Method for predicting ionic conductivity of solid electrolytes using quantum algorithms

Abstract

A method according to the present invention can predict the ionic conductivity of solid electrolytes, the method comprising the steps of: calculating stable structures of the solid electrolytes by using density functional theory (DFT); clustering the stable structures to generate input values for quantum algorithms; and inputting the generated input values to the quantum algorithms to calculate the ionic conductivity of the solid electrolytes.

Description

Method for Predicting Ionic Conductivity of Solid Electrolytes Using Quantum Algorithms

[0001] The present invention provides a method for improving the accuracy of predicting the ionic conductivity of a solid electrolyte by combining Density Functional Theory (DFT) and a quantum algorithm.

[0002] Solid electrolytes are attracting attention as next-generation battery materials due to their advantages over liquid electrolytes, such as high safety, durability, non-toxicity, and high energy density. In particular, solid electrolytes can be used in conjunction with metallic lithium anodes, offering the potential to achieve even higher energy densities.

[0003] However, the low lithium-ion conductivity of solid electrolytes is a significant technical obstacle to commercialization. Currently, the ionic conductivity of solid electrolytes is predicted using Density Functional Theory (DFT), but there is a problem in that the results vary significantly depending on the approximation method of the Exchange-Correlation Functional used.

[0004] For example, Li, a representative solid electrolyte 10 GeP2S 12 When calculating the lithium ion diffusion barrier energy of (LGPS), the lithium ion diffusion barrier energy is calculated to be approximately 0.15 eV when using the PBE approximation, approximately 0.17 eV when using the wB97 approximation, and approximately 0.23 eV when using the lc-wPBE approximation, showing different results depending on the approximation method used. This variability can cause an error of more than 10 times in lithium ion conductivity prediction, and therefore, a prediction method with higher accuracy is required.

[0005] The present invention provides a method for predicting the ionic conductivity of a solid electrolyte with high accuracy by utilizing a quantum algorithm.

[0006] The present invention provides a method for accurately calculating the diffusion characteristics of lithium ions in a solid electrolyte by combining density functional theory and quantum algorithms.

[0007] The present invention provides a method for effectively selecting an active space related to lithium ion diffusion in the structure of a solid electrolyte and converting it into an input value for a quantum algorithm.

[0008] The present invention provides a method for calculating the ionic conductivity of a solid electrolyte by utilizing quantum algorithms such as a Variational Quantum Eigensolver (VQE), Quantum Filter Diagonalization (QFD), or Quantum Phase Estimation (QPE).

[0009] The technical problems to be solved in this document are not limited to those mentioned above, and other technical problems not mentioned will be clearly understood by those skilled in the art to which this invention belongs from the description below.

[0010] A method for predicting the ionic conductivity of a solid electrolyte according to one embodiment of the present invention may include: a step of calculating stable structures of the solid electrolyte using Density Functional Theory (DFT); a step of clustering the stable structures to generate input values ​​for a quantum algorithm; and a step of inputting the generated input values ​​into the quantum algorithm to calculate the ionic conductivity of the solid electrolyte.

[0011] The step of calculating the above stable structures may include the step of calculating the diffusion path of lithium ions in the solid electrolyte using the Nudged Elastic Band (NEB) method.

[0012] The clustering step may include a step of obtaining a plurality of cross-sections of the lithium ion diffusion path by performing a cylindrical cut centered on the diffusion channel of the solid electrolyte.

[0013] The clustering step may further include the step of selecting an active space corresponding to a plurality of molecular orbitals that are directly involved in the formation and destruction of chemical bonds around the lithium ion diffusion path in each of the plurality of cross-sections.

[0014] The step of selecting the active space may further include the step of selecting orbitals for each of the plurality of cross-sections in which the occupancy number of MP2 natural orbitals calculated from the def2-SVP basis function is 0.1 to 1.9 as the active space.

[0015] The above clustering step may include a step of calculating an integral value based on molecular orbitals.

[0016] The step of calculating the integral value further includes the step of calculating a 1-electron integral value and a 2-electron integral value based on molecular orbitals for Hamiltonian configuration; wherein the 1-electron integral value is the kinetic energy of an electron in the active space and the attractive energy between the nucleus and the electron, and the 2-electron integral value may be the repulsive energy between a plurality of electrons in the active space.

[0017] The above quantum algorithm may include at least one of a Variational Quantum Eigensolver (VQE), Quantum Filter Diagonalization (QFD), or Quantum Phase Estimation (QPE).

[0018] The step of calculating the ionic conductivity of the solid electrolyte may include calculating the ionic conductivity of the solid electrolyte by calculating the barrier energy between the initial state and the transition state of the solid electrolyte.

[0019] In a storage medium storing at least one instruction according to one embodiment of the present invention, when the at least one instruction is executed by a processor, the processor may perform the steps of: calculating stable structures of a solid electrolyte using Density Functional Theory (DFT); clustering the stable structures to generate input values ​​for a quantum algorithm; and inputting the generated input values ​​into the quantum algorithm to calculate the ionic conductivity of the solid electrolyte.

[0020] When the above at least one instruction is executed by the processor, the processor may calculate the diffusion path of lithium ions in the solid electrolyte using a Nudged Elastic Band (NEB) method in the step of calculating the stable structures.

[0021] When the above at least one instruction is executed by the processor, the processor may perform a cylindrical cut centered on the diffusion channel of the solid electrolyte in the clustering step to obtain a plurality of cross-sections of the lithium ion diffusion path.

[0022] When the above at least one instruction is executed by the processor, the processor may select an active space, which is a plurality of molecular orbitals directly involved in the formation and destruction of chemical bonds around the lithium ion diffusion path in each of the plurality of cross-sections during the clustering step.

[0023] When the above at least one instruction is executed by the processor, the processor may configure the active space by selecting orbitals with an occupancy number between 0.1 and 1.9 from among the natural orbitals included in each of the plurality of cross-sections obtained by MP2 / def2-SVP calculation.

[0024] When the above at least one instruction is executed by the processor, the processor may calculate a molecular orbital-based integral value in the clustering step.

[0025] When the above at least one instruction is executed by the processor, the processor calculates a molecular orbital-based 1-electron integral and a 2-electron integral for a Hamiltonian configuration when calculating the integral value, the 1-electron integral may be the kinetic energy of an electron in the active space and the attractive energy between the nucleus and the electron, and the 2-electron integral may be the repulsive energy between a plurality of electrons in the active space.

[0026] When the above at least one instruction is executed by the processor, the processor may cause the quantum algorithm to perform energy calculations including at least one of a Variational Quantum Eigensolver (VQE), Quantum Filter Diagonalization (QFD), or Quantum Phase Estimation (QPE).

[0027] When the above at least one instruction is executed by the processor, the processor may calculate the ionic conductivity of the solid electrolyte by calculating the barrier energy between the initial state and the transition state of the solid electrolyte in the step of calculating the ionic conductivity of the solid electrolyte.

[0028] According to one embodiment of the present invention, the ionic conductivity of a solid electrolyte can be predicted with higher accuracy than the conventional density functional theory method.

[0029] According to one embodiment of the present invention, the barrier energy of lithium ion diffusion in a solid electrolyte can be accurately calculated without relying on the selection of an exchange-correlation function.

[0030] According to one embodiment of the present invention, the electronic correlation effect of a solid electrolyte can be considered more precisely through a quantum algorithm.

[0031] According to one embodiment of the present invention, the time and cost of developing battery materials can be reduced by accurately predicting the performance of a new solid electrolyte material in advance using quantum computing technology.

[0032] According to one embodiment of the present invention, by quantitatively analyzing the relationship between the structural characteristics and ionic conductivity of a solid electrolyte, guidelines necessary for designing a higher-performance solid electrolyte material can be provided.

[0033] FIG. 1 is an overall flowchart of a method for predicting the ionic conductivity of a solid electrolyte using a quantum algorithm according to one embodiment of the present invention.

[0034] Figure 2 is a flowchart illustrating the steps of applying density functional theory in Figure 1.

[0035] Figure 3 is a flowchart illustrating the clustering and input value generation steps of Figure 1.

[0036] Figure 4 is a flowchart illustrating the steps for applying the quantum algorithm of Figure 1.

[0037] FIG. 5 is a block diagram of an ion conductivity prediction device for a solid electrolyte according to one embodiment of the present invention.

[0038] The embodiments described in this document and the configurations illustrated in the drawings are merely preferred examples of the disclosed invention, and various modifications that may replace the embodiments and drawings of this specification may exist at the time of filing this application.

[0039] The terms used in this document are for describing the embodiments and are not intended to limit or restrict the disclosed invention.

[0040] For example, in this specification, singular expressions may include plural expressions unless the context clearly indicates otherwise.

[0041] In this document, each of the phrases such as "A or B", "at least one of A and B", "at least one of A or B", "A, B or C", "at least one of A, B and C", and "at least one of A, B, or C" may include any one of the items listed together in the corresponding phrase, or all possible combinations thereof.

[0042] The term "and / or" includes a combination of multiple related described components or any of the multiple related described components. For example, "A and / or B" may include only "A," only "B," or both "A and B."

[0043] Additionally, terms such as “include” or “have” are intended to express the existence of the features, numbers, steps, actions, components, parts, or combinations thereof described in the specification, and do not exclude the additional existence or addition of one or more other features, numbers, steps, actions, components, parts, or combinations thereof.

[0044] When it is said that a component is "connected," "combined," "supported," or "in contact" with another component, this includes not only cases where the components are directly connected, combined, supported, or in contact, but also cases where they are indirectly connected, combined, supported, or in contact through a third component.

[0045] When it is said that a component is located "on" another component, this includes not only cases where one component is in contact with the other, but also cases where another component exists between the two components.

[0046] Meanwhile, terms such as "front," "rear," "left," "right," "top," and "bottom" used in the following description are defined based on the drawings; however, the shape and position of each component are not limited by these terms. For example, the front side may be defined as the +X side and the rear side as the -X side. For example, based on the drawings, the right side may be defined as the +Y side and the left side as the -Y side. For example, based on the drawings, the top side may be defined as the +Z side and the bottom side as the -Z side.

[0047] In addition, terms including ordinal numbers, such as "first," "second," etc., are used to distinguish one component from another and do not limit the components.

[0048] In addition, terms such as "~part," "~unit," "~block," "~part," and "~module" may refer to a unit that processes at least one function or operation. For example, the terms may refer to at least one piece of hardware such as an FPGA (field-programmable gate array) or ASIC (application specific integrated circuit), at least one piece of software stored in memory, or at least one process processed by a processor.

[0049] An embodiment of the disclosed invention is described in detail below with reference to the attached drawings. Identical reference numbers or symbols in the attached drawings may indicate parts or components that perform substantially the same function.

[0050] The operating principle and embodiments of the present invention will be described below with reference to the attached drawings.

[0051] FIG. 1 is an overall flowchart of a method for predicting the ionic conductivity of a solid electrolyte using a quantum algorithm according to one embodiment of the present invention.

[0052] Referring to FIG. 1, the method for predicting the ionic conductivity of a solid electrolyte using a quantum algorithm according to the present invention may include a step of applying density functional theory (10), a clustering and input value generation step (20), and a quantum algorithm application step (30).

[0053] In the density functional theory application step (10), the density functional theory (DFT) method is used to calculate the stable structures of the solid electrolyte. The density functional theory method can be a method for calculating material properties while reducing computational costs by expressing many-body quantum mechanics problems as a function of electron density.

[0054] The basic principle of density functional theory is based on the Hohenberg-Kohn theorem, and the total energy of a system can satisfy the following mathematical equation 1 as a function of electron density ρ(r).

[0055] [Mathematical Formula 1]

[0056]

[0057] T[ρ] is the kinetic energy of the electron (eV), V ne [ρ] is the electrostatic interaction energy (eV) between the nucleus and the electron, V ee [ρ] is the Coulomb interaction energy (eV) between electrons, and E xc[ρ] can be the exchange-correlation energy (eV).

[0058] In the present invention, the diffusion barrier energy of lithium ions in a solid electrolyte can be calculated using various exchange-correlation function approximation methods. Table 1 below shows Li, a representative solid electrolyte material. 10 GeP2S 12 This is the result calculated using various approximation methods for (LGPS).

[0059] Exchange-correlation function approximation method lithium ion diffusion barrier energy (E a , eV)PBE0.15wB970.17Ic-wPBE0.23B3LYP-D30.22wB97M-V0.18MP2 (reference value)0.32

[0060] Referring to Table 1, the lithium ion diffusion barrier energy calculated according to the approximation method of the exchange-correlation function used varies from 0.15 eV to 0.23 eV, which can show a significant difference from the MP2 reference value of 0.32 eV.

[0061] The MP2 reference value may be the result of calculations based on Møller-Plesset perturbation theory 2nd order. This method may be one of the accurate methods for calculating electron correlation.

[0062] In Table 1 above, the MP2 reference value is 0.32 eV, and the MP2 reference value may be a reference value for comparison with the diffusion barrier energy height (e.g., 0.15–0.23 eV) calculated by various density functional theory calculation methods (e.g., PBE, wB97, lc-wPBE, etc.).

[0063] MP2 values ​​can calculate electronic correlations more accurately than density functional theory. Therefore, MP2 values ​​can be a more reliable reference value for calculating the lithium ion diffusion barrier energy height of solid electrolytes.

[0064] The Eyring-Polanyi equation can express the rate constant of a chemical reaction as a function of temperature and activation energy (barrier energy). In solid electrolytes, this equation can be used to calculate the diffusion rate of lithium ions, and the lithium ion diffusion barrier energy height (E) calculated by density functional theory methods a By applying ) to this equation, the diffusion coefficient (D) can be obtained, and through this, the ionic conductivity (σ) can be finally predicted.

[0065] LGPS is Li 10 GeP2S 12 It is a solid electrolyte with a specific chemical formula and is one of the materials with the highest ionic conductivity among currently known solid electrolytes. It has a tetragonal structure, and lithium ions can move primarily through channels formed along the c-axis.

[0066] Lithium ions diffuse within this channel through cooperative motion, which is induced by Coulomb repulsion between lithium ions. The LGPS structure contains multiple (e.g., 32) possible lithium sites that can be located at Wyckoff positions such as 4c, 4d, 8f, and 16h, and only some of these (e.g., 20) can actually be occupied by lithium ions.

[0067] In the clustering and input value generation step (20), stable structures obtained by density functional theory calculations can be clustered to generate input values ​​suitable for the quantum algorithm. In this process, cylindrical cutting can be performed around the diffusion channels of the solid electrolyte.

[0068] Cylindrical cleavage can be a method for generating clusters containing only atoms within a certain radius centered on the lithium ion diffusion channel. Although these generated clusters are smaller in size than the overall periodic structure, they can retain the key characteristics of lithium ion diffusion.

[0069] The present invention may be an experiment in which a cluster composed of a plurality of atoms (e.g., about 62) is generated from an LGPS solid electrolyte. This cluster includes a lithium diffusion channel and surrounding atoms and may be suitable for studying the diffusion pathway and barrier energy of lithium ions.

[0070] In addition, the process may include selecting active spaces directly related to lithium ion diffusion within the cluster. Active spaces are sets of molecular orbitals that quantum algorithms must intensively compute, which can improve computational efficiency and accuracy.

[0071] The present invention may configure an active space by selecting orbitals with an occupancy number between 0.1 and 1.9 from natural orbitals obtained by calculating MP2 / def2-SVP.

[0072] In the quantum algorithm application step (30), the generated input value can be input into the quantum algorithm to calculate the ionic conductivity of the solid electrolyte. The quantum algorithm may be an algorithm that performs electronic structure calculations in a quantum computer including a quantum simulator and quantum hardware that operates by utilizing quantum mechanical principles.

[0073] The quantum algorithm used in the present invention is as follows:

[0074] A Variational Quantum Eigensolver (VQE) can be a method for calculating the ground state energy of a Hamiltonian by combining parameterized quantum circuits with classical optimization algorithms.

[0075] Quantum Filter Diagonalization (QFD) can be a method that uses a time-evolution operator to generate a Krylov subspace and, through this, computes the eigenvalues ​​and eigenvectors of the Hamiltonian.

[0076] Quantum Phase Estimation (QPE) is a method that uses quantum circuits to estimate the eigenvalues ​​of unitary operators and can be a method for accurately calculating the eigenenergies of the Hamiltonian.

[0077] The specific implementation methods and characteristics of these quantum algorithms are explained in detail in the quantum algorithm application stage.

[0078] The energy values ​​calculated through this quantum algorithm can more accurately predict the diffusion barrier energy of lithium ions in a solid electrolyte, and based on this, the ionic conductivity can be calculated using the Eyring-Polany equation.

[0079] Figure 2 is a flowchart illustrating the steps of applying density functional theory in Figure 1.

[0080] Referring to FIG. 2, the density functional theory application step (10) includes a solid electrolyte structure optimization step (11), a lithium ion diffusion path determination step (12), a Nudged Elastic Band (NEB) method application step (13), a lithium ion path-specific stable structure calculation step (14), and a lithium ion diffusion energy barrier calculation step (15).

[0081] In the solid electrolyte structure optimization step (11), the basic structure of the solid electrolyte can be optimized using a density functional theory method. In this step, the energy of the structure, including the lattice constant, atomic position, bond length, bond angle, etc. of the solid electrolyte, can be adjusted to a value that minimizes it.

[0082] In the present invention, 'stable structure' refers to atomic structures in which the energy has a local minimum when the lithium ion diffusion barrier energy height is calculated using the density functional theory method, and the interatomic force is less than or equal to a predefined threshold value (e.g., 0.01 eV / Å). In particular, structures corresponding to the initial state, transition state, and final state in the diffusion path of lithium ions within a solid electrolyte are included.

[0083] The initial state is the original lattice site where the lithium ion is stably located, which can correspond to a local energy minimum.

[0084] The transition state is the energy maximum that lithium ions pass through while moving along a diffusion path, and this can be a saddle point on the diffusion path.

[0085] The final state is a new stable site reached by the movement of lithium ions, which can correspond to another local energy minimum.

[0086] The energy difference between these three states, particularly the energy difference between the transition state and the initial state, can be the diffusion barrier energy.

[0087] Li 10 GeP2S 12 In the case of solid electrolytes such as (LGPS), they have a tetragonal structure and can be optimized based on lattice parameters a = 8.72 Å and b = 12.64 Å.

[0088] Lattice parameters are values ​​that define the size and shape of the basic unit cell of a crystal structure and can consist of length parameters (a, b, c) and angle parameters (α, β, γ). Length parameters represent the lengths in the three axial directions of the unit cell, and angle parameters represent the angles between the three axes.

[0089] Structural optimization can generally be performed iteratively by setting specific threshold values ​​for energy, force, pressure, etc., until convergence criteria are satisfied.

[0090] For example, in the structural optimization process, multiple threshold values, namely system energy change, interatomic force, and system pressure, are each 10 -5 Ry or less, 10 -4 Iterative calculations can be performed by adjusting atomic positions and lattice parameters until Ry / au and 0.5 kBar are simultaneously satisfied. This optimized structure can serve as the basis for calculating the diffusion path of lithium ions.

[0091] In the lithium ion diffusion path determination step (12), possible paths through which lithium ions can move within the solid electrolyte can be determined. In solid electrolytes such as LGPS, lithium ions mainly move through channels formed along the c-axis.

[0092] These channels connect lithium sites coordinated in a tetrahedral shape, and these sites can be classified into 4c, 4d, 8f, 16h, etc. based on Wyckoff positions.

[0093] To determine the diffusion path, the initial and final states must first be defined. The initial state is the original site where the lithium ion is stably located, and the final state can be the new stable site after the lithium ion has moved.

[0094] For example, by considering cooperative movement in the lithium channel within the LGPS, an initial state and a final state containing three lithium ions can be set.

[0095] The cooperative movement of lithium ions described above may mean a phenomenon in which a single lithium ion does not move independently, but rather multiple lithium ions within the channel move sequentially or simultaneously through Coulomb interactions with one another.

[0096] In the NEB method application step (13), the diffusion path and barrier energy of lithium ions can be calculated using the NEB method. The NEB method is a method for finding the Minimum Energy Path (MEP) between two stable states (initial state and final state).

[0097] The accuracy and computational efficiency of the NEB method can depend on computational parameters such as the number of images used and the power threshold. If the number of images is insufficient, the diffusion path cannot be accurately represented, while if it is too large, computational costs may increase unnecessarily. The appropriate number of images can be determined based on the complexity of the system to be computed and the required accuracy.

[0098] Similarly, the force threshold should be set by considering the balance between path convergence and computational efficiency. These parameters can be selected to obtain the energy profile of the lithium ion diffusion path with reliable accuracy.

[0099] In this invention, the diffusion path and barrier energy of lithium ions were accurately calculated by appropriately adjusting the NEB calculation parameters based on the aforementioned principles.

[0100] In the step (14) of calculating the stable structure for each lithium ion path, various structures in which lithium ions can be located along the diffusion path obtained by the NEB method can be calculated. In this step, the position of the lithium ion at each point on the diffusion path and the rearrangement of other atoms accordingly can be considered.

[0101] The structure of the transition state can be the structure of the point with the highest energy during the diffusion path. To accurately determine the transition state structure, the image with the highest energy among those obtained from NEB calculations can be identified, and additional structure optimization can be performed using this as a starting point.

[0102] In this process, a special algorithm (e.g., the dimer method) can be used to accurately find the saddle point of the energy surface.

[0103] In the lithium ion diffusion energy barrier calculation step (15), the energies of the initial state, transition state, and final state are calculated, and the diffusion barrier energy can be determined from this. The diffusion barrier energy is defined as the energy difference between the transition state and the initial state, which can represent the barrier energy that lithium ions must overcome to move from one site to another.

[0104] In an embodiment of the present invention, the diffusion barrier energy of LGPS can be calculated using various exchange-correlation function approximation methods (e.g., PBE, wB97, lc-wPBE, B3LYP-D3, wB97M-V) as shown in Table 1 above.

[0105] The diffusion barrier energy calculated in this way can be converted into the diffusion coefficient and ionic conductivity of lithium ions through the Eyring-Polanyi equation. The diffusion coefficient (D) can satisfy the following mathematical equation 2.

[0106] [Mathematical Formula 2]

[0107]

[0108] Here, D0 is the pre-exponential factor, E a is the lithium ion diffusion barrier energy (eV), k B ε is the Boltzmann constant, and T is the absolute temperature (k).

[0109] Referring to Equation 2, when the diffusion barrier activation energy increases by 0.1 eV, the diffusion coefficient (D) can decrease by about 50 times.

[0110] In the lithium ion diffusion energy barrier calculation step (15), the lithium ion diffusion barrier energy can be determined using the accurate energy values ​​of the initial state, transition state, and final state provided in the previous step.

[0111] The final output of this stage is the lithium ion diffusion barrier energy (E a ) is passed to the clustering and input value generation step (20) of FIG. 1, and can subsequently be used as a reference point for configuring a cluster model for a quantum algorithm and selecting an active space.

[0112] The diffusion barrier energy of lithium ions is used as an important indicator to determine the size and boundaries of the active space in the subsequent clustering and input generation stages, and the precise barrier energy value can also be utilized as a comparison standard to verify the performance of the quantum algorithm.

[0113] In summary, in the solid electrolyte structure optimization step (11), density functional theory can be used to output an optimized structure including the lattice constant, atomic position, bond length, bond angle, etc. of the solid electrolyte.

[0114] This optimized structure data is input into the lithium ion diffusion path determination step (12) and can serve as a basis for determining possible paths through which lithium ions can move. The optimized structure can identify possible locations and movement channels of lithium ions within the solid electrolyte.

[0115] In the lithium ion diffusion path determination step (12), atomic coordinate data of the initial and final states can be output along with possible movement paths of lithium ions within the solid electrolyte. This data can be input into the NEB method application step (13) and used as the starting and ending points for NEB calculation.

[0116] In the NEB method application step (13), structure and energy data of various intermediate states (images) along the lithium ion diffusion path can be output. In particular, the point with the highest energy along the diffusion path can be identified.

[0117] This data is input into the lithium-ion pathway-specific stable structure calculation step (14), which allows for more precise optimization of the atomic arrangement in each intermediate state. The result of the NEB calculation provides the approximate location of the transition state, which allows for finding a more accurate transition state structure in the next step.

[0118] In the lithium ion path-specific stable structure calculation step (14), the detailed atomic structure of each point on the diffusion path, particularly the transition state, can be output. This data is input into the lithium ion diffusion energy barrier calculation step (15) to calculate the accurate energy difference between the initial state and the transition state.

[0119] In the lithium ion diffusion energy barrier calculation step (15), the lithium ion diffusion barrier energy (E a ) can be output. This data is passed to the clustering and input value generation step (20), which is the next step of FIG. 1, and can be used as a reference point for configuring a cluster model for a quantum algorithm and selecting an active space.

[0120] Figure 3 is a flowchart illustrating the clustering and input value generation steps of Figure 1.

[0121] Referring to FIG. 3, the clustering and input value generation step (20) includes a solid electrolyte clustering step (21), an active space selection step (22), a 1-electron integral value calculation step (23), a 2-electron integral value calculation step (24), and a quantum algorithm input value generation step (25).

[0122] In the solid electrolyte clustering step (21), a cluster model can be generated from a periodic model. The cluster model is smaller in size than the periodic model but must maintain the key characteristics of the lithium ion diffusion mechanism.

[0123] For example, a cluster composed of multiple atoms (e.g., about 62) can be generated from an LGPS solid electrolyte. This cluster includes lithium diffusion channels and surrounding atoms and may be suitable for studying the diffusion pathways and barrier energies of lithium ions. In the cluster model, only the diffusing lithium ions and coordinated sulfur atoms are allowed to relax, while the remaining atoms are immobilized to minimize surface effects.

[0124] The solid electrolyte clustering step (21) can use the lithium ion diffusion path and barrier energy data calculated in FIG. 2 as direct inputs. In this step, based on the atomic structure of the initial state and transition state confirmed by density functional theory, a small-sized cluster model can be generated that can increase computational efficiency while maintaining the key characteristics of lithium ion diffusion.

[0125] More specifically, cylindrical cleavage is performed around the lithium ion diffusion channel, and atoms that strongly interact with lithium ions in the transition state are preferentially included.

[0126] In addition, the size of the cluster and the number of atoms to be included are adjusted according to the magnitude of the barrier energy calculated in Figure 2, allowing for position optimization of sulfur atoms directly coordinating with lithium ions and fixing the remaining atoms to minimize boundary effects.

[0127] In the active space selection step (22), molecular orbitals directly related to lithium ion diffusion in the cluster model can be selected to configure the active space. The active space is a set of molecular orbitals that the quantum algorithm must intensively compute, thereby increasing computational efficiency and accuracy.

[0128] Selecting the active space involves obtaining the natural orbital (NO) and its occupancy through MP2 / def2-SVP calculation. Orbitals with occupancy very close to 0 (completely empty state) or 2 (completely filled state) may have a low contribution to electron correlation effects.

[0129] On the other hand, orbitals with occupancy numbers between 0.1 and 1.9 contribute significantly to both static correlation and dynamic correlation, so orbitals in this range can be selected to perform the calculation of the fully active space magnetic coordinate field (CASSCF), and the natural orbitals obtained as a result can be selected as the final active space.

[0130] In addition, during the process of selecting an active space, the orbital with the lowest occupancy among natural orbitals with an occupancy of 0.1 or more and 1 or less can be selected preferentially, and orbitals with an occupancy of more than 1 and less than 2 having electronic structural characteristics combined with it can be identified and added through visual analysis, etc.

[0131] The orbital set configured in this way is used in the calculation of the fully active space magnetic unconstrained field (CASSCF), and the natural orbitals derived from the calculation results can be determined as the final active space.

[0132] Static correlation is an electronic correlation effect arising from multiple determinant characteristics, and it can be important in systems where strong electron-electron interactions exist and cannot be sufficiently described by a single Slater determinant.

[0133] Dynamic correlation is a correlation effect related to the instantaneous movement of electrons, which may describe the phenomenon of electrons avoiding each other due to Coulomb repulsion.

[0134] Accordingly, in the present invention, an orbital with an occupancy range of 0.1 to 1.9 is used as the active space.

[0135] For example, active spaces such as CAS(2,2), CAS(4,4), CAS(6,6), and CAS(8,8) can be configured depending on the size of the active space. Here, the first number may represent the number of electrons in the active space, and the second number may represent the number of molecular orbitals.

[0136] In the step (23) for calculating the 1-electron integral value, the 1-electron part of the Hamiltonian required as an input value for the quantum algorithm can be calculated. The 1-electron integral value can satisfy the following mathematical formula 3.

[0137] [Mathematical Formula 3]

[0138]

[0139] Here h pq ε is the 1-electron integral (eV) between molecular orbitals p and q, p and q are molecular orbitals, h is the Hamiltonian operator, T is the electron kinetic energy operator, V ne is a nuclear-electron attraction potential operator.

[0140] The 1-electron integral represents the kinetic energy of the electron and the Coulomb attraction between the nucleus and the electron, and is an important component of the Hamiltonian.

[0141] In the step (24) for calculating the 2-electron integral value, the 2-electron part of the Hamiltonian required as an input value for the quantum algorithm can be calculated. The 2-electron integral value can satisfy the following mathematical formula 4.

[0142] [Mathematical Formula 4]

[0143]

[0144] g pqrs ε is the 2-electron integral (eV) between molecular orbitals p, q, r, and s, φ is the molecular orbital function, r1 and r2 are the position vectors of the two electrons, and r 12 is the distance (Å) between two electrons.

[0145] The 2-electron integral represents the Coulomb repulsion between electrons, which is essential for accurately calculating electron correlation effects.

[0146] In the quantum algorithm input value generation step (25), data can be converted into a form that can be input to the quantum algorithm based on the calculated 1-electron and 2-electron integral values. In this step, the Hamiltonian of the active space is mapped to a quantum bit (qubit) and expressed in a form that the quantum algorithm can process.

[0147] Methods for mapping Hamiltonians to qubits include the Jordan-Wigner transformation, the Bravyi-Kitaev transformation, and parity mapping. For example, an active space with n molecular orbitals can be mapped to 2n qubits. This is because each spin-orbital (each space orbital with an alpha spin and a beta spin) corresponds to one qubit.

[0148] The mapped Hamiltonian can be expressed as a sum of Pauli operators as shown in Equation 5 below.

[0149] [Mathematical Formula 5]

[0150]

[0151] Here, H is the mapped Hamiltonian (eV), c j is the coefficient, P j is the tensor product of Pauli operators (combination of I, X, Y, Z).

[0152] The Hamiltonian transformed in this way can be used as a direct input to a quantum algorithm.

[0153] The input values ​​generated through this process can be used in the next step, the application of a quantum algorithm, to accurately calculate the diffusion barrier energy of lithium ions within the solid electrolyte.

[0154] In summary, in the solid electrolyte clustering step (21), a cluster model can be generated from a periodic model. In this step, cylindrical cutting can be performed around the diffusion channels of the solid electrolyte by utilizing the lithium ion diffusion path and barrier energy data calculated in FIG. 2.

[0155] For example, a cluster containing a lithium diffusion channel and surrounding atoms composed of about 62 atoms can be generated from the LGPS solid electrolyte. This generated cluster model can serve as the basis for the next step, active space selection.

[0156] In the active space selection step (22), the active space can be configured by selecting molecular orbitals directly related to lithium ion diffusion in the cluster model. In this step, the natural orbital (NO) and its occupancy number are obtained through MP2 / def2-SVP calculation, and the active space can be configured by selecting orbitals with an occupancy number between 0.1 and 1.9.

[0157] Depending on the size of the active space, various active space configurations such as CAS(2,2), CAS(4,4), CAS(6,6), and CAS(8,8) are possible, which can determine the range and complexity of the electronic integral value to be calculated in the next step.

[0158] In the step (23) for calculating the 1-electron integral value, the 1-electron part of the Hamiltonian required as an input value for the quantum algorithm can be calculated. The 1-electron integral value is a value representing the kinetic energy and nuclear-electron attraction between molecular orbitals p and q, and this integral value satisfies the mathematical equation 3 described above. The 1-electron integral value calculated in this way provides basic energy information of electrons in the active space.

[0159] In the step (24) for calculating the 2-electron integral value, the 2-electron part of the Hamiltonian required as an input value for the quantum algorithm can be calculated. The 2-electron integral value is a value representing the electron-electron Coulomb repulsion between four molecular orbitals p, q, r, and s, and satisfies the above-described mathematical formula 4.

[0160] The 2-electron integral is an essential element for accurately calculating electron correlation effects, and can capture complex interactions between electrons, particularly during the lithium ion diffusion process. Along with the 1-electron integral, the 2-electron integral is required to fully describe the Hamiltonian and is key data directly input into quantum algorithms.

[0161] In the quantum algorithm input value generation step (25), data can be converted into a form that can be input into the quantum algorithm based on the calculated 1-electron and 2-electron integral values.

[0162] In this step, the Hamiltonian of the active space is mapped to quantum bits (qubits), and an active space with n molecular orbitals can be mapped to 2n qubits. Hamiltonian mapping methods that can be used include the Jordan-Wigner transformation, the Bravi-Kitaev transformation, and parity mapping.

[0163] The mapped Hamiltonian satisfies the above-described mathematical equation 5 and can be expressed as a sum of Pauli operators, and the Hamiltonian converted into a sum of Pauli operators is used as a direct input in the quantum algorithm application step (30) of FIG. 4 described later to accurately calculate the diffusion barrier energy of lithium ions in the solid electrolyte.

[0164] Figure 4 is a flowchart illustrating the steps for applying the quantum algorithm of Figure 1.

[0165] Referring to FIG. 4, the quantum algorithm application step (30) includes a molecular orbital-based integral input step (31), a quantum algorithm selection and application step (32), and an energy result output step (33).

[0166] In the molecular orbital-based integral input step (31), molecular orbital-based integral values ​​calculated in the clustering step can be provided as input to the quantum algorithm. These integral values ​​form a Hamiltonian, through which the energy of the system can be calculated.

[0167] In the molecular orbital-based integral input step (31), the Hamiltonian in the form of a Pauli operator generated in the last step of FIG. 3 can be used as an input to the quantum algorithm. This Hamiltonian can be expressed in a second-order quantization form including 1-electron and 2-electron terms as shown in Equation 6 below.

[0168] [Mathematical Formula 6]

[0169]

[0170] H is the Hamiltonian (eV) of the system, h pq ε = 1-electron integral (eV), g pqrs is the 2-electron integral value (eV), a †p is the production operator for the p-th spin orbital, a †q is the production operator for the q-th spin orbital, a q is the annihilation operator for the q-th spin orbital, a r is the annihilation operator for the r-th spin orbital and a s is the annihilation operator for the s-th spin orbital.

[0171] In the quantum algorithm selection and application step (32), a calculation is performed using the selected quantum algorithm. The characteristics of the three quantum algorithms are as follows:

[0172] Variational quantum eigenvalue analyzers can operate at shallow circuit depths, making them suitable for current noisy intermediate-scale quantum (NISQ) devices. In implementation, quantum number conservation (QNP) ancepts are used to conserve particle number and spin, thereby restricting the search to a physically meaningful solution space.

[0173] The variational quantum eigenvalue analyzer adjusts the circuit depth according to the size of the active space and uses a minimum of n / 2 circuit layers for a CAS(n,n) active space. For the classic optimization algorithm, the BFGS or Adam optimizer is used, and the initial parameters are derived from the results of MP2 or CCSD calculations to accelerate the optimization process.

[0174] Quantum filter diagonalization is more efficient for calculating energy barriers because it can simultaneously calculate the energies of the ground state and excited state without an optimization process. The Trotter step size (△t) is adjusted to the energy scale of the system, and the Krylov subspace size is selected proportionally to the active space size, but includes at least 8 states. Second-order Trotter decomposition is used for the Trotterization of the time evolution operator to improve accuracy.

[0175] Quantum phase estimation can provide the most accurate results among the three algorithms, but it may require the deepest circuitry and a quantum computer capable of error correction.

[0176] For example, calculating the lithium ion barrier for a CAS(8,8) active space using a variational quantum eigenvalue analyzer and quantum number conservation (QNP) answers allows the lithium ion diffusion barrier of the LGPS to be calculated with an error of within 3% compared to the MP2 reference value using multiple layers of quantum circuits (e.g., 7) and multiple parameters (e.g., 98).

[0177] The aforementioned Ansatz refers to the trial wave function or quantum circuit structure used in quantum algorithms, and can be used as an initial estimate for energy minimization in variational calculus. In other words, in quantum algorithms, Ansatz represents a parameterized quantum circuit structure for preparing a quantum state.

[0178] The aforementioned quantum number-conserving anthts are special types of anthts that conserve electron number and spin, and may be particularly suitable for quantum calculations of chemical systems.

[0179] Using a variational quantum eigenvalue analyzer (VQE) may include steps of preparing an initial quantum state (e.g., a Hartree-Foc state), applying a parameterized quantum circuit (e.g., Anserts), measuring energy expectations, finding parameters that minimize energy using classical optimization algorithms, and updating parameters and repeating the process:

[0180] For example, variational quantum eigenvalue analyzer calculations can be performed using quantum number conservation ancepts. Quantum number conservation ancepts are suitable for chemical systems because they conserve electron number and spin, and can use two 4-qubit gates as the basic unit. These ancepts can increase accuracy by increasing the number of layers as the size of the active space increases.

[0181] Using Quantum Filter Diagonalization (QFD) involves preparing the initial quantum state and the time evolution operator e (-i△tH) It may include steps of generating a Krylov subspace by applying powers of, calculating Hamiltonian and nested matrix elements in the subspace, and solving the generalized eigenvalue problem.

[0182] Unlike variational quantum eigenvalue analyzers, this method can calculate the energies of the ground and excited states without an optimization process, and generally may require deeper quantum circuits.

[0183] Using quantum phase estimation (QPE) may include the steps of preparing an initial state (e.g., an approx. of the ground state), applying a phase estimation circuit, and calculating the system's eigenenergies through measurement.

[0184] Quantum Filter Diagonalization (QFD) is more efficient for energy barrier calculations because it can simultaneously calculate the energies of the ground state and excited state without an optimization process. However, it requires deeper circuitry than VQE and demands more quantum gate operations.

[0185] Quantum Phase Estimation (QPE) can provide the most accurate results among the three algorithms, but it requires the deepest circuitry and a quantum computer capable of error correction. Therefore, it is the most difficult method to realize at the current level of technology.

[0186] In the energy result output step (33), an energy value calculated through a quantum algorithm can be output. Specifically, the electron energy of the initial state and the transition state can be calculated, and the diffusion barrier of lithium ions can be determined through the difference.

[0187] Lithium ion diffusion barrier height (E a ) can be calculated as shown in the following mathematical formula 7.

[0188] [Mathematical Formula 7]

[0189]

[0190] Here, E a is the lithium ion diffusion barrier height (eV), E TS is the electron energy of the transition state (eV) and E I is the electron energy (eV) of the initial state.

[0191] The calculated diffusion barrier can be related to ionic conductivity through the Eyring-Polany equation as shown in Equation 8 below.

[0192] [Mathematical Formula 8]

[0193]

[0194] Here, σ is ionic conductivity (S / cm), n is the charge number, e is the electron charge (C), and k B ε is the Boltzmann constant (eV / K), T is the absolute temperature (K), D is the diffusion coefficient (cm² / s), and c is the concentration of mobile ions (mol / cm³).

[0195] The diffusion coefficient (D) can satisfy Equation 2 described in Figure 2 above. The ionic conductivity calculated in this way can be directly used to predict and evaluate the performance of the solid electrolyte.

[0196] For example, ion conductivity calculations are the height of the lithium ion diffusion barrier (E) between the initial state and the transition state. a It may include the step of calculating ) through a quantum algorithm, the step of calculating the diffusion coefficient (D) using the Eyring-Polany equation, and the step of calculating the ionic conductivity (σ) using the Nernst-Einstein relationship.

[0197] When the energy difference between the initial state and the transition state calculated by the quantum algorithm is 0.31 eV, the diffusion coefficient (D) at room temperature (298 K) is 7.8 × 10⁻⁶ -7 It is cm² / s, and this is the lithium ion concentration of LGPS (1.64X10⁻⁶ 22 When considered together with ions / cm³, the ionic conductivity is 7.2X10 -3 It is calculated in S / cm.

[0198] When the energy difference between the initial state and the transition state calculated by the quantum algorithm is 0.31 eV, ionic conductivity can be predicted under various temperature conditions (e.g., 3.1 x 10⁴ S / cm at 250 K, 7.2 x 10³ S / cm at 298 K, 9.5 x 10² S / cm at 350 K).

[0199] In summary, in the molecular orbital-based integral input step (31), the molecular orbital-based integral values ​​calculated in the clustering step of FIG. 3 can be provided as input to the quantum algorithm.

[0200] In this step, the Hamiltonian in the form of a Pauli operator generated in the last step of FIG. 3 can be used as the input to the quantum algorithm. This Hamiltonian is a second-order quantized form containing 1-electron and 2-electron terms as shown in Equation 6 described above.

[0201] The Hamiltonian expressed in this way is in a form that quantum algorithms can process directly and can serve as a basic input for calculating the system's energy.

[0202] In the quantum algorithm selection and application step (32), calculations can be performed using the selected quantum algorithm. In this step, one or more algorithms such as variational quantum eigenvalue analyzer (VQE), quantum filter diagonalization (QFD), or quantum phase estimation (QPE) can be applied.

[0203] In this step, the energy state of the system can be calculated by constructing and executing a quantum circuit according to the selected algorithm.

[0204] In the energy result output step (33), an energy value calculated through a quantum algorithm can be output. Specifically, the electron energy of the initial state and the transition state can be calculated, and the diffusion barrier of lithium ions can be determined through the difference.

[0205] Lithium ion diffusion barrier height (E a ) can be calculated as the difference between the energy of the transition state and the energy of the initial state, as shown in Equation 7 above. The calculated diffusion barrier can be related to the ionic conductivity through the Eyring-Polany equation, as shown in Equation 8 above.

[0206] Such temperature-dependent ionic conductivity predictions can provide an important indicator for evaluating the performance of solid electrolytes within the battery operating temperature range. In particular, since ionic conductivity at low temperatures is directly related to the winter performance of electric vehicles, accurate predictions enable the prior evaluation of material performance in actual application environments.

[0207] FIG. 5 is a block diagram of an ion conductivity prediction device for a solid electrolyte according to one embodiment of the present invention.

[0208] Referring to FIG. 5, the prediction device of the present invention includes a quantum computing device (40). The quantum computing device (40) may include a processor (41) and a memory (42). This device may be connected to a computing device (50) and a user interface (60).

[0209] The memory (42) is connected to the computing device (50) and can serve to store the calculated value of the actual quantum hardware, and the processor (41) is a physical device that performs calculations by directly utilizing quantum mechanical principles and can use quantum bits (e.g., qubits) as the basic unit of operation.

[0210] The computing device (50) can perform density functional theory calculations on the solid electrolyte structure and then transmit the results to the memory (42) of the quantum computing device (40). The quantum processor (41) receives data from the memory (42), generates a clustering model, and thereby generates input values ​​for the quantum algorithm.

[0211] The processor (41) can execute quantum algorithms such as variational quantum eigenvalue analyzer, quantum filter diagonalization, or quantum phase estimation based on the generated input value.

[0212] The processor (41) can finally calculate the ion conductivity based on the output of the quantum algorithm. The calculated result can be provided to the user through the user interface (60).

[0213] The user interface (60) may be a graphic-based tool that allows the user to perform tasks such as selecting a solid electrolyte model, setting calculation parameters, visualizing results, and analyzing results.

[0214] The user interface (60) can be implemented in various forms, such as a graphic-based desktop interface, a web-based interface, a command-line interface, a mobile application interface, a virtual reality or augmented reality interface, a programming API, or an interactive dashboard.

[0215] Through this interface, users can access and manage solid electrolyte structure libraries, set and adjust computation parameters, monitor computation progress, visualize results such as energy barrier graphs, diffusion path maps, and ionic conductivity, export data and generate reports, compare performance between various quantum algorithms, and analyze the accuracy and uncertainty of computation results.

[0216] Through the prediction device of the present invention, the user can predict the ionic conductivity of a solid electrolyte with high accuracy, thereby reducing time and costs during the battery material development process and efficiently designing a solid electrolyte with improved performance.

[0217] To evaluate the accuracy of the quantum algorithm calculation results, a high-level MP2 calculation is used as a reference value. The computational error represents the absolute difference between the energy value calculated by the quantum algorithm and the reference energy value calculated using the MP2 method. This evaluation method can serve as an objective verification of the quantum algorithm's accuracy.

[0218] Tables 2 and 3 below are experimental examples of the present invention.

[0219] Table 2 and Table 3 may be the results of quantum algorithm calculations for active spaces of various sizes in the initial structure and transition state, respectively. Here, the active space consists of orbitals with occupancy numbers ranging from 0.1 to 1.9, and in the CAS(n,m) notation, n represents the number of electrons in the active space and m represents the number of orbitals.

[0220] The number of quantum circuit layers and parameters represents the complexity of the quantum algorithm, and the computational error is the absolute value (eV) of the energy difference relative to the MP2 reference value.

[0221] Initial structure-active space quantum circuit layer count parameter calculation error (eV)(2,2)129.90 * 10 -11 (4,4)2124.27 * 10 -4 (6,6)4403.65 * 10-3 (8,8)7983.92 * 10 -3

[0222] Transition state, active space, quantum circuit, number of layers, number of parameters, calculation error (eV) (2,2) 129.31 * 10 -9 (4,4)2124.71 * 10 -4 (6,6)5502.99 * 10 -4 (8,8)7981.12 * 10 -2

[0223] Referring to Tables 2 and 3, it can be seen that as the size of the active space increases, the number of layers and parameters of the required quantum circuit also increases. In the case of CAS(2,2), a numerically very low error (10¹¹–10⁻¹⁻¹ eV) was achieved with only one layer and two parameters, but this may paradoxically be a phenomenon that occurs because the active space is too small.

[0224] In other words, because the active space is very limited, the calculation converges accurately, but it may not provide physically meaningful results as it fails to sufficiently capture the complex electronic correlation effects occurring during the lithium ion diffusion process.

[0225] Numerical accuracy (e.g., CAS(2,2)) and physical validity must be considered separately, and a larger active space may provide physically more valid results, even if the numerical error is somewhat larger.

[0226] Since only a limited number of electrons and orbitals are considered in the small active space, it cannot fully represent the interactions between lithium ions and surrounding atoms and is closer to a simplified model.

[0227] Larger active spaces, such as CAS(6,6) or CAS(8,8), contain more electrons and orbitals associated with the lithium ion diffusion process, so they can more completely represent the physical properties of the system.

[0228] All results from CAS(4,4) to CAS(8,8) showed an error within the chemical accuracy (4.3X10² eV = 1 kcal / mol), and in particular, CAS(6,6) achieved a low error of 3.65X10³ eV in the initial structure and 2.99X10⁴ eV in the transition state.

[0229] These results demonstrate that quantum algorithms provide much more accurate energy barrier calculations than traditional DFT methods. As can be seen in Table 1 above, DFT functions showed a large error of 0.09–0.17 eV compared to the MP2 reference value (0.32 eV), whereas quantum algorithms using the CAS(6,6) active space achieved an accuracy improvement of 25 to 500 times compared to DFT methods. This improvement in accuracy allows for more reliable prediction of the ionic conductivity of solid electrolytes, as it has an exponential effect on the diffusion coefficient according to the Eyring-Polany equation.

[0230] Meanwhile, the disclosed embodiments may be implemented in the form of a recording medium that stores instructions executable by a computer. The instructions may be stored in the form of program code and, when executed by a processor, may generate a program module to perform the operation of the disclosed embodiments. The recording medium may be implemented as a computer-readable recording medium.

[0231] Computer-readable recording media include all types of recording media that store instructions that can be decoded by a computer. Examples include ROM (read-only memory), RAM (random access memory), magnetic tape, magnetic disk, flash memory, optical data storage devices, etc.

[0232] Additionally, computer-readable recording media may be provided in the form of non-transitory storage media. Here, 'non-transitory storage media' simply means that it is a tangible device and does not contain a signal (e.g., electromagnetic waves), and this term does not distinguish between cases where data is stored semi-permanently and cases where it is stored temporarily. For example, 'non-transitory storage media' may include a buffer in which data is stored temporarily.

[0233] According to one embodiment, the method according to the various embodiments disclosed herein may be provided as included in a computer program product. The computer program product may be traded between a seller and a buyer as a product. The computer program product may be distributed in the form of a device-readable recording medium (e.g., compact disc read-only memory (CD-ROM)), or distributed online (e.g., download or upload) through an application store (e.g., Play Store™) or directly between two user devices (e.g., smartphones). In the case of online distribution, at least a portion of the computer program product (e.g., downloadable app) may be temporarily stored or temporarily created on a device-readable recording medium, such as the memory of a manufacturer's server, an application store's server, or a relay server.

[0234] As described above, the disclosed embodiments have been explained with reference to the attached drawings. Those skilled in the art will understand that the present invention may be practiced in forms different from the disclosed embodiments without changing the technical spirit or essential features of the invention. The disclosed embodiments are illustrative and should not be interpreted restrictively.

Claims

1. In a method for predicting the ionic conductivity of a solid electrolyte, A step of calculating stable structures of the solid electrolyte using Density Functional Theory (DFT); A step of clustering the above stable structures to generate input values ​​for a quantum algorithm; and A step of calculating the ionic conductivity of the solid electrolyte by inputting the generated input value into a quantum algorithm; A method for predicting the ionic conductivity of a solid electrolyte including 2. In Paragraph 1, The step of calculating the above stable structures is, A method for predicting the ion conductivity of a solid electrolyte, comprising the step of calculating the diffusion path of lithium ions within the solid electrolyte using the Nudged Elastic Band (NEB) method.

3. In Paragraph 1, The above clustering step is, A method for predicting the ion conductivity of a solid electrolyte, comprising the step of obtaining a plurality of cross-sections of a lithium ion diffusion path by performing a cylindrical cut centered on the diffusion channel of the solid electrolyte.

4. In Paragraph 3, The above clustering step is, A method for predicting the ion conductivity of a solid electrolyte, further comprising the step of selecting an active space corresponding to a plurality of molecular orbitals directly involved in the formation and destruction of chemical bonds around the lithium ion diffusion path in each of the plurality of cross-sections.

5. In Paragraph 4, A method for predicting the ion conductivity of a solid electrolyte, further comprising the step of selecting the active space, wherein among the natural orbitals included in each of the plurality of cross-sections obtained by MP2 / def2-SVP calculation, orbitals having an occupancy number of 0.1 to 1.9 are selected as the active space.

6. In Paragraph 1, The above clustering step is, A method for predicting the ionic conductivity of a solid electrolyte, comprising the step of calculating an integral value based on molecular orbitals.

7. In Paragraph 6, The step of calculating the above integral value is, The method further includes the step of calculating molecular orbital-based 1-electron integral values ​​and 2-electron integral values ​​for Hamiltonian configuration, The above 1-electron integral is the kinetic energy of the electron in the active space and the attractive energy between the nucleus and the electron, and The above 2-electron integral value is a method for predicting the ionic conductivity of a solid electrolyte, which is the repulsion energy between multiple electrons in the active space.

8. In Paragraph 1, The above quantum algorithm is, A method for predicting the ionic conductivity of a solid electrolyte comprising at least one of a Variational Quantum Eigensolver (VQE), Quantum Filter Diagonalization (QFD), or Quantum Phase Estimation (QPE).

9. In Paragraph 1, The step of calculating the ionic conductivity of the above-mentioned solid electrolyte is, A method for predicting the ionic conductivity of a solid electrolyte, comprising calculating the ionic conductivity of the solid electrolyte by calculating the energy barrier between the initial state and the transition state of the solid electrolyte.

10. In a storage medium storing at least one instruction, when the at least one instruction is executed by a processor, the processor, A step of calculating stable structures of solid electrolytes using Density Functional Theory (DFT); A step of clustering the above stable structures to generate input values ​​for a quantum algorithm; and A storage medium that enables the step of calculating the ionic conductivity of the solid electrolyte by inputting the above-mentioned generated input value into a quantum algorithm.

11. In Paragraph 10, When the above at least one instruction is executed by the processor, the processor, A storage medium that calculates the diffusion path of lithium ions in the solid electrolyte using the Nudged Elastic Band (NEB) method in the step of calculating the above stable structures.

12. In Paragraph 10, When the above at least one instruction is executed by the processor, the processor, A storage medium that obtains multiple cross-sections of lithium ion diffusion paths by performing cylindrical cutting centered on the diffusion channel of the solid electrolyte in the clustering step.

13. In Paragraph 12, When the above at least one instruction is executed by the processor, the processor, A storage medium that selects active spaces, which are multiple molecular orbitals directly involved in the formation and destruction of chemical bonds around lithium ion diffusion pathways in each of the plurality of cross-sections, in the clustering step described above.

14. In Paragraph 13, When the above at least one instruction is executed by the processor, the processor, A storage medium configured to form an active space by selecting orbitals having an occupancy number between 0.1 and 1.9 from among the natural orbitals included in each of the plurality of cross-sections obtained by MP2 / def2-SVP calculation when selecting the active space.

15. In Paragraph 10, When the above at least one instruction is executed by the processor, the processor, A storage medium for calculating molecular orbital-based integral values ​​in the clustering step described above.

16. In Paragraph 15, When the above at least one instruction is executed by the processor, the processor, When calculating the above integral value, calculate the molecular orbital-based 1-electron integral and 2-electron integral for the Hamiltonian configuration, and The above 1-electron integral is the kinetic energy of the electron in the active space and the attractive energy between the nucleus and the electron, and The above 2-electron integral value is a storage medium that is the repulsion energy between multiple electrons in an active space.

17. In Paragraph 10, When the above at least one instruction is executed by the processor, the processor, A storage medium that enables the above quantum algorithm to perform ion conductivity calculations including at least one of a Variational Quantum Eigensolver (VQE), Quantum Filter Diagonalization (QFD), or Quantum Phase Estimation (QPE).

18. In Paragraph 10, When the above at least one instruction is executed by the processor, the processor, A storage medium that calculates the ionic conductivity of the solid electrolyte by calculating the energy barrier between the initial state and the transition state of the solid electrolyte in the step of calculating the ionic conductivity of the solid electrolyte.

Images