Co-adsorption energy prediction method and co-adsorption energy prediction program

The method and program predict co-adsorption energy by approximating initial adsorption energy using electric dipole moments and polarizability, addressing the inefficiencies of quantum chemical calculations, thereby enhancing catalyst design.

JP2024156380APending Publication Date: 2024-11-06SUZUKI MOTOR CORP
View PDF 0 Cites 0 Cited by

Patent Information

Application Number
JP2023070794
Authority / Receiving Office
JP · JP
Patent Type
Applications
Current Assignee / Owner
Filing Date
2023-04-24
Publication Date
2024-11-06

AI Technical Summary

Technical Problem

Existing methods for predicting co-adsorption energy, such as quantum chemical calculations, are time-consuming and costly due to the vast number of structural models that need to be evaluated, especially when multiple types of gas molecules are adsorbed on a metal surface.

Method used

A method and program that utilize equations to predict co-adsorption energy by determining parameters through minimal chemical quantum calculations, considering the interaction between adsorbed molecules on a metal surface, using electric dipole moments and polarizability to approximate initial adsorption energy.

Benefits of technology

This approach allows for efficient prediction of co-adsorption energy with reduced computational time and cost, improving catalyst design efficiency by accurately estimating interactions between different gas molecules on a metal surface.

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure 2024156380000001_ABST
    Figure 2024156380000001_ABST
Patent Text Reader

Abstract

To provide a co-adsorption energy prediction method and co-adsorption energy prediction program, which enable prediction of co-adsorption energy by setting parameters through minimum required chemical quantum computation.SOLUTION: Co-adsorption energy per each of multiple types of adsorption molecules adsorbed on a metal surface is predicted from co-adsorption energy ΔEads., energy variation ΔEC.T. associated with charge transfer, initial adsorption energy ΔE0, and energy variation ΔEdip..SELECTED DRAWING: Figure 2
Need to check novelty before this filing date? Find Prior Art

Description

[Technical field]

[0001] The present invention relates to a method for predicting coadsorption energy and a program for predicting coadsorption energy. [Background technology]

[0002] A model of the interaction between the electric dipoles of molecules adsorbed on metal surfaces is presented in an approximate analytical form suitable for estimating the coverage dependence of the adsorption energy. [Prior art documents] [Patent documents]

[0003] [Non-Patent Document 1] Takahiro Hirai, Masaki Ohkoshi, Atsuyuki Ishikawa, Hiroshi Nakai, Journal of Computer Chemistry, Japan, 18 (2019), p.70-77 [Non-Patent Document 2] Robert G. Parr and Pratim K. Chattaraji, Journal of the American Chemical Society, (US), 113 (1991), p.1854-1856 [Non-Patent Document 3] RG Parr and RG Pearson, Journal of the American Chemical Society, (US), 105 (1983), p.7512-7516. [Non-Patent Document 4] Anton Kokalj, Physical Review B, 84 (2011), 045418. [Non-Patent Document 5] Computational Chemistry Comparison and Benchmark DataBase, [online], National Institute of Standards and Technology, Internet<URL:https: / / cccbdb.nist.gov / introx.asp> Summary of the Invention [Problem to be solved by the invention]

[0004] Non-Patent Document 1 discloses a technique for approximating the adsorption energy with a linear expression of the coverage rate of adsorbed molecules. This technique targets a structural model in which a large number of one type of gas molecules are adsorbed on a metal surface. Non-Patent Document 1 does not take into account the interaction between adsorbed molecules when multiple types of gas molecules are adsorbed (co-adsorbed) on a metal surface in a mixed state.

[0005] By the way, quantum chemical calculations are one method for calculating the coadsorption energy. The problem with this calculation method is that it requires an enormous amount of calculation time. In addition, when the types and coverage of the coadsorbed molecules vary widely, the number of possible structural models becomes enormous. Therefore, if one tries to obtain the coadsorption energy of interest using quantum chemical calculations, an enormous amount of calculation time is required for all of the enormous number of structural models. In other words, performing chemical quantum calculations for all of these enormous number of structural models is extremely difficult in terms of time and cost.

[0006] Therefore, a method for predicting the coadsorption energy is needed. If the coadsorption energy could be predicted without spending time on quantum chemical calculations, the efficiency of catalyst design, for example, could be improved.

[0007] An object of the present invention is to provide a method for predicting coadsorption energy, which is capable of predicting coadsorption energy by determining parameters through the minimum necessary chemical quantum calculations, and a program for predicting coadsorption energy. [Means for solving the problem]

[0008] In order to solve the above problems, the method for predicting coadsorption energy according to the present invention predicts the coadsorption energy per one of multiple types of adsorbed molecules adsorbed on a metal surface using equations (1) to (13).

[0009]

number

[0010]

number

[0011]

number

[0012]

number

[0013]

number

[0014]

number

[0015]

number

[0016]

number

[0017]

number

[0018]

number

[0019]

number

[0020] In addition, in order to solve the above-mentioned problems, the method for predicting coadsorption energy according to the present invention predicts the coadsorption energy per one of multiple types of adsorbed molecules adsorbed on a metal surface using the above-mentioned formulas (2) to (15).

[0021]

number

[0022]

number

[0023] Further, in order to solve the above-mentioned problems, the present invention provides a method for predicting coadsorption energy, comprising the steps of: selecting a metal having a surface and a plurality of types of adsorbed molecules identified by a natural number subscript i; and calculating a coverage θ of each of the adsorbed molecules adsorbed on the surface. i Assuming that the polarizability α of each of the adsorbed molecules is i is extracted from the given data, and the electric dipole moment P 0 inside i is extracted from the given data, and the distance d between the center of each of the adsorbed molecules and the interface is calculated. i Estimate the initial adsorption energy ΔE of each of the adsorbed molecules. 0 i Assuming that multiple electric dipole moments P are excited at the interface between the metal and each of the adsorbed molecules, 0 interface i Assuming that the coadsorption energy ΔE ads. The first term calculated by the formula (2) and the electric dipole moment P calculated by the formula (3) are calculated. 0inside i Weighted average of P 0 inside and the electric dipole moment P calculated by the formula (4) 0 interface i Weighted average of P 0 interface The coadsorption energy ΔE calculated using the prediction formula, which is the sum of the second term including ads. The coadsorption energy ΔE calculated by quantum chemical calculation ads. The initial adsorption energy ΔE 0 i and the electric dipole moment P 0 interface i Repeat the correction of the coadsorption energy ΔE calculated by the above prediction formula. ads. The coadsorption energy ΔE calculated by quantum chemical calculation ads. When the initial adsorption energy ΔE 0 i and the electric dipole moment P 0 interface Using the coadsorption energy ΔE ads. The above prediction formula is determined.

[0024] In addition, in order to solve the above-mentioned problems, the coadsorption energy prediction program of the present invention enables a computer to realize the function of predicting the coadsorption energy of multiple types of adsorbed molecules adsorbed on a metal surface using formulas (1) to (13).

[0025] Further, in order to solve the above-mentioned problems, the present invention provides a program for predicting coadsorption energy, which includes a function of selecting a metal having a surface and a plurality of types of adsorbed molecules identified by a natural number subscript i, and a function of calculating a coverage θ of each of the adsorbed molecules adsorbed on the surface. i The function of assuming polarizability α of each of the adsorbed molecules i from given data, and the electric dipole moment P 0 inside i from the given data, and the distance d between the center of each of the adsorbed molecules and the interface iand the initial adsorption energy ΔE of each of the adsorbed molecules. 0 i and a function of assuming a plurality of electric dipole moments P excited at the interface between the metal and each of the adsorbed molecules. 0 interface i The coadsorption energy ΔE is calculated by quantum chemical calculation. ads. The first term obtained by the formula (2) and the electric dipole moment P obtained by the formula (3) 0 inside i Weighted average of P 0 inside and the electric dipole moment P calculated by the formula (4) 0 interface i Weighted average of P 0 interface The coadsorption energy ΔE calculated using the prediction formula, which is the sum of the second term including ads. The coadsorption energy ΔE calculated by quantum chemical calculation ads. The initial adsorption energy ΔE 0 i and the electric dipole moment P 0 interface i The function of repeatedly correcting the coadsorption energy ΔE calculated by the above prediction formula ads. The coadsorption energy ΔE calculated by quantum chemical calculation ads. When the initial adsorption energy ΔE 0 i and the electric dipole moment P 0 interface Using the coadsorption energy ΔE ads. and a function of determining the prediction formula. Effect of the Invention

[0026] According to the present invention, it is possible to provide a method for predicting coadsorption energy and a program for predicting coadsorption energy, which are capable of predicting coadsorption energy by determining parameters through a minimum necessary amount of chemical quantum calculation. [Brief description of the drawings]

[0027] [Figure 1]1 is a block diagram of a vehicle to which the co-adsorption energy prediction method and the co-adsorption energy prediction program according to the present embodiment are applied; [Diagram 2] FIG. 4 is a conceptual diagram of a prediction formula for predicting adsorption energy according to an embodiment of the present invention. [Diagram 3] 1 is a flowchart showing a procedure for predicting coadsorption energy. [Figure 4] FIG. 2 is a diagram showing an example of a structural model for determining coadsorption energy. [Diagram 5] FIG. 2 is a diagram showing the relationship between the internal energy and the number of electrons of a carbon monoxide molecule. [Figure 6] 1 is a graph comparing the coadsorption energy calculated by the prediction formula and the coadsorption energy calculated by quantum chemical calculation. [Figure 7] 1 is a graph comparing the coadsorption energy estimated in a comparative example with the coadsorption energy calculated by quantum chemical calculation. [Figure 8] This graph shows the change in adsorption energy versus coverage in an adsorption structure where the sum of the coverage of carbon monoxide molecules and the coverage of oxygen atoms is 100 percent. DETAILED DESCRIPTION OF THE PREFERRED EMBODIMENTS

[0028] An embodiment of a method for predicting coadsorption energy and a program for predicting coadsorption energy according to the present invention will be described with reference to Fig. 1 to Fig. 8. Note that the same or corresponding components are denoted by the same reference numerals in the drawings.

[0029] FIG. 1 is a block diagram of a vehicle to which the co-adsorption energy prediction method and the co-adsorption energy prediction program according to the present embodiment are applied.

[0030] As shown in FIG. 1, a vehicle 1 includes an internal combustion engine 2, an exhaust pipe 3 for guiding exhaust gas discharged from the internal combustion engine 2, and a catalyst 5 provided in the exhaust pipe 3.

[0031] The catalyst 5 is, for example, a three-way catalyst (TWC). The catalyst 5 purifies harmful components contained in the exhaust gas of the internal combustion engine 2, such as hydrocarbons (HC), carbon monoxide (CO), and nitrogen oxides (NOx), through oxidation and reduction reactions, and converts them into water (H 2 O), carbon dioxide (CO 2 ), and nitrogen (N 2 The catalyst 5 carries a precious metal such as platinum (Pt), palladium (Pd) or rhodium (Rh). In addition to harmful substances, the exhaust gas also contains oxygen (O 2 )

[0032] The coadsorption energy prediction method and coadsorption energy prediction program of this embodiment are intended to predict coadsorption energy when the type of coadsorbed molecule and the coverage rate of the metal surface by the coadsorbed molecule change in a variety of ways, such as in the catalyst 5 of a vehicle 1.

[0033] The coadsorption energy prediction method according to this embodiment will hereinafter be simply referred to as "this prediction method", and the coadsorption energy prediction program according to this embodiment will hereinafter be simply referred to as "this prediction program".

[0034] Here, the derivation of the prediction equation for predicting the coadsorption energy will be explained. The meanings of the symbols used in the derivation process are shown in Tables 1 to 3. In Tables 1 to 3, the unit [-] is a dimensionless quantity. In the symbol column of Table 1 to the symbol column of Table 3, physical quantities with dimensions are written in italics, and vector quantities with dimensions are written in bold italics.

[0035] [Table 1]

[0036] [Table 2]

[0037] [Table 3]

[0038] FIG. 2 is a conceptual diagram of a prediction formula for predicting adsorption energy according to an embodiment of the present invention.

[0039] In FIG. 2, various characteristics of one gas molecule 11 are identified by adding the suffix "A", and various characteristics of the other gas molecule 12 are identified by adding the suffix "B". When generally discussed in the specification, the characteristics of the gas molecules 11 and 12 are identified by adding the suffix "i". In addition, the gas molecules 11 and 12 adsorbed on the surface 16 of the metal 15 may be called adsorbed molecules 11 and 12. For example, the gas molecules 11 and 12 are different types of substances, such as one gas molecule 11 being carbon monoxide (CO) and the other gas molecule 12 being nitrogen oxides (NOx). Each gas molecule 11 and 12 contains at least one gas molecule, and may contain multiple gas molecules of the same type.

[0040] FIG. 2(a) shows an initial state in which the gas molecules 11 and 12 and the surface 16 of the metal 15 are sufficiently separated from each other and do not interact with each other. Each of the gas molecules 11 and 12 has an electric dipole moment P 0 inside i and the chemical potential μ i The surface 16 of the metal 15 has a chemical potential μ Metal It shall have the following.

[0041] The initial state of FIG. 2(a) is changed to the state shown in FIG. 2(b). FIG. 2(b) shows the state where the gas molecules 11 and 12 and the surface 16 of the metal 15 reach thermal equilibrium. The chemical potential μ whole The energy change due to the charge transfer is assumed to be a uniform value. C.T. Let us assume that.

[0042] The state shown in Fig. 2(b) transitions to the state shown in Fig. 2(c). Fig. 2(c) shows a state in which the gas molecules 11 and 12 are adsorbed on the surface 16 of the metal 15, ignoring the interaction between the gas molecules 11 and 12. An electric dipole moment P 0 interface i The electric dipole moment P 0 inside i is excited. The energy change in this process is the initial adsorption energy ΔE 0 Far away.

[0043] Finally, as shown in FIG. 2(d), the interaction between the gas molecules 11 and 12, which was ignored in FIG. 2(a), is corrected. The electric field vector E inside i The electric field vector E inside i By the electric dipole moment P 0 inside i is the electric dipole moment P inside i In addition, the electric field vector E interface i The electric field vector E interface i By the electric dipole moment P 0 interface i is the electric dipole moment P interface i These changes are called dielectric polarization. The energy change in this process is the energy change ΔE per one of the adsorbed molecules 11 and 12 due to the interaction between the electric dipoles. dip. It is called.

[0044] From Fig. 2(a) to Fig. 2(d), the coadsorption energy ΔE ads. is estimated by equation (16).

[0045]

number

[0046] Coadsorption energy ΔE ads. , the energy change due to charge transfer ΔE C.T. , initial adsorption energy ΔE 0 , and the energy change ΔE dip. is defined per one of the adsorbed molecules 11 and 12.

[0047] Since the interaction between gas molecules 11 and 12 is ignored, the initial adsorption energy ΔE 0 can be expressed as a constant that depends only on the type of adsorbed molecules 11 and 12. Therefore, the weighted average of the initial adsorption energy ΔE 0 are the initial adsorption energies ΔE of the adsorbed molecules 11 and 12. 0 i , and the coverage θ of each of the adsorbed molecules 11 and 12 i It can be calculated using equation (17).

[0048]

number

[0049] Energy change of the entire system due to charge transfer ΔE C.T. is estimated by the HSAB (Hard and Soft Acids and Bases) rule, and the energy change ΔE dip. is estimated using Maxwell's theory of electromagnetics.

[0050] First, according to the HSAB law described in Non-Patent Documents 2 and 3, the chemical properties of the gas molecules 11 and 12 can be expressed by a chemical potential μ and a chemical softness S (Softness). i is the internal energy E of gas molecules 11 and 12 i , the number of electrons in gas molecules 11 and 12, N i The chemical softness S of each gas molecule 11 and 12 is defined as follows: i is defined as in equation (19).

[0051]

number

[0052]

number

[0053] The chemical potential μ of each gas molecule 11 and 12 due to the charge transfer i Change in Δμ i can be calculated using equation (20).

[0054]

number

[0055] Number of electrons in gas molecules 11 and 12, N i is based on the number of electrons before the charge transfer. The chemical potential μ of the entire system after the charge transfer whole becomes a constant value. In other words, the chemical potential μ i (ΔN i ) is the chemical potential μ whole Since this coincides with the above, the relationship in equation (21) holds.

[0056]

number

[0057] Furthermore, the change in internal energy ΔE i can be calculated using equation (22).

[0058]

number

[0059] As shown in [Equation 21], the amount of electron movement ΔN i If is zero, (ΔE i ÷ΔN i) is the chemical potential μ of the charge-neutral gas molecules 11 and 12 i (0) C 1 The chemical potential μ of the charge-neutral gas molecules 11 and 12 i (0). In addition, the relation ΔN obtained by approximating equation (19) in terms of difference under the assumption of transformation of equation (22) is i =S i ×Δμ i was used.

[0060]

number

[0061] From the above, the energy change ΔE C.T. can be calculated using equation (23).

[0062]

number

[0063] Chemical potential μ of the entire system after charge transfer whole can be calculated as follows:

[0064] First, the total charge on the surface 16 of the metal 15 is the amount of electron movement per metal atom, ΔN metal and the number of metal atoms exposed on the surface 16, n metal The sum of the charges of the adsorbed molecules 11 and 12 is the amount of electron transfer ΔN i and the coverage of the adsorbed molecules 11 and 12, θ i and the number of metal atoms exposed on the surface 16, n metal Since the number of electrons in the entire system is conserved, we obtain equation (24).

[0065]

number

[0066] In addition, the chemical potential μ of the surface 16 of the metal 15 in the charge neutral state metal (0), Chemical softness S of metal 15 surface 16 metal Equation (24) was rearranged using [equation 24].

[0067]

number

[0068] While Non-Patent Document 2 and Non-Patent Document 3 consider a chemical reaction between two substances, this prediction method predicts the energy change accompanying a chemical reaction between two gas molecules 11 and 12 and the surface 16 of the metal 15, that is, three substances, as shown in Equation (22) and Equation (23). Furthermore, in this prediction method, the reactants are not limited to the two gas molecules 11 and 12 and the surface 16 of the metal 15, and the energy of a chemical reaction due to charge transfer between three or more substances can be predicted.

[0069] Next, the energy change ΔE per one of the adsorbed molecules 11 and 12 due to the interaction between the electric dipoles dip. This article explains:

[0070] According to Maxwell's electromagnetism, the electric field vector E that the electric dipole moment P creates on the position vector r is expressed by equation (25).

[0071]

number

[0072] For simplicity, if we approximate the adsorbed molecules 11 and 12 as one type, we obtain equation (26). From the position vector R of other adsorbed molecules 11 and 12 measured with respect to any adsorbed molecule 11 or 12 as the reference, Rd is defined by equation (27), and R2d is defined by equation (28). Rnd is the position vector of other electric dipole moments measured with respect to any electric dipole moment.

[0073]

number

[0074]

number

[0075] The first term in equation (26) is the internal electric dipole moment P inside is the electric field vector excited, and the second term is the electric dipole moment P excited at the interface between the surrounding adsorbed molecules 11 and 12 and the metal 15. interface is the exciting electric field vector, and the third term is the electric dipole moment P inside is the exciting electric field vector.

[0076] Furthermore, the adsorbed molecules 11 and 12 are aligned vertically, and the electric dipole moment P inside , and the electric dipole moment P excited at the interface between the metal 15 and the adsorbed molecules 11 and 12. interface The electric field vector E generated at the center of the adsorbed molecules 11 and 12 is assumed to be perpendicular to the surface. inside is oriented vertically, and we obtain equation (29).

[0077]

number

[0078] The calculation of equation (29) can be performed by defining equation (30) with reference to Non-Patent Document 4.

[0079]

number

[0080] Using equation (30), equation (29) is transformed into equation (31).

[0081]

number

[0082] On the surface of a hexagonal crystal, it is approximated by equation (32).

[0083]

number

[0084] Here, equation (33) is defined.

[0085]

number

[0086] In a similar manner, the electric field vector E interface can be calculated using equation (34).

[0087]

number

[0088] Here, we consider the dielectric polarization. The electric dipole moment P 0 inside This electric field vector E inside The internal electric dipole moment P of the adsorbed molecule is given by inside In this case, we obtain equation (36).

[0089]

number

[0090] Similarly, equation (36) also holds.

[0091]

number

[0092] By solving the above equations, equations (37) and (38) are obtained.

[0093]

number

[0094] Here, A, B, and C in equations (37) and (38) are defined by equations (39) to (41).

[0095]

number

[0096] When an electric dipole moment P exists in an electric field vector E, the potential energy stabilizes to [Equation 38]. However, when an electric dipole moment P is excited by an electric field vector E, the stabilizing energy decreases by [Equation 39]. The initial electric dipole moment P 0 and the electric dipole moment P have the relationship shown in [Equation 40]. Therefore, the stabilizing energy is shown in [Equation 41] (Non-Patent Document 4). From the above, the energy change ΔE dip. In addition, due to the limitations of the application format, formula (42) is divided into three fractional expressions for addition and subtraction, and enclosed in brackets {}.

[0097]

number

[0098]

number

[0099]

number

[0100]

number

[0101]

number

[0102] The first term on the right hand side of equation (42) before transformation represents the stabilizing energy due to the electric dipoles inside the adsorbed molecules 11 and 12. The energy that stabilizes the mirror-image electric dipoles is also the same value, so the first term is doubled. The second term on the right hand side of equation (42) before transformation represents the energy change due to the dipoles excited at the interface. To specifically calculate equation (42), the distance R between the nearest adsorbed molecules 11 and 12 is nn , the weighted average of the distance between the center of the adsorbed molecules 11 and 12 and the interface d, the weighted average of the polarizabilities of the adsorbed molecules 11 and 12 α, and the weighted average of the internal electric dipole moments of the adsorbed molecules 11 and 12 assuming no dielectric polarization P 0 inside , the weighted average P of the electric dipole moment excited at the interface between the metal 15 and the adsorbed molecules 11 and 12, assuming no dielectric polarization. 0 interface is determined. On the (111) surface of a hexagonal crystal, the area occupied by one metal atom on the surface 16 is [Equation 43], so the area occupied by one adsorbed molecule is [Equation 44]. The coverage θ is calculated using equation (43). Note that the right-hand side of equation (42) after transformation can be summarized as one fractional equation, but due to the constraints of the width of the application format, it is divided into three fractional equations for addition and subtraction and enclosed in {}.

[0103]

number

[0104]

number

[0105]

number

[0106] Assuming that the adsorbed molecules 11 and 12 are also arranged in a hexagonal crystal, the distance R between the nearest adsorbed molecules 11 and 12 is nn is estimated by equation (44).

[0107]

number

[0108] The distance d between the center of the multiple types of adsorbed molecules 11 and 12 and the interface, the polarizability α of the multiple types of adsorbed molecules 11 and 12, and the electric dipole moment P inside the adsorbed molecules 11 and 12 assuming no dielectric polarization. 0 inside , electric dipole moment P excited at the interface between the metal 15 and the adsorbed molecules 11 and 12 0 interface is approximated by the average value of the values ​​of the adsorbed molecules 11 and 12. That is, the electric dipole moment P 0 inside i , the electric dipole moments P excited at the surface 16 and interface of the adsorbed molecules 11 and 12 and the metal 15, respectively. 0 interface i , the distance d between the center of each adsorbed molecule 11, 12 and the interface i , the polarizability α of each of the adsorbed molecules 11 and 12 i The average value is defined as shown in equations (45) to (48). The weighting coefficient is the coverage θ of the corresponding adsorbed molecules 11 and 12. i It is.

[0109]

number

[0110] Therefore, the electric dipole moment P 0 inside i , the electric dipole moment P excited at the interface between each of the adsorbed molecules 11 and 12 and the surface 16 of the metal 15. 0 interface i, the distance d between the center of each adsorbed molecule 11, 12 and the interface i , the polarizability α of each of the adsorbed molecules 11 and 12 i , and the coverage θ of each of the adsorbed molecules 11 and 12 i Equation (42) can be calculated by specifying

[0111] Polarizability α of each adsorbed molecule 11 and 12 i , and the electric dipole moment P of each of the adsorbed molecules 11 and 12. 0 inside i There are literature values ​​for the coverage θ of each of the adsorbed molecules 11 and 12. i is determined from the adsorption structure under consideration. The distance d between the center of each adsorbed molecule 11, 12 and the interface is i is estimated, for example, by the molecular radius of the adsorbed molecules 11 and 12.

[0112] The electric dipole moment P excited at the interface between each of the adsorbed molecules 11 and 12 and the surface 16 of the metal 15 is 0 interface i is treated as a free parameter.

[0113] The present prediction method and program differ from Non-Patent Document 4, which discloses a method for calculating the electric dipole moment between molecules adsorbed on the surface of a material, in two respects: one molecule is expressed by three electric dipole moments, i.e., the interaction of n coadsorbed molecules is approximated by 3n electric dipole moments, and a formula describing the interaction has been analytically derived.

[0114] From the above, formula (16) can be rewritten as prediction formula (49). Note that due to the limitations of the application format, the third term of prediction formula (49) is divided into three fractional expressions, an addition and a subtraction, and enclosed in brackets {}.

[0115]

number

[0116] Using the prediction equation (49), the coadsorption energy ΔE ads. can be predicted.

[0117] Figure 3 shows the coadsorption energy ΔE ads. 1 is a flowchart showing a procedure for predicting

[0118] As shown in FIG. 3, this prediction method includes seven major steps.

[0119] The first step S1 includes a procedure for selecting a plurality of types of adsorbed molecules 11, 12, each of which is identified by an integer subscript i greater than 0, and a metal 15 having a surface 16, and a procedure for determining the coverage θ of each of the adsorbed molecules 11, 12 adsorbed on the surface 16. i In other words, the first step S1 is a procedure for selecting candidates for the assumed structural model.

[0120] The second step S2 is to calculate the chemical potential μ i and the chemical softness S of each adsorbed molecule. i The procedure for calculating the polarizabilities α of the adsorbed molecules 11 and 12 using quantum chemical calculations. i The procedure to extract from the given data, the electric dipole moment P 0 inside i from the given data, and the distance d between the center of each of the adsorbed molecules 11 and 12 and the interface i This is a procedure for estimating the following:

[0121] In the third step S3, a plurality of electric dipole moments P are excited at the interfaces between the surface 16 of the metal 15 and each of the adsorbed molecules 11 and 12. 0 interface i The procedure assuming that the initial adsorption energies ΔE of the adsorbed molecules 11 and 12 are 0 i This is a procedure that assumes the following.

[0122] The fourth step S4 is to calculate the coadsorption energy ΔE ads.This is the procedure for determining

[0123] The fifth step S5 is to calculate the coadsorption energy ΔE ads. is the coadsorption energy ΔE calculated by the quantum chemical calculation in the fourth step S4. ads. This is a procedure for determining whether the value is close to or not.

[0124] Then, the coadsorption energy ΔE calculated using the prediction equation (49) ads. The coadsorption energy ΔE calculated by quantum chemical calculation ads. In the sixth step S6, multiple electric dipole moments P 0 interface i and the initial adsorption energy ΔE 0 i Then, steps S4 and S5 are repeated.

[0125] Coadsorption energy ΔE calculated using the prediction formula (49) ads. is the coadsorption energy ΔE calculated by the quantum chemical calculation in the fourth step S4. ads. If it is approximated to, in the seventh step S7, the electric dipole moment P 0 interface , and the initial adsorption energy ΔE of each gas molecule 11 and 12 0 i Using the coadsorption energy ΔE ads. The prediction formula (49) is determined.

[0126] Next, a specific procedure will be described.

[0127] First, the coadsorption energy ΔE ads. The execution of the fourth step S4 is premised on the execution of the first step S1.

[0128] FIG. 4 is a diagram showing an example of a structural model for determining the coadsorption energy.

[0129] As shown in FIG. 4, a structure was assumed in which carbon monoxide molecules (CO) and oxygen atoms (O) as adsorbed molecules 11 and 12 were adsorbed onto a surface 16 of rhodium (Rh) as a metal 15 (first step S1). The rhodium (Rh) surface 16 is a (111)-(2×2) surface with periodic boundary conditions imposed. Three types of typical adsorption sites on the surface 16 were assumed: FCC sites, HCP sites, and Top sites. The coverage θ of the respective adsorbed molecules 11 and 12 i was varied between 25 percent and 100 percent. The ratio of the adsorbed molecules 11 and 12, that is, the ratio of carbon monoxide molecules (CO) and oxygen atoms (O), was varied between 0 percent and 100 percent.

[0130] For the assumed adsorption structure, the coadsorption energy ΔE per adsorbed molecule is ads. was calculated from equation (50) by quantum chemical calculation.

[0131]

number

[0132] The meanings of the symbols used in Eq. (50) are shown in Table 4. In Table 4, the unit [-] is a dimensionless quantity. The coadsorption energy ΔE per adsorbed molecule calculated from Eq. (50) by quantum chemical calculation is ads. is shown in Table 5.

[0133] [Table 4]

[0134] [Table 5]

[0135] Next, to perform the fifth step S5, the coadsorption energy ΔE per adsorbed molecule is calculated using the prediction equation (49). ads. Request.

[0136] Polarizability α of each adsorbed molecule 11 and 12 i , and the electric dipole moment P of each of the adsorbed molecules 11 and 12 0 inside i The value described in Non-Patent Document 5 was adopted for (second step S2). The weighted average α of the polarizabilities of the multiple types of adsorbed molecules 11 and 12 is given by equation (46). The electric dipole moment P 0 inside is given by equation (47).

[0137] Next, the distance d between the center of each of the adsorbed molecules 11 and 12 and the interface i The distance d between the center of the carbon monoxide molecule (CO) and the interface was estimated (second step S2). i was estimated as the sum of the atomic radius of the rhodium atom (Rh), the atomic radius of the carbon atom (C), and half the bond distance of the carbon monoxide molecule (CO). The distance d between the center of the oxygen atom (O) and the interface i was estimated as the sum of the atomic radius of rhodium (Rh) and the atomic radius of oxygen atom (O). The distance d i The weighted average d is given by equation (45).

[0138] The chemical potential μ of each gas molecule 11 and 12 i was calculated by quantum chemical calculation using Equation (18) (second step S2). In addition, the chemical softness S i was calculated by quantum chemical calculation using formula (19) (second step S2).

[0139] Figure 5 shows the internal energy E of carbon monoxide (CO). i and the number of electrons Ni.

[0140] As shown in Figure 5, the internal energy E of the carbon monoxide molecule (CO) is calculated by changing the number of electrons Ni. i was calculated by quantum chemical calculations.

[0141] As shown in FIG. 5, there is a relationship between the internal energy E and the number of electrons N of a carbon monoxide molecule (CO) that is approximated by the curve of [Equation 50].

[0142]

number

[0143] From equation (18), the chemical potential μ of a carbon monoxide molecule (CO) is calculated as follows:

[0144]

number

[0145] Moreover, from equation (19), the chemical softness S of a carbon monoxide molecule (CO) can be calculated as shown in [Equation 52].

[0146]

number

[0147] The initial adsorption energy ΔE of each gas molecule 11 and 12 0 i , and multiple electric dipole moments P excited at the interfaces between the surface 16 of the metal 15 and the respective adsorbed molecules 11 and 12. 0 interface i is a free parameter. Therefore, the two free parameters (ΔE 0 i , P 0 interface i The value of ΔE ) was determined to be the value that reproduces the quantum chemical calculation (third step S3). The weighted average ΔE 0 is given by equation (17). The electric dipole moment P excited at the interface between the surface 16 of the metal 15 and each of the adsorbed molecules 11 and 12 is 0 interface i Weighted average of P 0 interface is given by equation (48).

[0148] Tables 6 and 7 show the parameters used in the analysis.

[0149] [Table 6]

[0150] [Table 7]

[0151] The coadsorption energy per adsorbate, ΔE, was calculated using the parameters shown in Tables 6 and 7 and the prediction equation (49). ads. asked for.

[0152] Figure 6 shows the coadsorption energy ΔE calculated by the prediction equation (49). ads. and the coadsorption energy ΔE in Table 5 calculated by quantum chemical calculation ads. 4 is a graph comparing the above (fifth step S5).

[0153] As shown in Figure 6, the coadsorption energy ΔE ads. and the coadsorption energy ΔE in Table 5 calculated by quantum chemical calculation ads. and compared.

[0154] Using the prediction method of Eq. (49), the coadsorption energy ΔE ads. It has been shown that it is possible to predict

[0155] Next, a comparative example will be described.

[0156] In Non-Patent Document 1, the coadsorption energy ΔE is expressed as a linear equation of the coverage θ. ads. By applying this concept, the adsorption energy ΔE ads.i can be calculated using equation (51). The coadsorption energy ΔE ads. can be estimated by equation (52).

[0157]

number

[0158] Coefficient a in Equation (51) i , coefficient b i is a free parameter. The coadsorption energies ΔE in Table 5 calculated by quantum chemical calculations ads. The coefficients a i , coefficient b i was fitted to the values ​​in Table 8.

[0159] [Table 8]

[0160] FIG. 7 shows the coadsorption energy ΔE estimated by Eq. (52) as a comparative example. ads. and the coadsorption energy ΔE in Table 5 calculated by quantum chemical calculation ads. 1 is a graph comparing

[0161] As shown in Figure 7, the coadsorption energy ΔE ads. By estimating the coadsorption energy ΔE ads. In other words, this prediction method can predict the coadsorption energy ΔE with higher accuracy than the method proposed in Non-Patent Document 1. ads. It has been shown that it is possible to predict

[0162] A remarkable example of the effectiveness of this prediction method appears in an adsorption structure in which a carbon monoxide molecule (CO) and an oxygen atom (O) are adsorbed on the surface 16 of a metal 15 .

[0163] Figure 8 shows the coverage rate θ of carbon monoxide molecules (CO). CO and the coverage of oxygen atoms (O) θ O The coverage θ in an adsorption structure where the sum of CO FIG. 13 shows the change in adsorption energy with respect to

[0164] As shown in Figure 8, according to quantum chemical calculations (indicated by "◇" in Figure 8), the coverage rate θ COis 50 percent, and the coverage of oxygen atoms (O) is θ O When the coadsorption energy ΔE is 50 percent ads. is the most stable.

[0165] The comparative example ("△" in FIG. 8) has a carbon monoxide (CO) coverage θ CO Coadsorption energy ΔE ads. It is predicted that changes linearly, and the prediction accuracy is poor.

[0166] This prediction method ("□" in Figure 8) is based on the coverage rate θ CO Coadsorption energy ΔE ads. This shows that the present prediction method is superior to the comparative example.

[0167] In addition, this prediction method uses two free parameters (ΔE 0 i , P 0 interface i ) must be fitted. In the comparative example, two free parameters (a i , b i ) needs to be fitted. In other words, the number of explanatory variables is the same in this prediction method and the comparative example. Despite the same number of explanatory variables, this prediction method can predict the quantum chemical calculations with higher accuracy, and the superiority of the proposed prediction formula (49) is clear.

[0168] The inventors also calculated the coadsorption energy ΔE ads. It was confirmed that it is possible to predict the energy change due to the charge transfer. Due to the limitations of the application format, the second term of the second prediction formula (53) was divided into three fractional expressions, the addition and subtraction of which was enclosed in brackets {}. The second term of the prediction formula (49), that is, the energy change due to the charge transfer, is ΔE C.T. The coadsorption energy ΔE calculated by the second prediction equation (53) ignoring ads. and the coadsorption energy ΔE in Table 5 calculated by quantum chemical calculation ads. The absolute mean error was superior to that of the comparative example.

[0169]

number

[0170] Furthermore, the dielectric polarization may be ignored and the weighted average of the polarizabilities of the adsorbed molecules may be set to α = 0. In this case, the coadsorption energy ΔE ads. In addition, due to the limitations of the application format, the second term of the third prediction formula (54) is divided into three subtractions and enclosed in brackets {}.

[0171]

number

[0172] In addition, when approximating as in [Equation 56], A, B, and C in the third prediction equation (54) are arranged as in [Equation 57]. In this case, the coadsorption energy ΔE ads. can be predicted.

[0173]

number

[0174]

number

[0175]

number

[0176] This prediction method uses the coadsorption energy ΔE ads.The adsorption energy is predicted with high accuracy. Therefore, this prediction method can be effectively used in the design of the catalyst 5 material, and enables the highly accurate design of the catalyst 5 material. The adsorption energy is also used as an explanatory variable in the search for sensor materials and water-repellent coatings. Therefore, this prediction method can be applied not only to the design of the catalyst 5 material, but also to the design of various functional materials.

[0177] The prediction program also causes a computer to realize a function for executing the prediction method. That is, the prediction program calculates the coadsorption energy ΔE using the prediction formulas (prediction formula (49), second prediction formula (53), third prediction formula (54), and fourth prediction formula (55)) determined by the prediction method. ads. The present prediction program also causes the computer to realize a function of executing the first step S1 to the seventh step S7 of FIG. 3. In other words, the present prediction program causes the computer to realize a function of determining prediction formulas (prediction formula (49), second prediction formula (53), third prediction formula (54), and fourth prediction formula (55)) through the execution of the first step S1 to the seventh step S7. Such a present prediction program causes the computer to realize a function of determining the coadsorption energy ΔE ads. It is provided as Computer Aided Engineering (CAE) software that predicts the following:

[0178] Therefore, according to the present prediction method and program, it is possible to determine parameters by the minimum necessary chemical quantum calculations and predict the coadsorption energy. [Explanation of symbols]

[0179] 1...vehicle, 2...internal combustion engine, 3...exhaust pipe, 5...catalyst, 11, 12...gas molecules, 15...metal, 16...metal surface.

Claims

1. A method for predicting coadsorption energy, which predicts the coadsorption energy per one of multiple types of adsorbed molecules adsorbed on a metal surface using equations (1) to (13). [0010] [0025] [0030] [0045] [0050] [006] [0070] [0080] [0090] [0089] ##EQU00011##

2. A method for predicting coadsorption energy, which predicts the coadsorption energy per one of multiple types of adsorbed molecules adsorbed on a metal surface using the above formulas (2) to (15). ##EQU00012## ##EQU00013##

3. A metal having a surface and a plurality of types of adsorbed molecules identified by a natural number subscript i are selected; The coverage θ of each of the adsorbed molecules adsorbed on the surface i Assume that The polarizability α of each of the adsorbed molecules i is extracted from the given data, The electric dipole moment P when it is assumed that there is no dielectric polarization inside each of the adsorbed molecules. 0 inside i is extracted from the given data, The distance d between the center of each adsorbed molecule and the interface i Estimate, The initial adsorption energy ΔE of each of the adsorbed molecules 0 i Assume that A plurality of electric dipole moments P excited at the interface between the metal and each of the adsorbed molecules 0 interface i Assume that The coadsorption energy ΔE was calculated by quantum chemical calculation. ads. Seeking The first term calculated by the formula (2) and the electric dipole moment P calculated by the formula (3) 0 inside i The weighted average of P 0 inside and the electric dipole moment P calculated by the formula (4) 0 interface i The weighted average of P 0 interface The coadsorption energy ΔE calculated using the prediction formula, which is the sum of the second term including ads. The coadsorption energy ΔE calculated by quantum chemical calculation ads. The initial adsorption energy ΔE 0 i and the electric dipole moment P 0 interface i Repeat the correction of Coadsorption energy ΔE calculated using the above prediction formula ads. The coadsorption energy ΔE calculated by quantum chemical calculation ads. When the initial adsorption energy ΔE 0 i and the electric dipole moment P 0 interface Using the coadsorption energy ΔE ads. A method for predicting coadsorption energy that determines the above prediction formula.

4. The method for predicting coadsorption energy according to claim 3 , wherein the second term is determined from the formulas (3) to (13) and (16). ##EQU00014##

5. 5. The method for predicting coadsorption energy according to claim 4, wherein the prediction formula is a sum of the first term, the second term, and the third term determined by the formula (15) and the formula (17). ##EQU00015##

6. On the computer, A function of predicting the coadsorption energy of multiple types of adsorbed molecules adsorbed on a metal surface using the formulas (1) to (13). A program for predicting the coadsorption energy to achieve this.

7. On the computer, A function of selecting a metal having a surface and a plurality of types of adsorbed molecules identified by a natural number subscript i; The coverage θ of each of the adsorbed molecules adsorbed on the surface i A function that assumes The polarizability α of each of the adsorbed molecules i from given data; The electric dipole moment P when it is assumed that there is no dielectric polarization inside each of the adsorbed molecules. 0 inside i from given data; The distance d between the center of each adsorbed molecule and the interface i The ability to estimate The initial adsorption energy ΔE of each of the adsorbed molecules 0 i A function that assumes A plurality of electric dipole moments P excited at the interface between the metal and each of the adsorbed molecules 0 interface i A function that assumes The coadsorption energy ΔE was calculated by quantum chemical calculation. ads. The function to request The first term calculated by the formula (2) and the electric dipole moment P calculated by the formula (3) 0 inside i The weighted average of P 0 inside and the electric dipole moment P calculated by the formula (4) 0 interface i The weighted average of P 0 interface The coadsorption energy ΔE calculated using the prediction formula, which is the sum of the second term including ads. The coadsorption energy ΔE calculated by quantum chemical calculation ads. The initial adsorption energy ΔE 0 i and the electric dipole moment P 0 interface i The ability to repeat corrections, Coadsorption energy ΔE calculated using the above prediction formula ads. The coadsorption energy ΔE calculated by quantum chemical calculation ads. When the initial adsorption energy ΔE 0 i and the electric dipole moment P 0 interface Using the coadsorption energy ΔE ads. A function of determining the prediction formula of A program for predicting the coadsorption energy to achieve this.