Highly versatile design method of transition metal chromophores, apparatus and storage medium

Abstract

The present application relates to the field of computational chemistry, in particular to the universality optimization in the process of computational design of transition metal chromophores. The present application is realized by the technical scheme that a high-universality design method of transition metal chromophores comprises the following steps: establishing a molecular database; performing scanning calculation on the molecules in the molecular database by using a calculation model, and regarding the molecules with calculation results meeting the screening requirements as candidate molecules, wherein the calculation model adopts DFA; and the calculation model adopts multiple DFAs for calculation. The purpose of the present application is to provide a high-universality design method of transition metal chromophores, equipment and storage medium, which can design functional molecules of metal chromophores in the face of different design systems; greatly improve the reliability of the output results of VHTS in the design process of different metal chromophores; provide better training data for subsequent AI model training, thereby improving the accuracy of the design results output by the AI model.

Description

Technical Field

[0001] The present invention relates to the field of computational chemistry, and in particular to universal optimization in the computational design process of transition metal chromophores. Background Art

[0002] Transition metal chromophores are groups or compounds containing transition metal ions or atoms that absorb photons in the visible light region, producing specific colors or luminescence. These colors or optical properties are primarily derived from the unique electronic structure and energy level distribution of transition metal ions.

[0003] Due to their unique optical properties, transition metal chromophores play a central role in diverse fields, such as light absorbers and harvesters. Transition metal chromophores are specifically designed to efficiently absorb specific wavelengths of light (particularly visible and near-infrared light), converting this energy into excited-state energy within the molecule. This is the basis for dye-sensitized solar cells, photocatalysis, photodynamic therapy, and other applications.

[0004] Another example: signal generators / reporters: These utilize significant changes in the light absorption or luminescence properties of metal chromophores (color change, fluorescence quenching / enhancement, wavelength shift, etc.) as signal outputs in response to external stimuli (specific molecular binding, pH, redox state changes, etc.). Based on this, corresponding sensors and probes are designed.

[0005] Designing a transition metal chromophore involves selecting and designing a corresponding transition metal chromophore molecule. However, this molecule must meet at least two conditions. Condition 1: The transition metal chromophore must be energetically stable in its ground state. Condition 2: It must possess an optimal absorption energy in the visible region. This means it must efficiently absorb ordinary visible light (such as sunlight or lamplight). Finding a transition metal chromophore that satisfies both of these conditions is not easy. Initially, the industry employed a standard trial-and-error approach, known as the "Edison method," which relies on empirical rules and extensive experimental trials to screen and refine designs. However, the design process is challenging due to the need to simultaneously optimize and consider multiple interrelated parameters, such as metal-ligand interactions, electron-donating / withdrawing effects, and ligand field strength. Therefore, designing a transition metal chromophore is a complex and complex task.

[0006] In order to match this complex design task, people often use automated computing models to perform calculations. The mainstream automated computing model is the combination of virtual high-throughput screening (VHTS) and machine learning (ML). VHTS uses the DFA method to calculate the properties of molecular structures, such as the ability to absorb light energy and stability mentioned above. ML uses AI models to screen, iterate, and optimize molecules, and in this process, the judgment results of VHTS will also become the training data for ML. In the calculation process of VHTS, DFA, density functional approximation, is involved. When calculating, DFA does not track the exact position and speed of each electron, but focuses on the distribution of electron density. The subsequent output judgment result is the base energy and electron density distribution map of the entire molecule. It can also calculate the output absorption energy and excited state properties.

[0007] However, in existing technologies, a single density functional is often used, meaning a single DFA is used in the calculation process. However, due to the inherent approximation of the exchange-correlation functional and the complexity of the electronic structure of transition metals, a single DFA cannot reliably predict the ground state spin, excitation energy, and multi-reference properties of TMC chromophores, also known as TMC Transition Metal Complexes. The output of the coordinated VHTS and ML results will largely depend on the choice of DFA. When using a single DFA approach in VHTS, the choice of DFA can lead to large deviations in the generated dataset, which in turn leads to deviations in the candidates recommended by the ML model. In other words, a DFA that performs well on some systems may fail significantly on others. Summary of the Invention

[0008] The purpose of the present invention is to provide a highly versatile design method, equipment and storage medium for transition metal chromophores, which can be used to design functional molecules of metal chromophores in the face of different design systems; greatly improve the reliability of the output results of VHTS in the design process of different metal chromophores, provide better training data for subsequent AI model training, and thus improve the accuracy of the design results output by the AI ​​model.

[0009] The present invention is achieved through the following technical solution: a highly versatile design method for transition metal chromophores, comprising the following steps:

[0010] Establishing molecular databases;

[0011] Scanning and calculating the molecules in the molecular database using a computational model, and considering the molecules whose computational results meet the screening requirements as candidate molecules, wherein the computational model adopts DFA;

[0012] The computational model uses multiple DFAs for computation, and the multiple DFAs are configured such that: in a specified DFA type set, each specified DFA type has at least one DFA; the specified DFA type set includes an LDA level, a GGA level, a hybrid functional level, and a bi-hybrid functional level;

[0013] The judgment result is determined by DFA consensus, and includes qualitative results and quantitative results. The qualitative result is generated based on the recognition ratio of all DFAs being greater than a preset threshold; the quantitative result is generated based on the average of the calculated values ​​of all DFAs.

[0014] As a preferred embodiment of the present invention, the LDA level may adopt one or more of the following DFAs: SVWN5, VWN, Xα;

[0015] The GGA level may use one or more of the following DFA: PBE, BLYP, PBEsol, revPBE;

[0016] The hybrid functional level can adopt one or more of the following DFA: B3LYP, PBE0, HSE06;

[0017] The dual hybrid functional level may adopt one or more of the following DFAs: B2PLYP, ωB97X-D, XYG3, PWPB95.

[0018] As a preferred embodiment of the present invention, it further includes a DFA supplementation step, in which the calculation model uses the candidate DFA to perform a secondary supplementation calculation on the judgment result.

[0019] As a preference of the present invention, in the DFA supplementation step, if the molecular spin state identification needs to be supplemented, TPSSh and SCAN are introduced for secondary supplementary calculations; if the molecular absorption ability identification needs to be supplemented, CAM-B3LYP and DSD-PBEP86 are introduced for secondary supplementation.

[0020] As a preferred embodiment of the present invention, it also includes an optimal DFA set determination step; in this step, the optimal DFA set is calculated based on the genetic algorithm GA, and the optimal DFA set reflects the optimal solution of the number and type of DFAs used in the calculation model. The constraints in the algorithm include constraint condition 1: comparison of the ground state spin misjudgment rate with a preset value; constraint condition 2: comparison of the fluctuation with a preset value.

[0021] As a preferred embodiment of the present invention, the constraint condition further includes constraint condition three: missed detection rate constraint, ie, comparison of the multi-reference missed detection rate with a preset value.

[0022] As a preferred embodiment of the present invention, when establishing molecular data, the following sub-steps are included:

[0023] bidentate ligand selection step;

[0024] The bidentate ligand is selected based on whether it is a common element and the number of atoms;

[0025] Basic complex selection step;

[0026] Combine the two ends of the bidentate ligand with two common metal elements to form a basic complex;

[0027] functionalization step;

[0028] Functional groups are added to the basic complexes to form the final molecular database.

[0029] As a preference of the present invention, in the basic complex selection step, the metal elements are iron and cobalt.

[0030] An electronic device comprises a processor and a memory; the processor is connected to the memory;

[0031] The memory is used to store executable program code; the processor runs a program corresponding to the executable program code by reading the executable program code stored in the memory, so as to execute the method.

[0032] A computer-readable storage medium stores a computer program, which implements the method when executed by a processor.

[0033] In summary, the present invention has the following beneficial effects:

[0034] 1. The coordinated use of four DFA types can meet the design requirements of different scenarios.

[0035] 2. The LDA level is the local density approximation, and the GGA level is the generalized gradient approximation. The combination of the two can adapt to design scenarios with higher efficiency requirements; while the hybrid functional level and double hybrid functional level can make more accurate corrections for charge transfer underestimation and hybrid spin misjudgment in the design process, which can ensure the requirements for precision and accuracy in the design scenario.

[0036] 3. The multi-DFA + consensus mechanism solution can improve anti-bias performance. The TMC error of a single DFA can reach >1eV, while the consensus of multiple DFAs can compress the prediction fluctuation to <0.3eV.

[0037] 4. The consensus results are used directly to train the subsequent ML model, rather than all the DFA judgment results, which significantly improves the active learning iteration speed.

[0038] 5. A DFA supplementation step has been added. If some important qualitative or quantitative results are inaccurate or the recognition is unsuccessful, the calculation model will introduce other DFAs for secondary calculation, further optimizing the "accuracy-efficiency" contradiction.

[0039] 6. Use a genetic algorithm model to determine the optimal DFA set, achieving a balance between computational cost and accuracy in quantitative design. In the selection and matching of specific DFAs, the optimal functional combination is matched according to molecular characteristics.

[0040] 7. The molecules in the molecular database finally have a good possibility of synthesis and are abundant on Earth, that is, the elements are abundant in the earth's crust, the mining cost is low, the supply is relatively stable and the environmental impact is controllable. BRIEF DESCRIPTION OF THE DRAWINGS

[0041] In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the following briefly introduces the drawings required for use in the embodiments or the description of the prior art. Obviously, the drawings described below are only some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying any creative work.

[0042] Figure 1 is a schematic flow chart of the steps of Example 1;

[0043] Figure 2 is a schematic flow chart of the steps of Example 3;

[0044] Figure 3 This is a schematic flow chart of the detailed steps of S01 in Example 4. DETAILED DESCRIPTION

[0045] The present invention will be further described in detail below with reference to the accompanying drawings.

[0046] The technical solutions in the embodiments of this specification will be described clearly and completely below in conjunction with the drawings in the embodiments of this specification.

[0047] Throughout this specification, the claims, and the accompanying drawings, the terms "first," "second," "third," and the like are used to distinguish between different items, not to describe a particular order. Furthermore, the terms "including," "having," and any variations thereof, are intended to cover non-exclusive inclusions. For example, a process, method, system, product, or apparatus comprising a series of steps or elements is not limited to the listed steps or elements but may optionally include steps or elements not listed, or may include other steps or elements inherent to the process, method, product, or apparatus.

[0048] The following description provides examples and does not limit the scope, applicability, or examples set forth in the claims. Changes may be made to the functions and arrangements of the elements described without departing from the scope of this specification. Various examples may appropriately omit, replace, or add various processes or components. For example, the described methods may be performed in an order different from the order described, and various steps may be added, omitted, or combined. Furthermore, features described with respect to some examples may be combined in other examples.

[0049] Example 1, as Figure 1 As shown in the figure, a highly versatile design method for transition metal chromophores is described. As described in the background of this specification, step S01 requires the establishment of a molecular database. This molecular database can be selected from existing databases or a fusion of multiple databases. Examples include the Cambridge Structural Database (CSD), the Inorganic Crystal Structure Database (ICSD), the Materials Project, and the Protein Data Bank (PDB). In this embodiment, the Cambridge Structural Database (CSD) is selected.

[0050] The computational model then scans and calculates the molecules in the molecular database, a process similar to traversal computation. Molecular databases contain a large number of molecules, often numbering in the tens of millions, meaning there are 30 to 50 million design options for photosensitive molecules. The computational model combines virtual high-throughput screening (VHTS) and machine learning (ML). The two work together collaboratively. Specifically, VHTS uses the DFA method to calculate molecular structural properties, such as their ability to absorb light energy and stability. Then, in S05, ML uses an AI model to screen, iterate, and optimize the molecules. During this process, the VHTS's judgment results also become training data for the ML. Simply put, VHTS uses ultra-large-scale, batch computation and screening of molecules in the entire molecular database, simulating and predicting the properties of these 30 to 50 million photosensitive molecules. The VHTS's calculations and predictions also become training data for the ML. The ML leverages patterns in this training data to guide the VHTS in screening the molecular database more efficiently and intelligently, and can even create new molecules from scratch.

[0051] Finally, in S07, the computational model produces the final result. Regardless of whether the final result is a screened result or a newly created result, the molecule corresponding to the final result is the candidate molecule for subsequent research and verification by engineering personnel.

[0052] The above can adopt mature content, mature VHTS, and mature ML systems from the existing technology, and this embodiment does not make any changes here. The core innovation of this embodiment focuses on the use of DFA. Specifically, in the existing technology, the calculation process of VHTS involves DFA, and it is a single DFA.

[0053] During their research and development, the inventors discovered that single DFAs inherently have certain inherent flaws. For example, single DFAs suffer from an approximation in the exchange-correlation functional (XC Functional). DFAs describe the exchange-correlation interaction between electrons through approximate models (such as local density approximation (LDA) and generalized gradient approximation (GGA). For transition metal systems with strong static correlations (Static Correlation), a single functional cannot accurately capture the multi-reference character (MRC), leading to deviations in energy and electronic structure predictions. For another example, single DFAs suffer from spin-state energy ordering errors. Single DFAs often misjudge the energy level difference between high-spin (HS) and low-spin (LS) states in transition metals.

[0054] Furthermore, the inventors discovered that a single DFA can propagate errors, leading to distorted predictions of molecular properties. The ground-state error of a single DFA can be amplified to the delta-SCF energy gap. A single DFA can also underestimate multi-reference properties.

[0055] Different DFAs have different characteristics and adaptability in different design tasks. This single DFA calculation method, regardless of the specific DFA selected, will always have a single method bias, resulting in large errors in the final calculation results or prediction results of the entire calculation model.

[0056] In this embodiment, different from the single DFA in the prior art, a multi-DFA+consensus mechanism solution is adopted, namely step S03 in the figure.

[0057] In this case, multiple DFAs are not simply a superposition of the number of DFAs, but rather a scheme in which four types coexist. DFAs are divided into different types, and among all DFA types, four types are selected: LDA level, GGA level, hybrid functional level, and bi-hybrid functional level. In the concept of the present invention, the computational model needs to be configured with at least one of each type. Among them, the LDA level is a local density approximation, and the GGA level is a generalized gradient approximation. The combination of the two can adapt to design scenarios with higher efficiency requirements; while the hybrid functional level and bi-hybrid functional level can make more accurate corrections for charge transfer underestimation and hybrid spin misjudgment in the design process, and can ensure the requirements for precision and accuracy in the design scenario. By using the four types of DFA in combination, not only can the design requirements of different scenarios be met, but also the design requirements of precision and efficiency can be met.

[0058] At least one DFA of each type must be configured, and designers can adjust the combination. Generally, the LDA stage can use one or more of the following DFAs: SVWN5, VWN, and Xα; the GGA stage can use one or more of the following DFAs: PBE, BLYP, PBEsol, and revPBE; the hybrid functional stage can use one or more of the following DFAs: B3LYP, PBE0, and HSE06; and the double hybrid functional stage can use one or more of the following DFAs: B2PLYP, ωB97X-D, XYG3, and PWPB95.

[0059] In actual implementation, a specified DFA type set can include 15-25 DFAs. Since the computational model includes numerous DFAs, there will naturally be numerous DFA judgment results. For example, if there are 23 DFAs, there will be 23 judgment results for the target molecule. In this case, rather than the VHTS providing all 23 DFA judgment results to the ML, a consensus mechanism is used to provide the ML with a single, finalized consensus result. This consensus mechanism reduces the computational burden on the subsequent ML while ensuring the accuracy of the ML learning data.

[0060] Specifically, we distinguish between qualitative and quantitative results. Qualitative results are generated based on the proportion of all DFAs identified being greater than a preset threshold. For example, analyzing the ground state spin state of a target molecule is a qualitative question. The question involves determining whether the target molecule is low-spin (LS) under normal conditions. If the preset threshold is 65%, then if 16 out of 23 DFAs are judged as low-spin, the qualitative result is considered low-spin; otherwise, it is considered non-low-spin.

[0061] The quantitative results are generated by averaging the calculated values ​​of all DFAs. This average value can be directly averaged, or it can be averaged using an optimized method. In this embodiment, the calculation method of "removing the extremes and taking the mean to calculate the fluctuations" is adopted. Specifically, when it is necessary to calculate the light absorption energy value of the target molecule, if there are 12 DFAs that calculate the light absorption energies as [2.1, 2.2, 2.3, 2.3, 2.4, 2.4, 2.5, 2.5, 2.6, 2.6, 2.7, 3.0]ev, remove the 4 extreme values ​​at both ends and only keep the 8 in the middle, then it is [2.3, 2.3, 2.4, 2.4, 2.5, 2.5, 2.6, 2.6,]ev, then the final consensus value of the light absorption energy value of the target molecule is 2.45±0.15ev, 2.45 is the average value, and 0.15 represents the fluctuation range.

[0062] In this example, multiple DFAs work together, and the resulting DFA judgment results are also differentiated between qualitative and quantitative. Ultimately, a single qualitative and quantitative consensus value is sent to the ML model. Through practical application, the multi-DFA + consensus mechanism solution has the following benefits: First, bias resistance: While a single DFA can achieve a TMC error of >1eV, the consensus of multiple DFAs compresses the prediction fluctuation to <0.3eV. Second, efficiency optimization: The consensus results directly train the subsequent ML model, rather than training with all DFA judgment results, significantly improving the speed of active learning iterations. Interpretability: The DFA standard deviation is used to identify "difficult molecules" (i.e., molecules with large fluctuations in quantitative judgment results as mentioned above), thus avoiding the blind spots of quantum chemistry methods.

[0063] Subsequent AI calculations for ML models can be the same as existing technologies and will not be described in detail in this article.

[0064] Example 2 differs from Example 1 in that it adds a DFA supplementation step. Specifically, if some important qualitative or quantitative results are inaccurate or the identification is unsuccessful, the computational model introduces additional DFAs for secondary calculation. These DFAs are known as candidate DFAs. While these DFAs often offer greater accuracy, they are more time-consuming to compute and are therefore not used in normal calculations.

[0065] Specifically, if spin state identification requires supplementation, for example, if the preset threshold is 65% as mentioned above, then only 12 of the 23 DFAs are judged as low-spin (LS), meaning the low-spin determination is unsuccessful. In this case, a candidate DFA, such as TPSSh or SCAN, can be introduced for a second supplementary calculation. Users can choose to use the candidate DFA's results as the basis, combine the candidate DFA with the previous DFA results to determine a consensus value, or adjust the weights of the candidate DFA and the standard DFA to generate a consensus value.

[0066] Similarly, if quantitative results require supplementation, such as molecular absorption capacity identification, CAM-B3LYP and DSD-PBEP86 are introduced for secondary supplementation. Similarly, the user can choose to base the consensus value on the judgment result of the candidate DFA, or combine the candidate DFA and the previous DFA results to obtain the consensus value, or adjust the weights of the candidate DFA and the standard DFA to generate the consensus value.

[0067] Example 3, as Figure 2As shown, based on Example 1, there is a further step to optimize the number and types of DFAs. This step occurs between S01 and S03, i.e., the step of determining the optimal DFA set. The design of this final DFA set requires two design decisions. On the one hand, the number of designs must be optimized, which requires a balance between computational cost and accuracy. On the other hand, the selection and combination of specific DFAs must be optimized based on the molecular characteristics.

[0068] In this example, a genetic algorithm (GA) is used for modeling. A GA is a meta-heuristic search and optimization algorithm inspired by natural evolution. It mimics the "survival of the fittest" principle of biological evolution and is used to find optimal or near-optimal solutions to complex problems.

[0069] At the beginning of the genetic algorithm's fitness calculation, it is necessary to determine the initial population. The initial population is the starting solution set for the genetic algorithm's iterative search, consisting of a set of candidate solutions generated randomly or regularly.

[0070] The specific way of setting up the initial population can be determined by the engineering personnel according to the actual design task. In this embodiment, considering the comprehensive consideration of covering diversity and controlling computing costs, the number of DFA combinations in the initial population can be configured to be 150 combinations. In subsequent calculations, the selection, crossover and mutation of DFA combinations are allowed. Furthermore, in order to avoid starting bias and avoid local optimal solution traps, it is defined that each DFA combination requires LDA level, GGA level, hybrid functional level and double hybrid functional level. The quantity ratio can also be customized by the user, such as the ratio of the four levels in each DFA combination is approximately 35%, 25%, 25% and 15%.

[0071] The fitness calculation of the genetic algorithm is the core definition of the fitness calculation algorithm. In this case, the fitness formula is: , where Cost is the computational cost. Different types of DFAs have different COST values. This assignment can be customized by the user, such as defining the LDA level, GGA level, hybrid functional level, and bihybrid functional level to be 1, 4, 10, and 22, respectively. 3 is the penalty weight coefficient. Penalty is the penalty term. As can be seen from the formula, when COST is smaller and the penalty term ΣPenalty is closer to zero, the fitness is greater, which is a better solution.

[0072] The algorithm's constraints, i.e., the design of the penalty term, can be customized by the user. In this embodiment, three are set: Constraint 1: comparison of the ground state spin misjudgment rate with the preset value; Constraint 2: Comparison between fluctuation and preset value; Constraint three: missed detection rate constraint, that is, comparison between multi-reference missed detection rate and preset value.

[0073] These three constraints are written into the penalty term formula above, and some constants are taken. In this embodiment, they are: , , .

[0074] Specifically, is the spin state prediction error rate, and 0.05, i.e. 5%, is the preset critical value, which enables the front-end algorithm to meet the prediction accuracy requirements.

[0075] Specifically, 0.15 is the threshold value preset in this embodiment. is the standard deviation of the reaction energy gap fluctuations, which forces the algorithm to meet the requirement of fluctuation smoothness.

[0076] is the missed detection ratio of strong multi-reference molecules in the model, and 0.03, or 3%, is the preset critical value in this embodiment. This ensures the completeness of the strong multi-reference molecule search.

[0077] Different DFA combinations generate different costs and penalty values. The DFA combination corresponding to the optimal Fitness solution calculated by the algorithm is the optimal FDA set. This step is performed before the computing system uses the FDA set for calculations. The calculation of the optimal FDA set relies on simulation testing of historical data. The test data used in the genetic algorithm calculation can be calculated and tested by the user in a pre-existing database.

[0078] The difference between Example 4 and Example 1 is that this example further defines the establishment of the molecular database in S01.

[0079] like Figure 2 As shown, S01 is refined into the following sub-steps:

[0080] S011, bidentate ligand selection step;

[0081] The bidentate ligand is selected based on whether it is a common element and the number of atoms.

[0082] For example, more than 800 bidentate ligands were screened from the CSD database, with the screening principles being common elements, atomic number ≤ 30, and clear charge.

[0083] Common elements can be limited to only those containing H, B, C, N, O, F, Si, P, S, Cl, Br, and I. A molecular number of ≤30 can control molecular complexity. Having a clear charge identifier ensures the computability of the electronic structure.

[0084] S012, basic complex selection step;

[0085] The two ends of the bidentate ligand are combined with two common metal elements to form a basic complex.

[0086] The two most common metal elements here are iron and cobalt. This is because iron and cobalt are both abundant elements on Earth, ensuring their practicality.

[0087] At this point, approximately 1.3 million basic complexes can be formed.

[0088] S013, functionalization step;

[0089] The basic complex is added with functional groups, such as -NH2, -CN, etc., to form the final molecular database.

[0090] At this point, the number of metal complexes finally formed in the design space is basically more than 30 million.

[0091] This design method ensures that the molecules in the molecular database have a better possibility of synthesis and are abundant on Earth, that is, the elements have abundant reserves in the earth's crust, low mining costs, relatively stable supply and controllable environmental impact.

[0092] The present invention also discloses an electronic device which may include: at least one processor judgment result, at least one network interface, a user interface judgment result, a memory and at least one communication bus judgment result.

[0093] The communication bus determination result can be used to implement the connection and communication of the above-mentioned components.

[0094] The user interface determination result may include buttons, and the optional user interface may further include a standard wired interface and a wireless interface.

[0095] The network interface may include, but is not limited to, a Bluetooth module, an NFC module, a Wi-Fi module, and the like.

[0096] Among them, the processor judgment result may include one or more processing cores. The processor judgment result uses various interfaces and lines to connect the various parts of the entire electronic device, and executes various functions of the electronic device and processes data by running or executing instructions, programs, code sets or instruction sets stored in the memory, and calling data stored in the memory. Optionally, the processor judgment result can be implemented in at least one hardware form of DSP, FPGA, and PLA. The processor judgment result can integrate one or a combination of CPU, GPU, and modem. Among them, the CPU mainly processes the operating system, user interface, and application programs; the GPU is responsible for rendering and drawing the content to be displayed on the display screen; and the modem is used to handle wireless communications. It is understandable that the above-mentioned modem may not be integrated into the processor judgment result, but may be implemented separately through a chip.

[0097] The memory may include either RAM or ROM. Optionally, the memory may include non-transitory computer-readable media. The memory may be used to store instructions, programs, codes, code sets, or instruction sets. The memory may include a program storage area and a data storage area. The program storage area may store instructions for implementing an operating system, instructions for at least one function (such as a touch function, a sound playback function, an image playback function, etc.), instructions for implementing each of the aforementioned method embodiments, etc.; the data storage area may store data related to each of the aforementioned method embodiments, etc. The memory may optionally be at least one storage device located remotely from the processor's determination result. The memory, as a computer storage medium, may include an operating system, a network communication module, a user interface module, and a broadcast upgrade application. The processor's determination result may be used to invoke the application stored in the memory and execute the steps of the highly versatile transition metal chromophore design method described in the aforementioned embodiments.

[0098] The embodiments of this specification also provide a computer-readable storage medium containing instructions that, when executed on a computer or processor, cause the computer or processor to perform one or more steps of the aforementioned broadcast upgrade method embodiment. If the components of the aforementioned electronic device are implemented as software functional units and sold or used as independent products, they may be stored in the computer-readable storage medium.

[0099] In the above embodiments, all or part of the embodiments can be implemented using software, hardware, firmware, or any combination thereof. When implemented using software, all or part of the embodiments can be implemented in the form of a computer program product. The computer program product includes one or more computer instructions. When the computer program instructions are loaded and executed on a computer, all or part of the processes or functions described in the embodiments of this specification are generated. The computer can be a general-purpose computer, a special-purpose computer, a computer network, or other programmable device. The computer instructions can be stored in a computer-readable storage medium or transmitted via the computer-readable storage medium. The computer instructions can be transmitted from one website, computer, server, or data center to another website, computer, server, or data center via wired (e.g., coaxial cable, optical fiber, Digital Subscriber Line (DSL)) or wireless (e.g., infrared, wireless, microwave, etc.) means. The computer-readable storage medium can be any available medium that can be accessed by a computer, or a data storage device such as a server or data center that integrates one or more available media. The available medium may be a magnetic medium (eg, a floppy disk, a hard disk, a magnetic tape), an optical medium (eg, a digital versatile disc (DVD)), or a semiconductor medium (eg, a solid state drive (SSD)).

[0100] Those skilled in the art will appreciate that all or part of the processes in the above-described method embodiments can be implemented by instructing the relevant hardware through a computer program. The program can be stored in a computer-readable storage medium. When executed, the program can include the processes of the above-described method embodiments. The aforementioned storage medium includes various media capable of storing program code, such as ROM, RAM, magnetic disks, or optical disks. The technical features of this embodiment and the implementation scheme can be combined in any manner unless they conflict.

[0101] The embodiments described above are merely preferred embodiments of this specification and are not intended to limit the scope of this specification. Without departing from the design spirit of this specification, various modifications and improvements made to the technical solutions of this specification by ordinary technicians in this field should fall within the scope of protection determined by the claims of this specification.

Claims

1. A highly versatile design method for transition metal chromophores, comprising the following steps: Establishing molecular databases; Scanning and calculating the molecules in the molecular database using a computational model, and considering the molecules whose computational results meet the screening requirements as candidate molecules, wherein the computational model adopts DFA; Its characteristics are: The computational model uses multiple DFAs for computation, and the multiple DFAs are configured such that: in a specified DFA type set, each specified DFA type has at least one DFA; the specified DFA type set includes an LDA level, a GGA level, a hybrid functional level, and a bihybrid functional level; The DFA judgment result is determined by DFA consensus. The judgment result includes qualitative results and quantitative results. The consensus generation of the qualitative result is based on the recognition ratio of all DFAs being greater than a preset threshold. The consensus generation of the quantitative results is generated by averaging the calculated values ​​of all DFAs; The method further includes an optimal DFA set determination step; in this step, an optimal DFA set is calculated based on a genetic algorithm (GA), wherein the optimal DFA set reflects an optimal solution for the number and type of DFAs used in the calculation model, and the constraints in the algorithm include constraint 1: comparison of the ground state spin misjudgment rate with a preset value; Constraint 2: Comparison of Δ-SCF fluctuation with the preset value.

2. The highly versatile design method for transition metal chromophores according to claim 1, characterized in that: The LDA level uses one or more of the following DFAs: SVWN5, VWN, Xα; The GGA level uses one or more of the following DFAs: PBE, BLYP, PBEsol, revPBE; The hybrid functional level adopts one or more of the following DFAs: B3LYP, PBE0, HSE06; The double hybrid functional level adopts one or more of the following DFAs: B2PLYP, ωB97X-D, XYG3, and PWPB95.

3. The highly versatile design method for transition metal chromophores according to claim 1, characterized in that: The method further comprises a DFA supplementation step, in which the calculation model uses a candidate DFA to perform a secondary supplementation calculation on the judgment result.

4. The highly versatile design method for transition metal chromophores according to claim 3, characterized in that: In the DFA supplementation step, if the molecular spin state identification needs to be supplemented, TPSSh and SCAN are introduced for secondary supplementation calculation; if the molecular absorption ability identification needs to be supplemented, CAM-B3LYP and DSD-PBEP86 are introduced for secondary supplementation.

5. The highly versatile design method for transition metal chromophores according to claim 1, characterized in that: The constraint conditions further include constraint condition three: missed detection rate constraint, ie, comparison of the multi-reference missed detection rate with a preset value.

6. The highly versatile design method for transition metal chromophores according to any one of claims 1 to 4, characterized in that: When establishing a molecular database, the following sub-steps are included: bidentate ligand selection step; The bidentate ligand is selected based on whether it is a common element and the number of atoms; Basic complex selection step; Combine the two ends of the bidentate ligand with two common metal elements to form a basic complex; functionalization step; Functional groups are added to the basic complexes to form the final molecular database.

7. The highly versatile design method for transition metal chromophores according to claim 6, characterized in that: In the basic complex selection step, the metal elements are iron and cobalt.

8. An electronic device comprising a processor and a memory; the processor is connected to the memory; the memory is used to store executable program code; the processor runs a program corresponding to the executable program code by reading the executable program code stored in the memory, so as to execute the method according to any one of claims 1 to 7.

9. A computer-readable storage medium having a computer program stored thereon, characterized in that: When the computer program is executed by a processor, the method according to any one of claims 1 to 7 is implemented.

Images