WO2023115639A1 - Procédé et appareil de calcul d'énergie de cristal moléculaire et support de stockage - Google Patents

Procédé et appareil de calcul d'énergie de cristal moléculaire et support de stockage Download PDF

Info

Publication number
WO2023115639A1
WO2023115639A1 PCT/CN2021/143749 CN2021143749W WO2023115639A1 WO 2023115639 A1 WO2023115639 A1 WO 2023115639A1 CN 2021143749 W CN2021143749 W CN 2021143749W WO 2023115639 A1 WO2023115639 A1 WO 2023115639A1
Authority
WO
WIPO (PCT)
Prior art keywords
energy
lattice
shell
sub
cluster
Prior art date
Application number
PCT/CN2021/143749
Other languages
English (en)
Chinese (zh)
Inventor
曾群
李柏辰
付文博
Original Assignee
深圳晶泰科技有限公司
Priority date (The priority date is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the date listed.)
Filing date
Publication date
Application filed by 深圳晶泰科技有限公司 filed Critical 深圳晶泰科技有限公司
Publication of WO2023115639A1 publication Critical patent/WO2023115639A1/fr

Links

Images

Classifications

    • G—PHYSICS
    • G16—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR SPECIFIC APPLICATION FIELDS
    • G16C—COMPUTATIONAL CHEMISTRY; CHEMOINFORMATICS; COMPUTATIONAL MATERIALS SCIENCE
    • G16C10/00—Computational theoretical chemistry, i.e. ICT specially adapted for theoretical aspects of quantum chemistry, molecular mechanics, molecular dynamics or the like
    • G—PHYSICS
    • G16—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR SPECIFIC APPLICATION FIELDS
    • G16C—COMPUTATIONAL CHEMISTRY; CHEMOINFORMATICS; COMPUTATIONAL MATERIALS SCIENCE
    • G16C20/00—Chemoinformatics, i.e. ICT specially adapted for the handling of physicochemical or structural data of chemical particles, elements, compounds or mixtures
    • G16C20/30—Prediction of properties of chemical compounds, compositions or mixtures

Definitions

  • the application belongs to the technical field of molecular crystals, and in particular relates to a molecular crystal energy calculation method, device and storage medium.
  • Molecular crystal is a kind of structure formed by stacking of organic molecules through weak non-bonding interactions. Accurate description of weak interactions usually requires high-precision methods with extremely high algorithm complexity, and such methods are usually difficult to apply to large molecular crystal systems. Molecular crystal calculations usually choose the cheaper density functional theory method, semi-empirical method or empirical force field method. In order to be able to apply high-precision methods to molecular crystals, people usually decompose the total energy of the crystal into the sum of several subsystems.
  • existing computing schemes usually discard clusters with smaller contributions according to certain rules, which may introduce certain errors.
  • the existing scheme in order to introduce the overall effect of the crystal, the existing scheme usually uses a low-precision method to calculate the energy of the crystal, which also introduces a certain error.
  • this application provides a molecular crystal energy calculation method, device and storage medium.
  • a molecular crystal energy calculation method By establishing a "core-shell" calculation scheme, the calculation efficiency and calculation accuracy of crystal energy can be improved.
  • the first aspect of the present application provides a molecular crystal energy calculation method, including:
  • the total crystal energy of the target crystal structure is calculated.
  • the determination of the central molecule, M clusters and M shell structures from the target crystal structure includes:
  • the central molecule is deleted from each of the clusters to obtain the corresponding M shell structures.
  • the selected atoms are located within the molecular cut-off radius to construct clusters, including:
  • the calculation of the molecular energy of the central molecule, the cluster energy of each of the clusters and the shell energy of each of the shell structures using quantum chemical methods includes:
  • the quantum chemical method is used to calculate the cluster energy of each of the clusters and the component energies that constitute the cluster energy;
  • Each of the shell structures is used as an input, and the quantum chemical method is used to calculate the shell energy of each shell structure and the component energies constituting the shell energy.
  • the quantum chemical method includes a tight-binding DFTB method based on density functional theory, and the energy calculation formula is:
  • E energy E 0 approx + E scc + E rep + E nuc + E disp_corr ;
  • the sub-item energies E 0 approximation , E scc , E rep , Enuc , and E disp_corr are: orbital energy under zero-order approximation, second/third-order electrostatic energy, short-range repulsion energy of valence bonds, and nuclear repulsion energy , the long-range dispersion correction energy.
  • the quantum chemical method includes a density functional theory DFT method, and the energy calculation formula is:
  • E energy E 0 +E j +E x +E c +E nuc +E disp_corr ;
  • the sub-items of energy E 0 , E j , Ex , E c , Enuc , and E disp_corr are: orbital energy, electron electrostatic energy, exchange energy, correlation energy, nuclear repulsion energy, and long-range dispersion correction energy.
  • the calculating the lattice energy of the target crystal structure according to the molecular energy, the cluster energy and the shell energy includes:
  • the partial energies of the molecular energy the partial energies of each of the cluster energies, and the partial energies of each of the shell energies, the relationship between the central molecule and each - Partial energies of the interaction energy between said shell structures;
  • the sub-items of the lattice energy of the target crystal structure are calculated using a second preset formula
  • the lattice energy of the target crystal structure is calculated using a third preset formula.
  • the first preset formula is:
  • E shell n interaction energy, i E cluster n,i -E molecule, i -E shell n,i ;
  • the n is the cluster number
  • the value of n is 1-M
  • the i is the number of each sub-item energy
  • the E cluster n,i is the cluster energy of the nth cluster
  • the i-th partial energy of the E molecule, i is the i-th partial energy of the molecular energy of the central molecule
  • the E shell n,i is the i-th shell energy of the n-th shell structure
  • Subitem energy, the E shell n interaction energy, i is the ith subitem energy of the interaction energy between the central molecule and the nth shell structure.
  • the second preset formula is:
  • the E lattice, i is the i-th sub-item energy of the lattice energy of the target crystal structure
  • the R n is the intercept radius adopted when constructing the n-th cluster
  • the A and The B is two different attenuation coefficients
  • the sub-items of the interaction energy between the central molecule and each of the shell structures are calculated using a second preset formula to obtain the sub-items of the lattice energy of the target crystal structure, including :
  • the second preset formula is:
  • E- lattice, i [E shell 1 interaction energy, i (R 1 ) 3 -E shell 2 interaction energy, i (R 2 ) 3 ]/[(R 1 ) 3 -(R 2 ) 3 ];
  • the E lattice, i is the ith sub-item energy of the lattice energy of the target crystal structure
  • the E shell 1 interaction energy, i is the interaction between the central molecule and the first shell structure
  • the i-th sub-item energy of the action energy, the E -shell 2 interaction energy, i is the i-th sub-item energy of the interaction energy between the central molecule and the second shell structure
  • the R1 is the construction of the first sub-item energy
  • the cut-off radius used when 1 cluster is used, and the R2 is the cut-off radius used when constructing the second cluster.
  • the n is 1, and the second preset formula is:
  • the E lattice, i is the ith sub-item energy of the lattice energy of the target crystal structure
  • the E shell 1 interaction energy, i is the interaction energy between the central molecule and the shell structure The energy of the i-th component.
  • the third preset formula is:
  • E lattice sum(k i E lattice, i );
  • the E lattice is the lattice energy of the target crystal structure
  • the E lattice i is the i-th sub-item energy of the lattice energy of the target crystal structure
  • the ki is the The contribution coefficient of the i-th component energy of the lattice energy.
  • the calculation method of the contribution coefficient of each sub-item energy of the lattice energy includes:
  • the total crystal energy of the target crystal structure is calculated according to the molecular energy and the lattice energy, including:
  • the total crystal energy of the target crystal structure is calculated.
  • the second aspect of the present application provides a molecular crystal energy comparison method, including:
  • the magnitude relationship of the crystal energies of the at least two crystal structures is determined.
  • the third aspect of the present application provides a molecular crystal energy calculation device, including:
  • An acquisition module configured to acquire a target crystal structure, and determine the central molecule, M clusters and M shell structures from the target crystal structure, where M is an integer greater than or equal to 1;
  • the first calculation module is used to separately calculate the molecular energy of the central molecule, the cluster energy of each of the clusters and the shell energy of each of the shell structures by using a quantum chemical method;
  • a second calculation module configured to calculate the lattice energy of the target crystal structure according to the molecular energy, the cluster energy and the shell energy;
  • the third calculation module is used to calculate the total crystal energy of the target crystal structure according to the molecular energy and the lattice energy.
  • the acquisition module determines the central molecule, M clusters and M shell structures from the target crystal structure, including:
  • the central molecule is deleted from each of the clusters to obtain the corresponding M shell structures.
  • the acquisition module selects molecules whose atoms are located within the cut-off radius to construct clusters, including:
  • the first calculation module includes:
  • the first calculation unit is used to use the molecular structure of the central molecule as an input, and calculate the molecular energy of the central molecule and the sub-energy of the molecular energy by using quantum chemical methods;
  • the second calculation unit is used to use the structure of each cluster as an input to calculate the cluster energy of each cluster and the energy of each subitem of the cluster energy by using a quantum chemical method;
  • the third calculation unit is used to use each shell structure as an input to calculate the shell energy of each shell structure and the component energies constituting the shell energy by quantum chemical method.
  • the quantum chemical method includes a tight-binding DFTB method based on density functional theory, and the energy calculation formula is:
  • E energy E 0 approx + E scc + E rep + E nuc + E disp_corr ;
  • the sub-item energies E 0 approximation , E scc , E rep , Enuc , and E disp_corr are: orbital energy under zero-order approximation, second/third-order electrostatic energy, short-range repulsion energy of valence bonds, and nuclear repulsion energy , the long-range dispersion correction energy.
  • the quantum chemical method includes a density functional theory DFT method, and the energy calculation formula is:
  • E energy E 0 +E j +E x +E c +E nuc +E disp_corr ;
  • the sub-items of energy E 0 , E j , Ex , E c , Enuc , and E disp_corr are: orbital energy, electron electrostatic energy, exchange energy, correlation energy, nuclear repulsion energy, and long-range dispersion correction energy.
  • the second calculation module includes:
  • the fourth calculation unit is used to calculate using the first preset formula according to the sub-items of the molecular energy, the sub-items of each of the cluster energies, and the sub-items of each of the shell energies obtaining the partial energies of the interaction energy between the central molecule and each of the shell structures;
  • the fifth calculation unit is used to calculate and obtain the lattice energy of the target crystal structure according to the component energies of the interaction energy between the central molecule and each of the shell structures by using the second preset formula. itemized energy;
  • the sixth calculation unit is configured to calculate the lattice energy of the target crystal structure by using a third preset formula according to the component energies of the lattice energy and the contribution coefficients of the component energies.
  • the first preset formula is:
  • E shell n interaction energy, i E cluster n,i -E molecule, i -E shell n,i ;
  • the n is the cluster number
  • the value of n is 1-M
  • the i is the number of each sub-item energy
  • the E cluster n,i is the cluster energy of the nth cluster
  • the i-th partial energy of the E molecule, i is the i-th partial energy of the molecular energy of the central molecule
  • the E shell n,i is the i-th shell energy of the n-th shell structure
  • Subitem energy, the E shell n interaction energy, i is the ith subitem energy of the interaction energy between the central molecule and the nth shell structure.
  • the second preset formula is:
  • the E lattice, i is the i-th sub-item energy of the lattice energy of the target crystal structure
  • the R n is the intercept radius adopted when constructing the n-th cluster
  • the A and The B is two different attenuation coefficients
  • the fifth calculation unit is specifically used to calculate the component energies of the interaction energy between the central molecule and each of the shell structures using the second preset formula to obtain multiple sets of E crystals Lattice, i , A and B; multiple sets of E lattice, i , A and B are fitted to obtain the sub-item energy of the lattice energy of the target crystal structure.
  • the second preset formula is:
  • E- lattice, i [E shell 1 interaction energy, i (R 1 ) 3 -E shell 2 interaction energy, i (R 2 ) 3 ]/[(R 1 ) 3 -(R 2 ) 3 ];
  • the E lattice, i is the ith sub-item energy of the lattice energy of the target crystal structure
  • the E shell 1 interaction energy, i is the distance between the central molecule and the first shell structure
  • the i-th sub-item energy of the interaction energy, the E -shell 2 interaction energy, i is the i-th sub-item energy of the interaction energy between the central molecule and the second shell structure
  • the R 1 is the cut-off radius used when constructing the first cluster
  • R 2 is the cut-off radius used when constructing the second cluster.
  • the n is 1, and the second preset formula is:
  • the E lattice, i is the ith sub-item energy of the lattice energy of the target crystal structure
  • the E shell 1 interaction energy, i is the interaction between the central molecule and the shell structure energy of the i-th component of energy.
  • the third preset formula is:
  • E lattice sum(k i E lattice, i );
  • the E lattice is the lattice energy of the target crystal structure
  • the E lattice i is the i-th sub-item energy of the lattice energy of the target crystal structure
  • the ki is the The contribution coefficient of the i-th component energy of the lattice energy.
  • the sixth calculation unit calculates the contribution coefficient of each sub-item energy of the lattice energy, including:
  • the third calculation module includes:
  • a calibration unit configured to perform energy calibration on the molecular energy to obtain the corrected molecular energy
  • a seventh calculation unit configured to calculate the total crystal energy of the target crystal structure by using the corrected molecular energy and the lattice energy.
  • the fourth aspect of the present application provides a molecular crystal energy comparison device, including:
  • an acquisition module configured to acquire at least two crystal structures to be compared
  • a calculation module configured to use the molecular crystal energy calculation device provided in the third aspect of the present application to calculate the crystal energy of each of the crystal structures, and obtain the total crystal energy of each of the crystal structures;
  • the determination module is configured to determine the relationship between the crystal energies of the at least two crystal structures according to the total crystal energy of each of the crystal structures.
  • the fifth aspect of the present application provides an electronic device, including:
  • the sixth aspect of the present application provides a computer-readable storage medium, on which executable code is stored, and when the executable code is executed by the processor of the electronic device, the processor executes the method provided in the first aspect of the present application.
  • the molecular crystal energy calculation method or the molecular crystal energy comparison method provided in the second aspect of the present application are examples of the present application.
  • the central molecule, at least one cluster structure and the corresponding shell structure can be determined from it, and the molecular energy of the central molecule, the The cluster energy of the structure and the shell energy of each shell structure are calculated based on the above energies to obtain the lattice energy of the target crystal structure, and then the molecular energy of the central molecule and the lattice energy are used to calculate the total crystal energy of the target crystal structure.
  • This application establishes a "core-shell” calculation scheme by calculating the energy of the central molecule (core) and cluster (shell), which can avoid the calculation of a large number of two-body, three-body and other multi-body clusters in the prior art, and can obtain Compared with the original direct low-precision calculation method, the energy accuracy is higher, and the calculation efficiency is also greatly improved.
  • Fig. 1 is a schematic flow chart of a molecular crystal energy calculation method provided in the embodiment of the present application
  • Fig. 2 is the central molecular structure diagram in a kind of Aspirin crystal provided by the embodiment of the present application;
  • Fig. 3 is the center molecule in Fig. 2 as the center of the sphere, with R 1 as the molecular structure diagram after the cut-off radius is carried out;
  • Fig. 4 is a structural diagram of cluster 1 after molecular completion of the molecular structure in Fig. 3;
  • Fig. 5 is a structural diagram of cluster 2 after the central molecule in Fig. 2 is used as the center of the sphere, and R2 is used as the cut-off radius for atomic interception and completion;
  • Fig. 6 is a schematic structural diagram of a molecular crystal energy calculation device provided in an embodiment of the present application.
  • FIG. 7 is a schematic structural diagram of an electronic device provided by an embodiment of the present application.
  • first, second, third and so on may be used in this application to describe various information, such information should not be limited to these terms. These terms are only used to distinguish information of the same type from one another.
  • first information may also be called second information, and similarly, second information may also be called first information.
  • second information may also be called first information.
  • a feature defined as “first” and “second” may explicitly or implicitly include one or more of these features.
  • “plurality” means two or more, unless otherwise specifically defined.
  • the embodiments of the present application provide a molecular crystal energy calculation method, device and storage medium.
  • a “core-shell” calculation scheme By establishing a “core-shell” calculation scheme, the calculation efficiency and calculation accuracy of crystal energy can be improved.
  • the embodiment of the present application provides a molecular crystal energy calculation method. As shown in Figure 1, the method may include the following steps:
  • M may be an integer greater than or equal to 1.
  • the target crystal structure may be a crystal structure composed of a single molecule, or a crystal structure composed of multiple molecules, which is not limited here.
  • the specific implementation of determining the central molecule, M clusters and M shell structures from the target crystal structure may include: selecting a molecule from the target crystal structure as the central molecule; taking the geometric center of the central molecule as The center of the sphere, and use M different preset radii as cut-off radii to intercept atoms of the target crystal structure, select molecules whose atoms are within the cut-off radius to construct clusters, and obtain M clusters; divide the central molecule from each cluster Delete the corresponding M shell structures.
  • the target crystal structure is a crystal structure composed of a single molecule
  • a molecule can be randomly selected from the target crystal structure as the central molecule, or a molecule can be selected from near the geometric center of the target crystal structure as the central molecule.
  • the target crystal structure is a crystal structure composed of two molecules (such as molecule A and molecule B)
  • the molecule A and molecule B that are closest to each other can be selected from the target crystal structure together as the central molecule.
  • molecules with all atoms located within the cut-off radius can be selected to construct clusters, and molecules with only some atoms within the cut-off radius are excluded;
  • the molecule can be completed, and the completed molecules and molecules with all atoms within the cutoff radius can be used to construct clusters together. That is to draw a sphere with the geometric center of the central molecule and the radius R, select all the atoms within the radius R, and ensure that the sphere contains at least all the atoms on the first neighbor molecules of the central molecule, according to the crystal arrangement, the atoms are filled as For a complete molecule, the set of molecules is called a "cluster", and the cluster without the central molecule is called a "shell".
  • the molecular structure of the central molecule can be used as input, and the molecular energy of the central molecule and the sub-item energies that constitute the molecular energy can be calculated by using quantum chemical methods; the structure of each cluster is used as input, and the same quantum chemical method can be used Calculate the cluster energy of each cluster and the sub-item energies that constitute the cluster energy; take each shell structure as input, and use the same quantum chemical method to calculate the shell energy of each shell structure and the various components that constitute the shell energy. Itemized energy.
  • the quantum chemical method can include a tight binding (Density Functional based Tight Binding, DFTB) method based on density functional theory, and the calculation method of each energy mentioned above can adopt the following formula:
  • E energy E 0 approx + E scc + E rep + E nuc + E disp_corr ; (1-1)
  • the sub-item energy E 0 is approximately the orbital energy under the zero-order approximation
  • E scc is the second/third-order electrostatic energy
  • E rep is the short-range repulsion energy of the valence bond
  • Enuc is the nuclear repulsion energy
  • the quantum chemical method may include a density functional theory (Density Functional Theory, DFT) method, and the calculation method of each energy mentioned above may adopt the following formula:
  • E energy E 0 +E j +E x +E c +E nuc +E disp_corr ; (1-2)
  • the sub-item energy E 0 is the orbital energy
  • E j is the electron electrostatic energy
  • Ex is the exchange energy
  • E c is the correlation energy
  • Enuc is the nuclear repulsion energy
  • E disp_corr is the long-range dispersion correction energy.
  • the quantum chemical method can also use other semi-empirical methods or empirical force field methods, etc., which are not limited here.
  • DFTB is taken as an example for further description. Taking the molecular structure of the central molecule as input, the molecular energy of the central molecule and its sub-items are calculated by formula (1-1).
  • the central molecule Take the geometric center of the central molecule as the center of the sphere, and use the preset radius R1 as the cut-off radius to perform atomic interception on the target crystal structure, and form a cluster with atoms within the range of the preset radius R1 to obtain cluster 1; Taking the structure of cluster 1 as input, the cluster energy and its sub-items of cluster 1 are also calculated by formula (1-1); the central molecule is deleted from cluster 1 to obtain the structure of shell 1, and the shell 1 structure as input, the shell energy and its sub-items of shell 1 structure are also calculated by formula (1-1).
  • the central molecule is deleted from cluster 2 to obtain the shell 2 structure, and The shell 2 structure is taken as input, and the shell energy and its sub-items of the shell 2 structure are also calculated by using the formula (1-1).
  • the central molecule is deleted from cluster 3 to obtain the structure of shell 3, and the The shell 3 structure is used as input, and the shell energy and its sub-items of the shell 3 structure are also calculated by using the formula (1-1).
  • the molecular energy of the central molecule, the cluster energy of each cluster and the shell energy of each shell structure can be used to determine the interaction energy between the central molecule and each shell structure, and then use the mutual The action energy is extrapolated to obtain the lattice energy of the target crystal structure.
  • the second preset formula can be used to calculate the target crystal structure The subitem energies of the lattice energy; and then according to the subitem energies of the lattice energy and the contribution coefficients of the subitem energies, the lattice energy of the target crystal structure is calculated by using the third preset formula.
  • the expression form of the first preset formula is as follows:
  • n is the cluster number, and the value of n is 1-M.
  • i is the number of each sub-item energy, taking the DFTB method as an example, when i is 1, the corresponding sub-item energy is approximately E 0 ; when i is 2, the corresponding sub-item energy is E scc ; when i is 3, The corresponding sub-item energy is E rep ; when i is 4, the corresponding sub-item energy is Enuc ; when i is 5, the corresponding sub-item energy is E disp_corr .
  • E cluster n,i is the i-th sub-item energy of the cluster energy of the nth cluster, E molecule, i is the i-th sub-item energy of the molecular energy of the central molecule, E -shell n,i is the n-th sub-item energy The i-th sub-item energy of the shell energy of a shell structure, E shell n interaction energy, i is the i-th sub-item energy of the interaction energy between the central molecule and the n-th shell structure.
  • the interaction energy between the shell structure and the central molecule under different cut-off radii (R n ) may be attenuated, and the attenuation conforms to the attenuation function related to R n .
  • the second preset formula can be expressed as follows:
  • E lattice, i is the i-th partial energy of the lattice energy of the target crystal structure
  • R n is the interception radius used when constructing the n-th cluster
  • a and B are two different attenuation coefficients, and Both A and B are unknown.
  • each component energy of the interaction energy between the central molecule and each shell structure can be calculated using the second preset formula to obtain multiple sets of E lattices, i , A and B; and for multiple sets of E Lattice, i , A and B are fitted to obtain the sub-items of the lattice energy of the target crystal structure.
  • the E -shell n interaction energy .i is used for energy extrapolation to obtain the E- lattice,i .
  • the E -shell n interaction energy.i and Rn in formula (1-4) are known, and the E lattice, i , A and B are unknown, so the interaction between the three shell structures and the central molecule can be used
  • the action energy is solved for the E lattice,i , A and B, and finally the E lattice,i is obtained.
  • a set of E lattices, i , A and B can be obtained correspondingly by using each 3 shell structures, so as to obtain multiple sets of E lattices, i , A and B, Then fit and optimize multiple groups of E lattice,i , A and B, and finally obtain E lattice,i .
  • performing fitting optimization on multiple sets of E lattices, i , A and B may be performing averaging on multiple sets of E lattices, i , A and B respectively.
  • R n increases accordingly, which leads to a rapid increase in the calculation amount.
  • the embodiment of the present application preferably has M within 3, that is, no more than 3 clusters.
  • a and B are two different attenuation coefficients, and both A and B are unknown.
  • the calculation process of E- lattice,i can be similar to the calculation process of formula (1-4), and will not be repeated here.
  • E- lattice, i [E shell 1 interaction energy, i (R 1 ) 3 -E shell 2 interaction energy, i (R 2 ) 3 ]/[(R 1 ) 3 -(R 2 ) 3 ]; (1-5)
  • E lattice i is the ith sub-item energy of the lattice energy of the target crystal structure
  • E shell 1 interaction energy i is the ith part of the interaction energy between the central molecule and the first shell structure
  • Subitem energy E shell 2 interaction energy, i is the ith subitem energy of the interaction energy between the central molecule and the second shell structure
  • R 1 is the intercept radius used when constructing the first cluster
  • R 2 is the intercept radius used when constructing the second cluster.
  • only one cluster structure can be used to calculate the shell interaction energy. At this time, extrapolation cannot be performed, but sub-items can still be used for fitting to obtain the E lattice,i .
  • M is 1
  • n is also 1
  • the second preset formula can be expressed as follows:
  • E lattice i is the ith sub-item energy of the lattice energy of the target crystal structure
  • E shell 1 interaction energy i is the i-th sub-item energy of the interaction energy between the central molecule and the shell structure .
  • the expression form of the third preset formula is as follows:
  • E lattice sum(k i E lattice, i ); (1-7)
  • E lattice is the lattice energy of the target crystal structure
  • E lattice i is the ith sub-item energy of the lattice energy of the target crystal structure
  • k i is the contribution of the i-th sub-item energy of the lattice energy coefficient.
  • the contribution coefficients of each component energy of the lattice energy can be set empirical values, or a certain amount of E -shell n interaction energy .i calculated before the calculation of the crystal energy, and compared with experiments or high
  • the precision lattice can be combined and the fit determined.
  • the sub-items of the extrapolated lattice energies are directly summed, physically they should be equal to the lattice energy, but in practical applications, it is found that: a) Although the overall energy may be consistent, different methods have different performances in terms of sub-items The trend appears to be different; b) The energy difference of the sub-items may lead to large errors in the calculation of the lattice energy. Therefore, the experimental lattice energy or high-precision lattice energy is used as a reference here, and each lattice energy sub-item of the lower-precision method is recalibrated to obtain a lattice energy closer to the reference.
  • N reference crystal structures and reference lattice energies E ref,j of these reference crystal structures can be collected from previous experimental results, literature reports and/or structure databases, where j is 1-N.
  • the reference crystal structure refers to the crystal structure or experimental crystal structure predicted by theoretical methods (such as DFT, DFTB, etc.), and the reference lattice energy is the crystal lattice energy predicted by high-precision theoretical methods (such as DFT) or Test measurements.
  • each reference crystal structure is used as input, and the sub-item energy of the lattice energy of each reference crystal structure can be obtained through the aforementioned series of calculations: E lattice, 0 approximation , E lattice, scc , E lattice, rep , E lattice, disp_corr , Enuc .
  • each sub-entry energy E lattice,i of the lattice energy of N reference crystal structures can be organized into a matrix, and the representation of the matrix (X) is as follows:
  • reference lattice energy E ref,j of each reference crystal structure can be expressed in vector form:
  • E [E ref,1 E ref,2 ... E ref,N ] T ;
  • Contribution coefficient ki is a parameter to be determined, which can be expressed in vector form:
  • K [k 0 approximate k scc k rep k disp_corr k nuc ] T ;
  • K [k 0 approximate k scc k rep k disp_corr 1] T ;
  • the sub-items of the predicted lattice energy of each reference crystal structure are obtained.
  • E XK
  • E* XK
  • E* contains the sub-items of the predicted lattice energies of each reference crystal structure.
  • E* is close to E
  • K is obtained when the error between E* and E (such as root mean square error RMSE, mean square error MSE, etc.) is the smallest.
  • the number N of the reference crystal structure is usually much larger than the number i of the sub-item energy, therefore, it can be obtained by directly fitting K using an optimization algorithm (such as the least square method).
  • the energy sum of the two can be calculated to obtain the total crystal energy of the target crystal structure.
  • the specific calculation formula is as follows:
  • E total energy is the total crystal energy of the target crystal structure
  • E molecule is the molecular energy of the central molecule
  • E lattice is the lattice energy of the target crystal structure.
  • an Aspirin crystal is taken as an example for illustration.
  • the Aspirin crystal has P21/c symmetry, and one unit cell contains 4 equivalent molecules.
  • the low-precision DFTB method is selected to extrapolate the 2-shell structure and predict the total energy of the crystal.
  • the Aspirin lattice energy calculated by the high-precision DFT method is -118kJ/mol.
  • R 1 4.0 angstroms as the cut-off radius, and use the geometric center of the central molecule as the center of the sphere to select atoms, as shown in FIG. 3 , which shows the structures of all atoms within the cut-off radius R 1 . It can be seen from the figure that the peripheral molecules are not complete, so the molecular completion operation is required;
  • Figure 4 shows the structure of cluster 1 after the molecules are filled.
  • the structure of the ball and stick model is represented as the central molecule, and the rest of the line model is the shell 1 structure.
  • Table 1 The sub-item energy calculated by DFTB method under 3OB parameters (unit: a.u.)
  • Table 2 The sub-item energies extrapolated by DFTB method under 3OB parameters (unit: a.u.)
  • the Aspirin crystal has P21/c symmetry, and one unit cell contains 4 equivalent molecules.
  • the combination of the PBE-D3BJ method with normal precision and the 6-31G* basis set in the density functional theory (DFT) method is selected to predict the crystal energy of the 1-shell structure.
  • R 1 4.0 angstroms as the cut-off radius, and use the geometric center of the central molecule as the center of the sphere to select atoms.
  • the structure of the ball-and-stick model is represented as the central molecule. Atoms selected with a cut-off radius of 4.0 angstroms, it can be seen that the peripheral molecules are not complete and need to be completed.
  • Table 3 The sub-item energy obtained under PBE-D3BJ and def-SV(P) (unit: a.u.)
  • the total crystal energy of the Aspirin crystal in the current method can be obtained: -1700088.5kJ/mol.
  • energy correction can be performed on the molecular energy of the central molecule first to obtain the corrected molecular energy; then, the total crystal energy of the target crystal structure can be calculated by using the corrected molecular energy and lattice energy.
  • the molecular energy and lattice energy are both calculated by the DFTB method, and then the molecular energy of the central molecule can be recalculated using a higher-precision method (such as the MP2/aug-cc-pVQZ method) to replace the previous
  • the molecular energy calculated by the low-precision method (DFTB) is used to obtain better and more reliable molecular energy, and then the total energy of the crystal with higher confidence is obtained to further improve the calculation accuracy.
  • correction methods may also be used, such as correcting at least one partial energy of the molecular energy, so as to obtain more reliable molecular energy.
  • the method provided in the embodiment of the present application establishes a "core-shell” calculation scheme by calculating the energy of the central molecule (core) and cluster (shell), which can avoid a large number of multi-body groups such as two-body and three-body groups in the prior art
  • the cluster calculation can obtain higher energy accuracy than the original low-precision calculation method directly, and the calculation efficiency is also greatly improved.
  • This application uses the technology of central molecule-peripheral clusters to evaluate crystal energy, which is compatible with mainstream quantum chemical calculation methods, can better balance calculation accuracy and cost, and can be used for fast calculation of crystal energy.
  • This application can directly use a large cluster for calculation, which reduces the number of calculations and avoids the steps of dividing, screening and calculating small clusters such as two bodies in the existing method.
  • this application uses the calculation results of the two shells for extrapolation.
  • the cluster can be fitted directly with a low-precision method, and the lattice energy can be directly obtained to further improve the overall calculation efficiency.
  • the embodiment of the present application also provides a molecular crystal energy comparison method, including the following steps:
  • the embodiment of the present application also provides a molecular crystal energy calculation device, which can be used to implement the molecular crystal energy calculation method provided in the foregoing embodiments.
  • the device may include:
  • An acquisition module 610 configured to acquire the target crystal structure, and determine the central molecule, M clusters and M shell structures from the target crystal structure, where M is an integer greater than or equal to 1;
  • the first calculation module 620 is used to separately calculate the molecular energy of the central molecule, the cluster energy of each cluster and the shell energy of each shell structure by using a quantum chemical method;
  • the second calculation module 630 is used to calculate the lattice energy of the target crystal structure according to the molecular energy, cluster energy and shell energy;
  • the third calculation module 640 is used to calculate the total crystal energy of the target crystal structure according to the molecular energy and the lattice energy.
  • the implementation of determining the central molecule, M clusters and M shell structures from the target crystal structure by the acquisition module 610 may include: selecting a molecule from the target crystal structure as the central molecule; taking the geometric center of the central molecule as The center of the sphere, and use M different preset radii as cut-off radii to intercept atoms of the target crystal structure, select molecules whose atoms are within the cut-off radius to construct clusters, and obtain M clusters; divide the central molecule from each cluster Delete the corresponding M shell structures.
  • the obtaining module 610 may further include:
  • an acquisition unit configured to acquire the target crystal structure
  • Selection unit used to select a molecule from the target crystal structure as the central molecule
  • the interception unit is used to take the geometric center of the central molecule as the center of the sphere, and use M different preset radii as cut-off radii to carry out atomic interception on the target crystal structure, select molecules whose atoms are within the cut-off radius to construct clusters, and obtain M a cluster;
  • the deletion unit is used to delete the central molecule from each cluster to obtain the corresponding M shell structures.
  • selecting molecules with atoms within the cutoff radius to construct clusters may include: selecting molecules with all atoms within the cutoff radius to construct clusters; or, if there are molecules with only some atoms within the cutoff radius , complete the molecule and build clusters using the completed molecule and molecules with all atoms within the cutoff radius.
  • the first calculation module 620 may further include:
  • the first calculation unit is used to use the molecular structure of the central molecule as an input, and calculate the molecular energy of the central molecule and the sub-energy of the molecular energy by using quantum chemical methods;
  • the second calculation unit is used to use the structure of each cluster as an input to calculate the cluster energy of each cluster and the energy of each subitem of the cluster energy by using a quantum chemical method;
  • the third calculation unit is used to use each shell structure as an input to calculate the shell energy of each shell structure and the component energies constituting the shell energy by quantum chemical method.
  • the sub-item energies E 0 approximation , E scc , E rep , Enuc , and E disp_corr are: orbital energy under zero-order approximation, second/third-order electrostatic energy, short-range repulsion energy of valence bonds, and nuclear repulsion energy , the long-range dispersion correction energy.
  • the sub-items of energy E 0 , E j , Ex , E c , Enuc , and E disp_corr are: orbital energy, electron electrostatic energy, exchange energy, correlation energy, nuclear repulsion energy, and long-range dispersion correction energy.
  • the second computing module 630 may further include:
  • the fourth calculation unit is used to calculate and obtain the central molecule and each sub-energy according to the sub-item energy of the molecular energy, the sub-item energy of each cluster energy, and the sub-item energy of each shell energy by using the first preset formula.
  • the fifth calculation unit is used to calculate the sub-items of the lattice energy of the target crystal structure by using the second preset formula according to the sub-items of the interaction energy between the central molecule and each shell structure;
  • the sixth calculation unit is used to calculate the lattice energy of the target crystal structure by using the third preset formula according to the component energies of the lattice energy and the contribution coefficients of the component energies.
  • n is the cluster number, the value of n is 1 ⁇ M, i is the number of each sub-item energy, E cluster n, i is the i-th sub-item energy of the cluster energy of the nth cluster, E molecule, i is the i-th partial energy of the molecular energy of the central molecule, E -shell n, i is the i-th partial energy of the shell energy of the n-th shell structure, E -shell n interaction energy, i is the center The energy of the ith component of the interaction energy between the molecule and the nth shell structure.
  • the second preset formula may be:
  • E lattice, i is the i-th sub-item energy of the lattice energy of the target crystal structure
  • R n is the intercept radius used when constructing the n-th cluster
  • a and B are two different attenuation coefficients
  • the fifth calculation unit can specifically be used to calculate the sub-item energies of the interaction energy between the central molecule and each shell structure using the second preset formula to obtain multiple sets of E lattices, i , A and B: Fit multiple groups of E lattice, i , A and B to obtain the sub-item energy of the lattice energy of the target crystal structure.
  • the second preset formula can be:
  • E- lattice, i [E shell 1 interaction energy, i (R 1 ) 3 -E shell 2 interaction energy, i (R 2 ) 3 ]/[(R 1 ) 3 -(R 2 ) 3 ];
  • E lattice i is the ith sub-item energy of the lattice energy of the target crystal structure
  • E shell 1 interaction energy i is the ith part of the interaction energy between the central molecule and the first shell structure
  • Subitem energy E shell 2 interaction energy, i is the ith subitem energy of the interaction energy between the central molecule and the second shell structure
  • R 1 is the intercept radius used when constructing the first cluster
  • R 2 is the intercept radius used when constructing the second cluster.
  • E lattice i is the ith sub-item energy of the lattice energy of the target crystal structure
  • E shell 1 interaction energy i is the i-th sub-item energy of the interaction energy between the central molecule and the shell structure .
  • E lattice is the lattice energy of the target crystal structure
  • E lattice i is the ith sub-item energy of the lattice energy of the target crystal structure
  • k i is the contribution of the i-th sub-item energy of the lattice energy coefficient.
  • the way for the sixth calculation unit to calculate the contribution coefficients of each sub-item energy of the lattice energy may include: obtaining multiple reference crystal structures and the reference lattice energy of each reference crystal structure; Each sub-item energy of the lattice energy; set the contribution coefficient of each sub-item energy of the lattice energy; according to each sub-item energy of the lattice energy of each reference crystal structure and the set contribution coefficient, get each reference Each sub-entry energy of the predicted lattice energy of the crystal structure; use each sub-item energy of the reference lattice energy of each reference crystal structure and each sub-item energy of the predicted lattice energy to fit the set contribution coefficient, The contribution coefficient of each sub-item energy after fitting is obtained.
  • the third calculation module 640 may further include:
  • a correction unit is used to perform energy correction on molecular energy to obtain corrected molecular energy
  • the seventh calculation unit is used to calculate the total crystal energy of the target crystal structure by using the corrected molecular energy and lattice energy.
  • the device in the embodiment of this application establishes a "core-shell” calculation scheme by calculating the energy of the central molecule (core) and cluster (shell), which can avoid a large number of multi-body groups such as two-body and three-body groups in the prior art
  • the cluster calculation can obtain higher energy accuracy than the original low-precision calculation method directly, and the calculation efficiency is also greatly improved.
  • the embodiment of the present application also provides a molecular crystal energy comparison device, which can be used to implement the molecular crystal energy comparison method provided in the foregoing embodiments.
  • the device may include:
  • an acquisition module configured to acquire at least two crystal structures to be compared
  • Calculation module for calculating the crystal energy of each crystal structure by using the molecular crystal energy calculation device provided in the foregoing embodiments, to obtain the total crystal energy of each crystal structure;
  • the determination module is configured to determine the relationship between the crystal energies of the above-mentioned at least two crystal structures according to the total crystal energy of each crystal structure.
  • an electronic device 700 includes a memory 710 and a processor 720 .
  • the processor 720 can be a central processing unit (Central Processing Unit, CPU), and can also be other general-purpose processors, digital signal processors (Digital Signal Processor, DSP), application specific integrated circuits (Application Specific Integrated Circuit, ASIC), on-site Field-Programmable Gate Array (FPGA) or other programmable logic devices, discrete gate or transistor logic devices, discrete hardware components, etc.
  • a general-purpose processor may be a microprocessor, or the processor may be any conventional processor, or the like.
  • the memory 710 may include various types of storage units such as system memory, read only memory (ROM), and persistent storage.
  • the ROM may store static data or instructions required by the processor 720 or other modules of the computer.
  • the persistent storage device may be a readable and writable storage device.
  • Persistent storage may be a non-volatile storage device that does not lose stored instructions and data even if the computer is powered off.
  • the permanent storage device adopts a mass storage device (such as a magnetic or optical disk, flash memory) as the permanent storage device.
  • the permanent storage device may be a removable storage device (such as a floppy disk, an optical drive).
  • the system memory can be a readable and writable storage device or a volatile readable and writable storage device, such as dynamic random access memory.
  • System memory can store some or all of the instructions and data that the processor needs at runtime.
  • the memory 710 may include any combination of computer-readable storage media, including various types of semiconductor memory chips (such as DRAM, SRAM, SDRAM, flash memory, programmable read-only memory), and magnetic disks and/or optical disks may also be used.
  • memory 710 may include a readable and/or writable removable storage device, such as a compact disc (CD), a read-only digital versatile disc (e.g., DVD-ROM, dual-layer DVD-ROM), Read-only Blu-ray Disc, Super Density Disc, Flash memory card (such as SD card, min SD card, Micro-SD card, etc.), magnetic floppy disk, etc.
  • a readable and/or writable removable storage device such as a compact disc (CD), a read-only digital versatile disc (e.g., DVD-ROM, dual-layer DVD-ROM), Read-only Blu-ray Disc, Super Density Disc, Flash memory card (such as SD card, min SD card, Micro-SD card, etc.), magnetic floppy disk, etc.
  • Computer-readable storage media do not contain carrier waves and transient electronic signals transmitted by wireless or wire.
  • Executable codes are stored in the memory 710 , and when the executable codes are processed by the processor 720 , the processor 720 can be made to execute part or all of the methods mentioned above.
  • the method according to the present application can also be implemented as a computer program or computer program product, the computer program or computer program product including computer program code instructions for executing some or all of the steps in the above method of the present application.
  • the present application may also be implemented as a computer-readable storage medium (or a non-transitory machine-readable storage medium or a machine-readable storage medium), on which executable code (or computer program or computer instruction code) is stored,
  • executable code or computer program or computer instruction code
  • the processor of the electronic device or server, etc.
  • the processor is made to perform part or all of the steps of the above-mentioned method according to the present application.

Landscapes

  • Theoretical Computer Science (AREA)
  • Engineering & Computer Science (AREA)
  • Computing Systems (AREA)
  • Life Sciences & Earth Sciences (AREA)
  • Bioinformatics & Cheminformatics (AREA)
  • Bioinformatics & Computational Biology (AREA)
  • Health & Medical Sciences (AREA)
  • General Health & Medical Sciences (AREA)
  • Spectroscopy & Molecular Physics (AREA)
  • Physics & Mathematics (AREA)
  • Chemical & Material Sciences (AREA)
  • Crystallography & Structural Chemistry (AREA)
  • Analysing Materials By The Use Of Radiation (AREA)

Abstract

Procédé et appareil de calcul d'énergie de cristal moléculaire et support de stockage. Le procédé consiste : à obtenir une structure de cristal cible et à déterminer une molécule centrale, M grappes et M structures d'enveloppe à partir de la structure de cristal cible (S110) ; à calculer séparément l'énergie moléculaire de la molécule centrale, l'énergie de grappe de chaque grappe et l'énergie d'enveloppe de chaque structure d'enveloppe à l'aide d'un procédé de chimie quantique (S120) ; à calculer l'énergie de réseau de la structure de cristal cible en fonction de l'énergie moléculaire, de l'énergie de grappe et de l'énergie d'enveloppe (S130) ; et à calculer l'énergie de cristal totale de la structure de cristal cible en fonction de l'énergie moléculaire et de l'énergie de réseau (S140). Selon la solution technique, l'efficacité de calcul et la précision de calcul de l'énergie de cristal peuvent être améliorées par l'établissement d'une solution de calcul « écorce-cœur ».
PCT/CN2021/143749 2021-12-24 2021-12-31 Procédé et appareil de calcul d'énergie de cristal moléculaire et support de stockage WO2023115639A1 (fr)

Applications Claiming Priority (2)

Application Number Priority Date Filing Date Title
CN202111600787.1A CN114464263A (zh) 2021-12-24 2021-12-24 一种分子晶体能量计算方法、装置及存储介质
CN202111600787.1 2021-12-24

Publications (1)

Publication Number Publication Date
WO2023115639A1 true WO2023115639A1 (fr) 2023-06-29

Family

ID=81407339

Family Applications (1)

Application Number Title Priority Date Filing Date
PCT/CN2021/143749 WO2023115639A1 (fr) 2021-12-24 2021-12-31 Procédé et appareil de calcul d'énergie de cristal moléculaire et support de stockage

Country Status (2)

Country Link
CN (1) CN114464263A (fr)
WO (1) WO2023115639A1 (fr)

Citations (5)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US20100023473A1 (en) * 2006-12-11 2010-01-28 Neumann Marcus A Tailor-made force fields for crystal structure prediction
CN108959842A (zh) * 2018-05-04 2018-12-07 深圳晶泰科技有限公司 用于有机分子晶体结构预测中高精度能量排位方法
CN112086132A (zh) * 2020-08-18 2020-12-15 深圳晶泰科技有限公司 有机分子晶体构建方法及系统
CN112216355A (zh) * 2020-10-22 2021-01-12 哈尔滨理工大学 一种基于机器学习的多组分晶体构型能预测方法
CN113807526A (zh) * 2021-09-26 2021-12-17 深圳市腾讯计算机系统有限公司 量子体系的本征态获取方法、装置、设备及存储介质

Patent Citations (5)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US20100023473A1 (en) * 2006-12-11 2010-01-28 Neumann Marcus A Tailor-made force fields for crystal structure prediction
CN108959842A (zh) * 2018-05-04 2018-12-07 深圳晶泰科技有限公司 用于有机分子晶体结构预测中高精度能量排位方法
CN112086132A (zh) * 2020-08-18 2020-12-15 深圳晶泰科技有限公司 有机分子晶体构建方法及系统
CN112216355A (zh) * 2020-10-22 2021-01-12 哈尔滨理工大学 一种基于机器学习的多组分晶体构型能预测方法
CN113807526A (zh) * 2021-09-26 2021-12-17 深圳市腾讯计算机系统有限公司 量子体系的本征态获取方法、装置、设备及存储介质

Also Published As

Publication number Publication date
CN114464263A (zh) 2022-05-10

Similar Documents

Publication Publication Date Title
Hoffmann et al. Molecular structures of unbound and transcribing RNA polymerase III
US20230054868A1 (en) Method and system for estimating ground state energy of quantum system
TWI512499B (zh) 使用快速評估技術之以有效窮盡路徑為基礎的靜態時序分析
US11651134B2 (en) Method of certifying safety levels of semiconductor memories in integrated circuits
US20160140272A1 (en) Method to measure edge-rate timing penalty of digital integrated circuits
WO2023115639A1 (fr) Procédé et appareil de calcul d'énergie de cristal moléculaire et support de stockage
WO2023065475A1 (fr) Procédé et appareil de prédiction de structure cristalline, et dispositif électronique
Tan et al. Locally weighted histogram analysis and stochastic solution for large-scale multi-state free energy estimation
CN106528905B (zh) 评估集成电路的系统及方法
CN116092593B (zh) 电解液成分预测方法、装置和计算机设备
WO2019091443A1 (fr) Procédé, appareil et dispositif d'ajustement basés sur un réseau neuronal
JP4646064B2 (ja) 分子軌道法計算方法及び計算装置
Hayashi et al. Calculation of the residual entropy of Ice Ih by Monte Carlo simulation with the combination of the replica-exchange Wang–Landau algorithm and multicanonical replica-exchange method
Lin et al. Bayesian hierarchical random-effects meta-analysis and design of phase I clinical trials
CN112243247A (zh) 基站优化优先级确定方法、装置及计算设备
CN113835307B (zh) 一种电子束邻近效应的校正方法及其装置
WO2022133895A1 (fr) Procédé et appareil de traitement de données de qualité d'équipement thermique à base de supervision d'équipement
WO2021004355A1 (fr) Procédé et appareil de construction de banque de leurres, procédé et appareil de construction de banque de leurres cibles, et procédé et appareil d'identification métabolomique des fdr
US20120253775A1 (en) Multidimensional Monte-Carlo Simulation for Yield Prediction
JP2006343244A (ja) デコンボリューション解析装置、デコンボリューション解析プログラム及びデコンボリューション解析方法
CN117497198B (zh) 一种高维医学数据特征子集筛选方法
Huang et al. Efficient estimation of the maximal association between multiple predictors and a survival outcome
TW201710978A (zh) 系統層級參數估測方法
LU503076B1 (en) High-precision method for inverting concentration of carbon dioxide
US20230205651A1 (en) Identification of optimal bit apportionments for digital functions subject to soft errors

Legal Events

Date Code Title Description
121 Ep: the epo has been informed by wipo that ep was designated in this application

Ref document number: 21968763

Country of ref document: EP

Kind code of ref document: A1