation parameters for controlling the calculation simulation process;

[0008] Based on the material information and the calculation parameters, a corresponding interface structure model of the cathode material and the electrolyte is constructed;

[0009] The equilibrium molecular dynamics MD simulation is performed on the interface structure model of the cathode material and the electrolyte by the isothermal and isobaric NPT ensemble algorithm to obtain the equilibrium lattice parameters, where the equilibrium lattice parameters are used to reflect the lattice structure state when the interaction between the cathode material of the battery and the electrolyte molecules reaches equilibrium under the equilibrium MD simulation conditions;

[0010] The constrained MD simulation is performed on the interface structure model of the cathode material and the electrolyte based on the equilibrium lattice parameters by the isothermal and isochoric NVT ensemble algorithm to obtain the average force on the reaction coordinate, where the reaction coordinate is defined as the distance between the dissolved metal atom and the reference atom at the junction of the cathode material of the battery and the electrolyte;

[0011] The thermodynamic integration calculation is performed on the average force to obtain the free energy curve of the dissolved metal atom at the junction of the cathode material of the battery and the electrolyte, and the free energy curve is used to reflect the energy change during the process of the metal atom dissolving from the surface of the cathode material of the battery into the electrolyte.

[0012] Optionally, in a possible implementation manner of the first aspect, the constructing a corresponding interface structure model of the cathode material and the electrolyte based on the material information and the calculation parameters includes:

[0013] Based on the material information and the calculation parameters, a symmetric surface structure model based on the Miller index of the cathode material surface is constructed, where there are an upper vacuum layer and a lower vacuum layer on both sides of the cathode material surface, and the same number of electrolyte molecules are added to the upper vacuum layer and the lower vacuum layer, and the number of the electrolyte molecules is calculated based on the volume of the vacuum layer and the density of the electrolyte;

[0014] The structure optimization process is performed on the symmetric surface structure model by the machine learning force field to obtain the corresponding interface structure model of the cathode material and the electrolyte, where the structure optimization process is used to eliminate the unstable contact phenomenon between atoms introduced during the modeling process, and the machine learning force field is a force field obtained by training using a first-principles atomic calculation data set through a machine learning method, and the first-principles atomic calculation data set includes: multiple atomic structures of the cathode material of the battery generated during the first-principles calculation process and their calculation results, and the output results of the machine learning force field are the total energy, the force on each atom, and the lattice stress of the symmetric surface structure model.

[0015] Optionally, in a possible implementation of the first aspect, the equilibrium molecular dynamics (MD) simulation of the interface structure model between the positive electrode material and the electrolyte by the isothermal-isobaric NPT ensemble algorithm to obtain the equilibrium lattice parameters includes:

[0016] During the process of performing the equilibrium molecular dynamics (MD) simulation, monitor the changes in the average potential energy, temperature, and lattice constant of the interface structure model between the positive electrode material and the electrolyte with the simulation time, and determine the unbalanced simulation duration when the interface structure model between the positive electrode material and the electrolyte is in an unbalanced state;

[0017] Based on the unbalanced simulation duration and the total length of the MD simulation trajectory, calculate the equilibrium lattice parameters and uncertainties of the interface structure model between the positive electrode material and the electrolyte.

[0018] Optionally, in a possible implementation of the first aspect, calculating the equilibrium lattice parameters and uncertainties of the interface structure model between the positive electrode material and the electrolyte based on the unbalanced simulation duration and the total length of the MD simulation trajectory includes:

[0019] Obtain the unbalanced simulation duration and the total length of the MD simulation trajectory of the interface structure model between the positive electrode material and the electrolyte;

[0020] Based on the unbalanced simulation duration and the total length of the MD simulation trajectory, calculate the remaining simulation trajectory length;

[0021] Divide the remaining simulation trajectory length into multiple trajectory segments;

[0022] Calculate the average value of the lattice parameters corresponding to each of the multiple trajectory segments to obtain multiple average values of the lattice parameters;

[0023] Calculate the average value of the multiple average values of the lattice parameters to obtain the equilibrium lattice parameters, and calculate the variance of the multiple average values of the lattice parameters to obtain the uncertainty of the equilibrium lattice parameters.

[0024] Optionally, in a possible implementation of the first aspect, the constrained MD simulation includes: MD simulation with a changing constraint value of the reaction coordinate and MD simulation with a fixed constraint value of the reaction coordinate. The MD simulation with a changing constraint value of the reaction coordinate is used to continuously increase the constraint value of the reaction coordinate during the simulation to provide the interface structure during the dissolution process of metal atoms; the MD simulation with a fixed constraint value of the reaction coordinate is used to perform configuration sampling on the interface structure under fixed reaction coordinate conditions and calculate the average force exerted on the reaction coordinate.

[0025] Optionally, in a possible implementation manner of the first aspect, the constrained MD simulation of the interface structure model between the positive electrode material and the electrolyte based on the equilibrium lattice parameter by using the isothermal isochoric NVT ensemble algorithm to obtain the average force exerted on the reaction coordinate includes:

[0026] During the MD simulation in which the constraint value of the reaction coordinate changes, the coordinates and velocities of the dissolved metal atoms are adjusted based on an improved interatomic distance constraint algorithm so that the distance between the metal atoms and the reference atoms is equal to the constraint target value, and the constraint target value of the reaction coordinate is continuously adjusted so that the reaction coordinate continuously increases after being balanced at the initial value for a predetermined time;

[0027] Based on the interface structure during the dissolution process of the metal atoms provided by the MD simulation in which the constraint value of the reaction coordinate changes, an initial structure is intercepted, where the initial structure is used for the MD simulation with the position of the fixed reference atoms and the constraint value of the reaction coordinate fixed;

[0028] After the MD simulation with the constraint value of the reaction coordinate fixed, based on the unbalanced simulation duration and the total length of the MD simulation trajectory, the average force exerted on the reaction coordinate and the uncertainty of the average force exerted on the reaction coordinate are calculated.

[0029] Optionally, in a possible implementation manner of the first aspect, the calculating the average force exerted on the reaction coordinate and the uncertainty of the average force exerted on the reaction coordinate based on the unbalanced simulation duration and the total length of the MD simulation trajectory includes:

[0030] Based on the unbalanced simulation duration and the total length of the MD simulation trajectory, the remaining simulation trajectory length is calculated;

[0031] The remaining simulation trajectory length is divided into multiple trajectory segments;

[0032] The ensemble average value corresponding to each of the multiple trajectory segments is calculated to obtain multiple ensemble average values;

[0033] The average value of the multiple ensemble average values is calculated to obtain the average force exerted on the reaction coordinate, and the variance of the multiple ensemble average values is calculated to obtain the uncertainty of the average force exerted on the reaction coordinate.

[0034] Optionally, in a possible implementation manner of the first aspect, the material information includes at least one of the following: the crystal structure of the positive electrode material, the Miller index on the surface of the positive electrode material, the composition and ratio of the electrolyte molecules, the type of the dissolved metal atoms, and the calculation parameters include at least one of the following: temperature, pressure, time step, trajectory length, equilibrium time, reaction coordinate range, and sampling interval.

[0035] According to a second aspect of the present application, a calculation system for the dissolution of metal atoms at the interface between a battery positive electrode and an electrolyte is provided. The calculation system includes: an input module, a structure construction module, a molecular dynamics (MD) simulation module, a post-processing module, and an output module, where:

[0036] The input module is configured to obtain material information of the battery positive electrode material and the electrolyte molecules, as well as calculation parameters for controlling the calculation simulation process;

[0037] The structure construction module is connected to the input module and is configured to construct a corresponding interface structure model of the positive electrode material and the electrolyte based on the material information and the calculation parameters;

[0038] The MD simulation module is connected to the structure construction module and is configured to perform equilibrium MD simulation on the interface structure model of the positive electrode material and the electrolyte through the isothermal-isobaric NPT ensemble algorithm to obtain equilibrium lattice parameters, and perform constrained MD simulation on the interface structure model of the positive electrode material and the electrolyte based on the equilibrium lattice parameters through the isothermal-isochoric NVT ensemble algorithm to obtain the average force exerted on the reaction coordinate. The equilibrium lattice parameters are used to reflect the lattice structure state when the interaction between the battery positive electrode material and the electrolyte molecules reaches equilibrium under the equilibrium MD simulation conditions. The reaction coordinate is defined as the distance between the dissolved metal atoms at the interface between the battery positive electrode material and the electrolyte and the reference atoms;

[0039] The post-processing module is connected to the MD simulation module and is configured to perform thermodynamic integration calculation on the average force to obtain the free energy curve of the dissolved metal atoms at the interface between the battery positive electrode material and the electrolyte;

[0040] The output module is connected to the post-processing module and is configured to output the free energy curve, which is used to reflect the energy change during the process of the metal atoms dissolving from the surface of the battery positive electrode material into the electrolyte.

[0041] The second aspect and any implementation manner of the second aspect respectively correspond to the first aspect and any implementation manner of the first aspect. The technical effects corresponding to the second aspect and any implementation manner of the second aspect can refer to the technical effects corresponding to the first aspect and any implementation manner of the first aspect above, and will not be elaborated here.

[0042] According to a third aspect of the present application, an electronic device is provided, including: a memory, a processor, and a computer program stored in the memory and executable on the processor. When the processor executes the computer program, the electronic device implements the method as described in any one of the above.

[0043] According to the fourth aspect of the present application, there is provided a computer-readable storage medium storing a computer program which, when executed by a processor, implements the method described in any one of the above.

[0044] According to the fifth aspect of the present application, there is provided a computer program product which, when running on an electronic device, causes the electronic device to execute the method described in any one of the above first aspect.

[0045] It can be understood that the beneficial effects of the above second aspect to fifth aspect can be referred to the relevant descriptions in the above first aspect, and will not be elaborated here.

[0046] The purpose of the embodiments of the present application is to provide a calculation method and system for the dissolution of metal atoms at the interface between the battery positive electrode and the electrolyte. The method includes: obtaining the material information of the battery positive electrode material and the electrolyte molecules, as well as the calculation parameters for controlling the calculation simulation process; constructing a corresponding interface structure model of the positive electrode material and the electrolyte based on the material information and the calculation parameters; performing equilibrium molecular dynamics MD simulation on the interface structure model of the positive electrode material and the electrolyte through the isothermal-isobaric NPT ensemble algorithm to obtain the equilibrium lattice parameters, where the equilibrium lattice parameters are used to reflect the lattice structure state when the interaction between the battery positive electrode material and the electrolyte molecules reaches equilibrium under the equilibrium MD simulation conditions; performing constrained MD simulation on the interface structure model of the positive electrode material and the electrolyte based on the equilibrium lattice parameters through the isothermal-isochoric NVT ensemble algorithm to obtain the average force on the reaction coordinate, and the reaction coordinate is defined as the distance between the dissolved metal atoms and the reference atoms at the interface between the battery positive electrode material and the electrolyte; performing thermodynamic integration calculation on the average force to obtain the free energy curve of the dissolution of metal atoms at the interface between the battery positive electrode material and the electrolyte, and the free energy curve is used to reflect the energy change during the process of metal atoms dissolving from the surface of the battery positive electrode material into the electrolyte.

[0047] The improvement of the reaction coordinate selection method in the examples of the present application makes the types of positive electrode materials that can be processed by the calculation no longer limited by the interface structure, and has stronger versatility. Using the machine learning force field to replace the first-principles calculation based on density functional theory DFT avoids the complex quantum chemistry calculation process, improves the efficiency of the calculation simulation, and thus better accurately calculates and analyzes the metal dissolution process at the interface between the positive electrode material and the electrolyte. It can solve the technical problems of excessive calculation amount, too low efficiency and insufficient versatility of the reaction coordinate when calculating and simulating the metal dissolution process at the interface between the positive electrode material and the electrolyte. Description of the Drawings

[0048] To more clearly illustrate the technical solutions in the embodiments of the present application, the following will briefly introduce the accompanying drawings required for use in the embodiments or the description of the prior art. Obviously, the accompanying drawings in the following description are only some embodiments of the present application. For those of ordinary skill in the art, without creative efforts, other accompanying drawings can be obtained based on these drawings.

[0049] Figure 1 It is a schematic flowchart of a calculation method for the dissolution of metal atoms at the interface between the battery positive electrode and the electrolyte provided by an embodiment of the present application;

[0050] Figure 2 It is a schematic flowchart of an alternative calculation method for the dissolution of metal atoms at the interface between the battery positive electrode and the electrolyte provided by an embodiment of the present application;

[0051] Figure 3 It is a schematic diagram of an alternative average force curve and free energy curve provided by an embodiment of the present application;

[0052] Figure 4 It is a schematic diagram of an alternative intercepted frame of the interface conformation provided by an embodiment of the present application;

[0053] Figure 5 It is a schematic diagram of another alternative average force curve and free energy curve provided by an embodiment of the present application;

[0054] Figure 6 It is a schematic diagram of another alternative intercepted frame of the interface conformation provided by an embodiment of the present application;

[0055] Figure 7 It is a schematic diagram of an alternative average force curve and free energy curve provided by Comparative Example 1;

[0056] Figure 8 It is a schematic diagram of an alternative intercepted frame of the interface conformation provided by Comparative Example 1;

[0057] Figure 9 It is a schematic structural diagram of a calculation system for the dissolution of metal atoms at the interface between the battery positive electrode and the electrolyte provided by an embodiment of the present application;

[0058] Figure 10 It is a schematic structural diagram of an electronic device provided by an embodiment of the present application. Detailed implementation manners

[0059] In the following description, specific details such as specific system architectures, technologies, etc. are presented for the purpose of illustration rather than limitation, so as to thoroughly understand the embodiments of the present application. However, those skilled in the art should clearly understand that the present application can also be implemented in other embodiments without these specific details. In other cases, detailed descriptions of well-known systems, devices, circuits, and methods are omitted to avoid unnecessary details from interfering with the description of the present application.

[0060] It should be understood that when used in the specification of the present application and the appended claims, the term "comprising" indicates the presence of the described features, wholes, steps, operations, elements, and / or components, but does not exclude the presence or addition of one or more other features, wholes, steps, operations, elements, components, and / or their combinations.

[0061] It should also be understood that the term "and / or" used in the specification of the present application and the appended claims refers to any combination and all possible combinations of one or more of the associated listed items, and includes these combinations.

[0062] As used in the specification of the present application and the appended claims, the term "if" can be interpreted as "when" or "once" or "in response to determining" or "in response to detecting" according to the context. Similarly, the phrase "if determined" or "if detecting [the described condition or event]" can be interpreted as meaning "once determined" or "in response to determining" or "once detecting [the described condition or event]" or "in response to detecting [the described condition or event]" according to the context.

[0063] In addition, in the description of the specification of the present application and the appended claims, the terms "first", "second", "third", etc. are only used for distinguishing descriptions and cannot be understood as indicating or implying relative importance.

[0064] The reference to "one embodiment" or "some embodiments" etc. described in the specification of the present application means that a specific feature, structure, or characteristic described in connection with the embodiment is included in one or more embodiments of the present application. Thus, statements such as "in one embodiment", "in some embodiments", "in other some embodiments", "in still other embodiments", etc. that appear in different places in this specification do not necessarily all refer to the same embodiment, but mean "one or more but not all embodiments", unless otherwise specifically emphasized in other ways. The terms "comprising", "including", "having", and their variants all mean "including but not limited to", unless otherwise specifically emphasized in other ways.

[0065] First, some terms in the embodiments of the present application are explained to facilitate the understanding of those skilled in the art.

[0066] Molecular dynamics (MD) simulation: For a material system composed of atoms and molecules, the evolution of atomic motion is simulated according to Newton's three laws of classical physics. During the MD simulation, time is discretized. At each time step, the forces acting on each atom are calculated, and then the velocities and coordinates of the atoms are updated. Repeating this step forms a simulation trajectory. MD simulation is an important and commonly used technique for calculating the dynamic and thermodynamic properties of a material system.

[0067] Reaction coordinate: The characteristic degree of freedom used to express the progress of a certain physical and chemical process of a system. Along the reaction coordinate, the system switches from one state to another. For example, in a chemical reaction, along the reaction coordinate, the system changes from reactants to products; during the lithium-ion migration process in the cathode material of a lithium-ion battery, along the reaction coordinate, Li + migrates from one site to another. The reaction coordinate can also be called the collective variable (CV).

[0068] First-principles calculation: It refers to calculating materials without relying on any empirical or experimentally derived parameters, but starting only from the most basic principles of quantum mechanics that electrons satisfy, to obtain the physical and chemical properties of materials. The input of first-principles calculation is the atomic structure of the material, and the output includes the electronic structure information of the material, as well as the total energy of the system, atomic forces, and other information that can be deduced therefrom. First-principles calculation has high accuracy and a wide range of applications, but it has a large amount of calculation and low efficiency.

[0069] Density functional theory (DFT): The mainstream first-principles calculation method applied in the field of materials.

[0070] Force field: The explicit functional relationship between the total energy of a material and the atomic forces as a function of atomic coordinates. Using this functional relationship, molecular dynamics simulation can be performed on the material to obtain the physical and chemical properties related to thermodynamics and kinetics of the material. Because this functional relationship is a simple operation, the calculation speed of the force field is fast. The force field can be obtained by fitting experimental data, which is called an empirical force field; it can also be obtained by fitting first-principles calculation data. Using the latter can inherit the high accuracy of first-principles calculation.

[0071] Machine learning force field: A method of using machine learning to fit first-principles calculation data to obtain the force field of materials. For traditional force fields, the functional form is artificially designed. After adopting the machine learning method, this functional form does not need to be explicitly specified but is automatically learned from the data set. Therefore, the machine learning force field has a wider application range and stronger generalization ability than traditional force fields.

[0072] Neural network force field: A machine learning force field fitted using a neural network. A neural network is a major category in machine learning methods, and its input-to-output process passes through a network composed of artificial neurons. The fitting ability of neural networks is strong and has become the absolute mainstream in the field of machine learning in recent years, and is also known as deep learning.

[0073] Pretraining: A way to train large-scale neural network models. First, train the model parameters on a large-scale and multi-type data set to learn general representation features; then fine-tune the parameters of the latter part of the model on a small data set related to a certain downstream task to improve the prediction accuracy of specific tasks.

[0074] Structural relaxation: By changing the atomic coordinates of materials, the total energy of the system is minimized. Structural relaxation is a basic operation commonly used in material calculations and is also called structural relaxation.

[0075] The cathode material is a crystal material. The most essential feature of a crystal material is translational periodicity in three directions, that is, a unit cell in the shape of a parallelepiped is repeatedly stacked along three directions to form a crystal material. The unit cell is the repeating unit of the material.

[0076] The primitive cell is a professional term in materials science and crystallography. The primitive cell is the smallest repeating unit that satisfies translational periodicity of the material, that is, the primitive cell is the smallest unit cell.

[0077] A supercell refers to a larger unit cell obtained by expanding the primitive cell and performing some other operations.

[0078] Miller index: In crystallography, in order to determine the direction of a certain crystal plane in a crystal, the Miller index is used for marking. The Miller index is the ratio of the reciprocals of the intercepts of the crystal plane on the three coordinate axes (usually the crystal axes) as relatively prime integers.

[0079] The above is a simple introduction to the terms involved in the embodiments of this application, and will not be elaborated below.

[0080] Currently, a key problem faced in the development of cathode materials for lithium-ion batteries is that with the progress of charge-discharge cycles, significant dissolution of active transition metals occurs at the interface between the cathode and the electrolyte. On the one hand, the dissolved transition metal ions can migrate to the surface of the anode, leading to an increase in battery impedance. On the other hand, they will induce the restructuring of the surface structure of the cathode material and exacerbate the loss of redox active sites, resulting in the attenuation of battery capacity. This metal dissolution problem is particularly severe in low-cost manganese-containing cathode materials, such as lithium manganate (LiMn 2 O 4 , LMO), lithium manganese iron phosphate (LiMn x Fe y PO 4 , x + y = 1, LMFP), etc.

[0081] Material calculations represented by the density functional theory (DFT) method are not restricted by experimental conditions. They can calculate and simulate the microscopic structure of materials to obtain the macroscopic physical and chemical properties of materials, and have developed into a key technical means for new material development on a par with experimental exploration.

[0082] This type of material calculation research combines first-principles calculations based on DFT with enhanced sampling molecular dynamics simulation techniques, realizing the microscopic mechanism simulation of the metal dissolution process at the atomic level, and playing an important role in understanding the influencing factors of the metal dissolution process. However, the current material calculation methods still have the following two main disadvantages. First, metal dissolution is a slow process occurring at the heterogeneous interface. Although assisted by enhanced sampling methods, long-time configuration sampling of complex interface models is still required. The first-principles method based on DFT has a large computational amount and low computational efficiency, so the time and resource costs required for calculation are too high. Second, the reaction coordinate describing the dissolution process in the current calculation method is defined by the bond length between the dissolved metal and the coordinated oxygen atoms. This is only applicable to materials with a relatively regular surface structure such as spinel-structured lithium manganate LMO and lithium nickel manganese oxide LiNi 0.5 Mn 1.5 O 4 (LNMO). It has poor applicability at the interface between general cathode materials and electrolytes and cannot accurately describe the metal dissolution process.

[0083] Therefore, there is an urgent need for a material calculation method that can efficiently calculate and simulate the metal dissolution process at the interface between the cathode material and the electrolyte, so as to promote the development of new cathode materials with lower costs and better cycling performance.

[0084] This application example provides an example of a calculation method for metal atom dissolution at the junction of the battery cathode and the electrolyte. Please refer to Figure 1 as shown Figure 1The figure shows a schematic flow chart of a calculation method for the dissolution of metal atoms at the interface between the positive electrode and the electrolyte provided by the present application. As an example but not a limitation, this method can be applied to or run in a calculation system for the dissolution of metal atoms at the interface between the positive electrode and the electrolyte. The method includes:

[0085] S101, obtaining the material information of the positive electrode material and the electrolyte molecules, as well as the calculation parameters for controlling the calculation simulation process.

[0086] S102, constructing a corresponding interface structure model of the positive electrode material and the electrolyte based on the material information and the calculation parameters.

[0087] S103, performing equilibrium molecular dynamics MD simulation on the interface structure model of the positive electrode material and the electrolyte through the isothermal and isobaric NPT ensemble algorithm to obtain the equilibrium lattice parameters, where the equilibrium lattice parameters are used to reflect the lattice structure state when the interaction between the positive electrode material and the electrolyte molecules reaches equilibrium under the equilibrium MD simulation conditions.

[0088] S104, performing constrained MD simulation on the interface structure model of the positive electrode material and the electrolyte based on the equilibrium lattice parameters through the isothermal and isochoric NVT ensemble algorithm to obtain the average force exerted on the reaction coordinate, where the reaction coordinate is defined as the distance between the dissolved metal atoms at the interface between the positive electrode material and the electrolyte and the reference atoms.

[0089] S105, performing thermodynamic integration calculation on the average force to obtain the free energy curve of the dissolution of metal atoms at the interface between the positive electrode material and the electrolyte, and the free energy curve is used to reflect the energy change during the process of metal atoms dissolving from the surface of the positive electrode material into the electrolyte.

[0090] The example of the present application can be applied to the field of calculation simulation of positive electrode materials for batteries, specifically involving the calculation simulation of metal dissolution at the interface between the positive electrode material and the electrolyte, and can be used for the development of positive electrode materials for high-capacity and high-cycle retention lithium-ion batteries or sodium-ion batteries.

[0091] Optionally, in the example of the present application, the material information includes at least one of the following: the crystal structure of the positive electrode material, the Miller indices on the surface of the positive electrode material, the composition and ratio of the electrolyte molecules, the type of dissolved metal atoms, and the calculation parameters include at least one of the following: temperature, pressure, time step, trajectory length, equilibrium time, reaction coordinate range, and sampling interval.

[0092] Specifically, the material information describes the specific characteristics of the cathode material and electrolyte molecules of the battery. For example, the crystal structure of the cathode material, surface characteristics (such as surface structure information related to Miller indices, etc.), elements contained, etc., as well as information such as the composition and ratio of electrolyte molecules. Since different material properties affect the interaction between the cathode material and electrolyte molecules at the interface and processes such as metal dissolution, the material information is crucial for accurately constructing the subsequent interface structure model of the cathode material and electrolyte.

[0093] In addition, reasonable setting of calculation parameters can make the simulation more in line with the actual situation and efficient. Calculation parameters are various conditions for precisely setting and controlling the entire calculation simulation process. For example, calculation parameters related to temperature, pressure, time step, trajectory length, etc. Temperature and pressure affect the thermodynamic state of the interface between the cathode material and electrolyte, and thus affect the behavior of the cathode material and electrolyte molecules; the time step determines the calculation accuracy, time resolution, and calculation efficiency of the MD simulation; the trajectory length is related to whether the configurational space sampling of the NPT and NVT ensembles is sufficient.

[0094] In the example of this application, using the detailed material information of the battery cathode material and electrolyte molecules obtained, combined with the set calculation parameters, through a series of steps, an interface structure model of the cathode material and electrolyte that can accurately reflect the actual situation is constructed. For example, starting from the given primitive cell of the cathode material, a symmetric surface structure model is constructed according to the predetermined Miller indices, a vacuum layer with a certain thickness is set on both sides of its surface, and then according to the information such as the density of the electrolyte and the volume of the vacuum layer, the appropriate number of electrolyte molecules is determined and added to the vacuum layer according to the calculation formula. Finally, the machine learning force field can also be used to optimize the structure of the constructed interface structure model, eliminate possible unstable atomic contacts, and make the structure model more reasonable and stable.

[0095] In the example of this application, the isothermal-isobaric NPT ensemble algorithm means that during the simulation, the number of atoms, pressure, and temperature of the interface structure model of the cathode material and electrolyte remain unchanged. During the simulation, physical quantities such as the average potential energy, temperature, and lattice constant of the interface structure model of the cathode material and electrolyte are monitored to determine whether the interface structure model of the cathode material and electrolyte reaches an equilibrium state.

[0096] When the interface structure model of the cathode material and electrolyte reaches equilibrium, the obtained equilibrium lattice parameters can accurately describe the lattice structure state when the interaction between the battery cathode material and electrolyte molecules reaches equilibrium under the current simulation conditions. In the subsequent isothermal-isochoric NVT ensemble MD simulation, the lattice parameters of the initial structure are set to the equilibrium lattice parameters and remain unchanged during the subsequent simulation process.

[0097] In the examples of this application, during the simulation of the isothermal and isochoric NVT ensemble, the number of atoms, volume, and temperature of the interface structure model between the cathode material and the electrolyte remain unchanged. Based on the equilibrium lattice parameters obtained from the above-mentioned NPT ensemble simulation, constrained MD simulation of the interface structure model between the cathode material and the electrolyte is carried out under this ensemble. During the simulation, constraints are imposed on a specific reaction coordinate (i.e., the distance between the dissolved metal atom and the reference atom at the junction of the cathode material and the electrolyte of the battery), and an improved constraint algorithm (such as the improvement of the RATTLE constraint algorithm) is used to ensure the accuracy and rationality of the simulation. Through the above simulation, the dynamic behavior of the interface structure model between the cathode material and the electrolyte and the average force exerted on the reaction coordinate can be observed as the reaction coordinate changes (due to the dissolution process of metal atoms) under fixed volume and temperature conditions.

[0098] In the examples of this application, the selection method of the reaction coordinate can more generally describe the process of metal dissolution at the interface between the cathode material and the electrolyte, and can overcome the limitations of some existing selection methods (such as the limitation of using the bond length between metal atoms and coordination atoms as the reaction coordinate). By imposing constraints on the reaction coordinate for simulation, the average force exerted on the reaction coordinate obtained can be further used to analyze the mechanical properties during the metal dissolution process. The average force exerted on the reaction coordinate reflects the average force exerted during the metal atom dissolution process due to the interaction between the cathode material and the electrolyte.

[0099] Thermodynamic integration calculations are performed on the average force exerted on the reaction coordinate obtained through the NVT ensemble constrained MD simulation. It should be understood that this is a mathematical operation process based on physical principles. By integrating the average force with respect to the reaction coordinate, a function of the reaction coordinate can be obtained, and this function is the required free energy curve. The free energy curve can intuitively reflect the energy change during the process of metal atoms in the cathode material of the battery dissolving into the electrolyte, and can also be used to judge the difficulty of metal dissolution. If the curve rises steeply, it indicates that the metal dissolution needs to overcome a high energy barrier and the dissolution is relatively difficult; otherwise, the dissolution is relatively easy. In addition, the free energy curve can also be used to guide the material screening and modification optimization of the cathode material and its interface with the electrolyte, etc. For example, by comparing the free energy curves under different material combinations, select the material combination that makes the metal dissolution process more favorable (such as difficult dissolution, which is beneficial to battery performance and stability) for battery research and development, etc.

[0100] Through the above calculation method provided by this application, it is possible to effectively solve the technical problems in the prior art that when calculating and simulating the metal dissolution process at the interface between the cathode material and the electrolyte, the calculation amount is too large, the efficiency is too low, and the generality of the reaction coordinate is insufficient. For example, by using the machine learning force field for structural optimization and MD simulation calculations, combined with a reasonably determined reaction coordinate and other solutions, the calculation efficiency is improved, and the selection of the reaction coordinate is more general, so as to better accurately calculate and analyze the metal dissolution process at the interface between the cathode material and the electrolyte.

[0101] In a possible implementation, please refer to Figure 2 as shown Figure 2 Fig. shows a schematic flow chart of a calculation method for the dissolution of metal atoms at the junction of a battery cathode and an electrolyte provided by this application. Based on material information and calculation parameters, a corresponding interface structure model of the cathode material and the electrolyte is constructed, including:

[0102] S201, based on material information and calculation parameters, construct a symmetric surface structure model based on the Miller indices of the cathode material surface.

[0103] Among them, there is an upper vacuum layer and a lower vacuum layer on both sides of the cathode material surface. The same number of electrolyte molecules are added to the upper vacuum layer and the lower vacuum layer. The number of electrolyte molecules is calculated based on the volume of the vacuum layer and the density of the electrolyte.

[0104] S202, perform structural optimization on the symmetric surface structure model through the machine learning force field to obtain the corresponding interface structure model of the cathode material and the electrolyte.

[0105] Among them, the structural optimization is used to eliminate the unstable atomic contacts introduced during the modeling process. The machine learning force field is obtained by training using a first-principles atomic calculation data set through a machine learning method. The first-principles atomic calculation data set includes: the atomic structures and calculation results of multiple battery cathode materials generated during the first-principles calculation process. The output results of the machine learning force field are the total energy, the force on each atom, and the lattice stress of the symmetric surface structure model.

[0106] Optionally, the machine learning force field is not limited to a specific model. For example, it may include, but is not limited to: the material graph neural network M3GNet considering three-body interactions, the general neural network potential function "Preferred Potential" (PFP), the deep potential energy model DPA with an attention mechanism, the crystal Hamiltonian graph neural network CHGNet, the atomic wire graph neural network force field ALIGNN-FF, and so on.

[0107] In the examples of this application, Miller indices are used to describe the crystallographic orientation and structural characteristics of the surface of the cathode material. Through the given Miller indices of the surface of the cathode material, the specific crystal plane directions and atomic arrangements of the symmetric surface structure model to be constructed can be determined. Different Miller indices correspond to different crystal planes, and the properties of different crystal planes (such as atomic density, chemical activity, etc.) will vary.

[0108] In an alternative example, in the constructed symmetric surface structure model, an upper vacuum layer and a lower vacuum layer are respectively arranged on both sides of the surface of the cathode material. It should be understood that the existence of these two vacuum layers is to more realistically present the interfacial environment between the cathode material and the electrolyte in the simulation. Moreover, the same number of electrolyte molecules are added to these two vacuum layers. Specifically, the number of electrolyte molecules is calculated based on the volume of the vacuum layer and the density of the electrolyte. Through this implementation method, the electrolyte components can be reasonably introduced into the model to form a complete interfacial structure model with the surface of the cathode material for subsequent studies of their interactions.

[0109] For example, the number n of electrolyte molecules is calculated according to the volume of the vacuum layer and the density of the electrolyte, and the calculation formula used is n = ρVN A / M, where ρ is the density of the electrolyte, V is the volume of the vacuum layer, N A is Avogadro's constant, and M is the molar mass of the electrolyte molecule.

[0110] In the examples of this application, the machine learning force field is a force field obtained by training through machine learning methods. Its training data comes from the first-principles atomic calculation dataset, which is generated during the first-principles calculation process. First-principles calculation starts from the most basic physical laws (such as laws related to quantum mechanics) and does not rely on empirical parameters to predict various properties of materials. During this process, a large amount of data on atomic structure, energy, force, etc. will be generated, and these data constitute the first-principles calculation dataset. For the trained machine learning force field, its independent variable is the atomic structure of the material, and the dependent variables are the total energy of the system and the forces on each atom. This means that when the specific atomic structure information of the cathode material is input, the machine learning force field can output the corresponding total energy of the model and the atomic force conditions.

[0111] Further, after constructing the symmetric surface structure model, a machine learning force field is used to optimize its structure to eliminate the unstable atomic contacts introduced during the modeling process. It should be noted that during the model construction process, due to various reasons (such as initial settings, atomic arrangements, etc.), some atomic contacts may not conform to the actual physical or chemical situations. These unstable contacts may affect the stability of subsequent MD simulation calculations. Through the structural optimization using the machine learning force field, the positions of each atom are continuously adjusted according to the force on each atom, so that the force on each atom is continuously reduced and the total energy of the system is continuously decreased. When the force on each atom is less than a set standard value, the structural optimization ends, thus obtaining a more reasonable and stable interface structure model of the cathode material and the electrolyte.

[0112] In a possible implementation, the equilibrium molecular dynamics MD simulation is performed on the interface structure model of the cathode material and the electrolyte through the isothermal-isobaric NPT ensemble algorithm to obtain the equilibrium lattice parameters, including:

[0113] During the process of performing the equilibrium molecular dynamics MD simulation, monitor the changes of the average potential energy, temperature, and lattice constant of the interface structure model of the cathode material and the electrolyte with the simulation time, and determine the unbalanced simulation duration when the interface structure model of the cathode material and the electrolyte is in an unbalanced state.

[0114] Based on the unbalanced simulation duration and the total length of the MD simulation trajectory, the equilibrium lattice parameters and uncertainties of the interface structure model of the cathode material and the electrolyte are calculated.

[0115] In the example of this application, the average potential energy reflects the energy situation of the interaction between atoms in the system. When performing MD simulation on the interface structure model of the cathode material and the electrolyte, there are various interactions between atoms, such as electrostatic interactions, chemical bond interactions, etc. The comprehensive effect of these interactions is reflected in the average potential energy. The change in the average potential energy can reveal the change in the stability of the system structure.

[0116] In the example of this application, temperature is another important thermodynamic parameter. In the isothermal-isobaric NPT ensemble simulation, although a fixed target temperature is set, there are still fluctuations in the total kinetic energy of the system and the actual temperature calculated therefrom. The change in the system temperature is one of the signs of whether it reaches equilibrium. When reaching equilibrium, the system temperature should oscillate around the set temperature.

[0117] In addition, in the examples of this application, for substances with a crystal structure such as the cathode material, the lattice constant describes the basic dimensional characteristics of its crystal structure, such as the side length of the unit cell, etc. During the simulation process, the change in the lattice constant reflects the structural adjustment of the cathode material under the interaction with the electrolyte and under simulation conditions (such as the influence of temperature T, pressure P (which can be taken as the atmospheric pressure of 1.0 bar), etc.). By monitoring the change in the lattice constant, it can be determined whether the density of the electrolyte layer in the system has reached equilibrium under the given temperature T and pressure P conditions.

[0118] As an alternative implementation, during the process of performing equilibrium molecular dynamics MD simulation, by continuously monitoring the changes in the above-mentioned average potential energy, temperature, and lattice constant with the simulation time, it is determined whether the system (i.e., the interface structure model of the cathode material and the electrolyte) is in an equilibrium state. When the values of these physical quantities do not stabilize within a certain range but show an obvious changing trend, it indicates that the system is in an unbalanced state. Record the time t from the start of the simulation until the system reaches equilibrium equ , which is the so-called unbalanced simulation duration.

[0119] After calculating the equilibrium lattice parameter and uncertainty based on the unbalanced simulation duration and the total length of the MD simulation trajectory, combined with the known total length of the MD simulation trajectory, the simulation trajectory can be reasonably divided and processed to calculate the equilibrium lattice parameter. Specifically, usually, the part of the trajectory in the unbalanced state at the beginning of the simulation (with a length equal to the unbalanced simulation duration) is removed, and then the remaining part of the trajectory in the equilibrium state is analyzed and processed. In the remaining part of the equilibrium trajectory, by statistical methods such as calculating the average value of the lattice constant of each frame structure (similar to the idea of taking the average in segments introduced before), the equilibrium lattice parameter representing the lattice structure characteristics of the system in the equilibrium state is calculated.

[0120] As an example rather than a limitation, the uncertainty is used to measure the reliability and accuracy of the calculated equilibrium lattice parameter. During the actual simulation process, due to the influence of various factors (such as statistical fluctuations in the simulation, approximation of the model, etc.), even the lattice parameter calculated in the equilibrium state may have a certain range of fluctuations. The uncertainty is a quantitative description of this range of fluctuations. Similarly, based on the analysis of the equilibrium trajectory part, by calculating statistical quantities such as the variance of the lattice constant of each frame structure in the equilibrium state (similar to the variance of each segment of data when calculating the equilibrium lattice parameter in the above example), the uncertainty of the equilibrium lattice parameter is obtained. This uncertainty can be used as a reference to help determine whether the obtained equilibrium lattice parameter is stable and reliable enough for reasonable use of this equilibrium lattice parameter in subsequent calculations.

[0121] In a possible implementation, based on the unbalanced simulation duration and the total length of the MD simulation trajectory, the equilibrium lattice parameters and uncertainties of the interface structure model between the cathode material and the electrolyte are calculated, including:

[0122] Obtain the unbalanced simulation duration of the interface structure model between the cathode material and the electrolyte and the total length of the MD simulation trajectory;

[0123] Based on the unbalanced simulation duration and the total length of the MD simulation trajectory, calculate the remaining simulation trajectory length;

[0124] Divide the remaining simulation trajectory length into multiple trajectory segments;

[0125] Calculate the average value of the lattice parameters corresponding to each trajectory segment among the multiple trajectory segments to obtain multiple average values of the lattice parameters;

[0126] Calculate the average value of the multiple average values of the lattice parameters to obtain the equilibrium lattice parameters, and calculate the variance of the multiple average values of the lattice parameters to obtain the uncertainty of the equilibrium lattice parameters.

[0127] In the example of this application, during the process of performing equilibrium molecular dynamics (MD) simulation on the interface structure model between the cathode material and the electrolyte through the isothermal-isobaric NPT ensemble algorithm, the system is determined to be in an equilibrium state by continuously monitoring the changes of the average potential energy, temperature, and lattice constant of the interface structure model between the cathode material and the electrolyte with the simulation time. The time period from the start of the simulation to before the system reaches equilibrium is the unbalanced simulation duration. This unbalanced simulation duration records the time length that the system experiences in the non-stable and non-equilibrium state at the initial stage of the simulation, and defines the non-equilibrium part that needs to be excluded from the simulation trajectory.

[0128] Optionally, in the example of this application, the total length of the MD simulation trajectory refers to the time range covered by the entire MD simulation process from start to end, representing the complete duration of the simulation's observation of the dynamic behavior of the system. Obtaining this total length is for accurately determining the part of the simulation trajectory in the equilibrium state based on it and the unbalanced simulation duration in the subsequent steps, and then for calculating the equilibrium lattice parameters and uncertainties.

[0129] In the example of this application, the purpose of calculating the remaining simulation trajectory length is to clarify the effective trajectory part that can be analyzed to obtain the equilibrium lattice parameters and uncertainties after the system reaches the equilibrium state during the simulation process. The calculation method is to subtract the unbalanced simulation duration from the total length of the MD simulation trajectory, and the obtained difference is the remaining simulation trajectory length.

[0130] By dividing into multiple trajectory segments, statistical analysis of the lattice parameters within each segment can be performed in subsequent steps, so as to more comprehensively capture the fluctuation characteristics and average situation of the lattice parameters of the system in the equilibrium state, in order to accurately calculate the equilibrium lattice parameters and uncertainties.

[0131] For each of the segmented trajectory segments, it is necessary to count the lattice parameters of each frame structure within the segment and calculate their average value. This is because within each trajectory segment, although the system is in an equilibrium state, the lattice parameters may still have certain fluctuations. By calculating the average value, a numerical value can be obtained that represents the overall level of the lattice parameters within the trajectory segment. For example, within a trajectory segment containing 100 frame structures, each frame structure has corresponding lattice parameters. After adding these 100 lattice parameters and dividing by 100, the average value of the lattice parameters corresponding to this trajectory segment is obtained. Performing such an operation on each segmented trajectory segment can obtain multiple average values of the lattice parameters, and these average values will serve as the basic data for the next calculation of the equilibrium lattice parameters and uncertainties.

[0132] Taking the average of the multiple average values of the lattice parameters obtained in the above example again, the result is the equilibrium lattice parameter. This equilibrium lattice parameter can be regarded as a comprehensive average level of the lattice parameters over the entire remaining simulation trajectory length (i.e., the trajectory part where the system is in an equilibrium state), and it can represent the lattice structure characteristics of the system in the equilibrium state. For example, if there are 5 trajectory segments, and the corresponding average values of the lattice parameters are a 1 、a 2 、a 3 、a 4 、a 5 , the equilibrium lattice parameter is (a 1 + a 2 + a 3 + a 4 + a 5 ) / 5.

[0133] The uncertainty of the equilibrium lattice parameter is obtained by calculating the variance of the multiple average values of the lattice parameters. Since the variance is a statistic used to measure the degree of data dispersion, in the example of this application, calculating the variance of the multiple average values of the lattice parameters reflects the dispersion of the average values of the lattice parameters corresponding to each trajectory segment relative to the equilibrium lattice parameter. The larger the variance, the greater the difference between the average values of the lattice parameters of each trajectory segment, that is, the higher the uncertainty in determining the equilibrium lattice parameter. For example, for the average values of the lattice parameters a 1 、a 2 、a 3 、a 4 、a 5 of the above 5 trajectory segments, by calculating their variance, the uncertainty of the equilibrium lattice parameter can be obtained, and this uncertainty can evaluate the reliability and stability of the obtained equilibrium lattice parameter.

[0134] In one possible implementation, the constrained MD simulation includes: MD simulation with the changing constraint value of the reaction coordinate and MD simulation with the fixed constraint value of the reaction coordinate. The MD simulation with the changing constraint value of the reaction coordinate is used to continuously increase the constraint value of the reaction coordinate during the simulation to provide the interfacial structure during the dissolution process of metal atoms. The MD simulation with the fixed constraint value of the reaction coordinate is used to perform configurational sampling on the interfacial structure under the condition of a fixed reaction coordinate and calculate the average force exerted on the reaction coordinate.

[0135] When studying the dissolution process of metal atoms at the interface between the cathode material of the battery and the electrolyte, the constrained MD simulation method is adopted to more accurately explore the relevant physical and chemical processes. This constrained MD simulation is mainly divided into two categories: MD simulation with the changing constraint value of the reaction coordinate and MD simulation with the fixed constraint value of the reaction coordinate. Both are carried out based on imposing constraint conditions on specific reaction coordinates, aiming to deeply understand the changes in the interfacial structure and related physical quantities under different constraint conditions during the process of metal atoms dissolving from the cathode material into the electrolyte.

[0136] In an alternative implementation, the operation in the MD simulation with the changing constraint value of the reaction coordinate: During the simulation, continuously increase the constraint value of the reaction coordinate, which is defined as the distance between the dissolved metal atoms at the junction of the cathode material of the battery and the electrolyte and the reference atoms. As the simulation progresses, continuously increase the constraint value of this interatomic distance to simulate the process of metal atoms gradually dissolving from the cathode material and moving into the electrolyte. By continuously increasing the constraint value of the reaction coordinate, it is possible to observe how the interfacial structure changes during the dissolution process of metal atoms. Because the dissolution of metal atoms will change their interaction with the surrounding environment (including the cathode material, electrolyte, etc.), thereby affecting the structural stability and atomic arrangement of the entire interface. This simulation method can provide researchers with information on the interfacial structure at a series of different dissolution stages, which helps to deeply understand the evolution mechanism of the interfacial structure during the dissolution process of metal atoms. For example, it can be understood how the distribution of atoms and the changes in chemical bonds at the interface change as metal atoms gradually move away from the surface of the cathode material.

[0137] In another alternative implementation, the operation in the MD simulation with the fixed constraint value of the reaction coordinate: Different from the MD simulation with the changing constraint value of the reaction coordinate, during this type of simulation, the constraint value of the reaction coordinate needs to be kept fixed. Once a specific distance between the dissolved metal atoms and the reference atoms is set as the constraint value of the reaction coordinate, this interatomic distance remains at a fixed value and does not change during the entire simulation process. It should be understood that the main role of the MD simulation with the fixed constraint value of the reaction coordinate is:

[0138] At a fixed reaction coordinate, MD simulations of a certain length in the NVT ensemble are carried out to achieve sufficient statistical sampling of the interface structure between the cathode material and the electrolyte under a specific dissolution state (defined by the constraint value of the fixed reaction coordinate), so as to provide accurate data for subsequent calculations of the average force and free energy. By comparing the simulation results under different fixed reaction coordinate values, a more comprehensive understanding of the characteristics of different stages in the process of metal atom dissolution can be obtained.

[0139] In one possible implementation, through the isothermal-isochoric NVT ensemble algorithm, a constrained MD simulation of the interface structure model between the cathode material and the electrolyte is carried out based on the equilibrium lattice parameters to obtain the average force exerted on the reaction coordinate, including:

[0140] During the MD simulation with the change of the constraint value of the reaction coordinate, the coordinates and velocities of the dissolved metal atoms are adjusted based on the improved interatomic distance constraint algorithm so that the distance between the metal atoms and the reference atoms is equal to the constraint target value, and the constraint target value of the reaction coordinate is continuously adjusted so that the reaction coordinate continuously increases after equilibrating for a predetermined time at the initial value.

[0141] Based on the interface structure during the metal atom dissolution provided by the MD simulation with the change of the constraint value of the reaction coordinate, an initial structure is intercepted, where the initial structure is used for the MD simulation with the position of the fixed reference atoms and the constraint value of the reaction coordinate fixed.

[0142] After the MD simulation with the constraint value of the reaction coordinate fixed, based on the unbalanced simulation duration and the total length of the MD simulation trajectory, the average force exerted on the reaction coordinate and the uncertainty of the average force exerted on the reaction coordinate are calculated.

[0143] Optionally, in the example of this application, when performing the MD simulation with the change of the constraint value of the reaction coordinate, an improved interatomic distance constraint algorithm is used to impose constraints on the reaction coordinate, and during the MD simulation with the change of the constraint value of the reaction coordinate, the constraint target value is continuously adjusted: first, it remains at the initial value d 0 for a period of time (t 0 ) to allow the system to reach equilibrium in the NVT ensemble; then, at a speed of v, the reaction coordinate is linearly increased. Specifically, the following formula is used,

[0144]

[0145] d is the constraint target value; d 0 is the initial value of the reaction coordinate when the metal atoms have not dissolved; t 0 is the simulation duration required for the system to reach equilibrium at the initial value of the reaction coordinate; t is the time of the MD simulation trajectory; v is the growth rate of the constraint value of the reaction coordinate, and the value range can be selected as

[0146] In the MD simulation with the constrained value of the reaction coordinate changing, during the process of continuously increasing the reaction coordinate, a series of interfacial structure changes during the dissolution of metal atoms are presented. For example, the arrangement of atoms, the state of chemical bonds, etc. evolve as the metal atoms dissolve. Based on these interfacial structures during the dissolution of metal atoms generated in the simulation process, a suitable initial structure is intercepted therefrom for specifically performing MD simulations with the positions of reference atoms fixed and the constrained value of the reaction coordinate fixed. It should be understood that specifically in such subsequent simulations with fixed constrained values, starting from this specific initial structure, further sufficient statistical sampling of the interfacial structure is carried out in the NVT ensemble to more accurately calculate the average force exerted on the reaction coordinate.

[0147] The unbalanced simulation duration records the time length during which the system is in an unbalanced state in the simulation process, while the total length of the MD simulation trajectory covers the entire time range from the start to the end of the simulation.

[0148] t slow =t 0 +(d f -d 0 ) / v;

[0149] Wherein, d f is the end value of the reaction coordinate at the end of the given dissolution. This formula is used to calculate the total length t slow of the MD simulation trajectory with the constrained value of the reaction coordinate changing, which includes two parts: (1) the pre-specified unbalanced simulation duration t 0 ; (2) the time required for the reaction coordinate to increase from the initial value d 0 to the end value d f : (d f -d 0 ) / v.

[0150] After completing the MD simulation with the constrained value of the reaction coordinate fixed, the average force exerted on the reaction coordinate and its uncertainty are calculated based on the unbalanced simulation duration and the total length of the MD simulation trajectory of this part of the MD simulation. By reasonably dividing and analyzing the simulation trajectory according to the unbalanced simulation duration and the total length of the MD simulation trajectory, the force exerted on the reaction coordinate under a given dissolution degree can be statistically obtained, and then the average force exerted on the reaction coordinate can be calculated. This average force reflects the magnitude of the average force exerted on the reaction coordinate (i.e., the distance between the metal atom and the reference atom) during the dissolution of metal atoms due to its interaction with the surrounding environment (including the cathode material, electrolyte, etc.).

[0151] Meanwhile, it is worth noting that the uncertainty in calculating the average force on the reaction coordinate can be used to measure the reliability and accuracy of the calculated average force. The uncertainty of the reaction coordinate is obtained by removing the unbalanced part from the MD simulation trajectory with the reaction coordinate constraint value fixed, and using the method of "taking the average in segments" to find the average value of the force on the reaction coordinate for each trajectory segment, and then finding the variance of each average value.

[0152] For example, the average value of the force on the reaction coordinate F(d) for a certain trajectory segment is calculated using the following formula:

[0153]

[0154] where F a is the force on the dissolved atom from the potential energy surface, n ba is the unit direction vector pointing from atom b to atom a, k B is the Boltzmann constant, T is the simulation temperature, refers to the ensemble average obtained from the constrained MD simulation when the reaction coordinate is fixed at d.

[0155] The variance of the above average values for each trajectory segment can evaluate whether the uncertainty of the obtained average force is stable and reliable enough, so that when performing operations such as calculating the free energy by thermodynamic integration based on the average force in the subsequent process, the accuracy and reliability of the data can be reasonably considered.

[0156] In addition, the average force F(d) at each reaction coordinate is integrated with respect to the reaction coordinate according to the following formula to obtain the free energy A(d) along the reaction coordinate:

[0157]

[0158] where A(d 0 ) is the free energy at the initial value d of the reaction coordinate 0 , which is a constant and can be taken as 0. The uncertainty of the free energy is calculated according to the linear error propagation based on the uncertainty of the average force.

[0159] The time step Δt of the simulation is limited by the time scale of high-frequency vibrations in the system. Since the commonly used carbonate electrolyte molecules in lithium-ion batteries often contain the lightest hydrogen atoms, Δt generally needs to be less than 1 fs, such as 0.5 fs. In the examples of this application, the method of increasing the mass of hydrogen atoms is adopted to increase the simulation time step and further improve the calculation efficiency. By setting the mass of hydrogen atoms to 2 - 3 amu, Δt can be increased to 1 - 2 fs.

[0160] Optionally, since the existing constrained MD simulation technology often uses the RATTLE algorithm to constrain bond lengths, the positions of the two atoms forming a bond can both change during the simulation. In the example of this application, the existing RATTLE algorithm is improved to make it compatible with the constraint of fixing the reference atom position. The implementation method is as follows: In each time step of the MD simulation, first perform the RATTLE-x step based on the coordinate update in the unconstrained case to make the coordinates satisfy the constraint conditions, and then perform the RATTLE-v step based on the velocity update in the unconstrained case to make the velocity satisfy the constraint conditions. Taking the i-th time step as an example, RATTLE-x is an iterative process that repeatedly performs the coordinate update of the following formula:

[0161]

[0162] where r ab = r a - r b is the coordinate difference between atoms a and b, r ab (t i-1 ) is the coordinate difference between the two atoms at the (i - 1)-th time step. The iteration terminates when the parameter x is less than a preset critical value. RATTLE-v is also an iterative process that repeatedly performs the velocity update of the following formula:

[0163]

[0164] where r ab (t i ) is the coordinate difference between atoms a and b after the RATTLE-x step is executed. The iteration terminates when the parameter y is less than a preset critical value.

[0165] In a possible implementation, based on the unbalanced simulation duration and the total length of the MD simulation trajectory, the average force on the reaction coordinate and the uncertainty of the average force on the reaction coordinate are calculated, including:

[0166] Based on the unbalanced simulation duration and the total length of the MD simulation trajectory, the remaining simulation trajectory length is calculated;

[0167] The remaining simulation trajectory length is divided into multiple trajectory segments;

[0168] The ensemble average corresponding to each trajectory segment in the multiple trajectory segments is calculated to obtain multiple ensemble averages;

[0169] The average of the multiple ensemble averages is calculated to obtain the average force on the reaction coordinate, and the variance of the multiple ensemble averages is calculated to obtain the uncertainty of the average force on the reaction coordinate.

[0170] In this implementation, specifically during the MD simulation in the isothermal and isochoric NVT ensemble with a fixed reaction coordinate constraint value, the unbalanced simulation duration records the time span that the system experiences in the non-equilibrium state, while the total length of the MD simulation trajectory covers the complete time range from the start to the end of the simulation. By subtracting the unbalanced simulation duration from the total length of the MD simulation trajectory, the remaining simulation trajectory length is obtained. This is the length of the effective trajectory part that can be analyzed to obtain the average force and uncertainty of the reaction coordinate after the system reaches the equilibrium state during the simulation process. For example, if the total length of the MD simulation trajectory is 1000 time steps and the unbalanced simulation duration is 200 time steps, then the remaining simulation trajectory length is 1000 - 200 = 800 time steps. This 800-time-step trajectory part will serve as the main data source for subsequent calculations of relevant physical quantities because it represents the behavior record of the system in the equilibrium state.

[0171] By dividing the remaining simulation trajectory length, statistical analysis of the relevant data within each segment can be carried out in subsequent steps, so as to more accurately capture the fluctuation characteristics and average situation of the average force exerted on the reaction coordinate by the system in the equilibrium state, in order to further calculate the accurate average force and its uncertainty.

[0172] For each divided trajectory segment, it is necessary to calculate the ensemble average value of the force corresponding to the reaction coordinate within this segment. This ensemble average value refers to an average force value obtained after comprehensively considering various factors (such as interatomic interactions, thermal motion of the system, etc.) that may affect the force on the reaction coordinate within this specific trajectory segment. During specific calculations, statistical averaging operations will be carried out within each trajectory segment based on the data of the force on the reaction coordinate recorded at different times and different atomic states during the simulation process. For example, if a trajectory segment contains 100 data on the force of the reaction coordinate at different times, the sum of these 100 data is divided by 100 to obtain the ensemble average value corresponding to this trajectory segment. Performing the above operations on each divided trajectory segment can obtain multiple ensemble average values, and these average values will serve as the basic data for the next step of calculating the average force and uncertainty of the reaction coordinate.

[0173] Taking the average of the multiple ensemble average values obtained in the above example again, the result is the average force exerted on the reaction coordinate. This average force can be regarded as a comprehensive average level of the force on the reaction coordinate within the entire remaining simulation trajectory length (i.e., the trajectory part of the system in the equilibrium state), which reflects the magnitude of the average force exerted on the reaction coordinate during the dissolution process of metal atoms due to their interaction with the surrounding environment (such as the cathode material, electrolyte, etc.). For example, if there are 5 trajectory segments, and the corresponding ensemble average values are F 1 、F 2 、F 3 、F4 , F 5 , the average force on the reaction coordinate is (F 1 + F 2 + F 3 + F 4 + F 5 ) / 5.

[0174] The uncertainty of the average force on the reaction coordinate is obtained by calculating the variance of multiple ensemble averages. Variance is a statistic that measures the degree of dispersion of data. In the examples of this application, this variance reflects the dispersion of the ensemble averages corresponding to each trajectory segment relative to the average force on the reaction coordinate. The larger the variance, the greater the difference between the ensemble averages of each trajectory segment, indicating a higher uncertainty of the average force on the reaction coordinate. For example, for the ensemble averages F 1 , F 2 , F 3 , F 4 , F 5 of the above 5 trajectory segments, by calculating their variances, the uncertainty of the average force on the reaction coordinate can be obtained. This uncertainty can help evaluate whether the obtained average force on the reaction coordinate is stable and reliable enough, so that when performing operations such as calculating free energy by thermodynamic integration based on the average force in the subsequent process, the accuracy and reliability of the data can be reasonably considered.

[0175] To facilitate understanding of the calculation method for the dissolution of metal atoms at the interface between the positive electrode and the electrolyte provided in the examples of this application, the method of this application is explained below through two alternative embodiments. For example, in the first embodiment, the positive electrode material of the battery is lithium manganese phosphate (LiMnPO 4 , LMP), and the electrolyte is dimethyl carbonate (CH 3 OCOOCH 3 , DMC) as an example, the calculation of manganese dissolution at the (010) surface of lithium manganese phosphate (LiMnPO 4 , LMP) and the interface with dimethyl carbonate (CH 3 OCOOCH 3 , DMC) is carried out. The temperature in the specific simulation process is 300 K, and the pressure is 1.0 bar. The crystal Hamiltonian graph neural network CHGNet (Crystal Hamiltonian Graph Neural Network) pre-trained neural network force field is used for structure optimization and MD simulation. The specific calculation process of manganese dissolution is as follows:

[0176] (1) Construct a structural model of the LMP-DMC interface. Starting from the primitive cell of LMP, first construct the upper and lower symmetric structures corresponding to the (010) index and satisfying LiMnPO 4Stoichiometric LMP surface structure model. The supercell of this surface model has 4 layers of metal atoms and 16 chemical formulas. The lengths of its a-axis and b-axis are respectively and The total thickness of the vacuum layer is According to the density of DMC ρ = 1.07 g / cm 3 and molar mass M = 90.078 g / mol, it can be calculated that 5 DMC molecules should be added to each of the upper and lower surface vacuum layers. The CHGNet force field is used to optimize the atomic positions and lattice vectors of the supercell. The convergence criterion for optimization is that the atomic force is less than

[0177] It should be understood that the atomic position refers to the coordinates of each atom in the supercell. The basic information required to express a unit cell includes: (1) Lattice vectors a, b, c: These are three vectors that represent the translational periodicity of the material lattice. (2) Each atom therein. The position of each atom can be represented by fractional coordinates relative to the lattice vectors. For example, the position of atom 1 can be expressed as: r 1 = xa + yb + zc, where x, y, z are the fractional coordinates of atom 1. Therefore, the structural optimization of the material includes adjusting two aspects: the lattice vectors and the fractional coordinates of each atom.

[0178] (2) Perform MD simulation on the above interface structure model in the NPT ensemble. Set the temperature to 300 K and the pressure to 1.0 bar. Set the mass of the hydrogen atom to 2 amu and the simulation time step to 1.0 fs. Set the total simulation time to 100 ps. According to the changes of the average potential energy, temperature and lattice parameters of the cathode material and electrolyte interface structure model with simulation time, determine the equilibrium time to be 50 ps. Use the method of segmental averaging to calculate the equilibrium lattice parameters and uncertainties on the trajectory after removing this part, that is: after removing the initial 50 ps trajectory, calculate the average value of the lattice parameters for each 5 ps trajectory segment; then take the average of the average values of 10 trajectory segments to obtain the equilibrium lattice parameters, and the variance of the average values of each of the 10 trajectory segments is used as the uncertainty of the equilibrium lattice parameters.

[0179] (3) MD simulations with a slowly increasing reaction coordinate: The structure of the last frame of the NPT simulation trajectory in step (2) was scaled to the obtained equilibrium lattice parameters to obtain the initial structure for the MD simulation in this step. The MD simulation was carried out in the NVT ensemble, and the target temperature was set to 300 K. The settings of the hydrogen atom mass and the time step were the same as in step (2). On the (010) surface of the LMP material, all the Mn-O bonds of the surface Mn atoms were not perpendicular to the surface. Therefore, it was not appropriate to use the Mn-O bond length to describe the Mn dissolution process. In this example, the distance between the surface-dissolved Mn atom (Mn15) and an internal Mn atom (Mn11, located in the 3rd layer) directly below it was selected as the reaction coordinate. The initial value of the reaction coordinate d 0 was set to The equilibrium time for staying at this initial value was set to 5 ps, and the growth rate of the reaction coordinate was set to The total simulation duration was set to 20 ps. During the MD simulation, the reaction coordinate was constrained by an improved bond length constraint algorithm, and the constraint target value of the reaction coordinate was modified every 5 fs to achieve the slow growth of the reaction coordinate. In addition to constraining the reaction coordinate, the positions of all the Mn atoms in the 3rd layer (Mn3, Mn6, Mn11, Mn14), including the reference atom Mn11, were also constrained.

[0180] (4) MD simulations with a fixed reaction coordinate: Starting from the reaction coordinate growth part (after 5 ps) of the simulation trajectory in step (3), structures were intercepted as the initial structures at intervals of to every (such as ) for MD simulations with a fixed reaction coordinate. The settings of the temperature, hydrogen atom mass, simulation time step, reaction coordinate constraint, and atomic position constraint in this step were the same as in step (3). The difference between this step and step (3) was that the constraint target value of the reaction coordinate remained unchanged during the simulation in this step. There were a total of 71 MD simulations with different reaction coordinate constraint target values in this step. The simulation duration was set to 10 ps. After the simulation ended, for each trajectory, the unbalanced part with a length of 2 ps was removed, and the remaining 8 ps trajectory was evenly divided into 5 trajectory segments with a length of 1.6 ps. The average force and its uncertainty exerted on the reaction coordinate were calculated by the method of taking the average of the segments.

[0181] (5) Thermodynamic integration: The average force exerted on the reaction coordinate obtained in step (4) was integrated with respect to the reaction coordinate to obtain the free energy change of the cathode material - electrolyte interface structure model along the reaction coordinate. Using the trapezoidal integration formula, the free energy calculation formula at each reaction coordinate was:

[0182]

[0183] Among them, n is the number of different values of the reaction coordinate. The uncertainty of the free energy at each reaction coordinate is calculated by the linear error propagation of the uncertainty of the average force. The specific calculation formula is:

[0184]

[0185] The average force curve and free energy curve along the reaction coordinate during the Mn dissolution process at the LMP-DMC interface obtained through the above steps are as Figure 3 shown, where the dashed line is the average force and the solid line is the free energy. The width of the shaded area represents the magnitude of the uncertainty. Based on Figure 3 a clear free energy plateau indicating the end of dissolution can be observed. Its height represents the free energy barrier for Mn dissolution, with a magnitude of 3.4 ± 0.1 eV. A frame of the interface conformation intercepted from the trajectory of the MD simulation with the fixed reaction coordinate of is as Figure 4 shown, Figure 4 and the coordination of Mn atoms in the DMC electrolyte after the dissolution is shown in Figure 4 The upper half of the interface in

[0186] shows only 4 DMC molecules coordinated with the dissolved Mn atoms, and the 4 DMC molecules form a penta-coordination with Mn. 0.5 Fe 0.5 PO 4 For another example, in the second embodiment, taking the cathode material of the battery as lithium manganese iron phosphate (LiMn 3 H 4 O 3 with Fe / Mn = 50:50 (LMFP)) and the electrolyte as ethylene carbonate (C 0.5 Fe 0.5 PO 4 as an example, the calculation of Mn dissolution at the interface between the (010) surface of lithium manganese iron phosphate (LiMn 3 H 4 O 3 with Fe / Mn = 50:50 (LMFP)) and ethylene carbonate (C

[0187] (1) Construct a structural model of the LMFP-DMC interface. Starting from the primitive cell of LMFP, construct an upper and lower symmetric structure corresponding to the (010) index and satisfying LiMn 0.5 Fe 0.5 PO 4Stoichiometric LMFP surface structure model. The primitive cell of LMFP contains 2 Fe atoms and 2 Mn atoms. Select the arrangement with the lowest energy, that is, Fe and Mn are arranged in layers along the a-axis direction. The supercell of the surface model has 4 layers of metal atoms and 16 chemical formulas. The lengths of its a-axis and b-axis are respectively and The total thickness of the vacuum layer is According to the density of EC ρ = 1.32 g / cm 3 and molar mass M = 88.06 g / mol, it can be calculated that 7 EC molecules should be added to each of the upper and lower surface vacuum layers. Use the CHGNet force field to optimize the atomic positions and lattice vectors of the supercell. The convergence criterion for the structure optimization is set to the atomic force being less than

[0188] (2) Perform MD simulation of the NPT ensemble on the above interface structure model. Set the temperature to 300 K and the pressure to 1.0 bar. Set the mass of the hydrogen atom to 2 amu and the simulation time step to 1.0 fs. Set the total simulation time to 60 ps. Determine the equilibrium time to be 30 ps according to the changes of the average potential energy, temperature, and lattice parameters of the cathode material and electrolyte interface structure model with the simulation time. Use the method of taking the average in segments on the trajectory after removing this part to calculate the equilibrium lattice parameter and uncertainty. The length of the trajectory segment is 5 ps, with a total of 6 trajectory segments.

[0189] (3) MD simulation with slow growth of the reaction coordinate: The initial structure of this step of the MD simulation is obtained by scaling the unit cell of the last frame of the trajectory in step (2) to the obtained equilibrium lattice parameter. The target temperature of the MD simulation is set to 300 K. There is also no Mn-O bond perpendicular to the surface on the (010) surface of the LMFP material. In this embodiment, the distance between the surface-dissolved Mn atom (Mn8) and an internal Mn atom (Mn4, located in the 3rd layer) directly below it is selected as the reaction coordinate. The initial value d of the reaction coordinate 0 is The equilibrium time for staying at this initial value is set to 5 ps, and the growth rate of the reaction coordinate is set to The total simulation duration is set to 36.6 ps. During the MD simulation, the constraint target value of the reaction coordinate is modified every 10 fs to achieve the slow growth of the reaction coordinate. In addition to constraining the reaction coordinate, the positions of all Fe and Mn atoms (Fe5, Fe6, Mn3, Mn4) in the 3rd layer including the reference atom Mn4 are also constrained.

[0190] (4) MD simulation with the reaction coordinate fixed: Starting from the reaction coordinate growth part (after 5 ps) of the simulation trajectory in step (3), according to the reaction coordinate value from to Every The interception structure is used as the initial structure, and MD simulations with fixed reaction coordinates are carried out. There are a total of 80 MD simulations with different reaction coordinate constraint target values in this step. The simulation duration is set to 15 ps. After the simulation ends, for each trajectory, the unbalanced part with a length of 3 ps is removed, and the remaining 12 ps trajectory is evenly divided into 6 trajectory segments with a length of 2 ps. The average force and its uncertainty received by the reaction coordinate are calculated by the method of taking the average of each segment.

[0191] (5) Thermodynamic integration: Integrate the average force received by the reaction coordinate obtained in step (4) with respect to the reaction coordinate to obtain the free energy change of the cathode material - electrolyte interface structure model along the reaction coordinate.

[0192] The average force curve and free energy curve of the Mn dissolution process at the LMFP - EC interface along the reaction coordinate obtained through the above steps are as Figure 5 shown, Figure 5 The dashed line in it is the average force, the solid line is the free energy, and the width of the shaded area represents the magnitude of the uncertainty. Based on Figure 5 a clear free energy plateau indicating the end of dissolution can be observed, and its height shows that the magnitude of the free energy barrier for Mn dissolution on this interface is 3.1 ± 0.1 eV. A frame of the interface conformation intercepted from the MD simulation trajectory with the fixed reaction coordinate of is as Figure 6 shown, and this Figure 6 can clearly show that at the above - mentioned free energy plateau, the Mn atoms have completely left the cathode - electrolyte interface and formed a square - pyramid - shaped coordination polyhedron with the EC molecules in the electrolyte. For clarity, only 5 EC molecules coordinated with the dissolved Mn atoms are shown in the upper - half interface in the figure. These 5 EC molecules form a five - coordination in the form of a square pyramid for Mn.

[0193] In addition, the embodiment of the present application also provides a comparative example to illustrate the effect of the improvement in the reaction coordinate selection method in the example solution of the present application. In the prior art for calculating metal dissolution at the cathode - electrolyte interface, generally, the bond length is selected as the reaction coordinate, and only the value of the reaction coordinate is constrained in the simulation. The cathode material - electrolyte interface structure model processed in this embodiment is the same as that in the second embodiment above. The overall calculation process is similar to that of the second embodiment, except that in the MD simulations of steps (3) and (4), the traditional RATTLE algorithm is used, and only the distance between the dissolved atom Mn8 and the reference atom Mn4 is constrained, and the positions of Mn4 and other Fe and Mn atoms (Fe5, Fe6, Mn3) in the same layer as it are not constrained.

[0194] Figure 7The average force and free energy curves along the reaction coordinate during the Mn dissolution process at the LMFP-EC interface calculated by only constraining the reaction coordinate according to the prior art (referred to as Comparative Example 1). The Figure 7 free energy plateau value is 2.5 ± 0.1 eV, which is significantly lower than the Figure 5 free energy plateau value obtained in the second embodiment shown. In addition, the reaction coordinate at which the free energy curve reaches the plateau value is also significantly earlier than that of the Figure 5 second embodiment in Figure 8 shows a frame of the interface conformation of the MD simulation fixed at the same reaction coordinate in the prior art as Figure 6 . It can be seen from Figure 8 that since the position of the reference atom (Mn4) is not constrained, it undergoes a large displacement under the action of the corresponding binding force of the reaction coordinate and has moved from the third layer to the lower surface of the cathode material, resulting in the dissolution atom (Mn8) not being able to leave the upper surface of the delithiated cathode material under such a large increase in the reaction coordinate . Thus, it can be seen that without fixing the reference atom, the increase in the reaction coordinate value cannot truly reflect the metal dissolution at the cathode-electrolyte interface.

[0195] Please refer to Figure 7 shown. Figure 7 In Comparative Example 1 shown, the average force and free energy curves along the reaction coordinate during the Mn dissolution at the LMFP-EC interface are calculated. The dashed line is the average force, the solid line is the free energy, and the width of the shaded area represents the magnitude of the uncertainty. Figure 8 In Comparative Example 1 shown, a frame of the LMFP-EC interface conformation intercepted from the MD simulation with the fixed reaction coordinate of is shown. For clarity, only one EC molecule coordinated with the dissolved Mn atom (Mn8) is shown in the upper half of the figure. Due to the significant displacement of the reference atom (Mn4), at this reaction coordinate, the Mn8 atom has not left the cathode-electrolyte interface.

[0196] The example of this application introduces a machine learning force field to replace the first-principles calculations based on DFT, greatly improving the computational efficiency of the metal dissolution problem at the cathode-electrolyte interface. Although the first-principles calculations based on DFT have high accuracy, their complex quantum chemical calculation process has a large computational amount, so the time and space scales that can be processed by using them for material calculations are limited. The characteristics of the metal dissolution problem at the cathode-electrolyte interface are as follows: First, a large supercell structure needs to be established to simulate the physical and chemical processes occurring at the interface; Second, to simulate the dissolution process, multiple long-time MD simulations are required to achieve sufficient sampling of the conformations of organic molecules in the electrolyte. These characteristics make the efficiency of using first-principles calculations to deal with the metal dissolution problem extremely low. On the contrary, the machine learning force field avoids the complex quantum chemical calculation process, and its computational efficiency can be improved by two to three orders of magnitude compared with the first-principles calculations. In addition, the machine learning force field pre-trained on a large number of first-principles calculation data sets can inherit the high accuracy of the first-principles calculations while improving the computational efficiency.

[0197] The example of this application improves the selection method of the reaction coordinate, enhancing the generality of the calculation method for different types of cathode materials. The prior art generally selects the bond length between a metal atom and a coordinating atom as the reaction coordinate. This selection method is only suitable for cathode materials in which the coordination polyhedron of the dissolved metal atom on the interface has a special shape, that is, there is a metal-coordinating atom bond perpendicular to the surface, so the applicable range is limited, such as the LMO material with a spinel structure. The example of this application selects the distance between the dissolved metal atom on the interface and a reference atom below it inside the cathode material as the reaction coordinate, which has no requirement for the interface structure of the cathode material, improving the generality of the method. In addition, the example of this application modifies the traditional RATTLE algorithm with bond length constraints to fix the position of the reference atom while fixing the reaction coordinate, so that the increase of the reaction coordinate can truly reflect the dissolution process of the interface metal atom, improving the accuracy of the calculation method for simulating the metal dissolution at the cathode-electrolyte interface.

[0198] The technical solution of the example of this application avoids the quantum chemical calculation process with low efficiency and improves the efficiency of material calculations for dealing with the metal dissolution problem at the cathode-electrolyte interface. In addition, the improvement of the reaction coordinate selection method in the example of this application makes the types of cathode materials that can be processed by the calculation no longer limited by the interface structure, with stronger generality. The metal dissolution at the cathode-electrolyte interface is one of the key problems affecting the stability and cycling performance of cathode materials. The technical solution of the example of this application helps to promote the research and solution of the interface metal dissolution problem by means of material calculations, and ultimately accelerates the development of new lithium-ion battery cathode materials.

[0199] In the technical solution exemplified by this application, it is pointed out that for electrolyte organic molecules containing hydrogen atoms, the time step is increased by increasing the mass of hydrogen atoms in MD simulations to improve the simulation efficiency. Another treatment method is to fix the bond lengths of all hydrogen-containing chemical bonds during the MD simulation, such as the C-H bond length in carbonate molecules, so as to eliminate the high-frequency stretching vibrations of the hydrogen-containing chemical bonds in the interface structure model between the positive electrode material and the electrolyte. This treatment method can also increase the time step.

[0200] The technical solution exemplified by this application uses a pre-trained machine learning force field to calculate the total energy and atomic forces of the interface structure model between the positive electrode material and the electrolyte. Currently, general pre-trained machine learning force fields are mostly trained on datasets calculated by first-principles based on DFT, and the DFT method itself has deficiencies in describing the dispersion interaction forces between molecules. To improve the accuracy of the machine learning force field trained using first-principles calculation data in dealing with the interactions between electrolyte organic molecules, the calculation system provided by this application example also implements the function of adding dispersion force correction on the basis of the machine learning force field for selection.

[0201] It should be understood that the magnitudes of the sequence numbers of the steps in the above embodiments do not mean the order of execution. The execution order of each process should be determined by its function and internal logic, and should not constitute any limitation to the implementation process of the embodiments of this application.

[0202] Corresponding to the calculation method for the dissolution of metal atoms at the junction of the battery positive electrode and the electrolyte in the above embodiments, Figure 9 is a schematic structural diagram of a calculation system for the dissolution of metal atoms at the junction of a battery positive electrode and an electrolyte provided by an embodiment of this application. This system can be implemented as part or all of a computer device by software, hardware, or a combination of both. This computer device can be Figure 10 the electronic device shown.

[0203] Referring to Figure 9 , the calculation system 600 for the dissolution of metal atoms at the junction of the battery positive electrode and the electrolyte includes: an input module 601, a structure building module 602, a molecular dynamics MD simulation module 603, a post-processing module 604, and an output module 605, where:

[0204] The input module 601 is used to obtain the material information of the battery positive electrode material and the electrolyte molecules, as well as the calculation parameters for controlling the calculation simulation process.

[0205] The structure building module 602 is connected to the input module 601 and is used to construct a corresponding interface structure model of the positive electrode material and the electrolyte based on the material information and the calculation parameters.

[0206] The MD simulation module 603, connected to the structure construction module 602, is used to perform equilibrium MD simulation on the interface structure model of the cathode material and the electrolyte through the isothermal and isobaric NPT ensemble algorithm to obtain the equilibrium lattice parameters, and perform constrained MD simulation on the interface structure model of the cathode material and the electrolyte based on the equilibrium lattice parameters through the isothermal and isochoric NVT ensemble algorithm to obtain the average force on the reaction coordinate. The equilibrium lattice parameters are used to reflect the lattice structure state when the interaction between the cathode material and the electrolyte molecules of the battery reaches equilibrium under the conditions of equilibrium MD simulation. The reaction coordinate is defined as the distance between the dissolved metal atoms at the junction of the cathode material and the electrolyte of the battery and the reference atoms.

[0207] The post-processing module 604, connected to the MD simulation module 603, is used to perform thermodynamic integration calculation on the average force to obtain the free energy curve of the dissolved metal atoms at the junction of the cathode material and the electrolyte of the battery;

[0208] The output module 605, connected to the post-processing module 604, is used to output the free energy curve, and the free energy curve is used to reflect the energy change during the process of the metal atoms dissolving from the surface of the cathode material of the battery into the electrolyte.

[0209] Further, on the basis of any of the above embodiments, as an example of the present application, the computing system further includes:

[0210] The control module, connected to each module in the computing system, is used to control the working process of each module.

[0211] Further, on the basis of any of the above embodiments, as an example of the present application, the structure construction module constructs the corresponding interface structure model of the cathode material and the electrolyte based on the material information and calculation parameters, and is configured to:

[0212] Construct a symmetric surface structure model based on the Miller indices of the cathode material surface based on the material information and calculation parameters, wherein there are an upper vacuum layer and a lower vacuum layer on both sides of the cathode material surface, and the same number of electrolyte molecules are added to the upper vacuum layer and the lower vacuum layer. The number of electrolyte molecules is calculated based on the volume of the vacuum layer and the density of the electrolyte;

[0213] Perform structure optimization processing on the symmetric surface structure model through a machine learning force field to obtain the corresponding interface structure model of the cathode material and the electrolyte, wherein the structure optimization processing is used to eliminate the unstable atomic contact phenomenon introduced during the modeling process. The machine learning force field is a force field obtained by training through a machine learning method using a first-principles atomic calculation data set. The first-principles atomic calculation data set includes: the atomic structures and calculation results of multiple cathode materials of the battery generated during the first-principles calculation process. The output results of the machine learning force field are the total energy, the force on each atom, and the lattice stress of the symmetric surface structure model.

[0214] Further, based on any of the above embodiments, as an example of the present application, the MD simulation module performs equilibrium molecular dynamics MD simulation on the interface structure model of the cathode material and the electrolyte through the isothermal and isobaric NPT ensemble algorithm to obtain the equilibrium lattice parameters, and is configured to:

[0215] During the process of performing equilibrium molecular dynamics MD simulation, monitor the changes of the average potential energy, temperature, and lattice constant of the interface structure model of the cathode material and the electrolyte with the simulation time, and determine the unbalanced simulation duration when the interface structure model of the cathode material and the electrolyte is in an unbalanced state;

[0216] Based on the unbalanced simulation duration and the total length of the MD simulation trajectory, calculate the equilibrium lattice parameters and uncertainties of the interface structure model of the cathode material and the electrolyte.

[0217] Further, based on any of the above embodiments, as an example of the present application, the MD simulation module calculates the equilibrium lattice parameters and uncertainties of the interface structure model of the cathode material and the electrolyte based on the unbalanced simulation duration and the total length of the MD simulation trajectory, and is further configured to:

[0218] Obtain the unbalanced simulation duration and the total length of the MD simulation trajectory of the interface structure model of the cathode material and the electrolyte;

[0219] Based on the unbalanced simulation duration and the total length of the MD simulation trajectory, calculate the remaining simulation trajectory length;

[0220] Divide the remaining simulation trajectory length into multiple trajectory segments;

[0221] Calculate the average value of the lattice parameters corresponding to each trajectory segment in the multiple trajectory segments to obtain multiple average values of the lattice parameters;

[0222] Calculate the average value of the multiple average values of the lattice parameters to obtain the equilibrium lattice parameters, and calculate the variance of the multiple average values of the lattice parameters to obtain the uncertainty of the equilibrium lattice parameters.

[0223] Further, based on any of the above embodiments, as an example of the present application, the constrained MD simulation includes: MD simulation with the constraint value of the reaction coordinate changed and MD simulation with the constraint value of the reaction coordinate fixed. The MD simulation with the constraint value of the reaction coordinate changed is used to continuously increase the constraint value of the reaction coordinate during the simulation to provide the interface structure during the dissolution process of metal atoms; the MD simulation with the constraint value of the reaction coordinate fixed is used to perform configuration sampling on the interface structure under the condition of a fixed reaction coordinate and calculate the average force exerted on the reaction coordinate.

[0224] Further, based on any of the above embodiments, as an example of the present application, the MD simulation module performs constrained MD simulation on the interface structure model of the cathode material and the electrolyte based on the equilibrium lattice parameters through the isothermal-isochoric NVT ensemble algorithm, obtains the average force exerted on the reaction coordinate, and is further configured to:

[0225] During the MD simulation where the constraint value of the reaction coordinate changes, the coordinates and velocities of the dissolved metal atoms are adjusted based on the improved interatomic distance constraint algorithm so that the distance between the metal atoms and the reference atoms is equal to the constraint target value, and the constraint target value of the reaction coordinate is continuously adjusted so that the reaction coordinate continuously increases after equilibrating for a predetermined time at the initial value;

[0226] Based on the interface structure during the dissolution process of the metal atoms provided by the MD simulation with the change of the constraint value of the reaction coordinate, an initial structure is intercepted, where the initial structure is used to perform MD simulation with the position of the reference atoms fixed and the constraint value of the reaction coordinate fixed;

[0227] After the MD simulation with the constraint value of the reaction coordinate fixed, based on the unbalanced simulation duration and the total length of the MD simulation trajectory, the average force exerted on the reaction coordinate and the uncertainty of the average force exerted on the reaction coordinate are calculated.

[0228] Further, based on any of the above embodiments, as an example of the present application, the MD simulation module calculates the average force exerted on the reaction coordinate and the uncertainty of the average force exerted on the reaction coordinate based on the unbalanced simulation duration and the total length of the MD simulation trajectory, and is further configured to:

[0229] Based on the unbalanced simulation duration and the total length of the MD simulation trajectory, calculate the remaining simulation trajectory length;

[0230] Divide the remaining simulation trajectory length into multiple trajectory segments;

[0231] Calculate the ensemble average value corresponding to each trajectory segment in the multiple trajectory segments to obtain multiple ensemble average values;

[0232] Calculate the average value of the multiple ensemble average values to obtain the average force exerted on the reaction coordinate, and calculate the variance of the multiple ensemble average values to obtain the uncertainty of the average force exerted on the reaction coordinate.

[0233] It should be noted that: for the calculation system of the dissolution of metal atoms at the junction of the battery cathode and the electrolyte provided in the above embodiments, only the above-mentioned division of each functional module is used for illustration. In practical applications, the above functions can be assigned to different functional modules according to needs, that is, the internal structure of the device is divided into different functional modules to complete all or part of the functions described above.

[0234] In the above embodiments, each functional unit and module can be integrated into one processing unit, or each unit can exist physically alone, or two or more units can be integrated into one unit. The above integrated unit can be implemented in the form of hardware or in the form of a software functional unit. In addition, the specific names of each functional unit and module are only for the convenience of mutual distinction and do not limit the protection scope of the embodiments of the present application.

[0235] It should be noted that, for the content such as information interaction and execution process between the above-mentioned devices / units, since it is based on the same concept as the method embodiments of the present application, for its specific functions and the technical effects brought, reference can be specifically made to the method embodiment part, and details are not described herein again.

[0236] The embodiments of the present application further provide an electronic device, which includes one or more processors and a memory;

[0237] The memory is coupled to one or more processors. The memory is used to store computer program code, and the computer program code includes computer instructions. One or more processors call the computer instructions to enable the electronic device to execute the calculation method for the dissolution of metal atoms at the junction of the battery positive electrode and the electrolyte as shown above.

[0238] Figure 10 FIG. 13 is a schematic structural diagram of an electronic device provided by an embodiment of the present application. The electronic device 700 can be a mobile phone, a smart screen, a tablet computer, a wearable electronic device, a vehicle-mounted electronic device, an augmented reality (AR) device, a virtual reality (VR) device, a laptop computer, an ultra-mobile personal computer (UMPC), a netbook, a personal digital assistant (PDA), a projector, or a communication device such as a server, a memory, a base station, or a smart vehicle, etc. The embodiments of the present application do not impose any restrictions on the specific type of the electronic device.

[0239] The memory 701 can be used to store computer software programs 702 and modules. The processor 703 executes various functional applications and data processing of the electronic device by running the software programs and modules stored in the memory 701. The memory 701 mainly includes a program storage area and a data storage area. Among them, the program storage area can store an operating system, application programs required for at least one function (such as a sound playback function, an image playback function, etc.); the data storage area can store data created according to the use of the electronic device (such as audio data, phone book, etc.). In addition, the memory 701 can include a high-speed random access memory, and can also include a non-volatile memory, such as at least one magnetic disk storage device, a flash memory device, or other volatile solid-state storage devices.

[0240] Among them, the processor 703 can include one or more of a central processing unit, an application processor (AP), a baseband processor, etc. The processor can be the nerve center and command center of the wireless router. The processor 703 can generate operation control signals according to the instruction operation code and timing signal to complete the control of fetching instructions and executing instructions. The memory 701 can be used to store computer-executable program codes, and the executable program codes include instructions. The processor 703 executes various functional applications and data processing of the network device by running the instructions stored in the memory. The memory 701 can include a program storage area and a data storage area, such as data for storing sound signals to be played. For example, the memory can be a double data rate synchronous dynamic random access memory DDR or a flash memory Flash, etc.

[0241] The embodiment of the present application also provides a computer-readable storage medium, in which computer instructions are stored; when the computer-readable storage medium runs on an electronic device, the electronic device is enabled to execute the calculation method for the dissolution of metal atoms at the junction of the battery positive electrode and the electrolyte as shown above.

[0242] The computer instructions can be stored in a computer-readable storage medium or transmitted from one computer-readable storage medium to another. For example, the computer instructions can be transmitted from one website, computer, server, or data center to another website, computer, server, or data center by wired means (such as coaxial cable, optical fiber, digital subscriber line (DSL)) or wireless means (such as infrared, wireless, microwave, etc.). The computer-readable storage medium can be any available medium that can be accessed by a computer or a data storage device such as a server or a data center that incorporates one or more media. The available medium can be a magnetic medium (such as a floppy disk, hard disk, magnetic tape), an optical medium, or a semiconductor medium (such as a solid-state disk (SSD)).

[0243] An embodiment of this application also provides a computer program product containing computer instructions. When the computer program product runs on an electronic device, it enables the electronic device to execute the calculation method for the dissolution of metal atoms at the junction of the battery positive electrode and the electrolyte as shown above.

[0244] The computer storage medium and the computer program product provided in the above embodiments of this application are both used to execute the method provided above. Therefore, the beneficial effects that can be achieved can refer to the beneficial effects corresponding to the method provided above and will not be elaborated here.

[0245] In the above embodiments, it can also be implemented in whole or in part by software, hardware, firmware, or any combination thereof. When implemented using software, it can be implemented in whole or in part in the form of a computer program product. The computer program product includes one or more computer instructions. When the computer instructions are loaded and executed on a computer, the processes or functions described in the embodiments of the present application are generated in whole or in part. The computer can be a general-purpose computer, a special-purpose computer, a computer network, or other programmable devices. The computer instructions can be stored in a computer-readable storage medium, or transmitted from one computer-readable storage medium to another computer-readable storage medium. For example, the computer instructions can be transmitted from one website, computer, server, or data center to another website, computer, server, or data center in a wired manner (such as coaxial cable, optical fiber, Digital Subscriber Line (DSL)) or a wireless manner (such as infrared, wireless, microwave, etc.). The computer-readable storage medium can be any available medium that can be accessed by a computer, or a data storage device such as a server or data center that includes one or more integrated available media. The available media can be magnetic media (such as floppy disks, hard disks, magnetic tapes), optical media (such as Digital Versatile Disc (DVD)), or semiconductor media (such as Solid State Disk (SSD)), etc.

[0246] In the above embodiments, the descriptions of the respective embodiments have their own emphases. For parts not detailed or recorded in a certain embodiment, reference can be made to the relevant descriptions of other embodiments.

[0247] Those of ordinary skill in the art can realize that the units and algorithm steps of the examples described in combination with the embodiments disclosed in this application can be implemented by electronic hardware, or a combination of computer software and electronic hardware. Whether these functions are executed in a hardware or software manner depends on the specific application and design constraints of the technical solution. A professional technician can use different methods for each specific application to implement the described functions, but such implementation should not be considered to exceed the scope of this application.

[0248] In the embodiments provided in the present application, it should be understood that the disclosed device / network device and method can be implemented in other ways. For example, the device / network device embodiments described above are merely illustrative. For example, the division of the modules or units is only a logical function division. In actual implementation, there may be other division methods. For example, multiple units or components can be combined or integrated into another system, or some features can be ignored or not executed. Another point is that the displayed or discussed couplings or direct couplings or communication connections to each other can be through some interfaces. The indirect couplings or communication connections of the devices or units can be in electrical, mechanical or other forms.

[0249] The units described as separate components may or may not be physically separated. The components displayed as units may or may not be physical units, that is, they may be located in one place or distributed to multiple network units. Some or all of the units can be selected according to actual needs to achieve the purpose of the solution of this embodiment.

[0250] The above embodiments are only used to illustrate the technical solutions of the present application and are not intended to limit them. Although the present application has been described in detail with reference to the foregoing embodiments, those of ordinary skill in the art should understand that they can still modify the technical solutions recorded in the foregoing embodiments or perform equivalent replacements for some of the technical features. These modifications or replacements do not cause the essence of the corresponding technical solutions to deviate from the spirit and scope of the technical solutions of the embodiments of the present application and should all be included in the protection scope of the present application.

Claims

1. A method for calculating the dissolution of metal atoms at the interface between a battery positive electrode and an electrolyte, characterized in that: include: Obtain material information of battery positive electrode materials and electrolyte molecules, as well as calculation parameters used to control the calculation simulation process; Based on the material information and the calculation parameters, a corresponding positive electrode material and electrolyte interface structure model is constructed; Performing equilibrium molecular dynamics (MD) simulation on the interface structure model of the positive electrode material and the electrolyte by an isothermal and isobaric NPT ensemble algorithm to obtain equilibrium lattice parameters, wherein the equilibrium lattice parameters are used to reflect the lattice structure state when the interaction between the positive electrode material of the battery and the electrolyte molecules reaches equilibrium under equilibrium MD simulation conditions; By using an isothermal and isochoric NVT ensemble algorithm, a constrained MD simulation is performed on the interface structure model of the positive electrode material and the electrolyte based on the equilibrium lattice parameters to obtain the average force on the reaction coordinate, wherein the reaction coordinate is defined as the distance between the dissolved metal atom and the reference atom at the interface between the positive electrode material of the battery and the electrolyte; The average force is thermodynamically integrated to obtain a free energy curve of the dissolution of metal atoms at the interface between the battery positive electrode material and the electrolyte. The free energy curve is used to reflect the energy change of the metal atoms during the dissolution process from the surface of the battery positive electrode material into the electrolyte.

2. The method according to claim 1, characterized in that The constructing of a corresponding positive electrode material and electrolyte interface structure model based on the material information and the calculation parameters includes: Based on the material information and the calculation parameters, a symmetrical surface structure model based on the Miller index of the positive electrode material surface is constructed, wherein an upper vacuum layer and a lower vacuum layer exist on both sides of the surface of the positive electrode material, and the same number of electrolyte molecules are added to the upper vacuum layer and the lower vacuum layer, and the number of the electrolyte molecules is calculated based on the volume of the vacuum layer and the density of the electrolyte; The symmetric surface structure model is structurally optimized by a machine learning force field to obtain a corresponding interface structure model of the positive electrode material and the electrolyte, wherein the structural optimization is used to eliminate the unstable contact phenomenon between atoms introduced in the modeling process, and the machine learning force field is obtained by training a force field using a first-principles atomic calculation data set through a machine learning method, and the first-principles atomic calculation data set includes: multiple atomic structures of the battery positive electrode materials generated in a first-principles calculation process and their calculation results, and the output result of the machine learning force field is the total energy of the symmetric surface structure model, the force on each atom, and the lattice stress.

3. The method according to claim 1, characterized in that The method of performing equilibrium molecular dynamics MD simulation on the interface structure model of the positive electrode material and the electrolyte by using an isothermal and isobaric NPT ensemble algorithm to obtain equilibrium lattice parameters includes: During the equilibrium molecular dynamics MD simulation, the average potential energy, temperature and lattice constant of the interface structure model of the positive electrode material and the electrolyte are monitored as a function of simulation time, and the unbalanced simulation time during which the interface structure model of the positive electrode material and the electrolyte is in an unbalanced state is determined; Based on the unbalanced simulation time and the total length of the MD simulation trajectory, the equilibrium lattice parameters and uncertainties of the interface structure model between the positive electrode material and the electrolyte are calculated.

4. The method according to claim 3, characterized in that Based on the unbalanced simulation time and the total length of the MD simulation trajectory, the equilibrium lattice parameters and uncertainties of the interface structure model of the cathode material and the electrolyte are calculated, including: Obtaining the unbalanced simulation time and the total length of the MD simulation trajectory of the interface structure model of the positive electrode material and the electrolyte; Based on the unbalanced simulation time and the total length of the MD simulation trajectory, a remaining simulation trajectory length is calculated; dividing the remaining simulated trajectory length into a plurality of trajectory segments; Calculating an average value of a lattice parameter corresponding to each trajectory segment in the plurality of trajectory segments to obtain a plurality of lattice parameter average values; The average of the plurality of lattice parameter average values ​​is calculated to obtain the equilibrium lattice parameter, and the variance of the plurality of lattice parameter average values ​​is calculated to obtain the uncertainty of the equilibrium lattice parameter.

5. The method according to claim 1, characterized in that The constrained MD simulation includes: an MD simulation with a changed constraint value of the reaction coordinate and an MD simulation with a fixed constraint value of the reaction coordinate. The MD simulation with a changed constraint value of the reaction coordinate is used to continuously increase the constraint value of the reaction coordinate during the simulation process to provide an interface structure during the dissolution of metal atoms; the MD simulation with a fixed constraint value of the reaction coordinate is used to perform configuration sampling of the interface structure under fixed reaction coordinate conditions and calculate the average force acting on the reaction coordinate.

6. The method according to claim 5, characterized in that The method of performing constrained MD simulation on the interface structure model of the cathode material and the electrolyte based on the equilibrium lattice parameters by using an isothermal and isochoric NVT ensemble algorithm to obtain the average force on the reaction coordinates includes: During the MD simulation in which the constraint value of the reaction coordinate changes, the coordinates and speed of the dissolved metal atoms are adjusted based on the improved interatomic distance constraint algorithm so that the distance between the metal atom and the reference atom is equal to the constraint target value, and the constraint target value of the reaction coordinate is continuously adjusted so that the reaction coordinate continues to increase after the initial value is balanced for a predetermined time; Based on the interface structure of the metal atom dissolution process provided by the MD simulation with the constraint value of the reaction coordinate changed, an initial structure is obtained, wherein the initial structure is used for the MD simulation with the reference atomic position and the constraint value of the reaction coordinate fixed; After the MD simulation with the constraint value of the reaction coordinate fixed, the average force on the reaction coordinate and the uncertainty of the average force on the reaction coordinate are calculated based on the unbalanced simulation time and the total length of the MD simulation trajectory.

7. The method according to claim 6, characterized in that The average force on the reaction coordinate and the uncertainty of the average force on the reaction coordinate are calculated based on the unbalanced simulation time and the total length of the MD simulation trajectory, including: Based on the unbalanced simulation time and the total length of the MD simulation trajectory, the remaining simulation trajectory length is calculated; dividing the remaining simulated trajectory length into a plurality of trajectory segments; Calculating the ensemble average corresponding to each trajectory segment in the plurality of trajectory segments to obtain a plurality of ensemble averages; The average of the plurality of ensemble averages is calculated to obtain the average force on the reaction coordinate, and the variance of the plurality of ensemble averages is calculated to obtain the uncertainty of the average force on the reaction coordinate.

8. The method according to any one of claims 1 to 7, characterized in that The material information includes at least one of the following: the crystal structure of the positive electrode material, the Miller index of the surface of the positive electrode material, the composition and proportion of the electrolyte molecules, and the type of dissolved metal atoms. The calculation parameters include at least one of the following: temperature, pressure, time step, trajectory length, equilibrium time, reaction coordinate range and sampling interval.

9. A calculation system for the dissolution of metal atoms at the interface between a battery positive electrode and an electrolyte, characterized in that: The computing system includes: an input module, a structure building module, a molecular dynamics MD simulation module, a post-processing module, and an output module, wherein: The input module is used to obtain material information of the battery positive electrode material and electrolyte molecules, as well as calculation parameters for controlling the calculation simulation process; The structure building module is connected to the input module and is used to construct a corresponding positive electrode material and electrolyte interface structure model based on the material information and the calculation parameters; The MD simulation module is connected to the structure building module, and is used to perform a balanced MD simulation on the interface structure model of the positive electrode material and the electrolyte through an isothermal and isobaric NPT ensemble algorithm to obtain a balanced lattice parameter, and to perform a constrained MD simulation on the interface structure model of the positive electrode material and the electrolyte based on the balanced lattice parameter through an isothermal and isochoric NVT ensemble algorithm to obtain an average force on the reaction coordinate, wherein the balanced lattice parameter is used to reflect the lattice structure state when the interaction between the positive electrode material of the battery and the electrolyte molecules reaches equilibrium under the balanced MD simulation conditions, and the reaction coordinate is defined as the distance between the dissolved metal atom and the reference atom at the interface between the positive electrode material of the battery and the electrolyte; The post-processing module is connected to the MD simulation module and is used to perform thermodynamic integral calculation on the average force to obtain a free energy curve of metal atoms dissolved at the interface between the battery positive electrode material and the electrolyte; The output module is connected to the post-processing module and is used to output the free energy curve, which is used to reflect the energy change of the metal atoms during the dissolution process from the surface of the battery positive electrode material into the electrolyte.

10. The computing system according to claim 9, characterized in that The computing system further includes: The control module is connected to each module in the computing system and is used to control the working process of each module.

11. The computing system according to claim 9, characterized in that: The structure building module constructs a corresponding positive electrode material and electrolyte interface structure model based on the material information and the calculation parameters, and is configured as follows: Based on the material information and the calculation parameters, a symmetrical surface structure model based on the Miller index of the positive electrode material surface is constructed, wherein an upper vacuum layer and a lower vacuum layer exist on both sides of the surface of the positive electrode material, and the same number of electrolyte molecules are added to the upper vacuum layer and the lower vacuum layer, and the number of the electrolyte molecules is calculated based on the volume of the vacuum layer and the density of the electrolyte; The symmetric surface structure model is structurally optimized by a machine learning force field to obtain a corresponding interface structure model of the positive electrode material and the electrolyte, wherein the structural optimization is used to eliminate the unstable contact phenomenon between atoms introduced in the modeling process, and the machine learning force field is obtained by training a force field using a first-principles atomic calculation data set through a machine learning method, and the first-principles atomic calculation data set includes: multiple atomic structures of the battery positive electrode materials generated in a first-principles calculation process and their calculation results, and the output result of the machine learning force field is the total energy of the symmetric surface structure model, the force on each atom, and the lattice stress.

12. The computing system according to claim 9, characterized in that: The MD simulation module performs equilibrium molecular dynamics MD simulation on the interface structure model of the cathode material and the electrolyte by using an isothermal and isobaric NPT ensemble algorithm to obtain equilibrium lattice parameters, and is configured as follows: During the equilibrium molecular dynamics MD simulation, the average potential energy, temperature and lattice constant of the interface structure model of the positive electrode material and the electrolyte are monitored as a function of simulation time, and the unbalanced simulation time during which the interface structure model of the positive electrode material and the electrolyte is in an unbalanced state is determined; Based on the unbalanced simulation time and the total length of the MD simulation trajectory, the equilibrium lattice parameters and uncertainties of the interface structure model between the positive electrode material and the electrolyte are calculated.

13. The computing system according to claim 12, characterized in that: The MD simulation module calculates the equilibrium lattice parameters and uncertainties of the interface structure model of the positive electrode material and the electrolyte based on the unbalanced simulation time and the total length of the MD simulation trajectory, and is further configured as follows: Obtaining the unbalanced simulation time and the total length of the MD simulation trajectory of the interface structure model of the positive electrode material and the electrolyte; Based on the unbalanced simulation time and the total length of the MD simulation trajectory, a remaining simulation trajectory length is calculated; dividing the remaining simulated trajectory length into a plurality of trajectory segments; Calculating an average value of a lattice parameter corresponding to each trajectory segment in the plurality of trajectory segments to obtain a plurality of lattice parameter average values; The average of the plurality of lattice parameter average values ​​is calculated to obtain the equilibrium lattice parameter, and the variance of the plurality of lattice parameter average values ​​is calculated to obtain the uncertainty of the equilibrium lattice parameter.

14. The computing system according to claim 9, characterized in that: The constrained MD simulation includes: an MD simulation with a changed constraint value of the reaction coordinate and an MD simulation with a fixed constraint value of the reaction coordinate. The MD simulation with a changed constraint value of the reaction coordinate is used to continuously increase the constraint value of the reaction coordinate during the simulation process to provide an interface structure during the dissolution of metal atoms; the MD simulation with a fixed constraint value of the reaction coordinate is used to perform configuration sampling of the interface structure under fixed reaction coordinate conditions and calculate the average force acting on the reaction coordinate.

15. The computing system according to claim 14, characterized in that: The MD simulation module performs constrained MD simulation on the interface structure model of the cathode material and the electrolyte based on the equilibrium lattice parameters by using an isothermal and isochoric NVT ensemble algorithm to obtain the average force on the reaction coordinates, and is further configured as follows: During the MD simulation in which the constraint value of the reaction coordinate changes, the coordinates and speed of the dissolved metal atoms are adjusted based on the improved interatomic distance constraint algorithm so that the distance between the metal atom and the reference atom is equal to the constraint target value, and the constraint target value of the reaction coordinate is continuously adjusted so that the reaction coordinate continues to increase after the initial value is balanced for a predetermined time; Based on the interface structure of the metal atom dissolution process provided by the MD simulation with the constraint value of the reaction coordinate changed, an initial structure is obtained, wherein the initial structure is used for the MD simulation with the reference atomic position and the constraint value of the reaction coordinate fixed; After the MD simulation with the constraint value of the reaction coordinate fixed, the average force on the reaction coordinate and the uncertainty of the average force on the reaction coordinate are calculated based on the unbalanced simulation time and the total length of the MD simulation trajectory.

16. The computing system according to claim 15, characterized in that: The MD simulation module calculates the average force on the reaction coordinate and the uncertainty of the average force on the reaction coordinate based on the unbalanced simulation time and the total length of the MD simulation trajectory, and is also configured as follows: Based on the unbalanced simulation time and the total length of the MD simulation trajectory, the remaining simulation trajectory length is calculated; dividing the remaining simulated trajectory length into a plurality of trajectory segments; Calculating the ensemble average corresponding to each trajectory segment in the plurality of trajectory segments to obtain a plurality of ensemble averages; The average of the plurality of ensemble averages is calculated to obtain the average force on the reaction coordinate, and the variance of the plurality of ensemble averages is calculated to obtain the uncertainty of the average force on the reaction coordinate.

17. An electronic device comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein: When the processor executes the computer program, the electronic device implements the method according to any one of claims 1 to 8.

18. A computer program product, characterized in that The invention comprises a computer program which, when executed, causes the method according to any one of claims 1 to 8 to be performed.

19. A computer-readable storage medium storing a computer program, characterized in that: When the computer program is executed by a processor, the method according to any one of claims 1 to 8 is implemented.

Images